New Internet Weather Forecast Website

U

usbm

New Member
Joined
17 Mar 2007
Messages
3
Visit site
Hi everyone,

I'm a university student currently working on a dissertation as part of my Geography degree.

I'm trying to find ways to improve the presentation and overall quality of internet weather forecasts - making them more user specific and applying the forecasted weather conditions to where you'll be and what you'll be doing. Ultimately, I would like to take all of this feedback and make my own internet weather forecast website, based on the comments and suggestions made.

As part of my background research, I'm just trying to get an insight into what you all think makes the perfect weather forecast for people who sail yachts and boats.

If you could spare a couple of seconds quickly noting down ideas regarding the following points, I'd be really grateful!

- What do YOU as a sailor require from an online weather website?
- What are some of the problems you've encountered with them?
- What changes would you suggest be made to them?
- What would you like to see more /less of?

Many thanks!!
usbm.
 
[ QUOTE ]
I'm trying to find ways to improve the ...overall quality of internet weather forecasts

[/ QUOTE ]
Please do update your Bio with some more information about yourself; this looks too much like a troll from an existing weather service.
 
Hi usbm,

You'll find most keen sailors are remarkably well informed about the weather, and often the thing most lacking is detail..... it would be nice for example to see wind direction and strength forecasts (perhaps in the format/style of a GRIB) alongside a synoptic chart or superimposed as some GRIBS formats are...

And then trends as well as fine detail...

For example, we all tend to know enough to know that its near impossible to predict precise wind strength four days in advance, but perhaps something that explains how a front coming through would affect the weather, eg wind will veer and freshen under the front and then back and decrease, along with an associated temperature drop for example.... and then we can watch the front advancing on various synoptic sources, with a better idea of what to expect from it....

Also.... sea state is equally important to sailors... and this is rarely covered in the main forecasting sites, along with visibility....

Final observation..... it would often be useful to know what model is used, eg is it GFS etc and it would make a huge difference to many forecast sites if we knew what time the data presented was run/calculated.... if its 24hours out of date, we know to treat it with more caution, plus can compare the forecast from several sources with a better appreciation for whether they are different interpretations, or just different time runs of the same model....

HTH
 
Hiya, thanks very much for your comments - thats exactly the sort of thing I'm looking for thanks.

The whole idea is that people can access weather information relating to their location and personal needs. The underlying idea is that people need different things from a weather forecast, not only in terms of its content but its presentation too.

So thanks very much again for your comments!

I hope to get an online questionnaire up soon to research this in more detail, and to eventually build a website that will take all of these commments and suggestions into account - aiding sailors in the long term.

Thanks again - please keep the comments coming!
usbm.

PS: I dont understand the second post here about trolls?
 
And it has to be free - none of this subscription cobblers. Perhaps it could be funded by the MCA or even RNLI as a call-out preventative service.
 
"PS: I dont understand the second post here about trolls? "

Suppose you ran a subscription weather service and wanted to publicise it on this forum (where you are not allowed to advertise). Then a good wheeze would be to ask the question that you have just asked, then say that you know just the site...

Your question:
I second all that's been said, particularly about stating the basic model and time. Many of us compare all the forecasts we can get hold of, and that gives us an idea of what is likely to happen, but also some measure of the uncertainty on the predictions. If, however, they are all using the same model, then we get a set of very similar forecasts, even if the situation is actually very uncertain.

Actually, I'd like error bars on each individual forecast (eg SW (+/- 45 degrees) F4 (+/- 1) gusting F5 (+/- 1)), but that is getting very cumbersome, especially when combined with the prediction based on another model which says (eg SSE (+/- 75 degrees) F5 (+/- 2) gusting F6 (+/- 2)). Simon Keeling is very good at telling us what is pretty certain and what he is less sure about - thanks Simon!
 
<< Also.... sea state is equally important to sailors... and this is rarely covered in the main forecasting sites, along with visibility.... >>

Absolutely. Two years ago I was stuck, along with many others, in Menorca waiting for an opportunity to head north to mainland Spain or France. A mistral was blowing itself out and there were large swells, 3 metres or so, pounding against the north coast. A French boat alongside me was using a subscription grib file service that included swell size. His predictions were remarkably accurate and consequently we waited an additional day, when the swell size was said to reduce from 1.5 metres first thing to insignificant later in the day. This proved to be exactly right. Unfortunately I never asked him for the source of the information.

I would welcome being able to receive this info in the Med, not the Atlantic where it is available from USA sources.
 
Few boats are equipped to access the internet whilst at sea, especially so for the graphics-heavy content that modern forecasting seems to love. So a site that concentrated on detailed short-term inshore forecasts with a plenty of granularity would suit the marina-hoppers very nicely. Granularity? Well, often there'll be different winds in the Eastern, Central and Western Solent, for example
 
[ QUOTE ]
A French boat alongside me was using a subscription grib file service that included swell size. His predictions were remarkably accurate and consequently we waited an additional day, when the swell size was said to reduce from 1.5 metres first thing to insignificant later in the day. This proved to be exactly right. Unfortunately I never asked him for the source of the information.

I would welcome being able to receive this info in the Med, not the Atlantic where it is available from USA sources.

[/ QUOTE ]

vyv,
the French system you mention is most likely Navimail, it is a subscription service offered by Meteo France (more info on their web site)

The underlying model is a mesoscale one (Aladin, a NH model), whereas most (all?) free available models from the US (like GFS, etc) are large scale synoptic models.

