Burrowing Abilities and Swash Behavior of Three

Journal of Experimental Marine Biology and Ecology 255 (2000) 229±245 www.elsevier.nl/locate/jembe

Burrowing abilities and swash behavior of three crabs, Emerita analoga Stimpson, Blepharipoda occidentalis Randall, and Lepidopa californica Efford (Anomura, Hippoidea), of exposed sandy beaches

J.E. Dugana,* , D.M. Hubbard a , M. Lastra b aMarine Science Institute, University of California, Santa Barbara, CA 93106, USA bDepartamento de Ecologia y Biologia Animal, Universidad de Vigo, Vigo, Spain Received 15 May 2000; received in revised form 12 September 2000; accepted 19 September 2000

Abstract

To investigate factors related to the distribution of intertidal species, and speci®c predictions of the swash exclusion hypothesis for exposed sandy beaches, we compared the burrowing abilities and swash behavior of three species of anomuran crabs in the superfamily Hippoidea (Emerita analoga, Blepharipoda occidentalis and Lepidopa californica) which commonly inhabit the intertidal and shallow subtidal zones of beaches along the California coast. Burrowing times in the laboratory increased signi®cantly with crab size for all species in ®ve sediment grain sizes ranging from ®ne sand to gravel (0.15 to 3.24 mm). For each species, burrowing times differed signi®cantly among sand grain sizes, ranging from 0.3 to 21.5 s. Burrowing times for the hippid crab, E. analoga, were relatively constant across sediment types, while those of the albuneid crabs, B. occidentalis and L. californica, were rapid in ®ne to medium sands, and much slower in coarser sediments. Our results indicate that E. analoga is a substrate generalist while L. californica and B. occidentalis are substrate sensitive. Pre-burrowing times and behavior, distance moved, and burrowing times differed among the species in the swash zone. Combined times of preburrowing and burrowing were shorter than the swash period (6 s) for most E. analoga individuals. Fifty percent of the individuals of L. californica reached the substrate and burrowed in the swash period, while no individuals of B. occidentalis burrowed in that time. Pre-burrowing behavior and time may be valuable in explaining spatial and temporal patterns in the distribution of hippoid crabs on California beaches. Our results support predictions of the swash exclusion hypothesis concerning the burrowing and locomotory abilities of sandy beach macrofauna. The substrate generalist characteristics, and unique orientation and swimming abilities of the hippid crab, E. analoga, in intertidal swash may help explain the success of this species and its congeners, and have important implications for understanding patterns of macrofauna community structure on

*Corresponding author. Tel.: 11-805-893-2675; fax: 11-805-893-8062. E-mail address: j [email protected] (J.E. Dugan). ] 0022-0981/00/$ ± see front matter  2000 Elsevier Science B.V. All rights reserved. PII: S0022-0981(00)00294-X 230 J.E. Dugan et al. / J. Exp. Mar. Biol. Ecol. 255 (2000) 229 ±245 exposed sandy beaches in California and other regions.  2000 Elsevier Science B.V. All rights reserved.

Keywords: Albuneidae; Hippidae; Beach morphodynamics; Sandy beach; Swash zone

1. Introduction

Exposed sandy beaches are composed of unconsolidated sediments subject to constant movement by waves and represent a challenging and physically unstable habitat for intertidal organisms. Macrofauna species which inhabit exposed sandy beaches generally exhibit behaviorial and morphological adaptations which allow them to withstand the harsh physical conditions. The ability to burrow rapidly in disturbed sediments of different grain sizes and to orient in surging swash are central to the success of individual species and ultimately to the composition of intertidal macrofauna communities on exposed sandy beaches. The structure of intertidal macrofauna communities of exposed sandy beaches is believed to be controlled largely by physical processes such as wave and sediment dynamics (McLachlan, 1990). In general, individual physical factors, such as beach slope or sand particle size, have not been broadly successful in predicting macrofaunal community structure. However, a dimensionless index (Dean's parameter, e.g. Short, 1996) which incorporates wave height, wave period and sediment fall velocity to describe beach morphodynamic state has been useful in predicting community structure (e.g., McLachlan, 1990, 1996). A number of studies of intertidal macrofauna communities have found that species richness increases linearly and abundance and biomass increase exponentially across a continuum from re¯ective to dissipative beaches as values of Dean's increase (McLachlan, 1990; Jaramillo and McLachlan, 1993; McLach- lan et al., 1993, 1995, 1996, 1998; Hacking, 1998). Intertidal macrofauna experience the wave regime of a particular beach type most directly as swash. The swash zone is the intertidal area landward of the surf zone that is alternately submerged and exposed by wave wash or swash. Swash climates are generally closely related to the morphodynamic state of the beach (McArdle and McLachlan, 1991, 1992). Long period swashes characteristically occur on ¯at, dissipative beaches while short period swashes occur on steep, re¯ective beaches (McArdle and McLachlan, 1991, 1992). McLachlan et al. (1993) proposed the `swash exclusion' hypothesis as a possible explanation for the observed patterns of decreasing richness, biomass and abundance of macrofauna from dissipative to re¯ective beaches. That hypothesis states that swash climate is a key factor in¯uencing the intertidal macrofauna (McLachlan et al., 1993, 1995). Across the continuum from dissipative to re¯ective beaches, swash climate changes from a relatively benign regime which can support almost all available macrofauna species to an increasingly inhospitable regime which excludes less robust species, until all but the supralittoral forms may be excluded on re¯ective beaches. This dynamic is proposed to result in decreasing values of species richness, abundance and biomass of macrofauna communities with increasingly re¯ective beach state (McLachlan et al., 1993, 1995). J.E. Dugan et al. / J. Exp. Mar. Biol. Ecol. 255 (2000) 229 ±245 231

