摘要
为了应对近海渔业资源日益衰退的问题,为实施基于空间的渔业管理提供必要信息,以江苏南部海域小黄鱼为研究对象,根据2019—2022年在该海域进行的季节性渔业资源综合调查,结合5个生物和非生物因子,采用组合物种分布模型(Ensemble species distribution model, ESDM),研究了该海域小黄鱼的空间分布特征及其主要影响因素。结果显示,相较于单一物种分布模型,组合物种分布模型具有更高的AUC值(春季:0.995±0.002;秋季:0.985±0.001)和TSS值(春季:0.935±0.038;秋季:0.903±0.029)。在春季,底层水温和底层盐度的重要性水平最高(0.40和0.38),而在秋季叶绿素a和饵料生物丰度对小黄鱼空间分布的影响较大,其重要性分别为0.53和0.46。春季小黄鱼主要分布在近岸浅海区域,整体呈条带状分布;秋季小黄鱼则主要分布于水深较深的远岸水域,且适宜分布的海域范围要大于春季,整体呈块状分布。此外,小黄鱼的空间分布特征亦呈现出明显的年际差异,例如在2021年,其适宜栖息地面积明显小于其他年份,分布范围也仅限于局部区域。研究表明,组合物种分布模型具有更优的预测性能,能够更好地反映小黄鱼的栖息分布特征及其影响因素;不同季节小黄鱼的适生区及影响因素各有差异。本研究可为揭示该海域小黄鱼的时空分布特征及其变化规律提供理论依据,为实施基于空间的渔业管理和保护区选划提供基础资料。
近年来,受到气候变化、过度捕捞、环境污染等自然和人为因素的共同影响,全球范围内的海洋生态系统普遍严重退化,由此出现了渔获量下降,生物多样性水平降低,渔获个体普遍低龄化、小型化等现
物种分布模型(Species distribution model, SDM)是预测物种潜在空间分布的有力工具,近年来已成为评估海洋生物的迁移路径和空间管理的重要手
小黄鱼(Larimichthys polyactis)隶属于鲈形目(Perciformes)石首鱼科(Sciaenidae)黄鱼属(Larimichthys),为暖温性底层鱼类,广泛分布于中国的黄渤海、东海以及朝鲜半岛以西海域,主要生活在26°N以北,126.5°E以西,水深40~80 m的近岸泥沙质海区,是底拖网等网具常见的专捕和兼捕对
本研究以江苏南部海域小黄鱼为研究对象,根据2019—2022年春季和秋季在该海域进行的渔业资源调查数据,基于组合物种分布模型(ESDM)研究了江苏南部海域小黄鱼的栖息分布特征及主要影响因素,旨在揭示该海域小黄鱼的时空分布特征及其变化规律,为实施基于空间的渔业管理和保护区优化提供基础资料。
样品采自2019—2022年春季(3—4月)和秋季(9—10月)在江苏南部海域(121.0°E~122.5°E,31.5°N~33.6°N)进行的季节性渔业资源底拖网调查,共8个航次。依据实际海况设置21个调查站位,各站位的分布如

图1 江苏南部海域采样站位分布图
Fig.1 Distribution of sampling stations in the southern coastal waters of Jiangsu Province
由于小黄鱼为底层鱼类,根据这一属性并结合相关历史文
为避免各影响因子之间存在多重共线性,采用方差膨胀因子(Variance inflation factor, VIF)对上述因子进行多重共线性分析,VIF表示变量回归参数的置信区间能膨胀为与模型无关的预测变量的程度。一般认为,若VIF大于10,则表明该因子存在多重共线性,需要剔
本研究构建的组合物种分布模型包含了10种常用的模型,即广义线性模型(Generalized linear model, GLM)、广义加性模型(Generalized additive model, GAM)、随机森林(Random forest, RF)模型、最大熵模型(Maximum entropy model, MaxEnt)、人工神经网络(Artificial neural network, ANN)、广义增强回归模型(Generalized boosted regression model, GBM)、分类树分析(Classification tree analysis, CTA)、表面分布区分室(Surface range envelope, SRE)模型、多元自适应回归样条(Multivariate adaptive regression spline, MARS)和弹性判别分析(Flexible discriminant analysis, FDA),使用biomod2软件包在R4.3.1环境中对这些模型进行参数拟
通过交叉验证评估各模型的预测性
Wi= | (1) |
式中:Wi为第i个单模型的权重;ri为第i个单模型的AUC值;h为AUC>0.8的单模型的个数。
不同因子对小黄鱼分布的影响各有差异,为反映各因子与响应变量的相关程度,需要对影响因子的重要性进行排序。本研究通过biomod2软件包中的get_variables_importance函数计算各因子置换后均方根误差损失(Root mean square error loss after permutation)的数值来评价因子的重要性,均方根误差损失越大,表明该因子对小黄鱼空间分布的影响越
