View Full Version : New Leadership at NHC?
rdale
07-03-2007, 08:54 AM
Sounds like Proenza's concern for the future of hurricane forecasting could be leading him to a nice job in the private sector...
Pressure builds for storm chief
The controversial director of the
National Hurricane Center came under additional scrutiny as a federal team of experts conducted an unannounced inspection.
BY MARTIN MERZER
http://www.miamiherald.com/460/story/158757.html
M Conder
07-03-2007, 10:14 PM
Forecasters want storm director ousted
"Three senior forecasters at the National Hurricane Center called Tuesday for the ouster of recently appointed director Bill Proenza, saying he has damaged public confidence in their forecasts, fractured morale and lost their support..."
BY MARTIN MERZER
http://www.miamiherald.com/459/story/159712.html
rdale
07-04-2007, 01:00 PM
Even more interesting... The QuikScat comments came from an unpublished study, and of "questionable" science. And now Congress is getting the impression that hurricane hunter aircraft produce inferior data!
http://www.wunderground.com/blog/JeffMasters/show.html
John Wetter
07-04-2007, 10:43 PM
It's quite interesting that all the leads have spoken out...
-John
Justin Turcotte
07-05-2007, 04:12 AM
It's quite interesting that all the leads have spoken out...
-John
Azz kissing, IMO. The career minded person will join the hunting party when the "witch" hunt is on faster than Chuck Schumer can find a television camera. Proenza fairly criticized NOAA for spending (or what many consider wasting) millions of taxpayer dollars for what amounts to a big party while technology used to more accurately forecast storm systems goes to the wayside. The merrits and expense of QuickScat can be debated but I see no reason why the public shouldn't be aware of potential implications should the satellite fail.
"Three senior forecasters at the National Hurricane Center called Tuesday for the ouster of recently appointed director Bill Proenza, saying he has damaged public confidence in their forecasts, fractured morale and lost their support..."
The only thing the public has lost confidence in is NOAAs ability to manage resources appropriately. I don't see how morale is fractured when your leader is lobbying for additional funding for your organization. The alleged loss of morale is probably a result of NOAA directors unwillingness to be challenged/debated in the court of public opinion. NOAA is not a publically traded firm that kisses up to shareholders. It is a government organization that must answer to the taxpayers. Proenza is the whistleblower on behalf of the taxpayers and for that I give him greater respect. It is unfortunate NOAA leadership has become extremely hostile in this regard. Government leadership at its finest. If only we had more "mavaricks" like Proenza floating around.
Glen Romine
07-05-2007, 09:58 AM
The career minded person will join the hunting party when the "witch" hunt is on ....
I'd be surprised - after all these guys are all senoir forecasters - so their careers are quite secure. I'd actually think if anything Proenza has insulted them by his statements. Sure, the data is helpful to the hurricane forecasters, and does have some value in improving at least some of the model guidance available to forecasters, but it's an insult to the skill of the forecasters themselves to suggest that their forecast skill is directly tied to the quality of NWP guidance, particularly from only one model. Forecasters I imagine take great pride in their skill - and for your boss to suggest that your skill is no better than the skill of the guidance that you use - well that's not going to win him any popularity contests. So, would loss of the Quikscat actually impact forecasts from NHC? Probably, but to quatify it is very difficult. It would appear though that the tool is not deemed as more valuable than forecaster pride for many of the senior forecasters at NHC.
Glen
Justin Turcotte
07-05-2007, 12:55 PM
... but it's an insult to the skill of the forecasters themselves to suggest that their forecast skill is directly tied to the quality of NWP guidance, particularly from only one model. Forecasters I imagine take great pride in their skill - and for your boss to suggest that your skill is no better than the skill of the guidance that you use - well that's not going to win him any popularity contests...
