Acoustic telemetry is an important tool for studying the behavior of aquatic animals and assessing the environmental impact of structures such as hydropower facilities. However, the physical size, signal intensity, and service life of off-the-shelf transmitters are presently insufficient for monitoring certain species. In this study, we developed a small, long-life acoustic transmitter with an approximate length of 24.2 mm, diameter of 5.0 mm, and dry weight of 0.72 g. The transmitter generates a coded acoustic signal at 416.7 kHz with a selectable source level between 159 and 163 dB relative to 1 μPa at 1 m, allowing a theoretical detection range of up to 500 m. The expected operational lifetime is 1 yr at a pulse rate interval of 15 s. The new technology makes long-term acoustic telemetry studies of small fish possible, and is being deployed for a long-term tracking of juvenile sturgeon.

Acoustic telemetry1,2 uses small transmitters (or tags) to monitor the behavior of fish.3,4 The tags transmit a unique acoustic signal to a network of receivers (hydrophones) at a selected pulse rate interval (PRI). The receivers detect these signals and extract the arrival time and identification code from the acoustic waveform.5 After post-processing the data and applying time-difference-of-arrival techniques, individual hosts are localized in three dimensions and tracked through the coverage area.6 The resulting tracks are then used to estimate the survival rates and behavior of fish passing through hydropower facilities.

There are three key parameters that determine the usability of an acoustic tag for a certain application:

  1. Tag size and weight. To reduce the tag burden (ratio of transmitter weight to fish weight) on tagged fish, acoustic fish tags are always preferred to be as light and small as possible.

  2. Acoustic signal strength. High signal strength is always desirable for greater detection range.

  3. Service life. A long service life is often desired for a long-term monitoring of certain species.

A wide range of acoustic tags are available for different host species and measurement objectives. For instance, the Juvenile Salmon Acoustic Telemetry System (JSATS),7–9 developed by the U.S. Army Corps of Engineers Portland District, employs acoustic transmitters and receiving systems to observe and assess the survival rates for juvenile Chinook salmon passing through the Federal Columbia River Power System. Various manufacturers offer JSATS-compatible tags with service lives of about 20 days at 3-s PRI, permitting the fish to be evaluated at multiple sites during migration. In a previous work, we developed a smaller tag that can be injected into the host rather than surgically implanted.10 This design reduces both the tag burden and deployment costs. The transmitter has a diameter of 3.38 mm, length of 15.00 mm, dry weight of 217 mg, and an average source level in the front 180° of about 155 dB (all values are relative to 1 μPa at 1 m). The operating lifetime is about 100 days at 3-s PRI.

To our knowledge, no solution presently exists for long-term monitoring of species such as juvenile (less than 1 yr old) sturgeon. This application requires a detection range of 500 m, combined with extended service life of up to 1 yr. The PT-1 acoustic transmitter from Sonotronics11 has a diameter of 7.1 mm, length of 16.0 mm, dry weight of 1.25 g, and detection range of 300 m. However, the service life is only 7 days at 1-s PRI, which is not suitable for long-term monitoring. The 795LG acoustic transmitter from HTI12 has an approximate lifetime of 220 days at 3-s PRI, or 400 days at 10-s PRI. The tag is 11 mm in diameter and 25 mm long. However, it is relatively heavy at 4.6 g and has a signal strength of just 152 dB. The Vemco V7 transmitter13 has a mass of 1.4 g in air, a diameter of 7 mm, a length of 18 mm, and a tag life of 220 days at 136 dB but with a transmission rate of every 90-s PRI. The injectable tag developed in the previous work has a significantly higher source level, enabling a detection range of up to 250 m in quiet environments. However, the detection range and service life are still insufficient. Therefore, new tag designs are needed to enhance the power source voltage, service life, and transmission range, and to reduce adverse effects and costs associated with thus broadening the range of potential environmental monitoring applications including, investigating behavior and habitat of juvenile sturgeon and other small species.

In this article, we report our effort to develop a new acoustic transmitter to address these needs. This design has a mass of less than 1 g and a service life of at least 1 yr at a PRI sufficient for tracking sturgeon. It also has a potential detection range of up to 500 m in quiet reservoir-type locations.

