ECOSYSTEM SERVICES OF PTEROPODID BATS, WITH SPECIAL ATTENTION TO FLYING FOXES (PTEROPUS AND ACERODON) IN SULAWESI, INDONESIA

By

SHEHERAZADE

A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2018

© 2018 Sheherazade

To my mom Thank you for teaching me how to be an independent, strong, and happy woman

ACKNOWLEDGMENTS

I would like to thank my advisor, Dr. Holly Ober, for her guidance and patience during my study. I thank her for always encouraging me to maximize all learning opportunities to gain skills and knowledge that will be important for my future. I thank Dr.

Bette Loiselle and Dr. Todd Palmer for their invaluable advice to this study and thesis.

I thank the Center for International Forestry and United States Agency for

International Development for funding my master program. I thank Dr. Steven Lawry,

Ms. Dina Hubuddin, Ms. Rahayu Koesnadi, Ms. Raya Soendjoto, Dr. Karen Kainer, and

Dr. Bette Loiselle for providing supports and managing administrative issues during my entire study. I thank Tropical Conservation and Development (TCD) University of Florida for allowing me to learn about the interdisciplinary aspects of conservation and awarding me a field research grant. I thank Rufford Small Grant, Bat Conservation

International, and IdeaWild for also funding my research. I thank Government Agency of

Maritime Affairs and Fisheries of Central Sulawesi, Integrated Permit Service Agency of

Central and West Sulawesi, Longkoga Barat/Timur village, Bualemo Subdistrict,

Batetangnga village and Binuang Subdistrict for permits. I thank Alliance for Tompotika

Conservation and my team for their tremendous assistance in the field.

I thank Dr. Susan M. Tsang and Asnim Alyoihana for being my lifetime mentors, collaborators, and best friends. I thank Tambora Muda Indonesia for trusting me as their president and working there voluntarily always give me the energy to finish my study. I especially would like to thank my best friends, Marsya, Ardian, Sapi, Nuy, Mita, Ridha,

Yulia, and Ana for their encouragements, lessons, and of course, laughter. Thank you for being really patient with a difficult person like me! I thank Eight Gators, especially

Ikbal, who have been good companies since the program started. Finally, I thank my

4

family, Mama, Oneng, and Ei, for all their support and unconditional loves. Thank you for respecting and believing in my dream to work for wildlife conservation in Indonesia.

5

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... 4

LIST OF TABLES ...... 8

LIST OF FIGURES ...... 9

ABSTRACT ...... 10

CHAPTER

1 INTRODUCTION ...... 12

2 BAT POLLINATION SERVICES TO DURIAN AND SIGNIFICANCE TO LOCAL ECONOMY IN SULAWESI, INDONESIA ...... 16

Background ...... 16 Methods ...... 18 Study Sites ...... 18 Results ...... 24 Floral Characteristics of Durian ...... 24 Pollination Biology of Durian ...... 24 Significance of Each Flower Visitor as Durian Pollinators ...... 25 Economic Contribution of Primary Pollinators to Durian Production ...... 27 Discussion ...... 27 Pollinator Assemblage of Durian in Sulawesi ...... 29 Resource Partitioning among Bat Species ...... 31 Appeal for Bat Conservation ...... 33

3 ECOSYSTEM SERVICES PROVIDED BY ACERODON CELEBENSIS AND PTEROPUS GRISEUS IN SULAWESI, INDONESIA ...... 48

Background ...... 48 Methods ...... 50 Study Sites ...... 50 Pollen Collection from Flying Foxes ...... 50 Resource Use of Flying Foxes ...... 52 Characterization of Ecosystem Services Provided by Flying Foxes ...... 54 Results ...... 55 Resource Use of Flying Foxes ...... 55 Ecosystem Services of Flying Foxes ...... 56 Discussion ...... 57 Ecosystem Services: Cultivated Crops ...... 58 Ecosystem Services: Native Plants ...... 60 Ecosystem Services: Unique Contributions ...... 61

6

4 CONCLUSION ...... 72

APPENDIX

A A LIST OF ANIMAL VISITORS AND PRIMARY POLLINATORS OF SEMI-WILD DURIAN THROUGHOUT SOUTHEAST ASIA...... 74

B RESULTS OF GENERALIZED LINEAR MIXED-EFFECT MODELS (GLMM) TO DETERMINE WHICH FACTORS CONTRIBUTE MOST TO DURIAN PRODUCTION ...... 76

LIST OF REFERENCES ...... 77

BIOGRAPHICAL SKETCH ...... 86

7

LIST OF TABLES

Table page

2-1 Duration of nocturnal visits by each animal to durian flowers...... 36

2-2 Comparison of number of durian fruit set under open pollination treatment across two pollinator groups...... 37

3-1 Morphometric measurement of a subset of the flying foxes found in the study sites: 18 adults of A. celebensis and 16 adults of P. griseus...... 65

3-2 Plants visited by flying foxes, the resources flying foxes used (flowers, fruits), the pollen load, and the value of each plant to humans in the local community...... 66

A-1 A list of animal visitors and primary pollinators of semi-wild durian throughout Southeast Asia...... 74

B-1 Results of Generalized Linear Mixed-effect Models (GLMM) to determine which factors contribute most to durian production...... 76

8

LIST OF FIGURES

Figure page

2-1 Study site in Batetangnga Village, West Sulawesi, Indonesia...... 38

2-2 A durian inflorescence (yellow circle) growing directly from the branch, consisting of tens of flowers...... 39

2-3 The treatments within the durian pollination exclusion experiment...... 40

2-4 An example of durian fruit set that was monitored at (a) Day 10; (b) Day 20; (c) Day 30); (d) Day 60...... 41

2-5 Mean nectar volume per durian flower (±SE) calculated cumulatively (n=10). ... 42

2-6 The number of durian fruit set under three pollination treatments...... 43

2-7 Different bat species fed on durian nectar...... 44

2-8 Temporal overlap of foraging activity on durian between pairs of species...... 45

2-9 Strigocuscus celebensis visited a durian inflorescence and consumed the flowers...... 46

2-10 Other durian visitors...... 47

3-1 The two study sites were islands off the coast of Central Sulawesi in Bualemo: Mantalu Daka Island and Tangkuladi Island...... 67

3-2 Some examples of plants used by flying foxes...... 68

3-3 Diet composition of Pteropus griseus during dry and rainy seasons...... 69

3-4 Diet composition of Acerodon celebensis during dry and rainy seasons...... 70

3-5 Species accumulation curve for plant species used by flying foxes...... 71

9

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science

ECOSYSTEM SERVICES OF PTEROPODID BATS, WITH SPECIAL ATTENTION TO FLYING FOXES (PTEROPUS AND ACERODON) IN SULAWESI, INDONESIA

By

Sheherazade

December 2018

Chair: Holly K. Ober Major: Wildlife Ecology and Conservation

Pteropodid bats, especially flying foxes, are threatened by hunting in Indonesia.

Scientific data on bat ecosystem services are lacking, yet these data are needed to promote their conservation. This study aims to investigate 1) bat pollination services for durian using a pollination exclusion experiment and camera trapping; and 2) flying fox ecosystem services by identifying pollen collected from fur. We conducted our study in

Sulawesi for eight months (May-December 2017). We found that bats are the primary durian pollinator. Three bat species pollinated durian flowers: Eonycteris spelaea (mean duration of visit: 116.87 sec/visit), and two flying foxes, Pteropus alecto (11.07 sec/visit) and Acerodon celebensis (11.60 sec/visit). Durian flowers visited by both small bats (E. spelaea) and flying foxes produced slightly more fruit than flowers visited only by the small bats, suggesting flying foxes are presumably more effective pollinators. Bat pollination services for durian are valued at ~$117/ha. Additionally, we collected pollen from 52 individuals of Pteropus griseus and 33 individuals of A. celebensis. We identified ~14 plant species used by flying foxes. These plants are economically valuable to the local livelihood and ecologically essential to the Sulawesi rainforest and mangroves. The two flying fox species provided distinct ecosystem services, with A.

10

celebensis functioning more as seed dispersers and P. griseus as pollinators.

Information learned through this research should be used to foster conservation of flying foxes to prevent the loss of productivity of plant species that rely on them, and preclude the loss of benefits they provide to human well-being.

11

CHAPTER 1 INTRODUCTION

Pteropodid bats play important roles in the ecosystems where they occur

(Hodgkison & Balding 2003, Sheherazade et al. 2017, Lim et al. 2018). There are ~200 species of these bats included in a family called Pteropodidae that occurs throughout the Paleotropics (Simmons 2005). Since the diet of pteropodid bats is mainly fruits, these bats contribute significantly to ecosystem health as seed dispersers, which is crucial for the persistence of species diversity in tropical rainforests (Hodgkison &

Balding 2003). Pteropodid bats play crucial roles in forest succession and regeneration as they feed on a considerable amount of figs (pioneer tree species), and seeds processed through their digestive system show high germination rates (Shilton et al.

1999, Hodgkison & Balding 2003). Because their diet is predominantly fruit, pteropodid bats are commonly called Old World fruit bats. However, floral resources (e.g. nectar) also constitute a portion of their diet, implying that these bats may also provide pollination services to native plants in the forests and even mangroves (Hodgkison &

Balding 2003, Nor Zalipah et al. 2016, Thavry et al. 2017).

Ecosystem services provided by pteropodid bats can also benefit human well- being (Ghanem & Voigt 2012, Stewart & Dudash 2016). Research conducted nearly 30 years ago documented 448 economically valuable forest products in the tropics derived from various plants pollinated by pteropodid bats (Fujita & Tuttle 1991). Plants dispersed and pollinated by bats are not only valuable in a cultural sense but also for the economic sector. Many of the tree species that benefit from pteropodid bats are important sources of timber (Kunz et al. 2011). Plants belonging to the family

Bombacaceae (now called Malvaceae) are primarily bat-pollinated, and are of great

12

economic value (Elmqvist et al. 1992, Fleming & Muchhala 2008, Fleming et al. 2009,

Win & Mya 2015). These species include, but are not limited to balsa (Ochroma pyramidale), kapok (Ceiba pentadra), and the cash crop, durian (Durio spp.) (Kunz et al.

2011). Therefore, the benefits bats provide are crucial to the livelihoods of the local people, especially in Southeast Asia (Fujita & Tuttle 1991, Kingston 2010).

Flying foxes are an exceptional group of large pteropodid bats. This common name refers to species belonging mainly to two genera, Pteropus and Acerodon, which make up approximately a third (70 of 200 species) of all recognized species in the

Pteropodidae (Simmons 2005). Flying fox distribution includes Asia, South Pacific, northern Australia, and islands off the coast of East Africa. Compared to other smaller pteropodid bats, flying foxes can fly over longer distances and cover larger areas

(Hengjan et al. 2018). This allows them to disperse seeds over larger regions, which is important for plant propagation in large intact forests and fragmented landscapes (Cox et al. 1992, Shilton et al. 1999, Luskin 2010, Roberts et al. 2012). The roles of flying foxes are especially crucial in island systems with low redundancy of frugivorous animals (McConkey & Drake 2015, Banack 1998). In Mauritius, about a quarter of the native wood species depend on flying foxes to disperse seeds, and these species account for 63% of the basal area in the island’s forests (Florens et al. 2017). In island ecosystems like Madagascar and Iriomote Island, Japan, flying foxes are prominent seed dispersers for figs that are essential for early succession (Oleksy, Giuggioli, et al.

2015, Lee et al. 2017), and for a large number of tree species endemic to the island

(Bollen & Van Elsacker 2002). Plants with large seeds in particular, depend on flying foxes for dispersal services (McConkey & Drake 2006, Florens et al. 2017).

13

Flying foxes also play an important functional role in pollinating plants in the tropics. For some of the flying foxes, floral-related parts are crucial diet components

(Baker et al. 1998, Banack 1998). Pteropus poliocephalus and P. scapulatus in

Australia depend largely on the nectar from flowers of Eucalyptus spp. and Corymbia citriodora (Marshall 1983, Birt 2004, Roberts et al. 2012). These bats demonstrate a preference for nectar over other food sources during the time of year when the eucalyptus species are abundant in the region (Birt 2004). In Myanmar, P. giganteus feed on nectar of flowers from plant species in the family Bombacaceae (Win & Mya

2015). In India, P. giganteus feed extensively on the nectar of Madhuca latifolia flowers

(Nathan et al. 2009). Pollination services provided by flying foxes are not only ecologically important, but also economically significant (Abdul-Aziz, Clements,

McConkey, et al. 2017, Abdul-Aziz, Clements, Peng, et al. 2017).

Although interest in documenting the ecosystem services provided by bats has increased over time, research remains sparse in some regions. The roles of flying foxes remain understudied in areas where flying foxes are highly threatened, which is where such information is needed most. As one of the important centers of bat diversity in the world, Indonesia represents a location where the ecosystem services provided by pteropodid bats may be crucial for the natural persistence of the rainforest. The nation comprises thousands of islands that likely need volant animals for long distance pollination and seed dispersal (Fleming et al. 2009). Sulawesi is particularly well suited to an investigation of the role of pteropodid bats because of its high diversity of bats and the small number of other medium to large herbivorous mammals, especially primates.

