Content uploaded by Erin Meyer-Gutbrod
Author content
All content in this area was uploaded by Erin Meyer-Gutbrod on Dec 20, 2023
Content may be subject to copyright.
Content uploaded by Erin Meyer-Gutbrod
Author content
All content in this area was uploaded by Erin Meyer-Gutbrod on Dec 20, 2023
Content may be subject to copyright.
PRIMARY RESEARCH ARTICLE
Uncertain recovery of the North Atlantic right whale in a
changing ocean
Erin L. Meyer-Gutbrod
1
|
Charles H. Greene
2
1
Marine Science Institute, University of
California, Santa Barbara, Santa Barbara,
CA, USA
2
Ocean Ecosystems and Resources
Program, Earth and Atmospheric Sciences,
Cornell University, Ithaca, NY, USA
Correspondence
Erin L. Meyer-Gutbrod, Marine Science
Institute, University of California, Santa
Barbara, Santa Barbara, CA, USA.
Emails: elg82@cornell.edu; elmg@ucsb.edu
Funding information
National Defense Science and Engineering
Fellowship; Cornell’s Atkinson Center for a
Sustainable Future, through its Sustainable
Biodiversity Fund
Abstract
Human activities have placed populations of many endangered species at risk and
mitigation efforts typically focus on reducing anthropogenic sources of mortality.
However, failing to recognize the additional role of environmental factors in regulat-
ing birth and mortality rates can lead to erroneous demographic analyses and con-
clusions. The North Atlantic right whale population is currently the focus of
conservation efforts aimed at reducing mortality rates associated with ship strikes
and entanglement in fishing gear. Consistent monitoring of the population since
1980 has revealed evidence that climate-associated changes in prey availability have
played an important role in the population’s recovery. The considerable interdecadal
differences observed in population growth coincide with remote Arctic and North
Atlantic oceanographic processes that link to the Gulf of Maine ecosystem. Here,
we build capture-recapture models to quantify the role of prey availability on right
whale demographic transitional probabilities and use a corresponding demographic
model to project population growth rates into the next century. Contrary to previ-
ous predictions, the right whale population is projected to recover in the future as
long as prey availability and mortality rates remain within the ranges observed dur-
ing 1980–2012. However, recent events indicate a northward range shift in right
whale prey, potentially resulting in decreased prey availability and/or an expansion
of right whale habitat into unprotected waters. An annual increase in the number of
whale deaths comparable to that observed during the summer 2017 mass mortality
event may cause a decline to extinction even under conditions of normal prey avail-
ability. This study highlights the importance of understanding the oceanographic
context for observed population changes when evaluating the efficacy of conserva-
tion management plans for endangered marine species.
KEYWORDS
capture–recapture, conservation, demography, endangered species, Eubalaena glacialis, marine
ecology, population modeling, right whale
1
|
INTRODUCTION
The North Atlantic right whale (Eubalaena glacialis) population has
approximately 524 individuals (Pettis & Hamilton, 2016) and ranges
from calving grounds in the southeastern US, along the coasts of
Florida and Georgia, to feeding grounds in the northeastern US and
Canada, primarily in and around the Gulf of Maine (GOM; Kraus &
Rolland, 2007). With its population decimated by centuries of com-
mercial whaling, this critically endangered species has exhibited vari-
able recovery rates since it first received protected status in 1935.
Despite protections, over half of its known mortalities in recent dec-
ades have been attributed to ship strikes, and over 80% of the
Received: 17 May 2017
|
Revised: 25 September 2017
|
Accepted: 27 September 2017
DOI: 10.1111/gcb.13929
Glob Change Biol. 2018;24:455–464. wileyonlinelibrary.com/journal/gcb ©2017 John Wiley & Sons Ltd
|
455
population has experienced entanglement in fishing gear, with entan-
glement mortalities on the rise (Campbell-Malone et al., 2008;
Henry, Cole, Hall, & Ledwell, 2016; Knowlton, Hamilton, Marx, Pet-
tis, & Kraus, 2012; Laist, Knowlton, & Pendleton, 2014). In addition
to these anthropogenic impacts on the species, variable feeding
conditions have been shown to influence right whale health and
demography.
The effects of prey limitation are most pronounced in reproduc-
tive females, which require increased caloric input to support preg-
nancy and lactation (Miller et al., 2011; Pettis et al., 2004).
Occurrences of sublethal entanglement have also been shown to sig-
nificantly increase energetic costs, competing with the energetic
requirements of reproduction to potentially increase calving intervals
(van der Hoop, Corkeron, & Moore, 2017). Reproductive females
also face decreased access to prey during migration to and time
spent on the calving ground (Fortune, Trites, Mayo, Rosen, & Hamil-
ton, 2013; Miller, Best, Perryman, Baumgartner, & Moore, 2012).
Birth rates correlate with changes in adult female health and body
condition (Rolland et al., 2016), and the interval between births has
been observed to lengthen during periods when prey availability is
low (Meyer-Gutbrod, Greene, Sullivan, & Pershing, 2015). Years with
few calf births directly correspond with years of decline or stagna-
tion in population growth, indicating the pronounced effect of repro-
ductive variability on species viability (Pace, Corkeron, & Kraus,
2017).
Since the right whale derives most of its nutrition from older
developmental stages of the lipid-rich copepod species Calanus fin-
marchicus (Mayo, Letcher, & Scott, 2001), the abundance of prey
species in the GOM provides a strong predictor of right whale repro-
ductive success (Greene & Pershing, 2004; Greene, Pershing, Ken-
ney, & Jossi, 2003; Meyer-Gutbrod & Greene, 2014; Meyer-Gutbrod
et al., 2015). In the GOM, decadal-scale changes in C. finmarhcicus
abundance have been linked to ecosystem regime shifts associated
with climate forcing from the Arctic (Greene et al., 2013; MERCINA
2012). During the 1980s, climatic conditions in the Arctic reduced
freshwater export from the Arctic Ocean into the North Atlantic,
and C. finmarchicus was abundant due to a favorable combination of
high local productivity and advective supply into the GOM from
upstream source regions (MERCINA 2004). In contrast, during the
1990s, significant declines in C. finmarchicus coincided with an
ecosystem regime shift in the GOM associated with the arrival of a
great salinity anomaly from the Arctic (Belkin, 2004; Greene, Persh-
ing, Cronin, & Ceci, 2008). This salinity anomaly altered the timing
and extent of seasonal stratification in the GOM throughout the
1990s, impacting the production and seasonal cycles of phytoplank-
ton, zooplankton and higher trophic-level consumers (Greene & Per-
shing, 2007; Greene et al., 2008, 2013). It has been hypothesized
that these changes in the ecosystem resulted in an increase in plank-
tivorous fishes, and their size-selective predation led to the decline
in C. finmarchicus abundance (Greene et al., 2013).
