2018-2019 Annual environmental monitoring summary for the ...€¦ · Lauren Fletcher APPROVED FOR RELEASE BY: Grant Hopkins ISSUE DATE: 4 June 2019 RECOMMENDED CITATION: Bennett

REPORT NO. 3333

2018-2019 ANNUAL ENVIRONMENTAL MONITORING SUMMARY FOR THE NGAMAHAU BAY SALMON FARM

CAWTHRON INSTITUTE | REPORT NO. 3333 JUNE 2019

2018-2019 ANNUAL ENVIRONMENTAL MONITORING SUMMARY FOR THE NGAMAHAU BAY SALMON FARM HOLLY BENNETT, EMILY MCGRATH, CARLOS CAMPOS, EMMA

NEWCOMBE, MAXIMILIAN SCHEEL, DEANNA ELVINES

Prepared for The New Zealand King Salmon Co. Ltd.

CAWTHRON INSTITUTE 98 Halifax Street East, Nelson 7010 | Private Bag 2, Nelson 7042 | New Zealand Ph. +64 3 548 2319 | Fax. +64 3 546 9464 www.cawthron.org.nz

REVIEWED BY: Lauren Fletcher

APPROVED FOR RELEASE BY: Grant Hopkins

ISSUE DATE: 4 June 2019

RECOMMENDED CITATION: Bennett H, McGrath E, Campos C, Newcombe E, Scheel M, Elvines D 2019. 2018-2019 Annual environmental monitoring summary for the Ngamahau Bay salmon farm. Prepared for the New Zealand King Salmon Co. Ltd. Cawthron Report No. 26 p. plus appendices.

© COPYRIGHT: This publication must not be reproduced or distributed, electronically or otherwise, in whole or in part without the written permission of the Copyright Holder, which is the party that commissioned the report.


i

TABLE OF CONTENTS

1. BACKGROUND ............................................................................................................. 1

2. KEY SAMPLING DETAILS AND RESULTS ................................................................... 2

2.1. Soft sediments ............................................................................................................................................... 2 2.1.1. Type 2 (annual) monitoring ....................................................................................................................... 2 2.1.2. Type 3 monitoring ..................................................................................................................................... 8

2.2. Water column ............................................................................................................................................... 12 2.2.1. Water column monitoring results ............................................................................................................ 14

2.3. Light effects monitoring ................................................................................................................................ 17 2.3.1. Summary of observations ....................................................................................................................... 17

3. ELIGIBILITY FOR FEED INCREASE ........................................................................... 18

3.1. Additional consideration - far-field effects .................................................................................................... 21

4. KEY FINDINGS ........................................................................................................... 22

5. REFERENCES ............................................................................................................ 24

6. APPENDICES ............................................................................................................. 27

LIST OF FIGURES

Figure 1. Monthly feed and nitrogen inputs at the Ngamahau Bay salmon farm for the 12 months preceding soft-sediment sampling (February 2018–January 2019). ..................... 1

Figure 2. Soft-sediment sampling locations at the Ngamahau Bay salmon farm site. ...................... 3 Figure 3. Time series of monthly feed discharge (tonnes, shown by shaded area under curve)

and maximum Enrichment Stage (ES) score (indicated by individual symbols) for each annual monitoring event at the Ngamahau Bay salmon farm since the farm was established .......................................................................................................................... 5

Figure 4. Approximate locations of Type 3 soft-sediment sampling stations for the Ngamahau Bay salmon farm. ................................................................................................................ 9

Figure 5. Water column sampling surveys for the three water column monitoring types (routine, full-suite and fine-scale), during 2018. ............................................................................. 12

Figure 6. NZ King Salmon and MDC routine and full-suite water quality monitoring stations in Tory Channel .................................................................................................................... 13

LIST OF TABLES

Table 1. Average Enrichment Stage (ES) scores and 95% confidence intervals (95% CI) calculated for indicator variables, and overall, for each of the Ngamahau Bay salmon farm stations sampled in February 2019 ............................................................................ 4

Table 2. Summary of visual assessment and indicator variables measured for each of the Ngamahau Bay salmon farm stations during the February 2019 monitoring survey ......... 6

Table 3. Total recoverable copper and zinc concentrations (mg/kg dry weight) in bulk sediment from Ngamahau Bay pen station samples, February 2019. ............................................... 8

Table 4. Water column sampling stations for the routine, full-suite and fine-scale monitoring components ...................................................................................................................... 13

JUNE 2019 REPORT NO. 3333 | CAWTHRON INSTITUTE

ii

Table 5. Summary of water column compliance for parameters measured at each of the Ngamahau Bay salmon farm monitoring stations ............................................................. 16

Table 6. Summary of consent conditions required to be met in order for the Ngamahau Bay salmon farm (NGA) to qualify for a feed increase. ........................................................... 19

LIST OF APPENDICES

Appendix 1. Methodology for soft-sediment sampling .......................................................................... 27 Appendix 2. Comprehensive discussion of results of the February 2019 soft-sediment monitoring

survey at the Ngamahau Bay salmon farm (NGA). .......................................................... 33 Appendix 3. Water column sampling methodology and compliance framework. ................................. 47 Appendix 4. Additional detail on the results of the 2018 Ngamahau Bay (NGA) salmon farm water

column monitoring. ........................................................................................................... 54 Appendix 5. Time series plots for eutrophication indicators, collected as part of the ongoing

monitoring programme ..................................................................................................... 66 Appendix 6. Datasheet used for underwater lighting observations. ..................................................... 76


1

1. BACKGROUND

This report presents the environmental monitoring results for the Ngamahau Bay

(NGA) salmon farm located in Tory Channel (consent number U140296). The NGA

farm was established in November 2015, making this the fourth annual monitoring

report for this site. Data presented include an assessment of depositional effects on

soft-sediment habitats, effects on the water column and effects of underwater lighting

on night-time feeding activity by fish, seabirds and marine mammals. Results from

reef habitat monitoring are reported separately in Dunmore (2019).

In terms of hydrodynamics, NGA is assessed as a high-flow site. The average mid-

water current speeds are 22 cm/sec. Water depth at the farm site is c. 30–35 m, with

the net pens extending from the surface to a depth of c. 20 m.

A total of 1,314 tonnes of feed was discharged at the NGA site in 2018 (similar to the

total feed discharged in 2017, shown by month in Figure 1). The highest monthly feed

input in the 12 months prior to sampling was during December (195 tonnes), while the

lowest was during May (72 tonnes). Nitrogen input averaged 7.8% of feed input from

January to December 2018 (range: 5.7 to 15.9 tonnes per month), totalling 103

tonnes for this period. In accordance with condition 34 of the farm resource consent,

the maximum annual tonnage of nitrogen that may be discharged in any year is 7% of

the maximum feed tonnage (1500 tonnes at NGA), equating to 105 tonnes of

nitrogen. Nitrogen input for 2018 was therefore within the maximum nitrogen

discharge allowable in any year.

Figure 1. Monthly feed and nitrogen inputs at the Ngamahau Bay salmon farm for the 12 months

preceding soft-sediment sampling (February 2018–January 2019). Feed and nitrogen input data provided by NZ King Salmon.


2

2. KEY SAMPLING DETAILS AND RESULTS

An overview of the key sampling details and results are provided in this section. More

comprehensive discussion of methodology and monitoring results are provided in the

relevant appendices.

2.1. Soft sediments

2.1.1. Type 2 (annual) monitoring

Annual soft-sediment monitoring at NGA was undertaken on 13 February 2019.

Sampling stations comprised three stations immediately adjacent to the net pens:

Pen 1, Pen 2 and Pen 3, as well as two stations to monitor enrichment within the

outer limit of effects: 300 N and 300 S (see Figure 2). Although not a requirement

under the Best Management Practice (BMP) guidelines, one station (75 N) was also

sampled at the Zone 2/3 boundary, 75 m along the north transect, to monitor the

enrichment footprint in the early stages of operation.

Three reference or ‘control’ stations were sampled, one near-field (TC-Ctl-1) and two

far-field (TC-Ctl-3 and TC-Ctl-6).

Sediments at all stations were assessed for organic content, redox potential, total free

sulphides and infauna community metrics (see Appendix 1 for all sampling details). In

addition, copper and zinc concentrations were also measured beneath the net pens.

The environmental monitoring results from soft-sediment habitats are used to

determine whether the farm is compliant with the benthic environmental quality

standards (EQS: benthic) specified in the consent conditions and the best

management practice guidelines developed for salmon farming in the Marlborough

Sounds region (see Appendix 1 for EQS: benthic).


3

Figure 2. Soft-sediment sampling locations at the Ngamahau Bay salmon farm site. ‘TC-Ctl’ = Tory

Channel Control. Position accuracy is ± 5 m.

Enrichment to soft-sediment habitats near Ngamahau Bay salmon farm

A summary of key findings is provided below, while detailed monitoring results are

provided in Appendix 2.

Measured levels of enrichment beneath the pens were within the allowable

Enrichment Stage (ES) scores (i.e. ES ≤ 5) at all three sampling stations (Table 1).

ES scores at all pen stations have decreased since the previous monitoring round

(see Figure 3 where ES scores are shown with feed levels over time). The average

overall ES scores at the three pen stations were ES 2.7, 2.7 and 2.9 at Pens 1, 2 and

3, respectively, indicating moderate enrichment levels. This was evident through both

sediment chemistry and macrofaunal community measures (Table 2 summarises all

observations for the NGA sites and see Appendix 2, Figure A2.2 for a comparison of

results between sites).

Further from the pens, ES scores at the 75 N (Zone 2/3 boundary; ES 3.0) and outer

limit of effects (300 N and 300 S, OLE; ES 1.9) remained unchanged from the

previous survey (February 2018) and were within the EQS (i.e. ES < 3.0 and ≤ 4.0,

respectively). The overall ES score at 75 N is higher than ES scores at the pen

stations. While ES scores at the OLE stations were the same as the previous year,

community composition and sediment chemistry suggest enrichment effects have

occurred at these sites. Macrofaunal abundance at the 300 S OLE station has

increased two-fold since the previous survey (average 869 individuals per core as


4

compared to 428 individuals per core in 2018). Meanwhile total free sulphides at the

300 N station have increased c. two-fold since last year.

We note that under the BMP guidelines, background / natural conditions are

assessed as enrichment stage (rather than individual variables), and the industry

operational goal is for the OLE to be ES < 3.0. In the context of ES scores both OLE

stations are compliant (i.e. ES < 3.0)1. The NGA consent requires that ES < 3.0 is

maintained at the OLE and that conditions remain statistically comparable with

relevant / appropriate reference stations. Conditions (assessment as ES scores as

per the BMP guidelines) at both OLE stations are statistically comparable with

relevant / appropriate reference stations2.

Table 1. Average Enrichment Stage (ES) scores and 95% confidence intervals (95% CI) calculated for indicator variables, and overall, for each of the Ngamahau Bay salmon farm stations sampled in February 2019. Full breakdowns of indicator variable contributions are provided in Appendix 2. All stations were compliant.

Station Organic

loading ES Sediment

chemistry ES Macrofauna

ES Overall

ES

Compliant with EQS?

Pen 1 2.3 (0.7) 2.8 (0.3) 2.7 (0.1) 2.7 (0.2) ✓

Pen 2 2.3 (0.7) 2.8 (0.8) 3.0 (0.3) 2.9 (0.4) ✓

Pen 3 2.0 (0.0) 2.3 (0.2) 2.9 (0.4) 2.7 (0.3) ✓

Zone of maximal effect (ZME); EQS ≤ 5.0

75 N* 1.0 (0.0) 2.8 (0.8) 3.3 (0.2) 3.0 (0.3) ✓

Zone 2/3 boundary; EQS ≤ 4.0

300 N 1.0 (0.0) 2.5 (0.6) 1.9 (0.3) 1.9 (0.1) ✓

300 S 1.0 (0.0) 2.3 (0.3) 1.9 (0.2) 1.9 (0.2) ✓

Outer limit of effects (OLE); EQS < 3.0

*Note that sampling at this station is not a requirement under the BMP guidelines.

1 See Section 2.1.2 for discussion relating to the use of ES <3.0 as a proxy for “natural conditions”. 2 300 S: one-way PERMANOVA: Pseudo-f (3, 11) = 0.96, p = 0.45, 300 N: one-way PERMANOVA: Pseudo-f (3,

11) = 1.3, p = 0.32.


5

Figure 3. Time series of monthly feed discharge (tonnes, shown by shaded area under curve) and maximum Enrichment Stage (ES) score (indicated by

individual symbols) for each annual monitoring event at the Ngamahau Bay salmon farm since the farm was established. ES scores reported are maximums recorded in: the zone of maximal effect (ZME)/Pen stations (pink diamond symbol), the outer limit of effects (OLE)/300 m stations (blue cross symbol), and relevant Tory Channel reference stations (coloured dots). The consented environmental quality standards (EQS) for the ZME (ES 5) and OLE (ES 3) are shown as red dashed lines. Feed data were provided by NZ King Salmon.


6

Table 2. Summary of visual assessment and indicator variables measured for each of the Ngamahau Bay salmon farm stations during the February 2019 monitoring survey. All farm comparisons are made to the TC-Ctl-1, TC-Ctl-3 and TC-Ctl-6 reference station values, unless otherwise stated. Reference station comparisons are made to the 2017-2018 values (Bennett et al. 2018b). %OM = percent organic matter. See Appendix 2 for representative images of the soft-sediment habitat at each site.

Farm Station

Bacteria Out- gassing

Observed epifauna Other observations Organic loading

Sediment chemistry Macrofauna

Pen 1 None No Snake star, anemone, cushion star

Dark grey, coarse sediments

%OM marginally elevated

Redox slightly negative, sulphides elevated

Total abundance elevated (average 1,718 individuals per core). Taxa richness slightly elevated (59-65 taxa per core). Moderate community compositional changes.

Pen 2 None No Hermit crabs Dark grey course sediments, fish faeces or feed pellets



Total abundance variable but average is very high (3,808 individuals per core). Taxa richness elevated (51-70 taxa per core). Moderate community compositional changes.

Pen 3 None No

Snake and cushion stars, apricot sea star, 11-armed sea star, sea cucumber, sea tulip, blue and green-lipped mussels

Dark grey, coarse sediments


Redox normal, sulphides elevated

Total abundance (average 1,846 individuals per core) and taxa richness marginally elevated (49-57 taxa per core). Moderate community compositional changes.

75 N None No Snake stars, finger sponge, attached and drift macroalgae

Dark grey sandy sediment with shell hash.

%OM normal


Total abundance elevated (average 1,253 individuals per core). Slightly reduced taxa richness in some samples (29-55 taxa per core). Marginal community compositional changes.

300 N None No

Shell hash, drift/attached macroalgae, sparse diatom mat coverage

%OM normal Redox slightly positive, sulphides elevated

Total abundance comparable (average 229 individuals per core). Taxa richness slightly reduced (38-41 taxa per core). No major change in community composition.

300 S None No Shell hash, drift/attached macroalgae

%OM slightly elevated

Redox and sulphides normal

Total abundance high (average 870 individuals per core). Taxa richness elevated (55-79 taxa per core). Marginal community compositional changes.


