Page 137 - Climate Change and Food Systems

chapter 4: an overview of climate change impact on crop production and its variability in europe, related uncertainties and research challenges
a. The lack of a comprehensive, well-tested methodology for the assessment of multiple risks to crop production under CC. Given
that changes in the means, variability and extremes of climate variables can imply changes in several of the relevant climate- induced risks that affect the growing conditions and yield formation of various food crops simultaneously (see Trnka et al., 2011; 2014),
it is necessary to account for not only heat and drought stress (as considered in some crop models) but also shifts in risks imposed by, for example, late frost, heavy rains that produce water-logging, or unfavourable precipitation, leading to yield losses during harvest time (Rötter et al., 2013b). In the MACSUR project, Trnka et al. (2014) began addressing this issue, using agroclimatic indicator approaches in combination with new climate projections from the CMIP5 ensemble.
b. With very few exceptions, most studies on the biophysical impacts of CC on crop production are incomplete. Usually, studies either address CC impacts on relative yield change for some key crops in their current cultivation areas, or analyse shifts in biophysical potential and risks for cultivation of a predefined set of different crops or crop groups, without simultaneously using crop models to estimate yields. In the ideal case, both types of studies should be combined, so that assessment of biophysical potential and relative yield changes is further linked with economic evaluation of a wide range of adaptation measures (not just changes in sowing dates, cultivar choice or irrigation) (see Rickards and Howden, 2012).
c. The fact that most CC impact assessments neglect to capture short-term climate
variability has a number of implications for their usefulness to address and quantify CC impacts on the different dimensions of food security, whether food supply stability or access (for further discussion, see below, and Section 3.5). Probably the most critical shortcoming is
emphasizing the influence of emission scenario uncertainty, climate model uncertainty or impact model uncertainty on the outcome of final impact projections while failing to account for the effects of short-term variability on food production and food security. It is well-established that a reliable and affordable supply of food is central to human well-being and the stability of societies (von Braun, 2008). However, CC is likely to reduce regional and global food security and the stability of agrifood systems because of increased short-term variability in supply (Lobell and Gourdji, 2012; Wheeler
and von Braun, 2013). The situation becomes especially critical if food supply shortages occur simultaneously in several important agricultural regions, as has sometimes been the case in recent years – e.g. 2007 and 2010 (Williams, 2012; Willenbockel, 2012; Lobell and Gourdij, 2012; Iizumi et al., 2013).
However, methodologies for assessing CC impacts on global and regional food supply have paid surprisingly little attention so far to capturing short-term variability occurring simultaneously at different locations (Williams, 2012). In addition, crop simulation models and other assessment tools have not been sufficiently suited to reliably capture climate variability and extremes and their impact on food production at regional and global scales (Wheeler et al., 2000; Rötter et al., 2011a; Lobell and Gourdij, 2012; Lobell et al., 2013; Wheeler and von Braun, 2013). Such information is required to effectively guide formulation of
trade and climate policies and inform decision- making on adaptation strategies at different levels of organization and different spatial scales (farm household, district, country, etc.). This research gap will require specific attention in the near future.
Figure 4 illustrates the development of impact projection results over time, from 1990s to present, using a few examples. The more recent CC impact projections for wheat and maize in Europe show less spread than earlier projections; this is probably largely the result of more consistent methodology, including choice of emissions and climate scenarios and better standards in impact model applications.
117