Annex 4
Deterministic Models of Climate Change,
Impact Assessments and Costs, and Uncertainty
DETERMINISTIC MODELS
AND IMPACT
ASSESSMENTS
Deterministic models—testable against
the historical and geographical patterns of climate change—are
used to reproduce the circulation of the atmosphere (hence General
Circulation Models, GCMs). They estimate climate change from projections
of natural and anthropogenic influences on climate including solar
input, atmospheric aerosol concentration and greenhouse gas concentrations.
Model and scenario uncertainties produce a range of estimates
of future temperature, precipitation and sea level rise. Local
impacts of climate on food production, disease incidence and water
availability are assessed using regional models embedded within
the GCMs. The Hadley Centre, supported by Defra and the UK Meteorological
Office is the leading world centre for this work.
The IPCC Third Assessment Report provides projections
of global average temperature rise ranging between 1.4 and 5.8ºC,
depending upon emissions scenario, by the end of the 21st century.
For global average sea-level rise, the projected range is 9 to
88 centimetres. The results from more recent research imply that
the IPCC range of global warming is too narrow—the upper
end of the range should perhaps be higher (eg, Stainforth et
al, [24]2005).
Further experiments using GCMs will be needed to confirm these
implications. A recent episode of the BBC's Horizon programme
("Global Dimming", 13 January 2005) suggested that short-term
cooling caused by man-made aerosol pollution in the atmosphere
could be partially offsetting some of the global warming caused
by greenhouse gases. This again is the subject of ongoing research.
Changes in global temperatures and sea level
summarized will have significant impacts on the natural environment,
ecosystems and human health and society. Many of these impacts
are already starting to be experienced. Estimates of these impacts
(whether in monetary terms or in physical metrics) are benchmarked
against projected climate changes (eg, global mean temperature,
precipitation, sea level raise) and can be cross linked to specific
emissions scenarios, for IPCC scenarios or other authors.
Most impact assessment studies in the literature
are based on the trend in climate. Considerable year to year and
decade to decade variability will in fact be superimposed on this
underlying trend. Anthropogenic changes could also initiate large-scale,
abrupt changes in physical and biological systems, such as changes
to ocean circulation and de-stabilisation of the polar ice sheets.
Such events are very difficult to predict but we cannot ignore
the risk.
Environmental stress induced or aggravated by
climate change could lead to impacts such as mass migration, severe
societal, economic and political crisis and ultimately conflict,
linked for example to access to water resources, or displacement
by flooding. Small to medium economies in sensitive regions of
the world (eg, parts of Africa, South East Asia, Central America,
small island states) are thought particularly vulnerable to these
socially contingent effects.
CLIMATE CHANGE
IMPACT COSTS
We already know that weather extremes carry
significant costs. In 2002, the severe floods in Europe caused
37 deaths and had an estimated direct cost of $16 billion, while
the heat wave in Europe in 2003 caused more than 30,000 early
deaths and had an estimated direct cost of $13.5 billion. Recent
research[25]
from the Hadley Centre indicated that such a heatwave is now four
times more likely as a result of human influence on climate, and
that by the 2050s one in two summers could be even hotter than
2003. Most climate change projections indicate that extremes such
as these are likely to occur more frequently in the future. Swiss
Re, the world's second largest insurer, said last year that the
economic costs of global warming threaten to double to $150 billion
per year in 10 years, hitting insurers with $30-40 billion in
claims, annually. A report[26]
from the ABI said that claims for storm and flood damages in the
UK have doubled to over £6 billion over the period 1998-2003,
compared to the previous five years, with the prospect of a further
tripling by 2050.
Climate change impacts assessments should consider
both projected climate changes and projected socio-economic changes,
and socio-economic scenarios have been developed for this purpose.
In the UK, corresponding climate change and socio-economic scenarios
have been produced under the UK Climate Impacts Programme for
researchers to use. However, quantitative research on climate
change impacts is sparse in many sectors, and qualitative studies
of climate impacts may often ignore the influence of possible
socio-economic changes. A particular difficulty is that to date
scenarios of climate change have been developed over relatively
long timescales (eg, 2020s, 2050s, 2080s), which are well beyond
the timeframe of most robust socio-economic modelling. Future
developments in climate change modelling will include provision
of more information on climate change over the shorter time-scales
(eg, 10 to 20 years) which are more compatible with socio-economics.
UNCERTAINTY
Estimates of climate change impacts represent
the final link in a long series of cause-effect relationships
and feedbacks which starts with assumptions about human activity
and their impacts on emissions of GHGs. Uncertainties tend to
increase at each step of this cause-effect relationship, along
what has been described as a "cascade of uncertainty"
(Jones, 2000, Corfee and Hone, 2003). So uncertainty on climate
change impacts reflects and compounds the following:
— uncertainty in socio economic
trends, including economic growth, population growth, land use,
etc;
— uncertainty in future levels
of GHG emissions given certain socio-economic trends, reflecting
uncertainty on technological progress and trends in energy and
carbon intensity of the global economy;
— uncertainty in future levels
of GHG concentrations in the atmosphere given certain levels of
GHG emissions, reflecting climate model uncertainty in the carbon
cycle and in atmospheric chemistry;
— uncertainty about temperature
and climate changes given certain level of GHG concentrations,
reflecting uncertainty about climate sensitivity;
— uncertainty about climate change
impacts for given levels of climate change, reflecting uncertainty
about vulnerability of human and natural systems, adaptive capacity,
impact of multiple-stresses, etc).
Monetary estimates of the damage costs of climate
change add another layer of uncertainty by introducing uncertainty
about monetary valuation (particularly in relation to non-market
impacts). In the case of global, aggregate estimates of damage
costs further drivers of variation include choices about how to
compare welfare impacts accruing in different regions of the world
and in different time periods.
24 Stainforth et al, "Uncertainty in predictions
of the climate response to rising levels of greenhouse gases"
Nature 433, 403-406, January 2005. Back
25
Stott et al, "Human contribution to the European
heatwave of 2003" Nature 432, 610-614, December 2004. Back
26
A changing climate for insurance, ABI, June 2004. Back
|