Direct tests of the swash exclusion hypothesis are problematic but some predictions of the hypothesis can be examined directly. One prediction is that burrowing and locomotory ability could determine which species may inhabit the swash zone in sandy beaches of different morphodynamic types thus contributing to the community patterns described above (e.g., McLachlan et al., 1995). For example, species which are rapid burrowers may be able to successfully inhabit a wider range of beach morphodynamic types than species which burrow slowly. Slow burrowing times relative to the swash period likely expose animals to higher swash velocities and additional turbulence from multiple swashes. Exposure to multiple swashes could physically dislodge and disorient animals leading to stranding, transport to the surf or impact zone, and lateral transport by longshore currents. Sediment grain size may also directly limit burrowing for some macrofauna species (Alexander et al., 1993; Nel et al., 1999). In general, re¯ective beaches tend to have coarser sediments and steeper slopes (Short, 1996) along with a harsher swash climate. Unfavorable sediment grain sizes in combination with harsh swash climates could increase such limitations (McLachlan, 1996). The anomuran crabs of the super family Hippoidea are important components of the macrofauna communities of exposed tropical and temperate sandy beaches (Efford, 1976; Trueman, 1970; Trueman and Ansell, 1969; Haley, 1982; Dugan et al., 1995). Three species of hippoid crabs occur on exposed sandy beaches along the California coast. The common sand crab, Emerita analoga (Hippidae) is a suspension-feeding tidal migrant which often dominates the abundance and biomass of the intertidal macrofauna of a wide range of types of exposed sandy beaches (Dugan et al., 1995, 2000). Highest densities of this species occur in the active swash zone. Two albuneid crabs, the spiny sand crab, Blepharipoda occidentalis, and the porcelain sand crab, Lepidopa californica, are less abundant and less widely distributed, occurring in the low intertidal zone and in sandy sublittoral habitats (Fager, 1968; Morris et al., 1980; Morin et al.,1985). The three species co-occur intertidally on a number of intermediate to dissipative type beaches in California (Dugan et al., 2000) We hypothesized that the burrowing abilities and swash behavior of the three species of hippoid crabs could vary, and that differences in distribution of the species among beaches of different morphodynamic types and, ultimately, macrofaunal community structure may be related to that variation. To test some of the predictions of the swash exclusion hypothesis, we compared burrowing rates of the three species in ®ve sediment sizes in the laboratory, and investigated pre-burrowing behavior and burrowing of the three species in the swash zone of a ®ne sand, re¯ective beach.

2. Methods

We collected a range of sizes of three species of hippoid crabs (Emerita analoga, Blepharipoda occidentalis and Lepidopa californica) from the intertidal zone of Pismo Beach, a dissipative sandy beach in central California, and at Santa Claus Lane Beach, an intermediate beach in southern California (Dean's parameter: 6.1 and 3.2, respectively), during July 1997. Recently molted and late pre-molt stage crabs were not used in the 232 J.E. Dugan et al. / J. Exp. Mar. Biol. Ecol. 255 (2000) 229 ±245 experiments. Crabs were maintained in clean sand and ¯owing, ®ltered seawater in the marine laboratory at the University of California, Santa Barbara. We used ®ve grades of intertidal beach sand in the burrowing trials: very ®ne, ®ne, medium, coarse and very coarse (Table 1). Those sand grades consisted of (1) sand from Pismo Beach and (2) four grades made up of sieved fractions of sand from Goleta Beach (an intermediate beach, Dean's parameter: 2.5). Samples of sediments (50 ml) were rinsed with distilled water, dried, sieved (0.5 Phi interval), and weighed to determine mean grain size and sorting. Burrowing times of crabs were measured in the laboratory ina4lbucket (205 mm in diameter) with 100 mm of sand and 50 mm of water column. Seawater temperatures during the burrowing trials were between 18 and 208C. Individual crabs were timed with a stopwatch from the initiation of burrowing (abdomen in contact with sediment, uropods active, penetration of the substrate by fourth pereopods) to the disappearance of the carapace under the sediment surface or the end of digging behavior (respiratory tube formed, antennae still). Crabs were removed from the sand immediately after burrowing. We measured the carapace length (CL, 0.1 mm), and blotted wet mass (0.01 g) of each crab. We also recorded the sex of each crab and reproductive state of female crabs (ovigerous or non-ovigerous). We calculated OLS regressions for burrowing time (s), and (1) CL (mm) and (2) wet mass (g) for each species. For each species, we compared the slopes and elevations of those regressions among the ®ve sediment sizes with ANCOVA. We compared the burrowing times of the three species in the ®ve sediment sizes using a two-way ANOVA with mass as a covariate. We calculated the burrowing rate index (BRI, see Stanley, 1970; Brown and Trueman, 1994; McLachlan et al., 1995) for each crab in each sediment size as

BRI 5 (Wet mass (g)0.33 /Burrowing time (s))*100

We examined the ability of BRI to remove the effect of crab size for interspeci®c comparisons of burrowing ability by determining the signi®cance of correlations of BRI with CL. We investigated the pre-burrowing and burrowing behavior of Emerita analoga, Blepharipoda occidentalis and Lepidopa californica in active swash in the intertidal zone of an east-facing beach protected from prevailing NW wind and swell near the

