Scaling of cumulative infiltration curves using pedotransfer functions

Surface irrigation is used more than pressurized irrigation due to low cost, more efficient energy requirements, and better aeration of the root zone. The design, evaluation and simulation of surface irrigation systems relies on a knowledge of the infiltration properties of the soil and of the movement of water over the field. Spatial variation of infiltration rates makes the management of surface irrigation systems a very complex process. Therefore, presentation of a general equation for infiltration is very difficult. One proper way to characterize a general infiltration equation is to use a scaling method. Cumulative infiltration was measured by double ring method in nine sites in the Ardabil area, Iran. Soil properties of the sites (particle size distribution, bulk density, particle density, and organic carbon) were measured. The essential part of a scaling method is to find the characteristic time tc and space scales Lc for the system to be modeled. To formulate the Lc and tc, effective parameters (saturated hydraulic conductivity, saturated water content and initial water content) must be considered. Direct measurement of the saturated hydraulic conductivity is time-consuming and costly. Alternatively, pedotransfer functions (PTFs) may be used to indirectly estimate the saturated hydraulic conductivity from easy-to-measure soil properties. Ability of some PTFs (ROSETTA, SOILPAR, and five regression PTFs) which estimate the saturated hydraulic conductivity were tested for merging of the diverse infiltration curves towards a single scaled curve. The merging of the diverse measured infiltration curves towards a single scaled curve showed that the use of the effective parameters estimated by ROSETTA just with particle size distribution as input in the scaling factors works better than the others PTFs. Finally, a regression software package was used to obtain a scaled infiltration equation (Philip´s infiltration equation) for the scaled infiltration curves.