Monthly mean sea level (MSL) data for 128 long-term National Water Level Observation Network (NWLON) stations of the Center for Operational Oceanographic Products and Services (CO-OPS) are analyzed in this report. All available data up to the end of 2006 are used to determine linear trends, average seasonal cycles, and interannual variability including estimated errors. The stations are located on the U.S. Atlantic and Pacific coasts, the Gulf of Mexico, Hawaii, Alaska, and on islands in the Atlantic and Pacific Oceans.
The linear trends obtained are relative MSL trends which are a combination of the absolute global rate of sea level rise (1.7 +/- 0.5 mm/yr in the 20th century) and the rate of any local vertical land motion. The variation in vertical land motion, ranging from rapid subsidence in Louisiana and eastern Texas to rapid uplift in Alaska, is primarily responsible for the regional differences in MSL trends and for the differing rates within regions. Separate pre- and post- seismic trends were calculated for some stations in Alaska and Guam with apparent seismic offsets in 1957, 1964, or 1993.
Time series plots of the monthly MSL data with the seasonal cycle removed are located in the appendices along with the 12-month average seasonal cycle for each station. The average seasonal cycles are used to derive the two tidal constituents that represent the regular seasonal variation which are then compared to the tidal constituents routinely used by CO-OPS to make the official tide predictions. The residual time series after the seasonal cycles and trends are removed represent the regional oceanic interannual variability, which is highly correlated from station to station. Using a 5-month running average of the residual, thresholds of +0.1 and -0.1 meters are defined for positive and negative anomalies.
Each calculated linear trend has an associated 95% confidence interval that is primarily dependent on the year range of data for each station. A derived inverse power relationship indicates that 50-60 years of data are required to obtain a trend with a 95% confidence interval of +/- 0.5 mm/yr. This dependence on record length is caused by the interannual variability in the observations. A series of 50-year segments were used to obtain linear MSL trends for the stations with over 80 years of data. None of the stations showed consistently increasing or decreasing 50-year MSL trends, although there was statistically significant multidecadal variability on the U.S. east coast with higher rates in the 1930s, 1940s and 1950s and lower rates in the 1960s and 1970s.
The long-term MSL changes at NWLON stations require that CO-OPS periodically introduce a new 19-year National Tidal Datum Epoch (NTDE) every 20-25 years to keep the datums up-to- date. In specific areas with rapid rates of vertical land motion, CO-OPS has adopted special 5- year Modified Tidal Datum Epochs (MTDEs) to prevent the datum elevations from becoming obsolete before the next nationwide update. In this report, it is recommended that CO-OPS