BookPDF Available

Cambodian Journal of Natural History Volume 2013 Issue 1

August 2013

Publisher: Royal University of Phnom Penh
Editor: Jennifer C. Daltry, Hang Chanthon, Neil M. Furey

Authors:

Re:wild and Fauna & Flora

Left: The Kingdom of Cambodia, showing the Phnom Kulen National Park, Siem Reap Province. Right: Phnom Kulen National Park, showing sampling localities and main water bodies. Maps designed using ArcGIS 9.3.

…

Location of botanical plots (dots) within Mondulkiri Protected Forest and Phnom Prich Wildlife Sanctuary, Mondulkiri, Cambodia. Semi-evergreen forests in the protected areas are indicated in black.

…

O Tauch Community Protected Area, Bokor National Park, Kampot Province.

…

Att ributes and levels used in the Choice Models.

…

Site A-Phnom Kbal Spean, Bantey Srei District, at 100 m a.s.l. (© T. Hartmann).

…

Figures - uploaded by Jennifer C. Daltry

Content may be subject to copyright.

Content uploaded by Jennifer C. Daltry

Content may be subject to copyright.

Cambodian Journal

of Natural History

August 2013 Vol 2013 No. 1

Identifying bats from echolocation calls

The real price of sustainable bamboo

Farming butterﬂ ies in Siem Reap

A survey of freshwater ﬁ shes

Describing new species

Deciduous forests

Cambodian Journal of Natural History

ISSN 2226–969X

Editors

Email: Editor.CJNH@gmail.com

• Dr Jenny C. Daltry, Senior Conservation Biologist, Fauna & Flora International.

• Dr Neil M. Furey, Research Adviser, Fauna & Flora International: Cambodia Programme.

• Hang Chanthon, Former Vice-Rector, Royal University of Phnom Penh.

• Dr Stephen J. Browne, Fauna & Flora International,

Singapore.

• Dr Martin Fisher, Editor of Oryx—The International

Journal of Conservation, Cambridge, United Kingdom.

• Dr L. Lee Grismer, La Sierra University, California,

USA.

• Dr Knud E. Heller, Nykøbing Falster Zoo, Denmark.

• Dr Sovanmoly Hul, Muséum National d’Histoire

Naturelle, Paris, France.

• Dr Andy L. Maxwell, World Wide Fund for Nature,

Cambodia.

• Dr Jörg Menzel, University of Bonn, Germany.

• Dr Brad Pe i , Murdoch University, Australia.

• Dr Campbell O. Webb, Harvard University Herbaria,

USA.

Other peer reviewers for this volume

The Cambodian Journal of Natural History (ISSN 2226–969X) is a free journal published by the Centre for Biodiversity

Conservation, Royal University of Phnom Penh. The Centre for Biodiversity Conservation is a non-proﬁ t making

unit dedicated to training Cambodian biologists and to the study and conservation of Cambodian biodiversity.

• Dr Arjan Boonman, Queen Mary University of

London, United Kingdom, and Indonesian Institute of

Sciences (LIPI), Bogor, Indonesia.

• Dr Stuart Davies, Smithsonian Institution.

Washington DC, USA.

• Dr James Guest, University of New South Wales,

Sydney, Australia.

• Dr Alice Hughes, University of Bristol, United

Kingdom.

• Dr Eriko Ito, Hokkaido Research Center, Forestry and

Forest Products Research Institute (FFPRI), Hokkaido,

Japan.

• Durai Jayaraman, Fauna & Flora International,

Phnom Penh, Cambodia.

• Prof. Maurice Ko elat, Cornol, Swi erland, and

Raﬄ es Museum of Biodiversity Research, National

University of Singapore.

• Alexander Monastyrski, Vietnam-Russia Tropical

Centre, Hanoi, Vietnam.

• Berry Mulligan, Fauna & Flora International, Phnom

Penh, Cambodia.

• Prof. Walter Rainboth, University of Wisconsin

Oshkosh, USA.

• Dr Nicholas J. Souter, Centre for Biodiversity

Conservation and Fauna & Flora International, Phnom

Penh, Cambodia.

• Tim Wood, Phnom Penh, Cambodia.

Cover photo: The Cambodian tailorbird Orthotomus chaktomuk made international headlines when it was named

in 2013, after being discovered at a construction site on the outskirts of Phnom Penh (© Ashish John, Wildlife

Conservation Society). In the Guest Editorial, one of the authors of the new bird, Jonathan Eames OBE, gives a

personal perspective on where and how to describe new species of animals.

International Editorial Board

Editorial

Guest Editorial—Describing new species

Jonathan C. EAMES

BirdLife International Cambodia Programme, #9, Street 29 Tonle Basac, Chamkarmon, P.O. Box 2686, Phnom Penh,

Cambodia.

Email Jonathan.Eames@birdlife.org

In the previous issue of the Cambodian Journal of

Natural History (Volume 2012, No. 2), the editors

provided guidance on how to draft a good scientiﬁ c

paper (Daltry et al., 2012). Following the recent publi-

cation of the type description of a new species of bird

from Cambodia in June 2013 (Mahood et al., 2013), I

thought it would be helpful to continue this theme by

providing guidance to authors on what to include in

a type description for a new animal. This editorial is

focused on bird and, to a lesser extent, mammal type

descriptions because this is a ﬁ eld where I have some

experience. The International Code of Zoological Nomen-

clature is a system of internationally accepted rules

and recommendations for naming animals. The Inter-

national Code of Nomenclature for algae, fungi, and plants

(McNeill et al., 2012), previously the International Code

for Botanical Nomenclature, contains some diﬀ erences

and those wishing to describe taxa from these groups

of organisms should refer to it instead.

It could be said that all humans are taxonomists. As

a species we devote considerable time in our lives to

sorting and classifying objects without perhaps even

thinking much about what we are doing. This would

include such mundane tasks such as sorting ripe from

unripe fruit, deciding which Premier League club is

worthy of our support, or even choosing a lipstick in a

department store.

The urge to discover is one of the strongest

motivations a human can experience. Discovery in

all its forms enriches our lives endlessly. Scientiﬁ c

discovery can be a revelation upon which the future of

civilizations turns. For example, the discovery of the

healing properties of penicillin in 1928 has since saved

millions of human lives. Species discovery is thrilling

too, but for many scientists it may come only once in

a career. Once we have discovered a new species we

must describe it. It is therefore important to try and

get it right.

Published guidance exists on what to publish in

a species description, for example Winston (1999).

One of the most useful papers on the subject was

wri en partly in response to the disappointment of

the authors with the quality of many of the descrip-

tions of new species of bird then appearing (LeCroy

& Vuilleumier, 1992). I commend both this paper and

a retort (Collar, 1999) to all, and unashamedly draw

heavily from the former in what now follows. I also

include recent examples of bad and good practice as

revealed in the type descriptions of mammals and

birds recently published from Asia and elsewhere.

When describing a new species the starting point is

to designate a “holotype” or two or more “syntypes”.

The holotype is a single voucher specimen used by an

author to deﬁ ne and represent the species, and this

may be the only one found or one of several individ-

uals found. When people talk about “the type” they

are referring to this specimen (Winston, 1999). The

syntypes are two or more specimens selected from the

available material to represent the species when no

single specimen has been identiﬁ ed as the holotype

(Winston, 1999). To facilitate future comparisons, the

holotype or syntypes should be complete specimens

and not unsupported illustrations, body parts, and

blood or tissue samples. A “type series”, comprising

the holotype or syntypes and additional specimens

(called “paratypes”) is desirable because this helps to

demonstrate variability within the new species.

While it is highly desirable to have the type

specimen(s) permanently deposited in a museum or

other publicly available collection, very occasionally it

may be impractical to kill an individual, for example a

highly endangered mammal (International Commis-

sion on Zoological Nomenclature, 2000). Authors and

editors that deviate from the convention of collecting

type specimens risk confusing the scientiﬁ c record and

the ire of their peers, as the following recent examples

clearly show.

In 2005 a new monkey was described from a photo-

graph of an individual animal (Jones et al., 2005). The

description of a new primate is a major event and

resulted in the type description being published in the

journal Science. However, the authors, working for a

Editorial

leading conservation organisation, chose to describe

their new species from photographs instead of killing

and preserving a specimen, perhaps because of the

rarity of the animal or because it was a monkey. Thus

the “holotype“ in this case was the animal depicted

in the photograph, but the International Code on

Zoological Nomenclature (International Commission

on Zoological Nomenclature, 2000) does not permit

photographs to be designated as types. In the absence

of a voucher specimen, and therefore being obliged

to describe the taxon from images only, the monkey

was placed in the genus Lophocebus, which contains

many species (Jones et al., 2005). Later research by a

leading museum, which chose to collect a specimen,

placed the monkey in the entirely new monotypic

genus Rungwecebus (Davenport et al., 2006) on the

basis of molecular and morphological data. Had the

type description been based on a voucher specimen,

a far more thorough piece of science could have been

published, establishing a new genus and new species

in a single paper and giving greater kudos to the

authors.

Another example serves to show how confusion

can arise when a complete specimen is not obtained

and described. In 2006 a new species of babbler was

described from India (Athreya, 2006). Whilst the

author’s reasons for not collecting a voucher specimen

were given in the paper, and whilst the type descrip-

tion may meet the provisions of the International Code

on Zoological Nomenclature, the absence of a complete

type specimen and the designation of the image of

the bird in photographs as type material renders it

ﬂ awed and of limited utility to others. Athreya (2006)

wrote: “The holotype is the bird from which a few

feathers were obtained and which is the subject in a

series of photographs presented in this paper. The

holotype was captured, photographed, measured and

released”. Since the holotype was released alive, do

the feathers, the photographs or the released bird or

all three represent the holotype? The absence of any

complete voucher specimens renders it almost impos-

sible for future researchers to make comparisons with

congeners.

In another, now infamous example, the Bulo Burti

boubou Laniarius liberatus, an African bushshrike,

was described from blood samples only and lacked

a specimen (Smith et al., 1991). New species can be

described on the basis of DNA sequences, but, while

not mandatory, it is strongly recommended that the

type specimen(s) from which the DNA was sequenced

is preserved and deposited in a museum with a type

label and data linking it to the sequence (International

Commission on Zoological Nomenclature, 2000).

The provenance of type material is also critically

important. In most recent cases involving vertebrates,

it is typical for the collector to be one of the authors

of the type description. Thus the provenance of the

type material is usually accurately known. Recently,

however, from this very region, there was an alarming

example where this was not the case. In 1994 the type

description of the khiting vhor Pseudonovibos spiralis

was based on preserved material purchased in a shop

(Peter & Feiler, 1994). Whilst the authors acted in good

faith, they were rash in rushing to publish a descrip-

tion based on material that was fraudulently crafted

from cow horn.

The information that should accompany a type

specimen includes a collection catalogue number,

the name of the institution where the type is depos-

ited, its age and sex, collecting locality (including

coordinates and altitude), date of collection, the name

of the collector(s), biometrics and a detailed word

description of the type. The inclusion of additional

information to help us to judge the validity of the

species is also advisable. This may include sonograms

of voice recordings (in the case of birds or frogs, for

example), tissue and blood samples and notes on

behaviour and ecology (LeCroy & Vuilleumier, 1992).

Two comprehensive examples of bird type descrip-

tions containing such comprehensive information

include Alstrom et al. (2010) and Mahood et al. (2013).

As the new species will bear a scientiﬁ c name

derived from Latin or Ancient Greek, the etymology

and gender of the proposed name must be given. For

most of us this means we must seek guidance from

a scholar in these obscure languages. The authors

must explain why the new species is included within

a particular genus, including any new genus they may

propose. Importantly, comparisons must be made

with closely related congeners, including sympatric

and allopatric forms, and maps showing geograph-

ical relationships included. This may render lengthy

and costly overseas trips to museums in Europe or

the United States necessary to examine specimens of

previously described species. The biogeography of

the new species should be discussed and an explana-

tion given as to why if the new taxon is allopatric, it

is a new species and not a new subspecies (LeCroy &

Vuilleumier, 1992).

In the case of Athreya’s (2006) babbler, the compar-

ison of this new taxon with its close congeners was

limited to comparing photographs only. This is not

reliable because photographs do not capture colour

precisely, and no direct comparison was made with

the two most closely related species. The comparison

Editorial

should have been undertaken with the holotype

specimens of the most closely related taxa under

museum conditions. In addition the type description

appeared in an obscure journal (although it was at

least published in the English language).

With a draft type description ﬁ nished, one must

consider to which journal it will be submi ed for

publication. It is important to pick a peer-reviewed

journal that is appropriate for the animal species being

described, and one that is familiar with publishing

type descriptions according to the International Code

of Zoological Nomenclature—the set of rules for naming

animals and the resolution of nomenclatural problems

(International Commission on Zoological Nomen-

clature, 2000). The International Commission on

Zoological Nomenclature acts as adviser and arbiter

for the zoological community by generating and

disseminating information on the correct use of the

scientiﬁ c names of animals.

The journal should ideally have an International

Standard Serial Number (ISSN) and be published in

the English language. However proud we may be of

our own languages, publication in any language other

than English will reduce the impact of the work within

the scientiﬁ c community at large. The publication of

type descriptions in books, where the description will

be more easily overlooked, should also be avoided.

Following these steps will help to ensure the widest

possible readership for your work (LeCroy & Vuille-

umier 1992).

The description of the Vietnamese pheasant Lophura

hatinhensis type specimen in a book in Vietnamese did

not help to clarify is existence. Only recently was its

invalidity as a taxon ﬁ nally established (Hennache et

al., 2012).

Where the holotype, syntypes and other type

material will be deposited is also a crucial considera-

tion. Type specimens should always be deposited in a

recognised museum collection that has good collection

management facilities and that also welcomes visiting

scientists. This is necessary to ensure the permanent

and safe storage of the priceless type material and to

ensure other scientists may have access to it for future

study (few museums will consider sending type

material on loan by post). To best serve the needs of

science, it may be appropriate to split the type series

so that some type material is stored in a collection

in the country of origin, thereby helping to promote

science locally, as well as in an internationally recog-

nised collection. In doing so, the risk of loss or damage

to the entire type series is spread and reduced. It is

also important that type specimens are labelled and

preferably stored separately from the main collection.

