ArticlePDF Available

Changing spatial distribution of fish stocks in relation to climate and population size on the Northeast United Sates continental shelf

Authors:
  • Northeast Fisheries Science Center

Abstract and Figures

We tested the hypothesis that recent oceanographic changes associated with climate change in the Northeast United States continental shelf ecosystem have caused a change in spatial distribution of marine fish. To do this, we analyzed temporal trends from 1968 to 2007 in the mean center of biomass, mean depth, mean temperature of occurrence, and area occupied in each of 36 fish stocks, Temporal trends in distribution were compared to time series of both local-and large-scale environmental variables, as well as estimates of survey abundance, Many stocks spanning several taxonomic groups, life-history strategies, and rates of fishing exhibited a poleward shift in their center of biomass, most with a simultaneous increase in depth, and a few with a concomitant expansion of their northern range. However, distributional changes were highly dependent on the biogeography of each species. Stocks located in the southern extent of the survey area exhibited much greater poleward shifts in center of biomass and some occupied habitats at increasingly greater depths. In contrast, minimal changes in the center of biomass were observed in stocks with distributions limited to the Gulf of Maine, but mean depth of these stocks increased while stock size decreased. Large-scale temperature increase and changes in circulation, represented by the Atlantic Multidecadal Oscillation, was the most important factor associated with shifts in the mean center of biomass. Stock size was more often correlated with the total area occupied by each species. These changes in spatial distribution of fish stocks are likely to persist such that stock structure should be re-evaluated for some species.
Content may be subject to copyright.
MARINE ECOLOGY PROGRESS SERIES
Mar Ecol Prog Ser
Vol. 393: 111–129, 2009
doi: 10.3354/meps08220 Published October 30
INTRODUCTION
Recently water temperatures have increased both
globally (Levitus et al. 2000, Knutson et al. 2006, Lozier
et al. 2008) and regionally (e.g. in the Northwest
Atlantic) (Friedland & Hare 2007, Belkin 2009). Cli-
mate models indicate that this warming trend is likely
to continue (IPCC 2007, Solomon et al. 2009). Many
recent studies have detected ecological impacts of cli-
mate change (Walther et al. 2002, Parmesan & Yohe
2003, Rosenzweig et al. 2008), but in many situations it
remains difficult to disentangle the effects of climate
change with those of other anthropogenic influences.
This fact is especially true of exploited species that
may exhibit responses consistent with both climate-
induced and harvest-induced impacts. Modeling stud-
ies suggest that both exploitation and climate change
will affect the dynamics of exploited species (Clark et
al. 2003, Fogarty et al. 2007, Hare et al. in press), and,
for successful management, it is important to docu-
ment the relative importance and synergistic effects of
these 2 stressors.
The first-order response of organisms to climate
change is a shift in distribution (Frank et al. 1990,
© Inter-Research 2009 · www.int-res.com*Email: janet.nye@noaa.gov
Changing spatial distribution of fish stocks in
relation to climate and population size on the
Northeast United States continental shelf
Janet A. Nye1,*, Jason S. Link1, Jonathan A. Hare2, William J. Overholtz1
1National Marine Fisheries Service, Northeast Fisheries Science Center, Woods Hole Laboratory, 166 Water St., Woods Hole,
Massachusetts 02543, USA
2National Marine Fisheries Service, Northeast Fisheries Science Center, Narragansett Laboratory, 28 Tarzwell Drive,
Narragansett, Rhode Island 02882, USA
ABSTRACT: We tested the hypothesis that recent oceanographic changes associated with climate
change in the Northeast United States continental shelf ecosystem have caused a change in spatial
distribution of marine fish. To do this, we analyzed temporal trends from 1968 to 2007 in the mean
center of biomass, mean depth, mean temperature of occurrence, and area occupied in each of 36 fish
stocks. Temporal trends in distribution were compared to time series of both local- and large-scale
environmental variables, as well as estimates of survey abundance. Many stocks spanning several
taxonomic groups, life-history strategies, and rates of fishing exhibited a poleward shift in their cen-
ter of biomass, most with a simultaneous increase in depth, and a few with a concomitant expansion
of their northern range. However, distributional changes were highly dependent on the biogeogra-
phy of each species. Stocks located in the southern extent of the survey area exhibited much greater
poleward shifts in center of biomass and some occupied habitats at increasingly greater depths. In
contrast, minimal changes in the center of biomass were observed in stocks with distributions limited
to the Gulf of Maine, but mean depth of these stocks increased while stock size decreased. Large-
scale temperature increase and changes in circulation, represented by the Atlantic Multidecadal
Oscillation, was the most important factor associated with shifts in the mean center of biomass. Stock
size was more often correlated with the total area occupied by each species. These changes in spatial
distribution of fish stocks are likely to persist such that stock structure should be re-evaluated for
some species.
KEY WORDS: Atlantic Multidecadal Oscillation · Climate change · Biogeography · Center of biomass ·
Northeast United States continental shelf · Distribution · Areaabundance relationships
Resale or republication not permitted without written consent of the publisher
Mar Ecol Prog Ser 393: 111–129, 2009
Shuter & Post 1990, McCarty 2001), which, in marine
species, may be manifested as a change in the center of
biomass, an expansion or contraction of the species’
range, or a change in depth distribution. These re-
sponses have recently been documented for fish com-
munities in the North Sea (Perry et al. 2005) and in the
Bering Sea (Mueter & Litzow 2008). Reponses of marine
fishes consistent with warming water bodies would be a
poleward shift in the mean center of biomass and an in-
crease in the mean depth of occurrence. The ranges of
‘cold’ species may be reduced by warming waters,
while ‘warm’ species might expand their range.
The potential importance of temperature regime
and climate on fish stocks, especially those that are
economically important, has been studied on rela-
tively short temporal and spatial scales in the NW
Atlantic (Frank et al. 1990, Murawski 1993, Drinkwa-
ter et al. 2003, Hare & Able 2007). Frank et al. (1990)
hypothesized that distributional changes in response
to warming would be most pronounced in Canadian
fish stocks at the southern extent of their range,
based on historical reports of distributional changes
due to warming in the 1950s (Taylor et al. 1957).
Murawski (1993) classified species into groups that
would be more or less likely to exhibit changes in
distribution in a warmer climate based on tempera-
ture preferences and historical distribution. Similarly,
Rose (2005) predicted that, with warming, American
shad, alewife, Atlantic mackerel, American plaice,
and winter flounder, among other species, would shift
their distribution north. Drinkwater (2005) predicted
that the George’s Bank and Gulf of Maine stocks of
Atlantic cod Gadus morhua would decline in abun-
dance and move poleward.
Since these studies, the Northeast United States con-
tinental shelf ecosystem has experienced a consistently
warm period in the last 10 yr and a rebuilding of many
overfished stocks, presenting an opportunity to reex-
amine the links between distribution, warming, and
population abundance. The purpose of the present
study was to quantify the relationship between the
spatial distribution of 36 NE United States fish stocks
and the changes in temperature regime and abun-
dance using a 40 yr trawl survey time series (1968 to
2007). We hypothesize that rising water temperatures
would induce at least one of the following responses: a
poleward shift in the center of biomass, an increase in
the mean depth of occurrence, a range expansion or
contraction depending upon biogeography of the spe-
cies, or no distributional shift consistent with warming,
but an increase in the mean temperature of occur-
rence. We also hypothesize that these responses would
be mediated by changes in the abundance of each
stock, many of which have occurred over the time
period examined.
MATERIALS AND METHODS
Fish data. Abundance and distribution data for 36
fish stocks were obtained from the Northeast Fisheries
Science Center (NEFSC) spring trawl survey, which
has been carried out on the NE United States continen-
tal shelf since 1968 (Fig. 1). The data collection and
sample processing methods have been described by
Azarovitz (1981). Briefly, the survey employs a strati-
fied random design, with stations allocated proportion-
ally to stratum area. A 12 mm mesh codend liner is
used to retain smaller bodied and juvenile fish. All fish
for each species were counted and weighed. Correc-
tion factors were applied for changes in vessels, gear,
and doors when appropriate. Only tows in which there
were no gear or duration problems from offshore trawl
survey strata consistently sampled over the spring sur-
vey time series (Strata 01010– 01300, 01360 01400,
01610– 01760) were used in statistical analysis. The
strata used are shown in Fig. 1. There has been no sig-
nificant change in the timing of the spring survey, and
the mean annual latitude and longitude of the stations
has not changed over time, using the strata that were
consistently sampled over the duration of this 40 yr
time series.
We chose 36 stocks for investigation, because indi-
viduals from these stocks were caught every year dur-
ing the spring survey, were consistently caught in rel-
atively high numbers, and represented a wide range of
taxonomic groups (Table 1, for species’ scientific
112
Fig. 1. Survey area with strata used in this analysis. GOM: Gulf
of Maine; GB: George’s Bank; SNE: Southern New England;
MAB: mid-Atlantic Bight; Q: fixed point used to calculate
distance, d, in each year
Nye et al.: Changing spatial distribution of Northeast USA fish stocks
names please refer to this table throughout). Most spe-
cies included here have only 1 identified stock in the
region, and were analyzed as such. Those species with
multiple stocks were analyzed by ecoregions, which
roughly correspond to stock boundaries. The 2 eco-
regions were the Gulf of Maine (GOM, ‘northern’) and
combined areas of George’s Bank, southern New Eng-
land, and mid-Atlantic Bight (GB, SNE, and MAB
areas, respectively, ‘southern’). Red hake, silver hake,
yellowtail flounder, winter flounder, Atlantic cod, and
haddock were analyzed by ecoregion. Additionally,
species limited to the GOM were considered cool-
water species for the study area and were classified as
‘northern’ species, while those species that were lim-
ited to GB, SNE, and MAB or that spanned the entire
range of the study area were classified as ‘southern’
species for ecoregion comparisons.
