In the past few decades, sea level rise (SLR) has been used as one of the most reliable proxies for evincing climate change impacts and significantly contributed to elevated coastal high-water levels around the globe. High tide flooding (HTF) has become more frequent along the U.S. coasts, and it is expected to become more frequent in the following decades. Thus, having an improved estimate of SLR along the coast is crucial for flood hazard mitigation and adaptation planning. There is a lack of a comprehensive framework that provides SLR and HTF flooding statistics at a reasonable spatial resolution that complements current point-based (tide gauge) estimations. To fill this gap, we developed a machine learning algorithm to extract the spatially distributed SLR and HTF thresholds using inputs from observational data. The outcome of this physics-informed machine learning methodology is SLR and HTF estimates under projected SLR by the mid-21st century Background