Download Adaptation to climate change starts with human–environment

DRAFT
Adaptation to climate change starts with human–environment interactions developed to
cope with climate variability: A risk management approach
Roger N. Jones1, Bo Lim2 and Ian Burton3
1
CSIRO Atmospheric Research, Aspendale Victoria 3195 Australia
2
UNDP National Communications Support Programme, New York 10017 USA
3
Meteorological Service of Canada, Toronto
Abstract
The most commonly used methods for national vulnerability and adaptation (V&A) assessments
are climate scenario driven, and focus on predicting future vulnerability based on biophysical
aspects of change. By discounting the role of human behaviour in shaping contemporary
responses to climate variability these methods have been unable to develop meaningful policy
responses. This paper describes an approach designed to assist countries in assessing the impacts
of, and developing adaptations to, climate change. The approach is vulnerability-based,
emphasising human-environment interactions under current climate and climate change. It shifts
from a linear, predictive climate changeimpactsadaptation methodology to one of risk
assessment that integrates the biophysical and socio-economic interactions of an activity or
system. The major questions under consideration are, 1) “To what extent will climate change
alter the present range of impacts, increasing the risk of vulnerability?” and 2) “How best can
development paths be modified to reduce current and future vulnerability to climate?” The
starting point is the relationship between current climate, the ability to cope and vulnerability.
Current climate variability and extremes are manifested as a coping range, encompassing
existing adaptations with a residual range of vulnerability separated by a critical threshold. The
degree to which climate change is likely to exceed a particular threshold will increase or
decrease the exposure to climate-related risk for a particular activity or system. Adaptation needs
are identified through critical threshold exceedance on an appropriate planning horizon and
managed by shifting the coping range using planning and policy mechanisms. This approach is
more robust than previous methods because it eschews a predictive methodology that analyses
single pathways, towards a systems approach that endeavours to distinguish sustainable
pathways amongst multiple possibilities.
Introduction
The Intergovernmental Panel on Climate Change’s (IPCC) Third Assessment Report (TAR)
concludes that climate change is happening now and some sensitive systems are responding
(IPCC, 2001b), and that serious and irreversible damages will occur within ranges of climate
change projected for this century (IPCC, 2001a). These conclusions increase the need to develop
policy for adaptation to climate change. Regional estimates of the magnitude and rate of climate
change vary widely, implying a need for a “wait and see” approach until forecasts improve, but
decisions that will be affected by climate change over their lifetimes are constantly being made.
For example, many developments in the agricultural, water, urban, industrial and forestry sectors
are constructed under assumptions of a stationary climate. In coming decades, these benefits may
be severely limited or even reversed under climate change unless adequate modifications are put
in place. Maladaptations (e.g. inappropriate forest clearance) may exacerbate climate risks.
Potential opportunities may be missed (Yohe and Dowlatabadi, 1999). How do policy makers
respond to a threat that is likely to occur but is highly uncertain in the way that it will manifest?
May 02. This paper will be submitted at end of June 02. This is a read-only draft. You can alter this document, but then you are
bound to send your comments to the authors (which will be gratefully received). Password is changeme – email comments to
[email protected]
DRAFT
This problem is particularly acute for developing nations. Many developing nations are
vulnerable to current climate (Smit and Pilifosova, 2001), most have highly variable climates and
all are limited in their capacity to adapt to climate extremes. Despite limited contributions of
historical greenhouse gas emissions, developing nations are highly vulnerable to future impacts
(AOSIS, 1999; Apuuli et al., 2000). Increasing adaptive capacity to climate change is a
development issue that competes for resources with other development issues, such as food
security, social equity, education and health (Munasinghe, 2000; Rayner, 2000). Critics of
international funds being spent on investigating long-term adaptation needs argue that reducing
the substantial current vulnerabilities should take precedence (e.g. Kelly, 2000).
More than 100 National Communications containing vulnerability and adaptation (V&A)
assessments have so for been submitted to the United Nations Framework for Climate Change
(UNFCCC), but their impact on adaptation policy has been limited. Most of the V&A
assessments were based on the seven step framework for impact assessment first described by
Carter et al. (1994) and elaborated on in UNEP (1998) and Carter and Parry (1998). This
method, referred to henceforth as the standard method, describes a cause and effect pathway that
projects climate change scenarios through impact models then formulates adaptation measures.
Different scenarios can lead to quite different adaptation strategies. Alternative approaches are to
try and distinguish the most likely pathway (what will happen) or to distinguish desirable and
undesirable pathways (what happens if). It is a linear methodology based on a causal chain of
events, the results of which are conditional upon the input scenarios and generally limited to
estimates of potential vulnerability. In this paper, we argue that the standard method has reached
its limits of utility for policy formulation.
This linear structure has contributed to, then been further emphasised by, the organisational
structure of the IPCC whereby Working Group I: The Scientific Basis feeds results onto
Working Group II: Impacts, Adaptation and Vulnerability. Furthermore, vulnerability, impacts
and adaptation are defined in the TAR as anomalies from the current state (see Box 1)
overlooking the role of vulnerability and adaptation to current climate as the basis for
recognising and addressing change.
Rather than addressing vulnerability addressed after moving from the physical environment to
social impacts, Hewitt (1983), Rayner and Malone (1997) and Adger (1999) advocate defining
vulnerability as the state of society–environment interactions under stress before determining
how this stress may change. Barnett (2001) describes the standard methods, in their anticipation
of impacts in an environment of uncertainty as unsuccessful, and suggests a strategy based on
understanding of vulnerability/resilience to current climate, particularly to climatic extremes (see
also Adger, 1996, 1999). While we agree in part with this approach, we propose a method by
which the analysis of present day vulnerability informed by historical perspective (Adger, 1996),
can be combined with specific methods for characterising risk under climate change in a manner
consistent with Article 2 of the UNFCCC. A similar approach is recommended by Pielke and
Bravo de Guenni (2002).
The approach presented in this paper is a contribution to the Adaptation Policy Framework
currently being developed under the auspices of the UN Development Programme (UNDP). The
Adaptation Policy Framework comprises five basic steps: scope project, assess current
vulnerability, characterise future conditions, prioritise policies and measures and facilitate
adaptation (Figure 1). We believe that the methods of characterising climate risk as described by
this paper provide a way to manage the uncertainty of climate change by bringing together the
social aspects of vulnerability, the technical requirements of climate impact assessment and the
2
DRAFT
contingencies required of policy development. A full description of this framework and
associated methods for application is currently being prepared (Burton, et al., in prep.).
We aim to augment and improve on the standard method of impact and adaptation assessment by
incorporating the four following aspects:
 Changing the focus of assessment from a climate-based approach to a vulnerability-based
approach.
 Managing uncertainty by moving from a predictive (prescriptive) approach to a risk-based
(diagnostic) approach.
 Recognising the role of behaviour in adaptation by taking adaptations developed to cope with
climate variability as the baseline upon which to adapt to climate change.
 Managing adaptation over appropriate time horizons, taking account of how both climatic
and socio-economic changes may alter vulnerability.
1. Scope Project
• Identify policy objectives of
national plans for key systems:
•existing assessments
•expert workshops
•national consultations
•Climate risks, impacts and damages
•Socio-economic drivers
•Natural resource drivers
•Adaptation experience and capacity
•Policy and development needs
3. Characterise Future Conditions
•Climate (scenarios, trends, risks and opportunities)
•Socio-economic (scenarios, trends)
•Environment (trends)
•Adaptation policy & development options
Adaptive Capacity
Stakeholders
2. Assess Current Vulnerability
4. Prioritise Policies and Measures
•Broad adaptation strategy
•Priority measures and projects among sectors
5. Facilitate Adaptation
•Incorporate climate risks into development plans
•Evaluate and monitor policies, measures and projects
Figure 1. Adaptation Policy Framework developed at the joint UNFCCC/UNDP GEF workshop
in Canada, June 2001 with subsequent modifications.
