Major Hurricane Irma

[larrye]

34 minutes ago, donsutherland1 said:

Hopefully, any sensationalism in the media will be avoided and, where present, be ignored for the time being. It's way too soon to speculate on landfall details. At this time, the general point that there is the potential for a strong hurricane to make U.S. landfall is about as far as one can go. That potential may be somewhat higher than climatology given the forecast pattern, but landfall remains uncertain. Outcomes, including scenarios where Irma would recurve safely away from the CONUS, remain on the table. By the middle of next week, details including whether landfall is likely, possible locations, and Irma's strength could begin to fall into place with a higher degree of confidence.

I dunno. The media's already been hyping it. Just this morning, CNN asked a rescue official in TX if he's ready for Irma. The media in this country is out for one thing - ratings.

[Eskimo Joe]

NHC advises no recon until 9/3, it's a long weekend.  Let's take a break.and enjoy some beer and football.

[olafminesaw]

9 minutes ago, larrye said:

I dunno. The media's already been hyping it. Just this morning, CNN asked a rescue official in TX if he's ready for Irma. The media in this country is out for one thing - ratings.

Thank goodness the Euro only goes out to hour 240.

[NJwx85]

1 minute ago, olafminesaw said:

Thank goodness the Euro only goes out to hour 240.

You still have the lower skilled GFS run four times a day out to 384hrs for clickbait.

[NJwx85]

12 minutes ago, larrye said:

I dunno. The media's already been hyping it. Just this morning, CNN asked a rescue official in TX if he's ready for Irma. The media in this country is out for one thing - ratings.

That idiot on CNN probably thinks the disturbance in the Bay of Campeche is Irma.

[harrisale]

Vis look at Irma this morning. Looks as though some drier air has wrapped around the north side of the circulation.

[NJwx85]

Just now, harrisale said:

Vis look at Irma this morning. Looks as though some drier air has wrapped around the north side of the circulation.

Negative, Microwave imagery confirms an ERC.

[RU848789]

1 hour ago, LakeEffectKing said:

I think it's worth reiterating....models are tools for meteorologists.  And have great value, especially in the shorter ranges.  But at lead times of 8-10 days, they ALL have much less value just because of time.  The synoptic setup at those leads, and the forecasting abilities (and thus value) of any model are greatly diminished.  I don't care if it's the Euro, NAVGEM, GFS....all are very susceptible to large errors at 10 days out....especially with the fragile conditions that are required to get a LF north of Hatteras.  Fish or EC LF....all are equally on the table, and will be for several more days.

Great post.  I'm sure you know this, but lots of new folks come to threads like this, so I'll elaborate a bit.  The fundamental reason for the large errors beyond 5-6 days for hurricane forecasts or any forecast, for that matter, lies in chaos theory, as it applies to non-linear deterministic systems, such as the weather.  Chaos essentially means that the small errors in initial and boundary conditions fed into the numerical models for weather simulation will propagate and grow ever larger, in time, such that the model spread in solutions becomes large beyond 5-6 days and massive beyond 8-9 days.  

The effect of chaos can actually be seen in ensemble runs, where stochastic (random) perturbations in initial and boundary conditions are fed into the "base" run of any model to see the effects, in time, of these perturbations (simulating small initial errors).  The ensemble runs show huge variations as time goes on, which basically shows that the range of possible outcomes beyond 8-9 days (or even beyond 5-6 days) is so large as to be not very useful.  It's why the NWS doesn't make deterministic forecasts beyond 7 days and why the NHC doesn't show tracks beyond 5 days - the uncertainty is just too high.  It's also why those 15+ day "forecasts" from AccuWeather and others are ****e.  

As an aside, this phenomenon is often known better as the "Butterfly Effect" a term coined by Lorenz back in the early 60s when he first published his work on Chaos Theory - he showed that small initialization errors in numerical weather models can propogate so much that a butterfly flapping its wings at t=0 can impact the outcome of a tornado weeks later. Kinda cool stuff...

https://en.wikipedia.org/wiki/Butterfly_effect

[Kevin Reilly]

55 minutes ago, larrye said:

I dunno. The media's already been hyping it. Just this morning, CNN asked a rescue official in TX if he's ready for Irma. The media in this country is out for one thing - ratings.

All about the money

[harrisale]

13 minutes ago, NJwx85 said:

Negative, Microwave imagery confirms an ERC.

