Leave the Science to the Scientists at the NHC

[MWatkins]

This has been an emotional debate for the most part...but I want to take a moment and reflect on the science behind this debate...and what I think is the real underlying problem is behind this mess.

First of all, here is a distinction between overall forecast skill, which is not up for debate, and landfall forecast skill. Intuitively, we know that QUIKSCAT is better than conventional satellite imagery at defining the location of a circulation center in some disorganized systems, especially if the center is obscured by high clouds.

We also know that track forecast skill is dependant on accuracy of the initial position. Being off by 60 nautical miles with the start position will result in a really bad forecast, for example. So it makes sense, to me, that in the absence of more reliable data, not having QUIKSCAT could impact forecast accuracy of storms way out at sea and out of recon range.

However…once the storm is in range of recon, the importance of QUIKSCAT essentially goes away.

So then the question becomes…would watch/warning areas need to be expanded without QUIKSCAT? The answer…no.

Recon is always available before any watches or warnings in almost every case, except for rapidly developing systems or mechanical delays etc. So one has to question the validity of the statement that watch/warning areas would have to be expanded without QUIKSCAT. If Proenza said this (and I think he did), his assessment was incorrect.

But still, because many forecasts are made without any recon data (not sure of the percent but I would guess 50% of NHC forecasts do not have recon data), there is still open for debate that overall forecast skill could be eroded.

Dr. Jeff Masters, who for the record has always been a huge contributor on the Talkin Tropics program…has been quoted in the Sun-Sentinel as saying:

"You have to start with the science – and if you don't understand the science in this business, you shouldn't be doing it,"

Which is exactly what he did in his blog entry written last week:

http://www.wunderground.com/blog/JeffMa ... amp=200707

He did some digging into the study Proenza has been citing about the degradation of forecast skill. Among other things, there were three major things that stood out:

1. The study was done on 2003 storms, and predominantly on storms out of recon range
2. The study compared skill of the GFS model…not against NHC forecast tracks
3. The results had not even been completed before Proenza was citing it publicly

Furthermore, the true impact on forecast skill may be significantly less…perhaps a statistically insignificant 2%. Dr. Masters did a tremendous amount of work to get this out…and I am glad he did.

After reading this detail, and looking at the other facts we know about…Proenza clearly jumped the gun. Yes…QUIKSCAT is dying and there is no replacement ready. That is a huge problem…but a scientific study has been misrepresented or misinterpreted.

So the new guy makes a mistake…that’s understandable. He took a risk and stood up for a resource that he felt was necessary, which is commendable. However, he did not do his scientific research, which is not so much commendable. I do think that if he were such a liar as he is being portrayed…wouldn’t that have come up during the interview process? This is a political move too…and many are doing all they can to make sure the American public is convinced this is the right move.

See…this highlights a much bigger problem. At the time, Bill Proenza was the “most qualified” person for the position. By that I mean:

1. He was the highest ranking government employee who wanted the job
2. He was the highest ranking person willing to put up with the off season stuff that comes with the package

Was it because he had the most hurricane forecasting experience? No…of course not. And there’s your problem.

The Director of the National Hurricane Center IS NOT by default the best hurricane forecaster available. Remember Bob Burpee?

I said this before Proenza was hired and I will say it again. We need to take the public interface stuff out of the job and let the head of the Hurricane Center focus on one thing: providing better hurricane forecasts. Get someone else to do the PR…hire a preparedness spokesperson. The Neil Frank model of director is antiquated. Get rid of it.

If they do this, the most qualified person for the job (Dr Rappaport) will stay as director.

Unless this happens, we will continue to put lesser qualified people in charge of making hurricane forecasts…unless we happen to get lucky and find someone like Max who had both the forecasting proficiency and the will to deal with the PR side of the job. And we know the PR and politics wore him down.

Get rid of the public interface (except when storms are actually coming). Keep the most qualified person in charge…and leave the science to the scientists.

MW

['CaneFreak]

AMEN!!!! Could not have said it better myself!!! Great post Mike!!!

[miamicanes177]

Dr. Jeff Masters, who for the record has always been a huge contributor on the Talkin Tropics program…has been quoted in the Sun-Sentinel as saying:

"You have to start with the science – and if you don't understand the science in this business, you shouldn't be doing it,"

Which is exactly what he did in his blog entry written last week:

http://www.wunderground.com/blog/JeffMa ... amp=200707

He did some digging into the study Proenza has been citing about the degradation of forecast skill. Among other things, there were three major things that stood out:

1. The study was done on 2003 storms, and predominantly on storms out of recon range
2. The study compared skill of the GFS model…not against NHC forecast tracks
3. The results had not even been completed before Proenza was citing it publicly

