Soil Erosion by Water/Rainsplash

← previous


The downslope component of the momentum of a single raindrop falling on a sloping soil surface is transferred in full to the soil surface. However, only a small proportion of the compoent normal to the surface is transferred, the remainder being reflected. The transfer of momentum to the soil particles has two effects:


It provides a consolidationg force compacting the soil and consequently intensifying crust formation. The formation of a surface crust results from clogging of the pores by soil compaction and by the infilling of surface pore spaces by fine particles detached from soil aggregates by the raindrop impact. Crusts have a dense surface skin or seal, about 0.1 mm thick, with well oriented clay particles. Beneath this is al layer, 1-3 mm thick, where the larger pore spaces are filled by finer washed-in material (Tackett and Pearson, 1965)[1]. Most aggregates on the soil surface are destroyed during an rainfall, while those in the lower layer of the crust remain intact, even though completely saturated (Farres, 1978)[2]. Although saturation reduces the internal strength of soil aggregates, they will not disintegrate until directly struck by raindrops.


The transfer of momentum to the soil particles produces a disruptive force as the water rapidly disperses from and returns to the point of impact in laterally flowing jets. The impact velocity of falling raindrops striking the soil surface varies from 4 m s−1 for a 1 mm diameter drop to 9 m s−1 for a 5 mm diameter drop. Local velocities of the lateral jets are about twice these (Huang et al., 1982)[3]. These fast moving water jets impart a velocity to some of the soil particles and launch them into the air, entrained within water droplets that are temselves formed by the break-up of the raindrop on contact with the ground (Mutchler and Young, 1975)[4]. Thus Raindrops are agents of both consolidation and dispersion.

The actual response of a soil to a given rainfall depends upon its moisture content and, therefore, its structural state and the rainfall intensity (kinetic energy). Le Bissonnais (1990)[5] describes three possible responses:

  • If the soil is dry and rainfall intensity is high, soil aggregates break down quickly by slaking. This is the breakdown by compression of air ahead of the wetting front. Infiltration capacity reduces rapidly and on smooth surfaces runoff can be generated after only a few millimeters of rain. With rougher surfaces, depression storage is greater and runoff takes longer to form
  • If the aggregates are initially partially wetted or rainfall intensity is low, mircocracking occurs and the aggregates break down into smaller aggregates. Surface roughness thus decreases but infiltration remains high because of large pore spaces between the microaggregates.
  • If the aggregates are initially saturated, large quantities of rain are required to seal the surface and infiltration capacity depends on the saturated hydraulic conductivity to the soil. Nevertheless, soils with less than 15 % clay are vulnerable to sealing if the intensity of the rain is high.

Over time, the percentage are of the soil surface affected by crust development intreases exponentially with cumulative rainfall energy (Govers and Poesen, 1985)[6], which, in turn, brings about an exponential decrease in infiltration capacity (Boiffin and Monnier, 1985)[7]. Crustability decreases with increasing contents of clay and organic matter since these provide greater aggregate stability to the soil. Thus loams and sandy loams are the most vulnerable to crust formation.

Soil Erosion by Rainsplash

Studies of the kinetic energy required to detach one kilogram of sediment by raindrop impact show that minimal energy is needed for soils with a geometric mean particle size of 0.125 mm and that soils with geometric mean particle size between 0.063 and 0.250 mm are the most vulnerable to raindrop detachment (Poesen, 1985)[8]: Silt loams, loams, fine sands and sandy loams. Coarser soils are resistant to detachment because of the weight of the large particles, finer soils are resistant because the raindrop energy has to overcome the adhesive or chemical bonding forces that link the minerals comprising the clay particles. Selective removal of particles by rainsplash can cause variations in soil texture downslope, affecting soil aggregates as well as primary particles.

The wide range in energy required to detach clay particles is a function of different levels of resistance in relation to the type of clay minerals and the relative amounts of  ,   and  Na cations in the water passing through the pores (Arulanandan et al., 1975)[9]. Overall, silt loams, loams, fine sands and sandy loams are the most detachable.

