) IN THE WATERS OF HONG KONG

Transcription

1 ASSESSMENT OF THE IMPACTS OF NOISE AND VESSEL TRAFFIC ON THE DISTRIBUTION, ABUNDANCE AND DENSITY OF CHINESE HUMPBACK DOLPHINS (SOUSA CHINENSIS CHINENSIS) IN THE WATERS OF HONG KONG A Thesis Submitted to the Committee on Graduate Studies in Partial Fulfillment of the Requirements for the Degree of Master of Science in the Faculty of Arts and Science TRENT UNIVERSITY Peterborough, Ontario, Canada Copyright by Michelle Nicole Klein 2015 Environmental and Life Sciences M.Sc. Graduate Program September 2015

2 ABSTRACT ASSESSMENT OF THE IMPACTS OF NOISE AND VESSEL TRAFFIC ON THE DISTRIBUTION, ABUNDANCE AND DENSITY OF CHINESE HUMPBACK DOLPHINS (SOUSA CHINENSIS CHINENSIS) IN THE WATERS OF HONG KONG Michelle Nicole Klein Marine mammals with near-shore distributions are susceptible to human-related recreational and commercial disturbances, particularly near densely populated and industrialized coastal areas. A population of over 2,500 Chinese humpback dolphins (Sousa chinensis chinensis) occupies the Pearl River Estuary in southern China. A part of this population uses Hong Kong s waters off of Lantau Island, where they are subjected to a number of anthropogenic threats, including vessel disturbance, fisheries interactions, and boat-based tourism. Previous research has shown that the abundance of this subspecies in Hong Kong s waters has declined about 60% since Using a combination of acoustic recordings, dolphin distribution and abundance data, and vessel traffic information I found that: 1) Four types of vessels common to the waters on Hong Kong generate noise that is audible to Sousa chinensis chinensis; 2) The spatial distribution of underwater noise in Hong Kong s waters does not significantly vary among the six sites sampled; 3) High-speed ferry traffic and passenger volume has increased dramatically during the study period; 4) There has been a significant decline in dolphin density in areas within and near vessel traffic; and 5) Dolphins are most at risk of vessel collisions and being exposed to vessel noise near Fan Lau and within the Urmston Road waterway just northeast of the Sha Chau and Lung Kwu Chau Marine ii

3 Park. These results can inform future acoustic studies on this species and guide conservation and management efforts in Hong Kong. Keywords: Humpback dolphins, Sousa chinensis chinensis, acoustics, noise, distribution, abundance, density, vessel traffic, human impacts, management iii

4 ACKNOWLEDGEMENTS I will do my best to properly acknowledge and thank all of the people that have provided the inspiration, assistance and instruction I have received during this adventure that is graduate school. First to my supervisor, Dr. Bradley White. He is an incredible advisor and I cannot thank him enough for all of the guidance, helpful suggestions, and pep talks while I worked on this project. Dr. Samuel Hung (Hong Kong Dolphin Conservation Society/Hong Kong Cetacean Research Project) also deserves my sincerest thanks for providing numerous opportunities for me to travel to Hong Kong and contribute to the Hong Kong Cetacean Research Project. Samuel has been instrumental in providing me with the data needed to complete this project and I appreciated his timely responses to all of my questions and requests for additional data, even when we were 12 times zones and 12,500 km apart! I would also like to thank my mentor, Dr. John Wang (Trent University/CetAsia Research Group). Without John, my knowledge of the biology and ecology of marine mammals would not be nearly as thorough or interesting if it were not for the casual and insightful discussions that I have had with him. I would like to thank my committee members, Dr. Leslie Kerr and Dr. Joe Nocera, for their insightful discussions, suggestions of additional literature to review, and iv

5 comments on my thesis drafts. I would also like to thank Dr. Raul Ponce-Hernandez for answering all of my questions relating to my GIS analyses. I would also like to thank my lab mates and friends, Lauren Dares, Jordan Hoffman, Shiva Javdan, and Claryana Araújo-Wang for their helpful suggestions along the way. To all of the Hong Kong Cetacean Research Project research assistants and interns, thank you for helping to collect the data used for this thesis and responding to all of my requests for additional data. To my family, your support, confidence and honest interest in my research made the rough times bearable and the good times even more rewarding. To my parents, you have always inspired my curiosity and instilled the confidence and motivation to pursue what most intrigues me thank you. To Steven Seyler, your love, inspiration and support are absolutely the most important things I have I cannot thank you enough. This research was supported by funding from the Hong Kong Agriculture, Fisheries and Conservation Department awarded to the Hong Kong Cetacean Research Project and an Ontario Graduate Scholarship funded my graduate studies at Trent University. v

6 TABLE OF CONTENTS Abstract / ii Acknowledgements / iv Table of Contents / vi List of Figures / vii List of Tables / x Chapter 1: General Introduction / 1 Effects of Elevated Underwater Noise on Marine Mammals / 3 Impacts of Human Activity on Humpback Dolphins / 5 Sousa chinensis chinensis distribution and abundance in Hong Kong / 6 Hearing Capabilities of Sousa spp. / 7 Previous acoustic research in Hong Kong / 8 Conservation Status / 9 Research Questions and Summary / 9 Chapter 2: Vessel noises in the waters of western Hong Kong relative to Chinese humpback dolphins (Sousa chinensis chinensis) / 13 Introduction / 14 Materials and Methods / 17 Results / 22 Discussion / 28 Chapter 3: Impacts of high-speed ferry route and traffic volume on the abundance and density of Chinese humpback dolphins (Sousa chinensis chinensis) in Hong Kong s waters / 42 Introduction / 43 Materials and Methods / 46 Results / 49 Discussion / 53 Chapter 4: Probability and mitigation of vessel collisions with Chinese humpback dolphins (Sousa chinensis chinensis) in Hong Kong s waters / 72 Introduction / 74 Materials and Methods / 77 Results / 82 Discussion / 84 Chapter 5: Discussion and Conclusions / 96 Literature Cited / 102 Appendix A / 110 vi

7 LIST OF FIGURES Chapter 1. Figure 1.1. Map of the Pearl River Estuary (PRE). Chapter 2 Figure 2.1. Map of Lantau Island in Hong Kong with recording stations, vessel recording locations and survey transect lines. Figure 2.2. Humpback dolphin and two bottlenose dolphin audiograms (modified from Li et al., 2012; Johnson, 1967; and Popov et al., 2007). Figure 2.3. Mean noise levels in six areas varying in amount of vessel traffic and types of vessels present. Figure /3 octave band sound pressure levels for a high-speed ferry at Northeast Lantau #1 (NEL #1); Beaufort sea state 2. Figure /3 octave band sound pressure levels for a small dolphin watching tour boat (Wala wala) in West Lantau; Beaufort sea state 1. Figure /3 octave band sound pressure levels for a shrimp trawling vessel at West Lantau #2 (WL #2); Beaufort sea state 3. Figure 7. 1/3 octave band sound pressure levels for a large shipping container vessel at Southwest Lantau #1 (SWL #1); Beaufort sea state 2. Chapter 3 Figure 3.1. Map of six survey areas in western Hong Kong. Figure 3.2. Temporal trends in abundance estimates of Chinese humpback dolphins in West, Northwest, and Northeast Lantau from (adapted from Hung, 2013). Figure 3.3. Annual number of high-speed ferries departing from and arriving at Hong Kong ports (excluding the Sky Pier) from vii

8 Figure 3.4. Annual trend in total number of high-speed ferry (HSF) trips and dolphin abundance from Figure 3.5. Annual number of high-speed ferries from Hong Kong to Macau and Non- Macau (i.e., Mainland China s cities) from Figure 3.6. Mean number of daily high-speed ferry trips from Figure 3.7. Annual number of passengers using high-speed ferry services within Hong Kong and to and from Macau and China s Mainland ports (excluding the Sky Pier) from Figure 3.8. Annual number of passengers traveling to Mainland China and Macau ports from the Sky Pier from Figure 3.9. Temporal trend in dolphin densities (DPUE values, i.e., number of dolphins per 100 units of survey effort) in Deep Bay (DB), Northeast Lantau (NEL), Northwest Lantau (NWL), West Lantau (WL), Southwest Lantau (SEL) and Southeast Lantau (SEL) from Figure Map of three areas that overlap with vessel traffic in either the North Lantau Vessel Fairway (NLVF) or South Lantau Vessel Fairway (SLVF). Figure Temporal trend in mean dolphin density (DPUE, number of dolphins per 100 units of survey effort) in seven 1km 2 grid cells around Fan Lau (near the South Lantau Vessel Fairway) from Figure Temporal trend in mean dolphin density (DPUE, number of dolphins per 100 units of survey effort) in seven 1km 2 grid cells around the northeast corner of the Hong Kong International Airport (near the Sky Pier and adjacent to the North Lantau Vessel Fairway) from Figure Temporal trend in mean dolphin density (DPUE, number of dolphins per 100 units of survey effort) in seven 1km 2 grid cells around Lung Kwu Chau (near the North Lantau Vessel Fairway) from Chapter 4 Figure 4.1. Map of Lantau Island in Hong Kong identifying six survey areas for humpback dolphins. Figure km resolution maps of western Hong Kong illustrating the study domain (black grid), survey area delineations (black dashed lines), and Hong Kong s territorial viii

9 border (solid black line) and showing the relative probability of a) observing a humpback dolphin, b) observing a vessel, and c) a vessel encountering a humpback dolphin in Figure km resolution maps of western Hong Kong illustrating the study domain (black grid), survey area delineations (black dashed lines), and Hong Kong s territorial border (solid black line) and showing the relative probability of a) observing a humpback dolphin, b) observing a vessel, and c) a vessel encountering a humpback dolphin in Figure km resolution maps of western Hong Kong illustrating the study domain (black grid), survey area delineations (black dashed lines), and Hong Kong s territorial border (solid black line) and showing the relative probability of a) observing a cargo or tanker vessel, b) observing a fishing vessel, and c) observing a high-speed craft in Figure km resolution maps of western Hong Kong illustrating the study domain (black grid), survey area delineations (black dashed lines), and Hong Kong s territorial border (solid black line) and showing the relative probability of a) observing a cargo or tanker vessel, b) observing a fishing vessel, and c) observing a high-speed craft in ix

10 LIST OF TABLES Chapter 2 Table 2.1. Descriptions of noise recording locations Table 2.2. Descriptions of common vessel types in Hong Kong Chapter 4 Table 4.1. Relative probabilities of observing humpback dolphins in western Hong Kong in 2012 and 2013 x

11 1 CHAPTER 1 GENERAL INTRODUCTION The concern about the effects of underwater noise on marine mammals has steadily increased in the past few decades. Humpback dolphins (Sousa spp.) are sensitive to underwater sound because of their very acute hearing, wide hearing frequency range, and responsiveness to sounds. Sousa spp. inhabit the shallow coastal and estuarine waters of the eastern Atlantic, western Pacific and Indian oceans. Because they inhabit areas close to land, often in areas of high human concentration, humpback dolphins are particularly susceptible to the impacts of human activities. Hong Kong s waters are particularly busy, with many sources of anthropogenic disturbance (e.g., construction, dredging, heavy vessel traffic, pollution, dolphin tourism, and fishing activities) throughout the area. Many of these activities generate considerable amounts of noise; however, the effects of underwater noise and disturbance from vessel traffic on Sousa chinensis cninensis are not well understood. Introduction Sound is an important sensory modality for many marine animals. In the viscous marine environment, other senses such as vision, touch, smell, or taste can be limited by the effective range and speed of signal transmission (Würsig and Richardson, 2009). Among marine mammals, cetaceans (whales, dolphins, and porpoises) use a wide band

12 2 of acoustic frequencies. For example, the massive blue whale (Balaenoptera musculus) produces low-frequency sounds down to 15 Hz, whereas on the other end of the spectrum, dolphins within the genus Cephalorhynchus can produce sounds up to 200 khz (Tougaard and Kyhn, 2010). This broad range of frequencies overlaps with many of the sounds humans introduce into the water (intentionally or otherwise), including ship noise, sonars of various types, and seismic activities (Hildebrand, 2005). Sources of anthropogenic sound are becoming both more pervasive and more powerful, increasing both oceanic background noise levels and peak sound intensity levels. Anthropogenic activities in the ocean have increased over the past 50 years, resulting in more low-frequency (<1,000 Hz) and mid-frequency (1 20 khz) noise (Hildebrand, 2005). There is growing concern that sound introduced into the sea by human activities can have detrimental effects on marine mammals. For example, there is mounting evidence suggesting that high-intensity anthropogenic sound from sonar and airguns have resulted in the stranding and subsequent mortality of beaked whales (Ziphiidae) (Hildebrand, 2005). In recent years, public concern about the possible effects of anthropogenic environmental noise has grown steadily within the scientific and conservation communities (Würsig et al., 2000; Jefferson and Hung, 2004; Jefferson et al., 2009), which has resulted in more research on and attention to mitigating these adverse impacts. However, to propose effective and scientifically-based measures for mitigating the impacts of human-generated noise on marine mammals, it is necessary to understand underwater noise and its possible effects on habitat use and behaviour of marine mammals.

13 3 Effects of Elevated Underwater Noise on Marine Mammals In the ocean, acoustic energy propagates efficiently, travelling rapidly and potentially over great distances. Sound travels almost five times faster through seawater than through air, and low frequencies (<500 Hz) can travel hundreds of kilometers with little loss in energy (Urick, 1983). Sound propagation can be affected by many factors, the most influential of which are: 1) frequency of the sound; 2) water depth; and 3) density differences within the water column, which vary primarily with temperature, pressure, and salinity (Urick, 1983). Therefore, before sounds are received by an animal, they can be subject to propagation conditions that can be quite complex and can alter the characteristics of the original sound energy. Noise can affect marine mammals in many ways. At low levels, it may be merely detectable, but at higher levels, it may interfere with animal communication and hinder acoustic signal detection (Nowacek et al., 2007). Noise also can alter an animal s behaviour (e.g., spatial avoidance, increase in swimming speed, changes in diving behaviour) and affect the auditory system by inducing a shift in hearing threshold (either temporary or permanent) (e.g., Au and Perryman, 1982; Kruse, 1991; Evans et al., 1992; Janik and Thompson, 1995; Allen and Read, 2000). Noise, particularly when it is impulsive, may also cause concussive effects, physical damage to tissues and organs, especially if they are gas-filled and even death. Behavioural responses to increased noise such as habitat displacement, behavioural changes and alterations in the intensity,

14 4 frequency and intervals of calls have been observed (Nowacek et al., 2007; Weilgart, 2007). Physiological responses to noise, such as increased levels of stress-related faecal hormone metabolites (glucocorticoids) have also been documented (Rolland et al., 2012). Prolonged exposure to stressors, such as anthropogenic noise, can lead to health problems such as suppressed growth, immune system function and reproduction (Romero and Butler, 2007; Sapolsky et al., 2000) as well as decreased fitness (Erbe, 2012). The effects of noise on marine mammals depend on the acoustic characteristics of the source, the medium through which the sound is traveling, and the receiver of the sound. The hearing threshold is the amplitude necessary for detection, and varies with frequency across the hearing range of a given individual (Nowacek et al., 2007). Both permanent threshold shifts and temporary threshold shifts represent actual reductions in the ability of an animal to hear, usually at particular frequencies, because of exposure to loud sounds of those frequencies. High-amplitude, broadband noise can also has the potential to mask communication sounds, echolocation signals, predator and prey sounds, and environmental sounds. Finally, behavioural responses are a demonstrable change in the activity of an animal in response to a sound. These effects can be difficult to detect due to the cryptic and variable nature of cetacean behaviour, but they can be demonstrated (Nowacek et al., 2007). Examples of behavioural effects include the abandonment of a location or important activity (e.g., feeding, nursing, reproduction) in response to some sound as well as changes in diving behaviour (e.g., swim speed, respiration rate, dive duration) vocalizations (e.g., increased call rate, increased

