Farm dam accounting for healthy and safe agricultural catchments

ABSTRACT

Farm dams are essential infrastructure for many farming operations, including livestock watering and crop irrigation. With demand for food increasing, many farmers are looking to capture, store and use more runoff to achieve higher yields under worsening climate extremes. The aim of this article is to advance understanding of farm dams in catchments and the current tools for accounting for them, providing an opportunity to consider a new runoff storage accounting framework to support integrated policy. The Australian dam management setting is provided for context. A longitudinal cohort survey of 254 South Australian farmers provides a snapshot of farmer perceptions of dams and dam management behaviours. Farmers express increasing concern about the future availability of water, how much is allocated to the environment and how this is accounted for. Whilst the field of IWRM and others have integrated disciplines for improved water decision-making, so far there has been limited attention to how to incentivise improved farm dam management through links to farm accounting and business. This study proposes a preliminary framework for assisting dam decision-making across scales to support the health and safety of environments and communities.

KEYWORDS:

• Water accounting
• farm dams
• catchment runoff
• water sharing
• integrated policy

Introduction

Supplies of freshwater for stock watering and irrigation are likely to worsen under anthropogenic climate change, with increasing droughts and evaporation, and heavier rainfalls. Demands on food production are growing and by 2050 arid countries like Australia will struggle to meet freshwater demands. Australia has over 1.8 million farm dams contributing to AU$17.7 billion of agricultural value (Malerba, Wright, and Macreadie 2022) and yet these dams can also cumulatively threaten downstream communities and environments if not managed well (Malerba, Wright, and Macreadie 2021; Tingey-Holyoak, Burritt, and Pisaniello 2013). The cumulative impact of farm dams¹ constructed across a catchment² is large because they alter the natural flow regime and capture all the runoff generated from their upstream area aggregating risk across two potential exacerbation extremes: (1) drought conditions during limited rainfall periods (Graymore and McBride 2013) – dams create barriers in a natural flow path and cumulatively create a water sharing issue for downstream users including the environment (Morden et al. 2022; Nathan and Lowe 2012; Schreider et al. 2002); and (2) flood conditions during intense rainfall (Pisaniello, Dam, and Tingey-Holyoak 2015). The age of most dams is reaching a critical point in many countries around the world and they were designed decades ago with very poor knowledge of extreme flood events, let alone with the data on how rapidly the climate is changing, which creates a greater safety issue for downstream communities (Wishart et al. 2020).

Examples of small dam water sharing and water storage safety problems include: farm dams along the Mekong River have been blamed for cumulatively worsening drought conditions and even held responsible for several ecological extinctions, including that of the Chinese River Dolphin (Bowmer 2014; Kibler and Tullos 2013; Lewis 2013). Recently, it has been found that farm dams in catchments cumulatively threaten downstream environments 280–380 per cent more than large dams (Morden et al. 2022). In China, the 1975 failure of the Shimantan and Banqiao dams was the result of the cumulative failure of 60 upstream dams and more than 230,000 people died (Xu, Han, and Fu 2022). In 1982 the failure of the 8 m high Lawn Lake dam in Colorado, United States (US) killed three people and caused US$31M in damage including loss of bridges, roads, a hydroelectric plant, and 177 businesses (Kinster 1987). Indeed, Graham’s (1999) study of US dam failures looked at the number of deaths from such incidences over the period of 1960–1998 finding that the failure of dams less than 15 m high caused 88 per cent of fatalities, while the failure of very small dams (less than 6 m high) caused two per cent of the deaths.

The challenge is not just the many millions of existing dams, of which the World Bank finds 90 per cent are small, but over the next decade it has been estimated that farm dam growth will exponentially increase around the world (Wishart et al. 2020). In Australia where there are already 1.8 million farm dams (650,000 of these in the Murray Darling Basin) it is estimated further farm dam growth over the next decade will reduce annual flows into the Murray-Darling system by 180 GL, more than one third of the capacity of Sydney Harbour.

Despite farm dams being so prevalent and with the potential for negative consequences for communities and environments downstream, in many places in the world there is a lack of:

1. Integrated agricultural water management and farm dams safety policy that can account for catchment run off at the appropriate levels, and

2. Understanding of farm dams – we do not know where roughly 15 per cent of farm dams in the world even are, let alone what they all were designed to hold, or whether they are managed appropriately (Malerba, Wright, and Macreadie 2021).

A common dam management behaviour that can compromise farm dams is intentional or negligent water ‘hoarding’ though artificial or natural spillway blocking.³ A spillway is a design feature of most farm dams.⁴ A blocked spillway wields the same environmental impact as less rain whilst also making dams unsafe (Pisaniello and Tingey-Holyoak 2017; Tingey-Holyoak, Burritt, and Pisaniello 2013). Yet, many farm dams are often considered too small to be of concern individually. Even though recent assessments indicate that farm dams make a disproportionately large adverse contribution to downstream biodiversity (Biggs, Fumetti, and Kelly-Quinn 2017; Hill et al. 2021) and community safety (Pisaniello 2016), policies have focused on larger dams, in-channel streams and major extractions (Morden et al. 2022; Pinto and Canno Ferreira Fais 2022).

The purpose of this article is to advance understanding of farm dams in catchments, the way they are managed, and current tools for accounting for them under current policy contexts, whilst also providing preliminary ways forward through accounting that could help support improved awareness of catchment runoff storage. Three research questions (RQs) are addressed:

• RQ1: What is the farm dam management setting in Australia?

• RQ2: What are farmers’ behaviours and perceptions relating to their catchment runoff storage management and accounting?

• RQ3: How can a preliminary runoff storage accounting framework be developed for farm dams to support effective integrated water storage policy?

The next section presents context for the study including dam management policy and accounting settings (RQ1), followed by methods and then results from two South Australian farmer cohort perception surveys conducted in 2012 and 2018 (RQ2), before proposing a potential runoff accounting framework for farm dams (RQ3). A discussion follows and conclusions made on the main themes covered.

Research context and significance

This section details the area of study, outlining context from the Australian farm dam policy environment, in addition to review of tools and methods from the hydrology and accounting disciplines. Finally, common challenges found with farm dam management at the farm level are considered.

Australian water storage policy context

Australia’s surface water management has led to huge over-allocation and use of water (Syme 2014; Young and McColl 2009; Baldwin et al. 2017). The 1985 review of Australia’s water resources was followed by the Council of Australian Governments (COAG) 1994 competition policy reforms,⁵ and led to the National Water Initiative (2004) and a push for improved water allocation policy for agriculture and ecological sustainability, including pricing water and return on assets calculations (Carroll 2010; Stoate et al. 2009). The Water Act of 2007 then made the Bureau of Meteorology responsible for developing a National Water Account and each state took the opportunity to revise their water laws to incorporate some form of water sharing regulations and a more holistic approach. The Water Act 2007 does not deal with overland flows and so problems with a lack of runoff sharing in catchments are left to the states (Baldwin, Ross, and Carter 2015).

