54 replies to this topic

[Stormchaser]

Hey everyone, old man winter is coming up on us quickly, and only 1 month down the road we'll be mostly leafless, 1.5 months we'll have seen our first flakes, and the first accumulating snowfall possible as early as 2 months from now. The winter outlooks I've seen so far are mixed, a few on one end of the spectrum w/ cold/snowy winter, and a couple on the warmer side for this winter. Regardless October is a very important month in terms of global trends WRT atmospheric circulation patterns as well as oceanic SSTA progression.

I'll note several of those factors here -- namely what we want to see happen this month (which thereby increases our chances for a cold and/or snowy Eastern winter).

1. Pacific SSTA:






2. Atlantic SSTA:





3. Cold October - when looking back in history, I've found that approximately 75% of colder than normal Octobers temp wise in NYC are followed by winters (DJF) which are colder than normal (Brief aside: Tony on Eastern had similar results for Philly, which indicated that about 3/4 of cool October yielded cool winters).

Here are the stats I ended up w/ when researching winters since 1950:

A) 31/59 (53%) years featured a warmer than normal winter in NYC.

B) Of these 31 winters, 19 had a warmer than average October preceding it. So 19/31: 61%

C) 28/59 (47%) years featured a colder than normal winter in NYC.

D) Of these 28 winters, 20 had a colder than average October preceding it. So 20/28: 71%

Moving to November's correlation (or lack thereof):

A) Of the 31 warmer than normal winters, 16 featured a warm November prior to it. So 16/31: 52%

B) Of the 28 colder than normal winters, 12 featured a cold November prior to it. So 12/28: 43%


Conclusion: November has little bearing on the ensuing winter, in fact there's a slight negative correlation between November and DJF for the colder than average winters. In other words, if I had to choose a temp anomaly for November it would be warmer than normal. And for October colder than normal is the preferred state for the following winter."


4. +NAO in October. I'll post my entire research results in the winter outlook (I plan on issuing one, 2-3 weeks from now) but the general idea is there's a strong inverse correlation b/t the NAO modality in October and its state in the winter DJFM. In other words, most of the time a +NAO in October will precede a -NAO winter, and vice versa. The signal is even more pronounced in years when ENSO is weak (weak nina, neutral, or weak nino). Only two years since 1950 -- 1951 and 1976 -- did not follow the NAO correlation in weak ENSO years.


There's many other factors too (northern hemisphere snow cover evolution, solar constant..) but this is a large chunk of it. Just some food for thought as we monitor important trends the next 2-3 weeks.

[Ehop]

Great explanation.
Even an idiot like me understood the ideas, of course, measuring thr factors would leave me lost.
Appreciate the tme and thought process with data to make up your theory.
Thanks

[Virgaman]

Ehop, on Sep 30 2009, 01:00 PM, said:

Great explanation.
Even an idiot like me understood the ideas, of course, measuring thr factors would leave me lost.
Appreciate the tme and thought process with data to make up your theory.
Thanks

yes i agree, thank God for the maps Storm posted or i would be lost

[Stormchaser]

Ehop, on Sep 30 2009, 09:00 AM, said:

Great explanation.
Even an idiot like me understood the ideas, of course, measuring thr factors would leave me lost.
Appreciate the tme and thought process with data to make up your theory.
Thanks

Virgaman, on Sep 30 2009, 09:36 AM, said:

yes i agree, thank God for the maps Storm posted or i would be lost

Thanks guys, the next 3 weeks is always an interesting time of year as we watch some global signals fall into place. The SSTA profile needs some improvement as I posted above but I'm liking the progged pattern for October:


IF this verifies (of course a big IF as it's Day 8-10) -- this is precisely the pattern we want to see w/ below normal temps overall in the Northeast but a +NAO (no blocking), and favorable pacific signal.

[robbbs]

Storm -- Sorry, appreciate your time and efforts but it's the same old issue on long term outlooks -- simply a game of what has followed in the past, which is never absolute and often close to a toss up and ignores the issue that there are factors at play that are unknown and have a huge bearing on the evetual outcome. It's a game of chance and odds and so it doesn't matter what happens in October, November, or whatever, as every possible outcome from one side to the other occured in the past. If the best we can do is say, given the past, it's more likely this happens (maybe), it highlights the shortcomings of long term outlooks. Case in point is last December when multiple factors pointed to a "vodka cold" and brutal December and temps wound up above normal.

