Regionalization of reference streamflows for the Araguaia River basin in Brazil

Abstract

Hydraulic projects and water management require reliable hydrological data. The Araguaia-Tocantins River basin, in addition to agricultural use, has great potential for hydroelectric exploitation. However, the streamflow monitoring network in the Araguaia River basin is composed of only a few stations, resulting in a lack of hydrological data. The regionalization of the reference streamflows is a technique that can help circumvent this lack of data, enabling the estimation of streamflows from easily obtainable explanatory variables. In this context, the objective of this study was to develop regional functions for the maximum streamflow (Qmax) applicable to different Return Periods (RP), the long-term mean streamflow (Qmlt) and the 95% streamflow permanence (Q95) of the upper and middle Araguaia River sub-basins. The dimensionless streamflow methodology was adopted with the drainage area as an explanatory variable. The tested regressive models were the linear, potential and quotient models. Leave-one-out cross-validation was used to assess the quality of the regional models. Ten statistical distributions of 2 to 5 parameters were used. (i) Satisfactory results were obtained for all reference streamflows. (ii) The cross-validation technique proved to be essential for the selection of the most robust model. (iii) The quotient model was shown to be superior to the potential linear model in most cases.