Hurricane Maria: NNW Of Bermuda Hurricane Watch Issued

[phil882]

Not going to happen. First, we have an agreement with the ECMWF for limited data exchange (we do not pay to receive ECMWF data, but only receive a specific subset of their earliest output with agreement not to distribute; or something along those lines). Additionally, I don't believe that creating an operational product (model) based on something that is completely out of their control (non operational, non-NOAA) is something that is going to happen.

There are technical issues involved as well, such as data timeliness, the necessary training data (for the statistical regression) [this would be especially important for anytime they would do an upgrade], and finally, they (ECMWF) would have to give us permission to create such a product/model.

Cool, thanks for the information! Interesting to hear that the NWS (and NOAA) is actually not paying for ECWMF but instead is in a data sharing agreement. Perhaps they realize the importance of NOAA's POES and other less routine obs such as G-IV data access during tropical cyclones.

So much for my idea of E-SHIPS

[Srain]

12Z...

[Juliancolton]

Recent view of the TD, which, as mentioned above, should be a TS at 11.

[Srain]

Latest ASCAT suggests the 'center' is further S than what guidance and best track suggests. It will be interesting to see where the NHC places the center and if they upgrade based on that pass.

[dtk]

Cool, thanks for the information! Interesting to hear that the NWS (and NOAA) is actually not paying for ECWMF but instead is in a data sharing agreement. Perhaps they realize the importance of NOAA's POES and other less routine obs such as G-IV data access during tropical cyclones.

So much for my idea of E-SHIPS

Despite the "competition", I think most of the operational centers have pretty good working relationships.

I think that an E-SHIPS thing could probably be done by someone, in a research/experimental/demonstration type mode, but a path to becoming an operational entity would be difficult. I think the biggest obstacle would be working out some sort of agreement with ECMWF (since in a way, it would be a sort of redistribution/repackaging of their product).

[Srain]

Maria is trekking along rather fast to the W. It's also interesting to note that the NHC no longer is expecting Maria to reach Hurricane strength as it makes an approach on the SE Bahamas...

[eyewall]

FAIL: MSNBC just mentioned Maria and said "you are looking at it right there". The disturbance in the Southern Gulf was on the screen.

[GaWx]

Some historical climatology data with respect to the region in which TD 14 formed (11.8N 37.0W):

Since 1950, 8 tropical cyclones formed in the region bounded by 9.8N-13.8N/34.0W-40.0W on September 6 +/- 20 days. A day outside that range was Hurricane Chris (1988), which peaked at Category 1 and recurved and Hurricane Andrew (1992), which made U.S. landfall as a devastating Category 5 hurricane.

Of the eight relevant storms, 3 (37.5%) made U.S. landfall. Those storms were Inez (1966), David (1979), and Frances (2004). 5 (62.5%) avoided U.S. landfall. The landfall percentage is 50% higher than that for all tropical cyclones that formed in the 34W-40W region on September 6 +/- 20 days. IMO, from that pool of eight storms, there are five analog candidates at this time (shown chronologically):

1. Flossie (1978)

2. David (1979)

3. Danielle (1998) **the solution closest to that shown on the 9/6 18z GFS**

4. Erin (2001)

5. Frances (2004)

I haven't done an extensive study of heading within 35-40W like I did for 25-30W. However, for the fun of it, I looked at the 8 mentioned by Don here. Out of these, the three with the lowest heading within 35-40W were David of 1979 (270 deg.), Frances of 2004 (274 deg.), and Andrew of 1992 (282 deg.). All three hit the US. Frances made it to 85.20W, David made it to 81.20W, and Andrew made it to 91.7W.

The other that hit was Inez of 1966 (293 deg.), which made it to 100.6W. So, the four hitters averaged 280 deg. and made it to 90W on average.

The four that missed were as follows: Chris of 1994 (283 deg./made it to 63.1W), Flossie of 1978 (315 deg./made it to 62.0W), Danielle of 1998 (284 deg/made it to 74.3W), and Erin of 2001 (284 deg./made it to 65.9W). So, the four missers averaged 292 deg. vs. the 280 deg. of the hitters, and they made it to only 66W on average vs. the 90W of the four hitters.

1) The results, though based on a fairly small sample, are intuitive and at least suggest a partial negative correlation between heading and how far west they ultimately get.

2) TD Maria had a heading of ~276 deg. in that area. So, these results at least suggest that it is still too early to call Maria a safe recurve east of the U.S. even if the odds may favor a recurve right now based on model consensus (trough likely not lifting out in time). Of course, dissipation in the tropics (before it even has a chance to recurve or hit) also has to be considered a decent possibility.

