Variation in hydrologic connectivity as a result of microtopography explained by discharge to catchment size relationship

In hydrological terms, raised bogs are often approximated by simple models as in the acrotelm–catotelm concept. However, raised bogs are often characterized by a pronounced surface topography, causing large changes in connectivity of contributing areas on the bog. In this study, daily regression of measured discharges versus catchment areas is used to quantify the impact of surface topography on catchment connectivity within a raised bog. The resulting coefficient of determination shows the strength of the relationship between the discharge and catchment area over time under different hydrological conditions. Monitoring of discharge, water table, transmissivity, and basic weather data on a raised bog (1.9 km²) in eastern central Estonia took place from May 2008 to June 2010. Contributing areas, calculated based on the outlet's discharge volume (V _Q ) divided by the net precipitation volume ( ), of the outlet containing the central pool‐ridge system varied between 1×10^?3 and 0.7 km², suggesting significant differences in connectivity between hydrological events. Correlation between discharge and theoretical catchment size was high (R ²>0.75) when the water table was close to the surface (less than 5 cm below peat surface), and consequently, transmissivities were also high (up to 1,030m²d^?1), which led to connectivity of local storage elements, such as pools and hollows. However, a water table below this threshold resulted in large parts of the catchment being disconnected. The importance of water table depths on catchment connectivity suggests the need to reconsider the hydrological concept of raised bogs; to incorporate these shallow flow components and better understand residence time and consequently transport of solutes, such as DOC, from patterned peatlands.     

Modelling two‐dimensional steady‐state groundwater flow and flow sensitivity to boundary conditions in blanket peat complexes

This study used a two‐dimensional steady‐state finite‐element groundwater flow model to simulate groundwater flow in two Newfoundland blanket peat complexes and to examine flow system sensitivity to changes in water table recharge and aquifer properties. The modelling results were examined within the context of peat‐forming processes in the two complexes. Modelled flow compared favourably with observed flow. The sensitivity analyses suggested that more highly decomposed bog peat along bog margins probably has/had a positive impact on net peat accumulation within bog interiors. Peat with lower hydraulic conductivity along bog margins effectively impedes lateral drainage, localizes water table drawdown to extreme bog margins, and elevates water tables along bog interiors. Peat formation and elevated water tables in adjacent poor fens/laggs currently rely on placic and ortstein horizons impeding vertical drainage and water flow inputs from adjacent bogs. Modest reductions in atmospheric recharge were found to govern bog‐flow‐system geometries in a way that would adversely affect paludification processes in adjacent fens/laggs. Copyright © 2004 John Wiley & Sons, Ltd.     

Towards a conceptual model of hydrological change on an abandoned cutover bog,Quebec

Cutover bogs do not return to functional peatland ecosystems after abandonment because re‐establishment of peat‐forming mosses is poor. This paper presents a conceptual model of bog disturbance caused by peat harvesting (1942–1972), and the hydrological evolution that occurred after abandonment (1973–1998). Two adjacent bogs of similar size and origin, one harvested and the other essentially undisturbed, provide the basis for understanding what changes occurred. The model is based on historical trends evident from previous surveys of land‐use, bog ecology and resource mapping; and from recent hydrological and ecological data that characterize the current condition. Water balance data and historical information suggest that runoff increased and evapotranspiration decreased following drainage, but tended towards pre‐disturbance levels following abandonment, as vegetation recolonized the surface and drainage became less efficient over time. Dewatering of soil pores after drainage caused shrinkage and oxidation of the peat and surface subsidence of approximately 80 cm over 57 years. Comparisons with a nearby natural bog suggest that bulk density in the upper 50 cm of cutover peat increased from 0·07 to 0·13 g cm^?3, specific yield declined from 0·14 to 0·07, water table fluctuations were 67% greater, and mean saturated hydraulic conductivity declined from 4·1 × 10^?5 to 1·3 × 10^?5 cm s^?1. More than 25 years after abandonment, Sphagnum mosses were distributed over broad areas but covered less than 15% of the surface. Areas with ‘good’ Sphagnum regeneration (>10% cover) were strongly correlated with high water tables (mean ?22 cm), especially in zones of seasonal groundwater discharge, artefacts of the extraction history. Forest cover expanded from 5 to 20% of the study area following abandonment. The effect of forest growth (transpiration and interception) and drainage on lowering water levels eventually will be countered by slower water movement through the increasingly dense soil, and by natural ditch deterioration. However, without management intervention, full re‐establishment of natural hydrological functions will take a very long time. Copyright © 2002 John Wiley & Sons, Ltd.     

