20251209 11525400 1.0 TRUE 150000000 5000 ISO19139 002 Habitat Condition Assessment System (HCAS) version 3.3 90m gridded habitat condition long-term epoch 1988-2024 HCAS33_HCB_1988_2024 2024-12-20T00:00:00 2025-09-29T00:00:00 3.3 20250929 https://data.csiro.au/collection/csiro:65549 CSIRO Commonwealth Scientific and Industrial Research Organisation Living Landscapes Principle Research Scientist GPO Box 1700 Canberra ACT 2601 hcas@csiro.au AU DCCEEW Australian Government Department of Climate Change, Energy, the Environment and Water Director Decision Support & Analysis, Environment Data & Analysis Branch, Environment Information Australia GPO Box 3090 Canberra ACT 2601 geospatial@dcceew.gov.au AU DCCEEW - Data Manager Australian Government Department of Climate Change, Energy, the Environment and Water Spatial Data Manager, Environment Data & Analysis, Environment Information Australia GPO Box 3090 Canberra ACT 2601 geospatial@dcceew.gov.au AU The data collection is available at: Valavi R, Levick SR, Lehmann EA, Liu N, Giljohann KM, Williams KJ, Collings S, Johnson S, Botha EJ, Munroe SEM, Van Niel TG, Newnham G, Paget M, Malley C, Carlile P, Gunawardana D, Lyon P, Richards AE, Tetreault Campbell S and Ferrier S (2025) HCAS 3.3 (1988-2024) base model estimate of Habitat Condition (90m grid), National Connectivity Index 2.0 (NCI) and annual time series for continental Australia. Data collection 65549. CSIRO, Canberra, Australia. Landing page: https://data.csiro.au/collection/csiro:65549. The following publication, which describes the data, will be released in 2026: Williams KJ, Liu N, Valavi R, Levick SR, Lehmann EA, Collings S, Giljohann KM, Johnson S, Botha EJ, Munroe SEM, Van Niel TG, Newnham G, Paget M, Malley C, Carlile P, Gunawardana D, Lyon P, Richards AE, Tetreault-Campbell S and Ferrier S (2025) Habitat Condition Assessment System (HCAS) version 3.3: Guide to the 90-metre data collection. Technical report EP2025-2979. CSIRO, Canberra, Australia. Biodiversity Habitat Condition Ecosystem Condition Satellite Remote Sensing To provide a high-resolution, nationally consistent dataset and time series of habitat condition for use by government, industry and other organisations. <DIV STYLE="text-align:Left;"><DIV><DIV>CSIRO has prepared this dataset in partnership with the Australian Government Department of Climate Change, Energy, the Environment and Water (DCCEEW) for use in biodiversity assessment and national reporting such as State of the Environment, Ecosystem Accounting, Ecological Knowledge System, and annual corporate reporting metrics for the environment and heritage portfolio. The HCAS v3.3 is a 90m gridded dataset depicting habitat condition on a scale between 0.0 (ecosystem removed) and 1.0 (ecosystem in reference condition) for continental Australia. The index represents the contribution that a given site (90m grid cell) makes to the effective area of ecosystem integrity remaining within any given spatial reporting unit, expressed as a proportion of the contribution made by a site in reference condition. Ecosystem integrity is defined as the ecosystem’s capacity to maintain its characteristic composition, structure, functioning and self-organisation over time within a natural range of variability. Reference condition is defined as the condition against which past, present and future ecosystem condition is compared to in order to measure relative change over time. Habitat condition (syn. ecosystem condition) is a generalised estimate for use in whole of biodiversity assessment metrics and may not represent the specific habitat quality requirements of all component species. The data provided is a long-term epoch (also known as the ‘base model’) derived from Geoscience Australia’s archive of Landsat Earth observation remote sensing variables averaged over 37 years (1988-2024) – HCAS33_HCB_1988_2024.tif.This long-term epoch provides the overall ‘best’ estimate of habitat condition derived from the HCAS method, as an expression of the long-term pattern of ecosystem dynamics and capacity to recover following disturbance. The data is provided as a cloud optimised GeoTIFF (COGs) to facilitate use in web applications, 90 m grid resolution, Geographic Datum of Australia (GDA) 1994 (Australian Albers, EPSG:3577). HCAS v3.3 (production version 3) was derived using 'ClassicHCAS' software version 0.2.0 and 'HCAS-workflow' software version 1.2.1.Data extent is defined by any data pixel intersecting a coastline polygon as defined by the land fraction dataset (Liu 2024). The data and no-data extent of each grid layer encompasses the area within the Australian Continental Exclusive Economic Zone (EEZ) (Alcock et al., 2020), excluding territories of Cocos, Christmas, Norfolk, Macquarie, Heard, and McDonald Islands, as well as Antarctica (Liu and Newnham, 2024).The following publication, which describes the data, will be released in 2026: Williams KJ, Liu N, Valavi R, Levick SR, Lehmann EA, Collings S, Giljohann KM, Johnson S, Botha EJ, Munroe SEM, Van Niel TG, Newnham G, Paget M, Malley C, Carlile P, Gunawardana D, Lyon