植物生态学报 ›› 2025, Vol. 49 ›› Issue (1): 42-58.DOI: 10.17521/cjpe.2024.0152  cstr: 32100.14.cjpe.2024.0152

• 研究论文 • 上一篇    下一篇

呼伦贝尔草甸草原退化特征因子识别与快速诊断指标体系构建

许梦真1,2, 卢正宽1,2, 谭星儒1,2, 王彦兵1,2, 苏天成1,2, 窦山德1,3, 潘庆民1,2, 陈世苹1,2,*()()   

  1. 1中国科学院植物研究所植被与环境变化国家重点实验室, 国家植物园, 北京 100093
    2中国科学院大学, 北京 100049
    3内蒙古锡林郭勒草原生态系统国家野外科学观测研究站, 内蒙古锡林郭勒 026000
  • 收稿日期:2024-05-13 接受日期:2024-12-10 出版日期:2025-01-20 发布日期:2025-03-08
  • 通讯作者: * 陈世苹: ORCID: 0000-0002-1934-2372 (spchen@ibcas.ac.cn)
  • 基金资助:
    中国科学院战略性先导科技专项(A类)(XDA26020101);国家自然科学基金(U22A20559);国家自然科学基金(32071565)

Identification of key factors and construction of a rapid diagnostic indicator system for evaluation of grassland degradation in Hulun Buir meadow grasslands

XU Meng-Zhen1,2, LU Zheng-Kuan1,2, TAN Xing-Ru1,2, WANG Yan-Bing1,2, SU Tian-Cheng1,2, DOU Shan-De1,3, PAN Qing-Min1,2, CHEN Shi-Ping1,2,*()()   

  1. 1State Key Laboratory of Vegetation and Environmental Change, Institute of Botany, Chinese Academy of Sciences, China National Botanical Garden, Beijing 100093, China
    2University of Chinese Academy of Sciences, Beijing 100049, China
    3National Field Research Station of Xilin Gol Grassland Ecosystem, Xilin Gol, Nei Mongol 026000, China
  • Received:2024-05-13 Accepted:2024-12-10 Online:2025-01-20 Published:2025-03-08
  • Supported by:
    Strategic Priority Research Program of the Chinese Academy of Sciences(XDA26020101);National Natural Science Foundation of China(U22A20559);National Natural Science Foundation of China(32071565)

摘要:

中国是草地资源大国, 但由于气候变化和人类活动的双重影响, 近70%的草原存在不同程度的退化, 因此明确草原退化特征因子, 构建草原退化快速诊断指标体系, 对准确评价草原的退化状况至关重要。通过在内蒙古呼伦贝尔草甸草原选取未退化、轻度退化、中度退化、重度退化样地进行群落调查, 获取相关植被和土壤指标, 运用随机森林模型进行退化指标的筛选和权重赋值, 并兼顾政府和牧民对生态系统服务的需求, 构建草甸草原退化快速诊断指标体系。随退化程度的增加, 群落地上生物量、凋落物生物量、群落高度、叶片厚度等指标均显著降低; 植物多样性、土壤全氮含量、有机碳含量等指标呈先上升后下降的趋势; 而群落氮磷含量、土壤密度等指标表现出显著上升的趋势。基于随机森林重要值和指标获取的难易程度, 该研究筛选出地上生物量、优质牧草比例、群落高度、凋落物生物量、物种丰富度、叶干物质含量、叶片厚度、土壤密度、土壤含水量、土壤无机氮含量共10个退化特征因子, 涉及牧草供应、侵蚀控制、多样性保护、植被抗逆、水分养分调节多个生态系统服务。以未退化样地作为参照, 构建了内蒙古草甸草原退化指数(DI), 并明确了不同退化程度下DI的变化范围, 为国家与地方现行标准指标选取的合理性提供了数据支持。

关键词: 草地退化, 生态系统服务, 随机森林算法, 退化指标

Abstract:

Aims China harbors extensive grassland resources, yet nearly 70% of these grasslands are afflicted by varying degrees of degradation under the combined pressure of climate change and human activities. Pinpointing the pivotal factors driving grassland degradation and establishing a rapid diagnostic system is imperative for precise condition assessments.

Methods This study was conducted in the Hulun Buir meadow steppe of Nei Mongol. The selected sites were categorized into four degradation levels: non-degraded, lightly degraded, moderately degraded, and heavily degraded. Vegetation and soil indicators were collected. Leveraging the random forests algorithm, degradation indicators were screened and weighted, with efforts made to reconcile ecosystem service priorities between the government and pastoralists.

Important findings This study identified ten key factors characterizing degradation, including aboveground biomass, proportion of high-quality forage, community height, litter biomass, species richness, leaf dry matter content, leaf thickness, soil density, soil water content and soil inorganic water content. These indicators encapsulate diverse ecosystem services, including forage supply, erosion control, biodiversity conservation, vegetation resilience, and water and nutrient regulation. Using non-degraded sites as a benchmark, a degradation index (DI) for the meadow steppes of Nei Mongol was developed, accompanied by delineated DI thresholds for different degradation levels. This study provides foundational data to support judicious selection of indicators for both national and regional standards.

Key words: grassland degradation, ecosystem services, random forests algorithm, degradation indicator