- Email: [email protected]
Varying components of productivity and their impact on fishing mortality reference points for Grand Bank Atlantic cod and American plaice
Fisheries Research 155 (2014) 64–73
Contents lists available at ScienceDirect
Fisheries Research journal homepage: www.elsevier.com/locate/fishres
Varying components of productivity and their impact on fishing mortality reference points for Grand Bank Atlantic cod and American plaice M.J. Morgan ∗ , P.A. Shelton, R.M. Rideout Fisheries and Oceans Canada, PO Box 5667, St. John’s, NL A1C 5X1, Canada
a r t i c l e
i n f o
Article history: Received 6 August 2013 Received in revised form 12 February 2014 Accepted 17 February 2014 Handling Editor George A. Rose Available online 20 March 2014 Keywords: Productivity Reference points Recruitment Maturity Weight
a b s t r a c t Population productivity is determined in large part by growth, maturity and recruitment. This determines the level of fishing that the population can sustain without declining. We examined how these components of productivity have varied during warm and cold periods for two Grand Bank fish populations, and how this variation affects fishing mortality reference points. Productivity of both Div. 3NO Atlantic cod and Div. 3LNO American plaice has varied considerably over time. Projections of population size under different conditions showed that the level of recruits per spawner played a major role in determining the level of fishing mortality that did not result in population decline. For plaice there was also a substantial impact of varying proportion mature at age. The impact of factors other than temperature on recruits per spawner and maturity meant that the association of productivity with temperature was not consistent. When productivity was at its lowest, the level of fishing mortality that could be sustained without causing rapid population decline was very low. The results of this study clearly demonstrate that the impacts of changing productivity can be rapid and very large and if fishing mortality reference points are not adjusted accordingly the results can be catastrophic. It may be helpful to develop fishing mortality reference points that incorporate all components of productivity and are updated using only recent data. Such reference points could be used in combination with buffers around a biomass limit reference point (Blim ) and/or low risk tolerances for declining below Blim to account for the uncertainties that would remain even in frequently updated reference points. Crown Copyright © 2014 Published by Elsevier B.V. All rights reserved.
1. Introduction The productivity of a fish population is the capacity of that population to produce biomass, and is a result of increases due to growth and reproduction, countered by declines due to mortality. The productivity of a fish population determines the level of fishing that can be sustained without a decline in population size. Major components of productivity are recruitment, weight at age, maturity at age and mortality. These components vary over time and therefore, so too does the overall productivity of the population. Recruitment is linked to the size (i.e. biomass or abundance) and composition of the spawning stock and is also affected by environmental conditions (Myers and Barrowman, 1996; Shepherd et al., 1984; Cushing, 1996). Growth varies with temperature and prey availability (Brett, 1979; Shelton et al., 1999; Bjornsson et al., 2001). Maturity has been shown to vary with temperature and mortality and is related to
∗ Corresponding author. Tel.: +1 709 772 2261; fax: +1 709 772 4188. E-mail address: [email protected] (M.J. Morgan). http://dx.doi.org/10.1016/j.fishres.2014.02.019 0165-7836/Crown Copyright © 2014 Published by Elsevier B.V. All rights reserved.
growth (Kjesbu et al., 1998; Morgan and Colbourne, 1999; Olsen et al., 2005). Estimates of natural mortality are few but it is not constant over time and can be influenced by environmental conditions and changes in predation level (Dutil and Lambert, 2000; Sinclair, 2001; Morgan and Brodie, 2001; Chouinard et al., 2005). Fishing mortality (F) reference points play an important role in the management of many fisheries and are determined by population productivity. As the components of productivity vary, so too do the reference points derived from them (A’mar et al., 2009; Brooks, 2013; Heino et al., 2013; Wayte, 2013). These reference points can be targets to be achieved or limits that should be avoided (Caddy and McGarvey, 1996). Three of the most commonly used fishing mortality reference points are FMSY , F0.1 and F40%SPR . F0.1 is generally used as a target, F40%SPR is a limit, while FMSY is used as both, depending on the fisheries management jurisdiction. FMSY is the fishing mortality giving the maximum sustainable yield from a population (Schaefer, 1954). Fishing at levels above FMSY will result in a population size that is lower than BMSY , the biomass giving maximum sustainable yield (MSY). F0.1 is determined as the F where the slope of the yield per recruit (YPR) curve is 10% of the slope at the
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73
65
Morgan et al., 2010). There is also evidence that recruitment rate is related to ocean climate in these two populations (Stige et al., 2006; Halliday and Pinhorn, 2009). The extended warm and cold periods, coupled with the fact that components of productivity are influenced by temperature, suggest the potential for long term changes in productivity of these two Grand Bank fish populations. In this study we examine time series of abundance and biological data on these stocks for indications of changes in weight at age, maturity at age and recruitment rate, during warm and cold periods. We estimate FMSY , F0.1 , and F40%SPR using the productivity from these periods and determine which components of productivity (recruitment, weight at age, maturity at age) are responsible for any differences. We also explore the implications of failing to recognize changes in stock productivity by examining the consequences of fishing the populations when they are at one level of productivity at reference points derived from a different level of productivity. Fig. 1. Rogers North Atlantic Oscillation (NAO) anomaly (millibars of pressure) relative to the 1981–2010 average. The warm and cold periods examined are indicated. Warm 1 is 1962–1966, Cold 1 is 1972–1976, Cold 2 is 1990–1994 and Warm 2 is 2001–2004.
origin. F40%SPR is the F that reduces spawner per recruit (SPR) to 40% of the unfished value. Changes in population productivity will influence the estimation of these three reference points differentially because they incorporate different components of productivity in their calculation. FMSY is determined by all components of productivity and will therefore be affected by changes in any component of the productivity of the population (Morgan et al., 2009; Brooks, 2013). Reference points derived from YPR or SPR do not include the recruitment component of productivity. YPR based reference points will respond only to variation in growth while SPR reference points will respond to changes in growth and/or maturity. If changes in productivity are short term, then the impact of assuming constant conditions is likely to be small. However, extended periods of low productivity could result in population decline if F is set at a reference point level based on an assumption of constant, more productive, conditions (Cook and Heath, 2005; Koster et al., 2009). Cold or warm ocean temperatures can persist for several to many years (Colbourne et al., 2012). To understand how productivity will differ and which reference points will respond to this variation, it is important to understand how the components of productivity may change with varying environmental conditions (Kell et al., 2005). Major variation in ocean temperature has been observed on the Grand Bank in the Northwest Atlantic off Newfoundland Canada over the last 50 years. There was an extended period of warm temperature in the 1960s, while the early 1970s and early 1990s were generally cold, with recent years once again characterized by warm ocean temperature (Colbourne, 2004; Colbourne et al., 2012). Thus there has been the potential for lengthy periods of high and low productivity of fish stocks in the area, driven by the direct and indirect effects of temperature. Northwest Atlantic Fisheries Organization (NAFO) Division 3NO Atlantic cod (Gadus morhua) and Div. 3LNO American plaice (Hippoglossoides platessoides) are two important groundfish populations on the Grand Bank. Both experienced severe population decline due to over fishing and spawning stock biomass (SSB) for both stocks is currently at a low level (Power et al., 2010; Rideout et al., 2011). They have also both exhibited major changes in maturity at age that may have been related, in part to temperature. Growth has also been shown to be related to temperature, although weakly, for Div. 3NO cod and for Div. 3LNO plaice, cohorts that grew faster and occupied warmer temperatures were those that matured earlier (Morgan and Colbourne, 1999;
2. Materials and methods 2.1. Productivity The Grand Bank has experienced warm and cold periods over the last half century. The 1960s and 2000s were generally warm while the early 1970s and the early 1990s were cold years (Colbourne, 2004; Colbourne et al., 2012). We chose years within these time periods based on anomalies in the North Atlantic Oscillation (NAO), an index of ocean climate on the Grand Bank (Colbourne, 2004), as a basis for comparisons of productivity. The years 1962–1966 (Warm 1), 1972–1976 (Cold 1), 1990–1994 (Cold 2) and 2001–2004 (Warm 2) were chosen to represent the 1960s, 1970s, 1990s and 2000s respectively (Fig. 1). These were years with consistently high (cold) or low (warm) NAO. For the first three periods five years were chosen in order to have the same number of years in each period. The 2000s were more variable and so there was only a 4-year period with a consistent NAO during which population data were available. All analyses and data are based on information from recent assessments of Div. 3NO cod (Power et al., 2010) and Div. 3LNO American plaice (Rideout et al., 2011). Model estimates of population numbers at age in each year were extracted for each stock. Stock weights at age (for calculating SSB) and catch weights at age (for calculating catch) were based on commercial sampling conducted for these assessments. Maturities at age in these assessments were modeled by cohort based on research vessel data. SSB was calculated as the sum of the product of model estimates of numbers at age (both sexes combined), model estimates of female proportion mature at age, and beginning of year stock weights at age (both sexes combined). Recruitment was calculated from the assessments as number at age 1 such that r = Nagerec × eM×a
