Linking climate change to community-level impacts on copepods via a new, trait-based model
Neil Banas,* Eva Møller, Torkel Nielsen, Lisa Eisner
*Univ of Strathclyde, Glasgow, UK neilbanas.com/projects
Bering Sea (60°N)
warm years; US Pacific Northwest (45°N)
warm decades
Disko Bay, West Greenland warm deep-water intrusions in 2000s
North Sea warming trend, 1960s–
Region-specific shifts in zooplankton community composition
Calanus spp. vs Pseudocalanus spp.
C. hyperboreus, C. glacialis vs C. finmarchicus
C. finmarchicus vs C. helgolandicus impacts on pollock, salmon, cod, forage fish like herring and sandeels, seabirds, whales….
survivorship Ndevelopmental stage D
Optimal annual routines (Varpe et al. 2007, 2009; Houston & McNamara 1999, Clark & Mangel 2000)
Emergent copepod communities (Record et al. 2013)
Past approaches
Coltrane (Copepod Life-history traits and adaptation to novel environments)
focus on reserves and timing
trait-based metacommunity
reserves R
structural biomass S population: annual cycle, net growth rate
small number of variable traits community associated with a given environment
×
survivorship N
developmental stage D
reserves R
structural biomass S
Q10 = 3
Q10 = 2.5; rates ~ S – 0.3
u0 (development rate corrected to 0°C) is the key trait generating size diversity (Banas and Campbell, MEPS, submitted; see poster)
net gain = ingestion – metabolism
mortality
Diapause is on/off based on a “myopic” criterion; turns off development, ingestion, and mortality, and reduces metabolism to 1/4 (Maps et al. 2012)
(Saiz and Calbet 2007; Forster et al. 2011)
survivorship N
developmental stage D
reserves R
structural biomass S
Two versions: “egg/reserves”: explicit model for income egg production (from ingestion) and capital egg production (from R)
“potential”: replace R with free scope 𝜑; look for the optimal date on which to spend it on eggs (i.e. the stable cycle that maximises egg fitness)
DiskoBay
M2
Newport
ShebaM8
E Be
ring S
ea
45°
60°
75°
180°
Large-scale biogeographic patterns
(e.g., poles to tropics)
Coexistence of multiple strategies in one environment
Time variability in one system (years to decades)
A general theory of large zooplankton in relation to environment ought to be able to reproduce
0 100 200 300
0
5
10
15
Disko Bay
Newport
Sheba
4+ gen/yr3210.5
Duration of prey availability ∂t
Annu
al m
ean
surf
ace
tem
pera
ture
(°C
)
Idealised “global biogeography” testbed
Generations per year vs. habitat in a C. glacialis/marshallae analog
(u0 = 0.007 d–1)
Gaussian window of prey availability; constant surface temperature;
deep temperature = 0.4 · surface
0 100 200 300
0
5
10
15
0 100 200 300
0
5
10
15
0 100 200 300
0
5
10
15
Duration of prey availability ∂t
Annu
al m
ean
surf
ace
tem
pera
ture
(°C
)
(a) (b) (c)
0 1 2 3 4
Maximum egg fitness FRange of viability
2 yr life cycle
1 yr life cycle
u0 = 0.01 d–1 u0 = 0.007 d–1 u0 = 0.005 d–1
Large-scale biogeographic patterns
(e.g., poles to tropics)
Coexistence of multiple strategies in one environment
Time variability in one system: Eastern Bering Sea
0 20 40 60 80 100 120 140 160 1800
1
2
3
4
5
6
7
8
Prey
sat
urat
ion
0
0.4
0.8
0
0.4
0.8
0
0.4
0.8
Jan 1 Dec 31
Tem
pera
ture
(°C
)
0
5
10
Jan 1 Dec 31
Tem
pera
ture
(°C
)
0
5
10 Surface
Bottom
Annu
al m
ean
surf
ace
tem
pera
ture
(°C
)
Date of ice retreat (yearday)
Jan 1 Dec 31 Jan 1 Dec 31Jan 1 Dec 31 Jan 1 Dec 31
0
0.5
1
1.5
2
2.5
3
Bloom follows ice retreat
Increasing ice algae
Bloom delayed by winter mixingPo
pula
tion
grow
th ra
te
73234
210 1100
5700 2600
1600
Southeast Bering Sea (M2),1971–2012
Northeast Bering Sea (M8),1971–2012
C. glacialis/marshallae abundance(log mean, ind m–2)
Bering Sea, C. glacialis/marshallae
0 20 40 60 80 100 120 140 160 1800
1
2
3
4
5
6
7
8
Prey
sat
urat
ion
0
0.4
0.8
0
0.4
0.8
0
0.4
0.8
Jan 1 Dec 31
Tem
pera
ture
(°C
)
0
5
10
Jan 1 Dec 31
Tem
pera
ture
(°C
)
0
5
10 Surface
Bottom
Annu
al m
ean
surf
ace
tem
pera
ture
(°C
)
Date of ice retreat (yearday)
Jan 1 Dec 31 Jan 1 Dec 31Jan 1 Dec 31 Jan 1 Dec 31
0
0.5
1
1.5
2
2.5
3
Bloom follows ice retreat
Increasing ice algae
Bloom delayed by winter mixing
Popu
latio
n gr
owth
rate
73234
210 1100
5700 2600
1600
Southeast Bering Sea (M2),1971–2012
Northeast Bering Sea (M8),1971–2012
C. glacialis/marshallae abundance(log mean, ind m–2)
