Click the map to compute local circumstances.
Engine: Besselian elements fitted to JPL DE440.
Ephemeris & ΔT
Lunar radius k
k2 < k1 is deliberate: light leaks through limb valleys at
C2/C3, so the effective umbral Moon is smaller. A single mean k
misclassifies beaded annulars as totals.
Solar radius
Nearly irrelevant on the centreline (~2 s); dominant near
the path edge (~20 s).
Observer
Sky view
—
Discs are drawn at refracted altitude — where they
actually appear. Skyline is ray-traced from Copernicus GLO-30 with Earth
curvature + refraction (k=0.13).
Live sky (now, not eclipse day)
True colour is reflected sunlight: it fades as the
Sun drops and is blank at night. The cloud mask is infrared, so it still works
at the 2–8° Sun of C2/C3 — trust it at eclipse hour.
Arrows point where the smoke goes, not where
the wind comes from. A plume rides the wind at ~1–3 km, which is not the 10 m
wind a weather station reports — over Burgos those differed by 23°, i.e. ~40 km
of plume displacement at 100 km. Model field (Open-Meteo), not stations.
nothingclosuresbadall
DGT loop detectors (~1 min old) colour each
diamond by how fast the road is actually moving; ✕
is a closure. Closures come from DGT national + Euskadi + Catalonia — DGT
excludes those two regions and both are on the path.
Both follow the map: pan or zoom and they reload for the new
view. This is today's sky. It forecasts nothing about 12 Aug — it is
here because in the last 24 h before totality, satellite nowcasting beats
every model, and because fire smoke is an August hazard in Spain that no
cloud forecast contains.
Terrain scan (grid)
Scores every ~30 m cell: can it see the Sun at C3
(the Sun is setting, so C3 is the binding constraint). Ray-marched toward
the Sun's azimuth only.