Demonstrating the Added Value of Crowdsourced Rainfall Data in Complex Terrain

Abstract

Validating high-resolution weather and climate models is challenged by insufficient spatial and temporal resolution of meteorological observations, particularly for the precipitation in complex terrain. Traditional datasets, which rely on sparse official weather stations and gridded datasets, often lack the spatio-temporal resolution needed for accurate localized studies. This study serves as a first step in investigating the potential of including Personal Weather Stations (PWSs) in the validation of high-resolution regional climate models. We performed a quality control on PWS data, flagging approximately 13% and retaining around 450 stations in Western Norway. Compared to 124 official meteorological stations (MET stations), PWSs provided significantly improved spatial coverage, especially in densely populated areas, revealing spatial variability often missed by MET stations and traditional gridded datasets. We validated simulations from the Weather Research and Forecasting (WRF) regional climate model using the combined PWS and MET observational dataset for two cases: multiple frontal passages in November 2022 and a record-breaking convective burst in August 2023, which were sparsely captured by official MET stations. Although biases existed in the WRF dataset, the incorporation of PWSs in the observational dataset revealed a more nuanced precipitation pattern and provided enhanced spatial validation opportunities. In conclusion, PWS networks significantly enhance observational coverage, aiding high-resolution model validation and opportunities for improved local precipitation understanding. As the number of PWSs grows, refined quality control measures will further solidify their role in meteorological research and emergency preparedness, particularly for localized extreme weather events. This integration is vital for advancing climate science and improving community resilience to weather-related challenges.