Glen
Forecast models when used appropriately can improve forecasts and when used inappropriately can hinder forecasts. I can honestly say my forecast verification would decrease without the aid of various NWP products and much more so without the aid of data. My verification would also decrease significantly if I solely relied on the model QPF output rather than my synoptic and mesoscale analysis, situational awareness, pattern recognition, knowledge of climo, etc. Lets take away all satellite data including IR and VIS and see how good the hurricane forecasts are. Only a fool could claim his forecast skill would be unchanged (assuming skill in this case is tied directly to verification). The NHC has boasted their forecast track errors have decreased significanly over the last 20 years. This is probably due to a combination of increased knowledge of tropical cyclones, better data, and better NWP. In short, no forecaster I know wants to lose data even if only of minor use.
Forecasters from my experience do take great pride in their skill but in this case I think the NHC folks have confused knowledge with skill. Nobody wants their intelligence insulted as these forecasters suggest happened. Unless I'm missing something between the media lines, all I see is a director who is trying to get more data so his forecasters can better utilize thier knowledge. I do get a sense that these forecasters believe the QuickScat satellite data is "overrated" in the eyes of Proenza. Perhaps it is. Science is full of debate and this one is worth continuing. Would this even be an issue if Proenza didn't chastize NOAA for the party spening? I still have the feeling there is some bandwagon hopping going on here.
rdale
07-05-2007, 02:05 PM
Looks like you do need to read in between the lines... As it's now coming out - for whatever reason he is saying that QuikScat is more important than recon planes and is willing to risk reducing funds for aircraft to develop a new satellite. I'm no tropical expert, but I see a major flaw there...
IAN GIAMMANCO
07-05-2007, 05:43 PM
Just a quick two cents, I don't think Proenza was trying to suggest that Quikscat data was more important than recon, that was misconstrued by the media and by congress. In regards to the forecasters, I still wonder how any of this actually underminds their abilities. I do know Rich Knabb wanted that job, and Proenza was basically just appointed to it.
Someone has to fight for hurricane research funds as they have been steadily cut. I will say more focus should have been put on the Joint Hurricane Testbed (JHT) rather than Quikscat. Although valuable over the open Atlantic, I think we can deal without for a while if that is the case.
rdale
07-05-2007, 05:52 PM
I don't think it's really going to hurt the public response to tropical warnings. Again a lot is being played out simply because it's in the media, but agree that while fighting for funds is the job of a center head -- make sure you're coming out fighting for the right tools.
MClarkson
07-05-2007, 10:25 PM
Didnt I say the quikscat was relatively unimportant in the last thread we had about this?
"QuikScat is another tool that we use to forecast," Lixion Avila said. "The forecast will not be degraded if we don't have the QuikScat."
http://www.cnn.com/2007/US/07/05/hurricane.furor/index.html
while not completely useless, quikscat is pretty near the bottom of things I look at for the tropics.
anyway, i know whos side im taking in this.
Franklin and Avila... i remember their names on NHC forecast discussions since i first started looking at them like 10 years ago, and I know Avila has been around a lot longer than that. Proenza... never heard of him till Max retired.
Jeff Snyder
07-05-2007, 10:54 PM
It's interesting that senior forecasters would say the loss of any semi-reliable data (I assume it's semi-reliable, otherwise I don't think we'd ever see it used in the forecast process or in the advisory discussions) would not have any negative impact on forecasts. From my casual forecasts through the past 6 years (chasing-related and otherwise), one of the things that I quickly learned is that it's important to realize that, sometimes, my forecasts are terrible. Sometimes it's my fault for ignoring something, and other times it's a situation that had very low predictability. In all cases, however, it's just as important to remember what we DON'T KNOW as it is to remember what we DO KNOW. In other words, realizing one's short-comings and limitations, IMO, is very important if one is to forecast well on a consistent basis (I know we've all been "cocky" from time to when we're on a stretch of "good" forecasts, only to be knocked down abruptly when Ma Nature shows us that we aren't as good as we think we are).