根据各年份、各季节的影响因子数据,在目标海域以0.01°×0.01°为单位进行网格划分,记录每个网格中心点的经纬度坐标值,使用克里金插值法对各网格中心点的环境和生物因子数据进行填充,将插值获得的数据代入已构建的组合物种分布模型中。模型通过内嵌的栖息地适宜性指数(Habitat suitability index, HSI)计算得到各年份、各季节小黄鱼在各网格点的HSI,HSI>0.7的海域是小黄鱼的适宜栖息地,最后使用surfer软件绘制小黄鱼的HSI分布图。HSI计算公式如
IHSI= | (2) |
式中:IHSI为栖息地适宜性指数HSI;Wi为第i个影响因子的权重;ISI⁃i为第i个影响因子的适宜性指数;j为影响因子的数量。
通过VIF检验各因子间的多重共线性水平,结果表明,春、秋季各影响因子的VIF值为1.06~5.05(VIF<10,
季节 Season | 底层水温 SBT | 底层盐度 SBS | 水深 Depth | 叶绿素a Chl.a | 饵料生物 Prey |
---|---|---|---|---|---|
春季 Spring | 2.46 | 4.83 | 1.62 | 2.25 | 1.10 |
秋季 Autumn | 1.15 | 5.05 | 1.48 | 4.27 | 1.06 |
通过100次交叉验证得到 biomod2 中10个单一物种分布模型在各季节的AUC值和TSS值(图

图2 江苏南部海域春季小黄鱼10个单一物种分布模型的AUC值和TSS值
Fig.2 Comparison of AUC and TSS of 10 single models for L. polyactis in the southern waters of Jiangsu Province in spring

图3 江苏南部海域秋季小黄鱼10个单一物种分布模型的AUC值和TSS值
Fig.3 Comparison of AUC and TSS of 10 single models for L. polyactis in the southern waters of Jiangsu Province in autumn
结果表明(
季节 Season | 用于集成的单一模型 Ensembled single models | 集成模型的AUC值 Value of AUC for the ensembled model | 集成模型的TSS值 Value of TSS for the ensembled model |
---|---|---|---|
春季 Spring | GAM、RF | 0.995±0.002 | 0.935±0.038 |
秋季 Autumn | GBM、GAM、RF | 0.985±0.001 | 0.903±0.029 |
通过构建组合物种分布模型分析了5个因子对各季节小黄鱼空间分布的重要性以及响应曲线(图

图4 江苏南部海域小黄鱼空间分布各影响因子的重要性
Fig.4 Importance of influencing factors in L. polyactis distribution in the southern waters of Jiangsu Province

图5 江苏南部海域春季小黄鱼空间分布5个影响因子的响应曲线
Fig.5 Response curves for five influencing factors for the spatial distribution of L. polyactis in the southern waters of Jiangsu Province in spring
黑色散点表示小黄鱼的栖息地适宜性指数(HSI);灰色阴影部分表示95%置信区间;红色曲线表示各因子对小黄鱼栖息地适宜性的响应曲线。
The black scatters represent the habitat suitability index (HSI) of L. polyactis; the gray shaded portions represent the 95% confidence interval; the red curves represent the response curve of each factor to the habitat suitability of L. polyactis.

图6 江苏南部海域秋季小黄鱼空间分布5个影响因子的响应曲线
Fig.6 Response curves for five influencing factors for the spatial distribution of L. polyactis in the southern waters of Jiangsu Province in autumn
黑色散点表示小黄鱼的栖息地适宜性指数(HSI);灰色阴影部分表示95%置信区间;红色曲线表示各因子对小黄鱼栖息地适宜性的响应曲线。
The black scatters represent the habitat suitability index (HSI) of L. polyactis; the gray shaded portions represent the 95% confidence interval; the red curves represent the response curve of each factor to the habitat suitability of L. polyactis.
