2019 Status of Akia-Maniitsoq caribou population, Central region, West Greenland. Technical Report No. 124

1 2019 Status of Akia-Maniitsoq caribou population, Central region West Greenland T Te ec ch hn ni ic ca al l R Re ep po or rt t N No o. .1 12 24 4, , 2 20 02 23 3 P Pi in nn ng go or rt ti it ta al le er ri if ff fi ik k – – G Gr re ee en nl la an nd d I In ns st ti it tu ut te e o of f N Na at tu ur ra al l R Re es so ou ur rc ce es s 2 Title: 2019 Status of Akia-Maniitsoq caribou population, Central region, West Greenland. Authors: Christine Cuyler1, Tiago A. Marques2, Iúri J.F. Correia3, Aslak Jensen4, Hans Mølgaard5 and Jukka Wagnholt6 1 Pinngortitaleriffik – Greenland Institute of Natural Resources, P.O. Box 570, 3900 Nuuk, Greenland 2 CREEM, University of St Andrews, School of Mathematics and Statistics, Scotland 3 University of Lisbon, Faculty of Sciences, Portugal 4 Solviaq 15, 3900 Nuuk, Greenland 5 P.O. Box 122, 3911 Sisimiut, Greenland 6 Tusass, P.O. Box 1002, 3900 Nuuk, Greenland Series: Technical Report No. 124, 2023 Date of publication: 06 April 2023 Publisher: Pinngortitaleriffik – Greenland Institute of Natural Resources Financial support: Government of Greenland and Pinngortitaleriffik – Greenland Institute of Natural Resources ISBN: 978-87-972977-7-3 ISSN: 1397-3657 EAN: 9788797297773 Cover photo: Aslak Jensen: One of many large caribou groups for the AkiaManiitsoq population, Narssarssuaq valley, Central region. Cited as: Cuyler, C., Marques, T.A., Correia, I.J.F., Jensen, A., Mølgaard, H. & Wagnholt, J. 2023. 2019 Status of Akia-Maniitsoq caribou population, Central region, West Greenland. Pinngortitaleriffik – Greenland Institute of Natural Resources. Technical Report No. 124. 93 pp. Contact address: The report is only available in electronic format. PDF-file copies can be downloaded at this homepage: https://natur.gl/forskning/rapporter/ Pinngortitaleriffik – Greenland Institute of Natural Resources P.O. Box 570, 3900 Nuuk, Greenland Phone: +299 36 12 00 E-mail: [email protected] www.natur.gl 3 2019 Status of Akia-Maniitsoq caribou population, Central region, West Greenland By Christine Cuyler1, Tiago A. Marques2, Iúri J.F. Correia3, Aslak Jensen4, Hans Mølgaard5 and Jukka Wagnholt6 1 Pinngortitaleriffik – Greenland Institute of Natural Resources, P.O. Box 570, 3900 Nuuk, Greenland 2 CREEM, University of St Andrews, School of Mathematics and Statistics, Scotland 3 University of Lisbon, Faculty of Sciences, Portugal 4 Solviaq 15, 3900 Nuuk, Greenland 5 P.O. Box 122, 3911 Sisimiut, Greenland 6 Tusass, P.O. Box 1002, 3900 Nuuk, Greenland Technical Report No. 124, 2023 Pinngortitaleriffik – Greenland Institute of Natural Resources 4 [Empty page] 5 Table of Contents Summary (English)……………………………………… 8 Eqikkaaneq (kalaallisut) ………………………………. 9 Resumé (dansk)………………………………………… 11 Introduction ……………………………………………. 13 Methods …………………………………………………. 15 Results…………………………………………………… 23 Discussion ……………………………………………… 47 Acknowledgements……………………………………... 55 Literature cited.………………………………………… 56 Figures 1. Borders of the Central region, … Page 13 2. Area covered by 2019 survey of the Central region (11,575 km2), which is… Page 16 3. The 50 line transects used in the 2019 survey of Central region… Page 19 4. Location and group size of caribou detections (truncated data) … Page 25 5. Exploratory analysis for the number of detections by sub-area … Page 26 6. Exploratory analysis for group size distribution among detections… Page 26 7. Exploratory analysis for caribou encounter rate per line transect… Page 27 8. Observer effect: histograms illustrating detected distances for Observers … Page 27 9. No. caribou detections per covariates heli.side, flat light, and camouflage … Page 29 10. No. caribou detections per covariates vegetation/ground or boulders… Page 29 11. No. caribou detections per covariates solar glare, dead ground, weather… Page 30 12. Summary of the frequency of elevations flown (truncated data). Page 30 13. Histogram of observed caribou distances for non-truncated data and … Page 31 14. Histogram for Hazard-rate with Camouflage as covariate of detected … Page 33 15. Estimated probabilities of detection for each observed group size… Page 33 16. Caribou density (left) and abundance (right) estimates with corresponding Page 35 17. Unusually high numbers of caribou, often in exceptionally large groups … Page 40 18. These 17 caribou could be sexed and aged owing to groups separation … Page 40 19. Observed frequency of cow-calf pairs for 96 groups for which… Page 42 20. Observed recent snowmobile activity in Akia (Nordlandet), which … Page 45 21. Past and present caribou population size estimates with confidence … Page 47 22. Past and present caribou density estimates for the Akia-Maniitsoq … Page 47 23. Past and present late winter bull to cow ratios for the Akia-Maniitsoq … Page 50 24. Past and present late winter calf (age 10-month) recruitment … Page 50 25. Akia sub-area, illustrating typical conditions. Page 58 26. Fog over the Maniitsoq Coast sub-area, which prevented planned flights. Page 58 27. Akia sub-area of Central region, illustrating just north of Greenland’s … Page 59 28. Akia sub-area of Central region, illustrating rugged Akia (Nordlandet) … Page 59 29. Akia sub-area, west end of line transect 10, view ENE. Page 60 30. Akia sub-area, thin mouth of Niaqungunaq (Fiskefjord) and conditions … Page 60 31. Akia sub-area in the valley, Narssarssuaq, illustrating ground conditions… Page 61 6 32. Akia sub-area illustrating ground conditions at east end of line transect 22… Page 61 33. Akia sub-area, valley north of Ilulialik fjord, illustrating ground … Page 62 34. Akia sub-area, view west across unnamed peninsula (which is … Page 62 35. Maniitsoq Coast sub-area, view to the north from line transect 55, in … Page 63 36. Maniitsoq Coast sub-area illustrating mountainous terrain and conditions… Page 63 37. Ujarassuit sub-area, illustrating typical conditions for terrain, snow, and … Page 64 38. Ujarassuit sub-area on the one day with sunshine, view west from the … Page 64 39. Ujarassuit sub-area, east end of line transect 35, view to west illustrating … Page 65 40. Ujarassuit sub-area, illustrating typical terrain, snow, and sunlight … Page 65 41. Ujarassuit sub-area terrain, snow, and flat light conditions at the east … Page 66 42. Ujarassuit sub-area, illustrating typical terrain, snow, and sunlight … Page 66 43. Ujarassuit sub-area, illustrating rugged highland terrain, snow depths … Page 67 44. Ujarassuit sub-area, xeric conditions of east end line transect 32, view N … Page 67 45. Ujarassuit sub-area, Ivisartoq highlands view east towards small bay … Page 68 46. Ujarassuit sub-area, Ivisartoq highlands illustrating almost non-existent … Page 68 47. Place names used regarding Central region (ca. 64°–66°N; 50°–53°W), … Page 69 48. Plot sampling grid example of total area A divided into smaller plots… Page 70 49. Example of a patch of tundra with the transect in the middle… Page 73 50. Half-normal (top row) and hazard-rate (bottom row) detection functions… Page 75 51. Possible shapes for the detection function when cosine adjustments are…. Page 76 52. A good model for the detection function should have a shoulder… Page 79 53. Eight caribou, several camouflaged against background, within 200 m of … Page 84 54. Three caribou camouflaged against background, within 100 m from … Page 84 55. Histograms for detected distances superimposed with estimated … Page 87 56. The three observers, Dr. C. Cuyler…, Aslak Jensen, …, Hans Mølgaard … Page 89 57. AS350 Helicopter viewing windows for the left and right sides, … Page 91 Tables 1. Late winter population parameters, Akia-Maniitsoq caribou population… Page 14 2. Summary of unprocessed results: Survey of Akia-Maniitsoq caribou … Page 23 3. Model comparison across three Conventional Distance Sampling models… Page 34 4. Encounter rate (ER) estimates per sub-area (stratum) for caribou groups… Page 35 5. Estimates of abundance per sub-area (stratum) for the Akia-Maniitsoq … Page 36 6. Estimates of density per sub-area (stratum) for the Akia-Maniitsoq … Page 36 7. Movement or non-movement of caribou reacting to helicopter fly-by… Page 37 8. Details for movement or non-movement of caribou reacting to helicopter … Page 38 9. Demographics for Akia-Maniitsoq caribou population, Central region … Page 39 10. Group size relative to composition Akia-Maniitsoq caribou population … Page 42 11. Approximate elevations for caribou groups observed… Page 44 12. Commonly used key functions and series expansions for detection function. Page 74 13. Detection function parameters’ estimates. Page 87 14. Recent caribou population estimates & minimum counts for West … Page 92 7 Appendices 1. Photos, Central region aerial survey conditions…March 2019 Page 59 2. Place names for the Central region Page 70 3. Statistical methods behind Distance Sampling Page 71 4. Distance Sampling Assumptions – short summary Page 84 5. Photos of camouflaged caribou observed March 2019 Page 85 6. Histograms for detected distances Page 86 7. Recommendations for improving future surveys Page 89 8. Recent caribou population estimates & minimum counts for West Greenland Page 93 Raw data may be accessed by contacting Pinngortitaleriffik – Greenland Institute of Natural Resources, Department of Mammals and Birds. 8 Summary (English) This report presents results from the aerial survey carried out by helicopter in early March 2019, for the Akia-Maniitsoq caribou population inhabiting the Central region in West Greenland. This population was last surveyed in March 2010. New estimates of abundance were overdue. Helicopter surveys in 2001, 2005 and 2010 used strip transect counts. The March 2019 helicopter survey, however, used Distance Sampling methods. For March 2019, the Akia-Maniitsoq caribou population abundance was estimated at 48,941 caribou (95% CI: 37,612 – 63,682; CV = 0.131; SE = 6390), with a density of 4.2 ± 0.5 caribou/km2 (95% CI: 3.2–5.5). The distance sampling estimate was precise given the excellent CV value (0.13). Survey coverage was 9.6% (truncated data), which is a ca. five-fold improvement from the 2001, 2005, 2010 strip transect count surveys, where coverage was always below 2%. Despite 18 years of harvest management aimed at controlling caribou abundance and density, by March 2019 Akia-Maniitsoq caribou population size was double what it was in 2010. Confidence Intervals for the 2010 and 2019 survey estimates do not overlap, therefore we conclude the Akia-Maniitsoq population size truly doubled over the nine-year period, 2010-2019. The overall density estimate for the Akia-Maniitsoq population was 4.2 caribou per km2. This value is much greater than the management recommended target of 1.2 caribou per km2 (Kingsley & Cuyler 2002, Cuyler et al. 2007). Further, density in the Akia sub-area averaged 6.6 caribou per km2. Exceeding the target caribou density is assumed to raise the risk of overgrazing and thus decline in caribou abundance. The late-winter calf percentage and calf recruitment were the highest ever recorded from helicopter surveys. Values remained high even after the unusually high number of orphan calves (n = 83) were removed from the calculations. The 2019 ratio of 49 bulls to 100 cows, reversed the downward trend (from 58 to 38 bulls per 100 cows) in 2001-2010 period for animals age > 1year. We conclude the 2019 demographics of the Akia-Maniitsoq population provides potential for further growth in abundance, albeit notwithstanding future catastrophic stochastic events, including extreme weather and pathogen outbreaks. 