UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Bio-acoustics of the gray whale (Eschrichtius robustus) Dahlheim, Marilyn Elayne 1987

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1988_A1 D34.pdf [ 12.52MB ]
Metadata
JSON: 831-1.0097975.json
JSON-LD: 831-1.0097975-ld.json
RDF/XML (Pretty): 831-1.0097975-rdf.xml
RDF/JSON: 831-1.0097975-rdf.json
Turtle: 831-1.0097975-turtle.txt
N-Triples: 831-1.0097975-rdf-ntriples.txt
Original Record: 831-1.0097975-source.json
Full Text
831-1.0097975-fulltext.txt
Citation
831-1.0097975.ris

Full Text

BIO-ACOUSTICS OF THE GRAY WHALE ( E s c h r i c h t l u s r o b u s t u s ) By MARILYN ELAYNE DAHLHEIM B . S c , San D i e g o S t a t e U n i v e r s i t y , 1976 M.Sc., San Diego S t a t e U n i v e r s i t y , 1980 A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES (Department of Z o o l o g y , U n i v e r s i t y of B r i t i s h C olumbia) We a c c e p t t h i s d i s s e r t a t i o n as c o n f o r m i n g t o t h e r e q u i r e d s t a n d a r d THE UNIVERSITY OF BRITISH COLUMBIA O c t o b e r 1987 ( c ) M a r i l y n Elayne Dahlheim, 1987 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 DE-6(3/81) ABSTRACT Gray whales (Eschrichtius robustus), while engaged in underwater signalling, circumvent noise in the acoustical channel by the structure and timing of their cal ls. Data yielding this conclusion were collected during an acoustical study on gray whales and their habitats (1981-1984). Sonographic analyses of tape recordings were used to quantify the acoustical repertoire, the ambient noise characteristics of the area, and the relationship between the animals' calls and the environment. The acoustical responses of whales to artif icial ly increased levels of noise were documented during playback experimentation in Mexico. Nine sound parameters were inspected and compared between control and experimental conditions: calling rates, call types, frequency range of signals (Hz), emphasized frequencies (Hz), received levels of sounds (dB re 1 yuPa), call duration (sec), percentage of calls exhibiting frequency modulation, number of pulses per series, and repetition rates of signals. The observed surface behavior of gray whales in response to noise ( i .e . , dive durations, movements and abundance) was also investigated. Analyses yielded: a description of gray whale call types; a characterization of the acoustical habitats occupied by this species, including a l ist of sources contributing to the ambient noise and a profile of the propagation characteristics of the study area; a determination of the relationship between whale calls and their habitats; and the acoustical capabilities and strategies of whales in response to noise. The plasticity observed in the overall behavior of this whale is of adaptive significance when considering the dynamic nature of noise in the environment. Typically, the multiple i i strategies employed by the whales when faced with various noise situations enable them to minimize the detrimental effect that noise has on their underwater signalling. Gray whale responses varied with the sound source and may also differ relative to the geographical range and/or general behavior of the animal. It is concluded that ambient noise (both natural and man-made) has a profound effect on the behavior of this coastal species and that acoustical calling is modified to optimize signal transmission and reception. TABLE OF CONTENTS Page LIST OF TABLES v i i LIST OF FIGURES ix ACKNOWLEDGEMENTS x i i i GENERAL INTRODUCTION 1 GENERAL MATERIALS AND METHODS 4 Study Area(s) 5 Research Equipment - Gray Whale Sound Production 5 Research Equipment - Ambient Noise 10 Research Equipment - Playback Experimentation 10 CHAPTER I - SOUND PRODUCTION, AMBIENT LEVELS AND THE RELATIONSHIP OF WHALE SIGNALS TO THEIR ACOUSTICAL HABITATS . . . . 14 INTRODUCTION 15 MATERIALS AND METHODS 23 Analysis Procedures - Gray Whale Sound Production 23 Analysis Procedures - Ambient Noise 24 Analysis Procedures - Relat ionship Between Whale Ca l l s and Ambi ent Noi se 25 RESULTS 27 Types of Sounds Produced 27 Laguna San Ignacio (Mexico) 27 S t . Lawrence Island (Alaska) 45 Washington State Waters 48 Integra ion of Regional Data Bases/Published Information . . 55i v TABLE OF CONTENTS, CONTINUED Page Ambient Noise 60 Laguna San Ignacio (Mexico) 60 St. Lawrence Island (Alaska) 80 Washington State Waters 91 Comparisons of Ambient Noise Throughout Range 96 Relationship Between Gray Whale Signals and Ambient Noise . . . 99 DISCUSSION 109 CHAPTER 2 - RESPONSES OF GRAY WHALES TO INCREASED LEVELS OF TEMPORARY (SHORT-TERM) NOISE 122 INTRODUCTION 123 MATERIALS AND METHODS 126 1983 Experimental Design 126 1983 Analysis of Gray Whale Sound Behavior 130 1983 Analysis of Observed Surface Behavior 132 RESULTS 134 1983 Cal ling Rates 134 1983 Call Types . 135 1983 Call Structure 138 1983 Observed Surface Behavior 141 DISCUSSION 151 CHAPTER 3 - RESPONSES OF GRAY WHALES TO INCREASED LEVELS OF CONTINUOUS (LONG-TERM) NOISE 157 INTRODUCTION 158 v TABLE OF CONTENTS, CONTINUED Page MATERIALS AND METHODS 161 1984 Experimental Design 161 1984 Observed Surface Behavior - Tracking Experiments 162 1984 D is t r ibu t ion and Abundance - Transect Experiments . . . . 163 1984 Analysis of Gray Whale Sound Behavior 169 1984 Analysis of Tracking Experiments 170 1984 Analysis of Transect Experiments 171 RESULTS 172 1984 Cal l i ng Rates 172 1983/1984 Comparisons of Ca l l i ng Rates 172 1984 Cal 1 Types . 175 1984 Ca l l Structure 175 1983/1984 Comparisons of Ca l l Structure 180 1984 Observed Surface Behavior - Tracking Experiments 184 1984 Observed Surface Behavior - Transect Experiments 214 Seasonal Comparisons (Abundance/Distribution/Movements) . . 214 Between-Season Comparisons (Abundance of Whales) 230 DISCUSSION 239 GENERAL SUMMARY/CONCLUSIONS 253 REFERENCES 258 ^ APPENDIX A A- l APPENDIX B B-l APPENDIX C C-l vi LIST OF TABLES Page TABLE I. A Summary of Published Data on Sounds Produced by the Gray Whale (1 967-1978) 19 TABLE I I . A C l a s s i f i c a t i o n of Sounds Produced by the Gray Whale in Laguna San Ignacio, Mexico 28 TABLE I I I . Sounds Produced by the Gray Whale Throughout I ts Range Based on Acoust ical Studies in Laguna San Ignacio and Published Accounts 56 TABLE IV. Comparisons of Gray Whale Sounds Throughout I ts Range. . . 59 TABLE V. Estimated Levels (dB re 1 juPa) and Their Associated Frequencies (Hz) for B io log ica l Ambient Noise Stat ions #1-32 in Laguna San Ignacio, Mexico 65 TABLE VI. B io log ica l Organisms Potent ia l l y Responsible for the High Levels of Ambient Noise in Laguna San Ignacio, Mexico. . . 68 TABLE VI I . Comparisons of Ambient Noise Pro f i l es Throughout the Range of the Gray Whale 97 TABLE VI I I . 1983 Ca l l i ng Rates of Gray Whales in Laguna San Ignacio, Mexico 136 TABLE IX. 1983 Ca l l Types Produced by Gray Whales in Laguna San Ignacio, Mexico 137 TABLE X. 1983 Structura l Changes in the Ca l l s of Gray Whales Occupying Laguna San Ignacio, Mexico 140 TABLE XI . Average Dive Times (sec) of Gray Whales During Various Experimental Conditions during the 1983 Season in Laguna San Ignacio, Mexico 146 TABLE XI I . 1984 Ca l l i ng Rates of Gray Whales in Laguna San Ignacio, Mexico 173 TABLE XI I I . Comparisons of the 1983 and 1984 Ca l l i ng Rates of Gray Whales Occupying Laguna San Ignacio, Mexico. . . . 176 TABLE XIV. 1984 Ca l l Types Produced by Gray Whales in Laguna San Ignacio, Mexico 177 TABLE XV. 1984 Structural Changes in the Cal ls of Gray Whales Occupying Laguna San Ignacio, Mexico 179 vi i LIST OF TABLES, CONTINUED Page TABLE XVI. 1983 and 1984 Comparisons of Structural Changes in the Calls of Gray Whales in Laguna San Ignacio, Mexico 181 TABLE XVII. Average Dive Times (sec) of Gray Whales Exposed to Various Experimental Conditions During the 1984 Season in Laguna San Ignacio, Mexico 189 TABLE XVIII. 1983/1984 Comparisons of Average Dive Times of Gray Whales in Laguna San Ignacio, Mexico 191 TABLE XIX. 1984 Trackline Data of Gray Whales Occurring in Laguna San Ignacio, Mexico 215 TABLE XX. Counts of Whales Occupying Laguna San Ignacio on or about 1 February and 2 March for the Years 1978-1982 (From Jones and Swartz, 1984) 233 vi i i LIST OF FIGURES Page Figure 1. Study Area - - Laguna San Ignacio, Baja California Sur, Mexico 7 Figure 2. Bottom-mounted Transducer in Laguna San Ignacio, Mexico. . 12 Figure 3. The Migration Route of the Gray Whale (Eschrichtius robustus) 17 Figure 4. Spectrograph of a Gray Whale SI Series 30 Figure 5. Spectrograph of Two SI Series Depicting Frequency Modulation 32 Figure 6. Spectrograph of a Gray Whale S2 Signal 34 Figure 7. Spectrograph of a Gray Whale S3 Signal 37 Figure 8. Spectrograph of a Gray Whale S4 Signal 39 Figure 9. Spectrograph of a Gray Whale S5 Signal (bubble blast). . . 41 Figure 10. Spectrograph of a Gray Whale S6 Signal (sub-surface exhalation) 44 Figure 11. Recording Locations of Gray Whales off St. Lawrence Island, Bering Sea, Alaska, Summer 1982 47 Figure 12. Spectrograph of a Sub-surface Exhalation (N6) Produced by a Surfacing Gray Whale off St. Lawrence Island, Alaska 50 Figure 13. Study Area - - Coastal Waters of Washington State 52 Figure 14. Recording Locations of Gray Whales in Washington State Waters 54 Figure 15. Locations of Biological Ambient Noise Stations in Laguna San Ignacio, Mexico 62 Figure 16. Spectra of Typical Biological Ambient Noise Levels in Laguna San Ignacio, Mexico 64 Figure 17. Knudsen Sea State Curves 67 Figure 18. Spectra of Typical Large Vessel (25-35 m in length) Noise Encountered in Laguna San Ignacio, Mexico 75 ix LIST OF FIGURES, CONTINUED Page Figure 19. Spectra of a 20-hp Mercury Outboard Engine Operating at (a - d) a) idle, b) quarter throttle, c) half throttle, and d) three-quarter throttle 77/79 Figure 20. Locations of Ambient Noise Stations off St. Lawrence Island, Alaska 82 Figure 21. Average Levels (dB re 1 uPa) of Ambient Noise Stations 1 and 2, St. Lawrence Island, Alaska 84 Figure 22. Average Level (dB re 1 uPa) of Ambient Noise at Station 3, St. Lawrence Island, Alaska (influenced by ship noise) 86 Figure 23. Average Levels (dB re 1 uPa) of Ambient Noise at Stations 4 and 5, St. Lawrence Island, Alaska (influenced by surf noise) 88 Figure 24. Average Levels (dB e 1 uPa) of Ambient Noise at Stations 6-12 (Southeast Cape), St. Lawrence Island, Alaska . . . . 90 Figure 25. Noise Spectra of a 7-m Boston Whaler Operating off Southeast Cape, St. Lawrence Island, Alaska 93 Figure 26. Spectra of Ambient Noise off La Push, Washington (coastal waters) 95 Figure 27. Composite Spectrograph of a Gray Whale Pulse Series (SI), Outboard Engine Noise and Biological Ambient Noise in Laguna San Ignacio, Mexico 101 Figure 28. Spectrograph of a Gray Whale Pulse Series (SI) and Outboard Engine Noise 104 Figure 29. Relationships of Gray Whale Sound Frequencies and Levels to Ambient Noise 106 Figure 30. Spectrograph of a Bottlenose Dolphin Whistle and Biological Ambient Noise in Laguna San Ignacio, Mexico . . 108 Figure 31. Immediate Study Area off Rocky Point Defined During Tracking Experiments in Laguna San Ignacio, Mexico . . . . 129 Figure 32. 1983 Frequency Distribution of Gray Whale Dive Durations 143 Figure 33. 1983 Frequency Distribution of Cow/calf Pairs and Single Whale Dive Times in Laguna San Ignacio, Mexico 145 x LIST OF FIGURES, CONTINUED Page Figure 34. Example of Line Display Depicting Behavior of Gray Whales and Environmental Effects During Tracking Experiments. . . 149 Figure 35. 1984 Vessel Survey Transect Route of Entire Lagoon System 165 Figure 36. 1984 Modified Transect Route - - Laguna San Ignacio, Mexico 168 Figure 37. 1984 Frequency Distribution of Dive Durations of Gray Whales Occupying Laguna San Ignacio, Mexico 186 Figure 38. 1984 Frequency Distribution of Cow/calf Pairs and Single Whale Dive Durations in Laguna San Ignacio, Mexico . . . . 188 Figure 39. 1983/1984 Frequency Distribution Comparisons of Dive Times on All Whales Occurring in Laguna San Ignacio, Mexico 193 Figure 40. 1983/1984 Frequency Distribution Comparisons of Dive Durations on Cow/calf pairs in Laguna San Ignacio, Mexico 195 Figure 41. 1983/1984 Frequency Distribution Comparisons of Dive Durations on Single Whales in Laguna San Ignacio, Mexico 197 Figure 42. Percentage of Time Spent Tracking Whales Per Experimental Condition 200 Figure 43. Trackline of Whale During Control Periods 202 Figure 44. Trackline of Whale in Presence of Real Sources 204 Figure 45. Trackline of Whale During Outboard Engine Playback Periods 206 Figure 46. Trackline of Whale During Oil-dri l l ing Playback Periods. . 209 Figure 47. Trackline of Whale During Killer Whale Playback Periods (no progressive movement through area) 211 Figure 48. Trackline of Whale During Killer Whale Playback Periods (depicting increased distance offshore) 213 Figure 49. A Comparison of the Total Number of Whales Occupying the Waters Between Parmenter Point and Yucca Plant During Control and Experimental Conditions 218 xi LIST OF FIGURES, CONTINUED Page Figure 50. A Comparison of the Number of Southbound Whales Occurring Between Rocky Point and Parmenter Point During Control and Experimental Periods 221 Figure 51. A Comparison of the Number of Northbound Whales Occurring Between Rocky Point and Parmenter Point During Control and Experimental Periods 224 Figure 52. A Comparison of the Number of Southbound Whales Occurring Between Rocky Point and Yucca Plant During Control and Experimental Periods 226 Figure 53. A Comparison of the Number of Northbound Whales Occurring Between Rocky Point and Yucca Plant During Control and Experimental Periods 228 Figure 54. A Comparison of the Number of Milling Whales Observed Between Parmenter Point and Yucca Plant During Control and Experimental Periods 232 Figure 55. Abundance of Gray Whales Occurring in Laguna San Ignacio, Mexico, for the Years 1978-1982 and 1984 236 Figure 56. Abundance of Single Whales (A) and Cow/calf Pairs (B) A and B Occurring in Laguna San Ignacio, Mexico, for the Years 1978-1982 and 1984 238 xi i ACKNOWLEDGEMENTS I am most indebted to the Mexican government (Departamento de Pesca) for allowing me to conduct this research. I thank Edith Polanco Jaime, Jorge Carranza Frasier, Alfonso Yanez, Alonso Lopez, Walter O'Campo, Pedro Mercado Sanchez, and Luis Fleischer. Special appreciation is extended to my University research committee; I thank H. Dean Fisher (advisor), P. LeBlond, A. Lewis, J . D. McPhail and W. Perrin for their support and advice. K. Norris and D. Siniff provided valuable external reviews. My gratitude is also extended to several agencies and numerous individuals who assisted at various stages of the research: Naval Acoustic Range, Bremerton, Washington; Hubbs/Sea World Research Institute, San Diego, California (F. Awbrey and W. Evans); American Cetacean Society (San Pedro Chapter); Applied Physics Laboratory, Seattle, Washington (S. Murphy); National Marine Fisheries Service, Graphics Department, Seattle, Washington (K. Conlan and S. Noel); NOAA ships and crew (R/V Miller Freeman and R/V Pi scoverer); Fisherman's Landing, San Diego, California, M/V Pacific Queen (Owner/Operator E. McEwen), M/V Royal Polaris (S. Loomis); M/V Searcher; H & M Landing, San Diego, California (R. Cooperstein, R. Glynn, J . Lobred, D. Mclntyre, C. and R. Miller, S. Porter); M. Bursk, T. Crawford, R. and W. Dahlheim, Sr . , J . Essley, J . Fish, R. Grotefednt, G. and S. Hewlett, J . Heyning, L. Hobbs, D. Ljungblad, L. Martin, R. Merrick, M. Moon, S. Moore, J . Rawlings, B. Reitherman, C. and F. Rodriquez, M. Stinson, J . Sumich, M. Symons, W. Walker, and W. Watkins. My deepest appreciation and sincerest thanks are extended to the five people who supported me throughout this research in countless ways: H. W. Braham, H. Dean Fisher, M. L. Jones, S. Swartz and husband, Bob. xi i i 1 GENERAL INTRODUCTION 2 GENERAL INTRODUCTION There is general agreement among scientists that the acoustical sense of marine mammals, especially that of the cetaceans, constitutes their most important sensory process. Although the sea provides a good medium for sound propagation and reception, communication channels within the aquatic enviroment are usually noisy. Thus, aquatic animals, dependent on acoustical signalling for spacing, attracting or alerting functions must overcome or be able to circumvent this noise in some way to remain effective. A reduction in acoustical transmission and/or its reception can aversely affect the reproduction or even the survival of a given species that is dependent on such a sensory process. In terrestrial studies, many features of animal signals have been shown to minimize the detrimental effect of noise on communication (Konishi, 1970; McGeorge, 1979). To date, quantitative studies addressing cetaceans and the effect of noise oh their acoustic signals have not been attempted. My hypothesis in undertaking this study was that the gray whale, while engaged in underwater signalling, circumvents noise in the acoustical channel by the structure and timing of its calls. To test this hypothesis it was necessary to 1) categorize the types of calls produced by the gray whale, 2) characterize the acoustical habitats occupied by this whale, and 3) determine the relationship between gray whale signals and ambient noise. With the descriptive information available (Chapter 1), experimental playback procedures were initiated to test the hypothesis. The acoustical responses, as well as the observed surface responses, of gray whales were quantified and compared between control and experimental 3 playback sessions. Gray whale responses to the projection of simulated noise sources are addressed in Chapter 2 (short-term playback experiments) and Chapter 3 (long-term playback experiments). A General Summary follows the final Chapter and describes the overall results and significant findings of the research. 4 GENERAL MATERIALS AND METHODS 5 GENERAL MATERIALS AND METHODS  Study Area(s) Laguna San Ignacio, Baja C a l i f o r n i a Sur, Mexico, located along the P a c i f i c Coast of Baja Ca l i f o rn ia approximately 680 km south of San Diego, C a l i f o r n i a , i s one of the major calv ing areas for gray whales ( F i g . 1) . Laguna San Ignacio was a nearly " i dea l " locat ion for the work. A dense concentration of whales occurs in a small area, ensuring a large sample s i z e . It i s a focal point of tour ism, allowing monitoring of gray whale phonations and ambient noise on days with and without vessel a c t i v i t y . Areas within the lagoon are re la t i ve l y undisturbed, ensuring minimum interference with the proposed playback experiments. This lagoon has also been the s i t e of long-term invest igat ions on the gray whale (Swartz and Cummings, 1 978; Swartz and Jones, 1980, 1981; Jones and Swartz, 1984). In addit ion to the primary research conducted i n Mexico, recordings of gray whale sounds and ambient noise were also co l lec ted in the Bering Sea (Alaska) and on the migration route (Washington S ta te ) . Research Equipment - Gray Whale Sound Production In 1981 (8-25 March) at Laguna San Ignacio, a KSP (HS-107/222) 1/ hydrophone was connected through a 2-dB step attenuator to a Nakamichi 550 cassette tape recorder. Maxell UD-90 tape was selected due to i t s low noise cha rac te r i s t i cs . Cal ibrated frequency response was 40 Hz to 19 kHz ± 3 dB. Recordings were co l lec ted from a 4.3-m in f l a tab le boat. Bearings using three landmarks (Weems and Plath hand-held bearing compass) were taken during each session to determine loca t ion . In a l l sampling 1/ Use of trade names does not imply endorsement by author. 6 Figure 1. Study Area - - Laguna San Ignacio, Baja Ca l i f o rn i a Sur, Mexico. 7 8 procedures, the hydrophone was placed at mid-water depth. Gain and attenuator sett ings as wel l as pertinent environmental and behavioral data were recorded on a narrat ive channel. The simultaneous narrat ive included: 1) time of day, 2) t ide condi t ions, 3) sea s ta te , 4) estimated distance of whales from recording platform, 5) behavior, 6) group composition, and 7) to ta l number of whales within visual range. Signals produced by bottlenose dolphins (Tursiops truncatus) were also recorded inc identa l ly during these sessions. Between the years 1982 and 1984 (12 Feb - 2 March 1982, 31 Jan -25 Feb 1983 and 26 Jan to 7 March 1984) at Laguna San Ignacio, a Nagra IV-SJS, ree l - to - ree l tape recorder was coupled with the KSP hydrophone and attenuator used in the 1981 experiments. Ampex 456 audio tape was se lected. This ca l ibrated system provided an e f fec t ive response of 25 Hz to 10 kHz + 2 dB at 9.5 cm/sec and 25 Hz to 20 kHz + 2 dB at 19 cm/sec. Recordings were made at both speeds to determine whether gray whale signals extended above 10 kHz. Cal ibra t ion tones were included in each recording. Beginning in 1982, a bottom-mounted hydrophone system was used to record whale sounds. The hydrophone was suspended from the bottom by a f l oa t set at 3 meters. Water depth was 8 meters. The associated cable was weighted down and extended 25 meters fol lowing the bottom contour of the lagoon terminating at the shore-based s tat ion at Punta Piedra (= Rocky Po in t ) . This system provided cer ta in advantages when compared to the f loa t ing platform used during the 1981 experiments. The bottom-mounted design provided 1) a reduction in system noise ( less d r i f t noise from recording vessel causing hydrophone acce lera t ion) ; 2) less chance of platform in te r fe r ing with whales' normal behavior; 3) an 9 elevated observation platform to document behavior and interactions of whales; and, 4) a 24-hour recording and monitoring system. Simultaneous narratives, similar to the 1981 data collected, accompanied each recording. Recordings were also collected from the 4.3-m inflatable boat at various locations in the lagoon. This portable system ensured overall documentation of the acoustical behavior of gray whales throughout the lagoon area. An LC-32 hydrophone was connected through an U.S. Navy-built amplifier which in turn was connected to a Nakamichi 550 tape recorder. The frequency response of this portable system ranged from 40 Hz to 19 kHz + 3 dB. In the Bering Sea, research was conducted from the NOAA ships R/V MILLER FREEMAN and R/V DISCOVERER during the periods 7-26 July, 1982 and 12-27 September, 1982, respectively. A small skiff (7-m Boston Whaler) deployed from the support vessel served as the recording platform. Whales were first located visually and the skiff was then moved toward them. The outboard engines were shut down, and the hydrophone lowered to mid-water depth, determined by fathometer readings. The calibrated system consisted of a Nagra IV-SJS reel-to-reel tape recorder, Ampex 456 magnetic tape, a KSP Industries hydrophone (Model HS-107/222), coupled to an in-line preamplifier and attenuator, identical to that used in Mexico. Recording locations were determined by radar from the support vessel. A simultaneous narrative accompanied all recordings. Attempts also were made to record gray whales within Washington State waters. A 4.3-m inflatable boat was used as the recording platform. The research equipment, approach to the whales, and subsequent hydrophone drop and narratives, was similar to the methods employed during the 1982 Bering Sea investigations. 10 Research Equipment - Ambient Noise Measurements Ambient noise measurements were collected using the same acoustical equipment and recording platforms as those described when collecting gray whale sound information. Ambient recordings took place either concurrently with attempts to record whales and/or independently. A narrative was included on each tape collected. Sessions were terminated when environmental interference caused excessive noise: for example, 1) Beaufort 2 conditions or greater occurred; 2) the effect of tide caused hydrophone acceleration (= movement of hydrophone causing low-frequency noise); and 3) vessels/skiffs entered the area. Sound spectra of large vessels were recorded at known distances from the port or starboard beam aspect when these vessels were underway and at anchor. Underwater sounds from small skiffs powered by outboard engines were also recorded as they operated at various speeds and idles. In Mexico, these non-biological, man-made sources were collected from both the floating recording platform as well as from the shore-based station. Research Equipment - Playback Experimentation During the 1983 and 1984 playback investigations, a Lubell Acoustic Transducer (Model 98) was bottom-mounted at a distance of 75 m off the farthest promontory available near the Rocky Point shore-based station (Fig. 2). To ensure proper depth, bottom placement and equipment safety, the transducer was affixed to a 1.5-m by 1.5-m square of PVC pipe. Several holes were drilled into this pipe, allowing water to enter into this structure as it was being submerged. A diver was used to guide the transducer cage (in an upright position) to the lagoon bottom. Once the structure was in contact with the bottom, the diver placed four 11 Figure 2. Bottom-mounted Transducer in Laguna San Ignacio, Mexi 13 sand-filled burlap sacks over each lower leg of the cage to secure its location. This anchorage system ensured that no movement of the equipment took place against the strong tidal currents. The measured depth at high tide of this transducer was 8 meters; within the depth specifications required by the manufacturer of this transducer. An average tidal range of 3 meters was documented in Laguna San Ignacio. The associated transducer cable was weighted down at 5-6 m intervals. This cable ran from the cage, along the bottom contour of the lagoon, to the shore-based station at Rocky Point. The design of the cage and cable anchorage system was necessary to ensure that a whale could not become entangled in the acoustical equipment. The shore-based equipment consisted of an Acoustics Systems, Inc., Transmitter (Model TS 107). This transmitter was driven by a 12-volt deep-cycle marine battery. Rotation and re-charging of batteries was neccessary each day to ensure full power to the system. A Briggs and Stratton (40-watt) generator was built for recharging purposes. The level of projected sounds could be controlled by the use of the gain control on the transmitter, spanning the range of 40 to 150 dB re 1 yU'Pa. The projected level varied with frequency. This transmitter was also equipped with a built-in calibration and/or test tone. This test tone swept from 15 kHz to 400 Hz and had a duration of 40 seconds. The input terminals of the transmitter were connected to a Nakamichi 550 cassette tape recorder. The overall frequency response of this system was 40 Hz to 19 kHz ± 2 dB. This equipment enabled the projection of pre-recorded, selected sounds to be played back underwater. 14 CHAPTER 1 SOUND PRODUCTION, AMBIENT NOISE LEVELS, AND THE RELATIONSHIP OF WHALE SIGNALS TO THEIR ACOUSTICAL HABITATS, 15 INTRODUCTION The gray whale, Eschr icht ius robustus ( L i l l j e b o r g , 1861), i s the sole member of the family Eschr ich t i idae (Suborder Mys t i ce t i ; Order Cetacea). Two stocks of gray whales occur in the North P a c i f i c : the "Ca l i fo rn ian" or eastern stock, which breeds along the western coast of North America, and the "Korean" or western stock, which apparently breeds off the coast of eastern Asia (Rice and Wolman, 1971). Although severely depleted in the early 1900s by commercial whaling, the eastern stock has recovered to nearly i t s pre-explo i ta t ion level of 18,500 animals (Re i l l y et a l . , 1983). Unfortunately, the western stock has not had a s im i la r recovery rate and is currently considered to be possibly ext inct (Rice and Wolman, 1971) and/or at c r i t i c a l l y low leve ls (Brownell and Chun, 1977; Blokhin et a l . , 1985). The gray whale formerly occurred in the North A t l a n t i c , as evidenced by subfossi l remains and scanty whaling records, and is considered ext inc t in that area (Mead and M i t c h e l l , 1984). Most mysticete whales undertake annual migrations from the i r feeding locat ions in high la t i tudes of both hemispheres, to t he i r breeding locat ions in more temperate and t rop ica l waters. The gray whale i s no exception and undertakes the longest known migration of any mammal, ranging from the Bering and Chukchi Seas to the waters of Baja C a l i f o r n i a , Mexico (Rice and Wolman, 1971; F i g . 3 ) . During th is annual migration, 90% of the population passes within 3.2 km of the North American coast l ine (Re i l l y et a l . , 1983). Because of i t s coastal habi tat , numerous studies have been accomplished on th is species. These studies have pr imari ly focused on estimates of population levels (Re i l l y et a l . , 1983; Rugh, 1984; Breiwick, York and Bouchet, 1985; Breiwick and Dahlheim, 1986); feeding habits (Ol iver et a l . , 16 Figure 3. The Migration Route of the Gray Whale (Eschrichtius robustus). Modified from Reilly et a l . , 1 983. i 7 18 1983, 1984; Nerini, 1984); studies on the abundance and distribution within breeding lagoons (Bryant et a l . , 1984; Jones and Swartz, 1984); and distribution and migration (Braham, 1984; Herzing and Mate, 1984; Miller et a l . , 1985). Despite the numerous contributions to the scientific literature on gray whales, l i t t le has pertained to their sound production behavior. Prior to 1967, only anecdotal information on the sounds produced by gray whales occurred in the literature. Aldrich (1889) reported that devilfish (gray whales) were known to "sing" on their northern feeding grounds. Tomilin (1957) reported that Soviet whalers operating in the Bering and Chukchi Seas detected "low-pitched roars" produced by gray whales. Eberhardt and Evans (1962), recording in Laguna Ojo de Liebre, noted "croaker-like grunts" and "low-frequency rumbles" which they ascribed to gray whales. Wenz (1964), in collaboration with Asa-Dorian, reported having detected "clicks" from migrating gray whales. Painter (1 963) tentatively identified a low "grunting" sound recorded in the coastal lagoons of Mexico as having been produced by gray whales. Recordings made by Rasmussen and Head (1965) in Laguna Ojo de Liebre and off San Diego, California contained no signals attributable to this species. However, recordings they made in the presence of gray whales off Todos Santos Bay, Baja California Norte, Mexico included a "series of intense sounds" they likened to sound produced by hammering against the hull of a wooden ship. Hubbs (1 966) obtained no recordings of gray whale sounds during work around solitary individuals and cow/calf pairs in Laguna San Ignacio. Between 1967 and 1978, various other researchers attempted to record sounds produced by the gray whale (Table I). In reviewing the literature, it became apparent that the acoustical Table I A Summary ol Published Data on Sounds Produced by the Cray Whale 11967-1978) Pulse 1 Frequency Peak energy Duration Pulses/ Pulse duration Reference Sludy area Sound type range (Hj) IHi) (sec) burst rate (sec) Asa Dorian and tn Laguna Ojo de Pulses 70-3000 4 0 0 - 8 0 0 5-22 Perkins. 1967 Liebre. Laguna Moans <2000 GiJerrfro Noriro. Whittles <2000 and oil Sim Diego. California Cummings ef a/.. O i l San Diego. Cal- Moans 2 0 - 2 0 0 _ 154 — — 1968 ifornia Blow sounds 15-175 100 Bubble-type 15-305 0 7 0 I Knocks Up to 350 Pouller. 1968 In Laguna Ojo de Grunts — — — Liebre Rumbles Cries Rasps Up to 12.000 100 Pulses 5-18/sec Chirps 2 0 - 5 0 8/sec Clicks U p to 12.000 Bong Up to 12.000 Fish ef at. 1974 (a) In captivity and Growl or moan 100-200.1500 100 — during open-water Pulse train 100-10000.1400 2 0 0 8-14 14/sec release Grunt 200-400.1600 020 Blowhole rumble Clicks 2000-6000 3400-4000 0 001-0 002 1-833 9-36/jec (b) In Wlckanlnnlsh Click trains 2000-6000 3500-4000 < 0 0 0 2 1-96 8-40/sec Bay.. Vancouver Island Ray and Schevlll. In capMvrly Noisy pulsation — — j 1974 Norris e( a / . 1977 In Bahla Magdalena Pulses 1 <100 Clicks 100-20000 0 i 5 2/sec Grunts Swan/ and Cum- In Laguna San Igna Knocks 200-1500 350 and 800 4 00 111 0 1 0 mings. 1978 CIO Pulses 400 O i O Snorts 6 6 - 6 0 0 1.00 -35/sec Slamming 50-2000 0 3 0 Bubble bursts Up to 1400 400-1200 4 0 0 1 Squeaking moan 700-1200 0 6 0 Comments In leal Pulses may be used in echolocaiion Soniferous day and n>ghi Assumed sounds wen) communicative Bong was unusual and may be alert signal Noisy pulsation heard during feeding 20 behavior of gray whales was not definitively documented and that additional long-term investigations could benefit our understanding of sound production by the gray whale. To test the stated hypothesis of this study, i t was also necessary to profile the acoustical habitats occupied by these whales. There has been an increasing awareness that ambient noise and its potential effect on marine mammals can no longer be ignored in the study of underwater bio-acoustics. Ambient noise is defined as "the contribution of all background oceanic noise resulting from either natural means and/or introduced sources" (Urick, 1967). Ambient noise therefore may consist of 1) sounds produced from a wide variety of biological organisms such as invertebrates, fish and marine mammals; 2) sounds associated with the physical properties related to ocean dynamics (Beaufort state, surf noise, weather, tidal currents, etc.); and 3) sounds introduced by man, such as vessel noise, aircraft noise, and noise related to industrial activities. A profile of ambient noise typically contains a statement of frequencies emphasized, as well as the associated levels at these frequencies. The ambient level is the intensity, in decibels, of this ambient background measured with a nondirectional hydrophone and referred to the intensity of a plane wave having a root mean square (rms) pressure of one micropascal (1 juPa). The sea furnishes a superior medium for sound propagation and reception, with sound waves travelling at 1500 m/sec, a value four times that of the speed of sound in air (Urick, 1967). Conversely, the sea can also represent a very dynamic environment for a travelling sound wave. A signal propagating through the aquatic medium can become delayed, distorted, enhanced and/or weakened. Propagation pathways can differ greatly depending upon the 21 physical characteristics of the area. Distinct differences in ambient levels and frequencies also exist regionally, depending upon the composition of sound-producing marine l i fe, the physical properties of the area, and the relative amount of human activities. Within a specific area, daily and/or seasonal changes can occur, reflecting the constant state of flux of ambient noi se. Extensive literature is available defining and identifying ambient noise in the world's oceans (Heindsman et a l . , 1955; Wilkenshaw, 1960; Wenz, 1961, 1962; Ezrow, 1962; Calderon, 1964; Green and Buck, 1964; Arase and Arase, 1966). In most cases, however, detailed measurements have not been made for a specific area. Rather, the information for a particular region may be extrapolated from other areas with similar oceanic characteristics and then modelled for the area in question. Since it was critical to this project to define, profile, and understand the acoustical environments occupied by gray whales, it was necessary to collect on-site measurements of ambient noise rather than rely on extrapolated information. In addition to collecting information on ambient noise levels in various parts of the gray whale's range, factors that could potentially influence a travelling sound wave were investigated as well. These included depth, temperature, salinity, and bottom substrate. To obtain precise information on propagation pathways in Laguna San Ignacio, a series of propagation experiments was conducted. These experiments provided a greater insight into the acoustical habitats of the whales and also added a critical element needed during the playback experimentation discussed in Chapters 2 and 3. Communication between individuals can be accomplished in a variety of ways. For those species inhabiting the aquatic environment (e.g., 22 cetaceans), certain restrictions are inherent on the use of different signalling methods. Vision appears to be well developed in those cetacean species studied (Breathnach, 1960; Kellogg and Rice, 1966; Caldwell and Caldwell, 1972). However, underwater vision is usually restricted to short distances. Chemical signals diffuse slowly and dissipate quickly, which could negate their use over long distances. In addition, the reduced and/or non-existent olfactory tracts in cetaceans imply l i t t le or no dependence on this sensory mechanism (Jansen and Jansen, 1953; Caldwell and Caldwell, 1972). Taking into account the advantages that a travelling sound wave has in the ocean environment, it appears that signalling and responding through the use of sound are by far the most effective means of exchanging information over a long distance in the aquatic environment. To determine if ambient noise has the potential of interfering with or masking the signals produced by an acoustically-dependent animal, the sound repertoire of that species, as well as the characteristics of its acoustical habitats, must be investigated and quantified. The relationship between the animals' calls and their environment can then be defined. This Chapter documents the types of calls produced by the gray whale and profiles the ambient noise conditions encountered by this species. The relationship of gray whale signals to ambient noise is inspected to determine i f the main sources of channel noise could affect the underwater signals produced by the gray whale. 