Oceanic Ambient Sound
APL/UW has been collecting ambient sound from omnidirectional sensors located throughout the North Pacific basin since 1994. An example of such long-term collection is the spectrogram shown below. This data was acquired at Sur Ridge, off the coast of California. The seasonal increases in intensity around 20 Hz are due to blue whales.
One of the significant findings provided by our collection program is the corroboration of the increase in ambient sound in the low-frequency regime. Our first paper on this increase (Andrew et al, ARLO 2002, see below) reported an increase of about 10 dB below 100 Hz for the location of Pt. Sur, California.
This finding was recently corroborated by McDonald et al (see Mark's website ) for a second site off San Nicolas Island.
The measurements off San Nicolas assisted and motivated us to rediscover the correct calibrations for more of the sites in our data collection program. We were able to correct the datasets from three more sites located off the North American west coast, bringing the total number of locations where we can compare contemporary noise levels to those measured by Gordon Wenz in the mid 1960s to four.
Comparisons of ambient sound levels between the 1960s and the present are shown in the figure below. The figure uses curves for 32, 40 and 50 Hz, frequencies at which the contribution due to distant merchant shipping is expected to dominate other sources. (Hence, this is the least ambiguous indication of the anthropogenic influence on oceanic ambient noise.) The markers for the mid-1960s represent about 1 to 2 year averages. The heavy dotted line represent a regression line calculated by Ross through a cloud of such data points (not shown here): this line had a slope of 0.55 dB/year. Ross speculated that this rate of increase would not be sustained through the end of the 20th century due to economic factors.
On the right side, a trend line estimated from the APL datasets is shown for each site for each frequency. These are connected by light dotted lines to their earlier Ross projections merely for convenience --- there is no suggestion that the ambient noise actually followed those curves. It can be seen that all four sites have had increases of roughly 10 dB at these frequencies. For the two southern sites (D and F), it appears that these levels are perhaps increasing slightly or staying steady, but for the two northern sites (G and H), the levels appear to have peaked prior to the initiation of our collection program and are currently decreasing.
It is not currently known how to interpret these contemporary trends.
A full description of these results appears in Andrew et al, "Long-time trends in ship traffic noise for four sites off the North American West Coast," J. Acoust. Soc Am., 129(2), DOI: 10.1121/1.3518770, accessible here. Alternatively, here is an electronic copy that may be downloaded for personal use only. Any other use requires prior permission of the author and the Acoustical Society of America. Copyright (2011) Acoustical Society of America).
A second intriguing application of this dataset is the validation of part of the Dynamic Ambient Noise Model (DANM). This was a joint collaboration between Kim Koehler at SPAWAR PMW155, Charlotte Leigh of the EIS Department at APL/UW, Tony Eller of SAIC, and the APL/UW NPAL group. An online description of the effort and the surprisingly accurate results are presented here.