Lineage statements for demand model outputs:
Inland flood mitigation demand is modelled as areas where flooding from rivers may negatively impact people and infrastructure. Three indicators contribute to the demand score:
flood extent, vulnerability of assets, and population size. These are weighted 2:1:1, respectively. A weighted sum of the rescaled indicators is the final demand heatmap output. Flood extent is scored from SEPA Flood Map Version 3.0 flood extent maps. Vulnerability of assets is scored based on the SEPA Flood Risk and Land Use Vulnerability Guidance (SEPA, 2024).
Pollination demand highlights which habitats are within travel range of an area in need of pollination. This is calculated using a cost distance approach (using costDist() from the R package 'terra'). Every habitat type is assigned a cost between 0-50, representing the difficulty for a pollinator to travel 10 m (one pixel; Blake & Baarda, 2018). However, unlike pollination capacity, demand instead assigns a 0 value to areas which require pollination (e.g. orchards and arable), rather than areas which provide pollination. The model then calculates the maximum potential 'catchment' for each area which requires pollination. Pollinators are assigned a total movement cost of 2214 m, and a patch in need of pollination must be at least 0.5 ha.
Access to nature demand is modelled from three indicators: distance to greenspace, population size, and population health. Areas further from greenspaces with a high population and low mean health score will have very high demand for access to nature. Greenspaces are defined in this model as public or restricted green infrastructure (GI; habitat baseline GI values of “Public” or “Restricted”) and non-private undetermined greenspace (GI value of “Undetermined Greenspace” and GIpublic value of N/A). The demand model searches for 2 ha greenspaces within a 300 m radius, which is representative of a 5-minute walk (Handley et al., 2003; Winn et al., 2018).
Air quality improvement demand is determined by four equally weighted indicators:distance to major roads, proportion of man-made surfaces, population, and health. To create the distance to major roads indicator, roads classified as either ‘Motorway’ or ‘A Road’ in OS Open Roads are selected and buffered by 40 m. This selection is then used to filter major roads from the habitat baseline by searching for intersections with road polygons (habitat code J511). A raster is created by calculating the distance of each pixel to a major road, up to 300 m from the road. These scores are then rescaled and transposed to reflect that areas closest to major roads have the highest demand for air quality improvement. The proportion of man-made surfaces indicator is created from the HabClass field of the habitat lookup. Habitats classified as either ‘Urban’ or ‘Infrastructure’ are selected from the baseline and assigned a value of 1. The proportion of land identified as one of these classes within a 300 m radius is calculated from the focal sum. The population indicator is created by summing the household population of each house in a pixel. Houses are identified as habitat code J360 (domestic buildings), and household population is assigned from the housePop field of the habitat baseline. The final population indicators score is the focal sum of the household population within a radius of 300 m. This score is then rescaled with a square root transformation as values from 0-1, where 1 is the highest value of the score nationally. The health deprivation indicator is created by calculating the mean health deprivation score for all houses in a pixel. Like the population indicator, the health deprivation score for each household is assigned from the health field of the habitat baseline. The final score is the focal mean of the health deprivation score, calculated over a radius of 300 m. This score is then rescaled with a min/max transformation as values from 0-1, where 1 is the highest value of the score nationally.The final demand score for air quality improvement is a sum of the four indicator scores, and a mask is applied to remove areas where the population density is less than 5 people per hectare.
Noise regulation demand is scored based on three indicators: distance to noise sources, population, and health. To create the first indicator, major roads, airports, and railways are defined as noise sources. These features are identified from OS VectorMap, with major roads including roads that are classified as ‘Motorway’, ‘A Road’ or ‘Minor Road’ in the dataset. Motorways are buffered by 15 m, airports are buffered by 500 m, and railways, minor roads, and A roads are buffered by 5 m. Within these buffered areas, each pixel is assigned its distance from the noise source. The scores are then converted to log10, and the inverse of the score is calculated to reflect that noise is greatest next to the source. A raster layer is produced for each of the five noise sources, and the five layers are summed to create the distances to noise sources indicator. The population and health indicators are created as described in Air Quality Improvement. The three indicators are summed to create the score map for noise regulation demand. Distance to noise sources is given a weight of 1 to reflect that proximity to noise sources drives demand for noise regulation, and the population and health indicators are each weighted 0.5. To prevent very sparsely populated areas from scoring highly, demand for noise regulation is set to 0 in areas that do not meet a population threshold of 5 people per hectare.
