New Grib Presentation

[did-g]

Originally Posted by bdbcat

Rick...

I assume on desktop we want the user to drag a rectangle region on the chart to select the area of interest. Not difficult, but will require an experienced c++/wxWidgets plugin author to build.

Dave

- There's already a half working area selection in grib plugin, half because it doesn't work if you pan viewport with cursor keys, IIRC.

- There's also the draw plugin with a fully functional one (may need some kind of rectangular mode ?).

[rgleason]

Did-g, you know, I really wouldn't mind a direct "fast download" from noaa that uses the screen coordinates as a first step.

I do think the selection box that we have for setting a more precise area could be simplified and implemented only in the second menu @ "Start Request" so that we don't have that funky red X come up and define the area in the first screen.
Suggest:
1. Hit @ "Start Request"
2. From Menu "Write and Send Email Request" check "Manual Request" or if it is already checked..(Perhaps just uncheck it when ever you go into the menu, unless the Start a definition of a selection box, "Use selection always" is checked.
That's it, no need for the other red X stuff.
3. Perhaps there is a better way to do it for email requests, but I haven't thought of it.

Now for a"Direct Noaa Download" and a "Box selection option" how to implement that?

Basically we have a selection for "Mail to [dropdown] Saildocs or Zygrib.
I think you would add a checkbox to "Mail to" and a new checkbox for "Direct Noaa download"
The configuration form might change a little but it would be similar.
The selection box routine would be the same as the emails I think.

[George64]

Is there any verification of the selected area vs. the available areas in NOAA?

[did-g]

Originally Posted by George64

Is there any verification of the selected area vs. the available areas in NOAA?

Isn't NOAA coverage worldwide?

[rgleason]

Many products are worldwide, but some products are focused on certain areas, particularly certain studies and experimental data sites.
NOAA is a primary source for many Met offices, as the NOAA satellites are an essential sensor for weather data.

I am still learning about Noaa after many years, but the databases get consolidated and the interpreters/prediction models get changed and improved with programming and tweaking to more closely resemble the real data being recieved. It is highly data intensive and people in one area don't often know much about other Noaa work. It is a big organization, doing essential work for business and the world, but they are very helpful and serious about their work.

Originally Posted by did-g

Isn't NOAA coverage worldwide?

[fogmachine]

NOAA has a number of models that are useful for sailors. Three of them are the GFS, the NAM, and the HRRR.

GFS is global, runs four times a day, and has forecasts to 10 days in the future (though it's not very accurate past a week).

NAM covers the continental US, runs four times a day, and forecasts for three days ahead.

HRRR covers the continental US at higher resolution, runs hourly and has 15-minute intervals for up to 17 hours ahead.

A good place to mess with them is http://www.tropicaltidbits.com

[did-g]

Originally Posted by fogmachine

NOAA has a number of models that are useful for sailors. Three of them are the GFS, the NAM, and the HRRR.

GFS is global, runs four times a day, and has forecasts to 10 days in the future (though it's not very accurate past a week).

NAM covers the continental US, runs four times a day, and forecasts for three days ahead.

HRRR covers the continental US at higher resolution, runs hourly and has 15-minute intervals for up to 17 hours ahead.

A good place to mess with them is http://www.tropicaltidbits.com

yes you're right but currently Opencpn can only decode GFS global model.

NAM store more than one parameter per record, I made a PR some months ago but I have to update it.

HRR is using a non regular grid and O can't decode it, I'v started working on it but never finished, a good stuff with HRR 15 minutes is the 5 minutes wind mean parameter, IMO better for us than only instantaneous wind available in all other models.

O could request HRR from saildoc though (saildoc format is a regular grid v1 grib) but it's only hourly.

[seandepagnier]

How to retrieve partial areas of GFS from noaa? I still don't understand how to do it. Is there a way with url manipulation or partial http GET?

From what I can tell other programs use their own server. This server gets the worldwide coverage, and breaks it down for users, for example, saildocs or zygrib. Is the true?

[bdbcat]

Sean...

See:
http://www.cruisersforum.com/forums/...ml#post2535085

Dave

[did-g]

Originally Posted by boat_alexandra

How to retrieve partial areas of GFS from noaa? I still don't understand how to do it. Is there a way with url manipulation or partial http GET?

From what I can tell other programs use their own server. This server gets the worldwide coverage, and breaks it down for users, for example, saildocs or zygrib. Is the true?

In my understanding the main reason saildocs and zygrib are using their own servers is for converting to grib version 1.

if you look at NOMADS-NOAA Operational Model Archive and Distribution System every 'grib filter' links available can display the right URL for a partial download.

[rgleason]

Sean, from bdbcat. Hope this helps.

[rgleason]

http://www.nws.noaa.gov/os/notificat...ave_multi1.htm
major upgrade to the Global Wave Forecasting System (Multi_1).
Upgrade highlights:

- Adding a North Polar Stereographic Grid extending to near the
North Pole
- Extending hourly grib2 output out to 120 hours

http://polar.ncep.noaa.gov/waves/vie...-US_eastcoast-

Sean it has Primary and Secondary wave height and periods, with Wind
Its more data but it is the major factors in routing all together, just missing current.

Are you out of the ice yet?

[SCarns]

Very happy to see you guys working on this. Hate the middle-man of the "email" and "open" portion of installing a grib. Other apps do the direct download thing direct from NOAA and it's cool/worthwhile pursuit...

[rgleason]

David Burch, Starpath Navigation reports several changes/improvements from NOAA regarding forecast of severe weather. Is Grib_pi ready?

HI Rick,


Not sure about this. I think i mentioned once that simulated radar reflectivity (REFC) is a better parameter for predicting squalls than CAPE, but i think i left it at i need to write an article to demo that. It is definitely true. I meant to say something about this again later when i saw someone post on OpenCPN forum that they thought precip would be just as good for squalls, and that too is not the case. The REFC is best of all. CAPE is a stability parameter that might lead to squalls, but there are other conditions required as well. See Table 3.4-3 (our weather book) and related discussion. REFC in effect combines these stability parameters along with other atmospheric parameters to yield a real squall forecast.


We have an example in our weather book, section 4.6 ( I think i sent you a copy)… and i have a blog post on a likely example but not proven. More to the point i am following a past student of ours as he rows to Australia and i compare the squall forecasts with what he experiences.


I will try to get to the article as soon as i can.


The other related news is NWS switches GFS to FV3-GFS on June 12. After that date GFS will include REFC globally. Last time i checked, the latest grib plugin for opencpn showed this parameter nicely. The only issue to resolve is the color bar as there is a standard one for this. Europe and US are slightly different, but the European one is better. I have notes on this that i was going to post but did not get around to it yet.


—david

[rgleason]

From David Burch's Blog

...promise to be a key parameter for inland and coastal sailing, in that they should be able to forecast squalls in an easy to understand manner. This parameter is also available offshore in FV3-GFS data.



He states that Luckgrib and Expedition are the only ones that can download REFC on that date. (Dec 13, 2018)


Can zygrib now select and download REFC? Saildocs?




PS: fogmachine's notes below detail some of the same models as David.

http://www.cruisersforum.com/forums/...ml#post2536698