101 MOOSE POPULATION DYNAMICS DURING 20 YEARS OF DECLINING HARVEST IN BRITISH COLUMBIA Gerald Kuzyk1, Ian Hatter2, Shelley Marshall3, Chris Procter4, Becky Cadsand5, Daniel Lirette6, Heidi Schindler7, Michael Bridger8, Patrick Stent1, Andrew Walker9, and Michael Klaczek3 1Ministry of Forests, Lands and Natural Resource Operations and Rural Development, 205 Industrial Road G, Cranbrook, British Columbia V1C 7G5, Canada; 2Nature Wise Consulting, 308 Uganda Avenue, Victoria, British Columbia V9A 5X7, Canada; 3Ministry of Forests, Lands and Natural Resource Operations and Rural Development, 2000 S Ospika Boulevard, Prince George, British Columbia V2N 4W5, Canada; 4Ministry of Forests, Lands and Natural Resource Operations and Rural Development, 1259 Dalhousie Drive, Kamloops, British Columbia V2C 5Z5, Canada; 5Ministry of Forests, Lands and Natural Resource Operations and Rural Development, Suite 400 - 640 Borland Street, Williams Lake, British Columbia V2G 4T1, Canada; 6Ministry of Forests, Lands and Natural Resource Operations and Rural Development, 300 S Cariboo Highway, 100 Mile House, British Columbia V0K 2E0, Canada; 7Ministry of Forests, Lands and Natural Resource Operations and Rural Development, 3726 Alfred Avenue, Smithers, British Columbia V0J 2N0, Canada; 8Ministry of Forests, Lands and Natural Resource Operations and Rural Development, Suite 400 - 10003 110th Avenue, Fort St. John, British Columbia V1J 6M7, Canada; 9Ministry of Forests, Lands and Natural Resource Operations and Rural Development, 102 Industrial Place, Penticton British Columbia V2A 7C8, Canada. ABSTRACT: Licenced harvest of moose (Alces alces) in British Columbia, Canada declined by approximately half over the 20-year period from 1996–2015. To better understand changes in moose populations coinciding with this period of declining harvest, we modelled population dynamics within 31 Game Management Zones (GMZs). We used aerial survey data (180 density and 159 composition surveys) combined with licensed harvest to develop 4 competing statistical models to assess popula- tion dynamics based on constant parameters and temporal trends in calf:cow ratios at 6 months, juvenile survival from 6–18 months, or cow survival. The models indicated that moose populations declined (λ < 1) in 7 GMZs (23%) from 1996–2005 and in 22 GMZs (71%) from 2006–2015. Over the 20-year period, the best model was fit with declining trends in calf:cow ratios in 8 GMZs, declining juvenile survival in 6 GMZs, and declining cow survival in 8 GMZs. Population growth rate was slightly reduced in those GMZs where licenced antlerless (cow and calf) hunting occurred but was not considered the primary factor causing population decline. Total licenced bull harvest influenced bull:cow ratios that were significantly lower in 2006–2015 (x = 37:100) than 1995–2005 (x = 48:100); significant predictive relationships existed between harvest rates and bull:cow ratios. Provincial moose numbers and harvest were highly correlated (r = 0.81) suggesting that declining harvest was a reaction to declining population trends. We found that the provincial moose population increased 6% from 1996–2005, subsequently declined 32% from 2006–2015, and declined 29% overall during the 20-year study period. ALCES VOL. 54: 101–119 (2018) Key words: Alces alces, bull:cow ratio, calf survival, cow survival, harvest, juvenile recruitment Moose (Alces alces) remain an impor- tant hunted species throughout their circum- polar range (Telfer 1984, Kelsall 1987, Karns 2007, Boman et al. 2011), but concern exists about declining populations in parts of North America (Lenarz et al. 2009, Decesare BC MOOSE POPULATION DYNAMICS – KUZYK ET AL. ALCES VOL. 54, 2018 102 et al. 2014, Timmermann and Rodgers 2017). Moose populations are vulnerable to factors such as predation and hunting, both of which can be influenced by human-in- duced landscape change (Murray et al. 2006, Brown 2011). Moose co-exist with predators throughout most of their range in Canada and Alaska, and in areas where predators are lightly hunted, moose density is typically low (<400 moose/1000 km2), and male (hereafter bull)—only hunting may be the only viable harvest option (Gasaway et al. 1992, Boertje et al. 1996). In these systems moose populations may fluctuate over time and decline in the absence of hunting (Gasaway et al. 1983). A recent assessment of the provincial moose population and licenced harvest trends in British Columbia (BC) documented that licenced harvest declined by approxi- mately half over a 20-year period from 1995–2014 as regional population declines of ~50–70% occurred in the central interior (Kuzyk 2016). These declines occurred con- currently with a mountain pine beetle (Dendroctonus ponderosae) outbreak and associated salvage logging and road build- ing, which presumably facilitated predator and hunter access to moose (Ritchie 2008). In 2013, the province initiated research to determine factors leading to the population declines in the central interior. Because the declines occurred over a relatively short period of 20 years, it was assumed that female moose (hereafter cow) survival was the most influential population parameter in the decline (Kuzyk and Heard 2014). These declines continue to create concern among stakeholders and First Nations who have requested more refined information about provincial moose population and harvest trends to help inform management decisions (Gorley 2016). To provide a consistent and objective assessment of relationships between moose population dynamics and declining harvest levels (Kuzyk 2016), we followed an approach similar to that of Hatter (1999) who assessed moose population trends in relation to harvest strategies in 19 Game Management Zones (GMZs) in northern and central BC from 1994–1996. GMZs generally share similar ecological characteristics and hunter harvest patterns providing a suitable geo- graphic unit for managing moose hunting. All GMZs are amalgamations of 1–13 Wildlife Management Units (WMUs) which is the geographic area used to collect hunting information. We expanded Hatter’s (1999) approach and included all GMZs (n = 31) with licenced moose hunting seasons and sur- vey data over a 20-year period (1996–2015). Our objectives were to: 1) evaluate compet- ing models of moose population dynamics in each GMZ to determine which parameters (i.e., calf:cow ratio at 6 months, juvenile sur- vival from 6–18 months, or cow survival) best predicted moose population trend, 2) assess the density, composition, and trend of moose populations based on the best model over the 20-year period within each GMZ, and 3) evaluate the influence of licenced moose hunting based on the best model within each GMZ. STUDY AREA We assessed moose population dynam- ics and licenced harvest in 31 GMZs where moose hunting is authorized in BC (Fig. 1). These areas are ecologically diverse (Meidinger and Pojar 1991) with moose occupying landscapes of differing topogra- phy and vegetation including northern boreal forests, dry interior forests of the cen- tral plateau, and some mountainous habitats (Eastman and Ritcey 1987, Kuzyk et al. 2016). A