Variability in the Backscattering Coefficient of Particles (b_bp) as a Function of Chlorophyll a in Case II Waters

A relationship may exist between chlorophyll a concentrations (chl a) and the backscattering coefficient of particles (b_bp) in Morel Case II waters. This relationship has been discussed in past literature within Case I waters (Kitchen and Zanaveld, 1990; Stramski and Kiefer, 1992), but it was not until recently that equipment was available to obtain a measurement of the backscattering coefficient, b_b. This study examines whether a correlation exists between discrete chl a measurements as quantified by a spectrophotometer, and the backscattering coefficient as measured by the HOBI Labs Hydroscat-6, in West Sound, WA (Case II waters). Once the chlorophyll a to backscattering coefficient comparison was preliminarily examined, a quantitative assessment of the robustness of chlorophyll-specific scattering coefficients (b*) was addressed. Theoretical background for the work was based on the Morel (1987) model of estimating chlorophyll-specific scattering from the relationship:

where

and N is the number of equally sized particles in a volume (V) of medium, Q_b is the efficiency factor for scattering (ratio of radiant energy scattered from a particle to the energy impinging on cross-section (s)), and

where c_i is the intracellular concentration of pigment (chl a), and v is the volume of the individual particles (((/6)*d³). Measurements of absorption by particles (a_p) and scattering by particles (b_p) were made with a WETLabs AC-9. Discrete chlorophyll concentrations were measured at three depths in the water column; 2 m, 9 m (chlorophyll maximum), and 15 m. The magnitude of spectral b_p increased with biomass between 2 m (3.56 mg m^-3) and 9 m (12.17 mg m^-3), as did the spectral backscattering coefficient for particles (b_bp), with little change in spectral shape. The ratio of b_bp/b_p was used as an indicator of particle size, and the 2 m and 9 m b_bp/b_p spectra were nearly superimposed, possibly indicating similar sized particles (diameter) at both depths produced the b_p and b_bp signals. The b_p spectrum at 15 m (chl a concentration of 9.32 mg m^-3) was located between the 2 and 9 m spectra and had a similar spectral shape, possibly confirming the effects of biomass (chl-containing particles or phytoplankton) on the magnitude of b_p spectra with depth. The b_bp spectrum for 15 m particles was nearly superimposed on the 9 m spectrum, showing the effects of scattering by particles other than chlorophyll-containing biomass, possibly detrital particles. Examination of the bb_p/b_p spectra for 2, 9, and 15 m indicated the 15 m spectrum had the highest values of b_bp/b_p, around 1.2-1.4 % of b_p, which may confirm the presence of smaller particles at 15 m which are more efficient backscatterers. To obtain potential closure for the relationship between effects of chlorophyll biomass on scattering and backscattering coefficients, the particle size distribution was examined from discrete water samples obtained from similar depths as chl a samples analyzed by a Coulter Counter. Slopes (gamma) of particle size distribution ranged from —2.189 at 2 m (larger particles), to -2.264 at 9 m (smaller particles, but close to the 2 m particle distribution), and —2.618 at 15 m, indicating the smallest particles (higher b_bp/b_bp) were present at 15 m, confirming the information inferred in the b_p, b_bp, and b_bp/b_p spectra. Finally, the robustness of chlorophyll-specific scattering coefficients (b*) was examined. Spectra for b* at 2 m, 9 m, and 15 m indicated that particles at 2 m have the most efficient scattering per chlorophyll, with lower b* at 15 m and 9 m and much less efficient in scattering. However, b* does not seem to be an accurate measurement of scattering per chl a, as a primary assumption in b* is that all scattering particles contain chlorophyll a. The b/C estimate ignores effects of pigment packaging which may affect scattering and may not be reflected in the b* spectrum. Morel (1987) found b* to be inversely proportional to intracellular pigment concentrations, and spectra from the three depths show that b/C does not follow this trend. Variables in the Morel (1987) equations could be measured and may provide a more accurate assessment of chl-specific coefficients in Case II waters.