极地研究 ›› 2017, Vol. 29 ›› Issue (2): 218-227.DOI: 10.13679/j.jdyj.2017.2.218

• 研究论文 • 上一篇    下一篇

基于形态学及傅里叶分析对两种南极冰鱼耳石外型的研究

魏联1  朱国平1,2,3,4  刘子俊1   

  1. 1. 上海海洋大学海洋科学学院, 上海 201306
    2. 国家远洋渔业工程技术研究中心, 上海 210306
    3. 极地海洋生态系统研究室, 大洋渔业资源可持续开发省部共建教育部重点实验室, 上海海洋大学, 上海 201306
    4. 远洋渔业协同创新中心, 上海 201306
  • 收稿日期:2016-02-22 修回日期:2016-04-18 出版日期:2017-06-30 发布日期:2017-06-30
  • 通讯作者: 朱国平
  • 基金资助:

    公益性行业(农业)科研专项(201203018)、教育部留学回国人员科研启动基金项目和国家科技支撑计划项目(2013BAD 13B03)资助

Otolith shape of two icefish species based on morphology and Fourier analysis

Wei Lian1, Zhu Guoping1,2,3,4, Liu Zijun1   

  • Received:2016-02-22 Revised:2016-04-18 Online:2017-06-30 Published:2017-06-30
  • Contact: Guoping ZHU

摘要:

根据2014年3—4月以及2015年4—9月在南设得兰群岛以及南乔治亚群岛周边水域采集的裘氏鳄头冰鱼(Champsocephalus gunnari)及雪冰(Chionobathyscus dewitti)样本, 对其耳石进行5种基础形态参数测量并转换为7种形态学参数, 比较分析了两种冰鱼间的耳石形态学差异, 再利用傅里叶分析法选取两种耳石的77个傅里叶系数进行判别。结果表明, 两种冰鱼耳石长、高、周长、面积及质量均与体长呈显著的幂指数关系(P<0.01)。由形态指标分析可知, 裘氏鳄头冰鱼较雪冰耳石环率更低, 即更趋近于圆, 更为规则, 且耳石厚度上略薄。两者各项形态参数间均存在显著性差异(P<0.01)。对耳石77个傅里叶谐值进行主成分分析, 其中前20个主成分解释总变异的82.491%, 两种冰鱼的因子分布图上重叠量较少, 可见区分度较好, 判别分析选取了其中的6个傅里叶值建立了判别函数, 总体判别率为96.15%。总体而言, 可利用耳石外型对两种冰鱼进行种类判别, 傅里叶分析更为直观清晰且较为准确。本研究可为南极冰鱼耳石形态学研究提供基础数据, 并就其种类鉴别提供备选方法。

关键词: 耳石形态, 傅里叶分析, 裘氏鳄头冰鱼, 雪冰

Abstract:

Using samples collected from the South Shetland Islands and South Georgia during March–April 2014 and April–September 2015, this study compares the differences in otolith shape for two icefish species (Mackerel icefish Champsocephalus gunnari and crocodile icefish Chionobathyscus dewitti) using seven morphological parameters which transformed from five basic shape measurements. Population discrimination has also been conducted using Fourier analysis of the otoliths. Results show a significant relationship with power functions between otolith shape measurements (length, height, perimeter, area and weight) and standard length (P<0.01) in the two species. Based on the analysis of shape measurements, and compared to C. dewitti, the circularity of the otolith of Champsocephalus gunnari is lower, its shape is more circular and more regular, and it is less wide. The morphological parameters of the otoliths show a remarkable difference between the two species (P<0.01). We analyzed 77 standardized EFDs by principal component analysis (PCA); the first 20 principal components had 82.8% cumulative contribution. There was less overlap and good separation in the factor distribution picture. Six harmonics of 77 EFDs were entered in the stepwise discriminant analysis (SDA) with Fourier analysis, and the total correct classification was 92.02%. In general, otolith morphology can be used for population discrimination of these two icefish species but Fourier analysis provides more direct, clear and accurate data. The results of the present study provide basic information for otolith morphological analysis and population discrimination of these two Antarctic icefish species.

Key words: otolith morphology, Fourier analysis, Champsocephalus gunnari, Chionobathyscus dewitti