AN ABSTRACT OF THE THESIS OF
Gregory L. Kowalke for the degree of Master of Science in Oceanography presented on
March 30, 2010.
Title: Transport of the Copepod, Calanus finmarchicus , in the Northwest Atlantic during
Diapause.
Abstract Approved:
Harold P. Batchelder
The copepod Calanus finmarchicus is the most important and biomass dominant mesozooplankter in the temperate-boreal North Atlantic. C. finmarchicus has an overwintering phase, termed diapause, during which it descends to great depths (300-
2000m) and is metabolically quiescent for up to ten months. Changes in the currents at depth due to climate variation may influence the spring distribution and abundance of C. finmarchicus . We evaluated the potential transport of C. finmarchicus during diapause in the Labrador Sea and the Scotian Shelf under positive and negative North Atlantic
Oscillation (NAO) index conditions with a coupled bio-physical individual-based model
(IBM). The physical simulation was a basin-scale ROMS model using mean forcing for
1980-1993 (1962-1971) from NCEP/NCAR re-analysis fields to generate idealized positive (negative) NAO conditions. The biological IBM consisted of a particle-tracking model with simulated biological behavior. We modeled four mechanisms that have been

proposed for diapause emergence: an external, photoperiod-dependent emergence trigger and three internal, temperature-controlled lipid consumption triggers. Copepods were seeded in the Labrador Sea and the Scotian Shelf for positive and negative NAO climatic conditions. An internal, lipid-based diapause emergence trigger based on Saumweber and
Durbin (2006; Deep Sea Res. II, 53 (23-24), 2597-2617) best reproduced field observations of the dates of C. finmarchicus appearance at the surface following diapause. The phase of the NAO had an effect on potential transport distance of C. finmarchicus during diapause on the Scotian Slope, and in the northeastern Atlantic. In the Labrador Sea, the phase of the NAO had little effect on potential transport and duration of diapause.

©Copyright by Gregory L. Kowalke
March 30, 2010
All Rights Reserved

Transport of the Copepod, Calanus finmarchicus , in the Northwest Atlantic during
Diapause by
Gregory L. Kowalke
A THESIS submitted to
Oregon State University in partial fulfillment of the requirements for the degree of
Masters of Science
Presented March 30, 2010
Commencement June 2010

Master of Science thesis of Gregory L. Kowalke presented on March 30, 2010.
APPROVED:
Major Professor representing Oceanography
Dean of the College of Oceanic and Atmospheric Sciences
Dean of the Graduate School
I understand that my thesis will become part of the permanent collection of Oregon State
University libraries. My signature below authorizes release of my thesis to any reader upon request.
Gregory L. Kowalke, Author

ACKNOWLEDGEMENTS
I‘d like to thank my advisor, Dr. Harold Batchelder for his assistance and encouragement, without which I could not have completed this thesis.
I‘d also like to thank my committee, Dr. Kelly Benoit-Bird and Dr. Yvette Spitz, for their support and assistance in producing this thesis and the research behind it.
I‘d like to acknowledge the help and sympathy I got from my labmates, Brendan Reser and Brie Lindsey. It was always well-timed and exactly what I needed, whether it was advice or a much needed break from the office.
I‘d like to thank my parents for, well, everything. Thanks Mom and Dad!
I‘d like to thank my best friend, Dr. Leanna Dudley. You always believe in me, even when I don‘t. Thanks for being a great friend.
Lastly, I‘d like to thank my girlfriend, Eden Rivers. Thank you for your unending support and encouragement, your patience as I pursued my degree, and for always providing a cheerful respite when things got rough. I can never thank you enough, but,
I‘ll try.
This research was funded by the National Science Foundation and the National
Oceanographic (award #NA03NES4400001) and Atmospheric Administration through the U.S. GLOBEC program, and is GLOBEC contribution number 662.

TABLE OF CONTENTS
Page
1.
Introduction ..................................................................................1
1.1.
Calanus finmarchicus and diapause .......................................2
1.2.
C. finmarchicus lifecycle and ocean physics .........................5
1.3.
North Atlantic Oscillation ......................................................8
1.4.
The transport of diapausing C. finmarchicus .........................11
2.
Transport of the copepod, Calanus finmarchicus in the northwest Atlantic during diapause ........................................13
2.1.
Introduction ............................................................................14
2.2.
Materials and methods ...........................................................17
2.2.1.
ROMS model ...................................................................18
2.2.2.
IBM model .......................................................................19
2.2.3.
Diapause emergence triggers within the
IBM ..................................................................................22
2.2.4.
Transport of diapausing C. finmarchicus in the northwest Atlantic ......................................................24
2.3.
Results ....................................................................................25
2.3.1.
Comparison of C. finmarchicus diapause emergence triggers ...........................................................25
2.3.2.
Dispersal of C. finmarchicus during diapause............................................................................26
2.3.3.
Duration of C. finmarchicus diapause ..............................27
2.3.4.
Potential transport distances of diapausing
C.finmarchicus
.................................................................30
2.3.5.
Multiple regression exploration of controls on diapause duration and transport ..................................32
2.4.
Discussion ..............................................................................33
3.
General conclusion.......................................................................55

TABLE OF CONTENTS (Continued)
Page
4.
Addendum ....................................................................................58
5.
Literature Cited ............................................................................61
6.
Appendix A ..................................................................................68

LIST OF TABLES
Table Page
1.
Means and standard deviations for diapause duration, total migration duration, total distance traveled, and net distance traveled. ..................................................................................39
2.
Mean DOY and calendar day of copepod reaching surface waters after emergence from diapause. ..................................41
3.
Results of stepwise multiple regression for diapause duration, total migration duration, total distance traveled and net distance traveled. .......................................................42

LIST OF FIGURES
Figure Page
1.
Region of interest for the NABM ROMS physical model
(dark line) and the IBM biological model (dashed line). .......43
2.
Spatial (A) and temporal (B) distribution of WOD05 temperature data used for comparison with output from the
1995-1998 NABM ROMS model. .........................................44
3.
Difference plot of NABM 1995-1998 output versus
WOD05 for temperature (A), and salinity (B). ....................45
4.
Seeding locations of particles representing C. finmarchicus in the IBM biological model. .................................................46
5.
Mean surfacing dates, binned to the nearest whole degree of latitude and longitude, for four different diapause emergence triggers: photoperiod (A), 75% wax ester kept in reserve (B), 38% wax ester kept in reserve (C), and
Saumweber and Durbin (2006) prosome length dependent emergence (D). .......................................................................
6.
Composite tracks of copepods released at all stations in positive NAO conditions (A & C) and negative NAO conditions (B & D) and with a July 15 th
release date (A &
B) and August 15 th
release date (C & D). ..............................48
7.
Tracks of copepods released on August 15 th
at station A for positive NAO conditions (A) and negative NAO conditions (B), and station B for positive NAO (C) and negative NAO (D). .................................................................49
8.
Mean day of the year (DOY) that copepods reached near surfaces waters. . ...................................................................54
9.
Plots of the prosome length (body size) dependent maximum total carbon weight using the Saumweber and
Durbin (2006) equation, the Kowalke replacement
(―Kowalke—this thesis‖) and the revised equation provided by Whitley Saumweber (pers. comm., 27 April
2010). ....................................................................................60
47

LIST OF APPENDIX TABLES
Table Page
1.
Means and standard deviations for diapause duration, total migration duration, total distance traveled, and net distance traveled of copepods in three weight classes and two mean diapausing depth bins at each station. Table contains the statistics for copepods released in July. .................................69
2.
Means and standard deviations for diapause duration, total migration duration, total distance traveled, and net distance traveled of copepods in three weight classes and two mean diapausing depth bins at each station. Table contains the statistics for copepods released in August. ............................71

LIST OF APPENDIX FIGURES
Figure Page
1.
Temperature anomalies between the 1995-1998 NABM and the WOD05 at 10m depth................................................73
2.
Temperature anomalies between the 1995-1998 NABM and the WOD05 at 400m depth..............................................74
3.
Position of the north wall of the Gulf Stream in the NABM under positive and negative NAO conditions. .......................75
4.
Annual mean temperature of the NABM at 10 m depth for
(A) positive NAO conditions and (B) negative NAO conditions. ..............................................................................76
5.
Annual mean temperature of the NABM at 400 m depth for (A) positive NAO conditions and (B) negative NAO conditions. ..............................................................................77
6.
Annual mean velocity fields of the NABM at 10 m depth for (A) positive NAO conditions and (B) negative NAO conditions. ..............................................................................78
7.
Annual mean velocity fields of the NABM at 400 m depth for (A) positive NAO conditions and (B) negative NAO conditions. ..............................................................................79
8.
Map of Labrador Sea and Scotian Shelf indicating the
AR7W sampling line (blue), and the Flemish Pass (red). ....80
9.
Annual mean temperature profile across the AR7W line in the NABM for (A) positive NAO conditions and (B) negative NAO conditions. ......................................................81
10.
Annual mean salinity profile across the AR7W line in the
NABM for (A) positive NAO conditions and (B) negative
NAO conditions. ....................................................................82
11.
Annual mean of velocity perpendicular to the AR7W line in the NABM for (A) positive NAO conditions and (B) negative NAO conditions. ......................................................83

LIST OF APPENDIX FIGURES (Continued)
Figure Page
12.
Annual mean temperature profile across the Flemish Pass in the NABM for (A) positive NAO conditions and (B) negative NAO conditions. ......................................................84
13.
Annual mean of velocity perpendicular to the Flemish Pass in the NABM for (A) positive NAO conditions and (B) negative NAO conditions. ......................................................85
14.
Temperature profile over time at Station B in the NABM for (A) positive NAO conditions and (B) negative NAO conditions. ..............................................................................86
15.
Temperature profile over time at Station C in the NABM for (A) positive NAO conditions and (B) negative NAO conditions. ..............................................................................87
16.
Distribution of prosome lengths in all experiments. ..............88
17.
Distribution of pre-assigned isopycnals for all experiments. ...........................................................................89
18.
Tracks of copepods released on July 15th at station A for
High NAO conditions (A) and Low NAO conditions (B), and station B for High NAO (C) and Low NAO (D). ...........90
19.
Depth of copepod versus one hour timestep (720 timesteps per month) of all copepods released at Station C on August
15 th
in (A) positive and (B) negative NAO conditions. . ......95
20.
Depth of copepod versus one hour timestep (720 timesteps per month) of all copepods released at Station I on August
15 th
in (A) positive and (B) negative NAO conditions.. ........96
21.
Depth over time profile of the journey of particle 897 during (A)positive and (B) negative NAO conditions.. .........97
22.
Peak abundance of C. finmarchicus in the North Atlantic, from Continuous Plankton Recorder data. .............................98

Transport of the copepod, Calanus finmarchicus , in the northwest Atlantic during diapause
1.
I
NTRODUCTION
In terms of biomass, the copepod Calanus finmarchicus is the dominant mesozooplankter in the temperate subarctic North Atlantic. Its biogeographic range spans the basin from the Labrador Sea to the Norwegian Sea, and from the Gulf of Maine in the south, to the
Barents Sea in the north. Within this vast domain, C. finmarchicus can attain concentrations so high that fishermen have reported that they rustle the surface of the sea as if it were raining (Marshall and Orr 1955). During the spring bloom, C. finmarchicus has been measured at concentrations of up to 165,000 individuals per square meter (Head et al. 2003). With this combination of ubiquity and high abundance, it is not surprising that C. finmarchicus is a key species of the ecosystem of the North Atlantic. Through sloppy feeding and excretion, it contributes greatly to nutrient cycling in the euphotic zone, and carbon export to the abyss through sinking feces (Roy et al. 2000). C. finmarchicus is also important to the ecosystem as a prey species. Its relative abundance during the spring diatom blooms is an indicator of the abundance of many economically important fish species, because of its significance as prey for larvae and juveniles of species such as Atlantic cod and haddock (Heath and Lough 2007). Right whales are supported throughout the winter and spring by feasting on diapausing and recently emerged C. finmarchicus (Wishner et al. 1995). However, C. finmarchicus has seen dramatic declines in abundance in its historic biogeographic range in recent years; this

2 may have severe effects on the ecosystem, and illustrates the need to understand the effects of climate upon C. finmarchicus (Helaouët and Beaugrand 2007)
1.1
Calanus finmarchicus and diapause
Depending on the generation and the region, the lifecycle of C. finmarchicus can include an optional resting stage, similar to diapause in insects (Hirche 1996). Low productivity in the North Atlantic during late summer and winter is harsh for populations of C. finmarchicus . To avoid this, as well as to avoid predation in the euphotic zone and reduce metabolic rate, C. finmarchicus sink to deep water and enter diapause for up to ten months (Kaartvedt 1996). In the late summer, late stage copepodites (C4 and C5) of C. finmarchicus accumulate energy as wax esters in an oil sac that may extend most of the prosome length. This oil storage may comprise up to 50% of the copepod total weight, depending on the size of the copepod (Miller et al. 2000). In each generation, some of the C4 and C5 copepodites descend to great depth, up to 2500 m deep, where they remain nearly motionless in a low metabolic, quiescent state, fueled only by stored wax esters
(Hirche 1996). By the end of the summer, most individuals of C. finmarchicus have entered diapause. The diapausing copepodites are transported by deep currents for up to ten months . Early in the year, usually corresponding to the spring phytoplankton blooms, C. finmarchicus arouse from diapause, and ascend to the surface to feed and reproduce. Wax ester stores that remain at arousal are utilized for molting to adult and gonadal development (Rey-Rassat et al. 2002).

