Forecasting Northwest Pacific Ocean neon flying squid abundance based on suitability of spawning and feeding grounds
-
摘要: 柔鱼(Ommastrephes bartramii)是北太平洋重要的经济头足类。短生命周期的特征使其资源丰度主要取决于补充量,早期生活史阶段的海洋环境将直接影响补充量大小。利用2004−2015年我国鱿钓船队在西北太平洋的生产统计数据,以及产卵场和索饵场海表水温,将产卵场和索饵场等分成不同数量的海区,运用相关性分析和随机森林模型,筛选出产卵期各月产卵场最适海表水温范围占总面积的比值(Ps)和索饵期各月索饵场最适海表水温范围占总面积的比值(Pf)与单位捕捞努力量渔获量(CPUE)显著相关的海区,将Ps值或Pf值作为神经网络模型的输入变量,分别构建基于产卵场、索饵场的资源丰度预报模型,分析模型的优劣与预报准确度。结果表明:产卵场划分方案为5°×5°,索饵场为2.5° (经)×4° (纬)较为合适。随机森林筛选出的海区与相关性分析筛选出的适宜海区范围大致吻合,且随机森林能够识别与CPUE相关的潜在海域。模型的预报结果表明,其预报准确度均达到90%以上,随机森林筛选出的海区的最适海表水温范围占该海区的比值作为神经网络模型输入因子构建的模型优于相关性分析,预报准确度更高。基于产卵场海域构建的模型相对于索饵场,模型精确度更高、更稳定。Abstract: Neon flying squid Ommastrephes bartramii is a cephalopod species of economically importance distributed in the North Pacific Ocean. Because of short lifespan, their abundances mainly depend on the recruitments and the marine environment during their early life stages will directly affect the recruitments. Using fishery data collected by Chinese squid-jigging fleets in the Northwest Pacific Ocean from 2004 to 2015 and the sea surface temperature (SST) of spawning and feeding grounds, which were divided into different numbers of subareas, correlation analysis and random forest model were used to screen out the subareas that CPUE has significant relationships with Ps (the proportion of favorable-SST in spawning grounds) of spawning grounds and Pf (the proportion of favorable-SST in feeding grounds) of feeding grounds during spawning and feeding periods. The Ps and Pf were used as input variables of the neural network model to forecast recruitments based on spawning ground and feeding ground, respectively, and the advantages and disadvantages of the model and forecast accuracy were analyzed. The results show that the schemes of dividing spawning ground into 5°×5° and feeding field into 2.5°(longitude)×4°(latitude) were optimal. The ranges of subareas selected by random forests are largely consistent with those selected by correlation analysis, both random forests and correlation analysis can identify potential subareas associated with CPUE, and both have forecasting accuracy of >90%. However, the subareas selected by the random forest are better and the forecast accuracy is higher than those selected by correlation analysis. In addition, the model based on the spawning ground is more accurate and stable than that based on feeding ground.
-
Key words:
- Northwast Pacific Ocean /
- Ommastrephes bartramii /
- abundance /
- random forests /
- neural network
-
表 1 2004−2015年各年份作业船数
Tab. 1 The numbers of fishing vessels from 2004 to 2015
年份 2004年 2005年 2006年 2007年 2008年 2009年 2010年 2011年 2012年 2013年 2014年 2015年 作业船数/艘 212 227 327 255 258 273 262 191 225 237 186 116 表 2 西北太平洋海域柔鱼产卵场Ps与CPUE时间序列的相关性和显著性
Tab. 2 The correlation and significance of Ps in the spawning grounds with CPUE time series of O. bartramii in the Northwest Pacific Ocean
方案 1月 2月 3月 4月 r p r p r p r p A − − − − − − − − B3 − − − − − − −0.660 <0.05 C5 0.600 <0.05 − − − − − − C6 − − 0.710 <0.05 − − − − D3 − − −0.670 <0.05 − − − − D9 0.650 <0.05 − − − − − − D11 − − 0.780 <0.01 − − − − D13 − − − − − − −0.600 <0.05 注:−表示该月份研究海域产卵场Ps与CPUE时间序列值相关性分析结果不显著。 表 3 西北太平洋海域柔鱼索饵场Pf与CPUE时间序列的相关性和显著性
Tab. 3 The correlation and significance of Pf in the feeding grounds with CPUE time series of O. bartramii in the Northwest Pacific Ocean
方案 8月 9月 10月 11月 r p r p r p r p A1 − − − − −0.580 <0.05 − − B1 − − − − −0.610 <0.05 0.590 <0.05 C1 − − − − −0.600 <0.05 − − C4 − − − − − − 0.840 <0.01 D1 − − − − −0.610 <0.05 − − D6 − − − − − − 0.750 <0.01 D7 −0.570 <0.05 − − − − 0.740 <0.01 D8 − − − − − − 0.710 <0.05 D10 − − − − − − 0.590 <0.05 注:−表示该月份研究海域索饵场Pf与CPUE时间序列值相关性分析结果不显著。 表 4 神经网络预报模型构建方案
