Predicting Seabed Sand Content Across the Australian Margin Using Machine Learning

18 spatial interpolation methods were compared for predicting seabed sand content within the Australian Exclusive Economic Zone (AEEZ). The study, using samples from Geoscience Australia's Marine Samples Database extracted in August 2010, found RFIDS and RFOK were among the most accurate methods across three tested regions. Model averaging further improved prediction accuracy, with the most accurate methods reducing error by up to 7%.

Use Cases

• Benchmarking spatial interpolation methods for marine sediment data based on the comparison of 18 methods.
• Predicting seabed sand content across the Australian Exclusive Economic Zone (AEEZ) using the recommended RFOKRFIDS method.
• Improving spatial predictions of marine environmental variables based on guidelines for secondary variable selection and model averaging.
• Assessing the impact of search window size and mtry parameters on geostatistical model performance.

Strengths

• Compares 18 distinct spatial interpolation methods, providing a broad methodological evaluation.
• Reports a reduction in prediction error by up to 7% for the most accurate methods.
• Evaluates performance across three distinct geographic regions within the AEEZ.
• Includes analysis of 36 combinations of input secondary variables, methods, and regions.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data is based on samples from a single extraction event in August 2010; temporal coverage is limited.

Provenance

Source

Geoscience Australia Data

Collection Method

Samples extracted from Geoscience Australia's Marine Samples Database.

Time Range

Samples extracted in August 2010.

Freshness

Last updated 2026-04-20 02:24:48.801929; freshness should be verified.

Geography

Australian Exclusive Economic Zone (AEEZ), specifically three study regions within it.