51 BIRTH RATES AND NEONATE SURVIVAL IN A PARASITE RICH MOOSE POPULATION IN VERMONT, USA Jacob DeBow1,5, Joshua Blouin1, Elias Rosenblatt1, Cedric Alexander2, Nicholas Fortin2, Katherina Gieder2, James Murdoch3, and Therese Donovan4 1Vermont Cooperative Fish and Wildlife Research Unit, Rubenstein School of Environmental and Natural Resources, University of Vermont, 302 Aiken Center, 81 Carrigan Drive, Burlington, Vermont 05405, USA; 2Vermont Fish and Wildlife, 271 North Main Street, Suite 215, Rutland, Vermont 05701, USA; 3Wildlife and Fisheries Biology Program, Rubenstein School of Environmental and Natural Resources, University of Vermont, 302 Aiken Center, 81 Carrigan Drive, Burlington, Vermont 05405, USA; 4U.S. Geological Survey, Vermont Cooperative Fish and Wildlife Research Unit, 302 Aiken Center, 81 Carrigan Drive, Burlington, Vermont 05405, USA; 5current address: New Hampshire Fish and Game Department, 629B Main Street, Lancaster, New Hampshire 03584, USA. ABSTRACT: Moose (Alces alces) populations are declining across much of their southern geographic range in North America. In Vermont and other northeastern states, measurable declines are attributed to low calf survival and reduced productivity associated with persistent winter tick (Dermacentor albipictus) parasitism. In 2017–2020, we studied 75 radio-collared female moose (38 calves and 37 adults) in Vermont to examine physiological, spatial, and temporal parameters relative to calf survival and adult productivity. Physiological measures included concentration of fecal glucocorticoid metab- olites (fGCM) which reflects stress, and urea nitrogen:creatinine ratios in urine (UN:C) which proxy nutritional state. The pregnancy rate at capture across years was 0.67 (95% CI = 0.50 – 0.80), and was negatively related to presence of lungworm (Dictyocaulus spp.). The birth rate calculated as the aver- age number of offspring delivered per adult female was <1.0 overall (2017–2020, LCI = 0.22, UCI = 0.86), similar across years, but increased with age. Logistic exposure models indicated that daily calf survival to 60 d increased as Julian birth date and days since birth increased (log odds = 0.0819, SE = 0.0215). The per capita independence rate, or rate that adult females add independent calves to the population, was negatively related to UN:C ratios and positively with fGCM. Further, this rate was related to autumnal habitat use of adult females; it was greater in home ranges character- ized by large amounts of mature (canopy) evergreen forests and wetland habitats, and small amounts of mixed forests and elevation than in ranges with abundant levels of mixed forest at high elevation. We conclude that winter ticks can negatively affect moose fecundity, and efforts to reduce host (moose) density through harvest or parasite (host) abundance through habitat manipulation may improve pro- ductivity and recruitment in local moose populations. ALCES VOL. 58: 51–73 (2022) Key words: Alces alces, birth rate, calf survival, moose, parasitism, population decline, winter tick The moose (Alces alces) is an iconic spe- cies ecologically, economically, and cultur- ally important in many regions including the northeastern United States where a major population expansion began in the 1970s in Maine, New Hampshire, and Vermont (Wattles and DeStefano 2011). However, populations in these states have declined measurably in the last 10–15 years, except in much of central and northern Maine (Timmermann and Rodgers 2005, 2017, Franzmann and Schwartz 2007, Jensen et al. 2018). More specifically, moose at their southern range boundary in northern Vermont and New Hampshire have undergone pur- poseful harvest reductions (VFWD 2009, 52 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. ALCES VOL. 58, 2022 Musante et al. 2010, Jones et al. 2017, Timmermann and Rodgers 2017, Ellingwood et al. 2019, Jones et al. 2019). The Vermont population peaked at ~ 4,800 individuals in 2005, with highest abundance and density in the northeast corner of the state (VFWD 2009). In response, increased hunting per- mits were allocated to maintain densities at ecological and cultural carrying capacity (VFWD 2009). However, over the past decade this population has continued to decline (45% since 2010) below the manage- ment goal despite reducing hunting permits. By 2017, the statewide population was only ~1,665 moose based on hunter sighting reports (VFWD 2018). Survival analyses of radio-collared moose from 2017 to 2019 indicated that calf survival from January to May was low (<0.50; DeBow et al. 2021), and associated with winter tick (Dermacentor albipictus) parasitism and other internal parasites like meningeal worm (Parelaphostrongylus tenuis) and lungworm (Dictyocaulus spp.) (DeBow et al. 2021, Rosenblatt et al. 2021). Winter tick epizootics, documented in Vermont as part of a larger study, refer to years in which severe parasitism results in calf mortality rates ≥0.5 within a given popu- lation (Musante et al. 2007). During their questing season in autumn, winter tick larvae attach to moose and take their last and largest blood meal in March–April when calves typ- ically succumb to malnutrition and acute anemia (Musante et al. 2007). The adult females detach after this last meal, lay eggs in the leaf litter, and their subsequent larvae quest in autumn at these sites (Samuel 2004). Blouin et al. (2021a) found that habitat selec- tion by maternal cows during the autumnal questing period (September–November) influenced winter survival of its calf. Birth rate, defined as the number of calves produced per adult female per year, is a function of age, body size, and individual health during the autumn breeding season (Saether and Haagenrud 1983, Testa and Adams 1998) with larger females producing calves earlier and more frequently (Adams and Pekins 1995, Testa and Adams 1998). Moose have a relatively low reproductive rate producing single calves most often, with twinning and triplets possible in ideal habitat conditions and population expansion (Boer 1992). For example, historic twinning rates in Vermont from 1993 to 2015 were ~ 21% (VFWD 2018) but twinning is rare currently. Yearlings (2-year-olds when birthing) are capable of reproduction but have a threshold of 200kg (field-dressed body weight), below which they rarely ovulate (Saether and Heim 1993, Adams and Pekins 1995, Sand 1996). Adult twinning and yearling pregnancy rates are useful indicators of the overall health of a population as they potentially indicate habitat status and/or resource limitation (Franzmann and Schwartz 1985, Boer 1992, Adams and Pekins 1995, Eberhardt 2002, Boertje et al. 2019). However, in adjacent New Hampshire where epizootics occurred in 3 successive years (2014–2016), yearling pregnancy and twinning rates were 0% and the adult birth rate was only 51% in optimal habitat (Jones et al. 2017). The survival rate of newborns influences population dynamics of large herbivores (Gaillard et al. 2000), with moose calves (<1 year of age) undergoing three distinct stages of development (Franzmann and Schwartz 2007). “Prenatal” is defined as pre- birth or gestational when growth is dependent on the nutritional condition of the dam. The “neonate” stage occurs in the initial 1.5–2 months of life when growth and survival are largely dependent upon the mother’s milk and direct care (Schwartz 1992b, 1992c). “Independence” is when the calf is no longer solely reliant on the direct care of its mother and weaning is mostly complete. Identifying and understanding factors that promote 53 ALCES VOL. 58, 2022 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. survival of neonates to independence is espe- cially important when annual recruitment is low as in the Vermont moose population. The rate at which independent calves enter a population strongly influences long- term population viability (Schwartz 1992b, 1992c). This rate, hereafter the per capita independence rate, encompasses pregnancy, birth, and neonate survival rates in aggregate, and may be measured at the population level as an annual rate or at the individual level. Highly stressed and/or nutritionally deprived adult females are less likely to bring a calf to independence (DelGiudice 1995). Further, pregnant cows that carry