Influences of Predator Cues on the Incidence of Ungulates, Mesopredators and Top Predators in the Greater Khingan Mountains, Northeastern China

Hamenya Mpemba1, Fan Yang 1, Kirsty J. MacLeod2,3, Dusu Wen1, Yan Liu1,4 and Guangshun Jiang1,*

1Feline Research Center of National Forestry and Grassland Administration, College of Wildlife and Protected Area, Northeast Forestry University, 26 Hexing Road, Harbin, Heilongjiang 150040, P.R. China

2Department of Ecosystem Science and Management, Pennsylvania National University, Forest Resources Building, University Park, PA 16802, USA

3Department of Biology, Lund University, Sölvegatan 37, 223 62 Lund, Sweden

4General Station for Surveillance of Wildlife-Borne Infectious Diseases, State Forestry and Grassland Administration, Shenyang, Liaoning Province, 110034, PR China

Hamenya Mpemba and Fan Yang have contributed equllay to this article.

ABSTRACT

Top predators can affect the behaviour of prey species via lethal (direct kill) or non-lethal effects (i.e., through predation risk). For example, prey species may move from areas perceived as risky to safer spaces where predation risk is lower, which can have important consequences for investment in foraging, movement, and mating, and for the behaviour and habitat use of other species, such as mesopredators. These changes in prey and mesopredator behaviours are likely mediated by the presence of predator cues in the environment. Here, we test how different predator cues (visual and odor) from familiar and novel predators (brown bear and Amur tiger, respectively) influence ungulate, mesopredator, and top predator visitation rates to camera trap sites in a national nature reserve in China. The comparison of these predator types is of particular interest in this region as Amur tigers may shortly be reintroduced here. We found that visual but not odour cues significantly affected ungulate visitation rates: ungulates showed reduced visitation to sites with either a novel or familiar visual predator cue. When combined, mesopredators and top predators also showed a small reduction in visitation rates to tiger cue sites compared to bear cue sites, suggesting a possible novel predator effect. The generalisation and contextual importance of predator cues for prey and mesopredators have been little studied. Understanding how species respond to novel cues may help to determine extinction probabilities and overall plasticity in the face of change. This study is, therefore, an important step forward in understanding predator cue responses at the community level. This is also the first study to test the ecological function of Amur tiger cues in the wild environment and may serve as essential information in the rewilding process of captive Amur tiger plans.

Article Information

Received 13 February 2019

Revised 10 March 2019

Accepted 15 March 2019

Available online 07 March 2022

(early access)

Published 27 October 2022

Authors’ Contribution

GJ, HM and FY conceived and designed the study. HM, FY, DW and YL conducted field work and collected data. KM and HM analysed the data and revised the manuscript. HM wrote the manuscript.

Key words

Predator cues, Ungulates, Mesopredators, Panthera tigris altaica, Camera trap, Fear ecology.

DOI: https://dx.doi.org/10.17582/journal.pjz/20190213140239

* Corresponding author: jgshun@126.com

0030-9923/2023/0001-269 $ 9.00/0

Copyright 2023 by the authors. Licensee Zoological Society of Pakistan.

This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

Introduction

Top predators can affect the behaviour of prey species via lethal (direct kill) or non-lethal effects (i.e., through predation risk) (Lima and Bednekoff, 1999; Tolon et al., 2009; Farnworth et al., 2016). Predation risk their activity patterns when confronted with risk (Lima and Bednekoff, 1999; Tolon et al., 2009). Differential use of habitats due to variation in predation risk is known as the landscape of fear phenomena (Laundré et al., 2001; Hernández and Laundré, 2005; Brook et al., 2012). Prey species such as ungulates tend to move from “risky” zones (e.g., where predator density is higher, or refuge scarcer) to safer spaces (e.g., reduced predator density, more refugia available) to reduce their predation risk (Tolon et al. (2009). Alterations in behavior can result in trade-offs between daily activities such as foraging, movement, and mating, and safety behaviours, especially in high risk habitats. For example, both ungulates and mesopredators adopt “safety” behaviours like vigilance, reduced foraging time, and shifting to safe habitats when under high predation risk (Creel et al., 2005; Li et al., 2011; Zheng et al., 2013; Kuijper et al., 2014). Thus, top predators can indirectly control the behavior and habitat use of mesopredators, as well as prey (Palacios et al., 2016).

Changes in prey and mesopredator behaviours are likely mediated by the presence of predator cues in the environment (Apfelbach et al., 2005; Caro, 2005). According to Creel et al. (2008) presence of cues indicating the availability of predators within the habitat can be sufficient enough for prey to assess the predation risk within the area. Top predator odour cues like feces, urine, and fur, are used by most mammalian prey species to detect the presence of danger during foraging; cues, therefore, help to recognize the extent of risk (Garvey et al., 2016, 2017). The presence of predator odours can trigger prey species to respond by altering their behavior (Apfelbach et al., 2005; Caro, 2005), such as decreasing visitation to specific areas (Nersesian et al., 2012) and increased protection of young (Schulte et al., 2013). In the presence of wolves, elk change behaviours like vigilance levels and movement (Creel et al., 2005, 2008). The accuracy of risk perception based on predator cues may increase once different cues are combined, or used in different contexts; for example lizards and newts use predator cues in a habitat-dependent manner, relying more on visual cues in an open area, and odour cues in a dense vegetative habitat (Mathis and Vincent, 2000). Visual and odour cues have the potential to convey different information (Smith and Belk, 2001), and may also differ in intensity and longevity (Brown and Cowan, 2000; Kats and Dill, 2016), as well as detectability by different species (e.g., some may have a greater ability to detect visual cues than others based on differences in visual acuity (Gonzálvez and Rodríguez-Gironés, 2013).

