79 ASSESSING MOOSE HUNTER DISTRIBUTION TO EXPLORE HUNTER COMPETITION Tessa R. Hasbrouck1, Todd J. Brinkman2, Glenn Stout3, and Knut Kielland2 1Alaska Department of Fish and Game, 2030 Sea Level Drive, Ketchikan, AK 99901, USA; 2University of Alaska Fairbanks, 1731 South Chandalar Drive, Fairbanks, AK 99975, USA; 3Alaska Department of Fish and Game, 1300 College Road, Fairbanks, AK 99701, USA ABSTRACT: Traditional values, motivations, and expectations of seclusion by moose (Alces alces) hunters, more specifically their distributional overlap and encounters in the field, may exacerbate perceptions of competition among hunters. However, few studies have quantitatively addressed over- lap in hunting activity where hunters express concern about competition. To assess spatial and tempo- ral characteristics of competition, our objectives were to: 1) quantify temporal harvest patterns in regions with low (roadless rural) and high (roaded urban) accessibility, and 2) quantify overlap in harvest patterns of two hunter groups (local, non-local) in rural regions. We used moose harvest data (2000–2016) in Alaska to quantify and compare hunting patterns across time and space between the two hunter groups in different moose management areas. We created a relative hunter overlap index that accounted for the extent of overlap between local and non-local harvest. The timing of peak har- vest was different (P < 0.01) in urban and rural regions, occurring in the beginning and middle of the hunting season, respectively. In the rural region, hunter overlap scores revealed a concentration in 20% of the area on 16–20 September, with 50% of local harvest on 33% of the area and 54% of non-local harvest on 18% of the area. We recommend specific management strategies, such as lifting the air transportation ban into inaccessible areas, to redistribute hunters and reduce overlap and con- cerns of competition in high-use areas. We also encourage dissemination of information about known hotspots of hunter overlap to modify hunter expectations and subsequent behavior. Our hunter overlap index should prove useful in regions where similar concerns about hunter competition, hunter satis- faction, and related management dilemmas occur. ALCES VOL. 56: 79 – 95 (2020) Key words: Alaska, Alces alces, competition, harvest patterns, hunting, moose, overlap index Local concern over competition for game with non-local hunters has been an ongoing policy issue for wildlife managers in most places hunting occurs, including Alaska (ADF&G 2016a). We define local hunters as people who hunt in the same area they reside, whereas non-local hunters are those who travel away from their resident management area for hunting opportunities. These two groups typically have the same hunting regu- lations, although in some situations rural res- idents in Alaska have priority on federal lands. Competition may be linked to conflict which occurs when the physical presence of an individual directly interferes with the goals of another individual (Jacob and Schreyer 1980), or when individuals pursu- ing the same goal have different values (Saremba and Gill 1991) or norms (Ruddell and Gramann 1994). We define competition as the rivalry between user groups for game or hunting space, either of which may be limited or perceived to be limited, regardless of harvest success. For example, hunters may experience negative consequences from competition even when they successfully MOOSE HUNTER DISTRIBUTION – HASBROUCK ET AL. ALCES VOL. 56, 2020 80 harvest game because the pursuit of seclu- sion or a certain traditional experience can be equally or more important than harvest success (Vaske et al. 1995, Brinkman 2014). Competition is exacerbated when different hunter types (local hunters or non-local hunters) overlap in the same area at the same time, creating a higher potential for direct encounters. Proposed solutions to address competi- tion concerns are often related to changes in the allocation of hunting opportunities. For example, 22 proposals were submitted to the Alaska Board of Game requesting changes in statewide or regional allocation of big game among hunter user groups between 2015 and 2018 (ADF&G 2016a). Many pro- posals (n = 12) focused specifically on moose (Alces alces) and described how local hunters are concerned that non-local hunters take too many moose and create excessive competition. Importantly, the extent of competition has not been objec- tively assessed in the areas where changes in allocation were requested. Information regarding the distribution and overlap of hunting activity by different stakeholder groups should provide insight about the extent of competition and inform potential solutions. Our research addresses this infor- mation gap by quantifying harvest patterns and overlap across space and time between two hunter types (local and non-local) in two hunting regions (accessible urban and inaccessible rural areas). It is important to address concerns about moose hunter competition because of the direct relationship with hunt satisfaction and success (Heberlein and Kuentzel 2002, Fix and Harrington 2012). Managers regulate hunter activity and maintain wildlife popula- tions to optimize hunting opportunities and maximize hunt satisfaction (Ericsson 2003), but satisfaction requires more than providing sufficient animal density (Hammitt et al. 1990, Brinkman et al. 2013). Hunters have expressed that the number of other hunters seen (i.e., perceived crowding) is an import- ant factor of hunt satisfaction (Heberlein and Kuentzel 2002). Overall, these findings sup- port multiple-satisfactions-approach-based management that recognizes multiple fac- tors contribute to hunt satisfaction (Hendee 1974). Conflict can occur when spatial and temporal overlap among hunters surpasses expected hunter density (Shelby and Heberlein 1986, Brinkman 2018), and when hunters encounter other hunters who exhibit, or are perceived to exhibit, hunting values and motivations that do not align with local norms (Fix and Harrington 2012). This is relevant in rural communities that may per- ceive that non-locals do not understand or respect traditional local practices (Kluwe and Krumpe 2003). Altered behavior in response to competition may increase the time, effort, and cost (e.g., fuel) of hunters, which is particularly salient to rural commu- nities with relatively weak cash economies (Brinkman et al. 2014). Assessing character- istics of competition may help better define the problem, enhance communication, and inform resolution of these issues (Decker and Chase 1997). Moose are an ideal species to explore hunter competition because of their critical nutritional, cultural, and economic impor- tance to Alaskan residents (Timmermann and Rodgers 2005, Northern Economics Inc. 2006). Hunter motivations may range from meat provision to trophy experiences, and from spending time with family to interact- ing with nature (Brinkman 2018). In 1987– 2007 in Alaska, 29,000 hunters harvested 7260 moose annually (on average) (Titus et al. 2009) that were used by > 90% of rural Interior Alaska community households (Brown et al. 2010), representing > $78 mil- lion annually in the state economy (McDowell Group 2014). ALCES VOL. 56, 2020 MOOSE HUNTER DISTRIBUTION – HASBROUCK ET AL. 81 With so much interest, importance, and investment in moose hunting, federal and state agencies create moose hunting regula- tions to sustain populations while optimizing diverse hunting opportunities. The Alaska Department of Fish and Game (ADF&G) has divided the state into 5 management regions with 26 Game Management Units (GMU). Some GMUs are subdivided into subunits allowing for more precise and localized man- agement of wildlife populations. Management seeks to mitigate biological and sociopoliti- cal issues (Bath 1995) with regulations that are often complex and differ in space and time. The Alaska Constitution ensures equal access to fish and wildlife for all Alaskans. However, the Alaska National Interest Lands Conservation Act of 1980 (ANILCA; P.L. [Public Law] 96-487) mandates that hunting and fishing priority be given to rural Alaskans on federal land. These contradictory pieces of literature have added complexity to hunting systems in Alaska. Although hunting opportunities exist for diverse interests of many stakeholder types, conflict from perceived competition and differences in value systems among hunter types occurs. Typically, all Alaskan residents (i.e., local and non-local hunters) recreate under the same hunting regulations. Most hunting in Alaska occurs on public land, with many hunters using the same areas year-after-year including setting up hunting camps and informal territories on public land (Johnson et al. 2016, Brinkman 2018). Hunting motivations vary by individ- ual, but local rural hunters place a high sig- nificance on meat provision, whereas non-resident and non-local hunters may be more motivated by novel experiences and trophy opportunities. It is common for people born in rural communities to move to urban areas for education or employment but return to rural communities to hunt (Kofinas et al. 2010). In addition to comparing different hunter groups, it is also important to consider how hunting patterns may change in areas with dif- ferent levels of accessibility (Brinkman et al. 2013). Access is a central logistical challenge to hunting in remote parts of Alaska and fac- tors into the allocation of hunting permits. For example, easily accessible regions in Alaska are more likely to be managed using draw per- mits (limited) because of higher hunter demand and the potential for hunter pressure and competition (Woodford 2014). Similar to areas outside of Alaska, harvest generally increases with proximity to roads (Fuller 1990) or in areas with higher road density (Hayes et al. 2002). General harvest permits (unlimited) in Alaska are more common in remote and inaccessible areas where hunter activity is reduced and overharvest less likely. The goal of our research was to explore hunter distribution in management areas with different accessibility and between hunter groups to assess competition concerns. Our objectives were to: 1) quantify harvest pat- terns over time in regions with low (roadless rural) and high (roaded urban) accessibility, and 2) quantify overlap in harvest patterns of 2 hunter groups (local, non-local) in rural regions where competition is more frequently expressed. By comparing local and non-local hunting patterns, our study addressed the spa- tial and temporal characteristics of user group issues that have not been studied extensively among moose hunters in Alaska. STUDY AREA We examined hunting patterns in GMUs 20, 21, and 24 in Interior Alaska (Fig. 1) where the main ecotype is the boreal forest com- prised mostly of white spruce (Picea glauca), black spruce (P. mariana), birch (Betula papy- rifera), aspen (Populus spp.), and willow (Salix spp.). The area contains low-lying wet- lands mottled with lakes, low scrub bogs, MOOSE HUNTER DISTRIBUTION – HASBROUCK ET AL. ALCES VOL. 56, 2020 82 herbaceous meadows, and forb-herbaceous marshes. Intense winters and summers cre- ate annual temperatures ranging from -40 to 22°C (Brabets et al. 2000). Fire is the pri- mary disturbance regime; however, fire sup- pression is concentrated on only 17% of the landscape, typically near roads and commu- nities (DeWilde and Chapin 2007). Fire alters moose habitat quality and has shaped nearly all of Alaska’s boreal forest, includ- ing the wildland-urban interface. To examine differences between urban and rural hunting regions, we compared 3 GMU subunits with high accessibility near Fairbanks (GMU 20A, 20B, and 20D; 34,600 km2) with 2 subunits (GMU 21D and 24D; 28,000 km2) with low accessibility near Koyukuk, ~250 km west of Fairbanks. Fig. 1. Map of study area depicting the high-access urban and low-access rural regions used to evaluate hunter competition in Alaska, USA. ALCES VOL. 56, 2020 MOOSE HUNTER DISTRIBUTION – HASBROUCK ET AL. 83 Although