The use of infrared-triggered cameras for surveying
phasianids in Sichuan Province, China
SHENG LI,
1
* WILLIAM J. MCSHEA,
2
DAJUN WANG,
1
LIANGKUN SHAO
3
&XIAOGANGSHI
4
1
Center for Nature and Society, College of Life Sciences, Peking University, Beijing, 100871, China
2
Conservation and Research Center, National Zoological Park, Front Royal, VA 22630, USA
3
Wanglang National Nature Reserve, Pingwu County, Sichuan Province, 622550, China
4
Wolong National Nature Reserve, Sichuan Province, 623006, China
We report on the use of infrared-triggered cameras as an effective tool to survey phasia-
nid populations in Wanglang and Wolong Nature Reserves, China. Surveys at 183
camera-trapping sites recorded 30 bird species, including nine phasianids (one grouse and
eight pheasant species). Blood Pheasant Ithaginis cruentus and Temminck?s Tragopan
Tragopan temminckii were the phasianids most often detected at both reserves and were
found within the mid-elevation range (2400?3600 m asl). The occupancy rate and detec-
tion probability of both species were examined using an occupancy model relative to
eight sampling covariates and three detection covariates. The model estimates of
occupancy for Blood Pheasant (0.30) and Temminck?s Tragopan (0.14) are close to the
na?ve estimates based on camera detections (0.27 and 0.13, respectively). The estimated
detection probability during a 5-day period was 0.36 for Blood Pheasant and 0.30 for
Temminck?s Tragopan. The daily activity patterns for these two species were assessed
from the time?date stamps on the photographs and sex ratios calculated for Blood
Pheasant (152M : 72F) and Temminck?s Tragopan (48M : 21F). Infrared cameras are
valuable for surveying these reclusive species and our protocol is applicable to research
or monitoring of phasianids.
Keywords: adult sex ratio, daily activity pattern, detection probability, occupancy model,
temperate forest.
Of the 179 species of phasianid (Gill & Wright
2006), 63 occur in China (Zheng 2000, Zheng
2005). Although these large and mainly terrestrial
birds have been closely associated with humans for
centuries in China, they are currently suffering
increasing threats from habitat loss, hunting,
human disturbance and hybridization with released
stock. Therefore, research, monitoring and conser-
vation of phasianids in China is becoming increas-
ingly important to the global conservation efforts
focused on this group (Fuller & Garson 2000,
Zhang et al. 2003).
Information on the basic ecology and distribu-
tion of most phasianids in China is poor (Fuller
& Garson 2000). Most species in southwestern
China are located within heavily forested habitats
(Li 1996, Zhang 1999, Zheng 2005) or open
habitats at higher elevations, such as sub-alpine
rhododendron scrub, alpine meadow and grass-
land (Long et al. 1998, BirdLife International
2001). Complex terrain, steep topography and
dense vegetation impede field research and moni-
toring activities. Traditional field survey methods
based on direct observation, such as transect
counts and behavioural observations, are difficult
to undertake due to the inaccessibility of remote
areas, lack of visibility in dense vegetation and
the birds? extreme sensitivity to human distur-
bance (Lu et al. 2003). Other methods based on
physical capture, such as radiotelemetry and
mark?recapture, can provide robust data for the
study of habitat selection, home-range and popu-
lation parameters of phasianids (Sun et al. 2003,
*Corresponding author.
Email: shengli@pku.edu.cn
? 2009 The Authors
Journal compilation ? 2009 British Ornithologists? Union
Ibis (2010), 152, 299?309
Jia et al. 2004, Ji et al. 2005) but are relatively
time consuming, costly and potentially harmful
to the individuals (Long et al. 1998, Cutler &
Swann 1999). The most common indirect survey
methods for phasianids in China include calling
counts and counts of moulted feathers (Lu &
Zheng 2001, Lu et al. 2007), but these are usu-
ally season-dependent (Conroy & Carroll 2000,
Lu & Zheng 2001, Lu et al. 2003). For example,
as with many avian species, the White Eared
Pheasant Crossoptilon crossoptilon only undergoes
a complete moult after breeding, in this case
between late July and early September (Lu &
Zheng 2001, Lu et al. 2003). Phasianids can be
potential indicators of habitat quality and human
pressure on the environment (Fuller & Garson
2000), and there is clearly a need for a method-
ology that can effectively detect these reclusive
birds with little disturbance and monitor the
trends in population dynamics (Conroy & Carroll
2000).
Infrared-triggered cameras, in which a passive
infrared sensor is triggered by an abrupt change in
temperature across a fan-shaped area in front of
the unit, activating an autofocus camera (Swann
et al. 2004), have proved to be an effective tool for
wildlife research and have been applied success-
fully to studies of numerous terrestrial mammals
(Karanth & Nichols 1998, Cutler & Swann 1999,
Moruzzi et al. 2002, Numata et al. 2005, Wang
et al. 2006). Infrared cameras have been used as
nest monitors to study the breeding behaviour and
nest predation of numerous birds (Cutler & Swann
1999) and to estimate the population demograph-
ics, habitat selection, occupied habitat proportion
and activity pattern of a number of terrestrial
species (Pei 1998, Jeganathan et al. 2002, Dinata
et al. 2008, Winarni et al. 2009). Infrared-triggered
cameras can fail to detect animals due to faulty
settings, low battery power, electrical malfunction
or short detection ranges (Swann et al. 2004), but
these problems are not influenced by habitat if the
unit is properly deployed. Camera units work
24 h?day with little if any disturbance to wildlife,
making the technique suitable to record and moni-
tor cryptic and reclusive terrestrial animals
(Carbone et al. 2001, Silveira et al. 2003, Karanth
et al. 2004). Standard methodologies for study-
ing phasianids (Bibby et al. 1992, Conroy &
Carroll 2000) do not include infrared-triggered
cameras, although phasianids have been detected
during mammal surveys in which these cameras
have been used (Pei 1998, Chan et al. 2005, Lu
et al. 2005).
We analysed photographs of phasianids obtained
during two large mammal surveys within nature
reserves in Sichuan Province, southwestern China,
with the intent of determining the suitability of
this tool for the detection of large, terrestrial bird
species, and to highlight ecological and behavioural
information that can be obtained using infrared-
triggered cameras.
