Conservation of migratory species at risk: Environmental conditions experienced from pre-gestation to parturition affect fall recruitment in migratory caribou

Conservation of migratory species at risk: Environmental conditions experienced from pre-gestation to parturition affect fall recruitment in migratory caribou | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 11 November 2025 V1 Latest version Share on Conservation of migratory species at risk: Environmental conditions experienced from pre-gestation to parturition affect fall recruitment in migratory caribou Authors : Roxanne Turgeon 0000-0001-5087-9373 [email protected] , Alexandre Carbonneau 0009-0002-4857-847X , Fanny Lescouzeres 0009-0007-2486-5518 , Barbara Vuillaume , Sandra Hamel 0000-0003-1126-8814 , and Joëlle Taillon Authors Info & Affiliations https://doi.org/10.22541/au.176286075.59304212/v1 608 views 129 downloads Contents Abstract Introduction Materials and methods Results Discussion Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Annual variations in environmental conditions can strongly affect vital rates such as survival and recruitment. These effects are likely to be exacerbated in highly seasonal environments and in species facing substantial energetic needs during specific seasons. For instance, pregnant migratory females must balance energy expenditure between long-distance travel and gestation. Little is known, however, on the environmental factors influencing recruitment in species exhibiting long-distance migrations. We aimed to fill this gap by evaluating the effect of environmental conditions from pre-gestation to weaning on fall recruitment of two migratory caribou (Rangifer tarandus) herds followed for over 35 years in northeastern Canada. Fall recruitment decreased with high precipitation experienced by adult females around estrus in previous fall (12% [95% confidence interval: 1-22]) and during gestation in winter (13% [3-21]), as well as with warm temperature (13% [3-23]) during gestation in winter and at calving (12% [2-21]). These environmental conditions experienced by both populations were positively correlated, suggesting these parameters could increase overall recruitment synchrony between nearby populations. We found no support for an effect of conditions experienced after birth, demonstrating that environmental conditions encountered from conception to calving are the strongest determinants of recruitment in migratory caribou. Our results highlight the importance of female condition throughout gestation with maternal allocation in fetal growth and newborn calf playing a major role in recruitment. Synchrony observed in recruitment between herds also underlines the importance of considering multiple populations facing similar environmental conditions in conservation strategies as they may exhibit simultaneous fluctuations. Introduction Wildlife conservation efforts rely on understanding extrinsic and intrinsic factors influencing demographic rates and how these rates contribute to fluctuations in population size. The relative influence of demographic rates on population dynamics has been evaluated in various species (Crouse et al., 1987; Dobson and Oli, 2001; Nater et al., 2023). In long-lived species, adult survival is the main driver of individual’s fitness (Clutton-Brock, 1988; Newton, 1989) and is thus canalized against temporal fluctuations (Gaillard and Yoccoz, 2003). Therefore, adult survival typically shows high elasticity with great potential influence on population growth rate (Gaillard et al., 1998; Gaillard et al., 2000; Owen-Smith and Mason, 2005), where a slight decrease in adult survival can result in a drastic population decline (Arthur et al., 2003; Coulson et al., 2005). Still, adult survival in long-lived species is usually high and stable, whereas reproductive success and juvenile survival show wide annual variability, and hence can play a key role in population dynamics (Eberhardt, 2002; Gaillard et al., 2000). In long-lived mammals, recruitment is usually more variable than adult survival, and annual variation in recruitment can have major demographic impacts (Coulson et al., 2005; Gaillard et al., 1998). Recruitment is a commonly used indicator of population productivity integrating numerous reproductive demographic parameters: the probability of gestation, the probability of parturition, neonatal survival, and juvenile’s seasonal survival. Thus, several factors may influence recruitment in large ungulates including population density (Clutton-Brock and Coulson, 2002; Festa-Bianchet and Jorgenson, 1998; Sand et al., 1996), maternal traits and previous reproductive status (Adams and Dale, 1998; Hamel et al., 2010; Hamel et al., 2009a; Swartout et al., 2023), predation (Bergerud and Page, 1987; Bonenfant et al., 2009; Festa-Bianchet et al., 1994), parasitism (Hughes et al., 2009), disease (Cassirer et al., 2013), and environmental conditions (e.g. weather and resource availability; Gese et al., 2023; Johnson et al., 2013). These factors can thus indirectly impact population growth, as shown in British Columbia where a reduction in wolf predation increased recruitment and population growth across several species (Bergerud and Elliott, 1998). Environmental conditions experienced by adult females before estrus and during gestation can affect recruitment through fecundity. Adult female fecundity typically relies on the onset and timing