a high winter tick load are compromised physiologically, and are likely more susceptible to other diseases, parasites, and predation (Murray et al. 2006, Musante et al. 2007, Pekins 2020). Given that these effects may be cumulative and carry through time (Wingfield et al. 1998, Keech et al. 2000, Parker 2003, Evans et al. 2006), sustained winter tick parasitism is potentially a long-term, disruptive impact to population recovery in Vermont and New Hampshire. In response, we monitored female moose to evaluate factors influencing reproduction and productivity in a population with heavy win- ter tick parasitism in northeastern Vermont. Our primary objectives were to estimate: 1) pregnancy rate, 2) per capita birth rate, 3) neonate survival to independence (60-days post-partum), and 4) the per capita indepen- dence rate. To better interpret these rates, we also evaluated physiological condition of moose in winter and their habitat use pat- terns during the autumnal questing period of winter ticks. STUDY AREA We conducted the study in northeastern Vermont, USA, from 2017 to 2020 within Wildlife Management Units (WMU) E1 and E2 (Fig. 1) due to the relatively high density of moose (~0.39 moose/km2; VFWD 2009). The study area occurred primarily within Essex County, encompassed 21 townships, and covered 1,738 km2. The outer boundar- ies were U.S. Rt. 2 north to the Canadian border and Rt. 114 east to the Connecticut River; WMU E1 and E2 are separated by VT Rt. 105 with E1 to the north and E2 to the south (Fig. 1). Forest composition of the study area was largely northern hardwoods. Yellow birch (Betula alleghaniensis), sugar maple (Acer saccharum), and American beech (Fagus grandifolia) were at mid and higher eleva- tions, with red spruce (Picea rubens) and balsam fir (Abies balsamea) on most ele- vated peaks. Lowlands and wet areas were primarily comprised of black spruce (Picea mariana), balsam fir, eastern white cedar (Thuja occidentalis), and speckled alder (Alnus incana). The study area included both federal and state managed forests, and pri- vately owned, actively managed commercial forest. These large ownerships created a mosaic of forest age classes with a constant presence of regenerating forest due to com- mercial forest harvests (Blouin et al. 2021b). Smaller parcels of private lands were con- centrated along the edge of the study area. Predation pressure on moose is generally low given the extirpation of cougars (Puma concolor) and wolves (Canis lupus) (DeGraaf and Yamasaki 2001), and few records exist of predation by other large carnivores, including black bear (Ursus americanus) and coyotes (Canis latrans) despite their common occur- rence throughout the study area (DeGraaf and Yamasaki 2001, Musante et al. 2010, DeBow et al. 2021). Vermont’s moose harvest is a lot- tery-based permit system, with total permits set by the Vermont Fish and Wildlife Department (VWFD) to meet population objectives. During the study (2017–2019), a moose hunt was held only in 2017 and 2018 with 40 and 10 moose harvested, respectively (VFWD 2020). Competition from white-tailed 54 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. ALCES VOL. 58, 2022 Fig. 1. Study area (1,738 km2) for assessing moose reproductive performance (2017–2019) in northeastern Vermont, USA. The northern boundary adjoins Quebec, Canada and the eastern boundary adjoins New Hampshire, USA. White circles represent locations of radio-collared animals (n = 126). Crosses (n = 4) represent locations of weather stations. 55 ALCES VOL. 58, 2022 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. deer (Odocoileus virginianus), the only other ungulate, was presumably minimal given the abundance of forage and low deer density (~7/mile2; VFWD 2020). We defined annual seasons as summer (June–August), fall (September– November), winter (December–February), and spring (March–May). Ambient tem- perature can reach 32 °C (~17 °C average) in summer and −40°C (~ −8°C average) in winter; average daily temperature was ~11 °C (National Climate Data Center [NCDC] 2019). Annual precipitation ranged from 100 to 110 cm per year (NCDC 2019) with average snow fall ranging 220–250 cm annually (NCDC 2019); depth varies with elevation. METHODS Capture and Health Assessment We measured pregnancy rates, birth rates, and neonate survival rates by capturing and radio-collaring female moose. We radio-collared adult cows (age ≥ 2) and calves (age ~ 6.5 months) and monitored them from 2017 to 2019 (Fig. 1). Less intensive monitor- ing occurred in 2020 when neonate survival to 60 d was recorded once in August (see beyond – Per Capita Independence Rate). Animals were captured by aerial net-gunning by heli- copter and physical restraint during 3–6 day periods in January, 2017–2019 (Native Range Capture Services, Elko, Nevada). Aerial dart- ing (3 mg Carfentanil, reversed by 300 mg Naltrexone; ZooPharm, Windsor Colorado, USA) was used only 3 times in the rare cir- cumstance where netting posed a threat to the health of a target individual. We fit each moose with a Survey Globalstar V7.1 GPS collar (VECTRONIC Aerospace GmbH, Berlin, Germany) equipped with a very high frequency (VHF) beacon that trans- mitted from 0600 to 1800 hr and a global posi- tioning system (GPS) sensor that transmitted a GPS location every 13 h. The GPS fix interval was chosen to ensure fixes would be staggered through time. Collars weighed ~ 0.85 kg; adult collars were sized to the individual, whereas calf collars were sized to 96 cm and retrofitted to individual size with duct tape and medical latex cord to allow for expansion at ~1 year as described in Musante et al. (2010). Collars were not designed to detach. Animal capture and handling followed guidelines of the American Society of Mammologists (Sikes et al. 2016) and all protocols were approved by the University of Vermont Institutional Animal Care and Use Committee (protocol #17-035). Biological samples and physical mea- surements were collected on all captured ani- mals to gauge individual condition. A body condition score was assigned from palpation of the spine, ribs, and rump: 1-very thin, 2-thin, 3-normal, 4-fat (Hundertmark and Schwartz 1998). Tick abundance was indexed from counts at the shoulder and rump where ticks were counted on four, 10 cm transects in each area (Bergeron and Pekins 2014). Blood (30 mL) was collected from the jugular vein in two, 10 mL clot activator vacutainers and a 10 mL EDTA vacutainer; clot activator vacutainers were spun in a centrifuge for serum and EDTA vacutainers were inverted immediately to ensure mixing. A fecal sample was collected from the rectum for screening to identify and estimate internal parasites at the University of Maine Animal Health Laboratory, Orono, Maine, USA. The McMasters flotation technique was used to determine the number of eggs/g fecal matter for tapeworms (Moniezia sp.) and round- worms (Nematodirus sp., Coccidia spp., and Strongylidea sp.), and the Baermann method was used to identify and quantify the number of lungworm larvae (Dictyocaulus sp.). Pregnancy Rate The pregnancy-specific protein-B test was used with serum samples to measure preg- nancy status at BioPRYN WILD, Moscow, 56 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. ALCES VOL. 58, 2022 Idaho, USA; accuracy is 99% in non-pregnant cows and 93–95% in pregnant cows. Due to small sample sizes, we evaluated 6 univariate logistic regression models to estimate the prob- ability of pregnancy at capture as a function of individual covariates (Table 1). Individual covariates included body condition score at capture, year of capture, fecal lungworm, roundworm (Nematodirus sp.), eggs per gram (EPG), and tapeworm counts. An intercept model was used to estimate the average rate of pregnancy across cows. Logistic regression analyses were conducted with the glm() func- tion in R (R Core Team 2017), and compared Table 1. Covariates used to assess birth rate, daily survival of neonates, and recruitment of calves to day 60 for moose in northeastern Vermont, USA (2017–2020). Objective number refers to the specifically numbered objectives in the introduction section. Predicted effects are predicted negative (-) or positive (+) effects of a given variable. Covariate Name Description Objective Predicted Effect References Individual Adults Year + Year of capture 1 +/- Adams and Pekins 1995 Condition Body condition score of each adult cow at capture (1 = very thin, 2 = thin, 3 = normal, 4 = fat). 