Annex A – Data Quality and Processing
A.1. Data Quality
Local authorities may not update all historic site information every survey, consequently some site information may not be current. For example, a site may be sold and change ownership type. A change like this could happen without a local authority’s knowledge.
A.2 Local Authority response rate
The overall data quality for a particular year will depend on how many Local Authorities have updated their survey for that year. If a Local Authority does not update their data then data for the previous year is rolled forward. This will provide a reasonable estimate but will not take account of any changes in the latest year. Data quality will also be affected by the number of years since the last update. Highland has not updated their survey since 2015 so their data is carried over from their latest update in 2015.
Annex Table 5 records local authority annual participation in the survey each year since 1996.
A.3 Removal and Addition of Sites
The means by which new sites are detected are largely centred on the expert knowledge of local authority planning officers supported by the SVDLS guidelines which clearly state the definitions for the recording of derelict and urban vacant land. The survey relies on the consistent application of definitions between local authorities and within a local authority through subsequent years. The guidelines are regularly reviewed and modified to assist in the provision of uniform returns. Some of the changes reported in Annex C - sites removed for definitional reasons is the result of a reappraisal of the survey guidelines by local authorities prior to the 2018 survey. Since the last survey, there were a total of 29 sites (36 hectares) removed for definitional reasons. SVDLS guidance documents can be found at: http://www.scotland.gov.uk/Topics/Statistics/Browse/Planning/DataSupplierArea.
A.4 Changes to Site Details
Changes in site details can cause difficulties in interpreting the data. The most notable are when two separate sites become joined by the inclusion of a further piece of vacant or derelict land, and alternatively when a single site has a central area brought back into use, such that the remaining disused area does not form a single contiguous site. The first of these scenarios may case difficulties if data regarding previous use etc. for the sites differ. Where this happens, details from the largest contributing site are used to describe the new, compound site. In addition, the introduction of GIS systems has also allowed planning officers to improve data quality, leading to some changes in site sizes recorded.
A.5 Changes in Historic Data, 2012-2018
As part of the 2018 survey, work was undertaken to improve the quality of the data supplied in the previous 2012 to 2018 returns. This was done to allow for improved time series analysis. This included removing sites identified by the local authority as being previously incorrectly included in the survey, adding in sites that the local authorities indicated should have been included in previous surveys and correcting historic sizes as a result of more accurate boundary mapping. A number of site sizes were changed during this years’ survey as a result of more accurate mapping of the sites by local authorities through increased use of GIS tools. The revised data for 2012 to 2017 is used throughout the tables and time series analysis of this bulletin.
The SVDLS has been running continuously since 1993. Historical data on the amount of derelict and urban vacant land in Scotland, for the years 1996 to 2011, is shown in Annex Table 1. 1996 to 2004 data has remained unaltered since the 2004 survey publication. 2005 data was last revised in 2011 and each subsequent publication revised historical data for the six years preceding the publication with the 2018 publication updating historical data from 2012-2017. Thus care should be taken when comparing 1996-2011 data in Annex Table 1 to the data shown in this bulletin.
Annex Table 1 - Derelict and Urban Vacant Land by local authority, 1996-2011[1,2,3]
|Local Authority||Total Derelict and Urban Vacant Land Area (ha)|
|Argyll and Bute||64||55||46||46||45||45||46||46||46||28||81||60||40||34||30||29|
|Dumfries and Galloway||365||371||372||364||361||364||360||358||358||225||474||465||456||466||461||491|
|City of Edinburgh||154||145||143||122||162||167||167||123||117||199||194||204||222||225||217||223|
|Loch Lomond and The Trossachs National Park4||-||-||-||-||-||-||-||-||-||-||-||-||-||-||-||35|
|Na h-Eileanan Siar||16||13||13||11||11||11||11||11||12||11||11||11||11||11||10||10|
|Perth and Kinross||94||98||132||149||139||106||111||109||114||96||88||78||40||51||49||50|
1. Figures may not sum due to rounding.
2. See Annex D for details of council participation in different years.
3. Care should be taken when comparing this data to that published for the years 2012-2018 in the main part of this bulletin. The data above has not been amended to take account of all land removed for definitional reasons or other previous reporting errors.
4. From 2011 Loch Lomond and the Trossachs National Park took responsibility for surveying vacant and derelict land within the park boundaries. These sites are no longer recorded in Argyll and Bute, Perth and Kinross, Stirling and West Dunbartonshire local authority boundaries, and are separately identifiable as Loch Lomond and the Trossachs National Park from 2011. Prior to 2011, these sites were classified within the relevant local authority boundary.
A.6 Land area eligible
Local authorities are asked to note all derelict land for the survey but are asked to limit the recording of vacant land to land located within any settlement as defined by the local authority in the latest council approved local plan. Consequently, the actual land area surveyed for vacant land is often considerably smaller than the total land area of each local authority, especially in more rural areas.
A.7 Coverage of Settlements
Local authorities were asked to survey for vacant land within all settlements defined in their latest council approved local plans. However, due to resource constraints some local authorities (particularly those covering large rural areas), were unable to do a full survey of every settlement for vacant land. All returned surveys covered every settlement within the planning authority boundary with a population of 2,000 or more. To ensure consistency, the statistics presented on vacant land throughout this bulletin refer only to land located within settlements that have a population of at least 2,000 (according to the local plan). Some local authorities also undertook either a full or partial survey for vacant land in settlements of under 2,000 in population - these are reported separately from the rest of the bulletin in AnnexTable 2.
