Tracking spatial and temporal changes in coastal wetlands of lakes using high-resolution remote sensing and machine learning
Wetlands are critical ecosystems that support biodiversity, improve water quality, and mitigate flooding risks, yet they are increasingly threatened by human activities and climate change. This project focuses on delineating and tracking changes in wetland boundaries around lakes to understand these impacts over time. By using high-resolution Landsat surface reflectance imagery and land cover maps as ground truth data, we develop a machine-learning model to analyze spatial and temporal shifts in wetland boundaries. Preliminary results show a significant difference in the distribution of water versus land pixels across the study period, indicating boundary changes that reflect environmental and anthropogenic influences. These findings suggest that our approach can effectively capture long-term wetland dynamics, providing insights to support conservation and restoration efforts.