Initiating conservation of a newly discovered population of the Endangered hog deer Axis porcinus in Myanmar
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Initiating conservation of a newly discovered
population of the Endangered hog deer
Axis porcinus in Myanmar
NGWE LWIN, MATTHEW LINKIE, ABISHEK HARIHAR, SAW SOE AUNG
A U N G K O L I N and F R A N K M O M B E R G
Abstract The unprecedented political and economic re- Introduction
forms taking place in Myanmar offer new opportunities
for biodiversity conservation. They also bring new chal-
lenges in the form of rapidly growing extractive industry F or decades much of Myanmar has been inaccessible to
conservation organizations because of political instabil-
ity and civil unrest. International sanctions have limited in-
and agriculture sectors that have been weakly regulated
and are often unsustainable. The Endangered hog deer vestment in government infrastructure and institutional
Axis porcinus epitomizes many of these conservation chal- development, including the Forestry Department, which is
lenges, and those facing most deer species in the Indo– the country’s leading agency for biodiversity conservation
Burma hotspot. The hog deer has disappeared from large (Rao et al., ). However, since Myanmar has under-
parts of its range as a result of overhunting and intense con- taken sweeping political and economic reforms aimed at
version of its floodplain grassland habitat for agriculture. giving foreign organizations access to the country, creating
We report on a population of hog deer that was discovered political stability and decreasing government control of the
in the Indawgyi landscape in central Myanmar in . We private sector (Schmidt, ). This comes at a time when
conducted the first rigorous assessment of a hog deer popu- significant attention is also being focused on Myanmar for
lation in Myanmar using an occupancy sampling protocol, its role as a major supply hub for illegal wildlife trade in Asia
tested the protocol’s robustness using a power analysis, and (Shepherd & Nijman, ; Nijman & Shepherd, ).
present the results to guide management intervention. The Myanmar remains rich in wildlife and other natural re-
results from our study site revealed widespread occurrence sources but has long and porous international borders that
of the species, with high precision. The population map was are poorly enforced or equipped to tackle illegal wildlife
then used to inform the development of a conservation trade. Furthermore, overexploitation by poorly regulated
management zone within a UNESCO Man and Biosphere mining and logging industries, subsistence hunting, and con-
Reserve around Indawgyi Lake. The importance of this version of floodplains and grasslands to rice fields have de-
population for the status of the hog deer in Myanmar re- pressed already stressed wildlife populations (Rao et al.,
mains unknown because documentation of the species has ). These threats have had a collective, although unquanti-
been sparse. Our survey protocol could make a significant fied, impact on the hog deer, which in many ways epitomizes
contribution to addressing this knowledge gap and setting the conservation challenges facing wildlife in Myanmar.
an informed agenda for conservation of the hog deer both The hog deer’s habitat preference for grassland in close
nationally and more widely across the Indo-Burma hotspot. proximity to wetlands often puts it in competition with
farmers wishing to expand rice fields (Dhungel & O’Gara,
Keywords Detection probability, hunting, mammal conser- ). It is also hunted for its meat. Over the past decades
vation, occupancy, South-east Asia, ungulate a once widespread species ranging from the Himalayan
foothills in Pakistan, Nepal and India to mainland
South-east Asia has declined by . % within its
South-east Asian range. This includes its complete dis-
appearance from Thailand, Laos PDR and Viet Nam
(Timmins et al., a). In Myanmar this once abundant
NGWE LWIN, SAW SOE AUNG, AUNG KO LIN and FRANK MOMBERG Fauna & Flora species has been reduced to small, isolated populations
International, Yangon, Myanmar
(Evans, ; Timmins et al., a) and there is a lack of re-
MATTHEW LINKIE* (Corresponding author) Fauna & Flora International, liable, or even current, population data for the hog deer, as
Singapore 247672, Singapore
for most other cervids and medium- or large-bodied mam-
ABISHEK HARIHAR Panthera, New York, USA, and Nature Conservation
Foundation, Mysore, India
mals in general. Thus, the news of the discovery of a hog
deer population in Myanmar by Fauna & Flora
*Current address: Wildlife Conservation Society (WCS), Indonesia Program,
Bogor, Indonesia. Email mlinkie@wcs.org International in was encouraging.