7

Table 2. continued. Summary of visual assessment and indicator variables measured for each of the Ngamahau Bay salmon farm stations during the February 2019 monitoring survey. All farm comparisons are made to the TC-Ctl-1, TC-Ctl-3 and TC-Ctl-6 reference station values, unless otherwise stated. Reference station comparisons are made to the 2017-2018 values (Bennett et al. 2018b). %OM = percent organic matter. See Appendix 2 for representative images of the soft-sediment habitat at each site. %OM = percent organic matter.

Reference station

Bacteria Out- gassing

Observed epifauna Other observations Organic loading

Sediment chemistry Macrofauna

TC-Ctl-1 None No Snake stars Fine sand, diatom mat coverage, burrow holes, trail marks

%OM elevated Redox marginally positive, sulphides elevated

Total abundance elevated (average 269 individuals per core), taxa per core (32-41) comparable. No major change in community composition.

TC-Ctl-3 None No Snake stars, hermit crabs, colonial ascidians

Fine grey sandy sediment with shell hash, burrow holes and mounds


Redox positive, sulphides marginally elevated

Total abundance slightly elevated (average 279 individuals per core). Comparable taxa richness (32-44 per core). No major change in community composition.

TC-Ctl-6 None No

Snake stars, colonial ascidians, bryozoans, sponges, hermit crabs

Coarse shell material and cobbles, drift/attached macroalgae


Redox positive, sulphides marginally reduced

Total abundance increased c. three-fold (average 639 individuals per core). Taxa richness marginally elevated (46-70 taxa per core). Impacted community composition.


8

Copper and zinc beneath the Ngamahau Bay pens

Total recoverable copper and zinc concentrations (5 to 5.8 mg/kg and 48 to 55 mg/kg

for copper and zinc, respectively; Table 3) were below the ANZECC (2000) ISQG-

Low trigger level (65 mg/kg and 200 mg/kg, respectively) for possible biological

effects. Both copper and zinc concentrations were similar to the previous monitoring

round at NGA, although the zinc concentration at Pen 3

(55 mg/kg) was slightly elevated compared to the previous survey (37 mg/kg; Bennett

et al. 2018b). Overall, both zinc and copper concentrations were similar to levels

reported within Tory Channel during the baseline survey in 2014 (Morrisey et al.

2015).

Table 3. Total recoverable copper and zinc concentrations (mg/kg dry weight) in bulk sediment from Ngamahau Bay pen station samples, February 2019.

Sample Copper Zinc

Pen 1 6.0 48

Pen 2 5.0 50

Pen 3 5.8 55

ANZECC ISQG-Low 65 200

ANZECC ISQG-High 270 410

2.1.2. Type 3 monitoring

In addition to Type 2 monitoring, a spatial footprint mapping exercise was carried out

(Type 3 monitoring) to reassess the appropriateness of the zone boundaries, and the

shape of the NGA depositional footprint after three years of farm operation at the

initial feed level (consent conditions 39b, 65i and 66j, also see Keeley & Taylor 2011).

Type 3 monitoring was carried out within, and just beyond, the EQS compliance

zones (see Figure 4 for sampling locations and types) to:

• map the distribution and extent of the depositional footprint at the initial feed level

specified in the consent

• determine whether the spatial arrangement of monitoring stations captures the

maximum extent of the footprint at this feed level

• cross-check the actual footprint against the predicted footprint.

Full sampling details are provided in the MEMAMP (Bennett et al. 2018b) and results

are provided in Appendix 2.


9

Figure 4. Approximate locations of Type 3 soft-sediment sampling stations for the Ngamahau Bay

salmon farm. Sampling locations were allocated throughout and beyond the outer limit of the predicted depositional footprint at 2,000 tonnes (note the initial feed level is 1,500 tonnes). All Type 2 monitoring stations (T2) except for reference stations are included. Consented maximum distances of EQS Compliance Zone 2/3 (75 m) and Zone 3/4 (300 m) boundaries are shown. Sampling undertaken at each station type: T3a samples = full suite3, T3b samples = full suite (with 2 infauna archived), T3c = redox, sulphides and odour / visual observations of sediment core4.

NGA depositional footprint – individual variables

Type 3 monitoring demonstrates that after three years of operation at the

recommended initial feed level (RIFL, ~1,500 tonnes per annum), the NGA footprint is

slightly larger than that predicted under both the 1,000 tonne (1KT) and 2,000 tonne

(2KT) modelled feed scenarios (we note the recommended initial feed level of 1,500

tonnes was not modelled in the initial assessment of effects [Clarke et al. 2011]).

Results compared to the 2KT footprint and compliance zones are discussed below,

although we note that the NGA farm has been operating at a lower feed level than

this (and that the modelled farm area differs slightly to the current farm).

Low-level enrichment effects were evident as elevated total free sulphide

concentrations c. 160 m inshore of the 2KT predicted footprint. Sulphide

concentrations were also elevated c. 60 m north of the 300 N OLE (and just outside

3 Full-suite analysis includes all parameters outlined in Appendix 1, Section A2.1 (except for copper and zinc). 4 Due to an administrative error total free sulphide values are missing for sites T3a1 and T3b2.


10

the predicted 2KT footprint)5. Macrofaunal abundance was elevated and changes to

community composition (as reflected by higher AMBI scores) were evident c. 40 m

inshore and c. 50 m offshore of the predicted 2KT footprint compared to reference

values. Communities east and west of the farm had an abundance of enrichment

tolerant taxa including amphipods, nematodes, worms belonging to the Paraonidae

family, and the capitellid Barantolla lepte. However, enrichment sensitive taxa such

as maldanid polychaetes were also abundant, suggesting the level of enrichment is

low.

Beyond the OLE in the north, macrofaunal abundance was elevated compared to

reference stations (except TC-Ctl-6). Macrofaunal abundance was also elevated at

and beyond the OLE in the south (and c. 70 m from the predicted 2KT footprint). Taxa

richness was elevated beyond the OLE compared to reference stations (except TC-

CTL-6 which also had high community diversity). Communities at and beyond

the OLE stations were similar to those east and west of the farm with an abundance

of enrichment sensitive taxa (including dorvilleids at 300 S) alongside enrichment

sensitive maldanid polychaetes, again suggesting enrichment effects are only minor.

As above, individual variables (total free sulphides, macrofaunal abundance and

community composition) demonstrate that the NGA depositional footprint extends at

least 60 m north of the 300 N OLE, 70 m south and 40 to 160 m east and west of the

predicted 2KT footprint (Clarke et al. 2011). Overall, the area experiencing minor to

moderate enrichment is at least 17 ha; this is 5 ha greater than the consented

depositional footprint (12 ha), which is greater than the 10% flexibility provided by

Condition 39b. An increase in feed use at this site is likely to result in further

enrichment beyond the present OLE monitoring stations, where no routine sampling

is currently undertaken.

Review of monitoring stations and EQS compliance zones

We note some ambiguity in the OLE EQS as set out in the consent:

• Under the BMP guidelines, the industry operational goal is for the OLE to be ES

< 3.0, and natural conditions are to be maintained. Importantly, background

conditions are assessed as enrichment stage (rather than individual variables).

• The NGA consent requires that ES < 3.0 is maintained at the OLE, and by

contrast, that conditions remain statistically comparable with relevant / appropriate

reference stations. In the context of the consent, it is not clear if the intent is to

measure conditions as ES, or as individual variables. We have assumed that

conditions are measured as ES as they are under the BMP guidelines (rather than

individual variables), and in this context, the OLE stations are statistically

comparable with relevant / appropriate reference stations (See Section 2.1.1).

5 Due to an administrative error total free sulphide values are missing for the sampling station south of the OLE

(T3a1), although concentrations measured at 300 S were similar to reference conditions.


11

Considering only ES scores at the compliance monitoring stations, the NGA

depositional footprint is well within the EQS across all zones (ES > 3.0 only measured

at one station at the Zone 2/3 boundary where ES 4.0 is acceptable). Conditions

(measured as ES, as per the BMP guidelines) also remain statistically comparable

with relevant / appropriate reference stations). On this basis, an amendment of zone

dimensions and area is not required at this stage. However, It is important to note that

‘background’ ES scores for this area (i.e. as measured at reference stations during

this monitoring round) ranged from 1.8 to 1.9, and therefore using ES < 3.0 as an

indicator of ‘natural conditions’ implies that a degree of enrichment outside of the

consented OLE is acceptable.

Recommendations

While the OLE stations were expected to receive low levels of deposition, farm-

related enrichment was detected beyond the predicted depositional footprint.

However, due to the dispersive nature of the NGA farm site, and the low background

enrichment levels at the site, this is unsurprising. It is for the regulatory body to

decide as to whether the observed level of enrichment, as indicated by total free

sulphides and macrofaunal community data, is acceptable beyond the OLE. If the

spatial extent of enrichment is not considered to be acceptable, additional sampling is

recommended to explicitly map the shape and extent of the actual depositional

footprint during the next annual monitoring to inform zoning amendments.

In addition, we recommend a footprint mapping exercise is undertaken at 5 yearly

intervals, or as otherwise required (i.e. prior to feed increase), as per the BMP

guidelines.


12

2.2. Water column

The water column monitoring results are used to determine whether the farm is

compliant with water quality standards (WQS) set out in the resource consent for this

farm (see Appendix 3 for EQS: water quality).

An overview of the annual sampling regime is provided below and illustrated in

Figure 5. Full sample collection details for each monitoring type are provided in

Appendix 3. Water sampling stations for NGA are summarised in Table 4 with station

locations shown in Figure 6:

• Routine (long-term) monitoring for chlorophyll-a (chl-a), dissolved oxygen (DO)

and total nitrogen (TN) was undertaken monthly.

• Full-suite (long-term) monitoring, for a larger suite of analytes (see Appendix 3

for the list of analytes sampled), was carried out in February, March, August and

September6.

• Fine-scale (targeted) monitoring was undertaken alongside long-term

monitoring in March and August.

Figure 5. Water column sampling surveys for the three water column monitoring types (routine, full-suite and fine-scale), during 2018.

6 According to the marine environmental monitoring - adaptive management plan (MEMAMP), full-suite monitoring should have taken place in July/August, however due to an administrative error this was postponed until August/September.


13

Table 4. Water column sampling stations for the routine, full-suite and fine-scale monitoring components. This design includes three of the Marlborough District Council (MDC) state of environment (SoE) monitoring stations (NZKS21/QCS3-6, QCS7 and QCS8) that are sampled by Marlborough District Council.

Description Station name

Net pen (down current) NZKS18

100 m (down current)* NGA 100 m

250 m (down current)* NGA 250 m

500 m (up & down current) NZKS19/20

Cumulative effect reference (CE-Ref) station NZKS21 (QCS3)**

Far-field reference (FF-Ref) station NZKS22

Far-field reference (FF-Ref) station*** QCS7 (inner Tory Channel)

Far-field reference (FF-Ref) station*** QCS8 (Opua Bay)

* Sampled during fine-scale monitoring, in March and August only.

** Also MDC SoE monitoring station. Hereafter referred to only using NZKS21.

*** Results are only included where the additional context is relevant.

Figure 6. NZ King Salmon and MDC routine and full-suite water quality monitoring stations in Tory

Channel. The 100 m and 250 m stations sampled during fine-scale sampling are not shown as their locations are tidally dependent, and therefore subject to change. Location of the net pen station is indicative only (location is tidally-dependent). Rectangle indicates location of the Ngamahau Bay farm.


14

2.2.1. Water column monitoring results

Box 1 provides a summary of water column monitoring results in the context of

specific monitoring objectives or other compliance measures as is detailed in the farm

consent. Key findings are expanded in the following paragraphs and a more

comprehensive account of the 2018 NGA water column results is provided in

Appendix 4.

Box 1. Compliance overview

Has the Ngamahau farm caused elevated nutrient concentrations beyond 250 m from

the edge of the net pens (outside of natural variation for that location/season)?

No. Fine-scale sampling shows elevated downstream concentrations of some nitrogen

species. However, the concentrations at the net pen itself were not outside historic

variation recorded at the same or similar locations for that time of year. Refer Appendix 4,

Section A4.3 for monthly results of TN and all fine-scale sampling results.

Has there been a statistically significant shift towards a eutrophic state?

No. Statistical significance testing on nutrient changes was undertaken by Broekhuizen and

Plew (2018). They found some evidence that nitrate concentrations have risen in the Tory

Channel, concurrent with a reduction in ammonium. The authors commented that, for most

water quality variables, linear trends could not be reliably determined and concluded that

‘the nutrient and chlorophyll concentrations are consistent with the view that the sounds are

near the oligotrophic-mesotrophic boundary, in terms of trophic classification’. Prior to the

salmon farm developments, the Marlborough Sounds were said to be in an oligotrophic-

mesotrophic state (EPA 2013 and references therein).

Have the TN WQS been breached in three successive months?

No. In all of the samples from Tory Channel in 2018, only one sample exceeded the TN

WQS (NZKS21 in May). See Appendix 4, Section A4.3, for monthly results of TN and all

fine-scale sampling results.

Have the DO WQS been breached in three successive months?

No. See Appendix 4, Section A4.2, for month-by-month and depth-related results.

Have the chl-a WQS been breached in three successive months?

No. All the chl-a measurements in the Tory Channel were below the threshold of

3.5 mg/m3. See Appendix 4, Section A4.4, for monthly results at all stations.

Key findings

Overall, water column profile and site average data indicate a well-mixed water

column in this area, owing to the high tidal currents experienced in Tory Channel.

There were no notable reductions in turbidity across any of the water column depth

profiles, except in the top 10 m at CE-Ref (NZKS21) station in February 2018.

Furthermore, there is no evidence of region-wide increases in turbidity across the

time series to date (see Appendix 5), that might indicate reductions in water quality

resulting from eutrophication.


15

Dissolved oxygen (DO) saturation levels at the net pen were within the applicable DO

WQS (i.e. > 70%, Table 5) in all months. DO saturation levels at both 500 m stations

(NZKS19 and NZKS20), as well as at the CE-Ref (NZKS21) and the FF-Ref

(NZKS22) stations, exceeded the applicable DO WQS (i.e. > 90%, Table 5) during

January, March and August. Do saturation levels were also below this DO WQS at

the NZKS21 station in June. The second step DO WQS threshold (WQS [2]) was

exceeded at NZKS19, NZKS20 and NZKS21 in March and at NZKS20 and NZKS21

in April. Although the time series of DO data does not indicate a reduction outside

historical levels and thus a shift in water quality associated with eutrophication (see

Appendix 5), it would be appropriate to undertake a more detailed analysis of monthly

DO profiling data to better understand the likely causes of DO WQS exceedances.

With one exception, TN concentrations at all stations were within the TN WQS

(i.e. ≤ 300 mg-N/m3; Table 5) in all months. The exception was the CE-Ref station in

the middle of Tory Channel in May (NZKS21; 309 mg-N/m3). TN concentrations were

also elevated at this station during 2016 and 2017. Nevertheless, although only a

limited time-series of data is available for TN (see Appendix 5), there is no evidence

to suggest an increased frequency of samples with high nitrogen concentrations in

this area since the farm began operating.