Table 1 Characteristics of the ®ve sediment grades used in the burrowing trials Sediment grade Mean grain size Sorting phi mm phi Very ®ne 2.73 0.15 0.27 Fine 2.46 0.18 0.38 Medium 0.87 0.55 0.48 Coarse 20.39 1.31 0.65 Very coarse 21.69 3.24 0.26 J.E. Dugan et al. / J. Exp. Mar. Biol. Ecol. 255 (2000) 229 ±245 233 marine laboratory of the University of California at Santa Barbara on July 28, 29, and 30, 1997. The study site was a modally re¯ective beach (Dean's parameter ,1.0) with narrow surf and swash zones, small breaker heights, a steep beach face, and prevailing longshore currents to the NE. Mean grain size was 0.28 mm, a value falling between the ®ne and medium size grades used in the laboratory trials. Sediments were not well sorted (sorting, I.G.S.D.51.28) compared to the sand grades used in the laboratory burrowing trials (Table 1). Swash zone dynamics and the slope of the beach at the ef¯uent line during the experiments were typical of conditions measured monthly at the site in the previous year (swash zone width, 5 to 10 m; period, 4 to 8 s; slope, 3 to 58) (Dugan and Hubbard, unpublished data). To observe the behavior of crabs in active swash, we marked the upper and lower limits of the swash zone and the release point for the crabs with PVC stakes. Release points were located 2 to 3 m above the lower limit of swashes. We adjusted the positions of the stakes approximately every 20 min to account for changes in tidal level. Crabs were released individually into slack water of 5 to 30 cm depth at the time of swash maximum. We measured several components of the responses of individuals released in the swash zone including: time elapsed between release and the initiation of burrowing, distance moved to a burrowing location, and burrowing times. The time between release and the start of burrowing, and the time to completion of burrowing was measured with a stopwatch. We measured the distance between release point and the location of burrowing in two components: perpendicular and parallel to shore. We also recorded behavior of each crab from the release point to the burrowing position as: (1) swimming in the water column, (2) drifting passively in the water column, (3) sliding in contact with the substrate, (4) tumbling across the substrate, and (5) orienting, pivoting to a head down current position in contact with the sediments.

3. Results

The sizes (CL) of individuals used in burrowing trials formed a continuous size distribution with some overlap in size between the three species: Blepharipoda occidentalis ranged from 25.3 to 65.5 mm CL, Lepidopa californica from 10.4 to 19.5 mm CL, and Emerita analoga from 5.7 to 33.9 mm CL (Table 2). The smallest animals tested in the laboratory were recently recruited E. analoga which were approximately the size of the coarsest sediment used in the laboratory burrowing trials. In the very coarse sediments, those small individuals burrowed successfully only when they initiated burrowing in the gaps between sediment particles. The fastest burrowing time observed in the laboratory was 0.3 s for Lepidopa californica in very ®ne sediments (Table 2). The slowest burrowing time observed in the laboratory was 21.5 s for Blepharipoda occidentalis in very coarse sediments. Burrowing time increased signi®cantly with carapace length and mass in each of the 15 treatments (three species in ®ve sediment sizes) (Table 3). No pattern of decreasing or increasing slope of that relationship was evident with sediment grade in Emerita analoga or Blepharipoda occidentalis. The slope of the relationships between burrowing 234 J.E. Dugan et al. / J. Exp. Mar. Biol. Ecol. 255 (2000) 229 ±245

