Premium
Defining and Delineating the Central Areas of Towns for Statistical Monitoring Using Continuous Surface Representations
Author(s) -
ThurstainGoodwin Mark,
Unwin David
Publication year - 2000
Publication title -
transactions in gis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.721
H-Index - 63
eISSN - 1467-9671
pISSN - 1361-1682
DOI - 10.1111/1467-9671.00058
Subject(s) - kernel density estimation , granularity , transformation (genetics) , computer science , geography , code (set theory) , categorical variable , index (typography) , kernel (algebra) , data mining , cartography , statistics , mathematics , machine learning , biochemistry , chemistry , set (abstract data type) , combinatorics , estimator , world wide web , gene , programming language , operating system
In the UK, the increasing availability of very high spatial resolution data using the unit post code as its geo‐reference is making possible new kinds of urban analysis and modelling. However, at this resolution the granularity of the data used to represent urban functions makes it difficult to apply traditional analytical and modelling methods. An alternative suggested here is to use kernel density estimation to transform these data from point or area ‘objects’ into continuous surfaces of spatial densities. The use of this transformation is illustrated by a study in which we attempt to develop a robust, generally applicable methodology for identifying the central areas of UK towns for the purpose of statistical reporting and comparison. Continuous density transformations from unit post code data relating to a series of indicators of town centredness created using Arc/InfoTM are normalised and then summed to give a composite ‘Index of Town Centredness’. Selection of key contours on these index surfaces enables town centres to be delineated.