2025-00450. Marine Mammals; Proposed Incidental Harassment Authorization for the Southern Beaufort Sea Stock of Polar Bears During Well Remediation Activities, North Slope of Alaska; Draft Environmental Assessment

Table 1—Temporary Threshold Shift (TTS) and Permanent Threshold Shift (PTS) Thresholds Established by Southall et al. (2019) Through Modeling and Extrapolation for “Other Marine Carnivores,” Which Includes Polar Bears

	TTS			PTS
	Non-impulsive	Impulsive		Non-impulsive	Impulsive
	SEL CUM	SEL CUM	Peak SPL	SEL CUM	SEL CUM	Peak SPL
Air	157	146	170	177	161	176
( print page 2723)
Water	199	188	226	219	203	232
Note: Values are weighted for other marine carnivores' hearing thresholds and given in cumulative sound exposure level (SEL CUM dB re 20µPa in air and SEL CUM dB re 1 µPa in water) for impulsive and nonimpulsive sounds, and unweighted peak sound pressure level in air (dB re 20µPa) and water (dB 1µPa) (impulsive sounds only).

During a Federal Aviation Administration test, test aircraft produced sound at all frequencies measured (50 Hz to 10 kHz) (Healy 1974). At frequencies centered at 5 kHz, jets flying at 300 m (984 ft) produced 1/3 octave band noise levels of 84 to 124 dB, propeller-driven aircraft produced 75 to 90 dB, and helicopters produced 60 to 70 dB (Richardson et al. 1995). Thus, the frequency and level of airborne sounds typically produced by aircraft are unlikely to cause TTS or PTS unless polar bears are very close to the sound source.

Although neither TTS nor PTS is anticipated during the specified activities, impacts from aircraft overflights have the potential to elicit biologically significant behavioral responses from polar bears. Exposure to aircraft overflights is expected to result in short-term behavior changes, such as ceasing to rest, walking, or running, and, therefore, has the potential to be energetically costly. Polar bears observed during intentional aircraft overflights conducted to study impacts of aircraft on polar bear responses, with an average flight altitude of 143 m (469 ft), exhibited biologically meaningful behavioral responses during 66.6 percent of aircraft overflights. These behavioral responses were significantly correlated with the aircraft's altitude, the bear's location ( e.g., coastline, barrier island), and the bear's activity (Quigley 2022; Quigley et al. 2024). Polar bears associated with dens exhibited various responses when exposed to low-flying aircraft, ranging from increased head movement and observation of the disturbance to the initiation of rapid movement and/or den abandonment (Larson et al. 2020). Aircraft activities can impact polar bears across all seasons; however, aircraft have a greater potential to disturb both individuals and groups of polar bears on land during the summer and fall. These onshore polar bears are primarily fasting or seeking alternative terrestrial foods (Cherry et al. 2009; Griffen et al. 2022), and polar bear responses to aircraft overflights may result in metabolic costs to limited energy reserves. To reduce potential disturbance of polar bears during aircraft activities, mitigation measures, such as minimum flight altitudes over polar bears and their frequently used areas and flight restrictions around known polar bear aggregations, will be conducted when safe to perform these operations during aircraft activities.

Effects to Denning Polar Bears

Known polar bear dens around the oil fields and other areas of the North Slope are monitored by the FWS. These dens may be discovered opportunistically or during planned surveys for tracking marked polar bears and detecting polar bear dens. However, these sites are only a small percentage of the total active polar bear dens for the SBS stock in any given year. Each year, many entities conducting operations on the North Slope coordinate with the FWS to conduct surveys to determine the location of any polar bear dens that may be located in close proximity to any of the operator's planned activities for that denning season. Under past IHAs and ITRs (Incidental Take Regulations), operators have been required to avoid known polar bear dens by 1.6 km (1 mi). However, an unknown polar bear den may be encountered during the BLM's activities. In instances when a previously unknown den was discovered near human activity, the FWS has implemented mitigation measures such as a 1.6-km (1-mi) activity exclusion zone around the den and 24-hour monitoring of the den site.

The responses of denning polar bears to disturbance and the consequences of these responses can vary throughout the denning process. We divide the denning period into four stages when considering impacts of disturbance: den establishment, early denning, late denning, and post-emergence; definitions and descriptions are provided by Woodruff et al. (2022) and are also located in the 2021-2026 Beaufort Sea ITR (86 FR 42982, August 5, 2021). The stage at which harassment occurs defines the level of disturbance response (Level B harassment, Level A harassment, or Lethal) attributed to either the sow or cub(s), along with the probability of the specific response occurring (see Denning Analysis).

