|Page tools: Print Page Print All|
2 These estimates explore concepts and methods while also assessing the quality and limitations of available data sources. The timing and frequency of future Land Accounts will be determined in consultation with stakeholders and the availability of data. A suite of State and Territory Land Accounts would be part of a set of integrated environmental accounts being developed by the ABS (see Towards an Integrated Environmental-Economic Account for Australia (cat. no. 4655.0.55.001)).
3 The Land Account incorporates data from different sources into a consolidated set of information making it possible to link physical data on land to economic data, such as those in Australia's National Accounts.
ENVIRONMENTAL ACCOUNTING FRAMEWORK
4 Environmental-economic accounting is a method of integrating environmental data with economic and, to a lesser degree, social data. The System of Environmental-Economic Accounting (SEEA) is an environmental-economic accounting framework developed by the United Nations Statistics Division. The SEEA Central Framework was adopted as an international statistical standard in early 2012, while a draft of the SEEA Experimental Ecosystem Accounting became available in November 2012. This publication has been produced in accordance with the SEEA and follows previous Australian Bureau of Statistics (ABS) environmental-economic accounting publications utilising this system - Land Account: Great Barrier Reef Region, Experimental Estimates, 2014 (cat. no. 4609.0.55.001), Land Account: Queensland, Experimental Estimates, 2013 (cat. no. 4609.0.55.003), Land Account: Victoria, Experimental Estimates, 2012 (cat. no. 4609.0.55.002), Land Account: Great Barrier Reef Region, Experimental Estimates, 2011 (cat. no. 4609.0.55.001), Information Paper: An Experimental Ecosystem Account for the Great Barrier Reef Region, 2015 (cat. no. 4680.0.55.001), Energy Account, Australia, 2012-13 (cat. no. 4604.0) and Water Account, Australia, 2012-13 (cat. no. 4610.0).
5 For further information on environmental-economic accounting please refer to the ABS Information Papers: Completing the Picture - Environmental Accounting in Practice, May 2012 (cat. no. 4628.0.55.001), Information Paper: Towards the Australian Environmental-Economic Accounts, 2013 (cat. no. 4655.0.55.002), Information Paper: What are Environmental Accounts?, 2008 (cat. no. 4655.0.55.001), Towards an Integrated Environmental-Economic Account for Australia, 2010 (cat. no. 4655.0.55.001), and the United Nations Statistics Division.
COVERAGE AND GEOGRAPHY
6 This release presents an experimental land account for SA and the associated Natural Resource Management (NRM) regions. The NRM regions include: Adelaide and Mount Lofty Ranges, Alinytjara Wilurara, Eyre Peninsula, Kangaroo Island, Northern and Yorke, South Australian Arid Lands, South Australian Murray Darling Basin, and South East. Figure 1 shows a map of the study region.
Figure 1: Study region - South Australia
7 The SA Land Account Tables (1-6) have been made available as Microsoft Excel Spreadsheets (.xls) and within the ABS.Stat Beta environment. The Land Accounting tables aim to follow structures specified within SEEA 2012. However, some columns and rows have been reordered within ABS.Stat to better suit that environment. Users should be aware of these differences when comparing the Excel and ABS.Stat data from the Downloads tab. It is recommended that ABS.Stat tables are viewed in their default layout.
8 ABS.Stat Beta is an interactive free online tool that presents data in a searchable, flexible and dynamic way. ABS.Stat provides users with two options: a web browser interface where you can view, query and download data; and a web services interface described in a machine-processable format using the Statistical Data and Metadata Standard (SDMX) allowing machine-to-machine mechanisms for accessing and sharing ABS data. ABS.Stat enables users to customise the view of data. Users should be cautious when using this functionality as manipulation of data dimensions can lead to misinterpretation of data.
INTERPRETING THE LAND ACCOUNT TABLES
Physical and monetary accounts
9 Tables 1, 2 and 5 in the Downloads tab are physical (hectares) and monetary (dollars) accounts for land use and land cover. These tables present the opening and closing stocks of rateable value or land area for different land uses or land covers. They also show additions and reductions in rateable value or land area and
the reasons for these additions and reductions. The monetary account for land use includes a revaluations category, which presents changes in the rateable value due solely to valuation changes over the accounting period. The accounts in this publication only present total additions and total reductions in rateable value or land area, with the exception of revaluations which are identified in the monetary account. The data used in this account do not provide sufficient contextual information to attribute reasons for change.