METHODS
Study area
The work was undertaken in two national reserves
(Wanglang Nature Reserve and Wolong Nature
Reserve) in Sichuan Province, southwestern China
(Fig. 1), from September 2004 to June 2007.
Wanglang National Nature Reserve (104C1763?E,
32C17656?N) is a 320-km
2
protected area established
in 1965 in the Min Mountains, 380 km northwest
of Chengdu, the capital of Sichuan Province.
Wolong National Nature Reserve (103C1768?E,
31C1766?N) is a 2000-km
2
protected area established
in 1963 in the Qionglai Mountains, 160 km west
of Chengdu. Both reserves were created to con-
serve Giant Panda Ailuropoda melanoleuca, Takin
Budorcas taxicolor and Golden Monkey Rhinopithe-
cus roxellarae. The elevational range of Wanglang
Nature Reserve is 2400?4980 m and major vege-
tation types are deciduous forest, conifer?decidu-
ous mixed forest, conifer forest, sub-alpine scrub
and alpine meadow. Wolong Nature Reserve has a
greater elevational range (1200?6250 m) and
comprises evergreen forest, evergreen?deciduous
mixed forest, deciduous forest, conifer?deciduous
mixed forest, conifer forest, sub-alpine scrub and
alpine meadow (Wolong National Nature Reserve
1987).
BirdLife International has identified both
reserves as Important Bird Areas (IBA Code:
CN194 for Wolong, CN188 for Wanglang), and
they fall within an Endemic Bird Area (EBA138)
because they host globally threatened and range-
restricted avian species including Chinese Monal
Lophophorus lhuysii (BirdLife International 2004).
Survey design
Trained staff in each reserve used two models of
infrared-triggered camera units: DeerCamC212 (Non
? 2009 The Authors
Journal compilation ? 2009 British Ornithologists? Union
300 S. Li et al.
Typical, Inc., Park Falls, WI, USA) and cameras
manufactured by project staff to similar specifica-
tions; both models had similar performance in tests
of sensor sensitivity and detection range. Ten Deer-
CamC212 and 20 self-manufactured camera units
were deployed in Wanglang and 20 self-manufac-
tured camera units were deployed in Wolong.
Camera units were positioned to minimize detec-
tion distances, with none exceeding 5 m. The
detection range of the sensors varies with the body
size of animals (Swann et al. 2004); for large ani-
mals, such as Takin and Giant Panda, the sensor
can be triggered at distances over 20 m, while for
small animals, such as Blood Pheasant and Snow
Partridge Lerwa lerwa, the maximum trigger dis-
tance is approximately 6 m. The reaction time for
both camera units (i.e. time taken between the
trigger of heat sensor and activation of the camera)
was 0.5?1 s. All the camera units were loaded with
400 ISO film and were set with a 2?3-min delay
between photographs, and set for 24-h monitoring.
Camera units were attached to trees 30?50 cm
above the ground and 3?5 m from a trail or
point where animal movement might be
expected. A time?date stamp accompanied each
photograph and at each sample site we recorded
GPS location, elevation, slope, aspect and habitat,
including forest type, canopy and shrub cover
and average diameter at breast height (DBH) of
trees. Field staff classified the habitat of the sam-
ple point into one of six forest types based on
tree species composition. At the end of each
monitoring session, the units were tested to con-
firm that they were still operational and had
unexposed frames; if not, the date on the last
photograph was taken as the last operational
date.
The reserves were divided into 1-km
2
blocks
within a Geographic Information System (GIS)
(ARC VIEW 3.2 and ARC GIS 9.0). One camera unit
was placed in each block for 1 month, and was
then moved to an adjacent block. Due to difficult
navigation in the field, two camera units were
occasionally placed in one block (14 blocks in
Wanglang and 16 blocks in Wolong). Within
each survey block, cameras were placed in likely
Figure 1. Location of sample sites within Wanglang National Nature Reserve (upper left) and Wolong National Nature Reserve (lower
left), China.
? 2009 The Authors
Journal compilation ? 2009 British Ornithologists? Union
Infrared camera phasianid survey 301
animal-use areas, as determined by field staff, and
> 400 m from other cameras. We concentrated our
survey effort in Wanglang (132 sites) within the
elevational range containing all forested habitat
(2400?3600 m). At Wolong, the 51 surveyed sites
were along 11 monitoring routes within an eleva-
tion range of 2000?4200 m. Field surveys were
conducted in Wanglang from September 2004 to
January 2005 and from March to October 2005; in
Wolong, we conducted surveys from April to
December 2006 and March to June 2007. Identifi-
cation of avian species was based on MacKinnon
et al. (2000), Zheng (2000, 2002), while names
and taxonomy follow Gill and Wright (2006).
We differentiated chicks, but not subadults, from
adult birds. For the majority of phasianid species,
the sex of adults could be distinguished by body
size, feather colour and pattern, or wattles on the
face or neck (MacKinnon et al. 2000, Zheng
2002). We were able to differentiate the sex of
adult individuals for five phasianid species: Sever-
tzov?s Grouse Tetrastes sewerzowi, Blood Pheasant,
Temminck?s Tragopan, Koklass Pheasant Pucrasia
macrolopha and Chinese Monal.
Photographic encounter rates were calculated
for each phasianid species for each 400-m eleva-
tional band. We defined detection at a sample
point as one individual photograph of one species
during a 30-min period. If more than one individ-
ual of the same species was identified on a single
photograph, we considered this one detection. All
detections for each species were summed for each
camera site, multiplied by 100, and divided by the
total sampling effort for that sample point (num-
ber of camera-days):
Photographic Rate ? No. of detections
C3 100=Camera-days
We used this photographic rate to compare detec-
tion in different reserves and elevational bands.
For the phasianid species that had sufficient
detections (i.e. Blood Pheasant and Temminck?s
Tragopan), the sampling record at each site was
divided into consecutive 5-day segments based on
the date stamp on the photographs. A detection
matrix of each species was established following
the approach proposed by MacKenzie et al.
