The impact of multibeam echosounder (MBES) operations on marine mammals has been less studied compared to military sonars. To contribute to the growing body of MBES knowledge, echolocation clicks of foraging Cuvier's beaked whales were detected on the Southern California Antisubmarine Warfare Range (SOAR) hydrophones during two MBES surveys and assembled into foraging events called group vocal periods (GVPs). Four GVP characteristics were analyzed Before, During, and After 12 kHz MBES surveys at the SOAR in 2017 and 2019 to assess differences in foraging behavior with respect to the mapping activity. The number of GVP per hour increased During and After MBES surveys compared with Before. There were no other differences between non-MBES and MBES periods for the three other characteristics: the number of clicks per GVP, GVP duration, and click rate. These results indicate that there was not a consistent change in foraging behavior during the MBES surveys that would suggest a clear response. The animals did not leave the range nor stop foraging during MBES activity. These results are in stark contrast to those of analogous studies assessing the effect of Naval mid-frequency active sonar on beaked whale foraging, where beaked whales stopped echolocating and left the area.

Over the last 20 years, there has been an increase in research focusing on mid-frequency (1–10 kHz) active sonar (MFAS) and its effect on toothed whales (McCarthy et al., 2011; DeRuiter et al., 2013; Jarvis et al., 2014; Manzano-Roth et al., 2016; Falcone et al., 2017; DiMarzio et al., 2019). This is largely due to concerns raised after several mass stranding events were linked to naval activities using high intensity MFAS sources (Frantzis, 1998; Evans and England, 2001; Fernandez et al., 2012; D'Amico et al., 2009). Less research has focused on the effect of higher frequency (>10 kHz) sonar (Vires, 2011; Cholewiak et al., 2017; Quick et al., 2017), such as multibeam echosounders (MBES), on toothed whales, despite similar source levels (216–245 dB re 1 μPa m) to MFAS, and an overlap in frequency range (10–400 kHz) with the most sensitive hearing range of toothed whales (10–150 kHz) (Ketten, 2004). In 2008, a stranding event of melon-headed whales in Antsihohy, Madagascar raised concern about the potential impact of MBES on marine mammals due to the temporal (< 24 h) and spatial association (65 km away) of the stranding event with a 12 kHz ocean mapping survey (Southall et al., 2013). While no direct cause of the stranding was determined, the investigators concluded that the animals most likely changed their behavior in response to the mapping survey, indirectly leading to the stranding (Southall et al., 2013). This stranding event as well as other observational studies of wild marine mammal reactions to high frequency echosounders (Cholewiak et al., 2017; Quick et al., 2017) has warranted further investigation into the potential effects that MBES signals may have on toothed whales. Furthermore, it raises the question, are MBES surveys any different than Naval sonar activity in terms of eliciting behavioral responses from toothed whales?

There are inherent differences in MFAS and MBES aside from operational frequency differences. MFAS are used to detect targets, like submarines, at distant ranges (10s of km). These systems generally have a wide vertical ensonification beam (40°) with 360° horizontal coverage, producing pings (1–2 s in length) for several minutes at intervals ranging from 6 to 15 min apart and source levels in excess of 235 dB re 1 μPa m (Hildebrand, 2009; Falcone et al., 2017). MBES are primarily used for seafloor mapping, requiring precise beam positioning and high horizontal and vertical resolution. These requirements equate to narrow (0.5° to 2°) downward directed beams in the ship's along-track direction, wide swaths across-track (commonly, 120° to 150°), and short operational pulse lengths (10–100 ms) that vary based on the ocean depth (Lurton and DeRuiter, 2011; Lurton, 2016; Kates Varghese et al., 2019). The resulting MBES geometry leads to a much smaller area of direct ensonification and orders of magnitude shorter pulses in comparison to MFAS, where exposure to harmful levels of sound are most likely to occur (Lurton, 2016). While the effect of MFAS on wild marine mammal behavior has been studied for some species with a controlled experimental design (McCarthy et al., 2011; Martin et al., 2015; Manzano-Roth et al., 2016), the effect of MBES on wild marine mammal behavior has not. The goal of this study was to assess the effect of MBES activity on beaked whale foraging behavior with a controlled experimental design.

In an effort to better characterize the radiation pattern of a Kongsberg EM 122 (12 kHz) multibeam echosounder, two MBES mapping surveys utilizing this system were run over the Southern California Antisubmarine Warfare Range (SOAR) hydrophone array of the U.S. Navy Southern California Offshore Range (SCORE), a multi-warfare training complex off of San Clemente Island, California (Mayer, 2017; Smith, 2019). A resident population of Cuvier's beaked whales (Ziphius cavirostris) are known to inhabit this region and produce sound when they forage (DiMarzio and Jarvis, 2016). This provided an opportunity to collect complimentary beaked whale data to assess the foraging behavior of beaked whales during the echosounder characterization study. The design of this study was analogous to studies assessing the effect of MFAS on the foraging behavior of beaked whales on the same (DiMarzio et al., 2019) and other Navy hydrophone ranges (McCarthy et al., 2011; Manzano-Roth et al., 2016), informing discussion and allowing for the comparison of results.

Cuvier's beaked whales generally forage in groups of 1–4 animals generating groups of echolocation clicks, collectively referred to as group vocal periods (GVPs). Animals have been tracked to depths up to 2992 m during foraging dives, foraging for an hour or more at a time (Schorr et al., 2014). During these foraging dives, Cuvier's beaked whales produce frequency-modulated echolocation clicks to detect and capture prey (Tyack et al., 2006). These highly directional clicks sweep through frequencies from 20 to 90 kHz, with most of the energy concentrated between 40 and 60 kHz, and which last between 200 and 600 μs. Inter-click intervals for this species range between 200 and 500 ms, and on-axis source levels for these clicks are between 214 and 224 dB re 1 μPa m (Johnson, et al., 2004; Zimmer, et al., 2005; Baumann-Pickering et al., 2013; Baumann-Pickering et al., 2014; Gassmann et al., 2015).

In previous work at SOAR comparing beaked whale click detections with visually sighted groups, an estimated horizontal detection range of 6.3 km over the course of a foraging dive cycle was used for Cuvier's beaked whales (DiMarzio and Jarvis, 2016). This detection distance was similar to the 6.5 km detection distance measured for Blainville's beaked whales in the Bahamas (Ward et al., 2008; Ward et al., 2011), a species with similar foraging clicks and dive behavior (Johnson et al., 2004; Tyack et al., 2006), and was assumed to be true for this study on the SOAR range as well. Since beaked whales generally begin foraging below 200 m and produce several thousand clicks in a foraging event, the spatial layout of the SOAR hydrophone range (<2000 m depth, < 6.5 km between adjacent hydrophones) was conducive for detecting a foraging event if it occurred on the range.

