Core Metadata (ANZLIC Version 2) for the Land Use of Australia 2010-11 Dataset Title Land Use of Australia 2010-11 Custodian Australian Bureau of Agricultural and Resource Economics and Sciences (ABARES) Jurisdiction Australia Description Abstract The Land use of Australia 2010-11 is the latest in a series of digital national land use maps at national scale. Agricultural land uses and their spatial distributions are based on the Australian Bureau of Statistics' 2010-11 agricultural census data. The spatial distribution of the agricultural land uses is modelled and was determined using Advanced Very High Resolution Radiometer (AVHRR) satellite imagery with training data to make agricultural land use allocations. The non-agricultural land uses are drawn from existing digital maps covering seven themes: topographic features, catchment scale land use, protected areas, World Heritage Areas, tenure, forest type and vegetation condition. The Land use of Australia 2010-11 is supplied as a set of raster datasets (in Esri grid format) with geographical coordinates referred to the Geocentric Datum of Australia 1994 (GDA94) with a 0.01 degree pixel size. These comprise a set of floating point grids with pixel values between 0 and 1 and an integer grid. The floating point grids are continuous probability surfaces that describe the spatial distribution of each of the agricultural commodity groups mapped. The integer grid is a categorical summary land use map, which has a value attribute table (VAT) with columns defining input layers and an output layer. The output layer specifies land use in terms of the Australian Land Use and Management Classification Version 7 and is an approximation to a maximum likelihood map. The Land Use of Australia datasets are recognized as Foundation Spatial Data by the Australia New Zealand Land Information Council and as an Essential Statistical Asset for Australia by the Australian Bureau of Statistics. Common applications of the datasets are in strategic planning and continental modelling. Refer to the user guide (Smart 2016) for more information. Search Words AGRICULTURE AGRICULTURE Crops AGRICULTURE Livestock AGRICULTURE Horticulture AGRICULTURE Irrigation BOUNDARIES BOUNDARIES Administrative BOUNDARIES Biophysical BOUNDARIES Cultural FLORA FLORA Exotic FLORA Native FORESTS FORESTS Agriforestry FORESTS Natural FORESTS Plantation HERITAGE HERITAGE World HUMAN ENVIRONMENT LAND LAND Conservation LAND Conservation Reserve LAND Cover LAND Topography LAND Use VEGETATION VEGETATION Structural WATER WATER Lakes WATER Surface WATER Wetlands North Bounding Latitude -9.995 South Bounding Latitude -44.005 East Bounding Longitude 154.005 West Bounding Longitude 112.505 Data Currency Beginning Date 2010-04-01 Ending Date 2011-03-31 Dataset Status Progress Complete Maintenance and Update Frequency As required Access Stored Data Format DIGITAL Esri ArcGIS Desktop 10.2 and ArcInfo Workstation 10.0 under Windows Available Format Type DIGITAL - Esri Grid Access Constraint Licence type: Copyright Copyright Commonwealth of Australia 2016 All material in this publication is licensed under a Creative Commons Attribution 3.0 Australia Licence, save for content supplied by third parties, logos and the Commonwealth Coat of Arms. Creative Commons Attribution 3.0 Australia Licence is a standard form licence agreement that allows you to copy, distribute, transmit and adapt this publication provided you attribute the work. A summary of the licence terms is available from creativecommons.org/licenses/by/3.0/au/deed.en. The full licence terms are available from creativecommons.org/licenses/by/3.0/au/legalcode. This publication (and any material sourced from it) should be attributed as: ABARES 2016, Land Use of Australia 2010-11, ABARES, Canberra, available at http://www.agriculture.gov.au/abares/publications. CC BY 3.0. Data quality Lineage I. Seven thematic layers were constructed as 0.01 degree rasters and overlain to determine the non-agricultural land uses and the distribution of potentially agricultural land. The themes and their main sources were: . Topographic features - 'GEODATA TOPO 250K series 3' (GA 2006) . Catchment scale land use - 'Catchment scale land use mapping for Australia - update January 2014' (ABARES 2014a) . Protected areas - 'Collaborative Australian protected areas database - CAPAD 2010' (DSEWPaC 2012b) and 'Indigenous protected areas (IPA) - declared' (DE 2014) . World heritage areas - 'Australia, world heritage areas' (DSEWPaC 2012a) . Tenure - 'Tenure of Australia's forests (2013)' (ABARES 2014b), 'GEODATA TOPO 250K series 3' (GA 2006), 'Catchment scale land use mapping for Australia - update January 2014' (ABARES 2014a) and 'Western Australia Aboriginal Lands Trust Estate as