Análise de eventos extremos de precipitação na estação pluviométrica no Município de Peixe (TO), código 83228, utilizando limiar dinâmico z-score (z ≥ 3)

Abstract

The statistical analysis of extreme precipitation events is an essential tool for understanding the phenomenon, particularly in identifying patterns (trends and anomalies) in time series. This study aims to identify and analyze extreme precipitation events at the PEIXE (TO) station during the period from 1980 to 2024. For this purpose, daily precipitation data from the PEIXE – TO pluviometric station, code 83228, are used. The data are organized and processed in Microsoft Excel 365 spreadsheets, where filters are applied to remove negative, null, and zero values. Extreme Value Statistics (EVS) is applied, with emphasis on the Z-score technique. The threshold corresponding to the Z-score (Z ≥ 3) is dynamically calculated by month and by decade. The results show a higher occurrence of extreme events in the summer months, especially in December, January, and March, as well as a recent intensification in typically dry months, such as July and August. The analysis captures seasonal and decadal trends, highlighting changes in the local rainfall regime. The results provide an important basis for management plans and adaptation strategies in the face of climate change.