MARINE MAMMAL SCIENCE, 21(3):536-549 (July 2005)

0 2005 by the Society for Marine Mammalogy

DUGONG (DUGONG DUGON) ABUNDANCE

ALONG THE ANDAMAN COAST OF THAILAND
ELLENM. HINES
Department of Geography and Human Environmental Studies,
San Francisco State University, 1600 Holloway Avenue,
San Francisco, California 94132, U.S.A.
E-mail: ehines@sfsu.edu

KANJANA ADULYANUKOSOL
Phuket Marine Biological Center, P. 0. Box 60, Phuket 83000 Thailand

DAVIDA. DUFFUS
Whale Research Lab, Department of Geography,
University of Victoria, Victoria,
British Columbia V8W 3P5, Canada

ABSTRACT
In 2000 and 2001, dugong abundance was estimated using aerial surveys in three
provinces along the Andaman coast of Thailand. A microlite aircraft was used to fly
aerial transects over seagrass areas. All surveys were done during rising tides as the
dugongs came to the seagrass beds to feed. The largest population was found in Trang
province. In Trang, the total number of sightings during 22 surveys was 264, out
of which 31.5% were single dugongs. The largest group seen in 2000 was 30, and
in 2001, 53. The maximum number of calves seen in one day was 13. The best
minimum estimate ofpopulation abundance is 123 animals (CV= 60.8%) in Trang
province. Higher numbers of dugong sightings and group sizes corresponded with
higher tides until water turbidity impeded sightings after the highest spring tide.
In other areas the number of animals seen was too small for population estimates.
Key words: dugong, Dugong dugon, population assessment, Thailand, Andaman Sea.

T h e dugong (Dugong dugon) is critically endangered. Outside of Australia, the

country with the largest estimated dugong population, dugongs only survive i n
fragmented populations in the Eastern hemisphere. Neither t h e number of dugongs
remaining in these populations nor its range is known beyond incidental sightings
and the reports of fishermen (Marsh et al. 1999, 2002). I n Thailand the dugong has
been under Federal protection since 1947 (Humphrey and Bain 1990). O u r study
along t h e Andaman coast is the first systematic dugong population survey in
Thailand (also see Hines 2 0 0 2 4 b).
There are five population centers along the Andaman coast: Ranong, Phuket,
Krabi, Trang, and Satun (Fig. 1). Aerial surveys were flown near Phuket, Krabi, and
Trang. Many of the seagrass beds surrounding the island of Phuket have been

536
 F i g w e 1. The circled areas are aerial survey locations along the Andarnan coast of
Thailand in 2000 and 2001.

degraded by coastal development (Poochaviranon 2000). Krabi, on the eastern coast

of Phang-nga Bay, has seagrass beds on Yao Noi and Yao Yai Islands in the Bay, also
at Sriboya, Pu, and Hang Islands (hereafter referred to as the Three Islands), and on
the eastern coast of Phang-nga Bay (Fig. 1). The largest seagrass beds along this
538 MARINE MAMMAL SCIENCE. VOL. 21. NO. 7. 2005

coast are in Trang province. There are two beds equaling 7 km2 at Libong island,
and 6.36 km2 in Had Chao Mai National Park (Chansang and Poovachiranon 1994)
(Fig. 1). In this area is also the largest known remnant population of dugongs in
Thailand (Adulyanukosol et at. 1997).
Our overall objective was to estimate dugong abundance and distribution along
the Andaman coast of Thailand. Our second objective was to develop practical field
protocols for future monitoring of dugong populations with appropriate spatial
scale, and sensitive to local budgetary constraints and logistical support (Aragones
et al. 1997).

