This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier?s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
Author's personal copy
Does science replace traditions? Correlates between traditional Tibetan culture
and local bird diversity in Southwest China
Xiaoli Shen a,c,?, Sheng Li a, Nyima Chen b, Shengzhi Li b, William J. McShea c, Zhi Lu a,b
aCenter for Nature and Society, School of Life Sciences, Peking University, Beijing 100871, China
b Shan Shui Conservation Center, Beijing 100871, China
c Smithsonian Conservation Biology Institute, National Zoological Park, 1500 Remount Rd., Front Royal, VA 22630, USA
a r t i c l e i n f o
Article history:
Received 27 April 2011
Received in revised form 26 October 2011
Accepted 30 October 2011
Available online 20 November 2011
Keywords:
Traditional ecological knowledge
Scienti?c ecological knowledge
Traditional practice index
Knowledge?attitudes?behaviors
Point count
a b s t r a c t
A positive relationship between traditional cultures and biodiversity exists worldwide, but when tradi-
tional and formal conservation institutions coexist, how they interact and affect biodiversity remains
poorly studied. From 2005 to 2007, we studied the relationship between Tibetan traditional practices
and biodiversity. Speci?cally, how traditional ecological knowledge (TEK) and scienti?c ecological knowl-
edge (SEK) affect local biodiversity by affecting people?s attitudes and behaviors towards conservation.
We interviewed 331 villagers in nine Tibetan villages in Sichuan Province, China. We used proxy ques-
tions to measure the traditional practices, TEK, SEK, conservation attitudes and behaviors of village
residents. Meanwhile, we assessed the bird diversity around the villages by strati?ed sampling and point
counts. The results indicate traditional practices exhibited a strong positive correlation with TEK, but a
negative correlation with formal education and SEK. The villagers with high traditional practices had
more positive attitudes towards conservation and more actively participated in conservation than villag-
ers with low traditional practices, and villagers with medium traditional practices were the least
concerned about, or participated in, conservation activities. Bird species richness, abundance, and the
Shannon?Wiener diversity index were positively correlated with the traditional practice index of each
village. The results of a negative binomial regression showed the traditional practice index was a positive
correlative factor of bird species richness, while formal education was not a signi?cant variable, after con-
trolling for other potential sampling and environment factors. Government-sponsored conservation edu-
cation was somewhat successful in raising people?s environmental awareness, but these efforts have yet
to correlate with enhanced biodiversity measures.
 2011 Elsevier Ltd. All rights reserved.
1. Introduction
The linkage between traditional cultures and biodiversity has
been studied and advocated by researchers during the last two
decades for the potential application to conservation management
(Gadgil et al., 1993; Loh and Harmon, 2005; Maf?, 2005).
Traditional ecological knowledge (TEK) rooted in indigenous cul-
tures plays an essential role in contributing to local biodiversity
conservation (Berkes et al., 2000). Numerous case studies of indig-
enous groups in India (Bhagwat et al., 2005; Jamir and Pandey,
2003; Ntiamoa-Baidu, 2008), China (Anderson et al., 2005; Christo-
pher, 2008; Luo et al., 2009), Southeast Asia (Wadley and Colfer,
2004), North America (Diemont and Martion, 2009; Turner et al.,
2000), South America (Castro and Aldunate, 2003) and Africa
(Decher, 1997; Mgumia and Oba, 2003) demonstrate the positive
effects of traditional cultures in biodiversity conservation.
Previous studies that examined the linkage between TEK and
biodiversity usually infer fromexaminations of beliefs and practices
to conclusions about ecosystem condition without direct measure-
ments of biodiversity. Studies that quantify the linkage between
TEK and biodiversity have mainly focused on comparing areas
under different conservation management (e.g., protected areas,
sacred groves and community forests) (Bhagwat et al., 2005; Mgu-
mia and Oba, 2003). Few studies have presented quantitative mea-
sures of both biodiversity and social perspectives (Liu et al., 2007).
Furthermore, different communities within the same indigenous
culture may differ in their level of TEK. The conservation outcomes
of TEK have seldom been compared among sites within a similar
culture in order to understand the effective role of TEK.
Dramatic economic development and urbanization has resulted
in the rapid livelihood change and the loss of traditional values in
many indigenous groups (Malhotra et al., 2001; Worldwatch
Institute, 2010). The loss of TEK and the decline in traditional
practices have coincided with evidence of serious habitat
0006-3207/$ - see front matter  2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biocon.2011.10.027
? Corresponding author at: Smithsonian Conservation Biology Institute, National
Zoological Park, 1500 Remount Rd., Front Royal, VA 22630, USA. Tel.: +1 540 635
0037; fax: +1 540 635 0085.
E-mail address: xlshen.pku@gmail.com (X. Shen).
Biological Conservation 145 (2012) 160?170
Contents lists available at SciVerse ScienceDirect
Biological Conservation
journal homepage: www.elsevier .com/locate /b iocon
Author's personal copy
degradationandbiodiversity loss (Sutherland, 2003). Preserving tra-
ditional knowledge and respecting the customary use of biological
resources by indigenous people have been recognized as important
solutions for preventing biodiversity loss (Rands et al., 2011). Some
scholars also argue integration of traditional with scienti?c ecologi-
cal knowledge (SEK, or so called ??western science??) can increase the
resilience of the social?ecological systems by combining adapta-
tions from different cultural traditions (Begossi, 1998). Studies have
shown that a combination of TEK and SEK can strengthen conserva-
tion planning, and assist co-management for sustainable resource
use (Becker and Ghimire, 2003; Drew, 2005; Moller et al., 2004),
but therehavebeen fewattempts todemonstratehowan interaction
of TEK and SEK impacts biodiversity.
Traditional Tibetan culture contains alternative knowledge and
perspectives that contribute to the conservation of both wild fauna
and ?ora, and their habitats and ecosystems (Anderson et al., 2005;
Salick et al., 2007). Shaped by Tibetan Buddhism, which is a
combination of Bon (a pre-Buddhist religion in Tibet) and Bud-
dhism, Tibetans hold a world view in which humans are part of
an interacting set of living beings and being kind to all other crea-
tures will gain good karma for one self (Feng, 2005; He, 2005).
These cultural values promote a harmonious relationship between
humans and nature by respecting and protecting all life beings. Key
elements of Tibetan cultural traditions that could facilitate conser-
vation include: ahimsa (no killing), encouraging care of the wildlife,
and the worship and protection of the sacred sites (the home of
deities, natural spirits and spiritual leaders) (Nan, 2001). The
importance of traditional Tibetan culture in conservation has been
recognized by scholars and conservation NGOs in China, but has
yet to be adopted by the Chinese government.
Tibetan communities differ in following their traditions due to
historical events and variable exposure to commercialization (Xu
et al., 2005). It is unknown how the pattern of biodiversity is cor-
related with the extent of the traditional practices in the Tibetan
region. In recent years, local Tibetan people?s attitudes and behav-
iors towards conservation have been in?uenced by the govern-
ment?s laws and regulations through formal education, including
both school education and environmental education activities con-
ducted by local government agencies (e.g., forestry bureaus and
environment bureaus). Extensive ?nancial and public resources
have been devoted to the basic education of youth in the Tibetan
area by the Chinese government during the last decades (Wu,
2001). There is a lack of understanding on how the traditional
and of?cial conservation paradigms interact and motivate rural
residents to undertake actions that protect wildlife and their hab-
itats, and whether these paradigms complement conservation
measures in the region.
