Assessing the Impacts of Multiple Breadbasket Failures

A relatively small area of the world accounts for a large proportion of total global cereal production, with most of the area devoted to the production of the world's three major cereal crops, rice, wheat and maize. An extensive literature of the sensitivity of agricultural productivity of these crops, and many others, has arisen over the past 25 years, with a general consensus that continued change in the physical climate system will very likely increase the difficulty of agricultural production in areas of the world that are already marginal with respect to production. But what this research only rarely does is assess the influence of extreme events in shocking agricultural production, and how the rest of the agricultural system reacts, in terms of prices, food insecurity, subsequent land-use change, and terrestrial carbon emissions, among many other possible responses. Because the agricultural system is interlinked with energy systems, food distribution and transportation systems, and economic systems, models that focus only on agricultural productivity can only provide a unidimensional view of the magnitude of potential impacts. We know such impacts can occur as a consequence of extreme climatic events, because they have - the impact of the severe regional drought and heat wave on the Russian and Ukrainian wheat harvests in 2010 had global consequences for food prices, just as one example. In this paper, we use an Integrated Assessment Model, the Global Change Assessment Model (GCAM), to investigate the potential outcomes of both moderate and severe shocks to agricultural productivity in the major breadbaskets of the world - both singly and in combination. The results demonstrate clearly that there are likely to be multidimensional consequences from the kinds of shocks that are possible from a rapidly changing climate system, especially when combined with other demographic and economic trends in the coming decades. These results are only one aspect of calculating a risk profile - it will require process, modeling, and statistical research to calculate the probability of such events occurring, but it is a first attempt to understand their potential magnitude.