these two regions have relatively similar habitats, climate, and September moose hunting regulations, they have vastly different social systems, infrastructure, and defining characteristics (Table 1), as well as moose and predator population dynamics. Precise estimates of predator numbers are unknown, but wolf (Canis lupus) and bear (Ursus spp.) predation are significant causes of moose mortality in both regions. Although moose counts (density estimates) were done during our study period, they were not regu- larly completed across the entirety of the study area and we did not incorporate them for this reason. These subunits were selected because of the importance of moose hunting in each region despite their differences in social sys- tems and infrastructure. In the study GMUs, hunters are typically allotted one moose during a September hunting season regardless of residency, ethnicity, or background; rural-priority hunts were not instituted during the study period. Although regulations around Fairbanks have been dynamic, harvest chronology has remained relatively constant over time in both study regions. High-access urban region The road-accessible urban region (GMU 20A/B/D) was situated around Fairbanks North Star Borough and was divided into 87 Uniform Coding Units (UCU) to provide finer spatial resolution to assess hunter har- vest. Trail systems are used to access parts of this region and hunters use road vehicles, ATVs, watercraft (motorized or man-pow- ered), and aircraft for access (Brinkman 2018). This GMU is subdivided into many smaller hunt areas with unique regulations that can change annually. For example, there were 64 different sets of regulations for moose harvest in 2017 (ADF&G 2016b) that included antlerless moose, any bull, and ant- ler or brow-tine restriction hunts. Law enforcement is low in the region but “peer-po- licing” may help limit illegal hunting activ- ity, although the regional poaching level is unknown. In 2016, 6,222 moose hunting per- mits were issued with 1,550 moose harvested (25% success rate; ADF&G 2020). Low-access rural region The rural region (GMUs 21D and 24D) was divided into 35 UCUs, not road-accessi- ble, and situated along the Yukon and Koyukuk Rivers (Fig. 2); local hunters are predominantly Koyukuk Athabascan. The Koyukuk Controlled Use Area (KCUA; 12,408 km²) straddles the northern GMU 21D and the southern GMU 24D. Although moose hunting occurs across the entirety of this region, the majority (especially non-lo- cals) occurs within the KCUA. In other com- munities on the Yukon River, all hunters Table 1. Difference in area, census, density, highway availability, and employment rate for the high access urban and low-access rural study regions, Alaska, USA. Census and employment rate was created by U.S. Census Bureau 2011. Urban Region Rural Region Area (km2) 34,600 28,000 Census (2000) 104,079 1,461 Density (# people/km2) 3.0 0.05 Highway length (km) 737 0 U.S. Census Bureau unemployment rate 7% 20% # of moose hunting regulations (2017) 64 10 MOOSE HUNTER DISTRIBUTION – HASBROUCK ET AL. ALCES VOL. 56, 2020 84 essentially travel and hunt by boat (Johnson et al. 2016). Although some trail systems exist, they are seldom used by moose hunt- ers. The KCUA has a mandatory ADF&G check-in station on the Koyukuk River and the reporting rate is believed high compared to other rural areas. Because all hunters coming from the lower Koyukuk River are required to stop at the check-in station, a comprehensive 20-year dataset of hunter and harvest information was available for the KCUA. The check-in process helps ensure that hunters are harvesting legally, and although minor violations are common, it is believed that egregious activity is minimal during the hunting season; the level of Fig. 2. Map of Uniform Coding Units (UCU) within Game Management Units 21D and 24D in the low-access rural study region used to evaluate hunter competition in Alaska, USA. ALCES VOL. 56, 2020 MOOSE HUNTER DISTRIBUTION – HASBROUCK ET AL. 85 poaching outside of the hunting season is unknown. Access to this region for non-local hunters requires considerable logistic effort and expense. For example, a Fairbanks resi- dent would drive 220 km north on the Dalton Highway to the Yukon River bridge and then boat 483 km down the Yukon River to access Galena near the mouth of the Koyukuk River; however, many non-local hunters travel much further to access this hunting region. This region had 10 different sets of moose harvest regulations in 2017 (Table 2; ADF&G 2016b). Under the registration hunt, Alaskan hunters (regardless of ethnicity or residency) can shoot any bull but are required to render the antlers unusable for trophy con- sideration by cutting one antler palm in half and forfeiting the cut portion to ADF&G. This “antler destruction” regulation was cre- ated to emphasize harvest for meat rather than trophy value (G. Stout, pers. commun.). Resident hunters can apply for a draw permit that allows them to harvest any size bull and to keep the antlers intact. Non-resident hunters can only participate if they receive a draw permit, and although they do not have to destroy an antler, they are mandated to harvest a bull with a minimum of 4 brow tines on at least one antler or with antler spread >127 cm. In 2016, 756 hunt permits were issued and 375 moose were harvested (50% success rate) (ADF&G 2020). METHODS Hunter database The best available information on hunter patterns was accessible from ADF&G annual harvest data. Although mandatory harvest reporting exists statewide, harvest data is likely incomplete due to underreporting in remote areas (Schmidt et al. 2015), but is not considered an issue in our rural region where hunters accept and are compliant with the check-in station. We assumed that hunters who report harvest are representative of all successful hunters within the GMU with respect to location of harvest, hunt patterns, and effort. Because unsuccessful hunters do not report fine-resolution details such as Table 