Hunting, Livelihoods and Declining Wildlife in the HponkanraziWildlife Sanctuary, North Myanmar

Madhu Rao • Saw Htun • Than Zaw •

Than Myint

Received: 17 December 2009 / Accepted: 14 June 2010

� Springer Science+Business Media, LLC 2010

Abstract The Hponkanrazi Wildlife Sanctuary, North

Myanmar and three contiguous protected areas, comprise

some of the largest expanses of natural forest remaining in

the region. Demand for wildlife products has resulted in

unsustainable exploitation of commercially valuable spe-

cies resulting in local extirpation of vulnerable species.

Camera trap, track and sign, and questionnaire-based sur-

veys were used to examine (a) wildlife species targeted by

hunters, (b) the importance of wild meat for household

consumption, and (c) the significance of hunting as a

livelihood activity for resident villages. Certain commer-

cially valuable species highly preferred by hunters were

either completely absent from hunt records (tiger, musk

deer and otter) or infrequently obtained during actual hunts

(bear, pangolin). Species obtained by hunters were com-

monly occurring species such as muntjacs with low com-

mercial value and not highly preferred by hunters. Fifty

eight percent of respondents (n = 84) indicated trade, 27%

listed subsistence use and 14% listed human-wildlife con-

flict as the main reason for hunting (n = 84). Average

amount of wild meat consumed per month is not signifi-

cantly higher during the hunting season compared to the

planting season (paired t-test, P [ 0.05). Throughout the

year, the average amount of fish consumed per month was

higher than livestock or wild meat (Friedman test,

P \ 0.0001). Hunting is driven largely by trade and wild

meat, while not a critical source of food for a large number

of families could potentially be an important, indirect

source of access to food for hunting families. Findings and

trends from this study are potentially useful in helping

design effective conservation strategies to address globally

prevalent problems of declining wildlife populations and

dependent human communities. The study provides rec-

ommendations to reduce illegal hunting and protect vul-

nerable species by strengthening park management through

enforcement, increasing the opportunity costs of poaching,

establishing no-take zones and research to determine the

economic significance of hunting for livelihoods.

Keywords Hunting � Myanmar � Wildlife trade �Wild meat � Livelihoods � Species decline

Introduction

Hunting of wildlife in tropical Asian forests has occurred for

several thousand years (Corlett 2007). Human populations in

the region engage in hunting to meet a range of nutritional,

economic, medicinal, cultural and recreational needs. Wild

species in trade are valued for ‘‘wild meat’’, traditional

medicines, pets, trophies, decorations, and other utilitarian

purposes. Increasing prosperity, population growth and

enhanced access to urban markets have resulted in escalating

demand for wildlife products. Overexploitation of wildlife

for the commercial trade has resulted in significant declines

and local extinctions for several wildlife species (Nooren

and Claridge 2001; Bennett and others 2002; Anon 2005).

Given that subsistence hunters, traders or consumers depend

on wildlife harvest as a source of food and income, the

implications of reduced availability of wildlife due to

unsustainable consumption are potentially significant but

remain poorly studied and largely unclear (Bennett 2002).

M. Rao (&)

Wildlife Conservation Society, 2300 Southern Boulevard,

Bronx, New York, NY 10460, USA

e-mail: mrao@wcs.org

S. Htun � T. Zaw � T. Myint

Wildlife Conservation Society Myanmar Program, Yangon,

Myanmar

123

Environmental Management

DOI 10.1007/s00267-010-9519-x

Myanmar, the largest country in mainland South-East

Asia, with a land area of 676,553 km2 is known more for

its isolation and rigid military rule rather than its biodi-

versity. Wide variation in latitude and longitude make the

country biogeographically distinct with high levels of

habitat and species diversity (Tordoff and others 2005).

The country still contains some of the largest expanses of

natural forest cover remaining in the region (Leimgruber

and others 2005) and several globally threatened species.

However, species and their habitats both within and outside

protected area boundaries are at risk due to threats such as

hunting and the extraction of non-timber forest products

(Rao and others 2002, 2005; Lynam and others 2006).

Historically, hunting, together with habitat loss has been

largely responsible for the depletion of large mammals

such as elephants, gaur, Sumatran rhinoceros and tigers

from many protected areas in Myanmar (Milton and Estes

1963). Growing regional demand for wildlife products is

directly linked with the exploitation of commercially

valuable populations where they are still available to be

hunted (Shepherd 2001; Shepherd and Nijman 2007). This

study examines (a) hunting patterns including wild animal

species being targeted by hunters, (b) the importance of

wild meat for household consumption and (c) the signifi-

cance of hunting as a livelihood activity for villages within

the Hpongkhanrazi Wildlife Sanctuary.

Study Area

The Hponkanrazi Wildlife Sanctuary (HWS) (2760 km2)

gazetted in 2003 lies within Kachin State, North Myanmar

(278300N, 978430E) (Fig. 1). The sanctuary, together with

three other contiguous protected areas, Hkakaborazi

National Park, Bumphabum Wildlife Sanctuary and the

Hukaung Valley Tiger Reserve forms the Northern Forest

Complex covering an area of 30,269 km2, constituting some

of the largest contiguous expanses of natural forest

remaining in the region. HWS supports largely intact veg-

etation over a wide altitudinal range and includes sub-

tropical moist hill forest, temperate forest, deciduous forest

and alpine forest in the higher altitudes. Habitats especially

at high altitudes, are relatively stable (Renner and others

2007). Undergrowth in tropical moist forest habitats is

dominated by bamboo, rattan and wild bananas. In addition,

there are unspoilt riverine habitats supporting globally

threatened species such as White-bellied Heron (Ardea in-

signis), Blyth’s tragopon (Tragopon blythii) and Slater’s

monal (Lophophorus sclateri). The Northern Forest Com-

plex supports a number of animal species characteristic of

the eastern Himalayas, including Red Panda (Ailurus ful-

gens), Serow (Capricornis milneedwardsi), Takin (Budor-

cas taxicolor), as well as populations of Black Muntjac

(Muntiacus crinifrons) (Rabinowitz and others 1998) and

Leaf Deer (Muntiacus putaoensis) (Amato and others 1999).