Table 2 Range of crab sizes and burrowing times in ®ve sediment grades in the laboratory and in the ®eld trials Species Size range Burrowing time range (s) CL (mm) Very ®ne Fine Medium Coarse Very coarse Emerita analoga Laboratory 5.7±33.9 1.7±7.2 1.1±8.2 1.0±4.5 1.2±5.6 0.7±7.5 Field 7.3±30.5 1.3±13.4 Lepidopa californica Laboratory 10.4±19.5 0.3±2.7 0.6±2.9 0.6±2.5 1.7±5.7 2.8±15.3 Field 11.3±19.7 1.4±3.6 Blepharipoda occidentalis Laboratory 25.3±65.5 1.4±11.4 1.0±9.9 2.1±8.8 3.4±13.5 9.2±21.5 Field 37.9±62.2 3.5±36.0 time, and carapace length and mass increased with increasing grain size for Lepidopa californica. The burrowing rate index (BRI) did not consistently remove the effect of size for use in interspeci®c comparisons of burrowing ability of crabs. For Emerita analoga and Blepharipoda occidentalis, BRI varied signi®cantly with crab size in regressions for eight of the 10 treatments (Table 3). BRI increased signi®cantly with crab size for ®ve of those eight treatments, suggesting that larger crabs may be relatively slower burrowers. For Lepidopa californica, the regressions of BRI and carapace length were negative and not signi®cant in the ®ve sediment grades (Table 3). Sediment size affected burrowing times for all three species. For each crab species, the slope of regressions of burrowing time and size (mass or CL) differed signi®cantly among the ®ve sediment grades (ANCOVA for CL: Emerita analoga, F 5 8.69, n 5 175, P , 0.001; Lepidopa californica, F 5 17.08, n 5 109, P , 0.001; Blepharipoda occidentalis, F 5 2.97, n 5 186, P , 0.001). Where those differences occurred varied among the species. Post hoc tests indicated no difference in slopes among the very ®ne, ®ne, and very coarse sediment grades for E. analoga. For L. californica, there were no differences in slopes among the very ®ne, ®ne and medium sediment grades. For B. occidentalis, only the ®ne and medium sediment grades did not have different slopes. The interaction between sediment size and species was signi®cant in the two-way ANOVA with mass as a covariate (P , 0.001, Table 4). Therefore, although the main effects of both species and sediment grade on burrowing time were also signi®cant (P , 0.001, Table 4), those effects on burrowing time must be interpreted with caution. The patterns of the response of burrowing times to increasing sediment size differed among the three crab species (Table 2, Fig. 1). A much greater range of actual burrowing times was observed for Blepharipoda occidentalis and Lepidopa californica across the ®ve sediment grades (|15-fold and |41-fold, respectively) than for Emerita analoga (about eight-fold) (Table 2). For standard sizes, the responses of the burrowing times of the two albuneid crab species (B. occidentalis, L. californica) to increasing grain size differed markedly from that of the hippid crab (E. analoga) (Fig. 1). In medium grade sand, the burrowing rates of the three species were similar. In ®ne sands, J.E. Dugan et al. / J. Exp. Mar. Biol. Ecol. 255 (2000) 229 ±245 235 0.31 ns 21 rn 2 species in ®ve 05.40 0.81*** 21 2 0.74 26.75 Slope Inter. 2 0.01 ns 22 0.34* ± 0.30 7.01 0.56*** ± rn 2 2 0.72 0.60** 22 0.95 3.30 0.74*** 37 0.12 10.42 0.38* 37 2 2 0.07 42.97 0.44 72.29 Slope Inter. 2 2 0.001) , P 0.30 ns 22 rn 2 0.01, *** , P 0.73 0.66*** 22 0.29 0.19 0.61*** 38 0.21 2 2 0.05, ** , 4.44 148.41 Slope Inter. 2 P 0.34 ns 22 rn 2 0.70 0.62** 22 0.17 3.33 0.73*** 45 0.08 2 2 values: ns, not signi®cant, * 4.59 163.85 2.06 212.67 0.52*** ± 0.22 81.42 0.09 ns ± r Slope Inter. 2 2 rn 0.92 0.66*** 22 0.15 4.40 0.80*** 29 0.14 2 2 2.02 177.07 0.63*** ± 10.7 288.74 0.39 ns 22 2 Slope Inter. 2 Table 3 Ordinary least squares regressions ofsediment burrowing grades time and (signi®cance (1) levels carapace for length,Species (2) mass, and Very of ®ne BRI (burrowing rate index) and carapace length for each Emerita analoga LengthMass 0.14BRI 0.26Lepidopa 1.32 californica 0.74Length 0.73*** 3.05 Fine 35 18.94 0.14 0.69*** 0.18 0.54*** 35 35 0.38 0.08 0.39 2.06 0.83*** 30.22 35 0.80*** 0.08 0.28 35 ns 35 0.17 Medium 0.57 1.52 1.32 0.79*** 35.04 35 0.83*** 0.07 35 0.58*** 35 0.14 1.67 1.56 2.47 0.66*** 11.86 Coarse 35 0.59*** 0.81*** 0.16 35 35 0.33 0.43 0.89 2.73 25.08 0.83*** 35 0.73*** 0.40* 35 35 Very coarse MassBRI 0.26 0.50 0.60** 22 0.33 0.69 0.63*** 22 0.36 0.85 0.65*** 22 0.60 1.98 0.63*** 22 2.08 3.54 0.83*** 21 Blepharipoda occidentalis Length 0.20 MassBRI 0.09 1.35 0.80*** ± 0.75 0.06 0.75*** ± 0.04 2.31 0.62*** ± 0.09 3.48 0.74*** ± 0.04 14.25 0.34* ± 236 J.E. Dugan et al. / J. Exp. Mar. Biol. Ecol. 255 (2000) 229 ±245

Table 4 Results of two-way ANOVA on crab species and sediment grade (grain size) with mass as a covariate Source SS DF MS F-ratio P Species 86.00 2 43.00 16.71 ,0.001 Sediment grade 2493.96 4 623.49 242.29 ,0.001 Species*Sediment grade 1840.94 8 230.12 89.43 ,0.001 Mass 470.34 1 470.34 182.77 ,0.001 Error 1168.29 454

E. analoga burrowed in longer times than L. californica and B. occidentalis. However, in coarse sediments, E. analoga burrowed in shorter times than the two albuneid species. Burrowing times for B. occidentalis and L. californica of standard lengths increased by 340 to 700% from ®ne to very coarse sediments, while burrowing times for E. analoga of standard sizes remained relatively constant across all sediment grades (Fig. 1). BRI values followed a similar pattern among the three species. Mean values of BRI for Emerita analoga varied over a narrower range, and were lower than those of the other two species in ®ne sand and higher in coarse sand. Mean BRI varied from 18 to 150 (range 11 to 301) across the ®ve sediment sizes tested for Lepidopa californica,22 to 109 (range 12 to 214) for Blepharipoda occidentalis and 34 to 66 (range 16 to 124) for Emerita analoga. The net movement of all crabs released in the swash zone was generally seaward and to the NE with the longshore current (Fig. 2). In general, Emerita analoga moved the shortest distance (m) from the release point prior to burrowing (Fig. 2). Blepharipoda occidentalis generally moved the greatest distance from the release point before burrowing successfully, while Lepidopa californica moved an intermediate distance (Fig. 2). The time before burrowing in the swash zone varied among the three species. Pre-burrowing time was not correlated with carapace length for any of the species. Mean pre-burrowing times were shorter for Emerita analoga than for the other two species, averaging ,5 s and ranging from 1.5 to 13.2 s (Fig. 3a). Pre-burrowing times were generally shorter than the average swash period (mean swash period 6.2 s10.38 s, n 5 3) for all but two individual E. analoga and three Lepidopa californica. For L. californica, pre-burrowing times averaged 25 s and ranged from 1.0 to 41.1 s. For Blepharipoda occidentalis, pre-burrowing times were longer than the average swash period for more than half of the animals observed, averaging .21 s and ranging from 6.0 to 83.0 s. Once individual crabs made contact with the sediments, burrowing times in the swash zone were somewhat longer but generally similar to those recorded in the laboratory in medium grade sediments for Lepidopa californica (1.4 to 3.6 s) and Emerita analoga (1.3 to 13.4 s, note: all but one individual burrowed in ,5.5 s) (Table 2). Burrowing times in ®eld conditions were longer than in the laboratory for Blepharipoda occidentalis, ranging from 3.5 to 36.0 s (Table 2). Burrowing times in the ®eld increased signi®cantly with carapace length for B. occidentalis and L. californica but not for E. analoga. However, we used a narrower size range of E. analoga in the ®eld trials due to J.E. Dugan et al. / J. Exp. Mar. Biol. Ecol. 255 (2000) 229 ±245 237

