The most reliable source of information for changes in the global mean net air–sea heat flux comes from the constraints provided by analyses of changes in ocean heat storage. The estimate of increase in global ocean heat content for 1971–2010 quantified in Box 3.1 corresponds to an increase in mean net heat flux from the atmosphere to the ocean of 0.55 W m–2. In contrast, closure of the global ocean mean net surface heat flux budget to within 20 W m–2 from observation based surface flux data sets has still not been reliably achieved (e.g., Trenberth et al., 2009). The increase in mean net air–sea heat flux is thus small compared to the uncertainties of the global mean. Large and Yeager (2012) examined global ocean average net heat flux variability using the CORE data set over 1984–2006 and concluded that natural variability, rather than long-term climate change, dominates heat flux changes over this relatively short, recent period. Since AR4, some studies have shown consistency in regional net heat flux variability at sub-basin scale since the 1980s, notably in the Tropical Indian Ocean (Yu et al., 2007) and North Pacific (Kawai et al., 2008). However, detection of a change in air–sea fluxes responsible for the long-term ocean warming remains beyond the ability of currently available surface flux data sets.