Coinciding with C. finmarchicus’decline, right whale birth rates
also decreased significantly during the 1990s, plummeting to a his-
torical low in 1999 and 2000 (Greene et al., 2003). Greene et al.
(2003) suggested that the reproductive failure observed during these
two years was driven by the 1998 crash of the C. finmarchicus popu-
lation in the GOM. They hypothesized that this crash was associated
with a decrease in the advective supply of C. finmarchicus into the
GOM after slope and shelf water circulation patterns in the North-
west Atlantic changed in response to the 20th century’s largest drop
in the North Atlantic Oscillation index during winter 1996 (MER-
CINA 2004).
At the end of the 1990s, the Arctic climate system shifted back
to a regime of freshwater storage (MERCINA 2012; Greene et al.,
2013). During the first decade of the 2000s, GOM plankton reverted
to an assemblage resembling that of the 1980s, including a rebound
in C. finmarchicus abundance. The right whale birth rate increased
rapidly and remained at a relatively high level for the rest of the dec-
ade (Meyer-Gutbrod & Greene, 2014; Meyer-Gutbrod et al., 2015).
Previously, it has been demonstrated that right whale reproduc-
tion since 1980 can be accurately replicated with models driven by
C. finmarchicus abundance anomalies determined from the GOM
Continuous Plankton Recorder survey (Meyer-Gutbrod & Greene,
2014; Meyer-Gutbrod et al., 2015). Here, we use a photographic
catalog of right whales compiled by the North Atlantic Right Whale
Consortium (Right Whale Consortium 2014) to develop a full demo-
graphic capture–recapture model, following individual whales in the
population as they transition between life stages at annual time
steps. This model enables us to evaluate the importance of prey
availability as well as the relative roles of birth and mortality rates in
determining right whale population growth during each of the past
three decades. It also provides the basis for population projections
into the future.
2
|
MATERIALS AND METHODS
2.1
|
Continuous Plankton Recorder data
The Continuous Plankton Recorder (CPR) is an instrument towed
behind ships of opportunity to collect and preserve plankton for sub-
sequent analyses in the laboratory (Warner & Hays, 1994). Since
1961, the NOAA National Marine Fisheries Service has operated a
CPR survey in the Gulf of Maine running between Boston, MA and
Cape Sable, NS at approximately monthly intervals (Jossi & Kane,
2013). In this study, we focus exclusively on the oldest stages of
C. finmarchicus (copepodite stages 5 and 6) due to their importance
in the diet of right whales (Mayo et al., 2001). Serving as a proxy for
annual variations in prey availability, an annually averaged transect-
wide index of late stage C. finmarchicus abundance anomalies were
calculated from the seasonal climatological cycle as described by
Pershing et al. (2005).
Although the CPR has limited geographic, depth, and temporal
coverage, its consistent use in long-term surveys has provided an
invaluable data set for studying broad ecosystem-wide regime shifts
on interannual and interdecadal time scales (Greene et al., 2013).
Despite its limitations, CPR-derived C. finmarchicus abundance has
been found to be significantly correlated with right whale sightings
456
|
MEYER-GUTBROD AND GREENE
in the Gulf of Maine (Pendleton et al., 2009; Pershing et al., 2009),
and used as a proxy for right whale prey abundance in multiple stud-
ies (Greene & Pershing, 2004; Greene et al., 2003; Meyer-Gutbrod
& Greene, 2014; Meyer-Gutbrod et al., 2015; Miller et al., 2011;
Patrician & Kenney, 2010).
2.2
|
Right whale capture–recapture model
North Atlantic right whales have been photographically catalogued
in a consistent manner since 1980, providing a history of sightings
and demographic states for each known individual (Figure 1; Right
Whale Consortium 2014). To study the demographic dynamics and
growth of the population, we built a state-based capture-recapture
demographic matrix model with females transitioning among five liv-
ing states and males transitioning among three living states (Caswell
& Fujiwara, 2004; Fujiwara & Caswell, 2002; Figure 2).
The newborn state accounts for the first year of life, and we
assume that mortality in this state is equal for both sexes. Whales
are classified as juveniles until they reach 9 years of age, have a
sighting history of more than 8 years or, in the case of a female,
until the year before the first known calving event (Hamilton, Knowl-
ton, Marx, & Kraus, 1998). Individuals are classified as dead either
when an identifiable carcass is recovered or after a whale has not
been sighted for five consecutive years.
An adult female sighted with a newborn enters the calving state,
and in the following year she enters the postcalving state. Females
never reproduce in consecutive years, and there are only 13 known
instances of a female exhibiting a 2-year spacing between births of
the 484 known calving events from 1980 to 2012 (Right Whale
Consortium 2014). By adding a 1-year postcalving state, we are able
to account for this difference in the probability of a 2-year birth
spacing compared to a longer birth spacing. Due to a lack of identi-
fied mortality events in calving and postcalving females, all adult
female states (resting, calving, and postcalving) are assumed to have
equal mortality rates.
Time steps between demographic states occur at 1-year intervals
based on a “right whale year,”which begins in December rather than
January. This definition allows calves born in December to be
included with their cohort, since right whale calves are born during
the winter months, usually December, January, and February.
