Identifying the occurrence and distribution of southern white cheeked gibbon nomascus siki delacour 1951 using autonomous recorders a case study in the khe nuoc trong proposed nature reserve

“IDENTIFYING THE OCCURRENCE AND DISTRIBUTION OF SOUTHERN WHITE- CHEEKED GIBBON NOMASCUS SIKI DELACOUR, 1951 USING AUTONOMOUS RECORDERS: A CASE STUDY IN THE KHE NUOC TRONG PROPOSED NATU

“IDENTIFYING THE OCCURRENCE AND DISTRIBUTION OF SOUTHERN

WHITE- CHEEKED GIBBON NOMASCUS SIKI (DELACOUR, 1951) USING

AUTONOMOUS RECORDERS: A CASE STUDY IN THE KHE NUOC

TRONG PROPOSED NATURE RESERVE’’

MASTER THESIS IN FOREST SCIENCE

Hanoi, November 2018

“IDENTIFYING THE OCCURRENCE AND DISTRIBUTION OF SOUTHERN

WHITE- CHEEKED GIBBON NOMASCUS SIKI (DELACOUR, 1951) USING

AUTONOMOUS RECORDERS: A CASE STUDY IN THE KHE NUOC

TRONG PROPOSED NATURE RESERVE’’

Major: Forest Science Code: 8620201

MASTER THESIS IN FOREST SCIENCE Supervisor: Assoc Prof Dr Vu Tien Thinh

Signature: ………

Hanoi, Novemeber 2018

ABSTRACT

It is necessary to have accurate information on a species occurrence and distribution

in order to monitor and protect endangered species While traditional survey methods have several limitations, using automatic recorders in monitoring wildlife provides significant promise for detecting the gibbon sounds because gibbon produce loud and elaborate sounds Autonomous recorder was used to record the gibbon calls and the associated sound analysis software (RAVEN Software) to determine the occurrence and distribution of the Southern white-cheeked gibbons

(Nomascus siki) in Khe Nuoc Trong proposed nature reserve The survey was

conducted in 8 compartments of the Khe Nuoc Trong proposed nature reserve for

27 days, 53 recording posts were set up in the field and the sounds were recorded for 1-3 consecutive days From 7 recording posts no gibbon call was detected and from 46 recording posts gibbons call was detected

The recorders were set up to record the sounds of gibbon in the nature reserve for all the day Gibbons usually produce loud sounds early in the morning that last for 10-30 minutes which starts from 5am to 9am Therefore, data was analyzed from 5am to 9am in the morning The male and female gibbons produce different sounds and have different song structures from which we can recognize them easily In this study 26 gibbon calls were detected from the recordings The gibbon groups are distributed to the west and near the central part of the nature reserve Weather has also a great effect on the gibbon singing which decreases the singing of the gibbon and the gibbon which did not sing while surveying remain undetected The forest of the surveyed area is broadleaf evergreen forest (rich forest, medium forest) which is the best habitat for gibbons The medium to high montane broadleaf evergreen forest contributes the highest percentage of the detected gibbon in the Khe Nuoc Trong proposed nature reserve

TABLE OF CONTENTS

ABSTRACT i

LIST OF FIGURE iv

LIST OF TABLE v

1 INTRODUCTION 1

1.1 Background 1

1.2 Goal 4

1.3 Specific objectives 4

1.4 Research questions 4

2 LITERATURE REVIEW 5

2.1 Occurrence and distribution 5

2.2 Gibbon in Vietnam 5

2.3 Distribution of Southern white-cheeked gibbon in Vietnam 7

2.4 Vocalization of gibbons 8

2.5 Calling time of gibbon 12

2.6 Factors effecting gibbon call 12

2.7 Using of automatic recorder 13

3 METHODS 16

3.1 Study area description 16

3.1.1 Location 16

3.1.2 Topography 16

3.1.3 Flora and Fauna 16

3.1.4 Population 17

3.2 Field survey 18

3.3 Data analysis and vocalization analysis Method 21

3.3.1 Sound analysis 21

3.3.2 Effect of weather on the singing behavior of gibbons 21

3.4 Vocalization analysis 22

3.4.1 Duet calls songs of Southern white-cheeked gibbon: 22

3.4.2 Female song of Southern white-cheeked gibbon 23

3.4.3 Male song of southern-white cheeked gibbon 23

4 RESULT 25

4.1 Gibbon calls detected 25

4.2 Distribution of gibbons in Khe Nuoc Trong proposed nature reserve 26

4.3 Structure of gibbon group 30

5 DISCUSSION 35

5.1 Occurrence and distribution of white-cheeked crested gibbon in KNT proposed nature reserve 35

6 CONCLUSION 39

ACKNOWLEDGEMENT 40

REFERENCE 41

APPENDICES 46

LIST OF FIGURE

Figure 1.1 Spectrograms of the six crested gibbon species of Vietnam 10 Figure.3.1 Flora images of the KNT proposed nature reserve 17 Figure.3.2 Elevation map of Khe Nuoc Trong proposed Nature Reserve, Quang Binh Province, Vietnam in May 2018 18 Figure.3.3 Elevation map of recording posts in the KNT proposed nature reserve in May 2017 19 Figure 3.4 SM3 (Wildlife Acoustic) automatic recorders was attached to the trees

in Khe Nuoc Trong proposed nature reserve 19 Figure.3.5 Duet call of the Sothern white-cheeked gibbon, great call is female call and short call is male call 23 Figure 4.1 Frequency distribution of the start time of gibbon song bouts 25 Figure.4.2 Percentage graph of the rain and no rainy days in the survey 26 Figure.4.3 Elevation and comparative map of the gibbon with forest type in Khe Nuoc Trong nature reserve in May 2018 27 Figure.4.4 Elevation and forest type map of the surveyed area with gibbons in the Khe Nuoc Trong nature reserve in May 2018 28 Figure.4.5 Percentage graph of the surveyed area forest 29 Figure.4.6 Elevation and distribution map of gibbon in the KNT proposed nature reserve in May 2018 29 Figure.4.7 Duet call of the southern white-cheeked gibbon from recording site at 5:55am 37 Figure.4.8.Female sonogram of Southern white-cheeked gibbon from recording site

at 4:47am 33 Figure.4.9 Male sonogram of the southern white-cheeked gibbon from recording site at 5:35am 34

