Bivalve shellfish harvesting and consumption in New Zealand, 2011: data for exposure assessment

Abstract

We provide an overview of bivalve molluscan shellfish (BMS) availability to New Zealanders. Data on commercial production (aquaculture) and harvesting, non-commercial (recreational and customary) harvesting and international trading (imports, exports) of BMS were assembled. These data were transformed to estimate the total weight (greenweight and meatweight) of each BMS species, and all BMS species, available to New Zealand domestic consumers in 2011. An estimated 68,000 tonnes greenweight, or 13,000 tonnes meatweight, of BMS were available. Mussels (mostly commercially-produced Perna canaliculus) accounted for 96% of the total available BMS by meatweight. Non-commercially harvested BMS were 1% of the total available BMS. BMS availability was estimated as 8 g/person/day for the total New Zealand population and 407 g/person/day for shellfish consumers. These estimates were comparable to published national consumption data. The results can be used as inputs to exposure assessment. Food availability is a useful estimate where consumption data are limited.

Keywords:

• shellfish
• mollusca
• New Zealand
• harvesting
• aquaculture
• diet
• exposure assessment

Introduction

Bivalve molluscan shellfish (BMS), such as oysters, scallops and mussels, exemplify the conflict of food risks versus food benefits (Nesheim & Yaktine 2007). BMS are filter feeders, which means that chemicals and microbes in their environment can accumulate and pose a risk for consumers. Examples include dinoflagellate toxins and norovirus (Rhodes et al. 2001; de Salas et al. 2005; Wall et al. 2011). On the other hand, shellfish are a popular food, a good source of nutrients, and wild populations can be harvested recreationally (Turner et al. 2005).

Exposure assessment is a fundamental part of quantitative risk assessment for food safety and nutritional dietary assessment (WHO/FAO 2008). Exposure assessment requires information on food consumption (the frequency, types, quantities and sources of food(s) consumed) and on the prevalence and concentration of the component of interest (hazard, nutrient) in foods. The principal source of information on shellfish consumption in New Zealand is that gathered by the Ministry of Health in the National Nutrition and National Children's Nutrition Surveys (Russell et al. 1999; Ministry of Health 2003). However, these surveys do not provide information on source (i.e. it is not possible to distinguish shellfish purchased from retail outlets or gathered from wild populations). Moreover, the limited sample size of these surveys (approximately 0.2% of the total population) means that infrequently consumed shellfish species may be missed.

An alternative to consumption data is production and harvesting information. The aim of this study was to provide a picture of BMS consumption by New Zealanders from data on the availability of BMS to the New Zealand consumer. The availability of BMS includes commercial production (commercial harvest of wild stocks and aquaculture), non-commercial harvesting of wild stocks (recreational or for customary purposes) and imports (Fig. 1). We investigated whether the estimated availability of BMS is a good alternative to consumption data by comparing our results with data from the two national nutrition surveys and published studies of shellfish consumption in New Zealand.

Figure 1  Sources of bivalve molluscan shellfish (BMS) available to the New Zealand consumer.

Methods

Data on the commercial harvesting of all wild shellfish stocks managed under the quota management system (tonnes or numbers of shellfish as reported on Catch Effort Returns, 2010/11 fishing year) were provided by the Ministry for Primary Industries. Landings reported against Destination Codes L and W were used in this analysis (Cartwright 2001). Aquaculture production statistics (tonnes, year ending 2011) were provided by Aquaculture New Zealand and were supplemented with data from a private company not represented by that association. Import, export and re-export statistics were obtained from Statistics New Zealand Infoshare (appropriate selections from harmonised system codes 3071, 3072, 3073, 3079 and 1605; tonnes, year ending December 2011) (Statistics New Zealand 2012). Recreational harvest estimates were taken from the 2000/01 National Marine Recreational Fishing Survey (Boyd et al. 2004). This was the most recent data available at the time of this research. Customary harvest data were provided by the Ministry for Primary Industries (reported harvesting, 2011). Individuals in each of these organisations were contacted to resolve queries or to request further information.