2.3
|
Sex and state uncertainty
The photographic identification of an individual right whale in a
given year does not necessarily indicate that the sex or demographic
state of that individual is discernible. Nevertheless, the incorporation
of all data, even incomplete sightings, is valuable in a capture–recap-
ture analysis. Of the 679 individual right whales that have been
1980 1984 1988 1992 1996 2000 2004 2008 2012
Number of females
0
50
100
150
200 Newborn
Juvenile
Adult resting
Adult calving
Adult post-calving
Unknown stage
1980 1984 1988 1992 1996 2000 2004 2008 2012
Number of males
0
50
100
150
200
250 Newborn
Juvenile
Adult
Unknown stage
(a)
(b)
FIGURE 1 Stacked bar graph showing
the demographic distribution of known
female (a), male (b) right whales over the
time series 1980–2012. Demographic state
is denoted by color: red =newborn,
green =juvenile, blue =all adults (male) or
nonbreeding adults (female), purple =
calving female, yellow =postcalving
female, turquoise =unknown state [Colour
figure can be viewed at
wileyonlinelibrary.com]
MEYER-GUTBROD AND GREENE
|
457
photographically identified in the catalog, 63 have an unknown sex
(Right Whale Consortium 2014). We assume a 50% likelihood that
each unknown sex individual is male or female. Individuals of
unknown sex are most commonly whales that died at a young age
before researchers had a chance to identify the sex. Therefore, it is
essential to include these individuals in the analysis to avoid under-
estimating newborn and juvenile mortality rates.
Among all years that an individual was sighted, the demographic
state was identified 85% of the time. For the other 15% of sightings,
we incorporate state uncertainty into the model by assuming the
individual is either a juvenile or a nonreproductive adult according to
the ratio of known, catalogued whales in those states over the time
series 1980–2012, separated by sex.
2.4
|
Model estimation
Capture probabilities, or the likelihood that an individual will be
sighted in a given time step, are estimated for each sex and demo-
graphic state. Since a newborn sighting is always the first sighting
(or “capturing”) of an individual, capture probabilities cannot be esti-
mated for newborns. Due to the high level of survey effort on the
calving ground and other survey efforts in the northern feeding
grounds, we assume that all calf births are recorded; therefore, the
capture probability of a calving female is fixed at 1.0. The capture
probability of an individual in the dead state (state 6 for females,
state 4 for males) is fixed at 0. Capture probabilities for all remaining
demographic states are modeled as logistic functions with depen-
dence on annual survey effort, which is designated as the total num-
ber of shipboard and aerial survey days each year (Right Whale
Consortium 2014). An annual metric of survey effort is appropriate
in this study because all capture and transitional probabilities esti-
mated in the models are annual averages.
Transitional probabilities, or the likelihood of an individual whale
transitioning between demographic states during an annual time
step, are modeled as a set of polychotomous logistic functions, fol-
lowing the protocol described in Caswell and Fujiwara (2004) (see
Supporting Information). In the prey-independent model, demo-
graphic transitional probabilities are modeled as constant over the
period 1980–2012. In the prey-dependent model formulations,
demographic transitional and survival probabilities are tested for
dependence on annually averaged C. finmarchicus abundance anoma-
lies aggregated across the Gulf of Maine CPR transect at a lag of 0-,
1- and 2-years. We tested the effects of prey dependence on all
demographic vital rates and compared model fitness using Akaike
Information Criteria (Akaike, 1974).
Parameter estimation was performed in AD Model Builder (Four-
nier et al., 2012), with subsequent analysis in R (R Core Team 2015).
For model-derived products, such as transitional probabilities (Fig-
ure 3a) and population growth rate (Figure 3b), 95% confidence
intervals were estimated from 10,000 parametric bootstrap samples
generated assuming a normal distribution for all model parameters.
2.5
|
Forward projections
Only female individuals contribute to future population growth
through breeding, so population growth and sensitivity analyses can
be performed using the female portion of the capture–recapture
model. The transition matrix estimated in the capture–recapture
analysis is converted to a population projection matrix following the
protocol outlined in Caswell and Fujiwara (2004). Population growth
is simulated over the next century under three different prey scenar-
ios corresponding to the three decades that right whales were moni-
tored for this study: 1980–1989, 1990–1999 and 2000–2009. Prey
abundance anomalies are resampled at each annual time step, and
FIGURE 2 State-structured model of
right whale demography. State 1 is the
newborn state for both males and females.
States 2,3,4 and 5 (pink; upper row)
represent the female states of juvenile,
adult, calving and postcalving respectively.
States 2 and 3 (blue; lower row) represent
the male states of juvenile and adult
respectively. Black arrows represent the
probability of transitioning between living
states, and red arrows represent mortality
rates. Vital rates were estimated in the
static prey-independent model; transition
rate from female state 3 to state 4 is prey-
dependent in later model formulations
[Colour figure can be viewed at
wileyonlinelibrary.com]
458
|
MEYER-GUTBROD AND GREENE
each run is projected over a 100-year period. Forward projections
are also simulated under increasing lethal removal scenarios demon-
strating population growth over the following century in the scenario
of 0, 2, 4 and 6 annual lethal removals of adult females. Under each
projection scenario, the distribution of right whales in the initial time
step is set to the observed population size and demographic distribu-
tion in 2012. To demonstrate stochasticity, 100 simulations are run
for each projection scenario.
3
|
RESULTS
The basic capture–recapture model provides an estimate of average
vital rate values over the time period 1980–2012. In this static,
prey-independent formulation of the model, the right whale annual
population growth rate is k=1.026, and resting adult females have
a 24% chance of calving each year (Figure 2, Table S1). The probabil-
ity of postcalving females transitioning into a calving state (i.e.,
females following a 2-year calving cycle rather than the typical 3+
year calving cycle) is much less common at 3%. Mortality rates are
highest for newborns, at 5%, consistent with previous demographic
analysis for right whales (Kraus, 1990). Population growth is most
sensitive to changes in the vital rates of mature adult females, who
have survived the newborn and juvenile stages and directly con-
tribute to reproduction (Table S1).