9 Environmental conditions during the 2019 survey provided extraordinary camouflage for caribou. Pooling environmental covariates into a single index for camouflage will improve detection function modelling. Skilled observers and flying helicopters low and slow were critical factors permitting detection of caribou, specifically because 25% of all groups remained stationary. Beyond population parameters, results of interest included relatively low elevations, mean 351 m, used by the Akia-Maniitsoq caribou population in early March. Among other things, this reflects the relative abundance of low elevations in the region. Further, although antlers are typical on female caribou, few (32%) cows possessed antlers in the Akia-Maniitsoq population. Eqikkaaneq (kalaallisut) Uani nalunaarusiami saqqummiunneqarput nunap immikkoortuani tuttutassiissutinik aqutsiviusuni qulimiguulik atorlugu Martsip aallartilaarnerani 2019-imi tuttunik kisitsinernit inernerit. Tuttut taakku pineqartut nunaatta kitaata qiterpasissuaniipput. Tuttutoqatigiiaat kingullermik kisitsivigineqarput martsimi 2010-mi. Taamanikkut tuttut amerlassusiisa missiliorneqarneranni amerlanaagaapput. Qulimiigulik atorlugu 2001-imi, 2005-imi, 2010-milu kisitsinerni periuseq atorneqartoq tassaavoq, qulimiguulimmit takusat aalajangersimasumik kisitseriaaseq, taaguuteqartinneqartoq Distance Sampling methods. Akia-Maniitsumi tuttut marts 2019 48,941-nik (95% CI: 37,612 – 63,682; CV = 0.131; SE = 6390) amerlassuseqarnissaat missiliuunneqarpoq, naatsorsuinertigullu tuttut kvadratkilometer-imut, km2-imut, eqimassuseqarnissaat 4.2 ± 0.5 tuttut/km2 (95% CI: 3.2–5.5) aalajangiunneqarpoq. The distance sampling missiliussineq, CV-kisitsit ajunngilluinnartoq (0.13-ulluni). Nuna kisitsiviusoq qulangiuaarneqartorlu 9.6%-iuvoq (truncated data), tassuunalu takuneqarpoq tuttut 2001-imi, 2005-imi 2010-milu kisinneqarneranni tallimariaammik amerleriarsimasut, kisitseriaaseq taannarpiaq atorlugu, kisiannili nunap kisitsivigineqartup angissusia tamatigut 2 % ataattarsimagaa. Naak ukiuni 18-ini tuttut ikilisarniarlugit kiisalu amerlassusiisigut eqimassusiisa aqussinnaalernissaat anguniarlugu aqutsisoqarsimagaluartoq, taamaattoq marts 2019-imi 2010-mut sanilliullugu marloriaammik amerlassuseqalersimapput. CI (confidence interval, tassa kisitsisit tutsuigina- 16 Figure 2. Area covered by the 2019 survey of the Central region (11,575 km2), which is inhabited by the Akia-Maniitsoq caribou population. Four different colours illustrate the four sub-areas, designated as Akia, Maniitsoq coast, Maniitsoq inland, and Ujarassuit. Greenland’s capital city, Nuuk, is the red diamond on the tip of grey peninsula arm in bottom left corner. 17 The Central region (ca. 64°/64°45’–66°N; 50°–53°W) is shared by the Qeqqata and Nuuk Kommunia. The only large settlement, which is situated on an island, is the coastal city of Maniitsoq with ca. 2,534 inhabitants. Otherwise, there are three small villages, also coastal, which combined contain a further ca. 650 people. In 2019, use of snowmobiles was still prohibited beyond settlement boundaries. See Appendix 2 for place name details. The Central region is seasonally ice-free and covers an area of 11,575 km2, (excluding lakes, rivers, sand, glaciers, and islands) (Fig. 2). Previous surveys reported a less precise larger land area of ca. 15,362 km2, which included lakes, rivers, and islands (Cuyler et al. 2003, 2005, 2011). The northern border is formed by two ice caps (i.e., Kangaamiut Sermiat (Sukkertoppen Ice Cap) and Tasersiap Sermia) and the western portion of the Kangerlussuaq Fjord. The southern border is framed by the Nuuk fjord (Godthåbsfjord), which currently is ice-free year-round to as far as Ilulialik Bay or the Ujarassuit paavat fjord-arm. The western border is the permanently ice-free seacoast of the Davis Strait, and the eastern border is the Greenland Ice Cap. The coastal topography in the north is mountainous. Elevations are often 1000 to 2000 m and glaciers predominate. Midway, the coast is mountainous with peaks from about 500 to 1200 m. In the south, coastal topography is rugged lowlands of elevations generally below 200 m. Field methods Since 2000, early March has been the chosen period for surveys because caribou dispersion is high, group size is small with low variability and daily movement is at the annual minimum (Cuyler et al. 2007, 2011, 2016; Poole et al. 2013). The former two reduce variance among transects, diminish counting error, and maximize precision, while the latter lowers movement between or along transects. The aerial survey period for the Akia-Maniitsoq caribou population was 01-12 March 2019. The platform for observation was a helicopter AS350. Pilot monitoring of helicopter radar altimeter made maintenance of a constant altitude possible by constantly adjusting for terrain features while flying low (40 m, ca. 120 feet) and slow (ca. 65 km/hour). Participants included three observers, all with previous survey experience: GINR’s senior scientist Christine Cuyler, professional hunter Aslak Jensen (Greenland Association of Professional Hunters (KNAPK)) from Nuuk and Sisimiut hunting officer Hans Mølgaard. Cuyler always sat in front and was the 18 data recorder. Cuyler (Observer 2) focused on detecting caribou directly on track line (center line, 0-line) before animals fled offline owing to approaching helicopter. Jensen and Mølgaard (Observer 1 and 3, respectively) were seated in the rear of the helicopter, on either side. The side they sat on alternated each time the helicopter was refueled, which was usually once daily and sometimes twice. Jensen and Mølgaard could not view the track line but observed animals for all distances beyond. Verbal contact among the observers permitted the digital audio recording of all observations and, most importantly, prevented any double counting of groups detected by more than one observer. Two audio devices (SONY IC recorder, ICD-SX712) were used to record separately the observations specific to the left and right side of the line transect. Audio recording devices were on continual recording for each line transect. At the end of each survey day, audio data was downloaded to computer for storage and back-up. Observations were later paired with Global Positioning System (GPS) coordinates of the helicopter at the time of observation. For each detection, the audio recording included distance to (see below) caribou group, as well as group size and behavior and name of the observer. Ground surface and weather conditions were also recorded. Manual click-counters, logging the number of caribou seen by an individual observer, provided low-tech back-up for doublechecking the digital audio observations for each line transect. Survey design Aligning line transects perpendicular to known gradients within the surveyed area can maximize precision of the resulting estimate by lowering the encounter rate variance (Buckland et al. 2001). Thus, the transect axis direction (east-west axis) was chosen as perpendicular to previously known animal distribution gradients in March and the west-east climate gradient from wet maritime to dry continental. An initial line transect was computer generated at random in each sub-area (see below), and the others followed at 10 or 20 km apart. The line transects flown provide the maximum area coverage possible given the financial resources available. Because some a priori transects became combined during the survey, line identification numbers are not consecutive. The surveyed Central region, 11,575 km2, was divided into four sub-areas, named Akia (4,150 km2), Maniitsoq Coast (3,375 km2), Maniitsoq Inland (2,011 km2) and Ujarassuit (2,039 km2) (Fig. 2). Sampling design for the 2019 survey considered 50 systematic parallel line transects (track lines flown) of variable length placed over the four sub-areas (Fig. 3). Line transects were separated by 10 km, excepting Maniitsoq Coast sub-area, which were 20 km. 19 Figure 3. The 50 line transects used in the 2019 survey of the Central region, Akia-Maniitsoq caribou population, employing the same four colours as applied to the four sub-areas in figure 2: Akia (red, 21 lines), Maniitsoq Coast (blue, 11 lines), Maniitsoq Inland (orange, 8 lines), and Ujarassuit (black, 10 lines). Line transects separated by 10 km, except Maniitsoq Coast, which were separated by 20 km. Line transect numbering is not consecutive, as some a priori lines became amalgamated during survey. The distance to a detected caribou group (object-of-interest) was before caribou movement occurred. Tightly cohesive behavior identified groups of multiple individuals. Excepting groups on the track line, which was distance 0 m, distance was the observer’s instantaneous and subjective estimate of the distance to center of the caribou group. Exact distance measurement from the track line (aka 0-line or center line) to a caribou group was effectively never possible because of practical considerations (details in Cuyler et al. 2021). Therefore, like all previous helicopter caribou surveys in Greenland, for distance measurement perpendicular to the track line, we approximated with rough “distance bins”, i.e., in meters, 0-50, 50-100, 100-200, 200-300, 300-400, 400-500, 500-600, 600-700, 700-800, 800-900, 900+. Bin value recorded for a group was always the upper limit of the bin applied. For analysis, we did not correct for the 40m altitude of the helicopter. Instead, we re-coded the reported upper distance values to mid-distance for a specific bin owing to three reasons. First, a caribou group could be at any distance within the bin., e.g., a group recorded in distance bin 300 m, was located somewhere between 200 and 300 meters. Second, placing 20 a caribou group within the correct bin relied heavily on observer ability to estimate distance to the observed animals in rugged terrain. Third, although for level ground (itself rare) the estimated direct line distance from observer (sitting in helicopter at 40 m altitude above ground) to a caribou group would be greater than the perpendicular distance from the track line to that group, those differences were small at 100 m and negligible beyond 200 m (i.e., in meters 8, 4, 3, 2, 2, 2, 1, 1, and 1). Regarding the 0-50 m bin, we assumed observer ability sufficient to compensate for 40 m altitude and assign a perpendicular 50 m distance correctly because immediately adjacent to the helicopter/track line. Further, to aid observer ability to estimate distances, before starting survey the helicopter hovered at 40m altitude while each observer used a “Leica laser range finder 1600” to gauge distances across level airport ground to a priori known perpendicular distances. Then observers marked their window with masking tape delineating the approximate distances for each bin. While on survey, in the absence of caribou and where vertical terrain features occurred, observers used the laser distance finders to test their ability to estimate distance, i.e., to the terrain feature. On rare occasions, observers were able to use the laser range finders to determine bin distance to a detected stationary group. Once all recorded distances were recoded to mid-distance, to model the detection function all the detections were pooled across observers with the helicopter functioning as a single observer. The pooled data were used to estimate a detection function, then estimate the detection probability and finally to estimate the density of the caribou within the surveyed area (Buckland et al. 2001). The detection function, 𝑔(𝑦), describes the probability of detecting an object-of-interest given that is at a distance 𝑦, from the track line, thus being a non-increasing function of 𝑦 (Buckland et al. 2015). For line transects, 𝑦 is the perpendicular distance from the track line to the detected object. Within DS methods, the probability of detection is explained recurring to these observed distances (Buckland et al. 2001). Distance sampling The caribou group was the selected object-of-interest on which detectability was modelled, i.e., individual caribou within a group were not considered. The individual line transects were the sample unit for design-based conventional DS analysis of the 2019 survey. Details for how this study’s DS analyses were performed are in Appendix 3. Thus, estimated CVs (Coefficients of Variation) from the models are referring to the transects, and total CV estimation is obtained by dividing the estimated standard error by the respective estimate. 