In the chasing realm, most of us who have been chasing or forecasting for more than a couple of years realize the limitations of numerical models and our understanding of the atmosphere. Perhaps, after decades of experience, some people don't really need every tool available to them, but I don't think that I'll ever get to the point at which I'm confident enough to say that I don't need satellite data (or other data, for that matter). Sure, some data are not weighed very heavily in my forecases, but I can't imagine that would be the case for satellite-derived winds in areas over the ocean where measured data is otherwise non-existant.
MClarkson
07-05-2007, 11:17 PM
The problems with quikscat data are many; The relatively low-orbiting satellite can survey a given spot of ocean only about once a day. Velocity data is often(always?) suspect with the quikscat winds and it cant peer into the CDO very well, especially the eyewall. Its resolution isnt quite what the plane can do either. The recon aircraft also return much more than just wind data, and at many levels if you count the occasional dropsonde.
If a plane is in the storm I probably wouldnt even look at the quikscat derived winds. Even in the deep ocean I think a strength estimate based on more conventional images(Dvorak technique, etc) is not only just as if not more accurate, but far more timely as well.
The one thing I do like to see a quikscat for is a developing deep ocean disturbance, especially to determine if circulation has closed off.
Jeff Snyder
07-06-2007, 12:27 AM
The one thing I do like to see a quikscat for is a developing deep ocean disturbance, especially to determine if circulation has closed off.
This is a different topic if it's an EITHER / OR situation. In other words, is it either recon or QuikSat?
At any rate, recon usually doesn't fly out for disturbances well in the central or eastern Atlantic. So, we could depend solely on ships that happen to provide enough evidence that the circulation has closed off, or we could use a tool that is at least better than nothing. From a human impact point of view, recon is very important, particularly since the storms that tend to garner the most attn from recon are those that are likely to impact land areas in the western Atlantic. However, I find it odd to hear that the forecast process will not be affected by losing a tool such as Quikscat, particularly for those disturbances and/or storms that are away from the recon domain. Gosh knows that some tropical storm forecasts are absymal to begin with. This seems particularly to be true of weak systems (and those undergoing cylcogenesis) and for those with rapid variation in intensity (i.e. those that are rapidly intensifying or rapidly weakening). Can you ever be confident enough on your forecast to say that the loss of this tool will have no negative impact on your forecast, particularly given that some tropical systems are poorly forecast to begin with? [I certainly don't think quiksat is the end-all and be-all, otherwise I would hope we'd see excellent forecasts for every system/storm that moves through the Atlantic each year. As it is, I'd have to think that more soundings would help... Can we set up a system by which ships or buoys can release RAOBs? LOL] Even if the Quikscat data are only used a handful of times during the season, it still seems imprudent to say "The forecast will not be degraded if we don't have the QuikScat." (per a quote from Avila in the CNN article below).
I'm not incredibly well-versed in tropical storm forecasting and analysis, so my thoughts above should not be weighed too heavily. However, I thought I'd give my 2 cents. Ah, the political... err... forecast process.
We have at least one (and, I think two) NHC forecasters on Stormtrack, so it'd be interesting to hear what they have to say. As it is, given the circumstances, though, I wouldn't be surprised if these members don't feel it appropriate to comment.
EDIT: This subject is now the lead story on CNN.com --> http://www.cnn.com/2007/US/07/05/hurricane.furor/index.html ... I really have to think there's more to this story than we know. Personality conflicts? I don't know, but it really sounds like politics is flirting with science again (I'm *shocked*... .... ....). I can't imagine a group petitioning to remove their boss while in hurricane season. Apparently, it's severe enough that it can't wait until after hurricane season. There has to be more to this story than just some folks upset that their boss said forecast quality would decrease if Quikscat "died".
