Skip to main content

Full text of "No evidence for trace metal limitation on anaerobic carbon mineralization in three peatland soils"

See other formats


Geoderma 314 (2018) 95-101 



ELSEVIER 


Contents lists available at ScienceDirect 

Geoderma 

journal homepage: www.elsevier.com/locate/geoderma 



No evidence for trace metal limitation on anaerobic carbon mineralization 
in three peatland soils 



Check for 
updates 


Jason K. Keller*, Jillian Wade 

Chapman University, Schmid College of Science and Technology, Orange, CA 92866, USA 


ARTICLE INFO 


ABSTRACT 


Handling editor: Junhong Bai 
Keywords: 

Anaerobic carbon mineralization 
Methane 

Ombrotrophic-minerotrophic gradient 

Peatland 

Trace metals 


Peatlands store roughly one-third of the terrestrial soil carbon and release the potent greenhouse gas methane 
(CH 4 ) to the atmosphere, making these wetlands among the most important ecosystems in the global carbon 
cycle. Despite their importance, the controls of anaerobic decomposition of organic matter to carbon dioxide 
(C0 2 ) and CH 4 within peatlands are not well understood. It is known, however, that the enzymes responsible for 
CH 4 production require cobalt, iron and nickel, and there is a growing appreciation for the potential role of trace 
metal limitation in anaerobic decomposition. To explore the possibility of trace metal limitation in peatlands, we 
washed 3 peat soils with either PbCl 2 , to remove available trace metals, or distilled water. Following these 
washes, we added trace metals (as CoCl 2 , CuCl 2 , FeCl 2 , PbCl 2 and NiCl 2 ) to each soil. We measured anaerobic 
CH 4 and C0 2 production in laboratory incubations over 4 weeks before adding glucose as a labile carbon source 
and measuring CH 4 and C0 2 production for an additional 4 weeks. In all 3 soils, neither CH 4 nor C0 2 production 
were limited by individual trace metals, even following the wash with PbCl 2 to remove available metals. Further, 
in response to the addition of a labile carbon substrate, all soils supported increased rates of CH 4 and C0 2 
production without progressive trace metal limitation. Taken together, our findings suggest that individual trace 
metals may not be limiting to anaerobic decomposition in many peatland soils. 


1. Introduction 

Peatlands are a diverse group of wetlands that store nearly 500 Pg of 
carbon in their soils, an estimated one-third of the terrestrial soil carbon 
(Bridgham et al., 2006; Kolka et al., 2016). Further, peatlands con¬ 
tribute a significant fraction of the flux of methane (CH 4 ) attributed to 
global wetland ecosystems (Bridgham et al., 2013; Keller and 
Medvedeff, 2016). Given that CH 4 has 45-times the sustained-flux 
global warming potential of C0 2 (on a per mass basis over the 100 year 
time period; Neubauer and Megonigal, 2015), CH 4 cycling within 
peatlands can have important implications for the climate. Under¬ 
standing the role of peatlands in the global climate hinges on our me¬ 
chanistic understanding of CH 4 and C0 2 dynamics within peatland 
ecosystems. 

At the landscape scale, peatlands are generally classified along a 
hydrogeomorphic gradient, ranging from precipitation-fed (ombro- 
trophic) bogs to predominately groundwater-fed (minerotrophic) rich 
fens. While this gradient is defined by the degree of groundwater in¬ 
fluence, a number of other factors, including: pH, dominant vegetation, 
nutrient availability, cation exchange capacity and trace metal avail¬ 
ability also co-vary along this gradient (Bridgham et al., 1996; Kolka 


et al, 2016). In addition, microbial carbon cycling varies along this 
gradient with more minerotrophic sites generally exhibiting higher 
rates of overall carbon mineralization and soils from minerotrophic rich 
fens producing more CH 4 than soils from ombrotrophic bogs (e.g., 
Keller and Bridgham, 2007; Updegraff et al., 1995; Ye et al., 2012). 
Understanding the mechanistic reasons for these differences in CH 4 
production among peatland types is crucial for understanding the po¬ 
tential feedbacks between peatland carbon cycling and global climate 
change. 

There is a growing appreciation for the role that trace metals may 
play in regulating the production of CH 4 in natural ecosystems. In 
particular, it is known that the enzymes responsible for the production 
of CH 4 require large amounts of iron, nickel and cobalt (Glass and 
Orphan, 2012; Jarrell and Kalmokoff, 1988). These trace metals are 
often found in low concentrations in peatlands (Basiliko and Yavitt, 
2001; Gogo et al., 2010; Gorham and Janssens, 2005), which may limit 
the potential for CH 4 production in these ecosystems. This trace metal 
limitation may be particularly pronounced in ombrotrophic bogs (often 
characterized by low CH 4 production) because of small inputs of trace 
metals from precipitation. Further, Sphagnum mosses, which dominate 
ombrotrophic peatlands, have a high cation exchange capacity and may 


* Corresponding author. 

E-mail address: jkeller@chapman.edu (J.K. Keller). 
https://d 0 i. 0 rg/l 0.1016/j. geoderma.2017.11.001 

Received 8 March 2017; Received in revised form 5 August 2017; Accepted 2 November 2017 

Available online 21 November 2017 

0016-7061/ © 2017 Elsevier B.V. All rights reserved. 















