A DYNAMIC-MODEL FOR THE LIFE-HISTORY OF MAUROLICUS-MUELLERI, A PELAGIC PLANKTIVOROUS FISH

The life history and vertical distribution of a female cohort of the m esopelagic fish Maurolicus muelleri is simulated using stochastic dyna mic programming. The environment is represented by vertical profiles o f zooplankton biomass, light intensity and temperature, all variables changing with season. The fish physiology is modelled by dynamic state variables that represent structural fish weight, energetic state and the age of developing oocytes. The model is used to simulate optimal d epth distribution (feeding vs. predation risk) and energy allocation ( somatic growth or reproduction). The optimal strategies predicted by t he model depend on structural fish weight, energetic state and seasona l factors in the environment. The different strategies predicted for d ifferent size groups of fish are consistent with field observations of M. muelleri. Small fish give higher priority to growth and tolerate h igher levels of predation risk than large fish. The strategies of smal l fish seem to be little affected by changes in energetic state or sea sonal factors in the environment. On the other hand, the predicted str ategies of large fish are largely dependent on energetic state and sea sonal changes in the environment. In the winter they do not reproduce and minimize visual predation risk by staying at depths with a low lig ht intensity. The low light intensities also result in a low food inta ke and a negative energy budget in the winter months. In spring, summe r and autumn, the predicted strategy of large fish is to stay at depth s that provide feeding rates sufficient to rebuild energy reserves los t in the winter and to provide energy for reproduction and somatic gro wth.