Regulations.gov

Status: Original and derived artifacts are available for this document.

Download: pdf

-(((-]ÿ‚y)ƒÿ„[

{/f4e15z{|

01ÿ34456578ÿ9 5ÿ34ÿ ÿ15ÿ7 ÿ31 1ÿ

ÿ34ÿÿ/01ÿ20015ÿ/01!"#$%&$'#$()(*$+,-*$)))+001.
#647
ÿ89:;<=ÿ?9:@<??<AB

=CDDEFGÿHICDÿJKEÿ<FGEIFLGMCFLNÿOMNPNMHEÿ=CEQMRGEFSE
BEGTCIU
V5 ÿWXÿ6ÿYMRKÿLFPÿOMNPNMHEÿ?EIZMSEÿ1ÿ[4Xÿ+\]ÿ()(-

ÿ!"#$%&$'#$()(*$+,-*.
/01ÿ!"#$%&$'#$()(*$+,-*$)))+.ÿ/01!"#$%&$'#$()(*$+,-*$)))+.ÿ2001
^_``abc

#ÿ446 ÿdef5.
gcchij`abck +

lm^nopqrsstuotaccaqovrbht
ÿw1f4 ÿ6x5yw1f4 5z7{/f4e15z{|!"#$%&$'#$()(*$+,-*$))+*44601}+zx d.

=CDDEFGÿ<~
!"#$%&$'#$()(*$+,-*$))+*
JILSUMFÿB€DEI
0x)$1dd7$x4*

6x5ywwwz7{/f4e15z{|001!"#$%&$'#$()(*$+,-*$))+*

+(

)1((1()'ÿfb&gÿ^!

`F3="HcFX

0122345ÿ73589

29553ÿ41

71 2345ÿ 53ÿ

3393 ÿ7853ÿ
!"#ÿ%&'ÿ(&()
*8+3ÿ01 45ÿ
%

,-./0 4/56ÿ8909ÿ8.:;5.9< ,AB;CDBE IB9J; LBM.J0E O,P S.TTB;0D;AÿU/D<9;CB
1"23ÿ 123=>?@="?ÿ1"FGHÿ1="Kÿ1?KNKÿ1Q"Rÿ1GHFVF3H?"G
WKHX"G#ÿYÿZG3KH#ÿ[HGÿ1NKHX"G#VHGÿ\ÿ]Kÿ[HGÿ13KVHGÿ\
^GGH2H=H#ÿZ"ÿ1"GGH2H=H#ÿ\ÿ^W_ÿ̀R3ÿaNb11NcF"cFX1"NH1KF3="HFX1ÿ\
de_^ÿaNb11@@@cF"cFX1KQKG1QK?VQVHQK"HV"GVQH"
Z3NNKÿ13NNK

aNb11@@@cKF3="HcFX1G1dhZVijVkZV(&(lV%m)lV&&%l

(1(

Supplemental Comment on Grizzly Bear Lower-48 DPS Rulemaking
Docket: FWS–R6–ES–2024–0038
Re: Endangered and Threatened Wildlife and Plants; Grizzly Bear Listing on the List of Endangered and
Threatened Wildlife With a Revised Section 4(d) Rule (90 Fed. Reg. 4234, January 15, 2025)
Submitted via: regulations.gov
May 1, 2026
Public Comments Processing
Attn: FWS–R6–ES–2024–0038
U.S. Fish and Wildlife Service
MS: PRB/3W
5275 Leesburg Pike
Falls Church, VA 22041

Introduction and Purpose of This Supplemental Comment
The Service is under a court-ordered deadline of December 18, 2026 to submit a final rule complying with
the ESA that revises or removes the lower-48 grizzly bear listing. This comment is submitted as
supplemental material for the administrative record supporting that final rule. Although the formal 60day comment period on the January 2025 proposed rule closed on May 16, 2025, the Service retains
discretion — and under ESA §4(b)(1)(A), an affirmative obligation — to consider the best scientific and
commercial data available at the time of final rulemaking. New evidence material to the adequacy of postlisting state regulatory mechanisms under ESA §4(a)(1)(D) has become available since the close of the
comment period, and this comment places that evidence in the administrative record.
The evidence consists of a completed analysis of Idaho's management of gray wolves under post-delisting
state authority. Gray wolves were delisted from the Northern Rocky Mountain DPS in 2011 under a
regulatory framework that, in structure, closely parallels what the proposed §4(d) rule and any successor
delisting rule would establish for grizzly bears: federal listing maintained or lifted, state primary
management authority, numerical population thresholds as the regulatory trigger for federal reintervention, and reliance on state monitoring to detect approach to those thresholds. Idaho's post-2011
wolf management provides a documented natural experiment in what that regulatory architecture
produces in practice. The results should inform the Service's evaluation of regulatory-mechanism
adequacy for grizzly bears, both under the DPS designation and the proposed §4(d) rule, and under any
forthcoming proposed delisting.
We do not argue that Idaho's wolf experience is directly transferable to grizzly bears as a biological matter
— the species differ in life history, reproductive rate, and behavioral ecology in ways that are material to
population dynamics. We argue that the regulatory framework is transferable, and that the specific failure

