赤水河流域生态系统服务权衡/协同的地形梯度解析与机器学习模型应用
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(42161052);贵州省科技支撑计划项目(黔科合支撑[2023]一般198、199, 黔科合支撑[2020]4Y008号)


Topographic gradient analysis and application of machine learning models for trade-offs and synergies of ecosystem services inthe Chishui River Basin
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
  • |
  • 文章评论
    摘要:

    地形梯度在景观格局的形成与演变过程中发挥着至关重要的作用,它深刻影响区域自然与社会资源的空间分布及要素流动,从而对生态系统服务及其相互关系产生深远影响。以赤水河流域为例,选取产水量、土壤保持、碳储量、生境质量、景观美学和粮食生产6类关键生态系统服务进行评估,运用基于对象的地形梯度分析法和空间叠置法探讨生态系统服务权衡/协同关系的时空演变规律,最后引入XGBoost-SHAP模型进行影响因素分析,揭示各特征变量的贡献程度和作用方向。结果表明:(1)2000年至2020年,赤水河流域生态系统服务平均物质量增加的为粮食产量、土壤保持和产水量,平均物质量下降的为碳储量、生境质量和景观美学,生态系统服务呈现显著的权衡特征。(2)权衡协同表现出地形梯度响应特征:20年来强权衡面积减少,弱权衡面积增加;高协同总体在中低梯度地形区稳定或略增,在高梯度地形区减少,而低协同在平地与低丘增加,在其他梯度减少;耕地减少、林地恢复会促进弱权衡与高协同,推动全域生态压力梯度递减。(3)植被、人类活动和地形因子是流域生态系统服务权衡/协同关系演变的关键驱动因素。NDVI作为核心生态参数在调控中发挥主导作用,高强度人类活动通过削弱植被健康度以加剧权衡,地形因子决定了生态系统服务空间分布的基础格局,三者共同构成“自然基底-生态表征-人为扰动”的驱动体系。

    Abstract:

    The terrain gradient plays a critical role in the formation and evolution of landscape patterns. It profoundly influences the spatial distribution and flow of natural and social resources, thereby having a far-reaching impact on ecosystem services and their interrelationships. Taking the Chishui River Basin as an example, six key ecosystem services—water yield, soil conservation, carbon storage, habitat quality, landscape aesthetics, and food production—are evaluated in this study. The study employs an object-based terrain gradient analysis method and spatial overlay technique to investigate the spatio-temporal evolution patterns of ecosystem service trade-offs/synergies. Finally, the XGBoost-SHAP method is introduced to analyze the influencing factors, revealing the contribution and direction of each feature variable. The results indicated that: (1) Between 2000 and 2020, the average values of food production, soil conservation, and water yield increased, while carbon storage, habitat quality, and landscape aesthetics declined, revealing a marked pattern of trade-offs among ecosystem services. (2) Trade-off and synergy patterns exhibited a gradient response: over the past two decades, strong trade-off areas reduced, whereas weak trade-off areas expanded. High synergy areas remained stable or slightly increased in low- to mid-gradient zones (Ⅰ—Ⅲ), but decreased in high-gradient zones (Ⅳ—Ⅴ). Low synergy areas increased in plains and low-hill zones (Ⅰ) and decreased in other terrain classes. The cultivated land reduction and forest restoration promoted weak trade-off and enhanced synergy, leading to a relief in ecological pressure. (3) Vegetation, human activities, and topographic factors were identified as key drivers of trade-off and synergy evolution. NDVI, as a core ecological indicator, played a dominant regulatory role. Intensive human activities exacerbated trade-offs by degrading vegetation health, while topographic factors determined the underlying spatial patterns of ecosystem service distribution. Overall, these components constituted a driving mechanism characterized by natural matrix, ecological outcomes, and anthropogenic disturbance.

    参考文献
    相似文献
    引证文献
引用本文

勾容,苏维词,董文卓,黄亮.赤水河流域生态系统服务权衡/协同的地形梯度解析与机器学习模型应用.生态学报,2025,45(20):9980~9996

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: