Water Account, Australia methodology

The general methods used to compile the physical supply and use tables of the water account are relatively straightforward. Data from the WSSS collection operates as a stabilising component, acting as a spine for the compilation. The basic premise behind this decision is that the WSSS collection is a relatively complete census of all water providers – a population who, by necessity, understand the subject matter. The EWS/EWES/EIS surveys are utilised to inform industry splits, excluding Agriculture, and provide detail that is lacking in the WSSS collection instrument. The Agricultural Census and surveys (such as REACS) filled a similar niche for Agriculture up until 2021-22.

With this in mind the following represents a broad outline of how the data feeds into each supply and use water category.

Self-extracted water: the sum of all self-extracted water use sets the supply figure for the environment to the economy.

Distributed and reuse water: derived from the WSSS collection, the vast majority of water is supplied by the Water supply, sewerage and drainage services industry. EWS/EWES water supply estimates set total supply for water providers outside of the Water supply, sewerage and drainage services industry, while total use by industry is used to allocate these data to states/territories.

Wastewater: currently the WAA does not include estimates for the supply of wastewater to sewerage systems, split by individual industries and households. This is a known data gap that will be explored in future iterations of the WAA. In the interim, the total wastewater supplied is estimated in the account. The total wastewater supply is aligned to the total wastewater received by subdivision 28, sourced from the WSSS collection.

Return flows: estimates are sourced from the WSSS collection (state/territory and national) and EWS/EWES (national) surveys, however no data is currently available for return flows from Agriculture and wastewater flows from households and other industries. These are known data gaps that will be explored in future iterations of the WAA. For the Hydroelectricity generation industry, return flows estimates were aligned with self-extracted surface water. For this edition of the WAA, water discharged to the sea (including seawater used for cooling) and discharges to land have been excluded from return flows in the Physical Supply and Use Tables. These flows do not re-enter inland freshwater systems and are treated as consumptive use to align with SEEA-Water standards. This change improves conceptual consistency and ensures water consumption figures reflect water retained within inland systems.

Self-extracted water: the WSSS, EWS, EWES, EIS, and REACS (discontinued after 2021-22) surveys all collect data on self-extracted water use – these data are utilised directly. For 2022-23 and 2023-24, agricultural irrigation water use is estimated using the CSIRO satellite-based evapotranspiration model rather than ABS survey data. Self-extracted use for hydroelectricity is modelled using a combination of data collected directly from hydroelectricity companies and National Greenhouse and Energy Reporting (NGER) data.

Distributed and reuse water: WSSS and EWS/EWES/EIS supply estimates set the total. WSSS sets control totals for ANZSIC divisions A, B, C, and D. EWS/EWES supply and the residual WSSS supply is allocated to other industries. Use by the Water supply, sewerage and drainage services industry is sourced from the WSSS and includes water losses in distribution.

Wastewater: all wastewater use/collection is currently assigned to the Water supply, sewerage and drainage services industry, and data is collected via the WSSS collection. While some other industries are likely treating a small amount of wastewater this would not involve wastewater transferring between institutional units and is therefore, out of scope.

Return flows: represent flows from the economy to the environment; under SEEA conventions (SEEA Water S 3.16) all return flows are attributed to the environment. Data collected via the WSSS, EWS, EWES and EIS.

Water use by industry: sum of self-extracted water, distributed water, wastewater and reuse water use.

Water consumption by industry: water use by industry minus the sum of supply of distributed water, supply of wastewater, supply of reuse water and supply of return flows to the environment. Note that for all industries and households, estimates of the supply of wastewater (usually to the sewerage system) are not available due to difficulties in the disaggregation of collected wastewater volumes by industry and households (wastewater service providers are usually able to provide estimates of total wastewater collected but find it difficult to report the origin of the wastewater, split by industry/households). This leads to an overestimation in water consumption at the industry and households level. This will be rectified in the future when the ABS is able to produce reliable wastewater estimates at industry and households level.

Note: environmental flows are out of scope in this edition of the WAA. Environmental flows relate to (a) statutory requirements to maintain specific water regimes or (b) specific entitlements, which accrue annual allocations of water that can be extracted (or left in-stream) for environmental purposes.