The prey-dependent capture–recapture models designed in this
study reveal the significant effects of prey abundance anomalies on
calving rates and projected population growth rates. Demographic
state-specific mortality rates were not found to be prey-dependent,
indicating that starvation is not currently a major threat to popula-
tion viability. However, incorporating prey dependence in the female
calving state reveals temporal variability in overall population growth
that coincides with right whale photographic observations, and sig-
nificantly improves model fitness compared to the prey-independent
model (DAIC =33.9). Periods of low prey coincide with reduced
birth rates as females strive to obtain adequate nutrition for breed-
ing and lactation. The probability of calving in a given year over the
1980–2012 time period ranges from a maximum value of 36% in
1988, during a period of high C. finmarchicus abundance, to a mini-
mum value of 11% in 1999, following the C. finmarchicus crash in
1998 (Figure 3a). Annual population growth rates vary with changes
in calving probability, reaching a maximum of k=1.040 in 1988, a
moderate rate of growth for this species, and a minimum of
k=1.004 in 1999, essentially population stagnation (Figure 3b).
Over the time series 1980–2012, population growth rates were typi-
cally positive with a period of stagnation at the end of the 1990s, as
evidenced by both the model results (Figure 3b) and the photo-
graphic catalog observations (Figure 1).
To explore future population growth under different scenarios of
prey availability, population trajectories were projected into the next
century using C. finmarchicus abundances randomly sampled from
the three decadal regimes observed: the high prey abundances of
the 1980s, the low prey abundances of the 1990s, and moderately
high prey abundances of the 2000s (Figure 4a).
The geometric mean and geometric standard deviations of the
annual population growth rates estimated by the capture–recapture
model for each of the three decadal prey regimes are: k=1.032
(std =1.006) for the 1980s, k=1.019 (std =1.007) for the 1990s,
and k=1.029 (std =1.003) for the 2000s. These projections mimic
the results of the historical capture–recapture model (Figure 3b),
with growth remaining positive under the lowest decadal prey per-
iod.
Although prey availability is a major driver of decadal differences
in the right whale population’s recovery, anthropogenic sources of
mortality are important and their mitigation should be adopted as
major elements in a conservation management plan for the species.
This is especially true since recent studies indicate that the incidence
and severity of injuries due to entanglement in fishing gear are
increasing (Knowlton et al., 2016; Robbins, Knowlton, & Landry,
2015). To assess the population’s sensitivity to future increases in
anthropogenic mortality, the population’s growth trajectory was pro-
jected over the next century with no change in mortality rates, or
with two, four, or six additional adult female mortalities each year
added to the base annual mortality rate of 0.03 (Figure 4b). Calanus
finmarchicus abundance was randomly sampled with replacement
from the 1979–2011 time series. Sampling from all three prey
0.10
0.15
0.20
0.25
0.30
0.35
0.40
Calving probability
–1.0 –0.5 0.0 0.5
1.00
1.01
1.02
1.03
1.04
Population growth rate
C. finmarchicus anomal
y
(a)
(b)
FIGURE 3 (a) The probability of a female transitioning from
resting to calving as a function of the annually averaged Calanus
finmarchicus abundance anomaly. (b) The population growth rate (k)
as a function of the annually averaged C. finmarchicus abundance
anomaly. Ticks above the x-axis mark observed annual
C. finmarchicus anomalies over the time period 1979–2011
MEYER-GUTBROD AND GREENE
|
459
regimes, the population was found to decline during the next cen-
tury with the annual removal of six or more adult females above the
base annual mortality rate.
Since two of the previous three decades occurred during regimes
of relatively high prey availability, birth rates resampled from the
previous three decades may favor an overly optimistic outcome. If
the low prey conditions observed during the 1990s return and inhi-
bit reproduction, the population’s growth will be more sensitive to
increased mortality rates. To analyze the combined effects of low
prey availability and increased mortality, the population’s growth tra-
jectory was projected over the next century with C. finmarchicus
abundance randomly sampled with replacement from the 1990–
1999 time series and with three mortality scenarios: no change in
mortality rates, or with two or four additional adult female mortali-
ties each year (Figure 4c). In this scenario, the population was found
to decline during the next century with the annual removal of four
or more adult females above the base annual mortality rate.
The simulations projecting population growth under increased
removal of adult female whales show how the mortality of just a
few additional animals each year can cause a shift between modest
growth and population decline. These additional deaths have an
especially profound impact when removed from the demographic
state with the highest impact on future growth (Table S1). However,
in the event of a mass mortality event, it can be difficult to quickly
identify the age and sex of a carcass, especially in cases where
necropsy is impossible (MacKinnon, 2017). In light of the recent
summer-2017 mass-mortality event, with a minimum of 13
confirmed deaths (Kassam, 2017; MacKinnon, 2017), we repeated
the forward projection simulations with additional annual whale
removals taken from a random demographic state rather than the
adult female state. Sampling from the entire 1979–2011 time series,
the population was found to decline toward extinction with the
annual removal of 13 or more animals taken at random from the
population above the modeled average annual mortality rates
(Figure 5a). When projecting under the lower prey regime of the
1990s, an annual removal of 10 additional whales taken from a ran-
dom demographic state above the average mortality rates will cause
a decline toward extinction (Figure 5b).
4
|
DISCUSSION
While it is impossible to accurately predict future growth, especially
in a population as small and volatile as that of the North Atlantic
right whale, demographic projections can be used as a tool to
explore the interactions of various drivers of growth and decline.
Right whale vital rates are heavily influenced by both environmental
conditions, such as prey availability, as well as human activities, such
as shipping and fishing. This analysis demonstrates multiple scenarios
of population trajectories under both low and high prey availability
and with consistent or increasing anthropogenic mortality rates.