Page 1 of 15

modes documented in Idaho's wolf monitoring are structural failures of post-delisting state monitoring
that would recur, with different numerical parameters but the same architecture, if grizzly bears were
delisted or if the §4(d) rule authorized expanded take under state monitoring regimes that share the
features we document.

Summary of the Argument
1. The Service's statutory obligation under ESA §4(a)(1)(D) is to evaluate whether existing regulatory
mechanisms are adequate to conserve the species. Based on the evidence summarized below,
that standard cannot be met for grizzly bears under any framework that delegates primary
management authority to the states on the terms established for Northern Rocky Mountain
wolves in 2011. The Service should decline to delist grizzly bears in the lower 48 states, should
maintain the threatened designation under a revised DPS, and should tighten rather than
expand the §4(d) rule's delegated-take provisions. For delisted or reclassified species, §4(a)(1)(D)
adequacy depends entirely on whether state monitoring can reliably detect approach to the
thresholds that condition federal non-intervention. Monitoring that cannot reliably detect decline
is not an adequate regulatory mechanism regardless of how the substantive take provisions are
written.
2. Idaho's post-2011 wolf monitoring framework has produced published abundance estimates that
are internally inconsistent with (a) the compensatory mortality threshold the agency itself cites
as the scientific basis for its harvest framework, and (b) multiple independent empirical indicators
derived from the agency's own mortality records. Under the agency's own numbers, implied
annual take rates exceeded the cited 29% compensatory threshold in five of six years from 2019
through 2024, reaching 38.0% in 2023. Concurrently, eleven independent empirical indicators
have shifted in directions consistent with accelerated population decline.
3. This is not a case where the monitoring is producing uncertain estimates that might be refined
with better methods. It is a case where the monitoring framework has structural features —
unpublished statewide extrapolation methods, retroactive revision of prior estimates without
methodological documentation, degrading age and genetic data collection, and the agency's own
operations contributing to that degradation — that prevent it from functioning as a reliable
regulatory trigger. A state management framework with these features cannot meet the
§4(a)(1)(D) adequacy standard.
4. The Service's January 2025 proposed rule (90 Fed. Reg. 4234) appropriately declines to delist and
appropriately maintains federal oversight. However, the proposed §4(d) rule authorizes expanded
take to be administered through state memoranda of understanding under state monitoring, with
the same structural features we document in the Idaho case. The administrative record for the
final rule must include an evaluation of whether those state monitoring frameworks can function

Page 2 of 15

as §4(a)(1)(D) regulatory mechanisms, not merely whether the substantive take provisions are
narrowly drawn.
5. Idaho's post-delisting management has also departed from the federal-tribal-state cooperative
framework that the 2009 delisting rule's §4(a)(1)(D) adequacy finding presupposed. The Nez Perce
Tribe led on-the-ground wolf recovery in Idaho from 1995 through 2016 under successive
federally-approved agreements, and retains treaty-reserved hunting rights on approximately 13
million acres of ceded territory under the 1855 Treaty of Walla Walla. Idaho's current
management operates toward a population target that materially diminishes the treaty-protected
resource without formal tribal consultation, in a departure from Secretarial Order 3206 and the
continuing trust obligations the Service retains post-delisting. Any grizzly rulemaking must
preserve and operationalize tribal consultation and co-management commitments if it is to avoid
the same trust-law failure mode.
6. The 2017 Habitat-Based Recovery Criteria supplement for the Northern Continental Divide
Ecosystem (approved October 5, 2017) is the operative habitat-monitoring framework that any
DPS designation or §4(d) rule must rely upon. It commits the Service and its federal partners to
biennial monitoring and reporting on specific habitat criteria — motorized route density, secure
core habitat, developed recreation sites, livestock allotments — measured against a 2011
baseline. We recommend specific strengthening of the monitoring and enforcement provisions
so that the HBRC can serve as an adequate §4(a)(1)(D) regulatory mechanism.
The balance of this comment develops these points.