Bibliographic reference to the published description

and the proposed name should be wri en on the label

(LeCroy & Vuilleumier, 1992).

Probably everyone reading this will be familiar

with the saola Pseudoryx nghetinhensis, which was

described in a le er in the prestigious journal Nature

(Dung et al., 1993). The discovery of a new large

mammal genus and the use of DNA analysis were

certainly factors that the editors of Nature consid-

ered when deciding to publish this type description.

The holotype was, however, deposited in the collec-

tion of the Forest Inventory and Planning Institute

in Hanoi, Vietnam. This could hardly be described

as a recognised museum collection with good collec-

tion management facilities or allowing easy access

by visiting scientists. It would have been be er for

science had the holotype been deposited in a museum

with a collection of bovid type material.

Although the golden age of vertebrate species

discovery has past, new species are described

regularly. In 2010, at least 208 species of higher plants

and vertebrate animals were apparently described

from the Greater Mekong region, of which at least

seven were described from Cambodia (Thompson,

2011). Thus the opportunity to discover a new species

in Cambodia is a very real one and I hope that many

of you will have in the future.

The Cambodian Journal of Natural History does not

normally accept formal descriptions of new species,

new subspecies or other new taxa. If you wish to

submit original taxonomic descriptions, please contact

the editors in advance. The journal editors have two

reasons for not accepting type descriptions. First, is

the need for expert peer reviewers. While the editors

have a good range of expert taxonomist contacts for

some taxa (e.g. reptiles, bats, amphibians, birds and

orchids), they may not be able to secure high calibre

reviews for some of the lesser taxa that may be

submi ed. The last thing any of us would want to do

is publish a taxon that turned out to be false or inade-

quately described and substantiated. That would be a

disservice to science— all of us are familiar with the

problems caused by poor descriptions.

Second, there are already many excellent places to

publish new species. As an author, I would be looking

for a robust, well established journal that has a strong

track record in publishing taxa and can reach the

global audience that need access to the descriptions,

both now and in the future (type descriptions stay

Editorial

forever relevant). For example, a journal like Zootaxa,

which is available as hard and soft copies in almost

every academic library.

I know from speaking with colleagues that

a number of new plant and animal species from

Cambodia await description. Some of these may

currently be known only from photographs and the

lack of appropriate type material may be holding up

publication in some cases. I urge that such issues are

overcome with haste and all eﬀ ort be made to describe

new species for Cambodia. This will help promote

Cambodian science and provide a service for science

globally.

References

Alstrom, P., Davidson, P., Duckworth, J.W., Eames, J.C., Le

T.T., Nguyen C., Olsson, U., Robson, C. & Timmins, R.

(2010) Description of a new species of Phylloscopus warbler

from Vietnam and Laos. Ibis, 152, 145–168.

Athreya, R. (2006) A new species of Liocichla (Aves: Timali-

idae) from Eaglenest Wildlife Sanctuary, Arunachal

Pradesh, India. Indian Birds, 2, 82–94.

Collar, N.J. (1999) New species, high standards and the case

of Laniarius liberatus. Ibis, 141, 358–367.

Daltry, J.C., Fisher, M. & Furey, N.M. (2012) How to write a

winning paper. Cambodian Journal of Natural History, 2012,

97–100.

Davenport, T.R.B., Stanley, W.T., Sargis, E.J., de Luca, D.W.,

Mpunga, N.E., Machaga, S.J. & Olson, L.E. (2006) A new

genus of African monkey, Rungwecebus: morphology,

ecology, and molecular phylogenetics. Science, 312,

1378–1381.

Dung V.V., Giao, P.M., Chinh N.N., Tuoc D., Arctander, P. &

MacKinnon, J.R. (1993) A new species of living bovid from

Vietnam. Nature, 363, 443–445.

Hennache, A., Mahood, S.P., Eames, J.C. & Randi, E. (2012)

Lophura hatinhensis is an invalid taxon. Forktail, 28, 122–129.

International Commission on Zoological Nomenclature

(2000) International Code of Zoological Nomenclature, Fourth

Edition. The International Trust for Zoological Nomen-

clature, The Natural History Museum, London, United

Kingdom. H p://www.nhm.ac.uk/hosted-sites/iczn/code/

[accessed 28 June 2013].

Jones, T., Ehardt, C.L., Butynski, T.M., Davenport, T.R.,

Mpunga, N.E., Machaga, S.J. & de Luca, D.W. (2005) The

highland mangabey Lopocebus kipunji: a new species of

African monkey. Science, 308, 1161–1164.

LeCroy, M. & Vuilleumier, F. (1992) Guidelines for the

description of new species in ornithology. Bulletin of the

British Ornithological Society. Centenary Supplement, 112A,

191–198.

Mahood, S.P., John, A.J.I., Eames, J.C., Oliveros, C.H., Moyle,

R.G., Hong C., Poole, C.M. & Sheldon, F.H. (2013) A new

species of lowland tailorbird (Passeriformes: Cisticolidae:

Orthotomus) from the Mekong ﬂ oodplain of Cambodia.

Forktail, 29, 1–14.

McNeill, J., Barrie, F.R., Buck, W.R., Demoulin, V., Greuter,

W., Hawksworth, D.L., Herendeen, P.S., Knapp, S.,

Marhold, K., Prado, J., Prud’homme van Reine, W.F.,

Smith, G.F., Wiersema, J.H. & Turland, N.J. (2012) Inter-

national Code of Nomenclature for algae, fungi, and plants

(Melbourne Code), Adopted by the Eighteenth International

Botanical Congress Melbourne, Australia, July 2011. Koel

Scientiﬁ c Books, Koenigstein, Germany.

Peter, W.P. & Feiler, A. (1994) A new bovid species from

Vietnam and Cambodia (Mammalia: Ruminantia). Zoolo-

gische Abhandlungen Museum für Tierkunde Dresden, 48,

169–176.

Smith, E.F.G., Arctander, P., Fjeldsa, J. & Amir, O.G. (1991) A

new species of shrike (Laniidae: Laniarius) from Somalia,

veriﬁ ed by DNA sequence data from the only known

individual. Ibis, 133, 227–235.

Thompson, C. (2011) Wild Mekong: New Species in 2010 From

the Forests, Wetlands and Waters of the Greater Mekong, Asia’s

Land of Rivers. WWF Greater Mekong, Hanoi, Vietnam.

Http://assets.wwf.org.uk/downloads/greater_mekong_

species_report_web_ready_version_nov_14_2011_1.pdf

[accessed 28 June 2013].

Winston, J.E. (1999) Describing Species: Practical Taxonomic

Procedure for Biologists. Columbia University Press, New

York, USA.

News

Cambodian Journal of Natural History 2013 (1) 5–6

News

The Kannitha Fund for female

Cambodian conservation

scientists

The Kannitha Fund was established in 2012 to provide

scholarships to Cambodian women for the Masters

of Science in Biodiversity Conservation course at the

Royal University of Phnom Penh. The Kannitha Schol-

arships cover basic materials and living expenses. By

removing some of the ﬁ nancial barriers for Cambodian

women to become scientists, the fund organisers hope

this will be a catalyst to beneﬁ t endangered wildlife

and poor forest communities for years to come.

The charitable trust fund was created in tribute

to Lim Kannitha, a talented conservation student

who tragically died from malaria in 2010, aged only

30 years. Ms Lim had been among the ﬁ rst students

to enrol on the Masters in Biodiversity Conserva-

tion course at the Royal University of Phnom Penh.

She was an outstanding young scientist, as Dr Carl

Traeholt, one of her lecturers, recalled:

“Kannitha came from a family of very modest

means and her educational background was mixed.

But Kannitha deﬁ ed the odds because she had deter-

mination, an indomitable spirit and a hunger for

knowledge. She studied hard to reach a level of scien-

tiﬁ c excellence both for herself and for her country.

She possessed a rare combination of qualities required

to make a good scientist: curiosity, commitment,

creativity, intelligence and altruism. Kannitha also

had the tenacity to break from traditional conserva-

tism and face new challenges with a smile and without

prejudice”.

Joe and Jade Heﬀ ernan launched the Kannitha

Fund with contributions made by friends and family

on their wedding day in lieu of gifts. Joe Heﬀ ernan

(Fauna & Flora International) ﬁ rst met Kannitha in

2006, when she conducted research with the Cambo-

dian Elephant Conservation Group. She later went

on to conduct her Masters thesis research on gibbons

with Dr Ben Rawson at Conservation International.

Ms Hoem Thavry, from Prey Veng Province,

became the ﬁ rst recipient of a Kannitha Scholar-

ship in 2013. Thavry said “The scholarship pays all

my tuition fees and this makes a big diﬀ erence as it

was really hard for my family to ﬁ nd money for this

before. Because of the award, I now don’t need to ﬁ nd

a job to support my study costs which means I have

a lot more time to study and improve my grades. The

scholarship has made me a be er student as it really

encouraged me to work hard to improve my knowl-

edge and skills”.

After she graduates, Thavry hopes to “share the

knowledge I’ve learned, especially with Cambodian

people who aren’t aware of all the beneﬁ ts we get from

biodiversity. I’d really like to develop educational

materials through radio, television and magazines

to encourage people to love and protect our natural

heritage. To do this, I’d like to learn more by working

with a good conservation organization and also if

possible to work as a researcher”.

Women interested in applying for a scholarship for

the Masters of Science in Biodiversity Conservation

should contact the Centre for Biodiversity Conser-

vation, Room 415, Main Campus, Faculty of Science,

Royal University of Phnom Penh, Confederation of

Russian Boulevard, Phnom Penh, Cambodia; Email:

mbiodiversity.info@rupp.edu.kh

To learn more about the Kannitha Fund and to

make donations to this important cause, please visit

the fund website:

H p://kannithafund.wordpress.com/

Announcing the publication

The Birds of Cambodia -

An Annotated Checklist

Frédéric Goes

Even though the international importance of the

Cambodian bird fauna is widely recognised, the only

review to date was conﬁ ned to pre-1970s records for

399 bird species. As information on Cambodian birds

has grown exponentially since the 1990s, the country

has needed a national treatment synthesizing latest

knowledge on its fascinating avifauna for some time.

The Birds of Cambodia - An Annotated Checklist is

a landmark publication that addresses this need. It

stems from 12 years of ornithological surveys and

observations, coupled with six years of collation and

review of all available records. The book exhaustively

documents the entire bird fauna of Cambodia and

News

Cambodian Journal of Natural History 2013 (1) 5–6

identiﬁ es all bird species of national conservation

concern. As such, it provides an authoritative basis

for a national red data book and future conservation

legislation.

The book consists of three parts and includes 48

colour plates illustrating major habitats, conservation

threats and over 80 bird species photographed in the

wild in Cambodia.

The introduction describes the country’s natural

geography, major habitats, protected areas, ornitho-

logical history and survey coverage, then goes on to

review conservation successes and challenges and

provide guidance for novice birdwatchers.

A systematic section forms the heart of the book

and presents peer-reviewed accounts for the 598

bird species currently conﬁ rmed for Cambodia. The

accounts synthesize latest knowledge on seasonal

occurrence, abundance, distribution and habitat,

including notes on breeding and conservation.

English, Latin, French and Khmer names, including

transliteration, are provided for each species. Detailed

reviews of records are also provided for rarities and

all species of conservation concern, together with a

proposed national conservation category.

Several appendices complete the book. These

include a reference checklist for Cambodian birds,

tables of nationally threatened species and potential

future species additions, census results for globally

threatened species and a geographical gaze eer.

In addition to stimulating interest and awareness

among the general public, The Birds of Cambodia -

An Annotated Checklist will undoubtedly become

an indispensable reference for conservationists

and ornithologists in Cambodia, as well as all bird

watchers visiting the Kingdom.

All proceeds from sales of the publication will be

used to build awareness and capacity for bird conser-

vation among young Cambodians.

Copies can be obtained from Fauna & Flora

International Cambodia Programme, #19, Street 360,

Boeng Keng Kong 1, Phnom Penh, Cambodia. Email:

birdsofcambodia@gmail.com

Giant ibis Pseudibis gigantea painting by Berry Mulligan, from the cover illustration of Frédéric Goes’ The Birds of

Cambodia - An Annotated Checklist.

Butterﬂ y farming in Siem Reap

Cambodian Journal of Natural History 2013 (1) 7–9

Short Communication

The Banteay Srey Butterﬂ y Centre: ﬁ ve years of endeavouring to

support conservation and poverty alleviation

Torsten VAN DER HEYDEN

Immenweide 83, D-22523 Hamburg, Germany.

Email tmvdh@web.de

Paper submitted 22 March 2013, revised manuscript accepted 24 May 2013.

Sustainable bu erﬂ y farms have been established in

several tropical developing countries with the aims

of supporting local rural livelihoods and conserving

forests with high biodiversity. For example, Kenya

(Gordon et al., 2011), Tanzania (Morgan-Brown et al.,

2010; van der Heyden, 2011) and Guyana (Sambhu &

van der Heyden, 2010).

In 2008, Ben Hayes, originally from the United

Kingdom, started the Banteay Srey Bu erﬂ y Centre

(BBC) near the Phnom Kulen National Park in Siem

Reap Province. It was based on the Zanzibar Bu erﬂ y

Centre, a similar project he started in Tanzania in 2006

(van der Heyden, 2011). Both projects operate within

or near protected areas where there is a lot of pressure

on natural resources from local communities. Ben

Hayes and two Cambodians, Nhoek Sakhaun and

Thoung Chantha, are the directors of the BBC. The

centre is managed by another Cambodian, Om Srey

Vat.

The BBC is in Sanday Village, near the Banteay

Srey Temple and the Cambodia Landmine Museum.

It oﬀ ers a live bu erﬂ y exhibition to residents and

tourists, which is the largest of its kind in Southeast

Asia. Hundreds of free-ﬂ ying bu erﬂ ies—all of them

native species of Cambodia—can be observed in a

ne ed tropical enclosure, approximately 30 m x 40 m.

