Search

Snowfall in elevated areas of the mid-latitudes has a strong impact on infrastructure, freshwater availability, and the climate system. The Cantabrian Mountains of the northwestern Iberian Peninsula are very vulnerable to climate change because of their moderate altitudes, which limits their snowfall. Monitoring snow events is essential for the evaluation of weather and climate prediction models. However, measurement networks are scarce in mountainous areas and have great uncertainties because of blizzards. In this study, a multiphysics ensemble of the Weather Research and Forecasting (WRF) model was designed using three microphysics and two planetary boundary layer (PBL) schemes to simulate nine snowfall events in the Cantabrian Mountains during autumn and winter 2021–2022. The WRF was validated using several snow characteristics, such as liquid water equivalent, snow cover, and snow depth. Liquid water equivalent was evaluated using snow-gauge networks and satellite products in an assessment of snow cover. In addition, a monitoring network of webcams and snow poles was implemented, improving the low density of snow observations in the mountains. The results showed good model performance for detection of snow cover and slight overestimation of liquid water equivalent and snow thickness, which may have been caused by under-catchment that is generally an effect of wind on the measurement systems and by snow compaction, respectively. Morrison microphysics and Mellor-Yamada-Nakanishi-Niino (MYNN PBL) yielded better results for liquid water equivalent at higher altitudes and output greater snow cover. The results help determine the best configurations for snow modelling in the study area to develop future studies of the spatiotemporal patterns of snow distribution. • Monitoring system of snow events in Cantabrian Mountains has been developed. • We selected nine snowfall events during autumn and winter 2021–2022. • Three microphysics and two PBL parameterizations were tested. • Morrison microphysics and MYNN PBL yielded better results at higher altitudes. [ABSTRACT FROM AUTHOR]

Precipitation is one of the most relevant fields in atmospheric modeling because of its environmental, social and economic implications. However, precipitation validation from weather model outputs presents substantial challenges, such as measurement uncertainties, use of gridded datasets vs. direct observations, and the selection of statistical goodness-of-fit measures. The main difficulty of working with precipitation is that it can be spatially irregular, especially in extreme events. High temporal aggregation smooths the field and reduces verification uncertainty. For this reason, validations are usually focused on a daily scale. However, many extreme events occur on shorter periods, for which a sub-daily precipitation assessment is required. In this paper, hourly precipitation verification of the Weather Research and Forecasting (WRF) model is explored for 45 extreme precipitation events (EPEs) recorded in northeastern Spain. For this, stations with recorded EPEs were classified according to the hourly distribution of precipitation. WRF simulations were established considering three microphysics and two planetary boundary layer (PBL) parameterizations. Finally, several statistical goodness-of-fit measures and spatial and temporal precipitation distributions were used for evaluating WRF performance. The results showed that microphysics were more important than PBL parameterizations. Goddard and Thompson together with Mellor-Yamada-Nakanishi and Nino PBL gave better results for most of the analyzed characteristics. However, an optimal combination of parameterizations was not obtained for all EPEs, because event characteristics had important effects on model performance. • Hourly precipitation verification of the WRF model has been explored. • We selected 965 stations where extreme precipitation was recorded. • Three microphysics and two planetary boundary layer parameterizations were tested. • Goddard and Thompson together with Mellor-Yamada-Nakanishi and Nino gave better results. [ABSTRACT FROM AUTHOR]