I. The Regulatory-Mechanism Question for Post-Listing Management
ESA §4(a)(1)(D) requires the Service, in any listing determination, to evaluate “the inadequacy of existing
regulatory mechanisms.” For species that remain listed under revised DPS designations, or for species
subject to §4(d) rules that delegate management flexibility to states, the same factor continues to
constrain the sufficiency of the regulatory framework. The Ninth Circuit and the District of Montana have
repeatedly remanded grizzly bear and wolf delisting rules on §4(a)(1)(D) grounds, finding that state
regulatory commitments were not supported by mechanisms capable of detecting and responding to
population decline. Crow Indian Tribe v. United States, 965 F.3d 662 (9th Cir. 2020), is directly on point;
the court affirmed vacatur of the 2017 Greater Yellowstone grizzly delisting rule in substantial part
because the Service had not adequately evaluated whether state monitoring would detect threats to
recalibration thresholds the delisting rule established as its own regulatory floor.
The principle emerging from these cases is straightforward: a numerical population threshold in a delisting
rule or §4(d) rule is a legally meaningful regulatory commitment only if the monitoring framework that
detects an approach to the threshold is itself reliable. If the monitoring can produce published estimates
that diverge substantially from empirical reality, the threshold is not a regulatory trigger but a nominal

Page 3 of 15

provision. The Service's evaluation of regulatory-mechanism adequacy must therefore extend to the
monitoring framework itself, not merely to the substantive population thresholds.
This principle applies with particular force to the current grizzly rulemaking. The proposed §4(d) rule
authorizes expanded take in Management Area C for threats to human safety, livestock, and property
“when non-lethal deterrence or live capture is not reasonably possible.” That authorization is administered
through state, tribal, and federal MOUs, with reporting back to the Service. The rule's conservation value
depends on the reliability of the reporting and on the Service's ability to detect whether the aggregate
take under these provisions is approaching levels that threaten the population. If state reporting and
monitoring frameworks have the structural features we document for Idaho wolves, the §4(d) rule
authorizes expanded take without a functioning regulatory trigger for corrective action.

II. The Idaho Wolf Evidence
The following findings are drawn from an analysis of the complete Idaho Department of Fish and Game
Big Game Mortality Records dataset, 6,995 verified wolf mortalities from 1995–2025, obtained through
successive Idaho public records requests. The full manuscript, currently in preparation for submission to
the Journal of Wildlife Management, is available to the Service on request and should be lodged in the
administrative record in its entirety. The findings summarized below are those most directly relevant to
the §4(a)(1)(D) evaluation.

A. The internal contradiction
The 2023 Idaho Gray Wolf Management Plan cites Adams et al. (2008) as the primary literature basis for
its harvest framework and asserts that wolf populations sustain human-caused mortality up to
approximately 29% through compensatory mortality. Using the agency's published abundance estimates
as denominators and its verified mortality counts as numerators, the implied annual take rate exceeded
29% in five of six years from 2019 through 2024. Under the agency's retroactively revised 2023 abundance
estimate, the implied 2023 take rate is 38.0%. The 2023 plan itself reports, on p. 7, that “Total documented
mortality in Idaho has averaged 33% over the last 5 years” — a number that exceeds, in the agency's own
words, the threshold the plan cites four pages later as the boundary of compensatory mortality. The
contradiction is internal to the agency's own published materials.
This contradiction is independent of any critique of the abundance estimation methods. If the estimates
are correct, the operating take rates exceed the literature threshold that the plan invokes. If the estimates
are biased high, the true take rates are even higher. There is no interpretation consistent with the agency's
own published numbers under which the wolf hunting, trapping, and control mortalities framework
operates within the compensatory range the plan defines as its scientific basis.

B. Concordance of empirical indicators

Page 4 of 15

Eleven independent empirical indicators derived from the mortality records — each from a different
underlying mechanism, each measured independently — have shifted since 2021 in directions consistent
with a population substantially smaller than the agency's published estimates imply:
Indicator

Baseline

Recent value

Change

Take rate vs. 29% cited threshold

<29% (Adams 2008)

27.3%–38.0% (2019–
2024)

above
threshold

Total annual mortality

515 (2021)

265 (2025)

−48%

Dedicated firearm CPUE, statewide

0.16 mort/day (2018)

0.08 mort/day (2025)

−50%

Dedicated firearm CPUE, noninjunction regions

0.19 (2018)

0.057 (2025)

−70%

October peak-month mortality

87 avg (2018–2023)

29 avg (2024–2025)

−67%

Government control kills

94 (2020)

28 (2025)

−70%

GMUs with any recorded kill

67 (2018–2020)

8 zero-kill GMUs
(2023–2025)

spatial
contraction

Pup + yearling share of aged
mortalities

75%–85%

75%–85%

structural

Firearm-only mortality

219 (2021)

157 (2025)

−28%

Seasonal peak month

October (6/6 yrs 2018–
2023)

October 5th–6th
(2024–2025)

restructured

Lolo zone aerial returns, GMU 10

20 wolves (Mar 2024)

6 wolves (Mar 2025)

−70%

None of these indicators is individually dispositive. The concordance among all of them — each derived
from a different mechanism — provides the inferential weight that single-indicator analysis cannot. A
planned-reduction interpretation of the decline does not account for the dedicated firearm CPUE collapse
in regions outside the 2023 federal trapping injunction, the 28% firearm-only decline (which cannot be
attributed to litigation-related or other trapping restrictions of any kind), or the structural pup-andyearling composition of the mortality sample that persists across all years.

