Skip to main content Accessibility help
×
Home

Information:

  • Access
  • Cited by 5

Actions:

      • Send article to Kindle

        To send this article to your Kindle, first ensure no-reply@cambridge.org is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle. Find out more about sending to your Kindle.

        Note you can select to send to either the @free.kindle.com or @kindle.com variations. ‘@free.kindle.com’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

        Find out more about the Kindle Personal Document Service.

        Population size, distribution and status of the remote and Critically Endangered Bawean deer Axis kuhlii
        Available formats
        ×

        Send article to Dropbox

        To send this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Dropbox.

        Population size, distribution and status of the remote and Critically Endangered Bawean deer Axis kuhlii
        Available formats
        ×

        Send article to Google Drive

        To send this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Google Drive.

        Population size, distribution and status of the remote and Critically Endangered Bawean deer Axis kuhlii
        Available formats
        ×
Export citation

Abstract

Conservation of rare ungulates requires reliable population size estimates and distribution maps for prioritizing investments and assessing the effectiveness of conservation measures. We used both camera trapping and a random encounter model approach, and faecal pellet group counts, to update the range and population size of the Bawean deer Axis kuhlii in the Bawean Island Nature Reserve and Wildlife Sanctuary, Indonesia. We studied 2-month periods to fulfil the assumption of population closure. Both methods provided similar population density estimates (higher in the dry season) of c. 227–416 individuals. The estimated range of the species is significantly narrower than previously reported. The main threats (habitat loss as a result of illegal logging, and disturbance by dogs and hunters) are ongoing. Based on these results we suggest that the species should retain its Critically Endangered status on the IUCN Red List.

Footnotes

*

Also at: Faculty of Forestry, Department of Forest Resources Conservation and Ecotourism, Bogor Agricultural University, Bogor, Indonesia

To view supplementary material for this article, please visit http://dx.doi.org/10.1017/S0030605316000429.

Introduction

Reliable information on population size and range, and any trends in these parameters, is required to assess the conservation status of a species using the Red List criteria (IUCN, 2001). In the absence of such information, conservation management is often based on crude estimates, expert opinion or educated guesses, which may result in erroneous decisions that can be counter-productive (Akҫakaya, Reference Akçakaya2002; Blake & Hedges, Reference Blake and Hedges2004; Murray et al., Reference Murray, Goldizen, O'Leary, McAlpine, Possingham and Choy2009). The Bawean deer Axis kuhlii (Temminck, 1836), the most isolated deer in the world and the only endemic deer species in Indonesian tropical rainforest, is categorized as Critically Endangered on the IUCN Red List (Semiadi et al., Reference Semiadi, Duckworth and Timmins2015). The Bawean deer is reported to range over a very small area restricted to the Bawean Island Nature Reserve and Wildlife Sanctuary and a peninsula on the north-west side of the island (Tanjung Cina; Lachenmeier & Melisch, Reference Lachenmeier and Melisch1996; Grubb, Reference Grubb, Wilson and Reeder2005). The protected area is relatively close to human settlements, and illegal logging is not uncommon in the forest habitat. Listed in Appendix I of CITES (2016), this taxon is legally protected and is one of 25 species prioritized for conservation by the Indonesian government on the basis of their threatened status (decree SK.180/IV-KKH/2015; Ministry of Environment and Forestry, 2015). Despite this status, and threats across its range, surprisingly little is known about the Bawean deer and no long-term monitoring has been implemented, partly because this is not a charismatic species.

Several methods have previously been used to study population trends and distribution in this species: faecal sampling (Blouch & Atmosoedirdjo, Reference Blouch and Atmosoedirdjo1978; Blouch, Reference Blouch1980; LIPI & IPB, 1999; Semiadi, Reference Semiadi2004; BBKSDA East Java, 2009), footprint (UGM & BBKSDA East Java, 2003) and call counts (BBKSDA East Java, 2009), and camera trap surveys (UGM & BBKSDA East Java, 2004). The latter study, in which 10 camera traps were installed at seven locations (Lang Pelem river, Lampeci river, Tambelang river, Mt Tinggi, Angsana block, Tanjung Putri block and Tanjung Cina) during 20 days, recorded no evidence of the Bawean deer, although this may be attributable to the short duration of the study and the placement of cameras in unsuitable locations.

Capture–recapture methods for estimating population size require individuals to be recognizable, either by rings or collars (e.g. Trolle et al., Reference Trolle, Noss, Passos and Oliveira2008; Oliveira-Santos et al., Reference Oliveira-Santos, André, Castro and Crawshaw2010) or by natural marks such as stripes, spots or scars (e.g. Kumbhar et al., Reference Kumbhar, Prabhu, Yogesh, Francies, Patwardhan, Qureshi and Jhala2013). They are not applicable to the many mammal species that lack distinctive marking, such as the Bawean deer, except when bucks are seasonally antlered.