We present a quantitative study on the relationship between
traditional practices and biodiversity and how traditional Tibetan
culture (TEK) and formal education (SEK) affect biodiversity by
changing people?s conservation attitudes and behaviors in a Tibetan
region. From2005 to 2007, we used proxy questions tomeasure tra-
ditional practices, TEK, SEK, conservation attitudes and behaviors of
residents, andmeasured the bird community richness and diversity
in nine Tibetan villages with different levels of traditional practices.
Our hypothesiswas that both TEK and SEK have a positive impact on
the attitudes of the villagers, which will be re?ected in the conser-
vation behaviors of individuals, and therefore have a measurable
impact on biodiversity measures at a landscape level (Fig. 1). We
discuss whether the traditional and conventional education could
be complementary and integrated in order to promote regional
environmental awareness and biodiversity outcomes.
2. Material and methods
2.1. Study area
Our study area was located in the eastern Tibetan plateau,
including Ganzi Tibetan Autonomous Prefecture (hereafter called
Ganzi) and Aba Tibetan-Qiang Autonomous Precture (hereafter
called Aba) in Sichuan Province (Fig. 2). We identi?ed nine Tibetan
villages across this region as our study sites (Fig. 2). These villages
were selected on the basis of similar pro?les on geography, vegeta-
tion and social-economic background, but different levels of tradi-
tional practices (Table 1). All villages spanned an elevation
between 3000?4000 m and exist within a similar landscape, includ-
ing buckwheat/corn farmland, coniferous forest, sub-alpine oak/
shrub, and alpine meadow (Zhang, 1997). All villages were similar
in size and had similar household numbers (range 46?57), except
for one village that had 74 households. Villagers in these nine vil-
lages subsist primarily on grain harvesting and livestock raising,
while obtaining additional income from collecting/selling natural
products and temporary employment in regional towns (Sichuan
Bureau of Statistics, 2006). There had been no active industries or
any other development projects around the villages. Residents of
the study sites are all Tibetans except for a few Han people in three
villages (N = 1 in Tangqiao (percentage Han = 0.4%), 3 in Jialazong
(1.3%), and 11 in Zhatuo (5.1%)) (Fig. 2).
2.2. Estimation on traditional practices and conservation
knowledge?attitudes?behaviors
At each village, we conducted face-to-face interviews with local
villagers, and used a predesigned questionnaire to estimate the
Fig. 1. A ?ow chart of how traditional practices and formal education link to local biodiversity. Traditional ecological knowledge (Berkes et al., 2000) is part of traditional
knowledge, which correlates to traditional practice measures. SEK is determined by formal education. Both TEK and SEK will affect the conservation attitudes of the villagers,
which furthermore affect the individual behaviors towards conservation. Conservation behaviors contribute to biodiversity measures at a landscape level through direct
protection on wildlife population and indirect contribution on conserving their habitat.
X. Shen et al. / Biological Conservation 145 (2012) 160?170 161
Author's personal copy
level of traditional practices and conservation knowledge?atti-
tudes?behaviors. We strati?ed the interview sampling by the age
and gender of the villagers. At each village we chose 6?8 interview-
ers in each age category:<26, 26?35, 36?45, 46?55, >55, and the ra-
tio of male: female in each category was controlled approximately
as 1:1. Interviews were conducted by the researchers and two
assistants, two of which are Tibetans. Questions were asked in
Mandarin, when the interviewers were conversant in the language;
otherwise interviews were conducted in Tibetan.
We tested the draft questionnaire in May?September 2005
through discussion with local experts on Tibetan culture and with
more than 60 villagers at four villages to identify potential prob-
lematic questions or ambiguous statements. The ?nal question-
naire comprised 29 questions organized in six sections: (1)
traditional practices, (2) TEK, (3) SEK, (4) conservation attitudes,
(5) conservation behaviors and (6) socio-demographic background
(Table A.1).
With regards to traditional practices, we asked villagers eight
questions (Questions 1?8 in Table A.1) that re?ected the traditional
world view and practices in their daily life but do not touch di-
rectly on their attitudes toward animals and habitats. The eight
questions were identi?ed through literature research (Childs,
2003; Ma and Chen, 2005; Mills, 2002; Norbu Chophel, 1983; Yu,
1997), interviews with Tibetologists and local religious leaders,
and an associated anthropological study (Chen, unpublished data).
We included different questions as proxy measures to quantify the
villager?s awareness of TEK and SEK, as well as their conservation
attitudes and behaviors (Table 2).
The question ??Why do you think you can?t hunt wildlife, such as
musk deer or white-eared pheasant??? was used to detect whether
the respondents were in?uenced by formal education or
traditions/religion towards conservation. The answers were sum-
marized into two categories: (1) for the reason of government laws
(such as because ??they were national protected animals??, ??afraid of
the penalty for killing protected animals??, or ??hunting wildlife was
prohibited after government con?scated the guns); and (2) for the
reason of traditions/religion (such as because ??killing any kinds of
life was not allowed to a Buddhist??, ??not good to yourself??, or
Table 1
Ecological and socioeconomic background of the nine surveyed Tibetan villages in Sichuan Province, China.
Villagea Elevation
range (m)
Number of
households
Distance to
major roadb
Average traditional
practice index
Average years of formal
education
Percentage who had completed
elementary school (%)
Jiuza (JZ) 2800?4000 46 Far 0.18 4.1 30.9
Dongma
(DM)
2800?3800 57 Medium 0.65 3.5 34.2
Jialazong
(JLZ)
3300?4100 47 Close 0.67 2.3 25.0
Zhatuo
(ZT)
3100?4000 46 Close 0.72 2.3 17.5
Riji (RJ) 3000?4200 74 Close 0.87 3.4 30.7
Tangqiao
(TQ)
3100?4200 48 Far 0.88 1.0 6.2
Liuba (LB) 3500?4200 51 Far 0.93 2.1 30.0
Quesuo
(QS)
3300?4100 51 Close 0.94 2.3 13.3
Xiadecha
(XDC)
3600?4400 46 Far 0.96 1.6 13.2
a Initial of each village was listed in the parentheses.
b Distance categories: ??close??, when the major road went through the village; ??medium??, when the village was not crossed by the major road and <5 km from the major
road; and ??far??, when the village was >5 km from the major road.
Fig. 2. The nine Tibetan villages surveyed in Sichuan Province, China. Village names: (1) Jiuza, (2) Quesuo, (3) Jialazong, (4) Dongma, (5) Riji, (6) Tangqiao, (7) Xiadecha, (8)
Liuba, and (9) Zhatuo.
162 X. Shen et al. / Biological Conservation 145 (2012) 160?170
Author's personal copy
??prohibited by the Rinpoche and the monastery??). Villagers who
provide both as their answers were counted equally in two
categories.
