5B_1_Neogeography for the rural urban classification of england and wales


Published on

  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

5B_1_Neogeography for the rural urban classification of england and wales

  1. 1. GISRUK 2010 – UCL LONDON UK Thursday, 15 th April 2010 <ul><li>Neogeography for the Rural Urban Classification of England and Wales </li></ul>Department of Geography Environment and Development Studies Birkbeck, University of London Malet Street London WC1E 7HX [email_address] j.shepherd@bbk.ac.uk  Maurizio Gibin and John Shepherd
  2. 2. Outline The Rural Urban Classification for England and Wales a brief explanation the methodology Neogeography for the classification what why who The web mashup architecture a brief tour of the features Conclusion
  3. 3. The Rural Urban Classification <ul><li>The Rural/Urban Definition was introduced in 2004 as a joint project between a number of Government Departments and was delivered by the Rural Evidence Research Centre at Birkbeck College (RERC) (John wave your hand if you are here!) </li></ul><ul><li>The definition has been developed for Output Areas (OAs), SOAs and Wards. </li></ul><ul><li>Focus on OA </li></ul><ul><li>The Definition adopts a settlement-based approach, comprising four settlement types, of which three are rural – MORPHOLOGY: </li></ul><ul><ul><li>Urban (population over 10,000) </li></ul></ul><ul><ul><li>Town and Fringe </li></ul></ul><ul><ul><li>Village </li></ul></ul><ul><ul><li>Hamlet and Isolated Dwellings </li></ul></ul><ul><li>The settlement types are assigned to either a 'sparse' or 'less sparse' regional setting to give eight classes - CONTEXT: </li></ul><ul><ul><li>Urban (Sparse) </li></ul></ul><ul><ul><li>Town and Fringe (Sparse) </li></ul></ul><ul><ul><li>Village (Sparse) </li></ul></ul><ul><ul><li>Hamlet and Isolated Dwellings (Sparse) </li></ul></ul><ul><ul><li>Urban (Less Sparse) </li></ul></ul><ul><ul><li>Town and Fringe (Less Sparse) </li></ul></ul><ul><ul><li>Village (Less Sparse) </li></ul></ul><ul><ul><li>Hamlet and Isolated Dwellings (Less Sparse) </li></ul></ul>http://www.ons.gov.uk/about-statistics/geography/products/area-classifications/rural-urban-definition-and-la-classification/rural-urban-definition/index.html
  4. 4. The Methodology <ul><li>The process of definition described here has as its initial ‘raw material’ all settlements where population is <10,000. </li></ul><ul><li>Royal Mail’s ‘Postcode Address File’ (PAF). </li></ul><ul><li>Grouping of every postal address on the basis of the 1 hectare (100m x 100m) cell within which it falls. </li></ul><ul><li>Density calculated at different bandwidths: </li></ul><ul><li>with an increase in the more areas of open space may be included and average densities will decline. Importantly, we can make use of this property in order to identify and classify rural settlements. </li></ul><ul><li>The rate at which density changes away from the ‘focus’ cell is a function of local settlement structure. </li></ul>
  5. 5. Morphology <ul><li>Furthermore, different morphologies or settlement forms can be shown to have different typical density ‘profiles’. </li></ul><ul><li>‘ Density profiles’ can thus be created using a series of different area or ‘window’ sizes. In other words, density profiles can be created by calculating densities at a series of fixed scales - in our case 200m, 400m, 800m and 1600m: </li></ul>Settlement 200m 400m 800m 1600m Peri-urban 0.3 0.59 1.57 2.8 Scattered dwellings 0.39 0.17 0.15 0.23 Hamlet 0.65 0.21 0.13 0.2 Village envelope 0.94 1.15 1.31 0.59 Village envelope (in peri-urban) 2.96 3.27 1.81 2.13 Village 3.81 2.28 0.83 0.58 Urban fringe 6.46 7.21 5.9 4.68 Small town 8.23 8.99 8.29 5.59 Urban Areas (above 10k) 16.09 15.17 13.78 11.89
  6. 6. Context <ul><li>Refers to the broader setting in which are located: </li></ul><ul><li>wider accessibility of a settlement, the sparsity of population and the potential costs of overcoming distance to supply that settlement with various public and private services. </li></ul><ul><li>Density profiles used at much larger scales 10km, 20km and 30km to characterize aspects of accessibility and population sparsity. </li></ul><ul><li>On the basis of these measures it is possible to identify areas where population is ‘sparse’ at the particular scale. By assigning these measures to 2001 Census Output Areas and focusing on the sparsest 5 percent in each case, three indicators of ‘sparsity’ are obtained. </li></ul>Context Residential delivery points density per ha Sparse at the 10km scale < 0 .3932 Sparse at the 20km scale < 0.41 Sparse at the 30km scale < 0.4224
  7. 7. The final classification…morphology and context <ul><li>Output Areas are classified by ‘hierarchical privileging’, that is, if an Output Area has 50 percent by area of a particular settlement morphology, this classification is used. </li></ul><ul><li>Where an Output Area did not contain a dominant morphological type then the largest settlement character is ‘privileged’ with the Output Area classification. </li></ul>
  8. 8. Neogeography <ul><li>“ Neogeography combines the complex techniques of cartography and GIS and places them within reach of users and developers.” </li></ul>Introduction to Neogeography (2006) Andrew J. Turner. O'Reilly Media, Inc. ISBN: 978-0-596-52995-6.
  9. 9. The web mashup <ul><li>Based on Google Maps and on the Rural Urban Classification, OA level </li></ul><ul><li>Layout similar to London Profiler </li></ul><ul><li>Ideal user: general public with an understanding of the classification </li></ul><ul><li>Easy to navigate with help at hand if needed </li></ul><ul><li>Detailed maps of settlements’ morphology and context </li></ul><ul><li>Information about the area </li></ul><ul><li>Possibility to mashup additional data: KML, SHP, CSV </li></ul><ul><li>Download of the data </li></ul><ul><li>User involvement through feedback </li></ul><ul><li>Future inclusion in the RERC website, after testing </li></ul>
  10. 21. Conclusions <ul><li>Improved visualization of the Rural Urban Classification </li></ul><ul><li>Better understanding </li></ul><ul><li>Provide contextual information </li></ul><ul><li>Add external data </li></ul><ul><li>User involvement </li></ul><ul><li>A good test for the 2011 Census data </li></ul><ul><li>Beyond the mashup </li></ul><ul><li>change the bandwidth </li></ul><ul><li>analysis on the fly (R, Python) </li></ul><ul><li>Technical problems to solve </li></ul>