Jeffrey R. White

A three-year study of controls on methane emissions from two Michigan peatlands

We investigate temporal changes in methane emissions over a three-year period from two peatlands in Michigan. Mean daily fluxes ranged from 0.6–68.4 mg CH4 m−2d−1 in plant communities dominated by Chamaedaphne calyculata, an eficaceous shrub, to 11.5–209 mg CH4 m−2d−1 in areas dominated by plants with aerenchymatous tissues, such as Carex oligosperma and Scheuchzeria palustris. Correlations between methane flux and water table position were significant at all sites for one annual cycle when water table fluctuations ranged from 15 cm above to 50 cm below the peat surface. Correlations were not significant during the second and third annual periods with smaller water table fluctuations. Methane flux was strongly correlated with peat temperatures at −5 to −40 cm (rs = 0.82 to 0.98) for all three years at sites with flora acting as conduits for methane transport. At shrub sites, the correlations between methane flux and peat temperature were weak to not significant during the first two years, but were strong in the third year.Low rates of methane consumption (−0.2 to −1.5 mg CH4 m−2 d−1 ) were observed at shrub sites when the water table was below −20 cm, while sites with plants capable of methane transport always had positive net fluxes of methane. The methane oxidizing potential at both types of sites was confirmed by peat core experiments. The results of this study indicate that methane emissions occur at rates that cannot be explained by diffusion alone; plant communities play a significant role in altering methane flux from peatland ecosystems by directly transporting methane from anaerobic peat to the atmosphere.

A synthesis of methane emissions from 71 northern, temperate, and subtropical wetlands

Wetlands are the largest natural source of atmospheric methane. Here, we assess controls on methane flux using a database of approximately 19 000 instantaneous measurements from 71 wetland sites located across subtropical, temperate, and northern high latitude regions. Our analyses confirm general controls on wetland methane emissions from soil temperature, water table, and vegetation, but also show that these relationships are modified depending on wetland type (bog, fen, or swamp), region (subarctic to temperate), and disturbance. Fen methane flux was more sensitive to vegetation and less sensitive to temperature than bog or swamp fluxes. The optimal water table for methane flux was consistently below the peat surface in bogs, close to the peat surface in poor fens, and above the peat surface in rich fens. However, the largest flux in bogs occurred when dry 30-day averaged antecedent conditions were followed by wet conditions, while in fens and swamps, the largest flux occurred when both 30-day averaged antecedent and current conditions were wet. Drained wetlands exhibited distinct characteristics, e.g. the absence of large flux following wet and warm conditions, suggesting that the same functional relationships between methane flux and environmental conditions cannot be used across pristine and disturbed wetlands. Together, our results suggest that water table and temperature are dominant controls on methane flux in pristine bogs and swamps, while other processes, such as vascular transport in pristine fens, have the potential to partially override the effect of these controls in other wetland types. Because wetland types vary in methane emissions and have distinct controls, these ecosystems need to be considered separately to yield reliable estimates of global wetland methane release.

A process‐based model to derive methane emissions from natural wetlands

A process-based model has been developed in order to calculate methane emissions from natural wetlands as a function of the hydrologic and thermal conditions in the soil. The considered processes in the model are methane production, methane consumption and transport of methane by diffusion, ebullition and through plants. The model has been tested against data from a three-year field study from a Michigan peatland. The interannual and seasonal variations of the modelled methane emissions and methane concentration profiles are in good agreement with the observations. During the growing season the main emission pathway proceeds through plants. Ebullition occurs whenever the water table is above the soil surface, while diffusion is only significant in the first 15 days after a drop of the water table below the peat surface.

Paleoecological investigation of recent lake acidification in the Adirondack Mountains, N.Y.

Paleoecological analysis of the sediment record of 12 Adirondack lakes reveals that the 8 clearwater lakes with current pH < 5.5 and alkalinity < 10 μeq l-1 have acidified recently. The onset of this acidification occurred between 1920 and 1970. Loss of alkalinity, based on quanitative analysis of diatom assemblages, ranged from 2 to 35 μeq l-1. The acidification trends are substantiated by several lines of evidence including stratigraphies of diatom, chrysophyte, chironomid, and cladoceran remains, Ca:Ti and Mn:Ti ratios, sequentially extracted forms of Al, and historical fish data. Acidification trends appear to be continuing in some lakes, despite reductions in atmospheric sulfur loading that began in the early 1970s. The primary cause of the acidification trend is clearly increased atmospheric deposition of strong acids derived from the combustion of fossil fuels. Natural processes and watershed disturbances cannot account for the changes in water chemistry that have occurred, but they may play a role. Sediment core profiles of Pb, Cu, V, Zn, S, polycyclic aromatic hydrocarbons, magnetic particles, and coal and oil soot provide a clear record of increased atmospheric input of materials associated with the combustion of fossil fuels beginning in the late 1800s and early 1900s. The primary evidence for acidification occurs after that period, and the pattern of water chemistry response to increased acid inputs is consistent with current understanding of lake-watershed acidification processes.