A previous review by DeRuiter (2010) suggests the hearing threshold of toothed whales is most sensitive between 10 and 150 kHz, indicating high sensitivity to the frequency range of deep-water MBES systems such as the one used in this study, which was the Kongsberg EM 122 with a 12 kHz center frequency. Understanding that Cuvier's beaked whales have been associated with MFAS-related stranding events on multiple occasions (Evans and England, 2001; Cox et al., 2006; D'Amico et al., 2009), Cuvier's beaked whales were an appropriate species to assess the effect of MBES activity, as well as compare with the effects of MFAS.

In order to assess the effect of MBES activity on beaked whale foraging behavior, echolocation clicks from Cuvier's beaked whales were detected and classified into GVPs, and compared across three exposure periods (Before, During, and After) with respect to MBES activity on the SOAR. This was an analogous design to studies that looked at beaked whale foraging behavior in response to MFAS exercises (McCarthy et al., 2011; Manzano-Roth et al., 2016; DiMarzio et al., 2019). In the present study, the EM 122 (12 kHz) MBES surveys were conducted on the hydrophone range of the SOAR located off of San Clemente Island, California. The hydrophone array consists of 177 bottom-mounted hydrophones at depths ranging from 840 to 1750 m spanning an 1800 km2 area (Fig. 1). The hydrophones, spaced between 2.5 and 6.5 km apart, have a receive bandwidth between 50 Hz and 48 kHz (DiMarzio and Jarvis, 2016), are sampled at 96 kHz, and are capable of receiving both the beaked whale clicks and the signals from the EM 122 MBES.

FIG. 1.

(Color online) Bathymetry of the Southern California Antisubmarine Warfare Range hydrophone array. 88 hydrophones are shown, indicated by circles. The position of the other 89 hydrophones have a similar but offset arrangement to those shown here.

FIG. 1.

(Color online) Bathymetry of the Southern California Antisubmarine Warfare Range hydrophone array. 88 hydrophones are shown, indicated by circles. The position of the other 89 hydrophones have a similar but offset arrangement to those shown here.

Close modal

Two surveys, one in January 2017 (Mayer, 2017; Smith, 2019) and the other in January 2019, were conducted as part of a MBES characterization project, each utilizing the UNOLS vessel R/V Sally Ride and its Kongsberg EM 122, a deep-water MBES. According to the manufacturer's specification, the EM 122 has a maximum source level of 239–242 dB re 1 μPa m and emits sound at center frequencies between 11 and 13.25 kHz. The EM 122 can be operated in single or dual swath mode, meaning each ping contains 8 or 16 pulses (Fig. 2), respectively, which are either continuous wave (gated, single-frequency) or frequency-modulated. The exact signal depends on user-defined input as well as depth information, which will vary based on the survey design and location (Kates Varghese et al., 2019). For the SOAR surveys, pulse lengths were between 8 and 15 ms, and each ping was spaced roughly 6–7 s apart. These were typical parameters for a mapping survey in the deep-water environment at the SOAR.

FIG. 2.

(Color online) Simplified representation and nomenclature for the structure of a possible EM 122 single swath mode ping. Each of the eight lines represents a pulse, while the series of eight pulses together constitutes a ping. This ping example contains only continuous wave pulses. An EM 122 single swath ping produces pulses at discrete frequencies between 11 and 13.25 kHz. An individual pulse length is between 2 and 100 ms, where longer pulses are used for deeper depths. A complete ping can last between 8 and 400 ms, again with longer ping durations associated with deeper depth modes.

FIG. 2.

(Color online) Simplified representation and nomenclature for the structure of a possible EM 122 single swath mode ping. Each of the eight lines represents a pulse, while the series of eight pulses together constitutes a ping. This ping example contains only continuous wave pulses. An EM 122 single swath ping produces pulses at discrete frequencies between 11 and 13.25 kHz. An individual pulse length is between 2 and 100 ms, where longer pulses are used for deeper depths. A complete ping can last between 8 and 400 ms, again with longer ping durations associated with deeper depth modes.

Close modal

The first MBES mapping survey was conducted 1/5/17 08:15-1/7/17 07:15 UTC. The majority of the survey was run in a “mowing-the-lawn” fashion (Fig. 3, left) using the EM 122 in deep, dual swath mode with CW pulses only. The vessel's survey speed was 10 kn except when turning the ship, when the speed was dropped to 5 kn. This was followed by a shorter period when multiple acoustic sources were in use. These included the EM 122, a Kongsberg EM 712 MBES (40 kHz), a Simrad EK 80 wide-band echo sounder (18, 38, 70, 120, 200 kHz), and a Knudsen sub-bottom profiler (3.5 kHz).

FIG. 3.

(Color online) Hydrophone range overlaid with the ship track lines. Left: 2017, right: 2019. Dots (green) indicate placement of the hydrophones. Lines (black) indicate track lines of the vessel.

FIG. 3.

(Color online) Hydrophone range overlaid with the ship track lines. Left: 2017, right: 2019. Dots (green) indicate placement of the hydrophones. Lines (black) indicate track lines of the vessel.

Close modal

The second MBES mapping survey was conducted 1/4/19 12:00-1/6/19 16:00 UTC during which the majority of the work was carried out in the southeastern corner of the range, restricting the EM 122 to single swath, CW only mode (Fig. 3, right). Following this, the EM 122 was set to operate in dual swath mode with FM pulses enabled and lines were run from the southeastern corner of the range to the center of the range and back twice before switching to “mowing-the-lawn” type survey lines. To the best of our knowledge, no MFAS activity was taking place on the range during either of these surveys.

The SOAR hydrophone data were processed at the Naval Undersea Warfare Center using a series of algorithms to obtain Cuvier's beaked whale GVPs (DiMarzio and Jarvis, 2016). These were similar processing procedures used to analyze the impact of MFAS on marine mammals (McCarthy et al., 2011; DiMarzio et al., 2019). Cuvier's beaked whale foraging clicks were first detected and classified from the hydrophone data with a class-specific support vector machine classifier. The clicks were then formed into click-trains on a per hydrophone and per class basis using a Java-based click train processor program. Clicks were added to each click train until at least three minutes passed without a click detection. If the click train had at least five clicks it was saved; otherwise it was discarded. The click trains were then used as input to a matlab-based autogrouper (AG) program which associated the click-trains into groups using a set of rules based on the time and location of the click trains (Moretti, 2019). The AG program was set to only use Cuvier's beaked whale click trains with inter-click intervals between 0.35 and 0.75 s (Frantzis et al., 2002).

A GVP is the time period from the first detected foraging click from the group to the group's last detected foraging click. The initial AG output was reformatted, filtered and summarized in r (R Core Team, 2018) to produce a final AG output, which contained a list of Cuvier's beaked whale GVPs detected along with information about each detected GVP. The filtering removed GVPs with group click counts less than 300 or greater than 43 200 clicks, and GVPs less than 5 min or greater than 90 min long. This was done to remove longer delphinid GVPs that may have been misidentified as Cuvier's beaked whales. “Edge-only groups,” or groups only detected on hydrophones on the edge of the range, were also removed. If an event was only detected on edge hydrophones, it was most likely from a group off the range and outside of the survey area. The detection statistics for the AG with the “edge-only groups” removed are 0.759 for the probability of detection, 0.241 for the probability of false negatives and 0.185 for the probability of false positives (DiMarzio and Jarvis, 2016).