Spatial metrics. To examine the relationship be-
tween changes in distribution and climate conditions,
6 distributional parameters were calculated: center of
biomass, maximum latitude, minimum latitude, mean
depth of occurence, mean temperature of occurence,
and area occupied. Center of biomass was calculated
in 4 steps. (1) Station latitude and longitude were con-
verted to across-shelf and along-shelf distances from
the 200 m isobath and Cape Hatteras, North Carolina,
in Matlab 7.7 (The Mathworks 2008). This transforma-
tion was needed since the survey follows the curvi-
linear shelf and straight averages of latitude and longi-
tude can result in centers of biomass off the shelf and
outside of the survey area. (2) Biomass-weighted mean
along-shelf and across-shelf location was calculated
using:
(1)
where Xis the parameter of interest (along-shelf
location, across-shelf location, temperature, etc.), jis
the survey year, wiis the log-transformed biomass [log
(x+ 1) in kg] for each station i. (3) Biomass-weighted
mean values of across-shelf and along-shelf distances
X
wX
w
j
iij
i
n
i
==
1
113
Species Scientific name Family Habitat Species historical range Ecoregion
Spiny dogfish Squalus acanthias Squalidae D Greenland to Argentina S
Little skate Leucoraga erinacea Rajidae D Newfoundland to North Carolina S
Thorny skate Amblyraja radiata Rajidae D Greenland to New York N
Winter skate Leucoraja ocellata Rajidae D Newfoundland to North Carolina S
Alewife Alosa pseudoharengus Clupeidae P Labrador to South Carolina S
American shad Alosa sapidissima Clupeidae P Labrador to Florida S
Atlantic herring Clupea harengus Clupeidae P Greenland to South Carolina S
Atlantic cod Gadus morhua Gadidae D Greenland to North Carolina N & S
Haddock Melanogrammus aeglefinus Gadidae D New Jersey to Newfoundland N & S
Pollock Pollachius virens Gadidae P Greenland to North Carolina N
Silver hake Merluccius bilinearis Gadidae P Newfoundland to South Carolina N & S
Red hake Urophycis chuss Merlucidae D Newfoundland to North Carolina N & S
Spotted hake Urophycis regia Merlucidae D Maine to Gulf of Mexico S
White hake Urophycis tenuis Merlucidae D Newfoundland to North Carolina N
Cusk Brosme brosme Lotidae D Newfoundland to New Jersey N
Goosefish Lophius americanus Lophiidae D Newfoundland to North Carolina S
American plaice Hippoglossoides platessoides Pleuronectidae D Labrador to Rhode Island N
Atlantic halibut Hippoglossus hippoglossus Pleuronectidae D Labrador to Virginia N
Yellowtail flounder Limanda ferruginea Pleuronectidae D Labrador to Virginia N & S
Winter flounder Pseudopleuronectes americanus Pleuronectidae D Labrador to Virginia N & S
Fourspot flounder Paralichthys oblongus Paralichthyidae D George’s Bank to Florida S
Summer flounder Paralichthys dentatus Paralichthyidae D Massachusetts to South Carolina S
Windowpane flounder Scophthalmus aquosus Scophthalmidae D Newfoundland to Florida S
Atlantic mackerel Scomber scombrus Scombridae P Labrador to North Carolina S
Ocean pout Zoarces americanus Zoarcidae D Labrador to Delaware S
Atlantic wolffish Anarhichas lupus Anarhichadidae D Greenland to New Jersey N
Blackbelly rosefish Helicolenus dactylopterus Scorpaenidae D Nova Scotia to Venezuela S
Longhorn sculpin Myoxocephalus octodecemspinosus Cottidae D Newfoundland to Virginia S
Sea raven Hemitripterus americanus Hemitripteridae D Labrador to Virginia S
Acadian redfish Sebastes fasciatus Sebastidae D Massachusetts to Newfoundland N
Table 1. Northeast USA fish stocks analyzed in the present study with scientific name, family, habitat (P: pelagic; D: demersal), spe-
cies historical range, and ecoregion by which stocks were separated for this study (N: northern; S: southern). Species are ordered by
taxonomy from most to least ancestral
Mar Ecol Prog Ser 393: 111–129, 2009
were then converted back to mean latitude and mean
longitude values for analysis. (4) Distance between
each center of biomass and a fixed point (35° N latitude
and 75° W longitude) was calculated. This fixed point
roughly corresponds to Cape Hatteras, NC, and is
below the southernmost extent of the NEFSC trawl
survey (Fig. 1). We used the great circle distance for-
mula to calculate the distance (d):
(2)
where mlat is the mean latitude and mlon is the mean
longitude at the center of biomass. All latitudes and
longitudes were converted to radians. The value of d,
hereafter called distance, was converted to kilometers
by assuming the earth is spherical with a radius of
6367 km. Changes in distance incorporate both along-
shelf and across-shelf shifts in distribution.
Biomass-weighted mean depth and mean tempera-
ture were calculated for each stock using Eq. (1). The
parameter of interest (Xij)was bottom depth or bottom
temperature at station iduring survey year j. The
minimum and maximum latitude at which each stock
was present in each year was derived from station
location data.
Area occupied by each stock in each year was calcu-
lated by dividing the survey area into 10-minute spa-
tial cells (261 km2) and then allocating each station to
the appropriate cell, similar to the method of Methratta
& Link (2007). The area of grids in which the species
was present was summed to represent the total area
that the stock occupied. In each year, some 10-minute
spatial cells were not sampled so this is a conservative
measure of ‘area occupied’ and an indicator of area rel-
ative to the survey time series.
Annual relative abundance and biomass for each
species was estimated as the stratified mean numbers
and biomass (kg) per tow respectively.
Environmental metrics. Two indices of temperature
variability over the ecosystem were used: mean bottom
temperatures (BT) during the spring survey and mean
annual sea surface temperatures (SST). Mean-strati-
fied annual BT of each survey was estimated from BT
measurements taken at each station during the spring
survey. Prior to 1990, BTs were measured with water
bottles, and, from then on, BTs were measured with
conductivity, temperature, and depth profilers (CTDs).
SST on the continental shelf was estimated using the
extended reconstructed sea surface temperature data-
set (ERSST Version 3, www.ncdc.noaa.gov/oa/climate/
research/sst/ersstv3.php). This dataset is based on the
SST compilation of the International Comprehensive
Ocean– Atmospheric Data Set and uses interpolation
procedures to reconstruct SST in regions with sparse
data (Smith & Reynolds 2004). While the dataset
extends back to 1854, we used values from 1968 to
2007, corresponding to the timing of the NEFSC trawl
survey where the 95% confidence interval of the data
was 0.1°C or less (Smith & Reynolds 2004). The spatial
resolution of the data was 2° longitude by 2° latitude,
and we used only grids that overlapped spatially with
our survey area to calculate the mean annual SST,
hereafter simply SST.
Two indices of longer term climatological conditions
were also used: the North Atlantic Oscillation (NAO)
and Atlantic Multidecadal Oscillation (AMO). The
NAO index was calculated as the difference between
surface pressure of the subtropical (Azores) high and
the subpolar (Iceland) low. The mean winter NAO
index was used because most of the variance in the
NAO occurs in the winter months and is the only
large-scale pressure and circulation pattern that is
evident throughout the year in the northern hemi-
sphere (Hurrell et al. 2003). Variability in the NAO
has been associated with changes in precipitation,
SST, wind fields, sea-ice formation, and, thus, with
ecosystem change (Beaugrand et al. 2003, Drinkwater
et al. 2003, Greene et al. 2003, Beaugrand 2004) and
fish recruitment (O’Brien et al. 2000, Reid et al. 2001,
Lindley et al. 2003).
The AMO is a 65 to 80 yr cycle in the North Atlantic
thought to be driven by ocean thermohaline circulation
(Sutton & Hodson 2005). The AMO is based on the
detrended Kaplan SST dataset (5° latitude ×5° longi-
tude grids) from 0 to 70° N. The linear effects of anthro-
pogenic climate change are removed to represent the
natural variation in SSTs that has been observed over
the last 150 yr and has been correlated with natural
fluctuations over the last millennium (Gray et al. 2004,
Sutton & Hodson 2005). The AMO is driven by thermo-
haline circulation and is associated with warmer land
and ocean temperatures, decreases in summer precip-
itation, and increases in the number of droughts (Sut-
ton & Hodson 2007). In the United States, the area rel-
evant to the present study, these climate effects are
most pronounced in the summer and somewhat less
prominent in the autumn and winter at lower latitudes
(Sutton & Hodson 2007).
Analysis. To statistically test for changes in spatial
distribution measures over time we conducted linear
regressions incorporating the appropriate temporal
autocorrelation into the error structure for each stock
(Proc Autoreg, SAS Institute, v9.1.3). We calculated a
separate linear regression over time for each stock and
for each distributional measure: mean distance, mini-
mum latitude, maximum latitude, mean depth of
occurrence, mean temperature of occurrence, and area
occupied. A change in mean distance represents a shift
in the center of biomass and is the strongest indicator
of distributional shifts. Minimum and maximum lati-
dA mlat
mlat
+
××
cos[(sin ) sin( )]
cos{ cos( ) cos[
35
35 mmlon −−( )]}75
114
Nye et al.: Changing spatial distribution of Northeast USA fish stocks
tude were also used as indictors of distributional shifts.
While the center of biomass for a species may shift due
to spatial differences in biomass, minimum and maxi-
mum latitude were determined only by presence, i.e. a
change in minimum or maximum latitude over time
may indicate the recent occurrence of a stock outside
its previous range or the local extirpation at the edge of
a stock’s range.
Additionally, maximum and minimum latitude can
be used in conjunction with area occupied to deter-
mine range contraction or expansion. Evidence of
range expansion includes a statistically significant
increase in area occupied, an increase in maximum lat-
itude, and/or a decrease in minimum latitude. Evi-
dence of range contraction includes a statistically sig-
nificant decrease in the area occupied, a decrease in
maximum latitude, and/or an increase in minimum lat-
itude. We examined the relationship between area
occupied and abundance (log-transformed number of
individuals) for each stock using a power function to
understand the effects of changes in abundance on
spatial distribution.
ArcGIS software was used to create smoothed maps
in 5 yr time blocks of several representative species
using the inverse distance weighting (IDW) method.
For each species, raster size was maintained at a con-
stant level and the smoothing power, p, was equal to 2.
The spatial pattern of mean annual BT for each 5 yr
time block was also depicted using IDW in ArcGIS. In
this analysis we used inshore and offshore strata to
visualize spatial trends.
Because there are many possible distributional
responses of fishes to changes in both temperature
regime and population abundance, we used canonical
correlation analysis (CanCorr) to examine the linear
relationship between these multiple response vari-
ables and multiple explanatory variables in each stock
(Proc Cancorr, SAS Institute). CanCorr is a multivari-
ate generalization of correlation analysis that relates 2
sets of multiple variables. A pair of canonical variates
is generated from each set of original variables, where
each canonical variate is a linear combination of the
original variables. The explanatory and response vari-
ables, and their abbrevations, are given in Table 2.
To evaluate the effect of biogeography on distribu-
tional responses, stocks were classified as either
‘northern’ or ‘southern’ depending on their range.
Stocks restricted to the GOM area and north were clas-
sified as ‘northern’ stocks and stocks whose ranges ex-
tended south of the GOM were classified as ‘southern’
(Table 1). The slopes from the linear regressions of dis-
tributional metrics and time were compared between
the 2 classifications using a Mann-Whitney U-test. His-
tograms of the slopes of the linear regressions were
compared for each of 6 distributional metrics.
RESULTS
Environmental metrics
Several metrics indicate a recent warming trend on
the NE United States continental shelf (Figs. 2 & 3).
The spatial distribution of BTs taken from the trawl
survey indicates that, over the last 40 yr, the coolest
time period occurred at the beginning of the survey,
from 1968 to 1972 (Fig. 2). Temperatures were the
warmest from 1973 to 1977 and then from 1998 to 2002.
The spatial patterns of BTs reflect the temporal trends
in both BTs and SSTs; temperatures have been consis-
tently warm since the late 1990s, except in the cool
years from 2003 to 2005 (Fig. 3a,b). SST anomalies
have been positive for the past 10 yr (Fig. 3b). Trends
in the winter NAO index are more variable, but, for the
past 10 yr, have been closer to the long-term mean
(Fig. 3c). AMO anomalies, on the other hand, have
been increasing steadily since the earlier 1970s, indi-
cating changes in circulation patterns and a general
warming over the entire North Atlantic, a spatial and
temporal scale that is larger than the NE United States
shelf, the scale to which BT and SST correspond.
Distributional metrics
There were clear poleward shifts consistent with
warming in many fish stocks of the NE United States,
most notably for alewife, American shad, silver hake
southern, red hakesouthern, and yellowtail flounder
southern (Fig. 4). Linear regressions of the annual cen-
ter of biomass indicated a statistically significant pole-
115
Variables Abbreviation
Explanatory
Bottom temperature BT
Extended reconstructed sea surface SST
temperature
Atlantic Multidecadal Oscillation AMO
North Atlantic Oscillation NAO
Relative biomass BIOMASS
Response
Distance the mean center of biomass distance
shifted poleward
Minimum latitude minlat
Maximum latitude maxlat
Mean depth of occurrence depth
Mean temperature of occurrence temp
Relative area of occupation area
Table 2. Explanatory and response variables used in statistical
analyses with corresponding abbreviations
Mar Ecol Prog Ser 393: 111–129, 2009116
Fig. 2. Spatial represen-
tation of bottom temper-
atures taken during the
Northeast Fisheries Sci-
ence Center (NEFSC)
surveys in 5 yr time
blocks
Nye et al.: Changing spatial distribution of Northeast USA fish stocks
ward shift in 17 of the 36 stocks, including a broad
range of taxa and life histories (Table 3). The center of
biomass for 4 species appeared to shift southward,
which would be inconsistent with a response to warm-
ing. These species included 2 elasmobranchs (winter
skate and little skate) and 2 stocks found primarily in
the GOM (Atlantic codGOM and winter flounder
northern).
