Journal of Plankton Research Vol.20 no.3 pp.585-601, 1998
Seasonal and tidal abundance patterns of decapod crustacean
larvae in a shallow inlet (SW Spain)
P.Drake, A.M.Arias and A.Rodriguez .
Institute de Ciencias Marinas deAndalucia (CSIC), 11510 Puerto Real, Cadiz,
Spain
Abstract. Planktonic larvae of decapod crustaceans were collected monthly from July 1991 to June
1992 by pumping during nocturnalfloodand ebb tides to establish seasonal larval abundance patterns
in an inlet of the Bay of Cadiz. Additional 24 h series of samples were collected seasonally (July 1991,
October 1991, January 1992 and May/June 1992) during spring and neap tides to analyse larval
abundance in relation to the main environmental cycles (diel, tidal and lunar phases) and vertical
position in the water column. First zoeae were the most abundant stage for most species, representing 97.6% of all individuals collected. Zoea I abundance was higher in spring and summer and, on
most sampling occasions, there was a net output from the inlet to the bay. Five species (Liocarcinus
arcuatus and Liocarcinus vemalis, Uca tangeri, Diogenes pugilator and Panopeus africanus) represented 60% of total individuals caught. The seasonal occurrence of first zoeae of the most abundant
species indicated two different reproductive patterns: species with a short reproductive period and
species spawning year round. Zoea I of several species (Panopeus africanus, Uca tangeri, Pachygrapsus marmoratus, Processa spp.) were significantly more abundant during ebb tides and their later
larval stages were scarcely collected, suggesting that these larvae are released in the inlet and exported
to the bay. Conversely, a net input of first zoeae was observed for other species (D.pugilator and
Pinnotheres pinnotheres), but their later larval stages were also scarcely collected. Such importation
could be a larval rhythm artifact due to release of larvae in the bay that drifted into the inlet by tidal
currents. The crab Ilia nucleus, whose later larval stages were collected frequently, was the only
species that seemed to complete its life cycle within the bay. These results suggest that the studied
inlet was primarily used by decapods as an adult habitat and spawning ground, while larval development occurred in open sea. Since vertical migration was not observed for exported larvae, the tidal
synchronization of female release seemed to be the most probable mechanism of larval exportation.
There were no significant differences between larval release during spring and neap tides.
Introduction
Decapod crustacean species inhabiting estuarine systems show two main life cycle
patterns: species which complete their larval development within the estuary, and
species whose newly hatched zoeae drift out from the estuary and subsequently
reinvade estuarine habitats mainly as megalops or juveniles. To be exported from
a tidal system, or to be retained within it, (i) larvae may synchronize their activity
and/or (ii) females may synchronize the larval release with the most suitable tidal
phases: high and ebb tides for being carried out, and low and flood tides to be
retained. The first mechanism may be accomplished by endogenously controlled
tidally rhythmic vertical migration of larvae (Cronin and Forward, 1979; Zeng
and Naylor, 1996); such migratory behaviour was observed for estuarine decapod
crustacean larvae that were retained within estuaries (Cronin, 1982) or exported
from them (Queiroga et ai, 1997); also flood-tide transport was suggested for
megalops reinvading estuaries (Tankersley et ai, 1995). In relation to the second
mechanism, synchronous larval release timed to minimize predation has been
proposed for decapod crustacean species inhabiting intertidal and shallow waters
(Forward, 1987; Morgan and Christy, 1995). Theoretically, nocturnal spring high
© Oxford University Press
585
P.Drake, AJVLArias and A.Rodriguez
tides are the best time to release larvae for species completing their larval
development in open sea. Under such conditions, adults and larvae are least likely
to be seen by predators, and larvae are transported more quickly to offshore
waters. A rapid larval transport from shallow to offshore waters may represent a
slower predator pressure by predatory fishes that are less abundant offshore
(Morgan, 1990). An alternative hypothesis is that larval exportation avoids stressful water conditions in estuarine areas. However, differences in adult habitats and
larval vulnerability should also be considered to understand the adaptive significance of the different larval release patterns that are shown by decapod crustacean species (De Vries and Forward, 1989, 1991; Morgan and Christy, 1995).
On the other hand, larval retention within estuaries appears to be restricted to
those species with adult populations confined to estuaries or which utilize upper
estuarine habitats in part of their life cycle (Cronin and Forward, 1979; Cronin,
1982). For these species, retention of larvae in the estuary may be the primary
mechanism of recruitment to estuarine adult populations (Sandifer, 1975). In
contrast, larvae of crustacean decapod species not restricted to estuarine habitats
are often flushed into the open sea by tidal currents (Johnson and Gonor, 1982;
Dittel et al, 1991); the major mechanism by which the parental population is
restocked is the immigration of megalops, juveniles or adults from coastal waters
(Sandifer, 1975; Wehrtmann and Dittel, 1990; Little and Epifanio, 1991).
The Rio San Pedro inlet was originally a part of the Guadalete River estuary,
but sedimentary processes and human activity together have changed it into a
marine tidal inlet. Despite these hydrodynamic changes, the ecosystem persists
in performing an important role in the early life of marine fish species as a nursery
area (Drake and Arias, 1991). Yet, there is very little knowledge about the use of
the inlet by larvae and post-larvae of macrobenthic species such as decapod crustaceans, except a recent study on the crabs Panopeusafricanus and Uca tangeri
(Rodriguez et al, 1997). Furthermore, the only information available about the
larval abundance rhythms of decapod crustaceans in the Iberian Peninsula is from
typical estuarine systems (Miiller, 1983; Fuste\ 1989; Paula, 1989,1993). Thus, the
present paper studies the larval abundance pattern of decapod crustacean species
in a shallow tidal inlet, where the freshwater inflow is insignificant.