The socio-demographic information included age, gender, eth-
nicity, religion, years of formal education, occupation, income,
experience of employment outside of the village, and ?uency in
spoken Mandarin.
2.3. Bird survey
Point countswereused to recordbird speciesoccurring in the two
major habitat types at eachvillage: 30points in conifer forest (there-
after called conifer), 30 in alpine oak and shrub (thereafter called
shrub). Two teamsof experienced investigatorswalkedaway thevil-
lage along different trails eachday, and set thepointswhen theymet
suitable habitat on their trail. The survey lasted until 30 points were
sampled in each habitat. Survey points were set >200 m apart for
sampling independence. At each point, all birds sighted and heard
withina50 mradiuswere recordedduringa10 min samplingperiod
(Hagan et al., 1997), while birds ?ying above the points were not
counted. The movements of the birds were tracked to minimize
duplicate counting. The sample sizes for both habitat types were
determined by examining the species accumulation curves derived
from our pilot study (Quesuo village, August 2006), and were a
compromisebetweendetecting rare species andreducing?eldeffort
to allow multiple sites to be sampled. The species accumulation
curves reached asymptotes with about 25 points for both habitat
types. The ?eld survey was mainly conducted 6:00?10:00 and
16:00?18:00 during the post-breeding season (August?mid-Sep-
tember) in 2006 in nine villages, and repeated during the breeding
season (May?June) in 2007 in ?ve of the nine villages.
2.4. Data analysis
For questions about traditional practices, all answers were
coded as binary variables, using 1 for ?Yes? and 0 for ?No?. The
summed score of answers to the eight questions (range 0?8) was
used as a traditional practice index (TPI) of individual respondents.
We considered respondents with higher TPI held more traditional
values and lived in a more traditional life. Years of formal educa-
tion was used as the index of formal education. In regards to
conservation knowledge, attitudes and behaviors, positive answers
were coded as plus score, answers of ?Don?t know? as neutral
(0 score) and negative answers as a minus score (see ??Measure??
in Table 2). Summed scores were calculated for the questions in
each section (i.e., TEK, SEK, attitudes and behaviors). The scores
of the two questions of SEK, scienti?c conservation terms and na-
tional protected animals, were calculated separately (Table 2).
We considered the higher scores to designate respondents that
were more knowledgeable and held more favorable attitudes and
behaviors towards conservation. We excluded female respondents
from question 8 (Table 2), as women in local communities were
not involved in patrolling activities.
We divided the respondents into three groups by their TPI value
based on a posthoc examination of their responses: low (TPI = 0?2),
medium (TPI = 3?5) and high (TPI = 6?8). We then used Mann?
Whitney test to compare the formal education, TEK, SEK, conserva-
tion attitudes and behaviors among the three groups. We used the
mean TPI score of all respondents in one village as village TPI, and
used a Pearson correlation test to examine the correlations
between the village TPI and the percentage villagers who reported
not hunting animals for the reason of government laws or
traditions/religion.
To explore the relationship between TPI and bird diversity, we
calculated bird species richness, abundance (number of individu-
als) and Shannon?Wiener Index of both conifer and shrub habitats
from the point count data of the post-breeding season for the nine
villages. We did not consider detection probability as our analysis
focused on relative differences in species richness and abundance
within each habitat type across different villages and not on abso-
lute estimates of these parameters. We used a Pearson correlation
test to examine the correlations between village TPI and species
richness, abundance and Shannon?Wiener Index of conifer and
shrub habitats at the village level. Point count data of the breeding
season were not included for this analysis as the bird survey was
not repeated in all the nine villages.
We used all the point count data of nine villages from two sur-
vey seasons for the regression to detect whether TPI and formal
education are signi?cant factors to the bird richness after control-
ling for other potential sampling and environmental factors. We
conducted a negative binomial regression for the 889 surveyed
points. We used species richness as the dependent variable, and in-
cluded 12 potential co-variables in the model: village TPI, mean va-
lue of years of formal education of the respondents in the village,
season (breeding or post-breeding), time (6:00?10:00, 10:00?
14:00, 14:00?18:00), elevation, habitat type (conifer or shrub),
Table 2
Proxy questions to measure conservation knowledge?attitudes?behaviors and scores allocated to possible answers.
Questions Measure
Traditional ecological knowledge TEK [0, 2]*
1. Do you believe that violating taboos on sacred mountains will bring harm to yourself? 1-Yes; 0-no or do not know sacred mountains or any taboos
2. Is there any difference between killing a big animal and a small animal? 1-Yes; 0-no or do not know
Scienti?c ecological knowledge SEK
Scienti?c conservation terms [0, 6]
3. Have you ever heard of: (1) class-I national protected animals; (2) law on the protection of
wildlife; (3) nature reserve; (4) forestry department; (5) forest protection and ?re prevention;
(6) biodiversity?
1-Yes; 0-no. The answers of 6 terms make a total score ranging
from 0 to 6
National protected animals [0, 5]
4. Are the following animals listed as nationally protected: (1) musk deer; (2) white-eared
pheasant; (3) large-billed crow; (4) mice; (5) blue sheep?
1-Correct answer; 0-do not know or wrong answer. The answers of
5 animals make a total score ranging from 0 to 5
Conservation attitudes [3, 3]
5. What would you do if you saw someone illegally cutting trees? 1-Stop it; 0-do not know; ??1??-do not care or leave it
6. What would you do if wild animals damaged your crops? 1-Non-lethal methods; 0-do not know; ??1??-lethal methods
7. What would you do if you met a wounded wild animal? 1-Rescue the animal; 0-do not know; ??1?? = do not care or eat it
Conservation behaviors [0, 1]
8. Have you ever taken part in the patrolling activities of your village? 1-Yes; 0-no or do not know there are patrolling activities
* Range of the scores summed from associated questions.
X. Shen et al. / Biological Conservation 145 (2012) 160?170 163
Author's personal copy
distance to village, distance to river, distance to monastery, slope,
aspect (warm or cold), and distance of the village to major road
(close, medium, far, Table 1). We included distance of the village
to major road as a surrogate measure of remoteness. All models
were examined and selected according to model AIC values (Akaike
Information Criterion. Akaike, 1973). We ranked the model with
the lowest AIC as the best model, and used differences in AIC be-
tween the best model and other candidate models (DAIC) to deter-
mine the relative ranking order. All models whose DAIC < 2 were
considered as equivalent best models (Burnharm and Anderson,
2002). We ran a ?nal negative binomial regression with the vari-
ables with most occurrences in all the likely models. We consid-
ered the variables in the negative binomial regression model
with a p < 0.05 as factors that signi?cantly affected the bird rich-
ness. We used SPSS 15.0 (SPSS, CA, USA) with a signi?cance level
of 0.05 for all statistical analyses.
3. Results
3.1. Socio-demographic characteristics
We completed 331 questionnaires in the nine villages, among
which 328 were considered valid (N = 328, rejection rate 1.0%).
Three questionnaires were excluded as the interviewees were not
local residents.
