Accessibility links

FEATURE: Understanding how rainfall affects food security and migration

Kevin Henry of CARE reports the findings of a wide-ranging study into how climate change could influence food security and migration in future

How does climate variability, especially changes to rainfall, affect the livelihoods of rural households in developing countries? To what extent are households already using migration as a short-term coping or longer-term adaptation strategy? How might climate change impact the food security of smallholder farmers in the coming decades? And under what circumstances might the numbers of ‘environmental migrants’ increase significantly as climate change impacts build?

These are the sort of questions that CARE and the UN University, with funding from the AXA Group and MacArthur Foundation, set out to explore in eight very diverse contexts in Asia, Africa and Latin America under the Where the Rain Falls initiative. I was part of a team that explored the relationships among rainfall, food security and migration by undertaking field research at district level in Guatemala, Peru, Ghana, Tanzania, India, Bangladesh, Thailand and Vietnam during 2011–2012.

In addition to the wide range of macro socio-economic conditions that these countries encompass, the selected sites were very diverse in terms of annual rainfall (from <600mm to >1700mm), elevation (near sea level to >4,000m above sea level), and proximity to cities and other centres of non-farm employment. Our partners at the Center for International Earth Science Information Network of the Earth Institute at Columbia University developed original maps for each research site.

The project employed mixed research methods to investigate the impacts of climatic variability on communities at each of the study sites. These included a household survey, participatory research approaches and expert interviews, with more than 1300 household surveys completed across the eight countries and nearly 2000 individuals participating in focus group discussions and interviews.

Rural people in each of the eight research locations perceived that climate change was making rainfall more variable, observations largely corroborated by an analysis of local meteorological data. The most commonly reported changes in rainfall patterns were: delayed onset and shorter rainy seasons; reduced number of rainy days per year; increased frequency of heavy rainfall events; and more frequent prolonged dry spells during rainy seasons. Significant changes in temperature and wind patterns were also reported at many of the research sites.

The largely agriculture-based households at the research sites overwhelmingly reported that rainfall variability negatively affected crop production and was a major contributor to food and livelihood insecurity. Levels of food insecurity varied significantly across the sites depending on such factors as: the total amount and seasonality of rainfall; the degree of agricultural intensification; the extent of livelihoods diversification; and the access of poor households to social safety nets and other support services.

We observed that rainfall had a more direct relationship with household migration decisions at research sites where the dependence on rain-fed agriculture was high and local livelihood diversification was low. Often these sites only yielded a single harvest per year. We found migration to be common across all the sites and noted it was: almost entirely within national boundaries and driven by livelihood-related needs; both rural-urban and rural-rural (to more productive agricultural areas); largely by individual household members (except at the India research site); and predominantly male but with growing participation by women (30% of the total).

To understand the potential for changes in rainfall to become a significant driver of human mobility in the future, we developed an agent-based model and applied it to the Tanzania research site. Agent-based modelling is a computational social simulation technique that enables the user to model the behavior of individual decision-making entities as well as their interactions with each other and the environment. The Rainfalls Agent-Based Migration Model represents vulnerability and migration decision-making at both the individual and household level, both of which could be generated from the household survey data collected in each case study location.

The initial application of the agent-based model in Tanzania showed that migration of vulnerable households is sensitive to changes in rainfall and that out-migration in that location could double in the next 25 years under an extreme drying scenario. By contrast, migration by less vulnerable or ‘contented’ households was found to be much less sensitive to changes in rainfall patterns.

Four different household profiles emerged from analysis of the research data. These ranged from households that are able to use migration to increase their climate resilience, by making successful moves to either areas where growing conditions are more favourable or urban centres or industrial estates with non-farm employment opportunities, to those ‘trapped’ populations unable to use migration as either a short-term coping mechanism or long-term adaptation strategy.

We found wealth, land ownership, dependency ratios and education to be important characteristics in determining whether or not households were able to use migration to increase resilience or, conversely if migration was either not possible (‘trapped populations’) or constituted an erosive coping strategy (i.e. one which increased vulnerability to future shocks or prevented households from escaping poverty).

Where the Rain Falls sought to contribute to a more nuanced understanding of how climate variability and change affect human mobility, that could be taken forward in future research and policy formulation related to “environmental migration.” We conclude, from our research, that pressure on rainfall-dependent livelihoods is likely to grow as a driver of long-term migration in the coming decades if vulnerable (including land-scarce) households are not assisted, much more than is the case today, in building climate-resilient livelihoods in situ.

Better understanding of how multiple factors, including both environmental changes plus food and livelihood security, interact to shape migration choices should be used to inform future adaptation investments and policies. This will to ensure that whatever choices poor households make, including migration, they become more resilient to climate change. CARE will be using the research findings to design and implement community-based adaptation projects in Peru, Tanzania, India and Thailand, beginning in 2013.

For more information: The detailed findings and policy recommendations from our research have been published by CARE and the UN University in a Global Policy Report and seven country Case Studies Reports, all of which were recently launched at COP18 in Doha and can be found at: For more information on CARE’s global climate change work, please go to:

Read the second part of this feature – Erratic rainfall makes the future uncertain.

We  invite bloggers from around the world to share their experiences and views. The views expressed here are those of the author[s], and not necessarily those of CDKN.

, , , , , , , , , ,

Comments are closed.