Publication

For government planners and other decision-makers, model simulations of future climate change pose a dilemma. Climate projections are consistent and unambiguous about coming stresses and the need to take adaptive action — but often do not provide enough detail to allow them to work out what action is needed.

Global climate models are the basis of much of the current scientific understanding about how and why the climate is changing. And they send very clear messages that the planet will continue to warm, sea levels will continue to rise and that rainfall and other climate parameters will change significantly in many places.

In short, increasing climate disruption is on the way, and major adaptations will be needed in sectors from agriculture and economic development to public health and disaster response. But to plan these responses, decision-makers need more specific information. Climate models agree on global trends, but disagree on just how fast and intense the changes will be, and sometimes on the patterns of change — which areas will become wetter or dryer, for instance. Taken together, the models suggest differing levels of confidence about different climate variables. And currently there is no objective way to tell which models are more trustworthy.

Because these models cannot be shown to be wrong and thus their projections can be considered as plausible future scenarios, results from multiple models must be combined to understand the range of possible climate futures we should prepare for. At the same time, adaptation planners need detailed projections of the likely changes in local areas — at a higher resolution than global climate models provide. Similarly, efforts to assess the likely impacts of climate change on factors such as water availability, ecological balance and food security also require highresolution data. It is possible to ‘downscale’ global models to yield detailed local projections, but downscaling many global models and making the results available requires significant computing and human resources.

As part of the Adapting to Climate Change in China research programme, partners at the UK Met Office Hadley Centre have used a novel method to try to get the best of both worlds — multi-model climate projections at high resolution — using fewer resources. The new approach meets decision-makers’ need for specific local scenarios, at lower cost. This method was applied to characterise the range of plausible changes within China, and the same technique can be transferred to any other country and has been applied for Vietnam and in Africa.