Update on frequency decline of Northeast Pacific blue whale (Balaenoptera musculus) calls.

Worldwide, the frequency (pitch) of blue whale (Balaenoptera musculus) calls has been decreasing since first recorded in the 1960s. This frequency decline occurs over annual and inter-annual timescales and has recently been documented in other baleen whale species, yet it remains unexplained. In the Northeast Pacific, blue whales produce two calls, or units, that, when regularly repeated, are referred to as song: A and B calls. In this population, frequency decline has thus far only been examined in B calls. In this work, passive acoustic data collected in the Southern California Bight from 2006 to 2019 were examined to determine if A calls are also declining in frequency and whether the call pulse rate was similarly impacted. Additionally, frequency measurements were made for B calls to determine whether the rate of frequency decline is the same as was calculated when this phenomenon was first reported in 2009. We found that A calls decreased at a rate of 0.32 Hz yr-1 during this period and that B calls were still decreasing, albeit at a slower rate (0.27 Hz yr-1) than reported previously. The A call pulse rate also declined over the course of the study, at a rate of 0.006 pulses/s yr-1. With this updated information, we consider the various theories that have been proposed to explain frequency decline in blue whales. We conclude that no current theory adequately accounts for all aspects of this phenomenon and consider the role that individual perception of song frequency may play. To understand the cause behind call frequency decline, future studies might want to explore the function of these songs and the mechanism for their synchronization. The ubiquitous nature of the frequency shift phenomenon may indicate a consistent level of vocal plasticity and fine auditory processing abilities across baleen whale species.

Introduction

The calls produced by multiple blue whale (Balaenoptera musculus) populations have been declining in frequency since they were first recorded in the 1960s [1]. This phenomenon was first reported by McDonald et al. [1], and subsequent studies have confirmed a downward frequency shift across blue whale populations [2-6] and a similar trend was noted in fin whales (Balaenoptera physalus) [2, 7] and bowhead whales (Balaena mysticetus) [8].

The Northeast Pacific population of blue whales is arguably the best studied, and calls from this region have been recorded for decades [9-13]. This population produces three types of low-frequency calls: A and B calls are long-duration (20 s), pulsed and tonal calls, respectively, that often occur in a sequence and are believed to serve a reproductive purpose [14-17] and D calls are downsweeps that last several seconds and are considered a social call, possibly associated with feeding [10, 14, 17]. In the first report on frequency shift in this population, the frequency of the 3rd harmonic of the B call had decreased from 65.7 to 45.5 Hz over 45 years; a rate of 0.4 Hz yr-1 from 1963 to 2008 [1].

A variety of hypotheses have been proposed to explain this frequency shift, with explanations ranging from environmental changes to population density and behavioral changes. However, there is currently no consensus on the underlying cause of this frequency decline, as many of the suggested hypotheses, such as an increase in body size, biological masking, and climate change have been refuted [1, 2, 7, 8, 18, 19]. The discovery that this long-term frequency decline was coupled with an intra-annual frequency decline that resets each year, provides further context for the potential mechanism behind this long-term trend [2-4], as does the occurrence of this phenomenon in other baleen whale species [2, 7, 8]. While it is possible that each of these findings have different causes, the most parsimonious explanation would account for both the intra- and inter-annual frequency decline in blue whales, as well as frequency decline in other baleen whale species.

Using passive acoustic recordings collected in recent years, it is possible to (1) examine whether frequency decline is occurring in Northeast Pacific blue whale A calls, (2) examine whether the pulse rate of A calls is changing, and (3) extend the dataset of McDonald et al. [1] to observe the current rate of frequency decline in Northeast Pacific blue whale B calls. The results provide insights to further support or refute existing hypotheses and determine if one hypothesis can account for all aspects of this currently unexplained phenomenon.

Methods

Data collection

Data were collected from 2006 to 2019 using high-frequency acoustic recording packages (HARPs) deployed at three locations in the Southern California Bight (Fig 1). The three recording locations were sites E (1,300 m), H (1,100 m), and N (1,265 m; Fig 1; S1 Table). HARPs were deployed on the seafloor and had a calibrated hydrophone suspended 10–30 m above the instrument [20]. Data were sampled continuously for all deployments but one, for which the recording was duty cycled (S1 Table). Data were collected at a 200-kHz sampling frequency but were decimated to a 2-kHz sampling frequency to allow more effective scanning of long-term spectral averages (LTSAs). LTSAs were created with 5-s time average and 1-Hz frequency resolution.

Fig 1

Map of recording locations offshore of Southern California.

Recording locations H, E, and N are shown as black circles. Color bar indicates depth in kilometers and thin grey lines mark 500 m bathymetry contours. Map generated using the marmap and ggplot2 packages in R [21–23] and bathymetry data from the ETOPO1 database hosted by NOAA [24, 25].

