Abstract This study examined behavioral patterns of three intertidal whelks in response to tidal variation. Nucella emarginata, Ocenebra circumtexta, Acanthanucella punctulata have widely overlapping habitat but significantly different resource utilization strategies. All three species were found in the primary study site, a 6m by 9m region of rocky intertidal shore with a vertical range of 0 to +2.5m above MLLW. Individuals were marked and observed at low tide for a two week period to discern basic patterns of movement and aggregation. Following this, surveys of microhabitat and activity were taken of the same area at daily high and low low tides for another two weeks. Acanthanucella were highly responsive to tidal patterns with 76% feeding at high tide and only 27% feeding at low. They also fed significantly more on sun-exposed patches of barnacles, often horizontal surfaces, during high tide (66% feeding in sun) than during low tide (21% feeding in sun). Ocenebra show a slight but significant tendency to feed less in the sun at low tide (27%) than at high tide (46%). However, the frequency of feeding in shaded areas does not vary significantly with the tide, possibly because these whelks are unique in utilizing undercut habitat. Nucella appear to have the most conservative behavior, feeding the least at high tide (61%) and showing non-significant responses to sun exposure and the timing of the low tide. Introduction Marine organisms that inhabit intertidal ecosystems must be capable of coping with an extreme range of environments. Most notably, they must be able to withstand exposure during low tide and the accompanying increase in temperature and desiccation rates (Somero, 2002). Because marine organisms are primarily adapted to an aquatic environment, the period spent exposed during low tide often requires dramatic responses and alterations in behavior. These pressures have produced a wide range of behavioral strategies in intertidal organisms. While some strategies are clearly different, such as between crabs and mussels, many more involve subtle differences in use of habitat. Even organisms with similar physiology have developed markedly different ways of surviving in the intertidal. This study examines the behavioral strategies of three intertidal whelks, Nucella emarginata, Ocenebra circumtexta, and Acanthanucella punctulata. Nucella is found in variable wave conditions from mid to high intertidal height and feed predominantly on Mytilus mussels and Balanus barnacles. Acanthanucella is typically found high in the intertidal in moderate wave conditions and feed on Balanus and Chthamalus barnacles as well as Littorina snails. Ocenebra is found commonly in heavy surf in the mid to low intertidal and feed predominantly on Chthamalus and Tetraclita barnacles. Despite these differences, these three species frequently occupy the same habitat. By observing these whelks when they are found together, this study was able to distinguish unique behavioral strategies that each species use to survive in the intertidal. Materials and Methods This study was carried out in the Hopkins Marine Life Refuge in Pacific Grove, California (36°37N, 121°54'W). The principal study site was a 6 by 9 meter region of rocky intertidal shore inhabited by Nucella emarginata, Ocenebra circumtexta, Acanthanucella punctulata and a small number of Acanthanucella spirata. The latter species was not included in this study due to its low abundance. The other three species are found in patchy abundance along the shores of the Refuge, but this site was ideal because it 1) was a place where all three species were found together in abundance, and 2) was easily accessible during high and low tides. To augment the data from this site, two isolated granite boulders with abundant Acanthanucella, one within 10m of the primary site and the other approximately 100 meters away, were also examined. All three of these sites had healthy stocks of Chthamalus spp. and Balanus glandula which are the principal prey for all three whelk species. The primary site also had a substantial population of Tetraclita rubescens and a limited population of the mussel Mytilus californianus. Initially, all the individual whelks within a subsection of the primary site were marked with enamel paint and observed at the daily low tide for 2 weeks. Twenty four Nucella, 19 Acanthanucella, and 21 Ocenebra were individually marked. In addition, two aggregations of 25 Acanthanucella were marked as groups, as was one Nucella aggregation of 8. During each observational period I noted the location of each marked whelk found within the study site. I also performed site-wide surveys of feeding activity and microhabitat conditions at the primary site and the two secondary sites. I conducted these surveys for two weeks at daily