G Model

ECOENG-4874; No. of Pages 12 ARTICLE IN PRESS

Ecological Engineering xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Ecological Engineering
journal homepage: www.elsevier.com/locate/ecoleng

Greenhouse gas emissions in natural and managed peatlands of

America: Case studies along a latitudinal gradient
Gert Veber a , Ain Kull a , Jorge A. Villa b , Martin Maddison a , Jaanus Paal c , Tõnu Oja a ,
Rodolfo Iturraspe d , Jaan Pärn a,e,f , Alar Teemusk a , Ülo Mander a,∗
a
Department of Geography, Institute of Ecology & Earth Sciences, University of Tartu, Estonia
b
Corporacion Universitaria Lasallista, Caldas, Colombia
c
Department of Botany, Institute of Ecology & Earth Sciences, University of Tartu, Estonia
d
National University of Tierra del Fuego, Ushuaia, Argentina
e
School of Geography, Geology and the Environment, Keele University, United Kingdom
f
School of Geography, Earth and Environmental Sciences, University of Birmingham, United Kingdom

a r t i c l e i n f o a b s t r a c t

Article history: Processes affecting CO2 and CH4 emissions and their budgets have been relatively well studied in north-
Received 11 May 2017 ern temperate peatlands, whereas similar studies are almost absent in southern Patagonia and the
Received in revised form 30 June 2017 high-altitude Andean peatlands, both of which are currently under heavy anthropogenic pressure. The
Accepted 30 June 2017
objectives of this study were to compare greenhouse gas (GHG) emissions in natural and managed peat-
Available online xxx
lands to examine the effect of management on GHG emissions and identify the environmental parameters
affecting them. We analysed CO2 , CH4 and N2 O emissions related to the physical and chemical condi-
Keywords:
tions of the peat: in a natural and managed transitional bog in Quebec, Canada, a natural páramo and
Bog
Carbon dioxide grazed peatland in the Colombian Andes, and a bog and a fen in Tierra del Fuego, Argentina. GHG fluxes
Fen were measured using the dark static chamber method. Groundwater table, temperature, O2 content,
Methane pH and redox potential were measured from observation wells, soil temperature was measured at four
Nitrous oxide depths, peat samples were analysed for pH, soil organic matter and dry matter content, P, K, Ca, Mg,
Páramo NH4 -N, NO3 -N, total N and C. In all regions, human-impacted peatlands showed significantly higher
Patagonia CO2 -C, N2 O-N and CH4 -C emissions than their natural counterparts. The Canadian managed transitional
bog showed the highest average CO2 -C (575 mg C m−2 h−1 ) and N2 O-N (0.08 mg N m−2 h−1 ) emissions,
whereas the Colombian pasture was the largest emitter of CH4 -C (2.35 mg C m−2 h−1 ). CO2 -C emissions
were controlled by soil temperature and C content, whereas CH4 -C flux was negatively correlated to
dissolved oxygen content in peat water, and positively to water table level and soil log (C/NO3 -N) ratio.
Total Inorganic Nitrogen (TIN), C/N ratio, and soil temperature were the main factors controlling N2 O
emissions. Intensive peatland management alters the soil C/N balance, and increases and leads to higher
variability of GHG emissions. Agricultural activities, especially crop production in peatlands, as well as
intensive grazing in mountain peatland pastures, are the main factors increasing GHG emissions in the
peatlands studied. Mitigation is possible via regulation of grazing intensity and replacing arable fields
with grasslands.
© 2017 Elsevier B.V. All rights reserved.

1. Introduction CO2 and N2 O emissions can increase significantly (Freeman et al.,

Undisturbed peatlands are important sinks for carbon dioxide
(CO2 ), sources of atmospheric methane (CH4 ) and nearly neutral to
nitrous oxide (N2 O) (Frolking et al., 2011). In drained organic soils,
depending on management intensity, CH4 fluxes are lowered, but
∗ Corresponding author.
E-mail address: ulo.mander@ut.ee (Ü. Mander).
1993; Kasimir-Klemedtsson et al., 1997; Maljanen et al., 2010). Pro-
cesses and parameters affecting CO2 and CH4 emissions and their
budgets have been relatively well studied in northern peatlands
(Frolking et al., 2011; Haddaway et al., 2014; Oertel et al., 2016).
The water table level has been found to be one of the key factors
governing methane emission from northern peatlands. Numerous
studies have reported a negative relationship between mean water
table depth and seasonal or annual CH4 emission from non-flooded
northern peatlands (Moore and Roulet, 1993; Bubier et al., 1993;

http://dx.doi.org/10.1016/j.ecoleng.2017.06.068
0925-8574/© 2017 Elsevier B.V. All rights reserved.