at May 2010' (DAA 2011) . Forest cover - 'Forests of Australia (2013)' (ABARES 2014b) . Vegetation condition - 'Catchment scale land use mapping for Australia - update January 2014' (ABARES 2014a) and 'VAST map for Queensland' (DSITIA 2012) II. The spatial distribution of agricultural land uses on non-forested agricultural land was modelled using area constraints based on the 2010-11 agricultural census data collected by the Australian Bureau of Statistics and reported at statistical local area (SLA) level (ABS 2012a and 2012b). Discrimination between land uses was provided by monthly AVHRR normalised difference vegetation index images with training data. The NDVI images covered the year represented by the dataset (1 April 2010 to 31 March 2011) and the four years represented by the training data (1 April 1997 to 31 March 2001). NDVI images were from the Bureau of Meteorology (BOM 2013) and splined by ABARES to reduce further the small proportion of no data. Training data were from an existing National Land and Water Resources Audit (NLWRA) database (NLWRA 2001) compiled by the then Bureau of Rural Sciences (BRS), now ABARES, for the NLWRA (Stewart et al. 2001). Agricultural land uses, dryland and irrigated, were allocated to non-forested agricultural land using an algorithm called SPREAD II, developed by Simon Barry, formerly of BRS (Smart et al. 2006). SPREAD II, like the SPREAD algorithm of Walker and Mallawaarachchi (1998), uses time series NDVI data with training data to spatially disaggregate agricultural census data. The SPREAD II methodology is statistically based, using a Bayesian technique - a Markov Chain Monte Carlo (MCMC) algorithm. To increase its discriminating power, SPREAD II was run using three spatial constraints. These and their main sources were: . Horticulture constraint - 'Catchment scale land use mapping for Australia - update January 2014' (ABARES 2014a) and 'GEODATA TOPO 250K series 3' (GA 2006) . Cultivation (cropping and horticulture) constraint - 'Native vegetation baseline 2004 version 1' (ABARES 2012b) and 'VAST map for Queensland' (DSITIA 2012) . Irrigation constraint - 'Catchment scale land use mapping for Australia - update January 2014' (ABARES 2014a) The SPREAD II outputs comprise agricultural commodity probability surfaces (floating point grids) and categorical summary land use maps for agricultural commodities and irrigation status (integer grids). Additional grazing land allocations were then made, outside SPREAD II, to pixels representing forested agricultural land with crown cover not exceeding 80%, in line with agricultural census area constraints. Allocations were determined separately for each SLA and were prioritised, firstly, by crown cover (woodland given priority over open forest) and, secondly, by slope (lowest slope given highest priority). Crown cover data were from the forest cover layer. Slope data were from a digital elevation model published by Geoscience Australia (GA 2008). III. The non-agricultural and agricultural land use data from steps I and II were combined in a single summary map with land use attribute values interpreted in terms of the Australian Land Use and Management Classification Version 7 (ABARES 2011). Positional Accuracy The data type and stated positional accuracy of the major source datasets used to determine the non-agricultural land uses and the distribution of agricultural land (as discussed in the lineage section) are as follows: . 'Australia, world heritage areas' (DSEWPaC 2012a) - positional accuracy not stated but the contribution of this dataset to the final product is negligible. . 'Catchment scale land use mapping for Australia - update January 2014' (ABARES 2014a) - original mapping scales are 1:250,000 or larger with national compilation as a 50 m raster dataset implying that positional errors do not exceed around 150 m. . 'Collaborative Australian protected areas database - CAPAD 2010' (DSEWPaC 2012b) - vector data, positional errors variable from < 1 m to 500 m. . 'Forests of Australia (2013)' (ABARES 2014b) - 100 m raster data; positional errors, in the main, do not exceed 100 m. . 'GEODATA TOPO 250K series 3' (GA 2006) - vector data, positional error less than 160 m for at least 90% of well-defined points. . 'Indigenous protected areas (IPA) - declared' (DE 2014) - a vector dataset with maximum positional errors varying from less than 1 m to 500 m. . 'Tenure of Australia's forests (2013)' (ABARES 2014c) - this is a 100 m raster dataset with positional errors for the underlying vector cadastral data stated as not exceeding 2 m in urban areas and 10 m in rural areas. It is expected that positional errors for the raster dataset, in the main, would not exceed 100 m. . 