基于海岸带成像仪CZI的连云港紫菜养殖动态监测  

Monitoring of Porphyra cultivation dynamics in Lianyungang based on coastal zone imager

作　　者：王鑫华 刘建强[3] 邢前国[1,2] 陈艳拢 WANG Xin-hua;LIU Jian-qiang;XING Qian-guo;CHEN Yan-long(Key Laboratory of Coastal Zone Environmental Process and Ecological Remediation,Yantai Institute of Coastal Zone Research,Chinese Academy of Sciences,Yantai 264003,China;University of Chinese Academy of Sciences,Beijing 100049,China;National Satellite Ocean Application Service,Ministry of Natural Resources,Beijing 100081,China;National Marine Environmental Monitoring Center,Dalian 116023,China)

机构地区：[1]中国科学院海岸带环境过程与生态修复重点实验室,中国科学院烟台海岸带研究所,烟台264003 [2]中国科学院大学,北京100049 [3]自然资源部国家卫星海洋应用中心,北京100081 [4]国家海洋环境监测中心,大连116023

出　　处：《海洋科学》2021年第7期9-17,共9页Marine Sciences

基　　金：中国科学院地球科学大数据先导专项(XDA19060203);自然科学基金(41676171,42076188)。

摘　　要：连云港海域的紫菜养殖遥感监测对于规划紫菜养殖空间分布具有重要意义。基于50 m空间分辨率的“海洋一号C”卫星(HY-1C)海岸带成像仪(coastal zone imager,CZI)数据,利用归一化植被指数(NDVI)和人工目视解译,获取了2018年10月—2020年4月连云港沿岸的紫菜养殖遥感监测面积,并分析了紫菜养殖的季节变化特征。结果显示,连云港紫菜养殖区主要分布于海州湾和连岛附近海域;养殖区自9月至次年5月在CZI图像上可见,紫菜养殖遥感监测面积呈先增加后减少的趋势,1—2月其遥感监测面积通常达到一个养殖周期的最大值,3月初面积迅速减少;基于CZI影像的2019年度遥感监测面积为123 km^(2),2020年度为160 km^(2)。建立HY-1C与哨兵二号(10 m)、高分一号(16 m)和Landsat-8(30 m)监测结果的线性模型,以Google Earth影像目视解译的紫菜养殖区遥感面积作为真实值,并将哨兵二号监测值转换为真实值。换算成真实值的2020年度紫菜养殖区真实面积为94 km^(2),较2015年度的42 km^(2)增长了1倍多。研究展示CZI可用于紫菜养殖区的业务化观测。建议利用其1—2月份的多期遥感影像监测结果作为年度紫菜养殖区遥感监测面积的基准。For planning the spatial distribution of Porphyra cultivation,remote sensing monitoring the Lianyungang sea area is of great significance.Normalized difference vegetation index(NDVI)and visual interpretation methods were used to extract the remote sensing monitoring area of Porphyra cultivation along the coast of Lianyungang from October 2018 to April 2020 based on Coastal Zone Imager(CZI)data of HY-1C satellite with a 50-m spatial resolution.Porphyra cultivation’s seasonal variation characteristics were analyzed based on the extracted data.Re-sults show that Porphyra cultivation area in Lianyungang was mainly distributed in the Haizhou Bay and the adja-cent waters of the Liandao Island.From September to April of the following year,the aquaculture area was visible on CZI images,which showed that the monitored area first increased and then decreased.The remote sensing area reached the maximum value over a breeding period from January to February;then rapidly reduced in early March.The total remote sensing area in 2019 and 2020 was 123 km^(2) and 160 km^(2),respectively.The linear models of HY-1C,Sentinel-2(10 m),and GF-1(16 m)were established based on satellite images of the Porphyra area.The Google Earth’s visual representation image was used to interpret the Porphyra cultivation’s remote sensing area as the true value,and the Sentinel-2’s monitoring value was converted into true value.The Porphyra cultivation area in 2020,which was converted into a true area,was about 94 km^(2);this is more than twice the monitoring results in 2015 when the true area of Porphyra was 42 km^(2).This study shows that CZI data can be used for operational monitoring of Porphyra cultivation zones.The multi-period remote sensing images of Porphyra in January to Feb-ruary were also suggested as a standard for the observed area of annual Porphyra cultivation.

关 键 词：HY-1C 海岸带成像仪(CZI) NDVI 连云港 紫菜养殖 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]