In ecology, a common form of statistical analysis relates a biological variable to variables that delineate the physical environment, typically by fitting a regression model or one of its extensions. Unfortunately, the biological data and the physical data are frequently obtained from eparate sources of data. In such cases there is no guarantee that the biological and physical data are co-located and the regression model cannot be used. A common and pragmatic solution is to predict the physical variables at the locations of the biological variables and then to use the predictions as if they were observations.We show that this procedure can cause potentially misleading inferences and we use generalized linear models as an example. We propose a Berkson error model which overcomes the limitations. The differences between using predicted covariates and the Berkson error model are illustrated by using data from the marine environment, and a simulation study based on these data.

Climate change is threatening tropical reefs across the world, with most scientists agreeing that the current changes in climate conditions are occurring at a much faster rate than in the past and are potentially beyond the capacity of reefs to adapt and recover. Current research in tropical ecosystems focuses largely on corals and fishes, although other benthic marine invertebrates provide crucial services to reef systems, with roles in nutrient cycling, water quality regulation, and herbivory. We review available information on the effects of environmental conditions associated with climate change on noncoral tropical benthic invertebrates, including inferences from modern and fossil records. Increasing sea surface temperatures may decrease survivorship and increase the developmental rate, as well as alter the timing of gonad development, spawning, and food availability. Environmental changes associated with climate change are linked to larger ecological processes, including changes in larval dispersal and recruitment success, shifts in community structure and range extensions, and the establishment and spread of invasive species. Loss of some species will trigger economic losses and negative effects on ecosystem function. Our review is intended to create a framework with which to predict the vulnerability of benthic invertebrates to the stressors associated with climate change, as well as their adaptive capacity. We anticipate that this review will assist scientists, managers, and policy-makers to better develop and implement regional research and management strategies, based on observed and predicted changes in environmental conditions.

Marine organisms are exposed not only to natural environmental stressors, but also the additional effects of anthropogenic stressors, notably increasing temperatures and reduced pH. Early life stages of marine organisms have been recognised as potentially vulnerable to the stressors associated with climate change and ocean acidification, but identifying patterns across studies, species and a range of response variables is challenging. This study is supported by the Marine Biodiversity Hub through the National Environmental Research Program and identifies knowledge gaps in research on multiple abiotic stressors and early life stage (embryo to larvae), while quantifying interactions based on life history. Temperature was the most common stressor (91% of studies), while the most common combination of stressors was temperature and salinity (66%), followed by temperature and pH (17.5%). All studies were conducted in the laboratory although four studies also undertook field experiments. Synergistic interactions (68% of individual tests) were more common than additive (16%) or antagonistic (16%) interactions. The meta-analysis yielded several key results: 1) Embryos are not more vulnerable to stress than larvae in combined stressor treatments. 2) Sub-lethal responses are not more likely to be affected by stress than lethal responses. 3) Interaction types vary among stressors, phyla, ontogenetic stages, and biological responses. 4) Elevated temperature is generally a greater stressor than ocean acidification, but this depends on ontogenetic stage and phylum. 5) Ocean acidification is a greater stressor for calcifying than non-calcifying larvae. Our findings will assist in monitoring and predicting the health of marine populations and communities by identifying sensitive and robust taxa.

This study was undertaken as part of a program to collect baseline data from the seabed environments over the Van Diemen Rise, which comprises a series of carbonate platforms and banks in the eastern Joseph Bonaparte Gulf, northwest of Darwin. Samples were collected during a survey on board the RV Solander in August and September 2009. The state of saturation for the different carbonate minerals (aragonite, calcite, high-magnesium calcite) was calculated for each sampling site from the ratio of the ion activity product and the solubility product. The carbonate ion concentration used for the ion activity product was calculated from total alkalinity and pH. The solubility products of the carbonate minerals were derived from literature data, e.g. the solubility for high-magnesium calcite as a function of the mol% MgCO3 was based on experimental results by Plummer and Mackenzie (1974, American Journal of Science vol. 274, p. 61-83). The calculated average state of saturation was 1.4 (range: 0.8-1.9) for high-magnesium calcite, 4.2 (range: 3.4-4.6) for aragonite, and 6.4 (range: 5.1-6.8) for calcite. Values close to 1 suggest the mineral is in thermodynamic equilibrium with ambient water, which is the case for high-magnesium calcite. In contrast, aragonite and calcite are distinctively supersaturated. Given the near-equilibrium state of high-magnesium calcite, this mineral phase will likely be lost over a time scale of decades as ocean acidification progresses. This ongoing process will alter the sediment composition significantly given the high abundance of high-magnesium calcite. This study supports the concept of using high-magnesium calcite as an indicator for the progression of ocean acidification where surface sediments have been sampled and preserved over time.

