Zooplankton are major primary consumers and predators in most aquatic ecosystems. They exhibit tremendous diversity of traits, ecological strategies and, consequently, impacts on other trophic levels and the cycling of materials and energy. An adequate representation of this diversity in community and ecosystem models is necessary to generate realistic predictions on the functioning of aquatic ecosystems but remains extremely challenging. We propose that the use of trait-based approaches is a promising way to reduce complexity while retaining realism in developing novel descriptions of zooplankton in ecosystem models. Characterizing zooplankton traits and trade-offs will also be helpful in understanding the selection pressures and diversity patterns that emerge in different ecosystems along major environmental gradients. Zooplankton traits can be characterized according to their function and type. Some traits, such as body size and motility, transcend several functions and are major determinants of zooplankton ecological strategies. Future developments of trait-based approaches to zooplankton should assemble a comprehensive matrix of key traits for diverse groups and explore it for general patterns; develop novel predictive models that explicitly incorporate traits and associated trade-offs; and utilize these traits to explain and predict zooplankton community structure and dynamics under different environmental conditions, including global change scenarios. © 2013 The Author.

Abstract. The daily vertical migrations of fish and other metazoans actively transport organic carbon from the ocean surface to depth, contributing to the biological carbon pump. We use an oxygen-constrained, game-theoretic food-web model to simulate diel vertical migrations and estimate near-global (global ocean minus coastal areas and high latitudes) carbon fluxes and sequestration by fish and zooplankton due to respiration, fecal pellets, and deadfalls. Our model provides estimates of the carbon export and sequestration potential for a range of pelagic functional groups, despite uncertain biomass estimates of some functional groups. While the export production of metazoans and fish is modest (∼20 % of global total), we estimate that their contribution to carbon sequestered by the biological pump (∼800 PgC) is conservatively more than 50 % of the estimated global total (∼1300 PgC) and that they have a significantly longer sequestration timescale (∼250 years) than previously reported for other components of the biological pump. Fish and multicellular zooplankton contribute about equally to this sequestered carbon pool. This essential ecosystem service could be at risk from both unregulated fishing on the high seas and ocean deoxygenation due to climate change.

Advances in sampling and observation are described including new optical and acoustical systems together with animal tagging and telemetry. Shipboard, laboratory, and in situ studies of zooplankton feeding, reproduction and development, zooplankton and larval fish growth, and zooplankton mortality are reviewed. The relation between individual zooplankton behaviours, such as feeding, predator escape and mate-finding, are considered in relation to population processes. Methods of retrospective data analysis applied in retrospective studies on past ecosystem states are described. Finally, potential future directions for marine ecosystem research using observational and experimental approaches are discussed.

Plankton imaging systems supported by automated classification and analysis have improved ecologists' ability to observe aquatic ecosystems. Today, we are on the cusp of reliably tracking plankton populations with a suite of lab-based and in situ tools, collecting imaging data at unprecedentedly fine spatial and temporal scales. But these data have potential well beyond examining the abundances of different taxa; the individual images themselves contain a wealth of information on functional traits. Here, we outline traits that could be measured from image data, suggest machine learning and computer vision approaches to extract functional trait information from the images, and discuss promising avenues for novel studies. The approaches we discuss are data agnostic and are broadly applicable to imagery of other aquatic or terrestrial organisms.