Impacts of the Specified Activities on Polar Bear Prey Species

Information on the potential impacts of the specified activities on polar bear prey species can be found in the supplemental information to this document (available as described in ADDRESSES ).

Estimated Take

Definitions of Incidental Take Under the Marine Mammal Protection Act

Below we provide definitions of three types of take of polar bears. The FWS does not anticipate and is not authorizing either Level A harassment or lethal take as a part of this proposed IHA; however, the definitions of these take types are provided for context and background.

Lethal Take

Human activity may result in biologically significant impacts to polar bears. In the most serious interactions ( e.g., vehicle collision, running over an unknown den causing its collapse), human actions can result in the mortality of polar bears. We also note that, while not considered incidental, in situations where there is an imminent threat to human life, polar bears may be killed. Additionally, though not considered incidental, polar bears have been accidentally killed during efforts to deter polar bears from a work area for safety and from direct chemical exposure (81 FR 52276, August 5, 2016). Unintentional disturbance of a female polar bear by human activity during the denning season may cause the female either to abandon her den prematurely with cubs or abandon her cubs in the den before the cubs can survive on their ( print page 2724) own. Either scenario may result in the incidental lethal take of the cubs.

Level A Harassment

Human activity may result in the injury of polar bears. Level A harassment, for nonmilitary readiness activities, is defined as any act of pursuit, torment, or annoyance that has the potential to injure a marine mammal or marine mammal stock in the wild.

Numerous actions can cause take by Level A harassment of polar bear cubs during the denning period, such as creating a disturbance that separates mothers from dependent cubs (Amstrup 2003), inducing early den emergence during the late denning period (Amstrup and Gardner 1994; Rode et al. 2018), instigating early departure from the den site during the post-emergence period (Andersen et al. 2024), or repeatedly interrupting the nursing or resting of cubs to the extent that it impacts the cubs' body condition.

Level B Harassment

Level B harassment for nonmilitary readiness activities means any act of pursuit, torment, or annoyance that has the potential to disturb a marine mammal or marine mammal stock in the wild by causing disruption of behavioral patterns, including, but not limited to, migration, breathing, nursing, feeding, or sheltering. Changes in behavior that disrupt biologically significant behaviors or activities for the affected animal are indicative of take by Level B harassment under the MMPA. Such reactions include, but are not limited to, the following:

• Fleeing (running or swimming away from a human or a human activity);
• Displaying a stress-related behavior such as jaw or lip-popping, front leg stomping, vocalizations, circling, intense staring, or salivating;
• Abandoning or avoiding preferred movement corridors such as ice floes, leads, polynyas, a segment of coastline, or barrier islands;
• Using a longer or more difficult route of travel instead of the intended path;
• Interrupting breeding, sheltering, or feeding;
• Moving away at a fast pace (adult) and cubs struggling to keep up;
• Temporary, short-term cessation of nursing or resting (cubs);
• Ceasing to rest repeatedly or for a prolonged period (adults);
• Loss of hunting opportunity due to disturbance of prey; or
• Any interruption in normal denning behavior that does not cause injury, den abandonment, or early departure of the female with cubs from the den site.

This list is not meant to encompass all possible behaviors; other behavioral responses may be indicative of take by Level B harassment. Relatively minor changes in behavior such as the animal raising its head or temporarily changing its direction of travel are not likely to disrupt biologically important behavioral patterns, and the FWS does not view such minor changes in behavior as indicative of a take by Level B harassment. It is also important to note that eliciting behavioral responses that equate to take by Level B harassment repeatedly may result in Level A harassment.

Surface Interactions

We analyzed take by Level B harassment for polar bears that may potentially be encountered and impacted during the BLM's oil well plugging and reclamation, soil sampling, snow trail, pad, and airstrip construction, and summer cleanup activities within the specified geographic region.

Impact Area

To assess the area of potential impact from the project activities, we calculate the area affected by project activities where harassment is possible. We refer to this area as an impact area. Behavioral response rates of polar bears to disturbances are highly variable, and data to support the relationship between distance to polar bears and disturbance are limited. Dyck and Baydack (2004) found sex-based differences in the frequencies of vigilance bouts, which involves an animal raising its head to visually scan its surroundings, by polar bears in the presence of vehicles on the tundra. However, in their summary of polar bear behavioral response to ice-breaking vessels in the Chukchi Sea, Smultea et al. (2016) found no difference between reactions of males, females with cubs, or females without cubs. During the FWS's coastal aerial surveys, 99 percent of polar bears that responded in a way that indicated possible Level B harassment (polar bears that were running when detected or began to run or swim in response to the aircraft) did so within 1.6 km (1 mi), as measured from the ninetieth percentile horizontal detection distance from the flight line. Similarly, Andersen and Aars (2008) found that female polar bears with cubs (the most conservative group observed) began to walk or run away from approaching snowmobiles at a mean distance of 1,534 m (0.95 mi). Thus, while future research into the reaction of polar bears to anthropogenic disturbance may indicate a different zone of potential impact is appropriate, the current literature suggests that the 1.6-km (1.0-mi) impact area will encompass most surface polar bear harassment events.

