Decision-making Uncertainties.

One Earth has a great article on decision-making for complex systems under uncertainties. “Identifying uncertainties in scenarios and models of socio-ecological systems in support of decision-making“
There are many sources of uncertainty in scenarios and models of socio-ecological systems, and understanding these uncertainties is critical in supporting informed decision-making about the management of natural resources. 

---

Three major types of uncertainties are defined:

Scenario uncertainty
The qualitative description of alternative worldviews and their development into the future and the quantification of model input parameters that are conditional on these descriptions.
– Linguistic uncertainty. The use of similar terms to mean different things in different research communities, e.g., pathways, ensembles, boundary conditions.
– Narratives storyline uncertainty. The limits to imagining unknown futures (e.g., unknown unknowns). This can relate, for example, to alternative worldviews or the uncertainties associated with participatory processes arising from internal consistency and knowledge limitations.
– Scenario parameter uncertainty. The estimation of quantitative parameters from narrative storylines that are subsequently used in models. Scenario parameter uncertainty follows from the interpretation of quantitative values from qualitative narratives, e.g., the number of people in a ‘‘high population growth’’ scenario.

Model uncertainty
The representation of processes in models and how this is done.
– Structural (epistemic) uncertainty. The uncertainties associated with the choice and the representation of processes in models.
– Input data uncertainties. The variability in baseline data conditions that are used to initialize a model, including thematic classification, i.e., how classes are defined in, for example, land-use maps.
– Error propagation uncertainty. The amplification (or dampening) of the transmission of errors across multiple coupled models. The role of meta-modeling and indirect effects (such as cross-sectoral interactions).

Decision uncertainty
Communicating and translating the results of scenario and modeling studies into decision-making.
– Data interpretation for decision-making. Selective use of data or information from different sources and their interpretation.
– Analyzing at relevant spatiotemporal scales. The selection of spatiotemporal scales at which simulated data are analyzed, and the granularity of derived indicators (e.g., level of integration across biodiversity facets, merging subsets of ecosystem services).
– Decision-making tools. The variety of decision-supporting methods, e.g., multi-criteria decision analysis.

Uncertainty is often seen as the problem, while instead it could be interpreted as a ‘‘space’’ to manage socio-ecological systems in more desirable directions. Uncertainty also helps to target future effort in model development and to identify areas that lack understanding and, so, are priorities for future research.
However, structural uncertainty needs to go beyond the improvement of model components and details, by re-evaluating the fundamental principles and assumptions of a model structure.
Furthermore, part of the total uncertainty in the future of socio-ecological systems actually derives from current and future decisions and, thus, from a decision-maker or citizen point of view, represents less of an ‘‘uncertainty’’ than our ‘‘societal leeway’’ or choices. Disentangling and documenting the different sources of uncertainties in socio-ecological systems is critical in allowing the design and initiation of informed and efficient actions.
Many things about the future will always be uncertain, but we may wish to avoid the foolish and the fanatical by adopting the wisdom of doubt. Data and knowledge about socioecological systems are increasing rapidly, and knowledge improvement is often concomitant with awareness raising about system complexity. This leads to the paradox that, as technical knowledge increases, what we ignore is increasingly more important than what we know.