1 + Bergeron et al. 2013; Hundertmark and Schwartz 1998 EPG Total parasite eggs per gram of fecal matter observed using the modified McMaster flotation technique. 1 - Murray et al. 2006 Lungworm Total number of lungworm (Dictyocaulcus sp.) eggs observed in feces using the modified McMaster flotation technique. 1 - Murray et al. 2006 Roundworm Total number of roundworm (Nematodirus sp., Cocidia spp., and Strongylidea sp.) eggs observed in feces using the modified McMaster flotation technique. 1 - Murray et al. 2006 Ticks at Capture Individual tick counts from capture. This is the total number of ticks counted on four 10 cm transects on the shoulder and rump 1 - Samuel 2004 Age + Age of adult cow during a given year (2+, 3+, 4+). 2 + Adams and Pekins 1995 UNC Urea nitrogen (UN) and creatinine (C; mg/dl) ratio obtained from urine collected in snow during winter months. 4 - Rosenblatt et al. 2021 fGCM Glucocorticoid metabolite concentrations (fGCM) collected from fecal pellets obtained during winter months. 4 + Rosenblatt et al. 2021 QPC1 First principal component describing female core home range habitat during the peak winter tick fall questing period. See Table 2. 4 0 Blouin et al. 2021a Table 1 (continued) 57 ALCES VOL. 58, 2022 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. with AIC model selection methods (Burnham and Anderson 2002). Per Capita Birth Rate We measured birth rates in 2017–2019 from field observations (hereafter “walk-ins”) conducted on all radio-collared adult cows, including calves that survived to reproductive age (≥2 years old). Neonates were tracked on foot with handheld telemetry equipment from 01 May to 31 July. Each animal was tracked to within sighting distance (typically ~ 20–50 m) Table 1 (continued). Covariates used to assess birth rate, daily survival of neonates, and recruitment of calves to day 60 for moose in northeastern Vermont, USA (2017–2020). Objective number refers to the specifically numbered objectives in the introduction section. Predicted effects are predicted negative (-) or positive (+) effects of a given variable. Covariate Name Description Objective Predicted Effect References QPC2 Second principal component describing female core home range habitat during the peak winter tick fall questing period. See Table 2. 4 - Blouin et al. 2021a QPC3 Third principal component describing female core home range habitat during the peak winter tick fall questing period. See Table 2. 4 + Blouin et al. 2021a QPC4 Fourth principal component describing female core home range habitat during the peak winter tick fall questing period. See Table 2. 4 - Blouin et al. 2021a Interval Neonates Birth date Date of birth for neonate moose as recorded by direct observation. 3 + Jones et al. 2017 Days Since Birth Number of days since neonate was born. 3 + Jones et al. 2017 Collar temperature Average temperature (Celsius) as recorded by the radio-collar affixed to an adult female. 3 - Schwartz 2007; McCann et al. 2013 Precipitation Average daily precipitation during the breeding season (15 Sep– 15 Oct) of the breeding season prior to obtaining pregnancy test. 3 +/- National Climate Data Center Distance traveled Average daily linear distance traveled between GPS points during exposure period 3 - Blouin et al. 2021b Mean Temp Average temperature (Celsius) of observation interval as estimated from the National Climate Data Center. 3 - Schwartz 2007; McCann et al. 2013 Min Temp Minimum temperature (Celsius) of observation interval as estimated from the National Climate Data Center. 3 - Schwartz 2007; McCann et al. 2013 Max Temp Maximum temperature (Celsius) of observation interval as estimated from the National Climate Data Center. 3 + Schwartz 2007; McCann et al. 2013 58 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. ALCES VOL. 58, 2022 to verify presence of a calf; care was taken to minimize disturbance. Previous fieldwork in New Hampshire and Maine (Jones et al. 2017) and our observations indicate that walk-ins have no apparent effect on calf survival or abandonment. Visual observation of each cow was attempted 2–3 times weekly each summer season. Presence of a calf was determined by actual sighting or obvious sign (i.e., tracks, beds, vocalizations). The frequency of walk- ins allowed for confident determination of presence or absence of unique events such as twinning. Parturition date was assigned by backdating the estimated age of a calf when first observed. Initial calf age (days) was based on coordination, mobility, wet or dry appearance, and presence of an umbilical cord (Larsen et al. 1989). Adults observed without a calf through 31 July were catego- rized as non-productive, producing 0 calves. We used generalized estimating equations (Harden and Hilbe 2003) to estimate popula- tion level birth rates (presence or absence of offspring) as a function of year and age, as both are known to influence per capita birth rates. Because teeth were not collected at capture, we assigned each adult cow to one of four age cat- egories: 2, ≥3, ≥4, and unknown. Individuals were classified initially as adult or calf (8–9 months old) at capture. All adults were of “unknown” age the first year but assumed to be ≥2 years old. In subsequent years, each adult advanced one year in age, and was classified as ≥3 or ≥ 4 depending on the year of study. Surviving female calves were assigned to their appropriate age class in successive years. Further, the age of 16 cows was assigned post-mortem after aging their central incisors. Analyses were conducted with R package, geepack (Højsgaard et al. 2006). Neonate Daily Survival to Independence Daily neonate survival was estimated from direct observations (2017–2019) during the walk-ins. A cow with calf was monitored 2–3 times weekly to document survival of the neonate(s). Loss of a neonate was confirmed from observation of a carcass or by its contin- ual absence in three subsequent walk-ins. In the latter case, the date of death was assigned as the mid-point between its last observation and date first missing (Musante et al. 2010, Jones et al. 2017). When mortality was sus- pected, we increased observations and field searches for calf sign (beds, tracks, scat). Deceased calves were assessed informally on site to possibly identify cause of death (e.g., predation, broken bones, environmental hazards, malnourishment). Daily survival rate of unmarked neo- nates was estimated using the Shaffer logis- tic exposure model, a generalized linear model that allows for varying visitation intervals (Shaffer 2004). Each period between walk-ins constituted a binomial trial where t = the number of days in the interval, and θ = st was the probability that the calf survives the interval, where s is the daily survival rate that depends on the value of explanatory variables such as individual or temporal covariates (Shaffer 2004). Exposure intervals were created for each neonate from the date of birth to date of death or 60 d (independence from mother), whichever came first. The interval length was computed as the number of days between walk-ins, with calf fate (survive or die) assigned each interval. The Shaffer logistic exposure model easily incorporates both interval-specific and individual covariates to assess their effects on daily survival. For each neonate, we measured interval-specific variables (averages) including calf age and local weather conditions from NOAA stations (precipitation, minimum, mean, and maxi- mum temperature; Table 1). We used the R package adehabitatLT (Calenge 2006) to compute the average distance moved by 59 ALCES VOL. 58, 2022 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. each cow in each interval. Individual neo- nate covariates included Julian birth date and age (d) (Table 1). We constructed 13 univariate and addi- tive models that combined temporal and individual covariates (informed by studies cited previously) to estimate daily survival probability (Table 1); models were ranked using AICc (Burnham and Anderson 2002). Assumptions described by Shaffer (2004) included: 1) neonates