An uncertain future, deep uncertainty, scenarios, robustness and adaptation: How do they fit together?
Introduction
Uncertainty has been considered extensively in the context of environmental and hydrological models for many years (Ascough et al., 2008, Durbach and Stewart, 2012, Refsgaard et al., 2007, Stewart, 2005). Approaches to dealing with uncertainty generally consider uncertainties in model inputs, model parameters, and model structure by way of probability distributions, resulting in a distribution of outputs around some “best-guess”. However, when faced with an uncertain future as a result of drivers such as climate, technological, socio-economic and political change, and corresponding policy and societal responses, the assumption that we can identify a “best-guess” output in the first place might no longer be appropriate (Haasnoot and Middelkoop, 2012, Walker et al., 2013a). This is because in such situations, there are multiple plausible future trajectories that generally correspond to distinct future states of the world that do not have an associated probability of occurrence or cannot even be ranked (Kwakkel et al., 2010). Consequently, when dealing with an uncertain future, a different conceptual approach to thinking about uncertainty is needed, which has resulted in the development of different terms that can be used to encapsulate the concept of multiple plausible futures, of which deep uncertainty (Lempert et al., 2003, Walker et al., 2013b) is arguably the most well-known.
Thinking about future uncertainty in terms of multiple plausible futures, rather than probability distributions, has implications in terms of the way uncertainty is quantified or described, the way system performance is measured and the way future strategies, designs or plans are developed. In terms of uncertainty quantification, consideration of multiple plausible futures generally necessitates the development of scenarios (e.g. Bárcena et al., 2015, Beh et al., 2015b, Gal et al., 2014, Greiner et al., 2014, Lan et al., 2015, Paton et al., 2013), rather than just sampling from probability distributions. In relation to system performance measurement, the presence of multiple plausible futures that cannot be characterised by probability distributions requires consideration of performance measures such as robustness (e.g. Kasprzyk et al., 2013, Matrosov et al., 2013, Mortazavi-Naeini et al., 2015, Paton et al., 2013, Whateley et al., 2014), which reward strategies, designs or plans that perform well under a range of future conditions, rather than performance measures that consider the probability of acceptable system performance for a “best-guess” future, such as reliability. When it comes to the development of future strategies, designs or plans, these generally need to be robust over long periods of time, making adaptive strategies (Beh et al., 2015a, Groves et al., 2014, Haasnoot et al., 2013, Haasnoot et al., 2014, Hamarat et al., 2014, Lempert and Groves, 2010, Ray et al., 2011) a viable alternative to their more commonly used static, fixed counterparts.
While each of these elements (i.e. thinking of future uncertainty as being represented by multiple plausible futures, using scenarios to quantify uncertainty, using robustness to measure system performance, and considering adaptive strategies as viable alternatives to fixed strategies) is not new in itself, they have generally been considered in isolation. This is exemplified by a number of recent synthesis papers, which have primarily focussed on one of these elements, without considering their connections. For example, Herman et al. (2015) mainly focus on measures of robustness, while Kwakkel et al. (2016a) and Dittrich et al. (2016) highlight different approaches to developing future strategies. While there are a number of review papers on scenarios (Bradfield et al., 2005, European Environmental Agency, 2009, Haasnoot and Middelkoop, 2012, Van Notten, 2005, Van Notten et al., 2005), and several examples of quantifying multiple plausible futures using scenarios (Fortes et al., 2015, Vervoort et al., 2014, Kok and Van Delden, 2009, Van Delden and Hagen-Zanker, 2009), recognition of these types of scenarios and their relevance for the quantification of multiple plausible futures have generally not featured in papers on deep uncertainty. Consequently, there is a need for a paper that offers a synthesis of how these elements fit together in the context of dealing with multiple plausible futures.
In order to address this shortcoming, the primary objective of this paper is to provide a multidisciplinary perspective on how the concepts of an uncertain future, deep uncertainty, scenarios, robustness and adaptation fit together to facilitate the development of strategies, designs and plans that are best suited to dealing with an uncertain future. The remainder of this paper is organised as follows. An outline of different paradigms for modelling the future is given in Section 2, followed by the articulation of some of the terms that encapsulate the concept of multiple plausible futures in Section 3. A classification of scenario types is given in Section 4, along with a discussion of their suitability for quantifying multiple plausible futures. A categorisation of the two main approaches to developing strategies for dealing with future uncertainties, as well as a discussion of the conditions that favour each of these approaches, is given in Section 5, followed by a discussion of the implications of considering multiple plausible futures on modelling in Section 6. Finally, a summary and concluding remarks are presented in Section 7.