The development of the random encounter model, a by-product of an ideal gas model (Hutchinson & Waser, Reference Hutchinson and Waser2007), has facilitated estimations of species densities from unmarked individuals with a known speed, and sensor detection parameters (Rowcliffe et al., Reference Rowcliffe, Field, Turvey and Carbone2008). The random encounter model has been implemented successfully for ungulate species by deploying cameras in systematic or fully randomized arrays (Rowcliffe et al., Reference Rowcliffe, Field, Turvey and Carbone2008; Rovero & Marshall, Reference Rovero and Marshall2009; Zero et al., Reference Zero, Sundaresan, O'Brien and Kinnaird2013; Carbajal-Borges et al., Reference Carbajal-Borges, Godínez-Gómez and Mendoza2014).

In this study we used two methods to estimate the abundance and map the range of the Bawean deer in the Bawean Island Nature Reserve and Wildlife Sanctuary, and assess its IUCN status. Density was estimated using both the random encounter model with camera trapping data and faecal pellet group counts; the latter technique was the most commonly used in previous studies.

Study area

Indonesia's Bawean Island (200 km2) is relatively isolated in the Java Sea (Fig. 1). Based on the classification of Smith and Ferguson, its climate is categorized as type C (Semiadi, Reference Semiadi2004). Rainfall is abundant during the north-west monsoon, from the end of October until April, and reaches c. 2,500 mm on the southern coast. Temperature is almost uniform throughout the year, with mean maximum and minimum temperatures of 32 and 22°C, respectively (Semiadi, Reference Semiadi2004). The study area encompasses 46.6 km2 of the Bawean Island Nature Reserve and Wildlife Sanctuary, which is characterized by steep topography (with slopes > 60°) and a wide altitudinal gradient (up to 630 m). The main vegetation type is evergreen tropical forest, which covers 23% of the island. A mosaic of closed and open forest as well as permanently dry and seasonally flooded habitat types occur in the study area, including gallery forest, semi-deciduous forest with understorey, shrub and grassland, and teak Tectona grandis plantations (60% of the area), which are all globally threatened by deforestation and climate change. The remaining natural forests are confined to the steep sides and tops of the higher hills and mountains, often occurring as islands surrounded by teak (Semiadi, Reference Semiadi2004).

Fig. 1 Camera trap grid and camera locations in the Bawean Island Nature Reserve and Wildlife Sanctuary, Indonesia.

Methods

Camera trapping

The Bawean Island Nature Reserve and Wildlife Sanctuary was divided into 20 4-km2 grid cells using the geographical information system ArcGIS 10.2.2 (ESRI, Redlands, USA). The camera trapping survey was conducted during both the wet (March–April and November 2014) and dry seasons (May–October 2014). Twenty Trophy Cam HD Max digital cameras (Bushnell Outdoor Products, Overland Park, USA) operating on passive infrared motion sensors were installed 30–50 cm above the ground, perpendicular to the ground, to record both small and large animals. Herbaceous vegetation in the vicinity of the cameras was cleared to avoid interference (Tobler et al., Reference Tobler, Carrillo-Percastegui, Leite-Pitman, Mares and Powell2008; Team Network, 2011; Rovero et al., Reference Rovero, Zimmermann, Berzi and Meek2013). The cameras were set at 1 minute video mode with 1 minute intervals. The total survey effort was 5,500 trap days. Sampling precision was assessed as the coefficient of variation of trap rates with cumulative trapping effort (cameras × days). The sampling precision for the Bawean deer increased up to an optimum trapping effort of 520–560 camera days (Fig. 2). One camera trap per grid cell was deployed, in open and accessible locations, applying a buffer equivalent to half of the mean maximum distance moved, (1/2MMDM) to reduce the likelihood of capturing the same individual twice (Karanth & Nichols, Reference Karanth and Nichols1998; Soisalo & Cavalcanti, Reference Soisalo and Cavalcanti2006).

Fig. 2 Camera trap sampling precision expressed as the coefficient of variation of Bawean deer trapping rates with cumulative trap effort (number of cameras × trapping days).

