Evaluating the utility of Sentinel-2 imagery in estimating water quality in an ecologically significant, shallow, and optically complex estuarine lake system
Year: 2024
Type: conference-paper
Abstract: Lake St Lucia is Africa’s largest estuarine lake system and is a Ramsar and World Heritage Site. Shallow coastal lakes are especially dynamic environments that shift between wet and dry periods where desiccation can result in hypersalinity and other significant changes in characteristics, influenced too by estuary mouth dynamics. Different sediment loads from the lake’s constituent surface water catchments, the distribution of benthic sediment and its resuspension by wind-induced wave action and biota means complex spatial and temporal changes in suspended sediment and turbidity. While much research has focussed on chlorophyll-a estimation in marine and reservoir environments, much less has been done in dynamic, optically complex environments such as this. The lake is routinely monitored for a range of physico-chemical parameters, and periodic more intensive field campaigns have included chlorophyll-a sampling. Using these data, and estimates from published sources, we have compiled a time series of chlorophyll-a for several sites. Cloud-free Landsat and Sentinel-2 images corresponding to field sampling have been identified. We extract reflectance values for individual multispectral bands and various ratios/indices previously used in remote sensing chlorophyll-a and suspended sediment estimation. These are used as inputs to a multiple linear regression model to establish the Landsat and Sentinel-2 bands and band ratios with the greatest potential to estimate chlorophyll-a and suspended sediment. The aim being to develop a spatial model to estimate these variables across Lake St Lucia and determine the utility of satellite-based remote sensing for future environmental management in the area.
Language: English
Subfield: Oceanography
Field: Earth and Planetary Sciences
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7204317035