Boon, J.D., 2012. Evidence of sea level acceleration at U.S. and Canadian tide stations, Atlantic Coast, North America.

Evidence of statistically significant acceleration in sea level rise relative to land is found in a recent analysis of monthly mean sea level (mmsl) at tide stations on the Atlantic coast of North America. Serial trend analysis was used at 11 U.S. Atlantic coast stations and 1 Canadian station (Halifax, Nova Scotia) with record lengths exceeding 75 years to examine change in the linear trend rate of rise over time. Deriving trend estimates that apply in the median year of fixed-length mmsl series, reversals in rate direction (increasing or decreasing) were observed around 1939–40 and again in the mid-1960s except at the northeasternmost stations in the latter period. What has not been observed until recently is a sharp reversal (in 1987) followed by a uniform, near-linear change in rise rate that infers constant acceleration at eight mid- to NE Atlantic tide stations, change not seen at SE U.S. Atlantic stations. Quadratic regression and analysis of variance applied to mmsl series over the last 43 years (1969–2011) confirms that addition of a quadratic term representing acceleration is statistically significant at 16 tide stations from Virginia to Nova Scotia. Previous quadratic model studies have focused on sea level series of longer spanning periods with variable serial trends undermining quadratic expression of either accelerating or decelerating sea level. Although the present 43-year analysis offers no proof that acceleration will be long lived, the rapidity of the nascent serial trend increase within the region of interest is unusual. Assuming constant acceleration exists and continues, the regression model projects mmsl by 2050 varying between 0.2 and 0.9 m above mean sea level (MSL) in the NE region and between −0.3 and 0.4 m above MSL in the SE region.