mountain pine beetle outbreak beginning in the early 1990s occurred over much of the central interior of the province leaving large amounts of dead pine, which ALCES VOL. 54, 2018 KUZYK ET AL. – BC MOOSE POPULATION DYNAMICS 103 led to increased salvage logging and road building (Alfaro et al. 2015). The increased number of roads and cutblocks were thought to facilitate hunter and predator access to moose (Ritchie 2008, Kuzyk and Heard 2014). Moose co-exist with wolves (Canis lupus), grizzly bears (Ursus arctos), and black bears (U. americanus) throughout most of their BC range and overlap with cougars (Felis concolor) in the central and southern areas (Spalding and Lesowski 1971, Mowat et al. 2013, Kuzyk and Hatter 2014). A diversity of ungulates including mule deer (Odocoileus hemionus), white- tailed deer (O. virginianus), elk (Cervus elaphus), bison (Bison bison), and caribou (Rangifer tarandus) also occur here (Shackleton 1999). Licenced hunting sea- sons for bulls occurred in all GMZs having a range of season dates between August 15 and November 30. Bull hunting was regu- lated with general open seasons with or without antler restrictions, limited-entry sea- sons (i.e., hunters must draw an appropriate authorization) with no antler restrictions, or a combination of general open and limited entry seasons. Hunting seasons for cows and calves were mostly limited-entry hunts in 7 GMZs between October 1 and December 10 over most of the 20-year period; general open seasons for calves existed in a few select areas. METHODS Moose density and composition Two winter aerial survey methods were used to collect reliable population information following provincial standards (RISC 2002). Fig. 1. Game Management Zones (n = 31) with licenced moose hunting from 1996-2015 in British Columbia, Canada. BC MOOSE POPULATION DYNAMICS – KUZYK ET AL. ALCES VOL. 54, 2018 104 Density surveys collected a combination of population size, density, and composition (i.e., bull:cow and calf:cow ratios) informa- tion, whereas composition surveys gathered only bull:cow and calf:cow ratios. Density surveys were typically conducted in 5–7 con- secutive days in December–March using stratified random blocks that could be remeasured to detect population trends 5–7 years later (Gasaway et al. 1986). Certain surveys were modified to include habitat- based stratification (Heard et al. 2008), and distance sampling surveys were used in more open habitats (Peters et al. 2014); all survey types produced comparable density estimates. A sightability correction factor developed in central BC (Quayle et al. 2001) was used to account for detection probabil- ity. These surveys followed established stan- dards for accuracy and precision (90% CI) with allowable error from ±15–25% of the estimated population size (RISC 2002). Composition surveys were conducted over 1–3 days in early winter (December– January) prior to typical antler drop. These surveys provided bull:cow and calf:cow ratios which were used as a general index of population trend to gauge progress towards harvest management objectives. Frequency of surveys varied among GMZs due to population objectives, and financial and logistical constraints. There were a total of 180 density surveys (GMZ range = 1–15) and 159 composition surveys (GMZ range = 1–24) used in our analysis. Over the 20-year period, the average num- ber of density and composition surveys per GMZ was 5.8 (min = 1, max = 15) and 5.1 (min = 0, max = 24), respectively. Licenced harvest Licenced resident harvest was estimated annually from 1996–2015 with a provincial hunter survey generating data from mail-out questionnaires sent to a random sample of moose hunters. These estimates (95% CI) were produced from an annual average of 15,477 questionnaires with an average response rate of 68%. Reporting of non- resident licenced harvest was mandatory and obtained from guide declarations. We used both resident and non-resident harvest in the models, but only used resident kill per unit effort (KPUE; resident kill/100 resident hunter days) as an indicator of population trend. We recognized that success rates dif- fered between resident and non-resident hunters, and that the overwhelming majority of moose hunters were residents. Unlicenced hunter harvest, which we defined as moose legally harvested by First Nations and moose harvested illegally, was not quantified as an annual harvest statistic due to the limited availability of these data. Unlicenced harvest mortality was incorpo- rated in estimates of annual non-hunting survival rates derived from radio-collared moose (Kuzyk et al. 2016), and was assumed as a constant proportion of annual mortality. Modelling approach Our models provided a reasonable trade-off between what can be measured practically by biologists and what is needed to help pre- dict moose population trends and responses to harvest management strategies (Hatter 1999). The intent was to utilize a statistically rigorous and objective modelling approach while maintaining a relatively easy-to- implement procedure for constructing population models from multiple types of population data. Because we wanted to make comparisons among GMZs, it was impor- tant that the standardized approach and assumptions were consistent among models. We used models built in Microsoft EXCEL® and model fitting was accom- plished using EXCEL’s built-in optimizer function, SOLVER. ALCES VOL. 54, 2018 KUZYK ET AL. – BC MOOSE POPULATION DYNAMICS 105 We constructed discrete-time, stage- structured population models for each of the 31 GMZs. Our model and model-fitting approach followed White and Lubow (2002), where the focus was on building a series of simple candidate models and selecting the most parsimonious model (i.e., the model that best fits the data with the fewest number of estimated parameters) using Akaike Information Criteria (AIC) (Burnham and Anderson 2002). An exception was that we included resident KPUE as a trend index. Although KPUE was available for all GMZs and was consistently collected over the 20-year study period, we did not use KPUE in GMZs 3b, 3c, 8a, and 8b because the trends in KPUE were inconsistent with the trends from periodic population estimates. For all other GMZs, we fit the KPUE index to the model using procedures described by Hatter (1998) and Haddon (2001). The data for each GMZ typically included estimates of absolute abundance in winter, which were extrapolated from density surveys, winter age and sex ratios (i.e., bull:cow and calf:cow ratios), KPUE, and the annual harvest (bulls, cows, and calves). The key parameters estimated by the model included: R, the recruitment rate or the calf:cow ratio when calves were ~6 months old which we define as the calf:cow ratio at 6 months; Sj, the calf survival rate from 6–18 months which we define as juvenile survival from 6–18 months; and Sf, the annual cow (>18 months) survival rate excluding licenced hunting which we define as cow survival. We used cow survival rates esti- mated from radio-collared cows from 5 study areas, and juvenile survival rates from 2 of these areas (Kuzyk et al. 2016). Survival rates were available