Great Barrier Reef Ecosystem Accounts, experimental estimates methodology

Ecosystem extent accounts describe the extent of the ecosystem types present in an accounting area and how the extent changes within the accounting period. The accounts record the total area of each ecosystem, classified by type within a specified area. Extent data is also used when deriving indicators relating to composition and change in ecosystem types across a region. Terrestrial, Marine, and Coastal ecosystems are measured by area (hectares). 

Spatial pre-processing

The final outputs adhere to the following specifications:

• 100 m grid resolution where possible (a) (b)
• ecosystem condition rasters resampled from 90 m to 100 m
• Statistical geographies were converted from polygons to 100 m rasters
• extent based on the NEAP Basic Spatial Unit (BSU) grid was aligned to NVIS version 7.0 grid and projection (DCCEEW)
• defined by a GDA2020 Australian Albers equal area projection.

(a) Mangroves and Saltmarsh data are at 10 m grid resolution.

(b) Marine extent data are at 250 m grid resolution.

Double counting

Due to differences in source data, some areas are double counted. The sum of extent across realms exceeds the total area associated with the GBR. The total area double counted in this release is approximately 282,000 hectares (0.36% of the total area associated with the GBR). 

Ecosystem classification

IUCN GET

The International Union for Conservation of Nature (IUCN) produces a classification system, the Global Ecosystem Typology (IUCN GET), as a framework to describe functionally different ecosystem types. The IUCN GET is a hierarchical classification which provides realm, biome, and EFGs. Each realm’s extent section will record the EFGs included. IUCN GET 2.0 is used for the GBR Ecosystem Accounts. 

The SEEA EA defines the condition of an ecosystem in terms of quantitative metrics describing its abiotic, biotic and landscape characteristics. Condition is then assessed with respect to an ecosystem’s composition, structure and function, which underpin its ecosystem integrity. 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.

Concepts

Ecosystem Condition Typology

The SEEA EA defines the condition of an ecosystem by measuring quantitative metrics describing biotic and abiotic characteristics. These characteristics can be classified according to the SEEA Ecosystem Condition Typology (SEEA ECT) shown in Table 2. Selecting for a range of metrics across the SEEA ECT will ensure the accounts are representative of overall condition.

Additional to the SEEA ECT, the SEEA EA also recognises the utility of including ancillary data, for example, recording environmental pressures as a proxy for condition, provided that this linkage is well documented.

Condition accounts

SEEA EA has a three-stage approach to developing ecosystem condition measures, with each stage building upon the previous stage:

• Stage 1: variable account
• Stage 2: indicator account
• Stage 3: condition index

Variables

Variables provide a quantitative metric for a characteristic of interest. Values are not compared against baselines, so do not signify good or bad ecosystem condition. Values generally reflect measured observations and require ecological expertise and/or local context to interpret meaningfully.

Indicators

Indicators provide normalised measures of condition variables, for example, by rescaling variables relative to a reference (ideal) level or condition. This makes indicators more intuitive to interpret and more practical for decision-makers.

Reference level

The reference level is an agreed-upon benchmark that describes the state of an ecosystem when its integrity is considered intact. It serves as a point of comparison for assessing ecosystem condition: if a measurement is close to the reference level, ecosystem condition is considered high; if a measurement is far from the reference level, ecosystem condition is considered low.

Indices

Indices are composite indicators that are aggregated from combinations of individual ecosystem condition indicators. Indices generalise a large amount of data and are useful in communicating general trends. 

The GBR Ecosystem Accounts include condition variables, indicators and indices.

Ecosystem services

Ecosystem services accounts show ecosystems’ contributions to economic and other human activities. Ecosystem services are central to the ecosystem accounting framework because they provide the link between ecosystem assets on the one hand, and the benefits derived from them and enjoyed by people on the other. This account uses ecosystem services as defined under SEEA EA.

Valuation

Ecosystems provide a range of contributions to society and the economy that may not be readily captured in market transactions, nor are these always recognised in economic activities or decision-making processes. Quantifying the economic value of services provided by ecosystems helps governments, businesses and communities to make informed decisions on resource use and allocation, including those coming from the natural world.

In ecosystem accounting, monetary valuation enables comparisons of ecosystem services and assets that are consistent with standard measures of services and assets as recorded in the national accounts. The SEEA EA framework has a tiered approach to deriving monetary values for ecosystem services. Below are the valuation methods in the order of preference:

1. methods where the price for the ecosystem service is directly observable;
2. methods where the price for the ecosystem service is obtained from markets for similar goods and services;
3. methods where the price for the ecosystem service is embodied in a market transaction (includes residual value, resource rent method, hedonic pricing);
4. methods where the price for the ecosystem services is based on revealed expenditures (costs) for related goods and services (includes averting behaviour, travel cost);
5. methods where the price for the ecosystem service is based on expected expenditures or markets (includes replacement cost, avoided damage, simulated exchange markets).

Several techniques have been considered to estimate the value of select ecosystem services, including:  

• observed market prices
• observed transactions 
• replacement cost 

In some instances, there may be several methods available to produce estimates for the value of ecosystem services. Where estimates are available from multiple methods, our estimates primarily draw on methods that rely on directly observable values. Future iterations of the accounts will continue to test valuation methods and different methods may be used if they are found to be more appropriate.

Concepts

Terrestrial realm

The Terrestrial realm includes all dry land, its vegetation cover, proximate atmosphere and substrate (soils, rocks) to the rooting depth of plants, and associated animals and microbes (IUCN GET). Terrestrial realm EFGs are listed in Table 3.

Freshwater realm

The Freshwater realm includes all permanent and temporary freshwater bodies as well as saline water bodies that are not directly connected to the oceans, and are characterised by water regimes defined by frequency, duration, flow velocity, depth and extent of inundation (IUCN GET). Freshwater realm EFGs are listed in Table 4.