Received February . Revision requested May .
Preliminary field surveys in northern Myanmar recorded
Accepted June . First published online October . the hog deer, using camera traps placed in grassland on the
Oryx, 2018, 52(1), 126–133 © 2016 Fauna & Flora International doi:10.1017/S0030605316000727
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 21 Sep 2021 at 03:04:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605316000727Hog deer conservation in Myanmar 127
eastern shore of Indawgyi Lake, the third largest freshwater grasslands and paddy fields). Hog deer sign comprised hoof
lake in mainland South-east Asia. Based on these records a prints and pellets, which are unmistakable as no other similar
conservation project was launched to produce a site action deer or livestock species are known to live in this area. Red
plan, form community conservation groups to define a hog muntjac Muntiacus vaginalis and sambar Rusa unicolor live
deer protection zone with village no-hunting agreements, outside the study area, in a nearby forest reserve.
and increase awareness amongst stakeholders of the species’ Within a grid cell, detection () or non-detection () of
plight. The first important step was to collect reliable data to hog deer sign was recorded at four -minute intervals
gain an understanding of the hog deer’s population status (sampling occasions). These replicates are considered to
and threats to the species in the Indawgyi area, which be spatial because the team tried to maintain a similar
would also provide a baseline for monitoring subsequent pace (, km per hour) in each cell and avoid overlapping
project intervention. Thus, the aims of this study were to with previously surveyed replicates, by not walking over a
() design and pilot a new hog deer sampling protocol previously walked trail within the same grid cell. For each
using detection/non-detection surveys, () conduct the replicate the team recorded whether surveys were conducted
first scientifically defensible population assessment of the off trail () or on footpaths or bullock cart paths (), as well
hog deer in Myanmar based on the species’ occupancy, as the dominant habitat type (tall grassland (elephant grass);
and () determine future sampling design parameters. grassland; burnt grassland; agricultural field, not planted; or
agriculture field, planted), all of which could influence spe-
cies detection probability. It did not rain during surveys, so
Methods the influence of this factor was not considered. Direct obser-
vations of hog deer hunting and recent burning were re-
Field methods corded seperately for each grid cell.
From preliminary field surveys conducted in and
in the northern state of Kachin the putative boundaries of Data analysis
hog deer distribution in the Indawgyi landscape were iden-
tified based on suitable habitat, and used to demarcate a A spatial database was constructed for the sampling grid
study area of km (Fig. ). This study area is characterized based on three site covariates: mean proximity to nearest vil-
by grassland located between the forested watershed hills in lage (Village), mean proximity to river or lake (Water), and
Indawgyi Reserve Forest and Indawgyi Lake. Several small evidence of recent burning (Burn; denoted as burnt () or
creeks flow through the grassland plain, and there are not burnt ()). Village and river data obtained from
patches of paddy fields in these communal lands. Myanmar Survey Department : . topographic maps
In Myanmar the hog deer falls under the protection of from were standardized using a logarithmic transform-
wildlife and protected areas law /. The preliminary ation. Sampling covariates were compiled along the surveys
field surveys, camera trapping and interviews with the to denote trail/no trail and the dominant habitat type in
local community indicated that the hog deer is solitary, each sampling replicate, which were constructed as categor-
and no herds were recorded. However, in Nepal hog deer ical covariates for both.
have been observed congregating during the fawning season Hog deer occupancy (ψ) was estimated using models that
or to forage on new shoot growth stimulated by anthropo- account for imperfect species detection (MacKenzie et al.,
genic burning of grasslands (Dhungel & O’Gara, ). An ). The occupancy status of grid cells was assumed to re-
individual’s home range is estimated to be .–. km main closed (i.e. constant) during the -hour sampling per-
(Dhungel & O’Gara, ; Odden et al., ). To measure iod. As the detection/non-detection sequence of hog deer
species occupancy the study area was divided into sam- sign was collected during four consecutive -minute sam-
pling units of × km, which should be sufficient to encom- pling occasions per grid cell, a first order Markov process
pass an individual’s home range. model was used to estimate the probability of hog deer oc-
From a possible selection of grid cells and considering currence (Hines et al., ). This model accounts explicitly
available field resources, cells were selected at random for for the potential dependence in detecting signs on consecu-
sampling, using a geographical information system (GIS). tive occasions by estimating two segment-level occupancy
Field surveys were conducted during – May , in the parameters (θ and θ’) and the probability of detecting
dry season, when it is easier to encounter hog deer sign. A signs (P) conditional on segment-level occupancy.
