Water quality in New Zealand rivers: current state and trends

River water quality, particularly in lowland catchments, is a matter of concern to the New Zealand public. We assessed river water quality and biological state and trends using data from more than 900 monitoring sites. Parallel state and trend analyses were carried out using all sites and a subset of lowland river sites. Median water-quality state in urban and pastoral land-cover classes was poorer than in exotic forest and natural land-cover classes, and lowland sites in the urban and pastoral classes had the poorest water quality. Nutrient and Escherichia coli concentrations increased and visual clarity and Macroinvertebrate Community Index scores decreased as proportions of catchments in high-intensity agricultural and urban land cover increased. Ten-year trends (2004–2013) indicated recent improvements in ammoniacal nitrogen, dissolved reactive phosphorus and total phosphorus in the pastoral and urban classes, possibly reflecting improved land management. In contrast, trends in nitrate-nitrogen in the exotic forest and cool-dry/pastoral classes indicated worsening conditions.     

Impact of land use on the faecal microbial quality of hill‐country streams

Faecal contamination of rural streams is of increasing concern in New Zealand. This study assessed hill‐country streams in the Whatawhata district that were impacted by pastoral farming, indigenous forest, or Pinus radiata forest; by measuring Escherichia coli bacteria at 14 sampling sites fortnightly for 2 years. E. coli concentrations were highest in streams flowing through grazed pasture. In both years there was a noticeable seasonal pattern in all streams irrespective of land use, with highest bacterial concentrations in summer and autumn and lowest in winter and early spring. There was no obvious correlation between E. coli concentration and rainfall or stream flow. In those streams impacted by a change in land use from pastoral to pines during the study, E. coli concentration fell rapidly and remained at levels lower than those in streams impacted by either indigenous or 7‐year pine forests. As E. coli was detected in all but two samples, the water in these streams is not suitable for human consumption. The pastoral streams consistently failed to meet stock drinking‐water guidelines (median concentration not greater than 100 E. coli 100 ml^–1) and the forest streams failed to do so in summer. Twenty‐eight percent of pastoral samples, 25% of indigenous forest samples, 14% of 7‐year pine forest samples, and 5% in New Pines stream samples (after planting) had E. coli concentrations associated with a high level of risk for contact recreation (>500 E. coli 100 ml^‐1) and the high concentrations usually occurred in summer.     

Water quality and sediment and nutrient export from New Zealand hill‐land catchments of contrasting land use

Measurements were made of suspended sediment (SS), volatile suspended solids, dissolved organic carbon (DOC), nitrogen (N) and phosphorus (P) concentrations, turbidity, black disk visibility, pH, alkalinity, and temperature, at monthly intervals for 2–5 years on nine streams draining catchments with pasture, pine plantation, and native forest land uses. Stream flow and flow‐weighted concentrations of SS, N, and P were also measured for up to 2 years from pasture, native forest, and mixed land‐use catchments enabling calculation of export (kg ha^‐1 yr^‐1). During 1996–97, export from the pasture stream was 2.5‐ to 7‐fold higher for SS (988), total P (1.50), total Kjeldahl N (5.65), nitrate N (4.37), and ammoniacal N (0.34) than from the stream draining native forest. In contrast, export of DOC (25.5) and dissolved reactive P (DRP) (0.25) from the pasture stream were within 20% of the native stream's values. Export of SS and nutrients (except DRP) from the pasture catchment was 4‐ to 15‐fold higher during the winters of 1995 and 1996 than winter 1997 when rainfall was half the normal level. Streams draining native forest had lower temperature, sediment, and nutrient concentrations (except DRP), and higher water clarity, than those draining pine forest and pasture. A pine/scrub stream had the highest SS and turbidity and lowest DRP, pH, and alkalinity. Pasture streams had the highest concentrations of all N species (geometric means 2‐to 4‐fold > native), total P, and DOC, and also showed the greatest variation in water quality attributes in relation to season and flow. The influences of land use were attributable to differences in both source materials of sediment and nutrients available for transport and changes in rates of in‐stream processing.     

