ArticlePDF Available

Abstract and Figures

Previous studies investigating organic-rich tundra have reported that increasing biodegradation of Arctic tundra soil organic carbon (SOC) under warming climate regimes will cause increasing CO2 and CH4 emissions. Organic-poor, mineral cryosols, which comprise 87% of Arctic tundra, are not as well characterized. This study examined biogeochemical processes of one-meter-long intact mineral cryosol cores (1-6% SOC) collected in the Canadian high Arctic. Vertical profiles of gaseous and aqueous chemistry and microbial composition were related to surface CO2 and CH4 fluxes during a simulated spring/summer thaw under light versus dark and in situ versus water saturated treatments. CO2 fluxes attained 0.8±0.4 mmol CO2 m-2 hr-1 for in situ treatments, of which 85±11% was produced by aerobic SOC oxidation, consistent with field observations and metagenomic analyses indicating aerobic heterotrophs were the dominant phylotypes. The Q10 values of CO2 emissions ranged from 2-4 over the course of thawing. CH4 degassing occurred during initial thaw, however all cores were CH4 sinks at atmospheric concentration CH4. Atmospheric CH4 uptake rates ranged from -126±77 to -207±7 nmol CH4 m-2 hr-1 with CH4 consumed between 0-35 cm depth. Metagenomic and gas chemistry analyses revealed that high-affinity Type II methanotrophic sequence abundance and activity were highest between 0- 35 cm depth. Microbial sulfate reduction dominated the anaerobic processes, outcompeting methanogenesis for H2 and acetate. Fluxes, microbial community composition, and biogeochemical rates indicate that mineral cryosols of Axel Heiberg Island act as net CO2 sources and atmospheric CH4 sinks during summertime thaw under both in situ and water saturated states.
Content may be subject to copyright.
Effects of simulated spring thaw of permafrost
from mineral cryosol on CO
2
emissions
and atmospheric CH
4
uptake
Brandon T. Stackhouse
1
, Tatiana A. Vishnivetskaya
2
, Alice Layton
2
, Archana Chauhan
2
,
Susan Pffner
2
, Nadia C. Mykytczuk
3
, Rebecca Sanders
4
, Lyle G. Whyte
5
, Lars Hedin
6
, Nabil Saad
7
,
Satish Myneni
1
, and Tullis C. Onstott
1
1
Department of Geosciences, Princeton University, Princeton, New Jersey, USA,
2
The Center for Environmental
Biotechnology, University of Tennessee, Knoxville, Knoxville, Tennessee, USA,
3
Vale Living with Lakes Centre, Laurentian
University of Sudbury, Sudbury, Ontario, Canada,
4
Department of Chemistry, North Central College, Naperville, Illinois, USA,
5
Department of Natural Resource Sciences, McGill University, Ste. Anne de Bellevue, Quebec, Canada,
6
Ecology and
Evolutionary Biology, Princeton University, Princeton, New Jersey, USA,
7
Picarro, Inc., Santa Clara, California, USA
Abstract Previous studies investigating organic-rich tundra have reported that increasing biodegradation
of Arctic tundra soil organic carbon (SOC) under warming climate regimes will cause increasing CO
2
and CH
4
emissions. Organic-poor, mineral cryosols, which comprise 87% of Arctic tundra, are not as well characterized.
This study examined biogeochemical processes of 1 m long intact mineral cryosol cores (16% SOC) collected
in the Canadian high Arctic. Vertical proles of gaseous and aqueous chemistry and microbial composition
were related to surface CO
2
and CH
4
uxes during a simulated spring/summer thaw under light versus dark
and in situ versus water saturated treatments. CO
2
uxes attained 0.8 ± 0.4 mmol CO
2
m
2
h
1
for in situ
treatments, of which 85 ± 11% was produced by aerobic SOC oxidation, consistent with eld observations
and metagenomic analyses indicating aerobic heterotrophs were the dominant phylotypes. The Q
10
values
of CO
2
emissions ranged from 2 to 4 over the course of thawing. CH
4
degassing occurred during initial
thaw; however, all cores were CH
4
sinks at atmospheric concentration CH
4
. Atmospheric CH
4
uptake rates
ranged from 126 ± 77 to 207 ± 7 nmol CH
4
m
2
h
1
with CH
4
consumed between 0 and 35 cm depth.
Metagenomic and gas chemistry analyses revealed that high-afnity Type II methanotrophic sequence
abundance and activity were highest between 0 and 35 cm depth. Microbial sulfate reduction dominated
the anaerobic processes, outcompeting methanogenesis for H
2
and acetate. Fluxes, microbial community
composition, and biogeochemical rates indicate that mineral cryosols of Axel Heiberg Island act as net CO
2
sources and atmospheric CH
4
sinks during summertime thaw under both in situ and water saturated states.
1. Introduction
Arctic permafrost is a large repository of soil organic carbon (SOC), with estimates of ~1100 Pg of carbon
within the top 3 m [Hugelius et al., 2014]. Subzero temperatures and limited access to liquid water inhibit
microbial degradation of SOC within permafrost [Schuur et al., 2008], sequestering this carbon until thawing.
In northern high latitudes, climate models predict increases of 3 to 7°C by 2050 [Rowlands et al., 2012] and
permafrost reduction by 2059% of its current areal extent by 2200 [Schaefer et al., 2011]. The conversion
of old SOC to CO
2
upon prolonged thawing of Arctic permafrost does occur [Mack et al., 2004; Schuur
et al., 2009], but Schuur et al. [2009] also reported net CO
2
uptake by surface vegetation for sites that had
experienced only moderate permafrost thaw. The release of CO
2
from old SOC, therefore, may be partly offset
initially by plant growth in Arctic tundra [Epstein et al., 2012]. The rate of CO
2
release is not simply a function of
temperature but also depends upon soil moisture, with any drying associated with warming, reducing CO
2
emission rates in organic-rich boreal soils [Allison and Treseder, 2008].
Permafrost thawing also leads to mass wasting of ground ice, affecting topography on the meter scale and
causingeitherdryingorooding in thermokarst terrains [Jorgenson et al., 2006; Osterkamp, 2007]. Flooding
during the formation of thermokarst lakes creates CH
4
-producing anaerobic environments [Walter et al.,
2006] and is predicted to increase the overall CH
4
ux from organic-rich Arctic histels and histosols [Zhuang
et al., 2004]. Because CH
4
has a global warming potential 75 times greater than CO
2
over a 20 year timescale
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 1
PUBLICATION
S
Journal of Geophysical Research: Biogeosciences
RESEARCH ARTICLE
10.1002/2015JG003004
Key Points:
Arctic mineral cryosols oxidize
atmospheric CH
4
under saturated
and dry states
CO
2
emissions result of aerobic
heterotrophy and governed by
water saturation
Permafrost thaw increases cumulative
carbon loss of soils up to a factor of 3
Supporting Information:
Supporting Information S1
Table S1
Table S2
Table S3
Table S4
Table S5
Table S6
Table S7
Table S8
Table S9
Table S10
Figure S1
Figure S2
Figure S3
Figure S4
Figure S5
Figure S6
Figure S7
Figure S8
Figure S9
Figure S10
Figure S11
Correspondence to:
B. T. Stackhouse,
bstackho@princeton.edu
Citation:
Stackhouse, B. T., et al. (2015), Effects of
simulated spring thaw of permafrost
from mineral cryosol on CO
2
emissions
and atmospheric CH
4
uptake,
J. Geophys. Res. Biogeosci.,120,
doi:10.1002/2015JG003004.
Received 31 MAR 2015
Accepted 6 AUG 2015
Accepted article online 13 AUG 2015
©2015. American Geophysical Union.
All Rights Reserved.
[Shindell et al., 2009], increases in the CH
4
emission rate will create a positive feedback leading to higher
temperatures and greater emissions [Davidson and Janssens,2006;Koven et al., 2011; Montzka et al., 2011].
The rate and timing of the positive feedbacks from CO
2
and CH
4
emissions with warming are difcult to pre-
dict because of the complexity of the interaction between temperature, soil characteristics, surface vegeta-
tion, and the microbial communities responsible for the degradation of SOC [Heimann and Reichstein,
2008]. Experimental investigations of subsamples from organic-rich active layer and permafrost cores have
revealed changes in the microbial populations after thawing for 7 to 11 days that were attributed to the
sudden availability of indigenous substrates [Coolen et al., 2011; Mackelprang et al., 2011]. Mackelprang
et al. [2011] also reported an abrupt release of CH
4
upon thawing that appeared to represent CH
4
trapped
in the frozen permafrost, as has been documented by Rivkina et al. [2007] and Allan et al. [2014]. These
incubation studies suggest that thawing of organic-rich permafrost provides a carbon source for microbial
respiration more rapidly than the annual to decadal timescales estimated from eld observations [Pautler
et al., 2010].
Mineral cryosols have lower SOC concentrations than those of histels and histosols and have been underre-
presented in experimental and eld studies of thawing permafrost. Yet they comprise 87% of the surface area
of the Arctic permafrost region and contain ~71% of the total SOC in Arctic permafrost and active layers
[Hugelius et al., 2014]. Unlike organic-rich permafrost, the mineral matrix of mineral cryosols will play a signif-
icant role in controlling the bioavailability of the SOC and how the rate of biodegradation varies with tem-
perature [Conant et al., 2011]. In vegetated tundra the uptake and respiration of modern CO
2
is promoted
by sunlight, whereas in sparsely vegetated, mineral cryosols sunlight can stimulate heterotrophic microbial
activity and respiration of CO
2
from SOC through mineral photocatalysis [Lu et al., 2012]. Czimczik and
Welker [2010] have reported net CO
2
emissions from the active layer of mineral cryosols in northern
Greenland during seasonal thaw, 2060% of which originated from the SOC. Lau et al. [2015]; Allan et al.
[2014], and Martineau et al. [2014] have reported atmospheric CH
4
uptake rates for mineral cryosols and
wetlands on Axel Heiberg Island, rates that increase with increasing temperature. CH
4
emissions, however,
are commonly reported for organic-rich Arctic sites [Verville et al., 1998; Christensen et al., 2000; Worthy and
Levin, 2000; Mastepanov et al., 2008; Olefeldt et al., 2012]. Jørgensen et al. [2014] have also reported atmo-
spheric CH
4
uptake rates by tundra in northeastern Greenland, rates that increase with temperature. When
Jørgensen et al. [2014] compared these rates to that the CH
4
emission rates reported by Mastepanov et al.
[2008, 2013] from the areally less abundant, organic-rich fens, they discovered that the ice-free northeastern
Greenland acted as a net atmospheric CH
4
sink. Jørgensen et al. [2014] also suggested that this atmospheric
CH
4
sink may increase with global warming.
Because the biogeochemical processes related to greenhouse gas ux from mineral cryosols may differ from
those of organic-rich active layer and permafrost, this study focused on understanding the relationships
between the CO
2
and CH
4
ux from mineral cryosols and temperature, soil geochemistry, sunlight, moisture
content, and the microbial community composition. Most incubation studies have been performed using 5 to
20 g samples that were isolated from interaction with the larger ecosystem. This makes it difcult to upscale
the observed rates spatially to the appropriate eld scales and temporally to the appropriate time scales for
climate models [Conant et al., 2011]. Intact cores collected from the active layer and underlying permafrost
were chosen over microcosms in the experimental design so that the measured surface uxes of CO
2
and
CH
4
from the cores could be directly compared to the vertical variation in temperature, soil properties, moist-
ure content, and microbial community structure [Knorr and Blodau, 2009]. To this end we conducted experi-
ments designed to simulate spring/summer thawing of the active layer and active layer deepening on
1 meter intact cores to examine greenhouse gas emissions under potential future soil conditions and to
understand the effects of permafrost thaw on the mineral cryosol system in its interconnected, in situ state.
We did this by (1) comparing the CO
2
and CH
4
uxes from this simulated thaw to ux measurements per-
formed in the eld during the spring/summer, (2) identifying the impact of different hydrological conditions
on CO
2
and CH
4
uxes through the active layer and into the atmosphere, (3) measuring the effects thawing
permafrost has on the surface CO
2
and CH
4
uxes as the frozen SOC becomes accessible to microorganisms,
(4) examining the relationships between the CO
2
and CH
4
uxes and the vertical zonation of the microbial
community composition, the temperature, the pore water chemistry, and the free energy for select anaerobic
microbial metabolic processes, (5) determining whether light input to the system would stimulate sufcient
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 2
plant growth to offset CO
2
emissions, (6) determining whether mineral cryosols are nitrogen limited, and (7)
comparing the δ
13
C of the CO
2
emissions with that of the underlying mineral cryosol SOC. The last goal
followed the approach of Biasi et al. [2005] who examined organic-rich cryosols from Siberia.
2. Materials and Methods
2.1. Field Site and Sample Collection
Core samples were collected in the high Arctic near the McGill Arctic Research Station (MARS) from an upland,
16 × 16 m
2
rectangular, ice-wedge polygon (N79°24.966W90°45.777) that lies within the watershed sur-
rounding Colour Lake on Axel Heiberg Island, Nunavut, Canada (Figure S1 in the supporting information).
The cryosols surrounding Colour Lake are classied as static and turbic cryosols and are sparsely vegetated
with sphagnum, sedges, and cotton grass [Buttle and Fraser, 1992]. The vegetation on the sampled polygon
consisted of Salix arctica,Polygonum viviparum,Dryas,Saxifraga, lichen, Papaver,Eriophorum, and other
grasses. The polygonal tundra surfaces are also encrusted with Ca and Mg sulfate minerals [Buttle and
Fraser, 1992] giving a slightly lighter color to the surface (Figure S1). The active layer thickness was 70 cm
in July 2010 and has varied from 60 to 73 cm between 2009 and 2011 [Allan et al., 2014]. The cryosol below
70 cm depth was therefore considered permafrost for this study.
The cores were collected from 30 April to 5 May 2011 when the ground was still frozen from the interior of a
single polygon after removing ~0.5 m of surface snow. A total of 40 cores were collected at a rate of 6 to 9
cores per day and were evenly distributed across the polygon from the interior to the edge (Figure S2 in
the supporting information). Coring was carried out using a Snow, Ice, and Permafrost Research
Establishment coring barrel (Jons Machine Shop, Alaska) [Wilhelm et al., 2012] lined with a 70% ethanol-
rinsed polycarbonate tube (San Diego Plastics, USA) of 7.6 cm diameter and 1 m length. By this means the
cores were collected directly into the polycarbonate tubes during drilling. Tubes were sealed immediately
with 70% ethanol-rinsed polyethylene end caps after removal from the coring barrel and were kept frozen
in coolers with blue ice during transport to Princeton University where they were stored at 20°C until the
thawing experiment. When drilling near the rim of the polygon, ice-rich cryosols were encountered near
the permafrost table (~70 cm) across a narrow depth prole and were interpreted as ice lenses extending
from the trough ice wedge into the polygon interior. These cores are referred to as the Edge cores in the text
below. The combined weight of the cryosol core and its tube and caps was 8 to 9.5 kg.
During July of 2011 and 2012, the CO
2
surface uxes were measured using opaque static chambers
(Figure S2) for 3 min intervals and a LiCOR LI-8100 (LiCOR, Nebraska). The CO
2
and CH
4
surface uxes were
also measured using dynamic chambers and a Cavity Ring-Down Spectroscopy (CRDS) Methane Isotope
Analyzer (Picarro, USA) during July of 2011 and 2012. Pore water samples were also collected in July 2011
using lysimeters (Soilmoisture Equipment Corp., CA) at depths of 15 cm, 30 cm, and 66 cm; were stored in
Ar-lled vials; and were frozen until analysis. The photosynthetically active radiation (PAR) ux at the site
during July 2012 was measured using a LightScout UV meter (Spectrum Technologies, USA) and ranged
from 100 to 1600 μmol m
2
s
1
depending on cloud cover and sun angle. These analyses overlap the values
of 300 to 900 μmol m
2
s
1
reported by Laeur et al. [2012] for Lake Hazen on Ellesmere Island, Canada for
July 2008.
2.2. Experimental Treatments and Design
The simulated thaw experiment was conducted in a 4.5°C walk-in cold room. Each core was removed from
the 20°C freezer and in the cold room their top caps quickly replaced with a clear polycarbonate cap that
provided ~0.8 to 1 L headspace and an inlet and outlet valve for gas sampling. Rhizon tubes (2.5 mm
diameter, 50 mm length) were installed to collect pore water at 5 cm (within the root zone and where the
SOC was high relative to the deeper active layer), 35 cm (the active layer above the water table), 65 cm
(the active layer at or below the water table), and 80 cm (within the permafrost) into each core while it was
still frozen. The cap at the bottom prevented pore water from leaving the system, essentially acting as a
new permafrost table after thawing was complete. Water left the system only through removal from
Rhizon tubes and evaporative loss from the surface during headspace collection. Four cores were also tted
with Xplorer GLX data logger thermocouples (Pasco Scientic, USA) to continuously monitor the core
temperatures at 5 cm, 35 cm, 65 cm, and 80 cm depths.
