3. Costs and Benefits of the Proposed NES

3.1 Identification of costs and benefits

Table 12 details the costs and benefits identified in the discussion document, along with further benefits identified by Harris Consulting, arising from the proposed NES. Where quantitative measures of cost and benefit can be made, these are reported. Other costs and benefits are discussed in qualitative terms, with quantitative data to support the discussion.

Table 12: Cost and benefits of the proposed NES

3.2 Quantification of costs

3.2.1 Capital and ongoing maintenance/calibration costs

A key component of the analysis is the quantification of the capital, installation, data-logging/transfer and maintenance costs associated with water measuring devices. The capital cost of measuring devices varies widely according to type (mechanical, ultrasonic, electromagnetic, indirect measurement), outlet type (pipe or channel), application (surface-water, groundwater, etc.) and size. Some measuring devices may not be suitable in some situations, and so the lowest-cost option may not always be the best solution. Installation costs also vary and are influenced by factors such as outlet type, choice of measuring device, access, and the extent to which existing head works or control structures require modification. Given that many regional councils already require that new water consents be measured, costs under the proposed NES are largely associated with retro-fitting measurement devices to existing structures.

The proposed NES does not specify that a flow-meter is required for measuring water take. A measurement method that can be independently certified as meeting the required accuracy standards will satisfy the terms of the proposed NES. Because it is not feasible to predict the extent to which alternative methods of measurement will be used, the cost analysis presented assumes that flow-meters are employed on pipe outlets and that fixed control structures are employed on channels to measure consented water take. We consider that flow-meters and fixed control structures will dominate the measurement of consented water take.

Tables 13 and 14 detail the capital and ongoing maintenance costs associated with measuring devices on pipe and channel outlets, respectively. Following are some of the key assumptions made.

• Although electronic data loggers are not a requirement of the NES, they are likely to provide the least-cost option in present value terms for meeting the requirements of the proposed NES. If more manual methods of recording are used, requiring say 5 to 15 minutes per day at a cost of $30/hr, this would equate to $2.50 to $7.50 per day. If the average irrigation season is 150 days in length it is clear that a data logger costing $500 to $700 would quickly prove its worth.⁵

• Flow-meters have limited life spans. A 20-year life span has been assumed, with the capital cost repeated at 20-year intervals.

• In many situations the internal mechanism of mechanical flow-meters will not last 20 years. The costing recognises the replacement of the mechanical register at five-yearly intervals as a surrogate for calibration.

• The costs associated with establishing measuring devices on channel outlets are very site-specific. The costing assumes that the larger channel outlets are already subject to measurement, and therefore the costing is for modest channel takes.

The installation costs assume measurement devices are installed by individuals who are able to provide satisfactory compliance certification to regional councils.

Table 13: Flow-meter costs for pipes

View flow-meter costs for pipes (large table).

Notes:

• Physical installation costs range from $150 to $1,500 per device depending on type and size. Although not described individually, this cost is included in the ‘Total installation costs’ given in Table 13.

• It is assumed that a proportion of consents with a take rate of over 100 L/s have more than one take point and therefore require more than one measuring device.

• Five-yearly maintenance and calibration costs are based on in-field diagnostic calibration of electromagnetic flow-meters, laboratory calibration of ultrasonic flow-meters or the replacement of the full mechanical register in mechanical flow-meters. The replacement of the full mechanical register recognises the fact that the internal mechanism of mechanical flow-meters has a limited life span in most situations. The replacement of the mechanical register eliminates the need for five-yearly calibration.

Table 14: Measurement device costs for channels

View measurement device costs for channels (large table).

3.2.2 Annual costs associated with reporting and compliance monitoring

The proposed NES requires the continuous volumetric measurement of water, recorded at least daily. It is also proposed that actual water take data be provided to regional councils by consent holders at least annually. Typically a variety of methods are employed by regional councils to collect consented water take data from measuring devices, including physical recovery by compliance monitoring staff, internet, email, “txting”, paper-based records and telemetry.

The costings presented in Tables 13 and 14 include those associated with an electronic data logger, but this does not rule out the use of other means of recording water take. Expertise and cost are involved in downloading the data recorded and providing it to the regional council. Environment Canterbury (2007b) report physical data collection costs in a distance-dependent range of $78 to $231 per consent for their Timaru office. This cost relates strictly to travel, accommodation (if necessary) and the downloading of the water take data. It does not include any post-processing. ECan report that an independent contractor offers a physical data collection and reporting service at a cost of $230 per consent for sites within 100 km of Timaru.

