o
Thi
M a r i N e P o P u l at i o N C o N N e C t i V i t y
r c
s a
o
ll
r
e
t
c
i
c
t
l
i
e h
v
e r
a
e
s b
di
s
e
t
e
ir
n p
b
u
ub
t
i
o
li
n o
s
h
Population Connectivity
e
f a
d in
n
y p
O
o
c
r
ea
t
i
n
o
o
n o
g
r
a
f t
p
h
h
i
y
s a
, V
r
o
and larval dispersal
t
i
lum
c
l
e b
e 20, N
y p
h
o
t
umb
o
c
o
p
e
y m
r 3, a q
using Geochemical Signatures in Calcified Structures
a
c
h
u
in
a
r
e
t
, r
e
r
e
l
p
y j
o
o
s
t
ur
in
n
g
a
, o
l o
r o
f The
t
h
e
r m
o
c
e
e
a
a
n
n
o
s i
g
s p
r
B y S i M o N r . t h o r r o l d ,
a
p
e
h
r
m
y S
i
t
o
d a N i e l l e C . Z a C h e r l ,
t
c
e
i
d o
e
t
y
nl
. C
a N d l i S a a . l e V i N
y w
o
p
i
y
t
r
h t
i
g
h
h
t 2007 b
e a
p
p
r
o
v
y The
a
The importance of larval dispersal to the population
l o
f The
o
dynamics and biogeography of marine organisms has been
c
e
a
o
n
c
o
e
recognized for almost a century (Hjort, 1914; Thorson, 1950).
g
a
r
n
a
o
p
g
h
r
y S
More recently, theoretical studies have highlighted the role that con-
a
p
h
o
y S
c
i
e
nectivity may play in determining the resilience of marine populations
o
t
y
c
.
i
a
e
t
ll r
(Hastings and Botsford, 2006). Effective spatial management of marine capture fish-
y
. S
i
g
e
h
n
t
d a
s r
eries, including the design of marine reserve networks, also requires an understanding
e
ll c
s
e
o
r
of population connectivity (Sale et al., 2005). However, remarkably few empirical esti-
v
r
e
r
e
d. P
s
p
mates of larval dispersal or population connectivity in ocean environments exist.
o
e
n
r
m
d
e
i
s
n
si
Direct and definitive estimates of larval dispersal in the ocean require the abil-
c
o
e t
n i
o: in
s g
ity to track microscopic larvae of benthic invertebrates and fishes through the pelagic
r
f
a
o@t
n
t
e
environment, from spawning locations to recruitment sites. Most marine species with
d t
o
s
.o
o c
r
pelagic larvae spawn millions of propagules that are released and then subjected to
g o
o
p
r Th e
y t
h
significant advection, diffusion, and mortality in vast volumes of seawater, making
i
s a
o
r
c
t
i
traditional mark-recapture approaches extremely difficult (Levin, 2006). However,
e
c
a
l
e f
n
o
o
g
r u
ecologists have embraced recent developments in probe-based mass spectrometry to
r
a
p
s
h
e in t
y S
examine the chemistry of calcified structures in marine invertebrates and fishes that
o
e
c
a
i
e
c
t
h
can be used as artificial or natural tags of natal origins. These geochemical tags are
y
in
, P
g a
o
B
n
revealing fascinating data on larval dispersal that are challenging widely held para-
d r
o
x 1931,
e
s
e
digms concerning the spatial scale of demographic connectivity in ocean ecosystems.
a
r
c
h
r
.
o
r
c
e
k
p
v
ub
ill
e
li
, M
c
a
t
d
i
o
20849-1931,
n
, s
y
s
t
e
mm
a
u
t
i
c r
S
a
e
.
p
r
o
du
80
Oceanogr
O
ap
ceanogr hy
ap
c
Vo
V l.
l. 20, No. 3
o
t
i
o
n
,

