Issues Magazine

Adaptation to Climate Change: The Case for Agriculture

By Bruce Campbell

Agriculture faces a massive challenge: feeding a growing population under future climates that will be less suitable for crop growth. Much effort needs to be given to adaptation of agricultural and food systems.

Our knowledge about the role of carbon dioxide for global temperatures goes back more than 100 years to Svante Arrhenius, winner of the 1903 Nobel Prize in Chemistry. But it has only been in the past few years that climate change has got the attention it deserves, though there has been little concrete action to reduce greenhouse gas emissions. Given the difficult-to-attain targets for emissions that will have to be met to control global warming, all sectors, including agriculture, will need to contribute to mitigation efforts.

To date, much of the interest in relation to climate change and agriculture has been on mitigation, but the focus is shifting. Since we know that global warming is unavoidable, much greater attention needs to be paid to adaptation. Developed countries like Australia have resources and experiences that can be applied to adapting agriculture to climate change. It will be in developing countries where the biggest challenges arise.

Agriculture has some stiff challenges ahead. While adapting to climate change, and helping to meet mitigation targets, agriculture also has to address the fact that one billion people go to bed hungry every day. On top of that, the global population will almost double by 2050, reaching nine billion people. In addition, food consumption patterns are changing as the average person in the world gets richer and consumes more food and more meat: ultimately, this requires more land under agriculture, greater agricultural productivity and different diets.

Dealing with Uncertainty
A major problem is that the current global climate models (GCMs) give rather different results, making planning for the future difficult. While there is quite a lot of agreement among models on how temperatures will change, there is little to no agreement for rainfall (Nelson G. et al., 2010; In their recent monograph, Gerald Nelson and colleagues from the International Food Policy Research Institute looked at scenarios for food security, farming and climate change to 2050. The researchers illustrate the differences among models by showing outputs for the CSIRO model and the Model for Interdisciplinary Research on Climate (MIROC) produced by the University of Tokyo’s Center for Climate System Research. The MIROC scenarios have greater increases on average but even then important agricultural regions of the world see decreased rainfall by 2050. However, for any particular region, the two models give quite different predictions.

Another problem with the GCMs is that they are ideally suited to exploring impacts over a 50–100-year period for large land masses, whereas farmers and those in agricultural R&D need information for specific areas for the next season and the next few decades. We need “downscaling” of the GCMs – information that can be applied to the time and space requirements of farmers, agricultural planners, development workers and researchers.

In addition, the models often do not give the information that is needed for specific crops and livestock. For example, researchers investigating the impacts of climate change on specific crops use crop models that often need daily data. Climate models do not provide such data.

On top of the uncertainty in future climates, we have to add the uncertainty related to the impacts of climate change on pests and diseases. Many pests and diseases of crops and animals are sensitive to climate, and we can expect these to change in currently unpredictable ways. Some will become prevalent in areas where they were previously unknown.

This all adds up to considerable uncertainty about the future. Our scientific endeavours have to try to reduce the uncertainty. We also have to get better at communicating uncertainty. This will be important for farmers so they can weigh up the odds and try to make decisions regarding their future farming practices. But it will also be important in terms of feeding appropriate information into what has become highly politically charged debates. Heavy disclaimers are needed in this uncertain world. For example, see the disclaimers on, which provides downscaled climate data for agricultural research and development.

What Will Climate Change Mean for Farmers?
Farmers are some of the first to acknowledge climate change, given what they perceive as its impacts on their practices and livelihoods. They talk of higher temperatures, strange weather patterns and new pests and diseases. (View the three photofilms “Two Degrees Up” at, where farmers from Colombia, Ghana and Kenya discuss the impacts of rising temperatures and what it means for them.) A 2°C rise will greatly alter production systems and agricultural markets. For example, coffee producers may need to switch to pineapples (Fig. 1). Local migration to “greener pastures” may be increased.

Farmers have always had to adapt, and have always been at the mercy of the weather, but there is now a new urgency given the predicted unprecedented rates of change. It is apparent that we will need accelerated adaptation.