J.K. Keller, J. Wade 


Geoderma 314 (2018) 95-101 


effectively bind trace metals (Gogo and Pearce, 2009a; Thomas and 
Pearce, 2004). In support of this enhanced trace metal limitation in 
ombrotrophic peatland soils, Basiliko and Yavitt (2001) reported that a 
mix of trace metals (iron, nickel, cobalt and sodium) increased rates of 
CH 4 production in an ombrotrophic bog peatland soil, but not in a 
minerotrophic fen soil. Similarly, the release of trace metals (e.g., iron, 
nickel and cobalt) from cation exchange sites (by saturation with lead 
(Pb 2 + ) or aluminum (Al 3 + )) stimulated CH 4 production in a Sphagnum- 
dominated bog soil, but not in a more minerotrophic fen soil (Gogo and 
Pearce, 2009b). 

These past projects have utilized various ‘cocktails’ of trace metals 
(in different combinations and concentrations) to explore for potential 
limitation of anaerobic carbon cycling. This approach is justified given 
that carbon mineralization could be co-limited by multiple trace metals; 
however, as suggested by Basiliko and Yavitt (2001), there is also a 
need to explore if individual trace metals can limit CH 4 production in 
peatland soils. In the current project, we tested for limitation of anae¬ 
robic carbon mineralization, as CH 4 and C0 2 production, by adding 
cobalt (Co), copper (Cu), iron (Fe), nickel (Ni), and lead (Pb) in¬ 
dividually to 3 peatland soils, ranging from an ombrotrophic bog to a 
minerotrophic rich fen. We tested for trace metal limitation under 
multiple experimental conditions in each soil. First, we attempted to 
induce trace metal limitation by saturating cation exchange sites with 
PbCl 2 and thoroughly washing soils to remove released trace metals. 
Second, we stimulated rates of anaerobic carbon mineralization by 
adding a labile carbon substrate (glucose) to test the possibility that 
progressive trace metal limitation would occur due to increased mi¬ 
crobial activity. We hypothesized that (i) the addition of trace metals 
would stimulate decomposition in ombrotrophic soils more than mi¬ 
nerotrophic soils and (ii) trace metal limitation would be exacerbated 
by both the removal of trace metals following the saturation of cation 
exchange sites as well as by the increased carbon mineralization fol¬ 
lowing the addition of a labile carbon substrate. 

2. Materials and methods 

2.1. Site description and sampling 

Soil samples for this project were collected from 3 peatlands located 
on the property of the University of Notre Dame Environmental 
Research Center in the Upper Peninsula of Michigan, USA. These sites 
represent a subset of peatlands selected as part of a larger project to 
represent the ombrotrophic-minerotrophic peatland gradient in this 
region based on differences in dominate vegetation, soil pH and anae¬ 
robic carbon cycling. These sites have been described previously (Ye 
et al., 2012), and a brief description is provided below. For consistency, 
we use the same site names utilized by Ye et al. (2012). 

“Bog 2” (N46°13 , 57 ,/ , W89°34 , 7") is dominated by > 90% cover by 
Sphagnum spp. mosses with scattered short-statured black spruce ( Picea 
mariana (Mill.) Britton, Sterns & Poggen) and ericaceous shrubs, in¬ 
cluding: leatherleaf ( Chamaedaphne calyculata (L.) Moench), small 
cranberry ( Vaccinium oxycoccos L.) and bog Labrador tea C Rhododendron 
groenlandicum Oeder). Average water-table depth (reported below 
hollow surfaces) during the growing season (~May-October) was 
- 16 cm and the pH was 4.1. Soil from Bog 2 had a von Post index of 
H3 and a rubbed fiber volume content of 38 ± 10% (values from Ye 
et al., 2012; mean ± 1 standard error) suggesting that the peat at this 
site was less decomposed than at the other sites. “Acidic Fen” 
(N46°12 , 48 ,/ , Wf89°30'2") has a Sphagnum spp. lawn with minimal cover 
from other species in the area of sampling. The average water-table 
depth was - 10 cm and the pH was 4.1. Soil from Acidic Fen had a von 
Post index of H3/H4 and a rubbed fiber volume content of 35 ± 9% 
(values from Ye et al., 2012). “Rich Fen” (N46°13 , 27", W89°29 / 53") is 
dominated by the upright sedge (Carex stricta Lam.) although leather- 
leaf shrubs are also present on tussocks. This site was consistently 
flooded with ~ 30 cm of standing water during the 2009 growing 


season. The average pH was 5.9 and soil from Rich Fen had a von Post 
index of H5 and a rubbed fiber volume content of 15 ± 5% (values 
from Ye et al., 2012), suggesting that this peat was the most decom¬ 
posed of the sites studied. 

Soil were collected from 30 cm below the water table measured in 
the field in each peatland in August of 2009 using 10-cm diameter PVC 
cores. Cores were 30 cm in length and were inserted into the soil with 
the aid of a serrated knife to minimize compaction. Cores were extruded 
into large Ziploc bags, frozen and shipped to Chapman University in 
Orange, CA. Prior to the initiation of this experiment, individual cores 
were allowed to thaw and large roots and living vegetation were re¬ 
moved by hand in the ambient atmosphere. The remaining root-free 
peat was refrozen. The length of time a core was thawed varied, but was 
generally < 1 week. 

2.2. Determination of water-extractable cations and cation exchange 
capacity 

Concentrations of water-extractable cations were measured by 
adding 20 g of field-moist, root-free peat to 20 mL of deionized water 
and shaking at 200 rpm for 1 h. Following the extraction, the slurry was 
centrifuged at 4100 rpm for 5 min and the supernatant was filtered 
through a P8 qualitative filter and frozen until analysis for cations at the 
Soil and Plant Tissue Testing Laboratory at the University of 
Massachusetts. 