(2002). We excluded sites where the sampling was
less than one 5-day segment. An occupancy model
(program PRESENCE, v. 2.2; MacKenzie et al. 2006,
Hines 2006) was used to estimate the site-
occupancy rate (w) and detection probability rela-
tive to eight sampling variables and three detection
variables (Table 1). Akaike Information Criterion
(AIC; Akaike 1973) values were used to rank the
occupancy models and all the models whose
DAIC ? 2 were considered as equivalent models.
The summed model weight of each covariate in
these models was used to determine the most
influential variables for each species. The sign of
logistic coefficient of each variable (positive or neg-
ative) was used to determine the direction of influ-
ence of the variable.
Table 1. Variables used to estimate the site occupancy rates and detection probabilities of Blood Pheasant and Temminck?s
Tragopan in the occupancy model.
Abbreviation Name Description
Sampling variables
NR Nature reserve Categorical (Wanglang, Wolong)
FOT Forest type Categorical (Broadleaved, Broadleaved?conifer mixed, Coniferous)
DBH Tree size (measured by DBH
a
) Categorical (< 30 cm, 30?50 cm, > 50 cm)
SCO Percentage shrub cover Categorical (< 25%, 25?50%, 50?75%, > 75%)
ELE Elevation Numeric (Range 1680?4220 m)
DTR Distance to nearest river Numeric (Range 3?1821 m)
DTT Distance to nearest road Numeric (Range 1?5100 m)
ASP Aspect Categorical (Warm ? NE, E, SE, S; Cold ? N, NW, W, SW)
Detection variables
CAM Camera model Commercially purchased or self-manufactured
SEA Season Breeding (March?July) or non-breeding (August?February)
b
LUR Scent lure durability Numeric (Days since application)
a
DBH, diameter at breast height.
b
Li (1996), Jia et al. (1999, 2003).
? 2009 The Authors
Journal compilation ? 2009 British Ornithologists? Union
302 S. Li et al.
The time and date printed on the photographs
has been used to determine the temporal patterns
of use for highway underpasses by different species
(Foster & Humphrey 1995) and the daily activity
pattern of individual species (Pei 1998). We used a
Daily Activity Index (DAI) of 2-h durations to
examine the daily activity level:
DAI ? No. of photographs within a duration
C3 100=Total no. photographs
A Chi-squared test using SPSS 13.0 (SPSS Inc.,
Chicago, IL, USA) was applied to determine the
significance of differences in the daily activity
patterns between species.
RESULTS
A sampling effort of 4908 camera-days across 183
sample sites (Fig. 1) was achieved in the two
reserves (3793 camera-days in Wanglang, 1115
camera-days in Wolong), resulting in 2750 photo-
graphs, 427 of which contained birds. During the
2-year survey, 30 bird species were recorded, nine
of which were phasianids (n = 308 photographs)
(Table 2, Appendix I). Within each reserve, six
species of phasianids were recorded.
Blood Pheasant and Temminck?s Tragopan were
the most frequently detected species for
both reserves within the mid-elevational range
(2400?3600 m). Snow Partridge, Tibetan Snow-
cock Tetraogallus tibetanus, and Chinese Monal
only occurred over 3600 m in sub-alpine shrub
habitats. Three species were detected in both
reserves: Blood Pheasant, Temminck?s Tragopan
and Koklass Pheasant. Of these, Blood Pheasant
was detected over the broadest elevational range
(2400 and 3800 m), with the highest photographic
rates (4.74 at Wanglang and 3.46 at Wolong)
between 2800 and 3200 m in both reserves. Tem-
minck?s Tragopan was detected over the elevational
range 2200?3200 m, and had the highest photo-
graphic rates (0.64 at Wanglang and 7.94 at
Wolong) at 2400?2800 m elevation in both
reserves. Koklass Pheasant was only detected at
2400?2800 m elevation in both reserves.
Blood Pheasant and Temminck?s Tragopan were
detected at least once at 49 sites and 23 sites,
respectively. The estimates of site occupancy rate of
both species were slightly higher than the naive esti-
mates (i.e. the proportion of sites where the species
was detected at least once, 0.27 for Blood Pheasant
and 0.13 for Temminck?s Tragopan), although the
estimated detection probabilities of both species
were smaller than 0.40 (Table 3), suggesting that
our sampling duration (30 days, maximum 45 days)
was long enough to detect the species at each site
when they were present. The estimation of occu-
pancy rate of Blood Pheasant (0.30) was higher than
that of Temminck?s Tragopan (0.14), indicating that
Blood Pheasant was more widely distributed. Den-
ser shrub cover, lower elevation and warmer aspect
were determined as the best predictors for the occu-
pancy of both species. Blood Pheasant occurred
more commonly at Wanglang and Temminck?s
Tragopan at Wolong (Table 4).
The DAI for Blood Pheasant and Temminck?s
Tragopan (160 and 46 photographs, respectively)
confirmed that both species are diurnal (Fig. 2).
Table 2. Phasianid species recorded with camera units in Wanglang National Nature Reserve and Wolong National Nature Reserve
(2005?2007).
Scientific name Common name IUCN threat status
a
Wanglang Wolong
Photos Sites Photos Sites
Tetrastes sewerzowi Severtzov?s Grouse NT 1 1 ? ?
Lerwa lerwa Snow Partridge LC ? ? 13 2
Tetraophasis obscurus Verreaux?s Monal-Partridge LC 4 4 ? ?
Tetraogallus tibetanus Tibetan Snowcock LC ? ? 4 1
Ithaginis cruentus Blood Pheasant LC 150 43 33 6
Tragopan temminckii Temminck?s Tragopan LC 25 8 43 15
Lophophorus lhuysii Chinese Monal VU ? ? 8 3
Pucrasia macrolopha Koklass Pheasant LC 5 2 10 3
Crossoptilon auritum Blue Eared Pheasant LC 12 9 ? ?
Name and taxonomy based on Gill and Wright (2006), Birds of the World.
a
LC, Least Concern; NT, Near Threatened; VU, Vulnerable.
? 2009 The Authors
Journal compilation ? 2009 British Ornithologists? Union
Infrared camera phasianid survey 303
Blood Pheasant showed a late morning activity
peak at 10:00?12:00 h, while there was no obvious
activity peak for Temminck?s Tragopan, but this
difference was not significant (v
2
= 8.06, df = 6,
P > 0.1).