Please cite this article in press as: Veber, G., et al., Greenhouse gas emissions in natural and managed peatlands of America: Case studies
along a latitudinal gradient. Ecol. Eng. (2017), http://dx.doi.org/10.1016/j.ecoleng.2017.06.068
G Model
ECOENG-4874; No. of Pages 12 ARTICLE IN PRESS
2 G. Veber et al. / Ecological Engineering xxx (2017) xxx–xxx
Pelletier et al., 2007; Munir and Strack, 2014; Webster et al., 2013)
but temperature can also exert a considerable influence on CH4
emissions in northern peatlands (Dise et al., 1993; Bellisario et al.,
1999; Levy et al., 2012). Dunfield et al. (1993) showed that CH4
production and oxidation in temperate and subarctic peat soils
attained an optimum at around 25 ◦ C. The magnitude of CH4 flux
from northern peatlands is a function of the availability and quality
of organic substrate, water table depth, temperature, pH, as well as
vegetation type and productivity (Lai, 2009; Turetsky et al., 2014).
Drainage for forestry or conversion to agriculture or pasture
exposes the peat soil to oxygen and accelerates aerobic decompo-
sition of organic matter with increased carbon dioxide emissions
rather than methane, which is partly due to attenuated methano-
genesis in the oxic soil layer but also because of the significant
potential for microbial methane oxidation in the aerated soil layer
(Segers, 1998; Kip et al., 2012; Preuss et al., 2013). Carbon stocks,
which are resistant to decay under the anaerobic conditions preva-
lent in wetland soils, can be lost by aerobic respiration after
drainage (Minkkinen and Laine, 1998). Historically, the destruction
of North American wetlands through land-use change has had the
greatest effect on carbon fluxes. The primary effects have been a
reduction in their ability to sequester carbon, oxidation of their soil
carbon reserves upon drainage and a reduction in CH4 emissions
(Bridgham et al., 2006).
Nitrous oxide emission from soils is derived from nitrification
and denitrification processes. Nitrogen availability is a growth-
limiting factor in many peatland ecosystems (Bridgham et al.,
1996), especially in natural wet conditions (Graham and Vitt,
2016). However, in some Eastern Canadian and European peat-
lands that have experienced chronically high nitrogen deposition
(5–40 kg ha−1 yr−1 ), they are thought to be phosphorus limited
(Limpens et al., 2006). Thus in general peatlands are a very weak
source of nitrous oxide (N2 O; ∼0.00002 Pg N2 O-N y−1 ), while its
share increases with anthropogenic disturbance (Frolking et al.,
2011; Carter et al., 2012; Salm et al., 2012). Climatic factors that
regulate N2 O emission, include temperature, precipitation and
freezing and thawing regimes (Burton and Beauchamp, 1994),
resulting in significant emissions outside the growing season or
even exceeding them (Wagner-Riddle et al., 1997, 2017).
A better understanding of local and regional effects and the
interaction between factors controlling the delicate balance of GHG
emissions is still required (Groffman et al., 2000; Bridgham et al.,
2008; Fritz et al., 2011; Frolking et al., 2011; Butterbach-Bahl et al.,
2013; Olson et al., 2013; Henneberg et al., 2015). In contrast to
the extensively studied greenhouse gas balances in northern peat-
lands, e.g. in Canada (Klinger et al., 1994; Liblik et al., 1997; Roulet,
2000; Turetsky et al., 2002; Cleary et al., 2005; Bridgham et al.,
2006; Petrescu et al., 2010; Kroetsch et al., 2011; Euskirchen et al.,
2014; Haddaway et al., 2014), similar studies about greenhouse
gas emissions in southern Patagonia (sphagnum and cushion bogs,
fens) and high-altitude Andean wetlands (páramos) are almost
non-existent. High-altitude mountain peatlands are typically small
compared to low-altitude peatlands, although they are abundant
across the Andean landscape and are likely a key component in
regional carbon cycling (Comas et al., 2017). The same applies
partly to Patagonian peatlands, which are often developed along
the narrow bottoms of valleys, on shallow depressions of highland
grasslands and on low-gradient slopes (Grootjans et al., 2010).
Both páramos and southern Patagonian peatlands are cur-
rently under heavy anthropogenic pressure, with the main threats
caused by agricultural land conversion, drainage, mining activity
and climate change (Buytaert et al., 2006; Grootjans et al., 2010,
2014; Benavides, 2014; Maldonado Fonkén, 2014; Salvador et al.,
2014; Ochoa-Tocac et al., 2016). Nevertheless páramos, of which
Colombia has the largest extent, covering a total of 35,000 square
kilometres in South America, provide important ecosystem ser-
Please cite this article in press as: Veber, G., et al., Greenhouse gas emissions in natural and managed peatlands of America: Case studies
along a latitudinal gradient. Ecol. Eng. (2017), http://dx.doi.org/10.1016/j.ecoleng.2017.06.068
vices, including water provision and carbon sequestration. Water
retained and released by the soil in páramos is used both by local
communities and downstream for irrigation, drinking water, and
hydroelectric power. Large Andean cities such as Mérida, Bogotá,
Quito, and Cajamarca, as well as many others, depend substantially
on páramos for their water (Mena-Vásconez and Farley, 2016). On
the other hand, human activities, such as agriculture, the inten-
sification of livestock grazing, Pinus plantations and tourism, are
increasing in the páramos (Buytaert et al., 2006). Grazing alters the
ground cover and can lead to soil compaction and erosion, while
also altering nutrient cycles and runoff. The changing land use to
pasture increases the rates of soil respiration, which is a direct
pathway for soil carbon loss (Buytaert et al., 2011; Houghton et al.,
2012).
Our literature review (see Material and Methods for details)
reveals that there are only three regions in Peru and two in the
Ecuadorian páramos and five areas in Patagonia where GHGs have
been reported in scientific papers (Table 1). Furthermore, only two
studies have reported all three major GHGs (CO2 , CH4 , N2 O) in
páramos. No studies of all three GHGs or N2 O measurements in
Patagonian mires have been published in scientific journals.
The first aim of our study was to contribute, with exploratory
in situ simultaneous measurements of CO2 , CH4 and N2 O with
static chambers, to filling in the gap in the modestly studied GHG
emissions of páramo and Patagonian region peatland ecosystems.
Although páramo peatland plant communities are considered a
crucial component of the Andes biodiversity hotspot, and Patag-
onian cushion bogs are regarded as a unique type of mire, there is
only modest understanding of the dynamics of plant communities
affected by human disturbance (Young and León, 2007; Heusser,
1995). Thus our second and third objectives were to compare green-
house gas (GHG) emissions along a north to south gradient – from
continental to Andean high-altitude and southern Patagonian mar-
itime climates in both natural and managed peatlands, and to
examine the effect of management intensity on GHG emissions, and
identify the environmental parameters that drive GHG emissions
in peatlands.
We hypothesize that more intensive peatland management (1)
alters the soil C/N balance, (2) increases emissions of CO2 -C, N2 O-N
and (3) lowers the emissions of CH4 -C and also (4) leads to higher
variability in GHG emissions.
2. Material and methods
2.1. Study sites
The data were collected in a global soil and gas sampling cam-
paign between July 2012 and June 2015, following a uniform
protocol (Pärn et al., 2015). We sampled six temperate peatland
sites in three regions across North and South America (Fig. 1).
According to their latitudinal and altitudinal gradients, these peat-
lands represent most characteristic types of extratropical peatlands
of the continents.
In our context “intensive management” means there has been/is
soil cultivation and fertilization, “moderate management” means
intensive grazing but no soil cultivation, and “low-intensity man-
agement” means low-intensity grazing. The distinction between
intensive and low-intensity grazing is based on the number of live-
stock units.
The study regions and sampling times were chosen to cover
the expected near-annual mean emission in Argentina in spring
(Peria et al., 2015) and in the continental summer conditions in
Canada, where the expected spatial variability of the GHG emis-
sions between the natural and managed peatlands is highest. There
is no significant seasonal variability in the Colombian Andes.
 ECOENG-4874; No. of Pages 12
G Model
along a latitudinal gradient. Ecol. Eng. (2017), http://dx.doi.org/10.1016/j.ecoleng.2017.06.068
Please cite this article in press as: Veber, G., et al., Greenhouse gas emissions in natural and managed peatlands of America: Case studies