Soil cation exchange capacity (CEC Ca ) was determined by compul¬ 
sive exchange with Ca 2+ (Gogo and Pearce, 2009a). Briefly, 0.15 g of 
air-dried peat was washed twice with 20 mL of 0.01 M HC1 for 5 min to 
remove background levels of Ca 2 + . Following each wash, the samples 
were centrifuged at 4100 rpm for 5 min and the supernatant was dis¬ 
carded. After both HCl washes, the remaining soil was washed twice 
with deionized water. Subsequently, the soil was saturated with 20 mL 
of 0.01 M CaCl 2 . This slurry was centrifuged for 5 min at 4100 rpm and 
the supernatant was discarded. After the CaCl 2 saturation, the re¬ 
maining soil was washed twice with deionized water. Finally, 20 mL of 
0.01 M HCl was added to the soil three times. After each addition, the 
soil was centrifuged at 4100 rpm for 5 min and the supernatant was 
collected. After all three washes, the combined supernatant was 
brought to 100 mL with 0.01 M HCl and this solution was analyzed for 
Ca 2 + at the Soil and Plant Tissue Testing Laboratory at the University 
of Massachusetts. 

2.3. Experimental design 

For logistical reasons, peat from each site was treated separately in 
this experiment (i.e., the treatments described below were applied to 
each peat at a separate time). This approach was appropriate as our 
intention was to focus on the importance of trace metals on carbon 
mineralization within a peatland while focusing on the more qualitative 
patterns (i.e., stimulation or inhibition by a given trace metal) between 
sites. 

2.3.1. Wash treatments 

To explore the role of trace metals bound to cation exchange sites, 
soils were initially washed with either deionized water or PbCl 2 . The 
water wash treatment was intended to remove dissolved cations. In 
contrast, the PbCl 2 wash treatment was intended to release soil-bound 
cations by saturating cation exchange sites with Pb 2+ (Gogo and 
Pearce, 2009b), and thus to induce trace metal limitation. Both wash 
treatments also removed dissolved organic matter, with important im¬ 
plications for the interpretation of our results (see Discussion section for 
additional details). For each replicate soil core, 100 g of field-moist peat 
was added to a Mason jar, amended with 100 mL of 2 mM PbCl 2 or 
100 mL of deionized water, and shaken at 200 rpm for 1 h. The peat 
was then transferred into 50-mL centrifuge tubes and centrifuged at 
3000 rpm for 5 min. The resulting supernatant was discarded and the 


96 


J.K. Keller, J. Wade 


Geoderma 314 (2018) 95-101 


peat was returned to the Mason jars and washed an additional 5 times 
with 100 mL of deionized water. Each wash cycle included shaking at 
200 rpm for 10 min, transferring to centrifuge tubes, centrifugation at 
3000 rpm for 5 min, discarding of the supernatant, and returning the 
peat to the Mason jar. The supernatant resulting from the fifth wash 
cycle tested negative for chloride (Chloride Test Kit, Model 8-P, Hach 
Company, Loveland, CO), suggesting that the chloride from the PbCl 2 
wash had been removed. Following the final wash, the peat was al¬ 
lowed to sit overnight at 4 °C. The following morning, the peat was 
centrifuged once more at 3000 rpm for 5 min and the supernatant was 
discarded. 

Ten gram subsamples of peat from each replicate core were added to 
twelve 160-mL serum bottles (6 contained PbCl 2 washed peat, 6 con¬ 
tained deionized water washed peat) along with 10 mL of deionized 
water which had been bubbled with N 2 for 15 min. The pH of each 
slurry was measured after 30 min. Subsequently, the bottles were 
capped with gray butyl septa and the headspaces were flushed with N 2 
for 15 min to establish anaerobic conditions. Additional subsamples of 
washed peat were dried at 60 °C for 48 h to determine percent moisture 
content. 

2.3.2. Phase I: equilibration 

All soils were incubated for 2 weeks in the dark at 22 °C. This 
equilibration phase was intended to allow microbial communities to re¬ 
establish after the wash treatments but before trace metal treatments 
were added to the soils. Headspace CH 4 and C0 2 concentrations were 
measured using a gas chromatograph (SRI 8610C, SRI Instruments, 
Torrance, CA) with a flame ionization detector and an in-line metha- 
nizer (to convert C0 2 to CH 4 ) on days 1, 3, 5, 7, 10 and 14. Dissolved 
CH 4 and C0 2 were calculated using Henry's Law adjusting for solubility, 
temperature and pH (Drever, 1997). Headspace and dissolved pools of 
both gases were summed to calculate total production. 

2.3.3. Phase II: trace metal amendment 

Following the equilibration phase, the bottles were opened in an 
anaerobic chamber (—95% N 2 and < 5% H 2 headspace; Coy 
Laboratory Products, Inc., Grasslake, MI) and the pH of each slurry was 
measured. Each bottle was amended with 10 mL of 0.2 mM trace metal 
solutions of CoCl 2 , CuCl 2 , FeCl 2 , NiCl 2 , or PbCl 2 . The use of only di¬ 
valent cations for this experiment normalized the amount of chloride 
added in each treatment. The amount of trace metals added by these 
treatments was 0.2 pmol g wet peat -1 (equivalent to 
—1-5 pmol gdw - 1 , depending on the moisture content of the different 
peat soils). These concentrations are higher than those added by 
Basiliko and Yavitt (2001) who added 0.02, 0.04 and 0.05 pmol g - 
wet peat - 1 of Co, Fe and Ni, respectively. They are more comparable to 
the concentrations added by Williams and Crawford (1984) who added 
0.5 and 3.7 pmol g wet peat - 1 of Co and Fe, respectively. All trace 
metal solutions were degassed with N 2 for five minutes prior to addi¬ 
tion. Ten milliliters of degassed, deionized water were added to the 
control treatment. The pH was recorded 30 min after treatment 
amendment. Bottles were capped, removed from the anaerobic 
chamber and flushed with N 2 for 15 min to ensure anaerobic condi¬ 
tions. The amended peat slurries were allowed to incubate in the dark 
at 22 °C for 4 weeks. Methane and C0 2 production were measured on 
days 1, 3, 5, 7, 10, 14, 21 and 28 as described above. 