23 MATERIALS AND METHODS Analysis Procedures - Gray Whale Sound Production Each underwater recording (magnetic tape) collected was edited and the pertinent information transferred to a written record. The edit sheet contained tape identification number, time and date of recording, tape count, type of sound emitted, subjective appraisal of the signal level (minimum, moderate, high), and general comments which included all pertinent behavioral or environmental data. Sound types were init ial ly categorized aurally. These categories were then visually inspected and verified through analysis on a Kay Elemetrics Corporation Sona-Graph (Model 7029A). Since onomatopoetic terms denote different things to people, I elected to use a numbering system to categorize gray whale sounds. Each distinct sound type was numbered. A letter was placed prior to the number to indicate the area in which this sound was recorded and documented. For example, an SI signal (southern range signal) would be designated as an NI signal on the northern range and/or an Ml signal if produced during the migration. During analyses, sounds of high quality, without interference of man-made noise, were selected, which provided better resolution and accuracy during quantification. The analyses of these high signal-to-noise calls resulted in the major conclusions drawn regarding sound categorization of gray whale signals. » Since only one omni-directional hydrophone was used, it was not possible to obtain source levels of gray whale cal ls. However, received levels were computed and subsequently used to profile this parameter of gray whale signals. Hydrophone sensitivity, amplifier gain relative to attenuator 24 settings and calibration tones recorded on tape during field collections were used to compute these received levels. A Nicolet Scientific Corporation FFT Computing Spectrum Analyzer, Model 446 was used to compute pressure levels. Displays for these spectra were made on an X-Y plotter. The results were then digitized using a plastic overlay designed by the author. The overlay format represented frequency values on the x-axis with dB levels (re 1 juPa) on the y-axis. Analysis Procedures - Ambient Noise To calculate biological ambient noise levels, one-third octave average sound pressure levels were computed using a Spectral Dynamics 345 Spectrum Analyzer set for 0-20 kHz bandpass and either 32 or 64 linear averages. The reported average sound pressure levels were calculated by averaging as many linear averages as possible from a single biological station. Using this method, the reported levels for the stations represent a range of actual sampling times equivalent to a flat weighted average sound pressure meter sampling from 1.28 to 14.0 sec. One-third octave levels for large vessels were calculated using a single output of 64 linear averages from a Spectral Dynamics 345 Spectrum Analyzer set at a 20 kHz bandpass. These data were not range corrected; thus, they represent levels received at the hydrophone. The outboard engine one-third octave levels were sampled using a General Radio 1995-9008 equipped with a 10 Hz to 80 kHz bandpass. The General Radio equipment was chosen for its high sampling rate which was needed during the analysis of the shorter-term signals produced by outboard engines. These levels were corrected assuming spherical spreading (20 log R) to 1 yd. Although Laguna San Ignacio represents a shallow-water environment (<30m), free-field 25 conditions (spherical spreading) were assumed during this phase of the research because of the close proximity of the outboard engines to the recording instrumentation. During analysis of the Bering Sea and Washington State data, hydrophone sensitivity, amplifier gain relative to attenuator settings, and calibration tones recorded on tape during field collections were used to compute ambient levels. Average sound pressure levels were computed using a Nicolet Scientific Corporation FFT Spectrum Analyzer, Model 446. The high pass f i l ter was set at 10 kHz, and recordings were averaged over three minutes. Displays of the calculated spectra for ambient sources were made on a X-Y plotter. Analysis Procedures - Relationship Between Whale Calls and Ambient Noise To determine the possible interference/masking effect of background noise on gray whale signals, several aspects of channel occupation were quantitatively examined. Sounds produced by the gray whale were examined, with particular attention to emphasized frequency bands. The received levels of sound at these emphasized frequencies were noted as well. Variations within each call type were listed to determine i f modifications occurred in call structure: e .g . , 1) repetition rates of calls; 2) amount of frequency modulation; and 3) duration of signals. Gray whale signals produced on different parts of this species' range were compared to determine if these call modifications could provide a selective advantage in different habitats or behavioral contexts. An inspection of the dominant frequencies and levels produced by natural background ambient conditions (e.g., biological organisms) was compared to the frequencies and levels emphasized by the gray whales. 26 In addition to the ambient noise profiles resulting from the biological component, an inspection of the frequencies and levels associated with different oceanic conditions was also reviewed. This was accomplished through my own field collections and a review of the appropriate literature. These data were compared to those collected on sound production by the gray whale. Man-made noise sources were inspected as well, and the dominant frequencies and associated levels were listed. When signals produced by the gray whale occurred at.the same time that a man-made event occurred, composite sonograms were made to illustrate the frequencies emphasized by both of these sources. 27 RESULTS TYPES OF SOUNDS PRODUCED Laguna San ignacio (Mexico) Based on 565 hrs of underwater recording, I have c l a s s i f i e d the sounds produced by the gray whale in Laguna San Ignacio into s i x d i s t i nc t types (Table I I ) . F i e l d observations provided assurance that these signals were in fact produced by gray whales. Typ ica l l y , signal strengths would increase as whales approached the hydrophone and decreased as whales moved away. In add i t ion , the or ig ina l recordings of a captive gray whale were obtained and compared to those ca l l s co l lec ted in Mexico. And f i n a l l y , recordings were co l lec ted in Laguna San Ignacio when whales were not present. The most prevalent phonation (83% of a l l s ignals) recorded during the invest igat ions was a pulsing sound which was termed SI (F i g . 4) . These t ransient pulses were broadband, with energy ranging from less than 100 Hz to 2 kHz, but usual ly having an emphasized band between 300 and 825 Hz. Although s ingle pulses were recorded, t yp i ca l l y SI signals occurred in a ser ies or burst, ranging from 2 to 30 with an average of 9.4 pulses per se r ies . The durations of these ser ies averaged 1.8 sec. Each pulse, however, was less than 0.05 sec long. Average pulse repet i t ion rate was 5.9 per sec, with a range of 2.2 to 14.7 sec. Frequently, these pulse modulated sounds were also frequency modulated, as shown in F i g . 5. Pulses were often meta l l i c sounding; however, t h i s aural charac te r i s t i c appeared to vary considerably. Received levels ranged from 90-160 dB re 1 uPa, with an average level calculated at 118 dB re 1 juPa. The second sound type was termed an S2 signal ( F i g . 6) . In a l l cases, S2 signals were characterized by a rapid FM up/down sweep. A meta l l i c Table I I . A C lass i f i ca t i on of Sounds Produced by the Gray Whale in Laguna San Ignacio, Mexico.* Number of Pulses Sound Frequency Range Energy Concentration Duration Pulses Per Per Received type n Low High Low High of Series Series Second Levels SI 1000 90 (120-1250) 1940 (1000-3000) 332 (100-1250) 824 (400-1600) 1.8 (0.3-4.6) 9.4 (2-30) 5.9 (2.2-14.7) 118 (90-160) S2 100 250 (100-300) 300 (200-350) 250 (200-275) 300 (250-350) 0.3 (0.2-0.4) — 102 (92-127) S3 100 125 (80-200) 1250 (750-1800) 170 (125-625) 430 (300-750) 2.0 (1.0-4.0) — ** 94 (90-113) S4 100 150 (125-200) 1570 (1500-1600) 225 (125-300) 600 (450-750) 0.9 (0.7-1.4) ** 92 (88-108) S5 100 130 (20-250) 840 (600-1500) 200 (20-250) 500 (300-750) 3.2 (1.8-4.5) — 1 18 (97-124) S6 100 250 (125-300) 850 (600-1000) 250 (125-300) 700 (300-800) 3.3 (1.2-3.6) — 102 (96-113) * Mean values are given and the i r associated ranges are shown in parentheses. A l l frequency values are given in Hz and time values in seconds. Received levels are shown in dB levels re 1 juPa. Pulse modulation occurr ing. 29 Figure 4. Spectrograph of a Gray Whale SI Seri 31 Figure 5. Spectrograph of Two SI Series Depicting Frequency Modulati Frequency (Hz) Ml 33 Figure 6. Spectrograph of a Gray Whale S2 Signal. 35 aural qual i ty was prevalent in a l l S2 s igna l s . A mean frequency range of 250 to 300 Hz was calculated with an average duration of 0.3 sec. Compared to a s ing le SI pulse, S2 signals had power emphasized frequencies, were of longer duration, had less var ia t ion , and did not occur in a se r ies . Five percent (5%) of the sample contained S2 s igna ls . Received levels ranged from 92-127 dB re 1 uPa, with an average of 102 dB. The th i rd type of s ignal (S3) was characterized by i t s low frequency range (125 Hz to 1250 Hz) and long duration (1.0 to 4.0 sec) . Energy was usual ly concentrated below 430 Hz. The spectrograph in F i g . 7 indicates that th is signal i s pulse modulated. S3 signals represented 1% of a l l c a l l s recorded. Received levels were considerably lower as compared to the SI and S2 signals and ranged from 90-113 dB re 1 uPa, with a mean of 94 dB. Some v a r i a b i l i t y occurred within th is s i g n a l , but to a lesser extent than that of the SI s i gna l . S4 signals ( F i g . 8) were t yp i ca l l y of higher mean frequencies (150-1570 Hz) than S3 s igna l s . Average energy was centered between 225 and 600 Hz. Average duration was short (< 1.0 sec) . Five percent (5%) of the sample contained S4 s igna l s . This sound type appeared to be pulse modulated as we'l l . L i t t l e var ia t ion occurred in th is s i gna l . Received levels ranged from 88-108 dB re 1 uPa, with an average of 92 dB. In Laguna San Ignacio, gray whales often release large amounts of a i r underwater. The released a i r r ises to the surface where i t develops into a large c i r cu la r bubble mass. This behavior has been termed a "bubble b last" or "bubble burst" by previous invest igators (Swartz and Cummings, 1978). The sound (S5) associated with bubble blasts i s acoust ica l ly d i s t i nc t and readi ly i den t i f i ab l e (F i g . 9). Mean frequency ranges were measured between 130 and 840 Hz, with most energy concentrated below 500 Hz. The average 36 F i g u r e 7. S p e c t r o g r a p h of a Gray Whale S3 S i g n a l . 2000 38 Figure 8. Spectrograph of a Gray Whale S4 S i g n a l . 39 40 Figure 9. Spectrograph of a Gray Whale S5 Signal (bubble b l a s t ) . 42 duration of this signal was variable, ranging from 1.8 to 4.5 sec. Bubble blasts could be heard underwater at estimated distances up to 2.5 km. A received level of 97-124 dB re 1 /jPa was calculated, with a mean of 118 dB. Frequently, gray whales also exhaled underwater trails of bubbles just prior to breaking the surface to breathe. Again, the sound (S6) was distinct aurally. Mean frequency ranges of the sub-surface exhalations were between 250 and 850 Hz, with major energy concentrated at frequencies less than 700 Hz (Fig. 10). Average duration was 3.3 sec. Levels received at the hydrophone ranged from 96-113 dB re 1 uPa, with an average of 102 dB. The six major sound types were heard under a wide variety of behavioral activities. No attempt was made to correlate sound production and behavior, with the exception of the bubble blast and sub-surface exhalation. To determine the periodicity of total sound production in the lagoon during February and March, the total number of sounds produced per hour was calculated for each month. In addition, the average number of whales occupying the lagoon each month was estimated. Out of 565 hours of recordings, 395 hours were recorded in February and 170 hours were recorded in March. In February, an average of 84 sounds were collected per hour. This value ranged from 62-128 sounds produced per hour. In March, total sound production per hour dropped to 38-86 sounds/hr, with an average of 50 sounds produced per hour. These monthly totals did not include the S5 or S6 signals. The abundance of whales calculated over a five-year period has been reported by Jones and Swartz (1984). Taking an average of their mid-February and March counts during the years 1978-1982 resulted in the following. The average number of whales occurring in the lagoon in February was 258 whales and in March 195 whales. If the total average number of 43 Figure 10. Spectrograph of a Gray Whale S6 Signal (sub-surface exhalati ¥9 45 sounds produced per hour by month i s divided by the average number of whales occupying the lagoon by month, the number of sounds produced per hour per whale i s obtained (sounds/hr/whale). In February, 0.33 sounds/hr/whale was ca lcu la ted . In March, 0.25 sounds/hr/whale was obtai ned. In comparing the types of signals produced and per iod ic i t y of phonations over a 24-hour per iod, no dif ferences in gray whale sound production were found (t = 1.27, p > 0.05). SI signals dominated the vocal reper to i re , and the level of sound a c t i v i t y remained the same. S t . Lawrence Island (Alaska) Despite considerable ef for t (40 hrs of recording), few signals were recorded in the presence of feeding gray whales. During J u l y , 30 hours of recordings were co l lec ted ; however, unfavorable weather in September l imited co l lec t ions to 10 hrs (F i g . 11). Although at times the whales were in close proximity to the recording platform ( < 1 0 m), the received levels of the i r sounds were considerably lower than the levels reported for Mexico. It is poss ib le , however, that the s ignals actua l ly recorded were not those of the whales within close range but represented sounds produced by distant whales. The low levels of these s ignals made spectral analyses impossible (s ignal - to-noise ra t io = 0 dB). In edi t ing these tapes, one can aural ly d is t ingu ish gray whale s igna ls ; however, the resultant visual display could not d is t inguish the signal from the background noise caused by the pi tch and r o l l of the recording platform and subsequent movement of the hydrophone. A to ta l of 53 sounds was recorded and aural ly c l a s s i f i e d into four d i s t i nc t c a l l types: 1) broadband pulses produced in a ser ies (NI s i g n a l ; 46 Figure 11. Recording Locations of Gray Whales off St. Lawrence Island, Bering Sea, Alaska, Summer 1982. 48 n = 14); 2) low frequency moans (N3 signal; n = 3); 3) grunt-like phonations (N4 signal; n = 6); and 4) sub-surface exhalations (N6 signal; n = 30). The received level of the N6 signal was of sufficient strength for analysis. This sound emphasized frequencies between 100 Hz and 1.0 kHz with energy concentrated below 600 Hz. Average duration was 4.2 sec (Fig. 12). Received levels were comparable to those documented in Mexico, averaging 100 dB (re 1 juPa). During 40 hours of recording, a total of 23 signals was documented (excluding sub-surface exhalations, n = 30). This reflects an hourly sound production rate of 0.58 signals/hr. If this value is then divided by the average number of whales occupying this area (approximately 100 animals, based on an visual estimate only), the resultant index of 0.006 sounds/hr/whale is obtained. Preliminary results indicate that sound activity is significantly reduced on the northern range as compared to sound production in Mexico. Due to the small sample size on the northern range, meaningful comparisons of sound periodicity over a 24-hour cycle could not be investigated. Washington State Waters Limited field work was accomplished in Washington waters. A total of 5 hours of recordings was collected from two locations within Washington State: 1) coastal waters (La Push, Washington; Fig. 13); and 2) inland waters (Langley, Whidbey Island, Puget Sound; Fig. 14). In coastal waters, three juvenile gray whales (approximately 10 m in length) were observed. One juvenile whale (less than 10 m) was observed during inland recordings at Langley. No sounds attributable to the gray whale were recorded during these sessions. At each location, it appeared that these juvenile whales 49 Figure 12. Spectrograph of a Sub-surface Exhalation (N6) Produced by a Surfacing Gray Whale off St. Lawrence Island, Alaska. 50 M (ZH) AouBobaoj 51 F i g u r e 13. Study Area -- C o a s t a l Waters of Washington S t a t e . 52 53 Figure 14. Recording Locations of Gray Whales in Washington State Waters. 54 55 were feeding, as evidenced by associated mud plumes and circling behavior (Moore and Ljungblad, 1984). Integration of Regional Data Bases/Pub1ished Information After comparing my spectrographs and recordings with those previously published and with the original tapes supplied by previous authors, a synthesis of all available data on calls produced by the gray whale was completed (Table III). Seven categories of sounds were established for the gray whale throughout its range. Signals produced by the gray whale (SI through S4) on its southern range emphasized the frequency ranges between 225-824 Hz. The received levels of these four southern call types varied, averaging 92-118 dB re 1 /jPa. The SI signal dominated the vocal repertoire of gray whales in Mexico. In addition, this sound also showed the most variation in call structure when compared to the S2, S3 and S4 signals. Variations within the SI signal were observed in: the number of pulses per series (2-30); the repetition rates (2.2-14.7 pulses/sec); the amount of frequency modulation; the duration of the total signal (0.3-4.6 sec); and in the range of received sound levels (90-160 dB re 1 uPa). Little variation occurred within the S2, S3 and S4 call types. When comparing the recordings collected over a 24-hour period in Mexico, no significant changes occurred in call types and/or calling rates. The overall calling rates, of course, were significantly higher in Laguna San Ignacio than in other areas. Comparisons of call structure from the northern and southern ranges revealed differences. On the northern range, as on the southern range, the NI signal exhibited the most variation among signals and represented the most frequently used phonation. NI signals were broadband, ranging 56 Table I I I . Sounds Produced by the Gray Whale Throughout I ts Range Based on Acoust ical Studies in Laguna San Ignacio and Published Accounts (only those accounts with su f f i c ien t data for comparison were inc luded) . Sound type' Geographical area Reference SI' S2 S3 S4 S5 S6 S7 S/N/M* S S/N/M S/N/M S/M S/N/M S/M This Chapter; Asa-Dorian and Perkins, 1967 (pu lses) ; Fish et a l . , 1974 (pulse t r a i n s ) ; Swartz and Cummings, 1978 (knocks and pu lses) ; Moore and Ljungblad, 1984 (knocks); Norris et al 1977 (c lack) . This Chapter. This Chapter; Cummings et a l . , 1968 (moans); Fish et a l . , 1974 (growls or moans); Moore and Ljungblad, 1984 (moan). This Chapter; F ish et a l . , 1974 (grunt); Swartz and Cummings, 1978 (snor t ) ; Moore and Ljungblad, 1984 (grunt). This Chapter; Cummings et a l . , 1968 (bubble-type); Swartz and Cummings, 1978 (bubble burs ts ) . This Chapter; Cummings et a l . , 1968 (blow sounds). Fish et a l . , 1 974 ( c l i cks and c l i ck t r a i n). * An SI signal produced on the northern range would be designated as an NI; the same type of signal produced during migration would be designated as M l . * * S, southern range; N, northern range; M, migration route. 57 from 200-4000 Hz, with most energy concentrated between 238-2584 Hz. Typ ica l ly th is signal contained 2-69 pulses per ser ies (x~ = 12), and was produced at a mean repet i t ion rate of 7.14 pulses per second. Signal duration averaged 1.4 sec, with a range of 0.4 to 2.3 sec. Upon inspection of the NI c a l l type, no evidence was found of frequency modulation. The d i f f i c u l t i e s during analysis (poor signal to noise rat io) suggested that these northern signal strengths were low and approximated those of ambient condit ions (estimated to be 65 dB re 1 uPa). A comparison between SI and NI s ignals revealed the fo l lowing. NI s ignals emphasized a broader and s l i gh t l y higher frequency band than SI s igna ls . The repet i t ion rates and number of pulses per ser ies of NI s ignals exhibi ted higher values; however, these northern signals were shorter in durat ion. The elaborate frequency modulation prevalent in the SI signal did not occur in the NI c a l l s . Received levels of sound were considerably higher on the southern range. In comparing the information ava i lab le on the S3 type c a l l s among the three areas of the gray whales' range, modif icat ions were also noted in ca l l s t ruc ture. A l l "3-type" c a l l s were pulse and frequency modulated. C o l l e c t i v e l y , these signals covered the frequencies of 20-1250 Hz, with the M3 c a l l representing the narrowest range (20-200 Hz), N3 showing an intermediate value (102-578 Hz), and S3 exhib i t ing the widest range (125-1250 Hz). Energy concentrations within the S3 and N3 signal were s im i l a r . Ca l l duration also varied among areas, with S3 signals showing the longest durat ion, followed by M3, and the shortest signal being the N3. Received levels were greatest in Mexico. Although Cummings et a l . (1968) reported a value of 152 dB re 1 /iPa for th is signal off C a l i f o r n i a , t h i s value represented a source l e v e l ; thus, va l i d comparisons could not be drawn. 58 The N4 signal exhibited a narrower range of frequencies than the S4 signal; however, the energy concentration of the S4 signal (225-600 Hz) included the 388 Hz mean value noted for N4 signals. N4 signals were considerably shorter in duration (x = 0.344 sec) than S4 signals (0.944 sec). Received levels of these calls were higher on the southern range (92 dB vs 65 dB re 1 /jPa). The tabulated data on these sound type comparisons are presented in Table IV. Table IV. Comparisons of Gray Whale Sounds Throughout Its Range. Energy Pulses Pulses Frequency Amplitude Received Sound Frequency Concentration Per Per Duration Modulation Modulation Levels dB Type Range (Hz) (Hz) Series Second (sec) Occurring Occurring re 1 pPa Reference SI 90-1940 332-824 9.4 (2-30) 5.9 (2.2-14.7) 1.8 (0.3-4.6) Yes Yes 118 (range) Chapter 1 NI 200-4000 963 (238-2584) 12.0 (2-69) 7.1 1.4 (0.4-2.3) No Yes 65 (range) Moore and Ljungblad, 1984; Ch.l S3 125-1250 170-430 — — 2.0 (1.0-4.0) Yes Yes 94 (range) Chapter 1 N3 102-578 325 — — 0.9 Yes Yes 65 (range) Moore and Ljungblad, 1984; Ch.l M3 20-200 — — — 1.4 Yes Yes 152* Cumrni ngs et a!.,1961 S4 150-1570 225-600 — — 0.9 (0.7-1.4) Yes Yes 92 Chapter 1 N4 136-748 388 — — 0.3 (0.2-0.5) Yes Yes 65 Moore and Ljungblad, 1984; Ch.l * Represents source level of signal. 60 AMBIENT NOISE Laguna San Ignacio (Mexico) Biological ambient noise was recorded at 32 stations, 14 in the outer lagoon and 18 in the inner lagoon (Fig. 15). Although noise spectra varied among the stations, typically they followed the pattern depicted in Fig. 16. Levels were lowest below 2 kHz, increased to high levels between 2 and 5 kHz and then declined gradually through 20 kHz. The average ambient noise levels attributable to biological sources (excluding cetaceans) ranged from 94 to 110 dB re 1 yuPa. The levels and associated octave-band frequencies for each of the 32 stations are listed in Table V. An inspection of the synoptic ambient samples (24-hour recordings) from Rocky Point revealed no significant differences in levels or frequencies over a 24-hour period (t = 1.17, p > 0.05). Increase in sea states showed a corresponding increase in ambient levels. These spectra were compared to the standard open-ocean reference spectra developed by Knudsen (1948), which provide a standard for judging relative noisiness (Fig. 17). When the spectra of ambient noise obtained in Laguna San Ignacio are compared with this standard, the resultant conclusion is that ambient noise levels are extremely high in the lagoon habitat. The biological component of noise dominates the spectra. Numerous biological organisms were identified and deemed possibly responsible for the high levels of noise received (Table VI). The only sources of anthropogenic noise occurring during these Mexican studies were vessel and skiff traff ic. Data were collected on six large vessels (25-35 m in length) and 14 skiffs powered by outboard engines. The sources of noise radiating from these large vessels can be grouped into 61 Figure 15. Locations of Biological Ambient Noise Stations in Laguna San Ignacio, Mexico. 6*2 113° 06'W 63 F i g u r e 16. S p e c t r a o f T y p i c a l B i o l o g i c a l Ambient N o i s e L e v e l s i n Laguna San I g n a c i o , Mexico. 6 4 V 3 —OCT A V E BANDWIDTH C O R R E C T I O N TO S P E C T R U M L E V E L IFOR C O N T I N U O U S S P E C T R A O N L Y I / 3 - O C T A V E - 8 A N O C E N T E R F R E Q U E N C Y Table V. Estimated Levels (dB re I /jPa) and Their Associated Frequencies (Hz) for Biological Ambient Noise Stations 1 - 32 in Laguna San Ignacio, Mexico Bi ologi cal Estimated F requency (Hz) Noi se Level (dB Station re 1 uPa) 1250 1600 2000 2500 3150 4000 5000 1 * 97 X X 2 94 X X 3 103 X X 4 107 X X X 5 98 X X X 6 109 X X 7 97 X X X 8 94 X X X 9 99 X 10 98 X 11 105 X X X 12 102 X X X X 13 100 X X X X 14 103 X 15 102 X 16 105 X 17 106 X 18 104 X 19 99 X 20 96 X 21 103 X X X 22 102 X X 23 102 X X 24 106 X X X X 25 103 X 26 110 X X X 27 106 X X X 28 105 X 29 109 X 30 108 X X X 31 99 X X X 32 95 X * For example, Station #1 had a maximum level of 97 dB in both the 4 and 5 kHz bands. uf 66 F i g u r e 17. Knudsen Sea S t a t e C u r v e s . 68 Table VI. Biological Organisms Potentially Responsible for the High-Levels of Ambient Noise in Laguna San Ignacio, Mexico (scientific and common name listed).  FISH LIST FOR LAGUNA SAN IGNACIO, BAJA CALIFORNIA SUR, MEXICO Manta hamiltoni Pacific manta Mobula lucasana Smoothtail mobula Myliobatis californica Bat ray Urolophus ha 1leri Round stingray Atherinopsis californiensis Jacksmelt Atherinops affinis Top smelt Hippocampus ingens Pacific seahorse Scorpaena guttata Sculpin Stereolepis gigas Giant sea bass Mycteroperca xenarcha Broomtail grouper Mycteroperca jordani Gulf grouper Epinephelus analogus Spotted cabri11 a Paralabrax maculatofasciatus Spotted sand bass Paralabrax sp. Grass bass Paralabrax nebulifer Barred sand bass Cynoscion xanthulus Orangemouth corvina Cynoscion parvipinnis Shortfin corvina Gi rel1 a ni gri cans Opal eye Hypsypops rubicundus Garibaldi Mugil cephalus Striped mullet Paralichthys californicus California halibut Alopias vulpinus Common thresher Carcharodon carcharias White shark Sphyrna lewini Scalloped hammerhead Isurus oxyrinchus Bonita shark Triakis semifasciata Leopard shark Carcharhinus leucas Bull shark Rhinobatos productus Shovelnose guitarfish Cheilotrema saturnum Black croaker Halichoeres semicinctus Rock wrasse Sphoeroides annulatus Builseye puffer Oxyjulis californica Senorita Pimelometopon pulchrum California sheephead Heterodontus francisci Horn shark Balistes polylepis Finescale triggerfish Anisotremus davidsonii Sargo Calamus brachysomus Pacific porgy Epinephelus itajara Jewfish Rypticus sp. Soapfish Rajidae Fami ly Skates Pomadasyidae Family Wavy line grunt Chaetodontidae Family Cortez angel 69 Table VI. Continued - -. . .conti nued MARINE INVERTEBRATES OF LAGUNA SAN IGNACIO, BAJA CALIFORNIA SUR, MEXICO Haliotus fu1 gens  Haliotus corrugata  Diodora di gueti Col 1iselI a strigatella Calliostoma gemmulatum Cal1iostoma gardanum Calliostoma keenae Cal1iostoma rema Tegula aureotincta Tegula fenebralis Tegula gal 1ina Turbo fluctuosus Turbo funiculosus Astraea undosa Homalapoma luri dum Nerita scabricosta Theodoxus luteofaciatus Littori a fasciata Vermiculari a pelluci da eburnea Cerithium stercusmuscarum Cerithidae californica Cerithidae albonodosa Cerithidae mazatlanica Cerithidae montagnei Epitonium tinctum Crepidula arenata Crepidula onyx Crepidula excavata Hi pponi x tumens Natica chemnitzii Poli ni ces uber Poli ni ces reeluzi anus Polinices altus Cassis centriquadrata Ficus ventricosa Malea ringens Muricanthus nigritus Eupleura muriciformis Ceratostoma monoceros Pteropupura festiva Pteropupura erinaceoides Ocenebra foveolata Acanthi na 1ugubri s Forreria belcheri Green abalone Pink abalone Keyhole 1 impet Limpet Gem top Gi1ded tegula Black turban Speckled tegula Speckled top Wavy top Mangrove periwinkle Turret California horn Horn shel1 Wentletrap SIi pper Onyx slipper Excavated sii pper Hoof shelI Moon shel1 White moon shel 1 Fi g shelI Tun shel1 Black murex Festive murex Unicorn Belcher's murex 70 . . .conti nued Table VI. Continued - -MARINE INVERTEBRATES, CONTINUED Thais biserialis Dogwinkle Thais speciosa Macron aethiops Macron orcutti Smooth macron Anachis adelinae Adaline's pyrene Anachis corunata Crown pyrene Melongena patula Fossil? Nassarius tegula Mud Nassa Fusinus irregularis Spindle shel1 01 iva i ncrassata Shouldered olive 01i vel1 a beati ca 01ivella Cancellaria ventricosa Pseudomelatoma pencil lata Turrid Terebra pedroana Pedro auger Terebra tiarel1 a Auger Ophiodermella ophioderma Turrid Bulla gouldiana Gould's bubble Haminoea virescens Green bubble Melampus mousleye Salt marsh snail Area pacifica Turkey wing Barbatia rostae Arc Barbati a alternata Arc Barbatia reevena Arc Andara tuberculosa Arc with nodes Andara grandis Massive arc Andara multicostata 34 ribbed arc Arcopsis solida Small arc Lithophaga attenuata rogersi Boring mussel Modiolus capax Fat horse mussel Modiolus pseudotulipusk Mussel Pinna rugosa Pen, center groove Atrina tuberculosa Pen shell Isognomon recognitus Western tree oyster Ostrea columbiensis Columbian oyster Ostrea i ridescens Brown metalic Pecten vogdesi Scallop Argopecten aequisulcatus Cali co seal lop Lyropecten subnodosus Lion's paw scallop Leptopectin monotimeris Kelp scallop Amonia peruviana Pearly ji ngle Cardita affinis Cardita Diplodonta subquadrata Felaniella sericata Chama pellucida Chama 71 Table VI. Continued - -MARINE INVERTEBRATES, CONTINUED Trachycardiium panamerise Papyridea aspera Laevicardium elatum Laevicardium substriatum Tri goniocardi a biangula Trigoniocardia granifera Ventricolari a fordi i Tivela stultorum Ami anti s callosa Dosina dunkeri Dosina ponderosa Megapitaria squalida Chi one californiensis Chione undatel1 a Chione fructifraga Chione gnidia Protothaca staminea Protothaca grata Perti col a californiensis Mactra californica Spisula hemphi1li Raeta undulata Tel 1i na bodegensi s Tel 1ina simulans Tellina ochracea Florimetis obesa Flori meti s cognata Macoma i dentata Macoma secta Tel 1idora burneti Donax californicus Donax gouldii Donax punctatostriatus Heterodonax pacificus Sangui nolari a te l l i noi des Sangui nolari a nutal1i i Semele desi ca Semele f1avescens Tagelus californianus Tagelus subteres Solen rosaceus Hiatel1 a arctica Periploma planiusculum Thracia curta Cryptomya cali forni ca Zi rfaea pi 1sbryi Dentalium sp. . . .conti nued Cockl e Thin cockle Smooth cockle Egg cockle Pismo clam White ami anti s Dosi na Heavy dosina Grey megapitaria California chione Wavy chione Chione Fri1 led chione Common littleneck Li ttleneck Nesting clam California mactra Dish clam Thin, ridged Bodega tel li n Red tel l in Sulphur te l l i n Yellow interior Indented macoma Bent nose macoma Dorsal edge fr i l led Wedge clam Bean clam Donax Heterodonax Purple clam Clipped semele Yellow semele Jackni fe Small jacknife Rosy razor Nesting clam ? Spoon clam Short thracia Piddock Toothed shel1 72 . . .conti nued Table VI. Continued - -MARINE INVERTEBRATES, CONTINUED Panulirus interruptus - California spiny lobster Alpheus sp. - Snapping shrimp Hippolyte sp. - Shrimp MARINE MAMMALS T u r s i o p s t r u n c a t u s Zalophus californianus Bottlenose dolphin California sea lion 73 three major classes: machinery, propeller, and hydrodynamic noise. Due to the variation in hull construction and engine types, each vessel and/or skiff had its own sound signature, thus only representative samples are shown. Noise from large vessels was broadband, ranging from 125 Hz to 20 kHz, with most energy concentrated below 10 kHz. Tonal peaks were typically observed below 1 kHz. Vessel noise was characterized by a continuous spectrum containing superimposed line components. Received levels of large vessels underway averaged 89 to 105 dB re 1 yuPa (Fig. 18). Received levels from vessels at anchor were 10 to 20 dB below those from the same vessel underway (pump and generator noise prevalent only when vessels at anchor). Fourteen skiffs (representing inflatables, wooden- and aluminum-hull craft) powered by five different models of outboard engines were recorded. Spectrum profiles were obtained for each model of engine. Overall noise levels were similar among models, but variations were prevalent in the sound signature of each engine. Noise levels and emphasized frequency bands also varied with the rpm of the engine. Source levels for a 20-hp Mercury outboard engine operating at idle, quarter throttle, half throttle, and three-quarter throttle are shown in Figs. 19a-d. The overall noise contributed by the outboard engine at idle is generally lower than that of the engine in gear. Estimated levels at idle occurred at frequencies below 1 kHz, with a noise level range of 92-117 dB re I uPa at 1 yd. Ranges of source levels (re 1 yuPa) for speeds other than idle were quarter speed, 104-124 dB; half speed, 108-126 dB; and three-quarter speed, 108-134 dB. The levels reported for the three-quarter speed represent a 3-sec average sound pressure output. All other plots are 4-sec average sound pressure levels. 74 Figure 18. Spectra of Typical Large Vessel (25-35 m in length) Noise Encountered in Laguna San Ignacio, Mexico. 7-5 ao TOC ON VERT FROM LEVELS RELATIVE TO 1(JPi TO LEVELS RELATIVE TO lyBi- SUBTRACT lOOdb 1/3 O C T A V E - B A N D C E N T E R F R E Q U E N C Y IN H E R T Z 76 Figure 19. Spectra of a 20-hp Mercury Outboard Engine (a - b) Operating at a) idle, and b) quarter throttle. 7:H 1/3-OCTAVE BANDWIDTH CORRECTION TO SPECTRUM LEVEL (FOR CONTINUOUS SPECTRA ONLY) 1/3-OCTAVE-BAND CENTER FREQUENCY IN HERTZ A 1/3-OCTAVE BANDWIDTH CORRECTION TO SPECTRUM LEVEL (FOR CONTINUOUS SPECTRA ONLY) 1/3-OCTAVE-BAND CENTER FREQUENCY IN HERTZ B 78 Figure 19. Spectra of a 20-hp Mercury Outboard Engine (c - d) Operating at c) half throttle, and d) three quarter throttle. 7(9 1/3-OCTAVE BANDWIDTH CORRECTION TO SPECTRUM LEVEL (FOR CONTINUOUS SPECTRA ONLY) s s s s ! 1/3-OCTAVE BANDWIDTH CORRECTION TO SPECTRUM LEVEL (FOR CONTINUOUS SPECTRA ONLY) "SO [ j 1 I 1-1 -1 I I I I I I 1 i t ' \ "j i J O 1 1 1 1 1 1 I " n i " " j [~T | 1 j 1 i~I 11 ~l I X ^ ' T T T ' •L 1 I . I I I I uo 1—|—i..-1... i -j J j I j T T T ' T ' "(' I I t F \ 1 ~i I 1~ • <o \-\- [ f tlv.i. !1" 1" 1 1 irT~"T~'T r 1 —t— 1 | I I I j 1 j " IOO | I I I I I t 1 j 1 | I 1 1 I [ 1 I I 1 "1 " T " I ' i I I * ° I j T \ ~f~~f~ | | | j | | j 1 —j—}• f "f""t—r—r~f"] " i 1 f { " t^7" f' \ m \ '-f— {• | j | ) | [—}• » t j" { j j 1 j I } 11 1 } | ' r "j ""j • \ j- 1 | I | -1—j- j I j - j {• t i l l I—r *° 1 Tnrml f r i r rfm i i i * I ' l ' l l l l i t l l i l l l i l l J l . i l t""T T 1 f 1 0 ' D I O O O O O O O"'O 1 O O O O L - 1 I 1 I 1 I L 1/3-OCTAVE-BAND CENTER FREQUENCY IN HERTZ 80 St. Lawrence Island (Alaska) Recordings of ambient noise were made at twelve locations in July off St. Lawrence Island, five on the west side and seven off Southeast Cape (Fig. 20). The selection of these locations was based on the previous occupation in these waters by gray whales. The levels of broadband ambient noise (dB re 1 juPa) at Stations 1 and 2 averaged 62 dB with most energy concentrated below 2 kHz (Fig. 21). An inspection of each station profile suggested l i t t le change above 10 kHz; therefore, the remaining profiles were analyzed between 0-10 kHz. The support vessel operated at a distance of 1.5 miles (speed = 3 knots). As a result, Station 3 (Fig. 22) was dominated by ship noise. Received levels ranged from 95-115 dB re 1 uPa at 20 Hz to 10 kHz. Stations 4 and 5 were located 0.5 miles and 1.5 miles off Gambell, respectively. Surf noise (one-meter surf break) dominated these two ambient stations (Fig. 23). A gradual increase in energy was noted with increasing frequencies. Bottom sediment consisted of gravel on the western side of the Island and beaches were steeply inclined. The ambient levels of Stations 4 and 5 were considerably higher than others due to surf noise (with the exception of Station 3, which was influenced by ship noise) and spectra ranged from 87-95 dB re 1 uPa. Recordings were made in water depths of 6.2-25.0 m. Stations 6 through 12 were located off Southeast Cape. Although some variations were observed among stations, these ambient sound levels were 52-66 dB re 1 uPa (Fig. 24). Frequency profiles were essentially f lat . Recordings were made in approximately 12.5 m (40 ft) and a coarse sand sediment dominated the bottom substrate. The density and species composition of sound-producing marine l i fe 81 Figure 20. Locations of Ambient Noise Stations off St. Lawrence Island, AIaska. S I B E R I A - | 6 60 i N 174°|W oo 83 Figure 2 1 . Average Levels (dB re 1 uPa) of Ambient Noise Stati 1 and 2 , St. Lawrence Island, Alaska. ,00 Frequency (kHz) 85 Figure 22. Average Level (dB re 1 /jPa) of Ambient Noise at,Station 3, St. Lawrence Island, Alaska (influenced by ship noise). Frequency (kHz) 87 Figure 23. Average Levels (dB re 1 juPa) of Ambient Noise at Stations 4 and 5, St. Lawrence Island, Alaska (influenced by surf noi se). 89 Figure 24. Average Levels (dB re 1 /jPa) of Ambient Noise at Stations 6-12 (Southeast Cape), St. Lawrence Island, Alaska. i 1 1 1 1 1 1 1 r 40h 0 J 1 I I I I I I L 5 Frequency (kHz) 10 91 were significantly reduced in the Bering Sea as compared to Baja. Species prevalent in the area were identified by scientists from LGL Limited and are reported in Thomson et a l . , 1986. In addition to the NOAA vessel, other non-biological sources were encountered and recorded. These included noise produced by five small launches powered by outboard engines. Each outboard engine exhibited its own sound signature and thus only one representative sample will be shown. The received levels ranged from 90-105 dB re 1 ;jPa and were broadband in frequency. Figure 25 depicts a sound spectrum of a 7-m Boston Whaler with two 90-hp Mercury outboard engines, operating at a distance of 1 mile at 1/4 throttle. Washington State Waters Ambient noise in the test area off La Push, Washington, was highly influenced by surf noise (Fig. 26). A 2.0-m surf break was observed during the recording sessions. The bottom contour sloped gently toward the beach with an associated sand substrate. Levels ranged from 78 to 89 dB re 1 yuPa, with a concentration of energy observed at 81 dB. Limited recording time was spent collecting ambient noise in this area (less than 5 hours). Varying weather conditions wi l l , of course, produce different levels and patterns of noise. No attempt was made to identify the biological organims responsible for ambient noise and/or collect information on depth, temperature and salinity in Washington State waters. 92 Figure 25. Noise Spectra of a 7-m Boston Whaler Operating off Southeast Cape, St. Lawrence Island, Alaska. Level (dB re l^uPa) 94 Figure 26. Spectra of Ambient Noise off La Push, Washington (coastal waters). 96 Comparisons of Ambient Noise Throughout Range Noise spectra among the three major areas of the gray whales' range (breeding, feeding and migratory grounds) showed marked differences in emphasized frequencies and levels (Table VII). These various profiles resulted from differences in the abundance and composition of sound-producing marine l i fe, oceanic conditions, and the amount of man's activities within the area. Recordings made in Laguna San Ignacio resulted in the highest levels of background noise (94-110 dB re 1 /iPa) documented during these studies. Frequencies between 2-5 kHz were emphasized in these spectra due to the biological contribution of noise. Migratory values of biological ambient noise resulted in intermediate values with respect to sound level (87-99 dB re 1 juPa); however, frequencies emphasized (2-6.3 kHz) were similar to those obtained in Mexico (2-5 kHz). In the Bering Sea, the biological contribution to overall background noise was negli gible. During this investigation, ambient noise measurements were generally made when Beaufort 3 or less conditions prevailed. This method, of course, precluded the collections of precise on-site measurements with varying sea states. Predicted patterns were thus based on the extensive work by other authors. Of those articles reviewed, Knudsen et a l . (1948) and Wenz (1962) provided what I believe to be the most comprehensive accounts with respect to ocean ambient noise profiles based on different physical events. In comparing the data collected during these investigations with this past literature, similar profiles were obtained for sea states 3 and below. However, due to the contribution from biological sources to the overall ambient noise in Mexico, the southern values reflect a somewhat higher 97 Table VII. Comparisons of Ambient Noise Profiles Throughout the Range of the Gray Whal e.  