A statistically significant increase in the area occu-
pied, indicative of a range expansion, was found for 10
stocks (Table 3). In addition, decreases in minimum
latitude and increases in maximum latitude, which are
also indicative of range expansion, were found in 1 and
5 stocks, respectively. Strong evidence of range expan-
sion includes a statistically significant increase in max-
imum latitude and a decrease in minimum latitude, in
addition to an increase in area occupied. American
shad and spotted hake fit all 3 criteria. The range
expansions (area occupied) of the largest magnitude
occurred in Atlantic herring and spotted hake and cor-
respond to increases of roughly 34 000 and 20 000 km2,
respectively, over the 40 yr time series.
Twelve species experienced range contraction, the
largest of which occurred in thorny skate, Atlantic
codGB, and yellowtail floundersouthern (Table 3).
The magnitude of these range contractions corre-
sponds to a reduction in the area occupied by these
stocks of from 18 000 to 21 000 km2over the 40 yr time
period. All stocks exhibiting a significant range con-
traction were northern species at the southern extent
of their range. Strong evidence of range contraction
includes statistically significant decreases in the area
occupied, a decrease in maximum latitude, and an
increase in minimum latitude. Atlantic codGB, Aca-
dian redfish, Atlantic wolffish, and cusk fit these crite-
ria (Table 3). Six stocks exhibited range contraction, as
well as poleward movement, and, again, all of these
stocks were northern species at the southern extent of
their range.
Changes in minimum and maximum latitude corrob-
orate the changes in area occupied and poleward
movement in most stocks, but also reveal more subtle
changes in spatial distribution of other stocks. Al-
though there was no change in the center of biomass or
area occupied for summer flounder, fourspot flounder,
and sea raven, shifts in their distribution may have
occurred, as indicated by changes in minimum and
maximum latitude. The maximum latitude of summer
flounder and fourspot flounder has shifted poleward,
suggesting a poleward movement and/or range ex-
pansion. The minimum latitude of sea raven has in-
creased, suggesting a poleward movement or range
contraction of this species.
Statistically significant increases in mean depth of oc-
currence were detected for 17 stocks (Table 3); 8 of
these stocks also exhibited poleward shifts in distribu-
tion (Table 3, Fig. S1 in the Electronic Supplement,
available at www.int-res.com/articles/suppl/m393p111_
app.pdf). Five stocks exhibited no significant change in
the center of biomass, while winter skate and Atlantic
codGOM appear to have shifted their biomass to
greater depths, while moving away from the poles.
Thorny skate, pollock, cusk, and Atlantic wolffish were
117
5
6
7
8
9
–2
–1
0
1
2
–6
–3
0
3
6
Sea surface
temperature, SST (°C)
Trawl bottom
temperature, BT (°C)
Winter NAO
index
AMO index
a
b
c
–0.4
–0.2
0
0.2
0.4
1968 1978 1988 1998 2008
d
Fig. 3. Time series of environmental variables representing the
temperature and climatological trends for the area of the
Northeast Fisheries Science Center (NEFSC) survey that we
examined. Trends in temperature regime are represented by
(a) mean annual bottom temperatures taken during the NEFSC
trawl survey and (b) standardized mean sea surface tempera-
ture for the area of the bottom trawl survey. Large-scale clima-
tological trends are represented by the (c) standardized winter
NAO (North Atlantic Oscillation) index and (d) standardized
AMO (Atlantic Multidecadal Oscillation) index
Mar Ecol Prog Ser 393: 111–129, 2009118
Yea r
Distance from southernmost point of survey (km)
1970
1980 1990 2000
600
700
Spiny dogfish
1970 1980 1990 2000
750
850
Little skate
1970 1980 1990 2000
970
1000
1030 Thorny skate
1970 1980 1990 2000
840
900
960
Winter skate
1970 1980 1990 2000
600
800
Alewife
1970 1980 1990 2000
500
700
900
American shad
1970 1980 1990 2000
500
700
900
Atlantic herring
1970 1980 1990 2000
940
980
1020
Atlantic cod-GOM
1970 1980 1990 2000
920
960
1020
Atlantic cod-GB
1970 1980 1990 2000
950
1050
Haddock-GOM
1970 1980 1990 2000
960
1020
1080
Haddock-GB
1970 1980 1990 2000
980
1040
1100 Pollock
1970 1980 1990 2000
980
1010
1040
Silver hake-northern
1970 1980 1990 2000
550
700
850
Silver hake-southern
1970 1980 1990 2000
950
980
1010
Red hake-northern
1970 1980 1990 2000
650
750
850
Red hake-southern
1970 1980 1990 2000
250
350
450
Spotted hake
1970 1980 1990 2000
850
950
White hake
Fig. 4. Time series of the mean center of biomass in 36 stocks caught during the Northeast Fisheries Science Center (NEFSC)
spring multispecies survey. Mean center of biomass is measured as the distance from the southernmost point of the survey
Nye et al.: Changing spatial distribution of Northeast USA fish stocks 119
1970 1980 1990 2000
1000
1060
Cusk
1970 1980 1990 2000
750
850
Goosefish
1970 1980 1990 2000
980
1000
American plaice
1970 1980 1990 2000
1000
1100
1200
Atlantic halibut
1970 1980 1990 2000
950
1050
Yellowtail flounder-northern
1970 1980 1990 2000
750
850
950
Yellowtail flounder-southern
1970 1980 1990 2000
900
1050
Winter flounder-northern
1970 1980 1990 2000
850
950
Winter flounder-southern
1970 1980 1990 2000
580
640
700
Fourspot flounder
1970 1980 1990 2000
300
500
Summer flounder
1970 1980 1990 2000
700
800
900
Windowpane
1970 1980 1990 2000
200
400
600
800 Atlantic mackerel
1970 1980 1990 2000
950
1050
Atlantic wolffish
1970 1980 1990 2000
760
800
840
Ocean pout
1970 1980 1990 2000
400
800
Blackbelly rosefish
1970 1980 1990 2000
800
900
Longhorn sculpin
1970 1980 1990 2000
900
940
980
Sea raven
1970 1980 1990 2000
1000
1020
Acadian redfish
Year
Distance from southernmost point of survey (km)
Fig. 4 (continued)
Mar Ecol Prog Ser 393: 111–129, 2009
found at increasingly greater depth, but had no appar-
ent shifts in the center of biomass. These fishes are gen-
erally found in the GOM. There were several stocks
that were found at increasingly deeper stations, but
that did not produce statistically significant linear in-
creases in depth of occurrence. The southern stocks of
red hake and silver hake may have shifted to deeper
water in more recent years (Fig. S1), but this was not
detected by linear regression (Table 3). Six stocks
were found at increasingly shallow depths, and all of
these species have increased in abundance and/or ex-
hibited a range expansion over the time series (Table 3,
Fig. S2 in the Electronic Supplement, available at www.
int-res.com/articles/suppl/m393p111_app.pdf).
Only 4 of the 36 stocks examined showed a statisti-
cally significant change in mean temperature over
time (Table 3). Of these, American shad and blackbelly
rosefish exhibited statistically significant decreases in
mean temperature concomitant with a poleward shift
in the mean center of biomass. However, Atlantic her-
ring and pollock showed no significant poleward
movement, but changes in temperature of occurrence
and area occupied. The range in mean temperature of
occurrence for all other stocks remained within a 2 to
5°C temperature range over the 40 yr examined,
despite observed spring BTs between 0 and 20°C.
Species’ range or area occupied in these fish stocks
can be predicted by abundance (Fig. S3 in the Elec-
tronic Supplement, available at www.int-res.com/
articles/suppl/m393p111_app.pdf). The relationship
between area occupied and abundance is very strong
for most stocks, but varies among all stocks. There
does not appear to be a strong relationship for some
stocks such as longhorn sculpin, sea raven, winter
flounder– southern and American plaice in which the
range of abundance over the time series is low.
120
Stock Poleward Area Maximum Minimum Mean Mean
movement occupied latitude latitude temperature depth
(km yr–1) (km2yr–1) (°lat yr–1) (°lat yr–1) (°C yr–1) (m yr–1)
Spiny dogfish
Winter skate –1.35 287.4 –0.066 0.78
Little skate –0.83
Thorny skate –517.5 0.012 0.48
Atlantic herring 868.6 0.031
Alewife 5.47 306.4 0.030 1.15
American shad 6.86 205.8 0.051 –0.031 0.98
Silver hake– northern 3.58 143.7 0.016
Silver hake– southern 0.57 407.7 –0.69
Atlantic codGOM –0.83 –0.0081 0.077 0.50
Atlantic codGB 1.48 –454.5
Haddock– GOM
Haddock– GB –246.3
Pollock –281.1 – 0.0066 0.040 0.024 1.36
White hake 2.10 279.8 0.033 0.85
Red hake– northern 0.61 161.5 0.027
Red hake– southern 5.52 330.5 0.41
Spotted hake 3.44 503.6 0.020 –0.46
Cusk –177.5 – 0.031 0.023 0.77
Atlantic halibut 2.57 0.022 1.05
American plaice –231.8 0.0051 0.53
Summer flounder 0.029
Fourspot flounder 0.042
Yellowtail flounder–northern 81.8 0.44
Yellowtail flounder–southern 4.32 496.0 –0.019 0.044
Winter flounder– northern –3.82 110.8 0.39
Winter flounder– southern 1.55 0.035
Windowpane flounder 3.81 0.31
Atlantic mackerel 250.2 –1.38
Acadian redfish –159.8 0.0063 0.037
Blackbelly rosefish 8.53 0.017 –0.04
Longhorn sculpin 1.87
Goosefish 1.61 1.19
Sea raven 0.021
Atlantic wolffish 188.8 0.025 0.51
Ocean pout 0.93 0.025 0.51
Table 3. Summary of trends in time series of distributional responses for 36 stocks on the Northeast United States continental
shelf. Only statistically significant (p < 0.05) slopes are reported. Significant slopes were not detected in any metrics for spiny
dogfish and haddock– GOM. GOM: Gulf of Maine; GB: George’s Bank
Nye et al.: Changing spatial distribution of Northeast USA fish stocks
Multivariate analyses
The factors associated with these shifts in distribu-
tion were examined using CanCorr analysis. There
were significant linear relationships between distribu-
tional measures (response variables) and explanatory
factors (environmental variables and biomass) for all
species. The first canonical variate was significant in
all stocks (p < 0.05). The second canonical variate was
significant in 26 of the 36 stocks (p < 0.05), and, cumu-
latively, the first 2 canonical variates explained at least
83% of the variability in each stock. The advantage
of CanCorr is that correlations between multiple re-
sponse and multiple explanatory variables can be
examined, and correlations within the response and
explanatory variables are also elucidated. These corre-
lations can be visualized in biplots of the loadings
(correlations between the variables and their respec-
tive canonical variates).
Biplots of the loadings are shown only
for 12 stocks in which the first 2 canoni-
cal variates were statistically significant
and the cumulative percentage of the
variance explained was >88% (Figs. 5
to 7). In most stocks, area was closely
loaded with BIOMASS (Figs. 5 to 7),
which would be expected from the tight
relationships described between area
and abundance (Fig. S3). BT and SST
were loaded similarly, but were always
loaded orthogonally to AMO, empha-
sizing the different scales at which
these environmental metrics operate.
Three broad patterns emerged from
the biplots depending on the biogeo-
graphy of the stock. Stocks at the south-
ern extent of their range responded neg-
atively to an increase in the AMO index
(Fig. 5). Stocks at the northern end of
their range responded positively to an in-
crease in the AMO index (Fig. 6). Stocks
confined to the GOM did not shift their
center of biomass in response to warm-
ing, but shifted their depth distribution
(Fig. 7). In stocks at the southern limit of
their range, distance is loaded similarly to
AMO and opposite to both BIOMASS
and area (Fig. 5). This can be interpreted
as the BIOMASS and area of the stock
decreasing, while the center of biomass
(distance) moves increasingly poleward
as warming increases. In southern
stocks, warming, as indicated by the
AMO, is favorable and leads to similar
loadings in AMO, BIOMASS, area, and
distance (Fig. 6). These loadings can be interpreted as an
increase in biomass, area occupied, and a poleward shift
in distribution in response to warming. Stocks limited to
the GOM appear to respond to warming by increasing
their depth preference (Fig. 7). The AMO is loaded sim-
ilarly to depth of occurrence and opposite to BIOMASS
and area, which can be interpreted as a decrease in stock
size, area occupied, and a shift to deeper depths as
warming increases. Some species that had range expan-
sion tend to have significant changes in mean tempera-
ture (Table 1), and BT and mean temperature of occur-
rence are not as highly correlated in CanCorr (Fig. 6).