The newly developed acoustic transmitter includes three main components: (1) a piezoelectric transducer that transmits a coded acoustic signal, (2) an electronic circuit board that controls the operation of the transmitter, and (3) a custom lithium carbon fluoride battery which powers the components of the acoustic transmitter.

The piezoelectric transducer is a Type VI lead zirconate titanate (PZT) ceramic tube. The dimensions of the PZT were chosen to achieve mechanical resonance at or near 416.7 kHz, which is the modulation frequency used in JSATS. The tube has an outer diameter of 2.54 mm, an inner diameter of 1.80 mm, and a length of 2.65 mm. An electrode is present on the inner and outer wall surfaces. Because the circuit board situated behind the PZT tube (Figure 1) blocks the sound waves emitted from the back of the tube, the acoustic signals propagating in this direction are considered largely “wasted” as they will have a much lower probability of being detected.10 To use the input electrical energy more efficiently, the center of the inner circle of the tube is purposely offset from that of the outer circle along the tube’s radial direction by 0.15 mm, so that the tube has a gradually changing wall thickness. The thicker portion of the tube faces the circuit board and the thinner portion of the tube is oriented towards the front of the transmitter. As a voltage is applied to the tube, the thinner portion of the tube experiences a higher electric field, so a higher fraction of the output acoustic energy is emitted from the front of the transmitter.

FIG. 1.

Design of the long-lifetime acoustic transmitter: (a) computer-aided design drawing, (b) photograph of the actual tag.

FIG. 1.

Design of the long-lifetime acoustic transmitter: (a) computer-aided design drawing, (b) photograph of the actual tag.

Close modal

The previous injectable tag uses a custom Li-CFx micro-battery to provide high energy density in a suitable form factor.14 The new transmitter uses the same internal design with larger dimensions to increase the total capacity. The battery has a cylindrical aluminum case that is 4.7 mm wide and 14.8 mm long. The nominal capacity of the battery is 56 mAh with output voltage between 3.0 V and 1.8 V. For comparison, the previous battery was 3.3 mm wide and 6.0 mm long with a nominal capacity of 7 mAh.

Figure 2 illustrates the overall electronics design of the new transmitter. A small 8-bit microcontroller PIC16LF182315 generates the coded waveforms to drive the PZT transducer. The waveforms employ phase-shift keying (PSK) to transmit a unique 31-bit tag code, which includes a 7-bit header, a 16-bit identification number, and an 8-bit cyclic redundancy check (CRC).8 A 10 MHz ceramic resonator provides a stable clock source for the microcontroller during transmission so that the modulation frequency achieves the target frequency of 416.7 kHz within 0.5% accuracy.

FIG. 2.

Block diagram of the electronic design of the new acoustic tag.

FIG. 2.

Block diagram of the electronic design of the new acoustic tag.

Close modal

To meet the requirement of 500-m potential detection range, a much higher source level than that of the previous injectable tag is required. Therefore, a dual boost converter circuit was designed to transform the battery voltage into higher driving voltages for the PZT transducer. Normally a boost converter consists of a capacitor, an inductor, a Schottky diode, and a transistor switch. The dual boost converter used in the new transmitter (Figure 3) combines two circuits with a shared inductor (L1) to minimize component area and generate two output voltages. One output voltage is higher than the battery voltage (e.g., about +7.0 V), whereas the other output voltage is lower than the battery voltage (e.g., about −3.0 V). This configuration allows a higher peak-to-peak voltage from the tag circuit to be delivered across the transducer. For example, when the two voltage potentials delivered by the dual boost converter are +7.0 and −3.0 V, the effective voltage delivered from the tag circuit to the PZT is about 20.0 V peak-to-peak.

FIG. 3.

Schematic of the dual boost converter that uses a shared inductor to minimize component area.

FIG. 3.

Schematic of the dual boost converter that uses a shared inductor to minimize component area.

Close modal

Before transmitting a tag code, the microcontroller activates each portion of the boost converter circuit for a predetermined length of time (on the order of 50 ms) to generate the desired voltage potentials. Then, the microcontroller activates two analog switches that alternately connect the two voltages onto either electrode of the PZT. The desired waveform is then applied to the transducer at the modulation frequency of 416.7 kHz. An additional inductor placed in series with the PZT further increases the effective driving voltage by creating an electrical resonance with the fundamental capacitance of the transducer. This resonance inductor also improves the efficiency of the circuit.