of estrus and on fetal survival from conception to calving. In large herbivores, summer environmental conditions, such as precipitation and temperature (Johnson et al., 2013; Proffitt et al., 2014), can affect female body condition just before reproduction and thus the probability of gestation (Cameron et al., 1993; Holmes et al., 2021; Tveraa et al., 2013). During gestation, energy required for locomotion and resource availability can modulate maternal reserves, with a decrease in body condition resulting in lower energy and nutrient allocation to fetal growth, which reduces the probability of parturition or neonatal survival (Bishop et al., 2009). In high latitudes, ungulates gestation occurs in harsh environments, characterized by deep, dense snow (Fancy and White, 1987; Holmes et al., 2021) and meteorological extremes (e.g. cold temperatures; Fancy and White, 1987; Forchhammer et al., 2001; Holmes et al., 2021; Johnson et al., 2013; Proffitt et al., 2014), leading to substantial compromises between reproduction, growth and survival. Thus, environmental conditions reducing female body condition before estrus and during gestation can lead to lower recruitment. Furthermore, environmental conditions encountered at birth and during the rearing period affect maternal condition, which can influence offspring survival and hence recruitment (Guinness et al., 1978; Holmes et al., 2021; Mahoney et al., 2015). Offspring survival increases with greater birth mass and pre-weaning growth, which depends on maternal allocation and cares (Barboza and Parker, 2008; Théoret-Gosselin et al., 2015) that hinge upon maternal condition (Lamb et al., 2023; Taillon et al., 2012). During the first weeks after birth, environmental conditions directly affect offspring survival through variations in thermoregulation costs (Dion et al., 2020; Vuillaume et al., 2023). In addition, because energy intake and growth rely entirely on maternal milk during that period (Parker et al., 2009), forage availability will indirectly influence offspring survival through its effect on maternal condition and milk quality (Barboza and Parker, 2008; Michel et al., 2018; Pettorelli et al., 2007; Post and Forchhammer, 2008). Because forage quality and availability affect female lactation (Renaud et al., 2020; Scornavacca et al., 2016), synchrony between parturition and the period of high primary productivity is also crucial to increase offspring nutrition and survival (Laforge et al., 2021; Stoner et al., 2016). Migratory species face substantial energetic needs required for seasonal migration (Milner-Gulland et al., 2011). This is an additional challenge for migratory females with limited food resource availability or quality, who must balance energy expenditures between the requirements of long-distance travel and the gestation cycle (Laforge et al., 2021; Maresh et al., 2015; Russell et al., 2024). Severe environmental conditions can exacerbate costs of migration, leading to reduced adult body condition (Fancy and White, 1987; Robson and Barriocanal, 2008). In migratory ungulates, seasonal migration occurs during the rut, with females in poor body condition skipping or delaying estrus, which may result in lower probability of conception (Cameron et al., 1993; Parker et al., 2009). Seasonal migration may also occur during the last trimester of gestation, with poor maternal condition impeding fetal growth and resulting in higher neonate mortality at birth (Bishop et al., 2009; Guinness et al., 1978; Mahoney et al., 2015). Although previous studies assessed the impact of environmental conditions on recruitment in various ungulate species (Gese et al., 2023; Hegel et al., 2010; Starns et al., 2014), little is known on the environmental factors influencing recruitment in ungulates exhibiting long-distance migrations. Moreover, large-scale environmental conditions are likely to simultaneously affect multiple populations, which can lead to synchrony in vital rates of spatially close but distinct populations, such as synchrony in recruitment, and result in synchronous population dynamics (Liebhold et al., 2004). This synchrony, named the Moran effect, was documented in multiple species (Sæther et al., 2007; Tedesco et al., 2004), including ungulates (Aanes et al., 2003; Grøtan et al., 2005; Mallory et al., 2018). Population dynamics’ synchrony can affect metapopulation persistence because populations facing similar drivers may present greater risk of global extinction (Hudson and Cattadori, 1999; Kahilainen et al., 2018; Liebhold et al., 2004). In the current context of climate change, assessing synchrony among populations is crucial because it can have implications for the management and conservation of species. We took advantage of long-term monitoring datasets from two migratory caribou ( Rangifer tarandus ) herds followed for over 35 years in northern Québec and Labrador, Canada, to evaluate the influence of environmental conditions on fall recruitment. Because the caribou herds studied inhabit a highly seasonal environment, we expected environmental conditions experienced either by adult females or calves to influence fall recruitment. Specifically, we expected fall recruitment to vary with environmental factors before and at the onset of estrus, during gestation, at calving and during the calves’ first months of life (Fig. 1). We predicted that local weather (temperature, precipitation, and snow