Fine-scale monitoring (carried out during March and August only) revealed a clear

trend of decreasing TN and Urea-N at the surface with increasing distance from the

farm. However, some far-field reference stations had relatively high concentrations of

these nutrients suggesting an external source (e.g. from Cook Strait). In March,

near-bed concentrations of total phosphorus (TP) and dissolved reactive phosphorus

(DRP) were shown to reduce with distance from the farm although differences were

not large. No other spatial gradients in nutrient concentrations were evident.

Chl-a concentrations at all stations were well below the WQS (i.e. ≤ 3.5 mg/m3;

Table 5) in all months. Average monthly fluorescence (a proxy for chlorophyll-a /

phytoplankton biomass) within surface waters (c. 5 m depth) at all sampling stations

shows no evidence of increased frequency of algal blooms in the area during the two

years of monitoring (see Appendix 5, Figure 5.9).

Phytoplankton biomass and community composition did not show any atypical

seasonal trends. Estimated phytoplankton biomass values around the NGA farm sites

in March, April, August and September 2018 were in the range of 2 to 102 mg C/m3

(see Appendix 4, Table A4.4), with the highest biomass estimates occurring in

September, including a particularly high reading at NZKS21 (although we note that

this value still within the range of data from the MDC monitoring stations, see

Broekhuizen & Plew 2018).

Diatoms dominated the phytoplankton biomass across all months, except in April

when overall biomass was particularly low (see Appendix 4, Table A4.4 and


16

Table A4.5). The particularly high biomass recorded at NZKS21 in September was

made up of 75.2% diatoms and 5.7% dinoflagellates, which are not unusual

proportions of these groups. Diatoms usually dominate the phytoplankton biomass,

except in winter when the category ‘other’ often dominates (Broekhuizen & Plew

2018).

Table 5. Summary of water column compliance for parameters measured at each of the Ngamahau Bay salmon farm monitoring stations. Ticks indicate measured concentrations were within the water quality standards (WQS) thresholds on all occasions. Sampling months during which WQS thresholds were exceeded are named.

NZKS18 NZKS19 NZKS20 NZKS21 NZKS22

Net pen 500 m 500 m CE-Ref FF-Ref

DO ✓ Jan, Mar,

Aug

Jan, Mar,

Apr, Aug

Jan, Mar, Apr,

June, Aug

Jan, Mar,

Aug

WQS > 70 % > 90 %

TN n/a ✓ ✓ May ✓

WQS n/a ≤ 300 mg-N/m3

Chl-a ✓ ✓ ✓ ✓ ✓

WQS ≤ 3.5 mg/m3


17

2.3. Light effects monitoring

Submerged artificial lighting (underwater lighting) may attract baitfish and increase

the visibility of prey during night-time hours. Increased aggregation and visibility of

prey could in turn increase rates of predation by the farmed salmon as well as by fish,

marine mammals (e.g. seals) and seabirds outside the pens. Furthermore, birds

attracted to the lights may be at an increased risk of collision/entanglement with farm

structures.

Light effects monitoring was carried out by farm staff at NGA over three years7 during

periods when underwater lighting was fully operational. The purpose of this

monitoring was to confirm that the magnitude of effects from underwater lighting on

night-time feeding activity by fish, seabirds and marine mammals in and around the

illuminated pens are generally as expected (condition 65j).

Surveys were carried out across 10 one to two-week periods during dark hours and at

least one hour following sunset or prior to sunrise (in some cases observations were

made both before and after sunrise). For each sampling period, farm staff recorded

observations on night-time activity by fish, seabirds and marine mammals in and

around the illuminated cages. Staff were required to be familiar with the species of

fish, birds and mammals likely to be seen around the farms, and the types of

behaviour that each is likely to display. An example of the data sheet that staff used is

provided in Appendix 6. A summary of survey data and discussion around whether

the observations were generally as expected is provided below.

2.3.1. Summary of observations

On most occasions, seals were observed swimming outside of the pens or resting on

farm structures. Although it was not specified whether seal activity was observed near

both illuminated and dark pens, site visits to other high-flow farms during dark hours

confirm that seal activity is similar between illuminated and dark pens (Cornelisen et

al. 2013, Bennett & Cornelisen 2018). There were no recorded observations of seals

chasing prey at night. This suggests there is no increased rates of predation by seals

as a result of the underwater lights.

Invertebrates including squid and crabs were frequently observed within the

illuminated pens. Squid, in particular, are a common prey items for seals, and

aggregations as a result of underwater lights may increase seal predation around

illuminated pens. Nevertheless, while seals were observed around the illuminated

pens at NGA, there was no evidence to suggest there is an increased rate of

predation by seals as a result of the underwater lights.

7 We note that while the consent requires only two years of farm staff observations, three years of data were

collected and are presented here.


18

Dolphin (bottlenose) were observed swimming around the farm at night on only one

occasion. Bottlenose dolphin are common in the Marlborough Sounds, and their

presence is unlikely to be related to the underwater lights.

Baitfish were almost always observed schooling and swimming inside the illuminated

pens. While observations were not made for the dark pens, baitfish schooling within

dark pens were observed during a site visit to the Kopaua salmon farm in the Pelorus

Sound (Bennett & Cornelisen 2018). Farm personnel are also reported to observe

baitfish in the non-illuminated pens at the Te Pangu Bay salmon farm (Cornelisen et

al. 2013), suggesting this activity is unrelated to the underwater lights. Other fish

observed swimming in the illuminated pens included garfish, dogfish, tarakihi and

barracuda. Occasionally pufferfish, kahawai and koheru were also noted.

Seabirds (commonly red-billed and black-backed gulls as well as shags) were

frequently observed to be resting on farm structures around the illuminated pens.

However, there were no recorded observations of seabirds foraging at night or

reports of collision/entanglement of seabirds with farm structures. This suggests there

is no increased activity of seabirds as a result of the underwater lights. As such, the

risk of seabird collision/entanglement with farm structures as a result of underwater

lighting is likely to be minor.

Overall, there is little evidence for enhanced aggregation of prey and predation within

illuminated pens and waters directly adjacent to illuminated pens at the NGA farm

site. While it is possible that such effects will occur as a result of underwater lighting

at this site, the area affected by the lights (as documented at the Kopaua farm; see

Bennett & Cornelisen 2018) is small and therefore the effects will be of little

ecological significance within the context of the wider Marlborough Sounds

ecosystem.

3. ELIGIBILITY FOR FEED INCREASE

After three years of operation at or near (±15%) the maximum initial feed discharge

level (1,500 tonnes per annum at NGA), the NGA farm may qualify for a feed

increase, if certain criteria are met. These criteria are compliance with consent

Conditions 36 to 44, which are summarised in Table 6 along with discussion around

whether the NGA farm meets these conditions.


19

Table 6. Summary of consent conditions required to be met in order for the Ngamahau Bay salmon farm (NGA) to qualify for a feed increase. Discussion around whether the NGA farm meets the conditions is provided along with references for further information.

Condition Summary Eligibility description Reference

36 Annual tonnage of feed may only be increased if Conditions 37-44

are met as well as any specifications from the 2018 MEMAMP.

See below (and also note additional consideration

section in regard to far-field effects).

McGrath et al. 2019,

McGrath & Campos

2019

37a Annual tonnage of feed may only be increased if:

a. The farm shall have operated at or near (±15%) its current

maximum annual feed discharge level for at least 3 years; and

Feed has been discharged near the maximum initial

feed level for at least 3 years.

Section 0 and

previous annual

monitoring reports

(Elvines et al. 2017,

Bennett et al. 2018b)

37b Annual tonnage of feed may only be increased if:

b. Annual monitoring results of the Enrichment Stage (ES) from the

most recent two successive years shall be comparable, based on

the monitoring undertaken in Condition 66, assessed as follows.

The Enrichment Stage (ES) from the annual monitoring,

assessed in accordance with Condition 40, shall statistically not

be significantly more than the ES from the previous year, based

on the average result for all sampling stations (Figure 3) within

each compliance Zone. This requirement must be met for each of

the Environmental Quality Standards (EQS) compliance Zones

for which ES are specified in Condition 40.

ES scores at all pen stations have decreased since

the previous monitoring round.

ES score at Zone 2/3 (75 N) were the same as the

previous year.

ES scores at Zone 3/4 (300 N and 300 S) were the

same as the previous year.

Type 2 monitoring

results (Section 2.1.1

and Appendix 2)

37c Annual tonnage of feed may only be increased if:

c. The marine farm complies with all the EQS specified in Condition

40 and is less than the relevant maximum EQS for each Zone.

Individual variables demonstrate natural conditions

have not been maintained at the OLE. However,

according to ES scores the NGA depositional

footprint is well within the EQS across all zones (ES

> 3.0 only measured at one station at the Zone 2/3

where ES 4.0 is acceptable).

Type 2 monitoring


and Appendix 2).

Type 3 monitoring

results (Section

2.1.2).


20

Condition Summary Eligibility description Reference

38 The discharge of feed, marine biofouling and antifouling at the

marine farm shall meet the requirements of Conditions 39 - 44

relating EQS at all times.

Discharge of feed and levels of copper and zinc

beneath pens meet requirements. Marine biofouling

was not assessed at NGA under this monitoring

programme.

Type 2 monitoring


and Appendix 2).

39, 40 EQS Compliance Zones shall be defined for the marine farm. At all

times, the seabed beneath and in the vicinity of the marine farm

shall comply with the EQS specified in Table 3 (of the consent).

EQS compliance zones defined. Amendment of

compliance zones may be required pending

decision on Type 3 monitoring results:

- Based on the individual variable

measurements the EQS compliance zones

may need amending,

- Based on ES scores alone, the NGA EQS

compliance zones do not need amending,

but additional monitoring stations further

than 300m are recommended when feed

levels are higher.

Type 3 monitoring

results (Section

2.1.2).

41, 42 Composite samples of sediments beneath and beside the net pens

shall be assessed against the ANZECC (2000) ISQG-Low criteria

for copper and zinc, as a first-tier trigger level. Where total metals

analysis of composite sediment samples exceeds the ANZECC

(2000) ISQG-Low criteria for copper and zinc, the MEM-AMP (refer

Conditions 65-66) shall include a hierarchical schedule of

monitoring of increasing focus and intensity and, ultimately,

management action based on the decision hierarchy contained in

Figure 5.

Compliant. Type 2 monitoring


and Appendix 2)

43, 44 The marine farm shall be operated at all times in such a way as to

achieve and comply with the Water Quality Objectives in the water

column.

Compliant. Water column

monitoring (Section

2.2 and Appendix 4)


21

3.1. Additional consideration - far-field effects

Farming in dispersive environments such as Tory Channel means that although

organic enrichment in the immediate proximity are more diffuse than they would be in

low-flow environments, effects from far-field dispersal of organic waste are more

likely. Enrichment effects in the far-field are also more difficult to attribute. The TC-

Ctl-4 reference site (situated in Ngaruru Bay) was established in 2013 (under the Clay

Point consent) to determine whether far-field enrichment effects were occurring as a

result of salmon farming in Tory Channel (Newcombe et al. 2013). While this

reference site is not monitored specifically under the NGA consent, we iterate its

relevant to all salmon farms operating within Tory Channel, because enrichment in

such side embayment’s is likely to be cumulative from multiple sources (e.g. other

farms as well as natural processes and other non-farm related events).

In this context, we note that average overall ES score at this station (ES 3.0) has

increased by 0.8 since the 2017 monitoring survey (Bennett & Elvines 2018).

Conditions of ES3.0 have not been encountered naturally within the Marlborough

Sounds to date (MPI 2015). Parameters driving the increase in ES score at this

station include elevated total free sulphides and decreased redox potential, as well as

decreased macrofaunal abundance and taxa richness when compared to the other

reference stations. The deterioration in conditions at this station coincided with the

establishment of the NGA farm (and a subsequent c. 1,300-tonne increase of feed

use in Tory Channel), thus farm related enrichment may be a contributing factor.

However, we also note the last two summers had anomalously warm and calm

weather patterns, which may also have been a contributing factor. As farm-related

effects are unable to be ruled out, a detailed analysis of all available Tory Chanel

data was recommended in the recent annual monitoring reports for the Clay Point

and Te Pangu Bay farms (McGrath et al. 2019, McGrath & Campos 2019). We

recommended this is done as a precautionary measure prior to any feed increases at

the NGA site.


22

4. KEY FINDINGS

All soft-sediment sampling stations at the NGA farm were compliant with the EQS

specified in the consent conditions and ES scores at all pen stations have decreased

since the previous monitoring round. While ES scores at the OLE stations were within

the consented EQS and remained the same as last year, changes to community

composition and sediment chemistry suggest natural conditions have not been

maintained at these sites. Additional sampling (for Type 3 monitoring) confirmed that

enrichment effects (elevated sulphide concentrations and macrofaunal abundance)

extend beyond the OLE stations (and the predicted depositional footprint modelled for

a higher feed level than what the farm is currently operating at). However, under the

BMP guidelines, background conditions are assessed as enrichment stage (rather

than individual variables), and the industry operational goal (and consent

requirement) is for the OLE to be ES < 3.0. With the exception of the 75 N station (at

the Zone 2/3 boundary), ES scores at all sampling stations were < 3.0.

Based on natural conditions being measured as ES, an amendment of zone

dimensions and area is not required at this stage. Nevertheless, clarification is

required on the EQS at the OLE as to whether the observed level of enrichment

beyond the OLE is acceptable, despite conditions being within the industry

operational goal (and consented EQS) of ES3.0, and comparable to reference sites

as measured by ES. If the spatial extent of enrichment is not considered to be

acceptable, additional sampling is recommended to explicitly map the shape and

extent of the actual depositional footprint during the next annual monitoring to inform

zoning amendments. We also recommend that a footprint mapping exercise is

undertaken at 5 yearly intervals, or as otherwise required (i.e. prior to feed increase),

as per the BMP guidelines.

Light effects monitoring found little evidence for enhanced aggregation of prey and

predation within illuminated pens and waters directly adjacent to illuminated pens at

the NGA farm site.

None of the WQS for total nitrogen (TN), dissolved oxygen (DO) and chlorophyll-a

(chl-a) were exceeded in three successive months, i.e. an amber state was not

triggered. There is no evidence of region-wide increases in turbidity across the time

series to date that might indicate reductions in water quality due to eutrophication.

Chl-a concentrations were well below the WQS in all months. Phytoplankton biomass

and community composition did not show any atypical seasonal trends. No

recommendations are made for the water column sampling design for the next

sampling round, pending finalisation of a working group review of the water column

approaches as they relate to the Marlborough Sounds salmon farming industry.

According to ES scores the NGA farm qualifies for a feed increase (i.e. consent

Conditions 36 to 44 are met). However, individual variables demonstrate natural


23

conditions have not been maintained at and beyond the OLE. As a result, ambiguity

remains as to whether Conditions 37c and 40 are met. It is for the regulatory body to

decide as to whether the observed level of enrichment is acceptable beyond the OLE,

and therefore whether the NGA farm qualifies for a feed increase. We note that an

increase in feed use at this site is likely to result in further enrichment beyond the

present OLE monitoring stations.

