Investigating the status of some macronutrients using digital mapping techniques in Maru’ak area, Lorestan Province, Iran

Abstract

Abstract The present study was conducted 1: to compare generalized linear model (GLM), random forest (RF), and Cubist, 2: to produce available phosphorus (AP) and potassium (AK) maps, and 3: to identify covariates controlling mineral distribution in Maru’ak area, Lorestan Province. To fulfill the goals, the location of 173 soil samples was determined by the cLHS method, in four different land uses including orchards, paddy fields, and agricultural and abandoned fields. The performance of models was assessed by the R2, RMSE, and MAE. Results showed that the RF model fitted better than GLM and Cubist models, and could explain 40 and 57% of AP and AK distribution, respectively. The R2, RMSE, and MAE for the RF model were 0.4, 2.81, and 2.43 for predicting AP; and were 0.57, 143.77, and 116.61 for predicting AK, respectively. The most important predictors selected by the RF model were valley depth and SAVI for AP and AK, respectively. The maps showed higher amounts of AP and AK in apricot orchards compared to other land uses, and no difference was observed between AP and AK content of paddy fields, agricultural and abandoned area. The higher amounts of AP and AK were related to orchard management, such as not removing plant residuals and fertilizer consumptions. It can be concluded from the present study that the orchards were the best land use for the study area, which increases soil quality and is in line with sustainable management. However, before generalizing the results, more detailed research is needed.