Exploring the Integration of 3D GIS Topographic Data with Computational Fluid Dynamics (CFD) Models
Computational fluid dynamics (CFD) models are very powerful computational and modeling tools to simulate dispersion of atmospheric releases at the urban or local scale. It has been used widely to simulate such events in various locations (e.g., New York City) to assist planning, emergency responses and assessment of potential impacts.The input data for CFD models are intrinsically spatial and 3-dimensional in nature, including various environmental characteristics such as topography. The outputs from CFD will typically include concentration levels of the modeled hazardous substance in the 3D space and at different time periods. It is logical to expect that CFD models should be tightly integrated with GIS. Unfortunately, the developments of GIS and CFD have been very much independent.Some GIS 3D data have been extracted to another database before they were used by CFD models. Most CFD model outputs are used for visualization and display only, and have not been used effectively for impact assessment. In this paper, we will show how topographic data in GIS can be used in CFD models and demonstrate the importance of using 3D GIS data in CFD modeling. Despite the technological advancement in GIS in recent decades, GIS still have limited capabilities in handling real 3D data and in visualizing 3D objects. We will also show how 3D results from CFD models can be handled and visualized in GIS in a 2D environment. FEFLO-Urban, a general-purpose CFD model will be used in this study. The 3D topography and building database of George Mason University Fairfax campus Virginia, USA are used as the test data. Results showing that using and not using topographical data in CFD models will generate significantly different results, even though the campus topography is not highly complex. The importance of topography in CFD modeling is substantiated. The proposed method in visualizing the 3D model outputs is not elegant, but can show the 3D distributions of the model results in an understanding manner.
computational fluid dynamics models transport and dispersion models 3D modeling topography
Fernando E.Camelli David W.Wong Mukul Sonwalkar
Department of Computational and Data Sciences, George Mason University,MS 6A2, 4400 University Drive Department of Earth Systems and GeoInformation Sciences, George Mason University,MS 6A2, 4400 Univer
国际会议
北京国际地理信息系统学术讨论会第七届会议（7th International Workshop Geographical Information System
北京
英文
1-11
2007-09-14（万方平台首次上网日期，不代表论文的发表时间）