| 摘要: |
| 【目的】近年中国多个大中城市突发极端暴雨灾害并导致严重内涝，探究城市暴雨水淹灾害的影响因子并对其进行模拟评估，以探讨城市对暴雨水淹灾害的承载能力。【方法】以杭州主城区为例，利用ArcGIS 10.2重分类中的手动分类法，结合地理探测器模型研究杭州主城区暴雨水淹灾害与高程、坡度、植被覆盖度、非渗透表面丰度、天然水体保持率和距河流距离6个影响因子的相关关系。通过地理加权回归预测6个影响因子作用下杭州主城区水淹灾害的空间分布。【结果】研究结果显示，杭州主城区水淹灾害与城市空间的高程、坡度、植被覆盖度、距河流距离呈负相关关系，与非渗透表面丰度呈正相关关系。杭州主城区暴雨水淹灾害的6个影响因子按影响程度排序为：植被覆盖度＞坡度＞高程＞距河流距离＞非渗透表面丰度＞天然水体保持率。所有影响因子的交互作用均增强了单因子对水淹灾害的影响，并呈现出两两相互增强或非线性增强的关系。从空间分布上看，杭州主城区暴雨水淹灾害呈现多核心分布状态，整体上主城区中部主要为高风险和次高风险区域，其余区域主要为中风险、次低风险和低风险区域。【结论】结合模拟成果和实践应用发现，当城市空间的高程大于40 m，坡度大于10°，植被覆盖度大于60%，非渗透表面丰度小于40%，距河流距离大于500 m时，能够对高风险与次高风险区域严重的水淹灾害起到有效的缓解作用。 |
| 关键词: 风景园林 暴雨水淹 地理探测器（GD） 因子阈值 风险预测 杭州主城区 |
| DOI：10.12409/j.fjyl.202208040466 |
| 分类号:TU998.4 |
| 基金项目:浙江省2021年度省软科学研究计划项目“城市智慧开放空间增强韧性的建设机制研究”（编号 2021C35026）；2023 年度浙江省社科联研究课题“以绿色空间为突破口缓解城市开放空间暴雨水淹灾害的对策”（编号 2023B015） |
|
| Simulation and Analysis of Factors Influencing Rainstorm Flooding Disaster in the Main Urban Area of Hangzhou |
|
WENG Yuanyuan, WEN Rikun, YANG Ling
|
|
Zhejiang A&F University
|
| Abstract: |
| [Objective] In view of the severe flooding that has recently occurred in a number of large and medium-sized Chinese cities due to sudden extreme rainstorm disasters, it is crucial for the research on urban sustainable development to analyze the factors influencing these disasters and simulate and evaluate the specific influence thereof. [Methods] Taking the main urban area of Hangzhou as an example, this research adopts the kernel density estimation method in ArcGIS 10.2 to generate a kernel density map of flooding hazards in Hangzhou during the period from 2019 to 2021 in combination with the flooding hazard points in the main urban area of Hangzhou, and by the natural breaks classification method, has the kernel density values of flooding hazards classified into the five levels of low risk, sub-low risk, medium risk, sub-high risk, and high risk, to generate a map of existing flooding hazard risk levels in the main urban area of Hangzhou. Additionally, the research adopts the manual classification method in ArcGIS 10.2 reclassification to, in combination with the independent influence of different influencing factors and relevant models such as the GeoDetector, explore the correlation between the flooding disaster in the urban area of Hangzhou and the six influencing factors of elevation, slope, vegetation coverage, impermeable surface abundance, natural water retention rate, and distance from river, and have the six influencing factors classified into 10 levels, with each level being corresponding to a stable interval of flooding hazard kernel density value, so as to determine the threshold interval between the most severe and the least severe factors influencing flooding hazard, and to rank the influence degree of the six influencing factors from high to low. Particularly, the research builds a fishing network in ArcGIS 10.2 to divide the main urban area into 500 m×500 m grid surface elements, generating a total of 13,425 elements, accompanied by 13,425 sampling points. Based on the sampling points, the research samples flooding hazard kernel density values and raster data from the six influencing factors and assigns them to the surface elements of the fishing network to obtain vector data that can be used for spatial analysis. On this basis, the research predicts the spatial distribution of flooding in the main urban area of Hangzhou under the effect of the six influencing factors by geographically weighted regression in ArcGIS 10.2, and compares the prediction results with actual observations to explore the reliability thereof. [Results] The findings of this research indicate that the flooding disaster in the main urban area of Hangzhou is positively correlated with the abundance of impermeable surfaces and negatively correlated with such factors as elevation, slope, vegetation coverage, and distance from river, and has no obvious positive or negative correlation with the natural water retention rate. The level of influence of the six influencing factors on the storm flooding disaster in Hangzhou’s main urban area can be expressed as follows: vegetation coverage > slope > elevation > distance from river > impermeable surface abundance > natural water retention rate. The interaction between all these influencing factors can increase the influence of a single factor on the flooding disaster and demonstrate the relationship between mutual and nonlinear enhancement. Spatially, there is a significant positive spatial correlation and a multi-core distribution of heavy rainfall flooding hazards in the main urban area of Hangzhou. In general, the core part of Hangzhou’s main urban area exhibits high and sub-high risk, while the remaining parts exhibit medium risk, sub-low risk, and low risk. Most of the frequently influenced sub-districts are featured by overall flat topography, low green coverage, high non-permeable surface, and dense river network. [Conclusion] In combination with the simulation results and practical application, the research concludes that the severe flooding disaster in high-risk and sub-high risk areas can be effectively mitigated when the elevation is higher than 40 m, the slope is higher than 10°, the vegetation coverage is higher than 60%, the impermeable surface abundance is less than 40%, and the distance from river is more than 500 m. In view of the low vegetation coverage and high abundance of non-permeable surfaces in the core part of Hangzhou’s main urban area, the urban planning can make full use of existing building roofs, building facades, community gray spaces, water systems and roads running through the city to build a blue-green sponge network covering areas with high risk or sub-high risk for storm flooding, increase the creation of permeable substrates in open spaces in risk areas, and solve the problem of uneven distribution of green patches in open spaces. Meanwhile, in addition to the measures for vegetation coverage, the climate-adaptive planning of urban space aiming to mitigate flooding hazards should also comprehensively consider the positive and negative factor curves and factor thresholds of flooding hazards for each influencing factor in urban space, so as to obtain a more reasonable indicator interval of planning control factors. |
| Key words: landscape architecture storm flooding GeoDetector (GD) factor threshold risk prediction main urban area of Hangzhou |