Furthermore, the true impact on forecast skill may be significantly less…perhaps a statistically insignificant 2%. Dr. Masters did a tremendous amount of work to get this out…and I am glad he did.
MW

What about the storms that recon will not fly because they are too far away? We need accurate accounts all storms for the record books. Wouldn't it be nice to have an accurate account for storms 500 years from ago? Think about that too. And Dr. Masters was in full support of Proenza until he gathered word of the hurricane forecasters not supporting him. In his blog June 22, just 17 days before calling for his firing and AFTER learning of the report Proenza was citing for the 10%/16% error numbers, Masters wrote this:

"Last night, I listened in to Proenza's comments an Internet radio show. I asked him where he got his numbers of 16% and 10% improvement for 72-hour and 49-hour hurricane track forecasts made using QuikSCAT satellite data. Proenza cited a study done of hurricane tracks from 2003 that showed these improvements. Margie came across a 2006 study which shows that for one storm studied (Hurricane Cindy of 1999), inclusion of QuikSCAT data improved track forecasts at 24 hours and 48 hours by 30-50%. There is also a 2007 study which showed improvements of 25%-50% for 24 hour - 48 hour model track forecasts of 2002's Hurricane Isidore using QuikSCAT data vs. no QuikSCAT data. We'll have more on the ongoing Bill Proenza hullaballo next week, with more info on just how important QuikSCAT is to hurricane forecasting."

CINDY 1999


ISIDORE 2002


I'm not sure what caused Masters to flip so quickly using the same information he had when defending Proenza. Politics, money and power my friend. A lot of people read Masters blog and he influences many people. NOAA wanted Proenza out badly. They would stop at nothing to see it through.

[Frank2]

Interesting post, Mike, though I have a few comments:

this is the only analogy I can come up with at this moment (I need to type fast this morning), but, as in the movie Volcano, the head of the agency is also usually the person who must deal with the press, so, to hire someone exclusively for this job (NOAA does have an Office of Public Affairs), doesn't always mean the person will not be called on - regulary - by the media (doesn't the President have a White House spokesperson - but still must deal with the press on almost a daily basis), so, while you are correct that good media skills do not always go hand-in-hand with good forecasting skills, as in most OCM's, it helps to have both, and, the NHC Director is no exception...

As for Bob Burpee, well, Dr. Burpee is well-respected in his field, but, like many of us, is perhaps more comfortable behind the camera, instead of in front of it (though he retired because it was his personal decision, as in the case of the other NHC Directors)...

That said - I was somewhat surprised when I heard that Max Mayfield was chosen as the NHC Director, but, as in the case of Dr. (Bob) Sheets, both are somewhat on the quiet side - but, like Jerry Jarrell, seemed to do well with the media, so, sometimes, what a person seems to be doesn't always mean that's how it will turn out...

Frank

[Berwick Bay]

I wish that there were an easy way to make happen what MWatkins is proposing. I think that (as usual) money is at the root of this. The higher ups don't want any pressure from the mainline forecasting dept. about how to spend the money allocated to them. The money spent involves contracts with the private sector so that the needs of NOAA (sat or otherwise) can be met. This involves cozy relationships of long standing between govt people in the Commerce Dept. and the private sector. So Proenza says "in the interest of public safety, money should be spent on Quikscat, as opposed to whatever else is in the budget". And marginally, public safety might be affected, or at least we can say that the money would be better spent here than on other matters. But he has made his statements publicly, shining a light on the priorities within the Commerce Dept. This is taboo. The Dept. has made a decision that any small effect on public safety from not updating the Quikscat system is not be compared with "other priorities" within the Dept. As I said, the effect on public safety would appear to be "rather small". But the question then becomes, "Well, If not for Quikscat, how is the money being spent?" Higher ups don't want that question asked. Like it or not, the NHC is now fully emersed within the Washington Establishment of lobbying and corporate contracts, and all of the unsavory things that go along with it.

[Frank2]

I'd comment on what you mentioned, Berwick Bay, but, I need to tread easy, since I'm still within...

...sometimes, what a person seems to be doesn't always mean that's how it will turn out...

By the way, per my earlier post above, the topic of whether someone is a good spokesperson has a Biblical origin, as in Moses' conversation with God (Exodus 4, Verse 10), when poor Moses pleads with God (concerning His selection of Moses for the task at hand), saying:

"O Lord, I have never been eloquent, neither in the past nor since you have spoken to [me]. I am slow of speech and tongue."

[And] Moses said, "O Lord, please send someone else to do it." (Exodus 4)

Perhaps this (or something similar) is what Ed Rappaport is thinking today, since he first declined the opportunity to be considered as NHC Director...

[Berwick Bay]

Really good post Frank. Speaking biblically, it also reminds one of the story of Saul and David. Saul wanted the Kingship (and had it for a number of years), but the Kingship wanted David! And spiritually speaking, there always more power in the latter situation.

