Future developments
Transcription
Future developments
LISTE DES AUTEURS ___________ H.Allender, Tant S. Master plan for waterways maintenance Artières O. Fiber optics monitoring solution for canal dykes Bois J.P. Restoring Navigability to the upper Rhone De Bruyn B. Improve the environmental studies quality of the projects of inland waterways Demenet P.F. Passing vessel interaction study with moored ships - Application to moored large containers Dupray S. New guidance for the use of concrete in maritime engineering Galiana M. Feedback on the repairing works on French navigable dikes and banks Gallicher Lavanne A. Squat measurement in Seine : current navigation conditions and impact of the channel deepening Lerouvillois D. Needs and technologies for port security Morellato D. Constitution of a numerical wave data-base along the French Mediterranean coasts through hind cast simulations over 19792002 Pareau P.E. The Rhone navigation management centre Pauli G. The greening of inland navigation - the case of Rhine navigation Raujouan P. Dredging in France - evolution of the practices Roux S. Determination of hawser forces using numerical and physical models for the third set of Panama locks studies Scherrer P. Port of Le Havre facing the challenge of gigantism of container vessels PIANC MMX Congress Liverpool UK 2010 MASTER PLAN FOR WATERWAYS MAINTENANCE by Henri ALLENDER1 ABSTRACT Devising the waterway maintenance master plan is tantamount to mapping out a maintenance policy and strategy around stated offer and quality of services objectives. It involves defining target network and equipment performance levels based on Voies Navigables de France (VNF) management goals, in terms of safety, availability and reliability. It then involves identifying the most appropriate methods, and sizing the resources and organisations, to achieve them. Developing preventive maintenance methods, in particular, should contribute to achieving performance objectives. Work mapping out this maintenance master plan addressed requirements in turn, namely: 1 : what property is requiring maintenance ? 2 : what state is this property ? 3 : how to bring it up to standard and maintain it ? 4 : what are the priorities ? all the infrastructure manager’s basic inventory status inspection management strategy ranking Mapping out the master plan then rolling it out locally also entails repositioning the maintenance function within the organisation. Today, maintenance is considered a general operating expense. Resources in general (and human resources in particular) are more and more tighter, and maintenance resources are therefore following suit. The main goal is to position maintenance as a function that makes a direct and essential contribution to network performance, enabling VNF to embark on a continuous-improvement drive and to harness modern maintenance methods. The human dimension will have to pervade absolutely every aspect of this project. And that, no doubt, will be one of the most promising action levers. This paper presents the methodological options that have been chosen, and explains why they match the different issues that we have identified. 1. Property inventory (the essential knowledge basis) The first objective was to consolidate a physical inventory of the main types of structures in the waterway network. We used all the sources of information available at VNF HQ and local branches, then ran a full network on-site inspection, to do so. Then we classified VNF’s extraordinary variety of property according to its function. Each structure is therefore classified based on function then type, as shown in table 1. Crossing falls – Locks Crossing falls – Water slopes and lifts Navigation weir – Fixed Navigation weirs – Movable and combined Alimentary systems – Fixed intake weirs Alimentary systems – Movable and combined intake systems Crossing – Moveable bridges Crossing – Fixed bridges Crossing – Canal bridges 1 Technical advisor, Voies navigables de France 1 of 17 PIANC MMX Congress Liverpool UK 2010 Crossing – Canal tunnels Crossing – Aqueducts under canals (straight or siphon) Outlet works – Side spillways (fixed) Outlet works – Siphons Outlet works – Valves Outlet works – Guard gates Alimentary systems – Storage dams Alimentary systems – Lock by-passes Alimentary systems – Gravity channels Alimentary systems – Pumping stations Alimentary systems – Lift stations Alimentary systems – Intake gates Alimentary systems – Regulator valves Table 1: structure classification Compiling this physical inventory was also an opportunity to gather the main information pertaining to each structure, including its location (GPS coordinates), standard dimensions (lock length, weir width, etc.), standard information and operation (manual, mechanical, automatic, etc.). There are now more than 4,300 structures (excl. linear sections) in the inventory. The largest categories (encompassing more than 100 structures) include locks (1800), crossings (960, incl. 700 aqueducts, 128 canal bridges and 28 tunnels), outlet works (640), navigation weirs (400), and alimentary systems (390, incl. 52 storage dams). Each structure category then has been divided in structural parts, as shown in fig. 1, to provide a framework for the method used to classify property condition, discussed in the next chapter. Structure (work) lock upstream head Part Equipement operating system gate sieve wall Fig. 1: structural decomposition We then completed the inventory with structure inspections to classify their functional status with a view to providing the management body with a full picture of its property’s residual value. 2. Inspecting property to ascertain its residual value We then completed the inventory by running inspections (visual inspections rather than in-depth expert reviews) in order to classify each structure’s and linear section’s functional status and thereby build a consistent picture of the structures as a whole. 2 of 17 P PIANC MMX Congress C Live erpool UK 2010 The firstt France-wid de inspection covered 100% 1 of the e discrete sttructures an nd 10% of the linear sections (the sample was chossen to provid de a satisfa actory and re epresentative e overview of global status). The metthod to classsify discrete e structures and linear sections s wass based on the FMECA A (Failure Mode, Effects, E and Criticality C Ana alysis) appro oach, which enabled e us to o: • Break down functions an nd run preliminary functio onal analysess on each strructure type; • Analyse failure modes (i.e. the syymptoms rev A vealing malffunction, sub o bstandard operation, s substandard d quality, etc.) then system matically ana alyse their efffect on overa all structure function; f • Use a critica ality scale bassed on expert input and rank r failure modes m for ea ach scenario. We then n directly obsserved structural degrada ation and ma aterial deterio oration, and assessed the risks of function failure (as well w as identiifying failure to comply with w regulatory requireme ents and saffety risks) uides drafted based on FM MECA resultts. using gu This metthod allowed d us to syste ematically asssess structu ure functiona al status by a assessing ea ach piece of equip pment’s funcctional statuss. The questtion, in a nu utshell, was “is this stru ucture fit to serve its purpose?” This ana alysis also en ntailed breakking down fun nctions into five f generic categories, c a as shown in table t 2. • Protects people (maintenanc ce ns staff, users and a staff, operation neighbours). Covers C all available safetty equipment (lifelines on manually-operate ed s on locks, dams, bollards ladders in cha ambers, footbridges an nd barriers, lifebelts, etc.) • Managing water le evels to ovide users with the water pro resources they req quire. Dams e especially impo ortant here. are maintening waterr leveels safety • Floatting-unit circulattion. Lock equip pment is especia ally important here. navig gation • Concern that natural ments should be environm handled sympathetically s y, maintainiing ecological continuity y. environm mental protectio on • Providing lon ng-term structure e service continuity. continuity Ta able 2: the five generic functions used u to inspect structurres 3 of 17 PIANC MMX Congress Liverpool UK 2010 The scale encompasses 4 levels: Level 1 – Functional, requires regular maintenance Level 2 – Deterioration can entail function failure over the medium term Level 3 – Deterioration can entail function failure over the short term Level 4 – Not functional, likely to fail over the very short term Each degradation was also classified according to its associated speciality civil engineering, hydromechanical, electrical, etc.) and extent. Each piece of equipment was therefore ranked according to functional status (level 1-4) and damage extent (<10%, 11%<<33%, 34%<<66%, 67%<<90%, >91%). A specific and global indicator for each discrete structure was built to provide an overview of its condition, as shown in fig. 2, to classify each structure on a scale from I (low risk of failure) to IV (high risk of failure): • • • • Class IV: structure is in very poor condition, with the equivalent of level-4 functional status symptoms on 2 major pieces of equipment, and the risk of total structure failure is high Class III: structure is in poor condition, with the equivalent of level-3 functional status symptoms on 2 major pieces of equipment, and the risk of total structure failure or substandard performance is high Class II: structure is in average condition, with the equivalent of level-3 functional status symptoms on 1 major piece of equipment, and is damaged to the point of entailing a risk of substandard performance Class I: structure is in good general condition and fit to fulfil its function. Equipement condition level : 1,2,3 or 4 Note per piece of equipement Nequipement Equipement importance Iequipement Level 1 : Nequipement = 1 Level 2 : Nequipement = 10 Level 3 : Nequipement = 100 Level 4 : Nequipement = 1000 For a structure: ∑i Iequipement = 100 Note per structure Nstructure Nstructure ∑i Nequipement × Iequipement 1/100 and classification (I, II, III, IV) Fig. 2: the overall functional status indicator 4 of 17 PIANC MMX Congress Liverpool UK 2010 3. Property refurbishing and maintenance as part of property management strategy Generally speaking, “policy” refers to what we want to do and “strategy” refers to how we do it. Under this initiative, VNF mapped out its maintenance policy to provide general objectives for its maintenance function and mapped out the main areas for improvement to develop it. It carved out its maintenance policy around three main principles: Æ Sustainable maintenance (by raising professional standards and promoting the maintenance function); Æ Sustainable waterways (exemplary water and property management, and strict environmental compliance); Æ Sustainable transport (tailoring services for commercial users, tourists and hydraulic management). Maintenance strategy, conversely, hinges heavily on the starting point, i.e. in particular on equipment obsolescence levels. Public authorities had paid little attention to network over the previous decades, and had not organised regular equipment maintenance or renewal properly. Property was therefore considerably degraded. VNF accordingly decided to base its maintenance strategy on three types of action plans, as shown in fig. 3. They follow: • Regular maintenance, i.e. regular property maintenance (including preventive, corrective and enhancement maintenance), to keep structures in proper working condition; • Refurbishing structures, i.e. restoring original structure functions and substance in order to offset past neglect (which is not the same as repair work, which is corrective action and comes under running maintenance); • Regenerating structures, i.e. running all renewal operations on equipment or parts thereof, and heavy-duty maintenance on long-life equipment, to extend property service life. Equipement condition New Minimum acceptable Period A Period B Period A: no regular and regeneration maintenance policy Period B: regular and regeneration maintenance policy Main Arrow : offsetting past neglect Orange arrows: regeneration Fig. 3: the global maintenance strategy 5 of 17 Equipement age PIANC MMX Congress Liverpool UK 2010 Each component’s relative importance will evolve over time (all other things being the same), as fig. 4 shows: • From a medium-term perspective (0-10 years), the main goal is to refurbish property; • From a long-term perspective (10+ years), once property has been refurbished, the main goal is to regenerate it in order to accommodate sustainable-management principles; • Running maintenance is essential regardless, but is nevertheless on a downward trend over the long term as refurbishing work spurs a virtuous circle around preventive maintenance and continuous improvement; • The total workload is also on a downward long-term trend in step with equipment lifecycle optimisation. Regeneration Refurbishing Regular maintenance corrective Regular maintenance preventive Regular maintenance improvement Medium term Long term Fig. 4: short- and long-term changes in running-maintenance, refurbishing and regeneration workloads We will not be going into details on this issue in this paper but it is nevertheless worth pointing out that this maintenance strategy itself has to evolve within a broader investment strategy encompassing a variety of modernisation projects (e.g. demolishing manually-operated weirs to build mechanically operated ones) and development projects (e.g. increasing lock capacity or widening waterways). 3.1 Assessing refurbishing requirements Work estimating property refurbishing or neglect-offsetting requirements is based on observed degradation (level 1-4 and extent), and on global reference unit pricing. Estimates are also adjusted to accommodate inspectors’ recommendations on the type of work required (repair, replacement, enhancement). Including provisions to refurbish weirs’ underwater structures and equipment using a statistical approach based on expert input on average weir age and residual service life (these structures and equipment accounted for a large portion of the inspection but visual inspections were deemed inadequate). Doing this doubled the estimated refurbishing requirements for the structures and equipment assessed using the statistical approach. As in the case of discrete structures, degradation observed on inspected linear sections entailed estimating refurbishing costs using a set of reference prices. Each inspected reach was then classified based on the cost of refurbishing one km of it. Extrapolation to 100% of the waterway’s linear section involves waterway-scale interpolation: the cost of refurbishing reaches between two inspected reaches is linearly deducted from the cost of refurbishing the linear sections on either side. 6 of 17 PIANC MMX Congress Liverpool UK 2010 We estimated refurbishing requirements without ranking refurbishing work per se. We then ran them through a tool to rank action priorities as explained in chapter 4 about analysing risks. 3.2 Assessing regeneration requirements We built two complementary approaches in order to cost property regeneration requirements. As shown in fig. 5, the first approach simply entails establishing yearly averages to ascertain requirements, and the second one involves piecing together theoretical regeneration peaks over time. Regeneration needs Approach 2 Approach 1 Initial catch-up phase Time Fig. 5: Changes in regeneration requirements – Approaches 1 and 2 The first approach is based on: • Equipment regeneration unit costs (unit cost base); • The equipment inventory; • The nominal theoretical service life for each type of equipment (adjusted based on measured equipment life). Then we calculated each piece of equipment’s: • Ct (regeneration cost) • Dt (theoretical service life) The resulting average annual global requirement is therefore: (1) The second approach determines the theoretical date by which a specific piece of equipment will require regeneration work based on its functional status at point in time ‘to’. This process involves three steps on each piece of equipment: 7 of 17 PIANC MMX Congress Liverpool UK 2010 Step 1: establishing probability levels Probability, here, refers to the probability of a piece of equipment failing, on a scale of 1 to 6. 1: extremely improbable 2: very improbable 3: improbable 4: probable 5: very probable 6: practically certain Degradation extent It is determined based on functional status (level 1-4) and the extent of observed degradation, using Matrix 1. Degradation level 4 3 2 1 91%-100% 6 5 4 1 67%-90% 5 4 3 1 34%-66% 4 3 2 1 11%-33% 3 2 2 1 0%-10% 2 2 2 1 Matrix 1: probability matrix Step 2: establishing theoretical service life and theoretical residual service life We have to include a new variable at this stage – whether the piece of equipment comes under civil engineering or not – to accommodate the considerable differences in degradation kinetics. Transfer curves (fig. 6 and fig. 7) determine the probability level associated with each given theoretical service life. Probability level Year Fig. 6: changes in probability (civil engineering) Probability level Year Fig. 7: changes in probability (others) 8 of 17 PIANC MMX Congress Liverpool UK 2010 Step 3: establishing regeneration dates We calculate average residual service life by subtracting theoretical residual service life (determined in Step 2) from nominal service life. That way, we can schedule regeneration work. 3.3 Assessing regular-maintenance requirements Regular maintenance requirements are estimated in terms of annual workloads. Estimates are based on: • Maintenance plans for five representative structures (a large lock, a small lock, a stop-plank weir, a manually-operated weir and a reach), focusing in particular on pending preventive and regulatory maintenance, and specifying the associated workload; • Estimated loads for each piece of equipment; • The property inventory. That way, we can calculate the total workload. This calculation factors in other adjustment variables. In particular: • Structure use (less frequent maintenance work on structures with lower stakes); • Structure size (less maintenance work on smaller structures). The total maintenance workload (Referent Technical Time or RTT) that we consolidate using that method can then be converted into full-time equivalent (FTE). Doing this involves taking into account several ratios to estimate total effective time (time spent on actual maintenance operations plus time spent on all ancillary operations). o o o o o o Preventive-maintenance ratio (Rpm): ratio preventive maintenance/running maintenance; Logistics ratio (Rlog): preparing operations and travelling; Management ratio (Rman): on-site maintenance team supervision; Presence ratio (Rpre): encompassing sick leave and training; Optimisation ratio (Ropt): room for improvement in organisations; Enhancement maintenance (Rim): extra workload for operations staff to enhance maintenance practices (structure maintenance plan, etc.). Structure functional status class (and the associated impact of unforeseen failure hampering scheduled preventive-maintenance work) is also used to multiply RTT by Rstate equal to: • 1.1 on Class II structures; • 1.4 on Class III structures; • 2.0 on Class IV structures. Staff (FTE)= RTT 1607 × Rlog × Rman × Rpre × Ropt × Rim × Rstate ÷ Rpm (2) Note: 1607 is the number of hours that an employee in France works per year (35 hours a week). 4. Ranking refurbishing work In VNF’s maintenance strategy, running maintenance and regeneration are an incompressible whole that is performed in full as per sustainable-management principles. Decisions on refurbishing-work action plans, however, are made using a different principle, based on accepting the risk of structure failure. To that end, we have used a global risk-management approach based on the large number of structures we manage, to rank priorities based on network technical and strategic performance levels depending on structure-related stakes and the waterways they are on. The refurbishing work action plan is also part of a global investment strategy spanning a wide spectrum of investment programmes (maintenance, modernisation, development and safety). 9 of 17 PIANC MMX Congress Liverpool UK 2010 4.1 Analysing risks – The principle The first result we got when we ran the risk analysis was a map of structure risks. Each structure has an associated risk index (Ir) according to the consequences should the risk materialise (gravity or G) and probability it will occur (P). 4: moderate consequence 5: serious consequence 6: catastrophe 6: practically certain 5: very probable 4: probable 3: improbable 2: very improbable 1: extremely improbable 3: minor consequence P 2: negligible consequence G 1: no consequence Ir = f (G; P) 6 22 27 31 33 36 5 17 23 28 32 35 4 13 18 24 29 34 3 10 14 19 25 30 2 8 11 15 20 26 1 7 9 12 16 21 Matrix 2: risk matrix Consequence is closely associated with a structure’s strategic weight, i.e. its importance in the light of its purpose. It is ranked on a scale from 1 (inconsequential) to 6 (catastrophic) as shown on Matrix 2. Probability is linked to a structure’s and its equipment’s functional status. It is ranked on a scale from 1 (practically impossible) to 6 (practically certain) as shown on Matrix 2. That way, we can rank risk on a scale from 1 to 36. Risk values are established on the matrix at the start. The two most noteworthy choices we made when building the matrix follow: • the 6 lowest risks are in column 1 (“no consequence”); • risk values in the matrix reflect the fact that consequence takes priority over probability; Classical risk-analysis methods typically focus on reducing risk by taking action to contain both probability and consequence. At VNF, however – and excluding situations where safety is at stake – the only option is to reduce probability as VNF’s overriding goal is to increase activity on its waterway network i.e., in a way, to increase consequence ! 4.2 Assessing strategic importance before analysing risks A structure’s strategic importance depends on the strategic importance of the corridor it is on. A corridor’s strategic importance does not hinge on a single factor: it varies according to the three main function categories that VNF uses to classify waterways, namely commercial navigation (merchandise transport, freight), water management (intakes and outlets, interaction with the environment, water-level management and hydroelectric power) and tourism (use of the waterway for touristic or leisure purposes), as shown in fig. 8. 10 of 17 PIANC MMX Congress Liverpool UK 2010 Corridor (or Itinerary) (e.g. Moselle) Strategic importance for commercial navigation Strategic importance for tourism Strategic importance for water management Fig. 8: a waterway’s 3 function categories Work on each waterway function category involved: • Establishing and choosing the most representative variables to quantify the importance of each function category. The main examples follow. o o o Commercial navigation: The amount of merchandise carried (in tonnes); The number of times goods-carrying river boats travel through locks; The weight of containers carried (in tonnes); Water management: The amount of water taken into and released from the network; The position of sensitive or protected natural environments; The use of land adjoining waterways; Total hydroelectric power plant output; Tourism: The number of times pleasure boats travel through locks; The presence of rental facilities and marinas; • Processing figures to be able to compare them on a single scale and rank waterway sections (sections are the smallest linear waterway divisions in the VNF geo-reference base). These sections are what make up the clusters used for the statistical analysis discussed below; • Running data through a statistical analysis to segment it and distil classes. Given the variety of variables, we used the k-means method. The advantage is that it segments a given group into homogeneous k-classes (minimising variance between class components and maximising differences between classes), as k is set from the start, and it weights each variable used in the analysis. This work led to three maps (one for each waterway purpose). We then rounded off this approach by blending findings to provide an at-a-glance picture of each corridor’s strategic importance. In this model, commercial navigation weighs 50%, water management 30% and tourism 20%. 11 of 17 PIANC MMX Congress Liverpool UK 2010 Unprocessed data for the three function categories Processing data for the three function categories, scoring sections K-means statistical analysis Fig. 9: An overview of the process to determine corridor strategic importance On the structure scale, the class used was the class of the corridor. By extension, the class on the equipment scale is the class on the structure scale. 4.3 Risk analysis Risk analysis focuses on the three waterway functions identified above, plus a component covering the safety of people and property. We assess structure-related levels by first of all assessing the pieces of equipment in it: A given piece of equipment’s probability level is based on their functional status: Pequipment = f (degradation level; degradation extent) 12 of 17 PIANC MMX Congress Liverpool UK 2010 Degradation extent Degradation level 4 3 2 1 91%-100% 6 5 4 1 67%-90% 5 4 3 1 34%-66% 4 3 2 1 11%-33% 3 2 2 1 0%-10% 2 2 2 1 Matrix 3: probability level Probability levels, in other words, are independent from waterway function category. Consequence levels are determined based on: o A piece of equipment’s contribution to the function under review, on a scale from 1 (no contribution) to 6 (strategic contribution); o Its strategic importance (the structure’s strategic importance vis-à-vis the function under review. Gequipment = f (contribution level; strategic importance) Contribution level Strategic importance A B C D E Level 6 6 5 4 3 1 Level 5 5 4 3 2 1 Level 4 4 3 2 2 1 Level 3 3 2 2 1 1 Level 2 2 2 1 1 1 Level 1 2 1 1 1 1 Matrix 4: consequence level (in this case, for commercial navigation) The resulting risk level for the structure is the maximum risk level found on its equipment. Defining acceptability thresholds It is important to define the acceptability threshold of a residual risk as it has a direct impact on the ensuing action plan. Fig. 10 shows how the acceptability threshold impacts action-plan sizing (the higher the acceptability threshold, the smaller the action plan). intermediate acceptability threshold minimal acceptability threshold Risk All risks are rejected : the plan of action is maximal (there is no organization into hierarchy) maximal acceptability threshold index An intermediate level of risk is accepted : the plan of action is ranked according to priorities All risks are accepted : the plan of action is empty Fig. 10: acceptability thresholds and action-plan sizing 13 of 17 P PIANC MMX Congress C Live erpool UK 2010 In practicce, the notion of accepta ability does no ot follow an entirely e binarry rationale. The ALA ARP (As Low w As Reasona ably Practica able) principle can be use ed to build sccenarios and d group risks into o three zoness, as shown in fig. 11: • An upper arrea, where risks A r are into olerable rega ardless of th he associate ed constraintts. In this a area, action to address risk has to be e taken regarrdless of cosst; • An intermediate area wh A here whetherr or not a ris sk is accepta able may be assessed on n a caseby-case basis, according g to feasibilityy and opporttunity; • A lower area a where riskss are deeme ed acceptable e by default. Here, the co ost of possib ble action t address riisks is out off proportion with to w the resulting risk reduction. Risk in ndex e zone Unacceptable Acceptable zone ALARP zone Fig. 11: the t ALARP principle 5- The impact off the maintenance strategy s on n the quallity of serv vice we prrovide for use ers As statted above, the mainte enance stra ategy aimin ng to imprrove river-network perfformance encompa asses three types t of worrk, namely re efurbishing, re egular mainttenance and regeneration n. It is also o important to o measure th he contribution that the maintenance m e function ma akes to impro oving the waterwa ay network’s performance e in order to o gauge action efficiency, and to ad djust strategies when and as required in lin ne with effortts to embark efforts on a continuous improvementt drive. We used d three indica ators to measure, as sho own in fig. 12 : Actual availlability : the actual a numbe er of days strructures are unavailable Safety : the e actual number of safetyy flaws obserrved on strucctures Theoritical availability : a theoretica al number off days structu ures are unavailable e F 12: netw Fig. work perform mance indic cators 14 of 17 PIANC MMX Congress Liverpool UK 2010 The rest of this chapter expounds on the principles we used to build the theoretical-availability indicator, which we can also use to extrapolate service quality forecasts. We use this indicator to bridge the gap between a structure’s functional status and the intrinsic risk of loss of function that it represents. This indicator at minimum reflects a given structure’s intrinsic reliability (which is inherent to each design). It increases in synch with structure obsolescence and use. Intrinsic reliability (design) Rintrinsic Functional status (obsolescence level) : 1-4 level and extent Use (number of times the structure is used during a period of time), includes an adjustment factor Theoretical availability of structure equipment Cuse Requipment Theoretical availability of structure R Fig. 13: the three factors determining theoretical availability To calculate this indicator, we accept that equipment failure entails structure failure (series model). Thus: R = Rintrinsic× ∏i Aequipment × Cuse (3) Theoretical availability of equipment Aequipment This variable is associated with a piece of equipment’s status (level 1-4) and the extent of its degradation (see chapter 2). Each degradation-extent category is based on expert input on theoretical downtime timeframes due to failure, as shown in table 3 (this example is for electric or hydraulic failures). Electric, hydraulic speciality Extent of degradation Theorical duration of failure (day) <10% 0,5 11-33% 0,75 34-66% 1 67-90% 2 Table 3: theoretical d (electric or hydraulic failure in this example) 15 of 17 90-100% 3 PIANC MMX Congress Liverpool UK 2010 Each degradation level (level 1-4) is based on expert input on the probability of failure occurring. For a given period, functional status level 4 is where the probability of failure occurring is the highest. At functional status level 3, the probability of failure occurring is 10 times lower than at level 4. At functional status level 2, the probability of failure occurring is 100 times lower than at level 4. Electric, hydraulic speciality Extent of degradation <10% 11-33% 34-66% 67-90% 90-100% Theorical duration of failure (day) 0,5 0,75 1 2 3 degradation level 4 1 chance on 2 1 chance on 2 1 chance on 2 1 chance on 2 1 chance on 2 Probability of occurrence degradation level 3 1 chance on 20 1 chance on 20 1 chance on 20 1 chance on 20 1 chance on 20 degradation level 2 1 chance on 200 1 chance on 200 1 chance on 200 1 chance on 200 1 chance on 200 Table 4: probability (electric or hydraulic failure in this example) Lastly, we determine theoretical annual downtime averages by dividing theoretical downtime by the probability of it occurring, as shown in table 5. Electric, hydraulic speciality Extent of degradation <10% 11-33% 34-66% Theorical duration of failure (day) 0,5 0,75 1 degradation level 4 1 chance on 2 1 chance on 2 1 chance on 2 Probability of occurrence degradation level 3 1 chance on 20 1 chance on 20 1 chance on 20 degradation level 2 1 chance on 200 1 chance on 200 1 chance on 200 0,2500 0,3750 0,5000 Theoretical average number of degradation level 4 0,0250 0,0375 0,0500 days per year of potential arrest - degradation level 3 degradation level 2 0,0025 0,0038 0,0050 Down time 67-90% 2 90-100% 3 1 chance on 2 1 chance on 2 1 chance on 20 1 chance on 20 1 chance on 200 1 chance on 200 1,0000 0,1000 0,0100 1,5000 0,1500 0,0150 Table 5: theoretical annual downtime average (electric or hydraulic failure in this example) We calculate theoretical availability using the formula below: (4) Required time –Down time Aequipement= Required time 16 of 17 PIANC MMX Congress Liverpool UK 2010 CONCLUSION The issues we have addressed at this point provide the first foundation underpinning a global maintenance policy and strategy for France’s waterway network, added to the extraordinary potential that nearly 1,500 members of maintenance staff can pool. Effectively rolling out the maintenance strategy will still require large-scale work to home in on the details on the various possible scenarios matching service-standard goals, and social and budget targets. The total figure is €1.3 billion (total, that is, without ranking priorities) at this point, to catch up (as per the method described above). At this point, the goal is to fine-tune the requirements, in particular by analysing risks and by establishing residual-risk acceptability thresholds. The first calculations aiming to map out global investment requirements point to roughly € 300 million per year, encompassing refurbishing work (including safety and compliance), modernisation and development (excluding the exceptional Seine-Nord Europe canal project). Implementing this strategy will also necessarily involve embarking on a continuous improvement process (a PDCA cycle) within the maintenance function. The main goals follow: • Keep the property inventory up to date; • Periodically re-inspect property to track developments; • Reconsider and optimise property maintenance plans; • Define criteria to trigger regeneration work; • Use available tools to foster progress, namely operation and performance indicators, and CAM (computer-assisted maintenance). Maintenance is no exception to the rule. “The wind only blows in the right direction for they who know where they are going.” 17 of 17 PIANC MMX Congress Liverpool UK 2010 FIBER OPTICS MONITORING SOLUTION FOR CANAL DYKES by 1 2 3 O. Artières , M. Galiana , P. Royet , Y.-L. Beck4, P. Cunat5, J.-R. Courivaud6, J.-J. Fry7, Y.H. Faure8,C. Guidoux9 ABSTRACT A fibre optics textile composite monitoring solution has been developed to detect and localize both leaks and early signs of failure (erosion, blocking, breaches, sliding, and settlements) of hydraulic works, and specially wet dikes of canals and harbours. This unique solution can detect at the same time changes in temperature and strain. It has been installed in several experimental and real sites. The performance of the GeoDetect® solution results in the combination of the high sensitiveness of the textile composite sensor connected to the relevant instrumentation to assess the soil properties changes, and the powerful mathematical model to analyse the measured data. Variations as small as 0.02% in soil strain and leaks as low as 0.1 l/min/m have been measured to detect and localise the early signs of failure. 1. INTRODUCTION Owners of waterways are managing a wide number of works with a large variety of ages and stages. The problem facing the canal owner today is to limit the impact of this aging phenomenon. The most serious task is the “safety” by eliminating the risk of failure. In the last decade a large number of dam and dike failures occurred. Management of the water resource is the second important topic: leakage control becomes more and more important. For this purpose, a technical solution that detects and localizes both malfunctions, precursors of failure (erosion, blocking, breaches, sliding, settlements), and leaks was developed. It is a reliable early localization and warning system for both gradual and catastrophic dam and dike failure. 2. GEODETECT® : THE GEOTEXTILE FIBER OPTICS MONITORING SOLUTION This development was handled through a partnership of companies and public research institutes within the Eureka labialized project “SafeDike”. The use of fibre optics in structural health monitoring systems for civil engineering applications have been widely used for many years. By integrating fibre optic sensing into a geotextile fabric as shown in Fig. 1, GeoDetect® is the first system designed specifically for geotechnical applications. It embodies a geocomposite fabric, fibre optics and instrumentation to provide a clearly innovative solution for the multi-functional requirements of a geotechnical application e.g. in-plane drainage capability, anchoring interface with the soil, protection of the fibre, reinforcement and data capture. It uses stimulated Brillouin or Raman scattering technology in single mode or multi-mode fibres to measure strain and/or temperature. A specific analysis of the raw data measured by the monitoring system often increases the accuracy and the speed of the early detection (Beck et al, 2009; Cunat et al., 2009a; Cunat et al., 2009b). GeoDetect® is designed to detect the first steps of internal erosion processes and hydraulic works instability. The detection of the leaks, which is the early stage of the internal erosion process, is 1 Global technology manager GeoDetect® solution, TenCate Geosynthetics, France, [email protected] 2 Engineer, Cetmef, France, [email protected] 3 Senior expert, Cemagref, France, [email protected] 4 Project Manager, EDF-DTG, France, [email protected] 5 PhD, EDF-DTG, France, [email protected] 6 Engineer Specialist, EDF-CIH, France, [email protected] 7 Senior expert, EDF-CIH, France, [email protected] 8 Professor, LTHE/University Joseph Fourier Grenoble, France, [email protected] 9 Fiber optics development manager, GeoPhyconsult, France, [email protected] 1 of 11 PIANC MMX Congress Liverpool UK 2010 assessed through the measurement of temperature changes using the passive method or the active heat pulse method (Radzicki et al., 2009). First stages of dike settlement or sliding are detected by strain measurement. In comparison to existing detection systems, the GeoDetect® solution is a distributed and continuous measurement along the whole canal length, which increases the accuracy and the speed of response, both crucial parameters to prevent collapse. It can provide a leak and deformation location with a spatial resolution of 1 meter, or even 0.5 m in some cases. The system is able to monitor several tenths of kilometres. The GeoDetect® solution has already been validated on several 1:1 scale experimental works and real dikes in-use. Different monitoring strategies may be designed, for example temporary monitoring, or continuous monitoring to be used as an early warning system. Figure 1: View of the GeoDetect® S-BR sensor with the embedded coloured optical cables 3. LEAK DETECTION THROUGH SOIL DYKES 3.1 The PERINE experimental basin in Aix-en-Provence (France) To test and validate the GeoDetect® solution, as well as other equipment and methods dedicated to detecting and quantifying leaks from a real structure, an experimental basin was built during the second quarter of 2006 on the Cemagref site in Aix-en-Provence, France. 2 of 11 PIANC MMX Congress Liverpool UK 2010 Figure 2: View of experimental PEERINE basin in Aix-en-Provence The main characteristics of the experimental basin (Fig. 2) are as follows: • Volume of materials: 1200 m3 of clayey materials (permeability at saturation of 10-11 m/s). • Perimeter: 118 m at the foot, 78 m at the head. • Coating: upstream geomembrane lining system • Approximate water volume: 200 m3 The GeoDetect® system was laid at three levels (referred to as T2, M2 and B2, respectively for top, middle and bottom) along the downstream embankment below a gravel abutment layer (Fig. 3). There are artificial leaks situated at two different heights. These local soil heterogeneities are made of coarse gravel. This provides a maximum flow rate of approximately 10 l/min. The main result from the experimental program is today the possibility to detect leaks with the order of magnitude of 1 l/min./m from an analysis of the residuals and the sum square error (SSE) (Fig. 4). Further results are given in Artières et al. (2007). GeoDetect® PANEL Artificial leaks Figure 3: Section of one dyke of the basin PEERINE basin with the artificial leaks Leaks Figure 4: Example of the results from the EDF/Cemagref temperature analysis model. M2 is the strip at the middle of the slope, B2 is the strip at the bottom of the slope. 3 of 11 PIANC MMX Congress Liverpool UK 2010 3.2 The Ijkdijk / piping experimental project : detection of internal erosion Four experimental dyke were built within the Dutch IJkDijk project (Smart Calibration Dyke), Piping part, to test the ability of sensoring systems to detect the first signs of internal erosion. This project funded by the Dutch government and a consortium of several companies and institutes. The IJkdijk piping project was carried out the second half of the year 2009 in North part of The Netherlands (IJkdijk, 2009). To investigate "piping", a test dike was built specially, containing advanced monitoring equipment. The phenomenon was initiated in a controlled way, resulting in the collapse of the test dike. The experiment and the measurements show that piping is a failure mechanism that needs to be taken seriously. The experiment took place in a section 4 m deep, 40 m long and 25 m wide. The subsurface in the section consisted of sand. Over the width of the section, there was a dike with high water on one side (more than 2.5 m) and low water on the other (0.1 m). The experimental dykes are about 15 m long and 4 m high, as shown on Fig. 5. The experiment was unique: this was the first time that the mechanism had been simulated and observed on this scale in a controlled setting. For the first time, it was also possible to demonstrate that there is a link between piping and the actual failure of a dike. The water head was progressively increased up to 2.6 m high up to the dyke collapses (Fig. 6). On the test number 4, six GeoDetect® S-BR strips were installed lengthwise the dyke closed to the interface between the sand and the dike, at a depth of 10 cm into the sand. Despite this depth from the interface were the flow channel occurs, the change of the temperature profile indicate the start of the piping channel (Fig. 7). Without data interpretation, the piping channel was detected 17 hours before collapse. With deeper data analysis carried out by EDF has shown an earlier detection threshold at least one or 2 days before the dyke collapses. Figure 5: The IJkdijk piping dyke during the test 4 of 11 PIANC MMX Congress Liverpool UK 2010 Figure 6: The IJkdijk Piping test after collapse due to internal erosion at the bottom interface Figure 7: The GeoDetect® temperature profiles analysed with the EDF/Geophyconsult model at different time, resp. 17 hours, 11 hours, 5 hours before failure and at the failure, from the top left to the bottom right. 5 of 11 PIANC MMX Congress Liverpool UK 2010 4. LEAK DETECTION THROUGH THIN LINING SYSTEM An experimental study was carried out in the PEERINE basin at the Cemagref in Aix-en-Provence, as previously described, to test the GeoDetect® performance to detect leaks through thin lining system, such as geomembrane, applied on the upstream face of the dike. Two configurations were tested, as described in Fig. 8: one slope covered with a granular drainage layer, one slope with a geocomposite drainage layer (geonet). These configurations represent very common design of canal lining. In this case, the purpose of GeoDetect® is the assessment of the amount of leakage through the liner in the perspective of a better control of the water resource. Artificial leaks were placed at different location through the geomembrane liner, as indicated in Fig. 9, able to control leaks as low as 0.2 l/min. These leaks were seen with the GeoDetect® system (Fig. 10). Concrete slabs (0,5 x 0,5 x 0,05) 2,5 m Geomembrane Artificial leaks 30 cm Protection geotextile 1 2 Location of the GeoDetect strips Filtration geotextile Granular drainage layer Drainage pipe 100 mm Protection geotextile Concrete slabs (0,5 x 0,5 x 0,05) Geomembrane 2,5 m Protection geotextile Top soil, 0,5 m GWB 1 2 GeoDetect strips Drainage geocomposite Filtration geotextile Drainage pipe 100 mm Figure 8: Two configuration on the use of GeoDetect® to detect leaks through a geomembrane lining system, with a granular drainage layer (top) and a geocomposite drainage layer (bottom). 6 of 11 PIANC MMX Congress Liverpool UK 2010 TOP VIEW Location of the leaks through the geomembrane Position des fuites à travers la me E Vue en Plan 36 m N 34,5 m 27 m 2% (Sol naturel) 24 m 11 m A GEM1 14 m GEB GEM2 21,5 m 23 m GEH 14 m 1m Vidange de fond GWB GWM2 GWH GWM1 Clôture Caniveau périphérique W O Figure 9: Location of the artificial leaks through the geomembrane of the PEERINE basin (green bars) Figure 10: Detection of the artificial leak (black arrows) with the GeoDetect® system for the strip installed at the toe of the slope (blue line) and at the middle of the slope (green line), data analysed by EDF and Cemagref. 7 of 11 PIANC MMX Congress Liverpool UK 2010 5. SOIL STRAIN DETECTION A first experimental dyke was built within the Dutch IJkDijk/Macrostability project in 2008 to test the ability of sensoring systems to detect the failure due to internal instability. The 100 m long and 6 m high dyke was made of an internal sandy core with a 1.5 to 2.4 m thick external clay revetment, common structure of the dikes along the Dutch canals. The GeoDetect® solution was installed inside the dyke, under the revetment, to measure the strain of the embankment at 4 different locations (Fig 11). During the test, the global factor of safety of the dyke was decreased step by step by digging a trench at the downstream toe and by increasing the internal water pressure till failure by slippage of an area of the downstream face (Figure 12). The FOS based monitoring system worked perfectly as it was able to detect and to localize the instable zone inside the dyke body before the failure occurred (Artières, Koelewijn et al, 2009). Soil strains as small as 0.02 % were measured indicating the first steps of the failure more than two days before the dyke collapses (Fig. 13). Figure 11: Installation of the GeoDetect® S-BR strip at the top of the IJkdijk/Macrostability dyke. Figure 12: The experimental IJkdijk/Macrostability dyke before failure on the left and three different views during collapse. 8 of 11 PIANC MMX Congress Liverpool UK 2010 Figure 13: The experimental dyke after failure and the corresponding measurement of the strain several hours before failure. 6. VALIDATION ON REAL CANAL SITES Several real sites are now equipped with the GeoDetect® monitoring solution. On earthworks and geotechnical applications, reinforced vertical walls and embankments over cavities are monitored for more than 5 years to measure any change of soil strain. In canals, a test section of about 100 m long and 3 m high along the canal from the Marne to the Rhine owned by the Voies Navigables de France (VNF) in the eastern part of France was monitored since 2008 (Fig. 14). Another section is monitored along the Rhine canal near Kembs (F) : this dyke owned by EDF is about 3 m high is built with permeable aggregates. The measurements are on-going on all of these real sites. 9 of 11 PIANC MMX Congress Liverpool UK 2010 Figure 14: The four GeoDetect® strips installed on a section of the Marne to Rhine canal. 7. CONCLUSION The GeoDetect® solution is an innovation that combines the benefits of geosynthetics materials with the latest sensing and measurement technologies This customizable solution provides objective, highly precise, and timely in-situ performance information, allowing the designers and owners to understand system performance in addition to providing alerts for negative “geo-events” (subsidence) and other potentially deleterious events. The GeoDetect® monitoring solution has already proved on experimental dikes the early detection and localisation of both leakage and settlement. 8. REFERENCES Artières O., Bonelli S., Fabre J.P., Guidoux C., Radzicki K., Royet P., Vedrenne Ch. (2007). Active and passive defences against internal erosion. 7th ICOLD European Club Dam Symposium, Fresing, Germany. 17-19 September 2007. Artières O., A.R. Koelewijn A.R.,Fry J.J., Royet P (2009). Early detection and localization of a failure zone in a dyke with a fiber optics monitoring system. Presented at the ICOLD Conference, Brasilia. 20-24 May 2009. Beck Y.L., Cunat P., Johanson S., Donstater J., Aufleger M. and Goltz M. (2009), Use of fibre optics in leakage detection: a review, European Working Group in Internal Erosion, St Petersburg (Russia), Cunat P., Beck Y.L., Fry J.J., Courivaud J.R., Fabre J.P., Faure Y.H. and Radzicki K. (2009a), Leakage detection based on temperature measurement with fibre optic : methods and results, HYDRO2009, Lyon Cunat P., Beck Y.L., Fry J.J., Courivaud J.R., Fabre J.P. and Faure Y.H. (2009b), Surveillance of dyke ageing, 2nd International conference on Long Term Behaviour of Dams, Graz (Austria) IJkdijk, (2009). 4 pages. Available on www.ijkdijk.eu. Radzicki K., Bonelli S., Beck Y.L. and Cunat P. (2009), Leakage and erosion processes identification by temperature measurements, in upstream part of earth hydraulic works using the impulse response function analysis method, European Working Group in Internal Erosion, St Petersburg (Russia) 10 of 11 PIANC MMX Congress Liverpool UK 2010 9. ADRESSES OF AUTHORS Olivier Artières, global technology manager GeoDetect® solution TenCate Geosynthetics, 9 rue Marcel Paul, 95873 Bezons Cedex, France Tel +33 1 34 23 53 74 / +33 6 8266 9570 Fax +33 1 34 23 53 64 [email protected] Mathieu Galiana, Engineer Centre d'Etudes Techniques Maritimes et Fluviales 2, Boulevard Gambetta – BP 60 039 - 60 321 Compiègne Cedex – France Tel : +33.344.92.60.69 Fax : +33.344.92.60.75 [email protected] Paul Royet, Senior Engineer Cemagref, CS 40061, 13182 Aix-en-Provence Cedex 5, France Tel +33 442 66 99 35 Fax +33 442 66 88 65 [email protected] Jean-Jacques Fry, Senior expert EDF / CIH - Savoie Technolac – F-73373 Le Bourget du Lac Cedex – France Tel : +33 4 7960 6178 Fax : +33 6 7070 1637 [email protected], [email protected] Jean-Robert Courivaud, Engineer specialist EDF / CIH - Savoie Technolac – F-73373 Le Bourget du Lac Cedex – France Tel : +33 4 7960 6178 Fax : +33 6 7070 1637 [email protected], [email protected] Yves-Laurent BECK, Project manager EDF / DTG - 21 avenue de l'Europe - BP41 – F-38040 Grenoble Tel : +33 4 7620 2406 [email protected] Pierre Cunat, PhD student EDF / DTG - 21 avenue de l'Europe - BP41 – F-38040 Grenoble Tel : +33 4 7620 2406 [email protected] Yves-Henri Faure, Professor UJF / LTHE - B.P. 53 – F-38041 Grenoble Cedex 9 – France Tel : +334 7663 5131 [email protected] Cyril Guidoux, Fiber optics development manager, GeoPhyconsult GeophyConsult - 12, allée du Lac de Garde - BP 231 - F 73374 Le Bourget-du-Lac cedex Tél. : +33 (0)4 79 62 55 40 [email protected] 11 of 11 PIANC MMX Congress Liverpool UK 2010 RESTORING NAVIGABILITY TO THE UPPER RHÔNE THE CNR’S COMMITMENT TO DEVELOPING TOURIST ACTIVITIES by 1 Gilles Delphin Poulat – Jacques Philippe Bois 2 1. INTRODUCTION th 1.1 The navigable Upper Rhone in the 19 century th In the 19 century, the navigable section of the Upper Rhone was a busy focal point for river traffic. As early as 1837, the 3 boats of the “Compagnie du Haut-Rhône” provided a regular service between Lyon and Lake Bourget via the Savières canal. At the beginning of the Second Empire (1860), seven companies competed with each other on the line between Lyon and Aix les Bains. River traffic steadied at around 10,000 passengers before undergoing a rapid decline in operations and traffic after 1900. Following the construction and commissioning in the eighties of hydropower schemes by the Compagnie Nationale du Rhône, navigation by pleasure boats on the Upper Rhone remained feasible, but the dams had to be crossed by transfer conveyors. 1.2 Wide-ranging demand and support There has been a continued boom in pleasure boating for the last three decades. Restoring navigability to the Upper Rhone is a major challenge for developing tourist and economic activities in the Valley and satisfies a threefold demand: - From elected representatives, who have supported this project since the 90s. Professionals in the tourist and leisure business, so they can develop their activities. The people of the Rhone Valley who wish to renew their links with the river and its banks for leisure purposes and who enjoy relaxing in a remarkable environment. Restoring navigability to the Upper Rhone is a decisive component that complements the existing tourist facilities (11 leisure centres, 5 passenger transport companies, 11 marinas and mooring points, 25 camp sites and 30 hotels). The project consists in extending the channel from Seyssel to Brégnier-Cordon along more than 50 km, resulting in a considerable extension of sailing time and providing access to more of the valley’s cultural and natural heritage. The bike track from Lake Geneva completes the river tourism activities and is part of an approach based on green tourism and sustainable development. 1.3 Restoring navigability on the Upper Rhone: a project carried out in the framework of CNR’s Missions in the General Interest After being validated on 20 July 2004 by its Supervisory Board and its governing authorities, the Ministry of the Economy, Finance and Industry and the Ministry of Public Works and Transport, the st Compagnie Nationale du Rhône launched its 1 plan of Missions in the General Interest (MGI) aimed at optimising hydropower production, developing river traffic, protecting the environment and reinforcing local ties. 1 2 Compagnie Nationale du Rhône, France, [email protected] Compagnie Nationale du Rhône, France, [email protected] 1 of 19 PIANC MMX Congress Liverpool UK 2010 Appended to its concession specifications, the MGI embody the company’s commitment to sustainable development. This voluntarist approach goes beyond the perimeter of its traditional community missions and requires that it implements a plan every five years up to the renewal of its concession in 2023. CNR devoted €125 million to its first plan and has programmed €160 million for its second. Restoring navigability to the Upper Rhone, written into the navigation and waterway development section of the plan, is one of a number of actions in the MGI whose goal is to achieve global and lasting territorial development. In addition, this project is part of the Rhone Plan supervised by the Prefect with responsibility for the watershed. One of its major challenges is to strengthen the bond between the population of the Rhone Valley and its river. Restoring navigability to the Upper Rhone will contribute to the valley’s economic development by boosting river tourism. The growth of economic activity will benefit business in the tourism and leisure sectors, local authorities and, as a result, generate employment. One method of assessing the socioeconomic value of projects intended to develop navigable waterways for tourist purposes is to situate the project in the far broader tourist development programme of which it is only one component. The success of the investments is due to the synergy generated between the different types of actors involved, which range from the organisation running the river infrastructure, the local authorities and the businesses whose activities are based on the river. This project spurs indirect effects (external to the project and related to the number of persons using the river) and spin-off effects (recorded due to the expenses allocated). The factors having an impact on the number of tourists to the site are in particular: - Cultural, environmental and gastronomic heritage. The other tourist facilities and the volume of hotel accommodation. Proximity with other tourist areas. The type of services provided to users. The quality of the water and its surroundings. Local dynamism. Regarding the equipment for the locks of Chautagne and Belley, the local context is positive due to the reasons mentioned below: There has been a resurgence in the popularity of pleasure boating on the Rhone over the past 30 years: downstream of Lyon 21,474 lockages were recorded on the river in 2005 versus 7,160 in 1972. On the Upper Rhone, pleasure boating now composes the major part of river usage: 2,500 lockages per year were recorded at Savières lock and more than 150,000 passengers a year were transported on Lake Bourget, Savières canal and the Rhone. The potential of the Upper Rhone for river tourism is considerable as it is close to large cities (Lyon, Geneva, Grenoble, Chambery, Aix-les-Bains, Annecy), which is a major advantage. The Upper Rhone has always been an important communication route for the valley. Although the river already draws a large number of tourists, new sources of tourism remain to be exploited. Packages that combine mountain and nautical leisure activities attract tourists from HauteSavoie (38 million every year of which 14 million come during the summer) to benefit from what the valley has to offer. The tourists and inhabitants of nearby urban areas and Switzerland also provide a high growth potential. 2 of 19 PIANC MMX Congress Liverpool UK 2010 2. RESTORING NAVIGABILITY TO THE UPPER RHONE – THE CROSSINGS OF CHAUTAGNE AND BELLEY 2.1 General presentation of the new developments (Figure 1) By constructing these two new crossings, the Compagnie Nationale du Rhône is extending navigable distance along 50 km of river by providing access for pleasure craft to the two adjacent bays of Belley and Chautagne via locks separated by an intermediate basin and by placing navigation signalling in the navigable channel. Figure 1 : Restoration of navigability on the Upper Rhone - Diagram Each crossing consists in constructing two Freycinet gauge locks (40 m long and 5.25 m wide) on the left bank of the existing hydropower plants at Belley and Chautagne ( Figures 2 and 3) . 3 of 19 PIANC MMX Congress Liverpool UK 2010 Figure 2 : Crossing the head at Belley The project in figures Start-up of works: start-up in 2008 at Chautagne and in mid- 2008 at Belley Provisional lead-time: 24 months for Chautagne and 20 months for Belley Commissioning: mid 2010 Cost: €35 million ex. tax financed wholly by CNR Figure 3 : Civil engineering works 4 of 19 PIANC MMX Congress Liverpool UK 2010 2.2 General design of the operation The locks of the new crossing will be built to Freycinet gauge specifications. This leads to designing the useful dimensions of the locks to permit boats to cross a height of 38.50 m: length: 40 m; width: 5.25 m; draught: 3 m. The heads of the development schemes are the following: - Chautagne: approximately 16 m with maximum of 17.30 m. Belley: approximately 17 m with a maximum of 18.20 m. The magnitude of the head and the choice of having the locks operated by the pleasure boat crews themselves (semi-automatic mode) led to staggering each crossing by the construction of two locks. Separating the two locks by a huge intermediate basin leads to better integration in the site, and attenuating variations in its water level following the staggered operation of the locks. 