SOUTHERN RANGE Received Levels Frequencies (dB re 1 yuPa) Emphasized Source Reference Stations 1-32 94-110 2-5 kHz Biological This study Man-Made 89-105 < 2.0 kHz Large vessels This study Man-Made 92-134 < 2.0 kHz Skiffs This study NORTHERN RANGE Stations 6-12 52-66 < 1.0 kHz Sea noise This study Man-Made 95-115 < 2.0 kHz Ship noise This study Stations 1-2 x = 62 < 2.5 kHz Sea noise This study Stations 4-5 87-95 < 2.5 kHz Surf noise This study Man-Made 90-105 < 2.0 kHz Boston Whaler This study Natural Ambient x = 65 < 2.0 kHz Sea noise Moore et a (1985) MIGRATION Washington State 78-89 < 2.5 kHz Surf noise This study Central Cali fornia 85 < 2.0 kHz Di stant shi ppi ng Mai me et a (1984) Centra 1 California 87-99 2.0--6.3 kHz Biological Mai me et a (1984) Central Califoria 90 6.3-16 kHz Sea noise Ma 1 me et a 1 (1984) 98 level than expected in the frequency range of 2-5 kHz. Conversely, the Bering Sea and C a l i f o r n i a ambient noise data c losely f i t the model. Introduced man-made sources t yp i ca l l y were broadband, emphasized frequencies less than 2 kHz, had higher associated l eve l s , and exhibi ted indiv idual sound s ignatures. Simi lar overa l l p ro f i l es for man-made sources were documented independent of the area being invest igated. Propagation experimentation and factors potent ia l ly a f fec t ing a t r ave l l i ng wave ( i . e . , temperature, s a l i n i t y and depth) were also invest igated during these s tud ies. Sound propagation in Laguna San Ignacio did not follow the cy l i nd r i ca l spreading law (10 log R ) . Sound transmission was enhanced in the southerly d i rec t ion (lower lagoon) and was reduced in the middle lagoon (northerly d i rect ion) due to the extensive sand bars causing shadow zones. As expected, major di f ferences were noted in water temperatures and depths between the Alaska and Mexico. S a l i n i t y values, however, were s imi la r between the two areas. Detai led resul ts of these invest igat ions are reported in Appendix A. 99 RELATIONSHIP BETWEEN GRAY WHALE SIGNALS AND AMBIENT NOISE A composite spectrograph representing a gray whale SI series, naturally occurring biological ambient noise and outboard engine noise, is shown in Fig. 27. Skiff noise predominates at frequencies below 500 Hz, with biological noise dominating the frequencies above 2 kHz. The major energy in the SI signal is below 1 kHz. All gray whale sounds analyzed from this area emphasized the frequency bands below the main concentration of energy of the biological ambient noise. Similar findings resulted when comparing the emphasized frequencies of gray whale signals and the main concentration of biological ambient noise off California. No comparisons were made in the Bering Sea since the biological contribution to the overall ambient noise was negligible. When emphasized frequencies of gray whale signals were compared to the sea noise models produced by Wenz (1962) and Knudsen et a l . (1948), and from my own field collections, an interesting pattern emerged. Sea noise was broadband; however, frequencies between 200 and 2000 Hz did not typically contain the major energy of the profile. Wenz (1962) noise profiles are based on shallow water measurements, 7-10 kt wind and moderate shipping. Increasing wind conditions, with a corresponding increase in sea state, would increase the overall level of the profile; however, l i t t le change would occur in the frequencies emphasized. A comparison of man-made noise and gray whale signals revealed a different pattern than those obtained during the comparisons between gray whale signals and naturally occurring sound sources. An inspection of frequencies of outboard engine noise and gray whale signals produced in Laguna San Ignacio indicated that similar acoustical channels were occupied 100 Figure 27. Composite Spectrograph of a Gray Whale Pulse Series (SI), Outboard Engine Noise and Biological Ambient Noise in Laguna San Ignacio, Mexico. 4000 N I X Biological ambient noise 102 (Fig. 28). Based upon the profiles obtained from other man-made sources (e.g., o i l , ship noise), and a review of profiles from other sources throughout the gray whale's range, overlapping frequencies are predicted between gray whale signals and these anthropogenic noise sources. In addition to comparing the frequencies emphasized by different sources, ambient levels were compared to the received levels of gray whale sounds. On the southern range, gray whales typically produce high level signals. The levels of ambient noise in Laguna San Ignacio also reflect high levels. Off California, ambient levels were intermediate in value. In the Bering Sea, low level ambient conditions were documented. Similar results were obtained for signals produced by these whales for this area. Frequency and level comparisons are graphically represented in Fig. 29. In addition to comparing the sounds of gray whales to ambient conditions in Laguna San Ignacio, I analyzed 50 signals produced by the bottlenose dolphin (Tursiops truncatus), the only other cetacean that occupies this lagoon area. This was done to determine the relationship of the dolphin signals to that of the high levels of natural occurring background noise (Fig. 30). In all cases, dolphin whistles occurred at frequencies above 7 kHz. The short duration whistles were frequency modulated, with the lowest frequency observed to rise when it encountered the ceiling of the biological ambient noise. Because dolphin signals were not treated in the same depth as those of the gray whale, these data should be interpreted with some caution. 103 Figure 28. Spectrograph of a Gray Whale Pulse Series (SI) and Outboard Engi ne Noi se. Frequency (Hz) frOT 105 Figure 29. Relationships of Gray Whale Sound Frequencies and Levels to Ambient Noise. 110 100H 9 0 1 80 WSQIISMfllSBIBHBBBSBii • • • • MBINH9H8JSIB9BIBSrBKf • • • • j i ! ' ! ' ! ' ! ' ! ' ! ' ! ' ! ' ! ' ! ' ! 1 ! 1 ! ' ! 1111111 t i t 111V ;:;1 -, 11,11111.1.1.! i •::: • Man-made noise P ^. Gray whale Ip^TfT^W Natural ambient ti^ ikkkkklUj Southern Range, Mexico •t Natural ambient I......mi Washington State Natural ambient Central California IsasEaaaa Natural ambient a i t c n l Northern Range, Bering Sea o 6S 70-60-50 H "T 2 T" 6 10 Frequency (kHz) 107 Figure 30. Spectrograph of a Bottlenose Dolphin Whistle and Biological Ambient Noise in Laguna San Ignacio, Mexico. 108 109 DISCUSSION Gray whales were more soniferous on their southern range than expected from previously published accounts. Recent advances in technology and the amount of time spent in the field compared to previous investigators may account for this. Table III indicates that gray whales produce at least seven distinct sound types. The S7 call type was not recorded during the current investigations; however, there appears to be sufficient information (Fish et a l . , 1974) to warrant its categorization. Although variations did exist within sound categories, each category was sufficiently distinct to be classified independently. Variations in these call types may reflect individual differences or responses to different behavioral situations. In addressing the overall acoustical activity of gray whales, hourly changes in sound periodicity and types of signals emitted by whales were inspected and compared. In Laguna San Ignacio, no obvious differences in periodicity or types of signals were detected over a 24-hr period. These results coincide with the observations of surface behavior of gray whales in this area presented by Swartz and Cummings (1978). These authors reported similar surface behavioral activities by gray whales over a 24-hr period. Although not significantly different, the slight decrease in the number of sounds produced per hour/animal in March did coincide with the departure of single, adult animals. These single whales represent the mating component of whales in Laguna San Ignacio and typically depart 30-45 days prior to the departure of the cow/calf pairs (Jones and Swartz, 1984). With the exception of the S2 signal, all other sound types produced by the gray whale on the southern range have been documented on other areas of its range. Moore and Ljungblad (1984) recently published on gray whale sounds collected in the Bering Sea. Two hours of sounds of sufficient 110 level to permit analysis were reported on by these authors. Their results indicated three distinct sound types termed NI, N3, and N4, following the classification scheme designed for this study. The NI signal dominated the vocal repertoire of feeding gray whales. A comparison of the results of these two northern studies (this study and Moore and Ljungblad, 1984) are similar and supportive. The only other available information on the acoustical behavior of feeding gray whales is noted in the Soviet literature (Bogoslovskaya, 1982). This author noted that single whales were usually silent; however, low-frequency sounds were occasionally exchanged between individuals. An adequate description of these sounds and the effort to obtain the recordings was not given in this report. Malme et al (1986) reported hearing no sounds attributable to gray whales in recent studies on feeding gray whales off St. Lawrence Island. Since the initiation of this project, additional attempts have been made by various researchers to collect sounds from migrating gray whales along the coast of California. Recordings made during the southbound migration off central California (30-day effort in January 1983 and 30 days in January 1984) resulted in no sounds that could be attributed to migrating whales If Malme, pers. comm.) . Three attempts have been made to collect sounds from northbound migrants, with the effort centered once again off the 2/ California coastline. Malme (pers. comm.) reported hearing no gray whale sounds during a 30-day investigation off central California in March 1983. However, during work in Monterey Bay in March of 1983, a series of sounds 2/ C. Malme, Bolt, Beranek and Newman, Cambridge, Massachusetts. Pers. Comm. to Dahlheim, July 1984. I l l 2/ were collected from gray whales (B. Wlirsig ) . This tape was reviewed by the author and was found to contain Ml signals. Detailed information regarding the number and/or behavioral aspects of the whales in the Monterey Bay region was not given. In April 1986, a two-week investigation took place off Piedras Blancas, California, during the northbound migration of cow/calf pairs. Sounds, reportedly produced by gray whales, were collected 1 / and are in the process of being analyzed (N. Cain, pers. comm.) . The five hours of recordings made in the presence of juvenile whales and/or lone whales in Washington State waters during this study resulted in no sounds produced by the whales. Based on the accounts with sufficient quantitative information, the most prevalent sound produced by the gray whale in Laguna San Ignacio, Mexico, and in the Bering Sea appears to be the SI (NI) signal. The SI call was also the most frequently heard signal recorded from the captive whale, "Gigi" (Fish et a l . , 1974). Cummings et a l . (1968) reported that the M3 call dominated the vocal repertoire of southbound gray whales off southern California. Unfortunately, either due to the absence of gray whale signals during recording sessions and/or the lack of detailed information, i t is impossible to conclude what type of sounds dominate the repertoire during the northbound phase of this whale's migration. In comparing the results on the periodicity of sound production among the three major areas, which reflect three distinct behavioral modes of the animals (feeding, migrating, breeding/calving), the following preliminary 3/ Wlirsig, Moss Landing Marine Laboratory, Moss Landing, California Pers. Comm. to Dahlheim, March 1983. 4/ N. Cain, UCSC, Santa Cruz, California. Pers. Comm. to Dahlheim, May, 1986. 112 interpretation was made. When the northern range is compared with the other two areas, a significant reduction in acoustical signalling occurs on the feeding grounds. During the southbound migration, Cummings et a l . (1968) reported a total of 231 sounds produced over a 24-hour period, which represented 218 whales. This would give a value of 0.04 sounds/hr/whale. This migratory value on sound production suggests a slight increase in sound activity during the southbound migration off southern California, as compared to the Bering Sea (0.006 sounds/hour/whale). The term "sounds/hr/whale" assumes that each whale within a defined area contributes equally to the sound environment. Although it may be erroneous to make this assumption, this term does allow an estimate or average of sound production to be made incorporating whale density. From the available data, it appears that sound production peaks on the whales' mating/calving grounds (rates of 0.33 for February and 0.25 for March) and possibly just prior to engaging in these activities (Southern California rates = 0.04 sounds/hr/whale). The annual concentrations of gray whales found within the coastal lagoons and adjacent waters of Mexico represent the peak of social activities for this species. Interactions among whales occur on a regular basis. Conversely, on the northern grounds, few social interactions occur. Whales are typically found either singly or in pairs (Bogoslovskaya et a l . , 1981, 1982, personal observations by the author). During migration, the average pod size is 2.3 (Reilly et a l . , 1983); however, pods of up to 15 animals have been observed during census work conducted by the National Marine Fisheries Service (unpub. data). On their breeding locations, acoustical signalling may be necessary to ensure successful matings or to maintain contact between mothers and their newborn calves. Conversely, during feeding and migrating, whales may not 113 need to interact and thus signalling may not be as important. To obtain a greater understanding of the acoustical behavior of gray whales, the literature pertaining to sound production by other cetaceans was reviewed. Odontocete (toothed) whales are well known for their extensive variety of cal ls , covering a wide range of frequencies (Evans, 1967; Fish and Winn, 1969; Watkins and Schevill, 1977; Watkins, 1977; Ford and Fisher, 1982; Dahlheim and Awbrey, 1982). A communicative as well as echolocative function has been ascribed and documented by experimentation for these whales (Bastian et a l . , 1966; Au, 1980; Au et a l . , 1985). The sound repertoires of the mysticete whales usually emphasize lower frequencies, have less variability, and cover a narrower range of frequencies when compared to the sounds of odontocetes. To date, echolocation capabilities have not been documented in the baleen whales. The available data acquired on baleen whale sound production do suggest that these whales can produce a variety of calls (Winn and Perkins, 1976). The elaborate calls of the humpback whales (Megaptera novaeangliae) have been well documented (Payne and McVay, 1971; Winn et. a l . , 1971). In addition, humpback whales have been shown to have marked seasonal differences in calling rates and types of signals produced between their breeding and feeding locations. Reports are also available on the sound behavior of other members of the family Balaenopteridae: minke whales, Balaenoptera acutorostrata (Beamish and Mitchell, 1973; Winn and Perkins, 1976); blue whales, Balaenoptera musculus (Cummings and Thompson, 1971); fin whales, Balaenoptera physalus (Schevill and Watkins, 1962; Thompson and Cummings, 1969; Watkins, 1981); and the Bryde's whale, Balaenoptera edeni (Thompson and Cummings, 1969). With the exception of the humpback whale, adequate data do not exist for valid comparisons between the sound activities of these whales for their 114 breeding/feeding locations. Whales of the family Balaenidae, the southern right whale (Balaena glacialis australis), northern right whale (Balaena  glacialis), and the bowhead whale (Balaena mysticetus), also are capable of producing a variety of phonations (Schevill and Watkins, 1962; Cummings et a l . , 1972; Ljungblad et a l . , 1980, 1982; Clark, 1982). Marked seasonal differences have also been demonstrated in their cal ls. When gray whale sounds are compared to those sounds produced by other mysticete whales, the results imply that gray whales, representing the sole member of the family Eschrichtiidae, may have more of a limited repertoire (7 call types) and frequency range in which to operate (40 Hz to 4 kHz). The predominant SI signal produced by the gray whale differs greatly both aurally and structurally when compared to those sounds produced by other baleen whales. Although some variation was shown to exist within sound types between mating and feeding areas, marked seasonal differences in call types were not apparent in these preliminary investigations. One fact that does seem consistent among the suborder Mysticeti is that calling rates increase when these species occupy their breeding grounds, which suggests a dependence on acoustical signalling during social activities and/or the rearing of their young. The high levels of biological ambient noise in Laguna San Ignacio are typical of coastal, shallow, warm-water lagoon environments (Johnson, 1943). This biological background noise is broadband in frequency, with frequency bands of 2-5 kHz emphasized. One-third octave band levels ranged from 94-110 dB re 1 juPa. The diversity and relative abundance of the invertebrates and fish (Table VI) occupying Laguna San Ignacio, are apparently the major contributors responsible for this high-level background noise. Sea noise, of course, also contributes. 115 Vessel and skiff traffic were the only sources of man-made noise encountered during these southern studies. These levels were higher than those reported for the naturally occurring ambient noise and emphasized different frequency bands (less than 2 kHz). The received levels of these sources reported do not truly represent the source levels of these vessels since the acoustical waves have undergone transmission losses and attenuation. Source levels would, of course, be greater that those received at some di stance. When compared to the ambient noise levels on the southern range, relatively low levels of noise were found in the waters off St. Lawrence Island (52-95 dB re 1 uPa). The noise spectra obtained for this northern area differed in their profiles (when compared to Mexico), emphasizing different frequency bands and levels. The contribution of noise resulting from the biological component on the northern range was negligible. Received levels and frequency profiles from man-made sources were similar when the two areas (northern vs southern range) were compared. Again, these man-made sources emphasized lower frequencies (less than 2 kHz). The limited recordings accomplished on the migration route off Washington State suggest an intermediate value (78-89 dB re 1 /uPa) of ambient noise occurring in this area. In recent studies off central California, Malme et a l . (1984) reported ambient noise at 85 dB re 1 /jPa at frequencies below 2 kHz. Between 2 and 6.3 kHz, levels increased from 87-99 dB and then gradually dropped off to 90 dB re 1 yuPa at 16 kHz. Malme et a l . (1984) reported that snapping shrimp were responsible for the high-level background noise obtained between 2 and 8 kHz, and that lower frequencies were influenced by ship traff ic. It is unclear how these authors 116 concluded that snapping shrimp were the source of high-level noise, since no collections were made. Their results are similar to my findings in Laguna San Ignacio; however, the levels off central California are lower. Again, the Washington and Californian values reflect an intermediate value between the Bering Sea and Mexico. Many factors influence sound transmission in the sea; however, temperature usually has the most profound effect (McLellan, 1965). Sound propagation in the two major study areas (Alaska and Mexico) would follow very different pathways and be considerably different based on these current preliminary investigations (Appendix A). One would expect temperature to be responsible for the reflections/refraction of sound sources in the Bering Sea, whereas in Mexico, the shallow water environment (sea surface and bottom) may be the primary cause of the numerous reflections observed. Previous reports suggest that sound propagation pathways in shallow-water environments typically follow the cylindrical spreading law (10 log R). The experimentation in Laguna San Ignacio has shown that this formula cannot be used to estimate the level reaching the whales during playback experiments (Appendix A). This is apparent in both the immediate study area as well as for areas north and south of the primary area. Sound propagation in the southerly direction (into lower lagoon) is apparently enhanced by the numerous reflections and refractions caused by the sea surface and bottom. Conversely, in the middle lagoon, sound sources were barely audible. The prevalent sand bars in this area caused sound shadow zones. These shadow zones were independent of height of tide. The acoustical characterization of ambient noise made during this 117 study represents a small window of time. Since gray whales only occupy Laguna San Ignacio between January and April (Jones and Swartz, 1984), the ambient values obtained more than likely represent a valid representation of noise during these months. A change in species composition (invertebrates or fish) would produce a different profile for this area. Major changes in bottom topography caused by storms/hurricanes would alter propagation pathways. The ambient noise profiles collected in the Bering Sea and off the coast of Washington represent extremely small time frames in which to predict ambient patterns. However, the measurements reported here are valuable since they do represent accurate levels obtained during selected time frames when whales were present. Moore et a l . (1985), during recent studies in the Bering Sea, reported ambient profiles similar to those documented during the present studies. Future investigations of ambient noise on the migration route and in the Bering Sea should replicate the in-depth work accomplished in Mexico. The dynamic nature of ambient noise within and among areas is well documented in our world's oceans (Urick, 1967 ; Clay and Medwin, 1977). The conclusions drawn from this investigation document major differences in the gray whale's acoustical environments throughout its range. As gray whales proceed south during their annual migration, ambient levels increase and differences occur in the frequencies emphasized. In addition, man's activities increase as well. Gray whales are thus exposed to a broad spectrum of noise in their preferred coastal habitat. An inspection of the sounds produced by this whale also documents changes in call structure dependent upon the area being occupied. In Laguna San Ignacio, high levels of natural ambient noise prevail throughout the gray whale season (January to March), emphasizing the 118 frequency bands of 2-5 kHz. On the northern range, these natural ambient p ro f i l es are considerably lower and emphasize d i f ferent frequency bands. When the sound charac ter is t i cs of gray whale signals are compared between these two areas, the southern acoust ical behavior of the gray whale can be summarized as fo l lows: 1) increased overal l rate of c a l l i n g , 2) high values of received l eve l s , 3) c a l l s emphasize a narrower and overal l lower frequency range, 4) the amount of frequency modulation within signals i s increased, 5) fewer number of pulses are produced per ser ies at a low repet i t ion ra te , and 6) c a l l s exhib i t a longer durat ion. Conversely, when northern signals are compared to southern s igna ls , these northern ca l l s exh ib i t : 1) a decrease in the overa l l rate of c a l l i n g , 2) a lower received l e v e l , 3) wider and higher emphasized frequency bands, 4) less frequency modulation within s i gna l s , 5) a higher number of pulses produced per ser ies with a corresponding higher repet i t ion rate, and 6) shorter ca l l durat ion. When p ro f i l es of gray whale sounds are compared to the spectra obtained for natural ambient noise condi t ions, an in te res t ing pattern emerges. A l l gray whale signals analyzed e i ther occupied frequency bands below the main concentration of ambient noise (e .g . , Laguna San Ignacio) or occupied d i f ferent frequency bands than those emphasized by the natural ambient condit ions (e .g . , Bering Sea). In add i t ion , while occupying high- level ambient habi ta ts , gray whale ca l l rates increased, the level of sound output was increased, c a l l s were t y p i c a l l y longer in durat ion, and the amount of frequency modulation within the signals was increased. These various ca l l modif icat ions may ensure the maximum transmission and reception of signals when high- levels of background noise characterize an area. When re la t i ve l y low-level ambient habitats are occupied, less demands are 119 placed on a calling species. This appears to be the case in the Bering Sea where calls of gray whales emphasized an overall broader and higher frequency range, call rates decreased, received levels of sounds were lower, and shorter call duration and less frequency modulation occurred in their signals. In comparing the acoustical behavior of a species between areas, the behavior of that animal cannot be ignored when evaluating differences in call structure. Although the behavior of this whale varies greatly between the two extremes of its range, i t is reasonable to assume that most signalling (independent of area) represents an important function for this species. Therefore a gray whale signal produced either on the feeding and/or breeding grounds should be so structured as to optimize its output to remain effective. It seems logical to postulate that cetacean sounds would have evolved to minimize the detrimental effect of noise on signalling. Natural ambient noise profiles in our oceans have remained relatively constant and somewhat fixed through time. With the exception of a catastrophic event, major changes in the overall spectra for an area are more than likely slow to occur. Noise due to the interference from other sounds and noise inherent in the transmission process delimit an acoustic niche for each calling species (Simkin and Il'ichen, 1965). The results of comparisons made between gray whale signals throughout its range and varying natural ambient conditions indicate that these signals occupy different acoustical channels than those dominated by the naturally occurring ambient conditions. This relationship suggests that a varying acoustical niche may "be utilized by the gray whale on different parts of its range, thus ensuring the maximum transmission and reception of signals. Man-made sources, when compared to natural noise, typically emphasize 120 lower frequencies (less than 2 kHz), have a corresponding high level of output at these low frequencies, and represent transient and/or temporary sound sources. When gray whale signals were compared to these man-made noise sources, it was found that these sources typically occupied the same frequency bands as gray whale signals. These results imply a high potential of man-made noise sources to interfere with and/or mask the signals produced by gray whales. To date, no quantitative attempts have been made to document the acoustical responses of whales to noise produced by man in the natural environment. Two studies, however, have been recently accomplished in captive environments with the bottlenose dolphin (Tursiops truncatus) and the beluga whale (Delphinapterus leucas), Au et a l . , 1974 and Au et a l . , 1985, respectively. These studies demonstrated that these two species of toothed whales were capable of shifting frequencies to circumvent noise in their acoustical channels. In Laguna San Ignacio, during these investigations, preliminary comparisons between sounds produced by T_. truncatus and ambient noise conditions suggested that these dolphin signals were typically higher than those emphasized by the ambient noise. These results support the work conducted in captivity by Au et a l . , 1974. To determine if gray whales can actively structure their calls in ways to maximize the transmission and reception of their signals in the presence of man-made noise and provide more insight into the relationship between gray whale sounds and natural noise, a quantitative study addressing the acoustical responses of these whales to increased levels of noise was warranted. Chapters 2 and 3 investigate the acoustical capabilities and possible strategies employed by these whales when faced with dynamic noise situations, including the potential masking of whale signals by the calls of conspecifics. 121 In addition, these Chapters inspect the other behavioral options available to this species when sound sources directly infiltrate its acoustical channels. 122 C H A P T E R 2 RESPONSES OF GRAY WHALES TO INCREASED LEVELS OF TEMPORARY (SHORT-TERM) NOISE 123 INTRODUCTION Studies on the effects of noise have been completed on a wide variety of terrestrial species (Saunders and Dooling, 1974; Lim and Melnick, 1975; Fletcher and Busnel, 1978; Holt, 1978; Nawrot, Cook and Hamm, 1978; Nielson, Burnham and Talley, 1978; Anonymous, 1979; Kluczek, 1979; Zielinski, 1979; Smiley andWilbanks, 1980). In most cases, noise has been shown to have a negative effect on the species being investigated. Physical adverse effects such as ear damage, and non-auditory physical effects such as physiological stress involving hormone responses leading to lowering of disease resistance, increased vulnerability to environmental stress, and hormone imbalances which may adversely affect reproduction, have been demonstrated. In addition, noise has the added potential of interfering with and/or masking the acoustical signals of animals. A reduction in the acoustic transmission and/or its reception can adversely affect the reproductive cycle or even the survival of a given species that is dependent on such a sensory process. Calls that have survival functions (e.g., providing appropriate information relative to food, competitors, potential mates and predators) must overcome or circumvent this noise to remain effective. In terrestrial studies, many features of animal signals have been shown to minimize the detrimental effect of noise on communication (Konishi, 1970). Studies addressing orthopterans (Otte, 1977); birds (Thorpe, 1961; Armstrong, 1963; Marler, 1977; Smith, 1977); and primates (Gautier and Gautier, 1977; Marler and Tenaza, 1977; Oppenheimer, 1977; McGeorge, 1979) have been conducted. A collective review of these studies suggests that these various taxa do minimize the effects of noise on their signalling 124 by 1) the re la t i ve t iming of the i r s i gna l s , and 2) the structure of t he i r c a l l s . To date, quant i tat ive studies on wi ld cetaceans and how these aquatic populations acoust ica l ly respond to noise are lack ing. The gray whale, due to i t s preferred coastal habi tat , i s exposed to a dynamic acoust ical environment. As demonstrated i n Chapter 1, gray whales may minimize the detrimental ef fect that natural ambient noise has on the i r s ignals by operating in an acoust ical niche. Operating in an acoust ical niche would represent an adaptive strategy on the part of these whales to overcome or circumvent the constant high levels of natural noise prevalent in the i r environments, thereby ensuring minimum interference and masking problems with the transmission and reception of t he i r own s igna ls . Man-made noise spectra (characterized by low-frequency p ro f i l es and high- level outputs) may, however, d i rec t l y in ter fere with and/or mask the s ignals produced by the gray whale. In add i t ion , when numerous whales are voca l i z ing , some level of acoust ical competition may ex is t among conspec i f i cs . With the l imi ted repertoire of c a l l s avai lab le to the gray whale, i s th is species capable of circumventing increased levels of noise, in s im i la r acoust ical channels, produced by man and/or i t s own conspeci f ics? To determine how th is species uses i t s acoust ical s igna l l i ng most e f f i c i e n t l y , noise levels were a r t i f i c i a l l y increased through a ser ies of playback experiments. Recordings co l lec ted during control and experimental condit ions were compared to determine i f a s ign i f i can t di f ference occurred in the acoust ical behavior of the gray whale in response to playback. Comparisons were also made between simulated vesse l / s k i f f noise (playback experiments) and real noise sources (vessels present in area) to determine the f e a s i b i l i t y of using playback techniques to assess the ef fect of noise on the acoust ical systems of marine mammals. Since gray 125 whales potentially could seek out areas of low-sound energy when subjected to increased levels of noise (thus avoiding the need to modify their acoustical signals), observations were also collected on the observed surface behavior of these whales. This Chapter investigates the effect of increased levels of short-term (temporary, stationary) noise stimuli on the behavior of gray whales in Laguna San Ignacio, Baja California Sur, Mexico. 126 MATERIALS AND METHODS  1983 Experimental Design During playback sessions, the experimental design employed a pre-trial (= control), t r i a l , and post-trial period. The duration of each session was 15 minutes. Sounds were projected back only during trial periods. The playback experiments included the projection of biological as well as non-biological sound sources. During biological sound playback periods, gray whale sounds (previously collected in Laguna San Ignacio) were projected back underwater. The biological playback schedule during this trial period consisted of one minute of playback, then one minute of silence, to prevent confusion during analysis. Thus for a 15-min interval, sounds were projected back for a total of eight minutes. Non-biological sound sources included: outboard engine noise (collected in Laguna San Ignacio, projected for 15 minutes) and test tones (calibration tone on acoustic amplifier, one tone produced every minute). Since these pre-recorded playback tapes already had undergone transmission losses (with the exception of the test tone) it was necessary to obtain a spectrum for each projected tape. Recordings of these playback tapes were made through the bottom-mounted hydrophone system to obtain an accurate spectra for each playback. All sound j stimuli were projected at maximum levels which ensured that whales within the immediate area (within 1/2 mile radius) were exposed to minimum sound levels of at least 130 dB re 1 /jPa (determined by propagation experiments conducted - Appendix A). Playback experiments were terminated when 1) Beaufort conditions created noise interference and 2) noise from unplanned vessel and skiff traffic interfered with the experiments. In addition to active playback periods, gray whale recordings were also 127 collected in the presence and absence of real man-made noise sources in the lagoon (e.g., vessel, skiff noise). The collection of gray whale sounds during the absence of vessel noise served as an additional control. Intra- and inter-experimental controls (e.g., pre-trials and absence of extraneous vessel sources) were compared to experimental conditions (playback trials and presence of vessels). Comparisons were also made between simulated outboard engine noise and real outboard noise to determine if the simulated noise source elicited similar responses by the whales. Post-trial periods were also compared among the various experiments. Although the main focus of this study was to determine if a change in whale phonations would occur in the presence of noise (see following sections for parameters of sounds measured), the potential existed that a gray whale could possibly move out of the intensified sound area and seek an area of lower sound energy. This could negate the need for gray whales to alter their acoustical repertoire. For this reason, the observed surface behavior of the whales also became an important component of this study. To document the observed surface behavior of whales, two observers collected data on whale orientation and dive profiles when the whales passed through the immediate study area (defined by the visual range of observers - Fig. 31). Observers were positioned 100 m from each other at the Rocky Point shore-station and were involved in tracking different individuals. As each whale was located, the observer recorded the time the whale came to the surface and respired. At this time, the orientation (direction of heading) was noted as well. Only well-marked whales were selected to track (readily identifiable due to characteristic scars and/or color patterns), thus avoiding erroneous data on individual behavior. Observers were not aware of when playback, 128 Figure 31. Immediate Study Area off Rocky Point Defined During Tracking Experiments in Laguna San Ignacio, Mexico. 129 0 1 nm 1 I I 130 i f any, took place. Observations were collected during both control and experimental conditions. 