two-person team, consisting of an experienced field biologist However, to test for the presence of spatial non-
and a local guide, surveyed each grid cell for a set sampling independence among replicates the constant model ψ(.),θ
effort of hours per cell. To increase detection probability (.),θ’(.),P(.),θπ(.) (Hines et al., ) was compared to a con-
within a cell, the team specifically targeted locations where stant single-season model, ψ(.),P(.) (MacKenzie et al., ).
sign was most likely to be encountered (e.g. animal trails, Candidate models were compared using Akaike’s
Oryx, 2018, 52(1), 126–133 © 2016 Fauna & Flora International doi:10.1017/S0030605316000727
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 21 Sep 2021 at 03:04:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605316000727128 N. Lwin et al.
FIG. 1 Habitat of a newly discovered population of hog deer Axis porcinus in the Indawgyi landscape of northern Myanmar, with the
survey grid and the locations where hog deer were recorded.
information criterion corrected for small sample sizes either assumed to be constant or allowed to vary with indi-
(AICc; Burnham & Anderson, ). All analyses were car- vidual or additively combined covariates, was modelled, with
ried out in PRESENCE v. . (Hines, ). ψ held in a general form for each model. After incorporating
When choosing between a model that explicitly ac- covariates influencing detection probability with the greatest
counted for spatial dependence (Hines et al., ) and Akaike weight (wi), the influence of spatial covariates on oc-
one that did not (MacKenzie et al., ), a two-step model- cupancy was assessed in the second step of the analysis. Here
ling approach was used to test the effects of the covariates. ψ was either assumed to be constant or allowed to vary with
Initially, the probability of detecting a sign, where P was individual or additively combined covariates.
Oryx, 2018, 52(1), 126–133 © 2016 Fauna & Flora International doi:10.1017/S0030605316000727
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 21 Sep 2021 at 03:04:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605316000727Hog deer conservation in Myanmar 129
TABLE 1 Effect of covariates on detection probability (P̂) of the hog deer Axis porcinus in the Indawgyi landscape of northern Myanmar
(Fig. ).
Model ID Model description No. of parameters ΔAICc1 Akaike weight −2L2
M.1.1 ψ(Burn + Water + Village)P(Habitat) 9 0.00 0.7377 180.02
M.1.2 ψ(Burn + Water + Village)P(Habitat + Burn) 10 3.00 0.1644 179.65
M.1.3 ψ(Burn + Water + Village)P(Habitat + Burn + Trail) 11 4.66 0.0719 177.72
M.1.4 ψ(Burn + Water + Village)P(Habitat + Trail) 10 6.70 0.0258 183.35
M.1.5 ψ(Burn + Water + Village)P(Burn + Trail) 7 18.32 0.0001 204.52
M.1.6 ψ(Burn + Water + Village)P(Burn) 6 19.28 0 208.32
M.1.7 ψ(Burn + Water + Village)P(Trail) 6 20.15 0 209.19
M.1.8 ψ(Burn + Water + Village)P(.) 5 20.47 0 212.21
M.1.9 ψ(.)P(.) 2 24.71 0 223.73
M.1.10 ψ(.)θ0(.)θ’(.)P(.)θ0π(.) 5 27.54 0 221.28
ΔAICc is the difference between each model’s AICc and the AICc of the top-ranking model.
L is the maximum log-likelihood.
Power analysis
Survey design recommendations were derived based on the
ψ and P estimates obtained from the field study. Where de-
tection probabilities varied depending on survey conditions,
the mean of these probabilities was used in calculations.