With the ABS Agricultural Survey and Census discontinued after 2021-22, ABS adopted a new approach developed by the Commonwealth Scientific and Industrial Research Organisation (CSIRO) that uses satellite-derived evapotranspiration data to estimate irrigation volumes. This method fills a critical data gap and aligns with the ABS Agriculture Statistics Modernisation Program, which aims to reduce respondent burden and leverage innovative data sources.

The CSIRO method estimates irrigation water use through a monthly water balance model applied at 100m resolution across Australia’s irrigated agricultural lands. It integrates:

• Land use data: Australian Bureau of Agricultural and Resource Economics and Sciences (ABARES) Catchment Scale Land Use of Australia (CLUM) and Australian Land Use and Management (ALUM) classification to identify irrigable areas.
• Crop phenology: Moderate Resolution Imaging Spectroradiometer (MODIS) vegetation dynamics to define irrigation periods.
• Actual evapotranspiration (AET): derived from MODIS satellite imagery using the CSIRO MODIS Reflectance-based Scaling EvapoTranspiration (CMRSET) algorithm.
• Climate and hydrology inputs: precipitation, potential evapotranspiration (PET), soil moisture, and runoff from the Bureau of Meteorology’s Australian Water Resources Assessment Landscape (AWRA-L) model.

Irrigation is calculated as the residual water needed to meet crop evapotranspiration demand after accounting for rainfall and soil moisture:

• Irrigation (IWU)=(ΔSoil Moisture+AET+Runoff)-Precipitation

If rainfall and soil moisture meet crop water needs, irrigation is assumed to be zero.

1. Identify irrigable land using CLUM and ALUM classifications.
2. Define irrigation periods by state using MODIS phenology data.
3. Compile monthly water balance inputs (AET, PET, rainfall, soil moisture, runoff).
4. Calculate irrigation water use for each pixel and month.
5. Aggregate monthly estimates to annual totals and geographic units (states, catchments).
6. Adjust irrigated areas using ABS survey benchmarks (2005-06 to 2020-21) and calibrated thresholds for other years.

The CSIRO evapotranspiration method has undergone rigorous validation against ABS survey-based irrigation data (Water Use on Australian Farms) across multiple spatial scales and time periods. Validation focused on assessing alignment in both magnitude and trend, using statistical performance metrics widely applied in hydrological modelling.

• Reference Data: ABS survey estimates of applied irrigation water for 16 financial years (2005-06 to 2020-21).
• Spatial Scales:
  • State/Territory level: Annual comparisons for all states and territories.
  • SA4 level: Aggregated comparisons for 11 years (2010-11 to 2020-21).
  • SA2 level: Census years only (2010-11, 2015-16, 2020-21).
• Metrics Used:
  • R² (Coefficient of Determination): Measures goodness of fit.
  • Percent Bias: Indicates average over- or under-estimation.
  • Nash-Sutcliffe Efficiency (NSE) and Kling-Gupta Efficiency (KGE): Assess predictive accuracy and consistency.

• State/Territory Level:
  • R² improved from 0.75 (Phase 1) to 0.96 (Phase 2).
  • Percent bias reduced from +12.1% to –4.6%.
  • NSE and KGE increased to 0.96 and 0.94, respectively.
• SA4 Level:
  • R² improved from 0.70 to 0.83.
  • Bias reduced from +16.8% to –3.0%.
• SA2 Level:
  • R² improved from 0.59 to 0.76.
  • Bias reduced from –15.2% to –9.4%.

These improvements demonstrate that the Phase 2 method significantly enhances accuracy compared to earlier versions, particularly at broader spatial scales where ABS data are consistently available.

• Strong alignment at state and regional levels confirms the method is fit-for-purpose for national and state reporting.
• Small-area estimates (SA2) show reasonable agreement but remain a priority for future refinement.
• Performance is highest in major agricultural states (NSW, Victoria, Queensland), with minor variability in South Australia and Tasmania. Western Australia shows greater volatility due to diverse climate and irrigation practices, which will be addressed in future model improvements.

Overall, the validation results provide high confidence in the reliability of the evapotranspiration-based irrigation estimates for inclusion in the Water Account, Australia. While the method is labelled experimental to reflect ongoing refinement, its accuracy at state and national levels meets ABS quality standards.