conditions) and vegetation productivity during summer and fall migration (t-1) would influence recruitment through their effects on female condition before and at the onset of estrus (H1; Fig. 1). We also predicted that winter harshness and local weather during winter and spring migration (t-1) would influence recruitment through their effects on female condition during gestation (H2; Fig. 1). During the calving period (t), we predicted that local weather and spring green-up will influence recruitment through their effects on the condition of both mothers and calves. We also predicted that local weather and vegetation productivity during summer (t) will influence recruitment through calf growth and survival during its first months of life (H4; Fig. 1). Finally, because the two studied populations lived in proximity and shared part of their ranges at specific times of the year, they should have encountered similar environmental conditions. We thus tested for a Moran effect by evaluating the synchrony in fall recruitment between the two populations, expecting fall recruitment to be correlated. Materials and methods 2.1 Study area We studied two migratory caribou herds in northern Québec and Labrador over a period of about 35 years: the Rivière-George herd (RGH) from 1991 to 2024 and the Rivière-aux-Feuilles herd (RFH) from 1994 to 2024. Whitin the last three decades, both herds increased in size and peaked in the 1990s (RGH) and early 2000s (RFH), before declining steeply (Fig. 2A). RFH numbers reached up to approximately 628,000 individuals in 2001 (Couturier et al., 2004) before decreasing to about 182,000 individuals in 2019 (MELCCFP, unpublished data), while RGH numbers peaked at 823,000 ± 104,000 individuals in 1993 (Couturier et al., 2004) and declined to 8,600 ± 344 individuals in 2024 (MELCCFP, unpublished data). Migratory caribou undertake large-scale seasonal migrations. Each spring in April, females migrate from their winter range located in the boreal forest to their calving grounds located in the tundra, which they used from late May to June (Taillon et al., 2016). Typically, the biological year for migratory caribou starts at calving, around 1st of June (Fig. 1). After using their summer ranges in the tundra from June to September, caribou migrate back to their wintering areas between October and December (Le Corre et al., 2017). To estimate annual migration departure dates, we used satellite radiotracking data according to the method developed by Le Corre et al. (2014). This method, based on a first passage time approach, allowed determining each year the start and end dates of the different seasons, i.e. spring migration, calving, summer, fall migration, and winter (Le Corre et al., 2014). For each herd and year, we generated a seasonal 95% kernel from adult female caribou locations to define the areas used by females in each season (n = 12 to 85, Kie et al., 2010; Vuillaume, 2023, Fig. 2). We extracted environmental conditions (see below) at the scale of these seasonal areas (95% seasonal kernels) to represent the conditions faced by the adult females and their young for each herd and year. 2.2 Fall recruitment Each fall, thousands of caribou (mean ± SE; RFH: 3220 ± 273; RGH: 3528 ± 444) have been classified by age- and sex-classes (calves, adult females, and adult males) during the rut (end of October) to assess annual recruitment (1991-2024 except 1999; Taillon et al., 2016). To be representative of the population, classified groups of caribou were evenly distributed throughout the range covered by caribou fitted with GPS or telemetry collars. We estimated annual fall recruitment using the number of calves per 100 adult females (≥ 17 months old) in the population (DeCesare et al., 2012; Taillon et al., 2016). Local weather and large-scale climate conditions We assessed local weather conditions in different seasonal areas by extracting weather data from the North American Regional Reanalysis (NARR) provided by the National Oceanic and Atmospheric Administration’s National Center for Atmospheric Prediction (NCEP; Mesinger et al., 2006). We applied a Québec Lambert projection (epsg:32198) to center the data over the province of Québec. The NARR predictions, generated from weather station data across North America, are the most reliable data available for northern Québec and Labrador (Mesinger et al., 2006). We found a positive correlation for average daily temperatures (r [95% confidence interval CI] =0.81 [0.80 – 0.82]) and precipitation (r [95% CI] =0.56 [0.54 – 0.58]) between NARR data and data from 14 weather stations in northern Québec and Labrador (2006 to 2019), further supporting the reliability of NARR predictions (Vuillaume, 2023). We extracted data providing the average daily temperatures (°C), the total daily precipitation (kg/m2), the daily snow cover (% of the area covered by snow), and the daily snow accumulation (kg/m2) at a resolution of 32 km2. We averaged these daily values to describe the weather conditions for each season (i.e. spring migration, calving, summer, fall migration, and winter) for each herd and year. Snow cover referred to the percentage of the area covered by snow whereas snow accumulation indicated the average quantity of snow on a 1m² surface at a given time. This measurement included information on snow cover, additional snowfall, and the volume of melted snow. We did not extract snow data in