For each GVP the output included (1) a timestamp for the start and end of each GVP, thereby providing the duration of the GVP; (2) “center hydrophone,” or hydrophone with the most detected clicks; (3) number of clicks detected on the center hydrophone; and (4) the sum of all clicks detected on all hydrophones in the group (“group click count”). This dataset provided information on the number of GVPs that occurred on the range, as well as specific details about each GVP, but did not discern whether these events were made by unique foraging groups or individuals.

The GVP is defined as “a temporally and spatially unique set of vocalizations that represent a single group of beaked whales vocalizing during a deep foraging dive” (McCarthy et al., 2011). Four GVP characteristics were used to assess foraging behavior: (1) number of GVP per hour, (2) number of clicks per GVP, (3) GVP duration, and (4) click rate. Click rate is defined as the number of clicks per minute of the GVP, whereas the “number of clicks” is the total number of clicks for a single GVP event. Each GVP characteristic was computed by summing (number of GVP per hour) or averaging (GVP characteristics 2, 3, 4) the detections from all of the range hydrophones for each of the three exposure periods, Before, During, and After, with respect to each mapping survey. Therefore, these metrics provide information about the temporal but not spatial distribution of foraging on the range throughout the study period.

The During period included the time when the EM 122 was first turned on until it was last turned off. The length of time of the During period in each year dictated how many hours of observation were selected for analysis Before and After the MBES activity to have a balanced analysis. So, for 2017 each of the three periods had 47 h of observation and for 2019 the three periods were each 52 h. To rule out potential diel foraging (Baird et al., 2008) or diel differences in detectability (DiMarzio et al., 2019) as factors for differences between the time periods, each of the three periods started and ended at the same time of day, with respect to one another. The exposure periods were separated by approximately 24 h to ensure appropriate comparison. For 2017, the Before period was 1/2/17 08:15-1/4/17 07:15 UTC and the After period was 1/8/17 08:15-1/10/17 07:15 UTC. The time separating each adjacent exposure period in 2017 was 25 h. In 2019, the Before period was 1/1/19 12:00-1/3/19 16:00 UTC, and the After period was 1/7/19 12:00-1/9/19 16:00 UTC. The time separating each adjacent exposure period of 2019 was 20 h.

Given that Cuvier's beaked whales forage on average for 40–60 min (Baird et al., 2006; Schorr et al., 2014; DiMarzio et al., 2019), an hour was chosen as the unit over which to average detections. Therefore, all detections within each hour of observation of a given exposure period were binned to compute each GVP characteristic hourly. Detections were binned based on the start time of each GVP, so that if a GVP started in a specific hour interval it was only accounted for once in that hour of observation, even if it lasted more than an hour. Using all hour observations in the analysis can increase the likelihood for temporal autocorrelation which increases the risk of a type I error—i.e., incorrectly concluding that there is a difference in foraging behavior across the Before, During, and After periods, when there is none. For this reason, a 99% confidence level for hypothesis testing was employed that is more stringent than the conventional 95% level.

The effect of MBES activity on foraging behavior of Cuvier's beaked whales was assessed by testing the following hypotheses with appropriate analysis of variance tests, either one-way analysis of variance (ANOVA) or, if the assumptions of the ANOVA were not met, a Kruskal-Wallis test.

H01—the number of GVP per hour was the same Before, During, and After MBES activity;

H02—the number of clicks per GVP was the same Before, During, and After MBES activity;

H03—the GVP duration was the same Before, During, and After MBES activity; and

H04—the click rate per GVP was the same Before, During, and After MBES activity.

In each case where the null hypothesis was rejected, a post hoc multiple comparison test was used to determine which of the exposure periods were different from one another.

Based on previous information about the foraging behavior of this resident population of Cuvier's beaked whales, it was hypothesized that the two years of data would not be different from one another. To test this, independent t-tests were run to compare each of the three exposure periods of 2017 to the respective exposure periods of 2019. Provided the two data sets were not different, they could be combined to add statistical power to the overall analysis.

GVP detection reports including 141 h of data were processed from 2017 and 156 h from 2019: 47 h for each of the three periods in 2017 and 52 h for each of the three periods in 2019. For any hour increment in which no GVPs were detected, the other GVP characteristics could not be calculated, reducing the number of observations for analysis for those characteristics.

There were 575 GVP detections made over the course of the 2017 study and 394 GVP detections in 2019. In general, there were significantly (p < 0.01) more GVPs per hour in 2017 (4.078 ± 2.36) than there were in 2019 (2.52 ± 1.95). Particularly in 2017, there were more (p < 0.01) GVPs per hour both During and After MBES activity compared with 2019 (Fig. 4, center and right), but no difference Before between the two years (Fig. 4, left). Compared across years, this metric may provide broader insight about foraging in this area, specifically, the inter-annual variability in animal presence on the range (DiMarzio et al., 2019). This could indicate that there were simply more animals overall on the range during 2017 in comparison to 2019. There was no difference between 2017 and 2019 in (1) the number of clicks per GVP, (2) GVP duration, or (3) click rate for any of the three exposure periods (Table I). Since there were no differences in the other GVP characteristics, the two years of data were assumed to come from similar distributions and were combined into one data set for further analysis. Figure 5 shows the hourly binned data across the three exposure periods for each year for comparison. However, an additional analysis was performed separately for each survey year which assessed the GVP characteristics across a finer-temporal scale with respect to the mapping activity that took place each year. These analyses are contained in supplementary material.1

FIG. 4.

(Color online) Boxplots of the number of GVP per hour for each exposure period by year. Left: Before, Center: During, right: After. Center line indicates the median, top of the box indicates 75th percentile, bottom of box indicates 25th percentile, whiskers cover 99.3% of the data and plus signs indicate outliers. Presence of an asterisk on a plot indicates that the two years were significantly different at a 99% significance level for that period.

FIG. 4.

(Color online) Boxplots of the number of GVP per hour for each exposure period by year. Left: Before, Center: During, right: After. Center line indicates the median, top of the box indicates 75th percentile, bottom of box indicates 25th percentile, whiskers cover 99.3% of the data and plus signs indicate outliers. Presence of an asterisk on a plot indicates that the two years were significantly different at a 99% significance level for that period.

Close modal
TABLE I.

Mean and standard deviation of each GVP characteristic for each year in each exposure period.

