Friday, May 24, 2013

We're Not Using Saffir-Simpson Anymore, And Neither Should Anyone Else

Back in 1971, the Saffir-Simpson scale for hurricanes had value and merit. The public wasn't as data-crazy as it is now, nor was the meteorological community. Our understanding of hurricanes, while pretty good, pales in comparison to what we know now as the science of the past 42 years has taught us tremendous amounts of information about how hurricanes work, the true relationship between pressure and wind speed, pressure differential, and size and scope of storms.

You won't see us use any Saffir-Simpson hurricane categories in 2013.

In 1971, Saffir-Simpson worked.   In 2013?  Well...let's just review a few recent examples:

Katrina (2005) had the wind speed of a Category 3 at landfall with the storm surge of a Category 5.  Sandy, regardless of whether or not you feel it was tropical or not at landfall, had the wind speed of a Category 1 but produced a much higher storm surge than a typical Cat 1 does.  Charley, a Category 4, had less of an impact on property and people than Sandy, which would have been a Cat 1 on Saffir had it been classified a hurricane, as Charley was a much smaller hurricane.

It lead us on more than one occasion to opine about the merits of Saffir-Simpson and its usefulness.

Saffir is good at one thing and one thing only...ranking storms by top wind speed.  It's simple, it's easy, and in 1971 that worked because that was where we were scientifically.  It was before the implementation of IKE, ACE, and other such metrics that measure potential impact of a storm, the life cycle strength of a storm, and so forth.  In the good old days, Saffir would apply as a nice way to rank storms.

Given what we know now about hurricanes and tropical cyclones, that no two storms are the same, that size matters in tropical cyclones, and that wind speed alone is not the be all and end all of tropical measurements (unless you're right in the eye), there needs to be a better way to track storms.

This year, is going to refrain from using Saffir-Simpson in any posts.  We encourage others in our field to do the same.  It is an outdated classification system given a storm like Sandy can produce more damage than Irene at the Shore, given both were similarly ranked storms, yet we communicate them both as "just" Cat 1 storms.  We can't communicate such a ranking to the public given that Sandy and Irene produced completely different impacts in our region -- Irene from flooding rains in Pennsylvania, Sandy from the wind and coastal flooding impacts in New Jersey and a lesser extent in Delaware.

No two storms are ever the same.  Throwing them into a simple ranking based on one factor is not beneficial to the public.  It may be easy for TV consumption but it's not helpful for public consumption.  People, we think, need to know more than that.

Storm impact -- the "what's in it for me" equation -- matters more than ranking a storm on a scale from one to five.  That's what we will discuss, especially if storms threaten the East Coast or our region.  The impacts we'll focus on are wind, coastal flooding (storm surge), and rain.  That's what matters.  We may mention the pressure is ranked at a certain point in comparison to other famed storms, if only for comparison's sake in history, or to make a point about the pressure gradient between a hurricane and high pressure nearby (Isabel's winds are an example of this).  These are things that matter to people, not whether it's a Cat 1, 2, 4, or whatever.

So, this summer and fall, when tropical systems develop, we'll mention the top winds, the pressure occasionally, but we're not going to fall into the trap of ranking them on a scale of one to five in the manner Saffir-Simpson does.  We will use some of the other metrics out there such as IKE or modeling of storm surge impact to show what the storm could do, may do, but we're going to avoid a simple scale.

We hope in time the rest of the community moves beyond Saffir.  Its usefulness and shelf life have expired.  It was a good system in a simpler, less data-intensive era, but as we have learned more about tropical cyclones we need to adapt our knowledge and place a priority on communicating impacts over merely putting them in classifications simply to achieve soundbyte status.