HYDROLOGY AND MICROCLIMATE OF A PARTLY RESTORED CUTOVER BOG,QUÉBEC

Peatlands do not readily return to functional wetland ecosystems after harvesting (cutting), because the harsh hydrological and microclimatic conditions are unsuitable for Sphagnum regeneration. In this study, drainage ditches blocked after harvesting restored the water balance to a condition similar to a nearby natural bog. Evaporation averaged 2.9 and 2.7 mm day⁻¹ on the cutover and natural bog, respectively. Evaporation consumed most of the rainfall input (86 and 80%, respectively), whereas runoff was minor at both sites (6 and 4%, respectively). However, the water table position was markedly different at these sites. Median water table depth was 0.05 m below the surface in the natural bog, compared with 0.44 m in the cutover bog (ditches blocked). Changes to the peak soil matrix owing to drainage and cutting reduced the specific yield (Sy) of the peat to 0.04–0.06 from 0.35–0.55, causing exaggerated water table changes in the cutover site. Nevertheless, volumetric soil moisture in the cutover site (0.67 ± .08) had low variability, and was maintained above moisture contents found in Sphagnum hummocks in the natural bog (0.48 ± .10), although less than on Sphagnum lawn (0.84 ± .11). Poor Sphagnum regeneration on cutover surfaces can therefore be attributed to its inability to extract water from the underlying peat, which retains water at matric suction greater than the non-vascular Sphagnum can generate. The corrupted iron pan under main ditches has permitted partial recharge of the underlying aquifer, reducing local hydraulic gradients, thereby decreasing vertical seepage loss.     

The impact of peat harvesting and natural regeneration on the water balance of an abandoned cutover bog,Quebec

Harvested sites rarely return to functional ecosystems after abandonment because drainage and peat extraction lower the water table and expose relatively decomposed peat, which is hydrologically unsuitable for Sphagnum moss re‐establishment. Some natural regeneration of Sphagnum has occurred in isolated pockets on traditionally harvested (block‐cut) sites, for reasons that are poorly understood, but are related to natural functions that regulate runoff and evaporation. This study evaluates the water balance of a naturally regenerated cutover bog and compares it with a nearby natural bog of similar size and origin, near Riviere du Loup, Quebec. Water balance results indicated that evapotranspiration was the major water loss from the harvested bog, comprising 92 and 84% of total outputs (2·9 mm day^?1) during the 1997 and 1998 seasons, respectively. Despite denser tree cover at the harvested site, evapotranspiration from the natural bog was similar, although less spatially variable. At the harvested site, evaporative losses ranged from 1·9 mm day^?1 on raised baulks and roads to 3·6 mm day^?1 from moist surfaces with Sphagnum. Although about half of the ditches were inactive or operating at only a fraction of their original efficiency, runoff was still significant at 12 and 24% of precipitation during the 1997 and 1998 study seasons, respectively. This compares with negligible rates of runoff at the natural bog. Thus the cutover bog, although abandoned over 25 years ago, has not regained its hydrological function. This is both a cause and effect of its inability to support renewed Sphagnum regeneration. Without suitable management (e.g. blocking ditches), this site is not likely to improve for a very long time. Copyright © 2001 John Wiley & Sons, Ltd.     