Before installation we collected evidence of the presence of the Bawean deer (footprints, faeces, food remains, antler rubbing on trees) throughout the grid in the Bawean Island Nature Reserve and Wildlife Sanctuary and also in Tanjung Cina. We selected camera locations using the two following procedures. Firstly, we superimposed a grid of 100 × 100 m cells over the study area using the Fishnet and Clip tools in ArcGIS 10.2.2 (ESRI, Redlands, USA). This generated 3,600 potential camera locations from which we randomly selected one for each 4 km2 grid cell in the field. In general, cameras were not moved during the study; however, if a camera did not capture any mammal in 2–3 periods of checking we moved it 300–500 m from the original location within the same grid cell. Secondly, we selected locations where we found evidence of the presence of the Bawean deer. In practice, few camera traps were placed preferentially in this way (only 21 locations), as signs of the deer were difficult to find in the field. Moreover these cameras, which were expected to confirm the efficacy of the equipment, did not take significantly more photographs than randomly set cameras. In total, 75 locations were sampled during the survey period. Cameras were checked once every 28–31 days, and batteries and memory cards were replaced as necessary. Malfunctioning cameras were replaced to avoid loss of data.

Faecal pellet group count

Within each grid cell we counted faecal pellet groups in four plots (7 × 7 m) around the camera trap, spaced 10 m apart, according to Acevedo et al. (Reference Acevedo, Ferreres, Jaroso, Durán, Escudero, Marco and Gortázar2010) and Alves et al. (Reference Alves, da Silva, Soares and Fonseca2013). A total of 300 square plots (4 × 75 camera trap locations) were surveyed during the wet (February–March 2014) and dry seasons (August–September 2014). After the initial removal of all pellets present in each plot we calculated the faecal accumulation rate by recording the monthly deposition of pellets after the initial removal of all pellets present in the plot (deposition time of faecal groups). This method is appropriate for rapid surveys and when it is quite difficult to find a new group of faecal pellets in the field (Prugh & Krebs, Reference Prugh and Krebs2004; St-Laurent & Ferron, Reference St-Laurent and Ferron2008).

Random encounter model

The random encounter model uses the rate of contact between moving animals and static camera traps to estimate species density. It requires estimation of species-specific camera trap detection (Carbone et al., Reference Carbone, Christie, Conforti, Coulson, Franklin and Ginsberg2001), along with a camera trap detection specified by radius and angle, and an estimated day range based on speed of movement and activity data (Rowcliffe et al., Reference Rowcliffe, Carbone, Jansen, Kays and Kranstauber2011, Reference Rowcliffe, Kays, Kranstauber, Carbone and Jansen2014). The model is based on three main assumptions: (1) animals conform adequately to the model used to describe the detection process (i.e. they behave like particles of an ideal gas, moving randomly and independently of one another), (2) photographs represent independent contacts between animals and cameras, and (3) the population is closed (Rowcliffe et al., Reference Rowcliffe, Field, Turvey and Carbone2008). To fulfil these assumptions, camera traps were set at least 300 m apart to reduce the likelihood of capturing the same individual twice, and increase independence of locations (Kays et al., Reference Kays, Kranstauber, Jansen, Carbone, Rowcliffe, Fountain and Tilak2009). We studied 2-month periods in the dry season (June–July and August–September) on the basis that in such short sampling times the probability of birth, death, migration or immigration events is low (Karanth & Nichols, Reference Karanth and Nichols1998; Silver et al., Reference Silver, Ostro, Marsh, Maffei, Noss and Kelly2004; Soria-Dίaz & Monroy-Vilchis, Reference Soria-Dίaz and Monroy-Vilchis2015).

We used the following equation to obtain density estimates from camera trap encounter rates (Rowcliffe et al., Reference Rowcliffe, Field, Turvey and Carbone2008):

$$gD\, = \,\displaystyle{y \over t}\displaystyle{\pi \over {vr(2 + \theta )}}$$

where y/t = trapping rate (number of independent photographic events per camera trap day), v = a species’ mean daily speed of movement (km day−1), r = radius of the camera trap detection zone (km) and θ = angle of the camera trap detection zone (radians). The outcome can then be multiplied by g (mean group size), as the independent unit recorded by the camera is the group rather than the individual (Rowcliffe et al., Reference Rowcliffe, Field, Turvey and Carbone2008; Zero et al., Reference Zero, Sundaresan, O'Brien and Kinnaird2013).

Independent photographic events were defined as individuals entering and exiting the field of view (Cusack et al., Reference Cusack, Swanson, Coulson, Packer, Carbone and Dickman2015). An individual's mean daily speed of movement was calculated as the speed recorded from camera trap videos multiplied by the proportion of time spent active (Rowcliffe et al., Reference Rowcliffe, Kays, Kranstauber, Carbone and Jansen2014). The radius of the camera trap detection zone was calculated by measuring the distance from the camera to the location of deer at the first trigger, based upon marked locations in the field. The angle of the camera trap detection zone was obtained in the field by detecting a stick in six paired approaches perpendicular to the sensor beam at a distance of 4 m and using a compass placed on a flat surface directly below the camera. The angle of maximal detection was then converted to radians for calculations. We used mean values of r and θ from all cameras in the calculations. The mean group size for this mostly solitary species was influenced mainly by does and their young.