[jfar57]

Virgaman, on Sep 30 2009, 09:36 AM, said:

yes i agree, thank God for the maps Storm posted or i would be lost

Hey wait....are you agreeing that Ehop is an idiot? Thats not nice...

[satellite_eyes]

Thanks for the effort storm. It's always enjoyable to read and whether you are wrong or right i think reading what you have to say helps many on here to learn how things work.

To me it's also enjoyable to see what people's 'gut' feeling is. Just for fun, it doesn't always have to be serious and based on science. Maybe at some point we can have 1 thread just for that type of outlook.

[Stormchaser]

robbbs, on Sep 30 2009, 12:07 PM, said:

Storm -- Sorry, appreciate your time and efforts but it's the same old issue on long term outlooks -- simply a game of what has followed in the past, which is never absolute and often close to a toss up and ignores the issue that there are factors at play that are unknown and have a huge bearing on the evetual outcome. It's a game of chance and odds and so it doesn't matter what happens in October, November, or whatever, as every possible outcome from one side to the other occured in the past. If the best we can do is say, given the past, it's more likely this happens (maybe), it highlights the shortcomings of long term outlooks. Case in point is last December when multiple factors pointed to a "vodka cold" and brutal December and temps wound up above normal.

Robbbs,

I hear you, of course there will be shortcomings as weather patterns are non-linear thus many unexpected curveballs may be thrown our way, but does that mean we should simply give up and not attempt to further the long range accuracy? I believe there's plenty of sound scientific tools out there to get a decent grasp on the upcoming season with at least some regularity. You'd be surprised how many forecasters/meteorologists on Eastern subscribe to the long range and for some of the energy mets they make a living by providing LT forecasts to certain companies.

There was 7-8 page superthread about the long range last week and a couple of energy mets explained that an accuracy rate of 55% of higher w/ LT forecasts is considered good and would make you a decent amount of money. If you were to tell your boss that percentage it wouldn't fly for about 95% of jobs out there but meteorology is notorious as an inexact science, especially further out in time. The idea is that as long as you can "beat climatology" for a seasonal forecast, then you're doing pretty well.

Also with regards to the value of the monthly calls debate -- here's an interesting scenario: say your winter forecast from October calls for a -2 December and a +2 January, with the exspectation of the cold regime beginning around 12/10. But December comes and that pattern change doesn't show itself until 12/20, 10 days later than you thought. December ends up being a bust on monthly temps because the cold was not able to overcome the warmth. This same cold regime hangs on until 1/20 and now your January warm call is also busted all because of that pattern change occurring 10 days later than anticipated in December. But the cold regime DID come, right? That's what's most important as nailing the exact timing from 2-3 months away is even more difficult than getting the overall departure correct for winter. So in the end, if you are able to hit the overall temp for a season from months away, that is a great call IMO regardless of the monthlies as timing is the biggest problem w/ the long range.

So I agree with you that there are definite short-comings but there's also plenty of value. It's hit and miss but apparently if you can rack up even slightly more hits than misses, you're doing pretty good (55-60% accuracy rate). With my 2007-08 call I was able to get the monthlies and the overall correct and that's what we're striving for every time (even though getting the months right is significantly harder).

[icehater]

Stormchaser, on Sep 30 2009, 09:32 PM, said:

Robbbs,

I hear you, of course there will be shortcomings as weather patterns are non-linear thus many unexpected curveballs may be thrown our way, but does that mean we should simply give up and not attempt to further the long range accuracy? I believe there's plenty of sound scientific tools out there to get a decent grasp on the upcoming season with at least some regularity. You'd be surprised how many forecasters/meteorologists on Eastern subscribe to the long range and for some of the energy mets they make a living by providing LT forecasts to certain companies.

There was 7-8 page superthread about the long range last week and a couple of energy mets explained that an accuracy rate of 55% of higher w/ LT forecasts is considered good and would make you a decent amount of money. If you were to tell your boss that percentage it wouldn't fly for about 95% of jobs out there but meteorology is notorious as an inexact science, especially further out in time. The idea is that as long as you can "beat climatology" for a seasonal forecast, then you're doing pretty well.