With regard to sex, 224 adult Blood Pheasants
were photographed including 152 males and 72
females (M : F = 2.11 : 1.00) (Table 5). Groups
with multiple adults (i.e. female?female pairs,
female?male pairs and male?male pairs) were
recorded. Aggregating behaviour was recorded in
eight photographs that contained more than two
adult individuals, with one photograph containing
a flock of eight adults. For Temminck?s Tragopan,
69 adult individuals were photographed including
48 males and 21 females (M : F = 2.29 : 1.00).
Only two photographs of Temminck?s Tragopan
contained more than one individual.
Camera unit reliability
We used data from 183 sites for our analysis,
but cameras placed at additional sites did not
yield useful information. Of 153 sites sampled
in Wanglang, 21 camera units were not working
at the 1-month check and the sites were not
considered in the analysis. At Wolong, 22 of 73
camera units were not operating after 1 month.
Electrical malfunction due to moisture was the
reason for failure of six camera units at Wang-
lang and nine units at Wolong. Low-quality
batteries purchased in rural areas and incorrect
recharging of lithium batteries caused camera
unit failure at 10 sites in Wanglang and nine
sites in Wolong. Two camera units, one in each
reserve, were attacked and damaged by Giant
Pandas, based on bite marks left on the camera
Table 3. The top models for predicting site occupancy of Blood Pheasant and Temminck?s Tragopan in Wanglang and Wolong
National Nature Reserves.
Models DAIC AIC weight No. par. ()2LL) est. w (? 1 se) est. P
Blood Pheasant
w (NR, SCO, ELE, DTR, ASP) P(.) 0.00 0.1311 6 533.0311 0.3016 (? 0.0111) 0.3623
w (NR, SCO, ELE, DTR) P(.) 0.75 0.0901 5 535.7777 0.3007 (? 0.0100) 0.3625
w (NR, ELE, DTR, ASP) P(.) 0.99 0.0799 5 536.0153 0.2994 (? 0.0099) 0.3629
w (NR, DBH, SCO, ELE, DTR, ASP) P(.) 1.43 0.0641 7 532.4641 0.3021 (? 0.0113) 0.3626
w (NR, FOT, SCO, ELE, DTR, ASP) P(.) 1.92 0.0502 7 532.9525 0.3018 (? 0.0111) 0.3621
w (NR, DBH, SCO, ELE, DTR) P(.) 2.00 0.0482 6 535.0332 0.3013 (? 0.0103) 0.3629
Temminck?s Tragopan
w (NR, SCO, ELE) P(.) 0.00 0.1358 4 238.5937 0.1427 (? 0.0122) 0.2983
w (NR, SCO, ELE, ASP) P(.) 0.40 0.1112 5 236.9865 0.1451 (? 0.0129) 0.2950
w (NR, ELE, ASP) P(.) 0.63 0.0991 4 239.2188 0.1434 (? 0.0120) 0.2943
w (NR, ELE) P(.) 1.18 0.0753 3 241.7730 0.1403 (? 0.0109) 0.2980
w (NR, DBH, SCO, ELE, ASP) P(.) 1.21 0.0741 6 235.7987 0.1421 (? 0.0125) 0.2981
w (NR, SCO, ELE, DTR) P(.) 1.85 0.0538 5 238.4408 0.1426 (? 0.0123) 0.2984
w (NR, SCO, ELE, DTR, ASP) P(.) 1.93 0.0517 6 236.5236 0.1454 (? 0.0131) 0.2947
We list all models whose DAIC ? 2 and present AIC weight, number of parameters (No. par.), twice the negative log likelihood ()2LL),
estimated occupancy rate (est. w) and estimated detection probability (est. P ) for each model. The key for the covariate codes used
is given in Table 1. None of the final models contained covariates used to measure detection variability (i.e. camera model, season,
or lure type) and P(.) was used to indicate this fact.
Table 4. Summed model weight of each sampling variable in the equivalent models listed in Table 3.
Species
Model variables
Nature
reserve
(NR)
Forest
type
(FOT)
Tree
size
(DBH)
Percentage
shrub cover
(SCO)
Elevation
(ELE)
Distance to
nearest
river (DTR)
Distance to
nearest
road (DTT)
Aspect
(ASP)
Blood Pheasant 0.4636 0.0502 0.1123 0.3837 0.4636 0.4636 0 0.3253
Temminck?s Tragopan 0.6010 0 0.0741 0.6010 0.4300 0.1055 0 0.3361
? 2009 The Authors
Journal compilation ? 2009 British Ornithologists? Union
304 S. Li et al.
unit casing and the last photographs. Seven units
(four at Wanglang and three at Wolong) were
stolen or damaged. In total, 19% of the sites
(Number of sites failed?Total number of sites
surveyed) resulted in no viable data due to cam-
era failure or loss.
DISCUSSION
Of the nine phasianids detected in this study, only
one species (Chinese Monal) is considered globally
endangered or threatened (Table 2). However,
three species (Severtzov?s Grouse, Verreaux?s
Monal-Partridge and Chinese Monal) are listed in
Category I of China Nationally Protected Animals,
and the other six species are listed in Category II
(Zheng & Wang 1998, MacKinnon et al. 2000). In
addition, four species (i.e. Severtzov?s Grouse, Ver-
reaux?s Monal-Partridge, Chinese Monal and Blue
Eared Pheasant) are endemic to central and south-
western China (Lei et al. 2002, 2003, Zhang et al.
2003). For Wanglang, our photographs are the first
documentation that Verreaux?s Monal-Partridge
occurs within this reserve despite 6 years of prior
monitoring activity. For Wolong, our work pro-
vided the first photographic evidence for Snow Par-
tridge and Tibetan Snowcock, both of which were
heard during 3 years of monitoring, but never
observed or photographed.
We recorded an additional 21 avian species dur-
ing the camera-trapping efforts, including some as
small as the Green-backed Tit Parus monticolus
(body length = 13 cm) and Golden Bush Robin
Tarsiger chrysaeus (body length = 14 cm).
Although we photographed multiple bird species,
there are problems inherent to infrared-triggered
cameras when comparing species of markedly dif-
ferent body size and foraging strata (Hernandez
et al. 1997, York et al. 2001, Moruzzi et al. 2002).