2. Regulations in the low access rural hunting region in 2016 (ADF&G 2016b). Permit types are registration (RP) and drawing (DP), and residency types are residents (R) and non-residents (NR). GMU Permit Type Residency Special Instruction Open Season 21D/24D, within KCUA RP R Any bull, destroy antler Sept 1–Sept 25 DP R Any bull Sept 5–Sept 25 DP NR Antlers ≥127 cm, OR ≥4 browtines on one side Sept 5–Sept 25 21D, outside KCUA RP R Any bull, destroy antler Aug 22–Aug 31, Sept 5–Sept 25 DP NR Any bull Sept 5–Sept 25 DP NR Antlers ≥127 cm, OR ≥4 browtines on one side Sept 5–Sept 25 21D, east of KCUA DP R Any bull Sept 5–Sept 25 DP NR Antlers ≥127 cm, OR ≥4 browtines on one side Sept 5–Sept 25 24D, remainder RP R Any bull, destroy antler Sept 5–Sept 25 DP R Any bull Sept 5–Sept 25 DP NR Antlers ≥127 cm, OR ≥4 browtines on one side Sept 5–Sept 25 MOOSE HUNTER DISTRIBUTION – HASBROUCK ET AL. ALCES VOL. 56, 2020 86 temporal and spatial details of their hunt, our analysis was limited to hunters who har- vested a moose. We acknowledge that unsuc- cessful hunters contribute to and are affected by hunter overlap and competition, but we assumed that successful hunters were a rea- sonable representation of all hunters because the hunter success rate was consistent over time, and many hunters use the same hunting area year-after-year, yet are unsuccessful 50% or more of the time. We analyzed all harvest data during the September hunting season from 2000 to 2016. We included the following harvest data fields in our analysis: hunter residency, success (yes or no), date of kill, and hunt location (UCU). Although harvest data included number of hunting days, we deemed these data insuffi- cient to assess hunter effort because of changes in reporting rates and possible issues with memory recall. We excluded data that were missing hunter residency or date of kill. Antlerless hunts and hunts that occurred out- side the normal September hunting season were not included in the analysis because these hunts did not occur regularly during the study period and were not comparable between regions. Special hunts introduce cir- cumstances that potentially influence harvest decisions that may introduce bias or inaccu- racy in our results. Analysis ESRI ArcGIS was used to map and visu- alize harvest locations and the underlying landscape. We used a Mann-Whitney U test to compare the date of kills in the high-ac- cess urban and low-access rural regions (Objective 1). We calculated descriptive sta- tistics on the proportion of successful local and non-local hunters in each UCU in the low-access rural region to measure spatial and temporal overlap between these 2 hunter groups (Objective 2). We compared the pro- portions using a Mann-Whitney U test to evaluate if local and non-local hunters used the same hunting space. We used a Kolmogorov-Smirnov test to compare the temporal overlap (i.e., distribution of date of kill) among hunters for the full study period. Further, we compared local harvest and non-local harvest over two even time peri- ods: 2000–2008 and 2009–2016. This split allowed us to examine harvest locations over time for each hunter type while maintaining adequate sample sizes. Due to non-normal distribution, we used paired Wilcoxon signed ranks tests to assess changes in UCU use for each hunter type between the two study peri- ods and used Mann-Whitney U tests to quan- tify the differences between local and non-local hunters use in each UCU for each time period. Considering that spatial overlap was compared among UCUs, we created a rela- tive overlap index (Eq. 1) that incorporated the non-local hunter density and the local hunter proportion in each UCU. Due to the low number of non-resident hunters and the similarity of patterns, we pooled non-resi- dents and non-local resident hunters. We used river length (Hydrology 1:1000000) within each UCU to calculate non-local hunter density because nearly all hunters in the rural region access their moose hunting areas by watercraft (Johnson et al. 2016) and areas away from navigable waters are seldom used. Current regulations prohibit airplane transportation for hunters within the KCUA portion of 21D/24D. We ranked each UCU in order of highest relative overlap index. The relative overlap index was calculated for each UCU across the entire study period, and between the two time periods (2000–2008 and 2009–2016). We used a Wilcoxon signed- ranks test to compare overlap in distribution between the two time periods to capture any temporal change in overlap levels. In our index, an increase in non-local hunter density within a specific UCU caused ALCES VOL. 56, 2020 MOOSE HUNTER DISTRIBUTION – HASBROUCK ET AL. 87 an increase in the relative overlap index score, but the increase was mediated by the level of importance of that UCU for local hunters. For example, a UCU with a high non-local hunter density that had high importance to local hunters had a higher score than a UCU with a high non-local hunter density that had low importance to local hunters. This approach was reasonable because this project was derived from and motivated by local hunter concerns in the rural region, and competition concerns between hunter types may be asymmetrical (i.e., locals are concerned about non-local presence, but not necessarily vice versa). # Non-Local Hunters CumulativeRiver Length(km) # Local Hunters Total Local Hunters Relative Overlap Index             ×     = (Eq. 1) Finally, we split the hunting season into 5 equal time periods (1–5, 6–10, 11–15, 16–20, and 21–25 September) and generated a relative overlap index for each UCU for each time block across the full study period. This step allowed us to simultaneously eval- uate the temporal and spatial overlap during the hunting season while maintaining ade- quate sample sizes. RESULTS A total of 25,113 harvest records from 2000 to 2016 were analyzed with 19,423 and 5692 moose in the urban and rural regions, respectively. The urban and rural regions exhibited different date-of-kill distributions (P < 0.01) based on the date of peak harvest (Fig. 3). Peak harvest occurred at the begin- ning of the season in the high-access urban region and in the middle of the season in the low-access rural region. Although local (n = 2286) and non-local hunters (n = 3156) in the rural region had different distributions in date of kill (P < 0.01), they had complete tempo- ral overlap across the hunting season (Fig. 3). In the rural region, 50% of the local har- vest occurred on 33% of the hunting area (5 UCUs), whereas 54% of the non-local har- vest occurred on 18% of the area (3 UCUs). Local and non-local hunters harvested fewer moose in the remaining UCUs (Fig. 4), revealing an uneven hunter distribution Fig. 3. The total daily number of moose harvested in September 2000–2016 by local, non-local, and non-resident hunters in a high-access urban hunting region, Alaska, USA. MOOSE HUNTER DISTRIBUTION – HASBROUCK ET AL. ALCES VOL. 56, 2020 88 across the region. Local and non-local hunt- ers overlapped spatially (P = 0.448) but with differences in importance level among UCUs (Fig. 5). Spatial distribution of non-lo- cal hunter was similar (P = 0.70) in the early (2000–2008) and late study periods (2009– 2016). Conversely, spatial distribution of local hunters changed (P = 0.02) as their proportional use declined in 5 UCUs and increased in 4 UCUs. Further, local and non-local hunters harvested a moose in sim- ilar locations in the early (P = 0.43) and late (P = 0.47) study periods, indicating that overlap existed across time. 0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35 Pr op or tio n N on -L oc al H ar ve st UCU (n = 35) 0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35 Pr op or tio n Lo ca l H ar ve st UCU (n = 35) A B Fig. 4. Proportional use of UCUs by local (A) and non-local hunters (B) within the low-access rural hunting region, Alasaka, USA. The X-axis refers to the most used UCUs for each hunter type and not a specific sub-area; therefore, the UCU value does not necessarily correspond to the same location for local and non-local hunters. Data points displayed as boxes sum together UCUs that constituted 50% of harvest by local (n = 5) and non-local hunters (n = 3). ALCES VOL. 56, 2020 MOOSE HUNTER DISTRIBUTION – HASBROUCK ET AL. 89 Hunter overlap scores ranged from 0 to 0.106 with a mean score of 0.010 (SD = 0.02) and median score of 0.003. In the rural region, high overlap existed in 12% of the hunting area, moderate overlap in 8%, and minimal overlap in the remaining 80% (Fig. 6). Twenty-seven UCUs had overlap scores less than the mean, 4 UCUs had scores ranging from 0.011 to 0.014, and 4 UCUs had the highest overlap scores of 0.033, 0.051, 0.054, and 0.106. The relative overlap index did not differ between the early and late study period (P = 0.92). Within the hunt- ing season, the relative overlap index showed that the extent of overlap (by 5-day periods) within each UCU ranged from 0 to 0.017 (mean = 0.0005, SD = 0.0016) and was high- est on 16–20 September (Fig. 7). DISCUSSION Our research focused on analyzing the spatial and temporal overlap between hunter groups because we assumed that the degree of overlap is directly related to the level of perceived competition. Our analyses indi- cate that rural and urban hunting regions exhibited different distributions of date-of-kill. The high-access urban region spiked soon after the opening of the hunting season, followed by steep decline and then moderate increase near the middle of the season, and then rapid decline; this pattern may reflect that specific regulations end on different dates (15, 20, 25, 30 September). Similar research with other species indicates that white-tailed deer (Odocoileus virgin- ianus) harvest is most closely associated with day of hunting season (Hansen et al. 1986) and red deer (Cervus elaphus) harvest increases on weekends and moon phase (Rivrud et al. 2014). We speculate that warmer weather early in the season might create more comfortable conditions for urban hunters less reliant on a successful annual harvest. Or, the beginning of the sea- son encompasses Labor Day weekend which provides urban hunters an extra hunting day in a region where employment is high rela- tive to remote communities. “Opening day rush” may be a dominant factor in urban areas where there are higher numbers of hunters targeting a limited number of moose. !. !. !. !. !. !. Legend !. Communities Rivers (simplified) >10% Harvest (33% actually) 5-9.9% Harvest 1-4.9% Harvest 0-.9% Harvest ¯ 0 60 12030 Kilometers !. !. !. !. !. !. Legend !. Communnities Rivers (simplified) >10% Harvest 5-9.9% Harvest 1-4.9% Harvest <1% Harvest ¯ 0 60 12030 Kilometers Fig. 5. Percent harvest by UCU in 2000–2016 by non-local (left) and local (right) hunters within the low-access rural hunting region, Alaska, USA. MOOSE HUNTER DISTRIBUTION – HASBROUCK ET AL. ALCES VOL. 56, 2020 90 We speculate that hunting patterns in the low-access rural region are driven by biolog- ical and environmental conditions more so than in the high-access urban area with higher hunter numbers. Fewer hunter num- bers/density in the rural region may afford hunters the opportunity to align their effort with ideal environmental conditions. Bull moose increase movement as rut approaches (Joly et al. 2015), with peak movement during the rut around 1–7 October (Brown et al. 2018) indicating that hunters may have a higher chance of encountering moose late in the season. Also, cooler temperatures in late September facilitate meat preservation in remote regions where several days may elapse between harvest and processing; lower ambient temperatures are critical as Fig. 6. Depiction of UCUs with the top 4 highest overlap scores, moderate overlap scores, and zero- low overlap scores in the low-access hunting region, Alaska, USA. ALCES VOL. 56, 2020 MOOSE HUNTER DISTRIBUTION – HASBROUCK ET AL. 91 meat begins to spoil at 4.4 °C (USDA 2011). Further, local hunters suggest that hunting moose is easier after leaf-fall in mid- to