Sustainability analysis of the management approach for six New Zealand lakes

Abstract

Jenkins B. 2016. Sustainability analysis of the management approach for six New Zealand lakes. Lake Reserv Manage. 32:101–115.

This paper describes a methodology for sustainability analysis based on failure pathways that can lead to the loss of system sustainability. For these failure pathways, critical variables can be identified by threshold values that define when the system changes from its sustainable state. The methodology also considers any management interventions being undertaken to address failure pathways and whether the extent of intervention is adequate to ensure the critical variables for the failure pathways remain below their threshold values. This methodology was then applied to the current management approaches for 6 New Zealand lakes: Lake Taupo, Lake Brunner, Lake Rotorua, Lake Omapere, Te Waihora/Lake Ellesmere, and Waituna Lagoon. Although diffuse nutrient pollution from land use intensification is affecting water quality of all of the lakes, the analysis identifies different failure pathways and different critical variables, implying the need for different management interventions to achieve desired water quality outcomes. Overall findings were (1) that all lakes will require reductions in land use intensification in their catchments to achieve sustainable water quality; (2) there is evidence of positive innovations that could potentially improve water quality; and (3) the level of management intervention is insufficient to achieve the desired water quality objectives for the lakes.

Keywords:

• critical variables
• failure pathways
• lake water quality
• management interventions
• nested adaptive systems
• sustainability strategies

The concept of sustainability as popularized by the Brundtland Commission (development that meets the needs of current generations without compromising the ability of future generations to meet their own needs; World Commission on Environment and Development 1987) has wide support but provides little guidance on how to make sustainability operational. In the context of this study, operational sustainability is the required management strategies to maintain lake water quality at desired levels. A methodology of sustainability analysis based on nested adaptive systems is defined as a basis for making sustainability operational.

Note that this research addresses a sustainability methodology (i.e., how research into sustainability is conducted and the principles that guide research practices) rather than sustainability methods (i.e., tools and techniques) or sustainability appraisal (i.e., the comparison of alternative actions with respect to sustainability criteria). As Griffioen et al. (2014) noted in discussing tools and concepts for sustainable management of the subsurface, “there is no assessment methodology that can be applied as an instrument for sustainable use.” There are examples of sustainability methods, such as the Gilboa et al. (2014) lake model and water quality assessment tool, to set management policy for a limited set of water quality criteria. Dalal-Clayton and Sadler (2014) recently reviewed the international applications of sustainability appraisal. However, the sustainability methodology in this paper is similar to the frameworks for analyzing social-ecological systems (Binder et al. 2013) and in particular the “Vulnerability Framework” of Turner et al. (2003).

The sustainability methodology we used is based on 7 elements: (1) establishing the adaptive cycle of exploitation, accumulation, disturbance/release, and reorganization for describing the response of biophysical and socioeconomic systems to disturbance; (2) defining socio-ecological systems as linked adaptive cycles of biophysical and socioeconomic systems; (3) specifying nested adaptive systems that operate at different spatial and time scales with linkages between the different scales; (4) using the first 3 steps as a basis, identifying possible failure pathways that can lead to system collapse; (5) defining critical variables on the possible failure pathways and the thresholds associated with system collapse; (6) using step 5 as a basis, identifying potential management interventions to address failure pathways that could lead to system collapse; and (7) combining management actions at multiple scales of, and linkages between, biophysical and socioeconomic systems into a sustainability strategy.

This paper describes the application of the methodology to the effects of land use intensification on water quality of 6 New Zealand lakes: Lake Brunner, Lake Taupo, Lake Rotorua, Lake Omapere, Te Waihora/Lake Ellesmere, and Waituna Lagoon (Fig. 1) with different lake characteristics (Table 1). The issue of water quality degradation of lakes due to diffuse nutrient pollution associated with land use intensification has been recognized for some time in New Zealand (Baber and Wilson 1972, Smith et al. 1993). Calls for addressing the issue also have a long history (Parliamentary Commissioner for the Environment 2004), and substantial research into technical methods have provided possible solutions to the problem (McDowell et al. 2013). Despite this historic knowledge and the availability of technical methods, a review of lake water quality found the poorest water quality was in lakes with catchments having high pasture land cover (Verburg et al. 2010). Modeling future land use change coupled with modeling nutrient loads associated with land use predicts increasing nitrogen (N) loads in nearly all New Zealand regions (Parliamentary Commissioner for the Environment 2013). This situation indicates that achieving sustainability is more than a technical analysis of the biophysical system; it is also a socioeconomic issue with respect to achieving the requisite level of management intervention.

Table 1. Lake characteristics.

Lake	Area (km^2)	Depth (m)	Trophic Level Index^*
Lake Brunner	41	109 (max)	2.9
Lake Taupo	622	60 (max)	2.1
Lake Rotorua	81	45 (max)	4.6
Lake Omapere	12	2 (avg)	6.1
Te Waihora / Lake Ellesmere	200	1 (avg)	5.9
Waituna Lagoon	16	3.3 (max)	4.4

^*Trophic level index (TLI) – a measure of the nutrient status of lakes incorporating total phosphorus concentration, total nitrogen concentration as well as visual clarity and chlorophyll a biomass (Burns et al. 2000).

Figure 1. Location of the 6 New Zealand lakes.

A discussion of results from the sustainability analysis of the 6 lakes is presented indicating that current levels of management intervention need to be increased to achieve sustainable water quality. An approach is proposed for the development of management interventions as a socioeconomic system to address the 4 phases of the adaptive cycle of a biophysical system (i.e., exploitation, accumulation, disturbance, and reorganization).