Many scientists don’t think we will be able to hold temperature rises to 2ºC, and some have started to talk about a “four degree world”. Such a scenario would be devastating to agriculture in many parts of the world, and especially in developing countries. Australia would also be hit hard. Phil Thornton from the International Livestock Research Institute and colleagues have analysed what agriculture and food systems in sub-Saharan Africa would look like in a 4°C+ world (Fig. 2). In order to get around the uncertainty of climate change models they used 18 of the models in their analysis. By 2090, agriculture in sub-Saharan Africa would be heavily impacted, with almost all parts of Africa registering a decline in the length of growing season. This is a truly frightening scenario for a region of the world characterised by poverty and under-development.

What Will Climate Change Mean for Food Systems and Society?
It is not only farmers who will be negatively impacted. Impacts on agriculture are likely to reverberate throughout society. At the International Food Policy Research Institute, Gerald Nelson and colleagues have simulated world food price changes in response to climate change for some of the main grain crops (Fig. 3). For this they use a complex simulation model of world agricultural systems and trade. They argue that world prices are a useful single indicator of the future of agriculture, with rising prices signalling imbalances in supply and demand and growing resource scarcity, driven by demand factors such as growing population and income or supply factors such as reduced productivity due to climate change. They start their analysis by recognising that an uncertain future means there is a range of plausible outcomes. Their study considers three combinations of income and population growth: a baseline scenario (with moderate income and population growth), a pessimistic scenario (with low income growth and high population growth), and an optimistic scenario (with high income growth and low population growth).

Their analysis shows that, unlike the price declines of the 20th century, the first half of the 21st century is likely to see increases in prices. Increasing demand driven by population and income growth is greater than supply. In Figure 3, the price increase called the economic growth effect is for a 2050 world with perfect mitigation (i.e. where climate change is not having an impact on agriculture). Income and demographic changes between 2010 and 2050 result in price increases that range from 10% for rice in the optimistic scenario to 54% for maize in the pessimistic scenario. These substantial increases show the pressure under which the world’s food system finds itself, even without climate change. With climate change, prices more or less double, ranging from 31% for rice in the optimistic scenario to 100% for maize in the baseline scenario.

What we don’t see in these results, or the projections in the previous section by Thornton and colleagues, are the impacts of extreme weather events. The GCMs or the downscaled data generated by Nelson and Thornton say nothing about how climate variability may change in the future. Many climate scientists suggest that extreme events will be more frequent and more severe.

Such events will drive food price spikes. The food price spike of 2008 led to food riots and political change in several countries. Recent food price rises in Tunisia have been one of the factors, among others, in driving political dissent – thus the significance of findings of Nelson and colleagues. Food price rises of the magnitude suggested by them do not bode well for society.

One solution, which they highlight, relates to international trade. We need to strengthen international trade arrangements to compensate for different climate change effects in different locations.

Investing in Agricultural Productivity Improvements
The good news is that many of the solutions to agricultural productivity problems are available. In addition, in most developing countries there are very large yield gaps (the gap between current productivity and the known potential of a crop). By removing the yield gaps with known technologies – such as improved crop, soil and water management practices and stress-tolerant varieties – we go a long way to increasing food production.

Of course, we have to address the reasons why these technologies are not used today. Climate change provides a massive and urgent incentive to intensify efforts to disseminate the fruits of past research, to adapt it to farmer contexts in different developing countries, and to put in place the necessary policies and incentives. The benefits of adopting many of the existing technologies could be sufficient to override the immediate negative impacts of climate change.

Farmers, ranchers and fishers have long adapted to annual climate variability and passed on that knowledge from generation to generation. However, the scope of climate change may change the conditions in a certain area more quickly and more drastically than traditional adaptation methods can handle.

Centuries-old coping mechanisms may be insufficient. As the temperature rises in one cool area (A), it will resemble a warmer climate (B), whose climate will also transform into another (C).

Andy Jarvis and colleagues at the International Centre for Tropical Agriculture are developing a system they call the analogue method, which hopes to use current and past climatic data, as well as projections into the future, to identify and connect these sites (A, B, C). The aim is to combine this method with farmer-to-farmer exchanges as well as via a web-based platform. This would then create a knowledge chain through which strategies and farming information could be passed down and shared. This will allow technologies currently practised successfully in one region to be transferred to other regions as climate changes and climate belts migrate.