Faecal accumulation rate

Population density (D, individuals per km2) was estimated using the equation proposed by Eberhardt & Van Etten (Reference Eberhardt and Van Etten1956): D = (NP × Dpg)/(T × dR), where NP = number of plots per km2, Dpg = mean number of faecal pellet groups, T = deposition time of faecal pellet groups, and dR = defecation rate. In the absence of field data we used the observed defecation rate of captive Bawean deer (13 faecal pellet groups per individual per day; Blouch & Atmosoedirdjo, Reference Blouch, Atmosoedirdjo and Wemmer1987). We extrapolated the calculated densities to the total protected area on Bawean island to estimate the population size. The standard error of the estimates was computed using the delta method (Seber, Reference Seber1982).

Distribution

We mapped the distribution and abundance based on the random encounter model and the faecal pellet count (map source: Bakosurtanal, 2002) using ArcGIS.

Results

Random encounter model

We recorded 118 photographs of Bawean deer (2.15 individuals per 100 trap-days), none during March–April, 10 in May (but only two independent photographic events), 19 in June, 23 in July, 32 in August, 22 in September, 6 in October and 6 in November. Random encounter modelling was performed bimonthly for June–July and August–September, with 32 and 50 independent photographic events, respectively. All variables and estimates for preferentially set, randomly set and all cameras are summarized in Table 1. Obtaining similar estimates for both sets of cameras, estimations for all cameras were 4.87 ± SE 1.05 individuals per km2 in June–July and 8.92 ± SE 1.17 in August–September, yielding population estimates of 227 ± SE 33 and 416 ± SE 55, respectively.

Table 1 Results of random encounter modelling to estimate the population of Bawean deer Axis kuhlii in the Bawean Island Nature Reserve and Wildlife Sanctuary (Fig. 1), for preferentially set (at locations with signs of deer), randomly set, and all cameras.

Faecal accumulation rate

We counted 30 and 50 faecal pellet groups after 60 days of accumulation in the wet and dry seasons, respectively. We estimated a density of 3.48 ± SE 2.61 individuals per km2 in the wet season (February–March) and 5.18 ±  SE  3.61 in the dry season (August–September). The population size over the sampled area was estimated to be 162 ± SE 122 and 242 ± SE 168 in the wet and dry seasons, respectively.

Distribution

The presence of Bawean deer was recorded in eight and 11 grid cells using camera traps and faecal pellet counts, respectively, including seven grid cells by both techniques (Fig. 3). The recorded range of the species was restricted to the south-western part of the Bawean Island Nature Reserve and Wildlife Sanctuary, from Mount Bulu to Mount Bengkuang, at 34–320 m elevation. Only old faecal pellets were found at Mount Tinggi and Mount Besar. No deer were recorded in Tanjung Cina or in the north-east of the island by these techniques, nor did we find any footprints or other sign of presence at Mount Tinggi, Mount Beringin, Kastoba Lake or Mount Payung-Payung. The highest estimated abundance was in the vicinity of Mount Dedawang (36.8 and 28.3 individuals per km2 by random encounter modelling and faecal accumulation rate, respectively). Abundance was lower at Mount Duren (1.1 individuals per km2 by faecal accumulation rate) and Mount Besar (3.4 and 3.3 individuals per km2 by random encounter modelling and faecal accumulation rate, respectively), and also around Mount Bajapati (4.1 and 2.2 individuals per km2). It was intermediate in the mixed secondary and teak forest of Mount Bulu (21.5 and 7.9 individuals per km2).

Fig. 3 Seasonal distribution of abundance (individuals per km2) of Bawean deer in the Bawean Island Nature Reserve and Wildlife Sanctuary (Fig. 1) based on (a) camera trapping and random encounter modelling and (b) faecal pellet counts and faecal accumulation rates.

Discussion

Density estimates

We successfully tested the suitability of camera trapping and random encounter modelling for monitoring the status of the Bawean deer in its tropical forest habitat. The absence of photographs during the first 2 months (March–April) may be related to the presence of a high number of researchers at the beginning of the survey period, unsuitable locations of cameras, and lower activity of the species in the wet season. The trapping rate and density estimate increased during the dry season, peaking in August. As most deer were photographed when feeding, this could be related to the availability of food plants, which can become more scarce during the dry season, even in tropical habitats (Pontes & Chivers, Reference Pontes and Chivers2007), leading to wider movements.