Also with regards to the value of the monthly calls debate -- here's an interesting scenario: say your winter forecast from October calls for a -2 December and a +2 January, with the exspectation of the cold regime beginning around 12/10. But December comes and that pattern change doesn't show itself until 12/20, 10 days later than you thought. December ends up being a bust on monthly temps because the cold was not able to overcome the warmth. This same cold regime hangs on until 1/20 and now your January warm call is also busted all because of that pattern change occurring 10 days later than anticipated in December. But the cold regime DID come, right? That's what's most important as nailing the exact timing from 2-3 months away is even more difficult than getting the overall departure correct for winter. So in the end, if you are able to hit the overall temp for a season from months away, that is a great call IMO regardless of the monthlies as timing is the biggest problem w/ the long range.

So I agree with you that there are definite short-comings but there's also plenty of value. It's hit and miss but apparently if you can rack up even slightly more hits than misses, you're doing pretty good (55-60% accuracy rate). With my 2007-08 call I was able to get the monthlies and the overall correct and that's what we're striving for every time (even though getting the months right is significantly harder).

Which is why individual months are really not important to an energy company and why a collective 3 month period is. But that's also why amount of snowfall isn't important either, and certainly not timing those snowfalls and it's those specifics that make LT forecasts look bad as that timing is pot luck. What an energy company was always looking for was a way to manage inventory and production thus it sought out seasonal forecasts, not monthly components. It's much the same way in finance. I just projected out 3 months of cash for my business. I couldn't care less if I'm wrong on the individual months as long as the three month cume is what I expect. That's what happened to me in the second quarter. I was low low and high with cash calls. Dead wrong on each individual month but by quarter's end the high cancelled out the two lows and I was within an eyeblink of my forecast for the quarter.

[robbbs]

Stormchaser, on Oct 1 2009, 01:32 AM, said:

Robbbs,

I hear you, of course there will be shortcomings as weather patterns are non-linear thus many unexpected curveballs may be thrown our way, but does that mean we should simply give up and not attempt to further the long range accuracy? I believe there's plenty of sound scientific tools out there to get a decent grasp on the upcoming season with at least some regularity. You'd be surprised how many forecasters/meteorologists on Eastern subscribe to the long range and for some of the energy mets they make a living by providing LT forecasts to certain companies.

There was 7-8 page superthread about the long range last week and a couple of energy mets explained that an accuracy rate of 55% of higher w/ LT forecasts is considered good and would make you a decent amount of money. If you were to tell your boss that percentage it wouldn't fly for about 95% of jobs out there but meteorology is notorious as an inexact science, especially further out in time. The idea is that as long as you can "beat climatology" for a seasonal forecast, then you're doing pretty well.

Also with regards to the value of the monthly calls debate -- here's an interesting scenario: say your winter forecast from October calls for a -2 December and a +2 January, with the exspectation of the cold regime beginning around 12/10. But December comes and that pattern change doesn't show itself until 12/20, 10 days later than you thought. December ends up being a bust on monthly temps because the cold was not able to overcome the warmth. This same cold regime hangs on until 1/20 and now your January warm call is also busted all because of that pattern change occurring 10 days later than anticipated in December. But the cold regime DID come, right? That's what's most important as nailing the exact timing from 2-3 months away is even more difficult than getting the overall departure correct for winter. So in the end, if you are able to hit the overall temp for a season from months away, that is a great call IMO regardless of the monthlies as timing is the biggest problem w/ the long range.

So I agree with you that there are definite short-comings but there's also plenty of value. It's hit and miss but apparently if you can rack up even slightly more hits than misses, you're doing pretty good (55-60% accuracy rate). With my 2007-08 call I was able to get the monthlies and the overall correct and that's what we're striving for every time (even though getting the months right is significantly harder).

Storm -- This is about as close as you and I have ever come to agreeing with respect to long range forecasting. In fact, it would be difficult for me to find anything in your long post to disagree with. The only point I'll add is that if too much wiggle room is left on the issue of calendar months versus season, a met can easily use that to his advantage and convenience and then claim verification on what was really a bust. Consequently, the "delayed but not denied" and similar comments and rationalizations from you know who (and who has made such claims when the forecasted cold for December actually arrived in March and he claimed he verified). I would also question the value of a 55% hit ratio, which consequently leaves a 45% miss ratio. Even a 60/40 split is not much better than a coin flip. I'll also add and agree with ice's point about the futility of projected snowfall totals as part of long range outlooks, which often have no relation to a projected pattern call, and where one storm alone can verify or bust an outlook.

