Focal animal and focal group follows provide the framework for data collection focusing on dolphin signal exchange and contact behavior. Simultaneous video and stereo audio recording of identified dolphins and real-time, all-occurrence documentation of specific behaviors and group activity provide a data record. A mobile video/acoustic system (MVA) is used to record dolphin behavior and sound. Definitions include:
-> Group – The “total available dolphins” observed within an imaginary ellipse, with a 200 m long axis & 100 m short axis for dimensions. At Mikura, the long axis placed parallel to the coastline.
-> Subgroup – Dolphins observed in apparent association, moving in the same general direction, although not all individuals necessarily heading in the same direction, and engaged in the same activity. Spatially, each individual in a single subgroup is within at least five dolphin-body widths of its nearest conspecific.
-> Short Encounter – Any underwater observations of dolphins between 30 s & 180 s. Underwater observations of dolphins that are greater than 180 s become true encounters by definition
The Mobile Video/Acoustic System–MVA (a.k.a. "array")
The mobile video/acoustic system (MVA) was developed to permit real-time synchronous recording of the sounds and behaviors of individual, free-swimming dolphins. Manually operated under water, the system consists of two omni-directional hydrophones cabled through a custom underwater housing into a stereo video camera. Hydrophone spacing on the housing is scaled to the human inter-aural distance based upon the speed of sound in water (4.5 times faster than in air). Location of the vocalizing dolphin is based upon associating visual distribution of animals with directions to sound sources as determined by aural psycho-acoustics.
Dudzinski and Gregg both used an ECD – echolocation click detector – that was added to the original design in a small second housing containing a digital audio (or recently an M-Audio) recorder together with a pre-processor circuit that allowed detection of the highly directional part of a dolphin’s echolocation ‘clicks’ and makes these audible and recordable.
Dudzinski affectionately calls the MVA her “array”–since it contains at least two hydrophones.
Read the publication on the creation and testing of the MVA.
Significance & Application
Observations of dolphin behavior are enhanced through the use of video recordings; valuable information is preserved and the dynamics of inter-animal relations, associations, behavior, and sociality can be examined. These data can be analyzed for evidence of acoustic, visual, and tactile signal exchange and behavioral coordination within varying social contexts. More importantly, comparisons of specific behaviors between individuals and groups of dolphins are facilitated. Video recordings are especially useful for documentation, and subsequent analysis, of subtle within-group signaling of dolphins underwater, where they spend over 85% of their time. Inter-observer reliability is also possible from videotaped records, thus providing a set of observations able to withstand the rigors of statistical machinations. Video records enable an examination of visually and acoustically described behavior that provide insight on the importance of contextual cues for how dolphin signal-receivers respond to signalers. Conclusions may then be drawn concerning the interplay between vocalizations, social behavior, and contextual variables.
In summer 2001, DCP increased the capability of our click detector (ECD) by adding a second frequency range. We use two circuit boards – one set at 70 kHz and one set at 120 kHz. This allows greater range to better determine how dolphins use their click (i.e., pulsed) sounds. The ECD boards were designed by Paul Lepper and Daryl Newborough of New Leap Ltd.
The following is the Abstract from Dudzinski et al’s presentation at the 30th Annual Meeting of the European Association for Aquatic Mammals held from 8-11 March 2002 in Aalborg, Denmark. For more information on the symposium program, go to www.eaam.org.
An examination of two frequency bands for echolocation click signals recorded from two species of wild dolphins with respect to behavior and potential function
Kathleen M. Dudzinski1, Paul Lepper2,3, Darryl Newborough3
1Dolphin Communication Project, 3600 S. Harbor Blvd., #429, Oxnard, CA 93035 USA
2Scripps Institute of Oceanography, Marine Physical Laboratory, Bldg. 4, San
Diego, CA 92106 USA
3New Leap Ltd., United Kingdom
ABSTRACT
Dolphin echolocation signals from two frequency bands have been processed for both wild bottlenose (Tursiops truncatus) and Atlantic spotted dolphins (Stenella frontalis) recorded around Mikura Island, Japan and in the Bahamas, respectively. These data are closely correlated with underwater video footage showing animal behavior around the acoustic receivers. Echolocation signals were detected for both species using a single wide-band hydrophone (-164 dB re 1V/uPa) and then processed in-situ using an envelope detection technique for energy content in two frequency bands centered on 70 kHz and 120 kHz. Each band had an approximate 15 kHz bandwidth with good rejection of energy between each band. This method allowed detailed analysis of variations in inter-click period for particular behaviors and changes in the click energy spectral distribution between the two frequency bands of interest. Variations in click rate are seen in both bands and appear to be correlated with changes in behavior. The shift in energy between frequencies is potentially related to specific dolphin head movements (e.g., head scanning); however, other behaviors were also concurrently documented with observed shifts in energy between both bands. This suggests that dolphins are capable of voluntary control over the energy content and repetition rate of high frequency click signals leading to the potential for communication among group members as one function for click signals.