15 5 amplitude, frequency modulation) (Weilgart, 2007). The repeated abandonment or disruption of vital activities can lead to detrimental consequences for the animal(s) affected. Impacts of Human Activity on Humpback Dolphins Humpback dolphins (Sousa spp.) are delphinid cetaceans found nearly exclusively in murky, shallow coastal and estuarine waters of the eastern Atlantic, Indian, and western Pacific oceans. The genus Sousa consists of the Atlantic humpback dolphin (Sousa teuszii), the Australian humpback dolphin (S. sahulensis), the Indian Ocean humpback dolphin (S. plumbea) and the Indo-Pacific humpback dolphin (S. chinensis) (Jefferson and Rosenbaum, 2014). Recently, humpback dolphins of the eastern Taiwan Strait were described as a distinct subspecies within S. chinensis, the Taiwanese humpback (or white) dolphin, Sousa chinensis taiwanensis, thus all remaining dolphins within the species, Sousa chinensis, are now referred to as the nominate subspecies, Sousa chinensis chinensis (the Chinese white or humpback dolphin) (see Wang et al., 2015). Because they inhabit nearshore waters, and often areas with high human concentrations, humpback dolphins are particularly susceptible to human activities. Hong Kong s waters are particularly busy, with many sources of anthropogenic disturbance (e.g., construction, dredging, heavy vessel traffic, chemical pollution,

16 6 dolphin tourism, and fishing activities) throughout the area (Morton, 1996; Jefferson and Hung, 2004; Jefferson et al., 2009). When the construction of the present Hong Kong International Airport on Chek Lap Kok Island began in 1993, there were concerns that local humpback dolphin habitat would be degraded and destroyed, and that the dolphins would be seriously threatened (Leatherwood and Jefferson, 1997; Liu and Hills, 1997). To understand the potential impacts of this project on local marine mammals, the Hong Kong Special Administrative Region (SAR) Government funded several long-term studies on the status and biology of the two resident cetaceans, humpback dolphins and finless porpoises, within Hong Kong s territories (Parsons, 1997; Porter, 1998; Jefferson, 2000a; Jefferson, 2000b; Jefferson, 2007; Jefferson et al., 2002; Jefferson et al., 2006; Jefferson and Hung, 2004; Jefferson and Hung, 2007), and these studies are still ongoing (Hung, 2013). The very significant increase in coastal development over the past 20 years, which is related to economic growth in China and Southeast Asia, has resulted in increasing impacts on the local coastal marine environment and its biota, including humpback dolphins (Liu and Hills, 1997). Many of these coastal development activities, including but not limited to, the ongoing construction of the Hong Kong-Zhuhai-Macau Bridge and its associated Border Crossing Facility, proposed expansion of the Hong Kong International Airport, and resultant increased vessel traffic servicing the airport passengers may contribute significantly more noise to Hong Kong s underwater soundscape.

17 7 Sousa chinensis chinensis distribution and abundance in Hong Kong Chinese humpback dolphins (S. c. chinensis) can be found throughout the Pearl River Estuary (PRE) (Figure 1.1). Recent abundance estimates from line-transect vessel survey data indicated there are approximately 2,500 animals (Chen et al., 2010) in the PRE, which includes the waters of Macau, Hong Kong and the People s Republic of China, with a resident group found in the waters around Lantau Island in Hong Kong (Hung and Jefferson, 2004). Long-term monitoring studies have shown that humpback dolphins mainly occur in the western waters of Hong Kong (near Lantau Island), and are only seen rarely in eastern waters where salinity is higher (Parsons, 1998; Jefferson, 2000a; Jefferson et al., 2002; Jefferson and Hung, 2004). The western waters around North and West Lantau Island represent the main distribution of humpback dolphins (Jefferson, 2000a; Jefferson and Hung, 2004; Hung, 2013). There is evidence that S. c. chinensis abundance has been steadily declining in the waters of Hong Kong (Hung, 2013), and human disturbances such as increased vessel traffic and major development projects may have resulted in some individuals avoiding certain areas that appeared to be important habitat in the past. Hearing Capabilities of Sousa spp. Since the first odontocete hearing was measured as a function of hearing threshold versus frequency of sound stimulus (i.e., an audiogram) in an Atlantic

18 8 common bottlenose dolphin, Tursiops truncatus (Johnson, 1967), audiograms of odontocete cetaceans have been measured using either psychophysical or evokedpotential methods (see review in Li et al., 2012). Li et al. (2012) described the hearing capabilities of a S. chinensis (now S. c. chinensis) using an evoked potential audiogram. Using a captive adult male, they measured responses to 1000 rhythmic 20 millisecond pip trains for each amplitude/frequency combination. Fourteen frequencies ranging from 5.6 khz to 152 khz were studied, and their results showed that most of the hearing thresholds were lower than 90 db re 1 µpa, covering a frequency range from 11.2 khz to 128 khz, and the lowest threshold of 47dB was measured at 45 khz. The audiogram, which is a function of hearing threshold versus stimulus carrier frequency, presented a U-shape with a region of highest hearing sensitivity (within 20dB of the lowest threshold) between approximately 20 khz and120 khz. This showed that Sousa spp. probably have hearing capabilities similar to that of the similarly-sized (skull and body) common bottlenose dolphin, T. truncatus. Extrapolation of the bottlenose dolphin audiogram to humpback dolphins in lower frequencies may be questionable based on some of the observed differences where humpback dolphin hearing threshold data overlap with those of bottlenose dolphins. However, individual variation in audiograms also exists among bottlenose dolphins (Johnson, 1967; Popov et al., 2007) so it is reasonable to assume that humpback dolphins likely exhibit similar individual differences as well.

19 9 Previous acoustic research in Hong Kong Ng and Leung (2003) and Hung et al. (2007) studied various anthropogenic disturbances that affect marine mammals in Hong Kong s waters. However, only a few researchers have studied the effects of noise pollution on these local species (Würsig et al., 2000; Würsig and Greene, 2002). Sounds made by vessel traffic in Hong Kong s waters appear to be a major source of anthropogenic underwater noise (Sims et al., 2012b; Würsig and Greene, 2002; Ng and Leung, 2003). Ng and Leung (2003) and Piwetz et al. (2012) have documented behavioural changes due to vessel disturbance. However, we still do not understand the effects of underwater noise (both natural and anthropogenic) on the distribution and abundance of S. c. chinensis in Hong Kong s waters. Conservation Status The global conservation status of Sousa chinensis was assessed as nearthreatened by the Red List of Threatened Species of the International Union for the Conservation of Nature (IUCN). The main threats to the species were primarily fishery interactions and habitat loss (Reeves et al., 2008). Although the species may not meet any of the criteria for Vulnerable at this time, it is likely to do so in the near future, especially considering the implications of the recent taxonomic revision that proposed three species (including the new species, the Australian humpback dolphin, S. sahulensis) within what was previously encompassed within S. chinensis, (Jefferson

20 10 and Rosenbaum, 2014) and the recent description of the Taiwanese humpback dolphin subspecies, Sousa chinensis taiwanensis (Wang et al., 2015). Research Questions and Summary Below are my research questions along with the chapters that address them: 1. What frequencies and intensities of noise are generated by various types of vessels within Sousa chinensis chinensis habitat in Hong Kong? (Chapter 2) 2. What is the spatial distribution of underwater noise in Hong Kong? (Chapter 2) 3. What is the temporal trend in high-speed ferry traffic and passenger volume since 1999? (Chapter 3) 4. How has Sousa chinensis chinensis density in Hong Kong been affected by vessel traffic? (Chapter 3) 5. What areas have the highest probability of encounters between vessels and dolphins? (Chapter 4) In the research presented in this thesis, I characterized and quantified the sound profiles of various types of vessels including a high-speed ferry, a dolphin watching tour boat, a shrimp trawler, and a shipping container vessel, as well as the mean noise levels within the dolphins habitat. I determined the relative sound contributions of various vessel types to Hong Kong s underwater soundscape and compared the noise profiles to

21 11 the hearing abilities (from published audiograms) of humpback and bottlenose dolphins. An understanding of the various sounds generated by these vessels will be useful in determining their contributions to the underwater soundscape and their effects on marine mammals in the area. I also analyzed high-speed ferry traffic volume and passenger data from the Hong Kong Marine Department to determine if there has been an increase in marine traffic within the dolphins habitat. To examine the impacts of high-speed ferry traffic on S. c. chinensis in Hong Kong, I used Hung s (2008) quantitative grid analysis approach to investigate trends in dolphin density in six different survey areas in western Hong Kong from To investigate the traffic patterns of vessels, the relative probability that a vessel occupies a 1 km 2 grid cell relative to other grid cells, was estimated for shipping, highspeed, and fishing vessels in western Hong Kong using Automatic Information System (AIS) data and plotted in a geographic information system (GIS). The relative probability of a dolphin occupying the same 1 km 2 grid cell was also calculated using available dolphin density data from line-transect surveys. Using an approach developed by Vanderlaan et al. (2008), the relative probability that a vessel and dolphin will occupy (encounter each other within) a grid cell was also calculated for 2012 and The resulting maps identified where vessel traffic and dolphin density are both high, and where mitigation measures may be most effective. Areas where both dolphin and vessel

22 12 densities were low were also identified and may offer options for the re-routing of certain types of vessels.

23 Figure 1.1. Map of the Pearl River Estuary (PRE) and delineation of Hong Kong s waters (solid black line). 13

24 14 CHAPTER 2 VESSEL NOISES IN THE WATERS OF WESTERN HONG KONG RELATIVE TO CHINESE HUMPBACK DOLPHINS (SOUSA CHINENSIS CHINENSIS) Abstract The waters of western Hong Kong are home to Chinese humpback dolphins (Sousa chinensis chinensis). This area is also dominated by intense vessel traffic that is hypothesized to be behaviourally and acoustically disruptive to dolphins. While behavioural disturbance from the presence of fishing and dolphin-watching vessels has been documented, there is a lack of information about the acoustic impacts of vessels on these dolphins. The relative sound contributions of four types of vessels, a highspeed ferry, a shipping container vessel, a fishing vessel, and a dolphin-watching tour boat were compared to the hearing thresholds of S. c. chinensis and nearby mean noise levels. Noise levels were also compared between areas of presumed high and low vessel traffic. Vessel sounds (varying between 200 and 48,000 Hz) were well within the audible range of dolphins. As vessels are numerous in western Hong Kong s waters, management of their speeds and distribution are suggested to mitigate potential effects on the dolphins. Future research should focus on the spatial overlap of the distribution of dolphins and different types of vessels. Keywords: Chinese humpback dolphins, Sousa chinensis chinensis, noise, vessels, hearing

25 15 Introduction Marine mammals with near-shore distributions are susceptible to humanrelated recreational and commercial disturbances, particularly near densely populated and industrialized coastal communities (Würsig, 1989; Jefferson et al., 2009). A population of over 2,500 humpback dolphins (Sousa chinensis chinensis) occupies the Pearl River Estuary (PRE) and adjacent waters of southern China (Chen et al., 2010). A part of this population uses Hong Kong s waters off Lantau Island, where they are subjected to a number of anthropogenic effects, including vessel disturbance, fisheries interactions, and boat-based tourism. This is also an important foraging area for S. c. chinensis (herein referred to as humpback dolphins) where these generalist feeders consume a variety of demersal and mid-water shoaling fishes supported by the PRE (Barros et al., 2004). Although the dolphins also engage in other biologically important activities in these waters, including socializing and resting, feeding appears to dominate daytime behavior (Hung, 2008). Increasing levels of vessel traffic and other anthropogenic activity off Lantau Island, including associated underwater noises that overlap with the dolphins vocalizations (Sims et al., 2012a), are of concern to the welfare of the animals in Hong Kong s waters (Reeves et al., 2008; Jefferson et al., 2009). Previous research (Ng and Leung, 2003; Hung et al., 2007) focused on anthropogenic disturbances that affected Hong Kong s marine mammals (i.e., S. c. chinensis and finless porpoise, Neophocaena phocaenoides), but only a few researchers have studied the effects of noise pollution on these species (Würsig et al., 2000; Würsig and Greene, 2002; Sims et al. 2012b). Würsig and Greene (2002) documented sound

26 16 pressure level (SPL) relationships to different frequencies associated with tankers and tugs either offloading, approaching, or departing the Aviation Fuel Receiving Facility (AFRF, Figure 2.1). Their findings showed that the waters north of Lantau Island are relatively noisy, but the vessels in question still meet airport authority requirements; however, they also noted that the effects of these sound disturbances to the cetaceans (almost exclusively humpback dolphins in North Lantau waters) inhabiting the area have yet to be documented. Sims et al. (2012b) found large differences in SPL between high traffic and no traffic areas, suggesting that vessels are the main contributors to these discrepancies. They documented the relative sound contributions of various high-speed vessels to nearby ambient and dolphin social sounds and found that the vessel sounds were well within the audible range of humpback dolphins, with sounds from ,000 Hz. Additionally, Sims et al. (2012b) found that vessel sounds at distances of 100 m exceeded the sounds made by humpback dolphins at closer distances, so vessels sounds may be masking or obscuring those made by dolphins at close distances. The objectives of this study were to examine the sound profiles of different vessel types including a high-speed vessel, a dolphin watching tour boat, a shrimp trawler, and a shipping container vessel in western Hong Kong s waters and characterize nearby mean noise levels. An understanding of the various sounds generated by these vessels will be useful in determining their contributions to the underwater soundscape and their effects on marine mammals in the area, as well as provide information for guiding potential mitigation measures. Recent dolphin abundance data indicated that the various activities of these vessels might be partially related to the recent decline in

27 17 humpback dolphin abundance in Hong Kong s waters (Hung, 2013). Thus, a summary of sounds from select vessels relative to mean background noise levels and dolphin hearing thresholds are provided.

28 18 Materials and Methods Field methods Vessel sounds and background noise levels (i.e., sounds recorded in both the presence and absence of vessels, and in the absence of dolphins) were recorded at various stations (Figure 2.1; see Table 2.1 for descriptions of noise recording locations) in the waters surrounding Lantau Island in Hong Kong, from May 2010 to July Recordings were taken in conjunction with a long-term sound monitoring program conducted by the Hong Kong Cetacean Research Project (HKCRP). HKCRP also conducts annual line-transect surveys throughout the Hong Kong Special Administrative Region, which is divided into twelve different survey areas, with line-transect surveys conducted among six of these areas (i.e., Northwest (NWL), Northeast (NEL), West (WL), Southwest (SWL), Southeast Lantau (SEL), and Deep Bay (DB) Figure 2.1). Solitary vessels (see Table 2.2 for description of vessel types) and other underwater sounds were recorded from the stern of a 15-m vessel, with a diesel inboard engine (the Standard ), with the engine and power off and the vessel drifting. Underwater sounds were recorded with a Cetacean Research Technology spotcalibrated hydrophone (model: CR1; sensitivity: db, re. V/μPa; linear frequency range listed as: khz 48 khz ± 3 db; usable frequency range listed as: khz 68kHz ± 3 db, but due to our linear frequency range, only sounds up to 48 khz were analyzed) and using a Fostex digital recorder (model: FR-2; frequency response: 20 Hz 80 khz ± 3 db) with a pre-amplified signal conditioner (model: PC200-ICP; precision gain: x0.1 x100; frequency range: >100 khz; system response: 1 Hz 100 khz ± 0.25 db) to

29 19 prevent overloading. The hydrophone was lowered into the water to a depth of between 3 to 7 m but suspended by a 2 m spar buoy, which acted to prevent excessive hydrophone movement from wave and boat motion. Recordings were made in Broadcast Wave Format using a sampling rate of 24-bit at 192 khz with recording duration, varying from 3 min and 1 s to 5 min and 2 s. During each sampling event, vessel type, distance from the recording vessel at cue time, vessel activity, and dolphin presence were recorded. The distance to vessels was measured using Bushnell Yardage Pro 800 laser range-finding binoculars (distance accuracy ±2 m up to 800 m). Date, start and end times, hydrophone and water depths, Beaufort sea state (see area, start and end location, gain, event, and any additional notes for each sampling event were also recorded. Locations were determined using a handheld GPS unit (Garmin etrex Legend H). 493 recordings were made between May 2010 and December 2013 in both the presence and absence of various vessel types; however, many recordings were made in the presence of multiple vessels. Acoustic data analysis Recordings of a high-speed ferry, a dolphin watching tour boat, a shrimp trawler, a shipping container vessel, and the mean noise level at various locations in the waters of western Hong Kong were analyzed using SpectraLAB software (version 4.32). Following the methods outlined in Sims et al. (2012b), vessel selections were divided into two categories (i.e., solitary and multiple vessels present) during the recording.

