Walchensee wind: forecast and measurements
30 days of forecast vs. what actually happened at the lake. Click a single day to see its details.
Last 30 days
Rule-based forecast (actually used)
ML-based forecast โ logistic (experimental)
ML-based forecast โ HistGradientBoosting (black-box model, experimental)
Actual measurements (session โฅ 1 h)
Thu Aug 20
Fri Sep 18
session (โฅ 1 h โฅ 11 kt)
marginal (โฅ 1 h โฅ 8 kt)
no wind
no data
thunderstorm (forecast rows: predicted ยท actual row: observed)
Sep 18, 2026
For today, Sep 18, 2026 ยท
as written at the time
NO GO
100% daytime cloud + only 219 W/mยฒ solar โ no insolation to drive a thermal
๐ค ML โ logistic
experimental
NO GO
GO 12% ยท
MAYBE 33% ยท
NO GO 55%
Strongest inputs โ
(+) dew-point spread; (โ) morning solar, pressure ฮ (thermik).
๐ค ML โ HGB ยท black box
experimental
NO GO
GO 24% ยท
MAYBE 12% ยท
NO GO 65%
Most distinctive inputs: morning solar, daytime cloud, Innsbruck pressure.
Experimental shadow models โ they never drive the verdict.
How good are they? โ
How the day was forecast at the time, before the rules were recalibrated โ shown for comparison only.