Ten randomized sampling blocks (10 km 9 10 km)

were established excluding high elevation areas in the

northeastern parts of the Hponkanrazi Sanctuary (Fig. 1).

Within the sampling blocks, biological information related

to wildlife and habitat was obtained through camera trap

and track and sign surveys. Socioeconomic surveys were

conducted in thirteen villages within HWS (Table 1). All

data were collected during November 2004 and June 2005.

Camera Trap Surveys

Direct observations of wildlife are typically quite rare in

tropical forests and hence camera traps were deployed to

obtain information on the presence/absence of medium and

large -sized mammals. Five camera traps were deployed

randomly in each of the 10 km 9 10 km sampling plot.

Sites for camera traps were selected based on observations

of tracks and signs in the plot. There were a total of 50

camera trap positions dispersed across the 10 plots. Camera

traps were left for an average of 24 days per camera trap

position. Depending on topography which influenced

accessibility, the camera traps were set as evenly as pos-

sible across the sampling plots. However, given that

accessibility in some sample plots was extremely difficult,

even distribution of camera traps in all sample plots could

not be achieved. (For example, a large stream with strong

water currents flows through one of the sample plots and

forced the survey team to focus survey activities on the

accessible side of the stream). Moreover, thick climbing

bamboo brakes are difficult to pass through and constituted

additional factors influencing the distribution of camera

traps. Given these factors, we established and strictly fol-

lowed a rule allowing a minimum distance of 500 m

between camera traps in all sample plots.

Independent events for camera trap captures of indi-

vidual animal species were defined following O’Brien and

others (2003). The number of trap nights was calculated for

each camera location from the day the camera was

mounted until the day it was retrieved. Following camera

retrieval, films were developed, and we identified each

photo of an animal to species, and recorded time and date.

We scored each photo as an independent or non-indepen-

dent event based on the following definition of an inde-

pendent event: (1) consecutive photographs of different

individuals of the same or different species; (2) consecutive

photographs of individuals of the same species taken more

than 0.5 hours apart; (3) non-consecutive photos of indi-

viduals of the same species. Independent events, as defined

above, were used to estimate average capture rate per 100

trap nights for species across all 10 plots.

Environmental Management

123

Strip Transect Surveys

Strip transect surveys for direct encounters or tracks/signs

were used to estimate relative encounter rates for large

terrestrial mammal and bird species along transects. In

each of the 10 sampling plots, a site representing the

dominant habitat within the plot was selected in order to

establish a permanent strip transect 1000 m in length and

2 m in width. There were 10 permanent strip transects

established within the wildlife sanctuary. Each transect was

divided into five 200 m sub-sections and surveyed 3–5

times over the course of the study period. The strip widths

for recording wildlife sightings and tracks/signs were 20 m

and 2 m respectively. Direct observations of wildlife and

counts of tracks and signs (faeces, soil scrapes, signs of

tusks and antlers, vocalizations, burrows, ground clearing

(for pheasants Lophura and Tragopan sp.) and exposed soil

mounds (for porcupines Hystrix brachyura) were recorded

by observers walking along a transect at a speed of

approximately 1 km h-1. Surveyors recorded tracks and

Fig. 1 Location map

Environmental Management

123

signs along strip transects between 0600 and 1000 h or

between 1500 and 1800 h, based on the assumption that

diurnal wildlife were most active at these times.

The effectiveness of strip transect estimation method

can vary depending on different habitat types and substrate

as abundance of a particular species can vary depending on

types of substrate recorded. For example, thick moss car-

pets in Rhododendron and Fir Forest and thick roots in

Dwarf Bamboo Forests hinder clear prints of track and

sign, making it difficult to establish strip transects. To

accurately estimate relative abundance and monitor long-

term trends in abundance, repeated surveys during a par-

ticular time-frame are necessary. The frequency of visits

and surveys at this site were limited due to logistical

constraints and hence results were only used to confirm

relative abundance trends as estimated by camera trap

surveys.

Questionnaire Based Interview Surveys

Interview surveys were conducted at the village, household

and individual levels. At the village level, groups of five to

eight hunters (belonging to the Lisu and Rawan commu-

nities) in each of the 13 villages were invited to attend

meetings to obtain information on prices of wildlife parts

and trade routes (Table 1).

The socio-economic surveys at the household level were

targeted at all 270 households in the 13 villages. However,

the survey team consisting of two to three individuals

conducted questionnaire-based interview surveys privately

in 250 households in 13 villages (92% of all households) to

obtain information on the following variables: Wealth

status (possession of assets such as durable goods (for

example, sewing machines), livestock, transportation,

agricultural implements, and hunting and fishing equip-

ment); Quantities of fish, livestock and wild meat con-

sumed per month per household. We used the data to

obtain information on the nutritional importance of wild

meat, fish and livestock during the planting (April-July)

and the hunting (October- May) seasons and the overall

economic importance of hunting for resident local com-

munities. For a subset of the households (n = 84), we also

obtained information on the five most commonly hunted

species (based on recall data), and the five most preferred

species.