The centre is visited by approximately 10,000 foreign

and 3,000 Cambodian visitors every year. They are

informed about the diﬀ erent species on display, the

bu erﬂ y life cycle and their ecology by trained local

staﬀ members. Through these talks, the BBC aims to

give visitors an increased knowledge of local Cambo-

CITATION: van der Heyden, T. (2013) The Banteay Srey Bu erﬂ y Centre: ﬁ ve years of endeavouring to support conservation

and poverty alleviation. Cambodian Journal of Natural History, 2013, 7–9.

dian biodiversity and hence stimulate interest in

conservation and protection issues.

The BBC also focuses on enabling local rural

communities to gain a livelihood by rearing bu erﬂ y

species and selling pupae to the BBC. Farmers in

Sanday Village and remote communities currently

farm 35 species of bu erﬂ ies and moths from various

families: Atrophaneura aristolochiae, A acus atlas (Fig.

1), Catopsilia pomona, C. scylla, Cethosia cyane, Charaxes

solon, Danaus genutia, Delias pasithoe, Dysphania

sagana, Elymnias hypermnestra, E. nesaea, Euploea core,

E. mulciber, Euthalia aconthea, E. lubentina, Graphium

agamemnon (Fig. 2), G. antiphates, G. doson, G. sarpedon,

Hebomoia glaucippe, Hypolimnas bolina, Junonia almana,

Lebadea martha, Lexias dirtea, Melanitis leda, Papilio

clytia, P. demoleus (Fig. 3), P. demolion, P. helenus, P.

memnon, P. polytes, Parantica aglea, Parthenos sylvia,

Polyura athamas and Tirumala septentrionis. All of these

species are farmed every year, but the number of

specimens reared and displayed may vary depending

on the season.

In small ne ed enclosures in the farmers’

backyards (Fig. 4), female bu erﬂ ies deposit their

eggs on the speciﬁ c food plants of the respective

species. The eggs are harvested by the farmers and the

hatched larvae are transferred to their food plants in

a “nursery”. After pupation, the pupae are sold to the

BBC and displayed in the centre (Fig. 5), where the

bu erﬂ ies emerge. The duration of the breeding cycle

varies depending on the species. Most species take

several weeks to complete the cycle from egg to adult

bu erﬂ y.

T. van der Heyden

Cambodian Journal of Natural History 2013 (1) 7–9

Fig. 1 A acus atlas (Saturniidae) (© T. van der Heyden). Fig. 2 Graphium agamemnon (Papilionidae) (© T. van der

Heyden).

Fig. 3 Papilio demoleus (Papilionidae) (© T. van der

Heyden).

Fig. 4 Breeding cage owned by a farmer with the Banteay

Srey Bu erﬂ y Centre (© B. Hayes).

Fig. 5 Pupae of Papilio memnon (Papilionidae) at the

Banteay Srey Bu erﬂ y Centre (© B. Hayes).

Because only a few bu erﬂ ies are caught from

the wild to start the farming process, these collec-

tions are not thought to harm the wild populations. A

number of the pupae reared by the farmers are used

for breeding purposes, thus avoiding unnecessary

consecutive collections from the wild. To prevent rare

or threatened species being caught, the BBC does not

buy any rare species and all farmers have been taught

to farm only common, non-threatened species.

The BBC buys pupae only from local farmers that

are members of the project. By rearing and selling

bu erﬂ ies from home, the farmers are able to increase

and diversify their income, which helps to alleviate

poverty. The additional monthly income is very

variable, depending on how many pupae the farmers

produce and of which species, but some farmers have

earned US$ 100 per month from this part time work.

This is twice the local average monthly income.

In addition, the farmers are able to gain the knowl-

edge that an intact natural environment is vital for

their business, which could motivate them to conserve

their natural surroundings instead of destroying them

for agriculture or other purposes. Forest clearance and

other forms of habitat destruction could potentially be

Butterﬂ y farming in Siem Reap

Cambodian Journal of Natural History 2013 (1) 7–9

reduced, having a positive impact on wild species,

both plants and animals.

As of March 2013, about 30 male and female

farmers work with the BBC. Their business of rearing

and selling bu erﬂ ies is an example of the sustain-

able use of natural resources. Revenue generated

from admission fees to the BBC is used to support

the farmers and their families with a supplementary

income as well. Additional people from local commu-

nities are employed by the BBC to manage the centre,

train and support the bu erﬂ y farmers, and guide

visitors. Part of the revenue from tourist admissions

is also used to support conservation projects, for

example, biodiversity surveys. The BBC is a member

of ConCERT (Connecting Communities, Environ-

ment & Responsible Tourism) based in Siem Reap, a

network of local partners involved in conservation.

Currently, there are no quantitative or qualitative

data to evaluate the impact of the BBC on the conser-

vation of natural resources. I therefore recommend

conducting a survey to investigate these aspects and

to determine if and how a itudes and behaviours

towards natural resources have changed within the

local communities. Generally, such a survey or evalu-

ation should be done for all bu erﬂ y farming projects

of this kind throughout the world to understand their

environmental impacts. Morgan-Brown et al. (2010)

examined a commercial bu erﬂ y farming project in

Tanzania and found bu erﬂ y farmers were signiﬁ -

cantly more active in forest conservation than other

community members because they “perceive a link

between earnings from bu erﬂ y farming and forest

conservation”. It is possible that similarly positive

results will be found in other sustainable bu erﬂ y

farming projects, including the BBC.

A future aim of the BBC is to farm more species

of Saturniidae, in addition to A acus atlas (Fig. 1).

Research is currently being carried out to increase

the number of species farmed and the BBC is also

planning to increase the number of farmers involved.

To enlarge this business, it will be necessary to export

pupae, for example, to bu erﬂ y exhibitions in Europe

or North America. The BBC is awaiting an export

license to do this. Finally, another challenge facing

the centre is the production of pupae all year round.

Because many of the bu erﬂ y farmers are engaged

in rice cultivation, li le or even no bu erﬂ y farming

takes place during the harvest period.

I agree with Gordon et al. (2011) and Morgan-

Brown et al. (2010) that initiatives like the BBC are

appropriate to support rural communities in tropical

developing countries to improve their living condi-

tions without harming nature.

Acknowledgements

I would like to thank Ben Hayes from the Banteay

Srey Bu erﬂ y Centre who provided me with informa-

tion about the project and two of the photographs.

References

Gordon, I., Fungomeli, M. & Githitho, A. (2011) Bu erﬂ y

farming as an NTFP: 19 years of Kipepeo: 1993-2012. Paper

presented to the Kenya Trees and Forests Conference, 27-29

February 2011, Brackenhurst, Kenya. H p://www.outremer.

total.com/os/content/NT00061F1A.pdf [accessed 24 May

2013].

Morgan-Brown, T., Jacobson, S.K., Wald, K. & Child, B.

(2010) Quantitative assessment of a Tanzanian integrated

conservation and development project involving bu erﬂ y

farming. Conservation Biology, 24, 563–572.

Sambhu, H. & van der Heyden, T. (2010) Sustainable

bu erﬂ y farming in tropical developing countries as an

opportunity for man and nature—the “Kawê Amazonica

Bu erﬂ y Farm” project in Guyana as an example (Insecta:

Lepidoptera). Sociedad Hispano-Luso-Americana de Lepidop-

terología (SHILAP) Revista de lepidopterología, 38, 451–456.

van der Heyden, T. (2011) Local and eﬀ ective: two projects

of bu erﬂ y farming in Cambodia and Tanzania (Insecta:

Lepidoptera). Sociedad Hispano-Luso-Americana de Lepidop-

terología (SHILAP) Revista de lepidopterología, 39, 267–270.

About the Author

TORSTEN VAN DER HEYDEN—apart from being

a teacher of Biology and Geography—is a German

independent researcher specialising in Lepidoptera

and Heteroptera. He has published papers on bu er-

ﬂ ies, moths and true bugs, focussing on their biology,

ecology and distribution, as well as papers on several

bu erﬂ y centres and farms and their impact on

conservation and protection. The author is a member

of various scientiﬁ c associations and societies, e.g. a

fellow of the Linnean Society of London and the Royal

Entomological Society, and a member of the Spanish

Real Sociedad Española de Historia Natural. He is a

member of the editorial boards of Atalanta—Zeitschrift

der Deutschen Forschungszentrale für Schme erling-

swanderungen, BV news Publicaciones Cientíﬁ cas and

Lepcey—The Journal of Tropical Asian Entomology.

T. Hartmann et al.

Cambodian Journal of Natural History 2013 (1) 10–15

Short Communication

Records of freshwater ﬁ sh species from Phnom Kulen National

Park, northwestern Cambodia

Timo HARTMANN1,*, Sebastian HÜLLEN1, Peter GEISSLER1, Markus HANDSCHUH2, SENG

Rattanak3, Friedrich Wilhelm MIESEN1 and Fabian HERDER1

1 Zoologisches Forschungsmuseum Alexander Koenig, Adenauerallee 160, D-53113 Bonn, Germany.

2 Angkor Centre for Conservation of Biodiversity (ACCB), Kbal Spean, Phnom Kulen National Park, P.O. Box 93 054,

Siem Reap, Cambodia.

3 General Department of Administration for Nature Conservation and Protection (GDANCP), Ministry of Environment,

48 Samdech Preah Sihanouk, Tonle Bassac, Chamkarmorn, Phnom Penh, Cambodia.

*Corresponding author. Email t.hartmann.zfmk@uni-bonn.de

Paper submitted 14 April 2013, revised manuscript accepted 15 June 2013.

Cambodia’s ﬁ sh fauna remains only partially

explored, with most of the ichthyological literature

dating back to the early 20th century when Cambodia

was under French jurisdiction (e.g. Leclère, 1901;

Durand, 1915; Chabanaud, 1926; Chevey, 1932). In

the 1960s, François d’Aubenton undertook the ﬁ rst

extensive scientiﬁ c collection of ﬁ shes from all over

Cambodia, which were later identiﬁ ed and revised by

Ko elat (1985). The following years of civil unrest in

Cambodia hampered further research. Besides a ﬁ eld

guide on the Fishes of the Cambodian Mekong (Rainboth,

1996) and an atlas on the Fishes of the Greater Mekong

Ecosystem (Rainboth et al., 2012), recent work has

focused mainly on the Tonle Sap lake and river system

(Lim et al., 1999; Motomura et al., 2002).

Phnom Kulen National Park is a 37,375-hectare

protected area in Banteay Srei, Svay Leu and Varin

Districts, Siem Reap Province, northwestern Cambodia

(Fig. 1). Hills dominate the park’s topography, with

most areas above 200 m elevation, but with few peaks

above 450 m. Most of the park is covered by disturbed

semi-evergreen forest that suﬀ ers from ongoing forest

clearance and selective logging. The national park

contains numerous small, often seasonal streams,

which are not mapped.

CITATION: Hartmann,T., Hüllen, S., Geissler, P., Handschuh, M., Seng R., Miesen, F.W. & Herder, F. (2013) Records of fresh-

water ﬁ sh species from Phnom Kulen National Park, northwestern Cambodia. Cambodian Journal of Natural History, 2013,

10–15.

Here we present recent records of freshwater

ﬁ shes from three sites in the Phnom Kulen National

Park: Site A—Phnom Kbal Spean, Bantey Srei District

(13°40’45.5”N, 104°01’25.0”E); Site B—Phnom

Kbal Spean, Bantey Srei District (13°41’13.0”N,

104°00’56.0”E); Site C—Phnom Kulen, Svay Leu

District (13°31’40.3”N, 104°07’10.8”E). These records

were a by-product of a wider herpetological survey.

Sites A and B are upland tributaries of the Kbal Spean

River in the Northwest of the national park. During

the early rainy season, in May and June 2011, Site A

was a shallow stream meandering between rocks,

with a moderate to slow current and a sandy bo om

(Fig. 2). Site B also had a sandy bo om and was up to

one metre deep, with a moderate current (Fig. 3). Site

C is a small tributary of the Phnom Kulen River, up to

one metre deep, with a very slow current and rocky

to sandy bo om (Fig. 4). At each site we captured ﬁ sh

opportunistically in all accessible micro-habitats.

Fish were caught using a dip net (50 cm x 70 cm;

mesh size 3.2 mm). After capture, life pictures were

taken, and then the ﬁ shes were anaesthetized with

clove oil, following the methods of Oetinger (2003), and

subsequently ﬁ xed in 95% ethanol. Specimens were

deposited at the Zoologisches Forschungsmuseum

Freshwater ﬁ sh in Northwest Cambodia

Cambodian Journal of Natural History 2013 (1) 10–15

Fig. 1 Left: The Kingdom of Cambodia, showing the Phnom Kulen National Park, Siem Reap Province. Right: Phnom

Kulen National Park, showing sampling localities and main water bodies. Maps designed using ArcGIS 9.3.

Fig. 2 Site A—Phnom Kbal Spean, Bantey Srei District, at

100 m a.s.l. (© T. Hartmann).

Fig. 3 Site B—Phnom Kbal Spean, Bantey Srei District, at

198 m a.s.l. (© P. Geissler).

Fig. 4 Site C—Phnom Kulen, Svay Leu District, at 315 m

a.s.l. (© P. Geissler).

T. Hartmann et al.

Cambodian Journal of Natural History 2013 (1) 10–15

Fig. 5 Variation in Puntius aurotaeniatus (© P. Geissler).

Fig. 6 Danio albolineatus (© P. Geissler).

Fig. 7 Rasbora paviana (© P. Geissler).

Fig. 8 Lepidocephalichthys hasselti (© P. Geissler). Fig. 9 Nemacheilus pallidus (© P. Geissler).

Fig. 10 Dermogenys siamensis (© P. Geissler). Fig. 11 Be a prima (© P. Geissler).

Freshwater ﬁ sh in Northwest Cambodia

Cambodian Journal of Natural History 2013 (1) 10–15

Ko elat’s (2000) observations on Danio albolineatus in

having a lateral line and 13 to 14 soft anal ﬁ n rays.

