JOURNAL OF PLANKTON RESEARCH
j
VOLUME
25
j
NUMBER
11
j
PAGES
1427–1439
j
2003
Topographically generated fronts,
very nearshore oceanography and
the distribution and settlement of
mussel larvae and barnacle cyprids
ANITA MCCULLOCH AND ALAN L. SHANKS*
OREGON INSTITUTE OF MARINE BIOLOGY, UNIVERSITY OF OREGON, PO BOX
*CORRESPONDING AUTHOR:
5389, CHARLESTON, OR 97420, USA
[email protected]
Flow patterns adjacent to shore may prevent or aid shoreward migration of benthic invertebrate
larvae. We hypothesized that a front at the mouth of Sunset Bay, Oregon, prevents shoreward
dispersal of larvae, significantly altering settlement of mussel larvae and barnacle cyprids. Settlement
was measured at three sets of moorings (three moorings per site) distributed across the front at Sunset
Bay. From 6 July to 4 September 2000, samples were collected roughly every other day.
Concurrently, we made vertical zooplankton tows adjacent to each mooring site and collected
physical oceanographic data. During upwelling-favorable winds, the front was always present at
the bay mouth, separating significantly cooler, saltier and denser offshore water from that within the
bay. During downwelling winds, the front broke down and we found no significant difference in the
surface physical oceanographic parameters across the bay mouth. During upwelling, the concentration of mussel larvae was higher seaward of the front than landward, but there was no significant
difference in concentration during downwelling, suggesting that the front may act as a barrier to the
shoreward dispersal of mussels. Mussel settlement was too low and sporadic to allow statistical
analysis. There was no difference in cyprid concentrations across the bay mouth whether the front
was present or not. Cyprid settlement was, however, nearly an order of magnitude lower at moorings
seaward of the front than at those landward. A significant cross-correlation was found between
settlement at the offshore mooring and tidal range (r = 0.464, lag = 0 days) and between settlement
at the mid and inner moorings and downwelling winds (r = 0.532 mid bay, r = 0.532 inner bay,
lag = 0 days). Seaward of the front, settlement varied with tidal range, while landward of the front,
most settlement occurred as brief pulses during downwelling winds, periods when the front was not
present. We found large differences in the distribution of cyprids, and mussel larvae and cyprid
settlement relative to the front; larval distributions and settlement varied with upwelling versus
downwelling winds and was due to differences in the very nearshore (i.e. within 100–1000 m of
shore) coastal oceanography.
INTRODUCTION
The majority of coastal invertebrates have complex
life cycles that include pelagic larval stages that must
return to the benthos prior to settlement. Transport by
coastal currents has been shown to affect larval supply,
recruitment and population dynamics of sessile adults
(Gaines and Roughgarden, 1985; Shanks, 1986; Shanks
and Wright, 1987; Farrell et al., 1991; Pineda, 1991;
Alexander and Roughgarden, 1996; Miller and Emlet, 1997).
Nearshore circulation patterns affected by shoreline
irregularities, such as headlands and embayments, can
modify current patterns to create eddies and fronts
(Okubo, 1973; Pingree et al., 1978; Pingree and Maddock,
1979; Wolanski and Hamner, 1988; Black et al., 1990)
that may affect larval recruitment. Several studies have
investigated circulation within larger bays and behind
headlands (Grundlingh and Largier, 1991; Graham
and Largier, 1997), but few have focused on small-scale
circulation within 100 m to a few kilometers from
doi: 10.1093/plankt/fbg098, available online at www.plankt.oupjournals.org
Journal of Plankton Research 25(11), Ó Oxford University Press; all rights reserved
JOURNAL OF PLANKTON RESEARCH
j
VOLUME
shore (Archambault et al., 1998, 1999; Archambault and
Bourget, 1999). Persistent oceanographic features (e.g.
fronts and eddies) in the very nearshore may aid or
prevent the shoreward migration of the pelagic larvae
of coastal fish and invertebrates; settlement at the coast
may vary with the presence or absence of the oceanographic feature. Variations in settlement caused by nearshore oceanography may ultimately affect community
structure. We report a study of nearshore, small-scale,
circulation patterns and their effects on dispersal and
settlement of mussel larvae and barnacle cyprids.
Several studies have shown that high concentrations
of plankton, including meroplankton, are often associated with fronts generated by the interaction of tidal
currents with topography (Alldredge and Hamner, 1980;
Hamner and Hauri, 1981; Wolanski and Hamner, 1988;
Willis and Oliver, 1990; Kingsford et al., 1991; Signell
and Geyer, 1991). With each change of the tide, the
fronts vanished and the concentrations of zooplankton
dissipated. However, if a persistent flow field generates a
front, then it may exist long enough to have measurable
ecological effects on the distribution of zooplankton, the
settlement of the larvae of benthic invertebrates and,
perhaps, the structure of benthic invertebrate populations and communities.
Along the Oregon coast, we have observed shoreparallel foam lines and associated changes in water coloration to persist for days at the mouth of small coves
and bays. The accumulation of foam suggests the existence of a convergence zone and the associated change
in water coloration suggests that the convergence delineates a front separating water masses, one in the cove or
bay and the other offshore. In a companion paper
(Shanks et al., 2003), we present data describing the
spatial structure of physical oceanographic properties
and zooplankton across very nearshore fronts at four
locations along the Oregon coast. One of these sites,
Sunset Bay, is the location where the following study
took place. The mouth of Sunset Bay faces west; when
waves were relatively small (<2 m amplitude) and the
winds were from the north (upwelling favorable), shoreparallel foam lines that extended completely across the
mouths of the bay were nearly always present. When
waves were large (2 m amplitude) or winds were downwelling favorable, foam lines extending across the
mouth of the bay were rare. Large waves at
the mouth of the bay push water into the bay, causing
the sea level to rise. To compensate for the increase in
sea level, a strong rip current forms in the cove and
extends hundreds of meters seaward from the bay
mouth (Shanks et al., 2003). At these times, water was
rapidly flushed from the bay and exchanged with coastal
water. In this and the companion paper (Shanks et al.,
25
j
NUMBER
11
j
PAGES
1427–1439
j
2003
2003), we show that the oceanographic data indicate
that the foam lines were associated with convergent
fronts that separated the surface waters in the bay from
those in the coastal ocean. When the front was present
during upwelling conditions, surface water on the landward side was warmer and generally exhibited lower
chlorophyll concentrations than on the ocean side
(Shanks and McCulloch, 2003; Shanks et al., 2003).
