Phytoplankton Group Determination using Hyperspectral Remote Sensing in Western Lake Erie

Document Type

Conference Paper/Presentation

Publication Date



Determination of phytoplankton groups (green algae, diatoms, cryptophytes, and cyanobacteria) in natural water bodies is important information for many stakeholders including scientists, water intake managers, and recreational users. These determinations can be made in the field using fluorescence instrumentation or in the laboratory using microscopy and particle imaging. There is potential to determine phytoplankton groups from remote sensing hyperspectral data measured on the ground, in the air, and from space. Remote sensing provides the unique ability to provide synoptic coverage of a water body without the need to be in the field. These techniques rely on the ability to detect specific spectral absorption features associated with different pigments that vary with phytoplankton group. An extensive data set was collected from May-October 2015 in western Lake Erie that included weekly coincident measurements of phytoplankton composition and water surface reflectance. This robust data set was used to evaluate and generate remote sensing based phytoplankton classification approaches with variable success. Supervised classification methods were able to determine the dominant phytoplankton type from others (cyanobacteria vs. diatoms) while complex machine learning techniques could differentiate types and concentrations.

Publisher's Statement

© 2016 International Association for Great Lakes Research. Publisher's version of record: http://iaglr.org/conference/downloads/2016_abstracts.pdf

Publication Title

IAGLR 59th Annual Conference