for a 2-year period in 5 study areas corresponding to GMZs 7Ob, 7Oc, 6c, 5d, and 3c; we used the documented survival rate from the corresponding study area for these GMZs each year. For the remaining GMZs, we used the average of the 5 study areas for each year. The juvenile survival rates were obtained from GMZ 7Ob and 7Oc in 2016. The estimates were based on the change in calf:cow ratios between early winter and late winter surveys, and included an adjustment for adult cow sur- vival between these 2 time periods (Skalski et al. 2005). All field-based estimates except harvest level included the standard error (SE). We fixed 3 model parameters at values reported in the literature: wounding loss equaled 15% of licenced hunting (Gasaway et al. 1983, Boer and Keppie 1988, Fryxell et al. 1988), calf sex ratio was 50:50 (Ballard et al. 1991, Boer 1992), and the natural adult bull survival rate was 96% of the cow sur- vival rate (Peterson 1977). We built 4 competing models for each GMZ (n = 124) and considered each as a potential hypothesis driving population dynamics. Model 1 held all 3 parameters constant; i.e., calf recruitment, juvenile sur- vival from 6–18 months, and cow survival. The other 3 models included a single para- meter as a linear trend while holding the other 2 constant (Table 1). For each model we calculated the corrected AICc for a small sample in order to determine the best model for each GMZ and AIC weight (wi) (Burnham and Anderson 2002). Changes in moose population dynamics over 4 time periods (1996–2015, 1996–2005, 2006–2015, 2011–2015) were evaluated. We chose 1996–2015 as a long-term (20 year) overview as it corresponded to the earlier moose population assessment (Kuzyk 2016) and when provincial survey methods became standardized. The period of 1996–2005 cor- responded to when the provincial harvest was relatively stable, and the 2006–2015 period corresponded to when annual har- vests declined by 37%. The 2011–2015 period corresponded to when additional har- vest restrictions were imposed for numerous BC MOOSE POPULATION DYNAMICS – KUZYK ET AL. ALCES VOL. 54, 2018 106 GMZs due to increasing concerns about declining moose densities (Kuzyk and Heard 2014, Kuzyk 2016). Annual population estimates were summed across all GMZs to provide the provincial estimates from 1996–2015. We used the best fit model estimates of post-hunt females (cows and calves) to assess the influence of cow and calf hunting on moose population growth. We calculated both the potential rate of population change without female hunting (λp) and the actual rate of change with female hunting (λh). A λp value >1 indicated potential growth and thus some opportunity for cow harvest, and a value <1 implied that licenced cow harvest had some negative influence on population growth. We used the modelled estimates of the post-hunt adult ratios to assess the influence of bull hunting on bull:cow ratios. It is unknown what adult sex ratio ensures that cows are synchronously bred to avoid imped- ing population growth (Timmermann 1992). Ministry policy is to maintain post-hunt, bull:cow ratios ≥ 30 bulls:100 cows in areas with densities > 200 moose/1000 km2, and 50 bulls:100 cows in areas with <200 moose/1000 km2 (BC MOE 2010). RESULTS Best models Model 1 was the best model (i.e., lowest AICc) in 7 (23%) GMZs, Model 2 in 12 (39%) GMZs, Model 3 in 2 (6%) GMZs, and Model 4 in 10 (32%) GMZs (Table 2). Model 1 had substantial support (i.e., ΔAICc <2) in 12 GMZs (39%), Model 2 in 16 GMZs (52%), Model 3 in 6 GMZs (19%), and Model 4 in 14 GMZs (45%). Models 2 and 4 had very strong support in 10 GMZs, each with a relative AIC weight of 1 in 5 GMZs. Twenty GMZs had a single top model with strong support (i.e., ΔAICc <2), 5 GMZs had 2 models, and 6 GMZs had 3 models (Table 2). Population dynamics We calculated the average (1996–2015) moose density, population composition, key population parameters, and rate of change by GMZ for the best fitting models, and the aver- age rate of population change (λ) in 1996– 2005, 2006–2015, and 2011–2015 (Table 2). The rate of population change varied spatially between 1996–2005 and 2006–2015 (Fig. 2). Population density in 1996–2015 varied from 6 (GMZ 4Wb) to 1078 (GMZ 7Ob), averag- ing 334 moose/1000 km2 (SD = 271) across Table 1. Sequence of models fit to moose population survey and harvest data for using parameters of calf:cow ratios at 6 months (R), juvenile survival from 6–18 months (SJ), or cow survival (SF) in each of 31 Game Management Zones in British Columbia, Canada. Model Calf:cow ratio (R) Juvenile survival (SJ) Cow survival (SF) Model structure 1 Constant Constant Constant Rt = R SJ,t = SJ SF,t = SF 2 Linear trend Constant Constant Rt = Rintercept + Rslope · (Yeart – Year0) SJ,t = SJ SF,t = SF 3 Constant Linear trend Constant Rt = R SJ,t = SJ,intercept + SJ,slope · (Yeart – Year0) SF,t = SF 4 Constant Constant Linear trend Rt = R SJ,t = SJ SF,t = SF,intercept + SF,slope · (Yeart – Year0) ALCES VOL. 54, 2018 KUZYK ET AL. – BC MOOSE POPULATION DYNAMICS 107 (C on tin ue d ) Ta bl e 2. T he a ve ra ge m oo se d en si ty , p op ul at io n co m po si tio n, k ey p op ul at io n pa ra m et er e st im at es , a nd r at e of c ha ng e ba se d on th e be st m od el s (l ow es t A IC c) in 3 1 G am e M an ag em en t Z on es fr om 1 99 6– 20 15 in B ri tis h C ol um bi a, C an ad a. T he a ve ra ge ra te o f c ha ng e is s ho w n fo r 1 99 6– 20 15 , 1 99 6– 20 05 , 2 00 6– 20 15 , an d 20 11 –2 01 5; th os e in g ra y ha ve λ <1 . G M Z M od el s1 ,2 w ith Δ A IC c < 2 A IC w ei gh t (w i)3 D en si ty Se x/ A ge R at io s K ey P op ul at io n Pa ra m et er s4 R at e of C ha ng e (λ )5 /1 00 0 km 2 B :1 00 C C a: 10 0 C R S J S F 19 96 –2 01 5 19 96 –2 00 5 20 06 –2 01 5 20 11 –2 01 5 3b 1, 2 , 4 0. 43 24 2 18 34 34 89 89 1. 03 1. 04 1. 03 1. 05 3c 1, 3 0. 57 20 9 29 37 37 68 89 1. 01 1. 01 1. 02 1. 02 3d 2 1. 00 16 1 69 47 15 –7 9 81 81 0. 97 1. 03 0. 91 0. 87 4E a 2 1. 00 20 4 46 30 6– 53 90 92 1. 02 1. 07 0. 97 0. 94 4E b 2 1. 00 12 8 40 27 7– 48 90 91 1. 02 1. 07 0. 97 0. 94 4E c 1 0. 60 94 60 27 27 86 86 0. 96 0. 95 0. 97 0. 98 4W a 2, 1 0. 61 85 55 49 23 –7 5 90 90 1. 10 1. 17 1. 04 1. 02 4W b 4 0. 99 6 49 30 30 81 –9 0 81 –9 9 0. 99 1. 01 0. 97 0. 95 5a 2, 1 0. 55 26 4 38 28 19 –3 6 90 90 0. 99 0. 99 0. 99 0. 98 5b 1, 2 , 4 0. 41 53 5 28 34 34 84 84 0. 98 0. 98 0. 98 0. 99 5c 4, 2 , 1 0. 47 35 2 34 40 40 79 –8 4 79 –8 4 0. 98 0. 96 1. 00 1. 01 5d 2 1. 00 22 4 38 35 25 –4 5 84 84 0. 99 1. 01 0. 96 0. 95 6b 2 1. 00 92 56 46 33 –6 0 79 79 0. 98 1. 01 0. 95 0. 93 6c 4 1. 00 52 6 35 37 37 77 –8 9 77 –8 9 0. 98 1. 02 0. 95 0. 92 6d 4 0. 96 36 1 47 46 46 25 86 –9 7 0. 97 0. 99 0. 95 0. 93 6e 4 1. 00 30 5 58 36 36 81 –9 0 81 –9 1 1. 01 1. 03 0. 99 0. 98 6f 2, 4 , 1 0. 51 51 1 56 31 27 –3 5 69 90 1. 00 1. 01 1. 00 0. 99 7O a 2 0. 68 22 5 43 25 6– 45 90 94 0. 98 1. 00 0. 96 0. 95 7O b 4 1. 00 10 78 35 32 32 64 81 –9 6 0. 97 1. 00 0. 94 0. 92 7O c 4, 3 0. 57 95 4 43 25 25 59 85 –9 9 0. 98 1. 00 0. 95 0. 94 7O d 3, 4 , 2 0. 46 56 8 49 35 35 50 –8 6 86 0. 98 1. 00 0. 96 0. 95 7O e 4 0. 99 21 5 49 25 25 79 –9 0 79 –9 9 1. 00 1. 04 0. 96 0. 93 7P a 4 1. 