• Bárdossy, A., and H. J. Caspary, 1990: Detection of climate change in Europe by analyzing European atmospheric circulation patterns from 1881 to 1989. Theor. Appl. Climatol., 42 , 155167.

    • Search Google Scholar
    • Export Citation
  • Bica, B., and R. Steinacker, 2005: High resolution analysis over complex terrain by using physical a priori knowledge. Hrvat. Meteor. Cas., 40 , 145148.

    • Search Google Scholar
    • Export Citation
  • Bica, B., R. Steinacker, C. Lotteraner, and M. Suklitsch, 2006: A new concept for high resolution temperature analysis over complex terrain. Theor. Appl. Climatol., in press.

    • Search Google Scholar
    • Export Citation
  • Bisci, C., M. Fazzini, F. Dramis, R. Lunardelli, A. Trenti, and M. Gaddo, 2004: Analysis of spatial and temporal distribution of precipitation in Trentino (Italian eastern Alps): Preliminary report. Meteor. Z., 13 , 183187.

    • Search Google Scholar
    • Export Citation
  • Cacciamani, C., T. Nanni, and S. Tibaldi, 1994: Mesoclimatology of winter temperature and precipitation in the Po Valley of northern Italy. Int. J. Climatol., 14 , 777814.

    • Search Google Scholar
    • Export Citation
  • Chapman, S., 1951: Atmospheric tides and oscillations. Compendium of Meteorology, T. F. Malone, Ed., Amer. Meteor. Soc., 510–530.

  • Daley, R., 1991: Atmospheric Data Analysis. Cambridge University Press, 457 pp.

  • Easterling, D. R., G. A. Meehl, C. Parmesan, S. A. Changnon, T. R. Karl, and L. O. Mearns, 2000: Climate extremes: Observations, modeling and impacts. Science, 289 , 20682074.

    • Search Google Scholar
    • Export Citation
  • Frei, C., and C. Schär, 1998: A precipitation climatology of the Alps from high-resolution rain-gauge observations. Int. J. Climatol., 18 , 873900.

    • Search Google Scholar
    • Export Citation
  • Hantel, M., Ed. 2005: Observed Global Climate. Landolt-Börnstein: Geophysics (Group V), Vol. 6, Springer-Verlag, 567 pp.

  • Kastendeuch, P. P., and P. Kaufmann, 1997: Classification of summer wind fields over complex terrain. Int. J. Climatol., 17 , 521534.

  • Knabl, T., 2004: Klimatologische Auswertung der VERA-Druckanalysen von 1989 bis 2001 (Climatological evaluation of VERA pressure analyses for the period 1989–2001). M.S. thesis, Dept. of Meteorology, University of Vienna, 92 pp.

  • Pielke, A. R., and P. Mehring, 1977: Use of mesoscale climatology in mountainous terrain to improve the spatial representation of mean monthly temperatures. Mon. Wea. Rev., 105 , 108112.

    • Search Google Scholar
    • Export Citation
  • Pöttschacher, W., R. Steinacker, and M. Dorninger, 1996: VERA—A high resolution analysis scheme for the atmosphere over complex terrain. MAP Newsletter, No. 5, 64–65.

  • Rakovec, J., S. Gabersek, and T. Vrhovec, 2004: Relief shapes and precipitation on the south side of the Alps—Part I: Relief characteristics and dry sensitivity simulations. Meteor. Z., 13 , 8390.

    • Search Google Scholar
    • Export Citation
  • Sanders, F., and E. G. Hoffman, 2002: A climatology of surface baroclinic zones. Wea. Forecasting, 17 , 774782.

  • Smith, R. B., and I. Barstad, 2004: A linear theory of orographic precipitation. J. Atmos. Sci., 61 , 13771391.

  • Steinacker, R., 1984: Area–height distribution of a valley and its relation to the valley wind. Beitr. Phys. Atmos., 57 , 6471.

  • Steinacker, R., C. Häberli, and W. Pöttschacher, 2000: A transparent method for the analysis and quality evaluation of irregularly distributed and noisy observational data. Mon. Wea. Rev., 128 , 23032316.

    • Search Google Scholar
    • Export Citation
  • Steinacker, R., and Coauthors, 2006: A mesoscale data analysis and downscaling method over complex terrain. Mon. Wea. Rev., 134 , 27582771.

    • Search Google Scholar
    • Export Citation
  • Trigo, I. F., G. R. Bigg, and T. D. Davies, 2002: Climatology of cyclogenesis mechanisms in the Mediterranean. Mon. Wea. Rev., 130 , 549569.

    • Search Google Scholar
    • Export Citation
  • Vergeiner, I., and E. Dreiseitl, 1987: Valley winds and slope winds: Observations and elementary thoughts. Meteor. Atmos. Phys., 36 , 264286.

    • Search Google Scholar
    • Export Citation
  • Vrhovec, T., J. Rakovec, S. Gabersek, G. Skoki, R. Zabkar, and G. Gegoric, 2004: Relief shapes and precipitation on the south side of the Alps. Part II. Heavy-rain cases during MAP and sensitivity to topography modifications. Meteor. Z., 13 , 201208.

    • Search Google Scholar
    • Export Citation
  • Wanner, H., and M. Furger, 1990: The Bise—Climatology of a regional wind north of the Alps. Meteor. Atmos. Phys, 43 , 105116.

  • Zängl, G., 2004: Upstream effects of an Alpine-scale mountain ridge under various flow angles. Meteor. Z., 13 , 291296.

  • View in gallery

    Physiographic map of the European Alps and the surrounding foreland area. The window displayed corresponds to the analysis domain of the climatology.

  • View in gallery

    Location of major geographic features labeled with terms referred to in the text.

  • View in gallery

    Flow pattern and wind field for a typical case of north Stau (0600 UTC 1 Nov 2001) over the Alpine region. The isobars show reduced sea level pressure in 1-hPa intervals. The arrows represent the wind direction and wind speed at the grid points. Here, 301 crosses (pressure) and 371 squares (wind) show the location of observing stations used for the analysis.

  • View in gallery

    Analysis of MSL pressure on 0300 UTC 31 Mar 2002 without fingerprint technique. The isobars are drawn in 1-hPa intervals. The crosses show the location of observing stations used for the analysis.

  • View in gallery

    Analysis of MSL pressure with fingerprint technique. The isobars are drawn in 1-hPa intervals. The crosses show the location of observing stations used for the analysis. Some stations referred to in the text are labeled and marked with black circles.

  • View in gallery

    Mean MSL pressure field over the Alpine region for January 1995 from ERA-40 analyses. Isobars are plotted in 1-hPa intervals. Data were taken from Hantel (2005).

  • View in gallery

    Mean MSL pressure field over the Alpine region for January 1995 from VERA analyses. Isobars are plotted in 1-hPa intervals.

