Investigating the Effect of Different Agricultural Land Uses on the Optimal Water Content for Tillage Using Decision Tree Model in Jiroft Plain

Optimum soil water content (OSWC) for tillage is the moisture at which, if tillage is done, the maximum amount of crushing (friability) is created in the soil with minimum energy consumption. The purpose of this research was to determine the OSWC for tillage operations under different agricultural land uses in Jiroft Plain, Iran, and to evaluate the effectiveness of the decision tree model in this regard for the region. For this purpose, soil samples (90 samples) were taken intact from 0 to 20 cm depth and some of its physical and chemical characteristics were determined. To determine the soil water characteristics curve, the soil water contents at matric suctions of 10, 30, 50, 100, 300, 500, 1000 and 1500 kPa were measured using a pressure plate device. The results showed that the type of cultivation did not have a significant effect on the physical and chemical characteristics of the soil, including the optimum weighted humidity. OSWC prediction was done with decision tree. In this research, three prediction accuracy criteria, namely, coefficient of determination (R2), root mean square error (RMSE), and normalized root mean square error (NRSME) were used to evaluate the model performance. The results showed that use of decision tree for predicting the OSWC was appropriate due to having a lower NRMSE (0.017) and a high R2 (0.75). Also, the results of OSWC modeling showed organic matter as the most influencing variable, followed by the equivalent calcium carbonate percentage, electrical conductivity, apparent density, and percentage of sand and silt.