| Summary: | Turbidity mapping and prediction using remote sensing has had limited success in the past. Previous research efforts have been conducted in temperate and even tropical climates where too many confounding factors affect the remote sensing signal. My research is focused on the development of an accurate and repeatable algorithm to predict turbidity in northern, glacial lakes using electro-optical satellite data. From an evolutionary perspective glacial environments are highly immature. Lakes found in these environments are typically classified as extremely oligotrophic resulting from their relatively recent formation and the surrounding harsh, northern climate. Unlike temperate or tropical lakes, northern glacial lakes do not contain significant amounts of biological material. Instead, these lakes are dominated by rock flour – suspended sediment originating from glacial rock weathering. This lack of biological influence makes satellite turbidity mapping and prediction more straightforward and potentially more accurate than similar efforts in temperate or tropical environments where biology typically drives these systems and strongly affects the remotely-sensed, electro-optical signal. The study site for my research was an ice marginal lake at the Bering Glacier located in coastal, south central Alaska. In situ turbidity data, collected using an autonomous robot buoy, was used to develop a model-based turbidity algorithm. Simple and multiple linear regression analyses were conducted using different Landsat 7 ETM+ bands to determine the best predictor(s) of turbidity in glacial lakes. The final algorithm utilized Landsat 7 ETM+ Band 3 (red portion of the electromagnetic spectrum) and Band 4 (near-infrared portion of the electromagnetic spectrum) data to predict turbidity concentrations. Turbidity maps created using the algorithm can be used to help determine inter- and intra-annual sediment dynamics of Vitus Lake. This information could be used to help researchers predict significant glacial events such as outburst ...
|
|---|