It’s Not Easy Being Blue-Green: Developing a Remote Cyanobacteria Detection Tool

EarthzineDEVELOP Summer 2014 VPS, Original

Project Team: Georgia Water Resources Team
Team Location: University of Georgia, Athens, Georgia

(Left) Image of Georgia Power reservoir study sites derived from Landsat 8; (Right) snapshots of ancillary data collection. Image Credit: Georgia Water Resources Team.

(Left) Image of Georgia Power reservoir study sites derived from Landsat 8; (Right) snapshots of ancillary data collection. Image Credit: Georgia Water Resources Team.

Authors:
Bradley Bartelme, Project Lead (University of Georgia)
Ike Sari Astuti (University of Georgia)
Elizabeth Benyshek (University of Georgia)
Shuvankar Ghosh (University of Georgia)
Danielle Haskett (University of Georgia)
Jiaying He (University of Georgia)
Benjamin Page (University of Georgia)

Mentors/Advisors:
Dr. Deepak Mishra (University of Georgia)
Dr. Susan Wilde (University of Georgia)

Abstract:
The effects of anthropogenic eutrophication are intensified in Georgia’s watersheds due to increasing temperatures, higher frequency drought events and higher availability of nutrients that increase primary productivity in reservoirs. These factors may ultimately lead to the formation of toxic Cyanobacterial Harmful Algal Blooms (Cyano-HABs or HABs). The abundance of phycocyanin, a phycobiliprotein, may be used as a proxy to assess the amount of cyanobacteria biomass that is present in a water body and is useful as a cyanobacterial bloom indicator. The Georgia Water Resources Project developed an early detection tool to aid in the identification and spatial distribution of phytoplankton and blue green algae (cyanobacteria) for Georgia inland waters using Landsat 8 Operational Land Imager (OLI) data paired with hyperspectral data. These data were calibrated using four spectral band ratio models: 1) Normalized Difference Chlorophyll Index (NDCI); 2) the 2-band Cyanobacteria Model; 3) Quasi-analytical Model; and 4) Phycocyanin detection model. In collaboration with Georgia Power Company (GPC), a model upscaling procedure demonstrated the feasibility of using Landsat 8 sensors to detect cyanobacteria reflectance patterns. This procedure will assist in the maintenance of water quality throughout Georgia and is imperative due to the shortage of freshwater resources present in man-made reservoirs. From this research, spatial and temporal distribution maps were delineated for the early identification of CyanoHABs in order to rapidly monitor and respond to these systems and aid in water management decision-making for Georgia reservoirs.

Back to VPS page.