Interviews were conducted with 123 individual hunters

in seven villages following their return from a ‘recent’

hunt. These interviews were conducted the day they

returned from the hunt for 73% of the hunters (n = 90) and

for the remaining, the hunting event had occurred less than

12 days previously. We obtained information on duration

of the hunt, species obtained, weapons used for hunting,

main occupations, reasons for hunting, trade routes and the

source of highest income.

It is particularly challenging to make socio-economic

surveys non-intimidating to villagers, especially in a

country such as Myanmar. Meetings and discussions were

held in every village prior to the surveys in order to clarify

survey objectives and to explain the significance of local

community participation in conservation efforts. These

meetings and discussions were critical as trust-building

activities and were aimed to increase the likelihood of

truthful disclosures on hunting patterns.

However, we acknowledge the shortcomings of the

socio-economic data obtained through the interview sur-

veys and three potential sources of error in our survey

methods (i) Non-truthful disclosures by survey participants

(for example, commonly hunted species), (ii) errors asso-

ciated with recall data (for example amounts of wildmeat,

fish and livestock consumed per month) and (iii) insuffi-

cient replication of surveys (due to monetary and access

constraints) to eliminate bias due to seasonality.

Results

Camera Trap and Strip-Transect Surveys

Cameras deployed for 1199 trap nights recorded the pres-

ence of a number of globally threatened and rare species

such as Takin (Budorcas taxicolor), Asiatic black bear

(Ursus thibetanus),Clouded leopard (Neofelis nebulosa)

Musk deer (Moschus fuscus), Gaur (Bos gaurus), Blyth’s

tragopon (Tragopon blythii) and Sclater’s monal pheasant

(Lophophorus sclateri). A total of 115 independent events

Table 1 Villages, households and population size in the Hpongk-

hanrazi Wildlife Sanctuary

Village name Number of households Population size

Awaddam-1 29 191

Awaddam-2 8 47

Htohulaw 11 59

Karlan 28 193

Lonnadam 25 141

Namhtumkhu 63 395

Namru-1 11 61

Namru-2 11 53

Namruhtu 5 22

Namze 30 190

Wanglingdam 16 93

Warsandam 16 87

Ziyadam 17 107

Environmental Management

123

recorded 19 species that included two species of birds

(Kalij pheasant and the Grey peacock pheasant). Three

species of muntjacs (M. muntjak (common muntjak), M.

crinifrons (black muntjac) and M. putaoensis (leaf munt-

jac) were most frequently captured in the camera traps

relative to other mammals accounting for nearly 40 % of

all the independent captures. Independent events were used

to determine the capture rate of a species per 100 trap

nights as an index of relative abundance. Serow and wild

pig were found to be the second and third most commonly

captured species after muntjac.

Strip transects used to survey an area of 84,000 m2

yielded 387 records of direct observations, tracks and signs.

Direct observations were made for primates such as the

Assamese macaque, Hoolock gibbon and stump-tailed

macaque, pheasant, hornbill and squirrel with occasional

sightings of wild pig. Tracks were recorded for all three

species of muntjac, serow, takin and red goral. Scats were

recorded as sign for small carnivores such as red panda

(Ailurus fulgens), yellow-throated marten (Ailurus fulgens),

stripe-backed weasel (Mustela strigidorsa), masked palm

civet (Paguma larvata) and spotted linsang (Prionodon

linsang).

Relative frequencies of independent events (based on

camera trap surveys) and encounter rates (based on track

and sign surveys) were found to vary in the same direction

for the following species (Fig. 2): serow, wild pig, gaur,

yellow-throated marten, wild dog, porcupine, sambar and

the following groupings of species: muntjacs (leaf muntjac,

black muntjac and common muntjac), macaques (assamese

and stump-tailed), pheasants (Argus sp.), small carnivores

(see above) and bears (Himalayan black bear and Malayan

sun bear) (Kendall Tau = 0.338, P = 0.009).

Preferred Species in Decline

Household-level questionnaire surveys were used to

determine species most preferred by hunters and species

obtained during hunts. Bears were cited as top priority for

forty nine percent of hunters interviewed (n = 74) fol-

lowed by pangolins (11%) and sambar (9%). A comparison

of species obtained during actual hunts (n = 125 individual

animals) with preferred species (those indicated as highly

preferred by hunters) indicated the following: (i) Certain

species highly preferred by hunters such as bear, pangolin

and sambar were infrequently obtained by hunters (ii)

Certain species highly preferred by hunters such as tiger,

musk deer and otter presumably for their high commercial

value in trade were completely absent from hunter offtake

(as recalled by hunters). (iii) Although not highly preferred

by hunters, muntjac (relatively more abundant than other

species) were the most commonly obtained species. Other

commonly obtained species such as masked palm civet,

pheasant, partridge, squirrel and rat species were not listed

as highly preferred by hunters (Fig. 3).

Livelihood Options: Farming and Hunting

Farming was stated to be the primary occupation for 96%

of interviewed hunters (n = 87). Farming was also listed as

the highest source of income by nearly 45% of interviewed

villagers (n = 84) followed by non-timber forest product

collection (31%) and hunting (24%). 58% of respondents

indicated trade, 27% listed subsistence use and 14% listed

human-wildlife conflict as the main reason for hunting

(n = 84) (Chi Square = 6, df = 2, P = 0.049). Crossbows

(70%) and jawtraps (12%) were the two most commonly

used weapons for hunting in the area (N = 118). 75% of

the respondents stated that the wildlife they traded was

destined for Putao, with the remaining destined for China

through Khaunglanphu, to Rattan collecting camps and

nearby villages. Table 2 provides information on the price

of traded wildlife parts obtained from group interviews at

the village level. Some of the most commercially valuable

species include bear (gall bladder), musk deer (musk),

pangolin (scales), takin, serow and sambar deer (antlers).

