The initial objective of the PELGAS integrated survey was to assess the biomass of small pelagic fish in the Bay of Biscay in spring. Sampling has been extended to study the spatial structure and dynamics of the pelagic ecosystem in springtime. The PELGAS survey has produced since 2000 long-term time-series of spatially-explicit data and indices, describing the main pelagic ecosystem components: hydrology, phytoplankton, mesozooplankton, fish and megafauna. In addition to small pelagic fish biomass estimates used for fish stock assessment, the survey now delivers two types of ecosystem products: standard gridded maps of ecosystem parameters, and a time series dataset of indicators of the Bay of Biscay pelagic ecosystem state. The evolution of the PELGAS survey and sampling sampling designs are presented and discussed, to highlight their impact on the survey ecosystem products. The potential of PELGAS survey products for ecosystem assessment is demonstrated with two case studies. Series of multivariate maps are first analysed to assess the pelagic ecosystem spatial structure. A map of consistent ecosystem seascapes boundaries and of their inter-annual variability in time is presented. Second, time series of potential ecosystem indices derived from the survey are jointly analysed to select the most continuous indicators within, and across, several ecosystem components. Trends in the selected time series are interpreted to derive information on marine ecosystem status and dynamics. Challenges and future directions are discussed, with a focus on the new big data flows that could be provided by the survey in a near future.