Seasonal and annual variation in survival is the consequence of changes in weather, food availability, predators and density (Loison and Langvatn 1998; Gaillard et al. 2000; Sæther et al. 2004; Garel et al. 2004; Altwegg et al. 2005). For a given life history strategy and environment, specific seasons are more demanding than others and include most mortality (Schaub and Vaterlaus-Schlegel 2001; Crespin et al. 2002; Schuyler et al. 2019). Identifying the most important seasons for variation in survival increases understanding of the causes of mortality and the impacts of different environmental factors (Giavi et al. 2014; Rockwell et al. 2017).
The non-breeding season is often physically the most demanding for resident species living in northern areas (e.g., Watts and Jonkel 1988; Lahti et al. 1998; Janke and Gates 2012) whereas it is not straightforward for migratory animals that move across vast distances during their annual cycle (Bauer and Hoye 2014) and face hugely different environments that have potential effects on survival (Calvert et al. 2009; Inamine et al. 2016; Rushing et al. 2017). Migration is an adaptation that optimizes reproduction and survival (Newton 2010; Avgar et al. 2013). Migrating animals can increase chances of survival by following spatio-temporal variation in food availability or other limiting resources such as water and nutrients but also by escaping predators and parasites (Fryxell et al. 1988; Lank et al. 2003; Bolger et al. 2008; White et al. 2014; Mysterud et al. 2016). During the journey itself, however, individuals may also be vulnerable to predators or be physically strained (Nicholson et al. 1997; Hebblewhite and Merrill 2007; Klaassen et al. 2014; Robinson et al. 2020). Understanding the causes of temporal variation in survival among migratory animals thus requires the consideration of the whole annual cycle. Yet, few studies address seasonal variation in demographic parameters of migratory species as it requires multiple sampling points during the year or tracking of individuals in time (Calvert et al. 2009; see also Marra et al. 2015).
Ungulates are known for their annual cycle that involves migration strategies between the more stationary periods at the reproductive and wintering grounds (Kauffman et al. 2021). Migrating ungulates are particularly susceptible to perturbations in the environment (Aikens et al. 2020) which underscores the need for research on survival in the separate phases of the annual cycle (Bolger et al. 2008). Survival of ungulates is primarily affected by density via resource limitation, predators and their interactions (Bolger et al. 2008). Predator populations can limit ungulate populations especially in the northern temperate zone where ungulates constitute the main prey species of large carnivores (Bergerud and Elliott 1998; Vucetich et al. 2005; Kojola et al. 2004, 2009; Gurarie et al. 2011). Harsh environmental conditions during winter, which result in lack of food, may increase predation (Tveraa et al. 2003; Kautz et al. 2020), and consequently cause seasonal variation in survival. Also, conditions during autumn may influence fat reserves and survival over the wintering period (Cook et al. 2004; Hurley et al. 2014).
The reindeer (Rangifer tarandus) is a migratory ungulate with a circumpolar distribution in the northern hemisphere and different subspecies living in forest and tundra biomes (Gunn 2016). The world population is decreasing and categorized as vulnerable (Vors and Boyce 2009; Gunn 2016). Especially, the North American woodland caribou (Rangifer tarandus caribou, Gmelin 1788) and the Eurasian subspecies, the wild forest reindeer (Rangifer tarandus fennicus, Lönn. 1909, hereafter ‘WFR’), have declined in the last decades (McLoughlin et al. 2003; Kojola et al. 2009; Hervieux et al. 2013; Danilov et al. 2018; Paasivaara et al. 2018). The declines in North America are linked to changes in calf and female mortality (Wittmer et al. 2005; Hervieux et al. 2013), which constitute the majority of population growth rate variation in woodland caribou populations (DeCesare et al. 2012). Importantly, nothing is known about long-term changes or annual, seasonal and spatial variation in female survival, or the main mortality causes in the WFR populations.
In North American caribou populations, variation in adult survival is linked to predation, apparent competition and weather patterns (Hervieux et al. 2014; Schmelzer et al. 2020). Predation by wolves (Canis lupus, L. 1758) is often the main cause of mortality in woodland caribou populations, which are ecologically similar to the WFR (Wittmer et al. 2005). Many caribou populations are declining as a consequence of increased predation (Wittmer et al. 2005). While this increase has been partly affected by anthropogenic habitat change induced apparent competition with moose (Alces alces, L. 1758) and white-tailed deer (Odocoileus virginianus, Zimmermann 1780) (Wittmer et al. 2010; Whittington et al. 2011; Latham et al. 2011), predation rates among caribou are commonly linked to the degree of spatial overlap with predators with predation being higher in areas and seasons of higher overlap (McLoughlin et al. 2003; Wittmer et al. 2005). WFR calves are also known to be significant prey of wolves and other large carnivores in Finland (Heikura 1997; Kojola et al. 2004, 2009; Gurarie et al. 2011) suggesting that large carnivores also consume adult females but published data on mortality causes are lacking.