30 20 Vessels were defined as solitary if there were no other vessels present within 2 km from the recording vessel throughout the duration of the recording. Recordings in which there were two or more vessels within 2 km of each other in the study area were classified as having multiple vessels. Solitary vessel selections were analyzed at specific cue times that described the vessel's distance and direction. These selections were analyzed over 5 second segments, ±2.5 seconds of the cue time to accurately capture their SPL without averaging out their sounds. 1/3 octave band SPLs were computed using SpectraLAB s compute average spectrum analysis for solitary vessel selections. A 1/3 octave bandwidth was used because of its general approximation to cetacean auditory bands (Greene, 1995a). The 1/3 octave band SPLs describe the SPLs of the individual vessel at specific distances, relative to the hearing range of S. c. chinensis and nearby background noise levels. For the background noise measurements, all recordings were sorted by sampling location (regardless of vessel presence/absence within 2km of the calibrated hydrophone) and 10 recordings for each (n=6) sampling location were randomly selected for analysis. 10-second non-overlapping section measurements were made throughout the recording starting at the beginning. Most recording times were not a multiple of 10, so only full 10-second clips were measured for these. To avoid sound selection bias, measurements were repeated starting from the end of the recording. Furthermore, 18 of these selections (for a total of 3 min) were randomly selected and averaged for each noise recording to compute 1/3 octave band SPLs. The mean 1/3 octave band SPLs for each sampling site were then calculated by averaging the 1/3

31 21 octave band SPLs from each of the 10 recordings. Standard deviation is also reported for each sampling location. To reduce geographic or nearby traffic differences between acoustic sampling locations and individual vessel recordings, sampling locations were selected near the individual vessel recording for mean background noise comparisons. The mean 1/3 octave band SPLs at each sampling location were used to assess individual vessel sound contributions relative to background noise levels. Recordings from NEL #1, NWL #1, NWL #2, WL #2, and SWL #2 were used for comparisons of nearby background noises levels to vessel sounds, while recordings from WL #3, considered a high vessel noise site, were used for comparison as it is adjacent to the South Lantau Vessel Fairway (SLVF). Dolphin audiograms were also compared to the mean sound levels recorded at the six aforementioned areas to describe the audibility of the mean noise levels. 1/3 octave band sound pressure levels above the audiogram curves are audible to the dolphins and that the larger the differences between the noise and hearing threshold sound pressure levels, the easier it is for the dolphins to hear a sound. While only one audiogram is available for humpback dolphins (Li et al., 2012), several exist for common bottlenose dolphins (Tursiops truncatus, hereafter simply bottlenose dolphins ) (Johnson, 1967; Popov et al., 2007). Popov et al. (2007) observed variation amongst individual bottlenose dolphin audiograms; as such, the single audiogram available for the humpback dolphin may not accurately represent the mean hearing sensitivity of the species. Past research on humpback dolphin communication frequencies (Sims et al.

32 a; Van Parijs and Corkeron, 2001a) indicates that they share similarities in repertoire and frequency range to bottlenose dolphins, suggesting that these two species may also share similar audiograms. Therefore, the published audiograms of both bottlenose dolphins (Johnson, 1967; Popov et al., 2007) and the single audiogram of a humpback dolphin (Li et al., 2012) were used for comparison with the received SPLs (see Figure 2.2). For the bottlenose dolphin audiograms, the average SPL for each frequency band was used since both audiograms gave multiple SPL thresholds per frequency unit. For the Johnson (1967) audiogram, db re 1 μbar was converted to db re 1 μpa by adding 100 to the recorded SPL (Greene, 1995a). Frequency and amplitude data for all sound selections were plotted using Microsoft Excel 2013.

33 23 Results Of the 493 recordings, four were used for analysis of solitary vessels. Specifically, recordings of an unclassified high-speed ferry (HSF), a small dolphin-watching tour boat, a shrimp trawler, and a shipping container vessel were examined. Mean noise levels of six areas (n=60) were also analyzed: Northeast Lantau #1 (NEL #1), Northwest Lantau #1 (NWL #1), Northwest Lantau #2 (NWL #2), West Lantau #2 (WL #2), West Lantau #3 (WL #3), and Southwest Lantau #2 (SWL #2); see Table 1 for site details. Mean Noise Levels Comparison of the mean noise levels for the six sampling sites revealed high variability but all exceeded the humpback dolphin hearing thresholds by approximately 6dB to 56dB, depending on the frequency. WL #3, a busy traffic area near the SLVF, had a mean SPL that was higher than a portion of the frequency range of the other sites, NEL #1, NWL #2, NWL #1, WL #2, and SWL #2 (i.e., 400 Hz-1.6 khz; Figure 2.3). However, SWL #2, NWL #2 and NEL #1 had increasingly higher mean SPLs compared to WL #3 for frequencies above 1.6 khz (Figure 2.3). NEL #1, located near the northeast corner of the Hong Kong International Airport and adjacent to the Sky Pier high-speed ferry lane had mean SPLs between 96 and 107 db re 1 µpa throughout the entire frequency range measured (50-48,000 Hz; Figure 2.3). Despite being near (~2 km) a high-traffic area, NEL #1 maintained a relatively low mean SPLs between 500 Hz and 10 khz. However, NEL #1 had higher SPLs for frequencies

34 24 under 300 Hz (between db re 1 µpa) than all other sampling locations (Figure 2.3). Despite being located directly adjacent to the busy North Lantau Vessel Fairway (NLVF), NWL #2, located within the Sha Chau and Lung Kwu Chau Marine Park (SCLKCMP), had the lowest mean SPLs of all of the recordings for frequencies between 50 and 500 Hz at around db re 1 µpa (Figure 2.3). NWL #1 is located to the west of Sha Chau within the SCLKCMP and experiences very little boat traffic. NWL #1 had the lowest SPLs between 2 khz and 12.5 khz (Figure 2.3). SPLs between 2 khz and 12.5 khz were approximately 5-10 db re 1 µpa lower than the high vessel noise site, WL #3. WL #2, located in an area with a natural coastline and rare boat traffic had a relatively low mean SPL below 250 Hz compared to the other sampling locations. However, the mean SPL increased greatly between 200 Hz and 1 khz by almost 10 db re 1 µpa to a peak of 103 db re 1 µpa at 1 khz. This location had mean SPLs between db re 1 µpa at frequencies greater than 1 khz (Figure 2.3). This area is located about 3 km from the busy SLVF. Lastly, SWL #2 was consistently in the relatively lower range of SPLs to about 1 khz. However, the mean SPL peaked at 104 db re 1 µpa at 16 khz and maintained the highest mean SPLs of all the sampling locations at frequencies above 16 khz. The Johnson (1967) bottlenose dolphin audiogram extended above all mean sound levels to around 300 Hz, where it dropped below the mean sound pressure levels of NEL #1, NWL #1, and WL #3 (Figure 2.3). All audiograms for the dolphins (humpback

35 25 and bottlenose) followed a declining pattern as frequency increased. There appeared to be intraspecific variation in the magnitude of the difference between dolphin hearing thresholds and SPL. Notably, the difference between the humpback dolphin audiogram and the SLVF sound pressure level (WL #3) was smaller as compared to the Johnson (1967) bottlenose dolphin audiogram and the SLVF (WL #3) sound pressure level. For example, near 5.6 khz the difference for humpback dolphins was ~6 db re 1 µpa compared to ~33 db re 1 µpa for bottlenose dolphins. This corresponds to the same ~20 db re 1 µpa difference between the humpback bottlenose dolphin audiograms observed by Sims et al. (2012b). However, this interspecific difference between audiograms rapidly decreased as frequency increased, with the audiograms of the two species converging around 32 khz. While this study did not extend to frequencies above 48 khz, it should be noted that the humpback dolphin audiogram diverged from the bottlenose dolphin audiogram and increased in SPLs at frequencies above 48,000 Hz (Figure 2.2). There was also intraspecific variation in hearing thresholds between the bottlenose dolphin audiograms. The Popov et al. (2007) audiogram declined at a slower rate as compared to the Johnson (1967) audiogram. While both bottlenose dolphin audiograms show a clear continuing trend of decline, the Popov et al. (2007) audiogram appeared to begin leveling out around 48,000 Hz. The data from Johnson (1967) audiogram did not extend beyond an upper frequency limit of 45,000 Hz; likewise, Popov et al. (2007) did not record responses to frequencies below 8,000 Hz. Because of these data gaps, both

36 26 bottlenose dolphin audiograms are shown for better clarity in frequency and sound pressure auditory thresholds. Vessel sounds At most distances, sounds from the high-speed ferry (Figure 2.4), small dolphinwatching tour boat (Figure 2.5), shrimp trawler (Figure 2.6), and shipping container vessel (Figure 2.7) were louder than the corresponding mean SPLs from either NEL #1, NWL #1, NWL #2, WL #2, or SWL #2. These higher SPLs were consistent throughout the range of the frequencies analyzed, though for all vessel types except the high-speed ferry, they usually declined to levels similar to those of the background sound in frequencies 4,000 Hz. The high-speed ferry SPLs were consistently higher throughout the range of frequencies measured than the mean noise levels at the sampling location nearest to where to ferry was recorded (NEL #1) as well as the SLVF (WL #3) (Figure 2.4). Sound pressure levels tended to peak between Hz for the high-speed ferry, 2,000-5,000 Hz for the dolphin-watching tour boat, Hz for the shrimp trawler, and Hz for the shipping container vessel. The highest SPL was 131 db re 1 µpa at 500 Hz and was associated with the shipping container vessel. These peaks were associated with a range of distances from m. All four vessel types generated broadband noise ranging from 50Hz-48 khz, however the shrimp trawler and the shipping container vessel had mean SPLs that were between 10 and 30 db re 1 µpa higher (depending on the distance of the vessel) than the small dolphin-watching tour boat and high-speed ferry at frequencies < 500 Hz. The orientation of the vessel (i.e.,

37 27 either approaching or traveling away) may have affected some of the received sounds. In one case, received sounds from a shipping container vessel were higher when the vessel was going away at further distance than when it was approaching the hydrophone (Figure 2.7, see lines corresponding to away at 243 m and approaching at 218 m from Hz). There were differences between vessel-generated sounds and the dolphin audiograms similar to those described for the mean noise levels of the area in which the vessel was recorded. However, most vessel sounds exceeded the mean noise levels, increasing the differences in sound levels between vessels and dolphin audiograms. The increased difference was apparent from frequencies 50-45,000 Hz (Fig. 2.4), 2,000-5,000 Hz (Fig. 2.5), Hz (Fig. 2.6), and 50-4,000 Hz (Fig. 2.7). While noise may not appear to be audible to bottlenose dolphins around frequencies 300 Hz, some vessel SPLs reached or exceeded the bottlenose dolphin auditory threshold at lower frequencies, from Hz (Figures 2.4, 2.6, and 2.7). The Li et al. (2012) humpback dolphin audiogram did not extend below 5,600 Hz, so we were unable to determine if humpback dolphins may also show similar decreases in hearing sensitivity to the lower frequencies as exhibited by bottlenose dolphins.

38 28 Discussion Vessels contribute considerable sound levels over a wide range of frequencies to the underwater soundscape in western Hong Kong. Similar to the previous findings of Sims et al. (2012b), greater vessel traffic appeared to be associated with higher sound pressure levels. The sound pressure levels were highly variable among the six sites studied, likely due to the random selection of recordings that were analyzed which contained varying numbers of vessels within recording range during the recording sessions. Differences in sound pressure level may also be partially attributed to Beaufort sea state, as the recordings were made in sea states ranging from 0 to 3. Because the noise recordings were made over several years, seasonal differences may be responsible for some of the observed differences in sound pressure levels; however, there were not enough recordings to determine if there were any seasonal or annual differences in sound pressure levels for each of the sampling locations. Nevertheless, the present study adds quantitative support to the findings of Sims et al. (2012b) and characterizes the sound profiles of several common vessel types that are important contributors to the underwater soundscape in western Hong Kong. Three audiograms (two bottlenose and one humpback dolphin) were used to compare vessel sound outputs to dolphin hearing. The Popov et al. (2007) audiogram was an average of 13 bottlenose dolphin subjects and may be a more accurate representation for that species. Hearing threshold comparisons were limited by the existence of only one available audiogram from a single humpback dolphin and individual variation may bias the observed differences. Extrapolation of the bottlenose

39 29 dolphin audiogram to that of the humpback dolphin audiogram in the lower frequencies may be questionable based on the overlap in hearing thresholds of humpback and bottlenose dolphins. However, individual variation in audiograms exists in bottlenose dolphins (Johnson, 1967; Popov et al., 2007) and so it is reasonable to assume that humpback dolphins also likely exhibit similar individual differences. Thus, any conclusions of species differences or similarities in hearing thresholds should be made cautiously until more data are available on individual variability in the hearing thresholds of humpback dolphins. It is unknown if the observed noise levels may cause physiological damage, increased stress, or behavioral changes in the humpback dolphins of Hong Kong s waters because such data are not available. However, humpback dolphins have shown behavioral changes in response to high levels of vessel traffic (a major contributor to noise level), with greater occurrences of longer dives associated with the presence of some oncoming vessels, particularly those at high speeds (Ng and Leung, 2003). Increasing diving duration in response to oncoming and high levels of vessel traffic may also result in elevated stress levels (Weilgart, 2007). Furthermore, Australian humpback dolphins increased their whistling rates after vessels passed closely (<1.5 km); this was hypothesized to function as attempts to re-establish group cohesion by Van Parijs and Corkeron (2001b). Thus, humpback dolphins likely experience some level of increased stress and both physical and communicative behavioral changes in busy traffic environments such as the SLVF.

40 30 Due to the random nature of noise selections for analyses from the recordings made at each sampling location, the represented noise levels are a mixture of both near and far ship distances. Ships closer in proximity will generate higher sound pressure levels, thus our estimated mean noise levels are representative of sound levels recorded from the average distance of ships relative to our sampling location. This is potentially problematic in determining the effects of noise on the local dolphins, since it is presently unknown what avoidance distances dolphins maintain (or attempt to maintain) from various types of ships. Ng and Leung (2003) documented differences in humpback dolphin responses to vessel types and distance; however, they did not describe dolphin responses to specific vessel types at varying distances. They reported higher rates of vessel avoidance by humpback dolphins to high-speed vessels, but it is unknown at what distances these behavioral changes were observed. Piwetz et al. (2012) found behavioral changes, such as mean swimming speed and reorientation rate, in response to small tour boats and trawlers within 1 km. However, many of the vessels present in the SLVF are high-speed ferries, known to make abrupt entrances into and departures from vessel fairways (Piwetz et al., 2012). These HSFs could quickly increase their proximity to dolphins, and sound pressure levels can elevate rapidly, potentially startling dolphins or causing other reactions while also leaving dolphins with little time to respond. Some research also suggested that increased unpredictability in vessel movement could have stronger effects on dolphin behavior (Constantine et al., 2004; Lusseau, 2003). This highlights a need for further research on humpback dolphin behavioral responses in the presence and the proximity

41 31 of various ships to determine potential differences in behavior to various vessel types at varying distances and orientation. High-speed ferries, tour boats, and shipping container vessels are numerous in western Hong Kong, and management of their speeds and distribution are important in mitigating potential effects on the local dolphin population. Data from the Hong Kong Marine Department show that there have been significant increases in vessel traffic, particularly high-speed ferries, over the past decade (see and Chapter 3 of this thesis). This increase in high-speed ferry traffic inversely correlates with the observed decrease in dolphin abundance in parts of western Hong Kong (see Chapter 3, Fig. 3.4). Trawling vessels (including pair, stern, shrimp, and hang trawlers) have been banned in Hong Kong s waters since December 31, 2012 (WWF Hong Kong, 2012) and this has reduced the impacts of trawling on the marine habitat in Hong Kong s waters. However, trawl boats continue to transit through and continue to contribute noise to the dolphins habitat within Hong Kong. However, it is uncertain if increased underwater noise associated with increased vessel traffic is the cause of the observed decline in dolphin abundance. Therefore, it will be important to determine how yearly-resident and seasonally-sighted individual dolphins in Hong Kong s waters respond to the increased vessel traffic and high levels of noise in specific areas; the upkeep of the Hong Kong Cetacean Research Project s photo-identification catalogue and sightings database can help answer these questions.