(1)
where Nagerec is the number of recruits and a is difference between the age at recruitment estimated in the assessment and age 1. For 3NO cod a is 1 and 4 for 3LNO American plaice. M is natural mortality and was set equal to 0.2 for 3NO cod for all years. For 3LNO American plaice M was 0.2 for all years except from 1989 to 1996 when it was 0.53, as is the case in the assessment, to include an increase in natural and other unaccounted for mortality over that period (Morgan and Brodie, 2001). Weight at age, maturity at age, recruits per spawner (RPS), and spawner per recruit (SPR) at F = 0 were examined to determine how they differed among Warm 1, Cold 1, Cold 2, Warm 2 and average conditions. For each period, average weights, maturities and RPS were calculated. Variation in recruitment was expressed as RPS as
66
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73
this metric accounts for some of the influence of SSB on recruitment without assuming a specific form of a stock recruit relationship, other than a constant recruitment rate during each period. SPR at F = 0 was calculated as SPR =
A
Na Wa Pa
(2)
a=1
where Na is the number at age a starting with one recruit, Wa and Pa are the weight at age and proportion mature at age, respectively, averaged over each time period. A is the terminal age in the assessment. Na is incremented as: Na+1 = Na × e−M
(3)
where M is the annual instantaneous rate of natural mortality. The percent growth rate in SSB at F = 0 (G0 ) reflects the combined impact of RPS and SPR on the productivity of the stock at F = 0 (Shelton et al., 2006). This is a metric similar to the intrinsic rate of increase but expressed in terms of SSB. It allows easy comparison of potential productivity across periods. It was calculated as G0 = average
SSB − SSB t t−1 SSBt−1
× 100
(4)
where N is incremented as in Eq. (3) and SSB is calculated using the average stock weight and maturity at age from each period. G0 is measured when the stock is below the break point of the hockey stick stock recruit curve and has a stable age composition. Recruitment was calculated as:
R=
˛ × SSB ˛×ˇ
if SSB < ˇ
(5)
if SSB≥ˇ
where R is recruitment, ˛ is RPS and ˇ is the maximum SSB estimated in the assessment for each population. 2.2. Fishing mortality reference points Projections of stock size were carried out to equilibrium over a range of F values to determine the F giving maximum sustainable yield (FMSY ). This reference point is affected by changing recruitment rate, weight and maturity at age. Population numbers at age were projected as: Na+1,y+1 = Na,y × e−(M+Fa,y )
(6)
where Na,y is the number alive at age a and the beginning of year y and Fa,y is the fishing mortality on age a in year y, obtained from: Fa,y = Ka Fy
N15,y+1 = N14,y e−(M+F14,y ) + N15,y e−(M+F15,y )
(8)
Yield in weight was calculated in each year as: n
3. Results
(7)
where Fy is the fully recruited fishing mortality in year y and Ka is the selectivity or partial recruitment to the fishery at age a (that is the proportion of Fy for that age). For 3NO cod, a = 1–12 with no plus group. For 3LNO plaice, a = 1–15, where age 15 is a plus group, which was updated in the simulation by applying:
Yield =
As in the calculation of G0 , recruitment was assumed to come from a hockey stick stock recruit relationship (Eq. (5)). This assumes the same break point for all periods. Density dependence in recruitment is assumed to occur at the same SSB for all RPS as a representation of what recruitment might be expected to occur under the prevailing conditions of each period. This allows results to be interpreted with respect to RPS without the confounding effect of different break points. In estimation of FMSY population size exceeded the break point by the end of the projection period for all time periods considered. All projections applied the same partial recruitment vector calculated from the matrix of fishing mortality at age for each stock. Natural mortality was (0.2) for all cases except that SPR at F = 0, G0 and the reference points were also calculated using an M = 0.53 for American plaice for Cold 2. This is done in the assessment to deal with unaccounted for mortality, some of which is thought to have been natural mortality (Morgan and Brodie, 2001). Weight at age and proportion mature at age averaged over each time period (Warm 1, Cold 1, Cold 2, Warm 2) and over the whole time series were used in analyses together with the average RPS calculated over the corresponding period. The YPR reference point F0.1 was calculated for each period using Eqs. (6)–(10) above, starting with 1 recruit and applying a range of F values, to produce a curve of yield across F. For American plaice the population was extended to age 30 to approximate the plus group in the assessment. F0.1 is determined as the F where the slope of the yield per recruit curve is 10% of the slope at the origin. This reference point is affected by variation in weight at age but not by RPS or maturity at age. Average weight at age in each time period and the overall average were used to examine how F0.1 varied across period. The F that gave 40% of the SPR at F = 0 (F40%SPR ) was determined by solving for the value of F giving this amount of SPR depletion per recruit using Eqs. (2), (6) and (7) above. This reference point is affected by changing weight and maturity at age but not recruitment rate. The impact of fishing at these reference points under varying productivity was examined. Populations were projected to equilibrium at FMSY determined for Warm 1, Cold 1, Cold 2 and Warm 2 and the average weights, maturities and RPS from the corresponding period. The response of the population was compared over 20 year projections from equilibrium at the three F reference points (FMSY , F0.1 , F40%SPR ) derived from the period being examined and from average conditions.
Ca,y × Cwa,y
(9)
a=1
where Cwa,y is the average catch weight at age a in year y and Ca,y is the catch number at age a in year y calculated as:
Ca,y = Na,y ×
1−e
−(M+Fa,y )
×
Fa,y M + Fa,y
(10)
3.1. Productivity Proportion mature at age for both stocks was lowest during Warm 1 and Cold 1, and highest during the most recent period, Warm 2. The variation was much greater for plaice than for cod (Fig. 2). Weight at age for older fish was markedly reduced during Warm 1 compared to the other periods for both stocks. For cod, ages 7–12 had lower weight at age in Warm 1, while for plaice it was ages 11–15 that had lower weight. Both 3NO cod and 3LNO plaice have exhibited substantial variation in RPS (Fig. 3). The timing of this variation and the association between RPS and SSB was not the same for the two populations although both stocks had extended periods when RPS was below average. There was a 24-year period of low RPS from 1981 to 2004 for cod and 11- and 10-year periods of low RPS for plaice from 1961 to 1971 and 1979 to 1988 respectively. During Warm 1 both SSB and RPS were high for cod while SSB was high and RPS low for plaice (Table 1). During Cold 1, RPS started below average for cod
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73
67
Fig. 2. Average estimated proportion mature at age and weight at age for cod and American plaice from each of four different time periods and for the entire time series.
and then increased rapidly as SSB declined to very low levels, with a resulting average RPS for the period that was above the long term average. For plaice, RPS was above average during Cold 1, while SSB was about average, although as for cod, RPS increased over the time period while SSB declined. During Cold 2, SSB and RPS were low for cod with low SSB and high RPS for plaice. In recent years when the NAO was again low and conditions warm (Warm 2), there has been
mainly below average RPS at low SSB for both American plaice and cod. Over all average RPS was much higher for plaice than for cod. Average SPR for plaice was an order of magnitude lower than for cod (Fig. 3) as a result of lower weight and later maturity at age for plaice. For both populations there has been a general increase in SPR since the beginning of the series but this increase is much greater for plaice. This results in higher SPR in Cold 2 and Warm 2,
Table 1 Average North Atlantic Oscillation anomaly (NAO, millibars of pressure), spawning stock biomass (SSB, ‘000 t), recruits per spawner (RPS), spawner per recruit (SPR), and fishing mortality for the selected periods used in the estimation of reference points. The estimated percentage SSB (G0 ) at F = 0 is also shown. For American plaice the two numbers for SPR and G0 during Cold 2 are for M = 0.2 and M = 0.53 (in brackets). All values are averages except for G0 which was calculated using the average values of RPS, maturity and weight at age for the different periods. NAO Cod Warm 1 Cold 1 Cold 2 Warm 2 Average
−13.6 4.84 5.3 −5.6 −2.2
American plaice Warm 1 Cold 1 Cold 2 Warm 2 Average
−13.6 4.84 5.3 −5.6 −2.2
SSB
RPS
SPR
G0
F
92.3 44.4 18.6 7.6 43.2
2.74 1.42 0.17 0.49 1.17
4.52 5.84 6.45 6.64 5.95
32.97 28.5 1.08 14.49 24.14
0.280 0.814 0.756 0.356 0.387
170 81.5 34.4 21.03 94.2
2.43 5.8 8.50 3.34 4.43
0.48 0.76 0.87 (0.02) 0.93 0.73
4.16 11.58 18.99 (−12) 11.81 11.67
0.122 0.407 0.752 0.269 0.328
68
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73
Fig. 3. Spawning stock biomass (‘000 t), recruits per spawner, and spawner per recruit for Div. 3NO cod and Div. 3LNO American plaice. The dashed horizontal lines give the biomass limit reference point established for each stock in the spawning stock biomass panels and average spawner per recruit and recruits per spawner over the whole time series in the other panels. The gray areas indicate the warm periods and the gray with stipple the cold periods used in this study. For American plaice the solid line in the spawner per recruit graph indicates the result using an M of 0.20 during 1989–1996, while the dotted line uses an M of 0.53 during that time.
