TURTLE HEARING CLASSIFICATION
FOR TURTLE EXCLUDER DEVICES DESIGN
ANTON YUDHANA
UNIVERSITI TEKNOLOGI MALAYSIA
TURTLE HEARING CLASSIFICATION
FOR TURTLE EXCLUDER DEVICES DESIGN
ANTON YUDHANA
A thesis submitted in fulfilment of the
requirements for the award of the degree of
Doctor of Philosophy (Electrical Engineering)
Faculty of Electrical Engineering
Universiti Teknologi Malaysia
FEBRUARY 2011
iii
To my beloved mother, father, wife, and children
iv
ACKNOWLEDGEMENT
Alhamdulillah. I praise and glorify be only to Allah SWT the Almighty, the
Most Beneficent, and the Most Merciful, whose blessings and guidance have helped
me to be able to finish this thesis. In particular, I wish to express my sincere
appreciation to my supervisor, Associate Professor Dr. Jafri Din for encouragement,
guidance, critics, advices, motivation, and friendship. Without his support and
interest, this thesis would not have been the same as presented here.
I am also indebted to Ministry of National Education (Depdiknas) Indonesia
and Ministry of Science, Technology, and Innovation (MOSTI) Malaysia for their
financial support during the term of this study. Many thanks to my colleague Mr.
Sunardi, Mr.Reza Firsandaya Malik, Lam Hok Yin, Mr. Tole Sutikno, Mrs. Lina
Handayani, and Mr. Ditto Prihadi. Also full thanks to Mr. Ahmad Syahril Mohd
Nawi for their supports. Thanks to Turtle and Marine Ecosystem Center (TUMEC)
Terengganu Malaysia for providing research infrastructure for facilities, researchers,
and staff, especially Syed Abdullah and YM Raja Bidin Raja Hassan, I am grateful
for their warm hospitality and for taking the time to share with me their knowledge
of the turtles of Malaysia. Mr. Jamil and TUMEC staffs are thanked for assisting
ABR measurements. Gratefully acknowledge to Department of Electrical
Engineering Universitas Ahmad Dahlan (UAD) Yogyakarta Indonesia for the
opportunity to pursue Ph.D. program at UTM. This thesis will not be completed
without their full encouragement and support as a solid team.
My sincere appreciation also extends to all my colleagues and others who
have provided assistance at various occasions. Their views and tips are useful indeed.
Unfortunately, it is not possible to list all of them in this limited space. I am grateful
to all my family members, especially my beloved mother Sri Kusnani. Finally,
specially thanks to my wife Erny Yuliati, my children: Enas Zakiya Yudhana, Enaya
Zahira Yudhana, and Elmas Zubair Yudhana for their encouragement and praying of
every time is my power finishing this research.
v
ABSTRACT
The process of reducing the accidental capture and subsequent mortalities of
sea turtles in regional captured fisheries through the use of Turtle Excluder Device
(TED) has been extremely important. The objectives of this study is to determine the
hearing ability for Green Turtle (Chelonia Midas), identify the green turtle and fish
behavior on various sound exposures, and determine the specification of TED using
underwater sound technique. Auditory Brainstem Response system is used to identify
the hearing threshold of the turtle. The measurements were performed at Turtle and
Marine Ecosystem Center (TUMEC) Terengganu, Perak, and Melaka Malaysia. It
was conducted at a turtle tank by employing click and tone stimulus. Turtle ages of 2
years, 5 years, 9 years, and 30 years were deployed. The measured data were
analyzed in time and frequency domains. It is found that the green turtle has a
hearing ability in the range of 50 Hz to 1200 Hz. The results showed that turtle
hearing sensitivity is about 300 Hz to 500 Hz. Also, the measurement of ambient
noise in the life habitat has been conducted. The measurements were divided into 3
different distances from the sea shore: 200 m, 400 m, and 800 m and within 3
different sea depth: 2 m, 5 m, and 10 m for each point, respectively. The frequency
and maximum magnitude of ambient noise are found to increase as the depth
increased. Finally, the behavior of the turtle and fish towards the emitted sound has
been observed. A group of signals that dispel turtles have been determined. The
types of signals are Low and High Frequency Modulation, white noise, and
sinusoidal signal. These signals do not give any respond to the fish. Therefore, this
information is very useful in the development of TED using sound.
.
vi
ABSTRAK
Proses mengurangkan kejadian penyu terperangkap dan kematian penyu
dalam industri perikanan dengan menggunakan Turtle Excluder Device (TED)
adalah sangat penting. Objektif kajian ini adalah untuk mengenalpasti kelakuan
penyu hijau (Chelonia Midas) kepada bunyi yang dipancarkan, menentukan kelakuan
penyu hijau dan ikan dalam menerima pelbagai bunyi, serta menentukan spesifikasi
TED menggunakan teknik bunyi bawah laut. Sistem ABR (Auditori Brainstem
Response) digunakan untuk mengenalpasti batas pendengaran penyu. Pengukuran
dilakukan di Turtle and Marine Ecosystem Center (TUMEC) Terengganu, Perak, dan
Melaka Malaysia. Pengukuran telah dijalankan dalam tangki penyu menggunakan
stimulus klik dan tone. Usia penyu yang digunakan adalah 2, 5, 9, dan 30 tahun.
Data terukur telah dianalisa dalam domain masa dan frekuensi. Kemampuan
pendengaran penyu hijau telah dikenal pasti pada julat frekuensi dari 50 Hz ke 1200
Hz. Hasil menunjukkan bahawa pendengaran penyu paling sensitif adalah pada julat
300 Hz ke 500 Hz. Pengukuran aras hingar di laut telah pun dilakukan. Pengukuran
dilakukan pada tiga jarak yang berbeza iatu, 200 m, 400 m, dan 800 m daripada
pantai dengan 3 perbezaan kedalaman pada masing-masing titik pengukuran iaitu, 2
m, 5 m, dan 10 m. Frekuensi dan magnitud maksimum daripada aras hingar didapati
meningkat dengan peningkatan kedalaman. Akhirnya, perilaku penyu dan ikan
terhadap bunyi yang dipancarkan telah pun diperhatikan. Kumpulan bunyi yang
boleh menghalau penyu telahpun dikenal pasti. Jenis bunyi tersebut adalah modulasi
frekuensi rendah tinggi, noise putih, dan isyarat sinusoidal. Isyarat bunyi ini tidak
memberikan sebarang kesan pada ikan. Oleh kerana itu, informasi ini sangat berguna
bagi perancangan TED menggunakan bunyi.
vii
TABLE OF CONTENTS
CHAPTER
1
2
TITLE
PAGE
DECLARATION
ii
DEDICATION
iii
ACKNOWLEDGEMENTS
iv
ABSTRACT
v
ABSTRAK
vi
TABLE OF CONTENTS
vii
LIST OF TABLES
xi
LIST OF FIGURES
xii
LIST OF ABREVIATIONS
xv
LIST OF SYMBOLS
xvii
LIST OF APPENDICES
xviii
INTRODUCTION
1
1.1 Research Background
1
1.2 Problem Statement
2
1.3 Objectives
3
1.4 Research Scope
3
1.5 Outline
3
LITERATURE REVIEW
5
2.1 Introduction
2.2 Types of Sea Turtle
5
2.2.1 Green Turtle Morphology
9
2.2.2 Evolutionary Trends and Conservation
9
2.2.3 Green Turtles Distribution in Malaysia
10
viii
3
2.3 Underwater Acoustic Units of Measurement
13
2.4 The decibel Scale in Underwater Acoustics
14
2.5 Turtle Auditory System
17
2.6 Turtle and Fish Hearing Threshold
20
2.7 Turtle Excluder Devices (TED)
25
2.8 Long Line and Gill Net
28
2.9 Sound Classification
31
2.10 Fast Fourier Transform (FFT) Algorithm
33
2.11 FFT Application on Spectral Analysis
34
2.12 Summary
35
RESEARCH METHODOLOGY
36
3.1 Introduction
36
3.2 Design of Study
36
3.3 Research Materials and Life Specimens
37
3.3.1. Research Tank
37
3.3.2. Green Turtle
40
3.3.3. Targeted Fish
42
3.3.4. SmartEp Software
43
3.3.5. Hydrophone for Recording Underwater
45
sound
3.3.6. Sound Generator
46
3.3.7. Underwater Speaker
48
3.3.8. Sound Recorder Software
49
3.3.9. Matlab Software
50
3.4 Measurements Procedures
50
3.4.1 Types of Measurement
50
3.4.2 ABR Response Measurement
50
3.4.3 Seawater Quality Measurement in Life
52
Habitat
3.4.4 Seawater Quality Measurement in the
53
Research Tank
3.4.5 Sound Profile Measurement
54
ix
3.4.6 Turtle and Fish Response Towards Sound
3.5 Spectral Analysis of The ABR Waveforms and
55
56
Sound Profile
4
3.5.1 ABR Waveforms Data Extraction
56
3.5.2 Sound Calibration Method
58
3.5.3 Sound Profile Data Extraction
58
3.5.4 Program Implementation and Analysis
60
3.6 Summary
61
RESULTS AND DISCUSSIONS
62
4.1 Introduction
62
4.2 Turtle Auditory Brainstem Response
62
4.2.1 Turtle Age 2 years
64
4.2.2 Turtle Age 5 years
67
4.2.3 Turtle Age 9 Years
69
4.2.4 Turtle Age 30 Years
71
4.3 Seawater Profile
77
4.3.1 Life Habitat Water Quality
77
4.3.2 Portable Instrument
78
4.3.3 Fixed Instrument
79
4.4 Sound Profile
81
4.4.1 Seawater Sound Profile
81
4.4.2 Research Tank Sound Profile
83
4.5 Sound Characterization
85
4.5.1 LFM Sound Profile
85
4.5.2. Sinusoidal Sound Profile
86
4.5.3. White Noise Sound Profile
88
4.6 Turtle and Fish Response
90
4.6.1 Turtle Response
90
4.6.2 Fish Response
91
4.7. Problem Encountered
91
4.8. Key Contributions
92
4.9. Summary
92
x
5
CONCLUSIONS AND FUTURE WORKS
93
5.1
93
Conclusion
5.2 Future Works
REFERENCES
Appendices A – F3
94
95
100-137
xi
LIST OF TABLES
TABLE NO.
TITLE
PAGE
2.1
Sound speed in some medium
14
2.2
The hearing ranges of fish
20
3.1
Hydrophone C304 Technical specification
45
3.2
Underwater Speaker Lubell 3400 Technical Specification
48
4.1
Turtle’s morphology in ABR measurement
61
4.2
ABR setup for green turtles ages 2 years
62
4.3
ABR response green turtles ages 2 years
64
4.4
Water quality recorded by using portable meter in life
habitat
4.5
Water quality recorded by using portable meter in the tank
4.6
Water quality recorded by using automatic meter (Half
Day)
4.7
Water quality recorded by using automatic meter (Two
days)
4.8
Water quality recorded by using automatic meter (Three
days)
76
76
77
78
79
4.9
Seawater sound profile in the life habitat
80
4.10
Seawater sound profile in the research tank
82
4.11
Frequency and magnitude of LFM sound profile
84
4.12
Frequency and magnitude of Sinusoidal Sound profile
85
4.13
Frequency and magnitude White Noise Sound profile
87
xii
LIST OF FIGURES
FIGURE NO.
TITLE
PAGE
2.1
Green turtle (chelonia midas)
5
2.2
The hawksbill turtle
6
2.3
The leatherback turtle
7
2.4
Lipas turtle
7
2.5
Sites of green turtle for hatchling in Malaysia Peninsula.
10
2.6
Maps of east Malaysia for egg and hatchling
11
2.7
The post breeding migration of three C.midas Female
(a,b,and C) and one C.midas (D) release from Ma’Daerah
12
Turtle Sanctuary Terengganu Malaysia
2.8
ABR waveforms
2.9
Diagrammatic representation of a Turtle Excluder Device
fitted to a trawl
17
25
2.9
Senangin II, research vessel, equipped with trawl net
28
2.11
Trawl net at senangin II
29
2.12
Trawl operation at Senangin II
29
2.13
Diver filming TED
30
3.1
The main site of research
36
3.2
Primary research tank with square shape
37
3.3
The second tank with round shape
37
3.4
The third tank with square shape
38
3.5
Research tank arrangement
38
3.6
Tank in cleaning
39
3.7
The species of research in the main tank
40
3.8
The green turtle places in stable condition
40
xiii
3.9
The weight measurement of turtle
41
3.10
The fish in the second tank
42
3.11
SmartEp main screen
43
3.12
Hydrophone C304
44
3.13
Generated sound software
46
3.14
Speaker type Lubell 3400
47
3.15
Sound recorder using edit audio software
48
3.16
Green turtle in ABR measurement
50
3.17
Green turtle was stationed by jacket for ABR measurement
51
3.18
Quality of Seawater measurement in life habitat
52
3.19
Quality of Seawater measurement in tank
53
3.20
Sound profile in the research tank
54
3.21
Acquired ABR data analysis flow
56
3.22
Reference sound analysis software
57
3.23
Monitored signal of water in tank profile
58
3.24
Extracted monitored signal in the tank
58
3.25
Acquired sound profile analysis
59
4.1
Green turtle morphology
61
4.2
ABR signal spectral of green turtle ages of 2 years
63
4.3
Turtle ages of 2 years hearing sensitivity
65
4.4
ABR signal spectral of green turtle ages of 5 years
66
4.5
Turtle ages of 5 years hearing sensitivity
67
4.6
ABR signal spectral of green turtle ages of 9 years
68
4.7
Turtle ages of 9 years hearing sensitivity
69
4.8
ABR signal spectral of green turtle ages of 30 years
70
4.9
Turtle ages of 30 years hearing sensitivity
70
4.10
Minimum frequencies of turtle’s hearing
72
4.11
Maximum frequencies of turtle hearing
72
4.12
Green turtle’s hearing bandwidth
73
4.13
Green turtle’s hearing range
74
4.14
Hearing ranges constellation
75
4.15
Noise floor in the life habitat(distance=200m,depth=2m)
79
4.16
Research tank sound profile
81
xiv
4.17
LFM sound profile
83
4.18
Sinusoidal sound profile
85
4.19
White noise sound profile
86
xv
LIST OF ABBREVIATIONS
ABR
-
Auditory Brainstem Response
AEP
-
Auditory evoked potentials
ANSI
-
American National Standard for Institute
ASCII
-
American Standard Code for Information Interchange
ASSR
-
Auditory Steady State Response
AULS
-
Autonomous Underwater Listening Station
CWT
-
Complex Wavelet Transform
dBnHL
-
dB scale relative to normal hearing level
DO
-
Dissolved Oxygen
EEG
-
Electro EnchepaloGraph
EFR
-
envelope following response
EFR
-
envelope following response
FFT
-
Fast Fourier Transform
hh
-
Hour
IHS
-
Intelligent Hearing System
L
-
length
LFM
-
Low Frequency Modulation
mm
-
Minute
MRTF
-
modulation rate transfer function
NMFS
-
National Marine Fisheries Service
NN
-
neural network
NOAA
-
National Oceanic and Atmospheric Administration
NPF
-
Northern Prawn Fishery
PTT
-
push to talk
RMS
-
Root Mean Square
xvi
RMSE
-
The Root Mean Squared Error
Sal
-
Salinity
sd
-
Stimulus Duration
SEAFDEC
-
South East Asian Fisheries Development Center
SNR
-
signal to noise ratio
SpCond
-
Sound propagation Conductivity
SPL
-
Sound Pressure Level
ss
-
Second
SSE
-
The Sum of Squares due to Error
Std
-
Standard Deviation
STFT
-
Short Time Fourier Transform
STFT
-
short time Fourier Transform
T
-
tall
TED
-
Turtle Excluder Devices
Temp
-
Temperature
TUMEC
-
Turtle and Marine Ecosystem Center
UTM
-
Universiti Teknologi Malaysia
W
-
width
WN
-
White Noise
xvii
LIST OF SYMBOLS
μ
-
micro
W
-
watt
m2
-
meter square
ms
-
mili second
re
-
relative
Pa
-
Pascal
L
-
liter
mg
-
mili gram
C
-
celcius
d
-
depth
ºC
-
degree celcius
Ω
-
ohm
Vdc
-
volt direct current
(Lx WxT)
-
LenghtxWidthxTall
xviii
LIST OF APPENDICES
APPENDIX
TITLE
PAGE
A
List of Author’s Publication
100
B
Source Code
101
C1
First measurement Procedure
107
C2
Second measurement Procedure
108
C3
Third measurement Procedure
109
C4
Fourth measurement Procedure
110
C5
Fifth measurement Procedure
111
C6
Sixth measurement Procedure
112
C7
Seventh measurement Procedure
113
C8
Eight measurement Procedure
114
D1
ABR and sound profile recording
115
D2
ABR System Setup and resulted data
116
D3
ABR spectral in smartEp
122
E1
Sound Profile in Life Habitat (distance=100m)
124
E2
Sound Profile in Life Habitat (distance=200m)
125
E3
Sound Profile in Life Habitat (distance=400m)
127
E4
Sound Profile in Life Habitat (distance=800m)
129
E5
Noise Floor in the life habitat (distance=200m)
131
E6
Noise Floor in the life habitat (distance=400m)
132
E7
Noise Floor in the life habitat (distance=800m)
133
E8
Peak of Magnitude for Life Habitat Noise Floor
134
F1
Sound Characterization Measurement
135
F2
Turtle response towards sound
136
F3
Fish response towards sound
137
CHAPTER 1
INTRODUCTION
1.1
Research Background
Sea turtles are important marine animals, not only under CITES (Convention
on International Trade in Endangered Species of Wild Fauna and Flora) agreement
but also traditional living resources in the ASEAN region. Most of the ASEAN
member countries have established national programs on the conservation and
enhancement of sea turtles. However, information on research, conservation and
enhancement of these animals in the region is rather fragmented.
Fisheries (Prohibition of Fishing) Regulations 1990 is released by Fisheries
Department Malaysia prohibits unsustainable fishing practices gill net (pukat hanyut)
for catching stingray or skates fish (ikan pari) at zone A or coastal marine (5 mile
from coast line) of the sea. One reason for this prohibition is the incidental captured
of the turtle in the net. Turtle and stingray have the same habitat and when the turtle
landed to reproduce and feed into the beach they would died in drawn out caused by
catch in the net. The number of landed turtle in Malaysia decrease every year.
(http://turtlemalaysia.gov.my/ancaman.html). Eckert (1999) reported that green turtle
(Chelonia mydas) is classified as endangered species, while the hawksbill turtle
(Eretmochelys imbricata) and the leatherback-turtle (Dermochelys coriacea) are
classified as critically endangered.