Table 1
Literature review on GHG emissions in páramos and southern Patagonia.
Region Location Altitude m a.s.l. Land use Dominant species Period CO2 CH4 N2 O Reference

g m−2 d−1 mg m−2 d−1 mg m−2 d−1

Southern Patagonia
Argentina, Tierra del Fuego Moat 40 Bog, natural Astelia pumila, Donatia December 2008, February <1 Fritz et al. (2011)
fascicularis 2009, late March 2009
Sphagnum spp. 1. . .14
Valle de Andorra 200 Sphagnum spp. 1. . .11

Chile, Punta Arenas SkyI <100 Bog, natural Sphagnum magellanicum, March and April, 2010 2.4 <3.2 Broder et al. (2015)
Empetrum rubrum, A.
pumila
SkyII S. magellanicum, E. rubrum 1.7 <3.2
PBr2 S. magellanicum, sedges 2.2 <3.2
and rushes

ARTICLE IN PRESS
Argentina, Santa Cruz Rio Gallegos <200 Magellanic humid grass Festuca gracillima, F. 4 seasons, once per season, 3.8 Peria et al. (2015)
steppe, moderate grazing magellanica, Poa spiciformis, 2011–2013

G. Veber et al. / Ecological Engineering xxx (2017) xxx–xxx

Magellanic humid grass Rytidosperma virescens, 2.7
steppe, intensive grazing Agropyron fuegianum, Carex
spp.
Argentina, Tierra del Fuego Valle de Pipo <100 Bog, natural S. magellanicum, E. rubrum January 2015 13.1 . . . 21.8 Lehmann et al.
(2016)

Páramo
Ecuador, Cajanuma Loja, Podocarpus National 2800–3000 Upper montane forest Cunoniaceae, Clusiaceae and May 2008–May 2009 −0.01. . .0.04 Wolf et al. (2011)
Park Melastomataceae 1.0. . .1.8 −0.44. . .−0.16 Wolf et al. (2012)

Peru, Andes South-eastern Peru 3500 Wet puna, bog January 2011 and June 10.0 ± 2.0 Jones et al. (2014)
2013, dry season
January 2011 and June 85.0 ± 15.0
2013, wet season
Wet puna, grassland January 2011 and June −0.05 ± 0.6
2013, dry season
January 2011 and June 0.4 ± 0.7
2013, wet season

Peru, Kosñipata Valley Manu National Park 1200–2200 Lower montane forest December 2010–December −0.64 ± 0.08 0.46 ± 0.12 Teh et al. (2014)
2200–3200 Upper montane forest 2011 −0.82 ± 0.08 0.12 ± 0.13
3200–3700 Montane grasslands 15.60 ± 2.14 0.04 ± 0.30
3200–3700 Montane grasslands, basin 63.99 ± 7.80 −0.27 ± 0.62
landforms

Peru, Kosñipata Valley Manu National Park 2200–3200 Lower montane forest February 2011 and June 4.1. . .4.3 −1.1. . .−1.0 Jones et al. (2016)
3200–3700 Upper montane forest 2013 2.9. . .4.0 −1.6. . .−1.1

Peru, Andes Wayqecha and Acjanaco 3085–3400 Montane grassland Calamagrostis longearistata, July 2011–July 2012 0.64. . .1.05 Oliver et al.
Scirpus rigidus, Festuca (2017)
dolichophy

Peru, Kosñipata Valley Manu National Park 3200–3700 Upper montane forest January 2011–June 2013 0.04 ± 0.07 Diem et al. (2017)
3200–3700 Montane grasslands 0.07 ± 0.08
3200–3700 Montane grasslands, basin −0.18 ± 0.16
landforms

Ecuador, Andes Mazar Wildlife Reserve 3200–3450 Paramo native forest Araliaceae, Asteraceae, 12–31 December 2013 5.59 ± 0.20 McKnight
Clusiaceae, Cyatheaceae, et al. (2017)
Myrsinaceae
Paramo grassland tussock grasses, Puya 4.75 ± 0.23
Paramo recently burned dava-herculis 3.75 ± 0.24
grassland