Respondents had a mean age of 41.0 years (SD = 15.2) and 52.0%
were male. There was no signi?cant difference in sample sizes of
different age categories (v2 = 1.726, df = 4, p = 0.786). All the
respondentswere Tibetans and 84.6% reported ??absolute faith in Ti-
betan Buddhism??, 3.4% ??partial faith in Tibetan Buddhism?? (i.e.,
practicing some, but not all, tenets of Tibetan Buddhism) and 12%
??little faith in Tibetan Buddhism??. The average year of formal school
educationwas 2.6, and 48.3% of the respondents did not receive any
school education, 42.9% had 66 years of formal education (elemen-
tary school), and 8.8% went to middle school, high school or college.
Themajority of the respondents (86.3%) were farmers or herdsman;
others were students, lamas, workers or of?cers.
3.2. Traditional practices and conservation knowledge-attitudes-
behaviors
Villagers with higher TPI received fewer years of formal educa-
tion (Fig. 3a). Older villagers practiced more traditions and re-
ceived less formal education (age  TPI, r = 0.136, p = 0.015;
age  years of formal education, r = 0.271, p < 0.001). There was
no difference between men and women in traditional practices
(Mann?Whitney U, Z = 0.908, p = 0.364), although men received
more formal education than women (Z = 3.386, p = 0.001).
All the measured conservation variables (i.e., TEK, SEK, and
conservation attitudes and behaviors) differed signi?cantly among
individualswith different TPI scores (Fig. 3). The villagerswith high-
er TPI scores were more aware of TEK, but less aware of scienti?c
conservation terms and had a lower knowledge of national
protected animals (Fig. 3b?d). With regard to conservation
attitudes, villagers with high TPI scores had more positive attitudes
towards conservation and more actively participated in conserva-
tion than villagers with low TPI scores, and villagers with medium
TPI scores were the least concerned about or participated in conser-
vation activities (Fig. 3e and f).
Fig. 3. Correlation between traditional practice index and formal education, traditional ecological knowledge, scienti?c ecological knowledge (scienti?c conservation terms
and national protected animals respectively), conservation attitudes and behaviors in nine survey Tibetan villages, Sichuan Province, China. Mann?Whitney test was used.
Differences among the three groups: low (traditional practice index, TPI = 0?2), medium (3?5) and high (6?8) are shown: 0.01 < p < 0.05; 0.001 < p < 0.01; p < 0.001.
164 X. Shen et al. / Biological Conservation 145 (2012) 160?170
Author's personal copy
Regarding whether the respondents? conservation attitudes
were in?uenced by formal education or traditions/religion, we
found village mean TPI score had a negative correlation with the
percentage of respondents in the village who reported hunting
wildlife was not allowed for the reason of government laws
(r = 0.934, df = 9, p < 0 .001), but a positive correlation with the
percentage of respondents who reported for the reason of
traditions/religion (r = 0.978, df = 9, p < 0.001). Villages with med-
ium mean TPI score had a roughly equal percentage of villagers
who reported not killing wildlife for the reason of government laws
and traditions/religion.
Patrolling activities were recorded in all the nine surveyed vil-
lages as the primary conservation response to threats to local wild-
life and nature resources (e.g., wildlife poaching, timber logging
and herb collecting). These patrolling activities were organized
and funded by four different parties: (1) government agencies,
(2) monasteries and religious leaders, (3) villagers and/or (4) out-
side conservation NGOs. Patrolling activities led by the local for-
estry department existed in all nine villages, during which men
were hired as forest guards and conducted regular patrolling to
protect forests and wildlife, and to report bush?re. Patrols con-
ducted by monasteries and/or villagers existed in six villages
(Fig. 4b). These two forms of community patrols were informally
organized, mainly for protecting the sacred mountains and as a
reaction to poaching threats in and around the village from outsid-
ers. Monks or villagers walked around in their spare time checking
for animal snares set by poachers. When local villagers encoun-
tered violators, or their signs, during their daily activities, they re-
ported to the monastery and a search was organized. The
percentage of respondents who knew about these activities varied
among villages. In the villages with highest TPI scores, most villag-
ers knew about the patrolling, while the least number of people
were knowledgeable within the villages with medium TPI scores
(Fig. 4a). The composition of patrols in the more traditional villages
was more diverse, with higher involvement of community mem-
bers and monasteries (Fig. 4b).
The percentage of the respondentswho had sworn in front of the
Rinpoche not to hunt (Question 20 in Table A.1) was positively cor-
related with the village TPI (r = 0.808, df = 9, p = 0 .008). No villagers
had pledged against hunting in the two least traditional villages.
3.3. Correlation between bird diversity and traditional practices
We recorded 128 bird species in the nine villages with a sam-
pling effort of 889 survey points. We recorded more species and
individuals of pheasants and raptors in the traditional villages (Ta-
ble 3). Species richness, and the Shannon?Wiener index, of both
the conifer and shrub habitats, and the number of individual birds
detected in the shrub habitat, had a signi?cant positive correlation
with the village TPI (Fig. 5). The relationship between diversity
measures and TPI was maintained when the village with the lowest
TPI, which might have dominated the results, was removed (Coni-
fer: species richness  TPI, r = 0.754, p = 0.031; Shrub: species rich-
ness  TPI, r = 0.896, p = 0.003, Shannon?Wiener index  TPI,
r = 0.762, p = 0.028).
We selected the following variables for the ?nal model: village
TPI, slope, aspect and survey time (Table A.2). All four variables sig-
ni?cantly contributed to the variability in species richness at the
survey point (Table 4). We detected more bird species at points
near villages with higher TPI scores. Sampling and environmental
factors were also important for predicting species richness at
points, as cold aspects exhibited 1.27 times the bird species of
warm aspects; and more species was recorded in the morning than
later in the day.
Table 3
Bird information of the post-breeding season of the nine surveyed Tibetan villages in Sichuan Province, China. Villages are presented in order of their TPI scores.
Village Average
traditional
practice index
No. of
species
detected
Average no. of
individuals per point
count
Species and individuals (number in the parenthesis) of pheasants and raptors recorded
Jiuza 0.18 39 4.0 n/a
Dongma 0.65 41 4.2 Northern Goshawk (Accipiter gentilis, 2)
Jialazong 0.67 44 5.8 n/a
Zhatuo 0.72 46 3.4 n/a
Riji 0.87 49 5.7 White Eared-pheasant (Crossoptilon crossoptilon, 6), Koklass Pheasant (Pucrasia macrolopha,
1)
Tangqiao 0.88 60 7.7 Blood Pheasant (Ithaginis cruentus, 2), White Eared-pheasant (3)
Liuba 0.93 55 5.2 Blood Pheasant (7), Common Buzzard (Buteo buteo, 1)
Quesuo 0.94 51 7.1 Blood Pheasant (13), Chinese grouse (Bonasa sewerzowi, 1), Black-eared Kite (Milvus
migrans, 1), Himalayan Vulture (Gyps himalayensis, 1), Northern Goshawk (1)
Xiadecha 0.96 58 7.4 White Eared-pheasant (15), Upland Buzzard (Buteo hemilasius, 1)
0%
20%
40%
60%
80%
100%
JZ DM JLZ ZT RJ TQ LB QS XDC
Pe
rc
en
t v
ill
ag
er
s w
ho
 k
ne
w
 
th
e 
pa
tr
o
lli
ng
 a
ct
iv
ite
s 
Villages
0%
20%
40%
60%
80%
100%
JZ DM JLZ ZT RJ TQ LB QS XDCPe
rc
en
t v
ill
ag
er
s w
ho
 
re
po
rt
ed
 th
e 
ca
te
go
ri
es
Villages 
Government agencies
Monasteries and religious leaders
Villagers
NGOs
(a)
(b)
Fig. 4. Percent villagers who knew the patrolling activities in the village and
percent villagers who reported the patrolling activities organized by four different
parties. Villages on the X axis were ranked by increasing traditional practice index.