• Estimates cover irrigated agriculture only; livestock water use is added separately.
• Currently unable to split surface and groundwater; disaggregation is planned for future releases.
• Small-area precision and crop-level detail will improve in later phases.
• The method assumes no lateral water movement between pixels and uses fixed irrigation periods by state, which may not capture year-to-year variability.

Planned improvements include:

• Disaggregation by crop type.
• Surface water vs groundwater split.
• Dynamic irrigation periods reflecting seasonal variability.
• Integration of administrative data (e.g., water licences) for improved accuracy.
• Backcasting of entire Water Account, Australia historical time series. 

The monetary water supply and use tables are displayed in purchasers’ and current prices. They generally follow the same format as the physical supply and use tables, however, there are a number of key differences, as outlined below:

• Self-extracted water and return flows are not currently valued as, generally, no volumetric charge is applied and there is a lack of available data. The majority of water that is extracted by the user and has a volumetric charge is included in distributed water in both the physical and monetary supply and use tables.
• Bulk water, or the supply of water between water suppliers, is valued in the monetary tables. As this is an intra-industry transfer the water flows/transactions are not included in the physical supply and use tables as this would lead to a double-counting of the supply of the water.
• The monetary supply and use tables also include relevant statistics from the Australian National Accounts, Input-Output Tables. These include import/export values, relevant taxes and subsidies, and social benefits paid in kind.
• Supply and use balancing is required as discrepancies exist between supply and use totals due to differing compilation techniques and data sources.
   

Self-extracted water: self-extracted water supply is currently not valued in the account.

Distributed water: revenue estimates for the Water supply, sewerage and drainage services industry are the sum of retail revenue received for distributed (potable and non-potable) water. Estimates for industries other than Water supply and sewerage services industry utilise household and industry derived unit values (prices) and the physical volume of distributed water supplied by these industries from the physical supply and use tables.

Reuse water: revenue estimates for the Water supply, sewerage and drainage services industry are the sum of retail revenue received for reuse water. Estimates for industries other than the Water supply, sewerage and drainage services industry utilise industry and household derived unit values (prices) and the physical volume of reuse water supplied by these industries from the physical and supply and use tables.

Bulk water, distributed: supply is the sum of the associated revenue from the WSSS collection. All revenue is allocated to the Water supply, sewerage and drainage services industry.
 
Bulk water, reuse: supply is the sum of the associated revenue from the WSSS collection. All revenue is allocated to the Water supply, sewerage and drainage services industry.

Sewerage wastewater: supply is set by total revenue received from sewerage charges and trade waste charges, as reported in the WSSS collection.

Return flows: currently not valued in the account.

Self-extracted water: currently not valued in the account.

Distributed water use for the Mining, Manufacturing, Electricity, Gas and Waste services industries are derived by multiplying the price of water paid by the physical use. The price is derived from data collected on the EWS, EWES and EIS surveys, by dividing expenditure on water by the volume of water used. 

Agricultural distributed water expenditure was calculated using data from the WSSS collection. An agricultural unit price was derived by dividing total revenue received for distributed water by the total volume of distributed water, at a state/territory level, for irrigation units. These unit prices were multiplied by the agricultural use of distributed water estimates from the physical supply and use tables.

Household estimates are derived by multiplying a unit price by physical use. The unit price is calculated using the Bureau of Meteorology’s National Performance Reports (NPR) data via the following methodology:

• The ‘Typical residential bill: water supply’ (data item P3, which includes fixed charges, usage charges and special levies) was multiplied by ‘Number of connected residential properties: water supply’ (data item C2) to derive total residential expenditure at a utility level.
• Total residential expenditure was then divided by ‘Total volume of water supplied to residential customers’ (data item W8) to derive a unit price ($ per kL) per utility.
• For all utilities without a unit price available using the above processes, the NPR weighted average unit price by state/territory was used; these unit prices were then multiplied by the total distributed water supply to residential customers for each utility.
• These were then aggregated to produce state/territory and national level household distributed water expenditure estimates.
• Note that in some cases data collected on the Water supply and sewerage services (WSSS) collection was used in preference to NPR data due to data quality issues. 

Reuse water: reuse water follows the same methodology as distributed water use.

Bulk water, distributed: expenditure on distributed bulk water mirrors the supply revenue values.

