Overview This gridded meteorological data set consists of climatologies (climate indicators) on a 30-year average basis for Austria and covers two historical periods with a high spatial resolution of 250m. The two 30-year periods provided for the observations allow the analysis of the climate change that has already occurred. The selection of climate indicators is optimized for the needs of ecological models. Resolution: 250x250mProjection: EPSG 31287Extent: AustriaPeriods: 1961-1990 and 1991-2020Format: GeoTIFFData sources: Station data and derived products List of climatologies (climate indicators) # Short name Name Description Unit Yearly (Y), monthly (M) or growing season (GS) 1 tasmin Average daily minimum temperature Arithmetic mean °C Y, M 2 tasmax Average daily maximum temperature Arithmetic mean °C Y, M 3 tas Average temperature Arithmetic mean °C Y, M 4 tas_warmest_month Average temperature mean in the warmest month Calculation of the mean temperature over the climate period for all months and then selection of the highest value for the warmest month °C Y 5 tas_coldest_month Average temperature mean in the coldest month Calculation of the mean temperature over the climate period for all months and then selection of the lowest value for the coldest month °C Y 6 tasmin_coldest_month Average temperature minimum in the coldest month Calculation of the mean minimum temperature over the climate period for all months and then selection of the lowest value for the coldest month °C Y 7 tasmax_warmest_month Average temperature maximum in the warmest month Calculation of the mean maximum temperature over the climate period for all months and then selection of the highest value for the warmest month °C Y 8 thermal_continentality Average annual amplitude of monthly mean temperature (Thermal continentality) Climatological monthly mean temperature in the warmest month minus climatological monthly mean temperature in the coldest month (04_tas* minus 05_tas*). °C Y 9 FD Average number of frost days per year Defined by 0°C daily minimum temperature at 2m height. days Y 10 GSL Average length of the growing season The growing season is the duration in days of the longest continuous period of days with an average temperature of at least 5°C. However, an earlier or later period of such warm days is included in the growing season if it lasts longer than the sum of all intervening cooler days days Y 11 tas Average temperature in the growing season Average temperature in the growing season defined as in climate indicator 10 (GSL). If a year does not have a growing season, this value is not defined. °C GS 12 GDD Average Growing Degree Days per year above 5°C Σ(Tmean – 5°C) per year. °C Y 13 FD_first Average date of the first frost occurrence Frost is defined by a temperature of 0°C at a height of 2 meters (arithmetic mean). Years without frost are excluded from the calculation of the mean. If no frost occurs at all, the value is indeterminate day of year Y 14 FD_last Average date of the last frost occurrence Frost is defined by a temperature of 0°C at a height of 2 meters (arithmetic mean). Years without frost are excluded from the calculation of the mean. If no frost occurs at all, the value is indeterminate day of year Y 15 HD35 Average number of extremely hot days above 35°C per year Defined by 35°C daily maximum temperature at 2m height. days Y 16 FD_10 Average number of days with hard frost below -10°C per year Defined by -10°C daily minimum temperature at 2m height. days Y 20 GLO_hori Average sum of global radiation on horizontal surface Also called irradiance or shortwave incoming radiation, taking cloud cover into account. kWh Y, M 21 GLO_real Average sum of global radiation on the real surface Also called irradiance or shortwave incoming radiation, taking cloud cover into account. kWh Y, M 31 vpd Average water vapor pressure deficit Calculated from daily dew point temperature and temperature. hPa Y, M 31 hurs Average relative humidity Calculated from daily dew point temperature and temperature. % Y, M 32 vpd Average water vapor pressure deficit Calculated from daily dew point temperature and temperature. hPa GS 32 hurs Average relative humidity Calculated from daily dew point temperature and temperature. % GS 33 pr Average precipitation sum mm Y, M 34 hygric_continentality Average hygric continentality according to Gams Defined as arctan of (elevation/annual_precip). Gams, H. (1931). Die klimatische Begrenzung von Pflanzenarealen und die Verteilung der hygrischen Kontinentalität in den Alpen. ° Y 35 pr1mm Average number of days with precipitation Daily precipitation of at least 1mm. days Y 36 pr1mm Average number of days with precipitation in the growing season Daily precipitation of at least 1mm. days GS 37 DP_3days, DP_5days, DP_7days Average number of days in dry periods in summer half-year Number of days in periods of at least 3, 5, or 7 days with a daily precipitation total of less than 1mm. Summer half-year: April to September days Y 39 ET0 Average annual potential evapotranspiration Calculation according to FAO Penman-Monteith: fao.org/3/X0490E/x0490e08.htm mm Y, M 40 WBAL Average climatic water balance Precipitation minus potential