Call measurements

For each year of data from 2006 to 2019, hourly LTSAs were manually scanned during September and October (S1 Table) for high-quality (high signal-to-noise ratio) blue whale A and B calls using Triton, a custom MATLAB program designed to allow visual scans of LTSAs [20]. The following analysis was conducted for A and B calls separately. When a call was identified in the LTSA, a 60-s spectrogram (2,000-point fast Fourier transform (FFT) length, 90% overlap), displayed up to 200 Hz, was used to confirm the quality of the call. When a call was selected for further analysis, custom MATLAB code was used to identify the start and end time of the call. For A calls, average peak frequency was calculated over the full duration of the most dominant call overtone (Fig 2). The number of pulses in each A call was also counted, and it was divided by call duration to determine the pulse rate. For B calls, the peak frequency was measured from the 3rd harmonic of the call every 0.5 s, for the duration of the call. The 3rd harmonic was used, as in McDonald et al. [1], because it typically contains the most energy and is therefore more consistently recorded than the fundamental frequency [10, 12]. The frequency measurement at 10 s from the start of the call was used for subsequent analysis (Fig 2). Before B call frequencies were measured, spectrograms were adjusted to an 8,000-point FFT length, resulting in a frequency resolution of 0.25 Hz. Due to the pulsed nature of A calls, a 2,000-point FFT length was used, providing a frequency resolution of 1 Hz.

Fig 2

Northeast Pacific blue whale A and B call.

Calls were recorded on October 17, 2015 at site N. The dashed box indicates the overtone (A call) and harmonic and time point (B call) for which frequencies were used to compare between years.

To select calls for this analysis, we started with visual inspection of data from October 1st and continued until frequency measurements were extracted for 30 calls separated by at least 24 h. If 30 calls were not selected in October, data were scanned starting at the end of September and working backwards until 30 calls had been measured. The months of September and October were selected for frequency measurements because this is a known period of high A and B call production in the Southern California Bight [9, 16, 26]. The sampling routine of 30 calls separated by at least 24 h each was selected to avoid oversampling any individual blue whale.

When selecting the time point to use for comparing B call frequency across years, consideration was given to the methods used by McDonald et al. [1], as the goal was to compare the rate of frequency decline in B calls established here to what had been observed previously. For B calls, the frequencies of the 30 calls selected each year were averaged and 95% confidence intervals were calculated along the length of the call (Fig 3). Confidence interval ranges for these yearly averages were lowest from 4.5 to 10 s into the call and highest from 12 to 15 s (Fig 3). The period of high variation is a result of Northeast Pacific blue whale B calls offshore of Southern California exhibiting occasional gaps during this portion of the call [27]. However, only the frequency at 10 s was used for interannual trend analyses. This time point was selected due to the aforementioned low confidence interval ranges and to avoid possible methodological confounds when comparing results to those reported by McDonald et al. [1], who measured frequencies from the estimated midpoint of the call. From the 420 calls measured in this study, the average call duration was 19.6 s, and so we assume that the measurements at 10 s in this study are comparable to those made by McDonald et al. [1].

Fig 3

Mean frequencies of Northeast Pacific blue whale B calls.

The first 20 s of the 3rd harmonic are shown each year from 2006 to 2019. Shaded areas represent 95% confidence intervals and the dashed line indicates 10 s point that was used for yearly comparisons.

For both A and B calls, the frequencies of the 30 calls measured for a given year were averaged, providing one representative call frequency per year, and 95% confidence intervals were calculated. In the same way, A call pulse rate was also averaged each year. The call measurements from all calls are available in S2 Table and S3 Table, for A and B calls, respectively. To examine interannual trends, the least squares method was used to determine the best fit to the data and the percent of change between years was calculated.

Results

A calls

Northeast Pacific blue whale A calls showed a 0.32 Hz yr-1 decrease in frequency over 13 years from 2006 to 2019 (Fig 4). The mean frequency was 85.6 Hz in 2006 and 81.5 Hz in 2019 (Fig 4; S1 Table). The best fit for these data was modelled with a second order polynomial (R2 = 0.97). Frequency did not consistently decrease each year; for example, in 2009, 2017, and 2019 the mean frequency was higher than in the previous year (Fig 4). There was no clear trend in the yearly percent of change, which ranged from 0 to 1% (Fig 5).

Fig 4

Mean frequency (a) and pulse rate (b) of Northeast Pacific blue whale A calls from 2006 to 2019. Error bars represent 95% confidence intervals and the dashed black line represents a quadratic fit to the data.

Fig 5

Percent of change of the mean frequency and pulse rate of Northeast Pacific blue whale calls.

The percent of change is shown for A calls each year from 2006 to 2019 and for B calls from 1963 to 2019. Data in black are from the current study and data in red are from McDonald et al. [1].

