Population Evaluation of Asian Elephant (Elephas maximus) in Commercially Exploited Habitats of Myanmar

2008·0.71 MB·English

by  Varma, Surendra

NAT. HIST. BULL. SIAM Soc. 56(1): 55--68，2 008
POPULATION EVALUATION OFA SIAN ELEPHANT
(ELEPHAS MAXIMUS) IN COl\品1ERCIALLYEXPLOITED
HABITATS OFM YANMAR.
Surendra 拘rma1，UY e Htut 2，3a nd U Ugd'
ABSTRACT
As tudy on the status of the Asian elephant (Elephas maximus) was initiated in Myanmar
with the objective of preparing ac onservation plan for elephants in selected locations of the
country. Five forest reserves of the Bago Yoma (central Myanmar)，s even forest reserves in
Rakhine Yoma (westem Myanmar) and Alaungdaw Kathapa National Park (AKNP) of northem
Myanm訂 we問 thesurvey sites. The indirect method of estimating elephant density合omthe
elephant d巴fecationrate/day，d ung decay rate/day，an d dung density巴stimateswas used. The
estimated mean daily defecation rate/day was 20.0 for Bago Yoma，1 9.0 for Rakhine Yoma
and 23.0 for AKNP. The mean daily dung decay rate/day was 0.072，0. 005 and 0.009 for Bago
Yoma，R akhine Yoma and AKNP，re spectively. The mean elephant density for Bago was 1.62
= =
(95% CI 1.49-1.75) and for Rakhine it was 0.05 (95% CI 0.04-0.005) and for AKNPit was
0.6 (95% CI =0 .528-D.74). Elephant dung density and elephant density varied substantially for
Bago Yoma and Rakhine Yoma Reserves. Thee stImated elephant density for both Bago Yoma
and AKNPa ppear to be very high comp紅'edto the earlier surveys. For Bago出eresult could be
due to av ery high dung decay rate that may not be representative of the entire wet season or for
regions outside the observation紅ea.Sample size for some r，邸erveswas very low，he nce，la ying
more tr釦 sectsis recommended for these reserves. Additional dung decay rate experiments，
particularly during the wet seasons，a nd similar systematic surveys covering other regions of
出ecountry，ar e desirable
Key Words: Habitat and population surveys，in direct census method，fo rest r凶erves
INTRODUCTION
The Asian elephant (Elephas maximus) is ag lobally threatened species and its survival
depends onm aintaining viable habitats and understanding the population status oft he species
(SUKUMAR，1 989; SANTIAPILLAI & JACKSON，19 90). The cu汀'entdistribution of the species
covers only as mall portion of its earlier extensive range (SUKUMAR & SANTIAPILLAI，19 96).
I Asian Elephant Research and Conservation Centre (A division of Asian Nature Conservation Foundation)，C/ o
Centre for Ecological Sciences，I ndian Institute of Science，B angalore -560 012，I ndia. Email:
[email protected]
2 Wildlife and Nature Conservation Division，Mi nistry of Forestry，Ya ngon，My anmar.
3 Current address: Park warden，A laungdaw Kathapa National Park Office，Y imabin Township，M agway
Division，M y釦m紅.
4 Chairman，Bi odiversity and Nature Conservation (BANCA)，1 4/A Bawga street，引mileMayangon Township，
Yangon，My mar.
組
Received 23 June 2007; accepted 5J anuary 2008.
55
56 SURENDRA VARMA，U YE HTUT AND U UGA
ALAl]SGD.W.K ATHArA ~' ..
、l!''''''