The trigger for arousing from diapause is still not confirmed. Several mechanisms have
3 been proposed for the cue that initiates arousal, including those that are external to the physiology of the individual (external triggers) and those dependent on the individuals physiological condition (internal triggers). A proposed external trigger for arousal is related to light intensity, wherein the dramatic change in light intensity or photoperiod at depth that occurs at high latitudes during the spring cues arousal (Miller et al. 1991).
Photoperiod at high latitudes will increase by 20% by the first of March as compared to the winter solstice (Clarke and Backus 1957; Miller et al. 1991). It has been questioned whether C. finmarchicus can sense these changes at depths greater than 500 m (Irigoien
2004). However, Clarke and Kelly (1965) demonstrate that light can penetrate to 500 m with an intensity of approximately 10
-5
μW cm
-1
. Although this is dim, the copepod
Acartia tonsa has been reported to be phototaxic at light intensities as low as 5.6x10
-5
μW cm
-1
, and it is possible that C. finmarchicus can sense the same intensity (Stearns and
Forward 1984). But, this dramatic increase in photoperiod occurs earlier in higher latitudes than lower latitudes, which is contrary to observed latitudinal effects on emergence date of diapausing C. finmarchicus (Heath et al. 2000; Irigoien 2000).
It has also been suggested that C. finmarchicus arousal is due to an internal physiological, genetic or hormonal process. One that has been proposed is the utilization and mobilization of the wax ester storage products (Hirche 1996). C. finmarchicus does not feed during diapause, and, therefore, must rely entirely on stored wax ester to support metabolism (Hallberg and Hirche 1980). Although metabolic rate is greatly reduced

during diapause, C. finmarchicus does exhibit a significant reduction in oil sac content
4 during diapause (Saumweber and Durbin 2006). Wax ester stores sustain C. finmarchicus during diapause, but most of the wax ester is needed after arousal for maturation and gonadal development. Estimates of the amount of wax ester that must be reserved for maturation and gonadal development vary, with Rey-Rassat et al. (2002) suggesting that
38% of the wax ester stored at the start of diapause may be necessary, whereas
Jónasdóttir (1999) suggests that C. finmarchicus may require 75% of stored wax ester for maturation and gonadal development. Saumweber and Durbin (2006) calculated minimum and maximum wax ester stores based on oil sac volumes observed from deepwater collected specimens, and proposed that a C. finmarchicus copepodite emerges from diapause when its wax ester stores are reduced to a prosome length specific minimum.
Their observations suggest that the threshold volume of the oil sac that results in arousal from diapause is not a fixed percentage of the original content, but, varies from 25-41% with prosome lengths between 2.0-3.0 mm.
Irigoien (2004) has presented a variation of the oil storage-based internal trigger hypothesis. Noting that wax esters compresses at depth, Irigoien (2004) suggested that the depth at which a C. finmarchicus diapauses is determined by the buoyancy of its wax ester stores. Wax esters are buoyant at surface pressures, but compress rapidly at depth, altering their buoyancy. Consumption of the wax ester stores and mobilization to a more readily available lipid might increase the buoyancy of the copepodite, initiating its arousal and ascent to the surface. Over time, natural selection tuned the amount of wax

esters needed to assure that the emergence of most C. finmarchicus from diapause corresponds with the spring diatom bloom in the native oceanic basin. Campbell (2004)
5 disagreed; based on observations that Neocalanus plumchrus could rapidly adjust their density to maintain neutral buoyancy, he suggested that C. finmarchicus may have the same capacity to regulate diapause depth.
Timing of entrance and termination of diapause may be controlled by an internal genetic and/or hormonal trigger. Ecdysteroids are known to control moulting in Arthropods, and are present and utilized in copepods. The concentrations of ecdysteroids are lower in diapausing Calanus pacificus than in active individuals (Johnson 2003). An assay of gene expression in diapausing and active C. finmarchicus indicated that active copepods expressed three genes associated with lipid synthesis, transport and storage that were not expressed in diapausing copepods, whereas diapausing copepods expressed genes associated with ferritin production and ecdysteroid receptors (Tarrant et al. 2008). The exact mechanisms behind diapause emergence triggers are still being investigated, but, many current hypotheses show promise.
1.2
C. finmarchicus lifecycle and ocean physics
As C. finmarchicus emerges from diapause and matures to an adult, it ascends to the surface. It often mates during the ascent, and females produce eggs upon reaching the surface, or, lacking the necessary lipid reserves, produce eggs after a short period of feeding (Heath 1999). Depending on the region, C. finmarchicus may go through one,

two, or three generations before most of the population enters diapause (Gislason and
6
Astthorsson 1998; Tande 1982). In multiple generation populations, at least a small proportion of each generation enters diapause (Irigoien 2000). The survival and emergence rate of these early diapausing individuals is unknown; they may be lost to the population, or they may mature in spring with the rest of the population.
In the northeast Atlantic, C. finmarchicus is found from the edge of the ice in the
Norwegian and Barents Seas, southward to the North Sea and the English Channel. The continental shelf regions of this biogeographic range, and especially the North Sea and
English Channel where C. finmarchicus nauplii and copepodites are historically dominant components of the zooplankton in spring and early summer, are too shallow for C. finmarchicus to successfully overwinter; these regions are recolonized each spring by copepods advected from populations that have overwintered in the deep Norwegian Sea and the Faroe-Shetland Channel (Heath et al. 2000). In the Norwegian Sea, there are resident populations that overwinter at depths of 150-380 m in the nearshore fjords that emerge from diapause in February and March (Falkenhaug et al. 1997). However, the main overwintering site for C. finmarchicus is in the Lofoten Basin and Tromsoflaket regions of the Norwegian Sea, where C. finmarchicus diapauses at depths of 700-1200 m.
C. finmarchicus emerge from these deep basins in February/March to feed on the spring diatom blooms of the Norwegian Sea. Once in surface waters C. finmarchicus may be advected into the North Sea (Halvorsen et al. 2003). Overwintering C. finmarchicus at
600-1000 m in the Faroe-Shetland Channel north of Scotland emerge from diapause in

7 early February and are advected into the productive North Sea (Heath 1999). Throughout the northeast Atlantic, the diapause depth of C. finmarchicus corresponds to the upper boundaries of Norwegian Sea Deep Water (NSDW). The NSDW is a cold (-2-0

C), relatively fresh water mass, with a potential density of σ t
=28, formed in the arctic that flows southward at depth throughout the northeast Atlantic (Heath and Jónasdóttir 1999;
Halvorsen et al. 2003).
The Labrador Sea is an important center of abundance for C. finmarchicus in the western
Atlantic (Tittensor et al. 2003). During peak abundance in late spring, C. finmarchicus may exceed 165,000 individuals m
-2
(Head et al. 2003). In the Labrador Sea, C. finmarchicus diapauses throughout the basin: C. finmarchicus diapause at 200-400 m in the western Labrador Sea, at deeper than 600 m on the Labrador slope, and at depths to
1000 m near the tail of the Grand Banks (Head et al. 2003; Head and Pepin 2008).
Temperatures are 2-4

C and salinity approximately 35 psu at these depths (Head et al.
2003). In the Labrador Sea, C. finmarchicus emerge from diapause in late spring, corresponding with the productive diatom bloom. Labrador Sea populations will proceed through one to two generations before entering diapause in late summer (Head and Pepin
2008).
The Gulf of Maine, Georges Bank and the Scotian Shelf represent the southernmost limit of C .
finmarchicus populations in the North Atlantic. Despite being at the edge of its biogeographic range, C. finmarchicus dominates mesozooplankton abundance in spring

and early summer (Johnson et al. 2008). These regions are recolonized each spring, mostly by C. finmarchicus populations that overwinter in nearby deep basins, such as the
8
Emerald Basin on the Scotian Shelf, or Georges, Wilkinson and Jordan Basins in the Gulf of Maine (Bucklin and Kocher 1996). C. finmarchicus may also overwinter in slope waters, north of the north wall of the Gulf Stream (Head et al. 1999; Miller et al. 1991).
C. finmarchicus is abundant in the Gulf of Maine and Scotian Shelf regions in late spring, usually coincident with the timing of the spring diatom bloom. The continued abundance of C. finmarchicus in these regions is a central concern, as the younger life stages of C. finmarchicus are important prey for the larvae of economically important fish species, such as cod and haddock (Heath and Lough 2007).
1.3 North Atlantic Oscillation
Climate variability can have a significant impact upon C. finmarchicus ' biogeographic range. Changes in water temperature and currents, both at the surface for active copepods and at depth for diapausing copepods, can have a profound effect upon the distribution and abundance of C. finmarchicus (Beaugrand et al. 2002). An important index of these climate changes is the North Atlantic Oscillation index (NAO), which is a measure of the difference between the low pressure center over Iceland and the high pressure center over the Azores. The NAO is most descriptive when averaged over the month or the season, with the average winter NAO index being highly correlated to basin-scale weather patterns. A positive (negative) NAO indicates stronger (weaker) than normal mid-latitude west winds across the Atlantic. This correlates to a variety of

atmospheric and oceanic conditions throughout the northern hemisphere. Increased southwesterly wind flow during positive NAO conditions increases the transport of the
9
North Atlantic Current, which brings warm water into the North Sea and the Norwegian
Sea. Ice flux east of Greenland is reduced, producing less ice cover, and earlier ice break up and retreat in the spring. This warmer water may increase stratification and reduce formation of North Atlantic Deep Water (NADW). On the west side of the North Atlantic basin, increased flow of the North Atlantic Current reduces the interaction of the
Labrador Sea with the rest of the basin, leading to cold, southerly flow in the Labrador
Sea, increased ice flux, and an increased production of Labrador Sea Deep Water.
Negative NAO conditions are nearly the opposite, with reduced westerly winds leading to colder conditions in the North Sea, an increase in Greenland Sea Water and NADW production, and increased ice flux in the East. In the Labrador Sea, there is an influx of warm water, leading to reduced ice flux, reduced Labrador Sea Deep Water production, and an increase in flow of the Labrador Current (Stenseth 2004; Hurrell 1995). Since
1980, the North Atlantic has been mostly in a positive NAO mode, and there is evidence that this may be linked to global climate change and may be representative of future conditions (Paeth et al. 1999; Shindell et al. 1999; Schaeffer et al. 2005).
The NAO index is also correlated to biological responses at multiple trophic levels, in the
North Atlantic. During positive NAO conditions, the increased air and sea surface temperatures cause an earlier retreat of sea ice in the northeast Atlantic, resulting in a lengthening of the growing season for phytoplankton (Reid et al. 1998). Marked effects

10 of the NAO on fish recruitment and size have also been observed. Ottersen et al. (2001) noted an increase in cod recruitment and size in the northeast Atlantic with high positive
NAO conditions, presumably because of temperature effects, but this may also be affected by the availability of zooplankton prey. There is also evidence that the prolonged positive NAO conditions of the last decades correspond to a regime shift in the biology throughout the northeast Atlantic (Reid et al. 1998). This regime shift is characterized by higher phytoplankton productivity, and a change in zooplankton community structure. C. finmarchicus , in particular, has shifted its biogeographic range north, and has been progressively replaced in the North Sea, at the southern extent of its historical biogeographic range, by the more southerly Calanus helgolandicus (Planque and
Fromentin 1996; Heath et al. 1999). This regime shift may be caused by both the increase in temperature and the change in prevailing currents altering the advective transport of zooplankton such as C. finmarchicus . In the northwest Atlantic, the effects of the NAO on C. finmarchicus seem to be the opposite of those seen in the northeast
Atlantic, with C. finmarchicus abundance exhibiting a positive correlation with the NAO
(Greene and Pershing 2000). But, as in the northeast Atlantic, the correlation between the
NAO and C. finmarchicus abundance changed in the 1990‘s. During the 1990‘s, the positive correlation shifted to a weakly negative correlation, with high positive NAO indices producing very little change in C. finmarchicus abundance (Greene et al. 2003).