2.3 Physical models A physical model of the prevailing currents at Chautagne was built in 2005, and has permitted adjusting the project. The objectives of this model were to: - Study the currents up and downstream of the locks. Study the conditions of entering the locks for the different modes of hydropower plant operation (flow rate in the headrace canal, turbine operation, trips). Modify, if necessary, the solution proposed by the project for the installation of the locks. th The model had a fixed bed and reproduced the site’s topography to a scale of 1/60 . The study included several test phases that permitted optimising the installation of the locks and the access conditions. Modifications to the installation of the upstream lock and the geometry of the lock lay-bys up and downstream were performed to remedy the problems that arose: For the upstream lock - The angle between the centre of the headrace canal and the centre of the upstream lock had to be reduced from 30 to 25. The lock had to be moved forty metres downstream. The upstream lay-by had to be enlarged. For the downstream lock - The opening of the downstream lay-by had to be narrowed. The shape of the dike of the downstream lay-by had to be modified. A physical model of the filling and emptying system of the locks (up and downstream) was built at CNR’s laboratory at Gerland. This was done to validate and calibrate the lock filling and emptying systems, by bypassing the lock heads for the downstream lock of Chautagne and by culverts for the other locks. The aim of the physical model was to validate and optimise the filling and emptying system dimensioned beforehand in the preliminary studies by using the numerical model, Flowmaster. Initially, the study consisted in optimising the geometry of the ports and the energy dissipation chamber in order to: - Minimise eddies in the chamber and the stresses exerted on the boats. 5 of 19 PIANC MMX Congress Liverpool UK 2010 - Minimise the eddies downstream of the chamber when emptying. - Avoid air and debris being drawn into the ports. - Maintain reasonable manoeuvring time (less than 10 minutes). Afterwards, the model was used to define the sluicing calculations required to fulfil the objectives set for normal operation and for abnormal operating conditions (breakdown of a filling or emptying sluice). 2.4 Risk study relating to the crossing A risk study was started at the beginning of 2005. It permitted identifying the risks relating to crossing the structures and validating the design of the latter. These characteristics led to implementing major safety measures to ensure that pleasure boat crews benefit from secure and adapted operation in an isolated environment. As for the ergonomic design study it permitted establishing: - Firstly, that crossing a high head did not have a significant impact on crew behaviour once the lashing, controls and safety facilities are installed. Secondly, that the semi-automatic mode can lead to “improper” utilisation by crews, the safety measures determined by the risk analysis make it possible to prevent this type of inappropriate situations. 2.5 The architectural part of the project is intended to ensure harmony with the nearby hydropower plant. Given the magnitude of the differences in level of the downstream locks, the project plans for the creation of planted terraces of about 3 metres in height, supported by scale type flexible concrete retaining walls and Reinforced Earth (Figure 4). The vegetation growing on these terraces changes with height: composed of covering plants and grasses on the first terrace, it is composed of shrubs and small trees on the last one. The entrance to the upstream locks from the lower bay is marked by two vertical masses that signal the passage. The morphology of the downstream locks is the same. As they are relatively low, only two terraces have been planned. Figure 4 : The architectural part : Reinforced Earth 6 of 19 PIANC MMX Congress Liverpool UK 2010 2.6 Civil engineering works (Figure 5) The design is classical with a U-shaped chamber and filling and emptying ensured by culverts supplied by ports built into the lock walls for three of the four locks planned for these crossings. For the fourth lock downstream of the Chautagne crossing, the problem of maintaining an open excavation dry has led to the construction of a cofferdam structure. The structure of the locks heads will be built of reinforced concrete inside the cofferdam structure that will be used as the external formwork. For the chamber, the confinement curtain will be used as the definitive lock wall. For all the structures, the upstream head houses the upstream bulkhead, the upstream gate, the filling shaft, the filling sluices and their bulkheads. The downstream head accommodates the downstream gate and its protection beam, the emptying sluices and their bulkheads. Figure 5 : Civil engineering works 2.7 General electric design The gates and the filling and emptying sluices are operated by electrical actuator. The choice of these cylinders has several advantages: - They permit avoiding hydraulic piping, which if broken could lead to a risk for the environment (oil leaks). They save space (bulky hydraulic power units are avoided). However, measures must be taken to ensure: - The acceleration and deceleration phases by variable speed regulators. Braking the filling and empty sluices if rapid closing is necessary. The 2 locks are considered as independent with regard to their operation. Therefore they are supplied separately by the nearby hydropower plant. What is more, the power supply is backed up by the emergency generating unit at the plant. 7 of 19 PIANC MMX Congress Liverpool UK 2010 2.8 General design of automation Each lock is equipped with an autonomous automation system controlled by a PLC. This automation system allows 3 modes of use depending on the choice of CNR’s Operations teams: - - A self-servicemode: it is the skipper of the pleasure craft who starts the lockage. If there is a fault or the skipper makes a deliberate call, the local Operations teams are summoned automatically. A semi-automatic mode by the operator: the CNR operator controls the lockage via a set of cycle controls (upstream cycle / downstream cycle). A maintenance mode: the CNR operator controls the different lock mechanisms via a control box located next to the mechanisms. In self-service mode, skippers activate pull knobs that allow: - Requesting a lockage and preparing the lock by operating the pull knobs located on the upstream and downstream pontoons. Starting lockage by using the pull knobs located in the chamber. The mechanisms are driven and controlled by the PLC. Stopping lockage in case of danger. In this case the PLC generates an automatic call to the Operations teams. Entry and exit authorisations are managed by navigation traffic lights. The operating cycles prohibit any dangerous operation by checking the positions of the mechanisms, the levels of the basins and by detecting boats in the gate opening and closing areas. 2.9 Vocal communication system The originality of the solution implemented in this project stems from linking the automation system on the one hand and the interphone, telephone and spoken alarms of the operating system on the other. The links between these systems are made possible by an optical fibre telecom system that links all CNR’s development schemes. Audio interface equipment is installed in each lock. This system permits vocal communications between the skipper and the operator on standby duty and the diffusion of messages from the automated system to the skippers or the standby operator: - - - - For every lockage step and every action performed by the skipper on the pull knobs, prerecorded audio messages in 3 languages are broadcast on loudspeakers installed on the edge of the chambers and on the upstream and downstream pontoons. These messages, of type “please check that all the boats are lashed correctly”, “your request has been taken into account, the lock is under preparation”, “waiting time: 10mn”, etc., inform the skipper of the situation, the wait, the means of action, the precautions to be taken and thus contribute to safety. An interphone is placed on each of the 2 mooring bollards in the chamber. This allows the skipper to communicate with the standby operator in case of need, wherever the latter happens to be (at a production site during working hours, at home outside working hours, or accessible via a mobile phone). Furthermore, the Operator can be called directly by the automated system which indicates which lock is subject to malfunctioning. The lock automation system then interfaces with the spoken alarms from Operations. Lastly, it is possible for the operator to call a lock and give instructions that are broadcast directly by the loudspeakers. 8 of 19 PIANC MMX Congress Liverpool UK 2010 Skipper Audio guide messages Interphone-telephone communication PLC Standby operator Vocal alarms 3. DETAILED DESCRIPTION OF THE CROSSING STRUCTURES OF CHAUTAGNE AND BELLEY 3.1 The locks (Figure 6) The upstream lock is located immediately next to the upstream dike in order to reduce the rockfill th required to link it with the latter. This detail was defined by using a physical model to 1/60 in order to facilitate the entry of the boats into the locks. The access navigation channel exits upstream of the plant. The downstream lock is located fairly close to the downstream bank, to reduce the excavation necessary for the downstream channel and maintain the surface area required to create the intermediate basin without widening it excessively. A catwalk with a footbridge is planned on the left bank on either side of the crossings. It permits reaching the locks and is where the control terminal is located. A crossing on the downstream heads of the locks permits restoring communications between the hydropower plant and the service roads. The geometrical dimensions of the main civil engineering elements and items of mechanical equipment are: Upstream lock CHAUTAGNE BELLEY Head (en m) (CE) / (MECH) Upstream gate 5.90 / 5,40 6.45 / 5.95 Lock chamber walls 14.75 14.75 Downstream gate 15.25 / 14.75 15.25 / 14.75 Downstream lock CHAUTAGNE BELLEY Upstream gate 5.05 / 4.35 5.05 / 4.55 Lock chamber walls 11.60 12,50 Downstream gate 12.10 / 11.40 13.00 / 12.50 9 of 19 PIANC MMX Congress Liverpool UK 2010 Figure 6 : The lock walls U-shaped chamber 3.2 The intermediate pool The materials necessary are taken from the rock excavated from the intermediate basin and the downstream channel. Given the materials required and the need to reduce level fluctuations when emptying the upstream lock and filling the downstream lock, a basin with a surface area of about 3 hectares has been planned. The height of the water in the intermediate basin is close to the natural terrain while the fluctuation is + / - 25 cm. Besides its role as the intermediate step in crossing the head, the decision was taken to give an ecological and landscaping dimension to the intermediate basin of Chautagne while the basin of Belley has been given a tourist and landscaping dimension. Furthermore, only the control terminals are installed in the basin. The basin crossed by the boats from one lock to the other provides a manoeuvring area 30 metres wide and 350 m long. It is 3 metres deep. The rest of the Chautagne pool is reserved for aquatic plants such as reeds and its depth varies from twenty centimetres to one metre. 10 of 19 PIANC MMX Congress Liverpool UK 2010 At Belley, the part outside the manoeuvring area will be devoted to aquatic plants such as reeds and have a depth of about one metre. The pool is watertight, sealed with natural materials or silt-clay (Chautagne) or molasse (Belley). Daily losses are estimated at 1,000 m³, meaning a daily variation of the water level in the pool of about 3 cm. A regulation system is planned on the upstream lock to supply the pool in the absence of lockages. A regulation system is planned for the intermediate pool, next to the downstream lock, to limit fluctuations of the water level when emptying the upstream lock. Figure 7 : The intermediate pool The upstream lock, aval vue 3.3 The mechanical equipment (Figure 8) The mechanical equipment comprises: - Gates, their fixed parts and their operating devices for the locks of Belley and Chautagne. Cofferdam elements for a chamber, and the fixed parts for the grooves, filling and emptying sluices and their operating devices. Water intake grids, culvert bulkheads, floating bollards and their fixed parts. Stop log grooves will be installed upstream of the upstream gates and downstream of the downstream gates on all the locks. The downstream lock at Chautagne also has stop log grooves downstream of the upstream gate and upstream of the downstream gate. The gates are of one-leaf type composed of a plate upstream with horizontal reinforcing beams with vertical stiffeners and wind bracing. They are supported by a spherical pivot bearing and held in place by a heel post. Protection is provided by IRON WOOD fitted up and downstream of the gate to protect it from impacts. Each lock has a filling unit equipped with a grid in the upstream apron, followed by 2 culverts bypassing the gate and each equipped with a penstock. Each penstock can be closed by a bulkhead. Emptying in each lock is carried out via an opening in the lock chamber wall, at the downstream head, at the end of the two culverts bypassing the gate. Both the culverts will be equipped with a sluice gate. The chambers of each lock will be equipped with 6 mooring spaces with floating bollards. Total manoeuvring time (opening and closing) takes about 1 minute 30 seconds. 11 of 19 PIANC MMX Congress Liverpool UK 2010 The total duration of the cycle is about 25 minutes. The equipment is fitted with a heating system. A pneumatic system with automatic and manual controls can be fitted to the upstream gate to clean the active parts in order to clear floating wood and sludge from the bottoms of the housings. Figure 8 : The mechanical equipment 3.4 Automation and PLCs The automation systems are installed in the technical building. The PLC permits autonomous control of the lock (self-service mode), or from the operator control point or by remote controlled operation (operator mode). It ensures the following functions: - Management of operating modes. Cycle management (upstream, downstream, fallback). The operation of the main lock components (gates, sluices). The operator also has a display that they can use to view statuses, faults and a record of the installation. 3.5 Traffic lights and control systems There are 4 panels of lights that authorise boats to enter and exit the lock chambers: Pull knob control systems dedicated to the lockage cycles are available to the lock users and are located on the pontoons (Figure 9) up and downstream and in the lock chamber. In addition, there is an operator control point located in the middle of the chamber (at the edge of the chamber wall) and a remote operation control also dedicated to lockage cycles. 12 of 19 PIANC MMX Congress Liverpool UK 2010 Figure 9 : The pontoons 3.6 Pull knob controls 4 control stations (Figure 10) are available to lock users: The control stations are located at the up and downstream entrances of the lock. Two control stations, located at opposite corners of the chamber, inform the central automation system that all the users are moored and therefore ready for lockage. Each control station takes the form of a rod, a green metal control, guided by steps and accessible to the user whatever the water level in the basin concerned. Each lock chamber control station is also equipped with a second control rod located next to the green one. This second rod is red and is used to stop the cycle in progress immediately. Figure 10 : Control station 13 of 19 PIANC MMX Congress Liverpool UK 2010 4. CARRYING OUT THE WORKS 4.1 The administrative procedure prior to the operation The structures crossing the heads of Belley and Chautagne fall within the framework of CNR's concession as operator of the Rhone. The projects are submitted in the form of specifications to an administrative procedure, a concession structure modification file and an impact study. The construction of the structures and the works proposed in the framework of this operation were preceded by a public inquiry. Authorisation was issued by Inter-Prefectural decree on 18 January 2008. 4.2 Distribution of the works in technical lots The operation comprises three main contracts: "civil engineering", "Hydromechanics" and "ElectricityAutomation". Additional contracts were signed with different companies during the works, in particular for: - The floating pontoons permitting mooring and starting the automated operation. Landscaping. For security reasons, the general programme planned, wherever possible, for the separation of the civil engineering lot and the electricity-automation lot. 4.3 Civil engineering (Figure 11) The services planned in the "Civil engineering – Earthmoving" contract were distributed between 4 construction trades: - Earthmoving. Reinforced concrete civil engineering structures. Sheet pile structures. Reinforced earth structures. The services were performed and coordinated by a consortium. The lead company, in charge of the "Reinforced Concrete Civil Engineering Structures", was responsible to the owner for this coordination. The technical coordination between the different contracts was ensured by the design and construction manager. 14 of 19 PIANC MMX Congress Liverpool UK 2010 Figure 11 : Civil engineering 4.4 General phasing and programming of the works (Figure 12) After a study and site preparation phase, the civil engineering works were programmed by taking care to avoid overlaps between identical tasks which would lead to increased costs, and by optimising earthmoving. The construction phase on the site was launched by a work order in May 2008. Works on the two sites were staggered by 6 months in order to optimise the human and material resources of the companies involved. 2008 2009 2010 Mar AprMai JunJuilAugSeptOct NovDec JanFeb Mar AprMai JunJuilAugSeptOct NovDec JanFeb Mar AprMai Jun CROSSING THE HEAD AT CHAUTAGNE Civil engineering works Preparing the works Upstream lock Downstream lock Earthmoving Opening the upstream dike Finitions Hydromecanical lot Studies and supplies material Assembly Electricity Automation Commissioning CROSSING THE HEAD AT BELLEY Civil engineering works Preparing the works Upstream lock Downstream lock Earthmoving Finitions Hydromecanical lot Electricity Automation Commissioning Figure 12 : General program 15 of 19 PIANC MMX Congress Liverpool UK 2010 Phase 1: preparing the works (2 months) This phase includes: - Starting the detailed design. Diverting utilities. Preparing platforms next to the future structures. Building access roads. Phase 2: excavating the locks (1.5 months) This phase included: - The full excavation of the upstream lock. Preliminary earthworks on the intermediate basin and the downstream lock. The drawdown of the groundwater next to the excavation of the upstream lock. Start-up of placing the rockfill for the future dike. Phase 3: start of the civil engineering concrete and sheet pile works (7 months) This phase included: - The construction of the apron of the upstream lock and the lock chamber walls. Pile driving the sheet pile and the start of excavation next to the downstream lock. Phase 4: The end of the civil engineering works of the locks and sealing the pool and channels (3 months) This phase included: - The end of building the chamber wall of the downstream lock with backfilling as works progressed. Sealing the intermediate pool with silt-clay. The end of the downstream channel. The end of building the future rockfill dike. Phase 5: Opening the upstream dike and filling the structures with water (2.5 months) - This phase included: - Earthmoving to open the dike, in water and upstream of the curtain wall. Depositing a gravel protection layer at the bottom. Earthmoving to open the dike, in water and downstream of the curtain wall. The application, in water, of sealing coats and a layer of gravel to protect them in the upstream channel. Cutting down the curtain walls. A specific phase concerning the link with the hydropower plant canal was carried out at the Chautagne development scheme in spring 2009. The high level of permeability within the dike and the importance of its tightness upstream to ensure its stability led to an in-depth study of the different methods that could be implemented. The solution chosen was to carry out earthworks to open it under dry conditions, after having installed a cofferdam at the entrance of the canal to lower the water level in the pool next to the opening below a height of 247.00. The lead-time for this operation was three weeks with the participation of several teams. Since it was necessary to shutdown the hydropower plant, the works were programmed during a period with low flow rates, thus advantage was taken of the reduction in the level of the pool to carry out maintenance works above water upstream of the plant, and verify the tightness of the canal banks. 16 of 19 PIANC MMX Congress Liverpool UK 2010 Figure 13 : Sealing works The phasing decided was as follows: - - 4.5 Shutting down the Chautagne hydropower plant. Pre-lowering phase: Lowering the pool from its normal height (252) to (250.50) by regulating the hydropower plant dam in operation. The construction of a gravel cofferdam at the head of the headrace canal at kilometre 145.750, adjusted to a crest height of 10 metres (253.5 NGF). The gravel was taken from stocks available on the left and right banks. Lowering the water height next to the site to (247.00), and building a riprap protection barrier at the outlet of the future filling channel. Massive earthworks to build the dike. Earthworks to form the junction. Sealing works (Figure 13). Installation of anti-wave action protection. Re-filling the hydropower plant canal with water via the channel on the crest of the cofferdam and filling the upstream channel. Removal of the gravel cofferdam. Starting-up the hydropower plant. The mechanical electricity and automation works After carrying out the detailed design and the completion of the lock platforms by the civil engineering team, the hydromechanical works were started at the Chautagne site in March 2009 and at Belley in October 2009. The first works concerned the installation of the different fixed parts sealed in the concrete and intended to house the various mechanical devices, in particular the grooves for the stop logs. The stop logs were then installed in the upstream and downstream grooves of the crossing thus making it possible to assemble the moving parts under dry conditions. The gates were brought to the site by abnormal load transport and installed by mobile crane for adjustment. The largest gate downstream of the upstream lock weighed nearly 40 tons. The assembly was completed with the sluices of the filling and emptying system. 17 of 19 PIANC MMX Congress Liverpool UK 2010 Tests without and then with filled with water were carried out before the technical acceptance of the structures. The hydromechanical lot took 8 months to complete. The works lasted from March to October 2009 on the Chautagne crossing and from October 2009 to May 2010 on the Belley crossing. 4.6 The electricity, automation, and instrumentation and control part The electricity, automation, and instrumentation and control part was carried out in parallel. The technical buildings were equipped with the different cabinets of related systems, with cables linked to the mechanical devices and control terminals. The system was then connected to the hydropower plant power supply network. All the PLC functions were tested under dummy dry and real conditions with water. The lockage cycles were performed before technical acceptance. The test period lasted 4 months. The total duration of the works was 9 months. The works were performed from March to November 2009 on the Chautagne crossing and from October 2009 to June 2010 on the Belley crossing. 5. SAFETY AND HEALTH PROTECTION The operation was subject to law 93-1418 of 31 December 1993 and the rulings applying directives supplementing the labour code. The works were then placed under the clauses of Ruling 94 and classified as belonging to a category 1 operation subject to: - The Preliminary Declaration (R 238-1). The obligation of a General Coordination Plan. Setting up an inter-enterprise college in charge of safety, health and working conditions. A safety and health protection coordinator monitored this category 1 operation in accordance with Ruling 94 C. GAIDATZIS of the ACCORD Consultancy. 6. SAFETY AND RESPECT FOR THE ENVIRONMENT A site that respects the environment and the neighbouring population During the works, the Compagnie Nationale du Rhône took pains to constantly minimise the impact of its construction sites on the environment. From the outset of the design phase, the project incorporated measures to limit the occupation of sensitive sites. This is in conformity with the aims of the Natura 2000 management of "Lac du BourgetChautagne-Rhône" related to the preservation of habitats and birds. 18 of 19 PIANC MMX Congress Liverpool UK 2010 In particular, CNR acted on the quality of the water taking care during the construction phases to do its utmost not to degrade the quality of the water during pumping and earthworks, by discharging clear water for the most part and returning used water to irrigate the surrounding woods during the summer pumping phase. The construction was adapted to preserve protected species, especially the beavers. The planting of reed beds on the intermediate pools of Chautagne and Belley facilitated the reproduction of species and guarantees the future of biodiversity in the Rhone Valley. CNR ensured that nuisance caused to the neighbouring population was minimised by making use of the construction site roads as much as possible. Regarding operations, CNR is concerned with preserving the natural environment and plans the maintenance of the lower banks by relying on plant engineering, and restricting boat speed to reduce nuisance caused to birds. An Upper Rhone River Traffic Observatory set up at the initiative of CNR, brings together local actors and ensures monitoring of river traffic, the management of maintenance works on the channel and marina sites, the study of sensitive ecological sites and the diffusion of information to pleasure boat crews by way of a guide to good conduct. 7. SITE ORGANISATION The Compagnie Nationale du Rhône was both the owner and the design and construction manager of this operation. The architectural project was entrusted to the cabinet 2BR of Lyon. The contracts for the works were awarded to the following companies following tenders: - Civil engineering lot: the consortium comprising LEON GROSSE – FAMY FONTAINE. The hydromechanical lot: ASTEN PETRISSAN. The electricity and automation lot: SOTEB. 19 of 19 PIANC MMX Congress Liverpool UK 2010 IMPROVE THE QUALITY OF ENVIRONMENTAL STUDIES FOR INLAND WATERWAYS PROJECTS by by B. de Bruyn1 and Ph. Rochette2 ABSTRACT In France, for most major projects environmental studies are mandatory. Such studies are needed when the project affects aquatic environment such as waterway (building or refurbishing of navigation weir, new waterway alignment, etc.) These studies generally provide a sound overview of the environment. Unfortunately, these studies are too focused on a protected species. It would be helpful to include an assessment of ecological services for human well-being. Usually, these studies do not provide any information on how ecosystems work. Therefore they are not adapted to stakeholders needs. Thus the project design does not take accurate account of the environment and prevent a good environmental fitting. To overcome this shortfall, an identification of environmental waterway functionalities has been completed. Improving the knowledge, the content, the method and the presentation of studies may lead to better consideration of environmental issues in navigation projects. It should help the stakeholders to manage their projects more sustainably. The inventory of environmental functionalities has been focused on the area in and around inland waterways estate. Adjacent ecosystems should not be severed from waterway ecosystem as they are interconnected. The studied area has been divided into several zones, navigable channel, submerged and emerged parts of the banks, towpath and hydraulic dependencies has shown in figure 1. Figure 1: studies areas (adapted from STCPMVN, 1982) The ecological role of each of the specific zones has been previously discussed and described (CETMEF, 2007). The dynamic aspect of ecosystems as well as the evolution of morphology and ecological functionalities are included in the description of each zone. The proposed approach for waterways environmental studies targets a wide audience. It should be helpful to most waterway operators who are generally not experts in aquatic ecology. Improving methods of organisation, presentation and operation of inventory data could help stakeholders to certain the environmental functionalities. It would also be useful to ensure more efficient projects. At the beginning of the XXI st century “The Millennium Ecosystem Assessment” was initiated by the United-Nation's General-Secretary. The aim was to identify the ecosystems' contributions to human well-being by distinguishing between the provisioning services, the regulating services, the cultural services and the support services. This assessment propose to look at these services from the waterways point of view. Identifying functionalities and services provided by waterways ecosystems should improve stakeholders' understanding of environmental studies. It could lead to a better integration of manmade projects in the environment. 1 2 CETMEF, France, [email protected] CETMEF, France, [email protected] 1 of 9 PIANC MMX Congress Liverpool UK 2010 1. INTRODUCTION In France most of the environmental studies performed in the context of development of inland waterways are mainly focused on the emblematic and charismatic species. These studies provide, at best, detailed biocenotic inventories. Usually, these inventories do not provide enough information on dynamic aspects of the environments under study. Without these informations, waterways stakeholders have difficulties making appropriate choices during waterway development. To improve environmental studies, we propose three complementary approaches. First a questionnaire has been developed on the inland waterways to easily establish an environmental diagnosis of waterways. Secondly a selection of measurable factors is still under development. These indicators would be used for a finer evaluation than the previous approach. It provides usable forward keys to improve the environment. Finally, the "Millennium ecosystem assessment" will be used as a template for environmental study. This will enable the study to put an emphasis on the “services provided” by ecosystems to the Human well being. We believe that this presentation will contribute to better managers' decisions concerning the environment. 2. ASSESSMENTS SCALES Rivers used for inland navigation are generally downstream of major watersheds. Indeed the characteristics of the downstream rivers (depth, width, water velocity, horizontal alignment etc.) are better suited for inland navigation. Therefore the area to be studied will need to be large. It is not easy to determine the relevant area to include in the study in order to understand the ecology associated with navigation structures. On one hand most systems and interactions change at different scales. On the other hand, the structures themselves are different and have different size. This is the case, for example weirs, embankments and dykes. The clear definition of an appropriate environmental boundary is difficult due to the above two aspects. Looking at the structure, four several scales seems to be relevant : – far superior to the structure; – superiors at structure; – of structure; – below the structure. Looking at the ecosystem, this concept can also be applied. The scales of space and time are correlated (Amoros C. & Petts G.E. 1993), and have to be taken into account. Thus, for hydro-system of navigable rivers it is possible to consider three levels (adapted from Wasson et al., 2000). By decreasing order: – the regional scale: this includes the regional network area or a navigable river watershed (because of its importance). At this scale, the suitable periods of phenomena are longer than hydrological cycle, such as the evolution of hydrogeomorphology and biological reservoir; – the linear scale which includes several levels: the large segment, the segment and the sequence. The evolution at this scale is affected by recurrent phenomena between the hydrological cycle and the daily cycles. Different cycle include evolution of riverbed, balanced living populations and several achievements biological functions (resting, feeding ,...); – the local scale: is concerned in small areas and short cycles that take place under a day such as ecological aspect of micro-habitats and feeding. The scales associated with water systems are not necessarily relevant for terrestrial ecosystems. 3. STUDIES AREAS The general method to improve the environment studies is independent of the kind of structure involved in navigation development. A reach, between two dams or navigation locks, can be an appropriate study zone for environmental assessment. This concept can be adjusted: when the reaches are very short it is wise to consider all 2 of 9 PIANC MMX Congress Liverpool UK 2010 reaches as only one entity. On the other hand when they are very long it may be better to consider them as several reaches, particularly when it has a singularity. The presence of a confluent for example can be considered as able to change the characteristics of the reach. The choice of the reference unit is guided by two factors: – as described above, scales are numerous and all can be relevant for the analysis; – reaches are particularly suited for management. This choice should not leave out other scales of study but it has been chosen because it is adapted to waterways managers habits. Five elements inside waterway have been selected. This selection has been guided by the similar characteristics and different functions of every units. Listed below are the five units selected: – navigation channel; – bank; – hydraulic dependency; – adjacent area; – structures for navigation. A diagram of the different elements chosen to be analysed are shown in figure 1 and 2. Figure 1: Extension of the functional units selected (cross section) 3 of 9 PIANC MMX Congress Liverpool UK 2010 Figure 2: Extension of the functional units selected (plan) 4. ASSESSMENT OF WATERWAYS ENVIRONMENT WITH AN EVALUATION GRID/TEMPLATE One method to characterise the environment of a waterways reach is the development of a rough expert system. The objectives of this method are: – an easy implementation (by managers themselves without any particular level of expertise); – a result that allows the characterisation of the waterways environment; – partial results that allow assessment of sections; – an educational perspective; – an operational under: diagnosis suggests ways to improve the quality of the waterway environment. This method is based on a list of preset questions. The questions are closed (answer by yes or no) to decrease subjectivity. Questions have been established so a low level of expertise is needed to answer. All answers can be provided by looking through literature and field survey. The availability of the data needed has been checked (CETMEF, 2007). The list of questions is the result of a three-step procedure. First of all a group of experts has identified parameters of importance to the characterisation (CETMEF, 2007). Secondly, the key issues have been identified as well as integrating elements of the main components of reach environment. All subjective references were then carefully removed from the list to retain only facts. For example, analysis such as SEQ Physics (Agence de l'eau, 1999) based on observer ratings were not included in the analysis. At this stage, the first draft of the expert system was field tested on the river Oise (El Bakali, 2008). This test revealed the vast amount of data available, unfortunately the use was difficult and a simplified approach was needed. In a third and final step, a list of closed questions are selected and a way of interpretation has been developed. 4 of 9 PIANC MMX Congress Liverpool UK 2010 The questions are grouped into five categories of items directly derived from the reaches elements mentioned above (cf. § 3.Studies areas): – generality of the reaches; – waterways and hydraulic annexes; – banks; – related areas; – structure for navigation. Each theme contains a variable number of questions (about 70 in total). In each question the impact is evaluated on the: – landscape; – biodiversity; – ecological functions. Thus the answer to each question provided a score for the three environment. The scores range from 0 (no effect) to 5 (positive effect). Different zones can be considered together to weight the overall environmental value on the waterway. They can provide a ranking for the several zone (waterway, banks, etc..) and for the reach itself. Questions themselves are designed with a didactic objective and have a concrete character. The questions have three aims: evaluate a reach environment characters, directly point to specific features of operating environment. The latter effect expected is to urge manager to lead toward solutions to improve situations if they are marginal or poor. For example, at the section "Banks" the question "Have you observed amphibians in the bank?" assesses some of the environmental characters of this reach. In this case a positive answer to this question does minute in terms of landscape but ecological functions and biodiversity are likely to be of good quality; thus these categories receive the maximum score of 5, they are top rated at 5. The result on this issue can be integrated at a higher level, taking into account other issues and weighted. The integrated results are intended to be analysed question by question. The results obtained allow comparison in landscape, in biodiversity and in ecological functions of the environmental characteristics of reaches. 5. ASSESS THE WATERWAYS ENVIRONMENT BY ESTIMATION OF INDICATORS A second approach has been to develop indicators which represent or integrate environmental aspects associated with the waterway. This approach had been derived from work out those used for hedges rating. Thus, series of indicators is proposed to contribute to the environmental diagnosis of the navigation infrastructure. The aim is to provide a means of estimating the potential contribution of a section (reach) to the local ecological wealth. These indicators do not require expertise resources in taxonomy. They link several aspects of the "system" waterway: the morphology of reach, the banks, the relative extent of various facies and contrast compared to the neighbourhood. The characteristics are chosen in order to be easily evaluated or measured. The data are also clearly identifiable in order to be collected by people non-specialist in environment, while characterizing the “ecological landscape”. The necessary data are from geometry or geography survey and are not the result as the previous method of a closed question. The implementation of indicators requires measurement on aerial photos and during field survey. This method is still under development. It completes the previous assessment by providing more quantifiable result. In this context, many indicators have been imagined to assess the various environmental components of a reach. These indicators can be sorted according several types: – Geometric indices of habitat; – Indices of banks; – Hedge plant indices; – Indices of occupations way; – Indices of annexes hydraulic; 5 of 9 PIANC MMX Congress Liverpool UK 2010 – – Connectivity indices; Pressure of navigation indices; Some indices have been developed in detail. Currently ten indices have been developed. Most of us are normalized to 1. This unit score describe an environment which is normally managed. 5.1 Occupancy banks index This index characterizes the ability of the riparian strip to ensure ecological functions. It draws on the Canadian Index of Quality Riparian (IQBR) which considers that a strip of land 15 m along the river provides important ecological functions (Saint-Jacques et al. 1998 ). Each kind of covered area is associated to a normed coefficient Ci (forest = 1; shrubs = 0.78, grass = 0.5, idle = 0.16, Culture = 0.03, uncover field = 0, infrastructure = 0, 03). The index value is calculated as a sum of surfaces (Pi) of a given type weighted by this coefficient. ∑ P i×C i occupancy banks index= i ∑ Pi i 5.2 Shorelines type index This index characterizes the nature of the interface between aquatic and terrestrial environments (Tales E. & Boet P., 2005; Wolter C. et al., 2003). Its principle is indented to the previous index calculated by weighting the length of the shoreline covered by a particular type and weighted by a coefficient characterizing the Ci (concrete lining= 0; pile = 0.14; riprap = 0.57, tree = 1). ∑ P i×C i shoreline type index= i ∑ Pi i 5.3 Hydraulic dependency index Quantifying the ecological function of hydraulic annexes on the mid waterway has not been possible. The role of hydraulic annexes is characterized by their surface relatively to the surface of the navigation channel. hydraulic dependency index= hydraulic dependency area reach chanel area 5.4 Tributary number index The ecological functions of tributaries flowing into the reach is important. To characterize these functions and compare with other reach, it is proposed to be assessed as follows: tributary number index= number of tributaries reachlength 5.5 Channel index The part of the river used as channel is strongly affected by navigation (current propulsion, depth, opacity) and is not the most convenient place for the development of aquatic beings/environments. To characterize this effect a channel index is proposed. It relates the length of the channel and banks. chenal index= 2×chenal length banks length 5.6 Straightness index The intend to use a water course for navigation has an effect on the layout plan of the navigation channel, it is moving towards an advantage of straightness. These morphological changes of rivers 6 of 9 PIANC MMX Congress Liverpool UK 2010 have effects on the community associated (normalization, erosion control, etc..). An index can evaluate this effect. straigthness index= straight line length reach length this index give also an indication on earth – water connectivity 5.7 Navigation index The navigation index allows to characterize the water surfaces where aquatic ecosystem may eventually taking place. It may be calculated as follow: navigation index= reach length×chenal width reach area 5.8 Shoal index The shoal is a rich aquatic environments (Madéore -Le Pichon C., 2006). This area reported to the total surface of the reach give the shoal index. shoal index= surface where the depthis less than 0.5m reacharea 5.9 Weir rate index The presence of weir is a problem for migratory fish species. To take this effect into account for a navigation reach, the following indicator is proposed. weir rateindex= weir number distance between reach and first dam or weir up stream of see 5.10 Navigation traffic index The traffic navigation had an incidence on the aquatic environment, particularly due to the wave and the drawdown which causes dewatering of shallow areas along the shoreline during vessel passage (Wolter et al., 2004). Thus an index was proposed as follow. navigation traffic index= navigationtraffic on the reach reach channel area These indices were applied at two reaches of the Oise-waterway, France (Renou, 2009). These reaches have some differences in their immediate environment. The first is located in peri-urban, highly industrialized and has overall embankment built and maintained. The second go through a forest, agricultural and sparsely populated environment, its banks are generally subject to a more spontaneous evolution. This test score index has yielded encouraging results. Thus, this trial showed that the indicators could be assessed using current aerial photographs, a GIS for use, summary field reconnaissance and literature easily available to the manager of the waterway. Such assessment required about one week for a reach; it may need less time for reaches well documented. This test has verified that the values of indicators were consistent with the overall perception "by an expert" of the environment of the reaches examined. Improvement steps must now be implemented. A data set large enough has to be collected to test the relevance of each indicator and to establish a catalogue of known cases to provide a database for comparison purposes. 6. STRUCTURE OF THE RESULTS : MILLENIUM ECOLOGICAL ASSESMENT A better identification of the functions (current) and changes that could occur by development works could be improved by a change of the modes of organization of data coming from environmental inventory. This proposal is primarily based on the approach promoted to conduct the "millennium ecosystem assessment" launched by the General-Secretary of the UN at the beginning of the millennium (Millennium Ecosystem Assessment, 2005). This inventory test and assessment attempts to identify the contributions of ecosystems to the well being of humanity. 7 of 9 PIANC MMX Congress Liverpool UK 2010 The many essential processes which contribute to the Human well-being are often called "services provided by ecosystems”. The millennium ecosystem assessment classifies them into four main categories of services: 1. Provisioning services: this is the goods production from ecosystem, goods such as genetic resources, food and fiber materials and freshwater. 2. Regulating services: what are the benefits of regulating ecosystem of major geothermodynamic cycles, geo-chemical and bio-demographic such as regulation of climate, water cycle and some diseases aetiology in human populations or species involved in the agro-pastoral economy. 3. Cultural services: these inputs are not material but deal with the spiritual enrichment, cognitive development, reflection, recreation and aesthetic experience, related to ecosystems. They contribute to social functioning and individual development. 4. Supporting services: they correspond to all the productions necessary to the functioning of all other services provided by ecosystems. They include the recycling of biomass, the production of atmospheric oxygen, training and retention of soil, nutrient cycling, the water cycle and supply habitat, population control species. The inventory of environmental studies should feed into identification of functions provided by medium examined. Register services provided to humanity by the ecosystem now seems like a way of a methodological progress to help stakeholders. Such a presentation should be into the focus of environmental waterways studies issue. It should also lead to a better understanding of ecological functions which support and develop opportunities for improving the environment as objectives. Reported at costs of implementing these should provide assistance to the decision to use by stakeholders. The inventory of services provided to humanity by the waterways began. It is accompanied by a thinking on the ecological functions puts at risks by development for navigation and the possible ways to improve them. 7. CONCLUSIONS AND PERSPECTIVES Additional tools to take into account the environmental aspects of waterways in a better way have been presented. A list of questions provide to waterway managers a diagnostic tool of navigation reaches. This tool is designed for a wide audience as possible. It should lead the manager to drive an environmental diagnosis by itself. It must bring the manager to a better understanding of the environment and functions that should be provided by the reaches he had to manage. Collecting indicators can improve this approach by providing a quantification of environments. These tools provide a feature as objective as possible. Because of its simplicity and transparency, they suggest in some cases means to improve the environmental reaches quality through the provided habitat sites. A proposed organization of environmental studies on developments works related to navigation thanks to environmental services provided by a waterway could complete this device. By identifying the services catered by the environment supported by waterways and providing stakeholders proposals for higher objectives it is possible to better quality of environmental studies. This raises decision making concerning these functions and probably would have, in the long term, an environment improving boosting effect. If tools exist, they are not yet completely operational. Developments are needed before to have a full range of tools. Including the questionnaire, they must be fully validated in the field, the indicator method should be supplemented and developed to have an higher level of integration. Also this method can be implemented on a sufficient number of reaches to have enough references. Finally methodological developments to enhance the effectiveness of environmental studies are still in development. They will allow to identified lacks in terms of knowledge on waterways ecosystem. 8 of 9 PIANC MMX Congress Liverpool UK 2010 8. REFERENCES Agences de l'eau (1999), Les outils d'évaluation de la qualité des cours d'eau (S.E.Q.) - Principes généraux. ISSN : 1161-0425 Amoros C. & Petts G.E. (1993), Hydrosystèmes fluviaux, Collection d'ecologie 24, Masson CETMEF, (2007) Evaluation environnementale des biefs de navigation, rapport intermédiaire de synthèse, 2007 – internal report El Bakkali M. (2008) Evaluation environnementale des biefs de navigation – Test d'une méthode d'approche, 2008, Travail de fin d'études – ENTPE. Madéore -Le Pichon C. (2006), Une approche “paysage aqautique” pour une meilleure connaissance du fonctionnement des écosystèmes fluviaux et l'amélioration de la conservation des peuplements de poissons, PhD Thesis Paris VI, Millennium Ecosystem Assessment (2005) Ecosystem and Human Well-being, MA, Island Press, Washington DC. Renou L. (2009). Evaluation environnementale des cours d'eau navigable, Test d'une méthode de diagnostic, 2009, Travail de fin d'études – ENTPE Saint-Jacques N. & Richard Y. (1998). Développement d’un indice de qualité de la bande riveraine: application à la rivière Chaudière et mise en relation avec l’intégrité biotique du milieu aquatique, pages 6.1 à 6.41, dans ministère de l’Environnement et de la Faune (éd.), Le bassin de la rivière Chaudière : l’état de l’écosystème aquatiques-1996. Direction des écosystèmes aquatiques, Québec, envirodoq n° EN980022. STCPMVN (1982), Notice STC n°82.1 Milieu aquatique et voies navigables – impact de la navigation intérieure sur les écosystèmes aquatiques. Tales E. & Boet P. (2005), Fonctionnalité écologique des berges et peuplements piscicoles - Effets des aménagements liés à la navigation, Etude CETMEF & CEMAGREF, 21p. Wolter C. & Arlinghaus R. (2003) Navigation impacts on freshwaterfish assemblages: the ecological relevance of swimming performance. Revue in fish biology and Fisheries 13: 63-89 Wolter C., Arlinghaus R., Sukhodolov A. & Engelhardt C., (2004), A model of navigation - induced currents in inland waterways and implications for juvenile fish displacement, Environnemental Management, Vol. 34, n°5, 12p. 9 of 9 PIANC MMX Congress Liverpool UK 2010 Etude du comportement de navires amarrés au passage d’un navire chenalant – Application à des grands porte-conteneurs par 1 P.F. Demenet et M. Le Bigot2, SOGREAH, France J.P. Guellec3 et L. Chéreau4, GPMH, France RESUME 1. INTRODUCTION Lorsqu’un navire chenalant approche d’un navire amarré, celui-ci est soumis à des efforts induits par les gradients de pression associés au champ d’écoulement. Ces efforts sont à l’origine de mouvements du navire amarré et d’efforts dans les amarres qui peuvent entrainer l’arrêt des opérations de chargement/déchargement et dans des cas extrêmes, la rupture des amarres. Cette interaction entre navires chenalant et amarré est une préoccupation du GPMH (Grand Port Maritime du Havre) dans l’ancien port du Havre du fait d’incidents déjà répertoriés à certains postes actuels (MCT6, CIM8) et d’un projet de création de nouveaux postes pour les grands porteconteneurs. SOGREAH a donc été chargée par le GPMH de réaliser une étude sur modèle mathématique concernant le comportement des navires amarrés au passage d’un navire chenalant. 2. ANALYSE DE L’ETAT ACTUEL Les postes actuels signalés comme étant sujets à des incidents pouvant entrainer des arrêts d’exploitation (phénomènes de ruptures d’amarres, mouvements importants du navire amarré,…) sont essentiellement le poste MCT6, qui accueille des navires minéraliers et le poste CIM 8, qui accueille des navires pétroliers (voir fig.1). Poste MCT6 Poste CIM 8 Quai de l’Asie Quai d’Osaka Fig. 1: Localisation des postes et quais 1 Directeur de Projet, SOGREAH, France, [email protected] Chef de Projet, SOGREAH, France, [email protected] 3 Directeur Technique Adjoint chargé des Projets, GPMH, France, [email protected] 4 Chef de Projet, GPMH, France, [email protected] 2 1 of 16 PIANC MMX Congress Liverpool UK 2010 L’analyse a essentiellement porté sur le poste MCT6 pour lequel on disposait d’un répertoire des différents incidents survenus. 2.1 Types de navires impliqués Pour le poste MCT6, les navires venant décharger sont de deux types: minéraliers de type « Panamax » (65 000 à 80 000 tpl) et minéraliers de type « Capesize » (135 000 à 195 000 tpl). Seuls les seconds types de minéraliers sont sujets aux ruptures d’amarres. Dans leur grande majorité, les incidents ont lieu pour des chargements proches du chargement à l’arrivée, donc pour des tirants d’eau importants. Le déchargement des cales situées aux extrémités de certains navires nécessite un ou plusieurs décentrages du navire. Le décentrage du navire qui implique une configuration d’amarrage différente de celle recommandée en début d’opération par la capitainerie, ne constitue pas un facteur favorisant la rupture d’amarres, car la majorité des incidents répertoriés ont lieu avant tout déhalage. Le clair sous quille, au moment des incidents, est assez hétérogène (variant entre 0,7 m et 14,2 m). Les navires chenalants impliqués dans les incidents sont des porte-conteneurs de deux types: porteconteneurs de faible tonnage (20 000 à 40 000 tpl) et de fort tonnage (50 000 à 90 000 tpl). Les tirants d’eau des porte-conteneurs, au moment des incidents, variaient entre 9 et 14 m. Les clairs sous quille restent assez importants (au minimum d’environ 30 % et le plus souvent supérieur à 50% du tirant d’eau). Les vitesses de passage des porte-conteneurs étaient le plus généralement de l’ordre de 5 à 7 nœuds. Aucune corrélation entre le sens de passage des navires dans le chenal et l’emplacement des amarres cassées n’est mise en évidence par les données. 2.2 Configuration d’amarrage Il existe une configuration d’amarrage type pour laquelle a été conçu le poste: 8 amarres à l’avant et 8 amarres à l’arrière (dont 4 gardes), comme le montre le schéma ci-dessous. Crocs à terre Ducs d’albe Bollards sur appontement Fig. 2: Configuration d’amarrage type – MCT6 Le choix de l’amarrage relève de la responsabilité du bord et en particulier, les tensions dans les amarres doivent être adaptées par l’équipage en fonction de la hauteur d’eau. Les reprises sur les amarres doivent être réalisées pour adapter la longueur des amarres et leurs tensions au chargement et à la variation de hauteur d’eau due à la marée. Si aucune reprise n’est effectuée ou du moins pas de façon systématique ou régulière dans le temps et que la longueur des amarres est trop courte, la tension dans les amarres augmente, ce qui peut se produire en particulier dans les gardes et les traversiers de faible longueur. La majorité des incidents ayant lieu à marée montante, la mauvaise reprise sur les amarres pourrait être une cause ou du moins un facteur aggravant de ruptures d’amarres. La mauvaise reprise des amarres concerne également le cas contraire, à savoir le cas des « amarres molles », pour lequel les amarres se trouvent à un moment donné, trop détendues. 2 of 16 PIANC MMX Congress Liverpool UK 2010 Enfin, on peut signaler que certains navires n’utilisent pas de façon systématique les crocs à terre pour les traversiers mais aussi les bollards sur l’appontement (voir fig. 2), ce qui conduit à des faibles longueurs d’amarres et ce, malgré les avis envoyés au commandant de chaque navire par la capitainerie. Les navires peuvent utiliser des amarres en acier (diamètre 38-40 mm) avec éventuellement des bouts nylon ou des amarres en nylon (diamètre 60-80 mm), voire les deux types en même temps. 2.3 Paramètres principaux D’après l’analyse de l’état actuel et les différentes références sur le sujet (cf. Réf. [1] à [8]), les paramètres les plus importants qui interviennent sur les efforts induits sur un navire amarré par un navire passant à forte vitesse à proximité d’un poste, sont les suivants : o La charge des navires amarrés. En général, les incidents se produisent pour des navires de fort tonnage à pleine charge ou au voisinage de la pleine charge, o Le clair sous quille du navire amarré qui est relié à la charge des navires et au niveau d’eau. Le paramètre adimensionnel à considérer est le rapport T/d avec T, tirant d’eau du navire amarré et d, hauteur d’eau. Les efforts hydrodynamiques augmentent avec l’augmentation de ce paramètre et donc avec la diminution du clair sous quille, o La distance de séparation entre les deux navires. Le paramètre adimensionnel à considérer est le rapport G/B avec G, distance de séparation entre les murailles des deux navires et B, largeur du navire amarré. D’après la référence [3], les efforts hydrodynamiques longitudinaux passeraient par un maximum lorsque ce paramètre est compris entre 2 et 4 tandis que les efforts transversaux et les moments atteindraient leur maximum pour des valeurs du paramètre comprises entre 1,5 et 3, o La vitesse de passage du navire chenalant. Les efforts et moments hydrodynamiques maximaux suivent globalement une loi quadratique avec la vitesse du navire chenalant. La configuration d’amarrage constitue également un paramètre important dont il faut tenir compte. D’après l’analyse précédente, on peut signaler, parmi les causes probables des incidents d’amarrage au poste MCT6 : 3. o La configuration du poste pour les grands navires, qui peut entraîner un plan d’amarrage qui n’est pas forcément optimal en termes de symétrie. En particulier, du fait du déhalage vers l’avant ou l’arrière pour atteindre, au cours du déchargement, les cales les plus extrêmes, il est souvent impossible d’utiliser un des crocs d’amarrage Est ou Ouest à terre et donc d’avoir le même nombre de traversiers à l’avant et à l’arrière. La dissymétrie de la longueur des gardes peut constituer aussi un facteur aggravant. o Des amarres de faible qualité et /ou mal reprises en fonction du niveau d’eau et du déchargement, o L’utilisation de bollards sur l’appontement pour les traversiers, entrainant de très faibles longueurs d’amarres et des efforts importants dans celles-ci. MODELISATION DU COMPORTEMENT DYNAMIQUE DES NAVIRES AMARRES Le phénomène déterminant dans les efforts appliqués sur le navire amarré est l’écoulement lié au déplacement d’eau dans le chenal à faible profondeur lors du passage des grands navires à proximité. La modélisation du système d’amarrage nécessite donc au préalable de connaître: o Les efforts induits sur le navire amarré par le passage de l’autre navire, o Les efforts hydrostatiques et hydrodynamiques du navire amarré, 3 of 16 PIANC MMX Congress Liverpool UK 2010 , puis à partir de ces efforts, de modéliser les réactions du navire amarré sous l’effet des différentes sollicitations. 