French med weather forecast suggestions?

[hoped4]

We've arrived in the French Med - we have Navtex and 3G phone (so far identified accuweather' on line) - what sources do others use to get a rounded view on weather forecasts please? I realise there are limits to online sites and they need to be treated with care/ scepticism.

 

[Sybarite]

As I posted on Scuttlebutt :

http://marine.meteoconsult.co.uk/marine-weather-forecast/shipping-forecast.php

 

[nortada]

As I posted on Scuttlebutt :

http://marine.meteoconsult.co.uk/marine-weather-forecast/shipping-forecast.php

There are tons of wx sites available, YR, Windgruu, Windfinder, Buoyweather to name but four.

What are these limits on on-line sites and why should they be treated with care/septicismThey forecast world wide using similar models to services nearer to home.

 

[GSL]

I always find Passage Weather very reliable.

Graham

 

[BobnLesley]

We've reached the western Atlantic/Caribbean now, but as when we were in the Med, we still use Passageweather and Windguru as our preferred/primary weather sources.

You mention having a Navtex unit; in the western/central Med (before regular/easy internet) we found the Meteo-France forecasts broadcast over that area to be perhaps the best forecasts that we've had anywhere. I recall sailing around Sardinia when we soon stopped heeding the Italian Navtex forecasts and instead looked at the French ones for the areas to the north and west of us and extrapolated from those.

 

[hoped4]

Thank you - these are some really helpful suggestions.

 

[GHA]

Our very on Frank's excellent site is worth a read for how it all works, most free sites are just different ways to look at the same GFS data, maybe ECMWF as well.

http://weather.mailasail.com/Franks-Weather/Home

 

[franksingleton]

As far as I can see, Metroconsult is derived from the US GFS. If you find the presentation OK, then use it. Personally, I prefer to use the GFS unadulterated - zyGrib on a laptop, PocketGrib or several other apps on a tablet. http://weather.mailasail.com/Franks-Weather/Grib-Files-Apps

ALWAYS know what the GMDSS forecasts are saying. Meteo France website on a laptop. Unfortunately, with a tablet, you have to use the Meteo France app. Rather slow, we find, with less information than on their website. http://www.meteofrance.com/appli-meteo.

Don't be conned into paying for forecasts that claim high precision. Weather is not precise, nor are or can be forecasts.

If I were in the Med now, I would take a look at the Turkmarine Weatherwise. This uses the WRF, a mesoscale model as good as any, but starts with ECMWF analyses rather than the rather more coarse GFS output. The better analysis should give a better forecast - on average, at least.

See links at http://weather.mailasail.com/Franks-Weather/Grib-And-Objective-Forecasts-Reviewed

Most GMDSS Med forecasts at http://weather.mailasail.com/Franks-Weather/Mediterranean-Marine-Weather-Forecast-Texts.

Passageweather, XCWeather and many others give the GFS in various guises. None can be saved for viewing later offline. Most GRIB services are automatically saved. That is useful in order to look at forecasts from day to day. Consistency over two or three days is a good guide to a forecast being reliable. An example is at http://weather.mailasail.com/w/uploads/Franks-Weather/ninedaystppdart.png. This shows a series of forecasts for a single location. See how the forecasts became very similar. Inconsistency implies uncertainty. Sometimes weather is more predictable than at other times.

 

[crisjones]

We have been using the standard US GFS Grib files downloaded via Zygrib in conjunction with the Open Skiron & Open WRF Grib Files available from http://www.openskiron.org/en/ I must admit that the Open Skiron Gribs do seem to be somewhat "variable" in their accuracy. However I do find that if the GFS and WRF gribs are giving much the same forecast then it can usually be relied upon for at least 2 or 3 days.
In general the US GFS Grib files are as accurate as anything else I have used in the Med particularly since they have reduced the grid size to 0.25 degrees. Obviously they still struggle with "local" effects such as headlands and small Islands, the WRF gribs can be better for forecasting these local effects.
We have never paid for weather info in 6 years in the Med and always feel we have enough info and accuracy from the above Gribs

 

[franksingleton]

We have been using the standard US GFS Grib files downloaded via Zygrib in conjunction with the Open Skiron & Open WRF Grib Files available from http://www.openskiron.org/en/ I must admit that the Open Skiron Gribs do seem to be somewhat "variable" in their accuracy. However I do find that if the GFS and WRF gribs are giving much the same forecast then it can usually be relied upon for at least 2 or 3 days.
In general the US GFS Grib files are as accurate as anything else I have used in the Med particularly since they have reduced the grid size to 0.25 degrees. Obviously they still struggle with "local" effects such as headlands and small Islands, the WRF gribs can be better for forecasting these local effects.
We have never paid for weather info in 6 years in the Med and always feel we have enough info and accuracy from the above Gribs