基于组合物种分布模型,绘制江苏南部海域小黄鱼在2019—2022年春季和秋季的栖息地适宜性分布图(

图7 2019—2022年春季江苏南部海域小黄鱼的栖息地适宜性(HSI)分布图
Fig.7 Distribution of HSI for L. polyactis during spring from 2019 to 2022 in the southern waters of Jiangsu Province

图8 2019—2022年秋季江苏南部海域小黄鱼的栖息地适宜性(HSI)分布图
Fig.8 Distribution of HSI for L. polyactis during autumn from 2019 to 2022 in the southern waters of Jiangsu Province
采用合适的模型是准确预测物种分布的前提,对于渔业资源的养护和管理具有重要意
然而尽管如此,经筛选得到的单一物种分布模型仍会因预测结果与真实结果的差异较大、过度拟合等而导致预测的精度较低。此外,单一模型常常只考虑部分因素对物种分布的影响,难以综合多种因素进行分析,这有可能无法反映物种分布的真实情
海洋是鱼类赖以生存的空间,通过海洋中水文环境要素信息的提取,可为鱼类空间分布的环境偏好分析提供数据支
饵料生物的分布也会影响鱼类对生境的选择,充足的饵料供应亦可以在一定程度上弥补水温、盐度等外界环境条件的波动造成的不利影
研究发现,江苏南部海域小黄鱼的分布特征存在明显的季节差异。在春季,小黄鱼主要分布在底层水温14~16.5 ℃、盐度5~27、水深3~20 m的苏南近岸浅海域;在秋季,则多分布于水深26~50 m的海域。这与小黄鱼的生态习性密切相关,春季为小黄鱼的产卵繁殖期,产卵场位于32°00′N~33°15′N和121°30′E~122°15′E之间的长江口以北邻近海域,该海域水深较浅,且受陆地径流影响,盐度偏低,为小黄鱼产卵繁殖提供了理想的环境条
目前,“吕泗渔场小黄鱼银鲳国家级水产种质资源保护区”的设立,对于江苏近海小黄鱼资源的保护效果显著,有利于保证小黄鱼生长、发育及繁殖,使其资源在一定程度上得到恢复。本研究通过识别江苏南部近海小黄鱼的适宜栖息地及其年际变化,可为现有保护区的评估提供科学依据。此外,随着全球气候变化日益加剧,极有可能导致受保护物种的栖息地发生变化,进而威胁到受保护的物
参考文献
ZHANG C L, CHEN Y, XU B D, et al. How to predict biodiversity in space? An evaluation of modelling approaches in marine ecosystems[J]. Diversity and Distributions, 2019, 25(11): 1697-1708. [百度学术]
KARNAUSKAS M, WALTER III J F, KELBLE C R, et al. To EBFM or not to EBFM? That is not the question[J]. Fish and Fisheries, 2021, 22(3): 646-651. [百度学术]
DYDERSKI M K, PAŹ S, FRELICH L E, et al. How much does climate change threaten European forest tree species distributions? [J]. Global Change Biology, 2018, 24(3): 1150-1163. [百度学术]
ZHANG Y L, ZHANG C L, XU B D, et al. Impacts of trophic interactions on the prediction of spatio-temporal distribution of mid-trophic level fishes[J]. Ecological Indicators, 2022, 138: 108826. [百度学术]
ZHANG Z X, MAMMOLA S, XIAN W W, et al. Modelling the potential impacts of climate change on the distribution of ichthyoplankton in the Yangtze Estuary, China[J]. Diversity and Distributions, 2020, 26(1): 126-137. [百度学术]
DU J G, DING L K, SU S K, et al. Setting conservation priorities for marine sharks in China and the Association of Southeast Asian Nations (ASEAN) seas: What are the benefits of a 30% conservation target? [J]. Frontiers in Marine Science, 2022, 9: 933291. [百度学术]
ZIMMER S N, HOLSINGER K W, DAWSON C A. A field-validated ensemble species distribution model of Eriogonum pelinophilum, an endangered subshrub in Colorado, USA[J]. Ecology and Evolution, 2023, 13(12): e10816. [百度学术]
刘静, 陈咏霞, 马琳. 黄渤海鱼类图志[M]. 北京: 科学出版社, 2015. [百度学术]
LIU J, CHEN Y X, MA L. Fishes of the Bohai Sea and Yellow Sea[M]. Beijing: Science Press, 2015. [百度学术]
ZHANG Y L, XU B D, JI Y P, et al. Comparison of habitat models in quantifying the spatio-temporal distribution of small yellow croaker (Larimichthys polyactis) in Haizhou Bay, China[J]. Estuarine, Coastal and Shelf Science, 2021, 261: 107512. [百度学术]
YANG W, HU W J, CHEN B, et al. Impact of climate change on potential habitat distribution of Sciaenidae in the coastal waters of China[J]. Acta Oceanologica Sinica, 2023, 42(4): 59-71. [百度学术]
王凯, 章守宇, 汪振华, 等. 马鞍列岛海域小黄鱼的食性[J]. 水生生物学报, 2012, 36(6): 1188-1192. [百度学术]