Not saying you're wrong but the storm is definitely moving into an area of mid-level dry air, as mentioned by NHC, which may hold back Irma from strengthening and possibly weakening after this ERC.

[olafminesaw]

8 minutes ago, RU848789 said:

Great post.  I'm sure you know this, but lots of new folks come to threads like this, so I'll elaborate a bit.  The fundamental reason for the large errors beyond 5-6 days for hurricane forecasts or any forecast, for that matter, lies in chaos theory, as it applies to non-linear deterministic systems, such as the weather.  Chaos essentially means that the small errors in initial and boundary conditions fed into the numerical models for weather simulation will propagate and grow ever larger, in time, such that the model spread in solutions becomes large beyond 5-6 days and massive beyond 8-9 days.  

The effect of chaos can actually be seen in ensemble runs, where stochastic (random) perturbations in initial and boundary conditions are fed into the "base" run of any model to see the effects, in time, of these perturbations (simulating small initial errors).  The ensemble runs show huge variations as time goes on, which basically shows that the range of possible outcomes beyond 8-9 days (or even beyond 5-6 days) is so large as to be not very useful.  It's why the NWS doesn't make deterministic forecasts beyond 7 days and why the NHC doesn't show tracks beyond 5 days - the uncertainty is just too high.  It's also why those 15+ day "forecasts" from AccuWeather and others are ****e.  

As an aside, this phenomenon is often known better as the "Butterfly Effect" a term coined by Lorenz back in the early 60s when he first published his work on Chaos Theory - he showed that small initialization errors in numerical weather models can propogate so much that a butterfly flapping its wings at t=0 can impact the outcome of a tornado weeks later. Kinda cool stuff...

https://en.wikipedia.org/wiki/Butterfly_effect

Something I've always wondered: Why do ensembles tend to group together so much, if the purpose of ensembles is to simulate chaos? Is it because there is an error (or errors) they're all missing which might have huge downstream implications?

[bdgwx]

A blend of the 0Z EPS and GEFS puts Irma at 23N/66W at HR 168 which is due north of Puerto Rico. The spread between the two is about 280 miles. Going out to HR 216 a blend puts Irma just to the northwest of the Bahamas with a spread between the two of no more than 300 miles. That's really not bad at all considering we're talking about a 9 day forecast.

 

[Kevin Reilly]

16 minutes ago, harrisale said:

Vis look at Irma this morning. Looks as though some drier air has wrapped around the north side of the circulation.

Yep should hold its own would not be surprised if it weakens a tad.   It's feeling the ridge to the north which will push Irma west southwest. 

[SN_Lover]

3 minutes ago, Kevin Reilly said:

Yep should hold its own would not be surprised if it weakens a tad.   It's feeling the ridge to the north which will push Irma west southwest. 

It has weakened

 

[NJwx85]

13 minutes ago, harrisale said:

Not saying you're wrong but the storm is definitely moving into an area of mid-level dry air, as mentioned by NHC, which may hold back Irma from strengthening and possibly weakening after this ERC.

Go back and look at the same graphic from three days ago and you'll see the same dry air in the same place. I'm not saying it couldn't have ingested a little dry air, but like I said the Microwave imagery shows a larger eye taking over. I'd post it but it keeps showing an old image from yesterday instead.

[NJwx85]

9 minutes ago, SN_Lover said:

It has weakened

 

That's standard during an ERC. It's not unusual to see a CAT 4 drop down to even a CAT 2 during one of these.

[NJwx85]

000
WTNT31 KNHC 011449
TCPAT1

BULLETIN
Hurricane Irma Advisory Number   9
NWS National Hurricane Center Miami FL       AL112017
1100 AM AST Fri Sep 01 2017

...IRMA FORECAST TO REMAIN A POWERFUL HURRICANE FOR DAYS...


SUMMARY OF 1100 AM AST...1500 UTC...INFORMATION
-----------------------------------------------
LOCATION...18.5N 37.8W
ABOUT 1580 MI...2540 KM E OF THE LEEWARD ISLANDS
MAXIMUM SUSTAINED WINDS...110 MPH...175 KM/H
PRESENT MOVEMENT...WNW OR 285 DEGREES AT 13 MPH...20 KM/H
MINIMUM CENTRAL PRESSURE...972 MB...28.71 INCHES


WATCHES AND WARNINGS
--------------------
There are no coastal watches or warnings in effect.


