We're already had Hurricane Stan, and Tropical Storm Tammy is now active. After just two more - Vince & Wilma - we'll be on to Greek alphabet named storms. And if there's just three more storms this season, 2005 will be the most active hurricane season on record for the Atlantic basin.
Listening to various "experts" on TV, I decided their statements left me with a lot of questions. So I decided to take a look at the data myself. This diary is the second entry in that effort. The first graph covered the period from 1886-present; since then I've explored another data source, and was able to extend back to 1851 with what I think is better quality data for the years before 1920.
Data from: weather.unisys.com
graph by me
poll follows
Since 1887 was a very active year, I was especially curious to see what happened in the years immediate preceding. That's been accomplished. This is still a work in progress, with more work to come.
What's the likelihood that this will be a record hurricane season? Consider the following for the top ten seasons for named storms 1851-present (or equivalent since naming only began in 1950).
Total # thru Total
Year Storms Oct. 5 Cat. 4+
1887 19 10 0
1933 21 18 2
1936 16 15 0
1969 18 13 1
1995 19 16 3
2000 15 12 2
2001 15 9 2
2003 16 12 2
2004 15 12 4
2005 19* 19 4
*Season still in progress, likely to increase
Five of the top ten have occurred in the past 6 years, since 2000. In comparison, 4 occurred in the 100 years from 1900-1999, and 1 in the 49 years 1851-1899.
The National Hurricane Center has produced a table which is frequently cited to support the idea that there haven't been any real changes in recent decades. That what we're seeing now is consistent for the period of record (since 1851). Their table selects and lumps data in a way which obscures what's shown in my graphs. Here's how:
- It only includes hurricanes, not tropical storms
- It excludes all storms & hurricanes which did not make U.S. landfall
- It lumps all data for a 10-year period together
- It shows Category 3,4&5 together (if one looks instead at Category 4&5, the trends are clearer)
I can't help wondering if this is deliberate. If there's a political agenda like at EPA, sending things back if they don't show the right results. I looked up the biography of the director of the National Hurricane Center (NHC), Dr. Max Mayfield, in google. It comes up on the White House website, and does appear to be a political appointment though that isn't state explicitly. Dr. Mayfield has been with NOAA since 1988 and has a relevant resumé for the job. He's no "Brownie". But one wonders how he could be in possession of the data in my graphs but insist there's no trends. One can speculate that his willingness to do so figured in his selection as Director of the NHC.
The thing about all this is that there are consequences. Not only in dealing with greenhouse gas emissions, but also in preparedness and other aspects of policy. It's hard to address a problem if you won't acknowledge its existence. If those in power (=oil industry) keep denying the situation at hand, more and more people are going to suffer. Just as we see in the little "Bushvilles" cropping up throughout the south, Florida to Texas, as people try and rebuild lives (and bury the dead) between record storm ravages both in terms of number and intensity. Tragic? Partly. Negligent? Certainly. Criminal? Perhaps, but certainly heading in that direction.
I did an earlier
diary on this subject, which included the following graph:
Data source: wunderground.com
Graph by me
wunderground.com, where the data for the earlier graph came from doesn't state clearly where their data came from. In contrast, weather.unisys.com has much better metadata. I sent e-mail to wunderground.com to ask about this, but have received no reply to date. Unisys shows more storms, more variation in storms, and extends back an additional 35 years. For these reasons, Unisys looks like a better source to me. Regardless, graphs based on the two different data sources show similar trends.