【摘要】 城市积水问题是城市资源集聚扩散、城市土地利用与城市水文环境之间重要矛盾的体现,探讨城市用地形态对城市积水的影响具有重要意义。以闽三角城市群洪涝灾害高风险区为例,采用多源数据和BP(Back Propagation)神经网络分别对城市积水的分布密度和深度进行测算,借助Arc GIS和fragstats4.2移动窗口量化城市用地形态,通过皮尔逊相关系数、逐步回归和空间自回归模型,剖析城市用地形态对城市积水的影响。根据研究结论提出缓解城市积水问题的城市用地形态规划管理建议。更多还原

【Abstract】 The problem of urban waterlogging is the embodiment of the important contradiction between urban resources agglomeration and diffusion, urban land use, and urban hydrological environment. It is of great significance to explore the impact of urban landform on urban water accumulation. Taking the high-risk area of flood disaster in the urban agglomeration of Min Delta as an example, Multi-source data and Back-Propagation neural network were used to calculate the distribution density and depth of urban waterlogging respectively. With the help of Arc GIS and fragstats4.2 mobile window, the influence of urban land use form on urban water accumulation was analyzed through the Pearson correlation coefficient, stepwise regression, and spatial autoregressive model. Finally, according to the research conclusions, the urban land-use form planning, and management suggestions were put forward to alleviate the problem of urban water accumulation.更多还原