low low and high low. The microhabitat data were categorized according to the following factors: rock orientation (vertical, horizontal, overhang), presence of protective cover (red algal growth or sea anemones versus bare rock), crevice or open rock face, barnacle cover (none, sparse, medium, dense), and whether the location was sun-exposed or shaded. The results of microhabitat usage were statistically analyzed using chi-square tests of goodness-of-fit, which compared the observed frequencies against the null hypothesis that all species utilize each habitat category in the same proportion. A 3-factor analysis of variance (ANÖVA) was used to compare feeding data, with species, tide height, and sun exposure as orthogonal fixed factors. Subsequent post-hoc comparisons were conducted using Student-Newman-Kuels tests (Underwood, 1997). The feeding frequency at high or low tide was compared against time of day using simple linear regression. Results Daily Observation of Marked Individuals Acanthanucella was the most active species. The two non-mating aggregates, which had been relatively stable for at least 2 days before marking, quickly disbanded. One aggregate fell to 7 Acanthanucella after one day and 1 after two, while the other dropped to 9 then 3 in the same time. These marked individuals were observed in the area for a few days but many were not seen again. Neither of these areas, which were barnacle-free undercuts, was ever repopulated. An unmarked mating aggregate within the study site remained relatively stable at approximately 20 individuals. The individually marked Acanthanucella were similarly mobile. Äfter two weeks, only 3 of the 19 (16%) were found on the rock were they were tagged initially. Those that had moved often had to travel through extensive barnacle-free rubble to reach their new location. On average, 33% of all marked individuals were found during daily observation. Acanthanucella frequently foraged in groups, typically of 4 to 10 individuals that emerged from shelter and traveled together into sun-exposed barnacle-covered areas. Ocenebra, like Acanthanucella, was highly mobile with an average recovery of only 23%. Äfter two weeks, only 2 of the original 21 (10%) were found on the same rock. These whelks were never found in aggregations, though they often occurred in pairs in what was possibly mating behavior. Ocenebra also traveled through rubble to reach new rocks. Nucella was less mobile than either Ocenebra or Acanthanucella. On average, 50% of all marked individuals were found each day and, unlike the other species, they were likely to be found on the same rock on successive days. Ten of the 24 (42%) remained on the same rock for the entire two weeks. Those that did travel remained on near-continuous barnacle cover. There was some aggregation behavior in Nucella but it was not as obvious or consistent as Acanthanucella. Several Nucella laid eggs, but these individuals were always solitary or with no more than 4 other conspecifics. Of the 8 individuals marked as a group, 6 dispersed individually and 2 remained at the original location. It was common for individual whelks to forage continuously for days or to remain in a barnacle-free crevice for weeks. The study site and the surrounding area were examined for marked snails after 1 month. Of the 24 Nucella originally marked, 11 were found within the site, 6 of which had not moved in two weeks. Four individually marked Acanthanucella were found within the site though all had moved to new rocks. One of the aggregations had relatively high returns: 4 of the 25 were found in a mating aggregation and 5 were moving as part of a foraging group. Only a single individual was found from the other aggregation. Two of the 21 marked Ocenebra were observed after a month and only one was inside the study site. Feeding and Microhabitat Relatively low proportions of Acanthanucella fed during low tide, but the percent feeding increased dramatically at high tide (Fig.1). These proportions varied significantly depending on the time of day when low tide or high tide occurred: there was a significant increase in the percent feeding at high tide as the tide progressed from sunrise to noon and a significant decrease in the percent feeding at low tide in the same progression (Fig. 2). The proportion feeding in shaded habitat remained low at high and low tides, but the proportion feeding in sun-exposed habitat increased significantly between low and high tide (Fig. 1, Table 1). When Acanthanucella are feeding at low tide, they were predominantly found on vertical surfaces, with a low percentage feeding on horizontal or undercut substrate (Fig. 3). At high tide, they fed in nearly equal proportions on vertical and horizontal surfaces and few were found feeding on undercuts (Fig. 3). They fed in similar