However, those species in which temperature of occur-
rence was highly correlated with BT were also experi-
encing range contraction and reductions in biomass
(Fig. 5).
A striking aspect of these results is the difference in
response between species located in the southern ex-
tent of the survey as compared to those that are re-
121
Silver hake-southern
distance
minlat
maxlat
depth
temp
area
BIOMASS
BT
NAO
AMO
SST
Atlantic cod-GB
area temp
depth
maxlat
minlat
distance
SST
AMO
NAO
BT
BIOMASS
Red hake-southern
distance
minlat
maxlat
depth
temp
area
BIOMASS
BT
NAO
AMO
SST
Yellowtail flounder-southern
distance
minlat
maxlat
depth
temp
area
BIOMASS
BT
NAO
AMO
SST
Loadings (correlations with first canonical variate)
Loadings (correlations with second canonical variate)
Fig. 5. Loadings (correlations with canonical variates) for the first 2 canonical
axes for the explanatory variables (triangles) and response variables (circles)
for 4 representative stocks responding unfavorably to warming. Unfavorable
responses include a decrease in population size (BIOMASS), decrease in area
occupied (area), and a poleward shift in the center of biomass (distance) with
an increase in Atlantic Multidecadal Oscillation (AMO) index. Scales for both
the x- and y-axes range from –1 to 1. NAO: North Atlantic Oscillation; SST: sea
surface temperature; BT: bottom temperature; depth: depth of occurrence;
maxlat and minlat: maximum and minimum latitude, respectively
Mar Ecol Prog Ser 393: 111–129, 2009
stricted to the GOM or northern extent
of the survey. To illustrate this differ-
ence, maps of red hake distribution
were created in 5 yr time blocks, be-
cause this species occurs both in the
GOM (red hakenorthern) and SNE
GB (red hakesouthern) regions. Red
hake was chosen because it was fished
heavily in both ecoregions until the
mid-1970s, after which fishing pressure
became low and was below FMSY. The
distribution of red hakesouthern has
changed much more noticeably than
the red hakenorthern stock (Fig. 8).
From the 1970s and into the late 1980s,
there were many high-density areas of
red hake in both the northern and
southern stocks. There were even high
densities of fish located south of Long
Island Sound in the MAB region. The
statistical findings that the center of
biomass of the southern red hake stock
shifted while its range contracted are
corroborated by these maps. These
maps also support the statistical find-
ings that the northern red hake stock
did shift poleward in the GOM, but this
shift was of much smaller magnitude
than the poleward shift in the southern
stock of red hake. The movements of
red hake mirror the spatial pattern in
warming observed along the NE United
122
distance
minlat
maxlat
depth
temp
area
BIOMASS
BT
NAO
AMO
SST
area
temp
depth
maxlat
minlat distance
SST
AMO
NAO
BT
BIOMASS
distance
minlat
maxlat
depth
temp
area
BIOMASS
BT
NAO
AMO
SST
distance
minlat
maxlat
depth
temp
area
BIOMASS
BT
NAO
AMO
SST
Summer flounder
Atlantic mackerel American shad
Spotted hake
Loadin
g
s (correlations with first canonical variate)
Loadings (correlations with second canonical variate)
Fig. 6. Loadings (correlations with canonical
variates) for the first 2 canonical axes for the
explanatory variables (triangles) and re-
sponse variables (circles) for 4 representative
stocks responding favorably to warming. Fa-
vorable responses include an increase in
population size (BIOMASS), increase in area
occupied (area), and a poleward shift in the
center of biomass (distance) with an increase
in Atlantic Multidecadal Oscillation (AMO)
index. Scales for both the x- and y-axes range
from –1 to 1. Other abbreviations as in Fig. 5
Ocean pout
distance
minlat
maxlat
depth
temp
area
BIOMASS
BT
NAO
AMO
SST
Pollock
area
temp
depth
maxlat
minlat
distance
SST
AMO
NAO BT
BIOMASS
American plaice
area
temp
depth
maxlat
minlat
distance
SST
AMO
NAO
BT
BIOMASS
Thorny skate
area
temp
depth
maxlat minlat
distance
SST
AMO
NAO
BT
BIOMASS
Loadings (correlations with first canonical variate)
Loadings (correlations with second canonical variate)
Fig. 7. Loadings (correlations with canonical
variates) for the first 2 canonical axes for the
explanatory variables (triangles) and re-
sponse variables (circles) for 4 stocks re-
stricted to the Gulf of Maine responding un-
favorably to warming by changing depth
distribution. Scales for both the x- and y-
axes range from –1 to 1. Other abbreviations
as in Fig. 5
Nye et al.: Changing spatial distribution of Northeast USA fish stocks 123
Fig. 8. Smoothed maps of
red hake spatial distribu-
tion (northern and south-
ern stocks combined) in
5 yr time blocks using
inverse distance weight-
ing. Units of biomass are
in kilograms per tow
Mar Ecol Prog Ser 393: 111–129, 2009
States continental shelf, as shown analogously in 5 yr
time blocks (Figs. 2 & 8). The distributional changes of
red hake reflect those of species with similar stock
structure, but also species that are traditionally found
in SNE and the MAB as compared to those that are
restricted to the GOM.
Distributional responses were different depending on
the biogeography of the species (Fig. 9). The poleward
shift in the center of biomass was much larger in the
southern stocks than the northern stocks (Fig. 9a). South-
ern stocks also exhibited shifts to deeper depths, but of a
smaller magnitude compared to northern stocks
(Fig. 9b). The temperature of occurrence did not change
over time for most stocks, as indicated by the majority of
observations being centered around zero (Fig. 9c). There
was also little difference between the northern stocks
and southern stocks in area occupied, but it does appear
that, in general, northern stocks were more likely to ex-
perience a range contraction (Fig. 9d). Maximum lati-
tude increased for more of the southern stocks (Fig. 9e)
and minimum latitude increased for all stocks except for
2 southern stocks (Fig. 9f). There was a significant differ-
ence between the distributions of slopes
in the 2 ecoregions for only 2 distribu-
tion metrics, distance moved poleward
and maximum latitude (Mann-Whitney;
Fig. 9).
In summary, 24 of the 36 stocks dis-
played statistically significant changes
consistent with warming, as indicated by
a poleward shift in the center of biomass,
an increase in mean depth of occurrence,
and/or an increase in mean temperature.
Additionally, fourspot flounder and
summer flounder show weak indications
of distributional changes consistent
with warming. Ten stocks had distribu-
tional changes that were possibly nonlin-
ear or were inconsistent with a response
to warming.
DISCUSSION
We have demonstrated clear shifts in
spatial distribution for the majority of
fish stocks that we examined on the NE
United States continental shelf. In 24 of
the 36 stocks, we detected statistically
significant changes associated with
large-scale warming. Poleward shifts in
the center of biomass and increases in
the depth distribution were consistent
with the predicted responses to climate
change that previous studies have doc-
umented in many ecosystems (South-
ward et al. 1995, Parmesan & Yohe
2003, Rosenzweig et al. 2008 for re-
views) and in marine fishes (Holbrook
et al. 1997, Perry et al. 2005, Dulvy et al.
2008). The long-term changes in distri-
bution that we detected were appropri-
ately correlated with large-scale warm-
ing and climatic conditions, specifically
the AMO, and were not associated with
measures of annual temperature fluctu-
124
Fig. 9. Histograms comparing distributional responses for northern (gray) and
southern (black) stocks in (a) distance shifted poleward, (b) mean depth, (c) mean
temperature, (d) area occupied, (e) maximum latitude, and (f) minimum latitude.
Significant differences detected with a Mann-Whitney U-test between species
found in the 2 ecoregions are indicated by p-values
Nye et al.: Changing spatial distribution of Northeast USA fish stocks
ations (BT or SST) as previous studies have docu-
mented (Taylor et al. 1957, Murawski 1993, Rose 2005).
Long-term trends in distribution were not correlated
with mean spring BT, indicating that fish populations
have shifted their distribution in response to a consis-
tent warm temperature regime over a large temporal
and spatial scale. A warmer temperature regime, as
indicated by the AMO, is supported by other studies in
the region showing that the range in annual water
temperatures has increased; winter temperatures have
remained fairly constant, while summer temperatures
have increased (Friedland & Hare 2007). Additionally,
annual surface temperatures have increased over time,
while annual BTs have not (Mountain 2003). The asso-
ciation of changes in distribution with large-scale cli-
mate variables rather than annual temperature vari-
ability suggests that these shifts are likely to persist, as
large-scale circulation and temperature changes con-
tinue as a result of global climate change (IPCC 2007,
Solomon et al. 2009). The lack of change in the mean
temperature of occurrence for most species indicates
that species shifted their distribution to remain within
their preferred temperature range.
The present study provides an empirical test of
Murawski’s (1993) hypothesized distributional shifts in
NE United States fish stocks according to a warming
scenario. Murawski (1993) predicted that warm-water
migratory species, such as Atlantic herring and mack-
erel, would be most sensitive to warming, while shal-
low-water sedentary fishes would exhibit smaller
changes in distribution on the NE United States conti-
nental shelf. However, we have shown that distribu-
tional changes were evident for species across many
families, life histories, habitat preferences, and histori-
cal fishing pressures. Even species that Murawski
(1993) termed ‘shallow-water sedentary’ such as yel-
lowtail flounder, winter flounder, windowpane, and
longhorn sculpin showed poleward distributional
shifts. Additionally, we provide preliminary evidence
that the range of summer flounder, also termed a
‘sedentary’ species, has expanded over time, that its
abundance increased, and that the center of biomass
was displaced poleward within the survey area.
Murawski (1993) identified species such as fourspot
flounder, goosefish, halibut, cod, alewife, and silver
hake that had low variance in their mean latitude,
depth, and temperature of occurrence. The present
analysis indicates that these species exhibited some of
the largest shifts in distribution concomitant with a
persistent warming trend along the NE United States
continental shelf. Consistent with Murawski’s (1993)
identification of life-history adaptations to annual tem-
perature change, warm-water migratory species alter
their geographical distribution annually, as indicated
by the close association between annual mean BT and
the mean temperature of occurrence. However, more
sedentary species do not adapt as quickly to interan-
nual temperature differences and, thus, display larger
distributional changes as a result of a consistent warm
temperature regime.
Murawski (1993) also identified a deep-water seden-
tary group that was, in general, restricted to the GOM.
While temperatures in the deep waters of the GOM
may not experience fluctuations in temperature as
large as the shallower parts of the NE United States
continental shelf, we did see range contraction in many
of these species (white hake, pollock, cusk, American
plaice, Acadian redfish, and Atlantic wolffish) and an
increase in the depth of occurrence (Atlantic cod
GOM, pollock, cusk, Atlantic halibut, and ocean pout).
The presence of distributional changes even in areas
where a response was not expected heightens concerns
over how species will respond if warming trends con-
tinue. Our analysis supports Murawski’s (1993) con-
tention that deep-water sedentary fish, particularly
those in the GOM, may not adjust their spatial distribu-
tion in response to warming, but may experience
greater changes in growth, reproduction, and recruit-
ment than those fish that have shifted their distribution.
Distributional responses were most pronounced in
southern species for which their centers of biomass
were historically in SNE and the MAB, as illustrated by
the comparison between the southern and northern
stocks of red hake. The center of biomass for most of
the southern stocks shifted to the northwest over the
time series. Northern stocks traditionally found in the
GOM, including the northern red hake stock, shifted
poleward only slightly, if at all. Thus, the bathymetry
and geography of the survey area was important in
detecting changes in spatial distribution. Spatial distri-
bution of temperature indicates a lack of suitable cool
deep-water habitats to which GOM stocks could move.
