Evaluation and Development of Pedotransfer Functions and Artificial Neural Networks to Saturation Moisture Content Estimation

Modeling of irrigation and agricultural drainage requires knowledge the soil hydraulic properties. However, uncertainty in direct measurement saturation moisture content (θs) has been generated several methodologies for its estimation, such as Pedotransfer Functions (PTFs) Artificial Neuronal Networks (ANNs). In this work, eight different PTFs were developed which relate to proportion sand clay, bulk density (BD) well saturated conductivity (Ks). addition, ANNs with combinations input hidden layers estimation θs. The results showed R2 values from 0.9046≤R2≤0.9877 PTFs, while ANNs, R2>0.9891 obtained. Finally, root-mean-square error (RMSE) was obtained each ANN configuration, ranging 0.0245≤RMSE≤0.0262. It found that particular characteristic parameters (% Clay, % Silt, Sand, BD Ks), accurate estimate θs is With development these models (PTFs ANNs), high 10 12 textural classes.