The zooplankton community changed dramatically
across the front. A variety of holoplankters (e.g. calanoid
copepods, copepod nauplii) and meroplankters (e.g.
mussel larvae, gastropod veligers, late-stage barnacle
nauplii) were significantly more concentrated at stations
on the seaward side of the front. We found a second
zooplankton community whose members were, in contrast, significantly more concentrated at stations on the
landward side of the front (e.g. harpacticoid copepods,
Littorina egg cases and Littorina veligers). A few organisms
(e.g. barnacle nauplii stages 1 and 2, cyprids) did not
seem to be affected by the presence of the front. We
have zooplankton data from one date at Sunset Bay
when winds were downwelling favorable and the front
was not present (Shanks et al., 2003). On this date, the
structure of the zooplankton community across the
mouth of the bay was more homogeneous. The distributions of zooplankton were clearly altered by the presence
of the fronts and there appears to be a difference in the
distribution of zooplankton communities depending on
whether the front was present or absent.
During summer, upwelling-favorable winds are common on the Oregon coast and the foam line and front at
the mouth of Sunset Bay are present much of the time
(Shanks et al., 2003). When winds shifted to directions
other than from the north, the front was no longer
present. Based on these observations, we hypothesized
(i) that during upwelling conditions, when the front is
present, settlement will be higher outside the bay than in
it. During periods of non-upwelling winds, when the
front is not present, we hypothesized (ii) that we would
find no difference in settlement rate across the mouth of
the bay. We further hypothesized (iii) that we would find
no difference in the settlement rate of cyprids across the
mouth of the bay whether a front was present or not.
The purpose of the following study was to test these
three hypotheses.
METHOD
Sunset Bay site description
The research was carried out within and adjacent
to Sunset Bay, Oregon (43 200 10000 N, 124 220 75000 W)
(Figure 1). Sunset Bay is located 4 km north of Cape
1428
A. MCCULLOCH AND A. L. SHANKS
j
VERY NEARSHORE OCEANOGRAPHY AND LARVAL SETTLEMENT
Fig. 1. Map of the study site at Sunset Bay, Oregon (43 200 10000 N,
124 220 75000 W ). During upwelling winds, a front forms across the
mouth of Sunset Bay (dashed line across the bay mouth). Larval
settlement was measured at three sets of moorings (n = 3 replicates at
each site, stars) spanning the front. Settlement was sampled roughly
every other day during the summer. At the same time, vertical plankton tows were made adjacent to the mooring sites (one tow per site) and
a CTD transect was made from near the beach in the inner bay past
the seaward moorings.
Arago and 3 km south of the Coos Bay estuary. The
most landward part of the bay is composed of a crescentshaped sand beach to either side of which are extensive
rocky intertidal zones. On the south side of the beach, a
stream empties into the ocean. About 300 m from the
beach, the bay becomes tightly constricted, then opens
up to form the outer portion of the bay (Figure 1). Most
of the shoreline is rocky; two reefs that extend to the sea
form the mouth of the bay. Depending on the tide, the
mouth of the bay is from 0.8 to 0.9 km from the waterline on the beach.
Tides along the Oregon coast are mixed semidiurnal,
with a mean annual range of 1.7 m. During the winter
(October–March), southwest winds produce downwelling conditions and frequent large swells. Beginning in
April or May, winds shift to the northwest, producing
upwelling conditions with occasional relaxation or
downwelling events due to weakening of the upwelling
winds or reversals in wind direction. These downwelling
or relaxation events usually last one to several days
(Barber and Smith, 1981). Swells tend to be smaller
during the spring and summer.
Physical characteristics of the nearshore
water column
In Sunset Bay and the adjacent shelf, determinations of
water temperature, salinity and density were made with
a Seabird Model 19 Conductivity–Temperature–Depth
(CTD) meter. On each sampling day, a CTD cast was
made at each of eight stations along a transect that
extended from 0.2 km from the beach to 1.5 km
offshore. The Noesys Transform program (Noesys,
1999) was used to generate contour plots of the water
column conditions.
Coastal wind data were obtained from the NOAA
Cape Arago Weather Station (CARO3) located 0.5 km
north of Sunset Bay on Gregory Point. Cross- and alongshelf wind stress were calculated using the standard
equation and a constant drag coefficient of 2 103
(Sverdrup et al., 1942). Because we used a constant
drag coefficient, the reported wind stress should be considered pseudo-stress values. The daily average wind
stress was used in time series analysis. The maximum
daily tidal range was obtained from tide tables and
defined as the maximum change in tidal elevation
between a high and adjacent low tide during a day.
Larval settlement time series
The pattern of larval settlement was measured by providing an artificial substratum attached to a set of moorings
that were distributed across the mouth of Sunset Bay.
Replicate moorings (n = 3) were established 3 m apart
at 0.23 km (inner bay, landward of the front, 3 m depth),
0.6 km (mid bay, landward of the front, 10 m depth) and
1.5 km (seaward of the front, 20 m depth) from the shore
(Figure 1). Attached to each buoyed mooring line were
four Plexiglas settlement plates (10 cm2) coated with
Safety-WalkTM tape (3M; Minneapolis, MN, USA) and
four artificial turf substrata units (TuffyTM scrub pads;
770 cm3) (Menge et al., 1997). The settlement plates were
used to measure the settlement of cyprids, and the artificial
turf substrata (hereafter tuffies) measured the settlement of
mussel larvae and other invertebrates. Settlement plates
and tuffies were affixed to the mooring floats and
remained 0.5 m below the surface through all stages of
the tide.