2.3.4. Phase III: trace metal and labile carbon amendment 

At the end of Phase II, all bottles were opened in an anaerobic 
chamber and pH was recorded. Each bottle was then amended with a 
second 10 mL treatment of the appropriate 0.2 mM trace metals solu¬ 
tion which also contained 10 mM of glucose. The glucose treatment 
added 0.6 mmol of carbon to the bottles, which was approximately 20-, 
5- and 3-times the amount of carbon released as both CH 4 and C0 2 in 
Phase I and Phase II of the experiments in the Bog 2, Acidic Fen and 
Rich Fen soils initially washed with distilled water, respectively (data 


from Fig. 1). These treatments were designed to explore the possibility 
that trace metal limitation was only present when labile carbon was not 
limiting to anaerobic carbon mineralization. The pH of each bottle was 
recorded after 30 min, and the bottles were recapped and flushed with 
N 2 for 15 min before being returned to a dark, 22 °C incubator. Methane 
and C0 2 production were measured on days 1, 3, 5, 7, 10, 14, 21, and 
28 as described above. After the final phase of the incubation, the 
bottles were opened and the final pH was recorded. 

2.4. Data analysis 

We present cumulative CH 4 and C0 2 production (pmol C gdw - *) in 
the unamended control treatment at the end of each experimental phase 
to provide a context for rates of carbon mineralization in our experi¬ 
ment. The effects of the different initial wash treatments (i.e., the 
deionized water wash and the PbCl 2 wash) on cumulative CH 4 and C0 2 
production at the end of each experimental phase were explored using 
independent samples t- tests within each soil. Differences in CEC Ca and 
water-extractable cation availability between soils were analyzed by 
one-way ANOVA followed by Fisher's LSD. Linear regressions were used 
to explore relationships between pH and CH 4 and C0 2 production 
during an experimental phase. 

To examine the effect of trace metal additions on rates of CH 4 and 
C0 2 production, we calculated the CH 4 Response and the C0 2 Response 
for each soil during Phase II (trace metal amendment phase) and Phase 
III (trace metal and labile carbon amendment phase) as follows: 

Response = [ (Treatment Rate+1) - (Control Rate+1) 

/(Control Rate+1) ] *100 . 

where Treatment Rate was the cumulative production of CH 4 or C0 2 in 
a soil amended with a trace metal and Control Rate was the cumulative 
production of the same gas in the unamended control soil from the same 
replicate core. In this approach, positive response values indicate a 
stimulation of CH 4 or C0 2 production in response to trace metal 
amendment and negative values indicate an inhibition. To test for 
significant effects of trace metal additions in Phase II (trace metal 
amendments) and Phase III (trace metal and labile carbon amendment), 
we used a one-sample t- test to compare the observed response to 0. 
Given the large number of comparisons within each peat type (40 in¬ 
dividual tests), we utilized a Bonferroni correction to set a = 0.00125. 
This approach is admittedly conservative, but given the lack of stimu¬ 
lation observed (see below), we felt it was appropriate. 

3. Results 

Cation exchange capacity (CEC Ca ), measured as Ca 2 + exchange, was 
higher in the minerotrophic Rich Fen soil (0.284 ± 0.023 mmol gdw - T ) 
than in the more ombrotrophic Bog 2 (0.190 ± 0.004 mmol gdw - *) and 
Acidic Fen (0.154 ± 0.004 mmol gdw - *) soils (Table 1). Total water- 
extractable cations ranged between 4.92 ± 0.61 pmol gdw -1 in the Bog 
2 soil and 6.30 ± 0.38 pmol gdw -1 in the Acidic Fen soil, but did not 
differ between the three soils (Table 1). Differences in individual water- 
extractable cations between soils varied for different cations. Calcium and 
Mg were highest in the Rich Fen soil; A1 was highest in the Acidic Fen soil; 
K was lowest in the Rich Fen soil and Fe was lowest in the Bog 2 soil. 
Water-extractable P was similar in all soils (Table 1). 

Cumulative CH 4 production was generally highest in the Rich Fen 
soil and comparable in the Bog 2 and Acidic Fen soils across all ex¬ 
perimental phases (Fig. 1A, C and E). Cumulative C0 2 production was 
also highest in the Rich Fen soil during the equilibration phase and the 
trace metal amendment phase, although the magnitude of the differ¬ 
ences between soils was less pronounced for C0 2 than it was for CH 4 
(Fig. IB, D). In the presence of labile carbon, cumulative C0 2 produc¬ 
tion was highest in the Bog 2 soil and comparable between the Acidic 
Fen and Rich Fen soils (Fig. IF). 