DCP's Protocol for Behavioral Data Collection (Observations & Recordings)
The following is a narrative related to DCP’s protocol for protocol for our behavioral and acoustic data collection. We use this protocol consistently between our field sites to document dolphin interactions and signal exchange. We maintain as non-invasive a process as possible.
Our studies into dolphin communication focus on signal exchange and use of tactile behaviors between individuals. DCP also promotes a comparative aspect to our studies. Because we use the same protocol for data collection at each of our field sites, we can directly compare data collected on each of our study groups. Some of the various research objectives and hypotheses are annotated in the following text.
Objectives
Hypotheses (examples as related to objectives)
H01: Bottlenose dolphins show no differential use among subgroups or groups in the shallow, near-shore waters near Mikura Island.
HA1: Bottlenose dolphins use the shallow, near-shore area around Mikura Island for different behavioral activities depending on subgroup type and size.
HA2: Bottlenose dolphins use the shallow, near-shore area around Mikura Island for foraging, social and play activities without regard to subgroup type and size.
H02: Contact behavior between bottlenose dolphins does not differ in use and expression according to dolphin age, sex or associates.
HA1: Contact behaviors differ in frequency related to the behavioral activity of each subgroup.
HA2: Contact behaviors differ in use and frequency according to group type and the age and sex of identified dolphins.
HA3a: The same contact behaviors are modified by dolphin postures within different behavioral activities, and thus will send varying messages (e.g., play versus aggression).
HO3: Sounds produced by bottlenose dolphins are not related to dolphin identification, age, sex or to subgroup behavioral activity.
HA1: Identified bottlenose dolphins possess individually distinct frequency-modulated pure tones (e.g., whistles).
HA2: Frequency-modulated pure tones vary according to dolphin age and sex but not necessarily to individual or behavioral activity.
HA3: Frequency-modulated pure tones vary according behavioral activity but not necessarily to dolphin identification, age or sex.
HA4: Pulsed vocalizations (e.g., clicks, squawks) vary according to the behavioral activity of subgroups, but not necessarily to subgroup type.
H04: All pulsed sounds possess a signal in the 120 kHz range and are used for investigation of the surrounding environment and for foraging only.
HA1: Pulsed sounds are used to exchange communicative as well as investigative information. That is, pulsed sounds are used in non-foraging and non-exploratory behaviors.
HA2: Only the click or pulsed sounds used during foraging, or exploratory behaviors possess a signal at about 120 kHz.
Related comparative questions focus on the similarities and differences in use and production of contact behavior and sounds as well as signal exchange between individuals between dolphins. Do individuals from both groups exhibit similar group types? Do they exhibit similar behavioral activities in their given habitats? Do individuals within both study populations use sounds differently? Do they interact with conspecifics similarly given similar group types or behavioral activities? Could the difference in habitat for the two locations – far from shore, white sand in the Bahamas versus near-shore, boulder-covered sea floor around Mikura – play a significant factor in any observed differences in behavioral activity or signal exchange?
Data Collection
Focal animal and focal group follows provide the framework for data collection focusing on dolphin signal exchange and contact behavior – communication. Two methods of collection will be followed for each swim encounter with dolphins: 1) simultaneous video and audio recording of identified dolphins; and 2) real-time, all-occurrence documentation of specific behaviors and group activity. Real-time, all-occurrence sampling will provide information on the overall behavioral activities of the dolphins. Information on the associations of vocal and tactile/visual signs of dolphins will be obtained primarily from video recordings, while corroborative evidence will be gathered in the form of anecdotal observations.
Focal individuals will be chosen opportunistically depending on individuals observed during swim encounters. The decision of which dolphin to watch is by chance with attention to an even representation of all age and both sexes. A focal session will start as soon as the observer is in position underwater and the group composition is assessed. Video recording will continue until the animal(s) leaves the field of view. Thus, duration of focal follows is dictated by dolphin behavior: video observations on individuals will continue for as long as possible while in the field.
A mobile video/acoustic system (MVA) will be used to record dolphin behavior and sounds (Dudzinski et al., 1995). The system is swimmer-propelled and facilitates localization of dolphin sound sources by associating video data of animal distributions with audio data from two hydrophones spaced relative to the human inter-aural distance as scaled to speed of sound in water. A small second housing containing a Sony TCD D8 digital recorder together with a pre-processor circuit which detects the highly directional part of a dolphin’s echolocation ‘clicks’ and makes these audible and recordable will also be used.