From Julian day 187 through 247 (i.e. July 6 through
September 4, 2000), samples were collected from the
moorings roughly every other day (n = 30 sample
periods). Four sampling periods were longer due to high
seas. The sampling period was extended to 3 days on
Julian day 219 and to 4 days on Julian days 198 and 212.
Settlement plates and tuffies were exchanged in the field
with fresh settlement plates and tuffies. Exposed settlement plates and tuffies were placed in separate bags and
frozen until analysis.
Settlement plates were examined under a dissecting
microscope; cyprids and juvenile barnacles were counted
and identified to species (Shanks, 2001). Typically, the
entire plate was counted; however, if there were
1429
JOURNAL OF PLANKTON RESEARCH
j
VOLUME
>200 settlers, then 10 random squares (2 cm2 each) were
counted. The tuffies were rinsed for several minutes with
fresh water into a 93-mm sieve to remove larvae
(repeated testing indicated that rinsing removed essentially all organisms within the tuffies), after which the
samples were preserved with CaCO3-buffered formalin.
Organisms from the tuffie samples were identified and
enumerated with a dissecting microscope equipped with
a pair of polarizing filters. The filters were placed
between the sample and the light source, and between
the sample and the microscope lens. The latter filter was
rotated until the shells of bivalves and gastropods
appeared to ‘glow’ due to the birefringence caused by
the crystalline structure of the shell (Gallager et al., 1989).
Lighting the samples in this way greatly facilitated sorting. Organisms that were small or difficult to identify
were identified with a compound microscope.
On each sampling day, a vertical plankton tow was
made adjacent to each set of three moorings. The plankton net (53 mm mesh net) had a mouth diameter of
0.25 m, too small a mouth for a flow meter. The volume
filtered was calculated from the length of the tow; the
net was lowered to a depth of 6 m three times for a
combined total tow length of 18 m. Approximately 0.9 m3
of water were filtered during each vertical plankton
tow. Samples were preserved in CaCO3-buffered formalin. Prior to analysis, samples were rinsed with tap water
into a 25 mm sieve and transferred to a beaker. With the
beaker on an electronic balance, tap water was added
until the volume was 200 ml (200 g). After vigorous
random stirring, three 10 ml subsamples were removed
with a Stempel pipette (Peterson et al., 1979; Omori and
Ikeda, 1984). At least 100 individuals of the target organisms (e.g. mussel larvae and cyprids) were counted when
possible. The sample standard deviation was 10% for
the more abundant organisms and between 10 and 20%
for the less common species (Venrick, 1978).
RESULTS
Physical oceanography
During the study, winds were typical of those observed
during a summer on the Oregon coast, with strong winds
from the northwest and weaker winds from the southeast.
There were five periods of downwelling-favorable winds
(i.e. from the south), 21% of the study (Figure 2).
During upwelling winds, a foam line was always
observed across the mouth of Sunset Bay. The foam
line extended from the rock reef on the north side of
the bay to the reef on the south side. Numerous observations (Shanks et al., 2003) suggest that the foam line does
not change position with the tide. During downwelling
25
j
NUMBER
11
j
PAGES
1427–1439
j
2003
Fig. 2. Physical parameters measured during the study (July 1
September 5, 2000 or Julian day 186–252). Wind data were obtained
from the NOAA Cape Arago weather station (CARO3) at Gregory
Point. Positive alongshore and cross-shore wind stress (dynes cm2)
represent winds to the north and east, respectively. Tide data are from
NOAA tide tables.
winds (winds from the south), however, the foam line
vanished from the bay mouth. Figure 3 presents contour
plots of temperature data collected on sample days with
downwelling winds compared with data collected on the
prior sample day when winds were from the north and
upwelling favorable. During upwelling-favorable winds,
surface waters seaward of the foam line were typically
0.5–1 C colder than surface waters landward of the
foam line. The warmer waters landward of the foam
line were present as a thin (<5 m deep) lens of water.
During days with upwelling-favorable winds, the foam
line at the mouth of Sunset Bay delineated a shallow
front where surface water temperatures changed from
warm waters landward to cooler waters offshore of the
foam line.
During days with downwelling winds, a different distribution of temperature was observed. On three dates
( Julian day 190, 204 and 222), surface temperatures in
the bay were either the same as those offshore, or were
cooler (Figure 3). On Julian day 236, the waters within
500 m of shore, well within the bay, were slightly
1430
A. MCCULLOCH AND A. L. SHANKS
j
VERY NEARSHORE OCEANOGRAPHY AND LARVAL SETTLEMENT
Fig. 3. Contour plots of temperature along CTD transects through Sunset Bay, Oregon. Plots on the right are from days with downwellingfavorable winds and those on the left are from the sample day just preceding these collections when winds were upwelling favorable. During
upwelling winds, a foam line was located at 0.7–0.9 km from shore (location indicated by the black arrowhead). During downwelling-favorable
winds, no foam line was present. Black diamonds along the top of each figure indicate station locations.
warmer (0.25 C warmer) than waters seaward of the bay
mouth. There was no change in surface water temperature at the mouth of the bay, as was typically found
during upwelling-favorable winds. On Julian day 246,
the temperature distribution was more complex. We
observed a lens of warm water in the bay, but it was
not connected to the beach; water adjacent to the beach
was cooler than that within the lens. Surface waters at
1431
JOURNAL OF PLANKTON RESEARCH
j
VOLUME
the bay mouth were slightly warmer in the bay than
seaward. During days with downwelling-favorable
winds, both the foam line and temperature front were
generally absent from the bay mouth.
Across the front, the largest differences in the physical
oceanographic variables were generally found at the
surface. To test for differences in the distribution of
physical variables across the front, we compared the
distribution of surface salinity, density and temperature
at stations on either side of the front on days with
upwelling and downwelling winds (i.e. winds from the
north and south, respectively; Figure 4). The purpose of
these tests was to determine whether, when the foam line
was present, there were significant differences in water
mass characteristics to either side of the foam line. If the
water mass characteristics were different, then it is concluded that the foam line delineates a front separating
water masses. The data were separated into two sets:
data collected on days with upwelling-favorable winds
and those collected when winds favored downwelling.