Microcoleus autumnalis and filamentous algae-dominated mats and chlorophyll-a increase with agricultural land use but respond differently to associated nutrient and sediment enrichment

ABSTRACT

This study investigated the impact of pastoral land use and nutrient and fine sediment inputs on Microcoleus autumnalis and filamentous algae-dominated mats, and benthic chlorophyll-a in streams (lower North Island, New Zealand). Surveying and sampling was undertaken monthly at 61 sites spanning a wide gradient in catchment cover and environmental conditions. Two boosted regression tree models were built. The first models included pastoral land cover and five environmental variables as predictors. In the second model pastoral land cover was replaced by nutrient/sediment data. The abundance of the two mat types and chlorophyll-a increased when pastoral land cover was between 20% and 70% (model 1). Replacement of pastoral land cover by nutrient/sediment data (model 2) slightly improved the model fit for all three periphyton variables. Microcoleus autumnalis-dominant mats increased with dissolved inorganic nitrogen concentrations up to ca. 0.6?mg?L^?1, and in streams with more frequent flushes. In contrast, filamentous algal-dominated mats increased with turbidity, and in streams with less frequent flushes. Chlorophyll-a generally followed the response of the dominant periphyton type. Increased knowledge on responses of specific periphyton types, rather than total biomass, to environmental variables is essential to guide effective management strategies.     

Growth and population dynamics of crayfish Paranephrops planifrons in streams within native forest and pastoral land uses

Population dynamics of crayfish (Paranephrops planifrons White) in streams draining native forest and pastoral catchments, Waikato, New Zealand, were investigated from September 1996 to July 1998. Crayfish densities were generally greater in native forest streams because of high recruitment over summer, but varied greatly between streams in both land uses. Peak densities in summer were 9 crayfish m^‐2 in native forest and 6 crayfish m^‐2 in pasture streams, but peak biomass in summer was much greater in pasture streams. Mark‐recapture data showed that crayfish, particularly juveniles, in pasture streams grew faster than in native forest streams, through both greater moult frequency and larger moult increments. Females reached reproductive size at c. 20 mm orbit‐carapace length (OCL) after their first year in pasture streams, but after 2 years in native forest streams. Annual degree days >10°C appeared to explain the differences in the timing of life cycles. Estimates of annual crayfish production (range = 0.8–3.4 g dry weight m^‐2 year^–1) were similar in both land uses, and P/B ratios were between 0.95 and 1.2. Despite deforestation and conversion to pasture, crayfish in these Waikato hill‐country streams have maintained similar levels of annual production to those in native forest streams, although juvenile growth rates have increased and longevity has decreased.     

Land use,associated eel production,and abundance of fish and crayfish in streams in Waikato,New Zealand

Abstract

The density and biomass of fish and crayfish, and the production of eels, was compared among streams in native forest, exotic forest, and pasture. Populations were estimated by multiple‐pass electroshocking at 11 sites in hill‐country streams in the Waikato region, North Island. Three sites were in native forest, four in exotic forest, and four in pasture. Length of stream sampled at each site was 46–94 m (41–246 m² in area), and catchment areas up stream of the sites ranged from 0.44 to 2.01 km².

A total of 487 fish were caught. The species were longfinned and shortfinned eels, banded kokopu, Cran's and redfinned bullies, and common smelt. Eels were the most abundant fish in all three land‐use types, and shortfinned eels were more abundant at pastoral sites (mean density 1.11 fish m^?2) than longfinned eels (mean density 0.129 fish m^?2). Banded kokopu were present only at forested sites. Mean fish densities were greater at pastoral sites (1.55 fish m^?2) than under either native forest (0.130 fish m^?2) or exotic forest (0.229 fish m^?2). Mean fish biomass was also greater at pastoral sites (89.7 g m^?2) than under native forest (12.8 g m^?2) or exotic forest (19.3 g m^?2). Longfinned eels made a greater contribution to the fish biomass at all sites than did shortfinned eels. Densities of crayfish were high (0.46–5.40 crayfish m^?2), but were not significantly different between land‐use types. Crayfish biomass ranged from 1.79 to 11.2 g m^?2. Total eel production was greater at pastoral sites (mean 17.9 g m^?2 year¹) than at forest sites (mean 2.39 gm^?2 year^?1).     