summer. Precipitation included the average amount of both rain and snow on a 1m² surface at a specific time. We used the Normalized Difference Vegetation Index (NDVI) to characterize spring green-up and summer resource availability. NDVI provides information about vegetation productivity and phenology (Pettorelli et al., 2011). We extracted NDVI data for the calving and summer areas of each herd annually from the AVHRR (Advanced Very High Resolution Radiometer) sensor processed by the GIMMS (Global Inventory Monitoring and Modelling System) with a resolution of 8 x 8 km over a 16-day period (NDVI-3G+, Pinzon et al., 2023) between 1991 and 2018 for the RGH and between 1994 and 2018 for the RFH. We used the integrated NDVI in June (i.e. the sum of NDVI values for June; iNDVI) of the calving area as an index for the green-up during the calving period (Garel et al., 2011; Hamel et al., 2009b). As a proxy for summer vegetation productivity, we computed a cumulated NDVI (TiNDVI; Campeau et al., 2019; Mahoney et al., 2021) by performing an annual smooth curve of the NDVI time series over the summer area of each herd and using the area under the curve for the summer period (Vuillaume et al., 2023). Because the length of summer periods varied for each year, we divided the area under the curve by the number of days of the summer season. To describe large-scale winter conditions, we used the North Atlantic Oscillation index (NAO), which is based on the differences in surface sea-level atmospheric pressure between subpolar and subtropical regions over the North Atlantic (Hurrell et al., 2003). We used NAO anomalies covering December to April, which is negatively correlated with winter precipitation and temperature in northern Québec (see Supporting information). 2.4 Statistical analyses We assessed the effects of environmental conditions on fall recruitment from 1991 to 2018 for RGH and from 1994 to 2018 for RFH (n = 51) using linear beta regression models. We used the ”betareg” function of the ”betareg” package (with the argument type=“BC”, a maximum likelihood estimator with bias correction; Cribari-Neto and Zeileis, 2010) in R (version 4.3.2; R Core Team, 2023). We divided the number of calves per 100 females by 100 to include recruitment as a proportion (calf/female) in the beta regression models, but we present the results as “calves per 100 females” unless otherwise stated. Given the numerous environmental variables and limited sample size, we built several models based on biological hypotheses related to female body condition and calf survival for each season (Table 1). Variables included in the models covered seasons from the previous summer (the year before calving t-1, biological year before fall recruitment record) to the end of the summer after calving (t, biological year of fall recruitment record) (Fig. 1). As multicollinearity was present in the system, we excluded highly correlated variables (> 0.8; see Supporting information). Given the strong correlation between snow accumulation and snow cover in each season, we selected snow accumulation during winter and spring migration, but selected snow cover during fall migration and calving because it was more representative of energetic cost and vegetation access (Vuillaume, 2023; Fig. 1). We included the variable population (RGH and RFH) as a covariate in all models, as well as in a null model accounting only for the population effect. We applied logarithmic (average snow cover at calving) or non-linear transformations (average daily temperatures during spring migration) to meet models’ requirements (data distribution and homogeneity of variance). The variance inflation factor (VIF) values were low, with the highest value being 3.13 for model 2.4 (Table 1). We used a model selection approach performed with Akaike Information Criterion corrected for small sample size (AICc; Burnham and Anderson, 2002), using the “aictab” function of the “AICcmodavg” package (Mazerolle, 2020) to rank models. We considered models with AICc ≤ 2 as providing equivalent support (Burnham and Anderson, 2002) and presented estimates with their 95% confidence intervals (CIs) for all parameters included in equivalent models. To assess synchrony in fall recruitment, we computed the Pearson correlation between the calves per 100 females ratio of the two herds from 1994 to 2024 (n = 29). We also assessed Spearman correlations between environmental factors affecting fall recruitment of the two study sites from 1994 to 2018 (n = 25) using the “spearman.ci” function of the “RVAideMemoire” package (Herve, 2023) to estimate their 95% confidence intervals. Results Model selection revealed that five models were equivalent (AICc ≤ 2, see Supporting information). One model was related to conditions faced by females before estrus (1.7), three models were related to conditions faced by females during gestation (2.2, 2.3, and 2.4) and one model to conditions before and during calving (3.4) (Tables 1, 2 and Supporting information). The supported models for explaining fall recruitment included winter temperature (t-1) (AICc weight = 0.17), winter precipitation (t-1) (AICc weight = 0.15), temperature during calving (t) (AICc weight = 0.13), and fall migration precipitation (t-1) (AICc weight = 0.07; see Supporting information). One model considering both winter precipitation and winter temperature (2.4) also came out among the most supported equivalent models (AICc weight = 0.08; see Supporting information). Compared with the models where winter precipitation and winter temperature were included separately (2.2 and 2.3), the estimates for these two variables were weaker in this model (95% confidence intervals included zero in model 2.4), likely because of reduced power and collinearity (Pearson’s [95% CI]: r = 0.67 [0.49, 0.80], n = 51). Still, all estimates provided a similar interpretation of a reduction of fall recruitment with an increase in winter temperatures and precipitation. The total daily precipitation during fall migration (t-1), when females are in estrus, was negatively related to fall recruitment the following year (Table 2, Fig. 3A): high precipitation in fall (3.5 kg/m2) was associated with a 12 % [95% CI: 1; 22] decrease in fall recruitment, from 35 to 23 calves per 100 females, compared with low fall precipitation (1.9 kg/m2). In winter, when females are in gestation (t-1), average daily temperatures and total daily precipitation were also negatively related to fall recruitment (Table 2, Fig. 3B-C). Warm winter (-12℃) was associated with a 13 % [3; 23] decrease in fall recruitment, from 38 to 24 calves per 100 females, compared with cold winter (-21℃). Similarly, fall recruitment decreased by 13 % [3; 21], from 37 to 24 calves per 100 females, during high precipitation winter (2.0 kg/m2) compared with low precipitation (1.0 kg/m2). High daily temperatures at calving (t) were related to lower fall recruitment (Table 2; Fig. 3D): a warm calving season (14℃) was associated with a 12 % [2; 21] decrease in fall recruitment, from 37 to 24 calves per 100 females, compared with a cold calving season (4℃). The four environmental conditions affecting fall recruitment were positively correlated between the two herds (precipitation during fall migration; r = 0.50 [0.10, 0.80], winter temperatures; r = 0.59 [0.19, 0.82], precipitation during winter; r = 0.52 [0.16, 0.78], and temperatures at calving; r = 0.59 [0.23, 0.79], n = 25, Fig. A2). The ratio of calves per 100 females between the two herds was also positively correlated (r = 0.54 [0.22, 0.76], n = 29; Fig. 4). Nonetheless, the confidence interval was large, with recruitment in some years (e.g. 2012-2013) being relatively different between the two herds. Discussion In highly seasonal environments, variations in environmental conditions can exert a strong effect on vital rates such as survival and recruitment. Environmental constraints on vital rates can be exacerbated in migratory species that perform long-distance movement at specific periods of the year. Our study shows that environmental conditions experienced by adult females around the time of estrus in fall and during gestation in winter affected fall recruitment in migratory caribou. Conditions experienced by calves and their mothers during their first weeks of life also affected fall recruitment, but there was no evidence for an influence of environmental factors during calf rearing in summer. These results support three of our initial hypotheses, showing that environmental conditions encountered from the onset of estrus to the first weeks post-parturition are determinants of fall recruitment. Precipitation during the period of estrus, which coincides with fall migration in migratory caribou, negatively influenced recruitment the following year. In northern subarctic and arctic regions, precipitation in fall is usually a mix of snow and rain-on-snow events. High precipitation in the form of snow and an increase in rain-on-snow events can reduce resource availability (Wilson and Festa-Bianchet, 2009) and limit forage access because of the ice layers created (Dolant et al., 2018; Hansen et al., 2011). Access to vegetation is essential during fall migration because energetic needs are high (Parker et al., 2005), especially when migration coincides with the breeding season. Together, higher traveling costs because of the energy required for locomotion in snow (Fancy and White 1987) and reduced forage access may decrease female body condition or lower resource available for reproductive allocation at the onset of fall migration and the breeding season. Body weight and body fat were shown to be positively related to the probability of gestation in the RGH (Pachkowski et al., 2013) as well as in other caribou herds (Cameron et al., 1993; Parker et al., 2009). Because females of such long-lived species usually favor their survival over reproduction (Festa-Bianchet and Jorgenson, 1998), a threshold body mass and body fat seem required for a female ungulate to conceive (Barboza and Parker, 2008; Parker et al., 2009). Poor body condition can also delay estrus in female ungulates (Cook et al., 2004) and lead to later parturition date (Adams and Dale, 1998). In large herbivores, late birth date is associated with lower offspring survival (Bishop et al., 2009; Côté and Festa-Bianchet, 2001) as shown recently for the RFH (Vuillaume et al., 2023). Absence or delay of estrus due to harsh environmental conditions encountered during the breeding period may therefore decrease female productivity leading to lower fall recruitment. Compared with non-migratory populations that lower their energy expenditure and rely more upon body reserves than energy intake during winter, migratory caribou move to winter ranges offering more abundant resources to maximize their energy intake (Joly et al., 2024). In winter, we