2
Fisheries Regulation 1990 was aimed to prevent turtle extinction when this
regulation had caused stingray fish production got decrease. The stingray fish is
quite popular in Malaysia and it is in line with the food industry variety. The demand
of this fish is increase in Malaysia. Recreational and sport fishing stingray is also
point of interest for tourism industry in Malaysia. The other benefit, its skin had
commercial values in exotic leather wear and very rare to found any of the stingray
skin wallets in Malaysia.
Meanwhile, the usage of gill net by fisheries had been monitored by local
authority because of many fisheries still used it in illegal fishing. The Turtle and
Marine Ecosystem Center (TUMEC) has been reported that in 2006 under Fisheries
Department they had confiscated more 20 gill nets in the inspection in Terengganu
reported by (http://web10.bernama.com/maritime/news). Fishermen still used the gill
net in the zone A that has much stingray population there. Plenty of fishes had been
collected, but incidentals capture of sea turtles in fishing gear is another major
problem. One interesting point to note is that hook and long lines do not seem to be
catching turtles within Malaysian territorial waters, although they are known to take
turtles in offshore areas. It is clear that prohibition is not really effective.
1.2
Problem Statements
The process of reducing the incidental capture and subsequent mortalities of
sea turtles in regional shrimp fisheries through the use of TED (Turtle Excluder
Device) has been extremely important. This is the considering to global significance
of Southeast Asia’s turtle populations and the importance of shrimp’s fisheries to
regional economies and fishing communities. Furthermore, little is known of sea
turtles' auditory behavior. Thus, the study on turtle’s hearing especially green turtle is
needed.
3
1.3
Objectives
The main purpose of this research is proposed Turtle Excluder Devices (TED)
using Sound technique. The proposed TED designed begin by conducted some basic
objectives of research step. The following objectives are:
•
To determine the hearing ability for green turtle.
•
To identify the green turtle behavior on various sound exposures.
•
To determine the specification of TED using underwater sound technique.
1.4
Research Scope
In order to achieve the research objectives, the following scopes will be
covered:
•
Measurement of turtle hearing threshold using Auditory Brainstem
Response (ABR) System.
1.5
•
Measurement of underwater sound profile.
•
Green turtle ABR data analysis.
•
Investigation of sound profile in sea water.
•
Measurement of the green turtle behavior on specific sound exposures.
Outline
The thesis is divided into five chapters. Following is an introductory chapter
that defines the importance of sea turtles, the aimed of fisheries regulations 1990, and
the usage of gill net by fisheries. Then the problem statements, objective, research
scope and thesis outline are given.
4
Chapter 2 begins by discussed on the types of sea turtles with green turtle
morphology, evolutionary trends and conservation and green turtle distribution in
Malaysia. Then, in the following section are discussed on underwater acoustic units
of measurement, the decibel scale in underwater acoustics, turtle auditory system,
turtle and fish hearing threshold, TED, long line and gill net, sound classification,
Fast Fourier Transform (FFT) algorithm. Finally, FFT application on spectral
analysis is discussed.
Chapter 3 presents the research methodology. This chapter begins by discuss
design of study, research materials and life specimens, measurements procedures,
spectral analysis of the ABR waveforms and sound profile.
Chapter 4 presents the results and discussion. The results consist of turtle
auditory brainstem response, seawater profile, sound profile, sound characterization,
turtle and fish response, and problem encountered. Finally, the contributions of the
research are given.
Chapter 5 presents the conclusion and future works. Simulated and measured
results are compared. The experimental verification process is explained with
numerical analysis given. The key contributions in this thesis are highlighted.
Finally, some recommendations on further work as well as a concluding statement
are given.
CHAPTER 2
LITERATURE REVIEW
2.1
Introduction
This chapter discusses on topics of sea turtle, turtle morphology, underwater
acoustic, turtle auditory system, turtle and fish hearing threshold, and also TED and
fishing gear. Furthermore, discussion on sound classification, FFT (Fast Fourier
Transform) Algorithm, and FFT application are presented. Emphasis is given on sea
turtle, where it is further divided into three subsections.
2.2
Types of Sea Turtle
Bonine and Smith (2003) have been reported that turtles are identified on all
continents (except in Antarctica). This species is greatest on diversity in the world,
there are 35 species dispersed in Southeastern United States while Southeastern Asia
around of 90 species.
Because of the shell and the ability to withdraw their heads and limbs into the
protective structure, thus turtles are easily to recognizable species of all reptiles.
Turtles include member of a monophyletic group comprising the order Chelonia or
Testudines. The term of Testudines is used to denote members of the order extinct
and extant, meanwhile Chelonia is often addressed extant turtles. Common
terminology is used for some groups of turtles has made some confusion. The word
‘tortoise’ is always subject to terrestrial species which have high domed shells, big
size and the high legs alike elephantine.
6
The term terrapin is used for freshwater animal that are commonly harvested
and feed, but in Britain it is used for all pool turtles. Penyu Agar or green turtle
(Chelonia midas) is depicted in Figure 2.1. This animal is member turtle species. It’s
shell length of 90 cm to 110 cm while the weight of 110 kg to 180 kg. Green turtle
could be recognized from its color olive brown to black or green. In addition,
dominant feed are sea grass or seaweeds which favorite food for this species. The
habitat of green turtle is ranging between coasts and islands in the tropics area.
Figure 2.1 Green Turtle (Chelonia Midas) (Bonine and Smith, 2003)
The Hawksbill Turtle (Eretmochelys Imbricata) is the other species of turtle
that have the local name Penyu Karah or Penyu sisik as shown in Figure 2.2. Karah
turtle has shell length of 70 cm to 90 cm with weight of 40 kg to 90 kg. Hawksbill
color is combined between dark brown and yellow. Sponge is kind of food for
Hawksbill. Moreover, the habitat for this turtle is at tropical oceans nearby coral
reefs.
7
Figure 2.2 Hawksbill Turtle (Bonine and Smith, 2003)
The other species of turtle is Leatherback (Dermochelys coriacea) as shown
in Figure 2.3. Its local name is Penyu Belimbing. Belimbing turtle has shell length of
150 cm to 180 cm and weight of 300 kg to 600 kg. In addition, the color of this turtle
is black mixed white or grey pattern. Leatherback has favorite feed jelly fish.
Meanwhile, leatherback has habitat at oceanic. In spite of the habitat is in the tropical
sites, however it moves to temperate water for food.
The identified 13 extinct families contains around of 300 species. Further,
turtles are classed into two distinctive clades, there are Cryptodira and Pleurodira. It
is presented that there are recognized method from neck retraction. The Pleurodira
or side-necked turtle would bend the head and neck to the side, meanwhile the
Cryptodira withdraw their necks in a vertical plane into their shell.
Turtles in North American are belonging the Cryptodira, meanwhile in
Arizona consists of five native species and four introduced species.
8
Figure 2.3 Leatherback Turtle (Bonine and Smith, 2003)
Lipas turtle as shown in Figure 2.4 has shell length of 50 cm to 70 cm with
weight of 30 kg to 60 kg and the color is olive brown. The favorite foods for this
species are crustaceans, mollusca, jellyfish, fish, and sea-grass. Coastal tropics area
is lipas turtle habitat.
Figure 2.4 Lipas Turtle (Bonine and Smith, 2003)
9
2.2.1
Green Turtle Morphology
Bonine et al. (2003) also reported that the turtle shell is comprised of bony
plates that are commonly shielded by with keratinized scutes. Moreover, the scutes
are the section of the shell which could be seen not the actual bones underneath. The
shell are comprised of two common portions, there are the dorsal carapace and the
ventral plastron.
The scutes are covered both halves of the shell are grouped and the features
named are useful for recognize the species. The head is covered with horny plates
whereas the neck generally lacks scales. In addition, the limb is shielded with scales
like tortoise desert or not like sonorant mud turtle. Also, for species in aquatic
habitat, the feet are recognized from webbed with clawed toes. Meanwhile, in
terrestrial habitat species is recognized from the feet lack webbing.
2.2.2 Evolutionary Trends and Conservation
Turtle evolution is the source of much inquiry and discussion. The marvelous
features of the shell from a comparative evolutionary and turtle morphology
standpoint are the evidence that the shell consist of the limb girdles and the rib cage
have been moved inside the rib cage. Somehow the ancestral form for the turtles
managed this feat and it has been so successful very little changes have been made to
the basic turtle morphology in over 200 million years. In fact, in South American
one turtle genus namely (Podocnemis) has not changed in shell morphology than 60
million years.
The evolutionary ‘conservativism’ proposed just how successful life in a
suitable of armor. Moreover, turtles have life history strategies that reflect this
morphological success. In fact, most turtle mortality occurs during the egg and
hatchling. As a result, the removal of adults for food, medicine or pet trade can have
devastating effects on wild populations.
10
Turtle populations everywhere are declining. Even though, there have been
turtle conservation successes in some sites. Turtle products are so popular in Asia
that three quarters of Asia’s native turtle populations are considered threatened and
two Chinese turtles are thought now to be extinct reported by Altyherr and Freyer
(2000) and Anon (2001).
Bonine et al. (2003) have concludes that in turtles reproductive success is
highly depended with increased age, while sexual maturity is delayed. Others report,
that sea turtle has exist for 200 millions of years and now day there are seven living
species around the world by Márquez (1990). Marques reported that C. mydas, the
Caretta caratta, and the Lepidochelys olivacea are classified as endangered species
whereas the Eretmochelys imbricata and Dermochelys coriacea have been
categorized into critically endangered by IUCN (2004).
Many human activities were responsible for the present conservation status of
sea turtles and the most evidence is on captured among them. Nowadays, some other
factors have contributed to continuation of the endangered status, a few that stand out
are the beaches’ disordered occupation (by hotels, houses, and crowding) fisheries
bycatch and sea pollution by Márquez (1990).
One significant subject have been studied is the effect of anthropogenic noise
towards sea turtles. Finally, for many years sound sources have been used by seismic
surveys to search the presence of oil and gas within the sea bottom by Jones (1999).
2.2.3
Green Turtles Distribution in Malaysia
In the Sabah and Sarawak Turtle Island, the green turtle is more widely
distributed with the most significant nesting populations occurring. Other nesting sea
shores are determined in Terengganu, primarily in Redang and Perhentian Islands,
Kemaman, and Kerteh. Furthermore, in Pahang popular nesting for green turtle are in
11
Chendor and Cherating, while in Perak is Pantai Remis. Finally, in Sabah green
turtles was most nesting at Sipadan Island as reported by Chan (2006).
Commonly, turtle landing for hatchling occurs in May until August for each
year. Some green turtle landing sites in Peninsular Malaysia is in Johor, Melaka,
Pahang, Terengganu, and Perak. Several sites for turtle hatchling in Malaysia
Peninsular are illustrated in Figure 2.5.
Figure 2.5 Sites of green turtle for hatchling in Malaysia Peninsular (TUMEC, 2009)
Meanwhile, there are two landing sites for the turtles lay their eggs in the
eastern Malaysian. Two sites in east Malaysia are Sabah and Sarawak. The main site
of turtle hatchling in East Malaysia is demonstrated in Figure 2.6.
12
Figure 2.6 Maps of green turtle in East Malaysia for egg and hatchling
(TUMEC, 2009)
Merwe et al. (2009) reported that male and female turtles remained within 30
km of the nesting beach during the breeding and inter-nesting periods, which
includes habitat beyond the 'no trawl zone' designed to protect turtles in this area. In
addition, following the breeding season, the tracked turtles migrated up to 1955 km
to four different foraging grounds such as in Vietnam, Indonesia, Peninsular
Malaysia and Borneo Malaysia. Moreover, during foraging, turtles occupied sites
threatened by human activities such as fishing and pollution.
The study also observed that the habitats used by the Ma'Daerah C. mydas
population during breeding are outside current local protection zones and extend into
unprotected international waters during migration and foraging. The study was aimed
to identify habitats used by C. mydas populations. It was obtained that the life habitat
is in a critical element of management and conservation of this endangered,
migratory species. The research concerned the need to increase offshore protection
around Ma'Daerah during the nesting season.
Finally, the study has identified the countries within South-east Asia that
Malaysia must cooperate with to ensure effective management of C. mydas
population. The information is particularly relevant to turtle sanctuary and then local
13
authority for manage in regions such as south east asia, where many coastal countries
occupy a small geographical area as depicted in Figure 2.7.
Figure 2.7 The post-breeding migration of three C.mydas females (A,B, and C) and
one C.mydas (D) released from Ma’Daerah Turtle sanctuary, Terengganu Malaysia
(vandeMerwe, 2009)
2.3
Underwater Acoustic Units of measurement
The unit in which noise is measured depends on the category of effect that is
of interest. Primary and secondary effects with the exception of auditory injury are
most commonly encountered with impulsive noise such as blast and have been
14
found by Yelverton (1972) to be associated with the impulse (I) or integral of
pressure over time, given by
(2.1)
where:
I is the impulse in Pascal-seconds (Pa.s)
P(t) is the acoustic pressure in Pa of the sound wave at time t,
t is time.
It is unfortunate that the term “an impulse” is also used to describe a sound
of short duration, but both terms are common in the literature. Impulse is thought of
as the average pressure of the noise impulsive then multiplied by it’s duration.
2.4
The deci Bel Scale in Underwater Acoustic
Sunardi (2010) has been reported that it is usual to express noise, whatever
the unit of measurement, in terms of deciBels (dB). The dB relates to the
measurement of noise to a reference unit. It expresses logarithmically the ratio
between the measurement and the reference unit. The word “level” is applied to any
unit expressed using the dB scale. Therefore, in a sound of peak pressure Pm (Pa) the
Sound Pressure Level (SPL) in dB is given by
(2.2)
where
Pref is the reference pressure, which for underwater applications is usually taken as 1
micro Pascal (µPa). For instance, a blast wave of 1 bar (105 Pa) would have a sound
pressure level, referred to 1 µPa, of
15
(2.3)
The reference unit must be specified when quoting a level. It is common to
append the unit as in the example or to specify the default reference unit within a
report. All measurements, unless it is otherwise indicated, are referred to 1 µPa as
reported by Nedwell (2003). In Rogers and Triyett (2007), sound speed in water is
about five times higher than that in air and the distances among the two ears in fish
are no more than a few centimeters. However, fundamentally different direction
mechanism must used by fish.
In addition, many things can effects the speed of sound, including not only
the nature of the medium (gas, liquid or solid) but also it’s temperature and any other
additive substances, such as salt in water. Sound propagates faster through denser
and hotter materials as listed in Table 2.1.
Table 2.1: Sound Speed in some Medium (http://www.earthlife.net/fish/hearing)
Medium
Temperature (0C)
Speed (m/s)
Air
0
331.4
Air
20
343.6
Air
30
348.7
Fresh Water
Unknown
1,493
Sea Water
Unknown
1,533
Diamond
Unknown
12,000
Acoustic is science of sound. In underwater acoustics, the unit of loudness is
dB. By definition, the decibel is a relative unit, not an absolute unit with physical
dimension. The dB is simply numerical scale to compare the values of like quantities,
usually power or intensity. A dB is one tenth of bel (B). In many situations, however,
the bel (B ) proved inconveniently large, so the dB has become more common.
16
Bel was devised by engineers of the Bell Telephone Laboratory to quantify
the reduction in audio level over a 1 mile (approximately 1.6 km) length of standard
telephone
cable,
named
in
honor
of
the
Bell
System’s
founder
and
telecommunications pioneer, Alexander Graham Bell. Acousticians introduced the
dB to devise of compressed scale to represent the large dynamic range of sounds
experienced by people from day to day, and also to acknowledge that humans-and
presumably other animals-perceive loudness increases in logarithmic fashion, not
linear.
Its logarithmic nature allows very large or very small ratios to be represented
by a convenient number, in a similar manner to scientific notation. An intensity ratio
of 10 translates into a level difference of 10 dB, 100 translate into a level difference
of 20 dB, and 1000 translates into a level difference of 30 dB, and so on.
The term “level” usually implies a decibel scale. The level difference in
decibels associated with two sound pressure values (measured in the same medium)
is determined by calculating the ratio of the pressures, squaring this number, taking
the logarithm (base 10), and multiplying by 10.
⎛ X ⎞
X dB = 10 log10 ⎜
⎟
⎝ Xo ⎠
X = 10
XdB
10
x X0
(2.4)
(2.5)
Where X0 is reference level and X is measured level.
If one chooses a standard reference value, then sound pressure levels can be
specified in decibels relative to that reference, but this should be stated along with
number, for clarity. In water, acousticians use a standard reference sound pressure of
1 micro pascal (i.e. 10-6 newton per square meter), abbreviated μPa.
The reference level itself is always at 0 dB, as shown by setting X = Xo in the
above equation. If X is greater than Xo then XdB is positive, if X is less than Xo then
17
XdB is negative. Since it expresses a ratio of two quantities, it is a dimensionless. A
level difference of +3 dB is roughly double (2x) power, -3 dB is roughly a half ( ½x)
power, and 10 dB is 10x power. The ratios of 0 until 10 dB are 1.0, 1.25, 1.6, 2.0,
2.5, 3.2, 4.0, 5.0, 6.3, 8.0, and 10.0 respectively.
The dB is not linear. Only summation and subtraction operator can be
deployed in the calculation of dB unit. Mean, deviation and other statistic value of
data (using x and : operator) must be converted and calculated in the linear unit.
In underwater acoustics, a source level usually represents the sound level at a
distance of 1 meter from the source, while received level is the sound level at the
listener’s actual position. There are two parameters of sound that relate to its
loudness, there are pressure and intensity.
Sound sensors (transducers) respond directly to pressure since the pressure
changes are caused by particle vibrations. In physical terms, pressure is force per
area (Newton/m2 = Pascal = Pa). 10-6 Pa = μPa is the standard unit of sound. Power is
measured in Watts and intensity in W/m2. Sound intensity is proportional to the
square of pressure.
I=
P2
A
(2.6)
Where p is power and A is area.
Measures of acoustic levels are pressure (force/area), total power in acoustic
wave, and intensity (power/area) which proportional to pressure squared. The unit
are Newton/m2 = Pa, Watts and Watts/m2 respectively.