P, Richards AE, Tetreault-Campbell S and Ferrier S (2025) Habitat Condition Assessment System (HCAS) version 3.3: Guide to the 90-metre data collection. Technical report EP2025-2979. CSIRO, Canberra, Australia.The data collection is available at: Valavi R, Levick SR, Lehmann EA, Liu N, Giljohann KM, Williams KJ, Collings S, Johnson S, Botha EJ, Munroe SEM, Van Niel TG, Newnham G, Paget M, Malley C, Carlile P, Gunawardana D, Lyon P, Richards AE, Tetreault Campbell S and Ferrier S (2025) HCAS 3.3 (1988-2024) base model estimate of Habitat Condition (90m grid), National Connectivity Index 2.0 (NCI) and annual time series for continental Australia. Data collection 65549. CSIRO, Canberra, Australia. Landing page: <A href="https://data.csiro.au/collection/csiro:65549" STYLE="text-decoration:underline;">https://data.csiro.au/collection/csiro:65549</A>. </DIV></DIV></DIV> The Habitat Condition Assessment System for Australia is an ongoing partnership between CSIRO and the Australian Government Department of Climate Change, Energy, the Environment and Water (DCCEEW). The work presented in this document and data collection was funded by DCCEEW through the Priority Improvements to the Habitat Condition Assessment System project, the HCAS v3.x Update to 2023 project, and the National Ecosystem Accounting Project. CSIRO and DCCEEW acknowledge the Traditional Owners of country throughout Australia and recognise their continuing connection to land, waters and culture. We pay our respects to their Elders past and present. View CSIRO’s vision towards reconciliation and DCCEEW’s Statement of Commitment to First Nations people. Geoscience Australia is acknowledged for the Digital Earth Australia Surface Reflectance NBART Landsat Collection 3 GeoMedian, GeoMAD and Fractional Cover derived products, version 4.0.0. Ross Searle and contributors to the Terrestrial Ecosystem Research Network (TERN) are acknowledged for the compilation of 90 m environmental covariates and the Soil and Landscape Grid for Australia (SLGA) collections. We also acknowledge ABARES for supporting the multiple lines of evidence approach to the spatial delineation of reference sites by providing a draft list of authoritative datasets (i.e. produced and managed by State and Territory agencies) relevant to land use mapping. The Vegetation, Assets, States and Transitions (VAST) classification of condition scores involved the reuse of expert-elicited data collected as part of the Regional Ecosystem Accounting Pilot (REAP) project: ethics clearance 115/22. The reuse of this data was approved by CSIRO’s Social Science Human Research Ethics Committee (CSHREC) in accordance with the National Statement on Ethical Conduct in Human Research (2007, 2023): 197/23 for Reuse of expert nominated reference sites, non-reference sites and condition data for the ‘Priority Improvements to the Habitat Condition Assessment System’ project to develop HCAS v3. We acknowledge contributors to that data as listed on page 3 of Williams et al. (2023: CSIRO publication EP2023-1426; DOI: https://doi.org/10.25919/r12v-ya32). A technical report series details methods and innovations involved in developing the HCAS version 3 series. For the latest publications see: https://research.csiro.au/biodiversity-knowledge/projects/hcas/, or contact us at hcas@csiro.au or geospatial@dcceew.gov.au. We acknowledge other contributors to the development of HCAS including authors of the internationally published overarching framework for the HCAS method (Harwood et al., 2016), and subsequent incremental improvements through two trial implementations: HCAS v2.0 beta (Williams et al., 2020), HCAS v2.1 (Williams et al., 2021a), and pilot applications to ecosystem accounts: HCAS v2.2 (Williams et al., 2021a) and HCAS v2.3 (Williams et al., 2023b), and acknowledgments therein. Specifically, we acknowledge Dr Thomas D Harwood who developed the code that formed the foundation of HCAS v2. Further details are provided in the accompanying "Habitat Condition Assessment System (HCAS) version 3.3: A guide to the 90-metre data collection" (publication EP2025-2979), to be released in 2026, which can be downloaded separately from CSIRO's publication repository. <DIV STYLE="text-align:Left;"><DIV><DIV>Creative Commons by Attribution 4.0 International (CC BY 4.0)No restriction, nil cost.