MANAGING NITRATE LEACHING TO GROUNDWATER: AN EMERGING ISSUE FOR CANTERBURY

Raymond Ford

Ken Taylor

Environment Canterbury, PO Box 345, Christchurch

Abstract

Most of Canterbury’s fluvial gravel aquifers are at risk of contamination from current and past land use activities. Over the last 20 years, land uses, such as agriculture, and rural residential subdivisions, have continued to intensify. Irrigation and nitrogenous fertilisers have also become an essential input to farming operations to satisfy market requirements for both the quality and quantity of agricultural products. During the same period, dairy cattle and deer numbers have increased five fold, and urea application has increased by a factor of nine.

A comprehensive review of groundwater nitrate-N data, carried out by Environment Canterbury in 2002, found that in areas where rivers and streams were the dominant source of recharge to groundwater and in the coastal confined gravel aquifers, nitrate concentrations were low. In areas where groundwater was shallow, unconfined, and recharged primarily by soil drainage, nitrate concentrations were significantly higher. The nitrate probably originates from a variety of sources, including past and present agricultural land uses, community sewerage schemes, industrial discharges to land, and septic tanks. The predominant trend is for increasing nitrate concentrations, mostly in wells on the lower, or eastern, parts of the plains.

Nitrate concentrations in coastal groundwater-fed streams are high, and exceed the guidelines for the control of periphyton growth, but are still mostly below the guidelines for the protection of aquatic ecosystems.

Environment Canterbury’s Proposed Natural Resources Regional Plan, notified in July 2004, establishes specific water quality outcomes for the region’s rivers, lakes, aquifers and community drinking water sources. The plan contains a range of measures, both regulatory and non-regulatory, to maintain and improve groundwater quality. One significant measure will be a requirement, when the plan becomes operative, for individual landowners to take responsibility for managing leaching of nitrate from their properties. Best management practices are to be applied to minimise nitrate nitrogen leaching, and to reduce nitrate nitrogen concentrations where high leaching rates occur.

The trend of increasing nitrogen loadings on the unconfined aquifers, spring-fed streams and coastal waters, as a result of the cumulative effects of land uses is potentially a significant issue for the region and may require catchment scale limits on nitrogen loads. Environment Canterbury is currently participating in a research programme to develop tools that will allow catchment-scale land use changes to be modelled and to predict changes in water quality.

1. Introduction

Over the last fifteen years there has been a major increase in agricultural production in New Zealand, while at the same the time there is increasing evidence that the country’s freshwater resources are becoming nutrient enriched and degraded as a result of pollution from non-point sources (Parkyn et al., 2002; PCE, 2004). If current trends persist, further decline in water quality is inevitable. Nationally, this issue is foremost among public concerns about the environment, and there is increasing pressure for public agencies to reverse the trend (PCE, 2004). This growing public unease is also mirrored in Canterbury, where water quality consistently ranks high among issues of environmental concern (ECan, 2003). This paper examines this issue in Canterbury in the context of nitrate enrichment of the region’s groundwater resources.

The Canterbury region covers a land area of 42,200 square kilometres, extending from the Clarence River in the north to the Waitaki River in the south, and bounded to the west by the Main Divide. Approximately 75% of this area is in some type of agricultural production, with the most intensive farming occurring on the down lands and plains. Around 80% of the water storage in the region is in the form of groundwater; much of this water is still in a high quality state, hence the fundamental significance of the groundwater resource as a component of the regional water cycle, and its importance to the community as a source of drinking water and as an input to agricultural and industrial production. Once regarded as a resource of unlimited quantity and invulnerable quality, in recent years there have been growing concerns about the impacts of human activity on the groundwater system, and a view that intervention is required to safeguard groundwater quality to ensure that it sustains human activities into the foreseeable future.

There is now a greater recognition of the linkages between groundwater quality and land use, and particularly of the potential impacts of nutrient leaching from farming activities. This recognition has focused attention on the state of the groundwater resource, and the science and policy required for its sustainable management (Lincoln Ventures, 1997).

2. Background

Canterbury’s groundwater resources

Groundwater in the Canterbury region occurs predominantly within fluvial gravel aquifers (Figure 1) (Brown, 2001). The largest area of groundwater is found under the Canterbury Plains, which cover an area of approximately 8,000 km². The groundwater is mostly unconfined or partially confined by beds of relatively impermeable sediments. Artesian conditions occur along the coast, on either side of Banks Peninsula, where the gravel aquifers are confined by layers of fine sediments.

The aquifers receive a continual recharge of water from the alpine rivers traversing the plains. This is supplemented by a variable recharge from rainfall. Water percolates downwards through the gravels and then flows laterally towards the coast where it emerges as springs and seeps, in the beds of many small streams on the lower reaches of the plains, in the beds of shallow coastal lakes such as Lake Ellesmere/Te Waihora (Taylor, 1996) and probably on the continental shelf (Brown, 2001; Taylor, 1996).