were accurately aged at first sighting, 2) fate was accurately deter- mined, 3) field observations did not influ- ence survival, and 4) neonates survived or failed independently of one another. We implemented the Shaffer logistic exposure model in R with the glm function that used the Shaffer logistic exposure link function provided by Bolker (2019). Per Capita Independence Rate We assessed the adult per capita indepen- dence rate (the probability an adult cow produces an independent calf) in two ways. First, for a subset of adult cows (n = 44), we collected urine and fecal samples from 20 January to April 7 (2017–2019) to measure fecal glucocorticoid metabolites (fGCM) concentrations and urea nitrogen:creatinine ratios (UN:C) as proxies for chronic stress and nutritional restriction, hypothesizing that neonate survival is related to the cow’s physiological condition. We used GPS loca- tions, VHF radio-telemetry, and tracking to collect random “fresh” samples of 5–7 fecal pellets and 10 mL of snow urine from each defecation or urination site located. Cow and calf samples were distinguished by the size of tracks, pellets, or back tracking for positive identification. Samples were col- lected with a rubber glove to minimize contamination. Samples were stored in a Whirl-Pak (Nasco, Madison, Wisconsin, USA) and frozen at −20°C. Fecal samples were shipped overnight for subsequent steroid and protein extraction; steroid metabolites were extracted by boiling approximately 0.2 g of dried material in 10 mL of 95% ethanol, after which the supernatant was dried and reconstituted in 1 mL of 95% methanol. Enzyme-linked immunosorbent assays were used to quan- tify fGCM concentrations (ELISA; Enzo Life Sciences ADI-900-071). We melted snow urine samples and subsampled 1mL of urine to measure urea nitrogen (UN) and creatinine (C) concentration (mg/dL) (Biovet, Inc., Barneveld, Wisconsin, USA). These data were expressed as a ratio (UN:C) and ratios were then loge-transformed to normalize variance (see Rosenblatt et al. 2021 for additional details). We used generalized estimating equa- tions (Harden and Hilbe 2003) to estimate per capita independence rate (at the popula- tion level) as a function of the annual median fGCM and UNC concentrations of the adult cow. Random effects were included because certain individuals were repre- sented multiple years. Second, we summa- rized the per capita independence rates per breeding cow as the total number of off- spring reared to day 60 out of total number of years observed. We included 2020 obser- vations in this analysis, although cows were observed only once (in August ~60–90 days post-partum) to determine presence of a calf. We related these rates to the habitat conditions in each cow’s average core home range during the autumnal questing period of winter ticks. To describe habitat in the core range, we used 4 recently developed Principal Component Analysis rasters that included 30 m2 land cover composition (deciduous forest, coniferous forest, mixed forest, and wetland from the National Land Cover Data, NLCD 2019), 30 m2 layers describing terrain characteristics (elevation and slope) (VCGI 2002), and 10 m2 lidar (light detection and ranging) variables 60 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. ALCES VOL. 58, 2022 (USGS 2016) that characterized forest age structure (Table 2; see Blouin et al. 2021a, 2021b). Each raster was rescaled to a 200 m2 resolution where pixel values provided the average of the underlying cells (Fig. 2). The home range cores were 50% fixed ker- nels (Worton 1989) using radio-collar loca- tions between 19 September and 15 November, the peak questing period of win- ter ticks in the region (Healy et al. 2018); analyses were performed in R package ade- habitatHR (Calenge 2006). Home ranges were overlaid on the 4 QPC (fall questing principal component) rasters, and average home range QPC scores were calculated by individual across years. We used binomial regression to analyze the successful rate to independence (fractions that represent the number of successes in n trials), where suc- cesses were the number of years in which an adult female successfully reared a calf to independence, and trials was the number of years the adult was alive in May to give birth. We used the glm() function in R to fit the model (successes, trials) ~ QPC1 + QPC2 + QPC3 + QPC4. RESULTS We captured and radio-marked 37 cows and 38 female calves (~8 months old); none were recaptured. Survivors remained in the study, increasing in age class through time. Tick counts ranged from 4–98 (median = 19.0) on cows and 0–84 on calves (median = 29.7). Mean body condition at capture was 2.83 and 2.57 (3 = normal) for cows and calves, respectively (see DeBow et al. 2021). Pregnancy Rate The average, 2-year rate (2017 and 2018) was 67% (24 of 36); 19 of 30 (63%) in 2017 and 5 of 6 (83%) in 2018. The intercept model predicted a similar pregnancy rate (0.67, 95% CI = 0.50 – 0.80, Table 3). Only lungworm level was negatively related to pregnancy rate (z = −2.31; Table 3). Table 2. A Principal Component Analysis of landcover and lidar variables that described moose habitat in northeastern Vermont, USA. Evergreen, deciduous, mixed, and wetland describe habitat composition as measured by the National Landcover Database (30 m2 pixel resolution; 2016). Canopy (vegetation > 6 m), Cover (vegetation between 3 – 6 m), Forage (vegetation < 3 m), and Shrub (vegetation between 0.2 and 2 m) describe habitat structure as measured through 10 m2 lidar imagery (USGS 2016) rescaled to 30 m2. The elevation raster from VCGI (Vermont Center for Geographic Information 2002) provided elevation in meters. The cumulative proportion indicates that the first four components explain 78% of the total variance of the data. Variable QPC1 QPC2 QPC3 QPC4 QPC5 QPC6 QPC7 QPC8 QPC9 Cumulative Proportion 0.33 0.55 0.67 0.78 0.87 0.94 0.99 1.00 1.00 Evergreen 0.15 0.42 0.54 0.19 0.53 0.03 0.11 0.44 0.00 Deciduous −0.35 −0.46 −0.03 0.27 −0.02 0.22 −0.45 0.59 0.00 Mixed 0.05 0.47 −0.57 −0.40 −0.07 0.21 0.02 0.49 0.02 Wetland 0.35 −0.07 0.32 0.08 −0.74 0.07 0.32 0.33 −0.01 Canopy −0.33 0.36 0.25 0.06 −0.21 0.73 −0.14 −0.32 0.00 Cover 0.25 0.42 −0.07 0.37 −0.25 −0.33 −0.64 −0.07 −0.17 Forage 0.53 −0.12 −0.14 0.17 0.13 0.30 −0.15 −0.10 0.72 Shrub 0.49 −0.22 −0.16 0.10 0.21 0.41 0.02 −0.07 −0.67 Elevation −0.21 0.14 −0.40 0.74 0.00 0.00 0.48 −0.01 0.02 61 ALCES VOL. 58, 2022 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. Per Capita Birth Rate The overall birth rate was 0.57 (63 of 110) ranging from 0.62 (26 of 42) in 2018 to 0.53 (21 of 40) in 2019. Yearling (age = 1–2) and adult (age ≥ 3) birth rates were 0.08 (1 of 13) and 0.64 (62 of 97), respectively. Twinning was documented once only (2017; rate = 0.01). The mean and median birth date were both 18 May, with 87% of births (55 of 63) within one week of this date. The annual mean birth date was 18, 17, and 21 May in 2017, 2018, and 2019, respectively. The earli- est and latest births were 07 May 2019 and 06 June 2019. The annual birth rate estimated with generalized estimating equations was simi- lar among years (2017 CI = 0.25 – 0.85; 2018 CI = 0.34 – 0.89; 2019 CI = 0.12 Fig. 2. Mapped distribution of four principal components describing habitat variables for moose in northeastern Vermont, USA. Each component is a combination of structural components (represented by lidar data; USGS 2016) and composition variables (represented by NLCD data; NLCD 2019), meters for elevation, and degrees for slope. See Table 2 for loadings. Cell size is 200 m2 and axes include UTM (Universal Transverse Mercator) Easting and Northing coordinates. 