1. Sensitivity of Global Upper Ocean Heat Content Estimates to Mapping Methods, XBT Bias Corrections, and Baseline Climatologies Ocean warming accounts for the majority of Earth’s current energy imbalance. Historic Ocean Heat Content (OHC) changes are important to understanding our changing climate. Calculations of OHC anomalies (OHCA) from in situ measurements provide estimates of these changes. Uncertainties in OHCA estimates arise from temporal and spatial inhomogeneity of subsurface ocean temperature measurements, instrument bias corrections, and the definitions of a mean ocean climatology from which anomalies are calculated. To quantify the uncertainties and biases these different factors contribute for different OHCA estimation methods, and the uncertainty in these different estimation methods, the same data set with the same quality control is used by seven groups calculating eight OHCA estimates for 0-700m depth and comparisons made. T. Boyer 1, C. M. Domingues 2,3,4, S. Good 5, G. C. Johnson 6, J. M. Lyman 6,7, M. Ishii 8, V. Gouretski 9, J. K. Willis 10, J. Antonov 11, S. Wijffels 12, J. A. Church 13, R. Cowley 12, N. Bindoff 2,3,4,12 1 NOAA/United States National Centers for Environmental Information, Silver Spring, Maryland, USA, 2 Institute for Marine and Antarctic Studies (IMAS), University of Tasmania (UTAS), Hobart, Tasmania, Australia, 3 Antarctic Climate and Ecosystems Cooperative Research Institute, Hobart, Tasmania, Australia, 4 Australian Research Council's Centre of Excellence for Climate System Science, Hobart, Tasmania, Australia, 5 Met Office, Exeter, Devon, United Kingdom, 6 NOAA/Pacific Marine Environmental Laboratory, Seattle, Washington, USA, 7 Joint Institute for Marine and Atmospheric Research, University of Hawai’i at Manoa, Honolulu, Hawai’i, USA, 8 Meteorological Research Institute, Japan Meteorological Agency, Nagamine, Tsukuba, Ibaraki, Japan, 9 University of Hamburg, Center for Earth System Research and Sustainability, CliSAP, Integrated Climate Data Center, 10 Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California, USA, 11 University Corporation for Atmospheric Research, Boulder, Colorado, USA, 12 Commonwealth Scientific and Industrial Research Organization, Hobart, Tasmania. Corresponding author: tim.boyer@noaa.gov NameMethod description DOMDomingues et al. (2008) LEVLevitus et al. (2009) ISHIshii et al. (2006) WILWillis et al. ( 2004) PMEL_MLyman and Johnson (2008) PMEL_RLyman and Johnson (2008) ENIngelby and Huddleson (2007) GOUGouretski et al. (2012) Methods for calculating OHC anomaly (OHCA) from in situ data in this study Each of the methods in this study calculates ocean heat content from in situ ocean profile data (Argo profiling floats,, expendable bathythermograph – XBT, Conductivity-Temperature-Depth (CTD), and bottle casts) by subtracting temperature at each specific depth from a monthly mean baseline, then employing a distinct mapping method to estimate full geographic coverage from actual in situ coverage (horizontally and vertically), then OHCA is calculated from temperature anomaly and integrated over the global ocean.. The Experiment : Yearly OHCA values can give a good measure of the Earths Energy Imbalance, but there are many sources of uncertainty in OHCA calculations. This study quantifies some of the uncertainties in OHCA estimation. The same data and quality control(ENv32a, http://www.metoffice.gov.uk/hadobs/en3/ plus Argo data (Barker et al., 2008) were used throughout to eliminate data/quality control choice differences in order to concentrate on uncertainty due to 1)Mapping method – differences due to technique for projecting irregular data to full geographic coverage, smoothing, etc. 2) XBT bias corrections – the main ocean observing system 1970-2001, systematic biases with different published methods for correcting. 