JB is confused and so am I....

[drezee]

From JB at 1430EST:
Station code or LAT, LON of starting point: ack Station code or LAT, LON of ending point: 39.1,-72.6 Distance: 321.5km, 173.6nm, 199.7mi Bearing: 222.8 degrees
The stats you see up there say that Nantucket is 199.7 miles northeast of the give center of Beryl. 222.8 degrees is 2.2 degrees away from a perfect southwest bearing of 225.
Now Miami has Beryl passing over or south of Nantucket, which is fine. It says Beryl is moving north-northeast...fine again. But Beryl is not by any stretch of the imagination south-southwest of Nantucket. If we look at a compass and 90 degrees per quad, south-southwest goes from 180 to 210 degrees, not 222.8, which is southwest.

Now this:
500pm Advisory position:

20/2100Z 39.6N 72.0W

Recon position

20/1925 39.32N 72.37W


So what they are saying is that Beryl will move

30 miles at 44 degrees in 90 minutes

OK so you mean Beryl will move at 20 mph due NE in that time...

[Derek Ortt]

radar shows that the advisory position is very close to accurate

the initial positions are subject to an error of roughly 20NM

[drezee]

radar shows that the advisory position is very close to accurate

the initial positions are subject to an error of roughly 20NM

True, but when you have recon in and radar all over it we should be able to do better

[Derek Ortt]

not when the center is this broad and far away from the radar range

[Aslkahuna]

Tha accuracy of the initial position on the advisory (which is in reality a forecast from the last reported actual position) will depend upon a number of factors-the first being how well defined a center is and secondly what the actual movement is between the last reported position and the current initial advisory position.

Steve

[wxman57]

I thought they'd made an error in the longitude. I had it at 72.3W not 72W. 72W is on the eastern edge of the center. Won't make much difference, though. Beryl won't have much of a punch by the time it crosses Cape Cod.

[drezee]

At 21:32:30Z

39 deg 31 min N
072 deg 11 min W


39.5N and 72.2W...is someone is trying to verify their forecast and ignoring recon?

[drezee]

UZNT13 KNHC 202317
XXAA 70238 99392 70714 11691 99009 24015 21049 00074 23412 21055
92754 19807 22554 85481 15800 23054 88999 77999
31313 09608 82312
61616 AF309 0502A BERYL OB 28
62626 MXWNDBND SPL 3926N07134W 2314 MBL WND 22056 AEV 20604 DLM W
ND 22556 008861 WL150 21555 080 =
XXBB 70238 99392 70714 11691 00009 24015 11850 15800 22843 15000
21212 00009 21049 11993 21558 22981 22055 33970 22059 44950 22558
55937 22554 66861 23557 77843 23051
31313 09608 82312
61616 AF309 0502A BERYL OB 28
62626 MXWNDBND SPL 3926N07134W 2314 MBL WND 22056 AEV 20604 DLM W
ND 22556 008861 WL150 21555 080 =

Measured 56mph surface winds and 1009mb pressure!


but yet they keep it at 50 mph....why have a drop if you aren't going to use it?

[Derek Ortt]

Dreezee, you're making a big deal out of nothing for 2 reaons

1. The advisory positions have absolutely nothing to do with verification. Only best track is used

2. That drop could be a spot wind and may not be representative of the total wind field. 45KT seems about right, and these winds won't even be felt on land, so there is no real concern arguing over 5 KT

[benny]

Dreezee, you're making a big deal out of nothing for 2 reaons

1. The advisory positions have absolutely nothing to do with verification. Only best track is used

2. That drop could be a spot wind and may not be representative of the total wind field. 45KT seems about right, and these winds won't even be felt on land, so there is no real concern arguing over 5 KT

Totally agree with Derek here....

on point #2... there was a spot surface wind of 49 kt. This could be a gust or a sustained wind. Looking at the WL150 wind.. which is the average of the lowest 150m of the drop... it is 55 kt... then a conversation has to be done to estimate sustained winds.. which average from .80 to .85 depending on how far down the drop penetrates. In this case it is closer to .85 of the WL150 wind.. which is like 46 kt.

[drezee]

2. That drop could be a spot wind and may not be representative of the total wind field. 45KT seems about right, and these winds won't even be felt on land, so there is no real concern arguing over 5 KT

By your argument, you only use it when it is within what you feel it should be. Take this example from Beryl...