Table 5. Occurrence by sex of Blood Pheasant and
Temminck?s Tragopan photographed in Wanglang National
Nature Reserve and Wolong National Nature Reserve,
September 2005?August 2007.
Sex(es) in
photograph
Blood Pheasant Temminck?s Tragopan
No. of photographs
(proportion)
n = 174
No. of photographs
(proportion)
n =68
Single # 91 (0.52) 47 (0.69)
Single $ 43 (0.25) 19 (0.28)
#?# Pair 14 (0.08) ?
#?$ Pair 11 (0.06) 1 (0.01)
$?$ Pair 3 (0.02) ?
$?C
a
2
b
(0.01) 1
c
(0.01)
C 2 (0.01) ?
Multi-adults (> 2) 8
d
(0.05) ?
No. of adult
individuals
152 # :72$ 48 # :21$
a
C, chick.
b
Two photographs each showed 1$?1C.
c
One photograph showed 1$?3C.
d
One photograph each showed 6#?2$,3#?2$,3#?1$,2#?1$,
2#?1$?1C, and 4$. Two photographs showed 3#.
Figure 2. Daily activity pattern of Blood Pheasant (n = 160 photographs) and Temminck?s Tragopan (n = 46 photographs), from com-
bined data of Wanglang National Nature Reserve and Wolong National Nature Reserve, September 2005?August 2007.
? 2009 The Authors
Journal compilation ? 2009 British Ornithologists? Union
Infrared camera phasianid survey 305
These issues are not as problematic for the nine
species of phasianids photographed, whose body
sizes (range 33?95 cm) are large enough to be
detected by the camera sensor at < 5 m and who
all forage primarily on the forest floor (Zheng et al.
1978, MacKinnon et al. 2000).
Our photographic survey did not detect several
phasianids that are on the species list maintained by
each reserve, for example Snow Partridge, Chinese
Monal and Common Pheasant Phasianus colchicus
at Wanglang (Liu et al. 2001), and Severtzov?s
Grouse, Verreaux?s Monal-Partridge, Golden Pheas-
ant Chrysolophus pictus and White Eared Pheasant
at Wolong (Yu & Deng 1993). It is possible the
missing species occur in habitats beyond our survey
area or elevational range. For example, the Chinese
Monal is reported to inhabit sub-alpine rhododen-
dron shrub, sub-alpine and alpine meadows, and
exposed cliffs above the treeline (Long et al. 1998,
BirdLife International 2001), an area beyond our
surveyed elevation at Wanglang. Likewise, there
are approximately 800 records in the last 3 years
for Severtzov?s Grouse in sub-alpine shrub habitat
above 3600 m in Wolong (X.G. Shi, Wolong
Nature Reserve unpubl. data), but our cameras
failed to detect them, probably due to our relatively
small sampling effort at this elevation (three sample
points, 115 camera-days). However, for species
such as Verreaux?s Monal-Partridge, where the pre-
ferred elevation is well within the range of our sam-
pling, it is probable that these species are either rare
or extirpated from Wolong. There have been no
sightings of Verreaux?s Monal-Partridge during the
reserve monitoring activities in the last 3 years
(X.G. Shi, Wolong Nature Reserve unpubl. data),
indicating the density of this species is currently
low within the reserve.
Of the three phasianids recorded in both reser-
ves, the Blood Pheasant and Temminck?s Tragopan
showed a broad distribution across the mid-eleva-
tional range (2400?3600 m). Other studies on the
ecology and distribution of these two species (Shi
et al. 1996, Li et al. 1998, Yu et al. 2000) indicated
that Temminck?s Tragopan occupies a broad range
of habitats (i.e. evergreen forest, deciduous forest,
deciduous and conifer mixed forest, and conifer
forest) and elevations (i.e. 650?3500 m). This is in
contrast to the Blood Pheasant, which occupies
deciduous?conifer mixed forest, conifer forest and
sub-alpine shrub at elevations ranging from 2300 to
4500 m (Li 1996, MacKinnon et al. 2000, Yu et al.
2000). In our survey, the Blood Pheasant was not
detected below 2400 m and was detected propor-
tionately more often in mature coniferous forest.
Temminck?s Tragopan was detected at a lower eleva-
tion and seemed to prefer broadleaved forest. Thus,
we found both species in less diverse situations than
those mentioned in these previous reports.
There were apparent differences in the social
organization of the two most abundant species, as
Blood Pheasants were observed in pairs or flocks
much more often than Temminck?s Tragopan. Both
Blood Pheasant and Temminck?s Tragopan are con-
sidered monogamous (Li 1996, Jia et al. 1999,
2004), but for Blood Pheasants, both single-sex and
mixed-sex flocks were photographed, whereas for
Temminck?s Tragopan no adult groupings of more
than a pair were photographed. We obtained sev-
eral photographs of Blood Pheasants in multi-adult
groups, as was previously reported in Wolong by Jia
et al. (1999) and in Shiqu by Lu et al. (2006), but
we obtained no photographs of such groups in Tem-
minck?s Tragopan. There is no published research
on the sex ratio or population structure of Tem-
minck?s Tragopan in China, but a study on a closely
related species, Cabot?s Tragopan Tragopan caboti,
estimated an even sex ratio within a population of
50 individuals (Zhang & Zheng 1990). A previous
study of Blood Pheasant in Wolong (Jia et al. 1999)
reported a sex ratio of 0.89 : 1.00, based on the
observation of 70 individuals in four winter flocks,
but we found male pheasants to be twice as abun-
dant as female pheasants in the photographs for
both species. Male-skewed adult sex ratios are com-
mon in birds (Donald 2007) but there are other fac-
tors that should be considered. One possibility is
that males of both species may be detected more
frequently than females because they spend more
time patrolling and defending their territories
(Zhang & Zheng 1999, Jia et al. 2003). A second
possibility is that males are detected more often
because they take the lead position when the pair
or flock moves through the habitat, with the trailing
animals being missed by the time-delayed camera.
Many studies using infrared-triggered cameras
are primarily designed for large mammal research
and the bird species detected during the sampling
are considered ?bycatch? detections (Chan et al.