EDIT 2: Here's a little better article (and includes the names of those who have signed the petition)--> Storm director's staff revolt widens: The staff rebellion at the National Hurricane Center grew dramatically, with nearly two dozen employees calling for the departure of director Bill Proenza. (http://www.miamiherald.com/459/story/161590.html) (Miami Herald). Per that article, the petition for resignation includes "Lixion Avila, James Franklin, Rick Knabb and Richard Pasch; hurricane forecasters Eric Blake, Dan Brown and Michelle Mainelli; meteorologists Wally Barnes, Robert Berg, John Cangialosi, Hugh Cobb, Martin Nelson, Gladys Rubio, Chris Sisko and Patricia Wallace; oceanographer Stephen Baig; executive officer Ahsha Tribble; administrative officer Vivian Jorge; and Proenza's administrative assistant, Evangelina Maruly." Now I'd REALLY be interesting in hearing from the NHC folks who are Stormtrack members...
Chris Hayes
07-06-2007, 02:51 PM
This is just rediculous in my opinion. He's the head of the NHC, and he lives in the United States. He has a right to voice his opionion about whether Quickscat should be replaced or not. Why are they complaining about him thinking it should be replaced. If he doesnt like the superiors in Washington, thats his perogative. I think there's more to the story than it sounds. Maybe they dont like the fact they werent promoted to head of the NHC. :rolleyes:
Scott Overpeck
07-06-2007, 09:36 PM
Chris is right, this may go a little deeper than one may think. Proenza and NWS HQ and NOAA have a lot longer history than needs discussed. I may not agree with the way Proenza has handled this, but let's just say it is not surprising. Proenza may have a rift with the higher ups, but I think at the same time he is trying to get funding for the satellite regardless of how useful it is. One might think that when GOES 8 or 9 go down, they will be replaced. Sadly, if Quicksat is not replaced, what makes you think there is funding to replace GOES 8 or 9 when the time comes? That may be how NOAA thinks. "We had good forecasts without quicksat, we'll have good forecasts even without GOES 8 or 9." Ya, right. If you think forecasting will not be hindered without Quicksat, do you think forecasts will not suffer without GOES 8 or 9? I hope not. You may think this is a bit extreme, but I like having precedent set of replacing satellites when they are dated and need replacing. It is nice having as much data as possible. Just my 2 cents here which may not amount to much.
richhorodner
07-06-2007, 11:20 PM
At first report of this "rift" at the NHC I thought the feds were coming down on the director as vengence for bucking their authority and criticising his higher ups in a publilc forum.
After reading the names of the employees that signed the petition to fire Mr. Proenza my opinion has changed. This list included the lady who has been the Administrative Officer and Assistant to the Director (many directors) for nearly 30 years (you know, that always busy and very loyal lady at the office that really runs the place), the Director's own secretary, several of the researchers in the back room who have been there several decades, and even a technitian (computer, electrical, radar, etc) that has worked at the NHC for several decades.
Based on this, my guess is that the quikscat and budget thing is not the core reason for the rebellion, just a vehicle to that rebellion. This is a very conservative, by the book type of people that are the least likely to publicly air their displeasures. My guess is that this new director is one of those bosses that just rubs everyone wrong, orders people around for the sake of it, alters long tested and refined procedures just for the purpose of putting his mark on things, despite coming from a meterological field that has little to do with tropical meteorology. The type of boss that makes you not want to drag yourself out of bed for, and go to work every day to deal with.
From Psycology 101-
Canebeard
Gregg Gallina
07-08-2007, 06:26 AM
Chris is right, this may go a little deeper than one may think. Proenza and NWS HQ and NOAA have a lot longer history than needs discussed. I may not agree with the way Proenza has handled this, but let's just say it is not surprising. Proenza may have a rift with the higher ups, but I think at the same time he is trying to get funding for the satellite regardless of how useful it is. One might think that when GOES 8 or 9 go down, they will be replaced. Sadly, if Quicksat is not replaced, what makes you think there is funding to replace GOES 8 or 9 when the time comes? That may be how NOAA thinks. "We had good forecasts without quicksat, we'll have good forecasts even without GOES 8 or 9." Ya, right. If you think forecasting will not be hindered without Quicksat, do you think forecasts will not suffer without GOES 8 or 9? I hope not. You may think this is a bit extreme, but I like having precedent set of replacing satellites when they are dated and need replacing. It is nice having as much data as possible. Just my 2 cents here which may not amount to much.
