CML data assimilation for precipitation forecasts over Austria
Alexander Kann1, Christoph Wittmann1, Phillip Scheffknecht1, Benedikt Bica1, Oliver Eigner2, Fabian Kovac2
1 GeoSphere, Austria, 2 St. Pölten University of Applied Sciences, Austria
Due to its sensitivity to the atmosphere’s moisture content, commercial microwave link data from mobile phone operators provide a valuable complementing data source to the conventional meteorological observation network. For precipitation analysis and forecasting, state-of-the-art models usually rely on rain radar and in-situ measurements from hydro-meteorological station networks. However, especially in Alpine regions, the observations have deficiencies, e.g. due to radar blocking effects or due to other local effects which are not represented well enough by the networks. The presentation will give a brief introduction of the CML data over Austria and its conversion to rain rates. The NWP model (AROME) and the analysis and nowcasting system (INCA) will be introduced, along with the assimilation procedure of the CML data. Finally, the impact of the CML data on the model performances will be discussed, ending up with an outlook on further activities and recommendations towards improved applications.
Combining commercial microwave links and weather radar for rain and dry snow detection
Erlend Øydvin1, Renaud Patrick Gaban1, Mareile Astrid Wolff1, Vegard Nilsen1, Nils-Otto Kitterød1, Christian Chwala3, Remco Van de Beek2, Jafet Andersson2
1 NMBU, Realtek, Norway, 2 Swedish Meteorological and Hydrological Institute (SMHI), Sweden, 3 Institute of Meteorology and Climate Research (IMK), Germany
Our study explores the synergy between Commercial microwave links (CMLs) and weather radar data to classify precipitation types. While CMLs estimate rainfall through signal loss, dry snow causes little signal dampening and appears as dry events. Addressing the limitations of weather radar in distinguishing precipitation types, we propose to combine weather radar wet events with CML dry events to accurately classify dry snow events. The estimates are compared with nearby disdrometer ground truth data across multiple locations in Norway.