首页 > 分享 > 黄河三角洲湿地入侵物种互花米草时空演化与景观格局分析

黄河三角洲湿地入侵物种互花米草时空演化与景观格局分析

花匠小妙招
2025-04-30 15:39

摘要： 近年来, 随着入侵物种——互花米草的爆发, 黄河三角洲自然保护区的湿地生态系统的结构和功能受到了严重破坏, 分析长时序互花米草入侵特征及其景观格局变化具有重要生态价值。本研究以长时序Landsat光学遥感影像为数据源, 以黄河三角洲湿地为实验区, 基于支持向量机分类算法获取了2000–2022年间的黄河三角洲湿地植被类型信息, 在此基础上揭示湿地的时空动态变化, 联合8种景观格局指数, 探究互花米草的爆发特征, 分析互花米草的空间格局、演变特点和生态因子。结果表明: (1)研究获取的23个时期的地物提取结果的平均总体精度为88.60%, 平均Kappa系数为0.85。支持向量机算法具有高于其他传统机器学习方法的分类精度, 其结果可以较好地满足湿地演化机理分析需求; (2)长时序湿地演化结果表明, 互花米草的扩张呈现出明显的阶段性特征: 2002–2010年, 互花米草扩张速度缓慢; 自2010年开始, 互花米草在黄河入海口两侧的扩张进入爆发期, 到2012年面积增长至22.63km2, 2012–2021年间互花米草面积持续稳定增长, 至2021年, 互花米草的面积扩张至最大, 达到50.42km2。景观格局指数的变化表明, 研究区内互花米草表现出较高的生态优势。破碎的互花米草斑块不断聚集连接, 逐渐入侵湿地本土植被并与其他植被群落形成了交错分布的空间格局; (3)黄河三角洲互花米草的扩张, 与水热条件、日照时长、地形与滩涂环境等多种自然因素表现出较高的相关性, 其中, 日照情况与离海距离与互花米草的生长极显著相关。本研究结果可为相关部门在黄河三角洲自然保护区湿地合理利用及互花米草治理和清除方面提供数据参考。

关键词: 互花米草, 长时序遥感监测, 黄河三角洲, 湿地分类, 演化机理分析

Abstract: Aims In recent years, with the outbreak of invasive species—Spartina alterniflora, the structure and function of the wetland ecosystem in the Yellow River Delta Nature Reserve have been severely damaged. It is of great ecological value to analyze the invasion characteristics and landscape pattern changes of S. alterniflora in long time series. Methods In this study, we use long-term Landsat optical remote sensing images as the data source and takes the Yellow River Delta wetlands as the study area. Based on the support vector machine classification algorithm, we have obtained the wetland type information of the Yellow River Delta from 2000 to 2022. On this basis, we revealed the spatiotemporal dynamic changes of the wetlands. By combining 8 landscape pattern indices, the outbreak characteristics of S. alterniflora were studied and further analysis on the spatial pattern, evolution characteristics, and ecological factors of S. alterniflora were conducted. Important findings The results showed that: (1) The average overall accuracy of ground object extraction results obtained in 23 periods is 88.60%, and the average Kappa coefficient was 0.85. The Support Vector Machine algorithm exhibits higher classification accuracy than other traditional machine learning methods, and its results can well reveal the needs for analyzing the wetland evolution mechanism. (2) The results of wetland evolution over long time periods indicate that the expansion of S. alterniflora obvious phased characteristics: from 2002 to 2010, S. alterniflora expanded slowly; Since 2010, the expansion of S. alterniflora on both sides of the Yellow River estuary has entered an explosive period. By 2012, its area had increased to 22.63 km2; And it continued to grow steadily from 2012 to 2021. In 2021, the area of S. alterniflora had expanded to 50.42 km2. The landscape pattern index indicates that S. alterniflora shows high ecological advantage in the study area. The fragmented patches of S. alterniflora continuously gather and connect, gradually invading the native wetland vegetation and constructing a staggered spatial pattern with other vegetation communities. (3) The expansion of S. alterniflora in the study area shows a high correlation with hydrothermal conditions, sunshine duration, topography and mudflat environment, among which, there is a extremely significant correlation between the sunshine conditions, distance from the sea, and the growth of S. alterniflora. The research results of this article offer theoretical underpinnings for relevant departments in the rational utilization of wetlands in the Yellow River Delta Nature Reserve, as well as in the management and eradication of S. alterniflora.

Key words: Spartina alterniflora, long-term remote sensing monitoring, Yellow River Delta, wetland classification, evolution mechanism analysis