97 


J.K. Keller, J. Wade 


Geoderma 314 (2018) 95-101 



Fig. 1. Mean ( ± 1 SE; n = 5) cumulative CH 4 and C0 2 production in Bog 2 (“B2”), Acidic Fen (“AF”) and Rich Fen (“RF”) soils which were initially washed with deionized (DI) water or 
PbCl 2 ; but were not amended with additional trace metals (i.e., the control treatment). Cumulative gas production over the 2-week equilibration phase (A. and B.); during the 4-week 
trace metal amendment phase (C. and D.); and over the 4-week trace metal and labile carbon amendment phase (E. and F.) are shown. Note that the axes scales are different for each 
panel. 


Table 1 

Mean ( ± 1 SE) CEC Ca and water-extractable cations in three peatland soils. Concentrations of Zn, B, Mn, Cu and Pb were also measured but were below 0.01 pmol gdw -1 . Total is the 
sum of all measured extracted cations. Different letters reflect differences between the soils (p < 0.05) based on one-way ANOVA followed by Fisher's LSD. 



CECca 

P 

K 



Ca 



Mg 



Fe 



A1 



Total 



mmol gdw - 1 

[imol gdw - 1 


















Bog 2 

0.190 ± 0.004 a 

0.76 ± 0.24 

2.16 

± 

0.46 b 

1.04 

+ 

0.08 a 

0.44 

± 

0.03 a 

0.10 

± 

0.01 A 

0.40 

+ 

0.10 A 

4.92 ± 

0.61 

Acidic Fen 

0.154 ± 0.004 a 

0.95 ± 0.20 

2.43 

+ 

0.40 b 

1.14 

+ 

0.05 a 

0.49 

+ 

0.02 a 

0.20 

+ 

0.04 b 

1.06 

± 

0.13 b 

6.30 ± 

0.38 

Rich Fen 

0.284 ± 0.023 b 

0.41 ± 0.14 

0.93 

+ 

0.21 a 

2.84 

+ 

0.1 6 b 

0.82 

+ 

0.07 b 

0.23 

+ 

0.02 b 

0.67 

+ 

0.12 a 

6.00 ± 

0.57 


Over the course of the experiment, rates of CH 4 production in¬ 
creased in the unamended soils (Fig. 1A, C), and showed an increase in 
response to the labile carbon amendment (Fig. IE). However, the sti¬ 
mulation of CH 4 production by the labile carbon addition was more 
dramatic in the Rich Fen soil than in either the Bog 2 or Acidic Fen soils 
(Fig. IE). In contrast, rates of C0 2 production decreased over the course 
of the experiment with comparable cumulative C0 2 production during 
the 2-week equilibration phase (Fig. IB) and the 4-week trace metal 
amendment phase (Fig. ID). The addition of labile carbon stimulated 
C0 2 production in all soils, but this effect was most pronounced in the 
Bog 2 soil (Fig. IF). 

In all soils, cumulative CH 4 and C0 2 production were lower in soils 
that were washed with PbCl 2 than in soils that were washed with 
deionized water during the experimental phase (Fig. 1). These 


differences were most pronounced for CH 4 production in the Bog 2 and 
Acidic Fen Soils, where cumulative CH 4 production in the PbCl 2 wash 
treatment was between 30% and 55% of the deionized water washed 
treatment during the trace metal amendment phase and the trace metal 
and labile carbon amendment phase (Fig. 1C and E). While the reduc¬ 
tion in CH 4 and C0 2 production in response to the PbCl 2 wash was 
consistent across all soils, the differences between the deionized water 
wash and the PbCl 2 wash were not significant in any experimental 
phase for any soil (p > 0.08 for all independent t-tests). 

There was little evidence of trace metal limitation in these peatland 
soils (Figs. 2-4). In the Bog 2 soil, the addition of trace metals generally 
inhibited CH 4 production (Fig. 2). This inhibition was more pronounced 
in the deionized water washed soils than the PbCl 2 washed soils when 
trace metals were added alone (Fig. 2A), but the difference between 


98 

























































J.K. Keller, J. Wade 


Geoderma 314 (2018) 95-101 



<D 

</> 

c 

o 

Q. 

CO 

CD 

dc 

CM 

o 

o 


0) 

CO 

c 

o 

a. 

(/) 

CD 

DC 

CM 

o 

o 


Fig. 2. Mean ( ± 1 SE; n = 5) response of cumulative CH 4 and C0 2 production in the Bog 2 soil, which was initially washed with deionized (DI) water or PbCl 2 . The responses during the 
4-week trace metal amendment phase (A. and B.) and over the 4-week trace metal and labile carbon amendment phase (C. and D.) reflect stimulation (positive values) or inhibition 
(negative values) by divalent cations, relative to unamended controls. Asterisks indicate a significant effect of the trace metal amendment based on a one-sample t -test (* = p < 0.005; 

** = p < 0.001). 


wash treatments disappeared when trace metals were added with a 
labile carbon source (Fig. 2C). Methane production was lowest fol¬ 
lowing the amendment of Cu in the Bog 2 soil, with a CH 4 response of 
- 38 and - 59% for the deionized water and PbCl 2 washed soils, re¬ 
spectively, in the trace metal amendment (Fig. 2 A) and - 69 and 


- 81% for the deionized water and PbCl 2 washed soil when trace me¬ 
tals were added with a labile carbon source (Fig. 2C). There was also no 
significant stimulation of C0 2 production in the Bog 2 soil in response 
to the addition of trace metals. Similar to CH 4 production, the most 
dramatic inhibition of C0 2 production was in response to the addition 


100 


— 50 

(D 

(/> 

C 

o 

Q. 

</> 

O' 


O 


o 


o - 


-50 


-100 

100 


w 50 

a> 

</) 

c 

o 

Q. 