1983 Analysis of Gray Whale Sound Behavior Each recording collected was edited and the information transferred to a written f i l e . These edit sheets included: tape identification number, time, type of sound produced and a written log of all pertinent comments (e.g., sea state, tidal conditions, presence of vessels in area, and general behavior of whales). Once completed, the edit sheets were grouped by experiment. Five experimental groupings were defined: 1) Control periods - no intervention of man-made noise into the envi ronment 2) Experiment A - real man-made sound sources prevalent in envi ronment 3) Experiment B - simulation of outboard engine noise 4) Experiment C - simulation of gray whale sounds 5) Experiment D - projection of test tone Within each experiment, data were further grouped for each test period (pre-trial, trial and post-trial). In the case of Experiment A, pre-trials did not always precede tr ia ls , since these experiments were conducted on an opportunistic basis when tour vessels were visiting the lagoon. This pre-trial, t r i a l , post-trial approach allowed acoustical comparisons to be made between and within each experiment. For each experimental grouping, the following measurements were made on gray whale signals: a) total number of sounds produced, b) the types of calls produced, and c) structural changes within cal ls . Variables approximated a normal distribution, thus allowing statistical comparisons using parametric tests following Sokal and Rohlf (1969) and Zar (1974). Statistical tests were conducted with the use 131 of an IBM/XT computer using the Number Cruncher Statistical System (Version 4.2; Jerry L. Hintze). Periodicity Change: The total number of whale sounds per experiment and per test period (pre-trial-post) were calculated by hand-scoring each signal produced (taken from edit sheet). The average number of sounds produced per 15-minute interval was then calculated. The resultant averages were compared for each test period and among experiments to determine if a significant difference occurred in sound periodicity when gray whales were exposed to different sound stimuli. When addressing overall acoustical activity under different control/ experimental conditions, whale density needed to be considered. During the projection of various sound stimuli, the zone of influence of potential acoustical interference exceeded the visual capabilities of the observers (outside immediate study area - Fig. 31). Thus no estimate could be made regarding the number of animals potentially impacted by increased noise levels. In addition, whales outside our visual range undoubtedly were contributing to the overall level of sound activity documented during these studi es. Call Types: To determine if gray whales switched call types in the presence of noise, data on the type of call produced were summarized from each edit sheet for each experimental condition described above. A percentage of a specific call type produced was then calculated, thereby illustrating the overall use of a specific call type for each event. These percentages were compared within and between each experiment. Structural Changes: The following six sound parameters were investigated: overall frequency range of signal (Hz); emphasized frequencies of signal (Hz); received levels (dB re 1 uPa); percentage of signals 132 showing frequency modulation; duration (sec); and number of pulses produced per series. Overall frequency ranges, emphasized frequencies, duration, and number of pulses per series were calculated directly off the sonograms. Repetition rates were calculated based upon the average duration of the call and the average number of pulses produced per series. In the case of frequency modulation, the presence and/or absence of frequency modulation within the signal was noted for each phonation. A percentage was then calculated for the number of calls exhibiting frequency modulation within each experiment. To determine the received level of the cal l , average sound pressure levels were computed using a Nicolet Scientific Corporation FFT Spectrum Analyzer, Model 446. Displays of calculated spectra were made on an x/y plotter and an average received level was calculated for each sound for all experiments. A matrix was established listing the average values calculated for each of the six acoustical variables under varying conditions. Statistical comparisons were made to determine if significant changes occurred in the call structure of gray whale signals between control and experimental conditions. 1983 Analysis of Observed Surface Behavior Data collected on the observed surface behavior of gray whales during different experimental conditions were organized into proper dates and times and by experiment. Data were then screened for obvious errors. Erroneous entries and dive data with less than three entries were dropped from the data base. Each whale tracked and its associated respirations were numbered chronologically. Average dive times per whale, as well as overall average dive time for all whales, were calculated. Data on an individual whale's respiration were then categorized into respective time 133 intervals to obtain a frequency distribution of observed dive times. Respiration rates of cow/calf pairs and single whales were compared to determine if significant differences occurred in their dive profiles. Mean respiration rates were then compared between control and experimental conditions to determine if a significant change in dive duration occurred. For each whale tracked, a line display was made. This allowed a graphic representation of a whale's trackline (= respiration pattern) to be inspected. Environmental data surrounding specific dive times were noted on these line displays. In addition to the analyses on gray whale respiration rates, the orientation (direction of heading) of the whale was also inspected. Each whale's trackline was investigated and the direction of heading for each whale was investigated for each surface blow. A percentage was then calculated indicating the total number of directional changes that occurred within an individual whale's trackline. The number of directional changes per trackline was compared among experiments. 134 RESULTS 1983 Calling Rates A significant increase in calling rates was documented when gray whales were exposed to increased levels of noise (= trial periods). Post-trial values, although lower than trial periods, were s t i l l somewhat inflated when compared to pre-trial control periods. An exception to the above occurred when gray whales were subjected to the test tone. During this trial period, all vocalizations ceased. The post-trial rate obtained was significantly lower than pre-trial values. Overt differences in the calling rates of gray whales were documented dependent upon the sound stimuli present in the environment. The presentation of outboard engine noise resulted in the greatest number of sounds produced with 45.5 + s.d. 11.3 sounds per 15-min interval. Real noise sources in the environment also caused whales to vocalize more (41.9 + s. d. 10.9 calls produced per 15-min interval); however, these rates were not as high as those documented in the presence of projected outboard engine noise (45.5 + s.d. 11.3). Significant differences were not found when calling rates were compared between real sources and outboard engine noise. Both outboard and real noise sources yielded higher calling rates than those produced by gray whales in the presence of their own sounds (23.9 + s.d. 5.9 sounds per 15-min interval). Although rates of calling varied per experiment, ranging from 23.9 to 45.5, all trial values were higher than control values (18.6 calls produced for the interval), with the exception of the test tone when all vocalizations ceased. The post-trial values indicated that baseline and/or control calling rates were reached at different times, once again depending upon the sound stimuli presented to the whales. Post-trial rates were highest for 135 real sources (29.6 + s. d. 3.8), followed by outboard engine noise (24.9 + s.d. 4.3), the gray whale signals (19.8 + s. d. 3.1), control values (18.0 + s. d. 5.2), and finally test tone values (15.2 + s.d. 3.7). Calling rates for each control/experimental condition are listed in Table VIII. Details on intra- and inter-experimental comparisons and associated statistics are included in Appendix B (1983 Calling Rates). 1983 Cal1 Types Table IX summarizes the results of the investigations conducted on call types. During control periods (which includes pre-trials), the SI signal dominated the sound repertoire of the gray whale, representing 83% of all calls produced. Of the remaining call types, S2 signals occurred 5%, S3 signals occurred 7%, and S4 signals occurred 5%. During all trial periods, with the exception of Experiment D when all vocalizations ceased, and post-trial periods, the SI signal continued to dominate the sound repertoire of the gray whale. 136 Table VIII. 1983 Calling Rates of Gray Whales in Laguna San Ignacio, Mexico (total number of sounds produced per 15-minute interval includes mean, standard deviations, and range in parentheses).* PRE-TRIAL TRIAL POST-TRIAL CONTROL 18.8 + s.d. 5.5 18.6 + s.d. 4.9 18.0 + s.d. 5.2 n = 10.5 hrs (8-26) (6-26) (10-26) EXPERIMENT A 17.3 + s.d. 5.2 41.9 + s .d. 10.9 29.6 + s .d. 3.8 Real Sources (9-27) (22-61) (23-37) n = 10.5 hrs EXPERIMENT B 19.7 + s.d. 3.4 45.5 + s.d. 11.3 24.9 ± s.d. 4.3 Outboard Engine (15-26) (26-65) (17-32) n = 10.5 hrs EXPERIMENT C 17.6 ± s.d. 4.4 23.9 + s.d. 5.9 19.8 + s.d. 3.1 Gray Whale (8-28) (14-36) (16-28) n = 10.5 hrs EXPERIMENT D 18.7 + s.d. 4.7 0 15.2 + s.d. 3.7 Test Tone (7-26) (9-23) n = 10.5 hrs * Total number of hours for all experimentation was 52.5 hrs. Each block represents 3.5 hrs which would be divided by 15-min periods to represent an individual experiment thus 14 experiments were accomplished for each block. 137 Table IX. 1983 Call Types Produced by Gray Whales in Laguna San Ignacio, Mexi co. PRE-TRIAL TRIAL POST-TRIAL CONTROL SI S2 S3 S4 = 85% = 5% = 5% = 5% SI S2 S3 S4 = 85% = 5% = 5% = 5% SI S2 S3 S4 85% 5% 5% 5% EXPERIMENT A Real Sources SI S2 S3 S4 85% 5% 5% 5% SI S2 S3 S4 = 85% = 5% SI S2 S3 S4 = 85% = 5% = 5% = 5% EXPERIMENT B Outboard Engine SI S2 S3 S4 85% 5% 5% 5% SI S2 S3 S4 85% 5% 5% 5% SI S2 S3 S4 85% 5% 5% 5% EXPERIMENT C Gray Whale SI S2 S3 S4 85% 5% 5% 5% SI S2 S3 S4 85% 5% 5% 5% SI S2 S3 S4 85% 5% 5% 5% EXPERIMENT D Test Tone SI S2 S3 S4 85% 5% 5% 5% No sounds SI S2 S3 S4 = 90% = 4% 138 1983 Cal1 Structure To determine if structural changes occurred in the calls of gray whales during experimental testing, signal characteristics were compared among different events. Since the SI signal dominated the acoustical repertoire under all conditions (Table IX), I elected to use this particular phonation when investigating structural changes. The following variables were measured and compared within and among experiments: frequency range of signal (Hz), emphasized frequencies of signal (Hz), received level of sounds (dB re 1 yuPa), percentage of SI signals exhibiting frequency modulation, duration (sec), and number of pulses produced per series. Call repetition rates (number of pulses/sec) were derived from the average duration of the signal and the average number of pulses produced per series. Three sounds per 15-minute interval for each experimental period were randomly selected and analyzed; thus n = 42 sounds for each experiment were exami ned. When trial periods of Experiments A, B, and C (Experiment D excluded since all vocalizations ceased) were compared to pre-trial values, significant increases were noted in received levels, number of signals with frequency modulation, number of pulses produced per series, and repetition rate of signals. Changes were not detected among the three test periods for call duration, frequency range of signal, and in the major concentration of energy of the signal. When post-trial test periods for each experiment were compared to the trial periods, an overall decrease was noted in received levels, number of signals frequency modulated, number of pulses produced per series, and repetition rates of signal. However, when comparisons were made between post-trials and pre-trials for each experiment for the four acoustical variables, the post-trial values 139 were s t i l l somewhat higher than control values, but were shown to closely approximate pre-trial values for certain sound stimuli. An exception occurred when post-trial periods of test tones were compared to pre-trial periods. In all cases, similar variables changed among experiments, however the values were different. The degree/amount of structural change within a call may be related to the type of sound stimuli present in the habitat. In all three trial experiments (A, B, and C), when compared to control periods, changes occurred in received levels, number of signals exhibiting frequency modulation, and the number of pulses produced per series. The presentation of outboard engine noise elicited the greatest received levels (156.6 + s.d. 9.5 dB), followed by real sources (148.8 ± s.d. 9.8 dB), then gray whale sounds (126.4 + s .d. 11.7). The presentation of gray whale sounds resulted in the greatest number of signals showing frequency modulation (95.2%) followed by real sources (85.7%) and then outboard engine playback (83.3%). When gray whales were subjected to real sources, the number of pulses per series was highest at 24.6 + s.d. 8.0, followed by outboard engine playback (22.6 + s.d. 6.8), and then gray whale sounds (12.8 + s.d. 6.7). No changes occurred in call duration, frequency range or emphasized frequencies. An inspection of the differences obtained during post-trial periods among the experiments suggests that pre-disturbance levels are reached at different times depending upon the type of sound stimuli causing the acoustical perturbation. Values associated with these structural changes are summarized in Table X. Statistical values obtained during these investigations are listed in Appendix B (1983 Call Structure). Table X. 1983 Structural Changes in the Calls of Gray Whales Occupying Laguna San Ignacio, Mexico.* Frequency Range (Hz) Emphasized Frequencies (Hz) Recei ved Levels (dB re 1 >uPa) Percent Frequency Modulation Duration (sec) Number Pu Ises/ Series Pulse Repetition Rates ** CONTROL AND PRE-TRIAL 100-2000 300-825 117.6 + 14.2 (90-156) 71.4 1.7 + 0.8 (0.3-4.6) 9.4 + 4.4 (2-30) 5.4 REAL SOURCES TRIAL 120-1800 280-700 148.8 ± 9.8 (131-170) 85.7 2.0 + 1.1 (0.5-5.2) 24.6 + 8.0 (14-48) 12.0 POST-TRIAL 120-1950 315-820 128.3 + 6.6 (118-142) 78.6 2.0 + 1.0 (0.4-5.2) 18.4 + 8.4 (8-40) 9.2 OUTBOARD ENGINE TRIAL 150-2000 300-800 156.6 ± 9.5 (138-172) 83.3 2.0 + 0.9 (0.5-5.0) 22.6 ± 6.8 (12-46) 11.3 POST-TRIAL 200-1800 350-780 121.5 + 6.2 (102-132) 73.8 1.9 + 0.9 (0.4-4.6) 13.2 + 5.1 (7-36) 6.7 GRAY WHALES SOUNDS TRIAL 100-2000 300-790 126.4 + 11.7 (104-153) 95.2 1.9 + 0.9 (0.4-4.6) 12.8 + 6.7 (4-32) 6.6 POST-TRIAL 125-1950 320-820 119.7 + 6.6 (107-140) 73.8 1.9 ± 0.8 (0.4-4.2) 10.8 ± 5.0 (5-34) 5.4 TEST TONES TRIALS 0 0 0 0 0 0 0 POST-TRIALS 125-2000 300-800 115.4 ± 9.3 (94-132) 71.4 1.8 + 0.8 (0.4-4.2) 9.6 + 4.9 (3-22) 5.3 * For each experiment a n value of 42 sounds were analyzed. * * Derived from average call duration and the average number of pulses produced per series. 141 1983 Observed Surface Behavior To determine if whales moved away from the sound source, thus negating the need for them to alter sound behavior, 101 individual whales were visually tracked by observers based at the Rocky Point shore-station. Average dive duration (amount of time spent underwater between two surface blows) of gray whales occupying Laguna San Ignacio in 1983 (exposed to a variety of experimental conditions) was 68 + s.d. 34.0 sec (n = 700 dives). A frequency distribution of these dive durations by 10-sec categories is represented in Fig 32. Out of the 101 whales tracked, 56 (439 dives) of these whales were females with accompanying calves (= cow/calf pairs). The average dive time of these cow/calf pairs (blow rates collected on adult female not calf) was 49.1 + s .d. 14.8 sec. This value was statistically different (t = 5.96, p < 0.05) from the average dive duration obtained for single whales at 98.2 + s.d. 59.3 sec (n = 45 whales for a total of 261 dives). Figure 33 graphically demonstrates the differences between the respiration rates of these two whale groups. Since dive durations were statistically different between cow/calf pairs and single whales, these data could not be lumped when comparing control and experimental conditions. Unfortunately, this decreased the sample size per experiment. When cow/calf pair and single whale dive times were compared among control, real vessel, outboard, and test tone trial periods, there were no significant differences in respiration rates (Anova, df = 3, F = 0.86, p > 0.05; Anova, df = 3, F = 0.52, p > 0.05, respectively). These rates are listed in Table XI. During the collection of post-trial respiration rates of Experiment A, B, and D, cow/calf pairs were somewhat harder to find (as indicated by the reduced numbers on dive data) and were harder to track (less predictable dive pattern; missed blows). During the post-trial 142 Figure 32. 1983 Frequency D is t r ibu t ion of Gray Whale Dive Durations. 143 Duration of Dive (seconds) 144 Figure 33. 1983 Frequency Distribution of Cow/calf Pairs and Single Whale Dive Times in Laguna San Ignacio, Mexico. 14'5 LEGEND B Cow/calf pairs 1983 • Single whales 1983 o o o o '- CM po ^ I I I I <— CN CO o o o o o o o o o o o o o o o o * — i r i f D N m r o O ' - N r o t LO co r-» oo <D o o CD o «— C M C O ' d - L O < o r > . ra oi A Duration of Dive (seconds) 146 Table XI. Average Dive Times (sec) of Gray Whales During Various Experimental Conditions During the 1983 Season in Laguna San Ignacio, Mexico (includes mean, standard deviations and and range in parentheses). SINGLE WHALES COW/CALF PAIRS CONTROL 92.7 + s.d. 34.2 (44-182) n = 87 dives 52.0 ± s.d. 16.2 (31-89) n = 167 dives EXPERIMENT A Real Sources 82.3 + s.d. 31.5 (49-136) n = 103 dives 45.8 + s.d. 12.7 (26-71) n = 149 dives EXPERIMENT B Outboard Engine 107.3 + s.d. 21.9 (90-132) n = 22 dives 54.1 + s.d. 10.6 (38-70) n = 86 dives EXPERIMENT D Test Tone 95.5 ± s.d. 67.0 (27-221) n = 49 dives 47.5 ± s.d. 15.7 (27-78) n = 37 dives 147 periods of Experiment D (test tone), typically cow/calf pairs could not be found. Similar trends were discovered when looking at post-trial periods (Experiment D) for single whales as well. Because of this, sample sizes were significantly reduced, thus comparisons among post-trial experiments were not attempted. Although statistical differences in dive duration were not obtained when whales were exposed to increased levels of noise, an inspection of dive times greater than 150 sec (n = 53 dives) revealed the following. For each whale tracked that had a dive time greater than 150 sec, the associated line display was inspected (Fig. 34). An allowance of one minute on either side of an environmental event was given for a whale to react to the perturbation. A collective review of dive times greater than 150 sec indicated that 52% of these dives were associated with real sound sources in the environment and 43% were associated with playback experiments. The remaining 5% had no associated comments. Included in the percentage for playback experiments were 28 cases in which whales being tracked disappeared at the start of the playback session. When this occurred, observers watched for a total of three minutes and then ended that particular tracking session. These dive times were thus presumed to be greater than 150 sees. Directional changes were also investigated during the 1983 season. During control periods, whales typically changed direction 46% of the time as they transited through the waters of the immediate study area. In the presence of real sources, an increase (69%) was noted in the number of directional changes made by the whales. Similar increases in directional changes were documented when whales were exposed to outboard engine playback (64%). The greatest number of directional changes occurred in the 148 Figure 34. Example of Line Display Depicting Behavior of Gray Whales and Environmental Effects During Tracking Experiments. Ponga passes ^ ^ Direction change ^ Tourist skiff approaches ^ _ Spyhop ^ Tourist skiff still in area Bubble blast w Sub-surface exhalation ^ Dolphins associated with whale ^ Ponga rapidly ^approaching from the south • Whale respiration % = 10 seconds Line Display 150 presence of the test tone (87%). Control periods were compared to experimental periods and yielded the following results. When control periods (46%) were compared to real sources (69%), a significant increase was noted in directional changes (Chi Square = 9.90, p < 0.05). Similar increases in the number of directional changes were detected when controls (46%) were compared to outboard playback (64%) (Chi Square = 5.83, p < 0.05) and test tone playback (87%) (Chi Square = 35.9, p < 0.05). When real sources (69%) were compared to outboard engine playback (64%), significant differences were not found (Chi Square = 0.35, p > 0.05). A significant difference was detected when real sources (69%) were compared to test tones (87%) (Chi Square = 8.42, p < 0.50) and outboard (64%) was compared to test tone (87%) (Chi Square = 13.08, p < 0.50). In each case, the test tone elicited the greatest number of directional changes. The above data suggest that whales transiting through the immediate study area increased the number of directional changes in the presence of noise sources. Although the observers subjectively believed that whales were moving away from the station at certain times, overall movements toward and away from the sound source could not be quantified in the 1983 season. However, this season's work did provide the basis for detailed studies on whale movements for the 1984 season (Chapter 3). 151 DISCUSSION Significant changes were documented in the acoustical signalling of gray whales when these whales were exposed to temporary, increased levels of noise in their environment. Typically, as noise levels were increased during trial periods ( i .e . , real sources, outboard engine playback, and projection of gray whale signals), a corresponding increase was documented in: I) calling rates; 2) received level of the signals; 3) number of frequency-modulated signals; 4) number of pulses produced per series; and 5) repetition rates of signals. During the post-trial periods of these three experiments, the values associated with these five acoustical variables were shown to decrease with time and in some cases to approximate those values obtained during control periods. Statistical differences were documented when the values obtained during trial and post-trial periods were compared among the experimental conditions. The combined results indicate that different sound stimuli dictate varying acoustical responses by gray whales. In the presence of real sources and the projection of outboard engine noise, similar changes were noted in calling rates and call structure for the two sound stimuli. Only the received level of the call was shown to be significantly higher in the presence of outboard engine noise. When comparing the post-trial periods between these two sound stimuli, significant differences were noted. Gray whales resumed baseline/control values for calling rates and call structure at a faster rate after exposure to simulated boat noise than to a real sound source present in the environment. Although the two vessel noise sources were somewhat similar in overall profiles, the ways in which these stimuli were presented were different. Outboard noise was presented as a rapid onset of sounds while real sources generally increased 152 gradually. This factor may be responsible for the different values obtained for received levels of cal ls, with a rapid onset of noise causing a higher value. A possible explanation for the different rates obtained during post-trial periods of these two sound sources could be that residual effects are s t i l l prevalent in the habitat with real sources (actual presence of boat) s t i l l emitting some noise, whereas during post-trials of outboard engine noise, the noise perturbation was completely terminated at the end of the trial period. Although some minor differences occurred when comparing the two sound sources, typically the two sources elicited similar changes in the acoustical behavior of whales. An inspection of the similarities obtained during these two trial experiments (real vs tape outboard) is of major importance. Apparently, gray whales respond in a similar acoustical manner to either real sources and/or simulated boat noise. These results suggest that playback techniques can be used to assess and monitor the acoustical behavior of gray whales to increased levels of noise. Similar sound parameters changed when the calling rates and call structure were compared between the projection of gray whales (biological sounds) and the presence of non-biological sources (real and outboard). However, the degree/amount of change was not as great in the presence of gray whale sounds as that noted for vessel sources. One acoustical variable did show an unusually high rate of change when these experiments were compared. The number of signals exhibiting frequency modulation was highest (95.2%) during the projection of gray whale sounds. The presentation of the test tone resulted in a radical departure when compared to other sound stimuli. When whales were exposed to this signal, all signalling ceased. During post-trial periods, gray whales typically 153 remained silent for approximately 10 minutes following exposure to this stimulus. When calling resumed, values associated with calling rates and call structure were somewhat lower than control values. When compared to the other three sound stimuli, this test tone signal was unique in several aspects. The signal covered a wide range of frequencies (15 kHz to 400 Hz), exhibited a transient nature, was projected on a random schedule, and did not normally occur in the lagoon environment. Certain sound parameters did not change in the presence of the temporary noise: types of calls produced, frequency range of signal, emphasized frequency bands and call duration. Since the SI signal typically dominates the acoustical repertoire and also shows the greatest amount of variation for all call types investigated, i t may be advantageous for gray whales to modify this signal rather than switch to other signals. Possibly their only available option would be either to modify the predominant call type or move out of the area to ensure the maximum transmission and reception of their signals. Unfortunately the relevance of the call with respect to behavior is unknown. For example, i t may be more beneficial in terms of circumventing noise to continue producing SI signals; however, it may be irrelevant to change call types in terms of what the whale was doing and/or trying to convey in terms of communication. The fact that frequency shifting was not noted may reflect certain physical limitations regarding their anatomical structure and/or reflect responses to the rapid onset of short-term noise. Significant differences in dive durations did not occur between control and experimental conditions. Statistical differences did, however, occur when dive times were compared between cow/calf pairs and single whales. The reduced cow/calf pair dive times and less variation in length 154 of dive could result from the adjustment made by cows to accommodate the limited breath-hold capabilities of their young calves. Overall, respiration was highly variable, ranging from 8 sec to 301 sec, with an average dive interval of 68 + s.d. 34 sec. Although there was a tendency for longer dives to be associated with man-made events, this could not be demonstrated statistically. Possibly the small sample size (due to the breakdown needed) and the highly variable respiration rates masked this fact. A total of 101 whales was tracked during the 1983 season. Although this data base is considerable smaller than that collected by Harvey and Mate (1984), overall dive profiles and their associated frequency distribution are similar. The average dive time calculated by Harvey and Mate (1984) for 3 single whales and 7 cow/calf pairs was 96 + s. d. 1.2 sees, which represents a slightly longer dive duration than documented during these studies. Dive durations of cow/calf pairs were collected in March and April, thus these calves would be approximately 4-6 weeks older than those reflected in the current data base. Harvey and Mate's dive data represent respiration rates of gray whales collected under a wide variety of environmental conditions including exposure to man-made sources. Directional changes occurred 46% of the time when observing whales under control situations. A significant increase in number of directional changes occurred in the presence of noise, with the test tone causing the greatest number of directional changes. Movements were subjectively noted to occur both both toward and away from the sound source, depending upon the sound stimuli being projected. On several occasions, during the projection of outboard engine noise, whales were observed to overtly change direction and head directly (east) to the transducer location. In both cases, these whales dove directly over the transducer cage, lingered, 155 and then eventually swam away. This is of particular significance when one considers the "friendly/curious whale" behavior exhibited by gray whales in Laguna San Ignacio. Apparently, the whales interpreted the projected outboard engine noise as a real (= true) source. The attraction of curious whales to the transducer, as well as the similar acoustical results achieved between real sources and outboard playback, supports the use of playback techniques and experimentation for future studies of noise impact on cetaceans. The directional changes observed in 1983 and the reduced number of whales available for tracking during post-trial periods suggests that gray whales avoid the immediate study area (transducer site location) during exposure to increased levels of noise. The experimental design employed in 1983 did not measure the precise trackline (distance offshore or actual movement) and/or the amount of time in area of the whale relative to the transducer site. This may be an important consideration since the responses of whales may vary depending upon this relative distance to the sound source. The combined results of these 1983 acoustical and observed behavioral investigations suggest that gray whales, when exposed to rapid onset of temporary noise, can modify their acoustical repertoire through call structure and timing. Several acoustical advantages for a calling species can be associated with these observed changes. If sounds are produced more often, there is certainly a greater chance of the signals being perceived. A similar result would occur if the level of sound production were increased. Frequency modulated signals have two advantages. These FM signals carry more information and travel farther distances (Wiley and Richards, 1978), thus enhancing the signal's transmission. A greater number of pulses per series and faster repetition rates would also 156 serve to enhance the signal. Under certain situations ( i .e. , test tone projection) a cessation of all vocalizations was documented. It may be that gray whales have a certain level of tolerance to various signals and/or how noise sources are presented to them. It may be more advantageous for a whale to move out of the ensonified area. The experimental design employed during 1983 represented a projection of an intermittent/temporary, stationary sound source (except real sources). The onset of the sounds projected during trial periods was rapid, with no phasing in or out. When whales were in the immediate vicinity (within 1/2 mile range), the onset of sounds did not accurately represent a natural noise source encountered by a whale in the environment. Sound would typically increase/decrease gradually as whales moved by. Although valid, the resultant behavior may represent a startle response and thus may not adequately reflect the acoustical capabilities of the gray whale. Whales may conceivably respond differently to continuous, moving sound sources and/or exposure times. By employing different experimental methods, a greater understanding of the acoustical capabilities/strategies employed by gray whales can be obtained, lending more evidence in support and/or rejection of the proposed hypothesis. 157 C H A P T E R 3 RESPONSES OF GRAY WHALES TO INCREASED LEVELS OF CONTINUOUS (LONG TERM) NOISE 158 INTRODUCTION Ocean ambient noise .is eminently characterized by variability (Urick, 1967). Much of this variability can be attributed to changes in the dominant sources of noise, such as sea state, amount of shipping, and/or composition/abundance of sound-producing marine organisms. The dominant sources of noise, prevalent in a specific environment, are responsible for the overall profile of noise with respect to emphasized frequencies and decibel levels. Ambient noise can also be classified as an intermittent/transient or a constant source of noise in an environment. The contribution to the overall spectra can thus be regarded as short-term (e.g., temporary, seasonal) or long-term (prevalent over extended periods of time). The origin of these signals can also be mobile or stationary. The onset of the signals can be rapid (indicating a fast rise-time) and/or occur gradually over time (slow rise-times). Another source of variability arises out of the changing sound transmission conditions of the aquatic medium. When considering this dynamic aspect of noise, the potential exists that gray whales may respond in a variety of ways dependent upon the type and nature of the sound stimulus in the environment. As demonstrated in Chapter 2, acoustical and observed surface behaviors of gray whales were documented to change when exposed to increased levels of short-term noise in Laguna San Ignacio. Responses varied with respect to the particular sound source present. These results suggested that the sound type, not the playback itself , was causative. The 1983 sound sources varied with respect to overall profiles; however, the presentation of most of these stimuli was similar across all experiments ( i .e . , rapid onset of short-duration sound sources). An exception occurred 159 when real sources were prevalent in the environment. The exposure of whales to these sources was gradual over time but s t i l l represented short-term exposure levels. When comparing the 1983 results between outboard engine playback periods and periods when real vessel noise occurred, differences were documented in the acoustical responses of the whales to these two sources. Could these differences reflect the whales' acoustical reaction to the rapid onset of short-term noise ( i .e . , outboard engine playback) vs their response to the gradual onset of short-term noise (real vessels in area)? At sea, when noise sources are encountered, typically this exposure is gradual over time. This allows the receiver adequate time to respond to and/or react to the source. Occasionally, in the case of certain signals, such as explosive charges (e.g., seismic blasts associated with oil exploration or natural sources such as isostatic rebound effects), the onset of sound is rapid and unpredictable. In these situations, adequate response time may not be given and the resultant behavior thus could be interpreted as a "startle response". Startle responses are, of course, valid and significant since they do represent an aspect of the whales' behavior; however, these responses may not fully represent the acoustical capabilities of the species being investigated. The possibility thus existed that the 1983 responses of gray whales (due to the experimental design employed) could have been described as startle responses. To determine if gray whales employ different acoustical strategies or responses when exposure time to the source is increased, a study was conducted in 1984 to investigate behavioral responses of gray whales to the projection of gradual onset of long-term (sounds projected for six to eight hours at a time) noise sources in Laguna San Ignacio. It was hoped 160 that these investigations would provide a better understanding of the acoustical capabilities/strategies of the gray whale. This Chapter investigates the effect of long-term noise on gray whales and compares the acoustical and observed surface responses of whales between the 1983 and 1984 seasons. 161 MATERIALS AND METHODS 1984 Experimental Design The experimental design employed control periods that were six to eight hours in duration. These control periods were defined as times without any intervention of man-made noise (either real sources or projected sounds). The playback experiments, conducted for six to eight hours at a time as well, included the projection of biological as well as non-biological sound sources. During biological playback, ki l ler whale (Orcinus orca) sounds were projected. Since ki l ler whales are known predators of gray whales (Dahlheim, 1981), it was assumed that this particular playback would result in an avoidance/aversive reaction on the part of the gray whales. This would allow documentation of the responses of gray whales to a known aversive stimulus, which could provide an important interpretive element to this study. Two types of man-made noise sources were projected back underwater during these 1984 studies. These included outboard engine noise (same tape used during 1983) and oi l -dr i l l ing sounds (collected in the Beaufort Sea). These two man-made sources dominated frequencies less than 2 kHz and thus potentially competed in the same frequency bands as signals produced by the gray whale. To verify that gray whales were not responding to the power generated/electrical energy produced by the transducer during playback experiments, the transducer was turned on for periods of up to eight hours, but no sounds were projected. As in 1983, acoustical recordings were also made in the presence of real sources prevalent in the environment and compared between control periods and other experimental conditions. Experimental testing was 162 terminated when: 1) Beaufort conditions created noise interference, and 2) noise from unplanned vessel and skiff traffic interfered with experimental on. 1984 Observed Surface Behavior - Tracking Experiments Based upon the experience gained in 1983, I decided that it was not only necessary to collect dive duration data on gray whales but also to collect precise data on the movements of the whales relative to the transducer site. The 1984 experimental design was modified from that of 1983 to allow for precise comparisons to be made of the surface activities of whales between control and experimental periods. A four-member team was necessary to carry out this phase of the research. One member was positioned at a northerly station, designated as North Marker. The other member was stationed at a southerly station, termed South Marker. The distance between these two stations was 100 m. The boundaries defined for the study area during these tracking experiments were based upon the effective visual range of the observers, which ensured that the observers could accurately identify an individual whale. Each observer was equipped with a pair of Steiner binoculars. These binoculars had built-in compasses, allowing the observers to collect the bearing to an individual whale as it came to the surface to respire. The North and South observers gathered data on the same whale, which was coordinated through the use of hand-held CB radios. Only well-marked, easily identifiable whales were selected to track. A third team member (positioned at South Marker) was designated as a recorder for the primary observers and collected the following information on a data entry form designed for this purpose: identification of observer North, South and recorder; date; whale identification number (numbered consecutively); 163 identification of whale grouping (e.g., cow/calf pair, singles); time of respiration; south and north bearing of each blow; environmental data to include information on weather, boats present and whale behavior. During these tracking experiments, the fourth team member was positioned near the shore-based playback equipment. The tracking team was not informed as to what, i f any, type of playback was occurring, thus avoiding bias in sampling. An effort log was maintained by the observers noting the amount of time on station per day. 1984 Distribution/Abundance - Transect Experiments To determine if a significant change occurred in the distribution and abundance of gray whales in the presence of increased levels of noise, vessel surveys were conducted during control and experimental conditions. The design of these surveys and the rationale employed followed those described by Jones and Swartz (1984), which represented a line transect methodology (Eberhardt, 1978). Transects were conducted from a 4.3-m inflatable boat powered by a 20-hp outboard engine, travelling at a speed of 6 knots. This speed ensured that a whale (which usually travels at 2-4 kts) would not move ahead of the vessel and thus be double counted. A complete survey of the lagoon took approximately 3 hours and 20 minutes. Transects were conducted along an imaginary line drawn through mid-lagoon (Fig. 35) from the breaker line at the lagoon entrance, north to I si a Garzas (16 nmi). This mid-line survey track ensured that all whale inhabitable waters were properly censused (both shorelines were clearly visible). Whales in the "North End" of the upper lagoon (north of transect termination point) were censused from a 5-m bluff on the northern tip of I si a Garza following the procedures developed by Jones and Swartz (1984). 