Impacto da assimilação de dados de sondagens e radiâncias em evento de ciclogênese

Abstract

An Assessment of the assimilation satellite radiances and the atmospheric soundings impact in the WRF simulation is presented in this work. The simulation is performed during two cases of cyclone propagation over southern Brazil, from August 20 to 23, 2020, and from June 27 to 29, 2021. The technique used in this study is a ThreeDimensional Variational Data Assimilation Scheme implemented on WRF Model (3DVar). The statistical metrics are used in order to verify the WRF variables, using INMET Weather Stations as ground truth. The statistical analyses are descriptive and present the variability at each point compared to the data observed using the coefficient of variation of the predicted values. The results are similar for both cases and show that the model adequately reproduced cyclogenesis events comparing to CPTEC analyses, satellite images, and GFS model. The assimilation procedure has a higher impact on the first 48 hours of forecasting for low and medium level of the atmosphere and, in the last 24 hours, assimilation has virtually no effect. Considering the sources of assimilated data, atmospheric soundings tend to reduce forecast errors more frequently than radiances, as they provide data on the vertical profile of the atmosphere; while satellite data has better temporal and spatial representation, although it is sources of error at low levels of the atmosphere. The statistical analysis shows that the estimation of the WRF model with assimilation procedure represents the trajectory of the ground truth. The forecasting of temperature and pressure overcoming moisture and windy simulations. In this simulation the model used to generate a phase delay in the cold wave propagation.