Data downloading could be performed by the individual consent holder/abstractor with the necessary tools, and could then conceivably be forwarded by email to the regional council. This may provide for a lower-cost option, but the extent to which this would occur is uncertain. Our determination of the annual cost of data download is therefore based on the service being provided by a third party, as follows:

The cost of downloading and subsequent processing is broadly based on the staffing levels given for this option in ECan’s 2007 business plan. Under the terms of the proposed NES, these costs will fall to the consent holder.

It is likely that councils will develop more cost-effective measures to capture water take data on a more frequent basis than that proposed by the NES. Under the NES, actual water take must be recorded daily and reported annually. Some councils require less frequent recording of take but more frequent reporting; others require real-time information. Telemetered methods or web-based initiatives, for example, may offer a means of doing this but involve significant capital expenditure. Although such methods are not recognised in the analysis presented, more refined implementation approaches may result in significantly reduced costs of data capture.

Clearly the storage and retrieval of a large amount of data requires database systems. Most councils possess database tools to support their hydrological resource management, and many already record water take data (to a greater or lesser extent). Many of the regional councils’ existing databases should be able to handle the increased levels of data storage required over time under the status quo, and initially under the proposed NES, but these systems would need to be enhanced sooner under the NES. No specific allowance has been made for this initiative beyond the implementation package proposed by the Ministry for the Environment, which will address this issue (see section 3.2.4).

Several regional councils have raised the prospect that the proposed NES will result in increased compliance monitoring costs. The costs established here include compliance with the NES. Section 35 of the RMA imposes a duty on regional councils to monitor compliance with consent conditions. The NES will effectively enhance or replace their compliance regime for take consents, and the costs associated with its implementation have already been fully costed in this analysis. We consider it unlikely that additional costs of compliance will be incurred over and above those already outlined.

3.2.3 Consent review costs

The gazetting of the proposed NES will require regional councils to initiate consent reviews. Under this process consent holders can request a full hearing. The cost of the review process will fall to the regional councils.

In the Auckland region the implementation of the NES will require minor changes to the consent conditions, relating primarily to the frequency with which a consent holder must record water take. At present these consents require a weekly reading by the consent holder, whereas under the proposed NES daily reading will be required. Auckland Regional Council (ARC) estimate that the total cost of the consent review process is likely to be in the order of $14,500 for some 800 consents ($18.00/consent), which includes the development of educational material. The ARC believes that requests for full hearings are unlikely because the process will be able to be managed to avoid such events.⁶

ECan estimate that if the consent review process can be managed administratively (i.e. as a “deemed consent” process) it would require two full-time equivalent (FTE) staff working for four to six months to administer changes to some 6,000 consented water takes. If it is assumed that one FTE is required for a year at an organisational cost of $80,000/annum, this would equate to a staff cost of approximately $13/consent. If a full review process, including possible hearings, were required ECan estimate the cost would be at least double that of a simple administrative process. This does not include the actual cost of any hearing.⁷

Although consent holders can seek a full hearing under the review process, the likelihood of successfully challenging the proposed NES is considered to be low. The extent to which hearings may be requested is unknown, but given the low probability of success it can be anticipated that the number of requests for hearings will be restricted. Hearings can be expensive ($15,000 to $20,000 per hearing6), the cost of which will fall to regional councils. The cost of hearings under the consent review process therefore represents a risk to councils.

We estimate that the initial notification (year 0) cost of the consent review process will be about $25 per consent, and that further costs ($5 per annum per consent) will be incurred over the following five years to support the review process.

Note that these consent review costs only apply to the “with NES” scenario (see box above), with the exception of ECan. Under the status quo, the implementation of ECan’s NRRP is recognised by including these review costs. The process of installing water-measuring devices under this plan will require consent review, the cost of which will fall to the regional council.

3.2.4 Central government costs

Central government costs are confined to those associated with gazetting the regulation, its implementation, and support and monitoring initiatives. Costs associated with gazetting the regulation are likely to be relatively minor and have not been able to be quantified. The Ministry for the Environment estimates that an implementation package of $300,000 per annum for five years will be required to support the introduction and implementation of the proposed NES. They also suggest that the costs of this package will be shared by central and local government. The proposed funding package will target education, communication, industry accreditation, database development, and a verification project.

Key informant interviews and submissions have consistently raised the issue of the certification of both measuring devices and technical installation staff. The availability of suitably qualified and competent technical service providers is raised as a key bottleneck in terms of meeting both the accuracy requirement and the implementation timeframe proposed by the NES. The New Zealand Water and Wastes Association (NZWWA) estimate that the costs of developing and running a series of educational seminars for industry professionals and installers is $250,000. The seminar series would be run over 12 months and allow for two seminars per council.

