Page 130 - Climate Change and Food Systems

climate change and food systems: global assessments and implications for food security and trade
forcing are uncertain for a number of reasons, including the incompleteness of climate models (e.g. certain processes are not taken into account), parameters that are difficult to estimate and, to a considerable degree, unforced natural variability (Rummukainen, 2012; 2014).
Shortcomings and uncertainties with respect
to climate modelling have been widely discussed (see Christensen et al., 2007; Boberg et al., 2012; Rummukainen, 2012). General Circulation Models, also called Global Climate Models (GCMs), have been applied to project the responses of the climate (i.e. changes in temperature, precipitation and other climatic variables) to increased GHG emissions in the atmosphere. In order to assess climate model uncertainty, multiclimate model datasets have been collected in huge international efforts such as the Coupled Model Intercomparison Project (CMIP). The fourth Intergovernmental Panel on Climate Change (IPCC) assessment report (AR4), released in 2007, utilized the multiclimate model dataset CMIP3 (Meehl et al., 2007),
which included simulations by over 20 GCMs for several emissions or climate forcing scenarios. There are numerous different GCMs, each using different numerical formulations and physical parameterizations of the atmosphere and its phenomena (Rummukainen, 2014). Whether based directly on GCMs or on climate scenario data that have been downscaled statistically (Wilby et al., 2004) or dynamically (through Regional Climate Models, or RCMs) (Rummukainen, 2010), climate projections generally show similar projected temperature or precipitation changes – i.e.
positive, negative or not evident. However, the exact magnitude of change varies widely when considering the outcome from ensembles of multiple models. Much of that variation can be attributed to discrepancies in internal variability simulated by the different models; however, the magnitude of change generally increases with higher emissions and over time.
When using the mean of the outcomes of climate model ensemble simulations for impact projections, it is important to keep in mind that, while means reveal the similarities of the various
models and their CC projections, outliers (i.e. the worst and best cases) should not be ignored, as they illustrate the (plausible) extreme responses. The best choice for a given impact study, therefore, is not to rely on the ensemble mean plus the
single most extreme outliers in both directions. Instead, if feasible, the multimodel mean should
be complemented by a number of individual projections that can delineate the model spread in the most relevant output variables for the particular study (Rötter et al., 2013a; Rummukainen, 2014).
The change in global mean temperature observed so far amounts to around +0.8o C since pre-industrial time. Increases in global mean temperature projected for the 2050s by IPCC (2007) as compared with the 2000s are between 0.5 and 3o C when considering different emission scenarios, climate models, and assumptions on feedbacks2; when considering warming since pre-industrial time, the projected increase is 1 to 3.5o C. Taking into account that observed GHG emission pathways currently follow the high-end emission scenarios (Peters et al., 2013), and given the 20-40 years
lag time in effective climate forcing of present-day emissions, it is very likely that projected changes
will exceed the international target of 2o C which was agreed upon by the United Nations Framework Convention on Climate Change (UNFCCC) to
avoid disastrous impacts. Regarding global precipitation changes, it has been suggested that the precipitation increase will be around 2 percent for each degree of warming (Rummukainen, 2014).
However, global climate response to the various forcings is not uniform. While warming will occur overall, some regions will warm more than others, and some may considerably exceed the global mean change.
In the recently released fifth IPCC assessment report (AR5), new sets of scenarios have been
2
Feedback mechanisms of the earth-atmosphere system frequently incorporate very complex processes with much detail that cannot be exactly described by or incorporated in climate change models and require simplification. Hence, climate change models employ (different) assumptions depending on how they simplify/represent certain processes.
110