MEmoDs
The only practical method to directly count dugongs is from the air (Hudson
1981, Marsh 1995a). Aerial surveys have been widely used in other parts of the
species range (Rathbun et al. 1988, Chambers et al. 1989, de Iongh 1995, Marsh
19956).
We first used a helicopter to perform extended area surveys along the entire coast
to determine the overall distribution and obtain a qualitative estimate of the
abundance of dugongs and seagrass (Lefebvre et al. 1995, Aragones et al. 1997).
Results of the helicopter surveys can be found in Hines (2002a).A microlite fixed-
wing aircraft was then used for more detailed surveys in areas that either had
sightings in the helicopter surveys, or were reported to have groups of dugongs by
interview respondents or Thai researchers (Aragones et al. 1997).
We used transect surveys to sample representative areas and estimate abundance
and densities of dugongs for the total area (Burnham et al. 1980). While line-
transect sampling is recommended for cetaceans (Barlow 1988, Forney et al. 1991),
dugongs are more difficult to sight because they are often solitary or in cow-calf
pairs, and as their surfacing time is from 1 to 2 s, are difficult to locate in turbid
water (Marsh 1995a; Pollack et al., in press). Strip, or fixed-width transect tech-
niques are recommended for dugong aerial surveys, and for other marine mammal
aerial surveys where the aircraft is moving rapidly. At the speeds required to safely
fly an open aircraft it is difficult to correctly assess perpendicular distances from the
aircraft while searching the field of view and counting animals (Marsh 1995a,
Miller et al. 1998).
The microlite aircraft was used to fly predetermined fixed transects to estimate
population distributions and counts in Trang province in both 2000 and 2001 and
in the Three Islands area in Krabi in 2000. in 2001 we flew transects along the
northeast coast of Phuket island in Phuket province, and the Yao islands in Phang-
nga Bay. Transects were perpendicular to shore and took advantage of visible
endpoints both on land and in the water (e.g., buoys, limestone outcroppings).
Transect lines should ideally be placed parallel to the density gradient of the
distribution of the group of animals (Yoshida et al. 1997). Dugongs are generally
found in shallow waters. While feeding they are mostly in sea grass beds 11-12 m
deep (Reynolds and Ode11 1991). Therefore, the density gradient is most probably
perpendicular to the shoreline, and we placed the transect lines accordingly. We
used a zigzag pattern to equalize the effort on all parts of the transect line
(Dahlheim et al. 2000, Buckland et al. 2001). In each survey area we estimated the
time it took to turn the microlite from one transect to another, and subtracted that
time from effort calculations. In 2001 we added a small area and four more transect
 HINES ETAL.: DUGONG DUGON 539

Figure 2. Transects and strata used for microlite surveys in Trang Province. The estuary
sampling block was added in 2001.

lines to the Trang transect to survey a nearby estuary where the seagrass had in-
creased since 2000. Local contacts reported dugongs during the past year.
In Trang we flew 45 transects in 2000 and 49 transects in 2001 (Fig. 2). In the
Three Islands area, we flew 20 zigzag transects (Fig. 3). In Phuket, we flew 9
transects, and in the Yao Islands, 36 transects (Fig. 4).To calculate differences in
density, we created sampling strata around groups of transects (Table 1) (Bayliss
1986, Marsh 1995a).
The transects were flown at between 400 and 500 m apart, starting from 1.5 to
2.5 h before high tide (Tide Tables, Hydrographic Department, Royal Thai Navy
2000). Transects were flown at an average height of 152 m and an average speed of
540 MARINE MAMMAL SCIENCE, VOL. 21, NO. 3 , 2005

F i g w e 3. Transects and strata used for microlite surveys in the Three Islands area, Krabi
Province.

46 kn (85 km/h). Visibility from the microlite was 200 m on each side. We
established this measurement by marking an area of the beach using a measuring
tape and then placing markers on the sand at 400 m as could be seen from the
microlite at survey altitude. We then marked the struts of the microlite
corresponding with that distance.
Two hundred meter visibility from an aircraft has been calculated for 137 m of
altitude by Marsh eta/. (1996). This altitude is considered sufficiently low to pre-
vent repeated sightings while allowing adequate visibility (Marsh and Saalfeld 1990,
Marsh 1995a). The higher altitude flown in our study was necessary for safety in
the microlite.
 HINES ET AL.: DUGONG DUGON 541

Figure 4. Transects and strata used for microlite surveys in Ban Pak Lok (Phuket
Province) and the Yao Islands (Phang-nga Province).