2.5
Turtle Auditory System
Auditory Evoked Potential (AEP) signals are transient electrical bio-signals
produced by various regions of the human brain in response to auditory stimuli such
18
as a periodic repetition of “clicks”. These signals are traditionally categorized into an
ABR which occurs for 11 ms after the stimulus, followed by a Mid Latency cortical
Response (MLR) which is typically confined to the next 70 ms, followed by a slow
cortical response which starts at about 80 ms after the stimulus. ABR signals have a
waveform morphology which typically exhibits five waves (peaks) in the 1.5 to 6 ms
post-stimulus interval studied by (Elvir et al., 2005; Jafari et al., 2005). In addition,
ABR signals have waveform within range of 1.2 to 7.5 ms.
Intelligent Hearing Systems (2007) reported that the ABR waveforms are
achieved from the species in order to asses and determine the cranial auditory nerve
function. In humans, similar techniques have been examined for assessing the
viability of cochlear implantation or the existence of any physical conditions aimed
to prevent implantation succeed.
Also stated, that the basic test of hearing consists of determining the ability of
a species for hearing pure tones at intervals throughout its hearing range. The
assessment is examined by training a species for respond towards a tone and then
reducing the tone’s intensity until response of the species is fail.
Figure 2.8 ABR waveforms (Jacquin et al., 2005)
19
Recorded ABR have a waveform which typically demonstrates five waves or
peaks in post-stimulus interval as depicted in Figure 2.8. The description of the
human ABR was made by Jewett and Williston in 1971. Their study have described
that method for acquiring recorded ABR from the acquired EEG are stated in Elvir et
al. (2005). Meanwhile, Parentel et al. (2006) studied that there is an overlap of those
frequencies and the audible frequency range of sea turtles which perceive sounds
from 60 to 1000 Hz.
Sneary et al. (2007) provided a basis method for analyzing auditory hair cell
frequency selectivity in turtle. Their studied presented that recorded ABRs from
multiple animals and life stages of live sea turtles deploying stimulus which coupled
directly to the turtle's tympanum. Moreover, hearing ability of turtle was indicated
morphometrically by testing variations in auditory anatomy and physiologically by
techniques of brainstem evoked potential was conducted by Ketten et al. (2005)
Lovell et al. (2004) have been studied that the sound reception mechanism
and ability of hearing of the prawn (P. serratus). Moreover, the studied also
proposed method deploying a combination of electron microscopic, anatomical, and
electrophysiological approaches, revealing that P. serratus was responsive to
perceived sounds is ranging from 100 - 3000 Hz. Other study by Bartol and Ketten
(2006) examined that recorded ABR successfully determined for 10 turtles (juvenile
and sub-adult of C. mydas and juvenile of L. kempi) and 2 tuna (Thunnus albacares)
by applying correlation technique for identifying the response towards perceived
sound.
Furthermore, Hecox and Galambos (1974) proposed estimation of the
modulation rate transfer function (MRTF) by applying method to measure the
temporal resolution of the auditory system implementing AEP. In addition, the AEP
is also useful to asses the hearing ability and temporal resolution of species. The
method requires no training of the species and was applied for assessing hearing
response in infants of human.
20
Popper (2000) hearing capabilities of several other clupeid species have
identified using ABR in order to ascertain whether ultrasonic hearing is determined
in Clupeiformes or whether it is only determined in a limited number of animals.
Moreover, the response of AEP to a single click or tone stimulus is actually the
summation of neurological response from multiple sources as obtained for human.
This reported by Kuwada et al. (2002).
Hearing capabilities is tested using evoked potentials by amplitude is
modulated of a tone at a typical rate. As the response of patient’s auditory system to
the tone, it follow the envelope of the loud-soft modulation of the tone by a
corresponding an envelope following response (EFR).
The MRTF provided by determining the amplitudes of the EFR at various
modulation rates. A typical MRTF is low pas in shape and the corner frequency of
that MRTF is inferred as the temporal resolution of the patient by Supin et al. (2001).
2.6
Turtle and Fish Hearing Threshold
Rogers and Triyett (2007) studied that fish can determine the direction and
range of underwater sound at frequencies range of 100 Hz to 1000 Hz even noise
background is presented. Lovella et al. (2005 reported that the statocyst of P.
serratus is sensitive to the motion of water particles displaced by low frequency
sounds of 100 Hz to 3000 Hz, with a hearing acuity similar to that of a generalist
fish. Meanwhile, Higgs et al. (2001) reported that there were no differences in
auditory threshold, bandwidth, or best frequency for zebrafish.
The Risso’s Dolphin followed perceived sound stimuli up to 1000 Hz with a
second peak response at 500 Hz. A weighted MRTF reflected that the animal
followed a broad range of rates from 100 Hz to 1000 Hz, but beyond 1250 Hz
hearing response of animal was simply an offset or onset respond by Mooney et al.
(2006).
21
In addition, amount of 12 hatchling through sub-adult turtles (Lepidochelys
kempi, Chelonia mydas, Caretta caretta) shows that juvenile green turtles have a
slightly broader hearing range of 100 Hz to 800 Hz with best sensitivity at 600 Hz to
700 Hz whereas sub-adults are ranging from 100 Hz to 500 Hz. Furthermore, Kemp's
Ridleys had a more restricted is ranging from 100 Hz to 500 Hz with most sensitive
hearing of 100 Hz to 200 Hz is examined by Ketten et al. (2005).
Fish can hear a wide range of sounds, however there is a great variety of
capabilities between species as studied by Lovell et al. (2005). The study also
reported that both of the structures involved in sound reception and the abilities of
hearing of the paddlefish and the lake sturgeon using a combination of physiological
and morphological approaches, revealing that both fish are responsive to sounds
ranging in frequency from 100 Hz to 500 Hz.
Moreover, the lowest hearing thresholds from both species were acquired
from frequencies in a bandwidth of between 200 Hz and 300 Hz, with higher
thresholds at 100 Hz and 500 Hz. For brief hearing ranges of fish is shown in Table
2.2.
Table 2.2: The Hearing Ranges of Fish (http://www.earthlife.net/fish/hearing.html)
Commercial Name
Scientific Name
Hearing Range (Hz)
Atlantic Salmon
Salmo salar
40 to 350
Bonito/Tuna
Euthynnus affinis
100 to 900
Red Piranha
Pygocentrus natereri
80 to 1,500
Goldfish
Carasius auratus
40 to 3,200
Brown Bullhead
Amereius nebulosus
100 to 4,000
Stone Moroko
Pseudorasbora parva
100 to 8,000
Atlantic Cod
Gadus morhua
20 to 38,000
American Shad
Alosa sapidissima
200 to 180,000
Gulf Menhaden
Brevoortia patronus
200 to 1,800
Howorth (2003) reported that the collective knowledge of sound detection
and marine hearing capabilities is limited. Much of knowledge on the hearing
22
frequencies of marine animals is based on the range frequency at which they vocalize
rather than the range at which they actually hear. Furthermore, fisheries, which has
jurisdiction over most marine mammals and all sea turtles in the region, has not set
any standard measurement for safe sound levels. Thus, this reason is caused limited
knowledge of true hearing thresholds of such animals and their sensitivity to various
sound levels especially of frequencies and durations.
As mentioned above, investigations into the hearing ability of marine fish
have most often yielded results exhibiting a narrower hearing range and less sensitive
hearing than specialists. This was first demonstrated in a variety of marine fish by
Tavolga and Wodinsky (1963), and later other research demonstrated in
taxonomically and ecologically diverse marine species reviews by (Fay, 1988;
Popper et al., 2003; and Ladich and Popper, 2004).
By examining the morphology of the inner ear of bluefin tuna (Thunnus
thynnus), Song et al. (2006) hypothesized that this species probably does not detect
sounds much over 1 kHz. This research concurred with the few other studies
deployed on tuna species. Iversen (1967) tested that yellowfin tuna (T. albacares)
can detect sounds from 50 Hz to 1100 Hz, with best sensitivity of 89 dB (re 1 μPa) at
500 Hz. Also, Kawakawa (Euthynnus affinis) appeared to be able to detect sounds
from 100 to 1100 Hz but with best sensitivity of 107 dB (re 1 μPa) at 500 Hz.
Additionally, Popper (1981) examined at the inner ear structure of a skip jack
tuna (Katsuwonus pelamis) and determined it to be typical of a hearing generalist.
Despite of only a few species of tuna have been studied, and in a number of fish
groups both generalists and specialists exist, it is reasonable to suggest that unless
bluefin tuna are exposed to very high intensity sounds from which they cannot swim
away and short and long term effects is not significant or nonexistent.
Song et al. (2006) observed that damselfish able to hear frequencies of up to
2000 Hz, with best sensitivity well below 100 Hz. Meanwhile, Egner and Mann
(2005) found that juvenile sergeant major effects of mid and high frequency sonar on
fish. In addition, damselfish (Abudefduf saxatilis) were most sensitive to lower
23
frequencies 100 to 400 Hz, moreover, larger fish (greater than 50 millimeters)
responded to sounds up to 1600 Hz. Also, the sergeant major damselfish is
considered to have poor sensitivity in comparison even to other hearing generalists.
Kenyon (1996) studied another marine generalist, the bicolor damselfish
(Stegastes partitus), and observed responses to sounds up to 1600 Hz with the most
sensitive frequency at 500 Hz. Further, larval and juvenile Nagasaki damselfish
(Pomacentrus nagasakiensis) were tested to hear at frequencies between 100 Hz and
2000 Hz, however, they are most sensitive to frequencies below 300 Hz by Wright et
al. (2005 and 2007).
Moreover, female oyster toadfish (Opsanus tau) apparently use the auditory
sense to detect and locate vocalizing males during the breeding season reported by
Winn (1967). Interestingly, female midshipman fish (Porichthys notatus) go through
a shift in hearing sensitivity depending on their reproductive status. Reproductive
females showed temporal encoding up to 340 kHz, while non-reproductive females
showed comparable encoding only up to 100 kHz by Sisneros and Bass (2003).
The hearing capability of Atlantic salmon (Salmo salar) indicates relatively
poor sensitivity to sound was studied by Hawkins and Johnstone (1978). Laboratory
experiments yielded responses only to 580 Hz and only at high sound levels. The
Atlantic salmon is considered to be a hearing generalist and this is probably the case
for all other salmonids studied to date based on studies of hearing (Popper et al.
2007; Wysocki et al. 2007) and inner ear morphology by Popper (1976 and 1977).
Furthermore, investigations into the inner ear structure of the long-spined
bullhead (Taurulus bubalis, order Scorpaeniformes) have suggested that these fish
have generalist hearing abilities, and this is supported by their lack of a swim bladder
by Lovell et al. (2005). Despite of it was impossible to extrapolate from this species
to all members of this large group of taxonomically diverse fishes, studies of hearing
in another species in this group, the leopard robin (Prionotus scitulus), suggested that
it was probably not able to detect sound much above 800 Hz, indicating that it shall
be a hearing generalist determined by Tavolga and Wodinsky (1963).
24
However, since the leopard sea robin has a swim bladder and the long spined
bullhead does not, this illustrates the diversity of species in this order and makes
extrapolation on hearing from two fish to all members of the group very difficult
been identified.
At the same time, in considering potential sources that are in the mid and
high-frequency range, a number of potential effects are clearly eliminated. Most
significantly, since the vast majority of fish species studied to date are hearing
generalists and cannot hear sounds above 500 Hz to 1500 Hz (depending upon the
species), there are not likely to be behavioral effects on these species from higher
frequency sounds.
Moreover, even those marine species that may hear above 1500 Hz, such as a
few sciaenids and the clupeids (and relatives), have relatively poor hearing above
1500 Hz as compared to their hearing sensitivity at lower frequencies. Thus, it is
reasonable to suggest that even among the species that have hearing ranges that
overlap with some mid and high frequency sounds, it is likely that the fish will only
actually hear the sounds if the fish and source are very close to one another.
Furthermore, since the vast majority of sounds that are of biological relevance
to fish are below 1000 Hz (Zelick et al. 1999; Ladich and Popper, 2004), even if a
fish detects a mid or high frequency sound, these sounds will not mask detection of
lower frequency biologically relevant sounds.
Thus, a reasonable conclusion, even without more data, is that there will be
few, and more likely no, impacts on the behavior of fish. At the same time, it is
possible that very intense mid and high frequency signals, and particularly
explosives, could have a physical impact on fish, resulting in damage to the swim
bladder and other organ systems. However, even these kinds of effects had only been
shown in a few cases in response to explosives, thus only when the fish had been
very close to the source.
25
Such effects have never been shown to any Navy sonar. Moreover, at greater
distances there appears to be little or no impact on fish, and particularly no impact on
fish that do not have a swim bladder or other air bubble that would be affected by
rapid pressure changes.
2.7
Turtle Excluder Devices (TED)
Actually, the process of reducing the incidental capture and subsequent
mortalities of sea turtles in regional fisheries using TEDs has been extremely
important, especially for stingray and shrimp fisheries. How to this is the case due to
global significance of Southeast Asia’s turtle populations and the importance of
stingray fisheries to regional economies and fishing communities. Most of the
ASEAN member countries have established national programs on the conservation
and enhancement of sea turtles such as turtle sanctuary. However, information on
research, conservation and enhancement of these species in the region is rather
fragmented.
The inventor of TED was Sinkey Boone, a Georgia shrimper known for
inventing Turtle Excluder Devices (TEDs) in 1969 published by Marine Turtle
newletter (2000). Since the TED, was first introduced to the U.S. shrimp fishery in
the late 1980's, research and development to improve TED performance has
continued. Scuba divers and video cameras have been attached to shrimp trawls
under actual working conditions, National Marine Fisheries Service (NMFS) gear
researchers working with shrimp fishermen and net shops have made improvements
to the hard or rigid-style TED system, improving performance for both turtle
exclusion and shrimp retention.
26
Figure 2.9 Diagrammatic representation of a TED fitted to a trawl Net.
(http://www.dpi.qld.gov.au/fishweb/10559.html)
Shrimp fishermen throughout the Southeastern U.S. have contributed to
improvements in TED design and techniques for handling TEDs at sea. As seen in
Figure 2.9, green turtle are sometimes caught in nets used in trawler fisheries,
probably because they are herding fish, which enter the gears of the trawl net. If the
green turtles do not leave the net in time and they fail go to the water surface to
breathe, thus they will drown.
Troeger et at. (1995) proposed the AusTED that was tested at five selected
locations, which ranged from shallow, estuarine, mud-bottom sites to deep-water,
oceanic, sand-bottom sites. The studied obtained that no significant differences were
observed in prawn catch rates between control and AusTED equipped nets. Sea
turtles and large stingrays were excluded from the AusTED equipped net and noncommercial by catch was significantly reduced at most sites trawled.
Brerwer et al. (1998) has shown that there are BRDs that can be of major
benefit to prawn trawl fishers in the NPF, by excluding much of the unwanted by
catch while maintaining catches of commercially valuable prawns. The studied
reported some benefits to the NPF fishery of using BRDs are such as fewer or no sea
turtles caught or killed, fewer small fish to be sorted from catches, and fewer large
animals (stingrays, sharks and sea turtles) were catch.
27
The examination also showed, resulting in a higher catch value and decreased
fishing impacts on by catch species, which helps to maintain ecological biodiversity
and resilience. Finally, failure of the NPF and other Australian prawn trawl fisheries
to voluntarily adopt BRDs may result in an involuntary adoption brought on by
pressures from community, conservation and trade bodies.
Meanwhile, McGilvray et al. (1998) reported The AusTED II reduced by
catch (including large animals) and slightly reduced byproduct. Variations in prawn
catches were dependent on the area being fished. In addition, large fluctuations in net
drag precluded any detailed analysis of changes in drag as a result of fitting the
AusTED II to standard commercial trawl gear. Moreover, the AusTED II equipped
net required no extra assistance or vigilance from the crew when tested under
commercial conditions.
Robins and McGilvray (1999) have proposed the modified version (the
AusTED II) and then tested under a range of conditions typical of prawn trawl
fisheries of north eastern Australia. Trials occurred during commercial trawling
operations to heighten industry awareness of research into BRDs and improve the
credibility of research results to industry. The results determined average by catch
reduction ranged between 15% and 49% depending upon fishery conditions, and the
capture of large animals such as stingrays and sea turtles was significantly reduced.
The other studied has reported that turtle repelling with sound is possible and
can be made practical by Marine Turtle Newsletter (2000). Meanwhile, Lenhardt
(2002) proposed continued exposure to existing high levels sound (noise) in vital sea
turtle habitats and any increase in noise could affect sea turtle behavior and ecology.
Brewer et al. (2006) have reported that nets with a combination of a turtle
excluder device and by catch reduction device reduced the catches of turtles by 99%,
sea snakes by 5%, sharks by 17.7%, rays by 36.3%, large sponges by 85.3%, and
others by catch by 8%, however, these results were largely attributable to the
influence of the turtle excluder devices.
28
Finally, the research also concluded that the introduction of TEDs and BRDs
into the NPF has almost fully removed the fishery's impact on sea turtles and greatly
reduced its impact on many of the highest risk sharks and rays
2.8
Long Line and Gill Net
The usage of gill net by fisheries had been monitored by local authority
because of many fisheries still used it in illegal fishing. TUMEC has been reported
that in 2006 under Fisheries Department they had confiscated more 20 gill nets in the
inspection in Terengganu published by http://web10.bernama.com/maritime/news
(2007). Fisheries still used as the gill net in the zone that has much stingray
population there.
Plenty of fish had been collected, but incidentals capture of sea turtles in
fishing gear is another major problem. The major incidental by catch problem exists
with gill nets that very effective in catching turtles and had banned, however the
fishermen still use them illegally. One interesting point to note is that hook and long
lines do not seem to be catching turtles within Malaysian territorial waters, although
they are known to take turtles in offshore areas. It is clear that prohibition is not
really effective
A gillnet consists of netting attached between a head-rope and a footrope. The
net is kept open vertically by the differences in buoyancy between the two ropes. The
head-rope is given positive buoyancy by using various floating devices. In shallow
waters, floating is typically applied by attachable cork or styrene floats or by using
head-ropes where styrene is embedded in the rope. For deep-water fisheries, hollow
metal or hard plastic rings are used to provide buoyancy.
Gillnets are very size selective there are specific mesh size tends to catch fish
of a limited size range. The mesh size may therefore be considered the most
important characteristic of a gill net. Mesh sizes are either given in bar length or as
29
stretched mesh. The bar length measure is often used by commercial fishermen and
by net manufacturers.
The results of this study shall be applied by attaching to gill net at the
research vessel. The vessel used for research collaboration between UTM and
SEAFDEC is Senangin II as seen in Figure 2.10.
Long lining is a hook and line fishery in which long lengths of baited hooks
are laid on the ocean floor to catch halibut, sablefish, rockfish, dogfish and other
species of ground-fish. In order to set long-line gear, a buoy topped with a marker
flag is tossed overboard. By attached net to the length of line at least as long as the
depth of water into which the gear is being set.
