Comparison of two methods for determining Q95 reference flow in the mouth of the surface catchment basin of the Meia Ponte river, state of Goiás, Brazil

Abstract

The preferred data for analyzing water availability are those of historical flow series of the sources of interest; however, most Brazilian watersheds do not have sufficient fluviometric monitoring. Such cases require techniques for transposing data from one region to another, otherwise known as ‘flow regionalization’. The present work aimed to compare the method proposed by Secretaria de Meio Ambiente e Desenvolvimento Sustentável (SEMAD) of the state of Goiás with the traditional method of regionalization for determining reference flow at the mouth of the surface catchment basin of the Meia Ponte river. Data from eight fluviometric stations were used for regionalization, with the regression equations being adjusted using four different models. The result revealed that the potential and linear models performed the best, both with R2 and R2a values of 0.996 and 0.995, respectively. The relative error in the application of the potential model and of the method adopted by SEMAD were below 30%. The reference flows obtained by the two best performing methods differed, with flow determined by the traditional method being 5.93% lower than that of the SEMAD equation. Therefore, a more detailed study is recommended to determine which equation models better fit the region.