In the states, regulations that successfully consider both safety and sharing issues usually only target larger dams with a considerable ‘stand alone’ hazard⁶ that hold significant water back in the catchment system. Many standalone existing farm dams deemed not holding enough water to create equity or safety risk to downstream communities and environments continue to be ignored⁷ (Pisaniello and Tingey-Holyoak 2017; Pisaniello, Tingey-Holyoak, and Burritt 2012). However, any dam regardless of size can incrementally contribute to the cumulative equity and safety threats downstream within a catchment if not designed and managed properly (Pisaniello 2010; 2016). The issue is whether the potential increased cumulative risk downstream is acceptable if most farm dams are left unaccounted or unsupervised, and past studies suggest this is not likely (Pisaniello and Tingey-Holyoak 2017).⁸⁹

Four Australian states represent the diversity in policy settings (see Pisaniello and Tingey-Holyoak 2017; Pisaniello, Tingey-Holyoak, and Burritt 2012).

1. New South Wales is the only state to have a specific Dams Safety Act 2015 and has a modern, community risk management and cost–benefit analysis approach (Pearse et al. 2020). Yet there is no integration with the Water Management Act 2000 (NSW) that considers fair water sharing with communities and environments. For Harvestable Rights dams, in some areas, only 10 per cent of the average annual regional rainfall runoff may be captured in dams. Metering is sporadic although progressing (NSW 2022), which does not assist with production of reliable accounts (Tingey-Holyoak 2014).

2. Victoria through its amended Water Act 1989 requires operating licences with enforceable conditions for water allocations and storage safety – but only imposed upon potentially hazardous dams. These include those with a wall height greater than 5 m and at least 50 ML storage capacity or those with a wall height greater than 10 m and at least 20 ML capacity, so larger dams.

3. Tasmania has a Water Management Act 1999 which fully integrates water sharing with dam safety state-wide through the Water Management (Safety of Dams) Regulations 2003 (updated 2015), for all dams including those as small as 1 ML.

4. South Australia has no dam safety laws. Water allocations laws are strong for dams over 5 ML in most planning areas, almost 90 per cent of dams in high dam concentration regions are of 5 ML or less capacity (Savadamuthu 2007). There is no mechanism for state or local government to ensure owners of dams maintain their dams and neither the Development Act nor the Landscapes SA Act 2019 currently provide an effective enforceable ongoing obligation to ensure all dams are properly maintained. However, in response to recent dam failures, the Department for Environment and Water has released an education and information campaign (DEW 2023) and Guidelines for how to manage a dam during a failure event (DEW 2018).

To investigate examples of how dam management is undertaken in the context of each policy environment, research has investigated practices in the four jurisdictions presented above. On-site data was captured by Pisaniello and Tingey-Holyoak (2017) involving 20 physical dam measurements in the four states so that flood risk and the amount of potentially extra water¹⁰ stored by blocked or under-designed spillways could be calculated (Table 1).

Table 1. Dam survey results synthesised from Pisaniello and Tingey-Holyoak (2017).

		South Australia	Victoria	New South Wales	Tasmania	Average/ total
% of dams at hazard rating	%HIGH	65	25	40	40	43
	%SIG	30	55	60	40	36
	%LOW	5	20	0	20	21
Total capacity of water held (at full supply level)	ML	792.2	1101.2	1094.8	2091.2	5079.4
Dams with flood Capability Acceptability^a	%	10	25	70	50	39
Dams with freeboard Acceptability^b	%	10	15	25	55	26
Dams with spillway Blockage^c	%	75	65	35	15	48
Dams with acceptability for ALL key spillway design/safety aspects	%	0	10	15	70	24
Dams storing extra runoff	%	90	85	75	45	74
Average extra runoff stored per dam	%	27	25	25	22	25
Total capacity of dams storing extra runoff	ML	726.4	859.3	674.5	813.1	3073.3
Total extra potential storage created^d	ML(%)	145.8 (20)	145.6 (17)	119.3 (18)	144.2 (18)	554.9 (18)

^a per ANCOLD guidelines = Acceptable = A, Unacceptable = U.

^b Need min. of 1.0 m or 1.1 m for cases with long reservoir fetch per Lewis (2002) and Nelson (1997) Acceptable = A, Unacceptable = U.

^c Yes or no, if yes artificial and/or natural and if artificial, type of blockage.

^d Extra storage from any artificial spillway blockage and/or under-designed spillway freeboard – this identified those ‘dams storing extra runoff’.

Results showed that across the four states 48 per cent of dams had their spillways blocked. There were 74 per cent of dams storing extra runoff from the catchment, with on average 25 per cent greater than their capacity, and these dams held a total of 3073.3 ML of water of which 18 per cent (554.9 ML) extra potential storage was created. For South Australia (which has no dam safety policy and only some accounting/allocations policy) only 25 per cent of dams had spillways unblocked and overall no dams that were fully safe, with results clearly demonstrating a trend concerning the number of dams found to be potentially storing extra water: South Australia 18/20 dams (90 per cent); Victoria 17/20 dams (85 per cent); New South Wales 15/20 (75 per cent) and Tasmania 9/20 (45 per cent).

The results demonstrate that on-site assessments of water stored and management behaviours are variable dependent on strength of dam management policy. If there was improved accounting of the increased storage at the policy level, then this could help with assisting dam decision-making at the farmer level to support the health and safety of environments and communities. The following section will consider the tools available to undertake farm dam accounting.

Tools available for farm dam accounting

Under these policy approaches, many water accounting tools can be refined to consider runoff interception by farm dams, for example: Standardised Water Accounting (SWA) (Hazelton 2013); System of Environmental-Economic Accounts for Water (SEEA-Water) (UNSD 2018); Integrated Water Resources Management (IWRM) (EC 2000); water footprinting and labelling (Hazelton 2013, 2015); and General Purpose Water Accounting (GPWA) (Chalmers, Godfrey, and Lynch 2012; Tello and Hazelton 2018). Australia initially adopted SEEA in 2013 resulting in the National Water Account and State Accounts employed by the Bureau of Meteorology, some government departments, industry associations and the National Water Commission. This has helped allocation and planning at the policy level but water accounting errors (i.e. ignoring much use, including surface and overland flow connectivity and return flows), mean there is limited awareness about how much water is really available for users and the environment. Australia also led the way in developing standards based on financial accounting for producing General Purpose Water Accounting Reports (GPWARs) (Chalmers, Godfrey, and Lynch 2012; Tingey-Holyoak and Pisaniello 2019). These standards are utilised at the basin scale in the Murray-Darling Basin and use volumetric information on both a physical flow basis (in the statement of water flows) and an accrual basis (in the statement of water assets and water liabilities, and the statement of changes in water assets and water liabilities). However, these too fail to consider return flows¹¹ (Seidl, Wheeler, and Zuo 2020), and so remain at a ‘gross’ level of account and only at the public-level scale (Tingey-Holyoak and Pisaniello 2019).