The lack of alternative vertebrate vectors here suggests pteropodid bats may contribute

14

heavily to pollination and seed dispersal. Since many pteropodid bats, including flying foxes, have a broad generalist diet, many ecologically and economically important plants potentially depend on these animals for pollination and seed dispersal (Fujita &

Tuttle 1991, Fleming & Sosa 1994). Moreover, Pteropodidae is the most threatened bat taxa in the world (Kingston 2010), and Sulawesi, Indonesia is the hotspot for hunting these animals for bushmeat (Sheherazade & Tsang 2015). Better understanding of the ecosystem services provided by these bats in Sulawesi can help to devise a more convincing case for flying fox conservation.

15

CHAPTER 2 BAT POLLINATION SERVICES TO DURIAN AND SIGNIFICANCE TO LOCAL ECONOMY IN SULAWESI, INDONESIA

Background

Durian is one of the most economically important fruits in Southeast Asia, with an export value of up to US$254.85 million (Indarti 2014). Indonesia is one of the primary producers of durian, and the fruit constitutes a significant portion of Indonesia’s gross domestic product (Indarti 2014). Durian was named by the Ministry of Agriculture as one of the five national fruit priorities along with banana, mango, citrus, and mangosteen, meaning its productivity and sale are national interests (Rafani 2013). In 2014, durian production reached up to 859,118 tons, and commanded a higher unit price than any other fruit commodity in Indonesia (Rafani 2013, Indarti 2014). Despite the popularity of durian, productivity of this cash crop for agricultural export or domestic consumption is much lower than that of other fruits. As a result of the huge domestic demand and the low productivity of domestic plants, Indonesia needs to import durian from Thailand,

Vietnam, and Malaysia (Rafani 2013). This durian import rate is increasing by 5 percent annually (Santoso 2012), widening the export-import gap between Indonesia and other

ASEAN countries (Indarti 2014).

Understanding the natural mechanism of durian production is essential to increasing fruit production and growing the Indonesian agricultural sector to decrease this trade deficit. However, compared to other Bombacaceae species, durian pollination ecology is poorly understood and studies have been limited to peninsular Southeast

Asia (Allen-Wardell et al. 1998, Yumoto 2000, Bumrungsri et al. 2009). Without additional knowledge, it is difficult for the Indonesian government to know where to

16

direct resources to improve crop production and establish Indonesia as a major global producer of durian.

Although plantations of some durian cultivars exist, most small-scale farmers in

Indonesia still rely mainly on semi-wild durian. Semi-wild durian grow in agroforestry systems without direct management and intervention by local farmers (Rafani 2013). It has been suggested that semi-wild durian may play a more significant economic role in domestic and local markets than previously reported (Djajanti 2006, Kunz et al. 2011).

However, quantification of the contribution of specific pollinators to the semi-wild durian trade has been estimated only in Thailand and Peninsular Malaysia (Bumrungsri et al.

2009, Stewart & Dudash 2016, Abdul-Aziz, Clements, McConkey, et al. 2017).

Durian is incredibly diverse, with 30 species and 76 varieties in Indonesia alone

(Santoso 2012). Durian flowers are generally self-incompatible and require animal vectors for pollination (Lim & Luders 1998, Yumoto 2000, Bumrungsri et al. 2009).

Some durian species possess the classically defined floral traits of a plant with chiropterophilous pollinator syndrome, suggesting that bats are their primary pollinators.

These floral traits include white or dull-coloration, strong fragrance, and nocturnal blooming (Troll 1975, Tschapka & Dressler 2002). Pteropodid bats, such as Eonycteris spelaea (common nectar bat), Macroglossus minimus (lesser long-tongued fruit bat), and Pteropus hypomelanus (island flying fox) are known to be durian visitors in other parts of Southeast Asia (Start 1974, Dobat & Peikert-Holle 1985, Marshall 1985, Fujita

& Tuttle 1991, Abdul-Aziz, Clements, McConkey, et al. 2017), but the assemblage of durian visitors and its primary pollinators differ among areas and durian cultivars

(Appendix A). In Malaysia, Durio grandiflorus and D. oblongus are primarily pollinated

17

by spiderhunter birds, while D. kutejensis is pollinated by bees, birds, and bats (Yumoto

2000). Durio zibethinus in southern Thailand and peninsular Malaysia rely on Eonycteris spelaea as their primary pollinator (Bumrungsri et al. 2009, 2013, Stewart & Dudash

2016, Abdul-Aziz, Clements, McConkey, et al. 2017).

Additional investigation of pollination systems in other major semi-wild durian producing areas will be a necessary component of efforts to improve crop production rates. Sulawesi will be a suitable place for the investigation because it has a different assemblage of volant wildlife than other regions where pollination ecology of durian has been studied, and because Sulawesi produces durian that are of interest to agricultural production. Our study aims to understand these pollination dynamics of semi-wild durian in Sulawesi, Indonesia, where durian is an essential local commodity. Our objectives were to:

1. Describe the floral characteristics of semi-wild durian in Sulawesi;

2. Investigate the pollination biology of durian, and confirm whether it is self- incompatible and requires animal pollinators;

3. Determine the contribution of each species of flower visitor as durian pollinators and their temporal use of durian flowers;

4. Estimate the economic contribution of primary pollinators to durian production.

Methods

Study Sites

This study was conducted in Batetangnga Village, Binuang subdistrict in Polewali

Mandar Regency, West Sulawesi (Figure 2-1), during the flowering and fruiting season of semi-wild durian (October-November 2017 and November 2017-January 2018, respectively). Batetangnga is the most geographically expansive village in the subdistrict (44.80 km2), comprising 45% of the total area (Badan Pusat Statistik

18

Kabupaten Polewali Mandar 2017a). It extends from the coastal lowlands into inland and montane areas dominated by secondary forest mixed with various agricultural plantations. The average annual rainfall in the area is 1,954 mm, with peaks in rainfall usually occurring in May, June, and December. Local people primarily cultivate cacao plantations together with rambutan (Nephelium lappaceum), langsat (Lansium parasiticum), mango (Mangifera indica), and notably durian (Durio zibethinus). All protocols were approved by the University of Florida under IACUC protocol No.

201709800.

Objective 1. Floral Characteristics of Durian

The durian species at our study site was Durio zibethinus. During the flowering season, durian trees have hundreds to thousands of inflorescences (a cluster of flowers on a stem). Durian flowers are cauliflorous, meaning tens of flowers are stacked per inflorescence and grow directly on branches (Figure 2-2). Understanding the pollination biology of durian requires an understanding of durian’s floral characteristics.

We observed 10 durian flowers, from five inflorescences in three durian trees.

Due to the height of durian trees (15-40 m), we selected relatively shorter trees with accessible flowers to conduct these observations. We recorded two sensory characteristics of durian flowers that might serve to attract pollinators, which were color and odor. The color of durian flowers was measured using a Nix Pro Color Sensor (Nix

Sensor Ltd.) that generated numerical data describing the Red, Green, and Blue (RGB) color spectrum for every flower. We input this RGB combination to online color tools, colorhexa.com and perbang.dk/rgb/, to generate the color names. We categorized the odor of each durian flower as odorless, mild, or strong (Watson 1983).

19

We carried out hourly observations and tests to determine five physiological attributes of the local durian species (Dobat & Peikert-Holle 1985, Lim & Luders 1998,

Yumoto 2000, Honsho, Yonemori, Sugiura, et al. 2004, Honsho, Yonemori, Somsri, et al. 2004, Bumrungsri et al. 2009). We assessed (1) when flowers bloomed; (2) the time each anther dehisced, by noting the presence of a slit on the anther as a sign that pollen grains had been released; (3) stigma receptivity, by placing a drop of hydrogen peroxide on each stigma and noting the presence of bubbles, which indicated that the stigma was receptive to pollen; (4) nectar production, using 100 l micro-capillary tubes to draw and measure nectar; and (5) sucrose concentration of nectar using a sucrose refractometer (Extech RF10 0 to 32% Brix Refractometer). All observations and tests began when durian flowers bloomed and ended when their corollas dropped from pedicels.

Objective 2. Pollination Biology of Durian

We chose seven 18-20 m tall durian trees for a pollination exclusion experiment, using three criteria to select focal trees: (1) a sufficient number of inflorescences to accommodate all the treatments (described below), (2) trees could be climbed without safety concerns (e.g. it was impractical to climb durian trees which were too slippery or had large trunk diameters), and (3) an absence of ant nests. For every target tree, we selected ~3-6 durian inflorescences and assigned them to one of three treatments: closed, insect, and open pollination (Figure 2-3). We picked inflorescences that had ~30 flowers.

For the closed pollination treatment, we bagged each inflorescence before the flower bloomed with a bag made of a mesh fabric that allowed light, rain, and gas

20

exchange between flowers and the ambient environment, but prohibited any animals from accessing the flowers. For the insect pollination treatment, inflorescences were enclosed with specialized nets made of the same garden fabric but manually perforated to allow only insects to access flowers (perforation size = 1.5 cm). For the open pollination treatment, we did not prevent access to the inflorescences by any potential visitors. All inflorescences were labeled to indicate replicate number, treatment, and experiment date. We had 36, 30, and 25 replicates (inflorescences) for open, insect, and closed treatments, respectively. Because durian flowers might have late acting self- incompatibility, and to compare pollination and reproductive success (Honsho,

Yonemori, Somsri, et al. 2004, Bumrungsri et al. 2009), we monitored and counted durian fruit set for all replicates on day 10, 20, 30 and 60 following the time each treatment was set up (Figure 2-4).

We used Kruskal-Wallis and Dunn’s tests (Dinno 2017) to determine whether each treatment produced significantly different amounts of durian fruit. We focused on fruit set on days 20 and 60 because these time intervals indicated pollination and reproduction success, respectively (Abdul-Aziz, Clements, McConkey, et al. 2017). We visualized data using ggplot from the package ‘ggplot2’ in R (Wickham & Chang 2016).

Objective 3. Significance of Each Flower Visitor as Durian Pollinators

We used camera traps to determine which animals visited durian flowers (Figure

2-3). We deployed ~18 camera traps (Bushnell Trophy Cam HD Essential E2 12 MP

Trail Camera) on the branches in front of durian inflorescences receiving open pollination treatments and rotated them to the next experimental set up a week later.

Camera traps were active during the entire flowering period for each inflorescence. The distance between inflorescences and camera traps ranged from 30 cm to 1.5 m.

21

Camera traps recorded day and night visitation rates of animals that fed on durian floral products (nectar or the whole flowers). We set up the camera traps to capture videos with a maximum length of 15 s, with 5 min intervals between consecutive recordings.

We used the total duration of visits by each animal species as a proxy for the contribution of each pollinator species to durian production. The videos also showed the foraging behavior of animal visitors, which helped us determine which species fed on nectar (and presumably benefitted durian by serving as primary pollinators) and which ate the whole flowers (and were thus detrimental to durian).

We used a generalized linear mixed-model (GLMM) to determine whether types of bat visitors, number of individual bat visits, and duration of bat visits affected durian fruit set. This approach is appropriate for analyses that combine numerical and categorical data as explanatory variables (Bilder & Loughin 2015). We divided types of bat visitors into two categories: 1) small bats only (E. spelaea), and 2) all bats (E. spelaea, Pteropus alecto, and Acerodon celebensis). We used GLMM with negative binomial distribution for durian fruit set on day 20 because there was an overdispersion, while we used GLMM with Poisson distribution for durian fruit set on day 60 because there was no overdispersion. We used Wilcoxon signed-rank tests to determine whether visitation by different groups of bats resulted in different amounts of fruit set.

We assessed the temporal aspect of durian foraging of each bat species, and determined whether there was temporal resource partitioning among bat species. We converted time of visitation to radians and used non-parametric kernel density estimates to display each species’ temporal foraging activity as a continuous distribution over a

24-hour cycle (Frey et al. 2017). We generated the estimates and density plots using

22

package ‘overlap’ in R (Meredith & Ridout 2017). We then used a descriptive measure of the degree of similarity to decide whether there was an overlap between each pair of species. The value ranged from 0 (no overlap) to 1 (complete overlap) (Ridout & Linkie

2009, Frey et al. 2017). All analyses were run in R 1.0.136 (https://cran.r-project.org).

Objective 4. Estimate the Economic Contribution of Primary Pollinators to Durian Production

We used a bioeconomic approach to estimate bat pollination services to durian, using a formula originally developed to calculate insect pollination services (Gallai et al.

2009). We modified the name of the formula ‘Insect Pollination Economic Value’ to

‘Pollination Economic Value (PEV)’:

퐼 푋 푃퐸푉 = ∑ ∑(푃 × 푄 × 퐷 ) 푖푥 푖푥 푖 (2-1) 푖=1 푥=1 where: i = crop x = region

Pix = price of crop i produced in x

Qix = quantity produced in region x

Di = dependence ratio of the crop i on bat pollination

Dependence ratio (0-1), Di, expresses the degree of durian dependency on pollination.

It is calculated by subtracting fruit set of bagged flowers (closed pollination) from fruit set of unbagged flowers (open pollination), divided by the fruit set of unbagged flowers

(Kasina et al. 2009). In our case, we calculated durian dependence ratios in a similar manner, but we used fruit set of bagged flowers subjected to insect pollination for closed pollination treatments since insects may also pollinate durian flowers.

23

Results

Floral Characteristics of Durian

Durian flowers had distinct attributes that conform to the chiropterophilous syndrome, which included dull coloration of the corolla and night-blooming behaviors.