00 57 4 42 30 30 74 –9 0 74 –9 6 0. 98 1. 05 0. 92 0. 88 7P b 4 1. 00 57 2 32 34 34 73 –8 8 73 –9 3 0. 97 1. 03 0. 92 0. 88 7P c 2 0. 79 90 3 28 17 13 –2 2 89 89 0. 96 0. 98 0. 95 0. 94 7P d 2, 3 0. 71 31 6 49 25 15 –3 4 87 87 0. 98 1. 01 0. 95 0. 93 BC MOOSE POPULATION DYNAMICS – KUZYK ET AL. ALCES VOL. 54, 2018 108 all GMZs. The adult bull:cow ratio in 1996– 2015 varied from 18:100 (GMZ 3b) to 69:100 (GMZ 3d), averaging 42:100 (SD = 12) across all GMZs. Bull:cow ratios were sig- nificantly lower (paired t test: t = 4.48, df = 30, P < 0.001) in 2006–2015 (x = 37:100, SD = 12.3) than in 1996–2005 (x = 48:100, SD = 15.3). Calf:cow ratios varied from 17:100 (GMZ 7Pc) to 49:100 (GMZ 4Wa), averaging 34:100 (SD = 8.0) across all GMZs. Across the 20-year period, calf:cow ratios at 6 months declined in 12 GMZs (39%), juvenile survival from 6–18 months declined in 9 GMZs (29%), and cow survival declined in 10 GMZs (32%). No GMZ experienced an increase in calf:cow ratio, juvenile survival, or annual survival of cows. The annual rate of population change (λ) varied from 0.96 (GMZ 7Pc) to 1.10 (GMZ 4Wa) from 1996– 2015; λ was ≥1 in 24 GMZs (77%) from 1996–2005, but only ≥9 in GMZs (29%) from 2006–2015 (χ2 = 12.7, df = 1, P < 0.001). Only 8 GMZs (3b, 3c, 4Wa, 5c, 8a, 8b, 8c, and 8d) had an average λ > 1 in the most recent 5-year period (2011–2015). The highest modelled population esti- mate across the province occurred in 2002 (258,532) and the lowest in 2015 (169,752) (Fig. 3). Percent change within the 4-time periods was: -29% from 1996–2015 (λ = 0.98), 6.4% from 1996–2005 (λ = 1.01), -32% from 2006–2015 (λ = 0.96), and -20% from 2011–2015 (λ = 0.95). There was a strong correlation between the decline in the population estimate and licenced harvest (r = 0.81, P <0.001). Licenced hunting The average annual licenced harvest during 1996–2015 varied from 18 (GMZ 8c) to 1490 moose per GMZ (GMZ 7Ob; Table 3). Harvest composition was predominantly bulls in all GMZs, with the antlerless harvest >10% of the total harvest in 8 GMZs: 3b (12%), 3c (15%), 3d (10%), 4Wb (31%), 7Oa (32%), Ta bl e 2. (C on tin ue d ) G M Z M od el s1 ,2 w ith Δ A IC c < 2 A IC w ei gh t (w i)3 D en si ty Se x/ A ge R at io s K ey P op ul at io n Pa ra m et er s4 R at e of C ha ng e (λ )5 /1 00 0 km 2 B /1 00 C C a/ 10 0 C R S J S F 19 96 –2 01 5 19 96 –2 00 5 20 06 –2 01 5 20 11 –2 01 5 7P e 2, 1 , 3 0. 38 17 3 57 29 22 -3 6 86 86 0. 98 1. 00 0. 96 0. 95 8a 1 0. 87 59 38 43 43 90 92 1. 09 1. 07 1. 13 1. 12 8b 1 0. 62 20 8 35 35 35 87 87 1. 01 0. 99 1. 02 1. 02 8c 3 0. 74 10 4 42 48 48 26 -9 0 95 1. 07 1. 12 1. 03 1. 02 8d 1 0. 64 91 30 45 45 83 83 1. 02 1. 02 1. 02 1. 02 1 M od el 1 . C al f: co w ra tio a t 6 m on th s (R ), ju ve ni le (S J) a nd c ow s ur vi va l ( S F ) r at es a re c on st an t. M od el 2 . L in ea r t re nd in c al f: co w ra tio a t 6 m on th s. J uv en ile a nd c ow s ur vi va l r at es a re c on st an t. M od el 3 . L in ea r t re nd in ju ve ni le s ur vi va l r at e. C al f: co w ra tio a t 6 m on th s an d co w s ur vi va l r at es a re c on st an t. M od el 4 . L in ea r t re nd in c ow s ur vi va l r at e. C al f: co w ra tio a t 6 m on th s an d ju ve ni le s ur vi va l r at es a re c on st an t. 2 M od el s lis te d st ar tin g w ith lo w es t Δ A IC c v al ue . 3 A IC re la tiv e w ei gh t o f b es t f itt in g m od el . 4 R = R ec ru itm en t r at e (c al ve s: 10 0 co w s) , S J = % c al f s ur vi va l r at e (6 - 18 m th s) , S F = % c ow s ur vi va l r at e (> 18 m th s) . 5 R at e of c ha ng e (λ ) b y tim e pe ri od . λ < 1 .0 a re h ig hl ig ht ed in g re y. ALCES VOL. 54, 2018 KUZYK ET AL. – BC MOOSE POPULATION DYNAMICS 109 7Ob (41%), 7Oc (32%), and 7Od (27%). The harvest rate of bulls relative to the population (bull harvest/pre-hunt population) averaged 5% (range = 2–12%), and the proportional bull harvest rate (bull harvest/pre-hunt bulls) averaged 17% (range = 5–36%). In compari- son, the proportional cow (range = 1–3% of cows) and calf harvest rates (range = 1–9% of calves) were low. Antlerless hunting generally had mini- mal influence on population growth rate. During 1996–2015, the average λp for GMZs 3b, 3d, 7Ob, and 7Oc was 1.00; λh was slightly lower in these and other units (0.97 in 7Ob and 0.99 in 3b, 3c, and 7Oc). When the 1996–2005 growth rate estimates were compared to those in 2006–2015, both λp and λh were > 1 in 1996–2005, and < 1 in Fig. 2. Population rate of change (λ) in 31 Game Management Zones having licenced moose hunting seasons from 1996–2005 (a.) and 2006–2015 (b.); λ values < 1 are highlighted in burgundy and values ≥1 in gray. (Footnote: λ for GMZ 8b was 0.993 from 1996–2005 and 1.024 from 2006–2015. GMZ 5c was 0.961 from 1996–2005 and 1.000 from 2006–2015). (a) (b) BC MOOSE POPULATION DYNAMICS – KUZYK ET AL. ALCES VOL. 54, 2018 110 2006–2015, with the exception of GMZ 3b where λp = 1.00 and λh = 0.99. Antlerless harvest may have reduced population growth rate somewhat during both time periods in certain GMZs, but was not a primary influ- ence in the overall decline of cows during 2006–2015. Modelled adult sex ratios averaged 43 bulls:100 cows during 1996–2015, 49 bulls:100 cows from 1996–2005, 37 bulls:100 cows from 2006–2015, and 35 bulls:100 cows from 2011–2015. Adult sex ratios were <30 bulls:100 cows in 4 GMZs from 1996–2015, 3 GMZs from 1996–2005, 11 GMZs from 2006–2015, and 13 GMZs from 2011–2015 (χ2 [1996–2005, 2006–2015] = 4.5, P = 0.034). Bull harvest rate (R2 = 0.82, P <0.001; Fig. 4) and moose harvest rate (R2 = 0.70, P < 0.001; Fig. 5) were strong pre- dictors of the bull:cow ratio (i.e., # post-hunt bulls:100 cows). A ratio of 30 bulls:100 cows was achieved at a population harvest rate of 6% and bull harvest rate of 23%. DISCUSSION We found that moose populations were declining (λ < 1) in 23% of the GMZs from 1996–2005, and in most (71%) GMZs from 2006–2015. While cow survival was declin- ing in certain GMZs, variation (decline) in the calf:cow ratio at 6 months and juvenile survival from 6–18 months was extensive and likely added to population decline in areas where cow survival was constant (Gaillard et al. 1998). The variation and decline measured in both parameters were consistent with previous studies indicating that their temporal variation is more common than in cow survival (Gaillard et al. 2000, Eberhardt 2002). Further, they may be more influential in local population change (Monteith et al. 2014), especially in variable seasonal environments (Raithel et al. 2007, Hurley et al. 2014). Others conclude that population growth rate is most influenced by calf and juvenile survival in predator-limited populations, especially in summer (Ballard et al. 1991) due to high calf predation by wolves, grizzly bears, and black bears (Gasaway et al. 1992, Boertje et al. 1996, Hayes et al. 2003), and presumably certain regions of BC are similar. Nutritionally- compromised cows also experience reduced pregnancy rate and calf survival (Murray et al. 2006, Schwartz 2007), both of which negatively influence recruitment and popula- tion growth rate. Predation is considered the major factor influencing cow survival in parts of Canada and Alaska (Hauge and Keith 1981, Mytton Fig. 3. Trend in modelled population size of moose (solid circles) and the licensed hunter harvest (open circles) from 1996–2015 in British Columbia, Canada. ALCES VOL. 54, 2018 KUZYK ET AL. – BC MOOSE POPULATION DYNAMICS 111 (C on tin ue d ) Ta bl e 3. S um m ar y of a ve ra ge h ar ve st p ar am et er s ba se d on th e be st m od el s (l ow es t A IC c) , a nd a ss es sm en t o f h un tin g in fl ue nc e (λ p, λ h , a nd b ul l:c ow ra tio ) i n 31 G am e M an ag em en t Z on es fr om 1 99 6– 20 15 in B ri tis h C ol um bi a, C an ad a. G M Z M od el s1 ,2 w ith Δ A IC c < 2 A IC w ei gh t (w i)3 To ta l H ar ve st % H ar ve st C om po si tio n % H ar ve st R at e R at e of C ha ng e4 B ul ls :1 00 C ow s5 B ul ls C ow s C al ve s B ul ls C ow s C al ve s To ta l λ p λ h 19 96 –2 01 5 19 96 –2 00 5 20 06 –2 01 5 20 11 –2 01 5 3b 1, 2 , 4 0. 