3
DRAFT
Approach of the Adaptation Policy Framework
The standard method has successfully been used to assess the level of threat posed by enhanced
climate change (see Watson et al., 1998; McCarthy et al., 2001). The knowledge that systems
have always responded to climate, and that significant vulnerabilities are on the horizon, shifts
the focus of adaptation assessments from “What are the potential adaptations to climate change”
to “How do we adapt to climate change?” Instead of assessments generating information about
potential adaptation, which is useful for inferring the need for mitigation, the demand is shifting
to practical application, a far more difficult role. This moves the focus of adaptation frameworks
away from scenario-based “what if” questions, towards providing guidance to policymakers.
The key issue for adaptation is not climate change itself but is vulnerability to climate
change. Climate change is a significant global issue because of vulnerability. Vulnerability is a
social response to climate change impacts and can be reduced by either adaptation, or mitigation.
We build on the relationship between vulnerability, impacts and adaptation as described by
Wheaton and MacIver (1999, based on Watson et al., 1996; Box 1) but also consider
vulnerability, impacts and adaptation under current climate to show how adaptation can be used
to reduce vulnerabilities to future climate over a variety of time scales. Adaptation to current
climate is linked with a coping range of climate (Hewitt and Burton, 1971; Smit et al., 2000). By
exceeding a critical threshold defining an unacceptable level of harm, climate occurrences
beyond that range will result in vulnerability (Swart and Vellinga, 1994; Parry et al., 1996;
Downing et al., 1997; Pittock and Jones, 2000). Future vulnerability is related to the changed
frequency of threshold exceedance under climate change (i.e. over long-term planning horizons).
The development of increased adaptive capacity to cope with future climate will be informed by
the risk of threshold exceedance over the long-term, but will build on adaptive strategies
developed to cope with current climate.
Better practical management of uncertainties by moving from a predictive approach to a
risk-based approach. Uncertainty about climate change is unlikely to be significantly reduced
in the short term, but neither can we adopt a “wait and see” strategy. Rather than trying to predict
impacts through individual scenarios, we focus on the triggers, or critical thresholds, that signal a
state of vulnerability. By using the same set of climate variables in threshold measurement,
scenario construction and impact assessment it is possible to determine where a threshold is
located within a range of future climate uncertainties. This approach to managing uncertainty
does not reduce the total uncertainty, but establishes agreed goals around which it is possible to
assess the overall risk of achieving (positive) or avoiding (negative) goals within the context of
projected climate change (Jones, 2001).
Planned adaptation is a behavioural response to information about the future. Planned
adaptation needs to take account of information about climate change and to develop appropriate
behavioural responses, which will be based on current individual, community and institutional
behaviour. Adaptation measures need to be consistent with current behaviour and future
expectations if they are to be accepted by stakeholders. Existing adaptation is a response to the
net effects of current climate (variability and change) as expressed by the coping range. An
understanding of current adaptation capacity is necessary to understand current vulnerability.
The analysis of behavioural responses to current climate variability also aids in the construction
of climate scenarios.
4
DRAFT
Management over appropriate time horizons. If adaptive capacity is to be addressed in a
proactive manner, both short-term and long-term vulnerabilities need to be managed (Adger and
Kelly, 1999). Climate change will manifest over the coming decades but current vulnerabilities
and national development plans will both manifest over the short term. Given the demands on
limited available resources (financial, technological, human) in the context of human
development, adaptation will need to demonstrate short-term gain, even if further benefits are
expected over the long term.
Box 1. Definitions of vulnerability, impacts and adaptation as defined by IPCC (2001b)
Vulnerability (V)
The degree to which a system is susceptible to, or unable to cope with, adverse effects of climate
change, including climate variability and extremes. Vulnerability is a function of the character,
magnitude and rate of climate variation to which a system is exposed, its sensitivity, and its
adaptive capacity.
Impacts (I)
Consequences of climate change on natural and human systems. Depending on the consideration
of adaptation, one can distinguish between potential impacts and residual impacts.
Adaptation (A)
Adjustment in natural or human systems in response to actual or expected climatic stimuli or
their effects, which moderates harm or exploits beneficial opportunities. Various types of
adaptation can be distinguished, including anticipatory and reactive adaptation, private and
public adaptation, and autonomous and planned adaptation.
Adaptive capacity
The ability of a system to adjust to climate change (including climate variability and extremes)
to moderate potential damages, take advantage of opportunities, or cope with the consequences.
V
Tc
I
A
Figure 2. Structure of vulnerability, impacts and adaptation where V and A are separated by a
critical threshold, (Tc) marking the level where impacts result in an unacceptable degree of harm.
IPCC (2001b) definitions imply the critical threshold is residual after adaptation whereas Pittock
and Jones (2000) define the critical threshold in terms of recognising the need to adapt.
Adopting a Vulnerability-based Approach
This paper integrates the biophysical and socio-economic aspects of climate variability and
change to develop an understanding of how climate risks may change over time, and to develop
5
DRAFT
adaptation options to manage those risks. The basic unit is the coping range, which relates the
major climatic drivers of a system with outputs that can be measured in terms of success or
failure (Smit et al., 2000; Jones, 2001). The coping range is distinguished as a range of
variability expressed in terms of climate or related system variable(s) that can be linked to
climate, such as streamflow and water supply, agricultural yield, forestry yield or levels of
income or profit. Within the coping range, conditions range from being desirable to tolerable
levels of harm, while beyond the coping range, tolerance (and resilience) levels are exceeded.
Combinations of climate variability and extremes can drive a system beyond its limit of
tolerance, defined by a critical threshold (Smit and Pilifosova, 2001). Because the starting point
for any adaptation is current behaviour, we analyse current climate vulnerability and how that
vulnerability is managed before proceeding on to an assessment of future risks. This exploration
is ideally carried out with stakeholders, and may result in different stakeholder-specific coping
ranges for the one activity. For example, dryland and irrigation farmers have very different
coping ranges with regard to seasonal rainfall, with dryland farmers requiring regular rainfall to
maintain soil moisture while irrigators are more dependent on volume in storage and can cope
with seasonal rainfall deficits.
The further climate change moves a system beyond its coping range, the greater vulnerability
will be. This idea is encapsulated in Smith et al. (2001) and Figures 2 to 4. The most vulnerable
systems will be those with critical thresholds that are exceeded at low levels of global warming.
As global warming increases, the damage levels to many individual systems will increase and the
number of systems at risk will also increase. Adaptation can reduce vulnerability by increasing
the coping range, maladaptation will reduce that range. Changing adaptive capacity can also
influence how the coping range evolves, independently of climate change (Smit et al., 2000;
Yohe and Tol, 2002). Therefore, it is possible to investigate how climate and socio-economic
factors influence the ability to cope with climate either individually, jointly as independent
variables and as interacting variables.
By placing adaptations within a framework of evolving policy for sustainable development both
short- and longer-term climate risks can be addressed. The major strategies for risk management
are adaptation and mitigation. Management through adaptation requires a better understanding of
adaptive behaviour in response to current climate risks, integrated with improved
characterisations of risk and if adaptation under climate change. Mitigation will reduce the
likelihood of climate breaching the coping range by reducing the rate and magnitude of climate
change, but is not dealt with any further here.
