The Need for Improved Hurricane Forecasting

For example, it can often become a major source of extra worry and stress when one has to travel out of town on business or vacation during the peak season. There's nothing like leaving the wife and kids at home, or just your home unattended when there is a storm swirling around in the Caribbean. Many of us wonder why the NHC can't give us better information about the probability of hurricane activity.

Hurricane season lasts from June1  to November 1, literally half the year, yet it is extremely rare for hurricanes to form in all but three of those six months -- August, September and October. But the media will breathlessly tell us about every little tropical depression that forms, but at times poses almost no threat of ever becoming a hurricane, yet alone reaching the North American Continent.

In over a quarter century of tropics watching, there are a number of things I've learned about hurricanes, their development and tracking. There are periods when world weather conditions do not favor the spawning of hurricanes. We've come to know these as the El Nino years. And there are periods when tropical cyclone activity greatly increases, such as in La Nina years. But, as we've come to learn, the number of hurricanes that develop has absolutely nothing to do with WHERE those storms will go.

Fortunately for those of us who live on the Atlantic and Gulf coasts, we've seen many years when virtually every storm that developed traveled across the Atlantic, only to curve harmlessly to the north, out into the central Atlantic long before they threaten the continent. This happens more often than not, and in some years virtually every storm does this. You know the routine, we've been through it dozens of times. 

Some of us remember the emphasis that used to be placed on the Bermuda high, a high pressure area that normally sits over the Atlantic somewhere in the vicinity of Bermuda. The positioning of the high was the thing that normally determined what track the storm would take. If that high shifted a bit further east, this tended to direct storms either onto, or up the US coastline. A little further west, it would send the storms into Florida or western Caribbean, creating a barrier high pressure ridge that prevents the northward curing.

Typically, it is the clockwise flow around the back (west) side of a high that causes the northward curve, so it's pretty easy to see how the Atlantic high controls direction. The high itself acts as a blocking and steering force, but always in conjunction with upper level winds that must be favorable to the storm's development (an upper level high over a surface low).

In recent years, we've heard little or nothing about the Bermuda high, but here in Y2K, the prominent feature is that there is no Bermuda high. Instead, the dominant weather pattern has been a series of deep troughs over the western Atlantic, extending down into the lower Caribbean. So far this year we've had four named storms and 8 depressions. Alberto became a hurricane and moved harmlessly into the central Atlantic. All the others were torn apart by strong upper level shearing winds caused by these troughs.

What struck me most about this weather pattern that persisted for nearly three weeks was that there was almost no chance of any tropical cyclone marching across the Atlantic through these troughs to reach the U.S. None. Every single one of them was promptly torn apart as soon as it came to the first of these three troughs, let alone being able to move through two more. Then came Debby, a minor hurricane, which got as far as Puerto Rico, but still had a trough ahead of her. At this point, the NHC was issuing strike probabilities charts with South Florida in the bullseye. I happened to be in New York at the time, so my only source of information was television. Needless to say, I was more than a little concerned, so I rushed my scheduling to return with grave consequences, only to arrive home with NO MORE DEBBY.

What happened to her? She rather predictably ran into a trough that tore it apart within 24 hours. Now why, I wondered, pulling up the old satellite photos, didn't they predict this when it was so plainly obvious what would happen? One of the main tools that hurricane forecasters use are a variety of satellite photos that clearly show strong upper level wind patterns. These are all made available on numerous web sites for anyone to see and use, assuming you know how to use them. 

Us old-fashioned sailor types are just the sort to read the weatherman's data and  second guess their predictions, since they so often hit the nail on the head!

Now here we are in the midst of the peak of the hurricane season with a highly unfavorable weather pattern, and yet the NHC does not issue any forecasting advisories about current probabilities about storm formation or tracking. Oh, sure, they say something like, "tropical storm development is not expected for the next two days."  That's not much help. Of course, we're all aware of the political realities that cause the NHC to be exceedingly conservative about what they say, particularly regarding forecasts of existing storms. But why they cannot foresee highly unfavorable conditions like we've experienced in the last month, and issue such  advisories or probabilities is something of a mystery to me since many people would be greatly benefited by such information.

Example: "The forecast potential for a hurricane reaching such and such an area for the next 7 days is estimated at 0% or 10%." Or whatever the case may be. My own experience suggests that making such  predictions is within the realm of possibility. Or they may say, "The forecast potential for a hurricane reaching such and such an area for the next 7 days is currently not predictable." in the event that they can't feel sufficiently confident about making a forecast. It would seem to me that it's possible to make such predictions with a high degree of reliability, far higher than the typical daily weather forecasts, the accuracy of which is probably something less than 50%.

The point being that there are periods when there is almost no chance of a storm forming and reaching a particular area, and times when conditions become more favorable. For the past month the tropical Atlantic been in a weather pattern that does not favor tropical cyclones reaching the US, as reality has proved out. I suspect that a lot of people would have liked to know that, considering the effect it could have on their daily planning.

Most of us are familiar with the famous Dr. Gray who issues his forecast of the number of storms that will form. Unfortunately, the number of storms has nothing to do with where those storms will go, and is basically of value to no one. 1995 saw the most active season on record with 19 named storms, but not one Cape Verde storm even came near the US coast. Four named storms did hit the US, ALL of these being late or early season western Caribbean formations.

Or we can look at it from the standpoint of strikes by decade where we note some remarkable variances. In 1992 there were the least number of named storms, but only one strike, Andrew, one of the worst storms of all, and it was the ONLY Cape Verde storm, which demonstrates that numbers has nothing to do what you and I may experience. The fact is that prevailing weather patterns at any given time yield a  rather predicable probability.

Since government bureaucrats are always seeking to expand their empires, and weather scientists are no different, I'd hope that they'd do a bit more to keep us better informed.