11
1.4 The potential transport of diapausing C. finmarchicus
In this work, I examine the hypothesis that climate variability, as represented by the
NAO, affects the transport of diapausing C. finmarchicus in the northwest Atlantic, and, thus, affects the timing and location of C. finmarchicus emergence from diapause. Shifts in the timing or location of awakening have potential to create a mismatch with the timing of the peak of the spring phytoplankton bloom, which is essential to the survival and growth of C. finmarchicus . The NAO is the strongest climate signal in the Atlantic, and serves as an indicator of profound changes in ocean circulation patterns, volume flow, and temperature, all three of which may alter the migration distance and duration of diapause in C. finmarchicus .
I investigated the potential transport of diapausing C. finmarchicus using a coupled
Eulerian-Lagrangian model. Flow fields and hydrographic conditions were generated using the Regional Ocean Modeling System (ROMS) and forced by National Centers for
Environmental Prediction/ National Center for Atmospheric Research (NCEP/NCAR) wind and surface fluxes (Kistler et al. 2001). Monthly surface forcings were averaged over two time periods, 1980-1993 and 1958-1971, to create idealized positive NAO and negative NAO surface forcing sets, respectively, for a ROMS simulation of the North
Atlantic (Chaudhuri et al. 2009). The output from these ROMS runs were used to force particle-tracking individual-based models (IBMs) of C. finmarchicus . Individual particles in the IBM were able to interact with their environment in ways that mimic the behavior of C. finmarchicus : descending into diapause, tracking an assigned density

during diapause, and using an internal or external trigger to emerge from diapause and
12 ascend to the surface. We initialized individuals into areas throughout the Labrador Sea, the Scotian Slope, and the open north Atlantic, and tracked the advective trajectories of copepods during diapause. This allowed an evaluation of the potential transport of C. finmarchicus through diapause as a function of the state of the NAO, and an assessment of population connectivity between the Labrador Sea and the Scotian Slope Sea populations of C. finmarchicus .

13
Transport of the copepod, Calanus finmarchicus , in the northwest Atlantic during diapause.
Gregory L. Kowalke and Harold P. Batchelder
To be submitted for publication.

1. Introduction
The biogeographic range of the subarctic copepod, Calanus finmarchicus is constrained
14 largely by temperature. It is possible that global climate change, through changes in temperature and currents may shift the biogeographic range of C. finmarchicus , and, change the ecological dynamics of the North Atlantic (Beaugrand et al. 2002; Beaugrand
2004; Beaugrand and Reid 2003). In fact, C. finmarchicus has declined in its historic biogeographic range in recent years, perhaps related to climate change (Beaugrand et al.
2009; Helaouët and Beaugrand 2007). This could have repercussions for the fisheries of the Georges Bank and Grand Banks, which are near the southern limits of C. finmarchicus ' biogeographic range.
The North Atlantic Oscillation index (NAO) may serve as a proxy for possible climatic conditions of global climate change in the North Atlantic (Hurrell 1995). Since 1980, the
NAO has been largely positive, driving stronger westerly winds, increasing the strength of the Gulf Stream as it crosses the Atlantic and increasing the temperature of waters in the northeast Atlantic, while reducing the exchange of cold Labrador Sea water with the rest of the northwest Atlantic (Stenseth et al. 2004). As anthropogenic climate change increases sea surface temperatures, it is likely that these increased temperatures will drive high positive NAO indices (Paeth et al. 1999; Shindell et al. 1999; Schaeffer et al. 2005).
Therefore, an analysis of the effects of the NAO, and the related changes in temperature and flow upon the biogeographic range of C. finmarchicus may offer a glimpse of the effects of global climate change on this species.

15
C. finmarchicus has an overwintering phase in its lifecycle, similar to diapause in insects, that further constrains its biogeographic range (Hirche 1996). In summer and early autumn, depending on the region, late stage copepodites (C4 and C5) of C. finmarchicus accrue energy stores in the form of wax esters. Then, the C4 and C5 C. finmarchicus copepodites descend to great depths, usually 400-1200m, and enter a quiescent diapause state which may last up to ten months (Hirche 1996). In spring or early summer, C. finmarchicus will rouse from diapause and ascend to the surface, metamorphosing into adults as they ascend.
The trigger for C. finmarchicus ' emergence from diapause is not definitively known.
Several hypotheses have been proposed, which can be divided into two categories: external triggers, which rely on cues from the environment, and internal triggers, which depend on the physiological state of the individual. Miller et al. (1991) proposed that light may act as an external trigger, after noting that diapausing C. finmarchicus awaken when brought into the lab. Both light intensity and photoperiod change dramatically in the spring at higher latitudes, and on sensing this change in light intensity, C. finmarchicus may emerge from diapause and begin its ascent to the surface. The ability of C. finmarchicus to sense the change in light intensity at the depths at which it diapauses has been questioned, considering the depths at which C. finmarchicus diapauses (Irigoien 2004).

Alternatively, internal physiological triggers have been proposed as mechanisms to arouse C. finmarchicus from diapause. C. finmarchicus does not feed during diapause,
16 and must rely on internal wax ester stores for metabolism, but, the wax ester stores must also fuel maturation, gonadal development, and the production of eggs (Hallberg and
Hirche 1980). Although metabolic rate is greatly reduced during diapause, there is evidence that C. finmarchicus does exhibit a significant reduction in oil sac volume during diapause (Saumweber and Durbin 2006). Hirche (1996) suggested that emergence from diapause may be triggered by the depletion of the wax ester stores to some critical amount necessary for maturation. Estimates of the amount of the initial wax ester reserves that are needed for maturation and gonadal development vary, with Rey-Rassat et al. (2002) suggesting that as little as 38% may be necessary, whereas Jónasdóttir
(1999) suggests that C. finmarchicus may require up to 75% for maturation and gonadal development. Saumweber and Durbin (2006) suggest that the reserves necessary for maturation may be body length-dependent.
Our research examines the possible effects of global climate change upon C. finmarchicus in the northwest Atlantic by modeling the advective transport of C. finmarchicus during diapause, in both positive and negative NAO conditions. We investigated the potential transport of diapausing C. finmarchicus using a coupled
Eulerian/Lagrangian model system, tracking particles representing copepods as they are advected through a modeled North Atlantic basin. We sought to address the following questions:


Which diapause emergence trigger provides model organisms with emergence dates that are most consistent with in situ observations of C. finmarchicus emergence
17 times?

How do the amount of initial wax ester stores and diapausing depth affect diapause duration and transport during diapause?

What are the effects of the different phases of the NAO on diapause duration and transport during diapause?

Is there increased transport of diapausing C. finmarchicus from the Labrador Sea to the region south of the Grand Banks in low NAO conditions?
2. Materials and Methods
We used a coupled Eulerian/Lagrangian model to evaluate the potential transport of diapausing C. finmarchicus . Temperatures and velocity were provided by a Regional
Ocean Modeling System (ROMS) simulation by Dr. Avijit Gangopadhay and Ayan
Chaudhuri for the North Atlantic basin. Simulations from this model have been used to evaluate the north-south displacements of the axis of the Gulf Stream and its relationship to warm core rings during differing NAO regimes (Chaudhuri et al. 2009). The
Lagrangian biological model is an individual-based model (IBM), based upon the model developed by Batchelder and Miller (1989) and used in Batchelder et al. (2002), consisting of a particle tracking model with added behaviors.

2.1 ROMS model
The North Atlantic Basin Model (NABM) is described in detail in Chaudhuri et al.
(2009). In brief, ROMS is a free-surface, terrain-following primitive equations ocean modeling system (Song and Haidvogel 1994; Haidvogel et al. 2008). The NABM used
18
National Centers for Environmental Prediction/ National Center for Atmospheric
Research (NCEP/NCAR) wind and surface flux forcing to generate flow fields and hydrographic conditions (Kistler et al. 2001). The 14-year monthly means of the
NCEP/NCAR fields for 1980-1993 (1958-1971) were chosen to represent positive
(negative) NAO conditions. NCEP/NCAR forcings were found to overestimate sensible and latent heat fluxes compared to the Southhampton Ocean Center (SOC) Ocean-
Atmosphere Atlas, so a regression model was used to correct these discrepancies and generate SOC adjusted positive- and negative-NAO monthly means, which were used to generate hydrography and circulation for positive and negative NAO patterns (Chaudhuri et al. 2009). Grid bathymetry was derived from the ETOPO2 topography database for the Gulf Stream region and the ETOPO5 topography database for the rest of the North
Atlantic basin (U.S. Dept. of Commerce, 2006). Ice dynamics and freshwater input were not included in the model. The model has 1/6

horizontal resolution, and 50 vertical sigma levels for a domain from 15

S to 75

N and 100

W to 20

E. The model was spun up for five years, at which time it had achieved steady-state (Chaudhuri et al. 2009).
Model fields for the NABM simulations forced by the negative NAO and positive NAO were saved as snapshots at three day intervals for a year, starting in July and ending in
June, in order to span the time period of diapause. To reduce memory requirements of

the biological IBM, the physical model fields from the NABM were clipped to a region
19 spanning 24ºW to 88ºW, and 31ºN to 75ºN, which includes the Labrador Sea, the Scotian
Shelf, and the Gulf of Maine (Fig. 1).
A similar process was used to generate a NABM simulation for the GLOBEC years of
1995-1998 using monthly forcing from the SOC Ocean-Atmosphere Atlas (Chaudhuri et al. 2009). Hydrographic data (temperature and salinity) fields from this simulation were compared to observed temperature and salinity data from the World Ocean Database 2005
(WOD05) (Boyer et al. 2006). Approximately 4000 points from approximately 250 profiles, spanning the GLOBEC years were selected from the 'verified' data in the
WOD05 and compared with NABM temperature and salinity interpolated to the WOD05 collection points (Fig. 2). Surface temperatures were often underestimated, and surface salinities overestimated by the NABM. At diapausing depths of 300-900m within the region of interest, however, NABM temperatures were within 1°C, and NABM salinities were within 1 PSU of WOD05 data (Fig. 3).
2.2 IBM model
The IBM model is a particle-tracking model, wherein the particles are able to react to and interact with the environment, mimicking the behavior of diapausing copepods
(Batchelder and Miller 1989; Batchelder et al. 2002). With a one hour time step, the model uses Runge-Kutta 4 th
order integration to move the copepods using the u, v and w velocities interpolated to the three-dimensional position of the copepod from the two