Bulk water, reuse: expenditure on reuse bulk water mirrors the supply revenue values.

Sewerage and drainage services: the EWS, EWES and EIS surveys provide estimates for the Mining, Manufacturing, Electricity, Gas and Waste services industries. For Agriculture and Other industries (ANZSIC Divisions E-S), estimates were available in the 2014-15 EWES and modelled forward using a Gross Value Added index by industry by state/territory. The gap between total supply and total industry use is allocated as household use.

Distributed water: total supply is used as the control total, with differences between supply and use allocated to use by Other industries (ANZSIC divisions E-S).

Reuse water: total supply is used as the control total. A weighted balancing approach was used to address discrepancies between the supply control total and use total, with the difference allocated between industries/households in the same proportions as reuse water use before balancing.

Bulk water, distributed: no balancing required.

Bulk water, reuse: no balancing required.

Sewerage and drainage services: total supply less total industry use is allocated to households' water use.

The summary tables aim to highlight key indicators of the WAA in a time series, and provide other relevant information linked with data from the supply and use tables. These indicators aim to enhance the ability of users to rapidly synthesise key stories. The data items presented in these tables, and not represented in the supply and use methodology described above, are described below:

• Total water use by industry: sum of self-extracted water, distributed water and reuse water use by industries.
• Total water use by industry (less Electricity, gas, water and waste services): sum of total self-extracted water, distributed water and reuse water use by industries, less the sum of total self-extracted water, distributed water and reuse water by the Electricity, gas, water and waste services industries. Note that a majority of the water used by these industries is either supplied to other users or returned to the environment (e.g. hydroelectricity water use).
• Total water use by households: sum of total distributed water and reuse water for households (note that self-extracted water use by households, including rainwater tanks and bores, is excluded from this edition of the WAA).
• Total water use: sum of self-extracted water, distributed water and reuse water use.
• Total water use less electricity and gas: sum of total self-extracted water, distributed water and reuse water use, less the sum of total self-extracted water, distributed water and reuse water by the Electricity, gas, water and waste services industries.
• Gross Value Added (GVA): Chain volume measures; sourced from Australian National Accounts: State Accounts.
• Estimated population as of 30 June: sourced from Australian Demographic Statistics.
• Estimated number of households at 30 June: based on a combination of latest estimated resident population; proportion of people living in private dwellings, from Censuses; average household sizes from Censuses; Census household counts; and Post Enumeration Survey household estimates.
• Total water use (less Electricity, gas, water and waste services) per capita: sum of total self-extracted water, distributed water and reuse water use, less the sum of total self-extracted water, distributed water and reuse water by the Electricity, gas, water and waste services industries divided by the estimated residential population.
• Total household water use per household: total water use by households (excluding self-extracted water) divided by the number of households.
• Total water consumption: total self-extracted water use less return flows.
• Water intensity: total water consumption (ML) divided by industry Gross Value Added ($ million).
• Water productivity: industry Gross Value Added ($ million) divided by total water consumption (ML).
• Expenditure per kL of water used by industry, distributed: expenditure on distributed water by industry, divided by total kL of distributed water used by industry.
• Expenditure per kL of water used by households, distributed: expenditure on distributed water by households, divided by total kL of distributed water used by households.
• Total expenditure by industry and households per capita, distributed: total expenditure on distributed water by industry and households, divided by the estimated residential population.
• Total household expenditure per household: total expenditure on distributed water by households, divided by households.
• Area-averaged rainfall: data sourced from Bureau of Meteorology, Climate change – trends and extremes
• Accessible volume in major dam storages as at 30 June: data sourced from Bureau of Meteorology, Water Storage Dashboard
   

A measured or estimated value of evapotranspiration under real-world conditions.

The Australian and New Zealand Standard Industrial Classification (ANZSIC) is the standard of classification used in Australia and New Zealand for the collection, compilation, and publication of statistics by industry.

A geologic formation which is capable of holding water and through which water can percolate. Aquifers are capable of yielding quantities of groundwater for economic activities.

The Australian Water Resources Assessment – Landscape model, developed by the Bureau of Meteorology to estimate water balance components such as soil moisture, runoff, and evapotranspiration across Australia.

Water supplied by a water provider/supplier/utility to another water provider/supplier/utility. Also known as "wholesale" water (as opposed to "retail" water).