Local temperature regulation demand is determined by three equally weighted indicators: proportion of manmade surfaces, population, and health. To determine the proportion of manmade
surfaces, habitats of the HabClass “Urban” and “Infrastructure” are selected from the basemap and assigned a score of 1; all other habitats are assigned a score of 0. The focal sum is then calculated with a radius of 200 m. The population indicator for this model is weighted by the proportion of the population that are 65 and older. The indicator is created from two raster layers: the sum of the household population per pixel, and the mean risk score per pixel. Like in the air quality improvement and noise regulation models, the population is obtained by summing the household population for each house in a pixel; the household population is stored in the housePop field of the habitat baseline. To obtain the mean risk score, the riskgroup value is assigned from the habitat baseline to each house, and the mean value per pixel is calculated. The population score is the focal sum of the household population, and the health risk score is the focal mean of the risk scores, both calculated at a radius of 200 m. To limit demand in sparsely populated areas, population scores of less than 50 are dropped. The population score is then multiplied by the health risk score to create the final version of the population weighted by risk group indicator. These scores are rescaled from 0-1 using a min/max transformation, where 1 is the highest value nationally. The third indicator, the health deprivation score, is calculated in the same manner as described in Air Quality Improvement. The focal mean is calculated for a radius of 200 m. Each of the three demand indicators are masked to built-up areas greater than 10 km2. The final demand score is the sum of the three indicators.
Lineage statement for Habitat Baseline:
The habitat baseline (hereafter 'baseline') is a habitat map with full coverage of Scotland. It is split into approximately 1120 tiles corresponding to OS grid squares (e.g. NY58). The baseline is built upon the OS MasterMap Topography layer. A classification system is used to derive a single habitat code for each polygon, taking into account agreements & disagreements between different sources of habitat information. Each constituent habitat dataset contributes an attribute to the table, and a new attribute (HabCode_B) contains the overall habitat code for that polygon. Additional habitat and landuse information is added from the following sources:
• CORINE land cover
• Scotland's Land Cover Map (SLAM)
• Habitat Map of Scotland (HabMoS; NVC no-overlaps version only)
• OS MasterMap Greenspace
• OS Open Greenspace
• National Forest Inventory
• Scottish Crop Map
The baseline mostly retains OS MasterMap's polygons. However, in certain upland areas, large OS MasterMap polygons broken up where SLAM offered several distinct habitat patches. The baseline attribute table also contains metrics specific to each polygon representing a domestic building, derived from Scotland's Census 2011 or the Scottish Index of Multiple Deprivation (SIMD) 2020, or both. These are:
• housePop Population estimates at postcode level are taken from Scotland's Census 2011, with results aggregated at Output Area level. SIMD 2020, which is at Data Zone level (broader than Output Area level), is joined to Output Area with a lookup (so several neighbouring Output Areas can have different population estimates but same the SIMD values)
• Health The health index we take from SIMD is HlthCIF, the Comparative Illness Factor (CIF). CIF is a combined count of the total number of people claiming one or more of: Disability Living Allowance, Attendance Allowance, Incapacity Benefit (not receiving DLA), Employment and Support Allowance, Severe Disablement Allowance, Income Support with disability premium, Personal Independence Payment, and Universal Credit claimants with an accepted restricted ability to work.
• Riskgroup Age estimates are taken from Scotland's Census 2011, with ages <10 and >65 considered groups at risk (for example, in the context of heat stress). A proportion of these age groups is calculated at the Data Zone level.
The baseline attribute table also contains metrics derived from Scotland soil carbon & peat depth:
• soil_label an integer 1-3 which joins to a lookup table with descriptions of peat content.
• peat_depth_cm_min minimum peat depth in cm.
• peat_depth_cm_median median peat depth in cm.
• deep_peat30_pct percentage of a polygon covered by peat with thickness >=30cm.
• deep_peat50_pct percentage of a polygon covered by peat with thickness >=50cm.
