Google Earth Engine, Open-Access Satellite Data, and Machine Learning in Support of Large-Area Probabilistic Wetland Mapping

Description: This journal article discusses a workflow for predicting the occurrence of wetlands on the landscape at a regional scale using Earth-observation data. When the probability-of-occurrence model is converted into a wet-dry classification, the model yields an overall accuracy of 85%. Given this accuracy, it provides a scalable foundation for province wide monitoring initiatives in Alberta.