3
G Model
ECOENG-4874; No. of Pages 12 ARTICLE IN PRESS
4 G. Veber et al. / Ecological Engineering xxx (2017) xxx–xxx
Fig. 1. Location of study areas (black dots) in temperate continental Canada, Colombian high-altitude and subpolar oceanic climate in Argentina according to Köppen climate
classification. Background image by Google Earth 2017.
A natural transitional bog (N45◦ 07 40.7 , W74◦ 13 05.5 ) and a
former arable land on a deep histosol, currently a fallow hayfield
(N45◦ 07 02.7 , W74◦ 12 07.7 ) in Quebec, Canada were sampled
from 19 to 24 July 2012. The study sites were located in the south-
western part of the province, approximately 60 km southwest of
the city of Montréal. The annual average temperature is 6.4 ◦ C, and
annual average precipitation is 975 mm. The warmest month of
the year is July, with an average temperature of 21.1 ◦ C, whereas
January is the coldest month of the year at an average of −9.8 ◦ C. The
typical species in the natural site were Chamaedaphne calyculata,
Ledum groenlandicum, Vaccinium myrtilloides with an understory
of Sphagnum mosses and in managed site Lythrum salicaria, Cype-
rus dentatus, Eriophorum angustifolium, Mimulus luteus and Phalaris
australis.
The natural transitional bog may feature indirect anthropogenic
effects (atmospheric deposition, general regionally lowered
groundwater level with increased nutrient content) and was con-
sidered as natural in our study in relation to the management
intensity gradient. These sites are not necessarily representative
of most Canadian peatlands but rather characterize the effect of
management on GHG emissions.
A natural Páramo peatland (N06◦ 40 11.8 , W75◦ 40 03.8 )
and an intensively grazed peatland pasture (N06◦ 38 35.1 ,
W75◦ 39 55.6 ) were sampled in the Department of Antioquia,
Colombia from 7 to 10 June 2015. The sites were located in the
northern portion of the Andes, in the north-western corner of
the Colombian Cordillera Central at altitudes of 3200 and 3000 m
respectively. The average annual temperature is approximately
10 ◦ C, and the average annual precipitation is 1990 mm, with pre-
cipitation peaks during the months of April and October (Urbina
and Benavides, 2015). The lowest and most diverse vegetation zone
of the alpine tundra is classified as subpáramo, a shrub-dominated
ecosystem that combines aspects of the grass páramo above and
the forest below (Buytaert et al., 2006). Typical species in the
natural site were Sphagnum sp, Puya trianae, Calamagrostis effusa,
Please cite this article in press as: Veber, G., et al., Greenhouse gas emissions in natural and managed peatlands of America: Case studies
along a latitudinal gradient. Ecol. Eng. (2017), http://dx.doi.org/10.1016/j.ecoleng.2017.06.068
Orthrosanthus sp, Baccharis tricuneata and in the managed site
Sphagnum sp, Drepanocladus, Ranunculus praemorsus, Carex spp.,
Eleocharis spp.
A natural bog in Haruwen (N54◦ 44 07.6 , W67◦ 54 40.5 )
and a managed fen Vega Esperanza (coordinates N54◦ 21 26.8 ,
W67◦ 45 22.7 ) in Tierra del Fuego, Argentina were sampled from
19 to 24 November 2014. Located in a subpolar oceanic climate,
the annual average temperature is 5.3 ◦ C and average precipita-
tion 578 mm. In January, the warmest month of the year, average
temperature is 9.0 ◦ C and in July, the coldest month, the average
temperature is 1.4 ◦ C.
The fen was managed as a low-intensity pasture. The typi-
cal species were Deschampsia antarctica, Alopecurus magellanicus,
Hordeum halophilum, H. lechleri, H. comosum, Koeleria fuegiana,
Carex macrosolen, Juncus schechzeroide, Ranunculus uniflorus and R.
fuegianus. In the natural bog the dominant species was Sphagnum
magellanicum associated with S. cuspidatum, S. falcatulum, Carex
magellanica, C. curta and C. banksii.
2.2. Gas sampling and analyses
The closed static chamber method (Hutchinson and Livingston,
1993; Mander et al., 2003) was used for the measurement of ecosys-
tem respiration (CO2 ), CH4 and N2 O fluxes. In Canada and Colombia
three sampling positions per site were established, and five posi-
tions per site in Argentina. The sampling positions followed the
water regime gradient. Each sampling position was instrumented
with three static gas samplers placed ten metres apart from each
other and a peat water observation well. PVC collars of 50 cm diam-
eter were installed at 10 cm depth at least 3 h before sampling,
in order to minimize disturbance. The gas samplers were closed
chambers made of white PVC, height 50 cm, diameter 50 cm, vol-
ume 65 L, sealed with water-filled PVC collars. Gas was sampled
using pre-evacuated (0.3 mbar) 50 mL glass vials every 20 min dur-
ing an one-hour session (at 0, 20, 40 and 60 min from the enclosure
G Model
ECOENG-4874; No. of Pages 12 ARTICLE IN PRESS
G. Veber et al. / Ecological Engineering xxx (2017) xxx–xxx
of samplers). At least four sampling sessions per position were per-
formed within two days. Samples were collected twice per day (in
the morning and in the afternoon).
The gas concentration in the collected sample was deter-
mined using the Shimadzu GC-2015 gas-chromatography system
equipped with an electron capture detector (ECD) and a flame
ionization detector (FID) at the laboratory of the Department of
Geography, Institute of Ecology and Earth Sciences, University of
Tartu, Estonia. The emission rate of the gas in chamber was calcu-
lated on the basis of a linear change in the gas concentration, taking
into account the volume of the chamber and the area covered by the
chamber. Emission rate for one sampling position was calculated
as the average of the three chambers at the sampling position. A
total of 1236 air samples were collected, giving 309 emission rate
values from chambers (108 in Canada, 81 in Colombia and 120 in
Argentina). In the statistical analysis, samples with the determi-
nation coefficient R2 ≥ 0.90 for goodness of fit to linear regression
and samples with near-zero fluxes below the gas-chromatographer
accuracy (change of concentration <20 ppb) with R2 < 0.90 were
used. As a result, 253 flux measurements of CO2 -C, 188 samples of
CH4 -C and 286 samples of N2 O-N from the original 309 chamber-
based emission values were used.
2.3. Soil and water sampling and analyses
During each gas sampling session at each sampling position,