While the incorporation of prey availability as an explanatory driver
of calving rate variability assists in building a mechanistic under-
standing of demographic changes, future efforts to build a temporal
2020 2040 2060 2080 2100
0
2,000
4,000
6,000
8,000
10,000
12,000
Projected growth under 3 decadal prey regimes
Population size
1980s
1990s
2000s
1980s 1990s 2000s
1.01
1.02
1.03
1.04
Growth rate
2020 2040 2060 2080 2100
0
2,000
4,000
6,000
Projections with non-breeding
adult female removals
Population size
0 Added deaths
2 Added deaths
4 Added deaths
6 Added deaths
0+ Deaths 2+ Deaths 4+ Deaths 6+ Deaths
0.90
0.95
1.00
1.05
Growth rate
2020 2040 2060 2080 2100
0
1,000
2,000
3,000
Projections with 1990s prey and
adult female removals
Population size
0 Added deaths
2 Added deaths
4 Added deaths
0+ Deaths 2+ Deaths 4+ Deaths
0.98
0.99
1.00
1.01
1.02
1.03
1.04
Growth rate
(a) (b) (c)
FIGURE 4 Projected population growth of the North Atlantic right whale under different scenarios of prey availability and mortality
increases taken from the nonbreeding adult female state. (a) Projected growth under three different decadal prey regimes: 1980s (red), 1990s
(green) and 2000s (blue) with no change in mortality rates. (b and c) Projected growth with simulated increases in annual mortality of resting
adult females. (b) Projected with prey randomly sampled from the complete time series 1980–2012 and no change in mortality (black lines),
and 2 (red lines), 4 (green lines), or 6 (blue lines) additional annual resting adult female removals. (c) Projected with prey sampled randomly
from the 1990s and no change in mortality (black lines), and 2 (red lines), 4 (green lines), or 6 (blue lines) additional annual resting adult female
removals. Each line on the upper plots corresponds to a 100-year population projection, simulated 100 times under each decadal prey regime.
Lower plots: Box and whisker plots showing the median (thick black bar), first and third quartiles (edges of box) maximum and minimum (ends
of whiskers) values and outliers of the resampled population growth rate, k, used in each annual time step to create the projections [Colour
figure can be viewed at wileyonlinelibrary.com]
460
|
MEYER-GUTBROD AND GREENE
index of anthropogenic risk to drive mortality rate variability would
be a valuable extension to this work.
The positive right whale growth rate projections in this study
provide a contrast to those reported in the modeling study by Fuji-
wara and Caswell (2001), in which the right whale population was
considered in decline and projected to go extinct within 200 years.
This reported decline was the result of the assumption that breeding
females were experiencing an increasing temporal trend in mortality
rate. The authors cited a mass mortality event of five recovered right
whale carcasses in 1996 to support this assumption. However, of
the carcasses recovered that year, two were newborns, two were
males, and the remaining individual’s sex was not identified (Waring
et al., 1999). Although it is a common practice to model temporal
trends in vital rates, forward projections that assume a constant rate
of change far into the future, and especially those that lack a mecha-
nistic link, should be treated cautiously (Coulson, Mace, Hudson, &
Possingham, 2001; Sutherland, 2006).
While the population projections reported here are more opti-
mistic than those reported by Fujiwara and Caswell (2001), they
must be interpreted with the caveat that the past is not always a
good predictor for the future. If the right whale population is
exposed to lower prey availability and/or higher anthropogenic mor-
tality rates than observed during the period 1980–2012, then the
fate of this species will become much more uncertain. With the bal-
ance between recovery and a gradual decline toward extinction so
precarious, a significant commitment to monitoring the right whale’s
small population and its prey environment remains essential in prop-
erly managing the recovery of this highly endangered species. Unfor-
tunately, recent evidence indicating changes in the species’foraging
habits and reproduction suggest that current monitoring efforts are
inadequate.
During recent years, the GOM has undergone a rapid warming
unprecedented in the historical record (Mills et al., 2013). This warm-
ing could have a major impact on the right whale population by
altering the distribution and abundance of C. finmarchicus in the spe-
cies’traditional feeding grounds (Greene, 2016; Reygondeau & Beau-
grand, 2011). In response, right whales may be shifting their habitat
use and exploring feeding grounds further north (Pettis & Hamilton,
2016). Sightings of the species in its traditional feeding grounds have
declined significantly, and annual birth rates are also lower than the
2020 2040 2060 2080 2100
0
1,000
2,000
3,000
4,000
5,000
6,000
Projected growth with removals
from random stage
Population size
0 Added deaths
5 Added deaths
10 Added deaths
13 Added deaths
0+ Deaths 5+ Deaths 10+ Deaths 13+ Deaths
0.90
0.95
1.00
1.05
Growth rate
2020 2040 2060 2080 2100
0
500
1,000
1,500
2,000
2,500
3,000
Projected growth with 1990s prey and
random stage removals
Population size
0 Added deaths
5 Added deaths
10 Added deaths
0+ Deaths 5+ Deaths 10+ Deaths
0.97
0.98
0.99
1.00
1.01
1.02
1.03
Growth rate
(a) (b)
FIGURE 5 Projected population growth of the North Atlantic right whale under different scenarios of prey availability and mortality
increases taken from any demographic state at random. (a) Projected growth with prey randomly sampled from the complete time series
1980–2012 and no change in mortality (black lines), and 5 (red lines), 10 (green lines) or 13 (blue lines) additional annual removals.
(b) Projected with prey sampled randomly from the 1990s and no change in mortality (black lines), and 5 (red lines) or 10 (green lines)
additional annual removals. Each line on the upper plots corresponds to a 100-year population projection, simulated 100 times under each
decadal prey regime. Lower plots: Box and whisker plots showing the median (thick black bar), first and third quartiles (edges of box) maximum
and minimum (ends of whiskers) values and outliers of the resampled population growth rate, k, used in each annual time step to create the
projections [Colour figure can be viewed at wileyonlinelibrary.com]
MEYER-GUTBROD AND GREENE
|
461
previous decade (Kraus et al., 2016; Pettis & Hamilton, 2016). If
females continue to use the calving grounds along the coast of Flor-
ida and Georgia, migration distances will increase as feeding grounds
move farther north. Given the known effects of nutritional limita-
tions on calving intervals, the increased time and energy used to
extend the winter calving migration may have a detrimental impact
on the population’s recovery rate.
Right whale monitoring and protection efforts have historically
been focused on the Gulf of Maine, with routine shipboard and aer-
ial surveys, passive acoustic monitoring infrastructure, and right
whale protected areas. These efforts have proven to be reasonably
effective in the past. While summertime foraging typically occurs in
and around the Gulf of Maine, during the summer of 2017 high
numbers of right whales shifted north of this area to the Gulf of St.