THE AIRCRAFT REPORTED A CENTRAL PRESSURE OF
1001 MB WITH MAXIMUM 850 MB FLIGHT-LEVEL WINDS OF 64 KT IN THE
SOUTHEAST QUADRANT. A DROPSONDE IN THIS AREA ALSO MEASURED 47 KT
SURFACE WINDS. BASED ON THIS...THE INITIAL INTENSITY REMAINS 50
KT.

So the HNC used the Dropsonde to set the winds to 50kts.


THE AIRCRAFT MEASURED 54 KT FROM 850
MB AT 2314 UTC... WHICH EQUATES TO ROUGHLY 45 KT AT THE SURFACE
USING STANDARD REDUCTION FACTORS. LOW-LEVEL WINDS FROM A DROPSONDE
VERY NEAR THAT LOCATION IN THE MAXIMUM WIND BAND ALSO SUPPORTED
SURFACE WINDS OF ABOUT 45 KT... A LITTLE SURPRISING GIVEN THE LACK
OF DEEP CONVECTION IN THAT AREA.

Also used the drop to verify the 45 kt winds at the surface (even with lack of convection or possiblity of it being a spot wind.


Then recon measure what I mentioned above and you try and tell me that oh it must have been a spot wind. How do you know that? How do they know that? If so, why use in the first place? The thing is if it is in line with what they think it should be +- some number they take it into account. Otherwise, they ignore it. If you like we could go on all day about this. In times past, the NHC will specifically mention spot winds on the discussion (in this case 0300).

[drezee]

I am not new. I can recall this happening on a number of occasions. In recent memory, Wilma and Katrina last year. Recon measured one thind and the NHC advisory was something els. I am not saying that they are doing this across the board, but it does happen...and it should never happen. If you measure it then use it...otherwise you lose credibility.

[x-y-no]

If you measure it then use it...otherwise you lose credibility.

If that's the standard, why do we even bother having human forecasters? Just report whatever is measured and whatever the models say, right?


I'm sorry, but there's still a big role for expert and experienced judgement in this field. I don't agree that applying that judgement to the data to come up with a synthesis costs any "credibility." In my mind, the approach you propose is far less credible.

[drezee]

If you measure it then use it...otherwise you lose credibility.

If that's the standard, why do we even bother having human forecasters? Just report whatever is measured and whatever the models say, right?


I'm sorry, but there's still a big role for expert and experienced judgement in this field. I don't agree that applying that judgement to the data to come up with a synthesis costs any "credibility." In my mind, the approach you propose is far less credible.

And your counter is not, People are subjective by nature and therefore when dealing in a scientific matter measurements are all that matter otherwise you have nothing to properly use as a control or reference. The same thing happens in this board or in the NHC discussions about the way to pronounce Beryl. People have opinion...instruments do not.

I am not purposing that we use only dropsondes and I never said that so you may need to correct your assertion...the very fact that you came to that conclusion only give additional credence to my point of humans being subjective and opinionated.

Opinions are not facts...


This discussion is about when using dropsonde measurements. I have clearly demonstrated that from forecaster to forecaster and discussion to discussion you do know if they will use the drops or not. There should be some type of standard that is used. Otherwise, is all subjective and if that forecaster likes to use dropsondes or not. We use the info that the NHC puts out for more than just that day. We need it for trend analysis over years for climate research. If we do not have a clear standard on how to use such instruments, they the data is by definition invalid!

[x-y-no]

And your counter is not, People are subjective by nature and therefore when dealing in a scientific matter measurements are all that matter otherwise you have nothing to properly use as a control or reference. The same thing happens in this board or in the NHC discussions about the way to pronounce Beryl. People have opinion...instruments do not.

I am not purposing that we use only dropsondes and I never said that so you may need to correct your assertion...the very fact that you came to that conclusion only give additional credence to my point of humans being subjective and opinionated.

Opinions are not facts...

Maybe you can explain to me what other option there is. You criticized them mentioning dropsondes as part of their reasoning in some cases while in others coming to a conclusion different than what a particular dropsonde indicates. Your statement was "if you measure it, use it."

I would point out that in none of the examples you gave where they used dropsonde data was that the only basis for their conclusion. In each case, they mentioned flight level data and then said that "also" a dropsonde indicated some wind strength. IOW, they were doing exactly what I said - using their experience and expertise to create a synthesis from the totality od data they had.

So what's your alternative? Discard all dropsonde data from consideration? Always rely on dropsonde data even if other data is contradictory? Precisely what different methodology do you propose?

This discussion is about when using dropsonde measurements. I have clearly demonstrated that from forecaster to forecaster and discussion to discussion you do know if they will use the drops or not.