25
j
NUMBER
11
j
PAGES
1427–1439
j
2003
With a Mann–Whitney U-test (two-tailed test), the
surface values of temperature, salinity and density at
stations seaward of the bay mouth or foam line were
statistically compared to values at landward stations.
During upwelling winds, the surface waters seaward of
the foam line were significantly colder (Figure 4) than
those landward of the foam line. The warmest surface
waters were, on average, found adjacent to the beach. In
contrast, on days with downwelling winds, there was not
a significant difference in surface temperature across the
bay mouth (Figure 4). On days with upwelling winds, the
surface waters seaward of the front were significantly
saltier and denser than they were landward of the front
(Figure 4). No significant differences in surface salinity or
density were found across the bay mouth on days with
downwelling winds (Figure 4). On both upwelling and
downwelling days, the water with the lowest surface
salinity and density was, on average, found closest to
shore. This low-salinity water was probably due to
input of fresh water from the small stream that empties
Fig. 4. Plots of average sea surface temperature, salinity and density ( SE) during sample days with upwelling winds (figures on left; n = 18) and
downwelling winds (figures on right; n = 5). The vertical dashed line in the upwelling plots indicates the approximate location of the foam line and
front present at the mouth of Sunset Bay during these conditions. The values in the corner of each figure are the results of a Mann–Whitney U-test
comparing the values observed seaward of the foam line or bay mouth with those landward.
1432
A. MCCULLOCH AND A. L. SHANKS
j
VERY NEARSHORE OCEANOGRAPHY AND LARVAL SETTLEMENT
into the bay adjacent to the beach. The foam line
formed during upwelling winds was associated with a
front that separated two surface water masses, one in the
bay and the other in the offshore coastal waters. During
downwelling-favorable winds, there was no foam line
and no front at the mouth of the bay separating coastal
surface ocean water mass from that in the bay.
The water column distributions of mussel
larvae and cyprids across the front
Mussel larvae (Mytilus spp.) were never abundant in the
zooplankton tows (Figure 5). The highest concentration
was 817 m3, observed at the seaward station, and
there were five sample days (19% of the days) when
none was caught at any of the three stations. Catch
was especially low prior to Julian day 225. The abundances at the inner station were very low and nearly
always less than that at the mid and seaward stations;
mussels were caught on only nine dates (33% of the
dates) at the inner station and the highest concentration
was 15 m3.
Given the low abundances encountered, only limited
statistical analysis is possible and the conclusions drawn
from this analysis should be taken as tentative. The
analysis (non-parametric Wilcoxon matched pairs test)
will be limited to data from the seaward and mid
stations, and will exclude those days when mussel larvae
were not caught at either station (6 out of 27 days).
Comparing this entire data set and those days with
upwelling-favorable winds, significantly more mussel larvae were caught at the seaward station than at the mid
station (entire set: z = 2.67, P = 0.007, n = 21; days with
upwelling winds: z = 3.07, P = 0.002, n = 16). There was
not a significant difference in the concentration of mussel
larvae between stations on days with downwelling winds
(z = 0.135, P = 0.89, n = 5). Tentatively, these results sug-
gest that the concentration of mussel larvae was lowest at
the inner station at all times, that during upwelling
conditions significantly more mussel larvae were present
at the station seaward of the front than at the mid station
landward of the front, and that there was no difference
in the concentration of mussel larvae at the seaward and
mid stations during downwelling winds when the foam
line and front were absent.
The concentrations of Balanus glandula cyprids ranged
from 0 to 140 m3 at the station seaward of the front,
from 0 to 328 m3 at the mid bay station and from 0 to
396 m3 in the inner bay (Figure 7). An analysis of
variance was used to test the effect of condition (upwelling versus downwelling) and mooring location (seaward
of the front, mid bay and inner bay) on the concentration of B. glandula cyprids (no. m3). The data were log
transformed prior to analysis and an Fmax test indicated
that after transformation the variances were homogeneous. No significant differences were found in the concentration of cyprids between stations or between
stations during days with upwelling- or downwellingfavorable winds (Figure 6).
Settlement on tuffies
A variety of larvae were recovered from the tuffies, including Mytilus spp. and other bivalves, nudibranchs, gastropods and megalopae. The following analysis will focus on
Mytilus spp. larvae. Settlements on the four tuffies on each
mooring were summed and the three moorings within
each zone (i.e. seaward, mid and inner) were averaged
(Figure 7). Most (68%) of the mussels settled seaward of
the front; only 26 and 6% of the mussels settled at the mid
and inner bay stations, respectively. Mytilus spp. settlement
was generally low throughout the study (Figure 7) with
settlement per period of 1 individual observed at all
three stations half the time. Settlement of >1 per period
did not occur regularly at the seaward and mid stations
until after Julian day 225. Average settlement per period
was 7.6, 2.5 and 0.6 for the seaward, mid and inner
stations, respectively. Because of the low settlement rates,
the data have not been analyzed statistically. Tentatively,
settlement rate appears to be positively related to the
concentration of mussel larvae in the zooplankton samples.
This settlement time series cannot be used to test whether
there was a difference in the pattern of settlement on days
with upwelling- versus downwelling-favorable winds.
Settlement plate data
Fig. 5. The concentration of mussel larvae in plankton samples collected at stations within Sunset Bay and landward of the front at the
mouth of the bay (mid and inner stations) and outside Sunset Bay and
seaward of the front (seaward station).
Nearly all (99.8%) of the barnacles settling on the plates
were B. glandula; the analysis is limited to this species.
The settlements during each sample period on the four
plates at each mooring were summed and averaged
across the three moorings within each zone (i.e. seaward
1433
JOURNAL OF PLANKTON RESEARCH
j
VOLUME
25
j
NUMBER
11
j
PAGES
1427–1439
j
2003
Fig. 6. Comparison of the average barnacle concentration ( 95% confidence interval) at the three mooring sites during the entire study (left-hand
figure, n = 23), sample periods with upwelling winds (middle figure, n = 18) and sample periods with downwelling winds (right-hand figure, n = 5).