164 Figure 35. 1984 Vessel Survey Transect Route of Entire Lagoon Syst From Jones and Swartz, 1984. 166 The survey team included two primary observers (left and right positions) and a vessel operator. Positions were rotated to prevent observer anticipation. All whales observed were counted as they were approximately abeam of the vessel to eliminate any confusion factor when counting whales. All sightings were recorded on a hand-held tape recorder. For example: when a cow/calf pair was observed east of the transect line heading north in the lagoon, the sighting was noted as C/CNE. Cow/calf pairs were counted as a single unit. If three single whales, as a group, were observed heading South on the west side of the transect line, the notation would be 3SW. Whales were also noted as follows: DU = direction unknown (observer was not able to clearly see the whale) and ND = no direction (whale was milling). Abundance estimates were based entirely upon these sighting data. Whale distribution was determined by noting (on the cassette tape): starting time of transect; time each whale was observed; passage of landmarks; and end time of transect. Transects were terminated when wave and wind conditions obscured whale blows which lowered the overall probability of whale detection (Beaufort 3 conditions = winds 12-18 km/hr). In addition to the full transects conducted, allowing comparisons of overall abundance between the 1984 season and the 5-year data base (1978-1982) collected by Jones and Swartz (1984), modified transects were also conducted. These modified transects (= partials) employed the same methods as the full census; however, the length of the transect line was shortened. Surveys were run between Parmenter Point and Yucca Plant (4 nautical miles - Fig. 36). These shorter surveys allowed transects to be conducted during control periods, as well as during playback periods. The vessel survey team would begin surveying at Parmenter Point and proceed north to Yucca Plant. During this period, no active playback took place 167 F i g u r e 36. 1984 M o d i f i e d T r a n s e c t Route -- Laguna San I g n a c i o , M exico. 168 169 (= control transect). The 4-mile transect took approximately 50 minutes to complete. As the northbound survey vessel passed the Rocky Point Station, the land-based observer would note the time of passage. An additional 35 minutes were given by the shore-based observer to allow adequate time for the survey vessel to reach the northbound limit (Yucca Plant) of the survey tract. Exactly 35 minutes after the vessel had been sighted abeam of Rocky Point and confirmation was obtained by radio that the vessel had reached Yucca Plant, sound playback was initiated. At this time the survey vessel began its southbound transect (= experimental transect), starting at Yucca Plant and proceeding south to Parmenter Point. Surveys were conducted at various tidal stages and times of the day to avoid the influence of these environmental factors. To avoid bias, the survey team was not informed as to what, i f any, playback was occurring. This experimental design allowed comparisons to be made between control and experimental transects for the 1984 season. 1984 Analysis of Gray Whale Sound Behavior The analyses procedure of the 1984 acoustical behavior followed the methods described for the 1983 acoustical analyses detailed in Chapter 2 (Materials and Methods section). Six experimental groupings were defined for the 1984 season. 1) Control - no intervention of real man-made noise into the envi ronment 2) Experiment A - real man-made sounds prevalent in environment 3) Experiment B - simulation of outboard engine noise 4) Experiment C - simulation of oi l -dr i l l ing noise 170 5) Experiment D - simulation of ki l ler whale sounds 6) Experiment E - equipment on, no sounds projected Since the 1984 design did not employ a pre-trial, trial and post-trial period, intra-experimental comparisons were not necessary. For each experimental and control condition the following sound parameters were inspected: Call periodicity, changes in call types and changes in call structure. The detailed procedures regarding the analyses and subsequent statistical comparisons of these sound parameters are described in Chapter 2 (Materials and Methods - Analyses of Gray Whale Sound Behavior). 1984 Analysis of Tracking Experiments The 1984 analyses on the respiration rates of gray whales followed the same procedures as those described in 1983 (Analyses of Observed Surface Behavior - Chapter 2). The 1984 data were also compared to the dive durations collected in 1983. In addition to the data collected on the dive durations of gray whales, detailed information was also available in 1984 on the movements (tracklines) of the whales as they passed through the study area. As each whale came to the surface to respire, two simultaneous bearings were taken from the North and South stations. Bearings collected were then converted into x and y distance values for each blow (= sighting) through triangulation. Data were screened for obvious errors and those bearings giving unrealistic distances and/or distance lines that did not intersect were deleted from the data base. A computer program was written which allowed conversion of these x/y coordinates. For each value obtained, a position (mark) was scored on a chart for each respiration/sighting. The series of sightings were then connected to illustrate the trackline of a 171 specific whale. Based upon these tracklines, calculations were then made to determine the average distance travelled offshore and to investigate the amount of time spent travelling vs milling by the whale. These data were compared between control and experimental periods. Also available was the start and end time of each whale's trackline. These values allowed calculations of the amount of time spent in the area by the whale to be compared between experiments. To obtain the amount of time spent tracking a whale under various conditions, an inspection of the effort log combined with the data entry form used during tracking experiments, was investigated. The percentage of time spent per experiment on a trackable whale as a reflection of the effort expended was thus obtained. 1984 Analysis of Transect Experiments Sighting data stored on cassette tapes were transcribed to a written edit sheet. The edit sheets contained not only whale counts but also identified the time of passage of landmarks and start and end times of the transects. The information on these edit sheets was then transferred to a daily summary sheet. From each summary sheet, the abundance and distribution of whales per specific area could be determined. In addition to the abundance and distribution values, each sighting record noted the direction in which the whales were heading. The coding system employed during these surveys allowed an inspection of different directions and milling rates of the whales to be made among experiments. Seasonal comparisons were made between control and experimental transects examining abundance, distribution, direction, and milling. Between-season comparisons (1984 to 1978-1982) were only accomplished on abundance estimates. 172 RESULTS 1984 Calling Rates Significant differences occurred between the calling rates obtained for each experimental test period (Anova, df = 4, f = 257.3, p < 0.05). Calling rates were highest when whales were exposed to real sources in the environment (159.6 + s.d. 35.5 sounds produced per hour). In the presence of outboard engine noise, these rates were 137.9 + s.d. 24.7 per hour. In both cases, vessel noise resulted in higher calling rates than control periods. Rates decreased dramatically from those collected during control periods in the presence of oi l -dr i l l ing sounds (14.9 ± s .d . 7.4/hr) and the projection of ki l ler whale sounds (all calling ceased). These results suggest that the amount of calling is dependent upon the type of sound source present in the environment. Table XII summarizes the rates of calling for each experiment. The statistics and detailed comparisons of these investigations are listed in Appendix C (1984 Calling Rates). 1983/1984 Comparisons of Calling Rates Significant differences occurred in the calling rates of gray whales dependent upon the signal presentation and/or duration of the perturbation. Calling rates for 1983 were based upon the number of calls produced per 15-min interval. For the purposes of these comparisons, the 1983 rates were adjusted to represent the number of calls produced per hour. The calling rates of gray whales for the 1983 control periods (Chapter 2) averaged 73.6 ± s .d. 19.2 and those obtained in 1984 control periods were 81.4 + s.d. 16.6. When compared, these control values were not significantly different (t = 1.45; p > 0.05). Calling rates produced by gray whales in the presence of real noise sources between 1983 173 Table XII. 1984 Calling Rates of Gray Whales in Laguna San Ignacio, Mexico (total number of sounds produced per hour includes mean, standard deviations and average range in parentheses). CALLING RATES (number of calls produced per hour) CONTROL n = 40 hrs 81.4 ± s .d. 16.6 (32-106) EXPERIMENT A Real Sources n = 40 hrs 159.6 ± s.d. 35.5 (90-246) EXPERIMENT B Outboard Engi ne n = 40 hrs 137.9 + s.d. 24.7 (100-200) EXPERIMENT C Oi 1 -Dri 11 ing Sounds n = 40 hrs 14.9 + s.d. 7.4 (3-32) EXPERIMENT D Killer Whale Sounds n = 40 hrs No sounds produced EXPERIMENT E Equipment on, no sounds projected n = 40 hrs 78.5 ± s.d. 17.0 (38-108) 174 167.7 + s.d. 43.6) and 1984 (159.6 + s.d. 35.5) were also similar (t = 0.68; p > 0.05). The presence of real sources for both years caused an elevation in calling rates when compared to control periods. When the call rates of gray whales were compared between the two years in the presence of outboard engine noise, significant differences were detected (t = 4.56; p < 0.05). The 1983 calling rate of 182.0 ± s.d. 45.2 was signficantly higher than the 1984 calling rate of 137.9 + s.d. 24.7. Although the same tape was used during the 1983 and 1984 seasons, the presentation and duration of this stimulus varied between the years. In 1983 this sound stimulus was presented as a rapid onset of short-term noise, whereas in 1984 this vessel stimu-lus was phased in over time and was presented for longer durations. The results of these comparisons suggest that a higher rate of calling occurs when gray whales are exposed to short-term, rapid onset of noise. For both years, however, a higher rate of calling was documented in the presence of outboard engine noise when these trial values were compared to control periods. An inspection of the projected sound stimuli that varied between the 1983 and 1984 field seasons (gray whale sounds and test tones in 1983 and oi l -dr i l l ing and ki l ler whale playbacks in 1984) yielded the following results. In the presence of increased gray whale phonations, a call rate of 95.6 + s.d. 23.6 calls per hour was established. When compared to the control values obtained in 1983 and 1984, an increase in calling rates was documented in the presence of their own signals. This rate was, however, lower than the calling rates obtained in the presence of vessel noise projected in 1983 and 1984 for both real source and outboard engine playback. 175 The projection of oi1-dri11ing sounds reduced the total number of calls produced per hour. In the presence of this noise stimulus, gray whales produced an averaged of 14.9 + s.d. 7.4 sounds per hour. This rate was significantly lower than those obtained for control periods, vessel noise, and gray whale signals. The most dramatic effect on calling rates was observed to occur in the presence of test tones and ki l ler whale playback periods. These two sources caused gray whales to terminate all calling. Table XIII lists the calling rates obtained for the 1983 and 1984 seasons. 1984 Call Types In 1984, as in 1983, the SI signal dominated the vocal repertoire of the gray whale during control periods, representing 85% of the calls produced. Of the remaining call types, S2 signals occurred 5%, S3 signals occurred 5%, and S4 signals occurred 5%. During all 1984 experimental periods, with the exception of ki l ler whale playback when all calling ceased, the SI signal remained dominant in the sound repertoire of the gray whale. Table XIV summarizes the results of these 1984 investigations on call types. 1984 Cal1 Structure Exposure of gray whales to real sources, outboard engine noise, and oi l -dr i l l ing sounds resulted in marked differences in gray whale call structure (increased values) when compared to control periods and test periods of Experiment E (equipment on, no sounds projected). Overall, similar acoustical variables and rates of change occurred when these three sound sources were compared. These results imply that similar mechanisms are employed by gray whales independent of the type 176 Table XIII. Comparisons of 1983 and 1984 Calling Rates of Gray Whales occupying Laguna San Ignacio, Mexico (mean and standard deviations). * 1983 Season * * 1984 Season CONTROL 73.6 + s .d. 19.2 n = 3.5 hrs 81.4 + s .d. 16.6 n = 40 hrs REAL SOURCES 167.6 + s.d. 43.6 n = 3.5 hrs 159.6 + s.d. 35.5 n = 40 hrs OUTBOARD ENGINE 182.0 ± s.d. 45.2 n = 3.5 hrs 137.9 ± s .d. 24.7 n = 40 hrs GRAY WHALE SOUNDS 95.6 + s.d. 23.6 n = 3.5 hrs No testing OIL-DRILLING No testing 14.9 + s.d. 7.4 n = 40 hrs TEST TONE Ceased al1 calli ng n = 3.5 hrs No testing KILLER WHALE No testing Ceased all calli ng n = 40 hrs * 1983 values multiplied by 4 to represent call rate per hour. ** Represents rates obtained during trial periods. 177 Table XIV. 1984 Call Types Produced by the Gray Whale in Laguna San Ignacio, Mexico. PERCENTAGE OF CALL TYPES PRODUCED SI SIGNAL S2 SIGNAL S3 SIGNAL S4 SIGNAL CONTROL 85% 5% 5% 5% EXPERIMENT A 85% 5% 5% REAL SOURCES EXPERIMENT B 85% 5% 5% 5% OUTBOARD ENGINE EXPERIMENT C 97% 1% 1% OIL DRILLING EXPERIMENT D KILLER WHALE EXPERIMENT D 85% 5% 5% 5% EQ.ON - NO SOUNDS PROJECTED 178 of sound source present in the environment. The 1984 values obtained for call structure are listed in Table XV. The presence of real sounds and the projection of outboard engine noise elicited similar structural changes in the gray whale's SI signal. Received levels, compared between these two experiments, showed comparable values (real sources = 148.5 + s.d. 12.2 and outboard = 150.2 + s.d. 12.8) and were not significantly different (t = 0.8, p > 0.05). The number of signals exhibiting frequency modulation was also similar (real sources = 90% and outboard engine playback = 86.2%) (Chi Square = 0.37; p > 0.05). As expected from previous comparisons, a significant difference in call duration was not found (real sources = 2.0 + s .d. 1.0 and outboard playback = 2.0 + s.d. 0.8) (t = 0.07; p > 0.05). An inspection of the number of pulses produced per series also yielded similar values (real sources = 25.4 + s .d. 8.5 and outboard = 23.1 + s.d. 6.9) (t = 1.8; p > 0.05). When comparing these two sound stimuli, repetition rates, although slightly higher in the presence of real sources (12.7 per sec), were close to the value calculated in the presence of outboard engine noise (11.5 per sec). The overall frequency range of the signal and the emphasized frequency bands were also similar between these two experiments. When comparing real and/or outboard engine noise against the structural values obtained in the presence of oi l -dr i l l ing sounds, the majority of comparisons suggested that similar changes were occurring among the three experiments. One acoustical variable, however, was significantly different among these testing periods. When the received level of the call was compared between the presence of real sources (148.5 + s.d. 12.2) and oi l -dr i l l ing sounds (143.1 + s.d. 9.4), significant differences were obtained (t = 3.1; p < 0.05). This was also the case Table XV. 1984 Structural Changes in the Calls of Gray Whales Occupying Laguna San Ignacio, Mexico. Emphasized Received Percent Number Pulse Frequency Frequencies Levels (dB Frequency Duration Pulses/ Repetition Range (Hz) (Hz) re 1 AiPa) Modulation (sec) Series Rates * CONTROL 100-2000 350-850 120.3+17.6 68.7 1.9 ± 0 . 8 9.9+4.0 5.2 n = 80 sounds (86-168) (0.3-4.6) (2-30) REAL SOURCES 150-1800 300-800 148.5+12.2 90.0 2.0+1.0 25.4+8.5 12.7 n = 80 sounds (124-180) (0.5-5.3) (14-52) OUTBOARD ENGINE 120-1900 320-860 150.2+12.8 86.2 2.0+0.8 23.1 ± 6 . 9 11.5 n = 80 sounds (120-180) (0.4-5.0) (12-50) OIL DRILLING 110-2000 310-850 143.1+9.4 92.5 1.8 +.0.9 25.4 +. 7.4 14.1 n = 80 sounds (129-178) (0.4-5.2) (10-48) KILLER WHALE 0 0 0 0 0 0 0 EQUIPMENT ON -NO SOUNDS 140-1800 340-860 119.6+14.3 66.2 1.9+0.8 9.3+5.4 4.9 n = 80 sounds (92-160) (0.4-4.2) (2-32) * Derived from average call duration and the average number of pulses produced per series. 180 when the received levels of the ca l l were compared between outboard engine noise (150.2 + s . d . 12.8) and o i l - d r i l l i n g noise (143.1 + s . d . 9.4) (t = 3.9; p < 0.05). Received levels were s i gn i f i can t l y higher in the presence of vessel noise (real and outboard engine) than those obtained in the presence of o i l - d r i l l i n g sounds. The s t a t i s t i c a l values obtained for the other acoust ical var iables tested during these comparisons are given below. Real vs o i l - d r i l l i n g comparisons: frequency modulation (Chi Square = 0.15, p > 0.05); duration (t = 1.05; p > 0.05); number of pulses per ser ies (t = 0.009; p > 0.05); and repet i t ion rates 12.7 vs 14.1. Outboard vs o i l - d r i l l i n g comparisons: frequency modulation (Chi Square = 1.45; p > 0.05); duration (t = 1.23; p > 0.05); number of pulses per ser ies (t = 1.96; p = 0.05); and repet i t ion rates 11.5 vs 14.1. The frequency range of the signal and the emphasized frequency bands were s im i la r for a l l three experiments. During the project ion of k i l l e r whale sounds, a l l gray whale vocal izat ions ceased; thus, s t ructura l comparisons among experiments could not be completed. Detai led comparisons and resultant s t a t i s t i c s are given in Appendix C (1984 Cal l St ructure) . 1983/1984 Comparisons of Ca l l Structure The combined resul ts of the 1983/84 seasons documented a measurable di f ference in received levels of c a l l s , amount of frequency modulation wi th in s igna ls , number of pulses produced per se r ies , and repet i t ion rates of s ignals in the presence of increased noise. Frequency range, emphasized frequency bands and c a l l duration was not al tered in the presence of noise. Table XVI l i s t s the values obtained for the 1983 and 1984 comparisons of ca l l s t ructure. Table XVI. 1983 and 1984 Comparisons of Structural Changes in the Ca l l s of Gray Whales in Laguna San Ignacio, Mexico (mean values and standard dev ia t ions) . * Frequency Range (Hz) Emphasized Frequencies (Hz) Received Levels (dB re 1 yuPa) Percent Frequency Modulation Duration (sec) Number Pulses/ Series Pulse Repet i t ion Rates ** CONTROL 1983 100-2000 300-825 117.6 t 14.2 71.4 1.7 + 0.8 9.4 + 4.4 5.4 1984 100-2000 300-850 120.3 + 17.6 68.7 1.9 + 0.8 9.9 ± 4.0 5.2 EXPERIMENT E 1 984 140-1800 340-860 119.6 + 14.3 66.2 1.9 ± 0.8 9.3 + 5.4 4.9 REAL SOURCES 1983 120-1800 280-700 148.8 + 9.8 85.7 2.0 +1.1 24.6 + 8.0 12.0 1984 150-1800 300-800 148.5 + 12.2 90.0 2.0 ± 1.0 25.4 ± 8.5 12.7 OUTBOARD ENGINE 1983 150-2000 300-800 156.6 + 9.5 83.3 2.0 + 0.9 22.6 + 6.8 11.3 1984 120-1900 320-860 150.2 + 12.8 86.2 2.0 + 0.8 23.1 + 6.9 11.5 GRAY WHALE 1983 100-2000 300-790 126.4 + 11.7 95.2 1.9 + 0.9 12.8 + 6.7 6.6 OIL DRILLING 1984 110-2000 310-850 143.1 + 9.4 92.5 1.8 + 0.9 25.4 + 7.4 14.1 TEST TONE*** 1983 0 0 0 0 0 0 0 KILLER WHALE*** 1984 0 0 0 0 0 0 0 * Sample s ize for 1983 = 42 sounds analyzed and in 1984 80 sounds per experiment were analyzed. * * Derived from average ca l l duration and the average number of pulses produced per se r i es . * * * A l l s igna l l i ng ceased in the presence of th is sound st imulus. 182 When control periods were compared between 1983 and 1984, the following results were obtained. The received level of the call was similar for both years (1983 = 117.6 + s.d. 14.2 and 1984 = 120.3 ± s.d. 17.6) (t = 0.8; p > 0.05). The number of signals that were frequency modulated was also similar between the 1983 (71.4%) and 1984 seasons (68.7%) (Chi Square = 0.06; p > 0.05). The number of pulses produced per series for control 1983 was 9.4 + s.d. 4.4, and 9.9 + s.d. 4.0 for the 1984 season. These values were also similar (t = 0.6; p > 0.05). Comparable repetition rates were derived for both the 1983 and 1984 seasons (5.4 and 5.2 pulses per sec, respectively). The similarities obtained for control periods of both 1983 and 1984 are suggestive of an optimal/preferred call rate and call structure for gray whales occupying Laguna San Ignacio. Real sources and their effect on call structure were compared between the 1983 and 1984 seasons and significant differences in call structure could not be obtained. The received level of the call in 1983 was 148.8 + s.d. 9.8 and 148.8 + s .d. 12.2 in 1984. These values were not statistically different (t = 0.09; p > 0.05). In 1983, 85.7% of the signals were frequency modulated. When this value was compared to the 90% obtained in 1984, once again significant differences were not detected (Chi Square = 0.5; p > 0.05). An inspection of the number of pulses produced per series was also similar between the two years (1983 = 24.6 + s.d. 8.0 and 1984 = 25.4 + s.d. 8.5) (t = 0.5; p > 0.05). Calculated repetition rates were 12.0 for 1983 and 12.7 for 1984. A comparison of the structural changes occurring in the SI signal in the presence of outboard engine noise between 1983 and 1984 yielded the following results. The 1983 received level was 156.6 + s.d. 9.5. This value was statistically higher than that obtained in 1984 (150.2 + s.d. 183 12.8) (t = 2.85; p < 0.05). Similar values were obtained for the remaining acoustical variables in the presence of outboard engine noise and are summarized below. Frequency modulation in 1983 equals 83.3% vs 86.2% in 1984 (Chi Square = 0.14; p > 0.05); number of pulses produced per series, 1983 = 22.6 + s.d. 6.8 and 1984 = 23.1 + s.d. 6.9 (t = 0.4; p > 0.05); and repetition rates for 1983 were 11.3 per sec vs 11.5 per sec for 1984. Overall the presence of outboard engine noise elicited similar responses for the 1983 and 1984 seasons. The differences obtained in the received levels of the call between the two seasons could be due to differences in signal presentation and/or duration. An inspection of the structural changes observed in the presence of increased gray whale phonations (1983) and oi l -dr i l l ing playback periods (1984) also documented changes in received levels, number of signals showing frequency modulation, number of pulses produced per series, and repetition rates of calls. The values obtained in the presence of gray whale signals, although modified from controls, did not typically match the higher rates obtained during the presence of vessel noise and/or oi l -dr i l l ing sounds. However, the number of signals exhibiting frequency modulation was 95.2% in the presence of their own signals, a value significantly higher than other test periods. 184 1984 Observed Surface Behavior - Tracking Experiments Average dive duration of whales during the 1984 season (when exposed to a variety of experimental conditions) was 92.5 + s.d. 58.1 sec (n = 1287 dives). A frequency distribution of these dive times by 10-sec categories is represented in Figure 37. Of the 128 whales tracked, 61 (768 dives) were cow/calf pairs. The average dive time of these cow/calf pairs (blow rate collected on adult female) was 70.9 ± s.d. 30.7 sec. This dive duration was statistically different (t = 4.57; p < 0.05) from the average dive duration obtained for single whales at 107.7 + s.d. 55.5 sec (n = 67 whales, total dives = 519). Figure 38 graphically represents the frequency distribution of dives for cow/calf pairs and single whales. When cow/calf pair and single whale dive times were compared among control, real sources, outboard engine playback, and oi l -dr i l l ing playback periods, significant difference in dive durations could not be established among the events (ANOVA, df = 3, F = 0.92; p > 0.05; ANOVA, df = 4, F = 0.76, p > 0.05, respectively). Valid comparisons could not be accomplished during ki l ler whale playbacks since cow/calf pairs did not enter the area. The dive durations calculated for both groups of whales during control and experimental conditions are listed in Table XVII. When the dive durations of gray whales were compared between the 1983 and 1984 seasons, significant differences in these rates were obtained. In 1983, the overall dive duration of all whales exposed to a variety of environmental conditions was 68.0 + s.d. 34.0 sec. In 1984, this rate increased to 92.5 + s.d. 58.1 sec. Dive durations in 1984 were significantly longer than those observed in the 1983 season (t = 3.7; p < 0.05). To determine if cow/calf pairs and/or single whales were responsible for these observed differences, comparisons were 185 Figure 37. 1984 Frequency D is t r ibu t ion of Dive Durations of Gray Whal Occupying Laguna San Ignacio, Mexico. Duration of Dive (seconds) 187 Figure 38. 1984 Frequency Distribution of Cow/calf Pairs and Single Whale Dive Durations in Laguna San Ignacio, Mexico. 18 8 >• CJ c cu LEGEND • Cow/calf pairs 1984 Q Single whales 1984 o o o o > - ( N ro t I I I «— CM CO o o o o o o o o o o o o o o o o *-Ln CD oo o) o <— CN oo n- L O C O oo o o o 5: ,7; ^ s 1 I I i I i i I I- I I I m O ' - C M n t i n i O N 00 05 A Duration of Dive (seconds) 189 Table XVII. Average Dive Times (sec) of Gray Whales Exposed to Various Experimental Conditions During the 1984 Season in Laguna San Ignacio, Mexico (includes mean and standard deviations). SINGLE WHALES COW/CALF PAIRS CONTROL AND EXPERIMENT E 120.1 + s.d. 32.7 n = 188 dives 67.1 + s.d. 14.8 n = 278 dives EXPERIMENT A Real Sources 125.8 + s .d. 28.6 n = 127 dives 59.4 + s.d. 21.3 n = 369 dives EXPERIMENT B Outboard Engi ne 94.3 + s .d. 46.7 n = 56 dives 83.8 + s .d. 16.2 n = 33 dives EXPERIMENT C Oil Drilling 81.0 + s.d. 42.6 n = 86 dives 72.9 + s .d. 12.4 n = 88 dives EXPERIMENT D Ki Her Whale 113.2 + s.d. 30.1 n = 62 dives Pai rs not present 190 made by whale group between the two seasons. When the dive durations of cow/calf pairs collected in 1983 (49.1 + s.d. 14.8 sec) were compared to the dive times collected in 1984 (70.9 + s.d. 30.7 sec), a significant difference was obtained (t = 4.8; p < 0.05). The dive times of cow/calf pairs in the 1984 season were significantly longer than those noted in 1983. When comparing the rates of single whales between the 1983 (98.2 + s.d. 59.3 sec) and 1984 seasons (107.7 + s.d. 55.5 sec), significant differences were not detectable (t = 0.86; p > 0.05). Similar dive durations were obtained for single whales between the two seasons. Table XVIII summarizes the values obtained and these data are graphically illustrated in Figures 39, 40 and 41. Shore-based observers experienced days when they spent a considerable amount of time tracking whales in the immediate study area. Conversely, on some days, a considerable amount of effort was expended, but the number of whales actually tracked was low. As suggested by an inspection of the sample sizes in Table XVII, fewer whales were tracked for certain experimental conditions. Since each observer noted the time spent on effort each day and also had a record of the time spent tracking a whale, an estimate of the effort expended per day and per experiment and the percentage of time spent on a trackable whale could thus be quantified. During control periods, observers spent 53% of the time tracking whales in the immediate study area. In the presence of real sources, this value was 46%. During the projection of outboard engine noise, this value decreased, and observers spent 38% of the time on trackable whales. A dramatic reduction in the number of available whales for tracking occurred during the playback of oi l -dr i l l ing sounds and ki l ler whale sounds. During the projection of oi l -dr i l l ing sounds, only 9% of the 191 Table XVIII. 1983/1984 Comparisons of Average Dive Times of Gray Whales in Laguna San Ignacio, Mexico (mean and standard deviations) 1983 Season 1984 Season ALL WHALES 68.0 + s.d. 34.0 n = 700 dives 92.5 + s.d. 58.1 n = 1287 dives SINGLE WHALES 98.2 + s.d. 59.3 n = 261 dives 107.7 + s.d. 55.5 n = 519 dives COW/CALF PAIRS 49.1 + s.d. 14.8 n = 439 dives 70.9 ± s.d. 30.7 n = 768 dives 192 Figure 39. 1983/1984 Frequency Distribution Comparisons of Dive Times on All Whales Occurring in Laguna San Ignacio, Mexico. 19 3 > c cu cr cu 25 24 23 22 21 20 19 18 17 16 ' 15 14 13 ' 12 11 [ 10 ' 9" 8 l\ 6 5 _ 4 3^ 2 ' 1 ' LEGEND  • Total whales 1983 • Total whales 1984 0 0 0 0 '— CN CO <— CN CO O O O O O O O O O O O O O O O O ' — in to N op oi o 1 - C N i r o ^ L n c D r v o o c T i O O i i s i i n i i i i i n n j Duration of Dive (seconds) 194 Figure 40. 1983/1984 Frequency Distribution Comparisons of Dive Durations of Cow/calf Pairs in Laguna San Ignacio, Mexi co. 1§5 > o c CT O) LEGEND B Cow/calf pairs 1983 D Cow/calf pairs 1984 o o o o o o o o o «— CNI n •q- in co oo o o o o o o o o o o o < — m to N oo o > o o I I I I I I I I I O) o c \ i o o » a - L O c o r ^ o o a ) A Duration of Dive (seconds) 196 Figure 41 . 1983/1984 Frequency Distribution Comparisons of Dive Durations on Single Whales in Laguna San Ignacio, Mexico. 197 O O O o ' — C N co ^ I I I I *— CN CO 0 0 0 0 0 0 0 0 0 0 o o o o o o L O C D r ^ O O O J O ' - C N C O ' a - L r j C D r ^ o O a j O O 0 ) 0 ' — C N C O " y u D C D r ^ 0 O < D A Duration of Dive (seconds) 198 time was spent tracking whales. During the playback of ki l ler whale sounds, only 7% of the time was spent tracking whales. A breakdown by the number and types of whales, as well as the time spent per experiment tracking whales, is shown in Figure 42. During control periods, whales typically were observed to travel directly through the study area with l i t t le evidence of milling taking place. The trackline depicted in Figure 43 demonstrates a typical movement pattern by a whale past the Rocky Point shore-station during control periods. The average perpendicular distance calculated offshore exhibited by these whales was 230.4 ± s.d. 69.4 meters. On the average, observers spent 6.4 + s.d. 4.0 minutes observing each whale. With real sources prevalent in the environment, whales once again showed a fairly direct trackline through the study area (Fig. 44). The average perpendicular distance travelled offshore was 243.9 ± s.d. 70.9 meters, which was not significantly different from control periods (t = 0.86; p > 0.05). The average time spent tracking a whale was 11.0 + s.d. 9.5 minutes, a value similar to that obtained during control periods (t = 0.42; p > 0.05). During the projection of outboard engine noise, the majority of whales were observed to transit directly through the study area; however, occasionally whales headed directly to the transducer site and were apparently attracted to the projected sounds (Fig. 45). The average perpendicular distance offshore was 261.5 + s.d. 92.0 meters. This value was found to be similar to that obtained during control periods (230.4 + s.d. 69.4 meters) (t = 0.39; p > 0.05). During outboard engine playback, observers spent an average of 13.2 + 9.6 minutes/whale, a value similar to that observed during control periods (t = 1.78; p > 0.05). 199 F i g u r e 42. Percentage o f Time Spent T r a c k i n g Whales Per E x p e r i m e n t a l C o n d i t i 201 Figure 43. Trackl ine of Whale During Control Per iods. Whale movement r ight to l e f t . 300 1' " ' " I I I 1 1 1 1 1 1 1 1 1 1 ! . „ . . ! . . 1 1 1 1 : i CONTROL—TRACKLINE _ i j _ _ i i i i ' i i i i 1 i r ] — —1 ^ . . . i i i i > V I 1 7 i i 1 i I —i i i i i 1 | \ ! ! ^ Rocky Point Shore Station ! ', i i i i i i i r- -i i i j i i 900 600 300 0 300 600 900 Distance south and north of shore station (meters) 203 Figure 44. Trackline of Whale in the Presence of Real Sources. Whale movement right to left. 9 0 0 - -R E A L SOURCES—TRACKLINE Rocky Point Shore Station -T-I 900 600 300 0 300 600 Distance south and north of shore station (meters) 900 205 F i g u r e 45. T r a c k l i n e of Whale D u r i n g Outboard Engine Playback P e r i o d s . Whale movement r i g h t t o l e f t . 9 0 0 - -CO •»-' CO ~ 600-CO CO o c CD 3 0 0 - - -T T" i i i i i i i . 1 . OUTBOARD—TRACKLINE Rocky Point Shore Station 900 600 300 0 300 600 Distance south and north of shore station (meters) 900 NO o en 207 During oi1-dri11ing playback periods, an increase was noted in the number of whales milling (64%) in the area vs travelling (36%) through the area (Fig. 46). The average offshore distance travelled by these whales was 238.7 + s .d. 108.7 meters. When distances were compared between this experimental condition and control periods (230.4 + s.d. 69.4 meters), a significant difference was not detected (t = 0.2; p > 0.05). An average of 13.6 +• s.d. 10.1 minutes/whale was spent by observers tracking whales during the presence of oi l -dr i l l ing sounds. This value was similar to control periods (t = 1.48; p > 0.05). During ki l ler whale playback periods, overt changes were documented in the movements of whales. Eighty-six percent (86%) of the whales tracked were observed to mi 11 in the area. Typically a progressive movement through the study area was not documented (Fig. 47). Of those whales observed to travel through the area, the average offshore distance maintained by these whales in the presence of ki l ler whale sounds was 335.1 + s.d. 72.5 meters (Fig. 48). This offshore distance was significantly different than that obtained during control periods (t = 3.21; p < 0.05). Since whales were positioning themselves in one location for long periods of time and/or the fact that whales farther offshore took a longer time to transit through the study area (greater distances involved), the average time spent tracking a whale increased to 36.0 + s.d. 14.7 minutes. When this value was compared to control periods, a significant difference was noted to occur (t = 2.3; p < 0.05). A comparison of the various tracklines obtained during these experiments yielded the following results. Significant differences were not detected between control and vessel noise (real and outboard engine) in the distance travelled offshore, amount of time spent by the whale in 208 Figure 46. Trackline of Whale During Oil-dri l l ing Playback Periods. Whale movement left to right. 9 0 0 H -E 600 -4 CO c 300-03 •*-> 900 600 "1 OIL-DRILLING—TRACKLINE i i +-i Rocky Point Shore Station i I i i i i i 300 0 300 3 600 900 Distance south and north of shore station (meters) 210 Figure 47. Trackl ine of Whale During K i l l e r Whale Playback Periods (no progressive movement through area). cu cu E o CO CO o c 6 0 0 -300-0-K I L L E R W H A L E — T R A C K L I N E j.. _.._L___ Rocky Point Shore Station 900 600 300 0 300 600 Distance south and north of shore station (meters) 900 H9 212 Figure 48. Trackline of Whale During Killer Whale Playback Periods (depicting increased distance offshore). Whale movement left to right. 