| 基于神经网络的南太平洋长鳍金枪鱼渔场预报 |
| |
| 作者姓名： | ESMAEILZADEH Marjan KARBASSI Abdolrez MOATTAR Faramarz |
| |
| 作者单位： | 上海海洋大学 海洋科学学院, 上海 201306;中国水产科学研究院 南海水产研究所, 广东 广州 510300,上海海洋大学 海洋科学学院, 上海 201306;上海海洋大学 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 201306;国家远洋渔业工程技术研究中心, 上海 201306,中国水产科学研究院 南海水产研究所, 广东 广州 510300 |
| |
| 基金项目： | 海洋局公益性行业专项（20155014）;国家科技支撑计划（2013BAD13B01） |
| |
| 摘 要： | 南太平洋长鳍金枪鱼是我国远洋渔业的重点捕捞对象;对南太平洋长鳍金枪鱼进行准确的渔场预报;可以提高捕捞效率;提高渔业的生产能力。本研究根据1993-2010年南太平洋长鳍金枪鱼的延绳钓生产数据以及海洋卫星遥感数据（海水表面温度;SST;海面高度;SSH）和ENSO（El Niño-Southern Oscillation）指标;采用DPS（data processing system）数据处理系统中的BP人工神经网络模型;以渔获产量（单位时间的渔获尾数）和单位捕捞努力量渔获量（CPUE;Catch per unit of effort）分别作为中心渔场的表征因子;并作为BP模型的输出因子;以月、经度、纬度、SST、SSH和ENSO指标等作为输入因子;分别构建4-3-1;5-4-1;5-3-1;6-5-1;6-4-1;6-3-1等BP模型结构;比较渔场预报模型优劣。研究结果表明;以CPUE作为输出因子的BP人工神经网络结构总体上较优;其中以6-4-1模型结构为最优;相对误差只有0.006 41。研究认为;以CPUE为输出因子的6-4-1结构的人工神经网络模型;能够准确预报南太平洋长鳍金枪鱼的渔场位置。
|
| 关 键 词： | 长鳍金枪鱼|神经网络|CPUE|中心渔场|南太平洋 |
| 收稿时间： | 5/8/2016 12:00:00 AM |
| 修稿时间： | 2015/11/17 0:00:00 |
Assessment of metal pollution in the Anzali Wetland sediments using chemical partitioning method and pollution indices |
| |
| ESMAEILZADEH Marjan,KARBASSI Abdolrez,MOATTAR Faramarz.Assessment of metal pollution in the Anzali Wetland sediments using chemical partitioning method and pollution indices[J].Acta Oceanologica Sinica,2016,35(10):28-36. |
| |
| Authors: | ESMAEILZADEH Marjan KARBASSI Abdolreza and MOATTAR Faramarz |
| |
| Institution: | 1.Department of Environmental Science, Faculty of Environment and Energy, Tehran Science and Research Branch, Islamic Azad University, Tehran 1477893855, Iran2.Department of Environmental Engineering, Graduate Faculty of Environment, University of Tehran, Tehran 1417853111, Iran |
| |
| Abstract: | Albacore tuna (Thunnus alalunga) is an important target species for Chinese pelagic fishery in the South Pacific Ocean, and the accurate predicting of fishing ground can enhance catch efficiency and improve fishing operation. In this paper, we apply the BP artificial neural network model in the DPS (data processing system) to forecast fishing ground, according to the fishing yield statistics from longline, the sea surface temperature (SST), sea surface height (SST), and ENSO (El Niño-Southern Oscillation) index obtained by satellite remote sensing from 1993 to 2010. We consider the fishing yield (unit is the number of tuna) and CPUE (catch per unit of effort) as the character factors of fishing ground, the same as the output factors in BP artificial neural network respectively, and we take month, latitude, longitude, SST, SSH and ENSO index as the input factors, building the 4-3-1, 5-4-1, 5-3-1, 6-5-1, 6-4-1, 6-3-1 model and choosing the better results. It is found that the model used by CPUE is better than that model used by fishing yield, particularly the 6-4-1 model, and the relatively error is only 0.006 14, which can forecast the fishing ground of Thunnus alalunga in the South Pacific Ocean accurately. |
| |
| Keywords: | albacore tuna|artificial neural network|CPUE|fishing ground|South Pacific Ocean |
| 本文献已被 CNKI SpringerLink 等数据库收录！ |
| 点击此处可从《海洋学报(英文版)》浏览原始摘要信息 |
| 点击此处可从《海洋学报(英文版)》下载免费的PDF全文 |
|
