Estudo da eficiência da previsão numérica do tempo de curto prazo para o município de Maceió/AL, utilizando o modelo WRF

Abstract

With the insertion of the numerical models, it became possible to carry out a more reliable and short-term forecast of the time in search of more efficient results besides being necessary to improve the model and to acquire knowledge of its performance. In this work we apply simple statistical methods to compare the performance of the model by comparing the meteorological variables with observed data from an automated weather station (AWS) and data simulated by the WRF model. The analyzed data of these variables were from July 10 to 19, 2017, using statistical and computational tools: atmospheric mesoscale model (WRF), spreadsheets, as well as specific programming languages and scripts for data execution. The (AWS) from which the observed data was extracted is located in Maceió. The WRF simulations were validated through data series and statistical analyzes and it was verified that in the variables there was efficiency of the WRF in the predictions. The results expected with correlation coefficient between average and strong for 24h in most simulated and a mean correlation for 48h and 72h in most of the simulated, it was possible to conclude that the model is adjusted to predict average values and that in some moments that the minimum and maximum results could not be simulated, it is therefore possible to indicate the model as a tool to be used to carry out short-term forecasting provided there is updating of topography and land use for the city of Maceió. In this research it was possible to obtain expected results, but the equipment of measurement and data processing can be improved to obtain even more satisfactory results.