---

GeoCheMiCal aPProaCheS
function of coastal geology, sources of
ited number of major rivers within the
Natural tags
pollution, atmospheric deposition, and
study location and to low rainfall and
Natural geochemical tags are generated
variable inputs from local watersheds.
its associated runoff. However, locations
by variations in environmental condi-
In southern California, for example,
of convergence between distinct ocean
tions, including temperature, salinity,
because of heavily urbanized develop-
currents (e.g., at Point Conception in
and seawater chemistry, that are sub-
ment, rainwater runoff should contrib-
central California and at Cape Hatteras
sequently recorded by the elemental or
ute high concentrations of trace ele-
in North Carolina, USA) are character-
isotopic composition of calcified struc-
ments associated with anthropogenic
ized by sharp gradients in temperature
tures (see Box 1). Therefore, shells, fish
activity into the estuaries and near-
and, to a lesser extent, salinity, making
ear-stones (called otoliths), and statoliths
shore waters. The magnitude of input
generation of distinctive geochemi-
(the invertebrate analog to otoliths) may
of these trace elements is a function of
cal tags likely (Zacherl, 2005). Seawater
act as dated "flight recorders" because
land use, the amount of impermeable
elemental composition may also vary
as organisms disperse across gradients
surface area within a watershed, num-
among locations at smaller spatial scales
in seawater composition or tempera-
ber and size of rivers, and frequency
(e.g., Becker et al., 2005) due to dif-
ture, their travels are being constantly
of rain events (Ackerman and Schiff,
ferences in coastal geology, mesoscale
recorded by the chemistry of the calci-
2003). Broadening consideration to the
oceanography, and variable inputs from
fied structure. Likewise, larvae develop-
entire west coast of the United States,
local watersheds.
ing in areas that have differing seawater
watersheds in Washington and Oregon
The tremendous advantage to using
characteristics build calcified structures
regularly receive > 100 inches of rain
natural geochemical tags is that every
whose elemental compositions reflect
annually, while watersheds in southern
larva is potentially tagged, eliminating
their sources (Zacherl et al., 2003a;
California rarely see > 50 inches annually
any concerns arising from dispersion of
Becker et al., 2007). Typically, geochemi-
(National Weather Service, 1961­1990
tagged larvae and subsequent low recap-
cal tags consist of either a combination
average annual rainfall). Because of this
ture rates. However, there can be a signif-
of minor and trace elements expressed as
extreme variation in land use, geol-
icant uncertainty in interpreting the ele-
a ratio to Ca (e.g., Mg/Ca, Mn/Ca, Sr/Ca,
ogy, and runoff contributions typically
mental variation in tags among locations.
and Ba/Ca), or stable isotope ratios such
associated with estuaries, these habitats
Unless obvious and strong broad-scale
as 18O (Killingley and Rex, 1985) and
would appear to provide ideal locations
regional gradients in tag composition
87Sr (Barnett-Johnson et al., 2005).
for generation of a natural geochemical
exist, the approach necessitates substan-
The ability of natural geochemical
tag (see "Progress to Date" below).
tial sampling effort to ensure that the
tags to track larval movement depends
Less attention has focused on using
geochemical tags of all potential source
upon the existence of substantial varia-
geochemical tags in open-coast species.
populations have been characterized.
tion in the elemental composition of
The limited emphasis on coastal ocean
those tags among locations of interest
habitats is probably because gradients in
artificial tags
(Thorrold et al., 2002). Many studies
oceanographic conditions can be more
Calcified tissues are also ideal reposito-
using geochemical tags have focused on
subtle, thus complicating the genera-
ries for artificial chemical markers that
species inhabiting estuaries because of
tion of unique geochemical tags. For
are used to tag embryos or lab-reared
significant variation in the elemental
example, Gillanders et al. (2001) found
larvae before dispersal from natal loca-
composition of their calcified structures
little variation in otolith chemistry of
tions (Jones et al.,1999; Moran and
(e.g., Swearer et al., 2003), probably
two-banded bream among locations and
Marko, 2005; Thorrold et al., 2006).
due to substantial differences in salin-
sites that ranged from ~ 10 to > 100 km
Fluorescent compounds such as tetra-
ity, temperature, and water chemistry
from one another along the coast of
cycline or calcein, elemental markers
among estuarine waters. Variation in
Spain. They attributed the lack of varia-
(e.g., rare earth elements), and radioac-
water chemistry among estuaries is a
tion in otolith chemistry to the lim-
tive isotopes have all been used to tag
Oceanography September 2007
81

---

calcified structures by immersing devel-
to local populations (Jones et al., 1999,
Spatial Variability in trace
oping larvae (or their food) in a solution
2005, and see "Emerging Technologies"
element Signatures
containing the target marker (reviewed
below), but has yet to be applied to
A key initial objective in using natural
by Thorrold et al., 2002). The marked
examinations of connectivity among
geochemical tags, and a prerequisite
larvae can then be released from a source
different subpopulations.
to its successful application to popula-
population and recaptured at a destina-
tion connectivity, has been to identify
tion of interest, where they are screened
ProGreSS to date
those habitats, oceanographic condi-
for presence of the artificial tag. To over-
Although initially applied to fish otoliths
tions, or specific locations that impart
come high larval mortality rates and dif-
(Campana, 1999), geochemical tagging
distinct multi-elemental signatures to
fusive processes that can dilute the con-
methods have now been expanded to
larvae. As mentioned above, estuaries
centrations of tagged larvae substantially,
include statoliths, protoconchs (larval
provide the ideal variation in chemical
significant portions of the total larval
shells) of molluscs (Zacherl et al., 2003a;
and physical characteristics to gener-
population must be tagged. To estimate
Arkhipkin et al., 2004), and whole bodies
ate significant variation in geochemi-
connectivity, the proportion tagged out
of decapod larvae (DiBacco and Levin,
cal tags. For instance, a clear distinction
of the total number of larvae spawned by
2000). However, applications for many
between embayment and nearshore
the focal population during the experi-
taxa remain difficult and labor intensive,
coastal otolith elemental signatures has
ment must be estimable. Artificial tag-
and only a handful of studies have suc-
been documented for larval California
ging has, therefore, been used success-
cessfully examined larval connectivity
halibut (Forrester and Swearer, 2002)
fully to estimate rates of larval retention
using natural geochemical tags.
and English sole (Brown, 2006) on the
BoX 1. CSi Seawater--do CalCiFied StruCtureS iNteGr ate Seawater ProPertieS?
use of geochemical tags is premised on the notion that the chemistry of
however, there is still some mystery involved with how simple and
calcified structures in some way reflects physiochemical properties of the
predictable findings from larval culturing studies play out in a field set-
surrounding seawater in which the calcified structure was formed. results
ting, where calcified structures form under the influence of several inter-
from controlled laboratory experiments on gastropod larvae by Zacherl
acting factors. Therefore, recent culturing experiments have focused on
et al. (2003b) and on fish larvae by Bath et al. (2000) clearly supported this
examining the interactive effects of two or more factors (e.g., Milton
assumption and yielded predictable results (Figure a-1). in their experi-
and Chenery 2001; Martin and Thorrold, 2005) and on ranking their
ments, both groups manipulated concentrations of Ba/Ca in rearing sea-
relative importance (elsdon and Gillanders; 2004; lloyd et al., in press).
water, raised larvae for several weeks under experimental conditions, and
Further, there is very recent evidence that not only seawater physical
then analyzed whole larval calcified structures using inductively coupled
and chemical properties but also maternal transmission of elements
plasma mass spectrometry (iCP-MS). linear relationships between con-
can influence the elemental composition of larval calcified struc-
centrations of elements in seawater and in biogenic aragonite are com-
tures. For example, there is evidence that trace elements
mon and have been documented for many elements including Ba, Sr, and
contained in the egg can be incorporated into larval-fish
Mg (e.g., lorens and Bender, 1980; elsdon and Gil anders, 2003).
otoliths (Kalish, 1990; Thorrold et al., 2006) and into
temperature and salinity also predictably influence element uptake
larval-gastropod statoliths (lloyd et al., in press).
into otoliths and statoliths (e.g., Martin et al., 2004; Zumholz et al., 2007).
This knowledge complicates our ability to draw
The combined results of al culturing studies suggest that calcified oto-
simple conclusions about the physical environment
liths, statoliths, and protoconchs, like foraminifera tests (lea et al., 1999),
based upon the chemistry of calcified structures
can indeed reliably integrate information about seawater physical and
but provides an exciting venue for future Calcified
chemical properties.
Structure investigations.
82
Oceanogr
O
ap
ceanogr hy
ap Vo
V l.
l. 20, No. 3
o

