You are here:
Comparative Sensor Fusion between Hyperspectral and Multispectral Remote Sensing Data for Monitoring Microcystin Distribution in Lake Erie
Citation:
Chang, N., B. Vannah, AND J. Yang. Comparative Sensor Fusion between Hyperspectral and Multispectral Remote Sensing Data for Monitoring Microcystin Distribution in Lake Erie. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing . Institute of Electrical and Electronics Engineers Incorporated (IEEE), Piscataway, NJ, 7(6):2426-2442, (2014).
Impact/Purpose:
Communicate to science and technical community on EPA's advances in using remote sensor to monitoring microcystin related to climate change and water quality
Description:
Urban growth and agricultural production have caused an influx of nutrients into Lake Erie, leading to eutrophic zones. These conditions result in the formation of algal blooms, some of which are toxic due to the presence of Microcystis (a cyanobacteria), which produces the hepatotoxin microcystin. Microcystis has a unique advantage over its competition as a result of the invasive zebra mussel population that filters algae out of the water column except for the toxic Microcystis. The toxin threatens human health and the ecosystem, and it is a concern for water treatment plants using the lake water as a tap water source. This presentation demonstrates the prototype of a near real-time early warning system using Integrated Data Fusion techniques with the aid of both hyperspectral (MERIS) and multispectral (MODIS and Landsat) remote sensing data to determine spatiotemporal microcystin concentrations. The temporal resolution of MODIS is fused with the higher spatial of MERIS and Landsat to create synthetic images on a daily basis. As a demonstration, the spatiotemporal distributions of microcystin within western Lake Erie are reconstructed using the band data from the fused products and applied machine-learning techniques. The performance of the models derived using fused hyperspectral and fused multispectral data are quantified using four statistical indices. The second task compared traditional two-band models against more complex genetic programming models for microcystin prediction. Analysis confirmed that genetic programs excel at accurately estimating microcystin concentrations in the lake, and the more detailed spectral reflectance data offered by hyperspectral sensors produces a noticeable increase in accuracy at low microcystin concentrations.
