Identifying the Causes of the Recent Sea Level Change Along the East Coast of North America

• Main Author: Kenigson, Jessica Sarah
• Format: Text
• Language: unknown
• Published: CU Scholar 2018
• Subjects: atlantic multidecadal oscillation; bayesian dynamic linear regression modeling; empirical mode decomposition; hycom-cice; north atlantic oscillation; sea level hotspot; Oceanography; Physical Sciences and Mathematics; Greenland; Ice Sheet; North Atlantic; North Atlantic oscillation; Sea ice
• Online Access: https://scholar.colorado.edu/atoc_gradetds/79; https://scholar.colorado.edu/cgi/viewcontent.cgi?article=1079&context=atoc_gradetds

A “hotspot” of accelerated sea level rise along the northeast coast of North America has been detected in tide gauge records and has been hypothesized to result from weakening of the Atlantic Meridional Overturning Circulation (AMOC). In modeling studies, buoyancy forcing in the North Atlantic Ocean, such as from surface ocean warming and accelerated Greenland Ice Sheet melt, weaken the AMOC. However, local and remote winds are thought to be major contributors to the interannual to decadal variability of sea level north of Cape Hatteras. In addition, there is controversy concerning the ability of common time series analysis methods to isolate acceleration due to modes of climate variability, such as the Atlantic Multidecadal Oscillation (AMO) and the North Atlantic Oscillation (NAO), from acceleration due to anthropogenic forcing. In this dissertation, the accuracy of two mathematical methods, Empirical Mode Decomposition and Ensemble Empirical Mode Decomposition, in quantifying sea level acceleration in the long-term trend is tested using a suite of experiments on synthetic and historical tide gauge records. The sea level acceleration in the hotspot is assessed across methods, record lengths, and other parameters. Multidecadal variability associated with the AMO is found to be a likely contributor to the recent sea level acceleration in the hotspot. In addition, the influence of the NAO on sea level in four geographical regions along the western boundary on interannual timescales is quantified using Bayesian Dynamic Linear Regression Modeling, which captures the time-varying regression relationship. A decadal shift of the NAO effect on sea level in the hotspot around 1987 is detected and attributed to changes in NAO-linked regional wind stress. Finally, a series of experiments with an ocean general circulation model (HYCOM) coupled to a sea ice model (CICE) are used to isolate the effects of buoyancy forcing from wind forcing on sea level in the hotspot on interannual to decadal timescales. The model experiments are interpreted in light of various high-quality in situ, reanalysis, and satellite data products. Buoyancy forcing is found to play a role on decadal timescales and wind forcing is found to be important on interannual to decadal timescales.