Risk management through adaptation is consistent with Article 2 of the UNFCCC which aims to
prevent dangerous anthropogenic interference with the climate system in terms of threats to (i)
the natural adaptation of ecosystems, (ii) the maintenance of food security and (iii) economically
sustainable development to proceed. However, in comparison to mitigation, which reduces
climate on a global scale because of the rapid mixing of greenhouse gases in the atmosphere,
adaptation is a bottom-up approach that needs to deal with system- and location-specific impacts.
This approach also begins at the local level, rather than the standard approach which is
downscaled from GCM output at the global level, so is more appropriate to the local scale
(Pielke and Bravo de Guenni, 2002). We investigate local and regional systems to assess which
activities are the most vulnerable and aim to reduce both current and future climate-related risks
through adaptation. While many of the tools described here are utilised in the standard method,
the overall approach is quite different, taking a system rather than a linear approach (Table 1).
6
DRAFT
Table 1. Table comparing the structure of the standard methods and adaptation policy
framework. The key difference is that the standard method projects each discipline from now
into the future separately, with each pass building on the last, whereas the APF builds an
understanding of system specific links between the biophysical and socio-economic under
current climate, then projects the whole structure through time.
Methodology
Standard
Structure
Linear,
causal
Begins
Starting focus
on climate
change science
Builds
Builds from one
discipline to the
next
Moves forward
Builds layers that project
forward in time: climate,
biophysical impacts, socioeconomic impacts, adaptation
APF
System,
relational
Starting focus
on stakeholders’
experience of
current climate
Builds using an
interdisciplinary
approach
Builds framework linking
biophysical to socioeconomic under current
conditions, then projects the
whole framework over
planning horizons
Output
Scenario-derived options for
adaptation to future conditions,
often based on biophysical
impacts, delivered towards the
end of the project
Knowledge of current adaptations
and adaptive capacity during the
project, comparison of current
and future climate risks mid
project, options for policy,
planning and management at the
end
Characterising climate risk
The coping range is used to identify climate-related hazards, and thus to characterise climate
risks. Climate risk associates a climate-related hazard with the likelihood of given climatic
conditions, with vulnerability being the socio-economic response to a given hazard (Downing et
al., 1999). The relationship between hazard, vulnerability and risk is not clear-cut and in this
paper we will provide the basis to several formulations that can be applied within V&A
assessments. Under current climate, most climate risks deal with recurrent phenomena, which are
frequency based. Characterising climate risk under climate change requires dealing with both
frequentist and single-event uncertainties. Understanding and dealing with these uncertainties is
a significant factor in the choice of analysis and needs to be clearly communicated amongst
researchers and stakeholders (e.g. Anderson, 1998).
Frequentist uncertainties are probabilistic descriptions of recurrent phenomena used to
characterise climate variability and extremes. Return events such as the 1 in 100 year flood,
likelihood of a specific extreme temperature, probability of a given severity of drought, cyclone
frequency and magnitude are all examples of recurrent climate phenomena. Thresholds linked to
specific magnitudes and frequencies of simple and complex climate phenomena have been
constructed to manage agriculture, water resources, energy management, emergency response
and so on. These tools are generally well established and lack of access or lack of resources to
adapt and apply them is largely a development issue. Providing that access or those resources
would qualify as a form of capacity building.
Climate change adds the much more problematic single-event uncertainties. These uncertainties
are associated with a single outcome, and are relevant to questions such as “will climate
change?”, “how much will the earth warm?”, “what is the direction and magnitude of rainfall
change in my region” or “will deep ocean circulation cease?”. Probabilities can be binary, for
example whether or not climate change is happening, a simple yes/no answer with attached
likelihoods (see IPCC, 2001a); or can be a single outcome within a range of uncertainty, e.g. the
range of global warming in 2100 (IPCC, 2001a; Schneider, 2001; Wigley and Raper, 2001;
Webster et al., 2001). The IPCC state the climate change is happening with a confidence of 66–
90%, (IPCC, 2001a) but the projected range of global average temperature (e.g. 1.4°C to 5.8°C
increase by 2100) has no likelihood attached to the entire range, nor to any distribution within
that range (IPCC, 2001a).
7
DRAFT
This limit to predictability restricts the standard method, which relies on a cause and effect
pathway that moves from climate change to impacts to adaptation. If all outcomes within the
above range of global warming are possible, then all adaptations projected from scenarios within
that range are also possible. This creates enormous uncertainty – how can we prioritise
adaptation options? It also implies that to prioritise adaptation options we must first overcome
these limits of predictability for climate change. However, this need not be the case. In the
section on characterising future risks we confront this limitation and show how it is possible to
assess climate risks and prioritise adaptation without first having to predict the most likely
climate change.
The characterisation of risk under climate change requires the development of both frequentist
and single-event probabilities to properly account for current and future vulnerability (Ahmad
and Warrick, 2001). Each impact is affected by different climatic variables. Characterising these
variables in the appropriate mode (mean, variability and extremes) is a challenge for climate
scenario developers. Most impacts will occur in response to new extremes, the product of both
variability and mean change (Smit and Pilifosova, 2001). All too often, climate variability and
change have been dealt with separately. Future climate is characterised in terms of changing
means while current climate is expressed in terms of variability and extremes, or “signal” and
“noise” (e.g. Mearns and Hulme, 2001). This distinction is useful for understanding how far
climate may move from its current state, for the attribution of climate change and for describing
global climate model output, but is not so useful for understanding and managing climate change
impacts.
Current vulnerability
Current vulnerability to climate can be assessed by linking the climatic stimuli affecting an
activity to outcomes that can be used to measure success or failure. These outcomes will provide
criteria that can be used to identify an appropriate level of harm. For example, climate variables
such as seasonal rainfall, degree-days (temperature), humidity and radiation can be linked to crop
yield and hence to income; and rainfall intensity and duration can be linked to flood and property
damage, loss of life and livelihood.
The characterisation of a climate hazard can be approached in two ways. The first is to fix a level
of hazard, such as a peak wind speed of 10ms-1, hurricane severity, or extreme temperature
threshold of 35°C then survey how vulnerability changes across different regions or through time
in response to that hazard. Different societies will show varying degrees of vulnerability
depending on their physical setting and social capacity. This is the approach most commonly
taken in disaster and geophysical hazards assessments. The second is a vulnerability-based
approach that determines the level of harm that is deemed unacceptable then links that to a
specific frequency and/or magnitude of climate events. A good example is drought, where the
amount and duration of rainfall deficit and the resulting impacts are location-specific. This
second method provides a better definition of the coping range because it is based on the mutual
interaction of climate and society. The range of climate affecting the system can then be divided
into the coping range and range of vulnerability separated by a critical threshold that marks a
change from acceptable to unacceptable conditions (Smit et al., 1999; Figure 2a).
8
DRAFT
Stationary Climate &
Coping Range
Vulnerable
Coping
Range
Vulnerable
Stationary Climate &
Coping Range
Changing Climate
Vulnerable
Adaptation
Coping
Range
Vulnerable
Planning Horizon
Stationary Climate &
Coping Range
Changing Climate
Vulnerable
Coping
Range
Vulnerable
Figure 3. Schematic diagrams showing relationship between (a) the coping range and baseline
critical thresholds and vulnerability under a stationary climate (after Hewitt and Burton, 1971
and others), (b) how climate change can lead to an increase in exceedance of a current threshold,
and (c) how adaptation can establish a new critical threshold, reducing vulnerability to climate
change.
A coping range can be simple, influenced by one or two climate drivers as in Figure 3, or
complex, influenced by multiple drivers. The critical threshold combines both behavioural
aspects and biophysical aspects of an activity because it marks a point where climate drives
impacts beyond a level that is considered tolerable (Smit et al., 1999; Pittock and Jones, 2000).
