The scientific journal Nature has retracted a high-profile article on the economic impacts of climate change after significant inaccuracies were discovered in economic data related to Uzbekistan. The study, originally published online on April 17, 2024, examined how rising global temperatures could affect economic output by mid-century.
According to the retraction notice, the dataset contained serious errors in Uzbekistan’s economic indicators from 1995 to 1999, which substantially altered the study’s findings.
The authors acknowledged that the results were highly sensitive to the inclusion of Uzbekistan. Upon review, they found the country’s reported economic figures during that period were inaccurate. They also noted additional complications, including inconsistencies caused by transitions between data sources, and the failure to fully account for long-term economic trends.
Another methodological issue, spatial auto-correlation, which affects uncertainty estimates, had not been incorporated into the original analysis. Once the authors corrected the dataset and adjusted their methodology, the study’s conclusions changed significantly.
Specifically, the projected range of climate-related economic damages by 2050 shifted from 11-29% to a broader 6-31%. The statistical likelihood that damages would differ significantly under various emissions scenarios by mid-century also fell, from 99% to 90%. Due to the scale of these changes, the authors concluded that a simple correction was insufficient and opted for a full retraction.
A revised version of the study, updated with corrected data and improved methodology, has been made publicly available but has not yet undergone peer review. The authors plan to resubmit the article, and Nature has stated it will update the retraction notice if the revised version is accepted for publication.
The researchers thanked the members of the scientific community who identified the discrepancies, and emphasized the importance of transparent corrections in climate science.