Mesoscale models are particularly efficient in the mediterranean as they take into account many of its peculiarities (mistral being one of the best known), whereas synoptic models may often be incapable of discerning the real extent of subsynoptic scale phenomena.

cheers
 
[ QUOTE ]
Mesoscale models are particularly efficient in the mediterranean as they take into account many of its peculiarities (mistral being one of the best known), whereas synoptic models may often be incapable of discerning the real extent of subsynoptic scale phenomena

[/ QUOTE ] Ooooh. Some lovely words there.

A very good example of this more detailed forecasting which takes account of diurnal and physical local effects is the Greek Posiedon system - Greek Sea area forecasts
Another example (polish up on your Greek alphabet first) is Greek meteo system

In these detailed sites, you always have to ask the question first, how did they create the models showing these variations from the synoptic data? Observation? Inspired guesswork? The detail looks great, but it might be misleading.
 
I sometimes wish the forecasters would install some really hi-tec equipment, such as a window. One look would tell them half of what they need to know. This morning's forecast promised a dry, warmish day for southern UK. As I listened to the radio, here in Portsmouth, I was watching the drizzle falling on the dinghy I was all preparing to spend the morning painting.....
 
[ QUOTE ]
In these detailed sites, you always have to ask the question first, how did they create the models showing these variations from the synoptic data? Observation? Inspired guesswork? The detail looks great, but it might be misleading.

[/ QUOTE ]

It depends what kind of details one is talking about.
A synoptic model usually takes into account phenomena between 100 and 1000km, whose evolution spans over periods of 24 hours
A mesoscale model takes into account phenomena whose size is between 10 and 100 km, with periods between a few hours and one full day.

Is is perfectly possible to interpolate datas from a synoptic model up to one cm spatial resolution to produce weather charts with time spans of a few seconds, but of course it would have little if any practical relevance.

With mesoscale models on the other hand, if one respects the detail scale they are built to provide (10-100km), even if far from perfect, their accuracy in the mediterranean will most probably be consistently better than synoptic models

IIRC (but to be confirmed) the Poseidon model from greece has been derived from ETA models, they are both mesoscale models so by definition more detailed than other synoptic models


/forums/images/graemlins/smile.gif
 
Thanks a lot for that, I was most impressed with the accuracy and often regret not finding out what it was.

Sorry for hijacking the thread but you can rest assured that it has been worthwhile /forums/images/graemlins/smile.gif
 
[ QUOTE ]
A French boat alongside me was using a subscription grib file service that included swell size. His predictions were remarkably accurate and consequently we waited an additional day, when the swell size was said to reduce from 1.5 metres first thing to insignificant later in the day. This proved to be exactly right. Unfortunately I never asked him for the source of the information.

I would welcome being able to receive this info in the Med, not the Atlantic where it is available from USA sources.

[/ QUOTE ]Very probably Buoyweather http://www.buoyweather.com/index.jsp

It does all you say and much, much more. There is a special Med model as well, and you can select the model you want and get it in grib format via email, by winlink and sailmail. A year's sub costs less than a main course in a Spanish restaurant and it is a sound investment. However I don't rely on it much. In Spanish waters I use the inm forecasts supplemented by the German dwd for the Med.

I was stuck in Mahon last August, wanting to go to Sardinia. We were waiting for friends to arrive and couldn't leave for a couple of weeks so for practice I obtained weathers from all sources including France Meteo (I managed to get wifi out by Isla Cuarantina!). On several occasions we had weather good for Sardinia - on those forecasts we would have left harbour at dawn. On each of those occasions a full gale with nasty gusts arrived less than six hours later. We would have been hammered. I decided that the weather conditions were not stable enough to be forecastable by any weather service, so we gave up the idea of Sardinia (at least, for that year) and sailed back to the mainland.
 
I have been looking at buoyweather but this is not the one that I saw in Menorca. That was a grib file service, which is what I am looking for. Access to Internet by wifi seems to be getting more rare as we move eastwards, coupled to the fact that we use marinas as little as possible. This means that GPRS is my No.1 means of accessing weather info, for which gribs are ideal. I have begun to prepare for the Meteo France system by downloading the Maxsea reader yesterday. Will sign up in the next day or so, costs are remarkably low at 4 Euros per month for 4-day forecasts.
 
[ QUOTE ]
I have been looking at buoyweather but this is not the one that I saw in Menorca. That was a grib file service, which is what I am looking for. Access to Internet by wifi seems to be getting more rare as we move eastwards, coupled to the fact that we use marinas as little as possible. This means that GPRS is my No.1 means of accessing weather info, for which gribs are ideal.

[/ QUOTE ]I think you'll find that you can get data from BW in grib format. You can also set it up to email you on 'subscription' (free, included in your fee) at scheduled times so all you need is to be able to receive text emails. Normally we don't have internet and I access the service by very low bandwidth email - winlink and sailmail. They might have changed it since last year - I haven't crawled over it yet this year.
 
Sailing in the Baltic I generally have access in most harbours to Danish, Swedish and German weather forecasts. The German - wetteronline.de - is ceratinlt the mosdt detailed, with a 7-day forecast, which contains a map of the area, showing windstrengths hour for hour. and also wave heights. Bothe the Danish and the Swedish are less detailed, but are generally fairly correct. If you want more details, send me a PM
 
You must log in or register to reply here.

Share this page

Top