Exposure PeriodBeforeDuringAfter
Year201720192017201920172019
Number of GVP per hour 2.7 ± 1.56 2.21 ± 2.00 4.43 ± 2.26 2.87 ± 1.9 5.11 ± 2.51 2.5 ± 1.91 
Number of clicks per GVP 2646 ± 1402 1903 ± 1046 2465 ± 1652 2327 ± 1197 3038 ± 1479 2853 ± 1889 
GVP duration (min) 48.02 ± 14.25 35.85 ± 11.82 40.43 ± 12.35 37.81 ± 12.54 43.77 ± 10.35 40.48 ± 16.28 
Click rate (clicks/min) 58.39 ± 30.88 57.47 ± 42.57 58.90 ± 31.76 59.36 ± 25.92 68.04 ± 24.28 70.47 ± 42.30 
Exposure PeriodBeforeDuringAfter
Year201720192017201920172019
Number of GVP per hour 2.7 ± 1.56 2.21 ± 2.00 4.43 ± 2.26 2.87 ± 1.9 5.11 ± 2.51 2.5 ± 1.91 
Number of clicks per GVP 2646 ± 1402 1903 ± 1046 2465 ± 1652 2327 ± 1197 3038 ± 1479 2853 ± 1889 
GVP duration (min) 48.02 ± 14.25 35.85 ± 11.82 40.43 ± 12.35 37.81 ± 12.54 43.77 ± 10.35 40.48 ± 16.28 
Click rate (clicks/min) 58.39 ± 30.88 57.47 ± 42.57 58.90 ± 31.76 59.36 ± 25.92 68.04 ± 24.28 70.47 ± 42.30 
FIG. 5.

(Color online) Barplots showing the hourly data of the four GVP characteristics across the three exposure periods of 2017 (left column) and 2019 (right column). First row: number of GVP per hour; second row: number of clicks; third row: GVP duration in minutes; last row: click rate. Color transition across each plot on the x axis shows transition between Before, During, and After periods from left to right, respectively.

FIG. 5.

(Color online) Barplots showing the hourly data of the four GVP characteristics across the three exposure periods of 2017 (left column) and 2019 (right column). First row: number of GVP per hour; second row: number of clicks; third row: GVP duration in minutes; last row: click rate. Color transition across each plot on the x axis shows transition between Before, During, and After periods from left to right, respectively.

Close modal

The combined dataset contained 99 h of observation for each of the three exposure periods. The average number of GVP per hour Before was 2.44 (SD = 1.81), which increased to 3.61 (SD = 2.21) During, and increased further to 3.74 (SD = 2.56) After (Table II). The data for this GVP characteristic met the assumptions of an ANOVA, so a one-way ANOVA was used to compare the exposure periods. The number of GVP per hour was not the same for the three exposure periods [F (2, 294) = 10.18, p = 0.00005]. There were more (p = 0.00067) GVP per hour During and more (p = 0.0001) GVP per hour After MBES activity than there were Before (Table III).

TABLE II.

Descriptive statistics for the four GVP characteristics of the combined analysis, including the mean and standard deviation for each exposure period and number of samples used to compute those values in parentheses.

BeforeDuringAfter
Number of GVP per hour 2.44 ± 1.81 (n = 99) 1.61 ± 2.21 (n = 99) 3.74 ± 2.56 (n = 99) 
Number of clicks per GVP 2312 ± 1301 (n = 81) 2395 ± 1432 (n = 94) 2946 ± 1690 (n = 92) 
GVP duration (min) 42.6 ± 14.49 (n = 81) 39.1 ± 12.45 (n = 94) 42.12 ± 13.67 (n = 92) 
Click rate (clicks/min) 57.98 ± 36.29 (n = 81) 59.14 ± 28.77 (n = 94) 69.26 ± 34.32 (n = 92) 
BeforeDuringAfter
Number of GVP per hour 2.44 ± 1.81 (n = 99) 1.61 ± 2.21 (n = 99) 3.74 ± 2.56 (n = 99) 
Number of clicks per GVP 2312 ± 1301 (n = 81) 2395 ± 1432 (n = 94) 2946 ± 1690 (n = 92) 
GVP duration (min) 42.6 ± 14.49 (n = 81) 39.1 ± 12.45 (n = 94) 42.12 ± 13.67 (n = 92) 
Click rate (clicks/min) 57.98 ± 36.29 (n = 81) 59.14 ± 28.77 (n = 94) 69.26 ± 34.32 (n = 92) 
TABLE III.

ANOVA tables for the four GVP characteristics of the combined analysis, including post-hoc comparison p-values. SS = sum of squares, DF = degrees of freedom, MS = mean square, F = F-statistic, Chi-sq = Chi-square statistic, Prob > test statistic = the significance of the test.

Number of GVP per hour
SourceSSDFMSFProb > FPost-hoc p-values
Groups 100.26 50.1313 10.18 0.00005 Before vs During 0.00067 
Error 1447.25 294 4.9226   Before vs After 0.0001 
Total 1547.52 296    During vs After 0.91 
Number of clicks per GVP 
Source SS DF MS Chi-sq Prob > Chi-sq  Post-hoc p-values 
Groups 46523.5 23261.8 7.86 0.0196 Before vs During 0.96 
Error 1521879 264 5786.6   Before vs After 0.034 
Total 1568402.5 266    During vs After 0.051 
GVP duration 
Source SS DF MS Prob > F  Post-hoc p-values 
Groups 659.2 329.58 1.8 0.1665 Before vs During 0.2 
Error 48212.8 264 182.62   Before vs After 0.97 
Total 48872 266    During vs After 0.28 
Click Rate 
Source SS DF MS Chi-sq Prob > Chi-sq  Post-hoc p-values 
Groups 76023 38022.5 12.75 0.0017 Before vs During 0.67 
Error 1510135 264 5720.2   Before vs After 0.002 
Total 1586158 266    During vs After 0.022 
Number of GVP per hour
SourceSSDFMSFProb > FPost-hoc p-values
Groups 100.26 50.1313 10.18 0.00005 Before vs During 0.00067 
Error 1447.25 294 4.9226   Before vs After 0.0001 
Total 1547.52 296    During vs After 0.91 
Number of clicks per GVP 
Source SS DF MS Chi-sq Prob > Chi-sq  Post-hoc p-values 
Groups 46523.5 23261.8 7.86 0.0196 Before vs During 0.96 
Error 1521879 264 5786.6   Before vs After 0.034 
Total 1568402.5 266    During vs After 0.051 
GVP duration 
Source SS DF MS Prob > F  Post-hoc p-values 
Groups 659.2 329.58 1.8 0.1665 Before vs During 0.2 
Error 48212.8 264 182.62   Before vs After 0.97 
Total 48872 266    During vs After 0.28 
Click Rate 
Source SS DF MS Chi-sq Prob > Chi-sq  Post-hoc p-values 
Groups 76023 38022.5 12.75 0.0017 Before vs During 0.67 
Error 1510135 264 5720.2   Before vs After 0.002 
Total 1586158 266    During vs After 0.022 

There were several hours within the study when there were no GVPs detected. Since the remaining GVP characteristics could only be calculated if there were GVP detections within a given hour bin, there were fewer observation hours for the other three GVP characteristics compared to the number of GVP per hour. In particular, there were 81 h with at least 1 GVP per hour Before, 94 h During, and 92 h After used to analyze the final three GVP characteristics (Table II).

