Over the last few decades, inland surface waters have been increasingly recognized as critically important to global climate, biodiversity, and security (Vörösmarty et al., 2010) as substantial changes have been observed in permanent and seasonal water bodies around the world. During that time, 90,000 km 2 of formerly permanent waters have disappeared, 72,000 km 2 have transitioned from permanent to seasonal, and 213,000 km 2 of new permanent water has come into existence (Pekel et al., 2016) underscoring a need for increased understanding of inland surface water dynamics at local, regional, and global scales. Trends in surface water change have been detected globally and are likely the product of climate change, natural variability, and human impacts (Pekel et al., 2016;Rodell et al., 2018), and yet our knowledge and understanding of these changes is limited due to measurement limitations at the global scale (Alsdorf et al., 2007). Historically, our primary source of water surface elevations has been in-situ hydrological gauge station networks installed at individual lake and reservoir sites. Unfortunately, due to costs and logistics, in-situ monitoring stations are only available for a small subset of lakes globally and suffer greatly from uneven spatial and temporal distribution (Alsdorf et al., 2007).Remote sensing has long been used to supplement in-situ networks by providing measurements of water surface area via single-band thresholds, water indices (Feyisa et al., 2014;McFeeters, 1996;Xu, 2006), tasseled cap wetness (Crist, 1985), principle component analysis (Lira, 2006), supervised and unsupervised classifications, as well as advanced methods recently developed including machine-learning, pixel-level fusion, and spectral matching of discrete particle swarms (