groundwater table depth (WT [cm]) in the observation wells
(Ø50 mm, polypropylene pipes [PP-HT] perforated to one meter
depth and covered with Typar SF 20 filter cloth) was determined.
Water temperature (Twater [◦ C]), dissolved oxygen content (O2
[mg/l]), dissolved oxygen saturation level (O2 [%]), pH and redox
potential (ORP [mV]) were measured in the observation wells using
a YSI Professional Plus handheld multi-parameter water quality
meter. Soil temperature was measured at four depths (10, 20, 30,
40 cm).
Soil samples of 150–200 g were collected from each chamber
location at two depths − from the permanently aerated 0–10 cm
layer and from the layer directly above the groundwater level (in
most cases 20–30 cm) at each sampling position (Canada n = 36,
Colombia n = 36, Argentina n = 60) after the final gas sampling.
The hermetically packed soil samples were kept in a thermal box
at a low temperature and transported to the Laboratory of Plant
Biochemistry of the Estonian University of Life Sciences in Tartu,
Estonia for chemical and physical analyses. All of the samples were
analysed for pH, dry matter content (DM), soil organic matter con-
tent (SOM), P, K, Ca, Mg, NH4 -N, NO3 -N, total N and total C. Soil pH
was determined from a potassium-chloride solution. SOM has been
determined as loss on ignition at 360 ◦ C for two hours of incubation.
Plant-available phosphorus (P) was determined on a FiaStar5000
flow-injection analyser (ammonium-lactate extractable). Plant-
available potassium (K) was determined from the same solution
using the flame-photometric method. Plant-available magnesium
(Mg) was determined from a 100 mL ammonium-acetate solu-
tion with a titanium-yellow reagent on a flow-injection analyser.
Calcium (Ca) was determined from a 100 mL ammonium-acetate
solution using the flame-photometric method. Soil ammonium
(NH4 -N) and nitrate (NO3 -N) were determined from a potassium-
chloride solution via flow-injection analysis. The total nitrogen (N)
and total carbon (C) content of oven-dry samples were determined
using the dry-combustion method on a varioMAX CNS elemental
analyser.
2.4. Statistical analysis of data
The statistical analysis was carried out using the STATISTICA 7.0
(StatSoft Inc.). The normality of the gas emission data was veri-
Please cite this article in press as: Veber, G., et al., Greenhouse gas emissions in natural and managed peatlands of America: Case studies
along a latitudinal gradient. Ecol. Eng. (2017), http://dx.doi.org/10.1016/j.ecoleng.2017.06.068
5
fied using the Kolmogorov–Smirnov, Lilliefors and Shapiro–Wilk’s
tests. Logarithmic transformation was applied to the C/NO3 ratio.
The distributions of GHGs differed from the normal, and hence non-
parametric tests were performed. The Kruskal–Wallis test was used
to detect statistically significant differences between the sites and
management intensities based on GHG fluxes. Considering the large
span between our study areas’ climate, topography, hydrogeologi-
cal conditions, plant and soil properties and management intensity,
we used the Principal Component Analysis (PCA) as a multivariate
exploratory technique to explain the main factors describing GHG
emissions along the north to south gradient from continental to
Andean high-altitude and southern Patagonian maritime climatic
conditions and to distinguish the environmental parameters that
best describe GHG emissions in different management intensity
conditions. Spearman’s rank order correlation and multiple regres-
sion analysis with forward stepwise inclusion were used to analyse
the relationship between the GHG fluxes and environmental con-
ditions. Outliers (mean plus or minus three standard deviations)
in the GHG emission data were excluded from the multiple regres-
sion analysis. Variables included in this analysis were selected from
the results of the correlation analysis according to the statisti-
cal significance. Variables which were expected to influence the
emission according to literature were also included to the analy-
sis. Forward stepwise inclusion method was used to add variables
to the model. In the first step, all statistically significant variables
were accepted in the model. In the next step, relevant variables
independently insignificant but improving the representativity of
the model (based on increase of adjusted R2 ) were included. Akaike
Information Criterion was used to prevent overparameterization of
the model. In all cases, p < 0.05 was the standard by which statistical
significance was accepted.
2.5. Literature review of GHG emissions in páramo and
Patagonian peatlands
A three-stage approach was used to extract the maximum num-
ber of scientific papers about measurements of CO2 , CH4 and N2 O
emissions in páramo and the Patagonian peatlands. The require-
ments were set as follows: only in situ measurements (either by the
chamber, eddy covariance or soda lime method), either CO2 , CH4
or N2 O measurement, and at least one original numeric emission
value for the considered ecosystem.
In the first stage, scientific papers were searched through liter-
ature databases (ISI Web of Science, Scopus, ScienceDirect) and the
Google Scholar search engine. Screening was performed on titles,
abstracts and full texts using geographical (Patagonia, South Amer-
ica, Andean, Tierra del Fuego) and ecosystem descriptors (peatland,
mire, bog, páramo, jalca, puna, bofedales, turberas) in combina-
tion with the following keywords: CO2 ; CH4 ; N2 O; carbon dioxide;
methane; nitrous oxide; GHG; emission; flux.
All full texts were critically appraised and data extracted accord-
ing to the requirements. In the second stage, the references of the
qualified papers were examined for relevant papers. In the final
stage (if available) papers citing the previously qualified papers
were searched for.
All reported GHG emission values were converted to g m−2 d−1
for CO2 -C, and mg m−2 d−1 for CH4 -C and N2 O-N.
3. Results
3.1. Soil and water properties
The water table was relatively uniform and high in both the
natural and managed sites in the humid climates of Patagonia
and páramo, but significantly lower in the Canadian natural bog
G Model
ECOENG-4874; No. of Pages 12 ARTICLE IN PRESS
6 G. Veber et al. / Ecological Engineering xxx (2017) xxx–xxx

