Does climate change mitigation modelling need its own R?

Two men install solar panels on a barn roof on Grange farm, near Balcombe.
Two men install solar panels on a barn roof on Grange farm, near Balcombe.
(c) Kristian Buus | Attribution 2.0 Generic (CC BY 2.0)

Dr Ajay Gambhir, Senior Research Fellow at the Grantham Institute, considers what climate scientists could learn from the COVID-19 crisis about communicating modelling results.

When Prime Minister Boris Johnson addressed the nation on Sunday 10 May, just over seven weeks after England’s lockdown began, he outlined five tests that must be passed before we could start moving on. These included protecting the NHS, providing enough personal protective equipment (PPE), and achieving sustained falls in infection and fatality rates. The fifth and final test was to ensure that any easing of the lockdown did not result in the reproduction, or R, rate of COVID-19 rising back above one. If it did, then each person infected would infect, on average, more than one other person, leading to an exponential rise in cases.

We’ve become only too aware of the R number over the first half of 2020, and its importance to understanding how the COVID-19 crisis could unfold. Arguably, it’s become a highly effective way to communicate the level of danger posed by the pandemic. Notwithstanding the potential misuses of R, as documented in a recent Nature news feature on its many nuances and uncertainties, it still has a simplicity and power in communicating an important aspect of this pandemic: if R is high, then we might need to take drastic action because the coronavirus is spreading faster and faster; if it’s low, and ideally far lower than one, then it’s in decline and we might be able to ease up.

More than a few parallels have been drawn between the near-term disruptive nature of the pandemic and the longer-term disruptive potential of climate change. Similarities range from the apparent importance of listening to experts and the advice stemming from their models, through to the potential to address these dangers via dramatic behavioural shifts.

Another intriguing parallel might come from the communication of modelling results. Could the climate change analytical community, and specifically the part of it that models the potential future pathways for greenhouse gas emissions, benefit from presenting its results using something like the R rate in epidemiological modelling?

Could the C rate or the I number capture the public’s imagination on climate change?

There are numerous differences between models that estimate the rise and decline of infectious diseases and those that estimate the rise and decline of greenhouse gas emissions. They operate on different timescales, often at different geographical scales, and with different objectives. But there are also useful similarities. Both allow the exploration of future scenarios around a threat whose impact varies depending on the degree of policy intervention. And both are becoming increasingly central to debates around sustainable economic development.

As such, if climate change modelling of future greenhouse gas emissions were communicated around a simple central number like epidemiology’s R, then it might capture the public’s imagination just as effectively. There are several potential candidates for such a number, including the current level of greenhouse gas emissions, the rate of decline of emissions, or any tax placed on activities that result in emissions (commonly known as the carbon price). Each of these indicates either the extent of the problem, or the extent to which a solution is in place.

For example, the expected rate of decline of emissions of the most significant greenhouse gas, carbon dioxide (CO2), over the course of 2020, has been estimated to be somewhere between five per cent and ten per cent compared to 2019 levels as a result of the coronavirus lockdowns. Some have pointed out that this rate of CO2 emissions reductions is required every year, for the next few decades, if we’re to avoid dangerous levels of climate change, as envisaged by the landmark 2015 Paris Agreement.

Another candidate number is the emissions intensity of our economic output i.e. the quantity of CO2 emitted for every unit of global Gross Domestic Product (GDP). We want our economies to continue to grow, to provide increasing goods and services and lift people out of poverty, but we don’t want emissions to increase as a result. The only way we can fully achieve such a “decoupling” of emissions and the economy is by reducing the CO2 emissions intensity of GDP to zero. Otherwise more and more CO2 will be added to the atmosphere, causing further global warming.

So, maybe those of us focused on emissions reductions modelling (known somewhat esoterically as climate change mitigation modelling) should try to popularise the notion of the C rate (i.e. the rate of year-on-year CO­2 reductions). If our C rate is less than 5% through the 2020s, we are less likely to be on track to meet the Paris Agreement’s goal of limiting global warming to 1.5oC above pre-industrial levels (as shown by Figure 1).

Figure 1: The C rate of emissions reductions in 1.5oC emissions reduction scenarios

Figure 1 shows the C rate of emissions reductions in 1.5oC emissions reduction scenarios.
Median CO2 emissions pathway that achieves a long-term temperature change of 1.5oC above pre-industrial levels, as envisaged by the Paris Agreement. Data from IPCC Special Report on 1.5oC scenario explorer 2020-2030.
 C rate shown is calculated as compound average annual reduction in CO2 over the period 2020-2030.
Median CO2 emissions pathway that achieves a long-term temperature change of 1.5oC above pre-industrial levels, as envisaged by the Paris Agreement. Data from IPCC Special Report on 1.5oC scenario explorer 2020-2030. C rate shown is calculated as compound average annual reduction in CO2 over the period 2020-2030.

Or maybe we should talk more about our I number (i.e. the CO2 emissions intensity of our GDP); if it’s more than zero, then emissions are still above the level needed for our economy to function without adding to the climate problem (as shown by Figure 2).

Figure 2: The I number of emissions intensity of GDP in 1.5oC emissions reduction scenarios

Figure 2 shows the I number of emissions intensity of GDP in 1.5oC emissions reduction scenarios
Median CO2 emissions pathway that achieves a long-term temperature change of 1.5oC above pre-industrial levels, as envisaged by the Paris Agreement. Data from IPCC Special Report on 1.5oC scenario explorer.
Median CO2 emissions pathway that achieves a long-term temperature change of 1.5oC above pre-industrial levels, as envisaged by the Paris Agreement. Data from IPCC Special Report on 1.5oC scenario explorer.

These numbers and concepts are currently obvious to those focused on producing possible pathways of CO2 emissions reductions that would address the climate change challenge. But they’re not yet tangible to the wider public. Regardless of the precise number chosen, the current crisis suggests there’s an opportunity for climate scientists to clearly and simply communicate their analysis to people who are now concerned about global risks in a way that their pre-COVID-19 lives arguably didn’t require them to be.


For the latest news, views and events from the Grantham Institute, sign up to our weekly update newsletter.

One thought on “Does climate change mitigation modelling need its own R?

Leave a Reply