900-cu CO ~ 600-CO o c CD 300-KILL ER WHALE —TRACKL INE i — _ _ i _ i — « - " ' i i i i i r 1 i — - l  i i • i i i 1 i— L ^ Rock j 1 1 y Point Shore Station | i i - • N i i i i i i i i O J 900 600 300 0 300 600 Distance south and north of shore station Meters) 900 214 the study area, and/or in the percentage of whales travelling vs milling. Oil-dri l l ing playback periods did not el ic i t a change in the distance travelled offshore or in the amount of time a whale spent in the area. However, an appreciable increase in the number of whales observed to mill in the study area in the presence of oi l -dr i l l ing sounds was documented. Killer whale playback caused the greatest change to occur and, when compared to the other three experimental periods and the control periods, a significant increase was documented in: 1) distance travelled offshore, 2) amount of time spent in the area, and 3) number of milling whales. The values calculated for these tracking experiments are listed in Table XIX. 1984 Observed Surface Behavior - Transect Experiments Seasonal Comparisons (Abundance/Distribution/Movements) Four pair transects were completed during the 1984 season. Oil-dri l l ing sounds were projected back on 6 February and 9 February 1984, and ki l ler whale sounds were projected during the surveys conducted on 13 February and 20 February 1984. In addition to investigating the abundance and distribution of whales between control and experimental transects, the direction and milling rates of the whales were also compared between the paired transects. A comparison of the total number of whales between control and experimental conditions inhabiting the waters of Parmenter Point to Yucca Plant resulted in a significant reduction of whales occurring in the area during experimental transects. On 6 February 1984, during control surveys, 73 whales were counted in these waters. During experimental periods, the number of whales occupying these waters decreased to 58. When Table XIX. 1984 Trackline Data on Gray Whales Occurring in Laguna San Ignacio, Mexico (Mean and standard deviations). Average Perpendicular Distance Offshore (m) Amount of Tracking Time (mi nutes) Percentage of Whales Travelli ng Percentage of Whales Milli ng CONTROL n = 466 dives 230.4 + 69.4 6.4 ± 4.0 92.0 8.0 REAL SOURCES n = 496 dives 243.9 + 70.9 11.0 + 9.5 91.0 9.0 OUTBOARD ENGINE n = 89 dives 261.5 + 92.0 13.2 + 9.6 90.0 10.0 OIL DRILLING n = 174 dives 238.7 + 108.7 13.6 + 10.1 36.0 64.0 KILLER WHALE n = 62 dives 335.1 + 72.5 36.0 ± 14.7 14.0 86.0 216 compared, a significant difference was obtained (Chi Square = 4.3; p < 0.05), with fewer whales occurring during experimental periods. On 9 February 1984, 63 whales occupied the, area during control periods. In the presence of oi l -dr i l l ing sounds, the count dropped to 36 whales. Experimental values were significantly lower than those obtained during control periods (Chi Square = 13.5; p < 0.05). On 13 February 1984, 80 whales were observed during control surveys and 49 whales were counted during the experimental transect. Once again, a significant decrease in the number of whales occurring in the area was documented during playback periods (Chi Square = 19.6; p < 0.05). On 20 February 1984, a similar trend was documented. During controls, 67 whales were present. In the presence of ki l ler whale sounds, this value dropped to 52 whales. A significant decrease occurred during experimental conditions (Chi Square = 4.06; p < 0.05). These data are graphically shown in Figure 49. The results from these comparisons document a decrease in the total number of whales occupying the waters of Parmenter Point to Yucca Plant, which indicates a shift in overall distribution by the whales in the presence of noise. Assuming whales were potentially disturbed by increased levels of noise in the area, a movement away from the sound source would be expected. The analyses on whale abundance accomplished above indicate that this was indeed the case. An inspection of the direction of heading of the whales should then support these results. For example, i f whales were located between Parmenter Point and Rocky Point (south of station), a southerly movement by the whales would 1) provide the best means of avoiding the sound source and 2) prevent the whale from being exposed to a higher source level of sound. Conversely, i f whales were occupying the 217 Figure 49. A Comparison of the Total Number of Whales Occupying the Waters Between Parmenter Point and Yucca Plant During Control and Experimental Condit ions. 2*18 219 waters between Rocky Point and Yucca Plant (north of station), one would expect a greater number of northbound whales to be counted during the experimental periods (movement away from source). Data collected on the direction of whales during vessel surveys were compared between control and experimental events to verify this fact. In the presence of oi l -dr i l l ing sounds (6 and 9 February - data pooled due to small sample sizes collected per experiment), a significant increase occurred in the number of southbound whales counted between the Rocky Point and Parmenter Point area (Chi Square = 17.2; p < 0.05). During control periods, four whales were noted as heading south. During experimental surveys, this was increased to 26 whales. A similar pattern was observed during the projection of kil ler whale sounds. During control periods (13 and 20 February) two whales were noted as southbound, whereas during playback periods, the number of southbound whales increased to 13. A significant increase in the number of southbound whales (Fig. 50) occurred during this experimental playback period as well (Chi Square = 7.2; p < 0.05). An inspection of the number of northbound whales between Rocky Point and Parmenter Point was also accomplished. One would expect that fewer northbound whales would be observed during the experimental surveys, i f whales were avoiding the sound stimuli. Statistical comparisons supported this assumption. On 6 and 9 February (oil-dri l l ing playback) 14 whales were noted as northbound during control periods. During experimental surveys, only one whale was observed northbound. A significant difference was, of course, obtained when these values were compared (Chi Square = 10.3; p < 0.05). The surveys conducted on 13 and 20 February (killer whale playback) also documented that fewer northbound 220 Figure 50. A Comparison of the Number of Southbound Whales Occurring Between Rocky Point and Parmenter Point During Control and Experimental Periods. 2 2>1 24-i 2 2 -2-6-84 2-9-84 2-13-84 2-20-84 Date of Surveys 222 whales occurred during experimental conditions (14 whales counted during control and eight during experimental phase of transect - F ig . 51). However, when these values were compared, a significant difference could not be obtained (Chi Square = 1.27; p > 0.05). Similar comparisons were accomplished on the number of southbound and northbound whales occurring between Rocky Point and Yucca Plant. If whales were avoiding the immediate study area, fewer southbound whales would be expected during playback periods. Conversely, a greater number of northbound whales should occur. On 6 and 9 February, 24 whales were observed southbound during control periods. During experimental transects, this number was 21. Although fewer whales were observed heading south during the experimental transect, this value was not significantly different than that obtained during control transects (Chi Square = 0.11; p > 0.05). When the paired transects conducted on 13 and 20 February were compared (control counts were 40 southbound whales vs 13 southbound whales during experimental transects), a significant difference was obtained (Chi Square = 17.3; p < 0.05), with fewer southbound whales observed during experimental conditions (Fig. 52). An examination of the number of northbound whales between Rocky Point and Yucca Plant suggested that more whales headed north during playback periods (Fig. 53). On 6 and 9 February, 15 whales were northbound during control periods vs 33 counted during experimental transects. When compared, these values were found to be significantly different (Chi Square = 7.9; p < 0.05). On 13 and 20 February, nine northbound whales were observed during control surveys. This number increased to 37 during experimental conditions. A greater number of northbound whales was documented during experimental conditions in 223 Figure 51. A Comparison of the Number of Northbound Whales Occurring Between Rocky Point and Parmenter Point During Control and Experimental Periods. 2^2 4 24-LEGEND Control transects Experimental transects | 18-CU +-. xi c 2-6-84 2-9-84 2-13-84 2-20-84 Dates of surveys 225 Figure 52. A Comparison of the Number of Southbound Whales Occurring Between Rocky Point and Yucca Plant During Control and Experimental Periods. 226 Dates of surveys 227 Figure 53. A Comparison of the Number of Northbound Whales Occurring Between Rocky Point and Yucca Plant During Control and Experimental Periods. Number of northbound whales observed between Rocky Point and Yucca Plant 229 the presence of ki l ler whale playbacks (Chi Square = 20.5; p < 0.05). On 6 February 1984 (during control conditions), a total of 73 whales was noted to occur between Parmenter Point and Yucca Plant. Of these, eight were observed either heading north or south in the waters between Rocky Point and Parmenter Point. Examining the total number of southbound vs northbound whales between Rocky Point and Yucca Plant yielded a value of 23 whales. Thus, for the whole area only 42% (31 vs 73) of the whales could be accounted for either actively moving south or north. This suggested that the additional 42 whales must have been scored as either milling (No Direction = N.D.) or Direction Unknown (D.U.). Conversely, an examination of the total number of whales occurring through the whole area on 6 February during experimental transects gave a value of 58. Of these, 93% (n =54) could be accounted for when comparing the total number of north vs southbound whales for the two areas (Rocky Point to Parmenter Point = 19 whales and Rocky Point to Yucca Plant = 35 whales ). These data suggested that a greater number of whales was actively moving during experimental conditions. To confirm this, comparisons were made between control and experimental conditions comparing the number of milling whales (scored as N.D.). On 6 February, considering the whole area (Parmenter Point to Yucca Plant), 28 whales were observed milling during control transects. This value declined during experimental testing to 17 whales. When compared, significant differences were obtained (Chi Square = 2.85; p < 0.05), with fewer whales milling during experimental conditions. On 9 February, 31 whales were noted as milling during control periods vs 13 during experimental transects. Once again, fewer whales were observed to mill in the presence of increased levels of noise (Chi Square = 8.42; p < 0.05). 230 Similar trends resulted when comparisons were completed for the experiments conducted on 13 and 20 February. On 13 February, 35 whales were scored as milling during control surveys. During experimental surveys, only 18 whales milled, a value significantly lower than that obtained during control surveys (Chi Square = 6.57; p < 0.05). On 20 February, 37 whales milled during control transects and 23 whales were observed exhibiting this behavior during experimental periods (Fig. 54). When compared, a significant decrease in milling rates occurred during experimental conditions (Chi Square = 4.02; p < 0.05). The results obtained for the paired transects documents 1) an overall reduction in the number of whales occurring in the area during playback periods, 2) active movement away from the sound source, 3) change in distribution, and 4) decrease in milling rates. Between-Season Comparisons - - Abundance of Whales Two full transects were completed on I February 1984 (prior to any playback) and on 2 March 1984 (after numerous playbacks had occurred). The total number of whales counted for each transect was compared to the five-year data base collected by Jones and Swartz (1984). Descriptive statistics (means, standard deviations, 95% confidence intervals) were calculated for the counts of whales collected on or about 1 February for the 5-year data base. Separate computations were completed for 1) all whales, 2) cow/calf pairs and 3) single whales. Similar statistics were calculated on or about 2 March for the 5-year data base as well. These calculations are summarized in Table XX. The counts obtained during the acoustical investigations (1 February and 2 March 1984 - ful1 transects) were compared to the 95% confidence 231 Figure 5 4 . A Comparison of the Number of Milling Whales Observed Between Parmenter Point and Yucca Plant During Control and Experimental Periods. 2 3 2 c cu cu 5 +-> cu X) u CO 4- i > c cu JS O tO o S3 » « > 5 ? O J CO . E: +-i zz c C CL o QJ CO E -5 s Z Q . c CO E 80-70-60-5 0 -4 0 -30-20-10-0-LEGEND Control transects Experimental transects 2-6-84 2-9-84 2-13-84 2-20-84 Dates of surveys 233 Table XX. Counts of Whales Occupying Laguna San Ignacio on or About 1 February and 2 March for the Years 1978-1982 (from Jones and Swartz, 1984). ALL WHALES SINGLE WHALES COW/CALF PAIRS FEBRUARY 1978 1979 1980 1981 1982 Mean and Standard Devi ati on 95% Confidence Interval 290 159 215 292 179 227.0 + 61.7 169.1 - 284.9 228 95 108 198 104 146.6 + 61.7 88.7 - 204.5 62 64 107 94 75 80.4 ± 19.5 62.1 - 98.7 2 MARCH 1978 1979 1980 1981 1982 131 310 231 193 267 37 209 69 11 63 Mean and Standard 226.4 + 68.6 Devi ation 77.8 + 76.8 94 101 162 182 204 148.6 + 49.0 95% Confidence 162.0 - 290.8 5.6 - 150.0 102.7 - 194.5 Interval 234 intervals calculated for the years 1978 to 1982. The 1 February 1984 count of all whales was 169. When compared to the confidence interval obtained for the 5-year data base (169.1-284.9), a significant difference in abundance of whales could not be detected. When the 1984 February count of single whales was compared with the confidence interval obtained for 1978-1982 (88.7-204.5), the 123 single whales observed in 1984 fell within the 5-year limits of the confidence interval. The number of cow/calf pairs counted on 1 February 1984 was 46. This value was found to be significantly lower than the 5-year average (62.1-98.7). On 2 March 1984, a total of 43 whales was observed occupying Laguna San Ignacio. This value was significantly lower than the average derived from the 5-year data base (162.0-290.8). When single whale counts for 1984 (19 whales) were compared to the 5-year confidence interval (5.6-150.0), the 1984 count fell within the limits of testing. Single whale counts were not significantly different from the 5-year mean. Conversely, when the 2 March 1984 cow/calf pair count of 24 was compared to the 5-year data base (102.7-194.5), a major reduction in the number of cow/calf pairs occupying the lagoon in 1984 was documented. The overall results of these abundance comparisons suggested that although single whales were somewhat on the low side, the counts of these whales were not statistically different from previous years. The cow/calf pair count in 1984 was, however, significantly reduced when compared to previous years (Figs. 55, 56A and 56B). 235 Figure 55. Abundance of Gray Whales Occurring in Laguna San Ignacio, Mexico, for the Years 1978-1982 and 1984. 2§6 LEGEND T i i r—i—m—i—i—i—i—r-n—i— i— i— i—|—rn—I—i—I 5 10 15 20 25 30|I 5 10 15 20 25|' 5 10 15 20 25 30|' 5 10 15 January February March April Date of observation 237 Figure 56. Abundance of Single Whales (A) and Cow/calf Pairs (B) Occurring in Laguna San Ignacio, Mexico, for the Years 1978-1982 and 1984. 2 378 A 300-i Single Whales LEGEND O) c 3 O o l/l -2? 10 0) _Q E 3 i i i i i n r ~ i i i — i — i — n — i — i — i — i — i m i 5 10 15 20 25 30i 5 10 15 20 25 J 5 10 15 20 2530] 5 10 15 January February March Date of observation April 239 DISCUSSION The acoustical repertoire of gray whales occupying Laguna San Ignacio was significantly modified as noise levels were increased in their environment. During the 1983 and 1984 field seasons, gray whales were subjected to a variety of acoustical stimuli. These noise sources varied with respect to 1) emphasized frequencies and decibel levels; 2) the way in which the sound stimulus was presented (rapid vs gradual onset); 3) exposure periods (short vs long-term); 4) the biological vs non-biological origin of the source; and 5) their regular occurrence in the lagoon habitat vs novel stimuli for the particular area. Although distinctly different from one another, the overall effect of these noise stimuli on the acoustical behavior of the gray whale was similar for all experiments in that the same acoustical variables were observed to change during experimental test periods. As noise levels purposely were increased, associated changes were documented in: 1) calling rates; 2) levels at which signals were produced; 3) percentage of signals exhibiting frequency modulation; 4) number of pulses produced per series; and 5) average repetition rates of calls. As described in Chapter 2, these modifications in the acoustical behavior of the gray whale during periods of increased levels of noise would serve to enhance signal transmission and reception. The fact that five of the nine acoustical variables measured were observed to change in the presence of noise implies a degree of f lexibil ity in the acoustical behavior of the whales and a capability of circumventing a variety of ambient conditions. Thts acoustical f lexibil ity may be required to cope with the dynamic ambient conditions encountered in coastal habitats. In addition, i f other forms of communication are of 240 limited use in the aquatic environment (e.g., visual and/or chemoreception) then sufficient plasticity in the dominant sensory mechanism would be requi red. The comparisons of the amount of change observed among experiments between the two seasons provided additional information and valuable insight into the acoustical capabilities and strategies of these whales when exposed to various sound stimuli present in their environments. For example, when comparing the acoustical responses of gray whales to outboard engine noise projected during the 1983 and 1984 studies, calling rates and received levels of signals were significantly higher in 1983. The 1983 results also documented significant differences among various experiments in gray whale calling rates and call structure. Conversely, in 1984, comparisons of experiments did not yield significant differences in thes,e acoustical variables. These collective results imply that gray whales vary their call behavior dependent upon 1) type of signal present; 2) duration of signal; and 3) presentation of signal. If we assume that the 1983 acoustical responses to the projection of the rapid onset of short-term noise could be labeled as a startled response, then the following conclusion could be drawn when comparing the 1983 and 1984 seasons. When startled, gray whales will actively modify their acoustical behavior. The types of changes observed are similar to those reported for the gradual onset of sounds; however, significant differences are apparent in the degree of change observed within the particular acoustical variable. A higher degree of change is documented during startle responses. Of particular interest is the fact that the 1983 acoustical responses did adequately reflect this species' acoustical capabilities and provided additional information on the range of their capabilities. 241 The novel stimuli (sound types that do not occur typically in the lagoon habitat) projected during these studies (test tones, oi l -dr i l l ing and ki l ler whale sounds) did, however, result in a radical departure in the acoustical behavior of the gray whale. In the presence of oi l -dr i l l ing sounds (although similar changes were observed to occur in call structure as described under vessel responses), overall calling rates were significantly reduced. The projection of test tones and ki l ler whale sounds had the most profound effect on the acoustical behavior of this species. In the presence of these two sources, all calling ceased. A preliminary interpretation of these results would predict that a significant reduction in calling and/or a complete cessation of calling would be detrimental for an acoustically dependent animal. However, a further examination and detailed interpretation of these results suggested that in some situations signal reduction could be beneficial. The benefits of the prey (= gray whales) remaining silent in the presence of predatory (= ki l ler whale) sounds are obvious. If gray whales transit undetected through an area occupied by these predators, then certainly this acoustical strategy provides a selective advantage to the prey. This silent behavior by gray whales, while actually being pursued by kil ler whales in the Bering Sea, has also been reported by Ljungblad and Moore (1983). When confronted with unknown sound sources (e.g., test tones), possibly the best strategy is to remain silent. When reviewing the responses of this species to oi l -dr i l l ing sounds projected in Laguna San Ignacio, other factors were considered. Due to the extensive range of gray whales and the fact that their preferred habitat has exposed 242 them to a wide variety of i ndus t r i a l l y - re l a ted sources, pr ior exposure to o i l - d r i l l i n g sounds was assumed. However, t h i s stimulus was novel in that gray whales do not t yp i ca l l y encounter th is type of noise in t h e i r calv ing locat ions. Ca l l i ng rates were reduced s ign i f i can t l y in the presence of th is s i g n a l ; however, when they did c a l l , s ignals were modified to ensure adequate t ransmission. This reduction in c a l l i n g rate in the presence of o i l - d r i l l i n g sounds may represent an addit ional strategy in the random timing of the projected s i g n a l . Ca l l reduction would, of course, provide cer ta in benefi ts in temporary s i t ua t i ons . However, i f prolonged, c a l l reduction could have a deleter ious ef fect on an acoustically-dependent animal. The overal l resul ts obtained during the 1983 and 1984 f i e l d seasons in Laguna San Ignacio also provided addit ional ins ight into the re la t ionsh ip between gray whale signals and the i r natural ambient environments. When comparing s ignals produced by the gray whale between the northern and southern range, s i gn i f i can t di f ferences in ca l l i ng rates and c a l l structure were obtained (Chapter 1). When the natural ambient condit ions of the northern and southern range were compared (Chapter 1), overt di f ferences were also documented. Laguna San Ignacio i s characterized by h igh- levels of b io log ica l noise, whereas the Bering Sea represents a low-level ambient environment. If , for example, these two areas are treated as a control (= Bering Sea) and experimental (= Laguna San Ignacio) hab i ta t , then responses of gray whales to d i f ferent natural ambient condit ions can be evaluated. When making comparisons of th is nature, cer ta in ly the overa l l behavior of the animal must be considered. However, i f we assume that e f fec t i ve sound communication would evolve independently 243 of the behavioral s ign i f icance of the c a l l , we would then predict that these c a l l s would be structured to optimize the i r transmission and reception independent of behavioral requirements. When treated in th is manner, va l id comparisons can be made between sound structures and c a l l i n g rates between the two areas. When these data were compared to those obtained during the 1983 and 1984 invest igat ions in Laguna San Ignacio, an important and in terest ing pattern emerged. C o l l e c t i v e l y , when exposed to h igh- level noise s i tuat ions ( i . e . , natural ambient condit ions on calv ing grounds and/or experimental cond i t ions) , s im i la r acoust ical variables were changed. A l l of these changes would benefi t the c a l l e r through sound enhancement. Two addit ional var iables were also modified when comparing c a l l structure between the Bering Sea and Mexico. Ca l l durat ion, although not documented to change in the presence of temporary noise sources (experimental), was s i g n i f i c a n t l y longer on the southern range vs the northern range. Cer ta in l y , increasing ca l l duration while inhabi t ing high-ambient environments would be benef ic ia l to the c a l l e r . An inspection of the frequency ranges emphasized by these whales also varied between the northern and southern range, with broader ranges observed on the northern grounds. This suggests that the high- level ambient condit ions in Laguna San Ignacio, emphasizing the frequency bands of 2-5 kHz, may have affected and/or l imi ted the frequency output of s ignals produced by the gray whale. These data imply addi t ional f l e x i b i l i t y in the acoust ical behavior of the gray whale and suggest that these whales may employ d i f ferent strategies when exposed to constant vs temporary h igh- level noise s i tua t ions . When attempting behavioral comparisons, a l l environmental factors should be considered and s ta ted . For example, the high repet i t ion rates and number of pulses produced per 244 series collected during the northern research by Moore and Ljungblad (1984) may have resulted from the effect of aircraft noise on the whales. The combined results documented during these studies support the acoustical niche hypothesis and demonstrate that environmental noise (either natural and/or man-made) directly influences and shapes the acoustical behavior of this coastal species. A study of the effect of noise on wildlife would not be complete unless a brief discussion on sound reception (hearing sensitivities) and sound-producing mechanisms was included. Unfortunately, l i t t le quantitative information is available on the auditory sensitivities of the marine mammals. Audiograms have been published for a select group of Odontocetes which include: the bottlenose dolphin, Tursiops truncatus (Johnson, 1966); the harbor porpoise, Phocoena phocoena (Andersen, 1970); the kil ler whale, Orcinus orca (Hall and Johnson, 1971); and the Amazon River dolphin, Inia geoffrensis (Jacobs and Hall, 1972). In most cases, however, only single animals from each species were studied. Information pertaining to the hearing sensitivities of the mysticete whales is non-existent in the published literature. Despite the fact that sufficient information is not available regarding hearing sensitivities in gray whales, certain assumptions can be made as to how and to what types of sounds are perceived by these whales. One can assume that gray whales hear sounds in the same frequency in which they produce sounds (100 to 2000 Hz). In most mammals studied, hearing range is typically broader than the vocalization range and we could assume that the gray whale is not an exception here. Actual sound-receiving mechanisms are not known in the baleen whales. In odontocetes, sound is apparently received through the fat-f i l led lower jaw and transmitted 245 to the internal ear. In mysticete whales, however, the lower jaw is solid. As demonstrated during these studies, gray whales were very capable of localizing underwater sound sources, as evidenced by their direct movements to the transducer location during outboard engine playback periods. This, of course, implies biaural and directional hearing. During the 1984 studies, an attempt was made to gather information on the hearing capabilities of gray whales. The primary objective of this work was to determine the feasibility of conducting hearing studies on a mysticete whale in its natural environment. The methodology employed holds some promise and the results of these preliminary investigations are reported in Dahlheim and Ljungblad, in prep. Although l i t t le is known on the sound-producing mechanisms in mysticete whales, there is general agreement that the site of sound production in the baleen whales is the pharyngeal pouches. In odontocetes, it is generally accepted that sound is produced through the extensive nasal sacs, a feature non-existent in baleen whales. The data suggesting that baleen whales produce sounds through the pharyngeal pouches are derived from studies conducted on sei whales (Balaenoptera borealis) and fin whales (Balaenoptera physalus), both representatives of the Family Balaenopteridae (Benham, 1901; Schulte, 1916; Hosokawa, 1950). Unfortunately, these structures have not been examined in gray whales and it may not be valid to extrapolate anatomical features between different families of whales. Although gray whales are capable of acoustically responding in a variety of ways, as demonstrated during these studies, there must be some physical (either anatomical and/or physiological) limitation regarding the nature and extent of these call modifications. In addition, there also 246 must be some level of sensitivity to particular sound sources due to the hearing sensitivities and/or past experience of these whales to the noise source. This level of tolerance/sensitivity may also vary with respect to 1) the individual; 2) group size; 3) age; 4) reproductive status; 5) behavior; and 6) part of the range occupied by this species. In addition to changes in call behavior in response to noise, changes were also documented in other aspects of their behavior. An inspection of the dive durations collected on gray whales during the 1983 and 1984 seasons in Laguna San Ignacio demonstrates the wide range of diving capabilities of this species. In 1983, as well as in 1984, significant differences were detected between the dive durations of females with accompanying calves vs the single whale component of the population. Apparently, mothers with young adjust the length of their dive to accommodate their young calves, resulting in shorter dive times. The dive durations exhibited by these adult females may change over time ( i .e. , increase in duration) with maturation of the calf. Comparisons conducted on the dive durations of cow/calf pairs and single whales in Laguna San Ignacio between control and experimental conditions did not detect significant differences when these whales were exposed to increased levels of noise. During recent studies by Malme et a l . , 1983 and 1984 addressing surface responses of gray whales to industrial noise along this species' migration route and feeding grounds, the following results were obtained. Blow rates of gray whales were not significantly different off California when comparing undisturbed vs disturbed periods. During these California experimental periods a variety of industrial sounds was projected, including noise from drillships, semi submersible dri l l rigs, dril l ing platforms, production platforms, helicopter noise and ki l ler 247 whale sounds. Conversely, when conducting similar experiments in the Bering Sea, a significantly longer dive time was obtained in the presence of drillship playbacks. However, during the projection of air-gun sounds, detectable differences were not found in the dive durations of these northern whales (Malme et a l . , 1986). The extent of the diving capabilities of the gray whale is shown when the dive behavior of this species is compared between different parts of its range. Wursig et a l . (1983) reported an average dive duration of 190.8 + s.d. 76.2 sec for feeding gray whales in the Bering Sea, a value significantly longer than that reported for other areas. During migration, Sumich (1983) noted an average dive length of 43.2 sec (0.72 breaths per min). When these data are compared to those collected on this species' southern range (this study; Harvey and Mate, 1984), marked differences are apparent. These differences may suggest that the overall behavior and/or habitat occupied by these whales dictates the dive profile. In any case, the wide range observed in the dive cycles of these whales is obvious upon inspection of these data. In addition to dive duration rates, several other parameters have been used by researchers to describe the dive cycle of the gray whale (e.g., blow interval, number of blows per surfacing, and duration of surfacing). The fact that the 1984 dive durations of cow/calf pairs collected in Laguna San Ignacio were significantly longer than those collected in 1983 is somewhat perplexing. Assuming cow/calf pairs would be more sensitive than single whales to changing environmental conditions, an observable change could be expected. Unfortunately, it is not clear if these longer 1984 dive times of cow/calf pairs were triggered by the presence of long-term noise sources projected into their environment in 1984 and/or resulted from other, inherent factors. 248 Additional data would be required to make a definitive statement and to determine the amount of variability of respiration rates between seasons. When inspecting the 1984 tracklines of gray whales in Laguna San Ignacio, despite the wide variety of behaviors exhibited by this species on its calving grounds, certain behavioral trends were documented in the presence of noise. Increased levels of noise were shown to have a direct effect on the travelling vs milling rates of these whales. This increase in milling by whales in the immediate study area may have resulted due to the close proximity of the whales to the sound source. In this situation, whales may have become confused, thus stopping all progressive movement past the sound source. Although precise tracklines were not collected in 1983, the increase in the number of directional changes observed during experimental periods suggests changes in movements and supports the 1984 tracking results. During studies conducted off central California (Malme et a l . , 1983, 1984), milling rates of migrating gray whales were also observed to increase in the presence of industrial playbacks. Similar work was conducted by this author in the Bering Sea, but unfortunately the sample size collected by Malme et a l . (1986) in the same area was limited and valid comparisons could not be made. In addition to the increase noted in milling rates of gray whales in the presence of ki l ler whale sounds, those whales actually travelling through the study area were noted to increase their distance from the sound source. The distances dealt with during these tracking experiments were restricted due to the visual limitations of the observers and initially were not that great. Possibly this fact was responsible for not detecting significant differences in the offshore distances of these whales to various experimental conditions. The results of these tracking experiments 249 noted that whales were, in fact, reacting to the sound source by changing their behavior and/or avoiding the area. In 1983, during post-trial periods, a significant reduction was noted to occur in the number of whales occupying the immediate study area. During these post-trial periods, whales were scarce and those that were present had unpredictable dive times, making them harder to track. In 1984, the results obtained from the effort log also suggested that whales were avoiding the immediate study area during certain experimental periods. Gray whale abundance and distribution were quantitatively examined during the transects conducted in 1984. These data supported the results obtained during the tracking experiments. The paired transects conducted in 1984 verified that whales were avoiding the immediate study area during oi l -dr i l l ing and ki l ler whale playback periods. During these playback periods, overall gray whale abundance in the study area decreased, reflecting a distributional change by these whales; directional changes were documented to occur during experimental transects, with an overall movement away from the sound source noted; and milling rates (outside the boundaries of the immediate study area) were observed to decrease. Milling rates obtained during these tracking and transect experiments were examined and yielded interesting results. During tracking, whales in the immediate area were observed to increase their milling rates during playback periods, whereas during transect surveys, whales outside the immediate study area exhibited a decrease in milling rates. Comparisons suggested that the distance of the whale relative to the source could be responsible for the different behavioral responses noted in these milling rates. These data may also be indicative of the hearing sensitivities/level of tolerance exhibited by these whales. 250 Between-season abundance comparisons (1978-1982 vs 1984) documented a significant reduction in the number of cow/calf pairs occupying Laguna San Ignacio in the 1984 season. The long-term playback experiments conducted during the 1984 season were shown to cause small scale and temporary changes in gray whale abundance, distribution and behavior. The effect of the long-term playbacks conducted in 1984 pertaining to the large scale impacts are not as clear. For example, the 1 February surveys, prior to any playback, demonstrated that the counts of cow/calf pairs were statistically lower than previous years, indicating that other factors may have been involved with respect to these low cow/calf pair counts. The 2 March count of cow/calf pairs, after repeated exposures to playback periods was, however, greatly reduced as compared to previous years. The zone of influence during the projection of the playback stimuli certainly extended beyond the art i f icial boundaries defined for the immediate study area. There is the possibility that whales outside the study area were also significantly affected; however, this does not seem reasonable when considering the transmission loss of these signals with range. In addition, other factors and their effects on whales must be considered. The amount of tour boat activity in Laguna San Ignacio has actually decreased over the years. Local traff ic, with respect to fishing efforts of.local people, has also decreased. In 1983, El Nino conditions prevailed along the Pacific coast. In 1984, residual effects of EI Nino were st i l l documented for the Eastern Pacific Ocean. It is unclear at present as to what effects, i f any, El Nino conditions would have on gray whales. Although the Laguna San Ignacio data base represents six years of data, this time window may not be adequate to predict the long-term natural fluctuations in gray whale 251 populations in the lagoon. The advantages of conducting long-term studies on these long-lived cetaceans are obvious. Considering the potential effect of these various factors on gray whales, and especially the fact that cow/calf pair counts were already reduced prior to any playbacks, a cause/effect relationship between the low counts and long-term playback experiments conducted in 1984 cannot be established. Due to the significant reduction of cow/calf pairs observed in Laguna San Ignacio during the 1984 season, a follow-up study addressing cow/calf pair abundance was considered necessary by the Marine Mammal Commission. In February and March of 1985, joint investigations were conducted to determine the overall abundance and distribution of whales in Laguna San Ignacio. The 1985 results are briefly addressed here due to their pertinence to this current study and are described in detail in Jones, Swartz and Dahlheim (1986). The transects conducted in 1985 (in the absence of acoustical work) documented an increase in the number of cow/calf pairs occupying the lagoon when compared to the 1984 season; however, this estimate was s t i l l 20% lower than the 5-year average. The conclusions based upon these acoustical investigations on gray whales have shown that ambient noise, either natural and/or man-made, has a profound effect on the acoustical behavior of this species. The sound behavior of the gray whale, documented during these studies, was modified in the presence of increased levels of noise, which was shown to enhance sound transmission and reception. Whale responses varied with respect to noise based on exposure time, the way in which the signal was presented, and the type of stimuli present in the environment. The plasticity observed in gray whale behaviors, when exposed to increased levels of 252 noise, enables this species to cope successfully with dynamic noise situations that typically characterize its coastal habitats. The hypothesis that gray whales, while engaged in underwater signalling, circumvent noise in their acoustical channels by the structure and timing of their calls has been accepted based upon the overall results obtained during these investigations. 253 GENERAL SUMMARY/CONCLUSIONS 254 GENERAL SUMMARY/CONCLUSIONS The results of this study have significantly increased our knowledge of the acoustical behavior of the gray whale. 1) Gray whales produce at least seven distinct sound types. Variations exist within sound categories; however, each call type is sufficiently distinct to be classified independently. 2) Sound activity increased on this species' mating/calving grounds with a corresponding increase observed in social activi ti es. 3) Similar call types were produced throughout the range; however, call structure showed more variation on the southern range. 4) Distinct differences were noted in the gray whales' acoustical habitats with low levels of natural ambient noise associated with the northern range (Bering Sea). Conversely, on the southern range (Mexico), high levels of natural ambient noise prevailed. Intermediate values of ambient noise were documented for the migration route (Washington State). 