Fig. 7. The average number of mussel settlers in tuffies on moorings
at stations within Sunset Bay and landward of the front at the mouth of
the bay (mid and inner stations) and outside Sunset Bay and seaward of
the front (seaward station).
of the front, mid bay and inner bay). During the first 2
weeks of the study, barnacle settlement was low (0–2
cyprids per plate per settlement period), but settlement
subsequently increased and ranged over three orders of
magnitude (Figure 8).
A total of 3905 individuals settled on the mooring
seaward of the front, 27 501 inshore of the front in the
mid bay and 22 392 in the inner bay (Figure 8); overall
settlement was much higher (by nearly a factor of 10) at
the stations landward of the front. An analysis of variance was used to test the effect of condition (upwelling
versus downwelling) and mooring location (seaward of
the front, mid bay and inner bay) on the settlement rate
of B. glandula cyprids (no. per plate). The data were log
transformed prior to analysis and an Fmax test indicated
that after transformation the variances were homogeneous. The ANOVA indicated that both the wind effect
and the station location effect were significant, but the
interaction term was not significant (Table I). During
sampling periods with upwelling winds (Figure 9), settlement was low and nearly the same across moorings, with
slightly higher settlement rates landward of the bay
mouth and front (i.e. at the mid and inner moorings).
During sampling periods with downwelling winds, however, settlement at the two moorings landward of the bay
mouth was higher by more than a factor of 10 than
settlement at the seaward mooring (Figures 8 and 9).
At stations landward of the front, much of the total
settlement occurred during a few large peaks in settlement that appeared to occur during downwelling or
relaxation events (Figure 8). Time series analysis was
used to investigate the relationship between barnacle
settlement and wind stress during the period when measurable barnacle settlement was occurring (i.e. Julian day
204–251). Prior to the time series analysis, the settlement
data were log transformed and detrended (Otnes and
Enochson, 1978; Dunstan, 1993; Thorrold et al., 1994).
Cross-correlations were run between alongshore and
cross-shore wind stress and B. glandula settlement per
sampling period. Cross-correlations were also investigated between the maximum daily tidal range over the
sample period and cyprid settlement.
No significant cross-correlations (P > 0.05) were
found between along- and cross-shore wind stress and
1434
A. MCCULLOCH AND A. L. SHANKS
j
VERY NEARSHORE OCEANOGRAPHY AND LARVAL SETTLEMENT
Fig. 8. Time series of B. glandula settlement at moorings through Sunset Bay, Oregon (solid lines). Values are the average settlement on three
replicate moorings per day. The inner and middle moorings were landward of the front at the mouth of Sunset Bay and the seaward mooring was
seaward of the front. Plotted with the settlement data is the average alongshore wind stress (dynes cm2) during the sample periods (dotted line).
Positive and negative wind stress values are upwelling- and downwelling-favorable winds, respectively.
Table I: Results of a two-way analysis of variance testing the effect of upwelling and downwelling
winds and mooring site (seaward of the front, mid bay and inner bay) on the settlement of
B. glandula cyprids
df
MS
F
Significance
Upwelling/downwelling winds
1
8.929
11.78
0.001
Sample site
2
2.448
3.230
0.0462
2
1.101
1.452
0.242
63
0.758
Interaction
Error
Larval concentrations were log transformed prior to analysis. The Fmax test indicated that after the transformation the variances were homogeneous.
barnacle settlement at the seaward moorings. However,
a significant positive correlation was found at a lag
of 0 days between the maximum daily tidal range and
settlement at the seaward station (r = 0.464, P < 0.05);
cyprid settlement tended to be largest around spring
tides.
No significant cross-correlations (P > 0.05) were found
between barnacle settlement in the mid and inner zones
1435
JOURNAL OF PLANKTON RESEARCH
j
VOLUME
25
j
NUMBER
11
j
PAGES
1427–1439
j
2003
Fig. 9. Comparison of the average barnacle settlement per sample period ( SE) at the three mooring sites during the entire study (n = 23),
sample periods with upwelling winds (n = 18) and sample periods with downwelling winds (n = 5).
and the maximum daily tidal range. At these stations,
however, there were significant positive correlations at
a lag of 0 days between alongshore wind stress and
barnacle settlement (mid bay: r = 0.531, P < 0.05; inner
bay: r = 0.532, P < 0.05). In addition, we found a significant positive correlation at a lag of 0 days between
cross-shore wind stress and barnacle settlement at the
mid bay moorings (r = 0.517, P < 0.05). Results of these
analyses suggest that, landward of the front, peaks in
barnacle settlement tended to occur during downwelling
winds.
DISCUSSION
During periods of relatively small waves (<2 m) and
upwelling-favorable winds (i.e. north or northwest
winds), a foam line and associated front were consistently
found at the mouth of Sunset Bay. When winds were
downwelling favorable, the foam line and front were not
present. During summer 2000, winds were typical for
the Oregon coast. The dominant winds were upwelling
favorable and from the north-west with occasional
periods of winds producing downwelling or relaxation
events. During the summer sampling period ( July
through September), small waves and north-west winds
were observed 84% of the time. The average duration of
these conditions was 8 days (range 2–20 days). If the
front acts as a barrier to the shoreward migration of
larvae that developed over the shelf, then, for much of
the summer, larvae may be prevented from settling in
the intertidal zone in Sunset Bay.
The concentration of mussel larvae in the zooplankton samples and the settlement rate of mussels to the
tuffies were disappointingly low. The data tentatively
suggest that the front at the mouth of Sunset Bay acts
as a barrier to the shoreward dispersal of mussel larvae;
during upwelling winds, when the front was present, the
concentration of mussel larvae was significantly higher
at the seaward than at the mid station and there was not
a significant difference in the concentration of larvae
during downwelling winds when the front was not present. Shanks et al. report similar results: during upwelling
conditions, mussel larvae were significantly more abundant seaward of the front at both Sunset Bay and Miller’s
Cove (the cove just to the north of Sunset Bay), and
at Sunset Bay, during a downwelling event, there was no
difference in the abundance of mussel larvae across the
mouth of the bay (Shanks et al., 2003). Because of sporadic and low settlement rates, the effect of the front on
mussel settlement could not be determined.