The data were in a variety of formats and units. Where data were in numbers of shellfish they were converted to tonnes of shellfish using the conversion factors presented in Table 1. These number/weight conversion factors were either legislative or generated from species-specific length/weight regressions. Where possible, more than one length/weight regression was used to represent stocks in different New Zealand locations, and the appropriate shellfish length was selected based on information in the literature describing the likely size of harvested shellfish. An exception to this approach is for the recreational harvest of scallops, for which area-specific mean weights were available (Boyd et al. 2004; Boyd & Reilly 2004). The customary harvesting data were not always reported as numbers or weight of shellfish. The following methods were applied to convert data to weight: one ‘bin’ = 25 kg (Fisher & Sanders 2009); one ‘bag’ = 10 kg; one ‘20 L bucket’ = 10 kg; one ‘10 L’ (assumed to be a 10 L bucket) = 5 kg; and where no units were provided the value was assumed to be the number of shellfish and converted to weights using the conversion factors in Table 1.

Table 1  Conversion factors used to convert numbers of shellfish into greenweight, and to convert greenweight into meatweight.

Bivalve molluscan shellfish	Weight per individual shellfish (g)	References	Proportion meatweight	References
Cockle (Austrovenus stutchburyi)	8	Hartill & Cryer 1999; Stewart 2008; Williams et al. 2009; Ministry of Fisheries 2011a	0.13	Fletcher et al. 2008
Deepwater clam (Panopea zelandica)	~		0.125	^1
Dosinia, ringed (Dosinia anus)	~		0.125	^1
Mussel, green-lipped (Perna canaliculus)	17	^2	0.3287	^2
Mussel, horse (Atrina zelandica)	209	^3	0.125	Fisher & Sanders 2009
Mussel, species unspecified	10	^4	0.125	Fisher & Sanders 2009
Oyster, dredge (Ostrea chilensis)	102	MAF 2012	0.125	^1
Oyster, Pacific (Crassostrea gigas)	67	MAF 2012	0.1352	^5
Oyster, rock (Saccostrea glomerata)	32	^2	0.1895	^2
Oyster, species unspecified	40	^6	0.125	^1
Pipi (Paphies australis)	12	Hartill & Cryer 1999; Williams et al. 2007; Ministry of Fisheries 2011a	0.2	Fisher & Sanders 2009
Scallop (Pecten novaezelandiae)	104	MAF 2011; Ministry of Fisheries 2011a	0.125	Fisher & Sanders 2009
Scallop, queen (Zygochlamys delicatula)	~		0.125	Fisher & Sanders 2009
Toheroa (Paphies ventricosa)	122	Beentjes 2010 ^7	0.125	^1
Triangle shell (Spisula aequilatera)	~		0.125	^1
Trough shell (Mactra discors)	~		0.125	^1
Trough shell, large (Mactra murchisoni)	~		0.125	^1
Tuatua (Paphies subtriangulata)	26	Cranfield et al. 2002 ^8	0.2	Fisher & Sanders 2009
Tuatua, southern (Paphies donacina)^9	~		0.125	^1
Tuatua, species unspecified	~		0.125	^1
Unspecified clams	~		0.125	^1
Notes:~This conversion factor was not required in this study; ^1a conversion factor of 0.125 (Fisher & Sanders 2009) was applied where no conversion factors were located in the literature or through personal enquiries; ^2Auckland Council, unpublished data (300 samples, 2006–2011); ^3K. Grange and C. Hay (DSIR), unpublished data (201 samples, 1984); ^4a conservative estimate used to convert non-commercially harvested mussels; ^5Auckland Council, unpublished data (996 samples, 2006–2011); ^6a conservative estimate used to convert non-commercially harvested oysters; ^7J. Williams (NIWA), unpublished data (1153 samples, 2010); ^8J. Williams (NIWA), unpublished data (1769 samples, 2009); ^9prior to September 2011 this species had the common name of deepwater tuatua (Ministry for Primary Industries, pers. comm., June 2012).

