Prey-predator interactions between the myctophid Bentosema glaciale and calanoid in the Labrador Sea

P. Pepin Northwest Atlantic Fisheries Centre • Throughout the North Atlantic, copepods of the genus are dominant players in the zooplankton community • C.finmarchicus is most important both numerically and in terms of integrated production

www.sahfos.org Previous modeling of C.finmarchicus had shown that paramaterization based on conditions in the Eastern Atlantic yielded inaccurate results, particularly in the Labrador Sea

Speirs et al. (2006) Previous modeling of C.finmarchicus had shown that paramaterization based on conditions in the Eastern Atlantic yielded inaccurate results, particularly in the Labrador Sea

The introduction of temperature-dependent mortality term yielded more realistic seasonal pattern in the spatial distribution

Speirs et al. (2006) • The assumption of a temperature-dependent mortality term in modeling C. finmarchicus appears reasonable but there are few data on which to base this decision

• To achieve strong predictive capacity requires strong causal relationship

• Under the assumption that the primary cause of mortality is predation, and with limited knowledge of the sources of mortality in sub-Arctic ecosystems, we turned our attention to the potential role of the Deep-Scattering Layer (DSL)

• Within the region, and other similar ecosystems, a major component of the DSL consists of myctophids, with Benthosema glaciale being a dominant species (Sameoto 1989; Bagoeine et al., 2001) and evidence indicated they could have a significant impact on the zooplankton community Benthosema glaciale

Pelagic Non-migratory Depth range 0 - 1100 m Planktivorous Max length ~ 10 – 11 cm High probability of occurrence throughout the North Atlantic

www.fishbase.org Observation program 8 • Completed two exploratory 5 surveys of NL Shelf and 36 37 Labrador Sea (June, August – 38 35 6 Sept 2006) 5 39 40 34 41 – Zooplankton collections 42 33 43 32 (integrated and vertically 31 4 5 30 stratified samples) to 29 28 measure condition (lipids), 27 16 16 E 26 egg production D 17 U 14 T 52 25 18

TI 19 – Hydroacoustic surveys to 13

LA 24 20 21 determine distribution of 22 12 23 11 potential predators 10

50 9 (planktivorous fish) 8 7 6 – Stomach sampling program 5 4 to estimate predator 3 8 2 consumption rates (some 4 1 depth stratification) • Focus on deep water and areas 46 of potential cross shelf transport 58 56 54 52 50 48 46 44 42 LONGITUDE Observation program Observation program Observation program

50 -6 ) 10 0 m 15 0 -6 . 5 (

ur 20 0 -7 L

25 0 og ( 30 0 -7 . 5 onde S f A o

35 0 -8 [m r 40 0 P 2

/ -8 . 5 m

) 45 0 -2 ]) m 50 0 -9 ( 55 0 th -9 . 5

p 60 0 e -1 0

D 65 0 70 0 -1 0 . 5 75 0 -1 1 0 3 0 6 0 9 0 1 2 0 15 0 1 8 0 21 0 2 4 0 2 7 0 3 00 33 0 T e leos t 673 ( 1 0- 12 June / J ui n 2006 ) 0 20 19 ) 50 18 17

(m 100 16 r Te

u 15 e 150 14 m d

13 pe n 12 o

200 r f 11 a t

10 u ro

250 r

9 e P 8 ( o

) / 300 7 C

m 6 )

( 5 350 4 th 3 p 400

e 2

D 1 450 0 -1 500 -2 0 3 0 6 0 9 0 120 1 5 0 180 210 24 0 2 70 3 0 0 330

Near surface scattering layer present on NL Shelf and in Labrador Sea – consists principally of amphipods with some medusae Deep scattering layer (DSL) is present throughout Labrador Sea Average temperature at 400-500 m ~ 4°C Observation program

Deep scattering layer (DSL) – 50-70% of biomass consists of myctophids

Background integrated Sv in Lab Sea ~ 10 – 80 times that found on NL Shelf

100 ids

ight Amphipods e 80 Hatchet fish r po Unidentified a Balck Giotre 60 Shrimp Viper

ge by w Boa age p

t 40 Lanceet ta Squid Snipe Eel 20 Myctophids rcen urcen e o

P 0 1000 P Lumi Sur e r

) 1e-3 e f ace l a -2 100 m l a 2 sur i 1e-4 ght (m A f ( ace (

10 µ S l Mo 1e-5 µ ata l m Mo gr -2 l

1 m s Inte 1e-6 -1 -2 )

s -1 ) 1e-7 0.1 0 30 60 90 120 150 180 210 240 270 300 330 Distance along transect (nautical miles) Distance le long du transect (milles nautiques) Abundance estimates

