Análisis descriptivo de caudales y precipitaciones a diferentes escalas de tiempo

Abstract

A descriptive analysis of the historical records of flows, levels and rainfall in 7 hydrometeorological stations located in different places of the country is performed in this project, the analysis of the statistical behavior of the time series is carried out through the use of integrated Autoregressive Models of Moving Averages (ARIMA). Three-time scales (daily, monthly and annual) are taken into consideration in each of the hydrological parameters to carry out the modeling and the predicted results of the forecasts are compared using two methodologies: Forecasts with models of the same time scale and models with models to different time scale. The results indicate that the best forecast performance occurs with the use of models of the same time scale when there are affected historical records, otherwise, the use of models at a different time scale may generate better results.