N
BAしり
1・NE. !\orth Ea.s1
2-SW 。SoUlh\ \出1
)-Mindo
φK.咽由剖
3・Ktuue
、
民AIH1M.:
~ IOU訓
2. S(lU出Za1nan
3.1'.叫由laman
" Yt:I同C
5I dυkon
守マ句
t.I>OSGA仇
2υ1)~SM-m I
J.l}D:o-I‘RAK : 2
4トOON\ι、、t'l
'.0旬、旬、\'I.l.l
か00河川、司P
7.()O~~1、T
Figure 1. Locationo f thes urvey sitesi n Myanma .rThe1 ocationo f thes urvey sites arem arkecl 31llongt he創easof
th巴e1ephantclistributioni n thec ouI1try
POPULATION EVALUTION OF ASIAN ELEPHANT (ELEPHAS MAXIMUS) 57
In白iscontext，af ter India，My anmar has the largest remaining habitat and population of both
wild and captive elephants. The captive elephant has always been considered the backbone
of the country's economy as nearly 50 percent of all timber in Myanmar is still extracted by
these elephants and about 3000 working elephants eneeded for the timber industry (ZAW，
紅
1998). To replenish captive stocks for timber extraction，el ephants have been captured from
the wild (SALTER，1 983; HTUT，19 93). Timber extraction has two negative effects on the
status of wild elephants: viable habitats are disturbed through extensive logging operations，
and there is as ubstantial decline of the wild elephant population. In 1972 the estimated
elephant number for Myanmar was 6000，w hich was reduced to 3000 by late 1970s and it
was predicted that the wild elephant population in Myanmar would decline by about 5%p er
year(CAUG且 EY，1980).
Estimating elephant numbers and mapping their distribution thus becomes imperative，
however，n o systematic surveys or studies on wild elephants have been carried out in the
regions of logging and elephant capture. There have been several attempts to estimate
elephant numbers，bu t these have no scientific basis (HTIπ1993; SUKUMAR & SANTIAPILLAI
1996). Estimating elephant numbers is difficult as visibility within the forests is very poor
and many of the forests are inaccessible. Most of the regions in this coun位y訂'every remote，
with rugged terrain，in fested with malaria，a nd have few and very poor logistic facilities.
However，th ese regions are very important due to the presence of globally threatened species
(SALTER，1 983; IUCN，1 989; HTUT，1 993). Although Myanmar provides crucial habitat
for Asian elephants，n o scientific study or survey on the species has ever been undertaken.
These aspects motivated us to conduct as tatus survey，po pulation evaluation and prepation
紅
of ac onservation action plan for the Asian elephant in commercially exploited habitats of
Myanmar. This paper reflects the first ever scientific approach to study the elephants in this
coun町 andthe findings should give scope for critical review of methods used and the results
obtained.
MATERIALAND METHODS
Survey Sites
The survey sites (Fig. 1) were Bago Yoma (170-200 N，9 60_970 E)，R akhine Yoma
(170-210 N，93 0-950 E) and Alaungdaw Kathapa National Park -AKNP( 220-230 N，9 40-
950 E).
The Bago，Ra khine and AKNPr egions have very extensive tracts of hills situated in the
central，w estem and northem regions of Myanmar，re spectively. The hill ranges of Rakhine
Yoma are as outhward extension of the Himalayas. AKNPi s in aw ell-forested mountainous
region situated west of the lower Chindwin River and Myittha Valley. The average elevation
of the Bago Yoma is about 700 m; the highest point is 900 m above sea level (asl). Rakhine
Yoma，wh ich runs for nearly 600 km，ra nges from 100 to 1400 m asl and average elevation
in AKNPi s about 1000 m; steep slopes and narrow ridges characterize all regions. Allh ave
good drainage systems: the Pegu and tributaries of Yenwe Chaung，a nd the Kun Chaung
are the major river sources in Bago Yoma. The Sandoway River (Sandoway Chaung) is血e
major river system in Rakhine. AKNPi s drained by an umber of tributaries of the Patolon
River，P etpa Chaung and Taungdwin Chaung being perennial among them. In all these
58 SURENDRA VARMA，U YE HTUT AND U UOA
regions，th e wet season lasts from May to October and rainfall is heaviest in August and
September. The annual rainfall for Bago averages 1700 mm，fo r Rakhine it is 1800 mma nd
for AKNP，1 500 mm. In all these regions，th e vegetation is largely mixed deciduous forest，
with semi-evergreen forests occurring in areas of high precipitation. Patches of evergreen
trees consisting mostly of secondary growth occur in af ew places.