Cyprinidae: Rasborinae

Rasbora paviana Tirant, 1885

Site A: ZFMK 44906–44916; Site B: ZFMK 44809–44826;

44836–44844 (Fig. 7).

Mekong Basin (Rainboth, 1996, Ko elat, 1998),

including Tonle Sap River and Tonle Sap Lake (Lim et

al., 1999, Motomura et al., 2002).

Cobitidae

Acanthopsoides hapalias Siebert, 1991

Site A: IAPG A5451, ZFMK 45232.

Lower Mekong (Rainboth, 1996; Doi, 1997; Ko elat,

1998; Ko elat, 2001).

Lepidocephalichthys hasselti (Valenciennes, 1846)

Site A: IAPG A5449, ZFMK 45233 (Fig. 8).

Mekong Basin (Ko elat et al., 1993; Rainboth, 1996;

Ko elat, 2001).

Balitoridae

Nemacheilus pallidus Ko elat, 1990

Site A: IAPG A5450, ZFMK 45234–45235 (Fig. 9).

Mekong Basin (Ko elat, 1990; Rainboth, 1996). Tonle

Sap Lake (Motomura et al., 2002).

Clariidae

Clarias spec. aﬀ . batrachus ‘Indochina’ Linnaeus,

1758

Clarias aﬀ . batrachus ‘Indochina’ (Ng & Ko elat,

2008)

Site B: ZFMK 44794.

Ng & Ko elat (2008) designated a neotype for Clarias

batrachus (Linnaeus, 1758), a species previously

thought to be widely distributed throughout South-

east Asia. It is now understood Clarias batrachus is

restricted to Java and the Indochinese form represents

a distinct species, Clarias aﬀ . batrachus ‘Indochina’,

but the whole complex requires further taxonomic

research. ZFMK 44794 has 77 soft dorsal ﬁ n rays and

53 soft anal ﬁ n rays; the ﬁ rst pectoral spine is thick-

ened and the inner edge is serrated. The dorsolateral

colouration of the preserved specimen is brown,

Alexander Koenig (ZFMK), Bonn, Germany, and the

Institute of Animal Physiology and Genetics (IAPG),

Libě chov, Czech Republic. Each collection number

refers to a single specimen.

Our collection contains 13 species, as listed below.

We also give the known distribution in Cambodia for

each of these species, assigning old locality names

from literature to currently used names where

possible, and provide further clarifying remarks on

species where necessary.

Compared to the immense species richness found

during ichthyological surveys elsewhere in Indochina

(e.g. Freyhof et al., 2000; Motomura et al., 2006), our

opportunistic and very short-term approach certainly

cannot determine the full diversity of the freshwater

ﬁ shes in Phnom Kulen National Park. An in-depth

survey of the park’s ichthyofauna would certainly

lead to the ﬁ nding of numerous additional species.

By presenting this list of freshwater ﬁ sh species, we

hope to encourage future Cambodian ichthyolo-

gists to undertake further work on this understudied

taxonomic group in the Phnom Kulen National Park

and elsewhere in Cambodia.

Cyprinidae: Barbinae

Puntius rhombeus Ko elat, 2000

Site A: ZFMK 44887–44888; Site B: ZFMK 44795–44807;

44869.

Mekong Basin (Ko elat, 2000).

Puntius aurotaeniatus (Tirant, 1885)

Site A: ZFMK 44892–44894.

Mekong Basin and in coastal drainages of the Gulf of

Thailand (Rainboth, 1996). Specimens vary remark-

ably in the pa ern and distinctiveness of their vertical

bars (Fig. 5 A, B).

Cyprinidae: Danioninae

Danio albolineatus (Blyth, 1860)

Site A: ZFMK 44899–44901; Site B: ZFMK 44845–44862;

Site C: ZFMK 44866–44868 (Fig. 6).

Mekong Basin (Rainboth, 1996; Ko elat, 2001); O Po

Kampon, Boum Long (= probably Banlung), Snoc

Trou, Toek Sap, Kirikum (= Kirirom), Sihanouk-

ville, Bokéo (= Bokor) and Sré Umbel (= Sre Ambel)

(Ko elat, 1985).

Remarks: Our specimens are in accordance to Fang &

T. Hartmann et al.

Cambodian Journal of Natural History 2013 (1) 10–15

without white dots. The ventral side is whitish to

brown in colour; ﬁ ns and head are darker.

Zenarchopteridae

Dermogenys siamensis Fowler, 1934

Site A: ZFMK 44918–44922; Site C: ZFMK 44886 (Fig.

10).

Mekong Basin (Ko elat, 2001; Meisner, 2001).

Channidae

Channa gachua (Hamilton, 1822)

Site A: ZFMK 44917; Site B: ZFMK 44827–44835;

44863–44865.

Mekong Basin (Ko elat, 1998; Ko elat, 2001).

Channa striata (Bloch, 1793)

Site A: ZFMK 44897–44905; Site C: ZFMK 44870–44873.

Mekong Basin (Ko elat, 1998), including Tonle Sap

River and Lake (Nao & Sina, 1998; Lim et al., 1999;

Motomura et al., 2002), Ratanakiri, Boum Long (=

probably Banlung), Kampong Chnang, Ream, Boeng

Kbal Damrey, Sihanoukville and Angkor (= Siem

Reap) (Ko elat, 1985).

Osphronemidae

Be a prima Ko elat, 1994

Site C: ZFMK 44874–44885 (Fig. 11).

Mekong Basin (Ko elat, 1994, 2001) and coastal drain-

ages in Cambodia and Vietnam (Allen, 2012).

Trichopsis vi ata (Cuvier, 1831)

Site A: ZFMK 44895–44896 (both juveniles) (Fig. 12).

Mekong Basin (Rainboth, 1996); Kirirom, Toek Sap,

Boeng Kbal Damrey, Stung Sen and O Po Kampon

(Ko elat, 1985).

Acknowledgements

We are grateful to His Excellency Chay Samith, General

Director of the General Department of Administration

for Nature Conservation and Protection (GDANCP),

Ministry of Environment of the Royal Government

of Cambodia, for kindly issuing the relevant permits.

We thank Sokkheng Novin, Director of National Park

and Wildlife Sanctuary Department of GDANCP, and

Hong Daravuth, Director of Phnom Kulen National

Park and Director of Angkor Cluster Protected Areas

of GDANCP, for their generous support. Special

thanks to Neang Thy (GDANCP) for facilitating the

permit process. We are grateful to Jörg Bohlen and

Tan Heok Hui for providing assistance with species

identiﬁ cation. We thank Dr Hiroyuki Motomura

(Kagoshima University Museum, Japan) for providing

literature, and Lara Mielke and Alistair Mould for

assistance during the ﬁ eld surveys. Field work was

supported by the Alexander Koenig Stiftung. The

manuscript beneﬁ ted from comments and sugges-

tions by Dr. Maurice Ko elat, Dr. Walter J. Rainboth

and an anonymous referee.

References

Allen, D. (2012) Be a prima. In IUCN Red List of Threat-

ened Species. Version 2012.1. H p://www.iucnredlist.org/

details/181251/0 [accessed 9 October 2012].

Chabanaud, P. (1926) Inventaire de la faune ichthyologie

de l’Indochine: première liste. Bulletin économique de

l’Indochine, 169, 561–581.

Chevey, P. (1932) Inventaire de la fauna ichthyologique

de l’Indochine. Notes de l’Institut Océanographique de

l’Indochine, 19, 1–31.

Doi, A. (1997) A review of taxonomic studies of cypriniform

ﬁ shes in Southeast Asia. Japanese Journal of Ichthyology, 44,

1–33.

Durand, P. (1915) Industrie de la Pêche Dans la Région des Lacs

au Cambodge. Saigon (Ho Chi Minh City), Vietnam.

Fang, F. & Ko elat, M. (2000) Danio roseus, a new species

from the Mekong basin in northeastern Thailand and

northwestern Laos (Teleostei: Cyprinidae). Ichthyological

Exploration of Freshwaters, 11, 149–154.

Fig. 12 Juvenile of Trichopsis vi ata (© P. Geissler).

Freshwater ﬁ sh in Northwest Cambodia

Cambodian Journal of Natural History 2013 (1) 10–15

Freyhof, J., Serov, D.V. & Nguyen T.N. (2000) A preliminary

checklist of the freshwater ﬁ shes of the River Dong Nai,

South Vietnam. Bonner zoologische Beiträge, 49, 93–99.

Ko elat, M. (1985) Fresh-water ﬁ shes of Kampuchea. Hydro-

biologia, 121, 249–279.

Ko elat, M. (1990) Indochinese Nemacheilines: A Revision of

Nemacheiline Loaches (Pisces: Cypriniformes) of Thailand,

Burma, Laos, Cambodia and Southern Viet Nam. Pfeil-Verlag,

Munich, Germany.

Ko elat, M. (1994) Diagnoses of two new species of ﬁ ghting

ﬁ shes from Thailand and Cambodia (Teleostei: Belon-

tiidae). Ichthyological Exploration of Freshwaters, 5, 297–304.

Ko elat, M. (1998) Fishes of the Nam Theun and Xe Bangfai

basins, Laos, with diagnoses of twenty-two new species

(Teleostei: Cyprinidae, Balitoridae, Cobitidae, Coiidae and

Odontobutidae). Ichthyological Exploration of Freshwaters, 9,

1–128.

Ko elat M. (2000) Diagnoses of a new genus and 64 new

species of ﬁ shes from Laos (Teleostei: Cyprinidae,

Balitoridae, Bagridae, Syngnathidae, Chaudhuriidae and

Tetraodontidae). Journal of South Asian Natural History, 5,

37–82.

Ko elat, M. (2001) Fishes of Laos. WHT Publications Ltd.,

Colombo, Sri Lanka.

Ko elat, M., Whi en, A.J., Kartikasari, S.N. & Wirjoatmodjo,

S. (1993) Freshwater Fishes of Western Indonesia and Sulawesi /

Ikan Air Tawar Indonesia Bagian Barat dan Sulawesi. Periplus

Editions, Hong Kong, China.

Leclère, A. (1901) La pêche dans le Grand Lac au Cambodge.

Bulletin Économique de l’Indochine, 38, 675–679.

Lim, P., Lek S., Touch S.T., Mao S.-O. & Chhouk B. (1999)

Diversity and spatial distribution of freshwater ﬁ sh in

Great Lake and Tonle Sap River (Cambodia, Southeast

Asia). Aquatic Living Resources, 12, 379–386.

Meisner, A.D. (2001) Phylogenetic systematics of the vivip-

arous hal eak genera Dermogenys and Nomorhamphus

(Teleostei: Hemiramphidae: Zenarchopterinae). Zoological

Journal of the Linnean Society, 133, 199–283.

Motomura, H., Tsukawaki, S. & Kamiya, T. (2002) A prelimi-

nary survey of the ﬁ shes of Lake Tonle Sap near Siem

Reap, Cambodia. Bulletin of the National Museum of Nature

and Science, 28, 233–246.

Nao T. & Sina, L. (1998) Review of the ﬁ sheries and aquacul-

ture sector in Cambodia. In Natural Resources-based

Development Strategy for the Tonle Sap Area, Cambodia.

Final Report, Volume 2. (ed. Cambodian National Mekong

Commi ee/ Nedeco), pp. 35–36. Mekong River Commis-

sion and United Nations Development Programme,

Phnom Penh, Cambodia.

Ng H.H. & Ko elat, M. (2008) The identity of Clarias batra-

chus (Linnaeus, 1758), with the designation of a neotype

(Teleostei: Clariidae). Zoological Journal of the Linnean

Society, 153, 725–732.

Oetinger, F.C. (2003) Betäubung von Regenbogenforellen

(Oncorhynchus mykiss) mit Nelkenöl und BHA—Stressbe-

lastung und Produktqualität. LMU München—Tierärztliche

Fakultät, Munich, Germany.

Rainboth, W.J. (1996) Fishes of the Cambodian Mekong: FAO

Species Identiﬁ cation Field Guide for Fishery Purposes. Food

and Agriculture Organisation, Rome, Italy.

Rainboth, W.J., Vidthayanon, C. & Mai D.Y. (2012) Fishes of

the Greater Mekong Ecosystem with Species List and Photo-

graphic Atlas. Miscellaneous Publications, Museum of

Zoology, University of Michigan. Ann Arbor, USA.

Phauk S. et al.

Cambodian Journal of Natural History 2013 (1) 16–26

Cambodian bat echolocation: a ﬁ rst description of assemblage

call parameters and assessment of their utility for species

identiﬁ cation

PHAUK Sophany1,*, PHEN Sarith1 and Neil M. FUREY1,2

1 Centre for Conservation Biodiversity, Room 415, Department of Biology, Faculty of Science, Royal University of

Phnom Penh, Confederation of Russia Boulevard, Phnom Penh, Cambodia.

2 Fauna & Flora International, Cambodia Programme, 19, Street 360, BKK 1, Chamkarmorn, Phnom Penh, Cambodia.

*Corresponding author. Email phauk.sophany@rupp.edu.kh

Paper submitted 3 May 2013, revised manuscript accepted 15 June 2013.

Abstract

Bats form a major component of mammal diversity in Southeast Asia and are potential indicators of wider

biodiversity impacts resulting from habitat loss and climate change. The development of eﬀ ective methods for

inventorying and monitoring Southeast Asian bats is critical if their conservation needs are to be determined and

their potential as bioindicators realised. To this end, we provide the ﬁ rst description of time-expanded echoloca-

tion calls from a Cambodian bat assemblage comprising 17 species from Phnom Kulen National Park. We further

evaluate the reliability of acoustic methods for identifying 13 of these taxa. Discriminant function analysis of

428 echolocation calls produced by the 13 bat species indicated that acoustic identiﬁ cation was feasible in most

instances by correctly classifying 85% of calls. The best models relied on two call parameters and were statisti-

cally signiﬁ cant. Further studies documenting geographical and other sources of variation in the echolocation

calls produced by Cambodia’s bat fauna are necessary to facilitate development of acoustic sampling as a tool for

their conservation.

