TY - JOUR

T1 - Sensitivity analysis of EUROSEM using Monte Carlo simulation II:

T2 - the effect of rills and rock fragments

AU - Veihe, A.

AU - Quinton, John

AU - Poesen, J.

PY - 2000/4/15

Y1 - 2000/4/15

N2 - A sensitivity analysis of the surface and catchment characteristics in the European soil erosion model (EUROSEM) was carried out with special emphasis on rills and rock fragment cover. The analysis focused on the use of Monte Carlo simulation but was supplemented by a simple sensitivity analysis where input variables were increased and decreased by 10%. The study showed that rock fragments have a significant effect upon the static output parameters of total runoff, peak flow rate, total soil loss and peak sediment discharge, but with a high coefficient of variation. The same applied to the average hydrographs and sedigraphs although the peak of the graphs was associated with a low coefficient of variation. On average, however, the model was able to simulate the effect of rock fragment cover quite well. The sensitivity analysis through the Monte Carlo simulation showed that the model is particularly sensitive to changes in parameters describing rills and the length of the plane when no rock fragments are simulated but that the model also is sensitive to changes in the fraction of non-erodible material and interrill slope when rock fragments were embedded in the topsoil. For rock fragments resting on the surface, changes in parameter values did not affect model output significantly. The simple sensitivity analysis supported the findings from the Monte Carlo simulation and illustrates the importance when choosing input parameters to describe both rills and rock fragment cover when modelling with EUROSEM.

AB - A sensitivity analysis of the surface and catchment characteristics in the European soil erosion model (EUROSEM) was carried out with special emphasis on rills and rock fragment cover. The analysis focused on the use of Monte Carlo simulation but was supplemented by a simple sensitivity analysis where input variables were increased and decreased by 10%. The study showed that rock fragments have a significant effect upon the static output parameters of total runoff, peak flow rate, total soil loss and peak sediment discharge, but with a high coefficient of variation. The same applied to the average hydrographs and sedigraphs although the peak of the graphs was associated with a low coefficient of variation. On average, however, the model was able to simulate the effect of rock fragment cover quite well. The sensitivity analysis through the Monte Carlo simulation showed that the model is particularly sensitive to changes in parameters describing rills and the length of the plane when no rock fragments are simulated but that the model also is sensitive to changes in the fraction of non-erodible material and interrill slope when rock fragments were embedded in the topsoil. For rock fragments resting on the surface, changes in parameter values did not affect model output significantly. The simple sensitivity analysis supported the findings from the Monte Carlo simulation and illustrates the importance when choosing input parameters to describe both rills and rock fragment cover when modelling with EUROSEM.

KW - soil erosion modelling

KW - sensitivity analysis

KW - Monte Carlo simulation

KW - rock fragment cover

KW - model uncertainty

KW - EROSION-MODEL EUROSEM

KW - SOIL-EROSION

KW - SEDIMENT YIELD

KW - OVERLAND-FLOW

KW - SURFACE

KW - COVER

KW - TRANSPORT

KW - POSITION

KW - RAINFALL

KW - RUNOFF

U2 - 10.1002/(SICI)1099-1085(20000415)14:5<927::AID-HYP979>3.0.CO;2-V

DO - 10.1002/(SICI)1099-1085(20000415)14:5<927::AID-HYP979>3.0.CO;2-V

M3 - Journal article

VL - 14

SP - 927

EP - 939

JO - Hydrological Processes

JF - Hydrological Processes

SN - 0885-6087

IS - 5

ER -