Transitional realms

Transitional realms represent areas where the core realms (Terrestrial, Freshwater, Marine, and Subterranean) meet and interact. The Terrestrial realm transitions to Palustrine wetlands, Shoreline and supralittoral and Brackish tidal biomes. Refer to Table 5 for the listing of EFGs within these Transitional biomes. Transitional realm biomes are reported with the Terrestrial realm, with the exception of the Brackish tidal biome, which is reported as part of the Coastal realm, please refer to Coastal realm.

Data sources

CSIRO MFT extent

Marine-Freshwater-Terrestrial realm spatial extent data were compiled at IUCN GET EFG level by CSIRO for the GBR Ecosystem Accounts. It uses:

• NVIS version 7.0 Extant Theme (DCCEEW) to define natural EFGs
• NVIS version 7.0 Pre-1750 Theme (DCCEEW) to define the pre-1750 extent of natural EFGs not in the extant dataset
• Geofabric version 3.3 (BoM) to define Lakes and wetlands biome EFGs
• OzCoasts (GA) to define Semi-confined transitional waters biome EFGs
• Land Use of Australia (ABARES) to define Intensive land-use biome EFGs
• DEA Land Cover (GA) to define annual changes between Land Use of Australia reporting years
• Interim Biogeographic Regionalisation for Australia Version 7 (DCCEEW) to identify natural EFGs across Terrestrial and Transitional realms (Terrestrial-Freshwater, Terrestrial-Freshwater-Marine)
• Note: while the extent of freshwater ecosystems was calculated alongside the terrestrial extent, the results are published separately in the freshwater section.

Output specifications:

• annual rasters for pre-1750 and 2010–11 to 2023–24 (a)
• 100 m grid spatial resolution

(a) Note that Terrestrial and Transitional realms’ extents were only published until 2020–21.

Methods

Spatial extent rasters classified at IUCN GET EFG were combined with geography rasters to derive the geographical extent for each EFG within the GBR area and NRM regions within the GBR area. The area of each EFG category was calculated for each region by counting the pixels present for each category in each region and multiplying by the area of the pixel. Each 100 m pixel represents one hectare.

Unclassified treatment

Pixels coded as Unclassified represent cases where an EFG could not be confidently assigned and where Marine realm EFGs that appeared in the Terrestrial extent outputs were reassigned to Unclassified. Unclassified pixels outside the IMCRA coastline were excluded from Terrestrial, Freshwater or Transitional (excluding Coastal) ecosystems data and reported under Marine ecosystems.

A small number of EFGs are not spatially distinct in the IUCN GET level 3 classification (EFG). The Permanently open riverine estuaries and bays EFG in the Freshwater-Marine realm spatially and conceptually overlaps a number of EFGs from the Marine shelf biome of the Marine realm. This results in approximately 409,000 hectares of Terrestrial, Freshwater or Transitional (excluding Coastal) ecosystems overlapping with other EFGs. The impact is minor (approximately 0.52% of the area associated with the GBR is reported as Terrestrial, Freshwater or Transitional (excluding Coastal) EFGs and also in another realm). The GBR extent data cube contains the following caveat: ‘Some areas reported as Terrestrial, Freshwater or Transitional EFGs are also reported within other realms. The sum of extent reported in separate realms may be greater than the total area associated with the Great Barrier Reef (GBR). The impact is minor (approximately 0.52% of the area associated with the GBR is reported as Terrestrial, Freshwater or Transitional (excluding Coastal) EFGs and also in another realm).’

Concepts

Remote sensing data was used to derive biotic and abiotic ecosystem condition variables. The variables represent a diverse set of condition characteristics to populate the ecosystem condition accounts following the SEEA ECT. The remote sensing variables included in the accounts are shown in Table 6.

Data sources

The terrestrial condition variables were compiled by CSIRO as part of their Habitat Condition Assessment System (HCAS) Version 3.3 dataset. HCAS provides a set of 14 condition variables, five of which were selected to provide a representative subset of typology classes from the SEEA ECT. This data combines inputs from a variety of sources:

• reference sites that represent the most intact examples of Australia’s ecosystems (CSIRO)
• environmental covariates (Terrestrial Ecosystem Research Network (TERN))
• Digital Earth Australia Surface Reflectance NBART Landsat Analysis Ready Data Collection 3 (GA).

Methods

The condition variables are represented in 3-year time periods (epochs), which are based on calendar years. The epoch with the end year that matched the end year of the financial year was used to represent the account years of interest, for example, the 2010–11 account year was used to represent the 2009 to 2011 epoch year. Each condition metric was summarised by EFG and output geographic region using area-weighted averages.

Concepts

Crop provisioning services represent the ecosystem contribution to the growth of cultivated plants that are harvested by economic units for various uses, including food and fibre production, fodder and energy. This is a final ecosystem service used by the agricultural industry for the purposes of generating an economic return. The contribution of the ecosystem occurs up to the point of harvest, and both the supply and use of ecosystem services are recorded in the location of harvest.

The account will contain an isolated case study focusing on sugarcane.

Data sources

Production and crop area

Total production and total crop areas for sugarcane were sourced from Australian Agriculture: Broadacre Crops (ABS), which provides statistics on the production and value of a range of broadacre crops. The 2022–23 sugarcane values were used to align with the scope of the account.

Financial performance

Comprehensive financial, physical and management information for sugarcane was sourced from the Australian Bureau of Agricultural and Resource Economics and Sciences (ABARES) Financial performance of sugarcane farms survey. This survey provides detailed income and expenditure data for 2020–21 across the growing regions of the Great Barrier Reef and was used in deriving the split of human and environmental contributions in the crop provisioning estimates.

