Paris

Researchers are unsure whether this summer's European heatwave — which may have caused 3,000 deaths in France alone, according to government estimates — can be attributed to global warming. But there's one thing they do agree on: that it demonstrates the need for better regional climate models, to help climatologists get to grips with the processes that drive extreme weather.

Existing models of the global climate system simulate observed, long-term changes in mean temperature reasonably well, researchers say. But models of regions such as Europe are less well developed, says Simon Brown, who studies climate extremes at the UK Met Office's Hadley Centre for Climate Prediction and Research in Berkshire. Researchers blame this weakness on a lack of resources to build the models, and a lack of understanding of the processes that underlie the weather.

Global climate models are made up of data points in a three-dimensional grid from the depths of the oceans to the upper reaches of the atmosphere. The ocean, atmosphere, land surface and ice components are linked by the fluxes between them.

But even the most powerful global models are not sufficiently fine-grained to yield useful signals for regional prediction. The global models' grid points are about 2.5 degrees of latitude apart; for regional simulations, climate researchers would like to zoom down to 0.5 degrees, at least. So researchers who model local effects have to select regions of interest for higher-resolution grids, and nest these within the global models.

This is suboptimal, points out Howard Cattle, who runs the climate-change programme at the Geneva-based World Climate Research Programme. “If you force a regional model with boundary conditions from a coarse resolution model, you aren't giving it information at the boundaries that is consistent with its internal dynamics,” he says. “It's a 'rubbish in, rubbish out' situation.”

An even bigger problem is a lack of understanding of the atmospheric processes that generate extreme climate events. The prolonged presence of a high-pressure anticyclone, which has characterized the current heatwave, is a common feature of European latitudes, says Hervé Le Treut, head of the lab of Dynamic Meteorology at the Ecole Normale Supérieure in Paris. But little is known about what controls the anticyclone's intensity, or what encourages it to hang around.

Europe's climate is particularly unpredictable, the researchers say. The El Niño effect can have a decisive effect on short-term climate trends in the United States, for example, but the influence of the Gulf Stream, say, on Europe's climate is tougher to untangle.

The fact that weather events create more 'noise' at a regional level than at a global level “makes it very hard to detect a climate-change signal” from regional data, says Ulrich Cubasch, a climate modeller at the Free University of Berlin, Germany.

The difficulty is compounded by the relatively meagre effort made to refine regional models. “Simulations by regional climate models have not been validated and analysed anywhere near as much as the global climate model simulations,” says Tim Osborn, a researcher at the University of East Anglia's Climatic Research Unit in Norwich, UK.

In these circumstances, it is difficult for anyone to figure out how far this summer's heatwave in Europe reflects the wider phenomenon of global warming. But data from the heatwave will at least help researchers to test and refine their models. “We should be looking at current extremes to test the models' ability,” Osborn says.