Sponge Species Richness Predictions with Environmental Variables and Geostatistical Models

A 2017 study by Jin Li et al. applied random forest, generalized linear models, and hybrid geostatistical techniques to predict sponge species richness. The dataset contains model outputs depicting non-linear relationships with environmental predictors and the spatial distribution of species richness. It was published in Environmental Modelling & Software and is hosted by the Australian Ocean Data Network.

Use Cases

• Predicting marine sponge biodiversity based on environmental variables mentioned in the description
• Comparing performance of random forest and generalized linear models for spatial count data
• Generating high-accuracy maps of species richness distribution for conservation planning
• Investigating the relationship between species richness and hard seabed features as revealed in the study

Strengths

• Study published in a peer-reviewed journal (Environmental Modelling & Software, Volume 97, 2017)
• Methods address specific issues with variable selection and model selection for count data
• Outputs include spatial distributions with high accuracy as stated in the description

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count and file size are unknown, which may limit suitability assessment

Provenance

Source

Australian Ocean Data Network

Collection Method

Model outputs from a study applying machine learning and geostatistical techniques to species richness data.

Time Range

Study published in 2017.

Freshness

Last updated 2026-04-10 22:01:55.169818; freshness should be verified

Geography

Likely focuses on Australian marine environments, given the source organization.