- Dates
- 1900 – 2026
- Eras
- Modern era
- Concepts
- environmentconservationforestswetlandsgeologysoils
- Within
- Beaver Island archipelago
Summary
This snapshot adds a public-observation layer to the archival, survey, and management records already in the knowledge base. It queried 36 named taxa already connected to Beaver Archipelago research, restricted the GBIF search to human observations with coordinates, and assigned records locally to Beaver, Garden, High, Hog, Whiskey, North Fox, or South Fox Island.
The first snapshot retains 5,097 records: 4,908 on Beaver, 65 on Garden, 112 on High, six on Hog, and six on South Fox. No retained records were returned for Whiskey or North Fox in the selected taxa. Those counts describe public GBIF coverage, not species richness, population size, survey effort, or absence.
What it adds
The strongest additions are observations that can be compared against dated historical and survey evidence:
- American woodcock: 106 retained public observations, providing a modern observation layer for the High Island woodcock research trail.
- Piping plover: 18 retained records, to be compared with the 1987 MNFI survey summary, the 2017 management plan, current protection rules, and the 2023 High Island nesting account.
- Bald eagle and Common loon: 967 and 872 retained records respectively, providing broad modern presence evidence that can be separated from the older island-distribution table.
- Ruffed grouse, terns, Merlin, bluebirds, sparrows, meadowlarks, and Wild Turkey: public records broaden the bird layer beyond rare-species and management-plan mentions.
- Mammals and invertebrates: records include white-tailed deer, snowshoe hare, smooth green snake, yellow-banded bumble bee, emerald ash borer, and the selected mollusk and bat taxa where GBIF returned public observations.
Spatial precision and interpretation
Every retained record preserves the original GBIF coordinate and
coordinateUncertaintyInMeters when supplied. The index also records a
precision class:
- high: uncertainty of 100 meters or less;
- medium: uncertainty from 101 to 1,000 meters;
- coarse_or_obscured: uncertainty above 1,000 meters;
- unknown: no uncertainty value supplied by GBIF.
The reported point is the basis for the island assignment. It is not a claim that the observation is an exact site record: generalized or obscured points can fall inside an island polygon even when the true location is uncertain. Sensitive species should therefore remain generalized in maps and prose, and the raw coordinate should not be treated as a public nest, roost, or rare-plant location without checking the original publisher’s terms.
How it connects to the environmental layers
The observation layer is deliberately separate from habitat inference:
- the GBIF record supplies a taxon, date, publisher, and reported location;
- the OSM polygon supplies the normalized island assignment;
- NLCD and tree-canopy rasters supply broad land-cover context;
- SSURGO, the 1974 soil survey, geology layers, and MNFI occurrence records supply soil, substrate, community, and field-survey context.
This allows a later analysis to ask whether observations cluster near open beach, wetland, forest edge, or particular soil/community settings. It does not allow the reverse claim that a land-cover or soil class proves a species presence.
Limits and next work
This is a named-taxon snapshot, not a complete download of every GBIF record or every species on every island. It is also not effort-corrected: a heavily visited island can have many more public observations than a remote island. The GBIF query used a bounding box and local polygon filtering because the API rejected some detailed island-polygon queries. The source API, query counts, retained counts, polygon hash, and record-level metadata are preserved in the JSON and provenance sidecar.
The next refinement is to link these records to the original Piping Plover and High Island woodcock reports and to separate exact public observations from coarse or observer-generalized records before generating maps.
Sources
See the sources: block above and the generated JSON index.
Sources
- The snapshot preserves GBIF keys, occurrence IDs, dates, coordinates, coordinate uncertainty, taxonomic fields, dataset, observer, license, and rights-holder metadata for retained records.research · wildlife · indices · gbif_archipelago_observations
- The query used a broad archipelago bounding box because some detailed polygon queries were rejected by GBIF; records were retained only when the reported point was covered by one of the seven checked-in island polygons.osm · islands
- Generated collector and provenance sidecar.The script is rerunnable and records query counts, filtered counts, polygon hash, and precision limitations. No public copy of this file is known; the archive holds one and can provide it on request.scripts · retrieve_gbif_archipelago_observations
- Project-generated GBIF wildlife observations and island ecology crosswalk (2026).Aggregates retained records by island and selected taxon, compares them with historical wildlife evidence, and attaches island-level soil, wetland, geology, historical-cover, and forest context without copying raw coordinates. No public copy of this file is known; the archive holds one and can provide it on request.research · wildlife · indices · gbif_archipelago_ecology_crosswalk
Added 3 August 2026. Recent changes
Cite this article
Beaver Island Archive, “GBIF public wildlife observations within Beaver Archipelago island boundaries”, https://beaverislandarchive.org/documents/gbif-archipelago-public-wildlife-observations/, accessed [date], citing GBIF occurrence search API, named-taxon HUMAN_OBSERVATION queries retrieved 2026-08-03. (and 3 further sources listed on the page).
Last revised 2026-08-03. Fill in [date] with the date you read it — this page is static and cannot know that.
BibTeX for “GBIF public wildlife observations within Beaver Archipelago island boundaries”RIS for “GBIF public wildlife observations within Beaver Archipelago island boundaries”