• Range of target strength measurements for myctophids – MacLennan and Forbes (1987) – TS = -61.7 dB – Torgersen and Kaartvedt (2001) – TS (modal) -63 -58 dB – Valinassab (2000) – TS = -60 -55 dB – Benoit-Baird and Au (2003) – TS = -44.8 dB

• Range of densities ~ 8 – 53 fish m-2

• Range of biomass ~ 15 – 190 mt km-2 Observation program

0.30

Mean (10th - 90th percentile) Median 0.25 Extensive diurnal vertical migration of DSL but only

0.20 in a fraction of DSL (10- 100m e 25% of integrated Sv)

0.15 abov on i t Difficult to determine 0.10

opor whether there is r P replacement of individuals 0.05 in upper layer

0.00 0 4 8 12162024 NDT Dusk to midnight

Dusk to midnight 0.030 0.5

0.025 Deep (300 - 600 m) 0.4 Surface (< 250 m) )

h 0.020 n s o fi i t

or 0.3 / w h c op a 0.015 m pr o e st v i w t 0.2 a ( l I 0.010 e R GF

0.1 0.005

0.000 0.0 <0.001 0.001 - 0.010.01 - 0.020.02 - 0.030.03 - 0.040.04 - 0.05 >0.05 Deep (300 - 600 m) Surface (<250m) GFI ( wstomach / wfish ) Midnight to dawn

0.030 0.5

Deep (300 - 600 m) 0.025 0.4 Surface (< 250 m) )

on h

0.020 i s t fi 0.3 por / w o h r

0.015 p e omac t v i s t

w 0.2 a

l ( e I

0.010 R GF

0.1 0.005

0.0 0.000 <0.001 0.001 - 0.010.01 - 0.020.02 - 0.030.03 - 0.040.04 - 0.05 >0.05 Deep (300 - 600 m) Surface (<250m) GFI ( wstomach / wfish ) Diurnal cycle (300 – 500 m)

0.06 Vertical migration motivated by hunger 0.05 level )

ish 0.04 f Relative proportion of / w

h migrants suggests that 0.03 omac

t turnover rates ~ 4 – 8 d s w

( I 0.02 GF

0.01

0.00 00:00 - 04:00 04:00 - 08:00 08:00 - 12:00 12:00 - 16:00 16:00 - 20:00 20:00 - 24:00 20

Amphipoda 10 Calanus spp. 9 ) 8 Calanus fhg m 7 CV I m 6 e ( 5 CV o mm)

i Calanus hyperboreus ( 4 CV I Calanoid pr th CIV a g CV n 3 CV I Copepoda

le CIII Euchaeta spp. de l CV CIV y r e Euphausiidae r 2 eu

P CIV g CIII CII Hyperidae Unidentified species

Lon CIII CII CI 1 0.9 CII CI C.hyperboreus 0.8 C. glacialis 0.7 CI 0.6 C. finmarchicus 0.5 10 20 30 40 50 60 70 80 90 Myctophid length (mm) Longeur du myctophide (mm) 20

Amphipoda 10 Calanus spp. 9 ) 8 Calanus fhg m 7 CV I Calanus finmarchicus m 6 Calanus glacialis e ( 5 CV o mm)

i Calanus hyperboreus ( 4 CV I Calanoid copepod pr th CIV a g CV n 3 CV I Copepoda

le CIII Euchaeta spp. de l CV CIV y r e Euphausiidae r 2 eu

P CIV g CIII CII Hyperidae Unidentified species

Lon CIII CII CI 1 0.9 CII CI C.hyperboreus 0.8 C. glacialis 0.7 CI 0.6 C. finmarchicus 0.5 10 20 30 40 50 60 70 80 90 Myctophid length (mm) Longeur du myctophide (mm) Fish length (20-35 mm) Fish length (45-55 mm) 0.35 0.5

