Skip to content

Weather Station Core

Turn any personal weather station into a complete weather intelligence system.

Weather Station Core (ws_core) is a Home Assistant custom integration that reads raw sensor data from any HA-integrated weather station and produces 170+ derived sensors through a guided setup flow.

What it does

You provide 7 sensor entities — temperature, humidity, pressure, wind speed, wind gust, wind direction, cumulative rainfall. ws_core does the rest.

Capabilities not available in any other HA weather integration

  • Precipitation nowcast with minutes-until-rain precision (sensor.ws_minutes_until_rain)
  • UTCI (Universal Thermal Climate Index, full Bröde 2012 polynomial)
  • Three fire danger systems: Canadian FWI, McArthur FFDI (Australia), Fosberg FFWI (US/global)
  • Zambretti barometric forecast — fully local, no network call, no API
  • Penman-Monteith ET₀ for irrigation scheduling (Smart Irrigation compatible)
  • 8 upload targets in one integration (WU, Weathercloud, WOW, CWOP, and more)
  • 8 translations at full entity-name parity, including translatable config-flow selectors (hemisphere, climate region) and localized human-readable sensor states (conditions summary, alerts, frost risk)

Always-on (core features, no API key required)

What How
Zambretti forecast From your station's pressure + wind data
36-condition classifier From illuminance, rain rate, temperature, wind
Wet-bulb, frost point, frost risk Stull (2011), Buck (1981) + frost-risk category
Pressure trend WMO No. 306 least-squares regression
Kalman-filtered rain rate De-noised tipping-bucket readings
ET₀ (Hargreaves-Samani) No solar radiation sensor required
Adaptive rain probability 90-day Brier-score blended local + NWP
Streak counters Dry days, heat days, frost days

Optional feature groups

Group What it adds
Comfort Indices Heat Index, Wind Chill, Humidex, UTCI, WBGT, VPD, Delta-T, THW, THSW, chill hours, clearness index, cloud cover
Precipitation Nowcast ws_minutes_until_rain, ws_minutes_until_dry, ws_rain_next_60min, ws_rain_expected_1h
Fire Risk FWI, FFDI, FFWI, fire risk score (1-10)
Lightning Detection Strike count, distance, rate, clearance countdown, proximity state
Indoor Sensors Indoor temp/humidity/CO₂, deltas, comfort score, plus named multi-room monitoring (per-room temp delta, humidity, CO₂ and comfort)
Degree Days HDD, CDD, GDD, leaf wetness
Moon Moon phase and illumination
Air Quality AQI, NO₂, ozone (Open-Meteo, free)
Network Uploads WU, Weathercloud, PWSWeather, WOW, AWEKAS, CWOP, OWM Stations, Windy
MQTT Discovery 70+ sensors as MQTT Discovery payloads
Station Diagnostics Drift detection, spike flags, spatial QC, data-quality score

Getting started

See the Quickstart for a 5-minute installation walkthrough. Then pick your first automation from Use Cases.

Which part should I try first?

If you care about... Start with
Not getting caught in rain Precipitation Nowcast and sensor.ws_minutes_until_rain
Saving water ET0 sensors and the Irrigation Rain Skip blueprint
Protecting awnings or exterior blinds High Wind Gusts blueprint
Frost-sensitive plants or pipes Frost risk sensors and Freeze Warning blueprint
Fire season FWI, FFDI, FFWI and Fire Danger Alert blueprint
Heat safety UTCI and WBGT comfort sensors
Data trust Station Diagnostics and data-quality score

Project status