Figure 2.10 Senangin II, Research Vessel, equipped with trawl net (SEAFDEC,
2008)
When enough line has been set out, an anchor is tied onto the buoy line, a
skate (or fixed length) of ground-line is secured to the anchor and the ground-line
begins running over the stern. Within set intervals along the ground-line, baited
hooks are attached by shorter lengths of line–a meter or so – known as gangions
reported by http://www.westcoastaquatic.ca/fisheries_overview.htm#Trawling
30
Commercially used hooks vary considerably in size, ranging in gape-width
from less than 0.5 cm to more than 10 cm in some shark fisheries. The bait is
believed to be the single most important factor determining the catch efficiency of
long-lines as reviewed by http://www.fao.org/docrep/005/x7788e/X7788E02.htm
Trawling or dragging is a commercial fishing method in which a trawl vessel
drags a cone-shaped net with a rectangular opening through the water to trap fish.
Trawling is used to take a wide variety of species in a number of separate fisheries
including shrimp, ephausiids, scallops and ground-fish. Two types of trawling
systems are used, the otter trawl and the beam trawl. The trawl in research Senangin
II vessel is shown Figure 2.11 and Figure 2.12.
Figure 2.11 Trawl net at Senangin II
Figure 2.12 Trawl operation at Senangin II
31
Since the TED was first introduced to the U.S. shrimp fishery in the late
1980's, research and development to improve TED performance has continued. TED
system, improving performance for both turtle exclusion and shrimp retention as
depicted in Figure 2.13.
Green Turtles are sometimes caught in nets used in trawler fisheries, probably
because they are herding fish which enter the gears of the trawl net. If the green
turtles do not leave the net in time it cannot go to the water surface to breathe and so
they drown.
Figure 2.13 Diver Filming TED (Marine Turtle Newsletter, 2000)
2.9
Sound Classification
Ahlstrom et al. (2006) used neural network classification for Systolic Heart
Murmur. The data were analyzed aim to find a suitable feature subset for automatic
classification on murmurs heart. McKinney and Breebaart (2003) have studied
models of human auditory processing begins with a bank of band-pass filters which
represent the frequency resolution of the peripheral human auditory system.
Seramani et al. (2006) using Independent Component Analysis (ICA) followed by
32
wavelet de-noising, dolphin whistles can be extracted successfully from a noisy
underwater environment.
The extraction of ABR from an EEG (Electro Encephalograph) signal did by
Jacquin et al. (2005). It is based on adaptive filtering of signals in the wavelet
domain, where the transform used is a nearly shift-invariant Complex Wavelet
Transform (CWT). They compare the algorithm to two existing methods. The first
simply consists of band-pass filtering the input EEG signal followed by linear
averaging. The second method uses signal-adaptive filtering in the Fourier domain
based on phase variance computed at each spectral component of the FFT.
Culver et al. (2007) have directed toward passive sonar and classification,
continuous wave (CW) and broadband signals, shallow water operation, both
platform-mounted and distributed systems, and frequencies below 1000 Hz.
Moreover, Davey et al. (2007) has classified ABR models by generated using time,
frequency and cross-correlation measures.
Furthermore, the classification of process both artificial neural networks
(NNs) and the decision tree algorithm was acquired. Process used a ratio of post
stimulus to pre stimulus power in the time domain, with power measures at 200 Hz,
500 Hz, and 900 Hz in the frequency domain.
Hyunh et al. (1998) used 300 examples of whale signals and 300 examples of
porpoises signals for the training of the classifier. The feed forward neural network is
used for the classifier. Wavelet representation is used for feature extraction.
Okumura et al. (2006) studied that classification of the vocalization patterns of
dugong calls and discusses the stability call sequence within and across local
population of dugongs. State transition probability of call sequence has been used to
classify the vocalization patterns.
Pires et al. (2007) have been developed constitutes a signal processing and
classification toolbox for exploratory data analysis of underwater sound. The sound
may be saved in several formats DAQ and WAV or raw acquisition data processed
33
feature vector in Matlab. Singleton and Poulter (2007) have been acquired peaks in
the estimated spectrum at each integer multiple of the fundamental frequency.
Elvir et al. (2005) introduced a new algorithm of adaptive filtering in the
time-frequency domain using a specific Wavelet Transform while traditionally ABR
extraction had been performed band-pass filtering followed by averaging as well as
adaptive filtering in the Fourier domain. Lenhardt (2002) introduced a turtle alerting
system and method utilizes an acoustic signal that may be provided low frequency
range, mid-frequency range, or high frequency range. One or more of those acoustic
signals may be accompanied and or modulated by either an ultrasonic signal or a
MHz signal.
Sakas et al. (2005) reported a sanctuary equipped with an Autonomous
Underwater Listening Station (AULS). Each AULS has the capability of recording
58 hours of continuous sound at depth with each deployment. One of the purposes of
the program is to identify vocalizations produced by marine mammals, invertebrates
and fish as well as anthropogenic sounds. Pompei (2006) introduced an acoustic
warning or alerting system for directing an audible warning signal to at least one
intended recipient, while reducing the chance that the warning signal could be
detected by others within the proximity of the system.
2.10
Fast Fourier Transform (FFT) Algorithm
Fourier analysis is useful for data analysis, as it breaks down a signal into
constituent sinusoids of different frequencies. It is particularly used in area such as
signal processing by (Graps, 1995; Shaker, 2005).
Meanwhile, Boashash and O’shea (1990) have been studied on analysis
technique for some underwater acoustic signals using time and frequency analysis.
Later, to deal with the noise, the non stationary and the diversity on the signals in the
34
underwater environment have been proposed time-frequency methods to constitute
an attractive solution by Iona et al. (2008).
There is better way to compute the Fourier transform of discrete data called
the FFT. The FFT was a truly revolutionary algorithm that made Fourier analysis
mainstream and made processing of digital signals commonplace. The power of the
FFT is that it allows computing the Fourier coefficients faster. The FFT has become
such a commonplace algorithm that it is built into Matlab. The coefficient FFT is
quite complex.
Use of complex numbers introduces some mathematical simplicity in Fourier
transform algorithms and provides a convenient representation. Real numbers are
often represented on the real number line and complex numbers are often visualized
on the two dimensional complex plane. In the complex plane it is clear to see that the
absolute value is simply the distance of the complex number from the origin.
The real part of the FFT corresponds to the cosines series and the imaginary
part corresponds to the sine. When taking an FFT of a real number, data set the
positive and negative frequencies turn out to be complex conjugates
2.11
FFT Application on Spectral Analysis
The study on the de noised light averaged ABR signals has been obtained by
process average valued in the Fourier domain by Fridman et al. (1982). Therefore,
Bailey et al. (1998) have been deployed short time Fourier Transform (STFT) using
a variety of “windows” with different relative advantages to address principally
difficulties in analyzing short term transient sound behavior.
The study on time and frequency classification of ABR response has been
conducted by Davey (2007). The study is demonstrated that time domain features of
recording ABRs were extracted and output to a text file for use by the time domain
35
software. Thus, FFT amplitude and power spectra of the time domain waveforms
were then obtained.
Jacquin et al. (2005) have been proposed a new method for the extraction of
human ABR from an EEG signal based on adaptive filtering signals in the wavelet
domain. The study was proposed that filtering used signal adaptive filter in the
Fourier domain based on phase variance have been computed at each spectral
component of the FFT. Storey (2006) presented two things that are different about
the FFT implementation in Matlab than the presentation the FFT uses complex
numbers and the FFT computes positive and negative frequencies.
2.12
Summary
In this chapter, types of sea turtles have been reviewed. From the literature
reviews, the green turtle morphology, evolutionary trends and conservation and the
green turtle distribution in the Malaysia have been studied.
Also underwater
acoustics have been reviewed that focused on units of measurement and the decibel
scale. The discussion on turtle auditory system and its hearing threshold have been
reported, meanwhile the fish hearing threshold also presented. Furthermore,
application of gill net by the fishermen has been discussed. Finally, the FFT
application in spectral analysis was reviewed.
36
CHAPTER 3
RESEARCH METHODOLOGY
3.1
Introduction
The
discussion
on
turtle
morphology,
underwater
acoustic,
ABR
measurement, recent TED, and sound classification have been reviewed in previous
chapters. In this chapter, the discussion is focused on such listed topics on design of
study, research materials and marine species, measurement method, and FFT
analysis. Also, procedure of measurement was conducted and FFT algorithm to
extracted data is explained.
3.2
Design of Study
The primary research has been conducted at Turtle and Marine Ecosystem
Center (TUMEC) Rantau Abang Dungun Kuala Terengganu Malaysia. Meanwhile,
the secondary research conducted at TUMEC Masjid Tanah Melaka and Segari
Perak. The secondary research aimed to enrich knowledge on green turtle and its life
habitat. All of measurements have been examined within Marc 2008 to November
2009.
The research is collaboration between Radio and Communication
Engineering Department, Fakulti Kejuruteraan Elektrik, Universiti Teknologi
Malaysia and TUMEC Malaysia. The main site of measurement is shown in Figure
3.1.
37
Figure 3.1 The main site of research
3.3
Research Materials and Life Specimens
The materials in this study were divided into two categories namely research
infrastructure and research instruments. The research infrastructure consist of the
main site of measurement, the main research tank with square shape , and two round
shape tanks with different radius. Meanwhile, the research instruments include
measuring instruments of ABR System, hydrophones, underwater speakers, sound
generators, and processing software. The primary specimen consists of two species;
there are turtles and fish that related to the research objectives. In addition, further
information of the materials and life specimen in this study is briefly described
below.
3.3.1 Research Tank
There were three fiber tanks is used in the research. The main research tank is
the biggest with square shape. The dimension of the main tank is length = 14 m,
38
width = 2.5 meters, and high = 1.50 m. The main research tank is shown in Figure
3.2.
Figure 3.2 Primary research tank (square shape)
The second tank is round shape with radius 2 m. The second tank is shown in Figure
3.3.
Figure 3.3 The second tank with round shape
39
The third tank is in square shape with dimension (length=2 m, wide=1m, height=80
cm). The third tank is shown in Figure 3.4.
Figure 3.4 The third tank with square shape
The water supply is obtained directly from nearby sea shore. Seawater
chemical profile was monitored regularly. Beside primary tank, there are also other
smaller tanks available for other purposes. The arrangement of these tanks is shown
in Figure 3.5.
Figure 3.5 Research Tank arrangement
40
TUMEC has regular schedule for monitoring the seawater profile in the tank.
Prior the research, seawater in the tank was conditioned similar with life habitat. The
regularly maintenance of the research tank is shown at Figure 3.6.
Figure 3.6 Tank in cleaning
3.3.2 Green Turtle
The species for this research is green turtle which chosen in various ages
aimed to obtain the hearing ability for different ages. It was housed in an open-air
and covered pool. The seawater in the research tank is kept at constant level with the
chemical , temperature, and volume of the water was monitored regularly.
The turtles were fed daily and the tanks maintained at 250C to 300C as normal
condition for its habitat. The turtles were kept on a light: dark cycle and monitored
daily. Turtles use under supervision of the TUMEC. The turtles ages 30 years was
used in this research is shown in Figure 3.7. The weight, shell width, and total length
were determined for each species used in this study
41
Figure 3.7 The species of research in the main tank
One of the primary sites for turtle conservation is TUMEC. The turtle jacket
has been employed to keep the turtle in stable position. This method was conducted
for turtle ages of 9 years and 30 years. Meanwhile, for the turtle ages of 2 years and 5
years the manner for keep stable position of the turtle by tight it. Deploying process
of the tightened and keep in stable position is shown in Figure 3.8.
Figure 3.8 The green turtle placed in stable condition
42
Turtles were measured for their dimension prior the ABR measurement was
conducted. The process of weight measurement is shown in Figure 3.9.
Figure 3.9 The Weight Measurement of turtle
3.3.3 Targeted Fish
There are several types of the fish were used in this study. The fish are main
commodity among the fishermen in Terengganu. Types of fish in this study were
namely Selar boops/selar kuning (Oxeye scad) and Megalaspis cordyla/cencaru
(Torpedo scad), Alapes djedaba/selar papan (Shrimp scad), and Decapterus
meruadsi/ selayang(Japanese scad) as reported by Sunardi (2010).
The fish were hatched and raised until the time of shipping, in main research
tank where the maneuvering was unrestricted. In the main tank the fish were put with
turtles together. Once in the main site of the research, the fish were categorized into
uncrowded (10 fish per tank) and crowded (more than 30 fish per tank) groups and
was kept in until used in experiments.
43
As fish were used in the measurements they were conditioned so that the
density of the fish in the tank remained unchanged. The water in the tank maintained
daily for monitored the quality of the water, especially the dissolved oxygen and
temperature by deploying aerator system. The fish use was under the supervision of
TUMEC Dungun Kuala Terengganu.
The purpose of the experiment using fish species was observed the behavior
of fish towards sound. This step was examined to ensure to the emitted sound has no
impact on fish. Fish use were tested in the measurement is shown in Figure 3.10.
Figure 3.10 The fish in the second tank
3.3.4 SmartEP Software
The instrument used to measure turtle hearing ability is SmartEP by
Intelligent Hearing System (2007). SmartEP is one of IHS System feature. The main
screen is composed of three functionality areas: the control panel, page menu, and
the main menu; and four informational areas: the indicator bar, the stimulus
information area, the SNR and noise graph and the waveform display area.
44
Up to two hundred and fifty objects may be shown on-screen for
manipulation, and only one object can be active at a time. The active recording may
be moved, filtered, have its peaks labeled, and have many other operations applied to
it. Although this limit imposes a restriction on how many recordings are displayed at
any time, the total number of recordings that can be acquired for each object is only
limited by the size of your computer’s hard drive. The main screen of SmartEP is
shown in Figure 3.11.
Figure 3.11 SmartEp Main screen
Once data has been acquired, it is needed to use either the ASSR analysis
window or the ASSR audiogram window to analyze the data. The ASSR analysis
window show the polar plot and the frequency domain representations of the
response which give the user a better visual understanding of the responses and noise
conditions of a recording. This window also show a table of numeric data, which is
used by the software to determine the presence of a response.
45
The ASSR audiogram windows build an audiogram based on the responses
that are currently being displayed on the page. Using the information from each of
the recordings, the window will show the estimated threshold in SPL.
3.3.5 Hydrophone for Recording Underwater Sound
A hydrophone is a microphone designed to be used underwater for recording
or listening to underwater sound. Most hydrophones are based on a piezoelectric
transducer that generates electricity when subjected to a pressure change. Such
piezoelectric materials or transducers can convert a sound signal into an electrical
signal since sound is a pressure wave in fluids. Some transducers can also serve as a
projector (emitter), but not all have this capability, and may be destroyed if used in
such a manner.
The hydrophone has many features such as mating connector molded to 15 m,
30m, or 50 m shielded cable, other lengths available on a custom order basis, cable
includes integral connector rated to a depth of 7000 meters. Furthermore, it has
directionality such as Omni-directional below 10 kHz, very directional at high
frequencies (e.g., ~15º solid angle at 200 kHz, 3dB), and directionality pattern is a
dipole. Hydrophone C304 is shown in Figure 3.12.
Figure 3.12 Hydrophone C304 (www. cetaceanresearch.com, 2009)
46
In spite of the main features have been listed above, hydrophone C304
also provides some addition features that is listed in Table 3.1.
Table 3.1: Hydrophone C304 technical specification
Technical Specification
Linear Frequency Range
Value
0.012 to 250
units
(±3dB) [kHz]
Usable Frequency Range
Transducer Sensitivity
0.005 to 250
-201
(+3/-20dB) [kHz]
dB, re 1V/µPa
Preamplifier Gain
33
dB
Effective Sensitivity
-168
dB, re 1V/µPa
SPL Equiv. Self Noise at
1kHz
Power Requirement
~ 63
dB, re 1µPa
RMS Overload Acoustic
Pressure
Maximum Operating
Depth
Operating Temperature
Range
Output Impedance
Dimensions
5 to 32
172 to 189
Vdc
dB, re 1µPa
460
m
-40 to 60
ºC
2.5
83 x 32 x (25 to 6) mm
(Lx WxT)
3.3.6 Sound Generator
The sound is emitted to test the hearing ability of the turtle and fish by the
sound generator. This system is included with underwater speakers. Software use is
dB Generator as seen in Figure 3.13.
47
Figure 3.13 Generated Sound Software
There are 3 types of sound from a total of 5 votes used in this test. Three
types of sound are LFM (Low and high Frequency Modulation), Sine wave, and
white noise. Two types of sounds that are not used were pink noise and wav files.
Types of pink noise source is not used because it represented a type of white noise,
while the wav files are not used because of the sound depends on the kind provided
that any voice in wav format, so it will be difficult to the later analysis.
While LFM sound bursts were played at 300 Hz for low frequency and 900
Hz for high frequency, sine sound was chosen at 300 Hz. Meanwhile, the white noise
was generated as a default type from the dB generator. The number of channel use
was adjusted in one channel. Finally, for all types of sound the gain was chosen at
0.5 dB.
Sounds were played and collected through hydrophone that direct connected
to the computer. Sounds were played from a computer with dB generator software
through a power amplifier connected to Lubell 3400 underwater speaker. Calibration
of frequencies was accomplished prior the measurement by analyzing the output of
each frequencies compared with software analyzer.
48
3.3.7 Underwater Speaker
The Lubell Labs System 3400 as depicted in Figure 3.14 is a high power
underwater sound system designed to reproduce voice, music, and broadband audio
in the pool, lake, or ocean. The System 3400 can be connected to standard 12 volt
marine electrical system, or connected directly to a 12 volt battery. The
mixer/amplifier features two microphone inputs and volume controls, an auxiliary
input, and includes a noise-cancelling microphone with locking push to talk (PTT)
switch.
Figure 3.14 Speaker type Lubell 3400 (www.lubell.com, 2009)
This speaker has many features such as equipped with protective cage,
internal resistor, and 25' PVC speaker cord. Also, it is assessed for oceanography,
marine biology, cinematography, diver interrogation and recall, scuba instruction,
underwater music and radio rebroadcast. The details features of underwater speaker
Lubel 3400 is listed in Table 3.2.
49
Table 3.2: Speaker type Lubell 3400 technical specification
Features
Typical range
Value
500
unit
m
Mixer/amplifier
60
Watt
Convenient operation
12
Volt
200 -20,000
Hz
Speaker efficiency (at 1kHz)
60%
U/W
SPL @ 1kHz
180
dB/uPa/m
System response
3.3.8 Sound Recorder Software
Sound recorder system is used to record the sound signal detected from the
water. By the hydrophone the sound signal is associated with a computer via the
audio input port. The detected sound signal was recorded in real time by audio
recorder software as depicted in Figure 3.15. The recorded sound was saved in wav
format.