Notable among the mammal species seen here are rhesus macaque (Macaca mulatta)，
hoolock gibbon (Hylobates hoolock)，P hayre's langur (Semnopithecus phayrei)，s nbar
釘
(Cervus unicolor)，ba rking deer (Muntiacus muntjak)，h og deer (Axis porcinus)，ga ur (Bos
gaurus)，T saine (saing) or banteng (Bos javanicus) serow (Naemorhedus sumatraensis)，
elephant (Elephas maximus)，S umatran rhino (Dicerorhinus sumatrensis)，As iatic black bear
(Ursus thibetanus)，Ma layan sun bear (Ursus malayanus)，le opard (Panthera pardus)，ti ger
(Panthera tigris) and Asian wild dog (Cuon alpinus) (common and scientific nam based
回
on CORBET & HILL (1992)，YI N (1993)，an d MENON (2003).
Current Habitat and Legal Status
All these regions have been su吋ectto intensive m agementfor logging during the
加
past 130 ye紅 s(SALTER，1 983; MY町T，1994; TUN，1 997). Under sustainable management
of forests since 1856，ti mber has been extracted and the system prescribed af elling cycle of
30 years in af elling series. Exc巴ptin av ery small part of as teep slope of Bago Yoma (UGA
1995)，ti mber extraction has been carried out in all the regions. The 130 ye sof logging has
紅
had negative effects and in many places this disturbance，al ong with other human activities，
has resulted in large areas being invaded by bamboo species. (SALTER，1 983; UGA，19 95).
Most of the regions have yet to evolve effective wildlife conservation and management
progr創ns.Out of白e16，000 km2a rea of Rakhine Yoma，on ly 1775 km2 has been gazetted
as Yakhine Yoma Wildlife Sanctuary，in 1997 (RAO ETA L，2 002; UGA，1 995). To preserve
the pristine nature of the teak and other forests，a s urvey for declaring 1500 km2B ago Yoma
Teak Nature Reserve was criedout in 1983，bu t the area has yet to be brought under legal
訂
protection or managemen .tAKNPw as legally gazetted as aw ildlife sanctuary in 1984.
Density Estimation
The indirect method of estimating elephant density (BARNES & JENSEN，19 87; DAWSON
& DEKKER，1 992; SANTOSH & SUKUMAR，1 993) was followed for the survey. Based on
village surveys and discussions with forest and Myanmar Timber Enterprise staff，t he
survey areas were divided into high，mo derate and low elephant use eas.Line transects of
訂
indefinite width (B URNHAM ETA L.， 1980) were used for estimating dung density in 5f orest
reserves of the Bago Yoma，7 f orest reserves of the Rakhine Yoma and 5r egions of AKNP.
Line transects of indefinite width are used when animal densities are low，an d the numbers
recorded from fixed-width transects are assumed to be too low for meaningful statistical
analyses (EBERHARDT，1 968)
Each tr加 sectwas cut afresh using af ield compass，a 卸値mnylon rope and knife. The
compass was used for fixing as traight line; the rope was to estimate the perpendicular distance
from the line to the dung piles，a nd the total distance covered. The rope was also useful in
assessing the topography and microhabitats at every 50minterval. The total number of
・
transects laid was 142 for Bago，1 48 for Rakhine，a nd 22 for AKNP. The totallength of
POPULATION EVALUTION OFA SIAN ELEPHANT (ELEPHAS MAXlMUS) 59
仕組sectsin ap articular reserve within ar egion was roughly proportional to the tota1 area of
the reserve. For exarnple Zarnari reserve in Bago constituted 30% of to凶 areaof the reserves
S創npledfor Bago and 25% of transects were laid in this reserve (Table 1). The sarne sarnple
design was followed for other regions surveyed and lines were well dis紅白uted，covering
di百erentregions of the reserves sarnpled. In ag iven site，no t more血an3g roups of workers
operated for cutting tr sectsand am inimum distance of 2k mw as maintained between the
佃
two groups.