LINEAGE STATEMENTS FOR PRE-EXISTING DATASETS
• CORINE land cover
Version 2020_20u1 Release date: 24-02-2020 File naming conventions simplified and better described. New file naming convention has been introduced based on user feedback on version 20. Filename is composed of combination of information about update campaign, data theme and reference year and version specification (including release year and release number). The French DOMs are provided in separate databases (files both for vector and raster version of data). All raster layers are back in 8 bit GeoTIFF. Modification is introduced based on the user feedback on version 20. In order to keep 8 bit resolution for raster change layers, they are divided into two files - representing consumption (from) and formation (to) part of change. Version 20 Release date: 14-06-2019 Vector CLC database was provided by National Teams within original CLC1990, I&CLC2000 update, FTSP/CLC2006 update, CLC2012 update and CLC2018 update projects. All features in original vector database were classified and digitised based on satellite images with 100 m positional accuracy (according to CLC specifications) and 25 ha minimum mapping unit into the standardized CLC nomenclature (44 CLC classes). European Corine Land Cover seamless DBs represent the final product of European data integration. The process of data integration started when national deliveries have been accepted and the Database Acceptance Report (DBTA) delivered . Delivered national data were produced in local national systems of all participating countries. Each national Coordinate Reference System (CRS) definition had to be known precisely together with its geometric relationship to a standard system in order to accurately transfer all national data into a standard European coordinate reference - ETRS89/LAEA1052. Mostly, the process itself was carried out by global equation-based transformation to ETRS89 (e.g. seven-parameters Bursa-Wolf methods). The accuracy of a particular transformation ranges from centimetres to meters depending on the method and the quality and number of control points available to define the transformation parameters, but, in any case, the accuracy is far above the actual CLC data resolution (for more details see the DBTA reports for particular country). National data, when transformed into the common European reference, are introduced into tiled pan-European structure and as final step seamless dataset is produced. In order to achieve production of the real seamless European database, the integration step includes also harmonization of database along country borders. It consists from edge-matching of land cover polygons from the national databases across national borders done by a verification / re-interpretation of the satellite images in the border regions (2 km wide strip along borders). The satellite images from IMAGE2000. CLC1990, CHA9000 and CLC2000 database were harmonized this way, but the order to priority was as following: CLC2000, both geometric and thematic adaptations of all polygons in a 2 km strip along national boundary lines; CHA9000 database to ensure that changes in CLC2000 are consistent with the change database; corrected CLC90 (if provided by the MS); corrections were focused to geometric adaptations in semi-automatical way based on CLC00 and CHA00 databases. Border harmonization step has been skipped for CHA0006, CHA0612, CLC2012 and CLC2018 datasets. Simplified border harmonization step for CLC2006 dataset has been created for these countries: CH, NO, KO, TR, IE. A simplified border matching has been applied: - <25 ha polygons are NOT systematically removed (see next bullet). - Sliver-like polygons (area < cca. 5 ha - soft limit) are generalised to largest or thematically most similar neighbour. - CLC-code differences in polygons along two sides of the border are NOT changed Only polygons with area <= 0,1ha were eliminated in CHA0006, CHA0612, CLC2012 and for CLC2018 datasets and CLC2006 dataset (besides the above-mentioned cases) and in parts newly added in campaigns 2006 and 2012 too Note: Some artificial lines (dividing polygons with the same code) can be still present in database due to technical constraints of current ArcGIS technology, but has no impact for dataset contents and can be dissolved for data extracts. Version 18 (V18) Release date: 19-09-2016 (see V18_5_1) Main purpose of the release: Publication of the final, corrected CLC 2012 data. The 4th CLC inventory for the reference year of 2012 was produced under the Copernicus Initial Operations (GIO). It has the shortest production time in history of CLC. Two high-resolution satellite image coverages (IRS Resourcesat-1/2, SPOT-4/5, RapidEye constellation) taken in 2011-2012 provided multi-temporal information to support the update. Computer Assisted Photointerpretation (CAPI) was the prevailing methodology applied in interpreting of satellite images. FI, DE, IC, IE, NO, ES and SE applied a semi-automatic methodology. UK has turned from semiautomatic processing to CAPI because no national hi-res dataset was available for 2012. Most of the QC was conducted in remote verifications. IT and ES were verified by regions. In producing the European products, a simplified border matching was applied (see Version 15). An independent validation of CLC and CLCC for CLC 2012 was carried out in 2016 and the results are available at https://land.copernicus.eu/user-corner/technical-library/clc-2012-validation-report-1. Changes from previous main release (Version 17):
• Inclusion of CLC 2012 layers for all the EEA39 countries.
• Production of CLC 2006 for Greece (in V18_3) and all CLCs for Channel Islands (V18_1).
