Abstract
The variation in the ambient sound levels in the deep ocean has been the subject of a number of recent studies, with particular interest in the identification of long-term trends. This paper describes a statistical method for performing long-term trend analysis and uncertainty evaluation of the estimated trends from deep-ocean noise data.Measurements of underwater ambient noise have been carried out since at least the 1960s. Most of the studies demonstrating an increase in the levels of low frequency sound in the deep-ocean have been undertaken in the Pacific Ocean. In part, the observed increasing trend has been attributed to increases in noise produced by shipping, but it is recognised that there is a variety of sound sources which contribute to the ambient sound field, both man-made and natural. The paucity of available data over the last 50 years has meant that attempts to determine trends have often been based on very few data points and relied on simple statistical techniques such as straight-line fits. In more recent studies covering the last 15 years, use has been made of much richer data sets where continuous monitoring has been undertaken. The measured data used here originate from the Southern Ocean and span up to a maximum of 15 years, from 2003 to 2018. The data were obtained from the hydro-acoustic monitoring stations of the Preparatory Commission for the Comprehensive Nuclear Test Ban Treaty Organization (CTBTO). The monitoring stations provide information at acoustic frequencies up to 105 Hz.The analysis method uses a flexible discrete model that incorporates terms that capture seasonal variations in the data together with a moving-average statistical model to describe the serial correlation of residual deviations, with uncertainties validated using bootstrap resampling. The main features of the approach used include (a) using monthly or daily aggregation intervals derived from 1 minute SPL averages, (b) using a model that includes terms to represent explicitly seasonal behaviour and that captures the serial correlation of the data-model differences, and (c) applying a non-parametric approach to validate the uncertainties of trend estimates that avoids the need to make an assumption about the distribution of those differences. Computation of the uncertainty associated with a parameter estimate is a fundamental requirement for statistical inference as it provides a degree of confidence in the precision of that estimate. The trend analysis is applied to time series representing monthly and daily aggregated statistical levels for five frequency bands to obtain estimates for the change in sound pressure level over the examined period with associated coverage intervals.For the CTBTO station at Cape Leeuwin, located off the southwest shore of Australia, the results showed that statistically significant reductions in SPL are observed for all statistical percentiles for the different frequency bands. Additionally, the relative differences between the trends for various percentiles were found to be remarkably similar for all frequency bands with higher percentiles following steeper trends than lower percentiles leading to a significant reduction of the dynamic range in the recorded noise. Also, it was shown that trends in the data are dominated by the trends at low frequencies. Strong seasonal variation is also observed, with a high degree of correlation with climatic factors such as sea surface temperature, Antarctic ice coverage and wind speed. Some possible explanations for the observed changes are postulated. Finally, for comparison, the results for the CTBTO station at Wake Island, located in the central Pacific Ocean, are presented for comparison.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.