Page 111 - Climate Change and Food Systems

Combining different analytical tools can also improve the robustness of the ability to predict farmer adaptation responses under climate change. Seo (2014) combines the Agro-ecological zones (AEZ)/Length of the growing period (LGP) methodology developed by FAO and IIASA (FAO/ IIASA 2005, 2012) with the GEF/World Bank
data set for African countries to test the model’s predictive power vis-à-vis farmers’ adaptation responses due to climate change. Seo classifies the farms in three categories – specialized crop farm, mixed farm, specialized livestock farm – and examines whether a farmer is more likely to choose a crop system when the LGP is deemed more suitable for crop production according to the AEZ methodology. The author also looks at whether the net revenue from crop production is higher when the AEZ is deemed more suitable for crops. Seo finds that the AEZ/LGP classification used alone, while well suited to predict crop patterns, is not a good predictor of non-cropping choices, such as such as diversifying portfolios to cope with varying climate factors, and is not a useful indicator of the major grains in the humid zones. Grain yields are found to be much lower in the humid zones then would be predicted by the AEZ method relying solely on LGP.
By examining the socio-economic data from the farm surveys, Seo finds that economic factors play an important role in explaining farmers’ decisions. Factors like travel hours to a port or a city, extension services, ownership of property,
and the number of household members were all important in explaining economic activities in Africa. For example, export possibilities (proximity to a port) tended to favour crop-only farms, but farms farther away from a nearest city favoured a crops- livestock farm (to take advantage of marketing opportunities in nearby urban centres). Also, a larger farm is more likely to be a livestock-only or
a crops-livestock than a crops-only (possibly from labour availability). Overall, the study revealed the type of incorrect inferences when relying solely on biophysical tools such as AEZ/LGP without taking into account the socio-economic determinants. Seo concludes with a strong recommendation for
a judicious integration of biophysical tools with economic analysis if we are to arrive at robust predictions of farm adaptation responses under climate change.
■ Food security
Most farm-level models do not yet perform integral analyses of climate change effects on food security (Van Wijk et al., 2014). Aside from food availability (which is readily modelled through changes in production), food access and stability aspects
are generally missing from household models. For example, household models typically lack proper accounting of food storage from one season to another season, as these are difficult to incorporate into a model (Van Wijk et al.). Farm household models rarely combine production with nutrition
or other socio-economic determinants. Also, most models tend to focus on a few important crops (e.g. maize, millet, sorghum, rice, many legumes) but neglect a host of other minor crops. The latter, however, can play an important role
in the diet and cash provision of smallholders
but are much more difficult to simulate with the existing models (Rodriguez et al., 2011). Most household models also give limited attention to the importance of non-agricultural activities (whether off-farm employment or ‘on-farm non-agricultural activities’), yet these can form important strategies of adaptation to climate change and are essential for improving access to food by poor rural households.
■ Risk analysis
To date, few household models have treated climate-related risk explicitly (Van Wijk et al.). A relevant question for adaptation is how climate (rainfall, temperature) variability affects crop management techniques and net farm returns. Traoré et al. (2014) examine the case of cotton, maize and small grains (millet, sorghum) in the Soudano-Sahel and find that a simple adaptation decision should give priority to planting cotton early; maize and sorghum can be delayed by up to a month without strong yield penalties; and millet should be planted last. Akponipké et al. (2010)
chapter 3: economic modelling of climate impacts and adaptation in agriculture: a survey of methods, results and gaps
91