Better accounting for farm dams has been on the minds of policymakers recently for a number of reasons. Firstly, investment in farm-level water use efficiency infrastructure, combined with water markets increasing efficiency faster than anyone thought possible, has resulted in dramatically increased water use and storage (Grafton et al. 2018; Wheeler et al. 2020). Secondly, there has been recognition of farm dams as a significant source of greenhouse gas to the atmosphere, ranking among the highest emitters per unit area among freshwater systems (Grinham et al. 2018; Ollivier et al. 2019). Subsequently, the IPCC have recommended farm dam accounting be included in national carbon inventories. Development of better runoff accounting frameworks and models for farm dams that can not only track all dams (Ausdams 2022), but also consider the local-level changes in storage that changed behaviours in regions can create, is now a pressing issue, beyond just water risk and security.

Knowledge of the hydrology on farms is limited and an accurate account of the volume of water stored in small dams is complex and challenging (Malerba, Wright, and Macreadie 2021, 2022; Srikanthan, Barua, and Hafeez 2015). Because farm dams are usually small in size with a wide distribution, it is only recently that the extent of farm dam development has begun to be systematically assessed. Farm dam water can be accounted for by department staff embarking on a bathymetric survey of each location, but this is time-consuming and costly (Hafeez et al. 2007). More realistic approaches are through aerial photography, satellite imagery and modelling (Lowe, Nathan, and Morden 2005; Malerba, Wright, and Macreadie 2021), and more recently these methods in combination (Fuentes, Scalzo, and Vervoort 2021).

Runoff accounting models and estimates

Moderately complex water balance algorithms have served to develop operational water balance models to estimate volumes of water stored in farm dams (Nathan and Lowe 2012), e.g. Tool for Estimating Dam Impacts (TEDI model) (SKM 2010), CHEAT model (Nathan, Jordan, and Morden 2005) and WaterCRESS (Clark et al. 2015). SKM was later refined to the Spatial Tool for Estimating Dam Impacts (STEDI model), a Windows-based computer program for simulating the impact of farm dams on streamflows within one basin (SKM 2012). The stock-and-flow dynamic model/Water Accounting System (WAS) is another innovative strategy for long-term water management (Turner, Baynes, and McInnis 2010). More recently, Australia’s BoM developed the operational daily water balance model, Australian Water Resources Assessment Modelling System (AWRAMS). This is a hydrological simulation system designed for water accounting (Elmahdi et al. 2015) which comprises a farm dam model (FDM), similar to STEDI and is used in the National Water Account (NWA). This is limited because all farm dams are assumed to be directly connected to the catchment outlet and the spatial variation is ignored (Srikanthan, Barua, and Hafeez 2015). The WaterSENSE project is producing strong results using Sentinel 2 satellites and advancing big data processing algorithms, however is limited by vegetation in dams and is being improved to link to whole of farm water use (Vervoort et al. 2022).

Whilst above farm level modelled dam account models and estimates are improving,¹² such accounting systems fail if all dams are not captured and properly documented, and secondly, if the complex interactions of weather, agricultural development and management practices are not accounted for (FAO 2017; Morden 2017; Nikkels et al. 2019). These estimates and models ultimately require validation via on-farm observation and calculated depth–area–volume relationships for accurate volume estimation (Chemin and Rabbani 2011; Malerba, Wright, and Macreadie 2021). As such, reliable information about the actual measured volume of farm dams in a catchment is extremely rare in any jurisdiction. The level of effort required to obtain the data is excessive and even with advancing neural networks and satellites in addition to building dataset/mapping resources (Ausdams 2022), there is a gap at the farmer level where dam management behaviours and storage can change within and between seasons and regions.

On-farm dam accounting methods

On-farm dam accounting can start simply with a dam inspection checklist that follows current international guidelines for small earthen dam review and inspection (see Pisaniello, Tingey-Holyoak, and Burritt 2012, 2015). In digital form, the use of individual cells as the ‘checklist’ can assist to ensure that the critical determinants for a specific behavioural intervention can be considered prior to implementing an intervention and can also help assess the results of an intervention (or the failure to intervene) (for example, Dreibelbis et al. 2013). Digital water management checklists can also serve as a baseline for additional technological, psychosocial and contextual data that can shape specific behavioural outcomes.

Further to checklists, simple and cost-effective regionalised farm dam storage and flood capacity review/design accounting tools exist.¹³¹⁴ Individual data can contribute to catchment scale accounts of water storage and safety. Such a tool can function in design mode, where the simple on-site input parameters can be varied selectively to satisfy flood capability, storage capacity and other practical on-site factors, e.g. a farmer’s minimum storage requirements for irrigation (Gaydon, Meinke, and Rodriguez 2012), any maximum storage requirements of water sharing regulations, etc. It is also applicable in review mode for an accountant or regulator to audit farmers’ assessments of dams, serving to protect the landholder and downstream public against both: (1) individual water hoarding and/or farm dam failures; and (2) cumulative/cascade water shortages or excesses within larger catchments (Pisaniello, Tingey-Holyoak, and Burritt 2012).

Sensing tools are also advancing at the farm level, that can monitor storage levels in dams and spillways in real time. These include IoT low cost flood inundation sensors, low pressure transducers to monitor pool levels, and soil tensiometers installed to measure strain on earthen dams (Johnson and Osborne 2020). In the US, Kentucky has a project using sensors on hazardous farm dams that are remotely located to feed data into a centralised data management platform to increase data collection and awareness to build community resilience to individual and cumulatively reduce flood failure (Johnson and Osborne 2020). In the UK, a new sensor project funded by the Department for Environment, Food and Rural Affairs worth $US 44 million was established in 2022 to create a digitally enabled research infrastructure for better awareness of farm dams and other water storage to reduce drought and flood risk – considering both sides of the dual extreme (Preventionweb 2021).