In the zooplankton tows, we did not find significant
variations in the abundance of cyprids by location or by
location crossed with wind condition (upwelling versus
downwelling winds), suggesting that the front at the
mouth of Sunset Bay did not act as a barrier to the
dispersal of cyprids, results identical with those reported
in Shanks et al. (Shanks et al., 2003).
Small peaks in settlement were observed seaward of
the front and these tended to occur around the spring
tides, suggesting that settlement was related to variations
in tidally driven shoreward transport, as has been
observed in southern California (Shanks, 1986; Pineda,
1991). In contrast, landward of the front, large peaks in
cyprid settlement were correlated with downwelling/
relaxation events, periods when the front at the mouth
of the bay was not present. When the front was present,
settlement in the bay was about as low as was observed
1436
A. MCCULLOCH AND A. L. SHANKS
j
VERY NEARSHORE OCEANOGRAPHY AND LARVAL SETTLEMENT
seaward of the front. When the front was absent, settlement at the mid and inner moorings was much higher
(>10 times) than seaward of the front. These short pulses
in settlement accounted for most of the barnacle settlement within the bay.
It is possible that B. glandula cyprids can sense the
relative proximity of shore and preferentially settle
close to it. This type of behavior could account for
higher settlement of cyprids at more nearshore moorings
(i.e. mid and inner moorings). If this behavior does exist,
then we would expect to see higher settlement at nearshore mooring throughout the study. Significantly higher
settlement at more nearshore moorings, however, only
occurred during downwelling-favorable winds; during
upwelling-favorable winds, the average settlement at
the seaward, mid and inner moorings was 7, 9 and 13
cyprids per settlement period (Figure 7), respectively. If
this hypothetical behavior exists, it is difficult to imagine
how it could vary with wind direction.
Landward of the front, peaks in cyprid settlement
occurred during downwelling/relaxation events. Settlement that occurs immediately following upwelling events
has been attributed to the upwelling front moving
back on shore and transporting larvae trapped in the
frontal convergence zone shoreward (Farrel et al., 1991;
Roughgarden et al., 1991; Wing et al., 1995a,b; Miller and
Emlet, 1997). If peaks in cyprid settlement had occurred
at all three moorings during downwelling/relaxation
events, then we could hypothesize that variations in
settlement were due to variations in onshore transport
due to upwelling fronts traveling to shore. However,
only at the moorings landward of the front did we find
settlement peaks associated with downwelling/relaxation
events; at the seaward mooring we did not find any
significant cross-correlations between settlement and
wind direction. At the mid and inner moorings, settlement
peaks during downwelling-favorable winds were, therefore, not due to the onshore transport of cyprids from the
offshore coastal waters by an upwelling front propagating
back toward shore. To attempt to explain these peaks in
settlement, we suggest a different mechanism.
Several studies have demonstrated that cyprids can
become concentrated in convergence zones (LeFevre,
1986; Shanks and Wright, 1987; LeFevre and Bourget,
1991; Pineda, 1994, 1999). The foam line and front at
the mouth of Sunset Bay are characterized by convergent flow; dye placed in the water to either side of the
foam line moved toward the foam line, where it downwelled, and abundant buoyant flotsam is found in the
foam line. During upwelling conditions, when the front
was present at the mouth of Sunset Bay, cyprids may
become concentrated in the frontal convergence. In
Chile, similar foam lines and fronts form across the
mouth of small coves during upwelling and neuston
tows indicated that meroplankton were concentrated in
the foam lines (A. L. Shanks and P. Manriquez, unpublished data). During downwelling/relaxation events, we
have observed the foam line at the mouth of Sunset Bay
bend into the bay. Usually, the northern end of the foam
line became detached from the rocks and extended
well into the bay. We have also observed foam lines in
other locations to propagate into bays and to the shore
during downwelling-favorable winds (El Quisco, Chile;
A. L. Shanks and P. Manriquez, unpublished data). In
other words, at the onset of downwelling conditions, the
foam line at the mouth of the bay appears to move into the
bay, where it eventually contacts the shore and disperses.
If cyprids are concentrated in the convergent front delineated by the foam line and remain with the foam line as
it moves into the bay during a downwelling/relaxation
event, then they could be carried into the bay, generating a pulse in settlement. In other words, the foam line
acts as a moving convergence capable of transporting
trapped larvae the several hundred meters to shore.
These settlement pulses would appear exactly like
what one would expect to see if settlement pulses were
due to an upwelling front relaxing to shore (Farrell et al.,
1991; Roughgarden et al., 1991; Wing et al., 1995a,b;
Miller and Emlet, 1997); however, the physical oceanography and geographic distribution of the fronts are
radically different. Along the Oregon coast, upwelling
fronts form 10–15 km from shore and delineate the
zone of separation between cold upwelled waters and
warmer surface waters displaced by upwelling. Upwelling fronts extend for tens of kilometers in the alongshore
direction. During a relaxation event, the upwelling front
moves back toward shore as a density current (Shanks
et al., 2000) composed of the warm offshore surface water
flowing over the top of the denser upwelled waters. During
upwelling relaxation, as the upwelling front moves back
toward shore, larvae may be transported kilometers shoreward (Shanks et al., 2000) and, due to the extent of the
upwelling front in the alongshore direction, settlement of
larvae transported by the front may be nearly synchronous
along an extensive length of shoreline.
How upwelling conditions generate the convergent
fronts at the mouth of bays is not understood. We do
know that they only form during upwelling winds (in
Chile, winds from the south generate foam lines and
fronts; A. L. Shanks and P. Manriquez, unpublished
data) and that the ends of the foam lines and convergent
fronts usually extend from shore to shore across the bay
mouth. At the onset of downwelling-favorable conditions, the foam line appears to move into the bay. We
found short intense settlement pulses associated with
downwelling conditions that we hypothesize were due
1437
JOURNAL OF PLANKTON RESEARCH
j
VOLUME
to the shoreward movement of cyprids trapped in the
foam line at the mouth of Sunset Bay. Where a similar
topographic front is not present, we predict that similar
settlement pulses will not occur. If this is true, then the
intensity of settlement along a coast could be a mosaic;
pulsed high settlement of cyprids may be seen in coves,
with lower and perhaps less variable settlement along
shorelines between coves.