late September in Interior Alaska (Appendix A) because of improved sightability along river networks and sloughs. Due to traditionally lower employment opportunities in rural regions, some local hunters presumably have flexibility when selecting hunting dates. Local and non-local hunters were not evenly distributed across the landscape as 50% of local harvest occurred in 5 UCUs, with the remaining 50% across 30 UCUs. Similarly, 54% of non-local harvest occurred in 3 UCUs with the remaining 46% in 32 UCUs. In the rural region we found that overlap was concentrated within 20% of the area, indicating that 80% of the area had minimal overlap. Figures 5 and 6 indicate that the greatest overlap and highest use by local and non-local hunters occurred in UCUs crossed by the Yukon and Koyukuk Rivers. Although tributaries exist throughout the area, these major rivers offer more con- sistent and safe access to moose. Overlap occurred later in the month, aligning with the peak harvest. Low overlap in the early season could reflect inferior weather conditions reducing hunt success or lower local participation. Local hunters may have more flexibility in their hunting dates whereas non-local hunters likely choose hunting dates in advance of the season because of the substantial time and effort needed to access the region. Also, some draw permits shorten the hunting season for non-resident hunters, therefore artificially bounding their hunting period. Our results indicated that local hunters changed hunt locations over time, but without further investigation, it is difficult to determine if this was a result of overlap, perceived com- petition, shifting moose densities, or other parameters. The “availability framework” uses hunter accessibility, game abundance, and seasonal distribution of game to inform local management decisions, and is also useful to assess hunting opportunities (Brinkman et al. 2013). Consistent with this framework, our findings suggest that along with game supply (e.g., moose abundance and distribution), it is critically important to account for hunter access when exploring how hunting systems function. For exam- ple, our research shows concentrations of Fig. 7. 3D plot showing when and where overlap is greatest between local and non-local hunters in the low access rural hunting region, Alaska, USA. The X-axis refers to individual UCUs, y-axis to the hunter overlap scores, and z-axis to the time period within the hunting season. Colors help visualize scores. MOOSE HUNTER DISTRIBUTION – HASBROUCK ET AL. ALCES VOL. 56, 2020 92 harvest mainly in UCUs with major naviga- ble rivers (Fig. 5). Similarly, Lebel et al. (2012) found that higher levels of access increased the number of harvested white- tailed deer. Hunters in areas with different levels of accessibility may also have differ- ent perceptions of acceptable levels of crowding (Shelby et al. 1989). Due to the near complete reliance on waterways for access, we used cumulative total river length within each UCU to determine hunter den- sity. Application in developed regions with uniform access could use road length, travel corridor length, or overall area as parame- ters. Identifying differences in regions with good and poor access may provide insight to employ local management strategies where hunter satisfaction is a concern. Tensions with non-local hunters are often discussed as issues in rural communi- ties, but the magnitude and scope of such attitudes and tensions have not been assessed quantitatively. Hunter survey research may provide insight into these characteristics and their influence (Brinkman 2018). Without hunter interviews, monitoring hunter inter- actions in the field, or conducting a robust moose behavioral study during the hunting season, we do not have estimates of the actual levels of competition. However, our findings support and inform future studies that may directly assess competition and its potential consequences. Local hunters are generally more toler- ant of other local hunters (Brinkman 2018). Future research might consider assessing proximity of harvest to communities, birth place of hunters instead of current address, and hunter expectations as to where other hunters will be encountered. For example, constructing maps with overlap index scores (see Fig. 2) would help local hunters distin- guish areas of potential conflict with high numbers of non-local hunters. Additionally, “unexpected” overlap may increase hunter conflict where hunters expect solitude but encounter others. Our ability to assess hunter competition was limited by the data that successful hunt- ers provided on harvest reports. Because we were unable to assess overlap by unsuccess- ful hunters, the role of hunter effort in over- lap and perceived competition is unknown. To fully understand competition and hunter satisfaction, we recommend that more data be collected from unsuccessful hunters, especially with regard to hunter effort. Questioning hunters about their own percep- tions as to why they were unsuccessful (e.g., bad weather, too many people, varying lev- els of effort, too many predators) may help managers address the root of satisfaction issues. This research focused on competition concerns of local hunters and we acknowl- edge that non-local hunters’ perceived com- petition by non-local hunters was not assessed, and that they may feel negative impacts from competition or may not agree with or appreciate local-societal norms. A more inclusive survey should assess the per- ceived competition in a management area by all hunters – local, non-local, and non-resi- dent. Yet, we provide novel information about previously untested hypotheses related to the issue of hunter competition. Our hunter overlap index can be modified and applied for different regions and game spe- cies to assess the existence and relative level of hunter competition. Because this overlap equation index does not include a temporal component, it would be important for researchers to assess the relative overlap