Developing priority variables ('ecosystem Essential Ocean Variables' - eEOVs) for observing dynamics and change in Southern Ocean ecosystems

Reliable statements about variability and change in marine ecosystems and their underlying causes are needed to report on their status and to guide management. Here we use the Framework on Ocean Observing (FOO) to begin developing ecosystem Essential Ocean Variables (eEOVs) for the Southern Ocean Observing System (SOOS). An eEOV is a defined biological or ecological quantity, which is derived from field observations, and which contributes significantly to assessments of Southern Ocean ecosystems. Here, assessments are concerned with estimating status and trends in ecosystem properties, attribution of trends to causes, and predicting future trajectories. eEOVs should be feasible to collect at appropriate spatial and temporal scales and are useful to the extent that they contribute to direct estimation of trends and/or attribution, and/or development of ecological (statistical or simulation) models to support assessments. In this paper we outline the rationale, including establishing a set of criteria, for selecting eEOVs for the SOOS and develop a list of candidate eEOVs for further evaluation. Other than habitat variables, nine types of eEOVs for Southern Ocean taxa are identified within three classes: state (magnitude, genetic/species, size spectrum), predator–prey (diet, foraging range), and autecology (phenology, reproductive rate, individual growth rate, detritus). Most candidates for the suite of Southern Ocean taxa relate to state or diet. Candidate autecological eEOVs have not been developed other than for marine mammals and birds. We consider some of the spatial and temporal issues that will influence the adoption and use of eEOVs in an observing system in the Southern Ocean, noting that existing operations and platforms potentially provide coverage of the four main sectors of the region — the East and West Pacific, Atlantic and Indian. Lastly, we discuss the importance of simulation modelling in helping with the design of the observing system in the long term.