EUMETSAT EUMETSAT Visiting Scientist cientist

Transcription

EUMETSAT EUMETSAT Visiting Scientist cientist
EUMETSAT Visiting Scientist Action
Investigation and formulation of criteria for the determination
of hazardous areas along WV black stripes
Mag. Thomas Krennert
Institute of Meteorology, University of Vienna
Acknowledgements
The study presented in this paper was carried out part time at the Austrian Central Institute for Meteorology and
Geodynamics between March 2006 and January 2007. The study was sponsored by EUMETSAT, as a part of the
SAFNWC Visiting Scientist Activity (VSA) program.
The author would like to thank EUMETSAT and ZAMG for making this VSA possible. Furthermore, I would like
to express my thankfulness to Dr. Veronika Zwatz-Meise, head of the synoptic department, who made this action
possible, and especially to Dr. Alexander Jann, head of the remote sensing branch, for his extended support and
supervision during my work.
Aim
In response to action IOPSG08-04, a "breakthrough performance level" for the "water vapour dark stripe" subproduct of PGE12 (Air Mass Analysis) was formulated as "objectively determined WV stripes match subjective
analysis; dark features clearly connected with convection-triggering are always represented in the product in
advance of the first formation of convection ". The possibility of scientifically further narrowing the marked regions
to those particularly endangered by developing convection was investigated.
In SAF/NWC/IOP/ZAMG/SCI/VAL/1, the problem of initiation along moisture gradients was described as: "[This
feature is not] by itself a trigger of convective activity. In fact, there is no doubt that in the majority of cases or over
most of the extension of such a phenomenon, no severe weather is found." The concept could be much more readily
applied if areas with a high probability for the formation of convection are highlighted.
The AMA dark stripe product was qualitatively evaluated in combination with other parameters to elaborate the
conditions for flagging a section of a dark stripe in a future release as "dangerous". Some of these parameters from
the NWC SAF have also been taken qualitatively into consideration. A comparison with corresponding products
from NWP models are showing reasonable results and are discussed thoroughly. The region of interest was the
greater Alpine region. The synoptic situation under discussion appears about twenty times a year over the Alps. Due
to a limitation of available NWP model data only for two years, a statistical evaluation seemed not reasonable.
Therefore a thorough listing of case studies is given in the appendix, showing the behaviour of significant
parameters under discussion.
The report allows suggestions on which basis the attribute "prone to convection" could be derived in order to
enhance the "dark stripe" product. Also a short outlook towards future research is given.
-2-
Contents
1. Introduction
5
2. The nature of Deep Moist Convection (DMC)
5
2.1 Atmospheric conditions for convection
6
2.2 Convection and Instability
7
2.3 Conditional versus potential instability
9
2.4 Inertial instability
10
2.5 Ingredients-based methodology
10
2.6 Entrainment
11
3. MSG Water Vapour channels
12
3.1 Interpretation of synoptic scale features in the WV imagery
4. Convection at satellite WV channel moisture
moisture gradients
13
14
4.1 Observations
14
4.2 Incoming solar radiation and diabatic heating
16
4.3 The vertical stratification of the air column
16
4.4 Dynamic initiation
18
4.5 Diagnosis with the Help of RGB imagery
21
4.6 Diagnosis with the help of the NWCSAF AMA product
25
5. Assessing Symmetric instability
28
5.1 Moist symmetric instabilities: Conditional and potential symmetric instability
29
5.2 The connection between the Mg–θes relation and moist geostrophic potential vorticity
30
5.3 Convective–symmetric instability
32
5.4 Slantwise convective available potential energy SCAPE
33
6. Applying the concept of MSI on WV boundary convection
34
7. Application of NWCSAF products and Global Instability Index
42
8. Conclusions and outlook
44
9. Case studies
45
10.
10. References
89
-3-
List of Symbols and Acronyms
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
AMA ………. Air Mass Analysis
CAPE ………. Convective available potential energy
CI ………. Conditional instability
cp ………. Specific heat of dry air at constant pressure
CSI ………. Conditional symmetric instability
DMC ………. Deep moist convection
EPV ………. Equivalent potential Vorticity, approximately equals MPV
f ………. Coriolis parameter
g ………. Gravity; subscript denoting geostrophic
GII .………. Global Instability Index
gr ………. Subscript denoting gradient
L ………. Latent heat of vaporization
LFC ………. Level of free convection
LFSC ………. Level of free slantwise convection
LPW ………. Layer Precipitable Water
M ………. Absolute momentum defined as υ + fx
Mg ………. Geostrophic absolute momentum defined as υg + fx
MPV ………. Moist potential vorticity defined as gη · gradθe
MPEF ………. Meteorological Products Extraction Facility
MPVg ………. Moist geostrophic potential vorticity defined as gηg · gradθe
MPVgs ………. Saturated geostrophic potential vorticity defined as gηg · gradθ*e
MSI ………. Moist symmetric instability
MSG ………. METEOSAT Second Generation satellite (METEOSAT 8)
NGM ………. Nested Grid Model
NWCSAF ………. Nowcasting Satellite Application Facility
NWP ………. Numerical Weather Prediction
PI ………. Potential instability (also called convective instability)
PSI ………. Potential symmetric instability
PV ………. Potential vorticity defined as gη · gradθ
PVg ………. Geostrophic potential vorticity defined as gηg · gradθ
PW ………. Precipitable Water
qυs ………. Saturation mixing ratio
RUC ………. Rapid Update Cycle
r ………. Distance along radius of curvature
SCAPE ………. Slantwise convective available potential energy
SAI ………. Stability Analysis Imagery
T ………. Temperature
TPW ………. Total Precipitable Water
WV ………. Water Vapour
υ ………. Component of wind in alongfront direction
υg ………. Component of geostrophic wind in alongfront direction
υgr ………. Component of gradient wind in alongfront direction
x ………. Cross-front coordinate
z ………. Height coordinate
Γd ………. Dry-adiabatic lapse rate
Γm ………. Moist-adiabatic lapse rate
ζg ………. Vertical component of geostrophic relative vorticity vector
η ………. Three-dimensional absolute vorticity vector
ηg ………. Three-dimensional geostrophic absolute vorticity vector
θ ………. Potential temperature
θe ………. Equivalent potential temperature
θes ………. Saturated equivalent potential temperature
θw ………. Wet-bulb potential temperature
grad ………. Gradient operator in x, y, and z coordinates
-4-
1. Introduction
Every year all over Europe, Thunderstorms cause damage, loss of property and even life. In literature the impact of
severe convective events is described by numerous articles, to mention some of the most recent ones: Doswell
(2003 and 2005) describes the social implications on social life and modern business. For operational national
weather services the ability to provide correct warnings to the public is of utmost importance. For this purpose exact
knowledge about convective processes has to be gained. This topic is still one of the most actively discussed in the
meteorological scientific community.
At present days with no significant advection of moisture or heat, also devoid of frontal systems, forecasting
convective storms is limited to the declaration that throughout the day convective cells of undefined strength will
form, initiating preferably over mountainous regions. This VSA has been undertaken in order to investigate whether
there is a preferred area for the onset of deep convection within this synoptic environment and if it can be connected
to moisture gradients at middle and upper tropospheric levels.
So, the topic under discussion is not whether WV gradients trigger DMC, but if the dynamics adherent to the WV
gradients are capable of modifying (additionally supporting) the onset of DMC.
Thus, in a fair weather situation with weak or moderate low level moisture supply (compared to other
typical synoptic situations prone to DMC) a lifted air parcel hardly reaches its level of free convection (LFC),
despite increased potential instability below WV dark zones and widespread conditional instability. In connection to
WV Boundaries, slantwise convection may contribute to further lifting of the parcel, making it more probable to
reach its LFC and lead to further thunderstorm development, even hail.
2. The nature of Deep
Deep Moist Convection (DMC)
Initiation of deep moist convection can be connected to a favourable thermodynamic environment, created by large
scale flows, and sufficient lift, usually provided by mesoscale processes (Doswell, 1987).
Thus, the single convective cell can appear in different synoptic environments (Krennert et al., 2001):
One type of convection is connected to fronts or convergence lines associated with upper level troughs or upper
level depressions. The intrusion of cold and dry air from higher levels of the troposphere is mostly seen as a dark
zone in the WV images behind a cold front. In combination with this process mesoscale upward motion results,
leading to rapid cyclonic development with possible convective activity at the rear and within frontal cloud bands.
A second type of convection appears at the leading edge of frontal cloud bands. Advection of cold and dry air at
higher levels overrunning a frontal cloud band can also be observed as a dark zone in the WV image. This
mechanism is responsible for severe storms for instance within prefrontal squall lines. All these mechanisms are
well indicated by typical parameters used in numerical weather prediction models (NWP) where the synoptic scale
environment for convection can also be easily determined (Doswell, 1987, Georgiev, 1999, Zwatz-Meise, et al.,
2002).
The type of deep moist convection which is under discussion in this report is clearly not associated with a frontal
cloud band and frontal dynamics. The typical synoptic environment is an isobaric ridge at 500 hPa and a ridge
including a synoptic boundary condition with overall “weak gradients” of usually used NWP parameters analyzing a
frontal cloud band. According to the absence of a frontal cloud band or any extended compact cloud area, this type
of convection is called “convection under fair weather conditions”.
As already mentioned a favoured area for the onset of deep moist convection is the transition zone between
relatively high and relatively low amounts of humidity above a level of 600 hPa, indicated by relatively high or low
pixel values within the WV image. These WV - “boundaries” may be part of a mesoscale or even synoptic scale
pattern in the WV image.
As WV – Boundaries are connected to dry air they could be mixed up with the “dry lines” which are often described
in literature as playing a role in initiation of DMC. They are described as “low level mesoscale boundary or
transition zone, hundreds of kilometres in lengths and up to tenth of kilometres in width, separating dry from moist
air” (Glossary American Met. Soc., p241).
-5-
Consequently both phenomena have to be distinguished: the “concept” of WV – Boundaries is related to middle and
upper tropospheric layers and the “dry lines” are related to the lower troposphere.
In general, physical concepts behind synoptic - and/or mesoscale cloud configurations are described using
hydrodynamic, quasigeostrophic theory with help of parameters such as temperature advection, vorticity advection
and many others. For small scale cloud features, like convective cells, the hydrostatic point of view becomes more
important. Large scale areas often contain synoptic conditions suitable for the development of convection; however,
convective features are of a much smaller scale. The concept of a buoyant air parcel is best understood using
hydrostatic information describing the state of the air column above.
2.1 Atmospheric conditions for convection
•
Convection in general is mass motion within a fluid resulting in transport and turbulent mixing of the
properties of that fluid by vertical movement, in matters of transport of energy (heat), water vapour and
momentum. As such, it is one of the three main processes by which energy (heat) can be transported
vertically: radiation, conduction, and convection (AMS Glossary, 2000).
Meteorologists typically use the term convection to refer to heat transport by the vertical component of the flow
associated with an updraft. Convection is best explained by the so-called lifted parcel theory. Here, the environment
of the parcel is not affected by this displacement, and the upward force affecting such a rising plume (i.e. parcel) of
air is buoyancy, which is defined by
The difference in density (left term) or in virtual temperature (right term) between the rising parcel (p) and the
parcel environment (e) causes the updraft. Consequently, a parcel with greater temperature and less density will be
controlled by buoyancy and displaced vertically. The virtual temperature is conservative in association with the
moisture within the parcel. Vertically integrated, the right term in the buoyancy formula gives the amount of
available convective energy CAPE. It is clear that the vertical displacement and acceleration of an air parcel is
directly associated to buoyancy. Atmospheric convection is nearly always turbulent:
•
•
Dry thermals, which do not reach their level of saturation. Mostly in the boundary layer convection is dry
with relative humidity less than 100%.
Buoyant plumes of moist warm air, reaching the saturation level and forming visible cumuliform clouds.
Depending on the conditions of the atmosphere they might grow into severe convective weather events
(thunderstorms resulting in large hail, damaging wind gusts, tornadoes, and heavy rainfall) which are
generally the result of released energy by phase changes of water.
During the condensation, latent heat energy is released and additionally contributes to buoyancy. Most of the energy
is expended against gravity, but the remaining portion of available energy is responsible for the severity of the
convective weather event. The vertical transport of moisture through convection above levels of 600 hPa can be
seen in the WV image. The term convection, commonly associated with this type, is henceforth referred to as deep
moist convection (DMC). Shallow convection with less vertical extent can be observed earlier in the VIS and the IR
channel.
-6-
Two main types of convection can be distinguished:
1. Free convection (also called gravitational or buoyant convection): motions that are predominantly vertical
and driven by buoyancy forces arising from hydrostatic instability, with locally significant deviations from
hydrostatic equilibrium. Free convection can mostly be related to diabatic heating through insolation (short
wave radiation) on surfaces with higher heat capacity than the neighbouring environment or thermal forcing
through greater diabatic heating at elevated surfaces.