5) Numerous fish and invertebrates were identified and deemed potentially responsible for the high-levels of natural ambient noise in Mexico. The biological contribution to ambient conditions was reduced on the migration route and on the feeding grounds. 6) Each respective vessel/outboard engine recorded exhibited its own sound signature. Profiles of man-made noise did not vary with range. 255 Salinity, temperature and depth values were inspected and compared between Mexico and Alaska. As expected, major differences were noted in water temperatures and depths with range. However, salinity values were similar between the two geographical areas. Sound propagation in Laguna San Ignacio did not follow the cylindrical spreading law (10 log R). Sound transmission was enhanced in the southerly direction (lower lagoon) and was reduced in the middle lagoon due to the extensive sand bars causing shadow zones. All gray whale signals analyzed either occupied frequency bands below the main concentration of ambient noise (e.g., Laguna San Ignacio) or occupied different frequency bands than those emphasized by the natural ambient conditions (e.g., Bering Sea). The relationship between gray.whale signals and natural ambient noise indicates that a varying acoustical niche may be utilized by the gray whale on different parts of its range. Man-made sources (e.g., ship noise, outboard engine noise, oi l -dr i l l ing sounds, etc.) were shown to have a high potential of masking/interfering with whale signals. In the presence of artificially-increased levels of noise in Laguna San Ignacio, significant changes were documented in gray whale calling rates and call structure. 256 Typically, as noise levels were increased, a corresponding increase was documented in: calling rates, received level of the signals, number of frequency-modulated signals, number of pulses produced per series, and repetition rates of signals. In the presence of test tones and ki l ler whale sounds, gray whales ceased all signalling. Significant changes were also documented in the observed surface behavior of whales in response to increased levels of noise. Changes were documented in the number of whales occupying the immediate study area, overall distribution and movements. Dive durations varied between cow/calf pairs and single whales with cow/calf pairs exhibiting shorter dive times. Increased levels of noise did not influence dive patterns of whales. The level of response (degree of change) varied depending upon the sound source projected (e.g., outboard engine noise, test tones, gray whale sounds, kil ler whale sounds, oi l -dr i l l ing sounds). The way in which the stimulus was presented (rapid onset of sound vs gradual onset of sound) also dictated the level of response. Although whale responses varied among experiments, similar acoustical variables were consistently changed. Comparisons of gray whale responses between real sound sources present in the environment and art i f icial ly produced outboard engine noise resulted in similar findings. These data 257 support the use of playback techniques as an adequate method of experimentation in studies addressing noise effects on marine mamma 1 s. The flexibil ity observed in gray whale behaviors, when exposed to increased levels of noise, enables this species to cope with most dynamic noise situations that characterize coastal habitats. Combined results of this investigation lend substantial evidence for acceptance of the hypothesis: The gray whale, while engaged in underwater signalling, circumvents noise in the acoustical channel by the structure and timing of its calls. 258 REFERENCES Aldrich, H. L. 1889. "Arctic Alaska and Siberia". Rand McNally, Chicago, I l l inois. Pp. 32-35. Andersen, S. 1970. Auditory Sensitivity of the Harbour Porpoise, Phocoena phocoena. In: Investigations on Cetacea. Volume 2. G. Pil leri (ed). Hirnanat. Inst., Berne, pp. 255-258. Anonymous, 1974. Acquisition, reduction and analysis of acoustical data. An unclassified summary of acoustical working group studies NADC Report No. AWG-SU. Department of the Navy. Naval Air Development Center, Warminster, Pa. Anonymous, 1979. Study reports on low-frequency noise. Japan Environmental Summary. 7(10) :2-3. Arase, E. M. and T. Arase. 1966. Correlation of Ambient Sea Noise. J . Acoust. Soc. Am. 40:205. Armstrong, E. A. 1963. A Study of Bird Song. Oxford University Press, London. Asa-Dorian, P. V. and P. J . Perkins. 1 967. The controversial production of sound by the California Gray whale, Eschrichtius gibbosus. Nor-Hvalfangst-Tid. 56(4): 74-77. Au, W. W. L., R. W. Floyd, R. H. Penner, and A. E. Murchison. 1974. Measurement of echolocation signals of the Atlantic bottlenose dolphin, Tursiops truncatus, Montagu, in open waters. J . Acoust. Soc. Am. 56:4, 1280-1290. Au, W. W. L. 1980. Echolocation signals of the Atlantic bottlenose dolphin (Tursiops truncatus) in open waters. In: Animal Sonar Systems, pp. 251-282 (Ed. R. G. Busnel and J . F. Fish). Plenum Press. Au, W. W. L, D. A. Carder, R. H. Penner and B. L. Scronce. 1985. Demonstration of adaptation of beluga whale echolocation signals. J . Acoust. Soc. Am. 77 (2): 726 - 730. Bastian, J . , C. Wall, and C. L. Anderson. 1966. The transmission of arbitrary environmental information between bottlenose dolphins. In: Animal Sonar Systems - Biology and Bionics, Vol. II, pp. 803 - 873 (ed. R. G. Busnel) Laboratoire de Physio logieus tique, Jouy-en-Josas 78, France. Beamish, P. and E. Mitchell. 1973. Short pulse length audio frequency sounds recorded in the presence of a minke whale (Balaenoptera acutorostrata). Deep-Sea Res., 20: 375 - 386. Benham, W. B. 1901. On the larynx of certain whales (Kogia, Balaenoptera and Ziphius). Proc. Zool. Soc. (London) 1901, 1, 278-300. 259 Rlokhin, S. A. , M. K. Maminov and G. M. Kosygin. 1985. On the Korean-Okhotsk Population of Gray Whales. Rep. Int. Whal. Commn. 35: 375-376. Bogoslovskaya, L. S. 1982. On the behavior of gray whales off Chukotka and Koryaka. Intl. Whal. Commn. Doc. #SC/36/PS5, 16 pages. Bogoslovskaya, L. S. , L. M. Votrgov, and T. N. Semenova. 1982. Distribution and feeding of gray whales off Chukotka in summer and autumn 1980. Rept. Intl. Whal. Commn. 32: 385 - 389. Bogoslovskaya, L. S. , L. M. Votrgov, and T. N. Semenova. 1981. Feeding habitat of the gray whale off Chukotka. Rept. Intl. Whal. Commn. 31: 507-510. Braham, H. W. 1984. Distribution and Migration of Gray Whales in Alaska. In: The Gray Whale. M. L. Jones, S. L. Swartz, and S. Leatherwood, eds. Academic Press, New York. pp. 249-266. Breiwick, J . M. and M. E. Dahlheim. 1986. Preliminary Results from the 1985/86 Gray Whale Census. Working paper submitted to the International Whaling Commission, SC/38/PS 14, 10 pages. Breiwick, J.M., A. E. York and G. C. Bouchet. 1985. Preliminary Analysis of the 1984/85 Gray Whale Census Data. Working paper submitted to the International Whaling Commission, SC/37/PS 25, 5 pages. Breathnach, A. S. 1960. The Cetacean Central Nervous System. Biol. Rev. 35: 187-230. Brownell, R. L. and C. Chun. 1977. Probable existence of the Korean Stock of Gray Whales (Eschrichtius robustus). J . Mammal. 58 (2): 237-239. Bryant, P. J . , C. M. Lafferty and S. K. Lafferty. 1984. Reoccupation of Laguna Guerrero Negro, Baja California, Mexico, by Gray Whales. In: The Gray Whale. M. L. Jones, S. L. Swartz and S. Leatherwood (eds). Academic Press, New York. pp. 275-387. Calderon, M. A. 1964. Probability density analysis of ocean ambient and ship noise. Naval Electronics Laboratory Report # 1248, Nov. 1964. Caldwell, D. K. and M. C. Caldwell. 1972. Senses and Communication. In: Mammals of the Sea - Biology and Medicine. S. H. Ridgway (ed). Charles C. Thomas, Publisher. Springfield, I l l inois, pp. 466-502. Clark, C. W. 1982. The acoustical repertoire of the Southern Right Whale, a quantitative analysis. Anim. Behav. 30: 1060 - 1071. Clay, C. S. and H. Medwin. 1977. Acoustical Oceanography: Principles and Applications. John Wiley and Sons, New York. 544 pages. Cummings, W. C , P. 0. Thompson and R. Cook. 1968. Underwater sounds of migrating gray whales, Eschrichtius giaucus (Cope). J . Acoust. Soc. Am. 44(5): 1278-1281. 260 Cummings, W. C , J . F. Fish, and P. 0. Thompson. 1972. Sound production and other behavior of southern right whales, Eubalaena glacialis. Trans. San Diego Soc. Nat. Hist. 17(1): 1- 14. Cummings, W. C. and P. 0. Thompson. 1971. Underwater sounds from the blue whale, Balaenoptera musculus. J . Acoust. Soc. Am. 50 (4, pt. 2):1193-1198. Dahlheim, M. E. and F. Awbrey. 1982. A Classification and Comparison of Vocalizations of Captive Killer Whales (Orcinus orca). J . Acoust. Soc. Am. 72: 661-670. Dahlheim, M. E. 1981. A Review of the Biology and Exploitation of the Killer Whale (Orcinus orca), with Comments on Recent Sightings from Antarctica. Rept. Int. Whal. Commn. 31: 541-546. Dahlheim, M. E., H. D. Fisher and J . Schempp. 1984. Sound Production by the Gray Whale and Ambient Noise Levels in Laguna San Ignacio, Baja California Sur, Mexico. In: The Gray Whale. M. J . Jones, S. L. Swartz and S. Leatherwood (Eds). Academic Press, New York. pp. 511-541. Dahlheim, M. E. and D. K. Ljungblad. In prep. Hearing studies on gray whales. Eberhardt, R. L. 1978. Transect Methods for Population Studies. J . Wildl. Manage. 42: 1-31. Eberhardt, R. L. and W. E. Evans. 1962. Sound activity of the California gray whale, Eschrichtius glaucus. J . Aud. Eng. Soc. 10(4): 324-328. Evans, W. E. 1967. Vocalization among marine mammals. In:Marine Bio-Acoustics. Vol. II. pp. 159- 186. (Ed. W. N. Tavolga) Pergamon Press. Ezrow, D. H. 1962. Measurement of the Thermal Noise Spectrum of Water. J . Acoust. Soc. Am. 34:550. Fish, J . F. and H. E. Winn. 1969. Sounds of marine mammals. In: Encyclopedia of Marine Resources ( Ed. F. E. Firth) Van Nostrand Reinhold Co., pp. 649 - 655. Fish, J . F., J . L. Sumich, and G. L. Lingle. 1974. Sounds produced by the gray whale, Eschrichtius robustus. Mar. Fish. Rev. 36(4):38-45. Fletcher, J . L. and R. G. Busnel. 1978. Effects of Noise on Wildlife. Academic Press: New York. 305 pp. Ford, J . K. B. and H. D. Fisher. 1982. Killer Whale (Orcinus orca) Dialects as an Indicator of Stocks in British Columbia. Rept. Intl. Whal. Commn. 32: 671-679. Gautier, J . P. and A. Gautier. 1977. Communication in Old World Monkeys. In: How Animals Communicate. T. Sebeok (ed). Indiana Press University, Bloomington. pp. 890-964. 261 Greene, C. R. and B. M. Buck. 1964. Arctic Ocean Ambient Noise. J . Acoust. Soc. Am. 36: 1218. Hall, J . D. and C. S. Johnson. 1971. Auditory Thresholds of a Killer Whale Orcinus orca Linnaeus. J . Acoust. Soc. Am. 4(1): 515-517. Harvey, J . T. and B. R. Mate. 1984. Dive Characteristics and Movements of Radio-Tagged Gray Whales in San Ignacio Lagoon, Baja California Sur, Mexico. In: The Gray Whale. M. L. Jones, S. L. Swartz, and S. Leatherwood (Eds). Academic Press, New York. pp. 561-575. Heindsman, T. E. , R. H. Smith, and A. D. Arneson. 1955. Effect of Rain upon Underwater Noise Levels. J . Acoust. Soc. Am. 27:378. Herzing, D. L. and B. R. Mate. 1984. Gray Whale Migrations Along the Oregon Coast, 1978-1981. In: The Gray Whale. M. L. Jones, 5. L. Swartz, and S. Leatherwood (eds). Academic Press, New York, pp. 289-307. Holt, P. G. 1 978. Auditory stress and the immune system. J . Sound and Vib. 59(1):131-132. Hosokawa, H. 1950. On the cetacean larynx with special remarks on the laryngeal sac of the sei whales and the aryteno-epiglottideal tube of the sperm whale. Sci. Rep. Whales Res. Inst. 3, 23-62. Hubbs, C. L. 1966. Comments In:" Whales, Dolphins, and Porpoises". K. S. Norris, Ed. p. 444. University of California Press, Berkeley and Los Angeles. Jacobs, D. W. and J . D. Hall. 1972. Thresholds of a Freshwater Dolphin, Inia geoffrensis, Blainville. J . Acoust. Soc. Am. 51(1): 530-533. Jansen, J . Jr . and J . Jansen. 1953. A Note on the Amygdaloid Complex in the Fin Whale (Balaenoptera physalus). Hvalrad. Skr. 39: 1-14. Johnson, C. S. 1966. Auditory Thresholds of the Bottlenose Porpoise (Tursiops truncatus, Montagu). U. S. Naval Ordinance Test Station Report (NOTS TP 4178). Johnson, M. W. 1943. Underwater sounds of biological origin. OSRD Sect. 6, No. 1270. 1-sr. 30-412. Jones, M. L. and S. L. Swartz. 1984. Demography and Phenology of Gray Whales and Evaluation of Whale-Watching Activities in Laguna San Ignacio, Baja California Sur, Mexico. In: The Gray Whale. M. L. Jones, S. L. Swartz, and J.S. Leatherwood, eds. Academic Press, New York. pp. 309-374. Jones, M. L., S. L. Swartz and M. E. Dahlheim. 1987. Gray Whale Abundance and Distribution in San Ignacio Lagoon During 1985, with a Comparison to Demographic Trends from 1978-1982. Draft Final Report to U. S. Marine Mammal Commission. Contract No. MM 2911023-0, 28 pages. 262 Kellogg, W. N. and C. E. Rice. 1966. Visual Discrimination and Problem Solving in a Bottlenose Dolphin. In: Whales, Dolphins and Porpoises. K. S. Norris (ed). University of California Press, Berkeley, pp. 731-754. Kluczek, J . P. 1979. Famines of noise emitted by machinery at pigs industrial farms. Eighth International Conference on noise control engineering. Warsaw, Poland. September 1 1-13, 1 979. Knudsen, V. 0., R. S. Alford, and J . W. Emling. 1948. Underwater Ambient Noise. J . Mar. Res. 7, p. 410. Konishi, M. 1970. Evolution of Design Features in the Coding of Species Specificity. Am. Zool. 10: 67-72. Lim, D. J . and W. Melnick. 1975. Mode of cochlear damage by excessive noise - An Overview. Advisory group Aerospace Research Development, Conference Proceedings, 171. Ljungblad, D. K., S. Leatherwood, and M. E. Dahlheim. 1980. Sounds recorded in the presence of an adult and calf bowhead whale. Mar. Fish. Rev. 42 (9-10): 86-87. Ljungblad, D. K. and S. E. Moore. 1983. Killer Whales (Orcinus orca) Chasing Gray Whales (Eschrichtius robustus) in the Northern Bering Sea. Arctic 36: 361-364. Ljungblad, D. K., P. 0. Thompson, and S. E. Moore. 1982. Underwater sounds recorded from migrating bowhead whales, Balaena mysticetus, in 1979. J . Acoust. Soc. Am. 71(2): 477 - 482. Malme, C. I. , P. R. Miles, C. W. Clark, P. Tyack, and J . E. Bird. 1983. Investigations of the potential effects of the underwater noise from petroleum industry activities on migrating gray whale behavior. Final report on Contract No. AA 851-CT2-39, to Minerals Management Service #5366. Malme, C. I., P. R. Miles, C. W. Clark, P. Tyack, and J . E. Bird. 1984. Investigations of the potential effects of the underwater noise from petroleum industry activities on migrating gray whale behavior Phase II: January 1984 migration. Contract No. 14-12-0001-29033, # 5586. Malme, C. I., B. Wursig, J . E. Bird, and P. Tyack. 1986. Behavioral Responses of Gray Whales to Industrial Noise: Feeding Observations and Predictive Modeling. Final Report to N0AA, National Ocean Service, 0MA, 0AD, Anchorage, Alaska. Contract No. 85-ABC-00141. Report No. 6265. Marler, P. 1977. The Evolution of Communication. In: How Animals Communicate. T. Sebeok (ed). Indiana Press University, Bloomington. pp. 45-70. Marler, P. and R. Tenaza. 1977. Signalling Behavior of Wild Apes, with Specific Reference to Vocalization. In: How Animals Communicate. T. Sebeok (ed). Indiana Press University, Bloomington. pp. 965-1033. 263 McGeorge, L. W. 1979. Circumvention of Noise in the Communicatin Channel by the Structure and Timing of the Calls of Forest Animals. PhD Dissertation, Duke University. 155 pages. McLellan, H. J . 1965. Elements of Physical Oceanography. Pergamon Press, New York. 151 pages. Mead, J . G. and E. D. Mitchell. 1984. Atlantic Gray Whales. In: The Gray Whale. M. L. Jones, S. L. Swartz and S. Leatherwood (eds). Academic Press, New York. pp. 33-53. Miller, R. V., J . H. Johnson, and N. V. Doroshenko. 1985. Gray Whales (Eschrichtius robustus) in the Western Chukchi and East Siberian Seas. Arctic, 38:1: 58-60. Moore, S. E. and D. K. Ljungblad. 1984. Gray whales in the Beaufort, Chukchi, and Bering Seas: Distribution and Sound Production. In: The Gray Whale. M. L. Jones, S. L. Swartz, and S. Leatherwood (eds). Academic Press. New York. pp. 543-559. Moore, S. E. , D. K. Ljungblad, and D. R. Schmidt. 1985. Ambient, Industrial and Biological Sounds Recorded in the Northern Bering, Eastern Chukchi, and Alaskan Beaufort Seas during the Seasonal Migrations of the Bowhead Whale (Balaena mysticetus), 1979-1982. Final Report prepared for the Minerals Management Service, Alaska Outer Continental Shelf Region, U. S. Department of the Interior. Prepared by Seaco, Inc., 2845 Nimitz Boulevard, San Diego, California 92106 Nawrot, P., R. 0. Cook, and C. W. Hamm. 1978. Effect of continuous, random onset and high frequency noise on embryo development in mice. Abstract of joint meetings of the Acous. Soc. of America and Japan. J . Acoust. Soc. Am. (Suppl) 64 (1). Nerini, M. 1984. A Review of Gray Whale Feeding Ecology. In: The Gray Whale. M. L. Jones, S. L. Swartz and S. Leatherwood (eds). Academic Press. New York. pp. 423-450. Nielson, D. W., J . Burnham and C. Talley. 1978. Squirrel monkey temporary threshold shift from 48-h exposures to low-frequency noise. J . Acoust. Soc. Am. 64(2):478-484. Norris, K. S. , R. M. Goodman, B. Villa-Ramirez and L. Hobbs. 1977. Behavior of the California Gray Whale, Eschrichtius robustus, in southern Baja California, Mexico. Fish. Bull. 75(1): 159-172. Oliver, J . S . , P. N. Slattery, M. A. Silberstein, and E. F. O'Connor. 1983. A Comparison of gray whale, Eschrichtius robustus, feeding in the Bering and Baja California. Fish. Bull. 81(3): 513-522. Oliver, J . S. , P. N. Slattery, M. A. , Silberstein, and E. F. O'Connor. 1984. Gray whale feeding on dense Ampeliscid Amphipod communities near Bamfield, British Columbia. Can. J . Zool. 62: 41-49. 264 Oppenheimer, J . R. 1977. Communication in New World Monkeys. In: How Animals Communicate. T. Sebeok (ed). Indiana Press University, Bloomington. pp. 851-889. Otte, D. 1977. Communication in Orthoptera. In: How Animals Communicate. T. Sebeok (ed). Indiana Press University, Bloomington. pp. 334-361. Painter, D. W. II. 1963. Ambient noise in a coastal lagoon. J . Acoust. Soc. Am. 35 (9):1458. Payne, R. and S. McVay. 1971. Songs of the humpback whales. Science 173:587-597. Poulter, T. C. 1968. Vocalizations of the gray whales in Laguna Ojo de Liebre (Scammon's Lagoon) Baja California, Mexico. Nor. Hvalfangst-Tid. 3 : 53 -62 . Rasmussen, R. A. and N. E. Head. 1965. The quiet gray whale (Eschrichtius . glaucus). Deep-Sea Res. 12:869-877. Ray, G. C. and W. E. Schevill. 1974. Feeding of a captive gray whale. Mar. Fish. Rev. 36 (4 ) :31 -38 . Reiily, S. B., D. W. Rice, and A. A. Wolman. 1983. Population assessment of the gray whale (Eschrichtius robustus) from California shore censuses, 1967-1980. Fish. Bull. 81 (2 ) : 267-281. Rice, D. W. and A. A. Wolman. 1971. Life history and ecology of the gray whale (Eschrichtius robustus). Spec. Publ. Am. Soc. Mammal. 3 , 1-42. Rugh, D. J . 1984. Census of Gray Whales at Unimak Pass, Alaska: November-December 1977-1979. In: The Gray Whale. M. L. Jones, S. L. Swartz, and S. Leatherwood (eds). Academic Press. New York, pp. 225-248. Saunders, J . and R. Dooling. 1974. Noise induced threshold shift in the parakeet (Melopsittacus undulatis). Proc. Nat. Acad. Sc i . U.S. A. 71(5) : 1962-1965. Schevill, W. E. and W. A. Watkins. 1962. Whale and Porpoise Voices. Woods Hole, Massachusetts: Woods Hoie Oceanographic Institution. Phonograph record and 24- page booklet. Schulte, H. 1916. Anatomy of a foetus of Balaenoptera borealis. Mem. Am. Mus. Nat. Hist. 1: 389-502. Simkin, G. N. and V. D. Il'ichen. 1965. Geographical Voice Variability in Animals as an Ecological and Evolutionary Problem. Zoologicheskii Zhurnal 44: 483-493 . Translated from Russian by National Science Foundation Office of Science Information Service. Smiley, C. S. and W. A. Wilbanks. 1980. Some effects of noise exposure on early development in the albino rat. J . Acoust. Soc. Am. 6 7 ( 1 ) . 265 Smith, W. J . 1977. Communication in Birds. In: How Animals Communicate. T. Sebeok (ed). Indiana University Press, Bloomington. pp. 545-574. Sokal, R. F. and F. J . Rohlf. 1969. Biometry: The Principles and Practice of Statistics in Biological Research. W. H. Freeman, San Francisco, California. Swartz, S. L. and W. C. Cummings. 1978. Gray whales (Eschrichtius robustus) in Laguna San Ignacio, Baja California, Mexico. U.S. NTIS Rept. PB-276319:1-38. Swartz, S. L. and M. L. Jones. 1980. Gray whales (Eschrichtius robustus) during the 1977-78 and 1978-79 winter seasons in Laguna San Ignacio, Baja California Sur, Mexico. U.S. NTIS Rept. PB80-202989:I-35. Swartz, S. L. and M. L. Jones. 1981. Demographic studies and habitat assessment of gray whales (Eschrichtius robustus) in Laguna San Ignacio, Baja California Sur, Mexico. U. S. NTIS Rept. PB82-123373, 1-56. Sumich, J . L. 1983. Swimming Velocities, Breathing Patterns, and Estimated Costs of Locomotion in Migrating Gray Whales, Eschrichtius robusutus. Can. J . Zool. 61(3): 647-652. Thompson, P. 0. and W. C. Cummings. 1969. Sound production of the finback whale, Balaenoptera physalus, and Eden's whale, B_. edeni, in the Gulf of California. Proc. Sixth Ann. Conf. on Biol . Sonar and Diving Mammals. Menlo Park, California. Stanford Research Institute, p. 109. Thomson, D. H. 1983. Distribution, Production, and Ecology of Gray Whale Prey Species. In: Feeding Ecology of Gray Whales (Eschrichtius robustus) in the Chirikof Basin, Summer 1982. D. H. Thomson (ed). Final Reports of Principle Investigators, Vol. 43, June 1986. Outer Continental Shelf Environmental Assessment Program. U.S. Dept. of Commerce. NOAA, Alaska Office, pp. 227-292. Thorpe, W. H. 1961. Bird-Song: The Biology of Vocal Communication and Expression in Birds. Cambridge University Press, Cambridge. Tomilin. A. G. 1 957. "Zveri SSSR i pri1ezhashchikh stran"(MammaIs of the USSR and Adjacent Countries. Kittobraznye, Vol. 9. Izd. Akad. Nauk SSSR, Moskva (Israel Program for Sci . Translation, Jerusalem, 1967). Urick, R. J . 1967. Principles of Underwater Sound for Engineers. McGraw-Hill Book Company, New York. 342 pages. Van Bergeijk, W. A . , J . R. Pierce, and E. E. David, Jr. 1960. Waves and the Ear. Doubleday and Company, Inc. Garden City, New York. 235 pages. Watkins, W. A. and W. E. Schevill. 1977. Sperm whale codas. J . Acoust. Soc. Amer. 62(6): 1485 - 1490. Watkins, W. A. 1977. Acoustical behavior of sperm whales. Oceanus, 20(2): Spring 1977, pp. 50 - 58. 266 Watkins, W. A. 1981. Activities and underwater sounds of fin whales. Sci. Rep. Whales Res. Inst. 33:83-117. Wenz, G. M. 1961. Some periodic variations in low-frequency acoustic ambient noise levels in the ocean. J . Acoust. Soc. Am. 33:64. Wenz, G. M. 1962. Acoustic ambient noise in the ocean: spectra and sources. J . Acoust. Soc. Am. 34:1936. Wenz, G. M. 1964. Curious noises and the sonic environment in the ocean. In: Marine Bio-Acoustics. W. N. Tavolga, ed. pp: 101-119. Macmillan, New York. Wiley, R. H. and D. G. Richards. 1978. Physical Constraints on Acoustic Communication in the Atmosphere: Implications for the Evolution of Animal Vocalizations. Behav. Ecol. Sociobiol. 3: 69-94. Wilkenshaw, H. M. 1960. Low-frequency spectrum of deep ocean ambient noise. J . Acoust. Soc. Am., 32:1497. Winn, H. E. and P. J . Perkins. 1976. Distribution and sounds of the minke whale, with a review of mysticete sounds. Cetology (19): 1-12. Winn, H. E. , P. J . Perkins, and T. C. Poulter. 1971. Sounds of the humpback whale. Proc. Seventh Ann. Conf. on Biol. Sonar and Diving Mammals. 1970. Menlo Park, California: Stanford Research Inst., pp. 39 - 52. W'uYsig, B., D. A. Crol 1, and R. S. Wells. 1983. Behavior of Summering Gray Whales. In: Feeding Ecology of Gray Whales (Eschrichtius robustus) in the Chirikof Basin, Summer 1982. D. H. Thomson (ed). Final reports of Principal Investigators, Vol. 43, June 1986. Outer Continental Shelf Environmental Assessment Program. U.S. Dept. of Commerce, NOAA. Alaska Office, pp. 333-376. Zar, J . H. 1974. Biostatistical Analysis. Prentice-Hall, Englewood Cl i f fs , New Jersey. Zielinski, K. 1979. Cueing properties of the decrease of white noise intensity for avoidance conditioning in cats. Acta Neurobiol. Exp. 39(5): 263-283. A-1 A P P E N D I X A A-2 APPENDIX A Propagation Experiments To determine the propagation characteristics of the underwater environment of Laguna San Ignacio, an underwater transducer (Lubell Speaker Model 98, capable of producing a signal of known frequency and level) was positioned approximately 75 m off the farthest promontory available at Punta Piedra/Rocky Point. This transducer was affixed to a 1.5-m by 1.5-m square of PVC pipe. This PVC structure (cage) arose out of the need to 1) float the transducer and 2) protect it from possible damage by the whales. Several holes were drilled into this pipe, allowing water to enter into the cage as it was being submerged. A diver was used to guide the transducer cage (in an upright position) to the bottom. Several burlap sacks f i l led with sand were placed over the lower bars of the cage to secure it to the bottom of the lagoon (to prevent drift due to tidal currents). The depth of the transducer cage was approximately 8 meters, within the tolerance level required for the transducer. The associated cables were weighted down (to prevent entanglement problems with whales) and were run along the bottom of the lagoon to the shore-based station at Rocky Point. The shore-based equipment consisted of an Acoustic Systems Inc. Amplifier driven by a 12-volt, deep-cycle marine battery. A Briggs and Stratton generator was used to recharge the batteries, since extensive battery drain occurred during playback periods. The amplifier was equipped with a built-in calibration (test) tone, which swept from 20 kHz to 400 Hz over a 40-sec time frame. Maximum output levels ranged from 80-150 dB re 1 juPa, which varied with frequency. The built-in test tone was the signal used during the underwater propagation experiments. Once the acoustical equipment was in place, a 4.3-m inflatable boat A-3 was dispatched from the shore-based station. Two team members were located in the boat and were equipped with all the necessary equipment to collect recordings of the projected test tones. The calibrated recording system used consisted of a Nakamichi 550 tape recorder and an LC-32 hydrophone. The frequency response of this equipment ranged from 40 Hz to 19 kHz + 3 dB. Communication between the shore-based personnel and the boat team was accomplished through hand-held CB walkie talkies. The boat team was requested to position its vessel at various distances and angles relative to the transducer site. Boat locations were determined by taking bearings from two locations (separated by 100 m) from the Rocky Point shoreline. An optical range finder was also used to estimate the distance of the vessel from the shoreli ne. Once the boat was on station, the hydrophone was lowered at 1.6 m (5 foot) depth increments until the bottom was reached. Recordings of the test tone (being projected by the shore-based team) were made at each station at these 1.6 m depth increments. Pertinent additional data recorded during the acoustical experiments included: 1) wind speed and direction; 2) sea state; 3) tide conditions; 4) time of day; 5) bottom depth; and 6) presence/absence of whales. Propagation experiments were terminated when Beaufort 2 conditions or above occurred, current caused hydrophone acceleration, and/or vessel or skiff traffic interfered with experiments. In addition to the test tones projected during propagation experiments in the immediate study area, sound sources used during the playback experiments (Chapters 4 and 5) were also projected back underwater to determine the received levels of these sources in areas outside the immediate study area. Recordings were conducted from the 4.3-m boat and A-4 utilized the same equipment as described for the collection of the test tone data. Average sound pressure levels were computed using a Nicolet Scientific Corporation FFT Computing Spectrum, Model 446. The bandpass f i l ter was set for 0-20 kHz range and 128 linear averages. Permanent displays of sound pressure levels calculated were obtained by connecting an X-Y plotter to the Spectrum Analyzer. Sound profiles were made for each station and compared against each other to determine the acoustical differences in frequencies and levels among station locations. An overlay grid was designed by the author which allowed conversion of the analog signals into digitized data. In addition to conducting the propagation experiments, those physical parameters that could potentially influence a propagating sound wave were investigated. Information was either collected or reviewed on depth profiles, bottom sediments, temperature, and salinity. These parameters were collected throughout the lagoon system. Data on depth profiles were provided by Jones and Swartz (1984), discerned from echograms made by a continuous-recording fathometer. Divers made visual observations of the lagoon floor and sampled bottom sediments. A Beckman Salinometer and thermometer was used to collect information on salinity and temperature. In the Bering Sea, temperature and salinity profiles were made using CTDs deployed from the NOAA vessels. Values were calculated by a NOAA officer and given to those scientists who requested the information. Unfortunately, the locations of these measurements did not coincide with the locations of the ambient noise stations, since the large ship was not able to enter the shallow water areas visited by the Boston Whaler. A-5 Results A to ta l of 61 propagation measurements was made in the immediate area of f Punta Piedra (designated as the study a rea) , representing a to ta l of 15 stat ions ( F i g . A l ) . Each stat ion locat ion was numbered and each associated 1.6-m drop was given a l e t te r designat ion. For example, Stat ion 2a = 1.6-m hydrophone depth; Stat ion 2b = same locat ion (however, depth of hydrophone was 3.2 m). Stat ion loca t ions , hydrophone depths, and other pert inent information recorded during these measurements are l i s t ed in Table A - I . A l l measurements were made at high slack water during Beaufort I or 2 sea s ta tes. A sound p r o f i l e was made for each of the 61 measurements d isplaying frequency on the horizontal axis and sound level on the ver t ica l ax i s . An example i s shown for Stat ion la ( F i g . A2). Each p ro f i l e was converted to d ig i t i zed data using a p las t i c overlay grid designed for th is purpose. The levels at in terva ls of 1 kHz measurements were col lected and are l i s t ed in Table A-I I. A comparison of the p ro f i l es among stat ions indicated that only minor f luctuat ions occurred in sound propagation pathways. Major areas of sound enhancement and/or cancel la t ion were not observed in the immediate study area. An inspect ion of the received levels of projected sources calculated for areas outside the immediate study area, however, did resul t in major acoust ical anomalies ( F i g . A3). Recordings of the projected signals and playback tapes to be used i n Chapters 2 and 3 projected into the lower lagoon area (n = 6 s tat ions) resulted in levels higher than expected for sound propagation fol lowing the cy l i nd r i ca l spreading law. Conversely, recordings made in the middle lagoon (n = 7 stat ions) of these projected sources resulted in received levels lower than expected for sound A-6 Figure A . l Locations of Propagation Stations in Laguna San Ignacio, Mexi co. A-7 A-8 Table A - I . Angles and Measured Depths of Propagation Test Tone Stat ions in Laguna San Ignacio, Mexico Angle South Marker Angl e North Hydrophone Bottom Stat ion No. Marker Depth ( f t ) Depth ( f t ) la 243 214 3 6 2a 222 212 5 22 2b 222 212 10 22 2c 222 212 15 22 2d 222 212 20 22 3a 218 214 5 25 3b 218 214 10 25 3c 218 214 15 25 3d 218 214 20 25 4a 234 224 5 23 4b 234 224 10 23 4c 234 224 15 23 4d 234 224 20 23 5a 233 225 5 22 5b 233 225 10 22 5c 233 225 15 22 5d 233 225 20 22 6a 240 222 5 18 6b 240 222 10 18 6c 240 222 15 18 7a 262 242 5 21 7b 262 242 10 21 7c 262 242 15 21 7d 262 242 20 21 8a 258 247 5 21 8b 258 247 10 21 8c 258 247 15 21 8d 258 247 20 21 9a 246 242 5 22 9b 246 242 10 22 9c 246 242 15 22 9d 246 242 20 22 lOal 276 255 5 21 lObl 276 255 10 21 1 Oc 1 276 255 15 21 lOdl 276 255 20 21 10a2 289 253 5 8 H a l 283 279 5 21 l l b l 283 279 10 21 1 l c l 283 279 15 21 l l d l 283 279 20 21 l la2 286 272 5 26 l lb2 286 272 10 26 l l c2 286 272 15 26 1 Id2 286 272 20 26 . . . conti nued A-9 Table A-I, Continued - -Angle Angle South North Hydrophone Bottom Station No. Marker Marker Depth (ft) Depth (ft) 12a 280 272 5 28 12b 280 272 10 28 12c 280 272 15 28 1 2d 280 272 20 28 13a 306 302 5 25 13b 306 302 10 25 13c 306 302 15 25 13d 306 302 20 25 14a 306 296 5 23 14b 306 296 10 23 14c 306 296 15 23 14d 306 296 20 23 15a 297 274 5 21 15b 297 274 10 21 15c 297 274 15 21 15d 297 274 20 21 A-10 Figure A.2 Sound Profile of the Received Test Tone at Station la in Laguna San Ignacio, Mexico. Table A-II. Digitized Data on Level (dB re 1 jdPa) and Frequencies (kHz) of Test Tones Received at Stations la - 15d.in Laguna San Ignacio, Mexico Leve l at S t a t i o n F R E Q U E N C Y p<HzT 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 8 104 105 100 101 99 98 97 96 96 96 96 96 96 96 95 94 94 94 94 8 105 105 102 102 101 101 100 100 100 99 98 97 97 97 96 94 94 94 94 8 106 108 102 103 101 101 100 100 100 99 98 98 97 97 96 94 94 94 94 8 104 103 102 103 102 102 101 101 100 100 99 99 98 98 97 94 94 94 94 8 103 103 102 102 101 101 102 100 100 100 99 98 98 98 97 94 94 94 94 8. 102 107 102 102 102 101 101 101 100 100 99 98 98 98 96 94 94 94 94 OO OO c 102 106 102 102 102 101 101 101 100 100 100 99 98 98 97 94 94 94 94 OO OO c 103 104 103 102 102 101 101 101 100 100 99 99 98 98 96 94 94 94 94 8 104 106 103 103 102 102 101 101 100 100 99 98 98 98 96 94 94 94 94 8 103 105 105 105 104 104 104 103 103 103 101 100 99 99 97 94 94 94 94 8 103 107 104 104 103 103 103 102 102 101 100 100 100 99 97 94 .94 94 94 i 00 00 104 107 105 105 104 104 104 103 103 102 101 100 100 100 97 95 94 94 94 > i 00 00 105 109 106 106 105 104 104 104 106 105 103 102 102 101 100 95 94 94 8 106 108 104 106 103 101 101 101 101 100 99 98 , 98 97 96 94 94 94 94 8 106 108 105 105 104 102 102 102 101 100 99 98 98 97 96 94 94 94 94 8 105 107 104 103 102 101 101 100 100 100 99 98 98 97 95 94 94 94 94 8 105 107 104 105 103 102 102 101 101 101 100 100 99 98 96 95 94 94 94 8 106 109 104 105 105 105 104 104 103 102 101 101 100 100 98 95 94 94 94 8 8 105 111 107 107 106 105 105 105 104 103 103 102 101 101 100 95 94 94 94 111 112 104 104 104 103 102 101 101 101 100 99 99 98 96 94 94 94 94 8 107 112 108 110 105 104 104 104 107 103 100 99 100 98 98 95 94 94 94 8 105 115 110 107 105 104 106 105 108 103 101 99 98 99 98 94 94 94 94 8 106 108 107 107 106 105 106 102 103 102 100 100 100 98 97 94 94 94 94 8 111 109 108 108 106 104 104 102 103 105 100 99 99 99 97 94 94 94 94 8 8 103 108 104 105 104 102 103 102 102 101 100 100 99 99 97 94 94 94 94 105 110 103 103 102 102 102 101 100 100 99 98 97 97 96 94 94 94 94 8 105 n o 104 104 103 102 102 101 100 100 99 98 98 97 96 94 94 94 94 8 103 107 106 105 102 101 101 101 102 100 99 98 98 98 97 94 94 94 94 6 102 104 103 103 102 102 101 100 100 100 99 98 98 98 96 94 94 94 94 8 102 105 103 103 103 102 102 101 101 100 100 99 98 98 97 94 94 94 94 8 102 107 102 103 102 101 101 100 101 100 99 98 98 97 96 94 94 94 94 8 103 108 103 103 103 101 101 100 101 100 99 98 98 98 96 94 94 94 94 l a 2a 2b 2c 2d 3a 3b 3c 3d 4a 4b 4c 4d 5a 5b 5c 5d 6a 6b 6c 7a 7b 7c 7d 8a 8b 8c 8d 9a 9b 9c 9d Table A-II.Continued — F R E Q U E N C Y ( i H z J a t S t a t i o n 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 Oa 1 118 108 107 101 107 104 105 101 102 100 100 99 99 98 98 97 94 94 94 94 1 Obi 118 105 109 103 106 105 103 102 101 100 100 98 97 97 97 95 94 94 94 94 l O c l 118 108 108 102 107 101 106 102 100 100 100 97 97 98 98 97 95 94 94 94 lOd l 118 110 110 101 105 105 104 101 100 100 100 99 98 98 98 97 96 94 94 94 10a2 118 105 110 108 104 103 102 101 99 100 100 100 97 97 96 95 94 94 94 94 11 a l 118 103 107 102 101 101 101 100 100 100 100 99 99 98 98 96 94 94 94 94 11 bl 118 103 105 103 102 101 101 100 100 100 100 99 98 98 97 96 94 94 94 94 1 l c l 118 103 103 101 101 101 102 100 100 100 99 98 97 97 96 95 94 94 94 94 1 l d l 118 104 105 101 101 100 100 100 100 100 99 98 97 97 96 96 94 94 94 94 1 la2 111 103 104 100 100 99 100 99 98 100 99 97 96 95 95 95 93 93 93 94 l l b 2 115 103 103 100 100 100 100 99 99 99 99 98 97 96 96 95 94 94 94 94 1 l c2 117 103 102 101 100 99 99 99 98 98 98 98 . 