During transmission, C1 and C3 supply the driving voltage and gradually decay as a result. To replenish the voltage, the microcontroller turns on the dual boost converter at the beginning of transmission, continuing for a predetermined time afterward.

The microcontroller firmware was developed in assembly language due to the number of time-critical operations required. The main module contains default configuration and executes instructions after the microcontroller reset, whereas the other modules define the various operations with the necessary routines. Upon reset, the microcontroller configures initial values of all the pins, and then goes into sleep mode to minimize current consumption. Applying a light source to an infrared sensor causes the microcontroller to wake up and prepare to receive additional data. A coded optical signal is then applied to the infrared sensor to specify configuration parameters such as the transmitted code and PRI. A special data transmission protocol ensures that the microcontroller is configured properly, as the infrared link provides no direct feedback. Upon success, the microcontroller may immediately start transmission or wait for a predetermined length of time first.

The new acoustic transmitter can automatically adjust the charging durations for the dual boost converter as the battery discharges such that the acoustic source level remains roughly constant over the majority of the operational lifetime. To this end, the configuration parameters include a table of charge durations for various battery voltages. A first set of values specify the charging durations for the lowest voltage range. A second set of values specify slightly shorter durations for the second lowest voltage range, and so forth. Higher values result in a higher source level but also increase energy consumption, allowing users to program the desired source level.

The developed long-life acoustic transmitter weighs approximately 720 mg in air. It possesses a primarily cylindrical body with a maximum diameter of 5.0 mm and a length of 24.2 mm (Figure 1). The front half of the tag is slimmed down in a relatively flat profile in order to reduce weight. Table I lists the source level and other parameters of the new transmitter.

TABLE I.

Exemplary parameters of the new acoustic tag.

Parameter Value
Dimensions  ø5.0 × 24.2 mm 
Dry weight  720 mg 
Volume  429 mm3 
Source levela  159-163 dB 
Tag lifetime  98 days (163 dB, 5-s PRI) 
127 days (161 dB, 5-s PRI) 
285 days (163 dB, 15-s PRI) 
365 days (161 dB, 15-s PRI) 
Detection range  Up to 500 m (163 dB) 
Parameter Value
Dimensions  ø5.0 × 24.2 mm 
Dry weight  720 mg 
Volume  429 mm3 
Source levela  159-163 dB 
Tag lifetime  98 days (163 dB, 5-s PRI) 
127 days (161 dB, 5-s PRI) 
285 days (163 dB, 15-s PRI) 
365 days (161 dB, 15-s PRI) 
Detection range  Up to 500 m (163 dB) 
a

All source levels referenced to 1 μPa at 1 m.

All measurements were evaluated at Pacific Northwest National Laboratory’s Bio-Acoustics and Flow Laboratory (BFL). BFL is accredited by the American Association for Laboratory Accreditation (A2LA) to the recognized International Standard ISO/IEC 17025:2005 for calibration and testing laboratories. The scope (Certificate Number 3267.01)16 includes hydrophone sensitivity measurements and power level measurements of sound sources for frequencies from 50 kHz to 500 kHz for both military equipment and commercial components.