  • View in gallery

    Mean course (1989–2001) of pressure at the 14/18 (black line, southern Valaisian Alps, Switzerland) and 14/26 (gray line, Swiss Plateau) grid points. The abscissa displays time from 1 to 7 Feb in 6-hourly intervals (dashed vertical lines), and the ordinate shows reduced pressure at the grid points.

  • View in gallery

    Mean 2D divergence of the 10-m wind field as a function of time for different months, averaged over the period of 1980–2001. The solid lines indicate divergence for (top to bottom) January, October, April, and August.

  • View in gallery

    Mean configuration of reduced sea level pressure for all 1200 UTC analyses of May 1989–2001; pmin = 1012.8 hPa, with isobar spacing 0.5 hPa.

  • View in gallery

    Mean configuration of reduced sea level pressure for all 0900 UTC analyses of January 1989–2001; pmax = 1027.2 hPa, with isobar spacing 0.5 hPa.

  • View in gallery

    Daily means of reduced pressure for two grid points for December, averaged over the period of 1989–2001. The first grid point is located in an Alpine Valley (Mur Valley at the border between Styria and Carinthia, Austria); the second one can be assigned to the Po Plain in northern Italy.

  • View in gallery

    Percentage of days with north Stau or south Stau relative to the length of a month (mean value for 1989–2001). Because the lengths of months differ, the number is given in percent of the length of the month.

  • View in gallery

    Pressure configuration for a typical south Stau situation with a pressure difference of 14–17 hPa between two grid points north and south of the Alps (black pluses). Isobars are plotted in 1-hPa intervals. Mean configuration of 67 analyses from the period of 1989–2001.

  • View in gallery

    Regions with an above-average number of grid points showing a WPM. The corresponding grid points are connected by thick black lines. The boldface numbers indicate how often the WPM criterion was met out of 37 984 cases.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 478 462 32
PDF Downloads 345 334 30

Thermally and Dynamically Induced Pressure Features over Complex Terrain from High-Resolution Analyses

View More View Less
  • 1 Department of Meteorology and Geophysics, University of Vienna, Vienna, Austria
© Get Permissions
Full access

Abstract

Within the Vienna Enhanced Resolution Analysis (VERA) Climatology (VERACLIM) project, the complex influence of topographic structures on the spatial distribution of meteorological parameters has been investigated and evaluated climatologically. VERACLIM is aimed to generate a set of high-resolution analyses (lower meso-β-scale) of various meteorological parameters on a climatological basis. It tried to combine both the high spatial resolution provided by the VERA scheme that was used and the high temporal resolution of a comprehensive synoptic dataset of the last two decades, which was retrieved from ECMWF’s Meteorological Archival and Retrieval System (MARS). In the present study, the interpolated fields of reduced pressure of 3-hourly synoptic data over the Alpine region are evaluated climatologically. Using high temporal and spatial resolution, the authors were able to investigate both thermally and dynamically induced mesoscale pressure phenomena such as “Stau,” associated with trans-Alpine flows, blocking by the Alps, and local pressure extrema, as well as thermal lows and thermal high pressure zones. Comparisons are made between the mean course of reduced pressure at given grid points and the averaged divergence of the 10-m wind field in the Alpine region. It is shown that, climatologically, Alpine pumping and thermally induced pressure patterns have a similar frequency and intensity. For the latter ones, the buildup and cutback processes are described. Moreover, the frequency and intensity of pressure-related mesoscale features in the Alpine region over the last decades are investigated.

Corresponding author address: Benedikt Bica, Dept. of Meteorology and Geophysics, University of Vienna, UZA II, Althanstraße 14, 1090 Vienna, Austria. Email: benedikt.bica@univie.ac.at

Abstract

Within the Vienna Enhanced Resolution Analysis (VERA) Climatology (VERACLIM) project, the complex influence of topographic structures on the spatial distribution of meteorological parameters has been investigated and evaluated climatologically. VERACLIM is aimed to generate a set of high-resolution analyses (lower meso-β-scale) of various meteorological parameters on a climatological basis. It tried to combine both the high spatial resolution provided by the VERA scheme that was used and the high temporal resolution of a comprehensive synoptic dataset of the last two decades, which was retrieved from ECMWF’s Meteorological Archival and Retrieval System (MARS). In the present study, the interpolated fields of reduced pressure of 3-hourly synoptic data over the Alpine region are evaluated climatologically. Using high temporal and spatial resolution, the authors were able to investigate both thermally and dynamically induced mesoscale pressure phenomena such as “Stau,” associated with trans-Alpine flows, blocking by the Alps, and local pressure extrema, as well as thermal lows and thermal high pressure zones. Comparisons are made between the mean course of reduced pressure at given grid points and the averaged divergence of the 10-m wind field in the Alpine region. It is shown that, climatologically, Alpine pumping and thermally induced pressure patterns have a similar frequency and intensity. For the latter ones, the buildup and cutback processes are described. Moreover, the frequency and intensity of pressure-related mesoscale features in the Alpine region over the last decades are investigated.

Corresponding author address: Benedikt Bica, Dept. of Meteorology and Geophysics, University of Vienna, UZA II, Althanstraße 14, 1090 Vienna, Austria. Email: benedikt.bica@univie.ac.at

1. Introduction

The scientific investigation of extreme weather events in the Alpine region has generally increased since several catastrophic events took place recently, widely discussed in the mass media because of their high impact (e.g., “century flood 2002,” “European drought summer 2003,” and “avalanche catastrophe in Galtür 1999”). The question whether there is an increasing tendency of catastrophic weather events has been addressed by several authors (e.g., Easterling et al. 2000).

Apart from extreme weather events, several investigations try to study the specific influence of mountains on the atmosphere in both a qualitative and quantitative way. Zängl (2004) investigates the influence of a mountain range on the wind and mass field under varying angles of the flow with respect to the orientation of the ridge. Smith and Barstad (2004) develop a linear theory of orographic precipitation; Bisci et al. (2004) analyze the temporal and spatial precipitation distribution in the Trentino region of Italy. In Slovenia, Rakovec et al. (2004) and Vrhovec et al. (2004) investigate the correlation between topographic features and precipitation on the southern side of the Alps.

Precipitation climatology in the Alps is certainly of major importance regarding flooding and other natural disasters. However, if we try to gain insight into the climatology of dynamic and kinematic features that cause these events, like divergence, flow splitting, or mesoscale vortices, we can detect regions where heavy precipitation tends to start or intensify or storms occur. Thus, the climatological evaluation of parameters like the mean temperature (e.g., Pielke and Mehring 1977; Cacciamani et al. 1994), air pressure, or surface wind field (Kastendeuch and Kaufmann 1997; Wanner and Furger 1990; Bárdossy and Caspary 1990) is of vital importance.