</> 0 

& 


-50 - 


-100 


A. 


o DI Wash 


F—f—^ 


C. 


h 




—-s- 

5 1 


** 


5 i ** 


-ii- 


-p- 




100 


50 ~ 

<D 

(/> 

c 

o 

Q. 

0 8 

DC 


-50 


- -50 


Co Cu Fe 


Ni Pb 


Co 


Cu 


Fe 


Ni 


Pb 


O 

o 


-100 

100 


50 o 

w 
c 
o 

Q. 
(/) 
<D 

DC 


O 

O 


-100 


Fig. 3. Mean ( ± 1 SE; n = 5) response of cumulative CH 4 and C0 2 production in the Acidic Fen soil, which was initially washed with deionized (DI) water or PbCl 2 . The responses during 
the 4-week trace metal amendment phase (A. and B.) and over the 4-week trace metal and labile carbon amendment phase (C. and D.) reflect stimulation (positive values) or inhibition 
(negative values) by divalent cations, relative to unamended controls. Asterisks indicate a significant effect of the trace metal amendment based on a one-sample t -test (* = p < 0.005; 

** = p < 0.001). 


99 













J.K. Keller, J. Wade 


Geoderma 314 (2018) 95-101 



0 

if) 

c 

o 

Q. 

(/) 

CD 

O' 

CM 

o 

o 


O 

(/> 

c 

o 

Q. 

(/> 

0 

0£ 

CM 

o 

o 


Fig. 4. Mean ( ± 1 SE; n = 5) response of cumulative CH 4 and C0 2 production in the Rich Fen soil, which was initially washed with deionized (DI) water or PbCl 2 . The responses during 
the 4-week trace metal amendment phase (A. and B.) and over the 4-week trace metal and labile carbon amendment phase (C. and D.) reflect stimulation (positive values) or inhibition 
(negative values) by divalent cations, relative to unamended controls. Asterisks indicate a significant effect of the trace metal amendment based on a one-sample t -test (* = p < 0.005; 

** = p < 0.001). 


of Cu in the Bog 2 soil (Fig. 2B, D); although the inhibitory effect of Co 
was almost as dramatic for the PbCl 2 washed soil in the absence of 
labile carbon (Fig. 2B). 

In the Acidic Fen soil, there were few increases in CH 4 or C0 2 
production in response to trace metal amendments (Fig. 3). In general, 
CH 4 production showed few responses to trace metal amendments with 
the exception of an inhibition by Cu in both the absence (- 44% and 
- 28% for deionized water washed and PbCl 2 washed soils) and pre¬ 
sence of labile carbon (- 72% and - 69% for deionized water washed 
and PbCl 2 washed soils; Fig. 3A, C). Average C0 2 production was fre¬ 
quently stimulated by trace metal addition; however, this stimulation 
was relatively minor. The most pronounced increases were observed in 
response to the addition of Fe in the absence of labile carbon and in 
response to the addition of Ni in the presence of labile carbon (Fig. 3B 
and D, respectively). 

The Rich Fen soil exhibited the highest average stimulation of CH 4 
production in response to the addition of Co (deionized water wash 
only) and Fe (both washes) in the absence of labile carbon (Fig. 4A). 
However, these results were driven by a single replicate core with low 
rates of CH 4 production in the control treatment. Similar to the other 
soils, there was an inhibitory effect of Cu in the Rich Fen soil, although 
this effect was not seen in the deionized water washed soil in the ab¬ 
sence of labile carbon (Fig. 4 A and C). There were few effects of trace 
metal amendment on C0 2 production in the Rich Fen soil, although 
there was an inhibitory effect of Cu in the presence of labile carbon 
(Fig. 4B and D). 

4. Discussion 

Contrary to our initial hypotheses, neither CH 4 nor C0 2 production 
were limited by any of the individual trace metals investigated in these 
peatland soils. The lack of stimulation of anaerobic decomposition by 
trace metals was consistent across three peatland soils representing a 
range of trace metal availabilities and cation exchange capacities 
(Table 1) as well as a range of cumulative carbon mineralization 


(Fig. 1). Even following a wash with PbCl 2 to remove available metals, 
the addition of individual trace metals did not lead to an increase in 
decomposition. Similarly, these soils supported increased rates of CH 4 
and C0 2 production in response to the addition of a labile carbon 
substrate (glucose) without progressive trace metal limitation 
(Figs. 2-4). 

These results appear contrary to past studies which have suggested a 
potential for trace metal limitation in peatlands, particularly ombro- 
trophic soils (Basiliko and Yavitt, 2001; Gogo and Pearce, 2009a; Gogo 
and Pearce, 2009b). It is worth noting, however, that trace metal lim¬ 
itation is not universal in these past peatland studies. For example, a 
trace metal solution (added simultaneously with a nitrogen/phosphorus 
solution and containing magnesium, iron, manganese, zinc and cobalt) 
inhibited CH 4 production in soil from a Sphagnum -dominated transition 
fen (Williams and Crawford, 1984). Even the trace metal solution 
(containing nickel, cobalt and iron) that resulted in 2- to 3-fold in¬ 
creases in CH 4 production in two bog soils had no stimulatory effect, or 
even an inhibitory effect, in 3 other peatland soils (Basiliko and Yavitt, 
2001). Trace metals when added alone, as in this study, or in combi¬ 
nation do not always result in the stimulation of CH 4 production in 
peatlands and the mechanisms for the variable responses in different 
peats remain elusive. 