Source level and beam pattern measurements were performed in a special water tank lined with 26-mm thick anechoic materials17 (Aptflex F48, Precision Acoustics, Ltd., Dorchester, Dorset, UK), which provides excellent ultrasound reflection reduction in the sub-MHz frequency range, thus minimizing the impact of echoes and noises inside the tank during the source level measurement. The receiving hydrophone (Model SC-001, Sonic Concepts, Bothell, WA, USA) had a sensitivity of −180.0 dB (re: 1 V/μPa) at the modulation frequency of 416.7 kHz.18 The hydrophone was connected through a four-channel JSATS signal conditioner with gain set at 10.6 dB and a data acquisition card (Model PCI-6111, National Instruments Corporation, Austin, TX, USA) to a computer for data processing. Prior to the measurements, the hydrophone was calibrated following IEC 60565:2006 standard (underwater acoustics–hydrophones–calibration in the frequency range 0.01 Hz–1 MHz) Section 919 with an omnidirectional broadband projector hydrophone20 (Model TC-4034, Reson A/S, Slangerup, Denmark) that was pre-calibrated by National Physical Laboratory (Middlesex, United Kingdom) with the uncertainty of 0.3 dB in sensitivity. For the calibration of the hydrophone, a 200-μs long pulse is generated by the pre-calibrated projector. Because the source level is known for the sound created by the projector, the sensitivity of the hydrophone can be obtained from the measured voltage of the data acquisition system and the known source level of the projected sound. The transmitter was mounted on a motion control unit located 1 m away from the receiving hydrophone in the tank. Because of the high operating frequency of the new transmitter at 416.7 kHz, the wavelength of the acoustic wave becomes 3.6 mm in water with the sound speed of 1500 m/s. This eliminates the near-field effect in the measurement of the source level, and allows for measuring the source level at 1 m directly. The motion control unit was rotated through 360° in 10° increments, moving the transmitter in pitch orientation with respect to the stationary hydrophone, while the hydrophone measured the source level continuously. A top view of the PZT transducer is shown in Figure 4 of the beam pattern, 0° is the transmitter orientation where the PZT was closest to the hydrophone. The representative source level is defined as the average of all source level measurements between 270° and 90° (the top half of the beam pattern).17 In this case, the average source level was 163.0 dB (Figure 4). In Figure 4, one can see that the source level was reduced by nearly 15 dB towards the back of the transmitter (at 180°) compared to that at the front of the transmitter (at 0°), due to the blocking of the tag body. With such a large reduction of source level, in a noisy water environment (e.g., near a dam), the energy emitted towards the rear of the transmitter would be more likely wasted. Therefore, the offset of the PZT would help to increase the detection probability in those conditions by reducing the amount of energy wasted. While in a relatively quiet water environment, a transmitter that uses a regular PZT tube (i.e., an “omnidirectional” transmitter) could potentially have better detection range towards the rear than one that uses an offset PZT tube.

FIG. 4.

Beam pattern and source level of a new transmitter at 163 dB.

FIG. 4.

Beam pattern and source level of a new transmitter at 163 dB.

Close modal

By changing the charging duration parameters, the transmitter can be configured to transmit signals at a selectable source level between 159 and 163 dB.

Aside from the battery capacity, in general, the transmitter lifetimes may be determined by (1) the PRI for transmission of the acoustic signal, (2) the source level of the acoustic signal, or (3) both the PRI and the source level of the acoustic signal.

Within the selectable range of acoustic source level between 159.0 and 163.0 dB, the new acoustic transmitter lifetime may be estimated from the empirical equation derived from our previous measured results of an injectable acoustic transmitter and the energy loss caused by the added booster converter,10 

T = V batt * C batt * t 0 * 1000 130 + 3 5 * 1 0 S L 155 10 + V batt * I s * t 0 × 1 24 .
(1)

Here T is the lifetime in days, Vbatt is the battery voltage in volts, Cbatt is the battery capacity in mAh, SL is the source level in unit of dB (re: 1 μPa at 1 m), Is is the constant static current that flows through the transmitter circuit in μA, and t0 is the PRI in seconds. This equation was obtained by assuming that three factors contribute to the energy drain from the battery: (1) the transmission of the acoustic signal, (2) the energy loss of the boost converter, and (3) the standby current constantly consumed by the microcontroller. The injectable transmitter10 used a same resonance circuit but it did not contain a boost converter. That injectable transmitter takes about 35 μJ of energy (per transmission) to produce a source level of 155 dB. Therefore, the energy consumption of the transmitter presented in this paper can be derived by scaling the required energy based on the source level range of 159–163 dB and adding the measured loss (130 μJ per transmission) of the boost converter. The number 24 is used to convert the tag life from hours to days.

Table II lists the experimental and projected tag lifetimes for acoustic tags at selected values for the PRI and source levels. For a given source level, the lifetime may be altered by varying the PRI. For example, at the source level of 163 dB at 1 m, the lifetime at 5-s PRI is approximately 98 days, whereas the lifetime at 15-s PRI is approximately 285 days. Alternatively, various values for the source level may be selected given a desired PRI to meet application requirements. For example, for the source level set to 161 dB, the projected tag lifetime is 13 days at a PRI of 0.5 s, 127 days at 5 s, and 365 days at 15 s.

