Visual Validation of the E-RUSLE Model Applied at the Pan-European Scale. Bosco, C., de Rigo, D., & Dewitte, O. 1:mri11a13+.
Visual Validation of the E-RUSLE Model Applied at the Pan-European Scale [link]Paper  doi  abstract   bibtex   
Validating soil erosion estimates at regional or larger scale is still extremely challenging. The common procedures are not technically and financially applicable for large spatial extents, despite this some options are still applicable. For validating the European map of soil erosion by water calculated using the approach proposed in Bosco et al. [1] we applied alternative qualitative methods based on visual evaluation. The 1 km 2 map was validated through a visual and categorical comparison between modelled and observed soil erosion. A procedure employing high-resolution Google Earth images and pictures as validation data is here shown. The resolution of the images, rapidly increased during the last years, allows for a visual qualitative estimation of local soil erosion rates. A cluster of 3x3 K m 2 around 85 selected points was analysed by the authors. The results corroborate the map obtained applying the e-RUSLE model. The 63\,% of a random sample of 732 grid cells are accurate, 83\,% at least moderately accurate with a bootstrap p ≤ 0.05). For each of the 85 clusters, the complete details of the validation also containing the comments of the evaluators and the geo-location of the analysed areas have been reported.
@article{boscoVisualValidationERUSLE2014,
  title = {Visual Validation of the E-{{RUSLE}} Model Applied at the Pan-{{European}} Scale},
  author = {Bosco, Claudio and de Rigo, Daniele and Dewitte, Olivier},
  date = {2014},
  journaltitle = {Scientific Topics Focus},
  volume = {1},
  pages = {mri11a13+},
  doi = {10.6084/m9.figshare.844627},
  url = {https://doi.org/10.6084/m9.figshare.844627},
  abstract = {Validating soil erosion estimates at regional or larger scale is still extremely challenging. The common procedures are not technically and financially applicable for large spatial extents, despite this some options are still applicable.

For validating the European map of soil erosion by water calculated using the approach proposed in Bosco et al. [1] we applied alternative qualitative methods based on visual evaluation. The 1 km 2 map was validated through a visual and categorical comparison between modelled and observed soil erosion.

A procedure employing high-resolution Google Earth images and pictures as validation data is here shown. The resolution of the images, rapidly increased during the last years, allows for a visual qualitative estimation of local soil erosion rates.

A cluster of 3x3 K m 2 around 85 selected points was analysed by the authors. The results corroborate the map obtained applying the e-RUSLE model. The 63\,\% of a random sample of 732 grid cells are accurate, 83\,\% at least moderately accurate with a bootstrap p ≤ 0.05).

For each of the 85 clusters, the complete details of the validation also containing the comments of the evaluators and the geo-location of the analysed areas have been reported.},
  keywords = {*imported-from-citeulike-INRMM,~INRMM-MiD:c-13340371,bootstrap,e-rusle,europe,featured-publication,land-cover,remote-sensing,rusle,scientific-topics-focus,semantic-array-programming,semap,soil-erosion,soil-resources,statistics,uncertainty,validation,visual-assessment,visual-interpretation},
  options = {useprefix=true},
  series = {Notes {{Transdiscipl}}. {{Model}}. {{Env}}.}
}

Downloads: 0