Figure 1. Canterbury’s groundwater resources

Economic value of groundwater

Groundwater is an important natural and economic resource for the region. Of all of the water allocated for out-of-stream uses in New Zealand, more than 60% is in Canterbury, and of this volume, approximately 33% is obtained from groundwater (LE, 2000). Most of the allocated water is used for irrigation (84%) to maintain agricultural production when potential evapotranspiration rates exceed rainfall during the summer. Smaller volumes of groundwater are used for public water supply (13%) and industry (3 %).

The Christchurch-West Melton groundwater system, the “Ashley Downs” and the Central Plains aquifers, which underlies about half the area of the Canterbury Plains, have been ranked among the top five aquifer systems in New Zealand in terms of their economic value as a source of water for domestic and industrial uses (White et al., 2004).

For many farming operations, irrigation has shifted over the last 15 years from a “drought proofing” operation to become an essential input to satisfy market demands relating to the quality and quantity of agricultural products (MAF, 2004).

Over a similar period, land uses in the region have continued to develop and intensify, in some instances at an accelerating rate. Between 1985 and 1999, the area of irrigated land in Canterbury was estimated to have increased from 150,000 ha to 350,000 ha (LE, 2000). Dairy cattle and deer numbers have increased rapidly over the last 10 years, while sheep numbers had decreased by 22 % to 7.5 million by 2004 (Figure 2).

Since 1985, there has been an enormous increase world-wide in the use of inorganic fertilisers. About half of all the inorganic fertiliser used on the planet has been used during this period (NRC, 2001), a trend which has also occurred in Canterbury. The application of urea to Canterbury farmland, for example, has increased by a factor of more than 9 between 1992 and 2004 (Figure 3). By 2004, fertiliser use in Canterbury had reached the same tonnage as the Waikato region, historically New Zealand’s heaviest user of fertiliser (Statistics NZ, 2006).

Groundwater is the principal source of drinking water in the region. Christchurch (population approximately 350,000) relies entirely on the Christchurch – West Melton groundwater system for high quality untreated groundwater for domestic drinking water and for commercial or industrial uses. Other communities, such as Timaru and Kaikoura, supplement their drinking water supplies from rivers with a groundwater source. Elsewhere on the Canterbury Plains and inland basins, most dwellings and businesses obtain their water supplies from their own wells. The demand for water for urban, commercial and industrial uses continues to grow as the population increases (LE, 2002).

3. Groundwater quality

Most of Canterbury’s aquifers are at risk of contamination from current and past land use activities, especially as land uses intensify over unconfined and semi-confined aquifers (ECan, 2004; Hanson, 2002; Hayward & Hanson, 2004). The vulnerability of aquifers to land uses varies according to their hydrogeological characteristics (Table 1).

Table 1: Principal aquifer types in the Canterbury Region and the risk of contamination

Aquifer Type	Water quality characteristics	Susceptibility to contamination
Shallow unconfined or semi-confined aquifers	Water quality variable, influenced by geology and overlying land uses	High risk, because of: · shallow water table · thin soils and relatively low proportion of organic matter with low capacity to assimilate contaminants · Gravel and sands overlying water table are highly permeable and are comprised of inert greywacke gravels that have low capacity to remove contaminants from water.
Deeper parts of unconfined or semi-confined aquifers	Generally high water quality. Very low concentrations of nutrients and microbiological contaminants. Water quality in some areas may be affected by local geology, e.g. buried peat deposits can increase iron and manganese concentrations.	Moderate risk. Depth to the water table provides some protection, but deep groundwater may be vulnerable to contamination from persistent or mobile contaminants, or land use activities in inland recharge areas. Within semi-confined layers, upwards pressure gradients and lower permeability confining layer may provide some degree of natural protection.
Coastal confined gravel aquifers	Generally very high water quality. Water quality in some areas may be affected by local geology, e.g. buried peat deposits can increase iron and manganese concentrations.	Relatively low risk. A combination of upwards groundwater pressures and layers of fine sediments. Over-abstraction could reduce groundwater pressures and cause downwards movement of contaminants or lateral salt water intrusion. Land use activities in the recharge area may threaten groundwater quality.
Non-alluvial aquifers	Low yields of water. Water quality is variable and is strongly influenced by local geology.	Relatively low risk, but dependent on the nature of the fracture system of the parent rock. Contaminant movement is difficult to predict.

Groundwater monitoring

Since 1986 Environment Canterbury (and its predecessor, the North Canterbury Catchment Board) has undertaken a regular sampling programme to assess groundwater quality in the region. A wide range of determinands are analysed, including major ions, microbiological indicators, nutrients, metals, trace elements, pH, and conductivity. The results are analysed with reference to the Drinking Water Standards for New Zealand 2000 (MoH, 2000), and published annually by Environment Canterbury (e.g., Abraham & Hanson, 2004a).

A comprehensive review of groundwater nitrate-N data held by Environment Canterbury was carried out in 2002, prompted in part by public concerns about the effects of new irrigation schemes and intensifying agricultural land uses on groundwater quality (Hanson, 2002). More detailed surveys were subsequently undertaken to define areas where nitrate-N concentrations exceeded the Ministry of Health’s Maximum Acceptable Value (MAV) for drinking water (Hayward & Hanson, 2004; Abraham & Hanson, 2004b). This value is a concentration of 50 mg/L nitrate, which is equivalent to a concentration of 11.3 mg/L nitrogen as nitrate-N (MoH, 2000).

