TAC allocation in mixed-fisheries: a bio-economic modelling. investigation applied to the New Zealand hoki fishery

Transcription

1 ICES CM 2008/I:10 NOT TO BE CITED WITHOUT PRIOR REFERENCE TO THE AUTHOR TAC allocation in mixed-fisheries: a bio-economic modelling investigation applied to the New Zealand hoki fishery Paul Marchal 1,2, Chris Francis 3, Philippe Lallemand 2, Sigrid Lehuta 4, Stéphanie Mahévas 4, Kevin Stokes 2, Youen Vermard 4 1. IFREMER, Channel and North Sea Fisheries Department, 150 Quai Gambetta, BP 699, Boulogne s/mer, France [tel: , fax: , paul.marchal@ifremer.fr] 2. Seafood Industry Council, 74 Cambridge Terrace, Wellington, New Zealand 3. NIWA, 301 Evans Bay Parade, Hataitai, Wellington 6021, New Zealand 4. IFREMER, Fisheries and Ecological Modeling Department, Rue de l Ile d Yeu, BP 21105, Nantes Cedex 03, France 1

2 ABSTRACT We evaluate, using a bio-economic modelling approach building on the ISIS-Fish software, the impact of relative TACs and deemed values on the sustainability of a selection of species exploited by the New Zealand hoki fishery. We investigate some aspects of the hoki mixed fisheries, consisting of four fleets and nineteen métiers, by considering the technical interactions between hoki and hake. The dynamics of effort allocation were modeled using the gravity model, using value per unit effort (VPUE) as attractivity coefficient. Eleven management scenarios, based on different levels of cach limits, effort limits and landing taxes have been defined, and their impact on spawning biomass, catches and VPUE have been preliminarily investigated. KEYWORDS Fleet dynamics, New Zealand hoki fishery, deemed value, fisheries management 2

3 INTRODUCTION Managing fisheries through individual property-right finds its roots in the early Two main instruments have been envisaged: the TURFs (Territorial User Rights in Fisheries) and the ITQ (Individual Transferable Quotas). While the implementation of TURFs is restricted to the management of sedentary species (e.g. shellfish), the ITQs have a broader applicability and are generally expected to benefit economic efficiency (Hentrich and Salomon 2006, Arnason 2007). ITQs have now been broadly implemented worldwide, including 10 major fishing nations, to regulate around 15% of the global marine fish catch (12 million tonnes) (Arnason 2007). In theory, ITQs create incentives for fishers to maximize the value of their catch and minimize harvest costs. They also create, in principle, economic incentives to avoid catch of species the fisher does not have quota for. Finally, in mixed fisheries, ITQs are expected to alleviate the discrepancy between the combined-stocks quota portfolios with actual species composition. Despite the flexibility brought about by ITQs, it is almost inevitable that the discrepancy will persist (Annala et al. 1991; Branch et al. 2006). A plaster in then needed to correct that mismatch, and that may take different forms depending on the country where the ITQs are implemented (Sanchirico et al. 2006). In New Zealand, the Total Allowable Commercial Catches are distributed to quota holders as ITQ shares. On the first day of the fishing year 1, each ITQ (expressed as a percentage of the TACC) generates for each quota holder, and each stock, a catching right (in kgs) referred to as the Annual Catch Entitlement (ACE), so that 1 For most of New Zealand stocks, including those harvested by the case studies fisheries investigated here, the fishing year n/n+1 starts the 1 st October in year n and finishes the 30 th September in year n+1. 3

