a stressed condition in the algae which caused this change in fluorescence. 

 Laboratory results which were reported by Blasco (ref. 27) are shown in fig- 

 ure 16. These data were taken over a 20-minute period of time, and they 

 exhibit a positive change in fluorescence with temperature. In contrast to 

 these two sets of data, Kiefer (ref. 26) found, while conducting experiments 

 in the Gulf of California, that temperature had no effect on the fluorescence 

 of chlorophyll a in vivo. It is assumed that within naturally occurring 

 limits, temperature is a relatively unimportant parameter in the determination 

 of chlorophyll a in vivo fluorescence. 



Additional Uncertainties in the Remote Quantification of 



Chlorophyll a In Vivo 



It has been assumed in the preceding derivation of the equation for the 

 power received by a laser fluorosensor system that measurements at different 

 excitation wavelengths are made for the same water sampling volume. Under this 

 condition, it would be difficult to predict the effective attenuation coeffi- . 

 cients to greater than 30-percent accuracy even if the values for the absorp- 

 tion and scattering coefficients were known precisely. This is because the 

 form of the equation for the effective attenuation coefficient is only an 

 approximation and because the attenuation coefficients vary depending upon the 

 scattering characteristics, and the amount and vertical distribution of tur- 

 bidity. If no knowledge of the scattering coefficient can be remotely obtained 

 for the sampling volume, the uncertainty in Y^ and Yf would be greater than 

 50 percent. Further, the effective attenuation coefficient at each excitation 

 wavelength has at least a 5- to 10-percent uncertainty relative to other exci- 

 tation wavelengths. If the measurements for all excitation wavelengths cannot 

 be made for the same sampling volume, the uncertainty in the attenuation coef- 

 ficients for each laser excitation becomes very large, unless the body of water 

 is uniform in turbidity over the distance traveled during one measurement 

 cycle. 



For a field measurement of laser output power and received fluorescence 

 power, it is assumed that a measurement accuracy of 5 percent is typical with 

 2.5 percent attainable. 



A greater than 40-percent variation in the fluorescence cross section for 

 various species within the same algal color group (ref. 14) can create large 

 uncertainties in determination of the chlorophyll a in vivo concentration by 

 the laser fluorosensor system. Some of this intragroup variation may only be 

 in absolute magnitude, and the relative spectral characteristics may be the 

 same. For example, an error in the measurement of chlorophyll a contained in 

 the algae can arise because of difficulty in complete extraction of all chloro- 

 phyll a from the algal cells. Part of the uncertainty in fluorescence cross 

 section can result from changes in the relative excitation spectra of various 

 species within a color group. The source and nature of this cross section 

 variation is not known; however, it is realistic to assume a 5- to 10-percent 

 uncertainty in the relative fluorescence cross section at each excitation 

 wavelength. 



21 