3.1 Efforts induits par le passage du navire chenalant Lorsque le navire chenalant approche du navire amarré (voir fig.3 ci-dessous), ce dernier commence par être « attiré » vers l’arrière (l’effort longitudinal est dirigé vers l’arrière, l’effort transversal tend à décoller le navire du quai et le moment a tendance à faire tourner le navire amarré vers le quai). Lorsque le navire chenalant arrive au droit du navire amarré, l’effort transversal tendant à écarter le navire du poste est maximal. Puis, lorsque le navire chenalant commence à dépasser le navire amarré, c’est l’avant de ce dernier qui est « aspiré » dans le sillage (l’effort longitudinal est dirigé vers l’avant, l’effort transversal tend à repousser le navire vers le quai et le moment a tendance à faire tourner le navire vers le chenal). Fig. 3 : Schéma des efforts induits par un navire chenalant Les efforts induits par le passage du navire ont été estimés : o En premier lieu, en utilisant un modèle 3D basé sur la théorie du potentiel d’écoulement en fluide parfait modifiée pour tenir compte de la présence simultanée de deux corps flottants (modèle DELPASS de l’Université de Technologie de DELFT ; cf. fig. 4 et réf. [5]). o En deuxième lieu, à partir d’un modèle semi-empirique intégrant des résultats de mesures en laboratoire (cf. réf. [3]) et qui a été calibré à partir des résultats de calcul obtenus avec le modèle précédent. Ce modèle a permis de faire varier certains des paramètres principaux afin de tester leur sensibilité. Fig. 4 : Exemple de champ d’écoulement calculé au droit d’un navire amarré (modèle DELPASS) 4 of 16 PIANC MMX Congress Liverpool UK 2010 3.2 Efforts hydrostatiques et hydrodynamiques Les efforts hydrostatiques et hydrodynamiques ont été obtenus pour le navire amarré en utilisant un modèle 3D de diffraction-radiation5 basé sur la théorie linéaire du potentiel d’écoulement en fluide parfait. L’écoulement est, dans ces conditions, défini par un potentiel initial en l’absence de tout obstacle et la déformation de cet écoulement par le corps immergé induit un potentiel perturbateur (diffraction-radiation). Pour les calculs, les navires sont schématisés en un certain nombre de facettes élémentaires (cf. fig. 5), les fonctions continues (potentiel, vitesses d’écoulement) étant remplacées par leur valeurs discrètes en ces points. Les calculs peuvent tenir compte également de la géométrie du chenal et d’une profondeur d’eau restreinte (faible clair sous quille). Ces calculs permettent d’avoir accès à la pression hydrodynamique sur chaque facette et ainsi au calcul global des torseurs des forces appliquées sur les navires (efforts et moments). Des coefficients d’efforts dus au vent et au courant sont également introduits pour chaque navire. Fig. 5 : Exemple de maillage de la coque d’un navire (porte-conteneurs de 180 000 tpl) 3.3 Efforts dans les amarres et mouvements du navire amarré Les efforts dans les amarres et les mouvements du navire amarré s’obtiennent à partir du modèle SHIP MOORINGS6. Ce modèle permet de simuler les systèmes d’amarrage et le comportement d’un navire amarré en fonction, entre autres, des aspects suivants: o Forces hydrostatiques amortissements), et hydrodynamiques du navire amarré (masses ajoutées, o Faible clair sous quille, o Caractéristiques non linéaires des amarres, des défenses et des ducs d’albe composant le poste d’amarrage, o Prise en compte des conditions d’environnement et en particulier : Vents variables permettant la reproduction des rafales ; Efforts hydrodynamiques dus aux courants ; Agitation due à la houle ou au clapot de multiples directions (calculs en houle irrégulière en tenant compte de tous les termes : efforts du premier ordre, efforts de dérive, termes du deuxième ordre, ondes longues, amortissement visqueux,…), o Prise en compte d’efforts induits par le passage d’un navire à proximité du navire amarré. Ce modèle permet de reproduire différentes configurations d’amarrage, en tenant compte notamment de la configuration du poste (positions dans l’espace à trois dimensions des points d’amarrage, des défenses et des ducs d’albe d’appui), des caractéristiques des amarres utilisées (types, tensions initiales), de la position du navire au cours du déchargement. 5 6 Modèle DIFFRAC de la société MARIN Développé par la société ALKYON 5 of 16 PIANC MMX Congress Liverpool UK 2010 Les calculs effectués permettent alors d’obtenir les efforts dans les amarres et les ducs d’albe, sous l’effet des différentes sollicitations, ainsi que les mouvements, les vitesses et les accélérations du navire amarré. 4. MODELISATION DE LA CONFIGURATION ACTUELLE La modélisation de l’état actuel a consisté d’une part à reproduire un incident réel de manière à vérifier le comportement du modèle de tenue à poste et d’autre part, à déterminer en fonction de la distance entre murailles droites des deux navires, les seuils de vitesse maximale à ne pas dépasser par le navire chenalant afin d’éviter une rupture des amarres. Pour toutes les simulations, les critères suivants, dictés par l’OCIMF, étaient appliqués: o Pour les défenses d’accostage : la déflection de la défense ne doit pas excéder 57.5% de sa hauteur ; o Pour les amarres, la tension dans la ligne ne doit pas excéder 55% de la charge maximale à la rupture, de manière à prendre en compte l’usure éventuelle ou le vieillissement de l’amarre. 4.1 Calage du modèle Le calage du modèle a consisté à reproduire l’incident d’amarrage du 5 décembre 2003 qui s’était caractérisé par une rupture des gardes avant suite au passage d’un porte-conteneurs à une vitesse de 7 nœuds dans le chenal. La configuration d’amarrage est illustrée par la fig.6 ci-dessous Fig. 6 : Configuration d’amarrage lors de l’incident du 5 décembre 2003 La plupart des amarres étaient en nylon (Ø 64 mm), sauf les gardes qui étaient constituées d’amarres acier (Ø 34 mm) avec des queues nylon. Il est à noter que deux traversiers étaient amarrés directement aux bollards de l’appontement et avaient donc une longueur très réduite. La reproduction de cet incident avec le modèle SHIPMOORINGS a permis de bien illustrer les phénomènes de rupture d’amarre induits par le passage d’un navire chenalant. 4.2 Détermination des seuils de vitesse maximale pour le navire chenalant La détermination de ces seuils a été réalisée pour le poste MCT6 avec les conditions suivantes: o Caractéristiques des navires chenalant et amarré : 6 of 16 PIANC MMX Congress Liverpool UK 2010 o Navire Chenalant Amarré Type de navire Porte-conteneurs de 80 à 90 000 tpl Minéralier 160 000 tpl pleine charge Longueur hors tout (m) 318 290 Largeur (m) 42.8 45 Titrant d’eau (m) 14 17.5 Profondeur de dragage (m/CM) (-15) (-16) Niveau d’eau (m/CM) +2 +2 Configuration d’amarrage : il s’agit de celle recommandée par le GPMH avec 12 amarres de pointe et traversiers sur les ducs d’albe et crocs à terre (6 à l’avant et 6 à l’arrière) et 4 gardes amarrées à l’appontement (2 à l’avant et 2 à l’arrière) (cf. fig. 7 ci-dessous). Les prétensions étaient de l’ordre de 7 à 10 t, ce qui représentait environ 10% de la charge à la rupture des amarres. L3, L4 L1, L2 L5, L6 L11, L12 L7 L8 L9 L13, L14 L10 L15, L16 Fig. 7 : Configuration d’amarrage testée pour le poste MCT6 o Différents types d’amarres ont été testés, dont : 1. Amarres en acier (Ø 40 mm, charge maximale à la rupture de 103 t), 2. Amarres en nylon (Ø 64 mm, charge maximale à la rupture de 72 t), 3. Amarres en acier (Ø 40 mm) avec 10 m de queue en nylon Ø 80 mm (charge maximale à la rupture de 103 t), 4. Configuration mixte avec amarres en nylon (Ø 64 mm, charge maximale à la rupture de 72 t), excepté les gardes qui sont en acier avec queue nylon (charge maximale à la rupture de 103 t). o Efforts induits par le navire chenalant : ces efforts ont été obtenus par la méthode décrite cidessus. Les figures suivantes n°8 et 9 présentent quelques résultats obtenus pour les torseurs des forces appliquées au navire amarré. On pourra remarquer la bonne comparaison entre les courbes d’efforts et de moment issues du modèle semi-empirique et celles obtenues avec le modèle DELPASS. 7 of 16 PIANC MMX Congress Liverpool UK 2010 Distance entre murailles droites =100 m 6000 Fx (kN) - (Modèle semi-empirique) Fy (kN) - (Modèle semi-empirique) 5000 Fx (kN) - (Delpass) Fy (kN) - (Delpass) 4000 Efforts appliqués au navire amarré (kN) 3000 2000 1000 0 0 50 100 150 -1000 200 250 300 350 Temps (s) -2000 -3000 Distance entre murailles droites =100 m 150000 Moment appliqué au navire amarré (kN.m) 100000 50000 Mz (kN.m) (modèle semiempirique) 0 0 -50000 50 100 150 200 250 300 350 Mz (kN.m) (Delpass) Temps (s) -100000 -150000 Fig. 8 : Efforts et moment appliqués au navire amarré (distance de 100 m et vitesse de 7 nœuds) Fig. 9 : Efforts et moment appliqués au navire amarré (distance de 100 m et vitesse de 5 nœuds) Les résultats obtenus en faisant varier la vitesse du navire chenalant et la distance entre murailles droites montrent que les amarres en nylon sont les plus critiques en termes de mouvements du navire amarré alors que les amarres en acier donnent les meilleurs résultats. Ceci s’explique par le fait qu’avec un système d’amarrage très souple (amarres nylon), le navire amarré a tendance à être « déplacé » de façon beaucoup plus importante par le navire chenalant. Ces grandes amplitudes de mouvement peuvent alors générer des efforts d’amarrage importants. La vitesse du navire chenalant ne doit pas dépasser 5 nœuds avec des amarres nylon et 6 à 7 nœuds avec des amarres acier, dans la mesure où la distance entre murailles droites du navire amarré et du navire chenalant est comprise entre 70 m et 150 m. En deçà de la distance de 70 m, la vitesse 8 of 16 PIANC MMX Congress Liverpool UK 2010 maximale admissible chute rapidement et cette distance semble donc constituer une limite minimale de passage. Au-delà de la distance de 150 m, la vitesse maximale admissible peut être un peu plus importante (augmentation de l’ordre de 1 nœud si l’on reste dans la « zone navigable » du chenal actuel qui est de l’ordre de 170 m). L’augmentation notable de la vitesse maximale admissible se fait sentir pour une distance entre murailles droites supérieure ou égale à 200-250 m. Il s’agit là d’une distance qu’on ne peut pas atteindre dans l’état actuel du chenal. Un décentrage vers l’Ouest ou vers l’Est du navire amarré par rapport à sa configuration d’amarrage optimale a tendance à aggraver la situation et la distance minimale à respecter entre murailles droites des 2 navires serait alors de 100 m, pour une même vitesse de chenalage. Bien qu’un système d’amarrage plus raide (amarres acier) permette d’augmenter les valeurs maximales admissibles de vitesse de passage du navire chenalant, à distance égale, ce système d’amarrage ne peut pas être considéré comme cas de base, un amarrage entièrement « acier » n’étant pas la norme pour ce type de minéralier et pouvant aussi présenter certains inconvénients lorsqu’il est soumis à des sollicitations dynamiques de fréquence plus élevée (vent notamment). Diverses mesures ayant pour but d’essayer d’améliorer la situation ont également été testées, comme : l’ajout de ducs d’albe d’amarrage afin de raccourcir la longueur des traversiers, la modification de la raideur des défenses d’accostage, l’augmentation de la prétension dans les lignes mais elles n’ont pas apporté de réelle amélioration des seuils de vitesse admissible (à distance égale entre navires). La mesure la plus efficace, en dehors de limiter la vitesse du navire chenalant ou d’augmenter le plus possible la distance entre navire amarré et navire chenalant, consisterait à augmenter le clair sous quille du navire amarré (en le rendant supérieur à 15% du tirant d’eau) par dragage de la souille et le maintien des contraintes opérationnelles en fonction de la marée, comme cela est déjà en vigueur à ce poste. 5. ETUDE DE LA CONFIGURATION FUTURE 5.1 Caractéristiques de la configuration future La configuration future est prévue pour accueillir des porte-conteneurs jusqu’à 180 000 tpl (18 000 TEU) ainsi que des minéraliers jusqu’à 260 000 tpl. Le futur quai à porte-conteneurs sera constitué d’un quai vertical équipé de crocs d’amarrage et de défenses d’accostage. Le poste ainsi que le chenal d’accès seraient dragués à (-18 m/CM), comme le montrent les schémas ci-dessous. Schéma B Bassin Th Ducrocq Schéma A Bassin Th Ducrocq -15 35 85 135 185 235 285 335 385 4 Futur quai 2 -15 35 135 0 -2 -2 -4 -4 -6 -6 -8 -8 -10 -10 -12 -12 -14 -14 -16 -16 -18 -18 235 285 335 -20 Profondeur (m/CM) Niveau d'eau Navire amarré Distance (m) Porte-conteneurs de 180 000 tpl Profondeur (m/CM) Niveau d'eau Fig. 10 : Profils de chenal considérés pour le futur port Types de navires considérés Les types de navires considérés étaient les suivants : 9 of 16 385 Futur quai Distance (m) 5.2 185 2 0 -20 85 4 Navire amarré Porte-conteneurs de 180 000 tpl PIANC MMX Congress Liverpool UK 2010 Type de navire Capacité Longueur hors tout (LOA) Largeur (B) Creux au pont principal (D) Tirant d'eau (T) Porte-conteneurs Porte-conteneurs Porte-conteneurs Vraquier 180 000 tpl - 18000 TEU 110 000 tpl - 8500 TEU 40 000 tpl - 2500 TEU 260 000 tpl 430 m 352 m 220 m 345 m 56 m 42.8 m 32.2 m 51 m 33 m 25 m 20 m - 17 m 14 m 11.5 m 19.5 m Tab. 1: Caractéristiques des navires modélisés Les navires chenalants étaient les porte-conteneurs de 180 000 et 110 000 tpl ainsi que le minéralier de 260 000 tpl. Tous les porte-conteneurs ont été considérés comme des navires amarrés au futur quai. 5.3 Configurations d’amarrage Les configurations d’amarrage retenues sont montrées sur les figures n°11 à 13 pour les différents types de porte-conteneurs. Il faut noter que pour les grands porte-conteneurs (110 000 à 180 000 tpl), trois configurations d’amarrage différentes ont été étudiées. L4, L5 L6, L7 L8, L9 L1, L2, L3 L10, L11 L12, L13, L14 L4, L5 L6, L7 L8, L9 L1, L2, L3 L4, L5 L10, L11 L12, L13, L14 L6, L7 L8, L9 L10, L11 L1, L2, L3 L12, L13, L14 Fig. 11: Configurations d’amarrage pour le porte-conteneurs de 180 000 tpl (14 lignes) 10 of 16 PIANC MMX Congress Liverpool UK 2010 L3, L4 L5, L6 L7, L8 L9, L10 L1, L2 L11, L12 L9, L10 L7, L8 L5, L6 L11, L12 L3, L4 L3, L4 L1, L2 L5, L6 L7, L8 L9, L10 L11, L12 L1, L2 Fig. 12: Configurations d’amarrage pour le porte-conteneurs de 110 000 tpl (12 lignes) L3, L4 L5, L6 L7, L8 L9, L10 L11, L12 L1, L2 Fig. 13: Configuration d’amarrage pour le porte-conteneurs de 40 000 tpl (12 lignes) Différents types d’amarres ont été considérés pour cette étude: ATLAS ; SUPERMAX Ropes ; Nylon. Les principales caractéristiques sont présentées dans le tableau ci-dessous et sur la figure n°14 suivante. Matériau / Type de ligne Atlas de Bexco ropes SuperMax Rope de DSR Corp Nylon Diamètre (mm) 78 40 60 Charge à la rupture (kN) 1 177 1 246 625 Tab. 2: Caractéristiques des amarres 11 of 16 PIANC MMX Congress Liverpool UK 2010 Courbes d'élongagtion 100% 90% 80% SUPERMAX ROPES (40 mm) ATLAS (78 mm) NYLON (60 mm) 70% % MBL 60% 50% 40% 30% 20% 10% 0% 0% 5% 10% 15% 20% 25% 30% élongation (%) SUPERMAX ROPES (40 mm) ATLAS (78 mm) NYLON (60 mm) Fig. 14 : Courbes caractéristiques des amarres en fonction du matériau Le graphe suivant n°15 présente les courbes de réaction- déflexion des défenses d’accostage modélisées. Système 1 - Réaction Système 2 - Réaction Système 1 - Energie Système 2 - Energie Comportement système de défenses Réaction (kN) RPD Système 1 4500 8000 4000 7000 3500 6000 3000 5000 2500 4000 2000 3000 1500 2000 1000 1000 500 0 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 Déflexion (m) Fig. 15 : Courbes réaction- déflexion des défenses d’accostage 12 of 16 Energie (kN.m) RPD Système 2 9000 PIANC MMX Congress Liverpool UK 2010 Les prétensions appliquées étaient de l’ordre de 50 kN sur chacune des lignes. Quelques essais ont été effectués avec une prétension de 100 t par amarre. 5.4 Efforts induits par les navires chenalants Les efforts induits par les navires chenalants ont été obtenus par la méthode décrite ci-dessus. Les figures suivantes n°16 et 17 présentent quelques résultats obtenus pour les torseurs des forces appliquées au navire amarré (ces efforts correspondent aux cas de passage présentés sur la figure 10 : niveau d’eau de +0,5 m/CM ; navire chenalant de 180 000 tpl et navires amarrés de 180 000 et 40 000 tpl). D’une manière générale, les calculs des efforts ont amené aux conclusions suivantes : o Les futurs navires chenalants étant beaucoup plus grands que ceux de la configuration actuelle, les efforts générés par ces navires sur les navires amarrés sont beaucoup plus élevés. o D’une manière générale, les efforts et moments générés sur le navire amarré par le passage d’un grand porte-conteneurs sont supérieurs à ceux générés par le passage d’un grand minéralier de dimensions équivalentes, bien que le déplacement de ce dernier soit plus important. En effet, la forme de la coque du porte-conteneurs, plus profilée, génère une vague d’étrave plus préjudiciable. o Il apparaît que le type de structure du poste (quai ou ouvrage sur pieux) a une certaine influence sur les efforts induits par le navire chenalant sur le navire amarré, les efforts transversaux et les moments induits sur un navire à quai étant plus faibles que ceux sur un navire amarré à un appontement. Ceci peut s’expliquer d’un point de vue hydrodynamique par la restriction de section hydraulique due à la présence du quai qui limite les surpressions entre le quai et la coque du navire amarré. Fig. 16: Efforts et moment: navires chenalant et amarré: porte-conteneur de 180 000 tpl; distance de 100 m et vitesse de 5 nœuds 13 of 16 PIANC MMX Congress Liverpool UK 2010 Fig. 16: Efforts et moment: navires chenalant: porte-conteneur de 180 000 tpl; navire amarré: porte-conteneur de 40 000 tpl; distance de 100 m et vitesse de 5 nœuds 5.5 Conditions d’environnement Lors des simulations de tenue à poste, outre les efforts induits par le navire chenalant, il a également été tenu compte des conditions environnementales (courant et niveau de marée, vent). Une étude de courantologie sur modèle mathématique TELEMAC a permis d’identifier les caractéristiques du courant au niveau des postes. Au niveau du futur quai à conteneurs, quatre conditions de courant ont été testées (pour une marée de coefficient 95). Ces conditions ont été sélectionnées de manière à ce que la combinaison (vitesse / direction / niveau d’eau) soit la plus critique possible pour la tenue à poste du navire amarré : o Courant C0: Vitesse de 0.10 m/s; Direction: N110°; Niveau d’eau= +0.5 m CMH (BM), o Courant C1: Vitesse de 0.42 m/s; Direction: N150 ; Niveau d’eau= +3.4 m CMH (PM-4 h), o Courant C2 : Vitesse de 0.27 m/s; Direction: N150°; Niveau d’eau= +2.0 m CMH (PM-4 h 30’), o Courant C3: Vitesse de 0.18 m/s; Direction: N110°; Niveau d’eau= +3.9 m CMH (PM-3 h 45’). Sur la base de la rose des vents au port du Havre, différentes conditions de vent ont été testées: o Vitesse de vent correspondant à la limite d’opérabilité des portiques: Vitesse de 38 nœuds avec une direction N230° ou N270°. o Vitesse de vent opérationnelle: Vitesse de 25 nœuds avec une direction N230°. 5.6 Conclusions Les conclusions suivantes, concernant les porte-conteneurs amarrés au futur quai peuvent être tirées au vu des résultats obtenus: o Les deux directions de vents testées sont plutôt favorables vis-à-vis des mouvements des navires à poste car ce sont des vents non décostants. Les vents peuvent cependant, dans certains cas, accentuer les mouvements dans la mesure où les rafales « accrochent » une période de résonance du système navire/configuration d’amarrage. o Bien qu’un système d’amarrage « très raide » n’est pas forcément favorable lorsqu’il est soumis à des sollicitations dynamiques de fréquence plus élevée, les lignes plus raides (de type SuperMax) permettent toujours d’obtenir des mouvements bien inférieurs à ceux obtenus avec des lignes plus souples (de type Atlas et à fortiori nylon), au prix cependant de tensions dans les lignes plus fortes et parfois légèrement au dessus de la limite admissible dans le cas du porte-conteneurs de 180000 tpl amarré. 14 of 16 PIANC MMX Congress Liverpool UK 2010 o La configuration d’amarrage n°1 pour le porte-conteneurs de 180000 tpl semble plus favorable que les deux autres tandis que la configuration n°2 est assez défavorable car elle conduit à des mouvements plus importants. o Lorsque ni le vent ni le courant ne sont pris en compte et de façon à respecter le critère limite de tension admissible, la distance entre navire chenalant (porte-conteneurs de 180 000 tpl ou minéralier de 260 000 tpl) et grand porte-conteneurs amarré (180000 tpl), pour une vitesse de passage du navire chenalant de 5 nœuds, devrait être au minimum de: - 150 m si le navire est amarré avec des lignes raides (de type Supermax), - 300 m si le navire est amarré avec des lignes nylon. (Note: pour les lignes Atlas dont la courbe caractéristique se situe entre celle des lignes Supermax et celle des lignes nylon, on pourrait s’attendre à une distance intermédiaire). o Lorsque l’on prend en compte le courant et le vent (qui rappelons-le est plutôt favorable), il apparaît une moindre différence entre les types d’amarres (hors amarres nylon). La distance minimum entre navire chenalant (porte-conteneurs de 180000 tpl ou minéralier de 260000 tpl) et navire amarré (porte-conteneurs de 180000 tpl ou 110000 tpl) pour respecter le critère limite de tension admissible, est de l’ordre de: - 200 m pour une vitesse de 5 nœuds, lorsque le navire amarré est un grand porteconteneurs de 180000 tpl - 150 m pour une vitesse de 6 à 7 nœuds, lorsque le navire amarré est un porte-conteneurs de 110000 tpl. o Les distances à respecter entre navire chenalant (porte-conteneurs de 180 000 tpl) et petit porte-conteneurs (40 000 tpl) amarré avec des lignes nylon, sont les suivantes: - 50 m pour une vitesse de passage de 5 nœuds, - 200 m pour une vitesse de passage de 6 nœuds, - 250 m pour une vitesse de passage de 7.5 nœuds. 15 of 16 PIANC MMX Congress Liverpool UK 2010 Références [1] Cornett A., Tschirky P., Knox P., Rollings S. (2008) – “Moored ship motions due to passing vessels in a narrow inland waterway” - Proceedings of 31st International Conference on Coastal Engineering, ASCE [2] Cohen S.B., Beck R.F. (1983) – “Experimental and theoretical hydrodynamic forces on a mathematical model in confined waters” - Journal of Ship Research – Vol. 27, n°2 – June 1983 [3] Kriebel D., Seelig W., Eskijian M. (2005) – “Mooring loads due to parallel passing ships” – Naval Facilities Engineering Service Center - Technical Report TR-6056-OCN [4] Muga B.J., Fang S. (1975) – “Passing ship effects – From theory and experiment” – Offshore Technology Conference, OTC 2368 [5] Pinkster J.A., Ruijter M.N. (2004) – « The influence of passing ships oh ships moored in restricted waters » - Offshore Technology Conference, OTC 16719 [6] Remery G.F.M. (1974) – “Mooring forces induced by passing ships” – Offshore Technology Conference, OTC 2066 [7] Spencer J., McBride M., Shennan D. (2008) – “Simulated moored and passing ship interaction for harbour and navigation design” – Proceedings of 31st International Conference on Coastal Engineering, ASCE [8] Wang S. (1975) – “Dynamic effects of ship passage on moored vessels” - Journal of the waterways harbors and coastal engineering division, ASCE 16 of 16 PIANC MMX Congress Liverpool UK 2010 NEW GUIDANCE FOR THE USE OF CONCRETE IN MARITIME ENGINEERING by by S. Dupray 12 3 , K. Gamst and B. Wood-Ballard 4 ABSTRACT This paper presents an overview of project that aims to deliver an up-to-date, self-supported and decision aiding guidance on the use of concrete for design / construction / repair and maintenance of maritime structures. Some of the key findings relevant to navigation structures are detailed. 1. INTRODUCTION Concrete has been extensively used as one the main material to construct navigation structures and the various structures associated to navigation, because it is versatile material, durable and widely available. Despite the long history of capital investment in concrete maritime structures, it is generally recognised that there are gaps in technical guidance documents available on the use of concrete for such structures. In particular, the need for consolidated guidance has been identified in the United Kingdom and France : joint action was needed to provide comprehensive guidance in particular considering key codes in Europe have appeared e.g. Eurocodes (in particular Eurocode 2) and standards on concrete (in particular EN 206) and concrete constituents (in particular EN197). The paper presents highlights on key conclusions from a collaborative project between UK and France. The project aimed at providing a holistic view on the subject, at summarizing good practice (Allen 1998, Concrete Society 1986, CETMEF 2008) and at transferring findings from recent researches into the day-to-day practice. The objectives of the research is to guide asset managers, designers and contractors on how to deliver concrete structures that are durable over their design life, to make an optimum selection from concrete materials and concrete construction options. For example the guidance provides guidance on how to avoid the risks of corrosion, how to deal with long design lives (typically greater than 50 years) and how to address the risk of attrition . It focuses on maritime and estuarine port structures ; although not focused on inland navigation structures, some of the findings may also be relevant. Generalities on concrete and concrete constituents, detailed concrete mix design procedures and detailed calculation of hydraulic conditions and loads are avoided but key references are identified. 2. APPROACHES TO MARITIME CONCRETE 2.1 Holistic approach The ‘ideal concrete’ does not exist and there is no ‘off the shelve’ concrete for a project: an optimum balance between factors and constraints of the project for design, construction, operation and maintenance should be achieved. This could be achieved by accounting for various aspects including fitness for purpose, ease of action, safety, environmental aspects, cost aspects, durability and strength. Hence the approach to concrete material and concrete structures should encompass asset management, understanding of deterioration agents and environmental forcing, design for durability and for loads, concrete technologies, construction technologies, specific testing, construction and quality control as well as monitoring / maintenance and repair. Other holistic considerations that are important include service life requirement, programming, health and safety, risks, whole life costs, environmental considerations including sustainability and architectural issues. Title 1 Coordinator of methods and research for structures, CETMEF, France, [email protected] 2 (at the time of the project) seconded engineer, HR Wallingford, United Kingdom 3 project manager, CIRIA, United Kingdom 4 project manager, HR Wallingford, United Kingdom 1 of 9 PIANC MMX Congress Liverpool UK 2010 2.2 Structure specific approach To provide a rational approach to structures, they have been grouped into families that display comparable functions and service exposures. For each group or family, information is given on structural forms, possible construction methods, key durability issues and key design issues. The relevant structures are to be found amongst some of the families that are solid vertical and battered walls, block works walls (such as quays), wave walls, crown walls, concrete armour units, caissons, revetments, piled concrete structures (such as jetties) and others structures (such as navigation towers, slipways, locks) … Some examples are given in Figure 1. Coastal works is also a family within the project. Crown wall on rubble mound breakwater (CETMEF) Concrete armour unit (CLI) Blockwork wall (Halcrow) Jetty (Halcrow) Maritime lock (Grand Port Maritime du Havre) Caissons (SAIPEM) Figure 1: example of concrete structure relevant to navigation 3. DEGRADATION OF MARITIME CONCRETE 3.1 Overview of degradations and their effects All maritime structures are exposed to aggressive environmental agents and other factors, including : • Seawater (with active components such as sulphate, chloride, dissolved CO2), • Moving waters (including tides, waves and currents), 2 of 9 PIANC MMX Congress Liverpool UK 2010 • Climate and climate change, • Abrasion induced by moving waters, sediments, propeller jets, and operations, • Impact induced by sediments or water alone, • Toe scour and erosion, • Unexpected incompatibility between concrete constituents, and • Other factors such as collision, spillage, biological attack (Dhir 2002, Toutlemonde 2007). All of the above should be taken into account during the design to ensure appropriate performance during construction, operation and maintenance. However, this is not sufficient. Actually – and less obviously – maritime concrete structures are also prone to specific internal degradation mechanisms that are associated with the forms of construction : • The use of slender elements requires reinforcement to resist the severity of service/hydraulic loads, which is consequently exposed to the corrosion risks associated with the maritime environment ; • Awkward shape are sometime used to optimize the structures mass performance but, in turn, plastic settlement of concrete and associated weakness may become a significant risk ; • An alternative is to use large and/or bulky (sometime non reinforced) elements, which in turn may lead to significant temperature increase in the concrete during setting (greater than 65°C) resulting in delayed ettringite formation (DEF) and internal sulphate attacks that may alter concrete strength without signs at early stage ; • High temperatures may lead to thermal cracking in particular if large thermal gradient exists ; • High density concrete is a means to increasing hydraulic stability, e.g. of concrete armour units. Durability and volume stability of heavy aggregates should be studied to avoid deteriorations in service ; • Structures are often build in stages which may lead to friction shrinkage cracks when fresh concrete is cast in long sections over hardened concrete, e.g. when casting (or recasting) a capping beam on top of a quay. If not correctly accounted for during design and construction, these may results in negative effects on the concrete and its reinforcement that are summarized in Table 1. 3.2 Rational approach to aggressiveness In Eurocode 2, a rational approach to assessing/designing for several of the external agents is given. Standard classes of exposure are given depending on the exposure, the location and the position of the structure or part of it, including: • XF for freeze and thaw ; • XA for chemical attack, e.g. induced by ground ; • XM for attrition. Other classes of exposure are specifically associated with corrosion: • XS for corrosion induced by seawater attack on reinforcement ; • XC for corrosion induced by CO2 and carbonation ; • XD for corrosion induced by other sources of sulphate. Other exposure classes have also been developed in specific European countries such as XH exposure classes for Delayed Ettringite Formation (LCPC 2007) or ACEC for aggressiveness of the chemical environment (BRE 2005). These classes of exposure allow standard selection of concrete constituents (nature and quantities) and reinforcement cover for a 50 year design life (and in some instance for a 100 year design life). 3 of 9 PIANC MMX Congress Liverpool UK 2010 Table 1 gives a more detailed inventory of deterioration agents and their associated effects. Overview of agents Overview of effects External agents Internal agents On reinforcement On concrete Ambient temperature Temperature (internal) Rebar corrosion Sulphate attack Traffic Chloride content (internal) Expansion Alkali-aggregate reaction Solar exposure Fire and chemical spillages Tide Section loss Sulphates (internal) Strength loss Leaching Alkalis (internal) Cracking Unstable constituents Expansion Wave action Spalling Current action Scaling Ice action Section loss Berthing/mooring forces Strength loss Off-loading forces Water tightness loss Impact Atmospheric CO2 Seawater (capillary rise/evaporation) Seawater (diffusion) Seawater (pressure head) Biological Ground sulphates Abrasion Table 1: Overview of agents acting on concrete and/or reinforcement at element and/or structure scale, and associated effects 4. CONCRETE DESIGN 4.1 Concrete technologies The various risks identified earlier can be met by selecting the most appropriate concrete constituents and quantities, the most appropriate concrete fabrication and placing techniques, which is discussed below. The designer can specify / adjust the reinforcement type (none, steel, fiber, others), the source of concrete (ready-mix supplier or in-situ mixer), the place of casting (in situ, precast or a combination of both using precast elements as permanent formwork for the cast in situ concrete), the type of placing of in situ concrete (e.g. underwater, sprayed, jumpformed, slipformed etc). The designer can also specify the concrete constituents / properties / performance such the concrete density, strength, porosity and permeability. Hence, the designer can select the most appropriate option to facilitate construction of the various parts of the structure, allowing for the best construction quality, and in fine the best durability. Concrete design and selection of the concrete constituents are important steps to fulfill the client needs. The designer has to identify and take into account the constraints of the site and the possible construction methods that would be envisaged to adapt the properties of both the fresh and hardened concrete to the project needs. In Europe, most of the concrete constituents, including cement and additions, various sizes of aggregates, admixtures, and fibres are subject to unified Euro-standards (such as EN 197 for cements) in substitution for previously used national standards. Maritime concrete constituents should 4 of 9 PIANC MMX Congress Liverpool UK 2010 be selected with reference to availability, cost, environmental aspects, risk of alkali aggregate reaction, etc. In addition, risks of deterioration are key drivers in the maritime environment, in particular: 4.2 • Cement should be able to resist attack from seawater (by having a limited C3A content or by complying to NF P 15317 in France) (Boutouil et al 1999), • Some cements/additives can offer improved durability, • Some cements/additives with lower heat of hydration can reduce the risk of DEF/ETC, • Sulphate resisting cements display a better resistance to DEF, • High quality aggregates can result in concrete that is resistant to the abrasive forces of water / sediment, • Fibres can improve concrete strength and resistance to abrasion, • High density aggregates can result in higher density concrete, and • Admixtures (used in small quantities) can improve the properties of concrete in the fresh and hardened state, and are possible ways to enhance both concrete workability and durability. Design for loads The mix design will benefit from empirical mix design relations but will require laboratory trials to confirm concrete performance during both the construction phase and in service, in particular for special concreting techniques such as underwater placing, spraying, slip-forming etc. Full scale trials (including trial pour or pumping trials) will at times be required as well as pilots or mock-ups for larger sites. Concrete design includes both design for loading and design for durability. Design for loading has many things in common with design of other civil engineering concrete structures, although the determination of the loading condition and internal stress may be challenging when the structure is exposed to wave impact loadings. Physical modelling may be the route required to refine loading conditions that would otherwise be determined using empirical relations. For other elements, such as concrete armour units, the actual forces on / stress within the individual elements have historically not been required during the structural design nor were such data readily available (Davidson and Magoon 1990). In such scenarios, concrete design has been based on field or laboratory trials and experiences of the unit inventor or licensee. 4.3 Design for durability Concrete generally offers a high durability ; nevertheless, the duration over which it should display a good performance without specific repair or maintenance activity, ie. the ‘design life’, should be defined and agreed upon as an input data to the design process. At least two routes are available for the design of concrete for durability, depending on the design life. For the design life of 50 years, concrete design can be carried out by using standard recommendations of EN206 and complementary standards. Mix design tables are available that allow to select concrete mix characteristics for the various classes of exposure to environmental agents. This includes the type of cement, the water to cement ratio, the concrete grade, the concrete dosage, the air content. Some national complementary standards may include the cement class, water absorption, class of compaction, class of consistence or further details on proportions of additions (e.g. British standard BS 8500 or French standard NF EN 206). The details for the minimum cover required for durability reinforcement are given by EN 1992 and may be detailed or adapted for various concrete qualities and grades in national complementary standards. For longer design lives of permanent maritime structures (between 50 to 120 years or even more than 120 years in exceptional situations), EN 206 does not offer direct guidance for concrete design. The designer should use a ‘performance based approach’ to set properties that the concrete should 5 of 9 PIANC MMX Congress Liverpool UK 2010 achieve with reference to buildability, structural strength in service and in particular durability. In this situation the designer does not specify concrete constituents and proportions but rather target characteristics and durability indicators to be achieved by the concrete. As an example, durability indicators for the concrete in relation to corrosion of the reinforcement might be porosity, permeability to water and gas, and chloride diffusion rates (AFGC 2004, AFGC 2007). Other specific durability indicators are required for other degradation mechanisms affecting concrete such as the amount of free silica for alkali aggregate reaction (LCPC 1994, BRE 2004), elevated temperatures for delayed ettringite formation (BRE 2001, LCPC 2007) or temperature differences for early thermal cracking (Bamforth, 2007). The values of the durability indicators are generally not readily available and need to be determined by testing, sometime over a sufficient period of time, say several months; this time requirement needs to be taken into account in the programme (Arliguie and Hornain 2007). Durability modelling may also help to refine estimates of the ingress of chloride, carbonation etc over the design life of the structure (Quillin 2001). In addition, durability indicators can be confirmed by regular monitoring of the concrete in the structure in service to refine the estimate of the actual remaining life of the structure (Buenfeld et al 2008). 4.4 Enhancement of durability Concrete material alone may not achieve the required durability ; enhancement measures are available to extend the concrete design life and can be very useful when exceptional design lives are sought. These enhancement measures comprise of special admixtures, cathodic prevention, coating, barriers but also specific construction techniques such as controlled permeability formwork (Price 2000). Some enhancement techniques can also be used on existing structures as part of maintenance activities. To address the specific aggressiveness of the maritime environment and specific forms of concrete, adhoc tests / trials are generally required in addition to more conventional tests/trials for durability design; such adhoc tests / trials may also be required during construction and for QA/QC. During the monitoring process of the structure in service, these tests are a way to assess the remaining life of the structure. The tests comprise of testing of seawater and chloride penetration, testing attrition and impact resistance. Specific tests are also available for underwater, slipformed, sprayed concrete as well as trial manufacturing of concrete units. 5. WORKING WITH CONCRETE 5.1 Construction Construction with concrete in the maritime environment is particularly challenging because some work may need to be carried out underwater. Dry construction windows can be reduced due to tides, concrete may be exposed at early age to water and waves and also delivery of concrete to its location in the works may be difficult. Construction may also need to be carried out while traffic is maintained in access channels, berth or infrastructures. To ensure successful construction of structures, early consideration of all these constraints must be recognized and construction activities in difficult areas should be minimized. Hence, the use of precast concrete is an attractive option in the maritime environment as it allows for some work to be done without contact with seawater. The key construction factors that need to be considered when selecting between cast in situ concrete, precast concrete or a combination of both are the position of the work with reference to water levels, waves and tide, lead time prior to work commencing on site, construction time on site, access and remoteness, space available on site or adjoining the site as well as workmanship and lifting equipments (including their access to site). Some concreting techniques are particularly adapted to the maritime environment including underwater tremmied concrete, self compacting concrete, horizontal and vertical slipformed concrete and also sprayed concrete. In situ concrete can be used in combination with precast elements that can act as permanent formwork and falsework as these offer increased flexibility and quality. Precast elements – factory manufactured or cast near to the site – may also be used as part of the permanent structure without in situ concrete e.g. in block work walls or armour layers. Quality control and quality assurance during concrete fabrication and construction are important because of the restriction on inspection (or repair) of the completed works. Such procedures should be adapted to the structure location and position, and to the form of the completed works. They should focus on those aspects critical to quality and durability of the construction. 6 of 9 PIANC MMX Congress Liverpool UK 2010 5.2 Repair and maintenance Many of the existing structures are aging and some of them are actually historical structures dating back to the XIXth century. They have been build with different materials, techniques than the recent ones and the knowledge of these structure is rarely comprehensive. Regular inspections and appropriate monitoring should be carried out as part of asset management, which allows optimized investigations, maintenance and repair of the structure. Also, it allows determination and refinement of the remaining life of the structure, in particular when managing for exceptional design lives. The various potential repair activities for maritime concrete, i.e. partial restoration, full restoration, protection and prevention and strengthening largely use concrete as a repair material. Some other specialist techniques / materials are also used for protecting the structure, protecting its concrete and preventing it from further deterioration. A multi-scale approach is essential as these activities can (and sometimes should) take place at both the material scale as well as the structure scale. After selective removal of the degraded component or deteriorated concrete, common methods of repair are replacement of the concrete / element, in-situ recasting, spraying concrete, and placement of concrete bags. 6. OUTCOME OF THE RESEARCH The findings of the research summarised above will be available in 2010 in a CIRIA publication titled ‘Use of concrete in maritime engineering – a guide to good practice’ that will also be available in French by CETMEF early 2011. The manual will be an integrated, self-supporting document comprising eight technical chapters as shown in Figure 2. It will help the asset manager / designer / contractor to address the key challenges associated with the use of concrete in the maritime environment, in particular: • Designing for target performance of concrete elements, • Ensuring durable concrete material which provides appropriate performance during the service life of the structure, • Providing strength to concrete, while avoiding the corrosion risk associated with conventional uses of reinforcing steel, • Designing with mass concrete, • Mitigating chloride-related corrosion of reinforcing steel, • Delivering concrete which is resistant to attrition, • Defining the extent to which non-reinforced concrete armour units may be used to structurally respond to hydraulic loads (see Figure 2), • Carrying out appropriate testing of concrete for the maritime environment – in particular for long term resistance to attrition, resilience to impacts, resistance to chloride attack, including laboratory testing and pilots, and • Repairing of maritime concrete structures. 7 of 9 PIANC MMX Congress Liverpool UK 2010 Figure 2: structure of the manual ACKNOWLEDGMENTS This project was managed by CIRIA primarily funded by the Environment Agency. HRWallingford and Halcrow were contractors for this project. Significant support in kind to this project was provided by CETMEF. Other funders include the Bam Nuttall, Civil and Marine, Concrete Centre, Institution of Civil Engineers, UK Quality Ash Association. Significant in kind contributions have also been provided by Aberdeen Harbour, CNR, Concrete Layer Innovation, Concrete Society, Eiffage, Port of Marseille, le Havre and Nantes, SAIPEM, US Army Corps of Engineers. The research has been carried out by a consortium comprising HR Wallingford and Halcrow. The project has been managed by CIRIA. References AFGC (2004). Conception des bétons pour une durée de vie des ouvrages, AFGC, Paris, 245pp. AFGC (2007). Concrete design for a given structure service life. Durability management with regard to reinforcement corrosion and alkali-silica reaction, AFGC, Paris, 245pp. Allen, R. (1998). Concrete in coastal structures, Thomas Telford, London, Great Britain, 301pp. Arliguie G. & H. Hornain (2007). GRANdeurs associées à la DUrabilité du BEton, Presses de l’Ecole Nationale des Ponts et Chausées, Paris, 437pp. Bamforth, P.B. (2007). Early-age thermal crack control in concrete, CIRIA, report C660, 112pp. Boutouil, M., Caminade, D., Henry, D. & Levacher, D (1999). Durabilité des mortiers et des bétons en milieu marin naturel et reconstitué – comportement mécanique et chimique à long terme, 182p BRE (2001). Delayed ettringite formation: in-situ concrete, BRE press, 8pp. BRE (2004). Alkali–silica reaction in concrete – Digest 330 BRE (2005). Concrete in aggressive grounds, BRE Press, 70pp. 8 of 9 PIANC MMX Congress Liverpool UK 2010 Concrete Society (1986). Marine Concrete, Proceeding of the International Conference on Concrete in the Marine Environment, London, 22-24 September 1986. Concrete Society, 446pp. Buenfeld NR, Davies, RD, Karimi, A & AL Gilbertson (2008). Intelligent monitoring of concrete structures. CIRIA, 116 p. CETMEF (2008). Guide sur utilisation des bétons en sites aquatiques, CETMEF, Compiègne, 53p. Davidsonn, DD & O. Magoon (1990). Stresses in concrete armour units, ASCE Publishing, New York (USA), 421pp. Dhir RK. (2002). Challenges of Concrete Construction: Concrete for Extreme Conditions, Thomas Telford, London (GB), 367pp. LCPC (1994). Recommandations pour la prévention des désordres dûs à l'alcali-réaction, LCPC, Paris, 51 pp. LCPC (2007). Recommandations pour la prévention des désordres dus à la réaction sulfatique interne, Collection Guide technique. LCPC, Paris, 60pp. Price W.F. (2000). Controlled permeability formwork, CIRIA, report C511, 102pp. Quillin, K. (2001). Modelling degradation processes affecting concrete, BRE Press, 85pp. Toutlemonde, F et al. (2007). Concrete under severe conditions – environment and loading – Actes de la 5ème conférence internationale sur les structures en béton sous conditions extrêmes d’environnement et de chargement. LCPC, Paris, 1866pp. 9 of 9 PIANC MMX Congress Liverpool UK 2010 Feedback on the repairing works on French navigable dikes and banks Retour d'expériences sur les réparations de digues et de berges de VNF by M. Galiana, D. Dubois, B. De Bruyn CETMEF 1 Y. Deniaud, CETE Nord-Picardie D. Goutaland, D. Le Bris, M. Massardi, CETE Lyon B. Dorbani, M. Hervé, CETE Est Abstract : The French navigable waterways belonging to VNF (Voies Navigables de France), which is the manager of waterways in France, are old and the repairing works are frequent. Concerning the dikes and the banks, the repairing works take place most of the time after a breach failure, a bank erosion or an inefficiency of the watertightness system. The technics used to solve these pathologies are various. It can be compacted clay, rocks, geomembranes, sheet piles, gabions mattresses or vegetative bank protection. In 2008, in order to make an assessment on the pathologies and the technics used, the Centre d'Etudes Techniques Maritimes et Fluviales (CETMEF) in collaboration with the Centre d'Etudes Techniques de l'Equipement (CETE) made a feedback on the repairing works of French navigable dikes and banks of VNF. Data were collected on 41 cases of repairing works and an information note was written for each case. This paper focuses on this feedback led by CETMEF and CETE. The context and a synthesis of this feedback will be presented and an analysis of repairing works will be made. Some examples of repairing works will also be given. Keywords: Dikes, banks, protection, watertightness, breach failure, erosion, repairing works, sheet pile, geomembranes, vegetative bank protection. Congress topic : Sustainable renovation – Renovation of waterway infrastructures Résumé : Les ouvrages de Voies Navigables de France ( VNF) sont anciens et les travaux de réhabilitation ou de reconstruction sont fréquents. Les réparations de digues ou d'aménagements de berges ont souvent lieu suite à une rupture d'ouvrages, à des problèmes d'affouillements ou à une déficience du système d'étanchéité. Diverses techniques existent : enrochement, corroi d'argile, géomembrane, rideau de palplanches, matelas gabions ou technique en génie végétale. En 2008, afin de faire un bilan sur les désordres affectant les digues et les berges de VNF et les travaux de réparations, le Centre d'Etudes Techniques Maritimes et Fluviales (CETMEF) a réalisé en collaboration avec les Centres d'Etudes Techniques de l'Equipement (CETE) un retour d'expériences sur 41 ouvrages de VNF. Cet article présentera ce retour d'expériences et quelques exemples d'aménagements et de réparations. Les désordres et les réparations seront étudiés et un bilan du retour d'expériences effectué. Mots-clés: Digues, berges, protection, étanchéité, brèche, érosion, réparations, palplanches, géomembranes, techniques végétales. 1 [email protected] 1 PIANC MMX Congress Liverpool UK 2010 Mathieu Galiana, Denise Dubois, Bertrand De Bruyn Centre d'Etudes Techniques Maritimes et Fluviales 2, Boulevard Gambetta – BP 60 039 - 60 321 Compiègne Cedex – France Tel : +33.344.92.60.69 Fax : +33.344.92.60.75 [email protected] [email protected] [email protected] Yann Deniaud Centre d'Etudes Techniques de l'Equipement Nord-Picardie Laboratoire Régional des Ponts et Chaussées de Saint-Quentin 151, rue de Paris – 02100 Saint-Quentin – France Tel +33 323 06 18 22 Fax +33 323 64 11 22 [email protected] David Goutaland, Michaël Massardi Centre d'Etudes Techniques de l'Equipement de Lyon Laboratoire Régional des Ponts et Chaussées d'Autun Boulevard Giberstein – BP 141 – 71 405 Autun – France Tel +33 385 86 67 91 Fax +33 385 86 67 79 [email protected] [email protected] Delphine Le Bris Centre d'Etudes Techniques de l'Equipement de Lyon 46, rue Saint Théobald - BP 128 - 38081 L'Isle d'Abeau– France [email protected] Matthieu Hervé Centre d'Etudes Techniques de l'Equipement de l'Est Laboratoire Régional des Ponts et Chaussées de Nancy 71 Rue de La Grande Haie - 54510 Tomblaine– France Tel +33 383 18 41 16 Fax +33 383 18 41 00 [email protected] 2 PIANC MMX Congress Liverpool UK 2010 Bruno Dorbani Centre d'Etudes Techniques de l'Equipement de l'Est Laboratoire Régional des Ponts et Chaussées de Strasbourg 11, rue Jean Mentelin - 67035 Strasbourg Cedex 2– France Tel +33 388 77 79 31 Fax +33 388 77 46 20 [email protected] 3 PIANC MMX Congress Liverpool UK 2010 Feedback on the repairing works on French navigable dikes and banks Retour d'expériences sur les réparations de digues et de berges de VNF by M. Galiana, D. Dubois, B. De Bruyn, CETMEF Y. Deniaud, CETE Nord-Picardie D. Goutaland, D. Le Bris, M. Massardi, CETE Lyon B. Dorbani, M. Hervé, CETE Est RESUME Les ouvrages de Voies Navigables de France ( VNF) sont anciens et les travaux de réhabilitation ou de reconstruction sont fréquents. Les réparations de digues ou les aménagements de berges ont souvent lieu suite à une rupture d'ouvrages, à des problèmes d'affouillements ou à une déficience du système d'étanchéité. Diverses techniques existent : enrochement, corroi d'argile, géomembrane, rideau de palplanches, matelas gabions ou technique en génie végétale. En 2008, afin de faire un bilan sur les désordres affectant les digues et lesberges de VNF et les travaux de réparations, le Centre d'Etudes Techniques Maritimes et Fluviales (CETMEF) a réalisé en collaboration avec les Centres d'Etudes Techniques de l'Equipement (CETE) un retour d'expériences sur 41 ouvrages de VNF. Cet article présentera ce retour d'expériences et quelques exemples d'aménagements et de réparations. Les désordres et les réparations seront étudiés et un bilan du retour d'expériences effectué. MOTS- CLES Digues, berges, protection, étanchéité, brèche, érosion, réparations, palplanches, géomembranes, techniques végétales. 1. INTRODUCTION Les digues et les berges de voies navigables constituent des ouvrages majeurs pour leur environnement proche et la pérennité de la voie d'eau. Ces deux types d'ouvrages se retrouvent à la fois sur les canaux et les rivières navigables et assurent de nombreuses fonctions: physique (étanchéité, soutènement d'infrastructures,...), hydraulique (échanges d'eau entre la voie d'eau et la nappe pour les berges par exemple), écologique (biodiversité, filtre de l'eau,...), socio-économiques (accueil d'activités telles que la pêche, le vélo...) et paysagères (attrait touristique de la voie d'eau). Les ouvrages de VNF sont anciens et il en résulte que les digues et berges sont soumises à de nombreuses pathologies telles que les affouillements, les fuites, les glissements,... De nombreuses techniques de réparations existent et diffèrent suivant la localisation et l'enjeu de l'intervention. Elles peuvent consister en une intervention sur : 2 le corps de digue après une brèche partielle ou totale. Un remblaiement en matériaux d'apport associés parfois avec des techniques de génie civil tels que les enrochements, les palplanches et le béton est alors réalisé; la berge suite à l'inefficacité de la protection, à des anses d'érosion, etc. Des techniques traditionnelles de génie civil telles que les palplanches, les enrochements ou des techniques en 2 génie végétal telles que le fascinage ou les boudins d'hélophytes peuvent être mises en œuvre. Les techniques en génie végétal sont des techniques utilisant des végétaux vivants ou parties de végétaux vivants. Ces techniques reposent sur l'aptitude des végétaux utilisés à se multiplier et à fixer le sol par le développement de leur système racinaire 4 PIANC MMX Congress Liverpool UK 2010 Des techniques mixtes, qui sont une combinaison de ces deux types de techniques peuvent également être réalisées; le dispositif d'étanchéité suite à l'inefficacité du dispositif d'étanchéité, des fuites, etc. Des techniques de génie civil telles que les palplanches, les géomembranes etc. peuvent être mises en œuvre. Afin de faire le bilan des pratiques actuelles dans le domaine des réparations de digues et berges de voies navigables et d'en tirer des enseignements, le Centre d'Etudes Techniques Maritimes et Fluviales a réalisé avec les Centres d'Etudes Techniques de l'Equipement deux études de retour d'expériences de 2007 à 2009. Au total, quarante et un cas de réparations de digues et d'aménagements de berges ont été analysés. Cet article est une synthèse des deux études de retour d'expériences mené par le CETMEF d'une part sur les réparations de digues et d'autre part sur les aménagements de berges Des exemples de travaux sont décrits et une analyse des cas ainsi qu'un bilan du retour d'expériences sont donnés. 2. PRESENTATION DU RETOUR D'EXPERIENCES SUR LES REPARATIONS DE DIGUES ET AMENAGEMENTS DE BERGES 2.1. Présentation générale Le retour d'expériences mené par le CETMEF avec les CETE a donné lieu à la publication d'une notice consacrée aux aménagements de berges en 2009 (recensement de seize cas de retour d'expériences) et à un guide sur les réparations de digues en 2010 (recensement de vingt-cinq cas de retour d'expériences). Le lecteur pourra se référer à ces deux études pour avoir une description et une analyse détaillée de l'ensemble des cas de retour d'expériences. Les voies navigables sur lesquelles un retour d'expériences a été effectué lors de travaux d'aménagement de berge ou de réparation de digue sont : la Meuse (2003, 2005); la Petite Saône (2003, 2004) ; la Grande Saône (1998, 2005, 2006) ; la Dérivation de la Saône (1998, 2004, 2007) ; le canal des Vosges (2004, 2008) ; la rigole de l'Arroux (en 2004) ; le canal de Briare (2002) ; le canal de la Colme (2005) ; le canal des Houillères de la Sarre (1995) ; le canal Latéral à la Loire (2000, 2002) ; le canal de la Marne au Rhin (2001, 2004, 2006, 2007) ; le canal du Midi (2004) ; la Moselle canalisée (2002) ; le canal de Neufossé (2004) ; le canal du Nord (2003, 2008) ; le Rhin canalisé (2004) ; le canal du Rhône au Rhin (2005, 2006) ; le canal de Roanne à Digoin (2007), le canal de Saint-Quentin (2001) ; le canal de la Sambre à l'Oise (2003, 2004), la Seine (2007). Les ouvrages recensés lors du retour d'expériences étaient soit sans protection anti-batillage et sans réel dispositif d'étanchéité (c'était le cas des seize ouvrages ayant fait l'objet d'un aménagement de berges en génie végétal ou mixte) soit constitués par une ou plusieurs des structures suivantes (cas des vingt-cinq cas de digues étudiées) : cuvelage béton ; perré ou murette ; rideau de palplanches ; corroi d'argile. Dans le cadre des berges sans protection, les aménagements étaient réalisés à la suite d'anses d'érosion ou d'affouillements. Pour les digues, la proportion de réparations induites par des ruptures était importante, quelle que soit la structure d'origine. Parmi les seize cas d'aménagements de berges et vingt-cinq cas de réparations de digues ayant fait l'objet d'un retour d'expériences : les techniques en génie végétal ont été employées dans sept cas. Il s'agit de techniques telles que le tressage, le fascinage, le caisson végétalisé, les boudins ou plages d'hélophytes. les techniques en génie mixte ont été mises en œuvre dans neuf cas. Il s'agit de techniques végétales (lits de plançons, fascines, boudins, plages ou matelas d'hélophytes) associés à des enrochements, plaquettes calcaires, tunage, merlons ou matelas gabions ; les techniques traditionnelles de génie civil (palplanches, géomembranes,...) ont été mises en œuvre dans vingt-cinq cas (il s'agit de réparations de digues, nécessitant d'importants travaux de terrassements). Trois exemples de réparations et d'aménagements sont présentés dans la partie suivante : une réparation de digue avec un Dispositif d'Etanchéité par Géomembrane après une rupture (Digoin, 2007); un aménagement de berges avec des boudins végétalisés et des plaquettes calcaires suite à des problèmes d'anses d'érosion (Montigny, 2005) ; une réparation de digues avec des palplanches et des enrochements suite à des affouillements et à un déversement de soutènement (Blaringhem, 2001). 5 PIANC MMX Congress Liverpool UK 2010 2.2. Exemples a) Réparations par DEG d'une digue à Briennon sur le canal Roanne à Digoin - 2007 (Galiana, 2009) Le canal de Roanne à Digoin est une voie d'eau au gabarit Freycinet, qui a pour fonction l'alimentation du canal latéral à la Loire. En juin 2007, suite à un mécanisme d'érosion interne, une brèche totale s'est produite dans la digue située à trois kilomètres en aval du port de Briennon. La digue initiale, d'une hauteur d'environ douze mètres, d'une largeur de quatre mètres en crête et avec des pentes de talus de 2V/3H, était un remblai homogène avec une étanchéité en corroi d'argile et une défense de berges en pieux bois. Illustration 1: vue de la brèche (source VNF) Les réparations ont consisté en la reconstruction de la digue et en la mise en place d'un DEG sur une cinquantaine de mètres. Le phasage des travaux a été le suivant : reconstitution du corps de digue en graves; réalisation d'un noyau d'argile; réalisation d'une tranchée drainante côté amont du noyau argileux avec un exutoire vers la vallée de la Loire pour rabattre la nappe dans la digue; mise en œuvre d'un DEG constitué d'une couche support en grave 0/31,5, d'un géotextile non tissé, d'une géomembrane bitumineuse et d'un géotextile anti-poinçonnement. La géomembrane a été ancrée en sommet de digue dans une tranchée d'ancrage remplie de graves et fixée avec des épingles au fond de la tranchée. Les géotextiles ont été ancrés en sommet de digue en arrière des matériaux qui recouvraient l'ensemble géomembrane-géotextile. Cet ensemble a été recouvert de matelas gabions fixés en tête par des filins reliés à des pieux métalliques de deux mètres. Chemin de halage 0/31,5 mm calcaire M u r d' a r gi le Tranchée d'ancrage de la géomembrane 0/31,5 mmMatelas calcairegabions Graves 0/40 mm Pieu d'ancrage métallique Tranchée drainante Ensemble géotextilegéomembrane llustration 2: Coupe type du projet de réparations de la digue (source CETE Lyon) 6 PIANC MMX Congress Liverpool UK 2010 Le coût des travaux s'est élevé à environ 280 000 euros. Compte-tenu des contraintes d'exploitation et de l'apparition de la brèche au début de la saison estivale, les travaux se sont déroulés en deux phases avec la pose des matelas gabions lors de la période de chômage du canal en novembre de la même année. Pour la première phase qui a duré un mois et demi, les travaux se sont néanmoins déroulés à sec grâce à la mise en place d'un batardeau installé dans le canal au niveau de la brèche. Jusqu'à ce jour, les réparations effectuées se sont révélées pérennes. Illustration 3: Vue des matelas gabions après les travaux de réparations (source VNF) b) Aménagements de berges en boudins végétalisés et plaquettes calcaires sur la Meuse à Montigny - 2005 (Galiana, 2009) La digue de Montigny se situe en milieu rural sur le bief 55 de la Meuse. Elle est située entre la Meuse et le canal de dérivation d'amenée à l'écluse. Elle n'est pas impactée par les crues de la Meuse mais par les ondes de batillage de l'ordre de quarante centimètres causées par les bateaux de plaisance, dont le trafic s'élève à 3 000 bateaux par an. Afin de résoudre les problèmes d'anses d'érosion, des travaux d'aménagements de berges se sont déroulés sur la digue de Montigny en 2005. Illustration 4: Vue des anses d'érosion (source VNF) Les travaux, qui ont été réalisés sur un linéaire de deux kilomètres ont consisté en la mise en place d'une technique en génie mixte constituée de branches de saules, de boudins d'hélophytes et de plaquettes calcaires. La hauteur totale de la protection était de quatre mètres et la hauteur émergée de trois mètres. Ce type de technique a été choisi sur la base de critères liés aux conditions géométriques du site et à l'insertion paysagère et écologique. 7 PIANC MMX Congress Liverpool UK 2010 Illustration 5: Vue des travaux (source VNF) Les travaux ont été effectués sur une durée d'un mois en période de chômage et le phasage a été le suivant : débroussaillage de la digue; terrassement pour la mise en forme de la plate-forme destinée à recevoir les plaquettes calcaires et les boudins; pose de branchage anti-affouillements; mise en place des plaquettes calcaires et des boudins en géotextiles coco et remplissage de ces derniers à l'aide d'une mini-pelle; pose des couches supérieures et talutage au dessus de la protection. Le coût des travaux s'est élevé à 750 000 euros. Aujourd'hui l'objectif de la protection de berge est atteint, les hélophytes ont bien repris et résistent au batillage. Le caractère naturel, végétal du site est préservé. Malgré cette réussite, quelques difficultés ont été rencontrées : le mauvais positionnement du boudin de base en phase travaux car la hauteur d'eau réelle pratiquée par le gestionnaire était plus basse que la hauteur théorique. La berge a dû être reformée par réalisation d'un piquetage; des dégradations des jeunes plantations par les rats musqués sont apparues. Des pièges ont été ajoutés. Illustration 6: Vue de la digue après aménagements en 2008 (source CETMEF) 8 PIANC MMX Congress Liverpool UK 2010 c) Réparations par palplanches et enrochement d'une digue à Blaringhem sur le canal de Neufossé en 2001 Le canal de Neufossé est une voie d'eau à grand gabarit intégrée à la la liaison Dunkerque-Escaut. La 3 défense de berge initiale de la digue était un perré de type Sensée , constituée par un perré, d'une pente de 3V/1H côté berge, fondé sur pieux. Ce dispositif était renforcé par un rideau de palplanches de quatre mètres cinquante de longueur, mis en œuvre devant le perré, les deux étant liés par un couronnement béton. En 2000-2001, d'importants désordres se sont matérialisés, sur un linéaire de cent quarante mètres. L'érosion importante de la risberme en pied de soutènement a généré un déchaussement de la partie en fiche du soutènement, diminuant la butée de pied et conduisant à la ruine de l'ouvrage par déversement. Les travaux ont été réalisés en 2004 et ils ont consisté à la réalisation de deux dispositifs. Un rideau de palplanches de dix mètres de hauteur, tiranté et ancré sur un contre rideau a été mis en œuvre au droit de la zone principale du désordre. Et sur l'ensemble de la zone affectée, un remblai en marne et la pose d'un enrochement sur fascine et géotextile ont été réalisés afin de reconstituer le talus sous eau, en butée de pied du rideau de palplanches existant. Illustration 7: Profil type des réparations du talus sous eau (source VNF) Les travaux de battage du rideau de palplanches et la reconstitution du talus sous eau en enrochements ont été réalisés à partir de la voie d'eau et les travaux de terrassement et de finition à partir du chemin de contre-halage. Les principales difficultés rencontrées pendant les travaux ont été la présence d'une ligne haute tension gênant les opérations de levage et de battage des palplanches ainsi que le caractère délicat de la pose d'un géotextile sous eau. 3 Murette en béton reposant sur des pieux bois 9 PIANC MMX Congress Liverpool UK 2010 Illustration 8: Mise en œuvre du rideau de palplanches et du contre-rideau (source VNF) Les travaux ont duré trois mois et leur coût s'est élevé à 685 000 € TTC (dont fourniture de palplanches 241 000 €). Environ cinq cent mètres de berges ont été réparées et jusqu'à à ce jour, les réparations se sont révélées pérennes. 3. ETUDE DES CAS DE RETOUR D'EXPERIENCES 3.1. Analyse des désordres et des phénomènes en cause Parmi les quarante et un cas répertoriés lors des deux études de retours d'expériences sur les réparations de digues et les aménagements de berges, plus d'un tiers des cas étudiés correspondent à des affouillements ou anses d'érosion (dix-sept cas) et un autre tiers à une brèche de l'ouvrage, partielle (sept cas) ou totale (six cas). Les autres types de désordres ayant conduit à des travaux de réparation sont les fuites (quatre cas), les déversements de soutènement ou de palplanches (trois cas), les affaissements (deux cas), les fontis (un cas), les glissements (un cas). 4 3 17 4 Anses d'érosion ou affouillements Brèches Fuites Déversement de soutènement ou de palplanches Affaissement, fontis, glissement 13 Illustration 9: Répartition des 41 cas étudiés par type de désordres Les cas d'affouillements ou anses d'érosion étudiés dans le cadre du retour d'expériences ont 10 PIANC MMX Congress Liverpool UK 2010 systématiquement eu lieu sur des ouvrages non protégés. Cette pathologie est le résultat d'une érosion externe sur plusieurs années, due aux courants érosifs (notamment sur les rivières navigables comme sur la Meuse, la Seine ou la Saône), au phénomène de batillage et à la variation du niveau d'eau dans le bief concerné. Dans certains cas, la présence d'arbres sur le talus amont et la crête de digue a pu contribuer par leur chute à l'érosion externe de la digue. Les brèches, qui ont conduit systématiquement à une vidange partielle ou totale du bief (entraînant le plus souvent l'interruption complète du trafic fluvial) s'expliquent par des mécanismes d'érosion interne, faisant suite à une augmentation des débits de fuite entraînée par la dégradation du dispositif d'étanchéité, par une insuffisance de fichage du rideau de palplanches ou par la présence de points singuliers (aqueducs, changements de dispositifs d'étanchéité) ; d'affouillement en arrière du parement amont suite à la déstructuration de celui-ci : à une surverse du dispositif d'étanchéité (infiltration d'eau par la partie supérieure de la digue nonétanchéifiée). Les déversements de soutènement ou de palplanches ont été constatés dans quatre cas étudiés dans le retour d'expériences mais ils ont été directement à l'origine de réparations dans trois cas. Les mécanismes à l'origine de tels désordres sont divers : insuffisance du fichage du rideau de palplanches ; rupture de tirants d'ancrage de palplanches ; érosion de risberme en pied de soutènement amont (liée notamment à des manœuvres de bateaux) ; effort de poussée associée à une mise en charge de la digue du côté aval suite à une surverse vers le canal pendant un épisode de crue d'une rivière proche. Les fontis ont été recensés dans quatre cas étudiés dans ce retour d'expériences. Dans un cas, ce désordre a été directement à l'origine des réparations et dans les autres cas, les fontis observés étaient connexes à d'autres désordres (fuites, affaissements) ayant occasionné les travaux de réparation. Les mécanismes identifiés à l'origine des fontis ont été : la présence d'ouvrages traversants, qui peut favoriser l'entraînement de matériaux du corps de digues, et occasionner un effondrement localisé du chemin de halage ; l'affouillement de la digue, qui peut conduire à un entraînement progressif des matériaux du corps de digue en arrière du dispositif d'étanchéité mis en place sur le talus amont et provoquer ainsi un affaissement du chemin de halage et des effondrements ponctuels ; la présence dans le sol support de formations sensibles à la dissolution, qui peut conduire à la création de cavités en profondeur, conduisant à la décompression des sols, et à un affaissement voire un effondrement du chemin de halage. Un seul cas de glissement a été identifié dans ce retour d'expériences et il s'est produit sur le talus aval de la digue suite à des travaux (mise en place d'un collecteur en bordure de chemin de halage et reprofilage du contre-fossé) et à des fortes précipitations. Parmi les cas étudiés dans le retour d'expériences, quatre cas de fuites et deux cas d'affaissement de chemin de halage ont directement conduit à des travaux de réparations. Ces désordres sont en fait souvent associés à d'autres désordres, dont ils apparaissent comme des phénomènes précurseurs (les brèches sont quasi-systématiquement précédées de fuites). Les fuites sont généralement associées à un vieillissement et à une dégradation progressive du dispositif ou à un défaut d'étanchéité de la digue, cette étanchéité pouvant être réalisée sur le talus amont (perré en pierres jointoyées ou cuvelage béton) ou dans le corps de digue (corroi d'argile ou palplanches). Les affaissements de chemin de halage sont accompagnés de fissurations et peuvent être associés à une dégradation du soutènement, à une faiblesse du sol support ou à la présence de cavités en profondeur. In fine, le retour d'expériences a permis de montrer que les digues les plus rigides, c'est à dire constituées d'un perré, une murette ou un cuvelage béton étaient les plus affectées par des ruptures et les digues les plus souples disposant d'un corroi d'argile ou d'enrochements étaient celles permettant le plus d'interventions préalables. Les digues présentant une discontinuité dans la technique de protection de berges et d'étanchéité d'origine, que cette discontinuité soit induite par la nature même de la technique 11 PIANC MMX Congress Liverpool UK 2010 (perrés et murettes) ou par des techniques mixtes, ont très majoritairement été réparées suite à la rupture du corps de digues causée systématiquement par un mécanisme d'érosion interne. 3.2. Analyse des aménagements de berges et des réparations de digues 9 9 Techniques génie mixte Techniques génie végétal 6 7 Techniques génie civil - géomembranes ou DEG Techniques génie civil - enrochements ou autres techniques (béton, tunage, remblai) Techniques génie civil - palplanches 11 Illustration 10: Répartition des 41 cas étudiés par type de techniques de réparation ou d'aménagement La figure suivante représente la proportion des techniques employées. a) Cas d'aménagements de berges avec des techniques génie mixte ou végétal Les techniques en génie mixte et génie végétal recensés dans le retour d'expériences ont été employées dans le cadre d'interventions programmées d'aménagements de berges. Les enjeux étaient généralement bien identifiés (protection de chemin de service, d'une route, préservation du chenal navigable,...) mais des lacunes existaient dans les diagnostics techniques et environnementaux du site, qui auraient pu être davantage développés et formalisés avant les travaux sur les aspects suivants: les actions hydrauliques et hydrodynamiques (connues sommairement) ; la géométrie de la voie d'eau, les flux dynamiques et la morphologie de la berge (des mesures étaient trop souvent réalisées au moment du calage de la protection plutôt qu'en amont pour comprendre les phénomènes) ; les caractéristiques écologiques de la berge (inventaire d'espèces non systématique) et les particularités d'usage. Dans la moitié des cas, il n'y a pas eu d'interruption de navigation. Pour les autres cas, la mise en œuvre de la technique d'aménagement a nécessité un chômage du bief ou un abaissement de la ligne d'eau avec interruption de la navigation. Les seize aménagements de berges ayant fait l'objet du retour d'expériences se sont révélés pérennes jusqu'à ce jour malgré quelques problèmes aléatoires qui sont apparus mais n'ont pas remis en cause la pérennité de la protection. Ces problèmes ont pu être : la mauvaise fixation de la protection en pied de berge, le mauvais calage de la protection par rapport au niveau d'eau réel du bief (réajustement en phase travaux) ; le développement d'espèces végétales invasives telles que les fougères aquatiques ; les dégâts occasionnés par les animaux fouisseurs (ragondins ou rats musqués) ; les dégâts liés aux activités humaines (piétinement des hélophytes par exemple). 12 PIANC MMX Congress Liverpool UK 2010 La durée moyenne des aménagements était d'un mois et demi (variable d'une semaine à quatre mois dans les cas étudiés). Le coût moyen par mètre linéaire des aménagements de berges étudiés a pu être estimé à 200-250 euros (incluant le coût des travaux préparatoires). b) Cas de réparations de digues avec des techniques en génie civil Les techniques en génie civil étudiées ont été mises en œuvre dans le cadre de réparations de digues. Le rideau de palplanches, souvent associé à une autre technique (injections, corroi d'argile ou pieux bois), a été la technique majoritairement mise en œuvre, suivi par les géomembranes. Les autres techniques ont consisté à la mise en place d'enrochement, de pieux bois, de tunage, de béton ou d'injections. Les réparations en urgence ont concerné quinze cas et dans la moitié des cas, des palplanches ont été mises en œuvre. Que la réparation fût programmée ou urgente, la technique mise en œuvre, dans environ la moitié des cas, a été identique à une technique déjà existante sur la digue. Les digues constituées initialement de rideaux de palplanches ont toutes été réparées par une technique similaire avec un rideau de palplanches mieux adapté. Les techniques de réparations par géomembranes associées ou non à un système de protection (Dispositif d'Etanchéité par Géomembrane) ont été majoritaires pour les réparations de digues en cuvelage béton. Il est apparu que la réalisation d'études géotechniques était assez rare lors d'interventions en urgence et les phases d'études étaient souvent délaissées au profit de la phase travaux mettant en œuvre des techniques simples (procédés rapides et usuels). Le diagnostic était souvent établi sur la base d'une unique inspection visuelle sans investigation géotechnique spécifique. Les études géotechniques étaient néanmoins plus fréquentes dans le cadre de travaux complémentaires. Dans la moitié des cas recensés, les réparations ont eu lieu sans vidange du bief mais avec parfois des restrictions de navigation ou une signalisation auprès des mariniers. Lorsque les travaux étaient liés à une brèche, ils étaient systématiquement réalisés à sec et ils nécessitaient la mise en place de batardeau en amont pour éviter la vidange complète du bief. Les réparations se sont révélées pérennes sauf pour quatre cas étudiés où d'autres pathologies sont apparues. Le désordre constaté était une persistance des fuites, traduisant un défaut d'étanchéité de la solution de confortement et qui pouvait être associé à : une mauvaise mise en place du DEG lorsque la géomembrane était mal plaquée sur le talus amont de la digue ou lorsque la géomembrane était mal encastrée lors de la réparation ; l'insuffisance de l'étanchéité lors de la mise en place du rideau de palplanches sans autre technique d'étanchéité. une fissuration du béton de protection ou à l'inefficacité du colmatage des fuites à l'aide de floculant. La durée moyenne des réparations était d'un mois (variable d'une semaine à six mois dans les cas étudiés). La durée moyenne pour les travaux de première urgence excèdaient rarement une semaine. Les coûts de réparations sont fonction de la technique employée, du délai de réalisation et des sujétions propres aux travaux (conditions d'accès et de mise en oeuvre...). A titre indicatif, le coût moyen par mètre linéaire de travaux de reconstruction de digues géomembranes a pu être estimé à 5000 - 6000 euros et des palplanches à 10 000 euros. 4. BILAN DU RETOUR D'EXPERIENCES 4.1. Enseignements sur les désordres Plusieurs “niveaux” de désordres sont à distinguer, essentiellement en fonction du degré de ruine de l'ouvrage : les désordres précurseurs (fuites, fissurations en crête de digue, affaissements), détectables lors des visites de surveillance de l'ouvrage. S'ils sont traités à temps, ils permettent d'éviter des désordres plus importants ; 13 PIANC MMX Congress Liverpool UK 2010 les désordres intermédiaires ou de transition (déversements de soutènement, glissement, fontis, anses d'érosion, renard), généralement associés à un(des) désordre(s) précurseur(s), pouvant conduire à une ruine totale de l'ouvrage si aucun travaux de confortement n'est entrepris à brève échéance ; la rupture de l'ouvrage, partielle ou totale (ruine ultime), nécessitant une reconstruction complète sur le linéaire de la brèche. Les structures les plus rigides sont affectées préférentiellement par des ruptures spontanées ne permettant pas d'intervention préalable ce qui peut s'expliquer en partie par la cinétique de rupture globalement rapide de ce genre de technique de protection et d'étanchéité d'origine (perré, murettes et cuvelages béton) et le sentiment de sécurité qu'elles peuvent conférer. A l'opposé, les ouvrages constitués principalement de structures plus flexibles en matériaux graveleux, sableux, ... ou en corrois d'argile sont toutes affectées par des mécanismes lents et sont réparées préférentiellement alors qu'elles sont atteintes par des désordres précurseurs ou intermédiaires. Cela peut s'expliquer par la cinétique globalement lente des mécanismes affectant ces structures relativement à l'organisation d'un chantier. De façon générale, la présence de points singuliers (dans la digue, dans le sol support), l'ancienneté des ouvrages, la végétation, les activités humaines (circulations d'engins de chantier par exemple) et les conditions climatiques (fortes précipitations ou gel) sont des facteurs contribuant à l'aggravation des désordres. 4.2. Enseignements sur les réparations de digues Le choix du type de réparation est souvent lié au degré d'urgence, au besoin de rétablir la navigation, à la possibilité de vider le bief partiellement ou complètement, à l'accessibilité et aux crédits. Le type de réparation le plus largement utilisé pour les digues est le rideau de palplanches, ce qui s'explique par ses facilités et rapidité de mise en œuvre, mais également par ses fonctions diverses de réparation de corps de digues, défense contre l'érosion des berges, et contribution à l'étanchéité des ouvrages. Il est apparu qu'en l'absence de technique complémentaire ou d'étanchéité des serrrures (absence de joints), les rideaux de palplanches ne garantissaient pas systématiquement l'étanchéité des ouvrages. La mise en place d'une géomembrane ou d'un DEG (souvent recouverte par un béton ou par un remblai avec protection par matelas gabion), est également une technique largement employée pour répondre à des problèmes de réparation d'ouvrage suite à une brèche et de réparation de dispositif d'étanchéité. Cependant, la mise en œuvre d'un DEG est délicate et peut s’avérer inefficace suite à une fixation insuffisante de la géomembrane ; un mauvais traitement des raccords ; une détérioration de la géomembrane pendant les travaux ; une couche de protection inefficace (fissuration du béton ou glissement des matériaux constituant les matelas gabions). Les travaux de réparations de digues ont souvent lieu dans une situation d'urgence. Les délais et la durée des réparations sont directement liés au choix de techniques de confortement, aux sujétions d'exécution propres au site et aux procédures administratives de choix et de contrôles des entreprises. Dans de telles situations d'urgence, le diagnostic ainsi que les contrôles de chantier sont en général succincts. Le critère de vitesse de mise en œuvre étant essentiel, la solution technique apportée entraine souvent la nécessité d'une intervention en deux phases, via l'engagement de travaux complémentaires. 4.3. Enseignements sur les aménagements de berges Le retour d'expériences a montré que lorsqu'une protection de berge a été pensée, adaptée et mise en œuvre au regard des forces et contraintes susceptibles de l'éroder, des réponses adaptées et concluantes ont été trouvées. Dans tous les cas réussis de protection en technique mixte ou végétale, il est apparu qu'il y avait une connaissance des techniques de base (tressage, tunage, fascines, plages, peigne, lits de plançons, caissons...), des caractéristiques des matériaux pouvant être utilisés (géotextiles synthétiques, toile coco, intissé...) et des végétaux (hélophytes, hydrophytes, caractéristiques racinaires, vivaces, annuelles...) pour les combiner au mieux et choisir les composants qui répondaient au mieux à l'érosion et au contexte (courants plus ou moins forts, batillage plus ou moins important, crues plus ou moins régulières...). Les critères à l'origine de la recherche d'une solution de protection de berges par technique mixte ouvégétale ont été l'urgence des travaux (lorsque le gestionnaire peut planifier ses travaux de protection 14 PIANC MMX Congress Liverpool UK 2010 de berges sans pression due à l'urgence, il s'avère que les techniques mixtes ou végétales seront plus facilement recherchées) ; les moyens à disposition du gestionnaire (en terme de délais d'étude, de mise en œuvre, et principalement d'ouvertures de crédits nécessaires en volume et en délais associées aux disponibilités foncières) ; la connaissance de l'environnement du site (lorsque le technicien appréhende les enjeux environnementaux des berges et possède la connaissance des apports et améliorations possibles par les techniques végétales, ce type de protection est souvent recherché voire privilégié ; la sensibilité aux problématiques écologiques (certains gestionnaires privilégient le rôle d'infrastructure de la voie d'eau, et n'utilisent que les techniques issues du génie civil, les critères environnementaux ne venant qu'en arrière plan) ; la volonté du gestionnaire (et surtout les directives sur le mode de protection des berges). 4.4. Propositions d'améliorations Une surveillance systématique et régulière (notamment en période hivernale) des ouvrages et des points singuliers est nécessaire pour repérer tout désordre précurseur, évaluer leurs évolutions vers des désordres de plus grande ampleur, et prévoir les travaux adéquats. La fréquence de surveillance doit dépendre de l'emplacement de l'ouvrage et des enjeux qui seront associés (par exemple la présence d'habitations ou d'infrastructures) ; des caractéristiques et de l'ancienneté de l'ouvrage ; de l'état des protections et du niveau des désordres. La surveillance doit être basée à la fois sur l'inspection visuelle dont l'objectif est de déceler des anomalies perceptibles à l'œil et sur l'auscultation qui permet de mesurer l'évolution de certains paramètres et d'établir une analyse du comportement de l'ouvrage sur le long terme (Perez, 2004). L'inspection visuelle doit porter particulièrement sur les points suivants : repérage des zones humides, de suintements et de fuites sur le talus aval, le pied de la digue, le contre-fossé ; repérage des fontis dans la cuvette, sur les digues ou dans les environs ; repérage des zones de talus instables particulièrement en pied de talus, côté contre-canal et en crête côté contre-canal et côté canal ; localisation de zones de tassements et retraits, présence de zones déformées ou de faible densité en surface (résultat de mécanisme d'érosion interne) ; suivi des points singuliers (joints d'étanchéité, de jonctions entre des protections de berges ou des dispositifs d'étanchéité différents, zones de raccordement entre les remblais et des zones rigides, traversée d'ouvrages comme les aqueducs). Dans la pratique, il s'est avéré que les réparations d'urgence de digues de voies navigables étaient plus nombreuses que les réparations programmées. Le principal problème rencontré dans les situations d'urgence est l'impératif de réactivité. Il est souvent nécessaire de rétablir partiellement ou complètement le fonctionnement de l'ouvrage, au plus vite (protection des personnes, enjeux financiers...) et plus la situation est extrême, moins de temps est laissé pour la réalisation des études préalables au confortement. Ce qui peut engendrer des problèmes de sécurité, de pérennité de l'ouvrage mais également de surcoût de travaux. Pour prévoir au mieux les situations d'urgence et les gérer de manière satisfaisante, il convient de s'y préparer en amont. Il paraît important, pour une gestion optimale de la crise d'avoir établi un plan de gestion des situations d'urgence, pour permettre une mise en œuvre rapide, efficace, répondant au mieux à la problématique et aux questions suivantes : comment réaliser un bilan efficace du sinistre ; comment choisir au mieux l'entreprise de travaux ; comment contrôler la mise en œuvre de la solution de confortement. Avant toute opération de travaux, il est aussi nécessaire d'effectuer un état des lieux complets de l'ouvrage et du site : recueil des informations disponibles à partir du dossier et registre d'ouvrages ; diagnostic technique et environnemental du site ; définition des enjeux (identification des contraintes du site, des besoins et exigences du gestionnaire) ; campagne de reconnaissances adaptées au projet (reconnaissances de type géophysique et/ou reconnaissances géotechniques in situ ; essais géotechniques ou mécaniques en 15 PIANC MMX Congress Liverpool UK 2010 laboratoire ; essais d'eau). Les connaissances précises de l’ouvrage et de son environnement permettront aux entreprises de travaux de définir au mieux les techniques à employer, le coût et le phasage des travaux, afin de réduire à la fois les risques (sécurité des personnes et tenue de l’ouvrage) et la durée du chantier (Poligot-Pitsch et al., 2006). De manière générale, la phase de terrassement est capitale pour tous les travaux et il sera nécessaire de vérifier la qualité géotechnique des matériaux mis en œuvre par les entreprises mandataires des marchés de travaux en s'appuyant sur les critères du Guide des Terrassements Routiers (GTR). Cette vérification devra être réalisée sur la nature des matériaux qui devront être aptes à l'utilisation souhaitée et sur leur mise en œuvre sur site notamment au niveau du compactage. Des précautions particulières devront être prises pour le raccordement avec les ouvrages existants de manière à obtenir la meilleure transition possible entre les partie anciennes et les parties neuves du canal. Dans le cadre de travaux d'aménagements de berges, la mise en place de protections en génie végétal ou génie mixte nécessitera des méthodes particulières (travail en eau ou travail de tressage de saule par exemple, ce qui est peu fréquent pour les entreprises de travaux), des connaissances de base en botanique, et un savoir-faire propre à ces techniques (par exemple, tous les éléments doivent être correctement fixés, les géotextiles agrafés). Dans le cadre de réparations de digues, la solution mise en œuvre devra être conçue afin de prévoir (CETMEF, 2002) : le contrôle des écoulements par répartition des zones de perméabilité croissante de l'amont vers l'aval, c'est à dire de la cuvette jusqu'aux talus extérieurs : l'étanchéité peut être assurée par le revêtement de la cuvette (perré, revêtement en béton, géomembrane protégée,...) ou bien par la digue proprement dite, par utilisation d'un matériau étanché (limon ou argile) ; la limitation des gradients hydrauliques, en deçà du « gradient hydraulique critique » pour le matériau de remblai. En l'état actuel des connaissances, on ne dispose pas de règles consensuelles sur ce sujet ; le compactage des remblais, de façon à garantir un bon serrage des matériaux ; l'utilisation de filtres et drains pour contrôler les écoulements internes et maîtriser les risques d'érosion du remblai ou de la fondation. La mise en eau se fera par paliers successifs, (leur hauteur unitaire étant faible : cinquante centimètres à un mètre) et lors de chaque palier, il conviendra de s'assurer de l'absence de désordre par des observations visuelles et des mesures piézomètriques et de débit dans les contre-canaux (Brygo, 1979). Lorsque les réparations ou les aménagements mis en place seront différentes de la structure d'origine, il faudra garder à l'esprit que cela induit généralement la création de points singuliers sur lesquels la surveillance devra être accrue. Aussi, après des travaux de réparations de digues, il se peut que visuellement aucun point singulier ne soit créé. Toutefois, il sera important de surveiller les zones frontalières à la réparation car les usages, compositions et propriétés des systèmes de protection seront différents et ces zones seront malgré leur apparence des zones singulières. Le retour d'expériences n'a pas pu évaluer de désordres très anciens, ce qui pose la question de l'archivage des données relatives à ce type d'événement. Il est donc recommandé de procéder à un archivage systématique de tous les documents relatifs aux désordres et aux opérations de réparations (dossier de récolement, photos...). 5. CONCLUSIONS Suite au retour d'expériences réalisé par le CETMEF avec les CETE, des données ont été recueillies pour seize cas d'aménagements de berges et vingt-cinq cas de réparations de digues de voies navigables ayant eu lieu ces dix dernières années et des enseignements ont été tirés. Plusieurs niveaux de désordres avec des cinétiques variables sont à distinguer : des désordres précurseurs (fuites, fissurations en crête de digue, affaissements), des désordres intermédiaires (déversements de palplanches, glissement, fontis, anses d'érosion, renard) et des désordres nécessitant une reconstruction de l'ouvrage (brèche partielle ou totale). Ces désordres peuvent apparaître suite à plusieurs facteurs : ancienneté de l'ouvrage, présence de point singulier, conditions climatiques défavorables (fortes 16 PIANC MMX Congress Liverpool UK 2010 précipitations ou gel) etc... Les réparations de digues ont souvent lieu dans l'urgence à la suite de brèches ou de déversement de palplanches avec comme impératif le maintien de la navigabilité. Quelles que soient les structures d'origine et les désordres, les réparations par palplanches sont majoritaires suivi des géomembranes. Pour les aménagements de berges, qui font généralement l'objet de travaux programmés, il est constaté que les techniques végétales et mixtes, encore peu utilisées il y a quelques années, sont de plus en plus utilisées par les services de VNF. Lorsqu'ils sont bien réalisées, ces aménagements de berges et ces réparations de digues sont efficaces en rivière mais aussi en section canalisée, petit ou grand gabarit. Cependant, il est à souligner que le recul n'est que de quelques années (entre cinq et dix ans en moyenne) pour les cas ayant fait l'objet d'un retour d'expériences dans le cadre de cette étude. Aussi, ce retour d'expériences doit être poursuivi à la fois sur les digues et les berges ayant déjà fait l'objet d'une étude mais aussi sur d'autres ouvrages. Afin de prévenir au mieux les désordres les plus préjudiciables pour les digues et berges de voies navigables, il convient d'assurer une surveillance régulière de ces ouvrages et de ne pas négliger leur suivi et leur entretien après des travaux. La réussite de toute opération de travaux passera par une définition approfondie des buts poursuivis, un diagnostic complet de site, une conception et une mise en œuvre rigoureuses. Pour y parvenir, les compétences, l'expérience et le savoir-faire du maître d'ouvrage, de son maître d'œuvre et de l'entreprise sont fondamentaux. 6. REFERENCES BRYGO Y. (1979), Les digues des voies navigables, STCPMVN, 65p. CETMEF (2002), Ouvrages de navigation et écoulements souterrains prévention et réparations, 144p. phénomènes, surveillances, CETMEF (2009), CIRIA, CUR, guide enrochement, 2009, 1340p. CETMEF, CETE Lyon (2009), Aménagement des berges des voies navigables – Retour d'expériences , 93p. CETMEF, CETE Lyon, Nancy, Strasbourg, Saint-Quentin (2010), Digues des voies navigables - retour d'expériences sur les désordres et les réparations. GALIANA M.(2009), Retour d'expériences sur l'utilisation des géosynthétiques dans les réparations de digues et berges de voies navigables, Rencontres géosynthétiques, Nantes, CFG, pp269-276. PEREZ M. (2004), La maintenance des digues et canaux, Actes du colloque « Sécurité des digues fluviales et de navigation », Orléans, CFGB, pp47-55. POLIGOT-PITSCH S., DALY F. (2006) Modification et réparation d'écluses ou de barrages de navigation, AIPCN, 21p. 17 PIANC MMX Congress Liverpool UK 2010 SQUAT EN SEINE : CONDITIONS DE SECURITE ACTUELLES ET IMPACT DE L’APPROFONDISSEMENT 1 by A. Gallicher Lavanne and P. Tournier 2 ABSTRACT Le Port de Rouen a pour projet d’approfondir son chenal d’accès. Dans ce cadre, une étude a été réalisée pour vérifier les conditions de navigation actuelles notamment liées aux tirants d’eau. Une campagne de mesures menée en 2008 a permis d’estimer, d’une part, la variation de tirant d’eau lors du transit due au phénomène de squat et d’autre part, par comparaison avec les formules issues de la littérature, d’estimer l’impact du projet sur ce phénomène. Les résultats de ces mesures et de leur exploitation valident les conditions actuelles de navigation ainsi que le bon dimensionnement du projet. 1. CONTEXTE ET OBJET DE L’ETUDE La présente étude a été réalisée dans le cadre du projet d’amélioration des accès maritimes du Grand Port Maritime de Rouen qui permettra d’augmenter la capacité nautique du chenal de navigation de 1 mètre en faisant passer les tirants d’eau maximaux de 10,30 m à la descente et 10.70 à la montée à Rouen à respectivement 11,30 m et 11.70. Les mesures présentées dans ce document ont pour objectif de valider l’adéquation des pratiques de navigation au gabarit du chenal actuel en évaluant en particulier le surenfoncement dynamique des navires et leur clair sous-quille effectif lors des transits dans le chenal d’accès à Rouen. Cette évaluation a permis de mettre en évidence les formules théoriques d’évaluation du surenfoncement dynamique les plus pertinentes pour le chenal de Rouen et d’en déduire les valeurs probables de surenfoncement des plus gros navires appelés à transiter dans le chenal approfondi. Ainsi, les profondeurs du chenal théorique défini dans le programme d’amélioration des accès nautiques du port de Rouen ont pu être validées en termes de clair sous quille disponible dans chaque section du transit. Pour le chenal d’accès à Rouen, ce sont les pilotes de Seine qui fixent au jour le jour les tirants d’eau maximaux de montée et de descente, en fonction des conditions bathymétriques et météorologiques du jour ainsi que du type de navire. Ces tirants d’eau intègrent un « pied de pilote » permettant de prendre en compte les variations dues à la houle (tangage, roulis, pilonnement) ainsi que l’incertitude sur la connaissance des fonds (sondage et variations naturelles). Le tirant d’eau d’un navire se déplaçant varie en fonction de nombreux facteurs dont, notamment: son clair sous quille, la largeur du chenal, sa vitesse, et les courants rencontrés. Cette variation est appelée « squat », venant de l’anglais to squat, « s’accroupir ». Ce terme désigne le surenfoncement dynamique d’un navire par rapport à son tirant d’eau statique, mesuré à l’arrêt. Les mesures présentées dans la suite de ce document ont été réalisées dans des conditions réelles d’exploitation sur des navires transitant avec de forts tirants d’eau en Seine. La démarche du Port de Rouen est constituée de cinq étapes : Réalisation d’une campagne de mesures sur des navires ciblés, Exploitation des résultats issus de l’analyse des données collectées, Comparaison des résultats avec les formules issues de la littérature, Synthèse et conclusions pour les conditions actuelles de navigation en Seine, Extrapolation des résultats aux conditions futures de navigation dans le chenal approfondi. 1 Ingénieur d’étude, Grand Port Maritime de Rouen, France, [email protected] Chef du service du chenal et des aides à la navigation, Grand Port Maritime de Rouen, France, [email protected] 2 1 of 18 PIANC MMX Congress Liverpool UK 2010 La campagne de mesures ciblée sur des navires approchant le tirant d’eau maximum admissible a été réalisée lors de différents coefficients de marée (marée moyenne, mortes eaux, vives eaux), à bord de navires montant et descendant ainsi que sur différents types de bateaux et donc différentes formes de carène (porte-conteneurs, vraquiers, pétroliers, Ro-Ro). 2. PRESENTATION DU DISPOSITIF DE MESURE 2.1 Principe de la mesure Pour effectuer avec une précision suffisante la mesure de squat, deux stations de mesures GPS différentiels ont été utilisées, l’une étant placée à l’avant du navire et l’autre à l’arrière afin de mesurer les variations des différents tirants d’eau ainsi que de l’assiette. La photo ci donnée en Figure 1 montre le dispositif de mesure. Figure 1 : Photographie de l'installation du dispositif de mesures Les mesures sont prises par pas de temps de 10 secondes. Cet intervalle entre deux mesures est issu d’un compromis entre la nécessité d’avoir une densité de points la plus importante pour s’assurer d’obtenir une bonne représentativité du phénomène squat et la contrainte matériel. Le trajet de Rouen à la mer étant long, près de 130 km et 7 h, le nombre de données enregistrées devient rapidement très important avec la fréquence d’acquisition. De plus, la fréquence de mesure influe directement sur l’autonomie de la batterie. Le matériel utilisé est le récepteur GPS Trimble 5700, couplé avec un carnet de terrain. Le système GPS permet d’obtenir avec l’intégration des corrections dues aux perturbations atmosphériques une incertitude de ± 5 cm. Ces corrections sont mesurées en permanence par le port aux quatre stations de référence situées le long de la Seine et permettent de couvrir l’ensemble du chenal d’accès, de Rouen à l’Engainement. Ces corrections sont transmises directement aux récepteurs GPS par signal radio et intégrées à la mesure automatiquement. 2.1.1 Variation du plan d’eau due à la marée Le Port dispose d’un réseau de 22 marégraphes répartis régulièrement le long de la Seine. Cette connaissance de la marée en Seine permet de calculer une marée « locale » en chaque point de mesure par interpolation géographique et chronologique des mesures de hauteur d’eau fournies par pas de 5 minutes. 2 of 18 PIANC MMX Congress Liverpool UK 2010 2.1.2 Variation du tirant d’eau due à la salinité La salinité varie pendant la montée en Seine, depuis l’engainement où l’eau est salée en permanence jusqu’à Aizier (PK 324), limite extrême de remontée du front de salinité. Cette variation entraîne un changement de densité de l’eau et, par conséquent, du tirant d’eau moyen du navire. Ce changement de densité s’accompagne souvent d’une modification de l’assiette du navire, cette modification variant en fonction de la forme de la carène et de son chargement. Le dispositif de mesure utilisé ne permet pas de dissocier dans la partie aval, le déjaugeage dû à un changement de densité de l’eau de la modification de l’assiette du navire et du squat. Aussi, dans la suite de l’étude, le terme « squat » comprendra la variation du tirant d’eau due à ces trois phénomènes concomitants et correspondra donc à la variation instantanée de tirant d’eau constaté lors du transit. 2.2 Navires étudiés Les navires ont été choisis en fonction de leur tirant d’eau proche des capacités du chenal et de leur forme de carène. Plusieurs types de navires ont pu être étudiés : Porte-conteneurs, Rouliers, Vraquier (Céréaliers, Pétroliers et Minéralier). Ce choix a fait l’objet d’une concertation avec les Pilotes qui ont aidé à cibler au mieux les navires présentant un fort potentiel de squat. Les conditions de navigation (vitesse) et hydrodynamiques (hauteur d’eau et courant) rencontrées en Seine étant différentes pour les navires montant et descendant à Rouen, ces deux types de parcours ont été étudiés. Le Tableau 1 synthétise les caractéristiques des navires étudiés, le sens montant ou descendant et le coefficient de marée. Nom Type Longueur Largeur Trajet Coefficient Marée Virana Delmas Ro-Ro 182.51 32.26 Montée 77 Lucie Delmas Ro-Ro 196.5 32.29 Montée 62 Bro Juno Pétrolier 119.92 20.99 Montée PJ 74 Maersk Rosyth Pétrolier 171.2 27.43 Montée PJ 52 Stones Minéralier 166.3 24.75 Montée 46 Richard Maersk Pétrolier 171.2 27.4 Descente 70 Kotor Céréalier 175.01 26.04 Descente 48 Patent Céréalier 176.86 29.4 Descente 55 30.2 Descente 100 Fort Saint Louis Porte conteneurs 197.2 Tableau 1 : Liste des navires étudiés et leurs caractéristiques 2.3 Principaux facteurs influençant le squat 2.3.1 Vitesse surface du navire La principale cause du squat est la vitesse du navire par rapport à la surface. Le squat croit très vite avec la vitesse, les deux étant liés par une loi en puissance de deux ou plus. Il faut également noter que même un navire à quai squatte dès lors qu’il est soumis à un courant de marée par exemple. Par contre, dès lors qu’il transite, le phénomène apparaît quelles que soient les conditions, en canal ou en pleine mer. La Figure 2, issue de nos mesures, illustre bien cette dépendance. 3 of 18 PIANC MMX Congress Liverpool UK 2010 Figure 2 : Influence de la vitesse surface sur le squat 2.3.2 Clair sous quille Pour un navire à vitesse donnée, le paramètre essentiel faisant varier le squat est le clair sous quille. En effet, plus la hauteur d’eau disponible est faible, plus le squat augmente. Lorsque la profondeur est faible, l’écoulement sous le navire se trouve accéléré sur toute la hauteur d’eau disponible sous la quille, ce qui entraîne une dépression amplifiée et donc un squat plus important. Parmi de nos mesures, le Patent nous permet d’apprécier clairement cette influence sur la partie aval de l’estuaire. Cette influence est représentée par la relation entre le squat et le rapport H/T, hauteur d’eau sur tirant d’eau. Les résultats sont présentés en Figure 3. Figure 3 : Influence du clair sous quille sur le squat 4 of 18 PIANC MMX Congress Liverpool UK 2010 3. PRESENTATION DES RESULTATS 3.1 Résultats présentés Les résultats issus des mesures et de leur traitement font apparaître plusieurs paramètres. Ces paramètres sont : le squat déterminant de chaque navire, la vitesse surface du navire au cours de son trajet et le clair sous quille. Ces paramètres sont présentés dans cette partie avant de voir les résultats proprement dits. 3.1.1 Choix du squat déterminant Pour chaque navire, les squats avant et arrière ont été mesurés. Parmi ces deux mesures, le squat déterminant est celui qui influe directement sur le tirant d’eau maximum. Il dépend des tirants d’eau avant et arrière ainsi que de leurs variations respectives. Le Tableau 2 présente les tirants d’eau observés ou mesurés en eau douce (à Rouen essentiellement ou à Port Jérôme) et le squat pris en compte pour la variation du tirant d’eau maximum. Navire TE avant TE arrière Squat déterminant Virana Delmas 7.4 8.26 Arrière Lucie Delmas 8.15 10 Arrière Bro Juno 9.75 9.49 Avant Maersk Rosyth 9.5 9.5 Avant Stones 10.40 10.55 Avant Richard Maersk 10.3 10.3 Avant Kotor 9.95 9.95 Avant Patent 9.73 9.82 Avant Fort St Louis 7.8 9.5 Arrière Tableau 2 : Tirant d’eau des navires et squat déterminant Remarque : Les mesures sont conformes à ce que prévoit la littérature : le squat déterminant dépend du type de navire, arrière pour les rouliers et porte-conteneurs (navires fins) et avant pour les vraquiers (navires pleins de formes). Il dépend également de l’assiette statique des navires mais, du fait que les rouliers et porteconteneurs ne sont chargés que partiellement cette influence peut être négligée. 3.1.2 Vitesse surface Le squat dépendant fortement de la vitesse, les résultats font également apparaître la « vitesse surface » du navire le long de son trajet. La « vitesse surface » se définit comme la vitesse du navire par rapport à l’eau. C’est cette vitesse qui est importante pour appréhender le squat. Pour l’obtenir à partir de nos mesures GPS donnant la vitesse sur le fond, nous avons utilisé un modèle hydrodynamique TELEMAC 2D, co-développé par le GPMR et le bureau d’étude SOGREAH. Ce modèle nous permet d’obtenir en tout lieu et pour tout coefficient de marée la vitesse du courant rencontré par le navire. 3.1.3 Choix de la sonde Afin de connaître le plus précisément possible le clair sous quille de chaque navire, la trajectoire issue des mesures a été étudiée et les sondes correspondantes extraites des levés bathymétriques du GPMR. De cette manière, il a été possible d’appréhender au mieux la pratique du chenal par les pilotes qui optimisent leur navigation en fonction des sondes contraignantes du chenal. 5 of 18 PIANC MMX Congress Liverpool UK 2010 La sonde relevée correspond à la sonde maximale présente sous le navire. Elle diffère de la sonde moyenne sous la quille qui représente le fond perçu par le navire. 3.1.4 Clair sous quille Afin d’évaluer la marge de sécurité sur les profondeurs effectivement disponible pour chacun des transits étudiés, le clair sous quille est également tracé. Le clair sous quille, noté UKC (pour « UnderKeel Clearance ») est calculé comme suit : 3.2 UKC = hauteur d’eau – (tirant d’eau + squat) Avec la hauteur d’eau calculée par la relation : h = hmarée – sonde Résultats par navire Les Figures 4 à 12 présentées infra synthétisent les résultats obtenus après traitement des mesures faites à bord. Les résultats sont présentés sous forme de profil en long du chenal de Seine. Les distances sont matérialisées classiquement par des PK (points kilométriques). Les principaux lieux sont donnés dans le Tableau 3 : Lieu PK Rouen 245 La Bouille (extrémité aval du port) 259 Duclair 278 Le Trait 300 Caudebec (station de relève des Pilotes) 310 Aizier (limite extrême de salinité) 325 Port Jérôme 330 Pont de Tancarville 338 Honfleur 355 Engainement 372 Tableau 3 : Localisation des principaux lieux Figure 4 : Résultats pour le Virana Delmas 6 of 18 PIANC MMX Congress Liverpool UK 2010 Figure 5 : Résultats du Lucie Delmas Figure 6 : Résultats du Bro Juno Figure 7 : Résultats du Maersk Rosyth 7 of 18 PIANC MMX Congress Liverpool UK 2010 Figure 8 : Résultats du Stones Figure 9 : Résultats du Richard Maersk Figure 10 : Résultats du Kotor 8 of 18 PIANC MMX Congress Liverpool UK 2010 Figure 11 : Résultats du Patent Figure 12 : Résultats du Fort St Louis 9 of 18 PIANC MMX Congress Liverpool UK 2010 3.3 Synthèse des résultats Le Tableau 4 synthétise les principaux paramètres représentés dans les graphiques. Navire Squat (m) Vitesse surface (nds) Clair sous quille (m) Moyenne Max Moyenne Max Moyenne Min Virana Delmas 0.44 0.77 10.26 13.26 6.23 3.59 Lucie Delmas 0.37 0.92 9.71 13.24 4.15 1.78 Bro Juno 0.41 0.79 10.73 12.98 4.64 1.77 Maersk Rosyth 0.41 1.02 9.62 12.27 3.80 1.42 Stones 0.35 0.79 9.24 11.30 3.19 1.13 Richard Maersk 0.66 1.13 9.21 12.13 2.18 0.17 Kotor 0.50 0.91 8.84 10.93 2.36 0.37 Patent 0.47 0.78 8.78 11.04 2.58 0.63 Fort St Louis 0.32 0.74 9.49 14.61 3.40 0.80 TOUS 0.44 1.13 9.43 14.61 3.49 0.17 Tableau 4 : Synthèse des résultats 3.4 Principales observations 3.4.1 Vitesse des navires Les navires effectuent le trajet Engainement – Rouen avec une vitesse proche de 10 nœuds avec des variations dues à des passages au droit d’appontements (ex : Port Jérôme), de croisement de navires, passages de courbes ou relève des pilotes. Il faut noter également une différence importante entre les navires montant et descendant : les montant gardent une vitesse quasi constante proche de 10 nds tout le long du trajet alors que les descendant ont une vitesse faible de l’ordre de 8-9 nds sur la partie amont (avant Aizier) et plus importante proche de 11 nds ensuite. Cela s’explique par le principe même de la navigation en Seine : les navires descendant naviguent en jusant jusqu’à Vatteville où ils rencontrent la basse mer à l’endroit le plus profond de la Seine puis doivent faire face au courant de flot pour sortir de l’estuaire, les navires montant transitent avec la marée de sorte à être en tout point de la Seine proches de la marée haute locale. 3.4.2 Squat On note que pour plus la moitié des navires suivis, le squat maximum enregistré est inférieure à 80 cm. Deux navires présentent des squats dépassant le mètre : le Richard Maersk et le Maersk Rosyth, deux navires pleins de forme avec des forts tirants d’eau. La moyenne des squats mesurés pour l’ensemble des transits en Seine est proche de 45 cm. Les valeurs du squat sont fortement dépendantes de la vitesse. Vu les contraintes de navigation, il existe de nombreuses zones où le navire doit réduire sa vitesse : au niveau des appontements de Honfleur (PK 355 environ) ou Port Jérôme Radicatel (PK 330- 336) et au moment de la relève des pilotes. De ce fait, le squat mesuré dans ces zones est toujours faible, inférieur à 45 cm. 10 of 18 PIANC MMX Congress Liverpool UK 2010 Ailleurs, les pics locaux pour tout navire atteignent 70 cm avec des maximums pouvant dépasser 1 mètre sur 3 secteurs bien identifiés : à la rencontre du front de marée à Vatteville pour les navires descendant où la montée rapide du plan d’eau doit avoir une influence importante sur le squat en dépit de vitesses faibles, dans la zone comprise entre les ponts de Tancarville et de Normandie, zone où les fonds sont généralement contraignants pour la navigation, à l’aval d’Honfleur, où la vitesse enregistrée est élevée et les fonds profonds. A Vatteville, pour les navires descendant, il s’agit également d’une zone où les navires montant avec un tirant d’eau faible croisent les forts tirants d’eau descendant. Il est connu que le croisement est un facteur aggravant le squat. Cela provient du fait que la section au droit du navire est brutalement réduite. De plus, cela peut introduire du tangage (oscillation du navire d’avant en arrière) qui accentue la perception du squat. Cela est très net dans le cas du Richard Maersk. Le graphique donné en Figure 13 montre les résultats obtenus dans cette zone après un croisement. Figure 13 : Illustration du tangage d’un navire En moyenne, c’est sur la partie aval que le squat est plus important. Cela est dû à deux facteurs prépondérants : la vitesse et la proximité du fond. A l’aval, que le navire monte ou descende, sa vitesse est supérieure à 10 nds, plutôt proche de 11 nds en moyenne et à l’estuaire, en aval de Tancarville, il existe des points hauts qui amplifient le squat. Sur la partie amont, il n’y a pas concordance de ces deux facteurs : les montant ont une certaine vitesse mais ont de l’eau sous la quille et les descendant naviguent plus près du fond (c’est la partie limitante d’un point de vue des tirants d’eau) mais avec une vitesse plus faible du fait qu’ils accompagnent le jusant. 3.4.3 Clair sous quille Les résultats concernant le clair sous quille font apparaître qu’en moyenne les navires transitent avec près de 3,50 mètres d’eau sous la quille. Les fonds du chenal d’accès au port sont irréguliers avec une alternance de fosses de points hauts. Les points hauts contraignants pour la navigation sont bien connus. Ce sont principalement : L’engainement (PK 365 – 370) et la zone située entre les deux ponts (PK 346 – 355) : ces lieux sont contraignants pour les navires à la montée et à la descente et sont navigués uniquement à pleine mer, La zone de Vatteville (PK 321 – 325) : les descendant rencontrent la basse mer peu avant et leur transit est calculé pour arriver sur zone en début de flot. Ils passent dans cette zone au plus tôt à vitesse faible et donc avec un clair sous quille réduit. Au minimum, le clair sous quille après déduction du squat est supérieur à 1mètre pour les navires montant et reste supérieur à 30 cm pour tous les descendant exceptés pour le Richard Maersk. Ce navire présente un clair sous quille réduit à la prise de flot vers Vatteville du fait d’un squat important dans cette zone, probablement lié au croisement de plusieurs navires. 11 of 18 PIANC MMX Congress Liverpool UK 2010 Ces résultats montrent : Que le chenal d’accès au port de Rouen est bien dimensionné en termes de gabarit, Que la connaissance de la Seine et des conditions complexes de navigation dans le chenal d’accès à Rouen (hauteurs d’eau et profondeurs variables dans un chenal long et sinueux) ainsi que l’expérience des pilotes permettent d’optimiser les capacités du chenal sans affecter la sécurité de la navigation. 4. DETERMINATION DES CONDITIONS DE NAVIGATION EN SEINE Le professeur C. Barass est un éminent spécialiste du phénomène. A ce titre, il a effectué des expertises sur de nombreux accidents maritimes et a formulé diverses formules selon les conditions de navigation. Nous avons comparé spécifiquement les résultats avec les formules d’évaluation du squat qu’il a établies car ce sont les plus utilisées dans le monde maritime, et notamment par les armateurs et les équipages. Cela s’explique par le fait que leur utilisation est simple et que les valeurs calculées majorent généralement le surenfoncement, ce qui garanti la sécurité de leur utilisation. Ces formules font intervenir peu de facteurs et peuvent être facilement calculées par les équipages. Ces facteurs sont : la vitesse surface du navire, le coefficient de bloc issu des tables hydrostatiques, et les caractéristiques du chenal, largeur du chenal et hauteur d’eau. 4.1 Hypothèses et notations 4.1.1 Coefficient de bloc Cb Le coefficient de bloc d’un navire est un critère de forme de carène. Plus un navire est fin et profilé, plus son coefficient de bloc est faible : c’est le cas des navires rapides type ferry ou porte-conteneurs. A l’inverse, les vraquiers sont des navires aux formes pleines avec des coefficients de blocs pouvant aller jusqu’à 0.9 pour les supertankers type ULCC (Ultra Large Crude Carrier). Ce coefficient se calcule à partir des tables hydrostatiques du navire en fonction du tirant d’eau. Il est défini par le rapport entre le volume immergé du navire et celui du parallélépipède rectangle dans lequel il s’inscrit. Le Tableau 5 ci dessous, issu de l’AIPCN, montre les Cb typiques des principales catégories de navires. Coefficient de bloc typique (Cb) AIPCN Tankers 0.85 Bulk Carriers 0.72 – 0.85 Container Ships 0.60 – 0. 80 General Cargo 0.72 – 0..85 Ro-Ro Vessels 0.70 – 0.80 Ferry 0.55 – 0.65 Tableau 5 : Coefficients de bloc par type de navire (source : AIPCN) 4.1.2 Notations Sont ici présentées les principales notations utilisées : V : vitesse surface du navire en nœuds Cb : coefficient de bloc du navire 12 of 18 PIANC MMX Congress Liverpool UK 2010 S : coefficient de blocage du navire, rapport entre les sections du navire et d’écoulement, S= AS 1 = B H AC × b T B : largeur du chenal au fond b : largeur du navire H : hauteur d’eau T : tirant d’eau statique du navire 4.2 Comparaison des résultats avec les formules de Barass 4.2.1 Présentation des formules S Formule globale (1981) : e = C b × 1− S 2/3 V 2.08 × 30 Formules simplifiées (les plus utilisées notamment par les armements) : e = Cb × V2 en mer ouverte 100 e = Cb × V2 en eau restreinte 50 4.2.2 Notions de mer ouverte et eau restreinte Le squat d’un navire dépend des conditions d’écoulement de l’eau autour de sa coque et la dépression créée par son passage. Ainsi, il est clair que, pour une profondeur donnée, plus les parois latérales sont proches, plus la vitesse de l’écoulement au droit du navire sera rapide et donc plus l’enfoncement sera important. Inversement, en mer ouverte, il n’y a pas d’influence de parois ou de restriction de section, l’accélération des filets d’eau est donc atténuée par le volume disponible. Cela étant dit, Barass propose la notion de largeur d’influence : cette largeur dépend du Cb du navire via la relation : Où CW = [ ] Weff = 7.7 + 4.5 × (1 − CW ) × B 2 1 2 + × Cb 3 3 Cette largeur d’influence correspond à la section maximale influencée par le passage du navire. Le Tableau 6 donne des exemples de largeur d’influence pour différents types de navires. Type de navire Coefficient de bloc Cb Largeur d’influence Navire virtuel ayant la forme d’un parallélépipède rectangle 1 7.7 x b Pétroliers 0.9 7.9 x b Cargo 0.7 9.5 x b Paquebots 0.625 10.51 x b Porte conteneurs 0.555 11.75 x b Remorqueur 0.5 12.81 x b Tableau 6 : Largeur d’influence par type de navire 13 of 18 PIANC MMX Congress Liverpool UK 2010 Dans le cas d’un calcul de squat en mer ouverte, le coefficient de blocage S est calculé avec cette largeur d’influence : S= 1 Weff H × b T De prime abord, on peut donc considérer que les conditions de navigation sont de type « mer ouverte » si la largeur du chenal est supérieure à la largeur d’influence. Remarque : le chenal d’accès au port de Rouen a une largeur de 120 m sur sa partie amont, jusqu’au PK 338, puis 200 m jusqu’au PK 365 avant de s’élargir considérablement dans sa partie la plus aval. (Voir Figure 14 ci après) Selon ce critère, la partie amont répond à la définition de condition en eau restreinte pour tous les types de navires et la partie l’aval, à celle de mer ouverte pour les navires les moins larges (b < 25 m). Figure 14 : largeur théorique du chenal NB : cette largeur est théorique, la largeur réelle du chenal dépend de la bathymétrie et de l’évolution des fonds. Elle peut être réduite comme à l’estuaire où actuellement des abandons sont admis. 