4 ACE(kg) = TACC(kg) x ITQ(%) ACE, like ITQ, is freely tradable on the open market, and accessible to any New Zealand citizen. Despite that flexibility, and even where fishers are allowed to acquire catch rights after landing fish, aggregate commercial catches may not always match up with TACCs. Discarding is prohibited in New Zealand for almost all species managed under the QMS and can hence not be considered as an option to balance catches as, e.g., in The Netherlands. Fishers and/or quota-holders have two options. If the mismatch between catch and quota is limited, quota-holders are allowed to carry forward up to 10% of their quota. If that mismatch is greater, fishers are allowed to land species in excess of their ACE, even when the overall TACC for these species has already been exceeded. In that case, fishers are charged at the end of the fishing year a landing tax, or deemed value, for each unit of catch they land above their ACE holdings at the time. The deemed value is set annually by the Minister of Fisheries, advised by the Ministry of Fisheries, at the same time as the TAC and the TACC. There is no clear policy or rationale as to how the deemed value is calculated. However, the level at which the deemed value is set may have dramatic consequences for the fisheries sustainability. While a high deemed value (i.e. well above the ACE price) may encourage fishers to shift target species once their ACE is exceeded, a deemed value set at a low level (i.e. close to, and a fortiori below, the ACE price), may incentivise fishers to pay the charge requested and continue targeting the same stock, even when they have no ACE. For instance, the deemed value for hoki has been lower than the ACE price in 2007, and the TAC set for that stock was exceeded at the end of the fishing year. 4

5 We evaluate, using a bio-economic modelling approach building on the ISIS-Fish software, the impact of relative TACs and deemed values on the sustainability of a selection of species exploited by the New Zealand hoki fishery. MATERIAL AND METHODS Bio-economic model The bio-economic model retained in this investigation is the version of ISIS-Fish ( ISIS-Fish is a spatial and seasonal simulation model describing the dynamics of resources, exploitation and management has been developed to explore the impact of a range of management measures upon fisheries dynamics. It allows the comparison the respective impacts of conventional management measures like catch and effort controls, and any spatialised measure. ISIS-Fish aims at being as generic as possible in order to be applied to different types of fishery. Existing knowledge about each fishery may be stored in a database included in the software, and may be easily modified. This includes the parameters describing each population and each fishing activity. ISIS-Fish allows for flexibility in several model assumptions (for instance stock-recruitment relationships, selectivity models, fleet dynamics,...) which make it possible to use it for most fisheries. Management measures and fisher s response to management measures and to economic conditions may be interactively implemented through a Script language. We summarise here the main features of ISIS-Fish. The full details and equations implemented in ISIS-Fish may be found in Mahévas and Pelletier (2004), Pelletier and Mahévas (2005) and Drouineau et al. (2006). ISIS-Fish is designed to assess the performance of local and temporal management measures involving spatial and seasonal control variables for regulating exploitation, e.g. fishing effort and catches. It is based on three sub-models, 5

6 namely a population dynamics model, a model for fishing activity and a model for management measures. Each sub-model is spatially and seasonally explicit (Figure 1). The biological sub-model is determined by a growth curve, condition factors, catchability coefficients, migration ogives and a recruitment function. The fishing activity sub-model builds on fleet parameters (fishing effort, economics), métier parameters (seasonal and spatial distribution, selectivity ogives, a standardization factor, target factors by species, seasons) and strategy parameters (set of métiers, proportion of fishing units per fleet, monthly distribution of fishing units among the métiers operated). Note that the standardization factor is used to get comparable effort units among gears, while the target factor associated to each species quantifies the strength with which the species is sought for by the métier. Finally, the management sub-model includes a series of management measures (TAC, mesh size limits, area access restrictions) and a fishers behavior gravity model. The fishery takes place in a region defined by its contour and a regular grid. The spatial resolution of the grid in latitude and longitude is chosen with respect to the dynamics being described, and depending on the precision of available information. Within the region, zones (i.e. sets of contiguous grid cells) are defined independently for each population, each fishing activity and each management measure. The model has a monthly time step. Seasons (i.e. sets of successive months) are also defined independently for each population, each fishing activity and each management measure. Within each zone and season, relevant variables such as fishing effort for a specific activity or abundance of a given population, are assumed to be homogeneous and uniformly distributed. Data 6