the most recent periods. The SPR for plaice during the Cold 2 is almost zero if an M of 0.53 is assumed, as it is for that time period in the assessment. The differing RPS resulted in very different levels of recruitment when applied as ˛ in a hockey stick stock recruit model (Fig. 4). The much larger RPS for plaice results in much greater recruitment than for cod. The highest levels of recruitment for plaice were during Cold 2 while for cod it is during Warm 1. Maximum recruitment
for cod in Warm 1 was 16 times that of the Cold 2 period. For plaice, maximum recruitment during the Cold 2 period was 3.5 times that of Warm 1. For both cod and plaice the second highest level of recruitment was during Cold 1. The potential growth in the SSB (G0 ) was very low for cod during Cold 2 (Table 1), at just over 1%. In comparison during Warm 1, G0 for cod was 33%. During Cold 1 G0 was just above average for cod while during Warm 2 it was below average. The highest G0
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73
69
Table 2 Estimated FMSY , F0.1 and F40%SPR for different time periods for Div. 3NO cod and Div. 3LNO American plaice. For American plaice the two numbers in Cold 2 are for M = 0.2 and M = 0.53 (in brackets). Period
Warm 1 Cold 1 Cold 2 Warm 2 Average
Cod
American plaice
FMSY
F0.1
F40%SPR
FMSY
F0.1
F40%SPR
0.33 0.29 0.02 0.22 0.30
0.21 0.19 0.18 0.20 0.20
0.12 0.12 0.12 0.12 0.12
0.08 0.29 0.33 (0) 0.48 0.37
0.19 0.17 0.16 (0.88) 0.16 0.17
0.05 0.05 0.06 (0.12) 0.07 0.06
Cold 1 and Warm 2 were periods of intermediate productivity for both stocks. 3.2. Reference points The estimated FMSY differed substantially for both plaice and cod across periods, by 80 and 95% respectively (Table 2). The variation in F0.1 and F40%SPR was much less. F40%SPR was much lower for plaice than for cod but varied more (28% vs 5%) reflecting large changes in maturity at age in plaice. During periods of low RPS, FMSY was very low for both cod and plaice (Cold 2 and Warm 1 respectively). Both F0.1 and F40%SPR were much higher than FMSY during the Cold 2 period for cod. But for plaice, while F0.1 was much higher than FMSY during the Warm 1 period, F40%SPR was similar to FMSY . For cod, fishing under the low productivity conditions of Cold 2 with reference points derived under conditions averaged over the whole time series has a very detrimental effect on SSB, causing a steep decline (Fig. 5). Both F0.1 and F40%SPR derived from conditions during Cold 2 also cause a steep decline in SSB under these low RPS conditions. During Warm 2 conditions, the FMSY estimated using average conditions also causes the SSB to decline over the projection period. For plaice, Warm 1 was the low productivity period. Average FMSY and F0.1 both result in population decline under the conditions of the Warm 1 period, while average F40%SPR allows the SSB to grow (Fig. 6). The F0.1 estimated using the weight at age from Warm 1 also results in a decrease in SSB. 4. Discussion
Fig. 4. Stock recruit relationships for Div. 3NO cod and Div. 3LNO American plaice used in the study. The relationship is based on the assumption of a segmented regression with a slope equal to the average recruits per spawner for each period and a plateau at that recruits per spawner times the maximum spawning stock biomass estimated in the assessment for each population.
for plaice was during Cold 2, G0 during Warm 1 being well below average. Cold 1 and Warm 2 had near average G0 for plaice. Overall, Warm 1 appears to have been a high productivity period for cod with high RPS and very high G0 , although SPR was below average because of the lower weights and maturity at age. The cold 1990s were a period of very low productivity for cod despite a higher SPR. For plaice, Warm 1 was a low productivity period, with low RPS, SPR and consequently low G0 . The Cold 2 period was one of high productivity for plaice. However, if natural mortality did increase for plaice during this time then it would have been a period of very low productivity for this population as well (Table 1).
The analyses presented here for Grand Bank cod and American plaice populations demonstrated substantial temporal variability in potential growth in SSB (G0 ), used here as a general indicator of population productivity. The variation in productivity stemmed from changes in all of the components of productivity examined: weight at age, maturity at age and recruitment rate. For both species, variation in recruitment rate was substantial and was the main contributor to changes in G0 , while changes in weight at age had little impact for either species. A larger impact on stock productivity of variation in recruitment than in growth was also reported by Kell et al. (2005). This is not unexpected given the large variation in recruitment experienced by most marine fish stocks. Both species exhibited changes in maturity at age but the variation was greater for plaice and therefore made a more substantial contribution to variation in G0 for that species than for cod. Average RPS is four times higher for plaice than cod, whereas average SPR is an order of magnitude higher for cod than plaice. The causes of the larger RPS for plaice are unclear but their lower SPR is the result of smaller weight at age and later maturation. In combination, RPS and SPR result in an average G0 value for cod that is twice as high as plaice, indicating that higher recovery rates could be expected from the depleted cod stock compared to the depleted plaice stock. The fact that neither stock has recovered above their respective SSB limit reference points since moratoria were applied to directed fishing in the early 1990s indicates that a large proportion of the
70
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73
Fig. 5. Spawning stock biomass (SSB ‘000 t) resulting from projections of population size of Div. 3NO cod. Projections start with equilibrium population size calculated using average weight at age, maturity at age and RPS for the period and FMSY specific to each period. Panels on the left use FMSY , F0.1 and F40%SPR derived under average conditions for the entire time series while those on the right use those reference points derived from the average conditions for the particular period.
surplus production under prevailing productivity conditions has been removed through bycatch mortality (Shelton and Morgan, 2005). Our results support previous work indicating that environmental conditions such as temperature can have a major impact on all of the components of productivity (Brett, 1979; Shepherd et al., 1984; Cushing, 1996; Kjesbu et al., 1998; Morgan and Colbourne,
1999). It is important to note, however, that temperature is not the only factor potentially influencing productivity. For example for cod, highest RPS was found in the warmest period, Warm 1 and the lowest RPS in the coldest period, Cold 2, with the opposite being the case for plaice. However, a fundamental aspect of population dynamics is the compensatory response of RPS to changes in population size (Ricker, 1954) and any influence of temperature would
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73
71
Fig. 6. Spawning stock biomass (SSB ‘000 t) resulting from projections of population size of Div. 3LNO American plaice. Projections start with equilibrium population size calculated using average weight at age, maturity at age and RPS for the period and FMSY specific to each period. Panels on the left use FMSY , F0.1 and F40%SPR derived under average conditions for the entire time series while those on the right use those reference points derived from the average conditions for the particular period.
interact with this response. Some evidence of this was seen for both species on the Grand Bank, with cod RPS increasing in response to declining SSB during Cold 1 and plaice RPS increasing in response to a decrease in SSB during Cold 2. For cod this led to an above average RPS for Cold 1 while RPS was well below average during Cold 2. Warm temperatures can lead to earlier age at maturity (Alm, 1957; Kjesbu et al., 1998; Morgan and Colbourne, 1999) but the dominant trend for both species in this study was a temporal one with fish
maturing younger in more recent years regardless of temperature. The change in maturity in both cod and plaice may be at least partly driven by fishing mortality (Barot et al., 2005; Olsen et al., 2005). This shift can lead to greater productivity for a population adapted to fishing (Heino et al., 2013). The patterns in variation in RPS and maturity meant that periods of high and low productivity were not the same for the two species. Warm 1 was a period of high productivity for cod with high RPS
72
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73
leading to a very high G0 despite lower weight and maturity at age. This same period was one of very low productivity for plaice, arising from low RPS, weights and maturity at age. Cold 2 was a period of low productivity for cod with very low RPS leading to limited potential for population growth. For plaice on the other hand, Cold 2 was a period of high RPS and high maturity at age leading to high productivity (under the assumption of M = 0.2). The impact of factors other than temperature on RPS and maturity meant that the association of productivity with temperature period was not consistent. For example, while Warm 1 had a very low G0 for plaice, the shift to maturation at an earlier age meant that Warm 2 had average G0 despite low RPS. Previous studies have examined the impact of changes in temperature on productivity and reference points in the context of climate change (Cook and Heath, 2005; Kell et al., 2005). The results here show that the impact of drivers other than temperature may make it difficult to predict how productivity will change as temperature changes. The variation in productivity had a large impact on the level of fishing mortality based reference points, particularly FMSY . When productivity was at its lowest, FMSY was very low. Low or even negative (i.e. population decline) productivity has been found in other stocks, sometimes lasting for extended periods of time (Shelton et al., 2006; Swain and Chouinard, 2008; Lambert, 2011). That different assumptions about productivity can lead to different estimates of fishing mortality reference points has been demonstrated previously (Beddington and May, 1977; Morgan et al., 2009; Brooks, 2013; Cervino et al., 2013). The results of this study clearly demonstrate that the impacts of changing productivity can be rapid and very large, and if fishing mortality reference points are not adjusted accordingly the results can be catastrophic. The variation in FMSY under changing productivity conditions was greater than that of F0.1 and F40%SPR . This is a result of the importance of changing RPS in both populations. The relative importance of the components of productivity leading to the variation in FMSY was not the same in the two populations. In cod, the main source of variation was change in RPS, while for plaice there was also a substantial change in SPR caused by varying proportion mature at age, which caused greater variation in F40%SPR for plaice compared to cod. Changes