2. Forced convection: motion induced by mechanical forces such as deflection by a large scale surface
irregularity, turbulent flow caused by friction at the boundary of a fluid, or motion caused by any applied
pressure gradient. Furthermore this type of convection is caused by lifting owing to dynamical convergence,
lifting through orography, lifting within the lee circulation of a mountain ridge or gravity waves.
The unstable stratification of the atmosphere is the cause of deep moist convection, whatever the reason may be for
convective initiation or the type of appearance of convective cells. Therefore the different types of instability are
now considered further:
2.2 Convection and Instability
The origin of instability is heat, both latent and sensible, that is produced as a result of solar heating and evaporation
of water vapour (also due to solar heating) in the lower troposphere. Convection transports the excess sensible heat
and water vapour from low levels into the upper troposphere, and transports cooler, dry air downward, thereby
reducing the instability. Thus the instability is initiating DMC, but convection will only continue until the instability
is removed.
Hydrostatic stability is a state variable for the stratification of the atmosphere or at least of a column of air. It can be
indicated by various variables like temperature, humidity or density. The change, usually a decrease of those
parameters, is shown in the lapse rate. In the hydrostatic approach the process lapse rate is distinguished from the
environmental lapse rate of the lifted air parcel. For parcel theory, the term "environment" means the immediate
vicinity to the parcel. This environment is not affected by the processes applied to the parcel. The dry adiabatic
lapse rate of an air parcel is the rate of decrease of temperature with height of a parcel of dry air which is lifted
through the atmosphere in hydrostatic equilibrium. That means if a parcel of air is vertically displaced without
saturation its temperature will change with the amount of Γ
During the process of lifting an air parcel will cool and may reach the point of saturation. If the parcel is still
buoyant and ascending, its temperature will decrease moist adiabatically. The moist adiabatic lapse rate is smaller
because of the release of latent heat through condensation:
where g is the acceleration of gravity, cp is specific heat, Lv is latent heat of air at constant pressure, R the gas
constant, rv is the mixing ratio of water vapour, T the temperature and ε the ratio of the two constants Rd/Rw.
-7-
The process lapse rate is
.
The relation between γ and Γ/Γm describes the stability:
Unsaturated
Saturated
γ>Γ
unstable γ > Γm
γ=Γ
neutral γ = Γm
γ<Γ
stable γ < Γm
The temperature T is dependent on changes of pressure and moisture during the lifting process. Therefore the
potential temperature Θ is defined, which is unchanged with adiabatic changes of pressure and thus conservative
when associated with the vertical pressure gradient:
According to adiabatic processes the equivalent potential temperature Θe is conservative associated to the vertical
pressure gradient under saturation:
Where L is Latent Heat, q is specific humidity and cp is the heat capacity at constant pressure.
According to the formula above, it can be seen that Θ increases with height in an air column within hydrodynamic
equilibrium and also Θe increases with height on a slightly different rate. The stability of the vertical stratification
can be directly derived from the vertical gradient of Θ and Θe. A decrease of Θe then causes vertical acceleration of
the air parcel due to instability.
Thus, an air parcel is regarded as absolutely stable if
γ < Γm
,
then, an air parcel has always the same temperature or is colder as its environment. Therefore there will be no lifting
force on that parcel.
An air parcel is regarded as absolutely unstable if
γ > Γm
,
-8-
This is the state of a column of air when the lapse rate of temperature is greater than the dry-adiabatic lapse rate.
The temperature of the air parcel is warmer than its environment, thus the parcel will be buoyant and will rise
further. An air parcel vertically displaced is accelerated in the direction of the displacement and the kinetic energy
consequently grows with the increasing distance from its level of origin.
An air parcel is regarded as potentially (convectively, thermally) unstable if
This condition represents the state of a humid but unsaturated parcel of air with an equivalent potential temperature
that decreases with height. That means that air parcel motion is stable with respect to unsaturated vertical
displacements and unstable with respect to saturated vertical displacements. The lapse rate in this layer steepens to a
rate greater than the pseudo-moist adiabatic lapse rate (pseudo-moist is moist adiabatic without its liquid water
content). If such a column is lifted until completely saturated, it will become unstable regardless of its initial
stratification. Its lapse rate will then exceed the moist adiabatic lapse rate.
The decrease of potential temperature with height as such is a sign for absolute instability. A column of air is
potentially unstable when the equivalent potential temperature is decreasing even if there is an increase of potential
temperature. But potential instability as such is not necessarily the cause of strong convective development and
heavy thunderstorm activity. The probability for severe weather and storms rises if the air parcel reaches saturation
and additional latent heat is released supporting the lift.
This situation is then called: Conditionally unstable
Γ
>
γ
>
Γm
valid for a saturated state
The atmosphere is said to be conditionally unstable where the environmental lapse rate is greater than the pseudomoist adiabatic lapse rate, but less than the dry - adiabatic lapse rate. The condition for instability when Γ > γ > Γm
is that the air parcel is saturated. When the environmental lapse rate is greater than Γ, an air parcel is unstable
whether or not it is saturated. If the environmental lapse rate is less than pseudomoist adiabatic lapse rate, the
atmosphere is conditionally stable. If γ = Γm, the atmosphere is neutral with respect to saturated vertical
displacements.
According to the saturation, the equivalent potential temperature Θe has to be replaced by Θes which is the saturation
equivalent potential temperature. It is the equivalent potential temperature an air parcel at the same pressure and
temperature would have when it is saturated.
2.3 Conditional
Conditional versus potential instability
At this point it should be mentioned, that an upward decrease of Θe (potentially unstable) implies conditional
instability, too, but if there is no saturation the layer of air might remain potentially unstable. Since Θe is not a state
variable for unsaturated air unless completely saturated, the air can be potentially unstable, neutral or quite stable,
when Θe decreases upward. When the parcel is lifted until saturation, the decrease of Θe is changed into a decrease
of Θes. When the air is not completely saturated, Θes is not conserved. But when the air is saturated and Θes = Θe,
both are conserved. At this point during the lifting process the air mass becomes more unstable due to the additional
release of latent heat: strong convective developments are highly probable. In other words: only if the vertical
displacement is sufficient enough and a potentially unstable layer or parcel is lifted to saturation it will become
conditionally unstable.
-9-
Normand (1938) understood that not all conditionally unstable atmospheres lead to unsettled weather. Because
moisture is not accounted for in assessing conditional instability (i.e., θes, the saturated equivalent potential
temperature, is a function of temperature and pressure only, not of humidity), some measure of the moisture profile
is needed to refine the classification of stability. Thus, the concept of available energy was introduced. So
conditional instability was subdivided into additional classifications based on what we now term convective
inhibition (CIN) and convective available potential energy (CAPE). Whereas initially the relative sizes of CIN and
CAPE were compared in order to determine stability, modern forecasters tend to consider each separately.
However, the layer-lifting process of potential instability is not typically associated with the development of
isolated upright deep moist convection. If it were, layer lifting initially would produce stable stratiform clouds,
which would then develop into deep moist convection. Although this process does not appear to be acting in
isolated convective storms, it does appear to occur in other circumstances. Schultz and Schumacher (1999), discuss
examples of so-called downscale convective–symmetric instability (Xu, 1986) in which the ascent occurring above
warm fronts is punctuated with isolated buoyant convective elements. It is important to recognize that the paths of
parcels in such situations are likely to undergo slantwise displacements to their lifting condensation levels (LCLs)
and LFCs before releasing the buoyant instability and becoming more upright; this process should be distinguished
from slantwise convection due to the release of conditional symmetric instability (CSI).
2.4 Inertial instability
The definition general of inertial instability is a instability in which the only form of energy transferred between the
steady state and the disturbance is kinetic energy (AMS Glossary, 2000, also called Helmholtz instability or
barotropic instability). Furthermore it is called a hydrodynamic instability arising in a rotating fluid mass when the
velocity distribution is such that the kinetic energy of a disturbance grows at the expense of kinetic energy of the
rotation. For a small plane-symmetric displacement (wave number zero) using the parcel method, this criterion for
instability is, that the centrifugal force on the displaced parcels is larger than the centrifugal force acting on the
environment. On the assumption that absolute angular momentum is conserved, this states that the fluid is unstable
if absolute angular momentum decreases outward from the axis; If this criterion is applied to rotation of the
Westerlies about the earth's axis, the angular speed of the earth is so large that the inequality fails and the
disturbance is usually stable during the basic state.
2.5 IngredientsIngredients-based methodology
Doswell (1987), Johns and Doswell (1992) describe an ingredients-based methodology which states that three
ingredients (lift, instability, and moisture) are required for deep, moist convection. The three ingredients were
chosen to imply the presence of CAPE via conditional instability and moisture, and to realize that convective
potential via the ascent of parcels to their LFCs. Clearly, if a sounding has no layer of conditional instability, deep,
moist convection is precluded (CAPE is zero if lapse rates do not exceed moist adiabatic somewhere in the
environmental sounding). The three ingredients are the necessary conditions for the initiation of deep, moist
convection. If CAPE is present and the lift is enough to attain an LFC, then these conditions also become sufficient
for deep, moist convection. The advantage of the ingredients-based methodology is that the lapse rates and moisture
are considered independently. Thus, forecasters anticipating the atmospheric changes allowing deep, moist
convection to develop can more easily visualize the destabilizing influences of lapse rate and moisture separately,
rather than trying to visualize the processes changing CAPE or potential instability.
- 10 -
2.6 Entrainment
Most convective clouds are driven by positive buoyancy, with virtual temperature greater than the environment, but
clouds with precipitation, evaporation, and/or melting can produce negatively buoyant convection. If there is not
sufficient moisture in lower levels, low moisture content in upper levels will cause too much evaporation and
therefore negative buoyancy. Evaporation is also caused through turbulent entrainment of dry air into a
Cumulonimbus cloud. That means if there is insufficient moisture in the lower levels to be lifted to saturation a
growing CB might "dry out" through entrainment. Here evaporation leads to a cooling of the rising air parcel and
thus to negative (downward) buoyancy and consequently stops or hinders DMC. The role of entrainment regarding
two different moisture regimes in the vicinity of a WV boundary has to be subject of thorough future research.
- 11 -
3. MSG Water Vapour channels
The METEOSAT Second Generation satellite operates by means of the Spinning Enhanced Visible and Infra Red
Imager (SEVIRI). It provides twelve different channels, among them two infrared channels centered at 6.2
micrometer wavelength (WV5 – channel) and at 7.3 micrometer (WV6 – channel).
In this way an investigation of the differential vertical and horizontal moisture distribution (qualitatively) becomes
possible. The MSG WV channels 5 and 6 display the humidity content in two different layers. Channel 5 has a
maximum absorption at a wavelength around 6.2 µm, with the maximum signal being received from around 350
hPa. The WV channel 6 has maximum absorption at 7.3 µm, with a maximum signal from around 500 hPa. See
schematic Fig.1.
Fig. 1: Weighting function of the 6,2 µm band (Channel 5) and the 7,3 µm band (Channel 6), respectively,
representing the height of the maximum signal for both WV channels.
Santurette and Georgiev (2005) describe the sensitivity range of the two channels, which provides an indication to
detect differences in the water content within any atmospheric layer at a certain altitude.
Table 1: Sensitivity range
WV Channel
Overall sensitivity
Layer of sufficiently large sensitivity range
Level of largest sensitivity range
Lower threshold of sensitivity
6.2
large
200 – 600 hPa
~ 400 hPa
~ 700 hPa
7.3
Medium
450 – 750 hPa
~ 600 hPa
~ 950 hPa
The authors of the study mentioned above suggest, that, since the radiation in the 6.2 µm band is highly absorbed by
water vapour content in mid- and upper troposphere, the imagery can be applied for synoptic air mass analysis as
well as for assessment of potential vorticity.
The WV6 – channel is more sensitive to lower moisture and therefore usable to compare NWP model low level
humidity fields.
- 12 -
3.1 Interpretation of synoptic scale features in the WV imagery
The basic principles are well described in Santurette and Georgiev (2005), Zwatz-Meise et al. (2005) and Weldon et
al. (1991).
Bright and dark features, respectively, can be connected to a physical state or process within the upper half of the
troposphere:
• Light and dark areas of image grey shades are associated with moist and dry air, respectively
• Lightening (darkening) of a feature is a sign of vertical ascent (descent)
• The synoptic scale boundaries between light/moist and dark/dry regions can be related to significant upper
flow features
The brighter WV imagery patterns can also be associated to mid – and upper- level dynamic processes like
ascending motions, regions of low values of vorticity and potential vorticity, higher dynamical tropopause. The
formation of a leaf like white pattern may be a precursor of surface cyclogenesis or intense convective weather.