Methods

Physical

The proportion of human contribution to the production of sugarcane was calculated for each ABARES growing region based on the financial performance survey. The costs assumed related to production were;

• Contracts
• Hired labour
• Fertiliser
• Fuel, oil and grease
• Repairs and maintenance
• Crop chemicals
• Electricity
• Motor vehicles
• Water.

 The proportion of economic inputs attributable to human inputs was calculated using the total costs and the costs related to production attributable to human inputs. It was assumed that the remainder of the input would come from environmental factors, representing the ecosystem contribution. These proportions were applied to the physical inputs.

Total sugarcane production was multiplied by the proportion of ecosystem contribution to estimate the ecosystem contribution to production:

Ecosystem Contribution to Production (tonnes) = Total Production (tonnes) × % ecosystem contribution

Statistical Area 2s (SA2) were mapped to the ABARES sugarcane growing regions. Ecosystem contribution to production and total crop area estimates were aggregated to SA4s and the Great Barrier Reef.

Monetary

Hedonic pricing is conceptually the ideal method for implementing the land‑rental approach because it breaks down value into individual parts and would more clearly separate human inputs from the land‑rent component. Land value reflects a bundle of economic and environmental attributes. However, applying a hedonic model requires detailed data on human inputs and land characteristics, which are not currently available.

To estimate the monetary value of crop provisioning services, we; 

1. Use the land area within ecosystem assets
2. Multiply these by a combination of land rental values and council rates per hectare.

Cash expenditure on land rent and council rates were used as a proxy for land rental value because most sugarcane farmland is owner‑operated rather than rented. Land rental value is typically revealed through market transactions such as rents and land sales. It is recognised that council rates are an administrative charge calculated using land value. Councils generally base these charges on unimproved land value (ULV) or site value, both of which are derived from market sources. Land values for agriculture will reflect, in part, climatic conditions conducive to crop growing.

Concepts

Coastal realm

The Coastal realm includes the Brackish tidal biome at the interface of Terrestrial, Freshwater, and Marine realms. These ecosystems range from coastal riverways and terrestrial-dominated coastal saltmarshes to marine-dominated intertidal forests (Table 7).

Data sources

Geoscience Australia’s (GA) DEA Coastal Ecosystems dataset provides extent data for multiple coastal ecosystems at 10 m resolution. The coastal product is derived from Sentinel-2 imagery and defines the extent of Saltmarsh and Mangroves for two years, 2021 and 2022. The model used to assign ecosystems in the DEA Coastal Ecosystems product is highly sensitive to small changes in the remotely sensed inputs. Slight changes in inputs can push the model across binary probability thresholds from year to year, causing significant changes in extent. Therefore, the comparison between reference years presented in this publication should be interpreted with caution and, accordingly, no change matrix is presented. Refer to the Caveats and Limitations section of the DEA Coastal Ecosystems documentation for more information.

Methods

Spatial extent data from GA’s DEA Coastal Ecosystems dataset has been aggregated according to GBR regions.

Mangroves are contained within MFT1.2 Intertidal forests and shrublands, and Saltmarsh within MFT1.3 Coastal saltmarshes and reedbeds (IUCN GET). MFT1.2 and MFT1.3 were included in the 100 m CSIRO MFT Extent, but were replaced by the higher resolution DEA Coastal ecosystems dataset. They have been labelled according to the GA dataset used, i.e. Mangroves and Saltmarsh.

It is acknowledged that this approach results in an overestimate of the total area where GA Mangrove and Saltmarsh overlap other EFGs from CSIRO data. Additionally, any CSIRO MFT1.2 (mangrove) and MFT1.3 (saltmarsh) that falls outside of GA’s Coastal Ecosystems extent and outside the CSIRO Benthic Marine extent will not be included. This method results in approximately 92,000 hectares of Mangrove and Saltmarsh ecosystems overlapping with other EFGs. Coastal extent data cubes contain the following caveat: Some areas reported as Coastal EFGs are also reported within other realms. The sum of extent reported across all realms may be greater than the total area associated with the Great Barrier Reef. The impact is minor (approximately 0.12% of the area associated with the GBR is reported as both Coastal EFGs and another realm).

Concepts

Mangrove canopy cover

Mangrove canopy cover indicates the overall health and productivity of mangrove forests. Monitoring changes to mangrove canopy cover density can help describe how mangrove forests are responding to environmental changes, as canopy cover is typically reduced when mangroves experience ecosystem disturbance.

Data sources

GA’s DEA Mangroves (Landsat) dataset measures Mangroves at a 30 m resolution, annually since 1987. The product includes the canopy density of mangroves across Australian coastlines.

Methods

Mangrove canopy cover is reported each financial year, from 2010–11 to 2021–22. The extent dataset (GA 10 m DEA Coastal Ecosystems) is more accurate and has higher spatial resolution than the canopy cover model, but is only reported for the two most recent years. To keep the condition account consistent and aligned with the more accurate extent model, canopy cover has been spatially constrained to the Mangrove extent dataset. A one-to-one mask was used for the last two years, where the time periods overlapped for both extent and condition. The earlier canopy cover models, between 2010–11 and 2019–20 inclusive, were applied to the closest available extent year, 2020–21.

The GA mangrove canopy cover dataset has 3 classifications, which are defined by canopy cover percentages, as described in Table 8. Any region without a classification was allocated as Unclassified. 

Mangrove canopy cover rasters were aggregated by geography rasters to derive the geographical extent for each level of cover. This was expressed as a percentage of the total regional Mangrove extent.

Concepts

Coastal protection services

Coastal protection services describe the protection afforded to coastal communities and infrastructure from tides and storm surges by coastal ecosystems. These ecosystems may include coral reefs, sand banks, dunes or mangrove ecosystems. This is a final ecosystem service. Measurements of coastal protection services included in this account are the counts of dwellings and residents protected by Mangroves and Saltmarsh in the event of a tidal or storm surge.

