Análisis de los patrones de precipitación mediante la transformada rápida de Fourier

Abstract

This study addresses the issue of incomplete precipitation records caused by the irregular maintenance of meteorological stations, which impacts the accuracy of climate analyses. The objective of this research is to evaluate the performance of the Fast Fourier Transform (FFT) in identifying behavioral patterns in precipitation data and improving the interpolation of missing data. The methodology applied adopts a quantitative approach with an analytical scope and a non-experimental research design. The study focuses on a literature review of the Fast Fourier Transform and its implementation in Python, using data collected from meteorological stations during the period 2019-2022. The results indicate that the Fast Fourier Transform demonstrates higher efficiency and accuracy in capturing the periodic nature of climatic data, as evidenced by a low RMSE in the performed interpolations, outperforming conventional interpolation methods such as linear and polynomial techniques. This study makes a significant contribution by providing a solid foundation for the effective application of the FFT in studying meteorological phenomena while also delivering practical tools for managing incomplete climate data. This has important implications for improving climate predictions and water resource planning in the Chimborazo province.