Do We Need Better Hurricane Ratings?

It’s no secret that the mass media relies on hype to stay in business, and disasters often provide ideal opportunities to broadcast on the edge of panic. For example, news outlets portrayed Hurricane Matthew as a superstorm destined to wreak havoc on the US coast, but instead the storm quickly weakened to Category 2 and peak wind gusts generally stayed below 100 mph after it made landfall. Still, Matthew destroyed 7,000 homes in North Carolina and killed 49 people in the US alone.

Disaster rating systems should provide a reasonable indication of potential damage, and Matthew’s mild ratings were something of an anomaly. As its name implies, the Saffir-Simpson hurricane wind scale (SSHWS) used to classify hurricanes takes only wind speed into account. Matthew’s damage was primarily caused by record flooding due to intense rainfall. Similarly, Hurricane Sandy registered as Category 1 but was the second-costliest hurricane the US has ever seen. These discrepancies beg the classic engineering question: is there a problem with the Saffir-Simpson scale, and if so can scientists design a better one?

For sure, wind speed is the most important metric relating to hurricanes. As shown in the video below, the difference between a Category 1 and a Category 2 storm can determine whether your trees lose their leaves or your house loses its shingles. But critics of the SSHWS say that rating storms without taking the storm’s sheer size and rainfall into account is potentially dangerous. Many favor developing an alternative rating system based on the Richter scale, which describes earthquake magnitude as a continuous logarithmic function. But while the Richter scale is much more accurate in its ability to describe earthquakes, the Saffir-Simpson scale’s neatly quantized categories are easier for the public to digest, even if they’re sometimes inaccurate.

It’s tempting to compare the SSHWS with the enhanced Fujita scale introduced in 2007 to classify tornadoes. The Fujita scale assigns ratings from EF0-EF5 based on the correlation between damage and wind speed. The original scale developed in 1971 attempted to smoothly join the Beaufort wind speed scale and the Mach scale for velocity, as seen here in the graph. Because it’s heavily based on years of observational damage, the EF scale provides fairly reliable expectations for tornado damage. However, factoring observational damage into hurricane ratings would be difficult because storms spend so much time at sea, where terrestrial damage data means very little. The Fujita scale was designed to be applicable to hurricane damage anyway, and hurricane engineers often describe damage using EF numbers.

One recent attempt to reform hurricane ratings is the Hurricane Severity Index, formulated by ImpactWeather. Private development of the HSI began shortly following the 2005 Atlantic hurricane season that caused record damage of over $159 billion. HSI uses a 50-point scale, with 25 points assigned to size—determined by the total coverage of several wind field levels—and intensity determined by wind speed. While the HSI has the potential to more accurately describe a storm, its broader numerical scale would probably be less intuitive to the public.

The HSI competes with a few other newer hurricane metrics, including the US National Weather Service’s accumulated cyclone energy (ACE) metric. ACE is the sum of the squares—taken at six-hour intervals—of the maximum estimated sustained velocity of a tropical storm. ACE effectively describes the most intense winds in a storm’s center, so it’s still a poor metric for understanding the potential for damage and destruction over a large area.

The recent metric with the most potential for describing damage might be NOAA’s integrated kinetic energy (IKE). Like the Fujita scale, IKE is heavily based on observational data collected from Hurricanes Andrew, Hugo, and Opal. It uses a five-point scale similar to the Saffir-Simpson, but it allows for continuous decimal gradations. A NOAA paper (pdf here) by Mark Powell and Timothy Reinhold describes how IKE is more relevant to damage caused by not just wind but also storm surge and waves. And unlike ACE, IKE takes the entire size of the storm into account, as shown in this calculator.

All rating scales are imperfect, and the Saffir-Simpson scale provides a simple means for warning those in a hurricane’s path. That said, incorporating new data and a few new metrics doesn’t hurt either.

Image credit: T.T. Fujita / Rugby471