in weight at age seemed to be of less importance than changes in maturity in both stocks. Changes in weight at age occurred mainly at older ages and would have a reduced impact on stock productivity under higher levels of F. During periods of low productivity, RPS was very low for both populations. The lack of inclusion of recruitment in F0.1 and F40%SPR meant that under conditions of low RPS they could be much higher than FMSY . Cervino et al. (2013) also found a discrepancy between FMSY and yield per recruit reference points using a Ricker stock recruit curve, in that case comparing different measures of reproductive potential. Kell and Fromentin (2007) determined that F0.1 performed better than F40%SPR when carrying capacity was variable. How well a particular type of reference point performs will depend on what components of productivity are changing. Given the large impact of RPS seen here and the generally large variation in recruitment experienced by fish stocks, MSY based reference points may be more robust than those based on YPR or SPR. Fishing at reference points derived from average conditions or with YPR or SPR based reference points when RPS was low resulted in rapid population decline. The periods of low RPS observed here in both populations was long enough to result in substantial population decline and even collapse. The level of constant fishing mortality that would ensure that the populations would not decline under low productivity conditions would be so low as to result in a large loss of yield if applied to all time periods, implying that target fishing mortality should be adjusted between low and higher productivity periods. We propose that this be done by applying recent average data in the
computation of fishing mortality target and limit reference points. Estimates of weight and maturity at age and recruitment need to be compared to those being used to calculate current reference points to determine if their variation warrants updating of the reference points (Brooks, 2013). In practice, it is usually YPR and SPR type reference points that are updated using recent data, while FMSY is usually estimated based on long term data. YPR and SPR based reference points do not incorporate recruitment and our study demonstrated that RPS can be so low as to make YPR and SPR based reference points dangerously high during periods of low recruitment. It may be helpful to develop fishing mortality reference points that incorporate all components of productivity and are updated using only recent data. One possible approach is to calculate FMSY using recent average RPS and a hockey stick stock recruit relationship as done in this study. Another potential approach is to compute G0 based on recent RPS and SPR and to apportion a fraction of that to the fishery and a fraction to population growth, the relative proportion depending on whether or not the population is depleted or healthy. Such reference points could be used in combination with buffers around a biomass limit reference point (Blim ) and/or low risk tolerances for declining below Blim to account for the uncertainties that would remain even in frequently updated reference points. Acknowledgements We thank all the sea going and laboratory staff involved in the collection of these data. A. Buren, L. Brooks and an anonymous reviewer provided helpful comments on an earlier version of the MS. This study was funded by Fisheries and Oceans Canada’s Aquatic Climate Change Adaptation Services Program. References Alm, G., 1957. Connections Between Maturity, Size and Age in Fishes. Reports from the Institute of Freshwater Research, Drottningholm, vol. 40., pp. 5–145. A’mar, Z.T., Punt, A.E., Dorn, M.W., 2009. The impact of regime shifts on the performance of management strategies for the Gulf of Alaska walleye Pollock (Theragra chalcogramma) fishery. Can. J. Fish. Aquat. Sci. 66, 2222–2242. Barot, S., Heino, M., Morgan, M.J., Dieckman, U., 2005. Maturation of Newfoundland American plaice (Hippoglossoides platessoides): long-term trends in maturation reaction norms despite low fishing mortality? ICES J. Mar. Sci. 62, 56–64. Beddington, J.R., May, R.M., 1977. Harvesting natural populations in a randomly fluctuating environment. Science 197, 463–465. Bjornsson, B., Steinarsson, A., Oddgeirsson, M., 2001. Optimal temperature for growth and feed conversion of immature cod (Gadus morhua L.). ICES J. Mar. Sci. 58, 29–38. Brett, J.R., 1979. Environmental factors and growth. In: Hoar, W.S., Randall, D.J., Brett, J.R. (Eds.), Fish Physiology, vol. III. Academic Press, New York, pp. 599–675. Brooks, E.N., 2013. Effects of variable reproductive potential on reference points for fisheries management. Fish. Res. 138, 152–158. Caddy, J.F., McGarvey, R., 1996. Targets or limits for management of fisheries? North Am. J. Fish. Manage. 16, 479–487. Cervino, S., Dominguez-Petit, R., Jardim, E., Mehault, S., Pineiro, C., Saborido-Rey, F., 2013. Impact of egg production and stock structure on MSY reference points and its management implications for southern hake (Merluccius merluccius). Fish. Res. 138, 168–178. Chouinard, G.A., Swain, D.P., Hammill, M.O., Poirier, G.A., 2005. Covariation between grey seal (Halichoerus grypus) abundance and natural mortality of cod (Gadus morhua) in the southern Gulf of St. Lawrence. Can. J. Fish. Aquat. Sci. 62, 1991–2000. Colbourne, E.B., 2004. Decadal changes in the ocean climate in Newfoundland and Labrador waters from the 1950 to the 1990. J. Northwest Atl. Fish. Sci. 34, 41–59. Colbourne, E.B., Craig, J., Fitzpatrick, C., Senciall, D., Stead, P., Bailey, W., 2012. An assessment of the physical oceanographic environment on the Newfoundland and Labrador shelf in NAFO Subareas 2 and 3 during 2011. NAFO SCR Doc. 12/009. Cook, R.M., Heath, M.R., 2005. The implications of warming climate for the management of North Sea demersal fisheries. ICES J. Mar. Sci. 62, 1322–1326. Cushing, D.H., 1996. Towards a Science of Recruitment in Fish Populations. Ecology Institute, Oldendorf, Germany. Dutil, J.-D., Lambert, Y., 2000. Natural mortality from poor condition in Atlantic cod (Gadus morhua). Can. J. Fish. Aquat. Sci. 57, 826–836. Halliday, R.G., Pinhorn, A.T., 2009. The roles of fishing and environmental change in the decline of Northwest Atlantic groundfish populations in the early 1990’s. Fish. Res. 97, 163–182.
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73 Heino, M., Baulier, L., Boukal, D.S., Ernande, B., Johnston, F.D., Mollet, F.M., Pardoe, H., Therkildsen, N.O., Uusi-Heikkilä, S., Vainikka, A., Arlinghaus, R., Dankel, D.J., Dunlop, E.S., Eikeset, A.M., Enberg, K., Engelhard, G.H., Jørgensen, C., Laugen, A.T., Matsumura, S., Nusslé, S., Urbach, D., Whitlock, R., Rijnsdorp, A.D., Dieckmann, U., 2013. Can fisheries-induced evolution shift reference points for fisheries management? ICES J. Mar. Sci. 70, 707–721. Kell, L.T., Fromentin, J.-M., 2007. Evaluation of the robustness of maximum sustainable yield based management strategies to variation in carrying capacity or migration pattern of Atlantic Bluefin tuna (Thunnus thynnus). Can. J. Fish. Aquat. Sci. 64, 837–847. Kell, L.T., Pilling, G.M., O’Brien, C.M., 2005. Implications of climate change for the management of North Sea cod. ICES J. Mar. Sci. 62, 1483–1491. Kjesbu, O.S., Witthames, P.R., Solemdal, P., Greer Walker, M., 1998. Temporal variations in the fecundity of Arcto-Norwegian cod (Gadus morhua) in response to natural changes in food and temperature. J. Sea Res. 40, 303–321. Koster, F.W., Vinther, M., MacKenzie, B.R., Eero, M., Plikshs, M., 2009. Environmental effects on recruitment and implications for biological reference points of Eastern Baltic cod (Gadus morhua). J. Northwest Atl. Fish. Sci. 41, 205–220. Lambert, Y., 2011. Environmental and fishing limitations to the rebuilding of the northern Gulf of St. Lawrence cod stock (Gadus morhua). Can. J. Fish. Aquat. Sci. 68, 618–631. Morgan, M.J., Brodie, W.B., 2001. An exploration of virtual population analyses for Divisions 3LNO American plaice. NAFO SCR Doc. 01/4. Morgan, M.J., Colbourne, E.B., 1999. Variation in maturity-at-age and size in three populations of American plaice. ICES J. Mar. Sci. 56, 673–688. Morgan, M.J., Murua, H., Kraus, G., Lambert, Y., Marteinsdottir, G., Marshall, C.T., O’Brien, L., Tomkiewicz, J., 2009. The evaluation of reference points and stock productivity in the context of alternative indices of stock reproductive potential. Can. J. Fish. Aquat. Sci. 66, 404–414. Morgan, M.J., Rideout, R.M., Colbourne, E.B., 2010. Impact of environmental temperature on Atlantic cod Gadus morhua energy allocation to growth, condition and reproduction. Mar. Ecol. Prog. Ser. 404, 185–195. Myers, R.A., Barrowman, N.J., 1996. Is fish recruitment related to spawner abundance? Fish. Bull. 94, 707–724.
73
Olsen, E.M., Lilly, G.R., Heino, M., Morgan, M.J., Brattey, J., Dieckmann, U., 2005. Assessing changes in age and size at maturation in collapsing populations of Atlantic cod (Gadus morhua). Can. J. Fish. Aquat. Sci. 62, 811–823. Power, D., Morgan, J., Murphy, E.F., Brattey, J., Healey, B., 2010. An assessment of the cod stock in NAFO Divisions 3NO. NAFO SCR Doc. 10/42. Ricker, W.E., 1954. Stock and recruitment. J. Fish. Res. Board Can. 11, 559–623. Rideout, R.M., Morgan, M.J., Maddock Parsons, D., Brodie, W.B., Healey, B.P., Power, D., Dwyer, K.S., 2011. An assessment of American plaice in NAFO Div. 3LNO. NAFO SCR Doc. 11/032. Schaefer, M.B., 1954. Some aspects of the dynamics of populations important to the management of commercial marine fisheries. Bull. Inter-Am. Trop. Tuna Comm. 1, 25–56. Shelton, P.A., Morgan, M.J., 2005. Is by-catch mortality preventing the rebuilding of cod (Gadus morhua) and American plaice (Hippoglossoides platessoides) stocks on the Grand Bank? J. Northwest Atl. Fish. Sci. 36, 1–17. Shelton, P.A., Lilly, G.R., Colbourne, E., 1999. Patterns in the annual weight increment for Div. 2J+3KL cod and possible prediction for stock projection. J. Northwest Atl. Fish. Sci. 25, 151–159. Shelton, P.A., Sinclair, A.F., Chouinard, G.A., Mohn, R., Duplisea, D.E., 2006. Fishing under low productivity conditions is further delaying recovery of Northwest Atlantic cod (Gadus morhua). Can. J. Fish. Aquat. Sci. 63, 235–238. Shepherd, J., Pope, J.G., Cousens, R.D., 1984. Variations in fish stocks and hypotheses concerning their links with climate. Rapp. P.-V. Reun. Cons. Int. Explor. Mer. 185, 255–267. Sinclair, A.F., 2001. Natural mortality of cod (Gadus morhua) in the southern Gulf of St Lawrence. ICES J. Mar. Sci. 58, 1–10. Stige, L.Chr., Ottersen, G., Brander, K., Chan, K.-S., Stenseth, N.Chr., 2006. Cod and climate: effect of the North Atlantic Oscillation on recruitment in the North Atlantic. Mar. Ecol. Prog. Ser. 325, 227–241. Swain, D.P., Chouinard, G.A., 2008. Predicted extirpation of the dominant demersal fish in a large marine ecosystem: Atlantic cod (Gadus morhua) in the southern Gulf of St. Lawrence. Can. J. Fish. Aquat. Sci. 65, 2315–2319. Wayte, S.E., 2013. Management implications of including a climate-induced recruitment shift in the stock assessment of jackass morwong (Nemadactylus macropterus) in south-eastern Australia. Fish. Res. 142, 47–55.
