Date of Award

Spring 2023

Project Type


Program or Major

Natural Resources and Environmental Studies

Degree Name

Master of Science

First Advisor

Russell G Congalton

Second Advisor

Benjamin T Fraser

Third Advisor

Amanda McQuaid


With the increasing occurrence and growing public health concern that cyanobacteria blooms pose, it is crucial that we continue to explore ways to improve our ability to accurately, efficiently, and safely monitor water quality in impacted lakes. Also known as blue-green algae, cyanobacteria are naturally occurring in many waters globally. Cyanobacteria harmful algal blooms (CHABs) release various toxins which can cause skin irritations and dog fatalities while the effects of long-term exposure to the neurotoxins for humans has stirred additional studies. Although possibly harmful if CHABs are present, monitoring this biological component is nonetheless an integral factor when studying freshwater ecosystems. The use of an unmanned aerial system (UAS), equipped with a high resolution multispectral dual imaging sensor, provides a novel and full waterbody approach to quantify CHABs. Using a DJI M300 RTK unmanned aerial copter equipped with a MicaSense dual camera system collecting data in 10 wavelengths, six NH lakes were monitored from May-September 2022. Five of these six lakes experienced cyanobacteria blooms during this field season. Using the UAS collected spectral data coupled with collected in-situ water quality data, we used the random forest algorithm to classify the remotely sensed data and predict water quality classification categories. The analysis yielded very high overall accuracies for cyanobacteria cell concentration (93%), chlorophyll-a concentration (87%), and phycocyanin concentration (92%). Reflectance data from the 475 nm wavelength, the normalized green blue difference index – version 4 (NGBDI_4), and the normalized green-red difference index – version 4 (NGRDI_4) indices in addition to a few others were the most important features for these classifications. Additionally, simple logarithmic regressions illuminated relationships between single bands and indices with water quality data. Particularly, cell concentration with NGBDI_4 (R2 = 0.31), chlorophyll-a concentration with 475 nm (R2 = 0.24), and phycocyanin concentration with NGBDI_4 (R2 = 0.27). Therefore, our proposed monitoring approach successfully classified cyanobacteria cell, chlorophyll-a, and phycocyanin concentrations in the sampled NH lakes using UAS multispectral data while identifying the multispectral properties most important for cyanobacteria identification.