4.2.3 Comparaison des résultats avec les diverses formules de Barass Les mesures présentées ici ne concernent que la partie amont de Tancarville (PK 338). Cette partie, communément appelée « rivière » en opposition à « l’estuaire », est la plus étroite. Le chenal a une largeur théorique de 120 mètres, exceptée autour du PK 330 où il fait 300 m. Exemple du Kotor Pour rappel, le Kotor est un navire céréalier de dimensions 175 m x 26 m. Il a effectué la descente depuis Rouen avec un tirant d’eau avant de 9.95 m. Son coefficient de bloc a été fixé à 0.85. La Figure 15 représente le squat calculé par les deux formules de Barass, simplifiée et générale, pour les deux conditions d’écoulement, eau ouverte ou eau restreinte, en fonction du squat mesuré. 14 of 18 PIANC MMX Congress Liverpool UK 2010 Figure 15 : Kotor : Comparaison des mesures avec les formules de Barass pour le choix des conditions d’écoulement Ce graphique montre bien que l’hypothèse de conditions d’écoulement en eau restreinte conduit à une surestimation forte du squat : + 49% avec la formule générale et même + 126 % avec la formule simplifiée utilisée par les bords. Avec l’hypothèse de mer ouverte, les deux formules présentent dans ce cas une bonne estimation du phénomène avec une surestimation de 13 % avec la formule simplifiée et une sous estimation de 12 % avec la formule générale. Exemple du Stones Le Stones est un navire minéralier de dimensions 166.3 m x 24.75 m. Il est monté à Rouen avec un tirant d’eau de 10.50 m. Le coefficient de bloc, relevé à bord, est de 0.815. Les courbes de comparaison des mesures avec les valeurs issues des formules de Barass sont présentées en Figure 16. 15 of 18 PIANC MMX Congress Liverpool UK 2010 Figure 16 : Stones : Comparaison des mesures avec les formules de Barass pour le choix des conditions d’écoulement Comme précédemment, les formules avec l’hypothèse d’eau restreinte conduit à une surestimation des valeurs de squat : + 52 % avec la formule générale et + 153 % avec la formule simplifiée alors qu’en condition de mer ouverte, les deux formules présentent une bonne estimation du phénomène avec une sous-estimation de 5 % avec la formule générale et une surestimation de 27 % avec la formule simplifiée. 4.3 Conclusion Les formules en eau restreinte surestiment largement voire très largement les valeurs observées alors qu’avec l’hypothèse d’eau ouverte, les prédictions sont proches de la mesure. Ainsi, pour la partie amont du chenal en Seine et donc a fortiori pour la partie aval de celui-ci (pour laquelle les largeurs sont plus importantes) les formules d’évaluation du squat les mieux adaptées sont les formules de type mer ouverte. 5. ESTIMATION DE L’IMPACT DE L’APPROFONDISSEMENT 5.1 Hypothèses et choix des navires testés Pour estimer l’impact de l’approfondissement en termes de squat, la formule de Barass est utilisée. Elle donne des résultats proches des mesures en nature. Compte tenu du fait que les conditions de navigation dans le chenal d’accès à Rouen sont considérées comme des conditions de type mer ouverte, la valeur de squat maximum calculée est indépendante de la taille du navire. Elle est fonction uniquement du coefficient de bloc. On considère donc un navire vraquier navigant avec le tirant d’eau maximum admissible soit 10.3 m en conditions actuelles et 11.3 après approfondissement. 16 of 18 PIANC MMX Congress Liverpool UK 2010 Le coefficient de bloc de ce navire varie avec son tirant d’eau. Il vaut 0.75 pour un tirant d’eau de 10.3 m et 0.78 pour un tirant d’eau de 11.3 m. Pour matérialiser l’amélioration des accès nautiques, le paramètre relatif à la hauteur d’eau, H, est augmenté d’un mètre. Le calcul est fait avec une hauteur d’eau initiale de 11.5 m, soit 10.3 m à laquelle s’ajoutent 70 cm de squat et 50 cm de clair sous quille. Cela correspond à des conditions défavorables pour le calcul du squat maximum. La vitesse de référence des calculs est de 10 nœuds correspondant à la vitesse moyenne d’un transit en Seine. 5.2 Résultats La formule générale de Barass est utilisée : S e = Cb × 1− S 2/3 × V 2.08 30 Le Tableau 7 présente les résultats obtenus pour les différentes conditions de navigations. Hauteur d’eau Tirant d’eau Coefficient de bloc Squat maximum Conditions actuelles 11.5 m 10.3 m 0.75 0.69 Après approfondissement 12.5 m 10.3 m 0.75 0.65 Après approfondissement 12.5 m 11.3 m 0.78 0.73 Tableau 7 : Impact de l’approfondissement du chenal 5.3 Conclusion Par application de la formule de Barass, l’impact de l’approfondissement peut être estimé : A une amélioration des conditions de navigation de 4 cm pour le même navire dans un chenal plus profond, A une augmentation du squat de 4 cm, pour un navire montant avec un tirant d’eau maximum. L’impact est donc faible au regard des valeurs actuelles et des hauteurs d’eau. Le squat attendu après approfondissement sera du même ordre de grandeur qu’actuellement. L’approfondissement du chenal d’accès au Port de Rouen n’est pas de nature à remettre en cause les conditions de sécurité et les pratiques actuelles. Cet impact sera également quantifié par une campagne de mesures similaires pour valider les résultats. 6. SYNTHESE ET CONCLUSION Dans le cadre de son programme d’approfondissement, le Port de Rouen a mené une étude afin de vérifier les conditions de navigation en Seine et notamment en termes de variation de tirant d’eau. Le phénomène de squat a ainsi été quantifié pour des navires à fort tirant d’eau transitant vers et depuis Rouen. Ces mesures ont permis après exploitation des données de vérifier le bon dimensionnement des accès et l’utilisation pertinente faite par les pilotes de la marée pour optimiser en toute sécurité le gabarit du chenal. Une comparaison avec les formules empiriques de prévision de squat montre une bonne corrélation avec les mesures et permet ainsi une estimation de l’impact de l’approfondissement sur le squat. La variation de squat liée aux changements de conditions de navigation est inférieure à 5 cm, ce qui est faible au regard des hauteurs d’eau concernées. Cet impact est pris en compte via le pied de pilote intégrant les incertitudes liées au tirant d’eau, à la connaissance exacte des fonds ou à l’hydrodynamique (houle). Cette estimation de l’impact valide le dimensionnement du projet tel qu’il est envisagé : amélioration des tirants d’eau admissible d’un mètre par un approfondissement du chenal d’un mètre également. 17 of 18 PIANC MMX Congress Liverpool UK 2010 REFERENCES rd Barass C. B. (2009), Trim and Squat, Proceedings 3 Squat Workshop, Elsfleth, Germany. Briggs, M.J. 2006. Ship Squat Predictions for Ship/Tow Simulator, Coastal and Hydraulics Engineering Technical Note CHETN-I-72, U.S. Army Engineer Research and Development Center, Vicksburg, MS. Dunker S., Gollenstede A., Härting A. et Reinking J. (2002), Analysis and Comparison of SHIPS Derived Squat, Proceedings Hydro 2002, Kiel. Härting A. et Reinking J. (2002), SHIPS : A new Method for Efficient Full-scale Ship Squat Determination, Proceedingd PIANC congress, pp 1805-1813, Sydney, 2002. Hervieu G., (1993), Le surenfoncement des navires : introduction à l’hydrodynamisme naval, Edition Masson, ISBN : 2-225-84025-3. Stocks D. T. et Dagget L. L (2002), Maximization of Ship Draft in the St. Lawrence Seaway : Volume 1, Squat Study, Fleet Technology Limited. 18 of 18 PIANC MMX Congress Liverpool UK 2010 Needs and Technologies for port security by by D. Lerouvillois1 ABSTRACT Concerns of the society about terrorism risks have become major since the September 11, 2001 attacks. This is particular true in the transportation sector, especially in the maritime transportation. Thus, the International Maritime Organisation (IMO) has defined and then approved a set of measures to enhance the security of the maritime sector against terrorism attacks (ISPS code) and the European Union has extended the security requirements (Regulation EC 725/2004, Directive EC 2005-65). To face up to terrorism threats, ports must implement organisational measures but also want to rely on technologies which can help them to protect their sensible assets and to manage the security of their site. This article first describes the main stakes and the general needs of port security. Then it presents technologies and systems implemented in three domains of port security, through studies and research projects in which the CETMEF (French Institute for maritime and waterways studies) has recently participated. They are : • screening of shipping containers, in the context of a future regulation sought by the United States : the 100% scanning, • perimeter protection of sensitive areas inside ports, • monitoring of maritime approaches to ports. 1. INTRODUCTION Maritime transport has always been a favourite target for intentional unlawful acts such as terrorism and piracy. But people really became aware of the risks after the terrorist attacks on September 11, 2001. So it urged the International Maritime Organisation (IMO) to define and approve in December 2002 the International Ship and Port Security (ISPS) code, which describes the preventive measures to be implemented by ports and ships. 2. STAKES OF PORT SECURITY 2.1 General With adoption of the ISPS code, the IMO justifies the arrangements of maritime security by reminding the main stakes : • Intentional unlawful acts, and more particularly terrorism, are among the most serious threats for the values of democraty, liberty and peace, • unlawful acts against the maritime transportation put in jeopardy the safety and the security of humans and goods, can damage seriously the maritime services and undermine the trust with the safety of maritime navigation. • the maritime transportation is very important for the world trade, so it justifies to make efforts to preserve the logistic chain from rupture due to terrorist attacks against ships, ports, off-shore terminals and other sites. Furthermore, the transport of dangerous goods, such as chemicals or radioactive materials for example, is particularly vulnerable against intentional unlawful acts, which can have serious consequences for citizens and the environment. Engineer, French Institute for Maritime and Waterways Studies (CETMEF), France, [email protected] 1 of 13 1 PIANC MMX Congress Liverpool UK 2010 The suicide attack against the U.S. Navy destroyer USS Cole on October 12, 2000, while it was harbored in the Yemeni port of Aden, has particularly revealed the terrorism threat inside ports. The modern economy is totally dependant on the sea. Over 80% of world trade, nearly 7 billion tons of freight, are transported each year on the oceans by more than 50 000 ships of more than 1 000 GT serving approximately 4 000 ports around the world, not include thousands of coasters. For the liberty of trade, the developed countries must control the main sea routes, especially the strategic points such as straits and channels, but also they must give security to ports and ships. Port infrastructures such as liquid fuel terminals, gas terminals and container terminals, are very vulnerable against attacks from ships or containers. Major attacks against the biggest ports of the world would have serious consequences on regional and global economy. The container traffic represents today almost 90% of the world maritime trade of goods. Fears of terrorists using containers began on October 25, 2001 by the discovery of a stowaway hidden in a container in the port of Gioia Tauro, in Italy. The container was loaded on a ship with the final destination to Canada. The man was identified as a presumed terrorist from Al-Quaida and he was suspected of preparing an attack in a plane on the liaison Montreal – Cairo. More recently, in March 2004, a suicide attack against the Israeli port of Ashdod, resulting in 11 deaths and 18 casualties, was perpetrated by terrorists who had penetrated in the port hidden inside a container transported by truck from the Gaza Strip. However, the major concern remains the use of containers for transportation and delivery of weapons of mass destruction (chemical, bacteriological, radioactive, nuclear). Other threats are taken into account by the authorities in charge of port security. Ports need to protect themselves against illicit acts such as theft of valuable goods, vandalism, etc. These acts are the most common risks they face. They also have to consider the risks of stowaways introducing inside ships, who hide most often into containers, trucks or ship structures. That's for example a real problem for the French ports of the Channel which face to candidates to immigration in Great Britain due to its high attractivity. The preferred ships for these activities are ferries, container ships, freighters and small ships. Finally, the customs authorities take special care to check goods passing through ports in the fight against illegal trafficking of goods (counterfeit, drugs, weapons, non declared valuable products, etc). 3. General needs of port security The ISPS code, adopted by the IMO on December 12, 2002, apply to passenger ships and cargo ships of 500 GT and upwards on international voyages, and to the port facilities serving such ships. Port facility is defined as the interface site between ship and port used for transferring cargoes or passengers. The ISPS code is based on three security levels corresponding to a normal situation, an increased risk, or a probable threat. It essentially defines procedures on security organization (responsibilities, methods, objectives, documents to produce, approbation and certification, training) which are made mandatory (part A). It also states as recommendations some lists of operational measures for the main domains of port activities (part B). 2 of 13 PIANC MMX Congress Liverpool UK 2010 It notably prescribes : • elaboration of security assessment and security plan for ships and port facilities. Security plans have to contain graduate measures in response for the three security levels of the code, in order to achieve security objectives assigned to the main port operations, • designation and training of ship security officers, company security officers and port facility security officers, • implementation of the plans at security level required by each state, • implementation of ship security alert systems inside ships, used discreetly when there are security incidents, and treatment of these alerts by each state of the flag. European Union has extended the application of the ISPS code to passenger ships on national voyages (class A) and optionally ships operating domestic services, to their companies and to the port facilities serving them (Regulation 725/2004). It has also made mandatory some arrangements of part B of the ISPS code. Then, it has strengthened the ISPS code by the Directive 2005/65 of 26 October 2005, laying down the extension of security arrangements to the ports which include one or more port facilities covered by an approved security plan pursuant to Regulation 725/2004. Thus, it prescribes port security evaluation and then port security plan by the port authorities. Each port has to appoint a port security officer, responsible for the implementation of port security measures. Port facility (port) security assessment : The port facility security assessment is essentially a risk analysis of all aspects of port facility operations. It consists in identifying the parts which are the most probable targets in case of attacks. The security risk is a function of the threat of an attack coupled with the vulnerability of the target and the consequences of an attack. The evaluation must include the following elements : • identification and evaluation of assets and infrastructure that should be considered important to protect, • identification of possible threats against the assets and infrastructure and their probability of occurrence, in order to establish and prioritise security measures, • identification, selection and prioritisation of countermeasures and procedural changes and their level of effectiveness in reducing vulnerability, • identification of weaknesses, including human factors, in the infrastructure, policies and procedures. Port facility (port) security plan : The port facility security plan sets the arrangements to be taken for the port facility security. It depends on the results of the security assessment. It clearly prescribes precise measures in the following domains : - Access to the port facility (or an area of the port) : The security measures may consist in : • establishing the security measures covering all means of access to the port facility, • identifying the appropriate locations where access restrictions or prohibitions should be applied for each of the security levels, • establishing for each security level the means of identification required to allow access to the port facility and for individuals to remain within the port facility without challenge, • developing an appropriate identification system, allowing for permanent and temporary identifications, for port facility personnel and for visitors respectively, 3 of 13 PIANC MMX Congress Liverpool UK 2010 • identifying the locations where persons, personal effects, and vehicle searches are to be undertaken. - Restricted areas within the port facility (port) : The purpose of restricted areas is to: • protect passengers, ship's personnel, port facility personnel and visitors, including those visiting in connection with a ship, • protect the port facility, • protect ships using, and serving, the port facility, • protect security sensitive locations and areas within the port facility, • protect security and surveillance equipments and systems, • protect cargoes and ship's stores from tampering. The security plan should identify the restricted areas to be established within the port facility and specify their extent, times of application, the security measures to be taken to control access to them and those to be taken to control activities within them. In case of temporary restricted areas, a security control should be made both before and after that area is established. The security measures should consist in controlling : • access by individuals, • entry, parking, loading and unloading of vehicles, • movement and storage of cargoes and ship's stores, unaccompanied baggage or personal effects. - Cargo handling : The security measures may include : • checking of cargoes, cargo transport units and cargo storage areas within the port facility prior to, and during, cargo handling operation, • checks to ensure that cargo entering the port facility matches the delivery note or equivalent cargo documentation, • searches of vehicles, • checking of seals and other methods used to prevent tampering upon entering the port facility and upon storage within the port facility. 4 of 13 PIANC MMX Congress Liverpool UK 2010 Checking of cargoes may be accomplished by some or all of the following means : • visual and physical examination, • using scanning/detection equipments, mechanical devices, or dogs. - Delivery of ship's stores : The security measures may consist in : • checking ship's stores and package integrity, • preventing ship's stores from being accepted unless ordered, • searching the delivery vehicle, • escorting delivery vehicles within the port facility. - Handling unaccompanied baggage : The security measures may consist in : • ensuring that unaccompanied baggage is screened or searched up to and including 100 percent, which may include use of x-ray screening, before it is allowed in the port facility, • Applying a 100 percent x-ray screening of all unaccompanied baggage. - Monitoring the security of the port facility (port) : It consists in monitoring the port facility and its nearby approaches, on land and water, at all times, including the night hours and periods of limited visibility, the restricted areas within the port facility, the ships at the port facility and areas surrounding ships. Such monitoring can include use of: • lighting, • security guards, including foot, vehicle and waterborne patrols, • automatic intrusion detection devices and surveillance equipments. 4. Technologies and systems for port security In order to implement the security measures described in the security plan, port facility and port authorities would like to rely on technologies and operational systems which can be adapted to the different types of risks and bring them assistance for daily protection of their sensitive sites. These systems facilitate prevention and dissuasion against unlawful acts, detection and impeding intruders, and can contribute to the protection of the sensitive elements and to the intervention of the public security forces. They also facilitate monitoring in real time of all the sensitive areas and help decision-making in case of imminent threats. Thus, against threats coming from sea or undersea, specific technologies are developed for monitoring the traffic, detection and tracking, dissuasion and protection. Against threats coming from land, ports can rely on systems for perimeter protection, video surveillance, intrusion detection, access control, screening of humans, luggage, cargoes and vehicles. Tendency is to take into account all the threats and to treat them globally inside an integrated port security system. We'll look at systems implemented in three domains of port security, through studies and research projects in which the CETMEF recently participated. They are : • screening of shipping containers, in the context of a future regulation sought by the United States : the 100% scanning, • perimeter protection of sensitive areas inside ports, • monitoring of the maritime approaches to ports. 5 of 13 PIANC MMX Congress Liverpool UK 2010 4.1 Screening of shipping containers Today, screening of shipping containers in French ports is mainly motivated by struggle against fraud and illegal trafficking, and is carried by the customs on the basis of a preliminary risk analysis. In the ports of Marseille and Le Havre, U.S. Customs officials cooperate with their French colleagues in the framework of the US program called CSI (Container Security Initiative) which aims to prevent the introduction of weapons of mass destruction on American ground through shipping containers. Checks are made before shipment of containers in foreign ports, by analysing risks and screening suspicious containers to the scanner (X-rays or Gamma). Impact study of 100% scanning : By the law of House Resolution 1 (HR1) approved on August 03, 2007, the United States plan to make compulsory, beginning July 1, 2012, the scanning of all U.S.-bound shipping containers at foreign ports before being loaded. Under this Act, called the 9/11 Act, scanning is the process of using radiation detection and nonintrusive imaging equipment (NII) to view the contents of a container and determine the presence of nuclear or other radioactive materials. The results of these controls are supposed to be transmitted in real time to a control centre located in the United States, which is responsible for analysis and taking appropriate decisions. The 9/11 Act recognizes that a set of technical and logistical challenges must be overcome in order for the scanning requirement to be achievable in all foreign ports. The deadline may be extended in two year increments, until the specific conditions necessary for a full deployment of the scanning equipment can be met. To cope with these new measures and anticipate the impact on port operation, the CETMEF was asked in 2008 to conduct an impact study of the 100% scanning in France. This study was carried out by the company CRIT T&L associated with the University of Le Havre. The main findings of the impact study were as follows : • French ports concerned by 100% scanning : There are seven major metropolitan ports in France. Among them, Le Havre and Marseille deal with more than 80 % of the container traffic (3,8 Million TEU in 2006). If we consider the US in-bound traffic (170 000 TEU in 2006), all the traffic is nearly concentrated in Le Havre and Marseille. Le Havre operates around 70 % of these exportations including 26 % of transhipment in 2006, whereas Marseille operates nearly 30 % of these flows and no transhipment. As for the modal shares, Marseille and Le Havre are on the same level : Road 73-75%, Rail 19-15 % and River 8-10%. Finally, there is another difference between these two ports : the actual terminal in Marseille-Fos (Graveleau terminal) and the future terminals (Fos 2, 3 and 4 XL) are very close to each other, whereas in Le Havre the terminals are more numerous (7 at the moment) but relatively far from each other. Thanks to port 2000 in Le Havre, the southern side of the port could raise up to 82% in 2012 for the US in-bound traffic. • Scanning technologies : In order to answer to the American demand, it will be compulsory to combine means of detection of radioactive goods with the radioscopic images of the containers. Three types of equipment exist : fixed systems with high detection power, mobile systems offering more flexibility of the scanning centre location and fixed portal systems to inspect more vehicles per hour (more than 100 per hour). These last ones are more adapted to the 100 % scanning requirements. 6 of 13 PIANC MMX Congress Liverpool UK 2010 • Experiments in progress : The US has launched a program called Secure Freight Initiative (SFI) to experiment 100 % scanning in port pilots. The first phase was conducted in small terminals at three ports (Southampton, Port Qasim and Puerto Cortes) from October 2007 to April 2008. According to the report of the US Department of Homeland Security (DHS) established in June 2008, 100% scanning of containers is possible on limited scale ports, but it generates very high costs impacting logistics chains. Furthermore, this process is particularly difficult to implement in case of containers transferred from inland waterways, and heavily penalizing in case of transhipment, which dramatically impact the flow of container traffic at major ports. • The different scenarios and their impact evaluation : Various assumptions have been considered for the impact study of 100% scanning at the ports of Marseille and Le Havre : estimated volumes to be processed in 2012 (170 000 TEU), an estimate of the typical configuration of a scanning module and costs of acquisition, installation and operation, estimation of operation conditions, different locations for implementation of scanning modules (in the terminal, in the port, outside the port). These scenarios have been considered from several angles, including costs by differentiating logistics costs (transfer of containers to the centre of scanning) and operational costs (investment, operation, maintenance). Depending on the assumptions made, to ensure the fluidity of traffic at each port and limit the risk of obstruction, the number of modules required for 100% scanning would be 2 to 3 in Le Havre and 1 in Marseille. In light of the criteria considered in the analysis and the constraints of space inside the container terminals, it was concluded that the scenarios based on the location of scanning modules inside the port (but outside the container terminals) present the best scores because they have such a better flow, better conditions of scalability, lower distortion of competition between terminals and lower risk of blocking the port. The scenarios positioning the scanning modules in the hinterland of the port involve too high costs to be considered at this point. • The findings of the impact study : In light of this study, and also experiments conducted in foreign ports, the economic impact of 100% scanning would be very high for the supply chain. Indeed, transhipment and inland waterway transport operations would generate an additional cost close to what companies charge for shipping at the level of THC (Terminal Handling Charges - fees charged by owners to their customers for dock work), about 150 Euro per container. At minimum, this would virtually double the expenses related to the port operation, with a strong impact on inland waterway transport in particular, which could be heavily penalized if no suitable solution was introduced (e.g. cost sharing). Impact assessment of 100% scanning on road traffic : In order to assess the impact of such an arrangement on the traffic of trucks arriving at the entrance of a container terminal (which congestion is caused by the deployment of a 100% scanning centre ?), it was proposed to conduct an experiment / simulation based on the actual traffic of trucks entering the container terminal in Fos-sur-Mer (Port of Marseille). This experiment is underway and results will be announced in early 2010. The experiment consists in counting the trucks entering the site during a period of two consecutive weeks and in choosing the most representative days. The collected data enable, by simulation of a stop corresponding to the estimated duration of scanning (variable value equal to 1mn30, 3 min and 5 min), to calculate the rise in the queue ahead of the entry site, and to deduct for lost time by truck depending on its arrival time. Researches for improvement on the existing road infrastructure, on location of the scanning centre and on the number of scanning devices will be sought. It is planned to consider several scenarios for scanning, especially considering only the containers for the USA, or any container, regardless of its destination. 7 of 13 PIANC MMX Congress Liverpool UK 2010 Research project VIP : The objective of the VIP project is to make consistent the performance of container control (for detection and identification of radiological and nuclear products) with the performance of port operation (in terms of traffic flow, quality of service, productivity). It is developed by an industrial consortium led by the company TL&A, including the CEA, CANBERRA, the port of Le Havre and the CETMEF. This project was selected in late 2008 by the French Agency for Research, ANR (Agence Nationale de la Recherche) under its aid program on global security, entitled CSOSG (Systems Concepts and Tools for Global Security). Started in early 2009, it is planned to be completed in early 2012. VIP seeks the establishment of a control system adapted to the identified risks. Risk analysis based on probable threats enable to define the specifications of controls (both in quality and quantity). The most suitable procedures and technologies will be identified, particularly in terms of sensitivity compared with the risk of "false alarms". They will be integrated into the dock work and will have to prevent disruption of port operations and the overall transport system, particularly in terms of duration. It will specifically : 4.2 • establish the control requirements depending on the identified risks, • determine the limits of current technologies with respect to these requirements and identify the most suitable, or those having the greatest potential for development, • improve a scintillation based detection technology, • test the selected technologies (and the improved) for implementation into the port environment (speed, reliability,...) and identify the constraints (safety requirements, energy needs, volume, operational conditions, interference, ...) • define the control procedures to implement and the intervention principles (for example extraction and neutralization in safety of suspected containers). Perimeter protection Perimeter protection is the outer protection of a sensitive site, so this is the first level of protection. From this position, it must ensure a number of essential functions : • Mark the property, • Informing the public, • Hide certain views, • Deter malicious acts, • Create a buffer zone upstream of buildings, • Detect and stop intruders, • Filter flows of personnel, passengers, vehicles and goods, • Facilitate removal of any doubt and intervention. 8 of 13 PIANC MMX Congress Liverpool UK 2010 While perimeter protection is a link in the overall security system, it is composed of various elements such as fence, access control, perimeter video surveillance,... As part of an infrastructure operation, these elements can be treated separately but must have been defined for the same level of protection and remain compatible. On the basis of a methodology for perimeter protection at airports, which was defined by the STAC (French Civil Aviation Technical Centre), the CETMEF launched a study in 2009 to design a methodology adapted to the port domain. It consists in defining a general method for analysing risks against the sensitive elements in ports and for orientation of perimeter protection solutions the best suited to cover these risks. This study, carried out by the French Institute INERIS, is conducted in four steps : 1) the contractor should be aware of the context by investigation in ten French ports of different sizes (the seven major ports and three decentralized ports) to : • identify the sensitive elements of the ports (restricted areas, port facilities managing flows of passengers or sensitive goods, storage areas of materials, loading docks, ...) that should be protected against intentional unlawful acts, • identify the types of threats to be taken into account (malice, theft, smuggling, terrorist attack, illegal immigration), • identify specific constraints to the ports (operating conditions, environment, topology, etc..) to be considered in the choice of perimeter protection solutions. 2) From the evidence previously gathered and the methodology developed for airports, the contractor should establish a general methodology for analysing risks to sensitive areas of ports. Depending on the type of threat and the level of protection to be achieved, it defines the functional and technical requirements of protection devices the most appropriate to the situation. These requirements must be sufficiently specific and quantifiable to enable port managers to define technical specifications the most appropriate for their perimeter protection. 3) In this step, the contractor should apply the methodology previously established in two ports (Marseille-Fos and Caen-Ouistreham), at a small scale, to confront its application with the security needs of the port, the specific conditions of port operation and the local environment. Application of the methodology will validate its adaptation to ports and verify the ability of implementation by port managers. The perimeter protection requirements derived from the methodology applied to these ports will be interpreted in light of the security needs expressed by the ports and they will be compared with existing perimeter protection devices to check their relevance. 4)The contractor should establish the final summary of its findings and conclusions from the results of previous steps, and its recommendations for the establishment of technical guidelines on implementation of perimeter protection applied directly by ports. The results of the study are expected during the first half of 2010. 4.3 Monitoring of maritime approaches The general principle applied to existing maritime surveillance systems is a combination of devices for monitoring ships into the maritime approaches (radar, satellites, AIS ,...) and cooperative procedures in which ships, on interrogation or spontaneously, provide information on their position, intention and characteristics of the ship (VHF radio, AIS, LRIT ,...). The difficulty of monitoring results not only technical problems of detection systems that do not always enable a view of all ships and boats, but the fact that cooperative procedures do not apply to all ships in the maritime area : • the AIS system applies to ships with a tonnage exceeding 500 GT, • the LRIT system applies to ISPS ships with sometimes opportunities to stop emissions towards the coastal state. 9 of 13 PIANC MMX Congress Liverpool UK 2010 Therefore, it is entirely proper that non-cooperative ships and boats (fishing, boating ,...) can navigate in sensitive maritime approaches. Monitoring the maritime approaches of ports is a key component for their protection. Once identified threats and restricted areas to monitor, it is necessary to specify a monitoring system capable of providing security information on the traffic into the maritime approaches and enabling the security forces to intervene if needed on these approaches to arrest intruders and prevent attacks. Research project SECMAR : The SECMAR project aims to provide a practical response to the detection of a terrorist threat coming from the sea surface or under the surface, in order to attack a sensitive coastal site. It foresees the development of a prototype system for detection intrusion in a sensitive maritime area in the south of France. It was set up by industry companies and research laboratories located in the south of France (supported by the competitiveness cluster “MER PACA”) and it was selected in 2006 by the French state for funding after a call for research projects. The industrial consortium is led by THALES (Underwater Systems) and the CETMEF participates as technical expert for the Ministry in charge of transport. The port of Marseille is also involved in this project by contributing to the definition of needs, and enabling the implementation of the SECMAR demonstrator in order to control the maritime approach to the oil terminal located at Fos-sur-Mer. In Phase 1, the consortium validated a classification of specific terrorist threats against this site and conducted experiments of detection equipment in 2007. These experiments and subsequent analysis enabled the definition of a prototype of the SECMAR system, suitable for sensitive sea areas, which is actually under development. 10 of 13 PIANC MMX Congress Liverpool UK 2010 This prototype includes a set of heterogeneous but complementary detection means to increase the probability of detecting terrorist actions and it considers the constraints due to the environment. The interest in combining heterogeneous sensors into a unique system of intrusion detection and analysis is twofold : • it enables to respond to various threats and different scenarios of intrusion, • it enables to adapt to complex environmental conditions : shallow seabed, large noise, heavy traffic, different maritime activities (trade, craft, fishing), low signatures, etc... This project is organized and structured to foster in each partner research and implementation of innovative technologies. Thus, it includes several components and technologies : • fixed underwater detection : monitoring of mobile on surface and underwater. During the experimental phase, it has been clearly demonstrated that the sonar, in addition to its ability to detect underwater noise-makers, is an indispensable complement to the radar and optronics for detection and localization of noisy elements on surface, especially when the visibility is degraded or when a mobile of large size masks other probable mobiles in the monitored area. Several types of underwater detection means were tested : a network of hydrophones up to 4 acoustic distributed bases, an experimental plane antenna, a vertical linear antenna. • Radar Detection : location of potential threats in the monitored area, using an adapted radar to detect small surface targets. The radar must detect an object with at least a low radar cross section (1 m x 1 m) at a distance between 200 m and 7500 m (with a radial resolution of 40 m). The surveillance area of the tested radar (Profiler Radar) was a cone of 60° wide / 8° high cut into four sections of 15° by 8°. The four sections were monitored in parallel and information was repeated every 5 to 15 seconds. • Optronics detection : composed of a warning unit and an analysis unit. The optronics sensor of the warning unit consists of 3 sensors each juxtaposed with a field of 20°. The lines of sight of each sensor are shifted by 20° in bearing. The warning unit provides to the supervision centre tracks associated with detection on all cameras. A fusion algorithm allows the pursuit of detection between adjacent fields covered by each camera. The analysis unit gives to the system a capacity for analysis and identification of the threat by producing a video sequence. The analysis ability is achieved in part by the capture and presentation to the operator of a video stream in the infrared band 9μm-14μm (IR3 band) and secondly by the use of binoculars - telemeter for manual designation of the target. Display and use of information are processed by the supervision centre. The initial position of the suspicious object is provided by the supervision centre to the analysis unit which becomes independent to ensure tracking of the object. The analysis unit provides real-time information to the operator (time between video recording and final display is less than 1 second). • Transmission network : provision of transmission means for the SECMAR system with an objective of maximum deployability. It includes : . radio connections between SECMAR sensors and the supervision center by point to point secured liaison up to 100 Mb/s in WIMAX technology. . radio connections between the supervision centre and a mobile unit for intervention by point to point secured liaison up to 50 Mb/s in WIMAX technology. . transmission network dedicated to SECMAR with interconnections to other communication networks of the port. . mechanisms for securing SECMAR system. • Supervision centre : located inside the port and dedicated to monitoring sensible maritime areas, it has the following features : . location, classification, multi-sensor fusion, . automatic development of the operational picture of the monitored maritime area, 11 of 13 PIANC MMX Congress Liverpool UK 2010 . detection of abnormal behaviour, . collection and exchange of information, . interface with port security related systems, . interoperability with the means / resources in charge of intervention in case of intrusion detection. From summer 2010, the port operators of Marseille will use during 6 months the SECMAR demonstrator to enable validation tests in a real environment and in operation 24H/24H and promote the SECMAR system. Research project MWPS : The MWPS project (Water Prevention & Warning System) aims to develop a protection system for restricted maritime area within a port with non-lethal means. This system must detect, deter and neutralize an accidental intrusion or malicious act. This project, supported by the competitiveness cluster “MER PACA”, was selected by the French state for funding after a call for research projects in 2008. It was set up by an industrial consortium led by CS and composed by CS, LACROIX, CESIGMA as industrial partners. MWPS provides a level of graduated and proportional response to threats by using rapid reaction capabilities on a maritime area about 1 sq. km. It can be implemented in a few seconds by a qualified operator. It includes the following functions: • detection and tracking day / night of boats navigating in the restricted area, • injunction by sound message and ultra-directional lighting, • intimidation by launching non-lethal pyrotechnic charges, • neutralization of threat by non-lethal means, • centralized control system. The main characteristics of these functions are: • detection and tracking by an optronics IR3 equipment, mounted on a 2-axis positioner, coupled with binoculars-telemeter, functioning day and night. • An acoustic wall audible up to several hundred yards without nuisance in other directions, based on two innovations consisting in : 12 of 13 PIANC MMX Congress Liverpool UK 2010 . using the non linearity of air to demodulate the low frequency from ultrasonic frequencies, . pointing to provide directional sound by electronic processing without mobility of the acoustic wall. • a light beam, bound to the optronics equipment, enabling to mark or blind the target at night, • a 2-axis launcher sending precisely for intimidation and neutralization charges of different nature (detonating, lachrymal, illuminating, rubber bullets ,...) with adaptable threshold effect. A bibliographical study on the effects of non-lethal weapons upon humans was conducted during this project, in order to reduce the danger of the device to a minimum. Neutralization (physical ability to stop the boat) is not a feature that will be available in the demonstrator at the end of the project. However, extensive research of solutions have been conducted. Work will be pursued for one solution using nets capable of blocking the propellers. Optimization of these nets and their integration into a pneumatic or pyrotechnic automated launcher are under consideration by the company LACROIX. A MWPS demonstrator should be presented in early 2010 on a site near Toulon. References European Commission (2004) Regulation (EC) No 725/2004 of 31 March 2004 on enhancing ship and port facility security. European Commission (2005) Directive 2005/65/EC of 26 October 2005 on enhancing port security. IMO (2002) International Ship and Port Facility Security Code (ISPS Code). 13 of 13 PIANC MMX Congress Liverpool UK 2010 CONSTITUTION OF A NUMERICAL WAVE DATA-BASE ALONG THE FRENCH MEDITERRANEAN COASTS THROUGH HINDCAST SIMULATIONS OVER 1979-2008 by 1 by David Morellato and Michel Benoit 2 ABSTRACT The purpose of this article is the extension of the wave data-base ANEMOC to cover the Western Mediterranean Sea through numerical hindcast simulations with the third generation spectral wave model TOMAWAC over the 1979-2008 period. Wind-fields used for these simulations come from the NOAA/NCEP reanalysis (version 2). The model is calibrated with satellite altimeter data and wave buoy data. The statistical analysis of these hindcast simulations allows to estimate the average and severe wave conditions over the Mediterranean Sea. Part of these numerical data will be available on the ANEMOC website. 1. INTRODUCTION Over the recent years, CETMEF and EDF R&D LNHE have been collaborating to build through numerical hindcast simulations a continuous data-base of wave conditions covering the North-East part of the Atlantic Ocean, the English Channel and the North Sea, with a refined resolution along the French coasts of the Atlantic Ocean and the English Channel (e.g. Benoit & Lafon, 2004, Benoit et al., 2008). This data-base, called ANEMOC, covers a period of time from 1979/01/01 to 2002/08/31. Some of the data obtained during this project are available on a Web site (http://anemoc.cetmef.developpement-durable.gouv.fr/). Among other applications, this data-base has been used to build charts for average and extreme wave conditions (Lafon & Benoit, 2006). The present study aims at following the same lines in terms of objectives, methodology and deliverables by focussing on the Western Mediterranean Sea, with particular interest for the Southern European coasts (France, Spain, Italy). 2. DESCRIPTION OF THE NUMERICAL MODEL USED FOR THE HINDCAST RUNS 2.1 The third-generation spectral wave code TOMAWAC The simulations are performed with the third generation spectral wave model TOMAWAC (Benoit et al. 1996), which is part of the TELEMAC hydro-informatics suite, developed at EDF-LNHE. TOMAWAC solves the wave action density balance equation (e.g. Bretherton & Garret, 1969, Komen et al., 1994) and models the evolution (in space and time) of the directional wave spectrum, under unsteady wind forcing. It takes into account the input of energy from the wind, nonlinear wave-wave interactions, as well as dissipation due to white-capping, bottom friction and depth-induced breaking in shallow-water. A feature of high interest of TOMAWAC for nearshore and coastal applications is the use of unstructured spatial grids, which allows to refine the mesh in areas of complex bathymetry and irregular shoreline. The model has already been validated for the hindcast of several real storms (Benoit et al., 1996, Aelbrecht et al., 1998). For this project, a numerical grid has been set up with a variable spatial resolution over the Mediterranean Sea (see next section). The wave spectrum grid uses 21 frequencies with a logarithmic scale with ∆f/f = 0.122 (between 0.04 Hz and 0.4 Hz) and 36 directions (constant angular resolution of 10 deg.). Output time-step for the results is 1 h. The model is run with steady-state water levels (corresponding to mean tidal level) and without current effects, although the code may deal with unsteady currents and water levels. Coupled runs with the flow model TELEMAC-2D allowing to study the interactions between waves, tides and storm surges will be addressed in a future phase of this project. 1 Dr Engineer, CETMEF, DELCE, Plouzané, France, [email protected] 2 Senior-scientist, Saint-Venant Laboratory for Hydraulics, Université Paris-Est (joint research unit EDF R&D, CETMEF, Ecole des Ponts ParisTech), Chatou, France, [email protected] 1 of 8 PIANC MMX Congress Liverpool UK 2010 LATITUDE (°) 2.2 The MEDITERRANEAN wave model The model (called MEDITERRANEAN model) covers the Mediterranean Sea (30°N to 45°N in latitudes and 6°W to 36 E in longitudes), with a gr id of variable mesh size, from about 2-3 km along French coasts and 6-10 km along Spanish and Italian coasts to a maximum resolution of 0.5° over the rest of the domain. The spatial grid comprises 8770 nodes and 16192 triangles (Figure 1). No wave spectra are imposed at the boundaries of this model: all the wave energy is generated inside the domain. Shallow-water processes (refraction, shoaling, bottom friction and breaking) are considered in this model. The computational time-step is 4 min and the required CPU time to model a whole year is about 7 days on a basic PC. Bathymetric data has been obtained from ETOPO 1 (Amante & Eakins, 2009, http://www.ngdc.noaa.gov/mgg/global/global.html), and from the French Oceanographic Service (SHOM) with a high resolution along the French coasts (Figure 1). LONGITUDE (°) Figure 1: Computational mesh used for the wave simulations with TOMAWAC 2.3 Wind data used for the hindcast runs The wave model is forced with wind-fields from the NOAA/NCEP reanalysis (version 2), provided by the NOAA/OAR/ESRL PSD, Boulder, Colorado, USA, available on Internet (http://www.esrl.noaa.gov/psd/, Kanamitsu et al., 2002). We use the two components of U10 (the mean speed at 10 m above sea level) provided every 6 hours over a global Gaussian grid T62 which mesh size is about 1.875 degrees. This data is linearly interpolated in space on the computational grid, and also in time. The model is used to hindcast sea-state conditions over a period of 30 years from 1979/01/01 to 2008/12/31. 3. CALIBRATION PHASE In order to calibrate the numerical model we used significant wave heights Hmo measured with radar altimetry (Queffeulou & Croizé-Fillon, 2009) from 1992 to 2008. The ground tracks of the set of satellites selected for this calibration (ERS1, ERS2, ENVISAT, TOPEX, POSEIDON, JASON and GFO) on the Western Mediterranean Sea are represented in Figure 2. When using the wave model with its default parameters and original wind data coming from the NCEP-2 re-analysis we obtained a mean underestimation of the wave height Hmo from 0 to 20 % along the Italian and the Spanish coasts and from 0 to 40 % along the French coasts. The analysis of the comparisons between the model’s results and the altimeter measurements revealed that the difference varies with the position (x,y) in the Mediterranean Sea and with the wind direction. Then we determined a correction coefficient to be applied to the wind velocity depending of the position (x, y) and the wind direction in order to correct the average bias. Justification to this type of correction is supported by the fact that for some fetches, the wind velocity is not well estimated by the atmospheric model due to the presence of mountains and irregular coastlines, taking account of its rather coarse spatial resolution. 2 of 8 PIANC MMX Congress Liverpool UK 2010 LATITUDE (°) At the end, we have thus two time-series of results for wave parameters, labelled “original wind” and “modified wind”. LONGITUDE (°) Figure 2: Ground tracks of satellites used for the calibration phase 4. COMPARISON PHASE 4.1 Wave buoys used The simulation results obtained with “original wind” and “modified wind” are compared by using measured wave data from 27 wave buoys located in the Western Mediterranean Sea (see Figures 3 and 4 for the period of measurement and the position of these buoys). Eleven of these buoys come from the French CANDHIS database (http://candhis.cetmef.developpement-durable.gouv.fr/) of the CETMEF, two of these buoys come from METEO-FRANCE, nine of these buoys come from the Italian RON (Rete ondametrica Nazionale, http://www.envirtech.org/ron.htm), four of them come from the Catalan XIOM (Xarxa d’Instrumentacion Oceanografica I Meteorologica, http://www.boiescat.org) and one of them from Monaco. YEAR Figure 3: Period of measurement of wave buoys in the Western Mediterranean Sea used for comparison with model’s results 3 of 8 LATITUDE (°) PIANC MMX Congress Liverpool UK 2010 LONGITUDE (°) Figure 4: Location of wave buoys in the Western Mediterranean Sea used for comparison 4.2 Results ATLAS TIME (HOUR SINCE 01/01/1948) Figure 5: Comparison of model results with buoy data of Mazara for the last quarter of 1999 4 of 8 PIANC MMX Congress Liverpool UK 2010 The comparison of measured and computed time series of wave parameters (the significant wave height Hmo, the mean period Tm02, the mean direction, etc) shows that the results are clearly improved when using the “modified wind” (example on Figure 5). The NRMSE (Normalised Root Mean Square Error) for all buoys is 40 % in the “original wind” case and 30 % the “modified wind” case. Figure 6 which represents the distribution of between measures and model for 6 buoys shows this decrease of the NRMSE with the modification of the wind field. In the same way, the slope of the best-fit straight line, passing through the origin, is closer to 1. Figure 6: Distribution of Hmo between measurements and model for 6 buoys Figure 7: Q-Q plot between measurements and model for 6 buoys 5 of 8 PIANC MMX Congress Liverpool UK 2010 The analysis of Q-Q plots (quantile-quantile) (Figure 7), which show the correspondence between measured and computed quantiles (from 1 % to 99 %) of the statistical distribution of Hmo over the simulated period, shows also best results of the model with the modification of the wind field. 5. PRODUCTION OF SYNTHETIC CHARTS OF SIGNIFICANT WAVE HEIGHT LATITUDE (°) Once the numerical data-base over the period 1979-2008, calibrated with altimeter data and compared with buoy data, has been built, one can use the time series of wave parameters to produce synthetic charts covering the Mediterranean sea. Examples are given here for the mean value of the significant wave height (Figure 8), the mean wave power (Figure 9) and the Q99 value of the significant wave height (i.e. the wave height which is on the average exceeded by 1% of the data) (Figure 10) LONGITUDE (°) LATITUDE (°) Figure 8 – Chart of the mean value of wave height Hmo computed over the period 1979-2008. LONGITUDE (°) Figure 9 – Chart of the mean wave power computed over the period 1979-2008. 6 of 8 LATITUDE (°) PIANC MMX Congress Liverpool UK 2010 LONGITUDE (°) Figure 10: Chart of the Q99 of wave height Hmo computed over the period 1979-2008. LATITUDE (°) 6. CONCLUSIONS AND FUTURE WORKS LONGITUDE (°) Figure 11: Three levels of data for the Western Mediterranean Sea available through the ANEMOC website The ANEMOC data-base on the Atlantic Ocean, the English Channel, the North Sea and the Mediterranean Sea is now complete in a first version. Through the present work, an extension to the Mediterranean Sea has been accomplished, covering a period of 30 years (from 1979/01/01 to 2008/12/31). Part of the results obtained during this work will be available on the present ANEMOC website. 3 levels of data are identified (Figure 11): • Level 1 : Time series of wave parameters with an hourly time-step on 250 sites. (Free for research applications, Non-free for commercial uses) • Level 2 : Synthetic charts of mean wave climate (year, summer or winter) on 50 sites. (Free) • Level 3 : Results of extreme value analysis for Hmo on 25 sites, based on POT (Peak-Over Threshold method. (Free) 7 of 8 PIANC MMX Congress Liverpool UK 2010 We are presently working (i) on a more detailed analysis and description of the sea-state parameters (first and second swell, wind-sea, archiving), (ii) on a better modelling of physical processes in the wave model, (iii) on the extension towards shallow water depth, (iv) on the possible use of meso-scale wind-fields and (v) on data assimilation (altimeter wave heights). 7. ACKNOWLEDGEMENTS This project is a joint collaboration between EDF-LNHE and CETMEF. We thank Météo-France, XIOM, RON, IFREMER, NOAA and SHOM for access to their data. 8. REFERENCES Aelbrecht, D., Benoit, M., Marcos, F. & Goasguen, G. (1998). Prediction of offshore and nearshore th storm waves using a third generation spectral wave model, Proc. 8 Int. Offshore and Polar Eng Conf. (ISOPE), vol. III, 71-76. Amante, C. & Eakins, B. W. (2009) ETOPO1 1 Arc-Minute Global Relief Model: Procedures, Data Sources and Analysis. NOAA Technical Memorandum NESDIS NGDC-24, 19 pp, March 2009. Benoit, M., Marcos, F. & Becq, F. (1996). Development of a third generation shallow-water wave th model with unstructured spatial meshing, Proc. 25 Int. Conf. on Coastal Eng., ASCE, Orlando (Florida, USA), 465-478. Benoit, M. & Lafon, F. (2004). A nearshore wave atlas along the coasts of France based on the numerical modelling of wave climate over 25 years. Proc. 29th Int. Conf. on Coastal Eng. (ICCE'2004), 19-24 September 2004, Lisbon (Portugal), pp 714-726. Benoit M., Lafon F. & Goasguen, G. (2008). Constitution et exploitation d’une base de données d’états de mer le long des côtes françaises par simulation numérique sur 23 ans. Base ANEMOC en Atlantique – Manche – Mer du Nord. European Journal of Env. and Civil Eng., Vol. 12/1-2, pp 35-50 (in French). Bretherton, F.P. & Garret, C.J.R. (1969). Wavetrains in inhomogeneous moving media, Proc. Roy. Soc. London, Series A, 302, 529-554. Kanamitsu, M., Ebisuzaki, W., Woollen, J., Yang, S-K, Hnilo, J. J., Fiorino, M. & Potter, G. L. (2002). NCEP-DEO AMIP-II Reanalysis (R-2)1631-1643, Bul. of the Atmos. Met. Soc. Komen, G.J., Cavaleri, L., Donelan, M., Hasselmann, K., Hasselmann, S. & Janssen, P.A.E.M. (1994). Dynamics and modeling of ocean waves, Cambridge Univ. Press, 532 p. Lafon, F. & Benoit, M. (2006). Estimation of extreme wave conditions from hindcast simulations with application to the wave climate along French coasts. Proc. 30th Int. Conf. on Coastal Eng. (ICCE'2006), 3-8 September 2006, San Diego (California, USA), pp 739-751. Queffeulou, P. & Croizé-Fillon, D. (2009). La mesure satellite de hauteur de vague par altimètre. État des lieux, application à la climatologie et à la modélisation des états de mer. AMA 2009. Les ateliers de modélisation de l'atmosphère, Toulouse, 27-29 janvier 2009. 