부산대학교 도서관

In this work, a time-dependent analysis is presented to identify climatic patterns that affect precipitation behavior in the lower and middle Paraguay River basin, corresponding to the territory of Paraguay. For this purpose, satellite precipitation images were used for the period from 2001 to 2020, available on the GIOVANNI - NASA platform. The results of the exploratory analysis of classical statistics allowed the identification of data distribution, probability density, and monthly, seasonal, and annual statistical measures. It was observed that the monthly average precipitation of the time series reaches a maximum in August 2015 (312.4 mm) and a minimum in August 2004 (6.5 mm). It was determined that December is the month with the highest monthly average precipitation in the series, while August has the lowest precipitation. On the other hand, stochastic models were applied to identify and evaluate the possibility of generating precipitation forecasts using the autoregressive ARIMA model, with unsatisfactory results because the series does not meet the necessary assumptions for the implementation of these models. A Fourier analysis was performed to detect periods of high and low flow, clearly identifying characteristic or dominant frequencies related to cycles of 1, 4, 5, 20, 10, and 0.5 years. The application of this technique allowed relating the occurrence of anomalies in the flows to hydroclimatic cycles during the two phases of the El Niño/Southern Oscillation (ENSO) phenomenon that occur in tropical South America, including Paraguay, among other countrie). The space image analysis identified the accumulation of the monthly average precipitation over 20 years, alternating between areas of Low Chaco with low precipitation (47 mm) and others with higher (332 mm) towards the east and center of the basin. Finally, Voronoi (Thiessen) interpolation was performed using monthly average precipitation data from the stations and the Isoyetas map, allowing visualization of precipitation patterns in the studied area.