Why are there no factors for Floodplain Harvesting?
Why didn’t we calculate factors for floodplain harvesting?
As at August 2018, there were no water entitlements associated with floodplain harvesting and there is no long-term data on this type of use. Therefore, the department is unable to calculate a factor that represents use per unit share at the present time.
To meet the goals of the Basin Plan, the Commonwealth Government buys water entitlements to reduce the long-term diversions within the Basin and meet the water recovery target. When the government buys an entitlement, the LTDLE factor indicates the long-term average volume of water the entitlement represents and that contributes to the recovery goal.
In NSW, a water entitlement doesn’t give the holder access to a fixed volume of water each year. Instead, it gives the holder a right to a share of the available water resources that exist at any point in time.
The entitlement is expressed as a ’unit share’ of the water resource, and water is allocated against each unit share based on a process known as the ‘available water determination’. The determination reflects the amount of water available in the system, based on a range of factors including dam storage levels, river flows and the current catchment conditions. The cap factors describe the long-term average use of these allocations, which is the long-term average volume that the Basin Plan seeks to adjust.
The water entitlements and the respective unit shares are assets that the Commonwealth Environmental Water Holder will own, as would any irrigator.
When will we calculate factors for floodplains?
The Commonwealth-funded Healthy Floodplains program will bring all floodplain harvesting activities within the NSW water licensing framework. One part of the project is to issue water entitlements for flood harvesting take.
The models that determine the floodplain harvesting entitlements will also estimate the long-term average use of the entitlements. They will be issued at the time the entitlements are made and become available for purchase from the Commonwealth.
More information on the Healthy Floodplains project.
Isn’t this whole process, and the Basin Plan more generally, invalid because you have ignored floodplain harvesting?
The Basin Plan and the 2750 gigalitre recovery target both include explicit estimates of some of the long-term average floodplain harvesting diversions. These diversions were included in the NSW hydrologic models supplied to the Murray–Darling Basin Authority (MDBA), and so incorporated into the MDBA’s modelling system.
The models also include an implicit representation of all the unknown diversions that occur. All NSW models are calibrated using a water balance technique that compares recorded flows and upstream and downstream gauges, and calculates a loss after other water balance components are accounted for, including metered diversions, evaporative losses, and tributary inflows.
This measurement of ‘loss’ includes some genuine loss, for example through seepage, but also represents other unknowns in the system— where we know the water has gone, but not necessarily where it went.
The floodplain harvesting estimates account for some of this water, and bring it into the existing licensing framework so it can be actively managed in the future. The new entitlements do not allow anyone to extract any more water than they have been to date, but represent better tracking of the existing diversions.
This also means that decisions and plans based on our earlier measurements of flows won’t change once floodplain harvesting is included in the modelling. The water has already been accounted for. The ‘loss’ component of the model hasn’t changed, but can now be separated into known diversions and other forms of loss.
Why don’t these LTDLE factors reflect all the low allocations I have been getting lately?
Basin Plan reference climate period
The reference to determine the baseline diversion limit (BDL), and therefore the sustainable diversion limit for the Basin Plan, is the long term average diversions from 1895 to 2009. The department has used the period 1895 to 2017 in determining the LTDLE factors.
This period was chosen because it includes a representative range of climate conditions. By using this time period, we include a variety of conditions from the major floods of the 1950s through to the drought of the 2000s.
The Basin Plan does not try to deliver the same amount of water each year, but instead seeks to provide volumes that reflect our highly variable climate. Because storage volumes and inflows change annually, so does the total amount of water allocated. In effect, all water entitlement allocations are indexed based on water availability, including entitlements held by the Commonwealth for the environment.
By using BDL models that cover the entire reference period 1895–2009, we can consider the effect of a range of climatic condition on water availability and use.
The method for determining the LTDLE factor also uses more recent historical information, including for very dry years, to determine how entitlement holders use any allocated water.
But why is my allocation so low?
In simple terms, entitlements are low at the moment because it has been extremely dry and there has been little rain in recent time. This has resulted in reduced flows in the rivers and inflows to dams, which means there is less water that can be distributed to entitlement holders via an available water determination.
For an independent article on the topic, see the ABC report published 11 Jun 2018 titled ‘How bad is the drought and why has it been so dry?’.
Why use a BDL model, and which one is being used, and why that one?
What is a BDL model?
The baseline diversion limit (BDL) model is one of a number of hydrologic river models used to inform decisions on how water is shared and what rules should be in place. The models do this by showing how those rules would have performed over the last 120 years of climate records.
The most common technique for using the models is called a two-model, long-term comparison. An existing model is configured to represent a reference case or condition, and the results are collected. The model is then reconfigured to represent a change of rule or policy, and run again. By comparing the results, modellers can look at the relative difference in outcomes from the two scenarios. This will show if a proposed change is likely to make things better or worse, compared to the desired outcome.
The models aren’t able to provide absolute data on the exact volumes of flows—for example they would not be able to give the exact amount of water at a given point on 3 June 1904. However, experience has shown that a two-model comparison can give an accurate indication of how much higher or lower a particular day’s flow might be with a different rule in place.
The Basin Plan uses a reference case called a baseline diversion limit (BDL), as defined by Schedule 3 of the Basin Plan. It represents the actual historical conditions up until 2009, and was developed by the MDBA based on hydrologic models provided by each of the states. This reference model was also used to assess how much consumptive water needed to be recovered to achieve a sustainable level of diversions.
More information on the MDBA’s report and their BDL models.
Why use a BDL model?
The department has used the BDL models in calculating LTDLE factors to be sure that the factors produced are consistent with the methods the MDBA used to calculate the required 2750 gigalitres recovery volume.
The department considered using other alternatives, including a:
- ‘current conditions’ scenario, where a model is configured to represent our best approximation of today’s conditions
- ‘full development conditions’ scenario, which would estimate how water might be used at a points in the future. For example, it would estimate water use in 2029, 10 years after the Basin Plan is implemented, or some point where all available allocations are fully utilised.
Many people have proposed that the full development conditions scenario should be used to ensure that there is never less than 2750 gigalitres of water available to the Commonwealth Environmental Water Holder.
However, the NSW Government believes the use of anything but the BDL is inconsistent with the assumptions in the Basin Plan. It would also be very likely to skew the balance the MDBA has struck between environmental, social and productive outcomes from water in the Basin.
Why are we using this version of the BDL?
NSW has developed new BDL model scenarios. They were used as part of the consultation process with stakeholder advisory panels for the development of water resource plans.
Under the Basin Plan, new BDL model scenarios may be developed at any time, due to the requirement to use the ‘best available information’. The department has given the MDBA its proposed new BDL scenario models and technical supporting material. The department is working closely with the MDBA to review the new model scenarios, to be sure that the proposed BDL scenario is a more accurate representation of the BDL condition, as described by schedule 3 of the Basin Plan.
NSW has decided that only the BDL scenario currently recognised as the best available information by the MDBA will be used to calculate LTDLE factors at any point. As with the choice to use the BDL model, this is intended to maintain consistency with the assumptions in the Basin Plan.
When would we change to a newer BDL scenario?
NSW intends to update the factors after the MDBA provides either a confirmation that our proposed new BDL scenarios are an improvement on the previous scenario, or it becomes clear that there are no known remaining issues identified in the technical review being undertaken and the recognition of ‘best available information’ has moved into a procedural activity.
Will changing to a newer version alter the factors?
Changing the BDL scenario will change the factors. However, advance trials with results from the proposed new BDL scenarios strongly suggest the changes will be small and, in most cases, immaterial. Most of the difference between the 2011 factors and the 2018 factors was related to moving the calculation onto a methodology that is consistent with the Basin Plan assumptions.
Testing and experience during the development work have also found altering the relative activation of entitlements within a BDL has only small effects on the overall recovery outcome because the environment ‘owns’ a wide range of entitlement products and if the value of one kind goes down, the value of another will go up. This tends to balance out, resulting in no material net change.
Why are there no LTDLE factors for all of the unregulated systems?
Both the Barwon–Darling and the intersecting streams are unregulated systems that have historically behaved very differently to regulated catchments. Factors for these areas have to be determined differently from those in other catchments. In the short term they will be estimated.
NSW is working with the Commonwealth and the MDBA to review the information that was originally used to agree on a recovery volume in these unregulated systems. This review will allow us to estimate factors on a case-by-case basis before the commencement of the water resource plans on 1 July 2019.
Why are these systems different from other river systems?
In the past, NSW did not require water take in unregulated systems to be tracked through meters. While water take was measured, the scale of diversions in these catchments was much lower than in a typical regulated system, and was thought to pose a lower degree of risk to the environment.
The relatively small scale of diversions also meant there was limited interest in more active management of water. Water sharing between users was managed by low-intervention rules, and there were few reasons to build models of long-term diversions in these systems and develop more active rule sets. Collectively, this has meant we have no models and no recorded diversion data that can be used to calculate the LTDLE for these systems.
Proposed requirements for metering in NSW mean that water take in unregulated water sources will be metered. Further, there are trials of new remote sensing techniques that have made it possible to build models for smaller unregulated valleys. In the future, this data will be used to calculate LTDLE factors for unregulated systems.
However, the Commonwealth has bought very few entitlements in the unregulated catchments, and it has relatively small entitlement shares. This means that the associated factors don’t significantly affect the overall water recovery balance. As a result, the NSW Government has decided there is no reason to delay the release of all other factors while waiting for Commonwealth factors for unregulated systems to be calculated.
What factors are we using in the meantime?
For the Barwon–Darling, factors were developed to reflect the unique nature of entitlements, which represent an individual share of the cap.
Within the intersecting streams, the National Parks and Wildlife Service are reconfiguring existing structural works at the Toorale property to meet environmental objectives. The water rights for this property were previously purchased for the environment, and the works have given new insight into how productive water was formerly used at Toorale. This information is being reviewed for use by NSW and the MDBA. This process will directly produce a new estimate of the baseline diversion limit (BDL) for the intersecting streams, which will inform the recovery volume for the area.
The LTDLE factors and interactions with efficiency projects
What is ‘up water’?
The term ‘up water’ is used to describe a component of the agreement between the states and the Commonwealth regarding implementation of the Basin Plan.
When the Basin Plan was established a water recovery target of 2750 gigalitres was chosen. Some jurisdictions proposed the water recovery target could be amended if the same environmental outcomes could be achieved with less water through projects that entailed additional works or rule changes. Implementing such projects would mean that a smaller volume of water than that specified in the recovery target would achieve the environmental outcomes.
In response, the sustainable diversion limit (SDL) adjustment mechanism was put in place. It allows the recovery target to be reduced by up to 650 gigalitres of ‘down water’—water that reduces the 2750 gigalitre target—where projects are set up to deliver the same outcome using less water.
There was an alternative view that the 2750 gigalitre recovery target was too low, and more water needed to be recovered for the environment. This water was meant to be recovered through the use of water efficiency projects.
For these projects, the idea was that increased efficiency in water use would let water users maintain their existing level of productive output while using a smaller amount of water. As the water recovered through efficiency would increase the recovery volume, it was called ‘up water’.
You can find further information on the following web pages:
- Federal Government releases 300-page Ernst Young report on Murray-Darling efficiency measures—The Land
- Sustainable diversion limit adjustment mechanism—MDBA
How will LTDLE factors be used with efficiency projects?
Each project will be assessed from first principles to determine the average annual amount of water saved as the result of new infrastructure or other measures, over a repeat sequence of the historical climate record. The long-term average amount of water saved will then be divided by the LTDLE factor to give the number of unit shares to be transferred. The volume of savings and the type of entitlement will determine the type of entitlement the project will recover.
