A new model for estimation of daily probability for the Pacific saury (Cololabis saira) encounter was proposed. The model performance was tested for the period of 2004–2018 (August–November) using the data from the Russian vessel monitoring system. The following physical oceanographic variables were used for encounter probability prediction: the absolute values and gradients (∇) of speed (V) of passive particles, imitating water parcels, and Lagrangian indicators. The positive effects on the encounter probability of saury were found for V, ∇V, and for the gradient of the finite-time Lyapunov exponent (∇Λ), while the effect of particle path length was negative. That means that saury preferred places close to the boundaries of the oceanographic features, where Lagrangian fronts are situated, but not inside the features themselves, because Λ is small in regular flows and large at Lagrangiam fronts. The model did not include information about years and volume of saury catches, but its monthly mean of catch probability in September had the highest correlation with Russian annual catches outside the national waters between Russia and Japan (r = 0.76, p = 0.001) and total annual catches there (r = 0.73, p = 0.002). Timeseries analysis of principle components (PC) from daily predictions of saury catch probabilities has also shown that the third PC correlated highly with the annual biomass of saury (r ≥ 0.8, p < 0.05). The model seems to be useful to manage Russian fishery and may help to explain the reasons for the saury biomass decline. The latter is very important to take into account for development of the stock assessment models.

一种估计太平洋秋刀鱼（ Cololabis saira）日概率的新模型) 相遇被提议。使用来自俄罗斯船舶监测系统的数据测试了 2004 年至 2018 年（8 月至 11 月）期间的模型性能。以下物理海洋学变量用于预测遭遇概率：被动粒子的速度（V）的绝对值和梯度（∇），模拟水包裹和拉格朗日指标。V, ∇V 和有限时间李雅普诺夫指数 (∇Λ) 的梯度对秋刀鱼的遭遇概率有积极影响，而粒子路径长度的影响是消极的。这意味着秋刀鱼更喜欢靠近海洋学特征边界的地方，即拉格朗日锋所在的位置，但不在特征本身内部，因为Λ在常规流动中较小，而在拉格朗日锋面较大。r  = 0.76，p  = 0.001）和那里的年总产量（r  = 0.73，p  = 0.002）。对秋刀鱼捕捞概率每日预测的主成分（PC）的时间序列分析也表明，第三个主成分与秋刀鱼的年生物量高度相关（r  ≥ 0.8，p  < 0.05）。该模型似乎对管理俄罗斯渔业很有用，可能有助于解释秋刀鱼生物量下降的原因。后者对于开发库存评估模型非常重要。