Further concerns have been voiced in relation to the possibility of inferior flow-meters entering the market. The NZWWA is aware that this is already occurring, and to prevent this they believe that adherence to some internationally recognised calibration/testing standard would be desirable. Possible standards include AS 3778 and ISO 4064.⁸

The implementation costs proposed by the Ministry would therefore be as follows, assuming a 50:50 split between central government and regional councils.

3.3 Potential benefits

Water is allocated as a flow rate and as a volume that can potentially be taken. Regional ability to monitor how much water is actually taken varies. As a result, many councils do not have a good understanding of the pressures on their rivers and aquifers to which observed changes in flows or levels can be related. The resulting uncertainty about the relationship between cause and effect is a significant impediment to robust water management decision-making and to the establishment of transparent and efficient allocation processes (Aqualinc, 2007).

The recognition that there are benefits associated with measuring consented water take is almost universal among stakeholders. The common thread is the potential for improved resource understanding and enhanced management of the resource at the catchment, in-stream and individual-user level. Approaches to implementing the knowledge gained from measuring consented water take are critical to deriving these benefits. The following looks at both the qualitative and quantitative benefits from installing water-measuring devices and the measurement of consented water take.

3.3.1 Improved compliance monitoring

Compliance monitoring and enforcement are important facets of resource management, helping to ensure equity and confidence in management regimes. Equity issues are as important to abstractors as they are to in-stream value stakeholders, and concerns relating to water resource management and compliance with consent conditions have been raised by a number of submitters to the proposed NES.⁹,¹⁰

In the absence of installed water-measuring devices, compliance monitoring of the consented take rate and/or volume may be based on intermittent measurement (ranging from annually to not at all) of rate and the recording of pump hours or power consumption, estimates of irrigated area and crop requirements. Although the Environment Court¹¹ considered that the use of records of power consumption might be suitable for ascertaining compliance with consent conditions, it is clear that this method can be imprecise and is limited to abstractions that employ electric motors.

The measurement of take rate combined with the recording of pump hours or power consumption can provide a coarse indication of usage, but neither method is an accurate measurement of actual water take. Field studies in South Africa report errors of up to 79 percent for pump hour methods, and errors of between ± 25 to 50 percent for kilowatt hour methods (van der Stoep, et al., 2005). Enforcing consent conditions based on such measurement would be difficult, and the seriousness of the errors involved would make challenges to attempted enforcement inevitable for anything other than blatant non-compliance, such as that occurring during times of restriction.

Comprehensive, reliable and transparent compliance monitoring and reporting can provide for more equitable and efficient allocation of the resource during water shortages and reduce transaction costs associated with consent applications. This is shown by Horizons Regional Council’s compliance monitoring system of consented water takes. Real-time data are superimposed on consented take rates/volumes for individual consents and catchments and displayed on the internet. The system is largely self-policing, with interested parties able to observe compliance. The transparent nature of the compliance monitoring regime gives parties confidence that allocation regimes are being complied with, and this can lead to reduced transaction costs during consent application processes. Peter Taylor, of Fish and Game Wellington Region, reports that their confidence in the management of the resource by Horizons and confidence in compliance with consent conditions has allowed his organisation to be less involved in consent applications and able to support less conservative assumptions in relation to the setting of environmental flows.¹²

Successful compliance monitoring regimes combine a number of tools and implementation approaches, and water measuring devices are an essential component of these. Improved confidence in water resource management through the measurement of consented water take would occur over time under the status quo, but more immediately under the proposed NES. The value of improved public confidence is likely to be manifested as reduced transaction costs in resource decision-making. The NES does not require the monitoring of take rate, and in some situations this may limit the extent to which it results in improved compliance monitoring. Nevertheless, the NES should assist compliance monitoring.

3.3.2 Environmental flows

The determination of environmental flows is often a contested process, but some understanding of natural flow is a necessary part of determining environmental flows. Where water is being abstracted there is often a need to understand actual take in order to accurately determine “naturalised flow”. In the absence of actual take data, assumptions regarding take are made or consented allocation is simply used as a proxy for actual take.¹³

Under the NES it is proposed that all consented water takes be measured, which means that actual take − and consequently naturalised flow − can be determined with greater accuracy than under the status quo. Although the uptake of water-measuring devices would increase over time under the status quo, it is unlikely to facilitate the comprehensive measurement of consented takes. For one thing, smaller takes are likely to be ignored in catchments other than those that are under significant stress from abstractors.

