Monitoring cyanobacteria blooms from drone based imaging systems

Session: 37. - Harmful Algal Blooms (HABs) and their Toxicity: Remote Sensing and Modeling Approaches

Ryan Ford, Rochester Institute of Technology, [email protected]
Anthony Vodacek, Rochester Institute of Technology, [email protected]

Abstract

A cyanobacteria bloom’s intensity can vary spatially on the order of meters, and temporally on the order of hours, making monitoring from remote sensing satellites difficult. Satellites collect imagery with spatial resolutions in the 10s to 100s of meters, on the order of 2-3 days at best while being subject to obscuration by clouds. Drone based imaging systems can overcome these limitations with their high spatial resolutions and ability to be implemented at any time, even under cloud cover. A drone was used to collect imagery over 2016 and 2017 blooms in a western New York lake. A model based spectrum matching algorithm was applied to the imagery to retrieve concentrations of bloom pigments. The retrieved concentrations were compared to measurements of water samples collected simultaneously with the imagery. The preliminary results of this comparison showed that chlorophyll-a could be retrieved with a percent error of 7%. This indicated that drone based imagery has great potential for monitoring cyanobacteria blooms, especially near beaches and public water supply intakes where the greatest hazards occur.

1. Keyword
remote sensing

2. Keyword
harmful algal blooms

3. Keyword
cyanophyta

4. Additional Keyword
Drone

5. Additional Keyword
Spectrum Matching

6. Additional Keyword
Hyperspectral/Multispectral