CITATION: Phauk S., Phen S. & Furey, N.M. (2013) Cambodian bat echolocation: a ﬁ rst description of assemblage call param-

eters and assessment of their utility for species identiﬁ cation. Cambodian Journal of Natural History, 2013, 16–26.

Bat echolocation calls

indispensable for maximising sampling completeness

in ﬁ eld surveys (Furey et al., 2009; Hughes et al., 2010,

2011), an essential requirement for eﬀ ective conserva-

tion planning.

Despite this, acoustic sampling has been rarely

employed in continental Southeast Asia to date and

detailed descriptions are lacking for the echolocation

calls of most bat species in the region. We address this

by providing the ﬁ rst description of time-expanded

echolocation calls for an assemblage of Cambodian

bats and evaluate the reliability of acoustic methods

for species identiﬁ cation. The study was undertaken

at Phnom Kulen National Park as part of a series

of ongoing bat surveys that primarily rely on harp

traps and mist nets in this area. As Cambodia’s bats

are poorly known relative to neighbouring countries

(Kingsada et al., 2011; Ith et al., 2011a), our overall

purpose was to provide information to assist future

bat research and conservation eﬀ orts in the country.

Methods

Study site

Phnom Kulen National Park is in Siem Reap Province,

Northwest Cambodia (Fig. 1). The region has a tropical

monsoon climate with a mean annual rainfall of 2,050

mm and an average annual temperature of 24°C

(Hou et al., 2004). The national park covers an area

of 37,350 hectares and encompasses lowland areas

and sandstone hills that culminate in two plateaus

reaching 450 m above sea level (a.s.l.). Habitats include

evergreen and semi-evergreen forests on hillsides

and plateaus, while lowland areas were originally

dominated by dry dipterocarp forest, of which only

small, degraded areas now remain (Neou et al., 2008).

Capture methods and species identiﬁ cation

Thirty-two nights of sampling were undertaken in

semi-evergreen forest of variable condition within

the Kbal Spean area (13°36’22’’N, 104°00’96’’E) of

the national park between April and July 2010. Live-

trapping was carried out from 18:00–21:00 h each

night using a four-bank harp trap (capture surface:

2.4 m2) and 70 denier mist nets (capture surface: 30

m2), giving a total sampling eﬀ ort of 234 m2 harp-trap-

Introduction

Bats form a major component of the Southeast Asian

mammal fauna, constituting approximately 30% of

the region’s mammal species and as many as half of all

mammal species in the tropical rainforest ecoregions

(Kingston, 2010). This group provides economically

signiﬁ cant ecosystem services in plant pollination,

seed dispersal and arthropod suppression (Kunz et al.,

2011). Bats also possess a variety of traits that support

their use as bioindicators, reﬂ ecting wider biodiversity

impacts from habitat loss and climate change (Jones

et al., 2009). Like much of the Southeast Asian fauna,

however, bats are severely threatened, with only

18% of species populations in the region presently

considered stable by the International Union for the

Conservation of Nature (IUCN) (Kingston, in press).

The global success, species richness and ability

of bats to exploit diverse niches are largely due to

their capacity for powered ﬂ ight and echolocation

(Jones & Teeling, 2006). Echolocation entails the use of

reﬂ ected sound waves whereby bats use the diﬀ erence

between the sounds they produce and the returning

echoes they hear to collect information about their

surrounding environment. This acoustic process is

largely ultrasonic and allows bats to navigate complex

three-dimensional spaces in complete darkness.

Echolocation tasks exert a strong selective pressure on

signal design, favouring species-speciﬁ c signal types

linked to ecological conditions (Schni ler et al., 2003).

As a consequence, once adequate reference recordings

have been obtained from bats of known identity, these

can be used to identify species exclusively by their

calls (Brigham et al., 2004).

The development of eﬀ ective methods for inven-

torying and monitoring bat populations is essential

if their conservation needs are to be determined and

their potential as bioindicators realised. Because

traditional sampling methods for bats—mist nets and

harp traps—are rarely employed more than a few

metres above ground level in surveys in Asia, they

typically fail to capture species that habitually ﬂ y in

open areas and/or above the forest canopy, even in

the most intensive studies. Detecting bats from their

calls is widely viewed as a means of overcoming

these limitations (Brigham et al., 2004). Recent studies

indicate that acoustic identiﬁ cation of Southeast Asian

bat species is feasible and that acoustic methods are

Keywords

Acoustic sampling, bat species identiﬁ cation, Phnom Kulen National Park.

Phauk S. et al.

Cambodian Journal of Natural History 2013 (1) 16–26

were obtained from motionless bats held in the hand,

whereas recordings for all other species were obtained

in ﬂ ight either from hand-released bats, a ﬂ ight cage

(measuring 10 x 2 x 1.5 m) or a tethered zip-line

(Sweczak, 2000). Because habitat structure induces

variation in echolocation calls (Schni ler et al., 2003),

we acknowledge that this means our sample is biased

towards call types that are characteristic of more

clu ered environments (i.e. broader bandwidths and

shorter durations).

Signal analysis was undertaken using BatSound

(vers. 3.31, Pe ersson Electronik AB, Sweden). To

avoid pseudo-replication, one call per bat was selected

for description of call parameters and subsequent

analysis. Because two species (Megaderma spasma

and Myotis annectans) were represented by only

two individuals, however, two calls were analysed

for each of these individuals. Additionally, as only

one individual was captured of each of four species

(Hipposideros cineraceus, Kerivoula hardwickii, Tylonyct-

eris pachypus and Miniopterus pusillus), four calls were

analysed for each of these individuals.

For each call, ﬁ ve parameters were measured:

call duration (duration of a single pulse), inter-pulse

interval (IPI, time from the start of one call to the

onset of the next), start frequency (frequency value at

the start of the call), end frequency (frequency value

at the end of the call) and peak frequency (FmaxE,

frequency of maximum energy for the whole call).

Call duration and IPI (ms) were obtained from oscillo-

grams, FmaxE (kHz) from power spectra, whereas the

start frequency (kHz) and end frequency (kHz) were

measured from spectrograms using a 512-size Fast

Fourier Transformation and a Hanning window. An

additional parameter, duty cycle (the amount of time

a bat spends calling relative to the amount of time it

is silent), was calculated by dividing the call duration

by the inter-pulse interval and multiplying by 100 (for

a percentage). All measurements were taken from

the call harmonic containing the greatest energy. The

position of the harmonic containing the most energy

and number of harmonics present in each call were

also noted for the purposes of describing the echoloca-

tion calls produced by each species.

Statistical procedures

To test the eﬃ cacy of acoustic data in correctly identi-

fying bat species, a discriminant function analysis

was performed. Species represented by a single

individual were excluded from the analysis (H.

cineraceus, K. hardwickii, T. pachypus and M. pusillus).

hours and 2,889 m2 mist-net-hours. A single night of

sampling was also undertaken from 18:00–19:30 h

using a harp trap and mist net (capture surface: 15

m2) at a cave entrance in a forest area (13°67’74”N,

104°02’01”E, entrance altitude 183 m a.s.l.) in June

2010. Sampling was avoided on consecutive nights at

the same location.

All bats captured were measured, photographed

and identiﬁ ed in the ﬁ eld using Borissenko & Kruskop

(2003) and Francis (2008). Where necessary to conﬁ rm

species identiﬁ cations, a minimum number of

non-reproductively active individuals were retained

as voucher specimens. All other bats were released as

near as possible to their capture site. Skulls and bacula

(where taxonomically important) of voucher speci-

mens were subsequently examined and all specimens

were deposited at the Centre for Biodiversity Conser-

vation Zoological Collection at the Royal University

of Phnom Penh. A full list of specimen material

examined is given in Annex 1. Nomenclature follows

Simmons (2005), with some modiﬁ cations (Soisook et

al., 2008).

Acoustic methods and call measurement

Time-expanded (x10) recordings of signals produced

by bats were made using a D240x ultrasound detector

with a sampling frequency of 307 kHz (Pe ersson

Electronik AB, Sweden) and stored digitally on

an Edirol R-09HR recorder (Roland, USA) using

a sampling rate of 44.1 kHz, with 16 bits/sample.

Recordings for rhinolophid and hipposiderid bats

Fig. 1 Location of Phnom Kulen National Park in

Cambodia.

Bat echolocation calls

One hundred and ﬁ fteen individuals (25%) were

captured in mist nets, and 345 (75%) were caught in

harp traps. Of the 18 echolocating bat species encoun-

tered, eight were captured in mist nets and harp traps

(Rhinolophus aﬃ nis, R. malayanus, R. pusillus, R. shameli,

R. microglobosus, Hipposideros galeritus, H. larvatus

and H. pomona), ﬁ ve in mist nets only (Megaderma

lyra, M. spasma, Hipposideros armiger, Hesperoptenus

blanfordi and Myotis annectans) and ﬁ ve exclusively in

harp traps (Hipposideros cineraceus, Hypsugo sp. A., T.

pachypus, K. hardwickii and Miniopterus pusillus).

Description of echolocation calls

Time-expanded recordings of 444 echolocation calls

were analysed for all but one of the 18 species captured

during the ﬁ eldwork. Recordings were not obtained

for a single individual designated as Hypsugo sp. A,

for which the correct speciﬁ c name has yet to be deter-

mined.

The ﬁ ve rhinolophid species in our sample

produced calls characterised by a long constant

frequency (CF) component which was preceded and

terminated by a brief frequency-modulated (FM)

component (Table 1, Fig. 3a). The second call harmonic

invariably contained the most energy and all ﬁ ve

species operated at high duty cycles, with mean values

ranging from 73.6 ± 13.9% in R. malayanus to 84.2 ± 3%

in R. pusillus. Peak frequency (FmaxE) values ranged

from 69.5 ± 1.7 kHz in R. shameli to 112.2 ± 1.3 kHz in R.

Because examination of covariance matrices using

Box’s M test indicated that these were not homoge-

nous (F = 13.653, P < 0.001), a quadratic discriminant

function analysis was applied. Cross-validation was

employed in the analysis. Multivariate analysis of

variance (MANOVA) was conducted to examine the

signiﬁ cance of the discriminant function analysis

models and Wilk’s λ values were used to determine

the discrimination power of each variable. All tests

were performed using MINITAB (vers. 15.0), with the

exception of Box’s M test which was performed using

SPSS Statistics (vers. 16.0). In all tests, values of P <

0.05 were considered signiﬁ cant.

Results

Species captured

Over the course of the ﬁ eldwork, 460 individuals

representing 18 echolocating bat species were

captured in ﬁ ve families (Megadermatidae: two

species; Rhinolophidae: ﬁ ve species; Hipposideridae:

ﬁ ve species; Vespertilionidae: ﬁ ve species; Miniop-

teridae: one species). Relative species abundance was

highly uneven with three species representing 71.1%

of all captures: Hipposideros pomona (140 individuals,

30.4% of captures), Rhinolophus shameli (111, 24.1%)

and R. malayanus (76, 16.5%) (Fig. 2).

Fig. 2 Relative abundance of echolocating bat species captured in Phnom Kulen National Park, Cambodia. Figures

show the total number of individuals captured.

Phauk S. et al.

Cambodian Journal of Natural History 2013 (1) 16–26

Fig. 3 Echolocation calls of 17 bat species in Phnom Kulen National Park: (a) Rhinolophidae; (b) Hipposideridae; (c)

Megadermatidae, Vespertilionidae and Miniopteridae. (Note diﬀ erences in x-axis values between ﬁ gures).

Bat echolocation calls

Table 1 Echolocation call parameters of ﬁ ve rhinolophid and ﬁ ve hipposiderid bat species at Phnom Kulen National

Park, Cambodia.

Species

Call

Structure

Start

Frequency

(kHz)

End

Frequency

(kHz)

Frequency

of Maximum

Energy

(kHz)

Call

Duration

(ms)

Inter-Pulse

Interval

(ms)

Duty

Cycle (%) n

RHINOLOPHIDAE

Rhinolophus

aﬃ nis

FM/CF/FM 70.5 ± 3.1

(66–76)

61.5 ± 2

(60–66)

77.1 ± 0.5

(76.5–78.3)

26.6 ± 5.9

(17.9–36.7)

36.9 ± 14

(20.8–63)

76.2 ± 12.5

(54.1–88.5)

Rhinolophus

malayanus

FM/CF/FM 76.1 ± 3.9

(71–82)

78 ± 4.8

(62–83)

83 ± 0.7

(81.1–84.7)

27.1 ± 8.8

(14.8–61.8)

40.9 ± 24.4

(19.1–131.4)

73.6 ± 13.9

(26.2–86.4)

Rhinolophus

pusillus

FM/CF/FM 101.4 ± 4.6

(95–111)

96.8 ± 3.2

(95–105)

112.2 ± 1.3

(108.9–114.1)

22.8 ± 5.3

(14.7–34.1)

27 ± 5.5

(18.2–38.6)

84.2 ± 3

(78.6–89.2)

Rhinolophus

shameli

FM/CF/FM 64 ± 3.1

(52–69)

61 ± 1.9

(59–69)

69.5 ± 1.7

(65.4–71.8)

31.9 ± 6.6

(15.5–46.4)

45.3 ± 22.1

(18.2–171.7)

76.8 ± 15.1

(22.8–90.9)

105

Rhinolophus

microglobosus

FM/CF/FM 94 ± 2.2

(90–97)

93.9 ± 2.1

(90–96)

98.3 ± 0.6

(96.4–98.9)

26.2 ± 7.1

(14.4–39)

34.8 ± 8.5

(21.7–49.4)

74.7 ± 3.4

(66.4–80.1)

HIPPOSIDERIDAE

Hipposideros

armiger

CF/FM 62.7 ± 0.7

(62–64)

55.7 ± 1.4

(53–58)

63.9 ± 0.8

(61.4–65)

10.4 ± 1.7

(8–14.4)

41.3 ± 14.8

(26.8–80.1)

26.7 ± 5.5

(15.7–35.8)

Hipposideros

cineraceus*

CF/FM 149 ± 0.8

(148–150)

129.3 ± 1.3

(128–131)

150 ± 0.8

(149.2–150.6)

5.5 ± 0.4

(5–5.9)

15.6 ± 2.8

(12.1–18.7)

36.2 ± 4.8

(31.6–41.3)

Hipposideros

galeritus

CF/FM 99.5 ± 1.4

(97–102)

90.1 ± 1.5

(87–92)

100.7 ± 1

(98.5–102.5)

5.9 ± 1

(3.9–8.7)

23.3 ± 6

(10.9–37.8)

26.5 ± 7.4

(17.8–53.2)

Hipposideros

larvatus

CF/FM 91.5 ± 0.8

(90–93)

81.5 ± 1.5

(80–87)

92.3 ± 0.8

(90.8–93.5)

6.6 ± 1.2

(5.1–9)

22.6 ± 9.2

(15.3–52.8)

31.5 ± 7.9

(13.6–46.2)

Hipposideros

pomona

CF/FM 134 ± 1.8

(128–139)

111.3 ± 2.9

(105–111.3)

134.8 ± 1.8

(128.3–139.7)

5.1 ± 0.7

(3.7–7.5)

12.2 ± 7.6

(7.1–93.6)

44.9 ± 7.5

(7.1–60)

135

kHz in H. cineraceus. Like the rhinolophids, FmaxE

values did not overlap between species indicating this

call parameter will also aid ﬁ eld identiﬁ cation of all

hipposiderids in our sample from Phnom Kulen.

