Download

Executive summary

The Parties to the United Nations Framework Convention on Climate Change (UNFCCC) are contemplating a range of goals for limiting the rise in global average temperature that is resulting from rising concentrations of greenhouse gases in the atmosphere.

Whatever temperature goal is eventually agreed by the Parties, its achievement will depend on the extent to which global emissions of greenhouse gases can be limited and reduced.

We have carried out a study that explores a variety of potential paths for annual global emissions of greenhouse gases which would offer a reasonable chance of limiting a rise in global average temperature to no more than 1.5°C above its preindustrial level.

As a first test, we explored simplified paths in which global emissions would be reduced to zero in 2021 and would remain at that level afterwards. This first test indicated that it would be very challenging to avoid a rise in global average temperature of more than 1.5°C. Even if global emissions fall from 47 billion tonnes of carbon-dioxide-equivalent in 2010 to 40 billion tonnes in 2020, and are then reduced to zero immediately afterwards, we estimate that there would be a maximum probability of less than 50 per cent of avoiding global warming of more than 1.5°C above the pre-industrial level.

However, if global emissions fall to less than 48 billion tonnes of carbon-dioxide-equivalent in 2020 and then are reduced to zero immediately afterwards, there would be at least about a 90 per cent chance that the global temperature would overshoot but return to within 1.5°C of the pre-industrial level within 50 years.

Therefore, if the global average temperature exceeds a rise of 1.5°C, it is possible that, after overshooting, it could fall back to the temperature goal over several decades if global annual emissions are reduced rapidly enough.

These first tests were not intended to be realistic scenarios, as reducing emissions to zero in the next decade would have enormous economic cost, but provide an indication of whether a goal of avoiding a temperature rise of more than 1.5°C could be theoretically achieved.

We have also explored emissions paths corresponding to more plausible rates of decline in annual emissions after 2020, instead of assuming zero emissions. We have analysed two sets of emissions paths, each of which used different assumptions about the emissions baselines to 2020 and the amount of anthropogenic emissions of sulphate aerosols. In all of these cases, the probability of avoiding a rise of more than 1.5°C above the pre-industrial level, without overshooting, was much less than 50 per cent.

However, we have identified a number of emissions paths that offer a probability of 50 per cent of global average temperature being no more than 1.5°C above its preindustrial level in the long term, but they involve temporarily overshooting the temperature goal for up to 100 years.

We have identified four key characteristics of emissions paths that offer at least 50 per cent probability of global average temperature being no more than 1.5°C above its preindustrial level in the long term, with a temporary overshoot of no more than 100 years.

The first characteristic of the paths is that they involve early and strong reductions in global annual emissions. Our findings suggest annual global emissions must begin to fall within five years (by no later than 2015) and reach levels by 2020 of no more than 48 billion tonnes of carbon-dioxide-equivalent. A review of the literature indicates that such early and strong reductions in global annual emissions are likely to be feasible and economically desirable, if appropriate policy measures are put in place immediately to correct a range of market failures.

The second characteristic of the paths is that they require rapid reductions in annual global emissions after 2020. Our findings indicate that annual global emissions could need to fall at rates of at least around 3 per cent per year after 2020. Annual global emissions in 2020 of 44 billion tonnes or more of carbon dioxide-equivalent would require reduction rates of at least 4 per cent per year afterwards. The feasibility of such rapid rates of emissions reductions is an area of active debate. We conclude that such rates could be technically feasible in certain circumstances, but are likely to entail considerably higher global costs than the 2°C goal, which would be reflected in higher global energy costs and lower world consumption than would be the case otherwise.

The third characteristic of the paths is that they require low annual global emissions by 2100, with a floor close to zero emissions in the long term. Annual global emissions would need to fall to much less than 5 billion tonnes of carbon dioxide-equivalent by 2100. This may require the deployment of technologies (such as carbon capture and storage for biomass burning) that have ‘negative net emissions’ and could offset residual emissions in sectors such as agriculture where emissions reductions would be particularly difficult. It is not clear whether such technologies are feasible on the scale needed, nor whether they will be safe and reliable over long time horizons.

The final characteristic of the paths is that they are based on the assumption that it is possible for the global average temperature to exceed the goal (i.e. overshoot) and then return over the course of a few decades to no more than 1.5°C above the pre-industrial level. It is important to note that there are significant uncertainties associated with modelling overshoot scenarios. There is currently a scientific debate over how quickly atmospheric concentrations of greenhouse gases would decline following large reductions in emissions from human activities.

It is very important to recognise that overshooting any temperature goal would generate risks of triggering feedback accelerations, such as the enhanced release of carbon from the thawing of soils that are currently frozen, or causing large-scale and potentially dangerous impacts that could be difficult to reverse, such as a loss of species, inundation of some land areas, or extensive bleaching of corals. More research is needed into the likelihood of triggering feedbacks or irreversible impacts, such as large rises in sea level, during temporary overshooting of a 1.5°C goal.

Overall, our results do not rule out the achievement of a 1.5°C goal in the long term. Our findings suggest that, given historical trends in global annual emissions of greenhouse gases and, even with early and strong action to reduce emissions, the likelihood of avoiding global warming of more than 1.5°C above the preindustrial level is low. However, it may be possible to limit the rise to no more than 1.5°C above the pre-industrial level in the long term, if the global average temperature is allowed to overshoot the goal and fall over a period of several decades.

Given the current uncertainties, one approach for policymakers may be to take actions that will allow the option of switching at some later point to an emissions path that is consistent with a 1.5°C goal. Our analysis suggests that the range of global annual emissions of 40 to 48 billion tonnes of carbon-dioxide-equivalent in 2020, which would be consistent with a 2°C goal, might also be compatible with a 1.5°C goal, if it is assumed that emissions reductions could be quickly accelerated after 2020. Our findings suggest that aiming for the bottom end of this range for 2020 (i.e. taking strong action now) would reduce the risk of closing down the option of switching to a 1.5°C goal.

Nicola Ranger, Laila Gohar, Jason Lowe, Alex Bowen and Robert Ward

Keep in touch with the Grantham Research Institute at LSE
Sign up to our newsletters and get the latest analysis, research, commentary and details of upcoming events.