X. Shen et al. / Biological Conservation 145 (2012) 160?170 165
Author's personal copy
4. Discussion
Previous empirical studies that examined the relationship be-
tween TEK and biodiversity often lack quantitative measurements
of both biodiversity and social perspectives. We chose birds as a
surrogate measure of biodiversity, as they are susceptible to land-
scape-level changes in the environment and relatively easy to
monitor (Blair, 1999). Measuring bird diversity was straightfor-
ward in our study, as standard protocols exist, but the quanti?ca-
tion of traditional practices was more challenging. Our proxy
questions for TPI were developed based on literature review and
tested during a trial period. We believed the TPI scores are an
appropriate measure representing the actual traditional status of
each village, which were also consistent with our subjective rank-
ing. The formal education level was low in the study area, thus
most villagers did not possess standard SEK. Instead, conservation
education by the government was the predominant form of knowl-
edge transfer in the local communities. Our SEK questions mea-
sured the villager?s comprehension of the policy statements
received from the government and were the local equivalent of
SEK. Possibly future increasing formal schooling will promote vil-
lager?s understanding of scienti?c ecological knowledge, but at
the time of this study government policy statements were the ex-
tent of their formal conservation knowledge.
As expected, villagers with more traditional practices were
more knowledgeable about TEK while less knowledgeable about
SEK. We consider this difference is due to how the villagers gained
their conservation knowledge: either primarily through commu-
nity education or through formal education. Traditional villagers
were more in?uenced by community regulations and religious
education while isolated from formal education. This dichotomy
was also supported by the answers to the question ??Why do you
think you cannot hunt any wildlife, such as musk deer or white-
eared pheasant??? which showed traditional villages had a high
percentage of people who followed the religious tenets against
hunting and a low percentage of people who followed the
government laws.
The attitude?behavior model is widely used in psychological
studies to interpret the complex mechanisms behind observed so-
cial activities (Ajzen and Fishbein, 1977; Kaiser et al., 1999) and
has been introduced to conservation biology researches as a
knowledge?attitudes?behaviors framework (Barney et al., 2005;
Kruse and Card, 2004). We adopted the knowledge?attitudes?
behaviors framework to explore how and to what extent TEK and
SEK determine local people?s behavior towards conservation. The
conservation attitudes and behaviors of the villagers re?ected a
combined in?uence of both TEK and SEK. The observed U-shape
Conifer habitat 
Shrub habitat 
Fig. 5. Correlation between bird diversity indices (species richness, number of individuals and Shannon?Wiener index) within the conifer and shrub habitats and traditional
practice index in nine survey Tibetan villages, Sichuan Province, China.
Table 4
Parameter estimates of variables in ?nal negative binomial regression model used to
predict the bird species richness.
Variables b S.E. P Exp. (b)
Traditional practice index 0.424 0.1285 0.001 1.53
Slope 0.009 0.0035 0.008 0.99
Aspect: Warma
Cold 0.239 0.0640 0.000 1.27
Time: 6 am?10 am a
10 am?2 pm 0.339 0.0744 0.000 0.71
2 pm?6 pm 0.484 0.0890 0.000 0.62
a The reference category of the variable.
166 X. Shen et al. / Biological Conservation 145 (2012) 160?170
Author's personal copy
relationship between individual TPIs and the measurements of
conservation attitudes and behaviors (Fig 3e and f) suggests that
conservation education by the government was successful in
raising people?s environmental awareness, but traditional ways
were comparatively more profound in promoting people?s conser-
vation attitudes. Villagers at the bottom of the U-shape curve were
neither strongly in?uenced by the traditions nor by the govern-
ment and their education should be considered a priority in future
conservation efforts. The conservation behaviors at the village level
showed similar trend as that of the individual level (Fig. 4).
In regards to bird diversity, we found a linear relationship be-
tween village mean TPI scores and species richness, abundance
and Shannon?Wiener index. Our regression analysis showed simi-
lar results: traditional practice was a positive correlative factor of
bird diversity, while formal education was not. We chose villages
with similar geography, vegetation and social-economic status to
control for environmental and social factors that could in?uence
bird communities. The land use of each village was similar with
farming, grazing, logging for fuel and housing materials, and col-
lecting non-forest product as the main human activities. We in-
cluded other anthropogenic factors, such as distances from roads
and monasteries, as covariates, but these factors were not selected
for our ?nal models. We consider the correlation between the tra-
ditional practices and bird diversity signi?cant, and these differ-
ences in bird diversity were a result of non-landscape attributes
within the villages that we attribute to differences in the villagers?
behaviors toward wildlife.
Traditional Tibetan culture has profound impacts on local peo-
ple?s attitudes and behaviors toward the protection of habitats
and wildlife mainly through two aspects: protecting sacred sites
and prohibiting hunting. We found a higher percentage of villagers
who understood the taboos and had pledged against hunting in the
traditional villages. Every Tibetan village has its sacred mountain
where hunting, logging and farming are forbidden. Strict taboos of-
ten exist for core areas of sacred mountains, such as prohibition of
livestock grazing and non-timber forest product collecting. These
practices often have real impacts, for example more mature forests
and endemic plant species are found on the sacred mountains than
non-sacred areas (Anderson et al., 2005; Salick et al., 2007). We
consider patrolling activities by the community an indicator of
the extent of local conservation efforts. Patrolling activity with
higher awareness and broader participation (Fig. 4) would be more
effective in preventing poaching. As a result, economically valuable
species, such as pheasants and raptors, were more often detected
in the villages with stronger patrolling activities and these tended
to be more traditional villages.
Although villagers in the lowest TPI villages had a higher score
in conservation attitudes and behaviors compared to the medium
TPI villages, these villages did not possess higher bird abundance
and diversity. Possibly this inconsistency was due to the insuf?-
cient time of government conservation efforts. Most (50 out of
51) nature reserves in Ganzi were established after 1995 (MEP,
2010). Government regulation did not reach these villages until
the late 1990s, with the logging ban in 1998, the prohibition of
guns around 2000, and conservation education by government
agencies (e.g., local forestry bureau and environmental protection
bureau). Loss of TEK, and increases in government regulation, occur
in conjunction with many other societal changes, such as loss of
traditional livelihoods, changing economic status, in?ux of differ-
ent cultures and individuals, which may also impact biodiversity
measures in the region (Malhotra et al., 2001; Bhagwat and Rutte,
2006).