The average number of clicks per GVP Before was 2312 (SD = 1301), 2395 (SD = 1432) During, and 2946 (SD = 1690) After (Table II). A Kruskal-Wallis test was used to compare the number of clicks per GVP since the data for this metric did not satisfy the normality assumption of an ANOVA. The test showed no difference between the three exposure periods [H (2) = 7.86, p = 0.02] at the 99% significance level (Table III).

The average GVP duration Before was 42.6 min (SD = 14.49), 39.1 min (SD = 12.45) During, and 42.12 min (SD = 13.67) After (Table II). The data for this GVP characteristic met the assumptions of an ANOVA, so a one-way ANOVA was used to compare the exposure periods. The test showed that there was no difference in GVP duration between the three exposure periods [F (2, 264) = 1.8, p = 0.1665] (Table III).

The average click rate Before was 57.98 clicks per minute (SD = 36.29), 59.14 clicks per minute (SD = 28.77) During, and 69.26 clicks per minute (SD = 34.32) After (Table II). The Kruskal-Wallis test was used to compare the click rate data between exposure periods since the data did not satisfy the normality assumption of an ANOVA. There was a difference between the three exposure periods [H (2) = 12.75, p = 0.0017]. The click rate was higher (p = 0.0021) After MBES activity than Before (Table III).

Of the four metrics used to assess beaked whale foraging behavior, only the number of GVPs per hour was statistically different During MBES activity versus a non-MBES period. In this case, there were more GVPs per hour During MBES activity than there were Before. This clearly shows that the animals did not stop foraging during the MBES surveys and did not leave the range since GVPs were still being detected on the range. More foraging events During the MBES survey is a stark contrast to the results of analogous studies assessing the effect of Naval sonar on foraging beaked whales (McCarthy et al., 2011; Manzano-Roth et al., 2016; DiMarzio et al., 2019). In the McCarthy (2011) study, Blainville's beaked whales not only stopped echolocating, but left the hydrophone range as well during a multi-ship Naval sonar exercise on the range (Tyack et al., 2011). A decrease in the number of GVPs during Naval MFAS exercises was also seen for groups of beaked whales from populations inhabiting waters on the Pacific Missile Range Facility in Hawaii (Manzano-Roth et al., 2016) and at the SOAR in southern California (DiMarzio et al., 2019).

The number of GVPs per hour and click rate both statistically differed between the Before and After periods, with more GVPs and an increased click rate occurring After than Before. The means of both of these GVP characteristics increased across the three exposure periods, while a similar trend was seen in the average number of clicks per GVP, although not statistically significant (Table II). Of the four GVP characteristics, three increased over the time period evaluated, which implies an overall increase in foraging effort during this study. The increase in foraging effort over time, even after MBES activity, suggests that this trend is most likely a function of some other change in the environment, either on or off the range, rather than a response to the MBES survey itself. Alternatively, this result may be viewed as a lag in response by the foraging animals to the MBES survey. For example, in the McCarthy et al. (2011) study, though the number of GVPs decreased during the MFAS period, it was not until roughly 30 h into the exercises that there was a pronounced decrease in the number of GVPs compared to pre-exposure numbers. These surveys were operated as typical deep-water mapping efforts using surface-borne MBES, so the duration of MBES activity in this study represents a typical length survey for this type of environment (and perhaps even longer as other sources were used when the survey was completed). Conducting an even longer survey in a controlled study may provide insight into whether the increase in the number of GVPs per hour and click rate After compared with Before is a lagged response to MBES. However, performing a longer survey would also not be an accurate representation of exposure to this noise source under normal operating conditions.

Regardless of whether the results are from a lagged response to the MBES survey, or an effect of some other factor not studied here, it is valuable to explore what an increase in foraging effort may mean. One explanation is that the MBES survey may have affected the behavior of the prey, squid, of the beaked whales. A recent study of squid abundance and distribution at the SOAR revealed large differences over small spatial scales that can have huge repercussions on a predator's decision to forage in a given area (Southall et al., 2018). Beaked whale foraging behavior is heavily dictated by prey abundance and distribution, so if MBES activity changed the dynamics of the prey, making them easier to hunt and/or capture, this could have led to the increase seen in foraging effort. Alternatively, it is possible that prey distribution at the SOAR, or the surrounding area, changed over the course of this study due to some external factor, such as the oceanographic conditions, that led to more favorable foraging conditions on the range. Local environmental conditions can affect prey densities and position in the water column. Due to the opportunistic nature of this work with the echosounder characterization study, only environmental parameters directly related to acoustic propagation were collected to support the source characterization work, which was not directly applicable to questions about prey dynamics. A third consideration is that the increase in foraging effort could represent unsuccessful or aborted foraging attempts, compensated for by an increase in the number and intensity of foraging attempts. One explanation for aborted foraging attempts could be that the signal from the MBES masks or jams the animal's ability to discern its own echolocation signal. This seems unlikely for two reasons: (1) the MBES signal frequency falls outside the octave band in which the echolocation signal of this species lies and (2) by this argument, an individual GVP would be shorter in duration and contain fewer clicks, which was not observed. In the absence of tagging or tracking individual animals and prey dynamics, these hypotheses or any additional interpretations of the cause and/or effect of increased foraging effort cannot be verified.

There can be significant variability in marine mammal behavior at the species, group, and individual level that can confound comparing results across studies. Environmental and physical conditions of the study site, such as the geometry of the seafloor, can affect how sound propagates and consequently how an animal will hear, perceive, and respond to a sound. Yet DiMarzio et al. (2019) showed a similar decreasing trend in the number of GVPs of Cuvier's beaked whales at the SOAR during MFAS activity as two other studies looking at a different species (Blainville's) of beaked whale (McCarthy et al., 2011; Manzano-Roth et al., 2016) at two other geographical locations (the Bahamas and Hawaii). Despite the potential for variability in marine mammal behavior, the equivalent results of the three MFAS studies (Manzano-Roth et al., 2016; McCarthy et al., 2011; DiMarzio et al., 2019) suggest a clear cessation of foraging response to MFAS activity by foraging beaked whales. Given the high site fidelity of Cuvier's beaked whales at the SOAR (Falcone et al., 2009), it is likely that at least some of the animals in the DiMarzio et al. study (2019) were the same as those in the present study. Recognizing this, the decrease in all indicators (number of GVP and GVP duration) of foraging effort during MFAS activity in comparison to the lack of a statistical difference in three of the four foraging metrics and the increase in the number of GVP per hour During and After MBES activity is a notable difference in response by foraging Cuvier's beaked whales at the SOAR.

