A rapid 3D reproduction system of dam-break floods constrained by post-disaster information
Introduction
The frequency and intensity of global extreme events have increased dramatically due to climate change and rapid economic growth. Floods are the most common and widespread type of natural disasters (Patel and Srivastava, 2013; Wan et al., 2014; Qiu et al., 2017). Dam-break floods are flood disasters that are usually associated with barrier lakes and obsolete reservoirs and are characterized by sudden occurrence, rapid expansion, high velocity and severe destruction (Li et al., 2013). As dam-break floods are a complex geographical phenomenon, the spatiotemporal process is often more important than the final spatial pattern, only the development process is clearly understood, and the deep-level change rules merit further exploration (Li et al., 2009).
Simulations have become a fundamental method for gaining insight into complex geographic processes (Bainbridge 2007; Chen et al., 2020). Moreover, developments in information science, earth observation technologies and geographical information science have strengthened the connections among geographical data, models and visualization, providing new perspectives and opportunities for spatiotemporal process modeling and the visualization of dam-break floods (Goodchild 2004; Li 2012; Chen et al., 2015; Achu et al., 2020; Kaur et al., 2019; Lü et al., 2019).
Many types of research have been conducted on flood simulation, with studies employing 1D hydraulic models, 2D hydraulic models, 1D-2D coupled hydraulic models and hydrological models. Most work has focused on simulation prediction and early warning of flood processes under ideal conditions (Ernst et al., 2010; Domingo et al., 2010; Pasquier et al., 2018; Wang et al., 2018). In practical use, it is difficult to use predictive models due to the uncertainty of data parameters, the instability of model structure, and the complexity of external conditions. Consequently, there is no universal method or software that can be applied to accurately predict floods under different conditions.
Disaster reproduction is a restored representation of the reason the disaster occurred, the process and the results. Whereas disaster simulation is forward-looking and focuses on the prediction of disaster evolution process, disaster reproduction emphasizes the restored representation of the disaster spatiotemporal process based on post-disaster information. The latter is mainly used for post-disaster education and media promotion, an intuitive and detailed 3D reproduction of the disaster process can improve the transmission efficiency of disaster information. Thus, disaster reproduction can overcome the shortcomings of traditional disaster education materials, which overly rely on ‘image and text publicity’, and ultimately improve people's disaster risk awareness and disaster prevention capabilities. After a dam-break flood, post-disaster information, such as the dam-break location, storage capacity, and affected area and buildings, can be used to help improve the simulation model of disasters (Li et al., 2015), thereby improving the spatiotemporal reproduction accuracy for this flood type. Moreover, transferring the disaster risk knowledge implied in numerical results to decision makers and the general public is a difficult task. The 3D virtual environment can intuitively display the simulation results of floods, which can improve users' perception and cognitive understanding of a dam-break flood disaster (Cheng et al., 2019; Chen et al. 2013, 2015, 2018; Havenith et al., 2019; Bailey and Chen, 2011; Lin et al., 2013a). Existing flood simulation software (e.g., FLO2D, HEC-RAS, etc.) emphasizes risk prediction, and problems with its application include tedious data processing, long simulation times and a high degree of specialization. Therefore, a rapid dam-break flood reproduction system that integrates data processing, simulation calculation and dynamic representation is urgently needed to support emergency decisions and the dissemination of risk knowledge.
This paper outlines the design and implementation of a rapid dam-break reproduction system. The system integrates a dam-break numerical model based on the cellular automata (CA) and the virtual geographic environment (VGE) framework, realizes rapid reproduction and augmented representation of a dam-break flood under the constraints of post-disaster information, and provides decision-making information support and risk knowledge for disaster-related persons. To realize the high-efficiency reproduction of a whole dam-break flood process, this system integrates data acquisition, model calculation and dynamic visualization functions. Through a user-friendly operation interface and flexible parameter configuration, the whole process of a dam-break flood can be represented in virtual 3D view. A dam-break flood that occurs anytime and anywhere can be rapidly reproduced by this system, and the reproduction time is reduced to less than 1 h, which greatly improves the 3D restoration efficiency of the spatiotemporal evolution of disasters.
