Seabed Mud Content Predictions for the Australian Margin via Machine Learning and Kriging

A 2008 study comparing 14 methods found machine learning was among the most accurate for predicting seabed mud content. This dataset likely contains results from a simulation experiment comparing 18 methods, including combinations of machine learning with ordinary kriging and inverse distance squared, across three regions of the Australian Exclusive Economic Zone. The work was produced by Geoscience Australia using samples from its Marine Samples Database (MARS).

Use Cases

• Benchmarking spatial interpolation methods based on the comparison of 18 statistical and machine learning techniques.
• Predicting seabed sediment composition based on sample data from the Australian Exclusive Economic Zone.
• Assessing the impact of topographic slope and search window size on prediction accuracy for marine geology.
• Creating composite prediction maps by averaging outputs from methods like RFOK and RFIDS.

Strengths

• Compares 18 distinct spatial interpolation methods, including novel combinations of machine learning with kriging.
• Based on samples from Geoscience Australia's authoritative Marine Samples Database (MARS).
• Evaluation covers three distinct geographic regions (N, NE, SW) of the Australian margin.
• Identified methods reduced prediction error by up to 19% compared to a control.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data files are in PDF and HTML formats, which may require extraction for computational use.

Provenance

Source

Geoscience Australia Data

Collection Method

Samples extracted from Geoscience Australia's Marine Samples Database (MARS) for a simulation experiment.

Freshness

Last updated 2026-04-20 02:18:02.710913; freshness should be verified.

Geography

Australian Exclusive Economic Zone, specifically N, NE, and SW regions.