---

Pacific coast. Fodrie (2006) found simi-
temperature, or watershed influence may
approaches to measuring connectivity,
lar patterns in juvenile California halibut
provide the needed chemical variation to
including coupled biophysical model-
otoliths, but went a step further in docu-
allow for the application of natural geo-
ing and population genetics. Despite
menting distinct signatures for embay-
chemical signatures (e.g., Zacherl, 2005).
the potential, there are remarkably
ments with different geomorpholo-
For instance, natal sites of larval rockfish
few studies that have used geochemi-
gies and for open-coast habitat. These
(Sebastes atrovirens) were distinguished
cal signatures successfully to deter-
findings were employed to address the
between mainland and island sites by
mine larval origins, although identi-
movements of juveniles within bays, and
Zn/Ca, Sr/Ca, Ba/Ca, and Pb/Ca signa-
allowed increasing refinement of nursery
tures in prehatch otoliths; mainland sites
SiMoN r. thorrold (sthorrold@
habitat determination for successful sub-
only 10 km apart also showed signifi-
whoi.edu) is Associate Scientist, Biology
adults as well as evaluation of the demo-
cant differences in elemental signatures
Department, Woods Hole Oceanographic
graphic consequences of using different
(Warner et al., 2005).
Institution, Woods Hole, MA, USA.
nursery habitats.
daNielle C. ZaCherl is Assistant
Detecting geochemical differences
identifying Natal origins
Professor, Department of Biological Science,
imparted to pelagic larvae that disperse
The ability to address the question,
California State University, Fullerton,
along the open coasts of continents
"where did newly recruited individu-
Fullerton, CA, USA. liSa a. leViN is
has proved a greater challenge. Recent
als come from?" is a real strength of
Professor, Integrative Oceanography
progress suggests that subtle differences
geochemical tags, and distinguishes
Division, Scripps Institution of
in water masses and eddies, upwelling,
the method from other indirect
Oceanography, La Jolla, CA, USA.
0
50
100
150
200
250
25
Figure a-1. The influence of ambient seawater on
the elemental composition of biogenic aragonite
is most easily quantified in the laboratory where
20
water chemistry and temperature can be accurately
)-1
constrained. For instance, Ba/Ca ratios (mean ±
o
l
standard error) in otoliths of laboratory-reared spot,
o
l
.
m
Leiostomus xanthurus, (cyan symbols) and of stato-
15
liths (green symbols) and protoconchs (red symbols)
(
µ
m
e
of laboratory-reared Kellet's whelk (Kelletia kelleti )
varied linearly with the Ba/Ca ratios of the rearing
10
water. Note that the primary x-axis depicts rearing-
a

s
t
r
u
c
t
ur
water Ba/Ca concentrations for statoliths and pro-
toconchs at 17.1°C and 34 salinity (from Zacherl
B
a
/
C
et al., 2003b), while the secondary x-axis depicts
5
rearing-water Ba/Ca concentrations for otoliths at
20.3°C and 20 salinity (from Bath et al., 2000).
0 0
5
10
15
20
Ba/Ca seawater (µmol.mol-1)
Oceanogr
O
ap
ceanogr hy
ap Sep
e te
t mb
e
er 2007
e
83

