Splash detachment Ds (g.s-1) is simulated as a function of soil aggregate stability, rainfall kinetic energy and the depth of the surface water layer. The kinetic energy can arise from both direct throughfall and drainage from leaves. Using splash tests in the original LISEM research the following general equation has been derived (unpublished data):
Ds = (2.82/As Ke exp(-1.48 WH) + 2.96) P A /dt
in which Ds is splash detachment (g.s-1), As is the aggregate stability (median number of drops to decrease the aggregate by 50%), Ke is rainfall or throughfall kinetic energy (J.m-2.mm-1), WH is the depth of the surface water layer (mm), P is the amount of rainfall or throughfall under the plant canopy in the timestep (mm), A is the surface over which the splash takes place (m2) and dt is the time step (s). Note that the empirical factor 2.82 converts the dimensionless drop number As into a factor with units g/J, while the factor 2.96 has as units g.mm-1.m-2. Although this suggests there is splash even when Ke is zero, this will not happen because in that case there will be no rainfall P.
Ds_rponded : A = (fpa)(1-cover)A, and Ke is Ker
Ds_tponded : A = (fpa)(cover)A, and Ke is Ket
Ds_rdry : A = (1-fpa)(1-cover)A, and Ke is Ker
Ds_tdry : A =(1-fpa)(cover)A, and Ke is Ket
The kinetic energy of free rainfall and throughfall from the plant canopy as respectively:
Ker = 8.95+8.44*log(I)
Ket = 15.8*sqrt(h)-5.87
where I is the rainfall intensity (mm/h) and h is the height of the plants (m). In equations 8a to 8d f is the splashdelivery fraction, a user defined fraction that determines the amount of splashed soil that is transported through the air from the dry part of the gridcell to the wet part of the gridcell, so that is can be transported). This means that although most splash occurs on the dry part of the gridcell because of the exponential decrease of splash with water height, only a part of it will be transported.