Method
Study site
This study was carried out in Rio San Pedro, a shallow inlet of the saltmarsh zone
situated to the east and south of the Bay of Cddiz (36°23'-37'N, 6°8'-15'W). This
sinuous 12-km-long inlet, characterized by semidiurnal mesotides (tidal range
1-3.5 m) and a soft muddy bed, crosses the north-eastern part of the salt marsh
and terminates in the middle east shore of the bay (Figure 1). The tidal current
from the bay flows along the inlet and supplies sea water to some saltmarsh fish
ponds situated on both sides of its course. Freshwater inflow is insignificant except
during heavy rains.
The sampling site was located 3 km from the inlet mouth, where the channel
width was -200 m during high spring tides, and at a point situated 50 m from the
586
Seasonal and tidal abundance patterns of decapod crustacean larvae in a shallow inlet
Rio San Pedro
inlet
Fig. 1. Location of the sampling site in the Bay of Qdiz, SW Spain. Arrows show water circulation
within the bay during the ebb tide (Miinoz and Sanchez-Lamadrid, 1994; Parrado el a/., 1996).
water's edge, where its bottom was 1.28 m below chart datum. There the inlet
cross-section showed a continuously submerged central channel (mean depth =
1.44 m below hydrographic zero; SD = 0.99 m), and lateral tidal flats partially
exposed at low tides (mean depth = 2.19 m above hydrographic zero; SD =
0.99 m).
Sampling
To estimate the annual patterns of larval abundance, decapod crustacean larvae
were collected from July 1991 to June 1992. Samples were taken monthly during
the night hours of spring tides. Each sampling consisted of two pumpings, lasting
1 h each (30 min at 0.5 m below the surface and 30 min at 0.5 m above the bottom),
and started 4 h before (flood) and 3 h after (ebb) the expected time of high water,
587
P.Drake, A.M.Arias and A.Rodriguez
with a gasoline-powered impeller pump whose outflow was filtered through a 300
urn mesh. The pump outflow was calibrated before each sampling period and kept
approximately constant for all samples, so each 30 min pumping allowed the
filtration of 10-12 m3 of water. To establish the patterns of larval stages in relation
to diel, tidal and spring/neap semilunar cycles, as well as vertical position in the
water column, additional samples were collected seasonally in July 1991
(summer), October 1991 (autumn), January 1992 (winter) and May-June 1992
(spring). During each sampling period, four 24 h series (two spring and two neap
tides) of eight samples, lasting 1 h each (30 min at 0.5 m below the surface and 30
min at 0.5 m above the bottom separately), were taken at 3 h intervals in synchronization with tidal phases (at each diurnal/nocturnal high, ebb, low and flood
tides), in order to take two samples for each diel, tidal and lunar situation, and
vertical position in the water column. Samples were preserved immediately in 5%
formaldehyde. Larvae were categorized by zoeal and megalopal stages, and
sorted into species.
Water temperature and salinity were measured at the start of each pumping
sample and current speed estimated during the sampling time with a Hydrobios
current meter (Figure 2).
Data analysis
The number of decapod crustacean larvae caught in each pumping was adjusted
to give their abundance per cubic metre of filtered water. Considering the different current speed and larval density at the surface and bottom of the water
column, we have estimated the input (flood) and output (ebb) of decapod crustacean larvae at the study inlet as the number of individuals passing every hour
through one square metre section. For each of the 12 most abundant species, a x2
test for heterogeneity was used to compare the monthly larval density at flood
and ebb tides. The number of larvae expected was estimated in proportion to the
volume of water filtered in each tide situation.
The contribution of the factors season and lunar phase to the variation
observed in the total number of zoeae caught during each 24 h series was ascertained using two-way nested (top factor = season; nested factor = lunar phase)
analysis of variance (ANOVA). Because the short-term (between consecutive or
alternate days) and long-term (between seasons) variation in zoeal density (24 h)
for each species was often considerable (see Results), zoeal abundance data were
expressed at each diel/tidal situation as a percentage (relative abundance) of the
total number caught during each 24 h series. After this transformation, and for
each species, a four-way ANOVA was performed to test the effect of diel (D),
tidal (T) and lunar phases (L) on the larval abundance, as well as the larval vertical position in the water column (P). Only second-order interactions of these
factors were analysed, and lunar phase was considered only to estimate interaction effects. Relative larval abundance (%) was arcsine Vx-transformed prior
to ANOVA analysis. Factors detected to be significant by ANOVA were analysed
further using a posteriori Student-Newman-Keuls tests set at the 5% significance
level.
588
Seasonal and tidal abundance patterns of decapod crustacean larvae in a shallow inlet
~ 3°
o
£ 20
10
45
S 40
CO
35
1.0
T
| 0,5
I 0,0
3 0.5
0,0
250
<? 200
E
100
50
0
ffi
B- n.n -
- 10
50
J A S O N D J
1991
FMAMJ
1992
Fig. 2. Monthly water temperature, salinity, current speed, zoea I density and zoea I flux of decapod
crustaceans in the Rio San Pedro inlet. For each month, bars on the left and right of the mark correspond to the flood and ebb values, respectively; the shaded part of each bar represents individuals
that were collected close to the bottom.