---

fication of juvenile nursery areas has
(Caley et al., 1996). The most compre-
ral tags in otoliths, that up to 50% of
been achieved in several instances (e.g.,
hensive assessments to date have been
bluehead wrasse (Thalassoma bifascia-
Thorrold et al., 2001).
conducted on coral reef fishes. Using
tus) recruits on St. Croix were spawned
One key question that can be
an artificial fluorescent tag incorpo-
locally. It is noteworthy that the levels
addressed with geochemical mark-
rated into embryonic otoliths, Jones
of self-recruitment documented by each
ers is the extent to which populations
and co-workers found that from 15 to
of the three studies were quite similar
are capable of replenishing themselves
60% of yellow damselfish (Pomacentrus
despite the fact that yellow damselfish
through recruitment of locally pro-
amboinensis) larvae recruiting to Lizard
and panda clownfish spawn benthic
duced larvae. For the few studies that
Island, Australia, originated from local
eggs and their larvae have relatively
have succeeded in identifying origins of
reefs (Jones et al., 1999), and up to
short pelagic durations, while blue-
settled, subadult, or adult individuals,
42% of panda clownfish (Amphiprion
head wrasse are pelagic spawners whose
a pattern has emerged suggesting that
polymnus) recruited to natal sites
offspring typically spend 40 to 60 days
local retention of larvae, or return of
around Schumann Island, Papua New
as pelagic larvae.
spawning adults to natal sites, may be
Guinea (Jones et al., 2005). Swearer
Determination of larval origins using
more prevalent than previously expected
et al. (1999) reported, based on natu-
geochemical signatures in marine inver-
BoX 2. MuSSelS iN MotioN
The use of natural geochemical tags in
two weeks after the outplanting, newly settled mussels (one to two
studies of larval connectivity requires
weeks old) were collected from intertidal rocks near the outplant moor-
knowledge of the signatures imparted to
ings. These recruits were 1­2 mm and retained their larval shell--the
larval structures by specific water masses
protoconch. recruit mussel tissue was analyzed by a PCr-based assay to
at different locations. This has been a seri-
identify species (either Mytilus californianus or Mytilus galloprovincialis).
ous challenge for free-spawning species that do
The chemical composition of recruit larval shells was determined by laser
not brood or hold their offspring on the seabed. a
ablation iCP-MS and individuals of each species at each sampling site
solution to this problem is presented in Becker et al. (2007). The authors
were assigned an origin. This information was used to assess larval con-
outplanted laboratory-spawned larvae of mytilid mussels to generate a
nectivity patterns for the California mussel and the bay mussel in San
reference map of chemical signatures at different locations in San diego
diego County. differences were found in larval connectivity patterns
County, including open-coast and bay settings along 75 km of shoreline
of the two species (Figure B-1); this was unexpected as their larvae are
(Figure B-1). larvae of two Mytilus mussel species were introduced into
thought to mix along the San diego coastline. M. californianus exhibited
PVC "homes" within 12 hours of spawning (before shell formation) and
asymmetric mixing with the majority of larvae originating in the northern
were transported to nearshore moorings where they were allowed to
part of the study area; there was high self-recruitment in the north and
develop for one week. during this period the larvae laid down arago-
high importation of larvae in the south. M. galloprovincialis recruits had
nite shells that retained a chemical signature of the waters in which they
more diverse origins from a mixture of north, south, and bay locations,
developed. The elemental composition of the larval shells was then ana-
but with substantial (40%) self-recruitment. This study provides informa-
lyzed using laser ablation inductively coupled plasma mass spectrometry
tion about larval connectivity for two species at one time in one region.
(iCP-MS). after analyzing shell signatures, Becker and co-workers deter-
it offers a first look at the origins of settled invertebrates using elemental
mined that the method resolution for identifying distinct water bodies
signatures in shells, and supports a growing paradigm of limited dispersal,
was greatest at the regional (20­30 km) scale and for individual bays.
even in species with long-lived larvae. however, many questions
remain about the underlying causes of species differences in
connectivity, the stability of these patterns over time, and
their relevance for other areas within the species' ranges.
84
Oceanogr
O
ap
ceanogr hy
ap Vo
V l.
l. 20, No. 3
o

---

tebrates began with elemental signatures
(M. californianus) were found to exhibit
Nonetheless, exciting new results using
of crab zoea from inside and outside San
differing connectivity patterns and
natural and artificial geochemical tags
Diego Bay (DiBacco and Levin, 2000).
rates of self-recruitment among study
clearly demonstrate the potential of the
While analyses confirmed behaviorally
sites (see Box 2).
approach (Almany et al., 2007; Becker
enhanced export of larvae (DiBacco et
et al., 2007). We also envisage applica-
al., 2001) and extensive mixing of lar-
eMerGiNG teChNoloGieS
tion of the techniques in novel environ-
vae from different sources (DiBacco
aNd Future direCtioNS
ments, including hydrothermal vents
and Chadwick, 2001), the researchers
The use of geochemical signatures in
and other ephemeral habitats in the deep
did not address the question of natal
calcified structures of marine organ-
sea, where connectivity is likely to be
origins of new recruits. Recently, natal
isms to estimate population connectiv-
extremely important to population per-
origins of new recruits were determined
ity in marine ecosystems is still in its
sistence and maintenance of biodiversity
in mytilid mussel populations along the
infancy. A number of significant chal-
(e.g., Neubert et al., 2006).
coast of southern California (Becker
lenges remain before routine estimates of
Despite some progress, the applica-
et al., 2007). Bay mussels (Mytilus gal-
population connectivity in coastal waters
tion of geochemical markers to the study
loprovincialis) and California mussels
will be possible using these approaches.
of larval dispersal in marine environ-
a
B
Figure B-1. (a) in the initial phase of a study to examine
population connectivity of mussels (Mytilus spp.) along the
coast of southern California, larval mussels were outplanted
into PVC homes at a number of locations and allowed to
develop for one week in ambient seawater. Natal origins of
newly settled recruits collected two weeks after the out-
planting were then determined based on reference maps
of chemical signatures developed from outplanted larvae
(from Becker et al., 2007). (B) The resulting schematic dia-
gram of population connectivity along the coast of San
diego County revealed unidirectional movement of larvae
from northern to southern locations for M. californianus
(top panel), while M. galloprovincialis (bottom panel)
populations showed more evidence of local self-recruit-
ment (from Becker, 2005). NC = Northern coastal region
(blue sites include al = agua hedionda lagoon, ah =
agua hedionda, Cr = Cardiff reef, lJdr = la Jolla dike
rock). SC = Southern coastal region (red sites include Sio =
Scripps institution of oceanography Pier, PB = Pacific Beach
(Crystal) Pier, oB = ocean Beach Pier, CaBr = Cabrillo
National Monument, iB = imperial Beach Pier, Shi = Shelter
island). MB = Mission Bay (pink site, CPMS = Crown Point
Mitigation Site). SB = San diego Bay (green sites include
hi = harbor island and CV = Chula Vista). labels point to
larval outplant stations, and Xs represent corresponding
intertidal recruit collection stations.
Oceanogr
O
ap
ceanogr hy
ap Sep
e te
t mb
e
er 2007
e
85