To estimate the weight of shellfish actually consumed, greenweights (unshucked shellfish) were converted to meatweights (meat only) using the conversion factors presented in Table 1. An exception to this is commercially harvested scallops, which are reported as meatweight (Cartwright 2001). When no species-specific conversion factor was identified, a default of 0.125 was applied, as commonly used in a database for New Zealand fisheries management (Fisher & Sanders 2009). The Statistics New Zealand data included a variety of whole and processed shellfish products, and the conversions from greenweight to meatweight were done as follows: no conversions were applied where the descriptor indicated that the product was without a shell (e.g. meat, dried); where the descriptor indicated that the product contained both shell halves (e.g. ‘chilled, whole’) the conversion factors were applied as in Table 1; where it was not clear whether the shell remained a part of the product (e.g. ‘whether in shell or not’) the product was assumed to be in the shell (except for frozen scallops which are usually traded as meat); where the descriptor indicated that the product contained only one half of the shell (e.g. ‘chilled, half-shell’) the following conversion factors were used to approximate meatweight: green-lipped mussels, 0.4948; mussels (species unspecified), 0.22; dredge oysters, 0.22; rock oysters, 0.3186; Pacific oysters, 0.2382 (all derived from the following equation: a/(a + 0.5(1−a)), where a=the proportion meatweight from Table 1, which assumes that both halves of the shell are equivalent in weight and does not take into account the variable weight of water in an unshucked shellfish); where oysters were described as ‘rock or Pacific’ they were assumed to be Pacific oysters; data for queen scallops and Pecten novaezelandiae are reported combined, so it was assumed that all data were for P. novaezelandiae (the majority of the commercial harvest, see Table 2); data for Mytilus spp. and Perna spp. are reported combined so it was assumed that all export data were for P. canaliculus (the dominant commercial species in New Zealand); all exports of tuatua were assumed to be southern tuatua (the majority of the commercial harvest, see Table 2).

Table 2  Estimated amount of bivalve molluscan shellfish (BMS) available to New Zealanders in 2011 by greenweight (all values are in tonnes).¹

	Commercial harvesting		Non-commercial harvesting		International trade (2011)
Species	Wild stocks (2010/11)	Aquaculture (2011)	Recreational harvesting (2000/01)	Customary harvesting (2011)	Imports	Re-exports	Exports	Amount of BMS available to the New Zealand consumer^2
Cockle	1,253.0	~	33.4	0.7	0.0	0.0	213.5	1,073.6
Deepwater clam	15.5	~	~	~	~	~	~	15.5
Dosinia, ringed	3.1	~	~	~	~	~	~	3.1
Mussel, all species	78.9	101,311.0	57.8	10.3	74.9	28.9	38,257.7	63,246.4
Mussel, green-lipped	78.2	101,311.0	29.0	^5 0.0	~	~	~
Mussel, horse	0.7	~	~	^5 0.0	~	~	~
Mussel, any species	~	~	28.8	10.3	74.9	28.9	38,257.7
Oyster, all species	1,119.7	1,824.0	61.1	34.9	75.3	0.2	1,706.4	1,408.4
Oyster, dredge ^3	1,119.7	20.0	13.2	34.9	4.6	0.0	146.3
Oyster, Pacific	~	1,804.0	8.1	~	67.3	0.0	1,133.2
Oyster, rock	~	~	5.2	~	0.0	0.0	43.8
Oyster, any species	~	~	34.6	~	3.4	0.2	383.0
Pipi	85.2	~	96.1	3.1	~	~	~	184.4
Scallop, all species	963.3	~	560.9	2.9	435.5	1.4	53.0	1,908.2
Scallop	962.1	~	560.9	2.9	~	~	~
Scallop, queen	1.1	~	~	~	~	~	~
Scallop, any species	~	~	~	~	435.5	1.4	53.0
Toheroa	~	~	1.2	0.8	~	~	~	2.0
Triangle shell	21.0	~	~	~	~	~	~	21.0
Trough shell	^50.0	~	~	~	~	~	~	^50.0
Trough shell, large	30.5	~	~	~	~	~	~	30.5
Tuatua, all species	42.7	~	150.6	0.3	0.0	0.0	^50.0	193.6
Tuatua	0.0	~	150.6	0.3	~	~	~
Tuatua, southern ^4	42.7	~	~	~	~	~	~
Tuatua, any species	~	~	~	~	0.0	0.0	^50.0
Unspecified clams	~	~	~	~	7.5	0.0	149.8	−142.3
Total	3,612.8	103,135.0	961.1	53.1	593.2	30.4	40,380.4	67,944.3
Notes: ~ Data are not applicable to this category; ^1values for commercial and non-commerical harvesting are in tonnes greenweight. Values for international trade are in tonnes of processed product; ^2sum of commercial harvesting, recreational harvesting and imports, minus exports and re-exports^; ^3including Foveaux Strait dredge oysters^; ^4prior to September 2011 this species had the common name of deepwater tuatua (Ministry for Primary Industries, pers. comm., June 2012); ^5green-lipped mussel, 0.03 tonne; horse mussel, 0.01 tonne; trough shell, 0.007 tonne; tuatua, 0.03 tonne.

