Chesapeake Bay Land Subsidence and Sea Level Change: An Evaluation of Past and Present Trends and Future Outlook.

摘要

Ten Chesapeake Bay water level stations presently have a combined total of 647 years of water level measurements with record lengths varying between 35 years (1975-2009) at the Chesapeake Bay Bridge Tunnel, VA, and 107 years (1903-2009) at Baltimore, MD. All ten stations, with the exception of Gloucester Point, VA, are active stations in the National Water Level Observation Network of water level stations maintained by the U.S. National Oceanic and Atmospheric Administration, Center for Operational Oceanographic Products and Services. New technologies such as sea surface range measurements from earth-orbiting satellites now provide a global assessment of absolute sea level (ASL) trends relative to the center of a reference ellipsoid rather than fixed points on the earths surface to which relative sea level (RSL) measurements refer. New methodologies have also been applied to derive spatial averages of ASL trends over large regions with greater accuracy. Notwithstanding these advances, there is still no substitute for an accurate time series of water level measurements obtained locally, preferably one spanning several decades, when assessing RSL trends that will affect a specific community or township in the coming decades. RSL trends will determine local inundation risk whether due to vertical land movement (emergence or subsidence) or the ASL trend found as the sum of RSL trend and land movement when both are measured positive upward. In Chesapeake Bay, RSL trends are consistently positive (rising) while land movement is negative (subsiding). By choosing a common time span for the ten bay stations evaluated in this report, we are able to compare differences in RSL rise rates with approximately the same degree of confidence at each station. Uncertainty has been reduced by extracting the decadal signal present at all ten stations before using linear regression to obtain new RSL rise rates with smaller than usual confidence intervals, permitting both temporal and spatial comparisons to be made.