---

ments remains hampered by techno-
ions that are then transported to a mass
isotope tracers that are less subject to
logical limitations. Elemental analysis
spectrometer for quantification (Russo
physiological effects and therefore more
of individual microscopic calcified
et al., 2002). In the future we may see
accurately reflect ambient values in the
structures must involve laser ablation of
femtosecond lasers coupled to plasma
environment. The number of these iso-
very small amounts of calcified material
source TOF-MS instruments producing
tope systems available to researchers has
(i.e., less than 5 µg). Such small quanti-
precise and accurate elemental data at
increased rapidly with the development
ties of analyte provide a transient signal
spatial resolutions less than 10 µm. This
of multiple collector arrays on ICP-MS
that is often insufficient for sequential
resolution could allow detailed examina-
instruments (Halliday et al., 1998).
analysis of isotopes with single collec-
tion of larval trajectories reflected in the
There may also be useful synergies with
tor inductively coupled plasma mass
chemical composition of distinct regions
researchers attempting to develop new
spectrometry (ICP-MS) instruments
of minute larval structures.
temperature proxies in biogenic carbon-
(Strasser et al., in press). However, time-
As mentioned above, studies attempt-
ates based on Mg, Ca, and Sr isotopes
of-flight mass spectrometry (TOF-MS)
ing to use natural geochemical mark-
that may not be subject to significant
allows for simultaneous detection of ions
ers in open marine systems commonly
biological fractionation (e.g., Nägler et
over a large mass range and is therefore
encounter subtle gradients in physical
al., 2000; Fietzke and Eisenhauer, 2006).
particularly well suited to analyses of
and chemical properties of coastal water
Interest continues in artificial tagging
transient signals generated by laser abla-
masses, at least compared to the stronger
approaches for marine larvae because
tion (Vázquez et al., 2002). New devel-
elemental signals imparted within river
few methods can provide unequivocal
opments in laser ablation also offer hope
and estuarine systems. Efforts to over-
estimates of population connectivity in
for more accurate chemical analyses of
come this problem include refinement
ocean ecosystems. However, the logistic
larval shells and otoliths. For instance,
of analytical techniques to increase
difficulties associated with tagging a large
UV femtosecond laser systems, likely to
the precision of geochemical variables.
proportion of the total larval produc-
tion from an area has, until very recently,
proved difficult to overcome. The
development of a TRAnsgenerational
with careful targeting of specific questions
Isotope Labeling (TRAIL) technique,
based on maternal transmission of an

to be addressed, and of focal species to be
enriched stable Ba isotope that is incor-
examined, studies using natural and artificial
porated in the embryonic otoliths of
tags in calcified structures are likely to lead to
larval fish, may help to overcome this
limitation (Thorrold et al., 2006). The

significant advances in our understanding of TRAIL approach represents a signifi-
population connectivity in ocean ecosystems.
cant advance from earlier artificial tag-
ging methods because it is possible to
tag a much higher proportion of the
be available commercially in the near
However, we may already be at a stage
total larval production from an area, the
future, will hopefully improve both the
where physiological or ontogenetic con-
technique can be used on benthic and
spatial resolution and accuracy of laser
trols on the elemental composition of
pelagic spawning fishes, and multiple
ablation ICP-MS analyses (Koch and
biogenic carbonates may introduce more
tags can be applied (Figure 1). The first
Günther, 2007). Femtosecond lasers
error into estimates of larval sources
field test of the method demonstrated
eliminate mass fractionation, at least in
from geochemical signatures than the
substantial (> 50%) self-recruitment of
theory, during the ablation process and
precision of the analytical technique. In
benthic and pelagic spawning fishes to
provide for stoichiometric conversion of
the future, researchers are likely to use
a small coral reef reserve in Papua New
86
Oceanography Vol. 20, No. 3