Furthermore, while not related to a specific consent condition, we note that potential

far-field effects have been observed at the TC-Ctl-4 reference station in Ngaruru Bay

(established under the Clay Point consent in 2013). A deterioration in conditions

(elevated sulphides, declines in macrofaunal abundance and changes to macrofaunal

communities) at this station coincides with the establishment of the NGA farm (and a

subsequent c. 1,300-tonne increase of feed use in Tory Channel). As farm-related

enrichment effects are unable to be ruled out, a detailed analysis of all available Tory

Chanel data is recommended as a precautionary measure prior to any feed increases

at the NGA site.


24

5. REFERENCES

ANZECC 2000. Australian and New Zealand guidelines for fresh and marine water

quality 2000 Volume 1. National Water Quality Management Strategy Paper

No. 4. Australian and New Zealand Environment and Conservation Council

and Agriculture and Resource Management Council of Australia and New

Zealand, Canberra.

Bennett H, Cornelisen C 2018. Effects of underwater lighting on the marine

environment at the Kopaua salmon farm. Prepared for New Zealand King

Salmon Co. Ltd. Cawthron Report No. 3149. 16 p.

Bennett H, Newcombe E, Elvines D, Dunmore R 2018a. Marine environmental

monitoring - adaptive management plan for salmon farms Ngamahau, Kopaua

and Waitata (2018-2019). Prepared for The New Zealand King Salmon Co.

Ltd. Cawthron Report No. 3211. 34 p. plus appendices.

Bennett H, Elvines D, Knight B 2018b. 2017-2018 annual environmental monitoring

report for the Ngamahau Bay salmon farm. Prepared for The New Zealand

King Salmon Co. Ltd. Cawthron Report No. 3144. 42 p. plus appendices.

Broekhuizen N, Plew D 2018. Marlborough Sounds water quality monitoring: review

of Marlborough District Council monitoring data 2011 – 2018. Prepared for

Marlborough District Council. NIWA Client Report 2018248HN. 159 p plus

appendices.

Clark D, Taylor D, Keeley K, Dunmore R, Forrest R, Goodwin E 2011. Assessment of

effects of farming salmon at Ngamahau, Queen Charlotte Sound: Deposition

and benthic effects. Prepared for New Zealand King Salmon Company

Limited. Cawthron Report No. 1993. 52 p.

Cornelisen C, Forrest R, Quarterman A 2013. Effects of artificial lighting on the

marine environment at the Te Pangu Bay salmon farm. Prepared for New

Zealand King Salmon Company Limited. Cawthron Report No. 3149. 18 p.

Cornet-Barthaux V, Armand L, Queguiner B 2007. Biovolume and biomass estimates

of key diatoms in the Southern Ocean. Aquatic Microbial Ecology 48(3): 295-

308.

Dunmore R 2019. Reef environmental monitoring results for the New Zealand King

Salmon Company Ltd salmon farms: 2018. Prepared for the New Zealand

King Salmon Company Ltd. Cawthron Report No. 3291. 84 p. plus

appendices.

Elvines D, Knight B 2017. Marine Environmental Monitoring - Adaptive Management

Plan for salmon farms Ngamahau, Kopaua and Waitata (2017-2018).

Prepared for New Zealand King Salmon Company Limited. Cawthron Report

No. 3050. 35 p. plus appendices.


25

Elvines D, Knight B, Berthelsen A, Fletcher L 2017. Ngamahau Bay salmon farm:

annual monitoring report (2016–2017). Prepared for The New Zealand King

Salmon Co. Ltd. Cawthron Report No. 3000. 38 p. plus appendices.

EPA (Environmental Protection Agency) 2013. Final report and decision of the Board

of Inquiry: Volume 1; New Zealand King Salmon requests for plan changes

and applications for resource consents. Decision date 22 February 2013.

Hillebrand H, Dürselen CD, Kirschtel D, Pollingher D, Zohary T 1999. Biovolume

calculation for pelagic and benthic microalgae. Journal of Phycology 35: 403-

424.

Karlson B, Cusack C, Bresnan E 2010. Microscopic and molecular methods for

quantitative phytoplankton analysis. UNESCO. 113 p.

Keeley N, Taylor D 2011. The New Zealand King Salmon Company Limited:

Assessment of environmental effects - benthic. Prepared for The New Zealand

King Salmon Co. Ltd. Cawthron Report No.1285. 73 p plus appendices.

Keeley N 2012. Assessment of enrichment stage and compliance for salmon farms–

2011. Prepared for New Zealand King Salmon Company Limited. Report No.

2080. 15 p.

Keeley N, Macleod C, Forrest B 2012. Combining best professional judgement and

quantile regression splines to improve characterisation of macrofaunal

responses to enrichment. Ecological Indicators 12: 154-166.

Menden-Deuer S, Lessard EJ 2000. Carbon to volume relationships for

dinoflagellates, diatoms, and other protist plankton. Limnology and

Oceanography 45(3): 569-579.

McGrath E, Bennett H, Campos C 2019. 2018-2019 Annual environmental monitoring

summary for the Te Pangu Bay salmon farm. Prepared for the New Zealand

King Salmon Co. Ltd. Cawthron Report No. 3324.14 p. plus appendices.

McGrath E, Campos C 2019. 2018-2019 Annual environmental monitoring summary

for the Clay Point salmon farm. Prepared for the New Zealand King Salmon

Co. Ltd. Cawthron Report No. 3325.13 p. plus appendices.

Morrisey D, Broekhuizen N, Grange K, Stenton-Dozey J 2014. Baseline monitoring

plan for new salmon farm sites, Marlborough Sounds, NIWA Client Report No:

NEL2013-015. 78 p. plus appendices.

Morrisey D, Stenton-Dozey J, Broekhuizen N, Anderson T, Brown S, Plew D 2015.

Baseline monitoring report for new salmon farm sites, Marlborough Sounds.

NIWA Client Report No. NEL-2014-020. Prepared for the New Zealand King

Salmon Co. Ltd. 247 p.

Ministry of Primary Industries (MPI) 2015. Best Management Practice guidelines for

salmon farms in the Marlborough Sounds: Part 1: Benthic environmental

quality standards and monitoring protocol (Version 1.0 January 2015).


26

Prepared for the Ministry for Primary Industries by the Benthic Standards

Working Group (Keeley N, Gillard M, Broekhuizen N, Ford R, Schuckard R,

Urlich S).

Rott E 1981. Some results from phytoplankton counting intercalibrations.

Schweizeriche Zeitschrift für Hydrologie 43(1): 34-62.


27

6. APPENDICES

Appendix 1. Methodology for soft-sediment sampling.

A1.1 Background

The following sub-sections provide detail on the soft-sediment sampling methodology,

described in the most recent marine environmental monitoring - adaptive

management plan (MEMAMP) for the site (Bennett et al. 2018a). Further rationale

and details related to the general monitoring procedures can be found in the Best

Management Practice (BMP) guidelines developed for salmon farming in the

Marlborough Sounds (MPI 2015).

A1.2 Sampling protocol

Three replicate sediment grab samples were collected at each sampling station using

a van Veen grab. Each grab sample was examined for sediment colour, odour,

texture and bacterial coverage. The top 30 mm of one sediment core (63 mm

diameter) was analysed for organic content as % ash-free dry weight (AFDW), redox

potential (EhNHE, mV), and total free sulphides (µM). In addition, composited triplicate

samples from the pen stations were analysed for total recoverable copper and zinc

concentrations. Laboratory analytical methods for sediment samples can be found in

Table A1.1.

A separate core (10 cm deep and 113 cm2 surface area) was collected from each

grab to describe the macrofaunal community assemblages. Core contents were

sieved to 0.5 mm and preserved in a solution of 95% ethanol and 5% glyoxal.

Animals were identified and counted by specialists at the Cawthron taxonomy

laboratory.

Two additional replicate samples (‘d’ and ‘e’ replicates) were collected from each pen

station to determine the redox potential (measured in the field), and to obtain organic

content and macrofauna samples for archive purposes.

Video footage of the seabed was taken at each station to qualitatively assess the

level of visible bacterial coverage, general seabed condition and presence of

sediment outgassing. The sea surface was also scanned for visible sediment

outgassing as this could provide further evidence of particularly enriched conditions.

General observations of epibiota (surface-dwelling animals) were also made.


28

A1.3 Data analysis: Assessment of Enrichment Stage

Seabed condition can be placed along an enrichment gradient which has been

quantitatively defined according to Enrichment Stage (ES). The ES assessment

references a selection of informative chemical and biological indicator variables8.

For each indicator variable (raw data), an equivalent ES score was calculated using

previously described relationships (MPI 2015)9. Average ES scores were then

calculated for:

• sediment chemistry variables (redox and sulphides).

• macrofauna composition variables: abundance (N), total number of taxa (S/core),

richness (d), Margalef richness index (d), evenness (J’), diversity (H’) and biotic

indices (AMBI, mAMBI and BQI).

• organic content (% AFDW).

The overall ES score for a given sample was then calculated by determining the

weighted average10 of those three groups of variables. Finally, the overall ES for the

sampling station was calculated from the average of the replicate samples with the

degree of certainty reflected in the associated 95% confidence interval.

Table A1.1 Laboratory analytical methods for sediment samples (February 2019) processed by either Hill Laboratories (a) or Cawthron Institute (b).

Analyte Method Default detection limit

Sediment samples

Organic matter (as ash-free dry weight) a

Ignition in muffle furnace 550°C, 6hr, gravimetric. APHA 2540 G 22nd ed. 2012. Calculation: 100 – Ash (dry wt).

0.04 g/100 g

Total recoverable copper & zinc a

Dried sample. Nitric/ hydrochloric acid digestion, ICP-MS, trace level. US EPA 200.2.

0.2 - 2 mg/kg (Cu)

0.4 - 4 mg/kg (Zn)

Total free sulphides b Cawthron Protocol 60.102. Sample solubilised in high pH solution with chelating agent and anti-oxidant. Measured in millivolt (mV) using a sulphide specific electrode and calibrated using a sulphide standard.

8 There are risks associated with placing emphasis on any individual indicator variables of ES. This is particularly

true for chemical indicators, which tend to be more spatially and temporally variable. As such, the derived overall ES value is considered a more robust measure of the general seabed state.

9 We note that ES calculations in the previous monitoring reports for this site did not implement the rules from appendix 10.2: bullet points 2b and 2c from MPI (2015).

10 Weighting used in the current assessment is the same as that used in previous years: organic loading = 0.1, sediment chemistry = 0.2, macrofauna composition = 0.7.


29

A1.4 Compliance framework for soft-sediment monitoring results.

The environmental monitoring results from soft-sediment habitats monitoring are used

to determine whether the farms are compliant with the environmental quality

standards (EQS: benthic) specified in the consent conditions.

A1.4.1 Enrichment

The EQS (benthic) are based on a seabed impact ‘zones concept’; an approach that

provides an upper limit to the spatial extent and magnitude of seabed impacts (see

Keeley 2012). The EQS in the consent conditions (Table A1.2) set precise

parameters for the allowable environmental states within the zones. In addition, best

management practice guidelines–benthic (BMP; MPI 2015) exist for salmon farming

in the Marlborough Sounds. The BMP was developed after the consent conditions

were written; thus, some aspects of the monitoring design or compliance framework

are inconsistent with that outlined in the BMP. Where discrepancies or uncertainty

exists as to the consistency or intent of the EQS or management action, the rationale

in the BMP will be consulted for guidance (as per Bennett et al. [2018a] and see

Table A1.2 3b).


30

Table A1.2 Environmental quality standards (EQS) for each zone at the Ngamahau Bay salmon farm (consent U140296,Table A1.2a). EQS descriptors for the OLE as worded in the consent conditions, and the best management practice guidelines (BMP; MPI 2015, Table A1.2b). The EQS descriptors from the BMP correspond to varying levels of management response.

Table A1.2a.

Compliance

Monitoring

Location

Consented EQS

Zones 1 & 2

Beside and

beneath the net

pens (ZME as

per the BMP)

Measured beneath

the edge of the net

pens

ES ≤ 5

No more than one replicate core with no taxa (azoic).

No obvious spontaneous outgassing (H2S/methane).

Bacteria mat (Beggiatoa) coverage not greater than

localised/patchy in distribution.

Zone 3*

Near to the net

pens

Measured at the

Zone 2/3 boundary

ES ≤ 4.0

Infauna abundance is not significantly higher than at

corresponding ‘Pen’ station.

Number of taxa > 75% of number at

relevant/appropriate reference station(s).

Zone 4

Outside the

footprint area

(OLE as per the

BMP)

Measured at the

Zone 3/4 boundary

stations

ES < 3.0

Conditions remain statistically comparable with

relevant/appropriate reference station(s).

Table A1.2b.

Consented EQS for

the OLE

BMP EQS for the OLE

ES threshold /

industry

operational goal

ES < 3.0 Overall ES < 3.011 (i.e. maintain natural conditions)

EQS descriptors Conditions remain

statistically

comparable with

relevant /

appropriate

reference station(s)

Alert: A statistically significant increase relative to

appropriate reference station(s)12

Minor: Overall ES ≥ 3.0,

AND

Mean ES 0.4 higher than previous year, and increase

is significant relative to appropriate reference stations

*Note that sampling at this station is not a requirement under the BMP guidelines

11 … Natural (i.e. non-farm impacted) seabed in the Marlborough Sounds varies from about ES 1.5–2.5 (but no

greater than ES 2.9) … (MPI 2015) 12 Statistically significant increase relative to appropriate reference station(s) implies the use of a BACI-type

analysis to test for a significant Station:Survey interaction term. More than one reference station may be included in the analysis.


31

A1.4.2 Copper and zinc

Compliance for copper and zinc levels follows the decision hierarchy in the BMP

guidelines (MPI 2015), as shown in Figure A1.1. The BMP guidelines state that the

ANZECC (2000) ISQG-Low criteria for copper and zinc are the most appropriate

trigger values for sediments beneath farms (Table A1.3, Figure A1.1). Therefore,

these guideline thresholds should be used to trigger further action if exceeded.

Table A1.3. ANZECC (2000) Interim Sediment Quality Guideline concentrations for copper and zinc

(mg/kg).

ISQG-Low ISQG-High

Copper 65 270

Zinc 200 410


32

Figure A1.1 Decision response hierarchy for metals tiered monitoring approach (from MPI 2015).

Copper is the example shown here.

Cage-edge or beneath-cage

composite sample

Sediment total metals analysis

Re-analyse individual triplicate

samples sieved at 250 m

No

Acid soluble Cu analysis of

individual triplicates of bulk

fraction e.g. 1M HCl

No further

action this

monitoring

round

Management action to

reduce inputs of copper

to benthic sediments

Spatial survey to delineate

ISQG-Low contour for AE-Cu

Below

ISQG-Low?

No

All below

ISQG-Low?Yes

Re-analyse individual

triplicate samples for bulk

sediment concentration

No

All below

ISQG-Low?Yes

Measure bulk sediment

recoverable copper

Measure fine sediment

recoverable copper

How much Cu is

contributed by the

coarse fraction?