LIST OF TABLE

Table.1.1 List of Vietnamese gibbons 7

Table.3.1 Area of each sub-zone in the study area 20

Table.4.1 Percentage graph of the gibbon call detected in each compartment 30

Table.4.2 Structure of the gibbon calls with recorded time 31

1 INTRODUCTION

1.1 Background

Geographically, Vietnam lies in the southeast Asia with an area of 331.689 km2and extends up to 1,600 km from north to south, along the eastern coast of Indochina from 230 N to 8030' N Altitude varies from the sea level to 3,143 m at the summit of mount Phan Si pan in the extreme north-west of the country (Geissmann, 2000b) Vietnam is rich in the biodiversity and many new species have been discovered by researchers recently Many endangered species are present in Vietnam and their populations are decreasing because of deforestation, hunting, and land encroachment (Geissmann, 2000a) In the north of Vietnam most of the primates which are endangered and critically endangered are found Among these one of the endangered primate is gibbon species which is only found in the north-central of Vietnam in six provinces (Harding, 2012)

Gibbon belongs to the family Hylobatidae and is among the most endangered Vietnamese primates Gibbon populations is expected to become extinct faster than other primates (Geissmann et al., 2001) Gibbons are arboreal apes living in the tropical rain forests of south-east Asia (Geissmann, 2000a) The impact of hunting

on gibbons in the recent past is undeniable and illustrated by the occurrence of suppressed populations within areas of largely intact habitat, suggesting hunting is a primary threat In some areas of central Vietnam there is no sufficient data available about gibbon species (Nadler et al., 2007a)

Monitoring programmes are the main source of information on the population status

of species of conservation concern, and have a significant role in setting

conservation action priorities The use of monitoring in conservation is important

for gibbon because gibbons are facing many problems so to evaluate the species problem and to estimate total population size of gibbons we need to do monitoring for gibbons By monitoring we can also identify the change in population size in a region and also helps to find the threats to the species Knowledge of population size is important in research which aimed at understanding interactions between species Monitoring programmes are not only designed to measure change in

abundance but also used to estimate density, and by extension population size, if certain assumptions are upheld (Stokes et al., 2010)

Monitoring a species may face a lot of difficulties Gibbon is difficult to monitor by the traditional method and lead to a range limitation and often does not give accurate information in the survey Therefore, by not having accurate information resulting in numerous difficulties for conservation (Vinh et al., 2010) The management of the gibbon species requires detailed information on the distribution and abundance which is often hard to obtain

Gibbons live in the upper canopy of the forest and are closely associated with closed canopy of broadleaf evergreen forests Gibbons are strongly arboreal and sometimes come to the ground (Sarma et al., 2013) Gibbon moves very quick in the upper canopy of the trees and people rarely see them in the forest Therefore, visual detections of gibbon are very difficult in the field, especially for short survey (Vinh

et al., 2010) Human interactions with primates may influence their behavior and using automatic recorders in the field has no direct contact with human observers thus reducing any possible impact of humans on their singing behavior (Muñoz-Delgado et al., 2004)

Using automatic sound recorder is an effective survey method for many reasons First reason is that it is suitable for surveying animals which do not sing daily Second reason is less human disturbance in the survey by using automatic recorder than traditional survey Third reason is it‟s beneficial in the difficult areas in the survey and no specialists are needed in the field survey (Thinh et al., 2017) Data could be analyzed easily from the recordings samples and each vocalization clip was then transferred to RAVEN PRO 1.4 sound analysis software (Cornell Lab of Ornithology Bioacoustics Research Program) to create spectrograms and measure attributes of the different vocal types (Cäsar et al., 2012)

Many researchers suggested automatic recording is the best way to collect the data

on animals in the field Automated acoustic monitoring can be used for studying various taxonomic groups of sound-producing animals and allows the detection of elusive, rare, and threatened species (Collen et al., 2009; Burbidge, A 2017)

Gibbons produce a loud and elaborate sound and can be detected by their loud and long song bout (Geissmann and Orgeldinger 2000a; Brockelman, 1987) From their loud sounds we can detect them easily by using automatic recorders Gibbon do not call every day therefore, using autonomous recorders in the survey are widely used and can provide a highly efficient method of species monitoring Automated examination of acoustic recording with a field listening survey, using commercially available autonomous recorder provide a highly efficient method of species monitoring and custom call detection and classification software This method relies

on loud sounds and is used by many researchers in their studies

Currently, in Vietnam there are many national parks and nature reserves with highly diversed gibbon species Khe Nuoc Trong proposed nature reserve is one of the most important areas on the basis of biodiversity and conservation which consists of

a 20,000-ha tract of Annamese Lowland Forest in Quang Binh Province, central Vietnam It is a part of a larger (82,700 ha) tract that extends into the neighboring Quang Tri Province It is estimated that the protection of the forest at Khe Nuoc Trong will sequester approximately 50,000 tonnes of carbon dioxide (CO2) per year for the next 30 years (Viet nature, 2018)

north-In Khe Nuoc Trong proposed nature reserve most of the species is endangered, due

to degradation or loss of their habitats In the nature reserve many endangered and critically endangered species are present (Merriman et al.,unpulished) Khe Nuoc Trong proposed nature reserve is home to the Critically Endangered Sunda Pangolin