A number of studies have shown the effects of predator cues on ungulate and/or mesopredator behavior (e.g., Hughes et al., 2010; Cremona et al., 2014; Kuijper et al., 2014; Wikenros et al., 2015; Natt et al., 2017; Suárez-Tangil and Rodríguez, 2017; Wikenros et al., 2017; Goullaud et al., 2018). However, less has been done to identify specific cues used to assess the presence of risk at the community level (Winnie, 2012; Beschta and Ripple, 2013; Kuijper et al., 2014). For example, although prey get information about the proximity of predators via some cue channels, for the most part in the type of visual and odour cues (Kelley and Magurran, 2003), most studies have considered only one cue. The relative importance of cue type for different species (of both predator and prey) is therefore relatively understudied. Further, the study on the effects of predator cues on prey species behavior vastly outweighs that on the effects on mesopredator species, though this is likely to be equally important in structuring communities, and some clear effects have been shown. For example, Palacios et al. (2016) showed that experimental modification of apex predator cues (e.g., predator odour and visual cues) reduced the distance swum, the area used and even foraging behaviours carried out by mesopredators. Mesopredators also respond especially quickly to predator species that have recently consumed conspecifics (Cremona et al., 2014).

The Siberia Tiger Park in Heilongjiang, China has bred a reasonable Amur tiger number and is planning a program to train and reintroduce them in the wild (Wang et al., 2018). Thus, comparison of these predator types is of particular interest in this region, as for how native prey and mesopredator species respond to these novel cues gives us insight into how predator-prey dynamics are likely to unfold in this region going forward. Currently, we are in a period of anthropogenic change; species are shifting ranges, and animals may be encountering novel predators that are moving into their range (Chen et al., 2011; Van Dyck, 2012; Wong and Candolin, 2015). Knowing how species respond to novel cues might help to determine extinction probabilities and overall plasticity in the face of change. To our current understanding, this is the first study that involves captive Amur tiger ecological influence on prey site visitation rate in the wild, which as we have noted is likely to be important in the future as this species regains some of its prevalence in the region.

Here, we test how different predator cues (visual and odour) from familiar and novel predators (brown bear vs. Amur tiger) influence ungulate, mesopredator, and top predator visitation rates to camera trap sites in Hanma National Nature Reserve, China. Previous work has demonstrated reductions in visitation rate to sites where predator cues are prevalent in both ungulates (Kuijper et al., 2014; Wikenros et al., 2015; Venter et al., 2017), and mesopredators (Palacios et al., 2016; Wikenros et al., 2017; Haswell et al., 2018; Sivy et al., 2018). We, therefore, predict that overall, ungulate and mesopredator visitation rates should be reduced at camera trap sites where predator cues are present, versus sites where control cues are present. We also predict that cue type might differentially influence prey and mesopredator visitation rates to camera trap sites. Different prey responds differently towards different predator cues, but also the intensity of predator cues determines the intensity of response (Gonzálvez and Rodríguez-Gironés, 2013). We predict that odour cues might induce a weaker response in terms of visitation rate because they are more affected by wind and are likely to be less long-lived than visual cues (Brown and Cowan, 2000; Kats and Dill, 2016). We also predict differences in prey and mesopredator response to predator cues based on predator type. Correct identification and response to predator cues are important for prey species fitness and often relies on experience (Carthey and Blumstein, 2017; Saxon-Mills et al., 2018). Due to this, prey species may show a very minimal or no response to cues from a novel predator or one that has not existed in a system for many generations (Saxon-Mills et al., 2018). Amur tigers have been absent from this region for many decades with no reported sightings in our study area in recent history (Turvey et al., 2017). We predict that prey and mesopredator response to familiar predator cues (brown bear visual odour cues) will be greater (a more significant drop in visitation rate) than the response to novel predator cues (tiger visual and odour cues).

Materials and Methods

Study site

This study was conducted between July 2016 and June 2017 in the Greater Khingan Mountains of Hanma National Nature reserve (51°20′02″‒51°49′48″N, 122°23′34″-122°52′46″E), close to the small city of Genhe, Inner Mongolia, North East China (Fig. 1). In total, Hanma comprises an area of 1073.48 km2. Habitats, including vegetation and species characteristics, are described elsewhere (Zhai-Penghui, 2015; Guo et al., 2017). Predators and mesopredators present in this system include the Eurasian lynx (Lynx lynx), wolf (Canis lupus), wolverine (Gulo gulo), and brown bear (Ursus arctos). Naturally occurring ungulate species include roe deer (Capreolus pygargus), moose (Alces alces), musk deer (Moschus sifanicus), reindeer (Rangifer tarandus), and red deer (Cervus elaphus) (Zhai-Penghui, 2015).

Experiment design

Design

First, we established treatment plots or grids, with one square plot for each treatment type: the visual cue experiment comprised three plots (Aa- tiger visual, Ab- bear visual, Ac- control), and the odour cue experiment comprised two plots (Ba- tiger, Bb- control; Fig. 1). Plot grids were determined randomly using the fishnet method in ArcGIS 9.3 (Krivoruchko, 2011), which laid grid cells of specified areas within the study area. Here, we set up camera traps, at least 1km from each other. At first, the odour group grids were designed to be the same as the visual cue experiment (i.e., with one plot for the treatment and another plot for the control), but given unavoidable differences in environmental conditions across the designated grid sites (i.e., mountains and swamp areas). We, therefore, redesigned the layout for this experiment so that each grid contained both treatment and control (i.e., Ba and Bb, respectively; Fig. 1) to avoid confounding effects of habitat type. We installed a total of 113 camera traps (LTL ACORN scouting camera; Ltl 5210 5MP and Ltl 5210A 12MP, and Nighthawk Bestguarder SG-990V infrared sensor trigger self-timer digital camera) in all over the Hanma national nature reserve. In our designed study grids there were a total of 86 cameras. We additionally set up 27 no treatment cameras to establish baseline visitation rates with no manipulation present; these were placed randomly within and outside the grids to the areas with high chances to capture mammals, i.e., nearby water points, areas with numerous animal ways and droppings. Cameras were set to a video mode to record for 20s at each 5s interval when triggered by an animal passing across. The treatments and controls used were: photos of Amur tigers Panthera tigris tigris (Kitchener et al., 2017) and brown bear Ursus arctos (Blanco et al., 2011) as visual cues and a blank model (vegetation) as their control. Lastly, we used Amur tiger feces as odour cue and soil as its control (Fig. 2).

