13 Error in spatial data
13.1 This report deals with the positional accuracy, error and uncertainty of spatial data. This is but one aspect of error inherent in spatial data. Some examples of other components are attribute accuracy, lineage, logical consistency, completeness and temporal accuracy. For more information see Burrough and McDonnell, 1998, pp220-240, 241-264, Chrisman, 1991, pp165-174 and Heuvelink, 1998.
13.2 As has been discussed, there are numerous factors that affect the positional accuracy and reliability of spatial data and contribute to its uncertainty. Many of these factors have been briefly described in this report. That is, the nature of spatial data and how it is conceptualised, how it is modeled, located in space, captured and manipulated, as well as computing and human factors. At each stage in the process error is present. That error is inherited by subsequent processes and is propagated throughout derived spatial data sets (Heuvelink, 1998, p 5 and Foote and Huebner, 1995). It is not well understood ‘how these errors contribute to the uncertainties in the results of GIS operations and computational models’ (Heuvelink, 1998, p 5).
13.3 Some aspects of the error, for example, that attributable to geodesy, surveying and photogrammetry are well understood and can be measured (Chrisman, 1991, p172). This error can be minimised through training, carefully targeted procedures and quality assurance checks.
13.4 Other aspects of the error present, for example, that attributable to the generalisation, approximation or abstraction of geographical phenomena, the conscientiousness of GIS operators, or through inappropriate people having write access to key datasets are less well understood and are of a random nature. In most cases this error can also be minimised through training, carefully targeted procedures and quality assurance checks.
13.5 Some aspects of the error are very difficult to detect and require a thorough and intimate knowledge of the spatial data, as well as of GIS principles and algorithms. Some examples are those errors attributable to faulty GIS algorithms or to an inappropriate algorithm being used for a particular task. Training, carefully targeted procedures and quality assurance checks may also detect these types of error.
13.6 “The usual view of errors and uncertainties is that they are bad. This is not necessarily so, however, because it can be useful to know how errors and uncertainties occur, how they can be managed and possibly reduced, and how knowledge of errors and error propagation can be used to improve our understanding of spatial patterns and processes. … A good understanding of errors and error propagation leads to active quality control.” (Burrough and McDonnell, 1998, p221).
To make suggestions or corrections to
this web page, please contact posacc@geoinnovations.com.au.
The information on this web page is © Copyright
1999 B.S.Bannerman.
All commercial rights reserved.
B.S.Bannerman materials published on the Internet are protected by copyright law. Apart from fair dealing for the purposes of private study, research, criticism or review, as permitted under the Copyright Act 1968, no part may be reproduced or reused for any commercial purposes whatsoever.