[flwxwatcher]

Interesting post MW!!

[Frank2]

So, true - as I heard recently, the person who is best suited is the one who least desires the title, since they are the usually the person who is the most humble of heart...

[Frank2]

Oops - sorry, flwxwatcher - guess you posted 1/4 second before I did...

[fci]

Good post and summation MW.

There have been some thought provoking posts both pro and con regarding this situation and I applaud many who have been involved in this discussion on the couple of threads that have been posted to discuss this situation.

I can't sum it up much better than the aforementioned posts on the various threads and MW's post that started this thread. I am NOT a scientist so my views represent a member of the General Public completely unbiased. Basically, I am the target audience of the NHC.

My thoughts come down to a couple of points:

- Mr Proenza took a calculated risk in going public with his concerns in an environment that wholly frowns upon such "public" behavior.

- He alienated his support team by using support data that they, simply; could not support. If he was going to step out like he did he HAD to have the support of his team which he obviously did not. In addition the public comments made by the foreacsters sure make it seem like this was not the only issue causing turmoil at the NHC.

- NHC suddenly became a political issue which is way beyond what it is meant to be. Sure, there had to be politics going on there behind the scenes as they are in virutally every workplace. However, as a member of the General Public, I would never have known about the political workings going on there. NOW, even the General Public saw the politics and that IS NOT a GOOD THING.

- In today's environment the "trust factor" is so critical. Whereas in the past, the General Public had pretty much blind faith in their Government and its agencies, that is no longer the case. There are several issues that exemplify this; Katrina response, The War.... Issues that, in this non-political forum; I will not delve into either positively or negatively. My point is that with "trust" no longer unconditionally given; the Government has to be all that much more trustworthy. Anything that erodes that trust has to be rooted out at once. When there is dissension seen PUBLICLY at a key agency such as the NHC right at the cusp of Hurricane Season it is a disaster. When the General Public sees the NHC director on television during "crunch time" these political issues have to be non-existant. If we think their eye is off the ball, it is a huge problem. EVEN THOUGH WE AS WEATHER ENTHUSIASTS AT S2K KNOW THAT THESE PROFESSIONALS ONLY HAVE THE BEST INTEREST OF THE PUBLIC AT HEART; the PERCEPTION might not be so. THAT is the problem and why Proenza had to be removed from his position.

While I understand WHY Mr. Proenza did what he did, I think it was not the correct thing to do. He miscalculated what he did and it was political suicide. In my opinion, his ouster was completely valid and a shame. But a necessary shame.

fci

[MWatkins]

Thanks Frank, Berwick, Canefreak and everyone else who added some very interesting thoughts to this idea.

There are a couple of things I want to expand on just a little bit.

First, I did not give full credit to the blog link above. Dr Masters made a point to stress that Margie Kieper contributed a significant amount of work in that entry...and I did not give her any credit for that in the post...

Second, I think it is important to remember that the staff there at the hurricane center, especially the forecasters, have been put in a very difficult position, should be given a little leeway. Clearly, they want to stick with the forecasting...and while they are government employees...they are not politicians. I am sure we have no idea what went on behind the scenes, and I doubt the full picture will ever come out.

So I think they deserve a pass...and at the end of the day I think the NHC is stronger than it was yesterday. The right guy is in charge (for now), and they can get back to work for someone they respect...and who is clearly qualified for the job.

MW

[Robjohn53]

I have read alot of this stuff about whats going on but yet don't understand it all. But i still wonder about a couple things.

1) Did NHC really want him out and if so why and i mean really why.

2) Did everyone respect him knowing he was on his way out.

3) I know in some other situations if someone sides with someone thats on there way out
there job could be effected as well. So i wonder this, Did they really respect
him, and was just looking out for there own jobs. Or did they really not respect him at all
and if so he must have really done alot to lose that respect of so many people. Like what
was said earlier we probly will never really know.

Just somthing to think about.

Robjohn53

[Frank2]

1) Did NHC really want him out and if so why and i mean really why.

2) Did everyone respect him knowing he was on his way out.

3) I know in some other situations if someone sides with someone thats on there way out
there job could be effected as well. So i wonder this, Did they really respect
him, and was just looking out for there own jobs. Or did they really not respect him at all
and if so he must have really done alot to lose that respect of so many people. Like what
was said earlier we probly will never really know.

1) The public only knows what we have learned from the media (if there are other reasons, they will likely remain an internal matter), but, it did not seem that the selection was well-received from the beginning, for reasons mentioned (any administrative issues, aside) by several forecasters - one issue was the fact that the former Director was away from operational forecasting for many years...

2) I'm sure it was an unhappy few weeks or even months at the NHC, but, I don't think it ever escalated to anything other than that (no one was ever notified of anything in that regard), so, as unhappy as they were, they likely still offered him the respect he deserved - though under duress, no doubt...

