Page 86 - Climate Change and Food Systems

climate change and food systems: global assessments and implications for food security and trade
3. Global climate models
Almost all of the economic assessments covered in this chapter at the sector level draw on the climate projections prepared for the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC, 2007), the most recent available at the time of this study. The Fourth Assessment synthesizes climate projections from 24 global climate models that participated in the World Climate Research Project’s third Coupled Model Inter-comparison Project (CMIP3).
Global climate models, or general circulation models (GCMs), are numerical models that apply known physical, chemical and biological principles to simulate the interaction of the atmosphere, oceans, land surface, snow, ice and permafrost
in determining the earth’s climate.3 GCMs have been applied to project the responses of the climate variables (changes in temperature, precipitation, etc.) to increased GHG emissions
in the atmosphere. Advances in scientific knowledge, data and computational capacity
have led to substantial refinements of GCMs
since their initial development in the 1960s, when atmospheric models described influences on climate in terms of the interrelationships between the atmosphere and a motionless ocean slab.
The models that participated in CMIP3 included coupled atmospheric-ocean GCMs that described complex, three-dimensional atmospheric and ocean interactions. The CMIP3 also included earth system models, which describe the carbon cycle feedback, whereby changes in temperature due to carbon dioxide (CO2) emission lead to changes in land use and vegetative cover, leading to feedback effects on CO emissions.
The GCM models use parameters to represent sub-processes that occur at smaller spatial and
3 Useful references on GCMs are Bader (2008), McClusky and Qadummi (2011) and Edwards (2011). LeTruet et al., 2007, describe the development of GCM capabilities from the beginning of the IPCC process through the CMIP3 used in the Fourth Assessment. Flato (2011) and Easterbrook (2011) discuss modelling advances that will be incorporated into the CMIP5 and IPCC’s Fifth Assessment.
temporal scales. GCMs are tested on their ability to explain climate over a historical training period. The fitted relationships are then used to simulate the effects of alternative climate-forcing scenarios (representative concentration pathways or RCPs)4 that describe various levels of human-induced GHG emissions. Their results describe projected changes in climate over the 21st century or more, including changes in temperature, rainfall and atmospheric pressure.
GCMs describe climate changes over relatively large spatial and temporal scales. The GCMs participating in CMIP3 describe the earth’s surface in horizontal grids of about 100 to 600 km width, and up to 30 vertical layers in both the atmosphere and ocean. Their predictions are strongest for temperature, a variable that is relatively consistent over these large spatial scales, while projections
for precipitation, a variable influenced by smaller- scale, topographical features and cloud formations, are less reliable (McClusky and Qaddami, 2011). Results at these coarse scales are too large relative to the input requirements of most pathway and economic models. GCM outputs therefore are usually downscaled using regional climate models, statistical techniques or spatial estimations.5
GCMs contributing to the third and fourth IPCC assessments simulated a common set of greenhouse emission scenarios that depict four broad story lines of alternative, stylized future paths and the interrelationships among five drivers of GHG emissions: population, economic and social development, energy technology, land use, and government policies (IPCC, 2000) (Table 1). The four story lines, with three subsets
2
4
For its Fifth Assessment Report, the IPCC adopted four RCPs (RCP2.6, RCP4.5, RCP6 and RCP8.5), each of which designates a given GHG concentration (not emission) trajectory (Moss et al., 2010). Each RCP represents a possible a possible range of radiative forcing values (increased radiative energy)
in 2100 relative to pre-industrial values (+2.6, +4.5, +6.0, and +8.5 Watts/m2, respectively) (Weyant et al. 2009).
5 Useful references on downscaling are Wilby and Wrigley, (1997), Leung et al., (2003), Strzepak and McClusky (2010), van Vuuren, et al., 2010 and McCloskey and Qaddumi (2011).
66