8 of 8 PIANC MMX Congress Liverpool UK 2010 LE CENTRE DE GESTION DE LA NAVIGATION SUR LE RHONE par 1 2 Pierre-Emmanuel Pareau , Romain Barthelet and Magali Neymarc 3 RESUME La Compagnie Nationale du Rhône concessionnaire du Rhône s’est engagée dans un projet de «Modernisation de la voie navigable» dont l’objet est de passer, par étapes, de la «manœuvre des écluses» à la «gestion du trafic», en créant un centre unique pour la gestion du trafic et la téléconduite des 14 écluses du Rhône. Les cinq premières écluses ont été mises en service en 2009, les 9 autres écluses le seront en 2010 et 2011. L’originalité de ce centre, outre le nombre d’écluses conduites depuis un seul point est la souplesse de fonctionnement qui permet de conduire n’importe quelle écluse depuis n’importe quel demi-pupitre, offrant ainsi une grande capacité d’adaptation en fonction du trafic et du nombre de personnes présentes. Le projet a été entièrement conçu et développé par les équipes d’ingénierie de la CNR Figure 1 : Photo d’un pupitre au CGN 1 Directeur du projet, Compagnie Nationale du Rhône, [email protected] Responsable de la maîtrise d’œuvre technique, Compagnie Nationale du Rhône, [email protected] 3 Chargé de maîtrise d’ouvrage Compagnie Nationale du Rhône, [email protected] 2 1 sur 17 PIANC MMX Congress Liverpool UK 2010 2009 2010 2011 Centre de gestion de la navigation (CGN) Figure 2 : Planning de mise en œuvre de la téléconduite des écluses sur le Rhône 1. INTRODUCTION Créée depuis plus de 70 ans pour gérer la concession du Rhône, la CNR a, au cours de ces décennies, conçu et exploite les aménagements à vocation multiple construits depuis la frontière suisse jusqu’à la mer Méditerrannée pour la production hydroélectrique et la navigation. Au titre de ses missions de Service Public, la CNR a souhaité conforter son engagement dans la modernisation de la navigation. A cet effet, La CNR a mis en œuvre dès 2004 un plan dont les actions principales sont : 2 sur 17 PIANC MMX Congress Liverpool UK 2010 L’amélioration de la fiabilité des écluses, La mise en service d’un système d’information pour les usagers(Inforhone.fr), la réalisation des cartes ECDIS, La création de nouvelles infrastructures (appontements, terminal à conteneurs au Port de Lyon). L’évolution récente du trafic et les perspectives de développement font apparaître la nécessité de s’adapter pour faire face à de nouveaux enjeux : La sécurité du transport (suivi des matières dangereuses, transport de passagers en hausse) et les règlements de plus en plus stricts, L’information des usagers et le suivi des marchandises au sein de la chaîne logistique, La charge supplémentaire pour les écluses. (doublement du trafic entre 1998 et 2009). Dans le cadre de son deuxième plan à 5 ans de Missions d’Intérêt Général, la CNR doit poursuivre son action dans les domaines suivants : Poursuite et accélération des programmes de modernisation et de fiabilisation des écluses, Gestion du trafic, apport de nouveaux services aux navigants, en accompagnant l’augmentation du trafic et en prenant en compte les nouvelles technologies de l’information, Elargissement des heures d’ouverture à la navigation de commerce (24h/24h). La Compagnie s’est donc engagée dans un projet de «Modernisation de la voie navigable» dont l’objet est de passer, par étapes, de la «manœuvre des écluses» à la «gestion du trafic», en créant un centre pour la gestion et la surveillance du trafic, et en accélérant le programme de fiabilisation et de modernisation des écluses. La première composante est la réalisation d’un Centre de Gestion de la Navigation (CGN) dont la mission sera, à terme, de gérer le trafic en temps réel et de téléconduire 24h/24h les écluses du basRhône. Une deuxième composante, menée en parallèle, est la poursuite de la mise à niveau des automatismes des écluses (sécurité des usagers et adaptation aux technologies nouvelles) et une fiabilisation générale de l’ensemble des organes pour assurer le niveau de service requis face à un trafic en forte croissance. 2. LES PRINCIPES GENERAUX La création du CGN répond aux besoins d’améliorer le niveau de service du trafic et à la nécessité de gérer au mieux la voie navigable et le fonctionnement des écluses. La gestion de trafic suppose de connaître en permanence l’état du trafic, les conditions hydrométéorologiques, la disponibilité des ouvrages, les conditions particulières (avis à la batellerie, incidents, avaries,…) la position des bateaux, les marchandises transportées, pour prévoir les passages aux écluses, fournir des informations et guider les navigants. Par ailleurs les écluses doivent être conduites dans les conditions les plus efficaces, pour limiter les temps d’attente et les fausses bassinées. La gestion de trafic doit recueillir, centraliser et analyser l’ensemble des informations nécessaires à la maitrise d’une situation en permanente évolution. La conduite des écluses nécessite des échanges avec les navigants, la prévision du trafic, la définition de l’ordre de passage des bateaux, le lancement et la surveillance des opérations d’éclusage, l’alimentation d’une base de données. 3 sur 17 PIANC MMX Congress Liverpool UK 2010 La gestion de trafic et la conduite des écluses sont donc complémentaires, tout en présentant des tâches communes notamment dans la collecte et la diffusion des informations et l’organisation des passages. La gestion permet de passer d’une vision parcellaire autour de chaque écluse à une vision globale sur l’ensemble du tronçon considéré. Dans le temps, la charge de travail de gestion, de conduite, de surveillance fluctue en fonction de la situation globale, de l’intensité du trafic, des sollicitations extérieures. SURVEILLANCE DU CHENAL ET DES ECLUSES ENREGISTREMENT DES ECLUSAGES ECHANGES AVEC INTERFACES PAM ET COUZON ECHANGES PORTS CONNAISSANCE EN TEMPS REEL POSITION BATEAUX GESTION DE CRISE PREVISION DES PASSAGES ET PLANNING DES OPERATEURS AVIS BAT ET ECHANGES AUTORITES ETAT DES INFRASTRUCTURES INFORMATION DES BATEAUX SUR HEURES DE PASSAGE AFFECTATION DES ECLUSAGES ET COMMUNICATIONS AVEC BATEAUX ECLUSAGES ET BASE DE DONNEES Figure 3 : Schéma fonctionnel de la gestion de trafic Concevoir une organisation de ces activités conduit à rechercher une solution qui : Procure la réactivité nécessaire pour s’adapter en temps réel, Assure la fluidité des opérations, Et garantisse le niveau de sûreté, la sûreté du trafic et la sécurité des biens et des personnes -dans ou à proximité des écluses- étant la priorité de la CNR et primant sur la gestion de la production hydroélectrique. 3. L’ORGANISATION DU PROJET Le projet est géré complètement par la CNR. La Direction de l’exploitation de la CNR assure le rôle de maître d’ouvrage du projet, elle a défini les objectifs, les fonctionnalités, suit les plannings de mise en œuvre et contrôle le budget global de l’opération. La Direction de l’Ingénierie de la CNR assure la maîtrise d’œuvre générale du projet, la conception technique générale et le suivi de la mise en œuvre. Elle réalise également les SCADA et automatismes 4 sur 17 PIANC MMX Congress Liverpool UK 2010 Les travaux d’adaptation électrique et de contrôle-commande, les développements des logiciels de gestion de la vidéo et de gestion des communications locales sont sous-traitées à des entreprises extérieures spécialisées, supervisées par la maitrise d’œuvre CNR. 4. PERIMETRE ET PRINCIPES GENERAUX DE FONCTIONNEMENT DU PROJET Le périmètre du projet concerne les 14 écluses du Bas-Rhône, y compris celles de Port Saint Louis et Barcarin. Chaque opérateur du centre de téléconduite a la possibilité technique de conduire n’importe quelle écluse raccordée au centre. Il peut superviser 2 éclusages simultanés, à l’exception de l’écluse de Port Saint Louis qui est commandée seule en raison de la présence d’un pont levant. L’opérateur en téléconduite conduit l’écluse selon les mêmes principes que localement (commandes de cycles). Il dispose d’interfaces similaires (mêmes boutons de commandes) et d’équipements vidéo et de sonorisation, comme en conduite depuis le mirador. Les écluses peuvent être conduites localement depuis les miradors. Le projet de téléconduite fait l’objet d’études de sûreté et d’une phase d’expérimentation sur 2 écluses. Le centre de téléconduite dispose d’une ergonomie de commande et d’information homogène pour l’ensemble des écluses. Figure 4 : Ecran de supervision d’une écluse Les équipements mis en œuvre au centre de téléconduite s’appuient sur une architecture similaire à celle de la téléconduite des usines hydroélectriques de la CNR, sur base microSCADA ABB. 5 sur 17 PIANC MMX Congress Liverpool UK 2010 La communication s’appuie sur le réseau informatique de la CNR existant, aussi bien pour les échanges de contrôle-commande que pour la vidéo, la sonorisation, etc. La téléconduite s’adapte aux systèmes existants. Les plus anciens sont modernisés, et de nouveaux équipements seront déployés ou adaptés sur les écluses de manière à permettre la téléconduite (équipements vidéo, interphonie, ….) et à répondre aux spécifications des études de sûreté et d’ergonomie. Les adaptations locales des automatismes sont mises en service : Sans gêne ni risque pour la navigation : les tests sont effectués sur plate-forme et lors des arrêts de navigation programmés (mois de mars), Avec le minimum de gêne pour l’ensemble des autres interventions concernant les écluses (maintenance génie-civil, mécanique, rénovations électriques, …). Cette contrainte sera prise en compte lors du phasage du déploiement. 5. L’ORGANISATION ET LE FONCTIONNEMENT DU CGN 5.1 Organisation La définition de l’organisation du centre de gestion de la navigation nécessite la connaissance et la prise en compte : Des éléments qui caractérisent le trafic fluvial, Des tâches à réaliser qui découlent du fonctionnement adopté. L’analyse du trafic a montré une forte variabilité saisonnière, le nombre d’éclusages varie de 4 000 en janvier à 11 000 en juillet Le trafic quelle que soit la saison est concentré sur la journée, le trafic nocturne ne représente que 13% du trafic total. Le nombre de techniciens navigation est défini pour que la navigation soit assurée dans les mêmes conditions qu’actuellement (en terme de délai de passage et de surveillance des opérations). Il y a donc le même nombre de techniciens navigation toute l’année, même si le trafic est moins important entre octobre et mars, et pourrait justifier un effectif inférieur. Le nombre de techniciens est fixé à 32, organisés en équipes effectuant des postes en 3 x 8 ou en 2 x 8. A terme, il y aura 7 techniciens en poste entre 6 h et 22 heures et 2 ou 3 techniciens la nuit suivant l’évolution du trafic nocturne. Trois cadres assurent le management des équipes. 5.2 Principes de fonctionnement Un technicien navigation peut sélectionner sur son poste n’importe quelle écluse (validée en mode conduite distante) qu’il va gérer en configuration CONDUITE ou en configuration SURVEILLANCE. Un technicien navigation peut effectuer au maximum deux tâches simultanément : Il peut surveiller deux sélections d'écluses, (une sélection par ½ poste), Il peut conduire deux écluses (1 écluse par ½ poste), Il peut conduire une écluse (sur un ½ poste) et surveiller une sélection d'écluses (sur l'autre ½ poste). Nota : fait exception l’écluse de Port St Louis du Rhône qui possède un pont levant : lorsqu'un technicien navigation sélectionne cette écluse en conduite, il ne peut plus conduire une autre écluse ni surveiller une sélection d’écluses. 6 sur 17 PIANC MMX Congress Liverpool UK 2010 Figure 5 : Implantation schématique des équipements d’un pupitre de téléconduite En cas de défaillance du système, toutes les écluses sont commandables depuis le mirador, la reprise en commande locale est effectuée en moins de 30 minutes. 6. LA SURETE DE FONCTIONNEMENT La sûreté de fonctionnement garantissant la sécurité des usagers et du personnel et la fiabilité des ouvrages est l’une des clés du succès du projet de téléconduite des écluses. Cet élément a été pris compte dès le début de la conception et le sera jusqu’à la mise en service du nouveau système. La démarche retenue est globale, elle s’articule autour : D’une action continue aux divers stades du projet pour définir les études de sûreté à réaliser, les points sensibles à étudier. Elle se concrétise par un appui au maitre d’ouvrage assuré par un consultant extérieur indépendant, D’un groupe de travail, dédié aux risques, composé de représentants des concepteurs (MOA et MOE), et de professionnels de l’activité navigation de la CNR (éclusiers, techniciens, managers locaux). Lors de ses différentes réunions, il a alimenté la réflexion globale autour des études de risques, De la réalisation d’études de risques du type loi SIST pour chaque écluse mise en téléconduite. L’objet des études réalisées est de présenter un Dossier Préliminaire de Sécurité conformément à loi SIST 2002-3 datée du 3 janvier 2002 relative à la sécurité des infrastructures et des systèmes de transport. Bien que le décret d’application relatif à la Sécurité des voies navigables n’ait pas été publié, la CNR a cependant décidé comme elle l’a fait pour la mise en œuvre de la téléconduite de l’écluse de Barcarin, de suivre les principes de la loi SIST en faisant réaliser : Un Dossier Préliminaire de Sécurité (DPS) avant la réalisation des travaux, Un rapport de sécurité établi par un expert ou un organisme qualifié, 7 sur 17 PIANC MMX Congress Liverpool UK 2010 Un Dossier de Sécurité (DS) avant la mise en exploitation. L’analyse de risques réalisée sur les écluses de Pierre-Bénite, Bourg les Valence et Avignon prenant en compte les installations actuelles, conduites en locale, a montré que le niveau de sécurité est globalement satisfaisant. Le système actuel a donc servi de référence pour évaluer le caractère GAME (Globalement Au Moins Equivalent) du niveau de sécurité du projet de téléconduite. Le principal impact, sur la sécurité, de la mise en place d’une téléconduite, est de supprimer à la fois la présence de l’éclusier pouvant intervenir au niveau du sas et le contrôle visuel effectué par l’éclusier dans un nombre important de situations plus ou moins dangereuses. Par conséquent, les 3 mesures envisagées, indispensables, qui permettent de compenser cet éloignement de l’éclusier sont les suivantes : Le renforcement du système vidéo composé de caméras, notamment aux têtes des écluses, qui a nécessité : o La réalisation d’une étude ergonomique sur l’aménagement du CGN et de la supervision, o La réalisation d’une étude sur le positionnement des caméras. L’appel à l’astreinte avec une intervention rapide en cas de problème technique ou d’incident, La présence de saisonniers à certaines écluses pour informer les plaisanciers sur la bonne marche à suivre. En conclusion, l’analyse de risques a démontré que pour chaque événement à risque identifié, les mesures de réduction du risque envisagées étaient jugées comme étant suffisantes pour que le niveau de sécurité de l’installation future soit globalement satisfaisant et équivalent au niveau de sécurité de l’installation de référence. 7. ARCHITECTURE L’architecture de téléconduite comprend 4 grands systèmes qui sont : Le contrôle-commande des manœuvres d’éclusage à distance, Le système d’arrêt d’urgence pour la mise en sécurité de l’écluse par coupure des alimentations de contrôle-commande et de puissance, La vidéosurveillance pour le contrôle visuel autour du sas et aux abords de l’écluse, Le système de gestion des communications vocales (radio VHF avec les navigants, sonorisation du sas et des garages pour diffusion de messages parlés ou pré-enregistrés, et téléphonie). L’ensemble de ces systèmes utilisent le réseau fibre optique de la CNR, réseau informatique d’un débit de 1Gbits/s permettant l’échange de flux vidéo en temps réel. Ce réseau est doublé par un réseau de secours qui permet en cas de coupure du réseau principal de garantir une bande passante de 1Gbits/s dédiée à la téléconduite. 8 sur 17 PIANC MMX Congress Liverpool UK 2010 Figure 6 : Schéma de principe des 4 systèmes 7.1 Le système de contrôle commande Il consiste à interfacer les automatismes locaux (automates PLC de conduite baptisés « APN », automates encadrant la sécurité ultime appelés « CSU », automates de services généraux APX) avec les systèmes informatisés d’interface homme-machine appelés « SCADA » et développés sur la base du logiciel micro-SCADA d’ABB. 9 sur 17 PIANC MMX Congress Liverpool UK 2010 Figure 7 : Système de contrôle-commande Les SCADAs locaux dialoguent avec les serveurs centraux SCADAs au CGN qui pilotent l’ensemble des affichages et commandes sur les 8 pupitres opérateurs du CGN. Le SCADA central CGN joue le rôle d’aiguilleur principal : il permet de mettre en relation un pupitre opérateur avec une ou plusieurs écluses. Op1 Op2 Op3 Op4 Op5 Op6 Op7 TCN (chef op) Système d’aiguillage du process (contrôlecommande) CGN Sites (écluses) PB VS SA GN BV BE LN CH BO CA AV BC BA SL Figure 8 : Système d’aiguillage L’opérateur du CGN dispose des informations du process des écluses qu’il conduit. Ces informations sont affichées sur 5 écrans. Ces 5 écrans permettent à l’opérateur de conduire 2 écluses : Sur le 1er écran : positions et états des organes d’une 1ère écluse, Sur un 2ème écran : défauts frappants les organes ou le système de la 1ère écluse, 10 sur 17 PIANC MMX Congress Liverpool UK 2010 Sur un 3ème écran : positions et états des organes d’une 2ème écluse, Sur un 4ème écran : défauts frappants les organes ou le système de la 2ème écluse, Sur un 5ème écran : sélection des écluses prises en charge. Les écrans 1 et 2 forment le ½ pupitre gauche. Les écrans 3 et 4 forment le ½ pupitre droit. L’opérateur peut également, en utilisant ses 5 écrans, adresser des macro-commandes interprétées et contrôlées par l’automate APN du site : Des commandes de cycles : « cycle montant », « cycle avalant », « arrêt », « validation », Des commandes de sélections (sélections d’organes, de fonctions, …). L’opérateur au CGN dispose d’un bouton poussoir placé sur chaque ½ pupitre de son poste. Ce bouton lui permet d’arrêter de manière sûre (avec coupure de l’énergie électrique) l’ensemble des organes de l’écluse qu’il conduit. Le traitement de cet arrêt dit « arrêt process » est assuré par la voie normale du système de contrôle-commande. Figure 9 : Poste opérateur CGN – partie contrôle- commande Les serveurs locaux et centraux sont redondés en nombre et dans des sites géographiques distincts de manière à assurer une grande disponibilité du système. Les développements de ces systèmes qui sont jugés comme faisant partie du « cœur de métier » de la CNR ont été réalisés par des équipes internes dans l’objectif de maîtriser pleinement la maintenance de ces outils. Ces mêmes outils (automates, serveurs SCADAs) sont d’ailleurs utilisés pour d’autres systèmes de conduite de la CNR, en particulier pour toute la gestion de la production électrique. 7.2 L’arrêt d’urgence distant L’arrêt process décrit précédemment utilise la voie normale et informatisée du système de contrôlecommande. Il peut difficilement, en termes de probabilité de défaillance, être accepté comme un véritable « arrêt d’urgence », au sens des normes de sécurité. C’est pourquoi la fonction arrêt d’urgence est développée à partir d’automates PLC spécifiques baptisés « APS ». Chaque écluse dispose d’un automate APS relié à son correspondant au CGN par le réseau de communication Ethernet IP de la CNR. La voie de sécurité ainsi créée permet d’acheminer la 11 sur 17 PIANC MMX Congress Liverpool UK 2010 demande d’arrêt d’urgence que l’opérateur du CGN peut activer depuis un bouton poussoir placé sur un synoptique mural. Le synoptique mural est équipé de 14 boutons poussoirs (1 pas écluse). La mise en œuvre, l’installation, la programmation et la maintenance des APS a été spécifiée conformément à la norme IEC 61-508. Le niveau de « SIL » (Safety Integrity Level) défini par cette norme a été évalué pour les écluses. Le niveau 3 a été retenu. L’ensemble du système d’arrêt d’urgence, depuis le BP de déclenchement, jusqu’aux actionneurs sur sites a été étudié, développé, et a obtenu la certification SIL3. Figure 10 : Le système d’arrêt d’urgence distant 7.3 Le système de Vidéo surveillance Basé sur un référentiel d’environ 16 caméras positionnées judicieusement autour du sas et dans les zones d’approche, le système de vidéo surveillance du CGN permet la commande des caméras et la visualisation des images sur 3 écrans dédiés sur chaque ½ pupitre de commande. Les caméras doivent permettre de contrôler la position des bateaux dans le sas, le bon amarrage des bateaux et notamment les plaisanciers moins aguerris à cette manœuvre, et de vérifier visuellement le bon déroulement de l’éclusage. Dans les écluses, des encodeurs sont mis en place. Ces équipements électroniques transforment le signal vidéo analogique reçu des caméras en un signal numérique, qui est ensuite compressé selon la norme MPEG4. Chaque encodeur diffuse ensuite les images en multicast sur le réseau IP. L’encodeur est paramétré de façon à diffuser les images avec : Une qualité CIF (352x288 pixels), 2 CIF (704x288 pixels) ou 4 CIF (704x576 pixels), Un rafraichissement 25, 12 ou 6 images / secondes, Un débit IP qui peut être bridé (300kbits/s à 3000kbits/s). Afin d’obtenir une bonne qualité, la plupart des images sont diffusées 4 CIF, 25 images/secondes sans bridage de débit. Au CGN, un serveur principal vidéo gère l’ensemble des images et assure leur diffusion depuis les encodeurs situés sur les sites. Ce serveur vidéo est en lien avec le serveur SCADA du CGN afin de 12 sur 17 PIANC MMX Congress Liverpool UK 2010 connaître les écluses conduites par l’opérateur de chaque pupitre. Il est redondé pour garantir une bonne disponibilité du système vidéo. Les images sont enregistrées numériquement sur un serveur et accessibles depuis un poste spécifique réservé uniquement à la maintenance. Les opérateurs ne disposent que des images temps réel. Le serveur d’enregistrement sert de serveur de secours en cas de perte de la liaison avec le serveur du CGN. Figure 11 : le système vidéo 7.4 Le système de communication vocale Ce système permet de disposer au CGN des mêmes moyens de communications que ceux dont disposait l’éclusier dans son mirador. Pour cela, les communications locales, qu’elles soient téléphoniques, par radio VHF, ou de la sonorisation par haut-parleurs, sont remontées via le réseau informatique CNR, jusqu’au CGN. L’architecture du système est un peu similaire à celle du contrôle-commande et de la vidéo, avec : Un système d’acquisition local : ici, pour l’audio, le système est disjoint : o Pour la téléphonie, l’acquisition se fait par des systèmes de téléphonie appelés « Media Gateway », qui sont des convertisseurs téléphonie analogique vers IP. Ils permettent la prise en charge des appels téléphoniques classiques aboutissant aux écluses 13 sur 17 PIANC MMX Congress Liverpool UK 2010 o Pour la radio VHF et la sonorisation, l’acquisition est faite par un équipement appelé « ATL » (adaptateur à la téléphonie) qui permet de convertir la VHF et la sonorisation (signaux de commande et signaux voix) sur l’IP. Un système central de gestion et d’aiguillage : le système est baptisé CTI (couplage téléphonie – informatique). Il s’agit d’un serveur central du CGN qui met en communication les interfaces audio de l’opérateur avec les systèmes des écluses dont il a la charge. L’opérateur dispose sur son pupitre d’une interface « Audio » unique et ergonomique qui lui permet d’utiliser n’importe laquelle des voies de communications raccordées : VHF, téléphone, sonorisation, interphonie. Figure 12 : Système de Communication Vocale 14 sur 17 PIANC MMX Congress Liverpool UK 2010 Figure 13 : Vue du poste de commandes vocales (VHF, téléphone, sonorisation) 8. LES TRAVAUX Les travaux comprennent essentiellement : L’aménagement d’une plate-forme de développement–maintenance à Pierre Bénite. Cette plateforme accueille l’ensemble de l’équipe projet, la maitrise d’œuvre et les équipes de développement. Elle permet également de tester l’ensemble des systèmes avant leur déploiement en production. Les tests sont conduits en simulant par logiciel le comportement hydraulique, électrique, et l’automatisme de 2 écluses. 2 postes opérateurs du CGN sont créés sur cette plateforme. Les travaux d’aménagement du centre à Châteauneuf et l’installation des équipements de conduite à distance. Ces travaux ont été menés de septembre 2008 à février 2009. Le développement des logiciels spécifiques et l’acquisition du matériel informatique (vidéo, communication vocale, conduite). Le développement des logiciels liés au contrôle-commande a été réalisé par les équipes CNR. Le développement des systèmes vidéo et de communication vocale ont été tous 2 confiés à des sociétés externes, sous maîtrise d’œuvre CNR. L’adaptation locale des écluses, comprenant notamment les travaux sur les automatismes, l’amélioration des équipements vidéo. Cette adaptation se fait selon un cadencement défini par les périodes annuelles des arrêts de navigation du Rhône qui ont lieu chaque année au mois de mars pour une durée variant de 7 à 10 jours. Les modifications électriques ou d’automatismes sont préparées l’année précédente (d’avril à septembre), puis déployées d’octobre à février, en parallèle des installations existantes et enfin raccordées lors des arrêts de navigation de mars. Ainsi, en mars 2008, 2 écluses, Avignon et Bourg-les-Valence ont pu être préparées localement. En mars 2009, 3 autres écluses ont été « adaptées ». 5 le seront en mars 2010 et les 4 dernières en mars 2011. Une fois adaptées, les écluses peuvent être raccordées au centre en dehors des arrêts de navigation. Les 2 premières, Avignon et Bourg- 15 sur 17 PIANC MMX Congress Liverpool UK 2010 les-Valence l’ont été en avril 2009, les 3 suivantes en novembre 2009. Les prochaines sont prévues à l’automne 2010, et les 4 dernières à l’automne 2011. Le raccordement des écluses de Port Saint Louis et Barcarin au réseau fibre optique de la CNR : ces 2 écluses au sud du Rhône ne sont pas raccordées directement au réseau Ethernet de la CNR. A l’heure actuelle, ce raccordement est en phase d’étude et plusieurs pistes sont explorées : liaisons par faisceau hertzien, fibre optique, liaison par opérateur télécom (Orange, SFR, …). Les aspects coûts, disponibilité, sécurité et délais de mise en œuvre sont pris en compte dans l’étude. 9. LE SYSTEME DE TELEMAINTENANCE Parallèlement au système de téléconduite, un logiciel de maintenance du système a été développé par les équipes de la CNR. Cette application permet de surveiller à distance les équipements du centre de gestion de la navigation et des écluses téléconduites. Plusieurs écrans permettent de suivre les opérations d’éclusage en temps réel de vérifier le fonctionnement du système global et faire un premier diagnostic ou de visualiser les défauts et les alarmes. Outre une ergonomie bien adaptée, les écrans sont accessibles à l’ensemble des personnels concernés (exploitation, maintenance) via l’intranet CNR. L’avantage est d’informer en temps réel les services concernés par la téléconduite et de faciliter les opérations de télémaintenance et télédiagnostic Figure 14 : Vue de l’écran de contrôle du CGN 16 sur 17 PIANC MMX Congress Liverpool UK 2010 Figure 15 : Vue alarmes 17 sur 17 THE GREENING OF INLAND NAVIGATION – THE CASE OF RHINE NAVIGATION by Gernot Pauli1 In the debate about the future of the world's economies and in particular its transport systems, the concept of 'greening' plays a prominent role. In general, greening describes a process, whereby giving consideration to environmental impacts, economic activities become more sustainable. In this context, inland navigation is often cited as the most environmentally friendly and sustainable mode of transport, with fewer harmful emissions and less consumption of resources than road or rail transport. Using navigation in the Rhine basin as an example, this paper sets out • to prove that the greening of inland navigation is already taking place, • to explain, how that has happened, and • to show, that the greening of inland navigation can be achieved without jeopardizing its economic advantage. As the paper will examine public policy as well as technical and economic aspects of greening, it should be of interest to a larger part of PIANC community. 1. INTRODUCTION – GREENING OF TRANSPORT: WHAT AND WHY? The words ‘green’ or ‘greening’ in relationship to economic activities are most likely understood and used differently, depending on the roles of the respective stakeholders. For example, customers may understand green products and services as being good for their health and their environment; politicians may have a broader view seeing green policies as those that improve the quality of life for their constituencies; scientists may focus mainly on emissions and the use of natural resources. Therefore, it seems necessary to find a generally acceptable reference as the basis for an analysis of the greening of inland navigation. The United Nations Environmental Programme (UNEP), has developed a concise understanding of greening. According to UNEP, “Greening the economy refers to the process of reconfiguring businesses and infrastructure to deliver better returns on natural, human and economic capital investments, while at the same time reducing greenhouse gas emissions, extracting and using less natural resources, creating less waste and reducing social disparities.”2 UNEP (2009) advocates developing “… more energy-efficient, less polluting transport modes and infrastructure, improved public transportation, and the use of greener vehicles”. Whereas rail is given as an example for investments into energy efficiency and low carbon mobility, inland navigation is not even mentioned in the relevant UNEP policy document. The International Transport Forum (ITF)3 refers to a report by the Organisation for Economic Cooperation and Development (OECD) (2009), which argues that the greening of the economy is a forward-looking structural measure to strengthen the economy. The report mentions France’s planned investments into river canal projects as an example of ‘green elements’ of current country stimulus policies. In the USA, the Department of Transportation (2009) also acknowledges green transportation in the context of a government stimulus package. The Department seems basically to understand green transportation as public transport with reduced consumption of fossil fuels. Of greater interest in the 1 2 3 Chief Engineer, Central Commission for the Navigation of the Rhine, [email protected] . This paper reflects the views of the author, which may differ from positions or statements by the CCNR or its member states. The UNEP Green Economy Initiative (GEI) webpage http://www.unep.org/greeneconomy/ . Initially envisioned as a two-year project, the GEI has been expanded to include a number of related UNEP and UN-wide initiatives focused on providing macroeconomic evidence for significantly increasing investments in the environment as a means of promoting sustainable economic growth, decent job creation, and poverty reduction. ITF, transformed from the European Conference of Ministers of Transport (ECMT), is an inter-governmental organisation within the OECD family. ITF is a global platform and meeting place at the highest level for transport, logistics and mobility. Key figures from government and politics, business and industry, research and civil society will meet in Leipzig each year to debate a transport topic of worldwide strategic importance. The topic of the 2008 forum was climate change. Further information can be found on ITF’s webpage www.internationaltransportforum.org PIANC MMX Congress Liverpool UK 2010 context of this paper is the proposal for a ‘Green Transportation Infrastructure Research and Technology Transfer Act’, which was presented to the US House of Representatives in 2008. The act defines “green transportation infrastructure” as infrastructure that: “(1) preserves and restores natural processes, landforms (such as floodplains), natural vegetated stream side buffers, wetlands, or other topographical features that can slow, filter, and naturally store stormwater runoff and floodwaters for future water supply and recharge of natural aquifers; (2) uses natural design techniques to manage stormwater; and (3) minimizes lifecycle energy consumption and air pollution.”4 Even though this definition is somewhat arbitrary with it focus on stormwater, at the same time, it enlarges the term greening by including aspects of land use and preservation as well as restoration of natural processes. The United Nations Economic Commission for Europe (UNECE), with its important transport programme, is supporting several initiatives related to transport and the environment5. UNECE (2009) has seen it as “important … to fully streamline climate change as well as the concept of the Green Economy/Green New Deal in all its work.” Nevertheless, UNECE does not yet seem to have developed a definition of green transport, nor does it even use the term. The European Commission (2008), to the contrary, has assigned the title ‘Greening Transport’ to a communication to the European Parliament and the Council. This communication has to be seen as an important policy paper “to make transport greener and more sustainable”. Within the communication, the Commission groups the proposed or already taken measures according to the major impact they address, namely climate change, local pollution, noise pollution, congestion and accidents. The Commission uses the communication as an explanation for its renewed initiative “to ‘get the prices right’ through internalising the external costs of transport”. The European Parliament (2009), while welcoming the Commission communication “as an important first partial step towards a more comprehensive approach making for more environmentally friendly transport …”, criticizes “… the Commission for the lack of an overall strategy on the greening of transport and the absence of a usable model for assessing external costs such as the environmental impact of transport.” Parliament, however, does not mention the absence of a definition for “Greening Transport”. With the reference to making transport “greener and more sustainable” and by mentioning not only environmental, but also economic and social impacts, Commission and Parliament seem to view ‘greening transport’ more or less the same as ‘making transport more sustainable’. The European Economic and Social Committee (EESC 2009) specifically refers to the ‘Greening of Inland Waterway Transport’ in one of its exploratory opinion, without giving any explanations of this term. In its opinion, the states that inland navigation is “competitive, sustainable and environmentally friendly” and that its environmental performance “… should be seen against the performance of EU land transport …” In 2009, the European Barge Union (EBU), the most important trade federation of the European inland navigation sector, organized a seminar entitled ‘Greening Transport and Inland Navigation’6, and dedicated to the aforementioned Commission communication. However, it is not clear, whether EBU wishes to use the term ‘Greening of Inland Navigation’ or to develop a definition or common understanding of its meaning. A common thread in the previous description of the term ‘greening’ is the demand to better use natural resources and to reduce pollution and greenhouse gas emissions. UNEP, UNECE, OECD and the European Commission also include economic and social aspects in their understanding of greening, creating a link to ‘sustainability’ and, in the context of this paper, to ‘sustainable transport’. ‘Sustainable transport’ or ‘sustainable mobility’, as opposed to ‘greening transport', is a well developed and widely accepted concept with clear definitions. According to the Centre for Sustainable Transportation (2005), “a sustainable transportation system is one that: 4 5 6 The information and quote was taken from the Library of Congress’ webpage http://www.thomas.gov/cgibin/bdquery/z?d110:HR05161:@@@L&summ2=m&#status Perhaps the most remarkable UNECE initiative related to transport and the environment is the Transport, Health and Environment Pan-european Programme (THE PEP), under the joint leadership of UNECE and the World Health Organization (WHO) in the European Region. This programme promotes environment- and health-friendly modes of transport and infrastructure. As it includes explicitly health objectives, this initiative stands out among those related to transport and the environment. More information on this initiative can be found on the programme’s webpage http://www.unece.org/thepep/en/welcome.htm The Secretariat General of the EBU drew the following conclusion from the seminar: “Modal shift towards inland shipping … does not only contribute to an improvement of the environmental performance of the transport chain but to developing a sustainable transport system in general.” More information on the seminar can be found on EBU’s webpage http://www.ebu-uenf.org/fileupload/ebuseminar/Conclusions_EBU-Seminar%202009.pdf 2 PIANC MMX Congress Liverpool UK 2010 • • • allows the basic access needs of individuals and societies to be met safely and in a manner consistent with human and ecosystem health, and with equity within and between generations; is affordable, operates efficiently, offers choice of transport mode, and supports a vibrant economy; limits emissions and waste within the planet’s ability to absorb them, minimizes consumption of non-renewable resources, limits consumption of renewable resources to the sustainable yield level, reuses and recycles its components, and minimizes the use of land and the production of noise.” This definition seems to be internationally accepted, not least by the European Union Council of Ministers7. In order to apply the concept of sustainable transport, several sets of indicators have been developed. The most important set in Europe is the one used for the Transport and Environment Reporting Mechanism (TERM) with its focus on the environmental performance of transport. TERM came into being at the request of the EU transport ministers in 1998 and now represents the key activity of the European Environmental Agency (EEA)8in the field of transport. The EEA publishes annually TERM reports, which – based on the TERM indicators – show the progress achieved on the transport and environment integration strategies in the EEA member countries. The EEA published its first TERM report in 2000, giving detailed background information on the TERM indicators. EEA (2009) explains the some 40 indicators that are actually used. The European research project SUMMA – Sustainable Mobility, Policy Measures and Assessment is more concerned than is TERM with sustainable transportation. According to Rahman, A. & van Grol, R. (2005), the first of the three project objectives was to “define and operationalise sustainable mobility and transport, develop an appropriate system, and define a set of indicators for monitoring the environmental, economic and social dimensions of sustainable transport and mobility.” The more than 60 outcome indicators9 developed within the SUMMA project are shown in Annex A. These outcome indicators enable the assessment of the results of measures taken to influence the functioning of a transport system. These measures were initially assumed to be policy measures, as the name of the project indicates. However, these indicators can be easily used to assess the outcome of administrative, technical or operational measures, as the driving forces behind these measures do not matter for assessment process. Both, the concept and the indicators have been criticized. The Centre for Sustainable Transportation (2005) summarizes these critics: The concept was said to be meaningless, and that attempts to define it were at best fruitless and at worst misleading. The list of indicators, because of their complexity, could be seen as a wish list, since some of them probably could not currently be measured or otherwise determined. Nevertheless, the author of this paper believes that a definition of the concept of sustainable transport is needed, not least as the term is used frequently in the debate about the future of our transport systems. (That is equally true for the concept of green transport, as this term seems to be used as a synonym for sustainable transport.) The indicators are also needed, as shown by the response of the European Parliament on the Commission’s Communication “Greening Transport”. A comprehensive list of indicators allows evaluating policies or other measures aimed at increasing the benefits of transport systems or ensuring their future. They also support objective and comprehensive comparisons between different modes of transports, for example when deciding about infrastructure investments. Finally, in the context of this paper, the indicators facilitate an analysis of the greening of inland navigation. 7 8 9 The EU Ministers responsible for Transport and Communications adopted the resolution “Strategy For Integrating Environment And Sustainable Development Into The Transport Policy” — also known as the April Resolution — at the 2340th meeting of the European Union’s Council of Ministers, held in Luxembourg, April 45, 2001. This resolution contains a definition for “sustainable transportation system”. The minutes of that meeting are available at http://corporate.skynet.be/sustainablefreight/trans-counci-conclusion-05-04-01.htm . The EEA explains the TERM policy context, process and concept on its webpage http://www.eea.europa.eu/themes/transport/term . SUMMA applies indicators for: 1) the external force, 2) functioning and performance of the system, and 3) outcomes of interest. For this paper, only the outcome indicators are relevant, because they describe the performance of a transport system. 3 PIANC MMX Congress Liverpool UK 2010 2. NAVIGATION IN THE RHINE BASIN The Rhine, with a watershed of some 185.000 km2 (see Figure 1) and a length of 1320 km, of which some 850 km are navigable, is not one of the largest rivers in the world. However, with more than 300 million tons of cargo and 2 million containers (TEU) transported each year, resulting in 45 billion tkm, it is one the busiest inland waterways worldwide. This is due to the Rhine’s geographical situation, as the Rhine and its tributaries, complemented by several canals, connect major European seaports, notably Rotterdam and Antwerp, with important European industrial areas, such as the “Ruhrgebiet”, the heart of German heavy industry and the Basel Region in Switzerland, known for its large chemical industries. Figure 1: Watershed of the Rhine10 Another factor contributing to the success of Rhine navigation is the well developed waterway infrastructure. Annex B shows the waterway profile of the Rhine and the permitted dimensions of vessels and convoys. As can be expected, the waterway profiles of the tributaries and the canals are less favourable. The permitted vessel and convey dimensions are generally determined by the usable length and width of the lock chambers, with a length ranging between 105 m and 170 m and a width of 11,5 m. The available water depth allows, in most cases, for a minimum draught of 2,5 m. The Central Commission for the Navigation of the Rhine (CCNR), the oldest international organisation worldwide, is the governing body for navigation of the Rhine. The Congress of Vienna established the CCNR in 1815, and it now has as its main legal basis the Mannheim Convention, which was signed 140 years ago. Table 1 shows the CCNR key legal instruments with implications for sustainability and the year in which they were originally introduced. These instruments are binding for the member states 10 www.iksr.de 4 PIANC MMX Congress Liverpool UK 2010 and thus for the users of the Rhine. The instruments are constantly adapted to meet technical progress and changing societal needs. 1838 First regulation concerning the transport of certain dangerous goods 1850 First police regulations adopted for navigation on the Rhine 1904 First inspection regulations with safety requirements for vessels 1922 First regulation for the issue of boatmaster's licences 1968 First radar rules for inland navigation 1971 First prescription for the transport of dangerous goods on the Rhine (ADNR) 1995 Revised Police Regulation, Inspection Regulation and ADNR, adapted to the technical progress in inland navigation 1996 Signing of a Convention regulating waste disposal in inland navigation 2000 First regulations for exhaust emissions from inland navigation engines Table 1: Key legal instruments of the CCNR with implications for aspects of sustainability Navigation in the Rhine basin has been and continues to be the subject of extensive research, much of it focusing on economic and environmental issues. This author is convinced that because of both its economic importance and the availability of relevant research results, navigation in the Rhine basin constitutes a suitable case study of the greening of inland navigation. Passenger transport by inland navigation in the Rhine basin is significant and this sector is growing. However, the economic importance of this sector is still relatively small compared with freight transport and it serves almost exclusively tourism, but not basic transport needs of people. Therefore, the case study is restricted to freight transport. 3. GREENING OF INLAND NAVIGATION: WHAT HAS BEEN ACHIEVED? As stated in the introductory chapter, the greening of inland navigation can basically be understood as making inland navigation more sustainable, and indicators are available to monitor this process. To assess the extent of greening of inland navigation, the outcome indicators of the SUMMA project will be used. The greening is only partly due to public policy measures, as will be shown in the following chapters. Therefore, the use of the more versatile SUMMA indicators is called for instead of the politically more important, but strictly policy oriented TERM indicators. However, for three reasons, the SUMMA indicators should be used with caution, mainly: First, the indicators are numerous and complex; second, for some of the indicators, data or even relevant methods are not (yet) available; third, for specific analysis, only certain indicators are relevant. Thus, a selection is required to identify those indicators that can support the analysis undertaken in the context of this paper. Annex C contains an overview on the selected indicators and explains the reasoning and process for the selection of the indicators. In addition, each of the following chapters mentions at the beginning the selected indicators, which are – for easy identification – set in cursive. Due to the character and scope of this paper, the extent of greening of inland navigation analysis must be limited to summarizing results of relevant research projects or similar activities. Because of limited data availability, for some of the indicators the analysis is based largely on the expert judgment of the author. An in-depth discussion of different aspects of inland navigation's sustainability should take place in a different context. 3.1 Accessibility A green or sustainable transport system allows basic access needs to be met. Accessibility in the freight sector means making regions accessible as a prerequisite for economic development and providing transport connections to important national and international destinations. Indicators for accessibility are intermodal terminal facilities and the accessibility of origins/destinations. Traditionally, inland navigation is used to provide accessibility of (direct) origins, such as transport of coal from mines, or (final) destinations, such as transport of iron ore to steel works. Bonnerjee, S. et al (2009) show two examples of this traditional use of inland navigation enabling the industrialisation of the Rhine basin. This traditional use of inland navigation is often ‘intramodal’, as inland navigation 5 PIANC MMX Congress Liverpool UK 2010 covers the entire transport chain between origin and destination of the cargo or, as is mostly the case in the Rhine basin, connects seaports with the origins or destinations of cargo. However, inland navigation can also play an important role in true intermodal transport, which is indicated by the number of intermodal terminal facilities. A large number of these facilities exists along the Rhine. For containers alone, more than 20 intermodal terminals11 are located between Rheinfelden in Switzerland and Emmerich on the German/Dutch border. That means, the Rhine averages some three container terminals per 100 km. On other rivers in the Rhine basin, the number of terminals is much lower and the space between them much greater. For example, for the entire Mosel, the Dutch Inland Shipping Information Agency (BVB, 2009) lists only one terminal, equalling roughly one terminal per 350 km. This vast discrepancy can be explained with specific transport demands; but it is also a result of different characteristic of the waterways. Between Rotterdam and Strasbourg, the air draught of bridges over the Rhine is at least 9 m, giving enough room for vessels to store containers in four or even more layers. Between Strasbourg and Basel the air draught is at least 7 m, allowing for three layers of containers. On the other hand, air draught for bridges over the Mosel is in some cases less than 6 m, often restricting carriage of containers to only two layers, which is hardly economical. As increasing the height of bridges is generally very costly, expanding container transport in inland navigation seems to be fairly limited. Besides physically integrating inland navigation with other modes of transport through intermodal terminal facilities, a need also exists to integrate information. Transport forwarders want to have easy access to all relevant information, in general or for specific transport tasks. PIANC (2004) shows, how the application of River Information Services (RIS) can support this integration. Navigation on the Rhine scores high on the indicators for accessibility, namely intermodal terminal facilities and the accessibility of origins/destinations. As there are plans to build more of these terminals, it can be concluded that Rhine navigation is providing ever more basic access needs in freight transport. Beside the availability of the terminals themselves, a prerequisite for this high score is an adequate infrastructure for economical transport of containers. 3.2 Transport operation costs A green or sustainable transport system is affordable and supports a competitive economy. Affordability in the freight sector means that the freight forwarders or cargo owners are able to cover their costs for transport services. Indicators for affordability are supplier operating costs and transport prices.12 As compiled by PLANCO (2007), Figure 2 indicates transport costs for bulk cargo on typical transport routes in Western and Central Europe, half of them passing through the Rhine basin. The figure shows that for the selected routes 1. transport costs for inland navigation are generally lower than those for rail and road, 2. transport costs for navigation on the Rhine (Rotterdam-Duisburg) are lower than those for transport activities involving other waterways. The same conclusions can be drawn from PLANCO (2007) for container transport. Operating costs and transport prices for inland navigation in Western Europe are increasing, according to the data compiled by the Secretariat of the CCNR (2008), with the main driving forces personnel and fuel oil costs. Personnel costs are largely determined by the CCNR regulations for ship personnel, as they prescribe the minimum number and qualification of crew on board, and fuel costs by prices for bunker fuel. One possibility to contain transport prices even when the main input costs are increasing is by using larger vessels. PLANCO (2007) shows that operating costs for the very large vessels13 and for convoys of a large motor vessel pushing one barge are considerably lower than the operating costs for smaller vessels or even for push convoys. Accordingly, analysis provided by the Secretariat of the CCNR (2007) concludes that a large majority of new vessels joining the Rhine fleet are very large and the average cargo carrying capacity of the Rhine fleet is increasing. PLANCO (2007) expects the number of very large vessels to increase more than twofold between 2006 and 2025, growing faster than any other vessel category. 11 As it seems impossible to develop reliable data for all intermodal facilities along the Rhine, the density of container terminals may serve as a proxy for the access to intermodal terminal facilities in general. 12 Due to the nature of this paper, a discussion of the definition of costs, e.g. transport costs vs. supplier operating costs is not possible. PLANCO (2007) gives different definitions and supports these definitions with data. However, for the purpose of this paper, the author's more general approach should be acceptable. 13 Very large vessels have a length between 110 m and 135 m and a width between 11,50 and some 17 m. Their draught is generally 3,5 m or more. 6 PIANC MMX Congress Liverpool UK 2010 Financial transport costs 60 50 Road Rail Inland Water Transport € per ton 40 30 20 10 Average Antwerpen - Rhine Ludwigshafen Hamburg Hannover Linz - Nürnberg Rotterdam - Rhine Dillingen Rotterdam - Rhine Großkrotzenburg Rotterdam - Rhine Duisburg Hamburg - Salzgitter Hamburg - Decin 0 Figure 2: Financial transport costs of different transport modes for bulk cargo on selected routes (PLANCO, 2007) The low transport operation costs of inland navigation ensure that the Rhine basin continues to enjoy the benefits of large industrial plants employing to tens of thousands of people. Bonnerjee, S. et al. (2009) demonstrate this with case studies for the chemical and steel industries. The effects of climate change have a direct impact on the transport operation costs. A constant, high volume water flow guarantees uninterrupted navigation with high load factors, keeping costs low. Adaptation to uneven and interrupted water flows is possible (PIANC, 2008). For the Rhine, expected negative impacts will be manageable and navigation will be able to compete successfully in the transport market.14 Jonkeren O.E. (2009), applying economic theories to detailed data of Rhine navigation, finds that “… under the given climate scenarios the effect of climate change on modal split in the Rhine area is limited.” For other rivers, the situation may be different. The Ubangi in central Africa gives a good example. Disruption of navigation due to low water levels has increased from 4 days each year in the middle of last centuries to more 200 days each since 2002 (Bonnerjee, S. et al., 2009). Navigation under those conditions is clearly not sustainable. Inland navigation transport is affordable and supports a competitive economy. Where adequate waterway infrastructure and sufficient water flow is available, inland navigation can satisfy transport demand with operating costs and transport prices lower than those of road or rail transport. Increasing vessel dimensions and using convoys of motor vessels with push barges help to compensate for increases in personnel and fuel costs. 3.3 Productivity and efficiency A green or sustainable transport system operates efficiently. Efficiency in the freight sector means that the transport equipment and infrastructure are fully used and that the energy consumption, which constitutes perhaps the main resource input for transport – personnel resources not included – is low. Indicators for productivity and efficiency are among others utilisation rates and energy efficiency. PLANCO (2007), examining typical transport tasks in Western and Central Europe, concludes from available transport statistics that for five of these tasks the inland navigation vessels return empty, for four of these tasks the vessels are loaded between 80% and 96% of their capacity on their return trips 14 CCNR Congress, “Rhine Navigation and Climate Change”, 24th / 25th June 2009, Bonn; http://www.zkrkongress2009.org/ 7 PIANC MMX Congress Liverpool UK 2010 and for four tasks the vessels return fully loaded. High return loads occur generally in the container trade; no return loads are common in transport of bulk goods. In this aspect, inland navigation is the same as road or rail transport. However, inland navigation is different in that the utilisation rate of an inland navigation vessel relies directly on the available water depth, which in a free flowing river such as the Rhine, is a function of the water flow and finally determined by precipitation. On critical stretches of the Rhine, the water level drops below 1,90 m on the average 20 days per year. This reduces the possible utilisation rate particularly for larger vessels. Jonkeren, O.E. (2009) provides in-depth analysis of the interdependency between the water level of the Rhine and the average load factor. At the gauging station of Kaub, he found for water levels of 2,60 m and more an average load factor of or above 80%. The load factor drops to just over 50% for a water level of 1,90 m. Thus, the utilisation rate of inland navigation is first of all determined by transport demand, an economic factor applying to all transport modes, but also by available water depth, a factor most often determined by nature, which applies only to inland navigation. RIS can help to improve the utilisation rates of both vessels and waterways. Vessels can make better use of available water depth by using Inland ECDIS15. Waterways administration can permit larger vessels to sail on waterways with restricted fairway dimensions when requiring the use of Inland AIS16 to avoid collisions. PLANCO (2007), examining the utilisation rate of transport infrastructure in Germany, concludes that inland navigation is the only transport mode that still offers spare capacity on all transport routes and especially in the Rhine basin. This has two important implications: 1. Inland waterways in the Rhine basin can accommodate future transport volume growth without any additional investment or land use. 2. Inland navigation is the only transport mode in the Rhine basin unhampered by congestion, caused by insufficient infrastructure capacity. The occurrence of spare capacity in inland navigation while at the same time the other modes of land transport suffer from congestion, is not restricted to the Rhine basin, as Bonnerjee, S. et al. (2009) highlight. Inland navigation in general and Rhine navigation in particular, are highly energy efficient. Table 2 presents data for typical transport tasks in Western and Central Europe showing that the energy per tkm or TEUkm is much lower for inland navigation than for the other modes of land transport. Data provided by the Dutch Inland Shipping Information Agency (BVB, 2009) shows higher values for energy consumption for transport modes, but reinforces the general pattern outlined by PLANCO. Kruse, C.J. et al. (2007) undertook the same comparison for land transport in the USA. Whereas the general pattern is again the same, the specific energy consumption of inland navigation in the USA is only some 30% lower than that of rail. Transport Mode Average energy consumption for selected transport tasks Bulk cargo [MJ per 100tkm] Container [MJ per 100 TEUkm] Road 92 594 Rail 43 431 Inland Navigation 22 237 Table 2: Average energy efficiency of different transport modes for selected transport tasks (PLANCO, 2007) Energy efficiency of inland navigation in the Rhine basin is not only very high, compared to other modes of transport, it is also steadily improving. In general, the more recent the vessel and convoy type, the larger is its carrying capacity and the lower its energy consumption per tkm. Chapter 3.8 on emissions to air and greenhouse gas emissions contains more detailed information. Inland navigation provides a very important tool to make the overall freight transport system more sustainable by increasing its productivity and efficiency. Utilisation rates of the vessels, as well as the waterway infrastructure, are often low. However, that means inland navigation very often has spare capacity to accommodate future transport growth, whereas other modes of land transport have to 15 16 Electronic Chart Display and Information System Automatic Identification System 8 PIANC MMX Congress Liverpool UK 2010 increase their capacity requiring investments and further land use for additional transport infrastructure. The comparable low utilisation of inland navigation also means no or very little congestion, avoiding additional transport costs and pollution. When it comes to energy efficiency, inland navigation practically outperforms road and rail by the factor 2 to 4, while at the same time providing significant opportunities for further improvement. 