Table 2
Soil and water parameters in study sites. Mean values with standard deviation (SD) are shown. For abbreviations see Material and methods.

Canada Colombia Argentina

Natural Managed Natural Managed Natural Managed

Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mean ± SD

WT (cm) −73.1 ± 5.09 −20.4 ± 10.50 −11.9 ± 5.68 −20.6 ± 10.79 −22.0 ± 7.83 −11.1 ± 5.48
O2 (mg/l) 4.02 ± 0.33 2.06 ± 0.09 1.78 ± 0.24 1.19 ± 0.11 3.95 ± 1.23 3.73 ± 0.84
pHwater 4.34 ± 0.48 4.79 ± 0.03 4.51 ± 0.27 4.92 ± 0.27 4.62 ± 0.33 5.92 ± 0.35
ORP (mV) 181.8 ± 20.24 166.0 ± 0.03 284.4 ± 4.05 352.1 ± 18.86 156.1 ± 76.55 185.3 ± 21.64
T10 (◦ C) 19.2 ± 1.25 21.8 ± 1.21 14.1 ± 0.69 17.0 ± 0.18 7.55 ± 0.90 7.68 ± 1.17
T40 (◦ C) 13.1 ± 0.59 19.6 ± 0.68 13.2 ± 1.02 16.5 ± 0.11 6.09 ± 1.06 6.72 ± 0.93
DM (%) 13.8 ± 0.88 26.3 ± 6.82 7.7 ± 2.90 23.0 ± 4.68 6.97 ± 1.70 13.1 ± 1.02
SOM (%) 81.8 ± 1.28 65.9 ± 8.50 70.7 ± 12.68 44.5 ± 10.18 80.5 ± 4.22 77.6 ± 2.96
N (%) 1.21 ± 0.04 1.59 ± 0.10 1.68 ± 0.16 1.73 ± 0.36 1.02 ± 0.26 2.96 ± 0.24
P (mg/kg) 31.9 ± 5.67 49.0 ± 15.13 59.7 ± 27.82 3.12 ± 1.54 96.9 ± 44.44 55.2 ± 17.87
C (%) 42.2 ± 2.23 35.6 ± 3.69 39.1 ± 6.61 26.3 ± 8.82 41.7 ± 2.14 44.8 ± 2.18
Ca (mg/kg) 997.8 ± 89.02 9326 ± 1721.4 198.7 ± 55.59 157.3 ± 39.14 1763 ± 1604.9 4683 ± 595.8
K (mg/kg) 90.2 ± 15.30 127.5 ± 45.76 289.5 ± 54.75 104.2 ± 19.29 249.4 ± 106.44 157.1 ± 44.31
Mg (mg/kg) 179.8 ± 17.26 2042 ± 497.6 66.1 ± 10.88 50.8 ± 13.05 911.3 ± 331.96 499.4 ± 103.44