Durian floral colors were variable, but most of them were grayish orange or light orange- gray, according to measurements with our color sensor. Our direct observations suggested the color of durian flowers was yellowish white. The flowers maintained this color during anthesis, but they turned brownish white a few hours before the corolla dropped the next morning. Other durian flowers in our study site showed similar coloration. We note that all flower color examination was conducted at night; the transparency of durian corollas may have impeded the measurement of the Nix Pro

Color, leading to consistent bias in color estimation. We could not quantify the odor of durian flowers, but report that durian flowers produced strong odor.

The flowers started to open at 1530 h, and fully opened around 1800 h. Pollen was released (anther dehiscence) after 1830 h, about the same time that stigmas became receptive. Stigmas were still receptive until the corolla dropped the next morning. During this time, the average amount of nectar produced by each durian flower was 167.4 l  15.11 (SE) per night (range: 79-240 l, n=10) (Figure 2-5). Nectar contained approximately 13.53%  0.48 (SE) (range: 9.95-16%, n=10) of sucrose.

Pollination Biology of Durian

Overall, durian subjected to the open pollination treatments had significantly higher fruit set compared to durian with insect and closed pollination treatments both on day 20 and 60 (Day 20-pollination success: 훸2₂ = 38.39, 푝 = 4.611푒 − 09; Day 60- reproductive success: 훸2₂ = 32.52, 푝 = 8.662푒 − 08) (Figure 2-6). The average number

24

of fruits set in open, insect and closed pollination treatments on day 60 was 1.69  0.37

(SE) (range: 0-10 fruits, n=36), 0.27  0.11 (SE) (range: 0-2 fruits, n=30) and 0.04 

0.04 (SE) (range: 0-1 fruit, n=25) respectively.

Significance of Each Flower Visitor as Durian Pollinators

Durian visitors consisted of four vertebrate species (three species of bats and one species of cuscus) and two invertebrate species (one species of bee and moth).

Bats (Eonycteris spelaea and two species of flying foxes, Pteropus alecto, the black flying fox, and Acerodon celebensis, the Sulawesi flying fox; Figure 2-7, Table 2-1) were the primary visitors to durian flowers in the open pollination treatment (6,828 sec out of

9,208 sec). These three species could be easily differentiated in the camera trap videos.

Eonycteris spelaea was the smallest species and had brightest (grayish) fur. Acerodon celebensis appeared smaller and brighter than P. alecto due to the differences in fur reflectance. Bats of all three species were observed drinking nectar from the floral tubes without destroying the flowers. Although flying foxes have large bodies, they hang from tree branches and bend their head towards the flowers when drinking, and therefore did not damage the inflorescences.

All three bat species showed a single nightly peak in visitation to durian.

Eonycteris spelaea and P. alecto showed similar temporal foraging activity on durian nectar (overlap = 0.88). However, both of these species showed different temporal foraging activity than A. celebensis (overlap estimate of E. spelaea-A. celebensis =

0.45; overlap of P. alecto-A. celebensis = 0.41). Eonycteris spelaea and P. alecto fed on durian nectar throughout much of the night, from 1900 h to 0300 h with the peak around

2300 h to 2400 h. In contrast, A. celebensis foraged at durian flowers from 2000 h to

25

2300 h with the peak around 2000 h to 2100 h (Figure 2-8). Pteropus alecto was always seen foraging alone. In contrast, we once observed two individuals of A. celebensis feeding together on the same durian inflorescences, and we saw 2-3 individuals of E. spelaea feeding together on the same durian inflorescences on several occasions. We never observed multiple individuals of P. alecto feeding on the same flowers, as we did with the other bat species.

Our GLMM showed that types of bat visitors, number of individual bat visits, and duration of bat visits did not significantly determine durian fruit set both on day 20 and

60 (Appendix B). However, the average durian fruit set was slightly higher on flowers that were visited by both small bats and flying foxes, compared to flowers that were only visited by small bats (Table 2-2), although these differences were not significant (Day

20: W = 43.5, 푝 = 0.1591; Day 60: W = 59.5, 푝 = 0.6147).

No animals other than bats were effective pollinators of durian. While

Strigocuscus celebensis (small Sulawesi cuscus) had the second longest average duration per visit, this species destructively consumed the flowers, and thus is likely not acting as an effective pollinator for durian (Figure 2-9). Flowers consumed by cuscus were not later visited by bats. Insects such as moths and bees rarely visited durian flowers. Durian flowers from insect pollination treatments had intermediate fruit set relative to the closed and open pollination treatments, suggesting that insects are capable of pollinating durian flowers, but not as effectively as bats. Camera traps might not record insects promptly, which possibly underestimated insect contribution as durian pollinators. During the day, camera traps also recorded bees and Aethopyga siparaja

(crimson sunbird) visiting durian flowers. The birds visited unopened flower buds of

26

durian and punctured the base of the flowers with their beak (Figure 2-10). The stigma and anthers were left untouched because the flowers were still closed. Thus, birds were unlikely acting as durian pollinators at this study site. These flowers were not later visited by bats, but we could not determine whether the bird-flower interaction affected durian fruit set because only a few flowers were visited by birds.

Economic Contribution of Primary Pollinators to Durian Production

The degree to which durian depends on bats as pollinators, the durian dependence ratio, was calculated as 0.84. In Batetangnga Village, durian fruiting season only occurs once per year and lasts for a couple of months. In our study, durian flowering season was in October to November 2017, and then the fruiting and harvesting season occurred January to early April 2018. During this period, the total durian production in Batetangnga Village was ~1,497,600 fruits, according to the locals.

The villagers sold durian at a per unit price rather than by weight, with each fruit commanding IDR 5000 (~US$0.35) during our study period. Thus, we estimate that bat pollination services were worth IDR 6.3 billion (~US$ 450,000). Because Batetangnga

Village encompasses ~44.8km2, bat pollination services are valued at ~$117/ha in this region.

Discussion

Our study provides the first evidence that durian flowers in Sulawesi, Indonesia were pollinated primarily by bats. Several observations support this evidence: (1) flowers were open and receptive during nighttime when bats were active; (2) bats had the greatest number of visits to flowers; (3) bats consumed nectar without damaging flowers; and (4) extremely low fruit set occurred via insect pollinators when vertebrates were prevented from accessing flowers during exclosure experiments. These results

27

corroborate findings from other studies in Thailand and Malaysia in which the same durian species, Durio zibethinus, showed similar floral traits and also relied on bats for pollination (Bumrungsri et al. 2009, Abdul-Aziz, Clements, Peng, et al. 2017). Our findings underscore the importance of bats for the pollination of this highly valuable fruit in Southeast Asia.

Among nocturnal animals, bats may increase the probability of successful durian pollination by transferring sufficient loads of pollen among different durian trees.

Previous studies have documented that pteropodid bats are long-distance pollen dispersers that can deposit large loads of pollen (Bumrungsri et al. 2013). For example,

E. spelaea, the main visitors of durian flowers in this study, travels up to 8-38 km while foraging (Start 1974, Acharya et al. 2015). Such long-distance pollen dispersers are important for low-density plant species such as the durian. Long distance pollination also increases the probability of pollen coming from genetically distant and distinct individuals, which is important to successful pollination and production of high quality durian fruits (Honsho et al. 2009).

Pteropodid bats in the non-seasonal parts of Southeast Asia are believed to occupy a more generalized feeding niche compared to the phyllostomid bats in the

New-World due to the lower resource diversity and lower spatio-temporal predictability in Southeast Asia (Fleming & Muchhala 2008). Hence, these generalist pteropodid bats are suspected to transfer heterospecific pollen and, thus, may be non-efficient pollinators. However, recent studies show that pteropodid bats deposit sufficient conspecific pollen to fertilize most ovules in a flower (Stewart & Dudash 2016, 2017).

Furthermore, observations demonstrate that only one bat visit is required to ensure

28

successful pollination for other plant species (Srithongchuay et al. 2008, Nor Zalipah et al. 2016). Our ongoing research on flying fox diet suggested that most bats (55%) visit only one plant species per night (Chapter 3), which increases the likelihood of conspecific pollen transfer. Unfortunately, we were not able to capture bats near durian trees to quantify pollen loads and verify the proportion of conspecific pollen carried by the bats. We expect that durian provides adequate and essential food resources to fulfill the energetic requirements of a bat because durian produces a massive number of flowers per night. For this reason, it is likely not necessary for bats to visit other plant species to meet their energetic needs, increasing the probability that bats transfer durian pollen from one durian tree to another. Indeed, a study conducted in Thailand found that 93% of D. zibethinus stigmas are loaded with conspecific pollen carried by the bats (Acharya et al. 2015). Eonycteris spelaea in other areas are known to primarily feed on durian nectar and repeatedly visit durian flowers every night during the flowering period; consequently, there is a high chance of transferring conspecific pollen within and among trees (Bumrungsri et al. 2013, Acharya et al. 2015, Thavry et al.

2017). The high-reward big-bang plant species, such as durian, that have limited phenological overlap with other big-bang species may promote high pollinator constancy and reduce interspecific pollen transfer (Stewart & Dudash 2017).

Pollinator Assemblage of Durian in Sulawesi

In Sulawesi, Indonesia, the pollinator assemblage of durian is more diverse than that in Malaysia and Thailand, likely because Sulawesi has the highest diversity of pteropodid bats (Mickleburgh et al. 1992, Simmons 2005). Our study supports previous findings that E. spelaea, a generalist nectarivore, is a durian pollinator in Sulawesi also, much like in other regions (Bumrungsri et al. 2009, Abdul-Aziz, Clements, McConkey, et

29

al. 2017). We also document durian pollination by the frugi-nectarivorous Pteropus alecto, which has been recorded a few times as a durian pollinator elsewhere (Marshall

1985). Our study provides the first record of durian pollination services by Acerodon celebensis, a species endemic to Sulawesi considered vulnerable by the IUCN due to extensive harvest for bushmeat (Sheherazade and Tsang 2015).

We cannot confidently state which bat species are most important for durian pollination in our study. Due to human safety concerns, we only placed camera traps in durian trees that were 18-20 m-tall. This restriction to shorter trees may have resulted in an underestimation of the role of flying foxes, which presumably visit durian flowers at higher canopy levels than the smaller E. spelaea (>20 m) (Abdul-Aziz 2016). Flying foxes are presumably more effective pollinators because they can deposit larger loads of pollen and move them over longer distances than smaller pteropodid bats. In

Australia, P. alecto individuals are known to forage up to 3-20 km (Vardon et al. 2001,

Markus & Hall 2004). Other species of flying foxes are also recognized as long-distance pollinators: P. poliocephalus has been known to move 17-25 km during foraging bouts

(Spencer et al. 1991); Pteropus tonganus - 5-22 km (Banack & Grant 2002); Pteropus rufus - 1-7 km (Oleksy, Racey, et al. 2015); and Pteropus vampyrus - 88-363 km

(Epstein et al. 2009). Further investigation is needed to quantify pollen loads carried by flying foxes and to better understand their individual contributions and differences to pollinator services for specific plant species.

Nectarivorous bat species, such as E. spelaea, are in general more important pollinators than the primarily frugivorous bat species, such as Pteropus and Acerodon

(Stewart & Dudash 2016). The potential role of flying foxes as pollinators should not be

30

overlooked, however, as flowers are an important diet component for these species during the times of year when fruits are scarce and flowers plentiful (Palmer et al. 2000,

Banack & Grant 2002). Flying foxes are opportunistic and sequential specialists, meaning they exploit preferred resources as each becomes seasonally available

(Marshall 1985, Banack 1998). Once durian blooms, their nectar becomes an abundantly available and nutritious resource for flying foxes. To date, all records of flying foxes using flower resources pertain to Pteropus species: this study presents the first evidence of nectar use by an Acerodon species.

Resource Partitioning among Bat Species

Within the bat pollinator assemblage, we might expect resource partitioning to occur among the three bat species that use durian nectar. Nectar can be a limiting resource for hundreds of thousands of bats in a short period of time during the flowering season. Some studies have reported territoriality of flying foxes when feeding on highly available and preferred resources (Gould 1977, 1978, Elmqvist et al. 1992, Banack

1998, Brooke 2001, Luskin 2010). In our studies, we observed greater aggressive interactions between P. alecto individuals than either E. spelaea or A. celebensis. The potential intra- and interspecific interaction among bats are worthy of additional investigation. In general, most of the bats, particularly the flying foxes, feed on the durian nectar solitarily. Flying foxes have been known to defend flowers where they are foraging from conspecifics by vocalizing and wing clapping (Brooke 2001, Abdul-Aziz,

Clements, McConkey, et al. 2017). This kind of feeding behavior may be potentially advantageous to the flowers, as the territoriality increases the chances of bats visiting different flowers of different trees (Gould 1977, Elmqvist et al. 1992).

31

Bats may exhibit both spatial and temporal resource partitioning. We could not test whether the bats partitioned their use of trees by height because we deployed all the camera traps at relatively similar heights, 18-20 m above the ground. However, we heard loud squeaking sounds and wing-clapping from the canopy of 30-40 m durian trees during the flowering season. Local people made similar observations. We believe these sounds were flying foxes that forage considerably higher in the trees compared to the smaller bats. Accessing flowers high in the trees would be easier and energetically less costly for the large pteropodid bats, which are less maneuverable than the smaller bats (Palmer et al. 2000). Most other bat studies corroborate this concept. Pteropus hypomelanus feeds at greater heights in durian trees (6-20 m) than Eonycteris spelaea

(6 m) in Peninsular Malaysia (Abdul-Aziz, Clements, McConkey, et al. 2017). Pteropus giganteus feeds at greater heights of Madhuca latifolia trees than Cynopterus sphinx with different temporal peaks for foraging activity (P. giganteus – 2030; C. sphinx –

1930) (Nathan et al. 2009). In the kapok trees, Pteropus giganteus (15-20 m) also forages at greater heights than C. sphinx with again distinct foraging times, 2000-2100 and 2300 respectively (Singaravelan & Marimuthu 2004).

