Scholarlink

<p>The variability in characteristics among different underlying surface combinations creates substantial challenges for accurately simulating urban rain-flood processes. As one of the most widely used models for urban stormwater simulation, the accuracy of the Storm Water Management Model (SWMM) is strongly influenced by the research spatial scale and sensitive parameters. Accurately calibrating hydrological parameters for catchment units with various land use combinations is crucial for enhancing the accuracy of urban flood simulations and forecasts. Taking the Xiaozhai area of Xi’an City as a case study, this research first analyzes the most sensitive hydrological characteristic parameters. Subsequently, the K-means clustering method is applied to classify 271 sub-catchment units based on features such as underlying surface type and distribution patterns considering the spatial heterogeneity of each catchment unit. Finally, the most sensitive parameters for different types of catchment units are calibrated. This study proposes a refined parameter calibration method that reflects the spatial heterogeneity of catchment units. Simulation results from the proposed method and the traditional method are compared with observed data. The findings indicate that the relative errors of the runoff simulation using classified calibration and assignment, as well as the conventional method, are 16.6% and 21.4% respectively, while the peak flow relative errors are 9.63% and 16.67% compared to the observed values. Furthermore, the proposed method improves runoff and peak flow simulation accuracy by 4.8% and 7.04% respectively, compared to the traditional parameter calibration method. In addition to enhancing simulation accuracy, the proposed method improves computational efficiency by grouping similar units and applying shared parameters, ultimately reducing model complexity. This research provides a novel approach for enhancing the modeling and simulation accuracy of urban rainfall-flood processes, which aiming to offer theoretical and technical support for urban flood management.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

SWMM Parameter Calibration Method for Urban Subcatchments Based on Spatial Heterogeneity

  • 期刊论文

  • 发表日期：

    2026年4月21日

查看全文
  • Li Jingsi,
  • Wang Tian,
  • Luan Guangxue,
  • Li Shan,
  • Zhou Wei,
  • Chen Guangzhao

摘要

The variability in characteristics among different underlying surface combinations creates substantial challenges for accurately simulating urban rain-flood processes. As one of the most widely used models for urban stormwater simulation, the accuracy of the Storm Water Management Model (SWMM) is strongly influenced by the research spatial scale and sensitive parameters. Accurately calibrating hydrological parameters for catchment units with various land use combinations is crucial for enhancing the accuracy of urban flood simulations and forecasts. Taking the Xiaozhai area of Xi’an City as a case study, this research first analyzes the most sensitive hydrological characteristic parameters. Subsequently, the K-means clustering method is applied to classify 271 sub-catchment units based on features such as underlying surface type and distribution patterns considering the spatial heterogeneity of each catchment unit. Finally, the most sensitive parameters for different types of catchment units are calibrated. This study proposes a refined parameter calibration method that reflects the spatial heterogeneity of catchment units. Simulation results from the proposed method and the traditional method are compared with observed data. The findings indicate that the relative errors of the runoff simulation using classified calibration and assignment, as well as the conventional method, are 16.6% and 21.4% respectively, while the peak flow relative errors are 9.63% and 16.67% compared to the observed values. Furthermore, the proposed method improves runoff and peak flow simulation accuracy by 4.8% and 7.04% respectively, compared to the traditional parameter calibration method. In addition to enhancing simulation accuracy, the proposed method improves computational efficiency by grouping similar units and applying shared parameters, ultimately reducing model complexity. This research provides a novel approach for enhancing the modeling and simulation accuracy of urban rainfall-flood processes, which aiming to offer theoretical and technical support for urban flood management.