How Many Relocations Needed To Do Animal Habitat Selection

• Loading metrics

Open up Access

Peer-reviewed

Research Article

Habitat Option by Eld's Deer following Relocation to a Patchy Landscape

• Duo Pan,
• Yan-Ling Song,
• Zhi-Gao Zeng,
• Benjamin D. Bravery

x

• Published: March 10, 2014
• https://doi.org/ten.1371/periodical.pone.0091158

Figures

Abstruse

An emerging issue in wildlife conservation is the re-establishment of viable populations of endangered species in suitable habitats. Here, we studied habitat pick by a population of Hainan Eld'south deer (Cervus eldi) relocated to a patchy landscape of farmland and woods. Hainan Eld'south deer were pushed to the brink of extinction in the 1970s, but their population expanded rapidly from 26 to more than k individuals past 2003 through effective reserve protection. As part of a wider relocation and population management strategy, 131 deer were removed from the reserve and reintroduced into a farmland-woods landscape in 2005. Habitat use nether a context of human being disturbance was surveyed by monitoring 19 radio-collared animals. The majority of deer locations (77%) were within 0.6–ii km of villages. Annual home ranges of these collared deer averaged 725 ha (SD 436), which was 55% of the size of the reserve from which they had originated. The annual home ranges contained 54% shrub-grassland, 26% forest and fifteen% farmland. The relocated deer population selected mural comprising slash-and-burn agronomics and forest, and avoided both intensively farmed areas and areas containing merely forest. Within the selected landscape, deer preferred swiddens and shrub-grasslands. Forests higher up 300 m in elevation were avoided, whereas forests below 300 m in elevation were overrepresented during the dry flavour and randomly used during the wet flavour. Our findings show that reintroduced deer can utilise disturbed habitats, and further demonstrate that subsistence agroforest ecosystems have the capacity to sustain endangered ungulates.

Citation: Pan D, Song Y-L, Zeng Z-K, Bravery BD (2014) Habitat Choice by Eld's Deer following Relocation to a Patchy Mural. PLoS ONE ix(3): e91158. https://doi.org/10.1371/journal.pone.0091158

Editor: K. Ullas Karanth, Wild animals Conservation Social club, Bharat

Received: August 4, 2013; Accepted: February 11, 2014; Published: March 10, 2014

Copyright: © 2014 Pan et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted employ, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: This research was supported by the National Natural Science Foundation of Red china (Grants 30770320 and 30470260) http://www.nsfc.gov.cn/Portal0/default152.htm. The funders had no office in study design, data collection and analysis, decision to publish, or grooming of the manuscript.

Competing interests: Benjamin D. Bravery is employed by Kexue Communications. There are no patents, products in development or marketed products to declare. This does not modify the authors' adherence to all the PLOS ONE policies on sharing data and materials, equally detailed online in the guide for authors.

Introduction

Human activeness is responsible for a refuse of wildlife populations and the deterioration of their habitats across the earth [1]. Protected areas established to safeguard wildlife constitute just a small fraction of the Globe's surface [2], and the spatial limitations placed on protected areas will likely result in management issues in the futurity. For example, protected areas may be forced to back up more animals than is platonic, specially for ungulates that respond well to conservation measures [three]. Farther, protected areas may non be constructive in restoring species in an ecological sense because limited spatial resources tin can support but a single or a few populations. Under certain circumstances, maintaining a species solely in a protected surface area may not reduce its adventure of extinction, whereas reintroducing populations outside a protected area can. Notwithstanding, maintaining populations outside reserves and next to homo populations poses a unique set of conservation challenges, particularly in relation to the disturbance of wildlife by humans.

Wild animals perceive repeated human disturbance as analogous to predation chance [four], [5]. Human activities increment the centre rates of wild animals, trigger the release of stress hormones and stimulate vigilance behavior [vi]–[8]. Man activities can too interrupt normal drinking, foraging and reproductive behavior [5], [9]–[11]. At the individual level, animals respond to repeated human disturbance past allocating extra time to restore their pre-disturbance state; at the population level, the stress of interactions with humans may decrease the reproductive success and survival rates of new-born individuals [12]. Given the impact of human being disturbance on the beliefs and ecology of wild animals, information technology is non surprising that many species shift toward areas containing few or no humans [xiii], [fourteen]. Berger's 'risk-disturbance hypothesis' predicts that a disturbed animal should follow the same economic principles that govern encounters of prey with predators [4], [15], [16]. Abstention is undoubtedly the most mutual anti-predator strategy [17]–[19]. Withal, some ungulates are known to adapt to or even benefit from human activity and seek out human-disturbed areas. For case, endangered primal deer (Odocoileus virginianus clavium) use urban areas more today than thirty years agone, and the survival rates of deer utilizing urban areas ofttimes are higher than for deer that utilize urban areas less frequently [twenty]. Roe deer (Capreolus capreolus) employ homo-disturbed areas as a refuge from lynx (Lynx lynx) predation [21]. White-tailed deer (Odocoileus virginianus) in North America exploit crops as a food resource, and as such unwanted deer-human interactions arise [22], [23].