In contrast, we did not find a clear temporal pattern in durian visitation by the bats. Eonycteris spelaea and P. alecto had similar foraging times, but these activity peaks occurred at different times than A. celebensis. Small pteropodid bats may partition resources from the larger bats spatially (by foraging predominantly at lower heights within trees), whereas the two large species partition resources temporally within nights. This partition may be a function of interference, smaller bats cannot visit

32

the flowers that are being used by larger bats, or roost location relative to the plants we monitored, which determines the time bats start foraging at these plants.

Appeal for Bat Conservation

Our study experimentally demonstrates the importance of pollination services provided by bats, including flying foxes, for the production of semi-wild durian in

Sulawesi, Indonesia. These pollination services are valued as much as US$ 450,000, for the production of 1,500 tons of durian fruit in Batetangnga village. However, this estimate should be treated with caution and considered a rough estimate since our study lasted only a single month and was conducted in only one village. Total durian production was based simply on an estimate provided by a knowledgeable local, because the official data about fruit production this year was not generated yet by the local government agency. The estimate did not take into account the proportion of fruits being consumed locally versus those traded within or outside the village, for which the price might vary. Binuang subdistrict, where the village is located, produces the greatest amount of durian in Polewali Mandar Regency. In 2016, the total production reached up to 103,036.5 tons or 86% of the total durian production in Binuang subdistrict (Badan

Pusat Statistik Kabupaten Polewali Mandar 2017b). Durian production here surpasses the production of other tropical fruits, and communal feasts of durian attract crowds of tourists, who in turn provide an additional source of income to the locals.

For most bat-pollinated plants, especially mass fruiting trees such as durian, high abundances of bats are indispensable to producing sufficient amounts of fruit to meet market demand (McConkey & Drake 2007, Lee et al. 2009, Nathan et al. 2009). Our research links the health of the local bat populations to the local economy. Conservation of bats in Sulawesi should be promoted to prevent the loss of productivity of plant

33

species that rely on them for pollination. This conservation is notably relevant to the flying foxes, A. celebensis and P. alecto, that have been intensely hunted throughout

Sulawesi for the bushmeat markets in North Sulawesi (Sheherazade & Tsang 2015).

The endemic A. celebensis is already listed as Vulnerable in the IUCN Red List (Tsang

& Sheherazade 2016). Additional loss of bats may have a profound impact both ecologically and economically. In our study region, bat pollination services are not only important to the local economy, but also culturally valuable during communal durian feasts.

We recommend the prioritization of bat protection by the Indonesian government and conservation NGOs. In our study area, flying foxes are considered by locals to be a pest of the economically important crops langsat and rambutan, and are often killed through poisons. After we informed locals of the role of bats in pollinating their durian, many opted to use non-lethal methods to deter bats from utilizing crop plants.

Specifically, they hung spoiled fish and chicken organs on the langsat and rambutan trees to deter the bats. This example demonstrates that there are alternative solutions available for bats and people to co-exist, and that locals are willing to institute these changes once they become aware of the importance of bats to durian.

This study provides information that could potentially be used to improve the productivity of durian in Indonesia to fulfil the high domestic market demand. More importantly, the government should start considering the potential of marketing the organic semi-wild durian pollinated by the bats. Organic fruits command higher prices in the market than non-organic ones, and an increase in perceived quality by consumers may increase the value of durian, and as a result, decrease the trade deficit without

34

having to increase crop yield. The Indonesian government may also target a larger, wealthier international clientele for the sale of organic durian, thereby increasing their market reach to an audience that has more disposable income. By highlighting the organic aspects of semi-wild durian, Indonesian durian could profitably be distinguished from the mass-produced durian from Thai farms, creating a separate niche market without having to compete directly for the same consumers. This “re-branding” of

Indonesian durian could be advantageous to the Indonesian economy and to bat conservation.

35

Table 2-1. Duration of nocturnal visits by each animal to durian flowers. Number of Total duration Mean duration of Family Species Common names inflorescence of visits (sec) visit (sec/visit) visited Pteropodidae Eonycteris spelaea Common nectar bat 6567 24 116.87 Pteropus alecto Black flying fox 140 7 11.07 Acerodon celebensis Sulawesi flying fox 121 5 11.60 Phalangeridae Strigocuscus celebensis Small Sulawesi cuscus 2335 2 583.75 Apidae Apis dorsata Giant honey bee 30 2 6 Erebidae Unknown Moth 15 1 15

36

Table 2-2. Comparison of number of durian fruit set under open pollination treatment across two pollinator groups. Durian flowers visited by both small bats and flying foxes produced more fruit when compared to the flowers visited only by the small bats. Small bats and flying foxes Average number of durian Small bats only (Eonycteris spelaea + Pteropus alecto + per inflorescence (Eonycteris spelaea) Acerodon celebensis) Day 20 5.41  1.48 (SE) 7.5  1.80 (SE) (range: 0-18) (range: 0-23) Day 60 1.12  0.36 (SE) 1.25  0.45 (SE) (range: 0-4) (range: 0-6)

37

Figure 2-1. Study site in Batetangnga Village, West Sulawesi, Indonesia.

38

Figure 2-2. A durian inflorescence (yellow circle) growing directly from the branch, consisting of tens of flowers. Photo courtesy of author.

39

Figure 2-3. The treatments within the durian pollination exclusion experiment: (a) Closed pollination treatment prevented access to all vertebrates and invertebrates, as an inflorescence was enclosed with a net; (b) Insect pollination treatment restricted access to flowers by vertebrates but not invertebrates, as an inflorescence was bagged with a net that was manually perforated (yellow arrows, perforation size = 1.5 cm); (c) Open pollination treatment did not restrict access to flowers in any way; (d) A camera trap was deployed in front of an inflorescence (yellow circle) receiving an open pollination treatment before the flowers bloomed. In this picture, durian fruits were several days old and corollas were already fallen. Photos courtesy of author.

40

Figure 2-4. An example of durian fruit set that was monitored at (a) Day 10; (b) Day 20; (c) Day 30); (d) Day 60. All of them received open pollination treatments. Photos courtesy of author.

41

Figure 2-5. Mean nectar volume per durian flower (±SE) calculated cumulatively (n=10).

42

Figure 2-6. The number of durian fruit set under three pollination treatments.

43

Figure 2-7. Different bat species fed on durian nectar. 1) Eonycteris spealea hung on durian flowers while feeding on nectar (a), sometimes accidentally pulled off flower reproductive organs (stigma and anther; b), and convincingly served as pollinator as bat head touched lower reproductive organs while feeding on nectar. 2) Pteropus alecto hung on the branch while feeding on nectar (a), grabbed a flower, but did not pull it off (b), and convincingly served as pollinator as its head touched flower reproductive organs. 3) Acerodon celebensis hung on the branch while feeding on nectar (a), bat licked its fur (b), and convincingly served as pollinator as its head touched flower reproductive organs. Notes: pictures of E. spelaea were enlarged relative to those of other species to show details. Photos courtesy of author.

44

Figure 2-8. Temporal overlap of foraging activity on durian between pairs of species. The gray shading indicates the time when pairs of species were both actively foraging at durian flowers. The y-axis shows the kernel density estimates.

45

Figure 2-9. Strigocuscus celebensis visited a durian inflorescence and consumed the flowers. A series of interactions included (a) grabbing a flower; (b) pulling a flower; (c) eating the stigma and anthers; and (d) chewing on flower parts. Photos courtesy of author.

46

Figure 2-10. Other durian visitors included a (a) male and (b) female of Aethopyga siparaja that stuck their beaks into the base of durian flowers to drink nectar at a time of day when flowers were not fully open; (c) a bee (yellow circle) came out from the base of a durian flower during the night; (d) a bee looked for remaining nectar shortly after sunrise. Photos courtesy of author.

47

CHAPTER 3 ECOSYSTEM SERVICES PROVIDED BY ACERODON CELEBENSIS AND PTEROPUS GRISEUS IN SULAWESI, INDONESIA

Background

Flying foxes (Pteropus and Acerodon) provide ecosystem services in tropical forests. These bats feed mainly on fruits, playing a critical role as seed dispersers

(Marshall 1985, Fujita & Tuttle 1991). At least 289 plant species rely on flying foxes to disperse their seeds (Fujita & Tuttle 1991). In addition to dispersing seeds, flying foxes enhance germination rates of those seeds that are processed by their digestive systems

(Oleksy, Giuggioli, et al. 2015). Floral products (e.g. nectar) are crucial diet components of flying foxes as well (Banack 1998), suggesting that another important ecological role of these bats is as pollinators. Through mutualistic relationships with flowering plants, flying foxes and other pteropodid bats act as pollinators of about 168 species in 41 plant families (Fujita & Tuttle 1991). These services are indispensable for the propagation of native plants and production of economically important cash crops particularly in the

Old-World tropics (Fujita & Tuttle 1991, Muscarella & Fleming 2007). Moreover, the long distances bats fly relative to other seed dispersing and pollinating organisms make them exceptionally important in both large intact forests and fragmented landscapes (Luskin

2010, Roberts et al. 2012).

Indonesia is a country with unusually high diversity of pteropodid bats (75 species) (Suyanto et al. 2002). Sulawesi has especially high fruit bat diversity (28 species) (Mickleburgh et al. 1992, Suyanto et al. 2002), and is particularly well suited to an investigation of the ecological role of flying foxes because of the small number of other medium to large phytophagous mammals there. The lack of competition, especially with primates, and high flying fox diversity suggests that flying foxes likely

48

contribute heavily to seed dispersal and pollination services on this island. Since many flying foxes have a broad generalist diet, many ecologically and economically important plants potentially depend on them (Fujita & Tuttle 1991, Fleming & Sosa 1994).

Another reason that Sulawesi is well suited for research on flying fox ecosystem services is because these bats are threatened by intense hunting and habitat loss

(Mickleburgh et al. 2009, Sheherazade & Tsang 2015). Many flying fox colonies in

Sulawesi are under huge pressure of hunting to supply bushmeat to North Sulawesi markets (Sheherazade & Tsang 2015). The decline of flying fox populations may negatively impact the reproductive success of plant species that depend on them

(McConkey & Drake 2006). Scientific data on flying fox ecosystem services in this region are needed to guide government efforts to conserve flying foxes. Tangible examples of flying fox ecosystem services may also help change attitudes and engage local communities in long-term bat and forest conservation (Scanlon et al. 2014)

Our study aims to understand the diet of two species of flying foxes by identifying the pollen on the flying fox fur. Because some tropical plants flower and fruit simultaneously, pollen could suggest which plants flying foxes visit both for floral and fruit rewards. The two flying foxes we investigated are 1) an endemic Sulawesi flying fox

(Acerodon celebensis) that occurs only in Sulawesi, and 2) a relatively widespread species, Gray flying fox (Pteropus griseus) that is found throughout Sulawesi, adjacent islands, and lesser Sunda. Knowing the diet of flying foxes will help us to better understand the ecosystem services they provide. Our research objectives are:

1. To understand resource use of flying foxes

2. To characterize ecosystem services provided by flying foxes, and describe the bats’ importance to Sulawesi rainforests and local livelihoods.

49

Methods

Study Sites

We conducted our study on two small islands in Bualemo subdistrict, Central

Sulawesi in June-July (dry season) and October-December 2017 (rainy season) (Figure

3-1). These two islands were 20 km apart and located in Local Sea Conservation Area

(Kawasan Konservasi Laut Daerah, KKLD). Each island had one colony of flying foxes.

The first island, Mantalu Daka or Besar Island, was ~5 km from the shoreline of

Longkoga Barat/Timur village. Three species of flying foxes roosted on this island:

Pteropus griseus, Acerodon celebensis, and Pteropus alecto. We focused our study on

Pteropus griseus because this was the only species that flew at a height that we were able to intercept. The island vegetation was dominated by mangroves and spiny forests, and also contained several coconut plantations. There was a swamp in the middle of the island where the bats roosted. The second island, Tangkuladi Island, was ~1.5 km from the shoreline of Taima village. It had one mixed colony of flying foxes that contained Acerodon celebensis and Pteropus alecto. We focused our study on

Acerodon celebensis for the same reason described previously. Tangkuladi Island was comprised entirely of mangroves and spiny forest.

Pollen Collection from Flying Foxes

We captured flying foxes following the procedures described in Methods of

Capturing and Handling Bats (Kunz et al. 2009), with slight modifications to reduce disturbance caused by the capturing activities (Sigit Wiantoro and Susan M. Tsang, pers. comm., January 2017). We set up mist nets to capture bats as they returned to their roost after foraging. We set up two stacked mist nets (Avinet 2.6 x 9 m, 38 mm mesh size, 4 shelves) at a height of 3.7 to 7.3 meters above the ground (Forest Filter

50

mist net pole system, Bat Conservation and Management) to capture flying foxes. We placed the mist nets along the coastline of each island, located >500m from each roost to prevent disturbance to the colony. The mist nets were opened from 0200 to 0600.

