Weather forecasting

Forecasting the weather is essential for paragliding and we have listed a number of free sources which you may find useful. – available as an app too. Great for forecasting everything (except thermal strength and soundings (use RASP)). Has different weather models – ECMWF, GFS, ICON-EU7, NEMS4.

Met Office – surface pressure charts are a good place to start with a general overview of the weather systems.

RASP models – These provide GFS forecasts with thermal models and soundings. Use either:

RASP site or XC Para or for the lazy use lazy RASP

XC Weather – still commonly used although the forecast resolution is not high (runs off GFS). It is most useful for live data, although the live data can typically be 1 hour or more old so local weather stations with trend data is usually better.

Brecon Beacons Mountain forecasts


Forecast models explained

Some important global weather forecasting models are GFS (US run), WRF (also US run), ECMWF (the “European model), the Met Office’s Unified model (run from out of Exeter) and NEMS4 (from Swiss meteoblue). ECMWF has generally been shown to be more accurate than other models at forecasting over 5 to 10 days, although at 2 days out there is little difference in which model will be most accurate.

GFS will react to changing weather systems quicker as GFS updates most frequently at 6 times per day whereas ECMWF and NEMS4 are updated twice per day.

The higher the resolution the forecast (i.e. 9 km is higher than 27 km), the more accurate the forecast will take account of topographical (e.g. mountain) and local (e.g. sea breeze) effects. This is especially true of mountainous areas like the Alps where the NEMS4 forecast may be most useful. NEMS4 has a resolution of 4km, compared with ECMWF at 9 km and GFS at 27 km. The GFS forecasts used in RASP (including thermal forecasts) are run at 2, 4 or 12 km depending how far out the forecast is.

Top Tips!

It’s a good idea to first look at the overall weather system – using the synoptic charts. Then try looking at the different forecast models. If they are all in agreement, it is less likely there will be change in the forecast and the forecasts are more likely to be accurate. If there is a big difference between the forecasts then this is probably because there is an important weather event which will impact significantly on the forecast (such as formation of a new low, or passage of a frontal system) – you should be able to identify this on the synoptics.

Make sure you don’t confuse the forecast with the live data (or “actuals”). You can find out what the weather is actually doing right now by looking at weather stations and webcams – check out out weather stations web page, or by using the SEW site reports group on Telegram. Also correlating the satellite images with the forecast can be useful for XC flying to determine where the good air masses and cloud cover are.

Remember the models only produce a prediction and do not take into account local conditions (especially when the resolution of the forecast is not as good). The forecast might give an accurate prediction of how the wind is flowing up high, but can’t necessarily predict how the wind is flowing around the hills and valleys down low. You need to imagine how the hills and valleys are funnelling, compressing and diverting the wind. After a while you will get better at correlating the forecast with what actually happens on the hill.

Please let us know if you think there’s some useful key information we should include on this page.