Under natural conditions, nitrate-N concentrations in groundwater are low, generally less than 3 mg/L (Hanson, 2002), and possibly as low as 1 mg/L (Close et al., 2001). Generally, low concentrations occur in groundwater where rivers and streams are the dominant source of recharge, and in coastal confined gravel aquifers between the Rakaia and Ashley rivers, where river recharge predominates and the confining layers of fine sediments protect the aquifers from direct contamination from the land surface. In other areas, away from the coastal confined gravel aquifers and areas dominated by river recharge, nitrate-N concentrations are generally higher (greater 3 mg/L) indicating that human activities are affecting groundwater quality (Figure 4).

Nitrate-N concentrations in shallow, unconfined groundwater fluctuate seasonally, between 2-6 mg/L, with the higher concentrations occurring during winter or spring, and the lower concentrations during autumn (Hanson, 2002).

In the unconfined aquifers north of the Ashley River, nitrate-N concentrations are generally low, but concentrations above the MAV have been found in localised areas (Abraham & Hanson, 2004b). In the Amuri Basin, median nitrate-N concentrations range from 3 to 8 mg/L. High concentrations of nitrate–N (exceeding the MAV) have also been found in areas near Cheviot, Culverden, Waipara, and in the Waimakariri District, between the Ashley and Eyre rivers and south of the Eyre River. The nitrate contamination is attributed to agricultural land use practices. Poor well-head protection, and in some cases, septic tank discharges may also be contributing factors. The relative contributions, however, from these different sources has not been established.

In the Central Plains, between the Rakaia and Waimakariri rivers, high nitrate-N concentrations occur south and west of Christchurch. Many wells have concentrations that exceed 0.5 MAV, and a number of wells have concentrations above MAV. Elevated nitrate-N concentrations occurred in the first (shallowest) aquifer in the south-west part of Christchurch City. A combination of factors is likely to contribute to the elevated nitrate-N concentrations. These include agricultural land uses, and the predominance of recharge by rainfall and irrigation, with local contributions from septic systems and sewage effluent disposal. In the south-west part of Christchurch City, historical land disposal of industrial processing waste probably contributed to the increase in nitrate-N concentrations. In the upper plains, where the water table is deep (> 30 m), the nitrate-N concentrations are less than 0.5 MAV.

In the southern part of the Canterbury Plains, Ashwick Flat near Fairlie, and the lowland areas south of Timaru, the average concentrations of nitrate-N range between 2 and 8 mg/L, with areas of locally high nitrate concentrations exceeding 0.75 MAV. Nitrate contamination originates from multiple sources, including agricultural activities in the catchment, land discharges from industry, and small discharges to land, such as those from domestic septic systems ( Hanson 2000, Abraham & Hanson, 2004b).

In the vicinity of Ashburton, concentrations in shallow groundwater are high for Canterbury. On the outskirts of Tinwald township, an area of nitrate-N concentrations exceeds the MAV. The extent of this contamination has not yet been determined (Abraham & Hanson, 2004b). North-west of the settlement of Rakaia, several wells are close to or exceed 0.75 MAV (8.5 mg/L nitrate-N), and in the coastal plain south-west of Ashburton and Rakaia, nitrate-N concentrations have exceeded the MAV in samples from numerous wells.

The source of nitrate contamination in the Ashburton District is attributed to the cumulative effects of agricultural land uses, compounded in some areas by large industrial discharges to land (Abraham, & Hanson, 2004b; Hayward & Hanson, 2004). Land disposal of effluent from meat processing plants north-east of Ashburton has created two long plumes of nitrate contamination in groundwater down-gradient of the disposal sites (Hayward & Hanson, 2004). Nitrate-N concentrations in many of the wells within the plumes are above the limit set by the Drinking Water Standards for New Zealand 2000 (MoH, 2000).

The highest concentrations of nitrate-N occur in wells less than 50 m deep, but there is no clear relationship between well depth and nitrate concentration (Figure 5) (Hanson, 2002). In wells deeper than 50 m, nitrate-N concentrations tend to be lower than for wells less than 50 m, and are generally less than the MAV. However, concentrations in some deep wells, including wells deeper than 100 metres, are near or exceed 0.5 MAV.

Figure 6. Trends in nitrate-N concentrations 1995-2004

There appears to be a long-term trend of increasing nitrate-N concentrations in groundwater (Figure 6). Of the 212 wells analysed by Environment Canterbury for trend, 40 (19%) showed increasing concentrations while in 19 wells (9%), the trend was for decreases over time (Environment Canterbury unpublished data[1]). Most of the wells with increasing trends were generally distributed across the lower, or eastern, part of the Canterbury Plains. Increases in nitrate–N concentrations with time have been observed in other places with unconfined aquifers, such as the lower Waitaki, Ashwick Flat, Culverden and Kaikoura, although in these areas the dataset covers a shorter period (5 -7 years). Wells with decreasing nitrate-N concentrations have been identified in a number of places on the Canterbury Plains; of these, five were located in an area of Christchurch where a meat processing plant discharged effluent on to land until the early 1990s.

