SIGNATURE ANALYSIS OF REFLECTANCE SPECTRA AND ITS APPLICATION
FOR REMOTE OBSERVATIONS OF THE PHYTOPLANKTON DISTRIBUTION
IN LAKE KINNERET
A, GITELSON 1 , M. MAYO 2 , AND Y. YACOBI 3
•^Ben-Gurion University of the Negev,
Sede-Boker Campus, 84993 (Israel)
2 Depart. of Geophysics and Planetary Sciences,
Tel-Aviv University (Israel)
3 Israel Oceanographic and Limnological Research,
Kinneret Limnological Laboratory
ABSTRACT:
To develop a quantitative technique for remote estimation of
chlorophyll concentration in Lake Kinneret (the Sea of Galilee),
optical models of the lake were devised and tested. Four
experiments were carried out in periods of low (3-20 mg.m J ) and
high (up to 200 mg.m -3 ) chlorophyll (Chi) concentration. Data
sets included high spectral resolution radiometric measurements
in the range from 400 to 850 nm, Chi and suspended matter
concentrations, Secchi disk transparency, and vertical
attenuation coefficients. For each period, the first data set was
used for development of the optical models, and the second was
employed for their validation. In the periods of low Chi, with a
background of variable and relatively high suspended matter
concentration, reflectance near 685 nm was used for Chi
assessment with an estimation error < 0.5 mg.m . For the high
Chi concentration period, the position and magnitude of the peak
of reflectance near 700 nm, were used for the Chi estimation
with an accuracy of about 7 mg.m -3 .
KEY WORDS: Optical model, Reflectance, Water quality
INTRODUCTION
Lake Kinneret (The Sea of Galilee) is the only large freshwater
body in Israel. Since the dynamics of water exploitation and
provision are not symmetrical, the lake undergoes cycles of
change in the water level. This fluctuation in the water level is
currently a major concern for water policy makers in Israel, as
the long-term impact of these changes on water quality is largely
unknown. A detailed monitoring program of the lake is the main
tool for acquisition of information needed for rapid response to
environmental questions. Algal density is a key parameter in
this monitoring program, being determined both by microscopic
count and measurement of chlorophyll (Chi) concentration.
The distribution of algae in the lake is irregular, both
spatially and temporally, causing severe sampling problems. An
effective method for monitoring the distribution of algae in Lake
Kinneret is remote sensing, which provides a synoptic view of the
investigated area. Often even a few images are useful as aids in
the design or improvement of point sampling programs through the
highlighting of the best sampling locations.
The fluorescence line height has been successfully used
for the remote detection of Chi (e.g., Neville and Gower, 1977;
Gower, 1980; Doerffer, 1981; Fischer and Kronfeld, 1990).
Quantitative accuracy of this technique is limited by the varying
277
