Stochastic modeling of basins microtopography: analysis of spatial variability and model testing

Microtopography is among the most important factors affecting the performance of basin irrigation system due to its influence on the advance and recession processes. This study is based on field-measured surface elevation of 116 basins in North China. The spatial variability of basins microtopography was analyzed using geostatistics; the spatial structure of basins microtopography can be characterized by a spherical semivariogram model. The correlations between selected basin geometry parameters, mainly the standard deviation (S _d ) of surface elevation differences (SED), and the semivariogram parameters were calculated and allow estimating the semivariogram parameters from basin characteristics. Considering the randomness of SED and, simultaneously, its spatial dependence, a procedure was developed to model the spatial distribution of SED using Monte-Carlo generation and kriging interpolation techniques. The required number of SED generations was also estimated depending upon the S _d of SED. The SED stochastic generation model was tested by comparing the advance, recession, flow water depths and performance parameters observed in an experimental basin with those simulated using measured and model generated SED data. Results show that estimation errors from using generated data are similar to those resulting from observations. Thus, SED generated data may be used for assessing the impacts of microtopography on irrigation performance.