0 20 40 60 80 100 120 140 160 1800
1
2
3
4
5
6
7
8
Prey
sat
urat
ion
0
0.4
0.8
0
0.4
0.8
0
0.4
0.8
Jan 1 Dec 31
Tem
pera
ture
(°C
)
0
5
10
Jan 1 Dec 31
Tem
pera
ture
(°C
)
0
5
10 Surface
Bottom
Annu
al m
ean
surf
ace
tem
pera
ture
(°C
)Date of ice retreat (yearday)
Jan 1 Dec 31 Jan 1 Dec 31Jan 1 Dec 31 Jan 1 Dec 31
0
0.5
1
1.5
2
2.5
3
Bloom follows ice retreat
Increasing ice algae
Bloom delayed by winter mixing
Popu
latio
n gr
owth
rate
73234
210 1100
5700 2600
1600
Southeast Bering Sea (M2),1971–2012
Northeast Bering Sea (M8),1971–2012
C. glacialis/marshallae abundance(log mean, ind m–2)
At both coarse and fine levels of detail, the threshhold for viability of high-latitude Calanus
is mainly a matter of timing, not temperature
Bering Sea, C. glacialis/marshallae
0 100 200 300
0
5
10
15
Disko Bay
Newport
Sheba
4+ gen/yr3210.5
Duration of prey availability ∂t
Annu
al m
ean
surf
ace
tem
pera
ture
(°C
)
global (idealised)
0 20 40 60 80 100 120 140 160 1800
1
2
3
4
5
6
7
8
Prey
sat
urat
ion
0
0.4
0.8
0
0.4
0.8
0
0.4
0.8
Jan 1 Dec 31
Tem
pera
ture
(°C
)
0
5
10
Jan 1 Dec 31
Tem
pera
ture
(°C
)
0
5
10 Surface
Bottom
Annu
al m
ean
surf
ace
tem
pera
ture
(°C
)Date of ice retreat (yearday)
Jan 1 Dec 31 Jan 1 Dec 31Jan 1 Dec 31 Jan 1 Dec 31
0
0.5
1
1.5
2
2.5
3
Bloom follows ice retreat
Increasing ice algae
Bloom delayed by winter mixing
Popu
latio
n gr
owth
rate
73234
210 1100
5700 2600
1600
Southeast Bering Sea (M2),1971–2012
Northeast Bering Sea (M8),1971–2012
C. glacialis/marshallae abundance(log mean, ind m–2)
Warming per se is not necessarily a stressor
Bering Sea, C. glacialis/marshallae
0 100 200 300
0
5
10
15
Disko Bay
Newport
Sheba
4+ gen/yr3210.5
Duration of prey availability ∂t
Annu
al m
ean
surf
ace
tem
pera
ture
(°C
)
global (idealised)
Large-scale biogeographic patterns
Coexistence of multiple strategies in one environment: Disko Bay, West Greenland
Time variability in one system (years to decades)
Disko Bay
1996–97 annual cycle + two axes of diversity: u0 (development rate → adult size) tegg (delay between maturation and egg production)
Adult size (µg C)
100 200 500 1000 2000
Gen
erat
ion
leng
th (y
r)
0
1
2
3
4
Adult size (µg C)
Gen
erat
ion
leng
th (y
r)
C. finmarchicus C. glacialis C. hyperboreus (Swalethorp et al. 2011)
Prey
con
cent
ratio
n P
(mg
chl m
–3)
Mar Apr May Jun Jul Aug Sep Oct
Rela
tive
egg
prod
uctio
n
C. g
laci
alis
C. h
yper
bore
us
C. fi
nmar
chic
us
0
0.2
0.1
10
5
15
0
P
Adult size (µg C) Adult size (µg C)
Adult size (µg C) Adult size (µg C)
50 100 200 500 1000 2000
Med
ian
spaw
ning
dat
e (y
eard
ay)
80
100
120
140
160
180
50 100 200 500 1000 2000
D a
t firs
t dia
paus
e
0.7
0.8
0.9
1
50 100 200 500 1000 2000
Cap
ital f
ract
ion
of e
gg p
rodu
ctio
n
0
0.2
0.4
0.6
0.8
1
50 100 200 500 1000 2000
Mea
n re
serv
e fr
actio
n
0
0.2
0.4
0.6
C4C5
(a) (b)
(c) (d)
Pure capital breeding
Pure income breeding
Wa50 100 200 500 1000 2000
tEcen jittered
100
110
120
130
140
150
160
170
Wa50 100 200 500 1000 2000
Ddiam
injittered
0.5
0.6
0.7
0.8
0.9
1
1.1
Wa50 100 200 500 1000 2000
capfrac
0
0.2
0.4
0.6
0.8
1
Wa50 100 200 500 1000 2000
rhom
ean
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Adult size (µg C)
Wax
est
er c
onte
nt (f
ract
ion
of to
tal C
)
C. finmarchicusC. glacialis
C. hyperboreus
1 yr/gen
2 yr/gen
3 yr/gen
( = Swalethorp et al. 2011)
Where this is headed
IPCC AR5
highly realistic, single-species copepod models current rules transposed to a new ocean
Coltrane metacommunity predictions all possible ways to be an anthropocene copepod
(actual adaptive capacity somewhere in between)
Summary
Many patterns in Calanus spp. (in latitude, time, and trait space) can be reproduced as a consequence of a handful of constraints in an individual’s energy budget…
total energy available in an environment per year; energy and time required to build a body; metabolic and predation penalties for taking too long to mature and reproduce; size and temperature scalings for vital rates
Phenology is crucial, but not (in these examples) through match/mismatch.
This approach constitutes a metacommunity model on top of which one can layer other species-level or region-specific constraints: cues for diapause, physiology of egg production, prey quality and selectivity, environmental dependence of predation, and so on.
neilbanas.com/projects/coltrane