Incentivising on-farm level dam accounting methods through links to farm accounting

Is it possible to account for runoff storage in another way and incentivise farmers measuring their own dams? When considering on-farm solutions and tools, ways to approach are at the policy level where investment pressure is on states to implement and support, or at the farmer level, couched in information and awareness raising or as part of accounting to incentivise improved dam management to minimise business risk (Greiner and Gregg 2011). Whilst not all farm dams are owned by businesses, research has found that a minimum of 70 per cent of farm dams are used for irrigation or stock watering for food production (Lewis 2002; Tingey-Holyoak 2014). For water accounting, tools are emerging that link soil sensing systems to farm accounting systems (e.g. MYOB, Xero etc.) to better incentivise monitoring and improve water-related decision-making for profitability (Tingey-Holyoak et al. 2020, 2021). Whilst improved farm dam management might not improve profitability, it does reduce business risk and potential liability, which is one of the key functions of farm accounting.

Key concepts from the management and financial accounting disciplines can allow for the water flows and stocks and water risk to the business to be understood in economic terms. Whilst full financial accounting rules and principles can be difficult and costly for farmers to implement (Argilés and Slof 2001; Seidl, Wheeler, and Zuo 2020), there are elements of accounting that can be simply linked to natural resource and dam decision-making, starting with management accounting (Phan, Baird, and Su 2017). Table 2 demonstrates on the left-hand side, the management accounting concept of dam planning, which considers water-related objectives across operational, tactical and strategic levels (Argilés and Slof 2001; Puig-Junoy and Argiles 2004). In the centre column, controlling considers budgeting for dam-related costs analysing results (Tomaszewski et al. 2000). Finally, in the right-hand column, dam-related decision-making based on both planning and controlling, demonstrates possible dam-related decisions from meeting daily watering objectives to longer term decisions such as dam removal and investment (Savić, Vasiljević, and Đorđević 2014) (Table 2).

Table 2. Management accounting concepts to support run-off accounting.

	Dam planning	Dam controlling	Example dam-related decision-making
Operational	Checking daily dam and spillway levels in line with objectives (e.g. stock watering, inundation minimisation) to gauge short-term risk.	Using checklist + sensor data fed into simple tool to derive a score that could be simply linked to likelihood of a budget item for repair, maintenance, loss, or liability etc.	Do I have enough water in my dam to achieve my farming objectives? Is my spillway functioning as needed? What are the risks created by my dam? When should I run checks and who has time and is least costly to do this?
Tactical	Using checks and measurements to manage medium term risk including considering opportunities for integrity checks, repair and vegetation clearing.	Link to budgeted items for repair, maintenance, loss, or liability etc., and analyse impacts to balance sheet and profit and loss statement.	How much insurance do I need? When should I plan dam repairs? How much will repairs cost? When and how do I clear vegetation?
Strategic	Using sensing data for objective setting/risk management over longer term, such as investment in hydropower or removal of old dam.	Using longer-term sensed data to forecast impacts to on farm infrastructure.	When should I build a new dam? When should I remove old dam? What else can my dam provide me with (e.g. energy)?

These aspects of management accounting interrelate with financial accounting as the decision-making column on the right-hand side will have direct impacts on the financial statements:

• balance sheet (Statement of Financial Position) impacts could include changes to asset valuation of the dam and surrounding infrastructure on farm, increased asset impairment (carrying cost-value), and increased continent liabilities (reliable estimate of probable loss caused by a dam), and

• profit and loss (Statement of Financial Performance) impacts include possible increased insurance expense, repairs and maintenance expense, fees for legal advice or advice from other professionals such as engineers, labour expense for repairing or maintaining dam, and possible increased impairment expense.

To create further accountability for these decisions and facilitate alignment with the needs of policymakers, financial accounting principles are available for adaptation to water management to guide rigorous and comparable basin-level runoff accounting including:

• nesting accounts within a farm for different field uses, sources and dams, and consolidating accounts and the elimination of inter-entity transactions, isolating links to other entities, reclassify data so water transfers within a basin between users and catchments are netted out and data adjusted to get an overall picture of the system’s health (Tello and Hazelton 2009, 2018),

• two reports/statement types that at a minimum embody financial accounting, ensuring there is an account of stocks (assets and liabilities) and flows (use), i.e. catchment capacity and direct monitoring of inputs to and outputs from them based on actual measurements of individual businesses/dams (Hazelton 2013, 2015; Tingey-Holyoak and Pisaniello 2019),

• double entry accounting concepts that reflect sensitivity to use and storage types, so that when non-entitled use increases (e.g. through farm dam accounting) then entitlements can be capped elsewhere (Chalmers, Godfrey, and Lynch 2012; Tello and Hazelton 2009, 2018), and finally

• auditing water information and a low-level monitoring (group/self, see Gunningham 2002) based on financial accounting concepts linking accounting and governance (Chalmers, Godfrey, and Lynch 2012).

The concepts presented in this section demonstrate that there is significant work being done toward better accounting for farm dams and the potential for integration with farm accounting principles. However, integration of tools and principles would require policymakers to appreciate the relevant and connected issues caused by surface water storage in on-farm catchment dams and better understand their impact. Carroll (2010), Graymore and McBride (2013) and others suggest that policymakers require additional research evidence about how farmers behave when managing their water and who they affect through their decisions, in addition to links to financial impacts of dams on a catchment’s health. This article proceeds to present research evidence about the perceptions of farmers related to their dams, in addition to developing a possible catchment runoff storage accounting solution that integrates physical and financial data. The following section details the methods used to achieve this.

Research method

The method comprises longitudinal South Australian farmer cohort surveys at two points in time. A questionnaire was originally developed in 2012 to ascertain farmers’ perceptions of dam safety management, including issues relating to accounting for water. This was then followed-up on and advanced in 2018. The study was conducted in the researchers’ home state of South Australia where farm dams represent one third of water stored in the state (Malerba, Wright, and Macreadie 2021) and there are many catchments where the dual potential for unfair sharing and unsafe storage of water exists (Pisaniello and Tingey-Holyoak 2017). The total population of South Australian farms was divided into strata through sampling of records in the Yellow Pages Online. Industry and regional strata were derived based on those operating in: (1) in industries known to require farm dams (e.g. irrigated agriculture, dairy, etc.); (2) regions of undulating topography typical of catchment dams where the potential threats and risks of unfair sharing/unsafe dams can exist. Online mapping technology gave the sampling more precision by confirming the existence of the catchment dams for each target in the frame from which random samples were drawn. Computer-assisted telephone interviews (CATI)¹⁵ were undertaken until the sampling frame was exhausted and an adequate response rate achieved. Survey interviews took on average 20 min each and no incentive was offered to participants. When surveys were then repeated in 2018, the same population was sampled as in 2012 but with additional questions related to accounting and business. In both surveys a 7-point Likert scale was used to optimise item reliability through maximising the level of measurement, with Likert items becoming significantly less accurate when the number of scale points drops below 5 (or above 7) (Colman, Norris, and Preston 1997; Neuman 2014).