During the dry season, we captured Pteropus griseus in June. We set up mist nets for four consecutive nights per week. We had 16 capture nights for four weeks in a row. Then, we captured Acerodon celebensis in July. Due to stricter regulations to capture bats on this island, we only set up mist nets for four consecutive nights. During the rainy season, we captured Acerodon celebensis in October for two consecutive nights with two-day intervals between captures for 16 capture nights total. After that, we captured Pteropus griseus in the end of November to early December. Due to extreme weather (e.g. big waves and storms), we only set up mist nets for six consecutive nights. The position of the mist nets was adjusted and varied every two or three nights to increase the probability of capturing bats and prevent shifts in the flight path flying foxes used to go back to their roost.

We collected the pollen immediately after disentangling each bat to prevent pollen loss. We followed the method described by Bumrungsri et al. (2013) and Stewart

& Dudash (2016) for pollen collection and analysis. We used a gelatinous cube to swab bat fur and collect all the pollen from each bat. The cube was made of 0.1 mL gel of mixed glycerin, fuchsin, and gelatin, enabling pollen coloring and preservation. Each cube was then fixed on a slide, heated and closed by a cover slide. Each slide was labeled with the field identification number for the bat. We weighed and measured each bat’s morphological characteristics (forearm length, ear length, hind foot length, maximum skull length, and weight). We recorded the sex (male, female) and age (adult,

51

juvenile) of each bat captured. After that, we put one small dot of nontoxic nail polish on one bat claw to allow recognition of individuals that were recaptured. Finally, after we finished all the processes, we released all the bats. All protocols were approved by the

University of Florida under IACUC protocol No. 201709800.

Resource Use of Flying Foxes

To determine flying fox diet, we examined the pollen samples from the bat fur using a light microscope. Pollen was categorized into different morphotypes and then identified to species level by comparing it with available literature and reference libraries, such as Australasian Pollen and Spore Atlas (http://apsa.anu.edu.au/), (Start

1974), the herbarium at the Indonesian Institute of Sciences (LIPI), and SEABCRU database. We determined pollen load, PL, by counting the number of pollen grains for each pollen morphotype using a hand counter. To understand diet composition of flying foxes, we calculated percentage of observations for each plant, which was the number of bats that carried pollen of a particular plant divided by the total number of bats then multiplied by 100. We also observed the temporal pattern of resource used by noting how frequently individual bats carried only a single species of pollen and how long they used a particular plant species before changing to another one.

In addition, we estimated diet breadth of flying foxes using this formula below:

Levins’ standardized diet breadth (퐵̂퐴) (Levins 1968):

퐵̂ − 1 퐵̂ = 퐴 푛 − 1 (3-1) where:

1 ̂ 퐵 = 2 (3-2) ∑ 푝̂푗

52

퐵̂ = Levins’ measure of diet breadth

푛 = Number of possible resource states

푝̂푗 = Proportion of individuals found in or using resource state j (∑ 푝̂푗 = 1.0)

The scale of diet breadth is 0-1, with values closer to 1 indicating broader diet breadth.

We calculated diet overlap between Pteropus griseus and Acerodon celebensis using this formula below:

Morisita’s index of diet overlap (Morisita 1959) between species j (P. griseus) and k (A. celebensis):

2 ∑푛 푝 푝 = 푖 푖푗 푖푘 (푛 − 1) (푛 − 1) ∑푛 [ 푖푗 ] ∑푛 푖푘 (3-3) 푖 푝푖푗 + 푖 푝푖푘 [( )] (푁푗 − 1) 푁푘 − 1 where pij = proportion of plant species i used by species j pik = proportion of plant species j used by species k nij = number of individuals of species j that use plant species i nik = number of individuals of species k that use plant species i

∑nij = Nj, ∑nik = Nk

We used pollen load (number of pollen grain collected for each plant species) to calculate the proportion of plant species used by each bat species. The Morista measure of overlap ranges from 0 (no resources used in common) to 1.0 (complete overlap).

53

Characterization of Ecosystem Services Provided by Flying Foxes

The presence of pollen on a mobile organism is often interpreted as an indication of pollination services provided by that animal. Because some tropical plants flower and fruit simultaneously, the presence of pollen on flying foxes may also indicate fruits eaten and thus potential seed dispersal services provided by the flying foxes.

To differentiate whether flying foxes were likely serving as pollinators or seed dispersers for each plant species visited, we first determined which parts of plants were used by flying foxes: the flowers, fruits, or both. To do this, we conducted a literature search to determine whether each plant species for which we found pollen had simultaneous flowering and fruiting, and whether fruit or nectar was previously reported to be part of flying fox diets. Key references were Fleming et al. (2009), Fleming &

Muchhala (2008), Banack (1998), van der Pijl (1956), and Cox et al. (1992).

Secondly, we determined whether the morphology of flowers and fruits conformed to chiropterophilous syndromes and thus had the potential to be exploited by flying foxes. For example, pollen from Sonneratia alba was collected from Pteropus griseus fur in this study. Published research indicated that S. alba was used extensively by pteropodid bats for its nectar, and that it did not flower and fruit at the same time, and the flowers demonstrated chiropterophilous features, such as night-blooming behavior, dull-color, and located at the outer part of the trees making them easily accessible to bats. Based on these criteria, we concluded that Pteropus griseus most likely used floral parts of S. alba and indicated pollination services were most likely for this mangrove species. Another example was pollen of Palaquium collected from Acerodon celebensis fur in this study. In the literature, the only use reported for Palaquium by flying foxes was for its fleshy fruits, this plant is reported to fruit and flower simultaneously, and

54

Palaquium’s flowers were so small that nectar could not be exploited by large flying foxes. We therefore concluded that bats likely got pollen on their fur as they visited

Palaquium trees to obtain fruit, and therefore indicated that seed dispersal services were most likely. We conducted these steps for every plant species identified in this study.

We then determined the human use of each plant species the bats visited to describe the value of flying fox’s ecosystem services to people. We used literature and our experiences interacting with the local community to know these uses and values.

Results

We captured 85 flying foxes (52 individuals of Pteropus griseus and 33 individuals of Acerodon celebensis) during 168 mist net hours. These two species were distinguished by fur color, weight, and morphological characteristics. Pteropus griseus was light to dark orange with a gray head. In contrast, A. celebensis was blonde with a brown to black face. It was larger and heavier than P. griseus (Table 3-1).

Resource Use of Flying Foxes

About 83% of flying foxes captured had pollen grains in their fur. It was less common for P. griseus not to carry any pollen (three individuals during the rainy season, one during the dry season), than for A. celebensis [14 individuals had no pollen (all during the rainy season)].

We identified 14 species and one family of plants used by flying foxes (Figure 3-

2). More than half of these plants (60%) were exploited by flying foxes for fruits, while

20% were used for flowers, and 20% for either fruits or flowers.

Pteropus griseus had a slightly wider diet breadth (퐵̂퐴: 0.33) than A. celebensis

(퐵̂퐴: 0.32), but their diet overlap was very low (C: 0.15). Pteropus griseus used 11 food

55

plant species; seven plant species during each season (Figure 3-3). Only three plant species were used during both seasons: C. nucifera, Sonneratia alba (mangroves or bakau), and Garcinia sp. (saptrees or manggis hutan). During the dry season, P. griseus fed on the same set of plants for the first and second week of the month:

Neonauclea excelsa, C. nucifera, and Syzygium sp. Then these bats added more plant species into their diet during the following weeks: Palaquium sp., Garcinia sp. and S. alba. In contrast, during the wet season, P. griseus used mostly S. alba flowers. The bats switched their diet from C. nucifera to S. alba after a few days. We frequently found pollen from two to three plant species on each bat every night during the dry season compared to only one plant species during the wet season.

In comparison, A. celebensis used six plant species, mostly for fruits (75%).

These bats exploited a different set of plants during each season. During the dry season, A. celebensis relied heavily on Ficus spp. (figs), but also used other resources used by P. griseus: N. excelsa, C. nucifera, and Garcinia sp. (Figure 3-4). Then, these bats used fruits of Palaquium sp. and flower resources of Duabanga moluccana during the wet season. We commonly found pollen of one plant species on each bat every night. The diet typically changed every few days to a week.

Ecosystem Services of Flying Foxes

Most of the plants used by flying foxes were native species in the forests (e.g.

Ficus and Palaquium) and mangroves (e.g. Duabanga moluccana and Sonneratia alba).

While more than half of the native plants were exploited for fruits, mangrove species were used by flying foxes for their nectar. Therefore, flying foxes primarily provided seed dispersal services for plants that grew in forests and pollination services for plants that grew in mangroves.

56

The primary human use of half of the plant species visited by flying foxes were as sources of timber for the local community (Table 3-2). Most were naturally occurring; only a few were purposefully grown in local plantations, such as Cocos nucifera

(Coconut trees or Pohon Kelapa in Bahasa Indonesia) and Carica papaya (Papaya or

Pepaya). Two of these species grown in plantations were highly valued by the locals as sources of edible fruits. Coconut tree had the highest human economic value, as almost all of its parts were utilized by locals.

Discussion

Our study provides the first scientific evidence regarding flying fox diet composition in Sulawesi by identifying pollen collected from the animal fur. These findings allow a deeper understanding of resource use of flying foxes, and highlight the importance of ecosystem services provided by bats on this island. These services include pollination and seed dispersal for native plants, which are essential components of Sulawesi ecosystems, and also economically valuable to the local human community.

The two flying fox species in our study sites, Pteropus griseus and Acerodon celebensis, are considered generalists, suggesting that they use many plant species.

Our results provide additional proof that flying foxes have a wide diet breadth and are able to exploit a variety of plants (Banack 1998, Nakamoto et al. 2015). However, the total number of food plants in the study site is considerably fewer than has been recorded in other places. Pteropus giganteus in India and Pteropys lylei in Thailand use

24 and 34 species of plants respectively, more than double the number of plants used by flying foxes in our study (Weber et al. 2015, Rao 2018). This discrepancy may be due to data collection methodology: we only used pollen to infer diet, while other research also collected faeces and spats or pellets to determine diet; and they collected

57

data for longer duration than we did. Future studies of diet breadth of P. griseus and A. celebensis may find it to be wider than we estimated in this study if additional dietary evidence is evaluated over a longer study period (Figure 3-5).

The diverse diet of flying foxes implies that they provide ecosystem services for various plants in Sulawesi. The services change seasonally as flying fox diets shift.

Flying foxes in Sulawesi use the same resources for a certain period of time then move to other fruiting or flowering plants. Previous research shows that pteropodid bats use some plants more extensively than others, but add more items to their diet when the main resources are depleted (Bumrungsri et al. 2013, Lim et al. 2018, Chen et al. 2017,

Weber et al. 2015, Banack 1998, Brooke 2001). Narrowing diet breadth to focus on abundant resources at a particular time is a strategy animals use to reduce the amount of energy needed to travel finding new and patchy resources (Páez et al. 2018). Flying foxes depend on spatio-temporal stability of food resources, thus they can flexibly change their diet according to food conditions (Nakamoto et al. 2015).

Ecosystem Services: Cultivated Crops

Several of the plants that benefit from flying foxes are economically important for the local people. The main livelihood of local people in our study region is farming. The landscape is dominated by coconut plantations, particularly in the coastal areas.

Coconut seems to be one of the main diet components of flying foxes, but we are still uncertain whether the flying foxes function as legitimate pollinators. Locals in Sulawesi report that flying foxes feed on coconut flowers, which is not a surprise as flying foxes are known to not only drink nectar, but also consume pollen and intact flowers

(Nakamoto et al. 2015). Other studies have found that coconut constitutes pteropodid diet (Pimsai et al. 2014, Abdul-Aziz, Clements, Peng, et al. 2017, Lim et al. 2018, Start

58

& Marshall 1976), but they do not explicitly state that bats contribute to coconut pollination. The small flower structure of coconut makes bat pollination seem unlikely, and a previous study suggests that coconut primarily depends on wind and bee pollination (Thomas & Josephrajkumar 2013). Flying foxes may accidentally pollinate a portion of coconut flowers while feeding on others. There is also a possibility that flying foxes may function as florivores, destroying inflorescences, disrupting pollination, and causing crop loss, but locals have never linked flying fox visitation to low coconut production. Further investigation into the relationship between flying foxes and coconut flowers is warranted, as coconut plays a substantial role in local economy and

Indonesian agricultural sector. Sulawesi is the major producer for copra, processed coconut fruits sold in the markets (Direktorat Jenderal Perkebunan (Directorate General of Estate Crops) 2017). In addition, coconut is culturally valuable for several reasons. Its leaves are used to make a container to cook rice during Moslem celebration days

(called Ketupat), and the boles of trees are used to build bridges and houses.

Locals may also benefit from flying foxes dispersing seed of papaya. We found that these fruits were eaten frequently during the rainy season, as has been reported in other places (Hengjan et al. 2018, Rao 2018). Local individuals in our study region did not express any contention toward flying foxes feeding on their papaya fruits. Rather, they expressed gratitude that they had many papaya trees in their backyards and farms without having invested labor to plant them. The seed dispersal service provided by flying foxes seems to compensate for the loss of a few papaya fruits (one or two papaya eaten per night). Most papaya is consumed locally; little is traded in the markets. Lack

59

of significant economic value of this fruit makes the conflict between farmer and bats unlikely.

