395 seabed samples collected between 2011 and 2017 were used to create a 10-meter resolution raster grid predicting seabed sand content (63-2000 µm) for Darwin and Bynoe Harbour. The dataset was produced by Geoscience Australia, the Australian Institute of Marine Science, and the Northern Territory Government's Department of Environment and Natural Resources. Predictions are based on seven environmental variables derived from multibeam sonar data and show high accuracy, with a VEcv of 39% for sand.
Use Cases
- Create thematic habitat maps for marine resource management based on predicted sand content.
- Analyze seabed composition patterns for environmental baseline studies using the 10-meter resolution raster.
- Model sediment distribution in harbour regions using predictions derived from multibeam bathymetry and backscatter.
- Compare predictive accuracy of sand content models using the reported VEcv metric.
- Identify areas of data absence for future survey planning based on noted gaps in predictive variables.
Strengths
- High-resolution 10-meter grid provides spatially continuous predictions.
- Predictive accuracy for sand is reported as VEcv = 39%, with an 84.8% improvement over a common interpolation method.
- Data compilation includes multiple surveys from 2011 to 2017, providing a multi-year baseline.
Limitations
- Absences in predictions occur due to non-availability of predictive variables.
- Column-level documentation is absent; field semantics must be inferred after download.
Provenance
- Source
- Australian Ocean Data Network
- Collection Method
- Predictions derived from 395 samples and seven environmental variables from multibeam sonar data.
- Time Range
- Survey data compiled from 2011 to 2017.
- Freshness
- Last updated 2026-04-16 14:31:40.558644; freshness should be verified.
- Geography
- Greater Darwin Harbour and Bynoe Harbour region, northern Australian marine margin.