2.1 Groundwater quality data
2.1.1 Monitoring sites and data sources
To facilitate comparison to the results in Daughney and Wall (2007), this report made use of the same set of monitoring sites (Table 1). SOE groundwater monitoring data were extracted from 14 different regional council databases by regional council personnel. SOE data from Gisborne District Council were not provided within the timeframe necessary for this investigation, and so are not tabulated or discussed in this report. In addition to the 14 regional SOE datasets, groundwater quality data collected through the National Groundwater Monitoring Programme (NGMP) (Daughney and Reeves, 2005) were provided by GNS Science. For the remainder of this report, the data from the NGMP sites are grouped together with the SOE data from the relevant region. Information pertaining to site location, bore depth, aquifer lithology and surrounding land use and land cover is available as a spreadsheet, downloadable from the Ministry for the Environment website.
It is important to note that data from the SOE and NGMP networks cannot be considered as representative of drinking water quality in New Zealand. Many of the monitoring sites considered in this report are not used for potable water supply, but rather are used for other purposes such as irrigation, stock watering, manufacturing, etc. For those monitoring sites considered in this report that are used for small scale supply of potable water, it is possible that water treatment methods may be used to improve water quality after abstraction and before human consumption. For detailed assessment of drinking water quality in New Zealand, readers are directed to the Annual Review of Drinking Water Quality reports produced by the Ministry of Health (e.g. Ministry of Health, 2009).
Table 1. Sources of groundwater quality data and number of sites considered in this investigation.
|ARC||Auckland Regional Council||18||6||0||24|
|EBOP||Environment Bay of Plenty||56||6||0||62|
|GDC||Gisborne District Council||04||6||0||6|
|GWRC||Greater Wellington Reg. Council||56||15||0||71|
|HBRC||Hawke’s Bay Regional Council||42||8||0||50|
|MDC||Marlborough District Council||11||13||0||24|
|MWRC||Manawatu-Wanganui Reg. Council5||28||4||0||32|
|NRC||Northland Regional Council||29||7||12||48|
|ORC||Otago Regional Council||94||7||0||101|
|TDC||Tasman District Council||6||10||0||16|
|TRC||Taranaki Regional Council||65||6||0||71|
|WCRC||West Coast Regional Council||0||8||0||8|
1 Total number of SOE and NGMP sites considered in each region; includes a small proportion of sites that are no longer actively monitored.
2 Certain regional datasets provided for and considered in this report included information from a small number of non-SOE wells that are monitored for site-specific investigations.
3 Daughney and Wall (2007) list 78 SOE sites in the Environment Southland dataset, but only 57 are unique (21 “sites” in the Southland dataset represent duplicate quality control sampling events at the main 57 SOE sites).
4 SOE data from Gisborne District Council were not provided within the timeframe necessary for this investigation
5 Trading name Horizons Regional Council.
2.1.2 Key indictors of groundwater quality and guidelines used
This report makes use of two water quality guidelines, the Drinking Water Standards for New Zealand (DWSNZ) (Ministry of Health, 2005) and the Australia and New Zealand Environment Conservation Council (ANZECC) guidelines for fresh and marine water quality (Australia and New Zealand Environment Conservation Council, 2000). The DWSNZ defines health-related maximum acceptable values (MAVs) and aesthetic guideline values (GVs) related to taste, odour, or colour. The ANZECC guidelines define trigger values (TVs) based on specified protection levels for aquatic ecosystems. This report uses TVs that correspond to the 95% protection level for freshwater ecosystems. Some ANZECC TVs (e.g. for heavy metals, ammonia) are directly related to toxicity to biota, whereas other TVs (e.g. for nutrients) are not directly related to toxicity, but if exceeded may lead to adverse ecological changes. The ANZECC guidelines also define TVs for stock drinking water, which are referred to in some sections of this report. Comparisons to both water quality standards are performed on a per-parameter basis, to determine the number and percentage of monitoring sites at which calculated medians exceed the relevant MAVs, GVs, or TVs.
It is important to note that exceedence of a DWSNZ threshold does not always indicate a threat to human health, because some DWSNZ guidelines are purely aesthetic, and in the case of health-related standards, water treatment methods can often be employed to remove or reduce the concentration of the parameter of concern. Similarly, exceedence of an ANZECC TV in groundwater will not necessarily lead to adverse ecological consequences in adjacent surface waters on all occasions, because groundwater discharging to a surface water body may mix with the surface water, leading to dilution and reduction of the concentration of the parameter of concern.