Under the proposed NES, more complete information on the volumes and rates taken in surface-water systems will:

• allow for a more accurate determination of naturalised flow and help to determine environmental flow regimes

• help to establish the influence of abstractions, if any, on the incidence and duration of breaches to these regimes

• assist water allocation committees (user groups) to more effectively manage the rationing of takes during times of low flow and to prevent environmental flow regimes from being breached through abstractive water take.

Knowledge of actual take is particularly important during times of low flow, allowing for a more efficient allocation of the water resource to meet both in-stream values and the needs of abstractors. Aqualinc maintain that the protection of in-stream values could be enhanced by the “integration of water use records in consent processes to provide estimates of actual demand in the assessment of cumulative demand” (Aqualinc, 2004a). Their report recommends that takes greater than 10 L/s in Waikato be the subject of “logging”, and that consent processes include the review of actual water taken.

In groundwater systems, complete information on take will allow for aquifer responses to be compared to actual extraction rates and provide opportunities for more accurate predictive modelling.

The case study (Box 1) illustrates the difficulties of attempting to set environmental flows in a contested situation in the absence of knowledge of actual water take.

Studies of non-market values (recreation use values, option values and existence values) suggest that New Zealand residents place a high value on the protection of the natural environment. Although it is not proposed that the values reported by these studies be included in the cost−benefit analysis, they are nevertheless useful to understanding the order of magnitude of likely benefits from improved water management and the protection of non-market values.

Sharp and Kerr (2005) provide a useful summary of New Zealand studies of non-market values, which indicates the potential magnitude of existence values associated with freshwater resources. The highest value reported per household ($203 per year) was produced by a local study, which looked at the values associated with reduced groundwater extraction on the Waimea Plains in Nelson (White, et al., 2001; cited in Sharp and Kerr, 2005). This figure was nearly matched ($197 per household per year, Net Present Value, NPV = $2 billion) by the national study of values associated with proposed Kawarau River hydroelectricity developments (Kerr, 1985; cited in Sharp and Kerr, 2005). A study of the Ashburton River estimated preservation values both for Ashburton ($118 per household per year) and for the rest of Canterbury ($70 per household per year). Sharp and Kerr (2005) maintain that “the smaller regional values support the hypothesis that existence values decline with distance. Aggregating that regional figure over all Canterbury households indicates (largely non-use) NPV benefits from preservation of Ashburton River flows in excess of $70 million.

They also report that the:

... NPV of flow protection on the Waimakariri River for Canterbury households is in the order of $60 million. Existence values are generally confounded with use values. Changes in river attributes such as flow, pollution levels and even impoundment, can affect the amenity gained from activities such as boating, fishing, picnicking and walking. Amenity users may have higher existence values than others because of their familiarity with and affinity for the amenity.

Sharp and Kerr (2005) conclude that “existing studies indicate that New Zealand residents can place high value on protection of the natural environment”, and “that changes in existence values typically arise because of impacts on the structure and functioning of the natural environment”. On this basis it is reasonable to assume that improvements in the management of water resources and environmental flows will be accorded a high value by New Zealand residents.

3.3.3 Potential allocative efficiency gains

In the absence of actual take data, the determination of the reasonable needs of consented water users is based on relatively simple models. The determination of the degree to which a resource is allocated is often based on these same models. However, actual take is often significantly different from consented allocation. In irrigation-type allocations this difference arises principally as a result of rainfall varying from year to year, but other factors such as crop maturity and types, stage of farm development and other economic factors also contribute to this difference. In industrial applications differences may arise simply because the industrial activity does not operate for the 365 days per annum assumed.

Aqualinc’s (2006b) report of actual take versus consented allocation for irrigation purposes shows consistent trends across the regions studied.

They show similar trends in terms of variability in use, low use on the irrigation season margins (early and late season) and considerable variations between seasons. This is to be expected as allocations are generally issued as a fixed peak take rate intended to provide a high level of supply reliability while demand varies according to climate, crop type and growth stage. On a catchment or groundwater zone, maximum use is up to 80 percent of allocation during periods or seasons of high demand. However, it is on average generally around 50–60 percent of allocation.

Aqualinc (2006b) report that in the Manawatu−Wanganui region:

... water use for municipal water supplies to several towns was typically up to a maximum of 60 percent of daily allocations (Marton, Halcombe and Ohakea). However, water use for rural water takes was more variable both in daily use and frequency of use, ranging from less than 20 percent to more than 100 percent.

