PS 65-84
Accuracy in detection and measurement of density-dependence in marine fish stocks: A simulation framework
Fish productivity is often used as an index in stock assessment to set robust management and conservation plans. Fishery scientists are attempting to better understand the underlying mechanisms shaping productivity. One of these mechanisms is a density dependence process defined as overcompensation. Overcompensation occurs at high abundance at the population or metapopulation scale. Although overcompensation has been identified in many species there is very little if any statistical evidence of this mechanism being detected in marine populations. This study uses simulations to test a method of detecting and measuring overcompensation. Data are simulated with process error and overcompensation varying from low to high levels using parametric relationships (e.g. Shepherd relationship). Then, we fit a model to the data and assessed the accuracy in detection and measurement of overcompensation.
Results/Conclusions
The study reveals several scenarios under witch overcompensation can be detected fairly accurately. The efficiency of detection depends on life history traits, levels of uncertainty in the data and levels of overcompensation. The detection accuracy is high under low uncertainty (CV <0.2) and with high overcompensation (>1.5). This project will hopefully lead to better understanding of an important ecological process and may lead to a reliable method to detect overcompensation with accuracy. The next step of this study would be to determine the intensity and frequency of overcompensation in marine fish stocks using a meta-analysis.