The area of land determined by topographic features, within which rainfall will contribute to run-off at a particular point. The catchment for a major river and its tributaries is usually referred to as a river basin.

Annually-reweighted chain Laspeyres volume indexes referenced to the current price values in a chosen reference year (i.e. the year when the quarterly chain volume measures sum to the current price annual values). Chain Laspeyres volume measures are compiled by linking together (compounding) movements in volumes, calculated using the average prices of the previous financial year, and applying the compounded movements to the current price estimates of the reference year.

Estimates are valued at the prices of the period to which the observation relates. For example, estimates for 2017-18 are valued using 2017-18 prices. This contrasts to chain volume measures where the prices used in valuation refer to the prices of the previous year.

A remote sensing algorithm developed by CSIRO to estimate actual evapotranspiration using MODIS satellite data.

Water supplied to a user including through a natural (e.g. river) or non-natural network (e.g. piped or open channel), and where an economic transaction has occurred for the exchange of this water. The majority of distributed water is supplied by the Water supply, sewerage and drainage services industry (ANZSIC Division 28). The water supply component consists of units mainly engaged in storage, purification or distribution of water by pipeline or carrier. It may also include the operation of irrigation systems that supply water to a farm and the supply of steam and hot water. Distributed water can include potable, mains and raw water but does not include reuse or bulk water.

A unit (e.g. business or organisation) that engages in production and/or consumption activities.

The process by which liquid water is converted to water vapour and removed from a surface - such as a lake, soil, or wet vegetation - into the atmosphere.

The part of the water cycle where water from vegetated areas is transferred to the atmosphere through transpiration and evaporation.

Exports of goods and services consist of sales, barter, or gifts or grants, of goods and services from residents to non-residents.

One thousand million litres, or 1,000,000,000 litres. 1 Gigalitre = 1,000,000 cubic metres.

The gross value of commodities produced. It is the value placed on recorded production at the wholesale prices realised in the market place.

Water supplied to other water providers and customers, plus losses, own use by water providers and environmental flows.

Water which collects in porous layers of underground formations known as aquifers.

The value of an industry's output at basic prices, minus the value of goods and services consumed as inputs during the process of production. Basic prices valuation of output removes the distortion caused by variations in commodity taxes and subsidies across the output of individual industries.

Imports of goods and services consist of sales, barter, or gifts or grants, of goods and services from non-residents to residents.

Water artificially applied to soils (i.e. does not include precipitation/rainfall).

One thousand litres (1,000 litres). 1 kilolitre = 1 cubic metre.

The Moderate Resolution Imaging Spectroradiometer, a satellite sensor that captures global imagery for monitoring land, ocean, and atmospheric processes, including vegetation, land cover, and evapotranspiration.

One million litres (1,000,000 litres). 1 megalitre = 1,000 cubic metres.

In the physical water supply and use tables, "Other industries" refers to the following list of industries, according to the Australian and New Zealand Standard Industrial Classification (ANZSIC):

• Construction
• Wholesale trade
• Retail trade
• Accommodation and food services
• Transport, postal and warehousing
• Information media and telecommunications
• Financial and insurance services
• Rental, hiring and real estate services
• Professional, scientific and technical services
• Administrative and support services
• Public administration and safety
• Education and training
• Health care and social assistance
• Arts and recreation services
• Other services.

An organisation that supplies water but whose main activity is in an industry other than the Water supply, sewerage and drainage services industry (e.g. Mining or Manufacturing).

The study of seasonal biological events, such as plant growth and flowering, in relation to climate.

Water that is suitable for human consumption (i.e. drinking water).

The amount of water that would be transferred from land to the atmosphere by evaporation and transpiration, assuming unlimited water availability.

The amount paid by the purchaser, excluding any tax deductible by the purchaser, in order to take delivery of a unit of a good or service at the time and place required by the purchaser. The purchaser’s price of a good includes any transport charges paid separately by the purchaser to take delivery at the required time and place.

Water extracted from the environment that has not been treated.

Potable or non-potable water supplied by a water supplier to customers not classified as water suppliers.

Return flows represent the flows of water from industries and households to the environment. This excludes flows of water from industries and households to wastewater treatment facilities but includes flows of water from treatment facilities directly to the environment. The SEEA records this as use by the environment.