21 The estimated standard error is obtained as a pooled estimate for entire region and accounting for transects and their variability, it incorporates the variance from the detection function (Buckland et al. 2001). The recorded distances to the observed caribou groups were used to estimate a detection function. With this, both the caribou detection probability and density within the surveyed area could be estimated (Buckland et al. 2001). The detection function, 𝑔(𝑦), describes the probability of detecting an object of interest (caribou group) given that it is at a distance 𝑦, from the track line, thus being a non-increasing function of 𝑦 (Buckland et al. 2015). For line transects, 𝑦 is the perpendicular distance from the track line to the detected object. Within DS methods, the probability of detection is explained recurring to these observed distances (Buckland et al. 2001). Prior to DS analysis, the raw data was first processed for inconsistencies. Then extensive exploratory data analysis was completed, including evaluation of observed distances, before proceeding to determining the detection function through model fitting and selection (Buckland et al. 2001; Marques et al. 2011; Thomas et al. 2010). To determine the detection function, several models were considered, (Thomas et al. 2010). The model presenting the lowest AIC value was chosen. Details regarding DS theory, methods and analysis are available in Buckland et al. (2001, 2015), and a briefer summary provided in Appendix 3, with a summary of DS assumptions in Appendix 4. For analysis, we used R Statistical Software (https://www.r-project.org/). Demographics Sex, age, and late-winter calf recruitment observations were obtained after most of the DS survey was completed. All caribou sighted were sexed and aged following a brief overpass with the helicopter. Sex and age criteria have remained unchanged since 2000 (details in Cuyler et al. 2011, 2016). Briefly, female sex was determined by the presence or absence of a vulva and/or urine patch on the rump of both adults and calves, i.e., antler size, shape, presence, or absence, were not used to determine sex. Two age classes were used, calf (age  10-months) and adult (age > 1-year). Age was determined by body size. 10month-old calves, male and female, being considerably smaller than all other age classes in March. Calf percentage is given relative to the total number of caribou sexed and aged. Calf recruitment is the value for late-winter and provided as the number of calves per 100 cows. Group size was based on proximity and group cohesion during possible flight response. To obtain 22 demographics and recruitment values, on 12 and 13 March, large areas of the Central region were flown, including the south coast of Ujarassuit and Ilulialik, Akia-Nordlandet between Fiskeand Nuuk fjords, Narssarssuaq Valley and peninsula, area surrounding line 32 and between line 32 and 30. Elevations where caribou detected Early March elevation use by caribou was approximated using GPS dataset for helicopter elevation/position and matching timestamps with those of the digital audio recording of caribou observations. GPS and digital recorder timestamps were synchronized before the survey began. Before analysis, the helicopter’s flight altitude of 40 m was subtracted from all elevations. Thereafter, and lacking a reliable constant correction factor, negative values were deleted. Natural mortality In the past, if locals/hunters observed several caribou carcasses in the terrain or on sea ice, this resulted in alarm about an assumed negative trend for the entire population. To put carcass observations into perspective, since 2000, all technical reports for Greenland caribou surveys have included, for that specific survey year, the expected number of annual adult caribou deaths resulting from natural mortality, i.e., not due to harvest. Age distributions among harvested Greenland caribou populations have suggested a natural mortality of from 8 to 10% per annum (Loison et al. 2000, Cuyler & Østergaard 2005). Meanwhile, natural mortality rates from 4 to 8% were reported for North American populations without predators (Bergerud 1967, 1971, Skoog 1968, Kelsall 1968, Heard & Ouellet 1994), albeit these are now considered low (Bergerud et al. 2008) and density-independent factors, e.g., adverse weather, can increase mortality (Gates et al. 1986). Bergerud (1980) proposed a standard adult mortality rate of 10% for all North American caribou populations, and more recently Bergerud et al. (2008) suggested 7.7% for an increasing population with predators. Large predators are absent in the Central region. Although natural mortality rates vary among years (Bergerud et al. 2008), given the above, an assumed standard natural mortality rate of 8-10% (Kingsley & Cuyler 2002) for Greenland caribou likely yields a reasonable estimate of annual mortality owing to natural causes. This rate is applied to the 2019 abundance estimates to provide wildlife managers with a rough number of expected caribou deaths due to natural mortality within the survey year. 23 Results Survey logistics & unprocessed data The aerial survey by helicopter of the Akia-Maniitsoq caribou population occurred within the period 01-14 March, which period was shared with the survey of the Ameralik caribou population. Poor weather made three days nonflyable, as did airport closures on two Sundays. DS data for the Akia-Maniitsoq caribou population was obtained over six days (01, 05, 06, 07, 09 and 12 March). Demographics data was obtained 12-13 March. Typical AS350 helicopters carrying three passengers and pilot, refueling was necessary after about 3 hours of flight time, an additional 15-20 minutes were possible when wind conditions and distance to nearest airport permitted. Table 2. Summary of unprocessed results: Survey of the Akia-Maniitsoq caribou population by helicopter in the Central region, 01-12 March 2019. Parameter Central region sub-area TOTAL Akia Maniitsoq Coast Maniitsoq Inland Ujarassuit Flight altitude (m) 40 40 40 40 40 Flight speed (km/hr) 60-70 60-70 60-70 60-70 60-70 Sub-area size (km2) 4,150 3,375 2,011 2,039 11,575 Number of lines 21 11 8 10 50 Distance flown (km) 423.12 197.26 244.17 241.02 1105.57 Strip width1 (m) 1000-1500 1000-1500 1000-1500 1000-1500 1000-1500 Surveyed area ca. (km2) 846 - 1,269 394 - 592 488 - 732 482 - 723 2,211 - 3,316 Coverage2 20.4-30.6 % 11.7-17.5 % 24.3-36.4 % 23.6-35.5 % 19.1-28.7 % Coverage post-truncation3 10.2 % 5.8 % 12.1 % 11.8 % 9.6 % Total caribou observed 1540 276 391 608 2,815 # Groups observed 385 73 116 175 749 Mean group size 3.99 3.78 3.37 3.47 3.75 Std Deviation group size ± 3.46 ± 3.80 ± 1.97 ± 2.04 ± 3.03 Median group size 3 3 3 3 3 Maximum group size 25 22 10 15 25 Minimum group size 1 1 1 1 1 1 Strip width provided is to one side of helicopter only. Must double for total strip width. 2 Coverage prior to truncation of strip width to 600 m. 3 Coverage after truncation of the strip width to 600 m for DS analyses (see page 24). The helicopter flight time totaled 34 hours and 38 minutes. This is 10 hours and 38 minutes more than flown in the last Central region survey, 2010. Time flown was divided between line transect DS survey (25 hours; 42 minutes) and the demographics survey (08 hours; 56 minutes). The 2019 survey used 50 line transects for a total distance flown of 1,106 km, i.e., Akia 423.1 km, Maniitsoq Coast 197.3 km, Maniitsoq Inland 244.2 km and Ujarassuit 241 km (Table 2). 24 Given the 1,106 km of line transects flown, an optimistic calculation of survey coverage of the Central region’s surveyed Akia-Maniitsoq area (11,575 km2) would be 19-29%, i.e., topography permitting and assuming maximum strip width of 1000-1500 m to either side of the helicopter (Table 2). However, for analyses (see DS analysis, page 24), the strip width was truncated to 600 m. Thus, coverage of the Central region surveyed was 9.6% for the final abundance and density estimates of the Akia-Maniitsoq caribou population. The observed raw totals were 749 caribou groups, which included 2,815 caribou (Table 2). Mean group size was 3.75 ±3.03 caribou, and median group size was 3 caribou. Data processing The raw data set was in Excel format containing the survey variables, including region, sub-area, respective areas (km2), transect identification, recorded distances, group size, and GPS coordinates. Sometimes included with caribou group observations were flight characteristics such as helicopter velocity and side, as well as survey characteristics such as solar glare, weather, snow covering and depth, dead ground, and surface conditions providing camouflage backgrounds for the caribou. All variables were properly restructured within R Statistical Software. The data set was subject to some prior processing before analysis. Comment fields were deleted. Variable names were re-coded to make them sensible in R. One caribou observation lacked distance. Given just one observation relative to the large amount the data (n = 749), and since the actual impact of using any given distance value is minor, the pragmatic solution was to use the average observed distance. No observations lacked group size, so no replacements were necessary. Data truncation was set at 0.6 km. Preliminary analysis distance sampling For reliable estimates of abundance, Buckland et al. (2001) suggests that sample size is at least 60 to 80 observations and from a minimum of 10 to 20 replicate line transects. The 2019 caribou survey for the Central region met these recommendations. Regarding observations (detections of groups of one or more caribou), for analysis the untruncated sample size was 749, while the truncated sample size was 734. Similarly, regarding the number of parallel transect lines separated by 10 or 20 km, there were 50 lines. Time required to complete a transect line depended on total length of the line. The following are the results for the truncated data (n = 734 detections). 25 Of the four sub-areas, the Akia sub-area dominated in observation frequency, i.e., number of detections (caribou groups) per sub-area (Fig. 4, 5). This was expected as Akia was the largest sub-area (ca. 4,150 km2) and received the greatest line transect distance flown relative to the sub-areas, Maniitsoq coast, Maniitsoq inland, and Ujarassuit. Further, of the four sub-areas, Akia possesses the most lowlands (elevation < 200 m), and the Akia-Maniitsoq caribou population are known to prefer lowland elevations in late winter (Cuyler et a. 2017). Although caribou were detected on most of the 50 line transects, six line transects lacked caribou detections, i.e., line 4 (Akia), lines 51, 52, 53 (Maniitsoq Coast) and lines 31, 34 (Ujarassuit). Figure 4. Location and group size of the detections (truncated data) observed along the line transects flown, 2019 survey of the Akia-Maniitsoq caribou population. The detected objects of interest, i.e., caribou groups, typically included no more than six animals, while the most observed group size was two animals (n = 193 observations) (Fig. 6). Groups consisting of less than five individuals made up 73% of the observations, while groups counting less than ten individuals made up 96%. Larger groups were scarce and typically observed at greater distances. 