Besides texture, the detachability of a soil also depends on the shear strength of the top soil (Cruse and Larson, 1977)[10]. The detachment of soil particles represents a failure of the soil by the combined mechanism of compression and shear under raindrop impact, which is most likely to occur under saturated conditions when the shear strength of the soil is lowest (Al-Durrah and Bradford, 1982)[11]. Detachment decreases exponentially with increasing shear strength.

During a storm Raindrops sometimes do not directly fall on to the soil surface, but on surface water in the form of puddles or overland flow. As the thickness of the surface water layer increases, so does splash erosion (Palmer, 1964)[12]. This is due to the durbulence that impacting raindrops impart on to the water. Only on sandy soils splash erosion does not increase with water depth (Poesen, 1981)[13].

There is, however, a critical water depth beyond which erosion decreases exponentially with increasing water depth because more of the rainfall energy is dissipated in the water and does not affect the soil surface. This critical depth is approximately equal to the diameter of the raindrops (Palmer, 1964)[12].

Rainsplash DetachmentEdit

Experimental Studies have shown that the rate of detachment of soil particles with rainsplash varies with the 1.0 power of the instantaneous kinetic energy of the rain (Free, 1960[14]; Quansah, 1981[15]). The detachment rate  D r on bare soil can be expressed by equations of the form:


  is the kinetic energy of the rain and   is the depth of surface water film.   is the local slope for a distance equivalent to only a few drop diameters from the point of raindrop impact (e.g. on the side of a soil clod), and not the average ground slope.   may vary from 0.8 for sands to 1.8 for clays (Bubenzer and Jones, 1971)[16]. Values for   are in the range of 0.2-0.3 (Torri and Sfalanga, 1986)[17], those for   in the range of 0.9-3.1 (Torri et al., 1987b)[18], varying with the texture of the soil.

Rainsplash TransportEdit

In contrast, average ground slope S is important when considering the overall transport of splashed particles. On a sloping surface more particles are thrown downslope than upslope during the detachment process, resulting in a net movement of material downslope. Splash transport  per unit width of slope can be expressed by the relationship:


where  (Meyer and Wishmeier, 1969)[19], and   decreases from 1.0 to minimum values of 0.8 and 0.75 where ground slope angles rise to 20 and 25°, respectively (Mosley, 1973; De Ploey et al., 1976)[20], and becomes negative on steeper slopes (Foster and Martin, 1969; Bryan, 1979)[21].

These relations ignore the role of wind. Windspeed imparts a horizontal force to a falling raindrop until its horizontal velicity component equals the wind velocity. As a result, the kinetic energy of the raindrop is increased. Detachment of soil particles by impacting wind-driven raindrops can be 1.5-3 times greater than resulting from rain of the same intensity without wind (Disrud and Krauss, 1971[22]; Lyles et al., 1974[23]). Wind also causes raindrops to strike the surface at an angle from vertical. This affects the relative proportions of upslope versus downslope splash (Moeyersons, 1983)[24].

Splash erosion is most important for detaching the soil particles that are subsequently eroded by running water (interrill overland flow, rill flow). However, on upper sections of hillslopes, splash transport may be the dominant erosion process. As runoff and soil loss increase, the importance of splash transport declines. Since splash erosion acts uniformly over the land surface its effexts are seen best where stones or tree roots selectively protect the underlying soil and splash pedestrals or soil pollars are formed.