Visual cue plots were specified as grids of 3.2 km × 3.2 km; the odour cue experiment plots were specified as grids of 1.6 km × 1.6 km. The latter was smaller due to the topography of this area having lower accessibility. Each visual cue plot had 18 camera traps, and the smaller odour cue plots had 8 camera traps each. In total we established seven plots; two for treatments (i.e., Fig. 1A, C) and one as control (Fig. 1C) for the visual cue experiment. The two other plots for odour experiment were divided each into treatment and its control due to the reasons stated above; making it four as previously stated. We chose to have separate plots for each treatment type as opposed to dispersing predator cue and control sites randomly due to the small size of the reserve which might have resulted in treatments influencing the visitation rates of nearby camera trap sites. The separation of treatment types ensured that this was not the case. We are confident that there was no effect of grid placement on visitation rates, again due to the small size of the reserve, and because all grids were similar in habitat and vegetation density.

Study 1: Visual cue experiment

In tiger visual grid (Fig. 1Aa), tiger visual cues (a life-size photo of a standing tiger printed on a canvas; Figure 2 following studies by Li et al. (2011) and Fischer et al. (2017)) were placed at pre-designated plot points by tying the print to two trees located adjacent to each other. Then, cameras were tied about 50cm height from the ground to one of the trees to record incidence of visitation and ungulate/mesopredator behaviours. The bear visual grid (Fig. 1Ac) was set up in the same way, with brown bear visual cues (life-size photo of a standing bear printed on a canvas). A control grid (Fig. 1Ab), contained canvases printed with a photo of vegetation/habitat-like environment. Each visual cue grid contained 18 camera trap plots (18 x tiger, 18 x bear, 18 x controls).

 

 

Study 2: Odour cue experiment

The Amur tiger feces from Harbin Siberian Tiger Park were collected by hand and refrigerated (maximum of 6 days in storage) before being transferred to 150 ml plastic bottles, pierced with small holes in order to let the tiger feces odour escape. Plastic bottles (150 ml) with the same volume of soil were used as controls. In both treatment and control plots, the plastic bottle containing either soil or feces was tied at a tree bark (Fig. 2D) at the height that is equivalent to the camera trap height (located at opposite tree) so that the camera could record animal incidences of visitation at the station and behaviours expressed. Each odour cue experiment grid contained 8 camera trap sites (i.e., 8 x tiger feces, and 8 x controls at Figure 1Ba; and 8 x tiger feces, and 8 x controls at Figure 1Bb).

Data collection

After one year of recording, we collected the memory cards (SD) from the cameras and brought them back to the laboratory for analysis. Using focal animal sampling and all-occurrence methodology (Altmann, 1974; Lehner, 1992; Martin et al., 1993; Margulis, 2016) we extracted the required data (namely – incidences in which an ungulate or mesopredator triggered the camera to record). Triggered recordings that were within 30 min of each other and contained the same animal were counted as one visit. Ungulate species detected were roe deer, moose, musk deer, red deer, and wild boar. Mesopredators were defined as predators or population of predators that are classified to be in the mid-size of the available predators within the given ecosystem (Groom et al., 2006; Prugh et al., 2009; Wallach et al., 2015); under this definition, we detected sable Martes zibellina and weasels Mustela nivalis in this system. Top predators (apex predators, alpha predators or mega-predators) were defined as large predators with no natural predators to feed on them within the food chain (Groom et al., 2006; Prugh et al., 2009; Sukhdeo, 2012; Wallach et al., 2015); we detected bear, lynx, and wolverine as top predators in this system.

For the visual cue experiment, data were collected from a total of 83 deployed cameras. Some cameras stopped functioning soon after deployment (within 30 days) and were not included in data analysis (n = 3 tiger photo cameras, n = 2 bear photo cameras, n = 6 control cameras) resulting in a sample of 69 cameras. In this sample, the mean length of camera activity was 280.0 days ± 11.1, and the total number of mammals captured on camera was 997.

For the odour cue experiment, data were collected from a total of 61 deployed cameras. Again, we excluded cameras that were active for fewer than 30 days (n = 1 control cameras, no predator odour cameras excluded), resulting in a sample of 57 cameras. In this sample, the mean length of camera activity was 251.6 days ± 14.5, and the total number of mammals captured on camera was 837.

We also collected data from 27 no treatment cameras, again excluding cameras that were active for fewer than 30 days (n = 3 cameras), resulting in a total of 24 cameras.

Statistical analysis

For analysis of both experiments, the number of a) ungulates, b) mesopredators, and c) top predators was set as the dependent variable in three separate generalized linear models (GLMs), with treatment, and number of days of camera activity as independent variables. For the photo cue experiment, treatment was a four-level factor: no treatment, tiger cue, bear cue, and control cue. For the odour cue experiment, treatment was a three-level factor: no treatment, tiger odour cue, control cue. Given the skew towards 0s and low numbers in mammal incidence data in both experiments, we used a negative binomial error structure to account for overdispersion, with a log link. All models were constructed using the glm.nb function (MASS package; Venables and Ripley, 2002) in the statistical software, R version 3.2.3 (R Core Team, 2015). We compared models (the full model, single-term models, a null model) using AIC, and report the best model (lowest AIC value). Significance values in models were determined using likelihood ratio tests. Where treatment was significant, we used post-hoc tests (Tukey tests) to determine whether some treatment types had stronger effects than others, as predicted.