Wastewater (including sewage) that is collected, on-supplied and used again by other businesses/households. It may have been treated to some extent. Reuse water is also known as recycled water or effluent re-use. Recycling within industrial sites (also known as “on-site recycling) is excluded.

Revenue from 'pay for use' and base rate charges (i.e. volumetric and fixed rates) for the provision of wastewater. It also includes revenue generated from trade waste charges.

Revenue from the provision of water supply services to residential and non-residential properties. Also includes revenue generated from bulk supply or bulk sale of potable and non-potable/raw water, treated wastewater or reuse water and desalinated water.

Water that flows over the land surface toward rivers, lakes, or other water bodies.

A process where salt is removed from water with a high salt content (usually sea water but sometimes other brackish water) to make it suitable for domestic or industrial use.

Water extracted directly from the environment for use (including surface water, groundwater and other bodies) for the purpose of supplying to others (including other water suppliers) and/or for own use. It also includes water produced from salt water desalination.

Infrastructure (e.g. pipe and drains) used to transport sewage or domestic wastewater.

Social transfers in kind consist of good and services provided to households by government and private, non-market producers free or at prices that are not economically significant.

The amount of water contained in the soil, influencing plant growth and water availability.

Current unrequited payments that government units, including non-resident government units, make to enterprises based on the levels of their production activities or quantities or values of the goods or services that they produce.

Physical water supply and use tables which provide information on the volumes of water extracted, supplied within the economy and discharged back into the environment by economic activity and households.

Water on the surface of continents such as in a river, lake, wetland, ice or snow. This excludes sea water.

A framework used to develop environmental accounts by integrating environmental information into an accounting framework. The primary manual is the SEEA Central Framework (SEEA-CF) which provides the conceptual basis for describing the interrelationship between the natural environment and the economy.

An elaboration of the SEEA framework which provides a conceptual framework for organising hydrological and economic information in a coherent and consistent framework. It is consistent with SEEA-CF.

An international framework which can be used to develop a comprehensive, consistent and flexible set of macro-economic accounts. The latest edition is the 2008 SNA.

Compulsory, unrequited payments, in cash or in kind, made by institutional units to government units.

The process by which water in plant tissues is released to the atmosphere, primarily through small openings in leaves called stomata.

Any water that has been used by an economic unit or household, and then transferred between economic units, usually for treatment. The most commonly understood form of wastewater is sewage; however, non-sewage wastewater products are also included (for example, untreated effluent and trade waste).

The amount of water that is extracted from the environment by the economy (industries and households) for use, which is not distributed to other economic units nor returned to an inland water resource (e.g. river, dam, lake, creek), because during use it has been incorporated into products, evaporated, transpired by plants, or consumed by households or livestock. It may include losses due to evaporation occurring in distribution and apparent losses due to illegal tapping as well as malfunctioning metering.

Water intensity is referred to as the ratio of physical flow to an economic aggregate (for example, ML/$m GVA) while water productivity represents the ratio of economic aggregate to physical flow (for example, $m GVA/ML). Water intensity is the inverse of water productivity.

Water that enters the water distribution system of a water provider but does not reach the end users/customers. Water losses can be attributed to seepage, leakage, evaporation (excluding evaporation from water storages), meter inaccuracies and theft.

A business or organisation that provides a reticulated water supply, irrigation water, reuse/recycle water and/or bulk water supply service. Water providers may be government or private and often operate water storage, purification and supply services. They may also provide sewerage or drainage services. Commonly referred to as a water utility or water supplier.

Water leaving/flowing out from an economic unit. Water supply is the sum of water supply to other economic units (distributed water, reuse water and wastewater) and water supply to the environment (return flows).

The provision of reticulated water supply (including desalinated water), irrigation water, recycled (reuse) water, bulk and/or retail water supply service.

Water intake by an economic unit, or the economy. Water use is the sum of the amount of water extracted from the environment (including surface water, groundwater and sea water for desalination) for use in the economy, plus water use within the economy (i.e. distributed water use, reuse water use and wastewater collection). Note that the direct use of precipitation on to land, which leads to the use of the water held in soil by plants (e.g. rainfed agriculture), is outside of the scope of this publication and is not included.