Scott, you may be a bit outdated on your lingo... the Goes satellites 8 and 9 ARE dead. We now use 12 (EAST) and 11(West) with 13 on central US backup (for when either EAST or WEST dies). I continue to fall into the old Goes8 speak myself because she was sooo reliable at 9.5yrs, but just call them GOES-EAST and GOES-WEST...that way you won't have some PITA (like me) give you a lecture.
Also, the GOES program is well stocked up...with the NOP(e) group coming up already funded...and the GOES-R series slated for 2014. GOES-R will be a revolution in satellite imagery with 5 minute scans for the CONUS, .5km resolution for visible imagery and 15 other channels (3 in the visible range to make "true" color images).
As for Quikscat... as an operational user it is a great satellite for TC ANALYSIS...which of course helps forecasting but is not as crucial for the long range forecast. Bigger deal is the not finding the center or max wind like the recon...but it is the large swath of good sfc wind structure. It will be sorely missed in the TC world, but much more at the Ocean Prediction Center, TAFB and Honolulu that forecast wind and waves...that is where it will be HUGE loss.
Sorry to be off topic in this thread, but I wanted to clear up some of the satellite issues you had. Also, I will stay out of the political arena on this one ;)
Rob Lightbown
07-08-2007, 09:57 AM
My thoughts at first were that Bill Proenza was bucking the "system" and trying to do what was right for his organization and the public it serves. Since last Tuesday, my opinions have changed and I'm not really sure what to think.
With at least half of the NHC staff revolting against Proenza, much like everyone else, I think that there is a lot more going on behind the scenes. One statement Proenza made kind of "sheds a light" on his thinking. He made a quote stating that his subordinates have signed a petition against him. Subordinates?? That really rubbed me the wrong way and I think that he may have a mindset that his staff is inferior to him. I could be wrong, but thats how I read it. I have had a few bosses like this and Rich is correct in the statement that this is the type of boss that you want to call in sick the rest of your life for and just quit.
In addition to this, I also read a rumour back on July 4th that there is talks of possible restrictions of hurricane track model data. The rumour is that some senior staffers at NHC, including Proenza, want to restrict access to NHC model track data. I don't know if this rumour is true b/c I haven't read it anywhere else or been able to verify it conclusively. Has anyone on this board heard anything about this rumour?? :confused:
richhorodner
07-08-2007, 10:16 AM
http://cbs4.com/video/?id=37023@wfor.dayport.com
Including the fact that the "surprise" inspection might have been a surprise to the Director, but not the NHC staff or a Miami Herald reporter who has been doing most of the hurricane related stories since 1988 for this paper.
rdale
07-08-2007, 12:49 PM
In addition to this, I also read a rumour back on July 4th that there is talks of possible restrictions of hurricane track model data. The rumour is that some senior staffers at NHC, including Proenza, want to restrict access to NHC model track data. I don't know if this rumour is true b/c I haven't read it anywhere else or been able to verify it conclusively. Has anyone on this board heard anything about this rumour?? :confused:
There's no way to pull that back... It'd be a violation of their Congressional mandate if they were to try - and you KNOW what kind of fuss AW would (correctly) start if this were to happen. It'd risk the entire NWS operation if Congress was involved because NWS willingly held back "potential life-saving information."
Mikey Gribble
07-08-2007, 01:34 PM
I think of this as two different issues and I am on board with everybody else that believes that there is a lot more to this story than we know. That interview (that was posted) was enlightening as to what is actually driving these people to call for their bosses removal.
