4 ǀ DISCUSSION
We found that as a residential species that have to share the same space with human, pairs of Himalayan marmots in Zoige wetland tended to dig more burrows relative to pairs in natural habitat, resulting in shorter inter burrow distances if they are suffered persistent but nonfatal disturbance from human activities (Bryant, 1996; Griffin et al., 2007) due to their high phenotypic plasticity (Huang et al., 1986; Poudel et al., 2015a; Poudel et al., 2015b). Besides, most breeding pairs in the region generally prefer to construct their reproductive burrows on mounds. Specifically, relative to pairs in HDH selectively dig their reproductive burrows on mounds whether the mound volume or the distance to road, pairs in the LDH did not show the same preference and on average dug their burrows farther away from roads.
Different from the fatal threatens like poaching and habitat loss that will directly kill affected animals (Tilson et al., 2004; Rija et al., 2020), residents in our study site never kill marmots due to their religious faith, but their daily activities are a persistent disturbance for the animals, and the level of disturbance differs between the habitats due to different road locations and the number of motorvehicles passing through. Stray dogs are a deadly threat to marmots (Poudel et al., 2015a), but although they are more abundant in the HDH and LDH, several garbage dumps in the area are capable of supplying enough food for them (Figure 1B), dogs around the village do not go out of their way to hunt marmots within their range (Altmann & Muruth, 1988). No stray dog predation on marmots was observed during our fieldwork, the main human influence on the marmot populations comes from motorvehicles that pass through the habitat. Seasonal fluctuation in the intensity of automobiles and dogs in the natural habitat occurs due to residents driving by with their dogs during the annual seasonal rotation of pasture, while daily trips between the village and pastures are done by motorcycles (Table 1).
All marmot breeding pairs dig a reproductive burrow for regular use to rest, reproduce and hibernate, but they also dig temporary burrows, which are occupied less frequently, throughout the home range as a refuge when threatened (Blumstein et al., 2001; Zhang et al., 2019). All breeding pairs around the village dig multiple burrows for shelter, nevertheless, pairs in HDH dug more pair-specific burrows (19.76 ± 1.4) than pairs in the LDH (13.4±0.96) and NH (10.8±0.73), probably due to they suffer the heaviest disturbance (Figure 2). Though nonsignificant, pairs in LDH generally dig two more burrows than NH pairs, more available refuges guarantee individuals have more chances of escape, and consequently, provide a survival advantage when threatened (Blumstein et al., 2001). Furthermore, shorter inter all burrow distances resulting from more burrows in the habitats enable marmots in two disturbed habitats to reach a potential refuge more quickly when threatened, increasing the likelihood of survival (Li et al., 2011; Zaman et al., 2019). Based on observations recorded from 2017 to 2020, no new temporary burrows were dug. It is possible that more burrows were dug in the HDH during the initial human settlement of the area, but marmots that had grown accustomed to humans’ daily activities no longer saw a benefit to digging new burrows (Mainini et al., 1992; Schell et al., 2018), which is energetically expensive.
Similarly, though only two more temporary burrows were dug, inter all burrow distance in LDH is far shorter than in NH, allowing the same reduction in distance and time required to reach a safe place for individuals in the habitat as their congeners in HDH. The different (number of burrows per pair) and same (inter burrow distance) patterns that emerge between two disturbed habitats may arise because the disturbances LDP individuals suffer are not intense enough to accustom them, but drive they selectively concentrate new burrows near reproductive burrows like urban woodchucks (Watson, 2009), the mean home range of LDH pairs (21.98 ± 2.86 are) is far smaller than pairs in NH (39.62 ± 2.55 are) with there are many unoccupied regions among different pairs in LDP (Figure S4), consequently, gain shorter inter all burrow distance to meet the requirements of flee efficiency and spend as little energy as possible on digging extra burrows simultaneously. Meanwhile, FID of HDP individuals (65.36 ± 4.45 m) are shorter than NP individuals (119.40 ± 8.11 m) as expected, nevertheless, even have the shortest inter all burrow distance, the FID of LDP individuals (105.00 ± 6.88 m) showed no coincident trend as HDH, but are as long as FID of marmots in NP (Dill & Houtman, 1989; Griffin et al., 2007). The differentiation may arise because the optimal strategy to survival for LDP individuals is to flee early like NP individuals when threatened however the distance to a potential refuge (Li et al., 2011). Shorter flee distance and longer FID guarantees the safety of unaccustomed LDP individuals under the disturbances of human activities (Zaman et al., 2019; Feng & Liang, 2020).