Evidence for chance

inclusion of large paint

flakes

No

Mean

AE-Cu Below

ISQG-Low?

Yes

Estimate of

bioavailability

Flagged to complete

next monitoring round

to at least level 3

1

2

3

4 Reduction of inputs

Establishment of

sediment Cu contours

Flagged to complete


to at least level 2

5

Yes

No

ISQG-L

contour <50m or zone

equivalent?

Yes

Flagged to complete


to at least level 5

Management action to further

reduce inputs of copper

to benthic sediments

Ecotoxicological studies to refine

site-specific trigger levels for long-

term protection of benthic habitats

Replacement of ISQG-L with

site-specific Cu trigger level

6Refinement of site-

appropriate trigger level

for sediment Cu

Distance of ISQG-L

contour from cage

edge

LEVEL


33

Appendix 2. Comprehensive discussion of results of the February 2019 soft-sediment monitoring survey at the Ngamahau Bay salmon farm (NGA).

A2.1 Qualitative description of soft-sediment habitats

Video footage of the seabed at the NGA stations showed relatively coarser sediments

than observed at most sampling stations throughout Tory Channel (except for TC-Ctl-

2 and TC-Ctl-6). Sediments here are predominantly sand with a considerable amount

of shell hash, including whole empty shells.

Sediments at the pen stations were dark grey and no Beggiatoa-like bacterial

coverage or outgassing was seen. Feed pellets or fish faeces were evident on the

surface of the sediment at the Pen 2 station (Figure A2.1). Noticeable epifauna

observed at Pen 1 included a solitary snake star (Ophiosammus maculata), anemone

(Anthothoe albocincta) and cushion star (Patiriella regularis) (Figure A2.1). The only

conspicuous epifauna noted at Pen 2 were hermit crabs (likely Pagurus sp.). Pen 3

had a higher diversity of conspicuous epifauna, including snake stars, cushion stars,

an apricot sea star (Sclerasterias mollis), 11-armed sea stars (Coscinasterias

muricata), a sea cucumber (Australostichopus mollis), a sea tulip (Pyura

pachydermatina), and clumps of blue and green-lipped mussels (Mytilus

galloprovincialis and Perna canaliculus) (Figure A2.1). Drift macroalgae (Ulva sp.)

was observed at all three Pen stations (Figure A2.1).

The substrate at the 75 N station was similar to beneath the pens, with predominantly

dark grey sandy sediment and a high proportion of shell hash. No Beggiatoa-like

bacterial coverage or outgassing was seen. Snake stars, a finger sponge, as well as

attached and drift macroalgae were observed, as were occasional burrow holes

(Figure A2.1).

Further from the farm structures at the OLE stations (300 N and 300 S), the sediment

was finer with some shell hash throughout. Sediments at the 300 S station had

greater amounts of shell hash, representative of transitional sandy sediments such as

those fringing reef areas. Snake stars were again abundant, and other obvious

epifauna included 11-armed and apricot sea stars, solitary and colonial ascidians, a

purple fan worm (Sabellidae sp.), and bryozoans. Drift and attached macroalgae were

also common, particularly at the 300 N station, which also featured sparse diatom

mat coverage (Figure A2.1).

The substrate at the TC-Ctl-1 reference station was predominantly fine sand, covered

in a rusty-coloured diatom mat (Figure A2.1). There were a number of burrow holes

and trail marks at this station, while the only epifauna observed were snake stars.

The TC-Ctl-3 reference station substrate was predominantly fine, light grey sandy

sediment with some shell hash present. Burrow holes and mounds were evident


34

(Figure A2.1). Noticeable epifauna included snake stars, hermit crabs and colonial

ascidians.

The TC-Ctl-6 reference station substrate contained considerable amounts of coarse

shell material and cobbles, more comparable to the stations adjacent to the NGA

farm. Epifaunal diversity was high at this station, compared with the other two

reference stations, and included colonial ascidians, snake stars, bryozoans, sponges

and hermit crabs. Attached and drift macroalgae were also observed.


35

Figure A2.1. Representative images of the seafloor at each of the Ngamahau Bay (NGA) salmon farm monitoring stations, February 2019.


36

Figure A2.1 continued. Representative images of the seafloor at each of the Ngamahau Bay (NGA)

salmon farm monitoring stations, February 2019.

A2.2 Assessment of enrichment to soft-sediment habitats

The average overall ES scores at the three Pen stations were ES 2.7, 2.7 and 2.9 at

Pens 1, 2 and 3, respectively (Table 1), well within the consented EQS (ES ≤ 5) for

this zone. While these values have decreased since the previous monitoring round

(2018 ES scores at Pens 1, 2 and 3 were ES 3.1, 3.2 and 3.4, respectively), they still


37

indicate moderate enrichment levels. At Pens 1 and 3, enrichment effects were

evident as slightly elevated (in an absolute sense) organic content and total free

sulphides levels. Redox potential at these sites have increased by c. two-fold from the

previous year and are now largely similar to the reference stations (particularly Pen 3;

Figure A2.2). At Pen 2, total free sulphides have decreased while redox potential

remains comparable to the previous year (Table A2.1; Figure A2.2).

Macrofaunal communities at all pen sites had high total abundances (average 1718,

3808 and 1845 individuals per core at Pens 1, 2 and 3, respectively) when compared

to reference stations (average abundance 269 to 639 individuals per core;

Figure A2.2) and baseline values (average abundance 130–180 individuals per core;

Morrisey et al. 2015). Taxa richness was also marginally elevated at the pen stations

(average 62, 59 and 55 taxa per core at Pens 1, 2 and 3, respectively) in comparison

to reference sites (average richness 37 to 59 taxa per core; Figure A2.2) and baseline

conditions (average richness 30–40 taxa per core; Morrisey et al. 2015). This is more

consistent with fertilisation effects seen at minor enrichment levels, as lower taxa

richness is expected with moderate levels of enrichment. Interestingly, the total

abundance at Pen 2 (3808 individuals per core) was c. two-fold greater than Pen 1

and Pen 3, while the taxa richness (average 59 taxa per core) was largely similar to

reference values (Figure A2.2). The average total abundance at this station has

increased from 1040 individuals and 35 taxa per core from the previous survey

(February 2018, Bennett et al. 2018b).

There was a notable shift in community composition across all pen stations since the

previous survey. This was characterised by an increase of polydorid polychaetes

(Spionidae), as well as higher densities of Oligochaete worms and enrichment

tolerant taxa including nematodes, which were notably abundant (particularly at Pen

2). Abundance of the opportunistic polychaete, Capitella capitata, has decreased at

all pen stations since they previous survey. Changes in community composition were

reflected by elevated AMBI scores (and relatively lower mAMBI scores).

The overall ES score at 75 N (Zone 2/3 boundary) was ES 3.0, the same as recorded

at this station in 2018. While within the consented EQS for this zone boundary, this

score indicates a moderate level of enrichment and is slightly higher than ES scores

measured at the pen stations. Observed changes in total free sulphides and

macrofaunal communities at this station were similar to those observed at the pen

stations (Figure A2.2). While total free sulphide values are elevated compared to both

the pen and reference stations, they have decreased slightly from last year. Similarly,

while macrofaunal abundances (average 1253 individuals per core) and taxa richness

values (average 43 taxa per core) remain elevated in comparison to the reference

stations, these values have decreased from the previous year (average 1357

individuals and 51 taxa per core in 2018, respectively; Figure A2.2). As with the pen

stations, higher numbers of enrichment tolerant nematodes, and to a lesser extent

Capitella capitata, were evident at this station. Marginal changes in community


38

composition were reflected by elevated AMBI scores (and relatively lower mAMBI

scores).

The ES score at both OLE stations was ES 1.9, similar to the range of average ES

scores at the reference stations (overall ES 1.8 to 2.0), and the same score as at the

OLE stations last year (ES 1.9, Figure 3). While ES scores at the OLE stations were

the same as the previous year, community composition and sediment chemistry

suggest natural conditions have not been maintained at these sites. Interestingly,

macrofaunal abundance at 300 S has increased c. two-fold since the previous survey

(average 869 individuals per core compared to 428 individuals per core in 2018).

Taxa richness was also slightly elevated (average 60 to 71 taxa per core). At 300 N,

total free sulphides have increased c. two-fold since last year and remain elevated

relative to reference conditions. Meanwhile other community composition metrics at

both OLE stations remain similar to reference stations, as supported by AMBI (and

corresponding M-AMBI) scores (Figure A2.2).

We note that under the BMP guidelines, background / natural conditions are

assessed as enrichment stage (rather than individual variables), and the industry

operational goal is for the OLE to be ES < 3.0. In the context of ES scores both OLE

stations are compliant (i.e. ES < 3.0)13. The NGA consent requires that ES < 3.0 is

maintained at the OLE and that conditions remain statistically comparable with

relevant / appropriate reference stations. Conditions (assessment as ES scores as

per the BMP guidelines) at both OLE stations are statistically comparable with

relevant / appropriate reference stations14.

13 See Section 2.1.2 for discussion on issues associated with using ES <3.0 as a proxy for “natural conditions”. 14 300 S: one-way PERMANOVA: Pseudo-f (3, 11) = 0.96, p = 0.45, 300 N: one-way PERMANOVA: Pseudo-f (3,

11) = 1.3, p = 0.32.


39

Figure A2.2. Sediment organic matter (% ash-free dry weight; AFDW), redox potential (EhNHE, mV),

total free sulphides (µM) and macrofauna statistics determined at the Ngamahau Bay salmon farm monitoring stations, February 2019. TC-Ctl = Tory Channel control. Error bars = ± 1 SE, n = 3.


40

Table A2.1. Detailed Enrichment Stage (ES) calculations for each station at the Ngamahau Bay salmon farm stations, February 2019. For details about how these values were calculated, see MPI (2015). Underlined values are cases where best professional judgement (BPJ; Keeley et al. 2012) was used. Note that ES calculations in previous annual monitoring reports did not implement the rules from Appendix 10.2: bullet points 2b&c from MPI 2015.

SITE INFORMATION

Date: Feb-19 Variable group weightings:

Farm/site: Ngamahau Bay 0.1 0.2 0.7

Flow environment: HF

RAW DATA (to be entered) ES equivalents

Station: Rep TOM Redox Sulphides N S j d H' AMBI M-AMBI BQI TOM Redox Sulphides N j d H' AMBI M-AMBI BQI

Organic

Loading

Sediment

chemistry

Macro

fauna Overall ES

Pen 1 A 4.4 165 351 1859 59 0.5 7.7 2.04 4.22 0.6 6.42 3 2.64 3.21 3.6 2.99 1.73 2.35 3.59 3.04 2.38 3 2.93 2.81 2.85

Pen 1 B 3.7 138 409 1694 65 0.55 8.61 2.29 3.99 0.67 6.94 2 2.89 3.31 3.53 2.75 1.58 2.13 3.35 2.61 2.18 2 3.1 2.59 2.63

Pen 1 C 3.7 266 379 1602 61 0.53 8.13 2.2 3.49 0.68 6.17 2 1.74 3.26 3.49 2.84 1.63 2.21 2.84 2.6 2.49 2 2.5 2.59 2.51

Pen 2 A 3.6 300 131 3913 70 0.49 8.34 2.1 4.05 0.66 7.34 2 1.43 2.58 4.17 3.03 1.6 2.29 3.41 2.66 2.04 2 2 2.74 2.52

Pen 2 B 3.9 114 379 5341 51 0.46 5.83 1.8 4.65 0.51 6.05 2 3.1 3.26 4.41 3.18 2.5 2.61 4.02 3.63 2.54 2 3.18 3.27 3.13

Pen 2 C 4 94 302 2171 57 0.49 7.29 2 4.35 0.58 6.67 3 3.28 3.12 3.72 3.03 1.85 2.39 3.72 3.18 2.28 3 3.2 2.88 2.96

Pen 3 A 3.5 247 177 1590 58 0.59 7.73 2.41 4.03 0.66 7.3 2 1.91 2.77 3.48 2.56 1.72 2.04 3.39 2.7 2.05 2 2.34 2.56 2.46

Pen 3 B 3.7 225 67 1385 49 0.44 6.64 1.72 3.96 0.54 5.3 2 2.1 2.14 3.37 3.27 2.1 2.71 3.32 3.47 2.91 2 2.12 3.02 2.74

Pen 3 C 3.5 295 476 2562 57 0.37 7.14 1.5 4.31 0.52 5.56 2 1.47 3.41 3.85 3.61 1.9 3.01 3.68 3.59 2.78 2 2.44 3.2 2.93

75 N A 3 92 810 478 29 0.37 4.54 1.26 3.89 0.41 4.06 1 3.3 3.76 2.56 3.61 3.26 3.37 3.25 4.35 3.67 1 3.53 3.44 3.21

75 N B 2.7 226 379 1108 46 0.35 6.42 1.35 3.79 0.49 5.08 1 2.1 3.26 3.2 3.7 2.2 3.23 3.15 3.79 3.03 1 2.68 3.19 2.87

75 N C 3.2 278 141 2174 55 0.33 7.03 1.31 3.81 0.52 5.03 1 1.63 2.63 3.72 3.8 1.94 3.28 3.16 3.58 3.06 1 2.13 3.22 2.78

300 N A 2.5 237 513 174 41 0.84 7.75 3.11 2.38 0.78 8.87 1 2 3.46 1.79 1.36 1.72 1.76 1.7 2.08 1.69 1 2.73 1.73 1.85

300 N B 2.1 197 379 169 38 0.78 7.21 2.85 2.05 0.76 8.83 1 2.36 3.26 1.76 1.65 1.88 1.82 1.37 2.18 1.69 1 2.81 1.76 1.9

300 N C 2.2 317 104 343 40 0.57 6.68 2.1 1.93 0.68 7.91 1 1.28 2.43 2.31 2.65 2.09 2.29 1.25 2.59 1.88 1 1.85 2.15 1.97

300 S A 3.3 141 67 847 79 0.72 11.57 3.16 2.31 0.94 9.45 1 2.86 2.14 3 1.93 1.5 1.76 1.63 1.74 1.62 1 2.5 1.88 1.92

300 S B 3.5 252 67 372 55 0.79 9.12 3.15 1.7 0.88 10.49 1 1.86 2.14 2.37 1.6 2 1.76 1.01 1.78 1.59 1 2 1.73 1.71

300 S C 3.4 197 164 1390 79 0.62 10.78 2.72 2.06 0.9 9.86 1 2.36 2.72 3.38 2.41 1.7 1.87 1.38 1.75 1.6 1 2.54 2.01 2.02

TC Ctl 1 A 4.7 171 156 271 38 0.7 6.6 2.55 1.98 0.72 9.76 1 2.59 2.69 2.12 2.03 2.12 1.95 1.3 2.34 1.6 1 2.64 1.92 1.97

TC Ctl 1 B 4.3 129 211 223 32 0.76 5.73 2.65 2 0.71 10.47 1 2.97 2.89 1.98 1.74 2.55 1.9 1.32 2.41 1.59 1 2.93 1.93 2.04

TC Ctl 1 C 4.5 151 181 313 41 0.71 6.96 2.62 1.94 0.75 10.01 1 2.77 2.79 2.24 1.98 1.97 1.91 1.25 2.22 1.59 1 2.78 1.88 1.97

TC Ctl 3 A 3.3 352 124 224 36 0.74 6.47 2.64 2.06 0.72 10.43 1 0.96 2.54 1.98 1.84 2.18 1.91 1.38 2.35 1.59 1 1.75 1.89 1.77

TC Ctl 3 B 3.2 186 156 217 32 0.71 5.76 2.48 2.04 0.68 10.84 1 2.46 2.69 1.95 1.98 2.54 2 1.36 2.54 1.61 1 2.57 2 2.01

TC Ctl 3 C 3.1 155 309 397 44 0.72 7.19 2.71 2.22 0.75 10.59 1 2.73 3.13 2.42 1.93 1.88 1.87 1.54 2.2 1.6 1 2.93 1.92 2.03

TC Ctl 6 A 2.9 278 71 653 61 0.68 9.26 2.79 1.67 0.86 10.12 1 1.63 2.18 2.8 2.12 2 1.84 0.98 1.82 1.59 1 1.9 1.88 1.8

TC Ctl 6 B 3 365 91 645 70 0.67 10.67 2.83 1.7 0.9 9.75 1 0.84 2.34 2.79 2.17 1.7 1.83 1.01 1.75 1.6 1 1.59 1.84 1.7

TC Ctl 6 C 3.3 143 71 620 46 0.61 7 2.34 1.37 0.77 9.43 1 2.84 2.18 2.76 2.46 1.96 2.09 0.67 2.14 1.62 1 2.51 1.96 1.97


41

Table A2.2. Summary of the average (SE) sediment physical and chemical properties, macrofauna variables and calculated indices for the Ngamahau Bay salmon farm stations during the February 2019 monitoring survey.

