Mapping seagrass habitats of potential suitability using a hybrid machine learning model

Seagrass meadows provide essential ecosystem services globally in the context of climate change. However, seagrass is being degraded at an accelerated rate due to ocean warming, acidification, aquaculture, and human activities. The need for more information on seagrasses’ spatial distribution health status a serious impediment their conservation management. Therefore, we propose new hybrid machine learning model (RF-SWOA) that integrates sinusoidal chaos map whale optimization algorithm (SWOA) with random forest (RF) accurately suitable habitat potential seagrasses. This study combines situ sampling data multivariate remote sensing train validate models. It shows RF-SWOA can predict suitability efficiently than RF. also two most important factors affecting Hainan Island China are distance land (38.2%) depth sea (25.9%). paper not only demonstrates effectiveness but provides accurate approach predicting research help identify areas thus develop strategies restore healthy ecosystems.