For analysis of the photo experiment, we initially constructed models containing either all four treatments, or the two predator cue treatments grouped (i.e., a predator cue treatment level), and tested which better fit the data using AIC comparison (Akaike, 2011). If the model containing the four-level treatment variable (including both predator cue types) was within 2AICc of the model containing the three-level treatment variables (predator cue types grouped), we maintained the separate cue types, as we were interested in potential differences in responses to different predators.

In models analysing the odour cue experiment, we additionally included cue ID as an independent variable to account for potentially varying effects of different individual cues, as two different tiger odour cues and two different control cues were used. In all three models, cue ID was not a significant predictor variable, so we do not report results for this here.

 

Results

Photo cue experiment

Treatment significantly predicted ungulate visitation to camera trap sites (X23 = 18.53, P <0.001; Table I; Fig. 3). Post-hoc tests revealed that visitation to sites with the tiger cue and the bear cue was significantly lower than visitation to sites where no cue was present (tiger photo Z = -4.51, P < 0.0001; bear photo Z = -2.97, P = 0.015, respectively). Visitation to no cue and control cue sites did not differ, as expected (Z = -1.65, P = 0.35). Interestingly, visitation responses to tiger and bear cues were not statistically different (Z = -1.46, P = 0.46). The length of camera activity (days) was positively correlated with ungulate visitation (X21 = 4.38, P <0.05).

 

Table I.- Results from best model (GLM) of ungulate visitation to camera traps.

		Estimate	S.E.	Z	P
Intercept		1.82	0.44	4.10
Treatment					0.0003
	No cue	0.00
	Tiger photo cue	-1.61	0.36	-4.51
	Bear photo cue	-1.01	0.34	-2.97
	Control cue	-0.66	0.40	-1.65
Days of camera activity		0.003	0.001	2.18	0.004

 

Table II.- Results from best model (GLM) of all predator (mesopredator + top predator) visitation to camera traps.

		Estimate	S.E.	Z	P
Intercept		0.68	0.31	2.22
Treatment					0.06
	No cue	0.00
	Tiger photo cue	-0.83	0.59	-1.41
	Bear photo cue	0.88	0.52	-1.70
	Control cue	-0.39	0.66	-0.59

 

 

Treatment did not significantly predict mesopredator visitation (X21 = 1.14, P = 0.77); the null model best described this data. Treatment also did not significantly predict top predator visitation (X21 = 2.81, P = 0.24); the null model best described this data. Combining mesopredators and top predators showed a small effect of treatment on visitation rates (X23 = 7.43, P = 0.06; Fig. 4): post-hoc testing showed that the only significant difference between cue types was that visitation to bear cue sites was slightly higher than visitation to tiger cue sites (Z = -2.62, P < 0.05).

Odour cue experiment

Odour cue treatment did not significantly predict ungulate visitation (X21 = 0.71, P = 0.70); mesopredator visitation (X21 = 1.14, P = 0.77); or, top predator visitation (X21 = 2.81, P = 0.24) to camera trap sites. The null model best described these data in all three cases. The null model also best described visitation rates of all predators combined; treatment was not significant (X22 = 2.12, P = 0.35).

Discussion

In this study, we tested how visual and odour cues from familiar and novel predators (brown bear vs. Amur tiger) influenced ungulate, mesopredator, and top predator visitation rates to camera trap sites in Hanma National Nature Reserve. We show that visual cues predicted ungulate visitation rates and ungulates responded equally to the novel and familiar predators. Odour cue, on the other hand, did not influence ungulate visitation rates. Mesopredator visitation was not influenced by either cue type, contrary to predictions, though mesopredators and top predators combined showed a weakly stronger negative response to tiger versus bear cues, suggesting a small novel predator effect on visitation rates. Predator avoidance has vital biological and transformative outcomes on communities and trophic cascades (Peacor and Werner, 2001; Gonzálvez and Rodríguez-Gironés, 2013); nevertheless, the generalisation and contextual importance of predator cues have also been little studied. This study is, therefore, an essential and exciting step forward in understanding predator cue responses at the community level, which may have implications for conservation in this region. There has been a study by Wang et al. (2018) on the captive Amur tiger responses towards its natural prey cues; here we test the response of prey towards the Amur tiger, providing insight into its ecological functioning in the wild environment to which it may soon be reintroduced.

Photo cue experiment

There has been work by various authors on studying the influence of predation risk to ungulates, but less has been done to identify the cue used by ungulates to assess presence of risk (Winnie, 2012; Beschta and Ripple, 2013; Kuijper et al., 2014), and equally little to understand whether mesopredators use the same cues within a community. Our results confirm that predation risk as assessed using predator visual cues (both Amur tiger and brown bear photos) significantly influences ungulate visitation rate at predator cues sites. There was a difference in visitation rate of ungulates (roe deer, moose and musk deer) to the camera sites with tiger and bear visual cues compared to camera sites at the control sites as well as at no treatment sites: more ungulates visited no treatment and control sites as opposed to sites where there were predator visual cues. This result is in line with a study on tammar wallabies (Macropus eugenii), showing that they respond to visual predator cues; tammar wallabies increased vigilance behaviours and reduced feeding rate after confronting fox visual cues (Blumstein et al., 2000). Our results are contrary to a study by Venter et al. (2017), which reported that large grazing herbivores (such as zebra, red hartebeest, and eland) do not appear to rely on visual cues while on foraging activities. How ungulate species respond to visual predator cues may have a size-dependent component. For example, small prey species such as impala, warthog, waterbuck, and kudu minimize their chances of encountering predators by evading the use of same space with all predators, while larger ungulate species like wildebeest, giraffe, and zebra only avoided sharing space with lion and leopards (Thaker et al., 2011). Our ungulate group contained moose, roe deer, and musk deer, with roe deer a majority. Moose are relatively large and comparable to zebra, red hartebeest, and eland, whereas roe deer and musk deer are much smaller. This may explain the difference in our results from those of Venter et al. (2017), indicating that visual cues inhibit visitation rate to the predator visual cue camera sites. Unfortunately, our small sample size of moose mean that we cannot further interpret our results based on ungulate size classes, but this would be a fruitful area of further study in this system.