3) baseline mean climatology – reference for anomalies; geographic, time period differences in chosen baselines Chosen study years are 1970-2008 and 1993-2008 (1970 near start of widespread use of XBT, 2008 is two years into the near-full global coverage of Argo floats, 1993 is the start of the altimeter era) A B NameDescriptionCorrection Type W08Wijffels et al. (2008)depth L09Levitus et al. (2009)temperature I09Ishii and Kimoto (2009)depth G11Good (2011)depth GO12Gouretski (2012)depth + temperature C13Cowley et al. (2013)depth + temperature C XBT correction methods in this study There are > 12 published corrections. The methods chosen for this study include representatives from each correction type (depth correction only, temperature correction only, depth and temperature corrections. NameDescriptionData UsedYears C1_HAlory et al (2007)Bottle, CTD, Argoall C2_HAlory et al. (2007)Same as C1_H plus XBTall C3_MLocarnini et al. (2013)All extant2005-2012 D Baseline mean choices in this study H=historical, M=modern. H climatologies have linear trend removed at each geographic grid to eliminate geographic bias as detailed in Wijffels et al. (2008). Levitus, S., J.I. Antonov, T.P. Boyer, R.A. Locarnini, H.E. Garcia, and A.V. Mishonov, 2009: Global Ocean Heat Content 1955-2008 in light of recently revealed instrumentation problems. Geophys. Res. Lett., 36, L07608, doi:10.1029/2008GL037155. Locarnini, R. A., A. V. Mishonov, J. I. Antonov, T. P. Boyer, H. E. Garcia, O. K. Baranova, M. M. Zweng, C. R. Paver, J. R. Reagan, D. R. Johnson, M. Hamilton, and D. Seidov, 2013: World Ocean Atlas 2013, Volume 1: Temperature. S. Levitus, Ed., A. Mishonov Technical Ed.; NOAA Atlas NESDIS 73, Silver Spring, MD, 40 pp. Lyman, J. M. and G. C. Johnson, 2008: Estimating global upper ocean heat content despite irregular sampling. J. Climate, 21, 5629-5641, doi:10.1175/2008JCLI2259.1. Lyman, J. M. and G. C. Johnson, 2014: Estimating global ocean heat content changes in the upper 1800 m since 1950 and the influence of climatology choice. J. Climate, 27, 1946–1958, doi: 10.1175/JCLI-D-12-00752.1. Rhein, M., S.R. Rintoul, S. Aoki, E. Campos, D. Chambers, R.A. Feely, S. Gulev, G.C. Johnson, S.A. Josey, A. Kostianoy, C. Mauritzen, D. Roemmich, L.D. Talley and F. Wang, 2013: Observations: Ocean. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp. 255–316, doi:10.1017/CBO9781107415324.010. Wijffels, S. E., J. Willis, C. M. Domingues, P. Barker, N. J. White, A. Gronell, K. Ridgway, J. A. Church, 2008: Changing Expendable Bathythermograph Fall Rates and Their Impact on Estimates of Thermosteric Sea Level Rise. J. Climate, 21, 56575672. doi: http://dx.doi.org/10.1175/2008JCLI2290.1 Willis, J. K., D. Roemmich and B. Cornuelle, 2004: Interannual variability in upper ocean heat content, temperature, and thermosteric expansion on global scales, J. Geophys. Res., 109, C12036, doi: 10.1029/2003JC002260. References: Alory, G., Wijffels, S. and Meyers, G. M., 2007: Observed temperature trends in the Indian Ocean over 1960-1999 and associated mechanisms. Geophys. Res. Lett., 34, L02606, 10.1029/2006GL028044 Barker, P. M., J. R. Dunn, C. M. Domingues, and S. E. Wijffels, 2011: Pressure Sensor Drifts in Argo and Their Impacts. J. Atmos. Oceanic Technol., 28, 1036–1049. doi:http://dx.doi.org/10.1175/2011JTECHO831.1 Cowley, R., S. Wijffels, L. Cheng, T. Boyer, S. Kizu, 2013: Biases in Expendable BathyThermograph data: a new view based on historical side-by-side comparisons. J. Atmos. Ocean Tech., 30, 1195-1225, doi: http://dx.doi.org/10.1175/JTECH-D-12-00127.1 Domingues, C. M., J. A. Church, N. J. White, P. J. Gleckler, S. E. Wijjfels, P. M. Barker, J. R. Dunn, 2008: Improved estimates of upper-ocean warming and multi-decadal sea-level rise, Nature, 453, 1090-1093. Good, S. A., 2011: Depth Biases in XBT Data Diagnosed Using Bathymetry Data. J. Atmos. Oceanic Technol., 28, 287–300. doi: http://dx.doi.org/10.1175/2010JTECHO773.1 Gouretski, V., 2012: Using GEBCO digital bathymetry to infer depth biases in the XBT data, Deep Sea Research-I, 62, 40-52 Gouretski, V., J. Kennedy, T. Boyer and A. Köhl, 2012: Consistent near-surface ocean warming since 1900 in two largerly independent observing networks. Geophys. Res. Lett., 34, doi:10.1029/2012GL052975 Ingelby, B. and M. Huddleston, 2007: Quality control of ocean temperature and salinity profiles – historical and real-time data, J. Marine Systems, 65, 158-175 Ishii, M., M. Kimoto, K. Sakamoto, and S.I. Iwasaki, 2006: Steric sea level changes estimated from historical ocean subsurface temperature and salinity analysis, J. Oceanogr., 62, 155-170. Ishii, M. and M. Kimoto, 2009: Reevaluation of Historical Ocean Heat Content Variations with an XBT depth bias Correction. J. Oceanogr. 