[icehater]

robbbs, on Sep 30 2009, 10:27 PM, said:

Storm -- This is about as close as you and I have ever come to agreeing with respect to long range forecasting. In fact, it would be difficult for me to find anything in your long post to disagree with. The only point I'll add is that if too much wiggle room is left on the issue of calendar months versus season, a met can easily use that to his advantage and convenience and then claim verification on what was really a bust. Consequently, the "delayed but not denied" and similar comments and rationalizations from you know who (and who has made such claims when the forecasted cold for December actually arrived in March and he claimed he verified). I would also question the value of a 55% hit ratio, which consequently leaves a 45% miss ratio. Even a 60/40 split is not much better than a coin flip. I'll also add and agree with ice's point about the futility of projected snowfall totals as part of long range outlooks, which often have no relation to a projected pattern call, and where one storm alone can verify or bust an outlook.

Think about that one Robbbs. If you turn out to be n accurate great forecaster and nail big-time patterns that are as repetritive as say the 95/96 winter was here, or if summer, say the pattern of the year with all the Mississippi flooding, would any such forecaster predict the level of snows seen here or rains seen in the midwest? I seriously doubt it. Even JB would have to have his head examined for predicting that level of extreme. But I doubt a forecaster would hesitate to predict record cold or coldest in 10-20 years, or heat in summer if he/she deeply believed in a pattern. When you really think about it L/T forecasts are really about temp calls vs average and the chances of them happening and precip calls in the same way. Predicting precip over a long period is hard enough without getting into percentage that is rain vs frozen. The more I think about it the more I realize NOAA's approach is the best approach as it's a pure climate call and any LT forecast is really about climate and not about specifics. They are heavily criticized for generality but those criticisms are really from weenies who think of LT forecasts the same way they think about 10 day forecasts.

[icehater]

robbbs, on Sep 30 2009, 10:27 PM, said:

Storm -- This is about as close as you and I have ever come to agreeing with respect to long range forecasting. In fact, it would be difficult for me to find anything in your long post to disagree with. The only point I'll add is that if too much wiggle room is left on the issue of calendar months versus season, a met can easily use that to his advantage and convenience and then claim verification on what was really a bust. Consequently, the "delayed but not denied" and similar comments and rationalizations from you know who (and who has made such claims when the forecasted cold for December actually arrived in March and he claimed he verified). I would also question the value of a 55% hit ratio, which consequently leaves a 45% miss ratio. Even a 60/40 split is not much better than a coin flip. I'll also add and agree with ice's point about the futility of projected snowfall totals as part of long range outlooks, which often have no relation to a projected pattern call, and where one storm alone can verify or bust an outlook.

Think about that one Robbbs. If you turn out to be n accurate/great forecaster and nail big-time patterns that are as repetritive as say the 95/96 winter was here, or if summer, say the pattern of the year with all the Mississippi flooding, would any such forecaster predict the level of snows seen here or rains seen in the midwest? I seriously doubt it. Even JB would have to have his head examined for predicting that level of extreme. But I doubt a forecaster would hesitate to predict record cold or coldest in 10-20 years, or heat in summer if he/she deeply believed in a pattern. When you really think about it L/T forecasts are really about temp calls vs average and the chances of them happening and precip calls in the same way. Predicting precip over a long period is hard enough without getting into percentage that is rain vs frozen. The more I think about it the more I realize NOAA's approach is the best approach as it's a pure climate call and any LT forecast is really about climate and not about specifics. They are heavily criticized for generality but those criticisms are really from weenies who think of LT forecasts the same way they think about 10 day forecasts.

[robbbs]

icehater, on Oct 1 2009, 03:02 AM, said:

Think about that one Robbbs. If you turn out to be n accurate/great forecaster and nail big-time patterns that are as repetritive as say the 95/96 winter was here, or if summer, say the pattern of the year with all the Mississippi flooding, would any such forecaster predict the level of snows seen here or rains seen in the midwest? I seriously doubt it. Even JB would have to have his head examined for predicting that level of extreme. But I doubt a forecaster would hesitate to predict record cold or coldest in 10-20 years, or heat in summer if he/she deeply believed in a pattern. When you really think about it L/T forecasts are really about temp calls vs average and the chances of them happening and precip calls in the same way. Predicting precip over a long period is hard enough without getting into percentage that is rain vs frozen. The more I think about it the more I realize NOAA's approach is the best approach as it's a pure climate call and any LT forecast is really about climate and not about specifics. They are heavily criticized for generality but those criticisms are really from weenies who think of LT forecasts the same way they think about 10 day forecasts.