Estimates obtained with the random encounter model were more precise (narrower confidence intervals) and higher than those obtained using faecal pellet counts. This discrepancy may be attributable to the decay of faeces, and the approximate values of some parameters, such as the speed of deer movement, and the camera detection zone used in random encounter modelling. All parameters, even those that are hard to obtain (Rowcliffe et al., Reference Rowcliffe, Carbone, Kays, Kranstauber and Jansen2012), should be measured more accurately for the Bawean deer in the future. The combination of camera trapping and global positioning system telemetry could improve the accuracy of estimates, not only for performing random encounter modelling but also for analysing how species’ home ranges can affect the required size of the area sampled. Such a study could also test the assumption of random distributions of cameras and wildlife (Cusack et al., Reference Cusack, Swanson, Coulson, Packer, Carbone and Dickman2015). The camera detection zone should also be investigated in different habitats and seasons; for example, we measured a lower detection radius in the wet season, although the difference was too weak to explain the absence of deer detection in that season.

Our findings suggest that random encounter modelling may yield accurate density estimates for elusive, rare and unmarked species, unlike photographic capture–recapture techniques, which require both unique markings and high-quality photographs for recognition of individuals (Soria-Dίaz & Monroy-Vilchis, Reference Soria-Dίaz and Monroy-Vilchis2015). Moreover, random encounter modelling is continually being improved (Rowcliffe et al., Reference Rowcliffe, Carbone, Jansen, Kays and Kranstauber2011). Both methods estimated the highest density in the dry season, as previously reported (Blouch & Atmosoedirdjo, Reference Blouch and Atmosoedirdjo1978), which supports the hypothesis that there is less movement in the wet season. The size of the Bawean deer population was estimated to be 242 individuals by the faecal pellet count method and 416 by random encounter modelling, which, compared to the previous estimate (250–300, by faecal pellet count; Semiadi, Reference Semiadi2004), suggests stability.

Distribution and conservation status

Our records indicate that the range of the Bawean deer has narrowed significantly. Camera trapping and faecal pellet counts proved to be complementary, with presence at Klumpang Gubuk recorded only by the latter technique. Unlike Blouch & Atmosoedirdjo (Reference Blouch, Atmosoedirdjo and Wemmer1987) and Semiadi (Reference Semiadi2004), we found Bawean deer only in the central mountain range and in the south-west of the Bawean Island Nature Reserve and Wildlife Sanctuary, around Mount Bulu. We assume that the deer is no longer present in Tanjung Cina, where a density of 11.8 individuals per km2 during the wet season was reported previously (Semiadi, Reference Semiadi2004). No sign of presence was recorded at Mount Tinggi, Mount Beringin, Kastoba Lake or Mount Payung-Payung. Records by Sitwell (Reference Sitwell1970), Blower (Reference Blower1975), Blouch & Atmosoedirdjo (Reference Blouch and Atmosoedirdjo1978) and Blouch (Reference Blouch1980) may indicate the existence of transient or survivor individuals rather than a stable population, possibly associated with increased habitat quality in some protected areas where routine patrol activities have reduced human disturbance and damage to vegetation.

The highest densities of Bawean deer were reported in secondary forests around Mount Dedawang, Mount Nangka, Mount Gadung, Mount Duren, Mount Mangoneng, Mount Bengkuang and Batulintang, a small area around Mount Bulu. The lowest densities were estimated in primary forests at Mount Besar, Mount Bajapati and Klumpang Gubuk. The population was centred around Mount Dedawang (cf. Blouch & Atmosoedirdjo, Reference Blouch, Atmosoedirdjo and Wemmer1987), with activity concentrated at low altitudes (< 300 m), where food and water are abundant.

Hunting activity was recorded at six of 20 camera trap locations (Rahman et al., Reference Rahman, Gonzalez and Aulagnier2016), and one snare was found in Batu Gebang block (the south-eastern part of Mount Dedawang), close to semi-open cultivated areas used by wild boar for foraging, and with the highest density of Bawean deer. Although wild boar were the poachers’ main target, deer could also be trapped, and die from stress-related causes (BBKSDA East Java, 2009). Furthermore, the Bawean Island Nature Reserve and Wildlife Sanctuary is close to human settlements, and some areas have been damaged by illegal logging. The continued presence of Bawean deer in harvested forests suggests some degree of tolerance to selective logging. The deer are attracted to settlements by agricultural crops (Semiadi, Reference Semiadi2004), which places them at risk from feral dogs (Blouch & Atmosoedirdjo, Reference Blouch and Atmosoedirdjo1978). We photographed feral dogs in 12 grid cells (Rahman et al., Reference Rahman, Gonzalez and Aulagnier2016) and recorded two cases of Bawean deer killed by feral dogs close to settlements. Although such events are rarer now than previously (Blouch & Atmosoedirdjo, Reference Blouch, Atmosoedirdjo and Wemmer1987), the threat should be taken seriously as feral dogs are the main predators of several deer species in South America (Weber & Gonzalez, Reference Weber and Gonzalez2003).

