Models Supporting Decision-Making and Policy Evaluation - Environmental Modelling: Finding Simplicity in Complexity - page 320

20

Models Supporting
Decision-Making and Policy
Evaluation

Mark Mulligan
Department of Geography, King's College London, UK

chaotic systems in which there is no apparent cause and
effect and which may never thus become known.

•

20.1 The complexity: making decisions
and implementing policy in the real
world

We have seen from Chapters 1, 2 and 18 that natural
systems are rarely simple and when they interact with
humans they are often complicated, complex or chaotic.
This fact means that decision-making around human
interventions to manage the environment, develop
economies, alleviate poverty and secure natural resources
have to be made in a situation of considerable uncertainty.
Though scientists are beginning to explore techniques for
managing and communicating this uncertainty through,
for example, ensemble approaches (see Chapter 25), there
remain considerable barriers to the application of models
in decision-making contexts such as those supporting
the development of environmental, developmental or
resource-management policy.
Nevertheless, it is important that policy is informed
by the best available knowledge provided by science
and even highly uncertain knowledge is better than no
knowledge for decision-making so long as the scientists
provide, and decision-makers use, knowledge with due
regard to its uncertainty. Most environmental systems are
so multilayered and complicated that traditional mech-
anisms for the communication of knowledge (reports,
scientific papers, policy briefs), whilst necessary, are not
sufficient to provide knowledge that must be both locally
relevant (geographically specific) and germane to the

As we have seen so far in this topic, for example in
Chapter 2, systems can be separated into:

simple systems in which cause and effect are known and
predictable because such systems are assembled from
what one might call 'known knowns'; 1

•

complicated systems in which cause and effect may not
be known but can be uncovered with expert knowl-
edge and suitable enquiry because such systems include
and become 'known knowns' but also some 'known
unknowns' that can be investigated and thus converted
to 'known knowns';

•

complex systems in which cause and effect may be
apparent with hindsight but prediction is usually not
possible because such systems include many so-called
'unknown unknowns';

•

1 This rather simple and effective communication of (lack of)
knowledge is often attributed to Former United States Secretary
of Defense Donald Rumsfeld in 2002 but has in fact been used on
a number of other occasions at least as far back as 1984: Epstein
(1984).