MFAS Naval sonar had a clear measurable effect on beaked whale foraging (McCarthy et al., 2011; Manzano-Roth et al., 2016; Falcone et al., 2017; DiMarzio et al., 2019) in comparison to 12 kHz MBES mapping sonar despite comparable source levels and fairly similar frequencies (3–10 kHz for MFAS) from a toothed whale hearing sensitivity perspective. This is quite reasonable given the distinct differences between these two sonar types: the directivity of transmission, the limited area of ensonification, and the short pulse lengths of MBES as compared to MFAS. These differences have a profound impact on the likelihood of interaction of these sources with marine life, and are important to consider in an animal's ability to detect, process, and potentially be affected by a signal (Southall et al., 2007; Southall et al., 2019). Additionally, the MFAS Naval sonar exercises (McCarthy et al., 2011) involved multiple ships, sources, and occupied a specific location on the range for several days at a time. A typical ocean mapping survey includes one vessel with mapping sonar that continuously moves in a “lawn-mowing” fashion over the survey area. Assuming the described operational paradigms are characteristic for these sonar types, along with the differences in pulse duration and ensonification volume, the two active sonar operations are inherently different which would factor into the different responses seen by foraging beaked whales.

This study took a coarse approach (1800 km2 area) to assessing the effect of MBES mapping sonar on foraging behavior of Cuvier's beaked whales in order to compare to the approach taken in studies assessing MFAS (McCarthy et al., 2011; Manzano-Roth et al., 2016; DiMarzio et al., 2019). Therefore, this study did not elucidate changes within a particular group of foraging animals or responses by individuals. Individual responses are an important consideration when assessing the full impact of an anthropogenic noise source on marine life. A few studies have reported changes in individual (Quick et al., 2017) and group (Cholewiak et al., 2017) behavior of marine mammals in response to high frequency scientific echosounders, though there are none to date on individual response to MBES. The value of this study is a coarse look at how MBES activity affects groups of beaked whales in an area (i.e., the SOAR), a species known to be highly sensitive to other sonar types. The results of this study show an increase in the number of GVP During and After MBES, an increase in the click rate After MBES, and no change in the number of clicks or the duration of the average GVP for Cuvier's beaked whales Before, During, or After MBES activity at the SOAR. There was not a uniform change in the four GVP characteristics During MBES activity, and two of the significant differences were found in relation to GVP characteristics during non-MBES periods. Both of these findings highlight the large amount of variability found among foraging events on the range and through time, while the latter result may suggest a lagged response to MBES activity. If the increase in foraging effort is, in fact, a response, albeit lagged, to MBES activity, this is the opposite response that beaked whales had to MFAS sonar. Since it is recognized that MFAS has a negative impact on beaked whale foraging, this result would suggest that there is not a negative impact of MBES activity on beaked whales foraging at the SOAR. If the significant differences in foraging behavior found in this study are due to the high variability seen in foraging activity, this would also suggest that there is no clear negative impact of MBES activity on beaked whale foraging at the SOAR. A finer temporal scale analysis of each year was conducted to assess some of these hypotheses and is provided in the supplementary material.1 This should be the first of many studies that take varying approaches (e.g., group/population versus individual, varying context and behaviors, environments, and mapping systems) to assessing the potential effects of MBES activity on marine mammals with a controlled experimental design.

This study would not have been possible without the collaboration of many entities including Scripps Institute of Oceanography and the Office of Naval Research for ship time on the R/V Sally Ride and the U.S. Navy for the time on the SOAR range and use of the hydrophone data. Thank you to Xavier Lurton for contributing his knowledge of MBES sonar and to Stephanie Watwood for her contributions to the work conducted at the SOAR. This material is based upon work supported by the National Science Foundation under Grant No. 1524585 and NOAA Grant No. NA15NOS4000200 provided to the Center for Coastal and Ocean Mapping at the University of New Hampshire. We would especially like to thank the two anonymous reviewers of this paper for their thorough comments and feedback.

1

See supplementary material at https://doi.org/10.1121/10.0001385 for a finer temporal and individual year analysis of beaked whale foraging behavior with respect to each of the two years of MBES survey work.