In contrast, the darker patterns may be associated with descending/ drying motions, regions of higher values of
Vorticity and potential Vorticity, latent tropopause anomalies and low heights of the dynamic tropopause, “dry
intrusion” regions with significant anomalies of potential Vorticity associated with rapid cyclogenic development
(Browning, 1993)
The sharp boundaries between elongated synoptic scale light and dark features can be associated to regions of
maximum upper level winds along its moist / bright side.
The overlay of model data fields over WV images provide a proper insight over upper level dynamic processes, also
shortcomings of the NWP models simulating the upper level circulation can be shown properly this way. Owing to
the widespread availability of literature about this topic (SATMANU, Santurette, 2005) the operational use of WV
imagery and its derived concepts is common state of the art among forecasters.
Also the use of upper tropospheric satellite data in NWP model assimilation as well as in model quality evaluation
becomes more intensive these days.
- 13 -
4. Convection at satellite WV channel moisture gradients
4.1 Observations
Krennert and Zwatz-Meise, (2003) investigated more than 50 “fair weather” – cases (1999 – 2001) over central
Europe developing more than 700 cells have been investigated in respect to their first appearance in the WV image.
Before they are seen in the WV image shallow convection occurs over mountainous terrain (above 400 meters
MSL) in the visible and infra red channels. The results of the investigation can be summarised as follows:
• The first appearance of deep convection happens in 66% of all cases at the WV - Boundaries.
• Convective cells within a dry or humid area of the WV structures generally appear somewhat later in time.
• About 22% of the cells appear first at weaker boundaries within a more humid area.
• About 12% of the cells appear first within a dry area.
• Nearly all cells seen first within a humid or dry area grow above mountainous terrain higher than 400m MSL
which might indicate orographic triggering.
• 24 % of all cells appearing first at a WV – Boundary are not connected to orography.
• There is no indication for the exact location where DMC appears at the WV – Boundary.
These observations may lead to the conclusion that development of deep convection out of widespread shallow
convection is influenced and modified by a certain physical environment at the WV – Boundaries.
As always with small scale convective phenomena, a lack of observations as well as the uncertainty of numerical
models has to be taken into account.
Further investigations made for about 40 cases between 2004 and 2007 show the same behaviour. From now on a
representative case will be shown to display parameter patterns under discussion.
The case of 28 July 2005 seems to be the ideal case for the onset of DMC at WV boundaries, because of the distinct
WV feature extending from Italy to the Ukraine, and also of the absence of significant frontal circulation over
central Europe.
Throughout the next chapters of this report the discussion will mostly consider this ideal case study. All other
investigated cases show the same parameter behaviour, only slightly different in magnitude and configuration
The area of interest is the anticyclonic region over the Alps and the immediate surroundings of Austria,
respectively.
The following three hourly sequences of images show the onset of DMC over Austria, beginning at 0600 UTC. On
the left side the MSG high resolution makes it possible to detect even small cumulus clouds over the Alps and its
surroundings. On the right side only DMC reaching above the height of 500 hPa is shown as white spots in relation
to the WV gradients. Also the motion and the deformation of the WV dark stripe over Austria can be seen properly.
The two convective cells can be clearly connected to the WV boundary.
A
B
- 14 -
C
D
E
F
G
H
Fig. 2, sequence A-H:
Left: MSG HRVIS channel 12, 28 July 2005, at 0600, 0900, 1200 and 1500 UTC, Austria indicated by red borderline;
Right: MSG WV channel 6.2µ, same times as left side
According to Doswell (1987) DMC depends on sufficient available moisture, conditional instability and a source of
lift (see also chapter 2). Sufficient lift is needed to move an air parcel to its level of free convection LFC, and with
available conditional instability initiating the deep convective process.
The following processes, being necessary conditions for convection shall be inspected in relation to their existence
at the WV - Boundaries.
- 15 -
4.2.
4.2. Incoming solar radiation and diabatic heating
One source of initiation of convection is heating of the surface and adjacent air layers by incoming solar radiation
during the daytime. This heating process causes an absolutely unstable shallow layer near the surface where
spontaneous convective overturning becomes likely (Doswell, 1987).
If enough lifting energy is provided the rising air parcel might reach condensation leading to strong deep moist
convection at the level of free convection in case of conditional instability. If the lifted parcel starts from higher
surfaces, e.g. mountain slopes, less energy is needed to reach the level of free condensation. This implies that
shallow convection will start earlier from elevated surfaces (Szoke, et al., 1984, Pielke, et al. 1986, Bluestein, 1993)
Incoming solar radiation is differently affected in the area with dry or wet upper level humidity, although the visible
part of the solar radiation (also known as “atmospheric window”) is not weakened remarkably by water vapour.
However there is considerable depletion in the near – IR band. According to Liou (2002, p87), water vapour is the
primary absorber in the near – IR, which contains about 50% of the incoming solar energy.
Consequently the reduction of solar energy reaching the surface might be higher below a humid than below the dry
area in the WV image.
Consequently, due to the differential energy of incoming solar radiation below dry and moist upper level layer,
decreased diabatic heating at surface levels below a humid region and increased heating below a dry region in the
water vapour should be expected.
Seen this mechanism alone and excluding other processes (e.g. advection), a certain fingerprint regarding the
surface temperature or the surface potential temperature (more convenient in mountainous regions) shall be seen.
However, even though any horizontal circulation during all of the investigated cases is very weak, no specific
pattern of surface temperature or surface potential temperature was measured.
Thus, it might be concluded, that the structure of the heated underground and its height (mountains) are more
important to super adiabatic overturning than the differential absorption of incoming solar radiation. This subject
has to be also addressed in future research.
4.3 The vertical stratification of the air column
A basic item necessary for the understanding of the physical process of deep convection is the vertical stratification,
especially in association with the vertical decrease of humidity.
A thorough investigation of all cases with the convection type under discussion has shown that the area under
consideration is widespread conditionally unstable in the low and middle troposphere. Near the surface a shallow
layer of absolute instability can be found in nearly all cases. Often the high reaching conditional instability in the
layer above is interrupted by very thin stable layers or weak inversions.
This shall be demonstrated with the example from the 28 July 2005 (compare Figures 2, A-H). As usual radio
soundings are not available at the exact location of deep convection.
Figure 3 (below) corresponds to Figure 2, E and F. Radiosonde stations are indicated with their WMO number. Near
to the station 11035 Vienna / Austria the onset of deep moist convection at a WV - Boundary can be seen at 1200
UTC. This station is closest to the development of the cell, both in distance and time. The vertical stratification
displayed here, is, in general, truly representative for all investigated cases.
- 16 -
Fig. 3: MSG WV5 image of the 28 July 2005 at 1200 UTC. WMO numbers indicate the location of radiosonde stations. The
station 11035 Vienna is situated at a WV - Boundary next to a developing cell in the northern part of Austria.
Figure 4 shows the stability analysis derived from the relevant lapse rates of the sounding. At ground levels the
stratification is absolutely unstable. Spontaneous convective overturning is highly possible within this layer. Above
that the air column is more or less conditionally unstable exceeding up to 350 hPa, except from interruptions by thin
inversions and stable layers.
Fig. 4: Stability analysis from the sounding of the 28 July 2005 at 1200 UTC at Vienna. Red: absolutely unstable (immediate
ground levels, super adiabatic overturning), yellow: conditionally unstable (up to 400 hPa, interrupted), green: inversion (thin
layer around 900 hPa), blue: stable
In Figure 5 the vertical distribution of relative humidity is plotted. The distinct amount of moisture in half of the
troposphere and the distinct gradient between about 650 hPa and 500 hPa is a typical condition observed throughout
all cases of deep convection at a WV – Boundary.
- 17 -
Fig. 5: Stratification of the relative humidity (%) derived from the radio sounding of Vienna, Austria, from 28 July 2005 at
1200 UTC. Maxima at about 900 and 700 hPa, strong gradient between 650 and 500 hPa.
In addition to the evaluation of radio soundings model derived soundings from the ALADIN LAM model have been
inspected. In general model data show similar results as the radio soundings. Most of the model derived soundings
show a strong vertical humidity gradient within the middle tropospheric layers (as confirmed by radio soundings).
In areas with homogeneous distribution of humidity at low levels having inhomogeneous distributions of humidity
above (shown in the WV imagery), horizontal gradients in potential instability will develop.
Higher potential instability is gained in connection with a higher vertical gradient of humidity. This is true for the
“black” (dry) region in the WV image in comparison to the “white” (humid) area. Consequently between these two
regions there is a horizontal gradient in potential instability which might correspond to the WV – Boundary. It is
possible to increase potential instability, while having no conditional instability at all, though this is not the case
here.
4.4 Dynamic initiation
In general dynamic processes described in chapter 3.1 are also responsible for the onset of DMC. Furthermore,
upper level dryness in the WV image also indicates processes of clear air turbulence, such as deformation (Bader et
al., 1995, the rate of darkening is proposed as a turbulence predictor), but also numerous examples from literature
are given to describe boundaries of the black stripes as areas favourable for convection. Martin et al. (1999)
describe a certain configuration of black stripes called mushroom configuration, according to Bader et al. (1995) dry
air overrunning moist air at lower levels can be prone to the formation of mesoscale convective systems.
Krennert et al. (2001) define upper level moisture boundaries also as favourable area of DMC while significant
frontal zones (indicated by cloudiness in satellite images and NWP fields) are absent. Under these conditions
convective developments are referred to as air mass thunderstorms (The preferred area of convection is represented
by the axis of the wedge of potential temperature or within the axis of a thickness ridge, see Bader et al, or Smith et
al. 1985).
Martin et al. (1999) describe cyclonic circulation which is also indicated in the WV image through a change from
anticyclonic and stable environment to an approaching upper level Vorticity maximum. Krennert et al. (2001)
identifies processes within a much smaller scale and also mostly within generally anticyclonic regimes. They
summarise that convection under “fair weather” shows a distinct diurnal cycle of development and decay, the cells
of DMC are initiated in the lower levels of the troposphere and deep convection develops and decays much faster
than the WV structures they are embedded in.
- 18 -
The following images give an overview of the synoptic situation of the case 28 July 2005, the relevant parameters
are derived from the LAM NWP model ALADIN, which is operationally available at ZAMG. The model grid point
resolution is about 16 km.
Fig6. :MSG WV5, 28 July 2005, 1200 UTC, Absolute topography at 300hPa
Figure 6 above shows a representing distribution of the 300 hPa heights of the ALADIN LAM model in
combination with the WV 6.2µ image (WV channel 5). A ridge with its axis from Corsica to Hungary can clearly be
seen, no frontal cloud bands are located within the immediate vicinity over the Alpine region. Figure 7 below shows
the same synoptic condition with the 500 hPa heights.
Fig. 7:MSG WV5, 28 July 2005, 1200 UTC, Absolute topography at 500hPa
The following Figure 8 shows the ability of the LAM model, resolving the humidity distribution at 300 hPa quite
well. Thus, a certain accuracy of the numerical values for this investigation can be assumed.
Fig. 8: MSG WV5, 28 July 2005, 1200 UTC, Relative Humidity at 300hPa
- 19 -
In Figure 9 the distribution of the equivalent potential temperature in the LAM model indicates a cold (and dry)
intrusion along with the black stripe in the WV image (see also Browning, 1993; Hoskins, 1985).
The WV boundary can be considered as a zone with higher gradients of equivalent potential temperature.
Fig. 9: MSG WV5, 28 July 2005, 1200 UTC, Equivalent potential temperature at 300hPa
Also according to Browning and Hoskins as well as Santurette and Georgiev (2005), the pattern of potential
vorticity (left) and relative vorticity fit the concept of tropopause intrusion.
Fig. 10: MSG WV5, 28 July 2005, 1200 UTC,
Left: Isopleths of 1,5 PVU at the given pressure level;
Right: relative Vorticity at 300 hPa, positive(cyclonic) values solid lines, negative (anticyclonic) values dashed lines
The behaviour of the zero line of shear vorticity at 300 hPa follows also Santurette et al. (2005). The rough
conclusion can be made that also at sub synoptic scale the dynamical concepts of the synoptic scale can be applied:
It has to be considered, that the zero line of shear Vorticity indicates a narrow zone of stronger vertical, and
assumed unidirectional increase of vertical wind shear. Zwatz-Meise, et al. (2006) and Santurette and Georgiev
(2005) connect this area with the axis of the jet streak. In connection to the WV boundaries under discussion it has
to be mentioned, that not only wind speed maxima but also minima can be indicated by this parameter.
