EXECUTIVE SUMMARY

This report was commissioned by Environment Southland (ES) to assess the water quality and ecological condition of six shallow Southland lakes. We examined the water quality and physico-chemistry, phytoplankton communities, zooplankton communities, aquatic macrophyte communities, benthic invertebrate communities, fish communities and expert assessed ecological integrity of these lakes.

We collated data from three sources for our analysis. Our primary database was collected as part of a 2004-2008 New Zealand-wide survey of the ecological integrity of shallow coastal lakes. In 2012, we conducted a repeat sampling in three Southland lakes from that survey (Lakes George/Uruwera, Vincent and The Reservoir) we also surveyed three new lakes (Lake Brunton and Lakes Sheila and Calder on Stewart Island/Rakiura). Additionally, we included water quality data collected by Environment Southland from three of the six lakes in 2000 and 2007 in our analysis.

Our research objectives were to (i) assess the ecological condition of the shallow lakes and (ii) to place them in a national context by comparing them to over 40 other shallow coastal lakes from around New Zealand. We also aimed to identify any special characteristics of the Southland lakes and to provide information on key knowledge gaps and management implications of our findings.

Overall, the mainland Southland lakes were of moderate to good water quality compared to other shallow lakes around New Zealand. It must be emphasised however, that such lakes in New Zealand have generally been heavily impacted by agricultural land use and introductions of numerous noxious invasive fish and macrophyte species.

Lake George/Uruwera has a very short water residence time and this probably prevents phytoplankton proliferation in this lake. It is the only lake in our dataset to contain Daphniacarinata, a native water flea which may play an important role in regulating phytoplanktonbiomass in the lake. Total phosphorus (TP) and total nitrogen (TN) both declined markedly between 2000 and 2004. It is not known why this occurred.

Lake Vincent is oligotrophic and the phytoplankton growth in this lake may be limited by phosphorus availability because nitrate levels in the lake tend to be high. Therefore, this lake would probably be particularly vulnerable to intensifying land use resulting in additional phosphorus input to the lake. However, a large decrease in phytoplankton biomass in this lake seems to have occurred between 2000 and 2004.

In 2012, The Reservoir was eutrophic and has high phytoplankton biomass dominated by the large desmid, Staurastrum. This contrasts with the prior sampling in 2004 when the lake hada lower phytoplankton biomass and was dominated by cyanobacteria and small green algae. The water quality data seem to suggest that phytoplankton in the lake are now nutrient saturated and it is not known what currently limits further increases in their biomass and growth.

Lake Sheila and Calder are pristine lakes which are strongly influenced by surrounding wetlands. They are humic-stained and have low nutrient concentrations and phytoplankton biomass. Lake Calder has an unusually low pH and very low solute concentrations suggesting that it is a rain-fed, seepage lake.

Lake Brunton is different to the other lakes in this study as it is an intermittently closed and open lake/lagoon (ICOLL), regulated by the natural breaching and closing of a seaward barrier. The lake undergoes large variations in salinity and when open to the sea, it is substantially flushed with sea water. At the time of sampling in 2012 it had just begun to fill after a short period of opening to the sea. It had low phytoplankton biomass and almost no zooplankton. However, it had high nitrate concentrations.

In 2004, the phytoplankton communities of the mainland Southland lakes were dominated by cyanobacteria and small green algae. To the best of our knowledge, the cyanobacteria taxa present in 2004 are not nitrogen-fixing and are not known to produce toxins. Casual observation in 2012, suggested that the phytoplankton community in The Reservoir had changed markedly, to be dominated by the large desmid, Staurastrum.

The metazooplankton communities of the Southland lakes showed low diversity and low biomass, being dominated by a calanoid copepod and small cladocerans. However, the native Daphnia carinata was present in Lake George/Uruwera in 2004 and 2012, the onlyshallow lake in New Zealand in which this zooplankter was found. While it was recorded at low densities both times, it is not clear how important a role Daphnia plays in regulating phytoplankton biomass in Lake George/Uruwera. An unusual cladoceran (a bosminid) was observed in the 2012 samples from Lakes Sheila and Vincent and we are not sure if it is a variant or sub-species of the common Bosmina meridionalis.

Overall, the submerged macrophyte communities of the Southland lakes have low taxonomic diversity and two of the lakes contain a non-indigenous species. While macrophyte communites were present in all six lakes, macrophyte beds in Lake George/Uruwera and The Reservoir were sparse. This is probably normal for Lake George/Uruwera due to its large fetch and exposure to winds, but this may indicate that macrophyte community in The Reservoir is vulnerable. The abundant macrophyte community of Lake Vincent no doubt confers some resilience to the effects of eutrophication for this lake. The macrophyte community of Lake Brunton consisted solely of Ruppia megacarpa, which has a broadsalinity tolerance and is an inhabitant in healthy ICOLL ecosystems.

Benthic invertebrate communities in the Southland lakes were quite varied. The pristine Lakes Sheila and Calder had low densities of invertebrates but had relatively high diversity. Lake Vincent had a high diversity of invertebrates, including the native freshwater mussel, Echyridella menziesi. This species was also found in Lake George/Uruwera, and while it has the potential to filter substantial amounts of phytoplankton from lake water, its importance in regulating phytoplankton biomass in these lakes is unknown. Overall, the macrobenthos of Southland shallow lakes tended to fall into three broad categories, either, those of lakes influenced by saline conditions, those with high macroinvertberate diversity associated with lakes dominated by macrophytes, and those with low diversity associated with lakes with . e.  little or no macrophyte communities).

Generally, both native and non-indigenous fish diversity in the Southland lakes is low. The nationally declining longfin eel was found in Lakes George/Uruwera, Vincent and The Reservoir, while the regionally declining giant kokopu was found in Lakes Vincent and The Reservoir, indicating that Southland lakes are a stronghold for these native species. The exotic perch was found in Lake Vincent and Lake George/Uruwera and its presence may play a role in the unusually low catch per unit effort of native fish in these two lakes.

In comparison to 41 shallow lakes around New Zealand, the ecological integrity of the mainland Southland lakes was ranked only moderately by an expert panel. Reasons for downgrading specific lakes include the presence of the non-indigenous macrophyte, Elodeacanadensis and non-indigenous perch, generally intensive catchment development, lownative fish catch per unit effort, and moderately high phytoplankton biomass.

The Southland lakes show moderate condition in terms of water quality and ecological condition. They rank well in some indicators and not so well in others. This report lists a number of knowledge gaps which, if addressed, would provide a better understanding of the vulnerabilities and degrees of resilience these lakes have to increasing anthropogenic pressures. The key aspects of these lakes to monitor and manage are the maintenance and enhancement of their macrophyte communities, controlling the downstream impacts of agricultural land uses in their catchments, and preventing the spread of invasive pest species into the lakes.