Discussion
Apex predator actively harvests prey population and suppress mesopredator community structure through top-down mechanisms (Ripple et al., 2014). However, there is a dearth of evidence for subordinate predator suppression due to lack of such natural experiment study systems. We got an opportunity to understand the impact of a solitary apex predator on competitively inferior social predator in a natural experiment setup.
The Asiatic wild dog/ dhole is listed under “Endangered” category by IUCN, with about only 949-2215 mature individuals surviving globally. Indian subcontinent harbors majority of the viable dhole population (Kamler et al., 2015). However, the species has been exterminated from 60% of its past distributional range in India (Karanth et al., 2010) due to decline in prey base, habitat loss, human persecution and intra-guild competition (Hayward, Lyngdoh, & Habib, 2014; kamler et al., 2015). In India, persistence of dholes is mostly confined to protected areas with infrequent presence records from secondary forests and agro-forest plantations (Srivathsa, Karanth, Kumar, & Oli, 2019). These remnant habitat patches are also shared by other large carnivores, thereby limiting dhole population by lethal intra-guild interactions (Steinmetz, Seuaturien, & Chutipong, 2013). Small size of PAs and lopsided conservation efforts can further result in over-inflated apex carnivore densities and be detrimental for subordinate predators like dholes (Rayan & Linkie, 2016; Kumar, Awasthi, Qureshi, & Jhala, 2019). Therefore, to conserve the endangered dhole, it is crucial to understand its response to tigers, itself an endangered and flagship species. A holistic understanding of guild interactions would improve our ability to optimize ecological triage while conserving charismatic apex predators and mid-ranking predators.
In this study, we found that the pack size was reduced in higher tiger density areas. This might be due to the influence of interference competition wherein a narrower niche is available for dholes to occupy which limits their persistence. Studies from Africa also show a similar trend where pack sizes of African wild dogs were reduced in areas of higher lion densities (Groom, Lannas, & Jackson, 2017). At high tiger density site (TATR) we also observed direct intraguild predation events i.e. tigers killing dholes (n= 5) however, no such events have been observed in NNTR, the low tiger density site. Similar events have been documented in African system by Creel and Creel (1996), where lions actively chase away dogs and in certain instances even kill them.
Conversely, in low tiger density area, larger pack size is indicative of a competitive release scenario (Groom, Lannas, & Jackson, 2017). A similar trend has been observed in spotted hyena clan size and pack size of African wild dog due to reduced interspecific competition in a lion depleted system and easy availability of food subsidies (Periquet, Fritz, & Revilla, 2014; Green, Farr, Holekamp, Strauss, & Zipkin, 2019; Groom, Lannas, & Jackson, 2017).
There was no significant difference between the individual prey densities at both studies, however, the higher density of a large-body size prey in NNTR is also interesting to note. This finding is in line with the idea that low numbers of apex predator lead to an increase in abundance of large-bodied prey in a system. The subordinate predators are limited by body-size and cannot target such large prey and this might be one of the possible explanations for our finding. Similar pattern is observed in study systems where wolf (Canis lupus ) populations are controlled (Burgar, Burton, & Fisher, 2019) that not only causes coyote (Canis latrans ) abundance to increase but also leads to stable number of adults in caribou (Rangifer tarandus ) population since coyotes are size-limited and cannot take down larger individuals (Lewis, Gullage, Fifield, Jennings, & Mahoney, 2017).
To examine whether reserve level patterns were concurrent with global scale, we did a distribution-wide assessment of pack size across dhole distributional range. We found support for our first two hypotheses, dhole pack size were positively associated with prey density and negatively associated with tiger densities. Statistically significant but moderate effect size reflects that the patterns were diffused at global scale. Patterns of subordinate predator response to apex predators numbers do not always mirror each other over temporal and spatial scales (Jachowski et al., 2020), pertaining to area specific ecological settings, anthropogenic disturbances, level of protection at reserve level and connectivity in the landscape at regional scale (Greenville, Wardle, Tamayo, & Dickman, 2014; Newsome, & Ripple, 2015). We did not find support for our third hypothesis, dhole pack size was not associated with high terrain ruggedness however, a low pack size has been reported from studies of high altitude and rugged terrain (Bashir, Bhattacharya, Poudyal, Roy, & Sathyakumar, 2014; Lyngdoh, Gopi, Selvan, & Habib, 2014).
Taken together, our reserve level and distribution wide assessment results are in support of the hypothesis that area specific ecological settings such as prey composition, abundance, and co-predator density are drivers of carnivore community structure, group size and behavior (M’soka, Creel, Becker, & Droge, 2016). Patterns of this study also adhere to trends that have been observed in other long-term studies from Serengeti National Park, Ngorongoro Conservation Area and Selous Game Reserve, Tanzania, where the group size of subordinate predators are positively associated with availability of food resources and negatively associated with, apex predator numbers (Creel & Creel, 1996; Creel & Creel, 1998; Woodroffe & Ginsberg, 1999; Pole, 2000; Périquet, Fritz, & Revilla, 2015; Green, Farr, Holekamp, Strauss, & Zipkin, 2019).
Tiger and dhole belong to the same carnivore guild because of similar diet spectrum (Hayward, Lyngdoh, & Habib, 2014), and therefore their persistence in an area is linked to each other (Johnsingh,1992). Studies based on non-invasive sampling such as camera trapping and occupancy surveys suggest that carnivores adapt to mechanisms of niche partitioning, depending on resource availability to facilitate sympatry (Karanth et al, 2017). Very few studies have demonstrated that intense intra-guild competition result in reduced density and spatial displacement of the subordinate predators (Harihar, Pandav, & Goyal, 2011; Mondal, Gupta, Bhattacharjee, Qureshi, & Sankar, 2012; Steinmetz, Seuaturien, & Chutipong, 2013). Our study provides a new direction in understanding how intra-guild competition can also potentially limit pack size of dhole.
Competition from dominant carnivores remains one of the major factors that drives densities of other sympatric carnivores in a system (Wrangham, Gittleman, & Chapman, 1993). NNTR and TATR are a comparable natural experiment setup that allows us to study how these two endangered tropical carnivores interact and understand consequences of dominance shift. We suggest that the increased pack size of a social subordinate predator (dhole) is an outcome of low density of apex predator and availability of wider niche. The decision to form large packs as compared to small packs is going to be more advantageous to limit the growth of top predator. However, future long-term studies are suggested based on multiple pack-year data, disease prevalence, predation, kleptoparasitism events and recruitment rate of dholes, across a gradient of apex predator density would further validate our hypothesis.