Contents lists available at ScienceDirect
Fisheries Research journal homepage: www.elsevier.com/locate/fishres
Varying components of productivity and their impact on fishing mortality reference points for Grand Bank Atlantic cod and American plaice M.J. Morgan ∗ , P.A. Shelton, R.M. Rideout Fisheries and Oceans Canada, PO Box 5667, St. John’s, NL A1C 5X1, Canada
a r t i c l e
i n f o
Article history: Received 6 August 2013 Received in revised form 12 February 2014 Accepted 17 February 2014 Handling Editor George A. Rose Available online 20 March 2014 Keywords: Productivity Reference points Recruitment Maturity Weight
a b s t r a c t Population productivity is determined in large part by growth, maturity and recruitment. This determines the level of fishing that the population can sustain without declining. We examined how these components of productivity have varied during warm and cold periods for two Grand Bank fish populations, and how this variation affects fishing mortality reference points. Productivity of both Div. 3NO Atlantic cod and Div. 3LNO American plaice has varied considerably over time. Projections of population size under different conditions showed that the level of recruits per spawner played a major role in determining the level of fishing mortality that did not result in population decline. For plaice there was also a substantial impact of varying proportion mature at age. The impact of factors other than temperature on recruits per spawner and maturity meant that the association of productivity with temperature was not consistent. When productivity was at its lowest, the level of fishing mortality that could be sustained without causing rapid population decline was very low. The results of this study clearly demonstrate that the impacts of changing productivity can be rapid and very large and if fishing mortality reference points are not adjusted accordingly the results can be catastrophic. It may be helpful to develop fishing mortality reference points that incorporate all components of productivity and are updated using only recent data. Such reference points could be used in combination with buffers around a biomass limit reference point (Blim ) and/or low risk tolerances for declining below Blim to account for the uncertainties that would remain even in frequently updated reference points. Crown Copyright © 2014 Published by Elsevier B.V. All rights reserved.
1. Introduction The productivity of a fish population is the capacity of that population to produce biomass, and is a result of increases due to growth and reproduction, countered by declines due to mortality. The productivity of a fish population determines the level of fishing that can be sustained without a decline in population size. Major components of productivity are recruitment, weight at age, maturity at age and mortality. These components vary over time and therefore, so too does the overall productivity of the population. Recruitment is linked to the size (i.e. biomass or abundance) and composition of the spawning stock and is also affected by environmental conditions (Myers and Barrowman, 1996; Shepherd et al., 1984; Cushing, 1996). Growth varies with temperature and prey availability (Brett, 1979; Shelton et al., 1999; Bjornsson et al., 2001). Maturity has been shown to vary with temperature and mortality and is related to
∗ Corresponding author. Tel.: +1 709 772 2261; fax: +1 709 772 4188. E-mail address: [email protected] (M.J. Morgan). http://dx.doi.org/10.1016/j.fishres.2014.02.019 0165-7836/Crown Copyright © 2014 Published by Elsevier B.V. All rights reserved.
growth (Kjesbu et al., 1998; Morgan and Colbourne, 1999; Olsen et al., 2005). Estimates of natural mortality are few but it is not constant over time and can be influenced by environmental conditions and changes in predation level (Dutil and Lambert, 2000; Sinclair, 2001; Morgan and Brodie, 2001; Chouinard et al., 2005). Fishing mortality (F) reference points play an important role in the management of many fisheries and are determined by population productivity. As the components of productivity vary, so too do the reference points derived from them (A’mar et al., 2009; Brooks, 2013; Heino et al., 2013; Wayte, 2013). These reference points can be targets to be achieved or limits that should be avoided (Caddy and McGarvey, 1996). Three of the most commonly used fishing mortality reference points are FMSY , F0.1 and F40%SPR . F0.1 is generally used as a target, F40%SPR is a limit, while FMSY is used as both, depending on the fisheries management jurisdiction. FMSY is the fishing mortality giving the maximum sustainable yield from a population (Schaefer, 1954). Fishing at levels above FMSY will result in a population size that is lower than BMSY , the biomass giving maximum sustainable yield (MSY). F0.1 is determined as the F where the slope of the yield per recruit (YPR) curve is 10% of the slope at the
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73
65
Morgan et al., 2010). There is also evidence that recruitment rate is related to ocean climate in these two populations (Stige et al., 2006; Halliday and Pinhorn, 2009). The extended warm and cold periods, coupled with the fact that components of productivity are influenced by temperature, suggest the potential for long term changes in productivity of these two Grand Bank fish populations. In this study we examine time series of abundance and biological data on these stocks for indications of changes in weight at age, maturity at age and recruitment rate, during warm and cold periods. We estimate FMSY , F0.1 , and F40%SPR using the productivity from these periods and determine which components of productivity (recruitment, weight at age, maturity at age) are responsible for any differences. We also explore the implications of failing to recognize changes in stock productivity by examining the consequences of fishing the populations when they are at one level of productivity at reference points derived from a different level of productivity. Fig. 1. Rogers North Atlantic Oscillation (NAO) anomaly (millibars of pressure) relative to the 1981–2010 average. The warm and cold periods examined are indicated. Warm 1 is 1962–1966, Cold 1 is 1972–1976, Cold 2 is 1990–1994 and Warm 2 is 2001–2004.
origin. F40%SPR is the F that reduces spawner per recruit (SPR) to 40% of the unfished value. Changes in population productivity will influence the estimation of these three reference points differentially because they incorporate different components of productivity in their calculation. FMSY is determined by all components of productivity and will therefore be affected by changes in any component of the productivity of the population (Morgan et al., 2009; Brooks, 2013). Reference points derived from YPR or SPR do not include the recruitment component of productivity. YPR based reference points will respond only to variation in growth while SPR reference points will respond to changes in growth and/or maturity. If changes in productivity are short term, then the impact of assuming constant conditions is likely to be small. However, extended periods of low productivity could result in population decline if F is set at a reference point level based on an assumption of constant, more productive, conditions (Cook and Heath, 2005; Koster et al., 2009). Cold or warm ocean temperatures can persist for several to many years (Colbourne et al., 2012). To understand how productivity will differ and which reference points will respond to this variation, it is important to understand how the components of productivity may change with varying environmental conditions (Kell et al., 2005). Major variation in ocean temperature has been observed on the Grand Bank in the Northwest Atlantic off Newfoundland Canada over the last 50 years. There was an extended period of warm temperature in the 1960s, while the early 1970s and early 1990s were generally cold, with recent years once again characterized by warm ocean temperature (Colbourne, 2004; Colbourne et al., 2012). Thus there has been the potential for lengthy periods of high and low productivity of fish stocks in the area, driven by the direct and indirect effects of temperature. Northwest Atlantic Fisheries Organization (NAFO) Division 3NO Atlantic cod (Gadus morhua) and Div. 3LNO American plaice (Hippoglossoides platessoides) are two important groundfish populations on the Grand Bank. Both experienced severe population decline due to over fishing and spawning stock biomass (SSB) for both stocks is currently at a low level (Power et al., 2010; Rideout et al., 2011). They have also both exhibited major changes in maturity at age that may have been related, in part to temperature. Growth has also been shown to be related to temperature, although weakly, for Div. 3NO cod and for Div. 3LNO plaice, cohorts that grew faster and occupied warmer temperatures were those that matured earlier (Morgan and Colbourne, 1999;
2. Materials and methods 2.1. Productivity The Grand Bank has experienced warm and cold periods over the last half century. The 1960s and 2000s were generally warm while the early 1970s and the early 1990s were cold years (Colbourne, 2004; Colbourne et al., 2012). We chose years within these time periods based on anomalies in the North Atlantic Oscillation (NAO), an index of ocean climate on the Grand Bank (Colbourne, 2004), as a basis for comparisons of productivity. The years 1962–1966 (Warm 1), 1972–1976 (Cold 1), 1990–1994 (Cold 2) and 2001–2004 (Warm 2) were chosen to represent the 1960s, 1970s, 1990s and 2000s respectively (Fig. 1). These were years with consistently high (cold) or low (warm) NAO. For the first three periods five years were chosen in order to have the same number of years in each period. The 2000s were more variable and so there was only a 4-year period with a consistent NAO during which population data were available. All analyses and data are based on information from recent assessments of Div. 3NO cod (Power et al., 2010) and Div. 3LNO American plaice (Rideout et al., 2011). Model estimates of population numbers at age in each year were extracted for each stock. Stock weights at age (for calculating SSB) and catch weights at age (for calculating catch) were based on commercial sampling conducted for these assessments. Maturities at age in these assessments were modeled by cohort based on research vessel data. SSB was calculated as the sum of the product of model estimates of numbers at age (both sexes combined), model estimates of female proportion mature at age, and beginning of year stock weights at age (both sexes combined). Recruitment was calculated from the assessments as number at age 1 such that r = Nagerec × eM×a