43 19 1 88 9 3 36 1 1 7 1. 03 1. 02 18 18 19 23 3c 1, 3 0. 57 13 7 85 12 3 26 1 1 6 1. 03 1. 02 29 36 21 21 3d 2 1. 00 97 89 8 2 11 1 0 4 1. 00 0. 99 69 87 51 45 4E a 2 1. 00 10 4 98 1 0 18 0 0 5 1. 05 1. 05 46 63 29 21 4E b 2 1. 00 38 97 1 2 16 0 1 4 1. 03 1. 03 40 44 35 28 4E c 1 0. 60 24 95 5 0 9 0 0 3 0. 97 0. 97 60 72 47 46 4W a 2, 1 0. 61 23 10 0 0 0 14 0 0 4 1. 11 1. 11 55 65 44 43 4W b 4 0. 99 75 69 24 7 18 3 3 7 1. 04 1. 00 49 67 31 29 5a 2, 1 0. 55 13 7 93 7 0 24 1 0 6 1. 02 1. 02 38 54 23 21 5b 1, 2 , 4 0. 41 43 3 99 1 0 25 0 0 6 0. 98 0. 98 28 29 27 26 5c 4, 2 , 1 0. 47 51 9 99 0 1 22 0 0 5 0. 98 0. 98 34 33 36 37 5d 2 1. 00 16 0 10 0 0 0 18 0 0 5 0. 99 0. 99 38 37 38 36 6b 2 1. 00 11 0 99 1 0 11 0 0 3 0. 98 0. 97 56 53 60 60 6c 4 1. 00 10 01 97 2 1 21 0 0 5 0. 99 0. 99 35 36 35 34 6d 4 0. 96 86 99 1 0 6 0 0 1 0. 98 0. 98 47 54 40 38 6e 4 1. 00 38 0 10 0 0 0 10 0 0 3 1. 02 1. 02 58 63 52 52 6f 2, 4 , 1 0. 51 19 1 10 0 0 0 5 0 0 2 1. 00 1. 00 56 56 56 56 7O a 2 0. 68 26 1 68 21 11 24 3 6 9 1. 06 1. 02 43 64 22 18 7O b 4 1. 00 14 90 58 15 26 17 2 9 7 1. 00 0. 97 35 37 33 32 7O c 4, 3 0. 57 80 9 68 15 17 11 1 5 4 1. 00 0. 99 43 53 34 32 7O d 3, 4 , 2 0. 46 42 4 73 16 11 10 1 2 4 0. 99 0. 98 49 50 48 46 7O e 4 0. 99 10 1 10 0 0 0 10 0 0 3 1. 00 1. 00 49 53 45 46 7P a 4 1. 00 37 2 10 0 0 0 12 0 0 3 0. 98 0. 98 42 40 43 42 7P b 4 1. 00 79 8 95 1 3 22 0 1 6 0. 98 0. 97 32 32 31 30 7P c 2 0. 79 36 1 10 0 0 0 14 0 0 3 0. 96 0. 96 28 27 29 28 7P d 2, 3 0. 71 17 2 10 0 0 0 6 0 0 2 0. 97 0. 97 49 45 53 51 7P e 2, 1 , 3 0. 38 18 4 99 0 0 6 0 0 2 0. 98 0. 98 57 56 57 56 BC MOOSE POPULATION DYNAMICS – KUZYK ET AL. ALCES VOL. 54, 2018 112 and Keith 1981, Larsen et al. 1989, Ballard et al. 1991, Gasaway et al. 1992). Given that a third of GMZs had declining cow survival, particularly in remote areas of northern BC, these areas probably reflect a natural preda- tor-prey system with minimal human influ- ence (Ballard et al. 1991). In such systems, the main cause of cow mortality is generally wolf predation (Hauge and Keith 1981, Ballard et al. 1991) with less predation by grizzly bears (Boertje et al. 1988). A current study with radio-collared cows in central BC is evaluating a landscape change hypothesis that assumes cow survival has a greater pro- portional effect on population growth rates than calf survival (Kuzyk and Heard 2014). In that study, annual cow survival rate ranges from 86–92% (Kuzyk et al. 2016) and is within the expected range of stable popula- tions (Bangs et al. 1989, Ballard et al. 1991, Bertram and Vivion 2002), and exceeds rates measured in the Northwest Territories (85%; Stenhouse et al. 1995) and northern Alberta (75–77%; Hauge and Keith 1981). The pri- mary causes of mortality (as of Kuzyk et al. 2016) were predation (43%), health-related (28%), and unlicenced hunting (16%). We found proportionally steeper popula- tion declines in the last 5 years (2011–2015) when 45% of populations were in >20% decline compared to 29% of populations declining ≤20% previously. These acceler- ated declines are consistent with other stud- ies in which ungulate species globally have declined (≤23%) in the past 40 years (Di Marco et al. 2014). For example, recent declines have been reported for mule deer (Lendrum et al. 2013) and pronghorn ante- lope (Antilocapra americana) in North America (Christie et al. 2015), ungulates in Africa (Western et al. 2009), and saiga ante- lope (Saiga tatarica) among other species in the Soviet Union (Milner-Gulland et al. 2001). The level of decline in our study was likely influenced by the rate of early calf Ta bl e 3. (C on tin ue d ) G M Z M od el s1 ,2 w ith Δ A IC c < 2 A IC w ei gh t (w i)3 To ta l H ar ve st % H ar ve st C om po si tio n % H ar ve st R at e R at e of C ha ng e4 B ul ls :1 00 C ow s5 B ul ls C ow s C al ve s B ul ls C ow s C al ve s To ta l λ p λ h 19 96 –2 01 5 19 96 –2 00 5 20 06 –2 01 5 20 11 –2 01 5 8a 1 0. 87 38 99 0 1 40 0 0 12 1. 11 1. 11 38 53 24 25 8b 1 0. 62 86 97 3 0 26 0 0 6 1. 03 1. 02 35 47 24 25 8c 3 0. 74 18 97 3 0 17 0 0 4 1. 09 1. 09 42 56 28 22 8d 1 0. 64 41 99 1 1 28 0 0 6 1. 02 1. 02 30 33 27 28 1 M od el 1 . C al f: co w ra tio a t 6 m on th s (R ), ju ve ni le (S J) a nd c ow s ur vi va l ( S F ) r at es a re c on st an t. M od el 2 . L in ea r t re nd in c al f: co w ra tio a t 6 m on th s. J uv en ile a nd c ow s ur vi va l r at es a re c on st an t. M od el 3 . L in ea r t re nd in ju ve ni le s ur vi va l r at e. C al f: co w ra tio a t 6 m on th s an d co w s ur vi va l r at es a re c on st an t. M od el 4 . L in ea r t re nd in c ow s ur vi va l r at e. C al f: co w ra tio a t 6 m on th s an d ju ve ni le s ur vi va l r at es a re c on st an t. 2 M od el s lis te d st ar tin g w ith lo w es t Δ A IC c. 3 A IC re la tiv e w ei gh t o f b es t f itt in g m od el . 4 λ p = p ot en tia l a nn ua l g ro w th ra te o f f em al es w ith ou t h un tin g, λ h = a nn ua l g ro w th ra te o f f em al es w ith h un tin g. λ < 1 is h ig hl ig ht ed in g re y. 5 B ul l:c ow ra tio s by ti m e pe ri od . A du lt se x ra tio s < 30 b ul ls :1 00 c ow s ar e hi gh lig ht ed in g re y. ALCES VOL. 54, 2018 KUZYK ET AL. – BC MOOSE POPULATION DYNAMICS 113 survival which is directly related to the level of predation (Larsen et al. 1989, Bertram and Vivion 2002, Patterson et al. 2013). Presumably, stochastic environmental events during the 20-year period added to the vari- ability of reproductive rates and juvenile and adult survival rates (Sæther 1997, Gaillard et al. 2000), and that decline in individual GMZs reflected the combined influence of these parameters and local anthropogenic disturbances (Brown 2011). A major challenge for wildlife mana- gers is maintaining sustainable harvests of ungulates in systems with multiple factors influencing population growth (Fryxell et al. 2014), which is especially relevant when populations are declining (Palazy et al. 2012). Moose in BC have high cultural and economic importance, and harvest manage- ment objectives are set to ensure that First Nations requirements are addressed, while maintaining diverse opportunities for licenced hunters (BC FLNRO 2015). Licenced antlerless harvest rates were low during the 20-year period, averaging ~1.5% for cows and 3% for