Homestake: wildlife, flora and the connected wetland mosaic
Research checkpoint September 23, 2026 · v0.3. This is a new research layer. The v0.1/v0.2 snapshots and unrelated project work remain unchanged. Prepared against the recovered current source register, 1041 evidence packet and September 23 hydrology findings. No website changes, publication, agency requests or external messages were made.
What the evidence now supports
The strongest conservation case concerns a connected but hydrologically varied landscape: groundwater-supported peatlands, seasonally wet meadows, willow cover, beaver ponds, stream habitat and adjacent forest. Direct inundation and construction effects must be assessed alongside seasonal changes in stream stage, groundwater, pond persistence and prey production. Colorado diversion studies show that riparian vegetation and channel form can change downstream; Colorado fen-restoration work shows that recovering water levels does not necessarily recover peat properties. Neither result supplies a Whitney-specific affected acreage or loss percentage. Caskey et al., 2015, abstract/Methods/Table IV · Schimelpfenig et al., 2014, abstract/Results
The useful addition to the 1041 record is a traceable chain: documented resource > measured hydrologic dependency > alternative-specific exposure > biological/function consequence > finding and enforceable requirement. Existing evidence supports serious, specific investigation of these pathways. Missing measurements do not demonstrate harmlessness, and suitable habitat alone does not prove presence or a legal violation. The published county criteria expressly encompass groundwater levels, wetland function, habitat and aquatic/terrestrial food webs. Eagle County Chapter VI, §6.04.01(15)-(20), Appendix A pp. iv-vi
Material additions to v0.2
Finding | Evidence label | What it changes |
|---|---|---|
BioBlitz refresh remains 1,400 observations, 229 terminal taxon concepts and 190 species-rank concepts; no audited record changes from v0.2 | Confirmed Homestake record, with individual identification grades | Preserve the previous counts; do not announce new BioBlitz species or abundance trends. |
233 screened GBIF records: 129 eBird-derived observations, 103 preserved specimens and one other occurrence | Confirmed Homestake record, provider-level identification and broad locality | Adds a separate historical and bird-record layer. These are records, not 233 additional species or independent checklists. |
Historical upstream boreal-toad breeding and Colorado River cutthroat population reported by CNHP | Confirmed Homestake record, historical drainage scale | Replace “no Homestake documentation found” with a dated, geographically bounded account. Do not move it into the Whitney footprint. |
Quantified Colorado moose distance-to-water evidence | Regional ecological evidence | Replace the unverified universal half-mile statement with a population-, season- and method-specific finding. |
Current CNHP ranks for mosses, plants and community leads | Regional ecological evidence; local observations separately graded | Adds conservation context without calling a rank a legal listing or constituent plants a rare-community occurrence. |
Older and “Fens 2025” GIS services return different feature sets | Unknown / conflicting | Reconcile map versions before changing existing public counts or overlap calculations. |
Sources: BioBlitz API query, seven pages plus species counts; GBIF discovery query, offsets 0/300; CNHP 2000 Eagle County survey, printed p. 159; moose study, Table 1; CNHP tracking list; older fen service; Fens 2025 service.
Evidence labels and geographic boundaries
Label | Meaning in this package |
|---|---|
Confirmed Homestake record | A reviewable local observation, specimen or agency record. Identification grade, date and spatial limits remain part of the claim; “confirmed record” does not upgrade a Needs ID identification. |
Project-area survey finding | A documented professional survey within a defined survey area, method and period. The new review did not obtain the underlying 2020 ERO memo/field GIS; inherited v0.2 survey figures remain a retrieval priority. |
Potential habitat / survey lead | Habitat suitability, historical taxonomic lead, or a regional candidate without adequate local occurrence evidence. |
Regional ecological evidence | A measured study, species account, classification or conservation designation whose geographic scope is stated. |
Homestake hypothesis | A testable local causal pathway, not a measured project effect. |
Unknown / conflicting | Missing, inaccessible or inconsistent evidence that changes interpretation. |
The BioBlitz collection area, historical “Homestake” localities, upper tributaries, 2019 drilling work areas, watershed inventory units, modeled reservoir alternatives, downstream map frame and gauge catchments are different spatial units. A record or acre total cannot be moved between them without a documented spatial join and adequate positional accuracy. Precise rare-species, breeding-site and camera coordinates are excluded from the deliverables.
