The 6-year [2005-2011] trend is calculated using a weighted least square fit and accounts for 0.54±0.1 Wm-2 for the ocean surface, i.e. 0.38±0.1 Wm-2 for the Earths surface. However, observed changes of GOHC have significant interannual to decadal variability associated with them, and this could mean that this variability may prevent detection of long-term trends. When time series of oceanic parameters are considered, linear trends are often computed to quantify the observed long-term changes. However, this does not imply that the original signal is best represented by a linear increase in time. To analyze the temporal evolution of GOHC time series is hence an important target to quantify significant interannual to decadal variability to in turn deliver reliable interpretation of global ocean changes.
...For an accurate estimation of GOHC, it is important to include its contributions from the deep ocean which are significant (Purkey and Johnson, 2010; Kouketsu et al., 2011), even if not dominant, and expected to grow with time as the abyssal ocean shifts (Wunsch et al., 2007). Analyses of observational surveys have shown significant deep ocean warming (Fahrbach et al. 2004; Johnson et al., 2007, 2008a, 2008b; Meredith et al., 2008; Böning et al., 2008; Leuliette and Miller, 2009; Jacobs and Giulivi, 2010; Masuda et al., 2010; Sutton and Roemmich, 2011), which are accompanied by changes of the general circulation system (Gille, 2008; Song and Colberg, 2011). In particular, Purkey and Johnson (2010) first quantify observed deep ocean temperature trends between the 1990s and 2000s. Warming in large areas of the global ocean accounts for a statistically significant fraction of the present global energy and sea level budgets, i.e. up to one third of those in the upper layer (Kouketsu et al., 2011; Song and Colberg, 2011).