Data were based on 164 interviews with hunters during

village group discussions (n = 13 villages). The top five

species most valued in trade (bear, pangolin, sambar, musk

deer and otter) coincided with the top five species listed as

preferred targets by hunters during household interview

surveys reinforcing the importance of trade as the primary

driver for hunting. While bear, pangolin and sambar were

0

0.5

1

1.5

2

2.5

3

3.5

4

0 0.2 0.4 0.6 0.8 1 1.2 1.4

Cap

ture

Rat

eper

100t

rap

nig

hts

(C

amer

a tr

ap c

aptu

res)

EncounterRateper 100sq.m (Track and sign counts)

Fig. 2 Comparison of trends in relative abundance of species as

determined by track and sign (encounter rate per 100 sq.m) and

camera traps (capture rate per 100 trap nights) for 12 species

Environmental Management

123

obtained in actual hunts, musk deer and otter were absent

from all hunting records as confirmed by hunters.

Patterns of Wild Meat, Fish and Livestock

Consumption

Interview surveys at the household level were used to

determine patterns of food consumption among villagers

specifically sources of protein such as fish, wild meat and

livestock meat. Fishing was a continuous activity that vil-

lagers engaged in throughout the year while hunting

occurred primarily during the dry season months

(November–February/March). Consumption data for fish,

wild meat and livestock were classified into four categories

based on the number of families consuming (in kg/month),

0 (no consumption), low (\5 kg), medium (5–8 kg) and

high ([8 kg) levels of fish, wild meat and livestock (Fig. 4

and Tables 3, 4).

(i) Throughout the year, relatively fewer families con-

sumed wild meat compared to livestock and fish. A

significantly higher number of families reported ‘zero’

consumption of wild meat compared to fish (n = 9)

and livestock (n = 2) (Chi square = 12.41, df = 2;

P = 0.002).

(ii) Most families consumed low levels of all three

sources of food (\5 kg per month). Patterns of fish

and wild meat consumption in terms of the number of

families consuming low, medium and high levels of

food (as defined above) remained the same but more

families reported consuming low levels of livestock

meat during the hunting season and fewer families

reported medium to high levels of consumption. As

can be expected, a few more families consume low

levels of wild meat during the hunting season.

(iii) Average amount of wild meat consumed per month

is not significantly higher during the hunting season

compared to the planting season (paired t-test,

t = 1.967, P = 0.05). However, average amounts

of fish and livestock consumed are significantly

higher during the planting season (paired t-test,

t = -7.113 (fish) and t = -23.018 (wild meat),

P \ 0.0001 for both).

(iv) Throughout the year, the average amount of fish

consumed per month was higher than livestock or wild

meat (Friedman test, Q = 243.365, P \ 0.0001).

(v) Average cost of livestock meat (kg/month) consumed

was found to be significantly more expensive than the

average cost of fish and wild meat with fish being

slightly more expensive than wild meat during both

seasons (Friedman Test Q = 138.721, P \ 0.0001).

(vi) Wealth status as measured by the possession of assets

such as durable goods (e.g. sewing machines),

livestock, transportation, agricultural implements,

and hunting/fishing equipment was most strongly

associated with livestock consumption compared to

fish and wildmeat during both seasons (Table 5). As

expected, livestock ownership was correlated most

strongly with livestock consumption during both

seasons. Ownership of hunting and fishing equipment

was more strongly correlated with fish consumption

compared to wildmeat consumption in both seasons.

0

0.1

0.2

0.3

0.4

0.5

0.6Preferred

Hunted

Rela

�ve

Freq

uenc

y

Fig. 3 Comparison of relative frequencies of species listed as top priority by hunters (n = 74) and those actually obtained during hunts

(n = .86). Preferred: relative frequency of species listed as top priority by hunters; Hunted: relative frequency of species obtained during hunts

Environmental Management

123

Discussion

Declining Species

Myanmar has been historically renowned as much for its

plentiful game as for the vast expanses of its dense jungles.

Large-scale depletion of wildlife occurred during and after

WW II and resulted in the rapid decimation of species such

as rhino both inside and outside protected areas (Milton

and Estes 1963). Although long-term monitoring data on

vulnerable species and offtake levels are necessary to

confirm species declines, evidence from this study supports

the likely decline of commercially valuable species such as

tiger, otter and musk deer. Similar declining trends for

particularly large mammals are reported from the Western

Indian Himalayas (Kaul and others 2004), in the Namdapha

National Park which adjoins the park across the Indian

border (Datta and others 2008) as well as the Hkakaborazi

National Park in North Myanmar (Rao and others 2005).

Decline in otter abundance has been documented by Than

Zaw and others (2008) and tiger populations appear to have

been drastically reduced across the country (Lynam and

Table 2 Data on wildlife parts traded and the average cost per unit of each part as determined through interviews of groups of hunters across 12