Fig. 1. Response of burrowing times to sediment grain size for standard sizes of each species of crab: (a) Emerita analoga,(b)Lepidopa californica and (c) Blepharipoda occidentalis. 238 J.E. Dugan et al. / J. Exp. Mar. Biol. Ecol. 255 (2000) 229 ±245

Fig. 2. Estimated distance moved before burrowing, shown as means and standard deviations of the positions of individuals of each species at the initiation of successful burrowing, following standardized release in the swash zone for (j) Emerita analoga,(d) Lepidopa californica and (m) Blepharipoda occidentalis. the dif®culty of tracking the smaller animals in the swash zone (data was collected on only one individual ,20 mm CL). When burrowing and pre-burrowing times were added together, differences between the three species remained evident in the ®eld trials (Fig. 3b). The proportion of individuals that reached the substrate and burrowed successfully in times less than or equal to the average swash period (6.2 s) varied among species. The majority of Emerita analoga individuals reached the substrate and burrowed in times less than or equal to the swash period. For Lepidopa californica, 50% of the individuals reached the substrate and burrowed in that time. For Blepharipoda occidentalis, the time to reach the substrate and burrow was greater than the swash period for all the individuals observed. The behavioral response to swash conditions also varied noticably between the three species (Fig. 4). Individual Emerita analoga were not observed tumbling, swimming at the surface, or sliding prior to orienting and burrowing and were not dislodged from the substrate by the swash once burrowing had been initiated. Lepidopa californica individuals typically swam or drifted at the water surface with the abdomen closest to the water surface prior to orienting and burrowing and were not observed tumbling or sliding (Fig. 4). This species was not dislodged by swash once burrowing was initiated. J.E. Dugan et al. / J. Exp. Mar. Biol. Ecol. 255 (2000) 229 ±245 239

Fig. 3. (a) Pre-burrowing times, measured as the time between release and the initiation of successful burrowing, for individuals of the three species released in the swash zone. (b) The sum of pre-burrowing and burrowing times for individuals of the three species released in the swash zone. For both ®gures: (s) Emerita analoga,(n) Lepidopa californica and (1) Blepharipoda occidentalis. 240 J.E. Dugan et al. / J. Exp. Mar. Biol. Ecol. 255 (2000) 229 ±245

Fig. 4. Pre-burrowing behaviors of individuals of each species prior to the initiation of successful burrowing following standardized release in the swash zone.

Blepharipoda occidentalis individuals tumbled and/or drifted in the swash, and one individual slid prior to burrowing. Thirty percent of the individuals of this species became dislodged by swash after burrowing had been initiated.

