This study presents a water accumulation diffusion algorithm to spatially simulate rainstorm-induced waterlogging for people's lives and property safety. Taking part of Jinfeng District in Yinchuan City, China, as a study area, a storm water management model (SWMM) model is constructed with the aid of geographic information system (GIS) and remote sensing (RS) technologies. GIS is used to divide subcatchments, generalize drainage system, set parameters, construct spatial geodatabase, and identify flood extents and depths. RS is used to obtain land-use/land-cover information. The water accumulation diffusion algorithm is then designed using the strategies of the dynamic interactions between pipes and surface and between central pixel and its neighbourhood pixels to transform water accumulation volume of sub-catchment into the submerged range and water accumulation depth. Positions, extents, depths, and volumes of water accumulation from pipe network and surface are simulated, respectively. The spatial simulation precisions of rainstorm waterlogging from the pipe network and surface are verified according to the measured and cyber rainstorm data, respectively. The results show that (1) the number of water accumulation nodes increases with the increase of rainfall intensity; (2) urban waterlogging is mainly distributed in the intersects of roads, low depressions and the aged drainage networks; and (3) spatial simulation of urban rainstorm waterlogging based on the GIS, RS, and SWMM techniques and the water accumulation diffusion algorithm is reliable. The results can provide decision-makings to predict rainstorm waterlogging, design drainage network, and construct a sponge city. ARTICLE HISTORY

With the acceleration of the urbanization process, the problems caused by extreme weather such as heavy rainstorm events have become more and more serious. During such events, the road and its auxiliary facilities may be damaged in the process of the rainstorm and waterlogging, resulting in the decline of its traffic capacity. Rainfall is a continuous process in a space–time dimension, and as rainfall data are obtained through discrete monitoring stations, the acquired rainfall data have discrete characteristics of time interval and space. In order to facilitate users in understanding the impact of urban waterlogging on traffic, the visualization of waterlogging information needs to be displayed under different spatial and temporal granularity. Therefore, the appropriateness of the visualization granularity directly affects the user’s cognition of the road waterlogging map. To solve this problem, this paper established a spatial granularity and temporal granularity computing quantitative model for spatio-temporal visualization of road waterlogging and the evaluation method of the model was based on the cognition experiment. The minimum visualization unit of the road section is 50 m and we proposed a 5-level depth grading method and two color schemes for road waterlogging visualization based on the user’s cognition. To verify the feasibility of the method, we developed a prototype system and implemented a dynamic spatio-temporal visualization of the waterlogging process in the main urban area of Nanjing, China. The user cognition experiment showed that most participants thought that the segmentation of road was helpful to the local visual expression of waterlogging, and the color schemes of waterlogging depth were also helpful to display the road waterlogging information more effectively.

<p><strong>Abstract.</strong> Urban waterlogging, as a common natural disaster in China, seriously restricted the development of society. Nowadays, while the computer technology is developing continuously, the urban waterlogging model is also constantly improved. These models can simulate the process of urban waterlogging, but the simulation results are not intuitive. So it is difficult for users to understand how the model works. Therefore, it is important to find a way to show the simulation results so that people can see the waterlogging simulation intuitively. Cesium, as a three-dimensional visualization platform, can reproduce the process of the urban waterlogging. It will make sense if we could show the simulation results on the Cesium platform. Nowadays, many studies focus on both urban waterlogging and visualization methods. However, there are fewer studies on the combination of the two, especially the interactive visualization of urban waterlogging under parameter adjustment. Therefore, this paper mainly focuses on urban three-dimensional interactive visualization method based on Cesium.</p><p>On the one hand, the three-dimensional visualization of the urban waterlogging simulation facilitates the intuitive expression of the simulation results. Without visualization, the results of the simulation are only some complicated and unintuitive figures for most non-experts. On the other hand, visualization based on the Cesium platform can better adapt to the cross-platform application. It can better meet the needs of different terminal devices of different users for the visualization platform, so that users can obtain the disaster information more accurately, consistently and intuitively. It is conducive for management departments to respond to sudden disasters more quickly and efficiently.</p><p>This research aims to propose a three-dimensional dynamic interactive visualization method for urban waterlogging. Particularly, we hope to find out how to integrate urban waterlogging model and 3D visualization platform. With this 3D visualization platform, we can combine the advantages of the SWMM (Storm Water Management Model) and Cesium platform. Using this platform, it will be easier and more effective to respond to disasters for the masses and management departments.</p><p>The following two issues are resolved in this study: i) How do model parameters affect the urban waterlogging simulation and visualization results? ii) How to integrate SWMM and Cesium 3D visualization platform?</p><p>In order to address the above research objectives, we will apply the following methodologies: i) We will analyse the parameters of the SWMM for the urban waterlogging visualization. Under the premise of understanding the development process of the urban waterlogging, we will analyse the modelling principle of the urban waterlogging, the mechanism of each part of the model separately. Then, we will find out the method of determining the model parameters of urban waterlogging and its influence on the simulation visualization results. ii) We will study integration methods of urban waterlogging model and 3D visualization platform. We will analyse the mechanism and process of urban waterlogging. We will also calculate the urban waterlogging process data by the SWMM, and establish a three-dimensional visualization platform by the node.js and Cesium, which can dynamically show the process of urban waterlogging. iii) We will complete the design and implementation of the interactive visualization platform of urban waterlogging. According to the above research, taking the Xianlin Campus of Nanjing Normal University as an example, we will build a dynamic interactive visualization system of urban waterlogging simulation based on Cesium. We will also verify the effectiveness of the system by comparing it with actual flood situation.</p><p>With this study, we expect to answer how model parameters affect the urban waterlogging simulation and visualization results. As expected results, we plan to build an interactive visualization system of urban waterlogging simulation based on Cesium, publish the flood calculation results into the 3D scene. This will make urban waterlogging process shown in the 3D scene. This visualization system is designed for different users, including specialists, government and individual. It means that you can use the system easily even if you are non-cartographers or non-IT-specialists.</p>

