Abhishek Savita, Catia M. Domingues, Tim Boyer, Simon A. Good, Viktor Gouretski, Masayoshi Ishii, Gregory C. Johnson, John M. Lyman, Josh K. Willis, Didier Monselesan, Simon J. Marsland, John Antonov, Susan E. Wijffels, Rebecca Cowley, Peter Dobrohotoff, William Hobbs, John A. Church
The Earth system is accumulating energy due to human-induced activities. More than 90% of this energy has been stored in the ocean as heat since 1970, with ~64% of that in the upper 700 m. Ocean heat uptake delays surface warming but causes sea level to rise. Important differences, however, exist among upper-ocean heat content (OHC) estimates from various international research groups at global, basin and regional scales. These differences may arise due to choices in mapping methods, bias-corrected data used, periods analysed, etc. In this internationally-coordinated comparison, we evaluate spread in upper OHC estimates arising from choices in instrumental bias corrections and mapping methods, in addition to the effect of using a common ocean mask. The same dataset was mapped by six research groups for 1970–2008, with six instrumental bias corrections applied to observations from eXpendable BathyThermographs (XBTs). We find that use of a common ocean mask may impact estimation of global OHC by 2–13%. Uncertainty due to mapping method dominates over XBT bias corrections at a global scale but is largest in the mostly sparsely observed Indian Ocean basin (68 MJ m-2, 1 MJ=106J), and also in eddy energetic regions of all basins. Uncertainty due to XBT bias correction is largest in the Pacific Ocean (40 MJ m-2) within 30°S–30°N. In both mapping and XBT cases, spread is higher since the 1990s. Differences in spatial trends among mapping methods are seen for the well-observed Northwest Atlantic and the poorly-observed Southern Ocean. Although our results cannot identify the best mapping or bias correction schemes, they identify where and when greater uncertainties exist, and so where further refinements may yield the largest improvements. Finally, our results highlight the need for a future international coordination to evaluate performance of existing mapping methods.