This dataset contains species identifications of crinoids collected during survey SOL4934 (R.V. Solander, 27 August - 24 September, 2009). Animals were collected from the Joseph Bonaparte Gulf with a benthic sled. Specimens were lodged at Museum of Victoria on the 19 April 2010. Species-level identifications were undertaken by Kate Naughton at the Museum of Victoria and were delivered to Geoscience Australia in December 2010. See GA Record 2010/09 for further details on survey methods and specimen acquisition. Data is presented here exactly as delivered by the taxonomist, and Geoscience Australia is unable to verify the accuracy of the taxonomic identifications.

Australia is increasingly recognised as a global hotspot for sponge biodiversity, with sponges playing key roles in habitat provision, water quality, bioerosion, and biodiscovery. Despite the intense focus on marine resource management in northern Australia, there is a large knowledge gap about sponge communities in this region. This study focuses on shelf environments of the Timor Sea, in particular the Van Diemen Rise and Londonderry Rise which are characterised by extensive carbonate terraces, banks and reefs, separated by soft sediment plains and deeply incised valleys. These carbonate terraces and banks are recognised as a Key Ecological Feature (KEF) in the marine region plans for northern Australia (North and Northwest Marine Regions) and are in part incorporated into the Oceanic Shoals Commonwealth Marine Reserve. To support the management of this marine reserve and its associated KEF, we use new datasets to investigate regional patterns in sponge assemblages and their relationships to seabed geomorphology. To do this, we use sponge assemblage data and multibeam-derived variables (depth, backscatter, slope, geomorphic feature) from seven survey areas located on the Van Diemen Rise (four sites) and Londonderry Rise (three sites), spanning approximately 320 km in an east-west direction. The dataset was collected during three collaborative surveys undertaken in 2009, 2010 and 2012 by Geoscience Australia, the Australian Institute of Marine Science and the Museum and Art Gallery of the Northern Territory as part of the Australian Government's Offshore Energy Security Initiative and the National Environmental Research Program Marine Biodiversity Hub. All surveys returned geophysical, biological, geochemical, and sedimentological data. Benthic biota were collected with a benthic sled across a range of geomorphic features (bank, terrace, ridge, plain, valley) identified from high-resolution multibeam sonar. Sponges were then taxonomically identified to 350 species, with the species accumulation curve indicating there may be over 900 sponge species in the region. Sponge assemblages were different between the Van Diemen Rise and Londonderry Rise, as well as between individual banks in the same area, indicating that different suites of species occurred at regional (east-west) and local (between banks) scales. Relationships between sponges and other multibeam-derived variables are more complex and warrant further research. The current study will help: i) facilitate integrated marine management by providing a baseline species inventory; ii) support the listing of carbonate banks of the Timor Sea shelf as a Key Ecological Feature, and; iii) inform future monitoring of marine protected area performance, particularly for areas of complex seabed geomorphology.

The Antarctic continental slope spans the depths from the shelf break (usually between 500-1000 m) to ~3000 m, is very steep, overlain by 'warm' Circumpolar Deep Water and life there is poorly studied. This study investigates whether life on Antarctica's continental slope is essentially an extension of the shelf or the deep-sea fauna, a transition zone between these or clearly distinct in its own right. Using data from several cruises to the Weddell and Scotia sea, including the ANDEEP (ANtarctic benthic DEEP-sea biodiversity, colonisation history and recent community patterns) I-III and BIOPEARL (BIOdiversity, Phylogeny, Evolution and Adaptive Radiation of Life in Antarctica) 1 and EASIZ II cruises as well as current data bases (SOMBASE, SCAR-MarBIN), we selected four different taxa (i.e. cheilostome bryozoans, isopod and ostracod crustaceans, and echinoid echinoderms) and two areas, the Weddell and the Scotia Sea, to examine faunal composition, richness and affinities. The answer has important ramifications to the link between physical oceanography and ecology, and the potential of the slope to act as a refuge and resupply zone to the shelf during glaciations (and therefore support or not glaciological reconstructions of ice sheets covering continental shelves).