2005). There may be three important reasons for
the lack of further application and analysis of cam-
era data for bird species in these studies. First,
camera settings for large mammals may not be
suitable to detect terrestrial birds. For both the
passive and the active camera units, cameras may
? 2009 The Authors
Journal compilation ? 2009 British Ornithologists? Union
306 S. Li et al.
be set at a height equivalent to the target animal?s
shoulder or chest. Therefore, cameras set for large
mammals may not be triggered by passing pheas-
ants. Secondly, phasianids cannot usually be individ-
ually identified in photographs, so mark?recapture
approaches, which are widely applied in studies
on large felids based on camera data (Karanth &
Nichols 1998, Cutler & Swann 1999), are not
appropriate for these phasianids. Thirdly, the rela-
tionship between camera photographic rate and
animal density remains unclear (Carbone et al.
2001, Jennelle et al. 2002). These potential limita-
tions do not preclude the incorporation of camera-
trapping into phasianid surveys.
Although our field protocol was developed as a
general survey of large- and medium-sized mam-
mals, it was effective in recording phasianids in for-
ested and remote areas. Our data demonstrate that
a combination of cameras and use of occupancy
models can be an effective approach for future
phasianid studies. Occupancy analyses suggested
that our survey protocol (30?50 cm above the
ground and 3?5 m from the trail or focal point, and
a 1-month sampling duration at each site) was suffi-
cient to detect the presence of several phasianids.
Given the relatively small home-range of phasia-
nids (e.g. 8.7?31.9 ha for Blood Pheasant in the
breeding season; Jia et al. 2004), we would recom-
mend a higher camera density (3?4 cameras?km
2
).
We do not recommend using the photographic rate
for direct comparison of the abundance of species
because of species-specific differences in behaviour
and population ecology (Hernandez et al. 1997,
York et al. 2001, Moruzzi et al. 2002, Stephens
et al. 2006, Rowcliffe et al. 2008), which may
affect detection probability by infrared-triggered
camera of any phasianids. Our analysis indicated
that the occupancy rate of phasianids could be esti-
mated using occupancy models and can be used as
a robust index for phasianid monitoring projects.
A standardized deployment of infrared-triggered
cameras should be a reliable tool for comparing
sites or time periods within a species, with minimal
variation among researchers and areas. At its most
basic level, the camera units provided information
on species? natural history that was not obtainable
using standard monitoring within these reserves
and will enhance decision-making for reserve man-
agement and conservation activities.
This work was supported by the Smithsonian?s
National Zoological Park, Friends of the National Zoo,
Peking University, World Wildlife Fund China, Sichu-
an Forestry Department, and China Wildlife Conserva-
tion Association. We thank Chen Youping, Jiang
Shiwei, Huang Junzhong, Xu Haibin, Zhang Qingyu,
Tang Hao and all the reserve staff who assisted with
data collection. We also thank John H. Rappole,
Pan Wenshi, Lu Zhi, Wang Hao, Shen Xiaoli, Liu
Yang, and several reviewers for fruitful discussions and
valuable comments on the manuscript.
REFERENCES
Akaike, H. 1973. Information theory and an extension of the
maximum likelihood principle. In Petrov, B.N. & Csadki, F.
(eds) Proceedings of 2nd International Symposium on
Information Theory: 267?281. Budapest: Akade?miai Kiado?.
Bibby, C.J., Burgess, N.D. & Hill, D.A. 1992. Bird Census
Techniques. London: Academic Press.
BirdLife International. 2001. Threatened Birds of Asia: The
BirdLife International Red Data Book. Cambridge: Birdlife
International.
BirdLife International. 2004. Important Bird Areas of
Asia: Key sites for conservation. Cambridge, UK: BirdLife
International.
Carbone, C., Christie, S., Conforti, K., Coulson, T., Frank-
lin, N., Ginsberg, J.R., Griffiths, M., Holden, J., Kamani-
shi, K., Kinnaird, M., Laidlaw, R., Lynam, A.,
Macdonald, D.W., Martyr, D., McDougal, C., Nath, L.,
O?Brien, T., Seidensticker, J., Smith, D.J.L., Sunquist,
M., Tilson, R. & Wan Sharuddin, W.N. 2001. The use of
photographic rates to estimate densities of tigers and other
cryptic mammals. Anim. Conserv. 4: 75?79.
Chan, B.L., Lee, K.S., Zhang, J.F. & Su, W.B. 2005. Notable
bird records from Bawangling National Nature Reserve,
Hainan Island, China. Forktail 21: 33?41.
Conroy, M.J. & Carroll, J.P. 2000. Estimating Abundance of
Galliformes: Tools and Application. Galliformes 2000: Pro-
ceedings of the 7th International Galliformes Symposium.
Kathmandu, Nepal.
Cutler, T.L. & Swann, D.E. 1999. Using remote photography
in wildlife ecology: a review. Wildl. Soc. Bull. 27: 571?581.
Dinata, Y., Nugroho, A., Haidir, I.A. & Linkie, M. 2008. Cam-
era trapping rare and threatened avifauna in west-central
Sumatra. Bird Cons. Int. 18: 30?37.
Donald, P.F. 2007. Adult sex ratios in wild bird populations.
Ibis, 149: 671?892.
Foster, M.L. & Humphrey, S.R. 1995. Use of highway under-
passes by Florida panthers and other wildlife. Wildl. Soc.
Bull. 23: 95?100.
Fuller, A.R. & Garson, P.J. 2000. Pheasants: Status Survey
and Conservation Action Plan 2000?2004. Switzerland and
Cambridge, UK: WPA?BirdLife? SSC Pheasant Specialist
Group.
Gill, F.B. & Wright, M. 2006. Birds of the World: Recom-
mended English Names. New Jersey: Princeton University
Press.
Hernandez, F., Rollins, D. & Cantu, R. 1997. An evaluation
of Trailmaster
C210
camera systems for identifying ground-nest
predators. Wildl. Soc. Bull. 25: 848?853.
Hines, J.E. 2006. PRESENCE v2.2 ? Software to estimate
patch occupancy and related parameters. USGS-PWRC.
? 2009 The Authors
Journal compilation ? 2009 British Ornithologists? Union
Infrared camera phasianid survey 307
Available at: http://www.mbr-pwrc.usgs.gov/software/
PRESENCE.html [accessed 25 November 2008].