Figure 3.15 Sound recorder using edit audio software
50
3.3.9 Matlab Software
Matlab software 7.0.4 is deployed to extract raw data from recording ABR,
seawater in life habitat profile, sound profile of seawater in the research tank and
types of sound that generated to turtle and fish.
3.4
Measurement Procedures
The research was achieved by conducting several measurements. The
achievement of the research could be reach through some procedures of
measurement. Details measurements are stated at the next subsection. In addition, the
basic test of hearing ability consists of determining the ability of a species to hear
pure tones at intervals throughout its hearing range by Heffner (2007).
3.4.1
Types of Measurement
Types of measurements consist of ABR measurement and sound profile
measurement. Meanwhile, for the support the data analysis the added measurement
and monitored were conducted, there are seawater profile measurement also turtle
and fish observation.
3.4.2
ABR Response Measurement
ABR were measured from the specimen turtles. During the ABR
measurement, the tank was conditioned empty and dried. Each species was equipped
with ABR system and hold in stable position as seen in Figure 3.16.
51
Green turtle was conditioned relax in the resting position. During a sound
presentation, the species is stationed. The transducer used to produce the stimuli is
position 0.5 m from the tank.
After turtle was stable and relaxed the transducer tip to some point
measurements. The species was kept for stable by tightened it in jacket for keep in
the recording ABR.
Turtle were equipped with ear electrode (earphones). Earphones are small
stimulating transducer that delivers the stimulus to turtle’s ear. Four earphones were
used to measured turtle hearing threshold. They were right ear electrode, left ear
electrode, low forehead electrode, and high forehead electrode. The click stimuli
records were divided into 15 to 20 time duration.
Figure 3.16 Green turtle in ABR measurement
Green turtle stationed by jacket is shown in Figure 3.17.
52
Figure 3.17 Green turtle was stationed by jacket for ABR Measurement
In this measurement click stimuli is used. It is a slight and sharp sound. Its
sound energy is evenly spread throughout the frequency spectrum. The range
frequency of ABR analyzer is ranging between 0 to 5 kHz. For presentation,
response averages were high-pass filtered at 30 Hz and low-pass filtered at 3000 Hz
to remove high frequency noise.
The ABR measurement also shows that the right and left ear electrodes were
placed on the preferred sites such as the mastoid, earlobes and ear canals. Electrodes
connected to the Amplifier transmitter box. The electrode leads were not in contact
with earphone transducers, cables, and electrode leads from other equipment. USB
lite or Opti-Amp was placed away from CRT computer monitors, electrical outlets,
and other possible sources of ambient noise. The recorded data is saved in three
modes. They are time domain, ASCII code, and data reported.
3.4.3 Seawater Quality Measurement in Life Habitat
Measurement of seawater quality in the life habitat is significant in order to
become reference on seawater in tank. The measurement was taken with Hydrolab
53
Data Sonde 4a. The primary parameters of interest are temperature (0 C), pH (pH),
salinity,
conductivity
(ms/cm),
and
dissolved
oxygen
(mg/L).
It
is
as
recommendation from the turtle conservation authority. The process of observation
of seawater in life habitat is shown in Figure 3.18.
Figure 3.18 Quality of seawater measurement in life habitat
3.4.4 Seawater Quality Measurement in the Research Tank
Observation of seawater quality in the research tank was conducted after
measurement in the life habitat was complete. The process of seawater quality
measurement in the tank is demonstrated in Figure 3.19.
54
Figure 3.19 Quality of seawater measurements in the tank
3.4.5 Sound Profile Measurement
Sound profile measurement consists of two step measurement. First,
measured the sound profile in the life habitat was conducted. The measurement in the
life habitat was conducted aimed to observe the basic parameter of the sound
characteristics.
The measurement was examined at KK Senangin Vessel II. Sound profile
measurement was tested in difference points of measurements with various
hydrophone positions on distance and depth.
After tested sound profile at the life habitat, we continued for conducting
measured sound profile at the tank. As the prior measurement at the life habitat, the
measurement was divided into difference of distance and depth. Measurement in the
tank consists of 14 point there are point 1 to point 14.
55
Sound profile measurements enabled to determine the characteristics of
ambient noise sea water in the tank. The result of sound profile measurement is
plotted in Figure 3.20. It shows that basic parameter seawater in the tank can be
identified by measuring water in the tank without emitted sound.
Further measurement was conducted by measuring tank with emitted sound.
The measurement conducted of difference hydrophone position from reference point
but in the same depth. The acquired signal was saved in wav format. Moreover, wav
format of the recorded signal could be retrieved in the Matlab. Further sound analysis
was deployed in Matlab.
Figure 3.20 Sound profile in the research tank
3.4.6 Turtle and Fish Response towards Sound
Observation of turtle response towards emitted sound carried out in the
research tank on the last Measurement at TUMEC Kuala Terengganu. There were
two species have been observed, there are turtle ages of 30 years male and female
56
turtles. This step was aimed to identify the turtle response in different sex. The types
of emitted sound consist were LFM, sinusoidal, and WN signal.
3.5
Spectral Analysis of the ABR Waveforms and Sound Profile
The FFT was deployed for wave analysis. It was taken the absolute value of
the FFT to remove the complex part of the transform so the plot function is
conducted. Complex value can’t be plotted in the real number plane. The detail of
whole measurement procedures are shown in appendix C1 to C8.
3.5.1
ABR Waveforms Data Extraction
First step, the measured ABR was analyzed using SmartEP software. Then,
the sound processor software (wavosaur) is deployed for comparison. In particular
case, conversion of ABR signals to its spectral using SmartEP should be compared
with other software analysis. Prior the ABR data analyzed in Matlab, ASCII data
from the ABR was processed first in Excel. Artifact information in numeric data is
removed. ABR value is selected start from 0 ms to 12.8 ms. Data extraction from
acquired ABR is shown in Figure 3.21.
57
ABR Recording
Recording Data format in txt , rpt, and pdf
Txt format converted into
ASCII in Excell
Rpt format processed in
SmartEP
ASCII format converted in
ANSI standard in notepad
Data in ANSI standard
processed in Matlab
Continuous and Spectral
plotted in Matlab
ABR in Time Domain
amplitude(uV)
6
4
2
0
-2
-4
0
2
4
6
8
time (ms)
Power%Spectrum
600
800
1000 1200
frequency (Hz)
0
10
12
14
Normalized Magnitude
10
0
200
400
1400
1600
1800
2000
Figure 3.21 Acquired ABR Data Analysis Flow
58
3.5.2
Sound Calibration Method
Calibration of emitted sound is conducted by using sound analyzer software.
The sound analyzer software was Wavosaur. It provides sampling frequency at
44100 Hz with type of sound is mono. In addition, the software provides feature to
cut and extract the raw signal. The comparison results were chosen as reference for
next analysis in Matlab. Reference software for calibration is depicted in Figure
3.22.
Figure 3.22 Reference sound analysis software
3.5.3 Sound Profile Data Extraction
Acquired sound was recorded in time domain wav format as shown in Figure
3.23. The data is recorded for 30 second. The size of recorded sound is too big in
computing therefore the data has been extracted in Wavosaur and then analyzed in
Matlab.
59
Figure 3.23 Monitored signal of water in the tank profile, depth (d) = 20 cm
Furthermore, the process of data extraction in the time domain is shown in
Figure 3.24.
Figure 3.24 Extracted monitored signal of water in the tank, depth (d) =20 cm
60
3.5.4 Program Implementation and Analysis
Finally, after the extracted data had acquired by applying FFT for the wave
file, then the results of transformation in the frequency domain is plotted and rescale
the axes to achieve specific frequency. The process of acquired sound analysis is
shown in Figure 3.25.
Sound Profile
recording
Sound save into
wave format
Wav file processed in
Matlab
Continuous forms and
Spectral plotted in Matlab
-3
Amplitude(V)
1
Time domain
x 10
0.5
0
-0.5
-1
0
2
4
6
8
time(ms)
10
12
14
5
x 10
Frequency domain
Magnitude(dB re uPa)
130
125
120
0
200
400
600
800 1000 1200
frequency(Hz)
1400
1600
1800
2000
Figure 3.25 Acquired sound profile analysis
61
3.6
Summary
In this chapter, design of study, research materials and life specimen, and
measurement procedure have been presented. Also the procedure of spectral analysis
of the ABR waveforms and sound profile has been reviewed. The scheme of sound
profile data extraction and program implementation had described the obtaining
sound profile of the seawater.
62
CHAPTER 4
RESULTS AND DISCUSSION
4.1
Introduction
Intensive investigation for various methodologies has been discussed in
Chapter 3. Results of the measurements and further discussion have been considered.
This chapter discusses the detailed analysis of turtle response towards sound based
on analysis of the ABR response. First, the hearing sensitivity and bandwidth of
different age of green turtle are presented. Second, the seawater profile is presented.
Third, the sound characterization of the research tank has been discussed and
compared to seawater sound profile. Finally, discussion on turtle and fish behavior
toward sound is presented. Turtle and fish behavior towards sound is useful for TED
design.
4.2
Turtle Auditory Brainstem Response
Four difference ages of green turtles namely 2 years, 5 years, 9 years, and 30
years have been measured. The turtle’s morphology for the ABR measurement is
summarized in the Table 4.1. Meanwhile, details morphology of the turtle is listed in
Figure 4.1.
63
Table 4.1: Turtle’s morphology in ABR measurement
No
Age
Sex
(year)
Total
Head
Shell Weight
Length Length Wide
(cm)
(cm)
(cm)
(Kg)
1.
2
NA
43
12
28
20
2.
5
NA
80
18
56
30
3.
9
NA
96
17
76
66
4.
30
Female
120
29
82
90
Figure 4.1 Green turtle morphology
The turtle morphology as seen in Figure 4.1 shows that the turtle weight is
increased as the ages increase. It is in line with common animal morphology that the
animal weight is increased as the increase of ages.
64
4.2.1
Turtle Age 2 Years
The type of transmitted stimulus for ABR measurement is click stimulus.
Observations focused on the right inner ear turtle due to it has represent the ABR of
the object. In addition, Blackman filter has been deployed in the recording in order to
reject the artifact. Details on ABR system setup is shown in Table 4.2.
Table 4.2: ABR setup for green turtle ages 2 years (sd = 100 μs -2000 μs)
Species
green turtle 2 years
Type of stimulus
click
Rate
70.00/s
Intensity
58 dB nHL
Number of Sweep
1000
Gain
5.0 K
Line filter Status
On
Ear
Right
Rejection Time
1.0 – 10.0 ms
Rej. Artefact
31 ms
Type of Filter
Blackman
ABR waveforms on turtle ages of 2 years have been successfully recorded. It
was indicated by the ABR response in the time domain. Furthermore, recording
signals were categorized into continuous signal (time domain) and converted signal
in its spectral (frequency domain).
Figure 4.2 illustrates the respond of turtle perceived to stimulus at 100 μs.
Recording signal obtain in ABR that captured in 1.6 ms to 6 ms. Next, ABR
recording is extracted using Matlab.
65
Figure 4.2 (a) ABR signal of Green Turtle ages of 2 years
Figure 4.2 (b) ABR signal spectral of Green Turtle ages of 2 years
Figure 4.2 shows that acquired ABR consist of continuous waveform and
converted into its spectral. Typical ABR recordings contain certain peaks and valleys
in the particular range. Furthermore, ABR response is commonly found between 1 to
15 ms from the time of stimulation. The spectral demonstrates the maximum
magnitude occurred at 355.5 Hz. In addition, the time setting is varied every 100 μs
from 100 μs to 2000 μs. The turtle ages of 2 years response to stimulus is listed in
Table 4.3.
66
Table 4.3: ABR response for turtle ages of 2 years (sd = 100 μs to 2000 μs)
Sample
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Stimulus
Duration
(μs)
100
200
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
sum
mean
Peak power
Spectral
in freq
Amplitude
Spectral
(Hz)
(%)
355.5
0.24
304.7
0.02
406.3
0.01
355.5
0.01
355.5
0.22
253.9
0.06
355.5
0.03
406.3
0.04
558.6
0.06
304.7
0.04
304.7
0.03
304.7
0.05
1219.0
0.03
304.7
0.02
863.3
0.05
507.8
0.08
457.0
0.01
304.7
0.01
1117.0
0.07
863.3
0.01
9902.70
0.95
495.13
0.05
Frequency
Range
(Hz)
75-1100
50-1000
75-1000
75-1100
75-1400
75-900
75-1000
100-950
100-1000
100-1100
75-1100
100-610
100-1500
100-1000
100-1000
100-1100
100-900
50-600
75-900
75-700
Most of occurrence (15 stimulus durations) of peak power about 300 to 500
Hz. It just two occurs occurred the peak power of turtle response achieved above of
1000 Hz. Turtle ages 2 years hearing ability is ranging from 50 to 1500 Hz that
obtained from the lowest to highest frequency in whole data and the average
sensitivity about 495.135 Hz. The sensitivity of turtle hearing is summarized at
Figure 4.3.
Figure 4.3 illustrates that hearing sensitivity of green turtle ages 2 years
mostly occurred about 300 to 500 Hz at 15 points although in particular stimulus
67
(sd=1500 μs and 1900 μs) the sensitivity achieved at frequency above 1000 Hz.
Finally, from the data analysis it has been obtained that the mean of turtle hearing
sensitivity is at 495.135 Hz with standard deviation of 286.3.
Hearing Sensitivity of Turtle Age 2 Years
1.6
data 1
1.4
frequency(kHz)
1.2
1
0.8
0.6
0.4
0.2
0
0.2
0.4
0.6
0.8
1
1.2
time stimulus(ms)
1.4
1.6
1.8
2
Figure 4.3 Turtle ages 2 years hearing sensitivity
4.2.2
Turtle Age 5 Years
The following measurement is ABR recording for turtle ages 5 years. The
type of click stimulus has been deployed in this recording. The measurement
procedure is similar with turtle ages 2 years that the observations focused on the right
inner ear turtle. Then, details of ABR system setup and resulted data are shown in
appendix D2.
ABR waveforms for turtle ages 5 years successfully recorded. Continuous
ABR is figured 0 to 12.8 ms. Next, the signal in the time domain converted into
frequency domain. The original signal and its spectral are plotted in Figure 4.4. It is
indicated that the turtle response has average sensitivity at frequency about 304.7 Hz
with frequency ranging from 5 Hz to 1900 Hz.
68
Figure 4.4 (a) ABR signal spectral for green turtle of 5 years
Figure 4.4 (b) ABR signal spectral for green turtle of 5 years
Later in the same manner by increased 100 μs duration time from 100 μs to
2000 μs the frequency spectral would be acquired as shown in appendix D2. The
result shows that turtle hearing is ranging from 10 Hz to 1900 Hz. Furthermore, the
hearing sensitivity of turtle ages 5 years is summarized at Figure 4.5.
.
Figure 4.5 illustrates that hearing sensitivity of green turtle ages 5 years
mostly occurred about 300 Hz to 400 Hz at 10 points even in particular stimulus the
sensitivity is achieved at frequency about 1000 Hz in 2000 μs stimulus and less than
200 Hz at 5 points. Finally, from the analysis has been conducted the mean of turtle
hearing sensitivity is at 354.9 Hz with standard deviation at 268.7.
69
Hearing Sensitivity of Turtle Age 5 Years
1.4
1.2
frequency(kHz)
1
0.8
0.6
0.4
0.2
0
0
0.2
0.4
0.6
0.8
1
1.2
time stimulus(ms)
1.4
1.6
1.8
2
Figure 4.5 Turtle ages 5 years hearing sensitivity
4.2.3
Turtle Age 9 Years
The next step for ABR recording is measurement for older turtle. Type of
click stimulus has been emitted in the turtle ages of 9 years. Observations focused on
the right inner ear turtle. Blackman filter deployed to get a better recorded signal. In
addition, range of used stimulus is adjusted at stimulus duration from 100 μs to 2000
μs.
ABR waveforms on turtle ages of 9 years have been successfully recorded.
Continuous ABR has been recorded for 0 to 12.8 ms. Original ABR waveforms and
its spectral for stimulus 100 μs is plotted in Figure 4.6. It is indicates turtle hearing
sensitivity is obtained at 520.480 Hz. Meanwhile, the frequency ranging is from 5 Hz
to 1600 Hz.
70
Figure 4.6 (a) ABR signal turtle ages of 9 years
Figure 4.6 (b) ABR spectral turtle ages of 9 years
Later in the same manner by varied every 100 μs durations time from 100 μs
to 2000 μs the frequency spectral of turtle ages 9 years turtle obtained as shown in
appendix D2. The result show that turtle hearing ability is ranging of 5 Hz to 1.600
Hz with average of sensitivity is in 520.48 Hz. Furthermore, sensitivity of turtle ages
9 years hearing is summarized at Figure 4.7.
Figure 4.7 demonstrates that hearing sensitivity of green turtle ages 9 years
mostly occurred ranging in 300 Hz to 500 Hz in 12 point stimulus although of in 2
point of stimulus the sensitivity is achieved at frequency above 1000 and one point of
stimulus less than 200 Hz. In addition, from the analysis has been conducted the
mean of turtle hearing sensitivity at 500.2 Hz with standard deviation at 262.67.
71
Hearing Sensitivity of Turtle Age 9 Years
1.4
1.2
frequency(kHz)
1
0.8
0.6
0.4
0.2
0
0
0.2
0.4
0.6
0.8
1
1.2
time stimulus(ms)
1.4
1.6
1.8
2
Figure 4.7 Turtle Age 9 Years Hearing Sensitivity
4.2.4
Turtle Age 30 Years
ABR measurement was conducted for female turtle ages 30 years. Similar
with younger turtles, this step is recorded in time and frequency domain too.
Measurements were taken with stimulus duration at 100 μs. Then, by increased every
100 μs per step, measurements were conducted until stimulus duration at 1500 μs.
(due to stimulus over 1500 μs the recording was failure).
ABR waveform in continuous time and its spectral form is shown in Figure
4.8. The figure shows that the average of turtles hearing sensitivity is determined at
534.887 Hz.
72
Figure 4.8 (a) ABR signal turtle age 30 years (sd at 100 μs)
Figure 4.8 (b) ABR spectral turtle ages of 30 years (sd at 100 μs)
Hearing Sensitivity of Turtle Age 30 Years
1
0.9
frequency(kHz)
0.8
0.7
0.6
0.5
0.4
0
0.5
1
time stimulus(ms)
Figure 4.9 Turtle Age 30 Years Hearing Sensitivity
1.5
73
Later in the same manner by increased 100 μs duration time from 100 μs to
2000 μs would be obtained the frequency spectral of turtle as shown in appendix D2.