7 In New Zealand, fisheries have been regulated by the Quota Management System (QMS) since A TAC is set by the Minister of Fisheries for each stock included in the QMS at the start of the fishing year. For most of these stocks, the fishing year starts the 1 st October and finishes the 30 th September in the following year. After an allocation has been granted to the recreational and the customary sectors, and also after provision has been made to account for other sources of mortality (mainly illegal fishing), the rest of the TAC (referred to as Total Allowable Commercial Catch, or TACC) is allotted to the commercial fishing sector. The TACCs are then distributed to quota holders as ITQ shares, and an ACE (catching right) is generated for each of them. The Minister of Fisheries also sets the deemed value of all QMS stocks, which fishers have to pay for each kg of fish landed above their ACE. The New Zealand exclusive economic zone is partitioned in ten Fisheries Management Areas (FMA) and in smaller-sized statistical areas (Figure 2). FMA 10 has mainly been established to prevent mis-reporting, as this area is normally subject to very little fishing activity. 629 fish stocks are managed under the QMS. The management areas specific to each stock are referred to as the Quota Management Areas (QMAs), and they usually consist of 1 or a combination of several FMAs. However, the QMAs of some aggregative species (e.g., orange roughy, oreos) may consist of sub-divisions of FMA. It should also be noted that the QMA, which are defined for management purposes, are not necessarily consistent with the stock areas defined for assessment purposes (ASA, Assessment Stock Areas). The hoki (Macruronus novaezelandiae) fishery (also called middle-depth fishery) is the largest fishery in New Zealand, mainly operated by bottom and mid-water trawlers. There are four main fisheries, two on hoki spawning grounds (west coast South Island and Cook Strait) and two on hoki feeding grounds (Chatham Rise and Sub-Antarctic). Hoki is assessed as two 7

8 separate stocks, an eastern component covering FMAs 1 and 2, and a western component covering FMAs 3 to 9 (Figure 1). Although hoki is managed as one single stock (HOK1), which covers FMAs 1 to 9, an informal agreement between the New Zealand Ministry of Fisheries and the fishing industry has established a TACC allocation key between the eastern and the western stocks. In response to a series of poor recruitments, the overall hoki TACC dropped from 250,000 t ( ) to 180,000 t ( ), and again to 100,000 t in (Annala et al. 2004). Depending on where and when fishing is taking place, a number of bycatch species are caught in variable proportions the hoki fishery, and these include in particular barracouta (Thyrsites atun), hake (Merluccius australis), jack mackerels (Trachurus sp.), ling (Genypterus blacodes), orange roughy (Hoplosthetus atlanticus), oreos (Pseudocyttus maculatus, Neocyttus rhomboidalis and Allocyttus sp.), red cod (Pseudophycis bachus), southern blue whiting (Micromesistius australis), squids (Nototodarus sp.) and warehous (Seriolella sp.). In this study, we investigate some aspects of the hoki mixed fisheries by considering the technical interactions between hoki and hake. Hake is not necessarily the most important bycatch of the hoki fishery. However, because of its relatively high deemed value, it has practically constrained the hoki fishery in recent years whenever its related TACC was exceeded. Hoki (HOK1) has since 1989 been assessed as two stocks, eastern hoki (covering FMAs 1-2) and western hoki (covering FMAs 3-9) (Francis 2007). Hake is managed and assessed as three stocks. The largest hake stock is that off the west coast of the South Island (FMA 7) and it is referred to as HAK7. The two other stocks, which are located in FMAs 1-3, 5-6, 8-9 (Sub- Antarctic hake or HAK1) and in FMA 4 (Chatham Rise hake or HAK4), are mainly by-catch 8

9 of the hoki trawlers (Anonymous 2007). For the five stocks, we have used the data used and outputs from the most recent validated assessments, which were performed using the CASAL package (Bull et al. 2005). Three runs were presented in the 2007 hoki assessment, and we selected the one which best fitted the requirements of the ISIS-Fish model used in this investigation. The selected run assumed, (1) a domed spawning selectivity, (2) an age-based selectivity, (3) no differentiation between males and females and, (4) natal fidelity (i.e. the fish to which a stock belongs is determined at birth) and the full independence between the two hoki stocks. The catch and effort data used to parameterise the exploitation sub-model were derived from mandatory New Zealand log-books. Data were available by fishing trip and by statistical area. The TACCs set for the different hoki and hake stocks in recent years are shown in Table 1. These have been used as a basis to parameterise the management model. Finally, we used in our model the landing prices, ACE prices and deemed values related to the hoki (HOK1) and hake (HAK1, HAK4, HAK7) stocks, and these are shown in Table 2. We used the 2003/2004 average figures, as hake landing price data in subsequent years were deemed unreliable. The deemed value of these stocks vary depending on the amount of fish landed above ACE. The deemed value we used here as reference to parameterise the management model is the charge paid when the amount of fish landed above the ACE is in different ranges: [0-20%, 20-40%, 40-60%, 60-80%, %, >100%]. Model parameterisation Population dynamics 9