(Manisjavanica) as well as three species of Endangered primates, the Red-shanked Douc (Pygathrixnemaeus), the Southern White-cheeked Gibbon (Nomascus siki) and the Haitinh Langur (Trachypithecus haitinhenis) (Viet nature, 2018)

In Khe Nuoc Trong proposed nature reserve, there are many studies conducted on birds and the flora and fauna of the nature reserve such as (Tuong et al., 2018);(Viet nature, 2018);(Merriman et al., unpublished) etc and on southern white-cheeked gibbon rare studies are conducted in the proposed nature reserve Khe Nuoc Trong proposed nature reserve is facing many difficulties in monitoring its diversity because it covers a large area and contains a variety of high mountains Surveying

of gibbon species by the traditional method may lead to a range of limitation and

often do not give accurate information Besides, the investors will have trouble if they research in bad weather Therefore, to have more accurate information in terms

of occurrence and distribution of gibbon in Khe Nuoc Trong proposed nature reserve; I decided to conduct this study:

„„Identifying the occurrence and distribution of southern white cheeked-gibbon

(Nomascus siki) using autonomous recorders: A case study in the Khe Nuoc Trong

proposed nature reserve‟‟

1.2 Goal

To add information on the occurrence and distribution of southern white-cheeked

gibbon (Nomascus siki) in Khe Nuoc Trong proposed nature reserve by using autonomous recorders

1.3 Specific objectives

The specific objectives of this study are:

 To determine the occurrence of southern white-cheeked gibbon in Khe Nuoc Trong proposed nature reserve by using autonomous recorders

 To determine the distribution of southern white-cheeked gibbon in Khe Nuoc Trong proposed nature reserve by using autonomous recorders

1.4 Research questions

 Where is the occurrence of southern white-cheeked gibbon in the study site?

 Where is the distribution of southern white-cheeked gibbon in the study site?

2 LITERATURE REVIEW

Gibbon is declining and facing many threats because of human activities in their habitat Therefore, it‟s important to conserve and monitor them for conservation For monitoring it‟s important to choose a best method that the habitat of the gibbon could not be disturbed In this study I will find out the occurrence and distribution

of Southern white-cheeked gibbons (Nomascus siki) by using automatic recorders

Using autonomous recorder is the best method for gibbon surveying to know about the gibbon population, problems, to conserve them and using of that method can be efficient for gibbon survey Therefore, previous related studies, available literature, documents and related materials have been reviewed

2.1 Occurrence and distribution

According to Friedmann & Regenbrecht, (2001) occurrence is defined as a subjective phenomenon that results from experiences induced by immersive virtual reality technology “Occurrence is a dimensional construct and describes the extent

to which a user feels present in an environment” Postulate that occurrence is the result of successful suppression of the mismatch between the predicted and the actual interceptive state (Suzuki and Critchley, 2012)

Krebs, (1972) point out that understanding pattern of distribution and abundance of organisms are basic concerns in ecology and wildlife conservation Distribution is the way in which something is shared out among a group or spread over an area Distributional patterns change over time and are influenced by both changes in overall abundance and the movement of individuals among habitats Distributional Consistency incorporates both spatial and temporal components, allowing researchers to quantify how consistently a population within a given region is distributed among an array of sites, while controlling for changes in regional abundance (Heath et al., 2012)

2.2 Gibbon in Vietnam

Some gibbon species is in Vietnam, Laos and all species of gibbon in Vietnam are

crested gibbons of the genus Nomascus These species are distributed in Vietnam

from the northern-most extent of the country almost to Ho Chi Minh City in the

south and all are listed as either endangered or critically endangered in the IUCN red list of threatened species (IUCN 2017) All Gibbons of Vietnam belong to a rare primate in Vietnam red list book (2007) by decision of 32 (2006)/ ND – CP, Appendix IB, Decision of 170 (2013), Group I2, Appendix I, CITIES (2013) named

in rare primates of Vietnam in Decision of 538 (QD – TTg on May 10th 2017) by Prime Minister about urgent activities to conserve Primate of Vietnam up to 2025, vision 2030 The gibbons (family Hylobatidae), finally, are not monkeys but apes (Hominoidea), and are more closely related to humans than to monkeys Gibbons are the only apes known to live on the Asian mainland Like all members of the apes, gibbons have no external tail (Geissmann et al., 2013)

Eames and Robson, (1993) stated that presence of primates in an area was usually first determined by conducting interviews with local people There is no population estimation available for Vietnam and even the distribution in Vietnam is not very clear yet The southernmost area with confirmed records is Quang Binh Province (Nhat, 2002) The most important populations are in Phong Nha-Ke Bang and Pu Mat National Parks (Ha et al., 2005) In Pu Mat National Park the population is believed to have declined in six years (1999 to 2004) about 40% A relatively large population exists in Dak Rong Nature Reserve, Quang Tri Province, with a minimum of 25 groups (Ha, 2004) But the systematic status of gibbons in this area still requires clarification (Nadler et al., 2007a) In Vietnam different species of gibbon are found in various places Gibbons are living in groups and each group consists of one male, females and young The group is leading by male which protect the group (Fuentes 2000)

