GIS, Its problems and potential solution approaches in the IT profession
“A geographic information system (GIS) is a computer system for capturing, storing, checking, and displaying data related to positions on Earth’s surface”. GIS can show many different kinds of data on one map, such as streets, buildings, and vegetation. This enables people to more easily see, analyze, and understand patterns and relationships. (Illustration courtesy U.S. Government Accountability Office)
“A geographic information system or GIS integrates data, hardware, software, and GPS to assist in the analysis and display of geographically referenced information.” (Lifewire, 2017)
“An information system that is designed to work with data referenced by spatial or geographic coordinates. In other words, a GIS is both a system with specific capabilities for spatially-referenced data, as well as a set of operations for working [analysis] with the data.”(Star and Estes, 1990)
In my view, GIS can be summarized as “A geographic information system (GIS) lets us visualize, question, analyze, and interpret data to understand relationships, patterns, and trends.”
Spatial data represents the location, size, and shape of an object on planet Earth such as a building, lake, mountain, or township. Spatial data may also include attributes that provide more information about the entity that is being represented. Geographic Information Systems (GIS) or other specialized software applications can be used to access, visualize, manipulate, and analyze geospatial data.
Microsoft introduced two spatial data types with SQL Server 2008: geometry and geography. Geometry types are represented as points on a planar, or flat-earth, surface. An example would be (5,2) where the first number represents that point’s position on the horizontal (x) axis and the second number represents the point’s position on the vertical (y) axis. Geography spatial data types, on the other hand, are represented as latitudinal and longitudinal degrees, as on Earth or other earth-like surfaces.
The availability of spatial databases and widespread use of geographic information systems has stimulated increasing interest in the analysis and modeling of spatial data. Spatial data analysis focuses on detecting patterns, and on exploring and modeling relationships between them in order to understand the processes responsible for their emergence. In this way, the role of space is emphasized, and our understanding of the working and representation of space, spatial patterns, and processes is enhanced. In applied research, the recognition of the spatial dimension often yields different and more meaningful results and helps to avoid erroneous conclusions.
THE APPLICATION APPROACH
A slight modification of the process-oriented approach yields a definition which categorizes GIS according to the type of information being handled. For example, Pavlidis’ classification scheme includes natural resource inventory systems, urban systems, planning and evaluation systems, management command and control systems, and citizen scientific systems (Pavlidis,1982). Applications in forestry may cut across several of these categories but are primarily concerned with inventory, planning, and management. An area of greatly increased attention in the field of land records, or multi-purpose cadaster, systems that use the individual parcels as basic building blocks (McLaughlin,1984). While defining GIS on the basis of applications may help to illustrate the scope of the field, it does not enable one to distinguish GIS from other forms of automated geographic data processing. Geographic information systems are independent of both scale and subject matter.
THE PROCESS-ORIENTED APPROACH
Process-oriented definitions, based on the idea that an information system consists of several integrated subsystems that help convert geographic data into useful information were formulated originally in the early 1970s by Tomlinson and others (Catkins and Tomlinson, 1977). Logically, the entire system must include procedures for the input, storage, retrieval, analysis, and output of geographic information. The value of such systems is determined by their ability to deliver timely and useful information. Although the intentions of this process-oriented definition are quite clear, the application of the definition is far too inclusive to help distinguish GIS from computer cartography, location-allocation exercises, or even statistical analysis. By applying such a broad definition one could argue that almost any successful master’s thesis in geography involves the creation of an operational GIS. Similarly, the production of an atlas also would seem to include all the necessary subsystems of a GIS. A process-oriented definition is, however, an extremely valuable frogman organizational perspective, as well as for establishing the notion that a system is something that is dynamic and should be viewed as a commitment to long term operation. Finally, any form of the process-oriented definition of GIS emphasizes the end use of the information and, in fact, need not imply that automation is involved at all in the processing (Pokier, 1985),
THE TOOLBOX APPROACH
The toolbox definition of GIS derives from the idea that such a system incorporates a sophisticated set of computer-based procedures and algorithms for handling spatial data. Published works by Tomlinson and Boyle (1981) and Dangermond (1983), for example, provide very complete delineations of the operational software functions that one should find in a full-featured GIS. Typically, these tools are organized according to the needs of each process-oriented subsystem (e.g., input, analysis, or output). The toolbox definition implies that all of these functions must be present and should work together efficiently to enhance the transfer of a variety of different types of geographical data through the system and ultimately into the hands of the end-user. Therefore, even though they are important components of automated geography, neither digitizing, image processing, nor automated mapping systems qualify as GIS because they do not possess all the necessary tools and do not provide the overall integration of functions. While checklists are very useful for evaluating different systems, they fail to provide a viable definition of the field.
THE APPLICATION APPROACH
A slight modification of the process-oriented approach yields a definition which categorizes GIS according to the type of information being handled. For example, Pavli is’ classification scheme includes natural resource inventory systems, urban systems, planning and evaluation systems, management command and control systems, and citizen scientific systems (Pavlidis,1982). Applications in forestry may cut across several of these categories but are primarily concerned with inventory, planning, and management. An area of greatly increased attention in the field of land records, or multi-purpose cadaster, systems that use the individual parcels as basic building blocks (McLaughlin, 1984). While defining GIS on the basis of applications may help to illustrate the scope of the field, it does not enable one to distinguish GIS from other forms of automated geographic data processing. Geographic information systems are independent of both scale and subject matter.