villages

Species Part traded Unit* USD Average SD n

Bear sp. Canines pair 0.38–1.54 0.87 0.40 8

Fat bottle 0.92–3.08 1.78 0.73 10

Meat viss 0.62–0.77 0.73 0.07 8

Gall bladder tical 11.54–53.85 27.12 13.75 20

Legs pair 7.69 6.15 3.08 9

Clouded leopard Skeleton, skin set 15.38 15.38 0.00 2

Hoolock gibbon Head set 0.77 0.00 0.00 1

Macaque sp. Gall bladder one 0.38–0.58 0.45 0.11 3

Legs pair 0.92–1.15 1.04 0.16 2

Meat viss 0.62–0.77 0.73 0.08 4

Skin sheet 0.38–1.54 0.96 0.82 2

Muntjac sp. Meat viss 0.62–0.77 0.68 0.12 7

Skin sheet 0.38–1.92 0.88 0.44 11

Musk deer Musk tical 61.54–115.38 71.15 20.78 5

Otter sp. Skin sheet 38.46 0.00 0.00 1

Pangolin Scales viss 34.62–53.85 42.31 7.69 5

Red goral Antlers pair 11.54 0.00 0.00 1

Gall bladder one 0.77–1.15 0.96 0.27 2

Meat viss 3.85 0.00 0.00 1

Skin sheet 0.77 0.00 0.00 1

Sambar Antlers pair 30.77 0.00 0.00 1

Heart one 0.77 0.00 0.00 1

Meat viss 0.62–0.77 0.69 0.11 2

Serow Antlers set 7.69–23.08 13.08 8.57 6

Gall bladder one 3.85 0.00 0.00 1

Heart one 0.58–0.77 0.67 0.14 2

Meat viss 0.31–0.77 0.73 0.27 6

Skin sheet 0.77 0.00 0.00 1

Small cat sp. Skin sheet 0.77 0.00 0.00 1

Takin Antlers pair 7.69 7.69 0.00 2

Dried meat viss 3.85 0.00 0.00 1

Tortoise sp. Whole individual 0.77–3.85 2.33 1.22 8

Wild pig Canines set 0.38–3.85 1.43 1.15 7

Meat viss 0.62–0.77 0.72 0.07 11

Yellow-throated marten Skin sheet 0.77 0.00 0.00 1

* 1 viss = 1.63 kg and 1 USD = 1300 Kyats

Environmental Management

123

others 2006). Intense hunting pressure on commercially

valued species such as bear, otter, musk deer, sambar and

pangolin can be expected to decimate their populations

unless significant measures are taken to stop illegal

poaching within HWS. Over time, it is expected that

smaller, more productive species should replace larger

bodied species of low productivity in hunter offtake

(Jerozolimski and Peres 2003). With trade as the primary

driver of hunting, species of high commercial value and

high vulnerability (low productivity) can be expected to be

depleted first with hunter offtake being increasingly dom-

inated by less commercially valuable (and hence less pre-

ferred) species.

Given that wildlife declines can happen very rapidly,

rigorous biological monitoring of target species will be

critical to detect declining populations (Albrechtsen and

others 2005). Accurate estimation of species’ population

trends for monitoring purposes is particularly difficult at

sites where they are hunted and where animals are either

extremely shy and/or occur at low densities (Nichols and

Karanth 2002). Validating estimates obtained from camera

trapping with independent estimates derived from line-

transects is a potentially useful approach to monitoring,

especially given limitations of access to sites and generally

low detection probability of wildlife in Myanmar’s forested

protected areas (Jennelle and others 2002; O’Brien and

others 2003).

Hunting and Livelihoods

The predominant livelihood activity, farming, was also

listed as being a higher source of income than hunting.

However, the small sample size warrants caution in inter-

preting these results. Villagers, aware that hunting was an

illegal activity may not have accurately represented the

facts. Furthermore, the study did not obtain data on agri-

cultural income. The importance of hunting as a source of

income for those listing farming as the primary source of

income although undetermined is clearly important to

examine. Given the non-separability of consumption and

production decisions in rural settings, it is particularly

important to determine the magnitude of livelihood

dependence on the harvest of wildmeat (Barrett and Arcese

1998). We need to examine factors influencing the decision

to hunt by villagers and understand the allocation of effort

between agriculture, non-timber forest collection and

hunting (Damania and others 2003). Participation in

hunting may be influenced by measures of relative wealth,

including ownership of livestock and farming land, means

of generating cash income and access to alternative sources

of meat (Loibooki and others 2002; Mcelwee 2008).

Overall, there is limited understanding of the microeco-

nomic logic that gives rise to livelihood heterogeneity

among forest peoples but researchers recognize that forest

people, even within the same community, vary markedly in

the way in which they incorporate forest resources into

their livelihoods (Agrawal and Gibson 1999; Barrett and

others 2001a, b). It is likely that a few of the poorest

families with no access to agricultural income could

engage in hunting to fulfill both subsistence and livelihood

0

50

100

150

200

250

0 0-3 3--5 >5 0 0-3 3--5 >5

Nu

mb

er o

f fa

mili

es

Planting Season Hunting Season

FishLivestock

Wildmeat

Fig. 4 A comparison of the consumption patterns (kg/month/house-

hold) of fish, livestock and wild meat in the planting and hunting

seasons (n = 242 households)

Table 3 Consumption patterns of wild meat, fish and livestock (kg/

household/month) for 242 households in 13 villages during the

planting and hunting seasons

Variable Minimum Maximum Mean SD

Planting Season

Wild meat 0.000 25.000 2.142 4.305

Fish 0.000 23.830 2.467 3.268

Livestock 0.000 11.170 1.464 1.616

Hunting Season

Wild meat 0.000 25.830 2.008 3.612

Fish 0.000 25.000 2.822 3.752

Livestock 0.000 9.750 2.154 1.809

Friedman’s test: P \ 0.0001

Table 4 Statistical tests to examine differences in the consumption

of fish, livestock and wild meat (kg/month per household) between

the planting and the hunting seasons

Consumption in the planting

vs. hunting seasons

Chi square df P

Fish 0.7 2 0.873

Livestock 16.99 2 0.00071

Wild meat 2.69 2 0.44

Environmental Management

123

needs especially if crops fail or other unforeseen circum-

stances preclude alternative forms of subsistence and/or

livelihood. Furthermore, data on actual offtake by hunters

will help confirm suspected declines in species abundance.

