SHIPS Model

[wjs3]

Hi:

This might be a dumb question...but I wondered if anyone knew where you can get a full output of current runs of the SHIPS model online.

I've seen output of what the SHIPS forecasts are...like here:

http://euler.atmos.colostate.edu/~vigh/ ... nsity1.png

And of course in some of the text products that folks post on these boards.

But I have read in various NHC discussions and other places that SHIPS includes a number of parameters and how each is working for or against development. For instance there are frequent references to SHIPS estimate of shear. I've also seen references to upper tropospheric (200 MB?) temps, among others.

Anyway, does anyone know where you can get the full output, including all the parameters, and not just the predicted windspeeds?

Anyone who's familiar with this, please let me know if I'm off base.

Thanks

[TWTC]

If your browser can handle ftp, then open up this URL:

ftp://ftp.nhc.noaa.gov/pub/atcf/stext/

scroll to the bottom for the latest detailed SHIPS files listed by date/time groups and by storm id and you might just see the detailed output including the main predictors and the relative contribution (positive or negative) of each predicter to the final intensity calculation.

[GulfBreezer]

The only dumb question is the one that is not asked

[Rainband]

The only dumb question is the one that is not asked

Agreed I tell my crew that all the time

[WindRunner]

If your browser can handle ftp, then open up this URL:

ftp://ftp.nhc.noaa.gov/pub/atcf/stext/

scroll to the bottom for the latest detailed SHIPS files listed by date/time groups and by storm id and you might just see the detailed output including the main predictors and the relative contribution (positive or negative) of each predicter to the final intensity calculation.

WHOA! Good find. Sorry if that seems a little excessive, but that's exactly what I said when I opened one of the files. Great find!!!

[TWTC]

Ah Grasshopper, much in life can be found with a bit of digging and much snooping around.

[wjs3]

Thanks, all. I'll take a look at the FTP!

[wjs3]

Let me add:

WHOA! That kills.

Thanks a ton, TWTC! Now I just need to spend time learning to interpret this. I know the terminology...but how do you combine the predictors down the columns to get to an intensity change estimate? Adding them doesn't seem to work???

And, how do you interpret the rapid intensification stuff at the bottom?

Thanks! This is really superb--the sort of thing that I love about this board!

[wxmann_91]

Sweet stuff, wish they had 2005 archives though. Thanks for the link!

[WindRunner]

Let me start off by saying that I don't know exactly how a few of these params pertain to intensity, but I think I've got most of it.

For the RI stuff at the bottom, just look at the percentages to keep it simple. They've done some calculations with the numbers in the first section to come up with the percentages in the second section. The "Prob of RI = XX%" is the good stuff And no, I don't know why a discriminated index makes a difference, or even how it makes a difference.

The intensity change stuff you ask for is the first thing, i.e.

Code: Select all

TIME (HR)          0     6    12    18    24    36    48    60    72    84    96   108   120
V (KT) NO LAND    30    35    40    45    50    58    64    66    62    57    50    43    34
V (KT) LAND       30    35    40    45    50    58    64    66    62    57    50    43    34
V (KT) LGE mod    30    32    34    36    38    42    44    46    44    41    37    32    27


First row is forecast time in hours (and yes, that is every 6 hours all the way out, and yes that's more detailed than what we see on the traditional BAMM/LBAR outputs)
Second row is the intensity at the above time in knots, without paying attention to any land interaction.
Third row is the same as the second, except it does account for land.
Fourth row "LGE mod" - I don't know, haven't seen that before, possibly the intensity estimates off of a large-grid model, which would explain why they are so much lower.


Of course, now I see that that isn't what you're asking at all (I think).
The charts in the middle are simple add-ups of the "contributions" or whatever they call them from each of the factors. Of course they don't say where those numbers came from, but that doesn't really matter. Adding up the "sub-total change" and the "total adjustment" gives the "total change (kts)", which is how much the system will intensify from T+0hr until that time frame.

The top stuff is just general predicted info for the storms environment, I believe.

[TWTC]

Yes Grasshopper, so many questions, but in life sometimes one must seek answers on their own with a little help along the way. Sometimes 1 + 1 does not equal 2 in computer models, especially multiple regression statistical/dynamical models which change quite a bit.

The key in the output is the major predictor values used in the model and then identifying the predictors that either enhance or detract the most from the intensity calculation and then by using meteological experience, one tries to determine if the model is on to something or determine if its "out to lunch."

For the rapid intensity portion, I suggest trying a GOOGLE search on the following: SHIPS RI JHT

But use RI with caution, predicting rapid intensification is a dicey business which no one has any skill, if they claim skill, then beware of any other advice they might offer.

So learn Grasshopper and Enjoy...

[wjs3]

LOL TWTC--

Thanks a ton

Windrunner, thanks too.

Grasshopper is off to google.

[wjs3]

Just found the orginal SHIPS paper which should help me figure out what all this stuff is...we are very much in the land of regressions, aren't we????

http://ams.allenpress.com/pdfserv/10.11 ... -0434(1994)009%3C0209:ASHIPS%3E2.0.CO%3B2