Storm surge

A storm surge is an abnormal rise in sea level above the  normal tide levels due to a storm. These powerful ocean movements are caused by strong winds, waves and atmospheric lows, piling water up against the coast, for example when a cyclone approaches. Erosion and damage to infrastructure, including loss of property, can be caused by coastal inundation during extreme high water levels and storms.

Mangrove and Saltmarsh ecosystems reduce the damage to dwellings during a storm surge by dissipating wave and wind energy.

Dwellings

For the dwelling count, a dwelling was defined as a structure which is intended to have people live in it, and which was habitable on Census night. Unoccupied residences of owners, managers or caretakers of caravan parks, marinas and manufactured home estates are also counted, but other unoccupied dwellings in such establishments are not counted, please refer to Census dictionary: dwelling.

Address Register Dwellings

The ABS Address Register is an up-to-date, comprehensive list of all known physical addresses within Australia and excludes non-physical addresses such as post office boxes and email addresses. For more information, please refer to: ABS Address Register, Users' Guide.

People

The counts of people are taken from the 2021 Census and are geocoded to ABS ASGS Mesh Blocks based on their usual place of residence (PURP). Population counts are perturbed to prevent the release of identifiable data.

Data sources

Mangrove extent

Geoscience Australia’s DEA Coastal Ecosystems product was used to define the extent of Saltmarsh and Mangroves for the coastal protection model. DEA Coastal Ecosystems was derived from Sentinel-2 imagery and is reported at a 10 m resolution for the years 2021 and 2022.

Coastline

Coastlines were sourced from the Geoscience Australia (GA) Digital Earth Australia (DEA) coastlines product, which includes the mainland and islands for both 2021 and 2022.

Coastal Connectivity

A coastal connectivity model was used to determine which coastal areas were at risk of inundation.  Prototype connectivity layers were created by Geoscience Australia for the ecosystem accounts. These combine coastal elevation with local Highest Astronomical Tides (HAT) and use connectivity filters to prevent isolated ‘bathtub’ inundation regions. This model was then filtered by the ABS to correspond to a 1-in-100-year coastal inundation event.

Population and dwelling counts

Population and dwelling data was sourced from the ABS Census Mesh Block counts. The usual resident population and total dwelling count per mesh block were both taken from the 2021 Census.

Dwelling location

Dwelling locations were sourced from the ABS Residential Address Register Common Frame, which contains spatial points for all residences in Australia. The November 2021 and November 2022 frames were chosen, to align with the time periods for mangrove extent.

Methods

Physical

Mangroves and Saltmarsh were reduced in size with a negative buffer to ensure that only the areas wide enough to offer protection were included in the protection model. A minimum belt width of at least 100 m for Mangroves and 40 m for Saltmarsh has been determined to offer coastal protection. The whole study area was then spatially constrained to the coastal connectivity model. Mangroves, Saltmarsh, dwellings and people were only included in the model if they were within the area at risk of a 1-in-100 year coastal inundation event.

The shortest distance from the coastline was then calculated for every point within the connectivity mask, for four different scenarios:

1. No Mangroves or Saltmarsh ‘barriers’, i.e. straight-line distance from the coast.
2. Mangroves providing barriers, through which water cannot travel.
3. Saltmarsh providing barriers.
4. Both Mangroves and Saltmarsh providing barriers.

The linear distance without mangroves or saltmarsh was subtracted from each of the three ‘protected’ scenarios to determine the extra distance required to travel to each point with the barriers in place. An area was considered protected if the distance from the coastline with barriers was greater than the distance without.

The areas at risk of inundation and areas protected were then overlaid with the Address Register Common Frame and Census Mesh Blocks (2021). The proportion of Address Register Common Frame dwellings protected per Census Mesh Block were used to scale the final Census dwelling and population counts. 

Results were finally aggregated to GBR regions based on a population-weighted correspondence between Mesh Blocks and these regions. This method was repeated for each Mangrove and Saltmarsh extent dataset, 2020–21 and 2021–22.

Mangroves and Saltmarsh can provide protection to the same dwellings or residents, so the total protection offered by both is less than the sum of the protection offered by each. Number of dwellings and people protected by Saltmarsh in Cape York and Wet Tropics was confidentialised and not available for publication.

Monetary

As experimental statistics, components may be refined, added or removed over time. In this release, monetary estimates of coastal protection service are not included due to further methodological development work. Physical estimates of the number of people and dwellings protected by Coastal ecosystems are unaffected by this decision.

Concepts

Global climate regulation

Global climate regulation services are the ecosystem contributions to reducing concentrations of greenhouse gases (GHG) in the atmosphere through the removal (sequestration) of carbon from the atmosphere and the retention (storage) of carbon in ecosystems. These services support the regulation of the chemical composition of the atmosphere and oceans to stabilise the climate, in turn avoiding damages that arise due to climate change. This is a final ecosystem service.

Carbon sequestration

The carbon sequestration component of the service reflects the ability of ecosystems to remove carbon from the atmosphere. 

Carbon retention

The carbon retention component of global climate regulation reflects the ability of ecosystems to accumulate and retain the stock of carbon – i.e. ecosystems supply a service through the avoided emission of carbon to the atmosphere.

Data sources

BlueCAM

Clean Energy Regulator’s (CER) Blue Carbon Accounting Model (BlueCAM) is available publicly as a downloadable Excel Workbook and has been used indirectly to source various parameters missing from sequestration focused code, such as baseline aboveground biomass and soil carbon stocks.

CSIRO supplied BlueCAM R functions, based on the BlueCAM model, have been used to calculate the net sequestration in cases where no ecosystem change occurs.