Age of the groundwater

Studies of groundwater in the central and northern parts of the Canterbury Plains generally place the age of the groundwater between 30 and 70 years, particularly in areas where rainfall recharge predominates. Younger groundwater is found in those areas where there is recharge from streams and rivers (Stewart et al., 2002).

A considerable amount of time may elapse between a land use activity and the appearance of nitrate-N in a well. The effect of current land uses on groundwater quality may not become apparent for many years. Measurements of the nitrogen isotope N¹⁵ suggest that the increase in nitrate-N concentrations in groundwater began around 1950 with the post-war intensification in agriculture (Stewart et al., 2002). Some of the measured concentrations of nitrate-N in the groundwater may have originated from land use activities several decades ago, before large areas of irrigation were developed. If rural land uses continue to intensify and more dry land areas are converted to irrigation and intensive dairying, nitrate concentrations can be expected to increase further (Hayward & Hanson, 2004).

Spring-fed streams of the lower plains

Spring-fed streams draining the lower or coastal parts of the Canterbury Plains are nutrient enriched with excessive concentrations of nitrate (Meredith & Hayward, 2002).

The median concentrations of nitrate-N in coastal streams draining the plains between the Rangitata and Rakaia rivers are above 5.6 mg/L (0.5 MAV) although mostly below the guidelines for the protection of aquatic ecosystems (Meredith et al., 2005). All of the streams greatly exceed the national guidelines for the prevention of excessive periphyton (algal) growth (Biggs, 2000). Enriched groundwater, as a result of irrigation and land use activities further up the plains, is likely to be the principal source of elevated nitrate-N concentrations (Meredith et al., 2005). Under low or base-flows, the annual loss of soluble nitrogen from these catchments to the coast is estimated to be about 1400 tonnes per annum[2].

Phosphorus concentrations in the spring-fed streams are generally very low, although close to the guideline values for controlling excessive growths of periphyton. The annual loss under low or base-flows is about 4.8 tonnes/ annum or 0.12 kg/ha.

Although the lowland waterways are enriched with nitrogen, excessive plant growths are not widespread, indicating that low phosphorus concentrations are probably limiting plant growth (Meredith & Hayward, 2002; Meredith et al., 2005).

4. Groundwater quality management and the Proposed Natural Resources Regional Plan

Environment Canterbury has statutory responsibility under the Resource Management Act 1991 for the sustainable management of the physical and natural resources in the Canterbury region, including managing the adverse effects of discharges and land use activities on water quality.

The Proposed Natural Resources Regional Plan (PNRRP)

The Council decided to implement its statutory responsibilities by preparing a regional plan to manage the impacts of human activities on the soil and water resources of the region. Public views were sought initially on the principal soil and water issues facing the region (CRC, 1995; ECan, 1999; ECan, 2001a). A draft regional plan setting out objectives, policies and methods for managing the region’s soil and water resources was prepared, and released for public comments in October 2001 (ECan, 2001b). Feedback from the community was considered by the Council, and further technical work was undertaken. The revised plan (Variation 1) comprising five chapters, was formally notified in July 2004. The period for public submissions on the plan closed in December 2005, and, as of February 2006, the submissions were being analysed by staff, with the view to beginning hearings by mid 2006.

Chapter 4 of the PNRRP contains three water quality objectives, which establish specific water quality outcomes for rivers, lakes and groundwater, and sources of community drinking water.

The groundwater quality objectives are based, in part, on the Drinking Water Standards for New Zealand 2000 (MoH, 2000). For aquifers where the groundwater quality is still high, the water quality is to be maintained in that state. Where groundwater quality has been affected by human activities, the maximum concentration of nitrate-N should not increase by more than 2 mg/L above the maximum concentration measured between 1996 and 2001, and should not exceed 11.3 mg/L. In community drinking water supply areas, the maximum concentration of nitrate-N in groundwater should not be greater than 5.6 mg/L.

Each objective is accompanied by policies that set out the position the Council will take on the use, development or protection of a natural resource. The policies are designed to achieve the objective. Underpinning each policy is a suite of regional rules and other methods. In most cases, a combination of methods will be needed, whereby each method complements the strengths and weaknesses of the others.

Broadly, the PNRRP sets out a three-pronged approach to managing nitrate contamination of groundwater, comprising a mix of regulation, education and advocacy, and investigations.

a) Regulation

(i) Point source discharges

The PNRRP proposes that prospective point source discharges to land should be considered from three aspects. Before allowing a discharge to land, the Council must be satisfied that measures have been taken to minimise the volume and concentration of contaminants to be discharged or that the applicant for the resource consent cannot use an existing treatment system or network. If a discharge to land is required, the discharge shall be applied in a way and a rate that matches the assimilative capacity of the soil. If the discharge is likely to result in contaminants reaching groundwater, the contaminant plume must be kept as small as practicable and there must be no adverse effects on other groundwater users.