Methodology

Adaptive cycle

The framework of our methodology is based on considering socio-ecological systems as nested adaptive systems. The first element is the “adaptive cycle,” which describes biophysical and socioeconomic systems in 4 phases (Gunderson and Holling 2002). The first phase is the “exploitation” phase, which is the use of resources from a biophysical or socioeconomic system. This leads to a second phase of “accumulation” in which a build-up of energy or material results from the exploitation of resources. The accumulation phase can be disrupted by a “disturbance” phase that leads to the release of accumulated energy or material and can potentially change the structure and function of the system. Following the disturbance phase is a “reorganization” phase involving the restructuring of the system. System response can be a recovery of the original system or a shift to an alternative system. The phases can be diagrammed as a Lissajous figure (Fig. 2).

Figure 2. Four phases of the adaptive cycle: exploitation, accumulation, disturbance/release, and reorganization.  Adapted from Gunderson and Holling (2002).

This framework provides an operational basis for sustainability, which is the maintenance of the structure, function, and relationships in adaptive cycles across different time and geographical cycles. A key property for sustainability is “resilience,” the capacity of a system to absorb disturbance and still retain its basic function and structure (Walker and Salt 2006).

Lake eutrophication is an example of an adaptive cycle. Land use intensification (i.e., exploitation phase) can lead to a build-up of nutrients in the lake water column or sediments (i.e., accumulation phase). Sufficient increase of nutrients can change the trophic state of the lake (i.e., disturbance phase) leading to increased algal blooms (i.e., release component of the disturbance phase). The lake restructures (i.e., reorganization phase): this can be algal die-off and loss from the lake (i.e., recovery of the original system) or a long-term decline in water quality (i.e., an alternative degraded system). Sustainability is the maintenance of water quality to retain the structure and function of the lake.

Socio-ecological systems as linked adaptive cycles

The second element is considering socio-ecological systems as linked adaptive cycles (Fig. 3) with a biophysical system linked to a socioeconomic system (Berkes et al. 2003). There are 4 types of sustainability issues: (1) the capacity of a biophysical system to adapt to disturbances independent of human activity (pathway 1 and the traditional study of ecological processes); (2) the effects of the socioeconomic system on the biophysical system (pathway 2 and the main emphasis of traditional environmental impact assessment); (3) the effects of the biophysical system on the socioeconomic system (pathway 3, which is receiving negative attention through the resilience of communities to natural disasters and positive attention through the ecosystem services that contribute to social well-being); and (4) the capacity of the socioeconomic system to adapt to socioeconomic disturbances (pathway 4, which includes issues such as the adequacy of institutional arrangements to implement management interventions). The effect of land use intensification on lake water quality is an example of pathway 2.

Figure 3. Four types of sustainability issues: 1) capacity of biophysical system; 2) effects of socioeconomic system on biophysical system; 3) effects of biophysical system on socioeconomic system; and 4) capacity of socioeconomic system.

Figure 4. Lake and catchment as nested adaptive cycles.

Nesting of adaptive cycles

The third element is the nesting of adaptive cycles (Holling et al. 2002). Systems operate at different spatial and time scales, which are linked. Describing eutrophication of lakes as an adaptive cycle must consider at least 2 geographic scales: the lake and the catchment upstream of the lake. The dominant cause of lake eutrophication is the increase in nutrient-intensive land uses in the catchment typically associated with agriculture (Abell et al. 2011). One linkage between the catchment and the lake is through the accumulation of nutrient levels in soils, leading to a release through soil erosion and runoff to nutrient accumulation in the lake downstream (Carpenter et al. 1998). The accumulation of nutrients in the lake and the lake sediments leads to the disturbance of eutrophication in the lake. Furthermore, reorganization of the degraded lake back to a higher water quality level usually requires reorganization in the catchment through a reduction in nutrient intensity of land use in the catchment (National Research Council 1992). Eutrophication of lakes is a nested system with the phases in the adaptive cycle as follows (Fig. 4):

• Exploitation of catchment: increase in nutrient-intensive land uses

• Accumulation in catchment: nutrient build-up in agricultural soils

• Release in catchment: soil erosion and transport of nutrients to rivers and lake

• Accumulation in lake: nutrient build-up in lake sediments and water column

• Disturbance in lake: eutrophication in lake

• Reorganization in lake: degraded lake unless there is a reduction in nutrient-intensive land uses

• Reorganization in catchment: reduction in nutrient-intensive farms.

There can also be other linkages between the catchment and the lake, such as nutrient-laden catchment runoff entering the lake (Carpenter et al. 1998) and nutrient leakage to groundwater from catchment land use entering the lake via the groundwater system (Morgenstern 2007). There can also be different time scales. For example, if the dominant nutrient input from a catchment to the lake is via groundwater, then there can be a considerable time delay between land use change and equilibrium with nutrient input to the lake (Morgenstern 2007).

Table 2. Categorization of failure pathways by type of sustainability issue and geographical scale.

Scale	Biophysical maintenance	Socioeconomic impact on biophysical	Biophysical impact on socioeconomic	Socioeconomic maintenance
Region	Climate variability	Climate change	New technology changing social relationships	Conflict: Hostile neighbours
				Integration: Loss of trade networks
Society	Cumulative depletion of natural resources	Cumulative environmental degradation	Natural disasters	Conflict: Internal warfare
				Integration: Diminishing marginal productivity
Individual	Local natural resource depletion	Local environmental degradation	Disease	Individual commitment to society

Failure pathways

For a society and its natural resource base, theorists like Tainter (1988), Diamond (2005), and Webster (2002) have identified failure pathways for societal collapse, which can be classified as linkages between socioeconomic and biophysical adaptive cycles at different spatial or time scales (Jenkins 2013). A tabular form of these failure pathways (Table 2) is the fourth element of the framework. Lake eutrophication is an example of “cumulative environmental degradation” as a failure pathway.