Estimated Harassment

We estimated Level B harassment using the spatio-temporally specific encounter rates and temporally specific harassment rates derived in the 2021-2026 Beaufort Sea ITR (86 FR 42982, August 5, 2021) in conjunction with the specified project activity information. Some portion of SBS bears may occur within the Chukchi Sea at a given time. However, the ITR rates do not explicitly account for this possibility, and the project area for this proposed IHA occurs only within the geographical boundary of the SBS subpopulation. Therefore, our analyses account only for SBS bears located within the SBS subpopulation boundary. Distribution patterns of polar bears along the coast of the SBS were estimated in Wilson et al. (2017) by dividing the North Slope Coastline into 10 equally sized grids and applying a Bayesian hierarchical model based on 14 years of aerial surveys in late summer and early fall. Wilson et al. (2017) estimated 140 polar bears per week along the coastline (a measurement that included barrier islands); however, not with uniform distributions. The study found that disproportionately high densities of bears occur in grids 6 and 9, which contain known large congregating areas such as Kaktovik and Cross Island; thus, the study has required polar bear density correction of factors in previously issued incidental take authorizations (ITAs). The vast majority of the coastline within the project area in this proposed IHA falls within grids 1-4 (although a small portion of the project area is located outside of Wilson et al.'s (2017) study area near the City of Wainwright). The Wilson et al. (2017) values for grids 1-4 are similar to those in the North Slope area where the 2021-2026 Beaufort Sea ITR (86 FR 42982, August 5, 2021) encounter rates were developed; therefore, we believe those values are applicable to the project area in this proposed IHA and do not require any correction factor for polar bear densities in our analyses.

Table 2—Definitions of Variables Used in Harassment Estimates of Polar Bears on the Coast of the North Slope of Alaska

Variable	Definition
Bes	Bears encountered in an impact area for the entire season.
ac	Coastal exposure area.
( print page 2725)
ai	Inland exposure area.
ro	Occupancy rate.
eco	Coastal open-water season bear-encounter rate in bears/season.
eci	Coastal ice season bear-encounter rate in bears/season.
eio	Inland open-water season bear-encounter rate in bears/season.
eii	Inland ice season bear-encounter rate in bears/season.
ti	Ice season harassment rate.
to	Open-water season harassment rate.
Bt	Number of estimated Level B harassment events.

Table 2 provides the definition for each variable used in the formulas to calculate the number of potential harassment events. The variables defined in table 2 were used in a series of formulas to ultimately estimate the total harassment from surface-level interactions. Encounter rates were originally calculated as polar bears encountered per square km per season. As a part of their Request, the BLM provided the FWS with digital geospatial files that included the maximum expected human occupancy ( i.e., rate of occupancy [ r_o] for each individual structure ( e.g., snow trails, snow pads) of their specified activities for each season of the IHA period. Using the buffer tool in ArcGIS, we created a spatial file of a 3.2-km (2-mi) buffer around all snow trails (3.2 km on either side of the proposed snow trail center line, i.e., 6.4 km [4 mi] total diameter) to account for up to 1.6-km (1-mi) deviations from the proposed center line of the routes, and around both well sites to account for the presently undetermined camp locations (within 1.6 km [1 mi] of well head). Additionally, we placed a 1.6-km (1-mi) buffer around all lakes that may be potentially utilized during operations. We binned the structures according to their seasonal occupancy rates by rounding them up into tenths (10 percent, 20 percent, etc.). We determined the impact area of each bin by first calculating the area within the buffers of 100-percent occupancy locations. We then removed the area of the 100-percent occupancy buffers from the project impact area and calculated the area within the 90-percent occupancy buffers. This iterative process continued until we calculated the area within all buffers. The areas of impact were then clipped by coastal and inland zone geospatial files to determine the coastal areas of impact (a_c) and inland areas of impact (a_i) for each occupancy bin. This process was repeated for both seasons (ice season and open-water [ice-free] season).

Impact areas were multiplied by the appropriate encounter rate to obtain the number of polar bears expected to be encountered in the impact area per season (B_es). Equation 1 provides an example of the calculation of polar bears encountered in the ice season for an impact area in the coastal zone.