We show that responses of ungulates to Amur tiger and brown bear cues in terms of site visitation are not statistically different, though tiger cues are relatively novel given their long absence in this region, contrary to our prediction that ungulates should show a stronger response to the cue with which they were more familiar (Wiles et al., 2003; Carthey and Blumstein, 2017; Saxon-Mills et al., 2018). This indicates a degree of generalization of predator visual cues that do not appear to rely strongly on previous experience. This is similar to Dunlop-Hayden and Rehage (2011) results which indicated no differences in prey reaction towards native vs. non-native predators. Alternatively, ungulates may have been responding to novelty per se, which in itself can elicit fear responses (Sneddon et al., 2003). An appropriate way to test this further might be to repeat the experiment using an additional novel cue such as a photograph of a train or other novel object, to determine whether the equal response to tiger cues as bear cues is because tiger cues are inherently novel, or because they are recognised as a predatory threat. It is worth to note that there are no any literature which show existence of tiger in the study area, even local people suggest that tiger has never lived in this ecosystem.

Mesopredator release theory suggests that an increase in apex predators also suppresses mesopredators in the area (Haswell et al., 2018). Mesopredators are likely to also use predator cues, for example, Switalski (2003) suggested that when wolves are within the area of study, coyotes use a visual cue to detect their predator (wolves) availability. Additionally, Palacios et al. (2016) proposed that that presence of any kind of apex predator cue (coral whether visual, chemical or even combined (coral trout Plectropomus leopardus) limits the distance its mesopredator prey (dottyback Pseudochromis fuscus) would swim or engage in other activities like foraging. Interestingly our study contradicts the above studies because both Amur tiger and brown bear cues showed no significant influence on visitation rates of mesopredators alone (though in combination with top predators, mesopredators showed slightly lower visitation to tiger cue sites). This may be because brown bear are few in our study area compared to the incidence of top predators in the coral trout and wolf systems, hence lessening the chance that bears interact directly with mesopredators, implying that mesopredators might not use bear visual cues (but may still respond to a novel cue, such as the tiger).

Odour cue experiment

Wikenros et al. (2015) studied the response of red deer and roe deer to Eurasian lynx olfactory cues, concluding that both red deer and roe deer decreased their visitation duration at the treatment sites. Also, Noell (2013) states that ungulates have a strong ability to detect the smell, but their response to predator cues vary (Apfelbach et al., 2005). We predicted that odour cue would predict ungulate visitation rates significantly in the wild, but this was not the case in our study. Nevertheless, our data support other previous findings that indicate that predator cues do not affect deer visitation rate to the chemical cue environment (Kimball et al., 2009; Elmeros et al., 2011). We believe that wetness and wind have a role in determining the effectiveness and longevity of odour cues. The odour cue experimental sites were located in the mountain that faces a large swamp area that contains water in summer and autumn. This might have influenced our result too due to the wet wind that blows towards the side of the camera trap experiment site. The current study acts as a foundational base for another ecologist to carry out further experiments on the ecological functioning role of captive Amur tiger odour in the wild.

Our result is in line with numerous other studies which reported that there is a weaker response of mammalian mesopredators (Garvey et al., 2016) towards the risk posed by their top predator. Tiger odour cue did not predict mesopredator visitation to the camera trap sites. Sih et al. (2010) suggested that prey species may need to have preceding experience toward the predator chemical cue before an experiment can be done. We agree with this idea; and we think that mesopredators of our study area might have evolved a lack of fear of tigers due to unfamiliarity (Suraci et al., 2017). This is because as discussed in the visual cue experiment that there is no clear record evidence that tiger ever existed in the area recently; but some archives indicate that there are signs for the tiger to have existed in the area in the distant past (Turvey et al., 2017). The same reasons (wetness and windward side) discussed in the above paragraph for ungulates applies here too. It would be informative to repeat the experiment using the familiar predator odour cue (brown bear), as the lack of response by both ungulates and mesopredators toward the tiger odour cue may have been due to lack of familiarity, rather than a lack of response generally to chemical cues. Unfortunately, we did not have access to suitable brown bear cues for this study.

Visual vs odour cues influence on visitation rate

Normally visual cue indicates the predator is present and prey/mesopredator will be in immediate danger if they come face to face with the cue, this means they have to minimize unnecessary movements or opting to flee from the area. However, feces may mean that there is a tiger in the area but investigating the cue could provide information on the sex of the donor, what it has eaten, etc. Tigers are territorial and the feces were fresh, it could mean that the donor is not likely to be back to this spot for a while and therefore the area is safe in the immediate future. Hence less fear, similar to our results. Apfelbach et al. (2005) indicate that body odour is a better test for antipredator responses than feces or urine, and studies of antipredator responses are moving towards using body odour if possible rather than feces as it is a less confusing cue of immediate predator presence. 

Conclusion

To our understanding, this is the first study to test the ecological function of Amur tiger cues in China and possibly worldwide in the wild environment. We demonstrate that captive Amur tiger visual and odour cues do not predict mesopredators and top predator visitation rates significantly, but both captive Amur tiger and brown bear visual cues predict the visitation rate of ungulates significantly to predator treatment sites. Our results have laid down foundations on the understanding of the ecological functioning of captive Amur tiger visual and odour cues. Therefore our study together with that of Wang et al. (2018) may save as key in the rewilding process of captive Amur tiger plans.

Acknowledgments

This work was supported by funds from the Fundamental Research Funds for the Central Universities (2572017PZ14), the fund from the National Key Programme of Research and Development, Ministry of Science and Technology (2016YFC0503200), and National Nature Science Foundation of China (NSFC 31872241; 31572285). We thank Bao Heng and Zhai Penghu for their support during fieldwork. Thanks to the staff at Hanma National Nature Reserve for their field work support, the management of Heilongjiang Amur Tiger Park who provided us with the fresh Amur tiger feces. This article is part of the principal author Ph. D. work.