The availability of shellfish (per person, per day) was calculated using the following estimates for the New Zealand resident population (all values mean for the year ended December 2011) from Statistics New Zealand (Statistics New Zealand 2012): total population, 4,407,400; children (5–14 years), 580,020; adults (15 + years), 3,513,100. The shellfish consuming population (87,215 people) was derived by adding 2.4% of the adult population to 0.5% of the child population; these percentages were sourced from an analysis of New Zealand national nutrition survey data and are for the proportions of respondents (adults, children) who reported consuming shellfish in the previous 24 hours (Russell et al. 1999; Ministry of Health 2003; Cressey et al. 2006).

Additional studies on shellfish consumption in New Zealand were identified through a search of scientific literature and the websites of Regional Councils and New Zealand research institutes.

Results

Table 2 summarises the commercial and non-commercial harvesting data and the estimated amounts of each species of BMS available to the New Zealand domestic consumer (greenweight). In addition to the species in Table 2, silky dosinia (Dosinia subrosea) and frilled venus shell (Bassina yatei) are managed under the quota management system, but were not reported as being harvested in 2011.

An estimated 67,944 tonnes greenweight of BMS were available to New Zealand domestic consumers in 2011. The BMS species available in the greatest amounts to New Zealand consumers, by weight, were mussels (93%; mostly green-lipped), scallops (3%; mostly P. novaezelandiae), oysters (2%; dredge and Pacific) and cockles (2%).

Aquaculture dominated commercial production of BMS in New Zealand (97% of the commercial harvest, by weight), and was dependent on two species, green-lipped mussels and Pacific oysters. Commercial harvesting of wild stocks was modest by comparison, but included a wider range of species. Oyster production was dominated by dredge oysters harvested from wild stocks (1120 tonnes) and Pacific oysters produced through aquaculture (1804 tonnes). Analysis of the export data showed that only 9% of the total weight of exported oyster product was from dredge oysters. This indicates that most of the commercially harvested dredge oysters were distributed through the domestic market.

Non-commercial harvesting appears to be an important source of tuatua, pipi and scallops relative to other sources. Table 2 shows the estimated recreational harvest of tuatua in 2000/01 (151 tonnes) to be over three times the commercial landings in 2011 (43 tonnes). Commercial harvesters target southern tuatua, which live in subtidal zones where they are not easily gathered by recreational harvesters. The reported commercial landing of tuatua in 2000/01 was also low (16 tonnes) (Ministry of Fisheries 2011a) so it is consistent that non-commercially harvested tuatua are consumed in higher amounts than commercially harvested tuatua. In previous years the commercial harvest of pipi has been higher (e.g. 184 tonnes in 2000/01 [Ministry of Fisheries 2011a]).

Table 3 shows that an estimated 12,947 tonnes meatweight of BMS were available to New Zealand domestic consumers in 2011. The data suggest that commercially harvested BMS were the main source in New Zealand, and that mussels dominated this market (96% of the total available BMS by meatweight). The next most important species were scallops (2%) and cockles (1%). Non-commercially harvested BMS made up 1% of the total meatweight.

Table 3  Estimated amount of bivalve molluscan shellfish (BMS) available to New Zealanders in 2011 by meatweight (all values are in tonnes).

Species	Commercial harvesting	Non-commercial harvesting	International trade^1	Amount of BMS available to the New Zealand consumer^2
Cockle	163	4	−28	140
Deepwater clam	2	~	~	2
Dosinia, ringed	^3<1	~	~	^3<1
Mussel, all species	33,327	14	^4−20,914	12,427
Oyster, all species	386	12	−321	77
Pipi	17	20	~	37
Scallop, all species	120	70	49	240
Toheroa	~	^3< 1	~	^3<1
Triangle shell	3	~	~	3
Trough shell	^3<1	~	~	^3<1
Trough shell, large	4	~	~	4
Tuatua, all species	5	30	^3<1	36
Unspecified clams	~	~	−18	−18
Total	34,028	152	−21,232	12,947
Notes: ~ Data are not applicable to this category; ^1imports minus the sum of exports and re-exports. A negative value indicates that more product was exported than was imported. Values are the sum of tonnes meatweight for meat, half or whole shell products, and tonnes of processed products (e.g. dried, in brine); ^2sum of commercial harvesting, recreational harvesting and international trade; ^3ringed dosinia, 0.4 tonne; toheroa, 0.2 tonne; trough shell, 0.001 tonne; tuatua,−0.004 tonne; ^4assuming that all exports are P. canaliculus (see Methods).

