Premium
Spectral reflectance and water quality of Adirondack mountain region lakes
Author(s) -
Vertucci Frank A.,
Likens Gene E.
Publication year - 1989
Publication title -
limnology and oceanography
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.7
H-Index - 197
eISSN - 1939-5590
pISSN - 0024-3590
DOI - 10.4319/lo.1989.34.8.1656
Subject(s) - reflectivity , irradiance , alkalinity , environmental science , dissolved organic carbon , water quality , mineralogy , environmental chemistry , remote sensing , chemistry , hydrology (agriculture) , geology , ecology , optics , biology , physics , organic chemistry , geotechnical engineering
The reflectance properties of lakes in the Adirondack mountain region are described. Five distinct types of reflectance spectra are identified and related to differences in water quality. No relationship was found between reflectance and acidification status. A reflectance model based on literature absorbance and scattering cross‐sections of water, plant pigments, dissolved organic carbon (DOC), and suspended inorganic matter is used to generate reflectance spectra of each reflectance type. Use of a variable Q factor during the conversion of remotely sensed reflectance to irradiance reflectance improves the correspondence between measured and modeled irradiance reflectance spectra. Comparison of measured reflectance with model reflectance also demonstrates the importance of small concentration changes as well as the interactive influence of plant pigments, DOC, and suspended material on lake color. Correlograms depict the wavelength‐dependent relationships between reflectance and water quality. Plant pigments, DOC, and suspended material are negatively correlated with reflectance from 400 to 600 nm and positively correlated from 600 to 750 nm; parameters associated with lake acidification, pH, alkalinity, and aluminum concentrations are poorly correlated with reflectance. Adirondack lakes exhibit a high degree of covariance between pigment, DOC, and suspended matter concentrations. Regression models predicting total pigments and DOC are developed which suggest that these parameters can be estimated remotely in these lakes, but remote estimation of pH, alkalinity, and aluminum is not possible with reflectance measures. A regression model estimating total pigments from the reflectance ratio of 525 : 554 nm is similar in slope and accuracy to one derived for the Coastal Zone Color Scanner (CZCS) for coastal marine waters. However, neither of the “recommended” CZCS pigment algorithms perform best with the Adirondack lake reflectance and pigment data.