A simulation study of the soil moisture content under a prairie field, in G
embloux Belgium, by the use of the one-dimensional model SWAP93, was carrie
d out. In order to determine the optimum relationship that S-max is followi
ng, two concepts of maximum water extraction rate were examined. The first
one assumed a linear variation of S-max with depth z and the second assumed
a homogeneous distribution of S-max with depth z. Five statistical criteri
a were used to compare the quality of simulation results, such as average e
rror (AE), root mean square error (RMSE), root mean square (RMS), modeling
efficiency (EF) and coefficient of residual mass (CRM). The differences bet
ween the criteria, showed that the assumption of the homogeneous distributi
on of S-max throughout the soil profile resulted in a more accurate predict
ion of soil moisture content. The agreement between measured and simulated
water content profiles, throughout the regarded period, was satisfactory fo
r depths >30 cm. The deviation between simulated and experimental values fo
r depth <30 cm could be explained by the influence from the cracks in the s
oil surface and wormholes that cause water losses and, consequently, undere
stimation of the experimental values. (C) 2000 Elsevier Science B.V. All ri
ghts reserved.