Due to the maneuverability and relative quietness of the aircraft, when an animal
or group of animals was seen, it was easy to circle at lower altitude to gather further
information. Species, number of calves, group size, and behavior were documented
before returning to the transect line.
Latitude and longitude of all sightings and routes were recorded using a global
positioning system (GPS). We also recorded the time, weather, tidal level, and
Beaufott sea state (see Hines 2002a).
We used the program DISTANCE for estimation of population abundance and
density (Thomas et af. 1998, Buckland et al. 2001). To analyze strip transects in
DISTANCE, we used a uniform detection function model with 0 adjustments terms
(Buckland et al. 2001). A detection function is g(x), or the probability of detecting
542 MARINE MAMMAL SCIENCE, VOL. 21, NO. 3, 2005

Table 1. Sampling strata and lengths of strip transects by area.

Strata Area Length of

Area Year number (kmz) transect (km)
Trang 200012001 1 29.24 59.11
Trang 200012001 2 3.25 6.73
Trang 2001 3 1.68 6.02
Trang 200012001 4 29.29 42.77
Trang 2000/2001 5 3 1.43 53.58
Three Islands 2000 1 60.28 105.19
Phuket 2001 1 58.56 55.13
Yao Islands 2001 1 20.01 29.40
Yao Islands 2001 2 8.82 15.10
Yao Islands 2001 3 2.95 14.62

an object, given that it is at a distance x from the random point or line (Thomas
et ul. 1998). Under these model definitions, the program assumes a probability of
detection of 1 between 0 distance from the observer and the truncation distance. In
sample sizes approaching 1, a probability density function for the width of the
transect [(fO)] cannot be calculated, and density and abundance cannot be estimated
reliably (Jefferson 2000, Buckland et ul. 2001).
Strip ttansects are a strip or strata of length L and width 220. In a formal strip
transect the assumption is made that all objects within the census area are detected
(Buckland et ul. 2001). The uniform detection function is the only model that fits
for strip transects as g(x) does not change over a set distance, and no adjustment
terms are needed.'
DISTANCE estimates D , or the number of animals per unit area as

D = n/2wL (1)
Therefore, we set the truncation distance at 200 m, or half the estimated width
of the strip.' We interpreted the maximum value of n as given in DISTANCE, or
estimate of animals in a specified area, as a minimum estimate of population abun-
dance (Jefferson 2000).
To examine the affects of tide level on our ability to see dugongs, we separated
the 2001 sightings into truncated tide level categories. We then ran abundance and
density estimates on each category.

RESULTS
Trung Province
In 2000 we conducted aerial surveys in TPdng Province for 10 d between 6 and
20 March. The average daily count of dugongs and calves was 38.2 (Table 2). All
dugongs sighted were in 5 5 - m depth lowest low water (sea level) as measured by
bathymetric maps from the Hydrographic Department of the Royal Thai Navy. We
recorded a total of 52 calves, an average of 5.2 daily. A minimum of 2 and a max-
imum of 13 calves were located in one day with a standard deviation of 3.6. Adults
' Personal communication from Stephen T. Buckland, CREEM, The Observatory, Buchanan
Gardens, St Andrews KYl6 9L2, Scotland, 14 October 2004.
 HINES ET AL.: DUGONG DUGON 543

Table 2. Estimates of abundance, density, and other associated parameters for the
microlite surveys in Rang province in 2000 and 2001. Symbols used: R, numbers of sam-
ples (the number of transect lines); L, total length of transects surveyed; # of animals seen,
the actual count of dugongs including calves; n , number of on-effort sightings; average
group size, average number of animals in group, nlL, encounter rate; D, estimate of density
of animals; N, estimate of animals in specified area; SE, standard error; CV, coefficient
of variation.