[donsutherland1]

Even as TD14 grew into Tropical Storm Maria, I have few changes from my initial thoughts. Odds somewhat favor recurvature, but the probability of U.S. landfall is greater than it was with Katia. Moreover, Maria could ultimately peak as a major hurricane (probably Category 3).

With respect to the storm's development, several quick thoughts are in order:

1. Development for the time being will likely be slow. Maria will be battling shear. Added to that, its forward motion is enhancing the impact of the shear (greater relative shear).

2. The shear impacting Maria should begin to relax after 72-96 hours.

3. Maria's forward speed should also be slowing at a time when the shear is beginning to relax, hence the relative shear will fall more quickly than the absolute shear.

Given those thoughts, I expect Maria will likely remain a tropical storm through 96 hours (perhaps a fairly weak one for the next 48-72 hours). After 72-84 hours steady development could commence. Given the difficulty in intensity forecasting, there is, of course, a lot of uncertainty. Nonetheless, I expect historic climatology associated with storms that form in the region in which Maria formed to provide some useful insight with respect to its peak strength. In addition, three models are indicating a dramatic ramping up of Maria's intensity after 84-96 hours.

Finally, in term's of Maria's track, the 9/7 12z GFS favors something resembling the track taken by Danielle (1998). Earlier runs of the GFS have been very consistent with such a track.

[nrgjeff]

Like very slow strengthening forecast due to wind shear in place. Westerly winds aloft may actually increase relative to Maria in a day or so as it tracks into an area of increasing shear. Plus the forward motion is so fast it creates shear from below by exacerbating the overall shear issue. Relative to the storm, add 20 kt to that westerly shear. The shear should greatly relax toward the Bahamas, leaving time for some strengthening.

However, I’m not too worried about the East Coast. General broad troughing forecast to remain over the East next week by global models, GFS, ECMWF, Canadian; so, would look for recurve if those are correct. For the sake of the drenched East let’s hope this is a fish storm.

Shear charts (current and tendency) from CIMSS: http://tropic.ssec.wisc.edu/

[phil882]

In the last few frames on visible you can see the center becoming exposed to the southeast of the convection near 13N and 43W. This was again completely expected as the low level flow is just accelerating too quickly for the upper level flow to catch up. The biggest question I think that needs to be answer for Maria is will it survive, not whether or not it will re-curve.

[bobbutts]

from the

Wiki Page on Danielle (1998)

The storm was generally well-forecast by the National Hurricane Center, with a low degree of errors compared to average; therefore, the storm's recurvature away from land was anticipated well in advance

and from the nhc at 11

MOST OFTHE TRACK MODELS...WHICH ARE IN EXCELLENT AGREEMENT.

wondering if this storm will feature higher than normal confidence track forecasts throughout it's life

[GaWx]

Even as TD14 grew into Tropical Storm Maria, I have few changes from my initial thoughts.

Don,

Would you please tell me the two nonhitting storms that you didn't list from the eight relevant ones in your other post? TIA

[donsutherland1]

Don,

Would you please tell me the two nonhitting storms that you didn't list from the eight relevant ones in your other post? TIA

GaWx,

I'll have to get the names tonight when I get home. I don't have the list with me. I'll post the full list this evening in this thread.

[Srain]

In the last few frames on visible you can see the center becoming exposed to the southeast of the convection near 13N and 43W. This was again completely expected as the low level flow is just accelerating too quickly for the upper level flow to catch up. The biggest question I think that needs to be answer for Maria is will it survive, not whether or not it will re-curve.

Interestingly this is exactly what the Euro was suggesting when that model brought the weak 850mb vort into the Eastern Caribbean and on W. Just perhaps this will end up nothing more than a depression (or less) further W than many are speculating.

[phil882]

Interestingly this is exactly what the Euro was suggesting when that model brought the weak 850mb vort into the Eastern Caribbean and on W. Just perhaps this will end up nothing more than a depression (or less) further W than many are speculating.

I'm in my office right now, and I just had a nice chat with another grad student who is doing his PhD. work on Kelvin Wave interactions with Tropical Cyclones. He actually thinks that conditions in the 48-72 hour period look quite favorable for development from a kelvin wave standpoint. While the storm is under the suppressed phase of the kelvin wave currently (which is causing the current degraded satellite presentation) by the time the system approaches the Lesser Antilles, he thinks we should see much improved conditions.

[donsutherland1]

GaWx,

The two recurving storms that were excluded were Frances (1976) and Hortense (1990).