Equation 1

B_es = a_c * e_ci

To generate the number of estimated Level B harassments for each area of interest, we multiplied the number of polar bears in the area of interest per season by the proportion of the season the area is occupied, the rate of occupancy, and the harassment rate (equation 2).

Equation 2

B_t = B_es * S_p * r_o * t_i

Aircraft Impacts on Polar Bears

Polar bears in the project area will likely be exposed to the visual and auditory stimulation associated with the applicant's fixed-wing and helicopter activities; however, these impacts are likely to be minimal and short-term. Aircraft activities may cause disruptions in the normal behavioral patterns of polar bears as either an auditory or visual stimulus, thereby resulting in incidental Level B harassment. To reduce the likelihood that polar bears are disturbed by aircraft, mitigation measures, such as minimum flight altitudes over polar bears and restrictions on sudden changes to aircraft movements and direction, will be required if this authorization is finalized. Once mitigated, such disturbances are expected to have no more than short-term, temporary, and minor impacts on individual polar bears.

Estimating Harassment Rates of Aircraft Activities

Harassment rates during aircraft activities were estimated using results from studies of fixed-wing aircraft and helicopter overflights (Quigley 2022; Quigley et al. 2024). In these studies, aerial searches along the northern coast of Alaska between Point Barrow and the western Canadian border were flown and polar bears were approached at different altitudes. Polar bears that did not exhibit behavioral changes consistent with harassment were then re-approached at progressively lower altitudes, reaching as low as 38 m (100 ft). Researchers recorded behavioral changes during these approaches and evaluated if and when Level B harassment occurred. Covariates examined were polar bear location (“barrier island” or “mainland”), initial behavior (“active” or “inactive”), group size, whether the polar bear belonged to a family group, and the number of previous overflights ( i.e., how many times the group was re-approached to elicit a behavioral change). A Bayesian imputation approach accounted for polar bears that exhibited a behavioral change consistent with harassment on their first approach, thus lacking an identified altitude at which no harassment occurred due to a lack of a “non-harassment” observation. Their final model included location, activity level, and the number of previous overflights as predictors of the altitude at which a polar bear was harassed. For our aircraft impacts analysis, we used harassment rates estimated for active polar bears observed on barrier islands, as they had the highest rates of harassment. We further assumed that no previous overflights were conducted.

We provide harassment rates for the following five categories of flights: take-offs, landings, low-altitude flights (defined as those between 122 m [400 ft] and 305 m [1,000 ft] altitude), mid-altitude flights (defined as those between 305 m [1,000 ft] and 457 m [1,500 ft] altitude), and high-altitude flights (defined as those between 457 m [1,500 ft] and 610 m [2,000 ft] altitude). Harassment rates were assigned to each of these flight categories using the harassment rate for the lowest altitude in the category ( e.g., for low-altitude flights, the harassment rate estimated for 122 m [400 ft] was used). This binning method of using the lowest altitude harassment rate in the bin allowed our estimates to be inclusive of possible changes in altitude due to variable flight conditions (table 3). ( print page 2726)

Table 3—Harassment Rates for the Five Categories of Flights for Fixed-Wing Aircraft and Helicopter Overflights

Flight category	Fixed-wing	Helicopter
Take-offs	0.99	>0.99
Landings	0.99	>0.99
Low-Altitude Flights (122-305 m)	0.86	>0.99
Mid-Altitude Flights (305-457 m)	0.03	0.82
High-Altitude Flights (457-610 m)	<0.01	0.05
Note : The rate in this table are based on Quigley et al. (2024).
We used the harassment rate associated with 30 m (100 ft) for take-offs and landings.

Estimating Area of Impact for Aircraft Activities

For each category of the flight path ( i.e., take-off, low-altitude travel, mid-altitude travel, high-altitude travel, and landing), we calculated an impact area and duration of impact using flight hours or flight path information provided in the Request. We used flights logs available through FlightAware ( https://www.flightaware.com/​), a website that maintains flight logs in the public domain, to estimate impact areas and flight hours for take-offs and landings. We estimated a take-off distance of 2.41 km (1.5 mi) that would be impacted for 10 minutes. We estimated a landing distance of 4.83 km (3 mi) per 305 m (1,000 ft) of altitude that would be impacted for 10 minutes per landing. To estimate the impact area of traveling segments, we subtracted the take-off and landing areas from the total area of the flight path. The duration of impact for traveling flights was either provided in the Request or calculated using the length of the flight and a conservative flight speed of 129 km per hour (80 mi per hour), which was approximately 1.5 minutes per 3.22 km (2 mi) of the flight path.

