STATISTICAL MONITORING OF SEA LEVEL VARIABILITY: A CONTROL CHART APPROACHAPPROACHES

Monitoring temporal changes in sea level is important in assessing coastal risk. Sea level anomalies at a tide gauge station, if kinematically conceived, include systematic variations such as trend, acceleration, periodic oscillations, and random disturbances. Among them, the non-stationary nature of the random sea level variations of known or unknown origin at coastal regions has been long recognized by the sea level community. This study proposes the analyses of subgroups of random residual statistics of a rigorously formulated kinematic model solution of tide gauge variations using X-bar and S control charts. The approach is demonstrated using Key West, Florida tide gauge records. The mean and standard errors of 5-year-long subgroups of the residuals revealed that sea level changes at this location have been progressively intensifying from 1913 to the present. Increasing oscillations in sea level at this locality may be attributed partly to the thermal expansion of seawater with increasing temperatures causing larger buoyancy-related sea level fluctuations as well as the intensification of atmospheric events including wind patterns and the impact of changes in inverted barometer effects that will alter coastal risk assessments for the future