- 20 -
Fig. 11: MSG WV5, 28 July 2005, 1200 UTC, zero line of shear vorticity at 300 hPa
4.5 Diagnosis with the Help of RGB imagery
RGB composites (RedGreenBlue) are frequently used by attributing 2 to 3 channels or channel combinations to
individual colour (RGB) beams. Red, Green, Blue: The three colours of light which can be mixed to produce any
other colour. Coloured images are often stored as a sequence of RGB triplets or as separate red, green and blue
overlays. These colours correspond to the three "guns" in a colour cathode ray tube and to the colour receptors in
the human eye. In this way it is possible to enhance specific properties of a physical process within the satellite
imagery. The IR channels of the MSG show DMC at an early stage (visible and infra red channels > 8µ) because of
its maximum absorption at lower tropospheric levels (see Fig. 12 below). Combining the properties of the MSG
channels 7 (8.7µ), 6 (7.3µ) and 5 (6.2µ) offers the possibility to observe the onset of DMC in direct relation to the
two MSG WV channels and classify the type of convection (shallow or deep) by addition of RGB colour intensities.
The usual grey shades of the IR channels are replaced by shades of the three colours each: IR7 by red, WV6 by
green and WV5 by blue. Thus, through shading and overlaying a completely new colour palette is produced, where
the moisture contents at different atmospheric layers is depicted.
Fig. 12: Weighting function of the MSG infra red channels, representing the height of the maximum signal for each channel.
The normalised weighting function for channel 7 is plotted as the blue line.
- 21 -
The following hourly sequence of images gives a clear impression of the onset of a deep moist convective cell over
the northern parts of Austria.
Blue colours indicate significant relative humidity at high levels, green colours indicate moist areas at mid levels
and reddish colours show cumulus clouds (already saturated) at low tropospheric levels. White colours indicate
saturated air parcels at high levels, dark areas indicate dry air at high and middle tropospheric levels. Significant
moisture at middle and high levels is represented by colours of bluish-green.
The stages of the convective cells developing along the WV boundaries over the eastern part of Austria can be
observed very well. Also the deep convective developments over the Czech Republic and Serbia, respectively, seem
to be connected to dynamic processes related to areas of transition between bluish and greenish colours in the
image.
A
B
C
D
E
F
Fig. 13, sequence A - F: RGB765, 28 July 2005, hourly 1000 to 1500 UTC, RGB combination image MSG IR channel 8.7µ
(red), WV channel 7.3µ ( green) and WV channel 6.2µ(blue).
- 22 -
Separating the pixel values of the three colours gives even more insight to the convective development under
discussion. In the cross sections below (Fig. 14, left side), the three different channels are represented by their
different colours. The cross sections are calculated every half hour starting at 1000 UTC on the 28 July 2005,
starting from left to right, indicated by the number of image pixels (horizontal axis). The vertical axis represents the
pixel value. Observing the first developing cell over the northern Parts of Austria (Fig. 14, A), various peaks of the
red curve (IR channel 7) indicate shallow convective development and shallow cumulus clouds, respectively.
The blue curve (WV channel 5) shows a distinct decline between pixel 24 and 85 with the minimum value around
55, indicating the dry air above 600 hPa and the associated WV boundary. The green curve also shows moisture
minimum, but slightly shifted and smoother than channel 5. In Fig. 14 B to F the further growth and development of
the convective cell towards deep moist convection is indicated by the increasing pixel values at the left flank of the
WV boundary, first in cannel IR7, followed by channel WV6 and WV5.
The cross section (images on the left side) is indicated by a red line in the images on the right side of the sequence.
The numbers represent pixels (abscissa on the left side, yellow numbers on the right side).
A
B
Fig. 14, sequence A - F: 28 July 2005, half-hourly 1000 to 1030 UTC;
Left: pixel value plots of the three channels IR channel 8.7µ (red), WV channel 7.4µ (green) and WV channel 6.2µ(blue).
Right: RGB765 composite, cross sections indicated by red line.
- 23 -
C
D
E
Fig. 14, sequence C - E: 28 July 2005, half-hourly 1100 to 1200 UTC;
Left: pixel value plots of the three channels IR channel 8.7µ (red), WV channel 7.4µ (green) and WV channel 6.2µ(blue).
Right: RGB765 composite, cross sections indicated by red line.
- 24 -
F
Fig. 14, sequence -F: 28 July 2005, 1230 UTC;
Left: pixel value plots of the three channels IR channel 8.7µ (red), WV channel 7.4µ (green) and WV channel 6.2µ(blue).
Right: RGB765 composite, cross sections indicated by red line.
4.6 Diagnosis of a WV boundary with the help of the NWCSAF AMA product
Jann (2002) describes a method of pattern recognition for the detection of ridge lines and stripes. This method is
capable of detecting automatically and objectively gradients within any two dimensional field. Thus, it can be
applied on the moisture gradients of the upper troposphere, which are indicated by pixel values in the image. This
way the positions of lines, stripes and filaments within the two MSG WV channels can be determined. A typical
member point of such a feature should have not more than two (out of eight) surrounding grid points, exhibiting a
more suitable value of the governing parameter (which is the brightness temperature in the present area).
This reasonable method has been applied on all available cases of WV boundary convection, but only the steepest
gradients are detected within an area set by the program thresholds. In most cases the gradients can be easily
detected by the human eye, however, a calculated position of the WV gradient seems arbitrary in most cases (i.e. the
exact location of a boundary is difficult to be diagnosed objectively on a flat transition zone, when the human eye
easily sees the boundary).
In the previous chapter 4.5 it was shown, that the WV channel 5 shows distinct gradients which are mostly analyzed
correctly. Whereas the WV channel 6 seems to have weaker gradients, therefore the exact numerical positioning of
a boundary line seems to be noisy.
Juxtaposition between the upper and lower boundary in order to derive an indication for instability becomes
unreasonable this way. It could not be concluded, that a certain position of the upper line compared to the lower
gradient line indicates the onset of DMC.
The method is highly valid for applications on bigger scales like synoptic systems and distinct air mass boundaries,
but no clear results can be retrieved from the analysis of meso – scale features and smaller, especially from WV6
imagery.
Further it has to be considered, that nearly all of the cases show distinct vertical moisture gradients anywhere
between 700 and 400 hPa (see radiosonde diagram Fig. 5). As Fig.1 shows the height of maximum signal of channel
6 around 500 hPa (well in between this vertical gradient zone) another source of uncertainty in the automatic
detection of WV boundaries occurs in connection to MSG channel 6. Also the moisture distribution in this channel
is influenced by temperature signals from below (snow cover from the Alpine ridge, reaching up to 700 hPa can be
seen in the 7.3µ channel).
- 25 -
The image sequence below shows the two WV channels with the automatic boundary detection. Only the major
gradients are analyzed, smaller features and moisture variations are not detected at the boundary. The detection
efficiency seems to be dependant on thresholds and general settings. However, the boundary zone consists of two
and more steps of intense transition between more moist and dryer filaments. Subjectively these multiple structure is
well observed; unfortunately the algorithm is constructed to detect only one gradient line. Therefore noisy
disturbances within the gradient line occur, especially on weak and flat gradients in WV6 (7.3µ).
A
B
- 26 -
C
Fig. 15, sequence A - C: 28 July 2005, 6 hourly from top to bottom, starting at 0600 UTC; Left: WV channel 7.4µwith
automatically detected boundary lines (red). Right: WV channel 6.2µwith boundary lines (blue).
A tuning of thresholds in the calculating program of the AMA products has also been performed within this VSA
investigation. The results did not show the desired accuracy of positioning the WV boundary more exactly in the
image, deviating the operational tuning of the program even resulted in a worse performance of the program.
Therefore these results are not discussed in this final report; the modification of the program and its thresholds has
to be subject of thorough investigations in the future.
- 27 -
5. Assessing Symmetric instability
The derivation of dry and moist symmetric instability is shown early by Helmholtz, Rayleigh, Kleinschmidt,
Sawyer, Eliassen and Kuo. The basics of the concept are presented by Holton (1992) and Bluestein (1993).
Derivations of MSI can also be found in the textbook presentations of Bluestein (1993), Houze (1993) and Emanuel
(1994).
Dry symmetric instability can be seen as a generalization of both inertial instability and dry gravitational instability.
The condition for inviscid, inertial instability is
∂Mg/∂x < 0, where Mg = υg + fx
is the geostrophic absolute momentum of a geostrophically balanced mean state, υg is the geostrophic wind in the
direction perpendicular to the temperature gradient (the direction along an elongated baroclinic zone), f is the
Coriolis parameter and x is the cross-front distance, increasing toward warmer air.
This condition is equivalent to the geostrophic absolute vorticity being negative.
Similarly, the condition for inviscid, dry gravitational instability (also referred to as dry absolute instability) is that
the hydrostatically balanced mean-state potential temperature θ decreases with height z
∂θ/∂z < 0.
Whereas a parcel may be inertially stable to horizontal (constant z) displacements
∂Mg/∂x > 0
and gravitationally stable to vertical displacements
∂θ/∂z > 0
separately, it may be unstable with respect to slantwise displacements by dry symmetric instability. The condition
for dry symmetric instability is that Mg surfaces slope less steeply than θ surfaces. Dry symmetric instability can be
viewed as either dry gravitational instability on a Mg surface or inertial instability on an isentropic surface.
Therefore, any slantwise displacement that occurs at an angle between the slopes of the Mg and θ surfaces will
release the symmetric instability.
In general, though, the atmosphere is not geostrophic and may, in fact, exhibit large departures from geostrophy in
regions susceptible to symmetric instability. In observational studies that assess symmetric instability from limited
observations such as single soundings, the geostrophic wind is often approximated by the total wind and stated
explicitly as such (Emanuel 1983b).
An additional complication occurs with mesoscale-model data in which the geostrophic wind (derived from the
geopotential height) is often much noisier than the total wind. It is common to approximate Mg by M
M = υ + fx,
where M is the absolute momentum calculated from the total alongfront wind υ, interchanging the total wind υ for
the geostrophic wind υg. The use of M instead of Mg is inconsistent with the theory for symmetric instability
discussed above. In areas with orography, large pressure changes, or intense frontogenesis, significant ageostrophy
in the observed lower-tropospheric wind can occur. In those cases, the geostrophic wind may be unidirectional but
the observed wind may have significant curvature (and not be nearly two-dimensional).
- 28 -
5.1 Moist symmetric instabilities
instabilities:
ies: Conditional symmetric instability and potential symmetric instability
Moist gravitational convection, conditional instability (CI) occurs locally at each height along a vertical sounding
where the environmental lapse rate lies between the moist- and dry-adiabatic lapse rates, or, equivalently, θes, the
saturation equivalent potential temperature decreases with height
∂θes/∂z < 0, with θes = θ exp(Lqvs/cpT),
where L is the latent heat of vaporization, qvs is the saturation mixing ratio, cp is the specific heat of dry air at
constant pressure, and T is the air temperature. The saturation equivalent potential temperature is the equivalent
potential temperature the air would have if it were saturated with water at the same pressure and temperature. When
the air is not saturated, θe is not conserved, but when the air is saturated, θe = θes and, hence, both are conserved.
Similarly, for slantwise convection, CSI occurs locally at each height where the environmental lapse rate along a Mg
surface lies between the moist- and dry-adiabatic lapse rates (is conditionally unstable along a Mg surface), or
∂θes/∂z|Mg < 0.
The instability is said to be conditional because saturation must be present locally in order for air to possess
parcel buoyancy.
As in moist gravitational convection where the potential instability (also known as convective instability) of a layer
along a vertical sounding can be defined
∂θe/∂z < 0,
it is possible to assess layer potential symmetric instability (PSI) along a Mg surface:
∂θe/∂z|Mg < 0.
The instability is said to be potential because the potentially unstable layer first must undergo a finite vertical
displacement to reach saturation and create the instability; release of the instability may then result, given sufficient
forcing for ascent. At saturation, the vertical gradients of θe, θes, and θw are equivalent, therefore, CI and PI, as well
as CSI and PSI, are equivalent.
Despite other examples in literature Schultz and Schumacher (1999), suggest that the term CSI should be used only
when employing θes and the term PSI only when employing θe.
Table 2. (below) summarizes the distinctions between moist and dry, as well as symmetric and gravitational
instabilities.
- 29 -
Table 2: Comparing gravitational, symmetric, and inertial instabilities, illustrating the parallelism in their definitions and
measures. Taken from Schultz and Schumacher (1999)
5.2 The connection between the Mg–θes relation and moist geostrophic potential vorticity
Certain assumptions are necessary in order to develop the Mg–θes relationship for the identification of CSI:
1) the geostrophic wind is constant in the direction along the elongated baroclinic zone,
2) the cross section for evaluating the Mg–θes relationship is perpendicular to the vertical shear of the geostrophic
wind (or, equivalently, the thermal wind or isotherms), and
It can be shown that extending the Mg– θe relationship for dry symmetric instability to three dimensions is
equivalent to computing geostrophic potential vorticity PVg (e.g. Hoskins, 1974; Mc Cann, 1995), where
PVg = gηg · gradθ,
g is gravity, ηg is the three-dimensional geostrophic absolute vorticity vector, and grad is the gradient operator in x,
y, and z coordinates. When PVg is negative (and inertial and dry gravitational instabilities are absent), dry
symmetric instability is present. Likewise, the three-dimensional form of the Mg–θes relationship for CSI is
equivalent to negative saturated geostrophic potential vorticity EPVs, also known as the saturated equivalent
geostrophic potential vorticity,
EPVg = gηg · gradθes.
when inertial and conditional instabilities are absent.