Critical variables and thresholds

For sustainability analysis of a nested adaptive socio-ecological system, failure pathways that create the greatest vulnerability for system collapse are given priority. Critical variables and thresholds related to system change are the fifth element of the framework. In lake eutrophication, critical variables include the nutrient loads from upstream catchments. Thresholds are the tipping points for critical variables that can change the state or function of a socio-ecological system, such as the lowering of dissolved oxygen (DO) in bottom waters of a lake to a point that triggers the release of nutrients from lake-bed sediments.

Potential management interventions

Critical variables and associated thresholds become the targets for potential management interventions, the sixth element of the framework. There are 3 generic outcomes for management interventions in relation to the management of a natural resource system such as a lake and its catchment (Chapin et al. 2009): (1) no action (or inadequate action) is taken leading to a degraded natural resource system; (2) appropriate action is taken to ensure that the threshold of change is not exceeded and the natural resource system retains its structure and function (i.e. retains its sustainability); and (3) action is taken to transform the system to an alternative state that has a sustainable structure and function.

Sustainability strategies

Chapin et al. (2009) identified 4 generic sustainability strategies to address the vulnerability of adaptive systems: (1) “reduce vulnerability,” which relates to reducing the degree of harm to which a system is exposed; (2) “enhance adaptive capacity,” which refers to the capacity of a system to respond to a disturbance; (3) “increase resilience,” which relates to increasing the ability of a system to absorb a disturbance; and (4) “enhance transformability,” which refers to creating a fundamentally new system which is sustainable. These strategies provide finer-scale approaches that can be classified as components of a nested socio-ecological system, the seventh element of the framework.

Results

The results of the sustainability analysis are presented in 2 parts. The first brings together steps 1, 3, 4, and 5 of the analysis of the biophysical system. It reviews the available scientific information for each lake and its catchment with respect to the disturbance to the lake water quality and the linkages between the lake and the catchment for the source of nutrients that present potential failure pathways for degrading lake water quality and identifies critical variables on these pathways and the thresholds associated with the desired water quality. The second part summarises information relating to steps 2, 6, and 7. It relates the socio-economic links to the biophysical system for each lake, focusing on the management interventions to address water quality failure pathways and whether the combination of interventions provides a basis for a strategy to achieve sustainable water quality.

Failure pathways and critical variables

Lake Brunner

Algal productivity in Lake Brunner is strongly limited by the availability of phosphorus (P) (Rutherford et al. 2008). Land use intensification from dairying has raised concerns that water quality will decline. Mean total phosphorus (TP) concentrations have increased from 5.1 mg/m³ in the early 1990s to 6.1 mg/m³ in the mid-2000s, with comparable increases in chlorophyll a (Chl-a) from 1.1 to 1.4 mg/m³. The lake has a long residence time (1.14 years), leading to relatively high retention of nutrients in the lake (estimated at 52%). Nutrient storage reduces sensitivity to P loading, but only when bottom water is sufficiently oxygenated. When lakes with long residence times have moved beyond nutrient loadings resulting in rates of algal productivity that cause anoxia in bottom-waters, it is very difficult to restore the waterbody to a state with acceptable water quality (Verburg 2009).

Lake Taupo

Bioassays indicate that Lake Taupo is sensitive to N, and algal growth in the lake increases in response to more N. The N load for pre-agricultural development is estimated as 650 t/yr and is currently estimated to be 1360 t/yr. It takes a long time for the effect of intensifying land use to be seen in the lake because the groundwater that carries much of the N from the land can take many decades to reach the lake. Studies estimate that between 20 and 80% of the current amount is yet to reach the lake before equilibrium is reached with current land use (Vant and Smith 2004).

The seasonal pattern of algal biomass indicates that August is typically the peak month. Both the annual mean and maximum values for Chl-a in the upper 10 m of the lake have increased since regular monitoring in 1994 (average 0.5 mg/m³, winter maximum 1.1 mg/m³) to peaks in 2003 (average 1.1 mg/m³, winter maximum 3.0 mg/m³). Values have remained around this level since then or slightly declined.

Lake Rotorua

Water quality deterioration in Lake Rotorua was observed between 1978 and 1983, including substantial algal bloom activity (Bay of Plenty Regional Council 2012b). Sources of nutrients comprise nutrients already in the lake and sediments, nutrients entering the lake, and nutrients in groundwater yet to reach the lake. The current N load entering the lake is estimated to be 755 t/yr, with nearly 80% from pastoral farming. A target N load has been set at 435 t/yr, yielding a Trophic Level Index (TLI) of 4.2. The P load is estimated to be 40 t/yr with a sustainable load of 37 t/yr (Bay of Plenty Regional Council 2012a).

Lake Rotorua has about 3 periods of stratification of 10 days duration each year when oxygen concentrations become zero in the bottom waters. Each stratification event releases the equivalent of about one-third to one-half of the annual incoming nutrient load from bottom sediments (Hamilton et al. 2004, Hamilton and Bruere 2010).

