R-factor in Italy

Rainfall erosivity in Italy: A national scale spatiotemporal assessment

Resume of the study

Soil erosion by water is a serious threat for the Mediterranean region. Raindrop impacts and consequent runoff generation are the main driving forces of this geomorphic process of soil degradation. The potential ability for rainfall to cause soil loss is expressed as rainfall erosivity, a key parameter required by most soil loss prediction models. In Italy, rainfall erosivity measurements are limited to few locations, preventing researchers from effectively assessing the geography and magnitude of soil loss across the country. The objectives of this study were to investigate the spatio-temporal distribution of rainfall erosivity in Italy and to develop a national-scale grid-based map of rainfall erosivity. Thus, annual rainfall erosivity values were measured and subsequently interpolated using a geostatistical approach. Time series of pluviographic records (10-years) with high temporal resolution (mostly 30-min) for 386 meteorological stations were analysed. Regression-kriging was used to interpolate rainfall erosivity values of the meteorological stations to an Italian rainfall erosivity map (500-m). A set of 23 environmental covariates was tested, of which seven covariates were selected based on a stepwise approach (mostly significant at the 0.01 level). The interpolation method showed a good performance for both the cross-validation dataset (R2ev = 0.777) and the fitting dataset (R2=0.779).

Figure 1. Spatial distribution of the meteorological stations employed in this study. The rainfall erosivity values (R-factor) are ranked into five classes determined through a quantile-based analysis.

 

Figure 2. Block-diagram of the seasonal R-factor spatial distribution.

 

Figure 3. Grid-based map of rainfall erosivity in Italy at a 500-m resolution.

 

Figure 4. Comparison between the rainfall erosivity map of Italy obtained in this study through a regression-kriging approach (0.5 × 0.5 Km cell size) (left) and the maps proposed by Van der Knijff, et al. (1999, 2000) (1Km x 1Km cell size) (middle) and by Bosco et al. (2015) (ca. 25 × 25 km cell size) (right). Rainfall erosivity values (R-factor) are ranked into five classes determined through quantile-based analyses.

References

Borrelli, P., Diodato, N., Panagos, P. 2016. Rainfall erosivity in Italy: a national scale spatio-temporal assessment. International Journal of Digital Earth. In Press. DOI:10.1080/17538947.2016.1148203

Displaying 1 - 1 of 1 |