4. Discussion

The three species of crabs investigated in this study were active and rapid burrowers in the laboratory and in the swash zone. Rapid burrowing rates have been reported for other hippid crab species and may be an important adaptation to life in the swash zone of exposed sandy beaches. Burrowing times of 1.5 s were reported for Emerita holthuisi and E. portoricensis on beaches with swash periods of 4 to8s(Ansell and Trueman, 1973). The crabs we investigated had larger BRI values (11 to 301) than those reported for beach bivalves (2 to 17) in their native sediments (McLachlan et al., 1995), suggesting that the hippoid crabs burrow at much higher rates than bivalves of similar sizes. For clams from various habitats, Stanley (1970) reported mean BRI values ranging from 0.01 to 20 and classi®ed those .6 as very rapid. Other intertidal crustaceans, such as mysids, have also been reported to be rapid burrowers, with BRI values of 6 to 24 (Nel et al., 1999). Our results suggest that modi®cation of the BRI formula may be needed to remove the effect of size for application to burrowing decapods. Infaunal species have been broadly classi®ed by the sensitivity of their burrowing J.E. Dugan et al. / J. Exp. Mar. Biol. Ecol. 255 (2000) 229 ±245 241 speeds to a range of sediment grain sizes (Alexander et al., 1993). The burrowing speeds of species which are substrate generalists are relatively insensitive to a wider range of sediment grain sizes compared to those of species that are substrate specialists or substrate sensitive (Alexander et al., 1993). The sensitivity of burrowing times to sediment size varied among the crab species in our laboratory trials, suggesting that the three species respond to sediment size differently. In ®ner sediment grades, the three species burrowed at similar rates. In coarser sediments, differences in burrowing rates were most apparent among the species. Those results suggest that the hippid crab, Emerita analoga, can be considered a substrate generalist, able to burrow rapidly across a range of sediment grain sizes (sensu Alexander et al., 1993), while the two albuneid crabs, Blepharipoda occidentalis and Lepidopa californica, can be classi®ed as substrate sensitive species which burrow more slowly in coarser sediment. The ability of substrate generalists, such as E. analoga, to burrow at similar speeds across a range of sediment grain sizes likely contributes to their success in the coarse sediments and harsher swash conditions typical of beaches with re¯ective characteristics (see Dugan and Hubbard, 1996; Dugan et al., 2000). Substrate generalists, such as E. analoga, may also cope better with temporal changes in sediment grain size, such as those occurring on intermediate beaches in Southern California (Dugan unpublished). A number of other sandy beach crustaceans, including several species of isopods (Nel, 1995) and Hippa paci®ca (Lastra et al., unpublished), are able to use a wide range of sediment sizes. Another crustacean, the beach mysid, Gastrosaccus psammondytes, although one of the most rapid burrowers reported, is considered a substrate specialist (sensu Alexander et al., 1993) because fastest burrowing times are reported in sediments with grain sizes between 0.25 and 1.00 mm and slower times occur in ®ner or coarser sediments (Nel et al., 1999). For the species that have been investigated, beach bivalves of the genus Donax have been reported as substrate sensitive with slower rates in coarse sediments (Alexander et al., 1993; Nel, 1995; de la Huz et al., unpublished). Comparisons of the sensitivity of the burrowing times of infaunal animals to a range of sediment grain sizes in still water addresses only one aspect of life in the intertidal zones of exposed sandy beaches. In our ®eld studies, at least two components appeared to be important for crabs inhabiting the swash zone: pre-burrowing time and behavior (e.g., orientation and movement before contact is made with the benthic substrate) and burrowing time. Our results suggest that differences in pre-burrowing times and behaviors among the three crab species may be signi®cant in determining natural distributions of these species. For example, Blepharipoda occidentalis individuals were transported greater distances and often dislodged by swash while attempting to contact the substrate and burrow in natural conditions on a ®ne sand, re¯ective beach. This, in combination with the longer burrowing times of B. occidentalis and L. californica in coarser sediments, may be important in excluding these crabs from the swash zones of many low intermediate to re¯ective beaches, particularly those with short swash periods and coarser sediments. Studies of the mechanics (Trueman, 1970; Ansell and Trueman, 1973), behavior patterns and neuromuscular coordination (Paul, 1981, 1991; Faulkes and Paul, 1997a,b) of digging in the hippoid crabs suggest a strong link between these attributes and the relative ability of the three species we studied to orient and burrow in the swash zone. 242 J.E. Dugan et al. / J. Exp. Mar. Biol. Ecol. 255 (2000) 229 ±245

Crabs in the superfamily Hippoidea use locomotory or walking limbs to burrow in soft sediments, and digging movements are generally believed to have evolved from decapod walking (Faulkes and Paul, 1997b). The coordination of movements of the legs, abdomen and tailfan in digging backwards differs between the hippid and the albuneid species (Faulkes and Paul, 1997a). Although basic similarities are present, the digging pattern of Emerita analoga diverges from the albuneid crabs in both the lack of a gait switch, and in the tighter coordination with the uropods and higher frequency of activity of leg 4 (Faulkes and Paul, 1997a). One visible result of these differences is that E. analoga exhibits a smooth descent into the sediments while the two albuneid species descend in a stepwise `ratcheting manner' (Faulkes and Paul, 1997a). Aspects of the neuromuscular coordination and anatomy may also help explain the differences we observed in the response of the three species to different sediment grain sizes. Emerita analoga has a signi®cantly larger tailfan than Blepharipoda occidentalis or Lepidopa californica. In E. analoga, more than 50% of the telson mass is comprised of a power stroke muscle which is absent in the other two species (Paul, 1981). The smaller, weaker albuneid tailfan is primarily used to induce thixotrophy in the sediments (Faulkes and Paul, 1997a), a feature we suggest which would be most advantageous in ®ner sediments and much less so in coarser sediment. In contrast, E. analoga use the large uropods in coordination with the strong tail to induce thixotrophy and to physically shovel sand (Faulkes and Paul, 1997a), a feature which would likely be advantageous in a wider range of sediment sizes. Unique aspects of the digging pattern, anatomy and neuromuscular coordination of the tailfan and legs of E. analoga may be important in allowing rapid burrowing in a wide range of sediment types as observed in our study. In addition to burrowing, differences in the swimming modes and relative abilities of hippids and albuneids may contribute to our results for the three California species. Emerita analoga swims by beating the uropods and using the long front legs as rudders without moving the thoracic legs or ¯ipping the tail (Paul, 1981). This unique feature allows strong and well-oriented backward swimming by this species (Paul, 1981) which may be advantageous for orientation and directed movement in the turbulent swash zone of exposed sandy beaches. Blepharipoda occidentalis swims weakly by ¯ipping its tail and rowing with the second and third thoracic legs (Faulkes and Paul, 1997a), thus is less able to orient or swim in currents typical of the swash zone of exposed sandy beaches. Lepidopa californica has not been studied, but its musculature is similar to that of B. occidentalis (Faulkes and Paul, 1997a). Results from our ®eld tests of swash behavior suggest this difference in swimming mode and ability among the two hippoid families investigated is critical and could in¯uence the distribution of these species among exposed sandy beaches with different swash climates. Our results on three decapod crabs support predictions of the swash exclusion hypothesis of McLachlan et al. (1993, 1995) for burrowing and locomotory abilities. The differences we observed in response to sediment grain size and swash among the hippoid crab species could help explain interspeci®c differences in the spatial distribution and relative abundances of these crabs observed among exposed sandy beaches of different morphodynamic types in California in other studies. Mean sediment particle sizes of the lower intertidal zone of 36 intermediate to re¯ective beaches in southern California ranged from 0.15 to 0.85 mm (Dugan et al., 2000). Mean grain sizes between J.E. Dugan et al. / J. Exp. Mar. Biol. Ecol. 255 (2000) 229 ±245 243