Statement of conflict of interest

The authors declare that they have no conflict of interests regarding the publication of this article.

References

Altmann, J., 1974. Observational study of behavior: Sampling methods. Behaviour, 49: 227-266. https://doi.org/10.1163/156853974X00534

Apfelbach, R., Blanchard, C.D., Blanchard, R.J., Hayes, R.A. and McGregor, I.S., 2005. The effects of predator odors in mammalian prey species: A review of field and laboratory studies. Neurosci. Biobehav. Rev., 29: 1123-1144. https://doi.org/10.1016/j.neubiorev.2005.05.005

Beschta, R.L. and Ripple, W.J., 2013. Are wolves saving yellowstone’s aspen? A landscape-level test of a behaviorally mediated trophic cascade: Comment. Ecology, 94: 1420-1425. https://doi.org/10.1890/11-0063.1

Blanco, J.C., Ballesteros, F., García-Serrano, A., Herrero, J., Nores, C. and Palomero, G., 2011. Behaviour of brown bears killing wild ungulates in the cantabrian mountains, southwestern europe. Eur. J. Wildl. Res., 57: 669-673. https://doi.org/10.1007/s10344-010-0464-z

Blumstein, D.T., Daniel, J.C., Griffin, A.S. and Evans, C.S., 2000. Insular tammar wallabies (Macropus eugenii) respond to visual but not acoustic cues from predators. Behav. Ecol., 11: 528-535. https://doi.org/10.1093/beheco/11.5.528

Brook, L.A., Johnson, C.N. and Ritchie, E.G., 2012. Effects of predator control on behaviour of an apex predator and indirect consequences for mesopredator suppression. J. appl. Ecol., 49: 1278-1286. https://doi.org/10.1111/j.1365-2664.2012.02207.x

Brown, G.E. and Cowan, J., 2000. Foraging trade-offs and predator inspection in an ostariophysan fish: Switching from chemical to visual cues. Behaviour, 137: 181-195. https://doi.org/10.1163/156853900502015

Caro, T., 2005. Antipredator defenses in birds and mammals. University of Chicago Press, Chicago, Illinois, United States.

Carthey, A.J.R. and Blumstein, D.T., 2017. Predicting predator recognition in a changing world. Trends Ecol. Evol., 33: 106-115. https://doi.org/10.1016/j.tree.2017.10.009

Chen, I.C., Hill, J.K., Ohlemüller, R., Roy, D.B. and Thomas, C.D., 2011. Rapid range shifts of species associated with high levels of climate warming. Science, 333: 1024-1026. https://doi.org/10.1126/science.1206432

Creel, S., Winnie, J., Maxwell, B., Hamlin, K. and Creel, M., 2005. Elk alter habitat selection as an antipredator response to wolves. Ecology, 86: 3387-3397. https://doi.org/10.1890/05-0032

Creel, S., Winnie, J.A., Christianson, D. and Liley, S., 2008. Time and space in general models of antipredator response: Tests with wolves and elk. Anim. Behav., 76: 1139-1146. https://doi.org/10.1016/j.anbehav.2008.07.006

Cremona, T., Crowther, M.S. and Webb, J.K., 2014. Variation of prey responses to cues from a mesopredator and an apex predator. Austral. Ecol., 39: 749-754. https://doi.org/10.1111/aec.12138

Dunlop-Hayden, K. and Rehage, J.S., 2011. Antipredator behavior and cue recognition by multiple everglades prey to a novel cichlid predator. Behaviour, 148: 795-823. https://doi.org/10.1163/000579511X577256

Elmeros, M., Winbladh, J.K., Andersen, P.N., Madsen, A.B. and Christensen, J.T., 2011. Effectiveness of odour repellents on red deer (Cervus elaphus) and roe deer (Capreolus capreolus): A field test. Eur. J. Wildl. Res., 57: 1223-1226. https://doi.org/10.1007/s10344-011-0517-y

Farnworth, B., Innes, J. and Waas, J.R., 2016. Converting predation cues into conservation tools: The effect of light on mouse foraging behaviour. PLoS One, 11: e0145432. https://doi.org/10.1371/journal.pone.0145432

Fischer, S., Oberhummer, E., Cunha-Saraiva, F., Gerber, N. and Taborsky, B., 2017. Smell or vision? The use of different sensory modalities in predator discrimination. Behav. Ecol. Sociobiol., 71: 143. https://doi.org/10.1007/s00265-017-2371-8

Garvey, P.M., Glen, A.S., Clout, M.N., Wyse, S.V., Nichols, M. and Pech, R.P., 2017. Exploiting interspecific olfactory communication to monitor predators. Ecol. Appl., 27: 389-402. https://doi.org/10.1002/eap.1483

Garvey, P.M., Glen, A.S. and Pech, R.P., 2016. Dominant predator odour triggers caution and eavesdropping behaviour in a mammalian mesopredator. Behav. Ecol. Sociobiol., 70: 481-492. https://doi.org/10.1007/s00265-016-2063-9

Gonzálvez, F.G. and Rodríguez-Gironés, M.A., 2013. Seeing is believing: Information content and behavioural response to visual and chemical cues. Proc. R. Soc. London B: Biol. Sci., 280: 886. https://doi.org/10.1098/rspb.2013.0886

Goullaud, E.L., de Zwaan, D.R. and Martin, K., 2018. Predation risk-induced adjustments in provisioning behavior for horned lark (Eremophila alpestris) in british columbia. Wilson J. Ornithol., 130: 180-190. https://doi.org/10.1676/16-150.1

Groom, M.J., Meffe, G.K. and Carroll, C.R., 2006. Principles of conservation biology. Sinauer Associates, Sunderland.

