Monitoring changes in the spatial distribution and health of biotic habitats requires spatially extensive surveys repeated through time. Although a number of habitat distribution mapping methods have been successful in clear, shallow-water coastal environments (e.g. aerial photography and Landsat imagery) and deeper (e.g. multibeam and sidescan sonar) marine environments, these methods fail in highly turbid and shallow environments such as many estuarine ecosystems. To map, model and predict key biotic habitats (seagrasses, green and red macroalgae, polychaete mounds [Ficopamatus enigmaticus] and mussel clumps [Mytilus edulis]) across a range of open and closed estuarine systems on the south-west coast of Western Australia, we integrated post-processed underwater video data with interpolated physical and spatial variables using Random Forest models. Predictive models and associated standard deviation maps were developed from fine-scale habitat cover data. Models performed well for spatial predictions of benthic habitats, with 79-90% of variation explained by depth, latitude, longitude and water quality parameters. The results of this study refine existing baseline maps of estuarine habitats and highlight the importance of biophysical processes driving plant and invertebrate species distribution within estuarine ecosystems. This study also shows that machine-learning techniques, now commonly used in terrestrial systems, also have important applications in coastal marine ecosystems. When applied to video data, these techniques provide a valuable approach to mapping and managing ecosystems that are too turbid for optical methods or too shallow for acoustic methods.

In early 2014 the RVIB Nathaniel B. Palmer conducted the first ever multidisciplinary study of the Sabrina Coast continental shelf. This area is remote and generally inaccessible, but biological significance is recognised by its initial inclusion within the proposed East Antarctic representative system of Marine Protected Areas. The datasets collected during this voyage allow analysis of the physical habitat parameters and benthic biota through interpretation of bottom camera images, high resolution multibeam bathymetry, sediment properties and oceanographic measurements, with satellite observations of sea ice also providing important environmental context. The suite of environmental and biological datasets provides evidence for a diverse, relatively high biomass continental shelf community that is strongly structured by the physical environment. The distribution of benthic taxa is most closely related ( = 0.592) to seafloor bathymetry, substrate type, latitude and the occurrence of phytodetritus. Phytodetritus accumulation is associated with muddy/sandy substrates, indicating long term sediment focussing in these areas, consistent with evidence of bottom recirculation features. These softer substrates contain relatively high abundances of mobile holothurians and amphipods. Scattered occurrence of dropstones creates habitat heterogeneity at fine-scales. Harder substrates have high abundances of brachiopods, bryozoans, polychaete tubeworms, a range of massive and encrusting sponges and sea whips. Several taxa are found only on areas of hard substrate, yet have a broad distribution across the sites, indicating that the density of dropstones is sufficient for most sessile invertebrates to disperse across the region. The occurrence of dropstones is associated with significant increases in taxa diversity, abundance and percent biological cover, enhancing the overall diversity and biomass of this ecosystem. This study illustrates how multidisciplinary studies can inform understanding of the drivers of benthic ecosystems, providing important constraints for generating realistic ecosystem models and contributing to our understanding of the sensitivity of this community to environmental change.

Identifying the spatial distribution and extent of biotic habitats in estuaries is important in managing our natural resources. Although a number of remote-sensing habitat distribution mapping methods have been successful in clear, shallow-water coastal environments and deeper marine environments, these methods fail in highly turbid and shallow estuarine environments. In this study, we used underwater video to record percent cover of biotic habitats of seagrasses, green and red macroalgae, polychaete mounds and mussel clumps in five estuaries on the south-west coast of Western Australia. Random Forest (RF) was used to model, predict and map their habitats across the estuaries based on the video data and spatially continuous environmental variables. RF is an ensemble, machine-learning method that is robust and able to handle non-linear relationships frequently encountered with ecological data. RF is often used within terrestrial ecosystems but less so in marine ecosystems (e.g. predicting seabed sediment and hardness for marine habitats). The results of this study show that RF models performed well for predictions of biotic habitats, with 79-90% of variation explained by depth, latitude, longitude and water quality variables. This study confirms that RF also has important applications in estuaries, with results refining existing baseline maps and highlighting the importance of biophysical processes driving plant and invertebrate species distribution. Underwater video and RF modelling offer a valuable approach to mapping the spatial distribution of highly turbid biotic habitats.