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 3
After attaching the headspace cap, Rhizon tubes, and thermocouples, each core was quickly placed into an
individual plastic trash bag (3 mil thickness) and then submerged into 20% ethanol baths contained in three
55 gal insulated barrels maintained at 3°C by recirculating thermochillers (ThermoCube 200-LT, Solid State
Cooling System, USA). The trash bag acted as an impermeable barrier between the 20% ethanol bath and the
cores. Each barrel held from 5 to 7 cores (Figure S3 in the supporting information). Because the installation
occurred in the 4.5°C walk-in cold room and because no core was out of the 20°C freezer for longer than
30 min during the installation process, the cores were still frozen when they were submerged into the bath.
The rst set of headspace gas samples were collected after the thermocouples had indicated that the cores had
equilibrated at 3°C approximately 1 week after they were submerged. Top-down active layer thawing process
was then initiated 3weeks later by progressive removal of the 20% ethanol allowing the exposed core tops to
equilibrate with the air temperature. Simulated progressive thawing down to the permafrost occurred over
9 weeks with the active layer attaining 4.2 to 4.8°C during the last 5 weeks of the experiment (the detailed tem-
perature history during thaw for the four different depths is shown in Figure S4 in the supporting information).
The 17 cores were divided into four treatments: in situ (seven cores, in situ water saturation conditions), dark
(two cores), saturated (four cores, mimicking thermokarst-affected terrain), and control (four cores, remain
frozen below 70 cm) (Table S1 in the supporting information). A subset of four in situ cores comprised the
Edge cores group. These were examined separately for differences in CO
2
and CH
4
emissions that could be
related to melting of the ice lenses. The dark treatment was designed to look for differences in the CO
2
ux
relative to the light exposed treatments that might reect uptake of CO
2
by vegetation as reported by Laeur
et al. [2012] for Lake Hazen on Ellesmere Island or for photocatalytic stimulation of CO
2
production [Lu et al.,
2012]. Dark cores were covered in aluminum foil to prevent light exposure but otherwise were treated in the
same manner as the in situ cores. Full-spectrum, uorescent bulbs (85 W, 5500 K, ALZO, USA) lit all remaining
core surfaces. The PAR ux through the polycarbonate core liner was measured by a LightScout UV meter
(Spectrum Technologies, USA) and was 300 to 800 μmol m
2
s
1
, comparable to that observed in the eld.
In the control treatment the permafrost (70 cm to 100 cm depth) was maintained at 2°C, equal to the
observed temperature of the top of the permafrost during the summer time, whereas the other cores were
fully thawed to 4.5°C (Figure S4). Saturated cores received weekly inputs of 40 mL of articial rainwater start-
ing at week 2, adding a total of ~500 mL of water. This was sufcient to ll pore space and maintain a water
table near the surface of the cores as thawing progressed down the core length. In situ, dark, and control
cores received no water input during thaw. Snow cover collected from Axel Heiberg Island was analyzed
to create articial rainwater of 200 μMNa
+
, 2.1 μMCa
2+
, 1.5 μMK
+
, 1.5 μMMg
2+
,58μMBr
, 2.1 μM HCO
3
,
140 μMCl
, and 2.4 μMSO
42
with the pH adjusted to 6.0 using HCl. The NaBr was added to track articial
rainwater penetration into the core to correct for any dilution effects on the aqueous species.
2.3. Headspace and Pore Water Sample Collection and Measurement
2.3.1. Gas Sampling
Core headspace was measured weekly. At the start of each week headspaces were ushed with ultrazero air with
an 80%/20% N
2
/O
2
gas mixture containing <5 ppmv CO
2
,<50 ppbv CH
4
,<30 ppbv H
2
,and<30 ppbv CO
(Airgas analyses). The detection limits for analyses are presented in Table S2 in the supporting information.
The ushing occurred at 20psi for 2.5 min, and the headspace was then immediately sampled. After 6days of
incubation the headspace was sampled again before ushing with ultrazero air and after ushing with ultrazero
air. All core headspace CO
2
concentrations were ~20 ± 12 ppmv after ushing with ultrazero air. The headspace
also contained 85 ± 41 ppbv CH
4
,310ppbvH
2
, and 82 ± 21 ppbv CO, and had O
2
and N
2
concentrations of
20.5 ± 0.1% an d 79.0 ± 0.5%, res pectively, after ushing with ultrazero air. The use of ultrazero air during the rst
13 weeks of thaw was selected to detect the release of H
2
,CO,CO
2
,andCH
4
during thawing given that Allan
et al. [ 2014] reported 60 to 130 nmols of CH
4
(g of cryosol)
1
and 850 to 550 nmols of CO
2
(g of cryosol)
1
emitted from active layer samples they collected from adjacent polygonal tundra. After the full core had been
thawed, from week 14 onward all core headspaces were ushed with an 80%/20% N
2
/O
2
gas mixture containing
400 ppmv CO
2
and 2 ppmv CH
4
, to replicate atmospheric conditions and to test for any CH
4
oxidation potential
and CO
2
uptake. The δ
13
CoftheCO
2
for this gas mixture was 12 ± 4VPDB (Vienna Pee Dee Belemnite).
2.3.2. GC Analysis and Flux Calculations
Weeks 0 to 3 gas samples were stored in ultra high purity (UHP) Ar-ushed serum vials prior to analysis, but
this approach did not permit replicate analyses to be performed. As a result, weeks 4 to 14 gas samples were
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 4
collected in SamplePro FlexFilm bags (SKC, USA) and analyzed on the same day. Gas composition was deter-
mined for O
2
and N
2
(thermal conductivity detector, TCD), H
2
and CO (reduced gas detector, RGD), and CO
2
and CH
4
(ame ionization detector) by gas chromatography (Peak Performer 1 series, Peak Laboratories, USA)
using UHP Ar as the carrier gas. Sample dilution to the instrumental linear response range was performed
using UHP Ar. The δ
13
CofCO
2
in the headspace was measured by iCO2-CRDS (Picarro G2101-i CRDS, USA).
The gas concentrations were converted into ux using the following formula:
Jimol m2h1

¼C1PtotV=RT A Δt½ (1)
where C
i
is the concentration of gas species i,P
tot
is the ambient headspace pressure (20 psi = 1.36 atm), Vis
the headspace volume (0.8 to 1 L), R= 0.082057 L atm mol
1
K
1
,Tis the ambient temperature in the cold
room in Kelvin (277.65), Ais the cross-sectional area of the cores (4.56 × 10
3
m
2
), and Δtis the duration
between sampling (typically 144 h).
The response of CO
2
ux to temperature was calculated as the Q
10
, the increase in the rate of a process
resulting from a 10°C increase in temperature, using the following formula:
Q10 ¼R2
R1
10
T2T1
ðÞ (2)
where T
1
is the starting time, T
2
is the ending time, R
1
is the rate at T
1
, and R
2
is the rate at T
2
. For this
experiment the Q
10
was calculated (1) using the temperature at 5 cm, (2) after the cryosol was over 0°C to
avoid articially inated Q
10
values due to the water-ice phase transition [Mikan et al., 2002], and (3) after
observed degassing from the cryosols upon thawing of ice trapping CO
2
.
Cumulative emissions of CO
2
from the cores were examined by treatment for time points at which signicant
changes in the rates of emissions occurred, using breakpoint analysis. A generalized linear model of CO
2
emissions was constructed in the statistical software R and then tested for a break in slope relationship using
the package segmented[Muggeo, 2008] and bootstrapped 100 times to avoid only discovering local optima
in slope changes.
2.3.3. Hydrological Sampling and Analysis
During progressive thawing syringes were connected to Rhizon tube sampling ports, and the syringe piston
pulled back for 8 h to draw in 3 to 5 mL of pore uid and gas through the Rhizon tubes. The combined volume
removed from each core for the six time points and four depths was ~4 × 4 × 6 = 96 mL, which was ~5% of the
~1.8 L pore water volume. The pH was measured using ~1 mL of sample within 2 h of collection without freez-
ing with a Ross pH microelectrode (Thermo Scientic, USA). One milliliter of collected water was injected into
a vial ushed and lled to 2 atm with UHP Ar for dissolved gas measurements by gas chromatography. The
dissolved O
2
and N
2
, however, could not be measured because their concentrations in 1 mL of water were
below the detection limit of the TCD. CO measurements in the pore water were also below the detection limit
for the RGD. The anion concentrations were measured by an ion chromatograph coupled to an electrospray
ionization-quadrupole mass spectrometer (Dionex IC25 and Thermo Scientic MSQ, USA). The cation concen-
trations were determined by inductively coupled plasma optical emission spectroscopy (Perkin Elmer Optima
4300 DV, USA). The NH
4+
concentrations were determined by the phenol/hypochlorite method [Parsons et al.,
1984]. The S
2
measurements were made by precipitating S
2
with 10 mM CdCl
2
,ltering the water on
0.1 μm, 27 mm diameter polycarbonate lters, dissolving the CdS precipitate in 4% HNO
3
solution and
measuring Cd by inductively coupled plasma atomic emission spectroscopy.
2.3.4. Calculations and Statistics
Mineral solubility, dissolved species activity, partial pressures, and the Gibbs free energy of microbial reac-
tions were calculated using the geochemical modeling program, the Geochemists Workbench version 8.0
[Bethke, 2008]. The pvalues for comparisons of gas ux and pore water geochemistry between the various
treatments were calculated using GraphPad QuickCalc as unpaired ttests (Pvalues, tratios, and Fratios
are presented in Table S3 in the supporting information). For the purposes of comparing pore gas concentra-
tions, saturated cores were compared against all other treatments combined into a single group to represent
unsaturated cryosols. A mass balance of hydrological and gaseous analytes in the system over the course of
thawing was used to infer the proportion of carbon emissions derived from particular respiration pathways. A
schematic overview of sample collection and analysis from the cores is presented in Figure S3d.
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 5
2.4. Soil Analysis and Sediment Collection
Soil moisture content (% H
2
O, w/w), soil organic carbon content (% SOC), total nitrogen (TN), the carbon-to-
nitrogen ratio (C/N), δ
13
C-SOC and the elemental composition was determined for a depth prole spanning
the top 1 m using representative cores collected from the polygon (see Figure S2). In a 30°C cold room the
core was partitioned into 2 cm sections using a miter saw that was sterilized with 70% ethanol between
cuts. Samples for total organic carbon (TOC) and TN were completely dried at 60°C and passed through a
1 mm sieve. The measurements were carried out on an elemental analyzer (CarloErba NC2500, Italy). The
δ
13
C-SOC was measured at Picarro, Inc. (USA) using a cavity ring-down spectrometer (iTOC-CRDS). Bulk soil
chemistry was determined by total dissolution and analysis by inductively coupled plasma mass spectrometry
(ActLabs, Ontario, Canada). A detailed description of the extraction and analysis of SOC composition is included
in the supporting information.
Ten gram samples (~5 cm
3
) of soil were also collected from 5 cm, 35 cm, 65 cm, and ~ 80 cm (10 cm below the
in situ permafrost table) of a core from each treatment and the Edge core for a total of 20 samples. These
samples were collected by drilling a small hole in the liner next to the Rhizon tube port with a sterilized drill
bit penetrating ~3 cm into the core and removing soil with a sterilized spatula. After collection the hole was
plugged with a 1.3 cm diameter butyl rubber stopper and sealed with waterproof tape. After 1week of thaw-
ing at each depth a second soil sample was collected from the same cores and from the previously drilled
holes. Given that the total core volume was ~4540 cm
3
, the removal of 2 × 4 × 5 = 40 cm
3
represents <1%
of a change in the total volume of the core and, therefore, should not signicantly affect the overall porosity
of the cores and their gas transport properties. SOC, TN, and C/N was also determined for these soil samples
(see Figure S2). Splits of these samples were shipped frozen to University of Tennessee Knoxville (UTK) for
genomic DNA extraction and sequencing. Unfortunately, due to a freezer failure at the UTK lab, the DNA
samples for the 35 cm depth after 1 week of thawing (ve samples) were exposed to room temperature for
a day and were, therefore, excluded from DNA analyses.
2.5. Metagenomic Analysis
Metagenomic analysis was performed on gDNA extractions from samples collected from three depths (5, 65,
and 80 cm) from ve cores representing the four different treatments and the Edge cores (Table S1,
Figure S2). Taxa were identied to the species level using both 16S rRNA and functional genes using the best
hit classication tool with default values of a maximum evalue of 10
5
, an 80% similarity cutoff and a mini-
mum alignment length of 15 bp or 15 amino acids for functional genes. The frequency data were exported
from MG-RAST (Metagenomic Rapid Annotations using Subsystems Technology) into STAMP [Parks and
Beiko, 2010] and analyzed statistically for means and standard deviations following normalization of hit abun-
dance data to the total sequences passing the QC pipeline. Normalized data were grouped by depth and
each group consisted of ve sample libraries each from a different permafrost core. See supporting informa-
tion for detailed description of genomic DNA extraction, metagenomic analysis, and data processing.
3. Results
3.1. Field Site Data
Based on analyses of a total of 38 samples collected from six cores, the SOC content was highly heteroge-
neous above 10 cm, ranging from 1 to 6%, and decreased to 1% below that depth where it remained to
the bottom of the cores (Figure S5a in the supporting information). These agree with the 0.98 to 1.1% SOC
concentrations reported by Wilhelm et al. [2011] for an active layer (14 to 19 cm and 55 to 60 cm depth)
and permafrost core (97 to 100 cm depth) from an adjacent polygon. All of these SOC values are low when
compared to the SOC concentrations for peat and fen sites in Greenland, Russia, and Alaska [Verville et al.,
1998; Christensen et al., 2000; Wagner et al., 2005; Olefeldt et al., 2012] but similar to values reported for pedon
2 from Ellesmere Island by Tarnocai and Bockheim [2011]. The SOC composition was found to be relatively
homogeneous along the 1 m prole, with the majority comprised of aliphatic compounds (69 to 84%)
and the remainder formed by lignin and carboxyl-rich alicyclic molecules (14 to 24%), aromatics (<2%),
and condensed aromatics (<4%). Geochemical characterization showed that the bulk inorganic com-
position of the soil was generally uniform with depth, with the exception of total sulfur, which was higher
in the top 10 cm and the permafrost table (0.29 ± 0.03%) than at other depths (0.18 ± 0.01%) (Table S4 in the
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 6
supporting information). Porosity was volumetrically estimated to be ~40% and the wet bulk density was
~1.8 ± 0.1 g cm
3
. Based on the density and the SOC concentration prole ~26 kg C m
2
exists in the top
1 m of the active layer and soil (equivalent to ~9.5 mol SOC in each core) similar to that shown by the map
of Hugelius et al. [2013a, 2013b]. Soil water content by weight of homogenized samples was 30.4 ± 2.0%
(010 cm), 11.0 ± 0.4% (3040 cm), 8.4 ± 0.6% (6070 cm), and 20.4 ± 0.7% (7080 cm). Material identiable as
root and leaf matter was observed to a depth of ~15 cm. The δ
13
C of the SOC ranged from 26.8 ± 0.2
in the top 2 cm to 25.7 ± 0.2at 65 cm (Figure S5a). Total nitrogen was 0.08± 0.02% and did not vary with
depth, similar to the 0.11% to 0.17% values reported by Wilhelm et al. [2011]. The C/N molar ratio of the six
cores examined ranged from 16 to 75 in the top 10 cm and declined to a uniform 17 ± 1 down to 1 m depth
(Figure S5b). No signicant difference was observed in the C/N of permafrost versus active layer samples. The
C/P and N/P values for the top 10 cm range from 103 to 244 and from 2.7 to 3.2, respectively. The C/P and N/P
values for the bottom 90 cm range from 39 to 61 and from 2.5 to 3.4, respectively. Pore water pH measured in
the eld in July 2012 decreased with depth from 6.0 at 15 cm to 5.2 at 65 cm. Acetate and Mn concentrations
increased with depth from 15 μMto54μM and 8 μMto14μM, respectively, whereas SO
42
decreased with
depth from 6 mM to 4 mM.
In situ CO
2
ux measurements were taken from transects of two ice wedge polygons that were adjacent
to the one that was cored for this study. The CO
2
uxes were 3.0 ± 2.9 mmol CO
2
m
2
h
1
in July of 2011
(12 to 13°C soil temperature at 0 to 5 cm depth) and 1.5 ± 0.4 mmol CO
2
m
2
h
1
in July 2012 (7 to 11°C soil
temperature at 0 to 5 cm depth). These polygons were found to be net sinks of atmospheric CH
4
with an
uptake ux of 400 ± 100 nmol CH
4
m
2
h
1
in July 2011 and 200 ± 100 nmol CH
4
m
2
h
1
in July 2012.