The two megadermatids in our sample produced

multi-harmonic FM calls (Table 2, Fig. 3c). Megaderma

lyra emi ed signals with a mean FmaxE of 64.7 ± 2.6

kHz, a mean call duration of 2.4 ± 0.8 ms and the third

harmonic contained the greatest energy. Megaderma

spasma produced calls of similar frequency with a

mean FmaxE of 65.4 ± 3.1 kHz (third harmonic), but

mean call durations were somewhat shorter at 1.1 ± 0.2

ms and the ﬁ rst harmonic appeared to be suppressed.

The four vespertilionids that were analysed

produced steep, downward FM calls dominated

by the fundamental harmonic (Table 2, Fig. 3c). All

four species produced relatively brief calls—mean

durations ranging from 0.5 ± 01 ms in K. hardwickii

to 2.4 ± 0.3 ms in Myotis annectans. Mean duty cycles

were generally higher than those of megadermatids

pusillus, whereas call duration ranged from 22.8 ± 5.3

ms in R. pusillus to 31.9 ± 6.6 ms in R. shameli. FmaxE

values did not overlap between species, indicating

this call parameter will be helpful for the ﬁ eld identi-

ﬁ cation of all of the rhinolophid species in our sample

from Phnom Kulen.

The ﬁ ve hipposiderids that were analysed

produced calls beginning with a relatively long and

almost CF component which terminated in a compar-

atively brief and downward FM component (Table 1,

Fig. 3b). This structure facilitates unequivocal separa-

tion of hipposiderid calls from all other bat families

in Phnom Kulen. All species produced calls with the

greatest energy in the second harmonic and operated

at lower duty cycles (mean values ranging from 26.5

± 7.4% in Hipposideros galeritus to 44.9 ± 7.5% in H.

pomona) than rhinolophids due to their shorter and

non-overlapping call durations. Mean values for the

la er ranged from 5.1 ± 0.7 ms in H. pomona to 10.4

± 1.7 ms in H. armiger, while mean FmaxE values

ranged from 63.9 ± 0.8 kHz in H. armiger to 150 ± 0.8

* One call per individual bat was analysed except for H. cineraceus, for which four calls from the same individual were analysed. CF = constant

frequency; FM = frequency-modulated. Values are given as mean ± SD (min–max).

Phauk S. et al.

Cambodian Journal of Natural History 2013 (1) 16–26

Table 2 Echolocation call parameters of two megadermatid, four vespertilionid and one miniopterid bat species at

Phnom Kulen National Park, Cambodia.

Table 3 Cross-validated classiﬁ cation matrix for species emi ing CF calls (genera Hipposideros and Rhinolophus).

Species

Call

Structure

Start

Frequency

(kHz)

End

Frequency

(kHz)

Frequency

of Maximum

Energy (kHz)

Call

Duration

(ms)

Inter-Pulse

Interval

(ms)

Duty

Cycle

(%) n

MEGADERMATIDAE

Megaderma

lyra

FM 72.4 ± 4.1

(66–76)

57 ± 4.1

(53–62)

64.7 ± 2.6

(61.6–67.5)

2.4 ± 0.8

(1.4–3.2)

93.6 ± 38.3

(38.9–134.1)

2.7 ± 0.7

(1.8–3.6)

Megaderma

spasma*

FM 70.8 ± 4.3

(65–74)

62.5 ± 2.1

(60–65)

65.4 ± 3.1

(61.6–69.3)

1.1 ± 0.2

(1–1.3)

68.3 ± 41.9

(25.9–104.4)

2.3 ± 1.5

(1–3.9)

VESPERTILIONIDAE

Hesperoptenus

blanfordi

FM 58 ± 10.8

(45–72)

39.3 ± 7

(35–54)

46.5 ± 6.6

(41.9–61.1)

1.5 ± 0.4

(1.2–2)

40 ± 25.6

(14.3–92.3)

4.7 ± 1.9

(2–7.7)

Myotis

annectans*

FM 50.8 ± 0.5

(50–51)

38 ± 1.2

(37–39)

39.8 ± 0.7

(39.2–40.8)

2.4 ± 0.3

(2.1–2.8)

54.4 ± 21.1

(31.1–80)

5.2 ± 2.8

(2.9–9)

Tylonycteris

pachypus**

FM 68.5 ± 3

(65–71)

46.3 ± 1.5

(45–48)

64.7 ± 1.2

(63.9–66.5)

1.8 ± 0.3

(1.5–2.1)

25 ± 12.6

(14.1–39.5)

8.7 ± 4.8

(3.8–14.1)

Kerivoula

hardwickii**

FM 114.8 ± 10.9

(104–126)

101.3 ± 1.5

(99–102)

103.3 ± 2.2

(100.7–106)

0.5 ± 0.1

(0.4–0.6)

15.9 ± 2

(13.8–18)

3 ± 0.4

(2.7–3.5)

MINIOPTERIDAE

Miniopterus

pusillus**

FM 73.5 ± 8.6

(63–84)

59.8 ± 0.5

(58–60)

60.8 ± 0.6

(60.2–61.6)

3.6 ± 0.3

(3.4–3.9)

48 ± 7

(39.7–54.3)

7.7 ± 1.2

(6.3–8.7)

One call per bat was analysed except for species marked * for which two calls per individual were measured, and ** for which four calls per

individual were measured. CF = constant frequency; FM = frequency-modulated. Values are given as mean ± SD (min–max).

Classiﬁ ed as

True Groups

Hipposideros

armiger

Hipposideros

galeritus

Hipposideros

larvatus

Hipposideros

pomona

Rhinolophus

aﬃ nis

Rhinolophus

malayanus

Rhinolophus

pusillus

Rhinolophus

shameli

Rhinolophus

microglobosus

Hipposideros armiger 1600000000

Hipposideros galeritus 0230000000

Hipposideros larvatus 0021000000

Hipposideros pomona 00013500000

Rhinolophus aﬃ nis 0000150000

Rhinolophus malayanus 0000061000

Rhinolophus pusillus 0000001700

Rhinolophus shameli 00000001050

Rhinolophus microglobosus 0000000015

Total n16 23 21 135 15 61 17 105 15

no. correct 16 23 21 135 15 61 17 105 15

% correct 100 100 100 100 100 100 100 100 100

The discriminant function analysis model relied on two parameters (Duration and FmaxE) and provided an overall correct classiﬁ cation rate

of 100% when cross-validated.

Bat echolocation calls

and mean FmaxE values ranged from 39.8 ± 0.7 kHz

in M. annectans to 103.3 ± 2.2 kHz in K. hardwickii. The

only miniopterid in our sample, Miniopterus pusillus,

produced steep, downward FM signals similar to

vespertilionids, but of longer mean call duration (3.6 ±

0.3 ms) and a mean FmaxE of 60.8 ± 0.6 kHz.

Discriminant function analysis

Thirteen bat species were assessed in the analysis.

As these could be unequivocally separated into two

groups by their call structures, quadratic discriminant

analysis was undertaken for (i) species whose calls

contained a CF portion terminating in an FM portion

(ﬁ ve rhinolophids and four hipposiderids) (“CF

group”); and (ii) species whose calls comprised an FM

signal (two megadermatids and two vespertilionids)

(“FM group”).

Quadratic discriminant function analysis for the

CF group resulted in a 100% correct classiﬁ cation

rate (408 calls correctly classiﬁ ed) which remained

unchanged when cross-validated (Table 3). The best

model relied upon two parameters (call duration and

FmaxE), which a MANOVA showed was signiﬁ cant

(Wilk’s λ = 0.00102, F = 1510.276, P < 0.001). Wilk’s

λ values indicated that the discrimination power of

the two variables in decreasing order was: FmaxE

(0.00271) and call duration (0.16016).

Despite the small sample sizes for species in the FM

group, the best model (relying on two call parameters:

FmaxE and IPI) produced a 90% correct classiﬁ ca-

tion rate (18 calls correctly classiﬁ ed out of 20) and a

70% correct classiﬁ cation rate when cross-validated

(14 calls correctly classiﬁ ed) (Table 4). MANOVA

demonstrated that the model was signiﬁ cant (Wilk’s

λ = 0.08927, F = 11.735, P < 0.001) and Wilk’s λ values

indicated that the discrimination power of the two

variables in decreasing order was: FmaxE (0.1224) and

IPI (0.65160).

Discussion

Ours is the ﬁ rst study to describe the echolocation

calls produced by a Cambodian bat assemblage and

in achieving a correct, cross-validated classiﬁ cation

rate of 85% overall, our results indicate that correct

acoustic identiﬁ cation of in-country bat species is

feasible using the call parameters we employed.

The call parameters we recorded for each species

are generally consistent with those of other studies

in the region (Soisook et al., 2008; Furey et al., 2009;

Douangboubpha et al., 2010; Hughes et al., 2010,

2011; Kingsada et al., 2011; Ith et al., 2011b), although

because sample sizes for megadermatid, vespertil-

ionid and miniopterid species were small (and for

reasons stated in the methods), it is highly unlikely

these encompass the full range of variation in calls

produced by these taxa. For instance, because Keriv-

oula spp. are known to produce signals with very high

starting frequencies (Kingston et al., 1999; Schmieder

et al., 2010), these were likely missed in our recordings

due to insuﬃ cient sampling frequencies, resulting in

lower starting frequencies and shorter call durations.

Notwithstanding this, the rate of correct classiﬁ ca-

tion we obtained in discriminant function analysis

is comparable to similar studies of bats around the

world. For instance, MacSwiney et al. (2008) achieved

a correct classiﬁ cation rate of 84% for 26 species in

Mexico; Russo & Jones (2002) a rate of 82% for 22

species in Italy; and Kofoky et al. (2009) a rate of 82%

for 15 species in Madagascar. Our results thus support

previous suggestions (Furey et al., 2009) that acoustic

identiﬁ cation of free-ﬂ ying bats is an equally achiev-

able goal in Southeast Asia.

Signiﬁ cant additional research will be required

to realise this goal, however. As intra-speciﬁ c varia-

tion occurs in echolocation calls due to geographical

location (Thomas et al., 1987), reference recordings

from every site under investigation will be required

to reliably identify species whose call parameters

Table 4 Cross-validated classiﬁ cation matrix for species

emi ing FM calls (genera Megaderma, Hesperoptenus and

Myotis).

Classiﬁ ed as

True Groups

Megaderma

lyra

Megaderma

spasma

Hesperoptenus

blanfordi

Myotis

annectans

Megaderma lyra 3100

Megaderma spasma 2310

Hesperoptenus blanfordi 0062

Myotis annectans 0002

Total n5474

no. correct 3362

% correct 60.0 75.0 85.7 50.0

The discriminant function analysis model relied on two parameters

(FmaxE, IPI) and provided an overall correct classiﬁ cation rate of

70.0% when cross-validated.

Phauk S. et al.

Cambodian Journal of Natural History 2013 (1) 16–26

may overlap with those of others in certain parts of

their range. Second, as habitat structure also induces

variation in echolocation calls (Schni ler et al., 2003),

recordings from a range of structural environments

will be required to elucidate the full repertoire of

calls produced by diﬀ erent species. This will require

signiﬁ cant ﬁ eld eﬀ ort to obtain suﬃ cient recordings

for less abundant (or simply rarely captured) taxa, as

demonstrated by the highly uneven relative species

abundances encountered in the present study.

Because acoustic methods are unlikely to improve

upon results provided by harp traps for bat species

that echolocate at very low intensities (e.g. species

within the Murininae and Kerivoulinae and Coelops

frithii) (Furey et al., 2009), this approach is perhaps

best regarded as an important complement to, rather

than a replacement of, traditional capture methods for

inventorying echolocating bats in Southeast Asia. As

the taxonomy of many Southeast Asian bats remains

uncertain (Francis et al., 2010), the need for live-

trapping and collecting voucher samples to ensure

correct assignment of names and recognition of species

limits will inevitably also continue. We nonetheless

recommend further studies to facilitate development

of acoustic sampling as a tool for improving under-

standing and conservation of Cambodian bats.

Acknowledgements

We are grateful to H.E. Chay Smith, Mr Hong Daravuth

and Mr Seng Ra anak (General Department for

Administration of Nature Conservation and Protec-

tion, Cambodian Ministry of Environment) for their

generous support and kindly facilitating the relevant

ﬁ eld permits. Additionally, the authors are grateful

to Markus Handschuh, Alistair Mould, Nikki Hulse

(Angkor Center for Conservation of Biodiversity) and

Ben Hayes for their invaluable help in the ﬁ eld and

to Martin Fisher, Stuart Paterson and Elizabeth Allen

of Fauna & Flora International who provided many

comments on an early draft of this paper. We are

similarly indebted to Phal Des and Meak Kamerane

(Royal University of Phnom Penh) for their assistance.