Gogo and Pearce (2009a) attributed the stimulation in CH 4 pro¬ 
duction in response to a wash with PbCl 2 to the release of trace metals 
from cation exchange sites; however, they also highlighted the fact that 
this apparent trace metal stimulation disappeared in the presence of 
labile carbon. Similar to our results, this suggests that carbon may be 
the fundamental limiting element for anaerobic decomposition. Given 
the apparent role of carbon limitation, it is important to note that the 
initial wash treatments used in this experiment also removed dissolved 
organic matter from the peat soils. There is mounting evidence that 
dissolved organic matter can be the primary source of carbon for me- 
thanogens in peatland soils. For example, the radiocarbon age of CH 4 is 
frequently more similar to dissolved carbon than solid-phase carbon in 
peatland ecosystems (e.g., Chanton et al., 2008; Tfaily et al., 2014; 


too 









J.K. Keller, J. Wade 


Geoderma 314 (2018) 95-101 


Wilson et al, 2016). Even after adding a labile carbon source in the 
form of glucose in Phase III of the current experiment, it appears that 
the active microbial community was able to mineralize that carbon 
using the available trace metals, even following the PbCl 2 wash treat¬ 
ment (Figs. 2-4). 

Another potential artifact of our experimental design was that the 
amendment of trace metals generally decreased the pH of the incuba¬ 
tions (Supplemental Table 1), possibly due to the release of H + from 
cation exchange sites in the presence of trace metals. Similar decreases 
in pH have been observed in past trace metal experiments (Gogo and 
Pearce, 2009a; Gogo and Pearce, 2009b). This acidification was more 
pronounced in the more ombrotrophic Bog 2 and Acidic Fen soils. 
Linear regressions suggest that in Bog 2 soil, these differences in pH 
explained between 34% and 40% of CH 4 and C0 2 production in the 
trace metal amendment and trace metal and labile carbon amendment 
phases. In the Acidic Fen soil, differences in pH explained 40% of CH 4 
production and 29% of C0 2 production during the trace metal 
amendment and labile carbon amendment phase. The small differences 
in pH in response to trace metal amendments did not explain difference 
in CH 4 and C0 2 production in the Rich Fen soil (Supplemental Table 2). 
The important role of pH in the more ombrotrophic soils is consistent 
with the recognized role of soil pH in regulating peatland carbon cy¬ 
cling and CH 4 production in particular (e.g., Dunfield et al., 1993; Ye 
et al., 2012). Future studies should consider the potential for these pH 
effects to mask trace metal limitation as the inhibitory effects of lower 
soil pH would counteract potential stimulation by trace metals. 

The most consistent effect of trace metal additions was the re¬ 
duction in both CH 4 and C0 2 production in all soils by Cu (Figs. 2-4). 
This reduction was dramatic - up to 81% reduction in CH 4 production - 
and suggests a direct inhibitory effect of Cu. The toxic effects of Cu on 
microbial activity are well documented (Beveridge et al., 1997), and a 
negative correlation between CH 4 emissions and dissolved Cu in rice 
soils suggest that the inhibitory effects of Cu may extend to the pro¬ 
duction of CH 4 in wetland environments (Jiao et al., 2005). While de¬ 
termination of the Cu concentrations necessary to inhibit microbial 
processes was beyond the scope of this study, it is important to note that 
the amount of Cu added to peats in the trace metal amendment and the 
trace metal and labile carbon amendment phases of this experiment far 
exceed the water soluble Cu in these soils (Table 1). However, Thomas 
and Pearce (2004) demonstrated that CuCl 2 treatment at ~ 10-times the 
concentration used in this study stimulated CH 4 production in bog from 
a peat soil, but only at intermediate depths. They suggested that this 
was due to the displacement of other trace metals coincident with 
copper binding to CEC sites in this soil. 

Taken together, our results suggest that individual trace metals do 
not limit anaerobic decomposition in many peatland soils. The lack of 
progressive limitation in response to the removal of metals by the PbCl 2 
wash and the addition of glucose suggest increased rates of decom¬ 
position are not likely to be limited by trace metals in these systems. 

Acknowledgements 

This work was supported by NSF grant DEB-0816743 to J.K.K. We 
acknowledge the support of Scott Bridgham, Rongzhong Ye and 
Brendan Bohannan for conceptualization of this experiment and fruitful 
conversations on our results. We appreciate comments from Cassandra 
Zalman and three anonymous reviewers on previous versions of this 
manuscript. The University of Notre Dame Environmental Research 
Center provided access to the peatlands sampled in this study. 

Appendix A. Supplementary data 

Supplementary data to this article can be found online at https:// 


doi.org/10.1016/j.geoderma.2017.11.001. 

References 

Basiliko, N., Yavitt, J.B., 2001. Influence of Ni, Co, Fe, and Na additions on methane 
production in Sphagnum -dominated Northern American peatlands. Biogeochemistry 
52, 133-153. 

Beveridge, T.J., Hughes, M.N., Lee, H., Leung, K.T., Poole, R.K., Savvaidis, I., Silver, S., 
Trevors, J.T., 1997. Metal-microbe interactions: contemporary approaches. Adv. 
Microb. Physiol. 38, 177-243. 

Bridgham, S.D., Cadillo-Quiroz, H., Keller, J.K., Zhuang, Q., 2013. Methane emissions 
from wetlands: biogeochemical, microbial, and modeling perspectives from local to 
global scales. Glob. Chang. Biol. 19, 1325-1346. 

Bridgham, S.D., Megonigal, J.P., Keller, J.K., Bliss, N.B., Trettin, C., 2006. The carbon 
balance of North American wetlands. Wetlands 26 (4), 889-916. 