589
P.Drake, AJVLArias and A.Rodriguez
Results
Larval stage composition
A total of 25 030 individuals (zoeae and megalops) of decapod crustaceans were
collected during the study period (Table I). They were categorized into 24 species,
five groups identified to genus and a residual group (1.2%) of unidentified individuals. Zoea I larvae were predominant, representing 97.6% of all individuals
collected. Later larval stages were rarely or never (50% of species) collected.
Only for a few species {Ilia nucleus, Palaemon spp. and Parthenope angulifrons)
did later zoeal stages represent >30% of the total number of individuals caught
for the species, while Penaeus kerathurus was the only species whose first larval
stage was not found. A total of 47 megalops and eight first crabs were collected,
Panopeus africanus (nine megalops) and Pinnotheres pinnotheres (eight megalops, two first crabs) being the best represented.
Liocarcinus spp. larvae (L.arcuatus and L.vernalis) were dominant, representing 20.8% of all collected larvae, followed by Uca tangeri (16.1%), Diogenes pugilator (11.7%), Panopeus africanus (10.6%), Pinnotheres pinnotheres (7.2%),
Carcinus maenas (6.2%), Processa spp. (5.6%) and Upogebia deltaura (4.0%).
Zoea I of several of these species (Processa spp., Upogebia deltaura, Diogenes
pugilator, Liocarcinus spp. and Pinnotheres pinnotheres) were collected during
most of the year, indicating a prolonged reproductive period (Table I). For others
(Cmaenas, Panopeus africanus and Uca tangeri), zoea I were found only during
a part of the year, suggesting a short reproductive period.
Seasonal larval abundance
Zoea I of decapod crustaceans were present in the study inlet during the whole
year and showed a marked seasonal trend, with maximal abundance in spring and
summer, and minimum in winter (Figure 2). In most samplings, zoea I were more
abundant at the surface of the water column and during ebb tide. Thus, there was
a net output of zoea I from the inner part of the inlet to the bay, primarily during
the warm period, but when monthly zoea I abundance was considered individually for the 12 most abundant species (Figure 3), onlyfivespecies (Processa spp.,
Liocarcinus spp., Panopeus africanus, Uca tangeri and Pachygrapsus marmoratus) showed individual patterns consistent with the global trend. Two species
(Pinnotheres pinnotheres and Upogebia deltaura) seemed to be imported into the
inlet during the warm period and one (D.pugilator) in spring and autumn, and a
last species (Cmaenas) was exported in winter. The rest of the most abundant
species (Hippolyte spp., Pisidia sp. and I.nucleus) did not show a clear output or
input trend for their zoea I.
When the total numbers of zoea I caught during each of the 24 h series were
compared (Table II and Figure 4), seasonal differences in abundance were statistically significant (two-way nested ANOVA; P < 0.05) for seven (Upogebia
deltaura, I.nucleus, Cmaenas, Panopeus africanus, Pinnotheres pinnotheres, Uca
tangeri and Pachygrapsus marmoratus) of the 12 most abundant species. For later
zoeae (>zoea I) and megalops, the maximal abundance was observed in the spring
590
X
XA
X
X
XA
A
X
X
Brachynolus sexdentatus (I, B)
Parthenope angulifrons (B)
Macropodia spp. (I, B)
Mo/a squinado (B)
Ma;'a sp.
NI
X
XA
X
XA
XA
XA
X
X
X
X
X
X
X
X
A
XA
X
XA
X
X
X
X
X
X
X
A
J
Pinnotheres pinnotheres (I, B)
L/ca langeri (I, B)
Pachygrapsus marmoratus (I)
Pachygrapsus transversus
Xantho sp. (B)
Penaeus kerathurus (I, B)
Sicyonia carinata (I, B)
Hippolyte spp. (1, B)
Processa spp. (I, B)
Palaemon spp. (I, B)
Crangon crangon (I, B)
Phylocheras fasciatus (I)
Callianassa sp. (I)
Upogebia deltaura (I, B)
Diogenes pugilator (B)
Spiropagurus elegans (B)
Porcellana platycheles
Pisidia sp. (I, B)
Medorippe lanata (I)
Elhusa mascarone (B)
//w nucleus (B)
Carcinus maenas (I, B)
Liocarcinus spp. (I, B)
Panopeus africanus (I)
Species
X
X
X
X
XA
X
X
X
X
X
X
X
X
XA
X
X
X
X
X
X
A
X
X
X
XA
A
X
o
s
X
X
X
X
X
X
X
X
X
X
X
N
X
X
X
X
X
D
X
X
X
X
XA
X
X
X
X
X
X
X
J
X
X
X
X
F
X
X
X
X
X
M
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
A
X
XA
X
X
X
XA
XA
XA
X
XA
X
XA
X
X
X
X
X
X
X
X
X
XA
XA
X
X
X
M
X
XA
X
X
X
X
X
XA
XA
X
X
X
XA
X
X
X
X
X
J
591
39
255
12
347
2
2
287
1740
4035
1
1516
5083
2650
43
162
856
28
67
440
342
994
2936
22
54
62
1391
36
435
"z.
2.4
0.2
1.0
0.0
1.4
0.0
0.0
1.1
16.5
0.0
7.1
20.8
10.8
0.2
0.7
3.5
0.1
0.3
1.8
6.2
1.8
5.7
0.1
0.2
0.3
1.4
4.0
12.0
0.1
%Z.
