Sources of uncertainty in ichthyoplankton surveys: modeling the influence of wind forcing and survey strategy on abundance estimates

A three-dimensional physical oceanographic model was used to simulate the temporal evolution of ichthyoplankton distributions in the Bornholm Basin, Baltic Sea. The Bornholm Basin is one of the major spawning grounds for cod (Gadus morhua callarias L.) and sprat (Sprattus sprattus balticus S.) in the Baltic Sea with a large number of historic survey data as well as material from ongoing investigations being available. The aim of the present study was to investigate the quantitative influence of different factors on abundance estimates of early life stages (eggs and larvae) of cod and sprat. Influences under consideration were advective losses, uncertainty due to non-sufficient spatial sampling resolution (e.g. number of stations) and effects of the sampling strategy (e.g. sequence of stations). The numerical simulations were performed for two time periods with contrasting meteorological forcing conditions. Errors in abundance estimates were highest at high wind speeds and when the organisms showed a patchy distribution. Advective losses were highest for the shallowest distributed life stages (cod and sprat larvae). Under strong west-wind conditions, losses due to transport out of the surveyed area reached up to >10% after 2 days. Different sampling strategies had no impact on overall accuracy. Errors due to sampling resolution amounted to 2–26% for the standard grid of 45 stations. The results suggest that wind forcing might serve as a first rough approximation for the reliability of historic abundance estimates.