Tab. 4 Construct the scenarios of neural network
海域 编号 输入因子 网络结构 相关性 产卵场海域 Ⅰ D9-1、D3-2、D11-2、D13-4 4∶5∶1 索饵场海域 Ⅱ D1-10、D6-11、D7-8、D8-11、D8-11、D10-11 6∶7∶1 随机森林 产卵场海域 Ⅲ D9-1、D10-1、D11-2、D13-4、D16-4 5∶6∶1 索饵场海域 Ⅳ D5-8、D7-8、D6-9、D1-10、D1-11、D6-11、D7-11、D8-11 8∶9∶1 -
[1] 王尧耕, 陈新军. 世界大洋性经济柔鱼类资源及其渔业[M]. 北京: 海洋出版社, 2005: 152−153.Wang Yaogeng, Chen Xinjun. The resource and biology of economic oceanic squid in the world[M]. Beijing: China Ocean Press, 2005: 152−153. [2] 陈新军, 田思泉, 许柳雄. 西北太平洋海域柔鱼产卵场和作业渔场的水温年间比较及与资源丰度的关系[J]. 上海水产大学学报, 2005, 14(2): 168−175.Chen Xinjun, Tian Siquan, Xu Liuxiong. Analysis on changes of surface water temperature in the spawning and feeding ground of Ommastrephes bartrami and its relationship with abundance index in the Northwestern Pacific Ocean[J]. Journal of Shanghai Fisheries University, 2005, 14(2): 168−175. [3] Chen Xinjun, Chen Yong, Tian Siquan, et al. An assessment of the west winter-spring cohort of neon flying squid (Ommastrephes bartramii) in the Northwest Pacific Ocean[J]. Fisheries Research, 2008, 92(2/3): 221−230. [4] Rodhouse P G. Managing and forecasting squid fisheries in variable environments[J]. Fisheries Research, 2001, 54(1): 3−8. doi: 10.1016/S0165-7836(01)00370-8 [5] Anderson C I H, Rodhouse P G. Life cycles, oceanography and variability: Ommastrephid squid in variable oceanographic environments[J]. Fisheries Research, 2001, 54(1): 133−143. doi: 10.1016/S0165-7836(01)00378-2 [6] 沈新强, 王云龙, 袁骐, 等. 北太平洋鱿鱼渔场叶绿素a分布特点及其与渔场的关系[J]. 海洋学报, 2004, 26(6): 118−123.Shen Xinqiang, Wang Yunlong, Yuan Qi, et al. Distributional characteristics of chlorophyll a and relation to the fishing ground in the squid fishing ground of the northern Pacific Ocean[J]. Haiyang Xuebao, 2004, 26(6): 118−123. [7] Boyle P R, Pierce G J, Hastie L C. Flexible reproductive strategies in the squid Loligo forbesi[J]. Marine Biology, 1995, 121(3): 501−508. doi: 10.1007/BF00349459 [8] 曹杰, 陈新军, 刘必林, 等. 鱿鱼类资源量变化与海洋环境关系的研究进展[J]. 上海海洋大学学报, 2010, 19(2): 232−239.Cao Jie, Chen Xinjun, Liu Bilin, et al. Review on the relationship between stock recruitment of squid and oceanographic environment[J]. Journal of Shanghai Oceans University, 2010, 19(2): 232−239. [9] Payne A G, Agnew D J, Pierce G J. Trends and assessment of cephalopod fisheries[J]. Fisheries Research, 2006, 78(1): 1−3. doi: 10.1016/j.fishres.2005.12.010 [10] 余为, 陈新军, 易倩, 等. 北太平洋柔鱼早期生活史研究进展[J]. 上海海洋大学学报, 2013, 22(5): 755−762.Yu Wei, Chen Xinjun, Yi Qian, et al. Review on the early life history of neon flying squid Ommastrephes bartramii in the North Pacific[J]. Journal of Shanghai Ocean University, 2013, 22(5): 755−762. [11] Murakami K, Watanabe Y, Nakara J. Growth, Distribution and Migration of Flying Squid (Ommastrephes bartramii) in the North Pacific[M]. Research Institute of North Pacific Fisheries, Hokkaido University, 1981: 161−179. [12] Nakamura Y. Distribution and maturity of neon flying squid, Ommastrephes bartrami (Lesueur), in the surrounding waters of Izu-Ogasawara Islands in spring[J]. Bulletin of Hokkaido National Fisheries Research Institute, 1988, 52: 139−149. [13] Okutain T. Systematics and distribution of larvae of Decapod Cephalopods collected from the sea surface on the Pacific Coast of Japan, 1960−1965[J]. Bulletin of the Tokai Regional Fisheries Research Laboratory, 1968, 55(2): 9−57. [14] Hayase S. Distribution of Spawning Grounds of Flying Squid, Ommastrephes bartrami, in