In the Waikato region a lack of knowledge regarding actual take is reported as reducing allocative efficiency:

Water use [is] lower than consented take rates and volume. Comparison of consents and water use records indicated that actual use is approximately 80% of the consented take (based on records of supply networks). This effectively reduced allocation efficiency by 10%. (Aqualinc, 2004a)

Because the models used for estimating take are reasonably coarse, data on actual take create significant potential for allocative efficiency gains. These gains arise through:

• identification of un-utilised allocation

• refinement of estimates of the degree of effective catchment allocation

• improved resource understanding, allowing for less conservatism in the setting of allocated volume at the catchment level.

Each of these is discussed below.

1. Identification of un-utilised allocation

Actual water take by consent holders is typically less than their consented allowances. Identifying actual take can therefore allow for water to be made available within the consented allocation. It is important to note that water made available in this way can be reallocated through a number of mechanisms: internally by the consent holder, through transfer to other parties, or clawed back and reallocated by the regional councils. The discussion below identifies cases in ground- and surface-water systems where measurement has identified and made available unused water within consented allocations.

Groundwater systems

It has been the experience of both ARC and Horizons that the measurement of actual water take has allowed un-utilised allocation to be identified and clawed back for reallocation to other uses.¹⁶,¹⁷ Unfortunately, the degree to which this has occurred and the end use to which the un-utilised volume has been put has not been documented. In Horizons’ case the application of volumetric charges under section 36 of the RMA, whereby consent holders are charged based on the volume allocated, has assisted the claw back.¹⁸

In the late 1980s ARC implemented the measurement of consented water take. As part of this process, consents were reviewed and conditions specifying annual volume allocations were imposed. Initial annual allocations for the Kaawa Aquifer were based on relatively simple calculations for maximum daily take rates and the number of days this take rate would be utilised. Subsequent actual consented take data from measurement allowed these initial allocations to be refined and reduced over time. Further refinement also included the development of soil and crop water utilisation models. In the Kumeu district, refinement of allocations based on actual water take for irrigation purposes showed the use factor initially developed for seasonal irrigation in the Kaawa Aquifer could be reduced, essentially freeing up the previously un-utilised component of the allocation. Actual consented water take data associated with all-year-round users allowed more refined allocations to be set for individual users, recognising that these uses may not operate for the 365 days at the peak daily allocation, as assumed in the initial allocations.¹⁹

Surface-water systems

Allocative efficiency gains can be made in surface-water systems when real-time data are available to monitor take. However, this requires telemetered measurement. Although the NES does not require telemetering of surface-water takes, the potential allocative efficiency gains are provided by the presence of a water-measuring device. An example of such gains is given in Aqualinc’s (2004b) report on the Waihou catchment for Environment Waikato, whereby knowing what was being taken, and when, could allow for the:

... establishment of a secondary class of take consents (B share), for either under-utilized allocations (for example seasonal irrigation takes) and or allocations above current water availability criteria. The secondary take would be restricted at a higher flow rate threshold than current allocations (i.e. above Q5), and therefore less reliable.

A further example of this is Synlait’s consent application on the Rakaia River, whereby they sought to identify unallocated water that can only be accessed when existing irrigators are not taking their maximum consented rate. Synlait report that their proposal would irrigate 6,000 ha.²⁰ Further benefits that result from such an approach include:

• reductions in energy use – pumping surface-water is significantly less energy intensive than pumping groundwater

• delaying the need for infrastructural development.

2. Refining the extent of allocation

In groundwater catchments where allocations are set using flow rates rather than volumetric allocations, the assumptions about the amount of water taken will determine whether the catchment is considered fully allocated. The following case study (Box 2) illustrates how better information on consented water take may lead to improved allocative efficiency via refined estimates of the extent to which a resource is actually allocated.

The consent sought by Lynton Dairy Ltd was granted by the Court, albeit with conditions on annual volume and duration. Accurate knowledge of actual water take would have greatly reduced the transaction costs associated with this hearing. It is evident from the Environment Court’s decision that they considered the measurement of actual water take to be a vital component of any water allocation policy.

3. Improved resource understanding

A further set of allocative efficiency gains may also be possible at the catchment level determination of total allocatable volumes. Regional water managers report that in the absence of information, conservative assumptions are employed in setting allocatable volume.²²,²³ The potential for allocative gains through better resource understanding, and therefore reduced conservatism, is evidenced in the in-house reports prepared as part of ECan’s Proposed Natural Resources Regional Plan (PNRRP).