Following (Rawson et al., 2011) classification report on Primate Conservation Status review a new Vietnam gibbon‟s taxonomy was introduced Current

taxonomic understanding identifies seven species of Nomascus, all distributed east

of the Mekong River in Cambodia, Laos, Vietnam and China; six of which are found in Vietnam This document has been the most detail one on Vietnam gibbons

up to present Therefore, we use scientific name and English name of Vietnam gibbons following (Rawson et al., 2011) The names of the six sub-species of gibbons in Vietnam are listed in (Table1.1)

Table.1.1 List of Vietnamese gibbons

1 Nomascus concolor Western black gibbon

2 Nomascus nasutus Eastern black gibbon

3 Nomascus leucogenys Northern white-cheeked gibbon

4 Nomascus siki Southern white-cheeked gibbon

5 Nomascus annamensis Northern yellow-cheeked gibbon

6 Nomascus gabriellae Southern yellow-cheeked gibbon

(Source: Rawson et al., 2011)

2.3 Distribution of Southern white-cheeked gibbon in Vietnam

Southern white-cheeked gibbon is distributed in southern Lao PDR and central Vietnam east of the Mekong River, and is found between 17° N and about 19.3° N (Geissmann, 2000b) In Vietnam, the most important populations of southern white-cheeked gibbons are in Phong Nha-Ke Bang National Park and Pu

north-Mat National Park (Ha et al., 2005) Nomascus siki was previously thought to occur

in five provinces of north-central Vietnam, with northern and southern distribution limits thought to be the Ca River (Nghe An Province) and Bach Ma National Park

(Thua Thien Hue Province) respectively The northern distributional limit of N siki

in Vietnam is unknown It is possible that some gibbons still occur in these nature

reserves and remained undetected, but the possibility is low (Rawson et al., 2007)

The Southern white-cheeked gibbon is one of the most endangered primates in Vietnam (Ha, 2007) Habitat loss, degradation and fragmentation caused by deforestation and the expansion of rubber plantations between the 1960s and the 1980s have been reported as the main causes of the decline in the gibbon population

in the area (Fei and Luo, 2014)

By Thinh et al., (2010) the southern white-cheeked gibbon, N siki, is now

understood to have a much more restricted range There is insufficient data to

quantitatively assess trends for N siki however all indications are that hunting being

a principle threat and there has been an on-going population decline for the species

in Vietnam Fortunately there remain relatively large populations in tracts of forest mainly centered in Quang Binh Province and along the border with Laos In

Vietnam, N siki occurs in parts of Ha Tinh, Quang Binh and Quang Tri Provinces

Populations have been documented in six sites: two protected areas (Phong Nha-Ke Bang National Park, Bac Huong Hoa Nature Reserve), two proposed nature reserves (Khe Ve, Giang Man) and two stateforest enterprises (Truong Son, Khe Giua)

Considerable new information on the taxonomy, distribution and ecology of N siki

in Vietnam has been documented since the first Vietnam status review of gibbons (Geissmann, 2000b)

The southern white-cheeked gibbon has recently been recognized as a distinct species; previously it was believed to be a subspecies of the northern white-

cheeked gibbon (Nomascus leucogenys) (Geissmann, 2000a) The white-cheeked

gibbon is identified as one of the most endangered primates in Vietnam (Geissmann

et al., 2003; Nadler et al., 2007b) Khe Nuoc Trong proposed nature reserve forest is one of the most important natural habitats for gibbon With special geographical conditions, it is one of the rare locations in Vietnam which have original biodiversity (Ha, 2007) In the NR hunting is widespread and heavy, and there is no recent record present on the occurrence and distribution of southern white-cheeked gibbon In Vietnam red list book, the status of this species critically endangered and globally endangered in the IUCN red list (Geissmann et al., 2001)

2.4 Vocalization of gibbons

Gibbons produce a loud and elaborate sound from which we can detect them (Geissmann, 1993) Therefore, using automatic recorders to record the calling of gibbons is the best way for monitoring gibbons This method relies on the loud sound of the gibbons calls (Brockelman, 1987; Brockelman and Srikosamatara, 1993) In wildlife survey people use different automatic acoustic recorders to record the sounds of animals because some animals do not produce sounds all the day

First, survey done in the study site and then automatic acoustic recorders are set up

in the field The recorder is attaching with trees or hangs in the trees canopy to record the sounds of the animals The automatic acoustic recorders offer a cost-effective way of increasing the reliability of surveys when compared with current approaches (Zwart et al., 2014)

Geissmann and Orgeldinger, (2000b) find out that if duetting fulfills a particular function in some gibbon species, but not in others, then a correlation between duetting and that particular function may disappear in a cross-species comparison

As a result, a cross-species comparison may not produce correlations for existing functions of duetting, even if the comparison were restricted to duetting species only Gibbon is stable and stereotyped (Jiang et al., 2017) The short phrases of males (and females) are usually less stereotypic than the great-calls Whereas female great-calls remain essentially unchanged throughout a song bout; successive short phrases are rarely identical Instead, males gradually build up their phrases during a song bout, beginning with single, simple notes Interestingly, gibbons appear to recognize, and may respond to, great-calls of other species All gibbon

species use a variety of different note types Crested gibbons (genus Nomascus),

represent the other extreme of the spectrum: they show the highest degree of specificity in their note type repertoire (Figure1.1) (Geissmann, 2002)

sex-Figure.1.1 Male and female spectrograms of the six crested gibbon species

of Vietnam (Hallam et al 2011)