32 Detection function models were fitted to the truncated data, i.e., strip width, 𝑤 = 0.60 km, to each side of the helicopter. For these models, every combination of key function and adjustment terms was tested (Appendix 8). Additional covariates assessed were Camouflage, Boulders showing through the snow surface, Group size, Observer, Flat light, Weather, Dead ground, Vegetation and Ground showing through the snow surface, Solar Glare, Helicopter side. Group size, as covariate, did little to explain caribou detection. Regardless, we did not consider converting Group size into a categorical variable, e.g., small, medium, large, because much information is lost for no perceivable gain, since we can estimate a probability of detection for each group size when this covariate is included in the model. A summary of the information from each model fitted to the data (Table 3) provides a simple overview of several models, and includes the respective key functions, adjustment terms, model formula, 𝜒2 Goodness-of-Fit test p-value, estimates of the detection probability, respective standard error (se (𝑃 a)), 𝐴𝐼𝐶, and Δ𝐴𝐼𝐶 comparison between each model and the model with the lowest AIC. The best model fitted to the data possesses the lowest change in AIC value (Δ𝐴𝐼𝐶 = 0). For the 2019 survey data for the Akia-Maniitsoq caribou population, this model has the Hazard-rate function as a key function, Camouflage as a covariate (AIC = 2590.006). The second-best model was the half-normal key with Camouflage as a covariate (AIC = 2590.042, i.e., Δ𝐴𝐼𝐶 = 0.037). This strongly suggests that Camouflage is the relevant covariate in detectability. Thus, we chose the ‘Hazard-rate with Camouflage as covariate’, which had the estimated averaged probability of detection for the Central region of 𝑃 a = 0.471 (se = 0.019) (Fig. 14). It is an averaged estimate since Camouflage is included in model. Consequently, each Camouflage level has its separate detection function, corresponding to different estimates for probability of detection (Fig. 15). The effect of Camouflage was marked on estimated probability of detection (Fig. 15). When Camouflage covariate was low, the estimated probability of detection was greatest. Interestingly, medium Camouflage had the lowest detection probability estimates, while those with high or extreme Camouflage had middle values for detection probability. Large group sizes (≥ 15 caribou) were typically detected when Camouflage was high, which may provide a partial explanation. 33 Figure 14. Histogram for Hazard-rate with Camouflage as covariate of detected distances with the estimated detection function overlaid. Figure 15. Estimated probabilities of detection for each observed group size per Camouflage obtained with the fitted model). 34 Table 3. Model comparison across the three Conventional Distance Sampling models and models considering different covariates further explaining detection. Key function Formula (variable) 𝒙𝟐 p-value 𝑷 a se (𝑷 a) AIC ∆AIC Hazard-rate Camouflage 0.000 0.471 0.019 2590.006 0.000 Half-normal Camouflage 0.000 0.453 0.012 2590.006 0.037 Half-normal Boulders showing through snow 0.000 0.456 0.013 2590.006 7.794 Half-normal Group size 0.002 0.458 0.013 2590.006 8.211 Half-normal Group size + Observer 0.001 0.458 0.013 2600.209 10.204 Hazard-rate Boulders showing through snow NA 0.448 0.020 2600.420 10.414 Half-normal with cosine adjustment term of order 2 1 0.000 0.426 0.019 2604.323 14.318 Hazard-rate with Hermite polynomial adjustment term of order 4 1 0.000 0.402 0.032 2604.566 14.561 Hazard-rate with cosine adjustment term of order 2 1 0.000 0.429 0.019 2604.722 14.716 Half-normal Flat light 0.000 0.461 0.013 2605.587 15.582 Hazard-rate with simple polynomial adjustment terms of order 4,6 1 0.000 0.398 0.034 2605.946 15.941 Half-normal with simple polynomial adjustment term of order 4 1 0.000 0.451 0.021 2606.378 16.372 Half-normal Weather 0.000 0.460 0.013 2606.469 16.464 Uniform with cosine adjustment terms of order 1,2,3 NA 0.000 0.427 0.020 2606.608 16.602 Half-normal 1 0.000 0.462 0.013 2606.810 16.804 Hazard-rate Group Size 0.000 0.458 0.020 2606.884 16.878 Half-normal Dead ground 0.000 0.461 0.013 2606.923 16.917 Half-normal Vegetation/Ground through snow 0.000 0.460 0.013 2606.991 16.985 Hazard-rate Group size + Observer 0.000 0.452 0.020 2607.879 17.873 Half-normal Solar glare 0.000 0.462 0.013 2608.513 18.508 Half-normal Helicopter side 0.000 0.462 0.013 2608.704 18.699 Half-normal Observer 0.000 0.462 0.013 2608.774 18.768 Uniform with simple polynomial adjustment terms of order 2,4,6,8 NA 0.000 0.458 0.023 2610.950 20.944 Hazard-rate Flat light 0.000 0.446 0.020 2614.757 24.751 Hazard-rate Dead ground 0.000 0.447 0.020 2615.115 25.109 Hazard-rate Weather 0.000 0.455 0.020 2615.783 25.778 Hazard-rate Observer 0.000 0.442 0.020 2617.808 27.802 Hazard-rate Helicopter side 0.000 0.449 0.020 2619.033 29.027 Hazard-rate Solar Glare 0.000 0.450 0.020 2619.190 29.184 Hazard-rate Vegetation/Ground through snow 0.000 0.454 0.020 2619.800 29.795 Uniform with Hermite polynomial adjustment term of order 2,4,6 NA 0.000 0.661 0.035 2708.504 118.498 Note: Formula, explanatory variables = 1 for no covariates. NA is for Uniform Key. Chi-square p-value, NA = not enough degrees of freedom for the GOF test, thus the ‘NA’ values. (Degrees of freedom calculated considering model parameters, these vary considering which key function is used and how many/which explanatory variables considered.) 35 The estimates for encounter rates indicate that the Akia sub-area had the most caribou, since its estimate was larger than the other sub-areas (Table 4). Concerning the designbased estimates for caribou abundance and density, Akia is also the sub-area presenting more caribou (Tables 5, 6, Fig. 16). Table 4. Encounter rate (ER) estimates per sub-area (stratum) for caribou groups of the Akia-Maniitsoq caribou population, considering four strata, seven bins, and a detection function fitted with Camouflage as a covariate. Sub-area Encounter rate Standard Error (se) Coefficient of Variation (cv) Akia 0.918 0.145 0.158 Maniitsoq Coast 0.409 0.131 0.321 Maniitsoq Inland 0.475 0.140 0.294 Ujarassuit 0.746 0.105 0.141 TOTAL 0.662 0.071 0.108 Figure 16. Caribou density (left) and abundance (right) estimates with corresponding confidence intervals for the four sub-areas, Akia, Maniitsoq Coast, Maniitsoq Inland, and Ujarassuit, and finally for the total Central region. In March 2019, the Central region had an estimated population size of 48,941 caribou (95% CI: 37,612 – 63,682) (Table 5, Fig. 16). Sum of abundance estimates for each sub-area equals total estimated abundance for the region. CV of 0.13 is excellent and indicates the AkiaManiitsoq caribou abundance estimate for 2019 is accurate. The design–based density estimate for the whole survey region was 4.23 caribou/km2, with 95% CI: 3.23 – 5.50 (Table 6, Fig. 16). 36 Table 5. Estimates of abundance per sub-area (stratum) for Akia-Maniitsoq caribou population in the Central region, March 2019, considering four strata, seven bins and Hazard-rate detection function with Camouflage as a covariate. Sub-area Abundance Estimate Standard Error (se) Coefficient of Variation (cv) 95% Confidence Interval Lower Upper Akia 27,286 5306 0.194 18,276 40,737 Maniitsoq coast 7,205 2323 0.322 3,456 15,024 Maniitsoq inland 6,037 1978 0.328 2,853 12,777 Ujarassuit 8,412 1257 0.149 5,992 11,809 TOTAL 48,941 6390 0.131 37,612 63,682 Table 6. Estimates of density per sub-area (stratum) for Akia-Maniitsoq caribou population in the Central region, March 2019, considering four strata, seven bins and Hazard-rate detection function with Camouflage as a covariate. Sub-area Density Estimate Standard Error (se) Coefficient of Variation (cv) 95% Confidence Interval Lower Upper Akia 6.575 1.279 0.194 4.404 9.816 Maniitsoq coast 2.135 0.688 0.322 1.024 4.452 Maniitsoq inland 3.002 0.984 0.328 1.419 6.354 Ujarassuit 4.126 0.616 0.149 2.939 5.791 TOTAL 4.228 0.552 0.131 3.249 5.502 Caribou flight reaction or lack thereof Like the North region survey of 2018 (Cuyler et al. 2021), the Akia-Maniitsoq caribou population survey of 2019 used digital audio recorders to collect observation data. The digital recorders permitted including in the dataset what, if any, was the behavioral reaction of the caribou group to the helicopter flying a line transect past or over them. Behavior could then be put in relation to group size and distance from the line transect. Behavior was recorded for a total of 683 groups, which involved 2,551 caribou. The size of caribou groups that exhibited movement was significantly different from those that did not move, mean group size 3.98 caribou for moving and 3.00 for non-moving (t Stat = -4.767; two-tailed t-testing: P < 0.0001, t = 1.965, df = 477). Further, non-moving caribou groups were generally 100 m further from line transect flown by the helicopter than caribou groups showing movement (Table 7). There was a significant difference between the mean distance for groups with movement, 199.02 m, and groups lacking movement, 299.42 m (t Stat = 7.577; two-tailed t testing P < 0.0001, t = 1.968, df = 304). Caribou groups reacting to the helicopter fly-by with movement made up 74.8% of all observations for which behavior was reported. Conversely, 25.2% of all caribou groups exhibited little or no movement. A quarter of all caribou groups observed lacked movement. Percentages change when considering the absolute number of caribou 37 involved. There were 2,551 caribou in the 683 groups (Table 7) for which behavior was reported, and of those caribou, 79.8% (n=2,035) moved and 20.2% (n=516) did not move. Table 7. Movement or non-movement of caribou reacting to helicopter fly-by, March 2019. Dataset that included group size, behavior, and distance from track line was n= 683. Parameter Akia-Maniitsoq caribou population Movement Non-Movement p – value Number of groups 511 172 % Group Observations 74.8% 25.2% GROUP SIZE Mean 3.98 3.00 < 0.0001 Confidence Level (95%) 0.2762 0.2975 Standard Error 0.1406 0.1507 Median 3 2 Mode 2 2 Standard deviation 3.1780 1.9765 Sample Variance 10.0997 3.9064 Maximum 25 16 Minimum 1 1 Number of caribou involved 2035 516 DISTANCE 1 Mean 199.02 m 299.42 m < 0.0001 Confidence Level (95%) 13.4252 22.4071 Standard Error 6.8335 11.3515 Median 200 300 Mode 50 300 Standard deviation 154.4726 148.8738 Sample Variance 23861.7858 22163.4027 Maximum 850 750 Minimum 50 50 1 Distance from the line transect flown by helicopter. Among the 441 ‘running’ groups (Runningaway, high speed, parallel, later standing, Table 8), 431 groups exhibited unabated flight, i.e., they never stopped while within view of the helicopter. Rough group composition (calf, adult) was determined for 202 of those groups and recorded calf presence in 85.6%, with remaining 14.4% groups adults only. Considering only the 113 caribou groups whose original position was on the line transect (n = 50) or within 50 m of the line transect (n=63), 110 of those groups (97.3%) never stopped running away from the helicopter. The remaining three groups (2.7%) initially ran away but then stopped, looked at the helicopter and ran no further. A calf was present in two of those three groups. 38 Table 8. Details for movement or non-movement of caribou reacting to helicopter fly-by, March 2019. Dataset of observations, which included caribou group size, behavior, and distance from line transect, was n= 683 groups, which contained n = 2,551 individual caribou. Akia-Maniitsoq caribou population Category Groups (n = 683) % Individuals (n = 2,551) % Exhibiting Movement Running away 405 59.3 1652 64.8 Running away high speed 13 1.9 65 2.5 Walking 38 5.6 124 4.9 Approach* 8 1.2 29 1.1 Confused (jostling/circling) 14 2.0 31 1.2 Running parallel to line transect 13 1.9 55 2.2 Running, later standing looking 10 1.5 49 1.9 Trotting away 9 1.3 25 1.0 Mixed: some moved, others did not1 1 0.1 5 0.2 TOTAL 511 74.8 2035 79.8 Lacking Movement Standing still 156 22.8 476 18.7 Standing, later walking 3 0.4 8 0.3 Lying down 7 1.0 13 0.5 Lying down, later stood up 4 0.6 11 0.4 Some lying, others standing still 1 0.1 6 0.2 Lying down, later walking 1 0.1 2 0.1 TOTAL 172 25.2 516 20.2 *Approach movement (walking, trotting, running) was towards helicopter while looking at helicopter. 