Curve number estimation from Brazilian Cerrado rainfall and runoff data

Abstract

The Curve Number (CN) method has been widely used to estimate runoff from rainfall events in Brazil; however, CN values for use in the Brazilian savanna (Cerrado) are poorly documented. In this study we used experimental plots to measure natural rain-fall-driven rates of runoff under undisturbed Cerrado and under the main crops found in this region, and derive associated CN values from the measured data using five different statistical methods. Curve numbers obtained from the standard USDA Natural Resources Conservation Service (NRCS) table were suitable to estimate runoff for bare soil, soybeans (Glycine max [L.] Merr.), and sugarcane (Saccharum L.). However, CN values obtained from measured rainfall-runoff data (CN calibrated) provided better runoff estimates than the CN values from the standard table. The best CN values for the bare soil (hydrologic soil group B), soybeans, and sugarcane were 81.2 (78.5 to 83.9), 78.7 (75.9 to 81.5), and 70.2 (67.8 to 72.6). The CN method was not adequate to estimate runoff for the undisturbed Cerrado, bare soil (hydrologic soil group A), pasture, and millet (Pennisetum glaucum).