摘要
基于2003—2017年黄渤海海域中分辨率成像光谱仪(moderate-resolution imaging spectroradiometer,MODIS)卫星遥感数据,利用自组织映射(self-organizing feature map,SOM)神经网络模型研究叶绿素a浓度(chlorophyll a concentration,Chl-a)的典型分布模式,分析Chl-a变化趋势,并利用广义加性模型(generalized additive model,GAM)研究其与环境因子的关系。结果表明:黄渤海Chl-a存在明显的季节性变化,7月份浓度最低,为2.41 mg/m^3,4月份浓度最高,为3.43 mg/m^3;Chl-a呈现从近海岸海域向深水海盆逐渐降低的变化趋势;将SOM模型提取的典型模式分为清澈、低浓度、中浓度和高浓度模式,这些模式有效地阐明了2003—2017年黄渤海Chl-a在时间上存在春季高、夏季低的变化趋势,Chl-a高值区主要分布在河流的入海口及近海岸;利用GAM模型发现海表温度(sea surface temperature,SST)、风速与Chl-a之间存在显著的非线性关系,SST、风速对Chl-a变化的解释率为39.3%,SST对Chl-a变化的影响比风速更大;人类活动的增加对黄渤海Chl-a变化也起着重要的作用。
The temporal and spatial distribution pattern of chlorophyll a concentration(Chl-a)derived from moderate-resolution imaging spectroradiometer(MODIS)satellite data in the Bohai Sea and Yellow Sea during 2003 to 2017 and its relationship with environmental factors were studied based on the self-organizing map(SOM)neural network model and generalized additive model(GAM).The results show a distinct seasonal variations of Chl-a along with a gradual increase in the study period.Chl-a reaches the lowest value of 2.41 mg/m 3 in July,and the highest value 3.43 mg/m 3 in April.The Chl-a decreases from the coast to deep water basin.The typical patterns extracted by SOM model are divided into clear,low concentration,medium concentration and high concentration modes,which effectively clarify that Chl-a in the Bohai Sea and Yellow Sea has high concentration in spring and low in summer during 2003 to 2017 and the highest value area of Chl-a is mainly distributed in coastal estuaries.By GAM,there is a significant nonlinear correlation between Chl-a,sea surface temperature(SST)and wind speed.The interpretation rate of Chl-a change is 39.3%,and the effect of SST on Chl-a change is greater than that of wind.Human activities also play an important role in the change of Chl-a in the Yellow Sea.
作者
赵娜
王霄鹏
李咏沙
姚凤梅
ZHAO Na;WANG Xiao-peng;LI Yong-sha;YAO Feng-mei(Remote Sensing Information and Digital Earth Center,College of Computer Sciences and Technology,Qingdao University,Qingdao 266071,China;College of Earth and Planetary Sciences,University of Chinese Academy of Sciences,Beijing 100049,China;Key Laboratory of Computational Geodynamics,Chinese Academy of Sciences,Beijing 100049,China)
出处
《科学技术与工程》
北大核心
2020年第17期7101-7107,共7页
Science Technology and Engineering
基金
国家自然科学基金(31571565,31671585)
山东省自然科学基金重大基础研究项目(ZR2017ZB0422)。
关键词
叶绿素a
自组织映射(SOM)神经网络
广义加性模型(GAM)
海表温度
海表风场
黄渤海
chlorophyll a
self-organizing feature map(SOM)neural network
generalized additive model(GAM)
sea surface temperature
sea surface wind
Bohai Sea and Yellow Sea