Forty Years of Wetland Status and Trends Analyses in the Great Lakes Using Landsat Archive Imagery and Google Earth Engine

Wetlands provide many benefits, such as water storage, flood control, transformation and retention of chemicals, habitat for species plants animals. The ongoing degradation wetlands in the Great Lakes basin has been caused by a number factors, including climate change, urbanization, agriculture. Mapping monitoring across large spatial temporal scales have proved challenging; however, recent advancements accessibility processing efficiency remotely sensed imagery facilitated these applications. In this study, historical Landsat archive was first employed Google Earth Engine (GEE) to classify (i.e., Bog, Fen, Swamp, Marsh) non-wetlands Open Water, Barren, Forest, Grassland/Shrubland, Cropland) throughout entire over past four decades. To end, an object-based supervised Random Forest (RF) model developed. All produced wetland maps had overall accuracies exceeding 84%, indicating high capability developed classification mapping. Changes were subsequently assessed 17 time intervals. It observed that approximately 16% study area changed since 1984, with highest increase occurring Cropland class decrease Marsh classes. mostly transitioned but also transition Cropland, Marsh, Swamp. A considerable amount converted into Cropland.