Lake Omapere

Lake Omapere has been affected by land use change and other human actions, including pastoral farming, lowering the lake, and draining wetlands, causing a decline in water quality (TLI = 6.1; 2005–2009 mean; Verburg et al. 2010). Oxygen weed was introduced in the 1970s, covering much of the lake, and collapsed in 1985 causing severe blue-green algal blooms. The lake alternates between algal-dominated and macrophyte-dominated states with periods of low Chl-a (around 10 mg/m³ 1994–1997 and 2008–2012) and periods of high Chl-a (maximum 250 mg/m³ 1998–2008; Gray 2012).

Nutrient loads in the lake are high (e.g., peak in-lake values in 2005–2006 of 2100 mg/m³ total N [TN] and 290 mg/m³ TP). The TN:TP ratio for the lake varies substantially over time, suggesting there is no consistent nutrient limitation. In times of high suspended sediment, light can be the limiting factor. There is an estimated 500 t of TN and 50 t of TP in the top 2 cm of lake bed sediments, which is frequently resuspended into the water column through wind and wave action. Despite recent efforts at riparian management, surface water inputs to the lake remain high, with P peaks after rainfall and N in stream baseflows with concentrations exceeding lake concentrations. DO levels are generally >6 mg/L, a level considered suitable for fish (Northland Regional Council 2007, Gray 2012).

Te Waihora/Lake Ellesmere

Te Waihora/Lake Ellesmere is artificially opened to the sea to manage lake level, originally to manage flooding of farm land, then to include water levels for bird habitat, and more recently to time openings to facilitate migration of longfin eels.

Lake sediment evidence indicates a decline in water quality from the time of forest clearance for agriculture 150 years ago (Kitto 2010). The lake is hypertrophic with high levels of N (TN average ∼2000 mg/m³) and P (TP average ∼200 mg/m³). Phytoplankton growth is light-limited (Chl-a typically 60–90 mg/m³), however, due to high suspended sediment levels from wind-driven resuspension of bed sediments and sediment inputs from stream inflows and lakeshore erosion. The lake does not regularly undergo severe oxygen depletion or toxic algae blooms (Hayward 2009).

The predominant source of nutrients is from tributary streams to the lake (N = 98%, P = 90%). However, these are groundwater-fed streams, so from a catchment perspective infiltration from land use to groundwater is a critical failure pathway. The current N load to the lake is estimated as 2650 t/yr, and the equilibrium load for the 2011 land use is estimated as 4100 t/yr, accounting for the time lag in travel of groundwater to the lake (Selwyn Waihora Zone Committee 2013). Although water quality monitoring shows an overall reduction in nutrient loadings to the lake, this is driven primarily by reduction in flows over the past 10 years rather than a reduction in instream nutrient concentrations.

The brackish nature of the lake (typically 4­6 ppt) results from sea water inflows during lake openings and waves overtopping the gravel bar separating the lake from the ocean. With reduced freshwater inflows to the lake from increased surface and ground water, abstraction has had the perverse effect of reducing salinity in the lake because there are fewer openings to the sea. This is the nature of nested adaptive systems.

Waituna Lagoon

As noted by the Lagoon Technical Group (2011), “through land development of the catchment over the past century (e.g., clearance of wetlands, drainage enhancement, and fertiliser inputs) and an opening regime managed for farm drainage, the lagoon is now experiencing a number of ecological problems. This includes a decline in abundance of Ruppia (seagrass) that is central to the lake's ecological functioning, increased abundance of nuisance filamentous algae, and reduced oxygenation of bed nutrients.”

Estimated nutrient inputs from surface water in the Waituna Lagoon catchment have increased from 179 t/yr in 1995 to 433 t/yr in 2009 for TN, and from 9.7 t/yr (1995) to 21 t/yr (2009) for TP (Hamill 2011). Nutrients from groundwater have been estimated to be 28–48 t/yr TN and 1.4–2.4 t/yr TP (Rissman et al. 2012). Based on dissolved inorganic N to TP ratios (DIN:TP), Waituna Lagoon is probably P-limited (Schallenberg et al. 2010); however, surface-water data from 2001 to 2010 suggest that both N and P could be limiting at different times (Lagoon Technical Group 2011).

Sediment rate monitoring shows elevated rates (2.5–3.0 mm/yr) of fine sediment deposition in localized areas since ∼1960 to present (Lagoon Technical Group 2011). Water clarity as measured by Secchi disk transparency of about 1.2 m is not sufficient for light to reach the bottom of the deeper parts of the lagoon (Hamill 2011). Sediment anoxia has become widespread throughout the lagoon since 2007 (Lagoon Technical Group 2011).

Waituna Lagoon is considered to have a high risk of “flipping” from its high value clear-water Ruppia-dominated state to a highly undesirable turbid algal-dominated (phytoplankton/epiphyte) state due to excessive inputs of N, P, and sediment (Lagoon Technical Group 2011). The reduction of macrophytes to below certain thresholds of biomass or percentage bottom cover facilitates wind-induced sediment resuspension, which further reduces water clarity, often initiating a collapse of macrophyte communities in shallow lakes (Schallenberg and Sorrell 2009).

Lake openings facilitate nutrient flushing, which also increases salinity and adversely affects Ruppia growth. Timing of lagoon openings must ensure closure prior to the main Ruppia growing and germination period (spring–summer; Lagoon Technical Group 2011).