(Table 2). This is explained by a prolonged hot and dry period pre-
ceding sampling in July 2012 which had a stronger effect on the
mainly precipitation-fed natural site than on the heavily degraded
and compacted managed peatland, where the groundwater level
was closer to the surface. That was also characterized by lower
dissolved oxygen content (2.06 vs 4.02 mg l−1 ) and a nearly uni-
form soil temperature profile in the upper 40 cm layer (from 21.8
to 19.6 ◦ C) in the managed site compared to a steep soil temperature
gradient (from 19.2 to 13.1 ◦ C) in the natural site. In the more humid
climate with smaller seasonal variation in Colombian páramo and
Patagonian mires peat water characteristics and soil temperature
profiles were relatively similar, with the only statistically signifi-
cant differences being in dissolved oxygen content and ORP in the
managed sites.
Mean total carbon content in our Colombian natural site was
39.1%. Mean soil organic matter content in the Colombian natu-
ral peatland (70.7%) is significantly higher than in the managed
peatland (44.5%). Lower SOM values for managed sites were also
recorded in Canada and Argentina, although they were not statisti-
cally significant in the latter. In Canada the peat in the managed site
had clearly higher nutrient (N, P, K) and Ca, Mg content, which were
residues of former fertilization. In the Argentinean sites Ca, Mg,
P and N content featured high internal heterogeneity and mainly
differed by local hydrological conditions of the peatlands.
3.2. Environmental factors controlling GHG fluxes
The PCA results showed that the first four components describe
more than 10% of the variation each (accordingly 29.4, 23.3, 16.3
and 11.5%). The first component (Fig. 2A) was mainly described
by variables affecting organic matter accumulation (soil P content,
SOM, C%, peat water temperature, soil temperature and dissolved
O2 content), correlating 39% of CO2 and 44% of CH4 emissions.
The second component described the environmental reaction and
conditions for organic matter decomposition (soil and peat water
pH, Ca, C/N, N%, dry matter content and water table depth) with
correlation coefficients of 48%, 30% and 30.6% with CO2 , N2 O and
CH4 respectively. N2 O had the strongest correlation (42%) with
the fourth component (main contributors: total inorganic nitro-
gen, NH4 , O2 and K). As in the natural peatlands under the cool and
humid climates (páramo, southern Patagonia), precipitation is not
a limiting factor, and water table depth contributed only modestly
to GHG emissions and had a relatively low factor loading in contrast
to soil and water temperature, nitrogen content, pH and dissolved
oxygen content.
As the first component was mainly described by variables
related to carbon accumulation, the Colombian managed site,
which was intensively grazed and thus erosion-prone due to tram-
pling (Fig. 2B), easily losing organic matter under intensive solar
radiation, was in sharp contrast to the least affected Argentinean
sites. However, the second component mainly distinguished nat-
ural and managed sites regardless of region, climatic or peatland
type, while the third component separated the cool humid sites
(Colombia, Argentina) from the Canadian sites under a continental
climate.
3.3. CO2 -C emissions
Median CO2 -C emissions varied between 26.8 in the Argen-
tinean natural and 572.4 mg C m−2 h−1 in the Canadian managed
sites (Fig. 3A1). While the median emissions in the natural sites
remained below 50 mg C m−2 h−1 , the emissions in the managed
sites increased more than ten times from Argentina to Canada along
the temperature and management gradient. The variability of the
fluxes increased with higher emissions and management intensity
(Fig. 3A1). In the natural sites the emissions varied from a mini-
mum of 6.4 to a maximum of 87.3 mg C m−2 h−1 , whereas in the
Canadian managed site the emissions varied between 397.1 and
874.3 mg C m−2 h−1 .
Emissions had the strongest positive correlations with water
temperature and soil temperatures at different depths (10, 20, 30
and 40 cm), in both the natural and managed sites (Table 3). The cor-
relations were stronger in the managed sites. Correlation between
CO2 -C and soil temperature at 10 cm depth (r = 0.92; p < 0.05) is pre-
sented in Fig. 3A2. Soil dry matter content played an important role
in explaining CO2 -C emissions (Table 3; Fig. 2A).
According to the statistical models created by multiple regres-
sion analysis with forward stepwise inclusion of variables, there are
no distinctive parameters that could efficiently predict CO2 -C emis-
sions in natural sites (Fig. 4AN). In the model (adjusted R2 = 0.43),
soil nitrogen content, soil water oxygen and dry matter content
had the strongest influence on emissions. In the managed sites the
regression model (adjusted R2 = 0.90) receives a major influence
from soil temperature at 20 cm depth and soil water pH. However,
in contrast to the natural sites where soil and water temperature
were insignificant, the temperature parameter overrides any other
model parameters in the case of managed peatlands, and Fig. 4AM
shows clearly how measured and predicted emission values are
grouped by study areas along the temperature gradient (lowest
emissions in Argentina and highest in Canada).

Please cite this article in press as: Veber, G., et al., Greenhouse gas emissions in natural and managed peatlands of America: Case studies
along a latitudinal gradient. Ecol. Eng. (2017), http://dx.doi.org/10.1016/j.ecoleng.2017.06.068
G Model
ECOENG-4874; No. of Pages 12 ARTICLE IN PRESS
G. Veber et al. / Ecological Engineering xxx (2017) xxx–xxx 7

Fig. 2. Ordination diagrams of PCA of environmental variables with GHG flux data in natural and managed areas, grouping by variables (A) and by sites (B).
Environmental variables: WT = water table, T-water = water temperature, O2 (mg/l) = dissolved oxygen content, pH-water = water pH, pH-soil = soil pH (KCl), ORP = redox
potential, T–10 cm = soil temperature at 10 cm depth, DM (%) = dry matter content, SOM = soil organic matter content, NO3 -N = soil nitrate, NH4 -N = soil ammonium, TIN = total
inorganic nitrogen, P = plant available phosphorus, K = plant available potassium, Mg = plant available magnesium, Ca = calcium, N (%) = total nitrogen, C (%) = total carbon,
C/N = C/N ratio, log (C/NO3 -N) = logarithm value of C/NO3 -N ratio. CO2 -C = CO2 -C emission (mg C m−2 h−1 ), CH4 -C = CH4 -C emission (␮g C m−2 h−1 ), N2 O-N = N2 O-N emission
(␮g C m−2 h−1 ).

Table 3
Correlations between CO2 -C, CH4 -C, N2 O-N emissions and environmental parameters in natural and managed sites across three countries. Marked correlations are significant
at p < 0.05.

Natural Managed

CO2 -C CH4 -C N2 O-N CO2 -C CH4 -C N2 O-N

(mg C m−2 h−1 ) (␮g C m−2 h−1 ) (␮g N m−2 h−1 ) (mg C m−2 h−1 ) (␮g C m−2 h−1 ) (␮g N m−2 h−1 )

