Chlorophyll a Spatial Microstructure Determination from Volumetrically Reconstructed Optical Serial Sectioned Fluorescence Images.

Knowledge of the spatial and temporal variability of the optical properties of coastal and oceanic waters has important applications in biological and physical oceanography. On a sub-meter scale the variation in absorption, scattering, and fluorescence properties indicates the distribution and interaction of biological particles and the effects of internal waves, currents, and localized density differences. In a demonstration of the capability to determine microscale phytoplankton distributions, an underwater serial sectioning tomography technique which measures induced Chlorophyll a fluorescence has been developed. The technique involves scanning a 457 nm illumination plane through a range of distances parallel to the imaging plane of a 1024 ^2 pixel digital CCD camera. Images of induced fluorescence at 685 nm in the sequentially illuminated planes are recorded. A theoretical model of the imaging process in the form of i = ln c + Ac + x shows that the image, i, is a function of a spatially varying c (Chl a) -dependent absorption component A, and a spatially invariant attenuation component, x. A numerical inverse method, incorporating an environmentally dependent calibration constant, is used to calculate Chl a values from the image intensities for three dimensional presentation. A stable reconstruction, not affected by noise induced error propagation, was demonstrated. Laboratory testing of a prototype system has shown that Chl a concentrations from 0.1 to 2.0 mg Chl a/m^3 at a 0.1 mg Chl a/m ^3 resolution can be determined. The limiting factor in the imaging process is the rapid attenuation of the fluoresced 685 nm light which determines the system optical depth, xi. With 0.1 mg Chl a/m^3 as a minimum, xi = 0.97. At this xi and c_{(it 685)} <= 0.7 m^{-1}, a 1 m ^3 volume of 0.1 mg Chl a/m ^3 can be imaged with a spatial resolution of 1 cm^3. At higher Chl a levels larger values of xi permit larger volumes to be imaged. The remote sensing aspect of this underwater tomography technique permits in-situ Chl a microstructure information to be obtained without disturbing the naturally occurring phytoplankton distributions.