'VAST map for Queensland' (DSITIA 2012) - positional accuracy not stated but this input contributes more to attribute accuracy than to positional accuracy. Its positional errors are expected to be small compared to the 0.01 degree pixel size of the output summary grids. . 'Western Australia Aboriginal Lands Trust Estate as at May 2010' (DAA 2011) - this is a vector dataset based on an October 2010 extract from the WA digital cadastral database. Positional errors are largest in the region close to the junction of the South Australia, Northern Territory and Western Australia borders, an upper bound with a very generous margin for safety for this area being 500 m. The output summary grids have 0.01 degree pixel size. Therefore, positional errors for the non-agricultural land uses in the summary grids, in the main, should not exceed 1 km. For the agricultural land uses in the summary grids and for the probability surfaces, positional accuracy is not relevant since there are very few well-defined points from which it can be determinated. Attribute Accuracy Quantitative assessments of the attribute accuracy of the summary and probability grids have been limited. Some considerations that reflect, qualitatively, on the attribute accuracy of the summary grids follow. The accuracy of the non-agricultural land use assignments in the summary grids depends mainly on the attribute accuracy of the source datasets on which they are based. Accuracy of all assignments also depends on the validity of the rules used for land use assignment. Quantitative statements of attribute accuracy are available for several of the source datasets: . 'Australia, world heritage areas' (DSEWPaC 2012a) - all available attributes are stated to have been checked and to be correct. . 'Indigenous protected areas (IPA) - declared' (DE 2014) - the metatdata implies that a process for checking attributes was undertaken and states that all anomalies were corrected. . 'GEODATA TOPO 250K series 3' (GA 2006) - the range of allowable attribute errors is from 0.5% to 5% at a 99% confidence level. The attribute accuracy of the other source datasets is expected to be high. The catchment scale land use data and the protected areas data are subject to rigorous quality control. The attribute accuracy of the catchment scale land use data was at least 80% at the time of mapping (ABARES 2011). The currency of the protected areas data, the world heritage data, the forest data and most of the tenure data closely approximates the April 2010 to March 2011 target currency period. However, the currencies of the topographic data (currency end date 2006 or earlier) and of the catchment scale land use data (currencies ranging from 1997 to 2014) are not a good match for the target currency period. Overall, it is expected that the attribute accuracy of the non-agricultural land use assignments should be reasonably high. The accuracy of the agricultural land use allocations based on automated interpretation of NDVI images is variable. The probability grids give an indication of the accuracy of the agricultural land use allocations other than those mapped outside SPREAD II (grazing of forest). The rules used for all land use assignments, non-agricultural and agricultural, appear to be sound. Logical Consistency The logical consistency of the automated, rule-based land use assignments has not been verified but the rule set used is essentially the same as that used in the construction of the Land Use of Australia, Version 4, 2005-06, with only a few minor changes. Further, the script used to implement the assignments was derived from that used for the same purpose in the construction of the 2005-06 dataset by systematically updating the code so that all changes in input and output attribute classification systems were accommodated. For the 2005-06 dataset, the attribute combination corresponding to each land use assignment in the summary grid was tested by inspection and the automated assignments were verified to be as intended and to be logically consistent. It is expected that the same will be true of the 2010-11 dataset. Completeness Coverage and classification and verification are complete. Verification took the form of expert review of a draft version of the final dataset by ABARES' partners in the Australian Collaborative Land Use and Management Program (ACLUMP - see http://www.agriculture.gov.au/abares/aclump for more information about ACLUMP). Steps were taken in the construction of the final dataset to correct the major problems identified in the review. The 'Land Use of Australia 2010-11' supersedes the 'Land Use of Australia, Interim Version 5, 2010-11', which was published in July 2014 and for which classification and verification were incomplete. Contact Organisation Australian Bureau of