1Mixed = different behavior by members within same group, e.g., some running towards others, which stood still and looked at the helicopter. Demographics & recruitment Sex, age, and late-winter calf recruitment data were collected in separate specific efforts that were not part of the line transect DS dataset. On the 12th and 13th of March 2019, and using ca. 8 hours helicopter flight time, we sexed and aged 276 groups of caribou, for a total of 1257 animals, in the Akia-Maniitsoq caribou population (Table 9). Cows were ca. 50% of the population, followed by bulls (age >1-year) at ca. 24% and calves at 26%. The calf percentage may be considered artificially high owing to the large number of calves lacking their dam, i.e., orphan calves. While obtaining demographics for Akia-Maniitsoq caribou population, we observed a loose aggregation of numerous caribou on the lowland flats (elevation < 200 m) at the north end of the Narssarssuaq Valley (Fig. 17). Most of these groups of caribou were not sexed and aged because they were too large, e.g., largest > 50 animals. The high group size combined with the confusion of constant jostling change of positions among the fleeing 39 caribou, made it difficult to sex and age all group members. Further, groups fleeing the helicopter often mixed or split apart to regroup with yet other groups before sex/age determination was complete. Where group separation and size permitted, demographics were obtainable (Fig. 18). Table 9. Demographics for Akia-Maniitsoq caribou population, Central region, March 2019. Parameter Akia-Maniitsoq caribou population Number of groups observed 276 GROUP SIZE Mean 4.55 Confidence Interval (95%) 0.3896 Standard Error 0.1979 Standard Deviation 3.29 Sample Variance 10.8079 Median group size 4 Mode group size 3 Maximum group size* 24 Minimum group size 1 DEMOGRAPHIC All data 83 orphan calves removed Total sexed & aged (n) 1257 100 % 1174 100 % Cow (age >1 year) 624 49.6 % 624 53.15 % Calves from previous spring 326 25.9 % 243 20.70 % (148 females) 11.77 % - - (171 males) 13.60 % - - (7 unknown) 0.56 % - - Bull (age >1 year) 307 24.4 % 307 26.15 % (145 adults age >3) 11.54 % (145 adults age >3) 12.35 % (162 juveniles 1< age <3) 12.89 % (162 juveniles 1< age <3) 13.80 % Recruitment (calves / 100 cows) 52.24 38.94 Sex ratio (Bull age >3 year/Cow) 0.23 0.23 Sex ratio (Bull age >1 year/Cow) 0.49 0.49 * This is maximum for only those groups that were sexed and aged. It was not possible to sex and age the individuals in the exceedingly large groups on the flats of the Narssarssuaq Valley (Fig. 17). 40 Figure 17. Unusually high numbers of caribou, often in exceptionally large groups were observed on the low elevation (< 200 m) flats of the valley, Narssarssuaq, Central region, early March 2019: Left are 45 caribou, above right 33 and below 14. Sex and age determination was only possible on the latter. Figure 18. These 17 caribou could be sexed and aged owing to group separation and relatively small group sizes, i.e., six (left), nine (foreground) and two (in distant background). 41 Calf lacking their dam The demographics data collection included a total of 326 calves (Table 9). Only 243 calves (74.5% of all calves observed) were in the company of cows, while 83 calves (25.5% of all calves observed) appeared to be orphans, i.e., lacking a dam/cow. Of the 83 calves, 52 (16% of all calves observed) were designated ‘true’ orphans. Primarily because they were solitary individuals (n = 4) or in the company of other calves (n = 30 calves, 10 groups), always with no older caribou nearby. Orphan pairs were observed five times, and triplet once. Four orphans together were observed three times. The maximum orphan group size was five calves and was observed once. The ratio of orphan females to males was almost 50:50, as there were 16 female and 18 male orphans. Secondly, ‘true’ orphans were also considered those in the company of bull only groups (n = 18, 14 groups). A single orphan calf was present in 11 of the bull groups. There were six orphans present in juveniles-only bull groups (n=5), eight present in adult bull groups (n=5) and four in mixed juvenile plus adult bull groups (n=4). The remaining 31 calves, out of the above 83, were designated ‘possible’ orphans because these were ‘extra’ calves observed in the company of cow-calf pairs, i.e., for 20 cow-calf groups (n = 117 caribou; 43 cows, 74 calves) the number of calves exceeded the number of cows. The group composition for half of these groups consisted of a single cow followed by two calves. Five cow-calf groups included multiple pairs and just one ‘extra’ calf. Two cow-calf groups included multiple pairs and two ‘extra’ calves. A further two cow-calf groups included multiple pairs and three ‘extra’ calves. Finally, one cow-calf group with multiple pairs included six ‘extra’ calves. The result was a 31 (9.5%) ‘possible’ orphan calves. Although only one orphan was in visibly poor body condition, owing to possible foraging difficulties we suspect that the 83 orphan calves have less chance of surviving their first winter than calves with their dams. If these orphan calves are removed from the dataset, then the resulting demographics becomes cows ca. 53%, bulls (age > 1-year) ca. 26% and calves ca. 21%, with calf recruitment reduced to ca. 39 calves per 100 cows (Table 9). Group size & composition The sex and age composition of caribou groups may have influenced group size. The results were almost identical for groups of all bulls (juveniles and adults) and those containing only adult bulls (age > 3-years) (Table 10). The groups of juvenile bulls (1-year < age < 3-year) had the lowest mean group size, 1.60 animals. The highest mean group size 48 Caribou detection Incomplete or patchy snow cover, substrates (including grass, bushes, ground) poking or showing through a thin snow layer, rocky terrain, light/shadow conditions, and occasionally fog, whether alone or in combinations are normal during aerial survey in West Greenland (Cuyler et al. 2005, 2007, 2011, 2016, 2021, 2023). Conditions were similar in 2019 and as usual made the well camouflaged caribou a challenge to detect. Stationary caribou exacerbated the detection difficulty. The almost west-east orientation of the line transects used in 2019 meant that in sunny conditions solar glare in the eyes of the observer on the south-facing side of the helicopter required polarized sunglasses. Despite the use of polarized sunglasses, detectability of caribou may have been reduced (Fig. 11). We recommend that future DS surveys combine all covariates contributing to the caribou becoming camouflaged into the terrain (e.g., Figs. 9, 10, 11), into a single index for camouflage (e.g., extreme, high, medium, low, none). Combined, we expect this will improve how caribou “invisibility” interacts with Key Functions to model the detection function. Regardless, given robust DS data, the influence of covariates on detectability is small and unlikely to significantly alter final abundance and density estimates. Caribou behaviour - flight response of caribou groups It is reasonable to expect that any survey for caribou would have some proportion of nonmoving caribou present in the surveyed area of the line transects. Since flight responses by the caribou may influence whether an observer detects them, in 2019 the line transect data included whether the helicopter fly-by elicited a flight movement response from the caribou group or whether they were stationary. About 75% of the observed caribou groups exhibited movement in response to the helicopter fly-by, while about 25% did not, which included some on the track line. This large proportion of stationary caribou groups underlines the importance of skilled observers able to detect caribou despite the extraordinary degree caribou are camouflaged for typical background conditions. From experience we know that observer skill can only succeed in detecting camouflaged caribou present if the helicopter is flying ‘low & slow’, as the current study did. Because stationary animals make up almost 1/4 of all caribou groups observed and include some on the line transect itself (where for DS analyses all caribou present must be detected), detecting non-moving caribou is essential to avoid underestimating population size. 49 Caribou groups that exhibited movement in response to the helicopter fly-by had a mean group size of 4. This was significantly (p < 0.0001) larger than the non-moving caribou mean group size of 3. Also, groups that exhibited movement were closer to the helicopter than groups that did not move (Table 9). As with previous aerial surveys, this attests to exceptional observer ability to detect caribou despite behavior displayed. Explanations for lack of movement among caribou further away from the helicopter would include that at greater distances the helicopter may be perceived by the caribou as less threatening. Additionally, group composition may be involved. For caribou groups that ran from the helicopter, most exhibited unabated flight reaction, i.e., they never stopped while we could see them. This was specifically true for those groups whose original position was the track line itself. Calf presence appears to be an important factor determining whether unabated flight occurs, since ca. 86% of unabated flight involved groups with calves. Demographics In early March 2019, the Akia-Maniitsoq caribou population’s demographics suggested a composition that provides potential for further growth in abundance, albeit notwithstanding future catastrophic stochastic events, including pathogen outbreaks and extreme weather. Ratio of bulls to cows appeared to be recovering from the 2001 to 2010 bull decline (Fig. 23). The 2019 calf percentage and recruitment (number calves per 100 cows) were the highest since helicopter surveys began in 2001 (Fig. 24). The 2019 values remained high even after removing the surprisingly high number of orphan calves (n = 83). Further, the 2019 percentage of calves was similar to the high 1998 level (Table 1) obtained from ground survey. The 2019 level of calf recruitment strongly suggests the possibility of future population growth (Bergerud et al. 2008). A discussion of Akia-Maniitsoq’s demographics in 2019 would not be complete without discussing that ¼ of all observed calves were orphans. Although the late winter calf recruitment was initially ca. 52 calves per 100 cows, the high incidence of orphan calves (n=83), ca. 25% all calves observed, brought calf recruitment value down to 34 calves per 100 cows. It is not unreasonable to assume that to create those orphans a similar number of cows died before the 2019 survey occurred. This suggests two possibilities. First, that some pathogen increased mortality among cows but not their calves, and not among bulls. 50 Common sense makes this unlikely since caribou pathogens are not so selective. The second possibility is that hunters harvesting from the Akia-Maniitsoq caribou population predominantly selected only the cow from cow-calf pairs. Figure 23. Past and present late winter bull to cow ratios for the Akia-Maniitsoq caribou population. Bull classification includes both juveniles, 1-year < age < 3-year, and adults, age > 3 years. Figure 24. Past and present late winter calf (age 10-month) recruitment (number calves per 100 cows) and calf percentage for the Akia-Maniitsoq caribou population. Column 2019a included all calves observed, while 2019b removed all orphan calves (n = 83). 58 45 38 49 100 100 100 100 0 20 40 60 80 100 120 2001 2005 2010 2019 Number of caribou Year Bull Cows 31 24 23,2 52,24 38,94 17 14 14,4 25,9 20,7 0 10 20 30 40 50 60 2001 2005 2010 2019a 2019b Calf recruitment or percentage Year Calves / 100 Cows Percentage 51 Elevation Albeit elevation data was only approximate given the limitations of the GPS device and mismatch between helicopter and caribou positions, during the 2019 aerial survey, caribou of the Akia-Maniitsoq population were observed at elevations of mean 361 m ±246 with the mode being 156 m. This use of relatively low elevations in late winter (March) is supported by 2008-2010 GPS telemetry data from for Akia-Maniitsoq cows, mean elevation ca. 200 m (Cuyler et al. 2017). This may reflect a caribou preference for low elevations or just that the Central region has more available lowland habitat than is usual in West Greenland, i.e., Akia sub-area is mostly < 200 m elevation (Figs. 1, 2). The caribou are likely to prefer low elevations because of a higher quality and quantity of available winter forage, which would be expected for lowland elevations at the latitude of the Central region (Körner 2007). Use of low elevations may also suggest that winter disturbance by humans on the Akia-Maniitsoq population is minimal. The winter 2019 caribou hunting season for the Central region was 01-15 February, ending 2-weeks before the 2019 survey began. The similar late winter elevation use across years suggests Akia-Maniitsoq caribou are not using high elevations to avoid hunters. Meanwhile, the Akia-Maniitsoq caribou population is not readily accessible to humans, regardless of season because the Central region can only be reached by boat. Specifically for winter, snowmobile use is prohibited over the entire region, rendering these vehicles illegal for winter hunting season transport or recreational use. Nevertheless, and despite being illegal, snowmobile use on the Akia (Nordlandet) area of the Central region was observed in March 2019 (Fig. 20). The same was observed during the previous survey of 2010, when snowmobile use was associated with high harvest number of caribou (Cuyler et al. 2011). In both 2010 and 2019, high numbers of caribou in unusually large groups were observed on the low elevation (< 200 m) flats in the northern portion of the Narssarssuaq Valley, Central region (Fig. 17). As suggested earlier (Cuyler et al. 2011), this caribou aggregation may be the result of animals avoiding the snowmobile activity in the extensive lowlands of Akia (Nordlandet). Unfortunately for the caribou, Akia (Nordlandet) is a prime winter foraging habitat (Cuyler et al. 2017), which is opposite Greenland’s capital city, Nuuk, and illegal snowmobile use may be common. The abnormally large aggregation of caribou in the Narssarssuaq Valley suggests that current levels of snowmobile disturbance are sufficient to cause avoidance behavior, albeit to another lowland area rather than to higher elevations. 