sources of mobile inshore forcast

[niccapotamus]

it has probably been asked before - but what are the best sources of a bog standard weather forcast please. mobile and ipad

many thanks
nick

 

[Rigger Mortice]

Met office for me.

 

[superheat6k]

I find the following very good ...

Met Office
BBC
Windfinder
Weather outlook

 

[LadyInBed]

I've found meteo consult marine app to be PDG.
I like the layout, prediction of % acuacy, strong wind warning and lots more when you start digging into it.

http://marine.meteoconsult.co.uk/shipping-forecast-services/shipping-forecast-iphone-application.php

 

[franksingleton]

Forecasts issued under the GMDS should always be heeded. They are based on numerical weather prediction models but with human experience and vetting. They can only be very general since you cannot describe weather over a large area or stretch of coast adequately in a few words. There are many sources depending on where you are. Other than VHF and NAVTEX, there are many internet links listed at http://weather.mailasail.com/Franks-Weather/Weather-Forecasts-On-The-Internet. For NW Europe, try http://weather.mailasail.com/Franks-Weather/Northwest-European-Marine-Weather-Forecast-Texts.

The various automated forecasts available from "unofficial" sources,, eg
MeteoConsult are generally based on the US GFS but with no human input or vetting. Even those produced by
National Weather Services should always be used in the light of GMDSS Forecasts.

 

[franksingleton]

I've found meteo consult marine app to be PDG.
I like the layout, prediction of % acuacy, strong wind warning and lots more when you start digging into it.

http://marine.meteoconsult.co.uk/shipping-forecast-services/shipping-forecast-iphone-application.php

See ¥5. I have seen MeteConsult seriously underestimate wind strengths and be dangerously optimistic about visibility. By al means, use them but use GMDSS forecasts as well. human input is invaluable.

 

[dylanwinter]

windguru

it has probably been asked before - but what are the best sources of a bog standard weather forcast please. mobile and ipad

many thanks
nick

deadly accurate to the nearest hour three days ahead

not bad two weeks ahead

very local

and gives cloud cover at three levels so as a cameraman I know what the light will be like

D

 

[Jamesuk]

Weather pro is pretty good in the UK, pretty awful in the med for wind. I'll never forget getting a forecast from a nearby town telling me the current wind on the coastal town was 1knot when in fact it was 30 gusting 52!!

Pocket Grib, met office, passage weather

 

[franksingleton]

Windguru, Pocketgrib,, XCWeather, Passageweather, Windfinder and many others are either just different representations of the US GFS or based on the GFS. None is any better or any worse than the others. WeatherPro is probably based on the ECMWF model. It should be slightly better than the rest. But, on any particular occasion, ECMWF may or may not be the best.

For planning ahead for the next week, I use the GFS. By comparing forecasts on successive days, I get an impression on whether the forecasts are reliable or not. Sometimes, the situation is finely balanced and forecasts will depend greatly on quite small detail in the analysis.

I show how to make these judgements in Reeds Weather Handbook, http://weather.mailasail.com/Franks-Weather/Reeds-Weather-Handbook.

 

[fergie_mac66]

deadly accurate to the nearest hour three days ahead

not bad two weeks ahead

very local

and gives cloud cover at three levels so as a cameraman I know what the light will be like

D

Plus 1 winguru amazingly accurate I wonder why the met
Office seems to get nowhere near the same with all there computing power.
I have noticed that when windguru is I bit out, its usually because the predicted weather has come a few hours early or later but the strength and direction changes are usually still valid

 

[franksingleton]

Plus 1 winguru amazingly accurate I wonder why the met
Office seems to get nowhere near the same with all there computing power.
I have noticed that when windguru is I bit out, its usually because the predicted weather has come a few hours early or 27 mklater but the strength and direction changes are usually still valid

Any numerical weather prediction model can only describe and predict weather detail on a scale of 4 or 5 grid lengths. The NOAA GFS uses a grid of c 27 km. it only provides forecasts on a 50 km grid. Like all the other services that provide forecasts for specific locations, Windguru simply interpolates between grid point values. In their case, as is usual, they interpolate from the GFS 50 km grid. It cannot be a precise or accurate representation of actual or forecast weather.