• Revised CLC 2000 and CLC 2006 layers were made available (V18_5).
• Change in rasterization algorithm (V18_2). Known problems:
• Some redundant lines between neighbouring polygons with the same code are still present, but only as result of persisting ‘adaptive tilling’ procedure (limitation of ESRI ArcGIS technology for large datasets).
• Polygons <25 ha can be present along national borders and along 'adaptive tilling' tiles boundaries. See https://land.copernicus.eu/user-corner/technical-library/clc-country-coverage-v18.5 for full information about the coverage of this version. See https://land.copernicus.eu/user-corner/technical-library/clc-and-clcc-release-lineage for full information about all sub-versions of this version.
Version 17 (V17) Release date: 02-12-2013 Main purpose of the release: Maintenance / Increased European coverage of CLC time series data. Changes from previous release (V16):
• Full CLC and CLCC data time series (from CLC 1990 to CLC 2006 including all CLCC datasets) has been included for the Autonomous Region of the Azores (PT). Version 16 (V16) Release date: 15-04-2012 Main purpose: Maintenance / Increased and improved European coverage of CLC time series data. Changes from previous release (V15):
• CLC 1990 coverage: TR has been delivered CLC 1990 and CLCC (1990, 2000) data. Still missing CLC 1990 data: AL, BA, CH, CY, FI, IS, MK, NO, SE, UK and the XK.
• CLC 2000_revised layer covering 27 countries was included (CLC 2000 data revised during production of CLC 2006).
• Shift in MT geographic position has been corrected. All CLC layers for MT have been re-projected.
• A few coding inconsistences were corrected.
Version 15 (V5) Release date: 20-07-2011 Main purpose: Publication of final CLC2 006 data. The 3rd CLC inventory for the reference year of 2006 was produced under GMES Fast Track Service on Land Monitoring. The CLCC database was considered as the primary product, and a uniform change mapping methodology was agreed. Dual date satellite imagery (SPOT-4/5 and IRS P6) taken in 2005-2007 provided enhanced change mapping capabilities. Some of the countries newly entering CLC have produced CLC 2000 datasets also during the project time frame. Scanned topographic maps and digital aerial ortho-imagery have become commonly available. CAPI was the prevailing method applied in interpreting of satellite images. Nevertheless, FI, IS, NO, SE and the UK applied a semiautomatic methodology. Most of the European QC was conducted by visiting national teams (see Version 2). In some cases, remote verification was applied (without mission to countries). ES and IT were verified by regions. Changes from previous release (V14 (V4)):
• CLC 2006 data covering Great Britain (part of UK) and TR were delivered. Thus, CLC 2006 European coverage includes 38 countries of the EEA39. Still missing CLC 2006 data for Greece.
• A simplified border matching was applied for countries new in CLC: XK, NO, CH and Türkiye: 1) <25 ha polygons along the borders are not removed systematically; 2) sliver-like polygons (area < cca. 5 ha) are generalised to largest or thematically most similar neighbour.
• For the rest of CLC 2006 countries a simple border-matching was applied. Code differences along two sides of borders are not changed. Only polygons with area ≤ 0,1 ha (sliver polygons) are eliminated.
• Data dissemination: CLC data become freely accessible from the EEA to any person or legal entity.
Version 14 (V4) Release date: 25-10-2010 Main purpose: Maintenance / Increased European coverage of CLC 2006 and CLC 2000 data. Changes from previous release (V13 (V3)):
• CLC 2006 European coverage includes 37 full countries of EEA39. New data for Northern Ireland (part of the UK), Madeira Islands (part of PT), CH, IS and TR were added to CLC 2006 data. Still missing CLC 2006: GR and the UK (except Northern Ireland).
• New data for Madeira Islands (PT), CH and IS were added into the European CLC 2000 coverage, which includes already the EE39. However, CLCC (1990, 2000) is available for 28 countries only.
• New data for Madeira Islands (PT) were added into CLC 1990 and CLCC (1990, 2000). Still missing CLC 1990 data: AL, BA, CH, CY, FI, IS, MK, NO, SE, TR, UK and XK. The seamless European database has been further improved addressing feedback from the EEA on V13 (V3):
• No-data buffer (code 999) outside of valid data area was deleted.
• Small gaps identified in V13 were corrected by tolerance adaptation in ArcGIS v10 geodatabase.
• Remaining neighbour polygons with the same code were resolved by additional dissolve operation.