</DIV></DIV></DIV> 90 ClassicHCAS software version 0.2.0; HCAS-workflow software version 1.2.1 Compressed Cloud Optimised GeoTIFF (COG) - floating point COG Extent in Decimal Degrees when projected to GCS_GDA_1994. Actual data is in projection ESPG:3577 GDA94 Australian Albers Coordinate system: GDA_1994_Albers False Easting: 0 False Northing: 0 Central meridian: 132° Standard parallels: -18°, -36° Latitude of origin: 0° Datum: D_GDA_1994 1 104.303216 167.85181 -47.902522 -6.098252 DCCEEW Australian Government Department of Climate Change, Energy, the Environment and Water Director Decision Support & Analysis, Environment Data & Analysis Branch, Environment Information Australia GPO Box 3090 Canberra ACT 2601 geospatial@dcceew.gov.au AU DCCEEW DCCEEW - Data Manager Australian Government Department of Climate Change, Energy, the Environment and Water Spatial Data Manager, Environment Data & Analysis, Environment Information Australia GPO Box 3090 Canberra ACT 2601 geospatial@dcceew.gov.au AU DCCEEW - Data Manager CSIRO Commonwealth Scientific and Industrial Research Organisation Living Landscapes Principle Research Scientist GPO Box 1700 Canberra ACT 2601 hcas@csiro.au AU CSIRO DCCEEW Australian Government Department of Climate Change, Energy, the Environment and Water Director Decision Support & Analysis, Environment Data & Analysis Branch, Environment Information Australia GPO Box 3090 Canberra ACT 2601 geospatial@dcceew.gov.au AU DCCEEW CSIRO staff have exercised due care and skill in preparing and compiling the information and data in this publication. CSIRO advises that the information contained in this publication comprises general statements based on scientific research. The reader is advised and needs to be aware that such information may be incomplete or unable to be used in any specific situation. No reliance or actions must therefore be made on that information without seeking prior expert professional, scientific and technical advice. To the extent permitted by law, CSIRO (including its employees and consultants) excludes all liability to any person for any consequences, including but not limited to all losses, damages, costs, expenses and any other compensation, arising directly or indirectly from using this publication (in part or in whole) and any information or material contained in it. Biodiversity, Habitat Condition, Ecosystem Condition, Satellite Remote Sensing Compressed cloud-optimised GeoTiff (COG) – floating point The raster dataset value is a floating point decimal between 0.0 and 1.0. The 0-1 range is derived by rescaling the uncalibrated output of the model using a monotonic non-linear calibration protocol between the end points and two internal points positioned close to the mid-point of the minimum and maximum ranges (smoothed using a Hermite spline). The choice of calibration protocol affects the grid cell value but not its ranking. As there is no ‘truth’ for the estimate of condition, it is not possible to specifically assess attribute accuracy. Instead, the dataset is compared with similar data generated by other organisations using a Type II regression analysis for continuous data and a concordance analysis (similar to map accuracy assessments) for ordered data, and other qualitative evaluations targeting suspect issues to provide a schedule of limitations to inform future improvements. These results are detailed in technical reports (in preparation) that document the HCAS version series 3 development. Note: There is no field program for specifically validating HCAS. There is no central aggregation and harmonisation of field surveyed condition data generated by a wide range of organisations. These gaps limit the ability to routinely and systematically validate the accuracy of HCAS estimates. Incomplete sources of information about the location of contemporary reference sites (where ecosystems continue to function within a natural range of variability – maintaining their characteristic composition, structure, functioning and self-organization over time), or uneven representation of sampled reference sites, can also affect attribute accuracy of the condition estimate. This depends on the validity and completeness of the inferred reference sites (Giljohann et al., 2024). Additional sources of information about the presence and validity of a reference site, including through consultation with regional experts, will improve the local estimation of habitat condition and derivatives. In order to produce a seamless coverage, gaps continent-wide, in the remote sensing input variables were filled. The fill method is detailed in technical reports and is summarised in documentation provided with the data collection. The fill method is not 100% consistent with surrounding areas in all places where a fill has been applied. We have noticed that the condition estimate may be less reliable where there was a gap in the remote sensing input variables. The fill could result in estimation bias when condition benchmarks are drawn from areas that have not been filled and are used to test areas that have been filled (and to some extent vice versa) – testing of this insight into the mechanism is underway. The impact of gaps in the remote sensing input variables is less evident in the 35-year long-term epoch because most gaps are filled across the time series during averaging – except in the case of fractional cover variables where water is treated as a data gap. Spatial artefacts associated with input remote sensing data may also be apparent. Other sources affecting completeness include how well the selected set of remote sensing input variables represent the three primary groupings of ecosystem characteristics required for an estimate of ecosystem condition: structure, function and composition. From a remote sensing perspective, structure refers