18
10
7
57
62
4
12
223
25
10
14
12
130
12
6
N>z.
3.0
1.7
1.2
9.5
0.7
10.3
4.2
1.7
2.3
36.9
2.0
2.0
21.5
2.0
1.0
%>z.
Table I. Monthly presence of zoea I (X) and advanced larval stages (A), total number of individuals of zoea I (Nzl) and later larval stages (/V>zi), and percentage
of abundance of zoea I (%zi) and later larval stages (%>zi) for decapod crustacean species caught at the Rfo San Pedro inlet. NI, no identified larvae. In
parentheses, I and B mean that adults of the species were observed in the study inlet and in the Bay of Cddiz, respectively (Arias, 1976; Drake et at., 1997)
1
3
IES
Ui
to
B'
i
g5S
n
Q
H
1
12,
•a
|
i
s
tr
1so
ft
On
5.
|
Vi
inlet
P.Drake, AJVLArias and A.Rodriguez
Hippotyte spp.
Caranus maenas
6
30
4
20
2
0
30
10
i
i i i i i i i
0
r
1
i r
Uocardnus spp. n
Processa spp.
200
20
100
10
0
18
O
O
O
•H-
r-F-rrtr-rrr
fTf
Upogebia deftaura
Panopeus africanus
45
12
30
6
15
0
T
rT T
i
0
r
Diogenes pugilator
CM
0
•
^
i i i i r
Pinnotheres pinnotheres
12
20
8
10
N
o
0
6
4
i
i i r
Piskfasp.
0
r i i n
*
160
Ucatangeri
4
80
2
0
i
/ffa nucleus
i i i i i i
r
Pachygrapsus marmoratus
3
2
1
0
592
i r
J A S O N D J FMAMJ
1991
1992
i
I i i i i
r
I
I
I
JASONDJFMAMJ
1991
1992
T
Seasonal and tidal abundance patterns of decapod crustacean larvae in a shallow inlet
Table n. Percentage of total variance corresponding to each source of variation for the zoeal
abundance [ln(l + no. zoeae)] of the most abundant species in each 24 h series. Season (top factor)
and lunar phase (nested factor) are categorical variables in two-way nested ANOVAs, and replicate
is error component
Species
Sources of variation
Replicate
Zoeal
Hippolyte spp.
Processa spp.
Upogebia deltaura
Diogenes pugilator
Pisidia sp.
Ilia nucleus
Carcinus maenas
Liocarcinus spp.
Panopeus africanus
Pinnotheres pinnotheres
Uca tangeri
Pachygrapsus marmoratus
Older larval stages
Ilia nucleus
Palaemon spp.
78.8
59.6
27.4
76.2
95.7
Lunar phase
0.0
31.3
23.8
0.0
0.0
4.3
10.0
44.5
2.7
7.7
13.1
9.1
72.6**
12.6
10.3
30.0
21.2
0.0
9.1
8.9
9.1
Season
23.9
13.3
12.1
10.9
0.0
92.2**
33.4
78.3*
87.3"
31.5
11.9*
78.9*
78.8*
70.0*
0.1
Wi < Au < Sp < Su
Au < Wi < Su < SP
Su < Au < Sp < Wi
Wi < Au < Su<Sp
Wi < Su < Au < Sp
Au<Wi < Sp < Su
Wi < Au < Su < Sp
Npt < Spt
53.5
Su, summer; Au, autumn; Wi, winter; Sp, spring; Npt, neap tide; Spt, spring tide.
Significance: *P < 0.05; **P < 0.01. Mean zoeal densities under the conditions joined by an underline
were not significantly different at the 5% level using Student-Newman-Keuls test.
tides (Figure 5) of autumn (mainly Palaemon spp. zoea V-VI) and spring
(primarily I.nucleus zoea II—III).
A significant (P < 0.05) positive correlation between monthly first zoea abundance and water temperature was observed for eight (Processa spp., Upogebia
deltaura, Diogenes pugilator, I.nucleus, Panopeus africanus, Pinnotheres
pinnotheres and Uca tangeri) of the 12 most abundant species. A negative correlation occurred in only one species (C.maenas). Conversely, no significant correlation between monthly larval abundance of each species and the other
hydrological variables (water salinity and current speed) was observed.
Diet and tidal patterns of larval abundance
The abundance of first zoeae of the 12 most abundant species did not show statistically significant differences depending on lunar phase (spring and neap tides)
(Table II). However, the short-term (between consecutive or alternate days)
and/or long-term (between seasons) variation of zoeal density (between 24 h
Fig. 3. Monthly zoea I flux for the 12 most abundant decapod crustacean species in the Rio San Pedro
inlet. For each month, bars on the left and right of the mark correspond to the flood and ebb values,
respectively; the shaded part of each bar represents individuals that were collected close to the
bottom. For each month, an asterisk means zoea I density that was significantly higher (x2 test; P <
0.05).
593
RDrake, AJVLArias and AJlodn'guez
Hippotyte spp.
Cardnus maenas
160
500
80
250
0
0
Processaspp.
200
100
0
I
f
Lkxardnus spp.
i
L
1200
600
0
Upogebia dettauna
Panopeus africanus
900
400
450
200
0
0
CM
N
e
Diogenes pugUator
Pinnotheres pinnotheres
300
150
0
1200
600
0
18a nucleus
Pachygrapsus marmoratus
120
200
60
100
0
0
Su
Au
r
I
Wi
Sp
Su
Au
Wi
i
Sp
Fig. 4. Number of zoea I for the 12 most abundant decapod crustacean species in the Rio San Pedro
inlet during each 24 h series (160 nr 3 ). For each season, empty and shaded bars correspond to the
spring and neap tides, respectively. Su, summer; Au, autumn; Wi, winter; Sp, spring.
