Decision-making about climate change adaptation

Abstract

Decision-making about and for climate change adaptation has fascinated the field of adaptation science for decades. We now have a solid set of decision-support tools, frameworks and approaches available to support decision-makers. Yet, in this piece, I argue that the black box of decision-making requires unravelling for us to better understand what heuristics (rule of thumb) are used about climate adaptation and how these impact the very decisions made. I also point out areas in need of more research including a new focus on decision quality, artificial intelligence and data, and the role of imagination in driving climate adaptation decision-making.

Introduction

How we make decisions on climate change adaptation has long fascinated the field. The Intergovernmental Panel on Climate Change (IPCC) dedicated chapters on this topic in both the Fifth Assessment (Jones et al., 2014) and the Sixth Assessment (New et al., 2022). The Sixth assessment also looked at different adaptation options and their feasibility, effectiveness and linkages to sustainable development goals. Common methods and approaches include dynamic adaptation policy pathways, scenario planning, robust decision-making, cost–benefit analysis, cost-effectiveness analysis, probability decision trees and multi-criteria decision analysis (Dittrich et al., 2016; Siders and Pearce, 2021). Approaches such as adaptation pathways focus on the trade-offs between different adaptation decisions and how to weigh these, when and why (Haasnoot et al., 2024). In other words, the field is certainly not ignorant about the decision-support tools, methods, options and approaches available.

But decision-making processes are shaped by ‘invisible’ factors, such as heuristics/rules of thumb, that have a huge impact on how the issue is framed in the first place. So, instead of focusing on the well-documented methods for decision-making in climate adaptation that others have covered eloquently (e.g. New et al., 2022; Siders and Pearce, 2021), I unpack in this Commentary the hidden factors that influence decision-making and the overall ways we conceptualise adaptation. This focus on the foundational underpinnings of decision-making and adaptation will hopefully raise new questions and generate lively debate in Sage Climate Change Dialogues going forward.

Doing away with the ‘black box’ of decision-making

Decision-making in adaptation has often been aligned with the concept of barriers: all the factors that impede and limit adaptation decisions and measures. The mainstream ‘linear and functionalistic view of decision-making’ states that once these barriers are lifted (e.g. lack of information), then decisions and implementation follow (Eisenack et al., 2014). Biesbroek et al. (2015) argue, however, that such a view distils complex dynamic decision-making processes into narrow simplistic conceptualisations that do not reflect the true nature of decision-making. In fact, decision-makers are faced with dynamic contexts with rules, norms and values under which they operate that have a bigger impact on how those decisions are made than just getting the ‘right’ information (Gorddard et al., 2016).

I argue that one way of doing away with the ‘black box’ in decision-making on adaptation is unravelling it: increasing the transparency of the box by articulating the assumptions made, and how these influence the outcomes. This is notoriously difficult to do but a case where the concept of adaptation heuristics can help. Heuristics are rules of thumb that help us to make faster decisions and offer guidance in conceptualising an issue (Hey, 2016). Each field of science (Hey, 2016) and practice (Crandall et al., 2006) has developed its own heuristics over time that guide the field, for example, acceptable methods, key concepts, and definitions. Heuristics set constraints in what we think is possible that in turn influence what decisions are even on the table. Research in finance has shown, for example, that clients’ core beliefs about money (a money script = how someone thinks about money) have a significant influence on financial outcomes and behaviour (Klontz and Britt, 2012).

While several cognitive heuristics and biases exist in the ‘formal’ decision-making processes themselves (Artinger et al., 2015), climate adaptation has its own array of heuristics that both enable and constrain the way adaptation decisions are made. For example, one of the longest-held heuristics is on the ‘localness’ of adaptation: ‘the accepted wisdom that adaptation should be local and context specific’ (Olazabal et al., 2024; emphasis added). Olazabal et al. (2024) warn us against the fallacy of thinking adaptation can be generalised and that it is this ‘generic’ nature of climate adaptation that is detrimental to our thinking.

Yet, using the ‘local’ heuristic has direct consequences on how decisions on adaptation are made: while this heuristic can refocus attention on local needs and specific contexts (Olazabal et al., 2024; Preston et al., 2015), it can push responsibility for adaptation, both for financing and implementation, to local governments/communities even when these are the most under-resourced and constrained by other levels (Nalau et al., 2015). Adaptation heuristics can direct decisions on the scale of adaptation considered and subsequently to whom responsibility is assigned even when this might not be the soundest choice. How heuristics play out in decision-making is increasingly an area of research we need to understand much better.

Adaptation decisions: ‘and/yes’

Yet, claiming adaptation to be ‘only local’, ‘only national’ or ‘only global’ demonstrates also our inherent incapability of understanding adaptation as ‘and/yes’ versus ‘either/or’ (Heath and Heath, 2013). In cognitive terms, it is much easier to frame decisions as ‘either/or’: choosing one and not the other. This is where we need more constructive dialogue as to why adaptation should be conceptualised one way or the other, or rather: why cannot it be both? Taking a broader perspective on adaptation (‘it can be this and that’), can lead to more nuanced conversations as to why we choose to frame adaptation at a particular scale, speed or depth, and how these drive decision-making. This can help us also reflect what decisions could be done at what scale and why.

