We assess an operational merged gauge–radar precipitation product over a period of 12 years, using gridded precipitation fields from a dense gauge network (WegenerNet) in southeastern Austria. We analyze annual data, seasonal data, and extremes using different metrics. We identify individual events using a simple threshold based on the interval between two consecutive events and evaluate the events' characteristics in both datasets.

Empirical high-resolution surface wind fields from two study areas, automatically generated by a weather diagnostic application, were intercompared with wind fields of different modeling approaches. The focus is on evaluating spatial differences and displacements between the different datasets. In general, the spatial verification indicates a better statistical agreement for the first study area (hilly WegenerNet Feldbach Region), than for the second one (mountainous WegenerNet Johnsbachtal).

Ensemble forecasting offers a useful method to simulate the uncertainty of a numerical forecast model for each individual forecast run. This study compares ALADIN-LAEF, a 16-member ensemble with a resolution of 11 km that combines several perturbation methods, with AROME-EPS, which downscales the members of ALADIN-LAEF to 2.5 km resolution. The verification shows that there are benefits of a higher-resolution ensemble, especially for highly localized precipitation and for mountainous terrain.

In this study a precipitation nowcasting method is developed which relies on satellite products and automatic weather station data only. It thus omits ground based radar observations which are not readily available in large parts of the world. The system shows improved skill when compared to numerical weather prediction models for analysis and for lead times up to one hour. This type of nowcasting could be valuable in data sparse regions where radar observations are lacking or of poor quality.

The effect of empirical-statistical bias correction methods, like quantile mapping (QM), on the simulated climate change signals (CCS) is currently strongly discussed and is often regarded as deficiency of bias correction methods. We demonstrate that, quite the contrary, QM can lead to an improved CCS and also has the potential to serve as an empirical constraint on model uncertainty in climate projections.