Effect of seasonal changes in the pathways of methanogenesis on the δ13C values of pore water methane in a Michigan peatland

The δ13C value of pore water methane produced in a Michigan peatland varied by 11‰ during the year. This isotopic shift resulted from large seasonal changes in the pathways of methane production. On the basis of mass balance calculations, the δ13C value of methane from CO2 reduction (average = −71.4 ± 1.8‰) was depleted in 13C compared to that produced from acetate (−44.4 ± 8.2‰). The dissolved methane at the site remained heavy (approximately −51‰) during most of the year. Tracer experiments using 14C-labeled CO2 indicated that during January 110 ± 25% of the methane was produced by CO2 reduction. Because of low-methane production rates during the winter, this 13C-depleted methane had only a slight effect on the isotopic composition of the methane pool. In early spring when peat temperatures and methane production rates increased, the δ13C value of the dissolved methane in shallow peat was influenced by the isotopically light methane and approached −61‰. Peat incubation experiments conducted at 15°C in May and June (when the peat reaches its maximum temperature) indicated that an average of 84 ± 9% of the methane production was from acetate and had an average δ13C value of −48.7 ± 5.6‰. Rising acetate concentrations during April-May (approaching 1 mmol L−1(mM)) followed by a rapid decrease in acetate concentrations during May-June reflected the shift toward methane production dominated by acetate fermentation. During this period, dissolved methane in shallow peat at the site returned to heavier values (approximately −51‰) similar to that produced in the incubation experiments.

Lead cycling in an acidic Adirondack Lake

Acidification of surface waters by acidic deposition has reportedly caused increased concentrations of Zn. Elevated concentrations of Zn are of interest due to potential toxicity to aquatic organisms. Researchers have also examined the chronology of Zn deposition to lake sediments as an indicator of changes in atmospheric deposition. However, there have been few studies that focus on Zn chemistry and transport in acidic lakes. Elevated concentrations of Zn (0.15-1.2 ..mu..mol L/sup -1/; mean = 0.39 ..mu..mol L/sup -1/) were observed in acidic Darts Lake. Although peak concentrations of 1.2 ..mu..mol of Zn L/sup -1/ were observed during snowmelt, spatial and temporal variations in Zn concentration were minor and limited to melt water in streams and at the lake surface. On the basis of mass balance calculations and sediment trap observations, Zn did not appear to be significantly retained in Darts Lake. Long- and short-term variations in in-lake retention of Zn, caused by surface water acidification, may complicate quantitative interpretation of Zn deposition in sediments. 41 references, 5 figures, 1 table.

Paleolimnological evidence for recent acidification of Big Moose Lake, Adirondack Mountains, N.Y. (USA)

Big Moose L. has become significantly more acidic since the 1950s, based on paleolimnological analyses of sediment cores. Reconstruction of past lakewater pH using diatom assemblage data indicates that from prior to 1800 to ca. 1950, lakewater pH was about 5.8. After the mid-1950s, the inferred pH decreased steadily and relatively quickly to about 4.6. Alkalinity reconstructions indicate a decrease of about 30 μeq · l-1 during the same period. There was a major shift in diatom assemblage composition, including a nearly total loss of euplanktonic taxa. Chrysophyte scale assemblages and chironomid (midge larvae remains also changed in a pattern indicating decreasing lakewater pH starting in the 1950s. Accumulation rates of total Ca, exchangeable and oxide Al, and other metals suggest recent lake-watershed acidification. Cores were dated using210Pb, pollen, and charcoal. Indicators of watershed change (deposition rates of Ti, Si, Al) do not suggest any major erosional events resulting from fires or logging. Accumulation rates of materials associated with combustion of fossil fuels (polycyclic aromatic hydrocarbons, coal and oil soot particles, some trace metals, and sulfur) are low until the late 1800s-early 1900s and increase relatively rapidly until the 1920s–1930s. Peak rates occurred between the late 1940s and about 1970, when rates declined.The recent decrease in pH of Big Moose L. cannot be accounted for by natural acidification or processes associated with watershed disturbance. The magnitude, rate and timing of the recent pH and alkalinity decreases, and their relationship to indicators of coal and oil combustion, indicate that the most reasonable explanation for the recent acidification is increased atmospheric deposition of strong acids derived from combustion of fossil fuels.