The Bawean deer has survived a decade of social turmoil, in which food scarcity triggered high levels of hunting and illegal logging (Semiadi, Reference Semiadi2004). Despite a stable population size, the species should retain its Critically Endangered status under criterion B1ab(ii, iii) and not C2a(ii) (which suggests a population decline) as in the most recent assessment (Semiadi et al., Reference Semiadi, Duckworth and Timmins2015). The extent of occurrence is < 100 km2, the area of occupancy is declining and the habitat is fragmented. Further study is needed, including a long-term monitoring scheme. A captive-breeding programme was established in 2006 with a founder population of two stags and five hinds, and this population had increased to 35 individuals by 2014 (Meijaard et al., Reference Meijaard, Leus and Burton2014).

Acknowledgements

We thank Balai Besar Konservasi Sumberdaya Alam of East Java and the Ministry of Environment and Forestry, Republic of Indonesia, for providing a licence to conduct research in Bawean Island Nature Reserve and Wildlife Sanctuary. Financial and logistical support was provided by the Rufford Small Grants Foundation. DAR received postgraduate scholarships and funding from the Directorate General of Higher Education, Ministry of Research, Technology and Higher Education, Republic of Indonesia.

Author contributions

DAR, GG and SA conceived and designed the study. DAR collected the field data and analysed the camera trap data. DAR, GG and SA wrote the manuscript.