Stakeholders and investigators jointly formulate thresholds that become a common and agreed
metric for an assessment (Jones, 2001). Thresholds can be defined simply, as in the amount of
rainfall required to distinguish good conditions from drought, or can be complex, such as the
accumulated deficit in irrigation allocations over a number of seasons (Jones and Page, 2001). In
most cases, critical thresholds are constructed using frequentist statistics. By providing a level
9
DRAFT
where conditions result in an unacceptable degree of harm, critical thresholds can be used for
risk assessment under both current climate and under climate change (Jones, 2001).
The width of the coping range is a function of historical adaptation, the key climate drivers
experienced over that time and system sensitivity to those drivers. In developing countries,
where the capacity to adapt has been limited by factors such as access to technology and
financial resources, climate variability is large and the reliance on climate is high (e.g. Ogallo et
al., 2000), the coping range will be small compared to the range of climate variability. Small
coping ranges are likely to be breached by numerous single events. The breaching of large
coping ranges, typical of developed countries where resilience is high, may require a sequence of
extreme events such as a string of droughts before impacts become unacceptable (e.g. Smit et al.,
1999). The ability to cope or the space within the coping range can also be defined as the range
of resilience, and is the inverse to the area of vulnerability beyond the coping range. Resilience is
the ability to “snap back” or return to previous conditions after a shock. Adaptive capacity is the
ability to adjust to change through adaptation, so is a potential that can be brought into play by
an experience of stress or information about a potential future stress.
Historically realised adaptation (usually developed in response to experienced stress) defines the
behaviour upon which any response to climate change is likely to be based. Adaptation
analogues show that adapting to a future climate is influenced by past behaviour (Glantz, 1996;
Parry, 1986; Warrick et al., 1986). This includes both autonomous and planned responses. Where
adaptive capacity to current climate is low, capacity building in the form of development
assistance will be required (Kelly and Adger, 2000).
Future vulnerability
Figure 2b shows how the coping range may be breached under climate change. For example,
represented in terms of temperature (or rainfall), the hot (wet) baseline threshold is exceeded
more frequently while the exceedance of the cold (dry) baseline threshold reduces over time.
Vulnerability will increase to extreme levels for the hot (wet) threshold over time. Figure 2c
represents the expansion of the coping range through adaptation and the consequent reduction of
vulnerability. The amount of adaptation needed is a function of the planning horizon under
assessment and the likelihood of exceeding a given threshold over a given planning horizon.
Planning horizons relate to the lifetime of decision-making associated with a particular activity –
how far into the future is it planned? Is climate change likely to occur with this planning
horizon? Do current planning decisions assume the continuation of historical conditions? How
do we incorporate climate change into long-term planning? Planning horizons have a bearing on
how far into the future an assessment makes its projections. The same activity can be affected by
several planning horizons used by different stakeholders (e.g. financial, urban planning and
engineering horizons for infrastructure). Often they can come into conflict, for example
ecosystem services and livelihood as considered in perpetuity by indigenous forest dwelling
communities and “shorter-term” forestry concessions. Water and agriculture are similarly
affected by different planning horizons, and not all will be relevant under climate change. While
individuals are perhaps less interested in the long-term collectively, community and industry will
have a longer-term outlook. Climate change assessments will be less urgent for activities that
function with short-term planning horizons and where there is a high adaptive capacity, giving
high system resilience. However, we believe that where both short and long-term vulnerability is
evident, both short- and long-term factors should be incorporated into planning decisions.
10
DRAFT
Figure 3 shows a time-slice through Figure 2 reflecting a particular planning horizon. It has a
structure similar to Figure 1 but where Figure 1 is based on anomalies of change, Figure 3a
accounts for vulnerability and adaptation under current climate and Figure 3b incorporates
climate change. Figure 3a shows current impacts (Ib), the coping range Cr and a reference
critical threshold (Tb). Under climate change impact events may exceed the coping range more
frequently (Figure 3b), so that potential vulnerability exceeds current vulnerability (Vb).
Adaptation will increase the coping range, raising the critical threshold and reducing potential
vulnerability. There are two ways to survey adaptation using this structure. One is to test the
likelihood of critical threshold exceedance under climate change under conditions where the
coping range remains constant or changes in response to socio-economic change. This refers to
the question “How much do we need to adapt from current conditions?” The other is to test how
much climate change, adaptive capacity may be able to accommodate, thereby identifying “safe”
levels of change for a particular activity, referring to question “How much can we adapt by a
particular time horizon?”.
For example, with increases in mean rainfall and rainfall intensity, and with catchment
modification, a baseline threshold for flooding would be exceeded more often under climate
change. Planned flood protection for managing current climate can be augmented to cope with
the larger floods expected under climate change. This could include larger structures but could
also include better land-use management of the catchment upstream.
V
Tb
Tc
Vb
I Tb
A
Ib
Cr
Cr
Figure 3a. Baseline impacts (Ib) relate to baseline Figure 3b. Structure of change where impacts
climate variability. The coping range (Cr) lies increase from baseline variability. Planned
below a baseline threshold and vulnerability
adaptation increases the level of the critical
threshold relative to impacts from baseline (Tb)
to that under climate change (Tc).
Characterising future risks
The previous section shows how to assess future vulnerability using a single scenario. The result
is plausible, but represents only one possibility out of many. The next step is to move to risk
assessment using a multiple scenario approach. Risk is assessed according to how likely a
threshold may be exceeded at a given date. Each climate scenario developed for a particular
V&A assessment will produce different values of vulnerability and impacts, suggesting different
rates and magnitudes of adaptation. Multiple scenarios produce significant uncertainties, so the
critical threshold serves as a common frame of reference around which these uncertainties can be
assessed. The more frequently a critical threshold is exceeded by individual (statistically
11
DRAFT
independent) climate change scenarios, the greater the risk an activity faces. Threshold
exceedance methods can utilise both aspects of a hazard described earlier (as a fixed geophysical
hazard, or as a function of a climatic driver and social vulnerability as in a critical threshold).
Likelihood can be estimated in ways that range from qualitative (e.g. expert judgement) to those
using highly applied statistical methods. In this section we demonstrate how multiple climate
scenarios can be used to assess the risk of climate change on individual activities or locations.
Two examples are used, based on sea level rise and coral bleaching. Coastal vulnerability
increases with rising sea level. The most physically vulnerable areas are those sensitive to the
smallest rise in sea level. They face inundation under almost any sea level rise scenario. Less
sensitive coastal zones will require progressively higher sea level rises before they face critical
levels of damage. Those that are sensitive to the upper limits of projected sea level rise are least
likely to be affected. A likelihood function to determine the risk of sea level rise requires an
estimate of the sea level rise sufficient to exceed a critical threshold of damage. For example,
locations A, B and C have coping ranges of climate variability that will be exceeded with sea
level rises of 25, 50 and 75 cm respectively. Given a total projected range of sea level rise from 9
to 88 cm, many scenarios will exceed the 25-cm threshold, some scenarios will exceed the 50-cm
threshold but few scenarios will exceed the 75-cm threshold (Table 1, Figure 5). This principle
holds irrespective of the probability distribution function of possible sea level rise. Coastal zones
where the coping range will be exceeded under the minimum possible scenario of sea level rise
face the greatest risk and those that can cope with the maximum sea level rise face the least risk.
Table 1. Probabilities of threshold exceedance for the three illustrative cumulative probability
distributions of sea level rise in Figure 5.