Here, we study survival of adult female WFR in the two distinct subpopulations living in Finland (Kainuu and Suomenselkä) using the fates of 271 females that were tracked using GPS transmitters over ten years (2009–2020). We determine cause of mortality for each death and evaluate the importance of predation as source of mortality compared to other causes (traffic and unknown causes) and consequently examine spatial, temporal (annual) and seasonal variation in cause-specific mortalities. Large carnivore populations in Finland, including wolf, rapidly recovered and expanded their distribution from Eastern Finland towards Western and Southwestern Finland as a result of increasing population trends due to conservation strategies and hunting control after Finland joined to the European Union (Kojola and Määttä 2004; Kojola et al. 2006; Chapron et al. 2014). In March 2020, there were 46 wolf territories in Finland (Heikkinen et al. 2022, see also Mäntyniemi et al. 2022). The brown bear (Ursus arctos, L. 1758) and the lynx (Lynx lynx, L. 1758) were more evenly distributed in Finland than the wolf population with about 2670–2800 individuals (Heikkinen et al. 2021) and 2155–2280 (Holmala et al. 2020), respectively (see Heikkinen et al. 2022, Chapron et al. 2014). In addition, there were about 390–400 wolverines (Gulo gulo, L. 1758) in Finland during winter 2021. The wolverine population has shown a positive trend from early 2000’s to early 2020’s, especially outside of the reindeer management area (Kojola et al. 2021b). Because WFR have been normal prey for all carnivores (see Heikura 1997) and especially for wolf (Kojola et al. 2004, 2021a; Gurarie et al. 2022), we hypothesise that predation by large carnivores is the main cause of mortality, and hence variation in survival, and especially mortality due to predation is linked to wolf densities in the distribution of the WFR.
Firstly, we test whether survival and cause specific mortalities differ between the subpopulations. During 2009–2020, the density of wolves in WFR distribution were approximately 3.7 times higher in Kainuu compared to Suomenselkä (Fig. 1b). On the basis of overlap between wolf territories and WFR distributions as well as the difference in wolf population density between WFR subpopulations, we predict that survival is lower and mortality caused by predation higher in Kainuu compared to Suomenselkä (Fig. 1b; Online Resource 1). Secondly, we test whether these subpopulations show inter-annual variation in survival and cause-specific mortalities. During the study years 2009–2020, the subpopulation in Kainuu declined slightly (λ = 0.986) whereas the subpopulation in Suomenselkä grew (λ = 1.049; Fig. 1a). Hence, we test for temporal trends in survival and cause-specific mortalities.
Finally, we focus on seasonal variation. The annual cycle of WFR is determined by movement between wintering, calving and rutting areas (Fig. 2). Migration occurs in early May between wintering grounds and calving sites, and vice versa after the rutting season from late October to December. Behaviour can be characterized as mixed or partial migration (Kauffman et al. 2021) since the distance between areas varies from few to hundreds of kilometres (Pulliainen et al. 1986). Wintering from January until late April takes place at relatively small areas which include good foraging sites (WFR can dig food under the snow, e.g., lichens) and open spaces, such as frozen lakes or open mires, where animals are better protected by the herd’s vigilance (Pulliainen et al. 1983). The timing of spring migration varies between years and individuals but usually occurs in the turn of the month from April to May. Moreover, it happens in a shorter time frame compared to autumn migration (Pulliainen et al. 1986). Calving season occurs during spring and summer. Females and calves spend time until August in areas characterized by a mosaic of lakes, mires and forests that provide diverse nutrition for the calf and lactating female (Helle 1980). Rutting develops during September and lasts until October. Females form herds with one dominant male with whom copulation occurs (Kojola 1986). This is followed by the long autumn migration period during which they move slowly towards the wintering areas. On the basis of temperature variation and spatial overlap between wolves and WFR, we predict that survival is lowest and predation rates are highest during winter.