Geoscience Australia undertook a marine survey of the Vlaming Sub-basin in March and April 2012 to provide seabed and shallow geological information to support an assessment of the CO2 storage potential of this sedimentary basin. The survey was undertaken under the Australian Government's National CO2 Infrastructure Plan (NCIP) to help identify sites suitable for the long term storage of CO2 within reasonable distances of major sources of CO2 emissions. The Vlaming Sub-basin is located offshore from Perth, Western Australia, and was previously identified by the Carbon Storage Taskforce (2009) as potentially highly suitable for CO2 storage. The principal aim of the Vlaming Sub-basin marine survey (GA survey number GA334) was to look for evidence of any past or current gas or fluid seepage at the seabed, and to determine whether these features are related to structures (e.g. faults) in the Vlaming Sub-basin that may extend up to the seabed. The survey also mapped seabed habitats and biota in the areas of interest to provide information on communities and biophysical features that may be associated with seepage. This research addresses key questions on the potential for containment of CO2 in the Early Cretaceous Gage Sandstone (the basin's proposed CO2 storage unit) and the regional integrity of the South Perth Shale (the seal unit that overlies the Gage Sandstone). This dataset comprises %carbonate and specific surface area of seabed sediments.

To date, a range of methods have been developed and applied to the processing and analysis of underwater video and imagery, in part driven by different requirements. For example, in Australia, the marine science community who are partnered by the National Environmental Research Program (NERP) and funded by the Marine Biodiversity Hub, has developed a national CATAMI (Collaborative and Automated Tools for Analysis of Marine Imagery and video) scheme. Technological advances in recent years have improved the usability and output quality of underwater video and still images used to identify and monitor underwater habitats and structures and as a result, these techniques are more frequently applied to marine studies. So far, a comprehensive review of underwater video and still imagery processing/analysis methods has not been completed, although the number of studies utilising underwater stills and video has increased dramatically. Difficulties in diver limitation and stringent regulations applied to the collection of diver-based imagery and video data from underwater benthic habitats. Therefore, remote sensing methods such as underwater video and still imagery are becoming increasingly pivotal for ground-truthing benthic biological and physical habitats in shallow and deep marine and freshwater habitats and are also providing a permanent archive for future analyses. This review focuses on post-processing observational methods used for underwater video and still image habitat classification and quantification. We summarise the main applications, advantages and disadvantages of video and still imagery scoring methods, and illustrate recent advances in this topic.

This paper presents an application of a spatially-balanced survey design using baited videos in one reserve in Australia's recently declared network of reserves to: (i) identify the types of demersal fish communities present, (ii) quantify their relationship with environmental gradients and, (iii) describe their spatial distribution across the shelf of the reserve as an essential first step towards developing a monitoring program. Baited videos recorded a range of species from multiple trophic levels, including species of commercial and recreational interest. The majority of species, whilst found commonly along the southern or south-eastern coasts of Australia, were endemic to Australia, highlighting the global significance of this region. We identified six distinct fish assemblages on the shelf of the reserve. The distribution of these assemblages was related to environmental variables, the most important being depth and habitat type. Assemblages could be broadly categorised as either a combination of reef or sand associated and shallow or deep. Species richness was greater on habitats containing some reef and declined with increasing depth. The trophic breath of species in assemblages was also greater in shallow waters. We further discuss how our results can be used to inform future surveys and monitoring within this reserve as well as comment more generally on the utility of our approach for establishing inventories and baselines when little prior knowledge is available.