Is Wild Meat a Critical Source of Food?

In the current study, wildmeat is likely to be an indirect

source of food for families in that the sale of meat probably

facilitates the purchase of food. Although a few more

families reported consuming wild meat during the hunting

season, it was not significantly more than those reporting

consumption during the planting season. Wild meat

appears to be less important than fish even during the

hunting season although fish is slightly more expensive

than wild meat in both seasons. A large number of families

appear to consume no wild meat throughout the year. The

availability of fish is perhaps closely linked to hunting

pressure as shown by Brashares and others (2004) in West

Africa where mammal declines (to wildlife trade) were

associated with spatial and temporal changes in fish sup-

plies. Higher costs associated with livestock consumption

make livestock a less preferred option compared to both

wild meat and fish. Also, livestock are reared for use during

emergencies and special occasions rather than for subsis-

tence (Bennett 2002). Increasing the productivity and

resilience of livestock could diminish the existing demand

for wild meat and, consequently, reduce the amount of

wildlife hunted for food. Fish is clearly an important source

of food and it is necessary to better understand the patterns

of dependence and impacts of extraction on fish popula-

tions. Overall, wild meat and other wildlife products could

play a larger role in household income than in direct

household consumption (De Merode and others 2004).

What Happens Once Wildlife Resources are Depleted?

Across Africa and Asia, although alternatives are available,

many people eat wild meat for cultural or taste reasons

(Bennett and others 2000; Lee 2000). Hence, if that wild

meat is no longer available, communities switch to alter-

native sources such as fish or livestock meat (Bennett

2002). In the Udzungwa mountains, Tanzania, depletion of

relevant wildlife populations has resulted in the relatively

low importance of bushmeat in terms of use and contri-

bution to fulfilling protein requirements. Neilsen (2006)

suggests that conservation efforts should focus more on

protecting wildlife species against exploitation and

encouraging a complete shift to domestic sources of meat,

by increasing the number of domestic animals in the

poorest parts of the population. However, increases in

livestock stocking rates are likely to have negative impli-

cations for habitat integrity and is less preferable to

increasing productivity and disease resistance of existing

livestock. Depletion of commercially valuable species

should have implications for those families dependent on

wildlife trade as a source of income. When the need for

income is immediate, it is unlikely that poor people will

refrain from exploiting decimated wildlife populations

(Kuhl and others 2009). However, a key consideration is

whether families continue to invest efforts in harvesting

and trading in less valuable species or if they switch to

alternative sources of livelihood. Understanding the factors

that drive the switch to alternative sources of income will

help design effective strategies to protect vulnerable spe-

cies of high commercial value while safe-guarding liveli-

hoods for the most severely affected families.

The following are some recommendations for conser-

vation action to help address the issue of declining wildlife

populations within HWS.

Park Staffing. Staffing the park adequately is critical for

effective conservation planning and implementation (Aung

2007). At the time of the study, there were no NWCD staff

stationed at the Park and the situation has not changed

since that time (Than Myint, personal communication,

2009)

Opportunity costs of poaching. As long as the expected

economic benefits from the sale of wildlife are greater than

Table 5 Relationships between wealth status (measured in terms of…) and the consumption of livestock, fish and wildmeat during the hunting

and planting seasons

ConSumption: hunting season Consumption: planting season

Livestock Fish Wildmeat Livestock Fish Wildmeat

Livestock 0.46 (\0.0001) 0.24 (0.000) 0.15 (0.016) 0.43 (\0.0001) 0.171 (0.007) 0.166 (0.009)

Hunting/fishing equipment 0.174 (0.006) 0.449 (\0.0001) 0.297 (\0.0001) 0.327 (\0.0001) 0.445 (\0.0001) 0.380 (\0.0001)

Total assets 0.424 (\0.0001) 0.390 (\0.0001) 0.241 (0.000) 0.528 (\0.0001) 0.351 (0.0001) 0.361 (0.0001)

Total assets (not incl.hunting/

fishing equipment)

0.494 (\0.0001) 0.336 (0.0001) 0.220 (0.000) 0.540 (0.0001) 0.283 (0.0001) 0.267 (0.0001)

Values in brackets are P-values

Environmental Management

123

the costs associated with a low probability of arrest, illegal

hunting for trade will continue (Hofer and others 2000).

Increasing the opportunity costs of poaching either through

enforcement or by increasing the returns to time spent in

alternative activities such as domestic livestock rearing

(see above) could help reduce the occurrence of poaching.

In practical terms, this might involve higher fines, or more

frequent and intensive patrols and other related policies

that require greater avoidance by poachers, thus rendering

poaching more difficult. Myanmar has committed to safe-

guard several species and representative ecosystems

through national legislation (Protection of Wildlife and

Protected Areas Law; State Law and Order and Restoration

Council 1994) and international conventions such as CBD

and CITES. Hence the provision of alternative forms of

livelihood (with high returns to time spent and minimum

impacts on habitats and species) to the poorest people who

depend on hunting and the cessation of illegal trade in

protected species should be a conservation priority for the

NWCD.

Extension services must encourage adoption of new

technologies that enable livestock to fulfill multiple roles

(for example, savings, insurance, food and income). These

new technologies must increase livestock production and

lower prices of domestic meat, while limiting the habitat

loss and degradation that are often associated with

increased livestock production.

