摘要
水生植被是湖泊生态系统的重要指示因子,受到气候变化及人类活动协同作用。基于Landsat系列卫星影像,构建了融合NDVI、AI和SVSI指数的决策树分类模型,提取了2000~2024年太湖水生植被时空分布,并综合采用传统方法和机器学习方法分析了气象因子和水质因子对水生植被面积的影响机制,揭示了水生植被时空动态的关键控制因子。结果表明,研究提出的水生植被遥感识别模型,在TM/ETM+/OLI传感器影像中均表现出良好的鲁棒性,分类总体精度达88%以上,Kappa系数超过0.75,其中OLI影像因更高的信噪比取得最优精度(总体精度92.59%,Kappa系数0.84)。太湖水生植被主要分布于东太湖、胥口湾和贡湖湾,夏季覆盖面积显著高于冬季(平均差值47.6 km^(2))。在时空动态方面,2010年被识别为水生植被面积变化转折点,夏季植被面积从2000年峰值217 km^(2)降至2012年最低值177 km^(2)之后回升,冬季植被面积则在2011年后稳定回升。驱动力分析结果表明,在夏季,气象条件和Chla共同影响水生植被生长,浮游植物作为Chla的主要来源对水生植被生长具有最主要的负面效果;在冬季,气象条件对水生植被生长的影响减弱,TN成为最主要的影响因子,并对水生植被生长具有显著的负面作用。因此,太湖水生态系统保护需要控制夏季藻华及冬季营养盐排放。研究结果为太湖治理提供了科学依据。
Aquatic vegetation serves as a critical indicator of lake ecosystems and is subject to the joint effects of climate change and human activities.This study developed a decision tree classification model integrating NDVI,AI,and SVSI indices based on Landsat imagery to extract the spatiotemporal distribution of aquatic vegetation in Lake Taihu from 2000 to 2024.The influence mechanisms of meteorological and water quality factors on vegetation coverage were analyzed using a combined method of the traditional and machine learning approaches.Results demonstrated that the proposed remote sensing identification model exhibited demonstrated robust performance across TM/ETM+/OLI images,with overall classification accuracy exceeding 88%and Kappa coefficients surpassing 0.75.OLI imagery achieved optimal accuracy(92.59%overall accuracy,Kappa coefficient 0.84)due to higher signal-to-noise ratios.Aquatic vegetation in Lake Taihu predominantly distributed in East Lake,Xukou Bay,and Gonghu Bay,with summer coverage significantly exceeding winter levels(the average difference is 47.6 km^(2)).A turning point was detected in 2010 in terms of the spatiotemporal variation of aquatic vegetation coverage.The summer vegetation coverage declined from the peak in 2000(217 km^(2))to the minimum in 2012(177 km^(2))before recovering,while the winter coverage showed a stable recovery after 2011.Driving force analysis indicated that in summer,meteorological conditions and chlorophyll-a(Chla)jointly influenced vegetation growth.Phytoplankton,as the primary source of Chla,exerted the strongest negative effects.During winter,meteorological influence diminished,and total nitrogen(TN)emerged as the dominant factor with significant negative impacts on vegetation growth.Therefore,effective protection of Lake Taihu ecosystem necessitated the dual control strategies targeting summer algal blooms and winter nutrient discharges.This study provided a scientific basis for its ecological remediation.
作者
吴弈秋
熊俊峰
黄金怡
马荣华
WU Yi-qiu;XIONG Jun-feng;HUANG Jin-yi;MA Rong-hua(School of Geographic Sciences,Nanjing University of Information Science&Technology,Nanjing 210044,China;State Key Laboratory of Lake and Watershed Science for Water Security,Nanjing Institute of Geography and Limnology,Chinese Academy of Sciences,Nanjing 211135,China;University of Chinese Academy of Science,Nanjing,Nanjing 211135,China;College of Tripical Agriculture and Forestry,Hainan University,Danzhou 571737,China)
出处
《长江流域资源与环境》
2026年第2期364-378,共15页
Resources and Environment in the Yangtze Basin
基金
国家重点研发计划项目(2021YFD1700600)
江苏省自然资源科技创新项目(JSZRKJ202409)
自然资源部国土卫星遥感应用重点实验室开放基金项目(KLSMNR-G202302)。