Lawrence (Kassam, 2017). A minimum of ten mortalities were
reported over a period of a few weeks in the Gulf of St. Lawrence,
with at least three additional mortalities near the Gulf of Maine
(Fisheries and Oceans Canada 2017; Kassam, 2017; MacKinnon,
2017). At present, the causes for these mortalities are unknown.
Regardless of their cause, the deaths of thirteen whales, greater than
two percent of the current population, is a staggering loss. It is criti-
cal that monitoring and protection efforts continue and expand
quickly to coincide with northern extensions in the species’range,
especially if these are likely to be permanent.
The population projections under conditions of increased
anthropogenic mortality rates demonstrate that the North Atlantic
right whale population may decline to extinction if an additional six
adult females (or 13 total individuals taken from demographic
states at random) are killed each year, assuming prey conditions
remain similar to those observed from 1980 to 2012 (Figures 4b
and 5a). However, under a regime of lower prey availability similar
to that of the 1990s, the additional removal of only 4 adult
females (or 10 total individuals taken from demographic states at
random) may cause a decline to extinction (Figures 4c and 5b). The
unprecedented right whale die-off during the summer 2017 is
especially disconcerting because calf births have plummeted with
only five calves counted the previous winter (Hain, 2017). Modeling
studies indicate that variation in survivorship is driving a shift in
the sex ratio to a male-dominated population, which poses addi-
tional challenges to achieving robust reproduction rates (Pace et al.,
2017). As right whale habitat occupation and sighting probabilities
have changed rapidly since 2010, it is too early to predict whether
these demographic changes are anomalous or whether the popula-
tion decline will continue.
It is important to note that while this study shows the threshold
in annual anthropogenic mortality increases that will cause popula-
tion decline, these are not conditions to strive for. Management
efforts should be aimed at achieving robust population growth rather
than simply avoiding population decline. While extinction risk is gov-
erned by environmental variables such as prey availability, vessel
traffic, and fishing gear use, it is also a product of stochasticity (Mel-
bourne & Hastings, 2008). Since it is impossible to accurately
account for all factors that increase extinction risk, especially in a
small population, it is essential that at-risk species, such as the North
Atlantic right whale, are managed conservatively.
Compounding the recent uncertainties about changes in right
whale foraging behavior, reproduction, and anthropogenic mortali-
ties, there are also new uncertainties about the availability of prey.
Due to budgetary constraints, the Northeast Fisheries Science Cen-
ter is no longer processing samples from the GOM CPR survey.
Therefore, at a time when right whales may be facing unprecedented
nutritional stress, scientists and managers lack the data necessary to
assess prey availability, the primary determinant of the species’
reproductive success. Recovery of the North Atlantic right whale
population may well depend upon management decisions arrived at
by using the best quantitative information available. It is unfortunate
that as the modeling framework used to generate such information
becomes available, the long-term monitoring program essential to
providing input data for such models is being compromised rather
than expanded. It is difficult to overemphasize the value of such
long-term monitoring programs as we attempt to understand and
predict the fates of both protected and exploited species of interest
in a changing ocean.
ACKNOWLEDGEMENTS
The authors thank the North Atlantic Right Whale Consortium for
access to the population data analyzed and reported in this paper.
We are especially appreciative of Philip Hamilton’s help and guid-
ance. Patrick Sullivan, Christopher Clark, Andrew Pershing, and three
anonymous reviewers provided valuable feedback on this research.
Support for E.M.G. to conduct this research was provided by the
Department of Defense, through a National Defense Science and
Engineering Fellowship, and Cornell’s Atkinson Center for a Sustain-
able Future, through its Sustainable Biodiversity Fund. We thank the
Whiteley Center at the University of Washington’s Friday Harbor
Laboratories for providing both authors with an inspiring setting dur-
ing the completion of this paper.
AUTHOR CONTRIBUTION
EMG built and analyzed the models, wrote the first draft of the
manuscript; CHG contributed to conceptual development and revi-
sion of manuscript.
ORCID
Erin L. Meyer-Gutbrod http://orcid.org/0000-0003-3184-5690
REFERENCES
Akaike, H. (1974). A new look at the statistical model identification. IEEE
Transactions on Automatic Control,19, 716–723.
Belkin, I. M. (2004). Propagation of the ‘‘Great Salinity Anomaly’’ of the
1990s around the northern North Atlantic. Geophysical Research Let-
ters,31, L08306. https://doi.org/10.1029/2003GL019334
462
|
MEYER-GUTBROD AND GREENE
Campbell-Malone, R., Barco, S. G., Daoust, P. Y., Knowlton, A. R.,
Mclellan, W. A., Rotstein, D. S., & Moore, M. J. (2008). Gross and
histologic evidence of sharp and blunt trauma in North Atlantic
Right Whales (Eubalaena glacialis) killed by vessels. Journal of Zoo
and Wildlife Medicine,39,37–55. https://doi.org/10.1638/2006-
0057.1
Caswell, H., & Fujiwara, M. (2004). Beyond survival estimation: Mark-
recapture, matrix population models, and population dynamics. Animal
Biodiversity and Conservation,27, 471–488.
Coulson, T., Mace, G. M., Hudson, E., & Possingham, H. (2001). The use
and abuse of population viability analysis. Trends in Ecology & Evolu-
tion,16(5), 219–221.
Fisheries and Oceans Canada (2017, Aug. 18). Right whale deaths in
Gulf of St. Lawrence. Retrieved from http://www.dfo-mpo.gc.ca/scie
nce/environmental-environnement/narightwhale-baleinenoirean/index-
eng.html
Fortune, S. M., Trites, A. W., Mayo, C. A., Rosen, D. A., & Hamilton, P. K.