Version 13 (V3) Release date: 02/2010 Main purpose: Publication of initial European coverage of CLC 2006 data. Changes from previous release (V2):
• Version numbering was changed to harmonise vector data (V3) and derived raster data (V13) releases.
• First seamless release in ESRI Geodatabase format.
• Initial coverage of CLC 2006 including 35 countries and Northern Ireland (part of the UK). Missing countries in CLC 2006: GR, CH, TR and the UK (except Northern Ireland).
• Two updates added to CLC 2000: a new version for NO and the first CLC dataset for TR.
• Sea buffer around land has been introduced (15 km as proxy to 12 nautical miles’ sea zone).
Version 2 (V2) Release date: 09/2009 Main purpose: Publication of final CLC 2000 coverages. The 2nd CLC inventory for the reference year of 2000 (CLC 2000) was carried out in the frames of CLC 2000 project. A single date Landsat-7 ETM satellite imagery taken in 1999-2001 was provided by JRC. The technology of drawing the interpretation on transparencies was discarded and replaced by CAPI (computer-assisted photo-interpretation). Prior to mapping changes CLC 1990 data had to be corrected: 1) bulk geometric mistakes removed and residual geometric errors >100 m and coding mistakes were corrected; 2) polygons smaller than the 25 ha MMU were generalised. European QC was conducted by visiting national teams (usually at the start and towards the end of the project). Computer-assisted verification has provided written, geo-located explanations regarding the mistakes and supported harmonized production of the database all over Europe.
Changes from previous release (V1):
• It was to deliver a single seamless layer, but was not feasible in ESRI environment. Therefore, seamless ESRI ArcInfo Librarian map tiles were produced again (but free of tiling artefacts reported in V1).
• New country deliveries integrated into European CLC 2000 ME, RS (incl. XK), IS and NO. Simple harmonization along national borders of these countries was done (small artefacts cleaned only).
• CLC 2000 data for MT have been updated to reflect changed geometry in CLC 2006 delivery.
• The dissemination and use of products was defined in an agreement between the EEA, the EC and the participating countries.
Version 1 (V1) Release date: 08/2005 Main purpose: Publication of initial European coverage of CLC 2000 and CLCC (1990, 2000) data. Changes from previous release (V0):
• The first consolidated version of European CLC data have been produced as integrated and harmonised seamless layer in ESRI ArcInfo Workstation Librarian map tiles.
• The production of the first CLCC database has started, but no consolidated methodology was available.
• Initial CLC 2000 coverage included 32 countries: AL, AT, BE, BA, BG, CY, CZ, DE, DK, EE, ES, FI, FR, GR, HR, HU, IE, IT, LV, LI, LT, LU, MK, MT, NL, PL, PT, RO, SI, SK, SE and the UK. Missing countries in CLC 2000: CH, IS, ME, NO, RS (including XK) and TR.
• CLC 1990 for most of the countries has been replaced by revised CLC 1990. Some additional countries have produced CLC 1990. Still missing in CLC 1990 European coverage: CY, LI, MT, SE and UK.
• Full harmonization (visual re-interpretation by keeping the 25 ha MMU) inside a 5-km wide strip along national borders was done including 32 countries for CLC 2000 and 24 countries for CLCC (1990, 2000).
• Semi-automatic harmonisation of 2-km wide strip along national borders was done for CLC 1990.
• Vector to raster conversion: “cell centre” method was applied.
• The 25 ha MMU is considered as hard limit. Polygons <25 ha were generalised.
• Dual ownership of CLC and CLCC data (EEA and the country) was introduced.
Version 0 (V0) Release dates: up to 12/2000 Main purpose: Distribution of country-level CLC 1990 data and creation of European raster products. The period of the first CLC inventory was rather long (1985-1996) and 1990 is considered as reference year. CLC 1990 data delivered by countries became part of GISCO database. Releases were provided bi-annually. Following political changes in Central and Eastern Europe 10 additional countries joined. The methodology was visual photointerpretation by drawing the CLC map on transparency, placed on top of satellite image hardcopy at scale 1:100.000.
• CLC 1990 vector and raster data were initially available for 12 countries: AT, BE, DE, DK, ES, FR, GR, IE, IT, LU, NL and PT. Raster only data were available for FI and UK.
• The EC Phare programme supported the implementation of CLC 1990 in 11 countries of Central and Eastern Europe between 1992 and 1998: BG, CZ and SK, EE, LV, LT, HU, PL, RO and SI.