Study context: weather and policy setting changes from 2012 to 2018

Table 3 summarises the changes in context during the study period from 2012 to 2018.

Table 3. Study context in South Australia 2012–2018.

Year	Weather	Policy
2012	Driest since 2008	Consultation on new Water Allocation Plans
2013	Record high temps	Plans implemented and altered including permit required for new dams and metering for some dams over 5 ML in several regions; ‘Securing Low Flows’ for environmental benefit below farm dams Project implemented in Mount Lofty region
2014	High temps, rainfall scarce from mid-year
2015	Dry, high temps
2016	Heavy rain events	Dam Safety regulation white paper developed; Resilient Hills and Coasts climate adaption plan considering environmental water and dams
2017	High rainfall
2018	High rainfall in some regions, dry in others	Most new Water Allocation Plans finalised; Kangaroo Island private dam maintenance and management emergency guidelines developed

During the study period, the weather ranged from dry and hot to heavy rains, creating a dual extreme setting for farm dams. The year 2012 started cool and wet for South Australia but then proceeded with higher than average day time temperatures meaning that water storage was still at a premium. Rainfall for South Australia was 77 per cent of average, the driest year since 2006 (BoM 2012). Some areas experienced very much below average rainfall meaning pressure on securing dam storage. In industries such as cropping, rainfall across the districts was 82 per cent of average, the driest since 2008. This was followed by a series of hot years with 2013 starting with very hot conditions across South Australia and the hottest year on record, with persistent extreme heat across large parts of the state for much of January (BoM 2013). Record high temperatures were observed with many areas experiencing extensive and prolonged heat events and yet total annual rainfall across the state tended to be near average. This is before rainfall became scarce through the second half of the year from August, with monthly totals across the state tending below average (BoM 2013, 2014, 2015). High temperatures and a need to water stock and crops can create pressure on dams and when combined with subsequent heavy rainfall. This results in the dual extreme situation of dams that are storing extra for thirsty crops and stock, can then face structural safety risks from sudden rains. In 2016, several heavy rain events brought flooding to parts of South Australia during the year yet temperatures were still above average overall (BoM 2016). Rainfall during 2017 and 2018 was above average for South Australia as a whole, although it was drier than average in some central and eastern areas (BoM 2017, 2018).

During the study period, water management policy was in a state of change. In 2012/2013 the SA Government adopted new water allocation plans for the Eastern and Western Mount Lofty Ranges, amalgamated plans for Southern Basins and Musgrave, and amalgamated a number of plans in the South East into one Lower Limestone Coast Plan. The government also adopted Water Affecting Activity policies for Kangaroo Island, Alinytjara Wilurara, Eyre Peninsula and SA Arid Lands regions. The sampling occurred in all high concentration dam areas in SA which included these areas, and whilst consultation on these plans occurred started in 2010, it continued into the survey period of 2012. There was discord reported in the media over some of the plans as they pertained to small farm dams (ABC 2011).

One region where there was significant community outrage was in the Adelaide and Mount Lofty Ranges Natural Resource Management (AMLR NRM) region which has the most complex landscape and greatest biodiversity of South Australia’s NRM regions (GIWR 2015). Out of South Australia’s 30,000 privately owned dams, 15,300 dams are located in the AMLR region capturing an average of 10 per cent of the annual surface water flow, with up to 70 per cent in some catchments (AMLRNRMB 2014). The AMLRNRMB plan and Water Allocation Plans (WAP) for prescribed surface water areas set out the criteria by which any application for a farm dam would be assessed. The ‘Securing Low Flows Project’ was a key part of the 2013 WAP and eleven trial sites were established to investigate practical ways to pass low flows around farm dams in the Eastern and Western Mount Lofty Ranges to provide water for the environment. The project aimed to maintain health of the 74 catchments across the Mount Lofty Ranges (SAMDBNRM 2018). Not long after in 2016 a white paper was commissioned concerning farm dam safety focused on education and awareness of the risks posed by dams (Environment SA 2019). The result was heightened attention on farm dams and fears of increased regulation, water for environmental needs and costs to repair or modify dams.

Another farm dam-dense region with specific changes during the study period was Kangaroo Island (KI). There are over 8590 dams on the 4405 km² island that assist in fulfilling the demand for water from residential, tourism, agriculture and industrial uses (DEWNR 2013; McMurray 2007). Through its council, KI established private dam maintenance and management emergency guidelines due to the large number of dams in a small region and in response to the 2009 failure of a 600 ML private farm dam (KINRMB 2018a). Also during the study period, a new KI NRM Plan 2017–2027 was developed (KINRMB 2017) and the KI NRM Board and the KI Council developed a regional climate change adaptation plan for the Adelaide Hills, Fleurieu Peninsula and KI Region (KINRMB 2018b; Resilient Hills and Coasts 2016).

The next section details the methods for developing the preliminary runoff storage accounting framework.

Development of a preliminary runoff storage accounting framework

The results from review of (1) available runoff accounting models and estimates, (2) on-farm dam accounting methods and (3) management and financial accounting principles that can incentivise tool uptake and interconnectedness, were integrated to develop a preliminary framework. This was done at a conceptual level using simple visual software. The framework was worked on in two stages. Firstly, the focus was placed on farm-level dam accounting methods and incentivisation through farm accounting for an individual dam. The second phase was representing the potential for integration of farm-level incentivised accounting with catchment-level data, and broader basin-level decision-making.

The following section discusses the results of the 2012 and 2018 farmer perceptions surveys in addition to presenting development of the preliminary runoff storage accounting framework.

Results

Farmer perceptions

The farmer perceptions surveys were undertaken in 2012 and 2018, principally on water management behaviour in response to regional, environmental, regulatory and funding/financial pressures (see Tingey-Holyoak, Burritt, and Pisaniello 2013). Final retained full survey data for the 2012 dataset (n = 100) and the 2018 dataset (n = 154) were compared with population data (current Census data, Australian Bureau of Statistics (ABS) 2012, 2018) to determine representativeness. The data was representative in terms of the type of farming undertaken in the area, farm size and farmer demographics. Data analysis consisted of descriptive statistics undertaken in SPSS v24 and a paired sample t-test was undertaken to compare the difference between the means of the two related samples from 2012 and 2018. Results show that in 2012 only five per cent of farmers believed that others in their region manage their dams in the same way they do and most did not believe that farmers around them share the water fairly (Table 4).