All take-offs, landings, and traveling segments were then spatially referenced to determine whether they were within the coastal or inland zones. The coastal zone is defined as the offshore and onshore areas within 2 km (1.2 mi) of the coastline, and the inland zone is defined as the onshore area greater than 2 km (1.2 mi) from the coastline. If no location or flight hour information was provided, flight paths were approximated based on the information provided in the Request. Of the flight paths that were described clearly or were addressed through assumptions, we marked the approximate flight path take-off and landing locations using ArcGIS Pro, and the flight paths were drawn. Once spatially referenced, all flight paths were buffered by 1.6 km (1 mi), which is consistent with aircraft surveys conducted by the FWS and USGS between August and October during most years from 2000 to 2014 (Schliebe et al. 2008; Atwood et al. 2015; Wilson et al. 2017). In these surveys, 99 percent of groups of polar bears that exhibited behavioral responses consistent with Level B harassment were observed within 1.6 km (1 mi) of the aircraft.

Table 4—Seasonal Polar Bear Encounter Rates by Zone

Coastal Zone Seasonal Encounter Rate
Ice Season (July 19-November 11)	0.05 bears/km2 .
Open-water Season (November 12-July 18)	1.48 bears/km2 .
Inland Zone Seasonal Encounter Rate
Ice Season (July 19-November 11)	0.004 bears/km2 .
Open-water Season (November 12-July 18)	0.005 bears/km2 .
Note : This table is adapted from the 2021-2026 Beaufort Sea ITR (86 FR 42982, August 5, 2021).

To calculate the total number of Level B harassment events estimated due to the specified activities, we calculated the number of flight hours for each flight category ( i.e., take-offs, low-altitude travel, mid-altitude travel, high-altitude travel, and landings) for each zone and season combination. These values were then used to calculate the proportion of the season that aircraft occupied their impact areas ( i.e., take-off area, landing area, or traveling segment impact areas). This proportion-of-season metric is equivalent to the occupancy rate ( r_o) generated for surface-level interaction harassment estimates. The total impact area for each of the flight categories was multiplied by the zone and season-specific polar bear encounter rate to determine the number of polar bears expected in that area for the season ( i.e., B_es, as seen in equation 1). This number was then multiplied by the proportion of the season to determine the number of polar bears expected in that area when flights are occurring, and the appropriate harassment rate based on flight altitude to estimate the number of polar bears that may be harassed as a result of the flights (as seen in equation 2). Table 5 shows a summary of aircraft operations during the specified activities and the values used to estimate Level B harassment of polar bears during aircraft operations. ( print page 2727)

Table 5—Summary of Aircraft Operations by Season and Activity During the Proposed IHA Period

Activity	Ice season (fixed-wing aircraft only)		Open-water season (helicopter only)
	Winter support—Cape Halkett	Winter support—Fish Creek	Site inspection— Deadhorse to Cape Halkett	Site inspection—Deadhorse to Fish Creek	Site Inspection—Cape Halkett to Fish Creek	Snow trail inspection and cleanup
Altitude *	High	High	High	High	High	Low
Total Flights	25	25	6	5	2	12
Proportion of Season	0.0026	0.0021	0.0020	0.0012	0.00017	0.01887
Proportion of Flight in Coastal Zone	0.60	0	.60	0	0.51	0.26
Proportion of Flight in Inland Zone	0.40	1	.40	1	0.49	0.74
Total Encounter Rate (bears/km2 /season) **	0.0316	0.004	0.89	0.005	0.7573	0.3885
Harassment Rate	0.001	0.001	0.05	0.05	0.05	0.99
Flight Time Harassment	6.570 × 10−07	6.744 × 10−08	0.000643	2.440 × 10−06	5.295 × 10−05	0.05909
Total Takeoffs and Landings	50	50	12	10	4	24
Landing Time/Season	0.001389	0.001389	0.000725	0.000604	0.000242	0.001449
Landing Time Harassment	0.0016283	0.0016283	0.025146	0.020955	0.008382	0.0502921
Takeoff Time/Season	0.001389	0.001389	0.000725	0.000604	0.000241	0.001449
Takeoff Time Harassment	0.001094	0.001094	0.016893	0.014078	0.00563	0.03379
Number Level B Harassment of Activity	0.002723	0.002723	0.042683	0.035035	0.014066	0.143164
Total number of level B harassment events across all aircraft activities				0.240
* High-altitude flight is defined as between 457 m [1,500 ft] and 610 m [2,000 ft] altitude. Low altitude is defined as between 122 m [400 ft] and 305 m [1,000 ft] altitude. There are no mid-altitude flights considered for this project.
** Accounts for unequal encounter rates over coastal and inland zones.