It is also worth noting that the inter reproductive burrow distance in HDH (118.31 ± 36.82 m) is shorter than that of the other two habitats (Figure 4A), a pattern may arise because the regions surrounding the HDH are uninhabitable due to improper soil and vegetation characteristics (Guo et al., 2020; Figure S1), HDP is actually an isolated population cannot freely communicate with other populations. The same as a reintroduced Alpine marmot population in Dolomiti Bellunesi National Park, Italy (Borgo et al., 2009), the HDH has been fully exploited by the growing breeding pairs since the village began to settle in the region. 72 pairs per km2 may be the maximum environmental carrying capacity for the species in such a ecosystem. In contrast, no similar variation emerged between LDP and NP, may because LDH is an open area conducive to free dispersal as the NH (Figure S2 to S5). This might explain why the inter burrow distance for reproductive burrows in the LDH was no different than that observed in the NH. LDH is an open space, marmots in the region have the freedom to actively avoid human influences in emigration, a strategy that is superior to the passive adaptation to human influence. The average inter reproductive burrow distances observed in LDH (143.73 ± 48.25 m) and NH (145.57 ± 38.66 m) may reflect more typical distancing between marmot pairs: reducing resource competition while maintaining regular contact between pairs. Together with the diversification on FID and two inter burrow distances, we concluded that compared with HDP, the reactions of LDP individuals may be the normal outcomes (dig more extra and concentrated temporary burrows and flee earlier to avoid potential dangers but also appropriate inter “family” distance) when Himalayan marmot affected by persistent, but nonfatal disturbances from humans.
The characteristics of reproductive den site selection also differed among the habitats. Most NH pairs constructed their reproductive burrows on mounds, and pairs in HDH also selectively dig their reproductive burrows on mounds (Figure 5A), even when those mounds were relatively close to a road and smaller than the mounds used by pairs in NH (Figure 5B, 5C). Marmots use their reproductive burrows giving birth to their offspring, and spending a lot of time resting/basking at the entrance to the burrow. This special preference to mound may because pairs build their reproductive burrows on mounds ensure better drainage relative to burrows dug on flat ground (Szor et al., 2008). Besides, similar to alpine marmots preferentially remaining near large stones that they climb to engage in surveillance to watch for predators more effectively (Borgo, 2003), Himalayan marmots in alpine meadow with less mound also selectively use site with many big stone to gain better vigilance and bask efficiency (Figure S6). However, in our site in Zoige wetland, due to the lack of large stone, rest or vigilance on mound higher than flat ground may also able to gain a better vision of the surrounding areas, improving their chances of detecting predators.