References

Acevedo, P., Ferreres, J., Jaroso, R., Durán, M., Escudero, M.A., Marco, J. & Gortázar, C. (2010) Estimating roe deer abundance from pellet group counts in Spain: an assessment of methods suitable for Mediterranean woodlands. Ecological Indicators, 10, 12261230.
Akçakaya, H.R. (2002) Estimating the variance of survival rates and fecundities. Animal Conservation, 5, 333336.
Alves, J., da Silva, A.A., Soares, A.M.V.M. & Fonseca, C. (2013) Pellet group count methods to estimate red deer densities: precision, potential accuracy and efficiency. Mammalian Biology, 78, 134141.
Bakosurtanal (2002) Peta Rupa Bumi Indonesia. Http://bakosurtanal.go.id/bakosurtanal/peta-rbi/ [accessed 5 April 2016].
BBKSDA East Java (2009) Monitoring satwa flagship jenis rusa Bawean. Unpublished report. Balai Besar Konservasi Sumberdaya Alam Jawa Timur, Surabaya, Indonesia.
Blake, S. & Hedges, S. (2004) Sinking the flagship: the case of forest elephants in Asia and Africa. Conservation Biology, 18, 11911202.
Blouch, R.A. (1980) Kuhl's deer, Bawean Island, Indonesia. Tiger Paper, 7, 2223.
Blouch, R.A. & Atmosoedirdjo, S. (1978) Preliminary report on the status of the Bawean deer (Axis kuhli). In Threatened Deer: Proceedings of a Working Meeting of the Deer Specialist Group of the Survival Service Commission, pp. 4955. IUCN, Morges, Switzerland.
Blouch, R.A. & Atmosoedirdjo, S. (1987) Biology of the Bawean deer and prospects for its management. In Biology and Management of the Cervidae (ed. Wemmer, C.M.), pp. 320327. Smithsonian Institution Press, Washington, DC, USA.
Blower, J. (1975) Report on a Visit to Pulau Bawean. UNDP/FAO Nature Conservation and Wildlife Management Project, ISN/73/013. FAO, Bogor, Indonesia.
Carbajal-Borges, J.P., Godínez-Gómez, O. & Mendoza, E. (2014) Density, abundance and activity patterns of the endangered Tapirus bairdii in one of its last strongholds in southern Mexico. Tropical Conservation Science, 7, 100114.
Carbone, C., Christie, S., Conforti, K., Coulson, T., Franklin, N., Ginsberg, J.R. et al. (2001) The use of photographic rates to estimate densities of tigers and other cryptic mammals. Animal Conservation, 4, 7579.
CITES (2016) The CITES Appendices. Http://www.cites.org/eng/app/index.shtml [accessed 23 June 2016].
Cusack, J.J., Swanson, A., Coulson, T., Packer, C., Carbone, C., Dickman, A.J. et al. (2015) Applying a random encounter model to estimate lion density from camera traps in Serengeti National Park, Tanzania. The Journal of Wildlife Management, 79, 10141021.
Eberhardt, L. & Van Etten, R.C. (1956) Evaluation of the pellet group count as a deer census method. The Journal of Wildlife Management, 20, 7074.
Grubb, P. (2005) Artiodactyla. In Mammal Species of the World. A Taxonomic and Geographic Reference. 3rd edition (eds Wilson, D.E. & Reeder, D.M.), pp. 637722. The Johns Hopkins University Press, Baltimore, USA.
Hutchinson, J.M.C. & Waser, P.M. (2007) Use, misuse and extensions of ‘ideal gas’ models of animal encounter. Biological Reviews of the Cambridge Philosophical Society, 82, 335359.
IUCN (2001) IUCN Red List Categories and Criteria: Version 3.1. IUCN Species Survival Commission. IUCN, Gland, Switzerland, and Cambridge, UK.
Karanth, K.U. & Nichols, J.D. (1998) Estimation of tiger densities in India using photographic captures and recaptures. Ecology, 79, 28522862.
Kays, R., Kranstauber, B., Jansen, P.A., Carbone, C., Rowcliffe, M., Fountain, T. & Tilak, S. (2009) Camera traps as sensor networks for monitoring animal communities. IEEE 34th Conference on Local Computer Networks, 811818.
Kumbhar, A., Prabhu, C.L., Yogesh, J.K., Francies, P., Patwardhan, G., Qureshi, Q. & Jhala, Y. (2013) Application of photographic capture–recapture sampling for estimating abundance of Indian mouse deer Muschiola indica . World Journal of Zoology, 8, 388391.
Lachenmeier, K. & Melisch, R. (1996) Der Bawean Hirsch (Axis kuhli): Status und Schutz einer wenig bekannten Hirschart Indonesiens. Zoologische Gesellschaft für Arten und Populationsschutz E. V. Mitteilungen, 12, 2325.
LIPI & IPB. (1999) Laporan akhir riset unggulan terpadu IV: Sifat biologi rusa Bawean (Axis kuhlii) dan pengembangan model pembudidayaan untuk tujuan konservasi satwa langka endemic. Puslit Biologi LIPI Press, Bogor, Indonesia.
Meijaard, E., Leus, K. & Burton, J. (2014) Think Pig: results from our first wild pig specialist group meeting for South-east and South Asia. Suiform Soundings, Newsletter of the IUCN SSC Wild Pig, Peccary and Hippo Specialist Groups, 12.
Ministry of Environment and Forestry (2015) Keputusan Direktur Jenderal Konservasi Sumber Daya Alam dan Ekosistem No. SK. 180/IV-KKH Tahun 2015 tentang penetapan dua puluh lima satwa terancam punah prioritas untuk ditingkatkan populasinya sebesar 10% pada tahun 2015–2019. Sekretariat Kabinet RI, Jakarta, Indonesia.
Murray, J.V., Goldizen, A.W., O'Leary, R.A., McAlpine, C.A., Possingham, H.P. & Choy, S.L. (2009) How useful is expert opinion for predicting the distribution of a species within and beyond the region of expertise? A case study using brush-tailed rock-wallabies Petrogale penicillata . Journal of Applied Ecology, 46, 842851.