Ecosystem Services: Native Plants

Flying foxes also benefit native plants in their natural habitat. They act as potential seed dispersers for forest plants, such as Neonauclea excelsa, Garcinia sp., and Palaquium sp. These species are essential components of tropical mountain forest communities in Sulawesi (Brambach et al. 2017). They are previously reported to be major components of flying fox diet in other areas, notably other Pacific islands (Banack

1998, Scanlon et al. 2014). Our findings add additional evidence that flying foxes provide important seed dispersal services to tropical forest plants. Future research that identifies pollen transported by flying foxes at greater taxonomic resolution may reveal that flying foxes have even more crucial roles than we were able to identify with our resolution to genera. There are 14 species of Garcinia and 9 species of Palaquium, some of which are endemic to Sulawesi (Keßler et al. 2002). Future research should explore whether flying foxes disperse seeds of endemic species.

Neonauclea excelsa, Garcinia sp., and Palaquium sp, are also very valuable to the local community as sources of timber. Trees are harvested at a sustainable rate only when the locals need them to build houses. Neonauclea excelsa is frequently used to make components of boats and houses. Similar to papaya’s case above, it is not necessary for the locals to keep planting the trees because flying foxes eat the fruits and help disperse seed. Additionally, flying foxes may provide pollination services to this tree because it has fruits and flowers at the same time. The tree had the highest amount of pollen (total number of pollen grains) on flying fox fur in our study. Considering the flower structure (Fig. 2.2), it seems unlikely that flying foxes purposefully visit plants to

60

feed on the nectar, but the high number of pollen grains we documented them transporting suggests they touch the flowers by chance while feeding on the fruits.

Pollen attached to the fur may then be brought to flowers located near other fruits.

Flying foxes are also known to disperse seeds of Garcinia and Palaquium species on other Pacific islands (Scanlon et al. 2014).

Ecosystem Services: Unique Contributions

Interestingly, low diet overlap between two species suggests that some services are unique to each flying fox species. Pteropus griseus and Acerodon celebensis have different diet composition, suggesting that they provide complementary ecosystem services. As an example, in regard to pollination services, both species pollinate mangroves, but P. griseus pollinates Sonneratia alba while Acerodon celebensis pollinates Duabanga moluccana. These are new records for flying foxes pollinating mangroves, because in other places of Southeast Asia, these plants are primarily pollinated by small pteropodid bats, such as Eonycteris spelaea and Macroglossus

(Bumrungsri et al. 2013, Lim et al. 2018, Start & Marshall 1976).

Although the low diet overlap may have been caused by the slight difference in time of capturing the bats, co-existing flying foxes are known to show disparate diet and habitat uses to moderate competition (Banack 1998, Stier & Mildenstein 2005). For instance, on Comoro Island, Pteropus livingstonii is highly dependent on forests and uses a limited number of plant species for food compared to P. seychellensis comorensis which is a habitat generalist and consumes a broader variety of plants

(Ibouroi et al. 2018). Moreover, in systems with two sympatric flying fox species, endemic flying foxes typically have a narrower nectar diet breadth than widespread species (Stier & Mildenstein 2005). For example, in the Philippines, the Philippine

61

endemic flying foxes, Acerodon jubatus, has narrower diet breadth than the relatively more widespread species, Pteropus vampyrus lanensis (Stier & Mildenstein 2005). Of the bats in our study, Pteropus griseus has a wider distribution (Sulawesi and Lesser

Sunda) than A. celebensis (endemic to Sulawesi). The former species has a slightly more diverse diet than the latter one in terms of plant species and type of resources

(fruits, flowers, and both). Moreover, we believe Acerodon celebensis functions more as a seed disperser than Pteropus griseus because individuals of the former species were more often captured without pollen. Conversely, this species may serve as a more efficient pollinator for those species it visits for pollen, since bats with pollen typically carried pollen from only one species, which may increase the likelihood of pollen being transferred to another plant of the same species rather than to one of a different species.

As P. griseus is more of a generalist, it can benefit many plant species. In contrast, although A. celebensis has a less diverse diet and thus benefits fewer plant species, some of the plants visited by these bats (Palaquium and Ficus) highly depend on A. celebensis for seed dispersal. These two genera were the most commonly found on A. celebensis, and are common tree species in Sulawesi forests (Brambach et al.

2017). Our findings also corroborate the role of flying foxes as seed dispersers for figs, as has been shown by other studies (Lee et al. 2017, Abdul-Aziz, Clements, Peng, et al.

2017, Rao 2018, Stier & Mildenstein 2005, Bollen & Van Elsacker 2002, Oleksy, Racey, et al. 2015, Hengjan et al. 2018, Javid et al. 2017), but this is the first case for Sulawesi,

Indonesia. Figs are keystone species, and some are pioneer species important for forest succession and regeneration (Chen et al. 2017, Oleksy, Giuggioli, et al. 2015).

62

Understanding that A. celebensis may play important roles in this forest dynamic gives a strong case to protect this species. Unfortunately, we could not identify the figs to species level due to lack of pollen references for the 37 fig species in Sulawesi (Keßler et al. 2002).

Our discoveries suggest that flying fox diversity is necessary to ensure the provision of different types of ecosystem services. Therefore, efforts to protect flying foxes must account for multiple species. The major limitations of our research are that we used only pollen to infer flying fox diet and used visual observation to identify pollen species. Future studies should investigate feces as well as pellets or spats, and use more modern plant identification approaches, such as metabarcoding. However, the current lack of a reference library in Sulawesi makes approaches reliant on DNA infeasible at this time. Building pollen, seed, and DNA barcode database for plants in

Sulawesi should be priorities for elucidating the ecosystem services provided by not only flying foxes, but also other frugivorous and nectarivorous animals in the region. In the meantime, our research should be viewed as a conservative characterization of flying fox ecosystem services in Sulawesi. We have demonstrated the link between flying fox populations, native plants, and the local economy.

Conservation of flying foxes in Sulawesi should be promoted to prevent further loss of productivity of plant species that rely on them. Outreach programs need to emphasize tangible examples of the roles that flying foxes provide to forests and livelihoods. This can allow people to feel more connected to bats and develop a positive attitude toward them. Protecting bat roost sites from hunting should be conducted at a local level so that people have a sense of community and obligation to participate.

63

Because most flying fox species are not included in the Indonesian protected species list, locals must establish their own village regulations to manage hunting. Organizing workshops where locals can learn and practice to enact and enforce the regulations they agree upon can equip them with necessary skills and empower them, so they are able to contribute when hunting pressure begins. All these aspects allow people to develop a good intention toward saving bats and make them more likely to engage in conservation and behave accordingly (Ajzen 1991). This is essential to the efforts of reducing hunting threats toward flying foxes in the region, so that locals continue to reap the ecosystem services these animals provide.

64

Table 3-1. Morphometric measurement of a subset of the flying foxes found in the study sites: 18 adults of A. celebensis and 16 adults of P. griseus. Species Weight (g) FA (mm) Ear (mm) GSL (mm) HF (mm)

Pteropus griseus 292.13  9.73 (SE) 126.06  1.01 (SE) 24.05  0.70 (SE) 60.79  0.90 (SE) 56.71  0.90 (SE) (range: 230-370) (range: 116.84-131.65) (range: 20.32-30.46) (range: 55.88-70.15) (range: 50.2-63.5) Acerodon 435.83  11.88 (SE) 137.87  0.92 (SE) 28.64  0.62 (SE) 66.13  1.05 (SE) 61.17  0.85 (SE) celebensis (range: 350-530) (range: 134.46-150.03) (range: 22.86-32.87) (range: 58.04-76.65) (range: 53.53-68.33) FA: Forearm, GSL: Greatest skull length, HF: Hindfoot.

65

Table 3-2. Plants visited by flying foxes, the resources flying foxes used (flowers, fruits), the pollen load, and the value of each plant to humans in the local community. Resources Pollen Flying fox used by flying Family Plant species Habitat Human values loads species foxes Flowers Arecaceae Cocos nucifera Plantation 4130 Pgr, Ace Fruits sold as copra; leaves used for local food container (Ketupat); timber Flowers Lythraceae Duabanga moluccana Mangrove 33 Ace Timber

Flowers Sonneratia alba Mangrove 70 Pgr Timber Fruits Burseraceae Canarium sp. Forest 51 Pgr Timber Fruits Caricaceae Carica papaya Plantation 341 Pgr Edible fruits, flowers, and leaves Fruits Clusiaceae Garcinia sp. Forest 7 Pgr, Ace Timber Fruits Ebenaceae Diospyros sp. Forest 22 Pgr Timber Fruits Moraceae Ficus sp.1 Forest 13 Ace Unknown

Fruits Ficus sp.2 Forest 83 Ace Unknown

Fruits Ficus sp.3 Forest 4 Ace Unkown Fruits Rubiaceae Neonauclea excelsa Forest 4295 Pgr, Ace Timber Fruits Sapotaceae Palaquium sp. Forest 431 Pgr, Ace Timber Fruits/flowers Lamiaceae Unknown Unknown 6 Pgr Unknown Fruits/flowers Myrtaceae Syzygium sp. Unknown 159 Pgr Edible fruits Fruits/flowers Pandanaceae Freycinetia sp. Forest 2 Pgr Unknown Pgr: Pteropus griseus; Ace: Acerodon celebensis

66

Figure 3-1. The two study sites were islands off the coast of Central Sulawesi in Bualemo: Mantalu Daka Island and Tangkuladi Island.

67

Figure 3-2. Some examples of plants used by flying foxes. Pollen was documented using a light microscope with a magnification of 40 x 0.65. Photos of S. alba flowers by Ton Rulkens © Creative Commons; S. alba trees by Ce Nuevo; Palaquium tree by Giuseppe Mazza.

68

Figure 3-3. Diet composition of Pteropus griseus during dry and rainy seasons.

69

Figure 3-4. Diet composition of Acerodon celebensis during dry and rainy seasons.

70

Figure 3-5. Species accumulation curve for plant species used by flying foxes.

71

CHAPTER 4 CONCLUSION

Our study primarily aimed to investigate the ecosystem services provided by pteropodid bats, with special attention to flying foxes (Pteropus and Acerodon), in

Sulawesi, Indonesia. Despite the high diversity and endemicity of pteropodid bats, this region lacks ecological studies about them. The paucity of data regarding these animals hinders our understanding of their ecological roles and their interaction with plant species in the community. Our study is the first in Indonesia that comprehensively explores the importance of bats in providing ecosystem services.

In the first chapter, using an experiment and camera traps, we found that bats are the primary pollinators of durian, a major fruit commodity in Indonesia and

Southeast Asia. Flowers visited by bats produced a higher number of durian fruits compared to flowers visited by insects or not visited at all. Our study also provides the first evidence of the role of two flying foxes, P. alecto and A. celebensis, as durian pollinators. Knowing that bats, including flying foxes, contribute to pollination of semi- wild durian is pivotal to gain more appreciation toward these animals in a region where durian is prominent to local livelihoods both economically and culturally.

In the second chapter, we focused more broadly on identifying ecosystem services provided by two flying foxes: P. griseus and A. celebensis. We collected pollen from the bat fur and then identified this pollen to plant species. We found that flying foxes functioned primarily as seed dispersers for native plants in the forests, while acting as pollinators for plant species found in mangroves. Many of these plants the bats visited were essential components of Sulawesi rainforests and valued by the local community as sources of timber. Each species of flying fox visited different plant

72

species. Thus, each species of flying fox provided unique ecosystem services for ecologically and economically important plants in Sulawesi, Indonesia.

Our findings will be used to foster conservation of pteropodid bats, particularly flying foxes. They have been hunted intensely in Sulawesi for bushmeat. Flying foxes are not protected under Indonesian list of protected species. Lack of regulation and scientific data prevents flying foxes from receiving conservation attention. Our studies highlight the importance of these animals to the ecosystem and human well-being on the island by providing specific examples of benefits bats provide to forests, mangroves, and people. Elucidating the link between bats and resources that are relevant and close to local people will help us to reduce the hunting and consumption threat. Moreover, the conservation of bats should incorporate in situ protection of bat roosts (e.g. mangroves, which are the primary roosts for flying foxes) and foraging areas (e.g. primary forests and mixed plantations), target bushmeat hunting in North Sulawesi, legislation for bat protection and hunting quotas, and outreach programs to raise awareness about the importance of bats. We hope to use our findings to generate more interest in fruit bats and their contributions to the ecosystem to reduce human prejudice against bats.

73

APPENDIX A A LIST OF ANIMAL VISITORS AND PRIMARY POLLINATORS OF SEMI-WILD DURIAN THROUGHOUT SOUTHEAST ASIA.