proportions on low and medium density barnacle cover and slightly higher proportions on high density barnacle cover (Fig. 4). When not feeding, Acanthanucella were found in similar abundance on horizontal and undercut substrate and slightly less on vertical (Fig. 5). In these three orientations they were found in equal proportions in crevices and on rock faces (Fig. 6). However, there was a greater percentage found in protective cover (i.e. red algal cover or at bases of anemones) than on rock (Fig. 7). Roughly half of observed Nucella were feeding at either low or high tide (Fig.1). These feeding proportions did not change significantly as the high and low tides progressed into the day (Fig. 8). Nucella did not feed in significantly different proportions in sun-exposed or shaded habitat at either high or low tide. They fed predominantly on vertical surfaces at both tides (Fig. 3) and were most often on medium density barnacle cover, though this percentage was highly variable from day to day (Fig. When not feeding, Nucella were found in roughly equal proportions on horizontal and vertical substrate but did not make use of undercuts (Fig. 5). Nucella were much more likely to be found in crevices than open rock faces when not feeding (Fig. 6) and were rarely found nestled in protective cover (e.g. red algae or the bases of anemones). Ocenebra fed in high proportions at both high and low tide (Fig. 1). The percent feeding at high tide did not vary depending on the timing of that tide, but the percent feeding at low tide decreased significantly as the low tide progressed into the day (Fig. 9). At high tide they fed in roughly equal proportions on sun-exposed and shaded habitat, but at low tide percent feeding in the sun was significantly less (Table 1). Ocenebra fed on undercut surfaces in approximately the same proportions as vertical at low tide and slightly less at high tide (Fig.3). They made little use of horizontal substrate at either tide (Fig.3). Ocenebra were characteristically found in higher proportions on high density barnacle cover than low or medium density (Fig. 4). Ocenebra were found predominantly on vertical surfaces when not feeding, but the proportion found on each orientation was highly variable from day to day (Fig. 5). Their use of crevices and protective cover was also highly variable and showed no clear trends (Fig. 7, Fig. 8). Discussion Acanthanucella are highly mobile predators and very responsive to changes in water level. As the tide rises, these whelks move into horizontal and sun-exposed habitat. As the tide recedes and these areas become increasingly stressful, they move back into protective cover (Morris et al, 1980). Acanthanucella also show dramatic variation in feeding frequency depending on the timing of the tide. Similar behavior can also be found in an Australian whelk, Thias orbita (Fairweather, 1988). In addition to this pattern of local movement, Acanthanucella show high dispersal rates and may travel across extensive rubble to reach new food sources. When foraging in this manner, they tend to travel in groups. Acanthanucella are also commonly found in aggregations when not foraging. The extent to which Acanthanucella form aggregates and travel in groups may depend on the time of year. This study was done during the peak of the Acanthanucella mating season (May-June) and they are known to mate in tight aggregations (Morris et al, 1980). One such aggregation was observed in the primary study site. However, other aggregations were loose and quickly dispersed. In addition, many foraging Acanthanucella traveled in groups without apparent mating activity. This likely indicates that coordinated behavior is characteristic of Acanthanucella and not entirely limited to the mating season. The influence of mating season on aggregation behavior warrants further investigation. Nucella display more conservative behavior. When they do move, which is less often than the other species, they rarely leave barnacle cover or face prolonged sun exposure. In addition, they feed the least at high tide and show no significance response to changes in the daily timing of the tides. When not feeding they typically seek protection in crevices, which may decrease stress due to emersion. A few Nucella were observed laying eggs, which supports the conclusion from the literature that they mate sporadically throughout the year (Morris et al, 1980). However, because this was limited to a few individual whelks, it is unlikely that mating had a large influence on the observed behavior of the entire population. The activity of Nucella varied dramatically within the population. This observation is supported by a previous survey of 128 Nucella emarginata (West, 1986). During a four month period of low tide observations, 19% of the marked individuals