Sea level acceleration and variability in the Chesapeake Bay: past trends, future projections, and spatial variations within the Bay

Fast sea level rise (SLR) is causing a growing risk of flooding to coastal communities around the Chesapeake Bay (hereafter, CB or "the Bay "), but there are also significant differences in sea level variability and sea level rise rates within the bay that have not been fully investigated in the past. Therefore, monthly sea level records for 1975-2021 from eight tide gauge stations, from the upper bay at Baltimore, MD, to the lower bay at Norfolk, VA, are analyzed and compared. The results show significant spatial variations within the Bay over a wide range of time scales. The largest contribution to the seasonal variations of mean sea level in the Bay is from the annual (S-A) and semiannual (S-SA) tides, while the contribution from thermosteric changes is relatively smaller. The lower Bay has a similar to 5 cm smaller mean annual sea level range than the upper Bay and has a secondary minimum in mid-year due to a larger semiannual tide than the upper Bay which is dominated by the annual tide. Variations in sea level anomaly (after removing the mean seasonal cycle) show anticorrelation between the upper and lower bay. Empirical mode decomposition (EMD) analysis reveals that variations with opposite phases at the two edges of the Bay appear mostly on decadal time scales that are linked with the North Atlantic Oscillation (NAO). Sea level trends vary along the Bay-linear SLR rates (4.5-6.1 mm y(-1)) increase from north to south, while sea level acceleration rates (all positive in the range 0.012-0.16 mm y(-2)) increase from south to north. The linear SLR pattern is driven by land subsidence rates, while the acceleration pattern suggests potential impacts from climate change signals that enter the mouth of the Bay in the southeast and amplified farther north by local dynamics. Monthly sea level projections until 2100, based on past trends and the seasonal cycle of each station, are compared with different SLR scenarios based on climate models. The results suggest that accounting for local sea level acceleration in projections can result in large differences in local future sea level rise.