Local records: refreshed and expanded
The live BioBlitz query was refreshed at 2026-09-23 14:40:29 UTC. The snapshot retains 618 Research Grade, 778 Needs ID and four Casual observations. All 1,400 observation IDs, observation dates, identification grades, taxon/report-taxon assignments and positional-accuracy values matched the v0.2 baseline in the audited comparison. The 229 reported terminal concepts include higher-rank identifications; 190 are species rank. 491 records lack a terminal report-taxon assignment in this snapshot and remain in the occurrence table rather than being forced into a species count. S01: live project
The GBIF search used “Homestake,” United States and Colorado, then screened county/locality fields. Of 379 discoveries, 242 matched Eagle County plus the locality name; nine “Homestake Peak School” records were excluded as place-name false positives. The resulting 233 records are a discovery set, not a complete drainage inventory. The 129 Cornell records contain 53 provider taxon-name concepts, including a genus-level identification; one additional historical bird specimen occurs outside that Cornell subset. Dataset lineage verifies submitted eBird-derived occurrences; the original checklist pages, hotspot inventory, checklist completeness, distance and duration were not retrieved. Merlin detections are not counted as eBird submissions. V3-35: query · V3-36: Cornell EOD dataset · S15: Cornell Sound ID guidance
Selected additions make the value and limits concrete:
Record | Date/source | Geographic and identification limit |
|---|---|---|
American Dipper, Cinclus mexicanus | Cornell/GBIF 5455274378, July 28, 2024 | Submitted bird occurrence; original checklist effort and local breeding not inspected. |
Wilson’s Warbler, Cardellina pusilla | Cornell/GBIF 2698947553, September 4, 2019 | Riparian-bird lead; a September observation does not establish nesting. |
Western jumping mouse, Zapus princeps | TCWC/GBIF 675755200, June 27, 1948; additional June specimens | Historical specimen label, broad Homestake locality; no current population or listed Preble’s-mouse inference. |
Water shrew, historical Sorex palustris navigator | TCWC/GBIF 675757484, June 29, 1948 | Modern S. navigator concept requires collection review. An ecological account supports an aquatic-insect link, not a new voucher determination. |
Hoary sedge, Carex canescens | COLO/GBIF 2573540395, August 1, 1985 | Published coordinate uncertainty is 10,000 m; unusable for assigning a specific wetland or reservoir footprint. |
The occurrence tables retain provider IDs, record dates, taxonomic strings, basis, licenses, uncertainty and GBIF quality flags. Multiple repositories or dates do not guarantee independent animals or populations. Specimens were not reexamined. Taxonomic conflicts: including a goshawk accepted-name mismatch and genus-level hawk record: remain visible. Some plant discoveries are introduced or of unresolved local origin; this package does not label the complete inventory “native.”
Historical amphibian and fish evidence
CNHP’s 2000 Eagle County survey describes a conservation area in the upper Homestake tributaries, reports boreal-toad breeding and a Colorado River cutthroat population, and lists an Altai cottongrass occurrence under its historical name. The publication is a meaningful local record, but its old conservation ranks and population descriptions are not current assessments. V3-17, printed p.159/PDF p.162
CPW’s 2006-2007 boreal-toad status report records historical breeding-site monitoring, including a pond drying in 2002, uncertain recruitment, and later visits with no evidence of breeding at an upstream site. A separate record named “Homestake Reservoir” contains a county/site-label inconsistency that requires agency reconciliation. The deliverables generalize location and do not reproduce breeding coordinates. Negative historical visits neither establish current absence nor a project effect. V3-24, printed pp.36 and 76
CPW’s 2023 and 2024 distribution schedules each list 1,000 approved CR1 fish for Homestake Creek water code 20622; the code legend denotes a Colorado River cutthroat hatchery strain. These are approved schedules, not evidence all fish were stocked, survived or reproduced. Historical population records and stocking plans therefore support a targeted request for fish surveys, completed stockings and genetics: not a blanket claim that every present Homestake trout is a native, genetically intact cutthroat. 2023 schedule, PDF p.34 and legend · 2024 schedule, PDF p.35 and legend · CPW taxonomy account
Species and conservation interpretation
The 339-row species matrix is a taxon-and-survey-target matrix, not a count of 339 locally established species. It includes every BioBlitz terminal concept, screened GBIF concepts and priority targets. Each row separates occurrence, ecology, status and the proposed local test. Unknown life-history or site-specific information is explicitly left unresolved. The accompanying flora table provides the plant indicator and community context; the observation tables preserve more detail than the summary rows.