The A calls showed a 0.006 pulses/s yr-1decrease in pulse rate over the course of the study. The mean pulse rate was 1.241 pulses s-1 in 2006 and 1.158 pulses s-1 in 2019 (Fig 4; S1 Table). The best fit for pulse rate was modelled with a second order polynomial (R2 = 0.84). However, the pulse rate did not consistently decrease each year, and actually increased in half of the years measured (Fig 4; S1 Table). There was no clear trend in the yearly percent of change, which ranged from -3.3 to 1.3% (Fig 5).

B calls

Northeast Pacific blue whale B calls showed a 0.27 Hz yr-1 decrease in frequency over 13 years from 2006 to 2019 (Fig 6A). The mean frequency at the 10 s point of the 3rd harmonic was 46.1 Hz in 2006 and 42.6 Hz in 2019 (Fig 6; S1 Table). The best fit for this data was modelled with a second order polynomial (R2 = 0.99; Fig 6B).

Fig 6

Mean frequencies of Northeast Pacific blue whale B calls.

Error bars represent 95% confidence intervals for mean frequencies at 10th second of the 3rd harmonic of the calls. Data from McDonald et al. [1] are shown in red and data from the current study in black for (a) 2006 to 2019 and (b) 1963 to 2019. The dashed red line represents a linear fit to data from McDonald et al. [1] and the dashed black line represents the quadratic fit to data from the current study.

From 2006 to 2008, our data overlapped with the dataset analyzed by McDonald et al. [1] (Fig 6A). The B call frequencies we measured fell within the 95% confidence intervals reported by McDonald et al. [1], with a maximum difference in mean frequency of 0.1 Hz over the three years (Fig 6A). The yearly percent of change showed more variation for the frequencies reported by McDonald et al. [1] (0–6.6%) compared to those reported here (0.21–1%), and appeared to stabilize after 2000 (Fig 5).

Discussion

Frequency and pulse rate decline in Northeast Pacific blue whale calls

From 2006 to 2019, both units of Northeast Pacific blue whale song have been declining, as has been observed in other blue whale populations [18, 19]. We know B calls have been declining in frequency since at least the 1960s [1], and even though there has not previously been a detailed examination of frequency shift in A calls, this decline has been occurring for longer than reported here; descriptions of these calls from offshore of California reveal that in 1997 the frequency of the overtone was around 90 Hz [10]. Although rates of frequency decline were comparable between the two calls, the annual change in pulse rate of the A calls did not show the same consistent decline.

For B calls, the rate of decline we report here, at 0.27 Hz yr-1, is slower than the rate reported for this population by McDonald et al. [1], which was 0.4 Hz yr-1 from 1963 to 2008 (Fig 6). These differences are likely not due to methodological inconsistences because, although McDonald et al. [1] measured frequency from the “midpoint” of the call and we used the 10-s mark, we found that the frequency of the call showed the least amount of variation from 4.5 to 10 s (Fig 3). Additionally, of the 420 calls that were measured, the average call duration was 19.6 s, suggesting that the call midpoint used by McDonald et al. [1] should have been close to the 10-s mark used here. We were also able to show that our measurements fell within the confidence intervals reported by McDonald et al. [1] for the years when the two studies overlapped (2006–2008; Fig 6A). Therefore, even if the point at which frequencies were measured was slightly different between studies, it would not explain the differences in rates of decline that we report here.

Another source of variation between our study and McDonald et al. [1] is the months that were sampled for calls. Because the frequency of Northeast Pacific blue whale B calls decreases intra-annually as well as inter-annually [4], call frequencies must be measured at the same time each year to accurately evaluate inter-annual trends. To avoid potential confounds from this seasonal influence, we only measured calls recorded in September and October. However, the calls measured by McDonald et al. [1] were sampled at various times throughout the years, often whenever data were available. While this may explain some variation with the McDonald et al. [1] dataset, during the three years that overlapped with our data, calls were measured in August and September, and we would not expect a large difference with the frequencies we measured in September and October. Again, this is supported by the comparable frequencies we report for the years where our data overlapped with those of McDonald et al. [1]. Therefore, we conclude that rate of frequency decline we report here is not different from that reported by McDonald et al. [1] as a result of methodology, and is instead an indication that Northeast Pacific blue whale B calls are declining in frequency at a slower rate than in previous years studied.

