Marine Science Faculty Publications

Global Ocean Warming and Sea Level Rise

Document Type

Conference Proceeding

Publication Date



The ocean observing system has progressed considerably over the past 50 years, enabling more accurate estimation of global ocean heat content and its impact on sea level (thermosteric sea level). For the entire 50-year record, estimates of global thermosteric sea level rise are about 0.4 mm/yr, much less than the total sea level rise of 1.8 mm/yr for the same period. However, this estimated 50-year heating rate may be biased low due to undersampling of the oceans, particularly in the Southern Hemisphere during the early decades. For the period 1993 – 2003, with high precision satellite altimetry and a quasi-global upper ocean thermal network, thermosteric sea level rise was about 1.6 mm/yr for the layer 0-750 m, out of a total sea level rise of 2.8 mm/yr. Considerable regional and interannual variability are evident. The large increase in the estimated heating rate relative to the earlier period may have been due partly to decadal variability and partly to improved global coverage by the measurements. The recent implementation of the global Argo array has now made it possible to estimate 1-year averages of thermosteric sea level with accuracy of 0.5 mm, and hence the error in 10-year change is less than 0.1 mm/yr. Global salinity measurements, also made by Argo, provide an important constraint on the oceanic freshwater budget, one component of which is the melting of continental ice that causes eustatic sea level rise. Data assimilation techniques offer the promise of more robust estimates of ocean thermal expansion but to date there are considereable poorly explored differences between the various approaches. GRACE gravity data can be used to estimate changes in mass of the ocean. GRACE, satellite altimetry and steric changes agree well for the annual cycle but there is significant disagreement on trends over the 3 year period from 2002 to 2006. Coupled ocean-atmosphere models are the central tool for computing future ocean thermal expansion. Many of these models no longer rely on flux corrections. However, different models produce significantly different amounts of thermal expansion. Some aspects of the regional distribution of sea level rise projected by these models are similar. The only high resolution coupled ocean-atmosphere model available produces similar results to the medium resolution version of the same model, but the high resolution model has finer scale features and also has an increase in eddy sea-level variability.

Was this content written or created while at USF?


Citation / Publisher Attribution

2006 UNESCO, presented at World Meterological Organization on 6-9 June in Paris, France.