Advancing Transparency and Accessibility: Implementing FAIR Data Principles in IPCC AR6 WGI Report
Bibliographic record
Abstract
The Sixth Assessment Report (AR6) marked the first time that the Intergovernmental Panel on Climate Change (IPCC) recommended and implemented FAIR data principles as part of the assessment process. While the critical importance and utility of FAIR principles are widely acknowledged, their implementation is not straightforward, especially in this unique context that involves the collaboration of hundreds of scientists around the world from different disciplines, utilising diverse information sources. We describe challenges and lessons learned when sharing data and other digital resources in the context of assessing the physical science basis (Working Group I, WGI). With the primary scope of ensuring transparency and reproducibility, and providing due credit to the data creators who collaborated on the report, authors were guided and supported to encourage the public availability of the data and the associated code used in post-processing. As part of this initiative, data and code for over 200 figures have been made accessible as well as all plotted data for the Summary for Policymakers. Additionally, the assessment foundation datasets, such as climate model simulations, have been curated, along with a novel category of data — datasets constrained through expert assessment (e.g., model-based projections of global surface temperature). Moreover, an innovative digital product was produced to support and expand the assessment done in the WGI AR6, building on these datasets and synthesising key findings for Climatic Impact Drivers: the Interactive Atlas. Its formal inclusion as part of the report was possible thanks to the implementation of FAIR principles, fully compliant to best data practices. Despite successes, challenges emerged with the novel FAIR implementation because of learning by doing and real-time development, managing diverse data, requiring the introduction of new roles and workflows. For authors, adapting dataset structures to fit repository constraints and addressing metadata requirements posed difficulties, as they didn't always align with actual dataset usage or user needs. Simultaneously, repositories faced challenges adapting dataset structures to their systems, ensuring adherence to standard conventions, managing references, reviewing licences, and awaiting author feedback. In this complex landscape, the role of data curators was crucial, serving as a facilitating bridge for information and requirements exchange, providing essential support to authors and collaborating with repositories to seek solutions that effectively integrated the needs of both parties. Flexibility and simplicity proved key allies in overcoming challenges. Prioritising clarity and acknowledging the limitations of a rigid structure enabled smooth navigation of obstacles, resulting in practical and useful outcomes. The new IPCC cycle starts now. FAIR principles need to be fully integrated into the climate assessment process to adhere to the highest standards in data access and stewardship. Our recommendations include the need to integrate data management workflows and engaging authors from the outset, highlighting the significance of data-related tasks for transparency, and enhancing the tools available to support authors. Providing authors clear instructions and timelines is crucial, along with technical assistance from data science experts. Actively endorsement and support of these initiatives by the IPCC leadership is vital for their effective integration into the assessment process.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.359 | 0.479 |
| Meta-epidemiology (narrow) | 0.001 | 0.002 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.008 | 0.010 |
| Science and technology studies | 0.005 | 0.013 |
| Scholarly communication | 0.034 | 0.027 |
| Open science | 0.008 | 0.024 |
| Research integrity | 0.007 | 0.011 |
| Insufficient payload (model declined to judge) | 0.007 | 0.003 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".