Firstly, the number of survey visits (K) required to deter-
mine species presence at an occupied site with a specified
probability was investigated. The probability of detecting
the species at an occupied site in at least one of the visits
is P* = – ( – P)K, where P is the mean detection probabil-
ity on any one visit. From this expression the number of vis-
its (K) required to achieve a given desired P* (.) was
derived as K = log( – P*)/log( – P) (McArdle, ; Kéry,
; Pellet & Schmidt, ).
Next, variations in the survey design that facilitated the FIG. 2 Mean probability of detection (± SE) of hog deer in
detection of changes in hog deer occupancy were assessed. various habitat types.
The assessments were based on the occupancy and detection
probability estimates from the study data set and assumed a
deer signs were detected on of sampling occasions
standard single-season, single-species sampling design in
and in of sampling units, producing a naïve occupancy
which S sites were sampled K repeated times (where K
estimate of ..
was determined to achieve P* = .). Based on the number
of replicates (in our case, the number of successive mi-
nute surveys), the number of sampling sites needed to Detecting hog deer signs Investigating the covariates that
detect a change in occupancy rate between two surveys influenced the probability of detecting hog deer signs we
for a given power given an actual proportional change, R found little support for spatial dependency in detection
(effect size), was calculated. Six effect sizes (R = , , , resulting from a study design based on four consecutive
, and %) at a significance level α = . were assessed -minute replicates. The Markovian detection model
under the constraint that the total number of sites (S) that (M..) had the lowest ranking in comparison to a
could be surveyed was (i.e. × km grid cells identified constant single-season model (M..; Table ). Further
as being within the putative boundaries of hog deer distribu- examination of the two segment-level occupancy parameters
tion in the Indawgyi landscape). All simulations were imple- (θ and θ’) highlighted non-significant dependency (θ = .,
mented using R code developed by Guillera-Arroita & % CI .–.; θ’ = ., % CI .–.). Hog deer
Lahoz-Monfort (). detection was strongly correlated with habitat type (M.) and
there was a large difference between this model and the
Results constant model (ΔAICc = .). Across the model set,
detection of signs was most strongly influenced by habitat
The sampling units were surveyed for a combined effort (AICc wi = .) in comparison to burning (AICc wi = .)
of . hours, with – grid cells completed each day. Hog or trail type (AICc wi = .). Among the habitat types, the
Oryx, 2018, 52(1), 126–133 © 2016 Fauna & Flora International doi:10.1017/S0030605316000727
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 21 Sep 2021 at 03:04:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605316000727130 N. Lwin et al.
TABLE 2 Hog deer occupancy (ĉ ) estimates obtained from the best model for the Indawgyi landscape of northern Myanmar (Fig. ).
Model ID Model description No. of parameters ΔAICc1 Akaike weight −2L2
M.2.1 ψ(Water)P(Habitat) 7 0.00 0.3017 181.66
M.2.2 ψ(.)P(Habitat) 6 0.50 0.2351 185.00
M.2.3 ψ(Water + Burn)P(Habitat) 8 1.53 0.1402 180.19
M.2.4 ψ(Burn)P(Habitat) 7 1.84 0.1202 183.50
M.2.5 ψ(Village)P(Habitat) 8 2.18 0.1014 183.84
M.2.6 ψ(Water + Village)P(Habitat) 7 2.93 0.0696 181.59
M.2.7 ψ(Burn + Water + Village)P(Habitat) 9 4.54 0.0311 180.02
M.2.8 ψ(Village + Burn)P(Habitat) 8 12.25 0.0007 190.91
ΔAICc is the difference between each model’s AICc and the AICc of the top-ranking model.
L is the maximum log-likelihood.
