Spatial modeling of soil water erosion associated with agroclimatic seasonality in the southern region of Rio Grande do Sul, Brazil

ABSTRACT

Water erosion is a serious soil degradation problem, with impact in several areas. Its measurement is extremely important and costly. Empirical models of soil loss estimation, such as the revised universal soil loss equation (RUSLE), are used to meet this demand. Few studies in Brazil are considered to evaluate the effect on agroclimatic seasonality in the estimates of soil loss due to water erosion in watersheds. Thus, the objective of this study was to evaluate the agroclimatic seasonality in the estimation of soil losses through RUSLE and to identify the factors that control erosion in the watershed of the Fragata stream. Rainfall erosivity (R) was calculated by means of precipitation data for four seasons and annual average. The soil erodibility factor (K) was obtained from soil information. The topographic factor (LS) was generated from the Digital Elevation Model (MDE) and the soil cover factor and conservationist practices (CP) through Landsat8/OLI satellite images. Seasonal variation had an effect on soil loss, with higher erosion rates in the summer and spring months. Soil losses between 5 and 50 Mg ha^-1ano^-1 were recorded in 24% of the BHAF, associated with periods of more erosive rainfall, higher slope and low vegetation cover. RUSLE factors with the greatest contribution to erosion were R, LS, and CP. The approach presented can be useful in quantifying soil losses in river basins.

Keywords:
geoprocessing; revised universal soil loss equation; seasonal variability; soil loss; soil conservation