We also thank Choun Phirom (Fauna & Flora Inter-

national) for preparing Fig. 1 and Dave Waldien (Bat

Conservation International), Tigga Kingston (Texas

Tech), Paul Bates (Harrison Institute) and Paul Racey

(University of Exeter) for their steadfast support.

The research was supported by the Darwin Initiative

(14-011; 14-037; 18-002; EIDPO028), the John D. and

Catherine D. MacArthur Foundation (US: 09-92411-

000-GSS), the Zoological Parks and Gardens Board

of Victoria (Australia) and the Critical Ecosystem

Partnership Fund (through a grant provided to the

Harrison Institute). Lastly, our special thanks to Arjan

Boonman and one anonymous reviewer who kindly

commented on the text.

References

Borissenko, A.V. & Kruskop, S.V. (2003) Bats of Vietnam and

Adjacent Territories: An Identiﬁ cation Manual. Joint Russian-

Vietnamese Science and Technological Tropical Centre,

Zoological Museum of Moscow M.V. Lomonosov State

University, Moscow, Russia.

Brigham, R.M., Kalko, E.K.V., Jones, G., Parsons, S. &

Limpens, H.J.G.A. (eds) (2004) Bat Echolocation Research:

Tools, Techniques and Analysis. Bat Conservation Interna-

tional, Texas, USA.

Douangboubpha, B., Bumrungsri, S., Satasook, C., Bu,

S.S.H., Harrison, D.L. & Bates, P.J.J. (2010) A taxonomic

review of Hipposideros halophyllus, H. ater and H. cineraceus

(Chiroptera: Hipposideridae) in Thailand and Myanmar.

Acta Chiropterologica, 12, 29–50.

Francis, C.M. (2008) A Guide to the Mammals of South-East

Asia. Princeton University Press, Princeton, New Jersey,

USA.

Francis, C.M., Borissenko, A.V., Ivanova, N.V., Eger, J.L.,

Lim, B.K., Guillén-Servent, A., Kruskop, S.V., Mackie,

I. & Hebert, P.D.N. (2010) The role of DNA barcodes in

understanding and conservation of mammal diversity in

Southeast Asia. PLoS ONE, 5, e12575; doi: 10.1371/journal.

pone.0012575.

Furey, N.M., Mackie, I.J. & Racey, P.A. (2009) The role of

ultrasonic bat detectors in improving inventory and

monitoring surveys in Vietnamese karst bat assemblages.

Current Zoology, 55, 327–341.

Furey, N.M., Phauk S., Phen S., Chheang S., Ith S., Bates,

P.J.J. & Csorba, G. (2012) New country records for ﬁ ve bat

species (Chiroptera) from Cambodia. Cambodian Journal of

Natural History, 2012, 141–149.

Hou K., Ken Serey R., Ung L. & Mao S. (2004) Management of

Pilot Watershed Areas in Cambodia: Baseline Survey. Working

paper 12b. Mekong River Commission and Deutsche

Gesellschaft für Technische Zusammenarbeit (GTZ)

Cooperation Programme, Vientiane, Lao PDR.

Hughes, A.C., Satasook, C., Bates, P.J.J., Soisook, P., Sritong-

chuay, T., Jones, G. & Bumrungsri, S. (2010) Echolocation

call analysis and presence-only modeling as conservation

monitoring tools for rhinolophid bats in Thailand. Acta

Chiropterologica, 12, 311–327.

Hughes, A.C., Satasook, C., Bates, P.J.J., Soisook, P., Sritong-

chuay, T., Jones, G. & Bumrungsri, S. (2011) Using

echolocation calls to identify Thai bat species: Vespertilio-

nidae, Emballonuridae, Nycteridae and Megadermatidae.

Acta Chiropterologica, 13, 447–455.

Bat echolocation calls

Ith S., Csorba, G., Bates, P.J.J. & Furey, N.M. (2011a) Conﬁ r-

mation of seven bat species for Cambodia. Cambodian

Journal of Natural History, 2011, 93–103.

Ith S., Soisook, P., Bumrungsri, S., Kingston, T., Puechmaille,

S.J., Struebig, M.J., Si Si, H.B., Vu D.T., Furey, N.M.,

Thomas, N.M. & Bates, P.J.J. (2011b) A taxonomic review

of Rhinolophus coelophyllus Peters 1867 and R. shameli Tate

1943 (Chiroptera: Rhinolophidae) in continental Southeast

Asia. Acta Chiropterologica, 13, 41–59.

Jones, G. & Teeling, E.C. (2006) The evolution of echoloca-

tion in bats. Trends in Ecology and Evolution, 21, 149–156.

Jones, G., Jacobs, D.A., Kunz, T.H., Willig, M.R. & Racey, P.A.

(2009) Carpe noctem: the importance of bats as bio-indica-

tors. Endangered Species Research, 8, 93–115.

Kingsada, P., Douangboubpha, B., Ith S., Furey, N., Soisook,

P., Bumrungsri, S., Satasook, C., Vu D.T., Csorba, G.,

Harrison, D., Pearch, M., Bates, P. & Thomas, N. (2011) A

checklist of bats from Cambodia, including the ﬁ rst record

of the intermediate horseshoe bat Rhinolophus aﬃ nis

(Chiroptera: Rhinolophidae), with additional information

from Thailand and Vietnam. Cambodian Journal of Natural

History, 2011, 49–59.

Kingston, T. (2010) Research priorities for bat conservation

in Southeast Asia: a consensus approach. Biodiversity and

Conservation, 19, 471–484.

Kingston, T. (in press) Response of bat diversity to forest

disturbance in Southeast Asia – insights from long-term

research in Malaysia. In Bat Evolution, Ecology, and Conser-

vation (eds R.A. Adams & S.C. Pedersen). Springer Science

Press, USA.

Kingston, T., Jones, G., Zubaid, A. & Kunz, T. (1999) Echolo-

cation signal design in Kerivoulinae and Murininae

(Chiroptera: Vespertilionidae) from Malaysia. Journal of

Zoology (London), 249, 359–374.

Kofoky, A.F., Randrianandrinina, F., Russ, J., Raharinante-

naina, I., Cardiﬀ , S.C., Jenkins, R.K. & Racey, P.A. (2009).

Forest bats of Madagascar: results of acoustic surveys.

Acta Chiropterologica, 11, 375–392.

Kunz, T.H., Braun de Torrez, E., Bauer, D., Lobova, T. &

Fleming, T.H. (2011) Ecosystem services provided by bats.

Annals of the New York Academy of Sciences, 1223, 1-38.

MacSwiney, M.C.G., Clarke, M.F. & Racey, P.A. (2008) What

you see is not what you get: the role of ultrasonic detec-

tors in increasing inventory completeness in Neotropical

bat assemblages. Journal of Applied Ecology, 45, 1364-1371.

Neou B., Khou E. & Touch S. (2008) Preliminary Study

of the Kulen National Park for Development of a Botanical

Garden. Ministry of Environment, Royal Government of

Cambodia, Phnom Penh, Cambodia.

Russo, D. & Jones, G. (2002) Identiﬁ cation of twenty-two bat

species (Mammalia: Chiroptera) from Italy by analysis of

time-expanded recording of echolocation calls. Journal of

Zoology (London), 258, 91–103.

Schmieder, D.A., Kingston, T., Hashim, R. & Siemers, B.M.

(2010) Breaking the trade-oﬀ : rainforest bats maximise

bandwidth and repetition rate of echolocation calls as they

approach prey. Biology Le ers, 6, 604–609.

Schni ler, H.U., Moss, C.F. & Denzinger, A. (2003) From

spatial orientation to food acquisition in echolocating bats.

Trends in Ecology and Evolution, 18, 386–394.

Simmons, N.B. (2005) Order Chiroptera. In Mammal Species

of the World: a Taxonomic and Geographic Reference, Third

Edition (eds D.E. Wilson & D.M. Reeder), pp. 312–529.

Johns Hopkins University Press, Baltimore, USA.

Soisook, P., Bumrungsri, S., Satasook, C., Thong V.D., Bu

S.S.H., Harrison, D.L. & Bates, P.J.J. (2008) A taxonomic

review of Rhinolophus stheno and R. malayanus (Chirop-

tera: Rhinolophidae) from continental Southeast Asia:

an evaluation of echolocation call frequency in discrimi-

nating between cryptic species. Acta Chiropterologica, 10,

221–242.

Sweczak, J.M. (2000) A tethered zip-line arrangement for

reliably collecting bat echolocation reference calls. Bat

Research News, 41, 142.

Thomas, D.W., Bell, G.P. & Fenton, M.B. (1987) Variation

in echolocation call frequencies recorded from North

American vespertilionid bats: a cautionary note. Journal of

Mammalogy, 68, 842–847.

About the authors

PHAUK SOPHANY is originally from Sihanoukville

and has worked as national coordinator for the Centre

for Biodiversity Conservation at the Royal Univer-

sity of Phnom Penh since 2011. He studied the use of

acoustic approaches for identiﬁ cation of Cambodian

bat species for his MSc degree and has a special interest

in the ecology of cave-dwelling bats and ﬂ ying foxes.

PHEN SARITH is a Cambodian national from Kampot

Province. After studying the eﬀ ects of forest distur-

bance on bat species in Phnom Kulen National Park,

he graduated from the Royal University of Phnom

Penh with an MSc in 2011. Following this, he worked

as a part-time ﬁ eld researcher for Fauna & Flora Inter-

national and the Wildlife Conservation Society. His

interests include the ecology of cave-dwelling bats.

NEIL FUREY has worked in Southeast Asia since 1997,

spending a decade in Vietnam and completing various

assignments in Cambodia, China, India, Indonesia

and Myanmar. A biologist by training, he studied

the ecology of Vietnamese karst bat assemblages for

his doctorate and has a special interest in community

ecology and systematics. Much of his work in South-

east Asia focuses on strengthening conservation and

research capacity.

Phauk S. et al.

Cambodian Journal of Natural History 2013 (1) 16–26

Annex 1

RHINOLOPHIDAE—Rhinolophus aﬃ nis: CBC00927, male, in

spirit, skull removed, collected on 25 April 2010, 13°41.028N,

104°01.366E, 82 metres above sea level (m a.s.l.); CBC00942,

CBC00943, males, in spirit, skulls removed, collected on 23

June 2010, 13°41.409N, 104°00.733E, 177 m a.s.l.; CBC00947,

CBC00948, CBC00949, males, in spirit, skulls removed,

collected on 24 July 2010, 13°40.301N, 104°01.510E, 72 m a.s.l.

(described by Kingsada et al., 2011). Rhinolophus malayanus:

CBC00904, male, in spirit, skull removed, collected on 21 April

2010, 12°46.887N, 103°27.806E, 205 m a.s.l.; CBC00921, male, in

spirit, skull removed, collected on 22 April 2010, 13°40.855N,

104°01.244E, 72 m a.s.l. Rhinolophus pusillus: CBC00933, male,

in spirit, skull removed, collected on 19 May 2010, 13°41.295N,

104°00.739E, 215 m a.s.l.; CBC00935, female, in spirit, skull

removed, collected on 20 May 2010, 13°41.189N, 104°00.642E,

182 m a.s.l. Rhinolophus shameli: CBC00905, CBC00906, females,

in spirit, skulls removed, collected on 21 April 2010, 12°46.887N,

103°27.806E, 205 m a.s.l.; CBC00926, female, in spirit, skull

removed, collected on 25 April 2010, 13°41.028N, 104°01.366E,

82 m a.s.l.; CBC00928, female, in spirit, skull removed, collected

on 18 May 2010, 13°41.340N, 104°00.668E, 171 m a.s.l. Rhinolo-

phus microglobosus: CBC00901, male, in spirit, skull removed,

collected on 21 April 2010, 12°46.887N, 103°27.806E, 205 m a.s.l.;

CBC00930, female, in spirit, skull removed, collected on 18 May

2010, 13°41.340N, 104°00.668E, 171 m a.s.l.; CBC00936, female,

in spirit, skull removed, collected on 20 May 2010, 13°41.189N,

104°00.642E, 182 m a.s.l.

HIPPOSIDERIDAE—Hipposideros armiger: CBC00923,

CBC00924, male and female, in spirit, skulls removed,

collected on 24 April 2010, 13°40.944N, 104°01.134E, 97 m a.s.l.;

CBC00938, male, in spirit, skull removed, collected on 23 May

2010, 13°40.598N, 104°01.506E, 66 m a.s.l.; CBC00941, female,

in spirit, skull removed, collected on 22 June 2010, 13°41.495N,

104°00.647E, 204 m a.s.l. Hipposideros cineraceus: CBC00944,

female, in spirit, skull removed, collected on 26 June 2010,

13°40.816N, 104°00.973E, 183, m a.s.l. Hipposideros galeritus:

CBC00898, CBC00899, CBC00900, two females and one male, in

spirit, skulls removed, collected on 21 April 2010, 12°46.887N,

103°27.806E, 205 m a.s.l.; CBC00931, male, in spirit, skull

removed, collected on 18 May 2010, 13°41.340N, 104°00.668E,

171 m a.s.l.; CBC00932, male, in spirit, skull removed, collected

on 19 May 2010, 13°41.295N, 104°00.739E, 215 m a.s.l.;

CBC00950, male, in spirit, skull removed, collected on 25 July

2010, 13°40.092N, 104°01.399E, 68 m a.s.l. Hipposideros larvatus:

CBC00925, female, in spirit, skull removed, collected on 27 April

2010, 13°40.949N, 104°01.416E, 80 m a.s.l.; CBC00929, female,

in spirit, skull removed, collected on 18 May 2010, 13°41.340N,

104°00.668E, 171 m a.s.l. Hipposideros pomona: CBC00902, male,

in spirit, skull removed, collected on 21 April 2010, 12°46.887N,

103°27.806E, 205 m a.s.l.; CBC00903, female, in spirit, skull

removed, collected on 24 April 2010, 13°40.944N, 104°01.134E, 97

m a.s.l.; CBC00934, female, in spirit, skull removed, collected on

20 May 2010, 13°41.189N, 104°00.642E, 182 m a.s.l.; CBC00939,

CBC00940, male and female, in spirit, skulls removed, collected

on 20 June 2010, 13°40.796N, 104°01.593E, 65 m a.s.l.

