山东科学 ›› 2025, Vol. 38 ›› Issue (3): 25-33.doi: 10.3976/j.issn.1002-4026.2025036

• 海洋遥感 • 上一篇    下一篇

基于遥感监测的黑泥湾筏式海带养殖区时空变化研究

李欣1(), 郝增周2,3,4,*(), 厉运周3,4,5, 黄海清2,4, 潘德炉2,4   

  1. 1.浙江海洋大学 信息工程学院,浙江 舟山 316022
    2.自然资源部 第二海洋研究所 卫星海洋环境监测预警全国重点实验室,浙江 杭州 310012
    3.齐鲁工业大学(山东省科学院) 海洋仪器仪表研究所,山东 青岛 266061
    4.山东省科学院 山东省院士工作站,山东 济南 250014
    5.崂山实验室,山东 青岛 266237
  • 收稿日期:2025-04-08 出版日期:2025-06-20 发布日期:2025-06-26
  • 通信作者: 郝增周 E-mail:xin.lii@foxmail.com;hzyx80@sio.org.cn
  • 作者简介:李欣(2001—),女,硕士研究生,研究方向为农业渔业信息化。E-mail:xin.lii@foxmail.com
  • 基金资助:
    山东省重点研发计划(2023ZLYS01);国家重点研发计划(2022YFC3104200)

Spatiotemporal variation in raft-based kelp aquaculture in Heiniwan Bay using a remote sensing and monitoring technique

LI Xin1(), HAO Zengzhou2,3,4,*(), LI Yunzhou3,4,5, HUANG Haiqing2,4, PAN Delu2,4   

  1. 1. School of Information Engineering, Zhejiang Ocean University, Zhoushan 316022, China
    2. State Key Laboratory of Satellite Ocean Environment Dynamics, Second Institute of Oceanography, Ministry of Natural Resources, Hangzhou 310012, China
    3. Institute of Marine Instrumentation, Qilu University of Technology (Shandong Academy of Sciences), Qingdao 266061, China
    4. Shandong Provincial Academician Workstation, Shandong Academy of Sciences, Jinan 250014, China
    5. Laoshan Laboratory, Qingdao 266237, China
  • Received:2025-04-08 Online:2025-06-20 Published:2025-06-26
  • Contact: HAO Zengzhou E-mail:xin.lii@foxmail.com;hzyx80@sio.org.cn

摘要:

掌握近海海洋养殖规模及其分布变化有利于近海区域养殖管理、规划与生态保护。基于遥感手段,利用哨兵二号A/B(Sentinel-2A/2B)卫星影像数据,通过U-Net深度学习模型自动提取,并结合人机交互修正,准确获取了2016—2024年黑泥湾海带筏架养殖区的空间分布。分析结果表明:近9年来,黑泥湾海带养殖区面积呈现“技术驱动扩张—政策调控收缩—适应性波动”三阶段演变,空间分布呈现“南北集聚—中部管控”的稳定格局。研究揭示了黑泥湾筏式海带养殖区在技术革新、政策调控与自然条件耦合作用下的时空演变特征,为具有相似养殖模式与环境特征的海湾养殖区的养殖规划、动态调整与生态管理决策优化提供了科学依据。

关键词: 哨兵二号A/B卫星, 遥感, 筏式海带养殖, 深度学习, 黑泥湾, 时空演变

Abstract:

Accurate assessment of the scale and distribution of offshore marine aquaculture is critical for effective management, spatial planning, and ecological protection. This study employed high-resolution Sentinel-2A/2B satellite imagery, a U-Net deep learning model for automatic feature extraction, and human-computer interactive correction to map the spatial extent of raft-based kelp farming in Heiniwan Bay from 2016 to 2024. The analysis revealed a three-phase development trajectory in the aquaculture area over the nine-year period. Spatial distribution exhibited a stable “north-south agglomeration with central sparsity” pattern. The observed spatiotemporal dynamics reflect the combined influence of technological advancements, policy interventions, and natural environmental conditions. These findings offer a robust scientific basis for optimizing aquaculture zoning, adaptive management strategies, and ecological governance in coastal regions with comparable aquaculture practices and environmental settings.

Key words: Sentinel-2A/2B, remote sensing, raft-based kelp aquaculture, deep learning, Heiniwan Bay, spatiotemporal evolution

中图分类号: 

  • TP79

开放获取 本文遵循知识共享-署名-非商业性4.0国际许可协议(CC BY-NC 4.0),允许第三方对本刊发表的论文自由共享(即在任何媒介以任何形式复制、发行原文)、演绎(即修改、转换或以原文为基础进行创作),必须给出适当的署名,提供指向本文许可协议的链接,同时表明是否对原文作了修改,不得将本文用于商业目的。CC BY-NC 4.0许可协议详情请访问 https://creativecommons.org/licenses/by-nc/4.0