Potential causes of frequency decline

Frequency decline in calls from blue whale populations worldwide [1-6] is evidence of call synchronization within populations, which may provide some adaptive advantages [28]. However, the question remains why this frequency decline is occurring and whether the new information presented, that A call frequency and pulse rate are also decreasing and that B call frequency is declining at a slower rate than reported previously, point towards one of the previously proposed hypotheses regarding this phenomenon. The decrease in the rate of decline may be a simple result of a physical limit being reached for vocalization frequency, but it is worth considering whether there is an alternative explanation for this observation. For the previously proposed hypotheses, we must consider whether they can explain all aspects of this phenomenon: the inter- and intra-annual frequency decline in blue whales worldwide, the decrease in the rate of frequency decline observed here in blue whales, and the observed frequency decline in other baleen whale species. While it is possible that multiple factors are responsible for these different observations, the most parsimonious explanation would provide a root cause that accounts for all these observations.

Hypotheses on seasonal changes in body condition [29] and calling depth [5] have been suggested as explanations for intra-annual frequency decline, but it is understood that these theories cannot explain the inter-annual trend [3, 29]. For calling depth, although call frequency can be impacted by depth, there is no biological basis to explain why blue whales, and other species for which frequency shift has been documented, would all collectively be calling at greater depths each year. Tag data also shows that calling depth can vary over a short time period [17], but because call frequency shows limited variation [1, 3, 18, 19, 28], it does not appear that call frequency is solely influenced by depth [19]. Therefore, we will not examine these hypotheses in further detail.

Although we know that ocean ambient noise has increased over the period that blue whale calls have declined in frequency [30-32], the expected response to avoidance of masking would be to call at higher frequencies, as a result of producing higher amplitude calls, as has been observed in right whales [33, 34]. Therefore, if blue whales were trying to be heard in a louder ocean, we would expect their calls to be increasing in frequency, not decreasing. Additionally, ambient noise has actually decreased in the Southern Indian Ocean, but blue whale calls in this region still exhibit a frequency decline [2], meaning that ambient noise does not provide a consistent explanation for changing blue whale call frequencies. However, Leroy et al. [2] suggested that seasonal changes in ambient noise could still explain intra-annual frequency decline because, at multiple sites in the Southern Indian Ocean, intra-annual frequency change in both Antarctic blue whales and fin whales was correlated with seasonal changes in ambient noise levels [2]. While it is possible that different factors are driving intra- and inter-annual frequency decline, the evidence presented is only that of two seasonal patterns. Blue whales are already known to exhibit seasonal changes in their acoustic behavior [16]; therefore, annual fluctuations in ambient noise should not necessarily be accepted as the driver of intra-annual blue whale frequency decline.

One theory originally proposed by McDonald et al. [1] was that call frequency is related to population density. This theory suggests that blue whales called at higher amplitudes, and consequently higher frequencies, to be heard by conspecifics after their populations were reduced by commercial whaling. As populations have increased, blue whales have been able to call at lower amplitudes because the distance between conspecifics has decreased. This assumes that there is a trade-off between frequency and amplitude that males exploit to produce lower frequency calls that may serve as an indicator of body size and fitness in inter- or intra-sexual selection [1]. Therefore, this theory could conceivably explain inter-annual frequency decline in other baleen whale species that were also impacted by commercial whaling, as well as account for the differing rates of frequency decline observed in different populations, as each population presumably recovers at its own rate [18]. However, offshore of Australia, inter- and intra-annual frequency decline has been documented from 2002 to 2017 in the spot call, which is presumed to be from a baleen whale, though the species has yet to be confirmed [35]. This call underwent a dramatic (>5 Hz) increase in frequency between 2006 and 2007 and then began to gradually decline again through 2017 [35]. This abrupt frequency increase could not be explained by the population density theory. Additionally, McDonald et al. [1] hypothesized that, if this theory were the mechanism behind blue whale frequency shift, song frequencies would stabilize in conjunction with population densities. The Northeast Pacific blue whale population was reduced by historical whaling [36, 37] and is still considered endangered. Although there is evidence the population may have returned to pre-whaling levels [38], recent data suggest that this population has experienced a recent growth after relatively stable numbers between the mid-1990s and mid-2010s [39]. It is not clear whether the change in the rate of decline we observed matches these variations in abundance trends.

The population density theory also assumes a relationship between source level and frequency that has not been observed in blue whales [5] and has been refuted in bowhead whales [8]. Finally, this theory does not account for intra-annual frequency decline, as it would require that animal densities steadily change within each year. Off the U.S. West Coast, blue whale distribution is not consistent between years, likely related to shifts in oceanographic conditions [40-42]. In bowhead whales, Thode et al. [8] found that a decrease in call frequency was predicted by an increase in call density, but this relationship has not been examined in blue whales. Therefore, the population density theory may not serve as a parsimonious explanation for all aspects of the frequency-shift phenomenon.

Finally, the decrease in pulse rate, documented here in A calls, has been previously observed in other blue whale populations’ songs [6, 19]. Changes in pulse rate have also been documented in fin whale populations [7, 43], though these are inter-call rates as opposed to the intra-call rates described for blue whales. These observations add an additional layer of complexity to the frequency shift phenomenon and are not explained by any of the previously discussed theories.

