Page 114 - Climate Change and Food Systems

climate change and food systems: global assessments and implications for food security and trade
specification of the characteristics of individuals. Imprecision in initial specifications can limit the range of outcomes and create path dependence.
Despite its advantages, the ABM model exhibits two weaknesses that require further attention. The first is the difficulty to extrapolate aggregate regional results from detailed individual farm cases. The second relates to the considerable model uncertainty. Van Wijk et al. (2014) reviewed 14 ABMs and found them to differ widely in their description of component processes and the detail with which climate is taken into account. Also, as most ABMs work on a yearly time-step, this can allow only for tactical and strategic decisions to
be modelled, while detailed climate risk analyses
in which drought periods and delays in the onset
of the rainy seasons occur cannot be easily captured unless transfer functions or adapted crop production values are used that can incorporate these climate effects.
6. Conclusions
Since the Stern review was published in 2006 as the first economic assessment of climate change impact, a growing economic literature on climate impact and adaptation has developed. The dominant strand of this literature was centered around IAMS to address a variety of biophysical processes translated into economic shocks. The IAMs were initially applied to climate mitigation problems and climate policy; and later applied
to adaptation – an issue that has become of increasing global concern with the realization
that climate change is already happening from past emissions and is unaffected by future emission levels. Parallel to IAMs, and given the local nature of adaptation, a growing number of household- or farm-level models are being applied to analyse local climate impacts and evaluate farmer adaptation decision processes. This paper surveys both strands of the literature focusing on agriculture. The survey is presented as a critical yet informative review that describes the results, strengths, weaknesses and gaps.
Results of the IAMs surveyed here report mostly negative and some positive outcomes of climate change on agriculture. These models also suggest, within the limitations and uncertainties
of these modelling frameworks, substantial capacity to offset negative climate change through adaptive supply-and-demand responses, productivity-enhancing investments and trade.
A number of characteristics are shared by the studies described here. First, they rest on carefully constructed foundations that are as much as possible transparent, timely and consistent in their integration with physical and biophysical models. Most of the assessments are based on climate change projections drawn from the 2007 IPCC Fourth Assessment. Several of the more recent models (that contributed to the Fifth Assessment) are also included and are shown to account for multiple pathways through which climate change affects economies.
■ Methodology issues
Nevertheless, the application of IAMs to adaptation presented a number of challenges. A starting difficulty is the modeller’s treatment of adaptation, which varies widely, ranging from implicit or indirect (through assumptions of higher agricultural productivity rate or water-efficiency rate) to
a simple specification of a few discrete crop management responses. A second difficulty relates to the issue of uncertainty about future climate outcomes. In IAMs uncertainty has been dealt with by combining several GCM outcomes and a range of socio-economic pathways agreed to under the IPCC framework. More recently, climate modellers have formed research networks to coordinate model scenarios and minimize model-based uncertainty (e.g. AgMip, MACSUR).
Improving economic analysis of adaptation require several model improvements. First, there
is the need to improve better representation and integration of biophysical processes into economic models. This require economists to increasingly work with researchers from other disciplines, recognizing that climate change impacts and
their analysis is a multi-dimensional problem. Also
94