evapotranspiration mm Y, M 41 WBAL Average climatic water balance in the growing season Precipitation minus potential evapotranspiration in the growing season mm GS Data sources The climate indicators were calculated using daily data from different sources: Temperature: SPARTACUS v2.1 (Gridded data set, 1x1km, https://doi.org/10.1007/s00704-015-1411-4 ) Precipitation: SPARTACUS v2.1 (Gridded data set, 1x1km, https://doi.org/10.1007/s00704-017-2093-x ) Radiation: APOLIS SHORT(Gridded data set, 100x100m, 2006-2020, https://adsabs.harvard.edu/abs/2012EGUGA..14.9705O)APOLIS LONG (Gridded data set, 100x100m, 1981-2016)SPARTACUS v2.1 (Gridded data set, 1x1km, daily sunshine duration, 1961-2020). Wind: Daily station data (https://doi.org/10.60669/gs6w-jd70) Humidity: Daily station data (https://doi.org/10.60669/gs6w-jd70) Digital elevation model: © Kooperation Länder, Bund (BEV, BML), 2022

The following daily gridded meteorological data were generated in order to calculate the 30-year averaged climate indicators: a) Precipitation, Temperature The daily data of SPARTACUS (maximum and minimum temperature, precipitation) was downscaled from 1 km to 250 m via a regional linear regression with elevation to account for topography-related effects on the finer resolution. A simple definition of the mean temperature is the average of the maximum and minimum temperatures. However, this definition is only a rough approximation of the real average. The more accurate daily mean temperature can be calculated with hourly data. In the SPARTACUS, hourly values are not available. Therefore, the mean temperature was corrected using climatologically determined values from the stations for each month. For each month and each station, the mean difference between the simple definition and the mean of hourly values was calculated. A dependence on the elevation can be observed. A linear regression of this difference with the elevation was performed for each month, and the values were mapped onto the grid. These residuals of the regression were spatially interpolated using a three-dimensional inverse-distance-squared method. This provides a spatial, monthly correction for the mean temperature that is added to the simple mean temperature. b) Global Radiation The requirements for the final dataset were continuous and homogeneous data from 1961 to 2020. APOLIS SHORT deviates significantly from the station measurements (typically 10 to 20% higher). Therefore, APOLIS LONG (1981-2016) was set as the reference, and the periods before and after were statistically adjusted to it. APOLIS SHORT was homogenized using APOLIS LONG. This was accomplished using empirical, multiplicative quantile mapping correction (Déqué, 2007, https://doi.org/10.1016/j.gloplacha.2006.11.030). For the period 1961-1981, we had to resort to gridded sunshine duration data from SPARTACUS v2.1 - a proxy that exhibits a high correlation with global radiation. The daily sunshine duration data were bilinearly interpolated from 1km to 100m (from SPARTACUS to APOLIS grid). The sunshine duration was corrected using empirical, multiplicative quantile mapping with global radiation from APOLIS LONG. c) Wind Speed The mean wind speed was interpolated from the weather stations (Publication in preparation). The time series had to be homogenized because some time series showed very obvious breaks. According to (Haimberger, 2007, https://doi.org/10.1175/JCLI4050.1), the Standard Normal Homogeneity Test (SNHT) was used to locate breaks in time series. For the wind speed measurements, we considered the newer part of the time series as "correct," and the older time series before the break was corrected using additive quantile mapping. The spatial interpolation of the homogenized data was separated into two parts: Data from the period 2009-2021 were utilized to calculate monthly climatologies. In this period, the density of the station network is almost constant. The monthly climatologies of mean wind speed at the stations are calculated with ridge regression, using elevation and proxies for the exposure which are calculated as the topography minus a smoothed topography, with smoothing performed using different radii (2, 5, and 8 km). The residuals from the ridge regression were interpolated using a three-dimensional inverse-distance-squared method. The gridded daily data are calculated as residuals from the monthly climatologies. For each day, the difference from the monthly climatology is calculated for each station that provided a valid value on that day. These differences are then spatially interpolated again using three-dimensional inverse-distance weighting. The final mean wind speed is calculated by adding the monthly climatology to the interpolated differences from the station climatology. d) Humidity The humidity (in the form of dew point) was generated following the methods described in Hiebl and Frei, 2016 (https://doi.org/10.1007/s00704-015-1411-4) and consists of Daily, non-linear vertical profiles for different regions in Austria using dew point from weather stations and A residual interpolation using the three-dimensional inverse-distance weighting which is also used above in the section "d) Wind Speed".