Geoscience Reference
In-Depth Information
Figure B5.5.1 A comparison of values of soil moisture at capillary potentials of -10 and -100 cm curves
fitted to measured data and estimated using the pedotransfer functions of Vereeken et al. (1989) for
different locations on a transect (after Romano and Santini, 1997, with kind permission of Elsevier).
an indication of the significant uncertainty in the parameters and moisture characteristics for
a typical sand and clay soil using this method, based on the analysis of a database of 2085
samples. The greater uncertainty for the clay soil reflects the smaller number of samples of fine
textured soils in the database.
Pedotransfer functions for different soil texture classes have been included in the US
STATSGO soils database (USDA SCS, 1992) and the European HYPRES soils database (W osten,
1999). SOILPAR 2.0 (Acutis and Donatelli, 2003) includes 15 regression-based methods for
estimating different soil characteristics and functional curves. ROSETTA (Schaap et al. , 2001)
uses a neural network as a method of estimating soil parameters and SINFERS (McBratney
et al. , 2002) is an inference system based on rules.
Box 5.6 Descriptive Equations for Surface Flows
We consider only a one-dimensional (downslope or downstream) description of surface flows
here, but the principles apply also to the two-dimensional, depth integrated St. Venant equa-
tions. In one dimension, it is assumed that the flow can be adequately represented by a flow
velocity, v [LT −1 ], averaged over the local cross-sectional area, A [L 2 ] such that discharge
Q
vA [L 3 T −1 ]. Thus, since both v and A vary with discharge, there are two solution variables
but for surface flow it is less usual to assume a simple functional relationship between them
(although see the discussion of the kinematic wave approximation below). Thus two equations
=
 
Search WWH ::




Custom Search