Group | Strongest supported relationship | Consequence for the project argument |
|---|---|---|
Moose, elk and mule deer | Moose have measured summer water association; elk/deer use broader seasonal forage and cover | Protect habitat mosaics and movement; verify local seasonal use. Do not call all ungulates fen obligates. |
Beaver | Direct aquatic habitat use plus engineering of ponds, groundwater and vegetation | Assess active complexes and their inflows/functions, not only dam footprints. |
Hare and lynx | Hare cover/prey and lynx forest/willow-corridor use | Connect vegetation and landscape continuity to prey habitat; verify lynx presence and final critical-habitat overlap separately. |
Bear, lion, fox and coyote | Broad habitat/generalist or predator relationships | Use documented occurrence and corridor/prey evidence where obtained, not generic wetland-dependence claims. |
Bobcat | CPW specifically describes avoidance of wetlands and heavy subalpine timber | Keep uncertain cat media unresolved; do not use a bobcat as evidence of fen dependence. |
Amphibians | Breeding-water duration, recruitment, connectivity and disease | Survey egg/larval/metamorph stages and pond networks, not adult sightings alone. Wood frog remains a lower-priority range/records question. |
Fish and aquatic insects | Flow, habitat connectivity, thermal conditions and prey delivery | Pair fish genetics and habitat work with benthos, drift and emergence; standing benthic abundance is not the same as prey reaching fish. |
Birds | Local records plus riparian-shrub and pond guild relationships | Confirm breeding and survey habitat strata; adjacent-forest birds are not automatically wetland specialists. |
Pollinators and small mammals | Added historical records and regional process leads | Obtain local host/flower, microhabitat and current-occurrence evidence before making species-specific loss claims. |
Pinpoints and URLs for these statements are in each species row. Key agency accounts include beaver, snowshoe hare, lynx, bobcat, boreal chorus frog, wood frog, river otter and CNHP riparian shrubland, wildlife list. These are regional accounts; local occurrence is supported independently by the record tables.
Status is not one category
Taxon/unit | Federal or state legal/administrative status | 2025 SWAP / current CNHP rank | Local evidence |
|---|---|---|---|
Canada lynx | Federal Threatened; Colorado Endangered | Tier 1; G5/S1 | 2019 suitable-habitat screen; no independently verified local animal record in this package |
Boreal toad, Southern Rocky Mountain population | Colorado Endangered; no federal listing inferred from historical candidate references | Tier 1; G4T1T2Q/S1 | Historical upper-drainage breeding; present occupancy unknown |
Colorado River cutthroat | Colorado Special Concern; distinct from federally listed greenback | Tier 1; current CNHP name Oncorhynchus virginalis pleuriticus, GNRT3/S3 | Historical upstream record; current genetics unresolved |
Mountain sucker | Colorado Special Concern | SGIN; GNR/S2? | 2019 potential-habitat screen only |
Northern leopard frog | Colorado Special Concern | Tier 1; G5/S3 | Potential-habitat screen only |
Wood frog | Colorado Special Concern | SGIN; G5/S3 | Range/records lead, not confirmed Homestake occurrence |
River otter | Colorado Threatened | Do not assign a tier from absence in the retrieved table | Local occurrence unresolved |
Thicket hairstreak | No ESA/state designation inferred from the retrieved dash fields | Tier 2; G4G5/S3; watchlisted | BioBlitz record; species-specific wetland dependence unverified |
Authority/date: CNHP current public tracking data, exact taxon rows, September 23, 2026; CPW 2025 SWAP, group/tier lists; CPW boreal toad; CPW lynx; CPW river otter. G/T ranks are global/taxon ranks; S ranks concern Colorado; ranges and question marks express uncertainty. SGIN means Species of Greatest Information Need. Special Concern and SWAP tiers must not be presented as ESA threatened/endangered listings.