The following discussion refers to first order, second order and third order approaches to setting allocation limits. The first order reflects a lower level of understanding about recharge sources, whereas with the third order approach, the groundwater system and sources of recharge are well understood. Therefore there is a lower level of certainty about the amount of groundwater that can be allocated with a first order approach. It is necessarily precautionary to reduce the risk of adverse environmental effects resulting from high allocation. With greater knowledge the second and third order approaches can be applied and in general are likely to result in an increase in the amount of groundwater available. (Environment Canterbury, 2004)

Knowledge of actual water take is an important component of second- and, in particular, third-order approaches to modelling groundwater systems and setting allocation limits. Third-order approaches should allow modelling to break free of relatively simplistic and often conservative assumptions, allowing for the effects of abstraction on the environment to be more accurately determined.²⁴ Third-order models are demanding of data. Actual water take is just one of the required inputs and as such is not the sole limiting factor in the development and use of such models.

Summary

In the absence of knowledge about the actual take, the conclusion is often drawn that a resource is fully allocated before it actually is. The refinement of allocation made possible through measurement − implementation mechanisms aside − has the potential to free up water in catchments that are presently considered to be highly allocated. Aqualinc estimate that knowledge of actual water take has the potential to free up 5 to 10 percent of allocated volume in what are considered to be highly allocated regions. They consider that achieving an allocative efficiency gain of 10 percent would take 15 to 20 years and would require the development of appropriate political processes along with more sophisticated models and management of the resource. A five percent allocative efficiency gain is, however, more readily achievable.24

Quantifying potential allocative efficiency gains

The quantification of the benefit arising from allocative efficiency gains has been made for a scenario in which an increased level of consumptive take by irrigators is enabled. Estimates made by MAF (2004) of the value of irrigation to New Zealand form the basis for evaluating the magnitude of gains associated with improved allocative efficiency.²⁵ These estimates of value to the economy (as Gorss Domestic Product, GDP) have been translated into an estimate of farm gate benefit²⁶ and applied to the likely additional water released as a result of efficiency gains. This water is likely to be made available as a result of:

• identifying water that is not being used by consent holders − this could be made available for further use by consent holders using it internally (e.g. by extending their irrigated area, by transferring the unused portion of water to another party, or by the consenting authority using claw-back provisions to take the water off the consent holder and reallocating it to other parties)

• improvement in understanding of the extent of allocation and the water resource in a catchment (particularly groundwater), leading to water being available within the allocation limit and less conservatism in setting the allocation limit.

Gains from allocative efficiency will only occur in areas where catchments are at or near full allocation. Using data from the New Zealand Business Council for Sustainable Development report (Aqualinc, 2007) which identified catchments that are approaching full allocation on a regional basis, the number of catchments in which significant gains in allocative efficiency may occur has been estimated. Regions where such benefits are possible include Canterbury and Otago, and to a lesser extent Tasman, Marlborough and Waikato.²⁷

The volume of water likely to be released for consumption is based on the 5 percent estimated by Aqualinc.²⁸ If 5 percent of the water in constrained catchments becomes available for reallocation, the value of water-based benefit gain is estimated at $48 million per annum (Table 15). To put this in context, 78 percent of water consents are used for irrigation (Aqualinc, 2006b), with the value contribution of irrigation to farm-gate GDP (2003) estimated as $920 million (MAF, 2004).

Table 15: Estimate of net benefit from a 5 percent increase in allocatable volume for water taken for irrigation

Such gains cannot be attributed entirely to the measurement of consented water take under the proposed NES, however. For one thing, the implementation approach adopted will be critical. There are transaction and other costs that will need to be offset against these gains, the extent of which will depend on how the gains are achieved. If the reallocation occurred internally to the farming system, then the transaction costs are relatively minor. Transaction costs increase with reallocation through external transfer and claw-back by regional councils. In the case of telemetry, this has an infrastructure cost over and above the established costs of the proposed NES.

There will also be constraints to these benefits being achieved. The availability of labour, capital and other resources is likely to be constrained if significant irrigation development occurs in a short period of time. There is also the possibility of environmental constraints on the ability to further intensify land use in some locations due to limits on the emission of nitrates, microbes and other contaminants. For these reasons, the estimate presented should be treated as indicative only, although it does provide an order of magnitude estimate of the likely benefits.

In conjunction with these benefits there are costs, including environmental costs. Water that was previously available for environmental flows because it was unused by consent holders would be extracted and used. Even though this would occur within an allocation limit specified by the planning authority, it will have some impact on the water body. The downstream impacts of intensification that result from increased water take will also create costs to users and ecosystems, even if these are within environmental limits. These costs include emissions of nitrates, microbes, phosphorus and greenhouse gases.