- 30 -
The schematic below (Fig. 16) displays the correlation between the Mg–θes relationship, the equivalent potential
Vorticity and the area of release of CSI.
Fig. 16: Relation between Mg–θes and EPVgs (MPV). Modified after Schultz and Schumacher (1999)
Therefore, assessing CSI using the three-dimensional form of EPVg does not require strict adherence to the same
assumptions as using the Mg–θes relationship in cross-section form. Owing to the potential confounds with assessing
Mg–θes relationships in cross sections, a more reliable assessment of CSI is obtained by employing EPVg or another
equivalent three-dimensional parameter.
The use of EPVg or another measure of CSI not dependent on cross-section orientation, e.g. SCAPE, seems to be
more convenient for a possible operational usage.
Finally, if both adequate moisture and lift are present in the absence of MSI, such that ascending air is forced to its
condensation level and beyond (forced convection), storm clouds may still form, with heavy precipitation being the
result (Doswell et al. 1998). Forced slantwise ascent leading to a single cloud/precipitation band can occur in the
absence of MSI over a mountain due to orographic lift or over a frontal zone due to secondary circulations
associated with frontogenesis, but this is not free convection. When EPVg is positive, multiple bands can be
generated externally only by pre-existing PVg or EPVg anomalies (Xu 1992). Observational documentation of these
features, however, has not occurred. Therefore, the absence of MSI does not preclude the formation of single-or
multiple-banded clouds and precipitation, much as the absence of potential or conditional instability to moist
gravitational convection does not preclude the same (Schultz and Schumacher, 1999).
Thus, like deep, moist gravitational convection, moist slantwise convection requires the simultaneous presence of
instability, moisture, and lift. If any one of these three is absent moist slantwise convection is prevented from
occurring. It is often observed in the atmosphere that regions of moist gravitational instability (CI or PI) may be in
the same vicinity as regions of MSI (CSI or PSI) (Emanuel 1983b; Snook 1992). In the warm sector, CI may exist,
whereas closer to the front, CSI, symmetric neutrality, and weak symmetric stability may be present. As discussed,
CI is a special case of CSI in which θes surfaces not only tilt more steeply than Mg surfaces, but are overturned, such
that
∂θes/∂z < 0.
Likewise, PI is a special case of PSI in which θe surfaces not only tilt more steeply than Mg surfaces, but are
overturned, such that
∂θe/∂z < 0.
As such, blindly employing the tests for CSI (EPVgs < 0 and the Mg–θes relationship) will identify regions of CI and
blindly applying the tests for PSI (EPVg < 0 and the Mg–θe relationship) will identify regions of PI.
- 31 -
5.3 Convective–
Convective–symmetric instability
The terms CSI and slantwise convection do not have interchangeable meanings.
For a deeper understanding of how convection (gravitational, slantwise, or both) organizes in the presence of both
CI/PI and CSI/PSI it shall be noted, that Xu and Clark (1985) argue for a continuum between gravitational and
slantwise convection, so, in a sense, the distinction that is drawn between gravitational and slantwise convection can
be considered arbitrary. As further noted by Jones and Thorpe (1992), “the strong distinction which is often made
between flows with positive and negative potential vorticity is an artefact of the use of balanced equations, rather
than a physical property of atmospheric flow.”
An initially gravitationally and symmetrically stable baroclinic atmosphere is destabilized by, for example, surface
heating or increasing the vertical shear of the geostrophic wind, CSI/PSI will arise before CI/PI (Emanuel 1994), but
owing to the larger growth rate and energy release of moist gravitational convection compared to slantwise
convection, gravitational convection, if initiated, is likely to dominate in time. Emanuel (1980), Jascourt et al.
(1988) describe a situation where CI/PI and CSI/PSI coexist convective–symmetric instability. Therefore, the
question arises as to the mesoscale circulations in the atmosphere to organize any resulting convection in such an
environment.
Xu (1986a) proposes two mechanisms for rainband development, mechanisms we now recognize as forms of
convective–symmetric instability. The first he refers to as “upscale development,” where small-scale moist
gravitational convection develops first, followed by mesoscale banded organization of clouds due to the release of
symmetric instability as the environment becomes gravitationally stabilized. It seems that this type of development
would be most likely to occur outside of frontal regions where small-scale moist gravitational convection organizes
in the absence of synoptic-scale air mass boundaries.
A likely observational example of upscale development in a situation of convective–symmetric instability is
documented by Jascourt et al. (1988). From a region of scattered cumulus over northern Louisiana, five parallel
cloud bands simultaneously grew to become lines of thunderstorms. The bands were aligned along the 700–500-mb
shear, a layer in which the moist symmetric stability was especially weak. The vertical stratification in the lower
troposphere, however, was conditionally unstable to gravitational convection with CAPE of more than 1000 J kg−1.
Jascourt et al. (1988) hypothesize that the initial latent-heat release by the scattered cumulus in the layer of weak
symmetric stability favoured the development of convective–symmetric instability and organized the convection
into the five bands. This case suggests that the nature and organization of convection can be modulated by the
symmetric stability.
These findings are the base for the further assumption, that also during the coexistence of moist gravitational
instabilities as a synoptic boundary condition and MSI in connection to WV gradients, a modification (triggering)
towards an onset of deep moist convection can be hypothesized.
Further, Schultz and Schumacher (1999) state that, given the typical inhomogeneous condition of the atmosphere,
areas favourable for MSI may be juxtaposed with areas favourable for gravitational convection such that convection
can possess characteristics of slantwise convection, gravitational convection, or both, and may evolve from one
form to the other. Consequently, convection can manifest itself in a variety of ways, depending on the
environmental stability, lift mechanism, amount of moisture, and other factors.
The coexistence of CSI/PSI and CI/PI, as well as adequate moisture and lift, may result in a mixture of moist
gravitational and moist slantwise convection associated with the release of convective–symmetric instability. And
also the existence of deep cumulonimbus, large precipitation rates, strong downdrafts, or lightning is not an
adequate discriminator between the existence of moist gravitational and moist slantwise convection. It is also
sometimes stated that the existence of deep cumulonimbus, large precipitation rates, strong downdrafts, or lightning
(i.e., thunderstorms) precludes moist slantwise convection. The following argument shows the insufficiency to that
claim.
In isolated thunderstorms, it is generally accepted that strong updrafts exceeding 5 m s−1 favour coexisting ice and
water phases that appear to be required for the generation of lightning. This would appear to exclude slantwise
- 32 -
convection as being associated with lightning, since most slantwise updrafts are assumed to be weaker (tens of cm
s−1 to a few m s−1), although direct observations are lacking. On the other hand, Williams (1991) notes that the
mechanisms that lead to charge separation are, in principle, independent of the existence of moist gravitational
instability, indicating the possibility that lightning could be associated with slantwise convection.
5.4 Slantwise convective available potential energy SCAPE
Analogous to convective available potential energy (CAPE) for gravitational convection, slantwise convective
available potential energy (SCAPE) can be defined (Emanuel 1983b, Shutts 1990; Emanuel 1994). These
parameters measure the amount of positive area on a thermodynamic diagram and represent the maximum kinetic
energy possible from adiabatic, inviscid parcel theory. In the case of CAPE, it is the maximum kinetic energy of a
vertical updraft; in the case of SCAPE, it is the maximum kinetic energy of an updraft along a Mg surface,
consisting of both, horizontal and vertical motions. Anyway, there are three possible explanations for an apparent
nonutility of SCAPE compared to EPVg:
First, CAPE typically represents available energy through the entire troposphere once the parcel reaches its LFC.
The LCL and LFC are usually fairly close to each other in gravitational convection. With SCAPE, however, the
vertical extent of the instability is usually much more shallow (less than a couple hundred hPa) and the horizontal
and vertical distances between the slantwise LCL and slantwise LFC can be much greater. The location of CSI with
respect to the location of the lift may be critical because of the typically shallow nature of the instability (and its
location far removed from the slantwise LCL). By using EPVg, the spatial relationship between the instability and
lift can be visualized. For example, knowledge of the relative locations between the region of EPVg < 0 and the
lifting mechanism allows the forecaster to know how strong the lift has to be to reach the instability. Operational
experience suggests that a separation of 100 or 200 hPa between the region of release of MSI and the negative EPVg
layer implies that a strong vertical circulation is required to access the instability.
Second, the computation and display of SCAPE is problematic. Unlike CAPE, where a single vertically integrated
value can be associated with each horizontal location and can be easily displayed on a horizontal map, SCAPE is an
integrated quantity along a slantwise path.
Third, observed values of SCAPE tend to be relatively small, whereas mesoscale numerical models tend to produce
larger values. Perhaps an analogy can be drawn between gravitational and slantwise convection to explain this
observation. Before convection starts, soundings in the environment would indicate positive available potential
energy. Once convection occurs, soundings taken within storms would indicate little, if any, CAPE/SCAPE.
Previous work on SCAPE has focused on the period when slantwise convection is already occurring (Emanuel,
1988). This may explain the absence of SCAPE (i.e., the symmetrically neutral state) in such observational studies.
In contrast, mesoscale models typically do not have the resolution or the parameterization to reduce SCAPE
(Schultz and Schumacher, 1999). Therefore, SCAPE may build up to values larger than those observed.
Previous work by Shutts (1990) and more recent research by Dixon (2000) suggests some utility of SCAPE in
numerical models, particularly in anticipating the development of cloud heads in oceanic cyclones.
Due to limited time it was not possible to address the topic of SCAPE sufficiently; the author of this report therefore
suggests more thorough investigations in the future. For the atmospheric layer between 925 and about 600 hPa
results should be similar to Browning et al. (2001). Nordeng, (1987) and McCann (1995), derive SCAPE from
applying the thermal wind equation on the equation of the moist potential geostrophic Vorticity. Thus, SCAPE can
be calculated from operationally used model fields.
SCAPE = CAPE + ∫
z1
z0
(
)
f d
(v − v0 )2 dz
2η dz
- 33 -
6. Applying the concept of moist slantwise instability on convection at upper
tropospheric moisture gradients
Assumption:
An overall conditionally unstable stratification during a period of weak low level pressure gradient (fair weather)
may be favourable to the onset of convective overturning above ground. Depending on the given topographical
height an air parcel situated at higher elevations (mountains, hills) needs less energy and heating in order to become
buoyant and starting to rise. Later in time, the onset of convective overturning will begin also in flat regions. If there
is enough lift, an air parcel may reach saturation – earlier or later in time given the initial level of topographical
height.
In order to reach the LFC two ingredients are needed in combination with given instability:
Sufficient moisture from ground levels, supplying the growing cell within regions of conditional instability,
and / or sufficient lift in case of inaccurate moisture supply from the ground.
It can be assumed that the fair weather convection deals with less favourable conditions regarding these two
ingredients:
• Overall moisture supply is less than e.g. during a prefrontal situation in summer season, when the formation
of thunderstorms at the southern Alpine flank is most likely, because of increased moisture advection
provided by a distinct southerly or a south – westerly stream at low and mid tropospheric levels.
• Although different regions of moisture are depicted in the WV images, nonetheless the strong gradient of
relative humidity between 600 and 400 hPa (shown in the soundings, see chapter 4.3). The investigation
showed that this distinct vertical moisture gradient can be found in 90% of all cases. In upper regions a
further horizontal gradient is marked by the moisture gradients in the WV image. Therefore it may be
concluded, that entrainment becomes stronger at the vertical moisture gradient and also seems to vary at
upper levels in the vicinity of the gradient between moist and dry regions in the WV image.
In the vertical cross section (VCS) of the ALADIN local area model a zone with positive values of potential
vorticity (PV) along the dark stripe in the WV image can be connected to a dry intrusion from tropopause levels
(Browning, 1993) (see Fig. 18 – 20). In the following the latitudinal cross section is discussed.
Fig. 17: MSG WV5, 28 July 2005, 1200 UTC, zonal and meridional vertical cross sections indicated by dashed line.
- 34 -
Fig. 18: 28 July 2005, 1200 UTC, ALADIN zonal vertical cross section. Positive values of the potential vorticity: solid blue
lines. Values of 0.5 PV units at a level of about 600 hPa indicate the dry intrusion of the WV dark stripe.
Fig. 19: 28 July 2005, 1200 UTC, ALADIN zonal vertical cross section. Positive values of the equivalent potential vorticity
(unsaturated): solid blue lines. Values of PVU = 1. 5 PV units at a level of about 300 hPa indicate the dry intrusion of the WV
dark stripe. Dashed lines: negative values, show distinct maxima between 850 and 600 hPa, also small positive maxima can be
seen.