found that increased temperatures and precipitation negatively affected recruitment in migratory caribou. Warm and snowy winters could lead to additional energy expenditure through higher traveling costs by increasing sinking depth (Fancy and White, 1987) and lower food intake by decreasing resource availability and accessibility (DeMars et al., 2017; Dolant et al., 2018; Hansen et al., 2011; Solberg et al., 2001). Similarly to the fall season, increase rain-on-snow events can also reduce resource availability and accessibility in winter (Dolant et al., 2018; Hansen et al., 2011) which can lower female body condition. During gestation, the quantity, type, and consistence of precipitation, which partly depends on ambient temperatures, may lead to a decline in female body condition. In large herbivores, poor maternal condition during gestation could result in fetal mortality (Bishop et al., 2009) or decrease in fetal growth in late winter, a time when 80% of fetal mass is deposited (Barboza and Parker, 2008), leading to lighter calves at birth (Keech et al., 2000; Taillon et al., 2012). Lower birth mass decreases calf survival and fall recruitment in large herbivores (Bishop et al., 2009; Guinness et al., 1978; Mahoney et al., 2015). Winter conditions experienced by pregnant females may therefore impact recruitment through calf birth mass and survival. In caribou and reindeer populations, high winter precipitation has been shown to reduce birth mass (Adams, 2005; Couturier et al., 2009; Weladji and Holand, 2003), calf survival, and recruitment (Solberg et al., 2001). Other studies, however, showed contrasting results, with warm winters with less precipitation either reducing (boreal and mountain caribou populations in western Canada; DeMars et al., 2021) or increasing recruitment (mountain caribou herds of the Yukon Territory; Gonet, 2020; Hegel et al., 2010). The effect of precipitation may not be the same in warmer compared with colder regions because the correlations between temperatures and precipitation differ between ecosystems (e.g. negative correlation in western Canada as opposed to positive correlation in eastern Canada – in our study area (r = 0.67; 95% CI [0.49, 0.80])). Because temperatures and precipitation in winter are highly correlated, a larger sample size would be required to clearly distinguish both effects. Impacts of environmental conditions on recruitment should therefore be interpreted in the light of the limiting factors specific to a species, population, or ecotype. Furthermore, calving is a critical period for offspring survival in ungulates (Bishop et al., 2009; Nobert et al., 2016) and environmental conditions during this short season can exert a strong influence on recruitment (Gese et al., 2023; Hegel et al., 2010). As observed in other caribou herds (Russel et al., 2018), we found that high temperatures during calving decreased fall recruitment. High temperatures in June may result in drought periods, increasing heat stress, which could reduce recruitment (Russel et al., 2018). Warmer spring may also lead to earlier vegetation growth that may translate into increased availability of high-quality forage for herbivores (Hamel et al., 2009b), but also result in a mismatch between the peak of lactation and high vegetation productivity (Post and Forchhammer, 2008). With climate warming, green-up tends to be earlier in northern latitudes (Gonsamo et al., 2018; Park et al., 2020), but caribou calving phenology has not changed accordingly (Couriot et al., 2023; Post and Forchhammer, 2008). A mismatch could reduce forage quality during calving season, a period with high protein requirements for milk production (Barboza and Parker, 2008; Parker et al., 2009), therefore impacting growth and survival of juveniles (Scornavacca et al., 2016). In our 26-year study of two migratory caribou herds, we found no evidence for the influence of environmental conditions and vegetation productivity during summer on fall recruitment. Environmental factors experienced from conception to calving seem to have a stronger impact on recruitment than those likely influencing offspring summer survival. Therefore, maternal allocation to fetal growth and the newborn calf seems to play the major role in recruitment, highlighting the importance of female condition throughout gestation. Our findings agree with a 20-year study of a reindeer population in Svalbard, where winter conditions impacting maternal body mass explained most of the variation in recruitment (Veiberg et al., 2017). For species undertaking long migrations at high latitudes, summer conditions may be less of a limiting factor for recruitment. The scarcity of resources during migration and winter, and brevity of the growing season may, in contrast, impose greater limitations compared to ungulate species living in lower latitudes (Brown, 2011; Lukacs et al., 2018; Wagler et al., 2023). Although fall recruitment differed between the two herds in some years, annual fall recruitment was strongly positively correlated, suggesting an overall synchrony between the two herds. Environmental conditions influencing fall recruitment were positively correlated between both herds, and increased synchrony in environmental conditions at large-scales has been shown to enhance synchrony in population dynamics of caribou herds (Post et al. 2004) and other species (Kahilainen et al., 2018; Koenig and Liebhold, 