Resumo:
Knowing the hydrological variables with a certain amount of time in advance is essential for planning strategies that relate to the management of water resources. The study area that covers this type of information is called hydrological forecasting, which consists of the representation of the physical phenomena of a hydrographic basin, through mathematical equations, aiming to estimate flows in the future. The present study focuses on the analysis of the flow tools by set in the sub-seasonal horizon in the hydrographic basin of the Paracatu River, which has irrigation as the main form of water use. Calibration of the MGB-IPH model was carried out, subsequent hydroclimatic forecasting as a whole, integrating the hydrological and climatic models, and finally, an analysis of the performance indicators was carried out. The results of the calibration of the hydrological model were considered satisfactory, based on the Nash, Nash-Log, Volume Error, Bias and dispersion graphs. The results of the probabilistic forecast are better when compared to the results of the deterministic forecast, obtained by the BS and BSS indicators. Through the indicators H, POD, FAR and BIAS, the results of the probabilistic forecast are better satisfactory for most of the sub-basins, but limited to the first months of forecast (January, February, April) and with greater efficiency in the first 20-30 days of forecast. The results were unsatisfactory and the flows were overestimated for some sub-basins, for example 42460000 and 42490000, (which claim interference from the Queimados HPP) mainly in the last months of the year (September, October) and in the last 30-40 days of forecast.