Because A calls are a pulsed signal and frequency shift was measured from an overtone, it is possible that the amplitude modulation of the signal was changing and causing the documented frequency shift of the overtone, rather than it being caused by the shift in the carrier frequency. However, although pulse rate declined over the years examined, the decline was not consistent each year and did not directly correspond to the frequency decline of the same year. For example, from 2008 to 2009 the frequency increased slightly but the pulse rate decreased, whereas from 2010 to 2011 the frequency decreased and the pulse rate increased. If future studies are to examine frequency decline in A calls, it might be worthwhile to measure frequencies of both carrier frequency and the most energetic overtone to be able to eliminate the possibility of frequency shift being caused by amplitude modulation changes, and to compare to the frequencies reported here. It might also be valuable to examine the received levels of the measured calls, as it is possible that variations in propagation and thus received level could have affected measured frequencies.

Much of the discussion around frequency shift has been focused on the production aspects and possible mechanisms and limits in sound production. However, it is also worth considering the role perception may play in baleen whales, especially for small frequency changes. Baleen whale auditory perception is poorly understood, although we know they have a thick auditory cortex [44] and we can assume that, as in other species, auditory processing depends on auditory as well as behavioral stimuli [45]. One of the unresolved questions is, can the whales perceive this gradual frequency shift? We do not know the frequency difference limen, the smallest detectable change in frequency, for baleen whales, but if we assume it is similar to humans’, their ability to distinguish small frequency changes could range between 0.2 and 0.09 Hz at 45 Hz [46]. At the high end, it would mean that whales might be able to perceive the overall seasonal change, but their consistent smaller frequency adjustment throughout the season [3] could be below the whale’s perceptive abilities of frequency differences. If the whales cannot, in fact, perceive this change, the gradual frequency shift may be an unintended consequence of a currently not understood song synchronization mechanism. It is clear we still have much to learn about whale auditory perception, but the ubiquitous nature of the frequency shift phenomenon may indicate a consistent level of vocal plasticity and fine auditory processing abilities across baleen whale species.

Conclusions

Examination of Northeast Pacific blue whale song units from 2008 to 2019 revealed a frequency decline in A calls, as well as a decrease in the rate of frequency decline in B calls from that reported for the second half of the 20th century. A decrease in the pulse rate of A calls was also documented. With this new information, we were able to revisit previously proposed hypotheses that have attempted to explain the worldwide frequency decline of blue whale song [1]. Although none of the currently proposed theories successfully account for all aspects of the frequency shift phenomenon, documentation of frequency decline and other changes to call characteristics in baleen whale species should continue in an effort to discover the mechanism underlying this trend.

High-frequency acoustic recording package deployment details.

The year, site, latitude, longitude, depth, and analysis period for each recording package deployment are provided, along with the associated mean frequency (and pulse rate for A calls) and 95% confidence intervals (CI) for both A and B calls each year.

(PDF)

Click here for additional data file.

Northeast Pacific blue whale A call measurements.

Call duration, average peak frequency, number of pulses, and pulse rate for each call measured each year from 2006 to 2019.

(XLSX)

Click here for additional data file.

Northeast Pacific blue whale B call measurements.

Frequency measurements of the 3rd harmonic for each call measured each year from 2006 to 2019. NaNs signify that the call did not extend to the specified time point.

(XLSX)

Click here for additional data file.

2 Dec 2021

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

William David Halliday, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that Figure 1 in your submission contain map images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

a. You may seek permission from the original copyright holder of Figure(s) [#] to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

b. If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Review of Frequency decline in blue whale calls

Firstly, my apologies to the authors for the delay in providing the review. Just too heavy a a workload.

The paper "Review of Frequency decline in blue whale calls" is well written and requires no grammatical edits. My suggestions are to do with: 1) the interpretation of frequency shifts between fundamental frequencies and harmonics or overtones; and 2) some additional ideas for discussion.

The authors have used techniques widely used by others. There is the potential for errors of interpretation in what is being observed (which applies to earlier works also) so I am suggesting some additional analysis as per below.

Main comment:

1/

Harmonics are caused by amplitude modulation of a carrier tone. The harmonic spacing relates to the frequency of amplitude modulation. If there is no amplitude modulation of the carrier tone then there are no harmonics or overtones. Thus if you measure frequency shift of harmonics there are potentially two causes which can act independently: 1) a change in frequency of the carrier tone which would result in a linear change in frequency of the harmonic, assuming the harmonic amplitude modulation rate did not change (this is what your analysis and those of others assumes); or 2) a change in the amplitude modulation rate which would result in the overtone or harmonic having a different frequency 'slope' through a call than the carrier frequency. Thus if you measure a frequency shift of the harmonic you need to be confident that the amplitude modulation rate which produces the harmonic has stayed constant through the call, if you are to assume the frequency shift applies to the entire call.