Units Pen1 Pen 2 Pen 3 75 N 300 N 300 S

Depth m 33 30 32 28 26 40

Sed

imen

ts

AFDW % 3.9 (0.2) 3.8 (0.1) 3.6 (0.1) 3 (0.1) 2.3 (0.1) 3.4 (0.1)

Redox EhNHE, mV 189.7 (39) 169.3 (65.6) 255.7 (20.7) 198.7 (55.4) 250.3 (35.3) 196.7 (32)

Sulphides* µM 379.7 (16.7) 270.7 (73.3) 240 (122.2) 443.3 (195.8) 332 (120.4) 99.3 (32.3)

Bacterial mat

- No No No No No No

Outgassing - No No No No No No

Odour - No No No No No No

Ma

cro

fau

na s

tati

sti

cs

Abundance No./core 1718.3 (75.2) 3808.3 (916.6) 1845.7 (363) 1253.3 (495) 228.7 (57.2) 869.7 (294.1)

No. taxa No./core 61.7 (1.8) 59.3 (5.6) 54.7 (2.8) 43.3 (7.6) 39.7 (0.9) 71 (8)

Evenness Stat. 0.5 (0.0) 0.5 (0.0) 0.5 (0.1) 0.4 (0.0) 0.7 (0.1) 0.7 (0.0)

Richness Stat. 8.1 (0.3) 7.2 (0.7) 7.2 (0.3) 6 (0.7) 7.2 (0.3) 10.5 (0.7)

SWDI Index 2.2 (0.1) 2 (0.1) 1.9 (0.3) 1.3 (0.0) 2.7 (0.3) 3 (0.1)

AMBI Index 3.9 (0.2) 4.3 (0.2) 4.1 (0.1) 3.8 (0.0) 2.1 (0.1) 2 (0.2)

mAMBI Index 0.6 (0.0) 0.6 (0.0) 0.6 (0.0) 0.5 (0.0) 0.7 (0.0) 0.9 (0.0)

BQI Index 6.5 (0.2) 6.7 (0.4) 6.1 (0.6) 4.7 (0.3) 8.5 (0.3) 9.9 (0.3)


42

Table A2.2. continued Summary of the average (SE) sediment physical and chemical properties, macrofauna variables and calculated indices for the Ngamahau Bay salmon farm reference stations during the February 2019 monitoring survey.

Units TC-Ctl-1 TC-Ctl-3 TC-Ctl-6

Depth m 24 31 30

Sed

imen

ts

AFDW % 4.5 (0.1) 3.2 (0.1) 3.1 (0.1)

Redox EhNHE, mV 150.3 (12.1) 231 (61.2) 262 (64.6)

Sulphides* µM 182.7 (15.9) 196.3 (57.1) 77.7 (6.7)

Bacterial mat

- No No No

Outgassing - No No No

Odour - No No No

Ma

cro

fau

na s

tati

sti

cs

Abundance No./core 269 (26) 279.3 (58.9) 639.3 (9.9)

No. taxa No./core 37 (2.6) 37.3 (3.5) 59 (7)

Evenness Stat. 0.7 (0.0) 0.7 (0.0) 0.7 (0.0)

Richness Stat. 6.4 (0.4) 6.5 (0.4) 9 (1.1)

SWDI Index 2.6 (0.0) 2.6 (0.1) 2.7 (0.2)

AMBI Index 2 (0.0) 2.1 (0.1) 1.6 (0.1)

mAMBI Index 0.7 (0.0) 0.7 (0.0) 0.8 (0.0)

BQI Index 10.1 (0.2) 10.6 (0.1) 9.8 (0.2)


43

Figure A2.3. Representative images of the seafloor at each of the Type 3 soft-sediment sampling

stations at the Ngamahau Bay salmon farm (NGA), February 2019. Drop camera footage was not obtained from stations T3a1, T3a4 and T3b1 through T3b6.


44

Figure A2.4. Sediment organic matter (% ash-free dry weight; AFDW), redox potential (EhNHE, mV),

total free sulphides (µM) and macrofauna statistics determined at the Type 3 soft-sediment sampling stations at the Ngamahau Bay salmon farm (NGA), February 2019. PS-Ctl = Pelorus Sound control. Error bars = ± 1 SE, n = 3.


45

Table A2.3. Detailed Enrichment Stage (ES) calculations for the Type 3 soft-sediment sampling stations at the Ngamahau Bay salmon farm (NGA), February 2019. For details about how these values were calculated, see MPI (2015). Underlined text are cases where best professional judgement (BPJ; Keeley et al. 2012b) was used. Note that ES calculations in previous annual monitoring reports did not implement the rules from Appendix 10.2: bullet points 2b&c from MPI (2015). Asterisks denote overall ES scores in which total free sulphide values were not included.

SITE INFORMATION

Date: Feb-19 Variable group weightings:

Farm/site: Ngamahau Bay 0.1 0.2 0.7

Flow environment: HF

RAW DATA (to be entered) ES equivalents

Station: Rep TOM Redox Sulphides N S j d H' AMBI M-AMBI BQI TOM Redox Sulphides N j d H' AMBI M-AMBI BQI

Organic

Loading

Sediment

chemistry

Macro

fauna Overall ES

T3a1 A 3.3 145 n/c 519 56 0.69 8.8 2.78 1.68 0.84 9.7 1 2.82 NA 2.62 2.08 1.57 1.85 0.99 1.87 1.6 1 2.82 1.8 1.92*

T3a1 B 3 209 n/c 410 48 0.7 7.81 2.73 1.39 0.82 10.68 1 2.25 NA 2.44 2.03 1.7 1.86 0.69 1.92 1.6 1 2.25 1.75 1.77*

T3a2 A 2.5 143 351 553 45 0.71 6.97 2.72 2.4 0.74 7.33 1 2.84 3.21 2.67 1.98 1.97 1.87 1.73 2.26 2.04 1 3.03 2.07 2.16

T3a2 B 2.3 115 n/c 636 61 0.68 9.29 2.8 2.41 0.81 7.54 1 3.09 NA 2.78 2.12 1.37 1.84 1.74 1.96 1.98 1 3.09 1.97 2.1*

T3a3 A 3 331 67 518 48 0.74 7.52 2.87 3.19 0.71 8.66 1 1.15 2.14 2.62 1.84 1.78 1.81 2.53 2.41 1.72 1 1.65 2.1 1.9

T3a3 B 3.9 243 67 809 64 0.71 9.41 2.96 2.65 0.82 9.16 2 1.94 2.14 2.96 1.98 1.36 1.79 1.98 1.92 1.65 2 2.04 1.95 1.97

T3a4 A 4 213 n/c 650 68 0.7 10.34 2.94 2.09 0.88 9.13 3 2.21 NA 2.8 2.03 1.24 1.79 1.41 1.79 1.65 3 2.21 1.81 2.01*

T3a4 B 3.6 202 n/c 743 76 0.73 11.35 3.16 2.05 0.94 9.16 2 2.31 NA 2.9 1.89 1.1 1.76 1.37 1.73 1.65 2 2.31 1.77 1.9*

T3b1 A 3.1 114 196 753 69 0.67 10.27 2.84 2.88 0.81 8.17 1 3.1 2.84 2.91 2.17 1.25 1.82 2.22 1.95 1.82 1 2.97 2.02 2.11

T3b2 A 3 157 n/c 997 80 0.53 11.44 2.34 1.65 0.89 9.39 1 2.72 NA 3.12 2.84 1.09 2.09 0.96 1.76 1.63 1 2.72 1.93 1.99*

T3b3 A 3.8 299 71 852 59 0.66 8.6 2.68 2.71 0.77 8.08 2 1.44 2.18 3 2.22 1.58 1.89 2.04 2.13 1.84 2 1.81 2.1 2.03

T3b4 A 3.2 -124 714 1187 57 0.47 7.91 1.88 3.47 0.62 5.93 1 5.25 3.67 3.26 3.13 1.68 2.51 2.82 2.94 2.6 1 4.46 2.71 2.89

T3b5 A 2.9 335 71 482 57 0.71 9.06 2.85 3.01 0.76 8.35 1 1.11 2.18 2.57 1.98 1.4 1.82 2.35 2.18 1.78 1 1.65 2.01 1.84

T3b6 A 3.6 145 115 381 43 0.65 7.07 2.44 2.09 0.72 7.46 2 2.82 2.49 2.39 2.27 1.93 2.02 1.41 2.34 2 2 2.66 2.05 2.17

T3b7 A 3.5 220 141 352 45 0.74 7.5 2.83 2.56 0.74 7.22 2 2.15 2.63 2.33 1.84 1.78 1.83 1.89 2.26 2.08 2 2.39 2 2.08


46

Table A2.4. Summary of the average (SE) sediment physical and chemical properties, macrofauna variables and calculated indices for the Type 3 soft-sediment sampling stations at the Ngamahau Bay salmon farm (NGA), during the February 2019 monitoring survey.

Units T3a1 T3a2 T3a3 T3a4 T3b1 T3b2 T3b3 T3b4 T3b5 T3b6 T3b7

Depth m 41 10 40 21.5 34.5 33 35.3 26 35 24.2 30

Sed

imen

ts AFDW % 3.1 (0.1) 2.4 (0.1) 3.5 (0.4) 3.8 (0.2) 3.1 3 3.8 3.2 2.9 3.6 3.5

Redox EhNHE, mV 177 (32) 129 (14) 287 (44) 207.5 (5.5) 114 157 299 -124 335 145 220

Sulphides* µM n/c* 351 67 (0.0) n/c* 196 n/c* 71 714 71 115 141

Bacterial mat** - N N N N N N N N N N N

Outgassing** - N N N N N N N N N N N

Odour - N N N N N N N N N N N

Mac

rofa

un

a s

tati

sti

cs

Abundance No./core 464.5 (54.5) 594.5 (41.5) 663.5 (145.5) 696.5 (46.5) 753 997 852 1187 482 381 352

No. taxa No./core 52 (4) 53 (8) 56 (8) 72 (4) 69 80 59 57 57 43 45

Evenness Stat. 0.7 (0.0) 0.7 (0.0) 0.7 (0.0) 0.7 (0.0) 0.7 0.5 0.7 0.5 0.7 0.6 0.7

Richness Stat. 8.3 (0.5) 8.1 (1.2) 8.5 (0.9) 10.8 (0.5) 10.3 11.4 8.6 7.9 9.1 7.1 7.5

SWDI Index 2.8 (0.0) 2.8 (0.0) 2.9 (0.0) 3.1 (0.1) 2.8 2.3 2.7 1.9 2.9 2.4 2.8

AMBI Index 1.5 (0.1) 2.4 (0.0) 2.9 (0.3) 2.1 (0.0) 2.9 1.7 2.7 3.5 3 2.1 2.6

mAMBI Index 0.8 (0.0) 0.8 (0.0) 0.8 (0.1) 0.9 (0.0) 0.8 0.9 0.8 0.6 0.8 0.7 0.7

BQI Index 10.2 (0.5) 7.4 (0.1) 8.9 (0.3) 9.1 (0.0) 8.2 9.4 8.1 5.9 8.4 7.5 7.2

*Total sulphide concentrations were not collected

**Sediment cores and grab footage were consulted for evidence of bacterial mat coverage and outgassing in the event that drop camera footage was

unavailable.


47

Appendix 3. Water column sampling methodology and compliance framework.

A3.1 Background

The following sub-sections provide detail on the water column sampling methodology.

The NGA water column sampling stations established in the baseline monitoring

(Morrisey et al. 2014; Morrisey et al. 2015) were for the ongoing NZ King Salmon

routine and full-suite monitoring programmes (see Table 4 and Figure 6 in Section 2.2

for an overview of the sampling regime). This design includes one of the MDC State

of the Environment (SoE) monitoring stations (NZKS21/QCS3) that is sampled by

MDC. In addition, results from two other SoE monitoring stations in Tory Channel

(QCS7, QCS8) are included where this additional context is useful. Extra sampling

stations for fine-scale water column monitoring were included in this design during the

fine-scale sampling months (March and August).

A3.2 Sampling protocol

On all sampling occasions, water column depth profile data were collected at each

station using a conductivity-temperature-depth (CTD) instrument with an attached

dissolved oxygen (DO) sensor. Parameters measured were salinity, temperature,

turbidity15 and DO. In addition, single, surface-integrated samples were taken over

the upper 15 m of the water column (obtained using a weighted hose) and analysed

for total nitrogen (TN) and chlorophyll-a (chl-a). Additional parameters measured in

February, March, July and August included silicate (DRSi), nitrogen species and

phosphorus species (Table A3.1), as well as phytoplankton composition and

biomass. Samples were stored on ice, except for phytoplankton samples, which were

preserved with Lugol’s acidified iodine solution and kept cool.

For the fine-scale monitoring (March and August), variations to the sampling

procedure were as follows:

• In addition to the surface-integrated sample at each station, a single water sample

from near the seabed (‘near-bed’) was collected using a van Dorn sampler.