The song structure of Nomascus spp differs in several respects from that of other gibbons ( Geissmann and Orgeldinger, 2000a) Adult female Nomascus produce

only great call phrases or fragments of great call phrases Fully developed song

phrases of adult male Nomascus typically consist of 3 different note types Crested gibbons (Nomascus) are 1 of 4 main taxonomic groups within the Hylobatidae and

are characterized by various morphologic, anatomic, karyologic, and vocal features (Konrad and Geissmann, 2004) Crested gibbons exhibit several unique characteristics in singing behavior that set them apart from other gibbons (Geissmann, 2000a)

According to Nhat, (2002) acoustic analysis could confirm the concordance between song structure and phylogenetic relationships as obtained from mitochondrial sequence data Significant differences are found between the songs of

N nasutus, N concolor and the four southern species, N leucogenys, N siki, N annamensis and N gabriellae Although latter four species revealed only subtle

differences in their songs, a detailed acoustic analysis was able to discriminate significantly between them This relation was positively correlated with their genetic relatedness found by recent molecular studies In many cases, species could

be distinguished by directly inspecting the spectrograms of their songs (Wich et al., 2002) From acoustic result we cannot conclude the origin of crested gibbons, because we have no evidence, which song structure is most ancestral However, we found the largest differences in the song structure between the most northern and most southern species with successive gradation between them (Hallam et al., 2007)

Wich et al., (2002) concluded that gibbon song vocalizations can be used as a vocal fingerprint to identify individuals Whether gibbons use these differences to discriminate among songs of neighbors and non-neighbors have not yet been demonstrated In contrast to the high classification accuracy among individuals in the same site, we found lower classification accuracy when data from both sites were combined Of course, more mistakes can be expected when the number of individuals increases (Sun et al., 2007) Based on the song structure of gibbons, we used vocalization software to analyze and determine the total number of individuals

of gibbons per group The sonograms could help to distinguish between the songs of male, female or juvenile white-cheeked crested gibbons (Thinh et al., 2010)

2.5 Calling time of gibbon

Gibbons often sing from their sleeping tree and frequently do not move far in the early morning during the singing bouts (Leighton, 1987; Cheyne, 2008) Preferentially, song bouts are uttered in the early morning and last for about 10.30 minutes (Geissmann et al., 2007) Gibbons usually begin to sing before dawn, i.e before the sun is visible in the sky (Tenaza, 1976) Gibbon neighboring groups interact when singing Groups have their territory boundaries and when one group started calling, the other, if present in the same general area, would start singing soon and both groups approached each other while continuing to sing (Xue-feng & Wu-jing, 2008)

In most gibbon species, mated individuals combine their respective songs in the form of well-coordinated duets (Konrad and Geissmann, 2004) Gibbons sing a maximum of 1 hour before sun rise (Cheyne et al., 2008) and is the only dominant activity during the dark hours Fan et al., (2009) point out that gibbon song bout last 7–15 min As gibbon groups do not call daily, several survey days are needed to detect almost all gibbon groups (Brockelman, 1987; Jiang et al., 2006)

2.6 Factors effecting gibbon call

Studying the structure and patterns of vocalizations in primates is important for our understanding of their evolution, responses to changing environments, and social behavior, as well as for the design of vocalization-based surveys Primates are known to alter their behavior in relation to environmental factors such as light intensity, temperature, humidity, precipitation, day/night length, moon phase, and weather induced by the geophysical solar and earth system (Erkert, 2003) Gibbons sing everyday throughout the year, although the time gibbons sing varies depending

on the season (Geissmann et al., 2003) Gibbons sing more during periods of high fruit availability (Bartlett, 2015) Gibbons sang less on rainy days, consistent with our original hypothesis and with previous studies Rainfall has been identified most often as a factor inhibiting singing events in gibbons (Dooley et al., 2013) The

frequency of singing varies between populations and may be influenced by variables, such as population density, availability of resources, sexual activity and weather Yet, roughly, most gibbon species duet almost daily (Geissmann et al.,

2005) Groups of Hylobates significantly reduced calling when disturbed by

loggers In Indonesia, where air quality varies seasonally with forest fires,

Hylobates albibarbis reduced singing during months of intense atmospheric smoke (Harrison et al., 2007)

Johnson et al., (2003) find out that gibbons were only heard singing in the absence

of rain, when cloud cover was generally less than 50% and generally when it was not windy Although it is accepted that gibbon singing can be affected by many factors, including weather, human disturbance and topography of the area few data are available on the effects of astronomical cues and/or the intensity of light or air quality at the time of singing (Cheyne 2008; Cheyne et al., 2008)

Delgado et al., (2004) highlight the possibility that human interactions with primates may influence their behavior All gibbons in his study were wild and had

no direct contact with human observers thus reducing any possible impact of humans on their singing behavior They found no decrease in the rate of observer directed threats over the course of his study Surprisingly, one group significantly increased their rate of observer-directed threats during the study period These threats were not accompanied by any other signs of fear and were often performed

at distances of 1 to 2 m from the observer (Crofoot et al., 2010)

2.7 Using of automatic recorder

Gibbons usually live in the upper forest canopy and are sensitive to human presence Therefore, seeing gibbons is difficult in the field, especially during short surveys Gibbons can be detected by their loud and long song bout (Geissmann and Orgeldinger, 2000a; Brockelman, 1987) Movement in the canopy requires vision, whereas singing does not Human interactions with primates may influence their behavior and using automatic recorders in the field has no direct contact with human observers thus reducing any possible impact of humans on their singing behavior (Muñoz-Delgado et al., 2004)

Gibbons are shy and difficult to approach in the wild, the recording method using the directional microphone was the most efficient method in the field Recording gibbon song is also useful for identifying the number of populations in research areas without scaring the gibbons By recording, individual groups can be studied more easily without many of the difficulties involved in observing primates in tall tropical rain forest (Geissmann et al., 2003) Vocal recognition might be a more suitable tool for surveying and studying this critically endangered primate than some traditional methods Using vocalization survey data techniques a population of