Carbon price

The price of carbon is sourced from the Australian Carbon Credit Units Scheme auctions (CER), the change in carbon price reflects the institutional settings at the time. It is acknowledged that carbon markets in Australia are not yet considered 'mature', yet prices such as those set through the Clean Energy Regulator’s auction can be considered suitable for estimates of climate regulation, as they reflect a market-based outcome and institutional settings. 

Methods

Physical

Carbon sequestration and retention have been estimated for blue carbon ecosystems: Mangroves, Saltmarsh and Seagrass meadows using the CER BlueCAM concepts. Extent was collated by GBR regions and allocated to BlueCAM climatic regions in accordance with Climate Change in Australia (CCA) definitions. Tidal range and elevation were assumed constant (1 m and 0 m respectively). Mangroves were assumed to be mature with a conservative age of 50 years. 

Net sequestration included sequestration by above and belowground biomass as well as soil sequestration, offset by carbon, methane, and nitrous oxide emissions. Emissions for Seagrass ecosystems were assumed to be insignificant, whereas such offsets were calculated for Mangrove and Saltmarsh ecosystems.

Carbon retention followed the same approach but applied baseline retention estimates reflective of mature ecosystems as available in the BlueCAM model. Carbon retention estimates relate to closing carbon stock, i.e. baseline retention (opening stock) plus sequestration over the financial year. Baseline retention includes carbon stock contained within aboveground biomass, belowground biomass and soil carbon stores. Aboveground biomass and soil carbon were estimated using BlueCAM’s inbuilt maximum parameters for mature ecosystems by climatic region. Mangrove belowground biomass was estimated from the aboveground stock using a “roots to shoots” ratio of 0.32 as per the BlueCAM technical manual and following similar processes as used in DCCEEW’s Full Carbon Accounting Model (FullCAM). Roots and rhizomes for seagrass and saltmarsh are typically less than 2 mm in diameter and fall within the soil carbon category, therefore separate belowground stock for these ecosystems was not calculated.

Monetary

The value of carbon retention is in mitigating the release of stored carbon into the atmosphere. Based on the work by Edens (2021), the carbon retention value can be represented by an annual yield that reflects a return to asset holders in maintaining the stored stock of carbon on behalf of society. In the estimation of the carbon retention value we:

1. estimate the carbon stock in ecosystem assets, 
   1. multiply this by a suitable carbon price and 
   2. convert this into an annual service flow by multiplying this value by a nominal suitable rate of return to generate an annual yield. 

The rate of return reflects a yield comprising a real discount rate of 4% plus the relevant nominal price level, in line with the national accounts. 

To avoid double counting of the sequestration component of retention, only carbon retention monetary values have been included. Valuation approaches have not yet been internationally adopted as a standard, as practitioners seek to test ways to best reflect carbon retention values.

Concepts

The Marine realm includes all connected saline ocean waters characterised by waves, tides and currents (IUCN GET). Within the Marine realm, the Marine shelf and Deep sea floors biomes represent benthic ecosystems, as shown in Table 9. The Pelagic ocean waters biome describes the water column above the benthic ecosystems.

Each EFG within the Pelagic ocean waters biome represents a depth range as follows:

• Epipelagic ocean waters (0 to 200 m depth)
• Mesopelagic ocean waters (200 to 1000 m depth)
• Bathypelagic ocean waters (1000 to 3000 m depth)
• Abyssopelagic ocean waters (3000 m to 6000 m depth).

Data sources

CSIRO’s Spatial Extent of IUCN ecosystem functional groups data provides the spatial extent of the Marine realm at the IUCN GET EFG level for Australia. Some EFGs in the IUCN GET are not included or have been combined in this dataset. Marine Pelagic ocean waters data was derived from depth measurements in the AusBathyTopo 250 m (Australia) 2023 Grid. Data for the Marine shelf and Deep sea floors biomes was derived from the Natural Values Ecosystems map as described in Methods for developing extent of IUCN Global Ecosystem Typology ecosystem functional groups across the marine, freshwater and terrestrial realms in Australia: pre-1750 and 2010–11 to 2023–24. The Natural Values Ecosystems map combines several data sources to provide broad-scale categorisation of Australia’s marine ecosystems. Cells are initially assigned to broad, sediment-based ecosystems based on depth ranges, on top of which, ecosystem data was added (e.g. reefs, seagrass) as described in Designing a targeted monitoring program to support evidence-based management of Australian Marine Parks: National Implementation.

Extent estimates of Photic coral reefs in this publication may exceed those from other sources due to differences in methodology and classification. The CSIRO's source use predictive model that includes both shallow and deeper mesophotic reefs that may not be reported as coral reefs in other data sources. The data source also assumes all reefs north of −32.69° latitude (approximately the latitude of Hawks Nest, New South Wales) are coral, whereas all reefs south of this latitude are assumed to be rocky reefs.

The Marine shelf and Deep sea floors biomes are contained in one raster with a reference period of 2021–22. The Pelagic ocean waters biome is in a separate raster with a reference period of 2022–23. Both are 250 m resolution rasters.

Methods

Marine benthic and pelagic extent rasters were intersected with geography rasters to summarise the area of each EFG by geographic region. For both benthic and pelagic Marine extent, the Integrated Marine and Coastal Regionalisation of Australia (IMCRA) geography was used to define the boundary between marine and non-marine ecosystems. 

Separate extent rasters are used for benthic Marine ecosystems and Terrestrial, Freshwater and Transitional ecosystems, so areas might be counted as more than one ecosystem at once. Ecosystem overlap occurs where benthic marine EFGs intersect transitional EFGs (e.g. Permanently open riverine estuaries and bays) or coastal EFGs. Approximately 338,000 hectares of benthic marine EFGs overlap with other EFGs. Benthic Marine extent data cubes contain the following caveat: Some areas reported as benthic Marine EFGs are also reported within other realms. The sum of extent reported in separate realms may be greater than the total area associated with the Great Barrier Reef (GBR). The impact is minor (approximately 0.43% of the area associated with the GBR is reported as benthic Marine EFGs and also in another realm).

