多源土地利用数据下景观格局对径流-泥沙影响的不确定性——以鄱阳湖区虎山流域为例
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:

第二次青藏高原科学考察与研究项目(2019QZKK0103);国家自然科学基金(42401030);三峡大学科研启动基金(2023RCKJ0025)


Uncertainty analysis on the effects of landscape pattern on runoff and sediment using multi-source land use data: a case study in the Hushan River Basin of the Poyang Lake Regions
Author:
Affiliation:

Fund Project:

  • 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:

    景观格局是表征土地利用(LUCC)空间分布特征的重要指标,人类活动导致的景观格局变化是改变流域径流-泥沙过程的重要因素之一。现有研究仅采用单一来源的LUCC产品探究景观格局对径流-泥沙过程的影响,忽略了不同LUCC产品给研究带来的不确定性。以鄱阳湖区饶河上游虎山流域为研究对象,基于1963—2022年气象水文数据和3种同分辨率的1990—2020年七期LUCC产品,分析流域降雨、径流、输沙和土地利用的变化特征,并结合相关分析和偏最小二乘回归法(PLSR)探究景观格局变化对径流-泥沙特征的影响及其不确定性。研究发现,1963—2022年,虎山流域年输沙模数(STM)和含沙量(SSC)呈显著增长趋势(P < 0.001),径流深(RD)呈不显著增长趋势。流域内主要的LUCC为林地和耕地,约占流域面积的(79.57±6.03)%和(15.72±2.36)%。1990—2020年,城镇扩张主要侵占林地和耕地,且存在较多林地-耕地之间的相互转化。不同LUCC产品得到的景观格局指数存在显著的差异(P < 0.05),主要由其对灌木和草地的识别差异所导致的。总体来说,研究区景观的最大斑块面积有所减小,景观破碎度有轻微增长,景观多样性有一定的增加。不同LUCC产品下得到的景观格局对径流-泥沙的影响存在差异,所有产品一致的结论是:斑块密度(PD)和蔓延度指数(CONTAG)与RD、STM和SSC呈负相关关系,边缘密度(ED)、Shannon多样性指数(SHDI)和Shannon均匀度指数(SHEI)与STM和SSC呈正相关关系;PLSR模型显示最大斑块占景观面积比例、斑块结合度、景观分割度、SHDI和修正的Simpson均匀度指数(MSIEI)对RD、STM和SSC的变化具有重要意义。选用ED表征面积-边缘特征、选用PD和CONTAG表示聚散度特征以及选用SHDI和SHEI表征多样性特征可以相对减少LUCC产品对研究结果带来的不确定性。通过增加流域内林地与耕地之间的连通性,减少零散分布的耕地,尽可能避免两种景观类型之间的相互转换,可以减少景观破碎化程度,缓解流域土壤侵蚀。研究成果可为鄱阳湖区水土流失防治及生态文明建设提供参考。

    Abstract:

    Landscape pattern is a significant indicator for representing the spatial distribution characteristics of land use and land cover change (LUCC). The landscape pattern changes caused by human activities are important factors that alter the hydrological processes within a watershed. Existing studies on the impact of landscape patterns on runoff-sediment processes only rely on single-source LUCC data, neglecting the uncertainty brought by LUCC data. In this study, the Hushan watershed in the Poyang Lake region was selected as the study area. Based on rainfall, runoff, and sediment data from 1963 to 2022, combined with seven periods of LUCC data from 1990 to 2020 at the same resolution obtained from three LUCC products, the characteristics of rainfall, runoff, sediment, and land use changes in the watershed were analyzed. Correlation analysis and partial least squares regression (PLSR) were used to investigate the impact of landscape pattern changes on runoff-sediment process characteristics and their uncertainty. The results showed that from 1963 to 2022, the annual sediment transport modulus (STM) and suspended sediment concentration (SSC) in the Hushan watershed showed a significant increasing trend (P < 0.001), while the Runoff Depth (RD) showed an insignificant increasing trend. The main LUCC types in the watershed were forest land and cultivated land, accounting for approximately (79.57±6.03)% and (15.72±2.36)% of the watershed area, respectively. From 1990 to 2020, urban expansion mainly encroached on forest land and cultivated land, with significant land conversions between the two. There were significant differences in landscape pattern characteristics among different LUCC types (P < 0.05), mainly due to the identification differences of shrub land and grass land by different LUCC products. In general, the maximum patch area of the landscape in the study area decreased, landscape fragmentation increased slightly, and landscape diversity increased to some extent. The influence of landscape patterns obtained under different LUCC products on runoff and sediment varies. The consistent findings across all products indicated that Patch Density (PD) and Contagion Index (CONTAG) are negatively correlated with RD, STM, and SSC, while Edge Density (ED), Shannon Diversity index (SHDI), and Shannon Evenness Index (SHEI) are positively correlated with STM and SSC. The PLSR model indicates that the proportion of the Largest Patch Index (LPI), Patch Cohesion (COHESION), Landscape Division (DIVISION), SHDI, and Modified Simpson Evenness Index (MSIEI) play significant roles in the variations of RD, STM, and SSC. Using ED to represent area-edge characteristics, PD and CONTAG to represent aggregation-dispersion characteristics, and SHDI and SHEI to represent diversity characteristics can relatively reduce the uncertainties caused by LUCC products in research results. By increasing the connectivity between forest land and cultivated land, reducing scattered cultivated land, and avoiding mutual conversions between the two landscape types as much as possible, the degree of landscape fragmentation can be reduced, and soil erosion in the watershed can be alleviated. The findings of this study provide a reference for soil erosion control and ecological civilization construction in the Poyang Lake region.

    参考文献
    相似文献
    引证文献
引用本文

魏冲,董晓华,马耀明,彭涛,喻丹,冷梦辉.多源土地利用数据下景观格局对径流-泥沙影响的不确定性——以鄱阳湖区虎山流域为例.生态学报,2025,45(3):1102~1115

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: