Long-Term Experimental Manipulation of Atmospheric Sulfate Deposition to a Peatland: Response of Methylmercury and Related Solute Export in Streamwater.

Changes in sulfate (SO42-) deposition have been linked to changes in mercury (Hg) methylation in peatlands and water quality in freshwater catchments. There is little empirical evidence, however, of how quickly methyl-Hg (MeHg, a bioaccumulative neurotoxin) export from catchments might change with declining SO42- deposition. Here, we present responses in total Hg (THg), MeHg, total organic carbon, pH, and SO42- export from a peatland-dominated catchment as a function of changing SO42- deposition in a long-term (1998-2011), whole-ecosystem, control-impact experiment. Annual SO42- deposition to half of a 2-ha peatland was experimentally increased 6-fold over natural levels and then returned to ambient levels in two phases. Sulfate additions led to a 5-fold increase in monthly flow-weighted MeHg concentrations and yields relative to a reference catchment. Once SO42- additions ceased, MeHg concentrations in the outflow streamwater returned to pre-SO42- addition levels within 2 years. The decline in streamwater MeHg was proportional to the change in the peatland area no longer receiving experimental SO42- inputs. Importantly, net demethylation and increased sorption to peat hastened the return of MeHg to baseline levels beyond purely hydrological flushing. Overall, we present clear empirical evidence of rapid and proportionate declines in MeHg export from a peatland-dominated catchment when SO42- deposition declines.

Accumulation of Methylmercury in Invertebrates and Masked Shrews (Sorex cinereus) at an Upland Forest-Peatland Interface in Northern Minnesota, USA.

Mercury (Hg) methylation is often elevated at the terrestrial-peatland interface, but methylmercury (MeHg) production at this "hot spot" has not been linked with in situ biotic accumulation. We examined total Hg and MeHg levels in peat, invertebrates and tissues of the insectivore Sorex cinereus (masked shrew), inhabiting a terrestrial-peatland ecotone in northern Minnesota, USA. Mean MeHg concentrations in S. cinereus (71 ng g-1) fell between concentrations measured in spiders (mean 70-140 ng g-1), and ground beetles and millipedes (mean 29-42 ng g-1). Methylmercury concentrations in S. cinereus increased with age and differed among tissues, with highest concentrations in kidneys and muscle, followed by liver and brain. Nearly all Hg in S. cinereus was in the methylated form. Overall, the high proportional accumulation of MeHg in peat at the site (3.5% total Hg as MeHg) did not lead to particularly elevated concentrations in invertebrates or shrews, which are below values considered a toxicological risk.

Stream invertebrate productivity linked to forest subsidies: 37 stream-years of reference and experimental data.

Riparian habitats provide detrital subsidies of varying quantities and qualities to recipient ecosystems. We used long-term data from three reference streams (covering 24 stream-years) and 13-year whole-stream organic matter manipulations to investigate the influence of terrestrial detrital quantity and quality on benthic invertebrate community structure, abundance, biomass, and secondary production in rockface (RF) and mixed substrates (MS) of forested headwater streams. Using a mesh canopy covering the entire treatment stream, we examined effects of litter ex'clusion, small- and large-wood removal, and addition of artificial wood (PVC) and leaves of varying quality on organic matter standing crops and invertebrate community structure and function. We assessed differences in functional feeding group distribution between substrate types as influenced by organic matter manipulations and long-term patterns of predator and prey production in manipulated vs. reference years. Particulate organic matter standing crops in MS of the treatment stream declined drastically with each successive year of litter exclusion, approaching zero after three years. Monthly invertebrate biomass and annual secondary production was positively related to benthic organic matter in the MS habitats. Rockface habitats exhibited fewer changes than MS habitats across all organic matter manipulations. With leaf addition, the patterns of functional group distribution among MS and RF habitats returned to patterns seen in reference streams. Secondary production per unit organic matter standing crop was greatest for the leaf addition period, followed by the reference streams, and significantly less for the litter exclusion and wood removal periods. These data indicate that the limited organic matter remaining in the stream following litter exclusion and wood removal was more refractory than that in the reference streams, whereas the added leaf material was more labile and readily converted into invertebrate production. Predator production and total production were tightly coupled in reference and treatment streams, indicating strong relationships between predators and their prey. Results from the artificial wood addition demonstrate that physical structure alone will not restore invertebrate productivity without detrital resources from the riparian forest. Our long-term studies conducted over three decades at the ecosystem scale unequivocally show the necessity of maintaining and restoring aquatic-terrestrial linkages in forested headwater streams.

Methylmercury declines in a boreal peatland when experimental sulfate deposition decreases.

Between 2001 and 2008 we experimentally manipulated atmospheric sulfate-loading to a small boreal peatland and monitored the resulting short and long-term changes in methylmercury (MeHg) production. MeHg concentrations and %MeHg (fraction of total-Hg (Hg(T)) present as MeHg) in the porewaters of the experimental treatment reached peak values within a week of sulfate addition and then declined as the added sulfate disappeared. MeHg increased cumulatively over time in the solid-phase peat, which acted as a sink for newly produced MeHg. In 2006 a "recovery" treatment was created by discontinuing sulfate addition to a portion of the experimentally treated section to assess how MeHg production might respond to decreased sulfate loads. Four years after sulfate additions ceased, MeHg concentrations and %MeHg had declined significantly from 2006 values in porewaters and peat, but remained elevated relative to control levels. Mosquito larvae collected from each treatment at the end of the experiment exhibited Hg(T) concentrations reflective of MeHg levels in the peat and porewaters where they were collected. The proportional responses of invertebrate Hg(T) to sulfate deposition rates demonstrate that further controls on sulfur emissions may represent an additional means of mitigating Hg contamination in fish and wildlife across low-sulfur landscapes.

Long-term nutrient enrichment decouples predator and prey production.

Increased nutrient mobilization by human activities represents one of the greatest threats to global ecosystems, but its effects on ecosystem productivity can differ depending on food web structure. When this structure facilitates efficient energy transfers to higher trophic levels, evidence from previous large-scale enrichments suggests that nutrients can stimulate the production of multiple trophic levels. Here we report results from a 5-year continuous nutrient enrichment of a forested stream that increased primary consumer production, but not predator production. Because of strong positive correlations between predator and prey production (evidence of highly efficient trophic transfers) under reference conditions, we originally predicted that nutrient enrichment would stimulate energy flow to higher trophic levels. However, enrichment decoupled this strong positive correlation and produced a nonlinear relationship between predator and prey production. By increasing the dominance of large-bodied predator-resistant prey, nutrient enrichment truncated energy flow to predators and reduced food web efficiency. This unexpected decline in food web efficiency indicates that nutrient enrichment, a ubiquitous threat to aquatic ecosystems, may have unforeseen and unpredictable effects on ecosystem structure and productivity.