For many years Unit of Wyoming has been the main/only source (select Europe / GIF/Skew-T) for these.
They are very useful because you get real data at 0000Z and 1200Z e.g. this is showing a very low base, with tops at 900mb and great vis above
The details are here.
I now see the UK is no longer supplying the Herstmonceux station data. France and Spain are supplying nothing although I don’t think this was always the case. So while windy.com is publishing previously “top secret” wx data from various national models in Europe, the Skew-T data is being shrunk.
I found e.g. this site but it just says Data not available.
German DWD provides the (pay) software “pc-met”. There’s a clickable map of accessible actual measurements taken:
The software also provides forecasted values, that seem to be based on the actual measurements. I use it extensively, was used to it when I flew gliders, and now I’m trying to guesstimate clouds and in particular ice from them. However already failed on it 
Camborne in the U.K. looks like this:
Windy:
Interesting.
I found the skew-t layer on Windy.com and yes it is great in that it shows France and Spain (basically what DWD shows) but mysteriously Herstmonceux is shown but with old data
The radiosonde layer is not easy to find
I am running on a PC however.
Based on the history of lanuches of that radiosonde location it looks like they’re not launching it every day of the week from there. Not sure why though.
Peter wrote:
I now see the UK is no longer supplying the Herstmonceux station data. France and Spain are supplying nothing although I don’t think this was always the case.
The problem is that lots of met organisations are actually abandoning the soundings in favour of model soundings instead. Not sure right now which ones still are regular but we lost quite a few over the years now.
I will try to find out more about it. Ours at Payerne is still going regularly.
Seems to me like their ‘Computer modelling can give results suitable for their purposes without the actual cost of sending up a balloon. For them what’s not to like. It’s probably a big saving.
In reality it’s only light GA that need the finer detail.
Those on the ground are probably well served by the current offerings and anything from bizjet upwards probably doesn’t really care.
It’s a shame as I use if for additional preparation/understanding when doing IFR in marginal to poor wx conditions
The idea behind the soundings has always (well since computers i.e. 1950s) been to get data points for the computer models.
A skilled meteorologist can read useful data directly e.g. the lapse rate driving convection and CB etc formation, rather than getting the cloud tops which are not really useful to anybody outside GA. Practically all commercial traffic is IFR start to finish, and Europe does not generally produce wx data for GA, beyond ICAO obligations.