(1)
where Nagerec is the number of recruits and a is difference between the age at recruitment estimated in the assessment and age 1. For 3NO cod a is 1 and 4 for 3LNO American plaice. M is natural mortality and was set equal to 0.2 for 3NO cod for all years. For 3LNO American plaice M was 0.2 for all years except from 1989 to 1996 when it was 0.53, as is the case in the assessment, to include an increase in natural and other unaccounted for mortality over that period (Morgan and Brodie, 2001). Weight at age, maturity at age, recruits per spawner (RPS), and spawner per recruit (SPR) at F = 0 were examined to determine how they differed among Warm 1, Cold 1, Cold 2, Warm 2 and average conditions. For each period, average weights, maturities and RPS were calculated. Variation in recruitment was expressed as RPS as
66
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73
this metric accounts for some of the influence of SSB on recruitment without assuming a specific form of a stock recruit relationship, other than a constant recruitment rate during each period. SPR at F = 0 was calculated as SPR =
A
Na Wa Pa
(2)
a=1
where Na is the number at age a starting with one recruit, Wa and Pa are the weight at age and proportion mature at age, respectively, averaged over each time period. A is the terminal age in the assessment. Na is incremented as: Na+1 = Na × e−M
(3)
where M is the annual instantaneous rate of natural mortality. The percent growth rate in SSB at F = 0 (G0 ) reflects the combined impact of RPS and SPR on the productivity of the stock at F = 0 (Shelton et al., 2006). This is a metric similar to the intrinsic rate of increase but expressed in terms of SSB. It allows easy comparison of potential productivity across periods. It was calculated as G0 = average
SSB − SSB t t−1 SSBt−1
× 100
(4)
where N is incremented as in Eq. (3) and SSB is calculated using the average stock weight and maturity at age from each period. G0 is measured when the stock is below the break point of the hockey stick stock recruit curve and has a stable age composition. Recruitment was calculated as:
R=
˛ × SSB ˛×ˇ
if SSB < ˇ
(5)
if SSB≥ˇ
where R is recruitment, ˛ is RPS and ˇ is the maximum SSB estimated in the assessment for each population. 2.2. Fishing mortality reference points Projections of stock size were carried out to equilibrium over a range of F values to determine the F giving maximum sustainable yield (FMSY ). This reference point is affected by changing recruitment rate, weight and maturity at age. Population numbers at age were projected as: Na+1,y+1 = Na,y × e−(M+Fa,y )
(6)
where Na,y is the number alive at age a and the beginning of year y and Fa,y is the fishing mortality on age a in year y, obtained from: Fa,y = Ka Fy
N15,y+1 = N14,y e−(M+F14,y ) + N15,y e−(M+F15,y )
(8)
Yield in weight was calculated in each year as: n
3. Results
(7)
where Fy is the fully recruited fishing mortality in year y and Ka is the selectivity or partial recruitment to the fishery at age a (that is the proportion of Fy for that age). For 3NO cod, a = 1–12 with no plus group. For 3LNO plaice, a = 1–15, where age 15 is a plus group, which was updated in the simulation by applying:
Yield =
As in the calculation of G0 , recruitment was assumed to come from a hockey stick stock recruit relationship (Eq. (5)). This assumes the same break point for all periods. Density dependence in recruitment is assumed to occur at the same SSB for all RPS as a representation of what recruitment might be expected to occur under the prevailing conditions of each period. This allows results to be interpreted with respect to RPS without the confounding effect of different break points. In estimation of FMSY population size exceeded the break point by the end of the projection period for all time periods considered. All projections applied the same partial recruitment vector calculated from the matrix of fishing mortality at age for each stock. Natural mortality was (0.2) for all cases except that SPR at F = 0, G0 and the reference points were also calculated using an M = 0.53 for American plaice for Cold 2. This is done in the assessment to deal with unaccounted for mortality, some of which is thought to have been natural mortality (Morgan and Brodie, 2001). Weight at age and proportion mature at age averaged over each time period (Warm 1, Cold 1, Cold 2, Warm 2) and over the whole time series were used in analyses together with the average RPS calculated over the corresponding period. The YPR reference point F0.1 was calculated for each period using Eqs. (6)–(10) above, starting with 1 recruit and applying a range of F values, to produce a curve of yield across F. For American plaice the population was extended to age 30 to approximate the plus group in the assessment. F0.1 is determined as the F where the slope of the yield per recruit curve is 10% of the slope at the origin. This reference point is affected by variation in weight at age but not by RPS or maturity at age. Average weight at age in each time period and the overall average were used to examine how F0.1 varied across period. The F that gave 40% of the SPR at F = 0 (F40%SPR ) was determined by solving for the value of F giving this amount of SPR depletion per recruit using Eqs. (2), (6) and (7) above. This reference point is affected by changing weight and maturity at age but not recruitment rate. The impact of fishing at these reference points under varying productivity was examined. Populations were projected to equilibrium at FMSY determined for Warm 1, Cold 1, Cold 2 and Warm 2 and the average weights, maturities and RPS from the corresponding period. The response of the population was compared over 20 year projections from equilibrium at the three F reference points (FMSY , F0.1 , F40%SPR ) derived from the period being examined and from average conditions.
Ca,y × Cwa,y
(9)
a=1
where Cwa,y is the average catch weight at age a in year y and Ca,y is the catch number at age a in year y calculated as:
Ca,y = Na,y ×
1−e
−(M+Fa,y )
×
Fa,y M + Fa,y
(10)
3.1. Productivity Proportion mature at age for both stocks was lowest during Warm 1 and Cold 1, and highest during the most recent period, Warm 2. The variation was much greater for plaice than for cod (Fig. 2). Weight at age for older fish was markedly reduced during Warm 1 compared to the other periods for both stocks. For cod, ages 7–12 had lower weight at age in Warm 1, while for plaice it was ages 11–15 that had lower weight. Both 3NO cod and 3LNO plaice have exhibited substantial variation in RPS (Fig. 3). The timing of this variation and the association between RPS and SSB was not the same for the two populations although both stocks had extended periods when RPS was below average. There was a 24-year period of low RPS from 1981 to 2004 for cod and 11- and 10-year periods of low RPS for plaice from 1961 to 1971 and 1979 to 1988 respectively. During Warm 1 both SSB and RPS were high for cod while SSB was high and RPS low for plaice (Table 1). During Cold 1, RPS started below average for cod
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73
67
Fig. 2. Average estimated proportion mature at age and weight at age for cod and American plaice from each of four different time periods and for the entire time series.
and then increased rapidly as SSB declined to very low levels, with a resulting average RPS for the period that was above the long term average. For plaice, RPS was above average during Cold 1, while SSB was about average, although as for cod, RPS increased over the time period while SSB declined. During Cold 2, SSB and RPS were low for cod with low SSB and high RPS for plaice. In recent years when the NAO was again low and conditions warm (Warm 2), there has been
mainly below average RPS at low SSB for both American plaice and cod. Over all average RPS was much higher for plaice than for cod. Average SPR for plaice was an order of magnitude lower than for cod (Fig. 3) as a result of lower weight and later maturity at age for plaice. For both populations there has been a general increase in SPR since the beginning of the series but this increase is much greater for plaice. This results in higher SPR in Cold 2 and Warm 2,
Table 1 Average North Atlantic Oscillation anomaly (NAO, millibars of pressure), spawning stock biomass (SSB, ‘000 t), recruits per spawner (RPS), spawner per recruit (SPR), and fishing mortality for the selected periods used in the estimation of reference points. The estimated percentage SSB (G0 ) at F = 0 is also shown. For American plaice the two numbers for SPR and G0 during Cold 2 are for M = 0.2 and M = 0.53 (in brackets). All values are averages except for G0 which was calculated using the average values of RPS, maturity and weight at age for the different periods. NAO Cod Warm 1 Cold 1 Cold 2 Warm 2 Average
−13.6 4.84 5.3 −5.6 −2.2
American plaice Warm 1 Cold 1 Cold 2 Warm 2 Average
−13.6 4.84 5.3 −5.6 −2.2
SSB
RPS
SPR
G0
F
92.3 44.4 18.6 7.6 43.2
2.74 1.42 0.17 0.49 1.17
4.52 5.84 6.45 6.64 5.95
32.97 28.5 1.08 14.49 24.14
0.280 0.814 0.756 0.356 0.387
170 81.5 34.4 21.03 94.2
2.43 5.8 8.50 3.34 4.43
0.48 0.76 0.87 (0.02) 0.93 0.73
4.16 11.58 18.99 (−12) 11.81 11.67
0.122 0.407 0.752 0.269 0.328
68
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73
Fig. 3. Spawning stock biomass (‘000 t), recruits per spawner, and spawner per recruit for Div. 3NO cod and Div. 3LNO American plaice. The dashed horizontal lines give the biomass limit reference point established for each stock in the spawning stock biomass panels and average spawner per recruit and recruits per spawner over the whole time series in the other panels. The gray areas indicate the warm periods and the gray with stipple the cold periods used in this study. For American plaice the solid line in the spawner per recruit graph indicates the result using an M of 0.20 during 1989–1996, while the dotted line uses an M of 0.53 during that time.