Based on the meatweight data and mean population estimates for 2011, an estimated 8 g of BMS were available per person, per day for the total New Zealand population. The BMS availability to estimated numbers of shellfish consumers was 407 g/person/day.

Discussion

This research has amalgamated five datasets to produce estimates of the availability of BMS to New Zealand domestic consumers. These results can be used as inputs for exposure assessment. This approach fills a current data gap by providing recent, national-level information on potential shellfish consumption by New Zealanders. An advantage of this approach over traditional food consumption surveys is that it considers all shellfish species that are available for consumption, from commercial and non-commercial sources, and estimates their relative availability. It is necessary for food consumption surveys to sample a population, which carries the risk of not detecting less common consumption activities (e.g. consumption of less popular shellfish species or recreational/customary harvesting).

The major challenges were standardising the datasets so that they could be aggregated, and identifying appropriate conversion factors to calculate tonnes from numbers of shellfish or convert greenweights to meatweights. The major sources of uncertainty are the recreational and customary harvest data because of their age and under-reporting, and the data on international trade because of the way these data are reported.

The data on imports and exports includes whole BMS and a variety of processed products from half-shells to dried or salted products. Export statistics show that the majority (83%, by weight) of the products are traded as half-shell and 6% are traded as whole-shell according to the descriptors. Therefore the export weights exclude waste shells, water and other trimmings (guts, skirts), so deducting the exported weight from total production (to provide greenweight domestic availability) overestimates domestic availability. This problem has been partially addressed by adjusting the estimates to meatweights (Table 3). However, a default value of 0.125 had to be used to convert the greenweight of 11 species to meatweight in the absence of specific data.

The recreational harvesting numbers are national estimates based on data provided by 7374 diarists from December 2000 to November 2001 (Boyd et al. 2004). The authors of the report estimated the harvest of each shellfish species by fishery management area. In some areas, some species were only harvested by one diarist, and in these situations the coefficient of variation for the national harvest estimate was very high (>100%). The recreational harvesting estimates were also reported as numbers of shellfish. The conversion factors applied (Table 1) are largely derived from commercially harvested shellfish, or use a preferred harvest size in the length/weight regressions. This results in an overestimated tonnage since recreational harvesters are still likely to collect smaller shellfish if the preferred size is not available (Hartill et al. 2005). In addition, these harvest data are now over a decade old and might not be representative of the current level of recreational harvesting.

The customary harvesting results will be an underestimate. In New Zealand, harvesting of shellfish for traditional or customary purposes is controlled through the Fisheries (Kaimoana Customary Fishing) Regulations 1998, the Fisheries (South Island Customary Fishing) Regulations 1999 or the Fisheries (Amateur Fishing) Regulations 1986 (Reeves 1986; Hardie Boys 1998, 1999). The two customary fishing regulations enable the Minister responsible for the regulations to confirm the appointment of custodians (Tangata Tiaki/Kaitiaki) for a customary food gathering area. These custodians manage customary food gathering within their area by having the power to authorise (or not) individuals to gather more seafood (kaimoana) than is permitted by rules set under the amateur fishing regulations (Reeves 1986; Hardie Boys 1998, 1999). The custodians must keep records of harvesting authorisations and the species and quantities harvested, and must report this information to the Ministry for Primary Industries. If there are no custodians appointed for a customary food gathering area, customary seafood gathering is controlled through regulations 27 and 27A of the amateur fishing regulations, which provide for authorising agents appointed by Māori entities to control such harvesting. There is no requirement for these authorising agents to report on the permits they issue, and because customary harvesting in much of the North Island (New Zealand) coastline operates under the amateur fishing regulations, these harvesting data are not reported (Ministry for Primary Industries, pers. comm., May 2012).

