NOAA Weather.gov API

The National Weather Service (NWS) API provides free, open access to weather forecasts, alerts, observations, radar data, and more for the United States. All data is public domain — no API key or account required.

Always check the official API documentation and OpenAPI spec for the latest endpoint details.

Base URL

https://api.weather.gov

Authentication

No API key is needed. However, a User-Agent header is required on every request to identify your application.

User-Agent: (myweatherapp.com, contact@myweatherapp.com)

The string can be anything — the more unique to your application, the less likely it will be affected by a security event. Requests without a User-Agent may be blocked.

Key Concepts

Core Workflow: Get a Forecast for a Location

The API uses a two-step pattern to get a forecast:

Step 1: Resolve coordinates to a grid point

GET /points/{latitude},{longitude}

This returns metadata including the WFO office, grid coordinates, and URLs for forecasts.

Step 2: Fetch the forecast using the grid info

GET /gridpoints/{wfo}/{gridX},{gridY}/forecast

This returns the human-readable 12-hour period forecast (7 days).

Example (curl)

# Step 1: Get grid info for Denver, CO
curl -s -H "User-Agent: MyApp" \
  "https://api.weather.gov/points/39.7456,-104.9994" | jq '.properties.forecast'

# Returns: "https://api.weather.gov/gridpoints/BOU/63,62/forecast"

# Step 2: Get the forecast
curl -s -H "User-Agent: MyApp" \
  "https://api.weather.gov/gridpoints/BOU/63,62/forecast" | jq '.properties.periods[0]'

Endpoint Reference

Points / Location Lookup

Forecasts

Alerts

Stations & Observations

Zones

Aviation

Radar

Text Products

Offices

Miscellaneous

Response Format

The default response format is GeoJSON (application/geo+json). Control the format with the Accept header:

All times are in ISO-8601 format.

Feature Flags

Optional headers to enable new features:

Code Examples

Get Forecast for a Location (Python)

import requests

BASE = "https://api.weather.gov"
HEADERS = {"User-Agent": "(myapp, contact@example.com)"}

# Step 1: Resolve coordinates
point = requests.get(f"{BASE}/points/39.7456,-104.9994", headers=HEADERS).json()
forecast_url = point["properties"]["forecast"]

# Step 2: Get forecast
forecast = requests.get(forecast_url, headers=HEADERS).json()
for period in forecast["properties"]["periods"][:4]:
    print(f"{period['name']}: {period['detailedForecast']}")

Get Active Alerts for a State (curl)

curl -s -H "User-Agent: MyApp" \
  "https://api.weather.gov/alerts/active?area=CO" \
  | jq '.features[] | {event: .properties.event, headline: .properties.headline}'

Get Current Observations (curl)

# Get latest observation from Denver International Airport
curl -s -H "User-Agent: MyApp" \
  "https://api.weather.gov/stations/KDEN/observations/latest" \
  | jq '.properties | {
    temperature: .temperature.value,
    windSpeed: .windSpeed.value,
    description: .textDescription
  }'

Get Hourly Forecast (Node.js)

const https = require("https");

const options = {
  headers: { "User-Agent": "MyApp (contact@example.com)" },
};

// Step 1: Get grid info
https.get("https://api.weather.gov/points/40.7128,-74.0060", options, (res) => {
  let data = "";
  res.on("data", (chunk) => (data += chunk));
  res.on("end", () => {
    const forecastUrl = JSON.parse(data).properties.forecastHourly;

    // Step 2: Get hourly forecast
    https.get(forecastUrl, options, (res2) => {
      let forecast = "";
      res2.on("data", (chunk) => (forecast += chunk));
      res2.on("end", () => {
        const periods = JSON.parse(forecast).properties.periods;
        periods.slice(0, 6).forEach((p) => {
          console.log(`${p.startTime}: ${p.temperature}°${p.temperatureUnit} - ${p.shortForecast}`);
        });
      });
    });
  });
});

Find Nearby Stations (Python)

import requests

BASE = "https://api.weather.gov"
HEADERS = {"User-Agent": "(myapp, contact@example.com)"}

# Get grid info for a point
point = requests.get(f"{BASE}/points/34.0522,-118.2437", headers=HEADERS).json()
stations_url = point["properties"]["observationStations"]

# List nearby stations
stations = requests.get(stations_url, headers=HEADERS).json()
for station in stations["features"][:5]:
    props = station["properties"]
    print(f"{props['stationIdentifier']}: {props['name']}")

Rate Limits

The rate limit is not publicly documented but is described as "generous for typical use."

Error Handling

Error responses include a JSON body with type, title, status, detail, and correlationId fields.

Known Issues

• /radar/queues endpoints may return 503 errors when results are large — filter with additional parameters.
• Station observations outside the central timezone may show null 24-hour max/min temperatures.
• Observation data can be delayed up to 20 minutes from the upstream MADIS source.
• Icon endpoints (/icons) are deprecated.

Resources

Changelog

• 0.1.0 — Initial release with forecasts, alerts, observations, stations, zones, aviation, radar, and text products.