Nearby polygons and wetlands with higher vegetation cover areas directly adjacent to Colour Lake yielded
CO
2
uxes in July 2010 and July 2011 ranging from 0.7 to 24 mmol CO
2
m
2
h
1
and CH
4
uptake uxes of
520 to 800 nmol CH
4
m
2
h
1
in July 2012 [Wilhelm et al., 2012; Allan et al., 2014]. The CO
2
ux from
the vegetated wetland was 10 times that of the polygons.
3.2. Gas Flux During Core Thaw
The ux of CO
2
from all the cores before the thawing procedure began (week 0, 3°C) ranged from
0.15 ± 0.05 to 1.09 ± 0.91 mmol CO
2
m
2
h
1
(Figure 1a) with signicant differences between treatments.
The CO
2
ux from the Edge cores (Table S3a) and the dark treatment (Table S3b) were higher than that
of the control cores, whereas that of the cores selected for the saturated and in situ treatments did not differ
from the control cores.
The 5 cm, 35cm and 65 cm depths thawed by the start of week 3, the middle of week 4, and the start of week
10, respectively. Permafrost in the thawing treatments also went above freezing by week 9, whereas perma-
frost in the control treatment remained frozen and after week 12 was a constant 2.2°C (Figure S4). From
week 0 to 5, the CO
2
uxes for all treatments decreased and by week 5 were not statistically different from
the control treatment and averaged 0.21 ± 0.12 mmol CO
2
m
2
h
1
(Figure 1a).
From week 5 to week 13 the CO
2
ux increased by a factor of 3 for the Edge cores (to 1.21
± 0.51 mmol CO
2
m
2
h
1
, A3c) and the dark cores (to 0.93 ± 0.22 mmol CO
2
m
2
h
1
, Table S3d) and were
higher than the control cores ux (0.45 ± 0.25 mmol CO
2
m
2
h
1
, for week 13). The CO
2
uxes for cores repre-
senting the saturated and in situ treatments were not signicantly different from that of the control cores and
increased only by a factor of 2 (0.40 ± 0.16 mmol CO
2
m
2
h
1
and 0.61 ± 0.26 mmol CO
2
m
2
h
1
, respectively
for week 13). By the end of week 15 the control and saturated cores had emitted a total of 4.5 ± 0.5 mmol CO
2
and 4.8 ± 0.4 mmol CO
2
, respectively, with no signicant difference between the treatments (Figure 2). By the
end of week 15 the dark and in situ cores had released a total of 9.5 ± 0.9 mmol CO
2
(Table S3e) and 10.6
± 1.5 mmol CO
2
(Table S3f), respectively, signicantly higher than the control treatment. The Edge cores had
a total cumulative emission by the end of week 15 of 13.8 ± 1.3 mmol CO
2
(Edge in Figure 2).
The cumulative CO
2
emissions of all treatments were found to have signicant break points in slope between the
weeks of 6 and 10 where the CO
2
emission rates increased (Table S5 in the supporting information). Break point
analysis [Muggeo, 2008] of the CO
2
emission provides an estimation of points of signicant change of rate and did
reveal an incr ease in week 10 fo r the in situ treatm ent from 0.29 ± 0.01 to 0.49 ± 0.03 mmol CO
2
m
2
h
1
and in
week 9 for the dark treatment from 0.38 ± 0.02 to 0.85 ± 0.03 mmol CO
2
m
2
h
1
. The Edge cores, which exhib-
ited the greatest CO
2
emission rates, recorded an increase in CO
2
emission rate during week 7 from 0.62 ± 0.02
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 7
Figure 1. Comparison of the average (a) CO
2
and (b) CH
4
uxes from core treatments during progressive thawing. The right
hand ordinate corresponds to the average concentration in the headspace given the surface area, 6day interval, 20 psi
pressure, and 4.5°C air temperature. A ux of 1 mmol CO
2
m
2
h
1
between the sampling time points is equivalent to a
measured concentration of ~9100 ppmv CO
2
in the headspace. A ux of ~100 nmol CH
4
m
2
h
1
would yield a headspace
concentration of ~0.9 ppmv of CH
4
during the 6 day incubation time. The bold vertical lines indicate the time that the
temperature at 5 cm, 35 cm, 65 cm, and 80 cm depths went above zero, which for 65 and 80 cm depths was the same time.
The black bar indicates the range of CO
2
ux values measured in the eld with static chambers. The dotted red line
indicates the time at which the headspace was ushed with 400 ppmv CO
2
and 2 ppmv CH
4
. Error bars are ±1 standard
deviation from measurements of all cores within a given treatment. The number in parentheses next to the treatment is the
number of cores averaged.
Figure 2. Comparison of the average cumulative emission of CO
2
from each core treatment by week. The bold vertical lines
indicate the time that the temperature at 5 cm, 35 cm, 65 cm, and 80 cm depths went above zero, which for 65 and 80 cm
depths was the same time. The colored symbols at the top of the gure indicate the time at which each treatment exhibited
a signicant change in the emission rate as derived by break point analysis (Table 2). Error bars are ±1 standard error of the
average from measurements of all cores within a given treatment. The number in parentheses next to the treatment is the
number of cores averaged.
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 8
to 1.07 ± 0.04 mmol CO
2
m
2
h
1
.The
saturated treatment exhibited a smaller
increase during week 7 from 0.23 ± 0.01
to 0.38 ± 0.02 mmol CO
2
m
2
h
1
. The
control treatment, which should not exhi-
bit an increase at week 10 due to perma-
frost thawing, revealed an increase in CO
2
emission rate during week 6 from 0.16
± 0.01 to 0.40 ± 0.02 mmol CO
2
m
2
h
1
(Figure 2). The increases in CO
2
emission
during week 6 to 7 coincide with the time
when the temperature rose in the 65 scm
layer from 3to1.5°C.
The correlation of CO
2
ux with tem-
perature was investigated from week 5
onward once CO
2
ux began increasing
again. During this period CO
2
emissions
were found to have a positive correlation
with temperature across all depths,
with the strongest correlation at 5 cm (Figure S6 in the supporting information). From measurements
at 5 cm, an increase of 4°C corresponded to an increase of 0.80 ± 0.12 mmol CO
2
m
2
h
1
(dark), 0.36
±0.12mmolCO
2
m
2
h
1
(in situ), 0.24 ± 0.08 mmolCO
2
m
2
h
1
(control), and 0.16 ± 0.08 mmol CO
2
m
2
h
1
(saturated). These correspond to Q
10
values of 88± 20 (dark), 5 ± 5 (in situ), 18 ± 10 (edge), 8 ± 6 (control), and
5 ± 4 (saturated). If temperature response is examined across the entire length of the thawing experiment,
including the initial outgassing phase (weeks 0 to 5), the Q
10
values of all treatments range from 2 to 4.
The δ
13
C-CO
2
was 25 ± 4at the start of thawing, becoming less negative toward week 7 (20 ± 4), and
then becoming more negative again by week 13 (25 ± 4) (Figure S7 in the supporting information). No
signicant difference was observed between the control cores and any of the treatments except for the
Edge core treatments during weeks 9 to12, where the Edge cores were isotopically heavier by an average
of 5, before becoming statistically similar again by week 13. The starting and ending δ
13
C emission values
were similar to that of the SOC (26.8 ± 0.2).
The O
2
uptake in the headspace ranged from 0 to 0.9 mmol O
2
m
2
h
1
in the saturated treatments to 0 to
3.7 mmol O
2
m
2
h
1
in the in situ, dark, and control treatments. The O
2
uptake from the headspace was
correlated to the CO
2
ux in the edge cores (m= 1.8 ± 0.2, R
2
= 0.74) and dark cores (m= 1.7 ± 0.2, R
2
= 0.76)
where the CO
2
ux was the greatest (Figure 3). The O
2
uptake was weakly correlated to the CO
2
ux in the
control cores (m= 1.5 ± 0.3, R
2
= 0.44) and poorly correlated to CO
2
ux in in situ cores (m= 0.8 ± 0.3,
R
2
= 0.23, Figure 3) and saturated cores (m= 0.8 ± 0.3, R
2
= 0.17, Figure 3) where the CO
2
ux was the least.
During the course of progressive thawing both the O
2
uptake and the CO
2
emissions increased, most notice-
ably in the in situ and dark treatments. The intersection of the linear regression of CO
2
emission against O
2
consumption yielded xaxis intercepts of 0.12 ± 0.09, and 0.14 ± 0.07 mmol CO
2
m
2
h
1
for the dark, and in
situ treatments, respectively, indicating the amount of respiration that was not accounted for by aerobic
heterotrophy was 0.14 ± 0.05 mmol CO
2
m
2
h
1
. The poorly correlated, saturated treatments exhibited an
xaxis intercept of 0.14 ± 0.13 mmol CO
2
m
2
h
1
.
The CH
4
ux for all the treatments at week 0 ranged from 54 ± 28 nmol CH
4
m
2
h
1
to 126
± 33 nmol CH
4
m
2
h
1
with no signicant difference from the control (Figure 1b, Table S3g). During the
course of thawing the CH
4
ux steadily decreased across all treatments from weeks 0 to 5 to near-zero
emission where it leveled off through week 13. None of the treatments differed from the control
(9 ± 12 nmol CH
4
m
2
h
1
, Table S3h) during weeks 10 through 13 when the permafrost thawed.
All cores exhibited erratic ux of H
2
and CO in the headspace. Substantial outgassing was not observed for
either for H
2
or CO (Figure S8 in the supporting information). The ux of CO and H
2
did not differ between
treatments at week 0 (Tables S3i and S3j, respectively) or at week 14 (Tables S3k and S3l, respectively).
Figure 3. Average CO
2
production versus O
2
uptake for individual cores
at each week. Error bars are the ±1 standard deviation associated with
the analyses. The black line describes the expected ratio (1.3) of CO
2
production to O
2
uptake based on aerobic oxidation of the SOC based
upon the observed average C/O ratio of 6 to 1. Least squares ts of the
data by the given treatments were 0.8 ± 0.3 (saturated), 1.5 ± 0.3 (control),
1.8 ± 0.2 (edge), 0.8 ± 0.3 (in situ), and 1.7 ± 0.2 (dark). The errors represent
±1 standard deviation derived from the residuals of the least squares t.
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 9
After week 13 when the headspace of the cores was ushed with air containing 400 ppmv CO
2
and 2 ppmv
CH
4
, the CO
2
ux showed no difference due to the switch in the ushing gas, but CH
4
release ceased and all
cores immediately became net consumers of CH
4
(Figure 1b). For week 14, CH
4
uptake rates were 155 ± 31,
207 ± 7, 126 ± 77, 147 ± 65, and 166 ± 7 nmol CH
4
m
2
h
1
for the in situ, dark, saturated, Edge, and
control treatments, respectively, with no difference between treatments (Figure 1b, Table S3m).
3.3. Pore Water Characteristics
The 5 cm pH values initially increased from 5.9 on week 4 to 6.3 on week 8 before decreasing to 5.5 by week
14. The 35 cm pH at week 8 after initial thawing was 5.6 before increasing to 5.8 by week 12 and decreasing to
5.5 by week 14. The initial permafrost pH was 5.5 at week 12. By this time the pH values for all depths had
converged to ~5.5, similar to eld measurements (Figure S9 in the supporting information).
Pore water dissolved CO
2
concentrations at 5 cm and 35 cm ranged between 0.4 mM and 1.1 mM (equivalent to
apCO
2
of 0.005 to 0.012 atm consistent with CO
2
headspace concentrations) and did not signicantly change
over the course of thawing and was no different between saturated and in situ treatments (Tables S3n, S6 and
S7 in the supporting information, Figures 4a and 4d). Upon initial thaw of the 65 cm and permafrost depths on
week 10, all of the in situ, saturated, and dark treatment cores exhibited dissolved CO
2
concentrations of 1.2 mM
to 2.1 mM (equivalent to a pCO
2
of 0.016 to 0.028atm), which are higher than that of the shallower depths, and
they remained higher through week 15 (Table S3o). The DIC at the 65cm depth increased in the saturated cores
by 0.6 ± 0.1 mmol between week 11 and week 14 (Figure 4a). The HCO
3
at pH 5.5 accounts for ~10% of total
DIC or ~0.04 to 0.21 mM for all measurements. The total DIC in the pore water accounts for 2 ± 1 mmol, equiva-
lent to ~45% of the CO
2
released into the headspace for the saturated andcontrol treatments and ~23% for the
in situ and dark treatments. The calculated saturation index, ln Q/K, for calcite (4.3), ankerite (7.1), and dolo-
mite (7.6) remained well below saturation during thaw.
Dissolved CH
4
concentrations for saturated cores at 5 cm (Figure 4e) decreased from 73 ± 155 nM (pCH
4
=32
±68μatm or 23 ± 49 ppmv CH
4
at the headspace pressure) on week 4 to below detection limit after week 6 in
Figure 4. Pore water concentrations by depth during thaw of saturated cores [4] for (a) CO
2
, (b) CH
4
, and (c) H
2
and for in situ cores [13] for (d) CO
2
, (e) CH
4
, and (f) H
2
.
The bold vertical lines indicate the time that the temperature at 5 cm, 35 cm, 65 cm, and 80 cm depths went above zero, which for 65 and 80 cm depths was the same
time. Error bars are ±1 standard deviation of the average from measurements of all cores within a given treatment. The number in parentheses next to the treatment
is the number of cores averaged.
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 10
all treatments (Figures 4b and 4e, pCH
4
<22 μatm). At 35 cm CH
4
concentrations ranged from 51 nM to 202 nM
but were not statistically different between treatments. CH
4
was highest in the permafrost upon initial thaw at
week 12, with concentrations of 390 ± 200 nM and 510 ± 450 nM for the saturated and in situ treatments, respec-
tively. Within 3 weeks the CH
4
concentrations of the permafrost in the in situ cores had decreased to 140 ± 140 nM
(Table S3p) (Figure 4e), whereas in the saturated cores CH
4
concentrations did not change (446 nM) (Figure 4b).
The dissolved H
2
values at the 5 cm and 35 cm depths after thawing were 14 ± 13 nM and 63 ± 22 nM, respec-
tively (equivalent to pH
2
=15 to66μatm or 11 to 48 ppmv H
2
at the headspace pressure), and displayed no
clear trend over time during thawing. Within the permafrost upon initial thaw at week 12, the H
2
concentra-
tions were 540 ± 440 nM and 390 ± 210 nM, for the in situ and saturated treatments, respectively (Figures 4c
and 4f). By week 15 these high H
2
concentrations decreased to 140 to 180 nM, and the H
2
concentrations did
not differ between treatments.
The active layer pore water NH
4+
concentrations were highest at 5 cm (8 ± 2.8 μM) and decreased over the
course of thawing (Figures S10a and S10b in the supporting information). Those for the 35 cm depth
remained below 2 ± 1 μMNH
4+
throughout the thaw, whereas at 65 cm depths the initial values were
6±3μMNH
4+
and decreased during the course of the experiment. On week 15 the NH
4+
concentrations at
all depths had decreased to <1μM with no differences between treatments. In the week after the permafrost
thawed, the pore water concentrations of NH
4+
for the saturated and in situ cores were 39 ± 27 μM and
22 ± 11 μM, respectively, with no difference between treatments. These concentrations were higher than active
layer concentrations during initial thawing in both saturated (Table S3q) and in situ treatments (Table S3r and
Figures S10a and S10b). From week 12 to week 15, the concentration of NH
4+
decreased in the thawed permafrost
of both the saturated (Table S3s) and in situ (Table S3t) cores to 0.8μM and 2 ± 2 μM, respectively.
Pore water NO
3
concentrations varied from below the detection limit of 0.5 μMto15μM across core treat-
ments but were higher at the 65 cm and permafrost depths than at the 5 cm and 35 cm depths (Tables S3u
and S6S9). This increasing NO
3
concentration with depth was also reported by Wilhelm et al. [2011] who
detected NO
3
in the permafrost but not in the active layer above 60 cm. From week 12 to week 15, when
NH
4+
released from the permafrost thaw diminished in concentration, no increases in the NO
3
concentra-
tions were detected and NO
2
concentrations were consistently below the detection limit of 1 μM.
Ca
2+
,Mg
2+
,andSO
42
were the major pore water ions with concentrations of 0.5 to 3.2mM, 0.5 to 3.8 mM, and
2 to 5 mM, respectively. Initial SO
42
concentrations were highest at the permafrost table and inthe permafrost
(4 to 6 mM) and decreased toward the surface (2 to 3mM). The pore water was below saturation for gypsum
and epsomite at all depths and times (ln Q/K= 12.5 to 1.9). The SO
42
concentrations in the active layer
depths (5 cm, 35 cm, and 65 cm) decreased over the course of the thawing in all treatments as well as the con-
trol (Tables S6S8 in the supporting information) with average losses of 0.4 mM, 0.3 mM, and 0.5 mM, respec-
tively, and with no signicant difference between treatments and the control. At the permafrost depth the
SO
42
concentration decreased by 1.3 ± 0.6 mM from week 12 to week 15 across all treatments. Surprisingly,
no S
2
was detected above the detection limit of 3 μM. Assuming that the pore water Fe concentrations of
0.006 t o 9 μM are completely Fe
2+
and S
2
concentrations are 3 μM, the system would be supersaturated with
respect to pyrite (ln Q/K= 4.2 to 5.2), but under saturated with respect to FeS minerals (ln Q/K= 4.1 to 5.2).
