An Evaluation of the Effects of Environmental Variables on Marine Plankton Primary Productivity by Multivariate Regression

Publisher Website
Google Scholar
Abstract
Stepwise multiple regression analysis has been used to evaluate the relative importance of environmental variables for primary productivity in oceanic areas off Washington and Oregon, USA. Special emphasis has been placed on the area affected by the Columbia River plume. In the plume, a large proportion of the total variability in primary productivity data is explained by only four variables: chlorophyll a, nitrate, depth of the photic zone, and temperature. In ambient sea water, the significant variables are salinity, temperature, solar radiation of the previous day, vertical stability in the upper 100 m, and phosphate. Analysis of the combined data indicates that, in the spring, primary productivity depends on chlorophyll a concentration alone. The results are explained from the biological and physical points of view. The usefulness of multiple regression analysis for marine ecosystem studies appears to be limited.