Copernicus Institute Utrecht University and Tyndall Centre for climate change research issued a new report reviewing state-of-the-art of methods and tools available in the literature in helping
inform adaptation decisions under uncertainty. The report reviews existing frameworks for decision making under uncertainty for
adaptation to climate change. It explores how different ways of including
uncertainty in decision making match with uncertainty information provided by various
uncertainty assessment methods. It reviews a broad range of areas of climate change impacts
and impacted sectors of society and economy that may require a response of planned
Decision frameworks and analysis tools can roughly be grouped into two schools of thought:
the predictive top-down approach and the resilience bottom-up approach. Some mixed
approaches are also discussed. The difference between top down and bottom up is in the
direction in which the causal chain is followed in the reasoning: Top down starts from the top
by exploring the accumulation of uncertainty from each step going from emission scenarios,
to carbon cycle response, to global climate response, to regional climate scenarios to produce
a range of possible local impacts in order to quantify what needs to be anticipated. Bottom up starts at the bottom: the impacted system and explores how resilient or robust this system is to changes and variations in climate variables and how adaptation can make the system less
prone to uncertain and largely unpredictable variations and trends in the climate.
Much more attention has been given to the prediction oriented (top-down)
approach.Various tools, techniques and methods used in the various steps of
climate change impact and adaptation assessments were reviewed. A range of strategies to account for uncertainty in decision making and
frameworks for decision making under uncertainty of relevance for adaptation decisions has been identified, ranging from 'acceptable risk' based approaches to precaution, resilience and adaptive management. Further, a collection of tools has been identified for uncertainty analysis of relevance for informing adaptation decision making processes and discourses. Both for the frameworks for decision making under uncertainty, and for the tools
for uncertainty assessment, it has been mapped how well each of them can cope with three levels of uncertainty distinguished in the report: statistical uncertainty, scenario uncertainty and
Roughly, the top down - prediction oriented approaches are strong in statistical uncertainty
and the resilience and robustness type of bottom up approaches are strong in coping with
recognized ignorance and surprises. An essential first step in the selection of an appropriate
decision making framework and appropriate methods for uncertainty analysis for a given
climate adaptation decision making problem will thus be a well argued judgment on the
policy-relevance of each of the three levels of uncertainty - along with a judgment of their
relative importance - to the particular decision making problem at hand.
The various uncertainty assessment tools where further mapped to the various frameworks for
decision making under uncertainty in an overview table, indicating methods that are key for
a given decision making framework, methods that are complementary to a given framework
and methods that do not match a given framework.
A hypothetical case-study sketches how
these tools might be applied in practice. The case study highlights a remarkable difference in
uncertainty range for precipitation changes between the latest Dutch KNMI scenarios and results from a perturbed physics ensemble using a general circulation model.
The report recommends that a plurality of approaches (using both top down and
bottom up) need to be tried in different contexts in order to learn what works and what
doesn’t. A few niches in the field of uncertainty and climate
change adaptation can be further explored, amongst others: robust decision making methods, development of
indicators for measuring resilience, development of a catalogue of wild cards and imaginable
surprises. Further the report recommends that the policy relevance of uncertainty related to the differences in predicted uncertainty range by different methods (as the one identified in the case study) need to be further explored and discussed in the climate adaptation community.
Click here to download the full report