594
Seasonal and tidal abundance patterns of decapod crustacean larvae in a shallow inlet
Total
120
60
0
Ma nucleus
10
s °
N
hi
50
o
Palaemon spp.
80 40 0
i
Su
r
Au
i
Wi
-IVSp
Fig. 5. Number of later zoeae (>zoea I) for all the species together (top) and for 1.nucleus and Palaemon spp. in the Rio San Pedro inlet during each 24 h series (160 nr 3 ). For each season, empty and
shaded bars correspond to the spring and neap tides, respectively. Su, summer; Au, autumn; Wi,
winter; Sp, spring.
series) for each species was considerable (Table II; Figure 4). Thus, the relative
abundance (percentage of total catch for each species in each 24 h series) was
used to analyse the effect of diel and tidal cycles on the zoea I abundance and
vertical position in the water column. Nevertheless, the lunar phase was introduced in ANOVA tests to estimate interaction effects because it is known that
some species may change the diel/tidal pattern of abundance depending on lunar
phases (Rodriguez et al, 1997).
Results of four-way ANOVA tests on the relative abundance of zoea I for the
most abundant species indicated that the effects of different diel, tidal or lunar
phases on larval density, as well as the larval vertical position in the water column,
were significant at least for one of these factors, with the exception of C.maenas
(Table III). Zoea I of Processa spp., Uca tangeri and Pachygrapsus marmoratus
were more abundant during ebb tide, and those of Panopeus africanus at high and
ebb tides, suggesting that they are released in the inlet and exported to the bay.
In contrast, first zoeae of D.pugilator, Pisidia sp., Lnucleus and Pinnotheres
pinnotheres were more abundant during high and flood tides than at ebb and low
tides, suggesting that they are imported into the inlet or just released and retained
there. In addition, zoea I of Upogebia deltaura, Pisidia sp., Liocarcinus spp. and
Panopeus africanus were more abundant during night than during the day, while
Pinnotheres pinnotheres showed the opposite diel pattern. In relation to the
595
P.Drake, A.M.Arias and A-Rodriguez
vertical position in the water column, Pisidia sp. and Panopeus africanus first
zoeae were mainly caught at the surface. On the other hand, for only five of the
12 most abundant species were one or more second-order interaction effects
statistically significant (Table III), the diel versus lunar phase being significant for
three of them. In fact, Processa spp. and Panopeus africanus larvae showed a
higher density during the night at neap tides, while Pinnotheres pinnotheres zoeae
were more abundant during the day at spring tides.
Because of the low abundance of later larval stages (>zoea I), their diel and
tidal patterns were analysed for only the two species (I. nucleus and Palaemon
spp.) whose later zoeal stages were the most abundant (Table I). Zoea II-III of
I. nucleus were significantly more abundant at flood and high tides and during the
day (Table III). No significant (P > 0.05) differences in the abundance of zoea
V-VI of Palaemon spp. were found in relation to diel and tidal cycles.
Discussion
Monthly and 24 h series samples showed a coincident pattern of first zoea export
from the Rio San Pedro inlet for four of the 12 most abundant species (Processa
spp., Panopeus africanus, Uca tangeri and Pachygrapsus marmoratus). Monthly
samples also suggested that first zoeae were exported for two other species
(Cmaenas and Liocarcinus spp.). All these species occurred as adults in the study
inlet (some of them also in the Bay of C&diz), but their later larval stages (>zoea
I) were rarely or never collected (Table I). These features indicated that these
species spawned in the study system and there was an export of their first zoeae
from the inlet to the bay. A rapid export of the first zoeae out of estuaries has
previously been reported for Cmaenas (Queiroga et al., 1994), Uca tangeri and
Pachygrapsus marmoratus (Paula, 1993), and other estuarine crab species (Dittel
and Epifanio, 1990; Dittel et al., 1991), but a retention of Panopeus africanus
larvae within the Mira Estuary was found by Paula (1993). Conversely, for two
of the most abundant species (D.pugilator and Pinnotheres pinnotheres), there
was an input of first zoeae into the inlet according to both monthly and 24 h series
samples. In addition, either monthly or 24 h series samples suggested that the first
zoeal stage was imported into the inlet for three other species (Upogebia deltaura,
at nocturnal spring tides; Pisidia sp. and I.nucleus, in 24 h series). Adult individuals of these species always occurred in the bay (some of them also in the study
inlet) and older larval stages (>zoea I) were rarely or never collected in the inlet,
except for the crab I.nucleus (Table I). Then, the input of zoea I of such species
did not seem to respond to a real mechanism of larval retention within the inlet.