Northern white cheeked gibbons Nomascus leucogenys has recently been recorded

in Vietnam (Vinh et al., 2010)

Hutto & Stutzman, (2009); Crofoot et al., (2010) suggested autonomous acoustic recorders offer an alternative not subject taxonomy of the biases of field surveys

By not requiring the presence of an observer, they can be less expensive, can be used in inaccessible habitats, avoid disturbance and minimize temporal bias through extended sampling Acoustic recorders can provide data easily and more rapidly than human-based survey techniques In comparison of the detection of nightjars via traditional human surveys with detection by bioacoustics recorders using the data recorded at the same time as the human surveys He found that the use of automated bioacoustics recorders to replace human wildlife Surveys is the best way to collect data in the field (Zwart et al., 2014)

Although songs of wild gibbon can often be heard over distances well exceeding 1

km, several factors confuse estimates of even skilled researchers All-occurrence sampling was used as it was the most effective way to look at the possible vocal function and behavioral responses of the whole family group, rather than just focusing on one individual (Brockelman, 1987; Brockelman and Srikosamatara, 1993) The infrequency of vocalizations meant it was possible to accurately monitor more than just the one individual at a time for the target behavior Vocalization studies were conducted only in a captive setting as in the wild vocalizations were rarely heard because of the distance between primates and observers (Agmen, 2014)

Many researchers suggested automatic recording is the best way to collect the data about animals in the field In an effort to establish a remote monitoring system for sound-producing animals species using autonomous recording units to collect soundscapes In comparison with classical survey methods, automated acoustic surveillance has many advantages For example, automated acoustic monitoring can

be used for studying various taxonomic groups of sound-producing animals and allows the detection of elusive, rare, and threatened species (Collen et al., 2009; Burbidge, A 2017)

It supports presence absence surveys as well as the estimation of species richness, population density and is suitable for virtually all terrestrial and aquatic habitats, including vast and remote areas and rough terrain (Jahn et al., 2017) The fixed-point method is the most suitable to estimate densities of gibbon populations In this method the loud morning songs of the gibbons are monitored from fixed listening points (Brockelman 1987; Brockelman and Srikosamatara, 1993)

to 106°48‟27‟‟E It is part of a larger (82,700 ha) tract that extends into the neighboring Quang Tri Province (Tuong et al., 2018)

3.1.2 Topography

The topography of the area is dominated by the foothills of the Annamite Mountains but rises in the west to elevations of over 1,000 m asl The site covers 19,898 ha area and land cover is predominantly forest consisting of a wide range of tree species characteristic of lowland evergreen-broadleaf forest To the south the site is contiguous with Bac Huong Hoa Nature Reserve in Quang Tri province (Tuong et al., 2018)

3.1.3 Flora and Fauna

Khe Nuoc Trong proposed nature reserve forms the southern part of the Truong Son Important Bird Area (IBA), which was identified by Birdlife International Vietnam Programme (2002) This IBA and the forests of Khe Nuoc Trong support a typical flora and fauna of the Annamese Lowlands Endemic Bird Area (EBA) The site is qualified as

an IBA under criteria A1, due to the presence of globally threatened species (at the time

of assessment, a number of species found in the nature reserve, such as Edwards‟s Pheasant and Crested Argus, were considered globally threatened and near-threatened species, respectively); A2, due to the occurrence of restricted-range species; and A3, due

to the occurrence of biome restricted species (Tuong et al., 2018) Land cover is predominantly forest consisting of a wide range of tree species including lowland evergreen-broadleaf forest and having complex biodiversity in the forest which is the home of many endemic species (Merriman et al., unpublished) This is the area of the forest itself the lowest continuous land in Vietnam It also preserves a large area of

primary evergreen moist tropical evergreen forest lowland (Tuong et al., 2018) The occurrence of white-cheeked gibbons in the area has been documented for many years recently However, the status of species in the area has not yet been studied full Some pictures of the surveyed area flora are given below figure (3.1)

Figure.3.1 Flora images of the KNT proposed nature reserve (Field data)

3.1.4 Population

The villages in the vicinity of Khe Nuoc Trong proposed nature reserve is Ho Rum (68 households, approx 300 people), Ha Lec (78 households, 346 people) and An Bai (38 households, 171 people) However, many „outsiders‟ come to take resources from the forest, including people from outside provinces and across the Laos border (Merriman et al., unpulished)

Figure.3.2 Elevation map of Khe Nuoc Trong proposed nature reserve, Quang Binh

Province, Vietnam in May 2018 (Abu Bakar)

3.2 Field survey

All gibbon species utter loud, long and well patterned song bouts from that loud songs we can detect them Gibbons usually sings in the early morning before dawn and after dawn (Geissmann 1993; Vinh et al., 2010) Durations of their singing varies from 15-30 minutes and, depends on species and environmental conditions, can be easily hear over long distance (i.e >2km) (Geissmann et al., 2005; Konrad and Geissmann, 2004) Therefore, using of automatic recorder in the field to detect the sounds of gibbons is the best and easy way to approach gibbon The fixed point count method was used in the field survey, this method relies on the loud sound of the gibbons calls whereby the loud morning songs of the gibbons are monitored from fixed recording posts (Geissmann 1993; Geissmann and Orgeldinger, 2000b) Field survey was conducted in April and May in 2017 from 14th April to 11th May for 27th day and 53 recording posts were set up in the field (Figure 3.3)