MEGADERMATIDAE—Megaderma lyra: CBC00919, female, in

spirit, skull removed, collected on 21 April 2010, 12°46.887N,

103°27.806E, 205 m a.s.l.; CBC00920, female, in spirit, skull

removed, collected on 24 April 2010, 13°40.944N, 104°01.134E,

97 m a.s.l. Megaderma spasma: CBC00945, female, in spirit, skull

removed, collected on 20 July 2010, 13°41.339N, 104°00.970E, 188

m a.s.l.

VESPERTILIONIDAE—Hesperoptenus blanfordi: CBC00907,

CBC00911, CBC00912, CBC00913, CBC00914, CBC00915,

CBC00916, four females and three males, in spirit, skulls removed,

collected on 21 April 2010, 12°46.887N, 103°27.806E, 205 m a.s.l.

Hypsugo sp. A: CBC00917, male, in spirit, skull and baculum

removed, collected on 21 April 2010, 12°46.887N, 103°27.806E,

205 m a.s.l. Myotis annectans: CBC00909, CBC00918, females, in

spirit, skulls removed, collected on 21 April 2010, 12°46.887N,

103°27.806E, 205 m a.s.l.; CBC00937, female, in spirit, skull

removed, collected on 23 May 2010, 13°40.598N, 104°01.506E, 66

m a.s.l. Tylonycteris pachypus: CBC00908, CBC00910, females, in

spirit, skulls removed, collected on 21 April 2010, 12°46.887N,

103°27.806E, 205 m a.s.l. Kerivoula hardwickii: CBC00946, female,

in spirit, skull removed, collected on 21 July 2010, 13°41.574N,

104°00.633E, 206 m a.s.l.

MINIOPTERIDAE—Miniopterus pusillus: CBC00951, male, in

spirit, skull removed, collected on 25 July 2010, 13°39.810N,

104°01.862E, 68 m.a.s.l (described by Furey et al., 2012).

Deciduous dipterocarp forest in the Eastern Plains

Structure and composition of deciduous dipterocarp forest in the

Eastern Plains Landscape, Cambodia

PIN Chanrattana1, PHAN Channa1,2, PRUM Sovanna1 and Thomas N.E. GRAY1,3,*

1 WWF Greater Mekong Cambodia Country Program, House #21, Street 322, Sangkat Boeung Keng Kang 1, Khan

Chamkamorn, Phnom Penh, Cambodia.

2 (Current address) Fauna & Flora International: Cambodia Programme, House #19, Street 360, Sangkat Boeung

Keng Kang 1, Phnom Penh, Cambodia.

3 (Current address) WWF Greater Mekong, House 39, Unit 05, Ban Saylon, Vientiane, Lao PDR.

*Corresponding author. Email tomnegray@hotmail.com

Paper submitted 30 April 2013, revised manuscript accepted 24 June 2013.

Abstract

Deciduous dipterocarp forests (DDF) are unique to Southeast Asia and the extensive DDF of northern and eastern

Cambodia are globally signiﬁ cant for biodiversity conservation. However there are few published studies on the

structure and botanical composition of DDF stands in Cambodia. We inventoried all tree stems >5 cm diameter

at breast height (dbh) in two one-hectare DDF plots in two protected areas (Mondulkiri Protected Forest and

CITATION: Pin C., Phan C., Prum S. & Gray, T.N.E. (2013) Structure and composition of deciduous dipterocarp forest in the

Eastern Plains Landscape, Cambodia. Cambodian Journal of Natural History, 2013, 27–34.

Pin C. et al.

Introduction

Deciduous dipterocarp forests (DDF) are endemic

to Asia and historically occurred widely across

continental Southeast Asia from Northeast India

through Myanmar and Thailand to Laos and Vietnam

(Bunyavejchewin et al., 2011). DDF landscapes are

restricted to lowlands below 900 m with a relatively

severe dry-season and poor (acidic, shallow, and

sandy) soils. A low diversity of tree species and

abundance of stems in small classes, compared

to other Southeast Asian forest types, are typical

(Rundel, 1999; Sahunalu, 2009, Tani et al., 2007;

Bunyavejchewin et al., 2011). One or more of four

deciduous species of Dipterocarpaceae (Shorea obtusa,

S. siamensis, Dipterocarpus tuberculatus and D. obtusi-

folius) dominate DDF (Bunyavejchewin et al., 2011).

Other important canopy species include Pterocarpus

macrocarpus (Fabaceae), Xylia xylocarpa (Fabaceae),

Gluta usitata (Anacardiaceae) and several species of

Terminalia (Combretaceae) (Rundel, 1999; Bunyave-

jchewin et al., 2011). The understory is dominated by

grasses and herbaceous bamboo (Vietnamosasa spp.)

that provide abundant forage for the region’s famous,

and historically abundant, large ungulates (Wharton

1957; Bunyavejchewin et al., 2011).

Historically, DDF was the most extensive forest

type across Thailand and Cambodia, but in common

with tropical dry forests globally, DDF landscapes

have been increasingly degraded and converted to

agricultural uses (Tordoﬀ et al., 2005). Indeed, tropical

dry forests are the most threatened major forest type

globally (Janzen, 1988) and the DDF of Indochina

represents one of the ﬁ ve largest areas of tropical

dry forest remaining. In a recent review, Indochi-

nese DDF was identiﬁ ed as amongst the four most

severely threatened dry forest types globally (Miles et

al., 2006). The DDF within the northern and eastern

plains of Cambodia form part of the Lower Mekong

Dry Forest Ecoregion and are globally irreplaceable

for conservation (Tordoﬀ et al., 2005), supporting

one of the most intact assemblages of megafauna in

Southeast Asia including the largest global popula-

tions of banteng Bos javanicus (Gray et al., 2012) and

white-shouldered ibis Pseudibis davisoni (Wright et

al., 2012). However, despite their wide occurrence

across Cambodia and high conservation value, there

are few recent published studies on the structure and

species composition of DDF. The aim of this study

was to undertake basic inventories, using repeatable

and simple methodologies, of two one-hectare plots in

DDF within the Eastern Plains Landscape, Mondulkiri

in eastern Cambodia. Our principal focus was to deter-

mine stem densities and basal areas of the dominant

species which each plot.

Methods

Study Sites

The study was conducted in two protected areas

within the Eastern Plains Landscape, Mondulkiri.

Phnom Prich Wildlife Sanctuary (2,225 km2) and

Mondulkiri Forest Protected (3,350 km2) (Fig. 1).

Within the two protected areas, DDF make up >75%

of forest cover with smaller areas of mixed deciduous

forest and, mostly in PPWS, semi-evergreen forest

(Fig. 1). The climate, soil and geology of the two

protected areas have not been extensively studied, but

are believed to be representative of lowland areas of

eastern Cambodia, with distinct dry and wet seasons

inﬂ uenced by the tropical monsoon.

Botanical survey

A one-hectare (100 m x 100 m) plot was established

Phnom Prich Wildlife Sanctuary) in the Eastern Plains Landscape, Mondulkiri. Both plots were dominated by

two species of deciduous dipterocarp: Shorea obtusa and Dipterocarpus tuberculatus. These two species represented

70–80% of individual stems >5 cm dbh and 75–80% of the basal area within the plots. Overall basal area was

similar between the two plots (14.2 m2/ha in Phnom Prich Wildlife Sanctuary; 15.0 m2/ha in Mondulkiri Protected

Forest) and within the range of published estimates from DDF plots in Indochina. Despite similar basal areas, the

Phnom Prich Wildlife Sanctuary plots had considerably more individual stems >5 cm dbh, particularly within

smaller size classes (5–10 cm dbh). We suggest that the structural diﬀ erences between the plots may be the result

of less frequent ﬁ res in Phnom Prich Wildlife Sanctuary. However, considerable further research is required to

understand the botanical and ecological processes that inﬂ uence DDF in Cambodia.

Keywords

basal area, botanical plot, deciduous dipterocarp forest, ﬁ re ecology, Indochina, Mondulkiri.

Deciduous dipterocarp forest in the Eastern Plains

within DDF in each of the core zones of Mondulkiri

Protected Forest (MPF; UTM 48N 737196E 1433940N,

190 m a.s.l.) and Phnom Prich Wildlife Sanctuary

(PPWS; UTM 48N 717388E 1409090N, 220 m a.s.l.)

(Fig. 1). The two plots were non-randomly situated

so as to be representative of typical DDF within each

protected area. Therefore both plots were placed on ﬂ at

land without rivers or streams and with continuous

homogenous DDF and limited habitat heterogeneity.

The disturbance history of the two plots is unclear

but both are within the core areas of protected areas

with no recent forest clearance. However, both plots

are likely to experience annual burning, which is

widespread throughout the DDF of the Eastern Plains

Landscape.

Plots were marked using a compass and measuring

tape with each plot divided into one hundred 10 m x 10

m quadrates marked using tape. Within each quadrat

all trees with a diameter at breast height (dbh, c. 1.3

m) equal to or more than 5 cm were measured (dbh

recorded) and identiﬁ ed to species. For species identi-

ﬁ cation, vernacular Khmer names were recorded and

checked against Dy Phon (2000) for scientiﬁ c names.

When no consensus on species identiﬁ cation was

achieved, individual trees were recorded as unidenti-

ﬁ ed forms. We acknowledge that this approach is not

optimal and that herbarium samples should have been

taken and properly preserved to allow for subsequent

identiﬁ cation. Given that a number of forms likely to

represent species were not identiﬁ ed, information on

tree species richness and diversity within the plots is

necessarily incomplete. Our study is therefore best

regarded as providing information on stand structure,

basal area, and dominant species composition rather

than a complete botanical inventory of tree species

richness.

Basal area (m2) was calculated for each individual

stem and summed in each plot for each species and

used to calculate total basal area. Fieldwork was

conducted during the dry season: February 2012 in

Phnom Prich Wildlife Sanctuary, and November 2012

in Mondulkiri Protected Forest.

Fig. 1 Location of botanical plots (dots) within Mondulkiri Protected Forest and Phnom Prich Wildlife Sanctuary,

Mondulkiri, Cambodia. Semi-evergreen forests in the protected areas are indicated in black.

Pin C. et al.

Table 1 Species list and structural summary based on inventorying all stems >5 cm dbh within one hectare plots in

deciduous dipterocarp forest in Mondulkiri Protected Forest and Phnom Prich Wildlife Sanctuary. Basal Area (BA) was

calculated using all stems and presented as m2.

Species and Family Stems Mean dbh BA % Stems % BA

Mondulkiri Protected Forest

Shorea obtusa Miq. Dipterocarpaeae 171 17.1 5.5 42.8 36.8

Dipterocarpus tuberculatus Roxb. Dipterocarpaeae 100 24.8 5.7 25.0 38.0

Unidentiﬁ ed species (seven forms) 42 15.3 1.1 10.5 7.3

Xylia xylocarpa (Roxb.) Taub. Fabaceae 25 16.5 0.7 6.3 4.9

Terminalia alata Heyne ex. Roth Combretaceae 20 23.7 1.2 5.0 7.9

Catunaregam tomentosa Blume ex DC. Rubiaceae 11 9.7 0.1 2.8 0.6

Terminalia chebula Re . Combretaceae 9 13.9 0.2 2.3 1.2

Dalbergia spp. Fabaceae 4 7.1 0.0 1.0 0.1

Bauhinia acuminata L. Fabaceae 4 15.0 0.1 1.0 0.5

Lagerstroemia speciosa (L.) Pers. Lythraceae 4 15.0 0.1 1.0 0.6

Pterocarpus macrocarpus Kurz. Fabaceae 3 9.7 0.0 0.8 0.2

Acacia intsia L. Fabaceae 2 31.5 0.2 0.5 1.1

Careya sphaerica Roxb. Lecythidaceae 2 22.8 0.1 0.5 0.6

Diospyros ehretioides Wall. ex G.Don Ebenaceae 1 8.0 0.0 0.3 0.0

Cratoxylum formosum Benth. Hypercaceae 1 5.7 0.0 0.3 0.0

Morinda citrifoila L. Rubiaceae 1 9.9 0.0 0.3 0.1

Phnom Prich Wildlife Sanctuary

Shorea obtusa Miq. Dipterocarpaeae 512 10.4 6.5 43.0 45.4

Dipterocarpus tuberculatus Roxb. Dipterocarpaeae 418 9.9 4.9 35.1 34.4

Terminalia alata Heyne ex. Roth. Combretaceae 80 12.4 1.3 6.7 8.9

Unidentiﬁ ed species (seven forms) 72 8.9 0.6 6.0 4.4

Albizia myriophylla Benth. Fabaceae 37 9.1 0.3 3.1 2.3

Terminalia chebula Re . Combretaceae 18 8.3 0.1 1.5 0.8

Xylia xylocarpa (Roxb.) Taub. Fabaceae 8 8.0 0.0 0.7 0.3

Pterocarpus macrocarpus Kurz. Fabaceae 8 7.2 0.0 0.7 0.2

Lagerstroemia speciosa (L.) Pers. Lythraceae 7 14.1 0.1 0.6 1.0

Dipterocarpus obtusifolius Teijsm ex Miq. Dipterocarpaeae 6 7.4 0.0 0.5 0.2

Catunaregam tomentosa Blume ex DC. Rubiaceae 5 7.0 0.0 0.4 0.1

Careya sphaerica Roxb. Lecythidaceae 4 8.2 0.0 0.3 0.2

Lophopetalum spp. Celastraceae 3 7.9 0.0 0.3 0.1

Diospyros ehretioides Wall. ex G.Don Ebenaceae 3 7.1 0.0 0.3 0.1

Phyllanthus emblica L. Phyllanthaceae 3 7.9 0.0 0.3 0.1

Shorea spp. Dipterocarpaea