0.20 and 0.30 mm occurred on the majority of those beaches, matching the ®ne to medium sediment grades used in our study. Our results suggest that none of the hippoid species investigated would be excluded on the basis of grain size alone in that range of sediments. Swash periods, where measured, varied considerably more among those beaches, ranging from 4.5 to 20 s (Dugan, 1999), and thus may play a larger role in the distribution of the hippoid crab species. Emerita analoga occurred on every beach surveyed and was the most abundant individual species on 22 of the 36 beaches (Dugan et al., 2000). Estimated mean abundances of E. analoga ranged from 360 to over 79,000 individuals m21 and this species comprised the majority of the macrofauna (40 to 99%) at a number of beaches (Dugan et al., 2000). Blepharipoda occidentalis and Lepidopa californica each occurred on ,45% of the 36 beaches with mean abundances of up to 190 and 220 individuals m21 , respectively (Dugan et al., 2000). These species never constituted the majority of the total macrofauna abundance on any of the 36 beaches sampled (Dugan et al., 2000). Our results for the hippoid crab species of California beaches suggest that an additional prediction or re®nement of the swash exclusion hypothesis could be made with regard to the intertidal and subtidal distribution of species with different burrowing abilities and swash behaviors. On beaches or during episodes with harsh swash climates and coarse sediments, the distribution of species less capable of dealing with those conditions could shift from the intertidal swash zone to the shallow subtidal. In support of this idea, the two albuneid species are not restricted to the intertidal zone and Blepharipoda occidentalis sometimes reaches peak abundances in the shallow subtidal (Fager, 1968; Morris et al., 1980; Morin et al., 1985) below the reach of regular intertidal swash. We suggest that the two albuneid species may extend into the intertidal only in more benign swash climates and ®ner sediments, conditions which could vary spatially and temporally, particularly on beaches of intermediate morphodynamic state. In contrast, Emerita analoga occurs almost exclusively in the intertidal zone much of the year and is subtidal only during periods of beach erosion or on extremely high tides (Jaramillo et al., 2000; Dugan, unpublished). The combination of substrate generalist characteristics and unique orientation and swimming abilities of Emerita analoga may be a key to the high abundance and wide distribution of this species in the intertidal zone of exposed sandy beaches of all morphodynamic types along the coast of California (Dugan et al., 1995, 2000) and in other latitudes on the Paci®c coast of South and North America. Although our results clearly support a number of predictions of the swash exclusion hypothesis, the relative dominance of populations of E. analoga and, potentially, of its congeners in other regions may also have important implications for understanding the structure of macrofauna communities of exposed sandy beaches. For example, the general predictions of macrofaunal community structure (species richness, abundance and biomass) with respect to beach morphodynamic state proposed by McLachlan (1990) have now been reported for many regions of the world (e.g., Jaramillo and McLachlan, 1993; McLachlan et al., 1993, 1996, 1998; Hacking, 1998). However, in southern California where populations of E. analoga occur on every beach type and often dominate the intertidal abundance and biomass of exposed sandy beaches, no relationships between the population biology, abundance and biomass of this species, and morphodynamic 244 J.E. Dugan et al. / J. Exp. Mar. Biol. Ecol. 255 (2000) 229 ±245 beach state were found (Dugan and Hubbard, 1996). In addition, macrofaunal community structure did not ®t the general predictions of McLachlan (1990) for the 36 intermediate to re¯ective beaches investigated to date in this region (Dugan, 1999; Dugan et al., 2000). Understanding the role of this uniquely adapted decapod species and its congeners in macrofaunal community structure could be important in the development of new concepts and the evaluation of existing ideas concerning ecological processes on exposed sandy beaches.

Acknowledgements

We thank M. Page and his students in the summer marine ecology course at University of California at Santa Barbara for assistance with ®eld collections. The manuscript bene®ted greatly from discussions with A. Brooks, M. Page and C. de la Huz. This research was partially supported by a grant from the Minerals Management Service, U.S. Dept. of the Interior, under MMS agreement No. 14-35-0001-30758 and the Coastal Marine Institute, University of California to J. Dugan and a research bursary from the University of Vigo to M. Lastra. The views and conclusions contained in the document are those of the authors and should not be interpreted as necessarily representing the of®cial policies, either expressed or implied, of the U.S. government. [RW]

