Automatic weather station data from Weisssee snow research site (2014-2018)

Meteorological and snow measurements of the automatic weather station (AWS) at Weisssee snow research site are presented for the period 1 October 2014 to 30 September 2018. The entire set comprises of 21 data series including precipitation, air temperature, relative humidity, incoming and outgoing short- and longwave radiation, wind speed and direction, snow depth, snow density and snow temperature in 15 minutes temporal resolution. On 6 October 2014, a Snow-Pack-Analyzer (SPA, Sommer Messtechnik) which measures different dielectric constants of ice, water, and air along flat ribbon sensors [Stähli et al., 2004, doi:10.3189/172756404781814933] was installed for continuous measurements of snow density including a diagonal sensor for measuring bulk snow density and three horizontally mounted sensors at 10, 30, and 50 cm above ground. A mean station SWE series can be calculated using both HS sensors and the SPA density values. Unrealistic density measurements are likely if SPA sensors are not (entirely) snow covered. Additionally, a series of 29 manual snow pit bulk density measurements can be used to check SPA density values. Comparing the daily mean values of the raw SPA bulk density values with the manual measurements results in a mean absolute error of 66 kg/m³ (R²=0.18). In the first winter 2014/2015 and especially in June the SPA sensor was strongly influenced by air pockets around the diagonal sensor which is why the entire SPA setup was turned by about 90° to be positioned parallel to the dominant wind direction [Schattan et al., 2017]. Thus, neglecting the data from June 2015 decreases the mean absolute error to 53 kg/m³ (R²=0.33). The mean absolute error decreases further to 40 kg/m³ (R²=0.8) when for snow depth values <0.7 m the density measurements of the horizontal sensors are used (e.g. for snow depth = 30 cm the 10 cm horizontal sensor is seen as relevant). Snow temperature is also measured at 10, 30, and 50 cm above ground and additionally at the snow-soil interface (0 cm) using unshielded PT100 sensors only giving meaningful measurements if they are covered by snow. Snow depth and precipitation measurements are made using identical sensors as described by Krajči et al. [2017]. The ultrasonic signal of the snow depth sensor is automatically corrected for temperature effects but still can show fluctuating measurements. Precipitation, shortwave radiation, air temperature and HS measurements are checked for errors and corrected if necessary using data from neighbouring stations. Other sensors measurements are mainly raw data. Missing values in the data series were not filled. Precipitation data were not corrected for e.g. wind induced undercatch which is known to be an issue. Relative humidity is unfortunately not available in the first winter (2014/15) at the AWS but records start in December 2015. To bridge the gap partly, additionally available relative humidity measurements from Weisssee research site are made available for the period 7 January 2015 to 7 June 2015. The data series (variable: relative humidity hobo) consists of mean values of three shielded low cost sensors which were placed around the AWS and were mounted 2 m above ground. The standard deviation of the measurements of the three sensors is provided and (if high) indicates data uncertainties caused by problems with individual sensors (e.g. measurement errors due to snow coverage of the sensor). To be used in combination with this data set: Fey, Christine; Schattan, Paul; Helfricht, Kay; Schöber, Johannes (2018): TLS snow distribution maps of the Weisssee snow research site (Kaunertal, Austria). PANGAEA, https://doi.pangaea.de/10.1594/PANGAEA.896843.