Assessing the climate change literature has become a challange of "big literature". In this presentation we use data science applications to understand, categorize and map out a landscape of more than 500,000 publications on climate change in the Web of Science and Scopus. We further use this rich dataset to re-assess a variety of biases in assessment reports by the Intergovernmental Panel of Climate Change (IPCC) that have been observed by scholars. We find that numerous claims in the literature no longer hold when adequately assessed in the context of the entire available science. We conclude on the arising changes in policy implications and outline how machine learning can help to revolutionize the mapping and synthesis of vast literatures as commonly done in IPCC assessments.

Download the slides for this talk.Download ( PDF, 6003.05 MB)