Figure 1. Relative densities of algal chlorophyll as indicated reflectance
ratio algorithm, R700/R670. Lighter colors indicate denser algal populations.
Figure 1. Relative densities of algal phycocyanin as indicated by a spectral trough near 625 nm. Warmer (red) colors indicate denser algal populations.
Figure 3. Hyperspectral remote sensing estimation of chlorophyll
Figure 4. Landsat Image of Sand
Introduction: To assess water quality in Nebraska’s 2500+ lakes and ponds, CALMIT is pioneering new techniques for using aerial and satellite remote sensing. Upwelling radiance of water contains information on the concentrations and composition of dissolved and suspended substances in water, and is the "raw material" for remote estimation of concentrations of water constituents. Here we focused on spectral features of the upwelling radiance determined by chlorophyll a absorbance and scattering by phytoplankton cells.
- to study the spectral features of reflectance of different water bodies during different seasons of the year, in order to find spectral features which are closely related to phytoplankton density.
- to devise and validate algorithms for chlorophyll estimation using reflectance data as the measured variables.
Methods: The project involves the use of aerial hyperspectral data (i.e., data collected in dozens of very narrow "color" bands). The data are obtained using an AISA imaging spectroradiometer carried aboard CALMIT’s Piper Saratoga aircraft. Data obtained with the AISA are calibrated using close-range spectroradiometers and water samples collected simultaneously in the field. Pigments from cyanobacteria are of special interest. Highly dense algal blooms can have severe impacts on human health, aquatic ecosystems and the economy. By quantifying cyanobacterial density, early warnings can be provided to the public for affected lakes.
Image Classification: Data processing was accomplished by using algorithms derived from close-range spectroradiometers and laboratory experiments. The data have a spatial resolution of approximately 1 meter.
Project Results: Remotely-sensed data revealed a high probability of toxic algae when (a) chlorophyll concentrations rose above 50 mg/m3 and (b) spectral features detected substantial algae-produced phycocyanin levels. These criteria successfully predicted the presence of toxic algal levels when water samples from identified lakes were tested. The images (left) were acquired on June 6, 2005 using CALMIT's AISAEagle hyperspectral imaging system over the Fremont State Lakes in Nebraska. Mapped reflected spectra indicative of chlorophyll imply denser algal colonies, highlighted in lighter colors (Figure 1). A similar map highlights pixels containing phycocyanin (Figure 2).
Information gained from several spectral bands in the red and near-infra-red ranges of the spectrum was sufficient to predict phytoplankton density.
Funding for this project was provided by grants from Israel National Academy of Sciences and Humanities, Water Commissioner's Office, Israel Ministry of Science and from the US-Israel Binational Science Foundation (BSF Grant N. 94-00087/1).
For more, read "Remote estimation of phytoplankton density in productive waters."