score within salient time periods (Fig. 7). MANAGEMENT IMPLICATIONS To minimize hunter dissatisfaction, we recommend that game managers employ strategies to distribute hunters from areas ALCES VOL. 56, 2020 MOOSE HUNTER DISTRIBUTION – HASBROUCK ET AL. 93 with a relatively high overlap index scores, especially during the peak and latter half of the hunting season. We encourage altering current regulations, specifically lifting the prohibition of using aircraft for transporta- tion into the KCUA with the caveat that air- craft use occur >1.6 km from the Yukon and Koyukuk River corridors. This should increase hunter distribution across the land- scape, provide unique opportunities for non-local hunters, and reduce conflict with local hunters who seldom use airplane trans- port or travel > 1.6 km from a navigable river. We further recommend providing information about harvest hot spots and “high overlap” to help hunters avoid areas with historically high hunter density and potential competition. Ultimately, both indi- vidual hunters and agencies should adapt to alleviate hunter competition and optimize hunt satisfaction. Our research provides a methodology to quantify hunter distribution which is broadly applicable in addressing concerns about hunter competition, a problem shared by many states and provinces. By using pre-ex- isting data (i.e., harvest records) that nearly all wildlife management agencies collect, our approach can help managers identify where new strategies might be useful to modify hunter distribution to ease hunter conflict. Managers could alter the timing of hunts (e.g., staggered entry) or modes of access (e.g., open or close roads, motorized or non-motorized access) to redistribute hunter activity. We speculate that structured interviews with hunters given hunter distri- bution maps may provide insight regarding support for proposed novel and adaptive strategies. Management responses to objec- tive information such as our hunter distribu- tion analyses, that identify the location and causes of competition, should foster public acceptance and compliance with related management decisions. ACKNOWLEDGMENTS Our research was part of a larger project, University of Alaska Fairbanks’s Community Research Partnerships for Supporting Sustainable Traditional Harvest Practices. Through this larger project, individual stud- ies were designed to work collaboratively with communities to design objective, rele- vant, and important research related to hunt- ing and fishing practices. Two entities, Koyukuk Traditional Council and Nulato Tribal Council, located in the rural GMUs partnered with us to address local research priorities related to hunter competition. We gratefully acknowledge Nulato Tribal Council (specifically A. Demoski), Koyukuk Traditional Council, Council of Athabascan Tribal Governments, and Tanana Chiefs Conference for their partnerships and input. Further, we thank K. Heeringa and the Community Research Partnerships for Supporting Sustainable Traditional Harvest Practices crew for their guidance and feed- back. Our research was funded by the National Science Foundation (award 1518563) and NASA (award NNX15AT72A). REFERENCES AlAskA DepArtment of Fish and GAme (ADF&G). 2016a. Interim reports moose – Year 2016. (accessed June 2018). _____. 2016b. 2016–2017 Alaska Hunting Regulations, No. 56. (accessed March 2018). _____. 2020. Moose hunting in Alaska: har- vest statistics. (accessed July 2020). BAth, A. J. 1995. The role of human dimen- sions in wildlife research in wildlife management. Ursus 10: 349–355. BrABets, T. P., B. WAnG, and R. H. meADe. 2000. Environmental and Hydrologic Overview of the Yukon River Basin, Alaska and Canada. Water-Resources https://secure.wildlife.alaska.gov/index.cfm https://secure.wildlife.alaska.gov/index.cfm http://hunt.alaska.gov http://hunt.alaska.gov https://adfg.gov/index.cfm?adfg=moosehunting.harvest https://adfg.gov/index.cfm?adfg=moosehunting.harvest MOOSE HUNTER DISTRIBUTION – HASBROUCK ET AL. ALCES VOL. 56, 2020 94 Investigations Report 99-4209. U.S. Geological Survey, Anchorage, Alaska, USA. BrinkmAn, T. J. 2014. Alaska Sheep Hunter Survey: Resident Sheep Hunter Responses. Report for Alaska Department of Fish and Game, Fairbanks, Alaska, USA. _____. 2018. Hunter acceptance of antlerless moose harvest in Alaska: importance of agency trust, proximity of hunter resi- dence to hunting area, and hunting expe- rience. Human Dimensions of Wildlife 23: 129–145. doi:10.1080/10871209.201 7.1399486 _____, G. koFinAs, W. D. hAnsen, F. S. ChApin III, and S. rupp. 2013. A new framework to manage hunting: why we should shift focus from abundance to availability. The Wildlife Professional 7: 38–43. _____, K. B. mArACle, J. kelly, m. VAnDyke, A. Firmin, and A. sprinGsteen. 2014. Impact of fuel costs on high-latitude sub- sistence activities. Ecology and Society 19: 18. doi:10.5751/ES-06861-190418 BroWn, C. l., A. r. Brenner, h. ikutA, e. h. mikoW, B. retherForD, l. J. slAyton, A. trAinor, J. pArk, D. koster, and M. L. Kostick. 2010. The harvest and uses of wild resources in Mountain Village, Marshall, Nulato, Galena, and Ruby, Alaska. Technical Paper Number 410. Alaska Department of Fish and Game, Fairbanks, Alaska, USA. _____, k. kiellAnD, t. J. BrinkmAn, s. l. GilBert, and e. s. euskirChen. 2018. Resource selection and movement of male moose in response to varying lev- els of off-road vehicle access. Ecosphere 9(9): e02405. doi:10.1002/ecs2.2405 DeCker, D. t., and l. C. ChAse. 1997. Human dimensions of living with wildlife: a man- agement challenge for the 21st century. Wildlife Society Bulletin 25: 788–795. DeWilDe, l., and F. s. ChApin III. 2007. Human impacts on the fire regime of Interior Alaska: interactions among duels, ignition sources, and fire suppression. 2007. Ecosystems 9: 1342–1353. doi:10. 1007/s10021-006-0095-0 eriCCson, G. 2003. Of moose and man: the past, the present, and the future of human dimensions in moose research. Alces 39: 11–26. Fix, p. J., and A. m. hArrinGton. 