Utilizing Hyperspectral Remote Sensing to Detect Concentration of Cyanobacteria in Freshwater Ecosystems

Harmful algal blooms (HABs) are a threat to freshwater quality, public health, and aquatic ecosystems. The economic losses suffered by the agricultural, fishing, and tourism industries as a result of HABs exceed billions of dollars worldwide annually, with cleanup costs from local and national governments reaching a similar price. Current manual field-based sampling methods followed by laboratory analysis to detect and monitor HABs in expensive, labor-intensive, and slow, delaying critical management decisions. Moreover, current detection methods have limited success documenting HABs in freshwater bodies and such attempts employ satellite-based multispectral remote sensing; however, satellite-based methods are limited by cost, low spatial and spectral resolution, and restricting temporal windows for on-demand revisits. Our study used relatively low-cost unpiloted aerial systems (UAS) and hyperspectral sensors to detect HABs with higher resolution while having the capacity to conduct near real-time detection. Additionally, our hyperspectral remote sensing can detect and differentiate between HABs that produce cyanobacteria and other chlorophyll-producing plants. We detected a spectral peak of 710 nm that is characteristic of cyanobacteria producing HABs. Principal components analysis (PCA) was useful to spatially highlight HABs over wide areas. By utilizing hyperspectral remote sensing with UAS, HABs can be monitored and detected more efficiently. This new state-of-the-art research methodology will allow for targeted assessment, monitoring, and design of HABs management plans that can be adapted for other impacted inland freshwater bodies.