DISCUSSION AND 48-HOUR OUTLOOK
------------------------------
At 1100 AM AST (1500 UTC), the center of Hurricane Irma was located
near latitude 18.5 North, longitude 37.8 West. Irma is moving toward
the west-northwest near 13 mph (20 km/h).  A turn toward the west is
expected by tonight, followed by a turn toward the west-southwest on
Saturday.

Maximum sustained winds are near 110 mph (175 km/h) with higher
gusts.  Fluctuations in strength, up or down, are possible during
the next few days, but Irma is expected to remain a powerful
hurricane through the weekend.

Hurricane-force winds extend outward up to 15 miles (30 km) from the
center and tropical-storm-force winds extend outward up to 90 miles
(150 km).

The estimated minimum central pressure is 972 mb (28.71 inches).


HAZARDS AFFECTING LAND
----------------------
None


NEXT ADVISORY
-------------
Next complete advisory at 500 PM AST.

$$
Forecaster Blake

 

[NJwx85]

000
WTNT41 KNHC 011450
TCDAT1

Hurricane Irma Discussion Number   9
NWS National Hurricane Center Miami FL       AL112017
1100 AM AST Fri Sep 01 2017

Irma is undergoing an eyewall replacement cycle.  The small eye is
becoming less distinct, with both microwave and visible imagery
indicating the presence of a forming outer eyewall. Satellite
intensity estimates are lower, so the initial wind speed is reduced
to 95 kt.  Some further weakening is possible over the next day or
so while Irma moves over marginally warm SSTs and continues the
eyewall replacement.  After that time, the environment should be
generally conducive for some restrengthening, although forecasting
the timing of eyewall replacement cycles is next to impossible.  The
biggest change from yesterday are the long-term wind shear
predictions from the global models, which shows a little more shear.
Still, the shear is not that strong, and the hurricane will be
moving over 29C SSTs.  Thus, the NHC intensity forecast is reduced
somewhat from the previous one, but remains near or slightly above
the model consensus.

Irma has turned a little bit to the left, now estimated to be
moving 285/11.  The general synoptic situation remains well
established due to a building mid-level high, which should cause
the hurricane to turn westward later today and then move
west-southwestward through the weekend.  An upper-level low will be
dropping southward on the east side of that high, and should be a
key feature to how far south Irma goes before eventually turning
westward and west-northwestward early next week.  There is a
noticeable clustering of guidance by day 5, with the ECMWF, HWRF and
corrected-consensus models to the south, and the UKMET, GFS, CTC and
HMON to the north. Since Irma is forecast to be a vertically deep
cyclone, it seems more likely to respond to the northerly flow from
the upper-level low, which leads me to believe the track will be on
the southern side of the guidance.  Thus, the forecast will stay
similar to the previous one, very close to the southern cluster
mentioned above.

[yoda]

The biggest change from yesterday are the long-term wind shear
predictions from the global models, which shows a little more shear.
Still, the shear is not that strong, and the hurricane will be
moving over 29C SSTs.  Thus, the NHC intensity forecast is reduced
somewhat from the previous one, but remains near or slightly above
the model consensus.

[NJwx85]

3 minutes ago, yoda said:

Hurricane Irma Discussion Number   9
NWS National Hurricane Center Miami FL       AL112017
1100 AM AST Fri Sep 01 2017

Irma is undergoing an eyewall replacement cycle.  The small eye is
becoming less distinct, with both microwave and visible imagery
indicating the presence of a forming outer eyewall. Satellite
intensity estimates are lower, so the initial wind speed is reduced
to 95 kt.  Some further weakening is possible over the next day or
so while Irma moves over marginally warm SSTs and continues the
eyewall replacement.  After that time, the environment should be
generally conducive for some restrengthening, although forecasting
the timing of eyewall replacement cycles is next to impossible.  The
biggest change from yesterday are the long-term wind shear
predictions from the global models, which shows a little more shear.
Still, the shear is not that strong, and the hurricane will be
moving over 29C SSTs.  Thus, the NHC intensity forecast is reduced
somewhat from the previous one, but remains near or slightly above
the model consensus.