The chapter also contains a list of community drinking water sources in the region, and includes a method for calculating the size of a wellhead protection zone around a community water supply well. Activities occurring within a well-head protection zone that are likely to pose a risk to drinking water quality will attract a much greater level of scrutiny prior to authorisation.

Nitrate can also enter groundwater as a result of a variety of discharges to land, including those from fertiliser, offal pits, human and animal effluent, and waste from industrial and trade processes. Several of these activities, such as the discharge of animal effluent to land, were controlled by regional rules in the Canterbury Transitional Regional Plan (CRC, 1991). As part of the development of the PNRRP, the existing rules were critically reviewed (LE, 2001; PDP, 2002), and most of the changes to the existing conditions recommended by the reviewers were incorporated into the new regional plan (e.g. discharge of animal effluent[3].

In general, small-scale discharges are permitted under the PNRRP. Authorisation of larger volumes is subject to consideration of the potential adverse effects on a case-by-case basis. However, the use of the aquifers as a direct method of disposing of animal or sewage effluent or hazardous substances is prohibited.

(ii) Managing non-point sources of nitrate leaching to groundwater

The general approach under the PNRRP is to manage nitrate leaching by using best management practices to minimise losses by matching nitrogen inputs to plant requirements and to prevent the build-up of mineral nitrogen in soils that have a high potential for leaching.

One of the most significant changes proposed is a new regional rule[4] to control nitrate leaching to groundwater in the gravel aquifers from general land uses. A resource consent will not be required (i.e., the discharge will be a permitted activity), but land owners will have to take responsibility to ensure that nitrate leaching from their properties does not exceed pre-determined thresholds.

Figure 7 illustrates the concepts underlying the rule. The two thresholds for nitrate concentration define the points above which a landowner will be required to undertake specific measures to reduce leaching losses. If land use activities result in nitrate concentrations exceeding Threshold 1, landowners must apply best management practices to minimise nitrate leaching. Where land use activities result in high concentrations of nitrate leaching, the landowner will be required to reduce nitrate concentrations to below Threshold 2. In all situations, the average annual nitrate leaching rate for the property will need to be estimated annually to determine the situation relative to the thresholds, and a record of the calculation kept.

Figure 7. Schematic outline of the nitrate leaching rule

The specific threshold values were determined on the advice of experts in soil-nutrient dynamics, and take into account a number of factors, including drinking water standards for nitrate (e.g. MoH, 2000) and the measured and modelled nitrate leaching losses under different land uses (Bidwell et al., 2003).

Both threshold values represent the average annual concentrations of nitrate-N directly below the plant rooting zone (150 cm). Threshold 1 is set at 8 mg/L, or about 70% of MAV. Land use activities with low leaching rates would fall below the threshold, while annual cropping involving soil cultivation and stocking rates of more than two dairy cows or three beef cattle per hectare are likely to exceed Threshold 1. Threshold 2 is set at 16 mg/L nitrate-N, or 70% of the World Health Organisation limit of 22.6 mg/L above which adverse health effects are observed. High stocking rates, e.g., more than 5 cows per hectare, will most likely exceed this threshold while more typical pastoral land uses are likely to fall below. Arable and vegetable cropping will be distributed above and below the threshold, depending on a range of factors such as crop type, soil drainage characteristics, and farm management systems (Bidwell et al., 2003).

The proposed rule will not take effect until the regional plan becomes operative. It is likely that the rule and the threshold values will be debated and challenged by submitters and different interest groups during hearings on Chapter 4.

b) Education and advocacy

The PNRRP recognises the importance of educating landowners about the relationship between land use practices and leaching losses, and of practices which minimise nutrient movement to groundwater. The principal methods in this regard are:

· promotion of the use of best management practices, whole farm nutrient management including the use of nutrient budgets, codes of practice, such as the “Code of Practice for Fertiliser Use” and the “Spreadmark Code of Practice”.

• support for further research on nitrogen leaching processes, including nutrient management tools, and the promotion of these amongst land users.

· encouragement, through education and advocacy, of land owners, groups or organisations, to modify their land use practices to reduce leaching losses, in areas where groundwater quality has declined.

c) Research and investigations - predicting cumulative effects of land use change on groundwater quality

There are still uncertainties about the effects of nitrate leaching at different spatial scales on groundwater quality from a mix of different land uses, including the time scales over which leaching is likely to impact on ambient groundwater concentrations.

Through the PNRRP, Environment Canterbury has identified a number of topics for future research. The Council is also participating in a project – the “Integrated Research for Aquifer Protection” (IRAP) - with a consortium of regional councils and research agencies. This project will take several years to complete. The outcome will be tools that will allow regional councils to model the cumulative effects of different types of land uses at a range of scales from paddock to catchment, and to predict nitrate concentrations in groundwater at different locations and points in time.

5. Issues associated with the implementation of the regional plan

The implementation of the PNRRP, and in particular a new regional rule to manage nitrate leaching to groundwater, will be a challenging task. It will require changes within Environment Canterbury and additional resourcing to implement the provisions of the plan, and acceptance from land owners of the need to change long-standing land use practices. The PNRRP represents the first steps at managing nitrate contamination at a regional level. As our knowledge of the problem grows, it is likely other, possibly more far-reaching, measures will need to be applied in the future.

