Oh, H.-J.; Syifa, M.; Lee, C.-W., and Lee, S., 2019. Ruditapes philippinarum habitat mapping potential using SVM and Naïve Bayes. In: Jung, H.-S.; Lee, S.; Ryu, J.-H., and Cui, T. (eds.), Advances in Remote Sensing and Geoscience Information Systems of Coastal Environments. Journal of Coastal Research, Special Issue No. 90, pp. 41-48. Coconut Creek (Florida), ISSN 0749-0208.

The aim of this study was to compare the performance of Support Vector Machine (SVM) and Naïve Bayes (NB) models for potential R. philippinarum habitat mapping in the Geunso Bay, Korea. R. philippinarum samples were collected during field observation. Remote sensing data were used to identify the factors controlling the distribution of R. philippinarum. Habitat potential maps were constructed and eight controlling factors were generated from satellite imagery, namely elevation of intertidal zone, aspect, exposure time, slope, density of tidal channel, distance from tidal channel, surface-sediment distribution, and near-infrared reflectance (NIR). Validation of the maps was conducted by comparison with surveyed habitat locations. Performance of the SVM model (AUC=0.777) is better compared with NB model (0.733). The GIS-based SVM and NB models combined with remote sensing techniques are efficient tools for mapping potential R. philippinarum habitat in tidal flats.

This content is only available as a PDF.
You do not currently have access to this content.