52 All the above may explain why the Akia-Maniitsoq caribou population averaged elevations ca. 300 m lower than the Ameralik caribou population, which were also surveyed in March 2019 (Cuyler et al. 2023). There should be less access by humans to the Akia-Maniitsoq caribou population, and their Narssarssuaq Valley appears to provide them with a remote and large lowland area where they can avoid winter hunters or snowmobile recreation. The current remoteness of the Narssarssuaq Valley will change if human access increases. Until then, elevation use by the Akia-Maniitsoq population will not likely be influenced by human disturbance. Late-winter antler possession Bulls As expected for caribou populations, adult (age > 3-years) bulls from the Akia-Maniitsoq population lacked antlers in March, while most, 87%, juvenile bulls retained one or both their antlers from the previous autumn. Thus, late-winter antler possession among bulls age > 1 year of the Akia-Maniitsoq population is similar to bulls in both the Kangerlussuaq-Sisimiut and Ameralik caribou populations (Cuyler et al. 2021, 2023). However, in those same three populations, antler possession varied for late-winter male calves (age < 1-year). Antler possession in male calves was poor, 40%, in the AkiaManiitsoq population. In contrast, antler possession in male calves was common and similar in male calves for the Kangerlussuaq-Sisimiut and Ameralik populations at 86.2% and 87.5%, respectively (Cuyler et al. 2021, 2023). Cows Although among wild caribou populations in North America, 98% of cows have antlers in late winter (Kelsall 1968, Reimers 1993, Bergerud et al. 2008), antler possession among cows in caribou populations of West Greenland is highly variable and often exhibits a high percentage of polled cows, i.e., no antlers (Thing et al. 1986, Cuyler et al. 2002, 2021). In North America, decline in the percentage of antlered cows has been attributed to overgrazed range, because that is a major factor causing poor cow body condition, which precludes antler growth (Gaare & Skogland 1980, Reimers 1983, Thing et al. 1986, Bergerud et al. 2008). In West Greenland, range condition is not the major factor influencing the number of polled cows (Cuyler et al. 2021). Albeit polled (antlers lacking) cows are common among caribou populations in West Greenland, the Akia-Maniitsoq population far exceeds the norm, i.e., just 32% of cows had antlers in late winter 2019. Instead, most cows are polled (68%). This was also reflected in antler possession or absence among female calves, and to a lesser extent even male calves. 53 The 2019 results for poor antler possession among Akia-Maniitsoq cows are supported by 1998 observations of 19% antlered Akia-Maniitsoq cows (Cuyler unpublished). Lack of antlers in Akia-Maniitsoq cows is in sharp contrast to the situation in North America. There, antler possession is assumed to confer dominance among large aggregations of caribou that must feed by cratering through deep snow (Kelsall 1968, Reimers 1993, Bergerud et al. 2008). Nevertheless, Bergerud et al. (2008) presented evidence that a high percentage of polled cows would be expected in populations that had small group sizes and little dependence on cratering. Akia-Maniitsoq caribou do have small group sizes, mean 3.75 ± 3.0 standard deviation. Further, xeric conditions occur in inland areas and can result in negligible snow depth covering food resources in the Ujarassuit sub-area and anywhere adjacent the Greenland Ice Cap (Appendix 1, Figs. 31-34, 37-46). Thus, the percentage of polled Akia-Maniitsoq cows may be unrelated to range condition, but rather reflect a reduced need for the dominance conferred by antlers. With only 32% of cows possessing antlers in 2019, the Akia-Maniitsoq population is in sharp contrast to either of the other two large caribou populations (KangerlussuaqSisimiut and Ameralik) that have been surveyed in West Greenland. For example, for the same survey year, 2019, 92% of Ameralik cows had antlers (Cuyler et al. 2023), and although not as great a difference, in 2018, antlered Kangerlussuaq-Sisimiut cows were 54% (Cuyler et al. 2021). It is notable that despite cows being predominantly polled in March 2019, the AkiaManiitsoq population had an excellent percentage of calves and calf recruitment. High calf numbers contradict any assumption that cows are in poor body condition (e.g., owing to overgrazed range) rather the opposite is indicated. We suggest that an inherited trait may be the major influence causing predominantly polled Akia-Maniitsoq cows. Snowmobile use Akia-Nordlandet Akia-Nordlandet is a peninsula of rugged lowlands just opposite Greenland’s capital city, Nuuk, on the north side of Nuuk (Godthåb) fjord. There are no permanent human habitations, but there are numerous recreational summer houses/cabins. Despite the continued prohibition on snowmobile use for this area, recent snowmobile activity was observed. Snowmobile use was first observed in March 1997 (Cuyler unpublished) and was extensive during the 2010 caribou winter harvest (Cuyler et al. 2011). 54 Other species observed During the early March 2019 survey multiple species other than caribou were observed. Ptarmigan and hares were relatively abundant and appeared to support a healthy arctic fox population. Gyr falcons, sea eagles and snowy owls were also observed. Although since ca. 2000 local knowledge has reported muskox presence in the Central region of West Greenland, like all previous aerial surveys of this region, no muskoxen were observed during the early March 2019 survey. There were two aerial surveys conducted in early March 2019, i.e., this study’s Central region and another in the South region (Cuyler et al. 2023). Although the surveys occurred almost simultaneously, each of the observed species (other than caribou) were more numerous in the Central region than in the South (Cuyler et al. 2023). For example, there were three times more ptarmigan in the Central region and almost twice as many hares. Not unexpectedly then regarding predators, there were four times the number of arctic foxes and several avian predators in the Central region. Still, the Central region’s area of survey effort (Table 2) was 2.4 times greater than for the South region (Cuyler et al. 2023). This partially explains the greater number of observations. However, it remains that the Central region had more ptarmigan and foxes as well as avian predators. Late winter 2019 habitat conditions in the Central region supported a greater abundance of several species than conditions in the South region. Acknowledgements This project was financed primarily by the Government of Greenland and by Pinngortitaleriffik – Greenland Institute for Natural Resources, Nuuk Greenland. Grateful thanks go to Air Greenland Charter and their helicopter pilot Stig Erick for his safe flying. For providing experienced observers, excellent at spotting caribou despite poor detection conditions, thanks are also due the Greenland Association of Professional Hunters (KNAPK) and the Greenland Fisheries and License Control (GFLK). We also thank Rikke Guldborg Hansen and Lars Witting for constructive review of the manuscript. 55 Literature cited Aronsson M., Heiðmarsson S., Jóhannesdóttir H., Barry T., Braa J., Burns C.T, Coulson S.J., Cuyler C., Falk K., Helgason H., Lárusson K.F., Lawler J.P., Kulmala P., MacNearney D., Oberndorfer E., Ravolainen V., Schmidt N.M., Soloviev M., Coon C. & Christensen T. 2021. State of the Arctic terrestrial biodiversity report. Conservation of Arctic Flora and Fauna International Secretariat, Akureyrik, Iceland.ISBN 9789935-431-94-3. 123 pp. Bergerud A.T. 1967. Management of Labrador caribou. J. Wildl. Manage. 31:621-642. Bergerud A.T. 1971. The population dynamics of Newfoundland caribou. Wildl. Monogr. 25: 55 pp. Bergerud A.T. 1980. A review of the population dynamics of caribou and wild reindeer in North America. Proc. 2nd Reindeer/Caribou Symp., 556-581. Trondheim: Direktoratet for Vilt og Ferskvannsfisk. Bergerud A.T., Luttich S.N. & Camps L. 2008. The return of caribou to Ungava.. McGill-Queen’s University Press, Montreal & Kingston, London, Ithaca. 586 pp. Brewer M.J., Butler A. & Cooksley S.L. 2016. The relative perperformance AIC, AICc and BIC in the presence of unobserved heterogeneity. Methods in Ecology and Evolution 7(6): 679–692. Buckland S.T., Anderson D.R., Burnham K.P. & Laake J.L. 1993. Distance Sampling: Estimating Abundance of Biological Populations. Springer. Buckland S.T., Anderson D.R., Burnham K.P., Laake J.L., Borchers D.L. & Thomas L. 2001. Introduction to Distance Sampling. Oxford: Oxford University Press. Buckland S.T., Anderson D.R., Burnham K.P., Laake J.L., Borchers D.L. & Thomas L. 2004. Advanced Distance Sampling: Estimating abundance of biological populations. Oxford University Press. Buckland S.T., Rexstad E.A., Marques T.A. & Oedekoven C.S. 2015. Distance Sampling: Methods and Applications. Springer. CARMA (Circum Arctic Rangifer Monitoring & Assessment network). www.carma.caff.is/ Correia I.J.F. 2020. Estimating caribou abundance in West Greenland using distance sampling methods. MSc. Thesis. University of Lisbon, Portugal. 63 pp. Couturier S., Dale A., Wood B. & Snook J. 2018. Results of a Spring 2017 aerial survey of the Torngat Mountains Caribou Herd. Technical report, Torngat Wildlife, Plants and Fisheries Secretariat. Cuyler C. 2020. Biologisk rådgivning for rensdyrog moskusoksefangst i 2020. Pinngortitaleriffik – Greenland Institute of Natural Resources harvest advice document to the Government of Greenland (GN 40-59/40-00-02-42), 4 pp. Supplementary materials, 26 pp. Cuyler C., Marques T.A., Correia I.J.F., Afonso B.C., Jensen A., Hegelund P. & Wagnholt J. 2021. 2018 status Kangerlussuaq-Sisimiut caribou, West Greenland. Greenland Institute of Natural Resources. Technical Report No. 117. 79 pp. Cuyler, C., Marques, T.A., Correia, I.J.F., Jensen, A., Mølgaard, H. & Wagnholt, J. 2023. 2019 Status of Ameralik caribou population, South region, West Greenland. Pinngortitaleriffik – Greenland Institute of Natural Resources. Technical Report No. 125. 109 pp. Cuyler C., Nagy J. & Zinglersen K. 2017. Seasonal movement and activity of Akia-Maniitsoq caribou cows in West Greenland as determined by satellite. Pinngortitaleriffik – Greenland Institute of Natural Resources. Technical Report No. 99. 94 pp. 56 Cuyler C., Nymand J., Jensen A. & Mølgaard H.S. 2016. 2012 status of two West Greenland caribou populations, 1) Ameralik, 2) Qeqertarsuatsiaat. Greenland Institute of Natural Resources Technical Report No. 98, 179 pp. Cuyler L.C., Rosing M., Egede J., Heinrich R. & Mølgaard H. 2005. Status of two West Greenland caribou populations; 1) Akia-Maniitsoq, 2) Kangerlussuaq-Sisimiut. Pinngortitaleriffik – Greenland Institute of Natural Resources. Technical Report No. 61. Part I-II, 64+44 pp. Cuyler C., Rosing M., Heinrich R., Egede J. & Mathæussen L. 2007. Status of two West Greenland caribou populations 2006, 1) Ameralik, 2) Qeqertarsuatsiaat. Greenland Institute of Natural Resources. Technical report No. 67. 