Validation of simulated precipitation forecasts with the BRAMS model for the La Libertad-Peru region

• Autores: Nina, Martha, Rubio, Obidio
• Formato: artículo
• Fecha de Publicación: 2021
• Institución: Universidad Nacional de Trujillo
• Repositorio: Revistas - Universidad Nacional de Trujillo
• Lenguaje: español
• OAI Identifier: oai:ojs.revistas.unitru.edu.pe:article/3713
• Enlace del recurso: https://revistas.unitru.edu.pe/index.php/SSMM/article/view/3713
• Nivel de acceso: acceso abierto
• Materia: Numerical model BRAMS; quality evaluation; forecast verification; precipitation; Modelo numérico BRAMS; evaluación de calidad; verificación de pronósticos; precipitación

The objective of this work is to validate the precipitation forecasts simulated with the Brazilian Regional Atmospheric Modelling Systems (BRAMS) numerical model in its version 5.3, for the La Libertad Region in Peru; this region of Peru has 80% of its land located in the northern highlands of Peru. In mountainous areas, precipitation is strongly influenced by the height of the terrain, a poor representation of the topographic elevations and depressions of the terrain can lead to an erroneous representation of the resolvable phenomena explicit by the model. Taking into account this characteristic of the study area, in the first instance the BRAMS model was configured to adequately represent the rugged orography of the local área at a horizontal resolution of 10 km. Then, the model was run to simulate precipitation forecasts at time horizons of 24, 48, 72, 96 and 120 hours, for the months of December 2019, January and February 2020. The validation of the forecasts was performed against observed data obtained from the National Meteorological and Hydrological Service (SENAMHI) of Peru, using quality indices for continuous and binary variables. From the results obtained, it is concluded that the BRAMS model performed well in forecasting the occurrence of precipitation for all time horizons. However, the model had difficulties in forecasting the occurrence of precipitation for higher thresholds, and predicted more false alarms for these thresholds. Finally, the model applied to the La Libertad Region in Peru, with a fairly rugged topography, had similar results to those obtained by other regional models applied in areas where there is little influence of terrain height.