FOCALIZE: a decision support system to integrate reserve selection
Transcription
FOCALIZE: a decision support system to integrate reserve selection
Journal of Conservation Planning Vol 8 (2012) 45 – 64 FOCALIZE: a decision support system to integrate reserve selection and land use planning through the use of complementary and supplementary criteria Martha Fandiño-Lozano; Willem van Wyngaarden Martha Fandiño-Lozano Ph.D General Director of the UPRA Ministry of Agriculture and Rural Development Avenida Jiménez 7-65 Bogotá, Colombia E-mail: [email protected] Phone: +571-3341199, Ext. 501 or +571-8742129 ABSTRACT: Willem van Wyngaarden Ph.D Scientific Director Grupo ARCO Centro Empresarial Potosí Of. 201 Km 19 Vía La Calera-Sopó La Calera, Cundinamarca Colombia E-mail: [email protected] Website: www.grupoarco.info FOCALIZE is a decision support system developed to select wildlife conservation areas. Using groups of ecosystems (chorological types) as an extended land surrogate, FOCALIZE provides a more complete habitat than a sample of individual ecosystems allows, especially for those animals that move across land units. Targets are based on the requirements of umbrella species and the characteristics of the ecosystems integrating each pattern. Compactness is achieved by concentrating the selected cells around the centers of each chorological type. Naturalness and social preferences are applied in a supplementary way; that is, the selection of each chorological type starts in the most natural and socially preferred sites but, if the target is not achieved, the selection continues on to fragmented or even transformed, and non-preferred areas providing information about how much area must be recovered and where to expect social conflicts between conservation and development. To present the results in a form suitable for decision making, we wrote two extensions of the program: FOCALIZE_Import_Scenario to display thematic maps of the selected planning units and OT-Plan to highlight the conflicts between conservation and economic activities, and to integrate the decisions made into a land use plan. FOCALIZE has been used in Colombia in several planning exercises at the national and regional level. Here the case of the Huila region is shown to illustrate how the system and its extensions operate. Keywords: Algorithm, coarse-filter, reserve selection, extended surrogate, non-arbitrary targets, connectivity, social preferences, wildlife conservation, land use planning. 45 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 INTRODUCTION In the early 1980s, systematic conservation planning came into being with the first reserve selection algorithm (Kirkpatrick 1983). Kirkpatrick´s idea was to divide a whole region into planning units containing comparable data about the features to be considered for selection (Pressey 2002). Worldwide, conservation areas do not include all the habitats necessary for the persistence of every species (Brooks et al. 2004). In Colombia, almost half of the ecosystem types today remain excluded from protection, and many natural parks are too small to support viable populations of the various species (Fandiño-Lozano and Wyngaarden 2005). In general, a reserve selection algorithm is a sequence of explicit steps through which selection criteria and rules are applied. Following this definition, ranking techniques would also be algorithms. The difference between systematic conservation planning and ranking is that ranking involves only a few sites, while systematic conservation planning considers whole countries or regions. The planning units are scrutinized one by one in a systematic search until the minimum targets are achieved, expressed as a percentage of the ecosystem’s extent or as the number of occurrences of the species under consideration (Pressey et al. 2003). The decision adopted by the Conference of the Parties to the Convention on Biological Diversity at its tenth meeting, which took place in Japan in 2010, stated that “by 2020, at least 17 percent of terrestrial and inland water areas, and 10 percent of coastal and marine areas, especially areas of particular importance for biodiversity and ecosystem services, are conserved through effectively and equitably managed, ecologically representative and well-connected systems of protected areas and other effective area-based conservation measures, and integrated into the wider landscapes and seascapes” (CBD 2010). As a series of fixed steps, reserve selection algorithms can be easily programmed utilizing conservation planning software known as decision support systems (DSS). These tools produce in a few minutes what otherwise would demand months or years of difficult manual calculations. Thus, any gap or missing components in a set of systematically selected reserves are no longer attributable to a faulty selection technique, and can be explained by either ineffective methods or low quality data. For such an ambitious political goal to have a real impact on wildlife persistence, the selection of new conservation areas must be highly effective. Effectiveness will depend on the validity of the selection method (Shrader-Frechette and McCoy 1993), the comparability (Margules and Pressey 2000), the quality of the data (Wyngaarden and Fandiño-Lozano 2005), and their application via proper computational techniques. How to select conservation areas for wildlife protection has been an important issue in conservation science for four decades in response to the evident destruction of natural areas and species decline. Two main types of methods have characterized reserve selection since algorithms emerged (Noss and Cooperrider 1994): the fine filter, in which selection is done on target species, and the coarse filter that uses land surrogates which, when conserved, would protect many species at a time. The fine-filter approach covers a large spectrum of species, including taxonomic groups, threatened organisms (Murphy and Noon 1992), endemics or rarities (Rabinowitz 1981), migratory species (CMS 2012), or other species for which data are available (Wright 1977, Stoms et al. 2005). The coarse-filter approach operates mainly on land units (Pressey 2004). Some species were also proposed as biological surrogates, including umbrella species, which are large area-demanding animals and consequently suitable for calculating the minimum extent of reserves (Wilcox 1984). Keystone species were considered because of functional reasons (Paine 1995), and focal species were suggested due to their sensitivity to habitat change (Lambeck 1997). Many reserves were declared as a result of ad hoc decision making (Pressey 1994) or as the default result of approaches in which only the lands useless for agriculture were assigned to wildlife conservation (Klingebiel and Montgomery 1961, FAO 1983). The discussion on how to undertake reserve selection started in the 1970s with the emergence of ranking techniques. Values were assigned to some sites based on various selection criteria and rules (Ratcliffe 1971, Tubbs and Blackwood 1971, Tans 1974, Tjallingii 1974, Gehlbach 1975, Wright 1977, Lesslie et al. 1988); places with higher scores became new conservation areas. Because only a few places were taken into consideration for ranking, significant gaps emerged in reserve systems. 46 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 To start, selecting conservation areas with a coarse-filter is strongly advised, as generating good data for the fine-filter approach is often not feasible (Pressey 2004). Not only must the species and their distribution be known, but the minimum number of individuals needed to maintain viable populations, the ecological requirements of each species, and the carrying capacity of all the existing ecosystems must be determined. For many countries, a complete list of species does not exist, much less the other data mentioned. ecosystems, and naturalness and social viability in a way that creates room for the recovery of ecosystems and reduces conflicts between conservation and development. The method was refined throughout the last fifteen years, although it officially became a DSS in 2002. We used FOCALIZE to select priorities for biological conservation in Colombia (Fandiño-Lozano and Wyngaarden 2003, 2005). After this work was done at a national level, some environmental institutions became interested in performing similar exercises at the regional level. Capacity-building programs started in 2007 to establish FOCALIZE in various decision-making institutions. In this article we use the Huila case to illustrate how the decision support system operates. We do not suggest that individual species should not have any role in reserve selection. The fine-filter approach is designed to complement the coarse filter (Noss 1987). Migratory animals, highly endemic species, and those surviving only in agricultural and urban landscapes after the destruction of their natural habitats, may not get protected in the selected fraction of the land surrogates used as a coarse-filter (Fandiño-Lozano 1996, 2000, 2001, Lombard et al. 2003). Discovering sensible ecological reasons to determine which species need to be targeted will increase the complementary power of the fine and the coarse filters. THE STUDY AREA The Huila region in Colombia is located at 2o south and 75o west (Figure 1), and covers about 2 million ha. It is a highly diverse province, with important gaps in the system of conservation areas. Only 35% remains covered with natural ecosystems, and the transformation of land into agricultural use, urban areas, and other artifacts continues at an alarming rate of about 10,000 ha per year (FandiñoLozano and Wyngaarden 2009); this rapid change is indicative of the situation in the Andean Region of Colombia. Here we present FOCALIZE: a new decision support system. This DSS uses groups of ecosystems as a land surrogate, umbrella species as biological surrogates to calculate minimum targets, compactness to concentrate the selected cells around the center of each group of Figure 1. Location of the study area. 47 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 selection begin? To answer these questions, the selection criteria and the rules are put into place. In FOCALIZE, the complementary criteria set the contents, size and shape of the new conservation areas, while the supplementary criteria help to find the best possible location. The center of the Huila region forms the upper part of the Magdalena river valley at an altitude of 400 to 1,000 m above sea level. It has a semi-arid climate in the north and is more humid in the south, exhibiting a relatively smooth topography and fertile soils. Because of these favorable conditions, irrigated agriculture and extensive livestock production dominate the area. Towards the east, south and west, various mountain ranges reach up to 5,750 m. With increasing elevation, temperature drops and rainfall increases. The upper slopes are extremely steep. The lower slopes have an undulating topography and are used intensively for agriculture; coffee is the main crop. As in many other regions of Colombia, petroleum production is an important economic activity in this region (IGAC 1995). What to Conserve? Many animals move through various ecosystems in search of the various components of their habitat and cannot survive within a single unit or ecosystem type (Jarman and Sinclair 1979, Sukumar 1989, Wiens 1992, Ims 1995, Wittemyer et al. 2007). For this reason, representing ecosystems in conservation areas, as is proposed generally, may ultimately have limited effectiveness in protecting wildlife (Noss 1983, Lombard et al. 2003, Dunning et al. 2006). Groups of spatially related ecosystems can be expected to be a better land surrogate because they provide the natural connectivity between individual units. We previously called such groups chorological types (Fandiño-Lozano 1996), and here we propose them as land surrogates to use in the coarse-filter approach. The variety of physical conditions in the area, together with its geologic history, has given rise to a large diversity in flora and fauna. But today only the higher parts of the mountains remain covered with large tracks of natural vegetation, equivalent to 19.8% of the region. The rest of the Huila only carries fragments spread over 15.2% of the region in lower-lying areas. The region contains sections of 6 national parks and 15 regional and local parks. Of the 31 existing ecosystems, 11 remain excluded from the conservation areas (Wyngaarden and Fandiño-Lozano 2008). The term “chorological” was developed by the Dutch ecologist Isaac Zonneveld (1989). He described the landscape on the basis of the topological (vertical) and chorological (horizontal) dimensions. The idea of using an “expanded coarse-filter” has been present in the literature on conservation planning since Noss (1987) proposed it. However, this suggestion was never put into practice. Problems related to water disequilibrium also affect the Huila region. Out of 33 towns, 9 face insufficient water supply in the dry season and flooding during the rainy season due to increased riverflow. The natural vegetation cover of the watersheds is partly responsible for such disequilibrium; in some cases it is <10%. To secure drinking water supply and reduce flooding, it has been proposed that 28 watersheds should be protected and their vegetation recovered (Fandiño-Lozano and Wyngaarden 2009). Priorities for water regulation can be brought together with the wildlife reserves into a more effective system of conservation areas. Assessing How Much Could Serve as an Effective Target? Effectiveness of conservation areas is not only a matter of features representation, but a matter of species persistence; this has to do with habitat size (Soulé 1987). In that sense, the commonly used arbitrary targets do not have “conservation merit” (Svancara et al. 2005). A politically set percentage of the remaining extent of the ecosystems does not mean anything for the species present. It can be too much for some and too little for others. Here we present a way of calculating non-arbitrary targets. THE SELECTION METHOD IMPLEMENTED IN FOCALIZE For a selection method to be effective, it must answer the following questions correctly: what to conserve, how much of each element should be conserved, and where should For the chorological type, as extended land surrogate, the targets should meet the appropriate size of each ecosystem 48 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 or minimum topological representativeness (MTR), and should also satisfy the spatial balance between those ecosystems integrating a pattern or chorological representativeness (Fandiño-Lozano 1996). same in the conservation areas as it