Uso de redes complexas como ferramenta de análise espaço-temporal de dados de precipitação

Abstract

This work aims to use complex systems methods to analyze the climatic phenomena of large - scale precipitation, which affects the states of Mato Grosso and Mato Grosso do Sul and the Pantanal, the largest wetland in the world. The variability, movement, force, beginning and end of the rains have enormous impact on the agriculture, livestock, economy and the life of the inhabitants of the region of study. Understanding the mechanisms of the rainy season is of great interest to society and also a scientific challenge. This study uses a Complex Network (CN) data analysis approach to analyze the dynamics and spatial-temporal visualization of extreme precipitation events during the pre-rain (october - november), rain (december - january - february) and post-rain (marchapril), using data from the Tropical Rainfall Measuring Mission (TRMM) satellite. The construction of the CN of extreme events is done using a nonlinear correlation technique called Event Synchronization (ES). Using ES, it is possible to determine the degree of synchronism of extreme precipitation events between the CN nodes and, through their measurements, to identify the precipitation patterns that occur in the study area.