10 The growth parameters, condition factors, natural mortality, recruitment, migration and spawning ogives of all hoki and hake stocks were drawn or derived from Dunn et al. (2006), Francis (2007) and Anonymous (2007). For hoki, the growth parameters, condition factors and natural mortality were provided for both sexes combined. For the hake stocks however, these parameters were provided for each sex separately, and the values we used were derived as the average over both sexes (Table 3). Recruitment was calculated the value leading to the virgin biomass (B 0 ), as estimated in the assessment reports (Table 3). Consistent with Francis (2007), we assumed that the hoki SSB for both stocks was simply the total biomass in the spawning area. Dunn et al. (2006) indicated that for all hake stocks, 50% of hake individuals were mature at 6-8 years. We assumed here knife-edge spawning, so that all hake aged 1-6 were immature and that all fish aged 7 and older were mature. For the hoki stocks, we kept the same number of age groups than that used in the 2007 assessment (i.e. 17 age groups with the latter being a plus-group). For all hake stocks, we used 7 age groups instead of the 30 given in the assessment. This simplification results from, (1) only spawning abundance (assumed to consist of fish aged 7 and above) was available and, (2) the complexity of the biological model had to be marginally reduced to allow for the development of processes inherent to both the fishing activity and the management submodels. Subsequently, for all hake stocks, age group 7 was made a plus-group, the biomass of which exactly corresponds to the SSB. Migrations have been evidenced for both hoki stocks and these are accounted for in the stock assessment (Francis 2007). Fish from the Eastern stock spawn in Cook Strait and have their home grounds in Chatham Rise. The Western stock spawn in west coast South Island and fish have their home grounds in sub-antarctic. Soon after being spawned, all juveniles move to Chatham Rise. Under the natal fidelity assumption, which we considered here, the stock to which a fish belongs is determined at birth. At some time before age 8, all Western hoki 10

11 juveniles migrate to their home grounds in sub-antarctic. The migration ogives used for both hoki stocks are given in Table 4. It is uncertain whether migrations are occurring within or across hake stock areas (Anonymous 2007), and these have been neglected in this study. Seasonal catchability could not be inferred directly and estimates were derived from an linear analysis of catch per unit effort (CPUE). The details of this analysis are given in Appendix A. Fishing mortality (F) series were required to derive catchability (Appendix A), and these were derived based on available information. For the hoki stocks, both abundance-at-age and catchat-age by season were available, so F was calculated by combining the VPA and the Baranov catch equations. F sq was calculated as the fishing mortality averaged over fishing years , and For the hake stocks, age- or season-based information could not be made available. Data available included aggregated annual time series of SSB and catch. Consistent with the spawning ogive used in this investigation, the hake SSB of all stocks corresponds exactly to the biomass of fish aged 7 and older. Based on the selectivity information given in Anonymous (2007) and Dunn et al. (2006), we assumed that fishing mortality was null over ages 1-3 and constant over ages 4-7 (HAK1 & HAK7), and null at age 1 and constant over ages 2-7 (HAK4). By combining the VPA and the Baranov catch equations, we could then derive the F matrix. F sq was calculated as the fishing mortality averaged over the past three years. Catchability estimates for the hoki and hake stocks are shown in Table 5. The initial hoki stock-at-age numbers have been drawn from the 2007 assessment. No recent assessment numbers were available for the three hake stocks. The initial hake stock-at-age numbers were then calculated as the equilibrium figures with status quo F. We also used B MSY proxies to evaluate the properties of the different management scenarios. For hake, we used the estimates (B MAY ) given in the assessment report (i.e. 19,810 t for HAK1, 7,500 t for HAK4, 26,300 t for HAK7). For hoki, no estimates of B MSY were available 11

