path analysis of effective soil properties on paddy soil saturated hydraulic conductivity

abstract the saturated hydraulic conductivity (ks) is an important physical property of soil. the direct measurement of this property in soils is a difficult and time consuming process. pedotransfer functions (ptfs) provide an alternative by estimating soil parameters from more readily available soil data. the objective of this study is to determine effective factors on soil hydraulic conductivity using multiple-linear regression methods and to determine direct and indirect effects of input soil properties of model using path analysis method. 70 soil samples with medium to heavy texture were randomly collected from paddy fields in guilan province after harvest of rice crop. range of ph indicated acidic to neutral condition of paddy soils. results showed that regression equation using geometric mean and standard deviation of soil particles diameter (dg and g), bulk density (b) and soil moisture at field capacity (0.033) as input variables, can estimate hydraulic conductivity with a good accuracy ( rmse= 0.5 and r2adj=0.84). although correlation between bulk density and saturated hydraulic conductivity was positive, but according to path analysis results, direct effect of bulk density on ks is negative. moreover the highest direct effect of soil properties on ks in paddy soil was through soil water content at field capacity that indicates importance of this factor to predict saturated hydraulic conductivity.