The primary organic acid observed across all treatments was acetate, which ranged from ~10 to ~300 μM.
Pore water acetate concentrations were initially highest in the in situ core treatment for all time points and
did not signicantly decrease during the course of thawing for any treatments (Tables S6S9 in the support-
ing information). Propanoate and formate were never detected (<1μM), whereas lactate was transiently
observed at concentrations up to 2 μM but showed no clear trend with time or depth (data not shown).
3.4. Respiration Pathways
The O
2
uptake ux into saturated cores was ~0.4 mmol m
2
h
1
, whereas that for the in situ treatments was
~4 mmol m
2
h
1
. Using Ficksrst law,
J¼ϕDdO
2
½
dZ(3)
where Jis ux in mol cm
2
s
1
,ϕis porosity (0.4), Dis the diffusion constant of O
2
in water at
4.5°C = 2.1 × 10
5
cm
2
s
1
[Cussler, 1985], d[O
2
] is the change in dissolved O
2
concentrationinmolcm
3
,
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 11
and dZis the depth in centimeters. The measured O
2
uptake uxes would correspond to a d[O
2
]/dZof
1.3 × 10
6
to 1.3 × 10
5
mol cm
3
-cm
1
. The top surface of the water layer will have an air saturation
concentration of 5 × 10
7
mol cm
3
. To sustain these uptake rates the thickness of the diffusion layer
would have to be 0.04 to 0.4 cm. The time required to diffusively penetrate this thin layer is given by
t¼Z2=ϕD(4)
yielding a tof 3 min to 5.5 h. The diffusion boundary layers are much thinner than the diffusion distances
corresponding to 15 weeks indicating that O
2
uptake rates and the corresponding CO
2
production rates are
diffusion limited. For the dark and in situ cores the O
2
uptake rates indicate that O
2
penetration is occurring
through the unsaturated pore space that exists in these mineral cryosols.
Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR-MS) analyses indicated that the SOC
extracted by acetone and pyridine from the bulk SOC was composed primarily of aliphatic compounds
with O/C ratios of 0.15 ±0.1 and H/C ratios of 1.5 to 2. As such, the extractable SOC average molecular weight
and the C/O of the SOC can be represented by the model molecule C
12
H
24
O
2
. Using this in a formula for
aerobic respiration,
C12H24 O2þ17O212CO2þ12H2O (5)
the estimated ratio of O
2
uptake to CO
2
released is ~1.3. If one includes the additional O
2
required for ~8%
growth yield for a cellular biomass with the formula C
4
H
7
O
1.5
N[Phelps et al., 1994], then the O
2
uptake to CO
2
released is ~1.5. Given that the slope of the measured uptake of O
2
versus the evolved CO
2
was 1.8 ± 0.1 for
the in situ treatments (Figure 3), the CO
2
production appears to be primarily governed by heterotrophic
respiration of the aliphatic SOC using O
2
as a terminal electron acceptor. Assuming an 8% growth yield, the 5
to 14 mmol of CO
2
emitted during the rst 15 weeks (Figure 2) would correspond to an increase of 0.005 to
0.014 g of biomass in the core, which would be equivalent to an increase of ~1 to 3× 10
7
cells g
1
assuming a
cellular mass of 5 × 10
14
g of C cell
1
. Enumeration of the bacterial abundance in these permafrost samples
by uorescence in situ hybridization yielded 9.7 × 10
8
cells g
1
[Vishnivetskaya et al., 2014]. The assumed
growth yield represents a maximum increase of at most 1 to 4% of the observed population since much of
the anabolic activity during the initial thaw probably goes to cellular repair [Price and Sowers, 2004].
The pore water analyses indicate that the cores lost 0.4 ± 0.1 mM SO
42
from the active layer and
1.3±0.6mMSO
42
from the permafrost. Assuming a maximum estimated water-lled volume of 40%, this
will result in the reduction of ~1.2 ± 0.4 mmol SO
42
in the entire core during thawing. Microbial sulfate
reduction coupled to oxidation of the SOC would be described by the following reaction:
C12H24 O2þ8:5SO42þ8:5Hþ
8:5HS þ12CO2þ12H2O (6)
According to this reaction the sulfate lost accounts for 1.7 mmol CO
2
evolved or 12% to 34% of the total
evolved CO
2
or equivalent to ~0.15 mmol CO
2
m
2
h
1
.
During the thawing the HS
was less than 3 μM despite the evidence of sulfate loss in both the in situ and
saturated cores. The Br
tracer was only detected in the 5 cm layers of the saturated cores and not at the dee-
per depths indicating that dilution by articial rainwater in the saturated cores cannot explain the loss of sul-
fate. As shown earlier this indicates that the system would supersaturated with respect to FeS
2
but not with
respect to FeS. This suggests that Fe(III) reduction coupled to SOC oxidation and FeS
2
precipitation must have
occurred to remove the HS
. This reaction can be expressed by the following equation:
262=3Fe OHðÞ
3þC12H24 O2þ451=3HSþ451=3Hþ
262=3FeS2þ12CO2þ46H2O (7)
yielding an additional 0.3 mmol CO
2
. An increase of 0.5 μmol Fe
2+
in the active layer over the 15 weeks of the
experiment (Tables S6S8) would contribute 0.08 μmol CO
2
assuming microbial Fe(III) reduction coupled to
SOC oxidation by the following formula,
68Fe OHðÞ
3þC12H24 O2 þ136H þ68Fe2þþ12CO2þ182H2O (8)
Total nitrication of the NH
4+
pool in the permafrost, which accounts for 45 to 78 μmol of reduced N, and
subsequent microbial NO
3
reduction would only yield a maximum of 69 μmol CO
2
assuming it is coupled
to SOC oxidation by the following formula:
13:6NO3þC12H24 O2þ13:6Hþ
6:8N2þ12CO2þ18:8H2O (9)
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 12
The combined anaerobic reactions above would contribute a maximum total estimate of ~2 mmol of CO
2
,
which is equivalent to ~0.2 mmol CO
2
m
2
h
1
emitted from the cores, with emissions from sulfate reduction
dominating. This rate is comparable to the inferred anaerobic respiration rates from Figure 3, which suggests
that anaerobic processes should contribute 0.14 ± 0.05 mmol CO
2
m
2
h
1
toward total emissions.
3.5. Microbial Community Structure
The cryosol microbial community was dominated by Actinobacteria and Proteobacteria, with Firmicutes,
Bacteriodetes, and Acidobacteria forming less abundant constituents (Table 1). Archaeal sequences com-
prised only 0.48 ± 0.30% of the total sequences. The archaeal sequences were primarily composed of
Euryarchaeota (0.31 ± 0.20%), which includes the methanogens, and Thaumarchaeota (0.12± 0.09%), which
includes ammonia oxidizers. Eukaryotic sequences comprised only 0.11 ± 0.06% of the total sequences,
and the largest phylum was Ascomycota (0.05 ±0.05%), a fungal phylum involved in the degradation of com-
plex organic compounds.
When the relative sequence abundance of the 65 cm and permafrost samples were compared to that of the
5 cm samples, signicant increases were observed for Firmicutes (Table S3v) and Actinobacteria (Table S3w)
and signicant decreases were observed for Acidobacteria (Table S3x), Proteobacteria (Table S3y), and
Verrucomicrobia (Table S3z) (Figure S11 in the supporting information). No signicant differences
between the relative abundance of these phyla were observed when comparing the 65 cm samples to the
permafrost samples. The relative abundance of methanotrophic sequences decreased with depth from
1.15 ± 0.25% at 5 cm to 0.35 ± 0.07% in the permafrost. Type I methanotrophs (family Methylococcaceae)
were present at ~0.17% across all depths, whereas the relative abundance of Type II methanotroph
Table 1. % of Total Sequence Reads
a
5 cm 65 cm Permafrost
Phylum
Acidobacteria 7.8 ± 1.2 3.2 ± 1.2 2.2 ± 0.7
Actinobacteria 20.8 ± 3.1 34.3 ± 4.3 39.6 ± 9.2
Bacteriodetes 3.6 ± 0.8 3.7 ± 0.9 8.4 ± 1.8
Cyanobacteria 1.8 ± 0.3 2.2 ± 0.3 1.4 ± 0.7
Firmicutes 3.5 ± 0.9 7.3 ± 1.4 5.0 ± 1.5
Gemmatimonadetes 1.1 ± 0.3 1.9 ± 0.6 0.7 ± 0.5
Proteobacteria 35.1 ± 4.0 22.5 ± 3.7 20.8 ± 4.9
Methanogens
All 0.12 ± 0.04 0.31 ± 0.05 0.20 ± 0.07
Acetoclastic 0.09 ± 0.03 0.18 ± 0.02 0.13 ± 0.05
Methanotrophs
All 1.15 ± 0.25 0.50 ± 0.10 0.35 ± 0.07
Type I 0.17 ± 0.01 0.15 ± 0.01 0.17 ± 0.05
Type II 0.84 ± 0.25 0.25 ± 0.09 0.13 ± 0.02
Verrucomicrobia 0.14 ± 0.01 0.10 ± 0.04 0.06 ± 0.04
Sulfate-Reducing Bacteria
All 0.85 ± 0.12 1.16 ± 0.18 0.76 ± 0.18
δ-Proteobacteria 0.69 ± 0.11 0.89 ± 0.17 0.59 ± 0.17
Firmicutes 0.16 ± 0.04 0.27 ± 0.06 0.17 ± 0.06
Nitrifying Bacteria
All 2.12 ± 0.25 1.31 ± 0.17 1.07 ± 0.25
Ammonia-oxidzing bacteria 0.47 ± 0.02 0.42 ± 0.02 0.57 ± 0.20
Nitrite-oxidizing bacteria 1.65 ± 0.25 0.89 ± 0.17 0.50 ± 0.15
Fe(III)-reducing Bacteria
All 2.52 ± 0.60 1.49 ± 0.20 1.08 ± 0.25
Acidobacterium 1.42 ± 0.54 0.47 ± 0.17 0.40 ± 0.17
Acidophilium 0.29 ± 0.06 0.14 ± 0.03 0.08 ± 0.02
Acidothiobacillus 0.07 ± 0.01 0.07 ± 0.01 0.07 ± 0.03
Geobacteraceae 0.74 ± 0.08 0.81 ± 0.10 0.53 ± 0.18
a
Based on the average of metagenomic sequence data from ve cores (one from each treatment) at each depth after
1 week of thawing. Error represents ±1 standard deviation from all metagenomes at that depth.
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 13
sequences (genera Methylocystis,Methylosinus,Methylocapsa, and Methylocella) decreased from 0.84 ± 0.12%
at 5 cm to 0.13± 0.02% in the permafrost, a shift largely driven by Methylocella. The genus Methylacidiphilum
accounted for the majority of the Verrucomicrobia sequences.
Sulfate-reducing bacteria (SRB) were present at all depths and were predominantly composed of the orders
Desulfobacteriales, Desulfovibrionales, and Syntrophobacterales (δ-Proteobacteria) and the genera
Desulfotomaculum (Firmicutes). At the order level the relative abundances of SRB sequences were lowest
in the permafrost (0.76 ± 0.18%) and highest at 65 cm (1.16 ± 0.18%). Across all depths the SRB community
was dominated by δ-Proteobacteria. Syntrophobacterales formed the largest component of the SRB at
5 cm (0.28 ± 0.03%), whereas at 65 cm all groups comprised a roughly equal share of the SRB sequences.
The genera Desultobacterium, an anaerobic bacteria involved in sulfur, nitrogen, and organohalogen cycling,
was found to comprise up to 0.24 ± 0.10% of sequences at the 65 cm depth.
Genera associated with the reduction of Fe(III) minerals in low pH environments [Blöthe et al., 2008] were
found to decrease with depth from 2.52 ± 0.60% of total sequences at 5 cm to 1.08 ± 0.25% of total sequences
in the permafrost. The decrease with depth was driven by Acidobacterium and Acidophilium, whereas
Acidothibacillus (0.07 ± 0.01%) and Geobacteraceae (0.74 ± 0.08%) were stable across the entire depth prole.
The relative abundance of nitrifying bacterial sequences decreased with depth from 2.12 ± 0.25% of total
sequences at 5 cm to 1.07± 0.25% of total sequences in the permafrost (ammonia-oxidizing genera:
Nitrosococcus,Nitrosovibrio,Nitrosomonas, and Nitrosopira; nitrite-oxidizing genera: Nitrobacter,Nitrococcus,
Nitrospira, and Nitrospina).
The relative abundance of methanogenic sequences (Methanobacteriales, Methanococcales, Methanomicrobiales,
Methanosarcinales, and Methanopyrales) increased from 0.12 ± 0.04% at 5 cm to 0.31 ± 0.05% at 65 cm depth
and decreased to 0.20 ± 0.07% within the permafrost. Acetoclastic methanogens (Methanosarcinales) formed
the largest single methanogenic order, comprising a minimum of 56% of the methanogenic sequences at
65 cm and a maximum of 77% of methanogenic sequences at 5cm. The next most abundant methanogenic
order was Methanomicrobiales, which comprised a maximum of 28% of total methanogenic sequences
at 65 cm.
4. Discussion
4.1. Release of CO
2
and CH
4
During Active Layer Thaw
During the initial 6 weeks of simulated progressive thawing, the active layer, regardless of treatment condi-
tions, released CO
2
and CH
4
to the headspace. This is supported by the following data:
1. Both the CO
2
and CH
4
ux from the cores decline with time (Figure 1), while the temperature in the active
layer is increasing (Figure S4).
2. The observed shift in the pore water pH from 5.9 to 6.4 at 2.5°C from week 3 to 7 (Figure S9) and the
simultaneous increase in the δ
13
CofCO
2
by 4.4(Figure S7) are consistent with the abiotic movement
of isotopically heavy HCO
3
into H
2
CO
3
and CO
2
during degassing and the carbon isotopic fractionation
between HCO
3
and H
2
CO
3
and CO
2
[Mook et al., 1974].
3. The 6 week time frame for this decline in ux is consistent with diffusive release from a water saturated
5 cm layer to mostly water saturated 35 cm layer using the effective diffusivity calculated from equation
(1) of Brummell et al. [2012] that takes into account the air-lled porosity of these cores.
4. The CO
2
concentrations in the active layer mineral cryosol range from 847 to 1224 nmol of CO
2
g
1
[Allan
et al., 2014]. This corresponds to a trapped CO
2
reservoir in the top 35 cm of the cores of 2.3 to 3.6 mmol
CO
2
. By week 5 when all the cores had thawed to a depth of 35 cm the cumulative CO
2
released ranged
from 0.8 to 3.2 mmol CO
2
(Figure 2).
5. The acetate concentrations did not signicantly change during the course of thawing consistent with low
heterotrophic activity during this time interval, though not necessarily proof by itself.
The CO
2
and CH
4
released from the cores during the initial thaw are likely to represent CO
2
and CH
4
trapped
in the cryosols during freeze-in and wintertime respiration. As temperatures decline during the fall, the
temperatures near the surface will be lower than those closer to the permafrost table. As a result the rate
of aerobic methanotrophy concentrated near the surface will be less than the rate of methanogenesis at
depth due to physical effects. As the surface freezes it will create an additional barrier to the diffusion of
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 14
atmospheric O
2
into the deeper active layer cryosol creating a more reducing environment for the anaerobic
generation of CO
2
and CH
4
. Some of this CO
2
and CH
4
may leak out during freeze-in, but some may remain
trapped in the ice as bubbles. Mastepanov et al. [2013] have proposed that these storage mechanisms for CO
2
and CH
4
in the active layers of organic-rich cryosols in Greenland could account for the interannual variability
of CH
4
emissions. At least one microbial isolates from AHI mineral cryosol has been shown to grow at 15°C
and actively metabolize at 25°C [Mykytczuk et al., 2013]. We propose that similar physical/microbial
mechanisms apply to the mineral cryosol of AHI but that the amount of CH
4
and CO
2
stored in the frozen
active layer is far less than that of the Greenland fen sites. In fact, as shown in Figure 1b, if the ultrazero air
with 2 ppmv of CH
4
had been used at the beginning of the experiment, instead of just ultrazero air, we
may not have detected any CH
4
release from the frozen cores. Rivkina et al. [2007] reported microbial CH
4
production from Siberian permafrost samples incubated at temperatures as low as 16.5°C. Although
subzero rates of respiration would be slow, wintertime respiration could signicantly add to the CO
2
and
CH
4
reservoir at AHI as has been reported for Svalbard heath and tundra meadow active layers by
[Morgner et al., 2010].
