5 A short history of monitoring and investigations and the implications for management
This summary of monitoring and investigations in Lakes Rotorua and Rotoiti is largely taken from a more extensive draft paper on 'Lake Rotorua Nutrient Load Targets'. This is being prepared for EBOP by Dr Kit Rutherford who is based at NIWA in Hamilton (Ref 4). This important paper refers to a large number of references, which are not cited in this report, covering work from the late 1960s up until the present.
It is important to summarise this work as it sets the scene for the recommendations covering gaps in understanding and the recommendations for short-term management options in section 7.
5.1 A list of monitoring and investigations
During the 1970s it was recognised that water quality was deteriorating in Lake Rotorua as a result of increased nutrient loads, mainly from the discharge of secondary treated sewage into the lake and catchment inputs from streams draining pasture and other land uses. The resulting water quality problems included: excessive growth of large invasive exotic plants (macrophytes) which smothered the native water plants; unwanted blooms of microscopic algae (phytoplankton) which reduced water clarity and were sometimes toxic; de-oxygenation during summer stratification leading to more frequent periods of anoxia; and more frequent releases of nutrients from the lake bed.
Public concerns led to a number of studies and management actions, which are listed below:
- Fish studied water quality and measured stream nutrient loads from 1968 to 1970.
- Taupo Research Laboratory (DSIR) conducted bioassay studies to determine nutrient limitation and measured 'internal loads' and lake quality during the 1970s and 1980s.
- Lake water quality was monitored from the mid 1970s by Fisheries Research Division (MAF), Taupo Research Laboratory, Hamilton Science Centre and EBOP.
- RDC has consistently monitored nutrient inputs from the Rotorua sewage treatment plant.
- Hoare conducted detailed stream nutrient load studies in 1976-1977.
- During the 1970s and 1980s, several catchment management measures were implemented through the Upper Kaituna Catchment Control Scheme to reduce nutrient inputs to Lake Rotorua.
- Williamson et al undertook a detailed study of flows and nutrient concentrations in the Ngongotaha catchment from 1987 to 1989 to assess the effectiveness of riparian retirement undertaken as part of the Kaituna Catchment Control Scheme.
- Vincent, Gibbs and Spigel have studied the effects of the flow from the Ohau Channel into Lake Rotoiti in the 1980s.
- Aquatic weed problems have been studied by the Aquatic Weeds section of MAF (now part of NIWA, Hamilton).
- More recently Rutherford et al have investigated the use of artificial and natural wetlands to strip nutrients.
- Rutherford, Elliot et al have also developed land use and catchment models to characterise nutrient losses from individual land uses and catchments, and this work is continuing.
It can be seen from this brief list that a large amount of work has been done to understand the processes that are causing the nutrient enrichment and algal problems in Lakes Rotorua and Rotoiti. This provides a firm foundation for the current work that EBOP is doing and many of the management measures that are now being contemplated.
5.2 Historical lake and nutrient targets
EBOP adopted lake water quality targets in the 1980s based, in the case of Lake Rotorua, on the acceptance that the water quality in the 1960s did not give rise to significant public concern. At the same time, the decision was made to divert treated sewage away from the lake as it was contributing 50% and 25% of the total (sewage plus catchment) load of total phosphorus (TP) and total nitrogen (TN) respectively.
The nutrient targets that were set are shown in Table 3 with the assumption that the catchment nutrient loads would remain steady.
Table 3. Total phosphorus and total nitrogen annual load targets for Lake Rotorua, 1985
| Nutrient sources |
Total phosphorus target |
Total nitrogen target |
|---|---|---|
| Sewage loads |
3.0 |
30.0 |
| Catchment loads |
34.0 |
405.0 |
| Total target loads |
37.0 |
435.0 |
As well as absolute targets, it is important to consider the N:P ratio as this can determine the type of algal blooms that can occur. Current thinking (Professor David Hamilton - pers comm) is that this ratio should be greater than 22:1 in the water body to prevent N limitation and the formation of nitrogen fixing blue-green algal blooms. This can be compared to an N:P ratio of about 1:12 for the catchment load targets in Table 3. The current N:P ratio in the lakes and its significance is discussed later.
These load targets are for both nitrogen and phosphorus and as discussed later it is important for both of these nutrients to be reduced, as well as managing the N:P ratio. This is because in phosphorus-limited systems - and at various times during the year even in nitrogen-limited systems - other algae blooms such as diatoms and dinoflagellates can occur and cause problems like reduced water clarity.
