Executive summary
There has been a large increase in the number of lakes being monitored in New Zealand in recent years, allowing a comprehensive comparison of water quality across land-use types, lake types and regions. This report summarises the current status of lake water quality in New Zealand, based on data from 121 lakes. No previous assessment of lake water quality has involved so many lakes, or incorporated data from lakes varying so widely in their water quality.
Data from lake water quality monitoring programmes of 10 regional councils and recent research data from NIWA and university research programmes were used to assess the current state (based on 2003–2006 data) and trends (from 1997 to 2006) in water quality in lakes throughout New Zealand. Water quality data were used to prepare a national picture of lake condition, and differences in condition were related to four land-cover classes: alpine, native forest/scrub, exotic forest and pasture. Differences in water quality between land-use types were also assessed using a recently developed lake classification scheme that allows the effects of lake morphology (size and depth) and climate to be identified. The range of parameters assessed i ncluded total nitrogen, total phosphorus, chlorophyll a, Secchi depth, ammonium, nitrate+nitrite, dissolved reactive phosphorus, suspended solids, and temperature. Two indices of water quality and ecological condition were used: the Trophic Level Index (TLI) method, based on lake water quality monitoring results, and the LakeSPI index, based on the depth and composition of the submerged aquatic plant community.
Water quality was found to be strongly correlated with land-cover classes, especially in comparing native vs pasture classes. Median values of total nitrogen, total phosphorus and chlorophyll a were four to six times higher in the pasture classes than in the native class. Water quality was also most variable within the pasture class, with many more extreme high values in this class than in the native or exotic forest classes. High exotic forest cover was usually associated with high phosphorus concentrations, but less frequently with high nitrogen concentrations.
Secchi depth data were lacking in many of the records, but those available supported the other data, with median values of 6.2 m in the native class, 2.7 m in the exotic forest class, and 1.2 m in the pasture class. Secchi depth data included many extreme high values in the native class, and extreme lows (< 1.0 m) in the pasture class. Values for ammonium and dissolved reactive phosphorus were also elevated in the pasture and exotic classes, as was conductivity and, to a lesser extent, turbidity. Nitrate-nitrite was low in almost all lakes.
Water quality was very high in the alpine class. TLI values indicated that alpine lakes were predominantly microtrophic or oligotrophic. TLI values for the other three categories varied depending on the proportions of land-cover classes: median values were oligotrophic where the native class predominated, and eutrophic where either the exotic forest or pasture classes predominated. All but two of the 18 supertrophic and hypertrophic lakes in the data set were in the pasture class.
The broad national picture is of high water quality in deep lakes at high altitude and in unmodified catchments, and of lower water quality in modified catchments, especially in small, shallow and warm lakes. Although lake water quality was degraded in both exotic forest and pasture land catchments, pasture use was associated with the worst water quality, most notably in the cases of extreme deterioration (supertrophic and hypertrophic lakes).
A regression tree analysis that related water quality to climate, geography, catchment land cover and lake morphometry (eg, depth, area) produced seven categories of lake water quality in different environments, as shown in the following table.
|
Group |
Number of lakes in data set |
Environment description |
Water quality |
Examples |
|---|---|---|---|---|
|
1 |
26 |
Very cold climates |
Excellent (all microtrophic or oligotrophic) |
Wanaka, Sumner, Te Anau, Tekapo |
|
2 |
21 |
Cold winters but milder mean annual temperatures |
Good (mostly oligotrophic and mesotrophic) |
Rotoma, Tikitapu, Brunner, Lady |
|
3 |
10 |
Mild climates, > c. 50% native catchment cover |
Good (some oligotrophic, most mesotrophic) |
Kai-iwi, Mahinapua, Ototoa |
|
4 |
7 |
Mild climate, native and pasture cover are both < c. 50%, far northern |
All very high TP, mostly high TN, but very low chla |
A small subset of far north (Aupori Peninsula) lakes |
|
5 |
25 |
Mild climate, native and pasture cover are both < c. 50% |
Poor (mostly eutrophic and some supertrophic) |
Rotorua (Bay of Plenty), Waiparera, Tomarata, Rotokawau |
|
6 |
20 |
Mild climate, > 50% pasture cover, lake area < 0.60 km2 |
Very poor (mostly supertrophic) |
Tomahawk Lagoon, Serpentine, many Northland lakes |
|
7 |
12 |
Mild climate, > 50% pasture cover, lake area > 0.60 km2 |
Extremely poor (all hypertrophic) |
Horowhenua, Ellesmere, Whangape, Ngaroto |
Extrapolation of these lake environment categories to the nationwide database of 3,820 lakes suggests that approximately 60% of New Zealand lakes are still likely to have excellent or very good water quality; these are lakes in cold regions with high native and low pasture cover. However, approximately 30% of lakes are likely to have very poor to extremely poor water quality. Lowland lakes are especially likely to have poor water quality.
Comparison of the water quality data with LakeSPI aquatic plant data showed strong decreases in the depth of plant colonisation with decreasing water quality and decreasing native catchment cover, consistent with lower water clarity and light penetration for plant growth. Water quality and land use did not correlate with native plant species diversity and numbers of introduced weed species, which appear to be driven by other pressures such as recreational boating activity and water-level fluctuations.
Recent trends over the last 10 to 15 years could be identified from 49 lakes that have had monitoring programmes in place for more than three years. Although most of these lakes are showing no detectable change, many have already become supertrophic and hypertrophic, so that additional nutrient run-off is not causing increased chlorophyll a and decreased clarity because algal growth is already nutrient-saturated. Others not showing trends include most large, deep lakes and lakes in regions with high native cover. Most of the lakes showing deterioration were North Island lakes at low altitudes. However, an encouraging feature of the trend analysis was that some pasture lakes that have long-term catchment management plans in place, involving actions such as restoration of marginal wetlands and riparian retirement of grazing activity, are showing probable improvement.