Endangered ungulates are candidates for man-dominated ecosystems considering of their broad diets and practiced response to conservation measures. All the same, they may as well damage crops and are often considered a nuisance. Eld's deer, a medium-sized tropical deer, is an herbivorous generalist patchily distributed across eastern India, Myanmar, Vietnam, Laos, Kingdom of cambodia and Hainan Island of China. Eld'due south deer is endangered in India, Vietnam, Lao people's democratic republic, Cambodia and Communist china, and is threatened in Myanmar [24]. On Hainan Island, People's republic of china, written records regarding Eld'south deer can exist traced dorsum to the yr 1511. Our lab has previously analyzed variation in mitochondrial DNA D-loop sequences of C. e. siamensis, C. e. thamin and C. eastward. Hainanus, and the results suggest that Eld'due south deer may have moved via a land bridge from the S-eastern Asian mainland to Hainan Island during the Pleistocene (0.69 Mya ago) [25]. Eld's deer was once widely distributed on Hainan Island. Considering their natural predators were in decline, man hunting drove the collapse of this island population of Eld'south deer and past 1976 just 26 individuals remained, in the region of Datian [26]. Consequently, the Datian Nature Reserve (DNR) was created to specifically protect Eld's deer in situ. To forestall human disturbances, a debate two.8 k high was synthetic around the perimeter of the 1314-ha protected expanse. Within the refuge, humans are unable to disturb the deer or deer habitat, except for daily patrols by rangers along regular routes. Eld's deer here drink water from a natural river and feed on natural vegetation. Wild Indian muntjac (Muntiacus muntjak nigripes) and boar (Sus scrofa chriodontus) are present every bit natural competitors, and Python molurus as a natural predator [27]. A self-sustaining population of Eld's deer has thrived in the reserve, recovering to more than 1000 individuals by 2003 [26]. A decision was then made to transfer animals to other parts of the isle. According to historical records, Hainan Eld's deer inhabited flat or gentle slopes below 200 m in elevation covered past woods and grass-shrubs [28], but such habitat has go deficient throughout the isle. In 2005, 131 Hainan Eld's deer were reintroduced to Chihao region, where forests are still numerous, seldom disturbed and mainly distributed at loftier altitudes. Intensive agricultural land and villages are present at lower altitudes, and slash-and-burn plots and successional stages, such every bit shrub-grassland, are scattered beyond the mural. Eld'southward deer went extinct from Chihao in the 1950s.

Because they seldom experience man disturbance or predators, reintroduced ungulates are oft naïve and may quickly modify their behavior and habits to adapt to new environments. Behavioral patterns of the reintroduced Eld'southward deer at Chihao are consistent with the take chances-disturbance hypothesis: deer flee humans upon visual or other contact, and more significantly, shift to a nocturnal foraging blueprint to avert act and disturbance [29]. The general furnishings of reintroduction on deer beliefs are well understood. All the same, ongoing management plans and future strategies for reintroducing this threatened species depend on data virtually habitat usage and spatial interactions with humans, data that is currently lacking merely urgently needed. Specifically, we first need to know what kind(s) of landscape this population chooses. Second, we demand to know how individual deer utilise patches within the selected mural. Here, nosotros aimed to determine (1) whether deer locations are positively correlated with altitude from human villages, and (2) whether deer mainly inhabit forests because they are less exploited and are able to conceal them, or use agricultural areas thereby creating spatial overlap and increased interactions with humans.

Equally reintroductions become a larger function of conservation strategies for a number of key species, and because relatively few empirical studies of human-mediated effects on the distribution of reintroduced populations exist, our results volition help the management and fundamental understanding of animals reintroduced from reserves.

Report Area and Methods

Study Area

Chihao is located in western Hainan Isle, China (108°44′-108°56′East, 18°54′-xix°ten′N; Fig. ane). The study expanse is 450 km², with elevation ranging from x to 680 m. The climate is hot and dry. The almanac average temperature is 24.6°C; July is the hottest calendar month (29.one°C) and January is the coldest (18.four°C). Annual precipitation is 1012 mm, and evaporation is 2522 mm. About rain occurs in the wet flavor from July to Jan.

Chihao habitats are heavily modified. There are 11 villages in the study area with a combined population of x,000 people. Well-nigh residents are Li indigenous minority and indigenous to Hainan. The villagers grow rice, bananas, mangos, rubber trees and eucalyptus. They besides raise a small number of goats, cattle and pigs. Hunting is strictly forbidden in the area. V forest rangers patrol weekly and have occasionally found steel traps for rabbits laid past some older villagers that could accidentally trap deer. Farming, firewood collection and livestock herding occur year circular. Beyond a few free-ranging dogs and wild boar, at that place are no other large or medium-sized mammals in the wild.

The original vegetation is monsoon forest dominated by Terminalia hainanensis, Albizia odoratissima, Albizia procera, Lannea grandis, Aporosa chinensis, Euphoria longan and Diospyros philippensis [30]. About 35% of the area continues to sustain forest, where is hilly, elevated, and located several kilometers from villages. Agriculture covers 50% of the report area. Rice, bananas and mangos are grown on flat areas and gentle slopes. In hilly areas, minor plots of Chinese edible bean, maize, sweet potato and cassava are created and managed by slashing and burning vegetation. These plots are referred to as 'swiddens'. After 2 years, fertility declines and the swiddens are abased and natural vegetation returns, dominated by Imperata cylindrica var. major, Hyptis suavedens, Lygodium microstadyum, Desodium dunnii, and Acacia pennata. This kind of successional vegetation, described here as 'shrub-grasslands', contains the greatest species richness in the study area and is occasionally used for livestock grazing. Across the landscape, there are ix patch types relevant to people and deer (Tabular array 1).

Ethical Statement

This work follows an brute-capture protocol (Hudonghan #2004-125) canonical past the National Forestry Agency of People's republic of china. The study was carried out under the potency of a scientific permit issued past the Animal Welfare Commission of the Chinese Academy of Sciences.

Deer Relocation and Radio Tracking

Male person Eld'due south deer are larger than females (male, 100 kg; female, 60 kg). Mating occurs during the dry season from Feb to June. Females give birth in the moisture season between late September and early December [31]. The female person:male person ratio of deer in DNR–all descendants of the original 26 survivors–is 1∶0.7 [32]. They range freely within the enclosure and seldom encounter humans apart from reserve wardens. Deer in DNR were captured at random by encircling with a nylon net. A total of 131 Eld's deer (79 females and 52 males) were transferred to Chihao from March to July 2005 in six batches; 19 adults (two–viii years old; 10 females and ix males) were fitted with radio-collar transmitters. Five deer were fitted with Modernistic-335 (150–152 MHz, 140 chiliad; Telonics, AZ, USA) and released in March 2005, and 14 were fitted with SMRC-one (151–152 MHz, 160 grand; Lotek Wireless Inc., Newmarket, Canada) and released in July 2005.

Collared deer were intensively tracked on human foot and by motorcycle in the daytime from July 2005 to November 2006 using a portable receiver (Telonics TR-iv) equipped with a iii-element Yagi antenna (Telonics). Animal locations were determined by triangulation, and approximately ane location was obtained for each collared deer every iii days.

Domicile Range Calculation

ArcView 3.3 (Environmental Systems Inquiry Institute, Redlands, CA, USA) and Home Range Extensions were used to gauge abode ranges. The 95% fixed-kernel method with least-squares cross-validation was used to calculate individual almanac home ranges [33]–[35]. The written report period covered ii wet seasons and one dry out flavour: wet¹ (August–December 2005), dry (February 2006–June 2006) and wet² (Baronial–December 2006). The 100% minimum convex polygon (MCP) method was used to calculate individual seasonal home ranges considering the number of observations was less than 50 and the emphasis was on relative size [36].

Habitat Evaluation and Mapping

Habitat qualities of the seven vegetation patch types (forest, shrub-grassland, swidden, grassland, paddy field, dry field and plantation) were roughly evaluated on the basis of anthropogenic disturbance intensity, edible food abundance and capacity to conceal deer. We interviewed 100 representative local villagers in 2006 to determine the number of days in the previous year they spent in each patch blazon. Nosotros so used the number of activity days per unit area to judge human disturbance intensity. Nosotros nerveless and weighed the edible parts of grasses, shrubs and trees from 30 (1×one thousand) random samples in each patch blazon to approximate edible food abundance. Hiding cover (concealment) was measured using a x×130 cm plastic bar, following the method of Griffith and Youtie [37].

Nosotros obtained a landscape map based on land cover from the Hainan Forestry Bureau and corrected it by walking along patch boundaries of core areas of deer habitat with a GPS and so mapping it using ArcGIS 9.2 (Ecology Systems Research Institute). We obtained a human disturbance map based on the landscape map by substituting disturbance intensity for each patch type. We also obtained a digital elevation map (DEM; 30-m resolution) from the Country Agency of Surveying and Mapping of China. By overlaying telemetry locations and home ranges on these maps, nosotros were able to clarify habitat use and selection for habitat patch type, pinnacle and human being disturbance.

Habitat Pick Assay

Considering that our focus was on grouping habitat selection, and habitat availability was the same for all reintroduced individuals, nosotros pooled collared individuals for analysis. We asked two questions: (1) what blazon of landscape practise reintroduced deer select within the study area, and (2) what patches do animals adopt inside their selected mural? To reply the first question, we represented the bachelor habitat as a circumvolve with its center at the release site and a radius respective to the distance of the farthest recorded location of a collared individual. We defined the 97% kernel grouping home range as 'used landscape' (Fig. 2a). To respond the second question, we defined 100% MCP group home ranges as available habitat, and used the numbers of telemetry locations in each patch category to evaluate patch utilise (Fig. 2b).

Figure 2. Habitat option of Eld's deer on a landscape scale (a) and stand up calibration (b).

(a) The available landscape is indicated by a circle centered on the release site with a radius corresponding to the distance to the farthest location recorded for a collared individual. The used landscape is denoted by the convex polygon (blackness line) inside the circle, which is the 97% kernel domicile range calculated from all locations for collared Eld'southward deer (due north  = 17). (b) All collared deer telemetry locations (black points) and their 100% MCP dwelling range (available habitat) are shown.

https://doi.org/ten.1371/journal.pone.0091158.g002

We performed a log-likelihood test to determine whether deer selectively used habitat initially [38]: (one) where yard is the number of habitat categories, is the quantity of habitat patch category i used past deer, and is the expected quantity of habitat patch category i used. The cipher hypothesis is that deer used each patch category in proportion to its relative abundance (randomly used). If the zip hypothesis is rejected, at least 1 habitat experienced significant selection. We so determined which habitats were selected by applying the Manly-Chesson selectivity alphabetize (2) and Bonferroni-adjusted 95% confidence intervals (3) [38] to obtain the selectivity index as (ii) where O_i is the proportion of habitat i used, and π_i is the proportion of habitat i bachelor. For landscape-scale selection, habitat is preferred if the selectivity index is >1 and avoided if <1, (3) where due north is the total number of comparisons being made. For stand-scale pick, habitat category i is preferred if the interval is >ane and avoided when <1. If the confidence interval includes i, the habitat blazon is randomly used. The standard fault (4) is (iv) where u is the full number of telemetry locations.

