The 17th Russian Norwegian Symposium Long term sustainable management of living marine resources in the Northern Seas Bergen, March 2016 Evaluating a harvest control rule of the NEA cod considering capelin Anatoly Filin
Starting points Cod and capelin are key species in the Barents Sea that are connected by trophic interactions. Evaluation of cod HCRs that take capelin into account cannot be fully considered if a single species model is used. This requires multispecies models. The capelin impact on cod stock dynamics is considered as a key element in all multispecies models for the Barents Sea.
Why is inclusion of capelin in cod HCR relevant? Capelin is a main prey item for cod in the Barents Sea. Cannibalism in cod stock, as well as growth and maturation rate of cod depends on capelin stock size. Stocks of cod and capelin both have strong year to year variations, which influences their interaction. Impact of cod as a predator on other commercial species increase when capelin stock is low.
Objectives Using the multispecies model: to perform simulations of long-term dynamics of cod stock under the HCRs including and excluding capelin; to evaluate the long-term effect of capelin-dependent HCRs on cod stock parameters and TAC; to evaluate effect of capelin-dependent HCRs on interannual stability of TAC.
Multispecies model STOCOBAR It describes stock dynamics and harvesting of cod in the Barents Sea, taking trophic interactions and environmental influence into account. Age structured, single area and single fleet model, one year time step. Model parameters were estimated using data 1984 2006.
Block-schema of the model Capelin stock Cod stock HCR (management scenario) Cod recruitment, growth, feeding, maturation and natural mortality Temperature scenarios
STOCOBAR description Филин А.А. Моделирование взаимоотношений трески и мойвы в экосистеме Баренцева моря: теоретические аспекты и практическое значение / А.А. Филин // Вопросы рыболовства. 2012. Т. 13. 2(50). С. 384 395. Daniel Howell, Anatoly A. Filin, Bjarte Bogstad and Jan Erik Stiansen/Unquantifiable uncertainty in projecting stock response to climate change: Example from NEA cod // Marine Biology Research. 2013. doi:10.1080/17451000.775452.
Simulation of capelin stock dynamics Observed data, 1973-2015 cod SSB >500 thousand tons capelin stock >3 mln. tons M= 3,30 n = 8 cod SSB > 500 thousand tons capelin stock <3 mln. tons M= 1,11 n = 12 cod SSB < 500 thousand tons capelin stock >3 mln. tons M= 1,91 n = 8 Capelin stock biomass vs. cod spawning stock biomass (SSB) in previous year. Brown circles denote years when capelin stock biomass in previous year was > 3 million tons cod SSB < 500 thousand tons capelin stock <3 mln. tons M= 4,85 n = 14 Historical replicates are drawn with equal probability from the 4 datasets depending on modelled cod and capelin stocks in previous year
Evaluated harvest control rules (HCR) Fishing mortality (Fbar) 0.6 0.4 0.2 0 HCR 1 (basic rule, no capelin) Fpa = 0,40 Bpa = 460 0 500 1000 1500 2000 Cod SSB, 1000 tonnes If SSB Bpa the F=0.40; if SSB < Bpa, then F is linearly reduced to F=0 at a SSB of 0 tons. Fishing mortality (Fbar) 0.8 0.6 0.4 0.2 0 Ftag Fpa = = 0,60 0,40 Fpa = 0,40 HCR 2 (capelin dependent) Bpa = 460 Bpa = 460 Btag 1 = 2Bpa Btag 2 = 3Bpa if capelin stock 1 million tons 0 500 1000 1500 2000 Cod SSB, 1000 tonnes If SSB > Bpa then F=0.40; if SSB < Bpa, then F is linearly reduced to F=0 at a SSB of 0 tons; If capelin stock 1 million tons then if cod SSB 2Bpa but 3Bpa F is linearly increased from F=0.40 at SSB=2*Bpa to F=0.60 at SSB=3Bpa; if SSB 3Bpa then F=0.60. Both HCRs were investigated in two variants: 1 without any restriction on year to year changes in TAC 2 with +/ 20% restriction on year to year changes in TAC when SSB> Bpa, min F=0.30
Simulations 100 iterations for each HCR. Modelled period 120 years, first 20 years were not used in analysis; Stochastic Ricker recruitment equation was used to couple the cod SSB and abundance at age 1. No uncertainties associated with fishery management and cod stock assessment.
