Searching for evidence of climate change in natural disasters

A key question in the debate around the economic impacts of climate change might have received a tentative answer last week by researchers studying economic losses from natural disasters.

Academics have sought for years to detect a "climate signal" in the tangled aftermath of hurricanes, floods, tornadoes and other perils. No one disputes that the number of catastrophes has risen fast, and that damage is skyrocketing. But most of that ruin is caused by humans: We are constructing new homes, condominiums, hotels and roads along dangerous coastlines.

Thus, we're encouraging more people to move into harm's way.

But the climate change signal hasn't been found in the piles of debris. It would be revealed by assembling economic losses from disasters from the past to the present, and then by adjusting those damages to neutralize the impacts of rising property values driven by population increases and other things.

Now, two researchers believe they might be hearing the first ping of that signal.


An unpublished paper released by the London School of Economics and Political Science last week found that insured damages in the United States rose 5 percent between 1973 and 2008, after adjusting for, or "normalizing," the rise in wealth in disaster-prone areas. German data revealed similar upticks.

That means climate change could be the remaining reason for climbing property losses related to hailstorms, lightning, floods and storm surges. Data from insured hurricane damage showed a 10 percent rise. The statistics remained true when the researchers removed the effects of Hurricane Katrina in 2005.

"Climate change could be one of the plausible explanations," said Eric Neumayer, a co-author of the paper and an environmental economist at the London School of Economics. "But we're very careful in the paper to say it's just one possibility. There are other possibilities behind this that for the time being we cannot sort of rule out."

Discovered: 'tentative evidence'

More research is needed, he said, to ascertain if other influences are driving the increased losses. Possibilities include changes in the way insurance companies calculate claims payments. Maybe they made it easier to receive cash. Fewer losses might also have been reported in early years of the data, for example, skewing the findings.

Other burps in the data, like the researchers' inability to account for climate resiliency efforts, could be confusing the signal.

"Our findings are interesting, but before any firm conclusions can be drawn from them, more research is needed to analyze which of these potential explanatory factors, of which anthropogenic climate change is but one possibility, or which combination of factors drive the observed upward trends," says the paper, which is awaiting peer review.

"With these caveats in mind, our findings only provide tentative evidence that anthropogenic climate change may possibly already have triggered more frequent and/or more intensive relevant natural disasters affecting Germany as well as the U.S. and thus, ironically, the biggest emitter of greenhouse gas emissions in the world."

If the results lead to a clear climate signal in catastrophe losses, they could influence one of the largest industries in the world. Many insurance companies are grappling with the effects of climate change and how it might affect their policies and their investments in electric utilities and other buffeted industries that reach into the billions. They're also hunting for emerging opportunities in the renewable energy field and elsewhere.

Several large European reinsurers are already bracing for climatic impacts and warning of the potential threats. Swiss Re, for example, issued a report last year warning that energy companies and other emitters could face lawsuits blaming them for global warming. The insurer cautioned that the litigation could rival asbestos liability in cost.

Some big insurers are already believers

Munich Re, meanwhile, hopes to get the attention of world leaders as international negotiations to reduce greenhouse gases begin this week in Cancun, Mexico. The company, which provided funding for the research covered in this article, confidently links catastrophe losses to climate change.

Flooding worldwide has produced a threefold increase in damages since 1980, the company says, and the frequency of windstorms, particularly Atlantic hurricanes, has doubled during that same period.

"This rise cannot be explained without global warming," Munich Re said in a statement directed to negotiators earlier this month.

The Obama administration released a report by the Global Change Research Program in June 2009 saying climatic impacts are already being felt. They include increases in downpours, rising sea levels, retreating glaciers, thawing permafrost and longer growing seasons.

It also suggests that climate change is already increasing insured losses, saying that population shifts and economic growth "do not fully explain the upward trend in costs or numbers of events."

"What is known with far greater certainty is that future increases in losses will be attributable to climate change as it increases the frequency and intensity of many types of extreme weather, such as severe thunderstorms and heat waves," the report adds.

Robert Hartwig, president and chief economist of the Insurance Information Institute, an industry group that closely follows loss trends, seems willing to accept the London study's findings. But he notes that if 5 percent of loss increases is caused by climate change, the other 95 percent is attributable to other impacts, like demographics.

"Where the value lies in this kind of research is if the climate is to become more volatile for anthropogenic reasons, then maybe that 5 percent in 20 or 25 years' time, maybe it'll be 10 percent by then," he said. "Maybe in 50 years' time, it'll be 50 percent. The point there is that if this technique can be validated, it might possible to extrapolate out to the future."

One researcher sees no signal

Roger Pielke Jr., a professor of environmental studies at the University of Colorado, Boulder, and an architect of the "normalization" method used to look for climate signals in natural disaster losses, doesn't believe the London study has uncovered evidence of global warming effects.

"People who assert a link to particular disasters and loss are just opening the door to legitimate criticism of their scientific claims," Pielke said. "The conclusion that I have reached is that you can't see that signal. It doesn't mean that there's no signal or that humans aren't affecting the climate. It just means that signal, if it exists, isn't large enough to be seen in the adjusted data."

He notes that the 1970s, the period during which the London researchers begin looking at U.S. data, is the lowest point in the century for disaster losses. Consequently, "If you start in 1973, you're definitely going to see an upward signal," he said.

Neumayer, the paper's co-author, wasn't surprised by Pielke's divergence. He expected the American researcher to be "suspicious" of the new results, but notes that the London study is using different data from a rich source: the insurance industry.

Yet he acknowledges that much more research will need to be done before anyone can say steadfastly that rising disaster losses are evidence of climate change.

"We're very careful not to jump to conclusions -- 'Oh, upward trends, hence climate change.' It would be silly to do that, because it's just one competing explanation," Neumayer said. "The work has only been begun."

Like what you see?

We thought you might.

Start a free trial now.

Get access to our comprehensive, daily coverage of energy and environmental politics and policy.



Latest Selected Headlines

More headlinesMore headlines

More headlinesMore headlines

More headlinesMore headlines

More headlinesMore headlines