Report of the Round Table Session

Blondel, P.1*, Sagen, H.2*, Martin, B.3, Pettit, E.C.4, Tegowski, J.5, Thodes, A.6, Tollefsen, D.7 and Worcester, P.8


1 University of Bath, UK
2 Nansen Environmental and Remote Sensing Center, Norway
3 JASCO Applied Sciences, Halifax NS, Canada
4 Dept. of Geology and Geophysics, University of Alaska Fairbanks, USA
5 Institute of Oceanography, University of Gdansk, Poland
6 Scripps Institution of Oceanography, USA
7 Norwegian Defence Research Establishment (FFI), Horten, Norway
8 Institute of Geophysics and Planetary Physics, Scripps Institution of Oceanography, USA

* Session Chairs and Corresponding Authors; E-mail: pyspb@bath.ac.ukhanne.sagen@nersc.no


This report can be referenced as:

Blondel, P., Sagen, H., Martin, B., Pettit, E.C., Tegowski, J., Thodes, A., Tollefsen, D. and Worcester, P. (2015). Report of the Polar Session, oceanoise2017, Vilanova i la Geltrú, Barcelona, Spain, 10-15 May. (Editors Michel André & Peter Sigray). Retrieved from https://2023.oceanoise.com


Polar Noise


The Polar Regions are very different but face similar challenges in terms of ambient noise, its impacts on marine life and how it will evolve with climate change and increased anthropogenic activities.


Growing scientific and societal concerns about the effects of underwater sound on marine ecosystems is now recognised through several international initiatives aiming at measuring the environmental impact of ocean noise at large spatial and temporal scales. This pressure and these concerns are particularly acute in the Polar Regions (Arctic and Antarctic), where climate change and increased human presence add new variables to little studied, and complex environments.

Sea ice reduction is facilitating resource exploration, marine transport and other economic activities (e.g. fishing) in these regions, adding to ambient noise. In the last decade, there has also been significant growth in offshore oil and gas exploration in several Arctic regions. An assessment made by the US Geological Survey estimates that 30% of the world’s undiscovered gas and 13% of the undiscovered oil are located in the Arctic (Klett and Gautier et al., 2009).Similarly, the Barents Sea is the most important fishing ground in Europe, and a recent report by Lloyds (Lloyd’s, 2012) suggest that over the coming decade the Arctic is likely to attract substantial investment potentially reaching $100bn or more. Increased exploitation of marine resources has been shown worldwide to increase ambient noise levels, and regulations have been put in place to address these concerns, from seismic industry to the 2020 goal of Good Environmental Status enshrined in the European Marine Strategy Framework Directive. The effects of climate change are also visible in the melting of glaciers, raising sea levels and increasing the amount of fresh water in fragile ecosystems, and the observation of changes in marine life, with species moving poleward as waters warm.

Several actors, academic and commercial, are collecting passive acoustic data in the Arctic and in the Antarctic, using local deployments, long-term moorings and observatories, ships and autonomous vehicles. But the different activities are not coordinated and hampered by the low level of dissemination of results to the different communities, the sharing of equipment in difficult-to-access regions and the exchange of good practice in very challenging and dangerous environments. There is a clear need for better sharing of knowledge of the current noise status in Polar Regions.


Following their presentations, the speakers participated in a lively panel discussion with input from the audience around key issues identified by the session organisers and informally presented as:

Theme 1: What is “the vision”?

What are people measuring and why? What are the issues?

What else could we do?

Theme 2: Filling the gaps

What are the knowledge gaps? The technology gaps?

Metrics? Models? Collaborations? Logistics? Regulations? Timeframe? Priorities?

Theme 3: Measuring – comparing

Data collection (e.g. NPL Good Practice Guide, Robinson et al., 2014)

Data processing (e.g. Merchant et al., Methods in Ecology and Evolution, 2015)

Data policies (sharing/recognising …)

Archiving (long-term)

Communication (web portal?)

The last 30 years have seen a huge increase in the number of expeditions and surveys, with the collection of new datasets encompassing a wide range of processes and frequencies. Overall, the Arctic is extremely quiet up to 100 Hz compared to other latitudes. In some areas, the lack of sound at certain frequencies is used by marine life for “acoustic niches” (e.g. van Opzeeland and Miksis-Olds, 2012), and the introduction of additional noise must be carefully monitored. One presenter discussed how seismic exploration did not seem to affect or block whale migrations, but definitely changed their vocal behaviour and increased it. In Arctic conditions, seismic activity is apparently audible 10% of the time, and this observation prompted observations from the floor that it might be much lower than in other parts of the world. One of the presentations showed sounds from an icebreaker decreasing linearly with distance but still audible 1,000 km away, then decreasing rapidly close to the ice edge. Natural ambient noise levels near continental shelves are well predicted by wind data, with a good correlation up to 50 Hz, not seen in other oceans. This correlation decreases with shipping and other human activities.