→ next


  1. Tackett, J. and Pearson, R. (1965). Some characteristics of soil crust formed by simulated rainfall. Soil Science, 99:407–413.
  2. Farres, P. (1978). The role of time and aggregate size in the crusting process. Earth Surface Processes, 3:243–254.
  3. Huang, C., Bradford, J., and Curshman, J. (1982). A numerical study of raindrop impact phenomena: the rigid case. Soil science of America Journal, 46:14–19.
  4. Mutchler, C. and Young, R. (1975). Soil detachment by raindrops. In Present and prospective technologyfor predictiong sediment yields and sources, volume 40, pages 113–117. USDA-ARS Publication.
  5. Le Bissonnais, Y. (1990). Experimental study and modelling of soil surface crusting processes. Catena Supplement, 17:13–28.
  6. Govers, G. and Poesen, J. (1985). A field-scale study of surface sealing and compaction on loam and sandy loam soils. part i. spatial variability of soil surface sealing and crusting. In Callebaut, F., Gabriels, D., and De Boodt, M., editors, Assessment of soil surface crusting and sealing, pages 171–182. Flanders Research Centre for Soil Erosion and Conservation, Gent.
  7. Boiffin, J. and Monnier, G. (1985). Infiltration rate as affected by soil surface crusting caused by rainfall. In Callebaut, F., Gabriels, D., and De Boodt, M., editors, Assessment of soil surface crusting and sealing, pages 210–217. Flanders Research Centre for Soil Erosion and Conservation, Gent.
  8. Poesen, J. (1985). An imporved splash transport model. Zeitschrift für Geomorphologie, 29:193–211.
  9. Arulanandan, K., Loganathan, P., and Krone, R. (1975). Pore and eroding fluid influences on the surface erosion of a soil. Journal of the Geotechnical Engineering Division ASCE, 101:53–66.
  10. Cruse, R. and Larson, W. (1977). Effect of soil shear strength on soil detachment due to raindrop impact. Soil science of America Journal, 41:777–781.
  11. Al-Durrah, M. and Bradford, J. (1982). Parameters for describing soil detachment due to single water drop impact. Soil science of America Journal, 46:836–840.
  12. a b Palmer, R. (1964). The influence of a thin water layer on water-drop impact forces. International Association of Hydrological Sciences Publication, 65:141–148.
  13. Poesen, J. (1981). Rainwash experiments on the erodibility of loose sediments. Earth Surface Processes and Landforms, 6:285–307.
  14. Free, G. (1960). Erosion characteristics of rainfall. Agricultural Engineering, 41:447–449,455.
  15. Quansah, C. (1981). The effect of soil type, slope, rain intensity and their interactions on splash detachment and transport. Journal of Soil Science, 32:325–332.
  16. Bubenzer, G. and Jones, B. (1971). Drop size and impact velocity effects on the detachment of soil under sumulated rainfall. Transactions of the American Society of Agricultural Engineers, 14:625–628.
  17. Torri, D. and Sfalanga, M. (1986). Some problems on soil erosion modelling. In Giorgini, A. and Zingales, F., editors, Agricultural nonpoint source pollution: Model selection and application, pages 161–171. Elsevier, Amsterdam.
  18. Torri, D., Sfalanga, M., and Del Sette, M. (1987b). Splash detachment: Runoff depth and soil cohesion. Catena, 14:149–155.
  19. Meyer, L. and Wishmeier, W. (1969). Mathematical simulation of the process of soil erosion by water. Transactions of the American Society of Agricultural Engineers, 12:754–758,762.
  20. Mosley, M. (1973). Rainsplash and the convexity of badland divides. Zeitschrift für Geomorphologie, Supplementband, 18:103–107.
  21. Foster, R. and Martin, G. (1969). Effect of unit weight and slope on erosion. Journal of the Irrigation and Drainage Division ASCE, 95:551–561.
  22. Disrud, L. and Krauss, R. (1971). Examining the process of soil detachment from clods exposed to wind driven simulated rainfall. Transactions of the American Society of Agricultural Engineers, 14:90–92.
  23. Lyles, L., Dickerson, J., and Schmeidler, M. (1974). Soil detachment from clods by rainfall: effects of wind, mulch cover and initial soil moisture. Transactions of the American Society of Agricultural Engineers, 17:697–700.
  24. Moeyersons (1983). Measurements of splash-saltation fluxes under oblique rain. Catena Supplement, 4:19–31.