C. Structural features of the monitoring framework
Three features of the Idaho monitoring framework are relevant to the §4(a)(1)(D) evaluation because each
is a structural failure mode rather than a transient methodological concern:
1. Statewide abundance estimation methods are not peer-reviewed in their statewide application.
Idaho's camera-based space-to-event model was validated for wolves in three study areas totaling
approximately 10,438 km² (Ausband et al. 2022, Ecosphere), but the statewide extrapolation is

Page 5 of 15

documented only in an internal technical report (Thompson et al. 2022) hosted on an agency SharePoint
site that is not publicly accessible. An independent technical critique (Creel 2024) documents that the
statewide implementation violates multiple stated assumptions of the underlying method, including nonrandom camera placement at predicted rendezvous sites, motion-triggered rather than time-lapse image
capture, and reliance on a 2010 resource selection function that may no longer reflect contemporary wolf
distribution. The newer approximate Bayesian computation method introduced in 2024 has also not been
published in peer-reviewed literature and is described only in commission meeting “goldsheet”
summaries.
2. Estimates are retroactively revised without published methodological explanation.
The 2023 abundance estimate produced by the ABC method was reported as approximately 1,350 in the
July 2024 goldsheet and revised to approximately 1,150 in the July 2025 goldsheet — a 15% reduction
with no accompanying methodological documentation. Retroactive revision of estimates is a routine
feature of scientific practice when accompanied by clear methodological explanation; revision without
explanation prevents independent evaluation of whether the new estimate better reflects reality or
reflects unstated changes in the underlying model.
3. Biological data inputs are degrading, and the agency's own operations are contributing to the
degradation.
Aging coverage of the mortality record declined from 64.6%–83.6% during 2011–2019 to 22.7%–50.7%
during 2020–2025, with zero age records in harvest year 2023 across multiple public-records releases. The
2023 plan explicitly commits to collecting age and genetic data from every killed wolf (p. 32). In March
2025, the agency's own aerial gunning operations in Game Management Unit 10 killed six wolves, five of
which were recorded as “No samples taken” — no age, sex, DNA, or tooth data collected. A population
estimation method that depends on age-structured mortality data as primary input cannot function when
the input data are not collected, and the inability to collect those data is most pronounced in the agency
operations most closely controlled by the agency itself.

D. Failure to honor tribal consultation commitments that conditioned delisting
The Service's regulatory-mechanism adequacy finding in the 2009 Northern Rocky Mountain delisting rule
rested not only on Idaho's state management plan but on the cooperative federal-tribal-state framework
that had delivered recovery. The Nez Perce Tribe operated the Idaho wolf recovery program on the ground
from 1995 forward under a Tribal Recovery and Management Plan approved by the Service in September
1995 — an arrangement that existed because the Idaho Legislature had statutorily prohibited the state
from participating in reintroduction. A May 2005 Memorandum of Agreement between Idaho and the
Tribe formalized a shared monitoring framework, with the Tribe assuming primary responsibility in the
Clearwater and McCall regions. The January 2006 federal-state MOA transferring day-to-day authority to
Idaho was executed alongside and in coordination with the continuing tribal role. The post-delisting five-

