Physical and biogeochemical correlates of spatio-temporal variation in the δ13C of marine macroalgae

Carbon isotope ratios (¹³C/¹²C) can be used to trace sources of production supporting food chains, as δ¹³C undergoes relatively small and predictable increases (∼0.5‰) through each trophic level. However, for this technique to be precise, variation in δ¹³C signatures of different sources of production (baseline sources) must be clearly defined and distinct from each other. Despite this, δ¹³C in the primary producers of marine systems are highly variable over space and time, due to the complexity of physical and biogeochemical processes that drive δ¹³C variation at the base of these foodwebs. We measured spatial and temporal variation in the δ¹³C of two species of macroalgae that are important dietary components of grazers over temperate reefs: the small kelpEcklonia radiata, and the red alga Plocamium preissianum, and related any variation to a suite of physical and biogeochemical variables. Patterns in δ¹³C variation, over different spatial (10 s m to 100 km) and temporal scales (weeks to seasons), differed greatly between taxa, but these were partly explained by the δ¹³C of dissolved inorganic carbon (DIC) and light. However, while the δ¹³C in E. radiata was not related to water temperature, a highly significant proportion of the spatio-temporal variation in δ¹³C of P.preissianum was explained by temperature alone. Accordingly, we applied this relationship to project (across temperate Australasia) and forecast (in time, south-western Australia) patterns in P. preissianum δ¹³C. The mean projected δ¹³C for P.preissianum in the study region varied by only ∼1‰ over a 12-month period, compared to ∼3‰ over 2000 km. This illustrates the potential scale in the shift of δ¹³C in baseline food sources over broad scales, and its implications to food web studies. While we show that those relationships differ across taxonomic groups, we recommend developing models to explain variability in δ¹³C of other baseline sources to facilitate the interpretation of variation in δ¹³C of consumers in food webs, particularly where data for baselines are absent over broad scales.