It is likely to be several years, as the plan works its way through the hearings and appeals process, before the nitrate leaching rule becomes operative. During this period, the Council will need to work actively with land owners and industry groups to explain the issue, to illustrate and promote the use of nutrient management tools in general, and to disseminate information on nutrient budgets, in particular.

Undoubtedly there will be resistance to such changes in land use management. There are, however, many advantages to a whole farm nutrient management approach. It offers greater flexibility, compared with a more prescriptive regulatory approach, and allows land owners to optimise production while minimising nitrate leaching by applying best management practices most suited to their individual circumstances. A range of best management practices can be used with the aim of preventing the accumulation of high concentrations of mineral nitrogen in excess of plant needs, particularly before or during winter and early spring when the soil has reached field capacity and drainage rates are high (LE 2001; Di and Cameron, 2001, Ledgard & Menner, 2005).

The application of planning controls can prove difficult when they are at variance with an individual’s economic goals. However, leachate management objectives can be aligned with considerations of profitability. The loss of plant-available nitrogen from a catchment and individual properties can have a significant economic cost (Meredith et al., 2005). The use of nutrient budgeting is likely to lead to significant savings in fertiliser costs. Measures that have both economic and environmental benefits are more likely to be accepted than those that provide environmental gains alone.

In most cases, computer models, which incorporate nitrate leaching, farm management practices and inputs (e.g. fertiliser and irrigation), plant nitrogen requirements, and local environmental characteristics, such as soil type and climate, will be required to provide reliable estimates of nitrate leaching losses. Decision support system models will allow farmers to look at different management options to improve production while reducing nitrate leaching (Di and Cameron, 2002). Estimates of the average annual nitrate loss will depend, not just on accurate input information for the leaching models, but are also likely to require the contribution of people with specialised skills in agricultural nutrient management, to run the models and design appropriate management strategies. While some land managers will have these skills, others may need to rely on outside expertise, such as farm management advisors and fertiliser company representatives.

Models of nitrate leaching fall into one of two categories (Bidwell et al., 2003): nutrient budget models (e.g. ‘Overseer’) which calculate nitrate leaching loss as the residual loss after taking into account all the other inputs or outputs, and nutrient discharge models (e.g., ‘Nitrogen Leaching Estimator” (NLE)) which are empirical models that have been calibrated with experimental data and calculate leachate rates directly.

The use of the models will raise awareness of landowners about the different factors that influence nitrate leaching, while determining whether a land use is above or below the thresholds set by the rule. Potentially, there are likely to be issues of consistency as Council determines whether a particular land use complies with the rule. A particular model may give quite different answers depending on what input values are chosen, or different models are likely to produce results that may be variance with each other.

There has been criticism that existing nutrient models are not suitable for arable or mixed cropping farming. To address this problem, Environment Canterbury is contributing to a programme led by Crop and Food Research Limited with the support of the Sustainable Farming Fund, to develop a nutrient management tool for arable or mixed cropping farmers.

If the nitrate leaching rule becomes operative in its current form, Environment Canterbury will probably focus its efforts, at least initially, on implementing the rule in those parts of the region where groundwater nitrate concentrations are of particular concern. No areas have been identified yet, but the priority is likely to be areas, such as the Ashburton and Selwyn districts where there has been a long history of elevated nitrate concentrations in groundwater, and areas where nitrate concentrations are exhibiting an increasing trend as a result of land use intensification, e.g. Ashwick Flat, near Fairlie.

6. Emerging issues

The coastal environment

In the marine environment, unlike the freshwater environment, nitrogen is the critical limiting nutrient for phytoplankton growth (NRC, 2001). Nitrate is constantly being transported through the region (Figure 8). A substantial quantity of nitrate, generated on the land, eventually finds its way to coastal waters, via spring-fed streams and large rivers, and groundwater emerging at the coast. Coastal waters, because of their location, are the sink for all the nutrients being transported from the land. Estuaries and coastal lakes and lagoons may act as temporary stores or traps and delay the entry of nitrate into the coastal waters.

What are the implications of large nitrogen inputs for the coastal environment?

Nutrient enrichment of coastal waters, or “eutrophication”, stimulates the growth of phytoplankton and can lead to algal blooms, including “red” and “brown tides”, degradation of habitat, oxygen depletion and the accumulation of toxins in shellfish used as a human food source. Nutrient enrichment of coastal waters is a significant problem in the United States (NRC, 2001), the North Sea, the Baltic Sea and the Black Sea (EEA, 2005). Algal blooms also occur periodically in New Zealand coastal waters, although there is no consensus as to their cause (Taylor & Smith, 1997). High chlorophyll a concentrations have been observed along the Canterbury Bight and there are anecdotal accounts of algal blooms during prolonged periods of calm weather (A. Meredith pers. comm., 2006). The discharge into the sea of nutrient-rich waters from Lake Ellesmere/Te Waihora has prompted complaints about its impacts on the local rock lobster fishery and marine farms.