(2013). Energetic requirements of North Atlantic right whales and the
implications for species recovery. Marine Ecology Progress Series,478,
253–272. https://doi.org/10.3354/meps10000
Fournier, D. A., Skaug, H. J., Ancheta, J., Ianelli, J., Magnusson, A., Maun-
der, M. N., ... Sibert, J. (2012). AD Model Builder: Using automatic
differentiation for statistical inference of highly parameterized com-
plex nonlinear models. Optimization Methods and Software,27(2),
233–249.
Fujiwara, M., & Caswell, H. (2001). Demography of the endangered
North Atlantic right whale. Nature,414(6863), 537–541.
Fujiwara, M., & Caswell, H. (2002). Estimating population projection
matrices from multi-stage mark–recapture data. Ecology,83(12),
3257–3265.
Greene, C. H. (2016). North America’s iconic marine species at risk due
to unprecedented ocean warming. Oceanography,29(3), 14–17.
https://doi.org/10.5670/oceanog.2016.67
Greene, C. H., Meyer-Gutbrod, E., Monger, B. C., McGarry, L. P., Persh-
ing, A. J., Belkin, I. M., ... Ji, R. (2013). Remote climate forcing of
decadal-scale regime shifts in Northwest Atlantic shelf ecosystems.
Limnology and Oceanography,58, 803–816. https://doi.org/10.4319/
lo.2013.58.3.0803
Greene, C. H., & Pershing, A. J. (2004). Climate and the conservation
biology of North Atlantic right whales: The right whale at the wrong
time? Frontiers in Ecology and the Environment,2(1), 29–34.
Greene, C. H., & Pershing, A. J. (2007). Climate drives sea change.
Science,315(5815), 1084.
Greene, C. H., Pershing, A. J., Cronin, T. M., & Ceci, N. (2008). Arctic cli-
mate change and its impacts on the ecology of the North Atlantic.
Ecology,89(11, Supplement), S24–S38.
Greene, C. H., Pershing, A. J., Kenney, R. D., & Jossi, J. W. (2003). Impact
of climate variability on the recovery of endangered North Atlantic
Right Whales. Oceanography,16,98–103.
Hain, J. (2017). The 2017 season: Two additional calves. Right Whale
News,25(3), 1–3.
Hamilton, P. K., Knowlton, A. R., Marx, M. K., & Kraus, S. D. (1998). Age
structure and longevity in North Atlantic right whales Eubalaena gla-
cialis and their relation to reproduction. Marine Ecology Progress Ser-
ies,171, 285–292.
Henry, A. G., Cole, T., Hall, L., & Ledwell, W. (2016). Serious Injury and
Mortality Determinations for Baleen Whale Stocks along the Gulf of
Mexico, United States East Coast, and Atlantic Canadian Provinces,
2010–2014. Northeast Fisheries Science Center Reference Document,
16–10.
van der Hoop, J., Corkeron, P., & Moore, M. (2016). Entanglement is a
costly life-history stage in large whales. Ecology and Evolution,7,92–
106. https://doi.org/10.1093/icb/16.4.699
Jossi, J. W., & Kane, J. (2013). An atlas of the dominant zooplankton col-
lected along a Continuous Plankton Recorder transect between
Massachusetts USA and Cape Sable NS, 1961–2008. US Dept Com-
mer, Northeast Fish Sci Cent Ref Doc, 13-12.
Kassam, A. (2017, August 3). Why do endangered right whales keep
dying off the coast of Canada? Retrieved from https://www.theguard
ian.com/environment/2017/aug/03/right-whales-10-dead-canada-enda
ngered-species
Knowlton, A. R., Hamilton, P. K., Marx, M. K., Pettis, H. M., & Kraus, S.
D. (2012). Monitoring North Atlantic right whale Eubalaena glacialis
entanglement rates: A 30 yr retrospective. Marine Ecology Progress
Series,466, 293–302. https://doi.org/10.3354/meps09923
Knowlton, A. R., Robbins, J., Landry, S., McKenna, H. A., Kraus, S. D., &
Werner, T. B. (2016). Effects of fishing rope strength on the severity
of large whale entanglements. Conservation Biology,30(2), 318–328.
https://doi.org/10.1111/cobi.12590
Kraus, S. D. (1990). Rates and potential causes of mortality in North
Atlantic right whales (Eubalaena glacialis). Marine Mammal Science,6
(4), 278–291.
Kraus, S. D., Kenney, R. D., Mayo, C. A., McLellan, W. A., Moore, M. J., &
Nowacek, D. P. (2016). Recent scientific publications cast doubt on
North Atlantic right whale future. Frontiers in Marine Science,3, 137.
Kraus, S. D., & Rolland, R. (2007). The urban whale: North Atlantic right
whales at the crossroads. Cambridge, MA: Harvard University Press.
Laist, D. W., Knowlton, A. R., & Pendleton, D. (2014). Effectiveness of
mandatory vessel speed limits for protecting North Atlantic right
whales. Endangered Species Research,23(2), 133–147.
MacKinnon, B.-J. (2017, August 16). ‘It’s very alarming.’Another dead
North Atlantic right whale found off Cape Cod. Retrieved from
http://www.cbc.ca/news/canada/new-brunswick/dead-north-atlantic-
right-whale-cape-cod-1.4249868
Mayo, C. A., Letcher, B. H., & Scott, S. (2001). Zooplankton filtering effi-
ciency of the baleen of a North Atlantic right whale, Eubalaena glacia-
lis.Journal of Cetacean Research and Management,3, 245–250.
Melbourne, B. A., & Hastings, A. (2008). Extinction risk depends strongly
on factors contributing to stochasticity. Nature,454, 100–103.
https://doi.org/10.1139/f54-039
[MERCINA] Marine Ecosystem Responses to Climate in the North Atlan-
tic Working Group (2004). Supply-side ecology and the response of
zooplankton to climate-driven changes in North Atlantic Ocean circu-
lation. Oceanography,17(3), 60–71. https://doi.org/10.5670/oceanog.
2004.31
[MERCINA] Marine Ecosystem Responses to Climate in the North Atlan-
tic Working Group (2012). Recent Arctic climate change and its
remote forcing of Northwest Atlantic shelf ecosystems. Oceanogra-
phy,25(3), 208–213. https://doi.org/10.5670/oceanog.2012.64
Meyer-Gutbrod, E. L., & Greene, C. H. (2014). Climate-associated regime
shifts drive decadalscale variability in recovery of North Atlantic right
whale population. Oceanography,27(3), 148–153. https://doi.org/10.
