bobbutts

Taking the under on 10-14" with 18" locally here.  Seems like only the NAM supports those kinds of numbers.

I'll be rooting against my forecast though, rather have the snow than be right.

18z NAM actually shaves off some qpf up this way vs. 12z but it's still a big hit.  Euro really not good though.  Seems like a really hard forecast for C. NH.  I can't decide what to expect yet.

This (experimental) graphic from GYX showing extreme uncertainty in the potential.. I'm looking at 0-18"

Despite the rather silly range of values there I think this is an excellent product.

Start time for Boston?

Improving appearance on satellite.

This was sort of a backdoor hit for San Juan.  Even though they were in the right front, the storm had to go 30+ miles over hilly terrain before reaching San Jose.  I don't know how to begin trying to calculate how much, but that will definitely temper the winds some.

4 hours ago, donsutherland1 said:

The probability of Maria's making landfall in the U.S. Mainland has declined overnight. The guidance came into somewhat better agreement concerning Jose's fate and, as a consequence, some of the uncertainty concerning Maria's fate was also reduced.

Based on hurricanes passing within 100 nautical miles of Maria's 8 am position (19.9°N 68.7°W), historical climatology (1851-present) showed approximately half making U.S. landfall. However, that figure significantly overstates the actual probability of Maria's making landfall.

When the pool is refined for those that made U.S. landfall and those that didn't, one comes up with the following numbers:

Average Longitude at 22.5°N Latitude:

U.S. Landfall: 74.6°W (σ=3.0°)
No. U.S. Landfall: 70.6°W (σ=2.3°)

Average Longitude at 25.0°N Latitude:

U.S. Landfall: 77.7°W (σ=3.3°)
No. U.S. Landfall: 71.9°W (σ=3.1°)

Maria is expected to be at more than 1 standard deviation to the east of average for storms that made U.S. landfall. It is also likely to be fairly close to the mean figure for those that did not make landfall. Overall, the adjusted data set implies a landfall figure of around 20%.

Fewer than 20% of the combined EPS and GEFS ensemble members show U.S. landfall. The 9/21 0z run of the ECMWF again indicates that Maria will not make landfall and it did not show the kind of closer call seen on the preceding 9/20 12z run.

Considering this data and the statistics for landfalling storms vs. those that did not make U.S. landfall from the climatological sample based on Maria's 8 am position, I believe the probability of U.S. landfall is now about 20%. U.S. landfall is very unlikely.

Finally, looking at past storms that were significantly weakened after crossing either Puerto Rico or Hispaniola (implying structural damage to their core), Maria probably won't see an increase in its maximum winds beyond 10 knots to 20 knots from its post-Puerto Rico figure (95 knots). The window of opportunity for intensification will likely begin to close within the next 18-36 hours, so unless a bout of rapid intensification occurs, it is unlikely that Maria will regain strength that exceeds maximum sustained winds of 105 knots (120 mph) to 115 knots (130 mph). However, intensity forecasting typically has low verification scores, so my confidence in these figures is no better than moderate.

I don't see any improvement on the reasoning above in the more recent posts.  You can predict that it is definitely OTS at this point and have a roughly 80% chance of being right.  

3 minutes ago, Rickg2 said:

the only difference will be how far the debris fly. My concern is that the storm will increase in size and spread out the hurricane force winds. I believe this happens after an eyewall replacement.

Yeah, see the post I made while you were making that one.

I stand by my original post though.  Per the damage wheel from the NWS 175 is twice as bad as 160.  You REALLY don't want to get far into cat 5 territory for winds when it's the difference between partial and total failure of the structures where people are sheltering. 

http://www.srh.noaa.gov/jetstream/tropics/images/saff_wheel.pdf

NW eyewall in the bigger eye looking very formidable.  A large swath of PR looks to face extreme winds.

2 minutes ago, downeastnc said:

Probably wont be enough to matter for Puerto Rico, though it might weaken a bit....I guess a 160 mph Cat 5 is better than a 175 mph Cat 5 .....inner eyewall still looks ok on radar....either way not going to be a pleasant morning in PR....

Every MPH counts extra at this point.  Big difference even though the 160 is still incredibly bad.

45 minutes ago, CoastalWx said:

Direct hit on PR.

Barring a change in direction it looks to make a close run at San Juan.

6 minutes ago, dan11295 said:

You can see the outer ring of convection on radar associated with the secondary wind max mentioned in the 5 pm NHC discussion. So far it doesnt appear to be choking off the eyewall though. I would assume that if an ERC did start now it probably would not complete before PR landfall.

Outer ring should pass over St. Croix.  Unlucky timing for them for it to be strong at this point.  Looks like two healthy eyewall features to me.

The problem with mountainous terrain is that there can be severe flash flooding and mudslides where they drain.

No much wiggle room for St. Croix.  I'm getting about 306 degree movement from radar.

Cloud tops really cooling per goes-16.  Looks extremely imposing even with the less filled in NW side.

Probably more importantly for PR, if Maria were to track over St. Croix it would likely also pass well East of San Juan.  If Maria stays on the current trajectory the inner circulation will miss St. Croix to the South/West or just brush it.

3 minutes ago, NJwx85 said:

Does anyone know what the terrain is like on St Croix? I'm wondering if a direct hit there could disrupt the core similarly to what happened with the impact to Dominica. Obviously it would be terrible news for St Croix but it might save PR from total catastrophic impact.

Far less mountainous, unlikely to be disruptive.

Last 5 hours and change from TJUA radar.

This has to be about worst case winds for Dominica with the eye appearing to hit directly through the most populated areas.

I imagine that waves/surge and freshwater flooding will all also be pretty dire.

Looks to be tightening as it comes ashore (accounting for parallax)

How much disruption to either the structure or trajectory can we expect assuming the core passes over Dominica with it's rugged topology?

There are mountains up to 4700' there.  I wonder if there will be a jog as it gets closer.  I assume a weakly steered storm is more likely to have it's path affected by the local geography than one under strong steering?  Also with a more compact core it may be more response to those effects vs. a larger storm?

1 minute ago, jpeters3 said:

Fortunately, preparations for a 115 kt storm are pretty much the same as those for a 130-140 kt storm (which is probably closer to reality).  GTF away from low-lying areas, take shelter from debris.

Looking at the island on Google Earth the one promising thing is they seem to build with hurricanes in mind.  There are relatively few low lying dwellings and the population center is on the SW side of the island.  Looks nothing like some of the US where large neighborhoods are placed on barrier islands with 10' or less of elevation.  Of course a powerful storm rapidly intensifying as it strikes them is going to be devastating regardless of all that.

wow, GFS keeps ramping up the qpf and now the NAM with the snow.. Can't look away from this one yet.

When you're near a sharp gradient of obscene snowfall totals there's huge bust potential.  You could give some huge range like 6-36" but that isn't very useful.  It's the kind of thing that's just impossible to summarize in one little map with numbers on it.  You need time to explain that changes in track that are very hard to predict could change the forecast drastically.