Daniel C. Reed

Coupled dynamics of iron and phosphorus in sediments of an oligotrophic coastal basin and the impact of anaerobic oxidation of methane.

Studies of phosphorus (P) dynamics in surface sediments of lakes and coastal seas typically emphasize the role of coupled iron (Fe), sulfur (S) and P cycling for sediment P burial and release. Here, we show that anaerobic oxidation of methane (AOM) also may impact sediment P cycling in such systems. Using porewater and sediment profiles for sites in an oligotrophic coastal basin (Bothnian Sea), we provide evidence for the formation of Fe-bound P (possibly vivianite; Fe3(PO4)2 .8H2O) below the zone of AOM with sulfate. Here, dissolved Fe2+ released from oxides is no longer scavenged by sulfide and high concentrations of both dissolved Fe2+ (>1 mM) and PO4 in the porewater allow supersaturation with respect to vivianite to be reached. Besides formation of Fe(II)-P, preservation of Fe-oxide bound P likely also contributes to permanent burial of P in Bothnian Sea sediments. Preliminary budget calculations suggest that the burial of Fe-bound P allows these sediments to act as a major sink for P from the adjacent eutrophic Baltic Proper.

Transient tracer dynamics in a lattice-automaton model of bioturbation

Mixing of sediments by benthic fauna represents a dominant transport process of solids in the majority of surficial marine deposits. Short-term mixing rates are often determined by fitting a transient, diffusive model of bioturbation to vertical profiles of introduced particles (e.g., luminophores). Previous field studies adopting this approach have noted that mixing intensity decreases with time and the nature of mixing progresses from advective or nonlocal transport to diffusive mixing. These observations have been attributed to “age-dependent” mixing. The present study employs a lattice-automaton model to investigate the short-term behavior of conservative transient tracers. Simulations demonstrate that despite constant, indiscriminate mixing ofparticles,mixingintensity,asquantifiedbythediffusionmodel,appearsasafunctionoftimedueto themodelbeinginvalidonshorttimescales.Failureofthemodel,however,isnotapparentfromtracer profiles. Furthermore, the errors incurred by misapplying the biodiffusion model can be considerable with estimates of mixing intensity reaching up to 3000% of the actual value. The transition from advective to diffusive mixing is also demonstrated and found to be the product of a boundary effect. Results presented here suggest that, on average, at least nine mixing events are required before the biodiffusionmodelbecomesanappropriatedescriptionofbioturbation.Thus,providingthetimescale of the tracer is sufficiently large relative to the time scale of mixing, the model provides a reasonable description of bioturbation.

Simulated fiddler-crab sediment mixing

Using a lattice-automaton model, we simulate the effects of fiddler crabs on the distribution of excess Pb in marsh sediments. Three previously-identified modes of bioturbation are investigated: (1) removal-and-fill, where material is excavated to the sediment-water interface and burrows, when abandoned, are subsequently filled by surface material, (2) removal-and-collapse, where the infilling occurs by collapse of the burrow walls, and (3) partial-compaction-and-collapse, where part of the excavated sediment is packed into the burrow wall and abandoned burrows subsequently collapse. These various mixing modes lead to somewhat different laterally-integrated Pbex profiles, which are also influenced by burrowing frequency, burrow dimensions, fraction of surface material replaced by new sediment (regeneration), and the fraction of material compacted during burial. Using parameters from a previous study in a South Carolina marsh, we find that data from low-marsh sites are best predicted by the partial-compaction-and-collapse process; this is consistent with the observation that burrow casts indicate far more material is excavated than is deposited as pellets at the sediment-water interface. The profile from the high-marsh site is best simulated by removal-and-fill mixing, with 50% regeneration of material at the sediment-water interface; this is consistent with less frequent flooding at this site. We have also calculated the exchange function for each of these mixing modes and show that they are highly asymmetric, indicating that the mixing is not diffusive. Only in the case of partialcompaction-and-collapse does the exchange function approach a diffusive form when the excavation rate decreases, i.e., the probability of compaction increases.