areapi

What area is this point in?

Give it a latitude and longitude, get back the Cellular Market Area, Basic Trading Area, Major Trading Area, Economic Area, Major Economic Area, Regional Economic Area Grouping and Partial Economic Area that contain it. These are the geographies the FCC uses to license wireless spectrum. It also returns the smaller FCC schemes (Regional PCS Areas, Economic Area Groupings, VHF Public Coast areas) and, for context, the county, state and Census metro/micro area.

• Website: https://areapi.netlify.app
• API: https://areapi.netlify.app/api/find?lat=38.9907&lon=-77.0261
• API docs (Swagger): https://areapi.netlify.app/docs/ (openapi.json)
• npm: npm install @brianfunk/areapi or npx @brianfunk/areapi 38.9907 -77.0261

No database, no server-side state, zero runtime dependencies. The polygons are simplified GeoJSON shipped with the package (about 10 MB for all fourteen types, loaded lazily one type at a time; the seven FCC market types alone are 3 MB) and the point-in-polygon test is forty lines of ray casting.

$ npx @brianfunk/areapi 38.9907 -77.0261
point   38.9907, -77.0261
status  OK
CMA     8  Washington, DC-MD-VA
BTA   461  Washington, DC
MTA    10  Washington-Baltimore
EA     13  Washington-Baltimore, DC-MD-VA
MEA     5  Washington
REAG    2  Southeast
PEA     5  Baltimore, MD-Washington, DC

Area types

| type | name | count | defined | source of the definition | | --- | --- | ---: | --- | --- | | cma | Cellular Market Area | 734 | 1990 | 306 MSAs (incl. Gulf of Mexico) + 428 RSAs, FCC PN CL-92-40 | | bta | Basic Trading Area | 493 | 1992 | Rand McNally Commercial Atlas, as modified by the FCC | | mta | Major Trading Area | 51 | 1992 | Rand McNally Commercial Atlas, as modified by the FCC | | ea | Economic Area | 176 | 1995 | BEA Economic Areas 1-172 + FCC 173-176 | | mea | Major Economic Area | 52 | 1995 | 47 CFR 27.6 | | reag | Regional Economic Area Grouping | 12 | 1995 | 47 CFR 27.6 | | pea | Partial Economic Area | 416 | 2014 | FCC PN DA 14-759 | | rpc | Regional PCS Area | 5 | 1994 | narrowband PCS regions | | eag | Economic Area Grouping | 6 | 1995 | 220 MHz / 700 MHz guard band groupings of EAs | | vpc | VHF Public Coast Station Area | 42 | 1998 | 9 maritime + 33 inland waterway areas | | cbsa | Core Based Statistical Area | 935 | 2025 | OMB metro + micro areas, Census 2025 boundaries | | county | County | 3,235 | 2025 | current county-equivalents, Census 2025 boundaries | | state | State | 56 | 2025 | states, DC and territories, Census 2025 boundaries | | cgsa | Cellular Geographic Service Area | 1,732 | rolling | licensed A/B block cellular service areas, keyed by call sign; a point is usually in two |

All of them are complete, including Puerto Rico, the U.S. Virgin Islands, Guam, the Northern Mariana Islands and American Samoa. The Gulf of Mexico (CMA 306, EA 176, MEA 52, REAG 12) is a water-only market; its polygon is the U.S. part of the Gulf from Marine Regions, which runs from the coastline out to the EEZ limit. (The FCC draws the EA/MEA/REAG Gulf boundary 12 nautical miles offshore rather than at the coast; that strip is attributed to the Gulf here.)

HTTP API

Interactive docs with try-it-out at /docs/; the OpenAPI 3 spec is site/openapi.json.

GET /api/find?lat=<lat>&lon=<lon>[&types=cma,bta,...][&format=json|xml|jsonp][&callback=fn]
GET /api/<type>/find?latitude=<lat>&longitude=<lon>
GET /api/types

| parameter | required | values | default | | --- | --- | --- | --- | | lat (or latitude) | yes | -90 to 90 | | | lon (or longitude, lng) | yes | -180 to 180 | | | types | no | comma-separated subset of the types above | all | | format | no | json, xml, jsonp | json | | callback | no | JSONP callback name | callback |

{
  "status": "OK",
  "point": { "lat": 38.9907, "lon": -77.0261 },
  "types": ["cma", "bta"],
  "areas": [
    { "type": "cma", "id": "8", "name": "Washington, DC-MD-VA", "asOf": "2026-10-08", "defined": "1990" },
    { "type": "bta", "id": "461", "name": "Washington, DC", "asOf": "2026-10-08", "defined": "1992" }
  ],
  "executionTime": 1.2
}

Each area carries asOf, the date the dataset was built and confirmed against its source (for the rolling CGSA licenses, the FCC's own as-of date), and defined, the year the FCC (or OMB) created that scheme. (vintage, the 1.0 name for defined, is still returned through 1.x but deprecated.) The FCC market definitions from the 1990s are still the ones in force; asOf is the date that was last checked.

status is OK, NONE (no area contains the point) or ERROR (HTTP 400 with an error message). Responses are CORS-enabled and cacheable for a day. The 2017 URL shape /api/<type>/<year>/find still works; the year is ignored.