Controls on methane production in a tidal freshwater estuary and a peatland: methane production via acetate fermentation and CO2 reduction

Rates of total methane production, acetate fermentation andCO2 reduction were compared for two different wetland sites. On aper-liter basis, sediments from the White Oak River estuary, a tidal freshwatersite in eastern North Carolina, had an annual methane production rate (53.3mM yr−1) an order of magnitude higher thanthat ofBuck Hollow Bog (5.5 mM yr−1), a peatlandinMichigan. Methane was produced in the White Oak River site on an annual basisbyboth acetate fermentation (72%) and CO2 reduction (28%) in a ratiotypical of freshwater methanogenic sites. Competition for acetate bynon-methanogenic microorganisms in Buck Hollow peat limited methane productionfrom acetate to only a few months a year, severely impacting annual methaneproduction rates. However, when acetate was available to the methanogens in thepeat during early spring, the percentage of methane production from acetatefermentation (84%) and CO2 reduction (16%) and rates of totalmethaneproduction were similar to those of the White Oak River sediments at the sametemperature. Rates of CO2 reduction and acetate fermentationconducted at both sites at various temperatures showed that Buck Hollow peatmethane production was also limited by a colder temperature regime as well asdifferences in the response of the CO2 reducing and aceticlasticmethanogens to temperature variations.

The selectivity of a sequential extraction procedure for the determination of iron oxyhydroxides andiron sulfides in lake sediments

A popular sequential extraction procedure (Tessier et al. 1979) designed t o extract metals partitioned in various sediment phases, was evaluated for its selectivity. Amorphous FeOOH, FeS, and FeS2 were added separately to natural lake sediments and sequentially extracted. The selectivity of the sequential procedure for the added solid phases was evaluated by determining the difference in the mass of Fe extracted from treated and control sediments. In the experiments where sulfide minerals were added, total S was measured in the residual solids in order to confirm selectivity of the method. Concentrations of total carbon remaining in the solid phase after each extraction step were also measured to determine the selectivity of the sequential procedure for carbon.The procedure was moderately selective for Fe added as FeOOH; a mean of 77 ± 12% (p < 0.05) of the Fe added was extracted in the step designed to reduce Fe-Mn oxyhydroxides. In experiments where FeS was added, a mean of 69 ± 11% (p < 0.05) of the Fe added as FeS was extracted in the fraction designed to oxidize sulfides and organic matter. Approximately 25% of the Fe added as FeS may have been extracted prematurely. Although less precise, total S analyses confirmed that much of the FeS was extracted in the oxidation step, yielding 104 ± 87% (p < 0.05) of the S added as FeS. The procedure was highly selective for FeS2; 92 ± 14% (p < 0.05) of the Fe added as pyrite was extracted in the sulfide extraction step. Extraction of 80 ± 54% (p < 0.05) of S added as pyrite confirmed that FeS2 were selectively extracted in the sulfide extraction step. Carbon in the sediments was also selectively extracted in the oxidation step (77 ± 2.4% of total C; p < 0.05). The applications and limitations of sequential extraction procedures as limnological research tools are discussed in light of our results.

Sediment biogeochemistry of iron and sulfur in an acidic lake

Abstract The biogeochemical cycling of Fe and S in the sediments of acidic Big Moose Lake, Adirondack Park, NY, U.S.A., was investigated. Sediment cores and porewater samples were collected along a depth transect from the hypolimnion (24 m water depth), metalimnion (17 m), and epilimion (8, 12 m). Four vertically distinguishable zones in the sediment environment were observed at each site: 1. 1) NO3− disappearance near the sediment/water interface; 2. 2) accumulation of solid-phase Fe in the top 5 cm; 3. 3) coincident accumulation of chromium reducible sulfur (CRS), disappearance of SO4− and minima in C:S and δ34S at or slightly below maxima in oxide-bound Fe; and 4. 4) homogenous background concentrations in S and Fe below 30 and 10 cm, respectively. Iron and sulfur accumulations in the upper 10 cm occurred at the same depths in cores of different ages, indicating that diagenetic rather than depositional processes played a dominant role in determining near-surface Fe and S profiles. Although sediment accumulation rates varied along the transect, the four zones were located at similar depths in the sediment column at all sites. Diagenetic processes play a major role in the development of these features in Big Moose Lake. The extent of Fe enrichment was considerably greater in the metalimnetic and epilimnetic sediments. In contrast, concentrations of Fe were lower in hypolimnetic sediments, which appeared to be losing Fe to the water column. At all sites, increases in total S from a background concentration of 60–80 μmol g−1 of dry mass occurred in sediments dated approximately 1850. More recent increases to 560 μmol Sg−1 occurred asynchronously in the cores, indicating an important role of sulfate reduction in adding S to sediments. However, organic sulfur accumulation accounted for 22–56% of the recent increase in sediment S. Concentrations of iron and organic carbon were high in lake sediments and probably do not limit S storage in this lake.