was in the original landscape to achieve chorological representativeness, it is possible to express the area of one ecosystem in terms of the area of another using such equivalencies, thus reducing the number of unknowns to one: The percent with which each ecosystem contributes to the landscape pattern in the original-potential condition probably influenced species distribution. Thus the relative contribution of an ecosystem to the pattern in the conservation area, if not equal, should be as close as possible to that in the landscape before it was transformed into agro systems, settlements and other artifacts; values which can be obtained from the map of the originalpotential ecosystems (Wyngaarden and Fandiño-Lozano 2005). The ideal chorological representativeness is 1, the case in which the proportions between (but not necessarily amounts of) ecosystems are the same in the conservation area and in the natural pattern. (2) Equivalencies areai = oai * area1 / oa1 areai = minimum area of ecosystem i needed MVP = minimum viable population oai = original area of ecosystem i The last step is to find the value of the MTR: (4) Minimum Topological Representativeness MTR = area1 / oa1 * 100% The MTR is expressed as the percentage of the original extension of the ecosystems (Fandiño-Lozano 1996). This percentage is shared by all the other ecosystems integrating the same chorological type, even by those where the umbrella species does not occur. In those cases the carrying capacity will be zero. The minimum area of each ecosystem will vary depending on its original extent. (1) General equation MVP = cc1 * area1 + cc2 * area2 +...+ cci * areai (3) Equation with a single unknown area1 = MVP / (cc1 + cc2 * oa2/oa1 +...+ cci *oai/oa1) Mathematically this principle can be written as follows: cci = carrying capacity of ecosystem i areai = minimum area of ecosystem i If, for example, the original area of ecosystem i is 200 ha, and the area of ecosystem 1 is 100 ha, then the minimum area of ecosystem i is 2 times the minimum area of ecosystem 1. After substituting equation 2 into equation 1 and rewriting it, we obtain the following: Regarding the MTR, we use the most area-demanding animals (umbrella species) merely to determine the minimum area of each ecosystem to be selected. If minimum viable populations (MVP) of umbrella species are expected to survive in the selected conservation areas, each ecosystem integrating a particular chorological type should have the minimum area to contribute effectively to the habitat needed according to the carrying capacity of the ecosystems (Fandiño-Lozano and Wyngaarden 2003). The balance between the ecosystems or chorological representativeness is automatically obtained because the equivalencies reduce the number of unknowns in equation 1. The MVP and the carrying capacities can be estimated using ecological and biological data. The minimum areas of the ecosystems are the unknowns to be solved in the equation. However, the number of unknowns is equal to the number of the ecosystems in the chorological type (area of ecosystem 1, 2…i), which makes it impossible to solve the equation. Nonetheless, since the contribution of each ecosystem to the chorological type must be the Where Should Selection Start? Wildlife reserves should be as round as possible to minimize the area exposed to external threat (Janzen 1986) and edge effects (Wilcox and Murphy 1985, 49 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 Schonewald-Cox and Bayless 1986). Compactness (Ball and Possingham 2000) and adjacency (Nicholls and Margules 1993) have been used in other algorithms to join together selected units, thus reducing the number of clusters or groups of cells. effect in preventing excessive clusters in fragmented landscapes. We use social needs and preferences as supplementary criteria as well. Social variables have been considered in reserve selection since the 1970s either to increase or lower the value of a site for conservation (Tans 1974, Gehlbach 1975, Wright 1977, Bedward et al. 1992, Wessel et al. 2000, Pressey and Taffs 2001, Reyers et al. 2002). MARXAN, a very popular DSS today, combines social variables into “a cost function” after expressing them in the same currency (Ball and Possingham 2000). FOCALIZE also uses compactness, but follows a logic that suits the land surrogate we developed. The ecosystems which are pulled together are those belonging to each chorological type. This is done by selecting the planning units as close as possible to the center of each chorological type, which happens to be the center of the ecosystems belonging to that pattern. In our method, if two sites show equal contents, the one that can be easily assigned to biological conservation is selected. The preference given to an area may have a positive character — for instance, the motivation to set a new reserve to regulate water supply — or a negative character, such as for areas where productive activities or development projects are planned. Planning units can also be excluded from selection in built-up areas, and existing conservation areas can be preselected, given a positive preference. In addition to compactness, the selection should make optimal use of natural remnants. In the 1970s it was proposed that conservation areas should be selected if they were natural or pristine (Ratcliffe 1971, Tubbs and Blackwood 1971, Spellerberg 1981, Wright 1977, Lesslie et al. 1988). Later, in systematic conservation planning, the selection of new reserves became made exclusively on natural remnants. When the targets were higher than the current natural cover, they were “truncated to present extant areas” (Pressey et al. 2003). Recovery of diminished ecosystems has not been considered for achieving minimum targets. THE ALGORITHM AND THE SOFTWARE FOCALIZE is a stand-alone computer program written in Visual Basic (Wyngaarden and Fandiño-Lozano 2011). It executes a heuristic algorithm written on the basis of the described selection criteria and rules. It chooses from square planning units or cells, into which the study area has to be pre-divided. Cells can be of any size with a maximum number of 32,000. For each chorological type the algorithm does the following: In our method, naturalness is used as a supplementary selection criterion. The selection starts in the areas with a natural cover and, if there is insufficient extent to reach the target per ecosystem, the algorithm continues with fragmented or totally transformed areas. This implies that the data pertaining to the original-potential cover of the ecosystems must be available. This selection consideration provides the information needed for protection and also for recovering diminished ecosystems. However, looking for the last natural fragments may split the chorological type into an unnecessarily high number of clusters. To avoid such an effect, it is possible to switch to less natural areas once a percentage of the target is reached on the basis of natural cover. The same compacting effect can be achieved by stopping the selection when the MTR is reached by summing the natural, the fragmented and the original-potential covers within the cells being selected. These options have no major effect on the dispersion of clusters in natural areas whereas they have a very positive 50 • chooses those cells that contain the ecosystem with the lowest MTR achievement