4.2. CO
2
Release From SOC Oxidation and During Permafrost Thaw
After week 5 the CO
2
uxes from the cores increased with increasing temperature and a deepening active
layer, while the δ
13
C decreased to 25by week 13 (Figure S7). The latter observation is consistent with
those of Biasi et al. [2005], who observed a shift of up to 10toward isotopically lighter CO
2
emissions with
increasing incubation temperature on microcosms of organic-rich cryosols (up to 10over a 22°C range).
The in situ and dark treatments exhibited the greatest O
2
drawdown, which correlated with the increase in
the CO
2
ux (Figure 2). After 15 weeks the vegetation on the surface of the cores exposed to light appeared
alive, and small amounts of new growth was observed. The vegetative cover on the dark cores, on the other
hand, had turned brown, and no new growth was visible. The dark treatment CO
2
emissions were greater
than those of the in situ cores but less than those of the Edge cores. These observations are consistent with
some uptake of CO
2
by the surface vegetation, but the spatial variation within the polygon and the presence
of ground ice has a more signicant impact on the range of CO
2
uxes observed from week 7 to 14. The CO
2
uxes from the intact cores after the active layer had thawed ranged from 0.5 to 1.8 mmol m
2
h
1
, overlap-
ping the 0.5 to 2.9 mmol m
2
h
1
summer CO
2
uxes measured in the eld (Figure 1a). In fact, the highest
CO
2
uxes are observed in the wetlands near Colour Lake [Allan et al., 2014]. Summertime emissions of
CO
2
at Colour Lake and from the intact core experiments indicate that photosynthetic xation of CO
2
is
currently unable to keep up with soil respiration even during peak growing season. This is consistent with
the eld results of Czimczik and Welker [2010] who observed
14
C-depleted CO
2
emissions from sparsely vege-
tated tundra in northern Greenland during summer active layer development from which they also inferred
heterotrophic respiration of thawed SOC. Well-studied organic-rich cryosol sites in Greenland, Siberia, and
Sweden are known to currently be net CO
2
sinks with uxes ranging from 0.4 to 2.8 mmol m
2
h
1
[Christensen et al., 2000; Kutzbach et al., 2007; van der Molen et al., 2007; Olefeldt et al., 2012; Mastepanov
et al., 2013], with high summertime CO
2
xation balanced by wintertime respiration and partial loss of
accumulated carbon stores.
The Q
10
values derived from the CO
2
ux measurements from week 0 to 15 ranged from 2 to 4 for tempera-
tures ranging from 0.8 to 4.8°C at 5 cm depth. Comparison of the intact core CO
2
uxes with those measured
in the eld at 9 and 12.5°C yielded Q
10
values ranging from 1.4 to 2.8. The CO
2
eld ux measurements
reported by Allan et al. [2014] from adjacent polygons were spaced over the season to capture the variations
in surface soil temperature, which ranged from 7 to 21°C. Their measurements yielded Q
10
values of 1.2 to 2.4
similar to the results from the cores. The detailed measurements provided by the simulated core thawing,
however, suggest that CO
2
stored in the active layer ice, either trapped during fall freezing or accumulated
during wintertime respiration, is released during the early summer. The Q
10
values observed for the control,
saturated, and in situ treatments after the initial 5 weeks ranged from 5 to 11 overlapping Q
10
values of 4.6 to
9.4 observed by Mikan et al. [2002] in microcosm experiments of organic-rich Alaskan cryosols. This raises the
question as to whether Q
10
values derived from eld measurements, like those of Allan et al. [2014], may be
underestimating the true value because of this early degassing.
The CO
2
released during the 15 weeks accounted (Figure 2) for ~0.05 to 0.15% of the bulk ~9.5 mol SOC in a
core. With CO
2
emission rates of 1.5 to 3 mmol CO
2
m
2
h
1
and assuming no additional carbon inputs into
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 15
the system, total turnover of SOC would occur in 320 to 640 years if respiration was restricted to the summer
season with a developed active layer (3 months). The degassing ux from the initial 5 weeks of thawing, if
interpreted as being solely comprised of wintertime respiration leaving the system, accounted for 0.8 to
3.2 mmol CO
2
(Figure 2) or 18 to 30% of total emissions. From the CO
2
emissions (Figure 1) and temperature
sensitivity (Figure S6) observed from the cores in this experiment, the high and low values for total CO
2
emis-
sions of these cryosols over a summer season (~3 months) would range from 7.4 molCO
2
m
2
(saturated con-
ditions, 9.5°C) to 15.4 mol m
2
CO
2
(in situ conditions, 9.5°C).
4.3. CH
4
Cycling in the Active Layer and During Permafrost Thaw
4.3.1. Location of Methane Consumption
Break point analyses did not reveal a signicant change in the CH
4
ux from the cores at week 10 for any of
the treatments, including the saturated treatment, even though pore water measurements recorded an
increase in CH
4
at depth upon thawing. In the in situ cores this pore water CH
4
rapidly declined
(Figure 4e), whereas in the saturated cores the high CH
4
concentrations remained stable (Figure 4b). This,
and the absence of elevated CH
4
concentrations at the shallower depths at this time, indicates that aerobic
methanotrophy is focused in the upper active layer and is oxidizing all of the CH
4
released at depth before it
reaches the surface. The surface CH
4
ux measurements are consistent with this observation. CH
4
uxes stea-
dily declined during the rst 5 to 6 weeks and then stabilized for weeks 6 through 13 (Figure 1b). Upon provi-
sion of atmospheric CH
4
concentrations on week 14 the cores rapidly consumed the headspace CH
4
. The
atmospheric CH
4
uptake ux measured in the intact cores overlapped the range of values reported from
eld observations.
4.3.2. Methanotrophic Community
Aerobic methanotrophs comprised 1.2% of the microbial community at 5 cm and declined with depth, with
Type II methanotrophs being most abundant in the top layer and more abundant than Type I methanotrophs.
This is compared to 0.25 to 0.65% relative sequence abundance of methanotrophs in Hess Creek Alaska
[Mackelprang et al., 2011] and 0.1% to 2.4% methanotrophs in Svalbard [Tveit et al., 2012]. High-afnity,
Type II methanotrophs are able to consume CH
4
to <200 ppbv [Duneld et al., 1999]. Lau et al. [2015], using
metagenomic, metatranscriptomic, and metaproteomic analyses of samples from the same polygon, recently
identied the methane monooxygenase enzyme of the active methanotrophic population of atmospheric
CH
4
oxidizing bacteria to be most closely related to the pmo genes of the upland soil cluster α. Samples from
a peat-rich active layer on Herschel Island in the western Canadian Arctic that exhibited aerobic CH
4
oxidation
contain a CH
4
oxidizing community dominated by Type II methanotrophs, primarily Methylocystis and
Methylosinus [Barbier et al., 2012]. Graham et al. [1993], observed that Type II methanotrophs are selected
by N limitation due to their ability to xN
2.
The prevalence of Type II methanotrophs near the top of the
active layer at AHI may be due to the nitrogen-poor C/N ratio in the top 10 cm of ~70:1, whereas at greater
depth in the active layer the C/N ratio is 16:1. Regardless of whether the cores were saturated with water, the
aerobic Type II methanotrophs near the top of the active layer consumed atmospheric CH
4
(Figure 1b). It is
likely that both Type I and II methanotrophs present within the active layer consumed the CH
4
released dur-
ing thawing of the permafrost. Methanogens comprised 0.2% of the microbial community in the permafrost
compared to 0.01 to 0.9% in Svalbard [Tveit et al., 2012] and 0.2 to 4.0% in Hess Creek, Alaska [Mackelprang
et al., 2011]. Nonetheless, microcosm experiments using mineral cryosols from a nearby polygonal terrain
and incubated anaerobically at 4°C yielded CH
4
production rates of 0.3 to 0.7 nmol CH
4
grams dry
weight
1
d
1
[Allan et al., 2014]. Applying this CH
4
production rate to the mineral cryosols of this study,
the observed 400 to 500 nM pore water CH
4
concentrations observed after permafrost thaw (Figures 4b
and 4e) could have been produced in less than 1 day. Our current observations cannot resolve whether
the CH
4
observed upon thawing of the permafrost represented trapped CH
4
or CH
4
produced after thawing
had occurred.
4.3.3. Competitive Inhibition of Methanogenesis
Assuming no source of SO
42
, the changes in the pore water SO
42
concentrations correspond to a microbial
sulfate reduction rate of 1.4 nmols SO
42
g
1
or 2 to 5 times greater than the rate of methanogenesis. SRBs
also comprise ~1% of the microbial community or on the order of 5 times that of the methanogens. These
ndings may be due to the fact that SRB have lower K
s
values for H
2
than methanogens, 1 to 4 μM versus
5to7μM[Kristjansson et al., 1982; Robinson and Tiedje, 1984]. The greater SRB activity may also reect the fact
that the ΔGvalues for H
2
oxidation and acetoclastic SO
42
reduction ranged from 103 to 130 kJ (mole of
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 16
SO
42
)
1
and from 105 to 110 kJ (mole of SO
42
)
1
, respectively (Table 2). The ΔGvalues for autotrophic
and acetoclastic methanogenesis, however, were lower, ranging from 52 to 76 kJ (mole of CH
4
)
1
and from
52 to 61 kJ (mole of CH
4
)
1
, respectively. These latter ΔGvalues lie close to the ΔGfor the generation of ATP
from ADP [Schink,1997].TheseΔG values are greater than those observed in simulated peat bog experiments
[Blodau et al., 2011] probably because of the higher observed H
2
concentrations recorded in our intact core
experiment. The Gibbs free energies for sulfate reduction, methanogenesis, and acetogenesis indicate the
following hierarchy in terms of favorability, H
2
oxidation by sulfate reduction reaction acetate oxidation by
sulfate reduction >H
2
/CO
2
methanogenesis acetoclastic methanogenesis >acetogenesis (Table 2).
Pore water H
2
concentrations were all sub-μM in the permafrost upon thawing where it reached at least 400
to 500 nM before it was rapidly consumed in both the in situ and saturated cores (Figures 4c and 4f), unlike
the CH
4
. The H
2
utilizing SRB and methanogens are energetically favored over the acetoclastic SRB and
methanogens soon after permafrost thaw, but by week 15, the ΔGvalues for H
2
versus acetoclastic SRB
and methanogens are comparable (Table 2). Autotrophic methanogens are typically able to outcompete
acetogenic bacteria for H
2
in the environment [Cord-Ruwisch et al., 1988]. The ΔGvalues for acetogenesis
were the most favorable in the permafrost in the rst week after thawing ranging from 10 to 23 kJ (mole
of acetate)
1
. Prior to thawing, the ΔGvalues for acetogenesis may have been comparable to the minimum
energy required for ATP production from ADP [Schink, 1997]. Kotsyurbenko et al. [2001] found that at tempera-
tures below 10°C and H
2
concentrations above 90 nM, some acetogenic bacteria were able to outcompete
certain autotrophic methanogens. Fermentative production of acetate and H
2
to support the high observed
H
2
in the permafrost was modeled using the following reaction:
C12H23 O2þ10H2O5Hþþ10H2þ6C2H3O2(10)
where the dodecanoic acid concentration was assumed to be 1 μM. For the observed pH of 5.5 and T= 4.5°C
and a 40 μM acetate concentration, the maximum H
2
that was energetically permissible is ~150 nM with
higher H
2
concentrations possible at higher pH. The free energy calculations appear to preclude the observed
high concentrations of H
2
and acetate in the permafrost. If H
2
and acetate are locally consumed by sulfate
reduction, methanogenesis, then fermentation could continue to produce H
2
and acetate in the system.
Acetate has been previously observed as a metabolic end product in Alaskan peat bogs [Duddleston and
Kinney, 2002], likely due to environmental conditions (e.g., temperature) slowing acetoclastic methanogen-
esis. A pH shift from 6 to 4.7 has also been shown to decrease the contribution of acetoclastic methanogen-
esis relative to H
2
/CO
2
methanogenesis [Kotsyurbenko et al., 2007]. Allan et al. [2014] observed for the Colour
Lake cryosols, however, that the CH
4
production rates from H
2
/CO
2
amended microcosms were comparable
to CH
4
production rates from acetate amended microcosms (T= 4°C). Allan et al. [2014] also found that the
methanogenic orders Methanococcales and Methanomicrobiales were highly associated with the high CH
4
-
producing microcosms. Combined, these same orders comprised ~10% and ~50% of methanogenic reads
Table 2. Free Energy of Microbial Reactions
In Situ Saturated
Microbial Redox Reaction Week 1112 Week 15 Week 1112 Week 15
65 cm Acetate + H2O CH4 + HCO354
a
57 56 61
4 H2 + H+ + HCO3CH4 + 3H2O 53 52 64 63
Acetate + SO422 HCO3+HS106 108 107 110
4 H2 + H+ + SO42HS+ 4 H2O 104 103 115 112
4 H2 + 2 HCO3+H+Acetate + 4 H2O 2 5 81
PF Acetate + H2O CH4 + HCO353 53 52 56
4 H2 + H+ + HCO3CH4 + 3 H2O 76 63 72 68
Acetate + SO422 HCO3+HS107 109 105 108
4 H2 + H+ + SO42HS+ 4 H2O 130 118 126 120
4 H2 + 2 HCO3+H+Acetate + 4 H2O 23 10 21 12
a
Free energy in kJ mol-CH
41
(acetoclastic methanogenesis and hydrogenotrophic methanogenesis), kJ mol-SO
41
(sulfate reduction with acetate, sulfate reduction with H2), or kJ mol-acetate
1
(acetogenesis). The values were calcu-
lated by GWB based upon the pore chemistry data and a pH of 5.5.
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 17
in this experiment, respectively, for the 65 cm and permafrost depths. These observations are consistent with
the similarity in the ΔGvalues for autotrophic versus acetoclastic methanogens in the 65 cm layer and the
permafrost at week 15.
4.4.4. Projected Timescale of Methanogenesis Inhibition by Sulfate
Although many sites in the Arctic have been shown to be net sources of CH
4
to the atmosphere (Table S10 in
the supporting information) and are feared to transition to a substantial CH
4
source with severe environmen-
tal and economic consequences as a result of climate change [Euskirchen et al., 2013], this study shows that
conditions specic to the site could mitigate future CH
4
emissions. Although CH
4
drawdown from the head-
space can be explained in in situ treatments by the deepening of aerobic active layer and inhibition of metha-
nogenesis with O
2
, water saturated soils with limited O
2
diffusion were also found to be sinks of CH
4
. In this
situation it is the chemistry of the system, specically the millimolar SO
42
concentrations, that allows SRBs to
compete with methanogenesis for H
2
and acetate to the extent that aerobic methanotrophy can maintain
the soils as a CH
4
sink. If the SO
42
in the pore water has no mechanism for replacement, however, these per-
mafrost soils could then run out of this electron acceptor over decadal timescales. This would occur before
the depletion of the SOC that sustains both sulfate reduction and methanogenesis through fermentation
under anaerobic conditions. This would result in a signicant shift in the methanotrophy/methanogenesis
balance and lead to CH
4
emissions from the soil. Alternatively, if the total S in the core soil (0.24 ± 0.06%
by weight; Table S4) replenishes the SO
42
pool in the pore water, then the rates of sulfate reduction
observed in this experiment would fully reduce the sulfate in ~1000 years given 3 months of thaw. This is
longer than the 320 to 640 year estimated turnover time for the SOC by aerobic heterotrophy and would
allow sulfate reduction to remain energetically competitive with methanogenesis over a long timescale.
4.4. Nitrogen Limitation in Mineral Cryosol
Organic-rich cryosols systems have been shown to be N limited and to efciently recycle N species [Schimel
et al., 2004; Sistla et al., 2012], although in some systems NO
3
export by ground water ow does occur
[Harms and Jones, 2012]. The metagenomic analyses of our mineral cryosol yielded functional genes for
the complete reduction of NO
3
to N
2
(nitrate reductase, nitrite reductase, nitric oxide reductase, and nitrous
oxide reductase) as well as functional genes associated with N
2
xation (nitrogenase), and nitrication
(ammonia oxidase and hydroxylamine oxidase).
The observed loss of NH
4+
in the permafrost at 80 cm depth upon thawing, in both the in situ and saturated
treatments, occurred without any corresponding increase in NO
2
or NO
3
. Measurements of N
2
O in the
headspace were made for the rst 2 weeks of thawing, but no emissions of N
2
O were observed above the
detection limit of 100 ppbv (data not shown). The existence of the full pathway for nitrication and denitri-
cation suggests that the initial NH
4+
pool released from the permafrost was either oxidized to NO
3
and sub-
sequently reduced to N
2
within 3 weeks or that the NH
4+
was taken up directly by the biomass during
anabolism or both. Aerobic respiration was estimated to add 0.005 to 0.014 g of biomass per core, which
would require ~50 to 150 μmol of additional N. This increase in biomass is sufcient to entirely consume
the ~70 μmol NH
4+
pool available in the core. This indicates that this environment is N limiting for microbial
growth and likely depends upon an N
2
-xing community for meeting additional growth demands.