Statistical Analysis

Values are expressed as ways ± standard deviations. All statistical analyses were performed using SPSS xviii.0. Pearson correlation analyses were used to describe relationships betwixt the number of telemetry locations and the altitude from villages, likewise as the relationship between annual domicile range size and the number of habitat categories it independent. Isle of mann-Whitney U-tests were used to compare differences in home range size between seasons. Habitat use in wet¹ season was influenced by the release of the animals, then only data from the dry out and wet² seasons were used in χ² tests.

Results

Overview of Location and Habitat

Two collared deer died very early on in the study, one from abdominal obstruction and one during a typhoon. In addition: ane deer (33F2) lost its collar; the batteries failed in the collars of three deer (36F5, 76F6 and 96M4); and the collar of one deer (71F2) remained undetected from January to March 2006. Ultimately, 12 individuals (6 females and vi males) were monitored over an entire year, and their locations were used to analyze dwelling house range and habitat utilize. Five collared individuals (3 females and 2 males) were monitored for less than a year, and their locations were merely used to analyze habitat selection (Tabular array 2).

In the Chihao region (CHR), act intensity is negatively related to elevation. Forests are mainly distributed at higher altitudes (mean = 272 thousand) and are the least disturbed landscape. Because forests also provide expert concealment, the forest is the safest of all patch types for deer. Among the three medium level disturbed patch types, shrub-grassland offered the best quality by virtue of its practiced food abundance and medium concealment. Swiddens were often adjacent to shrub-grasslands or embedded in them. Intensive agricultural areas were all heavily distributed and occupied lower altitudes (mean  = 94 m) (Table 3).

Spatial Distribution of Eld'southward Deer

Although our recordings showed that Eld'due south deer could travel up to 3.3 km (directly-line distance) within a single day, the boilerplate altitude from a known Eld's deer location to the nearest village was merely 1.3 km. The majority of deer locations (77%) were betwixt 0.6–2 km from villages. Only 5% of locations were farther than ii.four km from villages (Fig. 3). Within i km of villages, the number of deer locations (32% of total) was positively correlated with distance from the village (r = 0.996, p<0.001); however, this correlation was negative at distances greater than ane km from a village (r = −0.932, p<0.001).

Annual and Seasonal Home Ranges

The average annual domicile range for a reintroduced deer was 725±436 ha (north = 12), which is 55% of the size of the reserve from which they originated. Almanac home range was highly variable betwixt animals (range = 137–1384 ha). The seasonal dwelling ranges were 483±225 ha (range = 241–883 ha; n = ten) in wet^ane flavour, 698±624 ha (range = 72–1920 ha; n = xi) in the dry season, and 272±158 ha (range = 57–531 ha; n = 10) in wet² flavor. Home ranges in wetⁱ season were significantly larger than those in moistureⁱⁱ (Z = 2.269, p = 0.023). Domicile ranges in the dry flavor trended towards being larger than those in the two moisture seasons, but differences were not significant (Z = 0.211, p = 0.833; Z = 1.620, p = 0.105).

Habitat Types within Home Ranges

The annual abode range of each collared deer contained four to 9 patch types. In that location was a positive human relationship between home range size and the number of patch types (r = 0.534, p = 0.09; n = 12). Shrub-grassland, forest, swidden and water were features of all 12 collared deer home ranges. Although the dry field category composed a larger proportion of the almanac home range than swidden on average (Fig. 4), it was constitute in the home ranges of nine individuals. Ii other heavily disturbed patch categories, paddy field and village, were included in the home ranges of 8 and four animals, respectively. Grassland was contained within the home range of 1 animal and plantation was found in the home range of 2 animals. On average, shrub-grasslands constituted the bulk of dwelling house ranges (54.1%), followed by woods (26.4%) and farmland (dry field, swidden and paddy field; 15% in full) (Fig. 4).

Habitat Selection

Mural scale.

At the landscape scale, deer preferred areas above 150 m in tiptop and avoided areas beneath 150 yard in peak. Medium and weakly disturbed patches were overrepresented and heavily disturbed areas were underrepresented. Deer showed a articulate preference for shrub-grassland and swidden, and a very slight preference for forest. Surprisingly, villages were also overrepresented at this scale. Deer exhibited an society of abstention of grassland>plantation>paddy field> water> dry field (Table four). These patterns propose that Eld'due south deer inhabit areas comprising traditional agriculture and wood. They avoided both landscapes consisting of but forest and those dominated past intensive farmland (Table 4; Fig. 2a).

Stand scale.

Inside the selected landscape (100% MCP group dwelling house range), reintroduced deer selectively used habitats with different altitudes (χ² = 1903.seven, df = vi, p<0.001), patch categories (χ² = 1136.7, df = 8, p<0.001) and disturbance (χⁱⁱ = 7127.3, df = two, p<0.001).

At this calibration, just areas at elevations of 150–225 m were overrepresented. Elevations of 225–300 m were used in proportion yr circular, and areas beneath 150 thousand and in a higher place 300 1000 in tiptop were definitely avoided (Table v). χ² examination showed no significant difference in elevation utilise betwixt seasons (p>0.1).

Deer preferred medium-disturbed habitats throughout the year; both weakly and heavily disturbed habitats were avoided (Table 5). Areas suffering different disturbance levels were used differently beyond the seasons (χ² = 9.234, df = two, p = 0.01). For case, deer used medium-disturbed habitats more and low-disturbed habitats less during the wet season than the dry flavor.