The allocative efficiency benefit of the proposed NES results from benefits arising earlier than they would in the status quo. The financial benefits identified have been modelled as arising over five to ten years following the installation of measuring devices. The extent of the benefits has been limited to the proportion of unmeasured takes versus total takes, and has been restricted to takes of greater than 20 L/s. Figure 3 describes the timing of benefits as they arise under the status quo and the proposed NES.

Allocative efficiency gains need not be limited to irrigators. Other potential beneficiaries may include municipal water supplies and industrial users, or improved environmental flows. Under this latter scenario, councils will have determined that the value of water as environmental flows is equal to or greater than its value as consumptive use.

3.3.4 Technical efficiency gains

Although there is some indication that measurement of consented water take may improve the technical efficiency of use through enhanced management and behavioural change, these gains are likely to be smaller and more difficult to identify than allocative efficiency gains.

The Foundation for Arable Research (FAR) maintains that “water measurement at one point may help measure use of water for irrigation but it does not ensure the efficient use of water for irrigation. Thus water measurement may have very limited impact on efficient water use.”³⁰ Despite this qualification, the measurement of actual water take is a necessary input to technical efficiency parameters such as ML/kg of yield, kW/ML, gross income/ML, etc.

Results from metered extraction trials in South Australia report benefits to irrigators from measuring water take at the point of extraction, allowing them to calculate efficiency factors and indices (Latcham et al., 2006). The authors state that “over 2,200 water meters have now been installed, costing licensees an estimated $7 million. However, many irrigators now claim it was money well spent as they have been able to identify costly inefficiencies coupled with the ability to make more informed decisions regarding their water use.” The Cooperative Extension Service of Kansas State University in the United States affirms that in most cases water-measuring devices will pay for themselves in “watering savings, optimum yields, and lowered energy costs” (Rogers and Black, 1992). Unfortunately, both publications are short on detail as to the magnitude of the benefits gained.

By measuring water take, abstractors are able to monitor the performance of pumps, intakes and wells. Any significant variation in flow rates may indicate that the pump, intake or well is no longer performing optimally. The ability to measure this before it becomes visually obvious can reduce pumping costs and avoid costly downtime. Dr Tony Davoren of Hydroservices Limited reports that in a typical irrigation season at least five of his 400 clients will suffer pump or well problems, resulting in costly losses in yield.³¹ The monitoring of pump flow rates and volumes would allow for any decline in performance to be identified early and rectified outside of the growing season, avoiding costly crop yield losses. Dr Davoren estimates that the loss of two weeks’ irrigation during a Canterbury growing season can result in the loss of 9−12 percent of potential yield (a loss of ≈ $300–500/ha) for a producer of ryegrass seed, and 2,500−3,000 kg/ha of dry matter in a dairying situation − the equivalent of $440−450/ha return in milk solids.

The PV10% loss of revenue for Dr Davoren’s clients is estimated at $5,300.³² The PV10% cost of a water-measuring device, including ongoing costs over a 36-year period, is estimated at $7,300. It is evident that, using these assumptions, the cost of the water-measuring device could to a large extent be offset by the benefit (avoidance of revenue loss) gained from the farming system. However, this benefit cannot be attributed entirely to the installation of a water-measuring device because there are other, perhaps more cost–effective, ways to monitor pump, well and/or intake performance. Such methods include periodic audit of irrigation systems and the installation of pressure gauges and/or an ammeter.

Behavioural change in response to the measurement of water take and in the absence of metered base charging has been reported in municipal and agricultural situations. The installation of urban water meters in Tauranga over the period 1999 to 2002 resulted in significant reductions in water take (Tauranga City, 2004). This was achieved in the absence of meter-based charging (which was introduced in 2002), but did include a significant education programme.³³

When working on conservation measures in surface-water-based irrigation districts in the Lower Rio Grande Valley, Dr Guy Fipps found that water measurement in itself reduced water take by 10 percent. When measurement was combined with training farmers in appropriate on-farm irrigation management, water take was reduced by 20−40 percent (Fipps, 2000). Although these results were achieved in the absence of volume-based pricing, Dr Fipps reports that “to get a grower to improve their on-farm water management, there must be some sort of incentive, sometimes this is the cost of water or energy for pumping, but may also be water allocation/supply restrictions, the promise of improved production, water shortages, etc.”³⁴