[GaWx]

GaWx,

The two recurving storms that were excluded were Frances (1976) and Hortense (1990).

Thanks, Don.

Between 35 and 40W, Frances of 1976 moved at 289 deg/made it to 55.2W, while Hortense of 1990 moved at 291 deg./made it to 51.7W. Now, I'll incorporate these two storms into my prior analysis:

I looked at the 10 storms that Don analyzed that developed near where Maria formed. Out of these, the three with the lowest heading within 35-40W were David of 1979 (270 deg.), Frances of 2004 (274 deg.), and Andrew of 1992 (282 deg.). All three hit the US. Frances made it to 85.2W, David made it to 81.2W, and Andrew made it to 91.7W.

The other that hit was Inez of 1966 (293 deg.), which made it to 100.6W. So, the four hitters averaged 280 deg. and made it to 90W on average.

The six that missed were as follows: Chris of 1994 (283 deg./made it to 63.1W), Danielle of 1998 (284 deg/made it to 74.3W), Erin of 2001 (284 deg./made it to 65.9W), Frances of 1976 (289 deg/made it to 55.2W), Hortense of 1990 (291 deg./made it to 51.7W), and Flossie of 1978 (315 deg./made it to 62.0W). So, the six missers averaged 291 deg. vs. the 280 deg. of the hitters, and they made it to only 62W on average vs. the 90W of the four hitters.

1) The results, though based on a fairly small sample, are intuitive and at least suggest a partial negative correlation between heading and how far west they ultimately get.

2) TD Maria had a heading of only ~276 deg. in that area, which is lower than 8 of the 10 in the study. So, these results at least suggest that it is still too early to call Maria a safe recurve east of the U.S. even if the odds may favor a recurve right now based on model consensus (due to trough being modeled to not lift out in time). With Maria staying south of many of the model tracks and when considering this analysis, it remains to be seen if it will actually recurve east of the CONUS. The Euro, which has had a further south track than the GFS in the W ATL, has come very close to hitting the U.S. Of course, dissipation in the tropics (before it even has a chance to recurve or hit) also has to be considered a decent possibility due mainly to shear.

[OHweather]

Tropical Storm Maria Forecast/discussion 1:

Tropical storm Maria is taking on the appearance of a sheared tropical cyclone. Outflow is significantly restricted on the southwestern side of the storm and the center is removed from the deepest convection. Infrared satellite imagery shows a consistent area of deep convection with cold cloud tops of -70 to -80C, however the most recent microwave images reveal that the center is well removed from the convection. Given this, Maria is not expected to intensify much in the near term and is clearly dealing with some shear.

Maria is currently on the southern edge of a large subtropical ridge, which is steering the tropical storm briskly to the west. The 21z motion estimate from the NHC is 275/20. This general motion is expected to continue for the next 48 or so hours as the low to mid level ridging remains quite strong to the north of Maria. In the short term, higher than normal heights over the western US/Canada combined with Lee's remnants will contribute to below normal heights over the eastern US. In addition, a +NAO will cause ridging over the central and western Atlantic to be rather weak. Because of this, Maria is expected to begin gradually turning WNW and then NW between 48 and 120 hours out.

It is too early to tell for sure which land areas may be impacted, however most global models take Maria through the Lesser Antilles and perhaps near Puerto Rico. All global models and ensembles, except for the notable exception of the European Ensembles brought Maria north of Hispaniola their 12z run, so a track through the northern Lesser Antilles, near Puerto Rico but well north of Hispaniola will be shown. From here the cyclone should continue gradually tuning to the right, as the heart of the ridging will be located rather far south, southeast of Bermuda, which will steer the cyclone more and more to the right, more than likely well off the US east coast and quite possibly north/east of the Bahamas as well. In addition, the 12z GFS and ECM ensembles both agreed on slowly decreasing heights over the Aleutians and increasing heights over the Greenland/Davis Strait area more than 5 days out, forcing an unseasonably strong longwave trough over the eastern US. This combination of weak ridging over the central/western Atlantic on top of a potentially rather persistent eastern US trough will make landfall highly unlikely along the east coast.