Another possible reason why traditional Tibetan practices were
more tightly linked to local biodiversity is that traditions might be
more effective at regulating and constraining people?s behavior
towards conservation than government laws and regulations
(Bhagwat and Rutte, 2006; Xu et al., 2005). The rationale, enforce-
ment and sanctioning mechanisms of traditional conservation
system remarkably differ from those of an of?cial conservation
system (Colding and Folke, 2001). Traditional ecological knowl-
edge exists as a knowledge-practice-beliefs complex passed down
within the community through generations (Berkes et al., 2000).
The practice of TEK is largely dependent on its cultural and institu-
tional context. In traditional Tibetan villages, villagers obey tradi-
tional rules and regulations as they believe doing so is good for
their own karma and the welfare of their community. Violations
of these customs will bring bad karma and punishment from the
deities and monasteries, and these violators will be looked down
by others within the community. Every community member is
not only the practitioner but the supervisor of their own behavior.
Secondly, lack of legitimacy of current environmental laws and
policies among villagers may limit the effectiveness of the formal
conservation system (Wilshusen et al., 2002). Although formally
educated villagers had a better understanding of conservation pol-
icies, they may put little effort into patrolling and other activities
that would uphold these policies due to the lack of incentives
and the con?icts that widely exist between the local livelihoods
and government conservation activities (Xu and Melick, 2007). A
wider participation in conservation (Fig. 4) and the self-enforced
system found within the traditional villages are lacking in the for-
mal conservation system, and may preclude a better conservation
outcome (i.e., increased biodiversity).
The inverse relationship between TPI and formal education, and
the correlation between bird diversity and TPI indicated the impor-
tance of Tibetan traditional practices in conservation, and scienti?c
knowledge as a complement rather than a substitute of TPI for con-
servation in the study area. The negative correlation between tradi-
tional and scienti?c knowledge does not suggest that those two are
incompatible. In fact, they havemuch in common since both knowl-
edge systems derive from the systematic observation of nature (Ber-
kes et al., 2000). Attempts and progress have beenmade in Tibet and
other parts of the world to integrate traditional and scienti?c eco-
logical knowledge in resourcemanagement and conservation (Arm-
strong et al., 2007; Becker and Ghimire, 2003; Fraser et al., 2006;Ma
and Basang Lhamo, 2009). In the case of our study area this integra-
tion has not been successful. The shift from TEK towards SEK among
our study sites has been reported in other traditional societies that
are under outside pressures, whether they be a capitalist economy
(Pretty et al., 2008), centralized political systems, or culturally-inap-
propriate education systems (Kothari, 2006).
Although traditional ecological knowledge and institutions
were more effective in conserving biodiversity than formal institu-
tions, they have been rarely recognized and involved in the conser-
vation planning and implementation in China (Xu and Melick,
2007). We suggest a complementary role for both Tibetan TEK
and government conservation efforts. Our most effective conserva-
tion tool is TEK and the maintenance of religious tenets toward
wildlife. These practices should be encouraged and supported, with
government assistance where possible. Secondly, it may be possi-
ble to increase SEK among Tibetan religious institutions by educat-
ing monks on ecological principles. Involving local religious
institutions and leaders in conservation education has been advo-
cated to enhance community participation in both collecting
knowledge about and protecting biodiversity (Sheikh, 2006). Local
communities are interested in formal conservation knowledge
which they could use to stop outside violations of their traditional
conservation practices (Ma and Basang Lhamo, 2009). Current
conservation education presented through government programs
in Tibetan area is more a policy statement than ecological educa-
tion. More emphasis on the ecological principles and disseminating
SEK through religious institutions can secure local conservation
activities, and help traditional communities to cope with the
X. Shen et al. / Biological Conservation 145 (2012) 160?170 167
Author's personal copy
changing environment (e.g., climate change and global warming)
(Becker and Ghimire, 2003; Zhang, 2006). Thirdly, government
conservation programs would bene?t by the support of local insti-
tutions to complement its limited funding and human resources
(Xie, 2004; Sheikh, 2006; Zhang, 2006; Xu and Melick, 2007).
??Right actions?? may have several underlying motivations that do
not all need to be based in modern science. There are caveats to
integrating these two systems, as it will require an initial effort
on regional agreements, ?nancial input, ethical guidelines and
law/policy framework, and has a high risk of failure (Mauro and
Hardison, 2000). Once established, however, the integrated man-
agement system would require low maintenance and few outside
input (Ma and Basang Lhamo, 2009; Tian, personal communi
cation).
Our study showed that both TEK and SEK support attitudes to-
ward conservation, and when the two systems were weak, biodi-
versity measures were low. Government education could serve as
an alternative way to guarantee people?s behavior towards conser-
vation in these areas, but reviving the cultural traditions and TEK
should be considered a primary conservation goal. Our study has
implications for the conservation of indigenous groups undergoing
extensive social and economic change and the loss of tradition.
When the ??social taboos?? (Colding and Folke, 2001) are weakening,
and government conservation efforts are missing or ineffective,
local communities are faced with habitat degradation and biodi-
versity loss. Assisting indigenous people in maintaining and prac-
ticing TEK is an urgent task. We recommend increased efforts to
promote the cultural traditions in the Tibetan area, especially
targeting the younger generation and local communities in?u-
enced by modern society. Those communities who are willing to
maintain their traditions should be encouraged. Conservation
training materials that appreciate the value of traditional Tibetan
culture could be embedded into formal education (Kimmerer,
2002) and strengthened both to inspire people?s pride on their
own tradition and to have a direct bene?t for biodiversity conser-
vation. Meanwhile, given the variable strength of traditional prac-
tices among communities across the vast Tibetan region, any policy
of uniformity will not be feasible (Ma, 2011). Policy makers and
conservation managers should respect local autonomy for practic-
ing their own conservation beliefs and practices and should not
consider policy pronouncements a replacement for local traditions.
Acknowledgements
This study was ?nancially supported by The Sacred Land Con-
servation Project of Conservation International that was funded
by Blue Moon Foundation. We thank Sichuan Forestry Department,
Ganzi Forestry Bureau, Aba Forestry Bureau for their administra-
tion support and consultation on the survey sites. We also thank
Tashi Dorje, Jangyong Pengtsuo, and Rinchen Sumdrup for the con-
sultation on Tibetan traditional knowledge. We especially appreci-
ate C. Wen, L. Zhang, K. Du, P. Que, G. Sun, Caiwang, B. Li, and all the
other forestry department staff and ?eld guides for their hard work
assisting with the data collection. We gratefully acknowledge
D. Wang, H. Wang, Q. Zhao, and other colleagues in Center for Nat-
ure and Society for their valuable comments on the subject. We are
grateful to the anonymous reviewers for their valuable and de-
tailed comments that greatly helped us to improve the manuscript.
Appendix A
See Tables A.1 and A.2.
Table A.1
Questionnaire on the traditional practices, conservation knowledge, attitudes and behaviors of villagers in nine survey Tibetan villages, Sichuan Province, China.