in its chorological type; • if more than one cell is available, maximizes naturalness by first selecting cells where the ecosystem has a natural cover, followed by cells that have only fragments or, if not available, a totally transformed cover; • if more than one cell available remains, maximizes social benefits and minimizes social conflict by selecting among preferred cells and, if the target Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 • cannot be reached, looks for cells without preference and ultimately for those with a negative preference; and how they will relate to the use of the territory and to development initiatives. if more than one cell available still exists, maximizes connectivity by selecting the cell closest to the center of the chorological type and, if available, neighboring to already selected cells. To make this critical information available and approachable for the non-expert, we programmed two extensions of FOCALIZE in the public domain GIS software ILWIS (52North 2011): “FOCALIZE_Import_Scenario” to display thematic maps of the selected planning units, and “OTPlan” to highlight the conflicts between conservation and economic activities and, based on the decisions made, to produce a land use plan. This process continues until the MTR for all ecosystems of one chorological type is reached; then the algorithm proceeds with the next chorological type. Finally, it includes cells that border selected cells on at least three sides, so that the perimeter of the selected areas is reduced. Figure 2 illustrates a model of how the algorithm operates. THE DATA The database for the Huila region consists of 16,250 cells of 400 ha each, containing information on the quantity of 7 chorological types, 31 ecosystems in 4 levels of naturalness, the presence of 3 social variables and the current conservation areas (Figure 3). Correct interpretation of the scenarios produced by FOCALIZE, or any DSS, requires training and in-depth knowledge of conservation planning methods, tools and data. However, decisions regarding new conservation areas are ultimately made by governmental agents, who usually are not experts in the topic and who do not have the time to examine details. To make sound conclusions, these decision-makers need as much clarity as possible regarding why the proposed new reserves are needed The ecosystems must be known as they are today (actual) and as they were originally or would potentially be, if recovered (original-potential). The original-potential Figure 2. Rules to apply the selection criteria. 51 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 Figure 3. Data needed to run focalize. Note: The square brackets indicate the data from which the other data are derived. The arrows show the data flow. Urban areas are not shown in the figure because of their small size. 52 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 cover is needed to differentiate the chorological types and find their centers, to calculate the targets, and to provide the levels of naturalness. distribution. Because they are competitors, the targets must be summed where they are co-present. Based on food chains, the carrying capacities of the ecosystems were calculated. The main prey of jaguar and puma are large- and medium-sized terrestrial herbivores. Because herbivore density in Colombia is unknown, we predicted the potential density using data from comparable African ecosystems (e.g., Bell 1982, Wyngaarden 1985, Bie 1991). Using these predictions and the known consumption of large African carnivores (Sinclair and Norton-Griffiths 1979), the maximum sustainable density or carrying capacity for jaguar and puma was estimated (Table 1, columns 5 and 6). The mapping of the actual ecosystems was done following standard procedures (Küchler and Zonneveld 1988). The original-potential distribution of the ecosystems was mapped by observing the recently modified limits on satellite images and by spatial modeling in those cases where it was not possible to see them anymore. Once they were mapped, the boundary length was calculated using GIS operations, and the obtained values were clustered with a principle component analysis (PCA), using CANOCO (Braak 1987). The chorological types were mapped by joining the ecosystems that shared a large proportion of boundaries (Fandiño-Lozano 1996, Wyngaarden and Fandiño-Lozano 2005). Finally, the centers of the chorological types were calculated using GIS operations. Equivalencies come from the ecosystems map on the basis of their extension in the original landscape (Table 1, column 4). With the data on equivalencies and the carrying capacity of the umbrella species, the targets were calculated using equations 3 and 4. The MTR varied from 15% to 100% of the original-potential extent of the ecosystems, and from 19 to 61,302 ha (Table 1, columns 7 and 8). The data about the existing conservation areas, and the social needs and preferences, were independent cartographic inputs. For the calculation of the MTR, we choose Panthera onca and Puma concolor as umbrella species; the two largest carnivores in Colombia. Their MVP was estimated using Belovsky (1987): 300 individuals for jaguars and 400 for pumas. The jaguar occurs in the more humid ecosystems at lower altitudes, while the puma has a much wider Chorological types Table 1. Data on the chorological types and ecosystems of the Huila Region. CT 38 CT 44 CT 45 CT 46 CT 48 CT 50 CT 51 Ecosystems Code Equivalence Description Central Cundiboyacense high plain and surroundings MPH5 Glaciated landscape with dense tussock grasslands MHH2 High andean low forests MHM2 High andean shrublands and dwarf forests MAH4 Evergreen andean forests MSH5 Evergreen sub-andean forests Eastern Cordillera: humid upper slopes MPH6 Glaciated landscape with dense tussock grasslands MAH6 Evergreen andean forests MAP3 Evergreen andean forests MSH6 Evergreen sub-andean forests Eastern Cordillera: sub-humid lower slopes MSS6 Semi-deciduous sub-andean forests MSA2 Deciduous sub-andean open forests MTS5 Semi-deciduous forests of the andes foothills MTA6 Deciduous forests of the andes foothills Tatacoa dry enclave VTB6 Evergreen forests of the seasonal river plains VTM1 Dense deciduous shrublands on terrace remnants VTM2 Open deciduous shrublands and grasslands on dissected terraces Upper part of the Magdalena river valley MSS7 Semi-deciduous sub-andean forests MTS8 Semi-deciduous forests of the andes foothills VBB2 Deciduous forests of on the alluvial fans VSB1 Deciduous forests of the low hills VTB5 Deciduous forests of the terraces VRB2 Riparian forests of the major floodplains Central Cordillera: páramos and upper slopes MNN1 Permanent ice and snow MNN2 Glaciated landscape without vegetation MUH4 Glaciated landscape with sparse tussock grasslands MPH7 Glaciated landscape with dense tussock grasslands MHH3 High andean low forests MHM3 High andean shrublands and dwarf forests Central Cordillera: middle and lower slopes MAH8 Evergreen andean forests MSH9 Evergreen sub-andean forests MTH8 Evergreen forests of the andes foothills 53 Carrying capacity (nr/km 2) Target (MTR) % of original extent Panthera onca Puma concolor ha 1.000 0.202 0.855 1.998 0.627 0.00 0.00 0.00 0.00 0.00 0.48 0.48 0.48 0.48 0.48 11,939 2,414 10,205 23,860 7,482 100.0 100.0 100.0 100.0 100.0 1.000 596.250 588.710 3,299.830 0.00 0.00 0.00 0.00 0.48 0.48 0.48 0.48 19 11,077 