This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Database Management system is a collection of related data. Data stored can be accessed in variety ways. A database is not application specific because it servers many purposes. The system can be used in shop's stores, telecommunications to mention few and any other scientific and non -scientific purposes.
In GIS world, a database term has gained importance after realizing the data management problems that were facing GIS with increased geographic information demand. Different database model has been invented and implemented.
Today, a geographic information system can be defined as the spatial database together with a set of operators (GIS interface). The most notable difference between the spatial database and traditional database is the nature of data to be stored. In a traditional database mostly deals with alphanumeric whereby in a spatial database in addition to alphanumeric are the geometry objects (point, polygon, lines).
This report focus on how the role of a database system has changed over past two decades in terms of data capture, storing, retrieval, computational, presentation and analysis. The report describes diverse database technologies that have been adopted and developed to make GIS more useful, powerful and reliable.
SPATIAL DATA ACQUISITION
It may be assumed that spatial data capture is not part of the database simply because the initial data acquisition process is done outside the database system but this is not realistic, in principle, regardless what capture devices are being used, the nature and form of data captured has a direct relation with the database structures.
Spatial data sources may be from remote sensing devices, direct survey or paper based maps. In past decades emphasis has been to integrate the spatial database and spatial data sources. The most notable achievement is the integration of spatial data sources with spatial databases (Arc info with ERDAS). A link between ESRI ARC/INFO and ERDAS 7.3 was introduced in 1988. The "ERDAS-ARC/INFO Live Link" allowed the mapping community to use technology from both companies to deliver high quality image display and image processing with powerful GIS capability.
Kressler et al. (2006) integrated LiDAR, image data and spatial databases to produce a higher resolution building/land classification map.
A Data model in RDMS is based on the concept proposed by Dr.Codd and availability of Structural Query language (SQL) for storing and retrieving data. The question has been if these concepts fit for spatial databases, this is because traditional RDMS is mainly for numbers, dates, strings types which in contrast spatial database has an additional shape data type made of points, polygon and lines and other type of vector representation. They often have a complex structure: a spatial data object may be composed of a single point or several thousands of polygons or various collections of polygons, lines, and points often with complicated consistency constraints. (Egenhofer; Glasgow; Gunther; Herring; Peuquet 1999)
Based on the nature of spatial data sometimes it is impossible to implement Dr. Codd rules, as for example spatial data is represented in( x,y,z) phenomenon, hence violet data atomicity behaviour of Relational Database management systems stated in Codd's rules. To address this weakness Object Database management System was initially suggested but has failed to gain popularity because of high vendor's investment in RDMS. Instead, the RDMS pioneer has integrated the ODBMS capabilities within their standard RDMS to create Hybrid Object Relational Database Management System (ORDBMS).
Two of the major commercial DBMS vendors have released spatial database extensions to their standard ORDBMS products: IBM offers two solutions - DB2 Spatial Extender and Informix spatial Datablade- and Oracle has a spatial option (Longley, Goodchild, Maguire and Rhind 2005)
SPATIAL DATA STORAGE AND RETRIEVAL
Initially back in 70's, GIS software was built in a file based system where by spatial and non- spatial data were stored in files and provided fast access to data. This was simply known as a first spatial database generation. This was easily facilitated by a stand-alone computer system architecture that existed during that time. Sharing data within the organisation from different vendors and other applications was still a problem.
The second spatial database generation (Hybrid GIS Model) was then improved to semi- database management system, where spatial data are stored in files based system and related attributes stored in a relational database management system. This system provides a substantial storage and retrieval mechanism; spatial data (maps) are stored in a file system that makes utilize fast direct input /output access technique. The GIS software provides a link between spatial data (maps) with the associated spatial attributes. In another hand, the system with one component stored outside RDMS may result in data inconsistency, integrity constrained and data redundancy.
With a development of computer technology, new database methods have been introduced to take advantage of relational database by storing both spatial data and attributes in a database. In the mid-'90s, new technology emerged that enabled spatial data to be stored in relational databases (often referred to as spatially enabling the database), opening a new era of broad scalability and the support of large, non tiled, continuous data layers. (ESRI White Paper - January 2003). Most of the spatial data today, are in form of vector objects (points, lines and polygon) that can be stored in tables together with its associated attributes.
Moreover, database storage has changed from a stand-alone computer system to client/ server architecture that also adds some other challenges in terms of storage and performance. As the spatial database becomes very large and equally important with increased in data accessibility and manipulation, the traditional client/ server dependent on a single central server has in addition become inadequate. Other methods of implementation have been proposed and implemented; today the distributed database is the mostly accepted technology where there are a large number of users accessing and querying data simultaneously.
The Structured Query language (SQL) is a widely accepted database language for interacting relational tables in storing and retrieving data. This language defines data types and operators to be used in relational tables based in its first concept of dealing with alphanumeric data. For many years, the Structured Query Language SQL has dominated the database market. There has been a long discussion in the literature as to whether SQL is suitable for querying spatial databases. (Egenhofer; Glasgow; Gunther; Herring; Peuquet 1999).
In 1997, standards were proposed by the Open Geospatial Consortium (OGC) for extending SQL for spatial data. Furthermore, most commercial and open-source DBMS vendors support extensions for handling spatial data. In extended SQL spatial allows a user to define data type and operation where in standard SQL data type is based on standard defined data types, the extension SQL helps users defines their data types and also complex geometric queries are as well possible. Unfortunately, standard query languages like Structured Query Languages, SQL, are restricted to supporting queries based on a relation join, sub-query, group functions and query combination operator (Healey. 1991).
The extended SQL operator as defined by OGC (buffer, isempty, equal, contains, cross, distance, intersection) are designed specifically for spatial geometry and attribute manipulations. Traditional RDMS uses B-tree structure technique for accessing, sorting data (numbers, strings and date). Spatial data that may consist of shapes that overlapping to each other, the B-tree method cannot be effectively applied.
Most of the database vendor and researchers have tried to find out suitable spatial accessing methods. The R-tree and quadtree have been deployed for spatial data indexing and querying.
SPATIAL DATA PRESENTATION AND ANALYSIS
With the computer technology breakthrough and need of enterprise data access and sharing functions, meaning that the spatial database has to be accessed in a client-server architecture environment using different platforms, this brings the concept of database independent to a platform and application programs. Geographic information results today can be accessed via computers in office or through the web interface and mobile devices.
Spatial data results from the database may be ranging from a simple geographical map to a numerical analytical result. The later has been a problem since most of the spatial database has inadequate analytical capabilities.
The question has been either embedding spatial statistical tools in GIS environment or integrating spatial tools in GIS. Zhang and Griffith (2000) suggest the possibility of integrating GIS components and spatial statistical analysis in a current proprietary Database Management System (DBMS: Microsoft Access), which can be easily extended to incorporate statistical analysis capabilities. In my opinion, I argue that though adaptability of MS- Access is practical but by itself, it is not widely used database in an enterprise environment. Therefore, further research should be conducted in the leading commercial database platforms.
From its inception, the GIS has been facing problems of not having a defined standards, as every vendor has its own standards that are making it difficult in working with spatial database and hence difficulty in integrating data from different sources.