Scottish Settlements Urban and Rural Areas in Scotland

Annex C - The Methodology for 2000

Future work

18. Tasks include:

• discuss with users any comments they may have. This will be initially by correspondence, email, etc. A meeting (or meetings) can be arranged if considered necessary;

• possible re-specification of the contents of the products described at paragraph 16;

• review the methodology in order to streamline the process, improve consistency of quantities used to define thresholds (addresses, residential addresses, population), etc;

• automate, as far as possible, the identification and filling in of rural 'holes' within candidate settlements. The term 'hole' could be taken to include isolated rural incursions into urban areas ('fjords') that might be closed off. This study could also look at combining two or more areas identified as settlements that are constituent parts of the same urban area. The parts are separately identified because they are separated by strips of low density. The resulting absorption of low density areas into a settlement may allow the density thresholds for some council areas to be increased with final settlements still showing some continuity with 1991 (if desirable);

• investigate ways of ensuring that depiction of boundaries meets requirement of forming basis for settlements (paragraph 3);

• consider possible refinements of method so as to use different density thresholds within a council area, for example, by defining densities by urban and rural or crofting, etc, for each council area. This may be done by examining how the size of land parcels associated with dwellings varies throughout the country. Conversely, we could try using a common threshold for the whole of every council area, i.e. use the 2.1 threshold and then fill in any 'holes'. We will have to do some analysis to see to what extent low density postcodes within or near groups of high density postcodes are swept up by a routine process to fill 'holes', close fjords and cross strips;

• investigate possible better ways of estimating population for each settlement, perhaps so as to be consistent with any other methods under development for estimating populations for small areas within council areas. Should population figures for each postcode be brought into the methodology at an earlier point e.g. as a substitute for the number of residential addresses? The latter is more reliable but is, in essence, a proxy for the former. Investigate the relationship of ratio between population and numbers of addresses. This may entail examining the relationship between each of the links in the following chain: address, residential address, residential address in use, residential address in use as main residence, number of residents;

• examine why some 17 1991 localities failed to materialise as settlements even though they each had an estimate population of at least 500 (refer to paragraph 18 of main paper). Each had at least one below threshold postcode that prevented a grouping of urban postcodes with sufficient addresses to become a settlement. This examination may be connected to that on filling 'holes', etc, above;

• split settlements into localities, so that OAs nest into localities instead of settlements; and

• considering detailed comments on particular settlements.