If fish shifted directly poleward or in a northwesterly
direction from their historical GOM habitat, the shal-
low waters of the Bay of Fundy and Scotian shelf would
be suboptimal for many of the cold deep-water fish of
the GOM. Although stocks that were restricted to the
GOM generally did not exhibit poleward shifts in their
centers of biomass, they exhibited increases in mean
depth of occurrence and temperature of occurrence, as
was the case in cusk and pollock. Although tempera-
tures in the GOM are generally more stable than in
other regions of the present study, temperature has
increased slightly in the GOM over the survey time
series (Friedland & Hare 2007) and there has been a
recent decrease in salinity and increase in stratification
(Drinkwater et al. 2009) to which GOM organisms will
have to adapt.
CanCorr effectively separated out the effects of
warming from changes in biomass for most of the stocks
125
Mar Ecol Prog Ser 393: 111–129, 2009
examined. The overwhelming response of these stocks
to warming was a poleward shift in the center of bio-
mass, and the response to an increase in biomass was a
range expansion. Conversely, a range contraction was
associated with a decrease in biomass more than with a
change in temperature. CanCorr analysis also empha-
sized that distributional responses involve the complex
interactions between fluctuations in biomass and envi-
ronmental forcing. Consistent with hypothesized spa-
tial shifts induced by warming (Hughes 2000, Walther
et al. 2002, Parmesan & Yohe 2003, Root et al. 2003),
many southern species shifted their range northward,
but they also experienced more favorable conditions for
growth and recruitment, resulting in abundance in-
creases and range expansion (Hare & Able 2007). More
northerly species made only slight shifts north, but their
ranges contracted, or they were found at increasingly
greater depths.
Previous studies have identified remarkably consis-
tent fish communities that have persisted over a long
time along the NW Atlantic coast (Overholtz & Tyler
1985, Gabriel 1992, Mahon et al. 1998). However,
Gabriel (1992) found that the boundaries of southern
MAB and GB species assemblages were the most vari-
able and seemed to be modified by annual differences
in temperature. Our analysis supports this finding, and,
while we only analyzed 36 stocks, we suspect an analy-
sis of the finfish community would indicate a poleward
shift in the southern boundary of the MAB and GB fish
community, as has been found previously along the NW
Atlantic coast (Gabriel 1992, Mahon et al. 1998). We
have shown here that fish species have shifted their dis-
tributions at different rates. Thus, the way in which dis-
tributional shifts have changed community structure
and species interactions should be evaluated.
Many fish species undergo ontogenetic shifts in
depth distribution (MacPherson & Duarte 1991, Me-
thratta & Link 2007, Labropoulou & Damalas 2008), a
well-documented relationship in the ecological litera-
ture termed ‘Heincke’s Law’ (Heincke 1913). Depth dis-
tribution might be expected to change as many heavily
fished stocks experience fluctuations in abundance as
well as in size structure. We did not investigate the ef-
fects of size structure on distributional responses, but, if
age structure were to be the main cause of distribu-
tional changes, we would expect heavily fished stocks
with truncated age structure to be found at increasingly
shallower depths and recovering stocks with expand-
ing age structure to be found at increasingly deeper
depths. However, no stock found at increasingly deeper
stations could be explained by an increase in abun-
dance. Only silver hake–northern was found at in-
creasingly shallow depths, which could be explained
by a decrease in abundance over the course of the time
series and a recent stock structure composed of more
Age 1 fish and few fish older than Age 3 (NEFSC 2001).
Additionally, in the CanCorr analysis biomass was
never loaded similarly with depth of occurrence.
Distributional shifts may be obscured by changes in
biomass, especially if species are found frequently out-
side the survey area or if their distribution is limited by
bathymetry and habitat availability (Spencer 2008).
For example, we observed no significant distributional
shifts in haddock found in either the GOM or GB/SNE
ecoregion. The abundance of haddock has increased
considerably since its lowest biomass levels in the
1990s. Therefore, shifts in the center of biomass by this
stock might be masked by its recent increase in popu-
lation size and may not be detected by linear statistical
methods. Similarly, we saw southward shifts in the
center of biomass with range expansion in winter
skate, spiny dogfish, and, to a lesser extent, little skate.
All 3 of these elasmobranch species experienced sig-
nificant increases in abundance in the 1990s, but the
ranges of these species extend beyond the boundaries
of our survey, making trends in spatial distribution dif-
ficult to interpret presently. In the future, it will be
important to compare surveys from adjacent areas to
accurately describe distributional changes in fish
stocks and to accurately assess their population sizes
(Blanchard et al. 2008).
We found a tight relationship between abundance
and the area occupied in each species, and this finding
was corroborated by the similar loadings between bio-
mass and area in the CanCorr. This relationship is well
documented in the ecological literature (Brown 1984,
MacCall 1990, Frank & Shackell 2001), so these tight
relationships are not surprising. By illustrating this
tight relationship in all 36 stocks examined, we empha-
size that changes in abundance are expressed in the
area occupied by each species and that changes in the
center of biomass are a response to changes in envi-
ronmental conditions. Most of the species in the pre-
sent study have been subjected to the direct or indirect
effects of fishing. However, spotted hake, blackbelly
rosefish, sea raven, and longhorn sculpin have not
been subjected to a direct fishery, and all indicated
poleward shifts in their distribution suggesting further
that these shifts are in response to climatic events and
not to fishing practices or changes in abundance.
A strength of this work is that we have shown distrib-
utional responses consistent with warming in multiple
stocks with a variety of life-history characteristics and
habitat preferences; this provides a kind of replicated
test of our hypothesis that recent warming has caused
shifts in spatial distribution. These shifts were most
highly correlated with the AMO, but the mechanisms
by which changes in distribution occur are still un-
known. In the NE United States, positive AMO phases
are associated with warmer land and ocean tempera-
126
Nye et al.: Changing spatial distribution of Northeast USA fish stocks
tures, increases in rainfall and river flow in the GOM
region, and decreases in rainfall and river flow in the
SNE and MAB areas (Enfield et al. 2001, Sutton & Hod-
son 2007). Both temperature and river flow have been
shown to influence fish distribution and recruitment dy-
namics (Hare & Able 2007, Wood & Austin 2009, Hare
et al. in press). The AMO has also been associated with
basin-scale circulation changes, changes in salinity and
temperature, and bio-geographical shifts in plankton,
fish, and whales in the NE Atlantic (Hatun et al. 2005,
Beaugrand et al. 2009). In contrast, the way in which
the AMO affects ecosystems of the NW Atlantic is
poorly understood. However, it has been hypothesized
that synchronous dynamics, but opposite in sign, in
functionally similar fish stocks in the NE and NW At-
lantic are connected by unknown mechanisms that in-
fluence the regions in opposite ways (Megrey et al.
2009). Thus, the AMO may be an index of oceanic cir-
culation that synchronously influences both sides of the
North Atlantic and causes poleward shifts in spatial dis-
tribution of marine organisms, now documented in both
eastern and western Atlantic ecosystems.
The AMO could influence fish distribution in several
ways, all centering on the thermal environments expe-
rienced by these fishes. We propose 4 potential mech-
anisms and suggest ways in which to test these hypo-
theses. (1) Large-scale changes in temperature might
induce directed movement of fish from warm waters to
cooler waters within their preferred temperature range
that are either deeper or poleward of their historical
distribution. While directed movement to more favor-
able habitat occurs in many species, it probably occurs
at a smaller spatial scale than the substantial shifts in
distribution that we observed. Large movements on
the order of 100s of kilometers and reductions in the
species’ range on the order of 20 000 km2are probably
not the result of directed movement alone. Large-scale
tagging experiments and continued temporally and
spatially intensive sampling over multiple years would
be needed to document directed movement.
(2) Migration patterns and timing may change in
response to increases in temperature that are better
represented by the AMO index than by snapshots of
BT at the time of our survey. Many of the fishes in the
present study migrate up and down the east coast of
the United States in response to seasonal changes in
temperature. However, as the temperature regime
warms, cold-water species may not migrate as far
south in the winter; thus, we would observe a poleward
shift in distribution when the species are sampled in
the spring. Similarly, warm-water species may be able
to move poleward and expand their range as summers
are consistently warmer. Large-scale tagging studies
might elucidate such a mechanism, but intensive sum-
mer- and winter-time sampling might also capture
changes in the northern and southern limits to which
stocks migrate.
(3) A third mechanism resulting in large population
shifts in distribution is that large-scale changes in tem-
perature and circulation represented by the AMO may
increase mortality at the southern extent of a species’
range, particularly in early life stages. While hatching
at lethal temperatures would directly increase the mor-
tality of eggs and larvae, reduced growth rate at subop-
timal temperatures may indirectly increase mortality
via an increase in predation and starvation (Houde
1987, 1989). Furthermore, changes in circulation may
transport eggs and larvae to suboptimal nursery habi-
tat, and alterations in the phenology of oceanographic
processes, such as stratification, may also have an im-
pact on the recruitment success of these fishes. Mortal-
ity at the southern extent of a species’ range, where
temperature is most likely to be above the species’ pre-
ferred temperature, may explain the shift in spatial dis-
tribution and would be reflected in stockrecruitment
relationships. An analysis of stockrecruitment dynam-
ics of meta-populations would clarify whether there are
differences in mortality and recruitment among meta-
populations of the same species. We hypothesize that
stocks at the southern extent of their range have experi-
enced unfavorable environmental conditions for re-
cruitment during the recent positive AMO phase, lead-
ing to the local extirpation of many of these stocks at the
southern range limit. Conversely, stocks at the northern
extent of a species’ range may have experienced favor-
able conditions for recruitment, leading to an increase
in abundance and range expansion.
(4) From an ecosystem perspective, changes in tem-
perature and salinity represented by the AMO may
cause changes in the phytoplankton or zooplankton
community, the effects of which may transfer through
the food web. Fish movement, migration, and recruit-
ment might all be directly influenced by abiotic factors
such as temperature, salinity, and circulation patterns,
but they may also be influenced indirectly by changes
in the distribution and abundance of lower trophic
level organisms. There may be some directed move-
ment by fish to follow prey. In addition, changes
in abundance, distribution, and composition of the
plankton community may allow high recruitments and
population growth in some species and the opposite
response in others. A holistic understanding of eco-
system functioning would not only explain the shifts
in fish distribution that we have documented, but
may elucidate a positive feedback mechanism that ex-
plains why we have observed shifts on the order of 1 to
8 km yr–1, similar to those in other studies on marine
fish (Perry et al. 2005, Mueter & Litzow 2008). For com-
parision, the average shift in terrestrial ecosystems is
about 0.61 km yr–1 (Parmesan & Yohe 2003).
127
Mar Ecol Prog Ser 393: 111–129, 2009
We examined the role of population abundance, but
were unable to directly examine the role of fishing in
distributional shifts. Fishing pressure may intensify the
effects of climate change (Hsieh et al. 2008, Planque et
al. in press), especially at the southern extent of the
species’ range in this region. Commercially important
species such as red hake, Atlantic cod, yellowtail
flounder, alewife, and American shad have historically
experienced intense fishing pressure, and all of these
stocks are at record low levels, particularly at the
southern limit of their range. For many of these spe-
cies, management has reduced fishing mortality, yet
the southern stocks have not been able to recover to
historic population sizes or to support viable fisheries.
We hypothesize that these species have failed to fully
recover from intense fishing at the southern extent of
their range, in part because of warming along the con-
tinental shelf. The observations that stock size has
decreased and the center of biomass has shifted pole-
ward in some southern stocks of these species, while
stock size has increased in northern stocks of these
species suggests that there may be movement of fish
from the southern stock into the northern stock. This
may be the case with the red hake, silver hake, yellow-
tail flounder, and winter flounder and has important
management implications. The effects of climate
and/or changes in distribution should be incorporated
into stock assessments and management scenarios.
