Commonwealth of Australia, 2005
- Torres Strait seabed and water-column data collation, bio-physical modeling and characterization - full report (PDF - 8.94 MB)
- Title, Contents, Executive summary, Introduction (PDF - 1 MB)
- Methods and results: 2.1 Collation of available datasets, 2.2 Oceanographic data (PDF - 1.91 MB)
- Methods and results: 2.3 Physical covariate data (PDF - 1.58 MB)
- Methods and results: 2.4 Biological survey data, 2.5 Acoustic survey data (PDF - 2 MB)
- Methods and results: 2.6 Spatial auto-correlation, 2.7. Statistical characterisation & bio-physical modeling, 2.8. Stratification and sampling design (PDF - 3.04 MB)
- Discussion, References, Project staff (PDF - 246 KB)
About the report
The main aims of this project were to compile and assess the current state of knowledge of the broadscale seabed and water-column ecosystem of Torres Strait, provide a preliminary characterisation of the region for use by the National Oceans Office (NOO), and design a sampling strategy for the Ecosystem Mapping Task of the Cooperative Research Centre for the Torres Strait. The project successfully collated and assessed relevant seabed & water-column data-sets, examined relationships between the biological and physical data, developed a bio-physical stratification of the seabed for the Torres Strait region, and completed a sampling design for the mapping project in time for the first field survey in January 2003. The major beneficiaries of the information include the National Oceans Office (NOO) and the Torres Strait people, and the Australian Fisheries Management Authority (AFMA) and the Torres Strait fishing industries. Funding was provided by the NOO and CSIRO.
Significant information on the physical environment was available from existing data. The Project collated 17 major datasets of physical and biological data for the region. Available relevant information included: physical environment (bathymetry, sediment grain-size and composition, water attributes & chemistry, ocean colour); basic seabed habitats; seagrass and algae; and some trawl samples. After checking quality and redundancy among sources, 32 physical variables were identified and mapped as potentially useful for modelling & stratification. A 0.01 degree resolution (~1.1 km) grid was established for analyses and sampling design, and the physical variables were re-sampled to this grid and mapped (interpolated where required), to provide a consistent set of full-coverage covariates at ~45,000 grid cells for the Project. The biological data was sourced from multiple legacy projects, each with different objectives, and was reconciled to useable common-denominator formats.
The broad-scale physical factors important in structuring patterns in the biological data were identified. Seabed current stress was the most important variable, and others included: chlorophyll, turbidity, oxygen, salinity, nutrients, sediment grain size, and depth. These bio-physical relationships were used to predict and map the categorical biological data to the whole Torres Strait region, with an estimate of the uncertainty. The Torres Strait region was characterised by weighting each physical covariate by its biological importance, then grouping the 0.01? grid cells into strata that had similar physical attributes. The stratification was mapped and represents an interim surrogate characterisation of Torres Strait. Sites for future sampling by the CRC-TS Mapping Task were selected from the biophysical strata to provide representative coverage of the Torres Strait environment.
The Project described the current state of knowledge of the physical marine environment and the seabed habitats and biota, identified the major knowledge gaps and the key information needs for regional marine planning and ecosystem-based management.