I agree that he should be removed strictly because of the unrest he has obviously caused among the other employees at NHC. It is apparent that he is not a good manager in this particular environment and he needs to be replaced if that many people are upset.
On the other hand, I strongly disagree with the forecasters that are claiming the loss of QuikSCAT wouldn't matter. Proenza is right in pointing out the misalocation of funding by his superiors and I think the regular employees at NHC used the QuikSCAT issue as a vehicle to further their main cause when Proenza's QuikSCAT comments really didn't have anything to do with what they were upset about. Whether the loss of QuikSCAT decreases forecast accuracy 15%, 10%, or 1% doesn't really matter IMO. It is an important forecasting tool and it fills a niche that will not be replaced by any other available products. Proenza was right to point out the fact that if we can afford to blow a bunch of money on a celebration then we should be able to afford to replace a satellite.
The gest of my feelings are that they should replace Proenza and QuikSCAT.
Rob Lightbown
07-09-2007, 02:43 PM
Breaking from CBS News "National Hurricane Center Director Bill Proenza temporarily re-assigned, Ed Rappaport Temporarily Placed In Charge"
http://cbs4.com/
Rob Lightbown
Crown Weather Services
http://www.crownweather.com
MClarkson
07-09-2007, 03:25 PM
I find it surprising that the removal of the NHC director is breaking news on the national networks.
Politics and forecasting combined, the result of Katrina. I dont think I like that.
cdcollura
07-10-2007, 10:59 AM
There is another thread about the Quickscat satellite's imminent failure on this site as well!
http://www.stormtrack.org/forum/showthread.php?t=12887
When it comes to POLITICS, simply asking a question can destroy your career.
Just WATCH - When the satellite does fail, and a 160-MPH storm is tearing away at the US with a 12 hour advance warning (instead of 48) ... Then, maybe the NHC / US Govenment will say "OEEPS, he was right, we should not have fired him."
I will never work for the govenment - EVER!!
I worked for the city back in 1993, and was let go for NO REASON ... Later, the person they hired (to replace me) was making $10,000 less than me - That WAS "the reason".
rdale
07-10-2007, 11:04 AM
Where did your example come from? QuickScat is NOT crucial to landfall forecasts. It's great for stuff over the middle of the ocean, and not too many people live in the middle of the ocean. Recon is crucial when hurricanes approach land, and because of his comments some in Congress openly stated they think recon funding should be cut so a new satellite can be built. The research he used was flawed, and his forecasters told him so, yet he went with it.
Bad idea.
Glen Romine
07-10-2007, 12:41 PM
The research he used was flawed, and his forecasters told him so, yet he went with it.
Would you care to elaborate on that statement? Have you even actually read the paper Proenza referenced? The work went through the standard peer review process - one that is believed to be pretty effective in ensuring that published research is credible. It is my impression that the study itself is fine - but that Proenza mis-represented the information he released. Clearly, he felt very strongly about wanting the Quikscat replaced and tried to leverage this study to overplay the impact once it is lost, whereas some of the NHC forecasters apparently do not share his level of interest and are rather pissed they weren't asked what they thought he should be campaigning for. Also, there almost certainly has to be more to the story than this one issue which we likely will never hear about - which is particularly clear since it was not just the forecasters on staff that wanted to see him removed.
rdale
07-10-2007, 01:35 PM
No - I'm going off Jeff Master's comments. And again there is more to the story, but nobody in the tropical world I've found says that the mainland US is in dire jeopary if QuickScat dies.
If you have the ability to put it in a short format - exactly what would cause the landfall errors if recon was good but QuickScat was gone?
Glen Romine
07-10-2007, 02:11 PM
If you have the ability to put it in a short format - exactly what would cause the landfall errors if recon was good but QuickScat was gone?