Overview of critical variables

Key critical variables for the 6 lakes were identified in relation to the main water quality failure pathways (Table 3). Although all lakes are subject to nutrient enrichment, the critical variables for the water quality failure pathways are different. In terms of the catchment, the type of nutrient varies (N, P, or both), as does the hydrology pathway (groundwater, surface water, groundwater-fed tributaries). This is related to both the nature of the catchment and the nature of the lake. Sediment can also be a critical variable. In terms of the lakes, different trigger points exist for change of state or function, including DO in bottom waters, algal blooms in upper waters, remobilization of sediments from deoxygenation (in deeper lakes) and wind resuspension (in shallow lakes), lake openings and levels (in coastal lakes), and macrophyte–algae switching.

Table 3. Critical variables on water quality failure pathways for the 6 study lakes.

Lake	Catchment variables	Lake variables
Lake Brunner	P loading in surface runoff	DO in bottom waters
Lake Taupo	N loading in groundwater	Algal blooms in lake
Lake Rotorua	N and P loading from surface and ground water	Algal blooms in lake
		DO in bottom waters
		Remobilization of N and P from sediments
Lake Omapere	N (baseflow) and P (peaks after rainfall) loading from surface water	Toxic algal blooms
		Algae macrophyte switching
		High N and P sediment load mobilized by wind
		DO levels for mussels
Te Waihora / Lake Ellesmere	N and P loading in groundwater-fed tributary streamsSediment from streams	Lake level and lake openingsWind resuspension of sediments
		High N, P and sediment loads
		Loss of macrophytes
Waituna Lagoon	N and P loading from surface water with groundwater contributionSediment from streams	Risk of flipping from macrophytes to algaeLake openingsWind resuspension

Current management interventions

The different critical variables and different failure pathways indicate the need for different management interventions to achieve sustainable water quality. Approaches are currently underway for each lake to assess whether the level of management intervention is likely to achieve sustainable water quality.

Lake Brunner

In 2004, the West Coast Regional Council notified a Proposed Water Management Plan identifying the Lake Brunner catchment as a Special Management Area (West Coast Regional Council 2010) with the objective of improving water quality to reach an average water clarity of 5.3 m by 2020 based on achieving the water quality levels that existed in 2004.

Farm plans were developed with farmers in the catchment in 2005. Some fencing and bridging work was undertaken, but not all participated, and lake nutrients continued to increase (Horrox 2009). Based on trend analysis of TP, research indicated that if the present land uses in the catchment (intensive dairy farming) continued to develop at the same rate using the same land use practices, the transition of the lake from oligotrophic to mesotrophic state would occur in 2040 (Verberg et al. 2013). The Regional Plan contains rules tightening dairy effluent requirements, agricultural land development, controls on access to riparian margins, controls on P applications (limited to 2005–2010 rates), and P water solubility. There are also nonregulatory measures for farm environmental plans with involvement of the New Zealand Landcare Trust and Dairy New Zealand.

However, specific P yields in the catchment are high, 2.4 kg/ha/yr compared to 1 kg/ha/yr for typical farms (Rutherford et al. 2008), and recent water quality monitoring shows an increase in hypolimnetic oxygen consumption (a critical variable for the lake) with the DO level in bottom waters continuing to decline to its lowest recorded level of 5 mg/L (45% saturation; West Coast Regional Council 2011).

Because current monitoring indicates ongoing decline in water quality with thresholds of critical variables being approached, we can reasonably conclude that inadequate action is being taken to achieve sustainable water quality.

Lake Taupo

The key management intervention in the catchment to return Lake Taupo to 2001 water quality levels is the establishment of a market for N discharge allowances (NDAs) for manageable N loads generated by land use intensification. A target has been set for a 20% reduction of current N leaching (Waikato Regional Council 2011).

The Lake Taupo Protection Trust has the task of permanent removal of this 20% of manageable N (186 of the 930 t/yr), to be achieved through land use change to lower N-leaching land uses. A public fund of NZ$81.5 million (about US$52.9 million at January 2016 exchange rates) with contributions from central, regional, and district government is managed by the trust, which can apply the fund to purchase N either by purchasing land or NDAs (Lake Taupo Protection Trust 2013). As of July 2011, the trust has permanently removed 100 t of N from the catchment through 13 completed deals, 5 consisting of whole farm purchases and 8 of NDA purchases (Duhon et al. 2011).

Because of the time lag of groundwater from past land uses, the goal of achieving improved water quality has been set for 2080; however, indications are that a 20% reduction in N leaching from current land use will be insufficient to reduce the catchment load to meet 2001 water quality levels. Scientific estimates for the exact percentage of the expected groundwater load range from 30 to 41% of current manageable load, with other estimates as high as 80% (Hadfield et al. 2007). These higher estimates mean greater reductions in N (and greater cost in purchasing land and NDAs) would be needed to meet the water quality target.

Lake Rotorua

For management interventions in Lake Rotorua, the Rotorua Lakes Strategy Group (Te Arawa Lakes Trust, Rotorua District Council, and Bay of Plenty Regional Council) prepared a Lakes Rotorua and Rotoiti Action Plan with an Action Plan Working Group of community and stakeholder organizations (Burns et al. 2009, Environment Bay of Plenty et al. 2009). The action plan is based on limiting catchment N loads to 435 t/yr and P loads to 37 t/yr; catchment inputs are currently 556 and 39 t/yr, respectively. With groundwater lag effects, the N load in equilibrium with current land use is estimated at 746 t/yr. In addition, nutrients retained in lake bed sediments are estimated to release 360 t/yr of N and 36 t/yr of P.