Distribution
25 cm
50 cm
75 cm
Top right (uniform)
80%
48%
16%
Lower left (central 92%
44%
4%
peak)
Lower right (skewed 77%
22%
2%
peak)
The principle is the same for activities where temperature increase is a strong driver, such as
coral bleaching, simulated as a function of increasing sea surface temperature (SST). Given two
critical thresholds for coral bleaching at +1°C and +2°C above current values, the lower
temperature threshold is much more likely to be exceeded and will always be exceeded first.
This likelihood is irrespective of the probability distribution function of SST increases. More
complex relationships between key climatic variables and impacts can be assessed in the same
way for activities within the agriculture, water resources and human health sectors, using riskresponse relationships in conjunction with critical thresholds (Jones, 2001). The impacts most
requiring adaptation will be those whose coping ranges are exceeded under current climate and
minimal levels of global warming.
This approach allows the risk of impacts to be assessed without first having to predict climate
change. Rather than trying to predict one outcome from many (e.g. the probability of climate
change being x), we can use the whole range of uncertainty to assess risk. Bayesian techniques
can also be used to great advantage. By testing how different probability distributions for a given
set of inputs affect the likelihood of threshold exceedance, we can determine how robust the
risks to a particular activity may be.
12
DRAFT
100
100
80
50 cm
25 cm
Sea Level Rise (cm)
Sea Level Rise (cm)
80
75 cm
75 cm
60
50 cm
40
25 cm
20
75 cm
60
50 cm
40
0
0
Probability (%)
Probability (%)
25 cm
20
75 cm
60
50 cm
40
25 cm
20
0
5
10
Probability (%)
60
50 cm
40
25 cm
20
0
0
100
5
10
Probability (%)
Probability (%)
75 cm
60
50 cm
40
25 cm
20
0
0
0
0
75 cm
Sea Level Rise (cm)
50 cm
40
Sea Level Rise (cm)
60
Sea Level Rise (cm)
Sea Level Rise (cm)
75 cm
80
80
80
80
100
100
100
100
0
100
0
100
25 cm
20
0
100
Probability (%)
Figure 5. Ranges of sea level rise from the IPCC Third Assessment Report (IPCC, 2001a; top
left), assuming uniform probability with the range (top right), and non-uniform probabilities
(lower left and right), together with cumulative probability distributions for threshold
exceedance and three thresholds of 25, 50 and 75 cm. These distributions have not been
produced by formal analysis and are for illustrative purposes only.
Recent debates on the need to provide probabilities for climate change have in part hinged on the
limits to prediction. If the causal sequence of events leading from emissions to climate change is
limited by its weakest link (e.g. predicting future economic activity and technology development
over the long-term), then the outputs dependent on those estimates are similarly limited. Grubler
and Nakicenovic (2001) and others use this reasoning to oppose probabilities being attached to
socio-economic projections affecting greenhouse gas emissions, thus preventing the downstream
estimation of climate change probabilities. Conversely, if probabilities are not calculated in any
formal sense, people will construct their own ad hoc conclusions based on their own social
construction of the component uncertainties (Schneider, 2001). This is equally of concern
because few people have an adequate understanding of the single-event uncertainties associated
with climate change.
This debate need not be an either/or argument. Using constructions that involve coping ranges,
thresholds and risk, can produce robust outcomes. For instance the left panel of Figure 5 shows
the single event probability of global warming as a probability distribution function consistent
with methods applied by Schneider (2001), Webster et al. (2001) and Wigley and Raper (2001).
The peak of the probability distribution function shows that a global warming of ~3°C in 2100 is
13
DRAFT
much more likely than either 1.5 or 5.5°C. However, by reconfiguring this into a probability
distribution function designed to assess the risk of warming exceedance (right panel, Figure 5),
all critical thresholds associated with a global warming of 1.4°C are 100% likely, reducing with
progressively higher temperatures. The probability of warming exceedance decreases with
increasing global warming and even if the numbers are uncertain, the structure is robust.
6
6
5
5
4
3
2
1
Increasing vulnerability
Probability of warming
exceedance
Global warming (°C)
Global warming (°C)
Probability of
warming
4
3
2
1
0
0
0
1
2
3
4
5
0
Frequency (%)
50
100
Frequency (%)
Increasing likelihood of
global warming
Figure 5. Probability of global warming calculated on the left to indicate the most likely outcome
in terms on temperature increase and on the right, to assess the likelihood of threshold
exceedance associated with specific levels of warming. Probabilities calculated by Monte Carlo
sampling of ranges of change in radiative forcing and climate sensitivity, similar to Schneider
(2001).
Can this principle be used to assess global risks? The approach described here is a bottom-up
method that assesses the risk to individual activities. It is complementary to the top-down ‘safe
corridors’ or ‘tolerable windows’ approach that aims to assesses the level of global warming at
which the aggregate costs and benefits of adaptation and mitigation are balanced and dangerous
climate change is avoided (e.g. Toth et al., 1997; Alcamo et al., 1998; Petschel-Held et al.,
1999). The synthesis from IPCC Working Group II in the TAR was constructed using
information from bottom up studies showing that risks increase as a function of global warming
(Smith et al., 2001). Figure 4 is a reproduction of Figure SPM-2 of IPCC (2001b) and links
global warming to increasing vulnerability for five reasons for concern. The activities most in
need of adaptation will be those with critical thresholds situated towards the base of the columns
Figure 4. The most robust paradigm is not one of prediction, i.e. which of the temperatures on
the left are the most likely? but is one of risk, where damages will increase with global warming
but probabilities will decrease. Risk assessment is needed to identify those activities facing the
highest risks and implement adaptation measures if they are to have a long-term, viable future.
14
DRAFT
Global Warming (°C)
6
5
Risks to
Many
Large
Increase
4
Negative
Net
for Most Negative
Regions
in All
Metrics
Higher
3
Markets
+ and -
2
1
0
1990
2010
2030
2050
2070
Risks to
Some
Increase
Negative
for Some
Regions
I
II
III
Very
Low
IV
V
2090
Year
A1
A1T
B1
Upper limit
Most
People
Worse
Off
A1F
A2
B2
Lower limit
I
II
III
IV
V
Risks to Unique and Threatened Systems
Risks from Extreme Climate Events
Distribution of Impacts
Aggregate Impacts
Risks from Large-Scale Discontinuities
Figure 4. The risk of adverse effects from global warming according to five reasons for concern.
The left part of the figure is global warming based on the six SRES greenhouse gas emission
marker scenarios, showing the estimates based on a mid-range climate sensitivity, and the whole
range. The right side shows the five lines of evidence, with increasingly negative impacts linked
to the magnitude of global warming (adapted from Figure SPM-2, IPCC, 2001b).
Adaptation
Adaptation to climate change can both ameliorate harmful impacts and exploit potential benefits.
In the IPCC TAR, adaptation is defined as an adjustment, change or a response of systems and
individual activities (see Smit and Pilifosova, 2001). Behaviour is the missing factor from most
of discussions of adaptation in the climate change literature. People cannot be expected to adjust
without investigating the social contexts that have influenced past adaptation, for example why
some measures are adopted and not others. The term adjustment in the IPCC definition takes the
role of behaviour for granted by neglecting what leads to the adjustment (e.g. information about
future climate or climate risks and the experienced stress of climate events) and how and why
that adjustment is carried out. If we consider planned adaptation to future climate as being
comprised of two major elements: information and response, then behaviour is the glue that links
them. We cannot realistically expect to adapt to climate change until without first understanding
how we have adapted to past climates. Current climate experience, the policy environment and
policy horizons, social cultural and economic considerations, and the perception of the risks
posed by climate change all influence future adaptation (Adger and Kelly, 1999).