42 32 Future research also should focus on understanding how individually-identified dolphins distribute themselves spatially relative to different types of vessels. The uncertainty in interspecific differences and/or similarities in audiogram hearing thresholds highlights the need for more Sousa spp. audiograms to help determine the sound pressure levels that are audible to some of the humpback dolphins at various frequencies. As an ultimate goal, determination of both the acute and chronic effects of different sound pressure levels on Sousa chinensis chinensis physiology, behavior, and communication throughout their range will help to assess and guide the management of anthropogenic ship disturbances to these animals.

43 Figure 2.1. Map of Lantau Island in Hong Kong showing recording stations, recording locations for various vessels and dolphin survey transect lines. HSF stands for an unidentified high-speed ferry and Wala wala is a local name associated with small dolphin-watching tour boats in the area. 33

44 1/3 octave band sound pressure level (db re 1 μpa) Frequency (Hz) Indo-Pacific humpback dolphin audiogram (Li et al., 2012) Bottlenose dolphin audiogram (Johnson, 1967) Bottlenose dolphin audiogram (Popov et al., 2007) Figure 2.2. Humpback dolphin and two bottlenose dolphin audiograms (modified from Li et al., 2012; Johnson, 1967; and Popov et al., 2007).

45 Table 2.1. Descriptions of noise recording locations LOCATION NORTHEAST LANTAU #1 (NEL1) SITE DESCRIPTION Located near the northeast corner of the Hong Kong International Airport and the construction site for the Boundary Crossing Facility (130 hectares of reclamation) NORTHWEST LANTAU #1 (NWL#1) Located in a relatively pristine area with natural coastline and little vessel traffic NORTHWEST LANTAU #2 (NWL #2) NORTHWEST LANTAU #2 (NWL #2) Located to the north of Lung Kwu Chau; adjacent to the North Lantau Vessel Fairway (NLVF) Within the Sha Chau and Lung Kwu Chau Marine Park with very little vessel traffic WEST LANTAU #3 (WL #3) Within the very busy shipping route at South Lantau Vessel Fairway (SLVF) SOUTHWEST LANTAU #2 (SWL #2) Located between the Soko Islands with very little vessel traffic 35

46 Table 2.2. Descriptions of common vessel types in Hong Kong VESSEL TYPE HIGH-SPEED FERRY SHIPPING CONTAINER VESSEL TRAWLING VESSEL DOLPHIN-WATCHING TOUR BOAT (WALA WALA) VESSEL DESCRIPTION Catamaran or hydrofoil hull shape; 25-50m in length; travel at speeds up to 50 knots (~93 km/h); propelled by diesel engines or water jets powered by jet engines; carry up to 418 people Variable in length and tonnage; mostly propelled by diesel engines; generally travel at speeds <25 knots (~46 km/h) Trawling vessels in Hong Kong include: shrimp, pair, hang, stern trawlers; generally powered by diesel engines; speed is variable depending on active fishing effort Small, skiff-like vessels propelled by an outboard engine; speeds are variable 36

47 1/3 octave band sound pressure level (db re 1 µpa) Frequency (Hz) NEL #1 Mean Noise NWL #2 Mean Noise WL #3 Mean Noise NWL #1 Mean Noise WL #2 Mean Noise SWL #2 Mean Noise Figure 2.3. Mean noise levels in six areas varying in amount of vessel traffic and types of vessels present. Error bars represent ± 1 standard error. See Table 2.1 for descriptions of noise recording locations.

48 1/3 octave band sound pressure level (db re 1 µpa) Frequency (Hz) 318 m 214 m 294 m Mean Noise (NEL #1) Mean Noise (WL #3) Indo-Pacific humpback dolphin audiogram (Li et al., 2012) Bottlenose dolphin audiogram (Johnson, 1967) Bottlenose dolphin audiogram (Popov et al., 2007) Figure /3 octave band sound pressure levels for a high-speed ferry at Northeast Lantau #1 (NEL #1); Beaufort sea state 2. Error bars represent ± 1 standard error. Bottlenose and humpback dolphin audiograms show the difference between noise levels and minimum audible levels for the dolphins.

49 1/3 octave band sound pressure level (db re 1 μpa) Frequency (Hz) 126 m 182 m 298 m Left area ~2 km Mean Noise (WL #2) Mean Noise (WL #3) Indo-Pacific humpback dolphin audiogram (Li et al., 2012) Bottlenose dolphin audiogram (Johnson, 1967) Bottlenose dolphin audiogram (Popov et al., 2007) Figure /3 octave band sound pressure levels for a small dolphin watching tour boat ( Wala wala ) in West Lantau; Beaufort sea state 1. Error bars represent ± 1 standard error. Bottlenose and humpback dolphin audiograms show the difference between noise levels and minimum audible levels for the dolphins.

50 1/3 octave band sound pressure level (db re 1 µpa) Frequency (Hz) 376 m 336 m 328 m 408 m Mean Noise (WL #2) Mean Noise (WL #3) Indo-Pacific humpback dolphin audiogram (Li et al., 2012) Bottlenose dolphin audiogram (Johnson, 1967) Bottlenose dolphin audiogram (Popov et al., 2007) Figure /3 octave band sound pressure levels for a shrimp trawling vessel at West Lantau #2 (WL #2); Beaufort sea state 3. Error bars represent ± 1 standard error. Bottlenose and humpback dolphin audiograms show the difference between noise levels and minimum audible levels for the dolphins.

51 1/3 octave sound pressure level (db re 1 μpa) Frequency (Hz) 350 m 218 m 243 m 565 m Mean Noise (WL #3) Mean Noise (SWL #2) Indo-Pacific humpback dolphin audiogram (Li et al., 2012) Bottlenose dolphin audiogram (Johnson, 1967) Bottlenose dolphin audiogram (Popov et al., 2007) Figure /3 octave band sound pressure levels for a large shipping container vessel at Southwest Lantau #1 (SWL #1); Beaufort sea state 2. Error bars represent ± 1 standard error. Bottlenose and humpback dolphin audiograms show the difference between noise levels and minimum audible levels for the dolphins.

52 42 CHAPTER 3 IMPACTS OF HIGH-SPEED FERRY ROUTE AND TRAFFIC VOLUME ON THE ABUNDANCE AND DENSITY OF CHINESE HUMPBACK DOLPHINS (SOUSA CHINENSIS CHINENSIS) IN HONG KONG S WATERS Abstract The total number of high-speed ferry trips between Hong Kong and Macau as well as Mainland China s ports increased by 35% from This increase was dominated by vessel traffic between Hong Kong and Macau particularly after the Sky Pier of the Hong Kong International Airport was opened in September An examination of temporal changes in Sousa chinensis chinensis densities at several sites near two major vessel fairways in North and South Lantau indicated that there was a significant decline in dolphin densities at Fan Lau and near the northeast corner of the Hong Kong International Airport between 2003 and Moreover, the increase in high-speed vessel traffic also correlated with the statistically significant decline in dolphin abundance in the Northwest Lantau, Northeast Lantau, and West Lantau survey areas. The ferry route from the Sky Pier, as well as the significant increase of high-speed ferry traffic between Hong Kong and Macau may have contributed to the observed decline in dolphin abundance. In light of this possibly serious problem, several mitigation measures are suggested, including the diversion of vessel traffic away from the South Lantau Vessel Fairway, attenuating marine traffic volume from the Sky Pier, as well as imposing speed restrictions along the Sky Pier route. Keywords: High-speed ferry, vessel traffic, Sousa chinensis chinensis, decline, mitigation

53 43 Introduction The effects of vessel traffic on cetaceans around the world have been well documented. Behavioural changes such as spatial avoidance, increase in swimming speed, changes in diving behaviour and acoustic behaviour have been studied extensively (e.g., Au and Perryman, 1982; Kruse, 1991; Evans et al., 1992; Janik and Thompson, 1995; Allen and Read, 2000; Van Parijs and Corkeron, 2001b). High levels of vessel traffic within cetacean habitat not only expose them to greater risk of vessel collisions (Van Waerebeek et al., 2007) but also add background noise to the underwater environment, which can result in the masking of dolphin sound production and reception and thus affecting their ability to forage and socialize (Bejder et al., 2006; Lusseau, 2005; Constantine et al., 2004). Such disturbance may lead to the displacement of dolphins from their important habitats and affect their energy and activity budgets. Humpback dolphins (Sousa chinensis chinensis) in Hong Kong live in aquatic habitats that are greatly influenced by various human activities (Hung, 2008; Jefferson et al., 2009). It has been hypothesized that the busy vessel traffic within the dolphins habitat around Lantau Island has led to increased mortality due to vessel collisions (Jefferson and Hung, 2004; Jefferson et al., 2009; Parsons, 2004). Several stranded dolphins in Hong Kong exhibited wounds that were consistent with blunt-force trauma injuries caused by collisions with vessels (Jefferson, 2000; Parsons and Jefferson, 2000), and a number of known individuals in the Hong Kong Cetacean Research Project s photo-identification catalogue also bear injuries that are almost certainly caused by boat propellers (Jefferson, 2000a; HKCRP unpublished data).

54 44 Within the dolphins habitat around Lantau Island, most of the vessel traffic is concentrated within two major vessel fairways: the North Lantau Vessel Fairway (NLVF), which includes the Urmston Road and extends to the Ma Wan Channel; and the South Lantau Vessel Fairway (SLVF) which stretches along the southern coast of Lantau Island from Chi Ma Wan Peninsula to Fan Lau. The Sky Pier, located at the northeast corner of the Chek Lap Kok Airport, opened in September 2003, and created another ferry route from this pier that is now adding significantly more high-speed ferry traffic to the NLVF. Both the NLVF and the ferry route from the Sky Pier overlap with major areas of dolphin occurrence in North Lantau waters (e.g., Lung Kwu Chau, the Brothers Islands), while the SLVF overlaps with areas of high occurrence of dolphins in West and South Lantau waters (e.g., Fan Lau) (Figure 3.1). Although vessel traffic in Hong Kong is composed of a variety of vessel types (e.g., container ships, fishing boats, high-speed and slow-moving ferries), the noise generated by high-speed ferries (e.g., catamarans and jetfoils) is thought to cause the most significant acoustic disturbance to humpback dolphins in Hong Kong, due to their fast speed and the loud, high-frequency sounds that they generate (Hung, 2008; Hung, 2011; Ng and Leung, 2003; Sims et al., 2012). Although the negative effects of vessel traffic on the dolphins habitat preference and behaviour were preliminarily examined (Ng and Leung, 2003; Hung 2008), the extent of such adverse impacts, such as acoustic disturbance, displacement from favorable foraging grounds, and long-term effects on population status remains unclear. Since 2003, there has been a 60% decline in dolphin abundance in Hong Kong s

55 45 waters (Hung, 2013), and one of the plausible explanations for such an observed decline was the increased amount of marine traffic within the dolphins habitat. To examine the impacts of high-speed ferry traffic on the dolphins in Hong Kong s waters, this chapter aims to investigate the temporal trends in dolphin density in areas near vessel traffic routes and high-speed ferry traffic. Evidence is provided on the adverse impacts of marine traffic on Sousa chinensis chinensis abundance and density in Hong Kong. Potential vessel traffic mitigation measures to deal with these impacts are discussed.

56 46 Materials and Methods Dolphin abundance Abundance of Sousa chinensis chinensis was estimated by line-transect analysis using systematic line-transect data collected by the Hong Kong Cetacean Research Project (see Hung (2010) for detailed explanation of survey methods). For the analyses, survey effort in each survey day was used as the sample. Estimates were calculated from dolphin sightings and effort data collected during conditions of Beaufort sea state 0-3 (see Jefferson, 2000a), using line-transect methods (Buckland et al., 2001). The estimates were made using the computer program DISTANCE Version 6.0, Release 2 (Thomas et al., 2009). To examine temporal trends in dolphin abundance from , the abundance values for Northeast (NEL), Northwest (NWL) and West (WL) Lantau were obtained from Hung (2013). High-speed ferry traffic volume and passenger data Information describing traffic routes, annual vessel traffic volume, and annual numbers of passengers for high-speed ferries departing from Hong Kong to major cities within the Pearl River Delta (e.g., Shekou, Shenzhen, Zhuhai, and Macau SAR) between 1999 and 2013 were obtained from the Hong Kong Marine Department through their website ( Data were recorded and graphed in Microsoft Excel 2013.

57 47 Quantitative grid analysis of dolphin density To conduct quantitative grid analysis of dolphin density, positions of on-effort sightings (i.e., sightings made while the survey vessel was on a primary transect line) of humpback dolphins were retrieved from the Hong Kong Cetacean Research Project s long-term sighting databases and then plotted onto 1 km 2 grids among six survey areas using ArcGIS 10.2 (see Figure 3.1). Dolphin density grids were then corrected for the amount of survey effort conducted within each grid. The total amount of survey effort spent in each grid was calculated by examining the survey coverage of each line-transect survey to determine how many times the grid was surveyed during the study period. For example, when the survey boat traversed a specific grid 50 times, 50 units of survey effort were counted for that grid. With the amount of survey effort calculated for each grid, the dolphin densities of each grid were then corrected (i.e., divided by the unit of survey effort). The unit for dolphin density, DPUE, represents the number of dolphins per 100 units of survey effort. Among the 1 km 2 grids that partially contained land, the percentage of sea area was calculated using GIS tools, and their DPUE values were adjusted accordingly. The following formula was used to estimate DPUE in each 1 km 2 grid within the study area: DPUE = ((D / E) x 100) / SA% Where: D = total number of dolphins observed during on-effort sightings E = total number of units of survey effort

58 48 SA% = percentage of sea area (within a grid) DPUE values are useful in examining dolphin density over different spatial resolutions. For this study, DPUE values were calculated in each 1 km 2 grid cell and temporal changes in mean dolphin density at several areas within and near high-speed ferry routes were examined from 2003 to These areas included a suite of seven grids at each of the following locations: Fan Lau near the SLVF, the northeast corner of the Hong Kong International Airport near the Sky Pier, and the eastern side of Lung Kwu Chau adjacent to the NLVF (Figure 3.10). Statistical analyses All statistical analyses (i.e., linear regressions, Pearson correlation) were performed using SYSTAT 13 ( with a significance level of α = 0.05.