---

30
30
Figure 1. a new approach for marking fish larvae
relies upon transgenerational transfer of enriched
Ba isotopes from females to the embryonic otoliths
of their offspring. to determine the potential of this
B
approach, black sea bass (Centropristis striata: dia-
monds), red and black anemonefish (Amphiprion
melanopus: squares), and brown-marbled grouper
B
20
(
20
Epinephelus fuscoguttatus: circles) females were
B
B
injected with either an enriched 137Ba (cyan symbols)
B
0.1 mm
or 135Ba (red symbols) solution (black symbols are
Ba
B
a
B
B
B
mean values [± 3] from controls for each species).
135
B
B
135
B
B
larvae spawned after injections were reared for two
BB
B
a/
B
to four weeks, and laser ablation inductively coupled
Ba/ 138
B
B
B B
B
B
B
B
B B
B
B
B
plasma mass spectrometry was then used to analyze
B
138
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B B
B B
B
Ba isotope ratios from a 50 µm x 50 µm raster cen-
B
10
10
tered on otolith cores from all three species (otolith
from A. melanopus shown in insert). Barium isotope
ratios in the cores of larval otoliths from injected
females plotted close to the theoretical mixing curves
(dashed lines) between the enriched isotope spikes
and natural Ba ratios, providing conclusive evidence
for maternal transfer of the unique Ba isotope tags.
00
00
22
4
4
66
8
8
138Ba/137Ba
138Ba/137Ba
Guinea (Almany et al., 2007). The suc-
mates of population connectivity, such
ing larval export from the reserve to
cess of the study was based on the abil-
information would have been useful
adjacent fished areas. Recent data sug-
ity to tag a high percentage of all larvae
when choosing protected-area locations
gest that the scale at which coral reef fish
produced in the reserve over a period of
in Australia's Great Barrier Reef Marine
populations can be both self-sustaining
two months. The TRAIL method is likely
Park or the Channel Islands off the coast
and capable of providing recruitment
to be particularly useful for species that
of California. However, there is also no
subsidies is considerably smaller than
form spawning aggregations at specific
denying that determining population
previously imagined (Almany et al.,
locations and times, and therefore pro-
connectivity using the approaches out-
2007). Nonetheless, the generality of
vide the opportunity for a considerable
lined here is both time-consuming and
these results for other systems, and even
percentage of the total spawning popula-
expensive, and is therefore only likely to
for coral reefs, is still in question, par-
tion to be captured.
be contemplated for a handful of species
ticularly for grouper and snapper spe-
at any location. One way to maximize
cies that contribute disproportionately
iMPliCatioNS
the generality of the results would be
to artisanal and commercial fisheries
Ultimately, scientists need to provide
to use the data to test the performance
on coral reefs. With careful targeting of
accurate information on population con-
of coupled biophysical models that are
specific questions to be addressed, and
nectivity, which can be used to optimize
more easily applied to a number of dif-
of focal species to be examined, stud-
spatial management approaches, includ-
ferent species over large spatial scales
ies using natural and artificial tags in
ing marine protected areas (MPAs), for
(e.g., Gilg and Hilbish, 2003; Galindo et
calcified structures are likely to lead
marine-capture fisheries. Although we
al., 2006). Similarly, it has proved almost
to significant advances in our under-
know of no management plans that have
impossible to verify the effectiveness of
standing of population connectivity
specifically incorporated empirical esti-
MPAs without a method for quantify-
in ocean ecosystems.
Oceanography September 2007
87

