Climate is changing. While worldwide societies experience climate-induced hazards, progress in climate change adaptation – both public and private – remains slow. Understanding who, when, and how pursues Climate Change Adaptation (CCA) is essential. Factors explaining adaptation by different actors (governments, communities, households, farmers, and individuals) may vary across countries and per type of adaptation (incremental vs. transformational). Worldwide rich qualitative research reveals factors influencing these different adaptation decisions. Thousands of articles report valuable insights, and the literature reporting mounting evidence constantly grows. To address this challenge, we use a novel approach to elicit patterns in decision factors for various actors and types of adaptation. Specifically, using natural language processing, thematic coding books, and network analysis, we consolidate empirical evidence fragmented across various textual sources. Here, we provide two exemplar datasets derived using this approach. Both databases provide systematic overviews of various adaptation factors associated (positively, negatively, or neutrally) with particular adaptation measures. The first dataset provides a systematic overview of factors associated with CCA to floods and sea-level rise by actor (see Dataset 1). The second set of data provides a systematic overview of factors associated with transformational vs. incremental adaptation by farmers (Dataset 2).
Both datasets rely on the textual information reported in peer-reviewed articles and derived using our algorithm Gil-Clavel & Filatova (2024). The coding book for adaptation factors per actor group is provided in Gil-Clavel et al. (2024), and the coding book for classifying CCA measures as transformational adaptation is described in Gil-Clavel et al. (2023). We go beyond the traditional automatic analysis of textual data focused on the word count and topic modeling. Instead, we strive to elicit meaningful associations between the type of adaptation and factors reported to influence those decisions. We do that based on the detailed code-books designed with domain experts and on the computational Natural Language Processing algorithms that elicit reported relationships and classify them as positive, negative, or neutral.
Related work:
Gil-Clavel, Sofia, and Tatiana Filatova. “Using Natural Language Processing and Networks to Automate Structured Literature Reviews: An Application to Farmers Climate Change Adaptation.” arXiv, July 3, 2024. http://arxiv.org/abs/2306.09737.
Gil-Clavel, Sofia, Thorid Wagenblast, Joos Akkerman, and Tatiana Filatova. “Patterns in Reported Adaptation Constraints: Insights from Peer-Reviewed Literature on Flood and Sea-Level Rise,” April 26, 2024. https://doi.org/10.31235/osf.io/3cqvn
Gil-Clavel, Sofia, Thorid Wagenblast, and Tatiana Filatova. “Incremental and Transformational Climate Change Adaptation Factors in Agriculture Worldwide: A Natural Language Processing Comparative Analysis,” November 24, 2023. https://doi.org/10.31235/osf.io/3dp5e
How to cite the datasets: Gil-Clavel, Sofia; Filatova, Tatiana, 2024, “Interrelated Climate Change Adaptation Measures and Factors”, https://doi.org/10.17026/SS/PYZCXK, DANS Data Station Social Sciences and Humanities
Funding This work was supported by the Netherlands Organization for Scientific Research NWO VIDI grant number 191015.
Python, 3.11
R, 4.3.3