Oliveira-Santos, L.G.R., André, C.Z., Castro, P.A. & Crawshaw, P.G. Jr (2010) Is it possible to individually identify mammals with no natural markings using camera-traps? A controlled case study with lowland tapirs. Mammalian Biology, 75, 375378.
Pontes, A.R.M. & Chivers, D.J. (2007) Peccary movements as determinants of the movements of large cats in Brazilian Amazonia. Journal of Zoology, 273, 257265.
Prugh, L.R. & Krebs, C.J. (2004) Snowshoe hare pellet-decay rates and aging in different habitats. Wildlife Society Bulletin, 32, 386393.
Rahman, D.A., Gonzalez, G. & Aulagnier, S. (2016) Benefit of camera trapping for surveying the Critically Endangered Bawean deer Axis kuhlii (Temminck, 1836). Tropical Zoology, http://dx.doi.org/10.1080/03946975.2016.1199763.
Rovero, F. & Marshall, A.R. (2009) Camera trapping photographic rate as an index of density in forest ungulates. Journal of Applied Ecology, 46, 10111017.
Rovero, F., Zimmermann, F., Berzi, D. & Meek, P.D. (2013) ‘Which camera trap type and how many do I need?’ A review of camera features and study designs for a range of wildlife research applications. Hystrix, The Italian Journal of Mammalogy, 24, 148156.
Rowcliffe, J.M., Carbone, C., Jansen, P.A., Kays, R. & Kranstauber, B. (2011) Quantifying the sensitivity of camera traps: an adapted distance sampling approach. Methods in Ecology and Evolution, 2, 464476.
Rowcliffe, J.M., Carbone, C., Kays, R., Kranstauber, B. & Jansen, P.A. (2012) Bias in estimating animal travel distance: the effect of sampling frequency. Methods in Ecology and Evolution, 3, 653662.
Rowcliffe, J.M., Field, J., Turvey, S.T. & Carbone, C. (2008) Estimating animal density using camera traps without the need for individual recognition. Journal of Applied Ecology, 45, 12281236.
Rowcliffe, J.M., Kays, R., Kranstauber, B., Carbone, C. & Jansen, P.A. (2014) Quantifying levels of animal activity using camera trap data. Methods in Ecology and Evolution, 5, 11701179.
Seber, G.A.F. (1982) The Estimation of Animal Abundance and Related Parameters. Macmillan, New York, USA.
Semiadi, G. (2004) Sifat Biologi Rusa Bawean (Axis kuhlii). Puslit Biologi LIPI Press, Bogor, Indonesia.
Semiadi, G., Duckworth, J.W. & Timmins, R. (2015) Axis kuhlii. The IUCN Red List of Threatened Species 2015: e.T2447A73071875. Http://dx.doi.org/10.2305/IUCN.UK.2015-2.RLTS.T2447A73071875.en [accessed 8 August 2016].
Silver, S.C., Ostro, L.E.T., Marsh, L.K., Maffei, L., Noss, A.J., Kelly, M.J. et al. (2004) The use of camera traps for estimating jaguar Panthera onca abundance and density using capture/recapture analysis. Oryx, 38, 148154.
Sitwell, N. (1970) Bawean Island expedition. Animals, 12, 389393.
Soisalo, M.K. & Cavalcanti, S.M.C. (2006) Estimating the density of a jaguar population in the Brazilian Pantanal using camera-traps and capture–recapture sampling in combination with GPS radio-telemetry. Biological Conservation, 129, 487496.
Soria-Dίaz, L. & Monroy-Vilchis, O. (2015) Monitoring population density and activity pattern of white-tailed deer (Odocoileus virginianus) in Central Mexico, using camera trapping. Mammalia, 79, 4350.
St-Laurent, M-H. & Ferron, J. (2008) Testing the reliability of pellet counts as an estimator of small rodent relative abundance in mature boreal forest. Journal of Negative Results, 5, 1422.
Sudarmadji (1985) Laporan inventarisasi rusa Bawean (Axis kuhlii). Proyek Pembinaan Kelestarian Sumber Daya Alam Hayati, Jawa Timur, Indonesia.
Team Network (2011) Terrestrial Vertebrate Protocol Implementation Manual, v.3.1. Tropical Ecology, Assessment and Monitoring Network. Center for Applied Biodiversity Science, Conservation International, Arlington, USA.
Tobler, M.W., Carrillo-Percastegui, S.E., Leite-Pitman, R., Mares, R. & Powell, G. (2008) An evaluation of camera traps for inventorying large- and medium-sized terrestrial rainforest mammals. Animal Conservation, 11, 169178.
Trolle, M., Noss, A.J., Passos, J.L. & Oliveira, L.F.B. (2008) Brazilian tapir density in the Pantanal: a comparison of systematic camera-trapping and line-transect surveys. Biotropica, 40, 211217.
UGM & BBKSDA East Java (2003) Studi Populasi Rusa Bawean. Fakultas Kehutanan UGM Press, Yogyakarta, Indonesia.
UGM & BBKSDA East Java (2004) Studi Ekologi Rusa Bawean. Fakultas Kehutanan UGM Press, Yogyakarta, Indonesia.
Weber, M. & Gonzalez, S. (2003) Latin American deer diversity and conservation: a review of status and distribution. Ecoscience, 10, 443454.
Zero, V.H., Sundaresan, S.R., O'Brien, T.G. & Kinnaird, M.F. (2013) Monitoring an Endangered savannah ungulate, Grevy's zebra Equus grevyi: choosing a method for estimating population densities. Oryx, 47, 410419.

Biographical sketches

Dede Aulia Rahman specializes in the ecology of deer species in tropical rainforest, and is working to develop the use of camera trapping and remote sensing in ungulate conservation projects in Indonesia. Georges Gonzalez studies the behaviour, management and conservation of ungulates in Western Europe, particularly in protected or managed areas. Stéphane Aulagnier's main area of interest is the evolutionary biology and conservation of Palaearctic mammals.