20
NABM snapshots that bracket the current time. Physical transport is assumed to occur by advection only; no random walk of smaller scale turbulence or diffusion is included.
Vertical movement was constrained to maintain copepods between 10 meters below the surface and at least 10 m above the bottom. Copepods were allowed to swim vertically to descend from the surface at the initiation of diapause and to ascend back to the surface following termination of diapause. We did not model the mechanism of entrance into diapause; instead, copepods in the specific simulations descended towards diapause depths on predetermined calendar dates. Sampling in the Gulf of Maine and Scotian
Shelf suggests that the onset of diapause in Calanus finmarchicus is variable between years and regions and while only roughly approximated due to low sampling frequency, the earliest onset of diapause appeared to be June on the Scotian Shelf and the latest onset of diapause was in September on the Newfoundland Shelf (Johnson et al. 2008).
Although there are few data for the Labrador Sea, Pepin and Head (2009) found C. finmarchicus near the surface at stations sampled in the Labrador Sea in June, and a larger proportion of C. finmarchicus at depths greater than 250 m in September, indicating that many copepods had entered diapause. Based upon this information, we examined the transport during diapause for copepods that entered diapause on July 15 th and August 15 th
by descending from an initial depth of 10m at 7.5 m day
-1
, slightly slower than observed ascent rates from diapause (Heath 1999). Each copepod was preassigned a target seawater density to which they would descend and remain during diapause. Rather than fix the depth of diapause, we fixed the isopycnal at which diapause

21 occurred, following the findings of Johnson et al. (2006) that showed higher retention of diapausing C. finmarchicus that followed a specified isopycnal than C. finmarchicus that diapaused at a constant depth in the Gulf of Maine. Pre-assigned isopycnals were distributed normally around 1030 kg m
-3
, with a standard deviation of 1 kg m
-3
, as that was the observed boundary of the Labrador Sea Deep Water within the NABM model.
Density was calculated in the IBM at each timestep using temperature and salinity from the NABM and the UNESCO 1983 equations of state (Fofonoff and Millard 1983).
Upon emergence from diapause, copepods ascend to the surface at a rate of 15m d
-1
(Heath 1999).
Metabolic consumption of wax ester reserves began when the copepod reached and began tracking its pre-assigned isopycnal. Each copepod was assigned a prosome length (an estimate of individual size) before the IBM simulations were run. The assigned prosome lengths were normally distributed about a mean of 2.7 mm (range ca. 2.4-3.0 mm), with a standard deviation of 0.125mm, corresponding to the distribution of C. finmarchicus C5 prosome lengths in the Labrador Sea in June (Pepin and Head 2009). Prosome lengths were randomly generated at each station. Prosome length (L; mm) was used to estimate the maximum (initial) oil sac volume (OSV; µL) and the maximum wax ester carbon weight (WE; µgC) using equations 1 and 2, (equations (6) and (12) from Saumweber and
Durbin 2006).
𝑂𝑆𝑉 𝑚𝑎𝑥
= 0.274052 𝐿 2 − 0.5987𝐿 + 0.254568
𝑊𝐸 = 0.74 × (
900 ×𝑂𝑆𝑉+10.8
1.44
) eq. 1 eq. 2

We assumed that all individuals initiated diapause with the maximum possible length
22 specific wax ester. Metabolism reduced the wax ester carbon content (WE t
; µgC) as
𝑊𝐸 𝑡
= 𝑊𝐸 𝑡−1
− 𝑅 ∙ 𝐶 𝑡𝑜𝑡𝑎𝑙
∙ ∆𝑡 , eq. 3 where R is the rate of wax ester carbon metabolism (µgC·µgC
-1
hr
-1
) estimated using the temperature (T; °C) and weight dependent C. finmarchicus diapause metabolism equation developed by Saumweber and Durbin (2006), adapted to total carbon content (equation
4).
𝑅 =
0.74×12.011× 10
10
(0.0412 ×𝑇+1.1585 )
6 eq. 4
Total copepod carbon content (C total
; µgC) was calculated from OSV using equation 5
(adapted from Figure 5 in Saumweber and Durbin 2006).
𝐶 𝑡𝑜𝑡𝑎𝑙
= 374.99 × 𝑂𝑆𝑉 0.536
eq. 5
2.3 Diapause emergence triggers within the IBM
We examined four potential diapause emergence triggers/mechanisms: one external and three internal. The external emergence trigger was based on photoperiod as suggested by
Miller et al. (1991). At each time step, local daylength was calculated for each copepod using equations from Brock (1981). The local daylength at winter solstice was stored for each copepod, and an individual‘s emergence from diapause was triggered when the current local daylength exceeded the stored solstice daylength by 20%.

23
Two of the internal diapause emergence triggers tested were similar, differing only in the threshold level of wax ester reserves that triggered arousal. We tested two different levels of wax esters held in reserve for maturation and gonadal development: the 75% suggested by Jonasdottir (1999), and the 38% suggested by Rey-Rassat et al. (2002). When the wax ester stores of the copepod reached 75% or 38% of the original weight, the copepod emerged from diapause and initiated ascent.
The last internal trigger was based on the observations of Saumweber and Durbin (2006) that C. finmarchicus aroused from diapause and began its ascent to the surface when the
OSV decreased below a minimum threshold (OSVmin; µL). Minimum OSV was calculated based on prosome length (L; mm) using equation 6 (Saumweber and Durbin
2006).
𝑂𝑆𝑉 𝑚𝑖𝑛
= 0.279779 𝐿 2 − 1.06674𝐿 + 1.060073
eq. 6
Modeling experiments comparing C. finmarchicus diapause emergence triggers were tested for the August 15 th
diapause entrance date in positive NAO conditions. A total of
100 copepods representing all combinations of 20 prosome lengths, distributed as described earlier, and 5 pre-assigned isopycnals to simulate different diapause depths, as described earlier, were seeded at 10 m deep at each of 12 locations at station B on August
15 th
(Fig. 4). The 1200 copepods were allowed to drift and enter and emerge from diapause for 300 days. Aside from the initial descent and final ascent, copepods moved only by advection on an isopycnal from the modeled u and v fields. The dates that the

24 copepods reached the surface after emerging from diapause were compared with reported
C. finmarchicus surfacing dates (Planque et al. 1997; Head et al. 2000) to determine which trigger produced realistic emergence dates.
2.4 Transport of diapausing C. finmarchicus in the northwest Atlantic
The diapause emergence trigger based upon internal oil storage volume of Saumweber and Durbin (2006) produced the most realistic emergence dates (see Results for details), and was the only trigger mechanism used in simulations to evaluate the transport of C. finmarchicus during diapause in the northwest Atlantic. Copepods were seeded at all stations in Figure 4 for diapause entry dates of July 15 th
and August 15 th
. At each seed station, 1200 individuals were seeded for all combinations of 12 starting locations (Fig.
4), 20 pre-assigned prosome lengths, and 5 pre-assigned diapausing isopycnals, as described for the earlier tests of different trigger mechanisms. These simulations were run for both positive and negative NAO conditions for up to 300 days from the date of diapause entry.
The results from the IBM were analyzed for mean distance traveled, mean diapause duration, and mean total migration duration for both positive and negative NAO conditions. Differences in these statistics between diapause onset dates, seeding location, mean depth during diapause, and starting weight were examined with paired t-tests, where appropriate.

25
3. Results
3.1 Comparison of C. finmarchicus diapause emergence triggers
Efficacy of the four arousal trigger mechanisms was evaluated by comparing the mean date that copepods appeared at the surface in the IBM with observed dates based on high surface concentrations of C. finmarchicus in the Labrador Sea. C. finmarchicus adults emerge in surface waters in the Labrador Sea in April/May (Head et al. 2000). Mean surfacing dates for each proposed trigger are shown in Figure 5. The trigger based on lengthening daylengths (photoperiod trigger) caused model copepods to emerge in a latitudinal pattern opposite to that expected from field observations, with copepods surfacing in February in the northernmost regions (Davis Channel), and emerging in
March off of the Grand Banks (Fig. 5a). This result is contrary to latitudinal patterns of emergence dates noted by Heath et al. (2000) and Miller et al. (1991).
The internal trigger proposed by Jonasdottir (1999), wherein a copepod emerges from diapause when wax ester stores are depleted to 75% of the original content, caused copepods to emerge in September through November of the same year that they descended in the IBM model (Fig. 5b). This is ca. 6 months earlier than the data suggest reproductive adults emerge at the surface. The internal trigger based on a reduction of wax ester stores to 38% of original content, as suggested by Rey-Rassat et al. (2002) approximated observed surfacing dates with a mean surfacing on DOY 90 (ca. April 1), and a standard deviation of 15 days (Fig. 5c). Copepods triggered to exit diapause by depleting wax ester stores to the minimum length-dependent content observed by

Saumweber and Durbin (2006) had mean surfacing on DOY 100 (ca. April 10), and a
26 slightly narrower distribution with standard deviation of 9 days (Fig. 5d). The diapause emergence trigger suggested by Saumweber and Durbin (2006) exhibited the latitudinal gradient and local synchrony expected from in situ observations. Based on these results, the Saumweber and Durbin (2006) trigger mechanism provided surfacing dates that best matched the field data (Miller et al. 1991, Heath et al. 2000); subsequent simulations that examined the effects of diapause depth, animal size (related to lipid stores), location of diapause entry, date of diapause entry, and NAO phase on diapause duration and transport only used the Saumweber and Durbin (2006) trigger.
3.2 Dispersal of C. finmarchicus during diapause.
Overall, particles representing C. finmarchicus transported by the NABM velocities followed the prevailing currents expected in the region as described by Luo et al. (2006) and Han et al. (2008). Diapausing copepods seeded on the slope of the Labrador Sea followed the counter-clockwise Labrador Current around the perimeter of the basin (Fig.
6). A small number of diapausing copepods seeded on the eastern slope near Greenland
(Station B) were advected into the Davis Channel (Fig. 7c,d). Copepods seeded on the western Labrador slope (Stations D and E) were advected along the slope and were deflected into the Atlantic Basin by the Flemish Cap (Fig. 7g-j), with a small number passing southward through the Flemish Pass, and surfacing near the Grand Banks (Fig.
7g). Copepods seeded in the central Labrador Sea (Station F) were largely retained within the basin, with some copepods escaping the slow central current and following the

27
Labrador Current (Fig. 7k,l). Copepods seeded to the southeast of Greenland (Station H) headed northeast into the Irminger Basin (Fig. 7o,p), whereas copepods seeded southwest of Greenland (Station G), remained relatively near their initial location (Fig. 7m,n).
Copepods seeded on the Scotian Slope (Station I), traveled northeastward along the slope, and were not entrained into or advected across the basin by the Gulf Stream prior to resurfacing (Fig. 7q,r).
3.3 Duration of C. finmarchicus diapause
There was little difference in mean diapause duration between copepod seeding regions in the Labrador Sea and the open Atlantic (stations A through H). Copepods seeded on the Scotian Slope had a shorter mean diapause duration, with mean diapause duration of
34 days during positive NAO conditions (Table 1). The longest mean diapause duration was at the west Labrador Sea Slope seeding location (station D) during positive NAO conditions, at 117 days. At all locations, the standard deviation of mean diapause duration was roughly 15 days (Table 1). The variability of the mean, as represented by the standard deviation, within a station and single simulation is due to the different conditions experienced by individuals having various lengths, which determine wax ester stores, and diapause isopycnals, which determine depth and temperature experienced.
Mean total migration duration (which includes time in descent, diapause and ascent) differed much less between regions in the Labrador Sea and open Atlantic Ocean, with a high mean total migration duration at station B of 238 days during positive NAO conditions, and a low mean total migration duration at station A of 215 days during

positive NAO conditions (Table 1). The standard deviations at all locations in the
Labrador Sea were approximately 7 days. The mean total migration duration of 158
28 days on the Scotian Slope (Station I) was substantially shorter, caused by the warmer temperatures at diapausing depths depleting wax ester reserves more rapidly.
The state of the NAO index had a small, but significant, effect on the mean diapause duration and the mean total migration duration, with positive NAO conditions causing an increase of 5 days in both mean diapause duration and mean total migration duration.
The standard deviation of mean diapause duration for both positive and negative NAO conditions (considering Stations A through I) was approximately 28 days, and the standard deviation of mean total migration duration for both conditions was 22.5 days.
C. finmarchicus copepods initiated on the Scotian Slope (Station I) had significantly shorter diapause duration than any other station. The similarity between positive and negative NAO conditions was reflected at each starting location, starting date, starting weight, and mean depth (Table 1).
Start date had little effect on mean diapause duration or mean total migration duration
(Table 1). The mean day of the year (DOY) for the end of migration is, thus, one month apart for the July 15 th
and August 15 th
start dates, with copepods seeded in the Labrador
Sea on July 15 th
surfacing in March/April, and copepods seeded on August 15 th
surfacing in April/May (Table 2). On the Scotian Slope (Station I), copepods seeded on July 15 th surfaced in late January, and copepods seeded on August 15 th
surfaced in late February.