As far as the scale of the phenomena investigated is concerned, the majority of climatological investigations focus on observed long- or short-term variations on a large to global scale, on climate scenarios under global change aspects, or on the performance of climate models. Investigations on a regional to local scale and their interpretation, however, are not found as frequently in literature. An example for a work in this field is Sanders and Hoffman (2002), who produced a climatological description (“climatology”) of baroclinic zones over the United States. Trigo et al. (2002) investigated cyclogenic processes in the Mediterranean on a climatologic base.

Mesoscale climatologies are produced either under increased spatial resolution with low temporal resolution or highly resolved temporal information is being presented for single stations, that is, with low spatial resolution. This is due to the fact that long time series of data are usually only available two or three times a day (e.g., the Mannheim observing hours), which prohibit a reliable representation of (aperiodic) diurnal variations. What is largely lacking in climatology is an evaluation of atmospheric fields with high temporal and high spatial resolution. The source for such studies is the synoptic data, which are available for a long time with a 3-hourly resolution. The lack of spatial resolution has to be compensated by an appropriate downscaling technique, because over the Alpine region the mean station distance lies on the order of 50 km.

The Vienna Enhanced Resolution Analysis (VERA) is a high-resolution analysis scheme (Steinacker et al. 2006; Bica and Steinacker 2005; Steinacker et al. 2000; Pöttschacher et al. 1996) that is based on the variational principle and does not need any first-guess fields, that is, it is data self-consistent. Furthermore, it includes a sophisticated data quality control tool. The downscaling approach consists of the “fingerprint” technique, that is, a utilization of physical a priori knowledge about the high-resolution fields over complex terrain. In this investigation, VERA has been used to interpolate a 13-yr set of 3-hourly synoptic data of reduced pressure to a regular grid.

The present paper is organized as follows: section 2 contains detailed information on the VERA analysis algorithm and the downscaling method in particular. In section 3, the VERA Climatology (VERACLIM) project that formed the framework for the present investigation is presented briefly. Section 4 contains information about the used synoptic dataset. In section 5, the evaluation of thermally and dynamically induced mesoscale pressure features, as retrieved from a comprehensive set of analyzed fields is discussed in detail. Conclusions are drawn in section 6.

2. VERA

The VERA interpolation method is used for transferring data to a regular high-resolution grid over complex terrain. It is intended for the application both on scalar and vector quantities. All modules of VERA run independent of a weather prediction model or other first-guess fields, using data self-consistency. The basic philosophy of VERA is to use the physical a priori knowledge about typical atmospheric structures in the atmospheric boundary layer and lower troposphere over complex topography for downscaling purposes.

In formal terms, the analysis can be divided into two different “process steps”: 1) the interpolation step, used for data quality control and for interpolation on a (regular) grid (Pöttschacher et al. 1996; Steinacker et al. 2000), and 2) downscaling using the “fingerprint” technique (Steinacker et al. 2006; Bica and Steinacker 2005).

In this work, VERA is exclusively used for analyses within the domain shown in Fig. 1. This domain covers the entire Alpine region (see section 4). In Fig. 2, the location and local names of some main geographical features, which will be referred to later in the text, are depicted.

a. Interpolation method

The VERA analysis method is based on the variational principle applied to higher-order spatial derivatives, which are computed from overlapping finite elements (Steinacker et al. 2000). For a scalar quantity Φ, the cost functional J(Φ) as the weighted sum of squared spatial derivatives on two dimensions is minimized:
i1558-8432-46-1-50-e1
where γi stands for the weight of the different (i) spatial derivatives SDi and σ denotes the area of the finite elements used for the derivation. This method minimizes the curvature and/or gradient of scalar fields and the kinematic quantities of vector fields, respectively. It is equivalent to the penalty function of thin-plate smoothing splines (Daley 1991).

b. Data quality control

In contrast to other operational quality control procedures we do not need to deal with any first-guess fields or with physical or climatological limits. We use the method with a decision-making algorithm to find out how strong an effect of “calming down” (minimization of curvature) the surface can be achieved by changing the value of one single station value. Hence, we get a deviation value for each station and each time, which can be further investigated statistically to find out gross errors, biases, mesoscale signals, and the meteorological noise.

c. Fingerprint technique

Because of the irregular spacing of observations and their specific situation with respect to topography (i.e., stations in valleys or basins, on slopes, passes, and mountains) the analyzed field may be very rough. Consequently, small-scale structures produced by topography cannot sufficiently be resolved by conventional analysis schemes, which tend to treat this roughness as noise and smooth it out. But mountainous topography actually produces small-scale structures of considerable amplitudes. Following Steinacker et al. (2006), at each grid point x, the analyzed variable [e.g., pressure p(x)] can be assumed to be composed of a synoptic part pS(x) and a part pT(x) that is due to orographic influence:
i1558-8432-46-1-50-e2
The coefficient c(x) is a variable weighting factor; pS(x) and c(x) are not a priori known; pT(x), however, is a precalculated, idealized field that is given with higher spatial resolution than the observational data. The scaling factor c(x) varies in space and from case to case. This is necessary because pT(x) represents values calculated for ideal conditions, but the actual influence of the topography on p(x) varies, resulting from diurnal, seasonal, and weather-type dependence. Formally, the fingerprint field is comprehended to the analysis as follows: Expressing pS(x) from Eq. (2) in terms of pT(x) and c(x) and substituting it for Φ(x) in Eq. (1) yields a system of linear equations that is then solved for c(x) and p(xj), where j denotes the unknown nodes. The synoptic part pS(x) is not necessarily of interest.

1) Calculation of the fingerprints

The task is now to make some assumption on the structure of the fingerprint field pT(x). In contrast to first-guess or background fields that are commonly used for conventional interpolation methods, our fingerprints are physical models that are based on known properties of meteorological fields over complex terrain.

The modification of the atmosphere by a mountain massif has been split up into two different physical processes so far—thermal effects resulting from different heating or cooling of the atmosphere over mountains (e.g., thermal low or thermal high pressure zones), and dynamic influences like blocking or leeside effects. These features of the Alpine atmosphere may be modeled at scales far below those resolved by observations, if a very high-resolution topographic dataset (on the order of 1 km) is available. For the analysis, individual observations determine the local intensity [i.e., the value of c(x)] of the thermal and/or dynamic topographic fingerprint (Steinacker et al. 2006).

Thermal fingerprint

The basic idea of the thermal fingerprint is to calculate the spatial temperature distribution for idealized radiative heating or cooling processes over mountainous terrain. The hydrostatic relation is then used in order to estimate the corresponding pressure field. Under undisturbed daytime conditions, both the reduced air volume over topography and a free volume over a plain will experience the same radiative flux at the top of the volumes. If the surface albedo, the sensible and latent heat fluxes, and the ground heat flux are assumed to be equal over the mountains and the plains, and further if the energy gain or loss is evenly distributed within the volume, the ratio of the overall heating or cooling will be exactly given by the ratio of the air volumes, which can easily be calculated from topographic information.

