GREEN LAKES FROM SPACE: MONITORING STORMWATER PONDS IN COASTAL SOUTH CAROLINA BY REMOTE SENSING

Stormwater ponds and lakes are engineered landscape features primarily built to mitigate flooding concerns due to increased surface runoff from impervious surfaces, but may also serve as an aesthetic anchor of a community. In coastal South Carolina, they primarily function as wet detention ponds. There is concern that these ponds act as gateways for the transport of contaminants from developed landscapes to receiving estuarine waters. Ecological and community-scale public risks may also be present since ponds and lakes receiving stormwater inputs may act as temporary sinks of harmful organic contaminants and metals, and be stressed by eutrophication (nutrient-enrichment) leading to a proliferation of algae, which may include harmful algal blooms (HABs). The objective of this research is to understand whether water quality parameters in dynamic, engineered stormwater structures can be adequately monitored using optical remote sensing. Optical measures may include color, clarity, and visual appearance—these signatures may be relatable to priority water quality measures (chlorophyll-a). Both satellite and near-ground remote sensing platforms (e.g. unmanned aerial vehicles/drones) may be applied for monitoring stormwater ponds. Usefulness of these platforms requires benchmarking data to in-field and in-lab measured water quality parameters. This work addresses a significant research opportunity, as stormwater ponds are landscape features that are smaller and shallower than water bodies traditionally monitored by remote sensing. These ponds are numerous and typically lie on private property, which makes traditional monitoring difficult. These considerations make enabling water quality monitoring by advanced approaches (i.e. by remote sensing) essential. Initial results show strong correlations for chlorophyll-a and total suspended solids when models are formulated on a site-specific basis; however, complications arise when attempting to develop generic models. Considering this, future work aims to consider multiple parameters to apply this approach to seasonal and location variation, and identification of harmful algal blooms.