基于渔船捕捞行为特征的远洋延绳钓渔场捕捞强度计算

渔场捕捞强度信息可以为渔业资源评估和管理提供帮助。本研究结合2017年10—11月船舶自动监控系统(AutomaticIdentificationSystem,AIS)信息和同期中国中西太平洋延绳钓渔船捕捞日志数据,通过挖掘延绳钓渔船作业航速和航向特征,建立渔场作业状态识别模型,提取渔场捕捞强度信息。以3～9节为航速阈值和0°～10°及300°～360°为航向阈值,渔船作业状态识别准确率为68.29%。阈值识别和日志记录的捕捞强度信息在空间上相关性很高(0.96),基于AIS信息挖掘的渔船捕捞强度空间分布特征和实际非常相似。阈值识别和日志记录的捕捞强度信息在空间上与单位捕捞努力量渔获量(catch per unite of effort, CPUE)、渔获尾数、渔获重量和投钩数的空间相关系数均大于0.62,基于AIS信息挖掘的渔船空间捕捞强度也可替代用于渔业资源分析。

Calculating the fishing intensity of offshore longline fleets on fishing grounds based on their fishing characteristics

A better understanding of the behavior of offshore fishing fleets is required in order to prioritize and enforce fisheries management and conservation measures worldwide. Satellite-based Automatic Information Systems (S-AIS) are now commonly installed on most ocean-going vessels and have been suggested as a novel tool to explore the movements of fishing fleets in near-real time. The fishing behavior and effort of vessels determined by vessel speed data obtained from AIS could assist in fishery resources analysis. In this study we used AIS data extracted from exactEarth Shipview^TM and fishing log data of longline vessels in the Western and Central Pacific Ocean; both types of data were collected from October to November 2017 and were analyzed together in order to establish a vessel status recognition model by evaluating the speed and heading characteristics of longline fishing vessels. The fishing effort model was defined, and the fishing intensity information of the fishing grounds was calculated based on the output of the fishing activity identification model. In order to test the fishing effort rationality data extracted from AIS, the spatial correlation coefficients of the fishing intensity obtained from AIS data mining and the catch per unit effort (CPUE), the total number of tuna, the total catch weight, and hook numbers were calculated. Our results indicated that the speed of longline vessels was mostly between 3 to 9 knots while fishing. The heading ranges of longline vessels were between 0 to 10° and 300 to 360°. The fishing activity was classified based on the speed and heading of vessels; the accuracy of the fishing vessel status classification was 68.29%. The spatial correlation of fishing intensity between threshold classification and logging was high (>0.96, P<0.000001). The spatial distribution characteristics of the fishing intensity based on AIS were similar to the actual ones but lower than later. The spatial correlation coefficients of the fishing intensity obtained from AIS and CPUE data, the total number of tuna, the total catch weight, and hook numbers were all greater than 0.62 (P<0.00001). Data on the fishing intensity of fishing vessels obtained from AIS could provide high-resolution information for scientists and decision makers and could be used as alternative data in fisheries stock assessment and management.