Modelagem de variáveis meteorológicas: um estudo de caso do Brasil

Abstract

This dissertation investigates climatic variability and its impacts in Brazil, highlighting the diversity of regional climate patterns and the relevance of forecasting extreme events, such as droughts and floods, that directly affect the country. The primary objective of this research is to study correlations of meteorological variables in Brazil through the cross-correlation coefficients ρDCCA and DMC2 x . The theoretical foundation includes a review of statistical models applied to meteorological data and time series, emphasizing the use of cross-correlation analyses and long-range memory methods, such as detrended fluctuation analysis (DFA). These methods are essential for developing strategic solutions aimed at mitigating the impacts of climate change and improving the understanding and forecasting of extreme weather events in Brazil. The study addressed climatic variability in Brazil, emphasizing the importance of analyzing meteorological conditions in different regions of the country, which present distinct climatic characteristics. The research utilized data from the National Institute of Meteorology (INMET), selecting 26 active stations between 2009 and 2019, allowing a comprehensive overview of solar radiation, temperature, humidity, and wind speed. The methodology included data processing and the application of statistical methods, such as cross-correlations, to investigate the relationships between the variables. Preliminary results indicated significant correlations, especially between solar radiation and temperature, with the expectation that the analyses will contribute to planning and decision-making in various fields of knowledge. The statistical analysis of the meteorological data revealed large variations in solar radiation, air temperature, relative humidity, and wind speed in different Brazilian cities. Using Tukey’s boxplot method, it was observed that cities like Natal and Jo˜ao Pessoa exhibited high solar radiation values, while relative humidity varied considerably between regions. The cross-correlation between solar radiation and air temperature showed a strong relationship in several stations, except in Cuiab´a and Teresina. The ρDCCA analysis highlighted significant correlations between solar radiation and relative humidity, especially in regions like the North and Northeast. The DMC2 x method allowed for a more comprehensive understanding of the interactions between the variables, essential for specialties such as sustainable management and adaptation to climate change, considering the particularities of each biome and their influences on regional climatic conditions. Although the research showed relevant results, limitations such as the collection time and the number of variables hinder definitive conclusions about global warming and its causes.