Now that climate science has convinced people that global warming is due to humans and that we need to act now if we are to prevent significant implications for the future, it's time to work out what to do. But that's trickier to get to grips with than the climate science itself as it involves a wide array of academic disciplines, including both physical and social sciences.

With that in mind, a recent conference in Exeter, UK, brought together nearly 200 researchers from the two spheres. Jointly organized by the UK's Met Office and the University of Exeter, the "Climate Change Impacts and Adaptation: Dangerous rates of change" conference forged some useful links.

"We've done a lot of the physics of climate change and some of the biology," Peter Cox of the University of Exeter, one of the organizers of the conference, told environmentalresearchweb. "We now need to connect it to the social responses – how do people react to the problem and what sort of policies might be effective?"

Cox says there are two parts to the issue – climate mitigation, which concerns how to limit the damage from climate change by the end of the century, and climate adaptation – adapting to the inevitable change already built into the climate system because of the timelags involved.

"We know we're going to get another half degree of warming on top of the three-quarters of a degree which has already happened, at least, and that's if carbon emissions stopped overnight," he explained. "The adaptation agenda has been neglected, partly because it's harder – it's more cross-disciplinary – and partly because people felt that focusing on the fact that we had inevitable climate change to deal with anyway would provide an excuse not to worry about the mitigation problem."

Martin Parry of Imperial College London, who was co-chair of working group II for last year's IPCC assessment, agreed that adaptation has been neglected. "If we had £0.5 m to do some research on adaptation, that would be more than we've ever had before," he said. "We need to try and close the loop on this."

Short-term gain
But in a rare piece of good news on the climate science front, Cox says that although we tend to look on the century scale at the moment, it's actually relatively easy to predict the rates of change over the next few decades. Serendipitously that's precisely the information needed for making decisions about climate adaptation measures. Indeed, it's currently possible to predict the rate of change over the next few decades with much more accuracy than it is to predict the long-term equilibrium change.

"If you are going to adapt you need to know the impacts and they're a function of how fast you do things – it's not just about the magnitude [of climate change], it's the rate," said Cox. "If you did a 3 degree warming slowly enough, ecosystems and human systems could adapt, but if you did it quickly, even it it's just half a degree, it would be very difficult to adapt to."

So, besides decadal predictions of the rates of change, what else is needed to plan for adaptation and mitigation? Cox says that better regional predictions will help, along with an assessment of risk to help policymakers make decisions, and a way of connecting to the decision-making process.

"Global average surface temperature is only relevant for sea level," said Brian Hoskins of Reading University and Imperial College London. "It's becoming just an index for how far along the road we are but doesn't show you what climate will be like regionally."

Hoskins spoke about his role as lone climate scientist on the UK climate change committee, where he joins two economists, an engineer, an energy specialist, an ecologist and a businessman. This body will set three five-year carbon dioxide budgets, as well as recommending an emissions target for 2050, under the auspices of the climate change bill the UK is likely to introduce on December 1st. "2050 is not the end of the world - at least we hope it isn't – so we should be thinking further beyond," said Hoskins.

Speaking off the record, a number of scientists appeared concerned that the levels of the target will be set too low – taking into account what's achievable relatively easily, say 50% cuts by 2050, rather than the cuts of 80% or so that climate science indicates we need to be relatively certain of remaining below the 2° barrier and avoiding dangerous levels of climate change. The UK will be the first country to introduce such binding emissions targets so many feel it's important that the target levels have a sound scientific basis and can lead the way for the rest of the world.

Great words
Speakers at the conference appeared unusually literate, at least on the science-fiction front, with at least two presenters quoting Douglas Adams and another name-checking Isaac Asimov. According to Richard Betts of the Met Office, Adams said that "trying to predict the future is a mug's game…but it's a game we all have to play because the world is changing so fast." Although Adams was talking about technology, Betts believes his views still hold true for climate change.

Betts detailed how the three communities of scenarios experts, climate modellers, and impacts specialists don't communicate well enough or ask each other for the data they require.

"The climate modellers don't really care about scenarios, and the impacts guys take the output and don't care about the details of the climate model," he said. But in fact impacts researchers could benefit by requesting the information they actually need from climate models rather than just making do with the outputs climate modellers provide as standard.

"Translating emissions to concentrations is an overlooked source of uncertainty," added Betts, who says that the uncertainty due to feedback effects in the resulting concentrations is similar to the range of emissions scenarios. For example, researchers will often say that their climate model used concentrations from the A1B emissions scenario. Betts' answer to this is "Yes, but which A1B concentrations?" The problem arises mainly because of the uncertainty in carbon sinks, which in turn depend on the climate itself.

Betts also spoke about the problems of dealing with climate model bias, how local-scale adaptation planning needs to include local drivers such as urban heat island effects and land use change, how near-term adaptation plans must consider natural variability, which needs new model techniques such as the recent work by Smith et al, and interactions between impacts, such as feedback between the effects of climate change on crops and the water cycle.

"Our mugs' game is also a team game," he stressed. "We need to work together to take all these feedbacks and interactions into account to get a better picture for both mitigation and adaptation."