Análise da ocorrência do granizo através das simulações no modelo WRF sobre o Rio Grande do Sul

Abstract

Hail is a form of precipitation that occurs from falling irregular pieces of ice. It is a short-term weather phenomenon that can occur in a small area in isolated events. Thus, as it is an event that varies locally and spatially, it was sought through the WRF model to find out if it is able to correctly indicate surface hail. Eight experiments with four CMP and two CCP were used in three cases selected from reports of hail events by SEDEC, magazine and newspaper website reports from 2009 to 2016. For the accomplishment of the work we used the NCEP data and known as Final Analysis. This data is meant to be run in WRF in the simulation containing 40 levels , the highest level being 50 hPa. The initial condition and boundary data are from NCEP with spatial resolution of 1 by 1 degree latitude by longitude. Three domains were used in the model with grid scales of 27, 9 and 3 km, with the center in the third (smaller) domain located in Santa Maria - RS. The simulations were started approximately 12 hours before the events and the outputs used in the results analysis were every 10 to 10 minutes for the smaller 3 km domain. The contingent table 2 x 2 was also used to analyze the GAS grid points at the end of simulation with the observation grid points defined as area of affected cities and thus statical indices are used to verify wich better produces hail. Analyzing the hail in the atmosphere and surface accumulated, the results indicated among all experiments that Exp1 and Exp2 were better in hail production for the three cases. Individually the Case 3 was the one that produced the best GAS in time and space. For the statistical indices, an unsatisfactory result was observed when analyzing the experiments using the CMP with CCP, mainly applying the high resolution modeling. The statistical indices were not satisfactory to study hail events mainly by the use of high resolution numerical modeling. Therefore, it is concluded that studying a local event such as hail is extremely difficult, as it varies greatly in time and space, especially in understanding the model’s cloud microphysics processes to produce exactly where the phenomenon occurred on the surface.