62 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. ALCES VOL. 58, 2022 – 0.67), but increased with age (Table 4). As with pregnancy rate, birth rate by yearlings was significantly lower (CI = 0 – 0.25) com- pared with older age groups (e.g., age > 4, CI = 0.43 – 0.9). The degree of uncertainty in classifying age groups combined with small sample size led to imprecise estimates of birth rates (Table 4); however, rates were <0.9 across years and ages. Neonate Daily Survival to Independence Neonate survival to 60 d post-parturition was 0.67 (42 of 63 calves) overall; annual survival was 0.63 (10 of 16), 0.69 (18 of 26), and 0.67 (14 of 21) in 2017, 2018, and 2019, respectively. Only one yearling produced a single calf during the study; that calf sur- vived to 60 d. The average date of death (n = 21) was 9.6 d post-partum (range = 1–57 days), Table 3. Logistic regression model selection results examining probability of pregnancy of adult female moose at capture (n = 36) in northeastern Vermont, USA (2017–2018) as a function of individual covariates. EPG = Total parasite eggs per gram of fecal matter using the modified McMaster flotation technique. Please see Table 1 for covariate terms. Model K LL AICc Delta AICc AICc Wt Parameter Estimate Std. Error z value Pr(>|z|) Lungworm 2 −17.24 38.84 0.00 0.96 Intercept 1.38 0.46 2.98 0.0048 Lungworm −1.72 0.75 −2.31 0.0280 Condition 2 −21.64 47.64 8.80 0.01 Intercept −2.93 2.35 −1.25 0.1834 Condition 1.29 0.83 1.55 0.1200 Intercept 1 −22.91 47.95 9.10 0.01 Intercept 0.69 0.35 1.96 0.0584 Year 2 −22.42 49.20 10.36 0.01 Year 2017 0.55 0.38 1.44 0.1409 Year 2018 1.06 1.16 0.92 0.2620 EPG 2 −22.79 49.94 11.10 0.00 Intercept 0.77 0.39 1.98 0.0560 EPG −0.01 0.03 −0.51 0.3509 Roundworm 2 −22.88 50.12 11.27 0.00 Intercept 0.73 0.37 1.94 0.0608 Nematodirus −0.01 0.03 −0.28 0.3836 Ticks 2 −22.91 50.18 11.34 0.00 Intercept 0.74 0.56 1.33 0.1651 Ticks 0.00 0.02 −0.11 0.3967 Table 4. Generalized estimating equation for birth rate of adult female moose in northeastern Vermont, USA (2017–2019). Coefficients are on the log-odds scale. LCI = lower 95% confidence interval; UCI = upper 95% confidence interval. Parameter Estimate Std Error Wald Pr(>|W|) Birth rate LCI Birth rate UCI Intercept (Age 3, 2017) 0.316 0.712 0.197 0.657 0.25 0.85 Year 2018 0.741 0.710 1.091 0.296 0.34 0.89 Year 2019 −0.620 0.688 0.814 0.367 0.12 0.67 Age > 4 0.960 0.640 2.246 0.134 0.43 0.90 Age 2 −3.266 1.105 8.731 0.003 0.00 0.25 Unknown −0.223 0.718 0.096 0.757 0.16 0.77 63 ALCES VOL. 58, 2022 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. with the majority (80%) occurring within 2 weeks; cause(s) of death was unknown. Relative to daily probability of calf sur- vival (Shaffer 2004), the “days since birth model” was the top-ranked carrying 55% of weight in the model set (Table 5); proba- bility of daily survival increased with calf age (Fig. 3). The second-best model included days since birth (+ effect) and Julian day of birth (- effect) and accounted for 30% of the weight in the model set; however, because confidence intervals for Julian day of birth crossed 0 and the model’s likelihood was negligibly better than the top model, it was dropped from consideration (Table 5). Neonate daily survival from the top model was estimated as 0.97 (0.94 to 0.98, 95% CI) on day zero (day of birth) and increased steadily over time. Daily survival was nearly 1.0 (0.997) by 30 d post-partum, and 0.999 at 60 d (Fig. 3). Cumulative survival from age 0 to 60 d was at 0.65 (95% CI = 0.44 to 0.79). We detected no signal related to year, daily movement, or temporal conditions. Per Capita Independence Rate The independence rate per breeding adult (success rate; calf survived to 60 d) was measured in 116 events with 43 cows. As expected, the yearly sample and rate (bino- mial trials) varied individually because annual captures of adult cows differed, cer- tain calves survived to reproductive age, and annual mortality of cows: 12 cows were observed 1 year, 8 in 2 years, 3 in 3 years, and 19 in all 4 years of the study. The indi- vidual success rate (binomial range = 0–1) ranged from 0 to 1, with just one female rearing a calf to independence all 4 years; only 7 cows had a success rate > 0.75. The rate increased with the median fGCM con- centration and decreased with the median UN:C ratio (2017: n = 3; 2018: n = 20; 2019: n = 21) (Fig. 4; Table 6). The success rate was positively related to questing home range as defined by QPC3 (Fig. 5, Table 6). Cows with core ranges characterized by large amounts of mature (canopy) evergreen forests and wetland hab- itats, and small amounts of mixed forests and elevation, had a higher average success rate than those with core ranges at high ele- vation with abundant mixed forest. None of the other PC variables (QPC1, QPC2, QPC4) were significant (alpha = 0.1; Table 6). The four uncorrelated components described 78% of the total variance of the habitat variables (Table 2). QPC1 was posi- tively related to younger (shrub/forage) wet- land habitat, and negatively related to the proportion of mature deciduous forest. QPC2 was positively related to mixed and ever- green forests that were more mature in struc- ture and negatively related to early succession (shrub/forage) deciduous forests. QPC3 was positively related to mature (canopy) ever- green forests and wetland habitats and nega- tively related to mixed forests and elevation. QPC4 was positively related to higher eleva- tion deciduous and evergreen forests of all age classes and negatively related to mixed forest composition (Table 2, Fig. 2). DISCUSSION Recent declines of moose populations in sev- eral regions along their southern range bound- ary in North America present unique management challenges in maintaining sus- tainable, viable populations (Timmermann and Rodgers 2017). These declines are com- pounded by a low-moderate fecundity and reflect the demographic vulnerability of peripheral populations to changing environ- mental conditions (Ruprecht et al. 2016). Not surprisingly, we found that birth rate and pro- ductivity increased with age, yet significant annual variation existed. The overall adult pregnancy rate was 0.67, a 48% reduction rel- ative to the average corpora lutea count of 1.3 64 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. ALCES VOL. 58, 2022 Table 5. AICc and parameter estimates of 13 univariate and multivariate logistic exposure models (Shaffer 2004) used to estimate moose calf survival to 60 days in northeastern Vermont, USA (2017–2019). Both univariate and additive models that combined temporal and individual covariates were considered. Coefficients are on the log-odds scale. UCI and LCI represent 95% upper and lower confidence intervals, respectively. Model LL AICc Delta_ AICc AICcWt Parameter Estimate Std. Error UCI LCI Days Since Birth −75.90 155.80 0.00 0.55 Intercept 3.37 0.36 4.08 2.66 Days Since Birth 0.08 0.02 0.12 0.04 Days Since Birth + Birth date −75.55 157.10 1.24 0.30 Intercept 8.58 5.76 19.87 −2.70 Birth date −0.04 0.04 0.04 −0.12 Days Since Birth 0.08 0.02 0.12 0.04 Year + Days Since Birth −75.50 159.00 3.21 0.11 Intercept 3.04 0.50 4.01 2.07 2018 0.43 0.56 1.53 −0.67 2019 0.51 0.61 1.71 −0.69 Days Since Birth 0.08 0.02 0.12 0.04 Day Of Year −78.50 161.00 5.17 0.04 Intercept −5.23 2.59 −0.15 −10.31 Julian 0.06 0.02 0.10 0.03 Max Temp −82.90 169.80 14.03 0.00 Intercept 0.87 1.08 2.98 −1.24 Max Temp 0.20 0.05 0.30 0.10 Min Temp −85.35 174.70 18.84 0.00 Intercept 3.70 0.50 4.68 2.73 Min Temp 0.18 0.07 0.31 0.06 Birthday + Temp + Precipitation −83.65 175.30 19.48 0.00 Intercept 13.26 5.75 24.53 2.00 Precipitation −0.07 0.07 0.06 −0.20 Min Temp 0.23 0.07 0.37 0.08 Birth Julian −0.07 0.04 0.01 −0.15 Min Temp + Precipitation −84.85 175.70 19.92 0.00 Intercept 3.77 0.50 4.75 2.79 Precipitation −0.07 0.07 0.06 −0.21 Min Temp 0.21 0.07 0.34 0.07 Intercept −89.60 181.20 25.42 0.00 Intercept 5.18 0.24 5.64 4.71 Birthdate −89.00 182.00 26.17 0.00 Intercept 11.91 5.77 23.21 0.60 Birth Julian −0.05 0.04 0.03 −0.13 Distance traveled −89.30 182.60 26.80 0.00 Intercept 4.97 0.35 5.65 4.29 Distance traveled 0.00 0.00 0.00 0.00 Precipitation −89.60 183.20 27.42 0.00 Intercept 5.15 0.34 5.82 4.48 Precipitation 0.01 0.08 0.16 −0.14 Year −89.10 184.20 28.41 0.00 Intercept (2017) 4.80 0.41 5.60 3.99 2018 0.53 0.56 1.63 −0.57 2019 0.52 0.61 1.72 −0.67 65 ALCES VOL. 58, 2022 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. in >3 year-old cows in 2003 when the popula- tion was >2 × larger and less impacted by winter ticks (VFWD 2009). The negative relationship between lungworm infestation and pregnancy may signify an additive and compounding effect to winter tick parasitism that reduces fitness and ovulation rate (Ballweber 2021). Likewise, the low birth rate of adult cows (0.57) and the single case of twinning suggests that nutritional