Section snippets
Three complementary paradigms for modelling the future
A fundamental purpose of modelling is to help understand the future, to support planning or adaptation. We focus here on quantitative models defined by a model structure and a set of parameter values. The model is applied to input data in order to obtain estimates of future system states. The models therefore have some temporal element (even if they do not generate time series), and are usually spatially situated (even if they are not spatially distributed). The quantitative model is usually
Terms used to encapsulate the concept of multiple plausible futures
There are different terms that can be used to encapsulate the concept of multiple plausible futures. In this paper, three of these will be discussed, including “deep” uncertainty, “global/local” uncertainty and “VUCA” (Volatility, Uncertainty, Complexity and Ambiguity). These three terms appear to have evolved more or less independently and are therefore of interest for illustration purposes and are used to highlight that the issue of dealing with multiple plausible futures is gaining
Methods for identifying multiple plausible futures
Arguably the most common approach to the identification of multiple plausible futures, or “states of the world” as they are referred to by Herman et al. (2015), is the use of scenarios. Mahmoud et al. (2009) defined scenarios as “possible future states of the world that represent alternative plausible conditions under different assumptions” and Van Notten et al., 2005, Van Notten et al., 2003 defined scenarios as “coherent descriptions of alternative hypothetical futures that reflect different
Coping with multiple plausible futures in model-based decision support
The aim of model-based decision support in the face of multiple plausible futures is to assist with the development of strategies, designs or plans (referred to as strategies hereafter) that perform adequately, irrespective of which of these futures actually occurs. This results in robust outcomes, where robustness can be thought of as a measure of the insensitivity of the performance of a given strategy to future conditions. This can be achieved by adopting two conceptually different
Implications for modelling
As with any other application of modelling, the treatment of multiple plausible futures is influenced by purpose and context and should be fit for purpose (Black et al., 2014, Jakeman et al., 2006), in this case particularly the needs of policy or planning (Van Delden et al., 2011, Walker, 2000, Walker and Haasnoot, 2011). There are some common model requirements, but there is also significant variation. A first common requirement is that most analyses make use of all three paradigms: use of
Summary and concluding remarks
The need to deal with an uncertain future as a result of changes in climate, technology, socio-economic conditions and politics has led to the realisation that traditional methods of dealing with uncertainty that are based on probability distributions surrounding a “best-guess” of the future are unlikely to be appropriate. This has precipitated the development of a number of concepts that consider multiple plausible futures, such as deep uncertainty, global uncertainty and VUCA, highlighting
Acknowledgements
The authors from the University of Adelaide gratefully acknowledge financial support from the Bushfire and Natural Hazards Cooperative Research Centre and an Australian Postgraduate Research Award. The authors would also like to thank the four anonymous reviewers of this paper, whose comments have contributed towards improving the quality of this paper significantly.
References (132)
- et al.
Best practices for conceptual modelling in environmental planning and management
Environ. Model. Softw.
(2016) - et al.
Future research challenges for incorporation of uncertainty in environmental and ecological decision-making
Ecol. Model.
(2008) - et al.
Selecting model scenarios of real hydrodynamic forcings on mesotidal and macrotidal estuaries influenced by river discharges using K-means clustering
Environ. Model. Softw.
(2015) - et al.
Scenario driven optimal sequencing under deep uncertainty
Environ. Model. Softw.
(2015) - et al.
Robust optimization – a comprehensive survey
Comput. Methods Appl. Mech. Eng.
(2007) - et al.
Towards best practice implementation and application of models for analysis of water resources management scenarios
Environ. Model. Softw.
(2014) - et al.
Scenario types and techniques: towards a user's guide
Futures
(2006) - et al.
The origins and evolution of scenario techniques in long range business planning
Futures
(2005) - et al.
Thinking inside the box: a participatory, computer-assisted approach to scenario discovery
Technol. Forecast. Soc. Change
(2010) - et al.
Assessing the robustness of adaptation decisions to climate change uncertainties: a case study on water resources management in the East of England
Glob. Environ. Change
(2007)
A survey of decision-making approaches for climate change adaptation: are robust methods the way forward?
