互花米草治理对黄河三角洲湿地利用及景观格局的影响——利用随机森林和改进的CA-Markov模型
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:

国家自然科学基金(41876202)


Using random forest and improved CA-Markov model to evaluate the effect of Spartina alterniflora clearing on wetland use and landscape pattern in the Yellow River Delta wetland
Author:
Affiliation:

Fund Project:

The National Natural Science Foundation of China (General Program, Key Program, Major Research Plan)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    随着2022年《互花米草防治专项行动计划(2022-2025年)》发布,黄河三角洲湿地互花米草清理工作迅速展开。通过对比模拟的自然发展情景下与互花米草大规模清理后的真实情景下黄河三角洲湿地利用及景观格局指数,分析清理工作生态效益及湿地未来潜在风险。基于时序Sentinel-1 SAR影像和随机森林分类器获取了2020-2023年黄河三角洲湿地利用/覆盖变化结果;针对短期湿地模拟预测驱动力因素缺乏定量化设定的问题,构建了一种联合随机森林与CA-Markov模型的湿地模拟预测方法,实现2023年湿地自然发展情景下更精准的模拟预测;最后计算了2023年湿地在模拟和真实情景下的景观格局指数。结果表明:1)2023年互花米草清理工作效果显著,其面积锐减至5.53km2,与2022年相比减少了55.36km2,为芦苇等本土植被释放了生存空间;2)与自然发展情景相比,互花米草“围栏”不复存在,本土植被连通性上升,芦苇分布更加琐碎且面积比自然发展情景下增加了16.37km2,湿地的优势植被类型已逐渐向芦苇转变;3)由于互花米草对新生滩涂的入侵能力强于其它本土植被以及沿岸受海浪冲击等问题,互花米草清除后的黄河三角洲湿地面临互花米草复萌、二次蔓延及海岸线侵蚀的新风险。

    Abstract:

    With the release of the Special Action Plan for the Control of Spartina alterniflora (2022-2025) in 2022, the cleanup work of Spartina alterniflora in the Yellow River Delta wetland has been carried out rapidly. This study analyzed the ecological benefits and future potential risks of Spartina alterniflora clearing by comparing the wetland use and landscape pattern indices of the Yellow River Delta wetland under the simulated natural development scenario and the actual scenarios post large-scale artificial clearing of Spartina alterniflora. Firstly, the wetland use and wetland cover in the Yellow River Delta from 2020 to 2023 were obtained based on time series Sentinel-1 SAR images and random forest (RF) algorithms. Subsequently, given the lack of quantitative setting of driving forces for short-term wetland simulation prediction, a method combining RF and CA-Markov model was constructed to to quantify the weight of driving factors and generate multi-criteria evaluation (MCE) atlas, which is introduced into CA-Markov model to achieve more accurate simulation predictions for wetland conditions in 2023 under natural development scenarios. Finally, the landscape pattern indices of wetland in 2023 under the simulated and actual scenarios were calculated. The results indicated that: (1) Compared with the original model, the Kappa coefficient of the improved CA-Markov model is improved by 0.014. From 2020 to 2022, Spartina alterniflora distributed in a large area along the coast of the wetland, occupying the living space of native vegetation. In 2023, the effect of clearing Spartina alterniflora was effect, and its area was reduced to 5.53 km2, which was 55.36 km2 less than that in 2022, freeing up the living space for native vegetation such as Phragmites australis. (2) In the natural development scenario, the distance factor has a greater impact on wetland evolution, while the terrain factor has a lesser impact. Compared with the natural development scenario, the "fence" of Spartina alterniflora disappeared in actual scenario, and the connectivity of native vegetation increased. The distribution of Phragmites australis became more fragmented, with an area increase of 16.37 km2 compared to the natural development scenario, gradually shifting the dominant vegetation type of wetlands towards Phragmites australis. (3) Due to the invasion ability of Spartina alterniflora to the new tidal flat was stronger than that of other native vegetation, and the coastlines of regions without Spartina alterniflora distribution move inward obviously when the coastal regions are affected by waves, the post-Spartina alterniflora-cleaning Yellow River Delta wetland faced new risks of recurrence, secondary spread, and coastline erosion.

    参考文献
    相似文献
    引证文献
引用本文

李振今,王志勇,刘晓彤,孙士昌,马晓康,左亚男.互花米草治理对黄河三角洲湿地利用及景观格局的影响——利用随机森林和改进的CA-Markov模型.生态学报,2024,44(18):8366~8382

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数: