A geostatistical approach to optimize the determination of saturated hydraulic conductivity for large-scale subsurface drainage design in Egypt

Measurements of saturated hydraulic conductivity (K-s) in the field are costly, time-consuming, and relatively cumbersome, chiefly as hydraulic conductivity exhibits a large spatial variability, so that it becomes difficult to find accurate representative values to correctly predict soil-water flow and design irrigation and drainage systems. K-s was measured in seven different soils in Egypt to evaluate its spatial variability and to develop a model for estimating its representative value for a large-scale subsurface drainage design. Published data from East Delta was also used. Results showed that the spatial structure of K-s is characterized by a high nugget effect with a correlation range varying from 1600 to 2700 m and is fairly correlated with the agricultural practices and geologic nature of field soils. Based on the concepts of geostatistics, a simple correlation model was developed for estimating reliable and rapid representative values of K-s. The validity of that model was tested statistically and on field data of one Nile Delta soil and one Nile Valley soil. The results indicated that the model will be practically valuable for estimating the representative value of K-s that could be used in the drainage design of small blocks or large areas. The model was applied to the design drain spacings used in Egypt and to estimate the minimum sample size required for estimating a mean value of K-s at a given precision level taking into account the spatial variability of K-s. The results showed that neglecting spatial variability of K-s may affect the design drain spacing by -27% to 3%, and overestimate the required sample size by about 76%. Such a model may be regarded as a helpful tool for drainage design oriented professionals without prior knowledge of geostatistics procedures. It is necessary to adequately characterize large areas to which hydrologic models, which require K-s, are to be applied. Furthermore, the magnitude of spatial dependence of K-s presented in this paper may be of meat help for a better understanding and modeling of water and solutes movement in, and through, the agricultural clay soils in Egypt. (C) 2000 Elsevier Science B.V. All rights reserved.