Page 6 of 15

year monitoring period, which the 2009 rule made a condition of delisting, was substantially carried out
by the Tribe through 2016.
Two commitments from this framework persist independent of the formal end of the federal-tribal
partnership in 2016 and are directly relevant to the §4(a)(1)(D) evaluation for grizzly bears:
First, the 1855 Treaty with the Nez Perces (12 Stat. 957), stipulated that the Nez Perce Tribe retained
treaty-reserved hunting, fishing, and gathering rights on approximately 13 million acres of ceded territory
in Idaho, Oregon, Montana, and Washington— territory that encompasses the core of Idaho's wolf range,
including the Frank Church–River of No Return Wilderness where the 1995–1996 reintroductions
occurred. The treaty right to hunt a species requires, as a matter of federal Indian law, that the resource
exist in sufficient abundance to be meaningfully harvestable. Idaho's openly stated objective of reducing
the statewide wolf population by approximately 60% — from an estimated 1,337 wolves in 2022 toward
a target of approximately 500 — is being implemented without formal consultation with the Tribe on
whether the resulting population level satisfies the treaty-reserved right, and without any analysis in the
2023 Management Plan of the effect of the reduction on tribal subsistence and cultural harvest. The plan's
brief "Nez Perce Tribe Contributions" section (p. 8) describes the historical recovery role but does not
address contemporary consultation obligations, treaty implications, or Tribal objections to the post-2021
take regime.
Second, the Service's own trust responsibility toward the Tribe, expressed in Secretarial Order 3206
(American Indian Tribal Rights, Federal-Tribal Trust Responsibilities, and the Endangered Species Act,
1997), requires the Service to consult with affected tribes on ESA actions that implicate treaty-reserved
resources. When the 2009 delisting rule transferred management to states that were expected to operate
within the cooperative framework that had produced recovery, the Service retained continuing oversight
obligations that include ensuring state regulatory mechanisms do not unilaterally abrogate treatyprotected interests. The Service's 2011 reissuance of the delisting rule (76 Fed. Reg. 25590) did not modify
these continuing obligations. Idaho's post-2021 management — enacted through state legislation (S.B.
1211, 2021) without federal consultation on the treaty implications, and implemented under a 2023 plan
that does not substantively address tribal consultation on the population target — is a departure from the
cooperative framework that the 2009 adequacy finding presupposed.
The relevance to the grizzly DPS rulemaking is direct and structural, not merely analogical. The proposed
DPS encompasses the ceded territories and reservations of multiple tribes whose treaty and trust interests
are implicated by any delisting, recalibration, or §4(d) authorization — including the Blackfeet Nation, the
Confederated Salish and Kootenai Tribes of the Flathead Reservation, the Confederated Tribes of the
Colville Reservation, the Nez Perce Tribe (including with respect to the Bitterroot Ecosystem NEP), the
Kootenai Tribe of Idaho, the Shoshone-Bannock Tribes, and the Eastern Shoshone and Northern Arapaho
of the Wind River Reservation. Any regulatory framework that authorizes expanded take — through
delisting, recalibration, or §4(d) MOUs — without binding tribal consultation commitments and

Page 7 of 15

enforceable mechanisms for sustained tribal participation in monitoring and management will produce,
under post-listing pressure, the same erosion we document in the Idaho wolf case. The Service's Section
7 consultation and trust obligations are continuing obligations, not one-time findings made at the moment
of listing or delisting.
We recommend that the final rule expressly preserve and operationalize the Service's ongoing trust and
consultation obligations. Specifically: (a) any MOU executed under the §4(d) rule authorizing expanded
take should require, as a condition of Service approval, evidence of meaningful government-togovernment consultation with each tribe whose treaty-reserved or trust interests are affected by the
MOU's geographic scope, with consultation records lodged in the administrative record; (b) the final rule
should specify that tribal wildlife programs operating under federally-recognized management authority
retain co-management status in their treaty territories regardless of state management decisions; and (c)
the Service should commit to reviewing state management plans for consistency with tribal treatyreserved rights on a recurring basis, not only at moments of ESA status change. These provisions would
distinguish the grizzly framework from the Idaho wolf framework in precisely the respects where the wolf
framework has failed.

E. What this means for regulatory-mechanism adequacy
The Idaho case demonstrates that a state monitoring framework can produce published abundance
estimates, cite peer-reviewed literature thresholds, operate under a formal management plan with
explicit numerical commitments, and still fail to detect that the operating take rate exceeds the
framework's own scientific basis. The failure is not concealed; it is evident in the agency's published
materials, cross-referenced against the agency's public records data. It has nonetheless persisted through
multiple successive management plans without correction.
The 2009 Northern Rocky Mountain delisting rule (74 Fed. Reg. 15123) established 150 wolves or 15
breeding pairs as the relisting trigger. That trigger is a regulatory commitment only if Idaho's monitoring
framework can reliably detect an approach to it. The evidence we document indicates it cannot. Idaho is
openly managing toward a population target of 500 wolves (approximately one-third of its 2022 estimated
population), and the monitoring framework has shown that it cannot even reliably determine whether
operating take rates are above or below the compensatory threshold the plan invokes.
This is the central regulatory-mechanism concern the Service must evaluate for grizzly bears. If a state
monitoring framework with these structural features is delegated the responsibility for detecting an
approach to a grizzly bear population floor under a delisting rule or §4(d) authorization, the regulatory
commitment represented by that floor is not meaningfully enforceable.