---

aCKNowledGeMeNtS
Becker, B.J., L.A. Levin, F.J. Fodrie, and P.A. McMillan.
Gillanders, B.M., P. Sanchez-Jerez, J. Bayle-Sempere,
2007. Complex larval retention patterns in
and A. Ramos-Espla. 2001. Trace elements in
Our sincere thanks to the Division of
marine invertebrates. Proceedings of the National
otoliths of the two-banded bream from a coastal
Ocean Sciences at the National Science
Academy of Sciences of the United States of America
region in the south-west Mediterranean: Are there
Foundation for funding larval connectiv-
104:3,267­3,272.
differences among locations? Journal of Fish Biology
Brown, J.A. 2006. Classification of juvenile flatfishes to
59:350­363.
ity research on K. kelletii (OCE 0351860
estuarine and coastal habitats based on the elemen-
Halliday, A.N., D.-C. Lee, J.N. Christensen, M.
to DZ), bivalve mussels (OCE 0327209
tal composition of otoliths. Estuarine, Coastal and
Rehkamper, W. Yi, X. Luo, C.M. Hall, C.J.
Shelf Science 66:594­611.
Ballentine, T. Pettkke, and C. Stirling. 1998.
and 0648656 to LAL), and coral reef
Caley, M.J., M.H. Carr, M.A. Hixon, T.P. Hughes,
Applications of multi-collector ICPMS to cosmo-
fishes (OCE 0424688 to SRT). We thank
G.P. Jones, and B.A. Menge. 1996. Recruitment
chemistry, geochemistry, and paleoceanography.
and the local dynamics of open marine popula-
Geochimica et Cosmochimica Acta 62:919­940.
B.J. Becker, J. Fodrie, and P. McMillan
tions. Annual Review of Ecology and Systematics
Hastings, A., and L.W. Botsford. 2006. Persistence of
for sharing unpublished data on mytilid
27:447­500.
spatial populations depends on returning home.
mussel and California halibut connec-
Campana, S.E. 1999. Chemistry and composition of
Proceedings of the National Academy of Sciences of
fish otoliths: Pathways, mechanisms and applica-
the United States of America 103:6,067­6,072.
tivity. Some of the ideas discussed here
tions. Marine Ecology Progress Series 188:263­297.
Hjort, J. 1914. Fluctuation in the great fisheries of
grew out of discussion among members
DiBacco, C., and L.A. Levin. 2000. Development and
northern Europe. Rapports et Procčs-Verbaux des
application of elemental fingerprinting to track the
Réunions du Conseil International pour l'Explora-
of the Coral Reef Targeted Research
dispersal of marine invertebrate larvae. Limnology
tion de la Mer 20:1­228.
(CRTR) Program Connectivity Working
and Oceanography 45:871­880.
Jones, G.P., M.J. Milicich, M.J. Emslie, and C. Lunow.
DiBacco, C., and D.B. Chadwick. 2001. Assessing
1999. Self recruitment in a coral-reef fish popula-
Group (http://www.gefcoral.org).
the dispersal and exchange of brachyuran larvae
tion. Nature 402:802­804.
between regions of San Diego Bay, California and
Jones, G.P., S. Planes, and S.R. Thorrold. 2005. Coral
reFereNCeS
nearshore coastal habitats using elemental finger-
reef fish larvae settle close to home. Current Biology
printing. Journal of Marine Research 59:53­78.
15:1,314­1,318.
Almany, G.R., M.L. Berumen, S.R. Thorrold, S. Planes,
Dibacco, C., D. Sutton, and L. McConnico. 2001.
Kalish, J.M. 1990. Use of otolith microchemistry to
and G.P. Jones. 2007. Local replenishment of coral
Vertical migration behavior and horizontal distri-
distinguish the progeny of sympatric anadromous
reef fish populations in a marine reserve. Science
bution of brachyuran larvae in a low-inflow estu-
and non-anadromous salmonids. Fishery Bulletin
316:742­744.
ary: Implications for bay-ocean exchange. Marine
88:657­666.
Ackerman, D., and K. Schiff. 2003. Modeling storm
Ecology Progress Series 217:191­206.
Killingley, J.S., and M.A. Rex. 1985. Mode of larval
water mass emissions to the Southern California
Elsdon, T.S., and Gillanders, B.M. 2003. Relationship
development in some deep-sea gastropods indi-
Bight. Journal of Environmental Engineering
between water and otolith elemental concentra-
cated by oxygen-18 values of their carbonate shells.
129:308­317.
tions in juvenile black bream Acanthopagrus butch-
Deep-Sea Research 32:809­818.
Arkhipkin, A., S.E. Campana, J. FitzGerald, and S.R.
eri. Marine Ecology Progress Series 260:263­272.
Koch, J., and D. Günther. 2007. Femtosecond laser
Thorrold. 2004. Spatial and temporal varia-
Elsdon, T.S., and Gillanders, B.M. 2004. Fish otolith
ablation inductively coupled plasma mass spec-
tion in elemental signatures of statoliths from
chemistry influenced by exposure to multiple
trometry: Achievements and remaining prob-
the Patagonian longfin squid (Loligo gahi).
environmental variables. Journal of Experimental
lems. Analytical and Bioanalytical Chemistry
Canadian Journal of Fisheries and Aquatic Sciences
Marine Biology and Ecology 313:269­284.
387:149­153.
61:1,212­1,224.
Fietzke, J., and A. Eisenhauer. 2007. Determination of
Lea, D.W., T.A. Mashiotta, and H.J. Spero. 1999.
Barnett-Johnson, R., F.C. Ramos, C.B. Grimes, and
temperature-dependent stable strontium isotope
Controls on magnesium and strontium uptake
R.B. MacFarlane. 2005. Validation of Sr isotopes
(88Sr/86Sr) fractionation via bracketing standard
in planktonic foraminifera determined by live
in otoliths by laser ablation multicollector induc-
MC-ICP-MS. Geochemistry Geophysics Geosystems,
culturing. Geochimica et Cosmochimica Acta
tively coupled plasma mass spectrometry (LA-
doi:10.1029/2006GC001243.
63:2,369­2,379.
MC-ICPMS): Opening avenues in fisheries science
Fodrie, F.J. 2006. Quantifying nursery habitat value
Levin, L.A. 2006. Recent progress in understanding
applications. Canadian Journal of Fisheries and
for the California halibut Paralichthys californicus:
larval dispersal: New directions and digressions.
Aquatic Sciences 62:2,425­2,430.
Distribution, elemental fingerprinting and demo-
Integrative and Comparative Biology 46:282­297.
Bath, G.E., S.R. Thorrold, C.M. Jones, S.E. Campana,
graphic approaches. Ph.D. Dissertation, University
Lorens, R.B., and Bender, M.L. 1980. The impact
J.W. McLaren, and J.W.H. Lam. 2000. Sr and
of California, San Diego.
of solution chemistry on Mytilus edulis calcite
Ba uptake in aragonitic otoliths of marine fish.
Forrester, G.E., and S.E. Swearer. 2002. Trace elements
and aragonite. Geochimica et Cosmochimica Acta
Geochimica et Cosmochimica Acta 64:1,705­1,714.
in otoliths indicate the use of open-coast versus
44:1,265­1,278.
Becker, B.J. 2005. The Regional Population Variability
bay nursery habitats by juvenile California halibut.
Lloyd, D.C., D.C. Zacherl, G. Paradis, M. Sheehy, and
and Larval Connectivity of Mytilid Mussels:
Marine Ecology Progress Series 241:201­213.
R.R. Warner. In press. Temperature, natal site and
Conserving the Populations of Cabrillo National
Galindo, H.M., D.B. Olson, and S.R. Palumbi. 2006.
seawater chemistry affect statolith element incor-
Monument (San Diego, CA). Ph.D. Dissertation.
Seascape genetics: A coupled oceanographic-
poration in Kelletia kelletii larvae. Marine Ecology
University of California, San Diego.
genetic model predicts population structure of
Progress Series.
Becker, B.J., F.J. Fodrie, P.A. McMillan, and L.A. Levin.
Caribbean corals. Current Biology 16:1,622­1,626.
Martin, G.B., and S.R. Thorrold. 2005. Temperature
2005. Spatial and temporal variation in trace
Gilg, M.R., and T.J. Hilbish. 2003. The geography
and salinity effects on magnesium, manganese,
elemental fingerprints of mytilid mussel shells: A
of marine larval dispersal: Coupling genetics
and barium incorporation in otoliths of larval and
precursor to invertebrate larval tracking. Limnology
with fine-scale physical oceanography. Ecology
early juvenile spot Leiostomus xanthurus. Marine
and Oceanography 50:48­61.
84:2,989­2,998.
Ecology Progress Series.
88
Oceanography Vol. 20, No. 3