# of Average
Area L animals group cv
Year Day (sq. km) R (km) seen n size nlL D N SE (%)
2000 6 March 93.2 45 162.2 62 19 3.3 0.1 1.0 89 39.8 44.7
7 March 36 16 2.3 0.1 0.6 52 17.5 33.7
8 March 37 12 3.1 0.1 0.6 53 25.9 48.9
9 March 34 13 2.6 0.1 0.1 49 16.1 32.9
10 March 23 11 2.1 0.1 0.4 33 14.0 42.3
13 March 15 10 1.5 0.1 0.2 22 7.9 35.9
17 March 23 15 1.5 0.1 0.4 33 8.8 26.8
18 March 45 13 3.5 0.1 0.7 65 22.5 34.6
19 March 66 13 5.1 0.1 1.0 95 51.9 54.7
20 March 41 18 2.3 0.1 0.6 59 16.2 27.5
Total 2000 382 140 550
Average
2000 38.2 14.0 2.7 0.1 0.5 55 22.1 38.2
2001 31 March 94.9 49 169.5 10 8 1.3 0.0 0.1 14 4.9 34.8
1 April 11 9 1.2 0.1 0.2 15 4.7 31.2
2 April 5 5 1.0 0.0 0.1 7 2.8 39.9
3 April 7 6 1.2 0.0 0.0 10 4.8 48.2
4 April 6 6 1.0 0.0 0.1 8 3.0 37.3
5 April 42 20 2.1 0.1 0.6 59 13.3 22.6
6 April 79 16 4.9 0.1 1.2 111 78.2 70.4
7 April 88 9 8.8 0.1 1.3 123 87.0 60.8
8 April 55 12 4.6 0.1 0.8 77 49.6 64.4
10 April 43 9 4.8 0.1 0.6 60 30.7 51.2
11 April 35 13 2.7 0.1 0.5 49 20.4 41.5
12 April 16 11 1.5 0.1 0.2 22 7.8 35.6
Total 2001 397 124 555
Average
200 1 33.1 10.3 2.9 0.1 0.5 46.3 25.6 44.8
Average
2000
and 2001 35.4 12.0 2.8 0.1 0.5 50.2 24.041.8
Totals for
both years 779 264

(n = 330) comprised 86% of the counts, and calves 14%. W e recorded 20.8 dugongs
per surveying hour. Of 186 sightings, single animals accounted for 75%. Only
three sightings (2%) were of groups over six (17, 26, and 30).
During the 2 0 0 1 field season i n Trang province, we completed 12 aerial surveys
between 3 1 March and 1 2 April. We counted an average of 33.1 dugongs and calves
per day (Table 2). All dugong sightings were again i n 55-m depths. T h e spatial
544 MARINE MAMMAL SCIENCE, VOL. 21, NO. 3 , 2005

Figure 5. Two examples of sighting locations in Trang province: (a) 6 April and (b) 7
April 2001.

distributions of sightings for two of the survey days are illustrated in Figure 5 . We
counted 46 calves in total, with mean sightings of 3.8 daily. Twelve percent of the
dugongs seen were calves and 88% ( n = 351) were adults. O n average we sighted
18.9 dugongs per hour.
We made a total of 124 separate sightings, out of which 31.5% were of single
animals. Groups of 5 3 constituted 42.6% of all sightings, and 45.6% were of
groups of six animals or larger. The largest group, 53, was seen on two
consecutive days in a row. For both years, the minimum estimate (the largest n) of
 HINES ETAL.: DUGONG DUGON 545

Table 3. Estimates of abundance, density, and other associated parameters for the
different strata on the microlite surveys in Trang province in 2000 and 2001. Symbols used:
R , numbers of samples (the number of transect lines); L total length of transects surveyed;
average n, average number of on-effort sightings; average group size, average number of
animals in a group, average nlL, encounter rate; average D , average estimate of density of
animals; average N,average estimate of animals in specified area; SE, standard error; CV,
coefficient of variation.