III. Application to the Grizzly Bear DPS Rulemaking
A. The proposed §4(d) rule
Page 8 of 15

The proposed rule authorizes expanded take through state, tribal, and federal MOUs in Management Area
C, including lethal removal of bears identified as ongoing threats “when non-lethal deterrence or live
capture is not reasonably possible.” Determinations of reasonable possibility are made by the state, tribal,
or federal authority operating under the MOU, subject to reporting. This structure is analogous to Idaho's
wolf “depredation” and “selective removal” authorities, which, under Idaho's current framework, produce
mortality records that are incorporated into the abundance estimation framework as an input.
The final rule should include, at a minimum, the following provisions to address regulatory-mechanism
adequacy:
1. Reporting and data collection standards as a condition of MOU execution.
Any state or tribal authority executing an MOU under the §4(d) rule should be required, as a condition of
the MOU, to collect complete biological samples from every grizzly bear taken under the MOU authority
— age, sex, DNA, and tooth samples at minimum — and to submit those samples to the Interagency Grizzly
Bear Study Team or an equivalent federally-coordinated data repository within 30 days of take. The Idaho
wolf case demonstrates that state commitments to collect biological data in management plans do not,
standing alone, ensure collection in practice. MOU-level requirements, with consequences for nonperformance, are necessary.
2. Monitoring framework peer-review requirement.
Population estimation methods used to evaluate compliance with §4(d) provisions or with any future
recalibration or relisting thresholds should be required to be published in peer-reviewed literature before
being relied upon as the regulatory monitoring framework. The Idaho case demonstrates that unpublished
statewide extrapolations of validated small-area methods can yield estimates that diverge materially from
empirical reality, with no mechanism for correction. A peer-review requirement at the methods level
would provide the structural discipline that has been absent in the Idaho wolf case.
3. Documentation requirement for revision of prior estimates.
Any retroactive revision of previously published abundance estimates by 5% or more should be required
to be accompanied by a published methodological explanation sufficient to permit independent
evaluation of the revision. This provision is modeled on standard scientific practice in peer-reviewed
journals and would prevent the pattern we document in the Idaho case, in which a 15% downward
revision of the 2023 abundance estimate was published in a commission goldsheet without any
methodological explanation.
4. Multiple-indicator reporting alongside model-based estimates.
The Interagency Grizzly Bear Study Team currently reports several empirical indicators of population
status, including conflict frequency, mortality records, and — for the Greater Yellowstone and Northern
Continental Divide Ecosystems — known and probable mortalities against the sustainable mortality
thresholds. This practice should be extended, formalized in the final rule or accompanying Conservation
Page 9 of 15

Strategy, and required of any successor state monitoring regime. The Idaho case demonstrates that
concordance among multiple independent empirical indicators can reveal population trajectories that
model-based estimates obscure. Requiring multiple-indicator reporting is a low-cost transparency
measure that substantially improves the diagnostic power of the monitoring framework.
5. Threshold for automatic re-evaluation.
The final rule should include a provision requiring that when any combination of empirical indicators —
including, but not limited to, total annual mortality, mortality concentration in specific demographic
cohorts, and spatial contraction of occupied range — exceeds specified thresholds, the Service will initiate
a formal re-evaluation of the regulatory framework. The Idaho case demonstrates that without an
automatic re-evaluation trigger, monitoring failures persist through multiple management-plan cycles
without correction.

B. Integration with the 2017 NCDE Habitat-Based Recovery Criteria
The 2017 Habitat-Based Recovery Criteria supplement to the Grizzly Bear Recovery Plan (approved
October 5, 2017) establishes the habitat-monitoring framework for the Northern Continental Divide
Ecosystem. The HBRC commits federal partners to maintain motorized route density, secure core habitat,
developed recreation sites, and livestock allotments at 2011 baseline levels, with biennial monitoring and
reporting. The HBRC is cited as an integral part of the recovery framework in the proposed DPS rule and
would continue to operate under any final rule.
We support the HBRC as a substantive matter and note three provisions that should be strengthened in
any implementation under the DPS final rule:
1. Biennial monitoring reports should be publicly accessible.
The HBRC specifies biennial monitoring of the three habitat criteria but does not require public posting of
monitoring reports in a readily accessible form. Reports should be posted to a public Service webpage
within 60 days of completion, with data in a format that permits independent analysis. This provision is
analogous to the data-accessibility concerns we document in the Idaho wolf case, where abundanceestimation reports are hosted on agency SharePoint sites that are not publicly accessible.
2. Temporary-change limits should be aggregated across subunits.
The HBRC permits 5% temporary increases in open motorized route density, 3% temporary increases in
total motorized route density, and 2% temporary decreases in secure core habitat per BMU subunit,
provided 10-year running averages remain within limits. Without an aggregate check across all subunits
within a BMU, a pattern of maximum-allowed degradation in every subunit could produce substantial
cumulative habitat loss that individually complies with the per-subunit limits. A BMU-level aggregate
check, calculated annually, should be added as a supplemental criterion.
3. The “valid existing rights” exception should be tightened.
Page 10 of 15

The HBRC provides that the motorized-access criterion does not apply to actions where “valid existing
rights preclude or restrict agency discretion (e.g., certain contracts, permits, leases, etc.).” This language
is potentially open-ended and could accommodate a substantial volume of development under existing
lease and permit structures without tripping the criterion. The final rule should specify that “valid existing
rights” is limited to legally enforceable vested property rights and does not include administrative permits,
leases, or contracts that the agency retains the discretion to modify. Without this clarification, the
criterion is vulnerable to the same drift we document for Idaho wolf monitoring — nominal commitment
maintained, substantive content eroded through accumulated exceptions.