3) There are always two sides to each argument, but, per what the staff said repeatedly about the Director (the Director did not say anything negative, in public, about the staff), it was the Director's lack of consideration for the staff that was unnerving to them - again, there are two sides to each story...

Of course, all of the above does not include any conflicts between the Director and his superiors that were mentioned by the media, so, it's important to consider that, as well, since this was not only a "staff versus boss" issue...

Hopefully all (including the former Director) will be able to go forward without any other conflicts - these things happen sometimes, and, hopefully, it's over now...

P.S. I agree, MWatkins...

[Frank2]

Here's the historical view of the entire NHC issue (I wonder who will play NHC's version of Robert the Bruce):

http://en.wikipedia.org/wiki/Bill_Proenza

it's interesting to realize that this is exactly how the histical process of every event in world history takes shape - makes you think...

P.S. May we all make only good contributions...

[superg77]

We need more satellites not less. We need more funding for the NHC and hurricane research not less. Proenza was right for sticking up for more research and for more data. Sticking up for a QUIKSCAT replacement was starting with the science. Why would we ever want less data or settle for less? The answer is always more, more, more data. The more data and funding we have the more we can learn about hurricanes. Settling for less is dumb. This is allowing the government to reduce science. We should never accept that. We also need all the hurricane information and data made public -- every single bit of it -- not taken away as MWatkins suggests. We are the ones paying for it.

[Derek Ortt]

we pay for cl;assified information as well, and that most certainly should not be made public, just ebcause we pay for it

To say that gov't should make something public just because we pay for it, to be honest, is 10 miles over the cuckoos nest. There are better arguments that can be made regarding this issue as to why data should be made publicly available.

My lunch break is over, back to more VRWs for the rest of the week

[superg77]

we pay for cl;assified information as well, and that most certainly should not be made public, just ebcause we pay for it

To say that gov't should make something public just because we pay for it, to be honest, is 10 miles over the cuckoos nest. There are better arguments that can be made regarding this issue as to why data should be made publicly available.

My lunch break is over, back to more VRWs for the rest of the week

Actually it was a perfectly logical argument. Why in the world would every single bit of hurricane and weather data not be public? You can't compare it to classified information.

[Jevo]

We need more satellites not less. We need more funding for the NHC and hurricane research not less. Proenza was right for sticking up for more research and for more data. Sticking up for a QUIKSCAT replacement was starting with the science. Why would we ever want less data or settle for less? The answer is always more, more, more data. The more data and funding we have the more we can learn about hurricanes. Settling for less is dumb. This is allowing the government to reduce science. We should never accept that. We also need all the hurricane information and data made public -- every single bit of it -- not taken away as MWatkins suggests. We are the ones paying for it.

Im having trouble following your point here... I dont see one time mike insisted that we reduce NHC funding... However he did put forth a much better argument backed by data and reasearch..... Weheras you are verball vomiting your beliefs and ideas... with nothing to back them.... I assure you Mike has forgotten more about Science and weather than you have learned.. I see you are new so maybe its time to learn which of the posters know what they are talking about.. As you can see others have posed their points of view as questions allowing comment....

JMHO

[superg77]

We need more satellites not less. We need more funding for the NHC and hurricane research not less. Proenza was right for sticking up for more research and for more data. Sticking up for a QUIKSCAT replacement was starting with the science. Why would we ever want less data or settle for less? The answer is always more, more, more data. The more data and funding we have the more we can learn about hurricanes. Settling for less is dumb. This is allowing the government to reduce science. We should never accept that. We also need all the hurricane information and data made public -- every single bit of it -- not taken away as MWatkins suggests. We are the ones paying for it.

Im having trouble following your point here... I dont see one time mike insisted that we reduce NHC funding... However he did put forth a much better argument backed by data and reasearch..... Weheras you are verball vomiting your beliefs and ideas... with nothing to back them.... I assure you Mike has forgotten more about Science and weather than you have learned.. I see you are new so maybe its time to learn which of the posters know what they are talking about.. As you can see others have posed their points of view as questions allowing comment....

JMHO

Sorry for any confusion although I don't appreciate your "verbal vomitting" comment. I didn't agree with Mike's argument that somehow having less data (no QUIKSCAT) is better science. If you read Mike's post it is clear he is trying to make an argument that we are ok without this instrument. I think that's a cop out and we should demand that the government replaces the satellite with an even better one. I think everyone should have stuck behind Proenza on this one. That's all I meant. I feel the fact that we have not and some of the NHC mets have not means we are allowing the government to reduce the amount of hurricane data. Proenza loses, the government wins and the amount of available hurricane data shrinks. It makes me sad To make it clear my position is pro Proenza and pro QUIKSCAT