Conservation Research

A clearer understanding of the economic factors that drive

hunting is useful to develop appropriate conservation and

development strategies. In general, predictors of use and

reliance can vary significantly across resource extraction

activities (Coomes and others 2004). In this case, given the

low population density, it should be straightforward to

identify those families that depend on wildlife for subsis-

tence and income. To effectively address the problem of

wildlife hunting, obtaining data on distribution, density and

rates of change in populations of hunted species is a clear

priority (Milner-Gulland and others 2002). We need to

determine if hunting is linked to poverty and low protein

intake by making comparisons between hunters and non-

hunters in terms of wealth measures.

No Take Areas

With appropriate incentives and controls in place, no-take

areas when appropriately designed (large enough) are a

promising way to achieve landscape-wide sustainable

hunting that can be enforced more easily than can restric-

tions on offtake. Studies have shown that game populations

can be sustained, despite high hunting pressure by

immigration from unhunted refugia via source-sink

dynamics (Ohl-Schacherer and others 2007; Naranjo and

Bodmer 2007). In particular, it is important to ensure that

the establishment of a no-take area does not exacerbate

biodiversity loss in surrounding areas. Local people are

more likely to support full protection of source areas if they

perceive direct benefits from them.

Conclusions

Commercially valuable species appear to be in decline or

locally extinct probably due to overexploitation for trade

which is the primary driver of hunting in HWS. Monitoring

vulnerable species and effective law enforcement will be

necessary to prevent further impoverishment. Fish is more

frequently consumed than wild meat or livestock

throughout the year. Wild meat plays a small role in

household consumption but a potentially important role in

household income. More research to determine social and

economic factors influencing livelihood choice will help

design effective approaches to conserving wildlife. Trends

and findings in this study are relevant both regionally and

globally where prevalent problems of declining wildlife

populations and dependent human communities require

carefully crafted solutions.

Acknowledgments We are grateful for the financial assistance

provided by the Foreign and Commonwealth Office, British Embassy,

Yangon and for the permissions granted and support provided by the

Nature and Wildlife Conservation Division, Ministry of Forestry,

Government of Myanmar. We are indebted to the local people who

shared their knowledge with us and local field assistants who spent

many long and arduous days in challenging field conditions. We

would like to especially thank U Kyaw Thinn Latt for his assistance in

creating Fig. 1 and to the administrative staff of the WCS Myanmar

Program for logistical assistance. The Wildlife Conservation Society,

New York provided additional financial assistance for this project.

We would like to thank three anonymous reviewers for their sug-

gestions to improve the manuscript.

References

Agrawal A, Gibson CC (1999) Enchantment and disenchantment: the

role of community in natural resource conservation. World

Development 27:629–649

Albrechtsen L, Macdonald DW, Johnson PJ, Castelo R, Fa JE (2005)

Faunal loss from bushmeat hunting: empirical evidence and

policy implications in Bioko Island. Environmental Science and

Policy 10:654–667

Amato G, Egan MG, Rabinowitz A (1999) A new species of muntjac,

Muntiacus putaoensis (Artiodactyla: Cervidae) from North

Myanmar. Animal Conservation 2:1–8

Anon (2005) Going, going, gone. The illegal trade in wildlife in East

and Southeast Asia. The World Bank, Washington, DC 23 pp

Aung M (2007) Policy and practice in Myanmar’s protected area

system. Journal of Environmental Management 84:188–203

Environmental Management

123

Barrett CB, Arcese P (1998) Wildlife harvest in integrated conser-

vation and development projects: linking harvest to household

demand, agricultural production, and environmental shocks in

the Serengeti. Land Economics 74:449–465

Barrett CB, Brandon K, Gibson C, Gjertsen H (2001a) Conserving

tropical biodiversity amid weak institutions. Bio-Science

51:497–502

Barrett CB, Reardon T, Webb P (2001b) Non-farm income diversi-

fication and household livelihood strategies in rural Africa:

concepts, dynamics, and policy implications. Food Policy

26:315–331

Bennett EL, Nyaoi AJ, Sompud J (2000) Saving Borneo’s bacon: the

sustainability of hunting in Sarawak and Sabah. In: Robinson JG,

Bennett EL (eds) Sustainability of hunting in tropical forests.

Columbia University Press, New York, pp 305–324

Bennett EL (2002) Is there a link between wild meat and food

security? Conservation Biology 16:590–592

Bennett EL, Milner-Gulland EJ, Bakarr M, Eves HE, Robinson JG,

Wilkie D (2002) Hunting the world’s wildlife to extinction. Oryx

36:328–329

Brashares JS, Arcese P, Sam MK, Coppolillo PB, Sinclair ARE,

Balmford A (2004) Bushmeat hunting, wildlife declines, and fish

supply in west Africa. Science 306:1180–1183

Coomes OT, Barhamb BL, Takasakic Y (2004) Targeting conserva-

tion–development initiatives in tropical forests: insights from

analyses of rain forest use and economic reliance among

Amazonian peasants. Ecological Economics 51:47–64

Corlett RT (2007) The impact of hunting on the mammalian fauna of

tropical Asian forests. Biotropica 39:292–303

Damania R, Stringer R, Karanth U, Stith B (2003) The economics of

protecting tiger populations: linking household behaviour to

poaching and prey depletion. Land Economics 79:198–216

Datta A, Anand MO, Naniwadekar R (2008) Empty forests: Large

carnivore and prey abundance in Namdapha National Park,

north-east India. Biological Conservation 141:1429–1435

De Merode E, Homewood K, Cowlishaw G (2004) The value of

bushmeat and other wild foods to rural households living in

extreme poverty in Democratic Republic of Congo. Biological

Conservation 118:573–581

Hofer H, Campbell KL, East ML, Huish SA (2000) Modeling the

spatial distribution of the economic costs and benefits of illegal

game meat hunting in the Serengeti. Natural Resource Modeling

13:151–177

Jennelle CS, Runge MC, Mackenzie DI (2002) The use of

photographic rates to estimate densities of tigers and other

cryptic mammals: a comment on misleading conclusions.