I. Traditional practices of the interviewee
1. Do you believe in reincarnation and karma?
2. Does pray or pilgrimage bene?t yourself?
3. Is there a Buddhist shrine in your house?
4. Are you wearing an amulet?
5. Do you or your family burn incense at festival or in good days?
6. Was your name, or that of your child, given by the lama?
7. Do you or your family pray every morning?
8. Do you invite the lama to select the day for getting married or building new houses?
II. Knowledge?attitudes?behaviors of the interviewee
9. Are there any sacred mountains in and around your village?
10. Do you believe violating taboos on sacred mountains will bring any harm to yourself?
11. Is there any difference between killing a big animal and a small animal?
12. Have you ever heard of: (1) class-I national protected animals; (2) law on the protection of wildlife; (3) nature reserve; (4) forest department; (5) forest protection
and ?re prevention; (6) biodiversity?
13. Are the following animals listed as nationally protected: (1) musk deer; (2) white-eared pheasant; (3) large-billed crow; (4) mice; (5) blue sheep?
14. What would you do if you saw someone illegally cutting trees?
15. What would you do if wild animals damaged your crops?
16. What would you do if you met a wounded wild animal?
17. Are there any patrolling activities going on in your village? If yes, who organizes the patrolling?
18. Have you ever taken part in the patrolling activities of your village?
III. Others
19. Why do you think you can not hunt wildlife, such as musk deer or white-eared pheasant?
20. Have you ever sworn in front of the Rinpoche not to hunt?
IV. Social ? demographic background of the interviewee
21. Age
22. Gender
23. Ethnicity
24. Religion
25. Years of formal education
26. Occupation
27. Income status in the village: rich; medium; poor (obtained from the village head)
28. Have you ever been employed outside of the village?
29. Is the interviewee ?uent in spoken Mandarin?
168 X. Shen et al. / Biological Conservation 145 (2012) 160?170
Author's personal copy
References
Ajzen, I., Fishbein, M., 1977. Attitude-behavior relations: a theoretical analysis and
review of empirical research. Psychological Bulletin 84, 888?918.
Akaike, H., 1973. Information theory and an extension of the maximum likelihood
principle. In: Proceeding of 2nd International Symposium on Information
Theory: Budapest, pp. 267?281.
Anderson, D.M., Salick, J., Moseley, R.K., Ou, X., 2005. Conserving the sacred
medicine mountains: a vegetation analysis of Tibetan sacred sites in northwest
Yunnan. Biodiversity and Conservation 14, 3065?3091.
Armstrong, M., Kimmerer, R.W., Vergun, J., 2007. Education and research
opportunities for traditional ecological knowledge. Frontiers in Ecology and
the Environment 5 (4), 12?14.
Barney, E.C., Mintzes, J.J., Yen, C.F., 2005. Assessing knowledge, attitudes, and
behavior toward charismatic megafauna: the case of dolphins. The Journal of
Environmental Education 36, 41?55.
Becker, C.D., Ghimire, K., 2003. Synergy between traditional ecological knowledge
and conservation science supports forest preservation in Ecuador. Conservation
Ecology, 8. <http://www.consecol.org/vol8/iss1/art1>.
Begossi, A., 1998. Resilience and neotraditional populations: The cnignras of the
Atlantic forest and caboclos of the Amazon (Brazil). In: Berkes, F., Folke, C. (Eds.),
Linking Social and Ecological Systems: Management Practices and Social
Mechanisms for Building Resilience. Cambridge University Press, Cambridge,
UK, pp. 129?157.
Berkes, F., Colding, J., Folke, C., 2000. Rediscovery of traditional ecological
knowledge as adaptive management. Ecological Applications 10, 1251?1262.
Bhagwat, S.A., Rutte, C., 2006. Sacred groves: potential for biodiversity
management. Frontiers in Ecology and the Environment 4, 519?524.
Bhagwat, S.A., Kushalappa, C.G., Williams, P.H., Brown, N.D., 2005. A landscape
approach to biodiversity conservation of sacred groves in the Western Ghats of
India. Conservation Biology 19, 1853?1862.
Blair, R.B., 1999. Birds and butter?ies along an urban gradient: surrogate taxa for
assessing biodiversity? Ecological Applications 9, 164?170.
Burnharm, K.P., Anderson, D.R., 2002. Model Selection and Multimodel Inference. A
Practical Information-Theoretic Approach, second ed. Spring, New York.
Castro, V., Aldunate, C., 2003. Sacred mountains in the highlands of the south-
central Andes. Mountain Research and Development 23, 39?73.
Childs, C., 2003. Names and nicknames in sKyid grong. Tibet Journal 28, 17?18.
Christopher, H., 2008. Biodiversity and sacred sites: vernacular conservation
practices in Northwest Yunnan, China. Worldviews: Global Religions, Culture,
and Ecology 12, 74?90.
Colding, J., Folke, C., 2001. Social taboos: ??invisible?? systems of local resource
management and biological conservation. Ecological Applications 11, 584?600.
Decher, J., 1997. Conservation, small mammals, and the future of sacred groves in
West Africa. Biodiversity and Conservation 6, 1007?1026.
Diemont, S.A.W., Martion, J.F., 2009. Lacandon Maya ecosystem management:
sustainable design for subsistence and environmental restoration. Ecological
Applications 19, 254?266.
Drew, J.A., 2005. Use of traditional ecological knowledge in marine conservation.
Conservation Biology 19, 1286?1293.
Feng, Z., 2005. The conservation perspective of Tibetan Buddhism. In: Ma, J.Z., Chen,
J. (Eds.), Tibetan Culture and Biodiversity Conservation. Yunnan Science and
Technology Press, Yunnan, China, pp. 83?89.
Fraser, D.J., Coon, T., Prince, M.R., Dion, R., Bernatchez, L., 2006. Integrating
traditional and evolutionary knowledge in biodiversity conservation: a
population level case study. Ecology and Society, 11. <http://
www.ecologyandsociety.org/vol11/iss2/art4/>.
Gadgil, M., Berkes, F., Folke, C., 1993. Indigenous knowledge for biodiversity
conservation. AMBIO 22, 151?156.
Hagan, J.M., McKinley, P.S., Meehan, A.L., Grove, S.L., 1997. Diversity and abundance
of landbirds in a northeastern industrial forest. The Journal of Wildlife
Management 61, 718?735.
He, J.H., 2005. The role of Tibetan Bon religion in conservation. In: Ma, J.Z., Chen, J.
(Eds.), Tibetan Culture and Biodiversity Conservation. Yunnan Science and
Technology Press, Yunnan, China, pp. 68?72.
Jamir, S.A., Pandey, H.N., 2003. Vascular plant diversity in the sacred groves of
Jaintia Hills in northeast India. Biodiversity and Conservation 12, 1497?1510.
Kaiser, F.G., Wol?ng, S., Fuhrer, U., 1999. Environmental attitude and ecological
behaviour. Journal of Environmental Psychology 19, 1?19.
Kimmerer, R.W., 2002. Weaving traditional ecological knowledge into biological
education: a call to action. BioScience 52 (5), 432?438.
Kothari, A., 2006. Community conserved areas: towards ecological and livelihood
security. In: Kothari, A. (Ed.), Community Conserved Areas. IUCN, Gland,
Switzerland, pp. 3?13 (pages 47?55).
