A size-structured ecosystem model was developed and fitted to an extensive set of observations of size-fractionated biomass, primary and heterotrophic bacterial production, respiration, particle flux, and zooplankton herbivory obtained on a cruise in the NE Atlantic during 20 days of the spring bloom. Rates of water-column-integrated primary production and heterotrophic bacterial production were used to force the model, and the model parameters were determined by optimizing the model to the remaining observations. A unique model solution was not possible in the absence of dissolved organic carbon (DOC) measurements. By setting a weak constraint on the DOC concentrations (that they should remain broadly constant), a solution was obtained that predicted a small DOC increase during the bloom period; this increase was similar in magnitude to observations in other areas and produced a good fit with the other data. Major conclusions: (1) gross primary production (GPP) was over twice the measured 14 C production estimates; (2) phytoplankton exudation was a major source of DOC, and 35% of the GPP flowed through the DOC and was utilized by the bacteria, with a low gross growth efficiency (17%); (3) 62% of the community respiration came from organisms Ͻ5 m in size; (4) the major source of food for microzooplankton was phytoplankton 1-5 m in size, but the mesozooplankton ingested more microzooplankton than phytoplankton; and (5) only 8% of the total net primary production was grazed by mesozooplankton. These results provide quantitative support to the emerging consensus on the relative roles of large and small organisms in the euphotic zone and demonstrate the necessity of measuring DOC to constrain the carbon budget within the euphotic zone.A prominent objective of biological oceanography is to determine and understand the processes controlling the fluxes of carbon in the ocean over time. The most important of these is the fate of CO 2 taken up by phytoplankton in the euphotic zone. We wish to quantify how this carbon passes through the food web to its eventual fate as carbon respired within the euphotic zone or carbon exported to the deeper ocean. An attempt to budget carbon was one of the aims of the Joint Global Ocean Flux Study (JGOFS) North Atlantic Bloom Experiment (NABE) carried out in 1989 at a number of stations along the 20ЊW meridian. The most extensive coverage was at 47ЊN, where Bender et al. (1992) calculated a carbon budget for the upper water column (0-50 m) over a 13-d period during the spring bloom. Of the total gross primary production of 1.83 mmol C m Ϫ2 , they estimated that 63% was respired, 16% was stored as particulate organic carbon (POC) within the upper 50 m, and 18% was sedimented out to depths Ͼ50 m. Bender et al. (1992) did not attempt to unravel the details 1 Corresponding author: mjf@soc.soton.ac.uk.
AcknowledgmentsWe thank those colleagues in the UK BOFS Program who provided data for this study and also the staff of the British Oceanographic Data Centre for assistance with accessi...