Section snippets
Related works
Information systems play a significant role in flood disaster management (Granell et al., 2013; Ding et al., 2015). Disaster management mainly includes four stages, prevention, preparedness, response and recovery, with different flood information systems serving different stages (Carter 2008; Pearce 2003). Serving the prevention and preparation stages of flood disaster management are flood prevention and early warning systems formed by integrating atmospheric models, hydraulic models,
Overall research framework
Fig. 1 shows the overall research framework of this paper, which includes three main parts. First, the required parameters for simulation of the whole dam-break flood process are obtained based on remote sensing images and digital elevation model (DEM) data, which include the landslide body, disaster boundary, affected buildings, etc. These data are the basis for the reproduction and representation of the dam-break flood. Second, the cellular automata model of a dam-break flood is constructed,
System architecture
The architecture of the dam-break flood reproduction system is divided into four main layers, as shown in Fig. 6. The data layer is the foundation of the system, which adopts the mixed storage of a file system and PostgreSQL database. The service layer is the core of the system and is responsible for the processing and integration of relevant models. The representation layer provides multidimensional spatiotemporal visualization and augmented representation of the dam-break flood process to
Discussion
In this paper, we aim to develop an efficient 3D reproduction and visualization system of dam-break floods. For additional inspiration and information to consider, the advantages and limitations of this system are discussed below.
First, the reproduction time of the flood process is reduced to less than 1 h by using our system. The simulation of the spatiotemporal process of a dam-break flood is a very complex process, and abundant input data and parameters are needed to achieve accurate
Conclusions and future work
The rapid dam-break flood reproduction system is a platform developed to effectively represent disasters and disseminate information about disasters involving a dam-break flood. This system has the following characteristics. First, it supports the rapid processing of disaster information such as the disaster boundaries, dam-break location and roughness. Second, it integrates a numerical model of a dam-break flood based on CA to realize the spatiotemporal process simulation with the constraint
Software availability
The dam-break flood reproduction system was developed by Southwest Jiaotong University using C++ language on the Visual Studio 2010 platform. This system can be run on a standard PC. The software and part of the experimental data that support the findings of this study are available at figshare.com under the identifier https://figshare.com/s/a5026c73887472c3024f.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This paper was supported by the National Natural Science Foundation of China (Grant Nos. U2034202, 41871289and 41941019) and Sichuan Science and Technology Program (Grant No. 2020JDTD0003).
References (49)
- et al.
The role of virtual globes in geoscience
Comput. Geosci.
(2011) - et al.
A virtual globe-based three-dimensional dynamic visualization method for gas diffusion
Environ. Model. Software
(2019) - et al.
Enhancing integrated environmental modelling by designing resource-oriented interfaces
Environ. Model. Software
(2013) - et al.
3D WebGIS-based platform for debris flow early warning: a case study
Eng. Geol.
(2015) - et al.
Technical review of large-scale hydrological models for implementation in operational flood forecasting schemes on continental level
Environ. Model. Software
(2016) - et al.
A fusion visualization method for disaster information based on self-explanatory symbols and photorealistic scene cooperation
ISPRS Int. J. Geo-Inf.
(2019) - et al.
Virtual geographic environments (VGEs): a new generation of geographic analysis tool
Earth Sci. Rev.
(2013) - et al.
An integrated flood management system based on linking environmental models and disasters-related data
Environ. Model. Software
(2017) - et al.
Integrated sediment transport process modeling by coupling soil and water assessment tool and environmental fluid dynamics code
Environ. Model. Software
(2019) - et al.
Implementation of tidal turbines in MIKE 3 and Delft3D models of pentland firth & orkney waters
Ocean Coast Manag.
(2017)