Concepts

The marine water quality index was derived by CSIRO from two indicators: Secchi depth and Chlorophyll-a concentration. 

• Secchi depth is a visual measure of water clarity and a proxy for light penetration. It is measured using a black and white patterned disc called a Secchi disc. The Secchi depth is the depth at which the Secchi disc is no longer visible underwater. A greater Secchi depth indicates more water clarity, so better water quality. 
• Chlorophyll a (Chl-a) concentration indicates phytoplankton biomass (algae) in a water body. Phytoplankton grow quickly in response to nutrient availability, so elevated values of Chl-a can indicate increased nutrient loading. Higher Chl-a concentration indicates lower water quality.

Data sources

Marine water quality indicators and index were compiled by CSIRO from eReefs model outputs. Chl-a concentration and Secchi depth data were taken from the eReefs biogeochemical model outputs and standardised as indicators between zero (very poor) to 100 (very good) relative to the corresponding water quality guideline value. Scores for each parameter were averaged by region and over the water year (October to September to generate annual scores for each region. The water quality index for each region was calculated as the average of Chl-a and Secchi depth indicators. A Reef-wide score was calculated as the area weighted average of regional scores.

Methods

For the full methodology, please refer to Reef Water Quality Report Card 2021 and 2022 Marine modelling methods.

Water quality indicators and index were divided by 100 to convert to a scale between 0 and 1. Data was published by GBR region and for the whole Reef.

Concepts

The long-term health of inshore seagrass meadows is measured through seagrass abundance and resilience, which are summarised as the Seagrass Index.

• Seagrass abundance is measured using the median seagrass percent cover relative to the site or reference guideline (habitat type within each GBR region). 
• Seagrass resilience is a combination of resistance (capacity to cope with disturbance) and recovery (adaptability to change). Resistance depends on species composition and meadow condition. Recovery considers reproductive effort (number of reproductive structures such as fruits, flowers, spathes) and seedbank reserves (counted after sieving collected cores).

Data sources

Seagrass condition data was compiled by the Centre for Tropical Water and Aquatic Ecosystem Research (TropWATER) at James Cook University (JCU) as part of GBRMPA’s Marine Monitoring Program (MMP) and provided to ABS. Inshore seagrass meadows are monitored using intertidal walkover surveys and scuba or drop-camera photo surveys, as well as core sampling. Seagrass condition is measured by assessing two indicators: abundance as percent cover, and resilience incorporating species composition and reproductive status estimated from sample cores.

Methods

For the full methodology, please refer to Marine Monitoring Program: Annual Report Inshore Seagrass Monitoring 2022–23.

Seagrass indicators and index were divided by 100 to convert to a scale between 0 and 1 for the GBR Ecosystem Accounts.

Concepts

The Coral Index is derived from five coral condition indicators, which are combined as a uniform weighted average.

• Coral cover is an indicator of corals' ability to resist the cumulative environmental pressures to which they have been exposed, but also the relative size of the population of corals as a source of larvae. 
• Macroalgae proportion within the algal community is an indicator of the risk of competition with corals – note that scores for this indicator decline with increasing levels of macroalgae. 
• Juvenile coral density is an indicator of the success of early life history stages in the replenishment of coral populations. 
• Rate of coral Cover change is an indicator of the recovery potential of coral communities due to growth.
• Hard coral community Composition is an indicator of selective pressures imposed by the environmental conditions around a reef.

Data sources

Coral monitoring data was compiled by the Australian Institute of Marine Science (AIMS) for GBRMPA’s Marine Monitoring Program (MMP) and provided to ABS. The MMP coral monitoring program combines multiple surveys, using photo transects, juvenile coral transects and scuba transects. A metric has been developed by AIMS for each of the five indicators, to score condition on a consistent scale between zero and one.

Methods

For full methodology, please refer to Marine Monitoring Program: Annual Report Inshore Coral Reef Monitoring 2022–23.

Reef level scores were calculated as the area weighted mean of regional scores based on the proportion of inshore coral reef area within the four regions: Wet Tropics (0.209), Burdekin (0.092), Mackay-Whitsunday (0.381) and Fitzroy (0.318). Cape York and Burnett Mary are not included due to sampling challenges. Sampling typically occurs over winter months, and coral condition data has been allocated to the preceding financial year to maintain the linkage with the summer disturbances, e.g. sampling from May-July 2018 was allocated to 2017-18 financial year.

Concepts

Wild fish provisioning services represent the contribution of ecosystems to the provision of fish and other seafoods. This is a final ecosystem service used by the fishing industry for the purposes of generating an economic return. The contribution of the ecosystem occurs up to the point of harvest, and both the supply and use of ecosystem services are recorded in the location of harvest. This ecosystem service excludes fish charters and recreational fishing.

Data sources

Estimates of physical catch and gross value of production (GVP) for wild fish provisioning services were provided by the Queensland Department of Primary Industries. The information provided was sourced from the Queensland Fisheries Data Mining Environment. Queensland's rent revenue share values were sourced from CSIRO’s supplementary National data for Australia’s ecosystem services: fisheries biomass provisioning services.

Methods

Physical

Commercial fisheries retained catch, by weight, was reported for GBR regions and aggregated by taxonomic group (finfish, crustaceans, and molluscs) for the financial years 2010–11 to 2022–23.

Monetary

The estimates of monetary value of wild fish provisioning services comprise: 

• the quota or licence fees for single species fisheries where available or 
• a rent revenue share for multi-species fisheries, based on the market value of fisheries, to distribute a quota fee across different wild fish catch.