But as dangerous climate change threatens, merely adapting current practices and applying current knowledge will not be enough. We need to stay ahead of climate change, and to do that we need to push agricultural science to new frontiers. Climate change promises novel conditions. We need to take rapid strides to understand what is going to happen to our farming systems, and what will be needed to maintain and expand food production. We will need to breed for the new conditions. We will need to devise practices that adapt to climate change, while also helping meet mitigation targets. We will need new techniques to get more crop per drop as water becomes an even more scarce resource.

Climate Risk Management
Many believe that the spectre of the future can be seen in the excessive heat and drought in Russia that led to wildfires and a grain embargo, as well as the unprecedented floods in Pakistan. Even the recent floods in Australia have been linked to climate change.

David Jones, head of climate monitoring and prediction at the Australia Bureau of Meteorology in Melbourne, has stated that we can expect stronger weather patterns in a hotter world. But, of course, making these causal links is difficult and we need to wait for further scientific evidence.

Whatever science concludes, we do not need to wait for the future – the climate today is already having significant negative impacts on the livelihoods of farming households around the world. Droughts, floods and excessive heat are not new phenomena, and farmers have developed various ways of coping with them.

But poverty limits options. In extreme events, farmers without insurance may be forced to sell off their assets, such as animals and farming equipment, thus hampering future production.

In addition, climate risk limits investment: farmers are not willing to invest their time and money in agriculture if there is a high risk of loss. There are thus lost opportunities in the good-rainfall years. Extreme events and the associated risk play a significant part in keeping farmers in developing countries poor.

We can immediately give greater attention to climate risk management for agriculture. This will have immediate benefits – in the coming seasons – while also preparing agriculture for future climate change. Climate risk management includes the systematic use of climate information in planning and decision-making from farmers to regional food security agencies; the use of climate-informed technologies that reduce vulnerability to weather variability and uncertainty; and climate-informed policy and market-based interventions that transfer risk from vulnerable populations.

Australia has plenty of lessons to offer the world in this regard. It has long been engaged in climate information services and agricultural insurance. Australia has some excellent systems for providing farmers with relevant meteorological data: short-term forecasts, seasonal forecasts and early warning systems. While many of these innovations are not yet feasible in some developing country contexts, they can be modified or may well soon be applicable given the growth in communications technologies.

A handful of projects in developing countries have been initiated. In Mali, for example, the national meteorological service has been providing climate information to farmers, and the farmers have also learned to monitor the weather themselves to improve their decision- making. As a result, yields and incomes are significantly higher.

Some interesting index-based insurance schemes have been piloted and scaled up in countries like Mexico and India. In this approach, insurance is purchased for specific risks that are defined by reference to measured weather conditions. Insurance is automatically paid out when the weather index goes below an established level. The insurance company does not need to visit farmers’ fields to assess losses and determine payouts; instead it uses data from, for example, nearby rain gauges. As well as reducing costs, this means that payouts can be made quickly, a feature that reduces or avoids distress sales of assets.

Such a product potentially enhances the capacity of farmers, banks, micro-finance lenders and agro-based industries to take risks. ICICI Lombard in India have scaled up such products from trials in the early 2000s to more than 500,000 clients today. In Mexico, risk insurance products are well-developed, especially focusing on the smaller farmers, and now some eight million hectares of farmland are covered by traditional and index-based products. Livestock risk insurance is being trialled in Kenya, and insurance for coastal dwellers – with production activities in fisheries, plantations and crops – is being trialled in India.

There are some major challenges ahead for agriculture in the context of multiple pressures – new climatic conditions including more frequent and severe extreme events, meeting tough mitigation targets, eradicating hunger, and continuing to feed a growing population with different food demands. The solutions lie in many different endeavours, from making scientific breakthroughs in the area of downscaling climate models to promoting accelerated agricultural adaptation and strengthening international trade, among others.

While the good news is that we do have many of the technologies to deal with climate change, we will need to do a better job of ensuring that the technologies are adapted, and produce methods where learning is possible between regions as climates migrate.

Risk management technologies and approaches, which have been well-developed in Australia, need to be adapted and trialled in developing country contexts. Most notably these include innovative climate information services for farmers and new forms of risk insurance.

The Research Program on Climate Change, Agriculture and Food Security is a strategic partnership between the Consultative Group on International Agricultural Research (CGIAR) and the Earth System Science Partnership (ESSP). For more information please visit