Jump to content
  • Member Statistics

    17,608
    Total Members
    7,904
    Most Online
    NH8550
    Newest Member
    NH8550
    Joined

Hurricane Sandy


Wonderdog

Recommended Posts

Which is probably why he showed the spaghetti plots first and foremost where most take the track you suggest.

And that's why I took back some of my initial criticism. IMO, the time he used showing the Euro Op. would have been better spent going into the actual forecast in more detail.

Link to comment
Share on other sites

  • Replies 1.5k
  • Created
  • Last Reply

And that's why I took back some of my initial criticism. IMO, the time he used showing the Euro Op. would have been better spent going into the actual forecast in more detail.

Should he now show the 18z GFS?

If he put caveats that it was just one of the possibilities then I see no problem. There is Tom T hype and there is JB hype and I think we all know who is worse.

Link to comment
Share on other sites

I know that people rag on the NOGAPS quite a bit (for good reason), but being a spectral model this type of situation plays to its strengths (oceanic storm, multiple significant waves). It's not terribly surprising to me that it has held onto its wrap-and-bomb solution for a few days now.

I personally think paying attention to the behavioral trends in spectral models and grid point models on a tricky situation like this can give a forecaster a nudge towards the most likely situation.

Link to comment
Share on other sites

I know that people rag on the NOGAPS quite a bit (for good reason), but being a spectral model this type of situation plays to its strengths (oceanic storm, multiple significant waves). It's not terribly surprising to me that it has held onto its wrap-and-bomb solution for a few days now.

I personally think paying attention to the behavioral trends in spectral models and grid point models on a tricky situation like this can give a forecaster a nudge towards the most likely situation.

GFS is a spectral model I believe...correct?

Link to comment
Share on other sites

GFS is a spectral model I believe...correct?

That's correct. There are others that fall into that category too (the RSM from the SREF ensemble for instance). The NOGAPS has inferior resolution to the GFS, but uses a superior data assimilation scheme (4D-Var vs the GFS's 3D-Var). In situations where large waves are present, the resolution advantage of the GFS is diminished somewhat. I believe this may be the case here.

At this distance, looking at ensembles is a good bet since they can tell you whether or not the deterministic run is likely or not. In this case, the model ensembles are telling us that the phasing situation is likely and extremely sensitive to timing (with respect to landfall point), but that Sandy escaping out to sea without much interaction is possible, but not very likely.

Link to comment
Share on other sites

That's correct. There are others that fall into that category too (the RSM from the SREF ensemble for instance). The NOGAPS has inferior resolution to the GFS, but uses a superior data assimilation scheme (4D-Var vs the GFS's 3D-Var). In situations where large waves are present, the resolution advantage of the GFS is diminished somewhat. I believe this may be the case here.

At this distance, looking at ensembles is a good bet since they can tell you whether or not the deterministic run is likely or not. In this case, the model ensembles are telling us that the phasing situation is likely and extremely sensitive to timing (with respect to landfall point), but that Sandy escaping out to sea without much interaction is possible, but not very likely.

I thought the GFS went away from 3D-Var earlier this year. If I remember correctly, its some 3D hybrid.

Link to comment
Share on other sites

That's correct. There are others that fall into that category too (the RSM from the SREF ensemble for instance). The NOGAPS has inferior resolution to the GFS, but uses a superior data assimilation scheme (4D-Var vs the GFS's 3D-Var). In situations where large waves are present, the resolution advantage of the GFS is diminished somewhat. I believe this may be the case here.

At this distance, looking at ensembles is a good bet since they can tell you whether or not the deterministic run is likely or not. In this case, the model ensembles are telling us that the phasing situation is likely and extremely sensitive to timing (with respect to landfall point), but that Sandy escaping out to sea without much interaction is possible, but not very likely.

But it looks like the 18Z gefs are not as bullish with respect to phasing as the 12z. Quite few look like they take sandy newd and then develop a second low on the baroclinic zone which then bombs out which would be a regular northeaster but probably would still have more moisture with it than a run of the mill one. I agree that landfall somewhere along the east coast is the more likely scenario than a complete miss but might raise the probability of the latter a little based on the GFS. I do think it interesting that the beta advection models that run off the gfs all make landfall. I'm not sure what that means, probably nothing but find it interesting.

Link to comment
Share on other sites

That's correct. There are others that fall into that category too (the RSM from the SREF ensemble for instance). The NOGAPS has inferior resolution to the GFS, but uses a superior data assimilation scheme (4D-Var vs the GFS's 3D-Var). In situations where large waves are present, the resolution advantage of the GFS is diminished somewhat. I believe this may be the case here.

At this distance, looking at ensembles is a good bet since they can tell you whether or not the deterministic run is likely or not. In this case, the model ensembles are telling us that the phasing situation is likely and extremely sensitive to timing (with respect to landfall point), but that Sandy escaping out to sea without much interaction is possible, but not very likely.

Many dismiss the NOGAPS due to its low performance overall, but as you said it is now 4D-var..upgraded within the last year or so I believe- so it is probably a good model to use for collaboration in situations such as this, where there is a rather wide spread in the models.

Link to comment
Share on other sites

How does the rapid strengthening that's being discussed over in the main forum play into the eventual path. I mean, if the actual strength now is beyond what was modeled, doesn't that affect the path later on? Or no?

I asked the same question earlier. I was wondering since it seems that a westward-moving storm will often start jogging north as it deepens, and thus perhaps something of the sort would also happen to a northward-moving storm. Or not...because of its northerly track and higher latitude.

Link to comment
Share on other sites

That's a good question and obviously a critical one. I did a quick google search and didn't come up with anything significant. One reference of Hurricane Irene (1999 version) which went through RI and saw increased storm motion from 10 to 18 m/s.

http://journals.ametsoc.org/doi/abs/10.1175/MWR-D-11-00293.1?journalCode=mwre

Link to comment
Share on other sites

But it looks like the 18Z gefs are not as bullish with respect to phasing as the 12z. Quite few look like they take sandy newd and then develop a second low on the baroclinic zone which then bombs out which would be a regular northeaster but probably would still have more moisture with it than a run of the mill one. I agree that landfall somewhere along the east coast is the more likely scenario than a complete miss but might raise the probability of the latter a little based on the GFS. I do think it interesting that the beta advection models that run off the gfs all make landfall. I'm not sure what that means, probably nothing but find it interesting.

Yeah, you're right. The GEFS members that are stronger and slightly further west when Sandy is in the Bahamas all end up with landfall, the weaker/east ones end up OTS. Tonight's RI episode argues for the stronger solutions there (pending Cuba interaction of course). Those beta advection models may be onto something with a large, NNW moving Sandy in the Bahamas.

Link to comment
Share on other sites

How does the rapid strengthening that's being discussed over in the main forum play into the eventual path. I mean, if the actual strength now is beyond what was modeled, doesn't that affect the path later on? Or no?

A question was asked in SNE about strength or size affecting the phase part(path) and it seemed the answer was more towards size. A larger system would not be picked up as easily.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.

×
×
  • Create New...