Ecol. Econ.
Modeling uncertainty in multi-criteria decision analysis
Eur. J. Operational Res.
Long-term energy scenarios: bridging the gap between socio-economic storylines and energy modeling
Technol. Forecast. Soc. Change
Dealing with uncertainty in ecosystem model scenarios: application of the single-model ensemble approach
Environ. Model. Softw.
Scenario modelling to support industry strategic planning and decision making
Environ. Model. Softw.
A new analytic method for finding policy-relevant scenarios
Glob. Environ. Change
Prediction under uncertainty as a boundary problem: a general formulation using Iterative Closed Question Modelling
Environ. Model. Softw.
A history of futures: a review of scenario use in water policy studies in the Netherlands
Environ. Sci. Policy
Dynamic adaptive policy pathways: a method for crafting robust decisions for a deeply uncertain world
Glob. Environ. Change
Fit for purpose? Building and evaluating a fast, integrated model for exploring water policy pathways
Environ. Model. Softw.
Lessons for model use in transition research: a survey and comparison with other research areas
Environ. Innov. Soc. Transit.
An exploratory approach for adaptive policymaking by using multi-objective robust optimization
Simul. Model. Pract. Theory
Ten iterative steps in development and evaluation of environmental models
Environ. Model. Softw.
Many objective robust decision making for complex environmental systems undergoing change
Environ. Model. Softw.
Multi-scale narratives from an IA perspective: Part II
Particip. local Scenar. Dev. Futur.
Multi-scale narratives from an IA perspective: Part I. European and Mediterranean scenario development
Futures
Combining participative backcasting and exploratory scenario development: experiences from the SCENES project
Technol. Forecast. Soc. Change
Improving scenario discovery for handling heterogeneous uncertainties and multinomial classified outcomes
Environ. Model. Softw.
Scenario-based robust optimization of a water supply system under risk of facility failure
Environ. Model. Softw.
Identifying and evaluating robust adaptive policy responses to climate change for water management agencies in the American west
Technol. Forecast. Soc. Change
A formal framework for scenario development in support of environmental decision-making
Environ. Model. Softw.
Robust decision making and info-gap decision theory for water resource system planning
J. Hydrol.
Robust optimization to secure urban bulk water supply against extreme drought and uncertain climate change
Environ. Model. Softw.
An updated scenario typology
Futures
The future shocks: on discontinuity and scenario development
Technol. Forecast. Soc. Change
Evaluating simulation-derived scenarios for effective decision support
Technol. Forecast. Soc. Change
Integrated framework for assessing urban water supply security of systems with non-traditional sources under climate change
Environ. Model. Softw.
Rethinking the 2×2 scenario method: grid or frames?
Technol. Forecast. Soc. Change
Uncertainty in the environmental modelling process - a framework and guidance
Environ. Model. Softw.
Visions for a sustainable Europe
Futures
Exploratory modeling for policy analysis
Operations Res.
Computer-assisted reasoning
Comput. Sci. Eng.
Exploratory modeling and analysis
Adaptive, multiobjective optimal sequencing approach for urban water supply augmentation under deep uncertainty
Water Resour. Res.
Info-gap Decision Theory: Decisions under Severe Uncertainty
What VUCA Really Means for You
Introduction to Stochastic Programming
Real-time deployment of artificial neural network forecasting models: understanding the range of applicability
Water Resour. Res.
Knowledge, uncertainty and physical geography: towards the development of methodologies for questioning belief
Trans. Inst. Br. Geogr.
Decision scaling: linking bottom-up vulnerability analysis with climate projections in the water sector
Water Resour. Res.
Cited by (320)
Optimizing Economic Performance of Rangeland Livestock Grazing Under Price and Climate Stressors
2024, Rangeland Ecology and ManagementDeveloping expert scientific consensus on the environmental and societal effects of marine artificial structures prior to decommissioning
2024, Journal of Environmental ManagementCollaborative modelling for goal-oriented scenario planning: A resilience planning case study in the context of greater Sydney
2024, International Journal of Disaster Risk ReductionImpact of land use changes on uncertainty in ecosystem services under different future scenarios: A case study of Zhang-Cheng area, China
2024, Journal of Cleaner ProductionEvaluation of emission reduction and other societal and environmental outcomes: Structured decision making for the Louisiana climate action plan
2023, Journal of Environmental Management