Management interventions in the Lakes Rotorua and Rotoiti Action Plan to achieve the reductions in N are heavily dependent on land management improvements in the catchment, with reductions of 30 t/yr through adoption of best management practices (by 2012), 56 t/yr through adoption of new but known technology (by 2019) and 84 t/yr by innovative (i.e., yet to be developed) new technology (by 2029). As of June 30 2013, reductions of 4.6 t of N and 0.24 t of P had been achieved Bay of Plenty Regional Council/Te Arawa Lake Trust (2015; Lake Rotorua achievements: www.rotorualakes.co.nz/lake_rotorua_achievements).

Another major potential contributor to N reduction was the diversion of Hamurana Stream from the lake to the Ohau channel downstream of the lake, with an estimated N reduction of 53–92 t/yr. Phosphorus reduction was focussed on lake-bed sediment treatment (25 t/yr) and flocculation of P in tributary streams (4 t/yr). Phosphorus locking has been successful, with 8.3 t of P removed, exceeding expectations (Environment Bay of Plenty et al. 2009).

Current implementation programmes are expecting a 320 t/yr N reduction from the catchment, with 270 t/yr from changes in pastoral land use and 50 t/yr from engineering solutions. This level would require a 51% reduction in relation to current N losses from pastoral land use. Farm interests have estimated that the farm costs amounted to $88 million and would result in a loss of farm value of $35 million (Omundsen 2013).

The 2008 Regional Plan (Rule 11) put in place nutrient benchmarks on pastoral properties mainly based on 2004–2005 land use. There was to be no net increase of the export of N and P, or the increase had to be offset on the property or within the same catchment. The Proposed Regional Policy Statement (RPS) includes the cap of 435 t/yr for N inputs to the lake, but was appealed by Federated Farmers in part because it did not address the economic consequences on land owners and imposed unachievable targets on the community. It also sought a collaborative approach to addressing water quality issues (Federated Farmers of New Zealand 2012).

A Funding Deed was agreed to among central, regional, and district governments for lake water quality improvements in the region, with $45.5 million allocated to Lake Rotorua. An implementation schedule based on the Lakes Rotorua and Rotoiti Action Plan has been developed with annual reports on progress. The most recent progress report indicated that success of flocculation of P in tributary streams was achieving greater P removal than predicted. It also noted that progress on targeted nutrient reduction from land use was behind schedule and land use negotiations were on hold until an integrated rules and incentive fund were developed. The report also indicated an improvement in the TLI for Lake Rotorua (down to 4.06), attributed to in-lake interventions and favorable climatic conditions. However, N and P levels continued to increase and clarity to decline (Bay of Plenty Regional Council et al. 2013).

An agreement has been reached between Federated Farmers, the Lake Rotorua Primary Producers’ Collective, and regional council (the Oturoa Agreement) in February 2013 to meet the N cap by 2032 with 70% of the required reduction reached by 2022 (the original RPS deadline). Further negotiations through the Rotorua Lakes Stakeholder Advisory Group have led to an agreement to achieve pastoral land use reductions for N of 270 t/yr through a combination of rules (140 t/yr), incentives (100 t/yr), and gorse removal (30 t/yr). This would require dairy farms to reduce N leaching to 35 kg/ha/yr and drystock farms to reduce N leaching to 13 kg/ha/yr, compared with current estimated averages of 54 kg/ha/yr for dairy and 15.7 kg/ha/yr for drystock farming (Omundsen 2013).

Although the proposed management interventions have uncertainties because they are dependent on new technologies, uncertain costs, modeled approaches, and implementation of major land use and land management changes, they represent a significant program to achieve sustainable water quality objectives.

Lake Omapere

Lake Omapere was vested to the Lake Omapere Trust in 1955. The Trust represents the Ngāi Puki-nui-toui, the local iwi (Māori tribe) for the region. Recent management interventions for the lake include the introduction of grass carp (Ctenopharyngodon idella) in 2000 and 2002 to address oxygen weed (Egeria densa), which at times covered the entire lake (e.g., in 1999), and a $0.6 million restoration and management project (2003–2010) administered by the Lake Omapere Project Management Group. The project included the voluntary adoption of farm plans, fencing and riparian planting, and a voluntary lake strategy (Northland Regional Council 2013).

Lake water quality has improved since 2007, but this is most likely a result of a natural switch between algal-dominated and macrophyte-dominated states rather than the project and restoration efforts. The available data suggest lake sediments still contain high nutrient levels that provide an internal nutrient source through wind resuspending sediment into the water column. The data also suggest that external inputs into the lake have not improved; nutrient levels in catchment streams and drains are still high. Mussel numbers are stable in the lake and, because they can filter algae from the water column, are likely to be one of the main reasons for the improving lake water quality (Gray 2012).

Waituna Lagoon

For the management of Waituna Lagoon, the regional council has formed a Waituna Partners Group of government agencies, industry groups, and community groups to sustainably manage the lagoon (Environment Southland 2013). The Lagoon Technical Group of scientists has been formed to advise on water quality and lagoon processes, the Lake Waituna Control Association manages lake openings, and a Waituna Liaison Committee engages stakeholders.

Drainage enhancement and rock protection works have been undertaken to reduce sediment loads in tributaries, and lake openings are managed for local flooding and nutrient flushing. The dairy industry has prepared a Sustainable Milk Production Plan for farmers to identify on-farm actions to address key environmental outcomes, and it supported a dairy farmer initiative for a Waituna Lagoon and Catchment Action Plan. Winter grazing trials have been initiated to reduce overland flow and sediment losses (Environment Southland 2013).