FIG. 3 Estimated relationship between predicted hog deer FIG. 4 Number of sampling sites needed to detect a change in
occupancy and distance to water. The vertical lines represent occupancy rate with varying effect size at a significance level of
% confidence intervals. α = ..
probability of detecting hog deer sign (given presence) was % confidence was five. This corresponds to a survey
highest in burnt grassland (Fig. ). effort of . hours per grid cell, rather than the -hour
field surveys conducted. Based on the model-averaged
Hog deer occupancy No sign of hunting of hog deer using occupancy (ĉ = . ± SE .) and detection probability
firearms was encountered during field surveys. The (P̂ = . ± SE .) estimated from the field data, the
covariate analysis revealed that species occupancy was number of sites that would be required in a future survey
higher in habitat closer to water (b̂ water = −. ± SE .; design to detect a change in occupancy varied considerably
M.; Table , Fig. ). There was also an effect from human under different scenarios (Fig. ). These results revealed the
activities, with a higher hog deer occupancy in recently burnt study area supporting a maximum of sites and surveyed
sites (b̂ burnt = . ± SE .) and sites further from villages over five occasions would not be able to detect changes in
(b̂ village = . ± SE .). The summed model weights for occupancy of , or % with a power of .. At least ,
each factor with respect to hog deer occupancy were as and sites would need to be sampled to detect changes
follows: distance to water (.%), recent burning (.%) in occupancy of , or %, respectively, with a power
and distance to village (.%). The model averaged of . ..
occupancy ĉ was . ± SE ..
Discussion
Power analysis According to the estimated mean
detection probability (P̂ = . ± SE .) the number of Several deer species have undergone an overall decline
replicates (-minute survey intervals) required to across their mainland South-east Asian range, most notably
determine hog deer presence at an occupied site with sambar, hog deer and Eld’s deer Rucervus eldii (Gray et al.,
Oryx, 2018, 52(1), 126–133 © 2016 Fauna & Flora International doi:10.1017/S0030605316000727
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 21 Sep 2021 at 03:04:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605316000727Hog deer conservation in Myanmar 131
; Timmins et al., a,b). Yet our findings indicate that Hog deer occupancy
the hog deer in our Myanmar study area have not followed
this trend. The sampling protocol developed and its success- The widespread occurrence of the hog deer in our study area
ful implementation proved to be robust. More importantly, raises several points for consideration. Firstly, without a
our study shed light on a previously little known hog deer historical baseline from the wider Indawgyi landscape we
population in Myanmar by revealing a high occupancy cannot fully rule out that the current population is in fact
across its putative habitat, and this information was used a remant of one that once extended across a larger range,
to influence conservation planning directly. This is welcome and is therefore declining. We recommend expanding fu-
news for a species that is under immense pressure. Security ture surveys beyond the putative range, which was identified
concerns notwithstanding, we recommend using this rapid through preliminary field surveys, including interviews with
survey technique across the wider Indawgyi landscape and the local community. Not only is this predicted to improve
mapping the status of the hog deer elsewhere in Myanmar, statistical rigour in measuring population trends, it would
especially in areas of tall grassland and riparian forest that also reduce the likelihood of individuals being missed and
may have been overlooked previously. excluded from subsequent conservation planning. There is
a seemingly suitable grassland habitat by the west side of
Indawgyi Lake but it was not possible to survey there be-
cause of security concerns.
Sampling protocol In the Indawgyi landscape hog deer occupancy was high-
In tropical Asia the applicability of occupancy survey tech- er in habitat that was closer to water. This finding fits with
niques for conducting range-wide assessments has been de- the species’ known preference for tall grassland and riparian
monstrated for entire species (Asian tapir Tapirus indicus), forest, which is associated with the floodplains of large rivers
subspecies (Sumatran tiger Panthera tigris sumatrae) and and lakes (Odden et al., ). Monsoonal flooding, as well
populations (sun bear Helarctos malayanus, which was ca- as anthropogenic burning of grass (for which we found a
tegorized as Data Deficient at the time; Linkie et al., , positive and significant correlation with hog deer occu-
; Wibisono et al., ). Our application of this technique pancy), would stimulate the growth of young and edible
to the hog deer yielded precise estimates of occupancy and shoots, thereby increasing food availability for hog deer, as
detection probability. Furthermore, the sampling protocol found for the chital Axis axis (Moe & Wegge, ).
required minimal training for a team with no prior experi- Although the results suggest that burning was not detrimen-
ence of conducting surveys within a capture–recapture tal to the hog deer (e.g. through reducing hiding cover), the
framework, although training in identification of species species did exhibit a tendency to avoid villages. This may
sign, especially where sympatric species occur, may be chal- have been to evade a threat not quantified in this study,
lenging for an inexperienced team. The baseline results are such as areas of higher human activity in Pakistan that
promising, and the power analysis identified where the were completely avoided by hog deer (Arshad et al., ).
protocol should be refined.
