Unialgal cultures were used to investigate relationships between cell volume and the carbon and nitrogen content of nondiatomaceous marine nanophytoplankton. Cell dimensions were determined by image-analyzed epifluorescence microscopy and particulate C and N by high-temperature dry combustion. Volumes were calculated by direct integration with published algorithms (biovolume), but could be estimated equally well from linear dimensions as prolate spheres. Preservation with 0.5% glutaraldehyde reduced cell volumes 29% on average. Correlations were highly significant between biovolume of preserved cells and C and N contents. Nonlinear regression models appeared most appropriate because smaller cells contained more C and N per unit volume than did larger cells. Suggested general C densities for estimating cell C from preserved volume were 0.36 pg pm-3 for 10' pm3 cells, 0.24 pg pm-3 for lo2 pm3 cells, and 0.16 pg C pm-3 for lo3 pm3 cells. Previous regression models substantially underestimated the C densities of nanophytoplankton of lo*-lo3 l.cm3. The explanation for these differences includes the method of determining mean population volumes, the use of preservatives, and the occurrence of significant vacuolar volume in larger phytoplankton.

No abstract

Particulate matter concentration (PM, often referred to as total suspended solids [TSS]) is an important parameter in the evaluation of water quality. Several optical measurements used to provide an estimate of water turbidity have also been used to estimate PM, among them light transmission, backscattering, and side-scattering. Here we analyze such measurements performed by the Alliance for Coastal Technologies (ACT) at various coastal locations to establish whether a given optical method performs better than others for the estimation of PM. All the technologies were found to perform well, predicting PM within less than 55% relative difference for 95% of samples (n = 85, four locations). Backscattering performed best as a predictor of PM, predicting PM with less than 37% relative difference for 95% of samples. The correlation coefficient (R) was between 0.96 and 0.98 for all methods with PM data ranging between 1.2 to 82.4 g m -3. In addition, co-located measurements of backscattering and attenuation improves PM prediction and provides compositional information about the suspended particles; when their ratio is high, the bulk particulate matter is dominated by inorganic material while when low, dominated by organic material.

No abstract

No abstract