We used a range of approaches to gain an understanding of spatial and species-related patterns in herbivory in five distinct studies. Using underwater video cameras and Sargassum myriocystum assays, 23 different fish species were observed consuming macroalgae, but seven species (Naso unicornis, Kyphosus sp., K. vaigiensis, Siganus doliatus, Scarus ghobban, S. schlegeli and initial-phase Scarus sp.) together accounted for 95% of the observed bites across five regions. Of these species, three were identified as the most important in consuming macroalgae: N. unicornis, Kyphosus sp. and K. vaigiensis. These results were supported by stable isotope analyses that incorporate nutrients from food sources over far longer periods than those examined using the assay approach. We quantified spatial patterns of macroalgal consumption and food sources across a range of scales. Firstly, across reef habitats separated by hundreds of meters, herbivory was always greatest in the structurally complex coral-dominated outer reef and reef flat habitats, which are also characterised by the highest biomass of herbivorous fish. Secondly, we showed a high degree of variability in grazing rates among regions separated by 100s km in the marine park, with different species responsible for macroalgal removal among those regions. Either N. unicornis or Kyphosus spp. were responsible for the majority of the grazing. Thirdly, we showed variability in the importance of different food sources across both habitats and regions for some consumers (e.g. Siganus spp.) but consistency for other species (e.g. Naso unicornis, Kyphosus spp.), which is likely to reflect shifts in food source availability or feeding preferences. Lastly, we found strong transcontinental differences between Keppel Islands in the Great Barrier Reef (GBR) on the east coast of Australia and Ningaloo Reef in both the diversity of the species observed feeding and on the species composition of the roving herbivorous fish community. In Ningaloo Reef, 23 species were observed biting on Sargassum, compared with just 8 in the Keppel Islands. Sargassum consumption in the Keppel Islands was dominated by a small number of species and supports the identification of Naso unicornis as a key browser species. The larger number of species feeding on macroalgae in Ningaloo Reef suggests that there may be higher functional redundancy among macroalgal consumers in this system. We also characterised the benthic community dynamics of the reef-flat and lagoon habitats to identify seasonal patterns and we experimentally determined the importance of herbivory on algae recruitment in these two habitats. Differences among habitats in algal biomass were strongly influenced by season. Lagoon habitats only had higher biomass than reef-flat habitats during part of the year (late summer/ early autumn). Herbivory had an equally strong effect on the community composition of algal recruits in the lagoon and reef flat habitats, despite the reef flat hosting a herbivorous fish community that was an order of magnitude greater in terms of biomass than the lagoon, which is characterised by younger and smaller fish (e.g. Scarus initial phase).

Ecosystem impacts of human usage and the effectiveness of zoning for biodiversity conservation: - Measures of the effectiveness of previously established sanctuary zones for protecting exploited subtidal fish and invertebrate populations. - Measures of the effectiveness of previously established sanctuary zones for protecting exploited intertidal invertebrate populations - Assessment of whether there is consistent evidence for trophic cascade effects in previously established sanctuary zones - Experimental assessment of potential for indirect effects of fishing on lagoonal and shallow water ecosystems - Assessment of adequacy of sanctuaries for exploited species and related ecological effects - Baselines for future assessment of the importance of zone size, age, configuration, location on effectiveness for protecting biodiversity - Recommendations on species and methods to be monitored to assess management effectiveness (monitoring protocols, management indicators) for intertidal and subtidal communities

Ecosystem process measurements were made on sediment sampled with a box core and water sampled from CTD profiles. For sediments, measurements were of wet/dry weight ratio, total organic carbon, chlorophyll and stable isotopes, nitrogen cycling (nitrogen fixation and denitirification) and oxygen demand. For water samples, measurements were of total suspended solids, chlorophyll and stable isotopes. The survey design was a natural experiment (i.e. control-impact, but not before-after) to test the long-term, broadscale effects of prawn fishing on non-target species and habitats. Deployments were stratified by fishing intensity (three strata) and day/night. The survey was repeated in three regions: (a) north of Groote Eylandt, (b) north-east of the Vanderlins, and (c) north of Mornington Island.