0.30 0.4

0.25 y y c n nc

e 0.3 u ue 0.20 C. hyperboreus q q e r fre f e ve C. glacialis v i

t 0.15 ti C. hyperboreus a a 0.2 l l e e R C. finmarchicus R 0.10 C. glacialis

0.1 C. finmarchicus 0.05

0.00 0.0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0123456789101112131415 Prey (thorax) length (mm) Prey (thorax) length (mm) Analytical results – stomach contents

• Generalized linear model • General linear model (Poisson error structure) (log-transformed GFI) • Number of prey per fish •GFI – Fish length ( P < 0.001) – Time of day ( P < 0.05 ) – Time of day ( P < 0.001) [highest midnight to [highest after dawn – midday] lowest at dusk] [lowest midday to midnight] – Positively affected by – Negatively affected by temperature (0-100m) temperature (300-500) ( P < 0.001) ( P < 0.001) for fish sampled in deep layer Q10~2.2 – Temperature effect Q10~1 significant for fish sampled – No significant effect of in surface layer and not temperature for fish significant for fish in deep sampled in surface layer layer Analytical results – stomach contents

• General linear model (log-transformed prey length) • Prey length – Significantly positively related to fish length ( P < 0.001) – Not significantly affected by temperature ( P > 0.1) – Significantly related to time of day ( P < 0.001) [larger prey more frequent in stomach sampled midday to dusk] – Significant seasonal (June vs September) effect ( P < 0.001) [probably a reflection of differences in availability – few larger copepods in August relative to June] 8 5

36 37 38 35 6 5 39 Vertical distribution of C. finmarchicus 40 34 41 42 33 43 32 31 54 30 Similar distributions for sister species 29 28 16 27 16 E 26 D 17 14 TU 52 25 18 -3 -3 TI 19 13 Density (# m ) Density (# m )

LA 24 20 21 22 12 23 11 0 500 1000 1500 2000 100 200 300 400 500 600 10

50 9 0 0 8 7 6 5 4 3 2 100 100 48 1

200 200 46 58 56 54 52 50 48 46 44 42

LONGITUDE ) m ( h

t 300 300 p De

400 400

500 500

600 600 0 500 1000 1500 2000 100 200 300 400 500 600 Analytical results – Mortality estimate

• Range of analytical assumptions – Fish density ~ 8 – 53 m-2 – Mean number of prey per fish (surface feeders) ~ 2 – 10 – Turnover rate ~ 0.10 – 0.25 d-1 – Proportion of prey C. finmarchicus CIII-CVI ~ 0.6 – C. finmarchicus density ~ 10,000 – 30,000 m-2 – Total consumption ~ <1 – 100 m-2 – Mortality 0.00004 – 0.008 d-1 Summary and Conclusions

• The deep-scatter layer (DSL) is widely (and almost uniformly) distributed through the western Labrador Sea, with an overall biomass 8 – 190 mt km-2 • On average, 70% of the biomass consists of Benthosema glaciale • Preferred temperature range: 3.7 – 4.1°C • Only 10 – 25% of the DSL migrates to surface waters daily • B. glaciale that migrate to the surface appear to be motivated by hunger • Digestion time / turnover rate ~ 4 – 10 days Summary and Conclusions

• Calanoid copepods make up 80% of the prey found in the stomachs • Late stage C. finmarchicus, C. glacialis and C. hyperboreus are more prevalent in the stomachs than earlier stages of each species • In contrast to expectations, measures of feeding activity (number of prey, GFI) are not equally affected by environmental temperature – – numbers of prey in surface feeders are positively affected by temperature (Q10 ~ 2.2) – GFI of fish in either surface or deep layers is negatively affected by temperature(Q10 ~ 1) • Overall impact of B. glaciale on C. finmarchicus results in mortality rates < 1% d-1 Issues for further research

• Analysis of feeding patterns in B. glaciale requires further assessment – Role of hunger in motivating activity and feeding pattern – Effect of local prey availability on feeding activity (numbers, biomass, selection) – Metabolic activity in cold environments???

• Other sources of mortality for Calanus in the region? ~100% Themisto libellula

In the Gulf of St. Laurence, A. Marion (University du Quebec, Rimouski) estimated that T. libellula could ingest ~2 % of the mesoplankton community, and exert a predation rate of ~3% d-1 on C. finmarchicus (CIV – CVI).

B. Glaciale may play a role in dynamics of C. finmarchicus but role of invertebrate predators needs to be investigated (greater overlap, turnover rates, local density) Thank you