the most recent periods. The SPR for plaice during the Cold 2 is almost zero if an M of 0.53 is assumed, as it is for that time period in the assessment. The differing RPS resulted in very different levels of recruitment when applied as ˛ in a hockey stick stock recruit model (Fig. 4). The much larger RPS for plaice results in much greater recruitment than for cod. The highest levels of recruitment for plaice were during Cold 2 while for cod it is during Warm 1. Maximum recruitment
for cod in Warm 1 was 16 times that of the Cold 2 period. For plaice, maximum recruitment during the Cold 2 period was 3.5 times that of Warm 1. For both cod and plaice the second highest level of recruitment was during Cold 1. The potential growth in the SSB (G0 ) was very low for cod during Cold 2 (Table 1), at just over 1%. In comparison during Warm 1, G0 for cod was 33%. During Cold 1 G0 was just above average for cod while during Warm 2 it was below average. The highest G0
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73
69
Table 2 Estimated FMSY , F0.1 and F40%SPR for different time periods for Div. 3NO cod and Div. 3LNO American plaice. For American plaice the two numbers in Cold 2 are for M = 0.2 and M = 0.53 (in brackets). Period
Warm 1 Cold 1 Cold 2 Warm 2 Average
Cod
American plaice
FMSY
F0.1
F40%SPR
FMSY
F0.1
F40%SPR
0.33 0.29 0.02 0.22 0.30
0.21 0.19 0.18 0.20 0.20
0.12 0.12 0.12 0.12 0.12
0.08 0.29 0.33 (0) 0.48 0.37
0.19 0.17 0.16 (0.88) 0.16 0.17
0.05 0.05 0.06 (0.12) 0.07 0.06
Cold 1 and Warm 2 were periods of intermediate productivity for both stocks. 3.2. Reference points The estimated FMSY differed substantially for both plaice and cod across periods, by 80 and 95% respectively (Table 2). The variation in F0.1 and F40%SPR was much less. F40%SPR was much lower for plaice than for cod but varied more (28% vs 5%) reflecting large changes in maturity at age in plaice. During periods of low RPS, FMSY was very low for both cod and plaice (Cold 2 and Warm 1 respectively). Both F0.1 and F40%SPR were much higher than FMSY during the Cold 2 period for cod. But for plaice, while F0.1 was much higher than FMSY during the Warm 1 period, F40%SPR was similar to FMSY . For cod, fishing under the low productivity conditions of Cold 2 with reference points derived under conditions averaged over the whole time series has a very detrimental effect on SSB, causing a steep decline (Fig. 5). Both F0.1 and F40%SPR derived from conditions during Cold 2 also cause a steep decline in SSB under these low RPS conditions. During Warm 2 conditions, the FMSY estimated using average conditions also causes the SSB to decline over the projection period. For plaice, Warm 1 was the low productivity period. Average FMSY and F0.1 both result in population decline under the conditions of the Warm 1 period, while average F40%SPR allows the SSB to grow (Fig. 6). The F0.1 estimated using the weight at age from Warm 1 also results in a decrease in SSB. 4. Discussion
Fig. 4. Stock recruit relationships for Div. 3NO cod and Div. 3LNO American plaice used in the study. The relationship is based on the assumption of a segmented regression with a slope equal to the average recruits per spawner for each period and a plateau at that recruits per spawner times the maximum spawning stock biomass estimated in the assessment for each population.
for plaice was during Cold 2, G0 during Warm 1 being well below average. Cold 1 and Warm 2 had near average G0 for plaice. Overall, Warm 1 appears to have been a high productivity period for cod with high RPS and very high G0 , although SPR was below average because of the lower weights and maturity at age. The cold 1990s were a period of very low productivity for cod despite a higher SPR. For plaice, Warm 1 was a low productivity period, with low RPS, SPR and consequently low G0 . The Cold 2 period was one of high productivity for plaice. However, if natural mortality did increase for plaice during this time then it would have been a period of very low productivity for this population as well (Table 1).
The analyses presented here for Grand Bank cod and American plaice populations demonstrated substantial temporal variability in potential growth in SSB (G0 ), used here as a general indicator of population productivity. The variation in productivity stemmed from changes in all of the components of productivity examined: weight at age, maturity at age and recruitment rate. For both species, variation in recruitment rate was substantial and was the main contributor to changes in G0 , while changes in weight at age had little impact for either species. A larger impact on stock productivity of variation in recruitment than in growth was also reported by Kell et al. (2005). This is not unexpected given the large variation in recruitment experienced by most marine fish stocks. Both species exhibited changes in maturity at age but the variation was greater for plaice and therefore made a more substantial contribution to variation in G0 for that species than for cod. Average RPS is four times higher for plaice than cod, whereas average SPR is an order of magnitude higher for cod than plaice. The causes of the larger RPS for plaice are unclear but their lower SPR is the result of smaller weight at age and later maturation. In combination, RPS and SPR result in an average G0 value for cod that is twice as high as plaice, indicating that higher recovery rates could be expected from the depleted cod stock compared to the depleted plaice stock. The fact that neither stock has recovered above their respective SSB limit reference points since moratoria were applied to directed fishing in the early 1990s indicates that a large proportion of the
70
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73
Fig. 5. Spawning stock biomass (SSB ‘000 t) resulting from projections of population size of Div. 3NO cod. Projections start with equilibrium population size calculated using average weight at age, maturity at age and RPS for the period and FMSY specific to each period. Panels on the left use FMSY , F0.1 and F40%SPR derived under average conditions for the entire time series while those on the right use those reference points derived from the average conditions for the particular period.
surplus production under prevailing productivity conditions has been removed through bycatch mortality (Shelton and Morgan, 2005). Our results support previous work indicating that environmental conditions such as temperature can have a major impact on all of the components of productivity (Brett, 1979; Shepherd et al., 1984; Cushing, 1996; Kjesbu et al., 1998; Morgan and Colbourne,
1999). It is important to note, however, that temperature is not the only factor potentially influencing productivity. For example for cod, highest RPS was found in the warmest period, Warm 1 and the lowest RPS in the coldest period, Cold 2, with the opposite being the case for plaice. However, a fundamental aspect of population dynamics is the compensatory response of RPS to changes in population size (Ricker, 1954) and any influence of temperature would
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73
71
Fig. 6. Spawning stock biomass (SSB ‘000 t) resulting from projections of population size of Div. 3LNO American plaice. Projections start with equilibrium population size calculated using average weight at age, maturity at age and RPS for the period and FMSY specific to each period. Panels on the left use FMSY , F0.1 and F40%SPR derived under average conditions for the entire time series while those on the right use those reference points derived from the average conditions for the particular period.
interact with this response. Some evidence of this was seen for both species on the Grand Bank, with cod RPS increasing in response to declining SSB during Cold 1 and plaice RPS increasing in response to a decrease in SSB during Cold 2. For cod this led to an above average RPS for Cold 1 while RPS was well below average during Cold 2. Warm temperatures can lead to earlier age at maturity (Alm, 1957; Kjesbu et al., 1998; Morgan and Colbourne, 1999) but the dominant trend for both species in this study was a temporal one with fish
maturing younger in more recent years regardless of temperature. The change in maturity in both cod and plaice may be at least partly driven by fishing mortality (Barot et al., 2005; Olsen et al., 2005). This shift can lead to greater productivity for a population adapted to fishing (Heino et al., 2013). The patterns in variation in RPS and maturity meant that periods of high and low productivity were not the same for the two species. Warm 1 was a period of high productivity for cod with high RPS
72
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73
leading to a very high G0 despite lower weight and maturity at age. This same period was one of very low productivity for plaice, arising from low RPS, weights and maturity at age. Cold 2 was a period of low productivity for cod with very low RPS leading to limited potential for population growth. For plaice on the other hand, Cold 2 was a period of high RPS and high maturity at age leading to high productivity (under the assumption of M = 0.2). The impact of factors other than temperature on RPS and maturity meant that the association of productivity with temperature period was not consistent. For example, while Warm 1 had a very low G0 for plaice, the shift to maturation at an earlier age meant that Warm 2 had average G0 despite low RPS. Previous studies have examined the impact of changes in temperature on productivity and reference points in the context of climate change (Cook and Heath, 2005; Kell et al., 2005). The results here show that the impact of drivers other than temperature may make it difficult to predict how productivity will change as temperature changes. The variation in productivity had a large impact on the level of fishing mortality based reference points, particularly FMSY . When productivity was at its lowest, FMSY was very low. Low or even negative (i.e. population decline) productivity has been found in other stocks, sometimes lasting for extended periods of time (Shelton et al., 2006; Swain and Chouinard, 2008; Lambert, 2011). That different assumptions about productivity can lead to different estimates of fishing mortality reference points has been demonstrated previously (Beddington and May, 1977; Morgan et al., 2009; Brooks, 2013; Cervino et al., 2013). The results of this study clearly demonstrate that the impacts of changing productivity can be rapid and very large, and if fishing mortality reference points are not adjusted accordingly the results can be catastrophic. The variation in FMSY under changing productivity conditions was greater than that of F0.1 and F40%SPR . This is a result of the importance of changing RPS in both populations. The relative importance of the components of productivity leading to the variation in FMSY was not the same in the two populations. In cod, the main source of variation was change in RPS, while for plaice there was also a substantial change in SPR caused by varying proportion mature at age, which caused greater variation in F40%SPR for plaice compared to cod. Changes