Mapping rainfall erosivity at a regional scale: a comparison of interpolation methods in the Ebro Basin (NE Spain)

Abstract

Abstract. Rainfall erosivity is a major causal factor of soil erosion, and it is included in many prediction models. Maps of rainfall erosivity indices are required for assessing soil erosion at the regional scale. In this study a comparison is made between several techniques for mapping the rainfall erosivity indices: i) the RUSLE R factor and ii) the average EI 30 index of the erosive events over the Ebro basin (NE Spain). A spatially dense precipitation data base with a high temporal resolution (15 min) was used. Global, local and geostatistical interpolation techniques were employed to produce maps of the rainfall erosivity indices, as well as mixed methods. To determine the reliability of the maps several goodness-of-fit and error statistics were computed, using a cross-validation scheme, as well as the uncertainty of the predictions, modeled by Gaussian geostatistical simulation. All methods were able to capture the general spatial pattern of both erosivity indices. The semivariogram analysis revealed that spatial autocorrelation only affected at distances of ∼15 km around the observatories. Therefore, local interpolation techniques tended to be better overall considering the validation statistics. All models showed high uncertainty, caused by the high variability of rainfall erosivity indices both in time and space, what stresses the importance of having long data series with a dense spatial coverage.

Rain event properties at the source of the Blue Nile River

“…For instance, kinetic energy, which is a commonly used indicator for the ability of rainfall to detach soil particles, is related to rain rate (see van Dijk et al, 2002 for an extensive review). The existence of such relation suggests that variability of rain event properties can lead to variations in kinetic energy in space and time (see Angulo-Martínez et al, 2009). Based on the results of the present study we speculate that rain events of mountain areas of Gilgel Abbay watershed have less soil erosive power (due to relatively small rain rate) than rain events of lowlands of this watershed.…”

Spatial interpolators for improving the mapping of carbon stock of the arboreal vegetation in Brazilian biomes of Atlantic forest and Savanna

“…MAE represents a measure of the average error, T indicates a prediction bias and is centered around zero and Willmontt's D index has been used for evaluation of accuracy of models/methods. This index reduces the outliers' effect in statistical value (Willmontt, 1984) and as higher is D value greater is the model (Angulo-Martinez et al, 2009). In addition to these statistical indices, the scatterplots of the predicted values around 1:1 line and Shapiro-Wilk test (at a = 0.05 of probability) were evaluated.…”

Improving the global applicability of the RUSLE model – adjustment of the topographical and rainfall erosivity factors

“…These factors represent the triggers for soil erosion by providing energy for soil to erode. They can also be seen as the natural components of the RUSLE model, as they include very little or no modification by human activities (Angulo-Martínez et al, 2009) apart from indirect effects on precipitation and extreme events due to anthropogenic climate change. In this way they represent the natural environmental constraints to soil erosion that are important to capture before the effect of human activities on soil erosion through land use change can be investigated.…”