Table 4. Comparison of South Australian farmer perceptions from 2012 to 2018 and additional accounting elements considered.

% Agree	SA 2012 %^a (n = 100)	SA 2018 % (n = 154)	% change	t- statistic
Water management behaviours in response to regional pressures
I follow the dam management approach used by other successful farmers in the area	21	51	↑30	−5.05***
Most of the farmers in my region manage their dams in the same way	5	62	↑57	−10.92***
The farmers around me share the water fairly	5	76	↑71	−15.30***
Water management behaviours in response to environmental pressures
The laws require me to provide too much water for the environment	56	21	↓35	6.16***
I am worried about the future of water availability to continue my farming operations	37	61	↑24	−3.84
Water management behaviours in response to regulatory pressures
I make a conscious evaluation of specific dam laws and choose to comply with them	17	64	↑47	−8.31***
I attempt to form an alliance with the regulators regarding my dam management	61	33	↓28	4.53
I negotiate openly with the regulators to obtain a mutually agreeable solution on the management of my dam	48	30	↓18	2.96***
I appear to comply with dam management laws but intentionally avoid certain aspects of the requirements	62	23	↓39	6.82***
I meet with elected officials to attempt to control the regulations on dam management issues	60	24	↓36	6.16***
I would challenge dam management requirements in court	54	28	↓26	4.31***
I am fearful of enforcement of dam management laws by regulators	27	72	↑45	−7.83
Links to farm accounting and business
I allow for a dam maintenance spend in my annual budget	33	45	↑12	−1.88***
I depend on government funding so that is why I comply with what regulators want for my dams	82	79	↓3	0.54
In the last 5 years this farm has had a positive productivity change		53
A maximum annual return from my property is my most important aim		49
Compared to other farmers in my region, my growth has been positive		42
I include water and dam management in my overall farm planning		74

^a *p-value < 0.05.**p-value < 0.01.***p-value < 0.001.

In 2012 over one third of farmers also had concerns about the future of water availability for farming (37 per cent) and many believed they must give up too much water for the environment (56 per cent, Table 4). Nearly two thirds of farmers attempt to form an alliance with regulators (61 per cent) and about half would negotiate openly (48 per cent). Many farmers agreed they attempt to control the regulations (60 per cent) or will avoid certain aspects (62 per cent), and over half would challenge the requirements in court (54 per cent). Many farmers reported being dependent on government funding or subsidies in some way (82 per cent) but only a third allowed for their dam management in their annual budget (33 per cent).

In the six years between the surveys in South Australia, farmers’ perceptions changed as did the context of their dam management. The results indicate several significant differences in perceptions between time periods. With stressful weather events and new Water Allocation planning and consultation there was a dramatic improvement in opinions on the subject of fair sharing between farmers with 76 per cent agreeing those around them share fairly compared to five per cent in 2012 (t(252) = −15.30, p = 0.001). But there was rising concern about the future of water availability which is to be expected amidst weather changes and general perception changes in the community, albeit this was not a statistically significant change (from 37 to 61 per cent: t(252) = −3.84, p = 0.53). Amidst renewed consciousness of environmental needs, farmers improved their impression of water sharing with the environment with only 21 per cent perceiving the environmental allocation was too high, a statistically significant decrease from 56 per cent in 2012 (t(252) = 6.16, p = 0.001). Fears rose about the enforcement of dam management laws by regulators (up 45 to 72 per cent) in line with new water allocations policy, consultation and revised planning in South Australia and interestingly, there was a significant decrease in the desire to try to influence (60 per cent in 2012 to 24 per cent in 2018, t(252) = 6.16, p = 0.001) or challenge the requirements (54 per cent in 2012 to 28 per cent in 2018, t(252) = 4.31, p = 0.001).

The findings show that since the first survey, there has been a statistically significant increase in the allowance for dam management expenditure in annual budgets (33 per cent in 2012 to 45 per cent in 2018, t(252) = −1.88, p = 0.001). Many of the farmers surveyed are increasingly focused on their annual return (49 per cent) and many are experiencing growth (42 per cent). Many have experienced a positive productivity change (53 per cent) and nearly three-quarters include water and dam management in their overall farm planning (74 per cent). This link to farm accounting and farm business is important to explore any consideration toward incentivised and integrated runoff storage accounting.

Development of a preliminary runoff storage accounting framework

In this phase, available runoff accounting models, on-farm dam accounting methods, and management and financial accounting principles were integrated to develop a preliminary framework, starting with farmer level dam accounting methods and incentivisation for an individual dam. On-farm, the dam accounting methods and incentivisation strategies through business and accounting aspects can be operationalised using the simple digital checklist and accounting software mapped together to create a risk score. The checklist captures baseline or changes to aspects of the dam including issues such as reservoir slides, downstream slope cracking or blocked spillway. This is linked directly with accounting software to capture data that can equate the likelihood of increased repairs and maintenance, or loss or potential liability based on the checklist assessment. This would financially quantify potential for increased expenses, contingent liabilities, liabilities and risk, with additional facility for tax considerations to be incorporated, such as increases in allowable deductions based on dam-related expenditure. The IoT low-cost flood inundation sensors are a further integration to facilitate awareness of when a height of reservoir has gone beyond capacity triggering a further risk warning alert.

Figure 1 demonstrates this possibility for an example dam in South Australia.

Figure 1. Operationalisation of farmer-level incentivised farm dam accounting for example dam.

The full development of the framework is through integration of the farm-level accounting with current catchment estimation methods and broader basin-level decision-making as is demonstrated in Figure 2.

Figure 2. A preliminary runoff storage accounting framework.

Figure 2 provides an illustration of the types of farm-level tools that can be extrapolated to the catchment, and ultimately, basin levels. Farm-level incentivised accounting seeks to provide an account of additional water held back from the catchment, plus documentation of the potential risks to safety and changes in them over time. The farmer-level physical checklist and sensor data could be accessible as a report for dam management regulators and be able to be utilised/integrated with other runoff accounting models and estimates and emerging resources at the catchment level (e.g. Ausdams 2022). These farm dam accounts nested at the catchment scale are designed to provide validated data on the estimate of stores in dams between and within seasons in any management scenario. On the right-hand side of the figure, basin-level accounts can provide a whole-of-dam picture of higher reliability than what can be captured from imagery and modelling alone. Reliance on high level data without the sensitivity of the left-hand side of the framework, means that behavioural aspects related to dam risk management are missed. These consolidated basin-level accounts can then be independently audited and ongoing monitoring implemented through a feedback loop.

The following section discusses the results.

