Regional decadal trends in sea level and implications for determining global mean sea level change and its causes

Author: C. Wunsch, Rui M. Ponte and P. Heimbach
Date: 
June 16, 2006
Type: 
Presentation
Venue: 
Understanding Sea-level Rise and Variability Workshop, UNESCO/ IOC, Paris, France
Citation: 

Wunsch, C., R.M. Ponte, and P. Heimbach, 2006. Regional decadal trends in sea level and implications for determining global mean sea level change and its causes, Understanding Sea-level Rise and Variability Workshop, UNESCO/ IOC, Paris, France, June 2006.

Estimates of regional patterns of sealevel change are obtained by combining the MIT general circulation model (1°x1°, 23 vertical levels, 80S-80N) and several hundred million ocean observations in a constrained optimization procedure for the period 1993-2004. The data include not only the available altimetry (TOPEX/Poseidon, Jason, GFO, ERS, Envisat), but most of the modern hydrography, ARGO float profiles, seasurface temperature and other observations. Adjustable parameters are initial temperature and salinity and NCEP-NCAR reanalysis forcing fluxes. Regional variations greatly exceed existing estimates of the mean rise in sealevel, including some regions with sealevel fall. Although thermal effects are largest, the contributions from salinity changes and mass redistribution within the oceans are both significant. Contributions below 1000m are also important, but not dominant. The inferred sealevel change is similar to other estimates derived strictly from data analyses, but there are quantitative differences. Given the complexity of the regional trends and the sparsity of the in situ hydrographic database, considerable uncertainty remains in the attribution of global mean sealevel change to thermal expansion or other effects. A major obstacle is the absence of atmospheric fluxes that are consistent in their conservation of freshwater and heat, and the lack of good regional uncertainty estimates for those fluxes.