Inadequate turfgrass irrigation management poses a significant challenge, resulting in considerable water loss through runoff and the transport of contaminants, ultimately jeopardizing surface and groundwater quality. This study introduces a Machine Learning (ML)-based Decision Support System (DSS) designed to optimize turfgrass irrigation, concurrently minimizing runoff and preserving turfgrass quality. A robust ML classifier, specifically the Radial Basis Function - Support Vector Machine (RBF-SVM) was trained on synthetic data generated through the Monte-Carlo (MC) technique, which was then used to specify a set of irrigation rules implemented in the irrigation controller. The synthetic data were derived from observations collected from irrigation plots at the Texas A&M University Turfgrass Laboratory in Texas, United States, with Soil Wetting Efficiency Index (SWEI) serving as the target variable. When tested against a commercially available irrigation controller, the ML-based controller significantly reduced runoff by an average of 74 % while maintaining high Green Cover (GC) in turfgrass, achieving an accuracy of 87 %. These findings highlight the potential of ML-driven irrigation systems to improve water use efficiency, reduce environmental impact, and maintain turf quality. Such systems could be beneficial for urban landscapes, sports fields, and agriculture, helping users conserve water while achieving sustainable turf management.