Yes, the GFS actually uses a grid about 1/2 of the output. That is as small or smaller than any other global model. OPenSkiron and Open WRF, like all other "detailed" models (as far as `I know) start with the GFS 0.25 degree analysis and boundary conditions. That is why claims for detailed forecasts are misleading. They may (or should) represent topography better than the GFS but they will not "know" about any small weather detail. By small, I mean about 4 or 5 grid lengths. A grid of points can only define shapes of about 4 or 5 grid lengths. In addition, small weather details have short lifetimes. Something the size of a large thunderstorm cloud, about 10-15 km, has a lifetime of only a few hours. That sets a limit 0n predictability.

For Cruising decisions, the GFS will not be beaten over all. On specific occasions any model may do better than the rest. When it comes to local detail, use experience plus the GFS. Remember that NWP models wil usually underestimate the strongest winds by at least one Beaufort. Models - or any objective forecast can be seriously wrong for poor visibility. That is why GMDSS forecasts are a must-have. Any forecast service that provides apparently detailed forecasts over large areas have will only be interpolating between grid points with NO human input. GMDSS forecasts have human input. Having said that, Italian GMDSS forecasts seem to have little human input. We used to prefer to use `French GMDSS forecasts to the west of Italy than the Italian.

 

[BobnLesley]

"Passageweather, XCWeather and many others give the GFS in various guises. None can be saved for viewing later offline."

I don't know about the others, but PassageWeather-lowbandwidth (it does what it says on the can) downloads and can be saved for ongoing comparison, that's exactly how we do it.

 

[JanZ]

I am fond of windy.com as linked to by Jim Baerselman from his website. Graphically superb and I have always found it close to what is actually happening!

Jan

 

[crisjones]

The standard US GFS Gribs that you download via Zygrib are saved to your computer so can easily be viewed offline, unless you really want the pretty pictorial output of sites like Passageweather, XCWeather etc then the GFS Grib is as good as it gets.
You can change the area and data that you download so file sizes can be very small if you are on limited bandwidth connections.

 

[GHA]

"Passageweather, XCWeather and many others give the GFS in various guises. None can be saved for viewing later offline."

I don't know about the others, but PassageWeather-lowbandwidth (it does what it says on the can) downloads and can be saved for ongoing comparison, that's exactly how we do it.

Isn't that just the images though?

The actual GFS grib data would be smaller still, plus see the gusts as well.

The weatherfax Opencpn plugin is good for comparing gribs data against synoptics & satellite images, have all one screen at once..

 

[franksingleton]

"Passageweather, XCWeather and many others give the GFS in various guises. None can be saved for viewing later offline."

I don't know about the others, but PassageWeather-lowbandwidth (it does what it says on the can) downloads and can be saved for ongoing comparison, that's exactly how we do it.

OK, that was a little sweeping.. You can save but only using screenshots, unless I have missed something. GRIB software saves files automatically. Sailors latched on to GRIBs early on (about 2000) because they were suitable for low bandwidth technology - using email via Saildocs - still a useful service, I find it convenient. We used to have to keep files down to a few 10s of kB. Now it is a few 100s. I have no idea about Passageweather or XCWeather sizes. Ashore with 3G etc it hardly matters. At sea using email over a satellite phone or HF/SSB it can be important. Also, there are areas - W Scotland where you might have to use 2G or GPRS.

If you are happy to use the screenshot technique, then OK. You are doing what many do not do and are comparing successive forecasts. That is one of my key recommendations. What I am really saying is that GRIBs make it easier and you have other built-in facilities. For exampLe, with zyGrib or PocketGrib, you can produce a meteogram for any location in your area of interest. You gain in flexibility against something that is, in fact, the same forecast, is easy to download but lacks flexibility.

 

[franksingleton]

I am fond of windy.com as linked to by Jim Baerselman from his website. Graphically superb and I have always found it close to what is actually happening!

Jan

Windy.tv is, as you will know, the GFS. As it provides a truly global picture, there must be some degradation otherwise files would be enormous. I have always thought of it as an impressive teaching facility rather than useful cruising decision making tool. You get more detailed information using GRIBs. For a true second opinion, there is MeteoEarth, a Meteo Group service. They use the ECMWF model again, presumably, downgraded.

A direct URL. Is https://www.windy.com/?43.517,-35.508,3