59 49 Results Dolphin abundance and high-speed ferry traffic volume and passenger data From 2003 to 2013, the abundance of Sousa chinensis chinensis in the Northeast, Northwest and West Lantau survey areas declined 60% from 158 individuals to 62 individuals (Figure 3.2). This represents an overall significant decline in abundance (R 2 =0.87, F(1, 9)=58.1, p<0.001). According to the information obtained from the Hong Kong Marine Department, the total number of HSF trips serving Hong Kong, Macau, and all Mainland China s ports increased 35% from 119,810 trips in 1999 to 161,776 trips in 2013 (Figure 3.3). This significant increase in HSF trips (R 2 =0.85, F(1, 13)=75.2, p<0.001) was also correlated significantly (r=-0.81, p <0.05) with the decline in dolphin abundance in the three survey areas of western Hong Kong (Figure 3.4). The Hong Kong-Macau HSF service increased by 60.8% from 68,060 trips in 1999 to 109,422 trips in 2013 (R 2 =0.93, F(1, 13)=90.2, p< 0.001; Figure 3.5), while the HSF traffic serving Hong Kong and mainland Chinese cities (i.e., excluding Macau) increased by 46% from 1999 to 2007, but the trend reversed from 2007 to 2013 with a decline of 31% (Figure 3.5) returning to levels similar to those in Notably, the number of Hong Kong-Macau HSF trips appeared to increase steeply after 2003 (9.9% between 2003 and 2004), and again after 2007 (8.5%-11% increase per year) (Figure 3.5). The number of daily trips going to/from Macau also increased dramatically from trips per day during to 330 trips in 2010 and 2011 and decreasing slightly from to 300 trips in 2013 (R 2 =0.93, F(1,13)=89.1, p< 0.001; Figure

60 50 3.6). In contrast, the increase in the mean number of daily ferry trips to mainland Chinese cities from Hong Kong remained relatively stable from ( trips), but a notable increase occurred starting in 2004 (186 trips) and subsequently peaking in 2007 at 207 trips. The trend then reversed with a marked decline from 2007 to 2009 (207 to 156 trips), with the mean number of daily trips in reverting to the earlier levels in (Figure 3.6). The annual number of passengers using HSF services within Hong Kong and to/from Macau and China s Mainland ports increased by about 81.4% from about 15,786,000 in 1999 to 28,628,000 in 2013 (R 2 =0.96, F(1,13)=156.4, p< 0.001; Figure 3.7). The increase was mainly attributable to passengers traveling to/from Macau, which increased by % from 9,623,000 in 1999 to 20,994,000 in 2013 (R 2 =0.97, F(1, 13)=180.4, p< 0.001; Figure 3.7). In contrast, the number of passengers traveling to/from China s Mainland ports fluctuated between 1999 and 2013, however there was an overall decrease of 19.4% from 1999 to 2013 from 6,163,000 to 4,965,000 (R 2 =0.72, F(1, 13)=14.4, p=0.002; Figure 3.7). The 2013 data also showed that 77.0% of the total passengers were attributable to the Macau ferry services, which dominated the overall temporal trend. The annual number of ferry trips to/from the Sky Pier was not available from the Hong Kong Marine Department, however, indirect evidence from the annual number of passengers using this ferry terminal indicated that the volume increased 215% from approximately 273,000 passengers in 2003 to 861,400 in 2004 (Figure 3.7). Moreover, the number of passengers using the Sky Pier ferry services has continued to increase

61 51 steadily since then, reaching the highest total in 2013 of 2,666,900 passengers (Figure 3.7). This represents an increase of 878% between 2003 and 2013 and was mainly attributed to passengers traveling to ports in Mainland China (Figure 3.8). The opening of the Sky Pier ferry terminal may have also provided travelers with a more convenient route to travel to/from HK to non-macau ports; the decrease in the non-macau passenger volume appears to be offset by the increase in the number of passengers using the Sky Pier ferry terminal (Figure 3.7). Quantitative grid analysis of dolphin density Temporal changes in dolphin density were analyzed in six survey areas in western Hong Kong (Figure 3.9). In addition to the six larger survey areas, temporal changes in dolphin density were analyzed in three smaller areas (i.e., Fan Lau near the SLVF, the northeast corner of the Hong Kong International Airport near the Sky Pier, and the eastern side of Lung Kwu Chau adjacent to the NLVF; see Figure 3.10) from 2003 to 2013 (Figures ). There was a significant decline (R 2 =0.56, F(1, 7)=7.31, p=0.03) in dolphin density (number of dolphins per 100 units of survey effort) in the seven grid cells near Fan Lau between 2003 and 2010 (Figure 3.11). There was also a significant decline (R 2 =0.47, F(1, 7)=5.43, p<0.05) in dolphin density (number of dolphins per 100 units of survey effort) in the seven grid cells near the Sky Pier between 2003 and 2010 (Figure 3.12). However, there was a significant increase (R 2 =0.99, F(1, 7)=21675, p<0.05) in dolphin density

62 52 (number of dolphins per 100 units of survey effort) in the seven grid cells near the Sky Pier between 2011 and 2013 (Figure 3.12). There was no significant decline (R 2 =0.10, F(1, 7)=0.69, p=0.44) in dolphin density (number of dolphins per 100 units of survey effort) in the seven grid cells near Lung Kwu Chau between 2003 and 2010 (Figure 3.13).

63 53 Discussion In a previous study, Hung (2008) analysed densities of humpback dolphins (i.e., the number of dolphins per 100 units of survey effort, DPUE) in 1 km 2 grids within and near the NLVF and SLVF from to examine if differences exist in dolphin densities within and near vessel traffic areas. That study revealed significantly lower mean DPUE values in grids with or near traffic along a section of the NLVF, suggesting that dolphins may avoid deep-water channels within intense vessel traffic areas despite the possible better feeding opportunities offered by these structures (Hung, 2008). In the present study, temporal changes in overall dolphin densities (DPUE values) at several areas from were examined including grid cells around Fan Lau, the northeast corner of the airport near the Sky Pier, and the eastern side of Lung Kwu Chau adjacent to the NLVF. At Fan Lau, where marine traffic was most intense (based on observations during shore-based theodolite tracking work (see Hung, 2011)) among these three areas, dolphin densities dropped steadily from the highest in 2003 (n=112.1) to the lowest in 2007 (n=18.5) with a slight rebound between (n=21.2 to 57.5) (Figure 3. 11). The notable decline in dolphin density between and coincided with the marked increase in volume of high-speed ferry traffic to and from Macau, implying that the dolphins may be avoiding this busy section of the SLVF as the number of high-speed ferries increased. Interestingly, the density of dolphins rebounded after Habituation to the high-speed ferry traffic by some dolphins may explain this observation but dolphin density was still much lower in the area when compared to the density of Furthermore, it should be noted that the

64 54 annual dolphin abundance in West Lantau has decreased since 2007 (Figure 3.2), which coincides with the increase in the number of high-speed ferry trips between Hong Kong and Macau. In addition to the decline in dolphin density around Fan Lau, the entire West Lantau area may have also been affected by the increased high-speed ferry traffic in the SLVF. Near the Sky Pier, dolphin densities at the northeast corner of the airport revealed a dramatic decline between 2003 and Since then, a steady but less steep decline in dolphin usage of this area was observed from The notable decline from 2003 to 2004 coincided with the opening of the Sky Pier, as well as a significant increase in ferry traffic traveling between China s mainland ports. Although the annual number of ferry trips to and from the Sky Pier was not available from Hong Kong s Marine Department, the annual number of passengers using this ferry terminal indicated that the passenger volume increased from approximately 273,000 passengers in 2003 to 861,400 in Moreover, the number of passengers using the Sky Pier ferry services has been steadily increasing since 2003, reaching the highest total in 2013 of 2,666,900 passengers; this represents an increase of 878% between 2003 and 2013 and is likely to continue to increase especially if the third runway expansion at the Hong Kong International Airport is approved and constructed. There was an increase in dolphin density in this area between 2011 and 2013, however the cause of this increase in currently unknown. The new route from the Sky Pier has contributed to the increase in vessel traffic in the NLVF, which likely contributed to the decline in dolphin density around Lung Kwu

65 55 Chau. The estimated dolphin abundance in the Northeast Lantau (NEL) survey area also dropped significantly from 2003 to 2004, and has remained at a lower level since then. As the traffic route of the vessels departing from the Sky Pier is situated at the boundary of the NEL and Northwest Lantau (NWL) survey areas, many individual dolphins that move frequently between these two survey areas may have been seriously affected by the increased amount of HSF traffic, forcing them to spend less time in the NEL survey area. Moreover, with the additional traffic in the NLVF from the opening of the Sky Pier, dolphin abundance estimates in the NWL survey area also dropped noticeably between 2003 and 2004, with a continuing decline since then (see Figure 3.2). The annual dolphin densities to the east of Lung Kwu Chau near the NLVF, an area identified as critical dolphin habitat in Hong Kong (Hung, 2008), fluctuated between 2003 and Two apparent declining trends in dolphin densities occurred during and The first decline in coincided with the marked increase in high-speed ferry traffic to China s mainland ports from , which was likely related to the opening of the Sky Pier in As the high-speed ferries departing from and arriving at the Sky Pier also travel through the NLVF, the decline in dolphin density around the eastern side of Lung Kwu Chau, adjacent to the NLVF during could be related to the disturbance from increased HSF traffic. However, both dolphin density in the grid cells examined around Lung Kwu Chau and HSF traffic from the Sky Pier to Mainland China declined between 2007 and 2010, suggesting that there may have been other factors contributing to the decline in dolphin occurrence in the area during that time. Despite the two periods of decline prior to 2009, there was no

66 56 statistically significant overall decline between 2003 and 2013 and dolphin density around Lung Kwu Chau increased between 2003 and 2004 and again between 2010 and 2011 and has remained relatively constant (n~60) since Suggestions for mitigation measures This study provides evidence that suggests that the increasing amount of vessel traffic within humpback dolphin habitats in Hong Kong s waters may have resulted in an overall decline in dolphin abundance in, and their usage of, certain areas. Based on this observed decline, suggestions for mitigation measures are outlined below. Since the SLVF overlaps with dolphin habitat near Fan Lau, the high-speed ferry traffic route could be re-aligned and diverted further south, preferably to the south of Shek Kwu Chau and the Soko Islands, or outside of Hong Kong s waters after passing near Cheung Chau. This will provide a less disturbed habitat for humpback dolphins as well as Indo-Pacific finless porpoises (Neophocaena phocaenoides) in the inshore waters of South Lantau. By diverting vessel traffic farther offshore, a safer passage will be provided for dolphins to move from Fan Lau to the Soko Islands or for porpoises to move from the waters near Shek Kwu Chau and the Soko Islands to Chi Ma Wan Peninsula and Shui Hau Peninsula while only minimally lengthening the travel times of the ferries. These are all areas where dolphins were observed more often prior to the increased high-speed ferry traffic. However, this may direct vessels into areas with high finless porpoise (Neophocaena phocaenoides) density, thus any vessel re-routing

67 57 feasibility studies that are conducted should consider this to avoid collisions with the other resident cetacean in the waters of Hong Kong. A feasibility study on such a management plan to re-route ferry traffic should be conducted immediately. Moreover, the proposed Southwest Lantau Marine Park and Soko Islands Marine Park should be established as soon as possible (see Hung, 2008). These two proposed marine parks should also be connected to provide a larger marine protected area with a protected movement corridor to allow the dolphins to move between the parks and prevent fragmentation of these important habitats.

68 Figure 3.1. Map of six survey areas in western Hong Kong. The Sha Chau and Lung Kwu Chau Marine Park boundary and vessel fairways are shown. Each survey grid has a spatial resolution of 1km 2. 58

69 Abundance Estimate Year WL NWL NEL Figure 3.2. Temporal trends in combined abundance estimates of humpback dolphins in Northeast, Northwest, and West Lantau from (modified from Hung, 2013).

70 Number of high-speed ferries Thousands y = x - 9E+06 R² = p < Year Figure 3.3. Annual number of high-speed ferries departing from and arriving at Hong Kong ports (excluding the Sky Pier) from

71 Number of high-speed ferry trips Thousands Dolphin abundance Pearson Correlation r=-0.81 p <0.05 y = x - 9E+06 R² = p < y = x R² = p < Year HSF Dolphin abundance Linear (HSF) Linear (Dolphin abundance) Figure 3.4. Annual trend in total number of high-speed ferry (HSF) trips and dolphin abundance in Northeast, Northwest and West Lantau from

72 Number of high-speed ferries Thousands y = x - 8E+06 R² = p < Year Macau Non-Macau Linear (Macau) Figure 3.5. Annual number of high-speed ferries from Hong Kong to Macau and Non- Macau (i.e., Mainland Chinese) ports from

73 Number of trips Year Total Macau Non-Macau Figure 3.6. Mean number of daily high-speed ferry trips from

74 Number of passengers Millions Year Total Macau Non-Macau Sky Pier Figure 3.7. Annual number of passengers using high-speed ferry services within Hong Kong and to and from Macau and China s Mainland ports from

75 65 Number of passengers Millions Year Mainland Macau Figure 3.8. Annual number of passengers traveling to Mainland China and Macau ports from the Sky Pier during The Sky Pier opened in September 2003 and the 2003 data have been extrapolated from 4 months to 12 months.

76

77 Figure 3.9. Temporal trend in dolphin densities (DPUE values, i.e., number of dolphins per 100 units of survey effort) in Deep Bay (DB), Northeast Lantau (NEL), Northwest Lantau (NWL), West Lantau (WL), Southwest Lantau (SEL) and Southeast Lantau (SEL) from

78 Figure Map of three areas that overlap with vessel traffic in either the North Lantau Vessel Fairway (NLVF) or South Lantau Vessel Fairway (SLVF). 68

79 DPUE Year Figure Temporal trend in annual mean dolphin density (DPUE, number of dolphins per 100 units of survey effort) in seven 1km 2 grid cells around Fan Lau (near the South Lantau Vessel Fairway) from

80 DPUE Year Figure Temporal trend in annual mean dolphin density (DPUE, number of dolphins per 100 units of survey effort) in seven 1km 2 grid cells around the northeast corner of the Hong Kong International Airport (near the Sky Pier and adjacent to the North Lantau Vessel Fairway) from

81 DPUE Year Figure Temporal trend in annual mean dolphin density (DPUE, number of dolphins per 100 units of survey effort) in seven 1km 2 grid cells around Lung Kwu Chau (near the North Lantau Vessel Fairway) from

82 72 CHAPTER 4 PROBABILITY AND MITIGATION OF VESSEL COLLISIONS WITH CHINESE HUMPBACK DOLPHINS (SOUSA CHINENSIS CHINENSIS) Abstract The high volumes of vessel traffic using the inshore waters of western Hong Kong present a serious collision threat to the local humpback dolphins (Sousa chinensis chinensis). A number of dolphins have been observed with serious and likely fatal injuries due to vessel collisions. Successful mitigation of vessel-dolphin collisions requires an understanding of vessel traffic, dolphin densities, and how vessel strikes occur spatially and temporally. Vessel transit paths from automatic identification system (AIS) data and humpback dolphin density data (1 km 2 resolution) from 2012 and 2013 were used to determine the relative probability of commercial shipping vessels, highspeed ferries, and fishing vessels encountering humpback dolphins in the waters of Hong Kong. The waters of the South Lantau Vessel Fairway are heavily traversed by high-speed passenger ferries that service the ports of Hong Kong, Macau, and other nearby cities in Mainland China. North of Lantau Island, the Urmston Road waterway (North Lantau Vessel Fairway) is even busier with intense commercial shipping as well as high-speed passenger ferry traffic. When considering all vessel types in both 2012 and 2013, the relative probability of a vessel-dolphin encounter was greatest in the areas just northeast of the Sha Chau and Lung Kwu Chau Marine Park within the Urmston Road waterway and just south of Fan Lau, which is within the South Lantau Vessel Fairway. These high encounter probabilities suggest that these areas should be the

83 73 focus of attention and efforts to reduce the risk imposed by vessel traffic on dolphins in Hong Kong s waters. Keywords: Humpback dolphin, Sousa chinensis chinensis, Hong Kong, Automatic Information System, vessel, ship, strike, encounter, mitigation, routing

84 74 Introduction Humpback dolphins (Sousa spp.) are widely distributed in the shallow, coastal water of the western Pacific Ocean and Indian Ocean (Ross et al., 1994). Sousa chinensis chinensis in Hong Kong represent a subgroup of a larger population, likely more than 2,500, which reside in the Pearl River Estuary (Chen et al., 2010). Probably because of feeding demand, they seldom go much beyond the boundary of the brackish water (Huang et al., 1978). In Hong Kong, resident cetaceans are protected from killing and deliberate harm under the Wild Animals Protection Ordinance (Cap. 170). However, much of the water occupied by dolphins is now used for a variety of human activities, including transportation, industrial waste disposal, commercial fishing, sand extraction, harbor dredging, and commercial dolphin watching (Morton, 1996). All these activities, together with the more general urban growth along the coastal fringe, translate into a massive flow of vessel traffic in the region. The abundance of humpback dolphins in Hong Kong s waters was estimated to be about 62 in 2013, which represents a 60% decline from 152 individuals in 2003 (Hung, 2013), and this decline is likely due, at least in part, to the increasing vessel traffic in western Hong Kong (see Chapter 3). In 2013, Hong Kong handled 22.4 million twenty-foot equivalent units of container vessels (TEUs), and was the fourth busiest port in the world (Hong Kong Government Yearbook, 2013). Some 376,100 vessels arrived in and departed from Hong Kong during the year, carrying 270 million tonnes of cargo and about 28.8 million passengers (Hong Kong Government Yearbook, 2013). Due to the high amount of vessel traffic in Hong Kong s waters, S. c. chinensis are at risk of injuries and death from

