The Saffir-Simpson hurricane scale - with categories from 1-5 - has been stripped down to indicate wind speed only. It provides no information on storm surge, rainfall rates/amounts, and tornadoes - which collectively cause far more damage and fatalities than wind alone when hurricanes come ashore.

Voices within the meteorological community are coming together to stress that the National Hurricane Center (NHC) must come up with new and innovative ways to communicate the hazards posed by hurricanes, emphasizing the limitations of the Saffir-Simpson wind scale.

Marshall Shepherd, president of the American Meteorological Society, wrote on his blog that we must do a better job conveying the hurricane rain and flood threat. Isaac, which was just a category 1 storm, produced significantly more rain than Katrina which was a category 3 storm. Shepherd uses the experience from Isaac to suggest supplementing the Saffir-Simpson scale with more rain-flooding information:

With Isaac, the storm was also very large in area and moving very slowly. This combination led to localized rainfall totals in the 20 to 30 inch range. The “water” from rainfall and flooding is just as significant as the wind and surge. Is it time to consider an augmentation of the Saffir Simpson scale to capture the rainfall-flood threat? It is a difficult science problem, but probably one worth investigating.

Broadcast meteorologist Nate Johnson (WRAL, Raleigh) wonders whether we should get rid of the Saffir-Simpson scale altogether. While acknowledging the merits of its simplicity, he concludes its single-hazard focus leaves out too much information:

... in a forward-looking sense, [the Saffir-Simpson scale] may be helpful to get us in the ballpark of what kind of damage we might expect.

But the ballpark is all we can expect such a categorization to give us. We must never forget that when we reduce storms to a category, this process is indeed a reduction – we’re removing data in exchange for the simplicity of a single, easily-comparable number. When we’re dealing with a single storm that’s threatening a populated coastline, that simple number belies the multitude of details that absolutely matter – details we don’t see when looking at a single category number. It also masks the truth that not all category 1 hurricanes are created equal.

Yet, Tim Heller, chief meteorologist of the ABC affiliate in Houston, reminds us that the Saffir-Simpson scale was initially designed to provide information about both wind and storm surge. He argues that if the scale were to be used as initially intended - to categorize a storm based on its peak wind and/or storm surge, whichever is higher - then it would be effective for making decisions about evacuations:

If hurricanes are ranked based on wind speed AND/OR storm surge, WHICHEVER IS GREATER, Hurricane Ike would have been ranked a “category four hurricane based on storm surge” [even though it was actually rated just a 2 based on wind]

Hurricane Isaac would have been ranked a “category three based on storm surge.” [even though it was only category 1 based on wind] Hurricane Andrew in 1992 would have been ranked a “category five based on wind speed.”

We don’t need another index.

Some have suggested we need to scrap the Saffir-Simpson Scale altogether and come up with something different.

But the public is familiar with the one-thru-five ranking and they do respond to a “category three” or higher hurricane. They don’t evacuate when a storm is producing category one winds, but they need to when the same storm is producing a 6-12’ storm surge.

CWG’s hurricane expert, Brian McNoldy, adds the Saffir-Simpson Scale is “very convenient for historical analyses but also very limited when it comes to describing the impacts that will be experienced when a hurricane approaches land.” He proposes the development of a new set of multi-hazard graphics to present hurricane risks:

Something that I think could help is a new suite of graphical products that are only generated when a storm has any probability of being near the U.S. coast. Most storms never affect land, so altering a scale to accommodate the landfalling ones doesn’t make sense. But a special set of landfalling products does. For severe weather over the U.S., the Storm Prediction Center creates a series of simple maps to illustrate the overall risk for severe weather, as well as the various components: tornadoes, straight-line winds, and large hail.

The National Hurricane Center could create a similar set of subjective maps for hurricanes: an overall impact map with separate maps for wind, rain, and surge. If the analyses are made available to the public in GIS formats, this single official set of maps could easily be utilized by any TV station or website for their own graphics. This visual depiction of the scale of the various risks could do a lot for increasing the tv-watching and internet-browsing public perception of what’s coming their way, and hopefully begin eroding away at the notion that Category 1 and 2 hurricanes are weak and not worth preparations or even evacuations.

Personally, I like a hybrid of the Heller and McNoldy approaches.

* Use the Saffir-Simpson scale to classify storms based on wind AND/OR storm surge. This gets around the fact some storms have relatively strong winds and small storm surges (e.g. Charley) while others have relatively weak winds and large storm surges (e.g. Isaac, Ike, Katrina). What matters is that people are prepared for the worst the storm has to bring. Granted, this would still not address rainfall. So...

* NHC should work with risk communicators and social scientists to develop all-hazard graphics to effectively convey the potential dangers of wind, storm surge, rain, and tornadoes from each and every landfalling storm.

What are your thoughts?