WANG K, ZHANG S Y, WANG Z H, et al. Feeding habits of small yellow croaker (Larimichthys polyactis) off Ma’an archipelago[J]. Acta Hydrobiologica Sinica, 2012, 36(6): 1188-1192. [百度学术]
RODRIGUES M, DE LA RIVA J, FOTHERINGHAM S. Modeling the spatial variation of the explanatory factors of human-caused wildfires in Spain using geographically weighted logistic regression[J]. Applied Geography, 2014, 48: 52-63. [百度学术]
张欣雨, 朱泽群, 袁雅欣, 等. 基于组合物种分布模型的黄河源区鹅绒委陵菜适宜生境及其对气候变化的响应[J]. 草业科学, 2022, 39(2): 254-267. [百度学术]
ZHANG X Y, ZHU Z Q, YUAN Y X, et al. Assessment of suitable Potentilla anserina habitat and its response to climate change in the source region of the Yellow River based on ensemble species distribution modeling[J]. Pratacultural Science, 2022, 39(2): 254-267. [百度学术]
TANAKA K, CHEN Y. Spatiotemporal variability of suitable habitat for American Lobster (Homarus americanus) in Long Island Sound[J]. Journal of Shellfish Research, 2015, 34(2): 531-543. [百度学术]
BHAT I A, FAYAZ M, ROOF-UL-QADIR, et al. Predicting potential distribution and range dynamics of Aquilegia fragrans under climate change: insights from ensemble species distribution modelling[J]. Environmental Monitoring and Assessment, 2023, 195(5): 623. [百度学术]
DREW C A, WIERSMA Y F, HUETTMANN F. Predictive species and habitat modeling in landscape ecology[M]. New York: Springer, 2011: 195-196. [百度学术]
EKUNDAYO T C, IJABADENIYI O A, IGBINOSA E O, et al. Using machine learning models to predict the effects of seasonal fluxes on Plesiomonas shigelloides population density[J]. Environmental Pollution, 2023, 317, 120734. [百度学术]
陈晓琳, 纪云龙, 李鹏程, 等. 基于组合物种分布模型的海州湾矛尾虾虎鱼空间分布特征及其影响因素研究[J]. 中国水产科学, 2024, 31(3): 343-355. [百度学术]
CHEN X L, JI Y L, LI P C, et al. Spatial distribution of Chaeturichthys stigmatias and influence factors in Haizhou Bay based on ensemble species distribution model[J]. Journal of Fishery Sciences of China, 2024, 31(3): 343-355. [百度学术]
GRITTI E S, GAUCHEREL C, CRESPO-PEREZ M V, et al. How can model comparison help improving species distribution models? [J]. PLoS One, 2013, 8(7): e68823. [百度学术]
QUESADA-RUIZ L C, RODRIGUEZ-GALIANO V F, ZURITA-MILLA R, et al. Area and feature guided Regularised random forest: a novel method for predictive modelling of binary phenomena. The case of illegal landfill in Canary Island[J]. International Journal of Geographical Information Science, 2022, 36(12): 2473-2495. [百度学术]
RAJKOVIĆ D, JEROMELA A M, PEZO L, et al. Artificial neural network and random forest regression models for modelling fatty acid and tocopherol content in oil of winter rapeseed[J]. Journal of Food Composition and Analysis, 2023, 115: 105020. [百度学术]
BUSBY J R. BIOCLIM: a bioclimate analysis and prediction system[J]. Plant Protection Quarterly, 1991, 6(1): 8-9. [百度学术]
MATHUR M, MATHUR P. Global distribution modelling of Macrophomina Phaseolina (tassi) goid: a comparative assessment using ensemble machine learning tools[J]. Australasian Plant Pathology, 2023, 52(4): 353-371. [百度学术]
RAMIREZ-REYES C, STREET G, VILELLA F J, et al. Ensemble species distribution model identifies survey opportunities for at-risk bearded beaksedge (Rhynchospora crinipes) in the southeastern United States[J]. Natural Areas Journal, 2021, 41(1): 55-63. [百度学术]
