An ensemble of ocean reanalyses for 1815-2013 with sparse observational input Academic Article uri icon


  • 2016. American Geophysical Union. All Rights Reserved. This paper describes a new eight-member ensemble of ocean reanalyses spanning nearly 200 years from 1815 to 2013 generated using the Simple Ocean Data Assimilation system with sparse observational input (SODAsi) to explore long-term changes in the oceans. The eight ensemble members assimilate surface temperature observations and use surface boundary conditions from an atmospheric reanalysis that is loosely coupled to the ocean reanalysis. Both surface and subsurface quantities, such as dynamic height and heat content, show a broad spectrum of variability. Surface temperature trends from 1815 to 2013 are positive in most regions, with some important exceptions; the central Tropical Pacific, around Antarctica, and in the Gulf Stream and Kuroshio extension regions all show cooling trends. A near-global average shows warming of about 0.8C over the full period, with most of the warming occurring after 1920. There is pronounced multidecadal variability in both the midlatitude and tropical oceans. In the North Atlantic Ocean, temperature variability is highly correlated with the meridional overturning stream function, with the largest correlation occurring when the stream function is advanced by 9 years. Trends of upper ocean heat content and dynamic height from the 1950s onward compare well with previously published values. Globally averaged heat content of the upper 700 m shows a nearly linear rise after the 1920s, requiring a net downward surface heat flux increase of 0.47 W m2 into the ocean. This is close to published estimates of the increased flux required to explain the heat content increase from 1971 to 2010.

published proceedings


altmetric score

  • 9.544

author list (cited authors)

  • Giese, B. S., Seidel, H. F., Compo, G. P., & Sardeshmukh, P. D.

citation count

  • 84

complete list of authors

  • Giese, Benjamin S||Seidel, Howard F||Compo, Gilbert P||Sardeshmukh, Prashant D

publication date

  • September 2016