Current state of water quality indicators in urban streams in New Zealand

ABSTRACT

National assessments have highlighted that urban streams have some of the poorest water quality in New Zealand, yet there has been no analysis focussing on these streams. In this study we defined urban streams as >15% catchment urban land cover and used monitoring data to assess several indicators of water quality including dissolved metals, nutrients, bacteria and the macroinvertebrate community index. We investigated associations between the indicators and urban land cover, imperviousness and geographic location. Water quality was highly variable, within and between monitored sites; and invariably poor when compared to water quality criteria. Dissolved zinc was positively related to the proportion of urban land cover and imperviousness in the upstream catchment. There were significant differences in ammoniacal-N, nitrate-N, turbidity, MCI, dissolved zinc and E. coli between geographic locations. The analysis indicates that if urban development continues in its current form, increases in urban land cover around New Zealand can be expected to result in further declines in water quality and a reduced likelihood that water quality objectives will be achieved at impacted locations.     

Land‐use effects on the hyporheic ecology of five small streams near Hamilton,New Zealand

Although the importance of the subsurface saturated interstitial zone (hyporheic zone) to the ecological functioning and maintenance of water quality of stream ecosystems is well known, there is little information on the impacts of different forms of land use upon this zone. Hyporheic physico‐chemistry and invertebrates were compared among small streams draining hill‐country catchments under pasture, exotic pine forest, and native forest near Hamilton, New Zealand. In streams draining native forest, the hyporheic zone harboured a relatively diverse invertebrate fauna comprising mostly taxa common in the surface benthos, although a few apparently obligate hyporheic taxa (ostracods, blind amphipods) were collected. Few individuals and taxa occupied the hyporheic zones of streams draining pasture with some groups such as water mites conspicuously absent. The hyporheos of the stream in exotic pine forest was similar in richness and abundance to that of the pasture streams. Hyporheic water temperatures were significantly higher in the pasture streams than those in pine or native forest. There were strong positive correlations between percentage saturation of dissolved oxygen in the hyporheic zones of all streams and both species richness and total invertebrate abundance. We suggest that land clearance for pasture leads to hill slumping and siltation that bury the lateral bars along the stream channels, rendering this habitat unsuitable for hyporheic invertebrates. Channel narrowing and incision may physically remove further hyporheic habitat, and the reduction of flushing flows to remove interstitial silt and clay leads to low hyporheic dissolved oxygen concentrations and reduced colonisation by surface benthos.     

Evidence for the effects of land use on freshwater ecosystems in New Zealand

ABSTRACT

To meet the challenges of preventing and reversing adverse effects of land use on ecosystems, management actions need to be founded on strong evidence. We used the pressure-state-impact (PSI) framework to assess evidence of land-use effects on New Zealand freshwater ecosystems. The evidence consisted of published quantitative and categorical associations linking land-use pressures to state changes and ecological impacts in rivers, lakes and aquifers. There was substantial evidence of land-use effects, particularly where land use/land cover (LULC) classes were used as pressure variables. Proportions of catchment area in urban and pastoral LULC were consistently, positively correlated with contaminant levels in water bodies and negatively correlated with ecological-health indicators. Other consistent PSI associations included positive correlations between cattle stocking rates and river contaminant levels, increased fine sediment and decreased ecological-health scores in rivers following forest harvest, and increased river contaminant levels at sites with stock access. Despite these consistent associations, the evidence base has four general shortcomings that should be addressed: (1) inadequate integration of data and models that link land use and contaminant loss to state changes and impacts in freshwater ecosystems; (2) weak inferences based on LULC; (3) reliance on categorical PSI associations; (4) gaps in reported PSI associations.     

Effects of riparian manipulation on stream communities in small streams: Two case studies