Average
Years Area L Average group Average Average Average CV
surveyed Strata (sq. km) k (km) n size nlL D N SE (%)
2000
and 2001 1 29.2 14 59.1 4.0 1.1 4.8 0.2 5.7 2.6 44.8
2000
and 2001 2 3.3 1 6.7 0.1 0.2 0.2 0.1 0.2 0.2 4.6
2001 3 1.7 4 7.3 0.3 0.4 0.5 0.2 0.3 0.3 33.4
2000
and 2001 4 29.3 17 42.8 3.6 5.1 20.4 1.2 34.8 25.1 67.0
2000
and 2001 5 31.4 13 53.6 4.4 1.9 9.8 0.4 13.2 6.7 55.9

the total population size in the area surveyed was 123 dugongs (Table 2). In the
tide level analysis, the numbers of dugong sightings and group sizes rose to a peak
level in the 3.0-3.4 medium tide level class, and fell at the highest spring tide
level.
Of the five sampling strata, the average population abundance estimate was
highest in stratum 4 (Table 3). The average group size and density of individual
animals was also highest here. The highest n for any survey day and therefore the
minimum population estimate for this sampling stratum was 98 animals.
Groups of dugongs gathered in the dense patches of sea grass, mostly Hulophiliu
ovalis, along the eastern end of Libong Island (Fig. 2). Smaller groups of dugongs
regularly fed on the more patchy sea grass along the south coast of the island. This
area, stratum 5, had an average group size of 1.9, and a minimum abundance es-
timate of 37 (from the survey day with the highest n). Between Muk Island and
Laem Yong Lam, another area with patchy sea grass, sampling block 1, an
estimated 12 dugongs foraged in average groups of 1.1 (Table 3) (see Hines
2002a).

Krabi Province
In Ktabi province, within the Three Islands area, after five transects, the max-
imum sighting was of four adult dugongs, plus one calf. The total number of
sightings over all five days was 12. Adults were 75% of the total, and calves, 25%.
The average number of sightings was 2.4, the minimum 0, and the maximum 5,
with a standard deviation of 2.5. The number of dugongs per hour in this area was
2.2. All dugongs were seen in 5 5 - m depths. The sampling block area here was
60.3 km’, and the transect length was 105.2 linear kilometers. The sample size in
this area was not sufficient for a reliable estimate of population abundance and
density.
5 46 MARINE MAMMAL SCIENCE, VOL. 21, NO. 3, 2005

Phuket and Pbang-nga PrwinceJ

We flew microlite surveys on the northeast coast of Phuket Island and around the
Yao Islands in Phang-nga Bay between March 23 and 25 in 2001. In Phuket, we
2
established one sampling block of 58.3 km , containing 55.1 linear kilometers of
transect. Three sampling blocks in the Yao Islands totaled 3 1.8 km’. The total linear
kilometers of the three transect paths equaled 88.5.
We sighted three dugongs, one near Ban Pak Lok on Phuket, and two in the Yao
Islands, none of which were calves. There were feeding trails in sea grass beds be-
tween Yao Noi and Yao Yai Islands, as well as in the sea grass in the sampling block
at the south end of Yao Yai Island. As in the surveys in other areas, no dugongs were
seen in waters 2 5 - m depth. As in Krabi, the sample size was too low to perform
statistical analysis.