That larval input could be the result of zoea I released by adult populations
inhabiting the inner part of the bay which drifted into the inlet (by tidal currents)
when they were being exported to the open sea (Figure 1). In fact, the bathymetric characteristics of the inlet studied classify it as a shallow flood-dominated
system, where the hydrodynamic regime could facilitate the passive penetration
of planktonic larvae (Drake and Arias, 1991). Thus, the results of this study
suggest that the Rio San Pedro inlet is used by decapod crustacean species
primary as an adult habitat and spawning area, with larval development taking
596
ns
ns
H<L<F<E
H<F<L<E
L<LE<H<F
ns
Uca tangeri
Pachygrapsus marmoratus
Older larval stages
Ilia nucleus
Palaemon spp.
ns
ns
ns
ns
ns
ns
ns
ns
B<S
ns
ns
ns
B<S
ns
Position
ns
ns
Diel x Lunar phase (DNp < DSp < NSp < NNp)
Tide X Diel (EN < LD < LN < ED < HN < FD < FN < HD^
Tide x Position (EB < LB < LS < ES < FS < HS < HB < FB)
Diel x Lunar phase (NSp < DNp < NNp < DSp)
Position x Lunar phase (BSp < SNp < SSp < BNp)
ns
ns
Tide x Diel (LN < LD < ED < H P <EN < FD < F N < H W
Diel x Position (NB < DB < P S < NS)
Diel x Lunar phase f DNp < NSp < DSp < NNp)
Interaction
E, ebb; F, flood; H, high tide; L, low tide; D, daytime; N, night-time; S, surface of the water column; B, bottom; Sp, spring tide; Np, neap tide.
Significance of Student-Newman-Keuls tests: ns, P > 0.05.
N<D
ns
ns
ns
D<N
ns
D<N
ns
ns
D<N
D<N
N<D
Diel
F<L<E<H
L_sUtLl£<E
L<F<E<H
L<E<F<H
L<E<H<F
L<E<F<H
ns
L<E<F<H
L<F<H<E
L<E<H<F
Tide
Main effects
Zoea I
Hippolyte spp.
Processa spp.
Upogebia deltaura
Diogenes pugilalor
Pisidia sp.
Ilia nucleus
Carcinus maenas
Liocarcinus spp.
Panopeus africanus
Pinnotheres pinnotheres
Species
Table III. Results of four-way ANOVAs of the effects of tidal, diel and lunar phases, and of vertical position on the relative abundance of zoeae in each 24 h
series for the most abundant species. The analysis was performed on arcsineVy-transformed data, where y is the percentage of zoeae that, from the total number
of zoeae in each 24 h series, were caught at the considered situation. The lunar phase was only considered to estimate interaction effects
P.Drake, AJVlArias and A.Rodriguez
place in the sea. Only I.nucleus seemed to use the inlet to complete its larval
development, but adults of this species were not observed in the inlet and there
was an input from the bay to the inlet of all zoeal stages (zoea I, II and III),
suggesting that the species completes its life cycle (larval development included)
within the Bay of Cadiz.
With the exception of I.nucleus, of the 12 most abundant species all those that
showed non-coincident exportation/importation patterns for monthly and 24 h
series samples (Cmaenas, Liocarcinus spp., Upogebia deltaura and Pisidia sp.)
had adult populations within and outside the inlet (Table I). Unless there was a
perfect synchronous larval releasing activity at both populations, such unclear
exportation/importation patterns may be due to a mixture of larvae from both
populations. Larval rhythms masked in estuarine systems because of a mixture of
larvae from different populations have previously been suggested at the Mira
Estuary (Paula, 1993). On the other hand, monthly samples were regularly spread
throughout the year, but they were taken only during nocturnal flood/ebb tides.
Since most dominant species in samples were significantly more abundant during
the night (Table III), monthly samples may be considered indicative of the net
flux at the sampling date (spring tide). In the case of 24 h series, samples represented the daily flux of larvae in the inlet more exactly, but they were collected
during shorter periods (7-9 days per season). Furthermore, because of the high
short-term and long-term variation in larval density, results of 24 h series have
been expressed as relative abundance. After such data transformation, all the
cycles had the same weight independently of real abundance. The latter feature
may have considerable consequences for species with a short period of abundance, such as C.maenas, because only one period of 24 h cycles (winter) was coincident with maximal releasing activity of the species. Changes in larval release
rhythms have been observed in C.maenas when their breeding season had not
reached its full peak (Queiroga et ai, 1994). Then both sampling strategies
showed different limitations and supplied complementary information.
To be exported from the study inlet, where there was a residual transport
pattern directed towards its inner part (Drake and Arias, 1991), (i) larvae should
stay near the surface during ebb tides and close to the bottom during floods (vertical migration) and/or (ii) females may synchronize the release of larvae with the
most suitable tidal phases (high and ebb tides). In the present study, the tidal
phase-vertical position interaction effect was not significant for the most
abundant species (except for the first zoea of Pinnotheres pinnotheres), indicating the absence of a tidal vertical migration. Then, as previously suggested for
Panopeus africanus and Uca tangeri (Rodriguez et ai, 1997), the tidal synchronization of female release of larvae seems to be a more probable cause of
exportation of larvae from the studied inlet. The observed patterns of first zoea
abundance of Processa spp., Uca tangeri and Pachygrapsus marmoratus, significantly more abundant at ebb tides, and for Panopeus africanus, greater during
high and ebb tides, clearly favour a passive larval exportation from the inlet to
the bay for these species.
Avoidance of stressful water conditions of estuarine areas and reduction of fish
predation have been claimed to explain larval exportation (Forward, 1987).
