Page 133 - 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
summer, Jun.-Aug.). Those for northern and southern Europe are based on the ENSEMBLES project (Harris et al. 2010)5; those for the northern hemisphere are based on individual GCM runs from the CMIP3 dataset (Meehl et al., 2007).
Output from 11 different RCMs has recently been presented by Sloth Madsen et al. (2012). Although they show similarities in terms of the strongest temperature and precipitation changes for broad regions in Europe, the specific spatial patterns of change can still be quite different in certain subregions. This also applies, though to a much lesser extent, if the same “parent” GCM has been the common source for different RCMs (Sloth Madsen et al., 2012).
As presented by Sloth Madsen et al. (2012) for the European continent, CC projections for the period 2030-2050 vary widely, depending
on the emissions scenarios and GCMs or RCMs considered. The general tendency, however, consistent with many studies, is to project that conditions will become wetter and warming will be stronger in northern Europe than in southern Europe or the northern hemisphere. This general picture is confirmed by probabilistic projections made for the SRES A1B scenario, based on ENSEMBLES RCMs, as well as by GCM runs based on CMIP3 (Figure 2). Even though thermal growing seasons will be extended, projected increases in the frequency of heavy rains, heat waves and drought (Christensen et al., 2007) may lead to higher variability in crop performance (Trnka et al., 2011; Trnka et al., 2014).
The ENSEMBLES project is 5-year funded project by the European Commission, which aims to provide probabilistic estimates of climatic risk through grouped (”ensemble”) integrations of Earth system models in which the uncertainties noted here are explicitly incorporated.
2. Climate change impact assessment methodology for agriculture
5
Current projections of CC impacts on crop yields and food production are almost entirely based on outcomes from process-based crop simulation models (Parry et al., 2004; Challinor et al., 2009; White et al., 2011; Müller and Robertson, 2014). However, for early impact assessments (e.g. Goudriaan et al., 1990), agroclimatic indices, such as “effective temperature sum” were applied to analyse the broad-scale sensitivity of agriculture
to CC and to determine shifts in agroclimatic suitability for given crops under different CC scenarios; among other crops, this was shown for maize and wheat in Europe (Carter et al., 1991). Over time a suite of biophysical impact assessment methods and models has been developed for analysing CC effects on land suitability and productivity of agricultural crops (Nix, 1985;
Rötter et al., 1995; Harrison and Butterfield, 1996; Schlenker and Roberts, 2009; Lobell and Burke, 2010; Lobell et al., 2011; Trnka et al., 2011, 2014; Challinor, 2011; Rötter et al., 2013a).
Here we describe three basic approaches, as well as combinations among these, or with other techniques and classification schemes:
a. Agroclimatic indices
b. Statistical crop weather models
c. Crop simulation models
a. Application of the agroclimatic index approach, in combination with environmental zoning for Europe, has been presented by Trnka et al. (2011), using 11 indices (selected from a large set of potential indices) to characterize climatic suitability and risks to the production of major crops, and their shifts under different CC scenarios. In this analysis, performed for 86 stations distributed over Europe, Trnka et al.
2.1
Different approaches to assessment
113