Net change matrices
10 Tables 3, 4 and 6 in the Downloads tab are net change matrices for land use and land cover in both physical and monetary units. Like the physical and monetary accounts described above, they present the opening and closing stocks of rateable value or land area for different land uses or land covers. In addition they present the net increase and net decrease of land uses and land covers according to the land use or land cover it was converted from (in the case of increases), or to (in the case of decreases). Finally, they present a total net change for each land use or land cover. The opening stock was calculated by summing all land use or land cover values by type for the beginning of the reference period. The closing stock was calculated by summing the opening stock and the total net change in rateable value or land area for each land use or land cover type. It is important to understand that the matrices show net changes, which may mask information. For example, if 100 hectares of 'Livestock Grazing' is lost in one place but added elsewhere, then no net change would be shown.
Land use by cover
11 The Summary of Findings shows a cross classification of 'Livestock Grazing' land use by cover in physical terms for 2006 and 2011. The table presents the opening and closing stocks of land area for combinations of land use and cover types. In addition the table shows the total change and percentage change of the land use and cover types between the two accounting periods.
12 This publication was produced using information from a number of Commonwealth and State Government Departments. The ABS gratefully acknowledges the assistance provided by these agencies. The data sources are listed below:
13 The two datasets used to produce the statistical tables are Land Valuations from the State of South Australia and Geoscience Australia's DLCD Version 2 (V2) beta product. The SA valuations data represents land use and value as at June 2006 and June 2011, while valuations have been weighted to the corresponding years in Australian System of National Accounts, 2013-14 (cat. no. 5204.0). ABS has used DLCD based on MODIS data from January 2005 to December 2006 and January 2010 to December 2011.
14 The primary aim of this publication was to develop a consistent and repeatable methodology for producing a regular suite of State and Territory Land Accounts. This aim drove the selection of the primary data sets used.
15 Tables 1-6 in the Downloads Tab are experimental, and as such the estimates should be used with caution. In compiling these tables the ABS has identified issues with the coverage, completeness and accuracy of the administrative data sets used. It is expected that these issues will be addressed as the methods and underlying datasets mature.
16 This publication has been produced from administrative information and satellite imagery. The validation of data accuracy has been a major challenge. For example, no other land use change information for the accounting period exists to benchmark the output tables against.
17 A considerable proportion of land reported in Tables 1-4 has changed to or from 'Unallocated' and 'Not Classified' land use. 'Unallocated' represents land uses that could not be allocated to the Australian Valuations Property Classification Code (AVPCC). 'Not Classified' represents areas where no land use information was available. The large movements to and from these classes highlight potential classification inconsistencies and poor data coverage in some areas.
18 The SA valuations cadastre was used as the foundation spatial boundary for the land use tables (Tables 1-4). These cadastral boundaries have undergone small and inconsistent shifts between the accounting periods. These shifts can cause small artificial changes between land use types. This issue was largely overcome by linking data for the two accounting periods using a tabular join (see paragraph 22). However a small percentage of properties are still affected by this issue.
19 The methodology of assigning a dominant land use where multiple cadastral properties overlap (see paragraph 23) has resulted in some artificial changes in land use and rateable value (Tables 1-4). Lower valued land use types are more vulnerable to changes due to the methodology of assigning a dominant land use based on the highest rateable value.
20 Tables 5 and 6 in the Downloads tab present changes in land cover. The supporting DLCD remains in a testing phase. As such, the information should be interpreted cautiously and with reference to the data custodians, Geoscience Australia (GA). The 'Built-up Surface', 'Extraction Sites', 'Irrigated Sugar' and 'Rainfed Sugar' land cover types were derived from supplementary datasets from a single period in time. These classes show no changes over time. The following are some known issues with the DLCD:
Land use tables
21 The South Australian (SA) cadastre was used as the building block for the land use tables (1-4). This provided a spatial index of digital cadastral boundaries. The land valuations data was provided to the ABS linked to the cadastre and was the source for land values and land use classification in this publication. This account has only presented rateable land values which are used for Local Government rating purposes and land tax assessment.
22 The cadastral boundaries for both accounting periods were linked together to enable the presentation of land use changes over time. The issue of cadastral polygon shifts has caused spatial misalignment between the two years of cadastre. To minimise this problem the accounting periods were linked using two separate approaches. Where possible, properties have been linked using a tabular join based on property identifiers. This was limited to properties that had no geometry changes during the accounting period or overlapping polygons in the individual years of cadastre. The tabular linked records for both periods were spatially joined to the 2011 version of valuations cadastre. This captured approximately 85% of valuation records within SA. The remaining 15% were linked together using a spatial overlay.