Current adaptive capacity influences the width of the coping range with regard to the magnitude
of historical impacts. In many systems, this capacity will have evolved to limit vulnerability to
climate variability and extremes, and the link between this variability and levels of harm will be
more or less understood. In systems where adaptive capacity has been eroded, then climate
variability will pose an increased risk, where extreme climate events (in the sense of linking
specific events to vulnerability) will be common. Where adaptive capacity is below a threshold
15
DRAFT
amount, general capacity building may be needed (Sen, 1981) before adaptation to specific
drivers of change such as climate can be managed.
General capacity building may increase adaptive capacity to climate change but without specific
measures that explicitly recognise the role of climate, this cannot be assured. For example,
development within a floodplain or coastal zone subject to an increasing risk of inundation may
have overall economic benefits but vulnerability to flooding will also be increased, increasing
the risk of economic and property loss. Therefore, we argue that knowledge of whether the
coping range is likely to be exceeded more frequently in the future will inform the need for
planned adaptation requiring specific measures to increase adaptation beyond that needed to
cope with current climate.
Adaptation almost always comes at an initial cost, although the benefits may give rise both to
short and long-term benefits (Tol et al., 1998; Scheraga and Grambsch, 1998). Given the
demands on limited resources (e.g. financial, technological, human), adaptation will need to
demonstrate short-term gain in the context of development, even if long term benefits are
expected. This choice often creates conflict within economic development planning, which is
usually trying to minimise initial costs. Some form of cost-benefit analysis for the short-term
benefits, however informal, will aid the allocation of resources since many adaptations will only
be implemented if they can be shown to be efficient. This is also relevant for the building of
adaptive capacity. Adaptation for a wide range of possible circumstances will be needed where
uncertainty is too great to develop specific adaptation measures, but knowledge of the relative
areas of risk within this uncertainty will inform priorities for capacity building.
Longer-term benefits may be derived from social and environmental factors in addition to
monetary gain. Cost-benefit analysis is very sensitive to discount rate, and is therefore highly
uncertain, whereas social and environmental benefits can be assessed as part of sustainable
livelihoods along with economic criteria. Irreversible changes to society and the environment
affecting intergenerational equity in particular need to be considered. Therefore, we see the
choice of adaptation measures as being helped by cost-benefit analysis in the shorter term (up to
several decades), subject to the consideration of longer term non-monetary benefits.
By relating climate change to threshold exceedance derived from an understanding of the current
coping range, policymakers can assess the need for adaptation over both short and long-term
planning horizons using critical thresholds as the common metric. The most effective adaptations
will be those that produce returns over a range of planning horizons and for multiple criteria. If
current and future climate vulnerabilities show a similar structure with regard to a coping range,
climate impacts and adaptation, they can be linked by applying planning horizons in the context
of sustainability.
Stakeholders
Experience shows that where adaptation assessments have involved stakeholders, their
effectiveness and implementation success is enhanced (e.g. McKenzie Hedger et al., 2000;
National Assessment Synthesis Team, 2000). To accommodate the needs of stakeholders, and to
increase the likelihood of risk management options being accepted, risk practitioners have
learned that behaviour must be directly incorporated into national risk assessments (Power and
McCarty, 1998). Pielke et al. (2000) identify communication and application in decision-making
as being as important as research in decision making under conditions of uncertainty.
Stakeholders are critical for the analysis of behaviour incorporated into historical adaptation, the
16
DRAFT
construction of impact criteria and critical thresholds, and the communication of risk (Jones,
2001). This requires the involvement of stakeholders throughout the assessment process.
Stakeholders can contribute important information about the policy environment early in the
process, help to design the assessment by their knowledge of coping mechanisms, the coping
range and what levels of harm can and cannot be tolerated.
Placing stakeholders in the centre of the assessment will help to shape the policy process
(Wheaton and McIver, 1999). The policy role of stakeholders is to help governments in choosing
or modifying their own development pathways by integrating adaptation options in
economically, sustainable manner under future climate. Stakeholders can supply information
about the policy environment and prepare institutional arrangements for dealing with assessment
outcomes. Local stakeholders have a role through the selection of culturally and environmentally
appropriate forms of adaptation (Faust and Smardon, 2001). Stakeholder involvement can also
benefit mitigation efforts (IPCC, 2001c). The links and feedbacks between adaptation and
mitigation processes for individual activities in the context of sustainable development are
important but have yet to be dealt with in detail. For instance, carbon can be sequestered in
managed forests but the risk of wildfire will also be increased, putting that sequestration at risk.
The management of carbon in agricultural soils under a warmer, higher CO2 climate is also very
important. By considering the socio-economic aspect of vulnerability at the national and
international scale, this framework can also be used to link greenhouse gas emissions to human
development as the main drivers of climate change.
Elements of an effective policy framework for adaptation to climate change
Any framework that deals with vulnerability and adaptation must have behaviour at its centre.
Rayner (2000) states Approaching the climate issue from a starting point of assessing human
vulnerability and societal adaptation may seem to be far less amenable to concerted rational
action by national governments than the current focus on implementing emissions reduction
targets but goes on to say it also opens the space for discussions of the adaptive strategies that
inevitable will be required, even to tackle the likelihood of climate change resulting from present
and past emissions.
Adaptation to climate change modifies the coping range, either directly or indirectly. The coping
range is an essential part of understanding the human response to historical climate conditions.
The key factor of this approach is the link between climate and behaviour both now and in the
future. Governments, in co-operation with other stakeholders, need to be able to choose or
modify their own development pathways by integrating adaptation options in economically,
sustainable manner under future climate. A vulnerability-based approach should also promote
the integration of adaptation and mitigation strategies, strongly linking greenhouse gas emissions
to human development as the main drivers of the climate system.
Concluding remarks
An approach for adaptation assessment under climate change is proposed that deals with
vulnerability in terms of risk, then aims to manage that risk through planned adaptation. By
understanding planned adaptation in terms of human development and behaviour, this approach
is designed as policy tool to help governments choose or modify their sustainable development
pathways under future climate through adaptation, consistent with the UN Framework Climate
Convention, and other environmental convention obligations. The ultimate objective of the
17
DRAFT
framework is to help countries to minimise vulnerability and risk of exposure to future climate in
the near, mid and longer term.
This approach is underpinned by four themes:
 Changing the focus of assessment from a climate-based approach to a vulnerability-based
approach.
 Managing uncertainty by moving from a predictive (prescriptive) approach to a risk-based
(diagnostic) approach.
 Recognising the role of behaviour in adaptation by taking adaptations developed to cope with
climate variability as the baseline upon which to adapt to climate change.
 Managing adaptation over appropriate time horizons, taking account of how both climatic
and socio-economic changes may alter vulnerability.
This approach signals a departure from the standard assessment methodology and closely
approaches the emerging science of sustainability (Kates et al. 2001), which strongly links
environment to development. We believe the approach described here provides guidance for a
new direction of adaptation assessments to meet the urgent needs of developing countries in
particular. Its underlying thrust is to understand what determines the vulnerability or resilience of
the nature-society systems under both current and future climate. Under a range of uncertainties
and possible future climates, there is much room for innovative methodological approaches to
support the Adaptation Policy Framework (Burton et al., in prep.) for decision-making. Although
a number of other authors have promoted a vulnerability-based approach as opposed to the
standard approach, we believe that the methods of characterising climate risk described in this
paper, provide the opportunity to improve upon the standard climate-driven approach.