Estimated Harassment From Aircraft Activities

Using the approaches described above, we estimated the total number of polar bears expected to be harassed by the aircraft activities during the proposed IHA period as a total of one bear (table 5).

Denning Analysis

Below we provide a complete description and results of the polar bear den simulation model used to assess impacts to denning polar bears from disturbance associated with all phases of the specified activities. In our denning analysis, we used the analytical method described in the 2023-2024 BLM IHA (88 FR 88943, December 26, 2023).

Additionally, on March 19, 2024, regulations promulgated in the 2021-2026 Beaufort Sea ITR (86 FR 42982, August 5, 2021) were challenged in Federal Court and the Ninth Circuit Court of Appeal issued a remand to FWS to conduct certain additional analysis. As a result of the Court's remand and ongoing scientific advancements, the FWS reexamined the denning analysis and incorporated newly available data since 2021 into the denning analysis model, allowing the continued inclusion of best available scientific information. Updates incorporated into the model adjust the impact area that can result in den disturbance, the probabilities of disturbance, and how FWS reports probabilities of different levels of take, i.e., Level B harassment, Level A harassment, and lethal take. Alterations to the denning model are described in greater detail below.

Den Simulation

We simulated dens across the entire North Slope of Alaska, ranging from the areas identified as denning habitat (Durner et al. 2006, 2013; Durner and Atwood 2018) contained within the National Petroleum Reserve-Alaska (NPR-A) in the west to the Canadian border in the east. To simulate dens on the landscape, we relied on the estimated number of dens in three different regions of northern Alaska provided by Atwood et al. (2020). These included the NPR-A, the area between the Colville and Canning Rivers (CC), and Arctic National Wildlife Refuge (NWR). Den simulations for this proposed IHA were conducted following the exact methodology described previously in the 2023-2024 BLM IHA (88 FR 88943, December 26, 2023).

Impact Area of Specified Activities

The model developed by Wilson and Durner (2020) provides a template for estimating the level of potential impact on denning polar bears during the specified activities while also considering the natural denning ecology of polar bears in the region. Previous iterations of the denning analysis model, including those utilized in the 2021-2026 Beaufort Sea ITR (86 FR 42982, August 5, 2021) and 2023-2024 BLM IHA (88 FR 88943, December 26, 2023), assumed that during all denning periods, any polar bears within dens within 1.6 km (1 mi) from project activities could exhibit a disturbance response if exposed to industrial stimuli. However, for this IHA, we refined that broad assumption to account for denning data that have been collected subsequent to the promulgation of the 2021-2026 Beaufort Sea ITR. Since 2021, four known dens (monitored in 2022 and 2023) have occurred near human activity. Of the four newly observed dens, three were extremely close to human activity (<50 m), yet the sows remained in their dens until the late denning period. We updated polar bear disturbance probabilities and litter size distributions with the information from these dens, then re-examined the historic dens that were used to create disturbance probabilities. We found that the distances between human activity and polar bear dens during the early denning period were considerably closer than those observed during other denning periods. Specifically, of the 15 dens within the case studies that were exposed to human activity during the early denning period, only one was potentially disturbed at a distance greater than 800 meters. This single den record also had imprecise information on the distance to human activity, so activity was assumed to occur within 1,610 m of the den and was likely closer. The historic dens analyzed during the den establishment, late denning, and post-emergence periods did not follow this pattern. For those dens, disturbance distances commonly exceeded 805 m. Evidence derived from dens exposed to human activity during the early denning period, including both new den records and historic dens, illustrates the reluctance of sows to abandon their maternal den/cubs in response to exposure to stimuli from nearby activity, and supports the concept that sows may be more risk tolerant during the early denning ( print page 2728) period. Additionally, sows may be less affected by sound from outside activities during the early denning period because dens are typically closed during that time, which can affect propagation of noise into the den (Owen et al. 2020). Given this evidence, we modified the denning analysis model to adjust the impact area for the early denning period to range from 0 to 805 m. As a result, dens that were simulated to be within 805 m of human activity could be disturbed during all denning periods, while dens between 806 and 1610 m way from human activity could only be disturbed during the den establishment, late denning, and post-emergence periods.

AIR Surveys

We assumed that all remediation and transit areas that will be utilized during denning season would have two AIR surveys flown prior to beginning any operations (figure 1). The first survey would occur between December 1 and December 25, 2024, and the second survey between December 15, 2024, and January 10, 2025, with a minimum of 24 hours between surveys. During each iteration of the model, each AIR survey was randomly assigned a probability of detecting dens using detection probabilities previously described in the 2023-2024 BLM IHA (88 FR 88943, December 26, 2023).