The effects of riparian manipulation in New Zealand are described for two case studies, one a short‐term study of the effects of the removal of riparian vegetation on fish, and the second, a long‐term study of the effect of re‐establishment of riparian vegetation on fish and benthic macro invertebrates. The first case study was an experiment carried out between November 2001 and May 2002. Overhanging bank vegetation and in‐stream wood were removed from short reaches of a small pastoral stream that had intact riparian margins, resulting in a change in stream structure with the formation of shallow uniform runs rather than pool and riffle structures as in unmodified reaches. The removal of bank cover and consequential instream habitat changes reduced inanga (Galaxias maculatus) densities by a factor of four within months of vegetation removal, showing the importance of instream cover and habitat to inanga. Adult longfin eel (Anguilla dieffenbachii) also became less abundant in the cleared reaches, but elvers (Anguilla spp.) became more abundant. In the second case study, pastoral sections in two small streams draining from native forest catchments were restored in 1995/96 by planting riparian vegetation and preventing stock access. After 10 years, the restoration efforts had more than doubled the numbers of giant kokopu (G. argenteus) and redfin bullies (Gobiomorphus huttoni), slightly increased numbers of banded kokopu (G. fasciatus), and decreased shortfin eel (A. australis) numbers by about 40%. The macroinvertebrate communities changed so that they became more similar to those at upstream native forest reference sites. These two case studies show that riparian margins can influence the composition of the fish and macroinvertebrate communities in small streams through the effects on cover, instream habitat and probably water temperature. Riparian restoration was most effective for the fish species that use cover and pool habitat.     

Regime shifts between clear and turbid water in New Zealand lakes: Environmental correlates and implications for management and restoration

We reviewed lakes in New Zealand reported to have undergone regime shifts between macrophyte‐dominated clear water states and de‐vegetated, turbid states. Regime‐shifting lakes (RSLs) occurred along a wide latitudinal gradient. We obtained catchment land‐use data as well as data on the occurrences of introduced (non‐indigenous) macrophytes and herbivorous and benthivorous fish for the 37 RSLs and for 58 lakes with similar maximum depths and climates, but which had not been reported to have undergone regime shifts. All RSLs had a maximum depth <20 m and mean annual surface air temperature between 9 and 16°G Regime shifts were positively related to the percentage of the catchment in pasture and negatively related to the percentage of the catchment in forest. The occurrences of the introduced macrophyte Egeria densa and the introduced fish, Ameiurus nebulosus (catfish), Carassius auratus (goldfish), Scardinius erythrophthalmus (rudd), Cyprinus carpio (koi carp), and Tinca tinca (tench), were significantly correlated to regime shifts in lakes. Although the presence of other introduced aquatic macrophytes was not significantly correlated with RSLs, the number of exotic fish taxa present in lakes was strongly positively correlated with increasing prevalence of regime shifts. The strength of the correlations between land use and introduced species versus regime shifts illustrates a number of factors which could be managed to reduce the susceptibility of lakes to regime shifts and to restore lakes that have become de‐vegetated. Our findings also suggests that regime shifts in lakes were unlikely to have been common in New Zealand lakes before anthropogenic deforestation and introductions of introduced aquatic taxa.     

Land use effects on habitat,water quality,periphyton, and benthic invertebrates in Waikato,New Zealand,hill‐country streams

Water quality, habitat, and biota were compared during spring amongst c. 100 m reaches on 11 streams draining pasture, native (podocarp‐broadleaf) forest, and exotic pine forest established on pasture 15 years previously. Differences were greatest between the pasture and native forest streams. Only 1–3% of incident light reached native and pine forest streams whereas 30% reached pasture streams. Pasture streams had 2.2°C higher mean temperature than the native streams, and 5‐fold higher nitrate, 30‐fold higher algal biomass, and 11‐fold higher gross photosynthesis. Native streams were 60% wider than pasture, with pine streams intermediate. Pine and pasture streams had 3‐fold higher suspended solids and fine sediment stored in the streambed than native streams. Woody debris volume was 17‐fold greater in pine than pasture streams, with native streams intermediate. Invertebrate taxa richness did not differ between land uses. Community composition differed most between pasture and native forest, with pine forest streams intermediate. Invertebrate densities were 3‐fold higher in pasture than native streams, mainly because of more chironomids and snails, but mayflies, stoneflies, and caddisflies densities were 2–3‐fold higher in forest streams than pasture.     