Each heuristic or ‘frame’ carries with it a diverse array of assumptions that will remain implicit unless articulated (Nalau et al., 2021). Practitioners in Queensland, Australia, participating in a reflective adaptation heuristic workshop, for example, were glad to have a space to use their second-order thinking (Thuraisingham and Lehmacher, 2013) to articulate the many assumptions about adaptation that are made (Nalau et al., 2021). Similarly, Singh et al. (2022) offer key observations how a variety of ‘frames’ exist on the effectiveness of climate adaptation that lead us to make decisions and evaluate outcomes in drastically different ways, depending how we decide to think about adaptation in the first place (Singh et al., 2022).

This is also the case when we talk about ‘transformation adaptation’ (TA) which is now also articulated as part of the Global Goal on Adaptation: we need to be clear about what our assumptions are, what scale we think they can have the most impact, and what makes TA different from other adaptation approaches. Much of the theoretical work can be helpful here: for example, Termeer et al. (2024) propose three specific key dimensions of transformative change (in-depth, system-wide and quick) with clearly articulated assumptions behind each. We need scholars to be leading on this intellectual front generating key heuristics that are sound while also stretching our cognitive abilities to grasp other conceptualisations or heuristics about adaptation.

Another area for research is the ‘why’ of adaptation decision-making: the assumptions people make about adaptation's potential to result in particular outcomes (Carr and Nalau, 2023). While many adaptation projects and programs aim to ‘reduce vulnerability’ and ‘increase adaptive capacity’, the rationale behind these outcomes (e.g. how to decide that action x is the most effective means to reduce vulnerability) is often unclear and/or hidden. Greater transparency on the rationale behind adaptation goals would be helpful to start creating an evidence base for climate adaptation benefits: the extent that adaptation actions result in positive outcomes and co-benefits, and how these decisions were made in the first place and why.

Focus on understanding decision process and decision quality

More research overall should also focus on understanding decision quality in adaptation. People commonly use ‘resulting’ as a cognitive shortcut where we equate a bad outcome to a bad decision. Yet, Duke (2020) challenges us to think more critically about the relationship between a decision process and an outcome. This, however, requires curiosity to investigate the process of making decisions rather than merely the outcome. Maladaptation literature, for example, often focuses on bad outcomes to demonstrate the pitfalls of adaptation rather than investigating how those decisions were made. Many factors that influence outcomes are far beyond the control of one project or program: even with a bad outcome, the decision itself could have been made with the best information at that time (Duke, 2020).

Focusing on improving ‘decision quality’ (Duke, 2020) therefore should be at the forefront of adaptation science. Questions include what does a good quality decision process look like for climate adaptation? What can we learn from the way decisions are made now, with what information, and to what extent do we revisit these decisions? What would a robust decision tool to learn from climate adaptation decisions look like? Useful approaches here can be building on decision trees and using decision diaries to track what information was available at the time of making the decision, what assumptions were made (including heuristics used), and what outcomes resulted and why (Duke, 2020). These questions are increasingly relevant given the recent shifts towards different types of ‘adaptation investing’, especially in nature-based and nature-positive solutions.

Artificial intelligence and data for decision-making

Another question is where we get the information and data that assist in making climate adaptation decisions. With the advances in artificial intelligence (AI), more data and tools are becoming available for climate adaptation decision-making. From AI, Augmented Reality/Virtual Reality, Internet of Things, Earth Observation and Drones, an increasing number of technologies are being deployed. AI systems such as ChatGPT can quickly provide guidance on ‘vulnerability metrics’ or ‘climate adaptation indicators’ albeit at a generic level. It can identify adaptation strategies regarding land use and even conservation (Bassetti, 2023). Large language models can be used to support scientific assessments, such as the IPCC, as they can generate quickly broad trends in the literature (Muccione et al., 2024; Nalau et al., 2024). Earth System Observations are increasingly sophisticated in providing data for decision-making and could be gamechangers in tracking and understanding climate adaptation implementation progress across different regions and scales.

We need, however, more debate and conversations around the role of AI in how we make decisions. Is there a way to provide adaptation guidance via AI that matches more of a tutor rather than a system that spits out conventional wisdom that might, or might not be, correct? Who would decide what information on climate adaptation the algorithms should be based on, and how would we check if the ‘guidance’ or information is indeed helpful? Would an advanced AI tutor focused on climate adaptation decision-making increase the potential for ‘good’ decision-making and make world-class adaptation information easily understood and accessible for a greater number of decision-makers? And who would ultimately determine what ‘good’ decisions for climate adaptation look like?

Role of imagination in adaptation

But perhaps the most unexplored space in adaptation decision-making is the role of imagination and how we use it in the process of crafting future visions (Comelli et al., 2024). Again, future visioning and scenario planning exercises are not value-free but a negotiation of multiple imaginary spaces we each hold to determine what is possible, what decisions need to be taken and how. So far we have not explored this area to its full potential even if future visioning/scenario planning has been a stable method in adaptation science (Nalau and Cobb, 2022). Projects such as Tomorrow's Cities (Pelling et al., 2024) and Imagine Adaptation (Olazabal et al., 2024) and initiatives such as United Nations Framework Convention on Climate Change's Resilience Frontiers are providing new insights into how to drive collective imaginations of climate adaptation and resilience. And many of the common decision-support tools and approaches require us to imagine what future could hold. Questions in this area include: What are the world changing actions that can accelerate adaptation? How do we reconcile and coexist with our different visions of the future? How do we create new spaces for shared imagination of a world that is resilient and well-adapting? How do these visions translate into tangible practical decisions?