12 in the assessment reports, and we used as a proxy 35% of the virgin biomass for each stock (i.e. 277,200 t for the eastern stock and 422,450 t for the western stock). Fleet dynamics By combining the hoki fisheries identified by Francis (2007) and the FMAs, we defined nineteen métiers (Table 7). Selectivity ogives by métier groups were given for the hoki stocks by Francis (2007). A simple knife-edge selectivity ogive was used for the three hake stocks based on information included in Dunn et al. (2006) and Anonymous (2007). The selectivity ogives implemented for the hoki and hake stocks are shown in Table 8. The standardization and the target factors associated to the hake and hoki populations were estimated using the GLM detailed in Appendix A, and the estimates are shown in Table 9. Four fleets (FL1, FL2, FL3 and FL4) were identified based on vessel size (trawlers below and above 46 m) and their main fishing strategy. Two strategies have been established by vessel size group, S1 and S2 for trawlers below 46 m (defining fleets FL1 and FL2 respectively), S3 and S4 for trawlers above 46 m (defining fleets FL3 and FL4 respectively). Each strategy consists of a monthly effort allocation by métier. The strategies have been identified by applying a cluster analysis to each fleet separately, using the Ward method. The variables analysed were, for each month, the proportion of days at sea allotted to each métier in fishing year 2005/2006. Two clusters were selected for each vessel size group, each of them representing one strategy. The main characteristics of the four fleets are shown in Table 10, while the four strategies used in this study are shown in Tables 11a and 11b. It should be noted that the actual average trip duration for the large trawlers as calculated from the log-books analysis was of 33 days, while the number of vessels was of 42 units. However, we were constrained to reduce the average trip duration for that fleet from 33 to 28 days, as ISIS-Fish does not allow fishing trips to exceed one month. In order to keep nominal fishing 12

13 effort invariant, we balanced the decrease in fishing trip duration by increasing the number of large trawlers from 42 to 50 vessels. The dynamics of the fleets were modeled using the gravity model. The overall nominal effort is assumed constant, but effort allocation across métiers may vary over time. The expected VPUE (Value Per Unit Effort) was chosen as the criterion to determine the effort allocation across métiers at each monthly time step, and that was estimated to be the VPUE obtained in the previous year and in the same month. The VPUE was calculated as the landing value minus ACE rental cost, and/or the deemed value. In the first year of simulation, the base strategies are defined as S1 and S2 (small trawlers), S3 and S4 (large trawlers). Management dynamics In this study, we investigate the impact of the two management tools applied to the New Zealand hoki fisheries: the TACC (i.e the commercial fishery share of the overall TAC) and the deemed value, as applied to the different hoki and hake stocks (Tables 1 and 2). For both management tools, we used three scenario levels: the current value (mean level), the current value minus 50% (low level) and the current value plus 50% (high level). We also investigated the effects of halving increasing by 50% the current level of nominal fishing effort. Each combination of TACC and deemed value has been evaluated. The different management scenarios are presented in Table 12. RESULTS Figure 3 shows the impact of the eleven management scenarios on the ratio between SSB and B MSY. The high effort scenario always lead to the lowest biomass for all stocks. The low effort scenario leads to a higher SSB for all stocks as expected. A consequence of decreasing the hoki deemed value and TACC is to shift fishing effort from the Chatham Rise to the West 13

14 Coast South Island. Compared to the base scenario, this shift results in an increase for the HAK4 and HOK1E SSBs, but more mixed results for the other stocks SSBs. The impact of the other scenarios is more limited. Note that the HAK1 and HOK1E spawning biomass is above B MSY for all years and all management scenarios. Figure 4 shows the impact of the eleven management scenarios on the ratio between catches and TACC. The most striking feature is that all management scenarios lead to an overshot of the HOK1W TACC. For the hake stocks (respectively eastern hoki stock), the greatest TACC overshot is caused by the low hake (respectively hoki) TACC scenario. As could be anticipated, high catch limis decrease the risk of TACC overshooting, for all stocks. A consequence of decreasing the hoki deemed value is to restrict the catches of the eastern hoki stock and of HAK4. Figure 5 shows the impact of the eleven management scenarios on the four fleets VPUE. The low hoki TACC scenario leads to negative VPUE. If VPUE is used as a proxy for fishing profit, then clearly this scenario is not viable. Likewise, the high hoki deemed value scenario is not viable for the large vessels (belonging to fleets FL3 and F4) for which hoki is the main target species. For all fleets, the low effort scenario always leads to large VPUEs. As explained above, a consequence of decreasing the hoki deemed value is to shift fishing effort from the east to the west, and the overall result is a large VPUE for all fleets. Increasing the hoki TACC does not have a marked effect on the hake and hoki SSBs (Figure 3), but it reduces the amount of deemed value paid for hoki. As a result, the VPUE obtained with the low hoki deemed value scenario is a large VPUE for most fleets. DISCUSSION 14