Weight, as determined by prosome length, had a predictable effect on mean diapause
29 duration and mean total migration duration, with lighter copepods, having smaller initial oil reserves, emerging earlier from diapause than heavier copepods (Table 1). But the effect of weight was relatively minor, with a difference of approximately 10 days between light and heavy copepods for both mean diapause duration and mean total migration duration.
Mean diapause depth had an impact on mean diapause duration, with deep diapausing copepods (greater than 350 m) diapausing for approximately 30 days less than shallow diapausing copepods (less than 350 m) (Table 1). Mean annual temperatures at depths greater than 350m are 1-3°C warmer than mean annual temperatures near the surface in the Labrador Sea, due to colder sea surface temperatures during winter. The warmer temperatures at depth in the Labrador Sea shorten diapause duration in deep diapausing copepods. Deeper diapausing copepods require more time to migrate to diapause depth and longer transit time to the surface after emergence from diapause, resulting in a difference between mean total migration duration of 10 days (Table 1). Deep diapausing copepods, therefore spent proportionately more time in transit, and less time in diapause, which aids the greater transport discussed in the next section.

3.4 Potential transport distances of diapausing C. finmarchicus
Seeding location had a significant effect on total distance traveled (Table 1). Copepods
30 seeded in the center of the Labrador Sea (Station F), the western open Atlantic (Station
G), the eastern open Atlantic (Station H) and the Scotian Slope (Station I) had the least transport, on average traveling 320 km to 590 km in total, with a mean net displacement of 150 km to 400 km. Copepods seeded on the slope near the midline of the Labrador
Sea basin (Station D) traveled the greatest distance, with mean total distance traveled of
2400 km during positive NAO conditions, and a mean net displacement of ca. 1500 km.
Some copepods seeded at stations B, C, and D were able to pass through the Flemish Pass
(Fig. 6). A fraction of the copepods from Station B and C were advected though the
Flemish Pass during negative NAO conditions, and a smaller fraction of copepods from
Station D passed through the Flemish Pass during positive NAO conditions. During negative NAO conditions 6% (Station B, July 15 th
release date) to 26% (Station C,
August 15 th
release date) of the copepods were advected though the Flemish Pass. All of the copepods from Station B and C advected through the Flemish Pass were in the ascent phase after termination of diapause. Only 3% (July 15 th
) to 8% (August 15 th
) of released copepods from Station D negotiated the Flemish Pass during diapause in positive NAO conditions. More copepods were advected through the Flemish Pass during negative
NAO conditions. However, the phase of the NAO had little effect on the mean total transport distance and the mean net displacement of the diapausing copepods, with differences due to the phase of the NAO being less than 60 km (Table 1). In the eastern open Atlantic (Station H), there were significant differences in both the mean distance

31 traveled, and the mean net due to the phase of the NAO (Fig. 7k,l). On the Scotian Slope, the northeastward transport was greater during the negative phase of the NAO (Fig. 7q,r).
The diapausing depth, as determined by the pre-assigned isopycnal (1030 kg m
-3
± 1 kg m
-3
), had an effect on the mean distance traveled (Table 1). Deeper diapausing copepods
(mean depth greater than 350 m) were advected greater distances than shallower diapausing copepods (mean depth less than 350 m). Differences between the mean net displacements of deep diapausing copepods and shallow diapausing copepods were up to
350 km, depending on the state of the NAO (Table 1). This is due to the increased transport during descent and ascent experienced by deep diapausing copepods.
There was an interaction between the phase of the NAO and release date: shallow diapausing copepods released on August 15 th
at Station A traveled significantly farther than deep diapausing copepods during positive NAO conditions whereas shallow diapausing copepods released on July 15 th
travelled a shorter distance than deep diapausing copepods. However, copepods released on August 15 th
at Station A in negative NAO conditions traveled a significantly shorter distance than copepods released on July 15 th under the same condition (Table 1).
Start date had a small, but statistically significant effect on distance traveled (Table 1).
Overall, copepods that began their descent on August 15 th
travelled farther than copepods released on July 15 th
, but, the difference in mean net displacement was less than 38 km.

Starting weight, as determined by prosome length, had an effect on mean distance travelled (Table 1). Heavy copepods, with a prosome length greater than 2.8 mm, travelled 96km (NAO-) to 171km (NAO+) farther (mean total distance traveled) than lighter copepods.
32
3.5 Multiple regression exploration of controls on diapause duration and transport
We ran a series of stepwise multiple regressions to determine the relative importance of each variable to duration and distance traveled. Prosome length, starting location, and depth, as indicated by the pre-assigned isopycnal, had the greatest effect on diapause duration, but condition of the NAO and release date increased the correlation of the multiple regression as well (r
2
= 0.6, p < 0.01; Table 3). The results of the stepwise multiple regression were much the same for total migration duration, except that depth had a greater effect than the starting location, and the condition of the NAO was of greater significance (r
2
= 0.8, p < 0.01; Table 3). Prosome length, starting location, and depth, as determined by the pre-assigned isopycnal, had the most significant effect on the total distance traveled; the condition of the NAO was not significant for total distance traveled (r
2
= 0.3, p < 0.01; Table 3). For net displacement, condition of the NAO had the most significant effect, followed by the prosome length, and the starting location (r
2
=
0.4, p < 0.01; Table 3). Overall, the stepwise multiple regressions indicate that prosome length, which determines the starting weight of wax ester stores, had the greatest effect on diapause duration and distance traveled, followed by starting location, and diapause depth, as determined by the pre-assigned isopycnal.

4. Discussion
In our IBM simulations, the photoperiod diapause emergence trigger based on Miller et
33 al. (1991) caused copepods to emerge in early February in the Davis Strait, and mid-
March near the Grand Banks (Fig. 7), which is not consistent with observed latitudinal gradients in surfacing dates (Miller et al. 1991, Heath et al. 2000). Moreover, a photoperiod trigger seems unlikely, because of the suggestion of Hind et al. (2000) that it is doubtful that C. finmarchicus can detect photoperiod changes at diapause depth, and because diapausing copepodites in the Labrador Sea can potentially be transported southward more than 20º of latitude (Fig. 6). This represents an enormous potential change in the timing of the spring phytoplankton bloom, which would not be well cued to by the winter solstice daylength at much higher latitudes. Two of the internal wax ester stores triggers produced emergence dates similar to those found in the field. The 75% wax ester reserve trigger proposed by Jonasdottir (1999) did not; copepods emerged 4 to
6 months too early. The method based on Rey-Rassat et al. (2002) 38% wax ester reserve trigger produced reasonable emergent times when used with the metabolism equation in the IBM. But, overall, the internal, lipid-based, emergence trigger based upon
Saumweber and Durbin (2006) exhibited the best agreement with field observations of surfacing and reproduction as a function of latitude (Planque et al. 1997; Miller et al.
1991). It was not necessary to impose an external latitudinal effect, where copepods emerged at set dates based on location, as in the models of Speirs et al. (2005) or
Tittensor et al. (2003) to achieve realistic latitudinal patterns in C. finmarchicus emergence. Within our model, the date of surface appearance after diapause termination

for C. finmarchicus is determined by the diapause entry date, the quantity of wax ester
34 stores, the minimum amount of wax ester stores that trigger emergence from diapause, and the temperature at diapause depths. In this model system, C. finmarchicus C5 copepodites were assumed to begin diapause with the maximum length-specific wax ester stores (according to Saumweber and Durbin, 2005). If copepods enter diapause with less than the length-specific maximum reserves, they would emerge from diapause earlier than predicted in these simulations. Descent rate and ascent rates used in the model also strongly impacted the total migration duration. Descent and ascent together accounted for more than 100 days (ca. 40%) of the total migration duration at all stations. Although based on ascent rates measured in the field, the descent and ascent rates are relatively slow, compared to typical diurnal vertical migration (e.g. swimming) rates (Heath 1999).
Most of what is known about the duration of diapausing C. finmarchicus in the Labrador
Sea is derived from observations of the dates that C. finmarchicus enters and emerges from diapause, and the depths that they can be commonly found during diapause.
Therefore the best available method of comparing the results of the IBM model with in situ conditions is by looking at the trends in surfacing dates. There is remarkable synchrony in emergence from diapause across the basin, with most C. finmarchicus emerging in March/April (Planque et al. 1997; Head et al. 2000). Planque et al. (1997), using Continuous Plankton Recorder data, delineated three areas of the North Atlantic where C. finmarchicus have seasonally different peak abundance patterns: (1) a
‗dominant‘ North Atlantic region, (2) the Labrador Sea, and (3) south of Newfoundland.

35
C. finmarchicus south of Newfoundland emerged from diapause and reached peak abundance earlier than C. finmarchicus in the dominant region. Conversely, C. finmarchicus in the Labrador Sea region reached peak surface abundance later than in the dominant region, but emerged from diapause at approximately the same time of year as in the dominant region (Planque et al. 1997). Spatial variation in emergence times from the
IBM model (Fig. 8) is similar to that found in the map of emergence based on field observations (Planque et al. 1997; their Figure 3). Additionally, within our model, copepods that surfaced in the Davis Strait emerged later than copepods throughout the
Labrador basin, which is similar to findings by Head and Pepin (2008). Model copepods seeded on the Scotian Slope emerged early in the year, as found by Johnson et al. (2008) and Planque et al. (1997).
We have developed a novel approach to estimate transport of C. finmarchicus during diapause in the north Atlantic. Previous models did not address the emergence trigger of diapause, and specified fixed dates of diapause emergence, either throughout the study region, as in Speirs et al. (2005), or with latitude dependence, as in Tittensor et al. (2003).
In addition, previous models did not include transports that occurred during descent and ascent. Johnson et al. (2008) modeled transport during diapause in C. finmarchicus , and, although they discuss the differing currents that an ascending copepod encounters, they only track diapausing copepods at depth. This model demonstrates that with ascent rates from field measurements ( Heath 1999) and assumed descent rates, the transport during ascent and descent can be considerable.

The NAO has been shown to influence North Atlantic temperatures and the strength of
36 the Labrador Slope Current (Greene and Pershing 2003) that transports C. finmarchicus throughout the basin during diapause, but, according to this study, the effects of the NAO on diapausing copepods are minimal. Considering that temperature at diapausing depths has a direct influence on diapause duration through the metabolic consumption of lipid reserves, and that typical diapausing depths are within the Labrador Slope Current, we hypothesized that during negative NAO periods when the Labrador Slope Current is stronger than usual and temperatures colder than usual, that southward transport of diapausing C. finmarchicus would be greater. Our Lagrangian simulations using the
NABM velocity and temperature fields suggest transports of copepods seeded initially in the Labrador Sea differ little with NAO phase. Transport of copepods seeded in oceanic regions more to the east (Stations G and H in Figure 4) and on the Scotian slope region
(Station I) exhibited more significant differences due to the phase of the NAO. Negative
NAO conditions transported Scotian slope diapausing copepods in the IBM much further northeast, as far as the Grand Banks, than transports during positive NAO (Fig. 6).
Negative NAO conditions also produced greater northeastern transport in the north
Atlantic southeast of Greenland. This suggests that, as indicated by previous research, the condition of the NAO has an effect on transport during diapause in the eastern
Atlantic, and in the Gulf of Maine and Scotian Shelf regions, but, little difference in the
Labrador Sea (Greene et al. 2003; Reid et al. 1998; Beaugrand et al. 2009).

