بخشی از مقاله انگلیسی:
Storm surge is an abnormal rise of sea surface caused by atmospheric forces, including wind stress and atmospheric pressure associated with extra-tropical and tropical cyclones (Flather, 2001). It has a great impact on coastal regions and may cause severe damage to coastal structures and loss of human lives and properties (Madsen and Jakobsen, 2004). Yangtze Estuary and Hangzhou Bay (YE-HB) are located in the east coast of China, facing open sea. They are subjected to frequent threats from tropical cyclones, suffering massive damage from resulting strong wind, storm surge and inland flooding. These tropical cyclones are mainly generated on the ocean surface, either east of the Philippine Islands, or near Guam. According to 1949–۲۰۰۸ statistics, about 3.5 typhoons occurred in these areas every year which made storm surge become one of the most serious oceanic disasters in the region. Storm surge forecasting is considered to be the best way to mitigate its damage. Surge generation by extra-tropical storms was studied in Europe (e.g., Kliem et al., 2006; Zampato et al., 2006), while it caused by hurricanes in the Gulf of Mexico and Eastern USA was studied in North America (e.g. Peng et al., 2006a, b; Weisberg and Zheng, 2008), and one by typhoons was studied in Asia and Oceania (e.g. McInnes et al., 2002; Jain et al., 2007). The forecasting methods are divided into two general categories: the empirical statistical forecasting methods (e.g., Conner et al., 1957; Harris, 1959) and the numerical prediction methods (e.g., Jelesnianski, 1966; 1967). The former methods require a sufficiently long-term observation, so they are greatly restricted; the latter one use numerical prediction results that provide information on storm forecasts, such as forecast surface wind, pressure fields, and seawater motion in the specific study area by solving a series of complex dynamic equations. With advances in computer technology, most countries adopted these methods for forecasting storm surges. Moreover, numerical prediction is one of primary measures for storm surge disaster control. For instance, the famous SLOSH model has been widely used for two-dimensional applications (Jelesnianski et al., 1992). Numerical experiments helped us to gain insight into storm surge mechanisms. They in turn contribute to improving storm surge theories. With the appearance of high-speed computers, numerical approximations to the governing equations have been developed. In early period, structured computational grids (generally squares) in two dimensions (vertically integrated) was used (Cao and Zhu, 2000; Dietsche et al., 2007; Hu et al., 2007). However, the 2D models may overestimate (or underestimate) bottom stress. Physically unrealistic parameterizations or other techniques of surface stress are necessary to calibrate the model. They essentially ignore the vertical flow structure that may have a significant effect on simulation (Weisberg and Zheng, 2008). Then, the 3D storm surge model was developed. It allows the inclusion of mass redistribution (the balance between the net water flux into an area and the change in water level) and the representation of more realistic shorelines and bathymetries. Research is also under way to include the influence of nonlinear interaction between astronomical tides and storm surge to reflect the obvious periodic oscillations seen in storm observations (Wang and Chai, 1989). Unstructured, finite element numerical techniques were developed for storm surge models that allowed the use of computational grids composed of unstructured triangles. These grids are easily configured to represent complex topographic and bathymetric features (e.g., irregular coastlines, rivers, inlets and barrier islands, etc.). Therefore, they can provide a very high resolution in area of interest (Chen et al., 2003;2007). The accuracy of wind field input during a tropical cyclone is crucial for the results of storm surge modeling. Some classic parametric pressure or wind models are frequently used to conveniently generate symmetric wind fields (Holland, 1980). These parametric cyclone models have demonstrated accuracy of a few feet wind fields when running past cyclones. With accurate wind fields, the 3D storm surge models are now able to fully account for the effect of the astronomical tide on the total water level. They also can forecast the potential flooding during a storm. In this study, the original Finite-Volume Coastal Ocean Model (FVCOM) model, which used the wet–dry grid point method, was revised by adding the water level change due to the atmospheric pressure field for the 3D storm surge model. In general, the storm model data input uses a DOS operating system mode. Output data files were comprised of the plotted static images. Thus, the accurateness and time effects of storm surge prediction are greatly affected. In addition, effective analysis and visual comprehension are also obstructed (Weisberg and Zheng, 2006a, b; Guo et al., 2009; Yin et al., 2009). This is a frontier research field through combing with improving meteorological observations and modeling, accuracy of bathymetric information, computational capabilities, and hurricane warning systems can enhance national disaster plans at last (Tsanis and Gad, 2001; Castrogiovanni et al., 2005). Geographic information system (GIS) enables users to handle a large amount of data in a short time frame, and allows them to allocate more time to study the engineering tasks instead of spending excess time on preliminary tasks. Numerous efforts have been made to integrate hydrodynamic models with GIS (e. g., Tsanis and Boyle, 2001; Gemitzi and Tolikas , 2004; Naoum et al., 2005; Ng et al., 2009). Such integrations provide not only efficient modeling pre-processing and post-processing, but also the system with spatial data management, analysis, and visualization functionalities (Ng et al., 2009). The paper attempted to establish the integration of a symmetric storm-induced wind and pressure field with a background, finite volume ocean model and GIS functionality. We integrated all of the modeling process into the same environment in order to fully use the advantages of both GIS and numerical modeling, including mesh calculation, model parameter selection and setting, model computation, and the final results visualization. The system can complete and improve the operability of model application and the decision-making efficiency. Thus, we solved the problems associated with traditional storm surge numerical method including issues with data collection, visualization, spatial query, and analysis of simulative results. As a synthetic measure and inversion/prediction for storm-surge disaster control, the system combines GIS with storm-surge numerical model felicitously, and provides well operating environment for visualization, spatial query and analysis of simulative results. Using the example of the storm surge which was induced by No. 8114 Typhoon Agnes crossing the YE-HB, the paper discusses the essential integrated technology in which we embedded ComGIS DLL (MarineTools Pro) of the storm surge numerical model inside the ArcGIS implemental software (Merkel et al., 2008; Tsanis and Gad, 2001; Ng et al., 2009). The development of the integrated system of storm surge numerical prediction supported by GIS is also included. The ways and specialties of the Desktop GIS development are summed up. The integrated methods of GIS and professional applied model are analyzed. The system shows seamless integration for GIS and all phases of the numerical model, including model preprocessing, model computation, model post-processing, and results visualization. For example, the Digital elevation model was utilized to make water landforms to generate the difference-grid data automatically. The spatial database technology of GIS was then applied to realize chart show and allow spatial query on information, such as water depth, typhoon route and tidal level. It also enabled to correlate with prediction of storm surge. GIS visualization technology was applied to simulate coastal dynamic environments (such as flow field, storm surge water accretion distribution) and their transformation process in typhoon-affected regions. The integrated model was used to reproduce the storm surge generated by Typhoon Agnes (No. 8114), and simulate typhoon-induced wind field and water elevations of YEHB. The results show that the system is good, easy to operate. It can improve the efficiency of decision-making for the storm surge numerical model. The paper is organized as follows. In Sect. 2, the numerical model coupled in system is described. The model consists of a symmetric storm-induced wind model and a revised FVCOM hydrodynamic model that involves the effect of the atmospheric pressure change. In Sect. 3, the method of coupling GIS with hydrological modeling is presented. Then, the process of YE-HB storm surge induced by Typhoon Agnes (No. 8114) is simulated to validate the developed MarineTools Pro, and the simulation results are compared with the observed data. The conclusions are outlined in Sect. 4.