condi- tion (fitness) of cows was possibly compro- mised early in the reproductive cycle. Calf age was the strongest predictor of daily survival, similar to observations in New Hampshire (Jones et al. 2017) and Minnesota (Severud et al. 2019), and as in white-tailed deer (Gaillard et al. 2000) and red deer (Cervus elaphus) (Smith and Anderson 1996). However, the overall survival rate to 60 d was ~0.67, measurably lower than elsewhere in North America (Boer 1992, Schwartz 1992c, VFWD 2009, Boertje et al. 2019) and expected at the study area latitude (Ruprecht et al. 2016). It was less than in New Hampshire (0.77) and Maine (0.94) in related studies conducted with alike methods (Jones et al. 2017), as well as in New Hampshire (0.71) in 2002 to 2005 (Musante et al. 2010). Although documenting births and calf survival can be challenging and uncertain at times, as in ear- lier research (Jones et al. 2017), we used highly trained field staff to maintain continu- ous and intensive monitoring each summer (~1700 observations) to ensure continuity with regional datasets. The low birth, calf survival, and indepen- dence rates we found suggest a declining population occupying sub-optimal habitat with limited resources (Franzmann and Schwartz 1985, Boer 1992, Malmsten et al. 2014, Blouin et al. 2021b). Moose in adjacent northern New Hampshire and central Maine have similar characteristics, yet regional hab- itat there and in our study area is considered near optimal for moose as commercial timber harvests provide constant availability of regenerating forest (Dunfey-Ball 2017, Jones et al. 2017, 2019, Blouin et al. 2021b). Fig. 3. Daily survival estimates of neonate moose in northeastern Vermont, USA based on days since birth as estimated from the Shaffer logistic exposure survival model (2004). Shaded areas indicate the 95% confidence intervals. 66 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. ALCES VOL. 58, 2022 Fig. 4. Relationships between binomial independence probability (the proportion of successful rearing of offspring to day 60 out of total years observed per female) and median fecal glucocorticoid metabolites (fGCM) concentrations, which is a proxy for stress, (left) and median UN:C ratios (right). Analyses were based on 25 individual adult female moose in northeastern Vermont, USA (2017–2019). Table 6. Generalized estimating equations relating the binomial probability of successfully rearing moose offspring to day 60 (independence rate) in northeastern Vermont, USA (2017–2020) as a function of fall questing habitat (Questing HR) and average physiological metrics from fecal and urine samples collected from January to April. 1fGCM = concentration of fecal glucocorticoid metabolites. 2UNC = urea nitrogen:creatinine ratio. Analysis Parameter Estimate Std. Error z Pr(>|z|) Questing HR Intercept −0.135 0.247 −0.550 0.584 QPC1 0.031 0.199 0.150 0.878 QPC2 −0.190 0.315 −0.600 0.546 QPC3 0.633 0.380 1.660 0.096 QPC4 0.669 0.502 1.330 0.183 Physiology Intercept −4.437 1.936 5.250 0.022 fGCM1 4.343 1.427 9.270 0.002 Median UNC2 −0.875 0.369 5.620 0.018 67 ALCES VOL. 58, 2022 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. Although habitat quality and forage resources influence reproductive performance, our study area consisted of ~35% optimal moose habitat based on availability of regenerating forest, including adequate winter range (Dunfey-Ball 2017). A more recent analysis of habitat using fine-scale (10 m2) lidar derived variables documented that 34% of the study area consisted of vegetation <3 m tall (Blouin et al. 2021b), providing an opti- mal source of moose forage and important component of a reproductive cow’s home range. Although we did not evaluate the nutritional landscape in terms of forage quality, moose forage on a wide variety of plants, with forage quantity (consumption) most important to nutritional condition (Shipley 2010). The regional frequency of winter tick epizootics – 5 in 6 years (2014 to 2019; Jones et al. 2019, Powers 2019, DeBow et al. 2021) – is historically unique (Samuel 2004) Fig. 5. Binomial independence probability (the proportion of successful rearing of offspring to day 60 out of total years observed per adult female moose) in relation to their average home range habitat condition during the winter tick fall questing period. The binomial success rates in adding a calf to independence was positively related to questing home range as defined by QPC3, which was positively influenced by mature (canopy) evergreen forests and wetland habitats and negatively influenced by mixed forests and elevation. Analysis was based on 25 individual adult female moose in northeastern Vermont, USA (2017–2019). 68 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. ALCES VOL. 58, 2022 and arguably, strongly influenced dynamics of the study population. The bioenergetic impacts associated with this sustained level of parasitism directly affects ovulation rate, gestation, birth rate, lactation, maturation, and recruitment in moose (Musante et al. 2007, Jones et al. 2017, 2019, Pekins 2020). Although we did not collect specific health-related metrics post-capture, the aver- age condition score of cows and calves at capture was below “normal,” indirect evi- dence that moose were compromised by winter tick parasitism in good habitat. Although we could not definitively parse out the relative influences of predation, habitat condition, or parasitism, the latter two fac- tors are presumably most important because predation is limited to nominal loss of neo- nates to black bears (Jones et al. 2017). Further, regional research indicates that climate change and macro- and microhabitat use influence winter tick abundance and infestation of moose (Dunfey-Ball 2017, Jones et al. 2019, Pekins 2020, Powers and Pekins 2020, Blouin et al. 2021a, 2021b, Rosenblatt et al. 2021). Blouin et al. (2021a) found a relation- ship between habitat use by maternal cows during the questing period of winter ticks (autumn) and winter survival of its calf. Cows with surviving calves used areas with greater proportions of young (shrub/forage) deciduous forests, and with higher propor- tions of mature (canopy) evergreen or mixed forests and wetlands at lower elevations (Blouin et al. 2021a, +QPC3 effect). Likewise, we found that these same habitat features were related to the independence rate (+QPC3; Tables 2 and 6). Healy et al. (2018) found overlapping habitat use by cows in autumn and spring, and more specif- ically, Blouin et al. (2021a) found a relation- ship between that specific overlap and cows that lost their calf. Both studies suggest a self-sustaining cycle of winter tick parasitism associated with regenerating for- est, specifically, optimal foraging habitat that harbors highest abundance of winter ticks (Powers and Pekins 2020) due to seasonal foraging preferences of moose. Further, signals in fecal stress metabo- lite concentrations and UN:C ratios reflected the low independence rate presumably related to declining nutritional condition from winter tick infestation (Pekins 2020). Specifically, UN:C ratios in calves (Ellingwood et al. 2019, Rosenblatt et al. 2021) are posi- tively related to the adult engorgement period of winter ticks that coincides with highest calf mortality. However, these met- rics can be influenced by multiple variables including weather, habitat, and nutritional condition that often change through time. We found a negative relationship between UN:C ratios and overall probability of recruiting calves to independence, but sur- prisingly, a positive relationship with fGCM. Elevated fGCM concentrations in cows that successfully recruit a calf may indicate maternal programming to better cope with adverse conditions (Sheriff et al. 2017), or may simply reflect the influence of preg- nancy on fGCM concentrations (Dantzer et al. 2014). The negative correlation between UN:C ratios and recruitment aligns with our understanding of the physiological toll of winter ticks on calves and cows during the final trimester of pregnancy (Keech et al. 2000, Pekins 2020). A plausible hypothesis explaining low birth and calf survival rates of moose is the nutritional demand of increased and compounding parasitism from winter ticks (Jones et al. 2019, Pekins 