WRF sensibility to microphysics and planetary boundary layer at micro-catchment scale: the case of the Quinuas catchment in the tropical andes of Ecuador

Abstract

Reliable high-resolution forecasts of precipitation are necessary to improve hydrological models forecasting. This is important in mountain regions, where complex convective and orographic processes occur. The present study evaluate the impact on precipitation, temperature and relative humidity forecasting of two sub-grid parameterization schemes: microphysics and planetary boundary layer (PBL). The results were compared for three resolutions (4km, 12 km and 36 km) in two nested domains. The Weather Research and Forecasting model was used. The analysis was based on two events of 20 and 15 mm cumulative rainfall during the dry and rainy seasons, respectively. Five allowable configurations were tested. The results show that microphysics and PBL’s schemes perform satisfactory forecasts for the two seasons. Specifically, there is a dependence of the PBL for each season. The Thompson microphysics scheme showed good results on both events, resulting in an adequate configuration for Tropical Andean Mountains precipitation forecast. In addition, precipitation is highly sensitive to domain resolution. Moreover, forecasted temperature and relative humidity (RH) are also very sensitive to domain resolution: temperature forecast has lower performances as the domain gets finer, meanwhile RH and precipitation improve their forecast.