Derived publication data for machine-learning prediction of marine vegetated habitats using multibeam echosounder-derived environmental features
Description
This dataset contains derived publication source data supporting a study of multi-frequency multibeam echosounder (MBES)-based prediction of marine vegetated habitats at Hujeong and Bongpyeong, Republic of Korea. The package includes source data underlying the main and supplementary figures and tables, validation summaries, candidate-selection results, mapping summaries, publication metadata, a data dictionary, a file manifest, and SHA-256 checksums. Raw MBES survey data, model binaries, proprietary imagery, manuscript artwork, and large derived raster products are not included. The associated analysis code is available at https://github.com/neobassist/MBES-Habitat-Framework.
Files
Steps to reproduce
The deposited files are derived publication source data rather than raw survey inputs. Figure- and table-level CSV files can be used directly to inspect and verify the reported values. Validation, candidate-selection, and mapping summaries provide the underlying numerical results. The associated publication-specific analysis code and configuration files are available at https://github.com/neobassist/MBES-Habitat-Framework. Full raw-to-publication reruns require restricted MBES survey inputs that are not included in this dataset.