Evaluation of Geostatistical and Response Surface Methods for Estimating Soil Saturated Hydraulic Conductivity Using Soil Physical Properties

Soil Saturated Hydraulic conductivity is one of the most important soils physical properties, which has particular importance in studying and modeling the movement of water and salts in porous media. Because these methods are expensive and time consuming. Therefore, it seems necessary to estimate saturated hydraulic conductivity using of fast, low-cost methods with acceptable accuracy. In this research, geostatistical and response surface methods were used to estimate the hydraulic conductivity of soil saturation, the data was prepared from 135 soil profiles related to soil science and land reform studies of Ardabil Agricultural Jihad Organization in Fatah Ali Plain area. The main factor included saturated hydraulic conductivity, soil texture, bulk density, porosity, degree of saturation, and field capacity were considered as secondary factors. Among the geostatistical methods, ordinary cokriging had the best fit and the best fitted variogram in this method was the exponential model with zero nugget effect and 156 of sill range. That showed the strength of the spatial structure of the region and the full impact of the spatial structure on region variogram.  In the ordinary cokriging method, the estimation accuracy was estimated with the regression coefficient (R2=0.93) and (RMSE=3.21). In the response surface model, the best saturated hydraulic conductivity model based on the effective variables, was quadratic model with regression coefficient of 0.94 and the adequacy of accuracy of 49.2 was estimated, to determine the maximum hydraulic conductivity assuming the change of other effective parameters in the range of measured values, a target with desirability of 0.67 was obtained. This goal calculated the hydraulic conductivity of the soil at 8.9 (%) clay and 86 (%) sand, equal to 57.15 (mm/hour). Therefore, both methods are highly accurate in determining the hydraulic conductivity of soil saturation, but the response surface method (RSM) was more accurate.