<p>This dataset contains identifications of polychaetes collected during surveys on the RV Solander in northern Australia: SOL4934 (27 August-24 September, 2009) , SOL5117 (30 July-27 August, 2010), SOL5463 (3-31 May 2012), and SOL5650 (12 September-6 October 2012). Sediment was collected with a Smith McIntyre grab or boxcore and elutriated over a 500um sieve. Elutriated material was then sorted back at Geoscience Australia, and all polychaetes were removed and sent to the Museum and Art Gallery of the Northern Territory (MAGNT). Chris Glasby and Charlotte Watson taxonomically identified each animal to the lowest possible level, differentiating species using established names and operational taxonomic units (OTUs). Specimens are lodged at the MAGNT. See relevant post-survey reports (GA Records 2010/09, 2011/08, 2012/66 and 2013/38) for further details on survey methods and specimen acquisition. This dataset is published with the permission of the CEO, Geoscience Australia. <p>This research is supported by the National Environmental Science Program (NESP) Marine Biodiversity Hub through Project D1.

This video is a compilation of three short data visualisations showing sea floor terrain in the MH370 search area. The visualisations were developed from a sample of bathymetric data collected as part of the search for missing Malaysia Airlines flight MH370.

This is a collection of multimedia products of data visualisations showing sea floor terrain in the MH370 search area and associated public communications regarding the search and Phase 1 data release. The video flythroughs and data visualisations were developed from a sample of bathymetric data from Phase 1 collected as part of the search for missing Malaysia Airlines flight MH370. This includes all visualisations (images and videos) used in the 'Data behind the search for MH370' Story Map. This also contains the translated versions of the MH370 Story Map on ESRIOnline ("the Data behind the search for MH370") into Malay and Chinese.

An integrated analysis of geoscience information and benthos data has been used to identify benthic biotopes (seafloor habitats and associated communities) in the nearshore marine environment of the Vestfold Hills, East Antarctica. High-resolution bathymetry and backscatter data were collected over 42km2 to depths of 215 m using a multibeam sonar system. Epibenthic community data and in situ observations of seafloor morphology, substrate composition and bedforms were obtained from towed underwater video. Analysis of the datasets was used to identify statistically distinct benthic assemblages and describe the physical habitat characteristics related to each assemblage, with seven discrete biotopes identified. The biotopes include a range of habitat types including shallow coastal embayments and rocky outcrops which are dominated by dense macroalgae communities, and deep muddy basins which are dominated by mixed invertebrate communities. Transition zones comprising steep slopes provide habitat for sessile invertebrate communities. Areas of flat sandy plains are relatively barren. The relationship between benthic community composition and environmental parameters is complex with many variables (e.g. depth, substrate type, longitude, latitude and slope) contributing to differences in community composition. Depth and substrate type were identified as the main drivers of benthic community composition, however, depth is likely a proxy for other unmeasured depth-dependent parameters such as light availability, frequency of disturbance by ice, currents and/or food availability. Sea ice cover is also an important driver and the benthic community in areas of extended sea ice cover is comprised of sessile invertebrates and devoid of macroalgae. This is the first study that has used an integrated sampling approach based on multibeam sonar and towed underwater video to investigate benthic assemblages across a range of habitats in a nearshore marine environment in East Antarctica. This study demonstrates the efficacy of using multibeam sonar and towed video systems to survey large areas of the seafloor and to collect non-destructive high-resolution data in the sensitive Antarctic marine environment. The multibeam data provide a physical framework for understanding benthic habitats and the distribution of benthic communities. This research provides a baseline for assessing environmental change across the nearshore marine environment. This research provides a baseline for assessing natural variability and human induced change on nearshore marine benthic communities (Australian Antarctic Science Project AAS-2201), contributes to Geoscience Australia's Marine Environmental Baseline Program, and supports Australian Government objectives to manage and protect the Antarctic marine environment.