Quota and licence prices established in the market reflect the expected resource rent generated by their use. However, not all fisheries have quota or license fee data and many fishing operations occur in multi-species fisheries where not all fish species caught have a quota or license fee.

The rent revenue share approach divides the estimated rent (annualised total quota sales value or licence asset value) at the aggregate fishery level by the total revenue of the entire fishery to produce a rent revenue share. This share is applied to the GVP in the fishery to produce an ecosystem service value.

Monetary estimates of wild fish provisioning services are provided for 2017–18 to 2021–22 as these years have more data available for recorded quota and license fees consistent with a market-based outcome for deriving monetary values of wild fish provisioning services.

Concepts

The SEEA EA defines cultural services as “the experiential and intangible services related to the perceived or actual qualities of ecosystems whose existence and functioning contribute to a range of cultural benefits… They are intended to capture the variety of ways in which people connect, and identify with, nature and the variety of motivations for those connections.” Recreation-related services are included as an example of cultural services and defined by the SEEA EA as “contributions of ecosystems, in particular through their biophysical characteristics and qualities, that enable people to use and enjoy the environment through direct, in situ, physical and experiential interactions with the environment. These include services both to locals and to non-locals (i.e. visitors, including tourists).”

Recreation-related services for the marine realm were estimated for the Great Barrier Reef Marine Park (GBRMP) from visitor and fee data collected by GBRMPA. GBRMPA charge an Environmental Management Charge (EMC) to commercial licensed operators for each visitor to the GBRMP based on length and type of visit, including: 

• Full day (≥3 hours)
• Part day (<3 hours)
• Non-Standard rates (including scenic flights or coral viewing boat tours)
• and Exempt (non-charged visitation). 

The EMC is likely to be an underestimate of the park's monetary value of recreational services as it only includes entry fees which reflects an administrative charge. There are a range of other exchange-based expenditures, such as travel, that may be directly linked to visiting a national park.

Data sources

Visitation and pricing data associated with the EMC was sourced from the GBRMPA. 

Data from Tourism Research Australia’s (TRA) International Visitor Survey (IVS) and National Visitor Survey (NVS) was used to inform the proportion of international and domestic visitation.

Methods

Quarterly visitation data by activity type (full day visits, part day visits, coral viewing, scenic flights and exempt visits) were collated from EMC data by GBR Management Area and financial year without alteration. Visitor numbers for each quarter and visitor type were multiplied by the relevant fee and aggregated by financial year to estimate monetary value.

Where there were 10 or less visits of a particular visit type during a quarter within an area, the source data was confidentialised. This has been reported as 'np', not available for publication, but included in totals. Some trips spanned more than one Management Area and some trips had locations that are not associated with a Management Area (e.g. airports). For these reasons, the sum of regions will not add up to the Great Barrier Reef total.

The EMC logbooks do not record whether the charge is paid by an international or domestic visitor. IVS and NVS tourism data was used to impute use by households (domestic visitors) or as exports (international visitors). Tourism data was available by Tourism regions, which were allocated to GBR Management Areas as shown in Table 10.

COVID-19 impacts and treatment 

EMC charges were waived during COVID-19 (1 April 2020 to 1 July 2023) to encourage visitation and support businesses. While no revenue was generated, the ecosystem service was still being provided, therefore past years’ fees were used to estimate monetary value over this period.

No IVS interviewing occurred from April 2020 to December 2022 due to COVID-19 and regional data on international visitors was unavailable for 2020-21 to 2022-23. For these years, international ratios for each Tourism region were scaled using Queensland international to domestic ratios. For back years where international splits were unavailable, total use was published but household/export splits were not.

Concepts

Supplementary data tables on tourism metrics and socio-economic statistics have been included to provide greater regional context to the accounts. All data is presented in financial years as of 30 June without alteration from the source publications.

Data sources

Socio-economic data was sourced from ABS Data by Region for SA2 and SA4 regions.

Tourism socio-economic data by Tourism region was sourced from TRA publications including:

• Regional Tourism Satellite Account (RTSA) (consumption, GVA, GRP, jobs), 
• National Visitor Survey (NVS) (Domestic visits (overnight and daytrips), 
• International Visitor Survey (IVS) (International visits), 
• Additional unpublished, regional data that was provided directly by TRA to ABS.

Methods

Socio-economic data at the SA2 level has been concorded to Tourism regions where appropriate based on the TRA SA2 to Tourism region concordance. This included SA2s outside the included SA4s to ensure comparison with tourism region data, particularly in Far North Queensland. The data has been collated for comparison, but otherwise no changes have been made to the data.

Data by region data included regional data on population, income, jobs by industry, number of businesses by industry, and gross value of agricultural production. 

• The estimated resident population is population census  data adjusted to include Australian residents temporarily overseas, natural increase in population, net migration, and amendments to the Census to calculate the year population by region as of 30 June.
• Estimates of Aboriginal and Torres Strait Islander population are similarly adjusted from census data as  of 30 June. 
• Employee income is computed from various data collected from individual income tax returns. This is presented regionally as a median and total. 
• Number of employee jobs by Industry refers to filled jobs in which an employee receives remuneration in wages, salary, payment in kind, or piece rates, excluding self-employed positions. These are calculated based on the linked employer employee dataset (LEED) and presented by industry. 
• Number of businesses by Industry includes actively trading businesses as at 30th of June for the past 5 years as extracted from the ABS Business Register. 
• Gross value of agricultural production includes crops, livestock slaughtered and other disposals, and other livestock products in wholesale prices. 