Discussion

The first research question asks: What is the farm dam management setting in Australia? Results show the integrated dam management setting in Australia is variable, with states taking different approaches to integration and most only considering relatively large dams. Tasmania reflects best current practice with fully integrated dam water sharing and safety policy for dams as small as 1 ML. The practice on farm in the states discussed is in line with the ‘strength’ of the policy integration and coverage. We do not suggest that there is a casual relationship but rather that there is a trend that warrants further consideration and investigation.

The second research question asks: What are farmers’ behaviours and perceptions relating to their catchment runoff storage management and accounting? Surveys from 2012 and 2018 showed that South Australian farmer perceptions changed considerably with heightened concerns about future water availability and how much surface water the government allocates to the environment. These results were in the context of several extreme weather in addition to consolation on new policy and trials on farm dam affecting initiatives such as low flow bypasses, which is an important lens through which to view these results. During the same time, fear of the regulators was also increased which is perhaps to be expected at times of water policy change. However, this is important to note given the placement of trust in regulators as principally important for water storage accounting success (PC 2020). Further exploration of accounting and business finds positive productivity changes and increased focus on annual return, including budgeting for dam management and most farmers including water in overall farm planning. This confirms other research in this area (Tingey-Holyoak, Burritt, and Pisaniello 2013, 2020, 2021) that suggests water management that effectively links to farm business activities is crucial for wide-spread take up of farm-level water accounting initiatives. Furthermore, it is necessary to have reliable data on how and why individual farmers store their water that is not subject to sampling and model specification errors (Carroll 2010).

The third research question asks: How can a preliminary runoff storage accounting framework be developed for farm dams to support effective integrated water storage policy? Data from farmer surveys was analysed for comparison with context. This was then linked back to key concepts from hydrology and accounting to provide the integration of available hydrology and accounting tools to address the needs identified in the farmer surveys. Results show that at the farm-level, useful runoff storage accounting tools are lacking, yet there is capacity to achieve this through budgeting and planning elements that could be better integrated for water and capitalised upon to feed into basin-wide initiatives. A framework is proposed that uses concepts from hydrological and accounting tools identified, linked to the results of farmers’ behaviours and perceptions.

We propose that development and use of a preliminary runoff storage accounting framework would mean that accounting for water stored in dams is incentivised at the farmer level, but is also scalable to the basin level. This would help toward awareness of the many high hazard dams being managed unsafely and create the ability to generate a more reliable statement of water position (stocks in dams) to assist policymakers, communities and environments (Baldwin et al., 2017). The impact of improved farm dam water accounts would not only be on planning for safety and water sharing but would also support better understanding of where the farm dams are and identify hotspots for methane¹⁶ emissions. Australia is committed to substantial reduction of these by 2050 and so better accounting for runoff stored in dams and where could help multiple areas of government (Malerba, Wright, and Macreadie 2021; Ollivier et al. 2019).

It is important to note that monitoring of farms is not systematic at this stage and is subject to the priorities of the department, some dams are ‘..monitored continually over decades …' .other sites are monitored over short to medium periods, and these are designed to meet the needs of a particular management action, business operation, regulatory approval condition or project (Environment SA 2021). Indeed, it is noted by Malerba, Wright, and Macreadie (2021) that there are approximately a quarter of a million farm dams completely undetected and monitored in Australia, with a high proportion of these in South Australia. So as communities continue to develop in catchments downstream of dams due to growing populations seeking housing, working upward toward a basin-level water account is required, as demonstrated on the right-hand side of Figure 2. This will account for community and environment-wide safety risks and deprivation in an integrated way. Farmers’ concerns about future water suggest that climate change issues are pervasive and encouraging some water management behaviours, such as hoarding through spillway blocking. It is therefore necessary to feed back to catchment and farm-level accounts through monitoring, preferably in ways that do not exacerbate enforcement fears, but instead capitalise on community connections, such as group–self monitoring. However, without links between on-farm accounting and efficiency data, and larger scale agro-meteorological, hydrometric and regional water productivity data, it will be difficult to make meaningful connections between information levels (FAO 2017; Nikkels et al. 2019).

Runoff storage accounting could be the link between (1) the generation of micro field-level accounts based on real farm business data, to (2) hydrological modelling at both the farm dam and catchment scale, generating a better baseline understanding about where all the farm dams really are, how much they are really storing and how safely – a current knowledge gap that is global. Such an undertaking requires linking to sample catchments of micro-level data for integration into basin-wide accounting so that superior auditing and subsequent monitoring is facilitated, as proposed by the framework. The linking of farm level business data could also influence behaviour of farmers as they will be aware of business risk and opportunities in their proximity created by their farm dams.

Conclusions

The research reported here demonstrates current farm dam policies and water accounting mechanisms in Australia are serving to hinder our understanding of farm dams and in many cases, have resulted in a lack of basic accounting information on farm dam density and location. Physical evidence shows many farmers are storing more water than they are entitled to, leading to unfair sharing of surface water with downstream users and the environment suffering. It also means that downstream communities are placed at risk due to unsafe dams. This is worsened by recent water allocation policy which fails to account for all related issues and factors, for example, return flows. Farmers’ perceptions and behaviours concerning dam management highlight how policy pressures, combined with climate change threats, are exacerbating behaviours like spillway blocking, much of which remains unseen due to gaps in accounting.

The framework proposed here would start to pave the way toward a mechanism that could look after not only the interests of farmers but also the downstream communities and environments against both: (1) individual water hoarding and/or farm dam failures which now occur regularly; and (2) cumulative/cascade water shortages or excesses (i.e. drought or flood exacerbation, respectively). This is in addition to the problem of unknown emissions caused by farm dams and a failure to include them in carbon accounts. The extent to which such scenarios present a real risk to downstream residents, the community and the environment reinforces the need for new approaches to accounting for farm dams, particularly those that are integrated and interdisciplinary. The data and resulting preliminary runoff storage accounting framework can potentially trigger a new way of thinking for Australian and international jurisdictions that have considerable on-farm surface water storage in catchment dams and high inter-annual variation in rainfall. However, further research is needed on how to advance and operationalise such a framework, especially because adequate accounting of surface water stored in farm dams and assurance of farm dam safety requires integration with better basin-level water accounting policies backed by legislation, funding and enforcement.