Routine objective verifications of the major models shows the ECMWF model as the best, followed closely by the UK the n closely by the GFS and JMA. In practice, there is little to choose between them in the short term. ECMWF performs better overall over the longer period, up to a week or do ahead.

My main criticism of the Met Office is that they do not make available to us their model output. In the Med, Turkey covers the whole of the area with ECMWF FOrecasts. ee http://212.175.180.126/DTS/sea.php. We cannot get such data west of Gibraltar. Several countries issue detailed (Meso-scale)forecasts, see http://www.dmi.dk/en/hav/#danmark as a good example.

The UKEuro Meso-scale model runs on a 5 km grid over a large area. It will be as good or better than any other but we can only see the output when in the Southern Baltic - courtesy of the Poles. See http://www.meteo.pl/um/php/pict_show_en.php?cat=4&time=0.

I have written a letter to be published in "Weather", a Royal Met Society Journal. I am hoping that it may rouse some people to take action but am not too hopeful.

 

[Sandy]

I've found meteo consult marine app to be PDG.
I like the layout, prediction of % acuacy, strong wind warning and lots more when you start digging into it.

http://marine.meteoconsult.co.uk/shipping-forecast-services/shipping-forecast-iphone-application.php

Thanks for point me to such a brillant app.

 

[jac]

Any numerical weather prediction model can only describe and predict weather detail on a scale of 4 or 5 grid lengths. The NOAA GFS uses a grid of c 27 km. it only provides forecasts on a 50 km grid. Like all the other services that provide forecasts for specific locations, Windguru simply interpolates between grid point values. In their case, as is usual, they interpolate from the GFS 50 km grid. It cannot be a precise or accurate representation of actual or forecast weather.

Routine objective verifications of the major models shows the ECMWF model as the best, followed closely by the UK the n closely by the GFS and JMA. In practice, there is little to choose between them in the short term. ECMWF performs better overall over the longer period, up to a week or do .

Just to correct one point re windguru, they don't use the GFS 50km grid but use a range of models depending on location. Details here.

http://www.windguru.cz/int/help_index.php?sec=models

Interesting that we get a 27km grid forecast - parts of Czech Republic are on 3km grid. Start point on this though is still GFS, just more computer power

 

[alahol2]

Interesting that we get a 27km grid forecast - parts of Czech Republic are on 3km grid. Start point on this though is still GFS, just more computer power

The Windfinder superforecast quotes a 12km grid. It also says that the forecast and superforecast are based on different physical models, whatever that means.

 

[franksingleton]

I think that I am correct in saying that their free version is an interpolation from the GFS.

My problem with their Meso-scale service, as with all the other "unofficial" sources of forecasts is that I do not think that they start with a detailed analysis. From all my queries, I understand that they all start with the GFS interpolated to a finer grid. Theyu will, I guess, be able to model topographic effects but their model will not know about amy detailed weather. If I am incorrect, I would like them to tell me what observational data they are using? National centres such as the UK use as much of the data sources at http://weather.mailasail.com/Franks-Weather/Weather-Observations-Nwp. They use 3DVar or 4DVar data assimilation schemes. I have to wonder what they use?

My message to all users of automatically generated forecasts, whether from apnational weather service or a private consultant, is to use them in the light of GMDSS forecasts. These start from state of the art models. For the UK, that is a global model using a 25 pm grid, a 5 km Meso-scale model covering an area fr om mid-Atlantic to the Urals, North Cape to North Africa and a 1.5 km model covering just the British Isles. The worded texts have the benefit of human experience. I have seen too many forecasts objectively produced and dangerously wrong to advise otherwise.

 

[franksingleton]

The Windfinder superforecast quotes a 12km grid. It also says that the forecast and superforecast are based on different physical models, whatever that means.

Windfinder forecast is the GFS, pure and simple. The Superforecast most probably uses the WRF - a development model freely available from the US. I have asked them and they are quite clear: they use no observational data. Their only input is the GFS 50 km data. They are quite open about it.

ECMWF, the UK, Norway, Sweden, Denmark etc all put a vast amount of effort into data assimilation. As far as I am aware, none of the "unofficial" firms have either the ability or the knowledge to handle the vast amount of data with a wide range of characteristics and times. As has always been the case, the quality of any weather forecast depends on the quality of the data input.