Results from the 14 land surface parameterization schemes involved in
the PILPS-RICE Workshop are compared for a soya crop growing season (f
rom June to September). During this period, the transpiration flux dom
inates the total surface evapotranspiration and observed data from HAP
EX-MOBILHY are available for comparison. Results indicate that during
the month of June half of the models fall within the uncertainty range
of the observations. The scatter between models behaviour is explaine
d by three major reasons: The functional dependency between soil moist
ure and transpiration; the initial moisture content at the beginning o
f the period; the vertical discretization within the soil and the exte
nsion of the root system that defines the soil water holding capacity
for plants Examination of diurnal cycles of evaporation reveals that f
ormulations based on the supply-demand concept are very sensitive to t
he specification of the root zone. This analysis underlines the need f
or more sensitivity experiments to be done with the current forcing da
ta set and more detailed datasets to be collected in future field expe
riments (e.g. latent heat flux during all the growing season, root zon
e distribution).