Environmental Monitoring of Cyanobacterial Harmful Algal Blooms Using Small Unmanned Aerial Systems
Microcystis , Cyanobacteria , Cyanobacteria Monitoring , Unmanned Aerial System , Harmful Algal Blooms
Cyanobacterial harmful algal blooms (CHABs) degrade water quality and may produce toxins. The distribution of CHABs can change rapidly due to variations in population dynamics and environmental conditions. Biological and ecological aspects of CHABs were studied in order to better understand CHABs dynamics. Field experiments were conducted near Hartington, Ontario, Canada, in ponds dominated by Microcystis aeruginosa during the summers of 2015 and 2016. Mixing of the water column followed by calm conditions resulted in over 90% of the Microcystis floating on the surface. Microcystis surface appearance formed four different assemblages: aggregate, ribbons, patches, and mats. Presence of CHABs on the water surface also depends on environmental influences such as direct and indirect wind effects. Surface coverage of CHABs was found to be reduced to half within an hour at wind speeds of 0.5 m/s. Our findings indicated that blooming involves surface display of cyanobacteria, therefore using surface imagery to quantify CHABs was justified. Traditional detection methods do not provide accurate distribution information, nor do they portray CHABs events in a real-time manner due to limitations in on-demand surveillance and delays between sample time and analyzed results. Therefore, a new CHAB detection method using small unmanned aerial systems (sUAS) with consumer-grade cameras was developed. When cyanobacteria were floating on the surface, CHABs detection through RGB band cameras and spectral enhancement techniques was efficient and accurate. sUAS were capable of providing coverage up to 1 km2 per mission and the short intervals between sampling and results (approx. 2 h) allowed for the rapid analysis of data and for implementing follow-up monitoring or treatments. sUAS were able to measure growth and decline curves of CHABs throughout 2016 which were then compared to water chemistry results. sUAS were cost-effective (estimated $50 CAD per mission) with an average cyanobacterial detection accuracy of 86% (8% of the error were false negatives, and 6% were false positives). Thus, it is a good candidate method to fill the urgent need for CHABs detection, providing cost effective, rapid, and accurate information to improve risk management at a local level and to help quickly allocate resources for mitigation.