Incorporating a broader set of considerations into stock
assessments by using an index of warming such as the
AMO might improve stockrecruitment relationships,
explain the lack of recovery to date for some stocks,
and provide more refined management advice.
Acknowledgements. This work was supported by a NOAA
NMFS Fisheries and the Environment (FATE) grant. We
thank all those who participated in the NEFSC bottom trawl
survey. We thank M. Fogarty, K. Friedland, D. Hart, L. Jacob-
son, C. Keith, R. McBride, and M. Taylor for thoughtful com-
ments on this work.
LITERATURE CITED
Azarovitz TR (1981) A brief historical review of the Woods
Hole Laboratory trawl survey time series. In: Doubleday
WG, Rivard D (eds) Bottom trawl surveys. Can Spec Publ
Fish Aquat Sci 58:62– 67
Beaugrand G (2004) The North Sea regime shift: evidence,
causes, mechanisms and consequences. In: Regime shifts
in the ocean: reconciling observations and theory. Prog
Oceanogr 60:245– 262
Beaugrand G, Brander KM, Lindley JA, Souissi S, Reid PC
(2003) Plankton effect on cod recruitment in the North
Sea. Nature 426:661– 664
Beaugrand G, Luczak C, Edwards M (2009) Rapid biogeo-
graphical plankton shifts in the North Atlantic Ocean.
Glob Change Biol 15:1790– 1803
Belkin IM (2009) Rapid warming of large marine ecosystems.
Prog Oceanogr 81:207–213
Blanchard JL, Maxwell DL, Jennings S (2008) Power of moni-
toring surveys to detect abundance trends in depleted pop-
ulations: the effects of density-dependent habitat use,
patchiness, and climate change. ICES J Mar Sci 65:111–120
Brown JH (1984) On the relationship between abundance and
distribution of species. Am Nat 124:255– 279
Clark RA, Fox CJ, Viner D, Livermore M (2003) North Sea cod
and climate change-modelling the effects of temperature
on population dynamics. Glob Change Biol 9:1669– 1680
Drinkwater KF (2005) The response of Atlantic cod (Gadus
morhua) to future climate change. ICES J Mar Sci 62:
1327–1337
Drinkwater KF, Belgrano A, Borja A, Conversi A and others
(2003) The response of marine ecosystems to climate vari-
ability associated with the North Atlantic Oscillation. In:
Hurrell J, Kushnir Y, Ottersen G, Visbeck M (eds) The
North Atlantic Oscillation, climatic significance and envi-
ronmental impact. Geophys Monogr 134:211–234
Drinkwater KF, Mueter F, Friedland KD, Taylor M, Hunt GL
Jr, Hare J, Melle W (2009) Recent climate forcing and
physical oceanographic changes in Northern Hemisphere
regions: a review and comparison of four marine eco-
systems. Prog Oceanogr 81:10–28
Dulvy NK, Rogers SI, Jenning S, Stelzenmuller V, Dye SR,
Skjoldal HR (2008) Climate change and deepening of the
North Sea fish assemblage: a biotic indicator of warming
seas. J Appl Ecol 45:1029– 1039
Enfield DB, Mestas-Nunez AM, Trimble PJ (2001) The
Atlantic multidecadal oscillation and its relation to rainfall
and river flows in the continental U.S. Geophys Res Lett
28:2077–2080
Fogarty MJ, Incze L, Hayhoe K, Mountain DG, Manning J
(2007) Potential climate change impacts on Atlantic cod
(Gadus morhua) off the northeastern USA. Mitig Adapt
Strateg Glob Change 13:453– 466
Frank KT, Shackell NL (2001) Area-dependent patterns of fin-
fish diversity in a large marine ecosystem. Can J Fish
Aquat Sci 58:1703– 1707
Frank KT, Perry RI, Drinkwater KF (1990) Predicted response
of Northwest Atlantic invertebrate and fish stocks to CO2
induced climate change. Trans Am Fish Soc 119:353–365
Friedland KD, Hare JA (2007) Long-term trends and regime
shifts in sea surface temperature on the continental shelf of
the northeast United States. Cont Shelf Res 27:2313–2328
Gabriel WL (1992) Persistence of demersal fish assemblages
between Cape Hatteras and Nova Scotia, Northwest
Atlantic. J Northwest Atl Fish Sci 14:29–46
Gray ST, Graumlich LJ, Betancourt JL, Pederson GT (2004) A
tree-ring based reconstruction of the Atlantic Multi-
decadal Oscillation since 1567 A.D. Geophys Res Lett 31:
L12205, doi:10.1029/2004GL019932
Greene CH, Pershing AJ, Conversi A, Planque B and others
(2003) Trans-Atlantic responses of Calanus finmarchicus
populations to basin-scale forcing associated with the
North Atlantic Oscillation. Achievements of the continu-
ous plankton recorder survey and a vision for its future.
Prog Oceanogr 58:301– 312
Hare JA, Able KW (2007) Mechanistic links between climate
and fisheries along the east coast of the United States:
explaining population outbursts of Atlantic croaker
(Micropogonias undulatus). Fish Oceanogr 16:3145
Hare JA, Alexander M, Fogarty M, Williams E, Scott J (in
press) Forecasting the dynamics of a coastal fishery spe-
cies using a coupled climate– population model. Ecol Appl
Hatun H, Sando AB, Drange H, Hansen B, Valdimarsson H
(2005) Influence of the Atlantic subpolar gyre on the ther-
mohaline circulation. Science 309:1841–1844
128
Nye et al.: Changing spatial distribution of Northeast USA fish stocks
Heincke F (1913) Untersuchungen über die Scholle. General
Bericht I. Schollenfischerei und Schonmaßregeln. Vor-
läufige Übersicht über die wichtigsten Ergebnisse des
Berichts. Rapp P-V Reun Cons Int Explor Mer 16:1– 70
Holbrook SJ, Schmitt RJ, Stephens JS Jr (1997) Changes in an
assemblage of temperate reef fishes associated with a cli-
mate shift. Ecol Appl 7:1299– 1310
Houde ED (1987) Fish early life dynamics and recruitment
variability. Am Fish Soc Symp 2:17–29
Houde ED (1989) Subtleties and episodes in the early life
stages of fishes. J Fish Biol 35:29– 38
Hsieh CH, Reiss CS, Hewitt RP, Sugihara G (2008) Spatial
analysis shows that fishing enhances the climatic sensitiv-
ity of marine fishes. Can J Fish Aquat Sci 65:947– 961
Hughes L (2000) Biological consequences of global warm-
ing: is the signal already apparent? Trends Ecol Evol 15:
56– 61
Hurrell JW, Kushnir Y, Ottersen G, Visbeck M (2003) An
overview of the North Atlantic Oscillation. In: The North
Atlantic Oscillation: climatic significance and environ-
mental impact. Geophys Monogr 134:1– 35
IPCC (Intergovernmental Panel on Climate Change) (2007)
Climate change 2007: synthesis report. Contribution of
Working Groups I, II and III. In: Team CW, Pachauri RK,
Reisinger A (eds) Fourth assessment report of the Inter-
governmental Panel on Climate Change. IPCC, Geneva
Knutson TR, Delworth TL, Dixon KW, Held IM and others
(2006) Assessment of the twentieth-century regional sur-
face temperature trends using the GFDL CM2 coupled
models. J Clim 19:1624– 1651
Labropoulou M, Damalas D (2008) Bathymetric trends in dis-
tribution and size of demersal fish species in the North
Aegean Sea. J Nat Hist 42:673– 686
Levitus S, Antonov JI, Boyer TP, Stephens C (2000) Warming
of the World Ocean. Science 287:2225–2229
Lindley JA, Reid PC, Brander KM (2003) Inverse relationship
between cod recruitment in North Sea and young fish in the
continuous plankton recorder surveys. Sci Mar 67:191–200
Lozier SM, Leadbetter S, Williams RG, Roussenov V, Reed
MSC, Moore NJ (2008) The spatial pattern and mecha-
nisms of heat-content change in the North Atlantic. Sci-
ence 319:800– 803
MacCall AD (1990) Dynamic geography of marine fish popu-
lations. Washington University Press, Seattle, WA
MacPherson E, Duarte CM (1991) Bathymetric trends in dem-
ersal fish size Is there a general relationship? Mar Ecol
Prog Ser 71:103–112
Mahon R, Brown SK, Zwanenburg KCT, Atkinson DB and
others (1998) Assemblages and biogeography of demersal
fishes of the east coast of North America. Can J Fish Aquat
Sci 55:1704– 1738
McCarty JP (2001) Ecological consequences of recent climate
change. Conserv Biol 15:320–331
Megrey BA, Hare JA, Stockhausen WT, Dommasnes A and
others (2009) A cross-ecosystem comparison of spatial and
temporal patterns of covariation in the recruitment of func-
tionally analogous fish stocks. Prog Oceanogr 81:63–92
Methratta ET, Link JS (2007) Ontogenetic variation in habitat
associations for four flatfish species in the Gulf of
Maine– Georges Bank region. J Fish Biol 70:1669–1688
Mountain DG (2003) Variability in the properties of the shelf
water in the Middle Atlantic Bight, 1977–1999. J Geophys
Res 108:3014, doi:10.1029/2001JC001044
Mueter FJ, Litzow MA (2008) Sea ice retreat alters the bio-
geography of the Bering Sea continental shelf. Ecol Appl
18:309– 320
Murawski SA (1993) Climate change and marine fish distrib-
utions: forecasting from historical analogy. Trans Am Fish
Soc 122:647– 658
NEFSC (Northeast Fisheries Science Center) (2001) The 32nd
northeast regional stock assessment workshop (32nd
SAW). Ref. Doc. 01-04, NEFSC, Woods Hole, MA
O’Brien CM, Fox CJ, Planque B, Casey J (2000) Climate vari-
ability and North Sea cod. Nature 404:142
Overholtz WJ, Tyler AV (1985) Longterm responses of demer-
sal fish assemblages of George Bank. Fish Bull (Wash DC)
83:507– 520
Parmesan C, Yohe G (2003) A globally coherent fingerprint of
climate change impacts across natural systems. Nature
421:37– 42
Perry AL, Low PJ, Ellis JR, Reynolds JD (2005) Climate
change and distribution shifts in marine fishes. Science
308:1912– 1915
Planque B, Fromentin JM, Cury P, Drinkwater KF, Jenning S,
Perry RI, Kifani S (in press) How does fishing alter marine
populations and ecosystems sensitivity to climate? J Mar
Syst
Reid PC, Borges Md F, Svendsen E (2001) A regime shift in
the North Sea circa 1988 linked to changes in the North
Sea horse mackerel fishery. Fish Res 50:163– 171
Root TL, Price JT, Hall KR, Schneider SH, Rosenzweig C,
Pounds JA (2003) Fingerprints of global warming on wild
animals and plants. Nature 421:57– 60
Rose GA (2005) On distributional responses of North Atlantic
fish to climate change. ICES J Mar Sci 62:1360– 1374
Rosenzweig C, Karoly D, Vicarellu M, Neofotis P and others
(2008) Attributing physical and biological impacts to
anthropogenic climate change. Nature 453:353358
Shuter BJ, Post JR (1990) Climate, population viability, and
the zoogeography of temperate fishes. Trans Am Fish Soc
119:314– 336
Smith TM, Reynolds RW (2004) Improved extended recon-
struction of SST (1954–1997). J Clim 17:2466 2477
Solomon S, Plattner GK, Knutti R, Friedlingstein P (2009) Irre-
versible climate change due to carbon dioxide emissions.
Proc Natl Acad Sci USA 106:1704–1709
Southward AJ, Hawkins SJ, Burrows MT (1995) Seventy
years’ observations of changes in distribution and abun-
dance of zooplankton and intertidal organisms in the
western English Channel in relation to rising sea temper-
ature. J Therm Biol 20:127–155
Spencer PD (2008) Density-independent and density-depen-
dent factors affecting temporal changes in spatial distribu-
tion of eastern Bering Sea flatfish. Fish Oceanogr 17:
396– 410
Sutton RT, Hodson DLR (2005) Atlantic Ocean forcing of
North American and European summer climate. Science
309:115– 118
Sutton RT, Hodson DLR (2007) Climate response to basin-
scale warming and cooling of the North Atlantic Ocean.
