UC Berkeley

As societies across the world increasingly experience the devastating impacts of climate change, there is an urgent need to implement and accelerate adaptation actions. Effective adaptation decisions need to be grounded in, and supported by robust science on expected climate change and its impacts. Despite several advances in climate science and modeling, the use of science in adaptation decisions is still very limited. This suggests that there is a gap between the production of climate science, and its use in adaptation i.e. there exists a climate science-usability gap for adaptation. My dissertation deconstructs the various aspects of this gap, by identifying the types of scientific knowledge and the processes of knowledge production that can lead to increased uptake of climate information in adaptation decisions. The overall aim of my research is to bridge the usability gap, by providing scientists with a better understanding of decision-makers’ climate information needs, providing decision-makers with an improved understanding of how science may be useful in their adaptation contexts, and recommending broader systemic changes that can sustain the development of usable science. Finally, this dissertation argues for and promotes ‘engaged’ models of research, where scientists and decision-makers jointly develop climate science that achieves more benefit to society.

Chapter 1 of this dissertation reviews on-the-ground adaptation projects from across the world to understand how adaptation is conceptualized and promoted by the international community, and the types of scientific information that are used in the planning and design of these projects. Using the case of ecosystems-based adaptation, we find that 65% of adaptation projects either did not use any information on expected climate change, or just used broad macro-scale climate projections that were not specific to local sectoral contexts. A majority of projects did not address uncertainty in future climate change or in adaptation benefits, nor did they track adaptation outcomes. This pervasive lack of use of climate science to inform adaptation actions is concerning, as it is difficult to ascertain the level of climate impacts that these adaptation projects can cope with.

Chapters 2.1 and 2.2 identify decision-makers’ climate information needs for undertaking adaptation action. Through semi-structured, exploratory interviews with perennial tree crop growers in California, Chapter 2.1 finds that farmers perceive long-term climate projections as an extension of weather forecasts, which can often lead to their skepticism of the utility of such information. Hence, the manner in which long-term projections are framed, presented, and discussed with farmers is critical to how they may perceive the potential utility of such information. We find that in-depth iterative conversations were essential to effectively understand the types of climate projections that are useful to farmers, and recognize that projections of crop-specific agro-climatic metrics are often more useful to farmers than projections of physical climatic metrics. Such conversations help in joint construction of meaning, or in this case joint understanding of useful climate information, where the farmer and the interviewer both improvise and iterate to find the best types of projections that fit specific decision-contexts.

Chapter 2.2 presents a case of co-production (Project Hyperion) wherein scientists and water managers jointly developed decision-relevant metrics for adaptive water management. We find that arriving at these actionable metrics is more complicated and iterative than is generally acknowledged in the literature. We identify engagement strategies that target both direct and indirect knowledge elicitation as effective approaches for translating managers’ needs into quantitative metrics. These strategies, along with the list of metrics we develop, provide tangible recommendations to both researchers and practitioners seeking to develop usable climate information.

Chapter 3 evaluates the skill of different Global Circulation Models (GCMs) in predicting decision-relevant climatic metrics (using the case of chill hours in California), and examines how differences in model choice may impact future projections. We find that the multi-model mean of GCMs is not the best predictor of this specific metric. Additionally, downscaled LOCA projections, which is a dataset recommended by the State of California, systematically underestimate the negative trend observed in historical chill hours. Further, we also find that good skill in predicting broader physical climate metrics does not guarantee skill in prediction of specific decision-relevant metrics such as chill hours. Since many decision-relevant metrics are non-linear derivations of primary physical quantities, approaching model evaluations through the lens of decision-relevant metrics can provide critical insights on model choice for adaptation decisions.

Finally, Chapter 4 reviews recent co-produced climate change adaptation projects alongside the theoretical scholarship about co-production – to compare the expected theoretical outcomes of co-production with those that are reported in practice. In bringing these two streams of thought and action together, we find that the practice of knowledge co-production is realizing improvements in knowledge utilization, but is not reporting on the more radical or transformational outcomes that the theory expects from co-production (such as changing power dynamics, transforming dominant knowledge paradigms and bringing institutional change). We identify five key reasons for why co-production practice may be falling short of its expectations. We propose that to address these issues and unleash the full potential of co-production, a more transparent, conversant, and interactive research agenda and discourse is required.