1126: Epsilon and Zeta

Explain xkcd: It's 'cause you're dumb.
(Redirected from 1126)
Jump to: navigation, search
Epsilon and Zeta
The average error in the NHC forecasted position of a hurricane three days in the future has shrunk to a third of what it was in 1990—a staggering accomplishment. However, as you may have gathered, forecasts of future storm *strength* have proved more difficult to improve.
Title text: The average error in the NHC forecasted position of a hurricane three days in the future has shrunk to a third of what it was in 1990—a staggering accomplishment. However, as you may have gathered, forecasts of future storm *strength* have proved more difficult to improve.


The Atlantic hurricane season normally runs from June to November. Randall is imagining the situation in the National Hurricane Center when the 2005 season was extended more than a month by the appearance of Hurricane Epsilon and Tropical Storm Zeta. He imagines the situation as NOAA meteorologists watch with amazement (and increasing annoyance as they were presumably unable to move off to other things such as post-season analysis) as Hurricane Epsilon and Zeta continued to exist far beyond the normal end-of-season date (November 30).

The monospaced text in most of the panels is material taken from actual NHC reports from that season. The commentary has been edited to fit the comic's format, but it's otherwise faithful to the actual reports. While the only change to Forecaster 1 is when he's celebrating New Year's Eve, Forecaster 2 is visibly losing it after the appearance of Zeta in late December, with unkempt hair and an unshaven beard.

The last report of the 2005 season was issued on January 6, 2006.

A full analysis of the 2005 hurricane season can be found here.

NHC reports on Epsilion and Zeta:

Randall has discussed the seemingly erratic nature of hurricanes before. This may, however, have been a response to the recent Hurricane Sandy.

The text also seems to be a parody of horror stories/movies, and their theme of writing.


The 2005 Atlantic hurricane season saw devastating storms like Katrina and Rita. But less well-remembered is just how strange the season got toward the end.
The forecasters at the National Hurricane Center are the best of the best.
Their predictions are masterpieces of professional analysis. But in November 2005, out in the center of the Atlantic — far from any land — the atmosphere stopped making sense.
And the forecasters — who'd expected the season to be long over by now — started to get a little…...unhinged.
This is their story, as seen through the actual 2005 NHC Advisories:
[Two men, one bald and one not, sit looking at their respective computers, at separate desks, back to back. The advisory is printed above them in caps small-caps Courier type.]
Tuesday, November 29th, 2005: Tropical Storm Epsilon ... The 26th named storm of the apparently never ending 2005 Atlantic hurricane season.
[The same scene, different text.]
10 PM Wed: The window of opportunity for strengthening should close in 12-24 hr. 4 PM Thu: Slow but steady weakening is expected to begin in 12-24 hours.
[The man with hair now has questions marks above his head.]
4 AM Fri: Epsilon does not appear weaker. 10 AM Fri: Epsilon has been upgraded to a 65-kt hurricane.
[The two still sit back-to-back.]
4 PM Sat: Epsilon has continued to strengthen against all odds ... [but] can not maintain the current intensity much longer since the environment is becoming increasingly unfavorable.
[The two still sit back-to-back.]
10 PM Sat: Epsilon might or might not still be a hurricane ... but in any case it likely will not be one on Sunday. 4 AM Sun: Epsilon is downgraded to a tropical storm.
[The two still sit back-to-back. The man with hair's fists are clenched.] :10 AM Sun: Morning satellite images indicate that Epsilon has restrengthened.
[A closer view of just the balding man at his desk.]
There are no clear reasons ... and I am not going to make one up ... to explain the recent strengthening of Epsilon and I am just describing the facts. However ... I still have to make an intensity forecast and the best bet at this time is to predict weakening ... Epsilon will likely become a remnant low. I heard that before about epsilon ... haven't you?
[The two men still sit back to back, but the man with hair is now turning his head toward the other man, with his arm resting on the back of his chair. The bald man is leaning forward in his seat, toward his computer while typing.]
4 PM Mon: The cloud pattern continues to be remarkably well-organized for a hurricane at such high latitude in December.
[The other man has turned back to his own screen.]
10 PM Mon: We have said this before ... but Epsilon really does not appear as strong this evening as it did this afternoon.
[Just the bald guy now.]
4 AM Tue: I have run out of things to say.
[The two of them again.]
10 PM Tue: The end is in sight. It really really is. But in the meantime ... Epsilon continues to maintain hurricane status. 4 AM Wed: The end is in sight ... yes ... but not quiet yet. I thought I was going to find a weakening system and instead I found that Epsilon is still a hurricane.
[The two of them still.]
10 AM Thu: Convection has vanished and Epsilon is now a tight swirl of low clouds. I hope this is the end of the long lasting 2005 hurricane season.
[This panel is blank and just reads: Nope.]
[The men are still at their desks. The bald man is leaning back on his chair and staring at his screen, taking his keyboard out of his desk; the other man's hair is noticeably disheveled, and he has started growing a five o'clock shadow.]
Enter Tropical Storm Zeta.
Friday, December 30th, 2005: An elongated area of low pressure ... which had its origins in an old frontal trough ... began developing organized convection overnight. Advisories are initiated on the 27th tropical storm of 2005.
[The men are still at their desks, the man with hair is even more bedraggled-looking.]
Any new storms would be in the 2006 season.
4PM Fri: Although the atmosphere seems to want to develop tropical storms ad nauseam ... the calendar will shortly put an end to the use of the Greek alphabet to name them.
[The bald man is now wearing a party hat and has a party horn in his mouth, and there is confetti in the air.]
But 2005's wouldn't end until Zeta did.
10 PM Sat: Zeta appeared on the verge of losing all of its deep convection a few hours ago ... but since about 21z the convection has been on somewhat of an increase again.
[A close view of the man with hair at his desk.]
10 PM Sun: This is like Epsilon all over again. Most of the conventional guidance suggested that zeta should have been dissipated by now ... well ... zeta is pretty much alive at this time. I have no choice but to forecast weakening again and again.
[The two of them again.]
4 AM Mon: By 24-36 hours ... a significant increase in westerly winds ... should act to shear away most of the associated convection ... and finally bring the 2005 Atlantic hurricane season to a merciful ending. 4 PM Mon: It is hard to conceive that a tropical cyclone will be able to survive for very long in such a hostile environment. therefore I have not backed off on the forecast of weakening.
[The two of them again. Both men have clenched fists rested back from their keyboards, frustrated.]
10 PM Mon: Zeta is stronger than yesterday.
10 AM Wed: As you can see... I ran out of things to say.
[Both men put up their keyboards...]
4 AM Thu: Satellite intensity estimates have decreased. Zeta is downgraded to a 30 kt tropical depression.
[...only to start typing on them again.]
10 AM Thu: Shortly after the previous advisory had been issued ... regretfully ... the intensity ... increased to 35 kt and Zeta is a tropical storm once again.
[The two of them again.]
10 PM Thu: Although it seems as if Zeta will never die ... the forecast continues to show weakening.
[Both men are now leaning back in their chairs, exhausted, their keyboards put away.]
4 PM, Friday, January 6, 2006: Zeta no longer meets the criteria of a tropical cyclone... which means that both it and the 2005 Atlantic hurricane season have ended. This is the national hurricane center signing off for 2005.
[The bald man still sits at his desk, the man with hair is no longer in his chair.]
[Again, we see the bald man at his desk and the other guy's empty desk.]
Bald guy: Actually, Zeta's cloud pattern is...
Hair guy: (out of panel) NO.
Bald guy: Ok.

