PREDICTION OF VERTICAL-DISTRIBUTION AND AMBIENT DEVELOPMENT TEMPERATURE OF BALTIC COD, GADUS-MORHUA L., EGGS

An artificial neural network (ANN) model was established to predict th e vertical distribution of Baltic cod eggs. Data from vertical distrib ution sampling in the Bornholm Basin over the period 1986-1995 were us ed to train and test the network, while data sets from sampling in 199 6 were used for validation. The model explained 82% of the variance be tween observed and predicted relative frequencies of occurrence of the eggs in relation to salinity, temperature and oxygen concentration; T he ANN fitted all observations satisfactorily except for one sampling date, where an exceptional hydrographic situation was observed. Mean a mbient temperatures, calculated from the predicted vertical distributi ons of the eggs and used for the computation of egg developmental time s, were overestimated by 0.05 degrees C on average. This corresponds t o an error in prediction of egg developmental time of less than 1%.