• Integrated European vector dataset was available as ESRI ArcInfo Librarian and derived raster products as ESRI grids in 100m and 250m resolution.
• Data dissemination policy was unclear.
Scotland's Land Cover Map (SLAM)
Training data were collected through using a combination of the following sources:
• Habitat Map of Scotland (ground polygons)
• 2022 National Forest Inventory
• Ordnance Survey
• High resolution imagery In all cases the ground data were not used naively: a careful combination of at least two data sources were used to create each polygon, and checking against recent high resolution imagery to ensure each polygon was ‘pure’ (i.e. included only one class) and up to date (for example, if it was a forest polygon, the trees had not been cleared since the data were collected). Satellite remote sensing datasets used for mapping were Optical Sentinel 2 (S2), Synthetic Aperture Radar (SAR) Sentinel-1, descending and ascending ALOS-PALSAR 2. A complex set of machine learning algorithms were used to produce a Prediction Model, and ultimately a prediction of a class for each pixel. Through the project duration the sophistication of the models used increased, increasing accuracy and efficiency. For commercial reasons the details of the final algorithms used will not be revealed here. The CHANGE dataset is a change map, which shows the land cover change that is predicted to have occurred between 2020 and 2022. The map is produced through a simple comparison between the 2020 and 2022 maps, where each instance of change identified is interpreted and assigned one of the following descriptors: (i) Afforestation (ii) Tree removal (iii) Agriculture related (iv) Urban development (v) Forest growth (vi) Water gain (vii) Water loss (viii) Other changes Please note, we believe these predicted changes, and others, are inaccurate, mainly due to inaccuracies we have identified in the 2020 map, along with improved methodologies and processes developed at Space Intelligence since the creation of the 2020 map.
Habitat Map of Scotland (HabMoS; NVC no-overlaps version only)
This dataset provides a non-overlapping polygon layer of the NVC survey coverage. This is derived from the NVCCONV_EUNIS_COVERAGE_MV dataset in geo.Store. It covers the NVC habitat surveys currently in the Habitat Map of Scotland as of October 2023. This dataset does not include the Saltmarsh survey 2013, Sand Dune Vegetation survey 2012 or Coastal Vegetated Shingle survey. Processing survey coverage overlaps: 1. Data was extracted from NVCCONV_EUNIS_COVERAGE_MV dataset in geo.Store then a process was done to identify which surveys overlapped with others. 2. Groups of overlapping surveys were identified, work was done one group at a time. 3. Prioritising overlapping surveys was assessed with the help of an habitat expert. Factors that influenced decision were: age of the survey, quality of the survey, quality of surveyors, expert knowledge, existing survey metadata. 4. Parts of lower priority surveys were removed where they were overlapped by a higher priority survey. 5. Final coverages without overlaps were used to clip their corresponding habitat polygons. Notes and caveats: • Processing survey coverages created slivers. • Some of the original NVC polygons had an empty description field. This is the reason why sometimes we have the coverage of a survey not covered completely with habitat polygons.
Lineage statement for the entire HabMoS (from which the NVC no-overlaps version was derived):
These data are a materialised view created in the database drawing data indirectly from the following tables.