The distribution of energy gain or loss, however, will hardly ever be even but rather will strongly depend on the static stability and the area–height distribution. Thus, the excess of the diurnal amplitudes of temperature over mountains as compared with the plains cannot be given by a constant ratio. Steinacker (1984) describes this effect for stably stratified air in valleys. The higher the stability and the more convex the area–height distribution of a valley, the larger the ratio of heating or cooling between valley and plain. The thermal separation of stably stratified layers in valleys is evident from observations. The temperature difference between the slope layer, where the diabatic heating or cooling takes place, and the “free” valley atmosphere remains roughly constant during the daytime (upslope winds) or nighttime (downslope winds). This means that the amount of heating/cooling in a slope layer is immediately transferred to the free atmosphere layer by layer by adiabatic heating/cooling. This concept has been introduced by Vergeiner and Dreiseitl (1987).

Dynamic fingerprint

When an airflow impinges on mountains, it is partly deflected around and partly deflected over the obstacle, or even sometimes becomes blocked. The entirety of these phenomena is termed “Stau” (German for congestion or traffic jam) herein. Figure 3 shows the pressure field (MSL) and the corresponding wind field for a typical case of north Stau (0600 UTC 1 November 2001) over the Alpine region. It can be seen very clearly from the wind arrows that, in this case, air masses are deflected around the Alpine arc to both sides of the windward pressure maximum.

Stau basically depends on the scale of the mountain, the static stability, and the speed of the air. A measure for this is the Froude number. Flow deflection or blocking of the air needs a corresponding pressure perturbation. For mountains of the size of the Alps a positive pressure perturbation can be observed reliably on the windward side (Stau) and a negative pressure perturbation on the leeward side. These pressure perturbations can be explained hydrostatically, that is, on the windward/leeward side we see a negative/positive temperature perturbation. These perturbations may then be used as an idealized dynamic fingerprint.

2) Fingerprints—A case study

This example documents how the interpolated field of a thermal high over the Alps is changed if a thermal fingerprint is used. Figure 4 shows the interpolated field of mean sea level pressure on 0300 UTC 31 March 2002 (approximately 1.5 h before sunrise). This interpolation was carried out without the use of fingerprints. The night was clear with weak winds in many parts of the Alpine region. There is a weak pressure gradient with a thermal high showing two centers—one in the eastern Alps and one in the western part. The highest pressure values reach about 1025 hPa.

If the analysis employs a thermal fingerprint we obtain an analyzed pressure field that appears to be much more structured (Fig. 5). The patterns over the plains have not significantly changed, whereas over the mountains, the centers of the high are more pronounced in shape and intensity (up to 1027 hPa, especially over the western and the Dinaric Alps, where the station density is very low). The pressure at the stations in the valleys (e.g., the stations annotated with 1, 5, and 8 in Fig. 5) is higher (up to 1025.9 hPa) than in the foreland (station 3, 1020.3 hPa). The lack of details in the analysis, which is due to the low station density in large areas of the western Alps, is compensated by the physical a priori knowledge of the thermal fingerprint; Fig. 5 shows more details in the analysis by enhancing the size, shape, and intensity of the thermal high. In the eastern Alps, the station density is higher than in the west. Note that the area of the thermal high is well equipped with stations. Therefore, its intensity and shape is well resolved by the measurements. In this case, only details of the thermal high are improved, like its extension toward the northeast. Around the three-corner point of Germany, Austria, and the Czech Republic, the shape of the thermal high is enhanced, too. The situation over the Balkan Mountains, where the station density is again very low, is similar to that of the western Alps—there is a weak signal of a thermal high that is interpreted by the fingerprint. The same applies for the Adige Valley and other major valleys.

3) Quality evaluation

To investigate the performance of the fingerprint technique objectively, a cross validation was carried out. For the analyses in Figs. 4 and 5, data from a set of 260 meteorological stations were used. For each case, one station datum after the other was left out and a value of the pressure field was interpolated to the location of the suppressed station. The difference between the measured station values and the interpolations, that is, the interpolation error, was statistically evaluated for the two fields (Table 1).

In general, the analysis without fingerprints yields larger interpolation errors than the fingerprint method. For example, the maximum difference between the measurements and the interpolated values occurs at the station in Sarajevo (9.82 hPa), because mesoscale patterns in the MSL pressure field are not resolved by the few stations in that area. Because of its specific location at the southeastern edge of the analysis domain, this grid point is susceptible to extrapolation errors if no terrain information is added to the analysis (Steinacker et al. 2000). Using the fingerprint technique, however, reduces the interpolation error to 2.09 hPa in Sarajevo. Table 1 confirms this result: the RMSE and the standard deviation of the interpolation errors are about 25% lower if the fingerprint is included to the analysis. The range of the differences is reduced by about 75%.

Not all of the examined stations are situated in areas with strong topographic influence. Statistics concerning the stations in mountainous terrain solely would presumably yield even more impressive results.

d. VERA versus ERA-40

To assess the potential of the VERA analysis method in contrast to other previously known schemes, we compared the European Centre for Medium-Range Weather Forecasts (ECMWF) 40-yr Reanalyses (ERA-40) pressure reanalysis and the mean VERA-analyzed pressure fields for January 1995 over the Alpine region. The ERA-40 data, which were taken from Hantel (2005), are given globally on a regular 1° grid (Fig. 6). In contrast, the VERA analysis (Fig. 7) was performed on a regular 20-km grid. Because of the higher resolution, the VERA-analyzed field appears to be much more structured than the smooth ERA-40 field. Thermally induced high pressure zones are very well resolved and can be seen over the eastern and western Alps, the Dinaric Alps, and the Apennine Mountains.

While the mean pressure of 1019.07 hPa is exactly the same for both fields, the pressure range is slightly higher in the VERA field (Table 2). If the gridpoint information is used to calculate absolute values of pressure gradients, we find that VERA significantly enhances gradients over mountainous topography and thus leads to a more realistic representation of terrain-induced pressure features.

3. VERACLIM

Mesoscale phenomena, such as pressure gradients across the Alpine barriers in connection with trans-Alpine flows (gap flows and Foehn), blocking by the Alps, intense isallobaric features, connected with wind storms, flow-splitting points, mesovortices, cold-air pools, frontal passages, etc., can be detected by the VERA system (Knabl 2004; Steinacker et al. 2006). Together with an appropriate data archive like the Meteorological Archival and Retrieval System (MARS; or ERA-40 archive) of ECMWF, the VERACLIM system leads to a synoptic climatological evaluation of the mesoscale phenomena mentioned above. It tries to benefit from both the high temporal resolution provided by the use of 3-hourly synoptic data and the high spatial resolution provided by the use of the fingerprint technique.