where not observed to feed while 40% made five or more sequential feeding attacks. This raises the question of whether individuals display uniformly different foraging behavior or cycle between periods of high and low activity. This question could be addressed by detailed long-term observation of individual whelks throughout the tide cycle. Ocenebra is unique in feeding heavily in both sun-exposed and shaded habitat, though they feed significantly less in the sun at low tide. Like Acanthanucella, the proportion feeding at low tide varies significantly depending on the time of day. Ocenebra do not, however, show any variation with the timing of the high tide, likely because nearly the entire population feeds when submerged. Ocenebra is also unique in utilizing undercut habitat for feeding as well as protection. Previous studies of Ocenebra found interindividual variation in preferred prey species (Dusek, 1999). If the species composition of barnacle cover is significantly influenced by substrate orientation or sun-exposure, this variation may produce different foraging strategies within an Ocenebra population whose individuals specialize on different barnacle species. For example, individuals that feed preferentially on barnacles found on sun exposed habitat may adopt a strategy similar to Acanthanucella and move with the tide. Conversely, Ocenebra that feed on barnacles found on undercuts or protected habitat may feed continually. Although this study did not analyze barnacle distribution by microhabitat within the study site, there is some evidence for such a hypothesis. Ocenebra feed predominantly on Tetraclita and Chthamalus barnacles. (Morris et al, 1980). Of the three barnacle species, Tetraclita occurs lowest on the shore and may therefore grow best in undercuts and similarly protected habitat at higher levels (Morris et al, 1980). If this is the case, then Ocenebra may be drawn towards undercuts out of food preference as well as for protection from physical stress. In contrast, Burrow (2003) found that Chthamalus is the dominant barnacle on sun-exposed rocks. As a result of these dispersal patterns, Ocenebra with different food preferences may be forced to adopt different behavioral strategies. Direct tests of the effect of individual food preferences on foraging strategies of Ocenebra would be of interest. A few general assumptions may require further investigation. When looking for marked whelks, I had to assume that the enamel paint would be sufficiently permanent for the duration of my study. I also took it for granted that applying the paint did not alter the whelks' behavior. However, as many marked whelks were not recovered and both Acanthanucella aggregations dispersed soon after being marked, these assumption may be unwarranted. In addition, I assumed that the behavior of each species was not dramatically influenced by the presence of the others. Although Acanthanucella have been known to occasionally prey on Nucella, this was never observed in my site (Morris et al, 1980). Previous studies have found that behavioral patterns of individual species vary depending on site conditions, local topography, and community structure. (Burrows and Hughes, 1989; Dahlhoff et al, 2001; Fairweather, 1988). An obvious extension of this comparative survey would be to examine behavior of these three whelks when they are found together in different habitat. In conclusion, this study has shown that Nucella emarginata, Ocenebra circumtexta, and Acanthanucella punctulata exhibit different resource utilization strategies when occupying the same habitat. Further information is needed from additional sites to determine if these patterns are general or if they vary in space and time. Regardless, my results suggest that further investigations to identify the ecological, behavioral, and physiological selection pressures that produce divergence among such ecologically similar, closely related sympatric species would be fruitful. Acknowledgements 1 would like to sincerely thank Jim Watanabe who helped me extensively throughout the entire process of this study. I would also like to thank Mark Denny for supplying equipment to study physical processes in the intertidal. Literature Cited Burrow, T. 2003. Field observations of Nucella emarginata: effects of temperature on feeding. Unpublished student paper, Hopkins Marine Station library, Stanford University. Burrows, M.T. and Hughes, R.N. 1989. Natural foraging of the dogwhelk, Nucella lapillus (Linnaeus); the weather and whether to feed. Journal of Molluscan Studies 55, 285-295. Dahlhoff, E., Bradley, B. and, Menge, B., 2001. Physiology of the rocky intertidal predator Nucella ostrina along an environmental stress gradient. Ecology 82(10), 2816-2829. Dusek, E. 1999. Ocenebra circumtexta in the intertidal: stratification