Soil moisture prediction plays a vital role in developing plants, soil properties, and sustenance of agricultural systems. Considering this motivation, in this study, an intelligent Multi-output regression method was implemented on daily values of meteorological and soil data obtained from Kemalpaşa-Örnekköy station in Izmir, Turkey, at three soil depths (15, 30, and 45 cm) between the years 2017 and 2019. In this study, nine different machine learning algorithms (Linear Regression (LR), Ridge Regression (RR), Least Absolute Shrinkage and Selection Operator (Lasso), Random Forest (RF), Extra Tree Regression (ETR), Adaptive Boosting (AdaBoost), Gradient Boosting (GB), Extreme Gradient Boosting (XGBoost), and Histogram-Based Gradient Boosting (HGB)) were compared each other in terms of MAE, RMSE, and R 2 metrics. The experiments indicate that the implemented Multi-output regression models show good soil moisture prediction performance. Also, the ETR algorithm provided the best prediction performance with an 0.81 R 2 value among the other models.

This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record.