Despite the very large differences in the oceanography
of typical upwelling fronts and the topographically generated fronts described in this paper, the temporal pattern of settlement generated by an upwelling front
relaxing to shore and a front such as that at the mouth
of Sunset Bay relaxing to shore would look similar; both
types of front move to shore during relaxation from an
upwelling wind. In a geographic setting such as Sunset
Bay, attributing settlement pulses to an upwelling front
relaxing to shore, however, could be incorrect. For
example, Miller and Emlet (Miller and Emlet, 1997)
looked at settlement of echinoderms in Miller’s Cove, a
cove just to the south of Sunset Bay with very similar
oceanography (Shanks et al., 2003). They found peaks in
the settlement of echinoderms associated with downwelling/relaxation events and attributed these peaks to the
shoreward transport of larvae due to upwelling relaxation events. Given the similarity of the front at the
mouth of Miller’s Cove to that at Sunset Bay, an alternate hypothesis would be that the peaks in settlement
were caused by the loss of the front at the mouth of
Miller’s Cove. It appears that careful investigation of the
very nearshore oceanography is needed to determine
whether temporal variations in settlement at coastal
sites depend on variations in the onshore transport of
larvae from the coastal ocean; variations in settlement
could be due to variations in onshore transport from the
coastal ocean to shore, variations in the very nearshore
oceanography (presence or absence of a topographic
front), or both.
The data reported in this paper demonstrate that the
front at the mouth of Sunset Bay, a small cove on the
Oregon coast, may alter the distribution of mussel larvae
and the settlement of barnacles. During upwelling-favorable
winds, the front appears to block the shoreward dispersal of mussels. During downwelling-favorable winds, the
front at the mouth of Sunset Bay is not present and
during these periods we observed large pulses in cyprid
settlement. Just hundreds of meters to the south of
Sunset Bay, the nearshore oceanography is quite different (Shanks et al., 2003); a shore-parallel front is present
but the mechanism of formation is quite different. This
change in oceanography could affect settlement. Perhaps
alongshore variations in larval settlement at the spatial
25
j
NUMBER
11
j
PAGES
1427–1439
j
2003
scale of hundreds to thousands of meters are driven by
variations in the very nearshore oceanography.
ACKNOWLEDGEMENTS
Field assistance was provided by J. April, D. Williams,
J. Miller and H. Shanks. This research was funded in
part by a Small Grant for Exploratory Research (OCE0002891) from the National Science Foundation. The
manuscript was written during a Fulbright Scholarship
to the Pontificia Universidad Catolica de Chile, Estación
Costera de Investigaciones Marinas in Las Cruces,
Chile. The paper was much improved by comments
from Drs Sergio Navarrete, Patricio Manriquez and
Juan Carlos Castilla.
REFERENCES
Alexander, S. E. and Roughgarden, J. (1996) Larval transport and
population dynamics of intertidal barnacles: a coupled benthic/
oceanic model. Ecol. Monogr., 66, 259–275.
Alldredge, A. L. and Hamner, W. M. (1980) Recurring aggregation of
zooplankton by tidal current. Estuarine Coastal Mar. Sci., 10, 31–37.
Archambault, P. and Bourget, E. (1999) Influence of shoreline configuration on spatial variation of meroplanktonic larvae, recruitment
and diversity of benthic subtidal communities. J. Exp. Mar. Biol. Ecol.,
238, 161–184.
Archambault, P., Roff, J. C., Bourget, E., Bang, B. and Ingram, G. R.
(1998) Nearshore abundance of zooplankton in relation to shoreline
configuration and mechanisms involved. J. Plankton Res., 20, 671–690.
Archambault, P., Mckindsey, C. W. and Bourget, E. (1999) Large-scale
shoreline configuration influences phytoplankton concentration and
mussel growth. Estuarine Coastal Shelf Sci., 49, 193–208.
Barber, R. T. and Smith, R. L. (1981) Coastal upwelling. In Longhurst, A.
(ed.), Analysis of Marine Ecosystems. Academic Press, London, pp. 31–67.
Black, K. P., Gay, S. L. and Andrews, J. C. (1990) Residence times of
neutrally-buoyant matter such as larvae, sewage, or nutrients on
coral reefs. Coral Reefs, 9, 105–114.
Dunstan, F. D. J. (1993) Time series analysis. In Fry, J. C. (ed.),
Biological Data Analysis: A Practical Approach. Oxford University Press,
New York, pp. 243–310.
Farrell, T. M., Bracher, D. and Roughgarden, J. (1991) Cross-shelf
transport causes recruitment to intertidal populations in central
California. Limnol. Oceanogr., 36, 279–288.
Gaines, S. D. and Roughgarden, J. (1985) Larval settlement rate: a
leading determinant of structure in an ecological community of the
marine intertidal zone. Proc. Natl Acad. Sci. USA, 82, 3707–3711.
Gallager, S. M., Bidwell, J. P. and Kuzirian, A. M. (1989) Strontium is
required in artificial seawater for embryonic shell formation in two
species of bivalve molluscs. In Crick, R. (ed.), Origin, Evolution, and
Modern Aspects of Biomineralization in Plants and Animals. Plenum Press,
New York.
Graham, W. M. and Largier, J. L. (1997) Upwelling shadows as
nearshore retention sites: the example of northern Monterey Bay.
Cont. Shelf Res., 17, 509–532.
1438
A. MCCULLOCH AND A. L. SHANKS
j
VERY NEARSHORE OCEANOGRAPHY AND LARVAL SETTLEMENT
Grundlingh, M. L. and Largier, J. L. (1991) Physical oceanography of
False Bay: a review. Trans. R. Soc. S. Afr., 47, 387–400.
zone: the cause of recruitment pulses in barnacle populations of
central California. Acta Oecol., 12, 35–51.