For each佐佃sect，the date of sarnpling，its location，ve getation type，to pography，st art
and end times of survey and distance along the transect were recorded. Thet opography
was recorded as small hill with slope，un dulating terrain，ft at land，a nd other types. The
categories of vegetation types were: semi-evergreen forest，ev ergreen forest，mo ist upper
mixed deciduous forest，a nd bamboo. The percentage of transects laid in each category
of terrain and forest type roughly matched their overall percentages within ag iven region
(Table 1). On sighting ad ung pile during at ransect walk，th e perpendicular distance of the
pileto白etr加 sectline was measured，an d the condition of the dung pile w拙 notedusing血e
stages of decay prescribed by BAR阻 S&JENS町(1987).
For both Bago and R北hineYoma，a s tudy of elephant defecation w出 C紅riedout in
elephant carnps (observing 32 eleph佃 tsat Bago and 57 at R北hine)，as these elephants feed
only on natural vegetation仕omthe surrounding forest ar開 s.In total 1704 man-hours were
spent for the defecation study. Fresh dung piles observed during the defecation study were
marked for monitoring of decay.rate，an d at ota1 of 100 dung piles were monitored in each
region. For AKNP，da ta 企omavailable literature (MYINT，1 994) on these two pararneters
were used.
Data Analysis
Thed ensity of dung piles，da ily defecation rate and dung decay rate were estimated
using an updated version of GAJAHA so食ware(ARCHANA & SUKUMAR，2 006). The
sarne prograrn was also used to compute eleph tdensity. Elephant density w estimated
佃 槌
separately for each reserve，an d for AKNP，th e data of all regions were pooled，as the sarnple
size of the dung piles for each region was very low.
The density (E) in elephants/km2w 部 calculatedusing the formula:
E=Y 'r/D
= = =
where，Y density of dung piles/km2，r dung decay rate/day，an d D defecation rate/day.
RESULTS ANDD lSCUSSION
Table 1s ummarizes the information on白e釘'eaof the reserves sarnpled，nu mber and
percen旬geof transects，th eir topography and microhabitat，di stance covered，an d the sarnple
size of the dung piles for each reserve in Bago，R akhine and different regions in AKNP.
Ther esults of dung density mean elephant density and mean elephant number for different
regions surveyed 'esummarized in Table 2. Thed efecation rate for the wet season in Bago
紅
=
was 20.02/day (Stand紅dE立orSE 0.55) and for Rakhine in白e世yseason it was 19.04/day
=
(SE 0.55). Overall decay rate for 100 fresh dung piles in Bago during the wet season was
Table 1. Forest reserves sampled，a rea，n umber and percentage oft ransects surveyed，白eirtopography and microhabitats，d istance covered 8
and number dung piles recorded for Bago，Ra khine and AKNPr egions.
Distance Dung
Regions Nameof Area % Noof % Topography along血E % Forest types along % cover芭d piles
Reserves 仕m2) 位置nsects transects 世letr加 sects (km) recorded
Bago INSYOdookeourknktatwahnhe.n ZZ 創倒nEr釘ii 77854291801524 LLa'I1J弓内弓，9467lh ，LAZ0.ヲ自ツ内 L'z333弓内今今3auεJZOL今今 222184556...5..2464 UHFRSUilninmladddaltuug l 1w11e1aaias hnttt iwdihlili nn stgglh o avpnaedl l setye ep slope 2Jaa51uai4『 253069 MELMMSEMUUGUGDMM M +DDDS++ ESEGEG G “おロ32Z29后339 7774n 20265 ・‘・‘113aa勾nuL44AUwAリ、，nyhζJ‘，園内， ZC岡田印
ロ
同
Tota1 2952 142 283.5 658 〉
〈
〉
Rakhine DDDDDDDDDDDDDDNNNNNNNMMSSAMG町YAARAAAYNATYWP*UDKU**1 122 * ** * 2732216150665720720050...0000528 ，、う&今&今&'AE且F4t112042:・--tEtt'OJQJA且。。，守 'Ii'Ii勾，今，“，‘，‘，aununULnu4，、，今勾 m初日&丘&&8351418 GFHUUliennlanddlttuu 11w11eaiaa ntttsdhiil nnosggpl eoa pned steep slope 1(341 、.hnU『，，，J包574.31 OEMSE血GUGe M路 D 5314301...7.148 142223140602440 HHaUA，且今&守，司令守u616102四月 Pcdc.Z『〉「開国→同玄
ロ
Tota1 11踊;0.5 148 296 354 C
C
AKNP MNSooiurnt曲dho--wnwee sstt TaauaT品邑 211788...322 GUUennnddtuu111eaa ttsiilnnoggp ea nd s同 ps10pe A守今ゐ57U536 BMRFUA+MMBD A+MB AM '-sA守今且3A司守今M551 ロ86 %00 O〉
kKu組nz曲ea t aaTdaT 1188..22 RSitedegpe ss lwoipteh val1ey 40・h51 EMRFGU MD ・'』1jaEL-L・且298 88 00

Tota1 1606 22 42 36
*p紅tof Thandwe Reserved Forest (DDNSAND ，)lSabyin &M indon (DDNARAKAN2)，pa rt of Gwa Reserved Forest (DDNGWA)，n orth of May Yu Reserved For-
est (DDNMAYU1)， south of MayY u Reserved Forest (DDNMAYU2)，pa rt of Miva Pya (DDNMYAP)佃 dpart of Sin Tanung Reserved Forests (DDNS刑T).