23 A small proportion of the remaining valuation records linked to properties that contained overlaps in the individual years of cadastre. These overlaps were treated by splitting and merging properties into new objects. This enabled all valuations to be included and prevented double counting of land area. Where multiple valuations fall on the same object their values are summed. These valuations can have different land use classifications reflecting multiple uses at the site. Where this occurred a dominant land use was allocated based on the land use type with greatest rateable value. The dominant land use method was chosen to facilitate reporting change between two time periods. The two accounting periods were then associated using a geographical union.
24 Both the tabular and union datasets were merged together and spatially intersected with the Natural Resource Management (NRM) regions in SA. In some cases this process split properties that crossed NRM boundaries. The rateable value was subsequently apportioned across these properties based on land area.
25 The land use classification used in this publication is shown in Appendix 1. In order to develop consistent Land Accounts for all jurisdictions in Australia, the land use classification applied needs to be consistent for all States and Territories. A decision was made to apply the Australian Valuation Property Classification Codes (AVPCC) to all States and Territories, as it is promoted as a national standard for land valuations. For the South Australian Land Account, a concordance between the SA land use classification and AVPCC has been created by the ABS.
26 'Primary Production' land use is split into AVPCC secondary level for 'Agriculture Cropping', 'Livestock Grazing' and 'Mixed Farming and Grazing', while 'Horticulture Fruit and Vegetable Crops' and 'Horticulture - Special Purpose structural Improvements' were merged into a single class. All remaining 'Primary Production' land uses are presented as 'Other Primary Production'. 'Community Services' and 'Sports, Heritage and Culture' have been merged as 'Community Services, Sports, Heritage and Culture', while 'Extractive Industry' and 'Infrastructure and Utilities' have been merged as 'Extractive Industry and Infrastructure/Utilities'. The ‘National Parks, Conservation Areas, Forest Reserves and Natural Water Reserves’ category includes a variety of SA land uses classes such as wooded areas, water areas, steep or rocky land and reserves. Not all uses in this category include a rateable value.
27 The SA Land Use classification system includes a 'Residential' land use class labelled 'Rural Living' (House with Primary Production). These land uses were allocated under both the 'Residential' and 'Primary Production' classes within the AVPCC classification. 'Rural Living' (House with Primary Production) properties under 20 hectares were classified as 'Residential', while all other categories were converted to their corresponding 'Primary Production' class. Some generic classes could not be placed under the current AVPCC 'Primary Production' classes. These have been added to 'Other Primary Production'.
28 All other categories are presented at the AVPCC primary level. A custom 'Unallocated' class has been added for properties with land use codes that could not be placed in the AVPCC. A 'Vacant Land' and 'Vacant Residential' class has also been added to capture any vacant land classes under the SA Land Use Classification. All other land area not covered by cadastre has been categorised as 'Not Classified'. As this is a custom classification not presented in the valuations data, it has not been valued in the monetary tables.
29 The Australian System of National Accounts (ASNA) publishes total land value for the State of SA for 'Residential', 'Commercial' and 'Rural' land use types. Tables 1 and 3 present data consistent with Australian System of National Accounts, 2013-14 (cat. no. 5204.0) as at June 2006 and June 2011. Appendix 1 shows the concordance between the land use classification used for the Land Account and ASNA published values.
30 The 'Commercial' and 'Rural' land uses in ASNA are sourced from SA valuations data. However there are small discrepancies between the raw land values provided for Land Accounting and ASNA purposes due to differences in extraction dates and method. To ensure rateable values presented in the SA Land Account align with the ASNA, Tables 1 and 3 were weighted to ASNA reported values for 'Commercial' and 'Rural' land. The value of 'Residential' land in ASNA is sourced from a method that uses a combination of data produced by the ABS and the Reserve Bank of Australia (RBA). The ABS and RBA methodology generates combined estimates of land and dwellings owned by households. The ABS value of dwellings owned by households is then subtracted to derive the value of residential land owned by households. 'Residential' and 'Vacant Residential' land in the SA Land Account has also been weighted to the ASNA report values.
31 Australian System of National Accounts, 2013-14 (cat. no. 5204.0) also includes an 'Other' land use, valued at $76.4 billion in South Australia for 2013. This is non-rateable land, including: crown land, National Parks and other land owned by government, as well as land held under native title and other types of arrangements. Tables 1-4 do not include any non-rateable land.
Land cover tables
32 The DLCD V2 beta was provided by GA as a 250 metre x 250 metre grid for Australia. Two reference periods of DLCD V2 beta as at 2005-06 and 2010-11 were used to analyse change over time. Within DLCD, each grid cell is attributed with a land cover type based on dominance for the reference period. The DLCD V2 beta is presented as 33 unique land cover classes. In consultation with GA, these were reclassified to 25 categories. These classifications are detailed in Appendix 2.
33 Using ESRI ArcMap™ software, the DLCD V2 beta for 2005-06 and 2010-11 were combined, generating a unique value for each combination of land cover change. The land area of each unique combination that changed from one land cover class to another was then calculated for each NRM region. This enabled the development of the physical account and net change matrix for land cover.
34 The 2005-06 and 2010-11 DLCD V2 beta have been resampled to align to the National Nested Grid (NNG). The NNG guideline has been developed by the Australian New Zealand Land Information Council (ANZLIC) NNG Work group. The guideline provides consistent standards/specifications to allow information sharing and effective use of grid cell data. The specifications detailed in the guideline have been used to resample each DLCD V2 beta in the reference period. The NNG guideline can be downloaded from the ANZLIC website: http://www.anzlic.gov.au/resources/national_nested_grid.
Land use by land cover
35 The DLCD V2 beta was provided by GA as a 250 metre by 250 metre grid for Australia. Using ESRI ArcMap™, the DLCD V2 beta for 2005-06 and 2010-11 was converted to a feature using the ‘Raster to Polygon’ function. The 25 land cover categories presented in Tables 5 and 6 have been further refined to nine unique classes (see Appendix 2) for this output table. The output of this conversion was then intersected with the valuations cadastre for 2006 and 2011. The output of this tool provides a spatial layer which includes information on land use and land cover for every overlapping polygon from the input features. This process can create areas of Land Cover smaller than the resolution of the original grid. Any value of less than 6.25 hectares in area is less than the resolution of the DLCD and should be used with caution. The output layer was converted to a tabular format and a pivot table was created between land use and cover for 2006 and 2005-06 and 2011 and 2010-11 respectively. The differences between 2006 and 2011 were then calculated and a summary table produced.
36 Users should be aware that discrepancies exist between data collection dates. DLCD provides a land cover reference based on 2 years of Moderate Resolution Imaging Spectroradiometer (MODIS) data for 2005-06 and 2010-11, while land use was derived from SA Valuer-General data extracted as at June 2006 and June 2011.
Land value grids
37 Grids of land value for 2006 and 2011 were built from SA Valuer-General data by apportioning the site value from cadastral properties to an overlaid grid. This approach assumed the distribution of land value within individual properties is homogenous. Where appropriate, data has been suppressed to prevent the release of any information that may identify any individual or organisation. In the land value grids suppressed cells are attributed as 'NoData'.
38 The GIS files are aligned to the National Nested Grid (NNG) standard for Australia in an Australian Albers Equal Area projection.
39 Where appropriate, data has been suppressed to prevent the release of any information that may identify any individual or organisation.
40 The following classifications were used in this publication:
41 Users should also refer to Land Account: Great Barrier Reef Region, Experimental Estimates, 2014 (cat. no. 4609.0.55.001), Land Account: Queensland, Experimental Estimates, 2013 (cat. no. 4609.0.55.003), Land Account: Victoria, Experimental Estimates, 2013 (cat. no. 4609.0.55.002) and Land Account: Great Barrier Reef Region, Experimental Estimates, 2011 (cat. no. 4609.0.55.001). These Land Accounts present the first experimental estimates produced by the ABS for Queensland and Victoria.
42 The Australian Statistical Geography Standard (ASGS) (cat. no. 1270.0.55.001) is the framework for understanding and interpreting the geographical context of statistics published by the ABS. Further information can be found at the ABS website - <www.abs.gov.au>.
43 Other ABS environmental-economic accounting publications which have been published in accordance with the SEEA include: Information Paper: An Experimental Ecosystem Account for the Great Barrier Reef Region, 2015 (cat. no. 4680.0.55.001), Energy Account, Australia, 2012-13 (cat. no. 4604.0) and Water Account, Australia, 2011-12 (cat. no. 4610.0).
44 Land, timber and sub-soil assets are also included on the national balance sheet contained in the Australian System of National Accounts, 2013-14 (cat. no. 5204.0). Additionally, information on the land valuation methodology used to produce the national balance sheet on land can be found in the feature article Australian System of National Accounts, 2005-06 - Valuing land and dwellings owned by households.
45 Other ABS publications related to this topic include:
np not available for publication but included in totals where applicable, unless otherwise indicated
- nil or rounded to zero (incl. null cells)
Appendix 1. Classifications of land use
Appendix 2. Classifications of land cover
These documents will be presented in a new window.