Data Sources¶

climagrid fetches from five U.S. government open-data APIs. All are free and require no registration except NOAA NCEI CDO (free token, issued in seconds).

NASA POWER¶

Module: climagrid.sources.nasa_power.NasaPowerAdapter

NASA’s Prediction of Worldwide Energy Resources (POWER) project provides hourly surface meteorology derived from the MERRA-2 atmospheric reanalysis at any lat/lon point globally.

Property	Value
Temporal resolution	Hourly
Spatial resolution	~50 km (MERRA-2 native)
Latency	~3 days
Coverage	Global
API key	Not required
Endpoint	`https://power.larc.nasa.gov/api/temporal/hourly/point`

Columns produced:

Column	Units	Description
`nasa_temperature_2m`	°C	2-metre air temperature
`nasa_wind_speed_10m`	m/s	10-metre wind speed
`nasa_solar_irradiance_ghi`	W/m²	Global horizontal irradiance
`nasa_relative_humidity_2m`	%	Relative humidity at 2 m
`nasa_precipitation`	mm	Hourly precipitation

Best for: Global coverage, no API key, consistent long historical records back to 1981. Preferred for rural assets outside HRRR coverage or for long-term trend analysis.

NOAA HRRR¶

Module: climagrid.sources.noaa_hrrr.HrrrAdapter

NOAA’s High-Resolution Rapid Refresh (HRRR) is a 3-km CONUS numerical weather prediction model updated hourly. It is the highest-resolution publicly available NWP product over the continental United States.

Property	Value
Temporal resolution	Hourly
Spatial resolution	3 km
Latency	~1 hour
Coverage	CONUS only
API key	Not required
Library	`herbie-data` (optional dep `[noaa-nwp]`)

Columns produced:

Column	Units	Description
`hrrr_temperature_2m`	°C	2-metre air temperature
`hrrr_wind_speed_10m`	m/s	10-metre wind speed
`hrrr_wind_direction_10m`	degrees	Wind direction (meteorological)
`hrrr_relative_humidity_2m`	%	Relative humidity
`hrrr_precipitation_rate`	mm/hr	Hourly precipitation
`hrrr_solar_irradiance_ghi`	W/m²	Downward shortwave radiation
`hrrr_snow_depth`	m	Snow depth

Best for: High spatial detail within CONUS. Required if your predictive model needs to distinguish neighboring substations separated by less than 50 km. Requires pip install "climagrid[noaa-nwp]".

NOAA NCEI CDO¶

Module: climagrid.sources.noaa_ncei.NceiAdapter

NOAA’s National Centers for Environmental Information Climate Data Online (CDO) API provides historical surface station observations from the Global Historical Climatology Network (GHCN) and Integrated Surface Database (ISD).

Property	Value
Temporal resolution	Hourly or daily
Spatial resolution	Station-based (nearest station matched)
Latency	~1 day
Coverage	Global
API key	Free, register at ncdc.noaa.gov/cdo-web/token
Env var	`NOAA_CDO_TOKEN` (set this and the adapter picks it up automatically)

Setup:

export NOAA_CDO_TOKEN="your_token_here"

Or pass it directly:

from climagrid.sources.noaa_ncei import NceiAdapter
adapter = NceiAdapter(api_token="your_token_here")

Columns produced:

Column	Units	Description
`ncei_temperature_max`	°C	Daily maximum temperature
`ncei_temperature_min`	°C	Daily minimum temperature
`ncei_wind_speed`	m/s	Average daily wind speed
`ncei_precipitation_daily`	mm	Daily precipitation
`ncei_relative_humidity`	%	Average daily relative humidity

Best for: Ground truth validation. Actual station measurements vs. model-derived data from NASA POWER or HRRR. Use for bias correction or audit trails in regulatory filings.

USDA NRCS AWDB¶

Module: climagrid.sources.usda_nrcs.NrcsAdapter

The U.S. Department of Agriculture’s Natural Resources Conservation Service (NRCS) operates the Soil Climate Analysis Network (SCAN) and SNOwpack TELemetry (SNOTEL) networks. The Automated Water Database (AWDB) REST API provides hourly soil moisture, soil temperature, and snow water equivalent from ~900 stations across the western U.S.

Property	Value
Temporal resolution	Hourly
Spatial resolution	Station-based
Coverage	Primarily western CONUS
API key	Not required
Endpoint	`https://wcc.sc.egov.usda.gov/awdbRestApi/services/v1`

Columns produced:

Column	Units	Description
`nrcs_soil_moisture_pct`	%	Volumetric soil moisture
`nrcs_soil_temperature`	°C	Soil temperature at 2-inch depth
`nrcs_snow_water_equivalent`	mm	Snow water equivalent
`nrcs_station_distance_km`	km	Distance to nearest NRCS station

Best for: Soil saturation risk (ground stability for pole foundations, underground cable performance), spring snowmelt flood risk, cold-climate cooperative service territories.

USFS NIFC WFIGS¶

Module: climagrid.sources.usfs_wfigs.WfigsAdapter

The National Interagency Fire Center (NIFC) Wildland Fire Incident Geospatial Service (WFIGS) provides near-real-time wildfire perimeters as GeoJSON polygons.

Property	Value
Temporal resolution	Updated multiple times daily
Coverage	CONUS + Alaska + Hawaii
API key	Not required
Endpoint	`https://services3.arcgis.com/T4QMspbfLg3qTGWY/...`

Columns produced:

Column	Units	Description
`wfigs_nearest_fire_km`	km	Distance to nearest active fire perimeter edge
`wfigs_fire_active`	bool	Any active fire within 50 km
`wfigs_fire_area_ha`	ha	Area of nearest active fire

Best for: Western U.S. cooperatives, transmission lines traversing WUI (wildland-urban interface) terrain, rapid situational awareness during red-flag conditions.

The compute_proximity() standalone function can be used independently of the adapter to compute fire risk scores from a pre-fetched perimeter DataFrame.