Abstract The interaction between hydrography, zooplankton and capelin larvae was studied during a 15-day survey in May 2001 in the southern part of the Barents Sea within the strategic programme Capelin and Herring in the Barents Sea Coexistence or Exclusion (BASECOEX). The spawning ground of capelin is found on shallow grounds close to the coast and after hatching the fish larvae are being transported offshelf towards the central part of the Barents Sea. The objective of the field study was to provide estimates of mortality and growth during the duration of the survey. Estimating vital rates requires sampling upon the same population over time. Areas feasible for studying these processes for capelin larvae would have to fulfil several conditions, such as moderate advection, high population numbers of larvae together with appropriate sampling techniques. This paper describes how a model approach may be used as a guideline for planning and executing a survey which has to meet the above conditions. Initially, the biological field was nowcasted with an Individual Based Model (IBM) at the start of the survey. The hydrodynamic model used in conjunction with the IBM required realistic weather forcing. Because of collecting and processing of the data, they are always lagging 1–2 days behind the current date. In order to solve the problem of nowcasting and forecasting we applied real data as far as we could and continued with forecasted weather for the next 5 days. This map was then utilized at sea as part of the information needed to decide where to place the sampling grid, and in the design of the layout of the grid. The model approach further enabled us to estimate the retention of water masses and consequently the capelin larvae and zooplankton populations within the region of study. The retention analysis shows that the advection in the selected area is significant, and after 2 weeks only a small fraction of the initial population is present in the original grid.