Furthermore, sensitivity of turtle ages 30 years hearing is summarized at Figure 4.9.
Appendix D2 demonstrates that hearing sensitivity of green turtle ages 30
mostly occurred ranging in 300 Hz to 500 Hz in 9 point stimulus of 15 data even of
in 3 of 15 of stimulus the sensitivity is achieved at frequency above 500 Hz. The
analysis has been resulted that the mean of hearing sensitivity is obtained at 534.887
Hz with standard deviation at 163.8
In addition, the hearing bandwidth for whole turtles is relatively narrow, 75 to
1200 Hz. It is obtained from intersection frequency range from turtle ages 2 years, 5
years, 9 years, and 30 years. Then the turtle sensitivity ranging from 300 Hz to 500
Hz. ABR responses for the 2 years, 5 years, 9 years, and 30 years turtle have been
recorded and analyzed. ABR signal in the time domain has been transformed into the
spectral form. Furthermore, each species has been summarized it hearing ability.
For all ages of turtles can be summarized that turtle has the lowest hearing
range of 5 Hz while the highest hearing is 1900 Hz. Detailed hearing ability each
species is discussed as follows. Turtle age of 2 years response to sound with
sensitivity at average of 495.135 Hz with ranging of 50 to 1500 Hz. Turtle ages of 5
years exhibits the response to stimulus on sensitivity at average 355.464 Hz with
hearing ability is ranging from 5 to 1900 Hz.
Moreover, the significant result for hearing sensitivity of turtle ages of 9
years is average at 520.480 Hz. The entire range of frequencies is summarized from
5 to 1600 Hz. Meanwhile, hearing ability for turtle ages of 30 years is average
534.887Hz with hearing range of 10 to 1200 Hz.
Furthermore, from the results of four turtle can take the value of minimum
frequency that occurs for each turtle. Thus, from data series of minimum frequency
could be plotted as depicted at Figure 4.10. The figure shows that minimum
frequency of turtles is plotted exponentially. Due to the exponentially results of the
74
graph function, the curve fitting have been informed that the data is confident in
acceptance at R-square: 0.9606.
Minimum Frequency of Green Turtle
frequency vs. age
Exponential
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
Age (year)
1
10
Figure 4.10 Minimum frequencies of turtles hearing
Maximum Frequency of Green Turtle
frequency vs. age
Exponential
1.8
1.6
1.4
Frequency (kHz)
Minimum Frequency (kHz)
0.09
1.2
1
0.8
0.6
0.4
0.2
0
1
10
Age (year)
Figure 4.11 Maximum frequencies of turtles hearing
75
Similar to the minimum frequency function, the data series of maximum
frequency for each turtle has been taken. The result is demonstrated in Figure 4.11.
The figure shows that maximum frequency of turtles plotted exponentially.
Due to the exponentially results of the graph function, from the curve fitting have
been informed that the data is confident in acceptance at R-square: 0.9344.
Next calculation, regarding the minimum and maximum frequency of data
series have been expanded to obtain the hearing bandwidth for each turtle. Thus,
bandwidth for each turtles is summarized at Figure 4.12.
Turtle Hearing Bandwidth
bandwidth vs. age
exp
1.8
Bandwidth (kHz)
1.6
1.4
1.2
1
0.8
0.6
1
Age (year)
10
Figure 4.12 Green turtle’s hearing bandwidth
Figure 4.12 shows that hearing bandwidth of turtles is plotted exponentially.
Due to the exponentially of graph function, the curve fitting has been obtained that
data is confident in acceptance with R-square: 0.9388.
76
Finally, hearing range for each turtles could be displayed. The hearing ranges
of green turtle could be seen at Figure 4.13. The result show there is some
intersection frequency between one turtles and others. The summary of hearing range
is plotted in one graph aim to determine the intersection zone. It is clearly that
intersection zone for all types of is occurs at frequency ranging from 50 Hz to 1200
Hz.
Green Turtles Hearing Ranges
40
35
Turtles Age (year)
30
25
Recommended frequency to be emitted
20
15
10
5
0
0 50 200
400
600
800 1000 1200
Frequency (Hz)
1400
1600
Figure 4.13 Green turtle’s hearing range
1800
2000
77
Hearing Ranges Constellation
Cod
Atlantik Salmon
Pinnipeds
Thooted Whale
Bhaleen Whale
Red ear turtle
Green Turtle
0
1
10
10
2
10
3
10
4
10
5
10
Frequency (Hz)
Figure 4.14 Hearing ranges constellation
The results of turtles hearing bandwidth have been plotted in one graph with
other sea animal. The hearing ranges constellation green turtle compare to other sea
animal could be seen in Figure 4.14. It is shown that hearing ranges of Green Turtle
overlap in some frequency with cod fish, Atlantic salmon, Pinipeds, Thooted whale,
and red ear turtle. However, this phenomenon is not occurs for all of the range of
hearing.
4.3
Seawater Profile
Measurement of water quality in the tank has been conducted. This
measurement is needed to determine the quality of seawater in life habitat and in the
research tank. Measurements were taken employing portable and fixed instruments.
4.3.1 Life Habitat Water Quality
Life habitat water quality in the sea has been measured by portable
instrument. The primary parameters has been measured are temperature (0 C), pH
78
(pH), salinity, conductivity (ms/cm), and dissolved oxygen (mg/L). Water quality in
the life habitat is listed in Table 4.4.
Table 4.4: Water quality recorded by using portable meter in life habitat
Date
Time
Temp
(mm/dd/yyyy)
hh:mm:ss
(°C)
8/26/2009
11:48:00
29.84
pH
SpCond
Sal
DO%
DO
(ppt)
(Sat)
(mg/l)
49.63 32.33 96.90
6.15
(Units) (ms/cm)
8.20
The results of seawater profiles measurement at life habitat is used for the
reference with the seawater profile in the tank. The analysis is important to ensure
that sea water in the research tank relatively similar with the life habitat water.
4.3.2
Portable Instrument
Chemical profile measurement at the tank used two instruments, there are
Portable YSI 556 MPS and Data Sonde 4a. The result of chemical profile that
measured by first instrument is shown in Table 4.5 Compare to the seawater in life
habitat, the water in the tank is not significant different. The temperature, pH,
spCond, Salinity , DO percentage, and DO in the life habitat is 29.84 °C, 8.20, 49.63
ms/cm, 32.33 ppt, 96.90 Sat and 6.15 mg/l respectively when in the tank were 30 °C,
7.95, 48.56 ms/cm, 31.54 ppt,87.90 Sat, and 5.58 mg/l respectively.
Table 4.5: Water quality recorded by using portable meter in the tank
Date
Time
Temp
(mm/dd/yyyy)
hh:mm:ss
(°C)
8/26/2009
15:18:00
30
pH
SpCond
Sal
DO%
DO
(ppt)
(Sat)
(mg/l)
48.56 31.54 87.90
5.58
(Units) (ms/cm)
7.95
79
4.3.3 Fixed Instrument
Data Sonde 4a that provided by Hydrolab was used to measured water quality
multiprobe in the fixed condition. It was stated at the holder. The instrument is
helpful to record the water chemical data for long time. Meanwhile, the last
measurement, chemical quality at the tank was measured per-15 minutes recording.
Measurement with Data Sonde was conducted for 3 conditions of water such
as new changing of water (half day), 2 days, and 3 days put into the tank
respectively.
a.
New Water (half day)
Water quality measurement for half day water is shown in Table 4.6. The
results show that the temperature, pH, spCond, Salinity , DO percentage, and DO in
are almost similar for all measurement. In addition, the measurement was conducted
for 45 minutes started from 14.45 pm until 15.30 pm.
Table 4.6: Water quality recorded using automatic meter (half day)
No
Date
Time
(mm/dd/yyyy) hh:mm:ss
Temp
pH
SpCond
Sal
DO%
DO
(°C)
(Units)
(ms/cm)
(ppt)
(Sat)
(mg/l)
1.
8/27/2009
14:45:00
30.23
7.76
49.1
32.16
107.1
6.65
2.
8/27/2009
14:50:00
30.24
7.76
49.1
32.19
106.1
6.58
3.
8/27/2009
14:55:00
30.24
7.77
49.2
32.21
105.9
6.57
4.
8/27/2009
15:00:00
30.22
7.77
49.2
32.23
104.1
6.46
5.
8/27/2009
15:05:00
30.24
7.77
49.2
32.23
103.4
6.41
6.
8/27/2009
15:10:00
30.23
7.77
49.2
32.25
103.3
6.41
7.
8/27/2009
15:15:00
30.21
7.76
49.2
32.25
102.4
6.36
8.
8/27/2009
15:20:00
30.21
7.76
49.2
32.25
102.9
6.39
9.
8/27/2009
15:25:00
30.18
7.76
49.2
32.25
101.7
6.32
10.
8/27/2009
15:30:00
30.17
7.76
49.2
32.24
104.0
6.46
80
b. Water (2 days)
After conducted measurement of water quality for half days, then have been
measured sea water in 2 days condition in the tank water quality measurement (2
days water). The result of measurement is listed in Table 4.7. The results show that
water quality for 5 hours. The measurement was conducted longer in order to obtain
more data in the evening time. Observation was relatively consistent for some
parameters. The exception was in temperature and DO percentage.
Table 4.7: Water quality recorded using automatic meter (2 days)
No
Date
Time
(mm/dd/yyyy) hh:mm:ss
Temp
(°C)
pH
SpCond
(Units) (ms/cm)
Sal
DO%
DO
(ppt)
(Sat)
(mg/l)
1.
8/26/2009
10:30:00
26.80
7.7
50.2
32.94
85.1
5.58
2.
8/26/2009
11:00:00
26.80
7.7
50.2
32.98
79.0
5.18
3.
8/26/2009
11:30:00
26.81
7.7
50.3
33.04
78.5
5.14
4.
8/26/2009
12:00:00
26.83
7.7
50.3
33.02
79.3
5.19
5.
8/26/2009
12:30:00
26.86
7.7
50.3
33.03
77.6
5.08
6.
8/26/2009
13:00:00
26.88
7.7
50.3
33.03
77.8
5.09
7.
8/26/2009
13:30:00
26.92
7.7
50.3
33.02
79.7
5.21
8.
8/26/2009
14:00:00
26.96
7.7
50.3
33.03
79.2
5.17
9.
8/26/2009
14:30:00
27.00
7.7
50.3
33.02
81.5
5.32
10
8/26/2009
15:00:00
27.05
7.7
50.3
33.02
80.3
5.24
11
8/26/2009
15:30:00
27.10
7.7
50.2
32.96
81.1
5.29
c. Water ( 3 days)
Water quality measurement (3 days water) is shown in Table 4.8. The results
show that water quality is relatively consistent most of parameters. The exceptions
were in temperature and DO percentage. The measurement was conducted for 1.30
hours due to the data was not significantly changed.
81
Table 4.8: Water quality recorded using automatic meter (3 day)
Date
Time
(mm/dd/yyyy) hh:mm:ss
Temp
(°C)
pH
SpCond
(Units) (ms/cm)
Sal
DO%
DO
(ppt)
(Sat)
(mg/l)
8/27/2009
10:00:00
26.32
7.59
50.3
33.06
70.2
4.64
8/27/2009
10:30:00
26.34
7.62
50.4
33.09
73.1
4.83
8/27/2009
11:00:00
26.38
7.63
50.4
33.09
74.1
4.89
4.4
Sound Profile
Sound profile measurements enabled to determine the characteristics of water
both in the seawater and research tank. Thus, this subsection is concern on ambient
noise of seawater. Sound profile measurement was performed by deployed
hydrophone.
4.4.1
Seawater Sound Profile
Due to green turtles habitat is significant parameter to be observed on sound
propagation measurement. Thus, this step is conducted by measuring the ambient
noise of seawater.
Figure 4.15 (a) Ambient noise in the life habitat (distance=200m, depth=2m)
82
Figure 4.15 (b) Ambient noise spectral in the life habitat (distance=200m, depth=2m)
Figure 4.15 (a) and (b) show that ambient noise of seawater with distance of
200 from sea shore and the depth of hydrophone is 2 m. As the results, the summary
of seawater ambient noise measurement in the life habitat is listed in Table 4.9 and
for the details plot for every point of measurement is shown in Appendix E1 to E8.
Table 4.9: Seawater Sound Profile in the life habitat
No
1
2
3
4
5
6
7
8
9
File
Frequency
(Hz)
sp2002mt
sp2005mt
sp20010mt
sp4002mt
sp4005mt
sp40010mt
sp8002mt
sp8005mt
sp80010mt
Average
432
859
601
657
1060
754
573
483
691
729.5
RMS
Magnitude
(dBre 1μPa)
88.97
90.01
88.10
84.47
85.36
86.36
71.81
89.77
87.89
85.758
Max
Min
Magnitude Magnitude
(dBre1μPa) (dBre1μPa)
127.2
127.6
129.1
120.1
125.0
129.5
125.1
128.6
129.5
126.57
60.95
63.68
57.57
54.28
54.33
58.64
40.03
63.92
58.91
56.587
The measurement of ambient noise in the life habitat has been conducted. The
measurements were divided into 3 difference distances from the sea shore of 200 m,
400 m, and 800 m. The various distance and depth was conducted to obtain the
ambient noise in the life habitat. The maximum magnitude of ambient floor is
increased as the increased of depth. The dominant frequency and it strength increases
with increases in depth. Exception occurs in depth of 5 m due to the underwater
current that affecting the transducer orientation. It is due to the fact that water
pressure increase with depth.
83
4.4.2
Research Tank Sound Profile
After got results from seawater sound profile, it is important to compare with
water condition in research tank. Sea water in the research tank important to be
measured to obtained the data comparative with natural habitat. Moreover, recorded
signal waveform is in time domain could be seen in Figure 4.16.
Figure 4.16 shows that ambient noise of seawater in the tank has maximum
magnitude at 957 Hz. The frequency and sound pressure level are is still in the range
of turtle hearing ability. As the results described, SPL for distance 1 m with depth 30
cm is obtained maximum at 132.76 dBre1μPa, minimum at 63.92 dBre1μPa and
RMS at 92.384 dBre1μPa respectively.
Figure 4.16 (a) Ambient noise in the research tank
Figure 4.16 (b) Ambient noise spectral in the research tank
Similar with measurement of sound profile in natural habitat, this step is
conducted on various point of measurement. The measurement is conducted at
various distances of 1 m to 14 m respectively with depth of 30 cm. Detail result of
seawater in the tank profile is listed in Table 4.10.
84
Table 4.10: Seawater sound profile in the research tank
No
File
Frequency
(Hz)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
sp1mt
sp2mt
sp3mt
sp4mt
sp5mt
sp6mt
sp7mt
sp8mt
sp9mt
sp10mt
sp11mt
sp12mt
sp13mt
sp14mt
Average
957
1037
958
739
808
779
760
991
640
869
960
619
659
741
822.6
RMS
Max
Min
Magnitude Magnitude Magnitude
(dBre1μPa) (dBre1μPa) (dBre1μPa)
90.84
131.5
63.58
89.17
131.4
58.16
89.72
132.7
61.22
92.25
131.8
63.69
92.03
132.0
61.10
90.81
133.1
64.50
93.04
133.4
63.89
90.66
132.1
57.17
93.92
133.1
70.29
90.45
133.6
62.36
90.91
132.5
64.54
92.98
133.9
65.15
95.11
134.0
67.96
101.4
133.6
71.37
92.384
132.76
63.92
Sound profile in the tank has been shown in Table 4.10. Acoustic
characteristics of sea water in the tank have been measured through signal analysis,
particularly in its spectral and SPL. Thus, as shown at the Table 4.10 SPL RMS is
ranging from 90 dBre1μPa to 101 dBre1μPa.
Furthermore, the comparison between SPL RMS of seawater in habitat and
seawater in research tank is particular observation. As shown at the Table 4.9 and
Table 4.10 above, Frequency average is of 729.5 Hz and 822.6 Hz, meanwhile SPL
RMS is 85.758 dBre1μPa and 92.384 dBre1μPa for seawater habitat and seawater
tank respectively.
Finally, the result exhibits that frequency average and SPL in the tank is
higher than in seawater habitat about 93.1 Hz and 6.626 dBre1μPa. This case could
be explained regarded of the difference of SPL is occurred caused by material of tank
reflection. Other reason is the dimension of the tank is too little compare to the
measurement area at the seawater habitat.
85
4.5. Sound Characterization
The subsection augments the detailed analysis of sound effects of behavioral
response of turtles to underwater sound from generator. Using the hearing ability
data from ABR analysis, thus, sound source in suitable with turtle hearing is emitted.
In order to enrich the variety of the data, thus various sounds were adjusted. In this
study, the sound types been emitted were LFM (Low and High Frequency
Modulation), sinusoidal, and White Noise (WN) due to provide by manufacturer. For
this step of measurement, the speaker is located in the middle of the research tank
(point 7) meanwhile the hydrophone is placed at radius 1 m to 7 m of right and left
side speaker respectively. The schematic diagram of measurement is shown in
Appendix F1.
4.5.1
LFM Sound Profile
LFM sound recording for radius 1 m from speaker is depicted at Figure 4.17.
Figure 4.17 demonstrates that recording LFM sound and it’s spectral. Spectral signal
has maximum magnitude at 166 Hz with SPL 140.9 dBre1μPa. There are harmonic
about of 500 and above 800 Hz.
Figure 4.17 (a) LFM sound profiles
Figure 4.17 (b) LFM spectral sound profiles
86
In similar method, the measurement is conducted until radius 7 m right side
and left side from speaker respectively. LFM sound analysis is listed in Table 4.11.
Table 4.11: Frequency and magnitude of LFM sound profile
No
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
File
Lfm1mtka
Lfm2mtka
Lfm3mtka
Lfm4mtka
Lfm5mtka
Lfm6mtka
Lfm7mtka
lfm1mtki
lfm2mtki
lfm3mtki
lfm4mtki
lfm5mtki
lfm6mtki
lfm7mtki
Average
Frequency
(Hz)
236
215
383
298
254
237
324
166
265
226
340
343
250
217
268.14
RMS
Max
Min
Magnitude Magnitude Magnitude
(dBre1μPa) (dBre1μPa) (dBre1μPa)
102.80
139.9
74.88
103.00
139.4
75.40
96.36
137.7
62.61
98.48
139.0
70.26
99.49
140.4
71.70
99.76
141.0
72.98
97.85
137.8
72.67
99.66
140.9
74.77
97.17
138.9
73.55
99.05
140.2
70.36
96.34
138.2
68.31
96.47
138.7
70.57
100.2
138.8
71.97
99.13
139.9
73.58
98.98
139.34
71.69
Recording and sound analysis for LFM sound has been conducted. The
results show that for all point of measurement the average of peak power and RMS
SPL occurred at 268.143 Hz and 98.983 dBre1μPa.
