Projeções hidrológicas em duas bacias hidrográficas a partir de cenários futuros impactados por mudanças climáticas no século XXI

Abstract

Bearing in mind the importance of water resources for individuals, as a means of subsistence and economic development, in addition to being one of the essential components for a balanced environment, the objective of this research was to obtain hydrological perspectives in the Lower Rio Ivaí Sub-basin (SBRI) and in the Upper Rio Tibagi Sub-basin (SART) in the state of Paraná in seasonal and annual periods, based on hydroclimatological variables, through future scenarios impacted by climate changes in the 21st century. Inicially, the ordinary kriging technique, using the stable geostatistical model, was applied to pluviometric data from the Águas Paraná Institute, INMET and IAPAR resulting in maps generated with the observed data showing gaps against the filled data. The verification of the modeling efficiency was conducted through the evaluation of the EMA, REQM, r, d (Willmott), NS (Nash-Sutcliffe) statistics and the 5% Wilcoxon test. In parallel, a deterministic-stochastic hybrid rainfall-runoff model was developed, using historical hydroclimatological records of the sub-basins obtained from the Águas Paraná Institute, in which initially the equivalent flow concept was applied for later application of a regressive model. between observed and estimated accumulated flows. The results were evaluated by the Shapiro-Wilk (S-W), Komogorov-Smirnov (K-S), Wilcoxon (WCX), and Brown-Forsythe (B-F) tests at the 5% significance level. Finally, for the generation of flows, considering two scenarios of possible climate changes (C1 and C2), the PGECLIMA_R program was used coupled to the deterministic-stochastic hybrid rainfall-runoff model and the precipitation results were represented spatially using the stable model while flow rates were represented on time. Regarding the results of the first analysis, there were relatively low values of EMA and REQM, as well as expressive values of the r, d and NS indices, mainly in SART, which had a lower percentage of failures. The statistical tests applied to the results of the rain-flow model indicated very good performance in the projection of flows, with optimal results for the K-S test. With regard to the last stage of this work projections of flows impacted by climatic changes predicted for the end of the century, indicated that there will possibly be an increase in flow, mainly in the rainy periods. Thus, through the research carried out, it was possible to estimate probable changes in precipitation and flows, which can be used as a tool for further studies, as well as a tool to support the action of decision makers in the future.