Fig. 19 shows the widespread distribution of negative values of EPV up to heights of about 500 hPa. The singular
small positive maxima will be subject of further scientific investigation. Yet it is unclear if the ALADIN LAM
model is noisy regarding this parameter, or if they have any meaning for the physical process under discussion. Fig.
22A shows the distribution of the EPV within the layer between 850 and 500 hPa with the coarser resolution of the
global ECMWF 0.5° model naturally without such small scale positive maxima. However, significant values of
negative EPV indicate a possible release of CSI.
- 35 -
If a still buoyant and saturated air parcel is reaching such a zone of negative EPV (e.g. around 850 hPa, see Fig. 19),
CSI might be released leading to an increase of buoyancy of the air parcel. Because of this additional lift the
saturated air parcel might reach its level of free convection (LFC). Due to conditional instability deep convection
will result.
We also have to keep in mind that the vertical acceleration induced by CSI reaches only values reaching from tens
of cm/s to a few m/s, whereas gravitational instability (CI) provides speeds up to tens of m/s. Thus, during the
coexistence of symmetric instabilities and gravitational instabilities, gravitational convection will dominate
slantwise convection. This could be the reason why deep moist convection (DMC) is observed which is not situated
at a WV – boundary at days with convection at WV – boundaries.
But if gravitational convection is weak, moisture supply is inefficient, entrainment causes negative buoyancy or the
rising parcel has to surmount a capping inversion, then the release of CSI might modify the convective process such
as being sufficient to accelerate the parcel towards its LFC (see also Jascourt, 1988 and Xu, 1985).
The areas within the immediate vicinity of the WV boundaries seem to fulfil all the preconditions necessary for the
release of CSI.
They are summarized:
1. The WV boundaries indicate zones with distinct vertical wind shear but little directional shear
2. Weak gravitational stability
3. The release of CSI favourably takes place in areas at or near saturation: if a saturated buoyant air parcel
reaches a zone of distinct negative EPV
4. Dynamical processes (Santurette and Georgiev, 2005) lead to local deformation at the boundary zone,
which promotes EPV becoming smaller or negative: EPV <0
5. A significant horizontal thermal gradient is induced by the dry and cold intrusion along with the WV dark
zone
If WV boundaries fulfil conditions favourable for the release of CSI, the question arises where exactly at a long and
elongated boundary the onset of DMC will take place. Except from favoured mountainous areas with initially low
gravitational stability those regions can be distinguished objectively by three ways:
•
•
•
The calculation of areas with favourable Mg–θe relationship (Schultz and Schumacher, 1999) (not used in
operational forecasting)
The calculation of SCAPE (no common use in forecasting, too)
The calculation of the equivalent potential Vorticity (EPV < 0) and the increase or decrease of its gradient
The latter shall be subject of the following discussion:
It is of interest to analyse the behaviour of conservative field quantities within the stream. EPV is such a
conservative field quantity and also an indicator of the existence of CSI.
Yet the diurnal cycle of the equivalent potential temperature leads to an increase or decrease of its vertical and
horizontal gradients. Further the movement of the WV features like dark stripes or filaments within the flow leads to
a change in vertical stability, gradient of vertical wind speed, vertical shear conditions and also the horizontal layer
distribution of the potential, and equivalent potential temperature.
In order to indicate such changing conditions and thus a change in the behaviour of the EPV field it is necessary to
develop an indicating parameter.
Bluestein (1993), Houze (1993), Emanuel (1994) stated, that if deformation terms in the geostrophic wind field are
dominant, the horizontal gradients of conservative field quantities such as: PV, EPV and Θe are strengthened or
weakened.
Starting from Ertel’s theorem for frictionless and adiabatic motion (Ertel, 1942)
DP ∂P r
≡
+ V2 ⋅ ∇P = 0
Dt
∂t
- 36 -
Here the approximation for a horizontal and geostrophic flow shall be applied, where v2 is the geostrophic velocity
for horizontal coordinates. To keep overview with the formulary terms EPV may be written as P.
Further, the gradient shall be applied onto the whole equation system, resulting in
D ∂P
∂P
× r = Ν⋅ r ,
Dt ∂x
∂x
where
Ν ≡ −(
∂v
)
∂x
represents the velocity gradient matrix.
Under the assumption that the tensor N only varies slowly compared to the gradient of the potential vorticity a
solution for the time dependence of the gradient of P in a coordinate system moving with the geostrophic wind
vector can be found (Skoda, 2007).
r
r
∇P = A exp[νt ]
r
The scalar ν is the eigenvalue of N, A its eigenvector. From trace N = 0 it follows that there is an option in
choosing the direction of the eigenvector and consequently the gradient of P is selected at t=0, as a constant
r
eigenvector which satisfies the equation above. The existence of a nontrivial solution of A ≠ 0 requires ν1 and
ν 2 to be the roots of the characteristic equation
r
r
0 = det(∇P −νI )
r
With the identity matrix I . It follows from the equation above, that the so called NUE – parameter can be
written as
(E 2 + F 2 − ξ 2 ) 2
] .
4
1
ν = ±[
Where E =
F=
∂u g
∂x
∂v g
∂x
ξ=
+
∂v g
∂x
−
∂v g
∂y
∂u g
∂y
−
∂u g
∂y
is the elongation deformation,
is the shear deformation and
is the relative Vorticity.
- 37 -
Thus, the magnitude of the P – gradient will tend to grow in time at an exponential rate, whenever ν 2 is positive.
When ν 2 is negative, the time evolution of the P gradient is oscillatory. ν = 0 means a loss of variability in time.
Therefore, values near zero will have no strong influence on the variation of gradient P.
In general more distinctions have to be made:
•
•
The dominance of the deformation terms ( E 2 + F 2 ) , or the dominance of the relative vorticity ξ .
Also a distinction has to be made between positive and negative values of the relative vorticity.
The case of a positive relative vorticity regarding the NUE – parameter will have influence on the growth of the
gradients of conservative field quantities on larger scales e.g. at frontal zones, postfrontal convection, Comma
clouds and enhanced cumuli.
In the case of negative relative vorticity mainly areas within pressure ridges will be affected.
Here, we distinguish between small negative values of ξ where the deformation terms will dominate and large
values of ξ , where ν 2 is again negative. The latter can be interpreted, that the airflow is mainly turbulent and
sensible to small disturbances (Skoda, 2007).
So, along the trajectory two solutions for ν 2 exist. To include the impact from the imaginary solution as well as
from the real solution, the resulting forcing from both, the deformation terms and the vorticity terms shall be
estimated.
In order to combine those two characteristics of the NUE – parameter and regarding the time invariability at ν = 0 ,
the calculation of the NUE – parameter shall be
1
(E 2 + F 2 − ξ 2 ) 2
] )
ν=(
4
for ξ < 0 , under the assumption of a geostrophic flow with zero divergence.
Seen in combination with the gradients in the WV image, a zone, favourable to DMC onset can be highlighted at a
time before even shallow convection occurs. Together with all the above described ingredients, deep moist
convection is likely to be expected at a WV boundary, resulting from shallow convection.
On the 28 July 2005 the first appearing deep convection took place around 47.5° N and 15° E. The ALADIN
vertical cross section (Fig. 20 below) shows significant values of the NUE – parameter between 850 and 600 hPa
around the same location.
- 38 -
Fig. 20: 28 July 2005, 1200 UTC, ALADIN zonal vertical cross section. NUE – parameter. Positive values represent the
increase of the gradient of EPV in case of ξ < 0 .
Fig. 21 shows the distribution of EPV in the layer between 850 and 500 hPa from the ECMWF 0.5° model at 0600
UTC. Over the western and northern part of Austria minima of EPV can be seen in the vicinity of the WV dark
stripe crossing the country from South to North.
Fig. 21: MSG WV5, 28 July 2005, 0600 UTC;
Layer EPVg (850 – 500 hPa), positive values: solid lines, negative values: dashed lines; significant maxima in the vicinity of
the WV boundary over Western Austria
The gradients around the minimum values of negative EPV have further increased 6 hours later (Fig. 22 A). At
constant pressure levels of 850, 700 an 500 hPa the NUE – parameter impacts the immediate vicinity of the first
growing deep convective cell in northern Austria. (Fig. 22, B – D).
Fig. 23 shows the resulting convective cells at their mature stadium at 1500 UTC.
- 39 -
A
B
C
D
Fig. 22: MSG WV5, 28 July 2005, 1200 UTC;
A: Layer EPVg (850 – 500 hPa), positive values: solid lines, negative values: dashed lines; significant maxima in the vicinity of
the WV boundary over Western Austria
B: NUE – parameter at 850 hPa, anticyclonic ( ξ < 0 ) intensification of EPVg
C: NUE – parameter at 700 hPa, anticyclonic ( ξ < 0 ) intensification of EPVg
D: NUE – parameter at 500 hPa, anticyclonic ( ξ < 0 ) intensification of EPVg
At all levels a further intensification of anticyclonic EPVg is observable, the onset of DMC is to be expected within this area
Fig. 23: MSG WV5, 28 July 2005, 1500 UTC; Time step of most intense convective activity at the WV boundary
- 40 -
The NUE – parameter seems useful at the lower half of the troposphere, convective cells above 500 hPa will have
already reached the mature stadium and further vertical displacement caused by slant convection may only affect
the stratiform precipitation amount from the trailing anvil. These considerations might be supported by the findings
of Jascourt (1988), Browning (2001), Dixon (2000) and Xu (1986).
- 41 -
7. Application of NWCSAF products and Global Instability Index
An additional aim for this VSA was to investigate if MSG NWCSAF products like LPW (layer precipitable water),
TPW (total precipitable water) and SAI (stability analysis imagery) can contribute to identify preferred areas at the
WV boundaries which are prone to deep convection.
Every available case has also been investigated applying these products, but the field values of the MSG NWCSAF
products showed no coherence or any specific pattern which could be related to the convective process under
discussion. All three products in question provide overall information about the unstable stratification and a certain
potential for precipitation over the region of interest. But no additional information was given in regard to the WV
boundary problem. Because of these obvious subjective results no objective method was developed to further
investigate the problem. Also a short look on the new product version (2.0) showed no different results. The author
of this report doubts if succeeding investigations will bring any results in connection to the WV boundary problem.
A
B
C
Fig. 24: NWCSAF parameters, 28 July 2005, 1100 UTC;
A: Precipitable water content,
B: Layer precipitable water – total,
C: Stability analysis
- 42 -
Fig. 25: MPEF parameter GII, 28 July 2005, 0600 UTC (left) and 1200 UTC (right);
Lifted Index
The MPEF parameter LI (Fig. 25 above) seems to indicate the convective potential more accurate compared to the
other parameters like TPW, PW and SAI. Observing two time steps (0600 UTC and 1200 UTC, Fig. 25) shows a
diurnal cycle over Austria, but still no pattern related to the moisture gradient at upper levels (i.e. see Fig. 2, chapter
4). Thus, no specific indication for DMC in the vicinity of the WV boundaries is given, although WV channels
contribute to the products under discussion.
So, without any knowledge of CSI, all instability indices give only overall readiness for convective onset but no
indication where it is going to happen exactly. Results are similar for all investigated cases; some examples are
given in the Appendix below.
- 43 -
8. Conclusion and outlook
Dark features (called water vapour boundaries, WV boundaries) in the MSG WV imagery are clearly connected
with triggering of deep moist convection (Krennert, Zwatz-Meise, 2003). Though WV boundaries are no trigger of
convective activity themselves, a further triggering of deep convection out of shallow convection can be well
observed within their vicinity. The aim of this Visiting Scientist Action was to scientifically further narrow the
regions particularly endangered by developing deep convection.
A general overview about the nature of convection, types of instabilities and the properties of the MSG WV
channels 5 and 6 have been given in the chapters 1 - 3. A thorough description of the physical characteristics of the
mechanism was given in chapter 4. Also all representative NWP model parameters have been discussed.
In order to identify regions of convective activity, many concepts regarding the lower levels of the atmosphere are
known and widely-used within the meteorological community. Concepts for triggering of deep convection from mid
and high levels of the troposphere are quite rare and mostly connected to pre- or postfrontal convection and upper
level troughs.
To solve the problem under discussion, the concept of Moist Symmetric Instabilities was introduced (chapter 5) and
applied to convective activity under “fair weather” conditions.
It became clear, that regions superimposed by distinct negative values of quasi geostrophic equivalent potential
vorticity EPVg will provide additional lift through slantwise convection onto an existing “shallow” convective cell.
Given small values of conditional stability a saturated buoyant air parcel will reach its level of free convection with
the help of slantwise convection. Therefore, combining WV boundaries with regions of distinct EPVg minima
increases the probability for the onset of deep moist convection.
A further parameter (so called NUE - parameter) indicating the increase of the EPVg gradient (thence intensifying
minima) was derived and was superimposed over zones with convective activity. The results clearly show, that a
steepening of the EPVg gradient as well as a succeeding deep convective development can be successfully
connected to this parameter.