37
The effects of the phase of the NAO on C. finmarchicus transport during diapause shown above are dependent on the assumption that the NABM adequately mimics the effects of the NAO in the real ocean. The NABM was developed to characterize the NAO related differences in the position of the north wall of the Gulf Stream. The Gulf Stream in the
NABM is further south during negative NAO conditions, which is consistent with longterm observations (Chaudhuri et al. 2009). However, the ability of the NABM to model all of the processes that contribute to the dynamics of the surface Labrador Current and the subsurface Labrador Slope Current, which are integral to the movement of C. finmarchicus in the northwest Atlantic, can be questioned. The NABM physical simulations do not include freshwater runoff or ice dynamics, which can affect stratification and deep water formation, and, consequently, currents throughout the water column (Hakkinen and Mellor 1992; Budgell 2005). As the copepods in this model follow the Labrador Sea Deep Water, alteration of the formation and transport of this deep water may impact the transport of diapausing C. finmarchicus . Increased stratification and decreased bottom water formation would likely reduce transport of diapausing C. finmarchicus and increase retention.
Greene et al. (2003) noted recent changes in the correlative effects of the NAO on C. finmarchicus abundance in the Northwest Atlantic, wherein it switched from a positive correlation to a negative correlation in the early 1990‘s. The results of this study suggest that the prevailing positive NAO conditions in recent years, which may be associated with global warming, will not result in substantial transport-related change in the range of

C. finmarchicus in the northwest Atlantic, assuming that the physical model adequately
38 represents the real ocean. But, if the surfacing date and location of recently aroused C. finmarchicus are altered because of changes in current or temperature they encounter, or through changes in the ability of C. finmarchicus to accrue and store lipids for diapause, then there is potential for temporal or spatial mismatch to occur between the peak of phytoplankton blooms and peak abundance of C. finmarchicus .

39

40

41

Table 3. Results of stepwise multiple regression for diapause duration, total migration duration, total distance traveled and net distance traveled. Latitude and longitude indicates the starting position of the copepod. Order indicates the order that variables were added to the stepwise multiple linear regression. Analysed with Matlab 7.0.
42
Diapause Duration r
2
= 0.61
NAO+/-
F = 10744 Release Date p < 0.01
Prosome L
Isopycnal
Latitude
Longitude
Total Migration Duration r
2
= 0.77
NAO+/-
F = 23206 p < 0.01
Release Date
Prosome L
Isopycnal
Latitude
Longitude
Total Distance Traveled r
2
= 0.35
NAO+/-
F = 4395 p < 0.01
Release Date
Prosome L
Isopycnal
Latitude
Longitude
Net Distance Traveled r
2
= 0.39
NAO+/-
F = 4495 Release Date p < 0.01
Prosome L
Isopycnal
Latitude
Longitude
5
3
4
1
6
2
5
3
6
1
2
Order
5
6
1
4
2
1
4
2
3
5
3
4 s.e.
6.6
0
32.0
1.2
0.5
0.6
s.e.
4.7
0
320
1.2
0.5
0.6
s.e.
2.6
0
12.4
0.5
0.2
0.2
s.e.
4.1
0
20.2
0.8
0.3
0.4
Beta
5.1
9.1x10
-6
459.0
-9.3
-54.7
63.5
Beta
53.9
6.2x10
-6
224.3
-6.5
-41.0
51.7
Beta
-66.4
1.0x10
-5
1068.3
-12.8
4.0
89.2
Beta
-118.8
2.8x10
-5
919.5
12.4
11.5
78.1
t
0.8
3.6
14.3
-7.8
-109.8
109.9
t
0.8
3.6
22.6
0.8
0.4
0.4
t
-46.5
-28.4
74.0
26.9
59.3
348.0
t
-16.0
6.4
6.4
-17.1
12.7
245.3
p-value
0.4
<0.01
<0.01
<0.01
<0.01
<0.01
p-value
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
p-value
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
p-value
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01

Figure 1. Region of interest for the NABM ROMS physical model (dark line) and the
IBM biological model (dashed line).
43

A
44
B
Figure 2. Spatial (A) and temporal (B) distribution of WOD05 temperature data used for comparison with output from the 1995-1998 NABM ROMS model.

A
B
Figure 3. Difference plot of NABM 1995-1998 output versus WOD05 for temperature
(A), and salinity (B). CTD data from the WOD05 collected during the time period between August 1, 1995 and May 30, 1998 within the model domain was compared to
NABM output extrapolated to the same latitude, longitude and depth. Plot is of the difference between observed results from the NABM model and the expected results of
WOD05 CTD casts, versus depth.
45

C
D
F
B
A
G
E
H
I
Figure 4. Seeding locations of particles representing C. finmarchicus in the IBM biological model. Labeled stations consist of twelve equidistant start points where 100 particles representing copepods were seeded.
46

47
A B
C D
Figure 5. Mean surfacing dates, binned to the nearest whole degree of latitude and longitude, for four different diapause emergent triggers: photoperiod (A), 75% wax ester kept in reserve (B), 38% wax ester kept in reserve (C), and Saumweber and Durbin
(2006) prosome length dependent emergence (D). Note that the negative dates in (B) indicate surfacing dates that occurred in the same year as the copepods were released.

A
NAO+
B
NAO-
48
C D
Figure 6. Composite tracks of copepods released at all stations in positive NAO conditions (A & C) and negative NAO conditions (B & D) and with a July 15 th
release date (A & B) and August 15 th
release date (C & D). There are nine tracks displayed for each site, constituting all combinations of the longest, median, and shortest prosome lengths, and the lowest, median and highest pre-assigned isopycnal. Red tracks are copepods descending to pre-assigned density, blue tracks are diapausing copepods, and green tracks are copepods emerging from diapause and ascending to the surface.

A
NAO+
B
NAO-
49
C D
Figure 7. Tracks of copepods released on August 15 th
at station A for positive NAO conditions (A) and negative NAO conditions (B), and station B for positive NAO (C) and negative NAO (D). Red tracks are copepods descending to pre-assigned density, blue tracks are diapausing copepods, and green tracks are copepods emerging from diapause and ascending to the surface. Tracks for copepods released on July 15 th
were similar, and, therefore, are not shown (see Figure 6).

E
NAO+
F
NAO-
50
G H
Figure 7 (continued). Tracks of copepods released on August 15 th
at station C for positive NAO conditions (E) and negative NAO conditions (F), and station D for positive
NAO (G) and positive NAO (H). Red tracks are copepods descending to pre-assigned density, blue tracks are diapausing copepods, and green tracks are copepods emerging from diapause and ascending to the surface. Tracks for copepods released on July 15 th were similar, and, therefore, are not shown (see Figure 6).

I
NAO+
J
NAO-
51
K L
Figure 7 (continued). Tracks of copepods released on August 15 th
at station E for positive
NAO conditions (I) and negative NAO conditions (J), and station F for positive NAO (K) and negative NAO (L). Red tracks are copepods descending to pre-assigned density, blue tracks are diapausing copepods, and green tracks are copepods emerging from diapause and ascending to the surface. Tracks for copepods released on July 15 th
were similar, and, therefore, are not shown (see Figure 6).

M
NAO+
N
NAO-
52
O P
Figure 7 (continued). Tracks of copepods released on August 15 th
at station G for positive NAO conditions (M) and negative NAO conditions (N), and station H for positive NAO (O) and negative NAO (P). Red tracks are copepods descending to preassigned density, blue tracks are diapausing copepods, and green tracks are copepods emerging from diapause and ascending to the surface. Tracks for copepods released on
July 15 th
were similar, and, therefore, are not shown (see Figure 6).

Q
NAO+
R
NAO-
53
Figure 7 (continued). Tracks of copepods released on August 15 th
at station I for positive
NAO conditions (Q) and negative NAO conditions (R). Red tracks are copepods descending to pre-assigned density, blue tracks are diapausing copepods, and green tracks are copepods emerging from diapause and ascending to the surface. Tracks for copepods released on July 15 th
were similar, and, therefore, are not shown (see Figure 6).

A B
54
C D
Figure 8. Mean day of the year (DOY) that copepods reached near surface waters.
Surface locations were rounded to the nearest whole degree of latitude and longitude, and dates averaged for each point. This plot shows mean DOY of surfacing for all copepods seeded on July 15 th
for positive (A) and negative (B) NAO conditions, and mean DOY of surfacing for all copepods seeded on August 15 th
for positive (C) and negative (D) NAO conditions. Note the negative surfacing dates in (A) and (B), which indicate December surfacing.

3.
G
ENERAL CONCLUSION
I examined a variety of hypothesized triggers for C. finmarchicus diapause termination
55 and found that the internal lipid-based trigger described by Saumweber and Durbin
(2006) provided the best agreement with in situ observations of emergence dates in the coupled NABM and IBM system.
The initial wax ester stores, as determined by prosome length, have a significant effect on diapause duration and, thus, transport during diapause, and are likely the foremost influence on the duration of diapause. The diapausing depth, as determined by the pycnoclines that a copepod follows, had less of an effect upon diapause duration and distance transported during diapause, but was still significant.
The effects of the phase of the NAO on the duration and transport during diapause were small but significant. The phase of the NAO had the greatest effect on the net distance transported. It also had an effect on distance and direction of transport on the Scotian
Shelf, and in the open ocean east of the Labrador Sea.
In our simulations, no copepods from the Labrador Sea stations were able to transit to the
Slope Sea region off of the Nova Scotian Shelf in a single diapausing generation. A small number of copepods passed through the Flemish Pass while in diapause or in ascent from diapause, but they were not transported far enough south to contribute significantly to the
Gulf of Maine and Scotian Shelf populations. These copepods also emerged from

diapause significantly later than the Scotian Shelf population, which may reduce the
56 impact of any interaction between the Labrador Sea and Scotian Shelf populations. This suggests that any connectivity between the C. finmarchicus populations in the Labrador
Sea and populations in the Slope Sea and Gulf of Maine does not occur through transport during diapause. If there is connectivity between the populations, it may be through southward transport by the Labrador Current or across the Grand Banks during the development of the surface active generation produced by the diapausing copepods.
Multiple surface active generations would increase the probability that connectivity would occur.
The IBM I used for this study used ascent and descent rates for copepods entering and emerging from diapause that were considerably slower than vertical swimming rates observed for diurnally migrating copepods. Despite being based upon in situ observations, the rates used in the IBM model contributed substantially to distance transported, and the total migration duration. The effects of varying ascent and descent rates upon transport needs further investigation.
It would be useful to extend the scope of the IBM beyond diapause transport. The results of this study point to possible vectors for gene exchange between populations of C. finmarchicus . Our model suggests that copepods from the three regions outlined by
(Bucklin et al. 2000) may interact at the margins of these populations, but, these interactions are likely to take multiple generations and multiple years. Transport of

57 diapausing C. finmarchicus from the Labrador Sea to the Slope Sea/Scotian Shelf region, did not occur in the NABM transport fields for either phase of the NAO. Modifying the
IBM to simulate both the surface active and diapause phase of C. finmarchicus over multiple years may provide additional insight into the extent and interaction of the biogeographic range of C. finmarchicus populations, but will require better information on the cues for descent and entry into diapause.

4.
A
DDENDUM
Since defending this document, it has come to the attention of Dr. Batchelder and I that
58 there are substantial errors in the equations of Saumweber and Durbin (2006). We found inconsistencies between their equations and some of the results presented in their paper.
We contacted Dr. Saumweber about these inconsistencies, which initiated several emails that clarified the typographic errors in the paper. The coefficients of their equation (9) were incorrectly reported; the correct equation for N total
is:
𝑁 = 1.63287𝐿 3 + 18.80358 × 𝑂𝑆𝑉 − 0.94817
eq. 7 where L is prosome length (mm) and OSV is the oil sac volume (μL). This correction should not impact the research presented here, as we based our model on carbon body weights using an equation (our equation 5; page 22) derived from Fig. 5 of Saumweber and Durbin (2006). Saumweber and Durbin used a different equation (their Eqn. 8),
𝐶 = 7.641879 𝐿 3 + 0.723668 374.99 𝑂𝑆𝑉 0.536
− 54.7409
eq. 8
Where L and OSV are as described above. Both the C total
equation in Saumweber and
Durbin (eqn. 8 above) and the equation we used (Eqn. 5; p. 22) yield maximum total body carbon weights, including the wax ester carbon, that are smaller than the maximum initial estimated wax ester carbon weights(from Eqn. 1) for large sized copepods. This is clearly unrealistic. For the Saumweber and Durbin equation, this occurs when prosome lengths are greater than 2.8 mm; for the Kowalke (this thesis) equation this occurs when
PL>2.6 mm. Dr. Saumweber was unaware of this inconsistency (personal

59 communication), until we informed him of this by email. He is of the opinion that this is the result of inadequate sampling (larger copepods were sampled late in diapause), differences due to the application of the Miller et al. (2000) technique for estimating oil sac volume (a copepod's oil sac is more 'football' shaped than cylindrically shape), and the problems associated with fitting an exponential curve to in situ data. Dr. Saumweber sent us a revised equation relating C total
to L and OSV, as follows,
𝐶 = 7.67365 𝐿 3 + 308.17835 𝑂𝑆𝑉 − 7.41910
eq. 9
As Figure 9 (below) shows, this new equation has heavier total carbon weights for larger sized Calanus finmarchicus , which will eliminate the situation where the wax ester content alone is greater than the total carbon weight. Because of this issue, the results of the simulations reported in this thesis are in error, but the magnitude of that error is presently unknown, but likely to be small. These equations in the model will be revised and new simulations run with the corrected equations, and the manuscript revised prior to submission of this thesis for publication in a peer-reviewed journal. We apologize to the readers of this thesis for the inaccurate results included here.