10,937 61,302 18.6 18.6 18.6 18.6 1.000 0.111 1.027 0.256 0.80 0.80 0.80 0.80 1.44 1.44 1.44 1.44 27,261 3,031 28,010 6,975 17.8 17.8 17.8 17.8 1.000 1.775 1.463 0.00 0.00 0.00 1.44 0.96 0.96 6,162 10,937 9,016 100.0 100.0 100.0 1.000 1.589 0.387 0.270 1.044 0.408 0.80 0.80 0.80 0.80 0.80 1.33 1.44 1.44 1.44 1.44 1.44 2.40 13,135 20,871 5,089 3,549 13,713 5,360 16.7 16.7 16.7 16.7 16.7 16.7 1.000 0.594 3.386 36.718 151.762 30.015 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.48 0.48 0.48 0.48 376 223 1,272 13,790 56,998 11,273 82.0 82.0 82.0 82.0 82.0 82.0 1.000 2.352 0.196 0.00 0.00 0.00 0.48 0.48 0.48 23,493 55,246 4,594 15.6 15.6 15.6 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 RESULTS OF A SCENARIO FOR THE HUILA REGION 541,788 ha were selected as priorities for conservation, corresponding to 28.5% of the study area. Priorities for water regulation were given a positive preference, and highly productive agricultural lands and areas assigned to petroleum exploitation, a negative one. Urban areas are always excluded because they cannot be reallocated to set new conservation areas. Existing natural parks were ignored during the selection. We omitted them in this example to avoid distortions in the results. The scenario is displayed in thematic maps and four tables. Table 2 contains the attributes of the selected cells and importing them into “FOCALIZE_Import_Scenario” produces the maps in Figure 4. The other tables contain the statistics on the chorological types (Table 3), the ecosystems (Table 4) and the clusters obtained (Table 5). Figure 4. Thematic maps in the scenario. 54 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 Table 2. The attributes of the selected cells. Cell …. Cell 1113 Cell 1114 Cell 1115 Cell 1119 Cell 1120 Cell 1121 Cell 1237 Cell 1238 Cell 1239 Cell 1240 Cell 1241 Cell 1242 …. Centre coordinates North West 903000 903000 903000 903000 903000 903000 901000 901000 901000 901000 901000 901000 940000 942000 944000 952000 954000 956000 938000 940000 942000 944000 946000 948000 Cluster number Chorological type Ecosystem Cluster 1 Cluster 1 Cluster 1 Cluster 1 Cluster 1 Cluster 1 Cluster 1 Cluster 1 Cluster 1 Cluster 1 Cluster 1 Cluster 1 CT 38 CT 38 CT 38 CT 38 CT 38 CT 38 CT 38 CT 38 CT 38 CT 38 CT 38 CT 38 MSH5 MSH5 MSH5 MHM2 MPH5 MPH5 MSH5 MSH5 MSH5 MAH4 MAH4 MAH4 Transformation (%) 0.0 0.0 34.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 16.7 Reason Reason 14 Reason 14 Reason 34 Reason 14 Reason 14 Reason 14 Reason 14 Reason 17 Reason 17 Reason 14 Reason 14 Reason 34 Note: The column “Reason” gives a code for the combination of criteria used to select the cell. Reason 14 means: cell without preference, with a natural cover and adjacent to an already selected cell; reason 17 means: cell with a negative preference, with a natural cover and adjacent to an already selected cell; and reason 34: cell without preference, with a fragmented cover and adjacent to an already selected cell. Some chorological types ended up in one cluster and in balance (38 and 46). Other chorological types were split into various clusters because of the ecosystems patterns in natural areas, the transformation of the landscape, the positive or negative preferences, and the limits of the study area (Table 3, Figures 4c and 4d). The possible inclusion of the clusters of a chorological type in one reserve depended on the situation that created the grouping of cells. Chorological type 44 was split because of the positive preference given to watersheds (Figure 4e). Chorological types 45, 48 and 51 ended up in various clusters because of the transformation of the landscape (in grey in Figure 3). Artificial clusters were produced in the case of type 50, located at the edge of the study area: two clusters were part of a large unit located in two regions, so the limit of the Huila created a separation that does not exist in reality (Figure 4c). based on the current extent of the natural areas. Reaching the MTR requires the recovery of 110,354 ha (Table 4, column 13). Figures 4a and 4b identify where and for which ecosystems such actions are needed, and the fragments that can serve as sources for recovery. In some cases, the achievement of the MTR was higher than 100% of the target. This occurred when an ecosystem was small in relation to the cell size. For example, in ecosystem MPH6, chorological type 44, the achievement had a value of 538% because the target was 19 ha and the selected cell contained 100 ha (Table 4, columns 7 and 8). Concerning shape, none of the clusters shows the ideal circular shape as expressed in the ratio BL/iBL = 1 (Possingham et al. 2000). Perfect round shapes are rare in nature and are seldom achieved in reserve selection; this ideal shape has to be seen as an indicator. The majority of the obtained values were not far from 1 (Table 5, column 5), with the exception of the two largest clusters bringing together several chorological types. Regarding targets, the MTR was always reached on the basis of the original-potential cover; not on the basis of the extant cover (Table 4, columns 11 and 12). Of the 31 ecosystems located in the Huila, 13 do not meet the target 55 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 Figure 5 (right side) shows how OT-Plan made use of these results to indicate the conflict between conservation and development. The legend indicates the conservation areas, existing or selected, that resulted in conflict either with highly productive agriculture (upper part) or with petroleum (lower part). From 319,817 ha of highly productive agricultural lands, 7.2% are in conflict with biodiversity and water conservation. And from 94,579 ha of interest for petroleum, 22.7% were selected in the scenario for conservation. The relation between the selected cells and the positive and negative preferences used to build the scenario is shown in Figure 4e. As in the scenario we gave a positive preference to some watersheds, the algorithm selected 58.7% of their extent to achieve the targets for biological conservation. But as those areas did not have all the ecosystems integrating the chorological types, 30.3% had to be selected in areas without positive preference and even 9.9% in areas with a negative preference. Table 3. Chorological information of the scenario. Chorological type Ecosystem CT 46 CT 46 CT 46 CT 50 CT 50 CT 50 CT 50 CT 50 CT 50 CT 50 CT 50 CT 50 CT 50 CT 50 CT 50 CT 51 CT 51 CT 51 CT 51 CT 51 CT 51 CT 51 CT 51 CT 51 CT 51 CT 51 CT 51 CT 51 CT 51 CT 51 VTB6 VTM1 VTM2 MNN1 MNN2 MUH4 MPH7 MHM3 MHH3 MNN1 MNN2 MUH4 MPH7 MHM3 MHH3 MAH8 MSH9 MTH8 MAH8 MSH9 MTH8 MAH8 MSH9 MTH8 MAH8 MSH9 MTH8 MAH8 MSH9 MTH8 Cluster identifier Chorological representativeness 2 2 2 6 6 6 6 6 6 16 16 16 16 16 16 7 7 7 11 11 11 14 14 14 15 15 15 16 16 16 56 1.00 1.00 1.00 25.04 16.46 7.94 0.06 3.06 0.44 0.09 0.47 0.84 0.96 0.99 1.02 1.23 0.98 0.00 0.00 1.16 4.07 0.00 0.37 13.61 3.53 0.00 0.00 1.69 0.78 0.00 Achievement MTR within cluster (%) 100.0 100.0 100.0 111.8 73.5 35.5 0.5 13.7 2.0 10.1 48.4 86.5 99.5 102.0 104.7 47.7 38.2 0.0 0.0 11.4 39.7 0.0 2.1 74.9 23.5 0.0 0.0 94.0 43.9 0.0 Ecosystem MPH5 MHH2 MHM2 MAH4 MSH5 MPH6 MAH6 MAP3 MSH6 MSS6 MSA2 MTS5 MTA6 VTB6 VTM1 VTM2 MSS7 MTS8 VBB2 VSB1 VTB5 VRB2 MNN1 MNN2 MUH4 MPH7 MHH3 MHM3 MAH8 MSH9 MTH8 Chorological type CT 38 CT 38 CT 38 CT 38 CT 38 CT 44 CT 44 CT 44 CT 44 CT 45 CT 45 CT 45 CT 45 CT 46 CT 46 