Deer showed a preference for shrub-grasslands and swiddens throughout the twelvemonth. 2-thirds of their locations were distributed within shrub-grasslands (Table v; Fig. 2b). Forest to a higher place 300 m in elevation was avoided all yr round. Forest below 300 g in peak was overrepresented during the dry out flavor and was randomly used in the moisture season. Approximately 20–29% of locations were in forest. Regarding intensively cultured state, deer randomly used paddy fields in the dry season and dry fields in the wet season. Plantations, grassland and villages were avoided all year (Table 5). Habitat utilise was different between the dry out and moisture seasons (χ² = 19.5, df = 4, p = 0.001). For instance, deer used shrub-grasslands more frequently and forests less ofttimes during the wet season than during the dry season.

Discussion

Our results bear witness that reintroduced deer were active around villages that maintained traditional agricultural practices. The distance of reintroduced deer from villages was small, not simply compared with their daily movements, merely also compared with the distances that other populations or deer species maintain from infrastructure. For example, a comprehensive analysis of Eld's deer populations in Kingdom of cambodia, Myanmar, Lao, Thailand and Vietnam showed that they were more likely to be found at a greater altitude from human being settlements and more than 10 km from the nearest village [39]. A similar pattern of mule deer (Odocoileus hemionus) distribution effectually drilling wells was also found: although a single well pad typically disturbs 1.2–1.six ha of habitat, areas with the highest probability of apply by female mule deer were 2.7, iii.1 and 3.seven km away from well pads during years i, ii and 3 after development, respectively [forty].

Presumably, Eld'south deer could inhabit areas shut to humans at Chihao because they had not experienced human disturbance while living in the reserve, except during antler harvest days in Oct. Stankowich [41] plant that animals subjected to greater hunting activity are more wary. If animals seldom experience human hunting, they do non care for humans as threats and develop less fear [42], [43]. Further, better quality food (crops and fresh plant growth in fallows) is associated with proximity to humans; thus, animals were attracted to populated areas. Spatial heterogeneity was high in Chihao, even in areas near villages, and then Eld's deer can conceal themselves in these heterogeneous microhabitats.

Our evaluation of habitat selection was guided by the framework adult past Johnson [44] and the scales in our study basically corresponded to the second- and third-order choice [44]. In our analysis, we did not strictly follow Johnson'due south method considering Eld's deer are not territorial; their individual domicile ranges are their scope of motility, not the space they defend. Accordingly, we did not use individual home ranges when evaluating habitat selection. The process used to evaluate habitat selection is often a source of controversy. To strengthen our conclusions, we used several different methods–MCP 96 to 99 and kernel 95 to 99–to judge the used landscape at a big scale. The results from these methods showed like trends: shrub-grasslands and swiddens were preferred; forests were very slightly preferred or used randomly; and grasslands, plantations and intensively cultured fields were avoided. Villages were overrepresented considering villages in Chihao are small (<l ha each) and deer locations were close to villages, making information technology easy to include this patch type in used landscapes using the kernel or MCP method. Only selection for water was affected by the method used: it was preferred if the used landscape was estimated by MCP 98, MCP 99 and kernel 99; randomly used co-ordinate to kernel 98; and avoided co-ordinate MCP 96, MCP 97 and kernel 95–97. This variation reflected the fact that used landscape boundaries crossed the largest reservoir in the study expanse.

The option index for shrub-grassland was higher than that for forest at both scales, and home ranges independent more shrub-grassland than forest. Plants in shrub-grasslands are new growth, and deer are well known to prefer fresh and tender grasses and shrubs [45]. Given that 2-thirds of deer feeding activeness occurs during the dark [29], shrub-grasslands were likely to be used even more frequently than revealed by our information, which were collected during the day. Past contrast, the three subspecies of Eld's deer in Kingdom of cambodia, Myanmar, Lao, Thailand and Vietnam all occupy open up forests. C. e. thamin are nigh oftentimes found in dry, deciduous dipterocarp forests with an open up understory in central Myanmar [39], [46]. These differences in habitat selection are probably attributable to the fact that Eld's deer on Hainan Island were not very sensitive to humans, and were able to withstand certain human disturbances to obtain high quality food in shrub-grasslands and swiddens. Also, traditional agriculture lands were closer to water than forests. Similarly, Eld's deer at some other reintroduced site (Houmiling Nature Reserve) bear witness a preference for habitat near h2o [47]. In addition, forests in Chihao often occupy hills above 250 m in elevation and McShea et al. reported detecting no Eld's deer populations at elevations above 400 m [39], [48].

Although forests offer better concealment, ample provender and are the least disturbed patch blazon, relocated deer selected medium-disturbed areas where high quality food and water were readily attainable. This is reverse to predictions from the 'risk-disturbance hypothesis', which posits that animals adopt habitats away from man activeness or infrastructure, even if these areas offer poorer grazing conditions [49], [50]. Roe deer in modified landscapes retain stiff links to woodland structures if woodland fragments are numerous and widely dispersed, but they adopt an open up field habit where the remaining woodland is clumped, with lilliputian edge [51]. Both roe deer and Eld's deer testify behavioral plasticity and ecological flexibility, which enables them to brand use of secondary matrix habitats. Past fleeing budgeted people and remaining concealed during times of superlative human action [29], but withstanding certain human activities, they are able to exploit resource and optimize their fettle. This indicates that avoidance is not the but principle that guides the behavior of wild ungulates. If human being activities are non-fatal, there will be tradeoffs and potential benefits from closer association.

In the dry out season, vegetation concealment, food and h2o deteriorated. Deer used forest more frequently, as it is the safest patch type, and visited paddy fields more for nutrient and water. Additionally, Eld's deer enter oestrus and seek mates in the dry season [52]. Therefore, the domicile ranges of Eld's deer were larger in the dry season than in the wet season, in understanding with a study on C. e. thamin, another subspecies of Eld's deer in the Chatthin Wildlife Sanctuary of Myanmar [53]. Comparisons between the two wet seasons indicated that efforts to locate suitable habitat during the early stages of reintroduction might event in increased deer motion.