Model Implementation

For each iteration of the model, we first determined which dens were exposed to the specified activities. Dens that were simulated to be within 805 m (2,641 ft) of human activity could be disturbed during all denning periods, while dens within 806-1610 m (2,644-5,282 ft) of human activity could only be disturbed during the den establishment, late denning, and post-emergence periods. Dens detected during AIR survey were excluded if activity did not occur prior to AIR survey. We identified the stage in the denning period when the exposure occurred based on the date range of the activities the den was exposed to: den establishment ( i.e., initial entrance into den until cubs are born), early denning ( i.e., birth of cubs until they are 60 days old), late denning ( i.e., date cubs are 60 days old until den emergence) and post-emergence ( i.e., the date of den emergence until permanent departure from the den site). We then determined whether the exposure elicited a response by the denning polar bear based on probabilities derived from the reviewed case studies (Woodruff et al. 2022a), which were updated with data from the dens monitored in 2022 and 2023 using the methods described in Woodruff et al. (2022a).

Specifically, we divided the number of cases that documented responses associated with either a Level B harassment ( i.e., potential to cause a disruption of behavioral patterns), Level A harassment ( i.e., potential to injure an animal), or lethal take (e.g., cub abandonment) of polar bears by the total number of cases with that combination of period and exposure type (table 6). Level B harassment was applicable to both adults and cubs, if present, whereas Level A harassment and lethal take were applicable to only cubs. AIR surveys were not considered to be a source of potential impact. In thousands of hours of AIR surveys conducted in northern Alaska over the last decade, we are not aware of a single instance of a polar bear abandoning its den during the early denning period due to an AIR survey overflight. These responses would be readily observable on the thermal cameras, and the fact that none have been observed indicates that den abandonment very likely does not occur given the brief duration of the aircraft overflight and the distance and altitude of the aircraft from the den site. Recent peer-reviewed research further supports the model assumption that AIR surveys are not a source of harassment (Quigley et al. 2024).

For dens exposed to activity, we used a multinomial distribution with the probabilities of different levels of take for that period (table 6) to determine whether a den was disturbed or not. If a lethal take was simulated to occur, a den was not allowed to be disturbed again during the subsequent denning periods because the outcome of that denning event was already determined.

The level of impact associated with a disturbance varied according to the severity and timing of the exposure (table 6). Exposures that resulted in emergence from dens prior to cubs reaching 60 days of age were considered lethal takes of cubs. If an exposure resulted in a Level A harassment during the late denning period, we first assigned that den a new random emergence date from a uniform distribution that ranged between the first date of exposure during the late denning period and the original den emergence date. We then determined whether that den was disturbed during the post-emergence period, but the probability of disturbance was dependent on whether or not a den was disturbed ( i.e., Level A harassment) during the late denning period (table 6). If an exposure resulted in a Level A harassment during the post-emergence period, we assigned the den a new time spent at the den site post-emergence from a uniform distribution that ranged from 0 to the original simulated time at the den post-emergence.

Recent research suggests that litter survival is related to the date of den emergence and time spent at the den post-emergence (Andersen et al. 2024), with litters having higher survival rates the later they emerge in the spring, and the longer they spend at the den site after emergence. To determine whether dens that were disturbed during the late denning and/or post-emergence period(s) experienced Level A harassment, we relied on estimates of litter survival until approximately 100 days post emergence, derived from the analysis of empirical data on the dates of emergence from the den and departure from the den site (Anderson et al. 2024). These estimates are dependent on the date of emergence and time spent at the den site post-emergence. For each den disturbed during the late denning and/or post-emergence periods, we obtained a random sample of regression coefficients from the posterior distribution and calculated the probability of a litter surviving approximately 100 days post-emergence with the following equation:

logit ( s ) = β₀ + β₁emerge + β₂depart

where s is the probability of at least one cub being alive approximately 100 days post-emergence, β₀ is the intercept coefficient, β₁ is the coefficient associated with the Julian date of emergence ( emerge ), and β₂ is the coefficient associated with the number of days the family group stayed at the den site post-emergence before departing ( depart). These probabilities are based on estimates of litter survival derived from the analysis of empirical data on the dates of emergence from the den and departure from the den site (Anderson et al. 2024).

