by Richard Allan.

Glancing at a graph showing the changes in global annual mean temperature (Fig. 1) you can't fail to notice that the rate of global warming appears to flatten off in the most recent decade. There is a good discussion in skepticalscience.com about the role of natural variability which can indeed explain decades of cooling despite continued build up of energy in the climate system in control climate simulations and anthropogenic warming scenarios. An alternative explanation for the warming hiatus is a combination of negative radiative forcings (e.g. volcanic eruptions, stratospheric water vapour changes or Asian aerosol emissions) that may have temporarily offset, to some extent, the net positive radiative forcing since 2000.

In a recent paper in Nature Geosciences we have combined CERES satellite measurements of changes in net radiative energy at the top of Earth's atmosphere with Argo ocean observations of heat content (down to a depth of 1800m) to assess the energy input to our climate in the past decade. Contrary to previous reports of missing energy in the climate system, but consistent with a recent assessment by Hansen et al., we determine a positive net radiative surplus of 0.5 Watts per metre squared over the last decade (equivalent to 250 billion 1 kilowatt electric heaters distributed over the globe, assuming that all the energy is retained by Earth's climate system). This indicates that while ocean surface warming has stalled in the recent decade, the climate is continuing to accumulate energy at a rate consistent with anthropogenic greenhouse gas increases (warming has been hidden beneath the ocean surface).

The mechanism behind changing ocean heat uptake in the last decade is unclear. Developing a capability for tracking energy flows through the climate system is highly important in understanding the relationship between climate forcing, feedback and response, decadal climate variability and the trajectory of climate change over the coming decades; this remains a considerable challenge for the scientific community.

Figure 1: Global annual mean surface temperature anomaly (relative to the 1951-1980 mean) from the HadCRUT3 and GISSTEMP blended land-surface air temperature and sea surface temperature observations.