Plan Change 13 to the Region Water Plan proposes to require consent for new dairy farms, and the regional council is developing a regional strategy with community involvement called Water and Land 2020 in response to National Policy Statement requirements. However, as Scanes (2012) summarizes in the last 10 years the ecological condition of Waituna has been in rapid decline, changing from a high value Ruppia-dominated state to a more degraded condition with nuisance epiphyte and algal blooms. Consequent sediment anoxia is causing additional stress to the keystone Ruppia species (Scanes 2012).

Current expert opinion is that unless urgent intervention occurs, the lagoon will almost certainly undergo a rapid change in state to an even more degraded phytoplankton-dominated system (e.g., with algal blooms), which would endanger the Ruppia community and change the fundamental values and character of the lagoon (Lagoon Technical Group 2011). The Scanes (2012) analysis of catchment loads from similar lagoons in New South Wales suggests that loads required to maintain a moderate environmental quality (some eutrophic symptoms but still supporting healthy seagrass and fish communities) would be a total N load of 9 t/km²/yr and a TP load of 0.57 t/km²/yr, representing a 52% reduction in TN load and a 23% reduction in TP load from 2010 conditions.

Te Waihora/Lake Ellesmere

The most significant management intervention for Te Waihora/Lake Ellesmere is the rehabilitation programme for the lake, Whakaora Te Waihora, a $12 million program supported by central and regional governments, Ngāi Tahu (Māori tribe for most of the South Island), and Fonterra (Te Waihora Co-Governance Group 2011). The lake opening regime has also been changed to incorporate longfin eel passage as well as wader habitat and to manage flooding of surrounding farmland.

Ngāi Tahu owns the lake bed under the Ngāi Tahu Settlement Act, and a Joint Management Plan was created with the Department of Conservation for the lake bed (Department of Conservation and Te Runanga o Ngai Tahu 2005), as well as a governance agreement between Ngāi Tahu and the regional council. Local organizational arrangements include the Lake Settlers Association involved in lake opening decisions and the Waihora Ellesmere Trust representing community interests. The Canterbury Water Management Strategy involves the Region Committee developing a Regional Implementation Programme and the Selwyn Waihora Zone Committee developing a Zone Implementation Programme (ZIP).

In the catchment, the Canterbury Water Management Strategy provides the regional framework for addressing land use intensification and water quality issues, and the Selwyn Waihora ZIP provides recommendations for managing the lake catchment, which is to receive statutory backing through the Regional Land and Water Plan.

Investigations for the ZIP have estimated the current N load from the catchment to the lake to be 2650 t/yr, with a time lag for further groundwater input to achieve equilibrium with current (2011) land use of 4100 t/yr of N. With the Central Plains Project and other intensification that can occur in the catchment, the N load is estimated to reach 5600 t/yr, and intensification would add an estimated $300 million in regional GDP (Selwyn Waihora Zone Committee 2013).

The Zone Committee has proposed adopting a solutions package to achieve farm management improvements 12.5% better than “good management practice.” This solution is designed to achieve a reduction in catchment N load to 4800 t/yr and is estimated to reduce regional GDP by $30 million. However, aspirations to achieve a macrophyte lake would require reducing the catchment N load to 800 t/yr (Selwyn Waihora Zone Committee 2013).

Discussion

Adequacy of interventions and emerging approaches

The management interventions were identified in relation to physical activities, regulatory activities, and organizational arrangements for the lakes and their catchments (Table 4). The analysis shows that while management interventions are occurring to reduce nutrient loads in each catchment, they are insufficient to achieve sustainable water quality outcomes for Lake Brunner, Lake Taupo, Lake Omapere, Waituna Lagoon, and Te Waihora/Lake Ellesmere. For Lake Rotorua, an ambitious nutrient target has been set and agreed to by key stakeholders to achieve sustainable water quality, but uncertainties exist with developing new technologies, costs, reliance on modeling, and ability to implement. Increased interventions and activities are needed to support interventions in order to achieve sustainable water quality in lake systems. This has also been observed by Abell and his colleagues, who stated that, “without a comprehensive shift towards sustainable land use practices, it is likely that water quality decline will continue for the foreseeable future” (Abell et al. 2011).

Table 4. Summary table of current management interventions.

Lake	Physical activities	Regulatory activities	Organizational arrangements
Brunner	Voluntary farm plans	Revised Regional Plan^1	NZ Landcare Trust
		P application limits	Dairy NZ farm plans
Appraisal: Inadequate action to achieve sustainability
Taupo	100 t of trades with Trust	Nitrogen discharge allowancesCap and Trade market in NDAs^2	Lake Taupo Protection Trust ($81.5 million)
		20% target reduction in N	“Partnerships of Innovation”
Appraisal: N load >20% in groundwater
Long-term return to 2001 levels by 2080 needs further intervention
Rotorua	Lake bed treatment	Nutrient benchmarks	Te Awara Lakes Trust
	P flocculation in streams	Reduce N: 755 to 435 t/y	Rotorua Lakes Strategy Group
	Best management practicesNew technologies	Deed of funding agreement ($45.5 million)	Stakeholder Advisory GroupLakes Water Quality Society
	Land use change to low N activities	Investigate nutrient trading	Chair in lake managementOturoa agreement
Appraisal: Significant programme underway to achieve catchment nutrient reduction and lake sediment source reduction: uncertainties with new technologies, costs, models and implementation
Omapere	Carp introduction to address	Voluntary lake strategy	Lake Omapere Trust
	oxygen weed		Lake Omapere project
	Voluntary farm plansFencing & riparian planting		management Group
Appraisal: Recent water quality improvements due to macrophyte/algae cycling
Tributary nutrient levels still exceed lake nutrient levels
Inadequate actions to achieve sustainability
Waituna	Drainage works	Region Plan Change:	Waituna Partners Group
	Sustainable milk production plansWinter grazing trialConstructed wetlands	consents for dairy farmsWater and Land 2020	Lagoon Technical GroupWaituna Liaison CommitteeLake Waituna Control Assn.
Appraisal: Load of 19 t/km2/y of N to be reduced to 9
Load of 0.74 t/km2/y of P to be reduced to 0.57
Lake openings to flush nutrients limited by salinity effects on seagrass
Inadequate actions to achieve sustainability
Te Waihora	Whakaora Te WaihoraChange in lake opening regime	Canterbury water management strategyZone implementation programmeRegional Land and Water Plan	Region CommitteeZone CommitteeNgāi Tahu / ECan agreementLake Settlers AssociationWaihora Ellesmere Trust
Appraisal: Current N load 2650 t/y, 2011 land use equilibrium load 4100 t/y
Central Plains and other intensification: 5600 t/y
Proposed solution package: 4800 t/y ($30 million reduction in regional GDP)
Inadequate action to achieve macrophyte lake of 800 t/y