Guo, K., Liu, H., Bao, H., Hu, J., Wang, S., Zhang, W., Zhao, Y. and Jiang, G., 2017. Habitat selection and their interspecific interactions for mammal assemblage in the Greater Khingan Mountains, northeastern China. Wildl. Biol., 2017: 00261. https://doi.org/10.2981/wlb.00261

Haswell, P.M., Jones, K.A., Kusak, J. and Hayward, M.W., 2018. Fear, foraging and olfaction: How mesopredators avoid costly interactions with apex predators. Oecologia, 187: 573–583. https://doi.org/10.1007/s00442-018-4133-3

Hernández, L. and Laundré, J.W., 2005. Foraging in the landscape of fear and its implications for habitat use and diet quality of elk cervus elaphus and bison bison bison. Wildl. Biol., 11: 215-220. https://doi.org/10.2981/0909-6396(2005)11[215:FITLOF]2.0.CO;2

Hughes, N.K., Price, C.J. and Banks, P.B., 2010. Predators are attracted to the olfactory signals of prey. PLoS One, 5: e13114. https://doi.org/10.1371/journal.pone.0013114

Kats, L.B. and Dill, L.M., 2016. The scent of death: Chemosensory assessment of predation risk by prey animals. Ecoscience, 5: 361-394. https://doi.org/10.1080/11956860.1998.11682468

Kelley, J.L. and Magurran, A.E., 2003. Learned predator recognition and antipredator responses in fishes. Fish Fish., 4: 216-226. https://doi.org/10.1046/j.1467-2979.2003.00126.x

Kimball, B.A., Taylor, J., Perry, K.R. and Capelli, C., 2009. Deer responses to repellent stimuli. J. chem. Ecol., 35: 1461-1470. https://doi.org/10.1007/s10886-009-9721-6

Kitchener, A.C., Breitenmoser-Würsten, C., Eizirik, E., Gentry, A., Werdelin, L., Wilting, A., Yamaguchi, N. and Johnson, W.E., 2017. A revised taxonomy of the felidae: The final report of the cat classification task force of the iucn cat specialist group. Cat News, NZP Staff Publications.

Krivoruchko, K., 2011. Spatial statistical data analysis for gis users. Esri Press, Redlands.

Kuijper, D.P., Verwijmeren, M., Churski, M., Zbyryt, A., Schmidt, K., Jędrzejewska, B. and Smit, C., 2014. What cues do ungulates use to assess predation risk in dense temperate forests? PLoS One, 9: e84607. https://doi.org/10.1371/journal.pone.0084607

Laundré, J.W., Hernández, L. and Altendorf, K.B., 2001. Wolves, elk, and bison: Reestablishing the landscape of fear in Yellowstone National Park, USA. Can. J. Zool., 79: 1401-1409. https://doi.org/10.1139/z01-094

Lehner, P.N., 1992. Sampling methods in behavior research. Poult. Sci., 71: 643-649. https://doi.org/10.3382/ps.0710643

Li, C., Yang, X., Ding, Y., Zhang, L., Fang, H., Tang, S. and Jiang, Z., 2011. Do pere david’s deer lose memories of their ancestral predators? PLoS One, 6: e23623. https://doi.org/10.1371/journal.pone.0023623

Lima, S.L. and Bednekoff, P.A., 1999. Temporal variation in danger drives antipredator behavior: The predation risk allocation hypothesis. Am. Natural., 153: 649-659. https://doi.org/10.1086/303202

Margulis, S., 2016. Sampling animal behavior. Animal Behavior Society Workshop, July 30, 2016.

Martin, P., Bateson, P.P.G. and Bateson, P., 1993. Measuring behaviour: An introductory guide. Cambridge University Press. https://doi.org/10.1017/CBO9781139168342

Mathis, A. and Vincent, F., 2000. Differential use of visual and chemical cues in predator recognition and threat-sensitive predator-avoidance responses by larval newts (Notophthalmus viridescens). Can. J. Zool., 78: 1646-1652. https://doi.org/10.1139/cjz-78-9-1646

Natt, M., Lönnstedt, O.M. and McCormick, M.I., 2017. Coral reef fish predator maintains olfactory acuity in degraded coral habitats. PLoS One, 12: e0179300. https://doi.org/10.1371/journal.pone.0179300

Nersesian, C.L., Banks, P.B. and McArthur, C., 2012. Behavioural responses to indirect and direct predator cues by a mammalian herbivore, the common brushtail possum. Behav. Ecol. Sociobiol., 66: 47-55. https://doi.org/10.1007/s00265-011-1250-y

Noell, S., 2013. Effects of brown bear (Ursus arctos) odour on the patch choice and behaviour of different ungulate species. Dept. of Wildlife, Fish and Environmental Studies, SLU, Umeå. Available at: https://stud.epsilon.slu.se/6003/ (Accessed on August 31, 2019).

Palacios, M., Warren, D.T. and McCormick, M.I., 2016. Sensory cues of a top-predator indirectly control a reef fish mesopredator. Oikos, 125: 201-209. https://doi.org/10.1111/oik.02116

Peacor, S.D. and Werner, E.E., 2001. The contribution of trait-mediated indirect effects to the net effects of a predator. Proc. natl. Acad. Sci., 98: 3904-3908. https://doi.org/10.1073/pnas.071061998

Prugh, L.R., Stoner, C.J., Epps, C.W., Bean, W.T., Ripple, W.J., Laliberte, A.S. and Brashares, J.S., 2009. The rise of the mesopredator. Bioscience, 59: 779-791. https://doi.org/10.1525/bio.2009.59.9.9

Saxon-Mills, E.C., Moseby, K., Blumstein, D.T. and Letnic, M., 2018. Prey naïveté and the anti-predator responses of a vulnerable marsupial prey to known and novel predators. Behav. Ecol. Sociobiol., 72: 151. https://doi.org/10.1007/s00265-018-2568-5