5670/oceanog.2014.64
Meyer-Gutbrod, E. L., Greene, C. H., Sullivan, P. J., & Pershing, A. J.
(2015). Climate-associated changes in prey availability drive repro-
ductive dynamics of the North Atlantic right whale population. Mar-
ine Ecology Progress Series,535, 243–258. https://doi.org/10.3354/
meps11372
Miller, C. A., Best, P. B., Perryman, W. L., Baumgartner, M. F., & Moore,
M. J. (2012). Body shape changes associated with reproductive sta-
tus, nutritive condition and growth in right whales Eubalaena glacialis
and E. australis.Marine Ecology Progress Series,459, 135–156.
https://doi.org/10.3354/meps09675
Miller, C. A., Reeb, D., Best, P. B., Knowlton, A. R., Brown, M. W., &
Moore, M. J. (2011). Blubber thickness in right whales Eubalaena gla-
cialis and Eubalaena australis related with reproduction, life history
status and prey abundance. Marine Ecology Progress Series,438, 267–
283.
Mills, K. E., Pershing, A. J., Brown, C. J., Chen, Y., Chiang, F. S., Holland,
D. S., ... Wahle, R. A. (2013). Fisheries management in a changing
MEYER-GUTBROD AND GREENE
|
463
climate lessons from the 2012 ocean heat wave in the Northwest
Atlantic. Oceanography,26(2), 191–195. https://doi.org/10.5670/ocea
nog.2013.27
Pace, R. M. III, Corkeron, P. J., & Kraus, S. D. (2017). State-space
mark-recapture estimates reveal a recent decline in abundance of
North Atlantic right whales. Ecology and Evolution, https://doi.org/
10.1002/ece3.3406s
Patrician, M. R., & Kenney, R. D. (2010). Using the Continuous Plankton
Recorder to investigate the absence of North Atlantic right whales
(Eubalaena glacialis) from the Roseway Basin foraging ground. Journal
of Plankton Research,32, 1685–1695. https://doi.org/10.1093/pla
nkt/fbq073
Pendleton, D. E., Pershing, A. J., Brown, M. W., Mayo, C. A., Kenney, R.
D., Record, N. R., & Cole, T. V. (2009). Regional-scale mean copepod
concentration indicates relative abundance of North Atlantic right
whales. Marine Ecology Progress Series,378, 211–225.
Pershing, A. J., Greene, C. H., Jossi, J. W., O’Brien, L., Brodziak, J. K., &
Bailey, B. A. (2005). Interdecadal variability in the Gulf of Maine zoo-
plankton community, with potential impacts on fish recruitment. ICES
Journal of Marine Science,62(7), 1511–1523.
Pershing, A. J., Record, N. R., Monger, B. C., Mayo, C. A., Brown, M. W.,
Cole, T. V., ... Woodard, L. A. (2009). Model-based estimates of right
whale habitat use in the Gulf of Maine. Marine Ecology Progress Series,
378, 245–257.
Pettis, H., & Hamilton, P. (2016). North Atlantic right whale consortium
2016 annual report card. Report to the North Atlantic Right Whale
Consortium.
Pettis, H. M., Rolland, R. M., Hamilton, P. K., Brault, S., Knowlton, A. R.,
& Kraus, S. D. (2004). Visual health assessment of North Atlantic
right whales (Eubalaena glacialis) using photographs. Canadian Journal
of Zoology,82(1), 8–19.
R Core Team (2015). R: A language and environment for statistical comput-
ing. Vienna, Austria: R Foundation for Statistical Computing.
Retrieved from http://www.R-project.org/
Reygondeau, G., & Beaugrand, G. (2011). Future climate-driven shifts in
distribution of Calanus finmarchicus.Global Change Biology,17, 756–
766. https://doi.org/10.1111/j.1365-2486.2010.02310.x
Right Whale Consortium (2014). North Atlantic Right Whale Consortium
Sightings and Identification Databases 12/11/2014. Boston, MA:
New England Aquarium
Robbins, J., Knowlton, A. R., & Landry, S. (2015). Apparent survival of
North Atlantic right whales after entanglement in fishing gear. Biologi-
cal Conservation,191, 421–427. https://doi.org/10.1016/j.biocon.
2015.07.023
Rolland, R. M., Schick, R. S., Pettis, H. M., Knowlton, A. R., Hamilton, P.
K., & Clark, J. S. (2016). Health of North Atlantic right whales Eubal-
aena glacialis over three decades: From individual health to demo-
graphic and population health trends. Marine Ecology Progress Series,
542, 265–282. https://doi.org/10.3354/meps11547
Sutherland, W. J. (2006). Predicting the ecological consequences of envi-
ronmental change: A review of the methods*.Journal of Applied Ecol-
ogy,43, 599–616. https://doi.org/10.5751/ES-00026-010201
van der Hoop, J., Corkeron, P., & Moore, M. (2017). Entanglement is a
costly life-history stage in large whales. Ecology and Evolution,7,92–
106. https://doi.org/10.1093/icb/16.4.699
Waring, G. T., Palka, D. L., Clapham, P. J., Swartz, S., Rossman, M. C.,
Cole, T. V., ... Odell, D. K. (1999). US Atlantic and Gulf of Mexico
Marine Mammal Stock Assessments–1999. Bulletin (US),96, 686–726.
Warner, A. J., & Hays, G. C. (1994). Sampling by the continuous plankton
recorder survey. Progress in Oceanography,34(2–3), 237–256.
SUPPORTING INFORMATION
Additional Supporting Information may be found online in the sup-
porting information tab for this article.
How to cite this article: Meyer-Gutbrod EL, Greene CH.
Uncertain recovery of the North Atlantic right whale in a
changing ocean. Glob Change Biol. 2018;24:455–464. https://
doi.org/10.1111/gcb.13929
464
|
MEYER-GUTBROD AND GREENE
