The 3-D coupled biophysical model ECOSMO (ECOSystern MOdel) has been applied to simulate the spatial and temporal variability of primary and secondary production and biomass in the North Sea in 1984, In order to assess the spatial and temporal dynamics of these components, statistical methods based on empirical orthogonal function analysis (EOF) are introduced to biological oceanography and ecosystem research as valuable tool to investigate spatial temporal variability. These methods are used to describe the spatial temporal characteristics of the seasonal signal of phytoplankton and zooplankton biomass and production in the North Sea. Employing these techniques made it possible to separate regional and temporal variability into the annual pattern, its temporal characteristics and some basic regional modulations of the average seasonal signal. The analysis was able to identify the modulation of average seasonal characteristics with for example earlier blooming in the south and later blooming in the northwestern North Sea. The simulated temporal development of diatom and flagellate blooms showed clear and unique temporal characteristics. The diatom bloom was characterized by a sharp peak occurring between middle of March and the end of April, with little to no diatom biomass in the second half of summer. Conversely flagellate biomass did not peak before the beginning of May and showed a relatively constant summer production and an autumn bloom. (c) 2006 Published by Elsevier B.V.