65, 287299, doi:10.1007/s10872-009-0027-7 NameMapping MethodXBT bias correctionBaseline Climatology DOM16.5 (17.1) ZJ16.7 (18.9) ZJ11.8 (7.1) ZJ LEV16.5 (17.1) ZJ11.1 (10.4) ZJ3.5 (2.7) ZJ PMEL_M16.5 (17.1) ZJ7.4 (11.5) ZJ14.5 (4.7) ZJ PMEL_R16.5 (17.1) ZJ12.1 (15.3) ZJ6.6 (4.9) ZJ ISH16.5 (17.1) ZJ9.5 (10.5) ZJ8.6 (3.9) ZJ EN16.5 (17.1) ZJ12.8 (15.1) ZJ11.8 (5.6) ZJ GOU16.5 (17.1) ZJ11.3 (15.2) ZJ11.5 (9.8) ZJ WIL* (17.1) ZJ* (12.8) ZJ* (3.1) ZJ Average16.5 (17.1) ZJ11.6 (12.2) ZJ9.8 (5.2) ZJ Since each baseline has a different mean OHC, C2_H and C3_M were referenced to C1_H value for 2008 for time series comparison. 1980 2008 E 1 2 3 456 78 9 Geographic Distribution: Earlier years such as 1980 (E1) have relatively sparse data distribution compared to later (Argo) years (E2). For E1,E2, red=data no deeper than 300m, black= data to 700m. OHCA distribution for 2008 )E3-9) show many of the same large scale features (cool La Nina conditions in the Eastern Pacific, warming north Atlantic), but also different OHCA distributions and magnitudes. (Indian subAntarctic front, mid south Atlantic). For E3-9, red=positive OHCA, blue=negative OHCA, contours at 5 x 10 18 joule intervals. PMEL_R has similar OHCA distribution to PMEL_M (E6) with mean 2008 OHCA in data sparse areas rather than zero OHCA in same areas form PMEL_M. (Shaded areas in E3 are not included in OHCA integral for DOM) F 1 2 3 Time Series Comparison Examples : Time series 1970-2008 of OHCA using W08 XBT corrections, C1_H baseline and each mapping method (F1) show similar OHCA increase, but year to year differences. DOM mapping method, C1_H baseline with different XBT corrections (F2) show difference due to choice of XBT correction. DOM has the largest variance due to XBT correction of all methods, ISH method, W08 XBT correction, three baseline means (F3) shows larger difference between historical (H) and modern (M) baseline cases for earlier years. This is due infill method and the warmer modern baseline (see Lyman and Johnson, 2014). 1 zetajoule (ZJ) = 10 21 J G 1 2 3 Variance/Uncertainty: Yearly standard deviation due to mapping method (G1) is similar for all XBT corrections, except late 1990s, early 2000s. Yearly standard deviation from different XBT corrections (G2) is generally larger for each method in later years while yearly standard deviation due to different baseline mean is generally larger in earlier years (see F3, explanation above). Straight lines on each graph are mean of yearly standard deviation. [Green curve in G1 is OHCA calculated using LEV method over areas where OHCA is not calculated in DOM (shading in E3).] H Summary : Table H summarizes the uncertainty (mean standard deviation) in OHCA estimates due to mapping method, XBT bias correction choice, and baseline mean. First values are for years 1970-2008, values in parentheses are for 1993-2008. Mapping method is a significant source of uncertainty in OHCA calculations inherent in the difficulty in using widely varying time/space data distributions. XBT bias correction also contributes uncertainty, as does baseline mean climatology. If the XBT community comes to a consensus on the most effective XBT correction, this source of uncertainty can be greatly reduced. The choice of baseline climatology can be important based on mapping method, with variants on the zero infill method needing a baseline mean which will not underestimate OHCA in earlier data sparse years. More work needs to be done to tighten uncertainties and explore regional effects, vertical resolution, data choice and quality control as additional sources of uncertainty for OHCA calculations. Summary of Uncertainties (mean yearly standard deviation ) for 1970-2008 (1993-2008)