Keywords

Map 10.1
figure 1

Climate stations (Data source: DWD, Deutscher Wetterdienst, ZAMG, Zentralanstalt für Meteorologie)

Full size image

1 Introduction

Almost all DANUBIA model components require values at ground level for meteorological parameters at high temporal and spatial resolution. Such data is already generated at 1 km2 resolution and at hourly intervals either via the mesoscale atmosphere model MM5 or by interpolation of observed time series. The measurement data from climate stations form the basis for the temporal and spatial interpolation of the meteorological parameters with the model AtmoStations and are used to validate the results of the DANUBIA component Rivernetwork (Chap. 29) and the models Snow (Chap. 30) and AtmoMM5 (Chap. 32).

In order to rely on a network of measurement stations with good coverage and of guaranteed data quality, homogeneity and continuity, data from all available stations of the German Weather Service (Deutscher Wetterdienst = DWD), the national meteorological service of the Federal Republic of Germany and the Central Institution for Meteorology and Geodynamics (Zentralanstalt für Meteorologie und Geodynamik = ZAMG) and the national meteorological and geophysical service of the Federal Republic of Austria were obtained for the Upper Danube catchment area and its surroundings. For the modelling period from 1960 to 2004, there were a total of 377 weather stations available for the region in question, of which 280 were German and 97 were Austrian. For the region within Switzerland, there was discharge data at the boundary gauges and climatologies for various meteorological parameters. Thus, there was no need for a costly acquisition of data.

2 Data Processing

The DWD data come from climate stations with up to three daily observations and from the main synoptic-climatological network (hourly measurements, enhanced with climatological data). The data obtained from the ZAMG include ground-level observations of meteorological parameters that are captured at conventional climate stations with manual measurements and at semiautomated weather stations. Table 10.1 presents the parameters for which data was captured at these stations.

Table 10.1 Parameters of weather stations with time-scheduled measurement or measurement duration
Full size table

On April 1, 2001, the method for capturing climatological data at the DWD was converted. New automatic data acquisition meant that additional data for calculating daily values could be utilized. One main aspect of the change in data management at the DWD was the shift in scheduled weather measurements from 7:30, 14:30 and 21:30 CET to 7:00, 14:00 and 19:00 CET. The actual observation time is 10 min prior to the reference time. This shift took effect at first for the main synoptic-climatological automated network. The new times were also adopted with the subsequent automation of the secondary climate and precipitation network (www.dwd.de). This change entailed only a minor shift in the time points for the temporal interpolation of the parameter with the model AtmoStations. However, the change entailed a significant change in the mean values (Gattermayr 2001). The model AtmoStations did not use the mean values from the weather stations, but instead calculated these from the interpolated values.

In addition to the change in measurement procedure during the study period, there is also the difference in the methods used by the DWD and ZAMG to be considered (e.g. different measurement times, timing according to CET or LMT or different units of the measurements). However, these differences are taken into account in the model AtmoStations as much as possible, so that the degree of inhomogeneity in the data is kept as low as possible.

For the period from 1960 to 2004, a total of over four million measurement days (data records) are available. The individual stations cover different time periods and also sometimes contain significant gaps in the data. Roughly a quarter of the stations (approx. 100) have a complete record for this period. Therefore, the number of stations per time increment available for spatial interpolation varies considerably. For the year 1960, data from 211 stations could be used for the model, whereas for the years 1994 and 1995, up to 300 stations were available (see Fig. 10.1).

Fig. 10.1
figure 2

Number of stations with daily measures per year

Full size image

3 Results

Map 10.1 shows the locations of the DWD and ZAMG climate stations within the study area. The station with the highest altitude is located in Austria on Sonnblick Mountain (3,105 m a.s.l.), and the second highest is in Germany on the Zugspitze (2,960 m a.s.l.); third highest is the device on the Pitztal Glacier at 2,850 m a.s.l. The two highest stations in the Bavarian Forest are on the Great Arber (1,437 m a.s.l.) and the Great Falkenstein (1,307 m a.s.l.). The lowest altitude measurements are taken in Mannheim at 96 m a.s.l. It is noteworthy that at several Austrian stations, two symbols are directly adjacent or virtually on top of each other. In part, this situation derives from the fact that these stations are carried forward under a new designation if the type of station changes (e.g. as a result of automation). Another reason is that stations on private property must be relocated and supplied with a new number if the observer changes.

The altitude distribution of the climate stations indicates that above 1,000 m a.s.l. there are significantly fewer stations in operation than below 1,000 m a.s.l.. Half of the sites are below 533 m a.s.l. and 75 % lie below 785 m a.s.l. (see Fig. 10.2). Eight stations are located above 2,000 m a.s.l., of which seven are in Austria. However, station density approximately corresponds to the area-altitude distribution of the catchment area.

Fig. 10.2
figure 3

Altitude distribution of the climate stations

Full size image

The high altitude and the strong small-scale variability in the Alpine region have an important impact on the climate and water supply in the Upper Danube catchment area. Relief enhances the thermal and dynamic processes in the atmosphere that can, for example, lead locally to more precipitation and hence can be very significant for the water supply. For the spatial interpretation and simulation of climatic state variables in regions with sharp relief patterns, there is therefore a need for a greater number of stations with regular high-quality data even at high sites and the most extreme locations of the study area. Unfortunately, at these sites logistical problems for the operation of the stations are often experienced. In addition to the measurements from the stations depicted on Map 10.1, data from climate stations at academic institutions were used to validate the model results. For example, the “Vernagtbach Climate Station” (at Vernagtferner) in Ötztal at an altitude of 2,640 m a.s.l., which belongs to the Commission for Geodesy and Glaciology of the Bavarian Academy of Sciences and Humanities, supplied year-round hourly data for all parameters listed in Table 10.1 except for cloud cover and visibility. Another station is located also in Ötztal on the Schwarzkögele Glacier (3,074 m a.s.l.), and there are several measurement sites on the glacier itself that are at 3,000 m a.s.l.