Ice cover is important, with younger and thinner ice improving propagation conditions at lower frequencies (when underwater ridges and roughness are more conducive to propagation). Melting affects higher frequencies, in particular near glaciers or in fjord environments. Sea ice generally gets younger, as shown in other studies (e.g. Wadhams, 2000): water below multi-year ice is quieter, with noise increasing as one moves toward its edges where weather and waves are directly audible. Thermal noise, fracturing and bubbles escaping from young and breaking sea ice and from freshwater glacier ice blocks all contribute to the noise budget, with transients at different levels and specific frequency bands. Interaction with the sea floor and the sea surface is by far the most important factor describing the attenuation of acoustic energy. How the acoustic signal is interacting with the sea floor and the sea surface (open water or sea ice) is determined by the sound speed profile. The interior Arctic is described by a typical Arctic profile with a strong 100-200 meter deep sound speed profile beneath the sea ice, causing rays near the horizontal to become trapped and continuously interact with the underside of the sea ice. Below the lower limit of the duct, a steadily increasing sound speed causes the steeper rays to be upward refracted.

Theme 1 (“what are people measuring and why”) shows there are different approaches to data collection, with a large part associated to exploration (discovering what is there, with increasing spatial and temporal resolution), and understanding baseline sound budgets in times of increasing human activities. A pervading theme of all articles is the need for long-term measurements, and the perception that there is still much to be learnt. Long-term datasets are necessary to understand variations, either because of the complexity and seasonal variability of what we study, or to identify how climate change is affecting the different sources, echoing recent statements by the European Polar Board and the European Marine Board (2013). All presenters also agreed that the effects of noise will vary dramatically between regions. So these long-term datasets have to be acquired in distinct places. We need to identify where best to gather these measurements. Is it better to go to places already studied, to add to the existing body of data, or to be more process-driven and select first what needs to be actually measured?

The European Marine Board (2013) states that: “A key requirement for both industry and science will be the collection of long-term environmental and biodiversity time-series data and improved modelling capabilities to enhance predictive capacities. This will require a significant investment to upgrade and expand the existing observing infrastructure, to support research on modelling and forecasting, and a major improvement in the sharing of data and information.” Is it possible to detect a trend between earlier and later studies? Comparisons in localised areas (e.g. Fram Strait) since the 1980’s show the link with the exploitation of resources, but there is no real handle on how human activities have changed the environment. Many changes are hard to assess because of the variations in propagation conditions, especially in the presence of ice at the sea surface. Bathymetry and knowledge of seabed geology are often not detailed enough for modelling of propagation, in particular in glacier-terminated fjords whose seabeds are rapidly evolving as glaciers retreat, but also in shallow-water regions where few measurements have been made, if ever. This was perceived as a significant knowledge gap, starting to answer questions in Theme 2 (“Filling the gaps”).

Discussion around Theme 3 (“Measuring –Comparing”) showed that comparison between distinct datasets (e.g. MIZEX ’84 and MIZEX’87 with recent data) cannot identify trends because older data is often only accessible from reports, not always with the full information (e.g. missing units or reference levels) or it was stored on old magnetic tapes, hard to work with. This was echoed by several panel members, with the feeling that there might be more datasets “out there” that no one knows about (e.g. from ice stations, UNCLOS seismic surveys looking for the edge of continental shelves, year-round monitoring). The use of common standards for calibrating hydrophones and setting up surveys is now emerging, e.g. with the recent Good Practice Guide of Robinson et al. (2014) and current efforts at international standard levels (e.g. British Standards, European standards, ISO). This is echoed by the recommendation for using rigorous metrics and associated data reporting (e.g. Merchant et al., Methods in Ecology and Evolution, 2015), with shared open-access software from individual articles (like the previous) or from community repositories, e.g. the Open Acoustics Library, which makes available benchmarked propagation models.

Long-term archiving is still an issue, along with the sharing of information in formats readable in the long term and with enough information to support future studies and/or reanalyses of older datasets. Data access policies vary between countries and funding agencies, although the current move to open-access increasingly allows recognition of datasets in publications and reports. Communication is also seen as a recurrent knowledge gap, with many datasets potentially being under-recognised or under-used, and no central portal to combine survey efforts in polar regions, for example to recover equipment left over long periods, or to set up moorings and oceanographic observations in places suitable for data exchange with other groups. Post-panel discussions and new collaborations initiated at Oceanoise’2015 enabled some researchers to share data, combine forces for surveys or to recover equipment (e.g. in the Antarctic) and be more aware of ambient noise research in the different regions and within distinct domains of application.


The wealth of scientific discussions made possible by Oceanoise’2015 are but a first step, and along with other conferences and collaborations, they aim at improving polar noise research and maximising its potential in times of climate change and increasing human pressure. Key points likely to be discussed at Oceanoise’2017 will include:

  • Measurements: what else have we learnt in the last years? How do the polar oceans adapt to warming and (possibly) acidification? How do the acoustic emissions from glaciers correspond to melting/calving processes? How do ecosystems adapt/change with general variations? How do the Antarctic soundscapes compare to the Arctic ones? Are long-term deployments possible, despite the general funding restrictions, and what do they tell us? What are the benefits of combining instruments/surveys in similar areas?
  • Modelling: propagation models in open water are well validated but often hampered by the poorer resolution of bathymetry and geology datasets, especially in shallow waters and in fjords with retreating glaciers. The recent, detailed observations of noise emitted by sea ice and melting glacier ice, including sub-glacial discharge, will be expanded in the years to come and this might lead to models of their acoustics.
  • Data access: how can long-term archival (and open access) be encouraged?


Readers are encouraged to also look at articles/reports published by each presenter to the Polar Noise session.