Acknowledgements

The paper was part of a detailed Australian Research Council Discovery Project (DP0987825) investigation of dam safety in addition to a Department of Human Services/SACOSS funded project investigating water security threats and challenges. The authors would like to acknowledge these bodies and the participants. Thank you also to Ms Kirsty Willis and her team at the Ehrenberg Bass Institute, Mr Arthur Spassis, Dr Mitali Panchal, Dr Elnaz Etihad and Square Holes for support with data collections. The discussants and participants at the 2020 Massachusetts Institute of Technology (MIT) J-WAFs Seminar Series (Boston) and 2022 International Conference on Irrigation and Drainage (Adelaide) are thanked for their helpful comments. Finally, we acknowledge the help of the anonymous reviewers whose comments greatly improved this manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Department of Human Services, Government of South Australia: [Grant Number DSCI1103]; ARC: [Grant Number DP0987825].

Notes

1 The farm dams being referred to are typically embankment dams that largely take water from local overland flows and exist in an area of undulating topography.

2 The landscape drainage area that channels water to a particular point along a network based on its surface topography is called a catchment (Wagener, Wheater, and Gupta 2004). Every part of the land on earth forms part of a catchment and so farms necessarily all reside within catchments. However, the catchments referred to in this article are those with undulating topographies where additional water stored in dams poses a greater risk to downstream communities.

3 A particular on-farm water storage management practice that creates a dual-extreme problem is spillway blocking and/or under-designing. By blocking spillways (either by artificial blockage or purposeful neglect of natural vegetation growth) or by under-designing them, farmers are able to store significant additional amounts of potentially non-entitled water; significant because small increases in storage height (resulting from a shallower spillway) at the top end of the reservoir, which has a triangular prism-like geometry, results in large increases in storage volume (Pisaniello 2016; McMurray 2004). Hence, suitable policy is required to plan, coordinate and manage elements of rural water systems effectively integrated to address equity and safety threats.

4 Most farm dams (which comprise earthen banks) are designed/built with a spillway to prevent overtopping and eroding during flood events. Most dam owners/designers/builders recognise that any farm dam with any catchment area (even if just its own reservoir area) must have a spillway of some kind (e.g. weir, channel, conduit) otherwise it would be highly vulnerable to overtopping, eroding and being washed away as soon as it fills. This was clearly demonstrated by Pisaniello, Tingey-Holyoak, and Burritt (2012, Table 4) where 504 Australian farm dams were randomly surveyed and 91 per cent had spillways.

5 The Council of Australian Governments at its meeting in Hobart on 25 February 1994 agreed to the principles of competition policy with the intent to achieve and maintain consistent and complementary competition laws and policies which will apply to all businesses in Australia regardless of ownership.

6 Hazard in this context refers to a dam’s individual ‘hazard rating’ or ‘consequence category’ as defined by ANCOLD (2012) Guidelines. These Guidelines assist engineers, consultants and departments working with dams to prepare consequence category assessments. Assessment of the hazard /consequence category has evolved since 2000 from a three-tier classification (Low, Significant and High) to a seven-tier system operating today (Very Low, Low, Significant, High C, B, A and Extreme) (ANCOLD 2012).

7 In terms of the dual safety-equity problem posed by farm dams, administration problems exist usually because qualified interdisciplinary personnel and funding are lacking. The sheer number of farm dams to be monitored poses a considerable challenge (Baillie 2008), and benefits are perceived to rarely outweigh the costs of monitoring most small farm dams, unless their failure would put a vulnerable population at risk (KPMG 2013).

8 For example, when numerous farm dams pose a cumulative threat within the catchment of a large hazardous public dam, such as South Australia’s Kangaroo Creek Dam. This is a large, high hazard public dam (65 m high, 19,000 ML capacity) in the River Torrens catchment of South Australia. A flood study found the dam’s peak inflow would increase four-fold assuming all small dams (>1000) in the catchment failed at the same time in a 1-in-200 years flood event: a reasonable assumption as supporting studies found most small dams cannot pass such an event. This additional flow to Kangaroo Creek dam would exceed its spillway capacity, which should otherwise be capable of passing at least a 1-in-10,000 years flood event, putting downstream communities and the environment at unacceptable risk (Pisaniello 2016).

9 From a more localised cascade perspective, the dam upstream of a more significant dam can certainly be of insignificant size in terms of its normal storage capacity posing any threat, but it may have a large catchment and associated large/deep spillway to safely pass the catchment flows, and so as soon as its spillway is blocked, its increased storage capacity that poses a threat to downstream equity and safety can become significant (Pisaniello 2010; Pisaniello and Tingey-Holyoak 2017). As such, original dam size and ‘stand-alone’ hazard are not necessarily the critical determining factors of the severity of dual-extreme cascade/cumulative risks within a catchment (de Lima et al. 2022). Severity depends more on physical catchment circumstances, such as relative locations of dams and downstream communities (ANCOLD 2012), as well as the jurisdictional circumstances that influence dam management practices (Pisaniello, Tingey-Holyoak, and Burritt 2012).

10 Extra water means the water that is stored above the amount the dam is designed to hold safely at full capacity.

11 These reports include volumetric information on both a physical flow basis (in the statement of water flows) and an accrual basis (in the statement of water assets and water liabilities, and the statement of changes in water assets and water liabilities). When these statements are analysed, captured natural inflows ratios and other ratios take consumption into account, but this is only a potential feature of GPWA and only extrapolatable/useful if more preparers conduct the analysis.

12 Recently in the Murray Darling Basin, Fowler et al. (2015) advanced an equation across a larger dataset to get a better understanding of the water held in the region’s 650,000 farm dams which had an accuracy to within 16 per cent of on the ground methods.

13 A full description of the tool is available in Pisaniello, Tingey-Holyoak, and Burritt (2012) and Pisaniello (2016).

14 It is important to note that farm dam storage capacity, spillway flood capability, and in turn cumulative storage and flood capacity are inter-related, i.e. storage capacity depends on reservoir area and height at full supply level, which affects the ‘flood attenuation’ ability of the reservoir, and hence, its spillway flood capability and downstream ‘domino effect’. The extent to which each variable can be altered for a single dam to meet (1) current acceptable standards for both spillway capacity and yield, (2) a farmer's minimum storage needs and even (3) the surrounding landscape, can be determined using Pisaniello, Tingey-Holyoak, and Burritt (2012).

15 The CATI system facilitates stratified probability sampling of the segmentation matrix, centralised data collection, standardised interviewer behaviour, and reduced survey costs (Vemuri et al. 2011).

16 Methane is a greenhouse gas that is 34 times more potent than carbon dioxide (Grinham et al. 2018). Ollivier et al. (2019) measured the GHG emissions from farm dams in south-eastern Australia and found them 3.43 times higher than temperate reservoir emissions. By modelling these results to the entire state of Victoria, the researchers found a farm dam CO₂-equivalent/day emission rate of 4853 tons, 3.1 times higher than state-wide reservoir emissions in spite of farm dams covering only 0.94 times the area comparatively.