2016). Greater risks of simultaneous decline and extinction of multiple populations could thus occur, as observed in three proximate caribou herds in northern Canada, where simultaneous decrease in abundance occurred under large-scale negative summer Arctic Oscillation intensity (Mallory et al., 2018). Because climate change is likely to change the variability and spatial extent of synchrony in environmental conditions at different scales, it can therefore affect synchrony in population dynamics (Hansen et al., 2020). Our study highlights the importance of environmental conditions, especially temperatures and precipitation, from conception to calving for recruitment in migratory caribou. High precipitation in fall and winter and warm temperatures in winter and during calving were found to individually decrease recruitment by 12% on average. Nevertheless, we could not assess their combined effect due to multicollinearity within the system and to avoid overparameterization. Thus, these effects are likely not additive and the large confidence intervals around the estimates (± 10%) show uncertainty of their strength. Overall, because conditions experienced by pregnant females influence recruitment and consequently population growth, our work emphasizes the benefits of long-term monitoring and precise large-scale environmental data for studying population dynamics and guiding conservation efforts. Synchrony observed in recruitment between herds also underlines the importance of considering multiple populations facing similar environmental conditions in conservation strategies as they may exhibit simultaneous fluctuations. With climate change, an increase in the frequency and intensity of extreme climatic events (e.g., rain-on-snow, early springs) might enhance their negative impact on population dynamics, especially for migratory species at high latitudes, where the speed and magnitude of changes are greatest (Robinson et al., 2009). Global changes therefore have the potential to affect synchronized populations similarly and could result in significant simultaneous demographic events. Data archiving statement: Data and R script will be available in the Open Science Framework depository upon acceptance. This private link can be used for peer review https://osf.io/u9pt4/?view_only=8cbef5fbf880406da9b821e3233ace79. Figure 1. Hypotheses reflecting on how environmental factors might affect fall recruitment of migratory caribou from the Rivière-George herd and the Rivière-aux-Feuilles herd in northern Québec and Labrador (Canada) . We expected environmental conditions throughout the year to influence fall recruitment; either by environmental factors faced by adult females before estrus (H1; summer and fall migration), by females during gestation (H2; winter and spring migration), by females and calves at calving (H3; calving period) or by calves during the first months of life (H4; summer). Local weather variables and large-scale environmental conditions used for each season are described in the arrows. NAO: North Atlantic Oscillation index; iNDVI: sum of the Normalized Difference Vegetation Index in June; TiNDVI: Normalized Difference Vegetation Index cumulated during summer; Snow cover: average snow cover; Cumulative snow: average cumulative snow; Temperature: average daily temperatures; Precipitation: average daily precipitation. t-1: period before calving – biological year before fall recruitment record; t: period from calving – biological year of fall recruitment record. Figure 2. A) Population estimates of two migratory caribou herds, Rivière-George (RGH, light blue) and Rivière-aux-Feuilles (RFH, dark orange) in northern Québec and Labrador, Canada. B) Historical ranges from 1987 (RFH) and 1991 (RGH) to 2023 and C) seasonal ranges in 2015 of RGH (light blue) and RFH (dark orange). The 90% confidence intervals are shown for the population estimates. Historical ranges are represented using 99.9% kernels while seasonal ranges are represented using 95% kernels. Black polygons represent the annual range of each herd in 2015. Figure 3. Changes in fall recruitment (calves per 100 females) with A) average daily precipitation during fall migration (t-1), B) average daily temperatures in winter (t-1), C) average daily precipitation in winter (t-1), and D) average daily temperatures at calving (t) of migratory caribou of the Rivière-George herd (1991-2018; n = 27; light blue dots) and the Rivière-aux-Feuilles herd (1994-2018; n = 24; dark orange dots) in northern Québec and Labrador, Canada. The average daily temperatures and the total daily precipitation were measured at the season scale. t-1: period before calving – biological year before fall recruitment record; t: period from calving – biological year of fall recruitment record. The black lines are the model predictions, and the shaded areas are the 95% confidence intervals. Dots are the raw data. Figure 4. Time series (A) and correlation (B) of fall recruitment of migratory caribou of the Rivière-George herd (RGH) and the Rivière-aux-Feuilles herd (RFH) in northern Québec and Labrador between 1994 and 2024. (A) Time series of the number of calves per 100 females for the RGH (light blue line) and the RFH (dark orange line). (B) Correlation between the number of calves per 100 females of the two herds for each year, where each point identifies the year. The red dotted line represents equality. Tables Table 1. Candidate