Surface water flow and storage play important roles in the regulation of water quality, yet the link between precipitation and surface water dynamics is poorly understood for agricultural watersheds like the Canadian Lake Erie Basin (CLEB). In this study, we developed a data-driven approach to address the following questions: (1) How do land cover and terrain affect the correlation between precipitation and inundated area? (2) How is antecedent precipitation related to inundated area? Specifically, we derived a near-monthly inundated area time series based on Landsat imagery acquired from 1984 to 2020 and quantified correlations with various precipitation indices for all quaternary watersheds across the CLEB. Strong correlations were observed between surplus precipitation (P-PET) and inundated area across the quaternary watersheds in the CLEB overall. However, this correlation was found to be considerably weaker in croplands compared to wetlands, suggesting a strong role of sub-surface drainage in agricultural areas. Finally, correlation between P-PET and inundated area were stronger when considering cumulative P-PET over three months in wetlands, pointing to the role of wetlands in longer-term storage and watershed memory. Our results underscore the importance of wetlands in hydrologic regulation in the CLEB, where loss of wetlands alters surface water response and decreases residence time. This study demonstrates a novel, observation-driven method to assess the link between precipitation and surface water dynamics, particularly in seasonally and ephemerally inundated environments, allowing for an improved understanding of the effects of climate and land use changes on surface hydrology in agricultural watersheds.