Eighty-five percent of the world’s population lives in areas impacted by human-induced climate change, according to an international team of researchers.
They used a new machine learning approach to identify more than 100,000 scientific studies on the effects of climate change across every continent. This massive literature review created a global map of impacts, which the team then compared to changing trends of surface temperature and rain caused by humans.
The take away? 80% of the planet’s land surface – containing 85% of the global population – is exposed to climate change impacts that can be directly attributed to humans.
In the age of big data, using AI is an important tool for climate scientists, the researchers say.
While it can’t substitute for expert assessments like the Intergovernmental Panel on Climate Change (IPPC), using machine learning to sort through climate studies is invaluable to helping map evidence in a systematic way.
In fact, since the IPCC’s fifth major assessment report was published in 2014, more than 46,000 new papers have been written – twice as many papers as the number published between the fourth and fifth report.
While systematic reviews of climate studies can help comprehensively describe the bigger picture of changes happening to our planet, they’re difficult to conduct.
“Their scope is often confined to very specific questions, covering no more than dozens to hundreds of studies,” the authors explain.
Instead, the team used a deep learning language model called BERT (Bidirectional Encoder Representations from Transformers) to trawl through two large databases – Web of Science and Scopus – and systematically identify literature on climate change. BERT found more than 600,000 papers, and the authors estimate that 100,000 of these are relevant to understanding the observed impacts of climate change.
BERT also revealed trends in the research published.
“Although the number of relevant studies in North America, Asia and Europe is much higher than in South America, Africa and Oceania, there is a large body of relevant studies available on all continents,” they note.
BERT also found that the majority of studies (34,794) documented impacts on terrestrial and freshwater ecosystems, while mountains, snow and ice had the smallest coverage, with only 6306 studies.
But this is still a huge number of studies for a person conducting a literature review to comb through.
“The use of machine learning means we consider more evidence than would otherwise be feasible, showing where evidence appears to be more prevalent and where important gaps can be observed,” the team notes.
The researchers hope that machine learning techniques can help create an automated, “living” map of climate impacts that can be readily updated as new research rolls in.
Lauren Fuge is a science journalist at Cosmos. She holds a BSc in physics from the University of Adelaide and a BA in English and creative writing from Flinders University.
Read science facts, not fiction...
There’s never been a more important time to explain the facts, cherish evidence-based knowledge and to showcase the latest scientific, technological and engineering breakthroughs. Cosmos is published by The Royal Institution of Australia, a charity dedicated to connecting people with the world of science. Financial contributions, however big or small, help us provide access to trusted science information at a time when the world needs it most. Please support us by making a donation or purchasing a subscription today.