J Clim 20:891– 907
Taylor CC, Bigelow HB, Graham HW (1957) Climate trends
and the distribution of marine animals in New England.
Fish Bull 57:293– 345
The Mathworks (2008) Matlab 7.7.0. The Mathworks, Natick,
MA
Walther GR, Post E, Convey P, Menzel A and others (2002)
Ecological responses to recent climate change. Nature
416:389– 395
Wood RJ, Austin HM (2009) Synchronous multidecadal fish
recruitment patterns in Chesapeake Bay, USA. Can J Fish
Aquat Sci 66:496– 508
129
Editorial responsibility: William Peterson,
Newport, Oregon, USA
Submitted: March 26, 2009; Accepted: July 10, 2009
Proofs received from author(s): October 20, 2009
... The development of most rights-based management schemes predates the emergence of strong climate change signals in the ocean. Climate change is now impacting fishing operations around the globe by altering species productivity [4,5] and shifting distributions [6,7]. Climate change is causing rapid directional shifts in the environment, and as the variance is overlaid on these trends, conditions are moving beyond the bounds of historical experience. ...
Article
Rights-based fisheries management – a management system that apportions harvesting rights of fisheries resources to individuals or groups of individuals based on pre-determined allocation criteria – has become more common in the past three decades due to its potential to achieve resource sustainability while improving economic efficiency. Policies that determine access and allocation of fishing opportunities are often complex as they are designed to achieve multiple management objectives. Climate change alters distribution and productivity of fishery resources, reducing the ability of many of the existing history-dependent allocation policies to achieve the original management objectives. This paper aims to develop an empirically grounded typology of allocation policies, synthesizing practical and academic understanding of fisheries allocation. We examined allocation policies of 54 randomly selected fisheries managed by rights-based management instruments. We use this typology to examine possible climate change impacts and responses on harvester-level allocations. Examining the 54 cases, we find that 89 % of these fisheries allocate fishing opportunities at least partly based on historical catch or access, and only 5 % use socio-economic criteria such as ethnicity or an equal share principle. We find 87 % of the systems are managed by catch quotas, and their ability to meet biological, social, and economic objectives depends on stock assessment quality. Yet, climate-informed stock assessments are critically lacking assessment and quota updates can take multiple years, creating a mismatch in sustainable and realized fishing opportunities. The developed typology and discussions provide useful terminology and lessons for managers and practitioners to prepare for climate challenge impacts.
... Reef fish may also be impacted by the changes in sea surface temperatures [15][16][17], as the differences in sea surface temperature will affect the reproduction of marine organisms and the growth of fish [16,18]. Moreover, it has a significant impact on the spatial distributions of species, as many have migrated to the poles and deeper seas due to increasing sea surface temperature, as a way to maintain constant climate conditions [19][20][21][22][23][24]. As a result, the presence and abundance of fish lead to shifts in marine ecosystems, changes in fishing operations, the distribution of catch shares, and the effectiveness of measures to manage the fisheries [19,25]. ...
Article
Full-text available
Climate change is becoming one of the main threats to fishery resources, with the attendant possibilities of decreasing income and food security. Sea surface temperature (SST) is considered a major environmental indicator of climate change, one that impacts the marine ecosystem and habitat. Studying the impacts of SST changes necessitates regular effective monitoring; remote sensing techniques provide researchers with the ability to track changes on various spatial and temporal scales. This study provides an integrated approach, using the advantages of remote sensing data and GIS tools, to assess the SST changes in the spatial potential aggregation zones of Plectropomus pessuliferus marisrubri and Plectropomus areolatus along the Red Sea’s Saudi coast. This study used SST satellite data for 2011 and 2021 to detect changes and develop suitability and risk assessment maps. The SST showed an increase of 0.46 °C from 2011 to 2021, particularly during the summer months. As a result, the suitability of spatial potential aggregation from 2011 to 2021 has dropped in the summer months. The risk assessment analysis revealed a decrease in the suitable potential aggregation zones in the summer months, as it reached about −35.7% in August, while it increased in the winter months, reaching +2.52% in January.
... For example, these stocks have been heavily exploited in the last 60 years [Pons et al., 2017]. In particular, the increase in fishing pressure coupled with the demography of the fish stocks has led to a collapse and recovery of populations with common trends among stocks as discussed by previous authors [Myers and Worm, 2003;Nye et al., 2009;Wang et al., 2020]. The amount of variance explained by the first two PCs for each fisheries databases (see Methods section 2.4) is very large ( Figure 1I). ...
Article
Full-text available
Previous studies have documented a strong relationship between marine ecosystems and large-scale modes of sea surface height (SSH) and sea surface temperature (SST) variability in the North Pacific such as the Pacific Decadal Oscillation and the North Pacific Gyre Oscillation. In the central and western North Pacific along the Kuroshio-Oyashio Extension (KOE), the expression of these modes in SSH and SST is linked to the propagation of long oceanic Rossby waves, which extend the predictability of the climate system to ~3 years. Using a multivariate physical-biological linear inverse model (LIM) we explore the extent to which this physical predictability leads to multi-year prediction of dominant fishery indicators inferred from three datasets (i.e., estimated biomasses, landings, and catches). We find that despite the strong autocorrelation in the fish indicators, the LIM adds dynamical forecast skill beyond persistence up to 5-6 years. By performing a sensitivity analysis of the LIM forecast model, we find that two main factors are essential for extending the dynamical predictability of the fishery indicators beyond persistence. The first is the interaction of the fishery indicators with the SST/SSH of the North and tropical Pacific. The second is the empirical relationship among the fisheries time series. This latter component reflects stock-stock interactions as well as common technological and human socioeconomic factors that may influence multiple fisheries and are captured in the training of the LIM. These results suggest that empirical dynamical models and machine learning algorithms, such as the LIM, provide an alternative and promising approach for forecasting key ecological indicators beyond the skill of persistence.
... Warming in the Northwest Atlantic is occurring faster than most of the global ocean Seidov et al. 2021). Climate-driven patterns in warming and ocean circulation in this region have been associated with declines in calanoid copepods (Record et al. 2019;Sorochan et al. 2019;Meyer-Gutbrod et al. 2021), and shifts in the abundance and distribution of higher trophic levels (Nye et al. 2009;Pinsky et al. 2013). These oceanographic processes have altered the foraging environment and behavior of the critically endangered North Atlantic right whale (Eubalaena glacialis), reducing the population's calving rate and exposing it to greater risks of serious injury and mortality from ship strikes and entanglement in fishing gear (Meyer-Gutbrod et al. 2021). ...
Article
Changes in the physical oceanography of the Northwest Atlantic stemming from both natural and anthropogenic climate change impact the foraging ecology and distribution of endangered North Atlantic right whales. In this study, right whale sightings from 1990 to 2018 were analyzed to examine decadal patterns in monthly habitat use in 12 high‐use areas. Depth‐integrated abundances of late‐stage Calanus finmarchicus and Calanus hyperboreus were also analyzed for decadal variations in the right whale foraging habitats. There were significant differences in the occurrence, seasonal timing, and persistence of foraging habitats across these three decades. In the decades of the 1990s and the 2010s, prey was less abundant than in the 2000s, corresponding to reduced use of the Southeast US calving grounds in the winter, increased use of Cape Cod Bay in winter and spring, and reduced use of Roseway Basin in the fall. In the 2010s, right whale sightings increased in Southern New England and the Gulf of St. Lawrence in the spring and summer, respectively. Summertime declines in the 2010s in late‐stage copepod abundances in the Gulf of Maine and surrounding regions, as well as in the Gulf of St. Lawrence, indicate that recent increased use of the Gulf of St. Lawrence is driven by a decline in prey in traditional foraging habitats rather than by an increase in prey in the new foraging habitat. This analysis of decadal‐scale differences in right whale sightings and prey abundance is critical for redefining right whale distribution patterns for the most recent (post‐2010) decade.
... movement poleward and/or deeper) in response to climate change are well documented across fish species in the region, including groundfish (e.g. Nye et al., 2009Nye et al., , 2011Pinsky et al., 2013). In addition, climateassociated changes in the timing of seasonal migrations relative to the timing of the survey could impact the availability of fish to spring or fall surveys (Langan et al., 2021). ...
Article
Full-text available
Failure to account for the impacts of climate and ecosystem change on stock dynamics can introduce uncertainty to stock assessments that can make meeting the objective of sustainable fisheries management challenging. The increased prevalence and magnitude of uncertainty in New England groundfish stock assessments (i.e. retrospective patterns) in recent years suggest that there may be common drivers impacting these stocks that are currently unaccounted for in the stock assessment. We examined the coherence in retrospective patterns across groundfish stock assessments and evaluated candidate drivers of retrospective patterns, including large-scale climate and ecosystem change, as well as significant management and monitoring changes. We found high coherence in moving window Mohn's rho time series for groundfish within the Gulf of Maine and Georges Bank areas. Fluctuations in Gulf of Maine groundfish Mohn's rho values were most strongly related to lagged bottom temperature and spiny dogfish biomass time series, whereas fluctuations in Georges Bank groundfish Mohn's rho values were strongly related to lagged time series of warm core rings formation from the Gulf Stream. Our identification of coherence in retrospective patterns across groundfish stocks by region supports the idea of common regional drivers with climate and ecosystem changes emerging as the leading contributing factors.
... The world's climate is changing at an unprecedented rate. Over the last century, global average temperature has increased by 0.85°C, resulting in biological responses across terrestrial, freshwater and marine environments (Cheung et al., 2015;Morley et al., 2018;Nye et al., 2009;Pecl et al., 2017). Species may respond to a changing climate in a variety of ways, including acclimatizing, adapting, moving to an area with a more suitable environment or even dying. ...
Article
Many marine species are shifting their distributions in response to changing ocean conditions, posing significant challenges and risks for fisheries management. Species distribution models (SDMs) are used to project future species distributions in the face of a changing climate. Information to fit SDMs generally comes from two main sources: fishery‐independent (scientific surveys) and fishery‐dependent (commercial catch) data. A concern with fishery‐dependent data is that fishing locations are not independent of the underlying species abundance, potentially biasing predictions of species distributions. However, resources for fishery‐independent surveys are increasingly limited; therefore, it is critical we understand the strengths and limitations of SDMs developed from fishery‐dependent data. We used a simulation approach to evaluate the potential for fishery‐dependent data to inform SDMs and abundance estimates and quantify the bias resulting from different fishery‐dependent sampling scenarios in the California Current System (CCS). We then evaluated the ability of the SDMs to project changes in the spatial distribution of species over time and compare the time scale over which model performance degrades between the different sampling scenarios and as a function of climate bias and novelty. Our results show that data generated from fishery‐dependent sampling can still result in SDMs with high predictive skill several decades into the future, given specific forms of preferential sampling which result in low climate bias and novelty. Therefore, fishery‐dependent data may be able to supplement information from surveys that are reduced or eliminated for budgetary reasons to project species distributions into the future.
... Peningkatan suhu perairan akan direspon dengan perubahan distribusi ikan ke daerah yang suhunya lebih cocok (Cheung dan Pauly 2016). Misalnya dengan bermigrasi ke daerah di posisi lintang tinggi atau ke perairan yang lebih dalam (Nye et al. 2009). Hal ini akan menyebabkan berkurangnya spesies tangkapan (Cheung et al. 2009) dan mempengaruhi kelimpahan dan distribusi ikan (Perry et al. 2005;Dulvy et al. 2009). ...
