Currently showing: Climate/natural disasters > Disaster risk

05 Jun 14 15:22

The start of the 2014 hurricane season coincided with the release of a study from the University of Illinois which states that historically, storms christened with female names caused more fatalities than storms with male names. One of the underlying reasons for this is because names produce a certain perception, and the perception is that daintily named hurricanes can't be too problematic. Their namesakes typically aren’t. 

In the interest of full disclosure, I haven't read the article, just the media coverage and the authors' responses to the media coverage, all of which intrigued me. While it seems as though the authors tried to eliminate certain variables which would skew their results (removing Katrina and Audrey from their data set since they were outliers), this historical analysis, by definition, cannot incorporate the multiple aspects of hurricane forecasting and communication which have improved in recent decades, leading to fewer fatalities. Today we have real-time social media updates connecting community members, satellite imaging of the storm and its projected track and round-the-clock media attention drawing attention to an event, as opposed to 40 years ago when perhaps radio and the 6 o'clock news were our sources of information.

To support their findings in the historical data, the authors also asked volunteers to react to a variety of questions and situations. One experiment asked them to predict hurricane intensity, without any other information but the name. The name "Omar," was predicted to be the most intense, while the name "Dolly," was predicted to be the least intense. Another experiment supplied the volunteers with two identical scenarios: A hurricane is threatening the coast and a voluntary evacuation order is in place. The only difference? One storm was named Danny, and the other was named Kate. More volunteers said they'd evacuate for Danny than Kate.

With the study run out of the University of Illinois, most of the volunteers were from a landlocked state, where the choice to evacuate is a purely hypothetical situation. Would volunteers from Texas, Louisiana, Florida, North Carolina, New Jersey, New York and Massachusetts, for example, respond differently? I venture they might, since the decision to evacuate has been a reality many in those states have faced. What would you do if a Category 4 hurricane was approaching your hometown? Would you be less concerned if it was named Belle instead of Barry?

Category: Climate/natural disasters: Disaster risk, Floods/storms, Resilience

Location: United States


Alicia Montoya - 12 Jun 2014, 5:55 a.m.

In his latest column, NYT's Nicolas Kristof highlights: "Researchers find that female-named hurricanes kill about twice as many people as similar male-named hurricanes because some people underestimate them."

However, Jeff Lazo from the National Centre for Atmospheric Research (NCAR) points out that, while we clearly respond to gender, the sample is too small and the historic evidence isn't as strong as it originally seemed: For a start, the researchers analysed hurricane data from 1950, but hurricanes all had female names at first. They only started getting male names on alternate years in 1979. This matters because hurricanes have also, on average, been getting less deadly over time. “It could be that more people die in female-named hurricanes, simply because more people died in hurricanes on average before they started getting male names,” says Lazo.

Personally, I don't care what a storm's called, I respond to images like this: and to advice from reputable scientists like those at NOAA. You?

Gavin Montgomery - 12 Jun 2014, 3:26 p.m.

While it is true that new communications technology is making it easier for us marshal a response to big storms and that recent hurricanes have been far less deadly than in previous years, there are areas for concern.

These include the continued degradation of natural storm barriers like mangroves, decades of under-investment in key infrastructure and the fractured regulatory landscape in the U.S., with blurred lines between local, state and federal governments all leap to mind. This manifested most recently during Sandy, which shut down one of the most important cities on our planet for days, revealing series planning deficiencies.

This might be the year that this lack of preparedness comes back to bite us. As this report shows - - there is a strong correlation between El Nino events and large Atlantic storms.

We have been extremely fortunate that there has not been a strong El Nino since 1998 - - but there is some evidence to point to this being a "Super El-Nino" year, an event we haven't seen since 1997.

How well are we really prepared for such a super-storm? Or a season of big storms?

Megan Linkin - 12 Jun 2014, 6:10 p.m.

Hi Gavin, La Nina events show the stronger correlation with large Atlantic storms. Table 2 of the Landsea paper demonstrates that; the mean loss during La Nina years is USD 5.8 bio (1997 USD), while the mean loss during El Nino years is USD 2.1 bio (1997 USD). Strong El Nino years, such as 1982 and 1997, corresponded to lower activity in the Atlantic.

It's important to be prepared for a hurricane in any year, because El Nino years do not equate to no losses, and there have been significant losses in El Nino years. Also, it only takes a single storm to make it a memorable year. However, it's also important to accurately communicate the potential impact of all these variations in the climate system on Atlantic hurricane activity.

If you would like to leave a comment, please, log in.