Artificial subsurface drainage systems (DS) exert significant impacts on agricultural production, local hydrology, and the transportation of agro-chemicals to aquatic environments. With increasing focus on technology driven farm management and environmental concerns, airborne and spaceborne remote sensing (RS) studies for DS detection are increasing. However, a systematic review detailing the methodologies for DS detection using RS systems is currently lacking. This study presents a comprehensive review of 19 remote sensing subsurface drainage system mapping studies, encompassing a diverse array of imagery, acquisition periods, and detection methods, with the aim of identifying best practices for detecting subsurface DS. These studies aim either to delineate the actual DS tile networks or to identify areas or fields where DS systems are likely installed. While DS detection has traditionally relied on visual interpretation by human analysts, the recent advent of machine learning and deep learning techniques in RS image analysis has enabled their application in DS detection, facilitating coverage of much larger areas. Our findings highlight the advantages of timing image acquisition in relation to rainfall and field conditions. As well as analyzing different methods for automatic detection and delineation of DS. However, disparities in or the absence of standardized evaluation methods pose challenges for robust comparisons of methodologies and datasets. Nonetheless, the integration of machine learning and deep learning holds promise for large-scale and automated DS detection. Based on our findings, we present recommendations for future research directions in the field of RS-based DS detection, emphasizing the necessity for standardized evaluation frameworks and ongoing advancements in analytical techniques.