Application of microcosm experiments for quantifying lateral flow and evapotranspiration on recovering bog ecotypes

The importance of characterizing the ecohydrological interactions in natural, damaged/drained, and restored bogs is underscored by the importance of peatlands to global climate change and the growing need for peatland restoration. An understudied aspect of peatland ecohydrology is how shallow lateral flow impacts local hydrological conditions and water balance, which are critical for peatland restoration success. A novel method is presented using microcosms installed in the field to understand the dynamics of shallow lateral flow. Analysis of the difference in water table fluctuation inside and outside the microcosm experimental areas allowed the water balance to be constrained and the calculation of lateral flow and evapotranspiration. As an initial demonstration of this method, a series of four microcosm experiments were set up in locations with differing ecological quality and land management histories, on a raised bog complex in the midlands of Ireland. The timing and magnitude of the lateral flow differed considerably between locations with differing ecological conditions, indicating that shallow lateral flow is an important determining factor in the ecohydrological trajectory of a recovering bog system. For locations where Sphagnum spp. moss layer was present, a slow continuous net lateral input of water from the upstream catchment area supported the water table during drought periods, which was not observed in locations lacking Sphagnum. Consistent with other studies, evapotranspiration was greater in locations with a Spaghnum moss layer than in locations with a surface of peat soil.     

Heuristic numerical and analytical models of the hydrologic controls over vertical solute transport in a domed peat bog,Jura Mountains,Switzerland

We report the results of numerical and analytical simulations to test the hypothesis that downward vertical flow of porewater from the crests of domed alpine and kettle bogs controls vertical porewater distributions of major solutes such as Ca and Mg. The domed Etang de la Gruère bog (EGr), Switzerland, characterized by a vertical downward gradient of 0·04 and stratified layers of peat, is chosen as a field site for the model calibration and evaluation. The middle 4‐m section of the 6·5 m thick bog peat is heavily humified and has a hydraulic conductivity of ～10^?5·6 cm s^?1. Above and below, peat is less humified with a hydraulic conductivity of ～10^?3 cm s^?1. Heuristic finite difference simulations, using Visual MODFLOW, of the bog hydraulics show that the higher conductivity peat at the bog base is critical to create the observed deep, local flow cells that substantively recharge porewater. Model results and Peclet number calculations show that before ～7000 ¹⁴C yr BP diffusion of solutes from underlying mineral soils controlled the vertical distribution of porewater chemistry. From 7000 to ～1250 ¹⁴C BP the porewater chemistry was probably controlled by both upward diffusion and downward advection, and after ～1250 ¹⁴C yr BP porewater chemistry was probably controlled by downward advection. Concentrations of conservative major solutes in the porewaters of alpine, ombrotrophic bogs are the net effect of both downward vertical porewater movement and upward vertical diffusion, the magnitudes of which are delicately poised to the configuration of the bog water table over time and subsurface peat stratigraphy. Copyright © 2002 John Wiley & Sons, Ltd.     

Changes in the discharge pattern of a cutover raised bog during rewetting

Although rewetting practices have been implemented in several cutover raised bogs, their effects on discharge patterns have not received much attention. In 1983, after peat extraction had stopped in the Leegmoor, a cutover raised bog in Germany, an experimental area was set up in order to monitor rewetting and subsequent restoration. Discharge was measured since 1984. The results suggest that rewetted raised bogs may not be effective as regulators of stream flow. An increase of low flows during a decade of rewetting could partly be attributed to weather variation. A comparison of the Leegmoor with a nearby control catchment produced similar results. A shift in the relationship between discharge and groundwater level over time was recorded. This is probably a result of the development of vegetation and a corresponding reduction of the area of open water, leading to an increase of resistance to water flow. The Leegmoor data demonstrate that the daily discharge pattern is most flashy immediately after the start of rewetting and will gradually smooth. Copyright © 1999 John Wiley & Sons, Ltd.     

