We examined the species composition and concentrat~on of planktonic algae as well as irradiance and near-surface temperature and sal~nity 3 times per week for 1 yr in a turbulent marine environment with con.tinuously high nutrient concentrations Seasonal variations in tempera.ture and sahn~ty were small, and diatoms composed the largest percentage of phytoplankton (25 pm) cell numbers throughout the year. We observed at least 96 diatom species, of which some were present throughout most of the year while others were observed for periods as short as 1 to 2 wk. Diatom species diversity reached minimum and maximum values during wlnter and summer months, respectively, and was positively correlated to temperature and daylength and negatively correlated to salinity. Winter and spring blooms occurred and were numerically dominated by small and large cells of Skeletonema costatum, respectively, and Pseudo-nitzschia delicatissirna, resulting in minimum seasonal values of species evenness, which peaked during summer months. The first 4 axes of a canonical correspondence analysis (CCA) explained 76 % of the variance in the diatom data; the measured environmental variables explained 90% of this variation, of which temperature had a dominant role. We discuss possible environmental optima of 44 diatom species based on results of CCA and make suggestions regarding altered measures of diversity.

ABSTRACT. A seasonal study was carried out to determine whether or not volume-frequency distributions of near surface nano-and microplankton were approximated by a power function model in temperate neritic waters. Measurements of buoyancy frequency, the nutrients NO3-N, NO2-N, NH,-N, urea-N, Si(OH),-Si, and PO,-P, phytoplankton photosynthehc pigments, and cell concentrations and volumes were made around a frontal zone in the Strait of Georgia, Canada, at ca 3 w k intervals from Apnl 1983 to September 1984, excluding October to February. The model accounted for a large percentage of the variance in spectra when nano-flagellates and dinoflagellates dominated phytoplankton crops in pre-diatom bloom and nutlient-depleted waters. When diatoms dominated crops, the fit was better in summer than in spring and often was positively correlated to temperature Possible covariance of temperature and grazing by large omnivorous zooplankters on diatoms was considered The exponent of the power function varied around -0.92, derived from linear regression analysis of a log,-transformed allometric equation A geometric mean or functional regression slope was used because the slope varied with the correlation coefficient and because the relationship descnbes an allometric dependence. Results support the use of a power function model to predict volume-frequency spectra of temperate nentic phytoplankton whenever diatoms are not abundant or from the time of the demise of the spring diatom bloom until at least the early fall season when diatoms are abundant. The presence of diatoms such as Laudena boreahs, Skeletonema costatun], and Thalassiosira nordenskiold~i in larger numbers than predicted by their cell volumes invalidates use of the model during spring.