5. Conclusions
Intact core experiments were used to simulate the seasonal development of an active layer and thawing of
sparsely vegetated, SOC-poor mineral cryosol from the Canadian high Arctic while being subjected to dif-
ferent moisture, temperature, and light regimes. The CO
2
and CH
4
uxes measured from the intact cores
were consistent with those determined in the eld from the same site. By characterizing the aqueous
and gaseous geochemistry and microbial community composition in a vertical prole for each regime
we were able to delineate the biogechemical processes that occurred at depth that accounted for the
observed surface uxes.
The results indicate that during the initial thaw the active layer, the CO
2
and CH
4
emissions represent the
release of gas trapped during the end summer freeze-in of the soil or formed during winter respiration.
This process has been reported for SOC-rich fens in Greenland, but in the case of the SOC-poor mineral cryo-
sols of this study the amount of CO
2
and CH
4
released was less.
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 18
The primary control on post thaw CO
2
emissions was water saturation. High saturation signicantly reduced
O
2
uptake into the soil and limited aerobic respiration that stoichiometrically converted the SOC to CO
2
. Over
the course of the experiment the CO
2
emissions from the saturated cores were 26 to 65% lower than that of
the in situ cores. Temperature increases were found to have a positive effect on CO
2
emission under all con-
ditions, with most Q
10
values ranging from 5 to 11, suggesting a high level of sensitivity to moderate changes
in the temperature range of 1 to 5°C. Photosynthetic uptake of CO
2
was observed, but it was not sufcient to
overcome the CO
2
emissions. This stands in stark contrast to Arctic sites in Greenland, Siberia, and Sweden,
where net CO
2
uxes show uptake by Arctic soils (Table S10). This study found that thawing of the permafrost
was also associated with higher cumulative CO
2
emissions during the course of the experiment as long as the
cores were not water saturated. Cores with intact, frozen permafrost released 31 to 67% less CO
2
than the in
situ cores with thawed permafrost. During the simulated summer thaw the δ
13
CofCO
2
emissions closely
matched the isotopic signature of the SOC, indicating full usage of the SOC without substantial fractionation.
Interestingly, saturation did not appear to play a major role in CH
4
ux, as the mineral cryosol under all con-
ditions, when exposed to atmospheric concentrations of CH
4
, was CH
4
sinks. The CH
4
produced at depth in
the active layer or released of from thawing permafrost was entirely oxidized by aerobic methanotrophs
before reaching the atmosphere. Although a rise in temperature is generally thought to favor an increase
in CH
4
emissions to the atmosphere, in Axel Heiberg Island mineral cryosols the rates of methanogenesis
are likely to remain subordinate to those of sulfate reduction due to the high S content of the mineral cryosol
providing a SO
42
source that allows sulfate reducers to successfully compete with methanogens for acetate
and H
2
.
In terms of CO
2
emissions and atmospheric CH
4
uptake, the sparsely vegetated mineral cryosols examined
from Colour Lake on Axel Heiberg Island are distinct from organic-rich cryosols that have been the focus of
most Arctic biogeochemical studies. Given the greater spatial extent of SOC-poor mineral cryosols compared
to SOC-rich cryosols, further studies are needed within the context of anticipated hydrological, geochemical,
and microbial community changes associated with global warming.
References
Allan, J., J. Ronholm, N. S. Mykytczuk, C. W. Greer, T. C. Onstott, and L. G. Whyte (2014), Methanogen community composition and rates of
methane consumption in Canadian High Arctic permafrost soils, Environ. Microbiol. Rep.,6, 136144, doi:10.1111/1758-2229.12139.
Allison, S. D., and K. K. Treseder (2008), Warming and drying suppress microbial activity and carbon cycling in boreal forest soils, Global
Change Biol.,14(12), 28982909, doi:10.1111/j.1365-2486.2008.01716.x.
Barbier, B., I. Dziduch,S. Liebner, L. Ganzert, H.Lantuit, W. Pollard, andD. Wagner (2012), Methane-cyclingcommunities in a permafrost-affected
soil on Herschel Island, Western Canadian Arctic: Active layer proling of mcrA and pmoA genes, FEMS Microbiol. Ecol.,82(2), 287302,
doi:10.1111/j.1574-6941.2012.01332.x.
Bethke, C. M. (2008), Geochemical and Biogeochemical Reaction Modeling, 2nd ed., Univ. of Illinois, Urbana, Illinois.
Biasi, C., O. Rusalimova, H. Meyer, C. Kaiser, W. Wanek, P. Barsukov, H. Junger, and A. Richter (2005), Temperature-dependent shift from labile
to recalcitrant carbon sources of arctic heterotrophs, Rapid Commun. Mass Spectrom.,19(11), 14011408, doi:10.1002/rcm.1911.
Blodau, C., M. Siems, and J. Beer (2011), Experimental burial inhibits methanogenesis and anaerobic decomposition in water-saturated peats,
Environ. Sci. Technol.,45(23), 99849989, doi:10.1021/es201777u.
Blöthe, M., D. M. Akob, J. E. Kostka, K. Göschel, H. L. Drake, and K. Küsel (2008), pH gradient-induced heterogeneity of Fe(III)-reducing
microorganisms in coal mining-associated lake sediments, Appl. Environ. Microbiol.,74(4), 10191029, doi:10.1128/AEM.01194-07.
Brummell, M. E., R. E. Farrell, and S. D. Siciliano (2012), Greenhouse gas soil production and surface uxes at a high arctic polar oasis, Soil Biol.
Biochem.,52,112, doi:10.1016/j.soilbio.2012.03.019.
Buttle, J., and K. Fraser (1992), Hydrochemical uxes in a high arctic wetland basin during spring snowmelt, Arct. Alp. Res.,24(2), 153164.
Christensen, T., T. Friborg, M. Sommerkorn, J. Kaplan, L. Illeris, H. Soegaard, C. Nordstroem, and S. Jonasson (2000), Trace gas exchange in a
high Arctic valley: 1. Variations in CO
2
and CH
4
Flux between tundra vegetation types, Global Biogeochem. Cycles,14(3), 701713,
doi:10.1029/1999GB001134.
Conant, R. T., et al. (2011), Temperature and soil organic matter decomposition ratesSynthesis of current knowledge and a way forward,
Global Change Biol.,17(11), 33923404, doi:10.1111/j.1365-2486.2011.02496.x.
Coolen, M. J. L., J. van de Giessen, E. Y. Zhu, and C. Wuchter (2011), Bioavailability of soil organic matter and microbial community dynamics
upon permafrost thaw, Environ. Microbiol.,13(8), 22992314, doi:10.1111/j.1462-2920.2011.02489.x.
Cord-Ruwisch, R., H. Seitz, and R. Conrad (1988), The capacity of hydrogenotrophic anaerobic bacteria to compete for traces of hydrogen
depends on the redox potential of the terminal electron acceptor, Arch. Microbiol.,149, 350357.
Cussler, E. L. (1985), Diffusion: Mass Transfer in Fluid Systems, Cambridge Univ. Press, New York.
Czimczik, C. I., and J. M. Welker (2010), Radiocarbon content of CO
2
respired from High Arctic tundra in northwest Greenland, Arct. Antarct.
Alp. Res.,42(3), 342350, doi:10.1657/1938-4246-42.3.342.
Davidson, E., and I. Janssens (2006), Temperature sensitivity of soil carbon decomposition and feedbacks to climate change, Nature,
440(7081), 16573, doi:10.1038/nature04514.
Duddleston, K., and M. Kinney (2002), Anaerobic microbial biogeochemistry in a northern bog: Acetate as a dominant metabolic end
product. Global Biogeochem. Cycles,16(4), 1063, doi:10.1029/2001GB001402.
Acknowledgments
The MG-RAST database sequence
identiers used for metagenomic data
analyzed in this manuscript can be
found in the Method section in the
supporting information. Geochemical
data, pore water concentrations, and
pore gas concentrations supporting
Figure 4 and used for the calculation of
ΔG values and the contribution of
metabolic pathways can be found in
Tables S4 to S9 in the supporting
information. This material is based upon
work supported by the U.S. Department
of Energy, Ofce of Science, Ofce of
Biological and Environmental Research
Genomic Science program under award
DE-SC0004902 to T.C.O., S.M.P., and
L.G.W. Canadian Polar Continental Shelf
Program, NSERC Northern Supplement,
and Northern Scientic Training Program
grants to L.G.W. provided additional
logistical support for high Arctic eld
research. Guillaume Lamarche-Gagnon,
Roland Wilhelm, and Heather Cray
provided eld assistance. We thank
M. Bender of the Dept. of Geosciences,
Prince ton Unive rsity fo r use o f his 30°C
ice core lab.
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 19
Duneld, P., W. Liesack, T. Henckel, R. Knowles, and R. Conrad (1999), High-afnity methane oxidation by a soil enrichmen t culture containing
a type II methanotroph, Appl. Environ. Microbiol.,65(3), 10091014.
Epstein, H. E., M. K. Raynolds, D. Walker, U. S. Bhatt, C. J. Tucker, and J. E. Pinzon (2012), Dynamics of aboveground phytomass of the
circumpolar Arctic tundra during the past three decades, Environ. Res. Lett.,7(1), 015506, doi:10.1088/ 1748-9326/7/1/015506.
Euskirchen, E., E. Goodstein, and H. Huntington (2013), An estimated cost of lost climate regulation services caused by thawing of the arctic
cryosphere, Ecol. Appl.,23(8), 18691880.
Graham, D., J. Chaudhary, R. Hanson, and R. Arnold (1993), Factors affecting competition between type I and type II methanotrophs in
two-organism, continuous-ow reactors, Microb. Ecol.,25(1), 117, doi:10.1007/BF00182126.
Harms, T. K., and J. B. Jones (2012), Thaw depth determines reaction and transport of inorganic nitrogen in valley bottom permafrost soils,
Global Change Biol.,18(9), 29582968, doi:10.1111/j.1365-2486.2012.02731.x.
Heimann, M., and M. Reichstein (2008), Terrestrial ecosystem carbon dynamics and climate feedbacks, Nature,451(7176), 28992,
doi:10.1038/nature06591.
Hugelius, G., et al. (2013a), A new data set for estimating organic carbon storage to 3 m depth in soils of thenorthern circumpolar permafrost
region, Earth Syst. Sci. Data,5(2), 393402, doi:10.5194/essd-5-393-2013.
Hugelius, G., C. Tarnocai, G. Broll, J. G. Canadell, P. Kuhry, and D. K. Swanson (2013b), The northern circump olar soil carbon database: Spatially
distributed datasets of soil coverage and soil carbon storage in the northern permafrost regions, Earth Syst. Sci. Data,5(1), 313,
doi:10.5194/essd-5-3-2013.
Hugelius, G., et al. (2014), Improved estimates show large circumpolar stocks of perm afrost carbon while quantifying substantial uncertainty
ranges and identifying remaining data gaps, Biogeosci. Discuss.,11(3), 47714822, doi:10.5194/bgd-11-4771-2014.
Jørgensen, C. J., K. M. Lund Johansen, A. Westergaard-Nielsen, and B. Elberling (2014), Net regional methane sink in High Arctic soils of
northeast Greenland, Nat. Geosci.,8,2023, doi:10.1038/ngeo2305.
Jorgenson, M. T., Y. L. Shur, and E. R. Pullman (2006), Abrupt increase in permafrost degradation in Arctic Alaska, Geophys. Res. Lett.,33,
L02503, doi:10.1029/2005GL024960.
Knorr, K. H., and C. Blodau (2009), Impact of experimental drought and rewetting on redox transformations and methanogenesis in
mesocosms of a northern fen soil, Soil Biol. Biochem.,41(6), 11871198, doi:10.1016/j.soilbio.2009.02.030.
Kotsyurbenko, O. R., M. V. Glagolev, A. N. Nozhevnikova, and R. Conrad (2001), Competition between homoacetogenic bacteria and
methanogenic archaea for hydrogen at low temperature, FEMS Microbiol. Ecol.,38(23), 153159, doi:10.1111/j.1 574-6941.2001.tb00893.x.
Kotsyurbenko, O. R., M. W. Friedrich, M. V. Simankova, A. N. Nozhevnikova, P. N. Golyshin, K. N. Timmis, and R. Conrad (2007), Shift from
acetoclastic to H2-dependent methanogenesis in a west Siberian peat bog at low pH values and isolation of an acidophilic
Methanobacterium strain, Appl. Environ. Microbiol.,73(7), 23442348, doi:10.1128/AEM.02413-06.
Koven, C. D., B. Ringeval, P. Friedlingstein, P. Ciais, P. Cadule, D. Khvorostyanov, G. Krinner, and C. Tarnocai (2011), Permafrost carbon -climate
feedbacks accelerate global warming, Proc. Natl. Acad. Sci. U.S.A.,108(36), 14,76914,774, doi:10.1073/pnas.1103910108.
Kristjansson, J., P. Schönheit, and R. Thauer (1982), Different Ks values for hydrogen of methanogenic bacteria and sulfate reducing bacteria:
An explanation for the apparent inhibition of methanogenesis by sulfate, Arch. Microbiol.,131, 278282.
Kutzbach, L., C. Wille, and E.-M. Pfeiffer (2007), The exchange of carbon dioxide between wet arctic tundra and the atmosphere at the Lena
River Delta, Northern Siberia, Biogeosciences,4(5), 869890, doi:10.5194/bg-4-869-2007.
Laeur, P. M., E. R. Humphreys, V. L. S. Louis, M. C. Myklebust, T. Papakyriakou, L. Poissant, J. D. Barker, M. Pilote, and K. A. Swystun (2012),
Variation in peak growing season net ecosystem production across the Canadian Arctic, Environ. Sci. Technol.,46(15), 79717977,
doi:10.1021/es300500m.
Lau, C. Y. M., B. T. Stackhouse, A. C. Layton, A. Chauhan, T. A. Vishnivetskaya, K. Chourey, N. C. S. Mykytczuk, P. C. Bennett, L. G. Whyte, and
T. C. Onstott (2015), An active atmospheric methane sink in high Arctic mineral cryosols, Int. Soc. Microb. Ecol.,9(8), 18801891.
Lu, A., et al. (2012), Growth of non-phototrophic microorganisms using solar energy through mineral photocatalysis, Nat. Commun.,3, 768,
doi:10.1038/ncomms1768.
Mack, M. C., E. A. Schuur, M. S. Bret-Harte, G. R. Shaver, and F. S. Chapin (2004), Ecosystem carbon storag e in arctic tundra reduced by
long-term nutrient fertilization, Nature,431(7007), 440443, doi:10.1038/nature02887.
Mackelprang, R., M. P. Waldrop, K. M. DeAngelis, M. M. David, K. L. Chavarria, S. J. Blazewicz, E. M. Rubin, and J. K. Jansson (2011),
Metagenomic analysis of a permafrost microbial community reveals a rapid response to thaw, Nature,480(7377), 368371, doi:10.1038/
nature10576.
Martineau, C., Y. Pan, L. Bodrossy, E. Yergeau, L. G. Whyte, and C. W. Greer (2014), Atmospheric methane oxidizers are present and active in
Canadian high Arctic soils, FEMS Microbiol. Ecol.,113, doi:10.1111/1574-6941.12287.
Mastepanov, M., C. Sigsgaard, E. J. Dlugokencky, S. Houweling, L. Ström, M. P. Tamstorf, and T. R. Christensen (2008), Large tundra methane
burst during onset of freezing, Nature,456(7222), 62830, doi:10.1038/nature07464.
Mastepanov, M., C. Sigsgaard, T. Tagesson, L. Ström, M. P. Tamstorf, M. Lund, and T. R. Christensen (2013), Revisiting factors controlling
methane emissions from high-Arctic tundra, Biogeosciences,10(11), 51395158, doi:10.5194/bg-10-5139-2013.
Mikan, C. J., J. P. Schimel, and A. P. Doyle (2002), Temperature controls of microbial respiration in arctic tundra soils above and below
freezing, Soil Biol. Biochem.,34(11), 17851795, doi:10.1016/S0038-0717(02)00168-2.
Montzka, S., E. Dlugokencky, and J. Butler (2011), Non-CO2 greenhouse gases and climate change, Nature,476(7358), 4350, doi:10.1038/
nature10322.
Mook, W., J. Bommerson, and W. Staverman (1974), Carbon isotope fractionation between dissolved bicarbonate and gaseous carbon
dioxide, Earth Planet. Sci. Lett.,22, 169176.
Morgner, E., B. Elberling, D. Strebel, and E. J. Cooper (2010), The importance of winter in annual ecosystem respiration in the High Arctic:
Effects of snow depth in two vegetation types, Polar Res.,29(2007), 5874, doi:10.1111/j.1751-8369.2010. 00151.x.
