Detecting Subtle Shifts in Ecosystem Functioning in a Dynamic Estuarine Environment

Abstract

Identifying the effects of stressors before they impact ecosystem functioning can be challenging in dynamic, heterogeneous 'real-world' ecosystems. In aquatic systems, for example, reductions in water clarity can limit the light available for photosynthesis, with knock-on consequences for secondary consumers, though in naturally turbid wave-swept estuaries, detecting the effects of elevated turbidity can be difficult. The objective of this study was to investigate the effects of shading on ecosystem functions mediated by sandflat primary producers (microphytobenthos) and deep-dwelling surface-feeding macrofauna (Macomona liliana; Bivalvia, Veneroida, Tellinidae). Shade cloths (which reduced incident light intensity by ~80%) were deployed on an exposed, intertidal sandflat to experimentally stress the microphytobenthic community associated with the sediment surface. After 13 weeks, sediment properties, macrofauna and fluxes of oxygen and inorganic nutrients across the sediment-water interface were measured. A multivariate metric of ecosystem function (MF) was generated by combining flux-based response variables, and distance-based linear models were used to determine shifts in the drivers of ecosystem function between non-shaded and shaded plots. No significant differences in MF or in the constituent ecosystem function variables were detected between the shaded and non-shaded plots. However, shading reduced the total explained variation in MF (from 64% in non-shaded plots to 15% in shaded plots) and affected the relative influence of M. liliana and other explanatory variables on MF. This suggests that although shade stress may shift the drivers of ecosystem functioning (consistent with earlier investigations of shading effects on sandflat interaction networks), ecosystem functions appear to have a degree of resilience to those changes.

Fig 1. Relationships between M. liliana and MPB in non-shaded (open circles) and shaded (black circles) plots.

Fig 2. Changes in the abundances of large (> 20 mm) bivalves (middle panel), bivalves including all size classes measured in cores (middle panel) and other macrofauna (four polychaetes and one amphipod, bottom panel) between non-shaded (open) and shaded (grey bars) treatments.

Asterisk denotes significance treatment effects (unpaired t-tests) at p <0.05 (*), <0.01 (**) and <0.001 (***).

Fig 3. PCA ordination showing changes in macrofauna community composition between non-shaded and shaded treatments.

All 51 species were included in analysis. Vector overlays show nine of the most influential species (At = Aonides trifida, As = Austrovenus stutchburyi, Ml = Macomona liliana, Na = Nicon aestuariensis, Op = Orbinia papilosa, Pa = Paphies australis, Pn = Paracalliope novizealandiae, Sc = Scolelepis sp., Ss = Soletellina siliquens, the rest omitted to improve figure clarity).

Fig 4. PCA ordination illustrates little effect of treatment (shaded versus non-shaded) on rates of individual ecosystems functions.

Vector overlays denote different proxies of ecosystem function (all listed in Table 1 were included in the analysis, though only a selection were presented on this figure for reasons of clarity).