Microtopographical and hydrophysical controls on subsurface flow and solute transport: A continuous solute release experiment in a subarctic bog

Resource extraction and transportation activities in subarctic Canada can result in the unintentional release of contaminants into the surrounding peatlands. In the event of a release, a thorough understanding of solute transport within the saturated zone is necessary to predict plume fate and the potential impacts on peatland ecosystems. To better characterize contaminant transport in these systems, approximately 13,000 L/day of sodium chloride tracer (200 mg/L) was released into a bog in the James Bay Lowland. The tracer was pumped into a fully penetrating well (1.5 m) between July 5 and August 18, 2015. Horizontal and vertical plume development was measured via in situ specific conductance and water table depth from an adaptive monitoring network. Over the spill period, the bulk of the plume travelled a lateral distance of 100 m in the direction of the slight regional groundwater and topographical slope. The plume shape was irregular and followed the hollows, indicating preferential flow paths due to the site microtopography. Saturated transport of the tracer occurred primarily at ~25 cm below ground surface (bgs), and at a discontinuous high hydraulic conductivity layer ~125 cm bgs due to a complex and heterogeneous vertical hydraulic conductivity profile. Plume measurement was confounded by a large amount of precipitation (233 mm over the study period) that temporarily diluted the tracer in the highly conductive upper peat layer. Longitudinal solute advection can be approximated using local water table information (i.e., depth and gradient); microtopography; and meteorological conditions. Vertical distribution of solute within the peat profile is far more complex due to the heterogeneous subsurface; characterization would be aided by a detailed understanding of the site‐specific peat profile; the degree of decomposition; and the type of contaminant (e.g., reactive/nonreactive). The results of this research highlight the difficulty of tracking a contaminant spill in bogs and provide a benchmark for the characterization of the short‐term fate of a plume in these complex systems.     

Spatio‐temporal changes in bog pool bottom topography – temperature effect and its influence on pool development: an example from a raised bog in Estonia

Increases in pool water and peat temperature in summer accelerate peat decomposition and production of biogenic gases, which can be trapped in peat pores and cause oscillation of peatland surfaces and the rise of peat from the bottom of bog pools. Associated changes in peat water conductivity, holding capacity and transpiration also affect bog hydrology. Our multi‐year study is the first to show in detail the extent and dynamics of changes in bog pool depth and bottom topography associated with changes in temperature, peat type and other factors. The true seasonal rise of peat from the pool bottom begins once the water temperature at the pool bottom exceeds 13–14 °C, although the speed and extent of the rise depends on peat properties, making the rise more erratic than its subsequent descent. The more rapid descent occurs after the first large drop in the temperature of the pool's surface water at the end of summer, resulting from the combination of reduced methane production and increased gas solubility with less influence by peat properties. Much higher dissolved organic carbon concentrations (216 ± 26 mg l^?1) in the pore water of peat risen from the bottom to the pool surface compared with that in the same type of peat at the pool bottom (62 ± 20 mg l^?1) indicate an acceleration of peat decomposition at the warmer pool surface. We show the extent and character of changes in pool depth and bottom topography and how annual differences relate to temperature. Only a few degrees' increase in pool water temperature could induce the pool bottom to rise faster and more extensively for a longer period and enhance decomposition in the peat at the pool surface. This should be evaluated in greater detail to assess the effects of temperature increase on the carbon budget and hydrology of peatlands. Copyright © 2012 John Wiley & Sons, Ltd.     

Hydrogeologic controls on peatland development in the Malloryville Wetland, New York (USA)

The Malloryville Wetland Complex, a small kettle-hole peatland, contains a diversity of peatland types. The wetland has a ‘rich’ side that contains wetland vegetation associated with solute-rich, near-neutral pH (minerotrophic) water, and a ‘poor’ side containing vegetation that grows in solute-poor and acidic (ombrotrophic) water. Vertical head gradients at piezometer clusters located in the rich side clearly show that groundwater is moving upwards towards the land surface, consistent with the vegetation types and surface water quality. In contrast, vertical head gradients also show that groundwater is moving upward in the poor side even though the vegetation and surface water chemistry are not minerotrophic. An incipient raised bog in the center of the poor side is the only site where groundwater moves consistently downward.