3.4 Costs to Economy A green or sustainable transport system causes no or few costs to the economy. Costs to the economy are often signs of distorted markets, which may increase transport demand through artificially low prices, lead to inequity within and between generations, or have other undesirable effects. Indicators for costs to the economy are public subsidies, infrastructure costs, or external transport costs. Public subsidies are often needed to build and to maintain inland waterways. As stipulated by the Mannheim Convention, navigation on the Rhine is free of charge. Also, no fuel tax is levied on gasoil for inland navigation. Therefore, all costs for construction and maintenance of its waterway infrastructure have to be covered by public subsidies. Since the Rhine is for the most part a free flowing river, these costs are low. The Rhine’s tributaries require dams and locks, which are costly to build, operate and maintain. It is not surprising, that therefore, contrary to what was intended when developing this waterway, the fees levied on vessels sailing on the Mosel do not cover these costs. PLANCO (2007) provides a comparison, reproduced in Table 3, of infrastructure costs and cost recovery in Germany for different transport modes. According to this comparison, inland navigation has the lowest infrastructure costs, but also the lowest level of cost recovery of land transport modes. infrastructure costs Infrastructure cost recovery Billion € € / 1.000 tkm Billion. € € / 1.000 tkm Level of cost recovery Road 6,467 45,21 - 2,826 - 19,76 56,3% Rail 2,876 48,42 - 2,483 - 41,80 13,7% Inland navigation 0,626 12,60 - 0,573 - 11,53 8,5% Transport mode Table 3: Comparison of infrastructure costs for freight transport in Germany for the year 1987 (PLANCO, 2007) External costs for different transport modes vary greatly. PIANC (2005b) analyses data for different modes, but does not attempt to compare them. PLANCO (2007) provides a comparison and gives more inside into different categories of external costs. Figure 3 shows that on selected transport routes, the external costs of inland waterways caused by accidents, noise and climate gases are much lower than those for rail and road transport. Only rail’s external costs for air pollution are lower than those of inland navigation. (More information on these important topics is provided in the chapters on emissions to air, noise, and safety). The external costs given by PLANCO are basically in line with those given by PIANC (2005b), while the latter also provides data for barrier effects and surface occupation. For those categories, inland navigation causes no external costs in contrary to road and rail transport. Comparisons of external costs for soil and water pollution or effects on biodiversity are not available yet, at least not for the Rhine basin. This is most likely due to the complex economic evaluation required. Comparing transport infrastructure costs, such as costs for construction, maintenance and operation, is very difficult. However, infrastructure costs recovery for inland navigation is clearly much lower than for other modes of transport. Therefore, infrastructure costs for inland navigation generally rely on public subsidies, whereas road and rail transport do claim to pay for some or all infrastructure costs via taxes and user charges. External costs of inland navigation, however, are very low, when compared to road and rail. Therefore, if all transport modes had to pay for their infrastructure and external costs, inland navigation may come out as the winner. The data provided in Figure 3 suggests just this. The European Commission (2008) strongly advocates a review of the transport costs to society and to internalize all external transport costs, convinced that this would make the EU transport system more sustainable. 9 PIANC MMX Congress Liverpool UK 2010 External costs on selected routes € per 100 tkm 0,9 0,8 Accidents Noise Air pollution Climate gases 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0 Min Max Average Road freight Min Max Railways Average Min Max Average Inland water transport Figure 3: External costs on selected transport routes (PLANCO, 2007) 3.5 Benefits to economy A transport system that does not create benefits is not needed. The benefits of different modes of transport are similar, for example public revenues from taxes. Inland navigation differs in that its infrastructure, the waterways, are attractive for numerous leisure or recreational activities, which lead to public revenues in addition to those from transport. A report by the Secretariat of the CCNR (2007) indicates that more than 2.000 daytrip vessels with a total capacity of almost 500.000 passengers, as well as some 370 cruise vessels with almost 40.000 beds are sailing on the waterways of the CCNR member states. River cruises may be the fastest growing segment in inland navigation in Europe. European waterways also provide opportunities for rowing, canoeing, yachting, swimming, fishing and several other leisure activities. All these activities create public revenues from taxes and employment. Bonnerjee, S. et al. (2009) describe similar observations for the Tennessee River in the USA. 3.6 Resource use A green or sustainable transport system minimizes consumption of non-renewable resources, limits consumption of renewable resources to the sustainable yield level and minimizes the use of land. Indicators for resource use of transport systems are energy consumption, consumption of solid raw materials and land take. Consumption of fossil fuels is probably the most important resource use for inland navigation. The current debate on ‘peak oil’17 starkly highlights that fossil fuel use is not sustainable. Dramatic rises in fuel prices as a precursor for fuel shortages can be expected. For a long time, liquid biofuels were seen as a substitute for fossil fuels. However, negative effects of biofuel production on the environment, especially on bio diversity, as well on food supply casts some doubt on biofuels as a viable alternative to fossil fuel. IEA (2009) expects biofuels to meet just 5% of road fuel demands, in 2030, compared to some 2% today. Currently, research is intensifying to use liquefied natural gas (LNG)18. IEA (2009) estimates the natural gas resources as sufficient to meet projected demand to 2030. However, these substitutes may offer only temporary relief as natural gas supplies will run out at one stage. 17 The World Energy Outlook 2009, prepared by the International Energy Agency (IEA), implies that the peak in oil production may occur around the year 2030. 18 MAGALOG, Maritime Gas Fuel Logistics, www.eu-magalog.eu 10 PIANC MMX Congress Liverpool UK 2010 Hydro, solar or wind power can be easily used as energy sources for rail transport, covering up to 100% of rail transport energy needs. This is not so for inland navigation, because the powerful propulsion engines of inland navigation vessels require their energy supply to be stored on board. Only for small (passenger) vessels and small auxiliary engines, solar power seems to be a real alternative (Grave, H. de, M. van Wirdum (redactie), 2006). Renewable energy fed into shore power supply for inland navigation vessels at berth reduces fossil fuel consumption as well as emissions of pollutants and noise, albeit to a very limited extend.19 Biogas may also substitute mineral oil based fuel in inland navigation. However, biogas production in general is very limited and so far, research on the use of biogas in inland navigation seems to be at a very early stage. Hydrogen, in contrast, is already powering a small passenger vessel in Hamburg as part of a project called Zemships – Zero Emission Ships20. A similar vessel, also using a fuel cell to produce electric energy from hydrogen, was baptised end of 2009 in Amsterdam.21 Hydrogen for this vessel will be generated with energy from a windfarm, making it a fuel that neither uses limited resources nor causes any pollution or GHG emissions. Consumption of solid raw materials is less import for inland navigation, as inland navigation infrastructure and equipment have long life cycles. Also inland navigation vessels are almost 100% recyclable as the vessels and all equipment on board are, with few exceptions, made from steel or other recyclable material. Land take for inland navigation infrastructure could be of concern, especially in port development. However, most often, inland navigation takes place on rivers and lakes, making land take unnecessary. This is in contrast to road and rail transport. Accordingly, different studies cited in PIANC (2005b) do not see the need to attribute external costs to land take of inland navigation. This chapter makes a very important conclusion for inland navigation. As long as there is no alternative for its energy consumption, which mineral oil currently fully meets, inland navigation is not sustainable. Using LNG may provide only a temporary relief. Consumption of solid raw materials and land take for inland navigation and its infrastructure, on the other hand, are on such a low level, as to be sustainable. 3.7 Direct ecological intrusion A green or sustainable transportation system is one that meets the transport demands consistent with ecosystem health. Indicators for direct ecological intrusion are fragmentation of land, damage of underwater habitats, losses of nature areas, proximity of transport infrastructure to designated nature areas, collisions with wildlife and introduction of non-native species. Inland Navigation usually does not contribute to fragmentation of land. Accordingly, studies referred to in PIANC (2005b) do not see the need to attribute external costs to land take of inland navigation. Inland navigation, however, may damage underwater habitats and lead to losses of nature areas. The Rhine provides a telling example. Over the last centuries, it has undergone dramatic changes from a free flowing river that changes frequently its bed to a river that stays in one place, supporting many human activities on the river and its shore, such as inland navigation, hydro power generation, agriculture. At the time, when these changes occurred, the communities along the river were mostly concerned with generating welfare from water use. Today, they wish to use the water in a way that is sustainable and protects the aquatic ecosystems. But, which and how much water use is sustainable? The most important tool of the EU for the protection and sustainable management of water is the so called Water Framework Directive (WFD)22. Its article 1 states “The purpose of this Directive is to establish a framework for the protection of inland surface waters … which: (a) prevents further deterioration and protects and enhances the status of aquatic ecosystems …; (b) promotes sustainable water use based on a long-term protection of available water resources; (c) aims at enhanced protection and improvement of the aquatic environment, inter alia, through specific measures for the progressive reduction of discharges, emissions and losses of priority substances and the cessation or phasing-out of discharges, emissions and losses of … hazardous substances …” 19 Robert Tieman, Practical experience with the adaptation of the inland navigation fleet to changes in the water discharge and with the reduction of fuel consumption, Presentation, CCNR Congress, “Rhine Navigation and Climate Change”, 24th / 25th June 2009, Bonn 20 www.zemships.eu 21 www.fuelcellboat.nl 22 Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy 11 PIANC MMX Congress Liverpool UK 2010 Because of these far reaching goals of the WFD, it seems reasonable to assume that the use of a water body is sustainable, including ecological aspects, as long as the goals of WFD are met. The WFD requires the EU member states to develop a River Basin Management Plan (RBMP), containing inter alia information on significant pressures and impacts of human activity on the status of surface water and on the measures adopted to reach the goals of the WFD. Table 4 contains a short summary of important pressures and impacts of inland navigation on the waterbodies in the Rhine basin and the foreseen measures, to address these pressures and impacts. This summary is based on the RBMP for the Rhine23, as published by the International Commission for the Protection of the Rhine (ICPR). The RBMP also contains a prognosis for improvements of the main river waterbodies. For example, in 2015 20% of the surface waterbodies will have a good ecological status or a good ecological potential compared to only 4% in 2009. Pressures or impacts of inland navigation Measures to avoid or to mitigate pressures or impacts Hydromorphological changes, including bank stabilization, to improve navigation conditions alter or reduce habitats in the water and on shore, Reactivation of overbank areas; reconnection of oxbow lakes and backwaters; increase of structural diversity on the banks; reestablishing of river continuity (installations of fish passes); redesign of river training works; improvement of sediment management; widening of riverbed; application of green technologies for waterway infrastructure Generation of waves and turbulences by vessels reduce quality of habitats Protection of shore / shallow areas against waves and vessel induced currents Current status and outlook Programs have been developed; first measures are implemented and show feasibility of program goals Table 4: Short summary of pressures or impacts of inland navigation on aquatic ecosystems and measures to avoid or mitigate them PIANC has published numerous technical reports with practical recommendations to minimize negative impacts of waterway infrastructure construction, maintenance and operation on water bodies and specifically on the aquatic environment. PIANC's most important contribution may be the development and promotion of ‘Working with Nature’24. This concept is based on the idea, that “… working with, rather than against, natural processes can result in … more sustainable solutions.” Inland ports are generally situated within city limits or close to large industrial plants and therefore far away from designated nature areas. Inland waterways can coexist with designated nature areas, as the Rhine basin demonstrates. The RBMP for the Rhine itself lists the areas established under the EU’s biodiversity protection legislation25. These areas are either situated on the shores of the Rhine or even include the waterbody itself, in some cases from shore to shore. PIANC (2005a) provides a guideline on bird habitat management in ports and waterways. These examples demonstrate that highly developed inland navigation inside a designated nature area does not undermine the goals of nature protection per se and that direct ecological intrusion, caused by the proximity of inland navigation infrastructure to designated nature areas, can be managed in a sustainable manner. PIANC (2008a) gives an insight into the most important impacts of vessels on inland waterways and their ecological effects. The report also deals with collisions with wildlife in inland navigation, such as fish larvae mortality from propeller impact. The report reveals that these impacts usually do not cause serious environmental problems and that highly efficient mitigation measures can be applied to minimize them. 23 International koordinierter Bewirtschaftungsplan für die internationale Flussgebietseinheit Rhein (Teil A = übergeordneter Teil), Internationale Kommission zum Schutz des Rheins, Koblenz, 2009 www.iksr.org/fileadmin/user_upload/Dokumente_de/Berichte/BWP-10d.pdf 24 http://www.pianc.org/workingwithnature.asp 25 Council Directive 79/409/EEC of 2 April 1979 on the conservation of wild birds, Council Directive 92/43/EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora 12 PIANC MMX Congress Liverpool UK 2010 Inland navigation has led to the introduction of invasive species from the black sea via the MainDanube Canal into the Rhine basin (Bundesanstalt für Gewässerkunde, 2009). Other non-native species invaded the Rhine from the seaports. According to Werner, S. and P. Rey (2009), the improvement and extension of natural habitats may help to control the expansion of invasive species. Inland navigation does not pose any significant threat to ecosystem health with regards to fragmentation of land and to proximity of transport infrastructure to designated nature areas. However, inland navigation may cause damage of underwater habitats, losses of nature areas, collisions with wildlife and introduction of non-native species. The implementation of the WFD shows, that these impacts can be overcome with dedicated programmes of measures and that in the long term inland navigation can become sustainable with regards to the use of surface waterbodies. 3.8 Emissions to air A green or sustainable transportation system is one that meets the transport demands consistent with human and ecosystem health, and limits emissions and waste within the planet’s ability to absorb them. Indicators for emissions to air are the transport emissions of greenhouse gases and of air pollutants as well as emissions of greenhouse gases and of air pollutants from manufacturing and maintenance of the transport equipment and infrastructure. The most important transport emissions to air are carbon dioxide (CO2), carbon monoxide (CO), nitrogen oxides (NOX), volatile organic compounds (VOCs) and particulate matter (PM). Table 5 summarises their impact and formation. Emission Formation Impact Carbon dioxide (CO2) Fossil fuels combustion Greenhouse gas Carbon monoxide (CO) Incomplete fuel combustion Health hazards Volatile organic compounds (VOCs); (Hydrocarbons (HC)) Incomplete fuel combustion; emission of chemicals Health hazards; smog Nitrogen oxides (NOx) Combustion in engines; oxidation of nitrogen in fuel or air Health hazards; acidification; eutrophication; N2O very potent greenhouse gas; ozone; smog Particulate matter (PM) Diesel engines Health hazards; smog Table 5: Formation and impact of transport emissions to air Transport greenhouse gas (GHG) emissions, such as CO2, continue to rise. Globally, the rise of transport GHG emissions is accelerating, usually in line with economic growth (ITF/OECD 2009). If no decisive measures are taken, transport GHG emissions will undermine overall GHG emission targets, as shown for the EU in Figure 4. The most important GHG in the transport sector is CO2, as 95% of the energy consumed by transport comes from fossil fuels (ITF/OECD 2009). CO2 emissions from transport have been the subject of many studies, often comparing CO2 emission intensity of different transport modes. The CO2 emission intensity or emission efficiency of transport describes the total emission from the activity in relation to the total transport work, expressed in tons of CO2 per ton*kilometre. Figure 5, which is compiled from data provided by Buhaug, Ø. et al. (2008), Christ, P. (2009), PLANCO (2007), Kruse C. et al. (2009), shows that the CO2 emission intensities of freight transport modes differ substantially, from inland navigation which has the lowest CO2 emission intensity. Large maritime vessels are the exception. EEA (2008) sees average CO2 emission intensity for rail (18 - 35 g/tkm) as being lower than the one of inland navigation (30 - 49 g/tkm). According to STREAM – Studie naar TRansport Emissies van Alle Modalieteiten, commissioned by the Dutch government (CE 2008), the CO2 emission intensity of inland navigation ranges from 20 up to 95 g/tkm. These differences testify that transport emissions are difficult to calculate and it is even more difficult to compare different transport modes.26 However, in all the studies, rail and inland navigation show far lower CO2 emission intensities than the road sector. 26 During the ITF Workshop on Reducing CO2 Emissions in Goods Transport, Erik van Agtmaal, Managing Partner of Green Logistics Consultants Group pointed out that there are large discrepancies between outcomes of different CO2 calculation tools, inconsistencies in available data and differences in results of more in-depth research. One of his recommendations to the ministers of transport was to create an international platform to consolidate, develop, maintain and disseminate common procedures and data. www.internationaltransportforum.org/Topics/Workshops/WS3vanAgtmaalSlides.pdf 13 PIANC MMX Congress Liverpool UK 2010 Growth of Transport GHG 120 Total GHG emissions (EU-27) Index (1990=100 100 80 60 -60 % 40 Annual growth rate: +1.4 % / year (avg. 2000-2005) -80 % Transport emissions 20 0 1990 2000 2010 2020 2030 2040 2050 Source: European Environment Agency Figure 4: European Union overall emissions trajectories compared with extrapolated transport emissions CO2 Intensity of Selected Freight Transport Modes: Log Scale (g/tkm) 50 RoRo 13 Container vessel (2000 - 8000 + TEU) 3 Container vessel (coastal < 2000 TEU) 2,5 Bulk carrier (10 k - 200 k Dw t) Products tanker (5 k - 60 k Dw t) 36 8 5, Crude oil carrier (10 k - 200 k Dw t) 60 20 24 2, 9 Inland barge (USA, average) 11 Inland barge (Western/Central Europe: 1500 - 18500 t) 10 Inland barge (Western/Central Europe: 70 - 400 TEU) 10 23 21 4 19 Rail Electric 69 15 Rail Diesel Road (heavy duty) 35 91 Air Long Haul (> 1600 km) 57 1 10 100 633 1000 Figure 5: CO2 intensity of selected freight transport modes (g/tkm) The CO2 emission intensity of inland navigation depends very much on the carrying capacities and the load factors of the vessels performing the transport task. This applies in general for all transport modes. Besides increasing the vessel size, there are numerous other measures, to further reduce CO2 intensity of inland navigation. PIANC (2008b) provides a comprehensive list of technical, operational and transport management measures for mitigation of GHG emissions from navigation. STREAM (CE (2008)), without giving any details, estimates the reduction potential of different techniques between 10 and 25%. The Bureau Innovatie Binnenvaart (Grave, H. de, M. van Wirdum (redactie), 14 PIANC MMX Congress Liverpool UK 2010 2006), Zöllner27 and Hazeltine28 provide estimates for individual measures to reduce CO2 in inland navigation, ranging from diesel-electric propulsion systems to optimal hull shapes for pushed conveys. The most effective measures probably relate to the operation of the vessel. A competition in the Netherlands in support of CO2 reduction in inland navigation resulted in an average fuel consumption reduction of more than 10%, and in an individual case of more than 50%29. As all of these measures basically increase energy efficiency and reduce fuel consumption, they also contribute to emission reductions of air pollutants, such as NOX and PM. This is very much needed, as inland navigation is not as clean with respect to air pollutants as other modes of land transport. A report by the World Health Organization (WHO 2005) describes the sometimes dramatic health effects air pollutants can have on people. In the Rhine basin, the population may suffer from such effects from inland navigation air pollution, as waterways pass through large cities, and the inland ports are often situated in the proximity of densely populated areas. Especially PM is of concern as it is seen as a reason for premature deaths30. PLANCO (2007) views inland navigation emissions to air not as critical, as its external costs calculation in Figure 3 shows. Also, because of the relative small size of the inland navigation sector, its environmental impact in general remains small. However, when air pollutants are put in relationship to the transport performance, inland navigation may not qualify as a green transport mode. Figures 6 and 7 show comparisons of NOX and PM emissions for different modes of land transport. Figure 6: NOX emissions for different modes of land transport (g/tkm) The relatively low NOX and PM emissions from road transport are the result of stringent regulation. Successive stages of lower emission limits have stimulated research and development for ever cleaner engines with more and more sophisticated exhaust treatment installations, along with sulphur free fuel. Relevant regulation for inland navigation in the EU and other parts of the world has fallen several steps behind. Thus, whereas transport emissions of greenhouse gases from inland navigation are comparatively very low, giving support to the claim that inland navigation is a green mode of transport, transport emissions of air pollutants from inland navigation are relatively high and, at least for PM, not sustainable. 27 Zöllner, J., Strömungstechnische Möglichkeiten zur Reduzierung des Kraftstoffverbrauchs und der CO2-Emissionen von Binnenschiffen, Presentation, CCNR Congress , “Rhine Navigation and Climate Change”, 24th / 25th June 2009, Bonn 28 Hazeldine, Pridmore, van Essen and Hulskotte (2009), Technical Options to reduce GHG for non-Road Transport Modes. Paper produced as part of contract ENV.C.3/SER/2008/0053 between European Commission, Directorate-General Environment and AEA Technology plc; see website www.eutransportghg2050.eu 29 CO2MPETITIE BESPAREN SAMEN 12,4% BRANDSTOF http://www.voortvarendbesparen.nl/content/deelnemers-co2mpetitie-besparen-samen-124-brandstof 30 www.euphix.org/object_document/o5637n27091.html 15 PIANC MMX Congress Liverpool UK 2010 Figure 7: PM emissions for different modes of land transport (g/tkm) To determine emissions of greenhouse gases and of air pollutants from manufacturing and maintenance of the transport equipment and infrastructure requires a life cycle assessment, which is a complex task and not one that has been undertaken yet for inland navigation. Inland navigation vessels have a very long utilization phase. In Germany, for example, the average age of the dry cargo motor vessels is some 55 years. External costs for construction and maintenance of waterway infrastructures are low compared to rail, and they are used over a very long period of time – in Europe often more than a hundred years. These factors indicate that for inland navigation emissions of greenhouse gases and of air pollutants from manufacturing and maintenance of the transport equipment and infrastructure are much lower than from their operation and therefore do not seem to significantly impact the sustainability of inland navigation. 3.9 Emissions to soil and water A green or sustainable transportation system is one that meets the transport demands consistent with human and ecosystem health, and limits emissions and waste within the planet’s ability to absorb them. Indicators for emissions from transport systems to soil and water are polluting transport accidents, runoff pollution from transport infrastructure, discharges of oil at sea, discharges of wastewater and waste at sea. Polluting transport accidents in the Rhine basin are seldom. A Dutch report on inland navigation accidents (Rijkswaterstaat, 2007) shows for the years 2003 till 2006 roughly one accident per year with serious environmental impacts. In the years 1997 to 2006 the accidents involving vessels with dangerous goods decreased from almost 200 to just 36. In 2003, a CCNR round table of government and industry representatives identified the interface between ship and shore as the main source of possible incidents of transport of dangerous goods. As a consequence the CCNR initiated together with the Oil Companies International Marine Forum (OCIMF) the ISGINTT project. ISGINTT stands for International Safety Guide for Inland Navigation Tank-barges and Terminals31. Within this project, experts of all industry associations concerned with the transport of dangerous goods on European inland waterways are developing the practical safety guide, using the International Safety Guide for Oil Tankers and Terminals (ISGOTT) as the basis. The CCNR will publish the ISGINTT in 2010. This safety guide will supplement the ADN32, Europe’s already very comprehensive regulation for the carriage of dangerous goods by inland waterways. Table 6 contains a summary of important emissions of inland navigation into the waterbodies and the foreseen measures addressing these effects, based on the RBMP for the Rhine. The table shows, that today all significant sources for pollution of water and soil by inland navigation are addressed. Outlawing polluting substances or processes either stopped the pollution or reduced it to a very low level. PIANC (2009a), striving for the reduction of environmental impacts from dredging projects, published a methodology for the selection of the ‘best management practices’ for specific projects. 31 32 www.insgintt.org www.unece.org/trans/danger/publi/adn/adn_e.html 16 PIANC MMX Congress Liverpool UK 2010 Effects of inland navigation Possible measures to avoid or mitigate effects Current status and outlook Contamination from sediments are set free during dredging for navigation purposes Improved sediment management Put in practice; treatment or disposal of contaminated sediments on shore (PAHs from burning of fuel etc. reach surface water as fallout from the atmosphere) Stringent regulation for emission of pollutants, such as heating, road transport See chapter on emissions to air; however, inland navigation’s share of these PAHs seem negligible PAHs from coatings of land structures or waterway infrastructure Replacement of PACs containing coatings Sale of PACs containing coatings in most countries of Rhine basin forbidden PAHs from bilge water and other waste Banning of bilge water discharge; collection of all bilge waters and waste; treatment on shore CDNI, legal base for collection and treatment of bilge water and other waste, practically fully implemented (see chapter on waste) TBT from underwater coatings of maritime vessels Banning of coatings containing TBT International regulation implemented Copper from underwater coatings of pleasure craft Replacement of copper in coatings Copper based coatings almost not sold anymore Zinc from underwater coatings of vessels or waterway infrastructure Zinc still in use, search for replacements Table 6: Pollution of water and soil caused by inland navigation Pollution from waterway infrastructure can be significant. Bearings in mechanical equipment of locks as well as in pumping stations feeding navigation canals need oil and grease for lubrication. These lubricants, grease in particular, often end up in the waterway. Current technology offers alternatives, such as bearings using water as lubricant or the replacement of mineral oil based lubricants with non polluting ones. PIANC (2009b) has addressed these other environmental issues related to the construction and maintenance of locks. Regulatory concepts, innovative management practices, materials and technologies are available to reduce polluting transport accidents, runoff pollution from transport infrastructure, discharges of oil, wastewater and waste from inland navigation to a level not posing a risk to human and ecosystem health. These measures are now being implemented in the Rhine basin, fueling the expectation, that the remaining emissions are congruent with the sustainability of inland navigation. 3.10 Noise A green or sustainable transportation system is one that meets the transport demands consistent with human health and minimizes the production of noise. An indicator for sustainability of noise emissions is the exposure to transport noise. PLANCO (2007) states that in Germany exposure to transport noise from inland navigation is seen as insignificant. Accordingly, PLANCO calculates the external costs from inland navigation noise emissions at 0,00 € cent per tkm, whereas those from road transport are calculated at 0,79 € cent per tkm and for rail transport at 0,84 € cent per tkm. The following factors limit exposure to noise emissions from inland navigation: • low to medium speed of inland navigation engines, • proper engine casing by the vessels' engine rooms, • absence of other transport noises, for example from brakes or tires, • long distance between noise emitters (vessel engines) and people on shore, • stringent noise emission regulations. 3.11 Waste Generation of non-recycled waste is not a significant issue for inland navigation. The only typical inland navigation waste is bilge water, the mixture of oil, fuel and water on the bottom of a vessel's engine room. In the Rhine basin, dedicated organisations systematically collect the bilge water; its components are separated and then either recycled (oil, fuel) or treated (water). The same applies for other wastes, such as cargo residues. The CDNI, la Convention relative à la collecte, au dépôt et à la 17 PIANC MMX Congress Liverpool UK 2010 réception des déchets survenant en navigation rhénane et intérieure33, which entered into force at the end of 2009 and is now being implemented by six countries in the Rhine basin, constitutes the strong legal base for the collection and treatment of bilge water and other waste. It also provides financial incentives to reduce the amount of bilge water produced in individual vessels. 3.12 Safety and security A green or sustainable transportation system is one that meets the transport demands safely. Indicators for the safety of a transport system are the accident related fatalities and serious injuries. PLANCO (2007), after an assessment of relevant studies for freight transport accidents in Germany, and concludes: “These analyses and calculations confirm the position of inland shipping as the safest transport mode compared to road and railways. … Of all economic costs caused in Germany by freight transport accidents in the period 2000 to 2005, 96,9% are due to trucks, 2,0% to railways and 1,1% to inland shipping. The specific economic accident cost per 100 domestic ton-km is €-Cents 42,9 for road freight (2005), 6,0 for railway freight (average 2000-2005) and 3,3 for inland shipping (average 2000-2005). Unit accident costs of road freight transport exceed those of inland shipping by a factor of 13; those of railway freight are 80% higher than for inland shipping.” Kruse C. et al. (2009) come to a similar overall conclusion for inland navigation in the USA. Slow speed and low traffic density are probably the main reasons for the low accident rates and costs of inland navigation. In the Rhine basin, constant improvements of the waterway infrastructure and the installation of vessel traffic management, too, contribute to the high safety standard. Stringent regulation also plays a major role. For example, the CCNR continuously updates its requirements for the vessels themselves, such as minimum maneuverability, or for their equipment, such as radar. Currently, the CCNR and its member states push the application of RIS to achieve even higher safety levels. Germany and the Netherlands are subsidizing the installation of Inland AIS transponders on all motorized vessels and the CCNR is updating its requirements for Inland AIS with a view of it becoming possibly mandatory around 2012. Inland navigation in the Rhine basin demonstrates that government and industry have many tools available to improve safety of inland navigation, even though, because of its 'inherent safety', it is already by far the safest land transport mode for freight. 3.13 Social cohesion Support of society is a critical issue for inland navigation. This is one of the lessons learned from the case studies, contained in the ‘Report on inland waterways and environmental protection’ (ECMT, 2006) commissioned by the European Council of Ministers of Transport (ECMT). The report lists a number of inland navigation projects, which failed, at least initially, because the projects were not coherent with the opinion of relevant social groups. However, the report also describes good practice examples of how to achieve the necessary social support. Ensuring cohesion between social values or expectations and navigation development projects is also at the core of PIANC’s concept of ‘Working with Nature’. These values and expectations most often relate to the protection of the environment and particularly of biodiversity. Thus, reducing direct ecological intrusion to a sustainable level, as described earlier, should also lead to long term public support for inland navigation. 3.14 Working conditions in transport sector A green or sustainable transportation system is one that meets the transport demands safely and in a manner consistent with human health as well as with equity within and between generations. Indicators for the working conditions in the transport sector are occupational accidents, precarious employment conditions and work absence due to accidents and illness. Due to the scope of this paper, the assessment of the working conditions in inland navigation will be brief. Inland navigation in the Rhine basin has few occupational accidents. The main cause for fatal accidents is drowning. In the last four decades, by increasing awareness, intensifying training and making it mandatory to wear life jackets and to install guardrails on all motor vessels, the German inland navigation sector could reduce fatalities from drowning to 0,15 cases per 1000 employees. This is a reduction to one tenth of the initial value34. Currently, the CCNR is discussing how to repeat this success story for vessels of other nationalities sailing on the Rhine. Recent data provided by the German Berufsgenossenschaft für Transport und Verkehrswirtschaft shows, that the number of cases of work absence due to accidents in relationship to the number of employees is also decreasing. 33 34 www.ccr-zkr.org Communication of the German Delegation, CCNR document RV/G (09) 26, 9th March 2009. 18 PIANC MMX Congress Liverpool UK 2010 Navigation in the Rhine basin is currently experiencing labour related developments that first occurred in maritime navigation some decades ago. Young people in Western Europe are less and less prepared to work on inland navigation vessels, and crews are increasingly recruited from Eastern European or Asian countries. At the same time social security systems are becoming more ‘flexible’, allowing ship owners to hire personnel without making any contributions to social security. Employment of personnel from outside the Rhine basin may be unavoidable in many cases in order to fulfil the manning requirements. Besides, these crews help the ship owner to keep costs low enough to be able to compete in the transport market. However, negative effects also result from this practice. Inadequate language and other skills reduce the ability of the crew to work safely and efficiently, increasing the risks of accidents. The social status and the attractiveness of work on inland navigation vessels decreases further, making it even more difficult to find sufficient personnel and negative effects on the social cohesion within the sector and between the sector and the public occur. Data from Germany indicates that fatal occupational accidents and related work absence in inland navigation is decreasing. Precarious employment conditions, however, seem to become more frequent and may pose a serious challenge with regard to the sustainability of inland navigation, if no appropriate measures will be taken. 4. CONCLUSIONS: WHAT WILL IT TAKE TO MAKE INLAND NAVIGATION A TRANSPORT SYSTEM THAT IS TRULY SUSTAINABLE? Table 7 summarizes the findings of the preceding chapters. A high score means that for aspects covered by the respective indicator, inland navigation seems to be sustainable. The key findings presented in Table 7 derive basically from an assessment of inland navigation in the Rhine basin. These findings apply also to other river basins with similar conditions. In general, the more a waterway profile favours navigation of large vessels, the greater the potential will be for inland navigation becoming truly sustainable. On the other hand, creating sustainable navigation on waterways with a limited profile for navigation or with very unreliable water flows is very challenging. Possibly the most critical issue for inland navigation’s sustainability is its current reliance on liquid fossil fuel. With mineral oil demand constantly increasing and supply reaching a plateau in the near future, costs for mineral oil will increase dramatically. Biofuels not being a viable option for the next decades, LNG seems currently to be the only realistic replacement fuel for inland navigation. LNG must be seen as a transitional fuel, as its use is also not sustainable. Therefore, in the very long-term, a source for renewable fuel will be needed. The burning of fuel in engines is responsible for inland navigation’s most severe environmental impact with the pollutants NOX and PM causing the greatest health and environmental risks. Very stringent emission regulations could reduce the risks to a sustainable level. Pauli and Schweighofer (2009) describe current regulatory developments as well as the technologies available to achieve the necessary reduction of emissions. A switch to LNG as the standard fuel would practically solve this problem, as exhaust emissions from LNG are almost free of the most critical pollutants NOX and PM. Using it as a fuel would also reduce inland navigation’s GHG emissions, as LNG contains relatively less carbon than does mineral oil. However, the research on LNG as fuel for inland navigation has only just started. Supply infrastructure, storage on board and safety seem to be the critical issues for which cost-effective solutions need to be found before LNG can become the main fuel for inland navigation. Reducing fuel consumption is, therefore, for the time being the most realistic option for a reduction of resource use and emissions to air, be it pollutants or GHG. Research aimed at improving fuel efficiency is increasing. Currently available data suggests that operational measures are most promising and fuel savings between 5 and 50% have been demonstrated, depending on individual circumstances. Direct ecological intrusion of inland navigation has been unsustainable in the past. However, inland navigation professionals together with ecologists have in recent years gained the necessary knowledge and developed concepts to reduce the intrusion to a sustainable level. The concept ‘Working with Nature’, as promoted by PIANC, or the programmes of the ICPR for revitalizing nature habitats in and along the Rhine are important examples. Now, waterway authorities must use this knowledge and apply these concepts. The implementation of the programmes of measures developed for the RBMP, as required by the WFD, should bring the ecological intrusion to a sustainable level. Reducing the ecological impact of inland navigation is also a key to improve cohesion between inland navigation development and social expectations and values. Implementation of new concepts, such as 19 PIANC MMX Congress Liverpool UK 2010 PIANC's 'Working with Nature' and well designed processes for stakeholder involvement are other key elements to secure public support for inland navigation. In the Rhine basin, precarious employment in inland navigation is rising. Combining the social security systems developed specifically for personnel on Rhine vessels with EU labour law would reduce this type of employment. Thereby the social status and attraction of inland navigation employment would improve, contributing to its sustainability. Indicators for sustainability Name Accessibility Score High for Rhine, in general mixed Critical Factors Infrastructure (bridge air draught for container transport) High (low costs) Transport operation costs Low in case of severe climate change impacts Outlook Improvements costly Improvements possible Waterway profile, vessel size, water flow Adaptation to negative climate change effects possible within certain limits Shift of transport growth to inland navigation would improve sustainability of overall transport system as inland navigation has generally spare capacities Mixed for equipment and infrastructure utilisation; Transport demand, water flow, RIS Very high for energy efficiency See GHG emissions Costs to economy High (low external costs) See emissions to air, safety, noise Benefits to economy Mixed Low revenues, extra benefits from leisure activities Resource use Very low (unsustainable resource use) Transport fuel Replacement for fossil fuel needed within next decades Direct ecological intrusion Low New concepts and ambitious programs of measures Substantial improvements expected High for GHG Vessel size, design, equipment, operation Improvements of 20% and more possible Very low for certain air pollutants Legal regulations Stringent regulation as for road transport Emissions to soil and water High Legal regulations Better implementation Noise Very high Physical characteristics, legal regulations Adapting regulations to technical progress Waste High Integrated concepts for waste collection and treatment, legal regulation Almost ‘zero’ waste Safety and security Very high Physical characteristics, legal regulations, RIS Better training and management, RIS Social cohesion Mixed Ecological intrusion, stakeholder involvement Implementation of new concepts and practices High for safety Legal regulation Mixed for social security Cost reduction, globalisation of labour force Productivity and efficiency Emissions to air Working condition in transport sector Improvements possible, internalization of external costs, extra benefits are increasing Better and more efficient regulation Table 7: Summary of key findings regarding the greening of inland navigation Offering its support to make inland navigation truly sustainable, PIANC may 1. perhaps on the basis of this paper, examine in more detail the aspects of inland navigation, which are not yet sustainable; 2. systematically and globally monitor the progress reached in respect to these aspects; 3. systematically collect and disseminate good and best practice examples for these aspects; 4. continue to initiate and guide work for these aspects, as it has done in the past. 20 PIANC MMX Congress Liverpool UK 2010 REFERENCES Bonnerjee, S. et al (2009). Inland Waterborne Transport: Connecting Countries. The United Nations World Water Assessment Programme, Insights, United Nations Educational Scientific and Cultural Organization, Paris, ISBN 978-92-104110-5 Buhaug, Ø., Corbett, J. J., Endresen, Ø., Eyring, V. Faber, J. Hanayama, S., Lee, D. S., Lee, D., Lindstad, H., Mjelde, A., Pålsson, C., Wanquing, W., Winebrake, J.J. & Yoshida, K. (2008). Updated study on greenhouse gas emissions from ships: Phase I Report. International Maritime Organisation, London Bundesanstalt für Gewässerkunde (2009). Jahresbericht 2008/2009. Koblenz Centre for Sustainable Transportation (2005). Defining Sustainable Transportation. Prepared for Transport Canada, Toronto CE (2008). L.C. (Eelco) den Boer, F.P.E. (Femke) Brouwer, H.P. (Huib) van Essen. STREAM – Studie naar TRansport Emissies van Alle Modalieteiten, Versie 2.0. CE Delft Christ, P. (2009). Discussion Paper No. 2009-11: Greenhouse gas emissions reduction potential from international shipping. Joint Transport Research Centre of the OECD and the International Transport Forum, Paris Department of Transportation (2009). $100 Million in Obama Administration Economic Recovery Act Funds Charts New Course for Green Transportation. News Release, US Government, Department of Transport, Federal Transit Administration, Washington DC BVB (2009). The Power of Inland Navigation. Dutch Inland Shipping Information Agency, Rotterdam EBU (2009). Inland Navigation – Greening Transport. Annual Report 2008/2009, European Barge Union, Rotterdam ECMT (2006). Inland Waterways and Environmental Protection. OECD, Paris, ISBN 92-821-1346-9 EEA (2000). Are we moving in the right direction? Indicators on transport and environment integration in the EU, TERM 2000. European Environment Agency; Copenhagen, ISBN 92-9167-206-8 EEA (2009). Transport at a crossroads, TERM 2008: indicators tracking transport and environment in the European Union. European Environment Agency, Copenhagen, ISBN 978-92-9167-991-1 European Commission (2008). Greening Transport. Communication from the Commission to the European Parliament and the Council (COM(2008) 433 final), Brussels European Commission (2009). A sustainable future for transport – Towards an integrated, technology-led and user-friendly system. Luxembourg, ISBN 978-92-79-13114-1 European Parliament (2009). Greening of transport and internalisation of external costs. European Parliament resolution of 11 March 2009 on the Greening of transport and the internalisation of external costs (2008/2240(INI)), Strasbourg European Economic and Social Committee (2009). The Greening of Maritime Transport and Inland Waterway Transport. exploratory opinion (TEN/365), Brussels Grave, H. de, M. van Wirdum (redactie) (2006). Brandstofbesparende en CO2 reducerende technieken in de binnenvaart, (edition Bureau Innovatie Binnenvaart, in cooperation with SenterNovem sponsored by Ministerie van Verkeer en Waterstaat) IEA (2009). The World Energy Outlook 2009. International Energy Agency and Organisation for Economic Co-operation and Development, Paris Jonkeren, O.E. (2009). Adaptation to Climate Change in Inland Waterway Transport. Vrije Universiteit, Amsterdam 21 PIANC MMX Congress Liverpool UK 2010 Kruse, C.; Protopapas, A.; Olson, L. E.; Bierling, D. H. (2009). A Modal Comparison of Domestic Freight Transportation – Effects on the General Public: Final Report. prepared for the U.S. Maritime Administration and the National Waterways Foundation; Texas Transportation Institute, Center for Ports & Waterways, The Texas A&M University System, College Station, Texas OECD (2009). Green Growth: Overcoming the Crisis and Beyond. Organisation for Economic Cooperation and Development, Paris PIANC (2004). Guidelines and Recommendations for River Information Services. Report of the Permanent of the Working Group 1 of the Inland Navigation Commission, Brussels PIANC (2005a). Bird habitat management in ports and waterways. Report of PIANC EnviCom Working Group 2, Brussels PIANC (2005b). Economic aspects of inland waterways. Report of PIANC InCom Working Group 21, Brussels PIANC (2008a). Considerations to reduce environmental impacts of vessels. Report of PIANC InCom Working Group 99, Brussels PIANC (2008b). Waterborne transport, ports and waterways: A review of climate change drivers, impacts, responses and mitigation. Report of PIANC EnviCom Task Group 3, Climate change and navigation, Brussels PIANC (2009a). Dredging management practices for the environment – A structured selection approach. PIANC report 100, Environmental Commission, Brussels PIANC (2009b). Innovation in navigation lock design. PIANC report 106, Inland Navigation Commission, Brussels PLANCO (2007). Verkehrswirtschaftlicher und ökologischer Vergleich der Verkehrsträger Straße, Bahn und Wasserstraße. PLANCO Consulting GmbH, Essen, in cooperation with Bundesanstalt für Gewässerkunde, Koblenz Rijkswaterstaat (2007). Monitor Nautische Veiligheid 2006 Landelijk UNEP (2009). Global Green New Deal, Policy Brief. United Nations Environmental Programme, Geneva UNECE (2009). Document E/ECE/1450, Climate Change Mitigation and Adaptation – the Work of the Economic Commission for Europe: Translating Global Objectives and Commitments into Regional Results. United Nations Economic Commission for Europe, Geneva Rahman, A. & van Grol, R. (2005). SUMMA, Final Publishable Report, version 2.0. Rand Europe, Leiden Secretariat of the CCNR (2007). Market Observation for Inland Navigation in Europe 2007-2. Strasbourg Secretariat of the CCNR (2008). Market Observation for Inland Navigation in Europe 2008-2. Strasbourg Werner, S. and P. Rey (2009). Helfen naturnahe Gewässer gegen die Dominanz von Neozoen? Natur und Mensch, 51. Jahrgang, Heftnummer 6 / 2009, Schweizerische Blätter für Natur- und Heimatschutz. Rheinaubund, Schweizerische Arbeitsgemeinschaft für Natur und Heimat, Schaffhausen WHO (2005). Health effects of transport-related air pollution. edited by Michal Krzyzanowski, World Health Organization, Copenhagen, ISBN 92 890 1373 7 22 PIANC MMX Congress Liverpool UK 2010 Annex A SUMMA outcome indicators Outcome of Interest EC1 ACCESSIBILITY EC2 TRANSPORT OPERATION COSTS EC3 PRODUCTIVITY / EFFICIENCY EC4 COSTS TO ECONOMY EC5 BENEFITS TO ECONOMY EN1 RESOURCE USE EN2 DIRECT ECOLOGICAL INTRUSION EN3 EMISSIONS TO AIR EN4 EMISSION TO SOIL AND WATER EN5 NOISE EN6 WASTE SO1 ACCESSIBILITY AND AFFORDABILITY (users) SO2 SAFETY AND SECURITY (users, drivers, the affected) SO3 FITNESS AND HEALTH (users) SO4 LIVEABILITY AND AMENITY (inhabitants, society, the affected) SO5 EQUITY (users and the affected) SO6 SOCIAL COHESION (Inhabitants, society and the affected) SO7 WORKING CONDITIONS IN TRANSPORT SECTOR (employees, drivers, operatives) EC11 EC12 EC13 EC14 EC21 EC22 EC23 EC31 EC32 EC33 EC34 EC41 EC42 EC43 EC44 EC51 EC52 EC53 EN11 EN12 EN13 EN21 EN22 EN23 EN24 EN25 EN26 EN27 EN31 EN32 EN33 EN34 EN41 EN42 EN43 EN44 EN45 EN46 EN51 EN61 SO11 SO12 SO13 SO14 SO15 SO21 SO22 SO23 SO31 SO41 SO42 SO43 SO44 SO51 SO52 SO53 SO61 SO62 SO63 SO71 SO72 SO73 Indicator name Intermodal Terminal facilities Accessibility of origins/destinations Access to basic services (SO11) Access to public transport (SO12) Supplier operating costs Transport- related expenditures of households (SO 14) Transport prices Freight haulage-related cost on product costs Utilisation rates Energy consumption efficiency of transport sector Energy efficiency Infrastructure costs Public subsidies External transport costs Final energy consumption (EN11) Gross value added Public revenues from taxes and traffic system charging Benefit from transport Energy consumption Consumption of solid raw materials Land take Fragmentation of land Damage of underwater habitats Losses of nature areas Proximity of transport infrastructure to designated nature areas Light emissions Collisions with wildlife Introduction of non-native species Transport emissions of greenhouse gases Greenhouse gas emissions from manufacture and maintenance Transport emissions of air pollutants Air pollutant emissions from manufacture and maintenance Hardening of surfaces Polluting transport accidents Runoff pollution from transport infrastructure Wastewater from manufacture and maintenance of infrastructure Discharges of oil at sea Discharges of wastewater and waste at sea Exposure to transport noise Generation of non-recycled waste Access to basic services Access to public transport Car independence Affordability Trip length Accident related fatalities and serious injuries Vehicle thefts & other crimes Security on public transport Walking and cycling as transport means for short distance trips Walkability, pedestrian friendliness Traffic calming Children’s journey to school Open space availability and accessibility Horizontal equity (fairness) Vertical equity (income) Vertical equity (mobility needs and ability) Public opinion profile on transport and transport policy issues Violation of traffic rules Long distance commuting Occupational accidents Precarious employment conditions Work absence due to accidents and illness 23 PIANC MMX Congress Liverpool UK 2010 Annex B 24 PIANC MMX Congress Liverpool UK 2010 25 PIANC MMX Congress Liverpool UK 2010 Annex C Selection of indicators for the assessment of greening of inland navigation As shown in the introductory chapter of this paper, greening of inland navigation can basically be understood as making inland navigation more sustainable and indicators exit to monitor this process. To assess the extent of greening of inland navigation, the outcome indicators of the SUMMA project will be used. The greening is only partly due to public policy measures. For that reason it is preferable to use the more versatile SUMMA indicators rather than the politically more important, but strictly policy oriented TERM indicators. For three reasons, the SUMMA indicators should be used with caution, mainly: First, the indicators are numerous and complex; second, for some of the indicators, data or even relevant methods are not (yet) available; third, for specific analysis, only certain indicators are relevant. The indicators to be used within this paper were selected on the basis of the following criteria: • The indicators are supportable by data. In addition, the available data is easy to reproduce and to understand. • The indicators are meaningful for inland navigation. (For example, the indicator ‘access to public transport’ is of little relevance for inland navigation.) • The indicators are of interest for the PIANC community as this paper is specifically prepared for the PIANC Congress 2010. • The indicators reflect the current debate on the future of transport systems. Table C1 indicates the topics put forward in the current debate on the future of transport systems. The European Commission (2008), (2009) addressed these topics as probably the most relevant for the future of transport systems in the Rhine basin. Policy topic Acronym Policy topic Acronym accessibility AC human capital development HC network integration NI social security SC network maintenance NM climate change CC integration of transport modes IM local pollution (air, soil, water) AP congestion CO noise pollution NP technological innovation TI land occupancy, bio-diversity BIO security SE SA consumption of non-renewable resources RC safety, accidents Table C1: Main topics of current policy papers by the European Commission (2008), (2009) Nevertheless, the selection should avoid discarding those indicators that show inland navigation weaknesses; otherwise, the assessment would be of little value. A weakness indicates an aspect of inland navigation, which may not be sustainable according to the definition of sustainable transport systems, or an area where inland navigation performs worse than rail or road transport. Accordingly, the strength of inland navigation is an aspect, which is sustainable, or an area, where it outperforms competing transport modes. Table C2 shows the results of the selection process. The process is based on data from literature, research projects or other sources, and – in the absence of relevant data – on the author’s expert judgment. 