Assessment of Pedotransfer Functions (PTFs)‎ in predicting soil hydraulic properties under arid and semi arid environments

Abstract EN

The main objectives of this study were to assess the applicability and the prediction accuracy of the most frequently cited and some recently developed PTFs that use soil data that we have such as : particle-size distribution, organic matter content, and dry bulk density to predict soil properties that we need such as saturated hydraulic conductivity, soil water content at field capacity and permanent wilting point, and available water content under arid and semi-arid Environments.

For this purpose eighteen widely used PTFs were selected.

In order to quantify the prediction accuracy, root mean square error (RMSE), mean error (ME), mean absolute error (MAE), and Pearson correlation (R) were used.

The PTFs showed good to poor prediction accuracy with RMSE ranged from 0.00208 to 0.01738 m3 m-3, ME values ranged from 0.00991 to 0.17364 m3 m-3 and with MAE ranged from 0.04707 to 0.17364 m3 m-3.

The validation indices showed the PTFs of BSSTOPSOIL and HYPERS were most accurate for our evaluation data set to estimate soil water content at field capacity, RAWLS, RAWLS-BRAKENIEK, MANRIQUE PTFs were the most accurate for soil water content at permanent wilting point, HYPERS PTF showed the best PTF in available water content estimation and COSBY PTF has the highest negative significant correlation, lowest ME, MAE, and RMSE in saturated hydraulic conductivity estimation.

The results indicated that saturated hydraulic conductivity, soil water content at field capacity and permanent wilting point, and available water content can be estimated for soils using PTFs when the laboratory measurements are not available.

However, a local evaluation is needed before using any available PTFs, moreover, further studies will be necessary to assess the validity of the estimation.