The two Irene Lakes below Sprague Glacier

Why we went looking

Our dataset originally listed a lake at 11,860 ft near Stones Peak as “Sprague Glacier Lake” — a descriptive name we’d coined ourselves (glacier name + “Lake”), never an official one. A user flagged that this invites exactly the confusion we built the site to avoid: it reads as a variant of the well-known, easily-accessible Sprague Lake off Bear Lake Road, six and a half miles away, when it is in fact a remote, off-trail pond with nothing but a shared surname in common.

We also had a second, unrelated lake in our data called “Irene Lake” — a separate off-trail lake near this same basin, itself already flagged on our Lake Irene page as commonly confused with the roadside picnic-area lake of the same words, reversed. Renaming one “Sprague” lake risked colliding with the other “Irene” lake sitting in the same small bowl. Before changing anything, we wanted to know what each lake is actually called, by whoever gets to decide that — and it turned out three different authorities give three different answers.

The bowl

All four named water bodies below sit within about a quarter mile of each other, fed by Sprague Glacier at 40.343°N, 105.732°W. USGS ImageryOnly satellite tiles are our own basemap imagery; the outlines are official government waterbody polygons from the USGS National Hydrography Dataset (NHD), queried live and plotted at their true surveyed position — not estimated by eye.

USGS ImageryOnly satellite tiles, zoom 16 (≈6 ft/px). Amber outlines are the two separate ponds that share a single GNIS record, 177804 “Irene Lake.” The red × marks that record’s official coordinate. Ice/snow is still visible floating on the tarn.

What each authority says

We checked three independent sources: the U.S. Board on Geographic Names (via the GNIS domestic names database, which also records whether a name has an actual Board decision behind it, or was just administratively catalogued), the National Hydrography Dataset (which stores the actual surveyed water polygons with GNIS linkage, not just a representative point), and OpenStreetMap (community-maintained, and the source of our own basemap).

GNIS — U.S. Board on Geographic Names

Retrieved from the official USGS bulk download, DomesticNames_CO_Text.zip, at prd-tnm.s3.amazonaws.com. The bgn_type/bgn_authority/bgn_date fields distinguish a genuine Board Decision from a name that was simply collected off a map and entered into the database.

“Sprague Tarn” does not appear in GNIS at all — it is not an official name in any sense. Notably, neither is the well-known roadside “Sprague Lake,” despite being one of the park’s most-visited stops; it was only administratively catalogued in 1998, never formally decided.

National Hydrography Dataset

Queried live from the USGS NHD map service at hydro.nationalmap.gov (layer 12, Waterbody — Large Scale), which stores actual surveyed polygons, each tagged with the GNIS ID it was matched to during compilation.

This is the single most useful record: USGS deliberately assigns the same GNIS ID, 177804 “Irene Lake,” to two physically separate ponds — the tarn at the glacier’s foot and a second pond about 600 ft northeast. It isn’t a coordinate error; it’s one federal name deliberately covering two water bodies.

OpenStreetMap — and its edit history

We pulled full version history for each way from the OSM API to see not just the current tags, but who named what, and when. This turned out to tell the most complete story.

Full tags on the current version (v5) of way 810052324:

natural = water
water = lake
name = Sprague Tarn
intermittent = no
tidal = no
salt = no

Our determination

The OSM version history tells a real, well-intentioned story. The tarn below Sprague Glacier was drawn without a name in 2020. On 27 July 2022, a mapper (VLD194) named it “Irene Lake” — independently arriving at the same association GNIS’s single representative point suggests. That created a real collision: two ponds 600 ft apart both called Irene Lake. On 5 August 2024, a more experienced local mapper (NW757, 1,400+ changesets at the time) deliberately renamed it to “Sprague Tarn” — a reasonable, well-constructed fix, following the same “[Glacier] + Tarn” pattern the Board itself approved once already in this park (Andrews Tarn, GNIS 177849, Board Decision 1961, below Andrews Glacier, GNIS 177848, Board Decision 1932).

We initially adopted “Sprague Tarn” on that basis. On reflection, we changed course, for one reason that outweighs the rest: “Sprague Tarn” has no footprint in any federal dataset whatsoever — not GNIS, not NHD. It is a single volunteer mapper’s coinage, however well-reasoned, made in one edit in 2024. “Irene Lake,” by contrast, is GNIS feature 177804 — catalogued in 1978, and the name USGS’s own surveyed NHD polygons actually carry for both ponds in this bowl. It is not Board-decided, but it is the controlling federal name on record, in a way “Sprague Tarn” simply is not.

GNIS’s error, as we see it, isn’t the name — it’s applying one name to two visibly separate ponds. That is a data-granularity problem we already have a house style for: this site splits several other multi-pond features the same way, using the shared base name plus a plain parenthetical — Haynach Lakes (main) / (north pond) / (south pond #1) / (south pond #2), Fay Lakes (Lower) / (Middle) / (Upper), Twin Lakes (lower) / (upper). Applying that same convention here was more consistent than either inventing a new name for one pond while the other kept the ambiguous shared name, or leaving both ponds indistinguishable on the site. So: Irene Lake (tarn) for the pond Sprague Glacier feeds directly, and Irene Lake (NE pond) for the other. Both entries are now marked "informal": false in our data — not because either has a Board Decision (neither does), but because “Irene Lake” is GNIS’s actual catalogued name, which is the same standard we already apply to “Sprague Lake” and “Lake Irene” elsewhere on the site: administratively collected but uncontested counts as our site’s working definition of non-informal, distinct from names like the original “Sprague Glacier Lake” or “Sprague Tarn” that no government source records at all.

Could this become official?

The U.S. Board on Geographic Names accepts two relevant kinds of request, both through the Domestic Names Committee: a new-name proposal for an unnamed feature, or a correction to the location/attribution of an existing one. Simple data errors in GNIS can also be reported directly to tnm_help@usgs.gov without a full proposal.

Our revised position actually sidesteps the obstacle we flagged in the original version of this page. BGN policy will not make official “a new name in a federally designated wilderness area… without an overriding need to do so,” and this bowl sits inside RMNP’s ~250,000-acre designated wilderness (2009) — a real problem for a brand-new name like “Sprague Tarn.” Asking GNIS to split feature 177804’s single point into two distinct records, by contrast, isn’t a new-name proposal at all: it’s a location/attribution correction to an already-catalogued name, which is the more straightforward of the two request types and doesn’t face the same wilderness restriction. We haven’t filed that correction ourselves — this page documents our own editorial decision, not a government submission — but it’s the realistic path if anyone wanted to pursue it.

Methodology

Every figure on this page came from a live source query, not a lookup table or prior knowledge:

• Basemap imagery: our own self-hosted USGS ImageryOnly PMTiles, the same tiles served on our interactive maps.
• Waterbody polygons: USGS NHD MapServer, layer 12, queried by bounding box, projected to image pixel space by the same Web Mercator transform MapLibre uses.
• GNIS records: bulk download of DomesticNames_CO_Text.zip from the USGS National Map staged products bucket, current as of the date on this page.
• OSM data: live queries against the official OSM API, including full element version history and changeset metadata, not just current tags.

For background on what GNIS, NHD, and OSM actually are, their official schemas, how NHD cross-references GNIS, and where to pull each dataset yourself, see GNIS, NHD, and OSM, explained.

Data pulled 2026-07-10, decision revised 2026-07-10 · part of an ongoing effort to keep lake names on this site consistent with their best available sources. See also: Irene Lake (tarn), Lake Irene, Irene Lake (NE pond), GNIS, NHD & OSM reference.