143 pp. (Part I: 1-74; Part II: 75-143). Cuyler C., Rosing M., Linnell J.D.C., Loison A., Ingerslev T. & Landa A. 2002. Status of the KangerlussuaqSisimiut caribou population (Rangifer tarandus groenlandicus) in 2000, West Greenland. Pinngortitaleriffik – Greenland Institute of Natural Resources. Technical Report No. 42. 52 pp. Cuyler C., Rosing M., Linnell J.D.C., Lund P.M., Jordhøy P., Loison A. & Landa A. 2003. Status of 3 West Greenland caribou populations; 1) Akia-Maniitsoq, 2) Ameralik & 3) Qeqertarsuatsiaat. Pinngortitaleriffik – Greenland Institute of Natural Resources. Technical Report No. 46. 74 pp. Cuyler C., Rosing M., Mølgaard H., Heinrich R. & Raundrup K. 2011. Status of two west Greenland caribou populations 2010; 1) Kangerlussuaq-Sisimiut & 2) Akia-Maniitsoq. Greenland Institute of Natural Resources. Technical Report No. 78. 158 pp. (Part I: 1-86; Part II: 87-158). Cuyler L.C. & Østergaard J. 2005. Fertility in two West Greenland caribou populations 1996/97: Potential for rapid growth. Wildlife Biology. 11(3): 221-227. Gaare E. & Skogland T. 1980. Lichen-reindeer interaction in a simple case model. Proc. 2nd International Reindeer/Caribou symposium 47-56. Trondheim: Direktoratet for vilt of ferskvannsfisk. Gibbons J.D. & Chakraborti S. 2011. Nonparametric Statistical Inferencing. Chapman & Hall. Gates C.C., Adamczewski J. & Mulders R. 1986. Population dynamics, winter ecology and social organization of Coats Island caribou. Arctic. 39(3): 216-222. Heard D.C. & Ouellet J.P. 1994. Dynamics of an introduced caribou population. Arctic. 47(1): 88-95. Jepsen B.I, Siegismund H.R. & Fredholm M. 2002. Population genetics of the native caribou (Rangifer tarandus groenlandicus) and the semi-domestic reindeer (Rangifer tarandus tarandus) in Southwestern Greenland: Evidence of introgression. Conservation Genetics. 3: 401-409. Kelsall L.B. 1968. The Caribou. Ottawa: Queen’s Printer. Kingsley M.C.S. & Cuyler C. 2002. Caribou harvest 2002: advisory document. Pinngortitaleriffik – Greenland Institute of Natural Resources, Nuuk. 12 pp. Körner C. 2007. The use of ‘altitude’ in ecological research. Trends Ecol. Evol. 22: 569–574. Linnell J.D.C., Cuyler C., Loison A., Lund P.M., Motzfeldt K.G., Ingerslev T. & Landa A. 2000. The scientific basic for managing the sustainable harvest of caribou and muskoxen in Greenland for the 21st Century: an evaluation and agenda. Technical Report 34, Greenland Institute of Natural Resources, Pinngortitaleriffik. Loison A., Cuyler C., Linnell J.D.C. & Landa A. 2000. The caribou harvest in West Greenland, 1995-1998. Pinngortitaleriffik – Greenland Institute of Natural Resources. Technical Report No. 28. 33 pp. 57 Miller D.L., Rexstad E., Thomas L., Marshall L. & Laake J. L. 2016. Distance Sampling in R. Journal of Statistical Software 89(1): 1–28. Marques T.A., Buckland S.T., Borchers D.L., Rexstad E. & Thomas L. 2011. Distance Sampling. International Encyclopedia of Statistical Science, 1: 398–400. Marques T.A., Thomas L., Fancy S.G. & Buckland S.T. 2007. Improving estimates of bird density using multiple covariate distance sampling. The Auk 124(4): 1229–1243. Naalakkersuisut. 2019. Press release 11 January 2019: Fangst af muskoxen og rensdyr vinter 2019 (Hunting of muskoxen and caribou winter 2019) Bilag 4. https://naalakkersuisut.gl/da/Naalakkersuisut/Nyheder Pollock K.H., Nichols J.D., Brownie C. & Hines J.E. 1990. Statistical inference for capture-recapture experiments. Wildlife Monographs 107: 3–97. Poole K.G., Cuyler C. & Nymand J. 2013. Evaluation of caribou Rangifer tarandus groenlandicus survey methodology in West Greenland. Wildlife Biology 19: 225–239. Reimers E. 1983. Growth rates and body size differences in Rangifer, a study of causes and effect. Rangifer 3: 3-15. Reimers E. 1993. Antlerless females among reindeer and caribou. Can. J. Zool. 71: 319-325. Skoog R.O. 1968. Ecology of the caribou (Rangifer tarandus granti) in Alaska. Ph.D. Thesis. University of California at Berkeley, 699 pp. Thing H., Olesen C.R. & Aastrup P. 1986. Antler possession by west Greenland female caribou in relation to population characteristics. Rangifer, Special Issue 1: 297-304. Thomas L., Buckland S.T., Burnham K.P., Anderson D.R., Laake, J.L., Borchers D.L. & Strindberg S. 2002. Distance Sampling. Encyclopedia of Environmetrics 1: 544–552. Thomas L., Buckland S.T., Rexstad E.A, Laake J.L., Strindberg S., Hedley S.L., Bishop J.R.B., Marques T.A. & Burnham K.P. 2010. Distance software: design and analysis of distance sampling surveys for estimating population size. J. Appl. Ecol. 47(1): 5-14. 64 Central region aerial survey conditions, March 2019 Figure 37. Ujarassuit sub-area, illustrating typical conditions for terrain, snow, and sunlight present for most of the line transects. Photo A. Jensen. Figure 38. Ujarassuit sub-area on the one day with sunshine, view west from the east end of line transect 27, illustrating rugged highland terrain and lack of snow. Photo C. Cuyler. 65 Central region aerial survey conditions, March 2019 Figure 39. Ujarassuit sub-area, east end of line transect 35, view to west illustrating snow depth and flat light. Photo A. Jensen. Figure 40. Ujarassuit sub-area, illustrating typical terrain, snow, and sunlight conditions for line transects 36, 35 and 33. Note the rather xeric landscape, view to the west with the large lake, Taserssuaq, in background. Photo A. Jensen. 66 Central region aerial survey conditions, March 2019 Figure 41. Ujarassuit sub-area terrain, snow, and flat light conditions at the east end of line transect 36, view to the west. Photo A. Jensen. Figure 42. Ujarassuit sub-area, illustrating typical terrain, snow, and sunlight conditions in a rather xeric landscape, view northwest from line transect 33 towards the large lake, Taserssuaq, in far background. Photo A. Jensen. 67 Central region aerial survey conditions, March 2019 Figure 43. Ujarassuit sub-area, illustrating rugged highland terrain, snow depths and sunlight typical of line transect 30. Similar conditions existed for line transects 27, 29, 31 and 28. This view southwest from line transect 30 towards Innajuattoq (Bird Mountain) in left background. Photo A. Jensen. Figure 44. Ujarassuit sub-area, Ivisartoq highlands, xeric conditions of east end line transect 32, view N across the many frozen lakes and edge of Greenland Ice Cap in foreground. Photo A. Jensen. 68 Central region aerial survey conditions, March 2019 Figure 45. Ujarassuit sub-area, Ivisartoq highlands view east towards small bay in the Ujarassuit Kangerluat fjord. Thin snow layer with ground showing through. Photo C. Cuyler. Figure 46. Ujarassuit sub-area, Ivisartoq highlands illustrating almost non-existent snow layer, view west towards Innajuattoq (Bird Mountain) in far-left background. Photo A. Jensen. 69 Appendix 2 Place names for the Central region Figure 47. Place names used regarding Central region (ca. 64°–66°N; 50°–53°W), which is inhabited by the AkiaManiitsoq caribou population. 70 Appendix 3 Statistical methods behind Distance Sampling This appendix presents the basic building blocks and reasoning behind Distance Sampling (DS) methods, followed by some details. This summary of statistical methods is from Correia (2020). Fundamental concepts Before entering into the detailed theory behind the DS methodology, we present a simpler design, which is quadrat or plot sampling (Buckland et al. 2001; Marques, 2009). In plot sampling, a region of interest with total area 𝐴, is divided into small plots of area 𝑎 𝑝𝑙𝑜𝑡 (Fig. 48). Some of these small plots are randomly chosen for sampling and the total number of individuals within these, 𝑛𝑝𝑙𝑜𝑡, is recorded. Figure 48. Plot sampling grid example of total area 𝐴 divided into smaller plots of area 𝑎 𝑝𝑙𝑜𝑡 . The density within each plot, 𝐷𝑝𝑙𝑜𝑡, is the number of individuals per unit area for the respective plot so, by definition, it is given by Equation (1) where 𝑎 is the total area sampled within 𝐴 . (i.e., 𝑎 = 4 ⋅ 𝑎 𝑝𝑙𝑜𝑡 = 4𝑘𝑚 2 for Fig. 48) Since a random design was used, the density is a representative estimate, by design, for the total area 𝐴 . Hence, an estimate for the abundance, 𝑁  , can b e obtained b y simply multiplying 𝐷  plot by the total a rea 𝐴 , Equation (2) 71 𝑃 The DS methodology is an extension of quadrat-based sampling methods. The detail that creates the bridge from one methodology to the other is the fact that the method described above assumes that every individual of interest is detected (Miller et al. 2016). Frequently, this assumption cannot be met, specifically if among the individuals of interest there are animals impossible to observe owing to low sightability. Several factors cause low sightability, including topographical barriers, weather conditions, ground surface conditions and many others related to observer training and survey design. The proportion of individuals that were not detected can be estimated using the detection function fitted to the observed distances (Thomas et al. 2002). Once this proportion is estimated, it can be considered to obtain more accurate estimates and then, an extrapolation for a wider region can be done similarly as shown in Equation (2). In DS, this proportion of detected objects in the area 𝑎 is defined as the probability of detection, 𝑃𝑎. Therefore, a density estimate can be obtained as per Equation (1) by adjusting 𝑛𝑝𝑙𝑜𝑡 by 𝑃𝑎, i.e., by correcting the detections for those that were missed. Since the latter cannot be known, in general, an estimate must be also obtained, thus: Equation (3) where 𝑃 󰆹 a is an estimate of 𝑃 𝑎 obtained from the distance data, and 𝑎 is the area of the sampled region. Usually 𝑎 = 2𝑤𝐿, with 𝑤 as the truncation distance, for both sides of the track line, and the total transect length 𝐿 = ∑𝑙 𝑗 𝑘 𝑗=1 , where 𝑙 is the length of transect 𝑗. Abundance can be determined using a reasoning analogous to that above (Equation 2). The truncation distance is defined as the distance beyond which distances are not recorded. This can be defined in the field or at the analysis stage. The coefficient of variation of 𝐷 , c 𝑣 (𝐷 ), is related with t w o random comp onen ts referred ab ov e, encoun ter rate ( 𝑛 𝑝𝑙𝑜𝑡 /𝐿 ), and 𝑃 󰆹 a , plus a third one that is the estimate of the exp ected size of detected clusters ( 𝐸 󰆹 ( 𝑠) ). Assuming independence b et w een these, the former is given by Equation (4) An approximation of the standard error of 𝐷 , 𝑠𝑒( 𝐷  ), is defined as Equation (5) 72 Once these are obtained, an approximate 100(1 − 𝛼)% confidence interval (CI) can be determined by Equation (6) Where is the quantile of the N(0,1) distribution 1.96 for a 95% confidence interval). However, the distribution of the 𝐷 is positively skewed, thus an interval assuming that 𝐷  is log-normally distributed has better coverage. According with Buckland et al. (2015), a 100(1-alpha)% confidence interval can be given by Equation (7) where Equation (8) and Equation (9) For further details see Buckland et al. (2001) and Buckland et al. (2015). Probability of detection Given the above, the probability of detecting an object, giving that it is within the area co vered b y the transects, 𝑃  a , needs to b e estimated. F or this pro ject, the ob ject of in terest consists in caribou groups. To illustrate the importance of this probability, consider that an observer walks across a large patch of tundra and detects 8 caribou (Fig. 49). While discussing with the local biologist, and considering the biologist’s experience, he/she will state that, on average, only one third of all caribou present are detected (i.e. , 𝑃  a = 1/3) meaning that probably there were around 24 caribou within that patch of tundra and 16 have been missed. That is where DS is useful, since it allows a rigorous framework for the estimation of P a and then an estimate of abundance can be obtained as shown in Equation (3). 