• GIS_SNH_OWNER.NVC_POLYGONS
• GIS_SNH_OWNER.NVCCONV_EC_COMMUNITY
• GIS_SNH_OWNER.HABITAT_SURVEY_METADATA
• GIS_SNH_OWNER.SALTMARSH_SURVEY_2013
• GIS_SNH_OWNER.HABMOS_SMSS_EUNIS_CONVERSION
• GIS_SNH_OWNER.HABMOS_SMSS_EUNIS_EXCEPTION
• GIS_CMEU_OWNER.SDVSS_2012
• GIS_SNH_OWNER.SDVSS_EC_COMMUNITY
• GIS_SNH_OWNER.HABMOS_NWSS_H9180_SNH
• GIS_SNH_OWNER.HABMOS_NWSS_H91A0_SNH
• GIS_SNH_OWNER.HABMOS_NWSS_H91AO
• GIS_SNH_OWNER.HABMOS_NWSS_H91C0
• GIS_SNH_OWNER.HABMOS_NWSS_H9E0_SNH
• GIS_SNH_OWNER.HABMOS_NFI_EUNIS_MINUS_NWSS
• GIS_EXTERNAL_OWNER.RCAHMS_HLA
• GIS_SNH_OWNER.HAMOS_EUNIS_I1_V2
• GIS_OS_OWNER.OS_VMD_BUILDINGS
• GIS_SNH_OWNER.HABMOS_ANNEXI_TB
• GIS_SNH_OWNER.HABMOS_ANNEXI_EUNIS
• GIS_SNH_OWNER.HABMOS_EUNIS_MASTER
The habmos AnnexI table is derived from the JNCC list of habitat types on Annex I of the EU Habitats Directive. http://jncc.defra.gov.uk/Publications/JNCC312/UK_habitat_list.asp - 19/06/2015 The mapping from NVC to Eunis classification is done on a site by site basis from the saltmarsh NVC to Eunis conversion tables edited by Susan Watt. Objective reference A1604626. Derived 07/07/2015 The Sand Dune Survey to Annex I mapping is derived from NVC conversions, processed by Carmen Mayo and extracted 2016-04-12 Updated 2017-08-29 The NVC Conversion data is derived from NVC conversions, processed by Carmen Mayo and extracted 2016-04-12 Updated 2017-08-29 The NWSS EUNIS conversions are currently at an annex I level extracted on the basis of NVC type by Forestry Commission for Scotland and in the case of H9180, H91A0 and H91E0 further processed by SNH analysts. Certain types have bee extracted from Historic Environment Scotland's Historic Land use Assessment and converted to EUNIS 22/09/2017 Type = Airfield J4.4 Type = Cemetery J2 Type = Maritime Installation J4.5 Type = Mining area J3.2 Type = Opencast site J3.2 Type = Quarry J3.2 Type = Railway J4.3 Type = Golf Course X40 Data from Scottish Governments IACS dataset have been processed so that all the large land parcels in upland areas have been deleted using a selection method based on an upland area mask and field size. Land parcels that do not have a SAF record and any land parcels that are woodland have also been removed to allow the use of more detailed habitat information for woodlands from the Forestry Commission Native Woodland Survey of Scotland and National Forest Inventory. The resulting data has been merged to remove internal divisions and classed as EUNIS I.1 Arable land and market gardens. 28/04/2017 Buildings data from Ordnance Survey Vector Map Districts buildings layer have also been incorporated and classed as Eunis J1. 31/05/2017 EUNIS code descriptions are obtained from the habmos Eunis tables which are derived from the correspondence tables produced by Iain Strachan Objective reference A2104366 derived 22/09/2017 The materialised view is scheduled to re-generate each morning meaning that there can be 1 days latency between data being loaded and it being reflected in these datasets.
• OS MasterMap Greenspace
Captured and maintained primarily by photogrammetric means with ground completion on a cyclic or continuous revision process.
• OS Open Greenspace
OS Open Greenspace is a generalised product which has been automatically simplified from Ordnance Survey large scale data.
• National Forest Inventory
The location and extent of all forests and woodlands (0.5 hectares and over) is stored and maintained as a digital map. The changes in the canopy cover are identified from: • Sentinel 2 imagery taken during spring/summer 2022 or colour aerial orthophotographic imagery available at the time of the assessment; • New planting information for the financial year 2021/2022, from grant schemes and the sub-compartment database covering the estate of Forestry England, Forestry and Land Scotland and Natural Resources Wales; Ordnance Survey MasterMap® (OSMM) features have been used as a reference for capturing the woodland boundaries.
• OS MasterMap Topography layer
Captured and maintained primarily by photogrammetric means with ground completion, on a cyclic or continuous revision process.
• SIMD
The methodology used to construct the SIMD is based on the approach developed by Oxford University for the Scottish Indices of Deprivation in 2003. Since that publication, the Scottish Government have published indices of multiple deprivation in 2004, 2006, 2009, 2012, 2016 and 2020. Full details of the methodology used to construct the SIMD 2020 are available in the technical notes (https://www.gov.scot/collections/scottish-index-of-multiple-deprivation-2020/).
• Scotland soil carbon & peat depth
Uses Landsat satellite imagery, soil, topography, climate and land cover maps for Scotland. Please refer to the publication "Aitkenhead, M.J., Coull, M.C., 2019. Mapping soil profile depth, bulk density and carbon stock in Scotland using remote sensing and spatial covariates. European Journal of Soil Science. 10.1111/ejss.12916" for more details.