Irma has turned a little bit to the left, now estimated to be
moving 285/11.  The general synoptic situation remains well
established due to a building mid-level high, which should cause
the hurricane to turn westward later today and then move
west-southwestward through the weekend.  An upper-level low will be
dropping southward on the east side of that high, and should be a
key feature to how far south Irma goes before eventually turning
westward and west-northwestward early next week.  There is a
noticeable clustering of guidance by day 5, with the ECMWF, HWRF and
corrected-consensus models to the south, and the UKMET, GFS, CTC and
HMON to the north. Since Irma is forecast to be a vertically deep
cyclone, it seems more likely to respond to the northerly flow from
the upper-level low, which leads me to believe the track will be on
the southern side of the guidance.  Thus, the forecast will stay
similar to the previous one, very close to the southern cluster
mentioned above.


FORECAST POSITIONS AND MAX WINDS

INIT  01/1500Z 18.5N  37.8W   95 KT 110 MPH
 12H  02/0000Z 18.7N  39.5W   90 KT 105 MPH
 24H  02/1200Z 18.5N  41.8W   90 KT 105 MPH
 36H  03/0000Z 18.0N  44.2W   95 KT 110 MPH
 48H  03/1200Z 17.3N  46.6W  100 KT 115 MPH
 72H  04/1200Z 16.2N  51.0W  105 KT 120 MPH
 96H  05/1200Z 16.7N  55.2W  110 KT 125 MPH
120H  06/1200Z 18.0N  59.5W  110 KT 125 MPH

$$
Forecaster Blake

Part of that is because the track has shifted slightly to the North. If Irma takes a path closer to the islands there will be less shear.

[bdgwx]

Looking at the last 4 model guidance spaghetti plots the official NHC forecast is consistently south of TVCN. Of the TVCN members HWRF is well south and COAMPS is well north so they're probably cancelling each other out. EPS ensemble strongly favors the official forecast over TVCN right now. This would make COAMPS an extreme outlier if the best track happens to verify the EPS. Normally I would say be careful about deviating too much from TVCN, but there's a lot to be said for the EPS so I can see why the NHC is staying south.

[NJwx85]

Through 24hrs on the 12z GFS the positioning is identical to 00z.

[LandofLincoln]

Not that this is a surprise or anything, but just starting the fact... GFS shows this continued weakness until late tonight/early tomorrow morning, when it starts to take that southern movement.

[NJwx85]

Through 48hrs, track is identical to 00z, a tick, and the slightest tick faster.

[Tatamy]

1 hour ago, Eskimo Joe said:

NHC advises no recon until 9/3, it's a long weekend.  Let's take a break.and enjoy some beer and football.

NOUS42 KNHC 011427
REPRPD
WEATHER RECONNAISSANCE FLIGHTS
CARCAH, NATIONAL HURRICANE CENTER, MIAMI, FL.
1030 AM EDT FRI 01 SEPTEMBER 2017
SUBJECT: TROPICAL CYCLONE PLAN OF THE DAY (TCPOD)
         VALID 02/1100Z TO 03/1100Z SEPTEMBER 2017
         TCPOD NUMBER.....17-092

I.  ATLANTIC REQUIREMENTS
    1. NEGATIVE RECONNAISSANCE REQUIREMENTS.
    2. SUCCEEDING DAY OUTLOOK: P-3 TAIL DOPPLER RADAR MISSIONS
       ON HURRICANE IRMA EVERY 12 HOURS STARTING AT 03/19Z.
    3. ADDITIONAL DAY OUTLOOK:
       A. BEGIN 6 HRLY FIX MISSIONS ON HURRICANE IRMA AT 04/1800Z
          NEAR 16.4N 52.5W.
       B. A G-IV SYNOPTIC SURVEILLANCE MISSION AROUND IRMA
          WITH TAKEOFF OF 04/1730Z.

[yoda]

Irma still moving WSW at hr 60

[NJwx85]

Just now, yoda said:

Irma still moving WSW at hr 60

It's hair NW and faster than the 00z run at that hour.

[LandofLincoln]

3 minutes ago, yoda said:

Irma still moving WSW at hr 60

Looks like it still is at 72, as well.

[JC-CT]

1 hour ago, olafminesaw said:

Something I've always wondered: Why do ensembles tend to group together so much, if the purpose of ensembles is to simulate chaos? Is it because there is an error (or errors) they're all missing which might have huge downstream implications?

They usually don't at the lead times they are useful. But important missing pieces of the initialization puzzle will always trump any attempt to perturb known initialization data.

[yoda]

W or slightly WNW movement is beginning for Irma at hr 84