A simple modification to increase confidence in confirm- Threats to the hog deer
ing the presence of hog deer would be to increase the survey
effort from to . hours per grid cell. A more challenging Hog deer may have thrived in the Indawgyi landscape be-
modification would be to increase the number of survey cause of lower levels of threats that have been identified as
sites, to detect a significant change in population trend. being important elsewhere, primarily the conversion of
Based on the current sampling framework and its results, floodplain grassland to rice paddy, attacks by domestic
a change in occupancy would only be detected if it were at dogs, the use of poisons to protect crops from wild herbi-
least ± %. To improve the detection level would require vores, and hunting for wild meat using snares and guns.
increasing the number of sampling sites beyond the current Other threats were absent; for example, competition from
range maximum of cells. This finding has important im- livestock grazing, competition with other wildlife (e.g. ele-
plications for similar studies in tropical landscapes where phant, sambar, gaur Bos gaurus) for grazing, and the pres-
the extent of habitat left to monitor is small, especially re- ence of large felid predators, such as the tiger and leopard
garding the need to understand the occupancy dynamics Panthera pardus (Bhowmik, ). Furthermore, informal
of colonization and extinction within grid cells. As a caveat, interviews with two local hunters revealed that home-made
grid cell size may need to vary based on varying habitat guns were typically used in the past to shoot hog deer, and
quality across the hog deer’s range to ensure that occupancy also spears or knives, particularly at night in paddy fields
is being measured, and where hog deer live in extensively during the rainy season, when it is easier to creep up
flooded landscapes rapid surveys should be conducted in close. However, these interviews also revealed that hog
both wet and dry seasons to avoid possible sampling biases deer hunting was considered to be minimal both in the
associated with seasonal movements. past and at present. Nonetheless, during a routine project
Oryx, 2018, 52(1), 126–133 © 2016 Fauna & Flora International doi:10.1017/S0030605316000727
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 21 Sep 2021 at 03:04:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605316000727132 N. Lwin et al.
site visit in the survey team found a hunting platform in significant habitat conversion). However, our sampling proto-
the grassland, and the occupancy team would have found it col could be extended to include other little-known deer spe-
difficult to detect signs of hunting with guns during surveys. cies, such as sambar and Eld’s deer, in South-east Asia.
It is difficult to quantify the threat posed by hunting, and we
recommend a fuller assessment; however, the widepsread
occupancy of hog deer recorded suggests that the level of Acknowledgements
threat is not high in our study area.
The absence of a covariate for the presence of dogs, as a We thank Patrick Oswald for GIS support, Sarah Brook for
proxy for hunting, may be a limitation in our study. We used sharing information on the hog deer, the Myanmar Forestry
distance to village as a proxy for human disturbance (a com- Department for survey permission, and the Conservation
monly used technique in hunting and deforestation studies Leadership Programme and the Helmsley Charitable Fund
in the tropics) and this should have partly controlled for the for funding the work.
presence of dogs, if indeed they had an effect. Thus, whereas
we cannot directly disprove the relationship between hog
deer and the presence of dogs, the widespread hog deer oc- Author contributions
cupancy (% of the study area) suggests that hunting was ML, NL and FM conceived and designed the study. NL, SSA
not influential. If dogs were critically important and hunting and AKL conducted the fieldwork. AH analysed the data.
was high then the widespread occupancy pattern we re- ML, NL, AH and FM wrote the article.
corded for this threatened species would not have been
found. The major threat to the hog deer population is
most likely to be the conversion of grasslands for agricultur- References
al use, primarily on the banks of the Indaw Chaung river
and in the mosaic of grasslands and agricultural fields be- A R S H A D , M., U L L A H , I., C H A U D H R Y , M.J.I. & K H A N , N.U.H. ()
tween the river and Mokso Reserved Forest. As part of the Estimating hog deer Axis porcinus population in the riverine forest
of Taunsa Barrage Wildlife Sanctuary, Punjab, Pakistan. Records:
ongoing management planning for Indawgyi wildlife sanc-
Zoological Survey of Pakistan, , –.
tuary, as well as through the designation and zonation of the B H O W M I K , M.K. () The causes of decline of hog deer (Axis
Indawgyi biosphere reserve, efforts to limit new conversion, porcinus) in protected areas of Himalayan West Bengal. Zoos’ Print
especially between Indaw Chaung and Mokso Reserved Journal, , –.
