🪂 Let's jump

Live conditions + forecast-based GO / CONSIDER / NO-GO for two Swedish dropzones.

Quick start · How it works · Container

"Should I drive ~70 min to the dropzone?" — one page that combines current observations with a GO / CONSIDER / NO-GO forecast for FK Aros (Västerås) and Skydive Stockholm (Gryttjom), for today and tomorrow. Wind, gusts, rain, cloud and thunder, scored hour-by-hour over the 09:00–21:00 window. The current hour prefers available live observations for wind, gust, direction, and temperature; later hours remain Open-Meteo forecasts. Temperature does not affect the verdict.

Decision aid only — always confirm actual ops and conditions at the DZ.

Quick start

# docker-compose.yml
services:
  letsjump:
    image: ghcr.io/d-ismlv/letsjump:latest
    ports: ["32325:32325"]
    restart: unless-stopped

docker compose up -d          # → http://localhost:32325

Stateless: nothing to persist, no secrets to set. Or run it locally:

npm install && npm run dev    # → http://localhost:3000

How it works

Everything runs server-side, in four steps:

1. Observe now. Pull the latest aviation observation from ESOW for Aros and ESCM as regional context for Gryttjom. Gryttjom's public weather page also supplies its onsite maximum gust over the last 30 minutes. The source and distance are shown, and both cards link to the DZ's live wind page.
2. Fetch the outlook. Pull the hourly forecast from Open-Meteo for each airfield's coordinates, using MET Nordic where available and best-match fields for the rest.
3. Score each hour. Every hour from opening until closing is rated go / consider / no-go on the worst of five factors: wind, gusts, rain, thunder, and cloud. Elapsed hours stay visible but muted and are excluded from the remaining-day outlook.
4. Reconcile. Live wind and gusts replace modeled current-hour values when available, and an adverse aviation ceiling can downgrade today's headline. Future-day tabs never show or apply current observations. Contradictory or incomplete forecast fields never score GO. The remaining hours become one outlook per dropzone. A short recoverable wind hold does not downgrade an otherwise strong day. Every usable window is retained (for example 09–13 · 16–20); terminal wind starting at least three hours before closing is flagged as a likely early stop.

Verdict	What it means
🟢 GO	Live conditions and a solid forecast window are within limits
🟡 CONSIDER	Jumpable but unsettled: showers possible, go early or watch the radar
🔴 NO-GO	A real stopper all day

The hard part is summer storms. A plain forecast can read "0 mm" at the field while a cell dumps rain a few kilometres away, so the engine also weighs rain probability and atmospheric instability (CAPE), not just millimetres. It also matches how jumpers actually behave: a shower chance only drops a day to CONSIDER (you jump the holes between clouds), while thunder, steady rain, or a solid cloud deck are the real NO-GO. An unexplained total-cloud/low-cloud disagreement is marked uncertain instead of silently becoming green. Mean wind and gust spread are scored separately: exactly 10 m/s mean wind is CONSIDER and anything above 10 m/s is NO-GO, before the formal 11 m/s operational hold. Maximum gust remains visible as data; a same-source spread below 7 m/s stays green, 7–8 m/s is CONSIDER, and above 8 m/s is NO-GO. The live METAR box also shows the lowest reported cloud base, toggling between feet and metres. All the thresholds live in LIMITS (lib/decision.ts) and are still being tuned against real jump days.

_{Forecast data © Open-Meteo (CC BY 4.0). Live aviation observations via AviationWeather.gov; onsite gust via Skydive Stockholm. Not an operational authority — check Skyview, onsite wind and SMHI radar + lightning.}