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 — logistic (experimental)
ML-based forecast — HistGradientBoosting (black-box model, experimental)
Actual measurements (session ≥ 1 h)
Mon Oct 12
Tue Nov 10
session (≥ 1 h ≥ 11 kt)
marginal (≥ 1 h ≥ 8 kt)
no wind
no data
thunderstorm (forecast rows: predicted · actual row: observed)
Oct 31, 2026
avg wind
gust
ignition (8 kt)
session (11 kt)
For Oct 31, 2026 ·
as written at the time
MAYBE
dew-point spread 0.9°C — air too moist, solar energy lost to evaporation
🤖 ML — logistic
experimental
NO GO
GO 33% ·
MAYBE 17% ·
NO GO 50%
Strongest inputs —
(+) dew-point spread; (−) pressure Δ (thermik), morning solar.
🤖 ML — HGB · black box
experimental
NO GO
GO 16% ·
MAYBE 5% ·
NO GO 79%
Most distinctive inputs: Innsbruck pressure, morning solar, daytime cloud.
Experimental shadow models — they never drive the verdict.
How good are they? →
How the day was forecast at the time, before the rules were recalibrated — shown for comparison only.