In the short term, the upper level environment will not favor much strengthening of Maria. The low level easterlies are rather strong to the south of the subtropical ridging steering Maria, at 20-30 kts. In addition, an upper level low to the northwest of the cyclone is causing strong southwesterly upper level winds overtop of Maria, creating moderate to strong southwesterly shear over the tropical storm. Not only is Maria being sheared right now, noted by the pulsing pattern exhibited by the convection displaced to the east of the low level center, but the water vapor also shows some dry air to the north of the storm. This will probably attempt to work into the circulation in the next day or two. So, the environment around Maria does not support much if any intensification over at least the next 48 hours. In fact, weakening is more likely than intensification in the short term, and will show Maria weakening to a 35 kt tropical storm in the short term, but will keep her as a TS for now as Maria is still firing some deep convection over the eastern quadrant. After 72 hours out, the GFS and ECM both weaken the upper low that is contributing fairly strong shear over Maria at the moment, and the GFS especially tried to develop an upper-level anti-cyclone over Maria by day 5. Maria will be moving over fairly warm waters days 3-5, so if it survives in the short term some stregnthening will be possible down the road, and will show this in the forecast.

Forecast track:

Intensity:

Initial...45 knots

12 hours (6z Thursday)...45 knots

24 hours (18z Thursday)...40 knots

36 hours (6z Friday)...35 knots

48 hours (18z Friday)...35 knots

72 hours (18z Saturday)...40 knots

96 hours (18z Sunday)...50 knots

120 hours (18z Monday)...60 knots

[donsutherland1]

Through 10 pm, Maria has been tracking rapidly westward. On account of a combination of its rapid forward speed and continuing shear, it has remained a relatively weak tropical storm. Despite the fairly hostile situation, it has not yet shown signs of fizzling.

Over the next 60-72 hours, little change in strength, and perhaps even a small amount of weakening, is possible. However, after 72 hours, the shear should be relaxing and Maria should be decelerating. As a result, it should begin to strengthen.

Down the road, Maria should recurve (my guess is that is a 67% probability). There is strong model support for recurvature, with many models turning Maria away from the U.S. not too far from the vicinity where Katia will recurve. As had been the case with Katia, the UKMET and ECMWF models have been farthest west. Although, the 12z ECMWF brings Maria west of 75W, I don’t believe Maria will get that far west. I do believe it will track a little closer to the U.S. than Katia will. A compromise between the ECMWF and GFS appears to offer a reasonable idea. The 18z AEMI consensus is perhaps closest to that kind of compromise track.

Track Estimates:

15.0N 55.0W

20.0N 66.4W

25.0N 72.5W

30.0N 72.0W

35.0N 69.0W

[donsutherland1]

With the 9/8 0z Euro coming into line with most of the other guidance on Maria, I have little change in my thoughts expressed above. More than likely, the 9/8 6z GFS is presenting an unlikely scenario, perhaps one that is resulting, in part, from the interaction it is forecasting between Nate (now taking it to the Gulf Coast when the modeling had turned against that idea) and Maria. For now, I don't think its overall scenario, which includes the forecast tracks of Maria and Nate, appears very likely.

[phil882]

Talk about disorganized... I'm not sure its going to make it through the day if this appearance doesn't improve soon. Recon is flying today, so we should no know this afternoon.

[Ed Lizard]

Talk about disorganized... I'm not sure its going to make it through the day if this appearance doesn't improve soon. Recon is flying today, so we should no know this afternoon.

Morning NHC disco did mention some doubts it had a closed circulation. Euro has it fairly weak through 5 days and still recurves it.

[Srain]

Morning NHC disco did mention some doubts it had a closed circulation. Euro has it fairly weak through 5 days and still recurves it.

The discussion mentioned that the trough may be weaker than progged and a turn to the NW as forecast may need to be watched as well. Maria looks sickly the morning. I'm wondering if RECON will even fly.

[donsutherland1]

The discussion mentioned that the trough may be weaker than progged and a turn to the NW as forecast may need to be watched as well. Maria looks sickly the morning. I'm wondering if RECON will even fly.

Given the storm's look, it's probably little more than a TD, if that. It will be interesting to see whether it regenerates should it deteriorate into a tropical wave.

[Eyewall2005]

12z runs do look rather interesting... Could be an open wave soon though.

[40/70 Benchmark]

The fact that Maria seems to be nearly flat lining def. lends some creedence to the western side of the track envelope....we'll see.

[MJO812]

11am update

Westward movement

[SnowGoose69]

This is what kept occurring in 2004. Tons of shear in that portion of the tropics and or strong easterlies which kept allowing storms to sneak well west.

[Eyewall2005]

This is what kept occurring in 2004. Tons of shear in that portion of the tropics and or strong easterlies which kept allowing storms to sneak well west.

Thats about the one and ONLY comparision to 2004. There's been no persistent high pressure over the eastern USA and western atlantic since what 2008.