Scientists warn against misuse of climate models in financial markets

LONDON, Feb.8 (Reuters) – Misuse of climate models could present an increasing risk to financial markets by giving investors a false sense of certainty about how the physical impacts of climate change will play out, according to the authors of an article published on Monday.

With heatwaves, wildfires, massive storms and sea level rise set to intensify as the planet warms, companies are under increasing pressure to disclose how the disruption could affect their activities.

But the authors of a peer-reviewed article here in Nature Climate Change warned that the drive to incorporate global warming into financial decision-making has surged ahead of models used to simulate climate by “at least a decade”.

“In the same way that a Formula 1 Grand Prix car is not what you would use to go to the supermarket, climate models were never developed to provide accurate information on financial risk,” said Andy Pitman, climatologist at the University of New South Wales and co-author of the article.

Misuse of climate models could lead to unintended consequences, such as “greenwashing” some investments by minimizing risk, or affecting the ability of companies to take on debt by inflating others, the authors said.

The problem is that existing climate models have been developed to predict temperature changes over decades, on a global or continental scale, while investors typically need location-specific analysis over much shorter time frames. .

Climate models are also not designed to simulate extreme weather events, such as storms, which can lead to sudden financial losses.

To bridge the gap, the authors called for the development of new forms of climate projection to support the financial sector, supported by trained “climate translators” to help regulators, investors and businesses make better use of science.

“Businesses like to use models because the numbers give them a sense of security,” said Tanya Fiedler, a lecturer at the University of Sydney and lead author of the article. “That doesn’t necessarily mean the numbers are reliable.” (Reporting by Matthew Green; Editing by Hugh Lawson)