<p><strong>Abstract.</strong> Urban flooding refers to the phenomenon of water accumulation in cities due to strong precipitation or continuous precipitation exceeding urban drainage capacity. At present, the problem of urban flooding has become another major urban disease after other urban problems such as crowded people, traffic congestion and environmental pollution (Weiwu W, et al. 2015). In order to better cope with the problem of waterlogging, there are many hydrodynamic models that can accurately simulate the process of pipe network drainage and water accumulation. However, there are currently two limitations that limit the practicality of these hydrodynamic models. First, these models often require a large amount of input data as a support. These data exhibit multi-source, heterogeneity, multi-scale, multi-resolution and other characteristics, which bring great difficulty to data acquisition and processing (Dongdong Z, et al. 2014). Second, the analysis results of the model contain a large amount of waterlogging related information. However, this information is usually represented by simple texts and tables, which is not conducive to interpretation, transmission and visualization. Especially for location-related information, such as flooding points, flow information of various parts of the drainage pipe, water depth in the flooded area, etc., non-interactive two-dimensional maps or tables can cause inconvenience to disaster management and decision-making.</p><p>Therefore, a proper data model, respecting the needs of integration of multisource input and output data of waterlogging simulation and analysis and the needs of 3D visualization in front end, is needed. As a matter of fact, what in the last years has proven to be an emerging and effective approach is the adoption of standard-based, integrated semantic 3D virtual city models, which represent an information hub for most of the above-mentioned needs (Agugiaro et al. 2018). In particular, being based on open standards (e.g. on the CityGML standard proposed by the Open Geospatial Consortium), virtual city models firstly reduce the effort in terms of data preparation and provision. Secondly, they offer clear data structures, ontologies and semantics to facilitate data exchange between different domains and applications, and 3D visualization which is essential for crisis management (Herman L, Řezník T, 2015). However, a standardized and omni-comprehensive urban data model covering also the waterlogging domain is still missing. Even CityGML falls partially short when it comes to the definition of specific entities and attributes for waterlogging-related applications.</p><p>This study aims to propose an Waterlogging Application Domain Extension (ADE) for CityGML 2.0, which is used to integrate the multi-source input data and the rich waterlogging related information of waterlogging simulation and analysis. In order to achieve the above objectives, the following contents will be investigated:</p><p>(1) Analysis of urban hydrodynamic models’ function and their data characteristics</p><p>Analyze the function of the hydrodynamic model in the city, and explore the source, precision and organizational structure of the input and output data during the simulation and analysis of waterlogging.</p><p>(2) Construction of waterlogging data integration model based on CityGML</p><p>Based on the CityGML2.0 and its ADE mechanism, we will construct a conceptual model of waterlogging data integration standard, to describe the geometric and spatial structure of urban buildings, vegetation, land-use, underground pipelines, waterlogging related entities, etc. And more information about waterlogging, such as flooding points, drainage network carrying pressure, catchment area, etc., will be integrated into the corresponding fields of the data integration model in a reasonable way to facilitate Internet transmission and visualization.</p><p>(3) Interactive 3D visualization method based on the proposed waterlogging ADE</p><p>Based on the rich waterlogging-related information in the waterlogging ADE, combined with 3D visualization technology, the 3D dynamic interactive visualization method is explored, in order to provide more intuitive data support for waterlogging disaster management and decision-making.</p><p>Through this study, we expect to improve the efficiency and practicability of the current waterlogging simulation and analysis by integrating the waterlogging related data from multi-source based on CityGML. As the result of the research, we intend to develop a prototype system based on the multi-source waterlogging related data integration method proposed in this study. With this standard, we can provide unified and standard data support for the waterlogging model. At the same time, the results of the waterlogging models are automatically integrated into the data set for dynamic 3D visualization.</p>