to the physical appearance of the ecosystem such as the stratification of canopies – their height, cover and volume. Function refers to the physical, chemical and biological processes that transform and translocate energy or materials in an ecosystem such as photosynthesis, water and nutrient cycling. Composition refers to the variety of life forms that make up an ecosystem, that may vary from place to place but perform similar roles in terms of overall ecosystem structure and function. The selected set of 14 remotely sensed ecosystem characteristic variables derived from the Landsat archive for use in HCAS v3.1 imperfectly characterise the above primary groupings of condition variables, due to optical sensor limitations. Therefore HCAS estimates of habitat condition may be thought of as a partial estimate of condition – Landsat sensors cannot depict belowground components of ecosystems or characteristics of ecosystem structure below dense canopy foliage or where ecosystems may have been impacted by invasive species where these impacts are not yet evident using remote detection methods. 20251216 7C3F9BC1-9F3F-4395-BE6A-FAA107D4802E DCCEEW Australian Government Department of Climate Change, Energy, the Environment and Water Director Decision Support & Analysis, Environment Data & Analysis Branch, Environment Information Australia GPO Box 3090 Canberra ACT 2601 geospatial@dcceew.gov.au AU DCCEEW DCCEEW - Data Manager Australian Government Department of Climate Change, Energy, the Environment and Water Spatial Data Manager, Environment Data & Analysis, Environment Information Australia GPO Box 3090 Canberra ACT 2601 geospatial@dcceew.gov.au AU DCCEEW - Data Manager CSIRO Commonwealth Scientific and Industrial Research Organisation Living Landscapes Principle Research Scientist GPO Box 1700 Canberra ACT 2601 hcas@csiro.au AU CSIRO ESPG The HCAS v3.3 product suite was developed at 90 m grid resolution in Australian Albers projection (GDA 1994) using a combination of: • 14 annualised time series of remotely sensed ecosystem characteristic variables, 1988 to 2024 (Levick et al., 2025), sourced from the Digital Earth Australia Surface Reflectance NBART Landsat Analysis Ready Data Collection 3 (Commonwealth of Australia, 2021) Derivative Products version 4.0.0 (Geoscience Australia, 2024; Lymburner, 2024) • 58 environmental covariates selected from more than 120 'non-anthropogenic' candidates largely sourced from the Terrestrial Ecosystem Research Network (TERN) compilation (Malone et al., 2025; Searle, 2023) with additional custom variables developed by Liu et al. (2025) • spatially inferred reference sites as training data (Giljohann et al., 2025b), sampled to represent the most intact remaining examples of Australia’s varied ecosystems and their environments (Valavi et al., 2025a) • spatially inferred reference sites as benchmark data (Giljohann et al., 2025b), sampled to represent both remotely sensed ecosystem characteristic variables and their environments from among the most intact remaining examples of Australia’s ecosystems, with an emphasis on individual water bodies and protected area properties (Valavi et al., 2025a). The HCAS v3.3 reference ecosystem model (REM) was developed using a generalised additive model (GAM) with 488,619 reference sites as training data, 46 environmental covariates (selected from 58), and 14 remotely sensed ecosystem characteristic variables summarised over 37 years (1988 to 2024). The REM was used in the HCAS v3.4 base model with the 1988 to 2024 long-term epoch and 590,342 benchmark reference sites. Short-term epochs for each of the 14 remotely sensed ecosystem characteristic variables were derived as 3-, 5- or 10-year antecedent rolling averages (the target year is the end year) utilising the time series (1988 to 2024). These processing steps are detailed in Valavi et al. (2025a). Model-based uncertainty estimates were derived from more than 100 bootstrap runs applied to the sample of training and benchmark reference sites to derive 95% confidence interval limits (upper 97.5th percentile condition estimate and lower 2.5th percentile condition estimate and the 95% confidence interval width) for each 90x90m pixel of the long-term epoch (1988-2024), and selected 3-, 5-, and 10-year epochs. All data products were derived using a 90m spatial grid in Australian Albers (GDA94 / Australian Albers). The "Habitat Condition Assessment System (HCAS) version 3.3: A guide to the 90-metre data collection", accompanying this data collection (see related links), outlines how the HCAS v3.3 differs from the HCAS v3.1 (Valavi et al. 2025; Williams et al. 2025), and includes a general discussion of limitations of particular note, and a data quality statement. Technical reports provide details about the inputs, processing methods, outputs and uncertainty quantification, applied in developing HCAS version 3 series. For the latest publications see: https://research.csiro.au/biodiversity-knowledge/projects/hcas/. Contact us at hcas@csiro.au or geospatial@dcceew.gov.au. Enterprise Geodatabase Raster Dataset 1