CT 46 CT 48 CT 48 CT 48 CT 48 CT 48 CT 48 CT 50 CT 50 CT 50 CT 50 CT 50 CT 50 CT 51 CT 51 CT 51 natural ha 11,939 2,414 10,205 23,119 6,421 100 45,731 57,483 102,508 44,673 5,689 5,390 6,788 1,239 8,750 7,213 2,470 6,353 516 2,697 17,277 712 458 272 1,551 16,817 68,333 13,747 117,750 72,980 2,654 Current condition of the ecosystems MTR Huila transformed total transformation ha ha % ha 11,939 11,939 2,414 2,414 10,205 10,205 741 23,860 3.1 23,860 1,061 7,482 14.2 7,482 100 19 13,894 59,625 23.3 11,077 1,388 58,871 2.4 10,937 227,475 329,983 68.9 61,302 108,213 152,886 70.8 27,261 11,310 16,999 66.5 3,031 151,698 157,088 96.6 28,010 32,331 39,119 82.6 6,975 4,923 6,162 79.9 6,162 2,187 10,937 20.0 10,937 1,803 9,016 20.0 9,016 75,965 78,435 96.9 13,135 118,275 124,628 94.9 20,871 29,875 30,391 98.3 5,089 18,495 21,192 87.3 3,549 64,609 81,886 78.9 13,713 31,292 32,004 97.8 5,360 458 376 272 223 1,551 1,272 16,817 13,790 1,174 69,507 1.7 56,998 13,747 11,273 32,490 150,240 21.6 23,493 280,316 353,296 79.3 55,246 26,726 29,380 91.0 4,594 natural ha 11,939 2,414 10,205 23,119 6,421 100 14,251 13,887 54,914 21,844 1,854 5,390 3,921 1,239 8,750 7,213 2,470 5,457 516 1,150 7,421 320 458 272 1,551 14,165 59,700 13,043 38,880 41,819 1,422 Selected area transformed ha 741 1,061 1,051 519 16,485 9,981 1,905 23,587 8,935 4,923 2,187 1,803 12,203 20,698 5,631 5,357 12,313 7,771 1,111 2,099 21,479 3,843 ha 11,939 2,414 10,205 23,860 7,482 100 15,302 14,406 71,399 31,825 3,759 28,977 12,856 6,162 10,937 9,016 14,673 26,155 6,147 6,507 19,734 8,091 458 272 1,551 14,165 60,811 13,043 40,979 63,298 5,265 total Achievement in % of target natural total 100.0 100.0 100.0 100.0 100.0 100.0 96.9 100.0 85.8 100.0 538.3 538.3 128.7 138.1 126.9 131.7 89.6 116.4 80.1 116.7 61.1 124.0 19.2 103.4 56.2 184.3 20.1 100.0 80.0 100.0 80.0 100.0 18.7 111.7 26.1 125.3 10.1 120.7 32.4 183.3 54.1 143.8 6.0 151.0 121.9 121.9 121.9 121.9 121.9 121.9 102.7 102.7 104.7 106.7 115.7 115.7 165.4 174.4 75.7 114.5 31.0 114.6 Minimum area to be recovered ha 741 1,061 6,388 5,418 1,177 22,620 3,054 4,923 2,187 1,803 10,665 15,414 4,573 2,399 6,293 5,039 13,426 3,172 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 Table 4. Topological information of the scenario. 57 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 Table 5. Characteristics of the clusters. Cluster identifier 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Cells no 347 151 27 22 10 14 96 23 51 46 38 27 15 66 17 449 14 14 10 21 Area Boundary 2 km km 1,190.3 412 588.6 140 92.9 68 88.0 60 40.0 36 41.6 36 359.4 128 92.0 56 204.0 76 184.0 68 152.0 88 108.0 84 46.6 40 264.0 116 55.4 44 1,644.8 528 55.7 36 56.0 40 40.0 28 66.5 68 BL / iBL km/km 3.36 1.62 1.99 1.80 1.60 1.57 1.90 1.64 1.50 1.41 2.01 2.28 1.65 2.01 1.66 3.67 1.36 1.50 1.24 2.35 Table 6. Land uses in OT-Plan. Land-use Biodiversity conservation Biodiversity and water conservation Water conservation Petroleum exploitation Highly productive agriculture Other agriculture and forestry uses Urban centers 58 ha 569,714 110,458 52,962 72,731 288,956 805,395 6,063 % 29.9 5.8 2.8 3.8 15.2 42.2 0.3 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 Figure 5. Procedure and results of the FOCALIZE extension OT-Plan. 59 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 DISCUSSION equivalencies that relate the area of one ecosystem to that of another, on which basis the equations could be solved. The chorological representativeness was the critical key to finding the minimum topological representativeness and presents a graceful solution to the traditional lack of biologically-sound targets. From the total extent of the Huila region, only 2.3% shows a conflict between conservation and economic activities. Figure 5 and Table 6 contain the results of combining the decisions made into a land use plan after analyzing the conflicts. Biodiversity and water conservation prevailed over the other uses, and petroleum predominated over agriculture. The order could change in other scenarios. The results obtained for the Huila region illustrate the importance of counting with non-arbitrary targets. The MTR varied from 15 to 100%. A general target, such as the 17% set by the CBD was insufficient in 71% of the ecosystems present in the Huila. The results obtained not only present a transparent process of reserve selection, but also for land use planning. Such integration improves the decision-making process by providing a more understandable mental picture about how to combine conservation and productive activities in a concrete territory. The low values obtained for areas in conflict probably will support completing the system of conservation areas because the constraints conservation imposes on development are perceived as marginal. In the MTR equation the concept of umbrella species also found a place. In this way, MVP and minimum sizes were brought together in solving the problem of targets; two approaches which have been traditionally seen as alternatives (Soulé 1987) instead of essential elements of the same solution. A serious effort must take place in the fields of genetics and ecology in order to refine the values in the MTR equation. The MVP should be calculated using the best available methods, and the carrying capacities of the ecosystems must take into consideration not only food, but all the ecological requirements as well as behavioral issues such as territoriality. Reducing conflicts can also be achieved in optimizing algorithms by joining social variables into a cost function, as MARXAN does. However, to express many variables in one currency is very difficult when dealing with things “that are not ready trade” (Brennan 1996, Shrader-Frechette 1984, 1987). Also, by adding up cost layers into a cost function, the selection process can become a black box for decision makers (Ardron et al. 2010). The equation does not solve the uncertainty about MVP which remains the worst methodological gap in conservation biology. As long as doubt shrouds the subject of minimum areas, we strongly support the recommendation of Noss et al. (1997): “conservation plans should be conservative in the sense of willing to err on the side of protecting too much than protecting too little”. His principle applied here would mean opting for the highest values of MVP and the lowest values of carrying capacities. We do not mean to suggest that FOCALIZE solves everything. The effectiveness of the chorological type, as an extended surrogate, must be evaluated. However, FOCALIZE is indeed a very good starting point as it is based on well-known patterns and processes. Who has not noticed that animals move across ecosystems or that the landscape is integrated by units delineated by boundaries? Our positive results, though immensely useful, do not suppose that all connectivity problems are solved. Some contacts of the landscape continuum are ignored in order to split it into chorological types. The contacts that are missed can be found in Figures 3 and 4c, by comparing the cells selected in the scenario with the original contacts of ecosystems and of chorological types. We were pleased whenever the chorological types ended up complete in one cluster or, when split into many clusters, could be included in a conservation area. Even better would be to achieve more than one chorological type in a cluster, even if it creates more elongated shapes increasing