A peat core collected at the bog center shows that the bog site was initially covered by minerotrophic vegetation, typically found in groundwater discharge zones, which was later replaced by ombrotrophic bog vegetation. Theoretical computer simulation experiments of the bog hydrogeologic setting through time suggest that the direction of vertical groundwater flow at the bog site permanently changed from up to down when a water table mound developed under a convex-shaped fen peat mound that probably formed because of differential peat accumulation. Ombrotrophic conditions and bog vegetation probably began when the fen water table mound grew sufficiently large enough to divert the upward movement of regional groundwater. The transition from rich to poor environments probably occurred when the wetland water table was substantially below the elevation of the surrounding regional water table.     

Experimental hydrological forcing to illustrate water flow processes of a subarctic ladder fen peatland

Large peatland complexes dominate the landscape of the James Bay Lowland in subarctic Ontario, Canada. However, there is not a thorough understanding of the hydrological processes occurring in these important systems, particularly how ladder fens connect large domed bogs to the aquatic ecosystems that drain the peatland complex. Ladder fens consist of a pool‐rib topography where flow downgradient is controlled by the peat ribs. Within the ribs, low‐lying preferential flow paths typically enhance the transmission of water, whereas the elevated ridge microforms impede water flow to downgradient aquatic ecosystems. To assess the hydrological connectivity, we study the role of the water table, peat transmissivity, and microtopography of a small ladder fen for 3 summers (2013–2015) in the James Bay Lowland. The system was manipulated with a sustained hydrological forcing (water addition) to the upslope boundary of the fen during 2014 (38 m³/day) and 2015 (30 m³/day). There was an exponential increase in transmissivity towards the peat surface due to extremely high‐hydraulic conductivities within the upper few centimeters of the peat deposit. At the maximum water table, the saturated hydraulic conductivity of the 0.1 m layer of peat below the water table varied depending on peat microtopography (preferential flow paths = 42–598 m/day and ridges = 16–52 m/day), resulting in high‐hydrological connectivity periods. Furthermore, during 2015, there was an abnormally large amount of precipitation (300 mm vs. long‐term average ~ 100 mm) that resulted in complete surface water connectivity of the site. This caused rapid movement of water from the head of system to the outlet (~15 hr) and runoff ratios >1, compared to low‐water table periods (runoff ratio ~ 0.05). This study highlights the profound importance of the transmissivity–water table feedback mechanism in ladder fens, on controlling the water retention and drainage of large peatland complexes.     

Water budget and surface‐layer water storage in a Sphagnum bog in central Sweden

A water budget was established for the open, undisturbed bog Stormossen, central Sweden, for the growing seasons of 1996 and 1997 as a part of the NOPEX project. The water budget was complemented with data on the spatial variation of groundwater levels and water contents in different microrelief elements (ridge, hollow and ridge margin). The seasonal (24 May to 4 October) rainfall, evaporation and runoff were 200, 256, and 43 mm in 1996, respectively, and 310, 286 and 74 mm in 1997, giving negative budgets of ?99 mm in 1996 and ?50 mm in 1997. Approximately 60% of the total budget was caused by storage changes in the upper 40 cm of the bog and 40% by swelling/shrinking in the layers below. This ‘mire breathing’ must be incorporated in future models of mire‐water dynamics. The water content varied diversely among the different microrelief elements, much depending on the properties of moss and peat together with distance to water table. There also was a strong hysteresis in the relationships between groundwater level and measured volumetric water content, depending partly on pore‐throat effects and partly on swelling/shrinking of the peat matrix. A seasonal variation of volumetric water content in a layer beneath water table was found to be larger than what could be justified by compression alone. We think that probable causes could be methane gas expansion together with temperature effects. The main conclusions of this study were: (i) water‐transport and storage characteristics are distinctly different among hummocks, ridges and hollows, (ii) mire wetness cannot be deduced from groundwater levels only, and (iii) an important part of the total water storage was caused by swelling/shrinking of the peat, not by changes in unsaturated water content. Copyright © 2002 John Wiley & Sons, Ltd.     

