info:eu-repo/semantics/bachelorThesis
Validación del modelo meteorológico WRF (Weather Research Forecasting) para Chimborazo
Fecha
2022-03-08Registro en:
Camino Carrasco, Teresa Ibeth. (2022). Validación del modelo meteorológico WRF (Weather Research Forecasting) para Chimborazo. Escuela Superior Politécnica de Chimborazo. Riobamba.
Autor
Camino Carrasco, Teresa Ibeth
Resumen
The aim of this study was to validate the WRF (Weather Research Forecasting) mesoscale model in the meteorological modeling of Chimborazo, it was necessary to start with the information collected by the meteorological stations which are part of the study area, this information was used to compare data; then, it was necessary to download and compile the WRF model through a virtual server; The analysis variables for both cases were temperature (degrees Celsius) and solar radiation (W/m2) with a monthly temporal resolution, from which data interpolation tools were applied in order to produce cartographic representations of the study area, considering a grid of approximately 6805 cells with a dimension of 1km by 1km, this revealed that the temperature ranges of the province vary from 11 to 18 degrees Celsius, in which the months from June to September are considered the hottest months, together with the radiation levels between 130 and 350 W/m2 according to INAMHI; The WRF model showed a temperature range from 10 to 15 degrees Celsius, with radiation levels between 90 and 100 W/m2. According to this study, the northern part of the province evidences the highest records of this variable throughout the year. Finally, the processed information was exposed to statistical tests of normality, root mean squared error (RMSE), mean absolute error (MAE) and correlation coefficients; showing that the difference between the mean of the evaluated data was minimal for the temperature variable, which represents a similarity between the data from INAMHI and WRF, and which are appropriate for the area of interest, opposite to the radiation variable, since the difference shown was considerable. Thus, it is recommended to reduce the temporal resolution of the model data seeking to have a greater representation.