97 97 96 95 94 94 94 94 11 d2 118 103 106 101 101 100 100 99 99 99 98 97 96 96 96 95 94 94 94 94 12a 113 100 104 102 101 101 101 100 100 100 100 99 98 98 97 96 94 94 94 94 12b 116 100 103 102 •101 101 101 101 100 100 100 100 99 99 98 96 94 94 94 94 12c 115 100 103 100 100 100 100 100 100 100 100 99 98 98 97 96 94 94 94 94 1 2d 118 101 102 100 100 100 100 100 100 100 100 99 99 98 98 97 94 94 94 94 13a 115 100 103 102 102 101 101 101 100 100 99 99 98 98 97 96 94 94 94 94 13b 112 101 102 101 101 101 100 100 100 100 99 99 98 97 96 95 94 94 94 94 13c 117 103 103 102 102 101 101 101 101 100 100 100 100 99 98 96 94 94 94 94 13d 117 102 103 102 102 102 102 102 101 101 100 100 100 99 99 98 94 94 94 94 14a 118 102 104 102 102 102 103 100 100 99 99 98 97 96 96 95 94 94 94 94 14b 118 105 105 101 103 101 101 100 100 100 99 98 98 98 97 95 94 94 94 94 14c 118 105 104 101 101 101 100 100 100 99 99 98 97 96 96 95 94 94 94 94 14d 118 102 104 100 100 100 101 101 100 100 99 98 98 98 97 96 94 94 94 94 15a 118 105 104 100 106 104 106 102 100 100 99 98 97 97 96 95 94 94 94 94 15b 118 107 110 102 103 103 104 102 100 100 99 99 98 97 97 95 94 94 94 94 15c 118 118 105 107 101 102 101 101 101 99 99 98 97 96 96 96 95 94 94 94 94 15d 106 108 102 103 102 102 101 100 99 99 98 97 97 96 95 94 94 94 94 A-14 Figure A.3 Locations of Propagation Measurements Collected Outside of Immediate Study Area in Laguna San Ignacio, Mexico. A-15 A-1 6 propagation fol lowing a cy l i nd r i ca l spreading law (10 log R) as proposed for shallow-water environments. Jones and Swartz (1984) divided the lagoon into f i ve d i s t i nc t areas based on lagoon bathymetry and sedimentology: the i n l e t , east channel, and the lower, middle, and upper lagoon ( F i g . A4). For the purposes of th is acoust ical i nves t i ga t i on , only the lower and middle areas of the lagoon are addressed. In the lower lagoon, a steep-walled channel cuts i t s way from the 3-km wide in le t to a const r ic t ion of 1.8-km wide at Punta Piedra, where i t terminates. A maximum depth of 25.9 m was reported and represents the widest, r e l a t i ve l y deep channel in the lagoon. The deepest regions of th is channel are covered with poorly sorted, hard-packed, f ine to coarse grained sand. West of Rocky Po in t , medium-grained sand ridges are interspersed with occasional rock outcroppings. Proceeding North, a system of three channels characterizes the middle lagoon. The channels range in depth from 7.6 to 21.3 meters. The sediments are poorly sorted, with f ine to coarse grained sand with crushed shel l and some rocky areas. Extensive sand bars support strands of eelgrass (Zostera marina); however, the channels are devoid of plant l i f e . Depth p ro f i l es in Laguna San Ignacio are depicted in F i g . A5. Temperature and s a l i n i t y measurements were made at various locat ions and depths throughout the lagoon system (F ig . A6). Temperature values ranged from 17*to 19° C and s a l i n i t y values averaged 32 ppt. A comparison of these values with depth indicated a homogenous, well-mixed water column. In the Bering Sea, water temperatures varied with depth from 7.49'C on the surface to 0.10°C at 20 meters. A thermocline was prevalent at about 5 m and at 10 and 20 meters ( F i g . A7). Sa l i n i t y values ranged between 31.5 ppt to 32.0 ppt (F i g . A8). Major var iat ions in s a l i n i t y A-17 F i g u r e A.4 Major G e o g r a p h i c a l D i v i s i o n s o f Laguna San I g n a c i o , Mexi From Jones and Swartz, 1984. A-18 A-1 9 Figure A.5 Depth Pro f i l es of Laguna San Ignacio Based on Echograms. Soundings are in Feet and Ver t ica l Exaggeration i s lOUx. Horizontal Scale Approximate. From Jones and Swartz, 1984. A-20 A-21 Figure A.6 Locations of Measurements of Temperature and S a l i n i t y in Laguna San Ignacio, Mexico. A-22 A-23 F i g u r e A.7 Temperature vs Depth P r o f i l e s f o r Nine S t a t i o n s o f f S t . Lawrence I s l a n d , A l a s k a . Depth (meters) A-25 Figure A.8 S a l i n i t y vs Depth Pro f i l es for Nine Stations off S t . Lawrence Is land, Alaska. ol 1 1 1 — 1 ' 1 J 19.9 22.2 30.0 30.5 31.0 31.5 32.0 32.5 Salinity (ppt) A-27 vs depth did not occur with the exception of Stat ion 9, which was inf luenced by fresh water. The source of th is fresh water i s unknown. Locations of these measurements on temperature and s a l i n i t y are l i s ted in Table A- I I I -Table A-111. Locations of Measurements Col lected on Temperature and S a l i n i t y of f S t . Lawrence Island Lat i tude/ Surface 5 M 10 M 20 M 30 M Bottom Date Time Station Longitude 'C ppt "C ppt °C ppt "C ppt "C ppt Depth(M) 7/1 1/82 1601 I 63°24.5' N, 5.06 31 .54 5.13 31.46 4.80 31 .56 1.79 31 .90 - - — 32 171° 51.2'W 7/11/82 2248 2 63°33.8'N, 5.30 31.55 5.00 31.48 4.96 31.45 1.14 31.61 — — 30 171° 53.8'W 7/12/82 2027 3 63°54.4'N, 6.23 31.38 6.10 31 .33 5.15 31 .34 4.36 32.08 0.69 32.51 37 171° 57.9'W 7/13/82 1501 4 63°37.3'N, 5.18 31.45 4.96 31.50 4.40 31.67 0.10 32.14 — - - 26 171° 52.1'W 7/14/82 1501 5 63°46.4'N, 5.32 31 .25 5.02 31 .52 3.60 31 .64 - - 23 171* 48.5'W 7/1 5/82 1723 6 63*45.1'N, 7.49 31 .53 6.19 31.50 5.63 31.42 — — — — 20 17f 28.3'W 7/1 6/82 1119 7 62*59.5'N, 2.48 32.04 2.41 31 .97 2.26 32.01 0.57 32.09 - - — 23 169'27.7'W 7/20/82 0932 8 62°59.0'N, 4.58 31 .76 4.40 31.74 2.40 31.67 1.41 31.92 — — 22 169°28.4'W 7/21/82 1111 9 62°59.5'N, 3.49 19.90 3.48 22.27 1.46 31.86 0.89 31.91 — — 22 169"28.0'W B-1 A P P E N D I X B B-2 APPENDIX B 1983 Ca l l i ng Rates  Control Periods During control periods (defined as p re - t r i a l s and/or no intervent ion of man-made noise) gray whales produced an average of 18.4 + s . d . 4.8 c a l l s per 15-minute interval (n = 24.5 hrs = 98/15-min i n t e r va l s ) . S t a t i s t i c a l dif ferences were not detected when comparisons were made between periods with no intervent ion of man-made noise and p r e - t r i a l periods (Anova, df = 4, F = 0.55, p > 0.05). During t r i a l and pos t - t r i a l per iods, the level of sound ac t i v i t y was observed to change when compared to control per iods. Experiment A - Real Sources In the presence of real sound sources ( t r i a l periods Experiment A) gray whales produced an average of 41.9 + s . d . 10.9 sounds per 15-minute in terva l (n = 3.5 hrs = 14/15-min i n t e r va l s ) . When th is t r i a l value was compared to p re - t r i a l periods (17.3 + s . d . 5.2 sounds/15-min i n t e r v a l , n = 3.5 h rs ) , a s ign i f i can t increase in c a l l i n g rates was s t a t i s t i c a l l y documented in the presence of real sources (t = 7.62, p < 0.05) . During pos t - t r i a l per iods, c a l l i n g rates were shown to decrease with respect to t r i a l s with an average of 29.6 + s . d . 3.8 ca l l s produced for a 15-min in terva l (n = 3.5 h rs ) . This pos t - t r i a l rate was s t a t i s t i c a l l y lower than t r i a l periods (t = 3.89, p < 0.05) but higher than p re - t r i a l periods (t = 7.32, p < 0.05). A pos t - t r i a l breakdown by 5-min in te rva ls (but weighted for 15 minutes) indicated a gradual reduction in c a l l i n g rates during th is pos t - t r i a l per iod. During the f i r s t f i ve minutes an average of 36 sounds was produced. An average of 30 ca l l s was produced during the second f ive minute category. During the f i na l f i ve minutes of the pos t - t r i a l per iod, an average of 24 sounds was produced. B-3 Experiment B - Outboard Engine Noise During the project ion of outboard engine noise ( t r i a l periods Experiment B ) , an increase in c a l l i n g rates (when compared to control periods = 19.7 + s . d . 3.4 ca l l s produced/15 min in terva l ) occurred with 45.5 + s . d . 11.3 ca l l s produced/15-min in terval (n = 3.5 hrs ) . This t r i a l rate was s i gn i f i can t l y higher than that documented during p re - t r i a l periods (t = 8.15, p < 0.05) . An inspection of the ca l l i ng rates during p o s t - t r i a l periods for Experiment B showed a reduction in the average number of c a l l s produced when compared to t r i a l s with an average of 24.9 ± s . d . 4.3 (n= 3.5 hrs) c a l l s emitted for the 15-minute i n t e r v a l . Pos t - t r i a l rates were s t a t i s t i c a l l y lower than t r i a l periods and were s i gn i f i can t l y higher than p re - t r i a l periods (t = 6.35, p < 0.05 and t = 3.60, p < 0.05, respec t i ve l y ) . A breakdown of pos t - t r i a l periods by f i ve minute categories (weighted for 15 minutes) gave values of 30, 27, and 18 sounds produced for each 15-min i n t e r v a l . By the end of the pos t - t r i a l period ( las t f i ve minutes), c a l l i n g rates had returned to p re - t r i a l values. Experiment C - Gray Whale Sounds During Experiment C ( t r i a l periods with gray whale sounds being projected, n = 3.5 h rs ) , the rate of c a l l i n g per 15-min in terva l averaged 23.9 + s . d . 5.9 sounds produced for the 15-min i n t e r v a l . This t r i a l value ref lected a s l igh t increase in c a l l i n g rates over p r e - t r i a l periods (17.6 + s . d . 4.4 sounds produced). When these values were compared, a s t a t i s t i c a l di f ference was noted (t = 3.06, p < 0.05). An overal l reduction in c a l l i n g rates occurred during p o s t - t r i a l periods with 24 sounds produced in the f i r s t f i ve minutes of the test per iod , 18 sounds for the second f i ve minutes and 18 sounds for the last f i ve minutes. These f ive-minute ca l l i ng rates were, of course, weighted for a 15-min i n t e r v a l . An average ca l l rate for B-4 the overal l 15 minute pos t - t r i a l period was 19.8 + s . d . 3.1 sounds produced. When th i s value was compared to t r i a l per iods, a s ign i f i can t reduction in c a l l i n g rates was documented in pos t - t r i a l periods (t = 2.26, p < 0.05). However, when comparisons were made between post and p re - t r i a l per iods, c a l l i n g rates were not s ign i f i can t l y d i f ferent (t = 1.37, p > 0.05). Experiment D - Test Tone Signals During the project ion of the test tone ( t r i a l periods Experiment D), a dramatic reduction in acoust ical ac t i v i t y was documented when compared to p r e - t r i a l per iods. In the presence of the test tone, a l l vocal ac t i v i t y ceased. An inspect ion of the pos t - t r i a l periods associated with t h i s pa r t i cu la r experiment indicated that gray whales remained s i l en t for at least 10 minutes fol lowing the acoust ical perturbat ion. Sound ac t i v i t y was shown to increase during the las t f i ve minutes of th is test period with 15.2 + s . d . 3.7 c a l l s produced (weighted for 15 minutes). This overa l l pos t - t r i a l c a l l i n g rate was s ign i f i can t l y lower than p r e - t r i a l rates (t = 2.16, p < 0.05). Comparisons of Ca l l i ng Rates Among Tr ia l Periods A comparison of the c a l l i n g rates among the various experiments y ie lded the fol lowing resu l t s . When t r i a l periods of Experiment A (real sources, ca l l rate equals 41.9 + s . d . 10.9) and Experiment B (outboard engine playback, c a l l rate equals 45.5 + s . d . 11.3) were compared, no s t a t i s t i c a l di f ferences in c a l l i n g rates were detected (t = 0.866, p > 0.05). When c a l l i n g rates were compared between t r i a l periods of Experiment A (real sources = 41.9 + s . d . 10.9) and Experiment C (gray whale sounds = 23.9 + s . d . 5 .9) , s ign i f i can t di f ferences occurred (t = 5.41, p < 0.05), with higher c a l l i n g rates documented in the presence of real sources. The c a l l i n g rates observed during t r i a l periods of outboard engine playback B-5 (45.5 + s . d . 11.3) and gray whale sound playback (23.9 ± s . d. 5.9) were a lso markedly d i f ferent (t = 6 .3 , p < 0.05) . Outboard engine noise e l i c i t e d a higher c a l l i n g rate than the project ion of gray whale s i gna l s . Since a l l vocal izat ions ceased during the project ion of the test tone, comparisons could not be made. Comparisons of Ca l l i ng Rates Among P o s t - t r i a l s Periods Comparisons made during pos t - t r i a l periods among the various experiments (n = 3.5 hrs for each experiment) resulted in d i s t i nc t di f ferences in c a l l i n g rates for each comparison and are summarized as fo l lows: real (26.6 + s . d . 3.8) vs outboard (24.9 + s. d . 4.3) (t = 3.24, p < 0.05); real (29.6 + s . d . 3.8) vs gray whale (19.8 + s . d . 3.1) (t = 7.6, p < 0.05); real (29.6 + s. d. 3.8) vs test tone (15.2 + s . d . 3.7) (t = 10.3, p < 0.05; and gray whale (19.8 + s . d. 3.1) vs test tone (15.2 + s . d . 3.7) (t = 3.56, p < 0.05). An analysis of variance applied to these pos t - t r i a l rates also indicated s t a t i s t i c a l dif ferences were occurr ing (df = 4, F = 28.58, p < 0.05). 1983 Ca l l Structure Control Periods During control and p r e - t r i a l per iods, the SI signal exhibi ted the fo l lowing signal cha rac te r i s t i cs . Frequency range of the signal extended from 100 Hz to 2 kHz, with most energy concentrated in a band of 300 to 825 Hz. The average received level of th is signal was 117.6 ± s . d . 14.2 dB re I juPa. Frequency modulation was noted to occur in 71.4% of the signals produced (30 signals out of 42). The average duration of th is c a l l was 1.7 + s . d . 0.8 sec. Gray whales t yp i ca l l y produced 9.4 + s . d . 4.4 pulses per se r i es . Calculat ions of number of pu lses/ser ies and the average ca l l duration suggested an average pulse repet i t ion rate of 5.4 per sec. B-6 Experiment A - Real Sources In the presence of real sound sources in the environment ( t r i a l experiments - Experiment A ) , the fol lowing structural changes occurred in the SI s i gna l . The received level of th is signal averaged 148.8 + s . d . 9.8 dB re 1 jjPa. Received levels documented during these t r i a l periods were s ign i f i can t l y higher than those observed during control periods (117.6 + s . d . 14.2 dB) (t = 11.7, p < 0.05). An increase was also noted in the number of SI s ignals exh ib i t ing frequency modulation during t r i a l periods calculated at 85.7%. This value was s i g n i f i c a n t l y higher than the 71.4% value obtained during control periods (Chi square = 5.24, p < 0.05). Although there was a general tendency for signal duration to increase in the presence of real sources (2.0 + s . d . 1.1 sec) , th is average value was not s t a t i s t i c a l l y d i f ferent from control values (1.7 + s . d . 0.8 sec) (t = 1.44, p > 0.05). The average number of pulses produced per ser ies during t r i a l periods was 24.6 + s . d . 8.0 pu lses /ser ies . The number of pu lses /ser ies produced during t r i a l periods was s i gn i f i can t l y higher than that obtained during control periods (9.4 + s . d . 4.4) (t = 10.7, p < 0.05). The increase noted in the number of pulses produced per ser ies and the fact that c a l l duration had not s ign i f i can t l y changed during t r i a l periods indicated that pulse repet i t ion rates also increased (average of 12 pu lses /sec) . A measurable change did not occur in the presence of real sources for overa l l frequency range of the signal (120-1800 Hz) and/or emphasized frequency range (280-700 Hz). An analysis of the pos t - t r i a l periods of Experiment A indicated that measurable dif ferences were s t i l l prevalent in the SI signal during th i s tes t per iod. The average received level was 128.3 + s . d . 6.6 dB re 1 p Pa. This pos t - t r i a l received level was s i gn i f i can t l y higher than control leve ls B-7 (117.6 + s . d . 14.2 dB) (t = 4.42, p < 0.05) but s i g n i f i c a n t l y lower when compared to t r i a l periods (148.8 + s . d . 9.8 dB) (t = 11.2, p < 0.05). An inspect ion of the number of s ignals that were frequency modulated during pos t - t r i a l periods resulted in 78.6%. This pos t - t r i a l percentage was not s t a t i s t i c a l l y d i f ferent from that obtained during contro ls (71.4%) (Chi square = 1.02, p > 0.05) and/or t r i a l periods (85.7%) (Chi square = 1.30, p > 0.05) . During pos t - t r i a l periods the number of pulses produced per ser ies was 18.5 + s . d . 8.4. The number of pu lses/ser ies was s ign i f i can t l y lower in pos t - t r i a l periods when compared to t r i a l periods (24.6 + s . d . 8.0) (t = 3.41, p < 0.05). A comparison between pre (9.4 + s . d . 4.4) and post t r i a l periods suggested that a s ign i f i can t di f ference also occurred in the number of pulses produced per ser ies for these events as well (t = 1.46, p < 0.05) with higher values obtained during pos t - t r i a l periods (18.5 + s . d . 8 .4) . Ca l l duration (2.0 + s . d . 1.0 sec) was not s i gn i f i can t l y d i f ferent between post and t r i a l periods (2.0 + s . d . 1.1) (t = 0.227, p > 0.05) and/or post and p re - t r i a l periods (1.7 + s . d . 0.8) (t = 1.25, p > 0.05) . No apparent dif ferences were found in pos t - t r i a l periods for overa l l frequency range (120-1950 Hz) and/or emphasized frequencies (315-820 Hz) of the s i g n a l . S imi lar values with respect to frequency ranges were obtained for a l l three test per iods. Based upon the average number of pu lses/ser ies and the average ca l l durat ion, an average repet i t ion rate of 9.25 pulses/sec was calculated for pos t - t r i a l per iods. This value was somewhat lower than t r i a l periods (x = 12) but higher than control periods (x = 5.4) . Experiment B - Outboard Engine Noise During the project ion of outboard engine noise (Experiment B t r i a l per iods) , the fo l lowing st ructura l modif ications were documented in the SI s i g n a l . An increase was once again noted in received level of SI signals B-8 with a value of 156.6 ± s . d . 9.5 dB re 1 / i P a . This t r i a l value was s ign i f i can t l y higher than the received level obtained during control periods (117.6 + s . d . 14.2) (t = 11.7, p < 0.05). The percentage of s ignals exh ib i t ing frequency modulation was 83.3% during t r i a l periods vs 71.4% obtained during control per iods. These values were shown not be be s t a t i s t i c a l l y d i f ferent (Chi square = 3.36, p > 0.05), but were s ign i f i can t at the 0.06 l e v e l . Average ca l l duration was 2.0 + s . d . 0.9 sec during t r i a l experiments. This value was not s ign i f i can t l y d i f fe rent from that obtained during p re - t r i a l periods (1.7 + s . d . 0.8 sec) (t = 1.44, p > 0.05). The number of pulses produced per ser ies during t r i a l periods was 22.6 +• s . d . 6.8. This value was markedly higher than the value obtained during control periods (9.4 + s . d . 4.4) (t = 10.7, p < 0.05). Repet i t ion rates (calculated from the average number of pu lses/ser ies and average c a l l duration) was 11.3/sec. Again no changes were observed to occur in frequency range (150-2000 Hz) and/or frequencies emphasized (300-800 Hz). An inspect ion of the pos t - t r i a l values (Experiment B) revealed that s t ructura l changes were s t i l l occurring in the SI signal during th is outboard test per iod. The average received level was 121.5 ± s . d . 6.2 dB compared to 117.6 + s . d . 14.2 dB in control periods and 156.6 + s . d . 9.5 dB in t r i a l per iods. P o s t - t r i a l values were not s i g n i f i c a n t l y d i f ferent than control levels (t = 1.61, p > 0.05) but were s ign i f i can t l y lower than t r i a l periods (t = 20.0, p < 0.05). During pos t - t r i a l per iods, the number of s ignals exh ib i t ing frequency modulation was 73.8%. This pos t - t r i a l value was not s i g n i f i c a n t l y d i f ferent than p r e - t r i a l (71.4%) or t r i a l periods (83.3%) (Chi square = 4.34, p > 0.05). The number of pulses produced per ser ies during pos t - t r i a l periods was 13.2 + s . d . 5 .1 , which was s ign i f i can t l y higher than p re - t r i a l s (9.4 + s . d . 4.4) (t = 3.64, p < B-9 0.05) and s i g n i f i c a n t l y lower than t r i a l periods (22.6 + s . d . 6.8) (t = 7.09, p < 0.05). Average c a l l duration during pos t - t r i a l s was 1.9 + s . d . 0.9 sec. S ign i f i cant dif ferences were not found when pos t - t r i a l ca l l duration was compared to p re - t r i a l c a l l duration (1.7 + s . d . 0.8 sec) (t = 1.22, p > 0.05) and/or t r i a l periods (2.0 ± s . d . 0.9 sec) (t.= 0.195, p > 0.05). An average repet i t ion rate during pos t - t r i a l s was 6.7/sec, somewhat lower than t r i a l rates (x = 11.3) but s im i la r to p re - t r i a l rates (x = 5.4). Frequency range (200-1800 Hz) and emphasized frequencies (350-780 Hz) were s im i la r in a l l three test per iods. Experiment C - Gray Whale Sounds During t r i a l periods when gray whale sounds were projected (Experiment C) , s t ructural changes in the SI signal were once again documented. When compared to p r e - t r i a l (117.6 + s . d . 14.2), a s i gn i f i can t increase (t = 3.08, p < 0.05) was noted in received l eve l s , averaging 126.4 + s . d . 11.7 dB re 1 yuPa. The number of s ignals exh ib i t ing frequency modulation was 95.2%, a value s ign i f i can t l y higher than control signals (71.4%) (Chi square = 18.43, p < 0.05) . The number of pu lses/ser ies was 12.8 + s . d . 6.7, which was also s ign f ican t ly higher than control values of 9.4 + s . d . 4.4 pulses/ ser ies (t = 2.77, p < 0.05). Once again, c a l l duration of 1.9 + s . d . 0.9 sec ( t r i a l periods) was not s i g n i f i c a n t l y d i f fe rent from controls (1.7 + s . d . 0.8 sec) (t = 1.08, p > 0.05). A repet i t ion rate of 6.6 pulses/sec was calculated for t r i a l per iods, a value close to that obtained in control periods (5.4) . No obvious di f ferences occurred between control and t r i a l periods with respect to overa l l frequency range (100-2000 Hz) or emphasized frequency bands of the s igna ls (300-790 Hz). During p o s t - t r i a l periods (Experiment C) , received levels were noted B-10 to be 119.7 + s . d . 6.6 dB re 1 / jPa. This pos t - t r i a l average value was not s ign i f i can t l y d i f ferent from p r e - t r i a l values (117.6 + s . d . 14.2) (t = 0.857, p > 0.05) but was s i g n i f i c a n t l y lower (t = 3.22, p < 0.05) than t r i a l periods (126.4 + s . d . 11.7 dB). P o s t - t r i a l s ignals exh ib i t ing frequency modulation were calculated at 73.8%. This value was not s i gn i f i can t l y d i f ferent from control values of 71.4% (Chi square = 0.50, p > 0.05) but was s i g n i f i c a n t l y lower than the 95.2% obtained during t r i a l periods (Chi square = 15.88, p < 0.05). During these pos t - t r i a l periods the number of pu lses/ser ies was 10.8 + s . d . 5 .0 . This average value when compared to p re - t r i a l periods (9.4 ± s . d . 4.4) was not s t a t i s t i c a l l y d i f ferent (t = 1.42, p > 0.05). When the average number of pu lses /ser ies was compared between post and t r i a l periods (12.8 +• s . d . 6 .7) , once again s ign i f i can t dif ferences could not be found (t = 1.52, p > 0.05). No s ign i f i can t di f ference could be detected in pos t - t r i a l ca l l duration (1.9 + s . d . 0.8 sec) between pre (1.7 + s . d . 0.8 sec) and pos t - t r i a l s (t = 1.13, p > 0.05) and post and t r i a l periods (1.9 + s . d . 0.9 sec) (t = 6.01, p > 0.05). An average repet i t ion rate during pos t - t r i a l s was 5.4 pulses/sec. A l l three test periods were s im i la r with respect to repet i t ion ra tes . The overal l frequency range of the signal during pos t - t r i a l s was 125-1950 Hz and frequency bands of 320-820 Hz were emphasized. These frequency ranges were s im i la r to those obtained during t r i a l and p re - t r i a l t imes. Experiment D - Test Tone Signals Since a l l vocal izat ions ceased during the project ion of the test tone (Experiment D) s t ructura l comparisons, of course, could not be accomplished. An inspection of the pos t - t r i a l periods of Experiment D did suggest that when gray whales began to vocal ize once again B-11 (approximately 10 minutes af ter the termination of the t r i a l period) the structure of the SI signal was s im i la r to that described for control per iods. The average received level of the s ignals during p o s t - t r i a l s was 115.4 + s . d . 9.3 dB re 1 juPa. When compared to the received values obtained during control periods (117.6 t s . d . 14.2 dB), a s ign i f i can t d i f ference in ca l l level did not occur (t = 0.864, p > 0.05). When the number of s ignals exh ib i t ing frequency modulation were compared between pre- and p o s t - t r i a l s , s ign i f i can t di f ferences could not be documented. Control and pos t - t r i a l values were both estimated at 71.4%. During p o s t - t r i a l per iods, an average of 9.6 + s . d . 4.9 pulses were produced per s e r i e s . This value was not s i g n i f i c a n t l y d i f ferent from the value of 9.4 + s . d . 4.4 obtained during control periods (t = 0.254, p > 0.05). As in a l l other comparisons, ca l l duration during pos t - t r i a l periods (1.8 + s . d . 0.8 sec) was not s i gn i f i can t l y d i f ferent from control periods (1.7 + 0.8 sec) (t = 0.357, p > 0.05). Repeti t ion rates calculated for pos t - t r i a l periods were 5.3 pulses/sec, which were s im i la r to the values obtained during p r e - t r i a l periods (5.4 pu lses /sec) . Overal l frequency range of the signal was 125-2000 Hz and the frequency bands of 300-800 Hz were emphasized. No apparent frequency sh i f t took place in these pos t - t r i a l per iods. 1983 Structural Comparisons of Ca l l s  Received Level of Signals Received levels of the SI signals were s ign i f i can t l y d i f ferent between the t r i a l periods of Experiment A (real sources = 148.8 + s . d . 9.8 dB) and Experiment B (outboard playback = 156.6 ± s . d . 9.5 dB) at t = 3.7, p < 0.05. When comparing the t r i a l experiments of real sources and gray whale playback (126.4 + s . d . 11.7 dB), once again s ign i f i can t di f ferences were obtained (t = 9.52, p < 0.05). A comparison of the received levels of gray whale B-12 c a l l s between the project ion of outboard engine noise (156.6 + s . d . 9.5 dB) and gray whale signals (126.4 ± s . d . 11.7) were shown to d i f f e r as well (t = 13.0, p < 0.05). The level of the received c a l l was shown to vary depending upon the sound st imul i present in the environment. When comparing the received levels of ca l l s among the four experimental pos t - t r i a l periods ( inc luding test tone p o s t - t r i a l s ) , major di f ferences were documented. These data are summarized as fo l lows: Real (128.3 + s . d . 6.6 db) vs outboard (121.5 + s . d . 6.2 dB) (t = 4 .8 , p < 0.05); real (128.3 +• s . d . 6.6 dB) vs gray whale playback (119.7 + s . d . 6.6 dB) (t = 5.96, p < 0.05); real (128.3 ± s . d . 6.6 dB) vs test tone (115.4 + s . d . 9.3 dB) (t = 7.35, p < 0.05); outboard (121.5 + s . d . 6.2 dB) vs tes t tone ( 115.4 + s . d . 9.3 dB) (t = 3.55, p < 0.05); and gray whale (119.7 + s . d . 6.6) vs test tone (115.4 + s . d . 9.3) (t = 2.46, p < 0.05). An exception occurred when pos t - t r i a l periods were compared between outboard (121.5 + s . d . 6.2) and gray whale sounds (119.7 + s . d . 6 .6) . In th is case s ign i f i can t di f ferences could not be found (t = 1.27, p > 0.05). An inspect ion of the average levels received during these pos t - t r i a l periods suggested that gray whales resume basel ine/contro l levels at a faster rate when exposed to the i r own signals and/or the rapid onset of outboard engine noise. P o s t - t r i a l levels were greatest fol lowing the presence of real sources in the environment (128.3 ± s . d . 6.6 dB), suggesting a slower recovery ra te . After exposure to the test tone (although s i l en t for 10 minutes), gray whale received levels (115.4 + s . d . 9.3) were s im i la r to those obtained in control periods (117.6 + s . d . 14.2 dB). The resul ts of these analyses suggested that the post-exposure received levels of ca l l s also varied with the type of sound st imul i that had been present in the envi ronment. B-13 Frequency Modulation Within Signals When invest igat ing the number of signals exhib i t ing frequency modulation during various t r i a l per iods, the analyses resulted in the fo l low ing . When t r i a l periods of real (85.7%) and outboard (83.3%) were compared, di f ferences were not found (Chi square = 0.074, p > 0.05) . However, when real (85.7%) and gray whale playback periods (95.2%) and outboard (83.3%) and gray whale playback periods (95.2%) were compared, s t a t i s t i c a l di f ferences were noted (Chi square = 4.02, p < 0.05 and Chi square = 6.08, p < 0.05, respec t i ve ly ) . In both cases, the number of signals exh ib i t ing frequency modulation were highest during gray whale playback periods at 95.2%. An analysis of pos t - t r i a l s with respect to the number of signals that were frequency modulated suggested that the values obtained for the various experiments were not d i f ferent (Chi square = 0.845, p > 0.05). The above pos t - t r i a l comparisons suggest that the number of s ignals that are frequency modulated af ter exposure to increased noise levels i s bas ica l ly s im i la r among various sound s t i m u l i . Ca i l Duration For a l l comparisons conducted on c a l l durat ion, s ign i f i can t di f ferences could not be found among the t r i a l or pos t - t r i a l periods among the various experiments. The s t a t i s t i c a l values calculated for these c a l l duration values are l i s t ed as fo l lows: I) t r i a l comparisons: real (2.0 f s . d . 1.1 sec) vs outboard (2.0 + s . d . 0.9 sec) (t = 0.183, p > 0.05); real (2.0 + s . d . 1.1 sec) vs gray whale (1.9 + s . d . 0.9 sec) (t = 0.457, p > 0.05); outboard (2.0 + s . d . 0.9) vs gray whale (1.9 ± s . d . 0.9 sec) (t = 0.30, p > 0.05); and 2) pos t - t r i a l comparisons: real (2.0 ±_s.d. 1.0 sec) vs outboard (1.9 + s . d . 0.9 sec) (t = 0.127, p > 0.05); real (2.0 + s . d . 1.0 sec) vs gray whale (1.9 + s . d . 0.8) (t = 0.169, p > 0.05); real ( 2.0 + s . d . 1.0 B-14 sec) vs test tone (1.8 + s . d . 0.8) (t = 0.914, p > 0.05); outboard (1.9 + s . d . 0.9) vs gray whale (1.9 + s . d . 0.8) (t = 0.04, p > 0.05); outboard (1.9 + s . d . 0.9) vs test tone (1.8 + s . d . 0.8) (t = 0.858, p > 0.05); and gray whale (1.9 + s . d . 0.8) vs test tone (1.8 + s . d . 0.8 sec) (t = 0.780, p > 0.05). Cal l duration was s im i la r for a l l experiments. Number of Pulses Within Signals A comparison of the number of pulses produced per ser ies among d i f ferent experiments y ie lded the fol lowing resu l t s . S t a t i s t i c a l di f ferences did not occur when the t r i a l periods of real (24.6 ± s . d . 8.0) and outboard playback (22.6 + s . d . 6.8) were compared (t = 1.25, p > 0.05). However, values obtained during the comparisons of t r i a l periods between real (24.6 + s . d . 8.0) and gray whale playback (12.8 + s . d . 6.7) (t = 7.26, p < 0.05) and outboard (22.6 + s . d . 6.8) and gray whale (12.8 + s . d . 6.7) (t = 6.56, p < 0.50) were d i f fe ren t . When an inspection of the number of pulses produced per ser ies was made among various pos t - t r i a l per iods, s t a t i s t i c a l di f ferences were found for each comparison as fo l lows: p o s t - t r i a l comparisons: real (18.4 + s . d . 8.4) vs outboard (13.2 + s . d . 5.1) (t = 3.46, p < 0.05); real (18.4 + s . d . 8.4) vs gray whale playback (10.8 + s . d . 5.0) (t = 5.01, p < 0.05); real (18.4 + s . d . 8.4) vs test tone (9.6 + s . d . 4.9) (t = 5.84, p < 0.05); outboard (13.2 + s . d . 5.1) vs gray whale (10.8 + s . d . 5.0) (t = 2.10, p < 0.05); and outboard (13.2 + s . d . 5.1) vs test tone (9.6 + s . d . 4.4) (t = 3.22, p < 0.05). An exception occurred when gray whale pos t - t r i a l s (13.2 + s . d . 5.1) were compared to test tone p o s t - t r i a l s (9.6 + s . d . 4 .9 ) . These values were not s i g n i f i c a n t l y d i f ferent (t = 1.11, p > 0.05). Again, the exposure to d i f ferent sound st imul i resulted in d i f ferent rates of decl ine during pos t - t r i a l events. B -15 Frequency Range and Frequencies Emphasized No apparent di f ferences were observed among experiments with respect to the overal l frequency range of the signal and the frequencies emphasized in the s igna l . C-1 A P P E N D I X C C-2 APPENDIX C 1984 Ca l l i ng Rates In 1984, 40 hours were spent co l l ec t i ng the sounds of gray whales in the absence of any noise disturbance (= control per iods) . Under control condi t ions, gray whales produced an average of 81.4 + s . d . 16.6 sounds/hour. When exposed to real sources in the environment (Experiment A; 40 hours), a change in c a l l i n g rate was documented. Ca l l i ng rates increased in the presence of real sources averaging 159.6 + s . d . 35.5/hour, a value s ign i f i can t l y higher than control periods (t = 12.59; p < 0.05). When gray whales were subjected to outboard engine noise (Experiment B; 40 hours), c a l l i n g rates averaged 137.9 + s . d . 24.7 sounds/hour. This experimental value ref lected a s ign i f i can t increase in c a l l i n g rates when compared to control values (t = 11.9; p < 0.05). In the presence of o i l - d r i l l i n g sounds (Experiment C; 40 hours), a major decrease in sound ac t i v i t y was documented. An average c a l l rate of 14.9 ± s . d . 7.4 sounds per hour was documented in the presence of th is sound source. When compared to control values (81.4 ± s . d . 16.6), a s ign i f i can t decrease in ca l l i ng rates was noted during the project ion of o i l - d r i l l i n g sounds (t = 23.0; p < 0.05). The project ion of k i l l e r whale sounds (Experiment D; 40 hours) created a greater ef fect on the acoust ical behavior of gray whales than the resul ts obtained in the presence of o i l - d r i l l i n g sounds. When these predatory signals were projected, a l l vocal izat ions by gray whales ceased. To ver i fy that gray whales were not responding to the short c l i ck produced when the sound equipment was turned on, and/or the possible acoust ical energy produced by the transducer, th is equipment was f u l l y C-3 operated without project ion of sounds (Experiment E; 40 hours). C a l l i n g rates obtained during th is experiment averaged 78.5 + s . d . 17.0. When these rates of c a l l i n g were compared to control periods (81.4 ± s . d . 16.6), a s ign i f i can t di f ference could not be obtained (t = 0.77; p > 0.05). 1984 Ca l l Structure Since the SI signal continued to dominate the acoust ical repertoire of the gray whale during the 1984 inves t iga t ions , s t ructura l comparisons were once again conducted on th is ca l l type. As in 1983, s ix sound parameters were measured and compared among the experimental events: frequency range of signal (Hz); range of emphasized frequencies (Hz); received levels of sound (dB re 1 / jPa); percentage of SI s ignals exh ib i t ing frequency modulation; duration (sec) ; and number of pulses produced per se r ies . Ca l l repet i t ion rates (number/sec) were derived from the average duration and average number of pulses produced per se r i es . Two sounds per hour for each experimental grouping were randomly selected and analyzed; thus a to ta l of 80 sounds for each experimental condit ion were examined. Control Periods During control per iods, the SI signal exhibi ted the fol lowing signal cha rac te r i s t i c s . Frequency range of the signal extended from 100 Hz to 2000 Hz, with most energy concentrated in a band of 300 to 850 Hz. The average received level of th is signal was 120.3 + s . d . 17.6 dB re 1 juPa. An inspect ion of the number of signals showing frequency modulation was 68.7%. Average signal duration was 1.9 + s . d . 0.8 sec. Gray whales t yp i ca l l y produced 9.9 + s . d . 4.0 pulses per se r ies . The calculated average pulse repet i t ion rate was 5.2 per sec. In the presence of increased noise l eve l s , marked dif ferences occurred in the structure C-4 of the SI signal when compared to control per iods. These resul ts are presented below. Experiment A - Real Sources In the presence of real sound sources (Experiment A ) , the fol lowing changes occurred in the SI s i g n a l . The received level of th is ca l l was noted to increase and averaged 148.5 + s . d . 12.2, a value s i gn i f i can t l y higher than that obtained during control periods (120.3 + s . d . 17.6) (t = 11.7; p < 0.05). The number of SI signals exh ib i t ing frequency modulation was 90%, compared to the 68.7% obtained during control per iods. When compared, these two values were s ign i f i can t l y d i f fe ren t , with frequency modulation occurring more often in the presence of real sources (Chi Square = 12.6; p < 0.05). Ca l l duration averaged 2.0 + s . d . 1.0 sec during Experiment A and was not shown to d i f f e r from that of control periods (1.9 + s . d . 0.8) (t = 0.89; p > 0.05). An increase was noted to occur in the number of pulses produced per s e r i e s , averaging 25.4 + s . d . 8.5 pulses. This value was s ign i f i can t l y higher (t = 14.6; p < 0.05) than that obtained during control periods (9.9 + s . d . 4 .0 ) . The derived repet i t ion rate, based on average c a l l duration and the average number of pulses per se r ies , was 12.7 pulses/sec, a value s i gn i f i can t l y higher than that obtained during control periods (5.2 pulses per sec) . A measurable change did not occur in the presence of real sources for overal l frequency range of the signal (150-1800 Hz) and/or emphasized frequency bands of the s ignal (300-800 Hz). Experiment B - Outboard Engine Noise During the project ion of outboard engine noise (Experiment B ) , a measurable change was noted in the received level of the c a l l , averaging 150.2 + s . d . 12.8 dB re 1 yuPa. This value was s i gn i f i can t l y higher than that obtained during control periods (120.3 + s . d . 17.6) (t = 12.2; p < 0.05). In the presence of outboard engine noise, the number of s ignals showing frequency modulation was 86.2%. When compared to control values of 68.7%, th i s experimental value was found to be s t a t i s t i c a l l y higher (Chi Square = 7.8; p < 0.05). During these experiments, average c a l l duration was 2.0 + s . d . 0.8 sec. When compared to controls (1.9 + s . d . 0 .8) , a s ign i f i can t di f ference in ca l l duration was not detected (t = 1.06; p > 0.05). The number of pulses produced per ser ies averaged 23.1 + s . d . 6.9, a value s i gn i f i can t l y higher than that obtained during control periods (9.9 + s . d . 4.0) (t = 14.6; p < 0.05) . The average repet i t ion rate calculated for th is experimental period was 11.5 pulses per sec, a considerable increase over that of 5.2 pulses per sec obtained during control per iods. Once again, a detectable dif ference did not occur when experimental periods for frequency range of signal (120-1900 Hz) and/or emphasized frequency bands (320-860 Hz) were compared to control per iods. Experiment C - O i l - D r i l l i n g Sounds During the project ion of o i l - d r i l l i n g sounds (Experiment C ) , gray whale ca l l structure was shown to d i f f e r from that described for control per iods. Received levels averaged 143.1 + s . d . 9.4 dB re 1^uPa. This value was s i gn i f i can t l y higher than control values (120.3 + s . d . 17.6) (t =10 .2 ; p < 0.05) . The number of signals exh ib i t ing frequency modulation was 92.5% vs the 68.7% obtained during control per iods. S t a t i s t i c a l di f ferences were documented when these two values were compared (Chi Square = 16.5; p < 0.05). No apparent di f ferences (t = 0 .2 ; p > 0.05) were found when ca l l duration was compared between experimental (1.8 + s . d . 0.9 sec) and control (1.9 ± s . d . 0.8) per iods. C-6 The number of pulses produced per ser ies averaged 25.4 ± s . d . 7.4, which re f lec ted a s ign i f i can t l y higher pulse rate than that noted during control periods (9.9 ± s . d . 4.0) (t = 16.2; p < 0.05). The average repet i t ion rate of the SI signal during the project ion of o i l - d r i l l i n g sounds was 14.1 pulses produced per sec, a value markedly higher than control periods (5.2 pulses per sec) . Frequency range of the signal (110-2000 Hz) and emphasized frequency bands (310-850 Hz) during exposure to o i l - d r i l l i n g sounds were s im i la r to control per iods. Experiment D - K i l l e r Whale Sounds Since the project ion of k i l l e r whale sounds resulted in a complete cessation of a l l gray whale voca l i za t ion , st ructural comparisons could not be accomplished for th is experimental condit ion (Experiment D). Experiment E - Equipment On; No Sounds Produced To test whether gray whales were responding to the acoust ical equipment used during these playback t r i a l s , comparisons of c a l l structure between Experiment E (equipment on, no sounds produced) and control periods were accomplished. S ign i f i cant dif ferences could not be obtained for any of the acoust ical var iables measured, suggesting that the acoust ical equipment i t s e l f had no ef fect on the behavior of gray whales. These comparisons are summarized below. Received levels (Experiment E) averaged 119.6 + s . d . 14.3 vs control (120.3 + s . d . 17.6) periods (t = 0.2; p > 0.05); number of signals exhib i t ing frequency modulation during experimental periods (66.2%) vs control (68.7%) (Chi Square = .05; p > 0.05); c a l l duration averaged 1.9 + s . d . 0.8 during experimental test ing and 1.9 + s . d . 0.8 during control periods (t = 0.04; p > 0.05); number of pulses produced per ser ies for experimental (9.3 + s . d . 5.4) vs 9.9 + s . d . 4.0 for control periods (t = 0.7; p > 0.05); repet i t ion rates of 4.9 per sec were calculated C-7 fo r Experiment E and compared to the 5.2 per sec obtained during control periods, and were found to be s im i l a r . The overal l frequency range of the signal was 140-1800 Hz and emphasized bands were 340-860 Hz during the test periods of Experiment E. No apparent sh i f t occurred with respect to frequency values for experimental vs control per iods. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0097975/manifest

Comment

Related Items