15 The results presented in the paper are of a preliminary form. More work is in particular needed to evaluate the impact of the different assumptions made on the model s outcomes. First, recruitment has been assumed constant for all stocks, and the impact of densitydependence on the model s outcome could be evaluated. An other important assumption made here was that the value of other species than hake and hoki is constant over time. However, other species (e.g. squids, barracuta, warehous, jack mackerel, ling, orange roughy) are believed to contribute substantially to fishers strategy and revenue. For most of these stocks, no assessment numbers are available, which adversely affect our ability to model the fluctuations of these species biomass and catches. A possible solution could be to envisage a variety of scenarios to project the other species dynamics in the future. Finally, while the model allows flexible effort allocation across métiers, the overall nominal effort was constant over time. We could contemplate to introduce a complimentary decision rule, which could reflect a decision to cease fishing when TACC is too low for the target species and/or when the expected profit is below some pre-determined level. Despite the limitations given above, which need to be addressed, this approach provides a useful canvas that may be used to evaluate the impact of administrative, as well as economic management, on mixed fisheries, and the resources they exploit. In this study, we could use this approach to rate the relative pro and cons of a variety of administrative (catch and effort limits) and economic (deemed value) management tools. This approach could in particular be useful in the context of setting deemed values based on conservation and economic considerations. 15

16 Future developments of the bio-economic model will include the implementation of (1) individual, transferable quota by fishing units, and (2) the development of an effort allocation model based on a variety of alternative determinants than VPUE, e.g. catch portfolios and traditions. ACKNOWLEDGEMENTS This work was funded through the TRANZEF project (Transposing New Zealand fisheries management experiences to EU fleets and fisheries: a bio-economic modeling approach) by the European Union (contract no: MOIF-CT ). The New Zealand Seafood Industry Council also provided assistance and funding. These supports are gratefully acknowledged. REFERENCES Annala, J.H., Sullivan, K.J., and Hore, A.J Management of multispecies fisheries in New Zealand by individual transferable quotas. ICES marine Science Symposium, 193, Annala, J.H., Sullivan, K.J., Smith, N.W.M., Griffiths, M.H., Todd, P.R., Mace, P.M., Connell, A.M Report from the Fishery Assessment Plenary, April 2004: stock assessments and yield estimates. Unpublished report held in NIWA library, Wellington, 690 p. Anonymous Report from the fishery assessment plenary, May 2007: stock assessments and yield estimates. New Zealand Ministry of Fisheries science group,

17 Arnason, R Advances in property rights based fisheries management: an introduction. Marine Resource Economics, 22, Branch, T.A., Rutherford, K., and Hilborn, R Replacing trip limits with individual transferable quotas: implications for discarding. Marine Policy, 30, Bull, B., Francis, R.I.C.C., Dunn, A., McKenzie, A., Gilbert, D.J., and Smith, M.H CASAL (C++ algorithmic stock assessment laboratory): CASAL user manual v /08/21. NIWA Technical Report 127, 274 p. Drouineau, H., Mahévas, S., Pelletier, D., and Beliaeff, B Assessing the impact of different management options using ISIS-Fish: the French Merluccius merluccius Nephrops norvegicus mixed fishery of the Bay of Biscay. Aquatic Living Resource, 19, Dunn, A., Ballara, S.L., and Phillips, N.L Stock assessment of hake (Merluccius australis) in HAK1 & 4 for the fishing year. New Zealand Fisheries Assessment Report 2006/11, 63 p. Francis, R.I.C.C Assessment of hoki (Macruronus novaezelandiae) in New Zealand Fisheries Assessment Report 2008/4, 109 p. Hentrich, S., and Salomon, M Flexible management of fishing rights and a sustainable fisheries industry in Europe. Marine Policy, 30, Mahévas, S., and Pelletier, D ISIS-Fish, a generic and spatially-explicit simulation tool for evaluating the impact of management measures on fisheries dynamics. Ecological Modelling, 171, Pelletier, D., and Mahévas, S Fisheries simulation models for evaluating the impact of management policies, with emphasis on marine protected areas. Fish and Fisheries, 6,