Library

import { find, feature, types } from '@brianfunk/areapi';

const r = await find({ lat: 38.9907, lon: -77.0261 });                 // every type
const r = await find({ lat: 38.9907, lon: -77.0261 }, { types: 'cma' }); // one or more
const f = await feature('cma', 8);   // GeoJSON Feature with geometry and bbox, for drawing
const m = await types();             // manifest: counts, as-of dates, sources

Ships TypeScript declarations. Works in Node 20+ and in the browser. In the browser, call setDataUrl('/path/to/data/') so it knows where to fetch the <type>.json files from; they are loaded lazily, one type at a time, and cached.

CLI

areapi <lat> <lon> [--types cma,bta,...] [--format table|json|xml]
areapi --list

Prints a table on a terminal and JSON when piped. Exit code 0 when at least one area matched, 3 when none did, 2 on bad input.

How the data is built

Every FCC market area is an aggregation of county-equivalents, so the polygons are built by dissolving county boundaries with the FCC's own county-to-market crosswalk rather than by redistributing the FCC shapefiles (which the FCC no longer publishes for most types).

| input | source | | --- | --- | | County boundaries | U.S. Census Bureau, Census 2000 generalized counties (co99_d00), plus the 2020 1:500k cartographic file for the four territories missing from the 2000 file | | Current counties, states, CBSAs | U.S. Census Bureau, 2025 cartographic boundary files 1:500k (the market layers deliberately stay on 2000 counties: that is what the FCC definitions are built from) | | County → market crosswalk | FCC OET FCCCNTY2K.txt (CMA, BTA, MTA, EA, MEA, REA, RPC, EAG, VPC per county FIPS, plus county and state names) | | PEA polygons | FCC FCC_PEAs_Website.zip shapefile | | CGSA polygons | FCC WTB Cellular Geographic Service Area map data (A and B block GeoJSON), downloaded in a browser since fcc.gov blocks scripted requests | | Gulf of Mexico | Marine Regions (VLIZ) EEZ × IHO sea areas, "United States part of the Gulf of Mexico" (MRGID 25281) | | Market names | FCC Universal Licensing System public data (l_market.zip, market table) |

To rebuild from scratch:

mkdir -p data/raw && cd data/raw
curl -O https://www2.census.gov/geo/tiger/PREVGENZ/co/co00shp/co99_d00_shp.zip
curl -O https://www2.census.gov/geo/tiger/GENZ2020/shp/cb_2020_us_county_500k.zip
curl -O https://www2.census.gov/geo/tiger/GENZ2025/shp/cb_2025_us_county_500k.zip
curl -O https://www2.census.gov/geo/tiger/GENZ2025/shp/cb_2025_us_cbsa_500k.zip
curl -O https://transition.fcc.gov/bureaus/oet/info/maps/areas/data/2000/FCCCNTY2K.txt
curl -O https://transition.fcc.gov/bureaus/oet/info/maps/areas/data/FCC_PEAs_Website.zip
curl -O https://data.fcc.gov/download/pub/uls/complete/l_market.zip
curl -o eez_iho_gulf.json "https://geo.vliz.be/geoserver/MarineRegions/wfs?service=WFS&version=1.0.0&request=GetFeature&typeName=MarineRegions:eez_iho&outputFormat=json&CQL_FILTER=marregion%20ILIKE%20%27%25Gulf%20of%20Mexico%25%27"
# A_block_CGSA.json and B_block_CGSA.json: download in a browser from
# https://www.fcc.gov/wireless/bureau-divisions/mobility-division/cellular-service/cellular-geographic-service-areas
unzip -p l_market.zip MK.dat | awk -F'|' '{print $6 "|" $9}' | sort -u > uls-markets.txt
cd ../..
node scripts/extract-names.js   # -> scripts/names.json
npm run build:data              # -> data/*.json, data/manifest.json

Geometry is simplified (SIMPLIFY=5% by default, visvalingam weighted, shapes preserved) and rounded to 4 decimal places, which is about 11 m. That is plenty for deciding which market a point is in, and is what keeps the whole dataset around 10 MB, 4 MB of which is the CGSA license polygons. data/manifest.json records the sources, counts, and the ids that have no polygon.

Development

npm install
npm test            # node:test, no framework
npm run build:site  # bundle src/ for the browser + copy data into site/
npm run dev         # local server: site + /api/* on a free port (prints the URL); --port N to pin one

npm run dev is a small Node server (scripts/dev.js) that serves site/ and routes /api/* to the same handler Netlify runs, so it never collides with other local servers and does not need the Netlify CLI. Production is Netlify, deployed from the dev branch: the site is static files, the API is one Netlify Function wrapping the same library. Merging to main publishes the package to npm and cuts a GitHub release automatically when the version is new.

License

MIT