Tourism data included number of domestic and international visitors, number of tourism jobs, tourism consumption, tourism gross value added, and tourism gross regional product. Domestic visits include day trips and overnight stays. Day trips involve a round-trip distance of at least 50 km and minimum of 4 hours. Overnight stays include at least one night away from home a minimum of 40 km away. International visits include short term (less than 12 months) visits to Australia by non-residents over 15 years old. Note that due to restricted international travel no data was available for the years 2019−20 to 2022−23. Tourism consumption is defined as the expenditure of visitors on tourism-related products in purchaser’s prices. Gross Value Added (GVA) measures contribution to the economy by an industry and is defined as the gross output of an industry minus intermediate consumption in basic prices.

Concepts

The Social and Economic Long-Term Monitoring Program for the Great Barrier Reef (SELTMP) collects and reports data on the human dimensions of the GBR. Selected SELTMP survey results for residents are included in the GBR Ecosystem Accounts as supplementary material to add social context to the GBR regions and demonstrate how perceptions and values have changed over time.

Data sources

Social attitude data was collected by CSIRO as part of the SELTMP and provided to ABS. CSIRO conducted surveys of residents in the GBR catchment region in 2013, 2017, 2021 and 2023. Data can be viewed by GBR region, refer to dashboard.

Methods

Response rates by region varied over time (Table 11). Cape York had the smallest number of respondents (n = 13 in 2023) with gaps in key demographics and is unlikely to be a representative sample. Burnett Mary had few respondents in 2017 (n = 46) compared to other years (n = 431 to 733), so comparisons between years must be applied with caution.

CSIRO administered surveys over four separate years, however due to survey question wording and delivery mode changes, only 2021 and 2023 results have been included in the GBR Ecosystem Accounts for most questions. For visitation frequency, 'More than once a week' included 'several times a week' and 'almost daily' from the 2013 survey and 'several times a week' and '> 3 times a week' from the 2017 survey. Respondents who had never visited the reef were recoded as 'Not at all'.

Survey results were collated by GBR Region as well as the overall Great Barrier Reef and presented by number and percentage of respondents for each survey year.

Beaman R (2023) AusBathyTopo 250m (Australia) 2023 Grid - A High-resolution Depth Model for Australia (20230004C). Geoscience Australia, Canberra.

Dunstan P and Hayes K (2023) Parks Australia's Natural Values Ecosystems [Data set]. CSIRO & NESP Marine and Coastal Hub. 

Dunstan PK, Woolley SNC, Monk J, Barrett N, Hayes KR, Foster S, Howe SA, Logan D, Samson CR and Francis SO (2023) Designing a targeted monitoring program to support evidence-based management of Australian Marine Parks: National Implementation. Report to the National Environmental Science Program, CSIRO, Australia.

Edens, B (2021) The treatment of global climate regulation service in SEEA EA and its potential for informing climate policy, United Nations Statistics Division.

Govekar PD, Griffin C and Beggs H (2022) Multi-Sensor Sea Surface Temperature Products from the Australian Bureau of Meteorology. Remote Sensing, 14(15), 3785. 

Jaffrés JBD, Cuff B, Cuff C, Knott M and Rasmussen C (2022) Hydrological characteristics of Australia: national catchment classification and regional relationships, Journal of Hydrology, 612 (Part A).

Liu N, Newnham G, Richards AE, Pascoe S and Tetreault Campbell S (2025) Methods for developing the 100 m resolution Basic Spatial Units (BSU) for national ecosystem accounts. Technical report EP2025-2871. A report for the National Ecosystem Accounting Project. CSIRO, Australia. 

Liu N, Newnham G, Richards AE, Pascoe S and Tetreault-Campbell S (2025) Basic Spatial Units (BSU) at 100 m resolution for national ecosystem accounts. Data collection 65932. CSIRO, Australia. 

Macfarlane C, Liu N, Newnham GJ, Richards AE and Tetreault-Campbell S (2025) Methods for developing extent of IUCN Global Ecosystem Typology ecosystem functional groups across the marine, freshwater and terrestrial realms in Australia: pre-1750 and 2010–11 to 2023–24. Technical report EP2025-2871. A report for the National Ecosystem Accounting Project. CSIRO, Australia.

Macfarlane C, Liu N, Newnham GJ, Richards AE and Tetreault-Campbell S (2025) Extent of IUCN Global Ecosystem Typology ecosystem functional groups across the marine, freshwater and terrestrial realms in Australia: pre-1750 to 2023-24. Data collection 65802. CSIRO, Australia. 

Phillips C, Newey V, Lyons M, Sagar S and Bishop-Taylor R (2025) DEA Coastal Ecosystems. Geoscience Australia, Canberra. 

Pascoe S, Evans D, Liu N, Scheufele G and Tetreault Campbell S (2025) Methods for developing a national dataset for Australia’s ecosystem services: fisheries biomass provisioning services (2000-01 to 2022-23). Technical report 2025-2392. A report for the National Ecosystem Accounting Project CSIRO, Australia. 

Pascoe S, Evans D, Liu N, Scheufele G and Tetreault Campbell S (2025) National data for Australia’s ecosystem services: fisheries biomass provisioning services (2000-01 to 2022-23). Data collection 65810. CSIRO, Australia. 

Valavi R, Levick S, Lehmann E, Liu N, Giljohann K, Williams K, Collings S, Johnson S, Botha H, Munroe S, Van Niel T, Newnham G, Paget M, Malley C, Carlile P, Gunawardana D, Lyon P, Richards A, 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, Australia. 

Wijffels SE, Beggs H, Griffin C, Middleton JF, Cahill M, King E, Jones E, Feng M, Benthuysen JA, Steinberg CR and Sutton P (2018) A fine spatial-scale sea surface temperature atlas of the Australian regional seas (SSTAARS): Seasonal variability and trends around Australasia and New Zealand revisited. Journal of Marine Systems, 187, 156–196. 

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, Australia.