• Duplicate surface-integrated samples were collected in March 2018 from the pen

station, and the 100 m, 250 m and 500 m downstream stations16. Replicate

samples were taken from a single, well-mixed, bucket of seawater. Each bucket

comprised a unique deployment of the 15 m hose sampler. The variability

between the replicates thus represents that introduced by ‘water parcel’ or

horizontal ‘spatial’ variability occurring at the sampling station during the time it

takes to be sampled. Because single samples were collected in August, means

15 Turbidity was used as a proxy for clarity, as turbidity data show the water column profile rather than just

surface characteristics. 16 Sometimes additional 500 m or reference stations were sampled with replication, but not consistently.


48

from the two samples were used for March data. Chlorophyll-a and phytoplankton

were not sampled in replicate or in near-bed samples.

• Monthly routine and full-suite samples were collected by MDC staff, at the same

time as wide-scale SoE monitoring in Queen Charlotte Sound (led by MDC).

Cawthron staff carried out sampling for fine-scale monitoring alongside MDC in

March and August 2018.

Table A3.1. Parameters to be measured at the sampling locations during the 2018-2019 monitoring. S = surface integrated water samples, N = near-bed water samples, CTD = profiles taken throughout the water column with electronic sensors. Italicised parameters (under fine-scale sampling) were removed from the sampling plan under the current MEMAMP (Bennett et al 2018a) and were therefore sampled in March only (under the previous MEMAMP; Elvines & Knight 2017). Note that July full-suite surveys were postponed until September. Nutrient samples from December are missing from the dataset as the samples were lost by the courier in transit. Phytoplankton sampling was delayed from February until early March but are labelled here as ‘February’ to distinguish them from the subsequent March sampling event.

Monitoring

component

Routine and full-suite Fine-scale

Station Net pen

(NZKS18)

500 m

(NZKS19,

20)

NZKS22 NZKS21 100 & 250 m

Parameter

TP Mar, Aug (N) - - - Mar, Aug (N)

DRP Mar, Aug (N) - - Mar, Aug (N) Mar, Aug (N)

DRSi Mar, Aug (S) - - Mar, Aug (S) Mar (N)

Phytoplankton Feb, Mar, Aug,

Sep (S)

- Feb, Mar, Aug, Sep (S) Mar (N)

NO2/NO3 Feb (S), Mar (S, N), Aug (S, N), Sep (S) Mar, Aug (S, N)

NH4 Feb (S), Mar (S, N), Aug (S, N), Sep (S) Mar, Aug (S, N)

Urea-N Mar, Aug (S) Mar, Aug (S)

PN Mar, Aug (S) Mar, Aug (S)

Turbidity * Feb, Mar, Aug, Sep (CTD) Mar, Aug (CTD)

Temp Feb, Mar, Aug, Sep (CTD) Mar, Aug (CTD)

Sal Feb, Mar, Aug, Sep (CTD) Mar, Aug (CTD)

Chl-a Monthly (S) Mar (N)

TN Monthly (S) Mar, Aug (S, N) Mar, Aug (S, N)

DO Monthly (CTD) Mar, Aug (CTD)

*Also proxy for clarity.


49

A3.3 Sample analysis

Samples were analysed for nutrients using routine methods (Table A3.2). In addition,

the remaining filtered sample water was archived at the laboratory in case follow-up

nutrient analyses were required (i.e. if an amber alert was triggered—see Elvines &

Knight 2017).

Algal taxonomic composition (species abundance) was determined from a subsample

of the 15 m depth integrated sample. Algal taxonomic composition was determined by

a modified Utermöhl method based on published Intergovernmental Oceanographic

Commission (IOC) methods (Karlson et al. 2010). For this process, each sample is

analysed using inverted light microscopy to identify and enumerate all taxa detected

in the sample to the finest practicable taxonomic level by IANZ accredited staff.

Sample bio-volume was estimated for recorded species and used to estimate cell

carbon content (biomass) (Table A3.2).


50

Table A3.2. Laboratory analytical methods for water samples processed by the NIWA laboratory in Hamilton.

Analyte Method Default

detection limit

Chlorophyll-a (c) (chl-a)

Acetone pigment extraction, spectrofluorometric measurement. A*10200H.

0.1 mg/m3

Dissolved reactive silicon (c) (DRSi)

Molybdosilicate / ascorbic acid reduction. APHA4500Si.

1 mg/m3

Total phosphorus (c) (TP)

Persulphate digest, molybdenum blue FIA. Lachat. 1 mg/m3

Urea nitrogen(c) (Urea-N)

Automated diacetyl-monoxime colorimetry. MSeawater.

1 mg/m3

Nitrate and nitrite nitrogen (c) (NO3-N)

DRP, NH4-N, NO3-N, Simultaneous Auto-analysis. Astoria.

1 mg/m3

Ammonium nitrogen (c) (NH4-N)


1 mg/m3

Dissolved reactive phosphorus (c) (DRP)


1 mg/m3

Total nitrogen (c) (TN)

Persulphate digest, auto cadmium reduction, FIA. Lachat.

10 mg/m3

Particulate nitrogen (c) (PN)

Calculation of TN–TDN (TDN determined by persulphate digest, auto cadmium reduction, FIA). Lachat.

10 mg/m3

Dissolved inorganic nitrogen (DIN)

Derived using NO3-N + NO2-N + NH4-N.

Phytoplankton biovolume (b)

From Morrisey et al. (2015): Estimated for each taxon using formulae representing the geometrical solids that approximated cell shape (Rott 1981, Hillebrand et al. 1999).

Phytoplankton carbon biomass (b)

From Morrisey et al. (2015): Cell numbers and biovolumes were used to calculate cell carbon using regression equations of Meden-Deuer and Lessard (2000) for dinoflagellates and cyanobacteria, and that of Cornet- Barthaux et al. (2007) for diatoms.


51

A3.4 Compliance framework for water quality monitoring results

The environmental monitoring results from water quality monitoring are used to

determine whether the farms are compliant with the environmental quality standards

(EQS: water) specified in the consent conditions.

A3.4.1 Water quality standards

Initial water quality standards (WQS) developed by Morrisey et al. (2015) for the

Waitata (WTA), Kopaua (KOP) and Ngamahau (NGA) sites set specific thresholds for

three parameters: chlorophyll-a, DO and TN. If these thresholds are exceeded in

three consecutive months, then an ‘amber state’ is reached, and further action must

be taken (see MEMAMP, Elvines & Knight 2017). The current WQS are discussed

and specified in the MEMAMP and are summarised in Table A3.3.

Because WQS only exist for TN, chlorophyll-a and DO, the additional

parameters/analytes that are measured in February and July (full-suite monitoring)

cannot be measured against WQS for compliance as required by the consent

(Condition 66c):

Monitoring in order to determine compliance with the WQS in Condition 44.

Throughout the term of the consent this shall include long-term water column

monitoring for nutrient (NH4-N, NO3-N, NO2-N, DRP, Si, TN and TP) and

chlorophyll a concentrations, phytoplankton composition and biomass,

salinity, clarity, temperature, turbidity and dissolved oxygen (DO) at locations

stipulated in Condition 63c. The precise location of the long- term monitoring

stations and the range of specific nutrient parameters monitored may,

however, be adjusted over time in response to monitoring results and/or in

response to modelling considered necessary by the Peer Review Panel in

accordance with Condition 70c. This monitoring is to be undertaken at least

four times per year with at least two surveys occurring during mid-summer

periods of highest salmon feed discharge rates and at least two surveys

occurring periods associated with winter/spring and/or autumn diatom

maxima.

Discussion of results from fine-scale sampling for additional parameters/analytes is

limited to spatial patterns, to fulfil Condition 66e:

Targeted water column surveys to quantify the localised effect of the marine

farm on surrounding water quality, for the purpose of obtaining information

regarding marine farm-specific, near-farm mixing properties in order to

provide a context for evaluating compliance with the WQS in Condition 44.

This shall involve a series of fine-scale surveys in the vicinity of the marine

farm (within 1km from the net pens) measuring: salinity, clarity, temperature,


52

chlorophyll a, turbidity, dissolved oxygen (DO), nutrient concentrations (NH4-

N, NO3-N, NO2-N, DRP, Si, TN and TP) phytoplankton composition and

biomass along transects that move away from the marine farm and span

potential nutrient gradients. The surveys shall be undertaken at least twice

per year and continued for at least two years after the marine farm has

reached stable maximum feed discharge levels and no future increases are

proposed. With respect the monitoring objective, the monitoring approach

may be adjusted over time in accordance with the written recommendation of

the Peer Review Panel.

Table A3.3. Water quality standards (WQS) for chlorophyll-a (chl-a), total nitrogen (TN) and dissolved oxygen (DO) for the Kopaua, Waitata Reach and Ngamahau Bay salmon farm consents, 2017–2018. The second step threshold takes into account reference values. Further discussion of the WQS and how they are applied can be found in the MEMAMP (Elvines & Knight 2017).

chlorophyll-a TN DO

WQS ≤ 3.5 mg/m3 ≤ 300 mg TN/m3 > 90% > 70%

Second step

threshold

n/a To be determined ≤ 1.2% lower than

applicable reference

stations (e.g. far-field,

upstream 500 m)

Sample 0-15 m depth integrated

sample

0-15 m depth integrated

sample

All depths,

bin mean of

1 m.

All depths,

bin mean of

1 m.

Location All stations Stations > 250 m from farm

(Stations < 250 m may

exceed these levels)

Stations

> 250 m

from farm

Stations

< 250 m

from farm

Tolerance Three consecutive months: at any one station, or at any station within the same

sound for three consecutive months

A3.4.2 Water quality objectives

In addition to WQS that are set for some parameters, there are also water quality

objectives specified in the consent (Condition 43):

43. The marine farm shall be operated at all times in such a way as to

achieve the following Water Quality Objectives in the water column:


53

a. To not cause an increase in the frequency, intensity or duration of

phytoplankton blooms (i.e. chlorophyll-a concentrations ≥ 5 mg/m3)

[Note: water clarity as affected by chlorophyll-a concentrations is

addressed by this objective];

b. To not cause a change in the typical seasonal patterns of

phytoplankton community structure (i.e. diatoms vs. dinoflagellates),

and with no increased frequency of harmful algal blooms (HABs)

(i.e. exceeding toxicity thresholds for HAB species);

c. To not cause reduction in dissolved oxygen concentrations to levels

that are potentially harmful to marine biota [Note: Near bottom

dissolved oxygen under the net pens is addressed separately

through the EQS – Seabed Deposition];

d. To not cause elevation of nutrient concentrations outside the

confines of established natural variation for the location and time of

year, beyond 250 m from the edge of the net pens;

e. To not cause a statistically significant shift, beyond that which is

likely to occur naturally, from an oligotrophic/mesotrophic state

towards a eutrophic state;

f. To not cause an obvious or noxious build-up of macroalgal (e.g. sea

lettuce) biomass [Note: to be monitored in accordance with

Condition 66h].


54

Appendix 4. Additional detail on the results of the 2018 Ngamahau Bay (NGA) salmon farm water column monitoring.

A4.1 Salinity, temperature, and turbidity

Overall, the depth profiles and average temperature and salinity data (see

Figure A4.1 and Figure A4.2) indicate a well-mixed water column in this area, owing

to the high tidal currents experienced in Tory Channel.

There is some evidence of warming of the waters with the consequent reduction in

salinity in February (Figure A4.1). Average salinity between stations did not vary by

more than 1.5 PSU (Figure A4.2). There was no evidence of temperature-driven

stratification at any of the stations.

There were no substantial reductions in turbidity with depth at any of the stations,

except the top 10 m at the CE Ref (NZKS21) in February. In addition, there was no

evidence of region-wide increases in turbidity across the time series to date that

might indicate reductions in water quality due to eutrophication (Appendix 5).


55

Figure A4.1. Water column profile of salinity (PSU), temperature (°C), and turbidity (NTU) (1 m depth

binned downcast data) at sampling stations in February, March, August and September 2018. August and March data were collected using an SBE 19+ CTD, and February and July data using a YSI EXO Sonde CTD. CE = cumulative effect, FF = far-field, Ref = reference.


56

Figure A4.2. Average temperature and salinity from 1.5–15 m depth for each routine sampling station

in Tory Channel. QCS7 & 8; further inside Tory Channel, are also plotted for context. Note data from some stations were excluded where a different CTD instrument was used (i.e. QCS7, QCS8 and NZKS21 in March and August). March and August data are from a Seabird 19+ CTD, while all other data are from a YSI EXO Sonde CTD. CE = cumulative effect, FF = far-field, Ref = reference.


57

A4.2 Dissolved oxygen

Monthly DO saturations at the NGA net pen (NZKS18) ranged from 82.5% (March) to

97.1% (October). The minimum DO saturations were within the WQS in all months

(i.e. > 70%; Table 5 and Table A4.1).

Both stations 500 m from the NGA farm (NZKS19 and NZKS20), as well as the CE-

Ref (NZKS21) and the FF-Ref (NZKS22) stations, exceeded the applicable WQS (i.e.

DO > 90%) during January, March and August (Table 5 and Table A4.1). DO

saturations at NZKS20 and NZKS21 were at the WQS threshold (i.e. 90%) in April.

NZKS21 also exceeded the WQS threshold in June. The second step DO WQS

threshold (WQS [2]) was exceeded at NZKS19, NZKS20 and NZKS21 in March and

at NZKS20 and NZKS21 in April. Although the time series of DO data does not

indicate a reduction outside historical levels and thus a shift in water quality

associated with eutrophication (see Appendix 5), it would be appropriate to undertake

a more detailed analysis of monthly DO profiling data to better understand the likely

causes of DO WQS exceedances.

Note that in March the water column profiles at the far-field control stations were

taken with the YSI EXO Sonde CTD instrument (used by MDC) while at the near-farm

stations parameters were measured with the Seabird 19+ instrument (used by

Cawthron). The YSI instrument consistently measures higher dissolved oxygen than

the Seabird, and this is very likely to have contributed to the exceedances of WQS [2]

in March. The tendency of the YSI instrument to record higher DO than the Seabird

was taken into account following the March monitoring; the far field reference stations

are now sampled with the Seabird CTD during fine-scale monitoring months (as

occurred in August) to improve comparability of far-field and near-farm data. This

tendency for the Seabird to measure lower DO than the YSI instruments also

potentially overstates the degree of reduction in DO observed from CTD profiles in

March, as described above. It may be appropriate to further consider the implications

of the differing instrumentation as part of the finalisation or implementation of the

BMP guidelines. However, the pattern of low DO, and extent of the footprint out to the

CE Ref remain an important observation.

No near-bed reductions in DO were observed in the CTD profiles for February,

March, August or September to indicate higher than usual biological activity on the

seabed (Figure A4.3). In September, DO saturations throughout the water column at

the net pen were lower than at all other stations. However, none of the DO

saturations exceeded the DO WQS threshold at 250 m.


58

Table A4.1. Minimum dissolved oxygen (DO) saturation (%) (1 m depth binned downcast data) at all stations. Both the first step (WQS [1]) and second step are provided (WQS [2]; see Appendix 3). WQS are shown where applicable. Bolded values indicate those below the WQS (1). CE = cumulative effect, FF = far-field, Ref = reference. March and August data are from a Seabird 19+ CTD, while all other data are from a YSI EXO Sonde CTD.

