Within Arctic deltas, surficial hydrologic connectivity of lakes to nearby river channels influences physical processes like sediment transport and ice phenology as well as biogeochemical processes such as photochemistry. As the Arctic hydrologic cycle is impacted by climate change, it is important to quantify temporal variability in connectivity. However, current connectivity detection methods are either spatially limited due to data availability constraints or have been applied at only a single time step. Additionally, the relationship between connectivity and lake ice is still poorly quantified. In this study, we present a multitemporal classification and validation of lake connectivity in the Colville River Delta, AK. We introduce a connectivity detection algorithm based on remote sensing of water color that is expandable to other high-sediment Arctic deltas. Comparison to validation datasets suggests that detection of high vs. low connectivity lakes is accurate in 69.5–85.5% of cases. Connectivity temporally varies in about 20% of studied lakes and correlates strongly with discharge and lake elevation, supporting the idea that future changes in discharge will be drivers of future changes in connectivity. Lakes that are always highly connected start and end ice break up an average of 26 and 16 days earlier, respectively, compared to lakes that are never connected. Because spring and summer ice conditions drive Arctic lake photochemistry processes, our research suggests that surface connectivity is an important parameter to consider when studying biogeochemistry of Arctic delta lakes.