the BL/IBL ratio. The importance of a round shape is a matter of scale. Special attention should be given to small patch features within a chorological type. The MTR obtained may be insufficient for the species restricted to small ecosystems. Avoiding small cell sizes might prevent this risk. In any case, it is essential to check closely the results. The chorological type concept did not solely become a novel extended land surrogate. It also turned out to be essential to calculate the MTR by providing the 60 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 LITERATURE CITED The supplementary character of naturalness and social variables is also an appealing characteristic of FOCALIZE. We saw the benefits evidenced by the low values of land in conflict. Regarding naturalness, in the present situation of the world, recovering ecosystems should be a core action in conservation planning. To start the selection in natural areas and continue through fragmented or transformed sites permits the achievement of the targets in any context, and provides information for directing the effort and allocating funds for protection as well as recovery. Ardron, J.A., H. Possingham and C.J. Klein (eds.). 2010. MARXAN Good Practices Handbook, Version 2. Pacific Marine Analysis and Research Association, Vancouver. Ball, I. and H. Possingham. 2000. MARXAN (v 1.8.2). Marine Reserve Design using Spatially Explicit Annealing. A Manual Prepared for the Great Barrier Reef Marine Park Authority. http:// www.ecology.uq.edu.au/index.html. Accessed July, 2007. Bedward, M., R.L. Pressey and D. Keith. 1992. A New Approach Restoration of ecosystems to achieve conservation targets has been questioned by some conservation planners arguing that “the target deficit should not be interpreted as accurate indicator of restoration needs. Previous transformation will have eliminated the portion of biodiversity in these broad habitat units that is unable to survive in heavily modified landscapes” (Pressey et al. 2003). That would be true if all remnants were gone a long time ago. But extinction is a process that implies time. Many species surviving today with enough variability to expand into viable numbers will face extinction if the habitat is insufficient – it is a matter of time (Tilman et al. 1994). for Selecting Fully Representative Reserve Networks: Addressing Efficiency, Reserve Design and Land Suitability with an Iterative Analysis. Biological Conservation 62:115-125. Bell, R.H.V. 1982. The Effect of Soil Nutrient Availability on Community Structure in African Ecosystems. In: B.J. Huntley and B.H. Walker (eds.) Ecology of Tropical Savannas. Ecological Studies 42. Springer-Verlag, Berlin. Pp. 193-216. Belovsky, G.E. 1987. Extinction Models and Mammalian Persistence. In: M.E. Soulé (ed.) Viable Populations for Conservation. Cambridge University Press, Cambridge. Pp. 35-57. Bie, S. de. 1991. Wildlife Resources of the West African Savanna. Wageningen Agricultural University Papers, Wageningen. Not having data for restoration cannot lead to ignoring its importance. Mapping the actual and the original-potential ecosystems (Fandiño-Lozano 1996) also made this aspect of the method possible and easy to execute. Braak, C.J.F. ter. 1987. CANOCO - A FORTRAN Program for Canonical Community Ordination by [Partial] [Detrended] [Canonical] Correspondence Analysis and Redundancy Analysis (version 2.1). TNO, Wageningen. To obtain FOCALIZE, contact the Research Group ARCO through our webpage: www.grupoarco.info/eng/products.htm. Brennan, A. 1996. Ethics, Ecology and Economics. In: N.S. Cooper and R.C.J. Carling (eds.) Ecologists and Ethical Judgements. Chapman & Hall, London. Pp. 13-26. ACKNOWLEDGEMENTS Brooks, T.M., M.I. Bakarr, T. Boucher, G.A.B. da Fonseca, C. We thank Dr. Ramya Rajagopalan for her helpful comments on the manuscript and the two referees for their constructive observations that improved the paper in several ways. Appreciation also goes to Javier González for his assistance editing this paper. Hilton-Taylor, J.M. Hoekstra, T. Moritz, S. Olivier, J. Parrish, R.L. Pressey, A.S.L. Rodriques, W. Sechrest, A. Stattersfield, W. Strahm, and S.N. Stuard. 2004. Coverage Provided by the Global Protected-Area System: Is It Enough? Bioscience 54:1081-1091. CDB. 2010. COP 10 Decision X/2. Strategic Plan for Biodiversity 2011-2020. http://www.cbd.int/decision/cop/?id=12268. Accessed May, 2011. CMS. 2012. Convention on Migratory Species. http://www.cms.int. Accessed January, 2012. 61 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 Dunning, J.B., M.J. Groom and H.R. Pulliam. 2006. Species and Jarman, P.J. and A.R.E. Sinclair. 1979. Feeding Strategy and the Pattern of Resource Partitioning in Ungulates. In: A.R.E. Sinclair and M. Norton-Griffiths (eds.) Serengeti. Dynamics of an Ecosystem. The University of Chicago Press, Chicago. Pp. 130-163. Landscape Approaches to Conservation. In: M.J. Groom, G.K., Meffe and C.R. Carroll (eds.) Principles of Conservation Biology. Sinauer Associates Inc. Sunderland. Pp. 419-465. Kirkpatrick, J. B. 1983. An Iterative Method for Establishing Priorities for the Selection of Nature Reserves. An Example from Tasmania. Biological Conservation 25:127-134. Fandiño-Lozano, M. 1996. Framework for Ecological Evaluation Oriented at the Establishment and Management of Protected Areas. A Case Study of the Santuario de Iguaque, Colombia. ITC Publication No. 45. ITC, Enschede. Klingebiel, A.A and P.H. Montgomery. 1961. Land Capability Classification. Agricultural Handbook 210. USDA. Soil Conservation Service, Washington D.C. Fandiño-Lozano, M. 2000. Propuesta de un nuevo sistema de categorías de áreas de conservación para Colombia. Departamento Nacional de Planeación, Bogotá. Küchler, A.W. and I.S. Zonneveld. 1988. Vegetation Mapping. Kluwer Academic Publishers, Dordrecht. Fandiño-Lozano, M. 2001. Ecological Evaluation for Conservation. A Way of Thought. In: D. van der Zee and I.S. Zonneveld (eds.) Lambeck, R.J. 1997. Focal Species: A Multi-Species Umbrella for Nature Conservation. Conservation Biology 11:849-856. Landscape Ecology Applied in Land Evaluation, Development and Conservation. ITC, Enschede. Pp. 195-219. Lesslie, R.G., B.G. Mackey and K.M. Preece. 1988. A ComputerBased Method of Wilderness Evaluation. Environmental Conservation 15:225-232. Fandiño-Lozano, M. and W. van Wyngaarden. 2003. Rol de la Fauna en la Selección de Áreas de Conservación Biológica. In: R. Polanco-Ochoa (ed.) Manejo de la Fauna Silvestre en la Lombard, A.T., R.M. Cowling, R.L. Pressey and A.G Rebelo. 2003. Effectiveness of Land Classes as Surrogates for Species in Conservation Planning for the Cape Floristic Region. Biological Conservation 112:45-62. Amazonia y Latinoamérica. Fundación Natura, Bogotá. Pp. 205209. Fandiño-Lozano, M. and W. van Wyngaarden. 2005. Prioridades de Conservación Biológica para Colombia. Grupo ARCO, Bogotá. Margules, C.R. and R.L. Pressey. 2000. Systematic Conservation Planning. Nature 405:243-253. Fandiño-Lozano, M. and W. van Wyngaarden. 2009. Área de estudio y diagnóstico del tema forestal. In: Plan general de Murphy, D.D. and B.R. Noon. 1992. Integrating Scientific Methods with Habitat Conservation Planning. Reserve Design for Northern Spotted Owls. Ecological Applications 2:3-17. ordenación forestal del Huila. Grupo ARCO, Bogotá. Pp. 6-22. FAO. 1983. Guidelines: Land Evaluation for Rainfed Agriculture. FAO Soils Bulletin No. 52. FAO, Rome. Nicholls, A.O. and C.R. Margules. 