and prey selection of a predatory gastropod. Unpublished student paper, Hopkins Marine Station library, Stanford University. Fairweather, P.G. 1988. Movement of intertidal whelks (Morula marginalba and Thais orbita) in relation to availability of prey and shelter. Marine Biology 100, 63-68. Morris, R.H., Abbott, D.P., and Haderlie, E.C., 1980. Intertidal invertebrates of California. Stanford University Press, Stanford, CA. 277-283, 515-521 Somero, G. 2002. Thermal physiology and vertical zonation of intertidal animals: optima, limits, and costs of living. Integrative and Comparative Biology 42, 780-789 Underwood, A.J. 1997. Experiments in ecology: their logical design and interpretation using analysis of variance. Cambridge Univ. Press., New York, NY. West, L., 1986. Interindividual variation in prey selection by the snail Nucella emarginata. Ecology 67, 798-809. Table 1: Results of 3-way ANÖVA of daily percent feeding with tide height, sun exposure, and whelk species. Tide height was high' or low', sun exposure was 'sun-exposed' or 'shaded'. Cochran's Test for homogeneity of variances was not significant (s'max - Esi 0.156); for Student-Newman-Keuls tests, means that are underlined are not significantly different (P2.05). Analysis of Variance Source Sum-of-Squares Mean-Square F-ratio Species 6312.055 3156.027 17.188 5047.147 Tide 5047.147 27.487 62.995 62.995 Sun Exposure 0.343 Species x Tide 609.409 3.319 1218.818 Species x Sun 7209.064 14418.128 39.261 Tide x Sun 2873.009 2873.009 15.647 1156.295 2312.590 Species x Tide x Sun 6.297 183.619 Within 17627.392 Student-Newman-Keuls tests: Acanthanucella: HighSun LowSun HighShade LowShade Nucella: HighShade Low Shade HighSun LowSun Ocenebra: LowShade HighShade HighSun LowSun 5.001 .001 559 040 5.001 .001 003 Figure Legends Figure 1: Percent feeding by whelk species, tide height, sun exposure. Percents are daily average proportion of snails feeding at each tide in each sun exposure category. Figure 2: Percent of Acanthanucella punctulata feeding versus time of day when high tide (open circles) or low tide (closed triangles) occurred. Vertical lines indicate time of sunrise and noon. High Tide: P «.001; r* =.92; Low Tide: P -.001, r’=.80 Figure 3: Percent of feeding individuals of each species found on horizontal, vertical, and undercut substrate at high and low tide. Error bars are + 1 standard deviation. Chi-square = 142.64 (df = 4, PS.001) Figure 4: Percent of feeding individuals of each species found on low, medium, and high-density barnacle cover (all tide levels combined). Error bars are + 1 standard deviation. Chi-square = 27.88 (df = 4, Pk.001) Figure 5: Percent of non-feeding individuals of each species found on horizontal, vertical and undercut substrate at high and low tide. Error bars are + 1 standard deviation. Chi-square = 152.20 (df = 4, PS.001) Figure 6: Percent of non-feeding individuals of each species found in crevices vs. open rock faces (all tide levels combined). Error bars are + 1 standard deviation. Chi-square = 84.57 (df = 2, PS.001) Figure 7: Percent of non-feeding individuals of each species found in protective cover (red algal cover or at bases of anemones) vs. -bare open rock. Error bars are + 1 standard deviation. Chi-square = 105.91 (df = 2, PS.001) Figure 8: Percent of Nucella emarginata feeding versus time of day when high tide (open circles) or low tide (closed triangles) occurred. Vertical lines indicate time of sunrise and noon. High Tide: P =.76, r’ =.02, Low Tide: P =.86, r .004 Figure 9: Percent of Ocenebra circumtexta feeding versus time of day when high tide (open circles) or low tide (closed triangles) occurred. Vertical lines indicate time of sunrise and noon. High Tide: P = .35, r’ =.14, Low Tide: P =.03,r 48 Figure 1: 100 D Sun □ Shade 80 60- 40 + 20 0 + Low High Acanthanucella Low High Nucella Low High Ocenebra O Figure 2: 100- 80 High Tide 4 Low Tide 60 2 40 A 20 X O 0 2 4 6 8 10 12 14 16 18 20 22 24 Time of Da Figure 3: 100- 80- 60 - 40- 20 0 + Orientation - Feeding □ Horizontal Undercut Vertical H I I Low High Low High Nucella Ocenebra Low High Acanthanucella 100 80 60 40- 20- Figure 4: Low □ Medium High Acanthanucella Barnacle Density Nucella Ocenebra Figure 5: 100- □ Horizontal Undercut 80 Vertical 60 40 20- 0 + Low High Acanthanucella Orientation - Not Feeding Low High Low High Nucella Ocenebra 100- 80- 60- 40 20- 0- Figure 6: Crevice usage when not feeding D On Face □ In Crevice Acanthanucella Nucella Ocenebra Figure 7: Cover g 100 Face 80- 60- 40- 2 20 Low High Acanthanucella Use of protective cover Low High Low High Nucella Ocenebra LL O Figure 8: 0 High Tide 100 4 Low Tide 80 60 4 —-+4—— 40 20 — O- O 2 4 6 8 10 12 14 16 18 20 22 24 Time of Day S. O Figure 9: 100 80 60 40 20 O a High Tide 4 Low Tide aa 0 2 4 6 8 10 12 14 16 18 20 22 24 Time of Day