598
Seasonal and tidal abundance patterns of decapod crustacean larvae in a shallow inlet
Larvae of species living in estuarine zones could be expected to be well adapted
physiologically to tolerate changes in water temperature and salinity, but such
species often show considerable changes in their survival rates depending on
environmental conditions (Nagaraj, 1993; Paula, 1993). In a typical estuarine
system, larvae have to minimize exposure to lethal combinations of high temperatures and low salinities (Christy, 1982; Forward, 1987). In the study inlet, with
hypersaline water conditions during most of the year (Figure 2), larvae should
avoid exposure to high temperatures and salinities that may cause periods of
hypoxia. This fact could explain the different larval patterns observed for
Panopeus africanus at the Mira Estuary, where zoeae were retained (Paula, 1989,
1993) and, in this study, where there was a first zoea export for the species. On
the other hand, the only species that seemed to complete its larval development
within the neighbouring bay was I.nucleus, a species with a relatively short larval
period (three zoeae) and with a peculiar disposition of zoeal spines that makes
them less vulnerable to planktivorous fish (Morgan, 1987,1989).
The studied population of Panopeus africanus showed a higher first zoea abundance generally near the time of high tide and often at night, which is common
among brachyurans living in tidal areas (DeCoursey, 1983; Forward, 1987; De
Vries and Forward, 1991). However, there were no significant differences
between daytime and night-time abundance of thefirstzoea oiProcessa spp., Uca
tangeri and Pachygrapsus marmoratus in this study, and maximal density was
mainly observed at ebb tides (Table III). Peaks of larval release for intertidal
crabs have often been correlated with the moon phases and local tidal/diel cycles
(Forward, 1987). Furthermore, a hierarchy of rhythms regulating reproductive
timing has been proposed recently to explain differences observed for populations in different tidal regimes (Morgan, 1996). Under the latter assumption,
zoeal rhythms should be similar at Mira Estuary (37°40'N, 8°40'W) and in the
study inlet (36°23'-37'N, 6°8'-15'W), which showed very close lunar/diel/tidal
cycles and tidal amplitudes (1-3 m at Mira Estuary; 1-3.5 m in the study inlet).
In the Mira Estuary, most species showed a semi-lunar rhythm of larval release,
centred on crepuscular high tides around the quarter moons (Paula, 1989). The
results of this study suggest that there are no clear differences in larval release
between spring and neap tides, so they do not seem to support the hypothesis
suggested by Paula (1989) that any crepuscular high tide should be suitable for
larval release. Nevertheless, larvae of several species were often more abundant
at neap or spring tides (Figure 4), but the considerable short-term variation in
larval abundance caused the differences observed to be statistically not significant (Table II). Consequently, longer sampling periods will be needed before we
can be more conclusive concerning this topic.
The study inlet seemed to be used by most decapod crustaceans as a spawning
area, while larval development occurred in open sea. However, megalops (the
stage expected to reinvade the inlet) were scarcely found for most species. As
previously suggested for Panopeus africanus and Uca tangeri (Rodriguez et al.,
1997), a possible hypothesis to explain the low abundance of megalops observed
in this study is that, due to the situation of the study inlet within the Bay of Cadiz
(Figure 1), most megalops immigrating from the sea settle on the mud flats which
599
P.Drake, A.M.Arias and A.Rodriguez
surround the mouth of the inlet, and the populations studied are restocked mainly
by immigration of juveniles or adults from outside the inlet. Another possible
explanation could be that the sampling method used (pumping) was unsuitable
for catching advanced larval stages, although this seems very unlikely because this
method has caught megalopae successfully in previous studies (Dittel and
Epifanio, 1990; Queiroga et al., 1994). Nevertheless, Wehrtmann and Dittel
(1990) found decapod megalops and juveniles which were attached to drifting
leaves, using them as a transport mechanism to reinvade estuaries. Larval and
juvenile decapod crustaceans might have been attached to floating macroalgae in
our study inlet, which would have reduced the probability of being caught.
Acknowledgements
This research was supported financially by Junta de Andalucia and Consejo
Superior de Investigaciones Cientificas. We are grateful to M.Espigares and
S.Gonzdlez for their invaluable help in sorting samples, and to I.L6pez de La
Rosa and I.Gonzdlez for supplying information about decapod crustacean distribution within the Bay of Cadiz.
References
Arias.A.M. (1976) Contribution al conocimiento de la fauna bentdnica de la Bahia de Cadiz. Invest.
Pesq., 40,355-386.
Christy,J.H. (1982) Adaptive significance of semilunar cycles of larval release in fiddler crabs (Genus
Uca): test of an hypothesis. Biol. Bull., 163,251-263.
Cronin.T.W. (1982) Estuarine retention of larvae of the crab Rhithropanopeus harrisii. Estuarine
Coastal Shelf ScL, 15, 207-220.
Cronin.T.W. and Forward.R.B. (1979) Tidal vertical migration: an endogenous rhythm in estuarine
crab larvae. Science, 205,1020-1022.
DeCoursey,PJ. (1983) Biological timing. In BlissJD., Vernberg,EJ. and Vernberg.W.B. (eds), The
Biology of Crustacea. Academic Press, New York, Vol. 7, pp. 107-162.
De Vries,M.C. and Forward.R.B.Jr (1989) Rhythms in larval release of the sublittoral crab
Neopanope sayi and the supralittoral crab Sesarma cinereum (Decapoda: Brachyura). Mar. Biol.,
100,241-248.
De Vries,M.C. and Forward.R.B.Jr (1991) Control of egg-hatching in crabs from different tidal
heights. /. Crust. Biol., 11,29-39.
Dittel,A.I. and Epifanio.C.E. (1990) Seasonal and tidal abundance of crab larvae in a tropical
mangrove system, Gulf of Nicoya, Costa Rica. Mar. Ecol. Prog. Ser, 65,25-34.