Targets for the trophic level index have already been discussed in section 3.5 and are 4.2 for Lake Rotorua and 3.5 for Lake Rotoiti.
5.3 Key issues for management
A number of key issues for management arise from the monitoring and investigations that have been undertaken for Lakes Rotorua and Rotoiti and these are summarised below.
Sewage disposal
Since 1991, sewage disposal has been by spray irrigation on pine forest in the Whakarewarewa Forest at the headwaters of the Waipa Stream. Monitoring of the stream since 1992 showed that N and P levels gradually rose to about 4 tonnes per year of total P and about 55 tonnes per year of total N in late 2001, exceeding the targets in Table 3.
It was originally intended that wetlands in the spray irrigation area would strip nutrients, but it appears that the spraying was too intense in each sub-area and the residence times too short in the wetlands for the nutrient stripping to be effective. Changes to the spray irrigation to reduce the time of application in each sub-area to 160 minutes appear to have markedly reduced the inputs of P and N into the Waipa Stream and hence into Lake Rotorua.
The implications for short-term management are discussed in section 7.
Septic tanks
Many lakeside settlements and holiday homes have septic tanks that have been shown to leach nutrients via groundwater into the lakes. This can cause localised nutrient enrichment and problems in some bays and lakeside areas.
These problems have been recognised by EBOP and RDC and priority areas will be connected to reticulated sewerage for treatment.
Lake water quality
As noted in section 3.5 and as shown in Kit Rutherford's paper (Ref 4), the TLI for Lake Rotorua does not show an overall increasing or decreasing trend from 1970 to 2002. It has not changed significantly since 1970, although some short-term variations can be seen with changes in sewage disposal and a long period of stratification in the lake over the summer of 1969-70. This is in contrast to public perceptions that water quality is steadily decreasing and indications that lake total nitrogen and total phosphorous concentrations are steadily increasing. On the other hand, David Hamilton's analysis of the long-term monitoring data using anoxia as an indicator does show a decreasing trend in water quality over a period of 50 years, especially in Lake Rotoiti.
Stream nutrient concentrations
Williamson et al showed in 1996 that the Kaituna Catchment Control Scheme had resulted in a decrease in total P and particulate N, but that total N had not changed because of an increase in soluble N (nitrate). This trend is supported by the monitoring of streams in the other Lake Rotorua catchments from the late 1960s to 2003, although as Rutherford points out (Ref 4), some caution must be observed in comparing the data from the various studies.
The overall conclusion from Rutherford's analysis is that stream nitrate concentrations in eight of the nine Lake Rotorua catchments have steadily increased, while total P concentrations have remained relatively steady. This increase of nitrate relates predominantly to baseflow, which a number of studies have shown contributes approximately 90% of the flow and a large proportion of the nutrient inputs to Lake Rotorua. Rainfall and stream-flow variations have been considered but further work may be needed to fully quantify the variations in the nutrient concentrations they cause.
Table 4 taken from Rutherford shows the stream names, monitoring sites, codes and base-flows for each of the nine catchments.
Table 4. Stream names, codes and mean baseflow
| Stream |
Site |
Code |
Fish |
Hoare |
Williamson |
EBOP |
|---|---|---|---|---|---|---|
| Awahou |
Hamurana Road Bridge |
AWA |
1.79 |
1.66 |
1.57 |
|
| Hamurana |
Hamurana Road Bridge |
HAM |
2.96 |
3.08 |
2.48 |
|
| Ngongotaha |
Town Bridge |
NGO |
2.29 |
1.98a |
1.38 |
1.94 |
| Puarenga |
Forest Research Institute |
PUA |
1.92 |
2.05 |
1.89 |
|
| Utuhina |
Lake Road Bridge |
UTU |
2.22 |
2.04 |
1.77 |
|
| Waiohewa |
Rangiteaorere Road Bridge |
WHE |
.48 |
.41 |
.33 |
|
| Waingaehe |
Walkbridge at SH 30 |
WNG |
.29 |
.27 |
.24 |
|
| Waiteti |
Arnold Street Walkbridge |
WTT |
1.51 |
1.39 |
.96 |
|
| Waiowhiro |
Aquarius Drive Recorder |
WWH |
ND |
.41 |
.32 |
a 1.79 m3/s over the whole study 1976-78.
The Waiohewa is a 'hot' stream receiving geothermal flow from the Tikitere geothermal field. It has historically been characterised by high ammonium concentrations that are progressively oxidised to nitrate between Tikitere and the lake, and provides an opportunity for short-term management which is discussed in section 7.