18 Sanchirico, J.N., Holland, D., Quigley, K., and Fina, M Catch-quota balancing in multispecies individual fishing quotas. Marine Policy, 30, Appendix A: Estimation of catchability, standardization and target factors A number of parameters (catchability, standardization and target factors) could not be inferred directly, and these were derived from the following GLM: Log(CPUE + 1.0e-8) = α year + β month + γ métier + ε The link function of the GLM is set to identity. The catch rates (CPUE) were aggregated over all fleets and vessels, and were calculated by species, statistical rectangle, target species, métier, month and fishing year. ε is a random noise assumed to be normally distributed. α, β and γ respectively refer to the year, month and métier effects. The year and month effects represent abundance dynamics, while the métier effect may be interpreted as a combination of standardization and target factors. The catchability q may then be derived from: q( stock, season, age) = F sq( stock, season, age) S( stock, métier, age) x E(métier) x exp( γˆ ) métier métiers where F sq is the status quo fishing mortality, S is the selectivity, E is the number of days at sea. 18

19 Table 1. TACC (in tonnes) set between fishing years 1996/1997 and 2005/2006 for hoki (HOK1) and hake (HAK1, HAK4, HAK7). We consider that 60% (respectively 40%) of the hoki TACC applies to the eastern (respectively the western) stock. Stock HOK1 HAK1 HAK4 HAK7 1996/ ,000 3,632 3,500 6, / ,000 3,632 3,500 6, / ,000 3,632 3,500 6, / ,000 3,632 3,500 6, / ,000 3,701 3,500 6, / ,000 3,701 3,500 6, / ,000 3,701 3,500 6, / ,000 3,701 3,500 6, / ,000 3,701 1,800 6, / ,000 3,701 1,800 7,700 Table 2. Average landing price, ACE prices and deemed values (all in New Zealand dollars per kg of fish) obtained in fishing year 2003/2004 for hoki (HOK1) and hake (HAK1, HAK4, HAK7). The deemed value shown here increases gradually in relation to the proportion of fish landed above the TACC. Stock Landing price ACE price Deemed value by category of TACC overshot <20% 20-40% 40-60% 60-80% % >100% HOK HAK HAK HAK

20 Table 3. Parameters characterizing growth, condition, natural mortality and recruitment of the hoki (HOK1E, HOK1W) and hake (HAK1, HAK4, HAK7) stocks, as implemented in the bioeconomic model. Stocks HOK1E HOK1W HAK1 HAK4 HAK7 No. age groups Growth L k t Condition factors a 4.79 x x x x x 10-6 W(kg)=a L(cm) b b Natural mortality M Recruitment R 3.38 x x x x x

21 Table 4. Migration ogives used in ISIS-Fish for the Eastern and Western hoki stocks. Age Eastern hoki Western hoki Chatham Rise Cook Strait Chatham Rise Sub-Antarctic Sub-Antarctic West Coast South Island 1 st July 1 st April 1 st July

22 Table 5. Catchability coefficients of the hoki (HOK1E, HOK1W) and hake (HAK1, HAK4, HAK7) stocks. HOK1E HOK1W HAK1 HAK4 HAK7 Oct.-Nov. Dec.-Mar. Apr.-Jun. Jul.-Sep. Oct.-Nov. Dec.-Mar. Apr.-Jun. Jul.-Sep E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E-06 22

23 E E E E E E E E-06 23

24 Table 6. Initial stock-at-age numbers (in million) as used in the simulations for the hoki (HOK1E and HOK1W) and hake (HAK1, HAK4, HAK7) stocks. Age HOK1E HOK1W HAK1 HAK4 HAK7 Chatham Rise Cook Strait Chatham Rise Sub-Antarctic West Coast South Island