We developed the code to run this model in program R (R Core Development Team 2020) and ran 10,000 iterations of the model ( i.e., Monte Carlo simulation) to derive the estimated number of dens disturbed and associated levels of harassment. We then determined the number of cubs that would have lethal take, Level A harassment, and Level B harassment, and the number of females that would experience Level B harassment. Table 6 shows the probability of an exposure resulting in the types of harassment of denning polar bears. ( print page 2729)

Table 6—Probability That an Exposure Elicited a Response by Denning Polar Bears That Would Result in Level B Harassment, Level A Harassment, Lethal Take, or No Take

Denning period	None (sow or cub(s))	Level B (sow)	Level B (cub(s))	Level A (cub(s))	Lethal (cub(s))
Den Establishment	0.750	0.250	0.000	0.000	0.000
Early Denning	0.860	0.140	0.000	0.000	0.130
Late Denning	0.510	0.490	0.000	0.490	0.000
Post Emergence—Previously Undisturbed Den	0.000	1.000	0.200	0.800	0.000
Post Emergence—Previously Disturbed Den	0.000	1.000	0.474	0.526	0.000
Note: Level B harassment was applicable to both adults and cubs, if present; Level A harassment and lethal take were applicable to cubs only and were not possible during the den establishment period, which ended with the birth of the cubs. Probabilities were calculated from the analysis of 60 case studies of polar bear responses to human activity. During the early denning period, there was no Level A harassment for cubs, only lethal take. We provide two sets of take probabilities for the post-emergence period. The first (Post-emergence—Undisturbed) is the set of probabilities when a den has not been disturbed during the late denning period. The second (Post-emergence—Disturbed) is the set of probabilities for a den that was disturbed during the late denning period (Rode et al. 2018; Andersen et al. 2024).

Model Results

Our analysis estimates a mean of 1.36 (median = 1; 95 percent CI: 0-4) land-based dens in the project area will potentially be exposed to disturbance from the specified activities during the 1-year period of the proposed IHA. Our den simulation analysis predicts this degree of potential exposure will have a zero (0) percent chance of incurring Level B harassment. Furthermore, our analysis predicts a zero (0) percent probability of the BLM's specified activities resulting in either Level A harassment or lethal take during the 1-year period of the proposed IHA.

Critical Assumptions

To conduct this analysis and estimate the potential amount of Level B harassment, Level A harassment, and lethal take, we made several critical assumptions.

Level B harassment is equated herein with behavioral responses that indicate harassment or disturbance, but not to the extent that cause the animal to experience significant biological consequences. Our estimates do not account for variable responses by polar bear age and sex; however, sensitivity of denning polar bears was incorporated into the analysis. The available information suggests that polar bears are generally resilient to low levels of disturbance. Females with dependent young and juvenile polar bears are physiologically the most sensitive (Andersen and Aars 2008) and most likely to experience harassment from disturbance. Not enough information on composition of the SBS polar bear stock in the specified project area is available to incorporate individual variability based on age and sex or to predict its influence on harassment estimates. Our estimates are derived from a variety of sample populations with various age and sex structures, and we assume the exposed population will have a similar composition, and that, therefore, the response rates are applicable.

The estimates of behavioral response presented here do not account for the individual movements of animals in response to the specified activities. Our assessment assumes animals remain stationary ( i.e., density does not change). Not enough information is available about the movement of polar bears in response to specific disturbances to refine this assumption.

The SBS polar bears create maternal dens on the sea ice as well as on land. The den simulation used in our analysis does not simulate dens on the sea ice. However, the specified activities will be conducted entirely on land and only a small percentage of the activities will occur within 1.6 km (1 mi) of the coastline. Therefore, the impact of the activities will be primarily limited to land-based dens within 1.6 km (1 mi) of the project impact areas used during denning season. Additionally, this impact area will be surveyed during AIR surveys to mitigate impacts on denning polar bears.

The specific combination of snow trail segments depicted in figure 1 that will be used for mobilization, resupply, and backhauling is not currently known. For the purposes of the above analyses and estimates of take by Level B and Level A harassment, and the risks of lethal take, we assumed that all routes within the AIR surveyed section (figure 1) of the project might potentially be used at some point during the denning season. This assumption results in a very conservative estimate of take for the 1-year IHA period that accounts for all possible operational scenarios.

Sum of Harassment From All Sources

Our analyses quantified the total number of Level B harassment, Level A harassment, and lethal take likely to result from the BLM's specified activities. We evaluated three potential sources of harassment/take, including surface interactions, aircraft overflights, and den disturbance of sows and/or cubs in our analyses. A summary of total estimated take via Level B harassment during the project by source is provided in table 7. We do not anticipate take by Level A harassment or lethal take to occur.

Table 7—Total Estimated Takes by Harassment of Polar Bears, by Source

Source and type of harassment	Number of estimated harassments
Bears on the surface—summer—Level B harassment	1
Bears on the surface—winter—Level B harassment	10
Aircraft activities—summer and winter—Level B harassment	1
Total	12