Ice -- I have been saying exactly that for several years. NOAA's approach is the intellectualy honest one as it inherently recognizes the shortcomings of long range forecasting and, therefore, presents it in terms of less, neutral, or greater than equal chances of above or below versus climatology with respect to temps and precip. That's all one can really do given the current state of the science. If, for example, you take our limited knowledge and ascertain a 60% analog and other factor correlation, you then present the outlooks in exactly that format with those odds reflected. Anything beyond that is voodoo science IMO and not to be taken seriously.

[Stormchaser]

NOAA's guidance is what it is -- exactly that. Personally I have more respect for folks who make a forecast and take the challenge rather than using a probability scheme. Their forecast is such that it's very easy to verify as above normal can mean +0.5 to +10, below normal -0.5 to -10, etc. The most important reason for a LR forecast is to get the magnitude of the warm or cold correct IMO (and at the very least attempt it). You're not going to get very far with a job in LT forecasting by simply saying "40% chance the winter temps will be above the mean" or "equal chances" which essentially the scientific term for "I don't know." There are plenty of valuable tools out there other than straight climatology or ENSO climatology (if we wanted that, just look at normals for a specific location, why even bother forecasting). Remember NOAA forecasted above normal every winter after 2001 from 2002-2005 in the east based upon global warming/1990-present climatology and ended up getting burned each year. Straight climo doesn't cut it in most cases.

I just see their stance as on the other end of the extreme -- much too safe.

[robbbs]

Stormchaser, on Oct 1 2009, 04:23 AM, said:

NOAA's guidance is what it is -- exactly that. Personally I have more respect for folks who make a forecast and take the challenge rather than using a probability scheme. Their forecast is such that it's very easy to verify as above normal can mean +0.5 to +10, below normal -0.5 to -10, etc. The most important reason for a LR forecast is to get the magnitude of the warm or cold correct IMO (and at the very least attempt it). You're not going to get very far with a job in LT forecasting by simply saying "40% chance the winter temps will be above the mean" or "equal chances" which essentially the scientific term for "I don't know." There are plenty of valuable tools out there other than straight climatology or ENSO climatology (if we wanted that, just look at normals for a specific location, why even bother forecasting). Remember NOAA forecasted above normal every winter after 2001 from 2002-2005 in the east based upon global warming/1990-present climatology and ended up getting burned each year. Straight climo doesn't cut it in most cases.

I just see their stance as on the other end of the extreme -- much too safe.

Storm -- I'd agree with you on that but it doesn't work, and that's the point. NOAA recognizes that the science is not at the level that many of the private outlooks attempt to be at and, therefore, NOAA's outlooks at least carry a higher level of accuracy (still not where we'd like it to be, but certainly higher). If the more detailed outlooks have a success rate not any/much better than flipping a coin, they offer no value, except to those that put them out, usually for self serving interests -- more hits on a web site, more subscriptions sold, etc. I have posted a complement to you in the past which I will repeat again -- you are either ahead of the science or the science is behind your efforts. Either way, with the current state of the science, flipping a coin has about as much chance at verifying as your detailed outlooks have. Not meant to be taken as a slight to you, but simply reflective of the shortcomings of the science.

[icehater]

Stormchaser, on Oct 1 2009, 12:23 AM, said:

NOAA's guidance is what it is -- exactly that. Personally I have more respect for folks who make a forecast and take the challenge rather than using a probability scheme. Their forecast is such that it's very easy to verify as above normal can mean +0.5 to +10, below normal -0.5 to -10, etc. The most important reason for a LR forecast is to get the magnitude of the warm or cold correct IMO (and at the very least attempt it). You're not going to get very far with a job in LT forecasting by simply saying "40% chance the winter temps will be above the mean" or "equal chances" which essentially the scientific term for "I don't know." There are plenty of valuable tools out there other than straight climatology or ENSO climatology (if we wanted that, just look at normals for a specific location, why even bother forecasting). Remember NOAA forecasted above normal every winter after 2001 from 2002-2005 in the east based upon global warming/1990-present climatology and ended up getting burned each year. Straight climo doesn't cut it in most cases.

I just see their stance as on the other end of the extreme -- much too safe.