Jeganathan, P., Green, R.E., Bowden, C.G.R., Norris, K.,
Pain, D. & Rahmani, A. 2002. Use of tracking strips and
automatic cameras for detecting critically endangered Jer-
don?s coursers Rhinoptilus bitorquatus in scrub jungle in
Andhra Pradesh, India. Oryx. 36: 182?188.
Jennelle, C.S., Runge, M.C. & MacKenzie, D.I. 2002. The
use of photographic rates to estimate densities of tigers and
other cryptic mammals: a comment on misleading conclu-
sions. Anim. Conserv. 5: 119?120.
Ji, T., Jia, C.X., Jiang, Y.X. & Sun, Y.H. 2005. Spring habitat
selection of the Chinese Grouse at Lianhuashan. Chinese
J. Zool. 40: 49?53.
Jia, C.X., Zheng, G.M., Zhou, X. & Zhang, H.M. 1999.
Social organization of Blood Pheasant (Ithaginis
cruentus) in Wolong Nature Reserve. Acta Zool. Sin. 45:
135?142.
Jia, C.X., Zheng, G.M., Zhou, X. & Zhang, H.M. 2003. Blood
Pheasant behaviour during the breeding season in Wolong
Nature Reserve, China. Chinese J. Zool. 38: 37?40.
Jia, C.X., Zheng, G.M., Zhou, X. & Zhang, H.M. 2004. Home
range and habitat characteristics of Blood Pheasant in sum-
mer. Sichuan J. Zool. 23: 349?352.
Karanth, K.U. & Nichols, J.D. 1998. Estimation of tiger densi-
ties in India using photographic captures and recaptures.
Ecology 79: 2852?2862.
Karanth, K.U., Nichols, J.D. & Kumar, N.S. 2004. Photo-
graphic Sampling of Elusive Mammals in Tropical Forests.
Sampling Rare or Elusive Species. Washington, DC: Island
Press.
Lei, F.M., Qu, Y.H., Lu, J.L., Yin, Z.H. & Lu, T.C. 2002. A
revision of China?s Endemic Bird List. Acta Zootaxonomica
Sin. 27: 857?864.
Lei, F.M., Qu, Y.H., Lu, J.L., Liu, Y. & Yin, Z.H. 2003. Con-
servation on diversity and distribution patterns of endemic
birds in China. Biodivers. Conserv. 12: 239?254.
Li, H.H., Yu, T.L. & Shen, L.T. 1998. The birds of Tragopan
Cuvior and Syrmaticus Wagler and their geological distribu-
tion in Guangxi, China. J. Guangxi Normal University 16:
76?80.
Li, X.T. 1996. Gamebirds of China. Beijing: China Forestry
Publishing House.
Liu, S.Y., Ran, J.H. & Lin, Q. 2001. Diversity of vertebrates in
Wanglang Nature Reserve. J. Sichuan For. Sci. Tech. 22:
10?14.
Long, T.L., Shao, K.Q., Guo, G., Cheng, C.Y., Zou, X.Y.,
Landel, H., Rimlinger, D. & Zhou, F.L. 1998. Study on the
winter ecology of Chinese Monal Lophophorus lhuysii. Sich-
uan J. Zool. 17: 104?105.
Lu, G., Dai, B., Li, R.G., Yang, J., Zhang, X.L., Feng, S.L.,
Ran, J.H. & Yue, B.S. 2007. Galiforme [sic] population den-
sities in Laojunshan Nature Reserve, Sichuan province.
Sichuan J. Zool. 26: 572?576.
Lu, Q.B., Wang, X.M. & Wang, Z.H. 2006. Correlation of
group and habitat requirement for alpine Blood Pheasants
in the initial mating period in Shiqu, Sichuan. Zool. Res. 27:
243?248.
Lu, X. & Zheng, G.M. 2001. Habitat selection and use by a
hybrid of White and Tibetan Eared Pheasants in eastern
Tibet during the post-incubation period. Can. J. Zool. 79:
319?324.
Lu, X., Cangjue, Z.M., Suolong, C.R. & Zheng, G.M. 2003.
Two field techniques for estimating relative abundance of
Galliformes. Wuhan Univ. J. Nat. Sci. 8: 459?462.
Lu, X.L., Jiang, Z.G., Tang, J.R., Wang, X.J., Xiang, D.Q. &
Zhang, J.P. 2005. Auto-trigger camera traps for studying
giant panda and its sympatric wildlife species. Acta Zool.
Sinica 51: 495?500.
MacKenzie, D.I., Nichols, J.D., Lachman, G.B., Droege, S.,
Royle, J.A. & Langtimm, C.A. 2002. Estimating site occu-
pancy rates when detection probabilities are less than one.
Ecology 83: 2248?2255.
MacKenzie, D.I., Nichols, J.D., Royle, J.A., Pollack, K.H.,
Bailey, L.L. & Hines, J.E. 2006. Occupancy Estimation
and Modeling. New York: Academic Press.
MacKinnon, J., Phillipps, K. & He, F.Q. 2000. A Field Guide
to the Birds of China. Oxford: Oxford University Press.
Moruzzi, T.L., Fuller, T.K., DeGraaf, R.M., Brooks, R.T. & Li,
W. 2002. Assessing remotely triggered cameras for survey-
ing carnivore distributions. Wildl. Soc. Bull. 30: 380?386.
Numata, S., Okuda, T., Sugimoto, T., Nishimura, S.,
Yoshida, K., Quah, E.S., Yasuda, M., Muangkhum, K. &
Noor, N.S.M. 2005. Camera trapping: a non-invasive
approach as an additional tool in the study of mammals in
Pasoh Forest Reserve and adjacent fragmented areas in
Peninsular Malaysia. Malay. Nat. J. 57: 29?45.
Pei, J.Q. 1998. An evaluation of using auto-trigger cameras to
record activity patterns of wild animals. Taiwan J. For. Sci.
13: 317?324.
Rowcliffe, J.M., Field, J., Turvey, S.T. & Carbone, C. 2008.
Estimating animal density using camera traps without the
need for individual recognition. J. Appl. Ecol. 45: 1228?
1236.
Shi, H.T., Zheng, G.M., Jiang, H. & Wu, Z.K. 1996. The study
on habitat selection of Temminck?s Tragopan. Acta Zool.