60
600
500
400
300
200
100
Saumweber & Durbin (2006) as published
Saumweber Revised (pers. comm., 2010)
Kowalke--this thesis
0
1.7
1.9
2.1
2.3
2.5
2.7
2.9
3.1
Prosome Length (mm)
Figure 9. Plots of the prosome length dependent maximum total carbon weight using the
Saumweber and Durbin (2006) equation, the Kowalke replacement (Kowalke—this thesis) and the revised equation provided by Whitley Saumweber (pers. comm., 27 April
2010). Note that the Saumweber Revised equation has substantially increased total carbon weights for larger sized copepods.

61
Literature cited:
Batchelder H.P., Edwards C.A. and Powell T.M., 2002. Individual-based models of copepod populations in coastal upwelling regions: implications of physiologically and environmentally influenced diel vertical migration on demographic success and nearshore retention. Progress in Oceanography , 53, 307-333.
Batchelder, H.P. and Miller, C.B., 1989. Life history and population dynamics of
Metridia pacifica : Results from simulation modelling. Ecological Modelling ,
48(1-2), 113-136.
Beaugrand, G., 2004. The North Sea regime shift: Evidence, causes, mechanisms and consequences. Progress In Oceanography , 60(2-4), 245-262.
Beaugrand, G., Luczak, C. and Edwards, M., 2009. Rapid biogeographical plankton shifts in the North Atlantic Ocean. Global Change Biology , 15(7), 1790-1803.
Beaugrand, G. Reid, P.C., Ibanez F., Lindley J.A., and Edwards, M., 2002.
Reorganization of North Atlantic Marine Copepod Biodiversity and Climate.
Science , 296(5573), 1692-1694.
Beaugrand, G. and Reid, P.C., 2003. Long-term changes in phytoplankton, zooplankton and salmon related to climate. Global Change Biology , 9(6), 801-817.
Boyer, T.P., Antonov, J.I., Garcia, H.E., Johnson, D.R., Locarnini, R.A., Mishonov, A.V.,
Pitcher, M.T., Baranova, O.K., and Smolyar, I.V., 2006. World Ocean Database
2005. S. Levitus, Ed., NOAA Atlas NESDIS 60, U.S. Government Printing
Office, Washington, D.C., 190 pp., DVDs.
Brock, T., 1981. Calculating solar radiation for ecological studies. Ecological Modelling ,
14(1-2), 1-19.
Bucklin, A. and Kocher, T., 1996. Source regions for recruitment of Calanus finmarchicus to Georges Bank: evidence from molecular population genetic analysis of mtDNA. Deep Sea Research Part II: Topical Studies in
Oceanography , 43(7-8), 1665-1681.
Bucklin, A., Astthorsson, O.S., Gislason, A., Allen, L.D, Smolenack, S.D., and Wiebe,
P.H. 2000. Population genetic variation of Calanus finmarchicus in Icelandic waters: preliminary evidence of genetic differences between Atlantic and Arctic populations. ICES J. Mar. Sci.
, 57(6), 1592-1604.
Budgell, W. 2005. Numerical simulation of ice-ocean variability in the Barents Sea region. Ocean Dynamics 55, no. 3 (December 1): 370-387.

62
Campbell, R.W., 2004. Reply to Horizons Article 'Some ideas about the role of lipids in the life cycle of Calanus finmarchicus ' Irigoien (2004): I. Journal of Plankton
Research , 26(8), 979-980.
Chaudhuri, A.H., Gangopadhyay, A. and Bisagni, J.J., 2009. Interannual variability of
Gulf Stream warm-core rings in response to the North Atlantic Oscillation.
Continental Shelf Research , 29(7), 856-869.
Clarke, G.L. and Backus, R.H., 1957. Measurements of light penetration in relation to vertical migration and records of luminescence of deep-sea animals. Deep Sea
Research (1953) , 4, 1-4, IN1-IN2, 5-14.
Clarke, G.L. and Mahlon G. K., 1965. Measurements of Diurnal Changes in
Bioluminescence from the Sea Surface to 2,000 Meters Using a New Photometric
Device. Limnology and Oceanography , 10, 54-66.
Falkenhaug, T., Tande, K. and Timonin, A., 1997. Spatio-temporal patterns in the copepod community in Malangen, Northern Norway. J. Plankton Res.
, 19(4), 449-
468.
Fofonoff, N. P., and Millard, R.C,, Algorithms for the computation of fundamental properties of seawater, UNESCO technical papers in marine sciences , 44, 1-53,
1983.
Gislason, A. and Astthorsson, O.S., 1998. Seasonal variations in biomass, abundance and composition of zooplankton in the subarctic waters north of Iceland. Polar
Biology , 20(2), 85-94.
Greene, C.H. and Pershing, A.J., 2000. The response of Calanus finmarchicus populations to climate variability in the Northwest Atlantic: basin-scale forcing associated with the North Atlantic Oscillation. ICES J. Mar. Sci.
, 57(6), 1536-
1544.
Greene, C.H., and Pershing, A.J. 2003. The Flip-Side of the North Atlantic Oscillation and Modal Shifts in Slope-Water Circulation Patterns. Limnology and Oceanography 48, no. 1 (January): 319-322.
Greene, C.H., Pershing, A.J., Conversi A., Planque. B., Hannah, C., Sameoto, D., Head,
E., Smith, P.C., Reid, P.C., Jossi, J., Mountian, D., Benfield, B.C., Wiebe, P.H, and Durbin, E. 2003. Trans-Atlantic responses of Calanus finmarchicus populations to basin-scale forcing associated with the North Atlantic Oscillation.
Progress In Oceanography , 58(2-4), 301-312.

63
Haidvogel, D.B., Arango, H.G., Hedstrom, K., Beckmann, A., Malanotte-Rizzoli, P., and
Shchepetkin, A. 2000. Model evaluation experiments in the North Atlantic Basin: simulations in nonlinear terrain-following coordinates. Dynamics of Atmospheres and Oceans , 32(3-4), 239-281.
Häkkinen, S., and Mellor, G.L. 1992. Modeling the Seasonal Variability of a Coupled
Arctic Ice-Ocean System. Journal of Geophysical Research 97, no. C12: 20285-
20304.
Hallberg, E. and Hirche, H., 1980. Differentiation of mid-gut in adults and over-wintering copepodids of Calanus finmarchicus (Gunnerus) and C. helgolandicus Claus.
Journal of Experimental Marine Biology and Ecology , 48(3), 283-295.
Halvorsen, E., Tande, K.S., Edvardsen, A., Slagstad, D., and Pedersen, O.P. 2003. Habitat selection of overwintering Calanus finmarchicus in the NE Norwegian Sea and shelf waters off Northern Norway in 2000-02. Fisheries Oceanography , 12(4-5),
339-351.
Han, G., Lu, Z., Wang, Z., Helbig, J., Chen, N., and de Young, B.. 2008. Seasonal variability of the Labrador Current and shelf circulation off Newfoundland. Journal of
Geophysical Research 113, no. C10 (10).
Head, E. and Pepin, P., 2008. Variations in overwintering depth distributions of Calanus finmarchicus in the slope waters of the NW Atlantic continental shelf and the
Labrador Sea. Journal of Northwest Atlantic Fishery Science , 39, 49-69.
Head, E.J.H., Harris, L.R. and Yashayaev, I., 2003. Distributions of Calanus spp. and other mesozooplankton in the Labrador Sea in relation to hydrography in spring and summer (1995–2000). Progress In Oceanography , 59(1), 1-30.
Head, E., Harris, L. and Campbell, R., 2000. Investigations on the ecology of Calanus spp. in the Labrador Sea. I. Relationship between the phytoplankton bloom and reproduction and development of Calanus finmarchicus in spring. Marine
Ecology Progress Series , 193, 53-73.
Head, E.J., Harris, L.R. and Petrie, B., 1999. Distribution of Calanus spp. on and around the Nova Scotia Shelf in April: evidence for an offshore source of Calanus finmarchicus to the central and western regions. Canadian Journal of Fisheries and Aquatic Sciences , 56(12), 2463-2476.
Heath, M.R., Fraser, J.G., Gislason, A., Hay, S.J., Jonasdottir, S.H., and Richardson, K.
2000. Winter distribution of Calanus finmarchicus in the Northeast Atlantic. ICES
J. Mar. Sci.
, 57(6), 1628-1635.

64
Heath, M.R. and Lough, R.G., 2007. A synthesis of large-scale patterns in the planktonic prey of larval and juvenile cod ( Gadus morhua ). Fisheries Oceanography , 16(2),
169-185.
Heath, M., 1999. The ascent migration of Calanus finmarchicus from overwintering depths in the Faroe Shetland Channel. Fisheries Oceanography , 8(s1), 84-99.
Heath, M., Backhaus, J.O., Richardson, K., McKenzie, E., Slagstad, D., Beare, D., Dunn,
J., Fraser, J.G., Gallego, A., Hainbucher, D., Hay, S., Jonasdottir, S., Madden, H.,
Mardaljevic, J., and Scacht, A. 1999. Climate fluctuations and the spring invasion of the North Sea by Calanus finmarchicus . Fisheries Oceanography , 8(s1), 163-
176.
Heath, M. and Jónasdóttir, S., 1999. Distribution and abundance of overwintering
Calanus finmarchicus in the Faroe Shetland Channel. Fisheries Oceanography ,
8(s1), 40-60.
Helaouët, P. and Beaugrand, G., 2007. Macroecology of Calanus finmarchicus and C.
helgolandicus in the North Atlantic Ocean and adjacent seas. Marine Ecology
Progress Series , 345, 147-165.
Hind, A., Gurney, W., Heath, M., and Bryant, A.D. 2000. Overwintering strategies in
Calanus finmarchicus . Marine Ecology Progress Series , 193, 95-107.
Hirche, H., 1996. Diapause in the marine copepod, Calanus finmarchicus -- a review.
Ophelia , 44(1-3), 129-143.
Hurrell, J.W., 1995. Decadal trends in the North Atlantic Oscillation: Regional temperatures and precipitation., Science , 269(5224), 676-679.
Irigoien, X., 2000. Oceanographic Research Papers: Vertical distribution and population structure of Calanus finmarchicus at station India (59°N, 19°W) during the passage of the great salinity anomaly, 1971–1975. Deep Sea Research Part I ,
47(1), 1-26.
Irigoien, X., 2004. Some ideas about the role of lipids in the life cycle of Calanus finmarchicus . J. Plankton Res.
, 26(3), 259-263.
Johnson, C.L., Leising, A.W., Runge, J.A., Head, E.J.H, Pepin, P., Plourde, S., and
Durbin, E.G. 2008. Characteristics of Calanus finmarchicus dormancy patterns in the Northwest Atlantic. ICES J. Mar. Sci.
, 65(3), 339-350.
Johnson, C., 2003. Ecdysteroids in the oceanic copepod Calanus pacificus : variation during molt cycle and change associated with diapause. Mar , 257, 159-165.