The region of interest was the greater Alpine region. The synoptic situation under discussion appears about twenty
times a year over the Alps. Due to the small number of available cases (about 20 per year) a mainly qualitative
discussion was performed within this report. Due to a limitation of available NWP model data only for two years, a
statistical evaluation seemed not reasonable.
Also the NWCSAF AMA dark stripe product was qualitatively evaluated in combination with other parameters to
elaborate the conditions for flagging a section of a dark stripe in a future release as "dangerous". Some of these
parameters from the NWC SAF have also been taken qualitatively into consideration.
A thorough listing of case studies is given in the appendix, showing the behaviour of significant parameters under
discussion.
The following items will have to be addressed in the near future:
Adopting the program thresholds of the AMA product could possibly influence the stripe detection. Due to different
characteristics of the moisture gradient in the WV image (flat or steep gradient, strong or weak gradient) it is not
clear at the current stage of development if the detection of the WV boundaries can be given by a constricted line at
all or should be indicated by a zone.
An objective investigation of the statistical behaviour of the model derived parameters EPVg and the NUE parameter is needed.
These results might contribute to an improvement of the parameterization of LAM models like ALADIN. It has to
be shown if an extra CSI parameterization of the model is necessary.
Also additional parameters indicating the release of CSI shall be taken into account: e.g. SCAPE
- 44 -
9. Appendix:
Appendix: Case Studies
The current report gives mostly a qualitative discussion about the investigated topic. Due to a small number of cases
and the limited data availability a thorough statistical investigation seemed unreasonable. This is the reason why a
rather extended appendix with 11 case studies is provided, in order to outline the behaviour of all involved major
parameters.
Throughout all case studies Fig. A shows the initial conditions hours before the first deep moist convective activity.
Also the distribution of the equivalent potential vorticity within a layer between 850 and 500 hPa is plotted, derived
from the global ECMWF model with a grid point resolution of 0.5°. According to the chapters 5 and 6 negative
values of EPVg are seen as a precursor for the release of CSI and thus as a source of additional lift.
Fig. B shows the NWCSAF AMA dark stripe product objectively and automatically derived from MSG WV
channel 6 (7.3µ) as an overlay at the time when deep moist convection is first seen in the WV channel. The AMA
product indicates a moisture boundary, which can be correlated to the first appearing convective cells.
Fig. C gives the same contents only for the MSG WV channel 5 (6.2µ).
In all cases the time of first appearance of the deep convective cells lies between the time points of the model time
resolution (ALADIN and ECMWF: 6 hours), due to limited time during the VSA a interpolation or temporal
downscaling was not performed.
Fig. D gives the same information as Fig. A, but 6 hours later. In all cases a distinct increase of the minima of the
negative values of EPVg can be seen, which is mostly supported by the indicating NUE - parameter (Fig. E). In Fig.
E also the AMA product can be seen derived form WV channel 5 and 6.
In Fig. F to H three MSG Nowcast - Satellite - Application - Facility products are shown. Every available case has
been investigated also with the NWCSAF products. No additional information about the area in the vicinity of a
WV boundary prone to convection could be derived. In general only a more or less intense unstable stratification is
indicated by the SAF products.
- 45 -
CASE 1, 060627:
On the 27 June 2006 0600 UTC, old thunderstorm cells can be seen over the northern part of Austria and
southern Poland. The area of interest lies to the south of the old thunderstorms, indicated by a distinct
minimum of negative EPVg (Fig. A). Fig. B and C show the first developing cell at the same minimum.
Also the increase of the EPVg gradient at 1200 UTC (Fig. D) is well supported by the NUE - parameter
(Fig. E).
CASE 1, Fig. A: 060627, 0600 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
- 46 -
CASE 1, Fig. B: 060627, 1000 UTC, MSG WV6, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 6.
CASE 1, Fig. C: 060627, 1000 UTC, MSG WV5, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 5.
- 47 -
CASE 1, Fig. D: 060627, 1200 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
CASE 1, Fig. E: 060627, 1200 UTC, MSG WV5, ECMWF 0.5° model,
cyan solid lines: NUE – parameter at 700 hPa, anticyclonic ( ξ < 0 ),
indicating intensification of the EPVg gradient.
Red solid lines: MSG WV6 NWCSAF AMA dark stripe product.
Blue solid lines: MSG WV5 NWCSAF AMA dark stripe product.
- 48 -
CASE 1, Fig. F: 060627, 1100 UTC, MSG SAFNWC, layer precipitable water, all layers
CASE 1, Fig. G: 060627, 1100 UTC, MSG SAFNWC, stability analysis
CASE 1, Fig. H: 060627, 1100 UTC, MSG SAFNWC, total precipitable water
- 49 -
CASE 2, 060704:
On the 4 July 2006 no deep convection can be seen at 1200 UTC over the Alpine region. Also overall
weak gradients of negative EPVg are shown in Fig. A. Only at 1600 UTC deep convection starts over the
central parts of Austria. Fig. B shows a weak gradient in the WV 6 image, which can be associated to the
convective cell. Also WV channel 5 shows a weak gradient and is also slightly better supported by the
AMA product (Fig. C). A more advanced stadium of the cells life cycle is reached at 1800 UTC, but still
no significant change in the EPVg gradient can be seen (Fig. D). Also no significant value of the NUE parameter is positioned in the immediate vicinity (Fig. E).
CASE 2, Fig. A: 060704, 1200 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
- 50 -
CASE 2, Fig. B: 060704, 1300 UTC, MSG WV6, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 6.
CASE 2, Fig. C: 060704, 1300 UTC, MSG WV5, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 5.
- 51 -
CASE 2, Fig. D: 060704, 1800 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
CASE 2, Fig. E: 060704, 1800 UTC, MSG WV5, ECMWF 0.5° model,
Cyan solid lines: NUE – parameter at 700 hPa, anticyclonic ( ξ < 0 ),
indicating intensification of the EPVg gradient.
Red solid lines: MSG WV6 NWCSAF AMA dark stripe product.
Blue solid lines: MSG WV5 NWCSAF AMA dark stripe product.
- 52 -
CASE 2, Fig. F: 060704, 1100 UTC, MSG SAFNWC, precipitable water, all layers
CASE 2, Fig. G: 060704, 1100 UTC, MSG SAFNWC, stability analysis
CASE 2, Fig. H: 060704, 1100 UTC, MSG SAFNWC, total precipitable water
- 53 -
CASE 3, 060705:
A rather distinct WV dark stripe is crossing the eastern part of Austria. First signs of deep convective
activity can only be observed at 1200 UTC over the central and western parts of Austria. Though the cells
are rather weak in the satellite image, they are well supported by the EPVg minimum and the NUE parameter (Fig. D and E).
CASE 3, Fig. A: 060705, 0600 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
The time of the first appearance of the deep convective activity is the same as displayed in CASE 3, Fig. E.
- 54 -
CASE 3, Fig. D: 060705, 1200 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
CASE 3, Fig. E: 060705, 1200 UTC, MSG WV5, ECMWF 0.5° model,
Cyan solid lines: NUE – parameter at 700 hPa, anticyclonic ( ξ < 0 ),
indicating intensification of the EPVg gradient.
Red solid lines: MSG WV6 NWCSAF AMA dark stripe product.
Blue solid lines: MSG WV5 NWCSAF AMA dark stripe product.
- 55 -
CASE 3, Fig. F: 060705, 1100 UTC, MSG SAFNWC, precipitable water, all layers
CASE 3, Fig. G: 060705, 1100 UTC, MSG SAFNWC, stability analysis
CASE 3, Fig. H: 060705, 1100 UTC, MSG SAFNWC, total precipitable water
- 56 -
CASE 4, 060706:
A distinct WV boundary is positioned across the eastern part of Austria with beginning deep convective
activity at its left boundary (Fig. A). Because of the sharp gradient in upper level moisture both of the
AMA products are set exactly at the boundary of the WV dark stripe (B and C). Growing cells are well
associated to minima of EPVg and the distribution of the NUE - parameter (D, E).
CASE 4, Fig. A: 060706, 0600 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
- 57 -
CASE 4, Fig. B: 060706, 1100 UTC, MSG WV6, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 6.
CASE 4, Fig. C: 060706, 1100 UTC, MSG WV5, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 5.
- 58 -
CASE 4, Fig. D: 060706, 1200 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
CASE 4, Fig. E: 060706, 1200 UTC, MSG WV5, ECMWF 0.5° model,
Cyan solid lines: NUE – parameter at 700 hPa, anticyclonic ( ξ < 0 ),
indicating intensification of the EPVg gradient.
Red solid lines: MSG WV6 NWCSAF AMA dark stripe product.
Blue solid lines: MSG WV5 NWCSAF AMA dark stripe product.
- 59 -
CASE 4, Fig. F: 060706, 1100 UTC, MSG SAFNWC, precipitable water, all layers
CASE 4, Fig. G: 060706, 1100 UTC, MSG SAFNWC, stability analysis
CASE 4, Fig. H: 060706, 1100 UTC, MSG SAFNWC, total precipitable water
- 60 -
CASE 5, 060707:
At 1200 UTC a convective development over the eastern part of Austria can be seen at the WV boundary
in channel 5 (Fig. C). Fig. D and E show no distinct support by the other parameters.
CASE 5, Fig. A: 060707, 0600 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
- 61 -
CASE 5, Fig. B: 060707, 1100 UTC, MSG WV6, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 6.
CASE 5, Fig. C: 060707, 1100 UTC, MSG WV5, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 5.
- 62 -
CASE 5, Fig. D: 060707, 1200 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
CASE 5, Fig. E: 060707, 1200 UTC, MSG WV5, ECMWF 0.5° model,
Cyan solid lines: NUE – parameter at 700 hPa, anticyclonic ( ξ < 0 ),
indicating intensification of the EPVg gradient.
Red solid lines: MSG WV6 NWCSAF AMA dark stripe product.
Blue solid lines: MSG WV5 NWCSAF AMA dark stripe product.
- 63 -
CASE 5, Fig. F: 060707, 1100 UTC, MSG SAFNWC, layer precipitable water, all layers
CASE 5, Fig. G: 060707, 1100 UTC, MSG SAFNWC, stability analysis
CASE 5, Fig. H: 060707, 1100 UTC, MSG SAFNWC, total precipitable water
- 64 -
CASE 6, 060709:
The following case shows convective development at rather weak WV boundaries. Only few cells can be
associated to the MSG channel 6 AMA product (Fig. B), at higher levels also no significant connection
can be seen (Fig. C). Though negative EPVg also has a field distribution with weak gradients, the NUE parameter is well supporting all convective activities (Fig. E).
CASE 6, Fig. A: 060709, 0600 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
- 65 -
CASE 6, Fig. B: 060709, 0900 UTC, MSG WV6, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 6.
CASE 6, Fig. C: 060709, 0900 UTC, MSG WV5, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 5.
- 66 -
CASE 6, Fig. D: 060709, 1200 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
CASE 6, Fig. E: 060709, 1200 UTC, MSG WV5, ECMWF 0.5° model,
Cyan solid lines: NUE – parameter at 700 hPa, anticyclonic ( ξ < 0 ),
indicating intensification of the EPVg gradient.
Red solid lines: MSG WV6 NWCSAF AMA dark stripe product.
Blue solid lines: MSG WV5 NWCSAF AMA dark stripe product.
- 67 -
CASE 6, Fig. F: 060709, 1100 UTC, MSG SAFNWC, layer precipitable water, all layers
CASE 6, Fig. G: 060709, 1100 UTC, MSG SAFNWC, stability analysis
CASE 6, Fig. H: 060709, 1100 UTC, MSG SAFNWC, total precipitable water
- 68 -
CASE 7, 060712:
The 12 of July 2006 at 0600 UTC is again a case with a weak EPVg gradient and also weak moisture
gradients in WV channel 5 and 6 (Fig. A). But none the less well supported by the AMA products channel
5 and channel 6 (Fig. B and C). At 1200 UTC the EPVg gradient is still weak (Fig. D), but the NUE parameter supports the deep convective development very well (Fig. E).
CASE 7, Fig. A: 060712, 0600 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
- 69 -
CASE 7, Fig. B: 060712, 1030 UTC, MSG WV6, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 6.
CASE 7, Fig. C: 060712, 1030 UTC, MSG WV5, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 5.
- 70 -
CASE 7, Fig. D: 060712, 1200 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
CASE 7, Fig. E: 060712, 1200 UTC, MSG WV5, ECMWF 0.5° model,
Cyan solid lines: NUE – parameter at 700 hPa, anticyclonic ( ξ < 0 ),
indicating intensification of the EPVg gradient.
Red solid lines: MSG WV6 NWCSAF AMA dark stripe product.
Blue solid lines: MSG WV5 NWCSAF AMA dark stripe product.