TABLE II.

Lifetime of the new acoustic tag.

Lifetime (days) at PRI
Source level (dB)a 0.5 s 5 s 0.5 sb 1 sb 5 sb 10 sb 15 sb
163  8.9  N/A  10  19.9  98  193  285 
161  12.9  124.9  13  25.9  127  249  365 
Lifetime (days) at PRI
Source level (dB)a 0.5 s 5 s 0.5 sb 1 sb 5 sb 10 sb 15 sb
163  8.9  N/A  10  19.9  98  193  285 
161  12.9  124.9  13  25.9  127  249  365 
a

All source levels referenced to 1 μPa at 1 m.

b

Projected values.

The actual service life of the new transmitters is consistent with the projected value in Table II obtained using the empirical Equation (1) described above.

The new transmitter has the capability of selecting a desired detection range by setting the appropriate source level. An empirical equation, derived from actual signal-to-noise ratio (SNR) data collected at various distances using a JSATS tag, may be used to estimate the source level needed for a 500-m detection range. The equation is based on assumptions of isovelocity water and the direct path propagation of acoustic signal. It is expressed in terms of the transmission loss that accounts for both the geometric spreading and the absorption of acoustic signal in water. Based on all of these and the passive sonar equation, the transmission loss can be related to the source level. According to the passive sonar equation,21,22 the transmission loss can be expressed as

T L = A * log 10 R + α × R 1 = S L S N R thd + N L ,
(2)

where TL is the transmission loss in dB, A is a coefficient ranging between 10 and 20 dependent upon the shape of the geometrical spreading (cylindrical or spherical) of the domain in which the acoustic wave propagates,23 R is the detection range in meters, α is the temperature dependent absorption coefficient in freshwater in dB/m, SL is the source level in dB, SNRthd is the threshold SNR for the detection in dB, and NL is the background noise level in dB referenced to 1 μPa, respectively. From the equation, one can see that the detection range of an underwater acoustic transmitter is heavily dependent upon its actual environment.

To evaluate the real-world performance of this transmitter, in early 2015, a prototype sample of the transmitter was tested for SNR as a function of distance in Snake River, Clarkston, WA. The sample transmitter had a source level of 159 dB re 1 μPa, due to the availability of the transmitter at the time. The water temperature was 3 °C and the depth of water was approximately 10 m. The background noise level was tested to be 93 dB re 1 μPa. The results are shown in Figure 5. In this experiment, the detection range was found to be approximately 225 m. This result indicated a spreading coefficient A of value 17.4, suggesting a more spherical spreading of the signal. As the spreading coefficient A depends on the domain in which the acoustic wave propagates, another previous experiment conducted at Little Goose Dam, WA suggested an empirical spreading coefficient A of value 14.5. Therefore, assuming that the background noise level at a quiet environment (e.g., the forebay of a dam), the SNR threshold for the detection, the water temperature, and spreading coefficient in Equation (2) are 96 dB,24 6 dB, 20 °C, and 14.5, respectively, the projected detection ranges at the two exemplary source level settings of 163 and 161 dB for three different spreading scenarios are listed in Table III, where one can see that a source level of 161–163 dB is sufficient to theoretically achieve a 500-m detection range. It is worth noting that due to the inevitable blocking of the transmitter body, such a detection range is only achievable in about 2/3 of the 360° angular range as indicated in Figure 4.

FIG. 5.

A coefficient that describes the effect of sound spreading is determined based on a curve fitting the SNR data to the passive sonar equation. The determined coefficient is 17.4.

FIG. 5.

A coefficient that describes the effect of sound spreading is determined based on a curve fitting the SNR data to the passive sonar equation. The determined coefficient is 17.4.

Close modal
TABLE III.

Projected detection range of the new acoustic tag.a

Detection range (m) for various models
Source level (dB) Sphericalb Empiricalb Cylindricalb
163  295  543  810 
161  267  504  766 
Detection range (m) for various models
Source level (dB) Sphericalb Empiricalb Cylindricalb
163  295  543  810 
161  267  504  766 
a

All source levels referenced to 1 μPa at 1 m.

b

Assuming water temperature is 20 °C, background noise level = 96 dB, spreading coefficient is 14.5.

