Noticias

Past performance is not a perfect guide to future performance, but it is the only guide we have.  It follows that scientific understanding of how agriculture will respond to future climates benefits tremendously from more precise information about past responses.  An inspiring example is the paper Nonlinear heat effects on African maize as evidenced by historical yield trials, by David Lobell, Marianne Bänziger, Cosmos Magorokosho and Bindiganavile Vivek, published in the inaugural issue of Nature Climate Change.

The authors’ innovation is to re-analyse 20,000 maize trials from research stations in southern, central and east Africa.  None of the trials were originally designed to investigate climatic effects; the post-hoc analysis was made possible by matching each trial to daily weather data extrapolated from the records of nearby weather stations.  The key finding is that each day above 30C reduced final yield by 1% under optimal conditions and by 1.7% under drought conditions – a surprisingly big impact for a crop considered relatively heat-tolerant.  Temperatures commonly exceed 30C in African sites with mean growing season temperatures of 23C and above.

The large data set allows a wealth of additional insights that are specific to maize in this region.  Hot days are especially damaging for the 21 days around silking (flowering), and hot days are worse than hot nights.  Different varieties respond differently to high temperatures, confirming that switching varieties is likely to be a vital adaptation strategy, at least within these temperature ranges.  Critically, breeding for heat tolerance should help maintain yields in drought years. 

Perhaps as interesting as the specific results are the method and the way of working that this study introduces.  The pairing of historic trial data with weather records can be applied to other crops, regions and climatic variables (e.g. a focus on precipitation rather than temperature).  In this case the data analysis was made possible by an active collaboration among Stanford University, the CGIAR’s International Maize and Wheat Improvement Center CIMMYT, national agricultural research systems, and private seed companies.

African agricultural research suffers from chronic under-investment.  FARA, the Forum on Agricultural Research in Africa, calculates from World Bank figures that 1% of the global public agricultural research budget is spent in Africa.  By 2005 only six countries (PDF) had met the promise of the 2003 Maputo Declaration to allocate 10% of their national budget to agriculture.  For instance Nigeria has a policy commitment to self-sufficiency in food production, but allocated less than 2% to agriculture in the 2011 budget.  Likewise international donors are not delivering on the aid money pledged at the Gleneagles G8 meeting in 2005 (PDF). 

Calls to reinvest in agriculture will amplify from all quarters, sharpened by growing evidence of future climatic impacts on farming and livelihoods.  Meanwhile African scientists need to be pragmatic about current constraints.  Studies like the one reported here are tricky – they require much in terms of skill, time and partnership – but they are relatively cheap.  Building on this and similar methodologies may be a great way to bring climate science and agricultural science together under African leadership.

Links:

• David B. Lobell, Marianne Bänziger, Cosmos Magorokosho & Bindiganavile Vivek. Nonlinear heat effects on African maize as evidenced by historical yield trials. Nature Climate Change, Volume: 1, 42–45 (2011).
• African Union. 10 Percent National Budget Allocation to Agriculture Development (PDF).
• Oxfam. Cooking the Books Won't Feed Anyone (PDF). Media Briefing, 18 May 2011.

This blog is the July installation of AgClim Letters, a monthly e-bulletin on science and policy written by Sonja Vermeulen, Head of Research for CCAFS. Sign up to receive AgClim Letters bulletin.  Your comments are welcome below.