Dittel,A.L, Epifanio.C.E. and Lizano.O. (1991) Flux of crab larvae in a mangrove creek in the Gulf
of Nicoya, Costa Rica. Estuarine Coastal Shelf ScL, 32,129-140.
Drake.P. and Arias.A.M. (1991) Ichthyoplankton of a shallow coastal inlet in south-west Spain:
factors contributing to colonization and retention. Estuarine Coastal Shelf ScL, 32,347-364.
Drake,P., Arias^A.M. and Conradi,M. (1997) Aportacibn al conocimiento de la macrofauna supra y
epibent6nica de los canos mareales de la Bahia de Cadiz (Espana). BoL Inst. Esp. Ocean., 23,
133-141.
ForwardJR.B.Jr (1987) Larval release rhythms of decapod crustaceans: an overview. Bull Mar. ScL,
41, 165-176.
Fuste'.X. (1989) Distribution of decapod crustacean larvae in the bays of the Ebro Delta. ScL Mar.,
53.763-770.
Johnson.G.E. and GonorJJ. (1982) The tidal exchange of Callianassa californiensis (Crustacea,
Decapoda) larvae between the ocean and the Salmon River Estuary, Oregon. Estuarine Coastal
Shelf ScL, 14, 501-516.
Little,K.T. and Epifanio.C.E. (1991) Mechanism for the re-invasion of an estuary by two species of
brachyuran megalopae. Mar. EcoL Prog. Ser., 68, 35-242.
600
Seasonal and tidal abundance patterns of decapod crustacean larvae in a shallow inlet
Morgan,S.G. (1987) Morphological and behavioral antipredatory adaptations of decapod zoeae.
Oecologia (Berlin), 73, 391-400.
Morgan.S.G. (1989) Adaptive significance of spination in estuarine crab zoeae. Ecology, 70,464-482.
Morgan.S.G. (1990) Impact of planktivorous fishes on dispersal, hatching, and morphology of
estuarine crab larvae. Ecology, 71,1639-1652.
Morgan.S.G. (1996) Plasticity in reproductive timing by crabs in adjacent tidal regimes. Mar. Ecol.
Prog. Ser., 139,105-118.
Morgan.S.G. and ChristyJ.H. (1995) Adaptive significance of the timing of larval release by crabs.
Am. Nat., 145,457-479.
Muller,K. (1983) Untersuchungen zur Populationsbiologie, Aktivitatsrhythmik und geographischen
Verbreitung von Uca tangeri (Decapoda, Brachyura). Zool. Jb. Syst., 110,221-266.
MunozJ.L. and Sanchez-Lamadrid,A. (1994) El medio fisico y biologico en la Bahia de Cadiz: Saco
interior. Informaciones T^cnicas, Consejeria de Agricultura y Pesca. Junta de Andalucia, Sevilla,
161 pp.
Nagaraj,M. (1993) Combined effects of temperature and salinity on the zoeal development of the
green crab, Carcinus maenas (Linnaeus, 1758) (Decapoda: Portunidae). Sci. Mar, 57,1-8.
ParradoJ.M., Gutierrez MasJ.M. and Achab,M. (1996) Determinacidn de direcciones de corrientes
mediante el analisis de 'formas de fondo' en la Bahia de Cadiz. Geogaceta, 20, 378-381.
Paula J. (1989) Rhythms of larval release of decapod crustaceans in the Mira Estuary, Portugal. Mar.
Biol., 100, 309-312.
Paulaj. (1993) Ecologia da fase larvar e recrutamento de crustaceos decapodes no estuario do rio
Mira. PhD Thesis, University of Lisbon, Lisbon.
Queiroga,H., CostlowJ.D. and Moreira.M.H. (1994) Larval abundance patterns of Carcinus maenas
(Decapoda, Brachyura) in Canal de Mira (Ria de Aveiro, Portugal). Mar. Ecol. Prog. Ser., Ill,
63-72.
Queiroga.H., Costlow, J.D. and Moreira,M.H. (1997) Vertical migration of the crab Carcinus maenas
first zoea in an estuary: implications for tidal stream transport. Mar. Ecol. Prog. Ser., 149,121-132.
Rodriguez,A., Drake,P. and Arias.A.M. (1997) Reproductive periods and larval abundance patterns
of crabs Panopeus africanus and Uca tangeri in a shallow inlet (SW Spain). Mar. Ecol. Prog. Ser.,
149,133-142.
Sandifer,P.A. (1975) The role of pelagic larvae in recruitment to populations of adult decapod crustaceans in the York River Estuary and adjacent lower Chesapeake Bay, Virginia. Estuarine Coastal
Mar. Sci., 3, 269-279.
Tankersley.R.A., McKelvey.L.M. and Forward,R.B.,Jr (1995) Responses of estuarine crab megalopae
to pressure, salinity and light: implications for flood-tide transport. Mar. Biol., 122, 391-400.
Wehrtmann.I.S. and Dittel,A.I. (1990) Utilization of floating mangrove leaves as a transport
mechanism of estuarine organisms, with emphasis on decapod crustacean. Mar. Ecol. Prog. Ser., 60,
67-73.
Zeng,C. and Naylor,E. (1996) Endogenous tidal rhythms of vertical migration infieldcollected zoea1 larvae of the shore Carcinus maenas: implications for ebb tide offshore dispersal. Mar. Ecol. Prog.
Ser., 132, 71-82.
Received on July 29, 1997; accepted on November 25, 1997
601