¹Items in italics are interventions in progress. ²NDAs = nitrogen discharge allowances.

Figure 5. Management intervention pathways to achieve sustainability.

Also evident is that new approaches are emerging that could provide the basis to improve water quality. There are a variety of technical innovations such as farm management plans, improvements in land management, riparian plantings, constructed wetlands, lake bed treatments, scientific investigations, water quality models, and improved monitoring. There are also indications of incorporating Māori resource stewardship concepts into management approaches, in particular, the restoration of mauri (maintenance of life-giving essence of a resource) for Lake Omapere and mahinga kai (maintaining healthy food systems) for Te Waihora/Lake Ellesmere.

An important component of the technical innovations for management intervention decision-making is the use of modeling. Much of the initial modeling was related to replicating the lake response to increased nutrients (i.e., failure pathways). However, modeling to predict the outcomes of management interventions (i.e., sustainability strategies) is increasing to include nutrient load reductions from changing land management practices, transport of nutrient accumulations in groundwater, tipping points for managing system resilience, and effects of rehabilitation program. In addition to biophysical modeling is financial modeling of the implications of changing farm management practices and modeling changes to regional economies.

There are also regulatory changes such as catchment nutrient limits, nutrient discharge allowances, and consent controls, as well as non-statutory strategies.

Of particular interest is the importance of collaborative organisational arrangements. This is evident in a variety of forms. These included intergovernmental partnerships, iwi-government governance and management agreements, strategy groups among key stakeholders, stakeholder advisory groups, community engagement mechanisms, Region and Zone implementation committees, technical advisory groups, and funding Trusts and financial Deeds of Agreement.

Management interventions to achieve sustainability

The analysis for sustainable lake management highlights the value of considering the lake and its catchment as a nested socio-ecological system. Both the biophysical pathways that degrade water quality and the socioeconomic system that develops and implements management interventions must be considered.

Management interventions in the biophysical system of water resources can occur at each of the phases of the biophysical adaptive cycle. In the exploitation phase, interventions are designed to reduce pressure on the resource ("reducing vulnerability"), such as reducing catchment N loads on a lake. In the accumulation phase, interventions are designed to address legacy issues of accumulated changes in the past ("enhancing adaptive capacity"), such as lake bed treatment to reduce remobilization of P. In the disturbance/release phase, interventions are designed to increase the resilience of systems to accommodate disturbance ("increasing resilience"), such as lake aeration to prevent stratification. Finally in the reorganization phase, interventions are designed to rehabilitate adverse effects of the system ("enhance transformability"), such as the reestablishment of macrophytes in a lake. This process can be shown diagrammatically (right side of Fig. 5) for interactions in the different phases of the biophysical system. The types of management interventions (i.e., reducing pressure, addressing legacy issues, increasing resilience, and, rehabilitating adverse effects) are consistent with terminology developed in a series of New Zealand workshops for the establishment of a national Centre for Research Excellence for water management in New Zealand.

Identifying the biophysical component of management interventions is not enough; the socioeconomic framework is also required to implement the needed management interventions. In the terminology of the adaptive cycle, this process involves the following steps: first, the use of human and economic resources for stakeholder, cultural, and community engagement to consider how to collaboratively address the issue of sustainability of our water resources as well as the investment of technical resources to understand the issues and financial resources to undertake actions (i.e., exploitation phase); second, the accumulation of knowledge, social, cultural, and economic capital to develop integrated approaches to sustainable strategies (i.e., accumulation phase); third, the formulation of new approaches to water management that change existing practices (i.e., disturbance/release phase); and fourth, the development of new institutional arrangements to implement the new approaches to water management (i.e., reorganization phase).

This framework has the potential to lead to the adoption of management interventions to the biophysical system of water resources that achieve sustainability. Similarly, failure to adopt appropriate management interventions will lead to ongoing degradation of water resources, as the 6 case studies in this study demonstrate. The adaptive cycle for implementing management interventions can be diagrammed as a Lissajous figure for the socioeconomic system (left side of Fig. 5). Linking the socioeconomic system and the biophysical system provides an overall framework for management intervention pathways to achieve sustainability (Fig. 5), highlighting the need to consider sustainability issues as nested socio-ecological systems.

Although this study provides evidence of positive innovations in water quality management, all of the lakes considered require reductions in land use intensification in their catchments to achieve sustainable water quality, and greater management intervention is needed to achieve sustainability. Maintaining our natural capital also requires building our social, cultural and economic capital to develop and implement management interventions to achieve sustainability.