Muggeo, V. (2008), Segmented: An R package to t regression models with broken-line relationships, R News,8(1), 2025.
Mykytczuk, N. C., S. J. Foote, C. R. Omelon, G. Southam, C. W. Greer, and L. G. Whyte (2013), Bacterial growth at 15 °C; molecular insights
from the permafrost bacterium Planococcus halocryophilus Or1, ISME J.,7(6), 12111226, doi:10.1038/ismej.2013.8.
Olefeldt, D., N. T. Roulet, O. Bergeron, P. Crill, K. Bäckstrand, and T. R. Christensen (2012), Net carbon accumulation of a high-latitude
permafrost palsa mire similar to permafrost-free peatlands, Geophys. Res. Lett.,39, L03501, doi:10.1029 /2011GL050355.
Osterkamp, T. E. (2007), Characteristics of the recent warming of permafrost in Alaska, J. Geophys. Res.,112, F02S02, doi:10.1029/2006JF000578.
Parks, D. H., and R. G. Beiko (2010), Identifying biologically relevant differences between metagenomic communities, Bioinformatics,26(6),
71521, doi:10.1093/bioinformatics/btq041.
Parsons, T. R., Y. Maita, and C. M. Lalli (1984), A Manual of Chemical and Biological Methods for Seawater Analysis, Pergamon Press, Potts Point
N.S.W. Australia.
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 20
Pautler, B. G., A. J. Simpson, D. J. McNally, S. F. Lamoureux, and M. J. Simpson (2010), Arctic permafrost active layer detachments stimulate
microbial activity and degradation of soil organic matter, Environ. Sci. Technol.,44(11), 40764082, doi:10.1021/es903685j.
Phelps, T. J., E. M. Murphy, S. M. Pffner, and D. C. White (1994), Comparison between geochemical and biological estimates of subsurface
microbial activities, Microb. Ecol.,28(3), 335349, doi:10.1007/BF00662027.
Price, P. B., and T. Sowers (2004), Temperature dependence of metabolic rates for microbial growth, maintenance, and survival, Proc. Natl.
Acad. Sci. U.S.A.,101(13), 46314636, doi:10.1073/pnas.0400522101.
Rivkina, E., V. Shcherbakova, K. Laurinavichius, L. Petrovskaya, K. Krivushin, G.Kraev, S. Pecheritsina, and D. Gilichinsky (2007), Biogeochemistry of
methane and methanogenic archaea in permafrost, FEMS Microbiol. Ecol.,61(1), 115, doi:10.1111/j.1574-6941.2007.00315.x.
Robinson, J., and J. Tiedje (1984), Competition between sulfate-reducing and methanogenic bacteria for H2 under resting and growing
conditions, Arch. Microbiol.,137,2632.
Rowlands, D. J., et al. (2012), Broad range of 2050 warming from an observationally constrained large climate model ensemble, Nat. Geosci.,
5(4), 256260, doi:10.1038/ngeo1430.
Schaefer, K., T. Zhang, L. Bruhwiler, and A. P. Barrett (2011), Amount and timing of permafrost carbon release in response to climate warming,
Tellus B,63(2), 165180, doi:10.1111/j.1600-0889.2011.00527.x.
Schimel, J. P., C. Bilbrough, and J. M. Welker (2004), Increased snow depth affects microbial activity and nitrogen mineralization in two Arc tic
tundra communities, Soil Biol. Biochem.,36(2), 217227, doi:10.1016/j.soilbio.2003.09.008.
Schink, B. (1997), Energetics of syntrophic cooperation in methanogenic degradation, Microbiol. Mol. Biol. Rev.,61(2), 262280.
Schuur, E. A. G., et al. (2008), Vulnerability of permafrost carbon to climate change: Implications for the global carbon cycle, BioScience,58(8),
701, doi:10.1641/B580807.
Schuur, E. A., J. G. Vogel, K. G. Crummer, H. Lee, J. O. Sickman, and T. E. Osterkamp (2009), The effect of permafrost thaw on old carbon release
and net carbon exchange from tundra, Nature,459, 5569, doi:10.1038/nature08031.
Shindell, D. T., G. Faluvegi, D. M. Koch, G. A. Schmidt, N. Unger, and S. E. Bauer (2009), Improved attribution of climate forcing to emissions,
Science,326, 716718, doi:10.1126/science.1174760.
Sistla, S., S. Asao, and J. P. Schimel (2012), Detecting microbial N-limitation in tussock tundra soil: Implications for Arctic soil organic carbon
cycling, Soil Biol. Biochem.,55,7884, doi:10.1016/j.soilbio.2012.06.010.
Tarnocai, C., and J. Bockheim (2011), Cryosolic soils of Canada: Genesis, distribution, and classication, Can. J. Soil Sci.,91(5), 749762,
doi:10.4141/cjss10020.
Tveit, A., R. Schwacke, M. M. Svenning, and T. Urich (2012), Organic carbon transformations in high-Arctic peat soils: Key functions and
microorganisms, ISME J.,7(2), 299311, doi:10.1038/ismej.2012.99.
Van der Molen, M. K., J. van Huissteden, F. J. W. Parmentier, A. M. R. Petrescu, A. J. Dolman, T. C. Maximov, A. V. Kononov, S. V. Karsanaev, and
D. A. Suzdalov (2007), The growing season greenhouse gas balance of a continental tundra site in the Indigirka lowlands, NE Siberia,
Biogeosciences,4(6), 9851003, doi:10.5194/bg-4-985-2007.
Verville, J., S. Hobbie, F. Chapin III, and D. Hooper (1998), Response of tundra CH
4
and CO
2
ux to manipulation of temperature and
vegetation, BiogEochemistry,4, 215235.
Vishnivetskaya, T. A., et al. (2014), Commercial DNA extraction kits impact observed microbial community composition in permafrost
samples, FEMS Microbiol. Ecol.,87(1), 217230, doi:10.1111/1574-6941.12219.
Wagner, D., A. Lipski, A. Embacher, and A. Gattinger (2005), Methane uxes in permafrost habitats of the Lena Delta: effects of microbial
community structure and organic matter quality, Environ. Microbiol.,7(10), 15821592, doi:10.1111 /j.1462-2920.2005.00849.x.
Walter, K. M., S. Zimov, J. P. Chanton, D. Verbyla, and F. S. Chapin (2006), Methane bubbling from Siberian thaw lakes as a positive feedback to
climate warming, Nature,443(7107), 715, doi:10.1038/nature05040.
Wilhelm, R. C., T. D. Niederberger, C. Greer, and L. G. Whyte (2011), Microbial diversity of activ e layer and permafrost in an acidic wetland from
the Canadian High Arctic, Can. J. Microbiol.,57(4), 3033015, doi:10.1139/w11-004.
Wilhelm, R. C., K. J. Radtke, N. C. S. Mykytczuk, C. W. Greer, and L. G. Whyte (2012), Life at the wedge: The activity and diversity of arctic ice
wedge microbial communities, AstrObiology,12(4), 34760, doi:10.1089/ast.2011.0730.
Worthy, D., and I. Levin (2000), Evidence for a link between climate and northern wetland methane emissions, J. Geophys. Res.,105,
40314038, doi:10.1029/1999JD901100.
Zhuang, Q., J. M. Melillo, D. W. Kicklighter, R. G. Prinn, A. D. McGuire, P. A. Steudler, B. S. Felzer, and S. Hu (2004), Methane uxes between
terrestrial ecosystems and the atmosphere at northern high latitudes during the past century: A retrospective analysis with a process-based
biogeochemistry model, Global Biogeochem. Cycles,18, GB3010, doi:10.1029/2004GB002239.
Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003004
STACKHOUSE ET AL. SIMULATED SPRING THAW OF PERMAFROST 21
... As an example, Antarctic, marine-derived lake communities have been shown to have evolved independently over their relatively short history of 3000-5000 years, adapting not just to low temperature but also to a variety of important environmental factors specific to each lake system (reviewed in Ref. [10]). Metagenomic analyses have also begun to be used to uncover the ways in which communities in polar environments respond to changing environmental conditions; for example, the effects of the seasonal polar sunlight cycle on Antarctic marine and marine-derived lake communities [11,12] and the roles that Arctic bacteria play in melting permafrost acting as a CO 2 source and atmospheric CH 4 sink [13,14]. ...
... To define variables that may explain the niche adaptation of Group C in the Axel Heiberg Island permafrost, available abiotic and biotic data were used from the permafrost study [13,14]. A range of physicochemical data were available for each of the four depths (5, 35, 65, and 80 cm), but as the timing of sampling for physicochemical data (0, 4, 6, 8, 11, and 12 weeks) did not align with the timing of sampling for the metagenomes (0, 0.25, 6, 12, and 18 months), the physicochemical data were ultimately not useful for interpreting Group C distribution. ...
... Attempting to identify specific niche conditions is not trivial. For the Axel Heiberg Island study, the permafrost microbial community was reported to be dominated by Actinobacteria and Proteobacteria, with significant increases at depth for Firmicutes and Actinobacteria and significant decreases for Acidobacteria, Proteobacteria, and Verrucomicrobia [14]. However, despite these taxonomic differences, we did not identify significant predicted functional differences by depth. ...
Article
Full-text available
Background Microorganisms drive critical global biogeochemical cycles and dominate the biomass in Earth’s expansive cold biosphere. Determining the genomic traits that enable psychrophiles to grow in cold environments informs about their physiology and adaptive responses. However, defining important genomic traits of psychrophiles has proven difficult, with the ability to extrapolate genomic knowledge to environmental relevance proving even more difficult. Results Here we examined the bacterial genus Arthrobacter and, assisted by genome sequences of new Tibetan Plateau isolates, defined a new clade, Group C, that represents isolates from polar and alpine environments. Group C had a superior ability to grow at −1°C and possessed genome G+C content, amino acid composition, predicted protein stability, and functional capacities (e.g., sulfur metabolism and mycothiol biosynthesis) that distinguished it from non-polar or alpine Group A Arthrobacter . Interrogation of nearly 1000 metagenomes identified an over-representation of Group C in Canadian permafrost communities from a simulated spring-thaw experiment, indicative of niche adaptation, and an under-representation of Group A in all polar and alpine samples, indicative of a general response to environmental temperature. Conclusion The findings illustrate a capacity to define genomic markers of specific taxa that potentially have value for environmental monitoring of cold environments, including environmental change arising from anthropogenic impact. More broadly, the study illustrates the challenges involved in extrapolating from genomic and physiological data to an environmental setting.
... Using these methods, nitrifying genes were found in the active layer of permafrost-affected soils in high abundance, comparable to other ecosystems, but with low diversity [120,174]. Nitrifiers have also been detected in frozen permafrost cores [101,150,174,183,184] with gene abundance increasing after thawing [185] but may have been lost in ice rich sediments (Yedoma, [87,186]). ...
... In addition, permafrost samples contained denitrifying genes, as shown in metagenome studies [135]. As in the active layer, the relative abundance of the last denitrification step was low in frozen permafrost, which may lead to the accumulation of N2O [101,174,183]. The gene abundance of denitrifiers increases after thawing [185], but not during long-term incubations [206]. ...
Article
Full-text available
Permafrost-affected tundra soils are large carbon (C) and nitrogen (N) reservoirs. However, N is largely bound in soil organic matter (SOM), and ecosystems generally have low N availability. Therefore, microbial induced N-cycling processes and N losses were considered negligible. Recent studies show that microbial N processing rates, inorganic N availability, and lateral N losses from thawing permafrost increase when vegetation cover is disturbed, resulting in reduced N uptake or increased N input from thawing permafrost. In this review, we describe currently known N hotspots, particularly bare patches in permafrost peatland or permafrost soils affected by thermokarst, and their microbiogeochemical characteristics, and present evidence for previously unrecorded N hotspots in the tundra. We summarize the current understanding of microbial N cycling processes that promote the release of the potent greenhouse gas (GHG) nitrous oxide (N2O) and the translocation of inorganic N from terrestrial into aquatic ecosystems. We suggest that certain soil characteristics and microbial traits can be used as indicators of N availability and N losses. Identifying N hotspots in permafrost soils is key to assessing the potential for N release from permafrost-affected soils under global warming, as well as the impact of increased N availability on emissions of carbon-containing GHGs.
... We employed read recruitment to compute the relative abundance of the UBA10452 lineage across the metagenomics datasets from which the MAGs were originally recovered. These datasets consisted of 10 Illumina NextSeq metagenomes from tundra soils in Rásttigáisá, Norway (this study); 69 Illumina NextSeq/NovaSeq metagenomes from tundra soils in Kilpisjärvi, Finland (Pessi et al. 2022); 13 Illumina HiSeq metagenomes from permafrost soils in Nunavut, Canada (Chauhan et al. 2014, Stackhouse et al. 2015; and three Illumina HiSeq metagenomes from polar desert soils in Wilkes Land, Antarctica (Ji et al. 2017). We used fasterqdump v2.10.8 (https://github.com/ncbi/sra-tools) to retrieve the raw metagenomic data from the Sequence Read Archive (SRA). ...
... carbon (C; 1.0%-7.3%), and N (0.1%-0.3%) content (Stackhouse et al. 2015, Ji et al. 2017, Pessi et al. 2022. Furthermore, the abundance profile of Ca. ...
Article
Full-text available
Ammonia-oxidizing archaea (AOA) are key players in the nitrogen cycle of polar soils. Here, we analyzed metagenomic data from tundra soils in Rásttigáisá, Norway, and recovered four metagenome-assembled genomes (MAGs) assigned to the genus ‘UBA10452’, an uncultured lineage of putative AOA in the order Nitrososphaerales (‘terrestrial group I.1b’), phylum Thaumarchaeota. Analysis of other eight previously reported MAGs and publicly available amplicon sequencing data revealed that the UBA10452 lineage is predominantly found in acidic polar and alpine soils. In particular, UBA10452 MAGs were more abundant in highly oligotrophic environments such as mineral permafrost than in more nutrient-rich, vegetated tundra soils. UBA10452 MAGs harbour multiple copies of genes related to cold tolerance, particularly genes involved in DNA replication and repair. Based on the phylogenetic, biogeographic, and ecological characteristics of 12 UBA10452 MAGs, which include a high-quality MAG (90.8% complete, 3.9% redundant) with a nearly complete 16S rRNA gene, we propose a novel Candidatus genus, Ca. Nitrosopolaris, with four species representing clear biogeographic/habitat clusters.
... Microorganisms are responsible for many important processes, including nutrient mineralization (bacteria) and dimethyl sulfoniopropionate (DMSP) production (phytoplankton named Phaeocystis and other ice algae and phytoplankton). All microorganisms produce greenhouse gas carbon dioxide (CO 2 ), and some produce methane (CH 4 ) [3][4][5][6][7][8]. Ny-Ålesund, called a "glacier museum", is located at the west of the Svalbard Archipelago (74~81° N, 10~35° E, Figure 2). ...
Article
Full-text available
Ny-Ålesund in Svalbard is a complex area with both continental and tidal glaciers. There are a lot of studies on prokaryotic and eukaryotic communities in coastal water and soil, but without studies in glacial-related waters. We make a distinctive and consolidated study on the structure of the prokaryotic and eukaryotic communities of pure glacial meltwater, glacial melting lake, glacial meltwater flowing via different types of soil at various elevations, estuarine glacial water and marine water. Moreover, we analyze the environmental–microbial relationships of the prokaryotic and eukaryotic communities via a canonical correspondence analysis and redundant analysis compared by a Pearson analysis. We found that there were distinct microbes in different environments. Altitude had significant correlations with prokaryotic and eukaryotic species in the 12 water samples (ppro = 0.001, npro = 1010, and peuk = 0.012, npro = 1651) (Pearson analysis). Altitude, temperature and salinity, respectively, accounted for 28.27%, 10.86% and 8.24% in the prokaryotic community structure and 25.77%, 17.72% and 3.46% in the eukaryotic, respectively, in water. Nitrogen, silicate and pH accounted for 38.15%, 6.15% and 2.48% in the prokaryotic community structure in soil and 26.65%, 12.78% and 8.66% in the eukaryotic. Eukaryotes were more stable than prokaryotes in changing environments. Cyanobacteria and dinoflagellates better adapt to a warming environment. Gammaproteobacteria and Chrysophysceae were most abundant in soil. Alphaproteobacteria, Bacteroidia, Mamiellophyceae and Prasinophytae were most abundant in water. Within these microbes, Bacilli and Chlorophyceae were only found in glaciers; Actinobacteria, KD94-96, Thermleophilia, Embryophyta, Trebouxiophyceae and Sordariomycetes were unique to soil.
... The site is characterized by a high centered icewedge polygon terrain. The soils are low in organic and water content and are sparsely vegetated with Sphagnum, sedges, and cotton grass 23,31 . The in situ CH 4 soil gas flux measurements were performed using a static chamber system and analyzed as previously described 24,32 . ...