Springs flowing into the Waingaehe, Hamurana and Awahou Streams have relatively high concentrations of phosphorus and also provide opportunities for short-term management, but similar springs flow into most of the other main streams. Hamurana is especially significant for short-term management as it contributes nearly 25% of the total phosphorus load.
Rutherford's draft report shows that phosphorus loads have not varied significantly over the period of monitoring, but that there are relatively high dissolved reactive phosphorus loads in the Waingaehe, Hamurana and Awahou streams. This form of phosphorus is readily available for algal growth and reductions in concentrations in these streams provide opportunities for short-term management.
Rutherford's draft report also shows the high ammonium concentrations in the Puarenga Stream, which have been discussed earlier, and the increasing trend in nitrate in all but the Waiohewa Stream, which is of real concern and has significant management implications.
Groundwater and its relationship to nutrient concentrations
Groundwater plays a dominant role in transporting nutrients from the catchments into the lakes, mostly via supply to the streams. Some of the volcanic soils are quite porous, so that water flows off land uses like dairying and can move down into the soil from recharge areas into relatively deep aquifers. These often reappear years later as springs flowing into streams or directly into lakes.
Aging of these deep groundwater flows shows they are up to 50 to 70 years old, indicating that the high nitrate concentrations in the streams have come from changes in land uses 50 to 70 years ago. EBOP is undertaking work to identify younger groundwater to determine the nitrate concentration trend for more recent land-use changes. This is important work that has significant implications for the medium- and longer-term management measures that focus on changing or modifying existing land uses.
Shallow groundwater generally flows laterally to the nearest stream and can also carry nutrients, especially nitrate. Together, these two groundwater flows provide nearly all of the stream base-flows and hence a significant fraction of the total nutrient inputs to the two lakes.
Storm-flows
While storm-flows appear to only contribute about 10% of the total flows and are more difficult to manage than base-flows, they may be important in causing algal blooms because of a relatively large input of nutrients over a short period. Subsequent weather conditions, especially in late spring, summer and autumn with sunshine and low wind speeds, could cause algal blooms including toxic blue-greens.
These 'event' driven blooms are common in other water bodies around the world and may occur in Lakes Rotorua and Rotoiti, but there is not enough information to determine if this is the case. Storm-flows may also be comparatively important in transporting sediment-bound phosphorus to the lakes.
Catchment nutrient loads, lake water quality and the N:P ratio
Rutherford (Ref 4) has used data from a number of the studies and the trends in the nitrate concentrations to estimate nutrient loads to Lake Rotorua.
Rutherford's draft report clearly shows the increasing trend in total nitrogen and a steady trend in total phosphorus. The former is concerning and has significant implications for management. The latter must not be ignored as it will be important to reduce both phosphorus and nitrogen.
The total load of nitrogen is about 700 tonnes per year and that of phosphorus is about 35 tonnes per year. This produces an N:P ratio for catchment inputs of about 20:1, although this is somewhat artificial as it is the ratio of N:P in the water column when an algal bloom occurs that is important.
Rutherford (Ref 4) elegantly analyses the effect of nutrient concentrations on lake water quality, particularly chlorophyll a and observes that some of the variations from model predictions could be due to factors other than catchment inputs, such as internal nutrient loads and nutrient cycling.
Rutherford has a figure in his draft report showing the ratio of annual average total N to total P in Lake Rotorua from 1970 to 2002. The N:P ratio is around 10 although there are some higher values from 1996 to 1999. As noted earlier, the N:P ratio in the water column at the time of an algal bloom may be more important than an annual average.
The difference between the catchment inputs may reflect inflows that have not been gauged but is more likely to reflect preferential loss of nitrogen from the lake, presumably through denitrification of oxidised nitrogen. This is important as it suggests that the denitrification process may play a key role in the proportion of blue-green algal blooms.
Internal nutrient loads
Internal nutrient loads may be similarly as important to algal blooms as external loads, especially over the summer period when stratification of Lake Rotorua happens and blue-green blooms are more likely to occur. In an average year, Lake Rotorua stratifies for periods of two to 10 days on several occasions each summer, although in the unusually calm and hot summer of 1969-70, it stratified for 20 to 40 days. Lake Rotorua is generally well mixed for the remainder of the year being relatively shallow (20 metres in the deeper basin with a mean depth of 8.9 metres) when compared to Lake Rotoiti (40 to 90 metres with a mean depth of 33 metres). The foldout diagram in Appendix 3 shows these differences in depth along a transect of both lakes.
