In this guest contribution to the SWAC blog, Kirsty Lewis of the UK Met Office Hadley Centre explores the relationship between climate change and food insecurity in developing and least developed countries. The research projections paint both stark and cautiously optimistic pictures. Failure to adapt to and mitigate climate change will drastically increase food insecurity, however; successful adaptation and mitigation efforts could actually reduce vulnerability. What do the results of this research mean for West Africa? What are the region’s priorities for decreasing vulnerability to food insecurity in the face of climate change?
While it is well-understood that climate change will mean increasing frequency and intensity of extreme weather events around the world, and that we are already seeing this trend in our observations of climate, what this trend will mean in detail is harder to evaluate. Increasing floods, droughts and storms pose a threat to the stability of food production, and therefore food security, in global terms, but to respond to this threat we need evidence on the scale and geography of the threat, as well as the effectiveness of mitigation and adaptation action to tackle it.
Climate and food insecurity index
Scientists from the UK Met Office Hadley Centre have been working closely with the World Food Programme over the last five years on the challenge of incorporating our climate science knowledge with food security expertise. Together we have developed an index of vulnerability of countries to food insecurity, as a result of flood and drought events, for developing and least-developed countries. The approach we took considered the IPCC1 definition of vulnerability as a combination of exposure, sensitivity and adaptive capacity. For each of these terms we included measures of food system function that correlated with undernourishment. For the exposure component, for example, this was a measure of the number of floods and droughts that occurred over agricultural production regions and populated areas, within each country. The sensitivity component included measures of how resilient agricultural production is to adverse weather events. This included metrics such as the percentage of rain-fed agriculture. Finally the adaptive capacity component included measures of economic resilience, such as the percentage of the population below the poverty line, or structural resilience, such as the percentage of paved roads. In this way the index included, not just the impact of weather on production, but also on access to food through markets.
The index values were normalised, and so are a relative, rather than absolute, measure of vulnerability to food insecurity. The results of this index in the present day climate are shown in the figure below.
Food insecurity and climate change vulnerability index: Present day climate
Source: UK Met Office Hadley Centre
The highest levels of vulnerability are in sub-Saharan Africa; there are medium levels across much of Asia, and lower levels in South and Central America. This pattern of vulnerability corresponds to global levels of undernourishment, as the index was designed to do. However, the index was also designed so that the exposure component could include not just observed drought and flood events, but also climate model projections based on RCP scenarios.2
In addition we developed adaptation scenarios, which effectively alter the remaining two components of the index: sensitivity and adaptive capacity. We defined a ‘high adaptation’ scenario where both the sensitivity of agricultural production and the economic resilience were improved by around 10-15% in the 2050s, and a further 10-15% in the 2080s. In this scenario the change was not applied equally to all countries, allowing the most vulnerable countries to improve more rapidly than the least vulnerable. A second ‘low adaptation’ scenario was also created, with improvements of around 5-10% per time period to the non-climate aspects of the index.
While using the climate model projections under the RCP scenarios was a straightforward choice, selecting adaptation scenarios was a lot more difficult. The chosen high and low adaptation scenarios are representative of plausible levels of adaption, but of course these are not a forecast of future behaviour.
The resulting index projections under the different climate change and adaptation scenarios provide some interesting and informative insights into both the geography of climate change impacts on food security and the relative importance of adaptation and mitigation to address these impacts.
Mitigation, no adaptation scenario
If we assume no adaptation action is taken, i.e. we just consider the impact of climate change, we see that vulnerability to food insecurity increases by the 2050s globally, regardless of which mitigation scenario is followed.
For example, under a scenario of rapid and sustained reductions in greenhouse gas emissions, known as RCP 2.6, (which is also consistent with a global average temperature rise of around 2°C) the level of increase of vulnerability to food insecurity is less than for the scenario of considerable future increasing emissions, known as RCP 8.5 (which is consistent with an end of the century global average temperature rise of 4°C or more). Nevertheless, both scenarios result in deteriorating food security conditions. This is because ‘inertia in the climate system’ (a delayed response of warming from previous emissions) means that we are committed to some level of climate change in the next few decades. Beyond the 2050s the two scenarios diverge. Under the RCP2.6 scenario, the rate of climate change levels off, and vulnerability to food insecurity stabilises. It is still worse than the present day, but no worse than the 2050s.
No mitigation, no adaptation scenario
Under the RCP8.5 scenario however, the levels of vulnerability to food insecurity increase considerably. This scenario of no action on either mitigation or adaptation is the worst case future, and the consequences for food insecurity are severe.
However, a scenario where no adaptation takes place is unlikely to be realistic, so we can compare these outcomes with an alternative ‘high adaptation’ scenario. In this case we see that adaptation makes a difference in reducing vulnerability to food insecurity at all timescales and with or without mitigation.
No mitigation, adaptation scenario
For example, under the RCP8.5 climate change scenario, adaptation almost keeps pace with climate change out to the 2050s, meaning that vulnerability to food insecurity remains at levels comparable, although a little worse, than the present day. After the 2050s though, the rate of climate change increases and outpaces adaptation efforts. In this scenario the 2080s are not as bad as they would be without adaption, but still worse than the present day.
Mitigation, adaptation scenario
There is one scenario where the future vulnerability to food insecurity is lower than the present day. This is the scenario where there are rapid and sustained reductions in greenhouse gas emissions (RCP2.6), and high levels of adaptation. In this scenario adaptation keeps pace with climate change to the 2050s, and beyond this timeframe, as the climate stabilises, continued adaptation leads to reductions in vulnerability and an improving food security outlook.
The messages from this research are clear. If we take no action to tackle climate change, the consequences for food insecurity in developing and least developed countries are severe. However, by both adapting and mitigating we can tackle climate change in a way that has positive consequences for future food insecurity.
The aim of the research that we conducted was to put information and evidence about the impacts of climate change and the effects of mitigation and adaptation efforts into the public domain. We created an interactive website to show these results, and so you can see for yourself how climate change and vulnerability to food insecurity interact.
Kirsty Lewis is Climate Security Science Manager with the Met Office Hadley Centre.
1 Intergovernmental Panel on Climate Change. Definition of ‘key vulnerability’: https://www.ipcc.ch/publications_and_data/ar4/wg2/en/ch19s19-1-2.html.
2 Representative Concentration Pathways are used for climate modelling and research. They describe four possible climate futures, all of which are considered possible depending on how much greenhouse gases are emitted in the years to come, https://en.wikipedia.org/wiki/Representative_Concentration_Pathways.