Information acquired by fishermen while fishing play a crucial role in their decision-making process. Those are used to form the basis for update of beliefs on productivity of a range of available fishing grounds while acting in an uncertain environment. In this paper, we focus on the formation of fishermen's beliefs about the productivity of fishing grounds based on signals obtained through harvest activity. We task ourselves with better understanding of the impact of prior experience on the at-sea day-to-day decision with respect to location choice. We investigate how obtained signals are propagated over time and space considering knowledge decay with increasing spatial and temporal separation, hence, incorporating intertemporal and spatial correlation between gained experience and future expectations. The parameters of the updating process are derived by fitting a dynamic Bayesian model to available data of location choices. The signal propagation parameters are derived by fitting the multinomial logit model to simulated expectations data paired with observed discrete location choices.
Our motivation is that successful fisheries management must be able to accurately predict the response of fishermen to fishery regulations with particular attention paid to flexible technologies where individuals can adjust effort and alter fishing behavior. Good understanding of fishermen location choice and location adjustment flexibility can contribute substantially to a design of management practices with spatial components. We apply our model to the Polish bottom trawler fleet in the Baltic Sea.