| 基于不同阶数灰色系统模型的北太平洋柔鱼资源丰度预测 |
| |
| 引用本文： | 解明阳,陈新军.基于不同阶数灰色系统模型的北太平洋柔鱼资源丰度预测[J].上海海洋大学学报,2021,30(4):755-762. |
| |
| 作者姓名： | 解明阳 陈新军 |
| |
| 作者单位： | 上海海洋大学海洋科学学院,上海海洋大学海洋科学学院 |
| |
| 基金项目： | 国家自然科学基金项目（面上项目，重点项目，重大项目）,国家科技攻关计划 |
| |
| 摘 要： | 柔鱼（Ommastrephes bartramii）为短生命周期种类，是西北太平洋经济头足类之一。优化资源丰度预测模型能够更科学、有效地为渔业生产提供依据。本研究利用1998—2016年北太平洋柔鱼生产统计数据，采用GM（1，1）模型对不同时间长度的资源丰度（CPUE）进行分析，选择相对误差和方差最小的CPUE序列作为母序列，与太平洋年代际震荡指数 （PDO）、产卵场平均海表温度（SGSST）、育肥场平均海表温度（FGSST）、产卵场平均叶绿素浓度（SGC）、育肥场平均叶绿素浓度（FGC）等因子进行灰色关联分析，并以此分别建立6个不同阶数的灰色预测模型GM（0，N）模型和GM（1，N）模型]，筛选误差最小的模型作为预测柔鱼资源丰度的最佳模型。结果表明，以8年CPUE序列的建模为最佳，其平均相对误差最小，为6.28%；同时，GM（0，N）模型的预测精度普遍比GM（1，N）模型的要高，其中包含2月SGSST、10月FGSST、8月FGC和10月PDO的GM（0，5）模型为最优，拟合相对误差为3.87%，预测相对误差为1.18%，可作为预测北太平洋柔鱼资源丰度的最优模型。
|
| 关 键 词： | 柔鱼 资源丰度 时序选择 GM模型 |
| 收稿时间： | 2020/2/2 0:00:00 |
| 修稿时间： | 2020/4/13 0:00:00 |
Prediction of abundance index of Ommastrephes bartramii in the North Pacific Ocean based on different order grey system models |
| |
| XIE Mingyang,CHEN Xinjun.Prediction of abundance index of Ommastrephes bartramii in the North Pacific Ocean based on different order grey system models[J].Journal of Shanghai Ocean University,2021,30(4):755-762. |
| |
| Authors: | XIE Mingyang CHEN Xinjun |
| |
| Institution: | College of Marine Sciences of Shanghai Ocean University,Collaborative Innovation Center for Distant-water Fisheries,Shanghai,;College of Marine Sciences,Shanghai Ocean University |
| |
| Abstract: | Ommastrephes bartramii is a kind of short-lived species which is one of the economic cephalopods in the Northwest Pacific. Optimizing the resource abundance prediction model can provide a scientific and effective basis for fishery production. This study used the fishing data of neon flying squid from 1998 to 2016. Firstly,GM (1,1) models are established for resource abundance (CPUE) sequences of different time lengths. The CPUE sequence with the smallest relative error and variance is selected to perform grey correlation analysis with the environment and climate factors of the spawning and fattening grounds, including Pacific Decadal Oscillation Index (PDO), average sea surface temperature at spawning ground (SGSST), average sea surface temperature at fattening grounds (FGSST), average chlorophyll concentration at spawning ground(SGC), average chlorophyll concentration at fattening ground (FGC)to evaluate the importance of environmental factors. And based on the results, we established 6 grey prediction models of different ordersGM (0, N) model and GM (1, N) model]. Finally,we selected the model with relatively small fitting errors and prediction errors as the best model to predict the abundance of neon flying squid resources. The results show that the average relative error of the GM (1,1) model of the 8-year CPUE sequence is the smallest (6.28%). The prediction accuracy of the GM (0, N) models is generally higher than that of the GM (1, N) models. The GM (0,5) model 4 which included SGSST in February, FGSST in October, FGC in August, and PDO in October have the best model effects. Its relative error of fitting is 3.87%, and the relative error of prediction is 1.18%.Therefore,we suggested that this model can be used to forecast the resource abundance of neon flying squid. |
| |
| Keywords: | Ommastrephes bartramii resource abundance timing selection GM model |
|
| 点击此处可从《上海海洋大学学报》浏览原始摘要信息 |
| 点击此处可从《上海海洋大学学报》下载免费的PDF全文 |
|