One major advantage of the fields generated in the framework of VERACLIM is that all information is given at grid points, that is, at any position within the complex topography of the investigated area and without any breaks because they do occur in time series from single-station observations.

Three of the main objectives of VERACLIM can be summarized in the following way:

  1. establishment of an Alpine-wide temporal and spatial high-resolution climatology of air pressure, temperature, humidity, and wind, covering ideally the time span of a normal climate period and mainly based on data collected by the synoptic network;

  2. detailed investigation of mesoscale features of various meteorological parameters over complex topography; and

  3. statistical exploration of the characteristics of mesoscale flow patterns and changes.

VERACLIM is embedded in the scope of the international Mesoscale Alpine Program (MAP).

4. Data

The data for this investigation were retrieved from ECMWF’s MARS. We were using a 13-yr set of 3-hourly synoptic data, starting on 0000 UTC 1 January 1989 and ending on 2100 UTC 31 December 2001. This gives a total of 37 984 sets of observations, which cover all European countries. During this 13-yr period, only four dates of observations were missing in the MARS archive: 0000 UTC 1 March 1995, 0000 UTC 4 January 1997, 0000 UTC 1 March 1997, and 1500 UTC 25 March 1992.

To avoid inhomogeneities in the raw dataset, these gaps were filled up by interpolating the neighboring observations to the missing date of observation. The synoptic data were then interpolated to a regular 20-km grid with 64 grid points (1260 km) in the zonal direction and 43 grid points (840 km) in the meridional direction, by making use of the VERA algorithm, including quality control and the use of thermal and dynamical fingerprints. The obtained grid covers the entire Alpine region (latitude φ = 42.86°–49.80°N, longitude λ = 5.02°–18.91°E). All together there is a complete 13-yr time series of reduced atmospheric pressure for each grid point in 3-hourly intervals on hand.

5. Results

a. Thermally induced pressure features

As a first step of data evaluation, the arithmetic means of reduced pressure were computed for different months and times of day (0000, 0300, 0600, 0900, 1200, 1500, 1800, 2100 UTC) for the entire 13-yr period. Figure 8 shows the mean course of reduced pressure for the first 7 days of February at the 14/18 (a valley floor in the southern Valaisian Alps, Italy, 388 m MSL) and 14/26 (Swiss plateau, 454 m MSL) grid points for this period.

At first glance, the regularity of the course of pressure over this 7-day period attracts attention and gives evidence of the higher amplitude of the diurnal pressure variation over the Alps as compared with the plains. What can also be seen easily is a superimposed trend of (synoptically) decreasing pressure by the order of 5 hPa over the displayed period, which means that on the 13-yr average, pressure decreased at this grid point during the first 7 days of February. As far as daily variations are concerned, there is a pronounced daily minimum of reduced pressure at 1500 UTC. After sunset (1600 UTC, 1700 local time), pressure increases steadily to a local maximum at about 0000 UTC. A second local maximum, which is partially higher than the midnight maximum, is reached in the morning hours around 0900 UTC.

The double oscillation in the daily course of pressure, often referred to as “atmospheric tides” (Chapman 1951), is particularly well pronounced on days with strong incoming solar radiation and nights with strong outgoing net radiation without any synoptic disturbances, because warming during daytime and cooling during nighttime is performed very effectively.

The result that was clearly retrieved from the data, namely, a larger pressure oscillation amplitude at intra-Alpine grid points as compared with extra-Alpine grid points, can be related to thermal or Alpine pumping. This is a wind circulation from the plains toward the mountains during daytime and vice versa during night that is commonly seen in the data of valley and foreland stations of the Alps. Alpine pumping also becomes evident if we look at the area mean divergence of the 10-m wind field (Fig. 9). The average mass flux (1980–2001) across an imaginary boundary (the solid line in Fig. 15) around the Alps shows a distinct wavelike pattern in the course of the day, with different amplitudes for different months. During the night, and especially in the cold seasons, there is on average a pronounced divergence in the wind field over the Alpine region, whereas for daytime conditions and especially in summer the mass flux becomes clearly convergent. Mean divergence ranges from about −1.3 × 10−6 to +2.2 × 10−6 s−1, which is approximately one order of magnitude less than the typical synoptic-scale divergence of the atmospheric wind field. However, it has to be taken into consideration that the divergence in Fig. 9 was computed as an average of all observations of 10-m wind for a 22-yr period of time, without filtering out clear and undisturbed days with a weak synoptic pressure gradient.

1) Thermal high and thermal low pressure zones

An example for a typical mesoscale feature in the pressure field that is due to thermal influence is a thermal high or thermal low pressure zone. Figure 10 shows the mean configuration of pressure over the Alpine region for the month of May, as computed as an average over all 1200 UTC observations of May in the averaging period (31 days times 13 yr yields 403 observations available for averaging). To inhibit a too-strong influence of pressure reduction, all sea level pressure fields of VERA are based on observations of stations below 750 m MSL. Moreover, the analyses were performed on the so-called minimum topography, that is, a special topography of the lowlands and valleys where high mountain peaks are smoothed out by considering the minimum height of topography within a given radius of influence (Bica et al. 2006). This was done in order to prevent the calculated fields from just being a one-to-one image of topography.

We observe two main centers of relatively low pressure—the eastern center with a larger spatial extension reaches from the Rhine Valley to the eastern border of the Alps, and the western center covers the French Alps southwest of the Ticino Alps. In between, over the St. Gotthard region right over the center of Switzerland, we observe a zone with a less pronounced thermal low, which we suggest should be called the “Alpine wasp waist,” because of the specific configuration of the isobars in this region. This phenomenon can be explained by the fact that on both sides of the Gotthard Pass comparatively short valleys lead directly toward the plains in the north and south. The short time constant of the valley wind system in such valleys as compared with the long ridge-parallel main Alpine valleys leads to a dominance of advective processes and hence a reduction of local heating and cooling. Heat lows in other summer months except for May do not differ qualitatively from Fig. 10.

The typical cold high pressure zone (Fig. 11) produced in a winter night was computed from all 0900 UTC observations of the month of January. Again, we find two centers of relatively high pressure over the western and the eastern part of the Alps. In the eastern part (pmax=1027.2 hPa), pressure is slightly higher than in the western part (1026.5 hPa). In addition, we observe a deformation of the isobars in the south Tyrolean Adige Valley. Apparently, the specific shape, expressed, for example, by the area–height distribution (Steinacker 1984) of this broad valley, creates reduced heating and cooling rates. Interestingly, high-resolution prognostic models often show a corresponding pattern in the Adige Valley. The region with the highest pressure for 0900 UTC in January is located in the upper Mur Valley in Styria, Austria. Also, the St. Gotthard region and the Swiss part of the Rhine Valley seem to form an unsuitable basin or valley for the formation of cold-air pools. Exactly over the St. Gotthard region, the thermal high is very shallow.