Forest, will be prioritized through core conservation zones B U R N H A M , K.P. & A N D E R S O N , D.R. () Model Selection and
Multimodel Inference: A Practical Information-Theoretic Approach.
and/or restricted buffer zones for livestock grazing only.
Second edition. Springer-Verlag, New York, USA.
These would then be demarcated and managed through col- D H U N G E L , S.K. & O’ G A R A , B.W. () Ecology of the hog deer in
laborative patrolling and law enforcement. Fauna & Flora Royal Chitwan National Park, Nepal. Wildlife Monographs, ,
International is working with the Nature Conservation –.
Department on the zonation of a wildlife sanctuary, in E VA N S , G.H. () Notes on the hog deer in Burma. Journal of the
part based on the results of the hog deer survey, which fore- Bombay Natural History Society, , –.
G R A Y , T.N.E., B R O O K , S.M., M C S H E A , W.J., M A H O O D , S.,
sees the designation of core and buffer zones. R A N J I T S I N G H , M.K., M I Y U N T , A. et al. () Rucervus eldii. The
In Myanmar the terrestrial protected area system has IUCN Red List of Threatened Species : e.TA.
been designed primarily to support the conservation man- Http://dx.doi.org/./IUCN.UK.-.RLTS.TA.
agement of the country’s biodiversity-rich rainforests, al- en [accessed May ].
though many of the protected areas have yet to be G U I L L E R A -A R R O I T A , G. & L A H O Z -M O N F O R T , J.J. () Designing
studies to detect differences in species occupancy: power analysis
officially gazetted (Myint Aung, ; Rao et al., ). under imperfect detection. Methods in Ecology and Evolution, ,
The government’s lack of awareness of the plight of grass- –.
land ecosystems coupled with demands to clear these spaces H I N E S () PRESENCE: software to estimate patch occupancy and
for agriculture means that little actual or presumed hog deer related parameters. USGS Patuxent Wildlife Research Center.
habitat is formally protected. As noted in the and Http://www.mbr-pwrc.usgs.gov/software/presence.html [accessed
September ].
IUCN Red List assessments, reliable information on the
H I N E S , J.E., N I C H O L S , J.D., R O Y L E , J.A., M AC K E N Z I E , D.I.,
range and population status of the hog deer in Myanmar G O P A L A S WA M Y , A.M., K U M A R , N.S. & K A R A N T H , K.U. ()
is lacking and prevents the formulation of a national species Tigers on trails: occupancy modeling for cluster sampling. Ecological
strategy (Timmins et al., a). The rapid and robust occu- Applications, , –.
pancy survey protocol applied in our study provides the ap- K É R Y , M. () Inferring the absence of a species: a case study of
snakes. The Journal of Wildlife Management, , –.
proach for conducting such a widespread national survey,
L I N K I E , M., D I N ATA , Y., N U G R O H O , A. & A C H M A D H A I D I R , I. ()
such as in Hukaung Valley (one of the largest floodplains in Estimating occupancy of a Data Deficient mammalian species living
tropical Asia) and Pidaung Wildlife Sanctuary (once consid- in tropical rainforests: sun bears in the Kerinci Seblat region,
ered important for hog deer but which has undergone Sumatra. Biological Conservation, , –.