Much too safe or with far more common sense? What they are saying is that it's not predictable to great detail and large regions with any specificity so why bother and the LT calls I have seen private forecasters make the last 5 years bear them out, often in the first month of those long range forecasts, many of which were issue later then NOAA's outlook. Again I think a true LT forecast should be for 3-4 months of weather collectively and it should simply be a call for above (below) or well above (below) normal values of temp and precip. Once you get into timing periods within a 3-4 month outlook (and think about it these should be viewed as outlooks not forecasts as the latter is specific and the former is general) and hoping to nail patterns and relaxations of them you are going to go wrong and I think that's what leads folks to hope for a big snow event on the last day or two of a month or a big cold spells or warm-ups in the final week. Those hopes are simply for verification of a bunch of mistimed pattern reloads and relaxations or completely missed patterns. If you do a forecast for January calling fo normal temps, folks are pretty much expecting a whole month of normal weather, not one which is mild and above normal for 3 weeeks and then gets balanced out by a week of cold weather. A forecaster will say he verified whereas most folks would look at it as 75% or more wrongly called.

[jfar57]

No need to resurrect this debate this year for the most part, but there is one place that I think you guys need to consider in your respective points of view.

A. NOAA has a public responsibility to be as accurate and informative as possible without creating hysteria and irresponsible information. Also, given their "stature" need to be able to verify politically.

B. The smaller and less well known the forecasting entity, the greater the risk they can take in making brash forecasts. If they are right, they gain following (and whatever bounty that leads to). If they are wrong, who cares.

There has to be a difference between who forecasts what publicly and the work going on with the science to improve the accuracy of any forecast, be it 2 days, 2 weeks or 2 years. Storm, the work you and presumably others are doing is critical to advancing the science. Even if your investigations turn out to be 100% invalid the value to science is that they no longer have to be looked at. Thats what science is all about.

So, from my vantage point, Storm, you are arguing about the science of forecasting and working to improve it. Robbbs and Ice are pretty much reacting to the "art" of forecasting as bound by an entities need for public perception. To Ice's point early on a true LT forecast (as from NOAA) has the ability to influence many things. As such, needs to be fairly conservative.

[icehater]

jfar57, on Oct 1 2009, 01:31 PM, said:

No need to resurrect this debate this year for the most part, but there is one place that I think you guys need to consider in your respective points of view.

A. NOAA has a public responsibility to be as accurate and informative as possible without creating hysteria and irresponsible information. Also, given their "stature" need to be able to verify politically.

B. The smaller and less well known the forecasting entity, the greater the risk they can take in making brash forecasts. If they are right, they gain following (and whatever bounty that leads to). If they are wrong, who cares.

There has to be a difference between who forecasts what publicly and the work going on with the science to improve the accuracy of any forecast, be it 2 days, 2 weeks or 2 years. Storm, the work you and presumably others are doing is critical to advancing the science. Even if your investigations turn out to be 100% invalid the value to science is that they no longer have to be looked at. Thats what science is all about.

So, from my vantage point, Storm, you are arguing about the science of forecasting and working to improve it. Robbbs and Ice are pretty much reacting to the "art" of forecasting as bound by an entities need for public perception. To Ice's point early on a true LT forecast (as from NOAA) has the ability to influence many things. As such, needs to be fairly conservative.

Great points. I think the spread of the internet has increased the desires of LT forecasters to predict weather with greater specificity than they'd ever have done without an internet. No other media would have ever bothered to put up so many forecasters POV's. That's why many calls are made to look more like 15 day forecasts than a seaonal call and it's also why those forecasters then grope to all ends to find any proof that verifies or comes close top verifying their calls even though the basis for the call was lost in left field.

[NittanyLion]

icehater, on Oct 1 2009, 02:18 PM, said:

Great points. I think the spread of the internet has increased the desires of LT forecasters to predict weather with greater specificity than they'd ever have done without an internet. No other media would have ever bothered to put up so many forecasters POV's. That's why many calls are made to look more like 15 day forecasts than a seaonal call and it's also why those forecasters then grope to all ends to find any proof that verifies or comes close top verifying their calls even though the basis for the call was lost in left field.

Also why it gives meteorologists everywhere bad names.

[icehater]

NittanyLion, on Oct 1 2009, 03:27 PM, said:

Also why it gives meteorologists everywhere bad names.

True. Any science that pauses to advance someone's name over the science itself gets a bad rap for its resident pros. And IMO, most LT forecasters are after glory, not better science.