We investigated whether low-abundance recruitment dynamics can change in time between compensation and depensation, the latter implying the presence of the Allee effect. For this, we studied the stock-recruitment time series of 17 gadid and flatfish populations in the RAM Legacy Stock Assessment Database using a Bayesian change point model. The recruitment dynamics were represented with the sigmoidal Beverton-Holt and the Saila-Lorda stock-recruitment models, and the parameters of the models were allowed to shift at a priori unknown change points.

The stock-recruitment data were extracted from the RAM Legacy Stock Assessment Database. The data set contains the spawning stock biomass (SSB) and recruit time series of the studied populations.  

The S-R models were fitted to the data with the Bayesian online change point detection method (BOCPD), combined with simulation-based filtering. The data were divided into segments by calculating their most likely segmentation (MLS). The methods are described more in detail in the supplementary material of the article.

The folder RAMLegacy contains the main scripts for fitting the change point model to the empirical data. The folder method_validation contains scripts for simulating S-R data and fitting the change point model to the simulated time series for method validation. Consider parallel computing for efficiency.