Conclusion - Data Warehouse Systems: Design and Implementation - page 569

integrate a 2D city map with a 3D city model in a spatial database by using
a unified modeling approach.
Further, in recent years, we have witnessed an increasing interest in
4D (i.e., 3D plus time) spatial database systems. For example, there is an
urgent need to move from old 2D cadastral systems to 4D ones [ 40 ]. Climate
modeling and disaster prevention are other applications that depend on 4D
modeling. In most cases, going from a 3D to a 2D model is possible, but
moving up from 2D to 3D or 4D is not. Therefore, 3D geomodeling is a
research area that is urgently required to produce new 3D/4D geoinformation
systems. Further, the design and implementation of geometric and topological
database operations for moving 3D (i.e., 4D) objects is a focus of interest for
research and industry. Although several conceptual models supporting 3D
objects have been proposed in the literature considering both geometrical and
topological aspects (e.g., [ 106 , 112 , 241 ]), no current database management
system supports 3D topological models, yet many support 2D topologies. A
key need for this consists in developing spatial indexes for topological models.
In summary, the directions in this field are the seamless integration
between 2D and 3D data models to be usable in both worlds and the
development of 3D/4D geographic information systems. A step in this
direction is CityGML, 1 an open information model for representing, storing,
and exchanging 3D city and landscape models. CityGML is implemented
as an application schema for the Geography Markup Language (GML)
[ 110 ], the standard for spatial data exchange issued by the Open Geospatial
Consortium, and provides a standard for describing the geometry, topology,
and semantics of 3D objects. CityGML is highly scalable supporting not
only buildings but also whole sites, districts, cities, regions, and countries.
CityGML provides 3D content, allowing visualization through several appli-
cations, but it also allows users to share virtual 3D city and landscape
models for sophisticated analysis, for example, environmental simulations,
energy demand analysis, city management, and disaster management. As an
example, the application of CityGML to a case study in the Netherlands can
be found in [ 216 ]. There is a series of conferences specifically devoted to 3D
GeoInformation. The volumes of these conferences are published by Springer,
the last one being [ 165 ].
Given the discussion above, the reader should at this point not be surprised
to know that very few publications have addressed the combination of data
warehouses and 3D objects. As an example, the BioMap data warehouse
[ 121 , 122 ] integrates biological data sources in order to provide integrated
sequence/structure/function resources that support analysis, mining, and
visualization of functional genomic data. Extending conceptual models for
data warehousing in this direction requires first the definition of 3D spatial
1 http://www.citygml.org/