The aims of this project are: (1) to identify and simulate key physical factors that have significant impacts on ecological processes on shelf and coastal areas of southwestern Western Australia (WA); and (2) to develop physical, ecological, and risk assessment models that can be used to assess impacts of multiple human use on coastal and shelf environments. The project consists of five main components: analysis of large scale climate forcing, development and application of regional and coastal circulation models, development of integrated biogeochemical/ecological models, development of coastal impact models, and risk assessment. This project will link existing field data, field observations from other SRFME projects, and output from new and existing models, with management objectives and needs defined by Western Australian stakeholders. Specific models to be developed include regional and coastal oceanographic models, biogeochemical / ecological models that links physical and ecological processes, and risk assessment models that link these models to human use of the marine environment. The project intends to build on methods and models already developed and/or used by other CMR projects such as the NWSJEMS and LWRDDC projects. These models will be adapted and extended to allow assessment of impacts of multiple use and natural forcing on nutrient cycling, production and habitat quality on shelf and coastal areas in southwestern WA. These tools will range from process-based simulation models to semi-empirical models, with a focus on making efficient use of existing data, and incorporating new data from large-scale observations such as acoustic and satellite data. The main deliverables of the project include analysis of large scale climate forcing, development and application of regional and coastal scale oceanographic, integrated biogeochemical/ecological, and coastal impact models and risk assessment methods.

The aim of the project is to examine how climate forcing influences nutrient, plankton and nekton dynamics across the continental shelf off Perth, Western Australia, with application to fisheries, management of marine protected areas and coastal processes. Objectives are: - Describe onshore-offshore biophysical ocean structure, its seasonal cycle and interannual variability based on remote sensing data and monitoring of temperature, salinity, nutrients, phytoplankton (chlorophyll and other pigments) and zooplankton based on monthly sampling off Hillarys; micronekton from quarterly cruises. - Measure primary productivity and parameters related to zooplankton grazing and productivity and food web links for key zooplankton species at selected stations/productivity regimes. - Use Hillarys transect temperature and chlorophyll data for validation of satellite derived surface temperatures and ocean colour (linkage with coastal project). - Apply acoustic methods to monitor zooplankton and micronekton and assess fine-scale distributions based on Tracor Acoustic Profiling System (TAPS) (6 frequencies, 300 kHz - 3 megaHz) and underway 38 and 120kHz frequencies; assess zooplankton and micronekton scattering strength and develop multi-frequency discrimination methods; - Explore optimised long term biophysical sampling strategies using remote sensing, acoustics and associated environmental variables for moored and other long-term monitoring linking process understanding with the numerical and empirical modelling. - Measure currents at a mooring site at 80 m depth off Perth and provide the data and oceanic process understanding for the numerical models. - Assess diet and nutritional status of western rock lobster phyllosoma with lipid biomarkers. - Input data and collaborate in development of biophysical NPZ models for coastal zone and shelf (modelling project); preliminary modelling of micronecton/necton.

This project has been designed to provide the Western Australian Government and its agencies with improved understanding of the coastal marine environment so that its decision making with regard to development in this zone is environmentally credible and sustainable. The project will deliver this result by the following sequence of research. First, existing data will be appraised in light of a simple system model for inshore coastal waters. This rudimentary understanding will be used to design environmental surveys for three representative coastal systems (chosen in consultation with WA departments and agencies). Whereupon, baseline data will be obtained on biodiversity, biogeochemical processes and environmental quality in these waterways. With this information, the research will then move on to consider the affects of selected stressors (localised sources and diffuse inputs) on the above ecological characteristics, and the potential for irreversible alteration. Where necessary, focussed investigations in the field or laboratory will be used to resolve key mechanisms and also the scale of response. Important outcomes for the project will be the development of validated environmental indicators for the use of coastal managers, and also other resources for them to better understand the complex interactions and inter-relations in coastal marine ecosystems (e.g. via conceptual models). This project will also work with other SRFME projects to improve capacity for prediction and scenario testing in environmental decision making via models and other tools.