Article
Full-text available
Greenhouse gas (GHG) emissions from Arctic permafrost soils create a positive feedback loop of climate warming and further GHG emissions. Active methane uptake in these soils can reduce the impact of GHG on future Arctic warming potential. Aerobic methane oxidizers are thought to be responsible for this apparent methane sink, though Arctic representatives of these organisms have resisted culturing efforts. Here, we first used in situ gas flux measurements and qPCR to identify relative methane sink hotspots at a high Arctic cytosol site, we then labeled the active microbiome in situ using DNA Stable Isotope Probing (SIP) with heavy ¹³ CH 4 (at 100 ppm and 1000 ppm). This was followed by amplicon and metagenome sequencing to identify active organisms involved in CH 4 metabolism in these high Arctic cryosols. Sequencing of ¹³ C-labeled pmoA genes demonstrated that type II methanotrophs ( Methylocapsa ) were overall the dominant active methane oxidizers in these mineral cryosols, while type I methanotrophs ( Methylomarinovum ) were only detected in the 100 ppm SIP treatment. From the SIP- ¹³ C-labeled DNA, we retrieved nine high to intermediate quality metagenome-assembled genomes (MAGs) belonging to the Proteobacteria , Gemmatimonadetes , and Chloroflexi , with three of these MAGs containing genes associated with methanotrophy. A novel Chloroflexi MAG contained a mmoX gene along with other methane oxidation pathway genes, identifying it as a potential uncultured methane oxidizer. This MAG also contained genes for copper import, synthesis of biopolymers, mercury detoxification, and ammonia uptake, indicating that this bacterium is strongly adapted to conditions in active layer permafrost and providing new insights into methane biogeochemical cycling. In addition, Betaproteobacterial MAGs were also identified as potential cross-feeders with methanotrophs in these Arctic cryosols. Overall, in situ SIP labeling combined with metagenomics and genome binning demonstrated to be a useful tool for discovering and characterizing novel organisms related to specific microbial functions or biogeochemical cycles of interest. Our findings reveal a unique and active Arctic cryosol microbial community potentially involved in CH 4 cycling.
... CH 4 is highly sensitive to temperature [44,[49][50][51]. Once global warming causes glacial collapse and permafrost melting, large amounts of stored CH 4 will be released into the atmosphere, causing irreversible effects on the global ecosystem [39,42,52]. CH 4 is usually formed by a microbiome-mediated process [40,44,[53][54][55] and is buried in subsea sediments and deep permafrost through long-term fermentation [44]. ...
Article
Full-text available
Climate change is having a profound impact on Arctic microbiomes and their living environments. However, we have only incomplete knowledge about the seasonal and inter-annual variations observed among these microbes and about their methane regulation mechanisms with respect to glaciers, glacial melting, snow lakes and coastal marine water. This gap in our knowledge limits our understanding of the linkages between climate and environmental change. In the Arctic, there are large reservoirs of methane which are sensitive to temperature changes. If global warming intensifies, larger quantities of methane stored in deep soil and sediments will be released into the atmosphere, causing irreversible effects on the global ecosystem. Methane production is mainly mediated by microorganisms. Although we have some knowledge of microbial community structure, we know less about the methane-correlated microbes in different land types in the Svalbard archipelago, and we do not have a comprehensive grasp of the relationship between them. That is the main reason we have written this paper, in which current knowledge of microorganisms and methane-correlated types in High Arctic Svalbard is described. The problems that need to be addressed in the future are also identified.
... These ancient permafrost sediments along coastline of the East Siberian Arctic Shelf are susceptible to rapid erosion and degradation due to global warming [77]. Previous studies have focused on the response of microorganisms to thawing of modern near-surface permafrost [67,68,78,79]. Due to thermal collapse and erosion of these carbon-rich Plio/Pleistocene coastline permafrost sediments, it is important to understand the metabolic status of the buried microorganisms in the deeper, older permafrost along the Arctic coastline across the Beringian region. ...
Article
Full-text available
Background Total DNA (intracellular, iDNA and extracellular, eDNA) from ancient permafrost records the mixed genetic repository of the past and present microbial populations through geological time. Given the exceptional preservation of eDNA under perennial frozen conditions, typical metagenomic sequencing of total DNA precludes the discrimination between fossil and living microorganisms in ancient cryogenic environments. DNA repair protocols were combined with high throughput sequencing (HTS) of separate iDNA and eDNA fraction to reconstruct metagenome-assembled genomes (MAGs) from ancient microbial DNA entrapped in Siberian coastal permafrost. Results Despite the severe DNA damage in ancient permafrost, the coupling of DNA repair and HTS resulted in a total of 52 MAGs from sediments across a chronosequence (26–120 kyr). These MAGs were compared with those derived from the same samples but without utilizing DNA repair protocols. The MAGs from the youngest stratum showed minimal DNA damage and thus likely originated from viable, active microbial species. Many MAGs from the older and deeper sediment appear related to past aerobic microbial populations that had died upon freezing. MAGs from anaerobic lineages, including Asgard archaea, however exhibited minimal DNA damage and likely represent extant living microorganisms that have become adapted to the cryogenic and anoxic environments. The integration of aspartic acid racemization modeling and metaproteomics further constrained the metabolic status of the living microbial populations. Collectively, combining DNA repair protocols with HTS unveiled the adaptive strategies of microbes to long-term survivability in ancient permafrost. Conclusions Our results indicated that coupling of DNA repair protocols with simultaneous sequencing of iDNA and eDNA fractions enabled the assembly of MAGs from past and living microorganisms in ancient permafrost. The genomic reconstruction from the past and extant microbial populations expanded our understanding about the microbial successions and biogeochemical alterations from the past paleoenvironment to the present-day frozen state. Furthermore, we provided genomic insights into long-term survival mechanisms of microorganisms under cryogenic conditions through geological time. The combined strategies in this study can be extrapolated to examine other ancient non-permafrost environments and constrain the search for past and extant extraterrestrial life in permafrost and ice deposits on Mars.
Article
Reducing atmospheric loads of greenhouse gases (GHGs), especially CO2 and CH4, has been considered the key to alleviating global crises we are facing, such as climate change, sea level elevation, and ocean acidification. To this end, development of strategies and technologies for carbon capture, sequestration, and utilization (CCSU) is urgently needed. Although physicochemical methods have been the most actively studied in the early stages of developing CCSU technologies, there have recently been growing interests in developing microbe‐based CCSU processes. In this article, we discuss advantages of microbe‐based CCSU technologies over physicochemical approaches and even plant‐based approaches. Next, various parts of the global carbon cycle where microorganisms can contribute, such as sequestering atmospheric GHGs, facilitating the carbon cycle, and slowing down the depletion of carbon reservoirs are described, emphasizing the impacts of microbes on the carbon cycle. Strategies to upgrade microbes and increase their performance in assimilating GHGs or converting GHGs to value‐added chemicals are also provided. Moreover, several examples of exploiting microbes to address environmental crises are discussed. Finally, we discuss things to overcome in microbe‐based CCSU technologies and provide future perspectives.
Article
Full-text available
In Arctic regions, thawing permafrost soils are projected to release 50 to 250 Gt of carbon by 2100. This data is mostly derived from carbon-rich wetlands, although 71% of this carbon pool is stored in faster-thawing mineral soils, where ecosystems close to the outer boundaries of permafrost regions are especially vulnerable. Although extensive data exists from currently thawing sites and short-term thawing experiments , investigations of the long-term changes following final thaw and co-occurring drainage are scarce. Here we show ecosystem changes at two comparable tussock tundra sites with distinct permafrost thaw histories, representing 15 and 25 years of natural drainage, that resulted in a 10-fold decrease in CH 4 emissions (3.2 ± 2.2 vs. 0.3 ± 0.4 mg C-CH 4 m −2 day −1), while CO 2 emissions were comparable. These data extend the time perspective from earlier studies based on short-term experimental drainage. The overall microbial community structures did not differ significantly between sites, although the drier top soils at the most advanced site led to a loss of methanogens and their syntrophic partners in surface layers while the abundance of methanotrophs remained unchanged. The resulting deeper aeration zones likely increased CH 4 oxidation due to the longer residence time of CH 4 in the oxidation zone, while the observed loss of aerenchyma plants reduced CH 4 diffusion from deeper soil layers directly to the atmosphere. Our findings highlight the importance of including hydrological, vegetation and microbial specific responses when studying long-term effects of climate change on CH 4 emissions and underscores the need for data from different soil types and thaw histories.
Article
Full-text available
Among the numerous studies of methane emission from northern wetlands the number of measurements carried on at high latitudes (north of the Arctic Circle) is very limited, and within these there is a bias towards studies of the growing season. Here we present results of five years of automatic chamber measurements at a high-arctic location in Zackenberg, NE Greenland covering both the growing seasons and two months of the following freeze-in period. The measurements show clear seasonal dynamics in methane emission. The start of the growing season increase in CH<sub>4</sub> fluxes were strongly related to the date of snow melt. The greatest variation in fluxes between the study years were observed during the first part of the growing season. Somewhat surprisingly this variability could not be explained by commonly known factors controlling methane emission, i.e. temperature and water table position. Late in the growing season CH<sub>4</sub> emissions were found to be very similar between the study years (except the extremely dry 2010) despite large differences in climatic factors (temperature and water table). Late-season bursts of CH<sub>4</sub> coinciding with soil freezing in the autumn were observed at least during three years between 2006 and 2010. The accumulated emission during the freeze-in CH<sub>4</sub> bursts was comparable in size with the growing season emission for the year 2007, and about one third of the growing season emissions for the years 2009 and 2010. In all three cases the CH<sub>4</sub> burst was accompanied by a~corresponding episodic increase in CO<sub>2</sub> emission, which can compose a significant contribution to the annual CO<sub>2</sub> flux budget. The most probable mechanism of the late season CH<sub>4</sub> and CO<sub>2</sub> bursts is physical release of gases, accumulated in the soil during the growing season. In this study we investigate the drivers and links between growing season and late season fluxes. The reported surprising seasonal dynamics of CH<sub>4</sub> emissions at this site show that there are important occasions where conventional knowledge on factors controlling methane emissions is overruled by other processes, acting in longer than seasonal time scales. Our findings suggest the importance of multiyear studies with continued focus on shoulder seasons.
Article
Full-text available
High-latitude terrestrial ecosystems are key components in the global carbon (C) cycle. Estimates of global soil organic carbon (SOC), however, do not include updated estimates of SOC storage in permafrost-affected soils or representation of the unique pedogenic processes that affect these soils. The Northern Circumpolar Soil Carbon Database (NCSCD) was developed to quantify the SOC stocks in the circumpolar permafrost region (18.7 × 106 km2). The NCSCD is a polygon-based digital database compiled from harmonized regional soil classification maps in which data on soil order coverage have been linked to pedon data (n = 1778) from the northern permafrost regions to calculate SOC content and mass. In addition, new gridded datasets at different spatial resolutions have been generated to facilitate research applications using the NCSCD (standard raster formats for use in geographic information systems and Network Common Data Form files common for applications in numerical models). This paper describes the compilation of the NCSCD spatial framework, the soil sampling and soil analytical procedures used to derive SOC content in pedons from North America and Eurasia and the formatting of the digital files that are available online. The potential applications and limitations of the NCSCD in spatial analyses are also discussed. The database has the doi:10.5879/ecds/00000001. An open access data portal with all the described GIS-datasets is available online at: http://www.bbcc.su.se/data/ncscd/.
Article
Full-text available
Methane (CH4) emission by carbon-rich cryosols at the high latitudes in Northern Hemisphere has been studied extensively. In contrast, data on the CH4 emission potential of carbon-poor cryosols is limited, despite their spatial predominance. This work employs CH4 flux measurements in the field and under laboratory conditions to show that the mineral cryosols at Axel Heiberg Island in the Canadian high Arctic consistently consume atmospheric CH4. Omics analyses present the first molecular evidence of active atmospheric CH4-oxidizing bacteria (atmMOB) in permafrost-affected cryosols, with the prevalent atmMOB genotype in our acidic mineral cryosols being closely related to Upland Soil Cluster α. The atmospheric (atm) CH4 uptake at the study site increases with ground temperature between 0 °C and 18 °C. Consequently, the atm CH4 sink strength is predicted to increase by a factor of 5-30 as the Arctic warms by 5-15 °C over a century. We demonstrate that acidic mineral cryosols are a previously unrecognized potential of CH4 sink that requires further investigation to determine its potential impact on larger scales. This study also calls attention to the poleward distribution of atmMOB, as well as to the potential influence of microbial atm CH4 oxidation, in the context of regional CH4 flux models and global warming.
Article
Full-text available
Arctic tundra soils serve as potentially important but poorly understood sinks of atmospheric methane (CH4), a powerful greenhouse gas. Numerical simulations project a net increase in methane consumption in soils in high northern latitudes as a consequence of warming in the past few decades. Advances have been made in quantifying hotspots of methane emissions in Arctic wetlands, but the drivers, magnitude, timing and location of methane consumption rates in High Arctic ecosystems are unclear. Here, we present measurements of rates of methane consumption in different vegetation types within the Zackenberg Valley in northeast Greenland over a full growing season. Field measurements show methane uptake in all non-water-saturated landforms studied, with seasonal averages of-8.3 ± 3.7 μmol CH4 m-2 h-1 in dry tundra and-3.1 ± 1.6 μmol CH4 m-2 h-1 in moist tundra. The fluxes were sensitive to temperature, with methane uptake increasing with increasing temperatures. We extrapolate our measurements and published measurements from wetlands with the help of remote-sensing land-cover classification using nine Landsat scenes. We conclude that the ice-free area of northeast Greenland acts as a net sink of atmospheric methane, and suggest that this sink will probably be enhanced under future warmer climatic conditions.
Article
Full-text available
Soils and other unconsolidated deposits in the northern circumpolar permafrost region store large amounts of soil organic carbon (SOC). This SOC is potentially vulnerable to remobilization following soil warming and permafrost thaw, but stock estimates are poorly constrained and quantitative error estimates were lacking. This study presents revised estimates of the permafrost SOC pool, including quantitative uncertainty estimates, in the 0-3 m depth range in soils as well as for deeper sediments (>3 m) in deltaic deposits of major rivers and in the Yedoma region of Siberia and Alaska. The revised estimates are based on significantly larger databases compared to previous studies. Compared to previous studies, the number of individual sites/pedons has increased by a factor ×8-11 for soils in the 1-3 m depth range,, a factor ×8 for deltaic alluvium and a factor ×5 for Yedoma region deposits. Upscaled based on regional soil maps, estimated permafrost region SOC stocks are 217 ± 15 and 472 ± 34 Pg for the 0-0.3 m and 0-1 m soil depths, respectively (±95% confidence intervals). Depending on the regional subdivision used to upscale 1-3 m soils (following physiography or continents), estimated 0-3 m SOC storage is 1034 ± 183 Pg or 1104 ± 133 Pg. Of this, 34 ± 16 Pg C is stored in thin soils of the High Arctic. Based on generalised calculations, storage of SOC in deep deltaic alluvium (>3 m to ≤60 m depth) of major Arctic rivers is estimated to 91 ± 39 Pg (of which 69 ± 34 Pg is in permafrost). In the Yedoma region, estimated >3 m SOC stocks are 178 +140/-146 Pg, of which 74 +54/-57 Pg is stored in intact, frozen Yedoma (late Pleistocene ice- and organic-rich silty sediments) with the remainder in refrozen thermokarst deposits (±16/84th percentiles of bootstrapped estimates). A total estimated mean storage for the permafrost region of ca. 1300-1370 Pg with an uncertainty range of 930-1690 Pg encompasses the combined revised estimates. Of this, ≤819-836 Pg is perennially frozen. While some components of the revised SOC stocks are similar in magnitude to those previously reported for this region, there are also substantial differences in individual components. There is evidence of remaining regional data-gaps. Estimates remain particularly poorly constrained for soils in the High Arctic region and physiographic regions with thin sedimentary overburden (mountains, highlands and plateaus) as well as for >3 m depth deposits in deltas and the Yedoma region.
Article
Tarnocai, C. and Bockheim, J. G. 2011. Cryosolic soils of Canada: Genesis, distribution, and classification. Can. J. Soil. Sci. 91: 749-762. Cryosols are permafrost-affected soils whose genesis is dominated by cryogenic processes, resulting in unique macromorphologies, micromorphologies, thermal characteristics, and physical and chemical properties. In addition, these soils are carbon sinks, storing high amounts of organic carbon collected for thousands of years. In the Canadian soil classification, the Cryosolic Order includes mineral and organic soils that have both cryogenic properties and permafrost within 1 or 2 m of the soil surface. This soil order is divided into Turbic, Static and Organic great groups on the basis of the soil materials (mineral or organic), cryogenic properties and depth to permafrost. The great groups are subdivided into subgroups on the basis of soil development and the resulting diagnostic soil horizons. Cryosols are commonly associated with the presence of ground ice in the subsoil. This causes serious problems when areas containing these soils are used for agriculture and construc