Two leading climate scientists have urged their colleagues to find the growing amount of "missing energy" that seems to be eluding climate sensors.

In a commentary in today's issue of Science, Kevin Trenberth and John Fasullo of the National Center for Atmospheric Research in Colorado, US, identify a large and growing amount of solar energy that appears to have been absorbed by the Earth – but has yet to turn up in terrestrial measurements.

Since 2001 scientists have used satellites to compare the amount of solar energy being absorbed by the Earth to the amount of infrared energy escaping from our planet. And just as predicted by the theory of manmade climate change, the amount of energy retained by the Earth has increased along with greenhouse-gas concentrations.

Losing energy since 2004

At first this extra energy seems to have boosted temperatures down here on Earth. Then something unexplained happened in about 2004 – and since then terrestrial measurements suggest that the planet is losing energy.

So are the satellites wrong? While Trenberth and Fasullo say that the satellites are not good enough to give accurate measurements of the net energy itself, they claim that the instruments are "sufficiently stable" to track changes in net energy, which are the important quantity.

Trenberth told that the discrepancy probably lies in the environment's largest heat reservoir. "I would say that the missing heat is mainly in the ocean," he argues.

Much of our understanding of how the oceans absorb energy comes from over 3000 "Argo floats" that gather temperature data at depths of up to 2000 m. However, Trenberth says he thinks that "oceanographers are fairly new at processing this kind of data and are still learning how to do it right". He also points out of that some of the floats deployed in the Atlantic have been problematic.

Lurking deep in the ocean

Despite the challenges involved in analysing the Argo data, Trenberth points to a recent study by Karina von Schuckmann and colleagues at CNRS in Plouzané, France that looks at Argo data gathered in 2003–2008. Unlike most calculations of energy changes in the oceans – which consider temperatures down to 700 m – the CNRS team looked all the way down to 2000 m. At these greater depths they appear to have found a significant chunk of the missing heat.

Trenberth believes that it is crucial to understand when this energy will return to the upper ocean, where it would have a significant effect on climate. Scientists already know the Southern Oscillation involves the absorption of solar energy by the Pacific Ocean during "La Niña" years and its release into the atmosphere during "El Niño" years – leading to significant changes in weather patterns in the Americas.

An El Niño began in 2009 and looks set to continue in 2010. Trenberth believes that it might result in much of the missing energy resurfacing – but adds that current data gathering and analysis techniques mean that it could be a year or two before we know. "One can argue that we should develop a system to do this in closer to real time as part of the new climate services," he said.