In the New Zealand context, both The New ZealandIrrigation Manual (Malvern Landcare Group, 2001) and the Irrigation Code of Practice and Irrigation Design Systems (Irrigation New Zealand, 2007) identify the potential for efficiency gains to be derived from measuring water use. Both publications identify savings in water use and energy use as potential benefits. Table 16 describes the level of energy saving required to offset the capital and ongoing costs associated with installing water-measuring devices on two Canterbury farm types. Importantly, it is a combination of tools (including soil moisture monitoring) that allows for any benefit to arise, and so the benefit cannot be entirely attributed to the installation of a device to measure actual water take. A similar result may also be gained from alternative measurement methods such as the use of rain gauges and/or the periodic audit of irrigation systems.

Table 16: Energy savings required to offset the cost of a flow-meter on two Canterbury farm types

It is evident from the stakeholder interviews and the literature that there may be some benefit in consented water users monitoring their water take and use. This benefit may even, in some situations, significantly offset the capital and ongoing costs associated with the installation of a water-measuring device.

3.3.5 Accounting for water

Sustainable development has been defined as “development that meets the needs of the present without compromising the ability of future generations to meet their own needs”.³⁶ The path to sustainable development requires readily available information about the links between the economy, the environment and society.

Natural resource accounts consist of stock and flow, measured in physical and monetary units. As such, they help provide a measure of New Zealand’s total natural wealth and provide information that can improve resource management. They can also help determine whether natural resources are being utilised efficiently on a national basis and across sectors. They may also be used to assess the physical and monetary extent of environmental depletion and degradation.

Natural resource accounts can help researchers analyse the effects of environmental policy on the economy, and of economic policy on the environment. Environmental accounts can, for instance, be used to establish which industries are reducing their reliance on natural resources relative to their contribution to GDP, a concept known as “decoupling”. This analysis can only be done if accurate data describing actual water take and use are available.

Statistics New Zealand reported in January 2006 that:

New Zealand is the only Organisation for Economic Co-operation and Development (OECD) nation that has not compiled a set of environmental accounts. Producing these accounts will, among other things, help New Zealand meet its commitments under various ratified international conventions.³⁷

The recently released Water Physical Stock Account 1995−2005 by Statistics New Zealand (2007) highlights the paucity of data surrounding actual water take. The authors state that :

there is insufficient data for the current stock accounts to quantify the volumes of water abstracted for the following purposes:

• irrigation
• livestock use
• private domestic use
• private industrial use
• geothermal electricity generation

Under the United Nations endorsed System of Environmental-Economic Accounting for Water (SEEAW), physical information from industries and households on water abstraction, use and supply within the economy and returns in the environment are key inputs.³⁸ Measurement of actual water take is also a fundamental requirement of the Sustainable Water Programme of Action (SWPOA), led by the Ministry for the Environment. SWPOA is, among other things, concerned with improving efficiency of use, and efficiency cannot be determined without comprehensive and accurate measurement of actual take. Therefore, such measurement is an important condition for achieving and demonstrating improved efficiency at all levels (individual, industry, regional and national).

McIndoe et al. (1998) report that “six of the sixteen economic and environmental indicators of sustainable irrigation recommended in MAF Policy Technical Paper 00/03 have water use as one of their components.” They go on to say that:

... to calculate these indicators, each day the total volume of water used on a farm for irrigation and the rate at which it is taken must be measured. This enables total seasonal volumes and depths of water applied to be calculated. Without flow measurements, the key indicators of sustainable irrigation cannot be determined.

McIndoe et al. argue that the ideal place for measuring water for irrigation efficiency calculations is at the irrigator. They recognise, however, that there are challenges in doing so, and that measurement at the pump is an acceptable method where a single irrigation unit is involved.

Further benefits arising from knowledge of actual water take reported in the New Zealand context include an improved ability to make on-farm allocation decisions when working within a volumetric allocation, and the ability to demonstrate reasonable and efficient use to regional councils and to markets.³⁹ The ability to be able to demonstrate reasonable use during consent application/review proceedings can substantially reduce the associated transaction costs, and the need for producers to be able to demonstrate “sustainability” and the efficient and reasonable use of resources is a growing trend in international markets.⁴⁰

The proposed NES will improve the ease and effectiveness with which actual water take can be reported. Under the status quo, comprehensive measurement of take is unlikely to be achieved within 15 to 20 years. The proposed NES will allow more complete physical water accounts to be compiled within five years of the regulation being gazetted. This will help the nation to meet its international obligations to report the status of its natural environment.