In this report, analytical results from the different databases were compiled into 32 parameter categories in order to facilitate assessment of groundwater quality at the national scale (Table 2, cf. Daughney and Wall, 2007). Overview statistics are provided for all 32 parameters. However, detailed interpretation is focussed on only the following six key indicators of groundwater quality:
- Nitrate-nitrogen (NO3-N). NO3-N is routinely monitored for health and environmental reasons. The DWSNZ specifies a health-related MAV of 11.3 mg/L. High concentrations can lead to blood disease, particularly in infants (commonly known as “blue baby syndrome”) (Ministry of Health, 2005). The ANZECC guidelines specify a TV of 7.2 mg/L, which is defined on the basis of direct toxicity to biota, and a TV of 0.17 mg/L, which is defined for protection of aquatic ecosystems. NO3-N is one of the two core indicators of groundwater quality employed by the Ministry for the Environment in their most recent national SOE report (Ministry for the Environment, 2007).
- Ammoniacal-nitrogen (NH4-N).1 Nitrogen in oxygen-rich groundwater exists predominantly as NO3-N, but under the oxygen-poor conditions that exist at about one third of the monitoring sites considered in this report (Daughney and Wall, 2007), nitrogen is converted to NH4-N by natural processes.2 NH4-N is therefore a useful indicator of groundwater quality because it shows whether the absence of NO3-N signifies a lack of human or agricultural impact on groundwater quality, or if the natural conditions in the aquifer might make evidence of such impact difficult to detect. The DWSNZ specifies an aesthetic GV for NH4-N of 1.5 mg/L to minimise odour. The ANZECC guidelines define two thresholds for NH4-N: a TV of 0.9 mg/L is set to protect against direct toxicity to biota, and a TV of 0.01 mg/L is set for protection of aquatic ecosystems.
- Eschericia coli (E. coli). E. coli is a species of bacteria that indicates the presence of faecal matter in groundwater. The DWSNZ specifies a MAV of 1 colony forming unit (cfu) per 100 ml for water that is used for human consumption, and the ANZECC guidelines include a TV of 100 cfu/100 ml for water that is used for livestock consumption. E. coli is the second core groundwater quality indicator used by the Ministry for the Environment (Ministry for the Environment, 2007).
- Iron (Fe). Elevated concentrations of dissolved Fe can impart an unpleasant taste to drinking water, and so DWSNZ includes an aesthetic GV of 0.2 mg/L. There are no recognised health or ecosystem risks associated with Fe, and so there is no MAV defined in DWSNZ or TV specified in the ANZECC guidelines. However, elevated concentrations of dissolved Fe in groundwater may indicate the possible occurrence of arsenic (Smedley and Kinniburg, 2002), which itself is not routinely monitored in groundwater in New Zealand. Fe is also a useful indicator because it is only soluble under oxygen-poor conditions, so complements NH4-N to understand measured concentrations of NO3-N.
- Manganese (Mn). Elevated concentrations of dissolved Mn can also impart an unpleasant taste to drinking water and cause staining of laundry and whiteware, and so DWSNZ includes an aesthetic GV of 0.04 mg/L. Due to risks to human health and freshwater ecosystems, the DWSNZ include a MAV of 0.4 mg/L, and the ANZECC guidelines include a toxicity-related TV of 1.9 mg/L. Mn is also only soluble in oxygen-poor groundwater, so it is a useful indicator for understanding measured concentrations of NO3-N.
- Electrical conductivity.3 Electrical conductivity provides a measure of the total dissolved solids (TDS) concentration in a groundwater sample, and so it provides a useful indicator for spatial and/or temporal changes in abstraction, salt water intrusion, recharge mechanism, etc. There are no health- or ecosystem-related standards for electrical conductivity specified in DWSNZ or ANZECC, however, there are aesthetic guidelines for TDS in the DWSNZ.
Table 2. Parameter category names, units and the abbreviations used in this report (cf. Daughney and Wall, 2007). Abbreviations in bold text are for the six key indicators of groundwater quality used in this report.
Parameter Category Name
|HCO3*||mg/L as HCO3||Bicarbonate1|
|TDS*||mg/L||Total Dissolved Solids2|
|Minor Constituents and |
|E. coli||cfu/100 ml||Escherichia coli 3|
* Parameters in the “base suite” that are recommended for regular SOE monitoring by Environment Waikato (2006) and agreed by the Regional Groundwater Forum.
** Parameters in the “base suite” that are recommended for at least occasional SOE monitoring by Environment Waikato (2006) and agreed by the Regional Groundwater Forum.
1 This category includes alkalinity results, after conversion of units.
2 Where TDS has not been measured, it is estimated by summation of major and minor element concentrations.
3 E. coli is the only microbiological parameter that is considered in the drinking water standards (Ministry of Health, 2005). Where E. coli concentrations are not known, this report makes use of a proxy variable such as faecal coliforms or total coliforms. Results reported as cfu/100 ml and MPN/100 ml are assumed equivalent.