The final lynx critical-habitat revision took effect August 17, 2026 and includes a Colorado Unit 6. Use final authoritative GIS and the legally controlling rule when testing actual alternatives. The 2019 report’s statement about then-current critical habitat is not current. Earlier investigation-point screening also is not a reservoir-boundary intersection. USFWS final rule, 91 FR 43732, Unit 6 and effective date
The moose-distance audit
Do not use “moose remain within half a mile of wetlands” as a universal fact. The source search did not establish an original primary study supporting that rule. Non-recovery is not proof that no such sentence exists; the unresolved source remains an explicit claim-audit item.
The stronger finding is a 2025 Colorado paper using 21 GPS-collared adults and 1,672 summer locations from 2017-2018: average distance to mapped water was approximately 110 m, with 90% of fixes within approximately 285 m. This is distance to mapped flowing/standing water, not a delineated wetland edge or a boundary beyond which moose cannot live. Its habitat-use data informed survey design. The exact extraction is ST01. Abouelezz & Hobbs, Methods and Table 1, DOI 10.1002/wlb3.01368
For a Homestake test, first resolve obscuring and positional accuracy privately. Compare observed distances with the distribution of available wetland, open-water, willow and forest habitats; stratify by season, temperature and snow. Use equal camera effort in wetland edges and comparable uplands. Opportunistic photographs along roads cannot establish habitat selection, a home-range limit or a population distribution. Report camera nights and uncertainty; do not publicly map cameras. This is a proposed analysis, not one completed here.
Moose presence also should not be presented as proof of an unchanged historical community: CPW describes transient historical records and establishment of breeding populations through translocations beginning in 1978. CPW, History of Moose in Colorado
Flora and communities: wider coverage, stricter matching
The new crosswalk contains 197 taxon/concept rows, including all 163 BioBlitz plant concepts, additional historical specimens and the historical cottongrass lead. Across the combined table, 125 exact names match the Corps regional list; 72 remain unmatched. The 163 BioBlitz concepts alone yield 106 exact matches, including 44 OBL/FACW concepts. These scopes differ from the v0.2 selected 43-taxon table and earlier synonym-expanded counts; they are not a biological gain or loss.
The fixed reference is final 2022 NWPL version 3.6, Western Mountains, Valleys and Coast, with 3,407 parsed entries and per-name PDF page locators. Ratings describe regional wetland occurrence tendency: OBL/FACW are wetland-leaning; FAC spans wetlands and non-wetlands; FACU/UPL are upland-leaning. They do not delineate a wetland, establish peat depth or distinguish a fen from a marsh. Draft update notices do not replace the final edition used here. USACE, 2022 WMVC list, cover/background and individual rows
No rating was assigned merely because a similar name exists. Examples needing a taxonomic review include Caltha chionophila versus the list’s C. leptosepala, Trollius albiflorus versus T. laxus, and Vaccinium cespitosum versus V. caespitosum. Bryophytes and genus/higher-rank records remain unrated when no exact entry exists. Unmatched is not automatically UPL.
Current CNHP exact matches add useful priorities: Aulacomnium palustre is G5/S1S3, fully tracked; Sphagnum angustifolium is G5/S3S4, watchlisted; historical GBIF Eriophorum chamissonis is G5/S3, watchlisted. The 33-row community-lead table includes Betula glandulosa / Sphagnum Shrub Fen (G2/S2) and willow/sedge associations with differing ranks. None is declared a confirmed Homestake community from plant co-occurrence. CNHP current tracking list, exact named rows
Use trained botanists and a bryologist to establish community composition, dominance and structure, identify difficult sedges/willows/mosses, and pair vegetation with peat and hydrology. A high-priority plant or association name is a reason to investigate carefully: not permission to relabel every potential-fen polygon.