As an example, Murray Irrigation Limited might choose to line a channel. A geotechnical study then finds that this would reduce seepage losses by 10 gigalitres per year for every year that the channel is filled.
The entitlement type is conveyance entitlement, for which the LTDLE factor is 0.918. So to have the project funded, Murray Irrigation Limited would end up transferring 10 / 0.918 = 10.89 gigalitres of conveyance entitlement to the project funder.
Is temporary trade counted against the users licence or the sellers licence?
Why is temporary trade data being used?
In the original modelling of LTDLE factors, the baseline diversion limit (BDL) scenario grouped together the estimated take from a range of entitlement types and used this to calculate a single value. Historical records of diversions were then used to split the complete diverted volume into different components that could be attributed to each of the entitlement types. The results produced were both unreliable and confusing.
Using historical diversion records included using an estimate of water availability based on the modelled long-term average reliability of allocations. In some cases this resulted in the records showing more water having been used than was available through allocations, or even the amount available at a valley scale.
There were also examples where usage was recorded against licence types in a way that didn’t match up with our general understanding of how entitlements were used in practice.
The department has looked at some key examples and determined that systematic temporary trades would explain the vast majority of the confusing results observed. Historically, it had been assumed that the amount of trade was small enough to ignore in determining the LTDLE factors. This work determined that this was no longer a reasonable assumption.
This change represents a significant difference to earlier methods of calculating cap or LTDLE factors.
Where is the use counted?
The NSW water accounting system uses a system similar to a bank account, where each water entitlement is linked to a single account. When water is allocated against the entitlement share from an available water determination, the volume of water is added to the account. When water is taken, the meter reading is deducted from the balance.
Similarly, when a water allocation is traded to an account, that account balance will go up, and when it is traded out, the account will go down. Diversion is recorded at the point where the water is taken. Traded water is therefore counted as use from the buyer’s account, because that is where the use is measured.
However, when determining LTDLE factors the accounting is more complex. The trade affects the volume of allocation water available, therefore it needs to be accounted for the seller and for the buyer involved in the transaction.
The following is a simplified example of how trade can affect the calculation of factors:
Farmer A has a 100 ML high-security entitlement. Over time, he has an allocation of 100 ML, and uses an average of 20 ML.
If we ignored trade, the LTDLE factor would appear to be 20/100 = 0.2. This would mean that by buying the entitlement, the Commonwealth would assume that there would be only 20 ML less irrigation taking place.
However, if Farmer A only uses 20 ML each year because he consistently sells 70 ML to Farmer B, this is not an accurate reflection of usage. A better measure of Farmer A’s usage would be 20/(100-70) = 0.66.
At the same time, Farmer B has a 100 ML general security entitlement. Over time, she has an average allocation of 60%, and uses an average of 55 ML of that entitlement. However, her overall usage is an average 125 ML (55 ML + 70 ML), because of the trade from Farmer A.
If we ignore the impact of trade, the direct calculation would give an LTDLE factor of 1.25, which clearly can’t occur. However, if we include the trade from Farmer A, the calculation becomes 125/(60+70)= 0.96
In practical terms, this results in some of the recorded long-term usage being counted for LTDLE purposes on both sides of the equation. This is done to give the best possible indication of the historical use of allocations, as expressed through trade and usage data in a range of conditions.
Why are the factors not simply calculated directly from long term BDL model outputs?
We don’t have direct representation of diversions by entitlement type
Most of the current generation of hydrologic models were designed and built approximately 20-years ago. They were designed to represent the volume of water being extracted from a regulated river. While they have been updated and recalibrated since they were first built, this continues to be their core focus.
All of the models use nodes to represent a point at which water is extracted. These nodes represent a group of entitlements whose account water is associated with a particular node. However, the calculated extraction is assigned only to the node, and is not linked back to the entitlement that provided the account volume to begin with.
For the original purpose of the models, the details around what water was associated with which entitlement was not a critical factor, and this information wasn’t built into the models. The entitlements were only relevant because they created an account balance that the irrigation node could draw from.
