Human-driven breakdown of predator-prey interactions in the northern Adriatic Sea

Abstract

Long-term baseline data that allow tracking how predator-prey interactions have responded to intensifying human impacts are often lacking. Here, we assess temporal changes in benthic community composition and interactions between drilling predatory gastropods and their molluscan prey using the Holocene fossil record of the shallow northern Adriatic Sea, which is characterized by a long history of human transformation. Molluscan assemblages differ between the Isonzo and Po prodelta, but both show consistent temporal trends in the abundance of dominant species. Samples of mollusc prey collected at high stratigraphic resolution indicate that drilling frequencies have drastically declined in the Po prodelta since the mid-twentieth century, while a weaker trend in the more condensed sediments of the Isonzo prodelta is not statistically significant. The decrease in drilling predation intensity and the community turnover are linked to the loss of predatory gastropods and the increased relative abundance of less-preferred prey during the most recent decades. Our results align with data showing the substantial depletion of marine resources at higher trophic levels in the region and indicate that the strong simplification of the food web initiated in the late nineteenth century accelerated further since the mid-twentieth century.

Keywords: Mediterranean Sea; conservation palaeobiology; drilling predation; eutrophication; hypoxia; molluscs.

Figure 1.

(a) Location of the sampling sites in the NAS and on the Po plain near Ferrara. Pan = Panzano. The inset shows the right valve of V. gibba with a predatory drill hole. The scale bar on inset represents 1 mm. (b) Median drilling frequencies (DFs) of the total assemblage and (c) median relative abundance of predators in time bins in both regions.

Figure 2.

DFs of the total assemblage in the cores decline in the late twentieth century and later in both regions. Note distinct peaks of DFs in the mid-twentieth century in both regions.

Figure 3.

Changes in abundance of drilling predators and DF in Po cores (data combined for all cores). (a) Vertical (temporal) changes in the relative abundance of drilling predators and DF in total assemblage as a function of depth in the core. Vertical lines represent the mean relative predator abundance and mean DF for data pooled across all levels in the cores. (b) A smoothed trend based on moving average computed as means of successive sets of five adjacent depth levels (core depth based on the middle of the five observations). Vertical lines are the same as in plot a. (c) Resampling distribution of Spearman rho values for rank correlation between the relative abundance of predators and depth in core. The vertical arrow indicates the location of the observed rho value.

Figure 4.

Changes in abundance of drilling predators and DF in Panzano cores (data combined for all cores). (a) Vertical (temporal) changes in the relative abundance of drilling predators and DF in total assemblage as a function of depth in the core. Vertical lines represent the mean relative predator abundance and mean DF for data pooled across all levels in the cores. (b) A smoothed trend based on moving average computed as means of successive sets of five adjacent depth levels (core depth based on the middle of the five observations). Vertical lines are the same as in plot a. (c) Resampling distribution of Spearman rho values for rank correlation between the relative abundance of predators and depth in core. The vertical arrow indicates the location of the observed rho value.

Figure 5.

Temporal change in the species composition of molluscan communities, visualized by NMDS, is associated with temporally decreasing DF for (a) the Po prodelta and (b) the Isonzo prodelta. Size of circles is proportional to DFs.

Figure 6.

Relative abundance and DFs of the four most abundant prey species in (a) the Po prodelta and (b) the Isonzo prodelta.