Shortwave Broadcast of GRIB's ?

[Adelie]

I have been searching on-line to see if GRIB's are being broadcast on shortwave frequencies like weather fax.

I have not found any information so I assume this is not happening.

Does anyone know different?

Also does anyone know why it wouldn't be happening?
The files too large for any reasonable broadcast area?
Data doesn't lend itself to decoding thru a computer's sound card like WxFax?
Nobody's gotten around to it yet?
???

Thanks

[Paul Elliott]

I don't believe that anyone is doing this, and I can think of a few reasons why not:

* The files are large, so it takes an aggressive protocol to be able to transmit them in a reasonable amount of time. The only sound-card decoders that might do the job are not really standardized, or too slow, so the idea of broadcasting wide-area GRIBs is questionable.

*The common digital protocols rely on retransmission of any data that has been corrupted by noise or interference. The way it is done now requires a two-way link for this type of error recovery. There are methods for forward-error-correction that can be used for broadcast, but these are not standard protocols for the currently available receive/decoding software. By the way, there are good reasons why the two-way error correction protocols are used.

* If you want broadcast, there are always the WFAX services. True, you can't plug these into a weather routing program...

[barnakiel]

1) One error in a fax transmission and you have a black dot where there would be a white one. One error in a grib file transmission and you may end up with a bad (useless) file.

2) The type of transmission required is completely different, you will end up transmiting a file attachment in any case, which is already available via SSB/modem or the Internet,

3) Once you have the fax onboard do you really need the grib? Other than for racing purposes.

4) Pactor is not the cheapest piece of equipment. You cannot open your beer with Pactor either (it is not IPX6, probably not even IPX5).

And so on / forth.

b.

[rhumbunctious]

Sailmail or ham radio equivalent plus saildocs.

SailMail
Home Page

[Auspicious]

Originally Posted by barnakiel

3) Once you have the fax onboard do you really need the grib? Other than for racing purposes.

If you have weather fax capability and a navigator who knows what to do with the products why on earth would you want gribs? A weather product that doesn't show lows, highs, and associated fronts isn't a viable primary weather product in my opinion.

[Paul L]

To the original poster as noted above, you can't get GRIBs via open HF, you do need a SSB with Pactor or an internet connection. With any internet connection you can request them from Saildocs.

As far as comparing a GRIB to a WxFax. For most cruising a GRIB is much more useful than a WxFax. Take a look at the WxFax for west of coast of the US, i.e. the northern pacific. It covers such a huge area that it will only tell you the very very big picture. With a GRIB file I can request wind (and rain and pressure) for the area I'm about to transit, say 150 to 500 miles. This will come in .5 or 1 degree increments. Much more useful than seeing a high pressure 1,500 miles away. I'm not saying that the WxFax's are useless, just that from a practical point of view, the GRIBS are lot more useful.

[LakeSuperior]

Originally Posted by Auspicious

If you have weather fax capability and a navigator who knows what to do with the products why on earth would you want gribs? A weather product that doesn't show lows, highs, and associated fronts isn't a viable primary weather product in my opinion.

This comment makes me wonder if you have ever looked at a GRIB product.

[Auspicious]

Originally Posted by LakeSuperior

This comment makes me wonder if you have ever looked at a GRIB product.

I have. I keep track of all the options available to me. On deliveries sometimes gribs are all I have to work with. Those are riskier.

If you can show me a grib model that at least shows fronts I'll be very grateful.

On the other hand, the RTOFS model for the Gulf Stream has been nice.

For the US East Coast, especially in winter when cold fronts boil off the land mass every few days -- well, not so much.

[Adelie]

Originally Posted by Paul Elliott

I don't believe that anyone is doing this, and I can think of a few reasons why not:

* The files are large, so it takes an aggressive protocol to be able to transmit them in a reasonable amount of time. The only sound-card decoders that might do the job are not really standardized, or too slow, so the idea of broadcasting wide-area GRIBs is questionable.

*The common digital protocols rely on retransmission of any data that has been corrupted by noise or interference. The way it is done now requires a two-way link for this type of error recovery. There are methods for forward-error-correction that can be used for broadcast, but these are not standard protocols for the currently available receive/decoding software. By the way, there are good reasons why the two-way error correction protocols are used.

* If you want broadcast, there are always the WFAX services. True, you can't plug these into a weather routing program...

I did quite a bit more research and found that local area Grib files can be as small as 10kb or less. The numbers I found for large areas were to quote "60kb for the Med up to 200kb for half an ocean."

Is 200kb large compared to a fax? I couldn't find info about weather fax data file sizes but I did find that a standard monochome fax is 1728x2200 pixels for I assume a whole page. That works out to about bit less than 3.8Mbits or 475Kb (8bit bytes). Assuming a typical WxFax is half a page, the file sizes should be comparable.