Algal blooms in coastal waters may be associated with substantial economic losses. Prohibitions on the harvesting of farmed shellfish in response to toxic blooms have impacted on aquaculture in New Zealand, and represent a critical issue for one the country’s most rapidly growing industries. The continental shelf off the Canterbury region has been identified has a potential area for new large scale aquaculture farms. Currently, there is one proposal for a large-scale mussel farm in Pegasus Bay.

The discharge of effluent or wastewater directly to the ocean also contributes to coastal nutrient loading. In recent years, there has been a shift away from land disposal of wastes to discharges into coastal waters. Christchurch City Council, and Waimakariri and Timaru District Councils are proposing to discharge sewage via ocean outfalls. Several large industries also discharge, or are proposing to discharge, their waste via ocean outfalls.

Further work is needed to understand the relationship between nutrient inputs to coastal Canterbury waters and the development of algal blooms (Bolton-Ritchie, 2006). The dynamics of the near-shore current systems in the Canterbury Bight, and the role they play in the dispersal of nutrients and the formation of algal blooms is not well known. A detailed investigation of the inshore coastal system will be required before any accurate predictions of the effects of nitrate loads on inshore coastal waters can be made.

Table 2 shows the area-specific load to the Canterbury Bight from the central part of the Canterbury region in comparison with coastal water bodies from Europe and the United States. The figure calculated for the Canterbury Bight is an underestimate because only low or base-flow loadings have been included. Because there are large areas of low intensity land use in the high country and inter-montane basins, the figure masks the yield on the plains where land use is more intensive. In the latter area yields can be in excess of 8 kg N/ha. Moreover, the estimate does not include nitrogen in the direct discharge of groundwater into coastal waters, which may be substantial. Unlike the overseas examples, nitrogen losses in the Canterbury region are almost entirely from background and agricultural non-point sources. The proportion of yield to the Baltic and North seas that is from non-point sources is about 60% of the total nitrogen load (EEA, 2005), whereas in Canterbury, non-point sources are likely to account for about 90% (calculated from data in Bolton-Ritchie, 2006).

Table 2. Total nitrogen loadings to the Canterbury Bight and some comparisons with other countries[5]

	Canterbury Bight	Chesapeake Bay, United States[6]	Baltic Sea 2000 (source EEA, 2005)	North Sea 2000 (source EEA, 2005)
Catchment area (million km^2)	0.026	0.167	1.6	0.53
Annual catchment yield (kg N/ha)	2.35	7.92	5.4	14.4

Managing cumulative effects of land use change on groundwater quality

In Canterbury, the trend is for greater agricultural production from the existing area of flat and gently rolling land. The increase in production is being facilitated by an increasing number of takes from groundwater, and it is likely to receive a further boost from proposals for new irrigation schemes and water storage projects (LE, 2002; Dommisse, 2005).

The expansion of intensive agricultural land uses over the unconfined gravel aquifers, such as the conversion from dryland farming to intensive dairying could potentially lead to a serious decline in the quality of the region’s groundwater resources. The trend is likely to be exacerbated if the soils become saturated with nitrogen, resulting in increased leaching of nitrate (PCE, 2004).

An extrapolation of current trends in nitrate concentrations suggests that within 30 years the nitrate concentrations of many shallow unconfined aquifers could exceed the limit set in the Drinking Water Standards for New Zealand 2000 (Hanson, 2002). This conclusion is also supported by modelling of land uses changes at a catchment scale (Di and Cameron, 2002, Di and Cameron, 2004). Already many spring fed streams in the lower Canterbury Plains show elevated nitrate concentrations, an indication of the effects of intensive land use activities occurring further inland. At times of low flows, the median nitrate concentrations in many coastal groundwater-fed streams between Ashburton and the Rangitata rivers are close to the guideline for aquatic ecosystems (Meredith et al., 2005).

Experience from overseas, where diffuse run-off from land uses and point source discharges have contaminated groundwater, has shown that aquifer remediation to drinking water standards is extremely difficult (Freeze and Cherry, 1989). Remediation of aquifers, even if it is technically feasible, is likely to take considerable time and resources, and therefore preventing a decline in groundwater quality is the only effective and economic way of managing the groundwater resource.

One of the major difficulties with managing groundwater contamination is that for most people, aquifers are literally “out of sight and out of mind”: a decline in groundwater quality does not attract the same level of attention until people are personally affected, unlike a river or lake where changes in water quality are readily observed. A feature of nitrate contamination is the time lag, possibly of several decades, between the leaching of the nitrate in the soil profile and its appearance in groundwater (Haynes, 1997). Therefore, the consequences of present land use activities may not become apparent for many years, and once elevated concentrations of nitrate occur in groundwater it may take many years before changes in land use practices bring about any measurable improvement in groundwater quality. This creates a problem when assessing an application for a resource consent and linking the relative contribution of nitrate from the proposed activity to the cumulative effects of land uses on water quality before the activity has commenced and its adverse effects are observed.

