EMPLOYING REMOTE SENSING TECHNIQUES TO UNDERSTAND SEASONAL CHANGES IN WATER QUALITY IN THE MUSKINGUM WATERSHED CONSERVANCY DISTRICT

The Muskingum Watershed Conservancy District (MWCD) encompasses several reservoirs in Northeast Ohio. In addition to its flood reduction benefits, it provides many recreational activities that contribute to the local economy. Lately, the MWCD has experienced a rise in harmful algae bloom (HAB) occurrences. Poorly regulated agricultural practices have caused nutrient oversaturation and heightened phosphorous load in local reservoirs, leading to an influx of seasonal algae growth. Concurrently, the MWCD has been subjected to varying degrees of coal mine drainage (CMD) due to extensive coal mining. If left untreated, these anthropogenic catalysts can severely affect aquatic ecosystems and local economies. As a result, early warning procedures and advanced water quality monitoring systems are critical in today’s evolving climate. To address this, we have applied a varimax-rotated principal component analysis (VPCA) to satellite imagery from Sentinel-2 A/B MSI using a Google Earth Engine (GEE) Python API combined with ground truthing and drone-based multispectral surveying to validate the satellite analysis. The VPCA uses unsupervised classification methods to extract six components from spectral images to be identified by a spectral library. Spectral reflectance is unmixed from water pixels to identify components such as various algae, suspended sediment, iron oxide/sulfide minerals, and semi-submerged surface vegetation. Employing Google Earth Engine (GEE) components generated by the KSU-VPCA, validated by ground-truthing field measurements, will allow for the identification of suspended sediment, HABs and other water column constituents. This research effort will demonstrate an advanced water quality tracking procedure for monitoring the spread of HABs and coal mine drainage over time.