No, you haven't demonstrated that. All you've demonstrated is that they tend not to bother mentioning every potentially contrary piece of evidence which for some reason was discounted in making a particular advisory. I don't have any problem with them doing that - it saves the discussions being cluttered with a lot of irellevant stuff.

There should be some type of standard that is used. Otherwise, is all subjective and if that forecaster likes to use dropsondes or not.

You seem to imagine that they simply arbitrarily choose to ignore dropsonde data in some cases. I don't believe that's the case at all. As I said, they look at the totality of the data and apply their experience and expertise to come up with a synthesis.

We use the info that the NHC puts out for more than just that day. We need it for trend analysis over years for climate research. If we do not have a clear standard on how to use such instruments, they the data is by definition invalid!

I don't understand this argument at all. It's not like reconnaisance data is discarded if a forecaster chooses to discount it in the making of an advisory. If you want to do such research, I suggest you go to the raw data rather than reading forecast advisories and discussions.

[drezee]

Quote:
So what's your alternative? Discard all dropsonde data from consideration? Always rely on dropsonde data even if other data is contradictory? Precisely what different methodology do you propose?


The alternative is to have a set standard of how to use dropsonde data. Not simply...we feel that that dropsonde was not representative. Put it into a set of parameters and go from there. I am an engineer. If I take old data, then I need to know how it was measured and that they used the same way to measure it each time. Otherwise, I cannot use the data. What I am saying is that in the past, the discussion has used words like feel and think, not within parameters. This brings us to the point below:

Quote:
You seem to imagine that they simply arbitrarily choose to ignore dropsonde data in some cases. I don't believe that's the case at all. As I said, they look at the totality of the data and apply their experience and expertise to come up with a synthesis.


You need to read my comments. I said the following before you made the above comment: "I am not saying that they are doing this across the board, but it does happen...and it should never happen." The measurement should be uniform. This is science not art.


Quote:
I don't understand this argument at all. It's not like reconnaissance data is discarded if a forecaster chooses to discount it in the making of an advisory. If you want to do such research, I suggest you go to the raw data rather than reading forecast advisories and discussions.

I am talking about using raw data in the advisory in a uniform way, i.e. don't think or feel it should be disregarded...know why it is because of parameters that have been set and strictly adhered to.

[x-y-no]

The alternative is to have a set standard of how to use dropsonde data. Not simply...we feel that that dropsonde was not representative. Put it into a set of parameters and go from there. I am an engineer. If I take old data, then I need to know how it was measured and that they used the same way to measure it each time. Otherwise, I cannot use the data. What I am saying is that in the past, the discussion has used words like feel and think, not within parameters.

Measurements on any given storm are neccesarily sparse, i.e. far from comprehensive. And dropsondes are but one piece of that sparse data. To insist that it be used in some mechanical fashion would substantially degrade the overall accuracy of advisories.

And again, nothing said in advisories or discussions should in any way affect your ability to do subsequent analysis based on the raw data. That raw data remains available and it is collected in a consistent manner. I'm really confused as to why you would think the content of the advisories and discussions would affect that in any way whatsoever.


...

You need to read my comments. I said the following before you made the above comment: "I am not saying that they are doing this across the board, but it does happen...and it should never happen." The measurement should be uniform. This is science not art.

I'm sorry, but meteorology is still in substantial part an art.

...

I am talking about using raw data in the advisory in a uniform way, i.e. don't think or feel it should be disregarded...know why it is because of parameters that have been set and strictly adhered to.

The net result of such a policy would be to significantly degrade the accuracy of advisories and forecasts.

And once again, I think your characterization that data is "disregarded" is false. As I understand the process, all data is considered, but to the extent that there are contradictory indications from different parts of the full data (and taking into account the overall sparseness of the data we do have), some resolution must be made. And that's the process of synthesis based on experience and expertise that I've repeatedly referred to.

[benny]

2. That drop could be a spot wind and may not be representative of the total wind field. 45KT seems about right, and these winds won't even be felt on land, so there is no real concern arguing over 5 KT

By your argument, you only use it when it is within what you feel it should be. Take this example from Beryl...

THE AIRCRAFT REPORTED A CENTRAL PRESSURE OF
1001 MB WITH MAXIMUM 850 MB FLIGHT-LEVEL WINDS OF 64 KT IN THE
SOUTHEAST QUADRANT. A DROPSONDE IN THIS AREA ALSO MEASURED 47 KT
SURFACE WINDS. BASED ON THIS...THE INITIAL INTENSITY REMAINS 50
KT.

