A third of oil reserves, half of gas reserves and over 80 per cent of current global coal reserves should not be used before 2050 if global warming is to stay below 2°C, according to a new model from UCL’s Institute for Sustainable Resources.

‘The constraint we applied was to keep temperatures within the two degrees that world policy makers say they would like to achieve,’ says Paul Ekins, director at the Institute. ‘This in turn restricts emissions and resource extraction.’ Coal is the most carbon intensive fossil fuel and so the model predicts the greatest restrictions on its extraction.

Restrictions in the model correspond to the IPCC’s carbon budget for 2050, says Ekins. Country-by-country impacts would depend on the resources in a region and market costs. According to Ekins, the countries with the largest unused coal reserves are the USA (with 95 per cent unused), the former Soviet Union (97 per cent unused) and China and India (both with around 77 per cent unused).

‘We compared several scenarios other than two-degree scenario. One had no carbon constraints, and it projected that a very great deal more of the fossil fuel base would be used,’ says Ekins. Existing policies and pledges would have to change to meet the two-degree target and preserve fossil fuel reserves.

There would be a significant regional economic impact if these resources stayed in ground. ‘Our model suggests, for example, that only a small amount of the Canadian tar sands would be used in the two-degree scenario because these are relatively expensive and carbon intensive. This would obviously have a large impact on the economy of that area,’ says Ekins. ‘If financial markets decided that there was going to be a low carbon future and only a small proportion of resources would be extracted, it would cause a decline in the resource value and the companies that own them.’ He adds that if oil companies invested in low carbon energy resources over 20 or 30 years this process could happen quite smoothly.

The model seeks the cheapest natural resources in each scenario. ‘It has thousands of different technologies all with different costs with assumptions based on the best expert opinion we can finds,’ says Ekins. The model starts with present costs, and then moves forward in five-year jumps.

New models will look at technological advance in low-carbon technologies, especially predicting when tech costs will fall. ‘We also have to keep up with new oil discoveries,’ Ekins says. ‘Carbon capture plays a minor role in the model, which surprises some people because it is thought of as a get-out-of-jail-free card. As it has not been demonstrated commercially we do not build it into the model until 2025.’ Carbon capture will also be expensive when it arrives, and the model chooses cheaper resources over these technologies.