Discriminating four temperate lakes using phytoplankton absorption spectra: Fid-Bau Portal

Discriminating four temperate lakes using phytoplankton absorption spectra

Kalenak, David

Photosynthetic algal particles are integral to the ecology and optical quality of lake water. Their absorption properties (photopigments, often >20 per species) are modified in response to drivers that affect the intensity and spectral distribution of incident solar energy, including the plankton itself, water, and chromophoric dissolved and suspended particulate matter. The present study was based on the hypothesis that this complex interaction imparts a distinguishing optical signature in lakes. A multivariate discriminant model was applied to in situ phytoplankton spectra from four temperate lakes of different trophy to test this hypothesis. The analysis identified a small set of wavelengths with significant discriminatory power that permitted near‐perfect lake classification solely on the basis of the aggregate spectral features in the absorption coefficient of phytoplankton. Furthermore, weighting the sample spectra by the corresponding standard deviation increased model robustness considerably, as shown in both unsupervised and supervised classification. Also presented are additional multivariate techniques that allow for the visualization of the data structure to help explain contributing factors. The results support the possibility of utilizing the optical properties of the phytoplankton to monitor ecological change in lakes.