At a worst case it is possible the whale call carrier frequency has not changed but the amplitude modulation rate has, which will be reflected in a frequency decline of the harmonics. While this is not the case with great whales and the call carrier frequency does decline over time there needs to be some better rigour applied when changes in frequencies of harmonics are analysed and in particular compared between authors, whom for the same species may be comparing frequency changes measured between harmonics and carrier frequencies, which are generated by different phenomena (carrier frequency by the physics of the sound generation apparatus and harmonics by the amplitude modulation rate of the carrier frequency).

If you are to compare frequency rates with the earlier study of McDonald et al, then you should be comparing the same harmonics.

While it means more analysis, it would also be good (and provide more rigour) if you could compare rates of change of carrier frequency with harmonic frequency (a reflection of amplitude modulation rate) through a call to determine which is changing - carrier frequency, amplitude modulation rate, or both? If the amplitude modulation rate does not change within a call then your analysis (and all the other published studies which use harmonic frequencies to demonstrate frequency changes over time) stands as they are. If the amplitude modulation changes through time then that opens new questions on what is going on and confuses the yearly comparisons, particularly the amount of frequency decline observed in different studies.

I do think you should do this analysis, otherwise we perpetuate the potential (am not saying it is always present or has occurred here) confusion of what is being measured - changes to carrier frequency, amplitude modulation rate if harmonics are measured, or both?

It would also help if you had a physicist to elaborate in the introduction, the production of harmonics or overtones and how this impacts what you are trying to measure (frequency shifts across a long time frame).

2/

Extra points for discussion:

Ward et al. (2017) show a 'resetting' of carrier frequency (this call type has no overtones) in a great whale species (see Fig. 5). This species should be 'resetting' its carrier frequency again, around now. It may be worth mentioning this. The 'resetting' implies a conscious effort on the part of the animal to shift its frequency as it occurs over such a short time period (one or two seasons) which is far too short to represent an environmental shift.

I disagree with the impact depth can have on whale carrier frequency. You have implied that depth has little impact. Tags attached to whales often have poor resolution (0.5 m is common), thus require high sample rates to infer small scale trends of changes in depth (ie. 20 s across a call length). We have a work in progress using a high sample rate tag which suggests call frequency within a blue whale species' call is related to the whale's depth. This would imply the animal can consciously change its call carrier frequency by changing its singing depth by a small amount. The species in question does this within a call. You can ignore this if you wish but I would suggest you at least say the animals singing depth will play a role in call carrier frequency changes.

Ward, R., Gavrilov, A. N., McCauley, R.D. (2017) “Spot” call: a common sound from an unidentified great whale in Australian temperate waters. J. Acoust. Soc. Am. Express Letters 142(2) EL231-236, doi.org/10.1121/1.4998608

Reviewer #2: Please see attachment for comments and plot of pulse rate vs. time.

Basic comments are that statistical regression needs to control for other factors, and analysis should also include pulse rate for type A calls.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Robert McCauley

Reviewer #2: Yes: Aaron Thode

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Review.pdf

Click here for additional data file.

Submitted filename: TypeARateVstime.jpg

Click here for additional data file.

14 Jan 2022

Journal Requirements: There was a copyright concern over the map used in Figure 1. However, the bathymetry data used in this map is from the ETOPO1 database that is publicly available from NOAA, so there are no copyright concerns with the use of this image. However, for clarity, we have added the source of the bathymetry data to the figure caption and include appropriate references for the use of this data, as requested by NOAA.

Reviewer 1 Responses:

Comment 1: Harmonics are caused by amplitude modulation of a carrier tone. The harmonic spacing relates to the frequency of amplitude modulation. If there is no amplitude modulation of the carrier tone then there are no harmonics or overtones. Thus if you measure frequency shift of harmonics there are potentially two causes which can act independently: 1) a change in frequency of the carrier tone which would result in a linear change in frequency of the harmonic, assuming the harmonic amplitude modulation rate did not change (this is what your analysis and those of others assumes); or 2) a change in the amplitude modulation rate which would result in the overtone or harmonic having a different frequency 'slope' through a call than the carrier frequency. Thus if you measure a frequency shift of the harmonic you need to be confident that the amplitude modulation rate which produces the harmonic has stayed constant through the call, if you are to assume the frequency shift applies to the entire call.

At a worst case it is possible the whale call carrier frequency has not changed but the amplitude modulation rate has, which will be reflected in a frequency decline of the harmonics. While this is not the case with great whales and the call carrier frequency does decline over time there needs to be some better rigour applied when changes in frequencies of harmonics are analysed and in particular compared between authors, whom for the same species may be comparing frequency changes measured between harmonics and carrier frequencies, which are generated by different phenomena (carrier frequency by the physics of the sound generation apparatus and harmonics by the amplitude modulation rate of the carrier frequency).