Most animals choose to locate reproductive dens at sites where they can conceal themselves to better protect themselves and their offspring (Ross et al., 2010; May et al., 2012; Sazatornil et al., 2016; Lai et al., 2020). Consequently, we predicted that pairs of Himalayan marmot would stay as far from the roads as possible, but breeding pairs in HDH still preferentially built their reproductive burrows on the mounds near roads despite the increased frequency of disturbance from the motorvehicles, which can be harmful (Whittington et al., 2019). This surprising result suggests that the availability of mounds is the primary determinants of site selection for reproductive burrows in Himalayan marmots. In Zoige wetland, mounds on the dry flat ground could be the limiting resource (Guo et al., 2020), as marmots always built burrows in the mounds that were present regardless of their size or distance from the road. For example, one occupied mound (HDH11) in HDH was only 2.2 meters from a road (Figure S3), and the average size of the occupied mounds in the HDH is smaller (2.14 ± 2.65 m3) than the occupied mounds in the NH (6.23 ± 5.13m3) (Figures 6A, 6B), indicating that marmots will use all the mounds they can find in an area, even smaller ones. There were no unoccupied mounds left in the HDH, and some breeding pairs that could not find a natural mound will built their own very small mounds around the entrance of their burrows (Figures 6C, 6D). There are no natural hiding places for marmots in the Zoige wetland (Zhang, 2019), and unlike predators, disturbances from daily human activities are nonfatal, consequently, sites that allowed for vigilance while resting were the only suitable choices for reproductive burrows, even if they were frequently disturbed by motorvehicles. Den site selection of American black bears (Ursus americanus ) and American badger (Taxidea taxus ), and the habitat utilization of Barbary macaques (Macaca sylvanus ) were also found to be unaffected by the distance to roads (Waller et al., 2013; Sunga et al., 2017; Waterman et al., 2019), suggesting that many species will tolerate persistent but non-life-threatening human disturbance to retain access to otherwise favorable habitat. The importance to the marmots of the vigilance and good drainage of mound-built burrows (Szor et al., 2008) outweighed disadvantages to digging reproductive burrows close to a road. Furthermore, dig their reproductive burrows near road may also arise because relative to other species sensitive to human disturbance (i.e. snowy plover Charadrius nivosus and Yunnan lake newt) (Webber et al., 2013; Amphibiaweb, 2021), marmots species are more able to endure nonfatal human disturbances (Neuhaus & Mainini, 1998; Griffin et al., 2007). Himalayan marmots disturbed in HDH for generations are highly accustomed to human activities are no longer selective pressures on the den site selection of the marmots in the habitat (Schell et al., 2018).
Site selection for reproductive burrows in the LDH showed a different profile relative to the other two populations, with burrows almost equally likely to be located on mounds or on flat ground. Moreover, the average volume of the mounds selected for reproductive burrows in the LDH is significantly larger (75.47 ± 78.69 m3) than the mounds in HDH and NH. This discrepancy might result from the radically different topography of the area. Aside from having many large mounds, the LDH is sloped, with some areas of the flat ground allowing for surveillance equal to the tops of mounds in the other two habitats (Figures 6E, 6F). Consequently, pairs in LDH are no longer limited by the availability of mounds. This is consistent with the greater average distances from reproductive burrows to the road in the LDH (98.06 ± 48.06 m) as opposed to the HDH (28.88 ± 12.29 m). Unlike the marmots of HDP, who are forced to prioritize vigilance and drainage, marmots in LDH have greater flexibility in sites where they can build reproductive burrows and so tend to avoid the roads.
Unlike reproductive burrows, temporary burrows were common on flat ground in all three habitats because they were used only to evade immediate threats. Good vision and drainage are not important for temporary burrows (Borgo, 2003; Szor et al., 2008). Consequently, Himalayan marmots dig temporary burrows in any location as needed and reserve their reproductive burrows for mounds when possible. This demonstrates the use of multiple habitat utilization strategies at once to cope with human disturbance and natural dangers.
Generally, relative to animals sensitive to human activities like Yunnan lake newt or some certain populations suffer extensive human disturbance like Asiatic lions, Himalayan marmot have a high plasticity, variation in habitat utilization in response to the varied intensity of nonfatal human disturbance of the species emerged, and heavier suffered population even gain a higher population density (Guo Cheng personal observation). Furthermore, it is also possible that other aspects of this species’ ecology, such as if the feeding range size of LDP individuals shows the same trend with their home ranges and if their time budget, body condition, may also change in response to human activity to improve survival, as has been observed in other animals require further study (Wright et al., 2009; Poudel et al., 2015b; Santini et al., 2019; Yang et al., 2019).