comment.png add a comment! ⋅ comment.png add a topic (use sparingly)! ⋅ Icons-mini-action refresh blue.gif refresh comments!


Official hurricane discussions for EPSILON and ZETA are here. I did read these discussions back when Randall made a blag post poking at them Odysseus654 (talk) 16:31, 26 October 2012 (UTC)

Does this have anything to do with the Italians convicting a bunch of scientists for failing to predict an earthquake? 16:46, 26 October 2012 (UTC)

Link? lcarsos (talk) 18:46, 26 October 2012 (UTC)
It was all over the news, but here's one account. I wonder if there are jurists in the Italian legal system aware of how much of a laughing stock this is making them. -- IronyChef (talk) 04:44, 27 October 2012 (UTC)
Note that the conviction was overturned by the appeals court. --MareCrisium (talk) 22:12, 11 June 2015 (UTC)
Actually, these "scientists" did issue a forecast: they assured that no quake was coming. This of course is ridiculous, since no one can currently make any such forecast. Out of 309 victims of the L'Aquila earthquake, at least 20 stopped taking the appropriate precautions, and would still be alive otherwise. This is why at the trial all 7 defendants were found guilty.
On appeal, 6 of them were acquitted because they raised a reasonable doubt of their involvement in the forecast (which was advertised as a result of the whole commission). The 7th was still found guilty in the appeal. The case is now to be discussed on the supreme court, so the remaining 6 might still find themselves in jail, where they deserve to be for making predictions which are totally un-scientific.
Unfortunately, many news outlets outside Italy reported the issue in an incorrect way, they said they were found guilty for failing to predict the quake. They did a prediction, which had no basis in science, and was totally wrong (maybe their crystal ball was foggy). Some people followed their advice, and died as a result. The commission members must be held accountable of their deeds.
Of course Randall couldn't be thinking of Italian seismologists. The NHC is a serious institution. And Randall knows perfectly well the difference between "failing to predict a quake" and "predicting that there won't be a quake".
--Lou Crazy (talk) 16:43, 31 August 2015 (UTC)

xkcd 980 (Money) also mentioned the fact that forecast accuracy has improved significantly: "Cost of hurricane forecast improvement funding since 1989: $440 million. Economic savings -- during Hurricane Irene alone -- due to limiting evacuations made possible by recent forecast advances: $700 million." S (talk) 00:16, 27 October 2012 (UTC)

I smile and occasionally chuckle at online comics, but this one had me guffawing with delight.--Noni Mausa (talk) 13:25, 27 October 2012 (UTC)

Good work to whoever linked the entire transcript! Trek7553 (talk) 21:24, 29 October 2012 (UTC)

I've added the rest of Zeta's saga, and added links in the transcript to each NHC message. Recommend the Quotes section be removed. David.windsor

The transcript section is just a transcript, but not the explain. I did a clean up. But that links like:
would be a great source for an proper explanation here. The explain itself is still a mess.
--Dgbrt (talk) 21:43, 6 December 2013 (UTC)

If somebody could add the links on the Zeta quotes, that would be great. Thanks! P.S. I'm not sure about transcript etiquitte, but if you do revert the links in it, please make them less bulky than the original. A list of "panel one: see here. Panel two..." might work. Kyt (talk) 19:04, 4 January 2014 (UTC)

It seems that the links in the transcript are not working correctly. For me it gives all 404 errors... what is going on? 18:58, 10 February 2014 (UTC)

Lots of NOAA links were incorrect, missing prefix zeroes, and missing in the Transcript. A different set were missing in the link list. I fixed all of both. Ioldanach (talk) 18:07, 14 May 2014 (UTC)