We calculated two estimates of shellfish availability from our data: 8 g/person/day for the total New Zealand population; and 407 g/person/day for shellfish consumers. These estimates are for raw meatweight and do not take account of factors such as weight loss from cooking or wastage during preparation. Our results are comparable to other estimates. The Food and Agriculture Organization of the United Nations (FAO) produces food balance sheets that estimate the amount of a food commodity available for consumption within a country, primarily based on production and import/export statistics (FAO 2012). The latest data available (2007) for the category ‘molluscs, other’ estimates 3 g/person/day for the total New Zealand population. The 1997 New Zealand National Nutrition Survey (NNS) and 2002 National Children's Nutrition Survey (CNS) asked respondents about consumption of shellfish (in these surveys the term ‘shellfish’ included the gastropod mollusc species pāua [Haliotis iris]; Russell et al. 1999; Ministry of Health 2003). From these surveys, the mean shellfish serving sizes for shellfish consumers were 106 g for adults and 49 g for children (Cressey et al. 2006). A more recent NNS of adults was conducted in 2008/09, but at the time of this research, individual record data to calculate shellfish consumption were not available (Ministry of Health 2011).

There are few other published studies on the consumption of shellfish in New Zealand. A 1981 survey of 75 oyster consumers in Bluff found that the frequency of consumption over the harvest period (March to August) ranged up to 192/week (McKenzie-Parnell et al. 1988). A 2011 survey of 22 ‘seafood consumers’ living around Northland's Rangaunu Harbour found that most respondents reported that they had consumed shellfish and finfish from the harbour all their life, on average 2.6 times/week, but data on shellfish were not reported separately (McNabb et al. 2012). In 2008, 30 participants from two Māori populations were interviewed about their consumption of wild food (kai), including food harvested from the sea (kaimoana) (Tipa et al. 2010a, b). Of the participants, 47% consumed mussels once a month or more often, and 7% consumed mussels 3–4 times per week; however, it was not clear whether this included shellfish purchased from retail outlets.

Understanding the level of non-commercial harvesting of BMS is important for food safety risk assessment, as wild shellfish stocks that support non-commercial harvesting are not subject to the same level of monitoring and risk management as commercially-harvested stocks (wild or cultivated). The NNS and CNS do not provide information on the source of shellfish consumed. In this study, an estimated 1% of the total available BMS was from non-commercial harvesting. This is an underestimate because customary harvesting is under-reported. In New Zealand, the commercial growing, harvesting, processing and sale of shellfish are highly regulated, as are food safety considerations. However, the regulations controlling recreational and customary harvesting do not include controls over food safety, although marine biotoxins are monitored in some recreational gathering areas (Reeves 1986; Hardie Boys 1998, 1999; Ministry for Primary Industries 2012).

When conducting an exposure assessment, these estimates of shellfish availability need to be accompanied by data on the frequency of shellfish consumption, distribution of serving sizes and food preparation. The NNS and CNS are currently the only sources of this information at a national level. Data on food preparation is important for assessing food safety risks and nutritional intake because preparation methods such as cooking, smoking, drying or fermentation will change the microbiological and nutritional qualities of the food. The NNS found that 7/59 (12%) of reported mussel servings and 13/22 (59%) of reported oyster servings were consumed raw (Greening et al. 2009), and raw consumption of mussels and oysters was also found to be popular in the wild kai study (Tipa et al. 2010b).

In summary, this study has shown how estimates of food availability are useful where consumption data are limited. This study's estimates of shellfish availability have captured data on a wide range of shellfish species plus information on their source, which has not been gathered through national-level surveys in New Zealand. These estimates can be used as inputs to exposure assessment, but would be improved by:

•	more accurate greenweight/meatweight conversion factors based on recent and appropriate data that also cover half-shell products;
•	better reporting of customary harvesting; and;
•	more recent estimates of recreational harvesting.

The Ministry for Primary Industries is undertaking a marine recreational fishing survey for the 2011/12 year and the results will be available in June 2013 (Ministry of Fisheries 2011b). National data on customary harvesting will improve if management of such harvesting is shifted from the amateur fishing regulations to the customary harvesting regulations. The estimates produced in this study can be easily updated as new information becomes available.

Acknowledgements

We thank the following people for provision of additional information and advice: J. Williams (National Institute of Water and Atmospheric Research [NIWA]), G. Fletcher (Plant and Food Research), the Ministry for Primary Industries (particularly M. Jones and K. Dunn), T. Haggitt (Coastal Research & Consulting Ltd), the Auckland Council, C. Marriott (Aquaculture New Zealand), M. Camden (Statistics New Zealand) and N. Phillips (previously of NIWA). This research was funded by the ESR Capability Development Fund.