85 75 collisions with vessels. Between 1993 and 1998, three stranded humpback dolphins were diagnosed to have been killed by boat strikes, and another dolphin was suspected to have been killed by a boat strike (Parsons and Jefferson, 2000). This represents 14% of all stranded humpback dolphins during this period. While Hong Kong, Macau, and other peripheral cities of the Pearl River Delta integrate into a mega-regional economy, the vessel traffic pressure on resident dolphins will inevitably increase in the future. Other Sousa populations also occur in areas of high boat traffic (e.g., Shanghai and Singapore) (Reeves et al., 2008), and the impact of shipping on humpback dolphins is a cause for concern for this species throughout its range. In Hong Kong s waters, there are two main shipping fairways, which pass directly through dolphin habitat: the North Lantau Vessel Fairway (NLVF) and South Lantau Vessel Fairway (SLVF) (Figure 4.1). The busy vessel traffic has already caused a negative impact on the animals. For example, Jefferson (2000a) reported that six dolphins (representing 2.8% of the animal catalog) indicated unmistakable evidence of propeller cuts on their bodies and 12.5% of stranding specimens had evidence of death from vessel collision. Apart from causing death and injury, vessel traffic can seriously displace dolphins and alter their behavior. Many humpback dolphins also demonstrated a strong tendency to follow fishing vessels as a foraging method (Jefferson, 2000; Parsons, 1998; Torey, 2001). Ng and Leung (2002) found that dive duration of humpback dolphins increased with proximity to vessels and increased density of vessel traffic. When dolphins were situated in an area of heavy vessel traffic or there was the presence of an oncoming vessel, they were believed to be in a state of anticipation so that longer

86 76 duration dives were performed (Ng and Leung, 2003). However, dive durations of positive, neutral, and undetermined responses were similar in length, indicating that further research is required to clarify the relationship between dive duration and external stress. Knowlton and Brown (2007) identified three primary means of reducing the likelihood of vessels striking large cetaceans including: the education of mariners; technological methods for detecting whales and alerting mariners of whales; warning whales of vessels; and changing vessel operations through altering routing and vessel speed restrictions. Several types of technological methods have been tested and implemented to reduce vessel collisions with large cetaceans (e.g., Mathews et al., 2001, Laurinoli et al., 2003; Vanderlaan et al., 2003; Brown et al, 2007; Mellinger at al., 2007) but mitigation methods for reducing vessel collisions with small cetaceans are lacking. The present study quantitatively addresses the problem of vessels striking dolphins by using vessel traffic and humpback dolphin density data to determine the relative probability of vessel and humpback dolphin encounters in Hong Kong s waters. In western Hong Kong, near Lantau Island, aggregations of humpback dolphins overlap with coastal vessel traffic patterns, including two major vessel fairways. The greatest overlap between dolphin and vessel traffic density was used as a proxy to determine the locations where the collisions are most likely to occur.

87 77 Materials and Methods Dolphin data An overview of dolphin survey methods by the Hong Kong Cetacean Research Project (HKCRP) are provided in Hung (2010), Jefferson (2000a; b) and Jefferson et al. (2002), though the relevant procedures are summarized here. The survey team used standard line-transect methods (Buckland et al., 2001) to conduct vessel surveys. The territorial waters of Hong Kong Special Administrative Region were divided into twelve different survey areas, of which line-transect survey data from six of these survey areas in western Hong Kong (i.e., Deep Bay (DB). Northwest (NEL), Northeast (NEL), West (WL), Southwest (SWL), and Southeast Lantau (SEL) (Figure 4.1) were used in this study to calculate dolphin densities. These six survey areas were analyzed because they encompassed the primary distribution of S. c. chinensis in Hong Kong s territorial waters. For each vessel survey, a 15 m inboard diesel vessel ( Standard ) with an open upper deck (about 4.5 m above water level) was used to make observations and the vessel traveled at km/h along N-S transect lines that were spaced 1 km apart. Only data collected during sea states 3 (Beaufort scale) were used in the present analysis (see Jefferson, 2000a). Dolphin density data (DPUE, number of dolphins observed per 100 units of survey effort in each 1 km 2 grid cell, where one unit of survey effort is equal to the vessel traversing into a specific grid cell one time), collected during surveys in 2012 and 2013, were plotted for each of the km 2 grid cells in the study area using ArcGIS

88 Survey effort was not uniform across each of the six survey areas, but calculations based on the DPUE counters biases that may be associated with the uneven geographic distribution of effort. Vessel traffic in western Hong Kong Automatic Information System (AIS) vessel-tracking data were obtained for 2012 and 2013 from Astra Paging Ltd. ( Due to the cost of obtaining vessel positional data for the western Hong Kong study area, one weekday and one weekend day were selected in each month for analysis (dates are listed in Appendix A). The chronologically first weekday and weekend day were selected for analysis, except in instances when either of those days fell on a public holiday (e.g., New Year s Day). In those cases, the next chronological weekday or weekend day was selected. Each record contained various fields that included date, time, a unique mobile maritime service identity (MMSI) number, latitude, longitude, course (degrees), speed (knots), ship name, call sign, draft, ship type, destination (if available) and estimated time of arrival (ETA) (if available). Only data that fell within the HKCRP survey area were used. AIS tracking is a requirement for all ships of 304 metric tonnes (300 gross tons) or more. In addition, AIS tracking is mandatory for passenger ships and tankers of or more metric tonnes (150 or more gross tons) ( and all self-propelled commercial vessels (excluding fishing and passenger vessels with

89 79 less than or equal to 150 passengers) greater than or equal to 19.8 meters (65 feet) in length and towing vessels greater than 7.9 meters (26 feet). Recreational boats or other small crafts are excluded from the AIS data. Vessel positions were recorded every 5 minutes. All vessel data were aggregated on a daily (24 h) basis (n=24 for 2012; n=24 for 2013) across the same 341 grid cells in which dolphin DPUE data and individual vessel locations were plotted using ArcGIS. Vessel locations were converted to individual vessel tracklines by using the points to line tool to connect the locations of the vessel (identified by the vessel s unique MMSI number) in chronological order. No smoothing of the vessel tracklines was conducted and the real course of the individual vessels may be slightly different from the point-to-line reconstruction in ArcGIS given the frequency of vessel positions (every five minutes) and speed of the vessel. The daily number of vessels transiting through each grid cell was calculated by spatially joining the individual vessel tracklines to the grid cell layer and summing the number of tracklines that intersected each grid cell for each day in each year. The mean daily vessel density for each grid cell was then calculated for 2012 and Estimating relative probability of encounter The probability of a dolphin encountering shipping, high-speed, and fishing vessels was calculated within a geographic information system. The resulting probability maps identified where vessel traffic is predicted to have above-average impact and

90 80 where mitigation measures may be most effective. It is assumed that the 2012 and 2013 DPUE estimates (DPUEi) provide the best estimate of relative probability, at 1km 2 resolution, that a dolphin occupies a grid cell, i relative to the other cells in a domain of n cells (simplification of the 2-dimensional nx,y grid) and is calculated as (Vanderlaan et al., 2008): P rel (Dolphin) i = DPUE i n DPUE i i=1 (1) Similarly, the relative probability that a vessel occupies a grid-cell, i relative to other cells in a domain of n cells, is calculated as: P rel (Vessel) i = V i n i=1 V i (2) where Vi is the number of vessels occupying grid cell, i (Vanderlaan et al., 2008). Using equations (1) and (2), the relative probability that a vessel and dolphin will occupy (encounter each other within) a given grid cell, i is then calculated: P rel (Encounter) i = P rel (Dolphin) i x P rel (Vessel) i n (P rel (Dolphin) i x P rel (Vessel) i ) i=1 where Prel (Encounter)i estimates are normalized such that their sums across the grid is equal to 1 (Vanderlaan et al., 2008). (3)

91 81 Spatial distribution data (e.g., Prel (Dolphin)i, Prel (Vessel)I, and Prel (Encounter)i) are presented as 1km 2 vectors (ArcGIS 10.2, ESRI). All map figures use a Transverse Mercator Projection. All statistical uncertainties are presented as ±1 standard deviation.

92 82 Humpback dolphins in western Hong Kong Results Based on the HKCRP survey grids, in 2012, NWL had the highest probability (42.6%) of observing a humpback dolphin and SEL had the lowest probability (0%) of observing a humpback dolphin (Figure 4.2a; Table 4.1). In 2013, NWL also had the highest probability (38.7%) of observing a humpback dolphin and SEL had the lowest (0%) probability of observing a humpback dolphin (Figure 4.3a; Table 4.1). Vessels in western Hong Kong An average of 838 ± 101 unique vessels was identified per day in 2012 and 820 ± 133 per day in In 2012, this included an average of 539 ± 75 cargo and tanker vessels, 33 ± 17 fishing vessels and 72 ± 15 high-speed craft. The relative probability of observing each of these three classes of vessels is shown in Figure 4.4. In 2013, this included an average of 540 ± 90 cargo and tanker vessels, 36 ± 17 fishing vessels and 67 ± 12 high-speed craft. The relative probabilities of observing each of these three classes of vessels are shown in Figure 4.5. The average number of vessels per grid cell per day in 2012 was 28 ± 7 and the average number of vessels per grid cell per day in 2013 was 27 ± 13. Approximately 62% of the cargo and tanker vessels, fishing vessels, and high-speed crafts were transiting through the NLVF and SLVF in 2012 resulting in a mean Prel(Vessel) of ±

93 83 (N=98) within the two vessel fairways (Figure 4.2b). In 2013, approximately 62% of the cargo and tanker vessels, fishing vessels, and high-speed crafts were transiting through the NLVF and SLVF resulting in a mean Prel(Vessel) of ± (N=98) within the two vessel fairways (Figure 4.3b). The remaining 38% of vessels were navigating elsewhere in western Hong Kong waters in 2012 and Approximately 20% (65 km 2 ) of western Hong Kong s waters were unoccupied by AIS-tracked vessels in 2012 and 17% (54 km 2 ) in Relative probability of a vessel encountering a humpback dolphin in the waters of western Hong Kong The highest relative probability of a vessel encountering a humpback dolphin, based on aggregated vessel number, was located within the Urmston Road waterway just northeast of the Sha Chau and Lung Kwu Chau Marine Park (Figures 4.2c and 4.3c). There were also elevated relative encounter probabilities just south of Fan Lau, within the SLVF (Figures 4.2c and 4.3c).

94 84 Discussion The highest probability of observing a humpback dolphin occurred in the Northwest and West Lantau survey areas. Therefore, it is within these areas that the presence of vessels is also assumed to pose the greatest risk to humpback dolphins. South Lantau waters were primarily and heavily transited by high-speed ferries that serviced the ports of Victoria Harbour (Hong Kong), Macau, and mainland Chinese ports. North Lantau waters were even busier due to intense commercial shipping traffic and high-speed ferry traffic to/from Hong Kong ports (including the Sky Pier at the Hong Kong International Airport) and mainland Chinese ports in the northern portion of the Pearl River Estuary. Currently, there are no restrictions on vessel speed within the Urmston Road waterway (NLVF) or the SLVF. One marine protected area, the Sha Chau and Lung Kwu Chau Marine Park (~12 km 2 ) exists in the Northwest Lantau survey area, however, according to the Marine Parks and Marine Reserves Regulation (Cap 476A, Section 10), any type of vessel is allowed to transit through the marine park but cannot exceed a speed limit of 10 knots. The vessel traffic patterns vary between the NLVF and SLVF in terms of (1) the types of vessels that use each vessel fairway, (2) the density of traffic in each fairway, and (3) the speeds of the vessels in each fairway. The SLVF is primarily dominated by high-speed ferries using the ports of Hong Kong, Macau, and mainland Chinese ports, whereas the NLVF is busier due to intense commercial shipping traffic as well as highspeed ferry traffic from Hong Kong ports. Commercial fishing vessels also use the NLVF

95 85 to some extent, but the density of these vessels is highest along the Hong Kong S.A.R territorial border with Mainland China. The most pragmatic means of reducing vessel strikes to dolphins are to: (1) reduce the probability of a vessel encountering a dolphin through modified vessel routing, (2) reduce the lethality of vessel strikes should a collision occur, through vessel speed reductions, and (3) reduce overall risk through modified routing coupled with speed restrictions (Vanderlaan et al., 2008). To reduce the risk to dolphins, the goals of vessel re-routing and/or speed restrictions should be to achieve the greatest reduction in risk, balanced by some minimal disruption to vessel operations while also maintaining safe navigation. A survey of 550 people conducted by Chan (2014) found that patrons of highspeed ferries were willing to spend an additional 23.5 minutes aboard a ferry (either due to a reduction in speed or increased travel distance resulting from re-routing) to reduce the vessel impact to the dolphins. At the same time, passengers were also willing to pay up to an additional $9.26 HKD (~$1.20 USD) surcharge on the ferry fare to support dolphin or marine conservation initiatives in Hong Kong. Based on the number of passengers using the high-speed ferry services in 2013 (N=28,628,000), this would result in an expected revenue of approximately $265 million HKD (~$34.2 million USD) that could be used to advance dolphin and marine conservation in Hong Kong. Although this is a reflection of the willingness of the public to support conservation of the declining dolphin population in Hong Kong, the Hong Kong S.A.R. government has not

96 86 yet implemented a conservation-surcharge on high-speed ferry ticket sales or any other vessel traffic mitigation measures. Chan (2014) used AIS data to calculate the average speed of high-speed ferries in the NLVF and SLVF as 28.3 knots and 39.7 knots, respectively, while the speeds of cargo and tanker vessels in both areas were generally 15 knots. Due to the high density of both dolphins and vessel traffic throughout the NLVF, re-routing traffic in this area would be extremely difficult. Thus, a speed restriction on vessels traveling through this area may be the most feasible mitigation option. A limit on the number of vessels traveling though the vessel fairways on a daily basis, in addition to the speed limit, should also be considered. 10 knots is regarded as the advisory speed limit internationally (IWC-ACCOMBAMS, 2011), and if a speed restriction on vessels traveling within the NLVF is implemented, cargo and tanker vessels will spend about 25.8 minutes more if they reduce their speed from 15 knots to 10 knots. Of course, for high-speed ferry passengers, the impacts of a speed restriction would be greater if traveling speed was reduced from about 28 knots to 10 knots in the NLVF (travel time would be almost 50 minutes greater). However, if speed restrictions are implemented, less highfrequency noise will be generated by vessels. Vessels and dolphins will also have more time to detect each other and react to avoid collisions. In the SLVF, a speed reduction for high-speed ferries from about 40 knots to 10 knots would result in an additional >50 minutes of traveling time. This would exceed the 23.5 extra minutes that surveyed passengers were willing to spend aboard a ferry

97 87 according to Chan (2014). Thus, it is suggested that the Hong Kong Marine Department consider re-routing high-speed vessels in the SLVF, preferably to the south of Shek Kwu Chau and the Soko Islands, or outside of Hong Kong s waters after passing near Cheung Chau, so that vessel operators would still be able to maintain their high speeds while avoiding areas of high dolphin density. However, this may direct vessels into areas with high densities of finless porpoises (Neophocaena phocaenoides) or other species, thus any vessel re-routing feasibility studies that are conducted should consider this to avoid creating other unforeseen issues with other cetaceans or wildlife in the waters of Hong Kong. Active acoustic devices have been successful in reducing incidental entanglements of harbour porpoises (Phocoena phocoena) (e.g., Kraus et al., 1997; Trippel et al., 1999; Culik et al., 2001) but similar acoustical devices have not been developed to alert small cetaceans, particularly delphinids, to approaching vessels. Despite the potential efficacy of using active acoustic devices, the two most practical options for reducing the likelihood of a vessel strike to a dolphin in Hong Kong may be: 1) altering vessel traffic routing in and around known dolphin habitats (to reduce the probability of dolphin encounter) or 2) reducing vessel speeds (to reduce the probability of collision and lethal injury in the case of a collision). The vessel re-routing option will reduce the occurrence, both spatially and temporally, of vessel and dolphin encounters. Although my study outlined one method to combine information on vessel traffic and dolphin distribution and density, several issues may need to be addressed before