Oscillating peat surface levels in a restiad peatland,New Zealand—magnitude and spatiotemporal variability

Hydrology, particularly the water table position below the surface (relative water level, RWL), is an important control on biogeochemical and ecological processes in peatlands. The surface elevation (SE) in a peatland oscillates in response to changes in effective stress on the peat matrix mainly caused by water level fluctuations. This phenomenon is called peatland surface oscillation (PSO). To investigate the spatiotemporal variability of PSO, surface elevation and the water level above sea level (AWL) were measured monthly (23 sites) over one year in a warm‐temperate restiad peatland, New Zealand. At one site peat surface elevation was measured indirectly by monitoring AWL and RWL continuously with pressure transducers. Annual PSO (the difference between maximum and minimum surface elevation) ranged from 3·2 to 28 cm (mean = 14·9 cm). Surface elevation changes were caused by AWL fluctuations. Spatially homogenous AWL fluctuations (mean 40 cm among sites) translated into RWL fluctuations reduced 27–56% by PSO except for three sites with shallow and dense peat at the peatland margin (7–17%). The SE‐AWL relationship was linear for 15 sites. However, eight sites showed significantly higher rates of surface elevation changes during the wet season and thus a non‐linear behaviour. We suggest flotation of upper peat layers during the wet season causing this non‐linear behaviour. Surprisingly, PSO was subjected to hysteresis: the positive SE‐AWL relationship reversed after rainfall when the surface slowly rose despite rapidly receding AWL. Hysteresis was more prominent during the dry season than during the wet season. Total peat thickness and bulk density together could only explain 50% of the spatial variability of PSO based on manual measurements. However, we found three broad types of SE‐AWL relationships differing in shape and slope of SE‐AWL curves. These oscillation types reflected patterns in vegetation and flooding. Copyright © 2007 John Wiley & Sons, Ltd.     

Continuous measurement of the depth of water table (inundation) in wetlands with fluctuating surfaces

Assessment of the water table in compressible and/or floating peatland mats requires a remote sensor that can measure simultaneously the elevation of the peat surface and the water level. A simple device, comprising two potentiometers, a float and ‘bog shoe’ was used to measure the water level and peat surface witin 0-3 mm. The water table (depth of water level relative to the peat surface) was then calculated.     

Stable isotopes of water show deep seasonal recharge in northern bogs and fens

Ground water recharge is assumed to occur primarily at raised bog crests in northern peatlands, which are globally significant terrestrial carbon reservoirs. We synoptically surveyed vertical profiles of peat pore water δ¹⁸O and δ²H from a range of bog and fen landforms across the Glacial Lake Agassiz Peatlands, northern Minnesota. Contrary to our expectations, we find that local‐scale recharge penetrates to not only the basal peat at topographically high bog crests but also transitional Sphagnum lawns and low‐lying fen water tracks. Surface landscape characteristics appear to control the isotopic composition of the deeper pore waters (depths ≥0.5 m), which are partitioned into discrete ranges of δ¹⁸O on the basis of landform type (mean ± standard deviation for bog crests = ?11.9 ± 0.4‰, lawns = ?10.6 ± 0.1‰, fen water tracks = ?8.8 ± 1.0‰). Fen water tracks have a shallow free‐water surface that is seasonally enriched by isotope fractionating evaporation, fingerprinting recharge to underlying pore waters at depths ≥3 m. Isotope mass balance calculations indicate on average 12% of the waters we sampled from the basal peat of the fen water tracks was lost to surface evaporation, which occurred prior to advection and dispersion into the underlying formation. These new data provide direct support for the hypothesis that methane production in deeper peat strata is fuelled by the downward transport of labile carbon substrates from the surface of northern peat basins. Copyright © 2013 John Wiley & Sons, Ltd.     

