Assessing Human Vulnerability to Climate Change from an Evolutionary Perspective - Reducing Disaster: Early Warning Systems for Climate Change - page 76

4.3.1 Impact Assessment

Climate-impact assessments effectively impose projected future climate conditions
on models of present-day sensitivity to predict the impact of climate change on dif-
ferent sectors or spatial scales. Some consider the forecasted net cost, usually given
as a percentage of GDP, to be indicative of the respective system's vulnerability
(Smit and Wandel 2006 ). For example, Nordhaus ( 1994 ), Fankhauser ( 1994 ), or Tol
( 2002 ) used sector specifi c economic models to value the change in output associ-
ated with different climate scenarios. Tol ( 2009 ) explained how these climate-
impact assessments work:

For agricultural products, an example of a traded good or service, agronomy papers are
used to predict the effect of climate on crop yield, and then market prices or economic
models are used to value the change in output (…). For non-market goods and services,
such as health, other methods are needed (…). Thus, the monetization of nonmarket climate
change effects relies on 'benefi t transfer,' in which epidemiology papers are used to esti-
mate effects on health or the environment, and then economic values are applied from stud-
ies of the valuation of mortality risks in contexts other than climate change.
The costs from each sector are then added up and extrapolated to the regional or
global level. Beyond the inherent questions concerning the accuracy of climate
models, there are number of problems with the economic modelling portion of this
approach. These models implicitly factor in systems' exposure and sensitivity to
climate change, but assume either that adaptation does not occur at all (the dumb
farmer trajectory in Fig. 4.2 ), or that perfect adaptation occurs (the clairvoyant
farmer trajectory): “(…) more recent studies (…) tend to assume agents have per-
fect foresight about climate change, and have the fl exibility and appropriate incen-
tives to respond” (Tol 2009 ). Moreover, the extrapolation of economic values for
particular locations to the regional or global scale fails to account for location-
specifi c aspects of vulnerability and adaptive processes.
Another group of climate-impact assessments (such as Mendelsohn et al. 2000a , b ;
Nordhaus 2006 ) measured observed variations in economic activity across space.
These studies assumed that variations across space were caused by differences in
climate, and that they will hold over time. They then estimated the impact that
projected changes in climate would have on economic activity, sometimes extrapo-
lating to other countries. Because these models are based on empirical observations,
they implicitly factor in a realistic level of adaptation. However, they fail to take into
account the variability of adaptive capacity over time. Most problematic, they “run
the risk that all differences between places are attributed to climate” (Tol 2009 ).
These large-scale impact assessments have been particularly useful in designing
mitigation policy. For example, they have been used to estimate the social cost of
carbon emissions. However, they offer little insight into the location-specifi c char-
acteristics and causes of vulnerability, or the processes of adaptation. More perti-
nent to this topic, they fail to isolate vulnerabilities associated with specifi c aspects
of climate, like extreme events (Tol 2009 ). It would be inappropriate to use impact
assessments to design adaptation policy.