2.2 Data analysis methods
An automated spreadsheet program (Daughney, 2007) was used to compute site-specific descriptive statistics for individual parameter categories. The following site-specific calculations were performed using 1) all available data for the period 1995 to 2008, and 2) data from each individual calendar year from 1995 to 2008:
- Median: a measure of central tendency, calculated using log-probability regression to deal with results reported as being below some analytical detection limit. Median values are calculated for each of the 32 parameter categories.
- Trend: rate of change in each parameter, based on Sen’s slope estimator for all trends that are detectable with the Mann-Kendal test at the 95% confidence interval (positive numbers indicate increasing trends), or tabulated as “N” for non-significant trends. Trend assessments are performed for 22 parameter categories; trends for trace elements are not determined because they are analysed at relatively few sites and many of the reported concentrations are near or below the detection limit.
Site-specific medians and trends were then used to compute regional and national statistics for individual parameter categories, again on the basis of 1) all available data for the period 1995 to 2008, and 2) data from each individual calendar year from 1995 to 2008:
- Number of sites: the total number of sites within the region and time period of interest for which sufficient data were available to determine site-specific medians or trends4;
- Percentiles: the 5th, 25th, 50th, 75th and 95th percentiles and maximum values in the set of site-specific medians and trends for the region and time period of interest (for the case where percentiles were determined within individual calendar years, trend tests were used to identify significant year-by-year increases or decreases in the percentile values);
- % Exceedence: percentage of monitoring sites for the region and time period of interest at which site-specific median values exceed relevant thresholds from the DWSNZ and ANZECC guidelines (for the case where exceedence was assessed within individual calendar years, trend tests were used to detect significant year-by-year increases or decreases in the percentage of sites exceeding guidelines); and
- % Trend: percentage of sites within the region at which sufficient data were available to perform trend tests, and at which significant increasing or decreasing trends were detectable (trend tests require several years of quarterly data, and so could not be performed for individual calendar years).
The national and regional statistics were then used to assess the influence of various categorical factors (e.g. aquifer lithology or confinement, surrounding land use, etc.) on calculated medians and trends.
See Daughney and Wall (2007) for a more detailed discussion of statistical methods.
2.3 Data limitations
Limitations associated with the data include the following (see Daughney and Wall, 2007, for detailed discussion):
- Length and continuity of data record: some sites considered in this report were sampled just two or three times, and other sites were sampled at an irregular interval, both of which complicate trend testing;
- Coverage of parameters: the data set used in this investigation did not include potentially important parameters such as pesticides, volatile organic compounds, petroleum hydrocarbons, pharmaceuticals, endocrine disrupting compounds, etc. simply because these are not routinely analysed in SOE groundwater monitoring programmes;
- Sampling and analytical methods: details of sampling and analytical methods were not provided, but can influence the data obtained, particularly for parameters such as Fe and Mn that tend to have different dissolved and total concentrations5;
- Representativeness of monitoring sites: it is not clear whether the sites considered in this report provide an accurate representation of groundwater quality in New Zealand, because many monitoring programmes target contaminated or at-risk aquifers, or aquifers that are used for water supply;
- Site details: important information about well construction, aquifer lithology, aquifer confinement, and surrounding land use at some monitoring sites is either not known or could not be provided within the necessary timeframe for this report;
- Capture zones and travel times: it is the land use in the capture zone at the time the groundwater was recharged that has the potential to influence current groundwater quality at any given monitoring site, but the capture zone and groundwater age are unknown for most monitoring sites in New Zealand.
1. There are two main naming conventions for this parameter. This report uses the name and abbreviation ammoniacal-nitrogen (NH4-N) because it is consistently employed by regional councils in New Zealand. However, many overseas guidelines recommend the use of the parameter name ammonia-nitrogen (NH3-N). The distinction is largely semantic, and the two parameters are directly comparable in terms of units of analysis.
2. Whilst NH4-N is the dominant form of nitrogen under oxygen-poor conditions, it is important to note that microbially mediated reactions can lead a small proportion of NH4-N to be converted to NO3-N, NO2-N or N2 even under oxygen-poor conditions.
3. Some overseas guidelines recommend the use of the parameter name “conductivity”. However, this report uses the parameter name electrical conductivity because it is the common parameter name used by regional councils, and also to clearly differentiate from hydraulic conductivity, thermal conductivity, etc., which are common parameters also used in hydrology.
4. It is assumed in this study that at least one result per year is required to determine median, and at least three results are required between 1995 and 2008 to determine trend. Median and trend values determined with so few measurements carry significant uncertainty, but these uncertainties are site-specific and have relatively little influence on statistics that are aggregated to the regional and national levels.
5. In practice, “dissolved” concentrations are operationally determined by performing the analysis on a filtered sample, and usually the filtration takes place in the field when the sample is collected by passing the sample through a membrane with pore size of 0.2 or 0.45 micrometers. In contrast, “total” concentration is measured in an unfiltered sample and hence is the sum of the dissolved concentration and the concentration derived from suspended solids.