Table A-1. A list of animal visitors and primary pollinators of semi-wild durian throughout Southeast Asia. Durian species Study sites Documented visitors Primary pollinators

Durio zibethinusa Thailand Eonycteris spelaea (common nectar bat)  Nectariidae (flowerpecker bird) Apis dorsata (giant honey bee) Coleoptera (beetle) Lepidoptera (moth) Durio zibethinusb Thailand Eonycteris spelaea  Rousettus leschenaultii (leschenault's rousette)  Durio zibethinusc Malaysia Eonycteris spelaea Pteropus vampyrus (large flying fox) Durio zibethinusd Malaysia Eonycteris spelaea Macroglossus (long-tongued nectar bat) Cynopterus (fruit bat) Pteropus (flying fox) Lepidoptera Durio zibethinuse Malaysia Eonycteris spelaea Durio zibethinusf Malaysia Eonycteris spelaea  Pteropus hypomelanus  Callosciurus notatus (plantain squirrel) Apis dorsata Lepidoptera Arachnothera robusta (long-billed Durio grandiflorusg Malaysia spiderhunter) 

74

Table A-1. Continued. Durian species Study sites Documented visitors Primary pollinators Arachnothera chrysogenys (yellow-eared spiderhunter)  Troides amphrysus (malay birdwing butterfly) Trigona (stingless bee) Durio oblongusg Malaysia Arachnothera robusta  Arachnothera chrysogenys  Arachnothera flavigaster  Sphingidae Durio kutejensisg Malaysia Eonycteris spelaea  Arachnothera robusta  Arachnothera flavigaster  Arachnothera longirostra (little spiderhunter)  Anthreptes simplex (plain sunbird)  Apis dorsata  Trigona  Durio griffithiih Malaysia Insecta (insect)

75

APPENDIX B RESULTS OF GENERALIZED LINEAR MIXED-EFFECT MODELS (GLMM) TO DETERMINE WHICH FACTORS CONTRIBUTE MOST TO DURIAN PRODUCTION

Table B-1. Results of Generalized Linear Mixed-effect Models (GLMM) to determine which factors contribute most to durian production. Days Model Estimate Std. Error z value Pr(>|z|) Day 20 (Intercept) 1.51E+00 3.73E-01 4.054 5.03e-05 *** bat.visitors -5.92E-02 1.34E+00 -0.044 0.965 individual -2.38E-02 1.00E-01 -0.238 0.812 duration 1.00E-04 2.34E-03 0.043 0.966 bat.visitors:individual 3.13E-01 3.84E-01 0.815 0.415 bat.visitors:duration -4.21E-02 7.15E-02 -0.589 0.556 individual:duration 3.25E-05 2.38E-04 0.137 0.891 bat.visitors:individual:duration 3.02E-03 5.14E-03 0.587 0.557 Day 60 (Intercept) -0.230813 0.451568 -0.511 0.6093 bat.visitors -1.403705 1.597906 -0.878 0.3797 individual 0.074704 0.104807 0.713 0.476 duration 0.002502 0.002362 1.059 0.2895 bat.visitors:individual 0.883674 0.51919 1.702 0.0888 bat.visitors:duration -0.120864 0.093704 -1.29 0.1971 individual:duration -0.000262 0.000241 -1.087 0.2771 bat.visitors:individual:duration 0.008521 0.006725 1.267 0.2052

76

LIST OF REFERENCES

ABDUL-AZIZ, S. 2016. Interactions between flying foxes, plants and people: implications for conservation. Museum National D’Histoire Naturelle.

ABDUL-AZIZ, S., G. R. CLEMENTS, K. R. MCCONKEY, T. SRITONGCHUAY, S. PATHIL, M. N. H. ABU YAZID, A. CAMPOS-ARCEIZ, P.-M. FORGET, and S. BUMRUNGSRI. 2017. Pollination by the locally endangered island flying fox (Pteropus hypomelanus) enhances fruit production of the economically important durian (Durio zibethinus). Ecol. Evol. 7: 8670–8684.

ABDUL-AZIZ, S., G. R. CLEMENTS, L. Y. PENG, A. CAMPOS-ARCEIZ, K. R. MCCONKEY, P.-M. FORGET, and H. M. GAN. 2017. Elucidating the diet of the island flying fox (Pteropus hypomelanus) in Peninsular Malaysia through Illumina Next-Generation Sequencing. PeerJ 5: e3176.

ACHARYA, P. R., P. A. RACEY, S. SOTTHIBANDHU, and S. BUMRUNGSRI. 2015. Feeding behaviour of the dawn bat (Eonycteris spelaea) promotes cross pollination of economically important plants in Southeast Asia. J. Pollinat. Ecol. 15: 44–50.

AJZEN, I. 1991. The theory of planned behavior. Organ. Behav. Hum. Decis. Process. 50: 179–211.

ALLEN-WARDELL, G. ET AL. 1998. The potential consequences of pollinator declines on the conservation of biodiversity and stability of fruit crop yields. Conserv. Biol. 12: 8–17.

BADAN PUSAT STATISTIK KABUPATEN POLEWALI MANDAR. 2017a. Kabupaten Polewali Mandar dalam Angka (Polewali Mandar Regency in Figures) 2017. Polewali.

BADAN PUSAT STATISTIK KABUPATEN POLEWALI MANDAR. 2017b. Kecamatan Binuang dalam Angka 2017 (Binuang Subdistrict in Figures 2017). Polewali.

BAKER, H. G., I. BAKER, and S. A. HODGES. 1998. Sugar composition of nectars and fruits consumed by birds and bats in the tropics and subtropics. Biotropica 30: 559–586.

BANACK, S. A. 1998. Diet selection and resource use by flying foxes (Genus Pteropus). Ecology 79: 1949–1967.

BANACK, S. A., and G. S. GRANT. 2002. Spatial and temporal movement patterns of the flying fox, Pteropus tonganus, in American Samoa. J. Wildl. Manage. 66: 1154.

BILDER, C. R., and T. M. LOUGHIN. 2015. Analysis of Categorical Data with R. CRC Press Taylor and Francis Group, Boca Raton.

BIRT, P. 2004. Mutualistic interactions between the nectar-feeding little red flying-fox Pteropus scapulatus (Chiroptera: Pteropodidae) and flowering eucalypts (Myrtaceae): habitat utilisation and pollination. The University of Queensland.

77

BOLLEN, A., and L. VAN ELSACKER. 2002. Feeding ecology of Pteropus rufus (Pteropodidae) in the littoral forest of Sainte Luce, SE Madagascar. Acta Chiropterologica 4: 33–47.

BRAMBACH, F., C. LEUSCHNER, A. TJOA, and H. CULMSEE. 2017. Diversity, endemism, and composition of tropical mountain forest communities in Sulawesi, Indonesia, in relation to elevation and soil properties. Perspect. Plant Ecol. Evol. Syst. 27: 68–79.

BROOKE, A. P. 2001. Population status and behaviours of the Samoan flying fox (Pteropus samoensis) on Tutuila Island, American Samoa. J. Zool. 254: 309–319.

BUMRUNGSRI, S., D. LANG, C. HARROWER, E. SRIPAORAYA, K. KITPIPIT, and P. A. RACEY. 2013. The dawn bat, Eonycteris spelaea Dobson (Chiroptera : Pteropodidae) feeds mainly on pollen of economically important food plants in Thailand. Acta Chiropterologica 15: 95–104.

BUMRUNGSRI, S., E. SRIPAORAYA, T. CHONGSIRI, K. SRIDITH, and P. A. RACEY. 2009. The pollination ecology of durian (Durio zibethinus, Bombacaceae) in southern Thailand. J. Trop. Ecol. 25: 85–92.

CALCAGNO, V. 2015. Package “glmulti.” CRAN Repos. 1–20. Available at: https://cran.r- project.org/web/packages/glmulti/glmulti.pdf [Accessed February 6, 2018].

CHEN, S. F., T. J. SHEN, H. C. LEE, H. W. WU, W. T. ZENG, D. J. LU, and H. C. LIN. 2017. Preference of an insular flying fox for seed figs enhances seed dispersal of a dioecious species. Biotropica 49: 511–520.

COX, P. A., T. ELMQVIST, E. D. PIERSON, and W. E. RAINEY. 1992. Flying foxes as pollinators and seed dispersers in Pacific Islands ecosystems. In D. E. Wilson and G. L. Graham (Eds.) Biological Report. pp. 18–35, U.S. Department of the Interior Fish and Wildlife Service, Washington DC.

DINNO, A. 2017. Package ‘dunn.test.’ CRAN Repos. 1–7. Available at: https://cran.r- project.org/web/packages/dunn.test/dunn.test.pdf [Accessed January 2, 2018].

DIREKTORAT JENDERAL PERKEBUNAN (DIRECTORATE GENERAL OF ESTATE CROPS). 2017. Statistik Perkebunan Indonesia (Tree Crop Estate Statistics of Indonesia) 2015- 2017 Kelapa (Coconut). South Jakarta.

DJAJANTI, D. 2006. Managing forest with community (PHBM) in Central Java: promoting equity in access to NTFPs. In S. Mahanty, J. Fox, M. Nurse, P. Stephen, and L. McLees (Eds.) Hanging in the Balance: Equity in Community-based Natural Resource Management in Asia. pp. 63–82, RECOFTC and East-West Center, Bangkok.

DOBAT, K., and T. PEIKERT-HOLLE. 1985. Bluten und Fledermause. Waldemar Kramer, Frankfurt.

78

ELMQVIST, T., P. A. COX, W. E. RAINEY, and E. D. PIERSON. 1992. Restricted pollination on Oceanic Islands: pollination of Ceiba pentandra by flying foxes in Samoa. Biotropica 24: 15–23.

EPSTEIN, J. H., K. J. OLIVAL, J. R. C. PULLIAM, C. SMITH, J. WESTRUM, T. HUGHES, A. P. DOBSON, A. ZUBAID, S. A. RAHMAN, M. M. BASIR, H. E. FIELD, and P. DASZAK. 2009. Pteropus vampyrus, a hunted migratory species with a multinational home-range and a need for regional management. J. Appl. Ecol. 46: 991–1002.

FLEMING, T. H., C. GEISELMAN, and W. J. KRESS. 2009. The evolution of bat pollination: a phylogenetic perspective. Ann. Bot. 104: 1017–1043.

FLEMING, T. H., and N. MUCHHALA. 2008. Nectar-feeding bird and bat niches in two worlds: pantropical comparisons of vertebrate pollination systems. J. Biogeogr. 35: 764–780.

FLEMING, T. H., and V. J. SOSA. 1994. Effects of nectarivorous and frugivorous mammals on reproductive success of plants. J. Mammal. 75: 845–851.

FLORENS, F. B. V., C. BAIDER, V. MARDAY, G. M. N. MARTIN, Z. ZMANAY, R. OLEKSY, G. KRIVEK, C. E. VINCENOT, D. STRASBERG, and T. KINGSTON. 2017. Disproportionately large ecological role of a recently mass-culled flying fox in native forests of an oceanic island. J. Nat. Conserv. 40: 85–93.

FREY, S., J. T. FISHER, A. C. BURTON, and J. P. VOLPE. 2017. Investigating animal activity patterns and temporal niche partitioning using camera-trap data: challenges and opportunities. Remote Sens. Ecol. Conserv. 3: 123–132.

FUJITA, M. S., and M. D. TUTTLE. 1991. Flying foxes (Chiroptera: Pteropodidae): threatened animals of key ecological and economic importance. Biol. Conserv. 5: 455–463.

GALLAI, N., J. M. SALLES, J. SETTELE, and B. E. VAISSIÈRE. 2009. Economic valuation of the vulnerability of world agriculture confronted with pollinator decline. Ecol. Econ. 68: 810–821.

GHANEM, S. J., and C. C. VOIGT. 2012. Increasing awareness of ecosystem services provided by bats. In H. J. Brockmann, T. J. Roper, M. Naguib, J. C. Mitani, and L. W. Simmons (Eds.) Advances in the Study of Behavior. pp. 279–302, Academic Press, Burlington.

GOULD, E. 1977. Foraging behaviour of Pteropus vampyrus on the flowers of Durio zibethinus. Malayan Nat. J. 10: 53–57.

GOULD, E. 1978. Foraging behavior of Malaysian nectar-feeding bats. Biotropica 10: 184–193.

79

HENGJAN, Y., V. SAPUTRA, M. MIRSAGERI, D. PRAMONO, S. KASMONO, C. BASRI, T. ANDO, Y. OHMORI, S. AGUNGPRIYONO, and E. HONDO. 2018. Nighttime behavioral study of flying foxes on the southern coast of West Java, Indonesia. J. Vet. Med. Sci. 80: 1146–1152.

HODGKISON, R., and S. T. BALDING. 2003. Fruit bats (Chiroptera: Pteropodidae) as seed dispersers and pollinatos in a lowland Malaysian rain forest. Biotropica 35: 491– 502.

HONSHO, C., S. SOMSRI, S. SALAKPETCH, T. TETSUMURA, Y. YONEMOTO, and K. YONEMORI. 2009. Pollen sources effects on seed formation and fruit characteristics in Thai durians. Trop. Agric. Dev. 53: 2009.

HONSHO, C., K. YONEMORI, S. SOMSRI, S. SUBHADRABANDHU, and A. SUGIURA. 2004. Marked improvement of fruit set in Thai durian by artificial cross-pollination. Sci. Hortic. (Amsterdam). 101: 399–406.

HONSHO, C., K. YONEMORI, A. SUGIURA, S. SOMSRI, and S. SUBHADRABANDHU. 2004. Durian floral differentiation and flowering habit. J. Am. Hortic. Soc. 129: 42–45.

IBOUROI, M. T., A. CHEHA, G. ASTRUC, S. A. O. DHURHAM, and A. BESNARD. 2018. A habitat suitability analysis at multi-spatial scale of two sympatric flying fox species reveals the urgent need for conservation action. Biodivers. Conserv. 27: 2395– 2423.

INDARTI, D. 2014. Outlook Komoditi Durian (Outlook for Durian as Commodity) by Center for Data and Agricultural Information System. Jakarta.

JAVID, A., T. L. GULRAIZ, M. ASHRAF, M. NADEEM, S. M. HUSSAIN, M. ALTAF, and I. BABU. 2017. Proximate analysis of Indian flying fox (Pteropus giganteus) natural food, with a note on its roost variations in urban areas of Lahore, Pakistan. Turkish J. Zool. 41: 714–721.

KASINA, J. M., J. MBURU, M. KRAEMER, and K. HOLM-MUELLER. 2009. Economic benefit of crop pollination by bees: a case of Kakamega small-holder farming in Western Kenya. J. Econ. Entomol. 102: 467–473.