Sinica 42: 90?95.
Silveira, L., Jacomo, A.T.A. & Diniz-Filho, J.A.F. 2003. Cam-
era trap, line transect census and track surveys: a compar-
ative evaluation. Biol. Conserv. 114: 351?355.
Stephens, P.A., Zaumyslova, O.Yu., Miquelle, D.G., Myslen-
kov, A.I. & Hayward, G.D. 2006. Estimating population
density from indirect sign: track counts and the Formozov-
Malyshev-Pereleshin formula. Anim. Conserv. 9: 339?348.
Sun, Y.H., Swenson, J.E., Fang, Y., Klaus, S. & Scherzin-
ger, W. 2003. Population ecology of the Chinese Grouse,
Bonasa sewerzowi, in a fragmented landscape. Biol. Con-
serv. 110: 177?184.
Swann, D.E., Hass, C.C., Dalton, D.C. & Wolf, S.A. 2004.
Infrared-triggered cameras for detecting wildlife: an evalua-
tion and review. Wildl. Soc. Bull. 32: 357?365.
Wang, D.J., Li, S., McShea, W.J. & Li, M.F. 2006. Use of
remote-trip cameras for wildlife surveys and evaluating the
effectiveness of conservation activities at a nature reserve
in Sichuan Province, China. Environ. Manage. 38: 942?951.
Winarni, N.L., O?Brien, T.G., Carroll, J.P. & Kinnaird, M.F.
2009. Movement, distribution, and abundance of Great
Argus Pheasants (Argusianus argus) in a Sumatran rainfor-
est. Auk 126: 341?350.
Wolong National Nature Reserve. 1987. Vegetation and
Plant Resource of Wolong. Chengdu: Sichuan Science and
Technology Press.
York, C.E., Moruzzi, T.L., Fuller, T.K., Organ, J.F., Sauvajot,
R.M. & DeGraaf, R.M. 2001. Description and evaluation of
? 2009 The Authors
Journal compilation ? 2009 British Ornithologists? Union
308 S. Li et al.
a remote camera and triggering system to monitor carni-
vores. Wildl. Soc. Bull. 29: 1228?1237.
Yu, Z.W.& Deng,Q.X. 1993. Findings report on birds in Wolong
Natural Reserve. J. Sichuan Teach. Coll. 14: 233?235.
Yu, Y.Q., Wu, J.P., Guo, S.T., He, P.J., Ji, M.Z. & Hu, Y.L.
2000. The population density and community structure of
the pheasants in northern Qinling Mountains. Chin. Biodivers.
8: 60?64.
Zhang, J.P. & Zheng, G.M. 1990. The study of the population
number and structure of Cabot?s Tragopan. Zool. Res. 11:
291?297.
Zhang, R.Z. 1999. Zoogeography of China. Beijing: Science
Press.
Zhang, Y.Y. & Zheng, G.M. 1999. A study on the
behaviours of Satyr Tragopan (Tragopan satyra)in
captivity in breeding season. J. Beijing Normal University
34: 508?512.
Zhang, Z.W., Ding, C.Q., Ding, P. & Zheng, G.M. 2003. The
current status and a conservation strategy for species of
Galliformes in China. Biodivers. Sci. 11: 414?421.
Zheng, G.M. 2005. A Checklist on the Classification and Distri-
bution of the Birds of China. Beijing: Science Press.
Zheng, G.M. & Wang, Q.S. 1998. China Red Data Book of
Endangered Animals: Aves. Beijing: Science Press.
Zheng, Z.X. 2000. A Complete Checklist of Species and Sub-
species of the Chinese Birds. Beijing: Science Press.
Zheng, Z.X. 2002. The Keys to the Birds of China. Beijing:
Science Press.
Zheng, Z.X., Tan, Y.K., Lu, T.C., Tang, C.Z., Bao, G.J. & Li,
F.L. 1978. Fauna Sinica, Aves. Vol. 4: Galliformes. Beijing:
Science Press.
Received 28 March 2009;
revision accepted 23 October 2009.
APPENDIX 1
List of all non-phasianid avian species recorded during our survey in Wanglang National Nature Reserve and Wolong National Nature
Reserve (2005?2007).
Order Family Scientific name Common name
Wanglang Wolong
# photographs # sites # photographs # sites
Passeriformes Corvidae Urocissa erythrorhyncha Red-billed Blue Magpie ? ? 2 1
Paridae Parus monticolus Green-backed Tit ? ? 1 1
Timaliidae Garrulax cineraceus Moustached Laughingthrush 1 1 ? ?
Garrulax lunulatus Barred Laughingthrush 1 1 1 1
Garrulax maximus Giant Laughingthrush 1 1 2 3
Garrulax ocellatus Spotted Laughingthrush ? ? 3 1
Garrulax elliotii Elliot?s Laughingthrush 3 3 2 3
Garrulax affinis Black-faced Laughingthrush 3 2 3 1
Sittidae Sitta nagaensis Chestnut-vented Nuthatch ? ? 1 1
Turdidae Myophonus caeruleus Blue Whistling Thrush ? ? 1 1
Zoothera dixoni Long-tailed Thrush 4 3 ? ?
Zoothera dauma Scaly Thrush 2 2 3 1
Turdus rubrocanus Chestnut Thrush 2 2 4 1
Turdus mupinensis Chinese Thrush 7 5 3 1
Muscicapidae Tarsiger indicus White-browed Bush Robin 1 1 ? ?
Tarsiger cyanurus Red-flanked Bluetail 5 3 3 3
Tarsiger chrysaeus Golden Bush Robin 3 2 7 1
Prunellidae Prunella immaculata Maroon-backed Accentor 2 1 ? ?
Fringillidae Fringilla montifringilla Brambling 2 1 ? ?
Carpodacus thura White-browed Rosefinch 1 1 ? ?
Pyrrhula erythaca Grey-headed Bullfinch ? ? 1 1
Name and taxonomy based on Gill and Wright 2006, Birds of the World.
The IUCN threat status of all species in Appendix I is Least Concern.
? 2009 The Authors
Journal compilation ? 2009 British Ornithologists? Union
Infrared camera phasianid survey 309