1.
Baird
,
R. W.
,
Webster
,
D. L.
,
McSweeney
,
D. J.
,
Ligon
,
A. D.
,
Schorr
,
G. S.
, and
Barlow
,
J.
(
2006
). “
Diving behavior of Cuvier's (Ziphius cavirostris) and Blainville's (Mesoplodon densirostris) beaked whales in Hawai'i
,”
Can. J. Zool.
84
(
8
),
1120
1128
.
2.
Baird
,
R. W.
,
Webster
,
D. L.
,
Schorr
,
G. S.
,
McSweeney
,
D. J.
, and
Barlow
,
J.
(
2008
). “
Diel variation in beaked whale diving behavior
,”
Mar. Mammal Sci.
24
(
3
),
630
642
.
3.
Baumann-Pickering
,
S.
,
McDonald
,
M.
,
Simonis
,
A.
,
Berga
,
A.
,
Merkens
,
K.
,
Oleson
,
E.
,
Roch
,
M.
,
Wiggins
,
S.
,
Rankin
,
S.
,
Yack
,
T.
, and
Hildebrand
,
J.
(
2013
). “
Species-specific beaked whale echolocation signals
,”
J. Acoust. Soc. Am.
134
(
3
),
2293
2301
.
4.
Baumann-Pickering
,
S.
,
Roch
,
M. A.
,
Brownell
,
R. L.
,
Simonis
,
A. E.
,
McDonald
,
M. A.
,
Solsona-Berga
,
A.
,
Oleson
,
E. M.
,
Wiggins
,
S. M.
, and
Hildebrand
,
J. A.
(
2014
). “
Spatio-temporal patterns of beaked whale echolocation signals in the North Pacific
,”
PLoS One
9
(
1
),
e86072
.
5.
Cholewiak
,
D.
,
DeAngelis
,
A.
,
Palka
,
D.
,
Corkeron
,
P.
, and
Van Parijs
,
S.
(
2017
). “
Beaked whales demonstrate a marked acoustic response to the use of shipboard echo sounders
,”
R. Soc. Open Sci.
4
,
170940
.
6.
Cox
,
T. M.
,
Ragen
,
T. J.
,
Read
,
A. J.
,
Vos
,
E.
,
Baird
,
R. W.
,
Balcomb
,
K.
,
Barlow
,
J.
,
Caldwell
,
J.
,
Cranford
,
T.
,
Crum
,
L.
,
D'Amico
,
A.
,
D'Spain
,
G.
,
Fernandez
,
A.
,
Finneran
,
J.
,
Gentry
,
R.
,
Gerth
,
W.
,
Gulland
,
F.
,
Hildebrand
,
J.
,
Houser
,
D.
,
Hullar
,
T.
,
Jepson
,
P. D.
,
Ketten
,
D.
,
MacLeod
,
C. D.
,
Miller
,
P.
,
Moore
,
S.
,
Mountain
,
D. C.
,
Palka
,
D.
,
Ponganis
,
P.
,
Rommel
,
S.
,
Rowles
,
T.
,
Taylor
,
B.
,
Tyack
,
P.
,
Wartzok
,
D.
,
Gisner
,
R.
,
Mead
,
J.
, and
Benner
,
L.
(
2006
). “
Understanding the impacts of anthropogenic sound on beaked whales
,”
J. Cet. Res. Man.
7
(
3
):
177
187
.
7.
D'Amico
,
A.
,
Gisiner
,
R. C.
,
Ketten
,
D. R.
,
Hammock
,
J. A.
,
Johnson
,
C.
,
Tyack
,
P. L.
, and
Mead
,
J.
(
2009
). “
Beaked whale strandings and naval exercises
,”
Aquat. Mamm.
35
(
4
),
452
472
.
8.
DeRuiter
,
S. L.
(
2010
). “
Marine animal acoustics
,” in
An Introduction to Underwater Acoustics: Principles and Applications
(
Praxis Publishing Limited
,
Chichester, UK
), pp.
425
474
.
9.
DeRuiter
,
S. L.
,
Southall
,
B. L.
,
Calambokidis
,
J.
,
Zimmer
,
W. M. X.
,
Sadykova
,
D.
,
Falcone
,
E. A.
,
Friedlaender
,
A. S.
,
Joseph
,
J. E.
,
Moretti
,
D.
,
Schorr
,
G. S.
,
Thomas
,
L.
, and
Tyack
,
P. L.
(
2013
). “
First direct measurements of behavioural responses by Cuvier's beaked whales to mid-frequency active sonar
,”
Biol. Lett.
9
,
20130223
.
10.
DiMarzio
,
N.
, and
Jarvis
,
S.
(
2016
). “
Temporal and spatial distribution and habitat use of Cuvier's beaked whales on the U.S. Navy's Southern California Antisubmarine Warfare Range (SOAR): Data preparation
,” Naval Undersea Warfare Center, Newport, RI.
11.
DiMarzio
,
N.
,
Watwood
,
S.
,
Fetherston
,
T.
, and
Moretti
,
D.
(
2019
). “
Marine mammal monitoring on navy ranges (M3R) on the Southern California Anti-Submarine Warfare Range (SOAR) and the Pacific Missile Range Facility (PMRF) 2018
,” Naval Undersea Warfare Center Newport, Newport, RI.
12.
Evans
,
D.
, and
England
,
G.
(
2001
). “
Joint interim report Bahamas marine mammal stranding event of 15–16 March 2000
,” U.S. Department of Commerce and Secretary of the Navy, Washington, DC.
13.
Falcone
,
E.
,
Schorr
,
G.
,
Douglas
,
A.
,
Calambokidis
,
J.
,
Henderson
,
E.
,
McKenna
,
M.
,
Hildebrand
,
J.
, and
Moretti
,
D.
(
2009
). “
Sighting characteristics and photo-identification of Cuvier's beaked whales (Ziphius cavirostris) near San Clemente Island California: A key area for beaked whales and the military?
,”
Mar. Biol.
156
,
2631
2640
.
14.
Falcone
,
E. A.
,
Schorr
,
G. S.
,
Watwood
,
S. L.
,
DeRuiter
,
S. L.
,
Zerbini
,
A. N.
,
Andrews
,
R. D.
,
Morrissey
,
R. P.
, and
Moretti
,
D. J.
(
2017
). “
Diving behaviour of Cuvier's beaked whales exposed to two types of military sonar
,”
R. Soc. Open Sci.
4
,
170629
.
15.
Fernandez
,
A.
,
Sierra
,
E.
,
Martin
,
V.
,
Mendez
,
M.
,
Sacchinni
,
S.
,
Bernaldo de Quiros
,
Andrada
,
M.
,
Rivero
,
M.
,
Quesada
,
O.
,
Tejedor
,
M.
, and
Arbelo
,
M.
(
2012
). “
Last ‘atypical’ beaked whales mass stranding in the Canary Islands (July, 2004)
,”
J. Mar. Sci. Res. Dev.
2
,
107
.
16.
Frantzis
,
A.
(
1998
). “
Does acoustic testing strand whales?
,”
Nature
392
(
5
),
29
.
17.
Frantzis
,
A.
,
Goold
,
J. C.
,
Skarsoulis
,
E. K.
,
Taroudakis
,
M. I.
, and
Kandia
,
V.
(
2002
). “
Clicks from Cuvier's beaked whales, Ziphius cavirostris (L)
,”
J. Acoust. Soc. Am.
112
(
1
),
34
37
.
18.
Gassmann
,
M.
,
Wiggins
,
S. M.
, and
Hildebrand
,
J. A.
(
2015
). “
Three-dimensional tracking of Cuvier's beaked whales' echolocation sounds using nested hydrophone arrays
,”
J. Acoustic. Soc. Am.
138
,
2483
2494
.
19.
Hildebrand
,
J. A.
(
2009
). “
Anthropogenic and natural sources of ambient noise in the ocean
,”
Mar. Ecol. Prog. Ser.
395
,
5
20
.
20.
Jarvis
,
S. M.
,
Morrisey
,
R. P.
,
Moretti
,
D. J.
,
DiMarzio
,
N. A.
, and
Shaffer
,
J. A.
(
2014
). “
Marine Mammal Monitoring on Navy Ranges (M3R): A toolset for automated detection, localization, and monitoring of marine mammals in open ocean environments
,”
Mar. Tech. Soc. J.
48
(
1
),
5
20
.
21.
Johnson
,
M.
,
Madsen
,
P. T.
,
Zimmer
,
W. M. X.
,
Aguilar de Soto
,
N.
, and
Tyack
,
P. L.
(
2004
). “
Beaked whales echolocate on prey
,”
Proc. R. Soc. B
271
,
S383
S386
.
22.
Kates Varghese
,
H.
,
Smith
,
M.
,
Miksis-Olds
,
J.
, and
Mayer
,
L.
(
2019
). “
Regulation consideration of ocean mapping multibeam echo sounders: A square peg in a round hole