Oryx, 2018, 52(1), 126–133 © 2016 Fauna & Flora International doi:10.1017/S0030605316000727
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 21 Sep 2021 at 03:04:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605316000727Hog deer conservation in Myanmar 133
L I N K I E , M., G U I L L E R A -A R R O I T A , G., S M I T H , J., A R I O , A., S C H M I D T , C. () As isolation ends, Myanmar faces new ecological
B E R TA G N O L I O , G., C H E O N G , F. et al. () Cryptic mammals risks. Science, , –.
caught on camera: assessing the utility of range wide camera trap S H E P H E R D , C.R. & N I J M A N , V. () The trade in bear parts from
data for conserving the endangered Asian tapir. Biological Myanmar: an illustration of the ineffectiveness of enforcement of
Conservation, , –. international wildlife trade regulations. Biodiversity and
M AC K E N Z I E , D.I., N I C H O L S , J.D., L AC H M A N , G.B. D R O E G E , S., R O Y L E , Conservation, , –.
J.A. & L A N G T I M M , C.A. () Estimating site occupancy rates T I M M I N S , R., D U C K WO R T H , J.W., S A M B A K U M A R , N., A N WA R U L
when detection probabilities are less than one. Ecology, , – I S L A M , M., S A G A R B A R A L , H., L O N G , B. & M A X W E L L , A. (a)
. Axis porcinus. The IUCN Red List of Threatened Species : e.
M AC K E N Z I E , D.I., N I C H O L S , J.D., R O Y L E , J.A., P O L LO C K , K.H., TA. Http://dx.doi.org/./IUCN.UK.-.
B A I L E Y , L.L. & H I N E S , J.E. () Occupancy Estimation and RLTS.TA.en [accessed May ].
Modeling: Inferring Patterns and Dynamics of Species Occurrence. T I M M I N S , R., K AWA N I S H I , K., G I M A N , B., L Y N A M , A., C H A N , B.,
Academic Press, New York, USA. S T E I N M E T Z , R. et al. (b) Rusa unicolor. The IUCN Red List of
M C A R D L E , B.H. () When are rare species not there? Oikos, , Threatened Species : e.TA [accessed May
–. ].
M O E , S.R. & W E G G E , P. () The effects of cutting and burning on W I B I S O N O , H.T., L I N K I E , M., G U I L L E R A -A R R O I TA , G., S M I T H , J.A.,
grass quality and axis deer (Axis axis) use of grassland in lowland S U N A R T O , P U S P A R I N I , W. et al. () Population status of a cryptic
Nepal. Journal of Tropical Ecology, , –. top predator: an island-wide assessment of tigers in Sumatran
M Y I N T A U N G , U. () Policy and practice in Myanmar’s rainforests. PLoS ONE, (), e.
protected area system. Journal of Environmental Management, ,
–.
N I J M A N , V. & S H E P H E R D , C.R. () Trade in tigers and other wild Biographical sketches
cats in Mong La and Tachilek, Myanmar—A tale of two border
towns. Biological Conservation, , –. N G W E L W I N has been working on conservation projects in Myanmar
O D D E N , M., W E G G E , P. & S T O R A A S , T. () Hog deer Axis porcinus for years. M A T T H E W L I N K I E has worked in Asia, mainly on tiger
need threatened tallgrass floodplains: a study of habitat selection in conservation, since . He is the Asia Coordinator for the IUCN
lowland Nepal. Animal Conservation, , –. Species Survival Commission Wild Pig Specialist Group. A B I S H E K
P E L L E T , J. & S C H M I D T , B.R. () Monitoring distributions using call H A R I H A R has worked on tiger conservation in northern India since
surveys: estimating site occupancy, detection probabilities and and currently works as a tiger population ecologist. S A W
inferring absence. Biological Conservation, , –. S O E A U N G coordinates biodiversity monitoring in Myanmar’s
R A O , M., H T U N , S., P L AT T , S.G., T I Z A R D , R., P O O L E , C., M Y I N T , T. & Tanintharyi landscape. A U N G K O L I N conducts biodiversity surveys
W AT S O N , J.E.M. () Biodiversity conservation in a changing in the Indawgyi and Imawbum areas of northern Myanmar. F R A N K
climate: a review of threats and implications for conservation M O M B E R G has worked on the development of conservation projects
planning in Myanmar. Ambio, , –. in Asia since .
Oryx, 2018, 52(1), 126–133 © 2016 Fauna & Flora International doi:10.1017/S0030605316000727
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 21 Sep 2021 at 03:04:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605316000727You can also read