The annual dwelling house range size of C. e. hainanus (seven.3 km²) is similar to that of C. e. thamin (8 km²) in Myanmar. Human action in the Chatthin Wild animals Sanctuary impacts deer to a lesser degree [48], [53]; here, man disturbance and habitat differences do not appear to take reduced or enlarged habitation range sizes overall. All the same, in Myanmar, the annual home range of females is smaller than males, a design consistent beyond a number of taxa [54]–[56]. At Chihao, the home range of females was larger than that of males. Possible differences in the ways male person and female C. e. hainanus respond to anthropogenic environments will be the bailiwick of our next study.

Management Implications

Our results provide managers with practical information on the reintroduction of Eld'southward deer to a patchy landscape. The reintroduced Eld'southward deer chose to inhabit landscapes comprising traditional agronomics and forest. These findings provide a reminder that preservation of traditional cultivation can be very important for deer survival but that ingather raiding past Eld's deer will be inevitable. To reduce human-deer conflict and clinch successful re-establishment of viable populations, we recommend that reintroduced populations exist co-managed by the DNR and local communities. Customs co-management is helpful in controlling anthropogenic disturbances to Eld's deer, specially during their rutting and breeding periods. Although forests were not preferred past deer as shrub-grasslands were, they appeared in each focal animal's home range. Considering forests are the to the lowest degree disturbed habitat, they might human action as corridor for dispersal of deer into and across agronomical areas [57], and provide adequate safety during pregnancy, calving and antler growth [58]–[threescore]. Our results support the thought that big sections of woods are predictive of Eld'southward deer presence [39], and suggest that maintaining forested areas is essential to the time to come recovery of this species. In addition, collared deer had a large annual home range (up to 1384 ha), indicating that Eld'southward deer in the fenced DNR have restricted movement and spatial beliefs. Although the fenced DNR allowed for the increase in population size, information technology should be enlarged by incorporating circumjacent regions with low disturbance and good habitat conditions.

Acknowledgments

We appreciate S. Y. Li, H. Zhang, L. C. Zhang, Z. C. Chen, Z. P. Yang, M. L. Fu and staff at Chihao Reserve Station for assistance in the field.

Writer Contributions

Conceived and designed the experiments: ZGZ YLS DP. Performed the experiments: DP ZGZ. Analyzed the data: DP. Wrote the paper: DP ZGZ YLS BDB.