C. Implications for any forthcoming proposed delisting
We understand, based on public statements by Service officials and by stakeholders including the Center
for Biological Diversity, that the administration is likely to issue a new proposed delisting rule in advance
of the December 18, 2026 final-rule deadline. Whether or not that occurs, the analysis we submit is
directly relevant to any delisting determination.
The 2020 Ninth Circuit decision in Crow Indian Tribe, 965 F.3d at 679–683, held that the Service's prior
delisting rule failed §4(a)(1)(D) in part because the Service had not adequately evaluated whether state
regulatory mechanisms would maintain the population above the recalibration threshold established in
the rule itself. The Idaho wolf evidence we document here is the most thorough empirical record available
of what state monitoring under a numerical-threshold regulatory regime actually produces. Any delisting
rule that relies on state monitoring under numerical thresholds, without addressing the specific structural
monitoring failures we document, would be vulnerable to remand on the same §4(a)(1)(D) grounds that
supported vacatur in Crow Indian Tribe.
We urge the Service, in preparing any proposed delisting or the December 2026 final rule, to include in
the administrative record a full evaluation of state wolf-monitoring performance in Idaho, Montana, and
Wyoming as the most directly analogous available evidence on state carnivore-monitoring performance
under delisted status. The evidence we summarize here is specific to Idaho; analogous evidence should
be examined for the other NRM states and for any state that would assume management responsibility
under the grizzly rule.

D. The §4(a)(1)(D) case against delisting on this record
The evidence in Section II establishes, on the most directly analogous available record, that the state
regulatory mechanism contemplated for post-delisting grizzly bear management is structurally incapable
of performing the detection and response functions that §4(a)(1)(D) requires. The evidence is not
speculative; it is derived from fifteen years of post-delisting state management of the species the Service
previously treated as the successful recovery precedent for grizzly bears.
The Ninth Circuit's decision in Crow Indian Tribe requires the Service to evaluate whether state regulatory
mechanisms will maintain the population above the recalibration threshold established in the rule itself.
Page 11 of 15

On the Idaho wolf record: operating take rates have exceeded the state's own cited compensatory
threshold in five of six years; eleven independent empirical indicators have diverged from published
abundance estimates in directions consistent with accelerated decline; biological data collection has
degraded by more than half; estimates have been retroactively revised without methodological
documentation; and tribal consultation commitments that conditioned the original delisting adequacy
finding have been abandoned. No element of this record would support a finding, under any reasonable
construction of §4(a)(1)(D), that Idaho's current regulatory mechanisms are adequate to conserve wolves
post-delisting. The Service has not proposed to relist wolves — a question beyond the scope of this
comment — but the analysis demonstrates that the regulatory framework cannot be relied upon as a
precedent for grizzly bear delisting.
The logical consequence for grizzly bears is direct. The Service cannot, on this administrative record, make
a defensible §4(a)(1)(D) finding that delisting is warranted, because the closest available empirical record
on how state regulatory mechanisms perform under this framework shows that they do not perform the
functions the finding would require them to perform. The Service's January 2025 determination that the
state petitions for DPS-specific delisting in the Greater Yellowstone and Northern Continental Divide
Ecosystems are “not warranted” should be maintained and extended: no DPS-specific or lower-48-wide
delisting of grizzly bears can satisfy §4(a)(1)(D) on this record.
We recognize that the Service is operating under a court-ordered deadline to issue a final rule by
December 18, 2026, and that political pressure to delist is substantial. Neither the deadline nor the
pressure alters the §4(a)(1)(D) analysis. The settlement agreement in Save the Yellowstone Grizzly requires
a final rule that “revises or removes” the listing; revision to confirm the threatened designation under a
defined DPS, with a tightened §4(d) rule, fully satisfies the settlement obligation while complying with the
ESA. Delisting on this record would not survive judicial review, and the Service should say so explicitly in
the final rule rather than produce a rule vulnerable to the same vacatur that has followed every prior
grizzly delisting attempt since 2007.
The recommendations in Sections III.A, III.B, and III.C above — strengthening MOU-level data
requirements, requiring peer review of monitoring methods, operationalizing tribal consultation
obligations, and tightening the HBRC's enforcement provisions — are not alternatives to maintaining the
listing. They are the minimum structural improvements that would make any future delisting analysis
defensible if conditions change. None of them remedies the §4(a)(1)(D) inadequacy that exists on the
current record.