BAM(Bamb∞)， MUMD+ B AM( Moist Upper Mixed Deciduous and Bamboo)，R F+ B AM( Riverian Forest and Bamboo)，R F( Riverian Fo問st)EG( Evergreen forest)
MUMD( Moist Upper Mixed Deciduous)，S EG (Semi Eve甲官enforest).
POPULATION EVALUTION OF ASIAN ELEPHANT (ELEPHAS MAXIMUS) 61
Table 2. Elephant density estimates for different reserves in Bago，Ra khine and AKNP.
Dung Mean elephant Mean number of
density density elephants
Region Reserve (SE) (95% CI) (959もCI)
Bago South Zamari 1 1380 (98.2) 4.95 (4.65-5.26) 542 (504-576)
South Zamari 2 121 (28.6) 0.43( 0.37--0.51) 70 (59-82)
South Zamari 3
North Zamari 151 (21.6) 0.54 (0.49--0.60) 388 (349-426)
Yenwe 300 (30.6) 1.07 (0.98-1.16) 852 (781-922)
Idokan 361 (42.2) 1.29 (1.18-1.42) 675 (613-737)
Okkan 392 (59.2) 1.41 (1.25-1.56) 56 (50-62)
Mean 1.62 (1.49-1.75) Tota12583
(2433-2805)
Rakhine DDNSAND 1 132 (22.4) 0.03 (0.03--0.04)
DDNARAK2 71 (8.9) 0.02 (0.01--0.02)
DDNGAW 165 (20.3) 0.04 (0.04--0.05
DDNMAYU 1 452 (66.0) 0.17 (0.12--0.14)
DDNMAYU2 225 (37.5) 0.06 (0.05--0.07)
DDNMYAP 27 (21.5) 0.01 (0.00--0.01)
DDNSINT 83 (18.6) 0.02 (0.02--0.03)
Mean 0.05 (0.04--00.5) Total722
(686.4ー758.7)
AKNP AKNP 1633.8 (375.8) 0.64 (0.52--0.74)
= =
O.072/day (SE 0.0035) and for Rakhine during the dry season，0. 0057/day (SE 0.0000).
=
官lemean defecation rate was 23.0/day (SE 1.5) and the mean decay rate of dung piles
were 0.0090/day (SE =0 .001) for AKNP( MYINT，19 94).
In Bago，fo r a2 85 kmt ransect survey with ad ung encounter rate of 2.41km， the elephant
density varied from 0.4 to 5e leph叩 tslkm2and an average density of 1.6e lephantslkm2
(95% Confidence Interval CI =1 .4-1.7e lephantslkm2)c ould be estimated for all the regions
surveyed in Bago. The elephant survey in Rakhine is based on a3 00kmtransect survey
開
with ad ung encounter rate of 1.21km. The density estimates for different regions in Rakhine
varied from 0.008 to 0.2 with an average density of elephant for all the regions surveyed
=
being 0.05 elephantslkm2 (95% CI 0.04--0.06). For AKNP，a d ensity ofO.61 elephantslkm2
=
(95% CI 0.5--0.7) was estimated through a5 0・kmsurvey with ad ung encounter rate of
0.851km. The encounter rate of dung piles indicated that the elephants were found to be using
only the southwest regions of the park.