1993. An Upgraded Reserve Selection Algorithm. Biological Conservation 64:165-169. Gehlbach, F.R. 1975. Investigation, Evaluation and Priority Ranking of Natural Areas. Biological Conservation 8:79-88. IGAC. 1995. Huila. Características Geográficas. Noss, R.F. 1983. A Regional Landscape Approach to Maintain Biodiversity. BioScience 33:700-706. Instituto Geográfico Agustín Codazzi, Bogotá. Noss, R.F. 1987. From Plant Communities to Landscapes in Conservation Inventories. A Look at the Nature Conservancy (USA). Biological Conservation 41:11-37. Ims, R.A. 1995. Movement Patterns Related to Spatial Structure. In: L. Hansson, L. Fahrig and G. Merriam (eds.) Mosaic Landscapes and Ecological Landscapes. Chapman and Hall, Noss, R.F. and A. Cooperrider. 1994. Saving Nature’s Legacy: Protecting and Restoring Biodiversity. Defenders of Wildlife and Island Press, Washington D.C. London. Pp. 85-109. Janzen, D.H. 1986. The Eternal External Threat. In: M.E. Soulé (ed.) Conservation Biology. The Science of Scarcity and Diversity. Noss, R.F., M.A. O´Connell and D.D. Murphy. 1997. The Science of Conservation Planning. Habitat Conservation Under the Sinauer Associates Inc., Sunderland. Pp. 287-303. 62 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 Endangered Species Act. Island Press, Washington D.C. Shrader-Frechette, K.S. 1987. The Real Risk of Risk-cost-benefit Analysis. In: P.T. Durbin (ed.) Technology and Responsibility. D. Reidel, Dordrecht. Pp. 343-357. Paine, R.T. 1995. A Conversation on Refining the Concept of Keystone Species. Conservation Biology 9:962-964. Shrader-Frechette, K.S. and E.D. McCoy. 1993. Method in Ecology. Strategies for Conservation. Cambridge University Press, Cambridge. Possingham, H,I. Ball and S. Andelman. 2000. Mathematical Methods for Identifying Representative Reserve Networks. In: S. Ferson and M. Burgman (eds.) Quantitative Methods for Conservation Biology. Springer-Verlag, New York. Pp. 291-205. Sinclair, A.R.E. and M. Norton-Griffiths (eds.). 1979. Serengeti. Dynamics of an Ecosystem. The University of Chicago Press, Chicago. Pressey, R.L. 1994. Ad Hoc Reservations: Forward or Backward Steps in Developing Representative Reserve Systems? Conservation Biology 8:662-668. Soulé, M.E. 1987. Introduction. In: M.E. Soulé (ed.) Viable Populations for Conservation. Cambridge University Press, Cambridge. Pp. 1-10. Pressey, R.L. 2002. The First Reserve Selection Algorithm – A Retrospective on Jamie Kirkpatrick´s 1983 paper. Progress in Physical Geography 26:434-441. Spellerberg, I.F. 1981. Ecological Evaluation for Conservation. Edward Arnold Ltd, London. Pressey, R.L. 2004. Conservation Planning and Biodiversity: Assembling the Best Data for the Job. Conservation Biology 18:1677-1681. Stoms, D.M., P.J. Comer, P.J. Crist and D.H. Grossman. 2005. Choosing surrogates for Biodiversity Conservation in Complex Planning Environments. Journal of Conservation Planning 1:26-39. Pressey, R.L. and K.H. Taffs. 2001. Scheduling Conservation Action in Production Landscapes: Priority Areas in Western New South Wales Defined by Irreplaceability and Vulnerability to Vegetation Loss. Biological Conservation 100:355-376. Sukumar, R. 1989. Ecology of the Asian Elephant in Southern India. I. Movement and Habitat Utilization Patterns. Journal of Tropical Ecology 5:1-18. Pressey, R.L., R.M. Cowling and M. Rouget. 2003. Formulating Conservation Targets for Biodiversity Pattern and Process in the Cape Floristic Region, South Africa. Biological Conservation 112:99-127. Svancara, L.K., R. Brannon, J.M. Scott, C.R. Groves, R.F. Noss and R.L. Pressey. 2005. Policy-driven versus Evidence-based Conservation: A Review of Political Targets and Biological Needs. BioScience 55:989-995. Rabinowitz, D. 1981. Seven Forms of Rarity. In: H. Synge (ed.) The Biological Aspects of Rare Plant Conservation. John Wiley & Sons Ltd., London. Pp. 205-217. Tans, W. 1974. Priority Ranking of Biotic Natural Areas. Michigan Botanist 13:31-39. Tilman, D., R.M. May, C.L. Lehman and M.A. Nowak. 1994. Habitat Destruction and the Extinction Debt. Nature 371:65-66. Ratcliffe, D.A. 1971. Criteria for the Selection of Natural Reserves. Advancement of Science 27:294-296. Tjallingii, S.P. 1974. Unity and Diversity in Landscape. Some Remarks on the Use of Soils and Vegetation Maps in Landscape Ecological Studies for Nature Conservation and Physical Planning in the Kromme Rijn Area, The Netherlands. Landscape Planning 1:7-34. Reyers, B., D.H.K. Fairbanks, K.J. Wessels and A.S. van Jaarsveld. 2002. A Multicriteria Approach to Reserve Selection: Addressing Long-term Biodiversity Maintenance. Biodiversity and Conservation 11:769-793. Tubbs, C.R. and J.W. Blackwood. 1971. Ecological Evaluation of Land for Planning Purposes. Biological Conservation 3:169-172. Schonewald-Cox, C. and J.W. Bayless. 1986. The Boundary Model: A Geographical Analysis of Design and Conservation of Nature Reserves. Biological Conservation 38:305-322. Wessel, K.J., B. Reyers and A.S. van Jaarsveld. 2000. Incorporating Land Cover Information into Regional Biodiversity Assessments in South Africa. Animal Conservation 3:67-79. Shrader-Frechette, K.S. 1984. Science Policy, Ethics and Economic Methodology. D. Reidel, Dordrecht. 63 Fandiño-Lozano, Wyngaarden / Journal of Conservation Planning Vol 8 (2012) 45 - 64 Wiens, J.A. 1992. Ecological Flows Across Landscape Boundaries: A Conceptual Overview. In: Hansen, A.J. and F. Di Castri (eds.) Landscape Boundaries, Consequences for Biotic Diversity and Ecological Flows. Springer, New York. Pp. 217-235. Wyngaarden, W. van. 1985. Elephants, Trees, Grass and Grazers. Relationships Between Climate, Soils, Vegetation and Large Herbivores in a Semi-arid Savannah Ecosystem (Tsavo, Kenya). ITC Publication No. 4. ITC, Enschede. Wilcox, B.A. 1984. In Situ Conservation of Genetic Resources: Determinants of Minimum Area Requirements. In: J.A. McNeely and K.R. Miller (eds.) National Parks, Conservation and Development. Proc. World Congress on National Parks. Washington D.C. Pp. 18-30. Wyngaarden, W. van and M. Fandiño-Lozano. 2005. Mapping the Actual and Original Distribution of the Ecosystems and the Chorological Types for Conservation Planning in Colombia. Diversity and Distributions 11:461-473. Wyngaarden, W. van and M. Fandiño-Lozano. 2008. Fortalecimiento de la capacidad de la CAM en la planificación sistemática de la conservación y en la consolidación del sistema regional de áreas protegidas. Grupo ARCO, Bogotá. Wilcox, B.A. and D.D. Murphy. 1985. Conservation Strategy: The Effects of Fragmentation on Extinction. American Naturalist 125:879-887. Wittemyer, G., W.M. Getz, F. Vollrath and I. Douglas-Hamilton. 2007. Social Dominance, Seasonal Movements, and Spatial Segregation in African Elephants: A Contribution to Conservation Behavior. Behavioral Ecology and Sociobiology 61:1919-1931. Wyngaarden, W. van and M. Fandiño-Lozano. 2011. FOCALIZE 1. User manual (version 1-0-8). Grupo ARCO, Bogotá. Zonneveld, I.S. 1989. The Land Unit. A fundamental Concept in Landscape Ecology, and Its Applications. Landscape Ecology 3:67-86 Wright, D.F. 1977. A Site Evaluation Scheme for Use in the Assessment of Potential Nature Reserves. Biological Conservation 3:293-305. 52North. 2011. ILWIS 3.7. ILWIS Software Community. http://52north.org/communities/ilwis/. Accessed January, 2011. 64