98 88 this information is ready to use in real-world management. The analyses presented in this study relied on the temporally aggregated DPUE data that do not incorporate daily or seasonal variability in dolphin distribution patterns and thus it was assumed that the spatial probabilities associated with the dolphins in western Hong Kong remained constant. There may also be daily variations in vessel traffic related to holidays that may not have been represented in the AIS data for the days that were selected for analysis to generate the mean daily vessel densities. Many vessels were also simply missing from the estimates of vessel density due to the lack of AIS tracking of such vessels. The most important of these missing sources in the vessel density maps is small boat traffic. Small boats do not log AIS positions (see Appendix A for more information), and can exist in large numbers in certain areas for recreational fishing, boating, or dolphin watching (Scheidat et al., 2004; Erbe, 2013). Repeated disturbance from small boats has been shown to disrupt feeding in killer whales (Orcinus orca) (Williams et al., 2006) and alter the behaviour of humpback whales (Megaptera novaeangliae) (Scheidat et al., 2004). Behavioural changes (i.e., changes in dive duration, swimming speed, and reorientation rate) due to vessel disturbance have also been documented for humpback dolphins in Hong Kong s waters (Ng and Leung, 2003; Piwetz et al. 2012). However, only the Piwetz et al. (2012) study examined the impacts of small vessels (dolphin watching tour boats) on humpback dolphins in Hong Kong s waters. The maps of the present study can inform marine spatial planning efforts. Future iterations of these maps could incorporate additional data from Hong Kong s waters and the rest of the Pearl River Estuary, as long as other datasets can be modeled to account

99 89 for spatial bias in opportunistic sightings, photo-id locations, or data from nonrandomized surveys. Otherwise, managers may end up inadvertently protecting areas only where it is convenient to collect data. The percentage of dolphin habitat in Hong Kong that is affected by vessel traffic can be read off the maps, and the areas where both vessel and dolphin densities are high can also be identified in such risk maps. Such areas would identify where focused attention for marine spatial planning efforts may be most effective. In conclusion, this chapter quantified the relative dolphin-vessel encounter probabilities within Hong Kong s waters, which represented the relative risk of vessel collisions for humpback dolphins. Policy makers in Hong Kong could use this information to assess the various options available to reduce the risk of vessel-dolphin encounters. At present, there are no vessel exclusion zones in western Hong Kong and speed restrictions only exist within the small Sha Chau and Lung Kwu Chau Marine Park in Northwest Lantau.

100 Figure 4.1. Map of Lantau Island in Hong Kong identifying six survey areas for humpback dolphins. 90

101 Table 4.1. Relative probabilities of observing humpback dolphins in six survey areas in western Hong Kong in 2012 and Survey Area P rel (Dolphin) i P rel (Dolphin) i DB NEL NWL WL SWL SEL

102 a b c Figure km resolution maps of western Hong Kong illustrating the study domain (black grid), survey area delineations (black dashed lines), and Hong Kong s territorial border (solid black line) and showing the relative probability of a) observing a humpback dolphin, b) observing a vessel, and c) a vessel encountering a humpback dolphin in

103 a b c Figure km resolution maps of western Hong Kong illustrating the study domain (black grid), survey area delineations (black dashed lines), and Hong Kong s territorial border (solid black line) and showing the relative probability of a) observing a humpback dolphin, b) observing a vessel, and c) a vessel encountering a humpback dolphin in

104 a b c Figure km resolution maps of western Hong Kong illustrating the study domain (black grid), survey area delineations (black dashed lines), and Hong Kong s territorial border (solid black line) and showing the relative probability of a) observing a cargo or tanker vessel, b) observing a fishing vessel, and c) observing a high-speed craft in

105 a b c Figure km resolution maps of western Hong Kong illustrating the study domain (black grid), survey area delineations (black dashed lines), and Hong Kong s territorial border (solid black line) and showing the relative probability of a) observing a cargo or tanker vessel, b) observing a fishing vessel, and c) observing a high-speed craft in

106 96 CHAPTER 5 DISCUSSION AND CONCLUSIONS Hong Kong s waters are particularly busy with many anthropogenic activities that generate considerable amounts of noise, which can disturb marine mammals (e.g., heavy vessel traffic, construction, dredging, tourism, and fishing activities). However, the effects of underwater noise and disturbance from vessel traffic on the local resident humpback dolphins, S. c. chinensis, are not well understood. My thesis addressed five questions (see Chapter 1) related to the impacts of noise and vessel traffic on the humpback dolphins inhabiting the Pearl River Estuary (PRE) of western Hong Kong. First, I wanted to determine if noises generated by vessels common to Hong Kong s waters were audible to the humpback dolphins. An analysis of noise generated by four different types of vessels (a high-speed ferry, a dolphin-watching tour boat, a shrimp trawler, and a container vessel) within Sousa chinensis chinensis habitat in Hong Kong showed that the frequencies of noise generated by these vessels were within the hearing range of Sousa chinensis chinensis (5.6 khz to 152 khz, measured by Li et al. (2012)). The intensity of the measured noise levels (the 1/3 octave band sound pressure level) from vessels exceeded the hearing threshold levels of the species by 6-33 db re 1 µpa depending on the frequency (Chapter 2). I also wanted to characterize the underwater soundscape of western Hong Kong. An analysis of the spatial distribution of underwater noise in Hong Kong s waters revealed that among the six areas sampled, the mean noise levels did not vary significantly across areas, however, there was a high

107 97 amount of variation around the mean that was likely attributable to the number and proximity of vessels to the hydrophone during recordings (Chapter 2). The uncertainty in interspecific differences and/or similarities in audiogram hearing thresholds (see Chapter 2 discussion) also highlights the need for more humpback dolphin audiograms to refine our understanding of the sound pressure level sensitivity of the species at various frequencies. The deployment of autonomous acoustic recording devices (such as High-frequency Acoustic Recording Packages (HARPs)) to record long-term background noise levels would be an invaluable contribution to determine the sound levels in a given area. This would be particularly useful in quantifying if there are diurnal differences in noise levels in the waters of western Hong Kong due to certain types of vessels (e.g., dolphin watching vessels, fishing vessels) being less active during the evening/early morning hours. Autonomous acoustic recording devices could also be used to record the presence of dolphins (if they are vocalizing) and the data obtained from these recording devices could be paired with data from land-based observations on the dolphins movements and behaviour during daylight hours. If the recording device has multiple hydrophones, it would be possible to track individual dolphins in three-dimensional space and estimate the number of individuals in the group; this has been done with schools of other species such as Delphinus sp. (Wiggins et al., 2013). If the vessel noise is audible to the humpback dolphins, then I assumed that the dolphins would be impacted at some level and I then proceeded to look for evidence of

108 98 an observed response by dolphins to this noise (i.e., an overall decline in abundance and/or localized declines in density) in the waters of western Hong Kong and areas located within or adjacent to existing vessel traffic routes. Vessel traffic, specifically high-speed ferry traffic, was used as a proxy for the amount of noise in the waters around Lantau Island in Hong Kong. First, an analysis of high-speed ferry traffic and passenger volume data from the Hong Kong Marine Department showed a significant increase in ferry trips and usage since The total number of high-speed ferry trips between Hong Kong and Macau as well as Mainland China s ports increased by 35% from This increase was dominated by vessel traffic between Hong Kong and Macau particularly after the Sky Pier of the Hong Kong International Airport was opened in 2003 (Chapter 3). I also found that there was a strong negative correlation between the significant increase in ferry traffic and the significant decline in S. c. chinensis abundance in Hong Kong s waters (Chapter 3). While the third chapter provides evidence that is consistent with my hypothesis that increased noise from vessel traffic has led to a decline in dolphin abundance, correlation, even when strong, does not imply causation and I was unable to test and reject other alternative hypotheses about the cause of the decline. Other possible causes of this decline in dolphin abundance may include, but are not limited to, changes in prey availability, habitat loss due to coastal development projects and land reclamation, or pollution-related habitat degradation. Realizing that data may be limited both spatially and temporally to address these potential causes, they should be the focus of future research.

109 99 An analysis of dolphin density in three areas within the high-speed ferry routes also indicated localized declines in two areas: one area near the Sky Pier at the Hong Kong International airport located within the North Lantau Vessel Fairway and another area near Fan Lau located within the South Lantau Vessel Fairway (Chapter 3). However, there was not a significant decline in the area near the northeast corner of the Sha Chau and Lung Kwu Chau Marine Park, which is partially within the North Lantau Vessel Fairway. This may suggest that the marine park regulations that are currently in place (see Chapter 3 discussion) may be successfully mitigating the impacts of vessel traffic in this area. Future research should identify what factors have made the Sha Chau and Lung Kwu Chau Marine Park and effective conservation area for humpback dolphins (and potentially other species of marine life) and whether or not these features could be replicated in other marine parks if they are established in western Hong Kong. Lastly, assuming that vessel traffic has caused the observed decline in dolphin abundance since 2003 and that greater vessel traffic will lead to a greater risk of collision to the dolphins, I identified where collisions are most likely to occur. Using Automatic Identification System (AIS) vessel positional data from 2012 and 2013, I calculated the relative probability of dolphin and vessel encounters within a 1 km 2 grid around Lantau Island and speculated on how imposing hypothetical speed restrictions within two vessel fairways in Hong Kong s waters would impact the transit time of shipping (i.e., cargo/container) vessels and high-speed ferries using each of the main vessel fairways. An analysis of dolphin density and Automatic Information System (AIS) vessel data showed that vessels and dolphins have the highest probability of

110 100 encountering each other in the Northwest and West Lantau survey areas near the eastern side of Lung Kwu Chau in the Urmston Road (also known as the North Lantau Vessel Fairway) and in the waters near Fan Lau (part of the South Lantau Vessel Fairway), respectively (Chapter 4). An understanding of where vessels and dolphins have the highest probability of encountering each other could guide conservation and mitigation measures. It is imperative that conservationists, researchers, managers and policy makers understand how current traffic patterns may be contributing to 1) collision-related deaths of S. c. chinensis in Hong Kong s waters and 2) the displacement of S. c. chinensis from Hong Kong s waters. The latter also highlights the need for more trans-national research collaboration between Hong Kong S.A.R. and mainland China because it is clear the political boundaries are not respected by the dolphins as transboundary movements of individually-recognized dolphins are well documented. Thus, a better understanding of the dolphins movement and residency patterns within the Pearl River Estuary is essential to our understanding of their population dynamics, behavior, and can inform the management of the human activities that are threatening this species. Furthermore, it would be beneficial to understand how individually identified dolphins distribute themselves spatially and temporally relative to different types of vessels and in the absence of vessels. This information would be helpful in determining if certain dolphins or classes (e.g., mother-calf pairs, females vs. males, young vs. old) of dolphins actively avoid or are attracted to certain types of vessels, which may pose

111 101 differential risks. Knowing this information would be useful for the management of vessel routing and speed restrictions in Hong Kong s waters for various vessel types. Despite some limitations, a decline in overall dolphin abundance and localized declines in dolphin densities in Hong Kong s waters due to vessel traffic was supported by my research. My thesis was able to 1) characterize the sounds generated by the main types of vessels using the waters of Hong Kong, 2) demonstrate a strong negative correlation between the increase in high-speed ferry traffic and the significant decline in dolphin abundance in Hong Kong s waters since 2003, and 3) identify areas where dolphins and vessels have the highest probabilities of encountering each other. The information from this research will be an important contribution to the conservation and management of S. c. chinensis in the waters of Hong Kong.

112 102 Literature Cited Allen, M. C. and A. J. Read Habitat selection of foraging bottlenose dolphins in relation to boat density near Clearwater, Florida. Marine Mammal Science 16: Au, D. W. K. and W. L. Perryman Movement and speed of dolphin schools responding to an approaching ship. Fishery Bulletin 80: Barros, N. B., T. A. Jefferson, and E. C. M. Parsons Feeding habits of Indo-Pacific humpback dolphins (Sousa chinensis) stranded in Hong Kong. Aquatic Mammals 30(1): Bejder, L., A. Samuels, H. Whitehead, N. Gales, J. Mann, R. Connor, M. Heithaus, J. Watson-Capps, C. Flaherty, and M. Krützen Decline in relative abundance of bottlenose dolphins exposed to long-term disturbance. Conservation Biology 20: Brown, M. W., S. C. Kraus, C. K. Slay, and L. P. Garrison Survey for discovery, science, and management. In: Kraus S. D. and R. M. Rolland. (eds). The urban whale: North Atlantic right whales at the crossroads. Harvard University Press, Cambridge, MA, pp Buckland, S. T., D. R. Anderson, K. P. Burnham, J. L. Laake, D. L. Borchers, and L. Thomas Introduction to distance sampling: estimating abundance of biological populations. Oxford University Press, London. Chan, H. Y The impacts of marine traffic on the Indo-Pacific humpback dolphins (Sousa chinensis) in Hong Kong waters. Master s thesis. University of Hong Kong. 149pp. Chen, T., S. K. Hung, Y. Qiu, X. Jia, and T. A. Jefferson Distribution, abundance and individual movements of Indo-Pacific humpback dolphins (Sousa chinensis) in the Pearl River Estuary, China. Mammalia 74: Constantine, R., D. H. Brunton, and T. Dennis Dolphin-watching tour boats change bottlenose dolphin (Tursiops truncatus) behaviour. Biological Conservation 117: Culik, B. M., S. Koschinski, N. Tregenza, and G. M. Ellis Reactions of harbor porpoises Phocoena and herring Clupea herengus to acoustic alarms. Marine Ecology Progress Series 211:

113 103 Erbe, C., R. Williams, D. Sandilands, and E. Ashe Identifying modeled ship noise hotspots for marine mammals of Canada s Pacific region. PLoS One 9(3): e Evans, P. G. H., P. J. Canwell, and E. J. Lewis An experimental study of the effects of pleasure craft noise upon bottlenose dolphins in Cardigan Bay. West Wales. Pages in Evans, P. G. H., Editor. European research on cetaceans. European Cetacean Society, Cambridge. Greene Jr., C. R. 1995a. Acoustic concepts and terminology. In Marine Mammals and Noise (eds. W. J. Richardson, C. R. Greene Jr., C. I. Malme, and D. H. Thomson.), San Diego: Academic Press. pp Greene Jr., C. R. 1995b. Ambient Noise. In Marine Mammals and Noise (eds. W. J. Richardson, C. R. Greene Jr., C. I. Malme, and D. H. Thomson.), San Diego: Academic Press. pp Hildebrand, J. A Impacts of Anthropogenic Sound. In: Marine Mammal Research: Conservation beyond Crisis, J.E. Reynolds et al. (eds). The Johns Hopkins University Press, Baltimore, Maryland. pp Hong Kong Government Yearbook Chapter 13: Transport. < Accessed 3 June Huang, Z., Y. Wen, and Z. Tang Preliminary studies on Sousa chinensis. Fudan Journal (Natural Science) 1: (in Chinese). Hung, C. L. H., R. K. F. Lau, J. C. W. Lam, T. A. Jefferson, S. K. Hung, M. H. W. Lam, and P. K. S. Lam Risk assessment of trace elements in the stomach contents of Indo-Pacific Humpback Dolphins and Finless Porpoises in Hong Kong waters. Chemosphere 66(7): Hung, S. K Habitat use of Indo-Pacific humpback dolphins (Sousa chinensis) in Hong Kong. PhD dissertation. University of Hong Kong. 266pp. Hung, S. K Monitoring of marine mammals in Hong Kong waters data collection: final report ( ). An unpublished report submitted to the Agriculture, Fisheries and Conservation Department of Hong Kong SAR Government, 117pp. Hung, S. K Monitoring marine mammals in Hong Kong waters data collection: final report ( ). An unpublished report submitted to the Agriculture, Fisheries and Conservation Department of Hong Kong SAR Government, 158pp. Hung, S. K Monitoring of Marine Mammals in Hong Kong waters data collection: Final report ( ) Part 1: An unpublished report submitted to the Agriculture, Fisheries and Conservation Department of Hong Kong SAR Government. 67pp.

