LEAVING FOSSIL FUELS IN THE GROUND: HOW MUCH, WHERE AND OVER WHAT TIME FRAME?

by Dr Christophe McGlade and Prof Paul Ekins, UCL Institute for Sustainable Resources

We have known for some time that to limit global warming, some of the world’s fossil fuel reserves are going to have to stay in the ground. All of the carbon dioxide that would result from burning current fossil fuel reserves is around three times the amount that gives us a decent chance of staying below the 2^oC threshold (the temperature rise accepted by the international community as associated with the possible onset of dangerous climate change).

Until recently people therefore frequently reported that two thirds of fossil fuels globally were ‘unburnable’. While it’s possible to get lots of interesting insights from this simple finding, this treats all the fossil fuels as the same. This is too simplistic.

The CO₂ emissions that result from burning a tonne of coal are much higher than getting the same amount of energy from gas. Oil also has a much higher energy density than coal or gas, which is why it’s so useful for transporting ourselves, and the things we produce and consume, around our own country and the world.

Our recent Nature paper therefore aimed to differentiate between the fossil fuels we can afford to burn while remaining within the 2°C temperature limit.

We found that a third of oil reserves, half of gas reserves and over 80% of current coal reserves globally should remain in the ground and not be used before 2050 if global warming is to stay below 2°C.

We also disaggregated fossil fuel reserves geographically and showed that the reserves that can and cannot be used are not spread around the world evenly. For example, 260 thousand million barrels oil reserves are not used before 2050 in the Middle East.

This is a huge volume of oil, roughly equivalent to all of the oil reserves held by Saudi Arabia. Meanwhile, the United States, Russia, and Australia all use less than 5% of their coal reserves meaning that over 95% of their coal reserves should remain unburnt, again by 2050. After 2050 the scope for carbon emissions, while staying within the 2^oC limit, is much reduced,

We also looked at the development of some of the more ‘unconventional’ sources of fossil fuel. None of the oil and gas resources in the Arctic (which are estimated to be very large) were produced and there is only very low levels of production of the oil sands in Canada. All Arctic resources and 85% of Canadian tar sands reserves should therefore be classified as unburnable.

Given the space constraints for an academic paper it wasn’t possible to put in as much information as we would have liked. One element that we have been asked for a number of times is the differences in production for each of the fossil fuels between our 2^oC and one that imposed no carbon constraint.

This scenario, in which the world abandons any efforts to reduce emissions, results in a level of global warming that reaches 4^oC by the end of the 21^st century (we avoid using the phrase ‘business-as-usual’ for such a scenario -see here for a discussion why). The world would then be hotter than has ever been experienced by humans, with unpredictable, but almost certainly very negative consequences for human societies, in terms of droughts, floods and sea-level rise, to name only three of the most likely impacts.

There is a real difference in coal and oil production between this 4^oC scenario and the one which gives us a reasonable chance of staying below 2^oC – around 220 billion barrels oil and 290 billion tonnes coal are produced in the 4^oC scenario. For coal, this represents a 250% increase in the amount of reserves produced.

However, there is much less difference in gas production between the two scenarios. Gas helps to offset some of the reduction in coal to the extent that up to 2035 gas consumption in the 2^oC scenario is actually higher than in the 4^oC scenario. This is gas acting as a ‘transition’ fuel to a low-carbon economy. However, for gas to a bridge in this way, there are a number of criteria that need to be satisfied – we have written about this in more detail here.

We also received some critical but constructive feedback on the paper and assumptions we made. We are in the process of writing more detailed responses to these and will post these here as soon as possible.