Bridgham, S.D., Pastor, J., Janssens, J., Chapin, C., Malterer, T., 1996. Multiple limiting 
gradients in peatlands: a call for a new paradigm. Wetlands 16 (1), 45-65. 

Chanton, J.P., Glaser, P.H., Chasar, L.S., Burdige, D.J., Hines, M.E., Siegel, D.I., Tremblay, 
L.B., Cooper, W.T., 2008. Radiocarbon evidence for the importance of surface ve¬ 
getation on fermentation and methanogenesis in contrasting types of boreal peat¬ 
lands. Glob. Biogeochem. Cycles 22, GB4022. http://dx.doi.org/10.1029/ 
2008GB003274. 

Drever, J.I., 1997. The Geochemistry of Natural Waters, 3rd ed. Prentice Hall. 

Dunfield, P., Knowles, R., Dumont, R., Moore, T.R., 1993. Methane production and 
consumption in temperate and sub-arctic peat soils - response to temperature and pH. 
Soil Biol. Biochem. 25 (3), 321-326. 

Glass, J.B., Orphan, V.J., 2012. Trace metal requirements for microbial enzymes involve 
in the production and consumption of methane and nitrous oxide. Front. Microbiol. 3, 
1 - 20 . 

Gogo, S., Pearce, D.M.E., 2009a. Carbon, cations and CEC: interactions and effects on 
microbial activity in peat. Geoderma 153, 76-86. 

Gogo, S., Pearce, D.M.E., 2009b. Saturation of raised bog peat exchange sites by Pb 2 + and 
Al 3+ stimulates CH4 production. Soil Biol. Biochem. 41, 2025-2028. 

Gogo, S., Shreeve, T.G., Pearce, D.M.E., 2010. Geochemistry of three contrasting British 
peatlands: complex patterns of cation availability and implications for microbial 
metabolism. Geoderma 158, 207-215. 

Gorham, E., Janssens, J.A., 2005. The distribution and accumulation of chemical ele¬ 
ments in five peat cores from the mid-continent to the eastern coast of North America. 
Wetlands 25 (2), 259-278. 

Jarrell, K.F., Kalmokoff, M.L., 1988. Nutritional requirements of the methanogenic ar- 
chebacteria. Can. J. Microbiol. 34, 557-576. 

Jiao, Y., Huang, Y., Zong, L., Zheng, X., Sass, R.L., 2005. Effects of copper concentration 
on methane emission from rice soils. Chemosphere 58, 185-193. 

Keller, J.K., Bridgham, S.D., 2007. Pathways of anaerobic carbon cycling across an om- 
brotrophic-minerotrophic peatland gradient. Limnol. Oceanogr. 52 (1), 96-107. 

Keller, J.K., Medvedeff, C.A., 2016. Soil organic matter. In: Vepraskas, M.J., Craft, C.B., 
Richardson, J.L. (Eds.), Wetland Soils: Genesis, Hydrology, Landscapes, and 
Classification, Second Edition. CRC Press, Boca Raton, FL, pp. 165-188. 

Kolka, R.S., Bridgham, S.D., Ping, C.-L., 2016. Soils of peatlands: histosols and gelisols. In: 
Vepraskas, M.J., Craft, C.B., Richardson, J.L. (Eds.), Wetland Soils: Genesis, 
Hydrology, Landscapes, and Classification, Second Edition. CRC Press, Boca Raton, 
FL, pp. 277-309. 

Neubauer, S.C., Megonigal, J.P., 2015. Moving beyond global warming potential to 
quantify the climatic role of ecosystems. Ecosystems 18, 1000-1013. http://dx.doi. 
org/10.1007/sl0021-015-9879-4. 

Tfaily, M.M., Cooper, W.T., Kostka, J.E., Chanton, P.R., Schadt, C.W., Hanson, P.J., 
Iversen, C.M., Chanton, J.P., 2014. Organic matter transformation in the peat column 
at Marcell experimental forest: humification and vertical stratification. J. Geophys. 
Res. Biogeosci. 119. http://dx.doi.org/10.1002/2013JG002492. 

Thomas, P.A., Pearce, D.M.E., 2004. Role of cation exchange in preventing the decay of 
anoxic deep peat. Soil Biol. Biochem. 36, 23-32. 

Updegraff, K., Pastor, J., Bridgham, S.D., Johnston, C.A., 1995. Environmental and sub¬ 
strate controls over carbon and nitrogen mineralization in northern peatlands. Ecol. 
Appl. 5 (1), 151-163. 

Williams, R.T., Crawford, R.L., 1984. Methane production in Minnesota peatlands. Appl. 
Environ. Microbiol. 47 (6), 1266-1271. 

Wilson, R.M., Hopple, A.M., Tfaily, M.M., Sebestyen, S.D., Schadt, C.W., Pfeifer-Meister, 
L., Medvedeff, C., McFarlane, K.J., Kostka, J.E., Kolton, M., Kolka, R.K., Kluber, L.A., 
Keller, J.K., Guilderson, T.P., Griffiths, N.A., Chanton, J.P., Bridgham, S.D., Hanson, 
P.J., 2016. Stability of peatland carbon to rising temperatures. Nat. Commun. 7, 
13723. 

Ye, R., Jin, Q., Bohannan, B., Keller, J.K., McAllister, S.A., Bridgham, S.D., 2012. pH 
controls over anaerobic carbon mineralization, the efficiency of methane production, 
and methanogenic pathways in peatlands across an ombrotrophic-minerotrophic 
gradient. Soil Biol. Biochem. 54, 36-37. 


101