Hamner, W. M. and Hauri, I. R. (1981) Effects of island mass: water
flow and plankton pattern around a reef in the Great Barrier Reef
lagoon, Australia. Limnol. Oceanogr., 26, 1084–1102.
Shanks, A. L. (1986) Tidal periodicity in the daily settlement of intertidal barnacle larvae and an hypothesized mechanism for the crossshelf transport of cyprids. Biol. Bull., 170, 429–440.
Kingsford, M. J., Wolanski, E. and Shoat, J. H. (1991) Influence of
tidally induced fronts and Langmuir circulations on distribution and
movement of presettlement fishes around a coral reef. Mar. Biol.,
109, 167–180.
Shanks, A. L. (2001) An Identification Guide to the Larval Marine Invertebrates
of the Pacific Northwest. Oregon State University Press, Corvallis, OR.
LeFevre, J. (1986) Aspects of the biology of frontal systems. Adv. Mar.
Biol., 23, 163–299.
LeFevre, J. and Bourget, E. (1991) Neustonic niche for cirripede larvae
as a possible adaptation to long-range dispersal. Mar. Ecol. Prog. Ser.,
74, 185–194.
Menge, B. A., Daley, B. A., Wheeler, P. A. and Strub, P. T. (1997)
Rocky intertidal oceanography: an association between community
structure and nearshore phytoplankton concentration. Limnol. Oceanogr.,
42, 57–66.
Miller, B. A. and Emlet, R. B. (1997) Influence of nearshore hydrodynamics on larval abundance and settlement of sea urchins Stronglylocentrotus franciscanus and S. purpuratus in the Oregon upwelling zone.
Mar. Ecol. Prog. Ser., 148, 83–94.
Shanks, A. L. and McCulloch, A. (2003) Topographically generated
fronts, very nearshore oceanography, and the distribution of chlorophyll, detritus, and selected diatom and dinoflagellate taxa. Mar.
Biol., 296, 113–126.
Shanks, A. L. and Wright, W. G. (1987) Internal-wave-mediated shoreward transport of cyprids, megalopae, and gammarids and correlated longshore differences in the settling rate of intertidal barnacles.
J. Exp. Mar. Biol. Ecol., 114, 1–13.
Shanks, A. L., Largier, J., Brink, L., Brubaker, J. and Hooff, R. (2000)
Demonstration of the onshore transport of larval invertebrates by the
shoreward movement of an upwelling front. Limnol. Oceanogr., 45,
230–236.
Shanks, A. L., McCulloch, A. and Miller, J. (2003) Topographically
generated fronts, very nearshore oceanography and the distribution of
larval invertebrates and holoplankters. J. Plankton Res., 25, 1257–1277.
Noesys (1999) Noesys User’s Guide and Reference Manual. Research Systems,
Boulder, CO.
Signell, R. and Geyer, W. R. (1991) Transient eddy formation around
headlands. J. Geophys. Res., 96, 2561–2575.
Okubo, A. (1973) Effect of shoreline irregularities on streamwise dispersion in estuaries and other embayments. Neth. J. Sea Res., 6, 213–224.
Sverdrup, H. U., Johnson, M. W. and Fleming, R. H. (1942) The
Oceans: Their Physics, Chemistry, and General Biology. Prentice-Hall,
Englewood Cliffs, NJ, 1087 pp.
Omori, M. and Ikeda, T. (1984) Methods in Marine Zooplankton Ecology.
Wiley & Sons, New York, 332 pp.
Otnes, R. F. and Enochson, L. (1978) Time Series Analysis: Basic Techniques.
Wiley and Sons, New York.
Thorrold, S. R., Shenker, J. M., Maddox, E. D., Mojica, R. and
Wishinski, E. (1994) Larval supply of shorefishes to nursery habitats
around Lee Stocking Island, Bahamas. II. Lunar and oceanographic
influences. Mar. Biol., 118, 567–578.
Peterson, W. T., Miller, C. B. and Hutchinson, A. (1979) Zonation and
maintenance of copepod populations in the Oregon upwelling zone.
Deep-Sea Res., 26A, 467–494.
Venrick, E. L. (1978) How many cells to count? In Sournia, A. (ed.),
Phytoplankton Manual. UNESCO, Paris, pp. 167–180.
Pineda, J. (1991) Predictable upwelling and the shoreward transport of
planktonic larvae by internal tidal bores. Science, 253, 548–551.
Willis, B. L. and Oliver, J. K. (1990) Direct tracking of coral larvae:
implications for dispersal studies of planktonic larvae in topographically complex environments. Ophelia, 32, 145–162.
Pineda, J. (1994) Spatial and temporal patterns in barnacle settlement
rate along a southern California rocky shore. Mar. Ecol. Prog. Ser.,
107, 125–138.
Pineda, J. (1999) Circulation and larval distribution in internal tidal
bore warm fronts. Limnol. Oceanogr., 44, 1400–1414.
Pingree, R. D. and Maddock, L. (1979) The tidal physics of headland
flows and offshore tidal bank formation. Mar. Geol., 32, 269–289.
Pingree, R. D., Bowman, M. J. and Esaias, W. E. (1978) Headland
fronts. In Bowman, M. J. and Esaias, W. E. (eds), Oceanic Fronts in
Coastal Processes. Springer-Verlag, Berlin, pp. 78–86.
Roughgarden, J., Pennington, J. T., Stoner, D., Alexander, S. and
Miller, K. (1991) Collisions of upwelling fronts with the intertidal
Wing, S. R., Botsford, L. W., Largier, J. L. and Morgan, L. E. (1995a)
Spatial structure of relaxation events and crab settlement in the northern California upwelling system. Mar. Ecol. Prog. Ser., 128, 199–211.
Wing, S. R., Largier, J. L., Botsford, L. W. and Quinn, J. F. (1995b)
Settlement and transport of benthic invertebrates in an intermittent
upwelling region. Limnol. Oceanogr., 40, 316–329.
Wolanski, E. and Hamner, W. M. (1988) Topographically controlled
fronts in the ocean and their biological influence. Science, 241,
177–181.
Received on June 2, 2003; accepted on August 9, 2003
1439