Schulte, L.M., Schulte, R. and Lötters, S., 2013. Avoiding predation: The importance of chemical and visual cues in poison frog reproductive behaviour. In: Chemical signals in vertebrates. Springer, pp. 309-321. https://doi.org/10.1007/978-1-4614-5927-9_25

Sih, A., Bolnick, D.I., Luttbeg, B., Orrock, J.L., Peacor, S.D., Pintor, L.M., Preisser, E., Rehage, J.S. and Vonesh, J.R., 2010. Predator–prey naïveté, antipredator behavior, and the ecology of predator invasions. Oikos, 119: 610-621. https://doi.org/10.1111/j.1600-0706.2009.18039.x

Sivy, K.J., Pozzanghera, C.B., Colson, K.E., Mumma, M.A. and Prugh, L.R., 2018. Apex predators and the facilitation of resource partitioning among mesopredators. Oikos, 127: 607-621. https://doi.org/10.1111/oik.04647

Smith, M.E. and Belk, M.C., 2001. Risk assessment in western mosquitofish (Gambusia affinis): Do multiple cues have additive effects? Behav. Ecol. Sociobiol., 51: 101-107. https://doi.org/10.1007/s002650100415

Sneddon, L.U., Braithwaite, V.A. and Gentle, M.J., 2003. Novel object test: Examining nociception and fear in the rainbow trout. J. Pain, 4: 431-440. https://doi.org/10.1067/S1526-5900(03)00717-X

Suárez-Tangil, B.D. and Rodríguez, A., 2017. Detection of iberian terrestrial mammals employing olfactory, visual and auditory attractants. Eur. J. Wildl. Res., 63: 93. https://doi.org/10.1007/s10344-017-1150-1

Sukhdeo, M.V., 2012. Where are the parasites in food webs? Parasites Vectors, 5: 239. https://doi.org/10.1186/1756-3305-5-239

Suraci, J.P., Roberts, D.J., Clinchy, M. and Zanette, L.Y., 2017. Fearlessness towards extirpated large carnivores may exacerbate the impacts of naïve mesocarnivores. Behav. Ecol., 28: 439-447. https://doi.org/10.1093/beheco/arw178

Switalski, T.A., 2003. Coyote foraging ecology and vigilance in response to gray wolf reintroduction in yellowstone national park. Can. J. Zool., 81: 985-993. https://doi.org/10.1139/z03-080

Thaker, M., Vanak, A.T., Owen, C.R., Ogden, M.B., Niemann, S.M. and Slotow, R., 2011. Minimizing predation risk in a landscape of multiple predators: Effects on the spatial distribution of african ungulates. Ecology, 92: 398-407. https://doi.org/10.1890/10-0126.1

Tolon, V., Dray, S., Loison, A., Zeileis, A., Fischer, C. and Baubet, E., 2009. Responding to spatial and temporal variations in predation risk: Space use of a game species in a changing landscape of fear. Can. J. Zool., 87: 1129-1137. https://doi.org/10.1139/Z09-101

Turvey, S.T., Crees, J.J., Li, Z., Bielby, J. and Yuan, J., 2017. Long-term archives reveal shifting extinction selectivity in china’s postglacial mammal fauna. Proc. R. Soc. B, 284: 20171979. https://doi.org/10.1098/rspb.2017.1979

Van Dyck, H., 2012. Changing organisms in rapidly changing anthropogenic landscapes: The significance of the umwelt-concept and functional habitat for animal conservation. Evolut. Applic., 5: 144-153. https://doi.org/10.1111/j.1752-4571.2011.00230.x

Venter, J.A., Prins, H.H., Mashanova, A. and Slotow, R., 2017. Ungulates rely less on visual cues, but more on adapting movement behaviour, when searching for forage. PeerJ, 5: e3178. https://doi.org/10.7717/peerj.3178

Wallach, A.D., Izhaki, I., Toms, J.D., Ripple, W.J. and Shanas, U., 2015. What is an apex predator? Oikos, 124: 1453-1461. https://doi.org/10.1111/oik.01977

Wang, Q., Liu, D., Holyoak, M., Jia, T., Yang, S., Liu, X., Kong, X. and Jiang, G., 2018. Innate preference for native prey and personality implications in captive amur tigers. Appl. Anim. Behav. Sci., 210: 95-102. https://doi.org/10.1016/j.applanim.2018.10.006

Wikenros, C., Jarnemo, A., Frisén, M., Kuijper, D.P. and Schmidt, K., 2017. Mesopredator behavioral response to olfactory signals of an apex predator. J. Ethol., 35: 161-168. https://doi.org/10.1007/s10164-016-0504-6

Wikenros, C., Kuijper, D.P., Behnke, R. and Schmidt, K., 2015. Behavioural responses of ungulates to indirect cues of an ambush predator. Behaviour, 152: 1019-1040. https://doi.org/10.1163/1568539X-00003266

Wiles, G.J., Bart, J., Beck, Jr. R.E. and Aguon, C.F., 2003. Impacts of the brown tree snake: Patterns of decline and species persistence in guam’s avifauna. Conserv. Biol., 17: 1350-1360. https://doi.org/10.1046/j.1523-1739.2003.01526.x

Winnie, J.A., 2012. Predation risk, elk, and aspen: Tests of a behaviorally mediated trophic cascade in the greater yellowstone ecosystem. Ecology, 93: 2600-2614. https://doi.org/10.1890/11-1990.1

Wong, B. and Candolin, U., 2015. Behavioral responses to changing environments. Behav. Ecol., 26: 665-673. https://doi.org/10.1093/beheco/aru183

Zhai-Penghui, L.Y., Zhengshan, L., 2015. The current situation and evaluation of the biodiversity of hanma national nature reserve. China Acad. J. Electron. House, (Original article in Chinese published 2015).

Zheng, W., Beauchamp, G., Jiang, X., Li, Z. and Yang, Q., 2013. Determinants of vigilance in a reintroduced population of pere david’s deer. Curr. Zool., 59: 265-270. https://doi.org/10.1093/czoolo/59.2.265