References

Alexander, R.R., Stanton, R.J., J., Dodd, J.R., 1993. In¯uence of sediment grain size on the burrowing of bivalves Ð correlation with distribution and stratigraphic persistence of selected neogene clams. Palaios 8, 289±303. Ansell, A.D., Trueman, E.R., 1973. The energy cost of migration of the bivalve Donax on tropical sandy beaches. Mar. Behav. Physiol. 2, 21±32. Brown, A.C., Trueman, E.R., 1994. The burrowing rate index. J. Molluscan Studies 60, 354±355. Dugan, J.E., 1999. Utilization of sandy beaches by shorebirds: relationships to population characteristics of macrofauna prey species and beach morphodynamics. Draft Final Study Report to Minerals Management Service and the UC Coastal Marine Institute. Dugan, J.E., Hubbard, D.M., 1996. Local variation in populations of the sand crab, Emerita analoga (Stimpson), on sandy beaches in southern California. Rev. Chil. Hist. Nat. 69, 579±588. Dugan, J.E., Hubbard, D.M., Page, H.M., 1995. Scaling population density to body size: tests in two soft sediment intertidal communities. J. Coastal Res. 11 (3), 849±857. Dugan, J.E., Hubbard, D.M., Engle, J.M., Martin, D.L., Richards, D.M., Davis, G.E., Lafferty, K.D., Ambrose, R.F., 2000. Macrofauna communities of exposed sandy beaches on the Southern California mainland and Channel Islands. Fifth California Islands Symposium, OCS Study, MMS 99-0038, pp. 339±346. Efford, I., 1976. Distribution of the sand crabs in the genus Emerita (Decapoda, Hippidae). Crustaceana 30 (2), 169±183. Fager, E.W., 1968. A sand-bottom epifaunal community of invertebrates in shallow water. Limnol. Oceanogr. 13, 448±464. Faulkes, Z., Paul, D.H., 1997a. Coordination between the legs and tail during digging and swimming in sand crabs. J. Comp. Physiol. A Ð Sensory, Neural, Behav. Physiol. 180 (2), 161±169. J.E. Dugan et al. / J. Exp. Mar. Biol. Ecol. 255 (2000) 229 ±245 245

Faulkes, Z., Paul, D.H., 1997b. Digging in sand crabs (Decapoda, Anomura, Hippoidea): interleg coordination. J. Exp. Biol. 200 (4), 793±805. Hacking, N., 1998. Macrofaunal community structure of beaches in northern New South Wales, Australia. Mar. Fresh. Res. 49 (1), 47±53. Haley, S.R., 1982. Zonation by size of the Paci®c mole crab, Hippa paci®ca Dana (Crustacea: Anomura: Hippidae), in Hawaii. J. Exp. Mar. Biol. Ecol. 58, 221±231. Jaramillo, E., McLachlan, A., 1993. Community and population responses of the macroinfauna to physical factors over a range of exposed sandy beaches in south-central Chile. Est. Coastal Shelf Sci. 37, 615±624. Jaramillo, E., Dugan, J., Contreras, H., 2000. Abundance, population structure, tidal movement and burrowing rate of Emerita analoga (Stimpson 1857) (Anomura, Hippidae) at a dissipative and a re¯ective beach in south central Chile. Mar. Ecol. Napoli. 21 (2), 113±127. McArdle, S., McLachlan, A., 1991. Dynamics of the swash zone and ef¯uent line on sandy beaches. Mar. Ecol. Prog. Ser. 76, 91±99. McArdle, S., McLachlan, A., 1992. Sand beach ecology: swash features relevant to the macrofauna. J. Coastal Res. 8, 398±407. McLachlan, A., 1990. Dissipative beaches and macrofauna communities on exposed intertidal sands. J. Coastal Res. 6, 57±71. McLachlan, A., 1996. Physical factors in benthic ecology: effects of changing sand particle size on beach fauna. Mar. Ecol. Prog. Ser. 131, 205±217. McLachlan, A., de Ruyck, A.M.C., Hacking, N., 1996. Community structure on sandy beaches: patterns of richness and zonation in relation to tide range and latitude. Rev. Chil. Hist. Nat. 69, 451±467. McLachlan, A., Jaramillo, E., Donn, T.E., Wessels, F., 1993. Sandy beach macrofauna communities and their control by the physical environment: a geographical comparison. J. Coastal Res. Spec. Issue 15, 27±38. McLachlan, A., Fisher, M., Al-Habsi, H.B., Al-Shukairi, S.S., Al-Habsi, A.M., 1998. Ecology of sandy beaches in Oman. J. Coastal Cons. 4 (2), 181±190. McLachlan, A., Jaramillo, E., Defeo, O., Dugan, J., de Ruyck, A., Coetzee, P., 1995. Adaptations of bivalves to different beach types. J. Exp. Mar. Biol. Ecol. 187, 147±160. Morin, J.G., Kastendiek, J.E., Harrington, A., Davis, N., 1985. Organization and patterns of interactions in a subtidal sand community on an exposed coast. Mar. Ecol. Prog. Ser. 27, 163±185. Morris, R.H., Abbott, D.P., Haderlie, E.C., 1980. Intertidal Invertebrates of California. Stanford University Press, Stanford, 690 pp. Nel, R., 1995. The effect of sand particle size on sandy beach macrofauna. MSc. Thesis, University of Port Elizabeth, Port Elizabeth. Nel, R., McLachlan, A., Winter, D., 1999. The effect of sand particle size on the burrowing ability of the beach mysid, Gastrosaccus psammodytes Tattersall. Est. Coastal Shelf Sci. 48, 599±604. Paul, D.H., 1981. Homologies between body movements and muscular contractions in the locomotion of two decapods of different families. J. Exp. Biol. 94, 159±168. Paul, D.H., 1991. Pedigrees of neurobehavioral circuits: tracing the evolution of novel behaviors by comparing motor patterns, muscles, and neurons in members of related taxa. Brain Behav. Evol. 38, 226±239. Short, A.D., 1996. The role of wave height, period, slope, tide range and embaymentisation in beach classi®cations: a review. Rev. Chil. Hist. Nat. 69, 589±604. Stanley, S.M., 1970. Shell form and life habits in the Bivalvia. Geological Society of America Memoir 125, 296 pp. Trueman, E.R., 1970. The mechanism of burrowing of the mole crab, Emerita. J. Exp. Biol. 53, 701±710. Trueman, E.R., Ansell, A.D., 1969. The mechanisms of burrowing into soft substrata by marine animals. Oceanogr. Mar. Biol. Annu. Rev. 7, 315±366.