December 11, 2013

Newsletter: 

Acoustics scientists are sharpening their focus on the sea floor, with new tools and techniques, a flurry of swath mapping, and a mission to frame international standards for consistency and ‘best practice’.

Several Australian research groups – including CSIRO, Geoscience Australia, Deakin University and the University of Sydney – are routinely employing acoustic instruments capable of collecting coastal and shelf bathymetry at metre (and sub-metre) resolution.

Scientists from Geoscience Australia, the University of Tasmania and CSIRO, have been working through the NERP Marine Biodiversity Hub to acquire new swath coverage of the Flinders, Freycinet and Oceanic Shoals Commonwealth Marine Reserves, and waters surrounding the Abrolhos and Solitary Islands (areas identified by the Department of the Environment as Key Ecological Features).

Acoustic information is being processed and interpreted at fine scale resolution on the shelf (one metre) and medium-scale resolution on the slope (20 m). The data is in much greater detail than previously used for marine bioregional planning which was as coarse as 250 metres. The data will be used to produce national maps that will reveal the structure of canyons and slopes, and offer clues to associated biodiversity.

‘The improved resolution of the multibeam acoustic data helps us identify finer scale features that might be resolved at a resolution that the biota respond to,” Marine Hub scientist Vanessa Lucieer of UTAS said. ‘It also reveals complexities in the seafloor such as sand waves, ripples and ledges that would have been “invisible” at coarser resolutions.’

Working with the Marine Hub team, Vanessa has compared habitat classes classified acoustically with those classified ‘visually’ from video data. Her findings highlight the importance of matching the appropriate validation technique to the sampling method.

‘To be able to use seabed acoustic mapping techniques for monitoring, we need to be able to accurately validate the seabed classifications we generate from the multibeam echo sounder (MBES) data,’ Vanessa said. ‘But what is an appropriate validation technique?’

‘For example, in the Flinders Commonwealth Marine Reserve, we found that due to the subtle transitional boundaries of some habitats such as sediment veneer (soft sand over the top of hard reef where biota such as sponges are supported), video is not the best validation method for the acoustic maps.’

Sounding out standard practice

The Marine Hub is also working to help provide national and international consistency in the collection and interpretation of acoustic data.

At a national workshop help at Hobart in June 2013, 80 scientists considered statistical and logistical challenges and research synergies associated with processing of MBES data and automated video analysis for fisheries and environmental monitoring.

Vanessa Lucieer is one of a several Australian scientists involved in an international backscatter working group that is developing procedural guidelines for multibeam acquisition and processing. She said the standardisation of the acquisition and processing of sonar backscatter data provided by MBES would help scientists around the world adopt this technology for applications including mapping for discovery and seabed monitoring.

‘Internationally, there is a growing interest in generating greater consistency, or at least a better understanding of best practices,’ Vanessa said. ‘The Australian acoustic research community is relatively small but we have the potential to make a unified contribution.’

‘Australia has one of the largest marine estates in the world. Consistency in data processing and analysis will ensure acoustic data can be combined to create one map of the seabeds around the continent.’

 


Further reading

 


Images

A. Location map of the Tasman Peninsula study area on the south east coast of Tasmania, Australia.
B. Extent of multibeam data collected between High Yellow Bluff and the Hippolyte Rocks. Black dots indicate AUV transects. Acoustic data shows relative backscatter intensity (dB).
C. Zoomed in section of O’Hara Bluff showing configuration of the AUV sampling design over the reef area (from ~20 m to ~60 m water
depth)

 Vanessa Lucieer, UTAS

 

Sediment veneer substrates were a dominant habitat in the Flinders Commonwealth Marine Reserve. These were characterised as smooth sandstone reef systems covered with a thin layer of sand. Without the presence of sponges and a bryzoan crust, it would look like sand habitat, but hard substrate could be found 1-2 mm under the surface. This example is from Patch 6 on the northern boundary of the MPA in water depths of 42-44 m.

  Vanessa Lucieer, UTAS


Contact:

Dr Vanessa Lucieer, University of Tasmania