New probabilistic wind speed charts

[GHA]

These look really useful.

http://www.opc.ncep.noaa.gov/prob_guidance.php?model=naefs&basin=atl&cycle=00&plot=15&loop=0#top

In addition to the output of a model run saying the wind will be this you can also have the probability of whatever wimdspeed you pick.

Looks very useful indeed.

 

[Angele]

Sorry, but if the colours on a weather chart totally hide the outline of the land to enable you to work out what is what, then I don't find that very useful.

Took me a while to work out that this is for the Atlantic. And I had to scroll down a long way before I could work out where the UK was.

 

[Quandary]

Probabilistic! word of the month for January?

 

[doug748]

These look really useful.

http://www.opc.ncep.noaa.gov/prob_guidance.php?model=naefs&basin=atl&cycle=00&plot=15&loop=0#top

In addition to the output of a model run saying the wind will be this you can also have the probability of whatever wimdspeed you pick.

Looks very useful indeed.

Great, thanks for pointing it out, I like it; probabilistically going to be a great favourite. Boom.

Would not trust the 384 hour forecast much tho.

 

[lw395]

This is 'just' the modeled forecast presented with a probability.
I.e. it is taking recent data and projecting up to 2 weeks ahead.
So it's a forecasting tool not a planning tool.
We need to be cautious of the definition of probability, if it's 90% probablity that winds will be say >10 knots, does that mean we'll spend two days in an atlantic crossing drifting?
Likewise if it's <10% prob of >30 knots, we could have two full days of gales?
It's very processed data from a model based on 'we see these lows building and moving this way at such a speed'.

Not to be confused with e.g. 'average and range of windspeeds observed on date DD/MM over the past ten years'.

 

[GHA]

Sorry, but if the colours on a weather chart totally hide the outline of the land to enable you to work out what is what, then I don't find that very useful.

Took me a while to work out that this is for the Atlantic. And I had to scroll down a long way before I could work out where the UK was.

The default 15kt isn't really a lot of use, try 30kts

 

[GHA]

Great, thanks for pointing it out, I like it; probabilistically going to be a great favourite. Boom.

This is a bit of a game changer, nothing else gives an indication of the confidence of accuracy that the models are producing.
A big deal

Someone generous will most likely have the Web addresses available in OpenCPN weatherfax plugin soon, will be a fast and powerful tool.

 

[Angele]

The default 15kt isn't really a lot of use, try 30kts

Ok, a bit better. But they really could do with making the coast a bit more obvious. Thicker outline. Could even allow the isobars to be toggled on or off. Then that would be much easier to read.

Although, I guess I just need to remember that the UK (and Europe/Africa) is somewhere on the right, and the USA is on the other side.

I do know of at least one weather site that gives an estimate of the accuracy of its forecasts. However, admittedly only locally on a "spot" basis (you get a forecast for a place and it tells you how confident it is about that forecast). Also nothing like "75% likely to be wind of more than x knots".