基于多源数据的广西珍珠湾红树林遥感监测与分析  

Remote sensing monitoring and analysis based on multi-sourced data of mangroves in Pearl Bay,Guangxi

作　　者：李梅娜 田雨晴 印萍[1,2,3,4] 段晓勇 张嘉琳 李雪 张尧 LI Meina;TIAN Yuqing;YIN Ping;DUAN Xiaoyong;ZHANG Jialin;LI Xue;ZHANG Yao(Geology Research Center for Coast and Shelf,China Geological Survey(CGS),Qingdao Institute of Marine Geology,CGS,Qingdao 266237,China;Zhoushan Marine Geohazards Field Observation and Research Station,CGS,Qingdao 266237,China;Engineering Technology Innovation Center for Marine Methane Monitoring,Ministry of Natural Resources,Qingdao 266237,China;Observation and Research Station of Zhejiang Coastal Urb-an Geological Security,Ministry of Natural Resources,Hangzhou 310007,China;Zhejiang Institute of Marine Geological Survey,Zhoushan 316000,China;College of Marine Geosciences,Ocean University of China,Qingdao 266100,China;School of Earth Sciences,China University of Geosciences,Wuhan 430074,China)

机构地区：[1]中国地质调查局海岸带地质和大陆架地质研究中心,中国地质调查局青岛海洋地质研究所,青岛266237 [2]中国地质调查局舟山海洋地质灾害野外科学观测研究站,青岛266237 [3]自然资源部海洋甲烷监测工程技术创新中心,青岛266237 [4]自然资源部浙江沿海城市地质安全野外科学观测研究站,杭州310007 [5]浙江省海洋地质调查大队,舟山316000 [6]中国海洋大学地球科学学院,青岛266100 [7]中国地质大学(武汉)地球科学学院,武汉430074

出　　处：《海洋地质前沿》2025年第4期82-92,共11页Marine Geology Frontiers

基　　金：亚洲合作资金项目“长江三角洲与红河三角洲海洋地质环境与灾害合作研究”;中国地质调查局项目“长江口海岸带地质环境调查与监测”(DD20242714);浙江省地质专项资金项目“浙江省海岸带综合地质调查(台州重点区)”(省资2023013);中央自然灾害防治体系建设项目“浙江省滨海地区浅层气灾害防治技术研究:以舟山绿色石化基地为例”(2023330053)。

摘　　要：为了解广西珍珠湾红树林近30年来的时空演化,以Landsat TM/OLI卫星遥感影像为数据源,结合Sentinel-2卫星遥感影像,基于地物光谱特征,构建了决策树分类规则;通过基于缨帽变换的决策树分类方法,提取1995、2000、2005、2010、2015和2023年红树林空间分布信息,计算动态度与景观格局指数,对研究区1995—2023年的红树林时空演化进行了动态分析。结果表明:近30年来,珍珠湾红树林在面积构成、空间位置等方面变化显著,整体呈现先减少后增多的趋势。其中,动态度最高的时期为2005—2010年,红树林面积增加了127.3 hm2,动态度为3.59%;景观格局指数计算表明,破碎度和景观分异程度总体上呈现先增后减的趋势,聚合度比较稳定;红树林空间演化特征与降水、气温、泥沙沉积、虫害等自然因素相关,与围塘养殖等人类活动密切相关。To understand the spatiotemporal evolution of mangroves in Pearl Bay,Guangxi over the past 30years,we used Landsat TM/OLI satellite remote sensing images as the data source and combined them with Sentinel-2 satellite remote sensing images.Based on the spectral characteristics of the ground objects,we constructed a decision tree classification rule.Through the decision tree classification method based on tasseled cap transformation,we extracted the spatial distribution information of mangroves in 1995,2000,2005,2010,2015 and 2023,calculated the dynamic degree and landscape pattern index,and analyzed the dynamic spatiotemporal evolution of mangroves in the study area from 1995 to 2023.Results indicate that over the past nearly 30 years,mangrove forests in Pearl Bay have changed significantly in terms of area composition and spatial location,presenting an overall trend of decreasing first and then increasing.The period with the highest dynamic degree is 2005-2010,during which the mangrove area increased to 836.21 hectares,with a dynamic degree of 3.59%.The calculation of the landscape pattern index shows that the fragmentation and landscape differentiation degree generally exhibit a trend of increasing first and then decreasing,while the aggregation degree is relatively stable.The spatial evolution of the mangroves are related to natural factors such as precipitation,temperature,sediment deposition,and pest damage,and are closely related to human activities such as pond culture.

关 键 词：红树林 遥感监测 时空演化 景观格局指数 驱动因素 

分 类 号：P737.1[天文地球—海洋地质]