In spring and autumn, thermally induced pressure features are not very well pronounced, because dynamical pressure configurations produced by Stau are superimposed over thermal ones, that is, the thermal forcing is missing. This can also be seen from the lower amplitudes for October and April in Fig. 9.

2) Temporal evolution and degeneration

The formation of a typical thermal low starts first in the western Alps and later on in the eastern Alps. Also, the pressure in the western center is lower than in the eastern one. In the evening, the degeneration of the thermal low starts first in the eastern Alps because the sun sets earlier there. A time lag can also be observed during the formation of a thermal high, which is formed later in the west than in the eastern Alps; however, the difference in intensity is similar in that case.

The time lag can certainly not be explained by the duration of insolation solely. The authors assume that it is also due to the fact that the Alps are significantly broader (in term of cross-sectional area) in the east when compared with the west, which results in a longer time constant for the buildup and cutback processes in the east.

b. Dynamically induced pressure features

Again, we consider time series at a grid point, but this time we make diurnal variations disappear by calculating daily mean values of atmospheric pressure from the eight available observations per day (0000, 0300, 0600, 0900, 1200, 1500, 1800, and 2100 UTC), which were calculated as a mean over the period of 1989–2001 themselves. In Fig. 12, the daily mean (1989–2001) of reduced pressure for December is plotted for two specific grid points. The first grid point is located in an Alpine valley (Mur Valley at the border between Styria and Carinthia, Austria); the second one can be assigned to the Po Plain in northern Italy.

There is a surprisingly well-pronounced periodicity of about 1 week in decrease and increase of air pressure during this month over the averaged period of time. While Fig. 9 shows thermally induced variations in pressure at a given grid point, Fig. 12 gives an idea of variations of the order of several days, which can be assigned to transient pressure systems, or, in general terms, to synoptic influence.

Because of the special shape of the Alpine arc, air masses are deflected in various ways, depending on the flow direction relative to the obstacle (Fig. 3). In this investigation we put our main focus on synoptic-scale flows approaching from the north or south because they have a significant impact on daily life in the Alpine region (e.g., precipitation resulting from Stau effects and Foehn on the leeward side of the mountain range).

1) Mean number of days with north Stau or south Stau

To detect Stau-like pressure configurations objectively, we sorted out all pressure analyses showing a pressure difference |Δp| > 6 hPa between two 360-km distant grid points (indicated by plusses in Fig. 14)—one in the north of the Alps, the other one in the south. Figure 13 displays the relative frequency of days with Stau for each month of the year, as an average over the 13-yr period. The maximum of the total of days with north and south Stau occurs during winter, whereas Stau in summer is a comparatively rare phenomenon. What is more, north Stau dominates in frequency over south Stau during the whole year, but its dominance is strongest in summer. As far as the strength of the trans-Alpine pressure gradient is concerned, for north Stau, the maximum pressure difference between the two selected grid points reached more than 22 hPa in 1995, and for south Stau more than 19 hPa in 1989 and 1995. An annual cycle can be observed, which is quite on track with the frequency of events, that is, the maximum intensity of Stau events occurs in summer and winter. Table 3 shows statistics of the relative frequency of different classes of pressure gradients between the two selected grid points.

2) Mean isobar configuration for Stau

In a second step, the analyzed fields with a Stau-like pressure configuration were divided into different classes, depending on direction and strength of the trans-Alpine pressure gradient. In Fig. 14, a typical south Stau situation (with Foehn blowing north of the Alps) is shown. This field was computed as a mean over 67 observations of south Stau with a pressure difference of 14–17 hPa between the north and south. In these cases, the area with the highest pressure can be found in the Po Plain with a center in Lombardy. A very well pronounced trough, reaching from southeastern Austria over Slovenia to Croatia gives evidence of a flow around the southeastern edge of the Alps. As a consequence, strong winds blowing up the Danube Valley in eastern Austria can be observed reliably in such situations.

3) Local pressure extrema

Climatological evaluation of atmospheric pressure at a single grid point relative to the neighboring grid points allows us to detect regions where mean pressure is relatively higher (lower) and where the flow consequently exhibits diffluence (confluence) in the presence of a Stau-type flow regime. To detect such windward pressure maxima (WPM; in the case of diffluence) or leeward pressure minima (LPM; in the case of confluence), we draw an imaginary line around the Alps (Fig. 15), connecting grid point to grid point. In contrast to other mountain ranges, the European Alps exhibit very well pronounced and sharp edges; they do not extend at all beyond the enclosing line chosen for our investigation. For this reason, the restriction to one single line appears acceptable. For a specific analysis, a grid point is counted as WPM if its pressure is higher than at the closest “neighbors” along the line and as LPM if it is lower. Again, all observations of the 1989–2001 period were used.

We found two distinct regions (Fig. 15) where the WPM criterion was predominately met: the first one is located in the Swiss Berner Oberland, around grid point 13/24, where the WPM criterion was met 5863 times out of 37 984 cases; the second one lies in the southeast of the Valaisian Alps around grid point 16/18, where the criterion was met 5815 times. There are some more regions that serve as WPM areas, namely, the Säntis in northeastern Switzerland, the Chiemgauer Alps, the Mangfallgebirge at the border between Germany and Austria, and the region south of the Julian Alps in Friuli (Italy) and Slovenia, all of which have more than 4000 WPM cases in the observing period. There is a good general collocation of areas with frequent WPM cases to areas with maximum precipitation along the Alpine rim as shown, for example, in Frei and Schär (1998).

Regions with the smallest number of WPM events are the western border of the Alpine ridge, the French pre-Alps, the region east of Salzburg in Austria, and Lombardy, at the border to Veneto, Italy.

On the other hand, grid points serving as LPM points (not shown in Fig. 15) are located in areas were confluence is frequently observed. These regions are often equivalent to regions where Foehn penetrates. Important LPM regions can be found in the northern Rhine Valley, the French Côte d’Azur, which may be favored by the geometric shape of the Alps, or the southern exit of the Adige Valley in the region of Lago di Garda.

6. Conclusions and outlook

In the framework of the VERACLIM project, a set of 3-hourly synoptic data was evaluated climatologically by making use of the high-resolution analysis scheme VERA. The fingerprint technique, that is, the utilization of physical a priori knowledge about the spatial distribution of meteorological parameters over complex terrain, was used for downscaling purposes.

One major advantage of the fields generated in the course of VERACLIM is that all information is given at grid points, at any position within the complex topography of the Alps and without any breaks because they do occur in time series from single-station observations.