Microbial enzymatic response to catchment‐scale variations in supply of dissolved organic carbon

A suite of exoenzyme activities was assayed in three New Zealand streams draining pasture, native forest, and a pine catchment. There were differences among catchments in activity of three of the five enzymes assayed (cello‐biohydrolase, N‐acetylglucosaminidase, and dihydroxylphenylalanine oxidation). A principal components analysis (PCA) demonstrates that patterns of enzyme activity can be used to separate the three stream types. An experimental addition of algal‐leachate, leaf‐litter leachate, and high dissolved organic carbon (DOC) water from a small seep also resulted in marked shifts in epilithic enzyme activities 1 day after DOC additions. Oxidative enzymes showed a particularly strong response to additions of humic DOC. As for the field samples, a PCA showed large differences among treatments indicating that exoenzyme patterns can be used to examine which DOC sources predominate in different streams. Application of this approach to describing differences among streams will require detailed seasonal sampling together with longer‐term experiments.     

Faecal microbial contamination of watercress (Nasturtium officinale) gathered by a Maori protocol in New Zealand streams

A study was carried out to determine the food safety status of watercress (Nasturtium officinale) harvested from rural streams under Maori protocol. Sixty‐five sample bunches (c. 500 g) were collected from four streams in the Waikato region of New Zealand each summer from 2005 to 2007. Two streams, Southern Enclosure and Te Waihou, were within reserves and the other two, Mangapiko and Piopio, flowed through pastoral farms. To assess faecal contamination status, Escherichia coli and thermotolerant Campylobacter were measured on watercress as collected and E. coli counts assessed against the New Zealand guideline for ready‐to‐eat foods (satisfactory: <3 E. coli per g). To determine whether washing would ensure the watercress met food safety standards, an additional 6 bunches were collected and washed in running tap water (household regime). A further 15 bunches were washed by a simulated commercial triple washing regime. At harvest, 16 of 17 watercress samples collected from the Southern Enclosure and 11 of 22 from Te Waihou met the satisfactory criterion for ready‐to‐eat food, but only 1 of 17 from Mangapiko and none of 9 from Piopio were satisfactory. No Campylobacter was recovered from any sample of watercress collected from the four sites. After washing in running tap water, E. coli numbers still exceeded the satisfactory criterion. Commercial triple washing was more effective in ensuring satisfactory watercress, but of the 15 samples subjected to this regime, only 6 met the satisfactory criterion. Escherichia coli remained firmly attached to watercress leaves after both washing regimes (presumably in biofilms). Overall, these findings suggest that it is not advisable to use watercress harvested from rural streams as a raw salad vegetable, particularly from those affected by pastoral farming.     

Modelling the time course of shade,temperature, and wood recovery in streams with riparian forest restoration

Action is increasingly being taken in New Zealand and elsewhere to restore ecological function to streams through planting of riparian zones. We used simulation modelling to explore the relative performance of three strategies to restore the riparian zone of a pastoral stream to native forest by: (1) passive regeneration; (2) planting then abandonment of a Pinus radiata plantation; and (3) active restoration by planting selected native trees. We linked the forest model LINKNZ with a shade and temperature model (sWAIORA), and a wood model (OSU_STREAMWOOD) to simulate recovery trajectories for key forest stream attributes in hypothetical streams (1.3–14.0 m channel width) in the central North Island, New Zealand. Both active restoration strategies outperformed passive regeneration in shade, temperature and stream wood volume for most of the simulation time (800 years). Although the abandoned pine plantation provided greatest shade initially (<100 years), active native planting provided the greatest benefits overall. In general, recovery of stream shade (and temperature) is expected within decades, is accelerated by deliberate planting, and is fastest in small streams in which thermal stress from sunlight exposure is greatest. However, full recovery of stream and riparian function may take centuries, being dependent on large trees providing wood to structure the channel.     

Effects of small ponds on stream water quality and macroinvertebrate communities

Six small constructed ponds (surface area 500–7500 m², catchment area 28–158 ha) in rural and native forest catchments in the Auckland region had poorer water quality than the streams they replaced. Temperature (24°C) and dissolved oxygen (DO) (4 mg/litre) criteria were exceeded for up to 46% and 84% of days, respectively, during a critical 40‐day summer period. The poor conditions found in ponds, even within undeveloped native forest catchments, indicated that the physical characteristics of ponds (e.g., lack of shade, organic sediments) affected water quality independently of other factors (e.g., land use, riparian protection). The frequency and severity of the exceedences were related to pond size, retention time, and catchment land use; the most degraded conditions were found in rural ponds with largest surface areas and longest retention times. Ponds affected water quality and macroinvertebrate communities downstream. Exceedences of temperature and DO criteria occurred more frequently and were more severe downstream than upstream of ponds. Ponds in rural catchments increased mean daily stream temperatures 3.1–6.6°C during the critical summer period, and temperature differences were three times higher than those in bush catchments (0.8–2.0°C). Elevated temperatures were observed for hundreds of metres downstream owing to the slow rate of cooling (1°C/ 100 m), expanding the extent of adverse effects well beyond the “footprint” of the pond. Macroinvertebrate community composition (sample area 1–3 m²) and values of four commonly used metrics appeared to be significantly affected by ponds in rural and native forest catchments. These finding have important management implications that should lead to modifications (e.g., breaching dams) of the estimated 4500 existing ponds in the Auckland region, where possible, and restrictions on proposals for new “on‐line” ponds.     