65
Jónasdóttir, S., 1999. Lipid content of Calanus finmarchicus during overwintering in the
Faroe-Shetland Channel. Fisheries Oceanography , 8, 61-72.
Kaartvedt, S., 1996. Habitat preference during overwintering and timing of seasonal vertical migration of Calanus finmarchicus . Ophelia , 44(1-3), 145-156.
Kistler, R., Kalnay, E., Collins, W., Saha, S., White, G., Woollen, J., Chelliah, M.,
Ebisuzaki, W., Kanamitsu, M., Kousky, V., van den Dool, H., Jenne, R., and
Fiorino. M., 2001: The NCEP-NCAR 50-Year Reanalysis: Monthly Means CD-
ROM and Documentation. Bull. Amer. Meteor. Soc.
, 82, 247-267.
Luo, Y., Prater, M.D., Durbin, E.G., and Rothstein, L.M. 2006. Changes in the Northwest
Atlantic circulation for the 1992-95 high NAO period from a numerical model.
Continental Shelf Research , 26(14), 1617-1635.
Marshall, S.M. and Orr, A.P., 1955. The biology of a marine copepod, Calanus finmarchicus Gunnerus , Oliver and Boyd.
Miller, C.B., Crain, J.A. and Morgan, C.A., 2000. Oil storage variability in Calanus finmarchicus . ICES J. Mar. Sci.
, 57(6), 1786-1799.
Miller, C. Cowles, T.J., Wiebe, P.H., Copley, N.J., and Grigg, H., 1991. Phenology in
Calanus finmarchicus ; hypotheses about control mechanisms. Marine , 72(1-2),
79-91.
Ottersen, G., Planque, B., Belgrano, A., Post, E., Reid, P., and Stenseth, N. 2001.
Ecological effects of the North Atlantic Oscillation. Oecologia , 128(1), 1-14.
Paeth, H., Hense, A., Glowienka-Hense, R., Voss, S., and Cubasch, U. 1999. The North
Atlantic Oscillation as an indicator for greenhouse-gas induced regional climate change. Climate Dynamics , 15(12), 953-960.
Pepin, P. and Head, E.J., 2009. Oceanographic Research Papers: Seasonal and depthdependent variations in the size and lipid contents of stage 5 copepodites of
Calanus finmarchicus in the waters of the Newfoundland Shelf and the Labrador
Sea. Deep Sea Research Part I , 56(6), 989-1002.
Planque, B. and Fromentin, J., 1996. Calanus and environment in the eastern North
Atlantic. I. Spatial and temporal patterns of C. finmarchicus and C. helgolandicus .
Marine Ecology Progress Series , 134, 101-109.

66
Planque, B., Hays, G.C., Ibanez, F., and Gamble, J.C. 1997. Large scale spatial variations in the seasonal abundance of Calanus finmarchicus . Deep Sea Research Part I:
Oceanographic Research Papers , 44(2), 315-326.
Reid, P.C., Planque, B. and Edwards, M., 1998. Is observed variability in the long-term results of the Continuous Plankton Recorder survey a response to climate change?
Fisheries Oceanography , 7(3-4), 282-288.
Rey-Rassat, C., Irigoien, X., R Harris, R., and Carlotti, F. 2002. Energetic cost of gonad development in Calanus finmarchicus and C. helgolandicus . Marine Ecology
Progress Series , 238, 301-306.
Roy, S., Silverberg, N., Romero, N., Deibel, D., Klein, B., Savenkoff, C., Vezina, A.,
Tremblay, J., Legendre, L., and Rivkin, R. 2000. Importance of mesozooplankton feeding for the downward flux of biogenic carbon in the Gulf of St. Lawrence
(Canada). Deep Sea Research Part II: Topical Studies in Oceanography , 47(3-4),
519-544.
Saumweber, W.J. and Durbin, E.G., 2006. Topical Studies in Oceanography : Estimating potential diapause duration in Calanus finmarchicus . Deep Sea Research Part II ,
53(23-24), 2597-2617.
Schaeffer, M., Selten, F.M. and Opsteegh, J.D., 2005. Shifts of means are not a proxy for changes in extreme winter temperatures in climate projections. Climate
Dynamics , 25(1), 51-63.
Shindell, D.T., Miller, R.L., Schmidt, G.A., and Pandolfo, L. 1999. Simulation of recent northern winter climate trends by greenhouse-gas forcing. Nature , 399(6735),
452-455.
Song, Y. and Haidvogel, D., 1994. A Semi-implicit Ocean Circulation Model Using a
Generalized Topography-Following Coordinate System. Journal of Computational Physics , 115(1), 228-244.
Speirs, D.C., Gurney, W.S.C, Heath, M.R., and Wood, S.N. 2005. Modelling the basinscale demography of Calanus finmarchicus in the north-east Atlantic. Fisheries
Oceanography , 14, 333-358.
Stearns, D.E. and Forward, R.B., 1984. Copepod photobehavior in a simulated natural light environment and its relation to nocturnal vertical migration. Marine Biology ,
82(1), 91-100.

67
Stenseth, N., Ottersen, G., Hurrell, J.W., and Belgrano, A. 2004. Marine ecosystems and climate variation : the North Atlantic : a comparative perspective , Oxford ;New
York: Oxford University Press.
Tande, K.S., 1982. Ecological investigations on the zooplankton community of
Balsfjorden, northern Norway: Generation cycles, and variations in body weight and body content of carbon and nitrogen related to overwintering and reproduction in the copepod Calanus finmarchicus (Gunnerus). Journal of
Experimental Marine Biology and Ecology , 62(2), 129-142.
Tarrant, A., Baumgartner, M.F., Verslycke, T., and Johnson, C.L. 2008. Differential gene expression in diapausing and active Calanus finmarchicus (Copepoda). Marine
Ecology Progress Series , 355, 193-207.
Tittensor, D.P., DeYoung, B. and Tang, C.L., 2003. Modelling the distribution, sustainability and diapause emergence timing of the copepod Calanus finmarchicus in the Labrador Sea. Fisheries Oceanography , 12, 299-316.
U.S. Department of Commerce, National Oceanic and Atmospheric Administration,
National Geophysical Data Center, 2006.
Wishner, K.F., Schoenherr, J.R., Beardsley, R., and Chen, C. 1995. Abundance, distribution and population structure of the copepod Calanus finmarchicus in a springtime right whale feeding area in the southwestern Gulf of Maine.
Continental Shelf Research , 15(4-5), 475-507.

APPENDIX
68

69

70

71

72

73
Figure A1. Temperature anomalies between the 1995-1998 NABM and the WOD05 at
10m depth. Points indicate the position of a CTD cast in the WOD05, and its color indicates the temperature anomaly.

74
Figure A2. Temperature anomalies between the 1995-1998 NABM and the WOD05 at
400m depth. Points indicate the position of a CTD cast in the WOD05, and it‘s color indicates the temperature anomaly.

75
Figure A3. Position of the north wall of the Gulf Stream in the NABM under positive and negative NAO conditions. From Chaudhuri, A. (pers. comm.).

76

77

78

79

80
Figure A8. Map of Labrador Sea and Scotian Shelf indicating the AR7W sampling line
(blue), and the Flemish Pass (red). The AR7W line is depicted in figures A9, A10, and
A11, and the Flemish Pass line is depicted in figures A12 and A14.

81
A
B
Figure A9. Annual mean temperature profile across the AR7W line in the NABM for (A) positive NAO conditions and (B) negative NAO conditions. The AR7W line runs from
Labrador, on the left, to Greenland, on the right.

A
B
Figure A10. Annual mean salinity profile across the AR7W line in the NABM for (A) positive NAO conditions and (B) negative NAO conditions. The AR7W line runs from
Newfoundland, on the left, to Greenland, on the right.
82

83
A
B
Figure A11. Annual mean of velocity perpendicular to the AR7W line in the NABM for
(A) positive NAO conditions and (B) negative NAO conditions. The AR7W line runs from Labrador, on the left, to Greenland, on the right. Warm colors on the right indicate the Labrador Current as it flows south along Labrador.

84
A
B
Figure A12. Annual mean temperature profile across the Flemish Pass in the NABM for
(A) positive NAO conditions and (B) negative NAO conditions. The Grand Banks are on the left, and the Flemish Cap is on the right.

A
B
Figure A13. Annual mean of velocity perpendicular to the Flemish Pass in the NABM for (A) positive NAO conditions and (B) negative NAO conditions. The Grand Banks are on the left, and the Flemish Cap is on the right.
85

86
A
B
Figure A14. Temperature profile over time at Station B in the NABM for (A) positive
NAO conditions and (B) negative NAO conditions.

A
Q
Jul1 Sep1 Nov1 Jan1 Mar1 May1 Jun1
B
Q
Jul1 Sep1 Nov1 Jan1 Mar1 May1 Jun1
Figure A15. Temperature profile over time at Station C in the NABM for (A) positive
NAO conditions and (B) negative NAO conditions.
87

Prosome Length (mm)
Figure A16. Distribution of prosome lengths in all experiments.
88

Pre-assigned isopycnal (kg·m
-3
)
Figure A17. Distribution of pre-assigned isopycnals for all experiments.
89

A
NAO+
B
NAO-
90
C D
Figure A18. Tracks of copepods released on July 15 th
at station A for positive NAO conditions (A) and negative NAO conditions (B), and station B for positive NAO (C) and negative NAO (D). Red tracks are copepods descending to pre-assigned density, blue tracks are diapausing copepods, and green tracks are copepods emerging from diapause and ascending to the surface.

E
NAO+
F
NAO-
91
G H
Figure A18 (continued). Tracks of copepods released on July 15 th
at station C for positive NAO conditions (E) and negative NAO conditions (F), and station D for positive
NAO (G) and negative NAO (H). Red tracks are copepods descending to pre-assigned density, blue tracks are diapausing copepods, and green tracks are copepods emerging from diapause and ascending to the surface.

I
NAO+
J
NAO-
92
K L
Figure A18 (continued). Tracks of copepods released on July 15 th
at station E for positive NAO conditions (I) and negative NAO conditions (J), and station F for positive
NAO (K) and negative NAO (L). Red tracks are copepods descending to pre-assigned density, blue tracks are diapausing copepods, and green tracks are copepods emerging from diapause and ascending to the surface.

M
NAO+
N
NAO-
93
O P
Figure A18 (continued). Tracks of copepods released on July 15 th
at station G for positive NAO conditions (M) and negative NAO conditions (N), and station H for positive NAO (O) and negative NAO (P). Red tracks are copepods descending to preassigned density, blue tracks are diapausing copepods, and green tracks are copepods emerging from diapause and ascending to the surface.

Q
NAO+
R
NAO-
94
Figure A18 (continued). Tracks of copepods released on July 15 th
at station I for positive
NAO conditions (Q) and negative NAO conditions (R). Red tracks are copepods descending to pre-assigned density, blue tracks are diapausing copepods, and green tracks are copepods emerging from diapause and ascending to the surface.

95
A
B
Timestep (hr)
Timestep (hr)
Figure A19. Depth of copepod versus one hour timestep (720 timesteps per month) of all copepods released at Station C on August 15 th
in (A) positive and (B) negative NAO conditions.

A
B
Timestep (hr)
96
Timestep (hr)
Figure A20. Depth of copepod versus one hour timestep (720 timesteps per month) of all copepods released at Station I on August 15 th
in (A) positive and (B) negative NAO conditions.

A
Q
Timestep (hr)
B
Timestep (hr)
Figure A21. Depth over time profile of the journey of particle 897 during (A) positive and (B) negative NAO conditions. Contours are temperature in Celsius. This copepod was released from Station C on August 15 th
, with a prosome length of 3.0 mm, and a preassigned isopycnal of 1030.1 kg m
-3
. This copepod traveled from Station C in the north
Labrador Sea to the region of the Flemish Pass during diapause.

98
Figure A22. Peak abundance of C. finmarchicus in the North Atlantic, from Continuous
Plankton Recorder data. Note that the green area reaches peak abundance, and therefore, arouses from diapause earlier than the red or blue areas. From Planque et al. (1997).