KEßLER, P. J. A., M. M. BOS, S. E. C. S. DAZA, A. KOP, L. P. M. WILLEMSE, R. PITOPANG, and S. R. GRADSTEIN. 2002. Checklist of woody plants of Sulawesi, Indonesia. Blumea J. Plant Taxon. Plant Geogr. 14: 1–160.

KINGSTON, T. 2010. Research priorities for bat conservation in Southeast Asia: a consensus approach. Biodivers. Conserv. 19: 471–484.

KUNZ, T. H., R. HODGKISON, and C. D. WEISE. 2009. Methods for capturing and handling bats. In T. H. Kunz and S. Parsons (Eds.) Ecological and Behavioral Methods for the Study of Bats. pp. 5–35, Johns Hopkins University Press, Baltimore.

80

KUNZ, T. H., E. B. DE TORREZ, D. BAUER, T. LOBOVA, and T. H. FLEMING. 2011. Ecosystem services provided by bats. Ann. N. Y. Acad. Sci. 1223: 1–38.

LEE, Y. F., Y. M. KUO, H. Y. CHANG, C. F. TSAI, and S. BABA. 2017. Foraging dispersion of Ryukyu flying-foxes and relationships with fig abundance in East-Asian subtropical island forests. BMC Ecol. 17: 1–12.

LEE, Y. F., T. TAKASO, T. Y. CHIANG, Y. M. KUO, N. NAKANISHI, H. Y. TZENG, and K. YASUDA. 2009. Variation in the nocturnal foraging distribution of and resource use by endangered Ryukyu flying foxes (Pteropus dasymallus) on Iriomotejima Island, Japan. Contrib. to Zool. 78: 51–64.

LEVINS, R. 1968. Evolution in Changing Environments. Princeton University Press, New Jersey.

LIM, T. K., and L. LUDERS. 1998. Durian flowering, pollination and incompatibility studies. Ann. Appl. Biol. 132: 151–165.

LIM, V.-C., R. RAMLI, S. BHASSU, and J.-J. WILSON. 2018. Pollination implications of the diverse diet of tropical nectar-feeding bats roosting in an urban cave. PeerJ 6: e4572.

LUSKIN, M. S. 2010. Flying foxes prefer to forage in farmland in a tropical dry forest landscape mosaic in Fiji. Biotropica 42: 246–250.

MARKUS, N., and L. HALL. 2004. Foraging behaviour of the black flying fox (Pteropus alecto) in the urban landscape of Brisbane, Queensland. Wildl. Res. 31: 345–355.

MARSHALL, A. G. 1983. Bats, flowers and fruit: evolutionary relationships in the Old World. Biol. J. Linn. Soc. 20: 115–135.

MARSHALL, A. G. 1985. Old World phytophagous bat (Megachiroptera) and their food plants: a survey. Zool. J. Linn. Soc. 83: 351–369.

MCCONKEY, K. R., and D. R. DRAKE. 2006. Flying foxes cease to function as seed dispersers long before they become rare. Ecology 87: 271–276.

MCCONKEY, K. R., and D. R. DRAKE. 2007. Indirect evidence that flying foxes track food resources among islands in a Pacific archipelago. Biotropica 39: 436–440.

MCCONKEY, K. R., and D. R. DRAKE. 2015. Low redundancy in seed dispersal within an island frugivore community. AoB Plants 7: 1–13.

MEREDITH, M., and M. RIDOUT. 2017. Overview of the overlap package. R Proj. 1–9.

MICKLEBURGH, S. P., A. M. HUTSON, and P. A. RACEY. 1992. Old World Fruit Bats: An Action Plan for their Conservation. IUCN, Gland.

81

MICKLEBURGH, S., K. WAYLEN, and P. RACEY. 2009. Bats as bushmeat: a global review. Oryx 43: 217–234.

MORISITA, M. 1959. Measuring of interspecific association and similarity between communities. Mem. Fac. Sci. Kyushu Univ. Ser. E 3: 65–80.

MUSCARELLA, R., and T. H. FLEMING. 2007. The role of frugivorous bats in tropical forest succession. Biol. Rev. 82: 573–590.

NAKAMOTO, A., K. KINJO, and M. IZAWA. 2015. Dietary plasticity in the Ryukyu flying fox on a subtropical ssland at the northern range limit of Pteropus. Acta Chiropterologica 17: 105–116.

NATHAN, P. T., T. KARUPPUDURAI, H. RAGHURAM, and G. MARIMUTHU. 2009. Bat foraging strategies and pollination of Madhuca latifolia (Sapotaceae) in Southern India. Acta Chiropterologica 11: 435–441.

NOR ZALIPAH, M., M. S. SHAHRUL ANUAR, and G. JONES. 2016. The potential significance of nectar-feeding bats as pollinators in mangrove habitats of Peninsular Malaysia. Biotropica 48: 425–428.

OLEKSY, R., L. GIUGGIOLI, T. J. MCKETTERICK, P. A. RACEY, and G. JONES. 2015. Fruit bats create extensive seed shadows and enhance germination success of pioneer plant species in deforested Madagascan landscapes. PLoS One 1–17.

OLEKSY, R., P. A. RACEY, and G. JONES. 2015. High-resolution GPS tracking reveals habitat selection and the potential for long-distance seed dispersal by Madagascan flying foxes Pteropus rufus. Glob. Ecol. Conserv. 3: 678–692.

PÁEZ, D. J., O. RESTIF, P. EBY, and R. K. PLOWRIGHT. 2018. Optimal foraging in seasonal environments: Implications for residency of australian flying foxes in food- subsidized urban landscapes. Philos. Trans. R. Soc. B Biol. Sci. 373: 20170097.

PALMER, C., O. PRICE, and C. BACH. 2000. Foraging ecology of the black flying fox (Pteropus alecto) in the seasonal tropics of the Northern Territory, Australia. Wildl. Res. 27: 169–178.

PIMSAI, U., M. J. PEARCH, C. SATASOOK, S. BUMRUNGSRI, and P. J. J. BATES. 2014. Murine rodents (Rodentia: Murinae) of the Myanmar-Thai-Malaysian peninsula and Singapore: taxonomy, distribution, ecology, conservation status, and illustrated identification keys. Bonn Zool. Bull. 63: 15–114.

RAFANI, M. I. 2013. Overview of Fruit Production, Marketing, and Research and Development System in Indonesia. Bogor.

RAO, S. 2018. Roosting, foraging and mating behaviour of Indian flying fox (Pteropus giganteus) in Rourkela, Odisha. SSRN 1–19.

82

RIDOUT, M. S., and M. LINKIE. 2009. Estimating overlap of daily activity patterns from camera trap data. J. Agric. Biol. Environ. Stat. 14: 322–337.

ROBERTS, B. J., C. P. CATTERALL, P. EBY, and J. KANOWSKI. 2012. Long-distance and frequent movements of the flying-fox Pteropus poliocephalus: implications for management. PLoS One 7: 1–11.

SANTOSO, P. J. 2012. Indonesia Berpotensi Produksi Durian Sepanjang Tahun (Indonesia’s Potential for Durian Production Year-round). SinarTani: Agroinovasi 19: 10–16.

SCANLON, A. T., S. PETIT, M. TUIWAWA, and A. NAIKATINI. 2014. High similarity between a bat-serviced plant assemblage and that used by humans. Biol. Conserv. 174: 111– 119.

SHEHERAZADE, and S. M. TSANG. 2015. Quantifying the bat bushmeat trade in North Sulawesi, Indonesia, with suggestions for conservation action. Glob. Ecol. Conserv. 3: 324–330.

SHEHERAZADE, YASMAN, D. H. PRADANA, and S. M. TSANG. 2017. The role of fruit bats in plant community changes in an urban forest in Indonesia. Raffles Bull. Zool. 65: 497–505.

SHILTON, L. A., J. D. ALTRINGHAM, S. G. COMPTON, and R. J. WHITTAKER. 1999. Old World fruit bats can be long-distance seed dispersers through extended retention of viable seeds in the gut. Proc. R. Soc. B Biol. Sci. 266: 219.

SIMMONS, N. 2005. Chiroptera. In D. E. Wilson and D. A. M. Reeder (Eds.) Mammals of the World: A Taxonomic and Geographic References. pp. 312–529, Johns Hopkins University Press, Baltimore.

SINGARAVELAN, N., and G. MARIMUTHU. 2004. Nectar feeding and pollen carrying from Ceiba pentandra by pteropodid bats. J. Mammal. 85: 1–7.

SPENCER, H. J., C. PALMER, and K. PARRY-JONES. 1991. Movements of fruit-bats in eastern Australia, determined by using radio-tracking. Wildl. Res. 18: 463–468.

SRITHONGCHUAY, T., S. BUMRUNGSRI, and E. SRIPAO-RAYA. 2008. The pollination ecology of the late-successional tree, Oroxylum indicum (Bignoniaceae) in Thailand. J. Trop. Ecol. 24: 477–484.

START, A. N. 1974. The feeding biology in relation to food sources of nectarivorous bats (Chiroptera: Macroglossinae) in Malaysia. University of Aberdeen.

START, A. N., and A. G. MARSHALL. 1976. Nectarivorous bats as pollinators of trees in West Malaysia. In J. Burley and B. T. Styles (Eds.) Tropical Trees: Variation, Breeding and Conservation. pp. 141–149, Academic Press, London.

83

STEWART, A. B., and M. R. DUDASH. 2016. Flower-visiting bat species contribute unequally towards agricultural pollination ecosystem services in southern Thailand. Biotropica doi:10.111.

STEWART, A. B., and M. R. DUDASH. 2017. Field evidence of strong differential pollen placement by Old World bat-pollinated plants. Ann. Bot. 119: 73–79.

STIER, S. C., and T. L. MILDENSTEIN. 2005. Dietary habits of the world’s largest bats: the Philippine flying foxes, Acerodon jubatus and Pteropus vamyprus lanensis. J. Mammal. 86: 719–728.

SUYANTO, A., M. YONEDA, I. MARYANTO, MAHARADATUNKAMSI, and J. SUGARDJITO. 2002. Checklist of the Mammals of Indonesia 2nd ed. LIPI-JICA-PHKA, Bogor.

THAVRY, H., J. CAPPELLE, S. BUMRUNGSRI, L. THONA, and N. M. FUREY. 2017. The diet of the cave nectar bat (Eonycteris spelaea Dobson) suggests it pollinates economically and ecologically significant plants in Southern Cambodia. Zool. Stud. 56: 1–7.

THOMAS, R. J., and A. JOSEPHRAJKUMAR. 2013. Flowering and pollination biology in coconut. J. Plant. Crop. 41: 109–117.

TROLL, R. 1975. Chiropterophily. Bios 46: 150–152.

TSANG, S. M., and SHEHERAZADE. 2016. Acerodon celebensis. IUCN Red List Threat. Species. Available at: http://dx.doi.org/10.2305/IUCN.UK.2016- 2.RLTS.T137A21988719.en [Accessed February 12, 2017].

TSCHAPKA, M., and S. DRESSLER. 2002. Chiropterophily: on bat-flowers and flower bats. R. Bot. Gard. 114–125.

VARDON, M. J., P. S. BROCKLEHURST, J. C. Z. WOINARSKI, R. B. CUNNINGHAM, C. F. DONNELLY, and C. R. TIDEMANN. 2001. Seasonal habitat use by flying-foxes, Pteropus alecto and P. scapulatus (Megachiroptera), in monsoonal Australia. J. Zool. 253: 523–535.

WATSON, B. J. 1983. Durian. Rare Fruit Counc. Aust. 4.

WEBER, N., P. DUENGKAE, J. FAHR, D. K. N. DECHMANN, P. PHENGSAKUI, W. KHUMBUCHA, B. SIRIAROONRAT, S. WACHARAPLUESADEE, P. MANEEORN, M. WIKELSKI, and S. NEWMAN. 2015. High-resolution GPS tracking of Lyle’s flying fox between temples and orchards in Central Thailand. J. Wildl. Manage. 79: 957–968.

WICKHAM, H., and W. CHANG. 2016. Package ‘ggplot2’ Create Elegant Data Visualisations Using the Grammar of Graphics Description. CRAN Repos. 1–189. Available at: https://cran.r-project.org/web/packages/ggplot2/ggplot2.pdf [Accessed April 3, 2017].

84

WIN, S. S., and K. M. MYA. 2015. The diet of the Indian flying fox Pteropus giganteus (Brünnich. 1782) (Chiroptera: Pteropodidae) in Myanmar - conflicts with local people? J. Threat. Taxa 7: 7568–7572.

YUMOTO, T. 2000. Bird-pollination of three Durio species (Bombacaceae) in a tropical rainforest in Sarawak, Malaysia. Am. J. Bot. 87: 1181–1188.

85

BIOGRAPHICAL SKETCH

Sheherazade graduated cum laude from Universitas Indonesia in August 2014 with a Bachelor of Science in biology. Then, she worked in Alliance for Tompotika

Conservation, Central Sulawesi from 2014 to 2016. She was involved in conservation education, patrol, and animal monitoring programs. Sheherazade pursued a Master of

Science in wildlife ecology and conservation at the University of Florida with a concentration on Tropical Conservation and Development, which she completed in the fall of 2018. She plans to go back to Indonesia and work as a conservation practitioner in Sulawesi. Being a leading conservationist in the Wallacea region has always been her lifetime goal.

86