Acknowledgements
Thanks to attendees of the UNDP-GEF Workshop for Developing an Adaptation Policy
Framework, Montreal Canada 11–14 June 2001 who provided feedback on many of the basic
concepts and approaches. Yvette Aguilar, Pasha Curruthers, Emilio Sempris, Cecilia Conde,
Oscar Paz; Joel Smith, Samir Safi, Isabelle Niang-Diop, Richard Klein, Rose Perez, Merylyn
McKenzie Hedger, Yasemin Biro, Pierre Giroux, Youssef Nassef, Roger Street, Helmi Eid,
Rawleston Moore, Carlos Corvalan, Pamela Kertland, Bettina Menne, Gina Ziervogel, Penny
Bramwell, Daniel Bouille, Xu Yinlong, Avelino Suarez, Rizaldi Boer, Irina Yesserkepova,
Andrew Githeko, Harald Dovland, Vute Wangwacharakul, David Cooper, Olga Pilifosova, Paola
Rossi, Louise Barbe, Liza Leclerc, Monirul Mirza, Elizabeth Malone, Sonja Vidic, Gary Yohe,
Patti Edwards and especially Barry Smit, for his singing and impetus to develop what eventually
become Figure 7.
References
Burton, I., Lim, B. and al Huq, S. (in prep.) The Adaptation Policy Framework.
Adger, W. N. (1996) Approaches to vulnerability to climate change. Global Environmental
Change Working Paper 96-05, Centre for Social and Economic Research on the Global
Environment, University of East Anglia and University College London.
Adger, W.N. (1999) Social vulnerability to climate change and extremes in coastal Vietnam,
World Development, 27, 249–269.
18
DRAFT
Adger, W.N. and Kelly, P.M. (1999) Social vulnerability to climate change and the architecture
of entitlements, Mitigation and Adaptation Strategies, 4, 253–266.
Alcamo, J., Leemans, R. and Kreileman, G.J.J. (1998) Global Change Scenarios of the 21st
Century: Results from the Image 2.1 Model. Pergamon & Elsevier Science, London,
United Kingdom.
Ahmad, Q.K. and Warrick, R.A. (2001) Mehtods and Tools, in McCarthy, J.J., Canziani, O.F.,
Leary, N.A., Dokken, D.J. and White, K.S. (eds.) Climate Change 2001: Impacts,
Adaptation, and Vulnerability, Contribution of Working Group II to the Third
Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge
University Press, Cambridge, 105–143.
AOSIS (1999) Communique: Third Summit of the Heads of State and Government of the
Alliance of Small Island States, Adopted at the United Nations, 25 September 1999,
http://www.sidsnet.org/aosis/statements/06.html.
Apuuli, B., Wright, J., Elias. C. and Burton, I. (2000) Reconciling national and global priorities
in adaptation to climate change: With an illustration from Uganda, Environmental
Monitoring and Assessment, 61, 145–159.
Anderson, J.L. (1998) Embracing uncertainty: the interface of Bayesian statistics and cognitive
Psychology,
Conservation
Ecology,
2,
Article
2,
http://www.consecol.org/Journal/vol2/iss1/art2
Barnett, J. (1999) Adapting to climate change in Pacific Island countries: the problem of
uncertainty, World Development, 29, 977–993.
Burton, I. (1997) Vulnerability and adaptive response in the context of climate and climate
change, Clim. Change, 36, 185–196.
Carter, T.R., Parry, M.L., Harasawa, H. and Nishioka, S. (1994) IPCC Technical Guidelines for
Assessing Climate Change Impacts and Adaptations, University College, London and
Centre for Global Environmental Research, Japan, 59 pp.
Downing, T.E., Ringius, L., Hulme, M. and Waughray, D. (1997) Adapting to climate change in
Africa, Mitigation and Adaptation Strategies for Global Change, 2,19–44.
Downing, T.E., Gawith, M.J., Olsthoorn, A.A., Tol, R.S.J. and Vellinga, P. (1999) Introduction,
In Downing, T.E., Olsthoorn, A.A. and Tol, R.S.J. (eds.), Climate, Change and Risk,
Routledge, London and New York, 1–18.
Faust, B.B. and Smardon, R.C. (2001) Introduction and overview: environmental knowledge,
rights, and ethics: co-managing with communities, Environmental Science and Policy, 4,
147–151.
Glantz, M.H. (1996) Forecasting by analogy: local responses to global climate change. In: Smith,
J., N. Bhatti, G. Menzhulin, R. Benioff, M.I. Budyko, M. Campos, B. Jallow, and F.
Rijsberman (eds.), Adapting to Climate Change: An International Perspective, SpringerVerlag, New York, NY, USA, 407–426.
Hewitt, K. ed. (1983) Interpretations of Calamity from the Viewpoint of Human Ecology, Allen
and Unwin, Boston.
Hewitt, K. and Burton, I. (1971) The Hazardousness of a Place: A Regional Ecology of
Damaging Events, University of Toronto, Toronto.
Hulme, M., Barrow, E.M., Arnett, N.W., Harrison, P.A., Johns, T.C. and Downing, T.E. (1999)
Relative impacts of human-induced climate change and natural climate variability. Nature
397, 688-691.
Jones, R.N. (2001) An environmental risk assessment/management framework for climate
change impact assessments, Natural Hazards, 23, 197-230.
Kates, R.W., Clark, W.C., Corell, R., Hall, M.J., Jaeger, C.C., Lowe, I., Schellnhuber, H.J.,
Bolin, B., Dickson, N.M., Faucheux, S., Gallopin, G.C., Grübler, A., Huntley, B., Jäger,
19
DRAFT
J., Jodha, N.S., Kasperson, R.E., Mabogunje, A., Matson, P., Mooney, H., Moore, B.,
O’Riordan, T. and Svedlin, U. (2001) Sustainability science, Science, 292, 641–642.
Kelly, M. (2000) Smoke and Mirrors, Tiempo 36/37.
Kelly, P. M and Adger, W. M (2000) Theory and practice in assessing vulnerability to climate
change facilitating adaptation, Climatic Change, 47, 325–352.
IPCC (2001a) Summary for Policymakers, in Houghton, J.T., Ding, Y., Griggs, D.J., Noguer,
M., Van Der Linden, P.J. and Xioaosu, D (eds.) Climate Change 2001: The Scientific
Basis, Contribution of Working Group I to the Third Assessment Report of the
Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge,
944 pp.
IPCC (2001b) Summary for Policymakers, In McCarthy, J.J., Canziani, O.F., Leary, N.A.,
Dokken, D.J. and White, K.S. (eds.) Climate Change 2001: Impacts, Adaptation, and
Vulnerability, Contribution of Working Group II to the Third Assessment Report of the
Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge,
1032 pp.
Lim, B., Jones, R.N., and Burton, I. (2001) A human development perspective: a conceptual
framework for developing national policy on adaptation to climate change. In:
Challenges of a changing earth: abstracts of scientific papers and posters presented at
the Global Change Open Science Conference, Amsterdam. Congrex Holland BV,
Amsterdam, 418.
McCarthy, J.J., Canziani, O.F., Leary, N.A., Dokken, D.J. and White, K.S. (eds.) (2001) Climate
Change 2001: Impacts, Adaptation, and Vulnerability, Contribution of Working Group II
to the Third Assessment Report of the Intergovernmental Panel on Climate Change,
Cambridge University Press, Cambridge, 1032 pp.
McKenzie Hedger, M., Gawith, M., Brown, I., Connell, R., and Downing, T.E. (eds.) (2000)
Climate Change: Assessing the impacts – identifying responses. The first three years of
the UK Climate Impacts Programme. UKCIP Technical Report, UKCIP and DETR,
Oxford. 168pp.
Mearns, L.O. and Hulme, M. (2001) Climate Scenario Development, in Houghton, J.T., Ding,
Y., Griggs, D.J., Noguer, M., Van Der Linden, P.J. and Xiaosu, D (eds.) Climate Change
2001: The Scientific Basis, Contribution of Working Group I to the Third Assessment
Report of the Intergovernmental Panel on Climate Change, Cambridge University Press,
Cambridge, 739–768.