It is estimated that approximately 16,872 hectares of land within settlements of under 2,000 in population were surveyed for vacant land in Scotland during 2018. From this area, local authorities reported a total of 319 hectares of vacant land across 92 sites. Aberdeenshire had the highest amount of reported vacant or derelict land within settlements of under 2,000 in population, with 176 hectares across 8 sites – this includes a large site at the former Edzell air base (170 hectares).
Of the 90 sites (331 hectares) of vacant land in settlements of under 2,000 in population reported in the previous survey (2017), 8 sites (10 hectares) were either fully or partially brought back into productive use.
Annex Table 2 - Vacant Land within settlements of under 2,000 in population, 20181
|Local Authority||Area (ha)||No. of Sites|
|City of Edinburgh||11||6|
|Loch Lomond and the Trossachs National Park||1||1|
|Perth and Kinross||2||3|
1. Figures may not sum due to rounding.
2. Highland has not updated their survey since 2015 so their 2018 figures have been carried over from 2015.
A.8 Estimate of the proportion of local authority population living within a defined distance of a Derelict Site
Tables 9 and 10 in the publication show estimates by local authority of the percentage of their population living within various distances of derelict land. For Table 9 a circular buffer zone based on the area of each derelict site was drawn around the grid co-ordinate points supplied for that site. This gives an estimated boundary for each site therefore data should be treated with care as actual site boundaries are not being used. Table 10 used the same estimated site boundaries and also estimated the proportion of local authority population living in close proximity to land that has been derelict prior to 1991 or 2001. In addition to derelict sites which were identified as being derelict prior to 1991 or 2001 the analysis also included sites where the length of time derelict is unknown but where the first site inspection occurred prior to 1991 or 2001.
To measure the proportion of each local authority population that lives within a certain distance of derelict land, a national data set was constructed that estimated the population of each property identified as likely to be residential in the latest National Records of Scotland (NRS) Address Register. Average household size figures were calculated by dividing NRS census (2011) population by census household counts at the Census Output Area (COA) level. All records in the Address Register that were identified as likely to be residential were assigned an estimated household size figure based on the average household size of the COA they fall within.
To ensure consistency with previously published population estimates, the latest available small-area estimates of population by NRS were used as a control factor on the calculated household size figures. For each property in the Address Register based dataset, the distance to the nearest estimated derelict site boundary was calculated, to highlight those properties within the defined distance of derelict land. Those properties’ estimated populations were then aggregated up by data zone to give a proportion of each data zone’s population (and hence each local authority’s population) estimated to live within the defined distance of derelict land.
Overall it is estimated that 29.1% of the population of Scotland live within 500 metres of a derelict site in 2018. The data published in Tables 9 and 10 on proximity will also be available at data zone level on the Scottish Government Statistics website at http://statistics.gov.scot/. The website will be updated with these results following the publication of this bulletin.
A.9 Deprivation Areas using Scottish Index of Multiple Deprivation (SIMD 2016)
Table 11 shows the amount of derelict and urban vacant land located within Scotland’s 15% most deprived data zones. Those data zones are identified using the 2016 Scottish Index of Multiple Deprivation (SIMD). The SIMD 2016 identifies small area concentrations of multiple deprivation across all of Scotland. It contains 38 indicators in seven domains: Current Income, Employment, Health, Education, Skills and Training, Geographic Access to Services, Housing and Crime. Sites in the SVDLS are identified as being located within Scotland’s most deprived areas if their grid co-ordinate point (i.e. the centre of the site) falls within a 15% most deprived data zone. Care should be taken with this data as there are instances of sites crossing data zone boundaries, hence part of the site may be in a 15% most deprived data zone, whilst the other part could be outwith. Further information on the SIMD 2016 project can be found on the Scottish Government’s web page at: http://www.scotland.gov.uk/Topics/Statistics/SIMD/Overview.
A.10 Land surveyed by Highland Council
Due to the large area of land covered by Highland council, a partial survey was carried out in each year between 2002 and 2005 (inclusive). Only derelict and urban vacant sites within settlements containing a population of 2,000 or over and derelict sites within the Inner Moray Firth area were surveyed. In 2006, the (then) Scottish Executive let a contract to consultants to survey derelict sites in outlying parts of Highland council area. This was the first time since 1993 that derelict sites were surveyed in the outlying areas of Highland council. Overall it meant that a complete survey was carried out in Highland during 2006 (in terms of what was presented in the main part of the 2006 bulletin on derelict and urban vacant land). Highland council surveyed all settlements of over 2,000 in population for derelict and urban vacant land and the Inner Moray Firth area for derelict land, the consultants surveyed the remaining outer rural areas for derelict land. In 2007 the previous practice was resumed with Highland council surveying for land within settlements containing a population of 2,000 or over and derelict sites within the Inner Moray Firth area. The 2006 results of the consultants’ work in most of the outer rural parts of Highland council were brought forward as the best possible estimate up to 2012. In 2013 Highland council submitted a survey based on visits made in preparation for the Inner Moray Firth Local Development Plan along with visits and knowledge from both Planning and Development and Housing and Property Services staff. Highland Council last updated their survey in 2015.
A.11 Land surveyed by Loch Lomond and the Trossachs National Park Authority From 2011 Loch Lomond and The Trossachs National Park took responsibility for surveying vacant and derelict land within the park boundaries. These sites are no longer recorded in Argyll and Bute, Perth and Kinross, Stirling and West Dunbartonshire local authority boundaries, and are separately identifiable as Loch Lomond and The Trossachs National Park from 2011. Prior to 2011 these sites were classified within the relevant local authority boundary.