So the HNC used the Dropsonde to set the winds to 50kts.


THE AIRCRAFT MEASURED 54 KT FROM 850
MB AT 2314 UTC... WHICH EQUATES TO ROUGHLY 45 KT AT THE SURFACE
USING STANDARD REDUCTION FACTORS. LOW-LEVEL WINDS FROM A DROPSONDE
VERY NEAR THAT LOCATION IN THE MAXIMUM WIND BAND ALSO SUPPORTED
SURFACE WINDS OF ABOUT 45 KT... A LITTLE SURPRISING GIVEN THE LACK
OF DEEP CONVECTION IN THAT AREA.

Also used the drop to verify the 45 kt winds at the surface (even with lack of convection or possiblity of it being a spot wind.


Then recon measure what I mentioned above and you try and tell me that oh it must have been a spot wind. How do you know that? How do they know that? If so, why use in the first place? The thing is if it is in line with what they think it should be +- some number they take it into account. Otherwise, they ignore it. If you like we could go on all day about this. In times past, the NHC will specifically mention spot winds on the discussion (in this case 0300).

As mentioned above by xy... one spot wind of a dropsonde doesn't mean a whole lot. It could be a gust or a sustained wind. That's why NHC forecasters try to get more than one sample if possible or using the lower-level winds to get a longer average wind then reduce it to the surface. This is by no means perfect but much better than nothing at all. One dropsonde measures 47 kt... another 49 kt at the surface... the low-level profile might be a lot stronger for the 47 kt wind ... that's why they used it to justify 50 kt winds. the 49 kt at the surface.. the low level winds supported about 45 kt. convection was a lot stronger during the earlier drop that measured the 47 kt surface winds.

what are you arguing about here.. 5 kt??? like anyone on this board or anywhere has that kind of accuracy. meteorology is still rather subjective at times but the differences are +- 5-10% or so.. in the noise level.

[drezee]

I am amazed how when people are confronted with differing opinions they attempt to trivialize the argument itself.

The significance in the difference between 5 kts:

1. History: In 5kts, you have the difference between every single classification from TD to CAT5. If you are off by 5 kts, then some hurricanes are TS and so on.

2. Evacuations: There are mandatory evacuations that take place per the category of hurricane or Storm classification.

3. Insurance: Some insurance companies are discussing about changing their policies to include "hurricane-induced water rise." If it isn't classified then your insurance claim may vary.

[brunota2003]

2. That drop could be a spot wind and may not be representative of the total wind field. 45KT seems about right, and these winds won't even be felt on land, so there is no real concern arguing over 5 KT

By your argument, you only use it when it is within what you feel it should be. Take this example from Beryl...

THE AIRCRAFT REPORTED A CENTRAL PRESSURE OF
1001 MB WITH MAXIMUM 850 MB FLIGHT-LEVEL WINDS OF 64 KT IN THE
SOUTHEAST QUADRANT. A DROPSONDE IN THIS AREA ALSO MEASURED 47 KT
SURFACE WINDS. BASED ON THIS...THE INITIAL INTENSITY REMAINS 50
KT.

So the HNC used the Dropsonde to set the winds to 50kts.


THE AIRCRAFT MEASURED 54 KT FROM 850
MB AT 2314 UTC... WHICH EQUATES TO ROUGHLY 45 KT AT THE SURFACE
USING STANDARD REDUCTION FACTORS. LOW-LEVEL WINDS FROM A DROPSONDE
VERY NEAR THAT LOCATION IN THE MAXIMUM WIND BAND ALSO SUPPORTED
SURFACE WINDS OF ABOUT 45 KT... A LITTLE SURPRISING GIVEN THE LACK
OF DEEP CONVECTION IN THAT AREA.

Also used the drop to verify the 45 kt winds at the surface (even with lack of convection or possiblity of it being a spot wind.


Then recon measure what I mentioned above and you try and tell me that oh it must have been a spot wind. How do you know that? How do they know that? If so, why use in the first place? The thing is if it is in line with what they think it should be +- some number they take it into account. Otherwise, they ignore it. If you like we could go on all day about this. In times past, the NHC will specifically mention spot winds on the discussion (in this case 0300).

I would like to point something out...in each of the discussions quoted above, they had recon AND dropsonde data to back it up...if there are no flight level winds to back up the dropsonde with its surface measurements then most likely it is a gust and that seems to be what they thought the dropsonde recorded was simplely a gust...