If you are to compare frequency rates with the earlier study of McDonald et al, then you should be comparing the same harmonics.

While it means more analysis, it would also be good (and provide more rigour) if you could compare rates of change of carrier frequency with harmonic frequency (a reflection of amplitude modulation rate) through a call to determine which is changing - carrier frequency, amplitude modulation rate, or both? If the amplitude modulation rate does not change within a call then your analysis (and all the other published studies which use harmonic frequencies to demonstrate frequency changes over time) stands as they are. If the amplitude modulation changes through time then that opens new questions on what is going on and confuses the yearly comparisons, particularly the amount of frequency decline observed in different studies.

I do think you should do this analysis, otherwise we perpetuate the potential (am not saying it is always present or has occurred here) confusion of what is being measured - changes to carrier frequency, amplitude modulation rate if harmonics are measured, or both?

It would also help if you had a physicist to elaborate in the introduction, the production of harmonics or overtones and how this impacts what you are trying to measure (frequency shifts across a long time frame).

Response: True harmonics, integer multiples of the fundamental frequency, are not caused by amplitude modulation of a carrier tone, as commented, but result from distortion of the signal wave in the vocal tract. The B calls of Northeast Pacific blue whale songs are not amplitude modulated signals and have true harmonics. An amplitude modulated signal, such as a Northeast Pacific blue whale A call, has non-harmonic overtones (or side bands). Therefore, measuring the 3rd harmonic of blue whale B calls is an accurate way to measure frequency shift, and is comparable across studies, regardless of which harmonic is being measured (provided the harmonic multiplier is accounted). Nevertheless, to demonstrate this, we used a subset of our data (20 calls, every other year) and performed the same analysis as described in our study but using the fundamental frequency of the B calls. We found that the fundamental frequency was declining at a rate of 0.09 Hz/yr, which corresponds to the 0.27 Hz/yr decline we report for the 3rd harmonic. In the figure below, which shows the average frequency measurements and 95% confidence intervals for the first 20 s of the calls for each year, you can see that the confidence intervals are larger than reported for the 3rd harmonic (shown in the manuscript in Figure 3) because the fundamental frequency is often faint and can be masked by other low frequency signals, such as from fin whale 20 Hz calls. As stated in the manuscript, the 3rd harmonic typically contains the most energy and can therefore be more consistently and accurately measured than the fundamental. Therefore, we believe that frequency shift should be measured using the 3rd harmonic instead of the fundamental.

Although we conducted this analysis, we feel it is not worth including in the manuscript, as it is simply a confirmation of what we would expect based on the physics behind the production of harmonics. Additionally, the McDonald et al. (2009) study that we compare our rate of frequency decline to was also measuring the 3rd harmonic. This was stated in the manuscript introduction but is also now restated in the methods, to avoid potential confusion (lines 98–99).

However, because blue whale A calls are amplitude modulated signals, and we were measuring frequency shift using an overtone for this call type, it is possible that amplitude modulation had some influence on its frequency decline. Reviewer 2 commented that there was a decline in the pulse rate of A calls over the years of the study, which was not something we investigated in depth. We have now included an examination of the decline in A call pulse rate in the manuscript as suggested, which includes discussion of the relationship between carrier frequency and overtones (lines 290–302). So, while it is certainly true that it will be important in future studies to take this into consideration when measuring frequency shift in blue whale A calls, we currently do not have the amplitude measurements in hand. We suggest that our proposed edits sufficiently address the reviewer concerns.

Comment 2: Extra points for discussion:

Ward et al. (2017) show a 'resetting' of carrier frequency (this call type has no overtones) in a great whale species (see Fig. 5). This species should be 'resetting' its carrier frequency again, around now. It may be worth mentioning this. The 'resetting' implies a conscious effort on the part of the animal to shift its frequency as it occurs over such a short time period (one or two seasons) which is far too short to represent an environmental shift.

I disagree with the impact depth can have on whale carrier frequency. You have implied that depth has little impact. Tags attached to whales often have poor resolution (0.5 m is common), thus require high sample rates to infer small scale trends of changes in depth (ie. 20 s across a call length). We have a work in progress using a high sample rate tag which suggests call frequency within a blue whale species' call is related to the whale's depth. This would imply the animal can consciously change its call carrier frequency by changing its singing depth by a small amount. The species in question does this within a call. You can ignore this if you wish but I would suggest you at least say the animals singing depth will play a role in call carrier frequency changes.

Response: We have added a brief discussion of the Ward et al. (2017) paper where we describe evidence to refute the abundance hypothesis (lines 262�  267).

