Tidal Zonation of Periwinkles

How does the abundance of marine molluscs vary across tidal zones?

Periwinkle

Hypothesis

Null: There will be no significant difference in mollusc abundance across the 3 tidal zones. Alternative: Mollusc abundance in the high intertidal zone will be significantly higher than in the low intertidal zone, reflected by significant correlations with abiotic variables.

Results

Fig. 1. PCA of mollusc count and environmental variables. Data are PC1 (Dim1) and PC2 (Dim2) of 6 environmental variables at low (L), middle (M) and high (H) intertidal zones (N=9), during the rainy spring season. L-sites are associated with higher elevation (-0.558) and temperature (-0.470), distinct from H-sites which are associated with increased salinity (0.368), mollusc abundance (0.375), and water depth (0.386). Quadrant overlap of M-sites suggests environmental heterogeneity can explain variance in mollusc abundance. PC2 lacks distinct zone clusters; variance may result from temperature (-0.410), pH (+0.647) and salinity (-0.532) interactions via evaporation.
Fig. 1 PCA

Fig 2. Differences and correlations between mollusc abundance and environmental variables. Matrix shows Spearman’s ρ between 6 variables (right) and scatterplots displaying predicted linear models (left). Asterisks represent significant correlations or differences ( ॱ = approaching significance, * = p<.05, ** = p<.005, *** = p<.001). Histograms represent sample distribution of each variable (N=27). Data in the leftmost column are the median and IQR of variables in low (L), middle (M) and high (H) intertidal zones (N=9).
Fig. 2 Correlations + Boxplot Matrix

Table 1. Shapiro Tests for Normality Salinity, mollusc abundance and elevation were normally distributed, so lms were used to assess correlations and parametric tests were used to assess zonation differences. Water temperature, depth, and pH were not normally distributed so glms and non-parametric tests were subsequently used for the same purposes.
Table 1. Shapiro tests

Table 2. Post-hoc analysis for pairwise zonation differences After significant initial model tests, post-hocs found significant differences between intertidal zones and water temp, depth, and elevation, as seen in Fig. 2. No significant zonation was observed in mollusc abundance, salinity, or pH.
Table 2. Post Hocs

Table 3. Poisson GLMs of mollusc count and abiotic vars Significance, strength, and direction of associations are denoted above (residual deviance = 293.05, 20df, AIC = 456.27).
Table 3. Poisson GLMs of mollusc count and abiotic vars

Fig. 3. Correlations with mollusc abundance Regression analysis of linear and exponential models revealed correlations between elevation, salinity, and mollusc count. Elevation explained 29.63% of variance and was significantly negatively correlated with mollusc count (S = 4200, rho = -.436, p = 0.02704). Although salinity explained significant variance (F(1;24) = 5.345, R2 adj = 0.1481, p = 0.02968) in mollusc count, no significant correlation was identified. Salinity significantly decreases along the elevation gradient [3A] due to less saltwater exposure, as does mollusc count [3C] so the observed relationship in [3B] may not reflect salinity preferences.
Table 3. Poisson GLMs of mollusc count and abiotic vars

Discussion

The alternative hypothesis was not supported by our model, as there was no significant zonation of mollusc abundance [Table 2]. This is likely due to landscape heterogeneity [Fig. 1]. However, H-sites were environmentally distinct from L-sites; the former significantly colder, deeper, and lower in altitude than the latter [Fig. 2, Table 2.]. GLMs found a significant negative correlation between mollusc abundance and temperature [Table 3.] and regression analysis revealed a significant negative correlation between abundance and elevation, indicating H-sites were preferential [Fig. 3]. Although species were not identified, the vast majority were periwinkles (Littorinidae) which are found to dominate high intertidal zones (Wickramasinghe et al., 2021). However, research finds they are absent in subtidal beds where submersion is permanent (Saier, 2000), and actively select the mid-intertidal (Apolinario et al., 1999), contradicting our findings. This suggests a bell curve gradient should be found; quadratic regression may have been more appropriate than linear models in this case. Periwinkle abundance was measured by shell count, but this does not indicate organism presence. Future experiments should check for a foot to confirm presence. Moreover, existing literature suggests rock characteristics affect periwinkle density, which was not integrated in our models. Rugosity and % bare rock were positively correlated with abundance (Carlson et al., 2006), indicating preferences for rough surfaces onto which they can firmly latch, withstanding waves in the middle-low intertidal zones. Our experiment highlights how mollusc zonation is impacted by landscape heterogeneity during low tide, where intertidal zones are exposed to abiotic changes (e.g. temperature, salinity, water depth). Further research should assess marine molluscs’ microhabitat preferences to comprehensively predict how population densities may shift under the growing threat of climate change, so preemptive decisions can be made regarding their conservation (eg. artificial rock introduction to increase pooled habitat availability in the mid intertidal zone).

References

Apolinario, M., Coutinho, R. and Baeta-Neves, M.H. (1999). Periwinkle (Gastropoda: Littorinidae) habitat selection and its impact upon microalgal populations. Revista Brasileira de Biologia, 59(2), pp.211–218. doi:https://doi.org/10.1590/s0034-71081999000200005.

Carlson, R.L., Shulman, M.J. and Ellis, J.C. (2006). Factors Contributing to Spatial Heterogeneity in the Abundance of the Common Periwinkle Littorina Littorea (.L). Journal of Molluscan Studies, 72(2), pp.149–156. doi:https://doi.org/10.1093/mollus/eyi059.

Saier, B. (2000). Age-dependent zonation of the periwinkle Littorina littorea (L.) in the Wadden Sea. Helgoland Marine Research, 54(4), pp.224–229. doi:https://doi.org/10.1007/s101520000054.

Wickramasinghe, M.P., Sudarshani, K.A.M. and Wegiriya, H.C.E. (2021). The diversity of marine invertebrate macrofauna in selected rocky intertidal zones of Matara, Sri Lanka. Asian Journal of Conservation Biology, 10(1), pp.15–21. doi:https://doi.org/10.53562/ajcb.ozdk5526.