Lake Rotoiti is strongly stratified from about November to March and, as discussed earlier, the bottom layers are severely de-oxygenated (anoxic). During these stratification events, nutrients, particularly phosphorus, are released from the sediments and may become available for algae growth. This aspect is not well understood, particularly as to the amount of nutrients that may move out of the hypolimnion to become available for algal growth. This is a key question that needs further understanding for both short- and longer-term management.
In Lake Rotorua the nutrients and algae can be transported through the Ohau Channel into Lake Rotoiti and add to the nutrients and algae already in that system.
Rutherford (Ref 4) has graphed the data for internal loads in Lake Rotorua from the existing monitoring data, which on limited data, suggests annual average loads of total phosphorus in the range of 10 to 15 tonnes and of total nitrogen of around 150 tonnes. Rutherford estimates that the internal loads will continue to supply nutrients for at least 20 years even if the external loads were reduced to the targets.
Professor Hamilton and a PhD student have been carrying out further work on this important topic and believe the internal loads could be considerably higher in Lake Rotorua perhaps around one half of the external loads. In other work, Professor Hamilton has shown that a significant fraction of the soluble nitrogen released from the lake bed in Lake Rotoiti is denitrified and lost from the system, thus lowering the N:P ratio.
Rutherford (Ref 4) reports the ratio of N:P from internal loads in Lake Rotorua as 4-5:1, which is quite low when compared to the ratio of about 22:1 to minimise or eliminate blue-green blooms. When coupled with the denitrification occurring in Lake Rotoiti, there is a clear disposition towards N limitation and nitrogen-fixing algal blooms in Lake Rotoiti over summer.
A better understanding of these processes is essential for management of algal blooms and this is discussed further in section 7.
The Ohau Channel and flows into Lake Rotoiti
It has already been stated that Lake Rotoiti is closely linked to Lake Rotorua with greater than 70% of the nutrients, along with sediments and algae, flowing through each year. Some work has been done on the fate of these flows as noted at the beginning of this section, but more work is needed to reach the necessary understanding to trial management options involving diversion of the channel.
Work by Vincent, Gibbs and Spigel, particularly in 1981-82, has shown that in winter (for approximately two-thirds of the year) the colder Lake Rotorua water flows into the bottom of Lake Rotoiti bringing oxygen, nutrients and particulates. It is believed that this winter 'underflow' has a benefit by bringing oxygen-rich water into the bottom waters of Rotoiti, particularly when the lake is stratified in spring and autumn, and therefore that any diversion during this period should be very carefully considered.
In summer, the limited evidence (primarily the studies using dye tracers in 1981-82) is that the warmer 'overflow' predominantly flows into the western arm of Lake Rotoiti and into the Kaituna River. More recent ideas are that there may be other warmer water flows into the surface layers (interflows) of Lake Rotoiti with concurrent transport of nutrients and algae at the best times for blue-green algal blooms (Max Gibbs, Noel Burns and Paul Scholes - pers comm).
A further complication is that the Ohau inflow is generally two to three degrees Celsius warmer when Lake Rotorua stratifies and hence is likely to be a surface overflow rather than an interflow.
Diversion and/or management of the Ohau Channel is a management option that has been considered for a number of years, but the lack of an adequate understanding of the processes discussed above is a major block to any trials.
Algal dynamics and succession
Rutherford's report (Ref 4) and the Rotorua Lakes Water Quality 2002 report both show considerable variations in chlorophyll a levels in Lakes Rotorua and Rotoiti. Some of the variations can be explained by the sewage diversion and increases in catchment nitrate loads, but other peaks are clearly related to shorter-term events (like stratification and destratification episodes) causing nutrient peaks in Lake Rotorua, which may be transported to Lake Rotoiti.
Depending on the time of the year and the weather conditions, these 'events' will provide opportunities for particular algae to bloom. In summer, especially in Lake Rotoiti, the result may be toxic blue-green algal blooms. These event-driven algal blooms are common in other parts of the world and effectively managing them requires an understanding of the events and processes that cause them.
In turn, this usually requires detailed monitoring of the processes and the succession of algae that bloom over the seasons. More intensive studies may also be needed to understand particular processes or interactions. So, for example, the blue-green algal blooms may have started from bottom sediment spores or living cells carried from Lake Rotorua, which bloomed in the western end of Lake Rotoiti and were then transported to the eastern end of the lake by wind or water currents.
Unfortunately, there is not adequate monitoring data to determine what happened, although there is good anecdotal evidence that the main blooms started in the western end of Lake Rotoiti. A better understanding of these processes is needed to focus management measures, but even with the existing knowledge some cosmetic management measures could be trialled.