2020). Further, adult reproduction is likely impacted through the nutritional costs of repeated winter tick parasitism (Samuel 2004, Pekins 2020), reducing a cow’s ability to recuperate with each event. To compensate, cows may give birth to 69 ALCES VOL. 58, 2022 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. compromised calves with low survival and predetermined mortality (Schwartz 1992a, 1992c) or fail to reproduce in suc- cessive years (Jones et al. 2019); either scenario aids subsequent recruitment through increased compensatory growth and fitness of that individual cow (Pekins 2020). Further monitoring of these tem- poral effects of winter tick parasitism may provide comprehensive understand- ing of the complex relationships among parasitism, reproduction, and population dynamics in moose. MANAGEMENT IMPLICATIONS Given that our study was conducted during a period of sustained winter tick epizootics, the low birth and calf survival rates were most likely a direct result of high and inces- sant parasitism. Winter tick abundance on the landscape would need to decline to coun- teract this effect by disrupting the host-para- site relationship. Moose (host) population reduction would theoretically decrease the distribution, abundance, and density of win- ter ticks on the landscape, effectively reduc- ing infestation of individual moose and increasing productivity. Although moose harvests in Vermont were purposefully reduced and terminated temporarily, harvest will restart in 2022 as a possible proactive management strategy to address winter tick parasitism (Ellingwood et al. 2020). Forest management that provides quality browse and optimal cover distributed widely across the landscape may improve condition of cows, increase recruitment, and minimize local abundance of winter ticks that help per- petuate epizootics. Because much of the regional moose population is contiguous, sharing similar habitat, resources, and con- straints, coordinated management strategies may best address the impact of winter tick parasitism. ACKNOWLEDGMENTS We thank W. Cottrell for veterinarian consul- tation which was funded by the Vermont Fish and Wildlife Department. We are also grate- ful for the support of the U.S. Fish and Wildlife Service National Wildlife Refuge System – Nulhegan Division, and land access from Weyerhaeuser Company, LandVest Inc., and many private landowners. Support from P. J. Pekins and L. E. Kantar was instru- mental in the design and follow through of this study as well as the hard work of a dedi- cated field staff including T. R. Smith, C. Lampart, L. A. Rossier, and D. P. Hotchkiss. At the time of publication, data were not pub- licly available from the Vermont Fish and Wildlife Department. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government. The Vermont Cooperative Fish and Wildlife Research Unit is jointly supported by the U.S. Geological Survey, University of Vermont, Vermont Fish and Wildlife Department, and Wildlife Management Institute. Funding was pro- vided by the Vermont Fish and Wildlife Department in cooperation with the U.S. Fish and Wildlife Service Division of Wildlife and Sportfish Restoration – Wildlife Restoration Program, the U.S. Geological Survey (Grant/ Cooperative Agreement No. G19AC00241), Safari Club International Foundation, and Vermont Electric Company. REFERENCES Adams, K. P., and P. J. Pekins. 1995. Growth patterns of New England moose: year- lings as indicators of population status. Alces 31: 53–59. BallweBer, L. R. 2021. Lungworm Infection in Animals. Merck Veterinary Manual. Merck & Co., Inc., Whitehouse Station, New Jersey, USA. 70 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. ALCES VOL. 58, 2022 Bergeron, D. H., and P. J. Pekins. 2014. Evaluating the usefulness of three indices for assessing winter tick abundance in northern New Hampshire. Alces 50: 1–15. _____, _____, and K. rines. 2013. Temporal assessment of physical characteristics and reproductive status of moose in new hampshire. Alces 49: 39–48. Blouin, J., J. deBow, E. rosenBlatt, J. Hines, C. alexander, K. gieder, N. Fortin, J. murdocH, and T. donovan. 2021a. Moose habitat selection and fit- ness consequences during two critical winter tick life stages in Vermont, United States. Frontiers in Ecology and Evolution 9: 642276. doi: 10.3389/ fevo.2021.642276 _____, _____, _____, C. alexander, K. gieder, N. Fortin, J. murdocH, and T. donovan. 2021b. Modeling moose habi- tat use by age, sex, and season in Vermont, USA using high-resolution lidar and national land cover data. Alces 57: 71–98. Boer, A. H. 1992. Fecundity of North American moose (Alces alces): a review. Alces Supplement 1: 1–10. Boertje, R. D., G. G. Frye, and D. D. young jr. 2019. Lifetime, known-age moose reproduction in a nutritionally stressed population. Journal of Wildlife Management 83: 610–626. doi: 10.1002/ jwmg.21613 Bolker, B. 2019. Logistic Regression, Accounting For Differences in Exposure. RStudio, RPubs. (accessed August 2019). BurnHam, K., and D. R. anderson. 2002. Model Selection and Multimodel Inference: A Practical Information- Theoretic Approach. Second Edition. Springer, New York, New York, USA. calenge, C. 2006. The package “adehabitat” for the R software: a tool for the analysis of space and habitat use by animals. Ecological Modelling 197: 516–519. doi: 10.1016/j.ecolmodel.2006.03.017 dantzer, B., Q. E. FletcHer, R. Boonstra, and M. J. sHeriFF. 2014. Measures of physiological stress: a transparent or opaque window into the status, manage- ment and conservation of species? Conservation Physiology 2: cou023. doi: 10.1093/conphys/cou023 deBow, J., J. Blouin, E. rosenBlatt, C. alexander, K. gieder, W. cottrell, J. murdocH, and T. donovan. 2021. Effects of winter ticks and internal parasites on moose survival in Vermont, USA. Journal of Wildlife Management 85: 1423–1439. doi: 10.1002/jwmg.22101 degraaF, R. M., and M. yamasaki. 2001. New England Wildlife. Pages 358–359 in New England Wildlife: Habitat, Natural History, and Distribution. University Press of New England, Lebanon, New Hampshire, USA. delgiudice, G. D. 1995. Assessing winter nutritional restriction of northern deer with urine in snow: considerations, potential, and limitations. Wildlife Society Bulletin 23: 687–693. dunFey-Ball, K. R. 2017. Moose density, habitat, and winter tick epizootics in a changing climate. M.S. Thesis. University of New Hampshire, Durham, New Hampshire, USA. eBerHardt, L. L. 2002. A paradigm for pop- ulation analysis of long-lived verte- brates. Ecology 83: 2841. doi: 10.1890/0012-9658 (2002)083[2841: APFPAO]2.0.CO;2 ellingwood, D., P. J. Pekins, and H. jones. 2019. Using snow urine samples to assess the impact of winter ticks on moose calf condition and survival. Alces 55: 13–21. _____, _____, _____, and A. R. musante. 2020. Evaluating moose Alces alces population response to infestation level of winter ticks Dermacentor albipictus. Wildlife Biology 2020: 1–7. doi: 10.2981/wlb.00619 evans, M. R., M. L. roBerts, K. L. BucHanan, and A. R. goldsmitH. 2006. https://rpubs.com/bbolker/logregexp https://rpubs.com/bbolker/logregexp 71 ALCES VOL. 58, 2022 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. Heritability of corticosterone response and changes in life history traits during selection in the zebra finch. Journal of Evolutionary Biology 19: 343–352. doi: 10.1111/j.1420-9101. 2005.01034.x Franzmann, A. W., and C. C. scHwartz. 1985. Moose twinning rates: a possible popula- tion condition assessment. Journal of Wildlife Management 49: 394–396. doi: 10.2307/3801540 _____, and _____. 2007. Ecology and Management of the North American Moose. University Press of Colorado, Boulder, Colorado, USA. gaillard, J. M., M. Festa-BiancHet, N. G. yoccoz, A. loison, and C. Toїgo. 2000. Temporal variation in fitness compo- nents and population dynamics of large herbivores. Annual Review of Ecologic Systematics 31: 367–393. Harden, J., and J. HilBe. 2003. Generalized Estimating Equations. Chapman and Hall/CRC, London, United Kingdom. Healy, C., P. J. Pekins, L. kantar, R. G. congalton, and S. atallaH. 2018. Selective habitat use by moose during critical periods in the winter tick life cycle. Alces 54: 85–100. Højsgaard, S., U. HalekoH, and J. yan. 2006. The R Package geepack for Generalized Estimating Equations. Journal of Statistical Software 15: 1–11. doi: 10.18637/jss.v015.i02 Hundertmark, K. J., and C. C. scHwartz. 