Smallholder rice production in Java, Indonesia

Marshall Burke

[NEW DELHI] Crops in South Asia and Southern Africa are likely to be worst hit by climate change and need greater investment in agriculture development and adaptation strategies, say US scientists.

The conclusions, reported today (1 February) in Science, are based on an analysis of climate risks for crops in 12 food-insecure regions.

The research team, led by David Lobell from the US-based Woods Institute for Environment at Stanford University, used statistical crop models and 20 climate change models for the year 2030.

The regions studied contain groups of countries with broadly similar diets and crop production systems, and a significant number of the world's malnourished people. The researchers calculated the 'hunger importance' of individual crops by multiplying the number of malnourished individuals by a crop's percentage contribution to calorie intake.

The more important crops — that more malnourished people depend on — in South Asia were found to be millets, groundnut, rapeseed and wheat. In the Sahel region of Africa it was sorghum, and maize in southern Africa.

The researchers say there are uncertainties in the crop and climate models, and several regions with poor climate and yield data, such as Central Africa, require further research to develop adaptation strategies.

But, they say, the analysis shows a clear impact of climate change on crops, and the data is particularly robust for South Asia and Southern Africa.

Increasing and sustaining attention on agricultural investment in the developing world "is one of the best things we can do for climate adaptation", says Marshall Burke, director of the Food Security and Environment Program at Stanford University.

He told SciDev.Net that several recent initiatives — such as the World Bank's increasing investment in agriculture, and the Rockefeller and Gates Foundations' work on improving Africa's seeds systems — "look like good investments in climate adaptation for agriculture; investments in improved crop varieties, in better ways to get them in the hands of farmers, investments in rural infrastructure".

The researchers say their study only intends to highlight major areas of concern and that finer-scale studies are needed to identify local 'hot spots'.

Burke says difficulties with this include most climate models not addressing the high uncertainty in how much the climate will change in a region and how crops will respond to that change. Models also tend to focus narrowly on a small subset of crops of importance to vulnerable regions.

Link to full paper in Science