References

1. 1. Vitousek PM, Mooney HA, Lubchenco J, Melillo JM (1997) Human domination of earth'southward ecosystems. Scientific discipline 277: 494–499.
   • View Article
   • Google Scholar
2. 2. Naughton-Treves L, Holland MB, Brandon Grand (2005) The role of protected areas in conserving biodiversity and sustaining local livelihoods. Annu Rev Environ Resour 30: 219–52.
   • View Article
   • Google Scholar
3. iii. Caro TM, Rejmanek Yard, Pelkey N (2000) Which mammals do good from protection in East Africa. In: Book Entwistle A, Dunstone Northward, editors. Priorities for the conservation of mammalian diversity: Has the panda had its day? Cambridge: UK Cambridge Academy Press. 3: pp. 221–238.
   • View Commodity
   • Google Scholar
4. four. Frid A, Dill 50 (2002) Human-caused disturbance stimuli every bit a form of predation risk. Conserv Ecol 6: 11 Available: http://www.consecol.org/vol6/iss1/art11.
   • View Article
   • Google Scholar
5. 5. Beale CM, Monaghan P (2004) Human being disturbance: people equally predation-costless predators? J Appl Ecol 41: 335–343.
   • View Commodity
   • Google Scholar
6. vi. MacArthur RA, Geist V, Johnston RH (1982) Cardiac and behavioral-responses of mountain sheep to homo disturbance. J Wildl Manage 46: 351–358.
   • View Commodity
   • Google Scholar
7. seven. Fowler GS (1999) Behavioral and hormonal responses of Magellanic penguins (Spheniscus magellanicus) to tourism and nest site visitation. Biol Conserv 90: 143–149.
   • View Commodity
   • Google Scholar
8. viii. Jayakody S, Sibbald AM, Gordon IJ, Lambin Ten (2008) Cherry-red deer Cervus elaphus vigilance behaviour differs with habitat and blazon of homo disturbance. Wildl Biol xiv: 81–91.
   • View Commodity
   • Google Scholar
9. ix. Boroski BB, Mossman AS (1998) Water use patterns of mule deer (Odocoileus hemionus) and the effects of human disturbance. J Arid Environ 38: 561–569.
   • View Article
   • Google Scholar
10. 10. Kerley LL, Goodrich JM, Miquelle DG, Smirnov EN, Quigley HB, et al. (2002) Effects of roads and human disturbance on Amur tigers. Conserv Biol 16: 97–108.
    • View Article
    • Google Scholar
11. xi. Approach B, Razin 1000 (2006) Effect of human being activities on bearded vulture behaviour and convenance success in the French Pyrenees. Biol Conserv 128: 276–284.
    • View Commodity
    • Google Scholar
12. 12. Vistnes I, Nellemann C (2008) The matter of spatial and temporal scales: a review of reindeer and caribou response to human activity. Polar Biol 31: 399–407.
    • View Article
    • Google Scholar
13. xiii. LaGory KE (1987) The influence of habitat and grouping characteristics on the alarm and flight response of white-tailed deer. Anim Behav 35: twenty–25.
    • View Article
    • Google Scholar
14. 14. Trombulak SC, Frissell CA (2000) Review of ecological furnishings of roads on terrestrial and aquatic communities. Conserv Biol 14: eighteen–30.
    • View Article
    • Google Scholar
15. 15. Berger J, Daneke D, Johnson J, Berwick SH (1983) Pronghorn foraging economy and predator avoidance in a desert ecosystem: implications for the conservation of large mammalian herbivores. Biol Conserv 25: 193–208.
    • View Article
    • Google Scholar
16. 16. Madsen J (1994) Impacts of disturbance on migratory waterfowl. Ibis 137: 567–574.
    • View Article
    • Google Scholar
17. 17. Cassirer EF, Freddy DJ, Ables ED (1992) Elk Responses to disturbance by cross-state skiers in Yellowstone National Park. Wildlife Soc B xx: 375–381.
    • View Article
    • Google Scholar
18. xviii. Bonenfant Chiliad, Kramer DL (1996) The influence of distance to burrow on flight initiation distance in the woodchuck, Marmota monax. Behav Ecol 7: 299–303.
    • View Article
    • Google Scholar
19. xix. De la Torre Due south, Snowdon CT, Bejarano M (2000) Effects of human activities on wild pygmy marmosets in Ecuadorian Amazonia. Biol Conserv 94: 153–163.
    • View Article
    • Google Scholar
20. twenty. Harveson PM, Lopez RR, Collier BA, Silvy NJ (2007) Impacts of urbanization on Florida Fundamental deer behavior and population dynamics. Biol Conserv 134: 321–331.
    • View Article
    • Google Scholar
21. 21. Basille K, Herfindal I, Santin-Janin H, Linnell JDC, Odden J, et al. (2009) What shapes Eurasian lynx distribution in human dominated landscapes: selecting prey or avoiding people? Ecography 32: 683–691.
    • View Article
    • Google Scholar
22. 22. West BC, Parkhurst JA (2002) Interactions between Deer Impairment, Deer density, and Stakeholder Attitudes in Virginia. Wild animals Soc B thirty: 139–147.
    • View Article
    • Google Scholar
23. 23. Drake D, Paulin JB, Curtis PD. Decker DJ, Julian GJS (2005) Assessment of negative economic impacts from deer in the northeastern U.s.. J Etension 43: 1RIB5 Available: http://www.joe.org/joe/2005february/rb5.php.
    • View Commodity
    • Google Scholar
24. 24. Timmins RJ, Duckworth JW (2008) Rucervus eldii. In 'IUCN 2012: IUCN Red List of threatened species'. Bachelor: http://www.iucnredlist.org.
25. 25. Zhang Q, Zeng ZG, Sunday LF, Song YL (2009) The origin and phylogenetics of Hainan Eld's deer and implications for Eld'southward deer conservation. Acta Theriologica Sinica 29: 365–371.
    • View Article
    • Google Scholar
26. 26. Zeng ZG, Song YL, Li JS, Teng LW, Zhang Q, et al. (2005) Distribution, status and conservation of Hainan Eld's deer (Cervus eldi hainanus) in China. Folia Zool 54: 249–257.
    • View Article
    • Google Scholar
27. 27. Fu ML, Ren Y, Hou R, Lin SL (2013) Effect of python population dynamic and distribution on Hainan Eld's deer population dynamics. Chinese Journal of Wildlife 34: 40–42.
    • View Article
    • Google Scholar
28. 28. Xu LH, Liu ZH, Liao WP, Li XH, Yu SM, et al.. (1983) Birds and mammals of Hainan Island. Beijing: Science Press. pp. 398–401.
29. 29. Pan D, Teng LW, Cui FJ, Zeng ZG, Bravery BD, et al. (2011) Eld's deer translocated to human-inhabited areas get nocturnal. AMBIO 40: 60–67.
    • View Commodity
    • Google Scholar
30. xxx. Wang BS, Zhang WY (2002) The groups and features of tropical forest vegetation of Hainan Island. Guihaia 22: 107–115.
    • View Commodity
    • Google Scholar