the North Pacific Ocean[J]. Japan International Research Center for Agricultural Science, 1995, 29(1): 65−72. [15] Cao Jie, Chen Xinjun, Chen Yong. Influence of surface oceanographic variability on abundance of the western winter spring cohort of neon flying squid Ommastrephes bartramii in the NW Pacific Ocean[J]. Marine Ecology Progress Series, 2009, 381(5): 119−127. [16] Bower J R. Estimated paralarval drift and inferred hatching sites for Ommastrephes bartramii (Cephalopoda: Ommastrephidae) near the Hawaiian Archipelago[J]. Fishery Bulletin, 1996, 94: 398−411. [17] Bower J R, Ichii T. The red flying squid (Ommastrephes bartramii): A review of recent research and the fishery in Japan[J]. Fisheries Research, 2005, 76(1): 39−55. doi: 10.1016/j.fishres.2005.05.009 [18] Breiman L. Random forests[J]. Machine Learning, 2001, 45(1): 5−32. doi: 10.1023/A:1010933404324 [19] Gall J, Yao A, Razavi N, et al. Hough forests for object detection, tracking, and action Recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(11): 2188−2202. doi: 10.1109/TPAMI.2011.70 [20] 曹正凤. 随机森林算法优化研究[D]. 北京: 首都经济贸易大学, 2014.Cao Zhengfeng. Study on optimization of random forests algorithm[D]. Beijing: Capital University of Economics and Business, 2014. [21] 方匡南, 吴见彬, 朱建平, 等. 随机森林方法研究综述[J]. 统计与信息论坛, 2011, 26(3): 32−38. doi: 10.3969/j.issn.1007-3116.2011.03.006Fang Kuangnan, Wu Jianbin, Zhu Jianping, et al. A review of technologies on random forests[J]. Statistics & Information Forum, 2011, 26(3): 32−38. doi: 10.3969/j.issn.1007-3116.2011.03.006 [22] 李磊. 随机森林及数据可视化在棉蚜等级预测中的应用研究[D]. 泰安: 山东农业大学, 2017.Li Lei. Application of random forest and data visualization in prediction of cotton aphid grade[D]. Tai’an: Shandong Agricultural University, 2017. [23] 张万军. 基于大数据的个人信用风险评估模型研究[D]. 北京: 对外经济贸易大学, 2016.Zhang Wanjun. Research on individual credit risk evaluation model based on big data[D]. Beijing: University of International Business and Economics, 2016. [24] 孟杰, 李春林. 基于随机森林模型的分类数据缺失值插补[J]. 统计与信息论坛, 2016, 29(9): 86−90.Meng Jie, Li Chunlin. Missing data imputation for categorical data based on random forest model[J]. Statistics & Information Forum, 2016, 29(9): 86−90. [25] 李欣海. 随机森林模型在分类与回归分析中的应用[J]. 应用昆虫学报, 2013, 50(4): 1190−1197. doi: 10.7679/j.issn.2095-1353.2013.163Li Xinhai. Using “random forest” for classification and regression[J]. Chinese Journal of Applied Entomology, 2013, 50(4): 1190−1197. doi: 10.7679/j.issn.2095-1353.2013.163 [26] Benediktsson J A, Swain P H, Ersoy O K. Neural network approaches versus statistical methods in classification of multisource remote sensing data[J]. IEEE Transactions on Geoscience and Remote Sensing, 1990, 28(4): 540−552. doi: 10.1109/TGRS.1990.572944 [27] Hush D R, Horne B G. Progress in supervised neural networks[J]. IEEE Signal Processing Magazine, 1993, 10(1): 8−39. doi: 10.1109/79.180705 [28] 汪金涛, 高峰, 雷林, 等. 基于产卵场环境因子的阿根廷滑柔鱼资源补充量预报模型研究[J]. 海洋学报, 2014, 36(12): 119−124.Wang Jintao, Gao Feng, Lei Lin, et al. Study on forecasting model of recruitment for Illex argentinus by using the environmental factors in the spawning ground[J]. Haiyang Xuebao, 2014, 36(12): 119−124. [29] 金龙, 况雪源, 黄海洪, 等. 人工神经网络预报模型的过拟合研究[J]. 气象学报, 2004, 62(1): 62−70.Jin Long, Kuang Xueyuan, Huang Haihong, et al. Study on the overfitting of the artificial neural network forecasting model[J]. Acta Meteorologica Sinica, 2004, 62(1): 62−70. [30] Young R E, Hirota J. Description of Ommastrephes bartramii (Cephalopoda: Ommastrephidae) paralarvae with evidence for spawning in Hawaiian waters[J]. Pacific Science, 1990, 44(1): 71−80.