DISCUSSION
We estimate that the present population of dugongs along the Andaman coast is
approximately 200 animals, with a minimum population abundance estimate of
123 animals in Trang province. Because we are, as of yet, unaware of the extent of
movement along the coast, protecting only some of these habitats, even those with
a larger population of dugongs, will not ensure the recovery of the entire pop-
ulation, even if local extinction is prevented. Both past and present habitat areas
need to be considered for protection (Preen 1998).
In the Kuraburi area, we cannot at this time make a reasonable minimum pop-
ulation estimate. Based on trails that we saw, the animals seen by villagers and
other researchers, and the extent and cover of the sea grass beds, the area can and
most likely does support a small population of dugongs.
The Yao Islands have dense beds of sea grass in two areas, the first on the north-
east coast of Yao Yai Island, and the second in a sheltered bay at the south end of
Yao Yai Island. We have located feeding trails during boat-based sea grass sampling
and aerial surveys at both these sites. There is sea grass in the middle of the west
coast of the same island, but it has been damaged by monsoon storms, and is sparse.
During the 2001 microlite survey, two dugongs were seen in the sea grass beds on
the northeast coast of Yao Yai Island. In 1999 Adulyanukosol saw a group of eight
dugongs between Yao Yai and Yao Noi Islands during a helicopter survey. Interview
respondents in these areas have reported regularly seeing groups of between 1 and
10 dugongs. Fishermen in this area think that there are approximately 15 dugongs
in the local sea grass beds. It is possible that a group of dugongs feed here regularly.
Adulyanukosol et a/. (1997) found five dugongs in the Three Islands area. In the
2000 microlite surveys, we saw up to four adults and one calf in one day. Villagers
told of seeing 10 dugongs in the past 3 yr. From the number of sightings, strandings,
and the condition and extent of sea grass, the Three Islands area has sea grass beds
that can support a small population of dugongs. We estimate that the minimum
population here at this time is between 5 and 10 dugongs.
In Trang province, for all strata combined, the best minimum estimate of the
population at this time is 123 dugongs. However, the coefficients of variation (CVs)
of the population estimates are relatively high (Table 2). Transect sampling design
and estimation of factors that control researchers’ ability to detect dugongs should
be further refined to reduce the CVs. In general, the highest levels of statistical
precision are indicated by CVs of <20%. At these percentages, the estimates are
 HINES ET AL.: DUGONG DUGON 547

considered more reliable, and can be dependably utilized as the bases for trend anal-
ysis in the future (Jefferson 2000).
To reduce the coefficients of variation i n the population abundance estimations,
the aerial transect methodology should be refined. Marsh and Sinclair (1989a, b)
and (Marsh 1 9 9 5 ~have) defined perception bias as animals potentially visible that
are overlooked by observers, and availability bias as the number of animals unseen
by observers because of uncontrollable factors such as turbidity, sea state, depth, and
animal behavior. Sighting availability along the Andaman coast was especially af-
fected by turbidity at various ride levels causing differences in abundance and
density estimates. At the highest spring tide levels, when the water depth was over
3.5 m , the level of turbidity of the water was too high to see animals that were not
at the surface.
Pollack et al. (in press) have refined merhods created by Marsh and Sinclair 1 9 8 9 b)~~~
to correct for these factors i n survey conditions and have reported lowered CV
estimates. W e recommend that these methods be adapted for local conditions and
implemented i n future surveys. W e also suggest that a larger aircraft be used with
room for at least one independent observer on each side of the plane to minimize
perception bias.
T h e transect methodology used here should be continued to increase the precision
of estimates. It is also important to establish transects and conduct surveys in the
secondary feeding sites. W h i l e a statistically powerful trend estimate is not currently
possible, more and regular surveys will provide a more complete picture of dugong
habitat and distribution along the Andarnan coast (Taylor and Gerrodette 1993,
Marsh 19956).

ACKNOWLEDGMENTS
We would like to acknowledge Ocean Park Conservation Foundation and the International
Development Research Centre for research funding. At the Phuket Marine Biological Center
we have been generously supported by the Director, Oonchit Bhatiyasevi, as well as Supot
Chantrapornsyl, Dr. Kongkiat Kittiwattanawong, and Sombat Poochaviranon. We thank
Dr. Paul Wade, Dr. Samantha Strindberg, Amy Kirk, Stephanie Olsen, Heather Patterson,
Monsiri Baird, Pisit Charnsoh, Suwan Pitaksintorn, Nimit Sittirod, Jim Enright, and two
anonymous reviewers. Dr. Cherdsak Virapat, Choomjet Karnjanakesorn, and Suriya
Saikrachang from the Thai Department of Fisheries were instrumental in getting us
support for our surveys.