114 104 Hung, S. K. and T. A. Jefferson Ranging patterns of Indo-Pacific humpback dolphins (Sousa chinensis) in the Pearl River Estuary, People s Republic of China. Aquatic Mammals 30(1): IWC-ACCOBAMS Report of the Joint IWC-ACCOBAMS Workshop on reducing risk of collisions between vessels and cetaceans. < ACCOBAMS_ShipStrikesWorkshop_1.pdf> Accessed 28 May Janik, V. M Acoustic communication in delphinids. Advances in the Study of Behavior 40: Janik, V. M. and P. M. Thompson Changes in the surfacing patterns of bottlenose dolphins in response to boat traffic. Marine Mammal Science 12: Jefferson, T. A. 2000a. Population biology of the Indo-Pacific hump-backed dolphin in Hong Kong waters. Wildlife Monographs 144: Jefferson, T. A. (ed.) 2000b. Conservation biology of the finless porpoise (Neophocaena phocaenoides) in Hong Kong waters: final report. Unpublished report prepared for the Hong Kong Agriculture, Fisheries and Conservation Department. Jefferson, T. A Monitoring of Chinese white dolphins (Sousa chinensis) in Hong Kong waters biopsy sampling and population data analysis: final report. An unpublished report prepared for the Agriculture, Fisheries and Conservation Department. Jefferson, T. A., S. K. Hung, L. Law, M. Torey, and N. Tregenza Distribution and abundance of finless porpoises in waters of Hong Kong and adjacent areas of China. Raffles Bulletin of Zoology, Supplement 10: Jefferson, T. A. and S. K. Hung A review of the status of the Indo-Pacific humpback dolphin (Sousa chinensis) in Chinese waters. Aquatic Mammals 30(1): Jefferson, T. A., S. K. Hung, and P. K. S. Lam Strandings, mortality and morbidity of Indo-Pacific humpback dolphins in Hong Kong, with emphasis on the role of organochlorine contaminants. Journal of Cetacean Research and Management 8: Jefferson, T. A. and S. K. Hung An updated, annotated checklist of the marine mammals of Hong Kong. Mammalia 71: Jefferson, T. A., S. K. Hung, and B. Würsig Protecting small cetaceans from coastal development: Impact assessment and mitigation experience in Hong Kong. Marine Policy 33: Jefferson, T.A., and H. C. Rosenbaum Taxonomic revision of the humpback dolphins (Sousa spp.), and description of a new species from Australia. Marine Mammal Science. 30:

115 105 Johnson, S. C Sound detection thresholds in marine mammals. In Marine Bioacoustics (ed. W. Talvoga), New York: Pergamon Press. pp Knowlton, A. R., M. W. Brown Running the gauntlet: right whales and vessel strikes. In: Kraus SD, Rolland RM (eds). The urban whale: North Atlantic right whales at the crossroads. Harvard University Press, Cambridge, MA, p Kraus, S. D., A. J. Read, A. Solow, K. Baldwin, T. Spradlin, E. Anderson, and J. Williamson Acoustic alarms reduce porpoise mortality. Nature 388:525. Kruse, S The interactions between killer whales and boats in Johnstone Strait, B. C. Pages in Pryor, K and Norris, K. S. editors. Dolphin Societies: Discoveries and Puzzles. University of California Press. Laurinolli, M. H., A. E. Hay, F. Desharnais, C. T. Taggart Localization of North Atlantic right whale sounds in the Bay of Fundy using a sonobuoy array. Marine Mammal Science 19: Leatherwood, S. and T. A. Jefferson Dolphin and development of Hong Kong: a case study in conflict. IBI Reports 7: Li, S., D. Wang, K. Wang, E. A. Taylor, E. Cros, W. Shi, Z. Wang, L. Fang, Y. Chen, and F. Kong Evoked-potential audiogram of an Indo-Pacific humpback dolphin (Sousa chinensis). Journal of Experimental Biology 215: Liu, J. H. and P. Hills Environmental planning, biodiversity and the development process: the case of Hong Kong s Chinese white dolphins. Journal of Environmental Management 50: Lusseau, D Male and female bottlenose dolphins Tursiops spp. have different strategies to avoid interactions with tour boats in Doubtful Sound, New Zealand. Marine Ecology Progress Series 257: Lusseau, D Residency pattern of bottlenose dolphins Tursiops spp. In Milford Sound, New Zealand is related to boat traffic. Marine Ecology Progress Series 295: Marine Parks and Marine Reserves Regulation (Cap. 276A). < 532/5220A58068B5BFF EF000B8917/$FILE/CAP_476A_e_b5.pdf> Accessed 28 May Matthews, J. N., S. Brown, D. Gillespie, M. Johnson, R. McLanaghan, A. Moscrop, D. Nowacek, R. Leaper, T. Lewis, and P. Tyack Vocalisation rates of the North Atlantic right whale (Eubalaena glacialis). Journal of Cetacean Research and Management. 3:

116 106 Mellinger, D. K., S. L. Nieukirk, H. Matsumoto, S. L. Heimlich, R. P. Dziak, J. Haxel, M. Fowler, C. Meinig, and H. V. Miller Seasonal occurrence of North Atlantic right whale (Eubalaena glacialis) vocalizations at two sites on the Scotian Shelf. Marine Mammal Science 23: Morton, B Protecting Hong Kong s marine biodiversity: present proposals, future challenges. Environmental Conservation. 23(1): Ng, S. L. and S. Leung Behavioral response of Indo-Pacific humpback dolphins (Sousa chinensis) to vessel traffic. Marine Environmental Research 56: Nowacek, D. P., L. H. Thorne, D. W. Johnston, and P. L. Tyack Responses of cetaceans to anthropogenic noise. Mammal Review 37(2): Parsons, E. C. M Hong Kong s cetaceans: the biology, ecology and behaviour of Sousa chinensis and Neophocaena phocaenoides. Ph.D. dissertation. University of Hong Kong, Hong Kong SAR, 257 pp. Parsons, E. C. M The behaviour of Hong Kong s resident cetaceans: The Indo- Pacific hump-backed dolphin and the finless porpoise. Aquatic Mammals 24: Parsons, E. C. M The behavior and ecology of the Indo-Pacific humpback dolphin (Sousa chinensis). Aquatic Mammals 30: Parsons, E. C. M. and T. A. Jefferson Post-mortem investigations on stranded dolphins and porpoises from Hong Kong waters. Journal of Wildlife Diseases 36: Piwetz, S., S.K. Hung, J.Y. Yang, D. Lundquist, and B. Würsig Influence of vessel traffic on movements of Indo-Pacific Humpback Dolphins (Sousa chinensis) off Lantau Island, Hong Kong. Aquatic Mammals 38(3): Popov, V. V., A. Y. Supin, M. G. Pletenko, M. B. Tarakanov, V. O. Klishin, T. N. Bulgakova, and E. I. Rosanova Audiogram variability in normal bottlenose dolphins (Tursiops truncatus). Aquatic Mammals 33(1): Porter, L. J The taxonomy, ecology and conservation of Sousa chinensis (Osbeck, 1765) (Cetacea: Delphinidae) in Hong Kong waters. Ph.D. dissertation. University of Hong Kong, Hong Kong SAR, 202 pp. Reeves, R. R., M. L. Dalebout, T. A. Jefferson, L. Karczmarski, K. Laidre, G. O Corry-Crowe, L. Rojas-Bracho, E. R. Secchi, E. Slooten, B. D. Smith, J. Y. Wang, and K. Zhou Sousa chinensis. The IUCN Red List of Threatened Species. Version < Retrieved on 18 July Rolland, R. M., S. E. Parks, K. E. Hunt, M. Castellotes, P. J. Corkeron, D. P. Nowacek, S. K. Wasser, and S. D. Kraus Evidence that ship noise increases stress in right whales. Proceedings of the Royal Society B doi: /rspb

117 107 Romero, L. M. and L. K. Butler Endocrinology of stress. International Journal of Comparative Psychology. 20: Ross, G. J. B., G. E. Heinsohn, and V. G. Cockcroft Humpback dolphins Sousa chinensis (Osbeck, 1765), Sousa plumbea (G. Cuvier, 1829) and Sousa teuszii (Kükenthal, 1892). In S. H. Ridgway & R. J. Harrison (Eds.), The handbook of marine mammals. Volume 5: The first book of dolphins (pp ). London: Academic Press. Sapolsky, R. M., L. M. Romero, and A. U. Munck How do glucocorticoids influence stress responses? Integrating permissive, suppressive, stimulatory, and preparative actions. Endocrine Reviews. 21: Scheidat, M., C. Castro, J. Gonzalez, R. Williams Behavioural responses of humpback whales (Megaptera novaeangliae) to whalewatching boats near Islas la Plata, Machalilla National Park, Ecuador. Journal of Cetacean Research and Management 6: Sims, P., R. Vaughn, S. K. Hung, and B. Würsig. 2012a. Sounds of Indo-Pacific humpback dolphins (Sousa chinensis) in West Hong Kong: A preliminary description. Acoustical Society of America 131(1): EL48-EL53. Sims, P., S. K. Hung, and B. Würsig. 2012b. High-speed vessel noises in West Hong Kong Waters and their contributions relative to Indo-Pacific Humpback Dolphins (Sousa chinensis). Journal of Marine Biology. 2012: Thomas, L., J. L. Laake, E. A. Rexstad, S. Strindberg, F. F. C. Marques, S. T. Buckland, D. L. Borchers, D. R. Anderson, K. P. Burnham, M. L. Burt, S. L. Hedley, J. H. Pollard, J. R. B. Bishop, and T. A. Marques Distance 6.0 Release 2. Research Unit for Wildlife Population Assessment, University of St. Andrews, UK. Torey, M Study on the interactions between cetaceans and fisheries in Hong Kong waters: final report. Unpublished Report of Ocean Park Conservation Foundation, Ocean Park, Hong Kong. Tougaard, J. and L. A. Kyhn Echolocation sounds of hourglass dolphins (Lagenorhynchus cruciger) are similar to the narrow band high-frequency echolocation sounds of the dolphin genus Cephalorhynchus. Marine Mammal Science, 26: Trippel, E. A., M. B. Strong, J. M. Terhune, and J. D. Conway Mitigation of harbour porpoise (Phocoena phocoena) by-catch in the gillnet fishery in the lower Bay of Fundy. Canadian Journal of Aquatic Science 56: Urick, R. J Principles of Underwater Sound. McGraw-Hill Co., New York.

118 108 Van Parijs, S. M. and P. J. Corkeron 2001b. Boat traffic affects the acoustic behaviour of Pacific humpback dolphins, Sousa chinensis. Journal of the Marine Biological Association of the United Kingdom 81(3): Van Parijs, S. M. and P. J. Corkeron. 2001a. Vocalizations and behavior of Pacific humpback dolphins Sousa chinensis. Ethology 107: Van Waerebeek, K., A. N. Baker, F. Felix, J. Gedamke, M. Iniguez, G. P. Sanino, E. Secchi, D. Sutaria, A. Van Helden, and Y. Wang Vessel collisions with small cetaceans worldwide and with larger whales in the Southern Hemisphere, an initial assessment. Latin American Journal of Aquatic Mammals 6: Vanderlaan, A. S. M., A. E. Hay, and C. T. Taggart Characterization of North Atlantic right-whale (Eubalaena glacialis) sounds in the Bay of Fundy. IEEE Journal of Ocean Engineering 28: Vanderlaan, A. S. M., C. T. Taggart, A. R. Serdynska, R. D. Kenney, and M. W. Brown Reducing the risk of lethal encounters: vessels and right whales in the Bay of Fundy and on the Scotian Shelf. Endangered Species Research 4: Wang, J. Y., S. C. Yang, and S. K. Hung Diagnosability and description of a new subspecies of Indo-Pacific humpback dolphin, Sousa chinensis (Osbeck, 1765), from the Taiwan Strait. Zoological Studies 54: Weilgart, L. S The impacts of anthropogenic ocean noise on cetaceans and implications for management. Canadian Journal of Zoology. 85: Wiggins, S. M., K. E. Frasier, E. E. Henderseon, J. A. Hildebrand Tracking dolphin whistles using an autonomous acoustic recorder array. Acoustical Society of America. 133(6): Wild Animals Protection Ordinance (Cap. 170). < 35/927CB8EAE EE004915CB/$FILE/CAP_170_e_b5.pdf> Accessed 5 April Williams, R., C. W. Clark, D. Ponirakis, and E. Ashe Acoustic quality of critical habitats for three threatened whale populations. Animal Conservation 17(2): World Wildlife Fund Hong Kong Trawling Ban in Hong Kong Waters takes Effect from Today: Estimated 20 to 30% Increase in Population of Marine Fish in Five Years. Web. Accessed 27 February < Würsig, B Cetaceans. Science. 244(4912):

119 109 Würsig, B. and C.R. Greene Jr Underwater sounds near a fuel receiving facility in western Hong Kong: relevance to dolphins. Marine Environmental Research 54(2): Würsig, B. and W.J. Richardson Effects of Noise. In: Encyclopedia of Marine Mammals, Perrin et al. (eds). Second Edition. pp Würsig, B., C. R. Greene Jr., and T. A. Jefferson Development of an air bubble curtain to reduce underwater noise of percussive piling. Marine Environmental Research 49(1):

120 110 APPENDIX A VT Explorer Vessel Tracking System Information VT Explorer operates the largest commercial AIS network in South Europe and is one of the leading international providers of AIS vessel tracking services. Their system processes over 200 million AIS ship reports per day received from over vessels. VT Explorer is a service of Astra Paging Ltd., a software development company established in A map of VT Explorer s AIS coverage areas can be viewed here: VT Explorer provides a real-time vessel tracking service as well as historical AIS data. Historical AIS data is a valuable data source used for vessel traffic analysis, port calling information, asset tracking, risk assessment, accident investigation, etc. The data are also used for analyzing the vessels movements on a global scale, potential trends in the shipping market or vessel behaviour patterns for potential prosecution of illegal actions. As one of the leading AIS networks, VT Explorer stores all AIS data received from vessel transponders in a historical database back to 2009 and provides: 1. Vessel movement reports; 2. Video simulation; and 3. Traffic density analysis Vessel movement reports were obtained for the analyses conducted in Chapter 4 of this thesis. Information in vessel movement reports Each vessel movement report contained to following information: 1. DATE / TIME - timestamp of the position report (UTC) 2. MMSI vessel's MMSI number 3. IMO vessel's IMO number 4. NAME vessel name 5. LATITUDE

121 LONGITUDE 7. COURSE course over ground (degrees) 8. SPEED speed over ground (knots) 9. CALLSIGN vessel call sign 10. AIS TYPE vessel type (according to AIS specification) 11. DRAFT current draft of the vessel 12. DESTINATION port of call 13. ETA estimated time of arrival Spatial resolution of the AIS data The vessel movement reports provide data on the positions of thousands of ships around the world, collected by terrestrial AIS stations since VT Explorer s position reports can be generated for: 1. One or more vessels for a given period of time (on a predefined list of Maritime Mobile Service Identity (MMSI) or International Maritime Organization (IMO) numbers); or 2. All vessel movements in a particular port or area (defined by geographical coordinates) The AIS vessel movement reports were requested for the area below (Figure 1), which corresponded to the area in which Sousa chinensis chinensis surveys were done. The coordinates for this area are: Latitude: N N; Longitude: E E

122 112 Figure 1. Map of the area where AIS vessel movement reports were requested. Temporal resolution of the AIS data In most cases, typical time resolution of a vessel movement report is either 5-minutes or 1-hour (time resolution is the time between two sequential position records of each vessel). Vessel position records were obtained with a 5-minute time resolution for all vessel types. The following table includes the days selected for analysis from 2012 and 2013: Date Day of week Date Day of week January 9 th Monday January 7 th Monday January 14 th Saturday January 12 th Saturday February 1 st Wednesday February 1 st Friday February 4th Saturday February 2 nd Saturday