4.5.2
Sinusoidal Sound Profile
Sinusoidal sound recording for radius 1 m from speaker is depicted at Figure
4.18. Figure 4.18 exhibits recording sinusoidal sound and it’s spectral. Meanwhile
87
the spectral signal inform that peak of power occurs at 356 Hz with SPL 137.4
dBre1μPa. There are harmonic about of above 100 Hz, 1600 Hz and 1900 Hz.
Figure 4.18 (a) Sinusoidal sound profile
Figure 4.18 (b) Spectral of sinusoidal sound profile
Table 4.12: Frequency and magnitude of sinusoidal sound profile
No
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
File
sin0mt
sin1mtka
sin2mtka
sin3mtka
sin4mtka
sin5mtka
sin6mtka
sin7mtka
sin1mtki
sin2mtki
sin3mtki
sin4mtki
sin5mtki
sin6mtki
sin7mtki
Average
Frequency
(Hz)
356
284
343
374
380
273
299
279
284
343
300
309
345
291
258
314.53
RMS
Max
Min
Magnitude Magnitude Magnitude
(dBre1μPa) (dBre1μPa) (dBre1μPa)
104.50
137.40
76.31
96.14
138.10
64.87
94.65
136.60
70.04
93.97
136.30
65.22
99.20
136.50
70.39
97.22
138.50
74.00
97.01
137.50
72.20
98.00
137.50
74.26
96.14
138.10
64.87
94.65
136.60
70.04
95.65
136.20
70.59
101.30
135.90
70.39
94.62
137.00
67.98
95.75
138.70
71.25
101.60
135.22
75.22
97.36
137.07
70.51
88
In similar method, the measurement is conducted until radius 7 m right side
and left side from speaker respectively. Sinusoidal sound analysis is listed in Table
4.12. Recording and sound analysis for sinusoidal sound has been conducted. The
results show that for all point of measurement the average of peak power and RMS
SPL occurred at 314.53 Hz and 97.36 dBre1μPa.
4.5.3 White Noise Sound Profile
The next measurement is conducted to record white noise sound. This type of
sound recording is monitored for radius 1 m from speaker as depicted at Figure 4.19.
Figure 4.19 illustrates recording WN sound and it’s spectral. This signal has
peak of power at 265 Hz with SPL 136.6 dBre1μPa, above 400 Hz, 600 Hz, 800 Hz,
and 1300 Hz.
Figure 4.19 (a) White noise sound profile
Figure 4.19 (b) Spectral of White noise sound profiles
89
Table 4.13: Frequency and Magnitude of white noise sound profile (depth = 30 cm)
No
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
File
wn1mtka
wn2mtka
wn3mtka
wn4mtka
wn5mtka
wn6mtka
wn7mtka
wn1mtki
wn2mtki
wn3mtki
wn4mtki
wn5mtki
wn6mtki
wn7mtki
Average
Frequency
(Hz)
265
391
335
278
330
316
347
183
391
165
324
253
260
346
304.27
Min
RMS
Max
Magnitude Magnitude Magnitude
(dBre1μPa) (dBre1μPa) (dBre1μPa)
114.7
136.6
82.16
111.2
135.6
84.30
112.2
137.5
86.67
113.3
137.1
83.02
112.2
136.8
84.96
113.2
138.7
87.26
112.0
137.6
82.89
117.2
139.5
94.44
111.2
135.6
84.30
114.5
139.8
89.02
111.2
138
84.57
113.8
138.8
88.13
113.4
136.3
87.25
111.7
137.1
87.07
112.99
137.09
85.72
In similar method, the measurement is repeated for radius 2 to 7 m right side
and left side from speaker respectively. WN sound analysis is listed in Table 4.13.
Recording and sound analysis for WN sound has been conducted. The results show
that for all point of measurement the average of peak power and RMS SPL occurred
at 304.27 Hz and 112.99 dBre1μPa.
Finally, sound propagation in the seawater in habitat and sweater in the
research tank has been performed. A group of sound source has been monitored too.
The research indicated that difference of measurement point did not affect for sound
profile especially for RMS magnitude and maximum magnitude value.
90
4.6
Turtle and Fish Response
Turtle and fish behavior towards sound is significant finding from this study.
The results of the observation to the species respond toward sound are basis
knowledge for the future TED design.
4.6.1 Turtle Response
Observation of turtle response towards emitted sound carried on the main
tank on the last measurement at TUMEC Kuala Terengganu. There were two species
have been observed turtle age 30 years male and female. The types of emitted sound
were LFM, sinusoidal, and WN. The response of turtle observations can be seen in
appendix E1.
The result shows that turtle behavioral avoidance for all types of emitted
sound. Turtle more sensitive towards LFM and WN, whereas sinus sound was less.
The female turtle response was slower than male turtle towards emitted sound.
Turtle response measurement later has been conducted in the same tank and
species. Advanced measurement was focused to determine the sensitivity of the turtle
to emitted sound that depend on distance changing. Type of sound was LFM at low
frequencies 300 Hz and 9000 Hz which the high frequency 1000 Hz is chosen. The
modulation is adjusted at 500 Hz and the gain 10 dB. Detailed response of the turtle
towards sound is shown Appendix F2.
Similar with previous method, turtle responses have been conducted. There
were 7 turtles put into the tank. Turtle ages above 7 years in male and female
combination. Low frequency of LFM signal changed into 300 Hz. The research show
that the male and female turtle excluded from the emitted sound. Changing of the
distance between the turtle and the sound source has no effect on the turtle behavior.
91
4.6.2
Fish Response
Monitoring of fish response towards emitted sound carried in the round small
tank (radius: 2m) on the last measurement at TUMEC Kuala Terengganu. Speaker
located at the edge of the tank while hydrophone at the middle. Hydrophone put in
the 40 cm depth.
There were three species have been observed selar, selar papan, and selayang.
LFM, sinusoidal, and WN signals were emitted into the tank. LFM sound was
emitted 2 channel combined frequency 300 and 600 Hz. Sinus signal is chosen in
300, 600, 900, and 1200 Hz. The response of fish observations can be seen in
Appendix F3. LFM set up until volume level max 3.
Fish gave no responses to all types of sound being emitted while the turtle
behavior avoidance in the same emitted sound. Next monitoring of fish response
towards emitted sound carried on the same tank. Speaker located at the edge of the
tank while hydrophone at the middle as seen in Appendix E. Location of hydrophone
was changed to 20 cm depth. Total amount of fish put into the tank were 10 of
different species.
Later the behavior fish towards sound continued in the same manner for
difference tank. The difference was in the number of fishes and tank deployed. The
results show that fish didn’t response towards all types of sound.
4.7
Problem Encountered
Problems encountered in this study are:
1.
Difficulty to positioned the turtle in stable and relax condition.
2.
Calibration tools in SPL measurement in underwater were not available.
3.
Difficulty to provide underwater camera for turtle orientation recording.
4.
The turtle ABR response to generated sound in life habitat depends on season
and availability of Research Vessel.
92
5.
Difficulty for catching stingray fish to be observed in the research tank.
4.8
Key Contributions
Key contributions of this study are:
1.
Enrich method of ABR analysis to assess the turtle hearing ability. The
deployment of FFT analysis on turtle ABR is the new finding on the area of
research.
2.
Proposed sound specification for TED design. The findings of turtle hearing
capability are become reference to the next working TED.
3.
Increase knowledge on marine turtle hearing ability, especially the green
turtle. Due to the limit of reference on turtle hearing ability, thus this research
shall give the added findings in the area of turtle hearing research.
4.
The results offer multiple opportunities for product development, thus
hearing pattern of green turtle algorithm can be used to identify the hearing
pattern of the others turtle and marine mammal.
5.
Through this project, a library of recordings and mammal underwater hearing
classification will be developed that can be used for educational product,
serve as research data sets, and a management tool. The library of underwater
known and unknown underwater sounds classification could be used to
international distribution research purpose.
4.9
Summary
In this chapter the results of measurements, ABR analysis, sound profile,
analysis, and sound characterization have been presented. Also, the response of turtle
and fish are the proof of the findings that the life specimen five the behavior as the
main objective of this research. Finally, based on the findings on the results and
discussion, it has identified the sound specification to be deployed in the
development TED without reducing the typical gill net performance.
93
CHAPTER 5
CONCLUSION AND FUTURE WORKS
5.1
Conclusion
Auditory Brainstem Response of turtles measurements have been conducted
at TUMEC (Turtle and Marine Ecosystem Center) Terengganu, Segari Perak, and
Melaka Malaysia. The measurements conducted at a research tank by employing
click and tone stimulus. The research focused on green turtle ages of 2, 5, 9, and 30
years.
Fast Fourier Transform (FFT) has been applied to ABR responses of green
turtle. As the results, the frequencies spectral of green turtles have been acquired. It
has demonstrated that turtles hearing ability is relatively narrow in bandwidth
ranging of (50-1200) Hz.
Moreover, the water chemical measurement demonstrates that the seawater in
the research tank and the seawater in life habitat obviously no significant difference.
Thus, this fact allowed us to conclude that the water condition in the tank can
represent the seawater in the area of research
Furthermore, sound characterization analysis has been obtained. It exhibits
that perceiving emitted sound at any point of measurement is relatively similar. In
other words, the sound pressure level (SPL) distribution for all types of sound
propagation in the research tank is sufficiently uniform.
In addition, turtles and fish behavior towards sound has been observed. A
group of sounds of different types that exclude green turtles have been determined.
Also, the fish didn’t response toward the same sounds. Finally, we have identified
94
the sound specification to be deployed in the development TED without reducing the
typical gill net performance.
5.2
Future Works
The future works will focus on implementing a TED prototype based on
sound technique. Prior to the implementation of the prototype, the study of in situ
sound propagation is also required. Also, in situ monitoring of turtle and fish
behavior towards sound should be conducted. In addition, the prototype performance
shall be evaluated by attaching it to the fishing gear.
95
REFERENCES
Ahlstrom, C., Hult, P., Rask, P., Karlsson, J.E., Nylander, E. (2006). Feature
Extraction for Systolic Heart Murmur Classification. Annals of Biomedical
Engineering. Vol. 34, No. 11, 1666–1677.
Bailey, T.C., Sapatinas, T., Powell, K.J., Krzzanowski, W.J. (1998). Signal Detection
in Underwater Sound Using Wavelets. Journal of the American Statistical
Association. Vol 93, No. 441, Application and Case Studies.
Bartol, S.M., Ketten, D.R. (2006). Turtle and Tuna Hearing. NOAA Technical
Memorandum NMFS-PIFSC-7.
Brewer, D., Rawlinson, N., Eayrs, S., Burridge, C., (1998). An assessment of
Bycatch Reduction Devices in a tropical Australian prawn trawl fishery,
Fisheries Research, Volume 36 (2-3):195-215.
Brewer, D., Heales, D., Milton, D., Dell, Q., Gry, G., Venables, B., Jones, P., (2006).
The impact of Turtle Excluder Devices (TED) and bycatch Reduction Devices
(BRDs) on Diverse Tropical Marine Communities in Australia's Northern Prawn
Trawl Fishery. Fisheries Research. Elsevier B.V. 81(2-3):176-188.
Boashash, B., O’shea, P. (1990). A Methodology for Detection and Classification of
Some Underwater Acoustic Signals Using Time-Frequency Analysis Technique.
IEEE Transactions on Acoustics, Speech, and Signal Processing. Vol 38, No.11.
Bonine, Smith, Stitt. (2003). Turtles. Ecol 483/583 Hall.
Chan, E.H. (2006). Marine Turtles in Malaysia: On the Verge of Extinction. Journal
of Aquatic Ecosystem Health & Management, 9(2):175-184.
Culver, R.L., Sibul, L.H., Bradley, D.L. (2007). Underwater Acoustic Signal
Processing. Pennsylvania State, Univ University Park, Applied Research Lab.
Davey, R., McCullagh, P., Lightbody, G., McAllister, G. (2007). Auditory Brainstem
Response Classification: A hybrid model using time and frequency features.
Journal on Artificial Intelligence in Medicine, 40: 1—14.
96
Eckert, K.L (1999). Research and Management Techniques for the Conservation of
Sea Turtles. IUCN/SSC Marine Turtle Specialist Group.
Fridman, J., John, E.R., Bergelson, M., Kaiser, J.B., and Baird, H.W. (1982).
Application of Digital Filtering and Automatic Peak Detection to Brainstem
Auditory Evoked Potential. Journal of Electroencephalogy and Clinical
Neurophysiology.
Graps, A. (2003). Introduction to Wavelets. IEEE Computer Society. 2: 2.
Hall, J.W. (1992). Handbook of Auditory Evoked Responses. Boston: Allyn and
Bacon.
Hecox K, Galambos R. (1974). Brain Stem Auditory Evoked Response in
Human Infants and Adults. Arch Oto. 99:30-33.
Heffner, H.E., Heffner, H. S. (2007). Hearing Rangesof Laboratory Animals, Journal
of the American Association for Laboratory Animal Science. 46 (1):21-30.
Higgs, D.M., Souza, M.J., Wilkins, H.R., Presson, J.C., Popper, A.N. (2001). Age
and Size Related Changes in the Inner Ear and Hearing Ability of the Adult
Zebrafish (Danio rerio). JARO. 03:174-184.
Howorth, P,(2003) Underwater Sound Measurements.
Huynh, Q.Q., Cooper, L.N., Intrator, N., Shouval, H. (1998). Classification of
Underwater Mammals Using Feature Extraction Based on Time-Frequency
Analysis and BCM Theory. IEEE Transactions on Signal Processing. 46 (5):
1202-1207.
Iona, C., Quinquis, A., Stephan, Y. (2008). Feature Extraction from Underwater
Signals using Time Frequency Warping Operators. The French Military Center
of Oceanography.
Jacquin, A., Causevic, E., John, R., Kovacevix, J. (2005). Adaptive Complex
Wavelet-Based Filtering of EEG for Extraction of Evoked Potential Responses.
ICASSP IEEE. Vol 5 : 393-396.
Jafari., A.H.. (2005). Automated Identification and Interpretation of Auditory
Brainstem Evoked Potentials. The 3rd European Medical and Biological
Engineering Conference, EMBEC'05 Prague, Czech Republic.
Ketten., D.R., Bartol, S.M. (2005). Functional Measures of Sea Turtle Hearing.
Woods Hole Oceanographic Inst Ma Biology Dept.
97
Kuwada S, Anderson, J.S, Batra, R, Fitzpatrick, D.C, Teissier N, D’Angelo, W.R.
(2002). Sources of the scalp-recorded amplitude modulated following response.
Journal Am Acad Aud, 13:188-204.
Lenhardt. (2002). United States Patent 6388949. Marine Turtle Acoustic Repellent/
Alerting Apparatus and Method.
Lovell, J.M., Findlay, M.M., Moate, R.M., Nedwell, J.R.,Pegg, M.A., (2005). The
inner ear morphology and hearing abilities of the Paddlefish (Polyodon
spathula) and the Lake Sturgeon (Acipenser fulvescens), Comparative
Biochemistry and Physiology.
Lovell, J.M., Findlaya, M.M., Moateb, R.M., Yan, H.Y. (2004). The hearing abilities
of the prawn Palaemon serratus, Comparative Biochemistry and Physiology,
Part A 140: 89– 100, Comparative Biochemistry and Physiology.
McKinney, M.F., Breebaart, J. (2003). Features for Audio and Music Classification,
John Hopkins University.
McGilvray, J.G., Mounsey, R.P., MacCartie, J., (1998). The AusTED II an Improved
Trawl Efficiency Device and Design Theories. Fisheries Research. Elsevier
B.V., 40(1):17-27.
Merwe, J.P., Ibrahim, K., Lee, S.Y., Whittier, J.M., (2009). Habitat Use by Green
Turtles (Chelonia mydas) Nesting in Peninsular Malaysia: Local and Regional
Conservation Implications. Wildlife Research. CSIRO Publishing 36:637-645.
Mooney, T.A., Tachtigall, P.E., Yuen, M.L. (2006) Temporal resolution of the
Risso’s dolphin, Grampus griseus, Auditory system. Journal Computer
Physiology.
Nedwell, C.H.J. (2003). Assessment of sub-sea acoustic noise and vibration from
offshore wind turbines and its impact on marine wildlife; initial measurements of
underwater noise during construction of offshore windfarms, and comparison
with background noise. COWRIE.
NOOA Technical Memorandum NMFS-SEFSC-366. (1995). The turtle excluder
device (TED) : A guide to better performance, U.S. Department of Commerce,
NMFS, Mississippi Laboratories.
Okumura, N., Ichikawa, K., Akamatsu, T., Arai, N., Shinke, T., Hara, T.,
Adulyanukasol, K. (2006). Stability of Call Sequence in Dugong’s vocalization.
IEEE Oceans Conference - Asia Pacific.
98
Parente, C.L., Lontra1, J.D., Araújo, M.A. (2006). Occurrence of sea turtles during
seismic surveys in Northeastern. Brazil.
Petersen, M.R., (2004). Description of Matlab’s Fast Fourier Transform Interface
Pires, F.J., Lobo, V., (2007). A user friendly toolbox for exploratory data analysis of
underwater sound. IEEE. 061215-064.
Pompei. (2006).United States Patent .Directional Acoustic Alerting System.
Popper. (2000). http://www.life.umd.edu/biology/popperlab/research.
Robins, J.B., McGilvray, J.G., (1999). The AusTED II an Improved Trawl Efficiency
Device Commercial Performance.Fisheries Research. Elsevier B.V.40(1): 2941.
Rogers, P., Triyett, D. (2007). The Fluid Mechanics of Fish Hearing. National
Science Foundation and Office of Naval Research.
Sakas, C.J., Goudey. C, Rountree. R. (2005). Sanctuary Sounds-Monitoring
Underwater Sounds in the National Marine Sanctuaries. Oceans 2005.
MTS/IEEE.
SEAFDEC, (2008), Terengganu Malaysia.
