Many considerably declined fish populations have not fully recovered despite reductions in fishing pressure. One of the possible causes of impaired recovery is the (demographic) Allee effect. To investigate whether low-abundance recruitment dynamics can switch between compensation and depensation, the latter implying the presence of the Allee effect, we analysed the stock–recruitment time series of 17 depleted cod-type and flatfish populations using a Bayesian change point model. The recruitment dynamics were represented with the sigmoidal Beverton–Holt and the Saila–Lorda stock–recruitment models, allowing the parameters of the models to shift at a priori unknown change points. Our synthesis study questions the common assumption that recruitment is stationary and compensatory and the high amount of scatteredness often present in stock–recruitment data is only due to random variation. When a moderate amount of such variation was assumed, stock–recruitment dynamics were best explained by a non-stationary model for 53% of the populations, which suggests that these populations exhibit temporal changes in the stock–recruitment relationship. For four populations, we found shifts between compensation and depensation, suggesting the presence of temporary Allee effects. However, the evidence of Allee effects was highly dependent on the priors of the stock–recruitment model parameters and the amount of random variation assumed. Nonetheless, detection of changes in low-abundance recruitment is essential in stock assessment since such changes affect the renewal ability of the population and, ultimately, its sustainable harvest limits.

中文翻译：

---

枯竭的鲽鱼和比目鱼种群的非平稳招募动态中的临时 Allee 效应

尽管捕捞压力降低，但许多大幅下降的鱼类种群仍未完全恢复。恢复受损的可能原因之一是（人口）阿利效应。为了研究低丰度招募动态是否可以在补偿和剥夺之间切换，后者暗示阿利效应的存在，我们使用贝叶斯变化点模型分析了 17 个耗尽的鳕鱼型和比目鱼种群的种群招募时间序列。招聘动态用 sigmoidal Beverton-Holt 和 Saila-Lorda 股票招聘模型表示，允许模型的参数先验地变化未知的变化点。我们的综合研究质疑招聘是固定的和补偿性的普遍假设，并且股票招聘数据中经常出现的大量分散只是由于随机变化。当假设这种变化量适中时，53% 的人群的非平稳模型最好地解释了库存-招聘动态，这表明这些人群表现出库存-招聘关系的时间变化。对于四个人群，我们发现补偿和剥夺之间的转变，表明存在暂时的 Allee 效应。然而，Allee 效应的证据高度依赖于股票招募模型参数的先验和假设的随机变化量。尽管如此，