Article
Full-text available
p align="center"> ABSTRACT Fisheries has significant roles for the Indonesian economy. Climate change influences Indonesian fisheries through a range of direct and indirect pathaway. A scientific based approach such as vulnerability is needed to determine the risks of climate change and adaptation strategies. Therefore, this study was conducted to analyze the vulnerability of fisheries to climate change on province scaled in Indonesia. Vulnerability index (VI) is obtained with composite index of exposure (EI), sensitivity (SI) and adaptive capacity (ACI) of ten provinces representing the eastern and western parts of Indonesia by using purposive sampling method. Source of data for indices variables were using recorded datas from relevant institutions. The results showed that fisheries status of North Sulawesi (VI = 0,78), Central Sulawesi (VI = 0,72) and Gorontalo (VI = 0,61) were very vulnerable despite the composition of constituent vulnerability index was different. This difference determined the specific policies to be taken to each province to reduce vulnerability. Short term policies are taken to reduce the vulnerability of the most vulnerable areas on Sulawesi Utara, Sulawesi Tengah, and Gorontalo. Medium term policy is carried out in high sensitivity areas, namely Kepulauan Riau, Sulawesi Utara, and Kalimantan Timur and in low adaptive capacity areas such as Jambi, Gorontalo and Bangka Belitung. Long term policy is conducted for areas with high exposure such as Sulawesi Tengah, Sulawesi Utara and Kalimantan Timur. Keywords : Climate change, fisheries, vulnerability, province ABSTRAK Perikanan tangkap memiliki peranan penting bagi perekonomian Indonesia. Adanya perubahan iklim akan berdampak merugikan secara langsung maupun tidak langsung pada perikanan tangkap Indonesia. Suatu pendekatan ilmiah diperlukan untuk menentukan risiko perubahan iklim dan strategi adaptasi perikanan tangkap, salah satunya adalah analisis kerentanan ( Vulnerability ). Oleh karena itu penelitian ini dilakukan untuk menganalisis kerentanan perikanan tangkap akibat perubahan iklim pada skala provinsi di Indonesia. Indeks kerentanan (VI) didapatkan dengan mengkompositkan indeks keterpaparan (EI), kepekaan (SI) dan kapasitas adaptif (ACI) dari sepuluh provinsi yang mewakili bagian timur dan barat Indonesia dengan metode purposive sampling. Sumber variabel penyusun indeks variabel menggunakaan rekaman data dari instansi terkait. Hasil penelitian menunjukkan bahwa provinsi Sulawesi Utara (VI=0,78), Sulawesi Tengah (VI=0,72) dan Gorontalo (VI=0,61) berstatus sangat rentan walaupun komposisi penyusun indeks kerentanannya tidak sama. Perbedaan ini menentukan bahwa jenis kebijakan yang diambil menjadi spesifik pada tiap provinsi untuk mengurangi kerentanan. Short term policy diambil untuk mengurangi dapak di daerah yang paling rentan yaitu Sulawesi Utara, Sulawesi Tengah, dan Gorontalo. Medium term policy dilakukan pada daerah yang kepekaannya tinggi yaitu Kepulauan Riau, Sulawesi Utara, dan Kalimantan Timur dan kapasitas adaptifnya rendah yaitu Jambi, Gorontalo dan Bangka Belitung. Long term policy dilakukan untuk daerah yang keterpaparannya tinggi yaitu Sulawesi Tengah, Sulawesi Utara dan Kalimantan Timur. Kata kunci : perubahan iklim, perikanan tangkap, kerentanan, provinsi </p
Article
Climate change is altering the distribution and abundance of fish species in ways not anticipated by current management policy. We created spatially explicit, dynamic models of marine habitats that can inform stock assessments for 25 commercial species on the US Northeast Shelf. The habitat models integrated substrate and seabed features along with the dynamic properties of the ocean. Changes in climate-mediated habitat can affect the survey results by altering the availability component of catchability. Changes in availability were examined (1980–2014) by combining species distribution models with hindcast ocean models. Three patterns in availability were evident: (1) the availability for most species varied over time with no trend; (2) for a number of estuary-dependent species, availability varied with no trend and then dropped dramatically in 2009 when the federal trawl survey changed vessels; and (3) for a set of mid-depth, non-estuary dependent species, availability showed a continuous decline over time. There were few changes in dynamic habitat as the bottom water temperature did not exhibit a strong trend over the time-period studied, resulting in little climate-attributed changes in catchability. Changes in survey design can also have dramatic impacts on catchability, highlighting the method’s ability to detect both climate driven and survey driven changes in catchability.
Article
Climate‐driven distributional shifts have been well‐documented for fisheries resources along the East Coast of the United States, yet little attention has been given to adjacent estuarine systems. The Chesapeake Bay is the largest estuary in the continental United States and serves as important habitat for a diversity of fishes and invertebrates, many of which are seasonal residents. Survey data indicate that relative abundance of finfish in Chesapeake Bay has diminished substantially, while coastwide stock status has remained unchanged. In response to warming, seasonal estuarine residents may remain in coastal waters or inhabit a northerly estuary, but the extent to which changing environmental conditions may drive exchange between the coastal ocean and estuarine systems remains unresolved. This study analyzed data collected from 2008 to 2019 by three fisheries‐independent trawl surveys to explore temporal patterns and associated environmental drivers of the estuarine–coastal ocean exchange in the Mid‐Atlantic for eight economically and ecologically important species. Relative habitat utilization of Chesapeake Bay declined for most species, while utilization patterns for Delaware Bay were largely constant or increasing over time. Broad‐scale, multispecies analyses of relative habitat utilization time series revealed that the North Atlantic Oscillation (NAO) was an important driver of Chesapeake Bay exchange, but that average Apr/May coastal ocean bottom temperature was significant for Delaware Bay. Collectively, the results demonstrate that several Mid‐Atlantic species have altered their estuarine habitat use over time, climate drivers associated with estuarine–coastal ocean exchange operate on different time scales, and that the impacts of warming within the Mid‐Atlantic vary spatially.
Article
Isotopic analysis of Micropogonias furnieri otoliths were used to get insight into palaeoceanographic conditions in the Guanabara Bay and Saquarema Lagoon, Rio de Janeiro state (RJ), located on the southeastern coast of Brazil, under upwelling influence of the Cabo Frio system. Archaeological otoliths come from two Holocene shellmounds (or sambaquis ): Galeão and Beirada. For the first time, radiocarbon analysis using high accuracy techniques were performed at Galeão. Age range was determined to be between 5820 and 4980 cal BP, which extends the chronology of human settlement in the Guanabara Bay. Micro-samples of the otoliths were collected sequentially from the core to the edge, to provide continuous δ ¹⁸ O and δ ¹³ C isotopic profiles over the lifetime of the individual fish. Derived-δ ¹⁸ O oto palaeotemperature estimates vary according to seasonality, resulting in a palaeoceanographic variation between 8 to 31°C for Guanabara Bay and 8 and 28°C for the Saquarema Lagoon. Our data indicate that whitemouth croakers were captured during the Middle Holocene from the Guanabara Bay and Saguarema Lagoon and resided in cooler temperatures compared to temperatures of current conditions.
Article
Full-text available
There is a significant (p<0.01) negative correlation between cod recruitment (age 1 year) and the numbers of young fish taken in the Continuous Plankton Recorder (CPR) Survey in the North Sea. This correlation is analysed with reference to hydrographic, climatic and biological variables in order to determine whether it results from opposite responses to physical forcing, trophic relationships or a combination of these. The results of correlation analyses are consistent with cod recruitment being determined mainly during the planktonic phase of development in the preceding year. Multiple regressions indicate that a combination of the North Atlantic Oscillation index in the preceding year and the abundance of CPR young fish is the pair of variables that provide the best fit to the data. The abundance of CPR young fish is best predicted by the Gulf Stream North Wall Index and the abundance of phytoplankton in the spring of the previous year. Opposite responses to climatic forcing and the regional environmental consequences are identified as of primary importance but interactions, possibly both predation and competition, occur between the planktonic stages of cod and both the planktonic stages of fish and the stocks from which they were derived.
Article
Full-text available
We show that the distributions of both exploited and nonexploited North Sea fishes have responded markedly to recent increases in sea temperature, with nearly two-thirds of species shifting in mean latitude or depth or both over 25 years. For species with northerly or southerly range margins in the North Sea, half have shown boundary shifts with warming, and all but one shifted northward. Species with shifting distributions have faster life cycles and smaller body sizes than nonshifting species. Further temperature rises are likely to have profound impacts on commercial fisheries through continued shifts in distribution and alterations in community interactions.
Article
The seasonal and interannual variability in the temperature, salinity, and volume of Shelf Water (SHW) in the Middle Atlantic Bight (MAB) is described for the period 1977-1999. Large interannual variations in the volume, salinity, and, to a lesser extent, temperature of the SHW occurred that were coherent over multiple year time periods. The variations in volume and salinity originated through processes acting outside of the MAB and were advected into the region from the Gulf of Maine. On a decadal average the SHW observed in the 1990s was approximately 1°C warmer, 0.25 PSU fresher, and 1000 km3 more abundant than during the 1977 - 1987 period. The warming during the 1990s was largest in the southern part of the bight during the winter, when the SHW was more than 2°C warmer then during the earlier decade.
Conference Paper
Future CO(2)-induced climate change scenarios from Global Circulation Models (GCMs) indicate increasing air temperatures, with the greatest warming in the Arctic and Subarctic, Changes to the wind fields and precipitation patterns are also suggested. These will lead to changes in the hydrographic properties of the ocean, as well as the vertical stratification and circulation patterns. Of particular note is the expected increase in ocean temperature. Based upon the observed responses of cod to temperature variability, the expected responses of cod stocks throughout the North Atlantic to the future temperature scenarios are reviewed and discussed here. Stocks in the Celtic and Irish Seas are expected to disappear under predicted temperature changes by the year 2100, while those in the southern North Sea and Georges Bank will decline. Cod will likely spread northwards along the coasts of Greenland and Labrador, occupy larger areas of the Barents Sea, and may even extend onto some of the continental shelves of the Arctic Ocean. In addition, spawning sites will be established further north than currently. It is likely that spring migrations will occur earlier, and fall returns will be later. There is the distinct possibility that, where seasonal sea ice disappears altogether, cod will cease their migration. Individual growth rates for many of the cod stocks will increase, leading to in overall increase in the total production of Atlantic cod in the North Atlantic. These responses of cod to future climate changes are highly uncertain, however, as they will also depend on the changes to climate and oceanographic variables besides temperature, such as plankton production, the prey and predator fields, and industrial fishing. (c) 2005 International Council for the Exploration of the Sea. Published by Elsevier Ltd. All rights reserved.
Article
[1] The seasonal and interannual variability in the temperature, salinity, and volume of Shelf Water (SHW) in the Middle Atlantic Bight (MAB) is described for the period 1977–1999. Large interannual variations in the volume, salinity, and, to a lesser extent, temperature of the SHW occurred that were coherent over multiple year time periods. The variations in volume and salinity originated through processes acting outside of the MAB and were advected into the region from the Gulf of Maine. On a decadal average the SHW observed in the 1990s was approximately 1°C warmer, 0.25 PSU fresher, and 1000 km3 more abundant than during the 1977–1987 period. The warming during the 1990s was largest in the southern part of the bight during the winter, when the SHW was more than 2°C warmer then during the earlier decade.
Article
Substantial changes have occurred in assemblages of nearshore reef fishes in the Southern California Bight during the past two decades. At two sites off Los Angeles, California, species richness of reef fishes fell 15-25%, and composition shifted from dominance by northern to southern species. Additionally, by 1993, 95% of the fish species had declined in abundance by an average of 69%. Concurrent declines of similar magnitude were observed for several trophic levels of the benthic ecosystem farther north at Santa Cruz Island where populations of surfperches (Pisces: Embiotocidae), the standing stock of their crustacean prey, and the biomass of understory macroalgae all declined by ∼80%. Abundances of fishes fell because declining recruitment of age-0 fish was insufficient to compensate for losses of older age classes. Annual levels of recruitment of age-0 fishes at all reefs examined fell more than one order of magnitude over two decades and was correlated among years with a broad indicator of Bight-wide productivity, the biomass of macrozooplankton in the California Current. Lower productivity of the coastal marine ecosystem, associated with a climate regime shift in 1976-1977, likely caused large, but unforeseen, impacts on population abundances and trophic structure in nearshore benthic communities.