Estimation of flow components by recursive filters: case study of Paracatu River Basin (SF-7), Brazil

The quantification of flow components is important for water resource management. The base, interflow, and runoff flows of nested basins of the Paracatu River (SF-7) were estimated with recursive signal filters in this study. At first, stationary analysis and multivariate gap filling were applied to the flow data. A methodology is presented, proposing filter calibration with the runoff influence and inflection of the recession curve for dry season. Filters are improved with a logic constraint that limits water flow overestimation within algorithm iteration. The results were coherent with previous studies and with hydrogeological and climatological maps.