Factors controlling litter input dynamics in streams draining pasture,pine, and native forest catchments

The effects of varying land use on the inputs of litter to streams were investigated in nine small Waikato, New Zealand, hill country streams between June 1995 and October 1997. Mass, timing, and composition of both vertical and lateral litter inputs were measured. Litter inputs to pasture streams were lower than those to streams in native or exotic pine forest. Litter inputs to native forest streams peaked in summer, with leaf material forming the dominant litter‐type throughout the year. The pine forest sites showed a winter minimum, with a pulse of reproductive litter (pollen heads) in spring. One pasture site, where riparian vegetation included willow and poplar showed an autumnal peak, with low inputs at other times of the year. Climate variables (air temperature, rainfall, and windrun) varied in their power of prediction of litter inputs. Mean air temperature showed a strong positive relationship with monthly litter input at the most intensively sampled native forest site. Annual litter inputs were positively related to canopy cover, although canopy‐type modified this relationship. At several sites lateral inputs of litter showed a positive relationship with the slope of the contributing area. Overall, lateral inputs were positively related to % unvegetated groundcover. In open pastures the combination of a lack of riparian trees, and the potential litter‐trapping capacity of pasture grasses, severely limits inputs of coarse particulate organic matter to streams.     

Summer light‐trap catches of adult Trichoptera in hill‐country catchments of contrasting land use,Waikato, New Zealand

The distribution of adult Trichoptera in light traps was investigated alongside nine streams draining catchments under native forest, pine forest, or pasture near Hamilton, Waikato, New Zealand. The aim of the study was to determine the relationship between abundance, taxonomic richness, and community composition with respect to land use during summer, and to evaluate the use of adult Trichoptera compared with benthic invertebrates as potential bio‐indicators of the effectiveness of land‐management changes. Adult Trichoptera faunas alongside the native streams were dominated by Hydrobiosidae, Conoesucidae, and Helicopsychidae (each >10% of total Trichoptera numbers for at least two of the three sites), whereas Leptoceridae, Oeconescidae, and Hydrobiosidae were relatively abundant alongside at least two of the pine sites. Adult Trichoptera faunas at the pasture sites were strongly dominated by Hydroptilidae which made up 47–85% of numbers caught at all sites. The mean number of individuals and taxa caught in light traps increased from November to January and then declined in February for all land‐use types. Overall, total numbers and taxonomic richness of adult Trichoptera were significantly lower at the pine sites compared to the pasture or native sites. TWINSPAN classification of benthic invertebrates collected in November clearly differentiated sites based on land use for presence/absence and percentage abundance data. A similar pattern was evident for most sites when adult Trichoptera faunas were used for the four sampling dates combined, suggesting that light trapping has potential as a tool for bio‐monitoring.     

Relationships between land use and stream invertebrate community structure in a South Island,New Zealand,coastal stream catchment

Macroinvertebrate community composition was compared across streams draining catchments dominated by either native bush, agricultural or urban land uses within the Water of Leith stream catchment near Dunedin, New Zealand. Land use was associated with differences in taxon richness and faunal composition of communities present in each stream. The mean abundance levels of Ephemeroptera, Plecoptera, and Trichoptera were highest in native bush catchments, and lowest in urban catchments. In contrast, the mean abundance of Oligochaeta exhibited the opposite pattern. Increasing dominance of the urban and agricultural streams by pollution tolerant taxa was reflected in the Macroinvertebrate Community Index and Quantitative Macroinvertebrate Community Index scores.