Kruse, C.K., Card, J.A., 2004. Effects of a conservation education camp program on
campers? self-reported knowledge, attitude, and behavior. The Journal of
Environmental Education 35, 33?45.
Liu, J.G. et al., 2007. Complexity of coupled human and natural systems. Science 217,
1513?1516.
Loh, J., Harmon, D., 2005. A global index of biocultural diversity. Ecological
Indicators 5, 231?241.
Luo, Y., Liu, J., Zhang, D., 2009. Role of traditional beliefs of Baima Tibetans in
biodiversity conservation in China. Forest Ecology and Management 257, 1995?
2001.
Ma, R., 2011. Social development and bilingual education in Tibet. China Tibetology
2, 108?139.
Ma, J.Z., Chen, J., 2005. Tibetan Culture and Biodiversity Conservation. Yunnan
Science and Technology Press, Yunnan, China.
Ma, H.B., Lhamo, Basang., 2009. Conservation concession: a new conservation
approach in Sanjiangyuan. Qinghai Economic Research 2, 35?39.
Maf?, L., 2005. Linguistic, cultural, and biological diversity. Annual Review of
Anthropology 29, 599?617.
Malhotra, K.C., Gokhale, Y., Chatterjee, S., Srivastava S., 2001. Cultural and Ecological
Dimensions of Sacred Groves in India. Indian National Science Academy, New
Delhi & Indira Gandhi Rashtriya Manav Sangrahalaya, Bhopal.
Mauro, F., Hardison, P.D., 2000. Traditional knowledge of indigenous and local
communities: international debate and policy initiatives. Ecological Application
10, 1263?1269.
MEP (Ministry of Environmental Protection), 2010. Nature Reserves in Sichuan
Province 2009. Ministry of Environmental Protection, Beijing, China. <http://
sts.mep.gov.cn/zrbhq/zrbhq/201012/t20101220_198880.htm> (accessed
04.11).
Mgumia, F.H., Oba, A.G., 2003. Potential role of sacred groves in biodiversity
conservation in Tanzania. Environmental Conservation 30, 259?265.
Mills, M.A., 2002. Identity, Ritual and State in Tibetan Buddhism: The Foundations
of Authority in Gelukpa Monasticism. RoutledgeCurzon, London and New York.
Moller, H., Berkes, F., Lyver, P.O., Kislalioglu, M., 2004. Combining science and
traditional ecological knowledge: monitoring populations for co-management.
Ecology and Society, 9. <http://www.ecologyandsociety.org/vol9/iss3/art2/>.
Nan, W.Y., 2001. The taboos in Tibetan areas and the role they have played in
environmental protection. North West Minorities Research 3, 21?29.
Norbu Chophel, 1983. Folk Culture of Tibet. Library of Tibetan Works Archives,
Dharamsala.
Ntiamoa-Baidu, Y., 2008. Indigenous beliefs and biodiversity conservation: the
effectiveness of sacred groves, taboos and totems in Ghana for habitat and
species conservation. Journal for the Study of Religion, Nature & Culture 2, 309?
326.
Pretty, J., et al., 2008. How do biodiversity and culture intersect? In: Paper Presented
to Sustaining Cultural and Biological Diversity in a Rapidly Changing World:
Lessons for Global Policy Conference in April 2?5, 2008, New York.
Rands, M.R.W., Adams, W.M., Bennun, L., Butchart, S.H.M., Clements, A., 2011.
Culture and Biodiversity Losses Linked ? Response, 331, 31.
Salick, J., Amend, A., Anderson, D., Hoffmeister, K., Gunn, B., Fang, Z., 2007. Tibetan
sacred sites conserve old growth trees and cover in the east Himalayas.
Biodiversity and Conservation 16, 693?706.
Sheikh, K.M., 2006. Involving religious leaders in conservation education in the
Western Karakorum, Pakistan. Mountain Research and Development 26, 319?
322.
Sichuan Bureau of Statistics, 2006. Sichuan Statistical Yearbook. China Statistics
Press, Beijing.
Sutherland, W.J., 2003. Parallel extinction risk and global distribution of languages
and species. Nature 423, 276?279.
Turner, N.J., Ignace, M.B., Ignace, R., 2000. Traditional ecological knowledge and
wisdom of aboriginal peoples in British Columbia. Ecological Application 10,
1275?1287.
Wadley, R.L., Colfer, C.J.P., 2004. Sacred forest, hunting, and conservation in West
Kalimantan, Indonesia. Human Ecology 32, 313?338.
Wilshusen, P.R., Brechin, S.R., Fortwangler, C.L., West, P.C., 2002. Reinventing a
square wheel: critique of a resurgent ??Protection Paradigm?? in international
biodiversity conservation. Society and Natural Resources 15, 17?40.
Worldwatch Institute, 2010. State of the World 2010: Transforming Cultures, from
Consumerism to Sustainability, Washington, DC.
Wu, M.H., 2001. Some thinking on the Greater Western Development and education
modernization in Tibetan region. Journal of Research on Education for Ethnic
Minorities 12, 5?10.
Xie, Y., 2004. Review on the management system of China?s nature reserves. In: Xie,
Y., Wang, S., Schei, P. (Eds.), China?s Protected Areas. Tsinghua University Press,
Beijing, pp. 315?338.
Xu, J.C., Melick, D.R., 2007. Rethinking the effectiveness of public protected areas in
southwestern China. Conservation Biology 21, 318?328.
Xu, J., Ma, E.T., Tashi, D., Fu, Y., Lu, Z., Melick, D., 2005. Integrating sacred knowledge
for conservation: cultures and landscape in southwest China. Ecology and
Society 10, 7.
Table A.2
Candidate negative binomial models whose DAIC < 2 used to predict the bird species
richness in the nine surveyed Tibetan villages, Sichuan Province, China. Models are
ranked in order of their AIC.
Model K* AIC DAIC
TPI, time, aspect, slope 4 3739.147 0.000
TPI, time, aspect, slope, habitat 5 3739.905 0.758
TPI, time, aspect, slope, season 5 3739.916 0.769
TPI, time, aspect, slope, distance to village 5 3740.341 1.194
TPI, time, aspect, slope, season, habitat 6 3740.603 1.456
TPI, time, aspect, slope, season, distance to village 6 3741.020 1.873
TPI, time, aspect, slope, distance to village, habitat 6 3741.047 1.900
* K = number of estimable parameters in the model.
X. Shen et al. / Biological Conservation 145 (2012) 160?170 169
Author's personal copy
Yu, C.J., 1997. Community survey and research in Lara village. In: the Institute of
Sociology and Anthropology, Peking University and the China Tibetology
Research Center (Eds.), Studies of Tibetan Society and Development. China
Tibetology Publishing House, Beijing, China, pp. 387?468.
Zhang, R.Z., 1997. Physical Geography of Hengduan Mountains. Science Press,
Beijing.
Zhang, X.M., 2006. A research of current situation of Tibetan Buddhism in Sichuan
Tibetan area. North West Ethno-National Studies 4, 105?111.
170 X. Shen et al. / Biological Conservation 145 (2012) 160?170