31. 31. Song YL, Zeng ZG (2003) Reproductive characteristics of female Eld's deer in their natural range in Hainan Isle, People's republic of china. Mammalia 67: 47–54.
    • View Article
    • Google Scholar
32. 32. Song YL, Li SY (1992) On population dynamics of Hainan Eld's deer (Cervus eldi hainanus) in Datian Nature Reserve, Hainan. Acta Zoologica Sinica 38: 165–171.
    • View Commodity
    • Google Scholar
33. 33. Worton BJ (1989) Kernel methods for estimating the utilization distribution in home-range studies. Ecology 70: 164–168.
    • View Article
    • Google Scholar
34. 34. Seaman DE, Powell RA (1996) An evaluation of the accuracy of kernel density estimators for home range analysis. Ecology 77: 2075–2085.
    • View Article
    • Google Scholar
35. 35. Powell RA (2000) Animal home ranges and territories and home range estimators. In: Boitani Fifty, Fuller TK, editors. Inquiry techniques in brute ecology: Controversies and consequences. New York: Columbia University Press. pp. 65–110.
36. 36. Dixon KR, Chapman JA (1980) Harmonic mean measure of animal activity areas. Ecology 61: 1040–1044.
    • View Article
    • Google Scholar
37. 37. Griffith B, Youtie BA (1988) Two devices for estimating foliage density and deer hiding cover. Wildlife Soc B sixteen: 206–210.
    • View Article
    • Google Scholar
38. 38. Manly BFJ, McDonald LL, Thomas DL, MacDonald TL, Erickson WP (2002) Resource selection by animals: statistical design and analysis for field studies. 2nd edition. New York: Kluwer Academic Publishers. 221 p.
39. 39. McShea WJ, Koy K, Clements T, Johnson A, Vongkhamheng C, et al. (2005) Finding a needle in the haystack: regional assay of suitable Eld's deer (Cervus eldi) forest in Southeast Asia. Biol Conserv 125: 101–111.
    • View Article
    • Google Scholar
40. 40. Sawyer H, Nielson RM, Lindzey F, McDonald LL (2006) Winter habitat selection of mule deer earlier and during development of a natural gas field. J Wildl Manage seventy: 396–403.
    • View Article
    • Google Scholar
41. 41. Stankowich T (2008) Ungulate flying responses to human disturbance: A review and meta-analysis. Biol Conserv 141: 2159–2193.
    • View Article
    • Google Scholar
42. 42. Dwyer CM (2004) How has the risk of predation shaped the behavioural responses of sheep to fright and distress? Anim Welfare 13: 269–281.
    • View Article
    • Google Scholar
43. 43. Blumstein DT, Daniel JC (2005) The loss of anti-predator behaviour following isolation on islands. Proc R Soc B 272: 1663–1668.
    • View Article
    • Google Scholar
44. 44. Johnson DH (1980) The comparing of usage and availability measurements for evaluating resource preference. Environmental 61: 65–71.
    • View Article
    • Google Scholar
45. 45. Yuan Xc, Liu XM, Wang J, Fu GA, Li SY (1990) Feeding habits of Hainan Thamin deer. Journal of Northeast Forestry University 18: 66–71.
    • View Article
    • Google Scholar
46. 46. McShea WJ, Leimgruber P, Aung K, Monfort SL, Wemmer C (1999) Range collapse of tropical cervid (Cervus eldi) and the extent of remaining habitat in central Myanmar. Anim Conserv 2: 173–183.
    • View Article
    • Google Scholar
47. 47. Wang LB, Yan HM, Lu XL, Yuan XC, Tao FW, et al. (2010) Habitat selection of released Hainan Eld'south deer in Houmiling Nature Reserve. Sichuan Periodical of Zoology 29: 929–932.
    • View Article
    • Google Scholar
48. 48. McShea WJ, Aung Yard, Poszig D, Wemmer C, Monfort SL (2001) Forage, habitat use, and sexual segregation by a tropical deer (cervus eldi thamin) in a dipterocarp forest. J Mammal 82: 848–857.
    • View Article
    • Google Scholar
49. 49. Nellemann C, Jordhøy P, Støen OG, Strand O (2000) Cumulative impacts of tourist resorts on wild reindeer (Rangifer tarandus tarandus) during wintertime. Arctic 53: 9–17.
    • View Article
    • Google Scholar
50. 50. Vistnes I, Nellemann C (2001) Abstention of cabins, roads, and power lines by reindeer during calving. J Wildl Manage 65: 915–925.
    • View Article
    • Google Scholar
51. 51. Hewison AJM, Vincent JP, Joachim J, Angibault JM, Cargnelutti B, et al. (2001) The effects of woodland fragmentation and homo activity on roe deer distribution in agricultural landscapes. Can J Zool 79: 679–689.
    • View Article
    • Google Scholar
52. 52. Zeng ZG, Song YL, Zhang Q (2011) Inter-individual variations in male and female mating bonds within a population of Eld'southward deer in Hainan Island, Communist china. Mammalia 75: 143–150.
    • View Commodity
    • Google Scholar
53. 53. Aung M, McShea WJ, Htung Due south, Than A, Soe TM, et al. (2001) Environmental and social organization of a tropical deer (Cervus eldi thamin). J Mammal 82: 836–847.
    • View Article
    • Google Scholar
54. 54. Ciuti South, Davini South, Luccarini Due south, Apollonio 1000 (2004) Could the predation risk hypothesis explain large-scale spatial sexual segregation in dormant deer (Dama dama)? Behav Ecol Sociobiol 56: 552–564.
    • View Article
    • Google Scholar
55. 55. Apollonio M, Ciuti S, Luccarini Due south (2005) Long-term influence of human presence on spatial sexual segregation in dormant deer (Dama dama). J Mammal 86: 937–946.
    • View Article
    • Google Scholar
56. 56. Schaefer JA, Mahoney SP (2007) Effects of progressive clearcut logging on Newfoundland caribou. J Wildl Manage 71: 1753–1757.
    • View Commodity
    • Google Scholar
57. 57. Naughton-Treves L, Mena JL, Treves A, Alvarez Northward, Radeloff VC (2003) Wildlife survival across park boundaries: the touch of slash-and-burn agriculture and hunting on mammals in Tambopata, Peru. Conserv Biol 17: 1106–1117.
    • View Commodity
    • Google Scholar
58. 58. Geist V (1971) Is big game harassment harmful? Oilweek 22: 12–thirteen.
    • View Commodity
    • Google Scholar
59. 59. Kuss FR, Graefe AR, Vaske JJ (1990) Visitor affect management: a review of research. Washington: National Parks and Conservation Clan. 256 p.
60. 60. Phillips GE, Alldrege AW (2000) Reproductive success of elk following disturbance past humans during calving season. J Wildl Manage 64: 521–530.
    • View Commodity
    • Google Scholar

Source: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0091158

Posted by: watkinsworick.blogspot.com

Share this post