,”
J. Ocean Tech.
14
(
3
),
40
46
.
23.
Ketten
,
D. R.
(
2004
). “
Marine mammal auditory systems: A summary of audiometric and anatomical data and implications for underwater acoustic impacts
,”
Polarforschung
72
(
2/3
),
79
92
.
24.
Lurton
,
X.
(
2016
). “
Modelling of the sound field radiated by multibeam echo sounders for acoustical impact assessment
,”
Appl. Acoust.
101
,
201
221
.
25.
Lurton
,
X.
, and
DeRuiter
,
S.
(
2011
). “
Sound radiation of seafloor-mapping echosounders in the water column, in relation to the risks posed to marine mammals
,”
Int. Hydrog. Rev.
2011
(
6
),
7
17
.
26.
Manzano-Roth
,
R.
,
Henderson
,
E.
,
Martin
,
S.
,
Martin
,
C.
, and
Matsuyama
,
B.
(
2016
). “
Impacts of U.S. Navy training events on Blainville's beaked whale (Mesoplodon densirostris) foraging dives in Hawaiian waters
,”
Aquat. Mamm.
42
(
4
),
507
528
.
27.
Martin
,
S.
,
Martin
,
C. R.
,
Matsuyama
,
B. M.
, and
Henderson
,
E. E.
(
2015
). “
Minke whales (Balaenoptera acutorostrata) respond to navy training
,”
J. Acoust. Soc. Am.
137
(
5
),
2533
2541
.
28.
Mayer
,
L.
(
2017
). “
University of New Hampshire/National Oceanic and Atmospheric Administration Joint Hydrographic Center performance and progress report 2017
,” NOAA Grant No. NA15NOS4000200, https://ccom.unh.edu/sites/default/files/progress_reports/2017-jhc-ccom-progress-report-web.pdf (Last viewed 12/1/2019).
29.
McCarthy
,
E.
,
Moretti
,
D.
,
Thomas
,
L.
,
DiMarzio
,
N.
,
Morrissey
,
R.
,
Jarvis
,
S.
,
Ward
,
J.
,
Izzi
,
A.
, and
Dilley
,
A.
(
2011
). “
Changes in spatial and temporal distribution and vocal behavior of Blainville's beaked whales (Mesoplodon densirostris) during multiship exercises with mid-frequency sonar
,”
Mar. Mamm. Sci.
27
(
3
),
E206
E226
.
30.
Moretti
,
D.
(
2019
). “
Estimating the effect of mid-frequency active sonar on the population health of Blainville's beaked whales (Mesoplodon densirostris) in the Tongue of the Ocean
,” Ph.D. thesis,
University of St. Andrews
, St. Andrews, Scotland, pp.
1
246
.
31.
Quick
,
N.
,
Scott-Hayward
,
L.
,
Sadykova
,
D.
,
Nowacek
,
D.
, and
Read
,
A.
(
2017
). “
Effects of a scientific echo sounder on the behavior of short finned pilot whales (Globicephala macrorhynchus
),”
Can. J. Fish. Aquat. Sci.
74
(
5
),
716
726
.
32.
R Core Team
(
2018
). “
R: A language and environment for statistical computing
,” version 3.5.1, R Foundation for Statistical Computing, Vienna, Austria.
33.
Schorr
,
G. S.
,
Falcone
,
E. A.
,
Moretti
,
D. J.
, and
Andrews
,
R. D.
(
2014
). “
First long-term behavioral records from Cuvier's beaked whales (Ziphius cavirostris) reveal record-breaking dives
,”
PLoS One
9
,
e92633
.
34.
Smith
,
M. J.
(
2019
). “
Analysis of the radiated sound field of a deep-water multibeam echo sounder using a navy hydrophone array
,” M.S. thesis, University of New Hampshire, https://unh.idm.oclc.org/login?url=https://search.proquest.com/docview/2273838102?accountid=14612 (Last viewed 12/1/2019).
35.
Southall
,
B.
,
Bowles
,
A. E.
,
Ellison
,
W. T.
,
Finneran
,
J. J.
,
Gentry
,
R. L.
,
Greene
,
C. R.
, Jr.
,
Kastak
,
D.
,
Ketten
,
D. R.
,
Miller
,
J. H.
,
Nachtigall
,
P. E.
,
Richardson
,
W. J.
,
Thomas
,
J. A.
, and
Tyack
,
P. L.
(
2007
). “
Marine mammal noise exposure criteria: Initial scientific recommendations
,
Aquat. Mamm.
33
(
4
),
411
522
.
36.
Southall
,
B. L.
,
Benoit-Bird
,
K. J.
,
Moline
,
M. A.
, and
Moretti
,
D.
(
2018
). “
Quantifying deep-sea predator-prey dynamics: Implications of biological heterogeneity for beaked whale conservation
,”
J. Appl. Eco.
56
,
1040
1049
.
37.
Southall
,
B. L.
,
Finneran
,
J. J.
,
Reichmuth
,
C.
,
Nachtigall
,
P. E.
,
Ketten
,
D. R.
,
Bowles
,
A. E.
,
Ellison
,
W. T.
,
Nowacek
,
D. P.
, and
Tyack
,
P. L.
(
2019
). “
Marine mammal noise exposure criteria: Updated scientific recommendations for residual hearing effects
,”
Aquat. Mamm.
45
(
2
),
125
232
.
38.
Southall
,
B. L.
,
Rowles
,
T.
,
Gulland
,
F.
,
Baird
,
R. W.
, and
Jepson
,
P. D.
(
2013
). “
Final report of the independent scientific review panel investigating potential contributing factors to a 2008 mass stranding of melon-headed whales (Peponocephala electra) in Antsohihy, Madagascar
,” https://www.cascadiaresearch.org/Hawaii/Madagascar_ISRP_Final_report.pdf (Last viewed 12/1/19).
39.
Tyack
,
P.
,
Johnson
,
M.
,
Aguilar Soto
,
N.
,
Sturlese
,
A.
, and
Madsen
,
P. T.
(
2006
). “
Extreme diving of beaked whales
,”
J. Exp. Biol.
209
,
4238
4253
.
40.
Tyack
,
P. L.
,
Zimmer
,
W. M. X.
,
Moretti
,
D.
,
Southall
,
B. L.
,
Claridge
,
D. E.
,
Durban
,
J. W.
,
Clark
,
C. W.
,
D'Amico
,
A.
,
DiMarzio
,
N.
,
Jarvis
,
S.
,
McCarthy
,
E.
,
Morrissey
,
R.
,
Ward
,
J.
, and
Boyd
,
I. L.
(
2011
). “
Beaked whales respond to simulated and actual Navy sonar
,”
PLoS One
6
(
3
),
e17009
.
41.
Vires
,
G.
(
2011
). “
Echosounder effects on beaked whales in the Tongue of the Ocean, Bahamas
,” Master's thesis,
Duke University
, Durham, NC.
42.
Ward
,
J.
,
Moretti
,
D.
,
Morrissey
,
R. P.
,
DiMarzio
,
N. A.
,
Tyack
,
P.
, and
Johnson
,
M.
(
2008
). “
Mesoplodon densirostris transmission beam pattern estimated from passive acoustic bottom mounted hydrophones and a DTag record
,”
J. Acoust. Soc. Am.
123
,
3619
.
43.
Ward
,
J.
,
Jarvis
,
S.
,
Moretti
,
D.
,
Morrissey
,
R.
,
DiMarzio
,
N.
,
Thomas
,
L.
, and
Marques
,
T. A.
(
2011
). “
Beaked whale (Mesoplodon densirostris) passive acoustic detection with increasing ambient noise
,”
J. Acoust. Soc. Am.
129
,
662
669
.
44.
Zimmer
,
W. M. X.
,
Johnson
,
M. P.
,
Madsen
,
P. T.
, and
Tyack
,
P. L.
(
2005
). “
Echolocation clicks of free-ranging Cuvier's beaked whales (Ziphius cavirostris
),”
J. Acoust. Soc. Am.
117
,
3919
3927
.

Supplementary Material