26 PIANC MMX Congress Liverpool UK 2010 Indicator name Relevant for IN Data available PIANC interest Policy debate Strength / weakness Selection EC11 EC12 EC13 EC14 EC21 EC22 EC23 EC31 EC32 EC33 EC34 EC41 EC42 EC43 Intermodal Terminal facilities Accessibility of origins/destinations Access to basic services (SO11) Access to public transport (SO12) Supplier operating costs Transport- related expenditures of households (SO 14) Transport prices Freight haulage-related cost on product costs Utilisation rates Energy consumption efficiency of transport sector Energy efficiency Infrastructure costs Public subsidies External transport costs X X O O + + AC, NI, IM AC O + X X X + O + X X + O + X X + + NI, CO O X X X X X + O O + + + O + CC, AP + + + X X X X EC44 EC51 EC52 EC53 EN11 EN12 EN13 EN21 EN22 EN23 EN24 EN25 EN26 EN27 EN31 EN32 EN33 EN34 EN41 EN42 EN43 EN44 EN45 Final energy consumption (EN11) Gross value added Public revenues from taxes and traffic system charging Benefit from transport Energy consumption Consumption of solid raw materials Land take Fragmentation of land Damage of underwater habitats Losses of nature areas Proximity of transport infrastructure to designated nature areas Light emissions Collisions with wildlife Introduction of non-native species Transport emissions of greenhouse gases Greenhouse gas emissions from manufacture and maintenance Transport emissions of air pollutants Air pollutant emissions from manufacture and maintenance Hardening of surfaces Polluting transport accidents Runoff pollution from transport infrastructure Wastewater from manufacture and maintenance of infrastructure Discharges of oil at sea X X X X X X X X X O O O O O O + + + + + + RC RC BIO BIO BIO BIO BIO O O + + + O O X X X X X X X X X X X X X X X O O + + - + O + O O - BIO BIO CC CC AP AP + + O X X X X X X X X O - O + AP AP O + X X X - O AP O X 27 SA, CC, AP, NP PIANC MMX Congress Liverpool UK 2010 Indicator name EN46 EN51 EN61 SO11 SO12 SO13 SO14 SO15 SO21 SO22 SO23 SO31 SO41 SO42 SO43 SO44 SO51 SO52 SO53 SO61 SO62 SO63 SO71 SO72 SO73 Discharges of wastewater and waste at sea Exposure to transport noise Generation of non-recycled waste Access to basic services Access to public transport Car independence Affordability Trip length Accident related fatalities and serious injuries Vehicle thefts & other crimes Security on public transport Walking and cycling as transport means for short distance trips Walkability, pedestrian friendliness Traffic calming Children’s journey to school Open space availability and accessibility Horizontal equity (fairness) Vertical equity (income) Vertical equity (mobility needs and ability) Public opinion profile on transport and transport policy issues Violation of traffic rules Long distance commuting Occupational accidents Precarious employment conditions Work absence due to accidents and illness Relevant for IN Data available PIANC interest Policy debate Strength / weakness Selection X X X + O O O O AP NP O + + X X X X O O SA + X X - + O X X X X - O - + O O X X X Strength / weakness Selection Symbol Meaning of symbol Relevant for IN X SA, SCC HC, SC SA, SC Data available PIANC interest Policy debate Yes Yes + Sufficient data High interest see Strong O Some data Some interest table Medium - No data Low interest Table C2: Selection of indicators for sustainability 28 Weak Dredging in France – evolution in the practices Philippe RAUJOUAN Centre d'Etudes Techniques, Maritimes et Fluviales Technopôle Brest-Iroise BP5 155, rue Pierre Bouguer 29280 – PLOUZANE - FRANCE 1 [email protected] Keewords : dredging operations, hydrodynamic dredging, dumping at sea, deposit sites, harbour, estuary, French regulation Introduction To fulfil the request of international conventions (London, OSPAR and Barcelone), the French Ministry of Ecology carries out every year a survey about dredging activities with the coastal offices (« services de police des eaux littorales ») in the metropolitan France and its oversea territories. Collected data give an overview of dredging practices according to the types of harbours (technical methods, disposal types) and dredged amounts on various coastlines (The Channel, the Atlantic ocean, the Mediterranean sea and oversea ). The results of the survey also indicate how the issue of the contamination of sediments is dealed and how it evolves according to current regulations. The survey also focuses on the contaminated sediments issue and its evolution with current regulation. The objective of this paper is to show, throughout the results of the annual survey, the state of the dredging activities in France as well as the ongoing discussions about the methods to reduce the impacts of the disposal of dredging material on the marine environment . Various harbour sites and activities Along its 6,000 km coastline, France has many harbour sites with varied activities . The state of siltation of the harbours and the trafic density justifies the regular dredging of access channels and berthing areas. The harbour sites can be classified in 2 categories : The « Grands Ports Maritimes » (GPM), ports of national interest defined by high trafic density. These ports enable to host large draught ships. The seven GPM, along the French coastline, are : Dunkerque, Le Havre and Rouen, along the Channel coast, Nantes-Saint Nazaire and Bordeaux along the Atlantic coast, and Marseille, along the Mediterranean. A subcategory can be defined : the estuarian harbours, located at the end of the estuary, with an access by a navigation channel dug in the river. There are three of them : Rouen, on the river Seine, Nantes-Saint Nazaire, on the river Loire and Bordeaux, on the river Garonne. Little and medium capacity harbours, in which are included regional commercial ports, fishing ports and marinas. 1 Constant volumes of dredged material Over the last five years the amount of dredged material from the French ports for maintenance operations is marked by few changes, from 35,5 Mm3 (in 2007) to 38,5 Mm3 (in 2004), e.g. an average of 37 Mm3. In the same period, the ratio of material dredged in each harbour category also does not change : « Grands Ports Maritime » 80 % (including estuarian ports :70 %) Other ports 20 % More than half of the amount of dredged material comes from the estuarian ports. The hydro-sedimentory process, characteristic of the estuaries, contributes to explain the significant siltation. The macrotidal nature of the three estuaries, their size (Gironde : 625 km², Loire : 60 km², Seine : 50 km²) and the distance of the ports from the mouth of the river (120 km for Rouen, 100 km for Bordeaux, 55 km for Nantes) are additional factors explaining the high volume regularly dredged to maintain the depth enabling navigation in the channels. Evolution in practices Most of usual dredging techniques are deployed during the operations carried out every year (135 maintenance operations in 2008). Hydraulic dredging is the most common technique but the practices start to evolve. Since three years now, the port of Nantes Saint-Nazaire tests on a large scale the hydrodynamic dredging process using the water injection method. The dredger is called « JEDSED ». This operation, subject to environmental monitoring, gives good results and will be renewed the years to come. The following graph presents the advantages and the drawbacks of the hydrodynamic technique Good results Not conclusive on silt recently settled in deep cavities in narrow spaces on sand or in solidified silt in not very deep cavity Advantages Drawbacks cost effective (no sediments transport) not appropriated in case of contamination dredging in places difficult to get to difficult to identify the deposit site reduced restrictions of navigation limited to very specific conditions dredging method with the lowest emission of CO2 final enrichment of the coarse fraction Graph 1 2 The following graph presents the different techniques used in the Grands Ports Maritimes in 2008 Graph 2 The situation is a bit different in the other ports. Dredging with water injection is not used. Only one department (French administrative division) located on the Atlantic coast (La Charente Maritime) experiments « le rotodévasage », method which mixes the silty ground to put into suspension the sediments. Only thousands of cubic meters are removed with this method. The following graph presents the different techniques used in the other ports in 2008 Graph 3 Main evacuation system : dumping at sea Dumping at sea is the most common method applied to treat sediments. This method is even the only one applied in the Grands Ports Maritimes. The low contamination of the sediment displaced during the maintenance of navigation channels allows the dumping operations which are, given the volumes involved, the most economical method and less harmful to the environment. The following graph is the map locating the « Grands Ports Maritimes » (estuarian ports in red) and the quantity dumped at sea in 2008 3 DUNKERQUE 0,9 MT ROUEN LE HAVRE 5,7 MT 2,1 MT NANTES SAINT NAZAIRE 3,1 MT LA ROCHELLE 0,2 MT BORDEAUX XXX 6,3 MT MARSEILLE 1,4 MT Dredged material dumping at sea in 2008 from « Gra Graph 4 In other ports, the distribution is different because contaminants levels may exceed the regulatory thresholds used to give the dumping at sea permit Some high contaminant levels measured in metals and organic/organo-metallic compounds can be explained by an accumulation of the silt due to the low density of maintenance operations in some semi-closed harbours. In these case, alternative solutions need a further investigation. The following graph presents the different sediments treatemant techniques used in the other ports in 2008 Graph 5 4 The following graph presents the location of dumping sites at sea used in 2008 English channel Atlantic Ocean Mediterranean Sea Graph 6 5 Regulation in accordance with europeen directives French regulation for dumping at sea has been drawn up in accordance with OSPAR guidelines for the management of dredged material. Contaminants action levels have been fixed for elements listed by OSPAR commission. The degrees of contamination are set in the Decree of the 9th of August 2006 that defines the contamination threshold level 1 and level 2. If the contaminations are below the level 1 threshold, the sediments can be seen as “lightly contaminated” and without any environmental impact. If the concentrations are between the level 1 and level 2 thresholds, the sediment is seen as “contaminated” and the dredging impacts have to be studied specifically. At last, if the concentrations are higher than the level 2 threshold, the sediment is considered “highly contaminated” with a great potential impact on the environment. The lightly contaminated sediments are usually dumped at sea. However, the contaminated sediments (contamination level between level 1 and level 2) are usually managed on land. Level 1 and level 2 have been defined for metals, PCB and TBT. Investigation is in progress to fix levels for PAH. The following graph presents lower action level (level 1) values for metals (mg/kg dry wt) in France st (1 column) and different countries, in the OSPAR area, using the same approach Arsenic Cadmium Chrome Cuivre Mercure Nickel Plomb Zinc As Cd Cr Cu Hg Ni Pb Zn Level 1 25 1,2 90 45 0,4 37 100 276 Belgique 20 2,5 60 20 0,3 70 70 160 Allemagne 30 2,5 150 40 1 50 100 350 Pays-Bas 29 0,8 100 36 0,3 35 85 140 Norvège 80 1 300 150 0,6 130 120 650 Espagne 80 1 200 100 0,6 100 120 500 Graph 7 The following graph presents upper action level (level 2) values for metals (mg/kg dry wt) in st France (1 column) and different countries, in the OSPAR area, using the same approach Arsenic Cadmium Chrome Cuivre Mercure Nickel Plomb Zinc As Cd Cr Cu Hg Ni Pb Zn Level 2 50 2,4 180 90 0,8 74 200 552 Belgique 100 12,5 300 100 1,5 350 350 800 Allemagne 150 12,5 750 200 5 250 500 1750 Graph 8 6 Pays-Bas 29 4 120 60 1,2 45 110 365 Norvège 400 5 1500 700 3 600 600 3000 Espagne 200 5 1000 400 3 400 600 3000 Management of dredged material : discussions and studies in progress GEODE group 2 In 1990, the French Ministry of transport created GEODE group in order to carry out an optimized management of maritime access integrating environmental, technical and economical issues. The GEODE group gives its expertise to the administration of maritime access of France's coastline commercial, fishing, marina and naval ports by : Taking into accounts the security of navigable depth as well as environmental and economy stakes, Developing expertise and surveys on dredging and environmental issues, Giving advices for the follow-up of dredging and relocation impacts, Contributing to define France's position within international groups. It integrates many scientific fields and aims at : Laying down common frames of reference between contracting authorities and the State departments in charge of investigating and following-up, Developing procedure manuals, Sharing best practices, pooling surveys and researchers, Sharing technical watch. GEODRISK 3 In order to ease the assessment of the risk linked with the sediments, the GEODE group has elaborated a decision support software, GEODRISK, based on the results of chemical analysis that are imposed by the 2000/06/14 decree. GEODRISK applies to all sediments among wich a contaminant exceeds level 1, and it allows to distinguish the dredging sediments according to their contamination level as well as their possible and measured toxicity. The results are given in the form of a risk score that allows the operation appraiser to follow a general view of the process on a decision tree. The toxicity results are expressed in risk stores and included in the process. Dangerous characteristic of sediments Discussions are in progress to establish the dangerous characteristic of contaminated sediments in reference to the list of 41 priority substances fixed by the water framework directive. The Ministry of Ecology has established a working group to develop a protocol for measuring the parameter H14 on the marine and continental sediments. Studies about sediments development SEDIMATERIAUX is a project which involves ports managers, researchers, industrialists. The final objective is to develop operations for management and valorisation of sediments dredged in harbours and navigation channels. As part of this study, the French public administration from the department of Var (on the Mediterranean coast) curried out the “SEDIMARD” project (Dredged Contaminated Sediments) to deal with the different issues related to the management of contaminated dredged sediments. Many Port Management Authorities from France and Italy were involved in the SEDIMARD project. The French public administration from the Var was supported by a group of scientific experts specialized in the management of contaminated sediments to guarantee the scientific quality of the 7 results, to validate the project orientations, the project actions and the results. The SEDIMARD project enabled to understand the mechanism of the studied pre-treatments and treatments, and their interactions. SEDIMARD made possible the comparison of different management treatments To obtain a substitution material from dredged sediment requires some criteria. It might : guarantee the same quality as the previous one it substitutes to, maintain long-term environmental safeguards, be cheap guarantee traceability Regional plan for hazardous waste disposal These plans coordinate, through a region, the actions of organisms who produce hazardous waste and those that ensure the disposal. Furthermore, they coordinate actions of local government or administrations concerned by these issues. These plans will take into account sea wastes, including hazardous dredged materials. Reference document on dredging management plans Some local governments establish a reference document on dredging management plans The aim of this document is to allow all partners involved, to find the best treatment for dredged material. It could be considered as a code of conduct allowing a large consultation with ports' managers, administrations, and environmental organizations. 4 So far, two departments are establishing such a document (Finistère and Morbihan, in Brittany). 1) The CETMEF is a national competence institute attached to the « Ministère de l'Ecologie, de l'Energie, du Développement durable et de la Mer en charge des Technologies vertes et des Négociations sur le climat » .http://www.developpementdurable.gouv.fr/ . The CETMEF advises the Ministry on technical points related to several issues of relevance to EIHA activities, including integrated coastal zone management, dredging, sand and gravel extraction, wind-farms, wave and tidal power, navigation, and also on issues related to inland waterways and flood mitigation. http://www.cetmef.developpementdurable.gouv.fr 2) The Port of Nantes-Saint-Nazaire authority is the leading partner of GEODE 3) Alzieu C. – 1999 - Dredging and marine environment: state of the art [Dragages et environnement marin: état des connaissances]. Ifremer: Plouzané, France. ISBN 2-84433-014-2. 223 pp. Alzieu C. – 2001 - GEODRISK : Software for the Assessment of Risks Linked to the Immersion of Seaports Dredging Sludge. 4) To download the document : http://www.finistere.equipement.gouv.fr 8 PIANC MMX Congress Liverpool UK 2010 DETERMINATION OF HAWSER FORCES USING NUMERICAL AND PHYSICAL MODELS FOR THE THIRD SET OF PANAMA LOCKS STUDIES 1 2 3 4 5 by S. Roux , P. Roumieu , T. De Mulder , M. Vantorre , J. De Regge and J. Wong 6 1. THE PANAMA THIRD SET OF LOCKS PROJECT 1.1 Main issues of the project The Panama Canal Authority (ACP) decided to build a new lane along the Panama Canal that will double capacity and allow more traffic. Along with this new lane, two sets of larger locks, referred to as the Third Set, is under construction as from 2009, one set of locks in the Pacific end and another one in the Atlantic side. Each set of locks will have three consecutive chambers with lengths varying between 427m and 488m, depending on the position of the inner gates, and a width of 55m. The design ship is a so-called Post-Panamax 12000 TEU container carrier (348x48.8x15.2m³; CB=65%). Because water consumption is a major issue, each of the 6 new Locks will be equipped with 3 Water Saving Basins (WSB). Figure 1: Third Set of Locks structure – Overall view This “3 locks & 9 WSB” configuration will help to save 87% of the water required for the transit of one ship between from Pacific ocean to the Gatun lake and the Atlantic ocean (as compared to a single lift lock with no WSB). Even though the New Locks will be wider and longer than the existing locks, they will consume 7% less water than the latter when the WSB are used. 1.2 Studies performed between 2002 and 2008 From 2002 to 2008, the Consortium Post Panamax (CPP) has performed several studies for ACP in order to achieve the preliminary design of the new Locks. Figure 2 shows the schedule and sequence of the studies carried out during this period and briefly presents the models used for each phase: 1 Compagnie Nationale du Rhône - CPP, France, [email protected] Compagnie Nationale du Rhône - CPP, France, [email protected] 3 Flanders Hydraulics Research – CPP, Belgium, [email protected] 4 Ghent University – Maritime Technology Division, Belgium, [email protected] 5 Technum-Tractebel Engineering – CPP, Belgium, [email protected] 6 Autoridad del Canal de Panama, Panama, [email protected] 2 1 of 15 PIANC MMX Congress Liverpool UK 2010 Conceptual studies 2/2002 2/2004 1D num. model Flowmaster Harmonization / actualization study 11/2004 6/2005 2D model Delft 2D Optimization study 1/2006 8/2007 2D model Delft 2D 3D model Fluent - Ananas Physical model study 04/2008 04/2007 1D model Flowmaster 3D model Fluent Figure 2: Study sequence between 2002 and 2008 1D, 2D & 3D numerical model studies were first carried out to select and pre-optimize a filling and emptying (F-E) system. After its completion, the physical model study was started in order to validate the selected system. The validation was based on the two main objectives which are to maximize the transit throughput and, at the same time, to limit hawser forces under an acceptable maximum threshold value. 1.3 Description of the numerical models used during the study Three main software packages were used in the hydraulic study: Flowmaster2 Software: 1D numerical model used for calculation of F-E times; The 1D calculations required to design the basis of the F-E system have been performed with the Flowmaster2 software. The code is able to predict the pressures, velocities and flow rates, the valves operating schedule, the water levels variations and the F-E times,. It can be run either in either steady state or transient simulation. The resulting data allowed to compare different F-E lock designs and to define the main components of the system. Delft2D numerical model: used for the calculation of the water movements in the lock chamber and in the assessment of the hydrostatic component of the hawser forces; The model has been set-up by Flanders Hydraulics Research based upon an accepted, state-of-theart, commercial, software package for shallow water flow (Delft3D software). The assumption of shallow water flow is acceptable because the horizontal dimensions are much larger than the vertical ones (hence vertical acceleration terms in the equations can be neglected) when simulating the translatory waves. This 2D numerical model solves the 2D shallow water equations (i.e. the so-called Saint-Venant equations) for the water flow in the lock chamber. It has been used in order to calculate the variation in time of the water levels in the lock chamber and to determine the hydrostatic pressure field acting on the ship hull during filling or emptying operations. Fluent Software: 3D numerical model used for the design of the most complex hydraulic shapes, for the calculation of head losses in particular lock components and for the calculation of the hydrodynamic forces acting on the vessel during a F-E operation; Fluent is a commercial 3D numerical flow simulation software package. It solves partial differential equations of fluid mechanics (RANS or Reynolds-Average Navier Stokes equations) using the finite volumes method. 2 of 15 PIANC MMX Congress Liverpool UK 2010 The 3D calculations of the hydrodynamic forces acting on the vessel have been carried out meshing the whole lock chamber, a section of the F-E system and the design vessel. Fluid structure interaction and interface tracking (using the VOF method) have been taken into account. Figure 3: Meshing of the chamber, F/E system and vessel 1.4 Description of the physical model A physical model at scale 1/30 was built in order to complete the design of the F-E system and to validate the two main issues which are the F-E times and the hawser forces. The model main elements of the model are: 2 lock chambers; 3 WSB associated with the lower lock chamber; A 250 m fore bay (Gatun lake side); A 250 m tail bay (ocean side). This “2 chambers & 3 basins” configuration allows to carry out all type of F-E operations, with and without using the WSB. Ocean side WSB Lower lock chamber Upper lock chamber Figure 4: General view of the physical model The model is equipped with different types of sensors in order to measure: The water levels in lock chambers, WSB, fore and tail bays; The longitudinal and transversal differential water levels in lock chamber; 3 of 15 Gatun side PIANC MMX Congress Liverpool UK 2010 The velocities and flow rate in culverts; The pressure in the culverts and downstream of the valves; The valve positions; The longitudinal and transversal hawser forces (i.e. the longitudinal and transversal components of the hydrodynamic force exerted by the water on the ship’s hull). A ship model, a Post-Panamax container vessel with 12,000 TEU nominal capacity, has been built at scale 1 to 30 to be used in the physical model of the locks. The ship model is neither equipped with propulsion, nor with a rudder since only lock F-E operations are simulated on the physical model. Figure 5: View of the Post-Panamax container ship model 4 of 15 PIANC MMX Congress Liverpool UK 2010 2. DEFINITION OF HAWSER FORCES AND SETTING OF A HAWSER FORCES CRITERION 2.1 Definition of the term ‘Hawser forces“ The term “hawser forces” can have multiple definitions which need to be detailed and specified in order to define a threshold value. The following definitions have to be distinguished: Forces exerted on the mooring lines These are actually the reaction forces which are opposing to the ship displacement and needed to sustain the ship in a given position in the lock chamber. They depend on the external forces exerted on the vessel, for example the hydrodynamic forces and wind forces, and on the vessel positioning system itself (number, size and position of the mooring lines). These are not the forces measured on the physical model since it would require representing the real configuration of the mooring lines system (which leads to many technical problems). However, these forces can be predicted by a numerical mechanical model of the ship and its vesselpositioning system, fed with time series of measured forces on the ship’s hull. A threshold value for the reaction can not be given if the mooring line system is not defined. Hydrodynamic forces exerted on the ship’s hull These forces are resulting from all the water movements exerted on the ship’s hull (pressure force due to the difference of water level, drag forces issued from the flow around the hull, turbulence force due to the energy dissipation in the lock chamber. These are the forces measured on the physical model with two components: a longitudinal one and a transversal one. Their numerical prediction would require a 3D RANS numerical model with a very refined mesh and taking into account Fluid Structure Interaction (FSI) aspect, which means long and costly numerical calculations. Hydrostatic forces exerted on the ship’s hull These forces are resulting from the hydrostatic pressure distribution along the ship’s hull, i.e. the water level differences along the ship’s hull. These forces are those predicted by the Delft2D numerical model during the initial studies (Delft2D numerical model solved the Shallow Water Equations in the lock chamber taking into account the discharge calculated with the 1D Flowmaster numerical model). The longitudinal and the transversal component of the hydrostatic forces result from these numerical calculations but are only one component (generally the main one) of the hydrodynamic forces. Forces deduced from the water surface slopes The water surface slopes can be used as an approximation of the hydrostatic forces exerted on the ship hull. Two components can be defined: a longitudinal slope and a transversal one. They allow to carry out rapid and easy evaluations of the “hawser forces” in situ since it is far easier to measure water surface slopes in the lock chamber than reaction forces in the mooring lines of a vessel, let alone the hydrodynamic forces on the ship’s hull. As both the water slopes and the “hawser forces” (i.e. the hydrodynamic forces exerted on the ship’s hull) are measured on the physical model, a comparison can result into a relevant correlation between the two parameters. 2.2 The “hawser force” criterion To evaluate the design of a F-E-system, a hawser force criterion is required. Basically, this is simply an attempt to quantify that the filling (and emptying) process should be sufficiently ‘tranquil’ or ‘smooth’ such that the ships moored in the lock chamber do not suffer from unacceptable displacements nor ruptures of mooring lines. Emptying in general gives rise to lower hawser forces than filling since the 5 of 15 PIANC MMX Congress Liverpool UK 2010 energy dissipation takes place outside of the lock chamber. Therefore, although both F-E operations were studied, only filling is discussed in this article. Traditionally the hawser force criterion is formulated by requiring the longitudinal force Fx not to exceed a certain threshold value, which is expressed (in promille) as a fraction of the ship’s displacement weight ∆. Verification of this criterion goes as follows: In the conceptual design phase, the maximum force Fx,max (only the hydrostatic force component) is estimated with a 1D in the present methodology or a 2D shallow water solver (which only accounts for the hydrostatic force component). In the final stages of the conceptual design phase, also a 3D RANS model was used (which allows to estimate also the complete hydrodynamic force). In the design phase, the maximum force Fx,max (including forces of different nature than the hydrostatic force component) is measured on the physical scale model. If Fx,max (expressed in promille as a fraction the ship’s displacement weight ∆) is below the threshold value, then the F-E-system is considered as complying with the hawser force criterion. If Fx,max exceeds the threshold value, then the design of the F-E system should be modified. Note that traditionally, neither in the numerical model, nor in the physical scale model, the real vesselpositioning system, which consists of several mooring lines, is explicitly taken into account in the calculation or the measurement of Fx,max. Stated differently, both in the numerical and the physical models, the design ship is kept in position by means of an artifical ‘vessel-positioning system’, which is only meant to quantify the forces on the ship’s hull. Yet, a real vessel-positioning system does play a role in the hawser force criterion, i.e. it determines the threshold level value. For inland navigation vessels, standard hawser force threshold values are put forward by international organizations (e.g. PIANC) or by national authorities. In some references in literature, different threshold values are specified, depending on the size of the vessel and on the vessel positioning system (e.g. different values for filling with mooring lines attached to fixed bollards vs. mooring lines attached to floating bollards). For ocean-going vessels, however, the references in literature related to hawser force threshold values are very scarce. Moreover, they correspond to a different mooring strategy than the one envisaged for the new Post-Panamax locks. This fact necessitates reflection upon the different strategies that could be adopted to define appropriate – i.e. reliable but not too conservative – hawser force threshold values for large maritime locks in general and for the new post-panamax locks in particular. 2.3 Methodologies available to determine the hawser forces threshold value The definition of a threshold value for the hawser forces requires previously having a fair idea on the mooring lines system to be used. In the case of the Post-Panamax locks, we considered a vesselpositioning system for ocean-going ships consisting of several mooring lines attached to fixed mooring bits on top of the lock wall and to winches on the vessel’s mooring deck. The (12,000 TEU) design vessel will be generally centred in the lock chamber. For the aforementioned system, different methodologies were considered to put forward threshold level values for the hawser forces in the (conceptual) design phase. Authority-based methodology The first method is to rely upon the authority of estimated (international or national) bodies, which suggest specific threshold values. As mentioned earlier, no such specifications were available in literature for Post-Panamax locks. 6 of 15 PIANC MMX Congress Liverpool UK 2010 Water surface slope-based methodology The second method to determine threshold values, is to get information from in situ observed (end-toend) water surface slopes in post-panamax and panamax locks, in which no operational problems related to hawser forces are known to the respective port or canal authorities: - Berendrecht lock (Port of Antwerp, Belgium): up to 0.40 ‰ - Miraflores lock (Existing lock on the Panama Canal using towing electrical locomotives as vessel positioning system): up to 1.30 ‰. In theory, the water surface slope, if measured between bow and stern, is a good measure for the longitudinal component of the hydrostatic force acting on the ship’s hull. This fact is exploited to estimate the hawser force by means of numerical models, based upon a (1D or 2D) solver for the shallow water equations. Force-based methodology The third alternative methodology consists of a rational way to quantify the hawser force threshold values for the aforementioned system, taking into account the characteristics of the vessel-positioning system. More specifically, the forces in the mooring lines are required to remain below certain limits, imposed by the vessel-positioning system. In the longitudinal direction, the threshold value, i.e. the maximum external force Fx,max that can be sustained by the reaction forces in the mooring lines, is given by: Fx , max = Tu fs fm ∑ cos(θ ) cos(φ ) i i i , where: - Tu denotes the minimum tensile strength of the lines, - fs the safety factor (with respect to Tu), - fm the magnification factor to account for the dynamic effects in the mass-spring system formed by the vessel and its mooring lines. A reasonable value is fm=2. N - ∑ cos(θ ) cos(φ ) i =1 i i is the geometrical efficiency (in the longitudinal direction) of the mooring lines, accounting for their geometrical orientation at the beginning of the filling process. The orientation is expressed by two angles θ i and φ i , which depend on the geometry of the lock chamber, the geometry of the vessel and its mooring deck, the (centred) position of the ship in the lock chamber and the water level at the beginning of filling. The optimal orientation of a mooring line is horizontal and aligned with the longitudinal axis of lock chamber and ship: cos(θ i ) cos(φ i ) = 1 . If a given mooring line configuration consists of 2N lines in total, then only N lines of it are active (i.e. are elongated and under tension) in sustaining an external force (e.g. in the direction of the positive x-axis, that is, the longitudinal axis). If those N lines were optimally oriented, then the maximum geometrical efficiency would be obtained: N max ∑ cos(θ i ) cos(φi ) = N . For the hawser force threshold value in the i =1 transversal direction, a similar expression was derived. 7 of 15 PIANC MMX Congress Liverpool UK 2010 Motion-based methodology Besides the previous force-based reasoning, one should also verify whether the ship’s motion, which is the response to the external forces acting on the ship’s hull and to the reaction forces of the mooring lines, to insure that it is within acceptable limits. This is not investigated in the classical numerical model studies. It requires an additional dynamic analysis model to calculate the ship’s motion (for which the classical numerical model studies might give the input time series of forces and moments acting on the ship’s hull). To illustrate the motion-based methodology, a simplified model was set up by the Maritime Technology division of Ghent University (Belgium) to simulate the six degrees of freedom of the 12,000 TEU design container vessel in the new third lane lock chamber, being controlled either in a passive (e.g. by mooring lines acting as linear springs, with limitation of forces) or in an active way (e.g. by tug assistance). The time series of hydrostatic forces and moments on the ship’s hull - as calculated by the 2D numerical model for the water flow in the lock chamber in which a ship is present - are used as an input to the dynamic model. 3. HAWSER FORCES MEASUREMENT AND CALCULATION METHODS 3.1 Assessment of the forces based on the calculation of the water slope In the conceptual design phase, the hawser force analysis boils down to the calculation of the longitudinal component, Fx, of the hydrostatic force on the ship’s hull. It is common to make Fx, nondimensional by taking the ratio of the force and the displacement weight of the design ship. In order to evaluate the variation of Fx in time, one needs the variation in time of the water level difference at bow and stern of the design ship. For this reason the 2D numerical model, described in section 1.3, has been set up to calculate the evolution in time of these water levels. This 2D numerical model solves the 2D shallow water equations (i.e. the so-called Saint-Venant equations) for the water flow in the lock chamber. The whole lock chamber, including the gate recesses, the rolling gates and all of the 40 ports (20 in each lock wall), is meshed. Discharge time series previously calculated by means of the Flowmaster 2 software, are applied as boundary conditions to the 2D model at the port positions. The computational grid is shown in Figure 6, and consists of square cells with size of 1 meter. Figure 6: Computational grid for 2D simulation of water flow in lock chamber 8 of 15 PIANC MMX Congress Liverpool UK 2010 The presence of the design ship is taken into account by defining a fictitious ‘atmospheric pressure field’ on top of the water surface, provoking a ‘trough” in the free surface having similar dimensions as the underwater volume of the ship’s hull. For the first calculations, that pressure field was dimensioned such as to lead to a box-shaped hull (left picture in figure 7), the dimensions of the box being the length, the beam and the draft of the design ship. The pressure field was later adjusted in order to represent better the exact shape of the Post-Panamax hull (right picture in Figure 7) Figure 7: Fictitious atmospheric pressure field in order to simulate the presence of the ship’s hull in the lock chamber After running the 2D model, in each point of the computational grid time series of the water (as well as time series of the two components of the depth-averaged water velocity) are available. From these water level time series, the hydrostatic forces upon the ship’s hull, and in particular the longitudinal component Fx, can be calculated. Figure 8 shows an example of results achieved after a simulation: Figure 8: Upstream & downstream water levels variation in the lock chamber and longitudinal hydrostatic force exerted on the ship hull 9 of 15 PIANC MMX Congress Liverpool UK 2010 An estimate of the transversal component Fy of the hydrostatic force can also be calculated from the differences in water level along the ship’s hull. This methodology based upon 2D and 1D models allows to assess the hydrostatic forces resulting from the water movements in the lock chamber rather quickly. Nevertheless, it can not take into account all components of the hydrodynamic force acting on the ship’s hull. It is consequently very efficient if used for comparison purposes rather than for accurate predictions of hawser forces. 3.2 Calculation of the forces with 3D numerical model 3.2.1 Selection of the numerical model In the conceptual design studies, the calculations were performed using both Flowmaster (1D software) and a Delft2D software model that solves the 2D shallow water equations in order to assess the so-called hawser forces and to compare different configurations of the lock filling and emptying (FE) system. After the calculations performed with Flowmaster and Delft2D software, 4 configurations have been retained. It was then decided to carry out additional analysis with 3D numerical CFD tools (Fluent and Ananas) solving partial differential equations of the fluid mechanics (Navier Stokes equations) using the finite volume method in order to finalize the F-E system design and to retain one configuration to be studied on physical model. The calculation domain that had to be taken into account included several specific issues which required analysis and validation on other cases. The analysis should consider the accuracy of the model but also the calculation time and cost. The main questions that were raised and that had to be answered were: What type of numerical model should be used, Large Eddy Simulation (LES) or Reynolds Average Navier Stokes (RANS)? How to take into account the water/air interface tracking in the calculation? Indeed, the water level is rising during the filling operation and need to be adjusted at every time step. The “Volume Of Fluid” (VOF) and “Level Set” method have been compared and used for this problem. How to take into account the Fluid-Structure Interaction (FSI)? It can be understood that the vessel and the water in the lock chamber interact with each other during a F-E operation. This FSI affects the magnitude of the forces acting on the ship hull and it was necessary to evaluate the effect on the final results. How to take into account the dynamic meshing? Indeed, since the water level and the ship hull are both rising during the simulation and so the boundary conditions are “moving”. It implies that the mesh has to be rebuilt at every time step which is a very costly and time consuming procedure. The capacities of several 3D CFD codes have been compared resulting into the selection of the ANANAS-Lemma code as the most suitable tool for these simulations. It benefits from the most recent and efficient developments based on the works by Inria and some French universities. It provides as standard features an implicit FSI model and a “Level set” interface tracking model. In addition, validation tests have been carried out by simulating the filling of the Zandvliet lock (Antwerp, Belgium). Comparisons have been made with results achieved on physical model by Flanders Hydraulics Research. Finally, it has been decided to run the calculation using the ANANAS-Lemma code with an Euler model for the simulations requiring fluid-structure coupling and the Fluent code with a RANS model to calculate purely free surface flows without fluid-structure interaction (using also the VOF method and an unstructured mesh allowing mesh deformation without remeshing). 10 of 15 PIANC MMX Congress Liverpool UK 2010 3.2.2 Application to the Panama Locks The calculation domain considered, shown in Figure 9, includes the lock chamber and the F-E system. The lock chamber and the gate recesses are represented. The F-E system comprises the main culverts, the secondary culverts, the central distributors and 20 ports on each side of the chamber. Figure 9: Calculation domain The meshing was composed of approximately 1 800 000 tetrahedral elements and 430 000 points. The discharge time series issued from the 1D model (Flowmaster) hydrographs were set as limit conditions in both main culverts. The initial water depth in the lock chamber was set to 18.30 m. The calculations allow to determine the flow distribution through the ports (Figure 10), the water movements in the lock chamber and to calculate the hydrodynamic forces exerted on the ship hull for each of the F-E systems resulting from the conceptual design studies (Figure 11). Figure 10: Example of flow distribution through the ports 11 of 15 PIANC MMX Congress Liverpool UK 2010 0.35 system system system system 0.25 1, standard 2, standard 6, standard 7, standard Force (‰) . 0.15 0.05 -0.05 -0.15 -0.25 -0.35 0 50 100 150 200 Time (s) Figure 11: Longitudinal forces exerted on the ship hull during a filling operation This method gave results with sufficient accuracy to compare the four different configurations of the FE system (so-called system 1, 2, 6 and 7 in Figure 11). The calculations also allowed to assess the magnitude of the expected forces acting on the ship and to evaluate the effect of the Fluid-Structure Interaction on these forces (forces have to be increased by 25% if the FSI is not included in the model). However, due to the calculation time and costs, only a limited number of 3D model simulations for different F-E scenarios (taking into account all hydrodynamic phenomenons) could be investigated. For this reason, a third method, coupling physical model measurements and mathematical model calculation was then set up. 3.3 Combination of physical model measurement and mathematical modelling (motionbased methodology) 3.3.1 Physical model measurements Presently, the physical model seems to be the best tool to represent reliably all the hydraulic phenomena that occur in the lock chamber and that generate and/or interact with the ship movements (provided the scale has been correctly chosen and the ship model is well modelled). On the other hand, the physical model of the lock itself does not permit to measure the forces in the mooring lines because of technical difficulties. Indeed, it is difficult to maintain the mooring lines tight on the model while the vessel is rising or lowering by almost 35 cm (at model scale) during a lockage operation. Moreover, the forces in the mooring lines are very sensitive to the geometric parameters (angles, length) and to the number and type of lines, which means that the results of every test is unique and can hardly be generalised. For instance, any change in the position of the lines would require new tests in order to update the results. Consequently, the methodology that has been retained consisted in: 1- Measuring the forces exerted on the ship during a F-E operation (which means that the force measured depends only on the hydraulic conditions such as initial head and on the ship position; and not on the type, number nor position of the mooring lines); 2- Using these results in a mathematical model developed to calculate the forces in the mooring lines and the ship motion. 12 of 15 PIANC MMX Congress Liverpool UK 2010 The longitudinal and transversal forces were measured on the physical model by means of three dynamometers (two for the transversal forces and one for the longitudinal forces) and positioned as shown on Figure 12. Longitudinal stern Transversal stern Transversal bow Figure 12: Location of the dynamometers This system allows the vessel to move vertically along vertical bars fixed into the lock bottom floor. The dynamometers are fixed on a plate maintained rigidly on the ship as shown in Figure 13. Figure 13: Details of the dynamometers installation This system does not model the elasticity of the mooring lines and all the forces exerted on the vessel are directly transferred to the measurement system. 3.3.2 Calculations of the forces in the mooring lines By means of a simplified mathematical model for the ship dynamics (set up at Ghent University), ship motions and hawser forces due to the longitudinal and lateral forces and the yawing moment measured on the horizontally fixed ship model has been simulated. Calculations have been performed for the most critical F-E scenarios. In the analysis, the response of a moored ship to the measured forces and moment has been simulated making use of (constant) estimated values for the added inertia and hydrodynamic damping coefficients. The vertical motion of the vessel during the lockage operation has been taken into account. The simulations are based on time histories of the vertical motion, the longitudinal force, the lateral force and the yawing moment acting on the ship during the lock operation. For each of the considered F-E operations, two types of time histories have been applied to the moored ship: 13 of 15 PIANC MMX Congress Liverpool UK 2010 The time histories as measured during the physical model tests; A modified time history, obtained by calculating a running average over a time span of 25 seconds for the lateral force and the yawing moment, while the original measured signals are used for the longitudinal force and the vertical motion. Prior to the calculations, a mooring arrangement was selected, taking into account the following requirements and operation assumptions: The ship’s own winches and mooring lines are used; A manual control of each winch by a simple, realistic criterion has been applied; During lock filling and emptying, the ship is kept centred in the lock chamber. The selected mooring configuration consists of eight mooring lines (see Figure 14): Four breast lines (fore starboard, fore port, aft starboard, aft port) Four spring lines (fore starboard, fore port, aft starboard, aft port) Figure 14: Mooring lines arrangement The mooring lines are assumed to be of the type Samson Proton-8 lines, 2-1/4” diameter, i.e. similar to the ones used in ACP tugs. For a number of representative cases, the mooring line configuration defined here was proven to be acceptable to stabilize the vessel position during lock F-E operation. Given the maximum force for the considered scenarios and given the suggested mooring line orientation – in which a separate subset is dedicated to the transversal hydrodynamic forces (4 breast lines) and a separate subset (4 spring lines) are dealing with the longitudinal hydrodynamic forces – it was advocated to define two independent threshold values for assessing physical model results: A longitudinal threshold value of 50 tons, which can directly be applied to the measurements of a physical model A transversal threshold value of 30 tons, which has to be applied to each of the running average (over a time span of 25 seconds, as the ship’s reaction appeared to be not sensitive to fluctuations with a smaller period) of each of the transversal force components measured in a physical model. This methodology, combining both numerical and physical model results, has been a perfect tool to validate the selected F-E system. Based on the results of physical model for the most critical F- E hydraulic scenarios, it leaves the designer with the necessary flexibility to decide on the best mooring system. 4. Conclusion In this paper, an overview has been given of the hawser force modelling efforts that were undertaken during the conceptual design of the Third Set of Panama Canal locks. Combining different types of tools – i.e. both 1D/2D/3D numerical models and a physical scale model for the water flow during F-E operations, as well as numerical models for ship motion and mooring 14 of 15 PIANC MMX Congress Liverpool UK 2010 lines – and a careful reflection upon the definition of hawser force criteria, has proven a valuable and efficient methodology to optimize and validate the selected F-E system. In the final stage of the conceptual design, the most critical hydraulic scenarios of the selected F-E system were simulated in the physical scale model. And the measured forces on the design ship were used to numerically simulate the ship motion, assuming a specific mooring line system. In further design studies, of course, a similar methodology can be exploited, leaving the designer the necessary freedom to decide upon a different, and more optimal mooring system. Finally, it worth mentioning that the ACP finally used the results from the conceptual design studies to define the performance criteria required for the final design and construction of the Third Set of Locks, in order to assure that the works will meet both the navigational safety requirements and the system performance and capacity goals. References TO1-Task1.2.5-CNR-R012 Numerical model of variations for the selected F/E system, CPP, AUGUST 2008 TO1-Task1.2.5-CNR-FHR-R013 Specification of hawser force criteria FINAL REPORT, SEPTEMBER 2007 TO5-Task5.1-CNR-R001 NTERIM REPORT VALIDATION TESTS, CPP, MARCH 2008 TO5-Task5.1-CNR-R002 OPTIMIZATION TESTS REPORT, CPP, JUNE 2008 Mooring Layout for Lock Filling/Emptying Operations, Prof. M. Vantorre, MAY 2008 15 of 15 CPP, P.SCHERRER Technical and Projects Director J.P. GUELLEC Projects Deputy Director PIANC MMX Congress 2010 Port of LE HAVRE facing the challenge of gigantism of container vessels PORT OF LE HAVRE FACING THE CHALLENGE OF GIGANTISM z Prospective on the development of vessel size z A first response: Port 2000 z Future developments: ► Improvement of « Port Rapide Aval » ( Strategic Plan 2009-2013) ► Improvement of draught of Port 2000 first berthes ( Strategic Plan 2009-2013) ► « Port Aval » development (Longer term studies) ► « Port Amont » development (Longer term studies) • New lock, • Restructurated and new terminals. 10 000 6 000 300,00 320,00 CMA CGM Fidelio (42.80) 8 000 340,00 360,00 Emma Maersk hyp - (54.40) 18 000 380,00 20 000 24 000 400,00 Loa (m) 420,00 440,00 Malacca max Post Suez max (60.00) Stx Shipbuilding (50.00) Malacca max (60.00) D3 (56.00) 22 000 Suez Max (50.00) Emma Maersk hyp + (54.40) Global Leader (54.20) Msc Kaline (51.00) D3 50.00) Hanjin Philipines (48.40) 16 000 Post Panamax (49.00) CMA CGM Vela (43.20) Cosco Océania (45.60) Cosco Asia (45.60) 14 000 Cosco Hellas: 349.64 12 000 Zin Los Angeles (45.60) Teu PROJECT VESSELS TEU FC LOA 17.000 Teu 13.000 Teu Teu Linéaire (Teu) 460,00 480,00 500,00 10 000 8 000 14 000 12 000 45.00 50.00 B (m) Emma Maersk hyp - (397.70 ) Global Leader (382.00) Suez Max (400.00) 18 000 55.00 Malacca max Malacca max: 400.00 Post Suez max (470.00) D3 (400.00) Emma Maersk hyp + (397.00) Stx Shipbuilding (450.00) 20 000 Msc Kaline (380.00) D3 (366.00) 22 000 Suez max (400.00) Hanjin Philipines (365.60) Cosco Océania (348.50) Cosco Asia (350.22) Zin Los Angeles (336.70) CMA CGM Vela (346.50) 16 000 Post Panamax (366.00) Cosco Hellas (349.64) 6 000 40.00 CMA CGM Fidelio (349.64) Teu PROJECT VESSELS TEU FC B ( BEAM) 24 000 17.000 Teu 13.000 Teu Teu Regression 13.000 Teu Regression Globale 60.00 65.00 8 000 10 000 6 000 13.00 14 000 14.00 15.00 16.00 Suez max (400.00) Te (m) 17.00 Malacca max: 400.00 D3 (400.00) Emma Maersk hyp + (397.00) 18 000 Emma Maersk hyp - (397.70 ) 20 000 Malacca max Post Suez max (470.00) 22 000 D3 (366.00) Hanjin Philipines (365.60) CMA CGM Vela (346.50) Msc Kaline (380.00) Post Panamax (366.00) 16 000 Zin Los Angeles (336.70) Cosco Océania (348.50) Suez Max (400.00) Global Leader (382.00) CMA CGM Fidelio (349.64 Cosco Asia (350.22) 12 000 Cosco Hellas (349.64) Teu PROJECT VESSELS TEU FC T ( DRAFT) 24 000 17.000 Teu 13.000 Teu Teu Regression 13.000 Teu Regression Globale 18.00 19.00 20.00 21.00 DESIGN VESSEL FOR NEW PROJECTS z 13.000 TEUs : LOA = 400 m B = 55 m (22 u.) T = 16.00 m z 17.000 TEUs : 440 m < LOA < 450 m B = 60 m (24 u.) T = 17.00m PORT OF LE HAVRE FACING THE CHALLENGE OF GIGANTISM z Prospective on the development of vessel size z A first response: Port 2000 z Future developments: ► Improvement of « Port Rapide Aval » ( Strategic Plan 2009-2013) ► Improvement of draught of Port 2000 first berthes ( Strategic Plan 2009-2013) ► « Port Aval » development (Longer term studies) ► « Port Amont » development (Longer term studies) • New lock, • Restructurated and new terminals. A five Kilometer Long New Breakwater July 2004 FIRST PHASE IN OPERATION SINCE APRIL 2006 November 2006 EUGEN MAERSK IN PORT 2000 February 2008 ENVIRONMENT Strong accompanying measures guaranteeing the Coherence of Natura 2000 Network (46 Meuros) Mudflat rehabilitation - 23 MEuros August 2006 DUNE BIRD RESTING AREA (2 M€) ADAPTATIVE MANAGEMENT 31. 01. 02 Birds Island 9M Euros 32 DIFFERENT SPECIES 10 YEARS SCIENTIFIC MONITORING Octeville Phase2: 2100m General aerial view TDF TDF TPO TPO Berth Berth 66 Berth Berth 10 10 September 2008 QUAY BUILDING PRINCIPLES (PHASE2: 2100M) DRAUGHT: 17 m DIAPHRAGM WALL - Building Principle Guiding wall Bentonite Clay 1 Guiding low wall 2 Panel being digged 3 Putting in place of reinforcement frame 4 Concreting of the panel Putting in Place of Reinforcement Frame Berthing Fenders Caping Beam building Tie Rods Installation 31 31 August August 2009 2009 11 December 2009 11 December 2009 PORT OF LE HAVRE FACING THE CHALLENGE OF GIGANTISM z Prospective on the development of vessel size z A first response: Port 2000 z Future developments: ►Improvement of « Port Rapide Aval » ( Strategic Plan 2009-2013) ► Improvement of draught of Port 2000 first berthes ( Strategic Plan 2009-2013) ► « Port Aval » development (Longer term studies) ► « Port Amont » development (Longer term studies) • New lock, • Restructurated and new terminals. Future developments short term René Coty Bassin - current situation Amerique quay 500m GDL*(-13.50) Asie quay 610m GDL*(-13.50) Atlantique quay 765m GDL*(-12.50) * GDL : Guaranteed Dredging Level Future developments short term René Coty Bassin - future situation New quay for 13.000 Teu 480m GDL*(-17.00) Deepening of Asie quay GDL*(-17.00) * GDL : Guaranteed Dredging Level Future developments short term Manœuvres test in the Le Havre Pilots simulator Pict : Le Havre pilots PORT OF LE HAVRE FACING THE CHALLENGE OF GIGANTISM z Prospective on the development of vessel size z A first response: Port 2000 z Future developments: ► Improvement of « Port Rapide Aval » ( Strategic Plan 2009-2013) ►Improvement of draught of Port 2000 first berthes (Strategic Plan 2009-2013) ► « Port Aval » development (Longer term studies) ► « Port Amont » development (Longer term studies) • New lock, • Restructurated and new terminals. PORT OF LE HAVRE TRAFIC CAPACITIES Former port : 2.1 M Teu Tomorrow Port 2000 : 4.2 to 6.3 M Teu Good Terminal capacity for Port of Le Havre at least up to 2020 PORT OF LE HAVRE FACING THE CHALLENGE OF GIGANTISM z Prospective on the development of vessel size z A first response: Port 2000 z Future developments: ► Improvement of « Port Rapide Aval » ( Strategic Plan 2009-2013) ► Improvement of draught of Port 2000 first berthes ( Strategic Plan 2009-2013) ►« Port Aval » development (Longer term studies) ► « Port Amont » development (Longer term studies) • New lock, • Restructurated and new terminals. Future developments Théophile Ducrocq Bassin - current situation Pierre Callet & Joannès Couvert quays 600 + 1530 m GDL*(-12.00) * GDL : Guaranteed Dredging Level Future developments Théophile Ducrocq Bassin - future situation New terminal for 17.000 Teu 1905 m GDL*(-18.50) * GDL : Guaranteed Dredging Level Future developments Théophile Ducrocq Bassin - future situation 3D Pilots manœuvres simulation Picts : Le Havre pilots PORT OF LE HAVRE FACING THE CHALLENGE OF GIGANTISM z Prospective on the development of vessel size z A first response: Port 2000 z Future developments: ► Improvement of « Port Rapide Aval » ( Strategic Plan 2009-2013) ► Improvement of draught of Port 2000 first berthes ( Strategic Plan 2009-2013) ► « Port Aval » development (Longer term studies) ►« Port Amont » development (Longer term studies) • New lock, • Restructurated and new terminals. Future developments François 1er Lock - current situation François 1er lock Future developments Objectives of New Lock Objectives: - Increased Capacities (Size and number of ships) - Improved Safety for maritime acces behind locks - Better use of all port behind the locks François 1er Lock: - Length between inner doors 328 m - Length between outer doors 400 m - Length of biggest ship accepted 348,5 m New Lock under study : - Length between inner doors 440 m Future developments New Lock - future situation New lock for 13.000 / 17.000 teu Future developments New Lock - future situation - principle for gate chamber Future developments New lock for restructurated and new terminals New development for Europe terminal Future terminal New development for Bougainville terminal LE HAVRE facing the challenge of gigantism THANK YOU FOR YOUR ATTENTION OUR WEBSITE: havre-port.fr Préparé par : E. RENAUD / CETMEF J-M. CARLIER / CETMEF S. DUPRAY / CETMEF / AIPCN JP