Figure.3.3 Elevation map of recording posts in the Khe Nuoc Trong proposed

nature reserve in May 2017 (Abu Bakar) The posts were setup apart 500m to 1000m depended upon the terrain of the study area The recording posts were setup in the west and near the central part of the proposed nature reserve Automatic recorders SM3 (Wildlife Acoustic) were setup

in the survey area and attached with trees to record gibbon call (figure.3.4) The coordinates of the recording posts were marked and saved by GPS

Figure.3.4 SM3 (Wildlife Acoustic) automatic recorders was attached to the trees

in Khe Nuoc Trong proposed nature reserve (Field data)

In the Khe Nuoc Trong proposed nature reserve eight compartments were selected (515, 516, 517, 532, 533, 534, 535 and 536) for surveying The total area of these compartments is 5,3532 hectares The area of each habitat of the Khe Nuoc Trong proposed nature reserve which is selected for survey is shown in the (Table.3.1)

Table.3.1 Area of each sub-zone in the study area

area

Source: Research data 2017

In 53 recording posts, 16 recording posts were surveyed for one day, 12 were surveyed for 2 days and 25 were surveyed for three days Picture and other activities were minimized from which disturbance occur The set up date and collection date

of the each recording post is shown in (Appendix 1)

The most important thing in the gibbon field survey is to choose a perfect timing of season because weather has a great effect on the vocal activity of gibbons Dry season is the best time for carrying out survey in the field because vocalization is frequent in dry season (Rawson, 2004) The gibbon‟s vocalization is effected by many factors (Cheyne et al., 2008), gibbons do not call when the weather is not good or when it‟s raining, and therefore perfect time of gibbon call is good weather because gibbon calls every morning when weather is good or when it‟s sunny (Brockelman, 1987) recommended that to reliably estimate density, that survey periods should be adequate to provide 90% probability of detection The weather of each day and forest type was recorded for every recording post see (Appendix2) Each sample site was sampled for 1-3 consecutive days and the sounds were recorded for all the day Before changing the sample sites, the recordings were saved in a disk and compressed at interval with a half hour The memory disk and batteries were changed with the changing of the sample sites

3.3 Data analysis and vocalization analysis Method

3.3.1 Sound analysis

The recording data was analyzed by RAVEN PRO 1.5 software (Cornell Lab of

Ornithology) to detect the spectrogram of the gibbon Four hour time interval was chosen from the all-day recording from 5am to 9am and was analyzed because it is suitable time for the behavior of gibbons singing

When gibbon call was detected from the recording site the coordinates and the starting and ending time of the gibbon call was checked to identify that how many groups and individuals are present in that area For the identification of the group from calls of gibbon when detected form the recording the starting time of the groups were checked and coordinates Two groups were identified from the starting and ending time of calling If the starting and ending time was different between the gibbon groups so it were consider two groups, and if the calling time was recorded same with the other recording post the group was considered one

The Group identification is very crucial in this survey The duets given out simultaneously or partially overlapping in time are easily distinguished as different groups The groups were identified by time and coordinates After detecting a call the day of singing was also noticed and the time of gibbon calls in a day, that how much time a pair utters a song in day Songs from each recording posts were noticed with recorder Id, coordinates and compartments (Appendix 4)

3.3.2 Effect of weather on the singing behavior of gibbons

Weather has also great effect on the singing behavior of the gibbons The survey was carried out in the April and May in 2017 In the April month weather was good only two days rain occur during this month from 14th April to 30th In April most of the days during survey were sunny and was good time for gibbon to sing Gibbon sings mostly on sunny day In May rain was noticed most of the days because it is a monsoon month in Vietnam and rain has great effect on the gibbon singing Gibbon does not sing when it rain or sing less on the rainy day

3.4 Vocalization analysis

Gibbons live in the upper canopy and difficult to observe them directly However, crested gibbons are well known to produce elaborate, loud, long sounds (Geissmann, 1993) The song structure of the gibbons is different between the species Based on the song structure of gibbons, we can determine the total number

of individual‟s male, female and sub-adults of gibbons in each group by using vocalization software to analyze gibbons groups The songs of male, female or juvenile of white-cheeked crested gibbon having different structures of sonogram,

we can distinguish between them with the help of sonogram The sonograms could help to distinguish between the songs of male, female or juvenile of white-cheeked crested gibbons Based on acoustic analysis we used RAVEN PRO 1.5 Software (Cornell Laboratory of Ornithology) to generate spectrograms, count individuals and to calculate minimum population density Crested gibbon songs consist of phrases from both sexes The adult female produce a great call which is the most stereotyped and most easily identifiable phrase of the gibbon song and consist of three phrases The male multi-modulated call is a series of notes, consists of more than one note type, uttered in succession and so related as to form a recognizable sequence or pattern in time and consist of three phrases When both sexes produce their loud sound call duet song and exhibit vigorous movement in an interactive manner (i.e performing a mutually cooperative and coordinated display) Young gibbons normally give calls with their parents at the same time and thus can easily

be counted in sonograms (Thinh et al., 2010) By the specific song structure of the sexes we can easily identified them from their different sonogram

3.4.1 Duet calls songs of Southern white-cheeked gibbon:

Every species of gibbons have different song structure from which we can differentiate between the species and sex of the gibbon The different song structure

of Nomascus is described in many previous studies (Geissmann, 2002) Southern

white-cheeked gibbon has different song structure from other gibbon‟s specie Duet call is the songs of mated pair, male and female gibbon when they start singing together In duet call of the Southern white-cheeked gibbon, male start singing first, then female combines her song with male song After, combining her song with the

male song, male pause his singing and then add a multi-modulated phrase with the end of the female great call (Figure 3.5) The difference between the sonogram of the male and female can be easily identified which is discussed below:

Figure.3.5 Duet call of the Sothern white-cheeked gibbon, great call is female call and short call is male call (Hallam et al., 2011)

3.4.2 Female song of Southern white-cheeked gibbon

Adult female of Southern white-cheeked gibbon only produce great call The female great call consists of three phrases: The first note a great call phrase of an adult

female The great call begins with oo notes (fa), followed by bark notes (fb) and ends with twitter notes (fc) (Geissmann, 2000a; Konrad and Geissmann, 2004) Female great calls begin with long notes of slowly increasing frequency (termed fa- notes) These notes are also called oo notes; this name is used in the present study

In the course of a great call, note durations and interval durations become

continuously shorter and oo notes gradually change to short notes of steeply increasing frequency (fb) These notes are referred to as bark notes After the climax

of the acceleration, bark notes tail off into a twitter (fc), The great call is female call

phrase (Konrad and Geissmann, 2004) (Figure 3.5)

3.4.3 Male song of southern-white cheeked gibbon

Male song of the Southern-white cheeked gibbon produce developed song The

fully developed song of adult male of Nomascus siki phrases typically consist of three different note types (i.e ma, mb, mc) The male song is starting from the boom

note which is a very deep note of constant frequency and usually produced as single notes, unlike other male notes, which usually occur in short series (phrases) The

boom note (ma) is short and then modulated to staccato note (mb), relatively

Time

continue repeated sounds, which are therefore also referred to as staccato notes The

most conspicuous part of the male song is the multi-modulated phrase (mc) This

phrase consists of several notes, which exhibit rapid and steep frequency modulations Adult males typically utter a multi-modulated phrase (coda) at or shortly after the climax of the female great call phrase In the course of a complete song bout, the male phrases are gradually built up At the beginning of the song bout, the male produces long hardly modulated notes that are precursors of the multi-modulated phrases Later in the song bout the phrases become more and more modulated and boom notes and staccato phrases are added (Konrad and Geissmann,

2004) (Figure 3.5)

4 RESULTS

4.1 Gibbon calls detected

From 53 recording posts 26 calls of southern white-cheeked gibbon were detected from recordings in the proposed nature reserve In 53 recording posts, no gibbon calls were detected from 7 recorders and from 45 recording posts gibbon songs were detected The earliest song bout was started at 5:15am and the late song was started

at 7:45 am No gibbon call was heard before 5:15am and after 7:45am The highest percentage of song bouts time in the morning is from 5:30am to 6:30am with the percentage of 70% calls detected in this time (Figure 4.1)

Figure.4.1 Frequency distribution of the start time of gibbon song bouts From all the recordings 26 southern white-cheeked gibbon calls were detected

Gibbons Nomascus siki is distributed to the western and near to the central part of

the Khe Nuoc Trong proposed nature reserve Other parts of the proposed nature reserve were not surveyed because of the limited time of the survey Gibbons were heard from the most of the recording posts most of the gibbon groups were heard from more than one posts because of the overlapping areas and of the short distance between the recording posts The highest number of gibbon groups counted from one recording post was two groups

Weather has also an effect on the singing behavior of gibbons From the 27 days of recordings survey, 8 days were recorded with rain and 19 days were recorded with

0 10 20 30 40 50 60

no rain Rain has a great effect on the gibbon calling when it rain gibbon does not sing or sing less than sunny days Gibbon statrted singing later than no rainy days and only three to four duet calls were heard on the rainy day and lasts for 4-7 minutes In the 8 days of the rain while analyzing data gibbon was detected in the rainy day In the 8 days of the rain only 3 days gibbon calls were detected and 5 days of the rain no gibbon call was detected In the survey time of 27 days, 30% was rainy days and 70% with no rainy days From this we know that the rain has effect on the gibbon calling The percentage graph of the survey period with rainy days and the days withoutrain (Figure 4.2)

Figure.4.2 Percentage graph of the rain and no rainy days in the survey

4.2 Distribution of gibbons in Khe Nuoc Trong proposed nature reserve

We found that Nomascus siki is present in the Khe Nuoc Trong proposed nature

reserve and having a high population of gibbons as compared to the other areas of the north central Vietnam In the nature reserve gibbon occurred in those areas of the forest which is not disturbed or less disturbed and where human interaction is not more The forest where gibbon is present in the proposed nature reserve is broadleaf evergreen forest and it is the suitable habitat for gibbons living because gibbon lives in the upper canopy of the trees (Figure 4.3)

0 10 20 30 40 50 60 70 80

Figure.4.3 Elevation and comparative map of the gibbon with forest type in Khe

Nuoc Trong proposed nature reserve in May 2017 (Abu Bakar)

In the study area all forest are evergreen broadleaves forest with two kinds of forest (medium and rich forest) Of the 53 recording sites, 46 recording sites were recorded with medium forest and 8 recording sites were recorded with rich forest Gibbon is distributed in medium and rich forest of the study area and all forest of the study site is broadleaves evergreen forest Most of the gibbons are distributed in the medium forest because of the insufficiency of the rich forest and medium forest

is covering most of the study area as compare to rich forest (Figure 4.4)

Figure.4.4 Elevation and forest type map of the surveyed area with recording posts

in the Khe Nuoc Trong proposed nature reserve in May 2017 (Abu Bakar) Medium forest covered a wide area in the nature reserve as compared to the rich forest The percentage of the rich forest is only 15% and the percentage of the medium forest is 85% The percentage of the medium forest is much higher than the rich forest (Figure 4.5)