Seramani. S, Taylor. E.A., Seekings. P.J., Yeo, K.P. (2006). Wavelet De-noising
with Independent Component Analysis for Segmentation of Dolphin Whistles in
a Noisy Underwater Environment. IEEE, Oceans Conference - Asia Pacific.
Shaker, M.M. (2005). EEG Waves Classifier using Wavelet Transform and Fourier
Transform. International Journal of Biological and Life Science. Vol 1 No 2.
Singleton, R.C., Poulter, T.C. (1967). Spectral Analysis of the Call of the Male Killer
Whale. IEEE Transaction on audio and electroacoustics.
Sneary, M.G., Lewis, E.R. (2007). Tuning properties of turtle auditory nerve fibers :
Evidence for suppression and adaptation. Hearing Research. 228 (207), 22-30,
Elsevier.
Statistical Toolbox User’s Guide”, Ver. 2.2. (2004). The Mathworks, Inc.
Sunardi (2010). Target Strenght Fish Identification for Scad Species of the South
China Sea. Doctor Philosophy, Universiti Teknologi Malaysia, Skudai.
Storey, B.D., Computing Fourier Series and Power Spectrum with Matlab.
Supin, A.Y., Popov, V., (2001). The sensory physiology of aquatic mammals.
Kluwer, Dordrecth.
Turtle and Marine Ecosystem Center. (2009). Segari Perak Malaysia..
99
Troeger, R. J.R., Buckworth, R., C., Dredge, M.C.L, (1995).Development of a trawl
efficiency device (TED) for Australian prawn fisheries. II. Field evaluations of
the AusTED, Fisheries Research, Elsevier B.V., 22(1-2):107-117.
-, 2007, User Manual, Intelligent Hearing Systems
-, Marine Turtle Newsletter No. 90, 2000 – Page 29.
http:// www.lubell.com/3400.html.
http://turtlemalaysia.gov.my/ancaman.html.
http://web10.bernama.com/maritime/news.
http://www. cetaceanresearch.com/hydrophones/c304-hydrophone.
http://www.earthlife.net/fish/hearing.html.
http://www.fao.org/docrep/005/x7788e/X7788E02.htm.
http://www.westcoastaquatic.ca/fisheries_overview.htm#Trawling.
http://www2.dpi.qld.gov.au/fishweb/10559.html.
100
APENDIX A
Lists of Publications
1. Yudhana, A., Sunardi, Din, J., Abdullah, S., Hassan, R.B.R., Design TED (Turtle
Excluder Device) based on Turtle Hearing Threshold, The 4th International
Conference on Information & Communication Technology and Systems (ICTS),
August 5, 2008, Surabaya-Indonesia.
2. Yudhana, A., Sunardi, Din, J., Abdullah, S., Hassan, R.B.R., Sound Stimulus for
Detection
Turtle
Hearing
Threshold,
International
Conference
on
Telecommunication, 19-20 August 2008, ITTelkom, Bandung-Indonesia.
3. Yudhana, A., Sunardi, Din, J., Abdullah, S., Hassan, R.B.R., Fast Fourier
Transform (FFT) Method for Turtle Hearing Signal Analysis, ICSTIE 2008, 12-13
Des 2008, UiTM Penang, Malaysia.
4. Yudhana, A., Sunardi, Din, J., Abdullah, S., Hassan, R.B.R., An Analysis of
Turtle Hearing Capability to Design TED (Turtle Excluder Device), IGCES 2008,
23-24 Des 2008, UTM Johor Bahru, Malaysia.
5. Yudhana, A., Sunardi, Din, J., Abdullah, S., Hassan, R.B.R.,Green Turtle Hearing
Identification Based on Spectral Analysis,
International Journal of Applied
Physics Research (APR), ISSN: 1916 9639 (print), 1916 9647 (online). May 2010
(Vol. 2 No. 1), Canadian Center of Science and Education, Canada. (published)
6. Yudhana, A., Sunardi, Din, J., Abdullah, S., Hassan, R.B.R., Auditory Brainstem
Response Method to Classify Turtle Hearing Capability, International Journal of
Telkomnika, Vol 8 No2, August, 2010
101
APPENDIX B
Source Code
% For Turtle 2 Years
%Stimulus 100us
load ag2100tes.txt;
sinyal= load('ag2100tes.txt');
sinyal1=sinyal';
x=sinyal1(1,1:512);
y=sinyal1(2,1:512);
p=fft(y,2048);
Pyy = p.* conj(p) /2048;
f = 10000*(0:128)/2048;
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);
plot(f,Pyy(1:129));title('Power%Spectrum');xlabel('frequency
(Hz)');ylabel('amplitude');grid on;
%Stimulus 200us
N=512;
load ag2101.txt;
sinyal= load('ag2101.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
%Stimulus 300us
N=512;
load ag2300tes.txt;
sinyal= load('ag2300tes.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
102
%Stimulus 400μs
N=512;
load ag2400tes.txt;
sinyal= load('ag2400tes.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
%Stimulus 500μs
N=512;
load ag2500tes.txt;
sinyal= load('ag2500tes.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
%Stimulus 600μs
N=512;
load ag2600tes.txt;
sinyal= load('ag2600tes.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
103
%Stimulus 700μs
N=512;
load ag2700tes.txt;
sinyal= load('ag2700tes.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
%Stimulus 800μs
N=512;
load ag2800tes.txt;
sinyal= load('ag2800tes.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
%Stimulus 900μs
N=512;
load ag2900tes.txt;
sinyal= load('ag2900tes.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
%Stimulus 1000μs
N=512;
load ag21001.txt;
sinyal= load('ag21001.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
104
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
%Stimulus 1100μs
N=512;
load ag21101.txt;
sinyal= load('ag21101.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
%Stimulus 1200μs
N=512;
load ag21200.txt;
sinyal= load('ag21200.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
%Stimulus 1300μs
N=512;
load ag21300.txt;
sinyal= load('ag21300.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
%Stimulus 1400μs
N=512;
load ag21400.txt;
sinyal= load('ag21400.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
105
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
%Stimulus 1500μs
N=512;
load ag21500.txt;
sinyal= load('ag21500.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
%Stimulus 1600μs
N=512;
load ag21600.txt;
sinyal= load('ag21600.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
%Stimulus 1700μs
N=512;
load ag21700.txt;
sinyal= load('ag21700.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
%Stimulus 1800μs
N=512;
106
load ag21800.txt;
sinyal= load('ag21800.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
%Stimulus 1900μs
N=512;
load ag21900.txt;
sinyal= load('ag21900.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
%Stimulus 2000μs
N=512;
load ag22000.txt;
sinyal= load('ag22000.txt');
sinyal1=sinyal';
x=sinyal1(1,1:N);
y=sinyal1(2,1:N);
p=fft(y,2*N);
spek=abs(p)/N;
spek=spek(1:N/4).^2;
freq=0:6500/128:6500;
freq1=freq(1:128);
subplot(2,1,1); plot(x,y);title('ABR in Time Domain');xlabel('time
(ms)');ylabel('amplitude(uV)');grid on;
subplot(2,1,2);plot(freq1,spek);title('Power%Spectrum');xlabel('freq
uency (Hz)');ylabel('amplitude');grid on;
107
APPENDIX C1
First Measurement Procedure, TUMEC Dungun
20 - 21 Februari 2008
108
APPENDIX C2
Second Measurement Procedure, TUMEC Dungun
3-4 June 2008
109
APPENDIX C3
Third Measurement Procedure, TUMEC Dungun
November 2008
110
APPENDIX C4
Fourth Measurement Procedure, Masjid Tanah Melaka
24 December 2008
111
APPENDIX C5
Fifth Measurement Procedure, Segari Lumut Perak
31 December 2008-2 January 2009
112
APPENDIX C6
Sixth Measurement Procedure, TUMEC Dungun
26 May 2009
113
APPENDIX C7
Seventh Measurement Procedure, TUMEC Dungun
10-12 August 2009
114
APPENDIX C8
Eight Measurement Procedure, TUMEC Dungun
26-28 August, 2009
115
APPENDIX D1
ABR and Sound Profile Recording
ABR Signal for Turtle Age 2 years (time domain, sd = 100 μs, click stimulus)
ABR signal for Turtle 5 Age Years ( sd = 100 μs)
ABR signal Turtle Age 9 years (time domain, sd = 100 μs, click stimulus)
ABR signal in time domain (sd = 100 μs)
116
APPENDIX D2
ABR System Setup and resulted Data
ABR setting for Turtle Age 5 years (stimulus duration = 100 μs – 2000 μs)
Type of stimulus
click
Time of stimulus
25 μs
Rate
70.00/s
Number of Sweep
1000
Time Scale
25.0 μs (12.8 time scale)
Gain
5.0 K
Line filter Status
On
Ear
Right
Rejection Time
1.0 – 10.0 ms
Rej. Artefact
31μV
Type of Filter
Blackman
117
ABR response turtle age 5 years (Click stimulus, sd = 100 – 2000 μs)
Sample
Stimulus
Peak power in
Spectral
Frequency
Duration
freq Spectral
Amplitude
Range
(μs)
(Hz)
(%)
(Hz
1
100
304.70
0.246
10-800
2
200
406.30
0.020
10-900
3
300
406.30
0.145
10-600
4
400
406.30
0.055
10-600
5
500
304.70
0.002
10-1100
6
600
50.78
0.019
10-600
7
700
50.78
0.029
10-900
8
800
50.78
0.110
10-900
9
900
304.70
0.010
10-1200
10
1000
50.78
0.887
5-900
11
1100
101.60
1.527
10-800
12
1200
304.70
0.274
5-1125
13
1300
355.50
0.058
10-1600
14
1400
558.60
0.122
50-1900
15
1500
355.50
0.093
10-1600
16
1600
0.00
0.082
10-1050
17
1700
964.80
0.056
5-1100
18
1800
355.50
0.177
5-1100
19
1900
355.50
0.058
5-1500
20
2000
1066.00
0.080
5-1200
Sum
6753.82
4.049
Mean
355.46
0.202
118
ABR setting for turtle age 9 years (stimulus duration = 900 μs -1500 μs)
Type of stimulus
Click
Time of stimulus
25 μs.
Rate
70.00/s
Intensity
58 dB nHL
Number of Sweep
1000
Time Scale
25.0 μs (12.8 time scale)
Gain
5.0 K
Line filter Status
On
Ear
Right
Rejection Time
1.0 ms – 10.0 ms
Rej. Artefact
31
Type of Filter
Blackman
119
ABR response turtle ages of 9 years (stimulus duration = 100 μs – 2000 μs)
Sample
Stimulus
Peak power in
Spectral
Frequency
Duration
freq Spectral
Amplitude
Range
(μs)
(Hz)
(%)
(Hz
1
100
101.60
20.56
5-1400
2
200
355.500
26.46
5-1400
3
300
406.300
4.589
50-1300
4
400
457.000
2.162
10-1500
5
500
457.000
1.5
25-800
6
600
1066.000
0.678
10-1200
7
700
1117.000
0.9919
15-1300
8
800
355.500
3.945
5-1250
9
900
355.500
8.252
5-930
10
1000
355.500
8.689
10-1500
11
1100
304.700
18.68
10-1600
12
1200
710.900
30.78
20-1600
13
1300
304.700
22.24
25-1600
14
1400
558.600
2.39
75-1500
15
1500
355.500
0.3988
100-1500
16
1600
710.900
5.073
100-1200
17
1700
304.700
5.574
100-1200
18
1800
304.700
18.19
100-1400
19
1900
710.900
3.895
150-1400
20
2000
710.900
3.094
100-1400
Sum
10409.600
188.142
Mean
520.480
9.407085
120
ABR setting for turtle age 30 years (stimulus duration = 900 μs -1500 μs)
Type of stimulus
Click
Time of stimulus
25 μs.
Rate
70.00/s
Intensity
58 dB nHL
Number of Sweep
1000
Time Scale
25.0 μs (12.8 time scale)
Gain
5.0 K
Line filter Status
On
Ear
Right
Rejection Time
1.0 ms – 10.0 ms
Rej. Artefact
31
Type of Filter
Blackman
ABR response turtle age 30 years (stimulus duration = 100 μs - 1500 μs)
Sample Stimulus
Peak power in
Spectral
Range freq.
Duration
freq Spectral
Amplitude
Spectral(Hz)
(μs)
(Hz)
(%)
1
100
507.800
10.77
10-1000
2
200
761.700
0.4668
10-1000
3
300
761.700
0.8309
10-1000
4
400
914.100
0.7846
10-1000
5
500
660.200
0.8902
10-1200
6
600
457.000
1.157
10-900
7
700
507.800
0.7409
20-920
8
800
406.300
8.705
20-1200
9
900
457.000
1.132
20-900
10
1000
457.000
0.4533
20-1200
11
1100
457.000
0.8238
20-910
12
1200
457.000
0.1229
10-940
13
1300
457.000
0.0322
1200
14
1400
457.000
8.026
NA
121
15
1500
304.700
0.5516
10-1200
16
1600
NA
NA
NA
17
1700
NA
NA
NA
18
1800
NA
NA
NA
19
1900
NA
NA
NA
20
2000
NA
NA
NA
Sum
8023.300
35.487
Average
534.887
2.365813
122
APPENDIX D3
ABR spectral in SmartEP
ABR signal spectral for Green Turtle 2 years (frequency domain, sd =
300 μs, click stimulus)
ABR signal spectral for Green Turtle 5 years (frequency domain, sd =
300 μs, click stimulus)
123
ABR signal spectral for Green Turtle 9 years (frequency domain, sd =
100 μs, click stimulus)
ABR signal spectral for Green Turtle 30 years (frequency domain, sd =
100 μs, click stimulus)
124
APPENDIX E1
Sound Profile in Life Habitat
Distance 100 meter from sea shore
(a) sp1002mt (time domain)
(b) sp1002mt (frequency domain)
125
APPENDIX E2
Sound Profile in Life Habitat
Distance 200 meter from sea shore
(a) sp2002mt (time domain)
(b) sp2002mt (frequency domain)
(a) sp2005mt (time domain)
(b) sp2005mt (frequency domain)
126
(a) sp20010mt (time domain)
(b) sp20010mt (frequency domain)
127
APPENDIX E3
Sound Profile in Life Habitat
Distance 400 meter from sea shore
(a) sp4002mt (time domain)
(c) sp4002mt (frequency domain)
(a) sp4005mt (time domain)
128
(b) sp4005mt (frequency domain)
(a) sp40010mt (time domain)
(b) sp40010mt (frequency domain)
129
APPENDIX E4
Sound Profile in Life Habitat
Distance 800 meter from sea shore
(a) sp8002mt (time domain)
(b) sp8002mt (frequency domain)
(a) sp8005mt (time domain)
(b) sp8005mt (frequency domain)
130
(a) sp80010mt (time domain)
(b) sp80010mt (frequency domain)
131
APPENDIX E5
Ambient Noise in the life habitat with distance of 200 m from the sea shore
Noise Floor in the sea(Distance 200m)
1000
data 1
quadratic
900
y = - 24*x 2 + 3.1e+002*x - 95
Frequency(Hz)
800
700
600
500
400
300
1
2
3
4
5
6
depth (m))
7
8
9
10
Ambient noise in the sea (distance 200m)
Noise Floor in the sea(Distance 200m)
1000
data 1
quadratic
900
Frequency(Hz)
800
700
600
500
400
2
3
4
5
6
depth (m))
7
8
9
Ambient noise in the sea (distance 200m)
10
132
APPENDIX E6
Ambient Noise in the life habitat with distance of 400 m from the sea shore
Noise Floor in the sea(Distance 400m)
1100
1050
data 1
quadratic
2
y = - 24*x + 3.1e+002*x + 1.4e+002
1000
Frequency(Hz)
950
900
850
800
750
700
650
2
3
4
5
6
depth (m))
7
8
9
10
Ambient noise in the sea (distance 400m)
Noise Floor in the sea(Distance 400m)
1100
data 1
quadratic
1050
1000
Frequency(Hz)
950
900
850
800
750
700
650
2
3
4
5
6
depth (m))
7
8
9
Ambient noise in the sea (distance 400m)
10
133
APPENDIX E7
Ambient Noise in the life habitat with distance of 800 m from the sea shore
Noise Floor in the sea(Distance 800m)
700
data 1
quadratic
2
y = 8.9*x - 93*x + 7.2e+002
Frequency(Hz)
650
600
550
500
450
2
3
4
5
6
depth (m))
7
8
9
10
Ambient noise in the sea (distance 800m)
Noise Floor in the sea(Distance 800m)
700
data 1
quadratic
Frequency(Hz)
650
600
550
500
450
2
3
4
5
6
depth (m))
7
8
9
Ambient noise in the sea (distance 800m)
10
134
APPENDIX E8
Peak of Magnitude for Life Habitat Ambient Noise
Peak of Magnitude for Noise Floor
130
129
Magnitude (dB re 1uPa)
128
127
126
125
124
123
122
Distance of 200 m
Distance of 400 m
Distance of 800 m
121
120
2
3
4
5
6
7
Depth of Measurement (m)
8
9
10
135
APPENDIX F1
Sound Characterization Measurement
Point
Point
Center of tank,Point 0
Point
Point
Schematic Diagram of Sound Characterization Measurement
136
APPENDIX F2
Turtle Responses towards Sound
Turtle Responses by generated sound in the Tank with L = 300 – 900 Hz (mod =500,
H = 1000 Hz, Gain = 10 dB).
Turtle Responses towards Sound (Main Tank)
No
Position of
Type of
speaker and
Sound
Response
hydrophone
1.
Point 12
2.
Point 5
3.
Point 5
LFM
Female dispelled
Sinusoidal Male dispelled.
WN
Male dispelled.
Female static
4.
Point 11
LFM
Female relatively static.
Turtle Responses towards Sound (Main Tank)
No
L (Hz)
Radius (m)
Response
1.
300
<1
dispelled
2.
400
1
dispelled
3.
500
1
dispelled
4.
600
6
dispelled
5.
700
6
dispelled,
Afraid to go near the speaker
6.
800
>6
dispelled
7.
900
>6
dispelled
137
APPENDIX F3
Fish Response towards Sound
Fish Response towards sound (Main tank)
NO
Sound type
Response
1.
LFM
No response
2.
Sinus
No response
3.
WN
No response
4.
LFM
No Response
up to level 10
5.
Sinus
No response
Fish Responses to Sound (Round tank, r = 2m)
No
Sound type
Response
1.
LFM
No response
2.
WN
No response
3.
Sinus
No response
Fish Responses by generated sound (Round tank, r = 3 m)
No
Sound type
Response
1.
LFM
No response
2.
WN
No response
3.
Sinus
No response