What the studies measured
The 14-row study table supplies design, sample, effect, limitation, geography, source inspection depth and transferability. Four complementary evidence lines matter most:
1. Diversion and vegetation: the Colorado paired-reach study connects altered flow with narrower/simplified channels and a shift toward upland vegetation. It motivates stage-elevation and plant transects, not a fixed universal buffer. ST02 2. Groundwater and peat: beaver-dam monitoring demonstrates effects beyond pond edges, while restored Colorado fens retained altered peat properties. These studies argue for source-water and functional measurements. ST03 · ST04 3. Breeding and food webs: pond quality influences amphibian dispersal/recruitment, while insect emergence and drifting prey respond to habitat and flow. More benthic insects need not mean more food transported to fish, and more emergence need not mean greater taxonomic richness. ST06 · ST08 · ST11 4. Multiple drivers and incomplete replacement: the RMNP willow study implicates browsing and beaver/floodplain feedbacks; European rewetting comparisons show persistent differences from near-natural fens. Local impact analysis must address interacting drivers and function over time. ST05 · ST13
The Scottish pollinator comparison is useful for designing a flower/insect survey, but belongs in a clearly labeled outside-region research note. It is not a Homestake species-loss forecast. Abstract-only studies are labeled as such; missing coefficients/sample details were not guessed.
Downstream: follow the water and test each connection
The current project mapping places French Creek above Gold Park, with the Missouri/Sopris system and other tributaries contributing along their appropriate reaches. Continue from the Whitney vicinity through lower Homestake/Blodgett toward the Red Cliff gauge reach, then the Eagle River. Red Cliff town, Gold Park and a full-mouth outlet are not interchangeable locations. Preserve the newer map-name and gauge corrections from the September 23 project work.
Wetland/habitat | Plausible water source | Main test of an additional diversion/release change |
|---|---|---|
Hillslope fen | Upslope groundwater/springs, possibly mixed supply | Measure hydraulic gradients and recharge pathways; a fen may discharge into the creek without depending strongly on creek stage. |
Valley-bottom fen or peat wetland | Alluvial groundwater, springs and/or stream exchange | Pair nested piezometers with stream/pond stage, peat cores and water chemistry. |
Wet meadow | Snowmelt, seasonal groundwater and flood wetting | Determine how long root-zone moisture and floodplain inundation persist after runoff recession. |
Willow carr | Shallow groundwater and variable floodplain supply | Measure shrub recruitment/height, water table and browsing together. |
Marsh/beaver complex | Pond inflow, dam stage and groundwater exchange | Track pond depth/area, dams and larval recruitment through summer; distinguish deep reservoir water from shallow vegetated habitat. |
Channel/floodplain/side pools | Discharge, tributary gains, groundwater, shade | Measure low-flow connectivity, temperature, oxygen, prey delivery and frequency/duration of overbank wetting. |
These are Homestake hypotheses until site-specific measurements resolve them. The seven-row pathway table links each to inspected sources, confounders and the local measurement needed. No individual fen has been assigned a source-water class in this review. CNHP fen · wet meadow · marsh · riparian shrubland
GIS findings and a required reconciliation
An attributes-only query used the existing downstream map’s public WGS84 display envelope against both CNHP services. The older service returned 18 features and the service named Fens 2025 returned six; every returned feature in both sets was classified Potential Fen. The saved GIS join CSV preserves service-specific object IDs, source classifications and whole-polygon acreage. It does not equate IDs across services, sum overlapping inventories, compute exposure or infer habitat disappearance. The count discrepancy may reflect changed mapping, scope or service content; its cause was not resolved. Older service · Fens 2025 metadata · existing map methods
Keep the existing map’s dated source attribution until CNHP clarifies which layer should govern and how source polygons correspond. A subsequent geometry comparison found five newer features overlapping older polygons, one without overlap and no exactly equal pairs. The attributes and reproducible queries are supplied; no new polygon layer is redistributed, because a clear redistribution license and service correspondence were not established. Existing modeled alternative overlaps also remain their own versioned analysis and must not be silently recalculated from the new feature count.
