Actualización de información hidrológica mediante la obtención de curvas IDF y procesos de clusterización correspondiente a zonas de interés de estudio del austro ecuatoriano

Abstract

The proper management of water resources and the planning of sustainable infrastructure in the region require a detailed understanding of precipitation patterns, especially due to the significant climatic variability that characterizes this area. This thesis aims to zone precipitation intensities in the southern region of Ecuador through the determination of IDF (IntensityDuration-Frequency) curves and the application of clustering techniques. Pluviometric data from stations operated by PROMAS-University of Cuenca, ETAPA EP, and INAMHI were compiled, and the data were validated using probability distributions such as Gumbel and Pearson type III. Goodness-of-fit tests, including the Kolmogorov test and Pearson's R² and R coefficients, were applied to generate IDF curves that reflect the relationship between precipitation intensity, duration, and frequency for different return periods.The clustering of data, carried out using techniques such as Spectral Clustering and K-means, was evaluated with quality indices like Silhouette, Davies-Bouldin, and Calinski-Harabasz, identifying zones with homogeneous precipitation regimes. Subsequently, interpolation using the Cressman method was applied, and intensity-duration-frequency maps were generated for various return periods. The results were integrated through the superposition of rasters to obtain a joint zoning of precipitation intensities. Finally, strategic zones for future monitoring were proposed to enhance the representativeness and precision of pluviometric information. This study provides essential tools for planning and design in water resource management, contributing to the development of sustainable infrastructure in the region.