I'm not sure, but i would say R is not the answer you are looking for. Soil or particle systems can be simulated with a pysical model called DEM. Google for LIGGGHTS, EDEM or PFC3D to get more information about the common simulation software. R could probably handle only a statistic approach but not a physical model of the gravel particles.
I'm not sure, but i would say R is not the answer you are looking for. Soil or particle systems can be simulated with a pysical model called DEM. Google for LIGGGHTS, EDEM or PFC3D to get more information about the common simulation software. R could probably handle only a statistic approach but not a physical model of the gravel particles.
There is a 2015 paper in the IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing titled Characterisation of Soil Erosion Indicators Using Hyper spectral Data From a Mediterranean Rained Cultivated Region by Thomas Schmidt, Manuel Rodríguez-Rastrero, Paula Escribano, Alicia Palacios-Orueta, Ella Ben-Dor, Antonio Plaza, Robert Milewski, Margarita Huesca, Ashley Bracken, Víctor Ticuéndez, Marta Pelayo and Sabine Chabrillat (available at http://www.umbc.edu/rssipl/people/aplaza/Papers/Journals/2015.JSTARS.Soil.pdf) which uses a couple of R packages for the spatial mapping of soil erosion. The authors might be a good place to begin.
1) Look at EcoHydrology and related packages (SWATmodel provides an implementation of the SWAT model. See http://swat.tamu.edu/ for more information about the SWAT model.)
https://r-forge.r-project.org/R/?group_id=956 | List of all packages provided by project EcoHydrology
2) Look in the Soil Science section of the CRAN Environmetrics Task View
https://cloud.r-project.org/web/views/Environmetrics.html | Soil Science section (search for it)