models for explaining variation in fall recruitment (calves per 100 females) of migratory caribou of the Rivière-George herd (1991-2018) and the Rivière-aux-Feuilles herd (1994-2018) in northern Québec and Labrador (Canada), according to four hypotheses (Fig. 1). Null Population Hypothesis 1 Female body condition before estrus 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10 Population + Temp_summer_t1 Population + Pcp_summer_t1 Population + Temp_summer_t1 + Pcp_summer_t1 Population + TiNDVI_t1 Population + Temp_summer_t1 + TiNDVI_t1 Population + Temp_FM_t1 Population + Pcp_FM_t1 Population + Snc_FM_t1 Population + Temp_FM_t1 + Snc_FM_t1 Population + TiNDVI_t1 + Snc_FM_t1 Hypothesis 2 Female body condition during gestation 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 2.11 2.12 Population + NAO_t1 Population + Pcp_winter_t1 Population + Temp_winter_t1 Population + Pcp_winter_t1 + Temp_winter_t1 Population + Acs_winter_t1 Population + Acs_winter_t1 + NAO_t1 Population + Pcp_SM_t1 Population + Temp_SM_t1 Population + Acs_SM_t1 Population + Temp_SM_t1 + Acs_SM_t1 Population + NAO_t1 + Acs_winter_t1 + Acs_SM_t1 Population + Temp_SM_t1 + Acs_winter_t1 + Acs_SM_t1 Hypothesis 3 Female and calf body conditions during calving 3.1 3.2 3.3 3.4 3.5 3.6 3.7 Population + iNDVI_t Population + Snc_calving_t Population + Pcp_calving_t Population + Temp_calving_t Population + Pcp_calving_t + Temp_calving_t Population + Snc_calving_t + Pcp_calving_t + Temp_calving_t Population + iNDVI_t + Snc_calving_t Hypothesis 4 Calf survival during summer 4.1 4.2 4.3 4.4 4.5 4.6 Population + TiNDVI_t Population + Pcp_summer_t Population + Temp_summer_t Population + Temp_summer_t + Pcp_summer_t Population + Temp_summer_t + TiNDVI_t Population + iNDVI_t + Snc_calving_t + TiNDVI_t NAO: North Atlantic Oscillation index; iNDVI: sum of the Normalized Difference Vegetation Index in June; TiNDVI: Normalized Difference Vegetation Index cumulated during summer; Snc: average snow cover; Acs: average cumulative snow; Temp: average daily temperatures; Pcp: average daily precipitation; SM: spring migration; FM: fall migration; Population: the Rivière-George herd or the Rivière-aux-Feuilles herd; t1: period before calving – biological year before fall recruitment record; t: period from calving – biological year of fall recruitment record. Table 2. Results of the model selection based on Akaike’s Information Criterion adjusted for small sample size (AICc) evaluating which environmental factors were most supported for explaining variation in fall recruitment (modelled as calf/female) of migratory caribou of the Rivière-George herd (RGH; 1991-2018) and the Rivière-aux-Feuilles herd (RFH; 1994-2018) in northern Québec and Labrador, Canada ( n = 51). Only equivalent models (ΔAICc < 2) are presented and compared with the null model (see Supporting information for the full list of models). Variables for which 95% confidence interval excluded zero are indicated in bold. Model 2.3 Intercept Winter temperature (t-1) Population (RGH) -1.86 ± [-2.73, -0.99] -0.07 ± [-0.12, -0.02] -0.20 ± [-0.48, 0.07] 4 0.00 0.17 Model 2.2 Intercept Winter precipitation (t-1) Population (RGH) 0.11 ± [-0.52, 0.74] -0.63 ± [-1.10, -0.17] 0.01 ± [-0.32, 0.33] 4 0.22 0.15 Model 3.4 Intercept Calving temperature (t) Population (RGH) -0.18 ± [-0.63, 0.28] -0.06 ± [-0.11, -0.01] -0.19 ± [-0.47, 0.09] 4 0.49 0.13 Model 2.4 Intercept Winter precipitation (t-1) Winter temperature (t-1) Population (RGH) -0.94 ± [-2.98, 1.10] -0.34 ± [-1.04, 0.36] -0.04 ± [-0.12, 0.04] -0.09 ± [-0.45, 0.28] 5 1.61 0.08 Model 1.7 Intercept Fall migration precipitation (t-1) Population (RGH) 0.21 [-0.58, 0.99] -0.37 [-0.69, -0.06] -0.17 [-0.46, 0.11] 4 1.76 0.07 Null model Intercept Population (RGH) -0.71 [-0.92, -0.50] -0.24 [-0.53, 0.05] 3 5.42 0.01 K: number of model parameters; ∆AICc: AICc differences; Wi: Akaike weights; RFH is the reference category represented by the intercept in all models. Information & Authors Information Version history V1 Version 1 11 November 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords environmental conditions gestation maternal allocation migratory caribou recruitment synchrony Authors Affiliations Roxanne Turgeon 0000-0001-5087-9373 [email protected] Université Laval View all articles by this author Alexandre Carbonneau 0009-0002-4857-847X Université Laval View all articles by this author Fanny Lescouzeres 0009-0007-2486-5518 Université Laval View all articles by this author Barbara Vuillaume View all articles by this author Sandra Hamel 0000-0003-1126-8814 Laval University View all articles by this author Joëlle Taillon Gouvernement du Québec Ministère de l’Environnement de la Lutte contre les changements climatiques de la Faune et des Parcs View all articles by this author Metrics & Citations Metrics Article Usage 608 views 129 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Roxanne Turgeon, Alexandre Carbonneau, Fanny Lescouzeres, et al. Conservation of migratory species at risk: Environmental conditions experienced from pre-gestation to parturition affect fall recruitment in migratory caribou. Authorea . 11 November 2025. DOI: https://doi.org/10.22541/au.176286075.59304212/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.176286075.59304212/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fff0ba53e1306f3',t:'MTc3OTQ4NjcyMA=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();