- 71 -
CASE 7, Fig. F: 060712, 1100 UTC, MSG SAFNWC, layer precipitable water, all layers
CASE 7, Fig. G: 060712, 1100 UTC, MSG SAFNWC, stability analysis
CASE 7, Fig. H: 060712, 1100 UTC, MSG SAFNWC, total precipitable water
- 72 -
CASE 8, 060713:
Convective development is taking place all around Austria, easily connected to distinct WV gradients.
Over the eastern part of Austria a rather flat gradient appears at 1000 UTC (Fig. B and C) and is
recognized by the AMA products. Though the intensities of the negative EPVg increase towards the
minima until 1200 UTC (Fig. D), the NUE - parameter can only be found more to the south in connection
to the developed cell over the south eastern parts of Austria (Fig. E).
CASE 8, Fig. A: 060713, 0600 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
- 73 -
CASE 8, Fig. B: 060713, 1000 UTC, MSG WV6, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 6.
CASE 8, Fig. C: 060713, 1000 UTC, MSG WV5, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 5.
- 74 -
CASE 8, Fig. D: 060713, 1200 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
CASE 8, Fig. E: 060713, 1200 UTC, MSG WV5, ECMWF 0.5° model,
Cyan solid lines: NUE – parameter at 700 hPa, anticyclonic ( ξ < 0 ),
indicating intensification of the EPVg gradient.
Red solid lines: MSG WV6 NWCSAF AMA dark stripe product.
Blue solid lines: MSG WV5 NWCSAF AMA dark stripe product.
- 75 -
CASE 8, Fig. F: 060713, 1100 UTC, MSG SAFNWC, layer precipitable water, all layers
CASE 8, Fig. G: 060713, 1100 UTC, MSG SAFNWC, stability analysis
CASE 8, Fig. H: 060713, 1100 UTC, MSG SAFNWC, total precipitable water
- 76 -
CASE 9, 060720:
Over the eastern half of Austria no distinct WV gradients can be found at 1200 UTC, but distinct gradients
exit in the field distribution of negative EPVg (Fig. A). At 1600 UTC WV boundaries are still very weak
and thus not detected by the AMA products. At 1800 UTC WV boundaries are still weak, but the model
parameters show good support (Fig. D and E).
CASE 9, Fig. A: 060720, 1200 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
- 77 -
CASE 9, Fig. B: 060720, 1600 UTC, MSG WV6, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 6.
CASE 9, Fig. C: 060720, 1600 UTC, MSG WV5, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 5.
- 78 -
CASE 9, Fig. D: 060720, 1800 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
CASE 9, Fig. E: 060720, 1800 UTC, MSG WV5, ECMWF 0.5° model,
Cyan solid lines: NUE – parameter at 700 hPa, anticyclonic ( ξ < 0 ),
indicating intensification of the EPVg gradient.
Red solid lines: MSG WV6 NWCSAF AMA dark stripe product.
Blue solid lines: MSG WV5 NWCSAF AMA dark stripe product.
- 79 -
CASE 9, Fig. F: 060720, 1100 UTC, MSG SAFNWC, layer precipitable water, all layers
CASE 9, Fig. G: 060720, 1100 UTC, MSG SAFNWC, stability analysis
CASE 9, Fig. H: 060720, 1100 UTC, MSG SAFNWC, total precipitable water
- 80 -
CASE 10, 060721:
WV boundaries over teh western part of Austria and a WV eddy over the eastern half of Austria mark the
initial condition on the 21 July 2006 at 0600 UTC. Also distinct minima of negative EPVg can be seen
over the region of interest (Fig. A). Though the WV gradients are not well detected by teh AMA product
(Fig. B and C), the model parameters give good support (Fig. D and E).
CASE 10, Fig. A: 060721, 0600 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
- 81 -
CASE 10, Fig. B: 060721, 1000 UTC, MSG WV6, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 6.
CASE 10, Fig. C: 060721, 1000 UTC, MSG WV5, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 5.
- 82 -
CASE 10, Fig. D: 060721, 1200 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
CASE 10, Fig. E: 060721, 1200 UTC, MSG WV5, ECMWF 0.5° model,
Cyan solid lines: NUE – parameter at 700 hPa, anticyclonic ( ξ < 0 ),
indicating intensification of the EPVg gradient.
Red solid lines: MSG WV6 NWCSAF AMA dark stripe product.
Blue solid lines: MSG WV5 NWCSAF AMA dark stripe product.
- 83 -
CASE 10, Fig. F: 060721, 1100 UTC, MSG SAFNWC, layer precipitable water, all layers
CASE 10, Fig. G: 060721, 1100 UTC, MSG SAFNWC, stability analysis
CASE 10, Fig. H: 060721, 1100 UTC, MSG SAFNWC, total precipitable water
- 84 -
CASE 11, 060723:
At 0600 UTC a well recognisable dark zone in the WV image is situated over the eastern part of Austria
and the EPVg distribution shows low gradients (Fig. A). During the next 6 hours the development of
various convective cells can be observed and is well diagnosed by the AMA products as well as by the
model parameters.
CASE 11, Fig. A: 060723, 0600 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid:
positive values, dashed: negative values
- 85 -
CASE 11, Fig. B: 060723, 1000 UTC, MSG WV6, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 6.
CASE 11, Fig. C: 060723, 1000 UTC, MSG WV5, NWCSAF AMA dark stripe product. Time of first appearance of
the deep convective activity in MSG WV channel 5.
- 86 -
CASE 11, Fig. D: 060723, 1200 UTC, MSG WV5, ECMWF 0.5° model, equivalent potential vorticity, solid: positive
values, dashed: negative values
CASE 11, Fig. E: 060723, 1200 UTC, MSG WV5, ECMWF 0.5° model,
Cyan solid lines: NUE – parameter at 700 hPa, anticyclonic ( ξ < 0 ),
indicating intensification of the EPVg gradient.
Red solid lines: MSG WV6 NWCSAF AMA dark stripe product.
Blue solid lines: MSG WV5 NWCSAF AMA dark stripe product.
- 87 -
CASE 11, Fig. F: 060723, 1100 UTC, MSG SAFNWC, layer precipitable water, all layers
CASE 11, Fig. G: 060704, 1100 UTC, MSG SAFNWC, stability analysis
CASE 11, Fig. H: 060723, 1100 UTC, MSG SAFNWC, total precipitable water
- 88 -
9. References
AMS Glossary, 2000, second edition, American Meteorological Society
BLUESTEIN, H. B., 1993: Synoptic-Dynamic Meteorology in Midlatitudes. Volume II: Observations and Theory of
Weather Systems. Oxford University Press, section 3.5.2.
BROWNING, K.A., 1993: Evolution of a Mesoscale Upper Tropospheric Vorticity Maximum and Comma Cloud
from a Cloud Free two – Dimensional Potential Vorticity Anomaly, Q. J. R. Meteorol. Soc., 119, 883 ff
BROWNING, K. A., Chapman, D., Dixon, R. S., 2001: Stacked Slantwise Convective Circulations, Q. J. R.
Meteorol. Soc., 127, 2513 ff
DIXON, R. S., 2000: Diagnostic studies of symmetric instability. Ph.D. dissertation, University of Reading, 128 pp.
[Available from Department of Meteorology, University of Reading, Earley Gate, P.O. Box 243, Reading RG6
6BB, United Kingdom.].
DOSWELL, C. A. III, 1987: The distinction between large-scale and mesoscale contribution to severe convection:
A case study example. Wea. Forecasting, 2, 3 ff.
DOSWELL, C. A., III, C. Ramis, R. Romero, and S. Alonso, 1998: A diagnostic study of three heavy precipitation
episodes in the western Mediterranean region. Wea. Forecasting, 13, 102 ff
DOSWELL III, C.A., 2003: Societal impacts of severe thunderstorms and tornadoes: Lessons learned and
implications for Europe. Atmos Res., 67-68, 135 ff.
DOSWELL III, C.A., 2005: Progress toward developing a practical societal response to severe convection. Nat.
Hazards Earth System Sci., 6.
EMANUEL, K. A., 1983b: On assessing local conditional symmetric instability from atmospheric soundings. Mon.
Wea. Rev., 111, 2016 ff.
EMANUEL, K. A., 1994: Atmospheric Convection. Oxford University Press, chapter 12.
GEORGIEV, C. G., 1999: Quantitative Relationship between Meteosat WV Data and Positive Potential Vorticity
Anomalies: a Case Study over the Mediterranean. Meteorol. Appl., 6, 97 ff.
HOLTON, J. R., 1992: An Introduction to Dynamic Meteorology. 3d ed. Academic Press
HOSKINS, B.J., McIntyre, M.E., Robertson, A.W., 1985: On the Use and Significance of Isentropic Potential
Vorticity Maps. Quart. Journal Roy. Met. Soc., 111, 877 ff
HOUZE, R. A. Jr., 1993: Cloud Dynamics. Academic Press, section 2.9.1.
HOSKINS, B. J., 1974: The role of potential vorticity in symmetric stability and instability. Quart. J. Roy. Meteor.
Soc, 100, 480 ff.
JANN, A., 2002: Use of a Simple Pattern Recognition Approach for the Detection of Ridge Lines and Stripes.
Meteorol. Appl., 9, 357ff
JASCOURT, S. D, S. S. Lindstrom, C. J. Seman, and D. D. Houghton, 1988: An observation of banded convective
development in the presence of weak symmetric stability. Mon. Wea. Rev., 116, 175 ff.
JONES, S. C., and A. J. Thorpe, 1992: The three-dimensional nature of “symmetric” instability. Quart. J. Roy.
Meteor. Soc., 118, 227 ff.
- 89 -
JOHNS, R. H., and C. A. Doswell III, 1992: Severe local storms forecasting. Wea. Forecasting, 7, 588 ff.
KRENNERT, T., Zwatz-Meise, V., 2003: Initiation of convective cells in relation to water vapour boundaries
in satellite images. Atmos. Res., 67-68, 353 ff.
LIOU, K.N., 2002: An Introduction to Atmospheric Radiation. Second edition, Academic Press, p87.
MARTIN, F., Elizaga, F., Riosalido, R., 1999: The Mushroom Configuration in Water Vapour Imagery and
Operational Applications. Meteorol. Appl., 6, 143 ff.
McCANN, D. W., 1995: Three-dimensional computations of equivalent potential vorticity. Wea. Forecasting, 10,
798 ff.
NORDENG, T. E., 1987: The Effect of Vertical and Slantwise Convection on the Simulation of Polar Lows, Tellus,
39A, 354 ff.
NORMAND, C. W. B., 1931: Graphical indication of humidity in the upper air. Nature, 128, 583.
PIELKE, R.A., Segal, M., 1986: Mesoscale circulations forced by differential terrain heating. Mesoscale
Meteorology and Forecasting, AMS, Ch.22.
SANTURETTE, P., GEORGIEV, C. G., 2005: Weather Analysis and Forecasting – Applying Satellite Water
Vapour Imagery and Potential Vorticity Analysis, Elsevier Academic Press
SCHULTZ, D. M., and P. N. Schumacher, 1999: The use and misuse of conditional symmetric instability. Mon.
Wea. Rev., 127, 2709 ff; Corrigendum, 128, 1573.
SHUTTS, G., 1990: SCAPE charts from numerical weather prediction model fields. Mon. Wea. Rev.,118, 2745 ff.
SKODA, G., 2007: Personal correspondence
SNOOK, J. S., 1992: Current techniques for real-time evaluation of conditional symmetric instability. Wea.
Forecasting, 7, 430 ff.
SZOKE, E.J., Weisman, M. L., Brown, J. M., Caracena, F., Schlatter, T.W., 1984: A Subsynoptic Analysis of the
Denver Tornadoes of 3 June 1981. Mon. Wea. Rev. 112, 790 ff.
WELDON, R. B., Holmes, S. J., 1991: Water Vapour Imagery: Interpretation and Applications to Weather Analysis
and Forecasting, NOAA Technical Report, NESDIS 57, NOAA, US Department of Commerce, Washington D.C.,
213 ff.
WILLIAMS, E., 1991: Comments on “Thunderstorms above frontal surfaces in environments without positive
CAPE. Part I: A climatology.” Mon. Wea. Rev., 119, 2511 ff.
XU, Q., and J. H. E. Clark, 1985: The nature of symmetric instability and its similarity to convective and inertial
instability. J. Atmos. Sci., 42, 2880 ff.
XU, Q., 1986: Conditional symmetric instability and mesoscale rainbands. Quart. J. Roy. Meteor. Soc.,112, 315 ff.
XU, Q., 1992: Formation and evolution of frontal rainbands and geostrophic potential vorticity anomalies. J. Atmos.
Sci., 49, 629 ff.
ZWATZ-MEISE, V., Krennert T., Winkler, R., Zeiner, B., 2005: Manual of Synoptic Satellite Meteorology.
ZAMG, CD-ROM, V6.0
- 90 -