CITED
LITERATURE
ADULYANUKOSOL, K., S. CHANTRAPORNSYL AND S. POOVACHIRANON. 1997. An aerial survey
of dugong (Dugong dugon) in Andaman coast, Thailand. Thai Fisheries Gazette 50:
3 59-3 74.
ARAGONES, L. V., T. A. JEFFERSON AND H. D. MARSH.1997. Marine mammal techniques
applicable in developing countries. Asian Marine Biology 14:15-39.
BARLOW,3. 1988. Harbor porpoise, Phocoena phocoena, abundance estimation for California,
Oregon and Washington: I. Ship surveys. Fishery Bulletin, U S . 86:417-43 1.
BAYLISS,
P. 1986. Factors affecting aerial surveys of marine fauna, and their relationship to
a census of dugongs in the coastal waters of the Northern territory. Australian Wildlife
Research 13:27-37.
BUCKLAND, S. T., D. R. ANDERSON, K. P. BURNHAM, J. L. LAAKE, D. L. BORCHERS AND L.
THOMAS. 2001. Introduction to distance sampling: Estimating abundance of biological
populations, Oxford University Press, Oxford, UK.
548 MARINE MAMMAL SCIENCE, VOL. 21. NO. 3. 2005

BURNHAM, K. P., D. R. ANDERSON AND J. L. LAAKE.1980. Estimation of density from line

transect sampling of biological populations. Wildlife Monograph 72:1-202.
CHAMBERS, M. R., E. BANIAND B. E. T. BARKER-HUDSON. 1989. The status of the dugongs
(Dugong dugon) in Vanuatu. South Pacific Regional Environmental Programme Topic
Review No. 37. South Pacific Commission, Noumea, New Caledonia.
CHANSANG, H., AND S. POOVACHIRANON. 1994. The distribution and species composition of
seagrass beds along the Andaman Sea coast of Thailand. Phuket Marine Biological
Center Research Bulletin 59:43-52.
DAHLHEIM, M., A. YORK,R. TOWELL, J. WAITEAND J. BREIWICK. 2000. Harbor porpoise
(Phocoena phocoena) abundance in Alaska: Bristol Bay to Southeast Alaska, 1991-1993.
Marine Mammal Science 1 6 : 2 8 4 5 .
DE IONGH,H. H . 1995. Aerial survey of the dugong (Dugong dugon Muller, 1776) in coastal
waters of the Lease Islands, East Indonesia. Marine and Freshwater Research 46:
7 59-76 1.
K. A., D. A. HANAN
FORNEY, AND J. BARLOW. 1991. Detecting trends in harbor porpoise
abundance from aerial surveys using analysis of covariance, Fishery Bulletin, U.S.
89x367-377.
HINES,E. 2002a. Conservation of the dugong (Dugong dugon) along the Andaman Coast of
Thailand: An example of the integration of conservation and biology in endangered
species research. Ph.D. dissertation, University of Victoria, Victoria, BC. 29 1 pp.
HINES,E. 20026. Conservation of the dugong (Dugong dugon) along the Andaman coast
of Thailand. Pages 155-180 in P. Dearden, ed. Environmental protection and rural
development in Thailand: Challenges and opportunities. White Lotus Press,
Bangkok, Thai land.
HUDSON,B . E. T. 1981, Interview and aerial survey data in relation to resource management
of the dugong in Manus province, Papua New Guinea. Bulletin of Marine Science 3 1:
662-672.
HUMPHREY, S. R., AND J. R. BAIN. 1990. Endangered animals of Thailand. Sandhill Crane
Press, Inc. Gainesville, FL.
JEFFERSON, T. A. 2000. Population biology of the Indo-Pacific hump-backed dolphin in
Hong Kong waters. Wildlife Monographs 144.
LEFEBVRE, L. W., B. B. ACKERMAN, K. M. PORTERAND K. H. POLLOCK. 1995. Aerial survey
as a technique for estimating trends in manatee population size-problems and
prospects. Pages 63-74 in T. J. O’Shea, B. B. Ackerman and H. F. Percival, eds.
Population biology of the Florida manatee. Information and Technology Report 1,
U. S. Department of the Interior, National Biological Service, Washington, DC.
MARSH,H . 1 9 9 5 ~ .Fixed-width aerial transects for determining dugong populations
sizes and distribution patterns. Pages 56-62 in T. J. O’Shea, B. B. Ackerman and
H. F. Percival, eds. Population biology of the Florida manatee. Information and
Technology Report 1, U. S. Department of the Interior, National Biological Service,
Washington, DC.
MARSH,H . 19956. The life history, pattern of breeding, and population dynamics of the
dugong. Pages 75-83 in T. J. O’Shea, B. B. Ackerman and H. F. Percival, eds.
Population biology of the Florida manatee. Information and Technology Report 1,
U. S. Department of the Interior, National Biological Service, Washington, DC.
MARSH,H . 1995~.The limits of detectable change. Pages 122-130 in C . G. Grigg, P. T.
Hale and D. Lunney, eds. Conservation through sustainable use of wildlife. Centre for
Conservation Biology, University of Queensland, Australia.
MARSH,H., AND W. K. SAALFELD. 1990. The distribution and abundance of dugongs in the
Great Barrier Reef Marine Park south of Cape Bedford. Australian Wildlife Research
17:511-524.
MARSH,H., AND D. E SINCLAIR. ~ . experimental evaluation of dugong and sea turtle
1 9 8 9 ~An
aerial survey techniques. Australian Wildlife Research 16:639-650.
 HINES ETAL.: DUGONG DUGON 549