Grouped diversion points
Most models don’t represent individual farms, but group them together based on similarities between how they operate, and geographical locations. In part, this is to manage computing workloads, but it is also simpler and typically more accurate to represent the average behaviour of a group of decision makers.
For example, typically all annual croppers between two river gauges will be represented by a single demand calculation. This means there could be a range of different entitlement types being represented at this single diversion point in the model, including stock and domestic, high-security, and general-security licences. So the model can’t be used to directly inform the calculation of the LTDLE factors.
I have seen use by entitlement figures quoted from models in the past?
Figures were provided in the past, but these were always informed estimates. There is no standardised way to divide grouped diversions into discrete parcels that can be attributed to specific entitlement types.
NSW models are typically more advanced, and some models can directly produce an estimated diversion by entitlement class. Most of the models can make, at a minimum, a distinction between general security, supplementary, and the high-security types of entitlement. But this distinction isn’t detailed enough to give a breakdown across all types of entitlement.
With this information, the department can make valley-specific assumptions that give a breakdown of the figures for each entitlement type. In the past, the department has provided this estimated breakdown, when requested by stakeholders.
However, when calculating LTDLE factors, there are cases where there are so few licences for some entitlement types that making these assumptions would mean directly selecting the final factor. It was decided this was neither transparent nor defensible.
Due to the strength of the models, the department has proposed to adopt an LTDLE design principle where models are used to describe the overall long-term diversion that is occurring. However, where possible, more detailed water accounting data will be used to divide the total diversion figure across entitlements.
Won’t the use of environmental water post 2009 distort results?
Ideally, diversion records that are not influenced by significant new water users entering the system would be used for calculations.
The department’s high-quality diversion records are associated with the commencement of the regulated river water sharing plans. In most cases the high-quality records start in 2004–05. However, in some valleys, the high-quality records begin from 2012.
If diversions records from after 2009 are excluded, in a number of cases this would reduce the period of record from 12 years to four years (2004 to 2009). Even more importantly, in the period 2004 to 2009 the regions were in the midst of a drought.
Excluding water records from after 2009 would mean that the records wouldn’t be representative of longer-term behaviours and decisions. Using the period 2004 to 2017 demonstrates diversions over a wider range of climatic conditions.
The total diversion component isn’t defined by account data
The total amount of LTDLE factor is set by the long-term baseline diversion limit (BDL) model, without being influenced by post-2009 environmental purchases. This prevents the recovery of environmental water unduly influencing the long-term diversions. Put another way, the account data is being used to define the relative volumes across entitlement types, while the total usage is defined by the modelled BDL long-term average usage.
The environment now owns a broad range of entitlements in most valleys. As the objective of the LTDLE factors is to determine if the Commonwealth has recovered enough entitlement to meet the volumetric recovery target, it was found that in practice the overall status of recovery wasn’t heavily influenced by the distribution of entitlement types. Where holdings of one type of entitlement went up, another type would go down, and the net effect of the Commonwealth portfolio largely balanced out.
Most of the change in recovery outcomes between the 2011 factors and the 2018 factors were driven by change to align factors with the BDL, not the adopted utilisation rates. The department found that the historical utilisation rates were most important for high-reliability entitlement types such as stock and domestic (S&D), local water utility (LWU) and high security (HS).
The observation was the utilisation of these entitlements didn’t change materially across years, and high security and conveyance entitlements were highly utilised in nearly every case. Environmental ownership is not materially altering the extent to which those entitlements are being used.
Supplementary entitlement usage is defined by the long-term model and is therefore unaffected by environmental usage. General security entitlements are then assigned the remainder of the BDL volume, so by extension they also aren’t directly affected by environmental water usage.
The proportion of take for environmental use
Ultimately, the overall use of environmental water to date isn’t significantly different from that of consumptive users. The pattern and timing of usage are different, but the overall usage from environmental water entitlements is within the range of use demonstrated by other single entitlement holders.
While the environment is often now the biggest single holder of entitlement in a valley, other water users continue to hold the majority of entitlements.