李国东, 李冬佳, 熊瑛, 等. 基于GAM的黄海南部越冬小黄鱼资源丰度与环境因子关系[J]. 海洋渔业, 2023, 45(4): 403-411. [百度学术]
LI G D, LI D J, XIONG Y, et al. Relationship between environmental factors and abundance of overwintering Larimichthys polyactis in the southern Yellow Sea based on GAM[J]. Marine Fisheries, 2023, 45(4): 403-411. [百度学术]
XUE Y, GUAN L S, TANAKA K, et al. Evaluating effects of rescaling and weighting data on habitat suitability modeling[J]. Fisheries Research, 2017, 188: 84-94. [百度学术]
李子东, 王燕平, 仲霞铭, 等. 江苏海域小黄鱼时空分布及生物学特征研究[J]. 海洋渔业, 2023, 45(1): 73-85. [百度学术]
LI Z D, WANG Y P, ZHONG X M, et al. Spatio-temporal and spatial distribution and biological characteristics of Larimichthys polyactis in Jiangsu sea area[J]. Marine Fisheries, 2023, 45(1): 73-85. [百度学术]
邹易阳, 薛莹, 麻秋云, 等. 应用栖息地适宜性指数研究海州湾小黄鱼的空间分布特征[J]. 中国海洋大学学报(自然科学版), 2016, 46(8): 54-63. [百度学术]
ZOU Y Y, XUE Y, MA Q Y, et al. Spatial Distribution of Larimichthys polyactis in Haizhou Bay Based on Habitat Suitability Index[J]. Periodical of Ocean University of China, 2016, 46(8): 54-63. [百度学术]
王雅丽, 王晶, 周永东, 等. 基于two-stage GAM的舟山渔场及邻近海域小黄鱼时空分布特征[J]. 中国水产科学, 2022, 29(4): 633-641. [百度学术]
WANG Y L, WANG J, ZHOU Y D, et al. Spatial and temporal distribution characteristics of Larimichthys polyactis in Zhoushan fishing ground and the adjacent waters based on two-stage GAM[J]. Journal of Fishery Sciences of China, 2022, 29(4): 633-641. [百度学术]
ZERBINI A N, FRIDAY N A, PALACIOS D M, et al. Baleen whale abundance and distribution in relation to environmental variables and prey density in the eastern Bering Sea[J]. Deep Sea Research Part Ⅱ: Topical Studies in Oceanography, 2016, 134: 312-330. [百度学术]
宁修仁, 刘子琳, 史君贤. 渤、黄、东海初级生产力和潜在渔业生产量的评估[J]. 海洋学报, 1995, 17(3): 72-84. [百度学术]
NING X R, LIU Z L, SHI J X. Assessment of primary productivity and potential fishery production in the Bohai, Yellow, and East China Seas[J]. Acta Oceanologica Sinica, 1995, 17(3): 72-84. [百度学术]
李建生, 严利平, 李惠玉, 等. 黄海南部、东海北部夏秋季小黄鱼数量分布及与浮游动物的关系[J]. 海洋渔业, 2007, 29(1): 31-37. [百度学术]
LI J S, YAN L P, LI H Y, et al. On the relationship between quantity distribution of small yellow croaker (Larimichthys polyactis Bleeker) and zooplankton in southern Yellow Sea and the northern East China Sea in summer and autumn[J]. Marine Fisheries, 2007, 29(1): 31-37. [百度学术]
YU F, REN Q, DIAO X Y, et al. The sandwich structure of the southern yellow sea cold water mass and Yellow Sea warm current[J]. Frontiers in Marine Science, 2022, 8: 767850. [百度学术]
傅晓婷. 渤黄海浮游植物群落结构的季节变化[D]. 天津: 天津科技大学, 2021. [百度学术]
FU X T. Seasonal variations of phytoplankton community structure in the Bohai Sea and the Yellow Sea[D]. Tianjin: Tianjin University of Science and Technology, 2021. [百度学术]
GILMOUR M E, ADAMS J, BLOCK B A, et al. Evaluation of MPA designs that protect highly mobile Megafauna now and under climate change scenarios[J]. Global Ecology and Conservation, 2022, 35: e02070. [百度学术]
CHEUNG W W L, WATSON R, PAULY D. Signature of ocean warming in global fisheries catch[J]. Nature, 2013, 497(7449): 365-368. [百度学术]