At this stage, Environment Canterbury has not determined how it will manage the total nitrate loads from individual catchments in the region. The Lake Taupo project has shown that good scientific, economic and social information, and robust models, will be essential to convince decision makers that management measures will be necessary (EW, 2005). The relative contribution of nitrate from natural sources, non-point and point sources within each catchment will need to be either measured or estimated before any policies can be put in place. These studies could also form a baseline for future monitoring against which the effectiveness of policies and management measures can be assessed.

The allocation of water for out-of-stream uses is an example of one policy approach that could be applied to manage the total nitrate loads. A maximum load or cap for nitrate discharges for each catchment would be determined with respect to some critical value, such as the MAV for drinking water, the toxicity limit for aquatic ecosystems (Hickey, 2002), or the assimilative capacity of coastal waters (NRC, 2001). The total load would then be apportioned among landowners and resource consent holders within the catchment, and could then be traded or assigned to other landowners within the catchment. The allocation of individual entitlements to nitrate discharge, however, raises challenging issues, such as that of determining an equitable basis for calculating the initial allocation to individuals, and, for Environment Canterbury, the practical problems associated with the monitoring of nitrate losses from land use activities and discharges, and administering any system of tradable permits (EW, 2005). Public involvement, including a wide range of groups such as district councils, landowners, research institutes, organisational groups, and Ngai Tahu, in the development of any caps on nitrate loads will be essential if management measures are to be successful.

7. Conclusions

Water quality monitoring has shown that groundwater in Canterbury is vulnerable to nitrate contamination and it is now recognised that management intervention is necessary if the high values of the groundwater resource are to be maintained in the long term. In the face of increasing intensification of land use, and upward trending nitrate concentrations in both the aquifers and coastal spring fed streams, it is also becoming clear that the assimilative capacity of the groundwater system is limited. There are still significant knowledge gaps – much needs to be understood about the transport and fate of inputs of nitrogen at the land surface, particularly in terms of the distribution of those inputs across both space and time within a heterogeneous hydrogeological system. Such knowledge will be critical in determining how the limited assimilative capacity is to be allocated.

More fundamentally, we need to improve our understanding of the nutrient fluxes and to acknowledge the linkages between groundwater, spring-fed streams and near-shore coastal waters. Future nitrate policy development will be required to encompass a more integrated or “whole hydrological systems” approach. High groundwater nitrate concentrations are not just a problem for groundwater users. Future management will need to consider the linkages between the different types of water bodies, and to identify those water bodies or values that are the most vulnerable to nitrate inputs. It may be that nitrate management will require the establishment and implementation of maximum loadings for particular catchments, and the allocation of discharge permits in a way that is analogous to the current allocation of water for abstractive use.

What may be particularly challenging to our thinking in this regard, is the prospect that such allocation limits may be determined not on the basis of the potential impacts of land-based activities on other groundwater users or the values in coastal spring-fed streams, but on the implications of those activities for the uses and values of waters overlying the continental shelf, and beyond.

Acknowledgements

Our paper draws on many years of work by the Environmental Canterbury groundwater quality monitoring team. We would like to acknowledge their contributions to our understanding of the issue, and to thank our colleagues for their comments and assistance with the preparation of this paper. Thanks also to Trevor Webb of Landcare Research for his comments.

References

Abraham, P.; Hanson, C. R. (2004a) Groundwater quality monitoring – annual summary 2003-2004. Environment Canterbury report no. U05/03. December 2004. 70 pp.

Abraham, P.; Hanson, C. R. (2004b). Nitrate contamination observed in Environment Canterbury’s annual groundwater quality survey. Environment Canterbury report no. U04/79. June 2004. 65 pp.

Bidwell, V., Cameron, K., Di, H., Francis, G. (2003) Discharge of Nitrate –nitrogen to groundwater from land use activities: recommendations for a permitted activity rule. Environment Canterbury report no. U03/27. May 2003. 18 pp.

Biggs, B.J.F. (2000) New Zealand Periphyton Guideline: detecting, monitoring and managing enrichment of streams. Ministry for the Environment, Wellington, June 2000.

Boesch, D.F.; Brinsfield, R.B.; Magnien, R.E. (2001) Chesapeake Bay Eutrophication: scientific understanding, ecosystem restoration, and challenges for agriculture. Journal of Environmental Quality 30:303 -320.

Bolton-Ritchie, L. (2006) Coastal water quality: Lake Ellesmere/Te Waihora to the Waitaki river mouth. Environment Canterbury technical report. (In prep.).

Brown, L.R.(2001) Canterbury. In ‘Groundwaters of New Zealand’, M.R. Rosen and P.A. White (eds). New Zealand Hydrological Society Inc, Wellington. p. 441- 457.

Close, M. E., Rosen, M.R, Smith, V.R. (2001) Fate and transport of nitrates and pesticides in New Zealand’s aquifers. In ‘Groundwaters of New Zealand’, M.R. Rosen and P.A. White (eds). New Zealand Hydrological Society Inc, Wellington. p. 185 -220.

CRC (1991 Transitional Regional Plan, Resource Management Act 1991 Canterbury Regional Council. October 1991.