Regimes de precipitação e dinâmica da vegetação em diferentes contextos geológico-geomorfológicos em ambiente tropical semiárido: Caatinga no Alto Sertão paraibano

• Ano de defesa: 2024
• Autor(a) principal: Araújo, Elânia Daniele Silva
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraíba; Brasil; Geografia; Programa de Pós-Graduação em Geografia; UFPB
• Palavras-chave em Português: Geologia e geomorfologia - Alto Sertão paraibano; Índices de vegetação; Sazonalidade; Comportamento da vegetação; Vegetação de Caatinga; Precipitação pluviométrica; Vegetation indices; Veasonality; Biomass; Vegetation behavior; CNPQ::CIENCIAS EXATAS E DA TERRA::GEOCIENCIAS
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/123456789/34139

The tropical semi-arid regions are characterized by the seasonality of rains alternating between dry and rainy periods. Thus, vegetation in these areas has developed adaptations that align with this seasonal cycle. Climate seasonality, especially precipitation, influences vegetation seasonality, causing the annual growth cycle of vegetation to be directly related to precipitation. In addition to precipitation, vegetation seasonality depends on other factors such as lithological and topographic characteristics, types of vegetation arrangements, soil properties, or a combination of these factors. Seasonally Dry Tropical Forests (SDTF) exhibit a typical pattern of the annual phenological cycle, with values influenced by the rainy season and delimited by phenological parameters marking the beginning and end of the rainy season. Studying the correlation between the dynamics of this vegetation and precipitation is of paramount importance, as it allows the identification of the period of the most significant influence of precipitation on its behavior. Therefore, this research aims to understand the behavior of Caatinga vegetation arrangements in different geological-geomorphological contexts of the Alto Sertão of Paraíba based on their correlation with precipitation data. Among the methodological procedures used, obtaining lithological, topographic, precipitation, and vegetation data stands out, using satellite images and the Normalized Difference Vegetation Index – NDVI. As a basis for the analyses, areas classified as native vegetation by Mapbiomas were considered. Statistical analyses were conducted using clustering, Pearson correlation, and simple linear regression between NDVI data and cumulative precipitation in different geological-geomorphological contexts. Areas with preserved vegetation arrangements were identified with the highest NDVI averages. These areas were mainly in granitoid complex lithology, with higher altitudes and slopes. Degraded areas had the lowest NDVI averages, predominantly featuring medium to low altitudes and slopes, with most metasedimentary lithology and sand/sandstone. Four clusterings were performed with geological-geomorphological, precipitation, and vegetation data separately, and one encompassed all these variables together. The clusterings proved effective, identifying areas with similar and distinct vegetation behaviors based on the variables used. The four clusterings expressed the behavior of vegetation in different scenarios. Groupings made with variables separately showed greater homogeneity among areas, while the grouping considering all variables exhibited more variation, especially concerning altitudes, slopes, and lithologies. The relationship between precipitation and vegetation proved inseparable, with precipitation being the variable that most influenced the behavior of Caatinga. Degraded areas showed a faster response to cumulative precipitation, while areas with greater diversity exhibited a slower response. R values were higher in clusterings where variables were used separately. For those involving all variables, the values were lower. Regressions showed that it is possible to predict the seasonal behavior of Caatinga vegetation. However, future research is needed to improve regression models and test other models to enhance predictions for SDTF in different geologicalgeomorphological scenarios. This response happens because these variables are important and influence vegetation behavior, but statistically, they reduce the precision of the models. More in-depth studies are required to integrate phenological and phytosociological data with estimates from sensors for a better understanding of the seasonal behavior of SDTF.