We did not mean to imply that depth has little impact on call frequency, only that the impact it could have does not adequately explain the inter-annual frequency shift. If a decrease in call depth were the explanation behind the frequency shift phenomenon it would mean that blue whales are collectively calling at greater depth each year. Without evidence to support this, or a biological precedent upon which to base such a claim, we did not want to speculate on this further. The evidence we have against this explanation is the fact that tag data has shown that calling depth can change over a short time period (Lewis et al. 2018). If frequency was progressively changing with depth, we would expect call frequencies to be much more variable within a year than they are reported to be. So, although depth may be a factor in frequency changes in some calls and at some times, it does not adequately explain the consistent frequency shift phenomenon based on the information currently available. We have made this assertion clearer in the text (lines 229�  234).

Reviewer 2 Responses:

Comment 1: Statistical regression: The paper performs a simple least-squares fit to the frequency data of both types of calls. Unfortunately, apparent peak frequency can be affected by factors other than the source (such as range and instrumentation), and the dataset is large enough that some of these factors can be checked. At a minimum the authors should use a generalized linear model (GLM) using received level as a predictor variable, HARP unit/location as a categorical predictor variable, and year as another predictor variable. The importance of received level is that it is a proxy for horizontal range, and there is abundant evidence that range and environmental propagation conditions impact peak frequency in other animals. Even if the water depths here seem deep (1 km in some locations), frequency-dependent attenuation and bottom interactions cannot be ruled out.

In addition, the paper should clarify whether the FFT length used to analyze the peak frequency of the type A call was shorter than the pulse rate. If the FFT length is too long, then the spectrum of the type A call will shift from the structure shown in Figure 2 to a “comb-like” spectrum where the separation between comb peaks is the reciprocal of the pulse rate. It’s important to ensure that the 75 Hz frequency peak measured is due to the “formant” shift and not from changes in the pulse rate.

Response: The influence of received level is not a concern for B calls because we were comparing frequencies that were measured at a point in the call when the signal becomes nearly tonal. Because these calls are essentially a tone and not a broadband signal, there would be no change to the peak frequency we measured as a result of propagation. Even if our measurements varied by a few seconds within the call, the frequency change would be so small (<0.5 Hz) that it would not be sufficiently impacted by propagation but would contribute to the relatively tight confidence intervals.

However, this is a potential concern for A calls, for which frequency measurements were averaged over the duration of the call. Unfortunately, determining the received level of each call would require substantial additional re-analysis because we did not extract received level information with our initial measurements and this re-analysis was not feasible at this time. Although we were not able to include the GLM as requested, we have included a more thorough discussion of changes in A calls, since we now incorporate an analysis of pulse rate changes, and used that to also comment on this concern (lines 284–302).

The FFT length used for the A call analysis was 2000 (sampling frequency was 2000 Hz, resulting in a 1-Hz resolution), as stated in the methods, and is therefore shorter than the pulse rate. There were 2 calls measured (out of 420) for which pulse rate was 1 s, and one where it was 0.9 s, but all other calls had a pulse rate above 1 s. This is certainly a factor that should be considered if future studies were to compare measurements of A calls to ours.

Comment 2: Pulse rate analysis: the attached figure, generated from supplemental table S2, shows that the pulse rate of type A calls is decreasing over time, to the point that by 2019 the rate has shifted one standard deviation from 2006. This would suggest that a more rigorous regression would find a significant relationship between pulse rate and time, an observation that may have a bearing on the extensive discussion.

Response: We now include an analysis of A call pulse rate in the manuscript. Thank you for this suggestion.

Comment 3: About that extensive discussion: I would make it shorter, as a detailed list of speculations does not enhance the paper quality.

Response: We removed one paragraph from the discussion that described 4 theories that have been refuted in all other papers on this subject. These are now very briefly referenced in the introduction (lines 57–59). For the other theories, we feel they are still worth discussing as previous publication have argued that they may explain some part of the frequency decline phenomenon, while we argue for a more parsimonious explanation.

Submitted filename: Rice et al.-Response to Reviewers.docx

Click here for additional data file.

22 Mar 2022

Update on frequency decline of Northeast Pacific blue whale (Balaenoptera musculus) calls

PONE-D-21-29018R1

Dear Dr. Rice,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements. Both reviewers assessed your revisions, and feel that you have addressed all of their comments appropriately.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

William David Halliday, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Thank you to the paper authors for considering and addressing comments. A comment, to decrease call frequency for a bubble the bubble depth must decrease, only by a small amount for the frequency shifts observed. This would apply to a whale lung space. Perhaps there are differing mechanisms for producing frequency shifts amongst different whale species, which would make the long term frequency decline more puzzling.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

24 Mar 2022

PONE-D-21-29018R1

Update on frequency decline of Northeast Pacific blue whale (Balaenoptera musculus) calls

Dear Dr. Rice:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. William David Halliday

Academic Editor

PLOS ONE