Geographic Names Information System (GNIS)
The Geographic Names Information System (GNIS) is a comprehensive gazetteer—a geographical dictionary or directory—containing name and location data for more than two million physical and cultural features. While primarily focused on the United States, its scope extends to the Marshall Islands, Micronesia, Palau, and Antarctica.
Developed by the United States Geological Survey (USGS) in collaboration with the United States Board on Geographic Names (BGN), the GNIS was created to standardize feature names across federal records. The system integrates topographic map names and bibliographic references, citing books and historic maps to verify place names. To ensure consistency, every feature is assigned a permanent, unique GNIS identifier. Once an entry is created, it is never removed unless it is an obvious duplicate.
Key Facts
- Scope: Covers over 2 million features in the US, Antarctica, and several Pacific island territories.
- Management: Jointly developed by the USGS and the US Board on Geographic Names (BGN).
- Identifiers: Each feature has a unique, permanent record identifier.
- Data Integrity: Entries are permanent except in cases of duplication.
- Public Input: New or changed names can be proposed via The National Map Corps.
Development Phases and Database Structure
The GNIS was built in stages to compile a massive array of geographical data from various sources.
Phase 1 (1978–1981)
Following a 1976 pilot project in Kansas and Colorado, Phase 1 produced five databases. This phase focused on USGS topographic maps but excluded features better documented elsewhere, such as airports, roads, individual buildings, and radio/TV masts. Early data was distributed via paper volumes, microfiche, and magnetic tape.
One notable error during this phase was the undocumented use of the feature class "tank" for lakes. A 1986 report by the US Army Corps of Engineers noted that large bodies of water, such as Cayuga Lake, were incorrectly labeled as "tanks," which the report assumed referred to reservoirs.
Phase 2 and Beyond
Phase 2 expanded the scope of data sources significantly. By 2003, data from 42 states had been fully processed. A proposed third phase intended to address discrepancies between local usage and official maps but was never initiated.
Specialized Databases
- National Geographic Names Database (NGNDB): Originally 57 files (one per state/territory), including a specific 1967 dictionary for Alaska. It includes subsets for the 50,000 most well-known features.
- USGS Topographic Map Names Database (TMNDB): Contained map names exactly as they appeared on 1:24,000, 1:100,000, and 1:250,000 scale maps. This was renamed the Geographic Cell Names Database (GCNDB) in the 1990s.
- Board on Geographic Names Database: A record of BGN Domestic Names Committee decisions since 1890 and names established by Acts of Congress.
- Antarctica Geographic Names Database (AGNDB): Added in the 1990s, this covers BGN-approved natural features in Antarctica and surrounding islands (e.g., South Georgia, Heard Island), excluding scientific outposts.
Data Accuracy and Populated Places
The GNIS does not differentiate between the size or significance of populated places; a tiny subdivision is treated with the same weight as a major city like New York City. This lack of granularity has led to some inaccuracies.
Research from the University of Connecticut in 2001 found that many "populated places" in Connecticut were actually road intersections or cartographic locators (survey corners marked by monuments or "bearing trees"), often identified by the word "Corner" in their name. Additionally, some entries refer to vanished railroad sidings, which can lead to errors in third-party tools like Google Maps.
Evolution of Place Names and Slur Removal
The GNIS is a living document. Name changes can be initiated by the general public, through the BGN, or via Presidential executive order (e.g., Executive Order 14172 in 2025 renaming the 'Gulf of Mexico' to the 'Gulf of America').
There has been a concerted effort to remove racial and ethnic slurs from the database. In 2015, a cross-reference against the Racial Slur Database identified 1,441 slur-based names across all US states, with Arizona having the highest concentration. Notable actions include:
- 1962 & 1974: Replacement of racial pejoratives for African Americans and Japanese Americans.
- 2021: An order by the Secretary of the Interior to remove the term "Squaw" from federal usage.
- Community Action: Students and linguists in Alaska successfully renamed features like "Negrohead Creek" to the Athabascan "Lochenyatth Creek."
| Database Component | Primary Focus | Key Characteristic |
|---|---|---|
| NGNDB | US States and Territories | Includes a top 50,000 features subset |
| GCNDB (formerly TMNDB) | Topographic Map Names | Based on specific map scales (e.g., 1:24,000) |
| AGNDB | Antarctica & Off-lying Islands | Natural features only; no outposts |
| BGN Database | Official Decisions | Records decisions since 1890 |
Frequently Asked Questions
How can a person propose a name change in the GNIS?
Proposals for new or changed names for U.S. geographical features can be submitted through The National Map Corps via the GNIS website, where the public can also review justifications for such changes.
What is a GNIS identifier?
A GNIS identifier is a permanent, unique feature record identifier assigned to every physical or cultural feature in the database to ensure it can be tracked regardless of name changes.
Why are some "populated places" in the GNIS not actually towns?
Some entries are historical records of vanished railroad sidings or are "cartographic locators," such as survey corners used by surveyors to mark land polygons.
Does the GNIS include man-made structures like airports?
No, during Phase 1, certain classes of features—including airports, roads, individual buildings, and broadcasting masts—were excluded because they were better documented in non-USGS maps.
How does the GNIS handle racial slurs in place names?
The federal government has issued orders to remove offensive terms (such as "Squaw" in 2021) and has historically replaced pejoratives with more acceptable terms through the BGN and executive action.