Acorn worms (Enteropneusta), which were previously thought to be a missing link in understanding the evolution of chordates, are an unusual and potentially important component of many deep-sea benthic environments, particularly for nutrient cycling. However very little is known about their distribution, abundance, or behaviour in deep-sea environments around the world, and almost nothing is known about their distribution within Australian waters. In this study, we take advantage of two large-scale deep-sea mapping surveys along the eastern and western continental margins of Australia to quantify the distribution, abundance and trail-forming behaviour of this highly unusual taxon. This is the first study to quantify the abundance and trail behaviour of acorn worms within Australian waters and provides the first evidence of strong depth-related distributions.

The Marine Biodiversity Hub was funded by the Australian Government Commonwealth Environmental Research Facilities (CERF) between 2007 and 2010. The Hub was developed to improve the scientific knowledge available to support marine bioregional planning and addressed two fundamental questions: 1. How can we predict the distribution of marine biodiversity; and 2. How can we use this improved capability to conserve and manage biodiversity in a multiple-use environment? This talk focuses on the Surrogates Program, one of four research programs in the Hub. The Surrogates Program addressed the above questions by testing and developing physical variables as surrogates of marine biodiversity, with a focus on seabed environments. In the program, we employed a range of marine survey technologies to collect high-quality and co-located benthic physical and biological data at four selected areas in temperate and tropical waters. We also developed advanced spatial and statistical approaches to test the degree of covariance between the physical and biological data, identify ecological processes, and generate prediction maps. During a number of field campaigns, we deployed a range of instruments to collect data including multibeam sonar, sediment grabs, benthic sleds, towed-video/still images and Autonomous Underwater Vehicles. GIS, machine-learning models and the SWAN hydrodynamic model were used to derive and predict a large number of physical variables as potential surrogates. The effectiveness of the surrogacy approaches were examined using multivariate analyses and spatial modelling techniques. In general, we found that using physical surrogates to predict marine biodiversity is a cost-effective approach. The new knowledge of surrogates and seabed ecological processes directly supports the management of the Australian marine estate. Other major outputs of the Surrogates Program include: - Thirty-seven new and updated national-scale marine physical environmental datasets; - High resolution bathymetry of targeted areas, covering almost 2000 km2, plus 171 km of underwater video transects, 402 sediment grab samples and 232 epifauna samples; - New seabed exposure and fetch models/datasets; and - Peer-reviewed reports and papers in scientific journals. The success of the Marine Biodiversity Hub has enabled the Hub to be refunded for a further four years through the new National Environmental Research Program. In this, Geoscience Australia (GA) is collaborating with the University of Tasmania, CSIRO Marine & Atmospheric Research, Australian Institute of Marine Science, Museum of Victoria, University of Western Australia and Charles Darwin University; GA is also leading Theme 3 Project 1 which focuses on identifying the functions and processes of shelf and canyon ecosystems. The project is expected to further advance marine biodiversity research in Australia by investigating the role of large-scale physical features on the shelf in influencing patterns of marine biodiversity.

Much of the deep sea comprises soft-sediment habitats dominated by low abundances of small infauna, and traditional methods of biological sampling may therefore fail to sufficiently quantify biodiversity. During feeding and burrowing, many deep sea animals bioturbate the sediment, leaving signs of their activities called lebensspuren ('life traces'). In this study, we use three criteria to assess whether the quantification of lebensspuren from high resolution still images is an appropriate technique to broadly quantify biological activity in the deep sea: 1) The ability to differentiate biological assemblages between geographic regions; 2) the ability to reveal known biological patterns across environmental gradients; and 3) correlation with other methods of biological characterisation often used in the deep sea (e.g. video). Lebensspuren were quantified using a univariate measure of track richness and a multivariate measure of lebensspuren assemblages from the eastern (1712 images, 13 stations) and western (949 images, 11 stations) Australian margins. A total of 46 lebensspuren types were identified, including those matching named trace fossils. Assemblages were significantly different between the two regions, with five lebensspuren types accounting for over 95% of the differentiation (ovoid pinnate trace, crater row, spider feature, matchstick feature, mesh feature). Track richness in the combined margins dataset was correlated to depth, chlorin index (i.e. organic freshness), and possibly mud, although the strength of the relationships varied according to the dataset used. There was no relationship to total organic carbon. Lebensspuren richness from still images was significantly related to lebensspuren from video but not to occurrence of epifauna. Based on these results, the quantification of lebensspuren from still images seems an appropriate measure to broadly characterise biological activity in deep sea soft sediment ecosystems.