A small and lightweight acoustic transmitter was developed to fill the void in long-term tracking of underwater hosts such as juvenile sturgeon. The tag offers a stronger acoustic signal and longer service life than existing alternatives. The source level and PRI may be configured for various monitoring purposes and applications. The design uses a center-offset PZT ceramic tube as the acoustic transducer and a 56-mAh tubular Li-CFx battery as the power source. A dual boost converter circuit generates the supply voltages necessary for long-range transmission. The design is 5 mm in diameter and 24.2 mm in length. It has a dry weight of approximately 720 mg, a programmable source level between 159 and 163 dB, a tag life of at least 1 yr at 15 s ping rate, and a potential detection range of approximately 500 m in quiet reservoir type locations. This is, to our knowledge, the first acoustic tag that can provide such a long detection range and service life with an in-air weight under 1 g and shorter than 25 mm.

The study was funded by the Laboratory Directed Research and Development Program at Pacific Northwest National Laboratory (PNNL). The data analysis was partially funded by the U.S. Department of Energy (DOE) Wind and Water Power Technologies Office. PNNL is operated by Battelle for the DOE. The authors thank Zach Booth, Ki Won Jung, Stephanie Liss, Dave Greenslade, Jason Reynolds, Ricardo Walker, and Yong Yuan of Pacific Northwest National Laboratory, and Phil Bates of Idaho Power Company for their help with this study.