2012. Measuring motivations as a method of mitigating social values conflict. Human Dimensions of Wildlife 17: 367–375. doi:10.1080/10871209.2012.682325 Fuller, T. K. 1990. Dynamics of a declining white-tailed deer population in north-cen- tral Minnesota. Wildlife Monographs 110: 1–37. hAmmitt, W. E., C. D. mCDonAlD, and M. E. PAtterson. 1990. Determinants of mul- tiple satisfaction for deer hunting. Wildlife Society Bulletin 18: 331–337. hAnsen, l. p., C. m. nixon, and F. loomis. 1986. Factors affecting daily and annual harvest of white-tailed deer in Illinois. Wildlife Society Bulletin 14: 368–376. hAyes, s. G., D. J. leptiCh, and p. ZAGer. 2002. Proximate factors affecting male elk hunting mortality in Northern Idaho. Journal of Wildlife Management 66: 491–499. doi:10.2307/3803182 heBerlein, t. A., and W. F. kuentZel. 2002. Too many hunts or not enough deer? Human and biological determinants of hunter satisfaction and quality. Human Dimensions of Wildlife 7: 229–250. doi:10.1080/10871200214753 henDee, J. C. 1974. A multiple-satisfaction approach to game management. Wildlife Society Bulletin 2: 104–113. JACoB, G., and r. sChreyer. 1980. Conflict in outdoor recreation: a theoretical per- spective. Journal of Leisure Research 12: 368–380. doi:10.1080/00222216.19 80.11969462 Johnson, i., t. J. BrinkmAn, k. Britton, J. kelly, k. hunDertmArk, B. lAke, and D. VerBylA. 2016. Quantifying rural hunter access in Alaska. Human Dimensions of Wildlife 21: 240–253. doi:10.1080/10871209.2016.1137109 ALCES VOL. 56, 2020 MOOSE HUNTER DISTRIBUTION – HASBROUCK ET AL. 95 Joly, k., t. CrAiG, m. s. sorum, J. s. mCmillAn, and m. A. spinDler. 2015. Variation in fine-scale movements of moose in the Upper Koyukuk River Drainage, Northcentral Alaska. Alces 51: 97–105. kluWe, J., and e. e. krumpe. 2003. Interpersonal and societal aspects of use conflicts. International Journal of Wilderness 9: 28–33. koFinAs, G. p., F. s. ChApin iii, s. BurnsilVer, J. i. sChmiDt, n. l. FresCo, k. kiellAnD, s. mArtin, A. sprinGsteen, and t. s. rupp. 2010. Resilience of Athabascan subsistence systems to Interior Alaska’s changing climate. Canadian Journal of Forest Research 40: 1347–1359. doi:10.1139/X10-108 leBel, F., C. DussAukt, A. mAsse, and s. D. Cote. 2012. Influence of habitat features and hunter behavior on white-tailed deer harvest. Journal of Wildlife Management 76: 1431–1440. doi:10.1002/jwmg.377 mCDoWell Group. 2014. The Economic Impacts of Guided Hunting in Alaska. Prepared for Alaska Professional Hunters Association. McDowell Group, Anchorage, Alaska, USA. northern eConomiCs inC. 2006. The Value of Alaska Moose. Prepared for Anchorage Soil and Water Conservation District and the Alaska Soil and Water Conservation District, Anchorage, Alaska, USA. riVruD, i. m., e. l. meisinGset, l. e. loe, and A. mysteruD. 2014. Interaction effects between weather and space use on harvesting effort and patterns in red deer. Ecology and Evolution 4: 4786–4797. doi:10.1002/ece3.1318 ruDDell, e. J., and J. h. GrAmAnn. 1994. Goal orientations, norms, and noise-in- duced conflict among recreation users. Leisure Sciences 16: 93–104. doi:10.1080/01490409409513222 sAremBA, J., and A. Gill. 1991. Value con- flicts in mountain parks. Annals of Tourism Research 18: 155–172. doi:10.1016/0160-7383(91)90052-D sChmiDt, J. i., k. A. hellie, and F. s. ChApin iii. 2015. Detecting, estimating, and correcting for biases in harvest data. Journal of Wildlife Management 79: 1152–1162. doi:10.1002/jwmg.928 shelBy, B., and t. A. heBerlein. 1986. Carrying Capacity in Recreation Settings. University of Oregon Press, Corvallis, Oregon, USA. _____, J. J. VAske, and t. A. heBerlein. 1989. Comparative analysis of crowding in mul- tiple locations: results from fifteen years of research. Leisure Sciences 11: 269–291. doi:10.1080/01490408909512227 timmermAnn, h. r., and A. r. roDGers. 2005. Moose: competing and comple- mentary values. Alces 41: 85–120. titus, k., t. l. hAynes, and t. F. pArAGi. 2009. The importance of moose, cari- bou, deer, and small game in the diets of Alaskans. Pages 137–143 in R. T. Watson, M. Fuller, M. Pokras, and W. G. Hunt, editors. Ingestion of Lead from Spent Ammunition: Implications for Wildlife and Humans. The Peregrine Fund, Boise, Idaho, USA. uniteDstAtesDepArtment oF AGriCulture (USDA). 2011. How Temperatures Affect Food. (accessed October 2018). u.s. Census BureAu. 2011. 2010 Census of Population and Housing. (accessed April 2018). VAske, J. J., m. p. Donnelly, k. WittmAn, and s. lAiDlAW. 1995. Interpersonal ver- sus social-values conflict. Leisure Sciences 26: 1–11. WooDForD, r. 2014. What Are Draw Hunts? How They Work and Improving the Odds. Alaska Fish and Wildlife News. . https://www.fsis.usda.gov/wps/portal/fsis/topics/food-safety-education/get-answers/food-safety-fact-sheets/safe-food-handling/how-temperatures-affect-food/ct_index https://www.fsis.usda.gov/wps/portal/fsis/topics/food-safety-education/get-answers/food-safety-fact-sheets/safe-food-handling/how-temperatures-affect-food/ct_index https://www.fsis.usda.gov/wps/portal/fsis/topics/food-safety-education/get-answers/food-safety-fact-sheets/safe-food-handling/how-temperatures-affect-food/ct_index https://www.fsis.usda.gov/wps/portal/fsis/topics/food-safety-education/get-answers/food-safety-fact-sheets/safe-food-handling/how-temperatures-affect-food/ct_index https://www.fsis.usda.gov/wps/portal/fsis/topics/food-safety-education/get-answers/food-safety-fact-sheets/safe-food-handling/how-temperatures-affect-food/ct_index https://www.fsis.usda.gov/wps/portal/fsis/topics/food-safety-education/get-answers/food-safety-fact-sheets/safe-food-handling/how-temperatures-affect-food/ct_index http://live.laborstats.alaska.gov/cen/dparea.cfm http://live.laborstats.alaska.gov/cen/dparea.cfm http://www.ADF&G.alaska.gov/index.cfm?ADF&G=wildlifenews.view_article&articles_id=645 http://www.ADF&G.alaska.gov/index.cfm?ADF&G=wildlifenews.view_article&articles_id=645 http://www.ADF&G.alaska.gov/index.cfm?ADF&G=wildlifenews.view_article&articles_id=645