Reality test Temperature and fishing mortality rate from the observed data Parameter Fishable stock biomass, mln. tons Spawning stock biomass, mln. tons Recruitment at age 3 Fishable stock abundance, 10 9 ind. Average stock weight at age 4 6, kg Maturation at age 6, % Maturation at age 7, % TAC (Landings), 1000 tons Mean annual values, 2000 2014 Modelled Observed ICES, 2015 Discrepancy, (%) 2,27 2,31 1,95 1,01 0,94 7,6 605,0 700,5 13,6 1,64 1,96 16,1 1,36 1,28 6,2 28,2 31,3 9,9 62 63,3 2,0 568,5 613,2 7,3 Capelin stock biomass 2,38 2,61 8,8
Fishery (mean values of 100 iterations) No any constraint on variations in TAC 20% constraint on variations in TAC M=615 HCR basic HCR capelin dependent
Stock biomass (mean values of 100 iterations) No any constraint on variations in TAC 20% constraint on variations in TAC HCR basic HCR capelin dependent
No any constraint on variations in TAC Recruitment and cannibalism (mean values of 100 iterations) 20% constraint on variations in TAC HCR basic The cannibalism mortality is calculated as a ratio of cod number at age 3 consumed by cod during the year to its modeled number at the beginning of a year.
Growth and maturation rate (mean values of 100 iterations) No any constraint on variations in TAC 20% constraint on variations in TAC HCR basic
Capelin (mean values of 100 iterations) No any constraint on variations in TAC 20% constraint on variations in TAC HCR basic
Simulated long-term consequences of using capelindependent HCR instead of basic HCR Parameter No constraint on TAC 20% constraint on TAC Cod growth rate 0,0% 0,0% Cod maturation rate +0,3% +0,1% Cannibalism in cod stock 3,7% 6,5% Cod fishable stock 1,9% +2,5% biomass Cod spawning stock 3,8% 0,5% biomass Cod fishing mortality +2,9% +7,4% (Fbar) Cod TAC +0,3% +8,1% Capelin stock size +0,5% +8,8%
Comparison of variability of cod stock and TAC Parameter Coefficient variation of cod TAC Mean interannual changes in TAC Probability of cod SSB < Bpa No restriction on TAC HCR basic HCR capelindependent 20% restriction on TAC HCR basic HCR capelindependent 23,3% 27,3% 25,4% 37,8% ±8,1% ±10,3% ±8,9% ±12,3% 0,2% 0,3% 0,8% 8,9%
Why variability in cod TAC will increase if capelin-dependent HCR with constraint on TAC is applied? The proportion of young individuals in cod stock will increase because the fishing mortality will increase and cannibalism will decline. The TAC is expressed in units of biomass. Fishing mortality refers to numbers. If constraint on variability in TAC is in place, the interannual changes in F increased as the rise in proportion of young cod in cod stock. Increased variability in F leads to increased variability in stock size, which results in a rise of annual variability in TAC.
Conclusions Increase of cod fishing mortality from F=0.40 to 0.60, when cod SSB is high and capelin is low, will not have negative long term consequences for cod stock and cod fishery. Effect of increased fishing mortality will be mitigated by decline in cannibalism. This will not lead to major changes if constraint on interannual variations in cod TAC is not be introduced. This will support increasing long term mean of cod TAC and capelin stock, but will reduce stability of cod TAC if 20% constraint on interannual changes in TAC will be applied.
Thank you for your attention