MARSH,H., AND D. F. SINCLAIR. 19896. Correcting for visibility bias in strip transect aerial
surveys of aquatic fauna. Journal of Wildlife Management 53:1017-1024.
MARSH,H., P. CORKERON, I. R. LAWLER, J. M. LANYON AND A. R. PREEN.1996. The status
of the dugong in the southern Great Barrier Reef Marine Park. GBRMP Research
Publication No. 41. Queensland, Australia. 80 pp.
MARSH,H., C. EROS,P. CORKERON AND B. BREEN. 1999. A conservation strategy for dugongs:
Implications of Australian research. Marine and Freshwater Research 50:979-990.
MARSH,H., H . PENROSE, C. EROSAND J. HUGUES. 2002. Dugong: Status reports and action
plans for countries and territories. UNEP/DEWA/RS.02-1.
MILLER, K. E., B. B. ACKERMAN, L. W. LEFEBVRE AND K. B. CLIFTON. 1998. An evaluation of
strip-transect aerial survey methods for monitoring manatee populations in Florida.
Wildlife Society Bulletin 26:561-570.
POLLOCK, K. H., H. D. MARSH,I. R. LAWLER AND M. W. ALLDREDGE. In press. Estimating
animal abundance in heterogeneous environments: An application to aerial surveys for
dugongs. Journal of Wildlife Management.
POOCHAVIRANON, S. 2000. Species composition and their depth distribution of seagrass beds
along the Andaman Sea coast of Thailand. Biologica Marine Mediterranea 7:412-416.
PREEN,A. 1998. Marine protected areas and dugong conservation along Australia’s Indian
Ocean coast. Environmental Management 22:173-181.
RATHBUN, G. B., R. L. BROWNELL, K. RALLS AND J. ENGBRING. 1988. Status of dugongs in
waters around Palau. Marine Mammal Science 4265-270.
REYNOLDS, J. E., AND D. K. ODELL.1991. Manatees and dugongs. Facts on File, New
York, NY.
TAYLOR,B. L. AND T. GERRODETTE. 1993. The use of statistical power in conservation
biology: The vaquita and the northern spotted owl. Conservation Biology 7:489-500.
THOMAS, L., J. L. LAAKE, J. F. DERRY,S. T. BUCKLAND, D. L. BORCHERS, D. R. ANDERSON,
K. P. BURNHAM, S. STRINDBERG, S. L. HEDLEY,M. L. BURT,F. MARQUES, J. H . POLLARD
AND R. M. FEWSTER. 1998.Distance 3.5. Release 6. Research Unit for Wildlife Popula-
tion Assessment, University of St. Andrews, UK. Available from http://www.ruwpa.
st-and.ac.uk/distance/
YOSHIDA, H., K. SHIRAKIHARA, H. KISHINOAND M. SHIRAKIHARA. 1997. A population size
estimate of the finless porpoise, Neophocaena phocaenoides, from aerial sighting surveys
in Ariake Sound and Tachibana Bay, Japan. Researches on Population Ecology 39:
239-247.
Received: 29 November 2003
Accepted: 18 November 2004