Munasinghe, M. (2000) Development, equity and sustainability (DES) and climate change, In
Pachauri, R., Taniguchi, T. and Tanaka, K. (eds.) Guidance Papers on the Cross Cutting
Issues of the Third Assessment Report of the IPCC, IPCC, Geneva, 69-110.
National Assessment Synthesis Team (2001) Climate Change Impacts on the United States:
Foundation Report, Cambridge University Press, Cambridge
Ogallo, L.A., Boulahya, M.S. and Keane, T. (2000) Applications of seasonal to interannual
climate prediction in agricultural planning and operations, Agric. For. Met., 103, 159 166.
Parry, M.L. (1986) Some implications of climatic change for human development. In Clark,
W.C. and Munn, R.E. (eds.), Sustainable Development of the Biosphere, International
Institute for Applied Systems Analysis, Laxenburg, Austria, 378–406.
Parry. M.L., Carter, T.R and Hulme, M. (1996) What is dangerous climate change? Global
Environmental Change, 6, 1–6.
Petschel-Held, G., Bruckner, T., Schellnhuber, H.J., Toth, F., and Hasselmann, K. (1999) The
tolerable windows approach: theoretical and methodological foundations, Climatic
Change, 31, 301–331.
Pielke, R.A. Jr., Saretwitz, D. and Byerly, R. Jr (2000) Decision making and the future of nature:
understanding and using predictions, In Pielke, R.A. Jr., Saretwitz, D. and Byerly, R. Jr
20
DRAFT
(Eds.) Prediction: Science, Decision Making and the Future of Nature, Island Press
Washington DC, 361-387.
Pielke, R.A. and L. Bravo de Guenni (eds.) (2001) How to evaluate the vulnerability of water
resources in changing environmental conditions, In: P Kabat (ed.) Biospheric Feedbacks
in the Climate System and the Hydrological Cycle, BAHC Synthesis Book, Stockholm,
Sweden, IGBP, in preparation.
Pittock, A.B. and Jones, R.N. (2000) Adaptation to what and why? Environmental Monitoring
and Research, 61, 9-35.
Power, M and McCarty, L.S. (1998) A comparative analysis of environmental risk
assessment/risk management frameworks, Environ. Sci. Technol., 32 (9), 224A-231A.
Rayner, S. (2000) Prediction and other approaches to climate change policy, In Pielke, R.A. Jr.,
Saretwitz, D. and Byerly, R. Jr (eds.) Prediction: Science, Decision Making and the
Future of Nature, Island Press Washington DC, 269-296.
Rayner, S. and Malone, E.L. (1998) Human Choice and Climate Change, Batelle Press,
Columbus Ohio, 4 vol.
Scheraga, J.D. and Grambsch, A.E. (1998) Risks, opportunities and adaptation to climate
change, Clim. Res., 10, 85–95.
Schneider, S.H. (2001) What is ‘dangerous’ climate change? Nature, 411, 17-19.
Sen, A.K. (1981) Poverty and Famines: An Essay on Entitlement and Deprivation, Clarendon,
Oxford.
Smit, B., Burton, I., Klein, R.J.T. and Street, R. (1999) The science of adaptation: a framework
for assessment, Mitigation and Adaptation Strategies, 4, 199–213.
Smit, B., Burton, I., Klein, R.J.T. and Wandel, J. (2000) An anatomy of adaptation to climate
change and variability, Climatic Change, 45, 223–251.
Smit, B. and Pilifosova, O. (2001) Adaptation to climate change in the context of sustainable
development and equity, In McCarthy, J.J., Canziani, O.F., Leary, N.A., Dokken, D.J.
and White, K.S. (eds.) Climate Change 2001: Impacts, Adaptation, and Vulnerability,
Contribution of Working Group II to the Third Assessment Report of the
Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge.
Smith, J.B., Schellnhuber, H-J. and Qader Mirza, M.M. (2001) Vulnerability to climate change
and reasons for concern: a synthesis, In McCarthy, J.J., Canziani, O.F., Leary, N.A.,
Dokken, D.J. and White, K.S. (eds.) Climate Change 2001: Impacts, Adaptation, and
Vulnerability, Contribution of Working Group II to the Third Assessment Report of the
Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge.
Swart, R.J. and Vellinga, P. (1994) The “ultimate objective” of the framework convention of
climate change requires a new approach for climate change research, Climatic Change
26, 343–349.
Tol, R.S.J., Fankhauser, S. and Smith, J.B. (1998) The scope for adaptation to climate change:
what can we learn from the impact literature, Global Environ. Change, 8, 109–123.
Toth, F.L., Bruckner, T., Füssel, H.-M., Leimbach, M., Petschel-Held, G. and Schellnhuber, H.J. (1997) The tolerable windows approach to integrated assessments. In: Cameron, O.K.,
Fukuwatari, K. and Morita, T. (eds.), Climate Change and Integrated Assessment
Models: Bridging the Gaps–Proceedings of the IPCC Asia Pacific Workshop on
Integrated Assessment Models, United Nations University, Tokyo, Japan, March 10-12,
1997. National Institute for Environmental Studies, Tsukuba, Japan, pp. 403-430.
UNEP (1998) Handbook on Methods for Climate Change Impact Assessment and Adaptation
Strategies, Version 2.0, Feenstra, J.F., Burton, I. Smith, J.B. and Tol, R.S.J. (eds.) United
Nations Environment Programme, Vrije Universiteit Amsterdam, Institute for
Environmental Studies,
http://www.vu.nl/english/o_o/instituten/IVM/research/climatechange/Handbook.htm
21
DRAFT
Warrick, R.A, Gifford, R.M. and Parry, M.L. (1986) CO2, climatic change and agriculture. In
Bolin, B., B.R. Döös, J. Jager, and R.A. Warrick (eds.), The Greenhouse Effect, Climatic
Change and Ecosystems, John Wiley and Sons, New York, 393–473.
Watson, R.T., Zinyowera, M.C. and Moss, R.H. (eds) (1996) Climate Change 1995: Impacts,
Adaptations and Mitigation of Climate Change: Scientific–Technical Analyses.
Contribution of Working Group II to the second assessment report of the
Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge,
878 pp.
Watson, R.T., Zinyowera, M.C. and Moss, R.H. (eds) (1998) The Regional Impacts of Climate
Change: An Assessment of Vulnerability. A Special Report of the Intergovernmental
Panel on Climate Change Working Group II, Cambridge University Press, Cambridge,
517 pp.
Wheaton, E.E. and McIver, D.C. (1999) A framework and key questions for adapting to climate
variability and change, Mitigation and Adaptation Strategies, 4, 215–225.
Webster, M.C., Forest, C.E., Reilly, J.M., Sokolov, A.P., Stone, O.H., Jacoby, H.D. and Prinn,
R.G. (2001) Uncertainty analysis of global climate change projections, MIT Joint
Program on the Science and Policy of Global Change Report No. 73, Massachusetts
Institute of Technology, Cambridge, 3pp.
Wigley, T.M.L. and Raper, S.C.B., Interpretation of high projections for global-mean warming,
Science, 293, 451–454, 2001.
Yohe, G. and Dowladabadi, H. (1999) Risk and Uncertainties, Analysis and Evaluation: Lessons
for Adaptation and Integration, Mitigation and Adaptation Strategies for Global Change,
4, 319–329.
Yohe, G. and Tol, R.S.J. (2002) Indicators for social; and economic coping capacity – moving
toward a working definition of adaptive capacity, Global Environmental Change, 12, 25–
40.
22