Automated coastline extraction for erosion modeling in Alaska.

Alaskan communities face the frontline impacts of climate change. Accurately predicting coastal changes is crucial for adaptation, but current methods overlook dynamic factors like declining sea ice and variations in tidal, temperature, and energy wave data. To enhance coastal modeling, more digitized coastlines are needed due to the inefficiency of manual interpretation. The project aims to develop an automated system for extracting coastlines from PlanetLab images by using techniques like Normalized Difference Water Index (NDWI) thresholding and DeepWaterMap algorithms. The main challenge is developing the NDWI index, a method for image labeling for the training data, specifically the thresholding value for the label. While also considering other preprocessing algorithms for atmospheric correction, mosaicking, etc. Additionally, devising a robust validation method that considers temporal variation is crucial. The project aims to have a model that automatically extracts a vectorized coastline representation from PlanetLabs satellite imagery with high accuracy.