IV. Specific Requests
In light of the foregoing, we request that the Service:

Page 12 of 15

1. Include this comment and the underlying Idaho wolf analysis in the administrative record for
FWS–R6–ES–2024–0038 and for any successor rulemaking, including any proposed delisting rule
preceding the December 18, 2026 final rule deadline.
2. In the final rule, incorporate the MOU-level biological sample collection requirement, the peerreview requirement for population estimation methods, the documentation requirement for
retroactive estimate revisions, the multiple-indicator reporting requirement, and the automatic
re-evaluation trigger described in Section III.A above.
3. Operationalize the Service's ongoing tribal trust and consultation obligations in the final rule by
requiring, as a condition of §4(d) MOU approval, documented government-to-government
consultation with each affected tribe; by preserving tribal co-management status on treaty-ceded
territories regardless of state management decisions; and by committing to recurring review of
state management plans for consistency with tribal treaty-reserved rights. These provisions
should reflect Secretarial Order 3206 and distinguish the grizzly framework from the Idaho wolf
framework in the respects where the wolf framework has departed from its cooperative federaltribal-state foundation.
4. In implementation of the 2017 NCDE Habitat-Based Recovery Criteria under the final rule,
strengthen the public-accessibility, BMU-aggregate-check, and “valid existing rights” provisions
described in Section III.B above.
5. Decline to delist grizzly bears in the lower 48 states in any form in the December 18, 2026 final
rule. The administrative record does not support, and cannot be made to support on available
evidence, a §4(a)(1)(D) finding that state regulatory mechanisms are adequate to conserve the
species post-delisting. The Idaho wolf evidence summarized here, together with analogous
evidence available for Montana and Wyoming wolf management, establishes that the regulatory
framework contemplated for post-delisting grizzly management has failed in its only comparable
application.
6. Maintain and extend the Service's January 2025 determination that the state petitions for DPSspecific delisting in the Greater Yellowstone and Northern Continental Divide Ecosystems are “not
warranted.” The evidence we document strengthens rather than weakens the basis for that
determination and applies with equal force to any lower-48-wide delisting.
7. Tighten, do not expand, the §4(d) rule's delegated-take provisions in the final rule. Authorization
of expanded take administered through state MOUs, under state monitoring regimes with the
structural features documented for Idaho wolves, is not consistent with the ESA's conservation
purpose or with the §4(a)(1)(D) adequacy standard. The proposed §4(d) rule's expanded take
authorities in Management Area C should be narrowed, not broadened, in the final rule.

Page 13 of 15

Closing
Grizzly bear recovery in the lower 48 states represents one of the most significant achievements of the
Endangered Species Act. That achievement was produced by five decades of federal listing, federal
funding, federal-tribal-state cooperative management, and the specific protections the Act provides. It
was not produced by the state regulatory architecture now being proposed as its successor, and the one
available natural experiment in that architecture — Idaho's management of wolves since 2011 — has
demonstrated that the architecture cannot maintain post-listing recovery.
The Service's obligation under the ESA is to make listing determinations on the best available scientific
and commercial data. The best available data on state management of a delisted large carnivore under
the regulatory framework contemplated for grizzly bears shows that the framework fails. Delisting grizzly
bears on this record is not scientifically defensible, will not survive judicial review, and would set back the
recovery that federal protection has made possible.
We urge the Service to decline to delist, to maintain the threatened designation under a defined DPS, to
tighten the §4(d) rule, to operationalize the tribal trust obligations the Service continues to hold, and to
strengthen the habitat-based recovery criteria as specified in this comment. These actions comply with
the December 2026 court-ordered deadline, satisfy the settlement agreement in Save the Yellowstone
Grizzly, and produce a final rule that can survive the judicial review that every prior grizzly delisting
attempt has failed.
We offer our continued assistance in providing the underlying Idaho wolf data, methodological
documentation, analysis, and tribal consultation context to support the administrative record for any rule
the Service issues.
Respectfully submitted,

Suzanne Asha Stone
Executive Director, The International Wildlife Coexistence Network
9169 W. State Street, Suite #1014, Garden City, Idaho 83714

Attachments:

Page 14 of 15

8. Stone et al. (in preparation). Internal contradiction in Idaho's wolf management: take rates exceed
the agency's cited compensatory threshold and empirical indicators support accelerated
population decline, 2018–2025. [Manuscript submitted for consideration; available on request.]
9. Supporting figures and tables from the Idaho BGMR analysis (available on request).

Page 15 of 15