1.
A. B.
Baggeroer
, “
Acoustic telemetry — An overview
,”
IEEE J. Oceanic Eng.
9
(
4
),
229
235
(
1984
).
2.
D. B.
Kilfoyle
and
A. B.
Baggeroer
, “
The state of the art in underwater acoustic telemetry
,”
IEEE J. Oceanic Eng.
25
(
1
),
4
27
(
2000
).
3.
A. R.
Childs
,
P. D.
Cowley
,
T.
Næsje
,
A. J.
Booth
,
W. M.
Potts
,
E. B.
Thorstad
, and
F.
Okland
, “
Do environmental factors influence the movement of estuarine fish? A case study using acoustic telemetry
,”
Estuarine, Coastal Shelf Sci.
78
(
1
),
227
236
(
2008
).
4.
C. J.
Michel
,
A. J.
Ammann
,
E. D.
Chapman
,
P. T.
Sandstrom
,
H. E.
Fish
,
M. J.
Thomas
,
G. P.
Singer
,
S. T.
Lindley
,
A. P.
Klimley
, and
R. B.
MacFarlane
, “
The effects of environmental factors on the migratory movement patterns of Sacramento River yearling late-fall run Chinook salmon (Oncorhynchus tshawytscha)
,”
Environ. Biol. Fishes
96
,
257–271
(
2013
).
5.
J. M.
Ingraham
,
Z. D.
Deng
,
T.
Fu
,
G. A.
McMichael
, and
B. A.
Trumbo
, “
A fast and accurate decoder for underwater acoustic telemetry
,”
Rev. Sci. Instrum.
85
(
7
),
074903
(
2014
).
6.
X.
Li
,
Z. D.
Deng
,
Y.
Sun
,
J. J.
Martinez
,
T.
Fu
,
G. A.
McMichael
, and
T. J.
Carlson
, “
A 3D approximate maximum likelihood solver for localization of fish implanted with acoustic transmitters
,”
Sci. Rep.
4
,
7215
(
2014
).
7.
G. A.
McMichael
,
M. B.
Eppard
,
T. J.
Carlson
,
J. A.
Carter
,
B. D.
Ebberts
,
R. S.
Brown
,
M.
Weiland
,
G. R.
Ploskey
,
R. A.
Harnish
, and
Z. D.
Deng
, “
The juvenile salmon acoustic telemetry system: A new tool
,”
Fisheries
35
(
1
),
9
22
(
2010
).
8.
M. A.
Weiland
,
Z.
Deng
,
T. A.
Seim
,
B. L.
Lamarche
,
E. Y.
Choi
,
T.
Fu
,
T. J.
Carlson
,
A. I.
Thronas
, and
M. B.
Eppard
, “
A cabled acoustic telemetry system for detecting and tracking juvenile salmon: Part 1. Engineering design and instrumentation
,”
Sensors
11
(
6
),
5645
5660
(
2011
).
9.
Z.
Deng
,
M. A.
Weiland
,
T.
Fu
,
T. A.
Seim
,
B. L.
Lamarche
,
E. Y.
Choi
,
T. J.
Carlson
, and
M. B.
Eppard
, “
A cabled acoustic telemetry system for detecting and tracking juvenile salmon: Part 2. Three-dimensional tracking and passage outcomes
,”
Sensors
11
(
6
),
5661
5676
(
2011
).
10.
Z.
Deng
,
T. Z.
Carlson
,
H.
Li
,
J.
Xiao
,
M. J.
Myjak
,
J.
Lu
,
J. J.
Martinez
,
C. M.
Woodley
,
M. A.
Weiland
, and
M. B.
Eppard
, “
An injectable acoustic transmitter for juvenile salmon
,”
Sci. Rep.
5
,
8111
(
2015
).
11.
See http://www.sonotronics.com/?page_id=1068 for Sonotronics, Tucson, AZ.
12.
See http://www.htisonar.com/acoustic_tags.htm for Hydroacoustic Technology, Inc., Seattle, WA.
13.
See http://vemco.com/products/v7-to-v16-69khz/ for Vemco, Nova Scotia, Canada.
14.
H.
Chen
,
S.
Cartmell
,
Q.
Wang
,
T.
Lozano
,
Z.
Deng
,
H.
Li
,
X.
Chen
,
Y.
Yuan
,
M. E.
Gross
,
T. J.
Carlson
, and
J.
Xiao
, “
Micro-battery development for juvenile salmon acoustic telemetry system applications
,”
Sci. Rep.
4
,
3790
(
2014
).
15.
16.
See http://www.a2la.org/scopepdf/3267-01.pdf for American Association for Laboratory Accreditation, Frederick, MD.
17.
H.
Li
,
Z.
Deng
,
Y.
Yuan
, and
T. J.
Carlson
, “
Design parameters of a miniaturized piezoelectric underwater acoustic transmitter
,”
Sensors
12
,
9098–9109
(
2012
).
18.
J. M.
Ingraham
,
Z.
Deng
,
J. J.
Martinez
,
B. A.
Trumbo
,
R. P.
Mueller
, and
M. A.
Weiland
, “
Feasibility of tracking fish with acoustic transmitters in the ice harbor dam tailrace
,”
Sci. Rep.
4
,
4090
(
2014
).
19.
International Electrotechnical Commission
, “
Underwater acoustics − hydrophones − calibration in the frequency range 0,01 Hz to 1 MHz
,” in
IEC 60565
(
IEC
,
Switzerland, Geneva
,
2006
).
20.
Z.
Deng
,
M. A.
Weiland
,
T. J.
Carlson
, and
M. B.
Eppard
, “
Design and instrumentation of a measurement and calibration system for an acoustic telemetry system
,”
Sensors
10
(
4
),
3090
3099
(
2012
).
21.
W. M.
Zimmer
,
J.
Harwood
,
P. L.
Tyack
,
M. P.
Johnson
, and
P. T.
Madsen
, “
Passive acoustic detection of deep-diving beaked whales
,”
J. Acoust. Soc. Am.
124
(
5
),
2823
2832
(
2008
).
22.
E. T.
Küsel
,
D. K.
Mellinger
,
L.
Thomas
,
T. A.
Marques
,
D.
Moretti
, and
J.
Ward
, “
Cetacean population density estimation from single fixed sensors using passive acoustics
,”
J. Acoust. Soc. Am.
129
(
6
),
3610
3622
(
2011
).
23.
F. B.
Jensen
,
W. A.
Kuperman
,
M. B.
Porter
, and
H.
Schmidt
,
Computational Ocean Acoustics
, 2nd ed. (
Springer-Verlag
,
New York
,
2000
).
24.
K. W.
Jung
,
Z.
Deng
,
J. J.
Martinez
,
D. R.
Geist
,
G. A.
McMichael
,
J. R.
Stephenson
, and
P. J.
Graf
, “
Performance of an acoustic telemetry system in a large fishway
,”
Anim. Biotelemetry
3
,
17
(
2015
).