THE DATABASE APPROACH
The database approach refines the toolbox definition of GIS by stressing the ease of the interaction of the other tools with the database. For example, Goodchild states, “A GIS is best defined as a system which uses a spatial database to provide answers to queries of a geographical nature. …The generic GIS thus can be viewed as a number of specialized spatial routines laid over a standard relational database management system” (Goodchild, 1985). Piquet would agree that a GIS must start with an inappropriate data model.
GIS and CAD
The following are the criteria on which we can differentiate CAD and GIS.
Modeling
CAD models things in the real world. GIS models the world itself. Therefore, GIS uses geographic coordinates systems and world map projections while CAD coordinates are relative to the object being modeled and are not usually relative to any particular place on earth.
Objects
CAD objects include lines, circles, arcs, text, etc. using layers, blocks, internal data, and dimensions. CAD objects don’t know about each other, even though they may touch or overlap.
GIS objects know about each other:
• GIS understands networks. For instance, the lines describing streets are related to one another.
• GIS understands enclosed areas (polygons) and their associativity with other objects.
• GIS understands connectivity, conductivity, and associativity which enable spatial analysis.
Topology
The primary difference between CAD and GIS is topology. GIS has it, CAD doesn’t. In a CAD environment, the objects (lines, polylines, points, etc.) have no relationships between them. Topology brings these objects together into logical groups to form real-world models.
• Node topology allows spatial analysis, such as buffering to determine other objects within a certain range.
• Network topology allows modeling of direction and resistance. Path tracing finds the fastest or best route. Flood tracing determines the maximum flow from a given point and network resistance. As with node topology, buffer analysis can be applied to networks too.
• Polygon topology enables polygons to have relationships. Polygons also have centroids which can be used to hold data relevant to the polygons. Polygon spatial analysis includes overlay analysis such as determining parcels in a floodplain. Polygons can be “dissolved” using attributes with common values to remove interior lines, in effect aggregating polygons within the same class.
Topology and spatial analysis differentiate GIS from CAD.
Data Management
GIS separates object storage from object display, combining data from multiple sources into a virtual data warehouse. That data can then be used in any number of separately defined analyses or presentations. CAD systems carry baggage such as line color, line width, etc. that is not relevant to the data itself.
GIS systems are usually disk-based and can model larger areas than CAD implementations which are usually memory-based. For instance, CAD files are typically smaller, such as product designs as compared to regional, state, or even world models in GIS.
The Trend
While the distinction between CAD and GIS is grey now, as features are added to CAD systems, the distinction will blur even more. And construction, GIS for initial planning, and layout.
The table summarizes the main differences.
In the last five years, DBMS made a large step toward the maintenance of geometries as GIS used to manage them. The support of 2D objects with 3D coordinates is adopted by all mainstream DBMS. The offered functions and operations are predominantly in the 2D domain. The DBMS spatial schemas have to be extended to fully represent the third dimension (first with the simple volumetric objects and later with more complex 3D data types). Concepts for 3D objects and prototype implementations are already reported, DBMS have to make the next step and natively support them. 3D operations and functions have to be developed not only for the volumetric object but also for all other objects embedded in 3D space. 3D functionality is next to be considered. It should be remembered that Spatial DBMS is a place for storage and management, and less intended for extensive analyses. The 3D functionality should not be completely taken away from frontend applications such as GIS and CAD/AEC. 3D Spatial DBMS should provide the basic (generic) 3D functions, such as computing volumes and finding neighbors. Complex analyses have to be attributed to the applications.
Some existing data types are clearly not sufficient for the purpose of some applications. A very typical example is multipoint. It was definitely not designed for large amounts of points as from laser scanning. DBMS fails to handle efficiently such amounts of data until now. Such points need special treatment. On the one hand, with the progress of data collection techniques, the amounts of points will only increase. Many laser scanning companies are increasingly getting concerned about the management of such data. On the other hand, the advances in 3D modeling would require more intelligent management of both raw and processed data. Clearly a new spatial data type with internal structure and index has to be developed.
Conclusion
Spatial analysis techniques in GIS that have been developed over the past half-century many topics can only be covered to a limited depth, whilst others have been omitted because they are not implemented in current mainstream GIS products. This is a rapidly changing field and increasingly GIS packages are including analytical tools as standard built-in facilities or as optional toolsets, add-ins, or analysts. In many instances, such facilities are provided by the original software suppliers (commercial vendors or collaborative non-commercial development teams) whilst in other cases, facilities have been developed and are provided by third parties. Many products offer software development kits (SDKs), programming languages and language support, scripting facilities, and/or special interfaces for developing one’s own analytical tools or variants.
References and Bibliography
Society, N. and Society, N. (2017). GIS (geographic information system). [Online] National Geographic Society. Available at: https://www.nationalgeographic.org/encyclopedia/geographic-information-system-gis/ [Accessed 13 Jun. 2017].
Lifewire. (2017). Geographic Information Systems — Definition and Uses. [online] Available at: https://www.lifewire.com/geographic-information-system-gis-1683309 [Accessed 14 Jun. 2017].
Arens, C, 2003, Maintaining Reality; Modelling 3D spatial objects in a GeoDBMS using a 3D primitive, Master’s Thesis TU Delft, 2003, 76 p
Hoefsloot, M. 2006, Storing Point clouds in DBMS, MSC Thesis, available at http://www.gdmc.nl/publications, 80p.
SearchSQLServer. (2017). What is spatial data? — Definition from WhatIs.com. [online] Available at: http://searchsqlserver.techtarget.com/definition/spatial-data [Accessed 16 Jul. 2017].
Fischer, M. and Wang, J. (2011). Spatial data analysis. Heidelberg: Springer.