For Bago，es timated mean density (21km2) and number (2600) appear to be very high.
Considering the amount of logging and other disturbances (SALTER，18 93; HTUT，19 94)，an d
also the numbers presented for Bago (MYINT，1 994)，on ly 300 elephants could be expected
for the regions surveyed. Earlier observation of MYINT (1994) also suggests that the density
62 SURENDRA VA RMA，U YE HTUT AND U UOA
estimated during the current study of Rakhine to be acceptable. For ah abitat area of 16，000
km2i n Rakhine，M YINT (1994) estimated 750 elephants. Although the habitat is disturbed
by logging and other activities，th e relatively low human population density，in accessibility，
and large and contiguous forest cover offer some hope for the conservation and management
of elephants in Rakhine.
For AKNP，t he overall elephant dung encounter rate was 0.8/km (varying between 0
and 3/km) and most other regions surveyed had no dung along the transects. MY町T(1994)
estimated ad ensity of 0.09 elephants/km (95% CI =0 .04-0.15) for AKNPa nd our density
estimate of 0.61 is six times higher than the earlier estimate. As observed by MYINT (1994)，
it is likely that elephants do not use some of the regions of由ep紅kand the estimated density
thus cannot be extrapolated for the entire park. The lack of sightings or low density of
elephants could be due to logging in the past，ot her disturbances，an d distribution of food and
other resources. Only 20% of locations surveyed had elephant dung piles，an d if the density
estimates are extrapolated to only as mall portion (say 10 to 20%) of AKNP，on ly 100 to 190
elephants would be predicted for the park.
Many other reasons also could be identified for the high density of elephants in some
of the reserved surveyed. Two extreme dung decay rates were found for the dry and the wet
seasons. The dung decay rate was high during the wet season，bu t low or slow in the dry
season. If the decay rate of the old dung piles is very slow but f詑shdung disappear faster
then the old during the wet season，s ome of the old dung piles from the dry season would
also remain in the wet season and increase the dung density，re sulting in over estimation of
density.
As this is the first scientific study of its kind in Myanmar，c omparative figures on
population density are not available from anywhere in the country. However，it is important
to mention that the encounter rates of dung piles are relatively lower than reported for some
of the elephant habitats in India. The encounter rate of dung piles/km was 0.14ー1.1times
that of elephant habitats in India for Bago，0. 07--0.59 that in Rakhine，a nd 0.05--0.44 that
in AKNP. Considering the encounter rate of dung piles/km and density estimates of the
elephant habitats in India，a nd inco中oratingthe elephant dung encounter rate/km of the
regions surveyed in Myanmar，a d ensity range of 0.1--0.3 elephants/km2f or Bago，0. 05--0.13
elephants/km2f or Rakhine and 0.03--0.1 elephants/km2f or AKNPw ould result (Table 3).
Table 3: Comparison of results across elephant habitats in India and Myanmar
Name of region Encounter rat Estimateddensity
怠
of dung/km (km-2)
India:
Mudumalai Wildlife Sanctuary，so uthem India 15.7 2且勺，内、4
Buxa Tiger Reserve，we st Bengal 8.7 ハυd5
Kalakadu Mundanthurai Tiger Reserve，s outhem India 2.0 nuF441
Myanmar:
Bago Yoma 2.3 1.61
Rakhine Yoma 1.2 0.051
AKNP 0.9 0.6
Mean elephant densities
I
POPULATiON EVALUTION OF ASIAN ELEPHANT (ELEPHAS MAXIMUS) 63
Figure 2. Logging operation.
Figure 3. Logging operation.
Figure 4. Remnants of the logging operation.
64
SURENDRA VARIVIA, U YE HTUT AND U UGA
Figure 5. Current status of the habitat.
Figure 6. Current status of habitat.
Figure 7. Training program for the saurvey.