calves. These rates were insufficient to initiate population declines, although may have minor additive effect on Fig. 4. Predictions of the post-hunt, bull:cow ratio based on the modelled bull harvest rate from 31 Game Management Zones in British Columbia, Canada; = −y eˆ 61.595 x0.031 where ŷ is the post-hunt bull:100 cows ratio, and x is the bull harvest rate (bull harvest/pre-hunt bull population). Fig. 5. Predictions of the post-hunt, bull:cow ratio based on the modelled population harvest rate from 31 Game Management Zones in British Columbia, Canada; = −y eˆ 67.339 x0.145 where ŷ is the post- hunt bull:100 cows ratio, and x is the population harvest rate (bull harvest/pre-hunt population of bulls, cows, and calves). BC MOOSE POPULATION DYNAMICS – KUZYK ET AL. ALCES VOL. 54, 2018 114 populations in decline. Our results are sug- gestive of areas in Alaska where predator density has been reduced or moose popula- tions are near their nutritional limitations, such that harvest mortality is partially com- pensatory to natural mortality (Boertje et al. 2007, Boertje et al. 2009). A sustainable har- vest in either system can include a combina- tion of bulls, cows, and calves. However, in areas where predation remains high, hunter harvest is often considered additive and restricted to bull-only harvest to avoid fur- ther population decline (Van Ballenberghe and Dart 1982, Gasaway et al. 1992, Boertje et al. 1996). Although a combination of harvest strat- egies may be required to adapt to variable factors influencing ungulate population growth (Fulton and Huntermark 2004, Stedman et al. 2004), an ongoing challenge is the lack of adequate data to implement and monitor diverse harvest strategies (Bunnefeld et al. 2011), especially with high hunter demand concurrent with declining popula- tions. Ministry policy in BC is to maintain post-hunt, bull:cow ratios ≥30 bulls:100 cows in densities >200 moose/1000 km2, and 50 bulls:100 cows in densities <200 moose/1000 km2 (BC MOE 2010). Although adult sex ratios were > 30 bulls:100 cows in most GMZs, bull:cow ratios were signifi- cantly lower during 2006–2015 (x = 37:100) than 1996–2005 (x = 48:100); declining recruitment may have been a contributing factor. We found significant predictive relation- ships between harvest rates and bull:cow ratios. Although the population harvest rate (bull harvest/pre-hunt population) accounted for 68% of the variance in bull:cow ratios for the time period 2006–2015, it was a poor predictor of adult sex ratios in the other periods. In contrast, the bull harvest rate (bull harvest/pre-hunt bulls) was a good pre- dictor of bull:cow ratios in all time periods. In 2006–2015 we found that average popula- tion harvest rate of 6% and bull harvest rate of 24% resulted in post-hunt, adult sex ratios of 30 bulls:100 cows. In areas with limited survey and monitoring data, our modelling suggests that a maximum population harvest rate of 5% and maximum bull harvest rate of 20% would be sustainable for most popula- tions in BC. These recommendations are similar to those in other northern systems where moose are limited by predation (Hayes et al. 2003). Importantly, calculation of sustainable bull harvest rates must account for bull selectivity in the unlicensed harvest and trends in calf recruitment. Our assessment provides a more rigor- ous and refined determination of moose population and licenced harvest trends in BC than previously available at the provin- cial and regional scales (Hatter 1999, Kuzyk 2016). For example, we found that the provincial moose population declined by 32% (λ = 0.96) from 2006–2015, which more closely aligns with views expressed by many First Nations and other stakehold- ers (Gorley 2016, Kuzyk 2016). Further, we determined that the provincial moose popu- lation trend and licenced harvest were highly correlated, suggesting that declining harvest was a reaction to the declining population, a conclusion similar to those in certain western United States where harvest and populations have declined in the past 15 years (Decesare et al. 2014, Nadeau et al. 2017, Timmermann and Rodgers 2017). Despite our efforts to improve under- standing of moose population trends by using a statistical model-fitting approach (White and Lubow 2002), we acknowledge that limited survey data in certain GMZs could lead to uncertainty in interpretation and extrapolation of our model results; e.g., 4 GMZs had only a single population esti- mate and 2 GMZs had only a single year of ALCES VOL. 54, 2018 KUZYK ET AL. – BC MOOSE POPULATION DYNAMICS 115 sex/age ratio data. Therefore, we used KPUE as a trend index to assist the models in deter- mining population growth rates, especially in certain northern GMZs with limited data. While we acknowledge the limitations of KPUE as a trend index (Crichton 1993, Bowyer et al. 1999, Hatter 2001, DeCesare et al. 2016), particularly in areas with increasing road access, we found similar trends in survey density estimates and KPUE in many GMZs. Another data concern was that the First Nation’s harvest (LeBlanc et al. 2011) was unknown and not quantifiable as an annual harvest statistic. Although field (i.e., radio-telemetry) estimates of cow survival rates have included unlicenced harvest (Kuzyk et al. 2016), sample sizes and distri- bution of study areas on the provincial land- scape were likely insufficient to adequately estimate losses at the provincial scale or account for temporal changes in harvest rate. The amount of unlicenced harvest of bulls and calves is also unknown at the provincial scale, and likely varied spatially and tempo- rally during the 20-year study period. A sys- tem of reliably estimating First Nations harvest would benefit provincial and regional moose population modelling by providing a more complete representation of harvest statistics. And, as with most studies, increas- ing the number and frequency of moose den- sity estimates and composition surveys would also provide more reliable data and population estimates. Despite these limitations, we were able to assess moose population dynamics and licenced harvest trends over a 20-year period suitable for management purposes. We used science-based and repeatable meth- ods to provide an assessment of provincial and GMZ moose populations, their tempo- ral trends, and predictive relationships useful for developing management strate- gies. Our approaches and ability to detect declining population trends will be benefi- cial when compiling future, broad- scale ungulate population trend assessments. ACKNOWLEDGEMENTS We thank D. Heard, M. Gillingham, D. Dupont, and V. Harriman for their review comments that improved this manuscript, and M. Anderson for producing the figures. REFERENCES AlfAro, r. I., l. vAn Akker, and B. HAwkes. 2015. Characteristics of forest legacies following two mountain pine beetle outbreaks in British Columbia, Canada. Canadian Journal of Forest Research 45: 1387–1396. BAllArd, w. B., J. s. wHItmAn, and d. J. reed. 1991. Population dynamics of moose in South-Central Alaska. Wildlife Monographs 114. BAngs, e. e., t. n. BAIley, and m. f. Portner. 1989. Survival rates of adult female moose on the Kenai Peninsula, Alaska. Journal of Wildlife Management 53: 557–563. BertrAm, m. r., and m. t. vIvIon. 2002. Moose mortality in eastern interior Alaska. Journal of Wildlife Management 66: 747–756. Boer, A. H. 1992. Fecundity of North American moose (Alces alces): a review. Alces Supplement 1 (1992): 1–10. _____, and D. M. kePPIe. 1988. Modelling a hunted moose population in New Brunswick. Alces 24: 201–217. BoertJe, r. d., w. C. gAsAwAy, d. v. grAngAArd, and d. g. kelleyHouse. 1988. Predation on moose and caribou by radio-collared grizzly bears in east central Alaska. Canadian Journal of Zoology 66: 2492–2499. _____, m. A. keeCH, d. d. young, k. A. kellIe, and C. t. seAton. 2009. Managing for elevated yield of moose in interior Alaska. Journal of Wildlife Management 73: 314–327. BC MOOSE POPULATION DYNAMICS – KUZYK ET AL. ALCES VOL. 54, 2018 116 _____, k. A. kellIe, C. t. seAton, m. A. keeCH, d. d. young, B. w. dAle, l. g. AdAms, and A. r. AdermAn. 2007. Ranking Alaska moose nutrition: signals to begin liberal antlerless harvest. Journal of Wildlife Management 71: 1494–1506. _____, P. vAlkenBurg, and m. mCnAy. 1996. Increases in moose, caribou, and wolves following wolf control. Journal of Wildlife Management 60: 474–489. BomAn m., l. mAttsson, g. erICsson, and B. krIstrom. 2011. Moose hunting val- ues in Sweden now and two decades ago: the Swedish hunters revisited. Environmental and Resource Economics 50: 515–30. Bowyer, r. t., m. C. nICHolson, e. m. molvAr, and J. B. fAro. 1999. Moose on Kalgin Island: are density-dependent pro- cesses related to harvest? Alces 35: 73–89. BrItIsH ColumBIA mInIstry of envIronment (BC MOE). 2010. Moose Harvest Management Procedure Manual. Fish and Wildlife Branch, Victoria, British Columbia, Canada. BrItIsH ColumBIA mInIstry of forests, lAnds And nAturAl resourCe oPerAtIons (BC FLNRO). 2015. Provincial Framework for Moose Management in British Columbia. Fish and Wildlife Branch, Victoria, British Columbia, Canada. Brown, g. s. 2011. Patterns and causes of demographic variation in a harvested moose population: evidence for the effects of climate and density-dependent drivers. Journal of Animal Ecology 80: 1288–1298. Bunnefeld, n. e. HosHIno, and e. J. mIlner-gullAnd. 2011. Management strategy evaluation: a powerful tool for conservation? Trends in Ecology and Evolution 26: 441–447. BurnHAm, k. P., and d. s. r. Anderson. 2002. Model Selection and Multimodel Inference: A Practical Information- Theoretic Approach, 2nd edition. Springer- Verlag, New York, New York, USA. CHrIstIe, k. s., w. f. Jensen, J. H. sCHmIdt, and m. s. BoyCe. 2015. Long-term changes in pronghorn abundance index linked to climate and oil development in North Dakota. Biological Conservation 192: 445–453. CrICHton, v. 1993. Hunter effort and obser- vations – the potential for monitoring trends of moose populations – a review. Alces 29: 181–185. deCesCAre, n. J., J. r. newBy, v. J. BoCCAdorI, t. CHIlton-rAd, t. t. tHIer, d. wAltee, k. Podruzny, and J. A. gude. 2016. Calibrating minimum counts and catch-per-unit-effort as indices of moose population trend. Wildlife Society Bulletin 40: 537–47. _____, t. d. smuCker, r. A. gArrot, and J. A. gude. 2014. Moose status in Montana. Alces 50: 35–51. dImArCo, m., m. l. BoItAnI, d. mAllon, m. HoffmAnn, A. IACuCCI, e. meIJAArd, P. vIsContI, J. sCHIPPer, and C. rondInInI. 2014. A retrospective evalua- tion of the global decline of carnivores and ungulates. Conservation Biology 28: 1109–1118. eAstmAm, d., and r. rItCey. 1987. Moose habitat relationships and management in British Columbia. Swedish Wildlife Research Supplement 1: 101–117. eBerHArdt, l. l. 2002. A paradigm for pop- ulation analysis of long-lived vertebrates. Ecology 83: 2841–2854. fryxell, J. m., w. e. merCer, and r. B. gellAtely. 1988. Population dynamics of Newfoundland moose using cohort analysis. Journal of Wildlife Management 52: 14–21. _____, A. r. e. sInClAIr, and g. CAugHley. 2014. Wildlife Ecology, Conservation, and Management. John Wiley and Sons, New York, New York, USA. fulton, d. C., and k. HundertmArk. 2004. Assessing the effects of a selective har- vest system on moose hunters’ behav- iors, beliefs, and satisfaction. Human Dimensions of Wildlife 9: 1–16. ALCES VOL. 54, 2018 KUZYK ET AL. – BC MOOSE POPULATION DYNAMICS 117 gAIllArd, J.-m., m. festA-BIAnCHet, and n. g. yoCCoz. 1998. Population dynamics of large herbivores: variable recruitment with constant adult survival. Trends in Ecology and Evolution 13: 58–63. _____, _____, _____, A. loIson, and C. toIgo. 2000. Temporal variation in fit- ness components and population dynam- ics of large herbivores. Annual Review of Ecology and Systematics 31: 367–393. gAsAwAy, w. C., r. d. BoertJe, d. grAngAArd, d. g. kellyHouse, r. o. stePHenson, and d. g. lArsen. 1992. The role of predation in limiting moose at low densities in Alaska and Yukon and implications for conservation. Wildlife Monographs 120. _____, s. d. duBoIs, d. J. reed, and s. J. HArBo. 1986. Estimating moose popula- tion parameters from aerial surveys. Biological Papers of the University of Alaska, Number 22. Institute of Arctic Biology, University of Alaska, Fairbanks, Alaska, USA. _____, r. o. stePHenson, J. l. dAvIs, P. e. k. sHePArd, and o. e. BurrIs. 1983. Interrelationships of wolves, prey and man in interior Alaska. Wildlife Monographs 84. gorley, A. 2016. A strategy to help restore moose populations in British Columbia. Prepared for the Ministry of Forests Lands and Natural Resource Operations. Fish and Wildlife Branch, Victoria, British Columbia, Canada. HAddon, m. 2001. Modelling and Quantitative Methods in Fisheries. Chapman & Hall/ CRC, London, United Kingdom. HAtter, I. w. 1998. A Bayesian approach to moose population assessment and har- vest decisions. Alces 34: 47–58. _____. 1999. An evaluation of moose har- vest management in central and northern British Columbia. Alces 35: 91–103. _____. 2001. An assessment of catch per unit effort to estimate rate of change in deer and moose populations. Alces 37: 71–77. HAuge, t. m., and l. B. keItH. 1981. Dynamics of moose populations in north- eastern Alberta. Journal of Wildlife Management 45: 573–597. HAyes, r. d., r. fArnell, r. m. P. wArd, J. CArey, m. deHn, g. w. kuzyk, A. m. BAer, C. l. gArdner, and m. o’donogHue. 2003. Experimental reduction of wolves in the Yukon: ungu- late responses and management implica- tions. Wildlife Monographs 152. HeArd, d. C., A. B. d. wAlker, J. B. Ayotte, and g. s. wAtts. 2008. Using GIS to modify a stratified random block survey design for moose. Alces 44: 111–116. Hurley, m. A., m. HeBBlewHIte, J. m. gAIllArd, s. drAy, k. A. tAylor, w. k. smItH, P. zAger, and C. BonenfAnt. 2014. Functional analysis of normalized difference vegetation index curves reveals overwinter mule deer survival is driven by both spring and autumn phe- nology. Philosophical Transactions of the Royal Society B 369 (1643): 20130196. kArns, P. d. 2007. Population distribution, density and trends. Pages 125–139 in A. W. Franzmann and C.C. Schwartz, edi- tors. Ecology and Management of the North American Moose, 2nd Edition. University Press of Colorado, Boulder, Colorado, USA. kelsAll, J. P. 1987. The distribution and status of moose (Alces alces) in North America. Swedish Wildlife Research Supplement 1: 1–10. kuzyk, g. w. 2016. Provincial population and harvest estimates of moose in British Columbia. Alces 52: 1–11. _____, and I. w. HAtter. 