The main conclusions from this paper can be summarized as follows:

  1. A distinct oscillation in daily time series of Alpine grid points, which is particularly pronounced on days with strong incoming solar radiation and outgoing net longwave radiation, gives evidence of the local forcing of atmospheric tides.

  2. The observed periodic wind circulation between the forelands and the mountains, often referred to as Alpine pumping, is in good accordance with the course and amplitude of reduced pressure at intra- and extra-Alpine grid points and with the divergence of the low-level wind field. Alpine pumping, in terms of divergence, is much more pronounced in the summer months than during wintertime.

  3. Both the mean pressure configuration for a thermal high and a thermal low pressure zone show two centers of high/low pressure over the eastern and the western part of the Alps, with a depression over the St. Gotthard region. The time of generation and of the cutback of thermally induced pressure zones differs for the eastern and the western maximum/minimum. Generally, the western low pressure center generates earlier and degenerates later than the eastern one, and it has a slightly stronger intensity. A similar time lag can be observed during the development of a thermal high, which forms in the eastern Alps first.

  4. As far as dynamically induced pressure features are concerned, there is a minimum of typical north or south Stau situations in the Alpine region from May to September. On average, the number of days that meet a criterion for south Stau is exceeded by the number of days with north Stau.

  5. There are distinct regions around the Alpine arc where, in presence of a Stau-type flow regime, windward pressure maxima (leeward pressure minima) in connection with diffluence (confluence) are predominately observed. Typical regions with pressure maxima are found in the Swiss Berner Oberland and in the Valaisian Alps, the Ticino, and the Julian Alps. Frequent confluence is observed in the northern Rhine Valley, the French Côte d’Azur, and the exit of the Adige Valley in Veneto, Italy.

Because of a number of major gaps in the synoptic data archives before 1989 and especially before 1980, our investigations were mostly limited to a period of 13 or 22 yr (1980/1989–2001).

These gaps are mainly due to the non–World Meteorological Organization (WMO)–conformed coding of synoptic data performed by several national weather services in the years before 1989.

Despite the fact that the evaluated period was short in comparison with usual climatological investigations or the 30-yr climate normal period defined by the WMO, our results generally coincide with what has been expected from previous investigations. On the other hand, new and improved insight is delivered into some pressure-related mesoscale features over complex terrain. In a new attempt, it is planned to fill up the gaps in the used data archives and finally to establish a comprehensive set of synoptic data of 30 yr at least. This would provide an even better basis for high-resolution climatological evaluations.

Acknowledgments

The VERACLIM project was funded by the Austrian Science Fund FWF (P15079). Synoptic data were supplied by the ECMWF through a special project.

REFERENCES

  • Bárdossy, A., and H. J. Caspary, 1990: Detection of climate change in Europe by analyzing European atmospheric circulation patterns from 1881 to 1989. Theor. Appl. Climatol., 42 , 155167.

    • Search Google Scholar
    • Export Citation
  • Bica, B., and R. Steinacker, 2005: High resolution analysis over complex terrain by using physical a priori knowledge. Hrvat. Meteor. Cas., 40 , 145148.

    • Search Google Scholar
    • Export Citation
  • Bica, B., R. Steinacker, C. Lotteraner, and M. Suklitsch, 2006: A new concept for high resolution temperature analysis over complex terrain. Theor. Appl. Climatol., in press.

    • Search Google Scholar
    • Export Citation
  • Bisci, C., M. Fazzini, F. Dramis, R. Lunardelli, A. Trenti, and M. Gaddo, 2004: Analysis of spatial and temporal distribution of precipitation in Trentino (Italian eastern Alps): Preliminary report. Meteor. Z., 13 , 183187.

    • Search Google Scholar
    • Export Citation
  • Cacciamani, C., T. Nanni, and S. Tibaldi, 1994: Mesoclimatology of winter temperature and precipitation in the Po Valley of northern Italy. Int. J. Climatol., 14 , 777814.

    • Search Google Scholar
    • Export Citation
  • Chapman, S., 1951: Atmospheric tides and oscillations. Compendium of Meteorology, T. F. Malone, Ed., Amer. Meteor. Soc., 510–530.

  • Daley, R., 1991: Atmospheric Data Analysis. Cambridge University Press, 457 pp.

  • Easterling, D. R., G. A. Meehl, C. Parmesan, S. A. Changnon, T. R. Karl, and L. O. Mearns, 2000: Climate extremes: Observations, modeling and impacts. Science, 289 , 20682074.

    • Search Google Scholar
    • Export Citation
  • Frei, C., and C. Schär, 1998: A precipitation climatology of the Alps from high-resolution rain-gauge observations. Int. J. Climatol., 18 , 873900.

    • Search Google Scholar
    • Export Citation
  • Hantel, M., Ed. 2005: Observed Global Climate. Landolt-Börnstein: Geophysics (Group V), Vol. 6, Springer-Verlag, 567 pp.

  • Kastendeuch, P. P., and P. Kaufmann, 1997: Classification of summer wind fields over complex terrain. Int. J. Climatol., 17 , 521534.

  • Knabl, T., 2004: Klimatologische Auswertung der VERA-Druckanalysen von 1989 bis 2001 (Climatological evaluation of VERA pressure analyses for the period 1989–2001). M.S. thesis, Dept. of Meteorology, University of Vienna, 92 pp.

  • Pielke, A. R., and P. Mehring, 1977: Use of mesoscale climatology in mountainous terrain to improve the spatial representation of mean monthly temperatures. Mon. Wea. Rev., 105 , 108112.

    • Search Google Scholar
    • Export Citation
  • Pöttschacher, W., R. Steinacker, and M. Dorninger, 1996: VERA—A high resolution analysis scheme for the atmosphere over complex terrain. MAP Newsletter, No. 5, 64–65.

  • Rakovec, J., S. Gabersek, and T. Vrhovec, 2004: Relief shapes and precipitation on the south side of the Alps—Part I: Relief characteristics and dry sensitivity simulations. Meteor. Z., 13 , 8390.

    • Search Google Scholar
    • Export Citation
  • Sanders, F., and E. G. Hoffman, 2002: A climatology of surface baroclinic zones. Wea. Forecasting, 17 , 774782.

  • Smith, R. B., and I. Barstad, 2004: A linear theory of orographic precipitation. J. Atmos. Sci., 61 , 13771391.

  • Steinacker, R., 1984: Area–height distribution of a valley and its relation to the valley wind. Beitr. Phys. Atmos., 57 , 6471.

  • Steinacker, R., C. Häberli, and W. Pöttschacher, 2000: A transparent method for the analysis and quality evaluation of irregularly distributed and noisy observational data. Mon. Wea. Rev., 128 , 23032316.

    • Search Google Scholar
    • Export Citation
  • Steinacker, R., and Coauthors, 2006: A mesoscale data analysis and downscaling method over complex terrain. Mon. Wea. Rev., 134 , 27582771.