Agricultural and Resource Economics and Sciences Contact Position Data Manager Mail Address GPO Box 858 Locality CANBERRA State ACT Country Australia Postcode 2601 Telephone +61 2 6272 4548 Electronic Mail Address info.abares@agriculture.gov.au Metadata Date 2016-06-23 Additional Metadata ABARES 2011, 'Guidelines for land use mapping in Australia: principles, procedures and definitions', edition 4, Australian Bureau of Agricultural and Resource Economics and Sciences, Canberra, available at data.daff.gov.au/anrdl/metadata_files/pe_abares99001806.xml. ABARES 2012a, 'Catchment scale land use of Australia - update November 2012', Australian Bureau of Agricultural and Resource Economics and Sciences, Canberra, available at data.daff.gov.au/anrdl/metadata_files/pb_luausr9abll07620121211.xml. ABARES 2012b, 'Native vegetation baseline 2004 version 1', Australian Bureau of Agricultural and Resource Economics and Sciences, Canberra, available at data.daff.gov.au/anrdl/metadata_files/pa_nvegbg9abll0042004_11a.xml. ABARES 2014a, 'Catchment scale land use mapping for Australia - update January 2014', unpublished dataset, Australian Bureau of Agricultural and Resource Economics and Sciences, Canberra. ABARES 2014b, 'Forests of Australia (2013)', Australian Bureau of Agricultural and Resource Economics and Sciences, Canberra, available at data.daff.gov.au/anrdl/metadata_files/pb_foa13g9abfs20140604_11a.xml. ABARES 2014c. 'Tenure of Australia's forests (2013)', Australian Bureau of Agricultural and Resource Economics and Sciences, Canberra, available at data.daff.gov.au/anrdl/metadata_files/pb_fta13g9abfs20140604_11a.xml. ABS 2012a, 'Agricultural commodities, Australia, 2010-11', cat. no. 7121.0, Australian Bureau of Statistics, Canberra, available at abs.gov.au/ AUSSTATS/abs@.nsf/Lookup/4618.0Main+Features12010-11?OpenDocument. ABS 2012b, 'Water use on Australian farms, 2010-11', cat. no. 4618.0, Australian Bureau of Statistics, Canberra, available at abs.gov.au/ AUSSTATS/abs@.nsf/Lookup/4618.0Main+Features12010-11?OpenDocument. BOM 2013, 'Normalized difference vegetation index (NDVI) - AVHRR, without atmospheric correction, Australia coverage', Bureau of Meteorology, Canberra, copy of 1 km monthly data covering relevant years supplied 24 September 2013. DAA 2011, 'Western Australia Aboriginal Lands Trust Estate as at 14 March 2011', Western Australian Government Department of Aboriginal Affairs, East Perth, Western Australia. DE 2014, 'Indigenous protected areas (IPA) - declared', Australian Government Department of the Environment, Canberra, available at www.environment.gov.au/fed/, accessed 6 April 2015. DSEWPaC 2012a, 'Australia, world heritage areas', Australian Government Department of the Environment, Canberra, available at www.environment.gov.au/fed/, accessed 23 January 2013. DSEWPaC 2012b, 'Collaborative Australian protected areas database - CAPAD 2010', Australian Government Department of the Environment, Canberra, available at www.environment.gov.au/fed/, accessed 24 January 2013. DSITIA 2012, 'VAST map for Queensland', 2 October 2012 compilation, unpublished dataset, Queensland Herbarium, Queensland Government Department of Science, Information Technology, Innovation and the Arts, Toowong, Queensland. GA 2006, 'GEODATA TOPO 250K series 3', Geoscience Australia, Canberra, available at www.ga.gov.au/metadata-gateway/metadata/record/64058/. GA 2008, 'GEODATA 9 second DEM and D8: digital elevation model version 3 and flow direction grid 2008', Geoscience Australia, Canberra, available at www.ga.gov.au/metadata-gateway/metadata/record/gcat_66006/. NLWRA, 2001, 'Control site database for the 1996/97 land use of Australia, version 2', unpublished dataset, Australian Bureau of Agricultural and Resource Economics and Sciences, Canberra. Smart, R 2016, 'User Guide for Land Use of Australia 2010-11', Australian Bureau of Agricultural and Resource Economics and Sciences, Canberra, available at, www.agriculture.gov.au/abares/publications. Smart, R, Knapp, S, Glover, J, Randall, L and Barry, S 2006, 'Regional scale land use mapping of Australia: 1992/93, 1993/94, 1996/97, 1998/99, 2000/01 and 2001/02 maps', version 3, unpublished report, Australian Bureau of Agricultural and Resource Economics and Sciences, Canberra. Stewart, JB, Smart, RV, Barry, SC and Veitch SM 2001, '1996/97 Land use of Australia: final report for project BRR5', National Land and Water Resources Audit, Canberra, available at www.agriculture.gov.au/abares/ aclump/Documents/Web_LU%20of%20Australia%201996_97.pdf. Walker, P.A. and Mallawaarachchi, T. 1998, 'Disaggregating agricultural statistics using NOAA-AVHRR NDVI', Remote Sensing of Environment, vol. 63, pp. 112-125, available at www.sciencedirect.com/science/article/pii/S0034425797001302.