Barnakiel's comment about 1 error screwing the whole file strikes me as wrong, it should only screw the wind direction or strength at one grid point.

I understand error correction is a big problem with no backhaul data channel but I could see a number of ways to minimize error, if nothing else the brute force method of sending the data 3 or more times and having the receiver vote on each forecast location.

Everything I have been able to find seems to point to a perceived lack of demand rather than any really big technical hurdles as the reason nobody is providing this.

Being a miser I was hoping there was a way to avoid buying the whole SSB and Pactor setup or an Iridium and still get the Gribs. Looks like I'll have to stick with WxFax untill I win the Lottery.

Thanks for the info everybody.

[Paul Elliott]

Originally Posted by Adelie

I did quite a bit more research and found that local area Grib files can be as small as 10kb or less. The numbers I found for large areas were to quote "60kb for the Med up to 200kb for half an ocean."

True, a small coarse grib is going to be about 10K bytes, but for a generally-useful broadcast service you're going to need to broadcast many small gribs in order to cover the areas people are going to want.

Is 200kb large compared to a fax? I couldn't find info about weather fax data file sizes but I did find that a standard monochome fax is 1728x2200 pixels for I assume a whole page. That works out to about bit less than 3.8Mbits or 475Kb (8bit bytes). Assuming a typical WxFax is half a page, the file sizes should be comparable.

WFAX broadcast uses an analog transmission format so we can't directly compare filesizes. The necessary time/bandwidth/signal-to-noise ratio would probably be equivalent to a few hundred KBytes of data, but that's just a guess -- I haven't done the math. You can get JPEG or TIF-compressed WFAX charts via email, and these are about 20KBytes.

Barnakiel's comment about 1 error screwing the whole file strikes me as wrong, it should only screw the wind direction or strength at one grid point.

Barnakiel is correct. One error can cause cascading coding errors, which might render the entire file useless.

I understand error correction is a big problem with no backhaul data channel but I could see a number of ways to minimize error, if nothing else the brute force method of sending the data 3 or more times and having the receiver vote on each forecast location.

As I mentioned previously, there are forward-error-correction methods (including the one you suggest), and these all significantly increase the amount of data that needs to be sent.

Everything I have been able to find seems to point to a perceived lack of demand rather than any really big technical hurdles as the reason nobody is providing this.

The technical hurdles aren't insurmountable, but they aren't trivial -- especially when we are talking about a limited-bandwidth, and noisy communications channel. There are already other methods that work and have been accepted. I think this is why you don't see a big demand for broadcast gribs.

Being a miser I was hoping there was a way to avoid buying the whole SSB and Pactor setup or an Iridium and still get the Gribs. Looks like I'll have to stick with WxFax untill I win the Lottery.

You might look at using ham radio and the WinMor protocol (which runs on a sound card). This is still pretty experimental, and I don't know how well distributed (or connected) the WinMor stations are, but this could be a relatively low-cost radio email solution. I honestly don't know if WinMor is at the "practical solution" stage though.

[estarzinger]

Originally Posted by Auspicious

A weather product that doesn't show lows, highs, and associated fronts isn't a viable primary weather product in my opinion.

If you can show me a grib model that at least shows fronts I'll be very grateful.

.

See below a current gfs grib off the US east coast with wind, pressure and rain layers.

You will see a high pressure system marked (lower left corner)

And you can clearly see the frontal system in the rain layer diagonally across the picture.

Rather useful IMHO

[conachair]

Slight thread drift, but has anyone had any success recieving images direct from satallites? Receiving weather satellite images

[Paul Elliott]

Originally Posted by conachair

Slight thread drift, but has anyone had any success recieving images direct from satallites? Receiving weather satellite images

Here's a link to an old discussion that touches on this question: http://www.cruisersforum.com/forums/...ery-22029.html

One of the posters describes his WXSAT direct-reception setup.

[conachair]

Originally Posted by Paul Elliott

Here's a link to an old discussion that touches on this question: http://www.cruisersforum.com/forums/...ery-22029.html

One of the posters describes his WXSAT direct-reception setup.

Cheers. Fred Rift has left the building.......

[barnakiel]

Originally Posted by Adelie

Barnakiel's comment about 1 error screwing the whole file strikes me as wrong, it should only screw the wind direction or strength at one grid point.

Barnakiel might have been wrong. He is a sailor, not an IT specialist.

Alas, when he takes a grib file and slightly manipulates some data in it (slightly, say replacing some bits with different values) he finds the manipulated file not to open.

I believe the way a file is sent - if a packet is lost - the file is useless. And to know a packet is lost one has to do the checksums. And to do them one needs a way to send back 'repeat packet' command if a packet is wrong or missing.

???

IT people now - is this picture good enough, or is Barnakiel COMPLETELY wrong?

He IS willing to learn too!

b.