    • Search Google Scholar
    • Export Citation
  • Trigo, I. F., G. R. Bigg, and T. D. Davies, 2002: Climatology of cyclogenesis mechanisms in the Mediterranean. Mon. Wea. Rev., 130 , 549569.

    • Search Google Scholar
    • Export Citation
  • Vergeiner, I., and E. Dreiseitl, 1987: Valley winds and slope winds: Observations and elementary thoughts. Meteor. Atmos. Phys., 36 , 264286.

    • Search Google Scholar
    • Export Citation
  • Vrhovec, T., J. Rakovec, S. Gabersek, G. Skoki, R. Zabkar, and G. Gegoric, 2004: Relief shapes and precipitation on the south side of the Alps. Part II. Heavy-rain cases during MAP and sensitivity to topography modifications. Meteor. Z., 13 , 201208.

    • Search Google Scholar
    • Export Citation
  • Wanner, H., and M. Furger, 1990: The Bise—Climatology of a regional wind north of the Alps. Meteor. Atmos. Phys, 43 , 105116.

  • Zängl, G., 2004: Upstream effects of an Alpine-scale mountain ridge under various flow angles. Meteor. Z., 13 , 291296.

Fig. 1.
Fig. 1.

Physiographic map of the European Alps and the surrounding foreland area. The window displayed corresponds to the analysis domain of the climatology.

Citation: Journal of Applied Meteorology and Climatology 46, 1; 10.1175/JAM2418.1

Fig. 2.
Fig. 2.

Location of major geographic features labeled with terms referred to in the text.

Citation: Journal of Applied Meteorology and Climatology 46, 1; 10.1175/JAM2418.1

Fig. 3.
Fig. 3.

Flow pattern and wind field for a typical case of north Stau (0600 UTC 1 Nov 2001) over the Alpine region. The isobars show reduced sea level pressure in 1-hPa intervals. The arrows represent the wind direction and wind speed at the grid points. Here, 301 crosses (pressure) and 371 squares (wind) show the location of observing stations used for the analysis.

Citation: Journal of Applied Meteorology and Climatology 46, 1; 10.1175/JAM2418.1

Fig. 4.
Fig. 4.

Analysis of MSL pressure on 0300 UTC 31 Mar 2002 without fingerprint technique. The isobars are drawn in 1-hPa intervals. The crosses show the location of observing stations used for the analysis.

Citation: Journal of Applied Meteorology and Climatology 46, 1; 10.1175/JAM2418.1

Fig. 5.
Fig. 5.

Analysis of MSL pressure with fingerprint technique. The isobars are drawn in 1-hPa intervals. The crosses show the location of observing stations used for the analysis. Some stations referred to in the text are labeled and marked with black circles.

Citation: Journal of Applied Meteorology and Climatology 46, 1; 10.1175/JAM2418.1

Fig. 6.
Fig. 6.

Mean MSL pressure field over the Alpine region for January 1995 from ERA-40 analyses. Isobars are plotted in 1-hPa intervals. Data were taken from Hantel (2005).

Citation: Journal of Applied Meteorology and Climatology 46, 1; 10.1175/JAM2418.1

Fig. 7.
Fig. 7.

Mean MSL pressure field over the Alpine region for January 1995 from VERA analyses. Isobars are plotted in 1-hPa intervals.

Citation: Journal of Applied Meteorology and Climatology 46, 1; 10.1175/JAM2418.1

Fig. 8.
Fig. 8.

Mean course (1989–2001) of pressure at the 14/18 (black line, southern Valaisian Alps, Switzerland) and 14/26 (gray line, Swiss Plateau) grid points. The abscissa displays time from 1 to 7 Feb in 6-hourly intervals (dashed vertical lines), and the ordinate shows reduced pressure at the grid points.

Citation: Journal of Applied Meteorology and Climatology 46, 1; 10.1175/JAM2418.1

Fig. 9.
Fig. 9.

Mean 2D divergence of the 10-m wind field as a function of time for different months, averaged over the period of 1980–2001. The solid lines indicate divergence for (top to bottom) January, October, April, and August.

Citation: Journal of Applied Meteorology and Climatology 46, 1; 10.1175/JAM2418.1

Fig. 10.
Fig. 10.

Mean configuration of reduced sea level pressure for all 1200 UTC analyses of May 1989–2001; pmin = 1012.8 hPa, with isobar spacing 0.5 hPa.

Citation: Journal of Applied Meteorology and Climatology 46, 1; 10.1175/JAM2418.1

Fig. 11.
Fig. 11.

Mean configuration of reduced sea level pressure for all 0900 UTC analyses of January 1989–2001; pmax = 1027.2 hPa, with isobar spacing 0.5 hPa.

Citation: Journal of Applied Meteorology and Climatology 46, 1; 10.1175/JAM2418.1

Fig. 12.
Fig. 12.

Daily means of reduced pressure for two grid points for December, averaged over the period of 1989–2001. The first grid point is located in an Alpine Valley (Mur Valley at the border between Styria and Carinthia, Austria); the second one can be assigned to the Po Plain in northern Italy.

Citation: Journal of Applied Meteorology and Climatology 46, 1; 10.1175/JAM2418.1

Fig. 13.
Fig. 13.

Percentage of days with north Stau or south Stau relative to the length of a month (mean value for 1989–2001). Because the lengths of months differ, the number is given in percent of the length of the month.

Citation: Journal of Applied Meteorology and Climatology 46, 1; 10.1175/JAM2418.1

Fig. 14.
Fig. 14.

Pressure configuration for a typical south Stau situation with a pressure difference of 14–17 hPa between two grid points north and south of the Alps (black pluses). Isobars are plotted in 1-hPa intervals. Mean configuration of 67 analyses from the period of 1989–2001.

Citation: Journal of Applied Meteorology and Climatology 46, 1; 10.1175/JAM2418.1

Fig. 15.
Fig. 15.

Regions with an above-average number of grid points showing a WPM. The corresponding grid points are connected by thick black lines. The boldface numbers indicate how often the WPM criterion was met out of 37 984 cases.

Citation: Journal of Applied Meteorology and Climatology 46, 1; 10.1175/JAM2418.1

Table 1.

Statistics of the differences between station measurements and interpolated values at station locations (hPa). Total number of stations n is 260.

Table 1.
Table 2.

Comparison of ERA-40 and VERA climatological mean values of MSL pressure (first four lines; hPa) and absolute values of pressure gradients [last four lines; hPa (100 km)−1] for January 1995.

Table 2.
Table 3.

Relative frequency (%) of different classes of pressure gradients (hPa) between two 360-km-distant grid points north and south of the Alps. Numbers were calculated from 37 984 analyses of the period of 1989–2001.

Table 3.
Save