I have not seen the study as it has not yet been published (though is near the top of the 'papers to appear' list so should soon be available). Speculating however, I don't get the impression that the study looked at landfall forecast errors at all - but simply forecast errors in TC positions. This could be done by having one experiment use all the available observations ingested into the initial condition used by the operational model, integrate the model to some time in the future and then look at the errors in the forecast relative to what came to be observed. So, this is an assessment that has to be done after the actual storm event. Then, the same experiment is done except the intial condition processing is denied the data from the Quikscat, was again integrated to some time in the future and was evaluated for the errors in the storm location relative to observations. Then, looking at the cumulative effect from many forecasts (a number of which for the same storms, but at different initial forecast times), the overall performance showed degraded forecast position estimates without the Quikscat data in the ingest.
The basic belief in model forecasts is that a better initial condition for the model at the beginning of a forecast cycle will generally (though not always) lead to a better forecast. The suggestion would then be that the information from the Quikscat is helpful if the forecast positions from the model run usining the Quikscat data in the ingest are generally closer to the observed storm locations than equivalent runs not ingesting the Quikscat data. Not knowing anything more about the study being referenced, I don't know if this improvement is most prevalent during any particular phase in the storm's development or geographic location - though with only 2 seasons considered I'm guessing this level of distinction was not pursued.
rdale
07-10-2007, 04:58 PM
http://www.wunderground.com/blog/JeffMasters/comment.html?entrynum=710&tstamp=200707
Gregg Gallina
07-10-2007, 06:57 PM
Would you care to elaborate on that statement? Have you even actually read the paper Proenza referenced? The work went through the standard peer review process -
The science is not flawed, it is the dataset... it was not expansive enough to show robust results. I believe it was only one season (2003) with limited intense hurricanes and TD cases. One season is not really enough, you need a good diversification...which is not typically the case across one season as global parameters such as MJO, ENSO, and placement of subtropical high tend not to vary enough. So a better database would expand to many more cases (over 100 would be good) with a diverse intensity gambit, as well as multiple seasons.
Overall, the science and methodology were good... and logical, but just like all results, you can view them anyway you want to prove your points ;)
Glen Romine
07-10-2007, 09:40 PM
I believe it was only one season (2003) with limited intense hurricanes and TD cases. One season is not really enough, you need a good diversification...which is not typically the case across one season as global parameters such as MJO, ENSO, and placement of subtropical high tend not to vary enough. So a better database would expand to many more cases (over 100 would be good) with a diverse intensity gambit, as well as multiple seasons.
The title of the study says it was done over two seasons - though still the number of forecasts looked at was only 19 (according to the review by Jeff Masters) for the 72 hr forecasts, he didn't mention the number of samples for the 48 hr forecasts, though you would assume more of these were available. I agree this is too small a number (at least for the 'known' 72 hr forecasts) to draw broad conclusions, though you can make decent statistical arguments with a sample size of around 30 if the data aren't too noisy, which is a little easier to compile than 100. Also, while I agree that there could be sensitivity of the data impact from differences in large scale pattern regimes, you could easily find a dozen or more other possible biases.
richhorodner
07-22-2007, 01:18 PM
http://www.palmbeachpost.com/storm/content/state/epaper/2007/07/20/m1a_PROENZA_0720.html#audio
This article recently in the Palm Beach Post has two audio files you can click on in the left margin of the article. The one titled; "NOAA Administrator Conrad Lautenbacher discusses the situation at the NHC" is quite informative of the situation.
Also found the following history-list of NHC Directors on Wikipedia (the user controlled "encyclopedia" of the Internet)
Gordon Dunn (1965–1967)
Robert Simpson (1967–1973)
Neil Frank (1973–1987)
Bob Sheets (1987–1995)
Bob Burpee (1996–1997)
Jerry Jarrell (1998–2000)
Britt Max Mayfield (2000–2007)
Bill Proenza Non-Grata (2007)
Edward (Ed) N. Rappaport (2007–Current
vBulletin® v3.7.2, Copyright ©2000-2008, Jelsoft Enterprises Ltd.