2014. Using ungu- late biomass to estimate abundance of wolves in British Columbia. Wildlife Society Bulletin doi: 10.1002/wsb.475. _____, and d. HeArd. 2014. Research design to determine factors affecting moose population change in British Columbia: testing the landscape change hypothesis. Wildlife Bulletin No. B-126. BC MOOSE POPULATION DYNAMICS – KUZYK ET AL. ALCES VOL. 54, 2018 118 British Columbia Ministry of Forest, Lands and Natural Resource Operations. Victoria, British Columbia, Canada. _____, s. mArsHAll, m. klACzek, C. ProCter, B. CAdsAnd, H. sCHIndler, and m. gIllIngHAm. 2016. Determining factors affecting moose population change in British Columbia: testing the landscape change hypothesis. Wildlife Working Report No. WR-123. Progress Report: February 2012 – 30 April 2016. British Columbia Ministry of Forest, Lands and Natural Resource Operations, Victoria, British Columbia, Canada. lArsen, d. g., d. A. gAutHIer, and r. l. mArkel. 1989. Causes and rate of moose mortality in southwest Yukon. Journal of Wildlife Management 53: 548–557. leBlAnC, J. e., B. e. mClAren, C. PereIrA, m. Bell, and s. AtlookAn. 2011. First Nations moose hunt in Ontario: a com- munity’s perspectives and reflections. Alces 47: 163–174. lenArz, m. s., m. e. nelson, m. w. sCHrAge, and A. J. edwArds. 2009. Temperature mediated moose survival in northeastern Minnesota. Journal of Wildlife Management 73: 503–10. lendrum, P. e., C. r. Anderson Jr., k. l. monteItH, J. A. Jenks, and r. t. Bowyer. 2013. Migrating mule deer: effects of anthropogenically altered landscapes. PLoS One 8: e64548. meIdInger, d., and J. PoJAr. 1991. Ecosystems of British Columbia. British Columbia Ministry of Forests, Special Report Series Number 6. British Columbia Ministry of Forests, Victoria, British Columbia, Canada. mIlner-gulland, e. J., m. v. kHolodovA, A. Bekenov, o. m. BukreevA, I. A. grACHev, l. AmgAlAn, and A. A. lusHCHekInA. 2001. Dramatic declines in saiga antelope populations. Oryx 35: 340–345. monteItH k. l., v. C. BleICH, t. r. stePHenson, B. m. PIerCe, m. m. Conner, J. g. kIe, and r. t. Bowyer. 2014. Life-history characteristics of mule deer: effects of nutrition in a vari- able environment. Wildlife Monographs 186. mowAt, g., d. C. HeArd, and C. J. sCHwArtz. 2013. Predicting grizzly bear density in western North America. PloS One 8: e82757. doi: 10.1371/journal. pone.0082757 murrAy, d. l., e. w. Cox, w. B. BAllArd, H. A. wHItlAw, m. s. lenArz, t. w. Custer, t. BArnett, and t. k. fuller. 2006. Pathogens, nutritional deficiency, and climate change influences on a declining moose population. Wildlife Monographs 166. mytton, w. r., and l. B. keItH. 1981. Dynamics of moose populations near Rochester, Alberta, 1975–1978. Canadian Field-Naturalist 95: 39–49. nAdeAu, m. s., n. J. deCesAre, d. g. BrImeyer, e. J. BergmAn, r. B. HArrIs, k. r. Hersey, k. k. HueBner, P. e. mAttHews, and t. P. tHomAs. 2017. Status and trends of moose populations and hunting opportunity in the western United States. Alces 53: 99–112. PAlAzy, l., C. BonenfAnt, J. m. gAIllArd, and f. CourCHAmP. 2012. Rarity, tro- phy hunting and ungulates. Animal Conservation 15: 4–11. PAtterson, B. r., J. f. Benson, k. r. mIddel, k. J. mIlls, A. sIlver, and m. e. oBBArd. 2013. Moose calf mortality in central Ontario, Canada. Journal of Wildlife Management 77: 832–841. Peters, w., m. HeBBlewHIte, k. g. smItH, s. m. weBB, n. weBB, m. russe, C. stAmBAugH, and r. B. Anderson. 2014. Contrasting aerial moose survey popula- tion estimate methods and evaluating sightability in West-Central Alberta, Canada. Wildlife Society Bulletin 38: 639–649. Peterson, r. o. 1977. Wolf ecology and prey relationships on Isle Royale. National Park Service Scientific Monograph Series, No. 11. ALCES VOL. 54, 2018 KUZYK ET AL. – BC MOOSE POPULATION DYNAMICS 119 QuAyle, J. f., A. g. mACHutCHon, and d. n. Jury. 2001. Modeling moose sight- ability in south-central British Columbia. Alces 37: 43–54. rAItHel, J. d., m. k. kAuffmAn, and d. H. PletsCHer. 2007. Impact of spatial and temporal variation in calf survival on the growth of elk populations. Journal of Wildlife Management 71: 795–803. resourCes InformAtIon stAndArds CommIttee (RISC). 2002. Aerial-based inventory methods for selected ungu- lates: bison, mountain goat, mountain sheep, moose, elk, deer and caribou. Standards for components of British Columbia’s biodiversity No. 32, Version 2.0. British Columbia Ministry of Sustainable Resource Management, Victoria, British Columbia, Canada. rItCHIe, C. 2008. Management and chal- lenges of the mountain pine beetle infestation in British Columbia. Alces 44: 127–135. sAetHer, B. e. 1997. Environmental stochas- ticity and population dynamics of large herbivores: a search for mechanisms. Trends in Ecology and Evolution 12: 143–149. sCHwArtz, C. C. 2007. Reproduction, natal- ity, and growth. Pages 141–171 in A. W. Franzmann and C. C. Schwartz, editors. Ecology and Management of the North American Moose. University Press of Colorado, Boulder, Colorado, USA. sHACkleton, d. 1999. Hoofed Mammals of British Columbia. Royal British Columbia Museum Handbook. University of British Columbia Press, Vancouver, Canada. skAlskI, J. r., k. e. rydIng, and J. J. mIllsPAugH. 2005. Wildlife Demography: Analysis of Sex, Age and Count Data. Elsevier Academic Press, Cambridge, Massachusetts, USA. sPAldIng, d. J., and J. lesowskI. 1971. Winter food of the cougar in south- central British Columbia. Journal of Wildlife Management 35: 378–381. stedmAn, r., d. r. dIefenBACH, C. B. swoPe, J. C. fInley, A. e. luloff, H. C. zInn, g. J. sAn JulIAn, and g. A. wAng. 2004. Integrating wildlife and human-dimensions research methods to study hunters. Journal of Wildlife Management 68: 762–773. stenHouse, g. B., P. B. lAtour, l. kutny, n. mACleAn, and g. glover. 1995. Productivity, survival, and movements of female moose in a low-density popu- lation, Northwest Territories, Canada. Arctic 48: 57–62. telfer, e. s. 1984. Circumpolar distribution and habitat requirements of moose (Alces alces). Pages 145–182 in R. Olson, R. Hastings, and F. Geddes, editors. Northern Ecology and Resource Management. University of Alberta Press, Edmonton, Alberta, Canada. tImmermAnn, H. r. 1992. Moose sociobiol- ogy and implications for harvest. Alces 28: 59–77. _____, and A. r. rodgers. 2017. The status and management of moose in North America – circa 2015. Alces 53: 1–22. vAn BAllenBerge, v., and J. dArt. 1982. Harvest yields from moose populations subject to wolf and bear predation. Alces 18: 258–275. western, d., s. russell, and I. CutHIll. 2009. The status of wildlife in protected areas compared to non- protected areas of Kenya. PLoS ONE 4: e6140. wHIte, g. C., and B. C. luBow. 2002. Fitting population models to multiple sources of observed data. Journal of Wildlife Management 66: 300–309. _Hlk524694781 pone.0163249.ref001