---

ultimately, scientists need to provide


accurate information on population
connectivity, which can be used to
optimize spatial management approaches...




for marine-capture fisheries.
293:223­232.
reserves. Trends in Ecology and Evolution 20:74­80.
Vázquez Peláez, M., J.M. Costa-Fernández, and A.
Martin, G.B., S.R. Thorrold, and C.M. Jones. 2004. The
Strasser, C.A., S.R. Thorrold, V.R. Starczak, and L.S.
Sanz-Medel. 2002. Critical comparison between
influence of temperature and salinity on strontium
Mullineaux. In press. Laser ablation ICP-MS analy-
quadrupole and time-of-flight inductively cou-
uptake in the otoliths of juvenile spot (Leiostomus
sis of larval shell in softshell clams (Mya arenaria)
pled plasma mass spectrometers for isotope ratio
xanthurus). Canadian Journal of Fisheries and
poses challenges for natural tag studies. Limnology
measurements in elemental speciation. Journal of
Aquatic Sciences 61:34­42.
and Oceanography Methods.
Analytical Atomic Spectrometry 17:950­957.
Milton, D.A., and S.R. Chenery. 2001. Sources and
Swearer, S.E., J.E. Caselle, D.W. Lea, and R.R. Warner.
Warner, R.R., S.E. Swearer, J.E. Caselle, M. Sheehy, and
uptake of trace metals in otoliths of juvenile bar-
1999. Larval retention and recruitment in an island
G. Paradis. 2005. Natal trace-elemental signatures
ramundi Lates calcarifer. Journal of Experimental
population of coral-reef fish. Nature 402:799­802.
in the otoliths of an open-coast fish. Limnology and
Marine Biology and Ecology 264:47­65.
Swearer, S.E., G.E. Forrester, M.A. Steele, A.J. Brooks,
Oceanography 50:1,529­1,542.
Moran, A.L., and P.B. Marko. 2005. A simple technique
and D.W. Lea. 2003. Spatio-temporal and inter-
Zacherl, D.C. 2005. Spatial and temporal variation in
for physical marking of larvae of marine bivalves.
specific variation in otolith trace-elemental fin-
statolith and protoconch trace elements as natu-
Journal of Shellfish Research 24:567­571.
gerprints in a temperate estuarine fish assemblage.
ral tags to track larval dispersal. Marine Ecology
Nägler, T.F., A. Eisenhauer, A. Müller, C. Hemleben,
Estuarine, Coastal and Shelf Science 56:1,111­1,123.
Progress Series 290:145­163.
and J. Kramers. 2000. The 44Ca-temperature on
Thorson, G. 1950. Reproductive and larval ecology
Zacherl, D.C., P.H. Manríquez, G. Paradis, R.W. Day,
fossil and cultured Globigerinoides sacculifer: New
of marine bottom invertebrates. Biology Review
J.C. Castilla, R.R. Warner, D.W. Lea, and S.G.
tool for reconstruction of past sea surface tempera-
25:1­45.
Gaines. 2003a. Trace elemental fingerprinting of
tures. Geochemistry Geophysics Geosystems 1: doi
Thorrold, S.R., G.P. Jones, M.E. Hellberg, R.S. Burton,
gastropod statoliths to study larval dispersal trajec-
10.1029/2000GC000091.
S.E. Swearer, J.E. Neigel, S.G. Morgan, and R.R.
tories. Marine Ecology Progress Series 248:297­303.
Neubert, M.G., L.S. Mullineaux, and M.F. Hill. 2006. A
Warner. 2002. Quantifying larval retention and
Zacherl, D.C., G. Paradis, and D. W. Lea. 2003b.
metapopulation approach to interpreting diversity
connectivity in marine populations with artificial
Barium and strontium uptake into larval proto-
at deep-sea hydrothermal vents. Pp. 321­350 in
and natural markers. Bulletin of Marine Science
conchs and statoliths of the marine neogastropod
Marine Metapopulations. J. Kritzer and P. Sale, eds,
70:S291­S308.
Kelletia kelletii. Geochimica et Cosmochimica Acta
Elsevier Academic Press.
Thorrold, S.R., C. Latkoczy, P.K. Swart, and C.M.
67:4,091­4,099.
Russo, R.E., X. Mao, J.J. Gonzales, and S.S. Mao. 2002.
Jones. 2001. Natal homing in a marine fish meta-
Zumholz, K., A. Klügel, T. Hansteen, and U.
Femtosecond laser ablation ICP-MS. Journal of
population. Science 291:297­299.
Piatkowski. 2007. Statolith microchemistry traces
Analytical Atomic Spectrometry 17:1,072­1,075.
Thorrold, S.R., G.P. Jones, S. Planes, and J.A. Hare.
the environmental history of the boreoatlantic
Sale, P.F., R.K. Cowen, B.S. Danilowicz, G.P. Jones,
2006. Transgenerational marking of embryonic
armhook squid Gonatus fabricii. Marine Ecology
J.P. Kritzer, K.C. Lindeman, S. Planes, N.V.C.
otoliths in marine fishes using barium stable iso-
Progress Series 333:195­204.
Polunin, G.R. Russ, Y.J. Sadovy, and R.S. Steneck.
topes. Canadian Journal of Fisheries and Aquatic
2005. Critical gaps impede use of no-take fishery
Sciences 63:1,193­1,197.
Oceanography September 2007
89

---