Ngamahau Bay farm (NGA)

NZKS18 NZKS19 NZKS20 NZKS21 NZKS22 WQS (2) *

Month Net pen 500 m

south

500 m

north CE-Ref FF-Ref

Jan 90.3 88 87.4 88.3 88 ≥ 87.1

Feb 92.8 92.4 93 91.3 98.3

Mar 82.5 71.4 74.6 70.8 84.2 ≥ 83.2

Apr 90.1 91.3 90 90 91.4 ≥ 90.3

May 93.6 93 94.1 93.1 94.6

Jun 89.8 90.2 90.8 89.8 92.9 ≥ 88.7

Jul 94.4 94 94.8 93.6 95.2

Aug 87.5 87.5 87 87.2 86.7 ≥ 85.7

Sep 92.9 97.3 97.7 98.9 97.1

Oct 97.1 98.6 98.7 99.2 98

Nov 96.4 96.8 96.4 96.8 97.4

Dec 96.6 94.5 94.9 93.6 95.2

WQS (1) > 70% > 90%

*The second step WQS threshold is month-specific, and is calculated by subtracting 1.2% from the

average of applicable reference station DO saturations (also see Appendix 3).


59

Figure A4.3. Dissolved oxygen (% saturation) (1-m binned depth binned downcast data) at routine and fine-scale sampling stations in February, March, August and September 2018. d/s = downstream, u/s = upstream, CE = cumulative effect, FF = far-field, Ref = reference.

A4.3 Nutrients

A4.3.1 Monthly results - total nitrogen (TN)

With one exception, TN concentrations at all stations were within the TN WQS

(i.e. ≤ 300 mg-N/m3) (Table A4.2). The exception was the NZKS21 station in the

middle of Tory Channel in May (309 mg-N/m3). This station had the highest

concentration of TN recorded in 2017, at 305 mg-N/m3 in June (Bennett et al. 2018b).

Similar one-off exceedances of the TN WQS were observed at this station in 2016

(see Appendix 5). Morrisey et al. (2015) showed that background concentrations of

TN > 300 mg/m3 can occur on an annual basis, albeit on ‘rare’ occasions. However,


60

the time-series of data available for TN (Appendix 5) does not suggest an increased

frequency of samples with high nitrogen concentrations in this region since the farm

began operating.

A4.3.2 Fine-scale sampling results – nitrogen

No gradient was evident in TN or Urea-N concentrations during fine-scale sampling in

March, and concentrations around the farm were similar to upstream and mid-channel

concentrations on the incoming tide. However, in the August sampling a clear trend of

decreasing TN and Urea-N at the surface was seen with increasing distance from the

farm. The greatest reduction in TN concentrations occurred at the surface in the first

100 m downstream of the farm. In March (when particulate nitrogen [PN] was

sampled), PN was highly variable across all sampling stations, with elevated

concentrations at 100 m and 250 m downstream of the farm and no obvious farm-

related gradient.

There were also near-bed reductions in concentrations of nitrate/nitrite

(NO3-N + NO2-N) in March and August downstream to 250 m from the farm. However,

near-bed concentrations of these nitrogen forms were considerably elevated at 500 m

from the farm and reference stations. Therefore, no trend with distance from the farm

can be established.

Concentrations of ammonium (NH4) at the net pen were slightly higher than

concentrations at stations within 250 m from the farm. However, concentrations in

surface water samples at the CE-Ref station (NZKS21) were at the same level as

concentrations at the net pen (Figure A4.4) and again no trend with distance is

apparent.

Urea-N showed a decreasing trend 500 m downstream from the farm during fine-scale

sampling in surface waters in August. However, concentration of this nutrient form

was also elevated at the CE-Ref station (NZKS21) on this occasion (Figure A4.4).


61

Table A4.2. Surface integrated results for total nitrogen (mg/m3) for all months. Shaded values indicate those above the WQS. CE = cumulative effect, FF = far-field, ref = reference.

NZKS18 NZKS19 NZKS20 NZKS21

(QCS3)

NZKS22

Month Net pen

500 m

north

500 m

south CE-Ref FF-Ref

Jan 189.0 209.0 229.0 199.0 176.0

Feb 165.0 171.0 143.0 158.0 152.0

Mar* 176.0 164.5 173.5 226.7 189.0

Apr 171.0 153.0 170.0 158.0 156.0

May 278.0 227.0 208.0 309.0 207.0

Jun 263.0 216.0 234.0 243.0 255.0

Jul 220.0 233.0 240.0 221.0 245.0

Aug 238.0 161.0 156.0 214.0 162.0

Sep 236.0 223.0 247.0 191.0 210.0

Oct 166.0 180.0 146.0 234.0 160.0

Nov 183.0 185.0 181.0 173.8 178.0

Dec 201.0 200.0 233.0 298.0 205.0

WQS n/a ≤ 300 mg-N/m3 n/a

* Mean value across duplicate samples.

A4.3.2 Fine scale sampling results - phosphorus

In March, near-bed concentrations of total phosphorus (TP) and dissolved reactive

phosphorus (DRP) reduced with distance from the farm although differences were not

large. In August, there was no spatial gradient evident in TP or DRP concentrations

across near-bed samples.


62

Figure A4.4. Concentrations (mg/m3) of nutrients in integrated surface (mean ±SE) samples and near-

bed samples in March and August 2018. Note: only surface integrated samples had replication. d/s = downstream, u/s = upstream, s/w = seaward, CE = cumulative effect, FF = far-field, Ref = reference.


63

A4.4 Chlorophyll-a

Chl-a concentrations across all stations and months of 2018 ranged from 0.2 mg/m3 to

2.1 mg/m3 and, consequently, were well below the WQS (i.e. ≤ 3.5 mg/m3)

(Table A4.3). Across the two years of monitoring, average monthly fluorescence (a

proxy for chlorophyll-a / phytoplankton biomass) in the surface c. 5 m at all sampling

stations shows no evidence of increased frequency of algal blooms in the area (see

Appendix 5).

Table A4.3. Surface integrated results for chlorophyll-a (mg/m3) from all sampled months in 2018. CE = cumulative effect, FF = far-field, Ref = reference. No values were above the chl-a WQS threshold (i.e. > 3.5 mg/m3).

Month

NZKS18 NZKS19 NZKS20 NZKS21

(QCS3)

NZKS22

Net pen 500 m

south

500 m

north

CE-Ref FF-Ref

Jan 0.5 0.4 0.3 0.3 0.2

Feb 0.8 1 1 1.1 0.9

Mar 0.5 0.5 0.5 0.8 0.4

Apr 1.1 0.7 0.5 0.4 0.6

May 0.6 0.6 0.7 0.7 0.7

Jun 0.3 0.3 0.3 0.3 0.3

Jul 0.3 0.3 0.3 0.2 0.3

Aug 0.4 0.5 0.5 0.5 0.3

Sep 1.8 1.7 2.1 2.1 1.4

Oct 1.2 1.1 1 1.1 0.8

Nov 0.8 0.9 1.1 1 1.1

Dec 1.6 1 1.1 1.2 1.1

WQS ≤ 3.5 mg/m3

A4.5 Phytoplankton biomass and composition

Estimated phytoplankton biomass values around the NGA farm sites in March, April,

August and September 2018 were in the range of 2 to 102 mg C/m3 (Table A4.4), with

the highest biomass estimates occurring in September, including a particularly high

reading at NZKS21. This high value was slightly lower than peak values recorded in

annual monitoring in Pelorus Sound (McGrath et al. 2019) and within the range of data

from the MDC monitoring stations (Broekhuizen & Plew 2018).


64

The proportion of biomass made up by diatoms was variable, particularly in April,

when they ranged from 12.8 to 74.8% of biomass (Table A4.5). However, overall

biomass was particularly low on this sampling occasion, so the relative dominance of

dinoflagellates was caused by low abundance of diatoms and other phytoplankton,

rather than a bloom of dinoflagellates (Figure A4.5). In other months diatoms

dominated, constituting between 44.2 and 99.6% of biomass. The high biomass

recorded at NZKS21 in September was made up of 75.2% diatoms and 5.7%

dinoflagellates, which are not unusual proportions of these groups.

Figure A4.5. Phytoplankton composition (as a percentage of total phytoplankton biomass) recorded in 2016–2018. The number of sites sampled has changed in the monitoring design since August 2017. In 2018, the full-suite monitoring was postponed from July to September.


65

Table A4.4 Surface estimated phytoplankton biomass (mg/cm3) for two major taxon groupings of diatoms and dinoflagellates in 2018. CE = cumulative effect, FF = far-field, Ref = reference, blank cells = not sampled.

NZKS18 NZKS21 NZKS22

Net pen (QCS3)

CE-Ref

FF-Ref

Mar 18

Diatom 15.1 20.2 Dinoflagellate 0.5 0.1 Other 0.0 20.2

Apr 18

Diatom 1.1 1.2 0.3 Dinoflagellate 6.4 0.2 2.1 Other 0.1 0.2 Aug 18

Diatom 25.8 22.8 2.1 Dinoflagellate 13.3 0.5 1.1 Other

1.6 Sep 18

Diatom 15.6 76.8 11.2 Dinoflagellate 14.8 5.8 5.0 Other 6.6 19.5 7.4

Table A4.5. Phytoplankton composition (as a percentage of total phytoplankton biomass) recorded

in 2018. CE = cumulative effect, FF = far-field, Ref = reference, blank cells = not sampled.

NZKS18 NZKS21 NZKS22

Net pen (QCS3)

CE-Ref

FF-Ref

Mar 2018

Diatom 96.87 99.68

Dinoflagellate 3.10 0.32

Other 0.03 0.00

Apr 2018

Diatom 13.91 74.83 12.84

Dinoflagellate 85.26 15.26 87.16

Other 0.83 9.90 0.00

Aug 2018

Diatom 65.97 91.56 66.22


Other 0.00 6.40 0.00

Sep 2018

Diatom 42.18 75.20 47.48


Other 17.81 19.08 31.34


66

Appendix 5. Time series plots for eutrophication indicators, collected as part of the ongoing monitoring programme. While these data are best displayed on the interactive platform https://cawthron.shinyapps.io/WQ-Marlborough/ (which has greater colour clarity, plot size and functionality) the main purpose of including the plots in this report is to give a general overview of the collective data series through time.

Figure A5.1. Turbidity (FNU: averaged across all depths) for the Tory Channel sampling stations. Measurements were taken from the YSI EXO Sonde instrument

used by MDC. CE = cumulative effect, FF = far-field, Ref = reference, QCS3 = NZKS21.

https://cawthron.shinyapps.io/WQ-Marlborough/


67

Figure A5.2. Dissolved oxygen (DO) saturation (%) in the surface 15 m of the water column at all Tory Channel sampling stations. Measurements were taken from an in situ DO sensor attached to a YSI EXO Sonde or seabird (SBE 19+) CTD. CE = cumulative effect, FF = far-field, Ref = reference, QCS3 = NZKS21.


68

Figure A5.3. Dissolved oxygen (DO) saturation (%) in the bottom 2 m of the water column at all Tory Channel sampling stations. Measurements were taken from an in situ DO sensor attached to a YSI EXO Sonde or seabird (SBE 19+) CTD. Saturation values are averaged from all records within the bottom 2 m of the water column profile at each station. CE = cumulative effect, FF = far-field, Ref = reference, QCS3 = NZKS21.


69

Figure A5.4. Total nitrogen (TN) concentrations (mg/m3) recorded from 15 m surface-integrated samples at all Tory Channel sampling stations. QCS3 = NZKS21. CE = cumulative effect, FF = far-field, Ref = reference, QCS3 = NZKS21.


70

Figure A5.5. Ammonium (NH4) concentrations (mg/m3) recorded from 15 m surface-integrated samples at all Tory Channel sampling stations. QCS3 = NZKS21. CE = cumulative effect, FF = far-field, Ref = reference, QCS3 = NZKS21.


71

Figure A5.6. Nitrate (NO3) + Nitrite (NO2) concentrations (mg/m3) recorded from 15 m surface-integrated samples at all Tory Channel sampling stations. QCS3 = NZKS21. CE = cumulative effect, FF = far-field, Ref = reference, QCS3 = NZKS21.


72

Figure A5.7. Dissolved reactive phosphorus (DRP) concentrations (mg/m3) recorded from 15 m surface-integrated samples from at all Tory Channel sampling stations. CE = cumulative effect, FF = far-field, Ref = reference, QCS3 = NZKS21.


73

Figure A5.8. Dissolved reactive silica (DRSi) concentrations (mg/m3) recorded from 15 m surface-integrated samples at all Tory Channel sampling stations. CE = cumulative effect, FF = far-field, Ref = reference, QCS3 = NZKS21.


74

Figure A5.9. Fluorescence (15 m depth average) at all Tory Channel sampling stations. Fluorescence measurements taken from the YSI EXO Sonde instrument used by MDC, except in March and August where measurements are from the Cawthron Seabird 16+ CTD. CE = cumulative effect, FF = far-field, Ref = reference, QCS3 = NZKS21.


75

Figure A5.10. Chlorophyll-a (mg/m3, surface to 15 m) at all Tory Channel sampling stations. Chl-a measurements taken from the YSI EXO Sonde instrument used by MDC, except in March and August where measurements are from the Cawthron Seabird 19+ CTD. CE = cumulative effect, FF = far-field, Ref = reference, QCS3 = NZKS21.


76

Appendix 6. Datasheet used for underwater lighting observations.

DATASHEET FOR UNDERWATER LIGHTING OBSERVATIONS

Date: 10 July 2016 Farm: Ngamahau

Time: 20:15 Staff member: John Smith

Conditions (e.g. cloud cover, winds): clear skies, calm with low winds (< 5 knots), full moon

Tide stage and sea state: Flooding tide, calm seas

Notes on farm operations: (e.g. lighting engaged in all pens?) Lights operational in two of three pens, last feeding at 18:00.

OBSERVATIONS

Seabirds

Species / type No. Behaviours Location(s) Comments Red-billed gull 12 Resting with some flying

above farm

On and above

farm structures

No evidence of birds feeding

Birds resting on structures

of both lit and unlit pens. Black-billed

gull

3 Resting On farm

structures

Shag 1 Flying past farm above farm

Seals

Species / type No. Behaviours Location(s) Comments Fur seal 4 2 observed resting and 2

swimming outside of farm

Sleeping on top

of predator net

No evidence of chasing prey

Dolphins

Species / type No. Behaviours Location(s) Comments None observed

Fish (other than salmon)

Species / type No. Behaviours Location(s) Comments garfish >12 Swimming near surface in

pens

In illuminated

pens

Baitfish likely yellow-eyed

mullet. Many were greater

than 10 cm. Baitfish 100s Schooling and swimming

inside pen

In illuminated

pens

Other observations: Please note here any observations of salmon feeding on baitfish or other observations not captured in the above. A few salmon were observed chasing baitfish near the surface.

Documents

2018-2019 Annual environmental monitoring summary for the ...€¦ · Lauren Fletcher APPROVED FOR RELEASE BY: Grant Hopkins ISSUE DATE: 4 June 2019 RECOMMENDED CITATION: Bennett