Reconstructing the time series of abundance of Northeast Arctic
cod (Gadus morhua), taking
cannibalism into account
Overview
• Cod cannibalism important factor in cod population dynamics
• Quantitative stomach content data available from 1984- present
• Qualitative stomach content data available from 1947-present
• Will describe cod cannibalism and its variations in time/space
• Long-term goal: Extend time series of cod abundance (VPA) down to age 1 back to 1947
Cod stomach data available
• Joint IMR-PINRO stomach content data base (1984-present), > 200 000 stomachs analysed (weight, prey composition etc.)
• PINRO investigations of qualititative stomach content (prey species found, degree of
fullness) 1947-present: > 1.5 million stomachs analysed
Definitions
• PFI: Partial Fullness Index (prey weight*10000/L3)
• FOcod
:Frequency of occurrence of cod in cod stomachs
• M2: Mortality due to cannibalism
• XSA: eXtended Survivors Analysis (a VPA-type assessment method)
Predator size vs. prey size
Prey size vs. predator size for cod preying on cod, for the period 1984-2005.
Proportion of PFI which is cod
0 0,1 0,2 0,3 0,4 0,5 0,6 0,7
17,5 32,5 47,5 62,5 77,5 92,5 107,5 122,5 Cod (predator) length
Proportion
Frequency of occurrence of cod in cod stomachs
0 6 12 18 24
0 4 8 12 16
1947 1952 1957 1962 1967 1972 1977 1982 1987 1992 1997 2002
Proportion of PFI, %
Frequency of occurrence %
Year
Proportion of cod in cod diet, entire Barents Sea
SBS+Sval Quantitative
Frequency of occurrence of cod and capelin
0 10 20 30 40 50 60 70
1947 1957 1967 1977 1987 1997
FO, %
Year
FO cod vs. FO capelin, 1947-2005.
y. cod capelin
Present use of cod cannibalism data in assessment
• Calculate amount of cod eaten per predatory cod (by half-year, predator and prey age group) for period
1984-present
• Assume that cannibalism mortality comes in addition to M=0.2
• Run XSA down to age 1 using number of cod eaten as additional catches
• Iterate until convergence
• Problem: Quantitative data not available before 1984
Survey abundance vs XSA estimate
y = 4,029x R² = 0,867
0 5000 10000 15000 20000 25000 30000
0 1000 2000 3000 4000 5000 6000
Age 1 XSA
Age 1 February survey
Cod age 1 survey vs XSA 1994-2006
y = 2,934x R² = 0,820
0 500 1000 1500 2000 2500 3000 3500
0 200 400 600 800 1000 1200
Age 2 XSA
Age 2 February survey
Cod age 2 survey vs XSA 1994-2006
y = 1,396x + 270,5 R² = 0,762
0 100 200 300 400 500 600 700 800 900
0 100 200 300 400 500
Age 3 XSA
Age 3 February survey
Cod age 3 survey vs XSA 1994-2006
Cannibalism mortality (M2) vs. capelin
abundance
Correlation between M2 for different
prey age groups
SSB-recruitment relationship, 1983- 2003 cohorts
y = 25,71x - 4171, R² = 0,449
0 5000 10000 15000 20000 25000 30000
0 200 400 600 800 1000
Recruitment Age 1(millions)
SSB (1000 Tonnes)
Age 1 y = 0,401x + 382,3
R² = 0,133
0 200 400 600 800 1000 1200
0 200 400 600 800 1000
Recruitment age 3(millions)
SSB (1000 Tonnes)
Age 3
Conclusions (I)
• Cannibalism mortality may be considerable on age 1-3 cod
• Length of predatory cod is at least twice length of cod prey
• Proportion of cod in cod diet increases with predatory cod size
• Large spatial variation in cannibalism
Conclusions (II)
• Survey estimates of age 1-3 cod (1994-2006) consistent with cannibalism data
• Cannibalism inversely related to capelin abundance
• Unexplained long-term trends in cannibalism level
• Including cannibalism improves stock- recruitment relationship
Future work
• Make model for cannibalism as a function of predator and prey (cod, capelin, other)
abundance based on quantitative stomach content data
• Hindcast cannibalism (and thus number at age) down to age 1 and back to 1947, based on
qualitative stomach content data
• Study stock/recruitment relationship and
harvesting strategies based on revised time series
