Page 145 - 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
ranging from 30 to 34 percent. Even if farmers
can adapt to CC by adjusting their management practices, such as planting decisions and choices among available cultivars, Deryng et al. (2011) found that, at the global scale, yield projections for spring wheat (-4 to -12%) and maize (-6 to -18%) would remain negative. Negative impacts of CC could be mitigated by the benefits of elevated CO2 concentration, specifically in the mid- and high- latitude areas, and in East and Southeast Asia. However, even in this case, average global crop yields were expected to decrease slightly, by 0
to 5 percent (Parry et al., 2004). As mentioned in Section 1.2, although SRES emission scenarios are not directly comparable to the new RCPs,
they both span a similar range of alternative
future emission scenarios (except for the low-
end emissions pathway, RCP2.6). According to Knutti and Sedlacek (2013), climate sensitivity
of high-end RCP8.5 (as applied in Müller and Robertson, 2014; see Figure 6), is fairly similar
to that of SRES A2. From a global perspective, Europe as a whole will be relatively little affected by CC. Most studies for 2050 indicate that, overall, the effects of changes in climate and CO2 concentration will lead to positive impacts on
crop production. This general picture is supported by the various global projections presented by Müller and Robertson (2014; Figure 6) as well as by various previous studies (e.g. Rosenzweig and Parry, 1994; Fischer et al., 2005; Nelson et al., 2010). Increases in crop yield and production are expected to be slightly to moderately positive in northern and western Europe, neutral to slightly positive in the central and eastern parts, and slightly negative to neutral for southern Europe (Olesen and Bindi, 2002; Easterling et al., 2007; Bindi and Olesen, 2011). However, for a given time horizon (e.g. 2050), the various studies show considerable differences in the spatial patterns
of projected crop yield gains or losses, as well
as in the rates of change, dependent on the choice of climate model, emission scenario (or concentration pathway) and impact model (see Section 3.3). According to Rummukainen (2014) and Hawkins and Sutton (2009), climate projection
uncertainty is particularly high for the near term
(up to 20 years), especially for regional projections, because of the enormous amount of uncertainty resulting from internal variability of the climate system. Overall climate projection uncertainty drops for lead times of 20-50 years, as climate model uncertainty decreases faster than emissions scenario uncertainty increases. Ultimately, total climate projection uncertainty rises again, due
to rapidly increasing scenario uncertainty (about GHG emissions, land-use change, socio-economic development). This should be kept in mind
when interpreting the various impact projections presented in Sections 3.2-3.4. The robustness
and level of certainty of impact models also varies, and decreases with increasing distance from the present situation – i.e. with more deviation from conditions for which they were initially built and evaluated. This is self-evident for the empirical- statistical crop-weather models (Lobell and Burke, 2010), but has been found for process-based
crop simulation models as well – e.g. in a model intercomparison with the largest ensemble of wheat models to date (Asseng et al., 2013).
In order to provide an indication of the
effects and relative importance of CC, CO2, and technological progress on crop production in Europe, we synthesized the results from various European-wide studies for wheat yields by mid- century (Figure 7). Considering CC only, impacts are slightly negative to neutral; including CO2 effects turns the picture slightly positive (around +10% for most studies). Considering technological progress/adaptation makes the biggest difference, resulting in high yield gains, although with a huge range of uncertainty. In the following section we illustrate the influence of different change variables, as well as that of using different climate models on the projected impacts.
Europe and its subregions
Effects of climate variables only
For Europe, yield projections are slightly negative if only CC impacts are considered (see Figures
125