73 Figure 49. Example of a patch of tundra with the transect in the middle. Blue dots represent eight observed caribou, while orange dots represent the 16 undetected ones. The lines perpendicular to the transect represent the recorded distances. Distance Sampling methods The detection function, 𝑔(𝑦), describes the probability of detecting an object of interest given that it is at a distance 𝑦, from the track line (also known as 0-line), thus being a nonincreasing function of 𝑦 (Buckland et al. 2015). For line transects, 𝑦 is the perpendicular distance from the 0-line to the detected object. Within DS methods, the probability of detection is explained recurring to these observed distances (Buckland et al. 2001). Sometimes covariates may be added to explain their relationship with the detection probability. In this situation, we are within the Multiple Covariate Distance Sampling (MCDS) framework (Buckland et al. 2001). Conventional Distance Sampling Conventional Distance Sampling (CDS) occurs when no additional covariates are added to the mo del. Once the detection function is estimated, 𝑃  a can b e obtained via the follo wing equation Equation (10) where 𝜋(𝑦)= 1 𝜔 and, therefore, used to estimate density using Equation (3). For 𝑔(𝑦) it is also specified a flexible semi-parametric model, composed by a key function and some 80 Mises test and the 𝜒2 Goodness-of-Fit test (GOF test). The likelihood ratio test can also be used but, since it is only applicable for nested models, AIC is the recommended method (Marques et al. 2007). A proper model should be simple with an adequate fit without overfitting the data. Akaike Information Criterion The relative fit of alternative models may be evaluated recurring to AIC, or AICc, in case of small samples, providing a small sample bias correction (Buckland et al. 2001). These criteria can be determined as follows Equation (15) Equation (16) where ℒ is the likelihood function, 𝑞 is the number of estimated parameters in the model, and 𝑛 is the sample size. This measure provides a trade-off between bias and variance. AIC includes two terms, one related with the fitted model, and the other working as a penalty considering the excess of parameters in the model (Brewer et al. 2016). Kolmogorov-Smirnov test The Kolmogorov–Smirnov test is one of the tests that can be applied to the detection function to assess model fit (Buckland et al. 2004). This test is only applicable for continuous data, being preferable to the 𝜒2 GOF test for MCDS methods. Considering the cumulative distribution function (c.d.f.) 𝐹 (𝑥) = 𝑃 (𝑋 ≤ 𝑥) and the empirical c.d.f. (e.d.f.) 𝑆(𝑥) , the null hypothesis to be tested is 𝐻 0 ∶ 𝐹 (𝑥) = 𝐹 0 (𝑥), ∀𝑥 . The alternative hypothesis states that both functions differ for at least some value of 𝑥 . In practice, 𝐹 (𝑥) is replaced by its estimate, and 𝐻0 states that the assumed model is the true model for the data ( Buckland et al. 2004 ). The largest absolute difference b et w een 𝐹 󰆹 (𝑥) and 𝑆 (𝑥) , denoted 𝐷𝑛, is the test statistic (Gibbons and Chakraborti 2011). The corresponding 𝑝-value can be approximated by Equation (17) 81 Cramér-von Mises test Similarly to the Kolmogorov-Smirnov test, the Cramér-von Mises test shares the same null hypothesis and basis on differences between c.d.f. and e.d.f.. However, instead of considering only the largest difference between the two functions, this test is based on their entire range (Buckland et al. 2004). The test statistic can be given by Equation (18) Chi-square Goodness-of-Fit test The 𝜒2 Goodness-of-Fit test (Buckland et al. 2001, 2015) compares the observed frequencies, ni, with the expected frequencies under the model E(ni) and it is given by Equation (19) under the null hypothesis (H0) of good model fitting, i.e., the difference between the observed (ni) and expected (E(ni)) counts is close to zero. In Equation (19), n is the total number of observations, u is the number of groups (or bins) within the distance data, and q is the number of model parameters estimated. Reject H0 if 𝑋𝑜𝑏𝑠 2 > 𝑋1−𝛼;(𝑢−𝑞−1) 2, with the latter representing the 1–𝛼 quantile from a 𝜒2 distribution with u - q - 1 degrees of freedom. As the number of parameters of the fitted model increases, the bias decreases, but the sampling variance increases (Buckland et al. 2001). While the Goodness-of-Fit test results should be considered in the analysis of distance data, they will be of limited value in selecting a model since these tests are sensitive to heaping. Therefore, care is needed in choosing suitable distance intervals. If data are collected with no fixed 𝜔, it is possible that a few extreme outliers will be recorded. These values are not useful, and the data should therefore be truncated. This can be checked using the distances’ histogram, and whether there is evidence of heaping or not (Buckland et al. 2001; Couturier et al. 2018). 82 Goodness-of-Fit tests allow formal testing of whether a detection function model provides an adequate fit to the data. Since the GOF test cannot be used on continuous data, unless grouped, it is of limited use for testing MCDS models (Buckland et al. 2015), being useful for testing models using CDS methods. However, if distances are not grouped, they must first be categorized into groups to allow the test to be conducted. Thus, there is a subjective aspect to the test, and different analysts, using different group cut points, may reach different conclusions about the model adequacy. In contrast, the Kolmogorov– Smirnov and Cramér–von Mises tests can only be applied to continuous data (Buckland et al. 2015). 83 Appendix 4 Distance Sampling Assumptions – short summary Line transect DS assumptions and design are described in Buckland et al. (1993) and a summary of the assumptions in relation to caribou survey in Greenland provided below are from Cuyler et al. (2016). 1. All caribou on the 0-line are detected. This is critical and must be true. 2. Caribou are randomly distributed. (Lacking this will not bias abundance estimates if the line transects are randomly placed, which they were.) 3. Detection of caribou is independent. (Although detection was dependent in our survey, the lines had random start-end points, so this assumption is not violated). 4. No caribou movement prior to detection. The method is a ‘snapshot’ method. In practice this assumption is not violated if the observer moves faster than the animal, e.g., if movement of caribou to the next line transect to be surveyed is rendered impossible, which it was. 5. Distance measurements are exact. Provided distance measurements are approximately unbiased, bias in line transect estimates tends to be small in the presence of measurement errors. In our survey we binned the observations into distance intervals which decreases measurement error. 6. Clusters (caribou groups) close to the 0-line are accurately sized. 7. Other assumptions include those for other survey types, e.g., that each population is closed, being confined within a clearly defined area. 84 Appendix 5 Photos of c amouflaged caribou observed March 2019. Figure 53. Eight caribou, several camouflaged against background, within 200 m of helicopter. Photo A. Jensen. Figure 54. Three caribou camouflaged against background, within 100 m from helicopter. Photo A. Jensen. 85 Appendix 6 Histograms for detected distances Histograms for detected distances superimposed with estimated detection functions for all truncated fitted models, presented order as in Table 3 86 87 Figure 55. Histograms for detected distances superimposed with estimated detection functions for all truncated fitted models. The parameter estimates and variability associated with them (Table 13), with Extreme as the reference level for comparison. Illustrates that High camouflage has a lower (negative) but similar (estimate close to 0) detection probability as Extreme camouflage. Medium camouflage has a more negative estimate and thus lower probability of detection relative to Extreme and High. Further, Low camouflage, being positive, means the probability of detection is greater when compared with Extreme camouflage (as also shown in Fig. 15). Table 13. Detection function parameters’ estimates. Estimate Standard Error Intercept -1.353 0.115 Camouflage High -0.089 0.122 Camouflage Medium -0.571 0.140 Camouflage Low 0.371 0.226 Note: Estimates are on log scale. 88 Appendix 7 Recommendations for improving future surveys Aerial survey methods & design The 9.6% survey coverage for the 2019 Distance Sampling (DS) surveys of the Akia-Maniitsoq caribou population promotes accuracy of abundance estimates and should be continued in the future to permit evaluating population trends. When flying line transects, distance and other factors often make identification of calves impossible, resulting in an underestimate of calf number. Demographic (sex, age, calf recruitment) data must continue to be collected in efforts separate from flying the line transects for DS. Flight altitudes from 30 to 40 m permit scanning the landscape for caribou even out to 1000-1500m from the track line without dead-ground interfering. Just be aware the degree to which the caribou are extremely camouflaged against the typical backgrounds. This can cause observer fatigue, mental exhaustion, even at the relatively slow speeds flown (60-70 km/hour). Any ‘dead’ ground causing caribou detections to be missed, will likely be mitigated by the DS analysis. Training and testing, observer ability to judge correct distance bin is necessary for improvement of this important variable. It is the author’s experience that without practice people commonly misjudge distance. Looking down from above can exacerbate this tendency. Flat terrain may provide a more (normal) horizontal line-of-sight to the animals, which may increase binning accuracy. However, terrain that slopes away, either up or down, confuses observers’ ability to judge distance from track line to animals. The steeper the slope, the greater the errors. The timing for aerial surveys could remain early March because that coincides with annual minimum caribou movement (avoids double counting), and enough day length for flying the pilot maximum of 7-hours per day. Experience from eight surveys since 2000 has illustrated that snow cover and depth is variable regardless of the winter period chosen. In Greenland, helicopters are seldom available at short notice. Book about 3-5 months ahead and reaffirm booking several times thereafter. For estimating the necessary window (dates) that helicopter is booked for survey, first calculate the number of days required for survey. Then, add days to allow for 89 several non-flying days owing to pilot flying hours going over weekly limit, airport closures on Sundays, and poor weather. For the latter a minimum 3-4 days should be allocated. Book survey observers early (Fig. 56), about six months in advance, to ensure the probability of obtaining the observers you require. Attributes would include previous experience detecting superbly camouflaged caribou, and proven lack of nausea, e.g., at sea or in helicopters. Even the usually sedate helicopter maneuvers for transect lines can illicit nausea in some. Meanwhile, the non-stop abrupt flying maneuvers required for caribou demographics cause nausea in most persons. Note: Even previous helicopter experience, including animal live capture, does not guarantee lack of nausea during sharp maneuvers specific to caribou demographics work. Figure 56. The three observers, Dr. Christine Cuyler, scientific leader (left), Aslak Jensen, commercial hunter (center), and Hans Mølgaard, Sisimiut hunting officer (right). Standardization of data collection regarding surface conditions Prior to 2019, the covariates (including degree of camouflage, % snow cover, snow depth, icing, visibility, lighting (e.g., flat light, shadow), presence of boulders and their size, vegetation poking through snow layers, etc.) were recorded without standardization and often ad hoc. In contrast the 2019 survey used specific standardized qualitative terms to make the covariates available for analyses. Evaluations for all environmental covariates were standardized to just five easy qualitative terms: Zero, Low, Medium, High, and Extreme. However, there were too many covariates to permit recording each with every detection of the object-of-interest (caribou).