25 Table 7. List of the 19 métiers implemented in ISIS-Fish. FMA Fishing ground Season Hoki stock partition Métier code 1 Chatham Rise All year round Non spawners 1 2 Chatham Rise All year round Non spawners 2 Cook Strait October May Non spawners 3 June - September Spawners 4 3 Chatham Rise All year round Non spawners 5 Cook Strait October May Non spawners 6 June - September Spawners 7 Sub-Antarctic area All year round Non spawners 8 4 Chatham Rise All year round Non spawners 9 5 Sub-Antarctic area All year round Non spawners 10 West Coast South Island All year round Spawners 11 Others All year round All 12 6 Sub-Antarctic area All year round Non spawners 13 7 Cook Strait October May Non spawners 14 June - September Spawners 15 West Coast South Island All year round Spawners 16 Others All year round All 17 8 All All year round All 18 9 All All year round All 19 25

26 Table 8. Selectivity ogives used in ISIS-Fish for the hoki and hake stocks by métier groups. The métier groups identified in relation to the hoki fishery consist of: Ensp (métiers no. 1, 2, 3, 5, 6, 9, 14), Wnsp (métiers no. 8, 10, 13), Esp (métiers no. 4, 7, 15), Wsp (métiers no. 11, 16) and others (métiers no. 12, 17, 18, 19). The codes refer to the métiers presented in Table 6. Age HOK1 (Eastern and Western stocks) / métier groups HAK1 HAK4 HAK7 Ensp Wnsp Esp Wsp Others

27 Table 9. Hake and hoki. Métier effect derived from the GLM, which is here interpreted as the product of the standardization factor and of the target factor. The codes refer to the métiers presented in Table 6. Métier code Métier effect Hake Hoki

28 Table 10. Main characteristics of the four fleets FL1, FL2, FL3 and FL4 investigated in the analysis, based on fishing year 2005/2006. The details of strategies S1, S2, S3 and S4 are given in Tables 10a and 10b. Fleet FL1 Fleet FL2 Fleet FL3 Fleet FL4 Main gear trawl trawl trawl trawl Length range m m m m Mean trip duration 4 days 4 days 28 days 28 days Number of vessels Strategy S1 S2 S3 S4 28

29 Table 11a. Monthly effort allocation (%) in strategies S1 and S2. Strategy Métier Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. S

30 S

31 Table 11b. Monthly effort allocation (%) in strategies S3 and S4. Strategy Métier Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. S S

32

33 Table 12. Different management scenarios simulated (DV = deemed value). Scenario 0 is the baseline. Scenario TACC hake TACC hoki DV hake DV hoki Nominal effort 0 current current current current current 1-50% current current current current 2 +50% current current current current 3 current -50% current current current 4 current +50% current current current 5 current current -50% current current 6 current current +50% current current 7 current current current -50% current 8 current current current +50% current 9 current current current current -50% 10 current current current current +50% 33

34 Figure 1. ISIS-Fish: interaction between the three sub-models of population dynamics, fishing activity and management. Spatial and seasonal fisheries dynamics Link between fishing effort and mortality Spatial and seasonal stock dynamics Fishing mortality, catch and biomass by month & area Spatial intersections Allocation of fishing effort Spatial and temporal management sub-model 34

35 Figure 2. Map of the New Zealand Exclusive Economic Zone including Fisheries Management Areas (FMA) and statistical areas. Relevant Statistical and Management Areas for MDF Metiers Challenger/ Central (Plateau) E 9 Auckland (West) Central (Egmont) Auckland (East) Kermadec S 3 35 S South-East 024 Coast Central 013 (East) 40 S 40 S South-East (Chatham Rise) S 45 S 175 E Southland Sub-Antarctic E Nautical Miles 175 E 1:10,000,000 Map Projection: Mercator

36 Figure 3. Ratio (SSB/Bmsy) by simulation type and by stock. 36

37 Figure 4. Ratio (Catch/TACC) by simulation type and by stock. 37

38 Figure 5. VPUE by simulation type and by fleet. 38