Natural post-fire bog recovery

The dynamics of natural recovery of bog massifs in Tver oblast, which suffered from fire in 2010, is discussed. The presented conclusions are based on three-year field observations at check sites of Galitskii mokh bog along with monitored variations of the physicochemical properties of peat ash and pyrolyzed peat, bog water chemistry, the development of microbial communities, and vegetation recovery. Moreover, some features of bog massif flooding are discussed.     

The dynamic balance between organic acids and circumneutral groundwater in a large boreal peat basin

Large raised bogs and patterned fens cover 56% of the landscape in the Glacial Lake Agassiz region of northern Minnesota (USA). Organic acids supply most of the acidity in the surface water of the bogs, but groundwater upwelling from the underlying glacial deposits neutralizes these organic acids within deep peat. Substantial concentrations of organic acids also occur in the surface waters of fens mixed with variable amounts of inorganic solutes contributed by groundwater discharge. We used a triprotic analog model to determine the extent to which organic acids in fen and bog waters behave as strong or weak acids. The modeling approach optimized charge balance by calibrating estimates of mole site density in the DOC (dissolved organic carbon) of surface and pore waters with estimates of triprotic acid dissociation constants. Before the calibration process, all of the bog waters and 76% of the fen waters had more than +20% imbalance in charge balance. After calibration, more than 75% of all waters were electrochemically balanced within 20%. In the best calibration, the mole site denisty of bog DOC was estimated as 0.05 mmol/mmol C., approximately six times smaller than that estimated for fen DOC or the DOC in the fen deeper fen peats that underlie all bog landforms. The three modeled de-protonation constants were; pK_a1=3.0, pK_a2=4.5 and pK_a3=7.0 for the bog DOC, and; pK_a1=5.2, pK_a2= 6.5 and pK_a3=7.0 for the fen DOC. Bog DOC, behaves as a strong acid despite its small mole site density. The DOC in bog runoff can therefore theoretically acidify the surface waters in adjacent fens wherever these waters do not receive sufficient buffering alkalinity from active groundwater seepage.     

Mercury and methylmercury biogeochemistry in a thawing permafrost wetland complex,Northwest Territories,Canada

In arctic and sub‐arctic environments, mercury (Hg), more specifically toxic methylmercury (MeHg), is of growing concern to local communities because of its accumulation in fish. In these regions, there is particular interest in the potential mobilization of atmospherically deposited Hg sequestered in permafrost that is thawing at unprecedented rates. Permafrost thaw and the resulting ground surface subsidence transforms forested peat plateaus into treeless and permafrost‐free thermokarst wetlands where inorganic Hg released from the thawed permafrost and draining from the surrounding peat plateaus may be transformed to MeHg. This study begins to characterize the spatial distribution of MeHg in a peat plateau–thermokarst wetland complex, a feature that prevails throughout the wetland‐dominated southern margin of thawing discontinuous permafrost in Canada's Northwest Territories. We measured pore water total Hg, MeHg, dissolved organic matter characteristics and general water chemistry parameters to evaluate the role of permafrost thaw on the pattern of water chemistry. A gradient in vegetation composition, water chemistry and dissolved organic matter characteristics followed a toposequence from the ombrotrophic bogs near the crest of the complex to poor fens at its downslope margins. We found that pore waters in poor fens contained elevated levels of MeHg, and the water draining from these features had dissolved MeHg concentrations 4.5 to 14.5 times higher than the water draining from the bogs. It was determined through analysis of historical aerial images that the poor fens in the toposequence had formed relatively recently (early 1970s) as a result of permafrost thaw. Differences between the fens and bogs are likely to be a result of their differences in groundwater function, and this suggests that permafrost thaw in this landscape can result in hotspots for Hg methylation that are hydrologically connected to downstream ecosystems. Copyright © 2016 John Wiley & Sons, Ltd.