1998. Predicting body mass of Alaskan moose (Alces alces gigas) using body measurements and condition assess- ment. Alces 34: 83–89. jones, H., P. J. Pekins, L. kantar, I. sidor, D. ellingwood, A. licHtenwalner, and M. o’neal. 2019. Mortality assessment of moose calves during successive years of winter tick epizootics in New Hampshire and Maine (USA). Canadian Journal of Zoology 97: 22–30. doi: 10.1139/cjz-2018-0140 _____, _____, _____, m. o’neil, and d. ellingwood. 2017. Fecundity and summer calf survival of moose during 3 successive years of winter tick epizoot- ics. Alces 53: 85–98. jensen, W. F., J. R. smitH, M. carstensen, C. E. Penner, B. M. Hosek, and J. J. maskey, jr. 2018. Expanding GIS analyses to monitor and assess North American moose distribution and den- sity. Alces 54: 45–54. keecH, M. A., R. T. Bowyer, M. V. H. jay, R. D. Boertje, B. W. dale, and T. R. stePHenson. 2000. Life-history conse- quences of maternal condition in Alaskan moose. Journal of Wildlife Management 64: 450–462. doi: 10.2307/3803243 larsen, D. G., D. A. gautHier, and R. L. markel. 1989. Causes and rate of moose mortality in the Southwest Yukon. Journal of Wildlife Management 53: 548–557. doi: 10.2307/3809175 malmsten, J., L. söderquist, C.-G. tHulin, D. gavier widén, L. yon, M. R. HutcHings, and A.-M. dalin. 2014. Reproductive characteristics in female Swedish moose, with emphasis on puberty, timing of oestrus, and mating. Acta Veterinaria Scandinavica 56: 23–33. doi: 10.1186/1751-0147-56-23 mccann, N. P., R. A. moen, and T. R. Harris. 2013. Warm-season heat stress in moose (Alces alces). Canadian Journal of Zoology 91: 893–989. doi: 10.1139/ cjz-2013-0175 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 cli- mate influences on a declining moose population. Wildlife Monographs 166: 1–30. doi: 10.2193/0084-0173 (2006)166 [1:PNDACI]2.0.CO;2 musante, A. R., P. J. Pekins, and D. L. scarPitti. 2007. Metabolic impacts of winter tick infestations on calf moose. Alces 43: 101–110. doi: 10.2981/09-014 _____, _____, and _____. 2010. Characteristics and dynamics of a 72 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. ALCES VOL. 58, 2022 regional moose population in the north- eastern United States. Wildlife Biology 16: 185–204. national climate data center (NCDC). 2019. Climate Data Online: Dataset Directory. (accessed August 2019). national land cover dataBase (NLCD). 2019. National Land Cover Database (NLCD) 2016 Products: U.S. Geological Survey Data Release. doi: 10.5066/ P96HHBIE Parker, K. L. 2003. Advances in the nutri- tional ecology of cervids at different scales. Ecoscience 10: 395–411. doi: 10.1080/11956860.2003.11682788 Pekins, P. J. 2020. Metabolic and population effects of winter tick infestations on moose: unique evolutionary circum- stances? Frontiers in Ecology and Evolution 8: 176. doi: 10.3389/ fevo.2020.00176 Powers, B. 2019. Assesing the relationships of winter ticks, weather, and a declining moose population in northern New Hampshire. M. S. Thesis. University of New Hampshire, Durham, New Hampshire, USA. _____, and P. J. Pekins. 2020. Abundance of winter ticks (Dermacentor albipictus) in two regenerating forest habitats in New Hampshire, USA. Alces 56: 1–13. r core team. 2017. R: A Language and Environment for Statistical Computing. R Foundation for Statisitcal Computing, R Core Team, Vienna, Austria. rosenBlatt, E., J. deBow, J. Blouin, T. donovan, J. murdocH, S. creel, W. rogers, K. gieder, N. Fortin, and C. alexander. 2021. Juvenile moose stress and nutrition dynamics related to winter ticks, landscape characteristics, cli- mate-mediated factors and survival. Conservation Physiology 9(1): coab048. doi: 10.1093/conphys/coab048 ruPrecHt, J. S., K. R. Hersey, K. HaFen, K. L. monteitH, N. J. decesare, M. J. kauFFman, and D. R. macnulty. 2016. Reproduction in moose at their southern range limit. Journal of Mammalogy 97: 1355–1365. doi: 10.1093/jmammal/ gyw099 saetHer, B.-E., and H. Haagenrud. 1983. Life history of the moose: fecundity rates in relation to age and carcass weight. Journal of Mammalogy 64: 226–232. doi: 10.2307/1380552 _____, and M. Heim. 1993. Ecological cor- relates of individual variation in age at independence in female moose: the effects of environmental variability. Journal of Animal Ecology 62: 482–489. doi: 10.2307/5197 samuel, B. 2004. White as a Ghost: Winter Ticks & Moose. Federation of Alberta Naturalists, Manitoba, Canada. sand, H. 1996. Life history patterns in female moose: the relationship between age, body size, fecundity and environ- mental conditions. Oecologia 106: 212–220. doi: 10.1007/BF00328601 scHwartz, C. C. 1992a. Physiological and nutri- tional adaptations of moose to northern environments. Alces (Suppl.) 1: 139–155. _____. 1992b. Techniques of moose hus- bandry. Alces (Suppl.) 1: 177–192. _____. 1992c. Reproductive biology of North American moose. Alces 28: 165–173. _____. 2007. Reproduction, natality 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. severud, W. J., T. R. oBermoller, G. D. delgiudice, and J. R. FieBerg. 2019. Survival and cause-specific mortality of moose calves in northeastern Minnesota. Journal of Wildlife Management 83: 1131–1142. doi: 10.1002/jwmg.21672 sHaFFer, T. L. 2004. A unified approach to analyzing nest success. The Auk 121: 526–540. doi: 10.1642/0004-8038(2004) 121[0526:AUATAN]2.0.CO;2 https://www.ncdc.noaa.gov/cdo-web/ https://www.ncdc.noaa.gov/cdo-web/ 73 ALCES VOL. 58, 2022 MOOSE REPRODUCTION IN VERMONT, USA – DEBOW ET AL. sHaPiro, L. S. 2010. Pathology and Parasitology for Veterinary Technicians. Thomson Delmar Learning, Clifton Park, New York, USA. sHeriFF, M. J., A. Bell, R, Boonstra, B. dantzer, S. G. lavergne, K. E. mcgHee, K. J. macleod, L. winandy, C. zimmer, and O. P. love. 2017. Integrating eco- logical and evolutionary context in the study of maternal stress. Integrative and Comparative Biology 57: 437–449. doi: 10.1093/icb/icx105 sHiPley, L. 2010. Fifty years of food and for- aging in moose: lessons in ecology from a model herbivore. Alces 46: 1–13. sikes, R. S., J. A. Bryan II, B. J. danielson, J. eggleston, M. R. gannon, W. L. gannon, D. W. Hale, B. R. jesmer, D. K. odell, L. E. olson, R. D. stevens, T. A. tHomPson, R. M. timm, S. A. trewHitt, and J. R. willougHBy. 2016. 2016 guide- lines of the American Society of Mammalogists for the use of wild mamals in research and education. Journal of Mammalogy 97: 663–688. doi: 10.1093/ jmammal/gyw078 smitH, B. L., and S. H. anderson. 1996. Patterns of neonatal mortality of elk in northwest Wyoming. Canadian Journal of Zoology 74: 1229–1237. doi: 10.1139/ z96-136 testa, J. W., and G. P. adams. 1998. Body condition and adjustments to reproduc- tive effort in female moose. Journal of Mammalogy 79: 1345–1354. doi: 10.2307/1383026 timmermann, H. R., and A. R. rodgers. 2005. Moose: competing and comple- mentary values. Alces 41: 85–120. _____, and _____. 2017. The status and management of moose in North America – circa 2015. Alces 53: 1–22. USGS. 2016. USGS Lidar Point Cloud VT Connecticut-River 2016 N5495E569 LAS 2018: U.S. Geological Survey. (accessed August 2019). Vermont Center for Geographic Information (VCGI). 2002. Online elevation data. (accessed August 2019). Vermont Fish and Wildlife Department (VFWD). 2009. Big Game Management Plan 2010–2020. Vermont Fish and Wildlife Department, Montpelier, Vermont, USA. _____. 2018. 2018 Vermont wildlife harvest report – Moose. Vermont Fish and Wildlife Department, Montpelier, Vermont, USA. _____. 2020. 2020 Antlerless Harvest and Youth Season Recommendation. Vermont Fish and Wildlife Department, Montpelier, Vermont, USA. wattles, D. W., and S. desteFano. 2011. Status and management of moose in the northeast- ern United States. Alces 47: 53–68. wingField, J. C., D. L. maney, C. W. Breuner, J. D. jacoBs, S. lynn, M. ramenoFsky, and R. D. ricHardson. 1998. Ecological bases of hormone-be- havior interactions: the emergency life history stage. American Zoologist 38: 191–206. doi: 10.1093/icb/38.1.191 worton, B. J. 1989. Kernel methods for esti- mating the utilization distribution in home- range studies. Ecology 70: 164–168. doi: 10.2307/1938423