Animal Conservation 5:119–120

Jerozolimski A, Peres CA (2003) Bringing home the biggest bacon: a

cross-site analysis of the structure of hunter-kill profiles in

Neotropical forests. Biological Conservation 111:415–425

Kaul R, Hilaludin Jandrotia JS, McGowan PJK (2004) Hunting of

large mammals and pheasants in the Western Indian Himalaya.

Oryx 38:1–6

Kuhl A, Balinova N, Bykova E, Arylov YN, Esipov A, Lushchekina AA,

Milner-Gulland EJ (2009) The role of saiga poaching in rural

communities: Linkages between attitudes, socio-economic circum-

stances and behaviour. Biological Conservation 142:1442–1449

Lee RJ (2000) Impact of subsistence hunting in North Sulawesi,

Indonesia, and conservation options. In: Robinson JG, Bennett

EL (eds) Sustainability of hunting in tropical forests. Columbia

University Press, New York, pp 455–472

Leimgruber P, Kelly DS, Steininger MK, Brunner J, Muller T, Songer

M (2005) Forest cover change patterns in Myanmar (Burma)

1990–2000. Environmental Conservation 32:356–364

Loibooki M, Hofer H, Campbell KLI, East ML (2002) Bushmeat

hunting by communities adjacent to the Serengeti National Park,

Tanzania: the importance of livestock ownership and alternative

sources of protein and income. Environmental Conservation

29:391–398

Lynam AJ, Khaing ST, Zaw KM (2006) Developing a National Tiger

Action Plan for the Union of Myanmar. Environmental Man-

agement 37:30–39

Mcelwee PD (2008) Forest Environmental income in Vietnam:

household socioeconomic factors influencing forest use. Envi-

ronmental Conservation 35:147–159

Milner-Gulland EJ, Bennett EL, the SCB 2002 Annual Meeting Wild

Meat Group (2003) Wild meat: the bigger picture. Trends in

Ecology and Evolution 18:351–357

Milton O, Estes RD (1963) Myanmar Wildlife Survey: 1959–1960.

Special publication No. 15. A report prepared for the Interna-

tional Union for Conservation of Nature and Natural Resources

and the American Committee for International Wildlife Protec-

tion, New York

Naranjo EJ, Bodmer RE (2007) Source–sink systems and conserva-

tion of hunted ungulates in the Lacandon Forest, Mexico.

Biological Conservation 138:412–420

Neilsen MR (2006) Importance, cause and effect of bushmeat hunting

in the Udzungwa Mountains, Tanzania: Implications for com-

munity based wildlife management. Biological Conservation

28:509–516

Nichols JD, Karanth U (2002) Statistical concepts: indices of relative

abundance. In: Karanth KU, Nichols JD (eds) Monitoring tigers

and their prey. A Manual for researchers, managers and

conservationists in Tropical Asia. Centre for Wildlife Studies,

Bangalore, India, pp 61–69

Nooren H, Claridge G (2001) Wildlife trade in Laos: the end of the

game. Netherlands Committee for IUCN, Amsterdam, The

Netherlands, 304 pp

O’Brien T, Kinnaird M, Wibisono H (2003) Crouching tigers, hidden

prey: Sumatran tiger and prey populations in a tropical forest

landscape. Animal Conservation 6:131–139

Ohl-Schacherer J, Shepard JG Jr, Kaplan H, Peres CA, Levi T, Yu

DW (2007) The sustainability of subsistence hunting by

Matsigenka Native Communities in Manu National Park, Peru.

Conservation Biology 21:1174–1185

Rabinowitz A, Amato G, Khaing ST (1998) The discovery of the

black muntjac, Muntiacus crinifrons, in northern Myanmar.

Mammalia 62:105–108

Rao M, Rabinowitz A, Khaing ST (2002) Status review of the

protected area system in Myanmar, with recommendations for

conservation planning. Conservation Biology 16:360–368

Rao M, Myint T, Zaw T, Htun S (2005) Hunting patterns in tropical

forests adjoining the Hkakaborazi National Park, North Myan-

mar. Oryx 39:292–300

Renner SC, Rappole JH, Leimgruber P, Kelly DS, Shwe NM, Aung T,

Aung M (2007) Land cover in the Northern Forest Complex of

Myanmar: new insights for conservation. Oryx 41:27–37

Shepherd CR (2001) Observations on Wildlife Trade at Golden Rock,

Myanmar. TRAFFIC Bulletin 19:7–10

Shepherd CR, Nijman V (2007) An assessment of wildlife trade at

Mong La market on the Myanmar–China border. TRAFFIC Bull

21:85–88

State Law and Order Restoration Council (SLORC) (1994) The

protection of wildlife and protected areas law (Law No. 6/94).

8th June 1994. Yangon, Myanmar

Zaw T, Htun S, Po SHT, Maung M, Lynam AJ, Latt KT, Duckworth

JW (2008) Status and distribution of small carnivores in

Myanmar. Small Carnivore Conservation 38:2–28

Tordoff AW, Eames JC, Eberhardt K, Baltzer MC, Davidson P,

Leimgruber P, Uga U, Than A (2005) Myanmar: investment

opportunities in biodiversity conservation. Birdlife International,

Yangon, Myanmar

Environmental Management

123