in weight at age seemed to be of less importance than changes in maturity in both stocks. Changes in weight at age occurred mainly at older ages and would have a reduced impact on stock productivity under higher levels of F. During periods of low productivity, RPS was very low for both populations. The lack of inclusion of recruitment in F0.1 and F40%SPR meant that under conditions of low RPS they could be much higher than FMSY . Cervino et al. (2013) also found a discrepancy between FMSY and yield per recruit reference points using a Ricker stock recruit curve, in that case comparing different measures of reproductive potential. Kell and Fromentin (2007) determined that F0.1 performed better than F40%SPR when carrying capacity was variable. How well a particular type of reference point performs will depend on what components of productivity are changing. Given the large impact of RPS seen here and the generally large variation in recruitment experienced by fish stocks, MSY based reference points may be more robust than those based on YPR or SPR. Fishing at reference points derived from average conditions or with YPR or SPR based reference points when RPS was low resulted in rapid population decline. The periods of low RPS observed here in both populations was long enough to result in substantial population decline and even collapse. The level of constant fishing mortality that would ensure that the populations would not decline under low productivity conditions would be so low as to result in a large loss of yield if applied to all time periods, implying that target fishing mortality should be adjusted between low and higher productivity periods. We propose that this be done by applying recent average data in the
computation of fishing mortality target and limit reference points. Estimates of weight and maturity at age and recruitment need to be compared to those being used to calculate current reference points to determine if their variation warrants updating of the reference points (Brooks, 2013). In practice, it is usually YPR and SPR type reference points that are updated using recent data, while FMSY is usually estimated based on long term data. YPR and SPR based reference points do not incorporate recruitment and our study demonstrated that RPS can be so low as to make YPR and SPR based reference points dangerously high during periods of low recruitment. It may be helpful to develop fishing mortality reference points that incorporate all components of productivity and are updated using only recent data. One possible approach is to calculate FMSY using recent average RPS and a hockey stick stock recruit relationship as done in this study. Another potential approach is to compute G0 based on recent RPS and SPR and to apportion a fraction of that to the fishery and a fraction to population growth, the relative proportion depending on whether or not the population is depleted or healthy. Such reference points could be used in combination with buffers around a biomass limit reference point (Blim ) and/or low risk tolerances for declining below Blim to account for the uncertainties that would remain even in frequently updated reference points. Acknowledgements We thank all the sea going and laboratory staff involved in the collection of these data. A. Buren, L. Brooks and an anonymous reviewer provided helpful comments on an earlier version of the MS. This study was funded by Fisheries and Oceans Canada’s Aquatic Climate Change Adaptation Services Program. References Alm, G., 1957. Connections Between Maturity, Size and Age in Fishes. Reports from the Institute of Freshwater Research, Drottningholm, vol. 40., pp. 5–145. A’mar, Z.T., Punt, A.E., Dorn, M.W., 2009. The impact of regime shifts on the performance of management strategies for the Gulf of Alaska walleye Pollock (Theragra chalcogramma) fishery. Can. J. Fish. Aquat. Sci. 66, 2222–2242. Barot, S., Heino, M., Morgan, M.J., Dieckman, U., 2005. Maturation of Newfoundland American plaice (Hippoglossoides platessoides): long-term trends in maturation reaction norms despite low fishing mortality? ICES J. Mar. Sci. 62, 56–64. Beddington, J.R., May, R.M., 1977. Harvesting natural populations in a randomly fluctuating environment. Science 197, 463–465. Bjornsson, B., Steinarsson, A., Oddgeirsson, M., 2001. Optimal temperature for growth and feed conversion of immature cod (Gadus morhua L.). ICES J. Mar. Sci. 58, 29–38. Brett, J.R., 1979. Environmental factors and growth. In: Hoar, W.S., Randall, D.J., Brett, J.R. (Eds.), Fish Physiology, vol. III. Academic Press, New York, pp. 599–675. Brooks, E.N., 2013. Effects of variable reproductive potential on reference points for fisheries management. Fish. Res. 138, 152–158. Caddy, J.F., McGarvey, R., 1996. Targets or limits for management of fisheries? North Am. J. Fish. Manage. 16, 479–487. Cervino, S., Dominguez-Petit, R., Jardim, E., Mehault, S., Pineiro, C., Saborido-Rey, F., 2013. Impact of egg production and stock structure on MSY reference points and its management implications for southern hake (Merluccius merluccius). Fish. Res. 138, 168–178. Chouinard, G.A., Swain, D.P., Hammill, M.O., Poirier, G.A., 2005. Covariation between grey seal (Halichoerus grypus) abundance and natural mortality of cod (Gadus morhua) in the southern Gulf of St. Lawrence. Can. J. Fish. Aquat. Sci. 62, 1991–2000. Colbourne, E.B., 2004. Decadal changes in the ocean climate in Newfoundland and Labrador waters from the 1950 to the 1990. J. Northwest Atl. Fish. Sci. 34, 41–59. Colbourne, E.B., Craig, J., Fitzpatrick, C., Senciall, D., Stead, P., Bailey, W., 2012. An assessment of the physical oceanographic environment on the Newfoundland and Labrador shelf in NAFO Subareas 2 and 3 during 2011. NAFO SCR Doc. 12/009. Cook, R.M., Heath, M.R., 2005. The implications of warming climate for the management of North Sea demersal fisheries. ICES J. Mar. Sci. 62, 1322–1326. Cushing, D.H., 1996. Towards a Science of Recruitment in Fish Populations. Ecology Institute, Oldendorf, Germany. Dutil, J.-D., Lambert, Y., 2000. Natural mortality from poor condition in Atlantic cod (Gadus morhua). Can. J. Fish. Aquat. Sci. 57, 826–836. Halliday, R.G., Pinhorn, A.T., 2009. The roles of fishing and environmental change in the decline of Northwest Atlantic groundfish populations in the early 1990’s. Fish. Res. 97, 163–182.
M.J. Morgan et al. / Fisheries Research 155 (2014) 64–73 Heino, M., Baulier, L., Boukal, D.S., Ernande, B., Johnston, F.D., Mollet, F.M., Pardoe, H., Therkildsen, N.O., Uusi-Heikkilä, S., Vainikka, A., Arlinghaus, R., Dankel, D.J., Dunlop, E.S., Eikeset, A.M., Enberg, K., Engelhard, G.H., Jørgensen, C., Laugen, A.T., Matsumura, S., Nusslé, S., Urbach, D., Whitlock, R., Rijnsdorp, A.D., Dieckmann, U., 2013. Can fisheries-induced evolution shift reference points for fisheries management? ICES J. Mar. Sci. 70, 707–721. Kell, L.T., Fromentin, J.-M., 2007. Evaluation of the robustness of maximum sustainable yield based management strategies to variation in carrying capacity or migration pattern of Atlantic Bluefin tuna (Thunnus thynnus). Can. J. Fish. Aquat. Sci. 64, 837–847. Kell, L.T., Pilling, G.M., O’Brien, C.M., 2005. Implications of climate change for the management of North Sea cod. ICES J. Mar. Sci. 62, 1483–1491. Kjesbu, O.S., Witthames, P.R., Solemdal, P., Greer Walker, M., 1998. Temporal variations in the fecundity of Arcto-Norwegian cod (Gadus morhua) in response to natural changes in food and temperature. J. Sea Res. 40, 303–321. Koster, F.W., Vinther, M., MacKenzie, B.R., Eero, M., Plikshs, M., 2009. Environmental effects on recruitment and implications for biological reference points of Eastern Baltic cod (Gadus morhua). J. Northwest Atl. Fish. Sci. 41, 205–220. Lambert, Y., 2011. Environmental and fishing limitations to the rebuilding of the northern Gulf of St. Lawrence cod stock (Gadus morhua). Can. J. Fish. Aquat. Sci. 68, 618–631. Morgan, M.J., Brodie, W.B., 2001. An exploration of virtual population analyses for Divisions 3LNO American plaice. NAFO SCR Doc. 01/4. Morgan, M.J., Colbourne, E.B., 1999. Variation in maturity-at-age and size in three populations of American plaice. ICES J. Mar. Sci. 56, 673–688. Morgan, M.J., Murua, H., Kraus, G., Lambert, Y., Marteinsdottir, G., Marshall, C.T., O’Brien, L., Tomkiewicz, J., 2009. The evaluation of reference points and stock productivity in the context of alternative indices of stock reproductive potential. Can. J. Fish. Aquat. Sci. 66, 404–414. Morgan, M.J., Rideout, R.M., Colbourne, E.B., 2010. Impact of environmental temperature on Atlantic cod Gadus morhua energy allocation to growth, condition and reproduction. Mar. Ecol. Prog. Ser. 404, 185–195. Myers, R.A., Barrowman, N.J., 1996. Is fish recruitment related to spawner abundance? Fish. Bull. 94, 707–724.
73
Olsen, E.M., Lilly, G.R., Heino, M., Morgan, M.J., Brattey, J., Dieckmann, U., 2005. Assessing changes in age and size at maturation in collapsing populations of Atlantic cod (Gadus morhua). Can. J. Fish. Aquat. Sci. 62, 811–823. Power, D., Morgan, J., Murphy, E.F., Brattey, J., Healey, B., 2010. An assessment of the cod stock in NAFO Divisions 3NO. NAFO SCR Doc. 10/42. Ricker, W.E., 1954. Stock and recruitment. J. Fish. Res. Board Can. 11, 559–623. Rideout, R.M., Morgan, M.J., Maddock Parsons, D., Brodie, W.B., Healey, B.P., Power, D., Dwyer, K.S., 2011. An assessment of American plaice in NAFO Div. 3LNO. NAFO SCR Doc. 11/032. Schaefer, M.B., 1954. Some aspects of the dynamics of populations important to the management of commercial marine fisheries. Bull. Inter-Am. Trop. Tuna Comm. 1, 25–56. Shelton, P.A., Morgan, M.J., 2005. Is by-catch mortality preventing the rebuilding of cod (Gadus morhua) and American plaice (Hippoglossoides platessoides) stocks on the Grand Bank? J. Northwest Atl. Fish. Sci. 36, 1–17. Shelton, P.A., Lilly, G.R., Colbourne, E., 1999. Patterns in the annual weight increment for Div. 2J+3KL cod and possible prediction for stock projection. J. Northwest Atl. Fish. Sci. 25, 151–159. Shelton, P.A., Sinclair, A.F., Chouinard, G.A., Mohn, R., Duplisea, D.E., 2006. Fishing under low productivity conditions is further delaying recovery of Northwest Atlantic cod (Gadus morhua). Can. J. Fish. Aquat. Sci. 63, 235–238. Shepherd, J., Pope, J.G., Cousens, R.D., 1984. Variations in fish stocks and hypotheses concerning their links with climate. Rapp. P.-V. Reun. Cons. Int. Explor. Mer. 185, 255–267. Sinclair, A.F., 2001. Natural mortality of cod (Gadus morhua) in the southern Gulf of St Lawrence. ICES J. Mar. Sci. 58, 1–10. Stige, L.Chr., Ottersen, G., Brander, K., Chan, K.-S., Stenseth, N.Chr., 2006. Cod and climate: effect of the North Atlantic Oscillation on recruitment in the North Atlantic. Mar. Ecol. Prog. Ser. 325, 227–241. Swain, D.P., Chouinard, G.A., 2008. Predicted extirpation of the dominant demersal fish in a large marine ecosystem: Atlantic cod (Gadus morhua) in the southern Gulf of St. Lawrence. Can. J. Fish. Aquat. Sci. 65, 2315–2319. Wayte, S.E., 2013. Management implications of including a climate-induced recruitment shift in the stock assessment of jackass morwong (Nemadactylus macropterus) in south-eastern Australia. Fish. Res. 142, 47–55.