The 2025 WRNF inventory’s confirmed/likely totals at HUC12 watershed scale are a separate product. Neither those totals nor this downstream cartographic frame gives the number of fens a reservoir would destroy. CNHP 2025, Appendix A and methods
Connect ecology to the latest flow accounting
Use the latest recovered gap assessment, not an earlier generalized statement that average exports cannot be estimated. It supports approximately 23,371 acre-feet/year of whole-tunnel exports for WY2007-2024. Across two recorded routes, the export share is 52.09% for Gold Park, WY2007-2024, and separately 43.04% for Red Cliff, WY2007-2011. With endpoint storage and recorded evaporation accounts, the Gold Park partial-balance comparison is 51.75%. The study distinguishes export intensity from conditional net gauge-flow reduction; they are not identical formulas. Neither is a current full-mouth percentage. Preserved latest findings
These figures establish relevant existing-flow context. Ecological prediction needs daily/seasonal hydrographs, the location and timing of physical releases, drought sequences, stage-to-groundwater relationships and alternative-specific changes. Do not transform annual export percentages into percentages of wetland, wildlife or biodiversity loss. Reservoir releases could produce different local temperature and stage responses depending on season, outlet depth, source water and operating rules; measure/model the actual case.
Seasonal survey and records plan
The detailed CSV specifies ten workstreams, priorities, agencies, repeat visits and metadata. The minimum useful sequence is:
Timing | Priority field/data work | Required output |
|---|---|---|
Now / before permit findings | Obtain actual alternatives/operations; original delineations and fen IDs; CPW fish genetics/stocking and toad monitoring; final lynx GIS; reconcile map services | Versioned source register and spatially correct baseline; precise sensitive records remain restricted |
Snowmelt and early breeding season | Stage/flow checks, groundwater gradients, seep/floodplain mapping; repeated amphibian visits | Hydroperiod and source-water evidence tied to resource IDs |
Early-mid summer | Three bird-count rounds; vegetation identification and plots; benthos/drift/emergence; beaver condition | Reproducible species/habitat observations with zero-detection and effort records |
Late summer / low flow | Fish/temperature/refugia assessment; amphibian metamorph checks; peat/water-table and plant repeat | Recruitment, pool persistence, thermal exposure and drawdown evidence |
Autumn/winter and repeat years | Camera continuity, snow/effort metadata, winter-range records; annual QA and reference-site comparison | Seasonal use and a defensible before/after baseline |
Aim for at least two pre-disturbance hydrologic years where feasible, reference reaches and consistent methods. A short initial camera deployment or one summer survey cannot establish absence or characterize drought sensitivity. Sample sizes beyond the proposed initial schedule should be adjusted to variance and detection after pilot data; the protocol is a feasible starting design, not a power analysis. Handling, trapping, tissue sampling and disease work belong with appropriately permitted specialists.
Remaining gaps, stated precisely
• No final alternative-specific ecological impact prediction: official current footprints and operating hydrographs were not recovered.
• No polygon-specific fen water-source test: public mapping and regional studies do not resolve hydraulic connection.
• No direct eBird checklist/hotspot refresh: submitted occurrence lineage was recovered through GBIF, but sampling-event data and current hotspot metadata remain a request.
• No current fish lineage or amphibian occupancy conclusion: historical documentation is stronger; modern agency records are still needed.
• No original universal half-mile moose source: use the verified summer study and a local test.
• No automatic synonym or rare-community assignment: unresolved plant and historical mammal names remain flagged.
• Some sources remain inaccessible: the source register records the failed source/URL and retrieval route. The ERO memo, several USGS full papers and the western jumping-mouse review were not used for new uninspected quantitative claims.
• No enacted-draft assumption: the official county page lists a September 23 recommendation hearing and October 27 BOCC adoption hearing for the 1041 amendment. Recheck actual adopted text and transition rules before filing. Official schedule
Using the deliverables
Start with the companion Homestake-Project-Integration-Plan-v0.3-2026-09-23.md for the 1041 crosswalk, site destinations, proposed copy and implementation order. The source register has 94 stable-ID rows, including all 51 preserved baseline sources; this is not a claim that 94 sources supplied new inspected findings. The claim ledger has 33 synthesized claims with explicit source/locator edges; species, flora and studies carry their own additional source links.
The database retains every original table and adds v03_ tables. It and the CSVs are the canonical research layer; public prose should be derived from the supported claim wording. The archive includes the research inputs needed to rebuild the layer, provenance, validation and handoff. See the validation file for exact checks and the remaining limits of those checks.