WT (cm) −0.3544 0.4295 0.0204 −0.3689 0.0249 −0.1254

p = .012 p = .002 p = .888 p = .003 p = .848 p = .331
Twater (◦ C) 0.592 0.0882 −0.2031 0.8991 0.3996 0.3978
p = .000 p = .542 p = .157 p = .001 p = .001 p = .001
O2 (mg/l) 0.1304 −0.4881 0.3511 −0.5572 −0.5506 −0.129
p = .367 p = .000 p = .012 p = .000 p = .000 p = .318
pHwater −0.0343 0.3558 −0.0254 −0.7632 −0.3968 −0.2838
p = .813 p = .011 p = .861 p = .000 p = .001 p = .025
ORP (mV) 0.1765 0.3684 −0.0557 0.0725 0.5637 −0.2381
p = .220 p = .008 p = .701 p = .576 p = .000 p = .062
T20 (◦ C) 0.5468 0.0513 −0.1808 0.9148 0.386 0.424
p = .000 p = .723 p = .209 p = .001 p = .002 p = .001
T40 (◦ C) 0.5439 0.2497 −0.2211 0.8892 0.4251 0.3867
p = .000 p = .080 p = .123 p = .002 p = .001 p = .002
DM (%) 0.4183 −0.3524 0.1218 0.661 0.1891 0.2448
p = .003 p = .012 p = .400 p = .000 p = .141 p = .055
SOM (%) −0.0871 0.052 −0.055 −0.2907 −0.4565 0.0996
p = .547 p = .720 p = .704 p = .022 p = .000 p = .441
NO3 -N (mg/kg) 0.1077 0.0061 0.6585 0.3211 −0.2943 0.189
p = .457 p = .967 p = .000 p = .011 p = .020 p = .141
NH4 -N (mg/kg) 0.159 0.6492 −0.1045 0.0565 −0.1233 0.3121
p = .270 p = .000 p = .470 p = .663 p = .340 p = .014
TIN (mg/kg) 0.181 0.6295 0.0649 0.1797 −0.2297 0.3591
p = .208 p = .000 p = .654 p = .162 p = .073 p = .004
N (%) 0.3761 0.4733 0.0114 −0.7555 −0.3605 −0.2865
p = .007 p = .001 p = .938 p = .000 p = .004 p = .024
P (mg/kg) −0.0585 0.2877 0.1764 −0.1354 −0.3728 0.0533
p = .687 p = .043 p = .220 p = .294 p = .003 p = .681
C (%) 0.0172 0.2356 0.0735 −0.44 −0.3898 −0.0107
p = .905 p = .100 p = .612 p = .000 p = .002 p = .934
Ca (mg/kg) −0.4129 −0.3716 0.0716 0.3509 −0.4235 0.3976
p = .003 p = .008 p = .621 p = .005 p = .001 p = .001
K (mg/kg) −0.0446 0.2048 −0.1601 −0.096 −0.1628 0.1076
p = .758 p = .154 p = .267 p = .458 p = .206 p = .405
Mg (mg/kg) −0.6445 −0.1948 −0.0394 0.6708 −0.213 0.5554
p = .000 p = .175 p = .786 p = .000 p = .096 p = .000
C/N −0.3787 −0.2467 0.0473 0.659 −0.0143 0.5028
p = .007 p = .084 p = .744 p = .000 p = .912 p = .000
Log(C/NO3 -N) −0.0156 0.4731 −0.4916 −0.1593 0.5285 −0.2957
p = .915 p = .001 p = .000 p = .216 p = .000 p = .020

Please cite this article in press as: Veber, G., et al., Greenhouse gas emissions in natural and managed peatlands of America: Case studies
along a latitudinal gradient. Ecol. Eng. (2017), http://dx.doi.org/10.1016/j.ecoleng.2017.06.068
G Model
ECOENG-4874; No. of Pages 12 ARTICLE IN PRESS
8 G. Veber et al. / Ecological Engineering xxx (2017) xxx–xxx

Fig. 3. CO2 -C (mg C m−2 h−1 ) (A1), CH4 -C (␮g C m−2 h−1 ) (B1) and N2 O-N (␮g N m−2 h−1 ) (C1) emissions from natural (Nat.) and managed (Man.) sites and the mean soil
temperature at 10 cm depth averaged over the time of measurement sessions in Canada, Colombia and Argentina. The strongest relationships in managed sites between
CO2 -C efflux and temperature (◦ C) at 10 cm depth (A2); CH4 -C emission and dissolved oxygen content (mg/l) in water (B2); N2 O-N emission and C/N ratio (C2).

3.4. CH4 -C emissions
Median emissions of CH4 -C varied between −0.8 in the Cana-
dian natural and 2312.2 ␮g C m−2 h−1 in the Colombian managed
sites (Fig. 3B1). Although the Canadian managed and the Colom-
bian natural site had the highest variability in methane emissions,
from −9.8 to 2414.3 and from −1.5 to 2252.2 ␮g CH4 -C m−2 h−1
respectively, in the Colombian pasture the highest emissions, vary-
ing between 1600.6 and 3128.3 ␮g CH4 -C m−2 h−1 , were measured.
While in the natural sites the CH4 -C emissions positively cor-
related with water table depth, in managed sites no significant
correlation was observed. In the natural sites soil temperatures at
different depths did not affect emissions, whereas in the managed
sites soil temperature positively correlated with methane emis-
sion (Table 3). Both in natural and managed sites soil water oxygen
content correlated negatively (r = −0.55; p < 0.05 in managed sites)
(Fig. 3B2) and soil log (C/NO3 -N) ratio positively with CH4 -C emis-
sion. Over the natural and managed sites the CH4 -C emission was
explained partly by the same environmental factors, most impor-
tantly temperature, as in the case of CO2 -C emissions (Fig. 2A) and
in addition correlated with water redox potential.
CH4 -C emission in natural sites is most affected by soil tempera-
ture at 40 cm depth, water table and soil carbon content (Fig. 4BN).
In the managed sites the number of significant parameters in the
regression equation is larger, as the model (adjusted R2 = 0.78)
encompasses soil pH, soil water oxygen content, soil temperature
at 40 cm depth, soil NO3 -N and organic matter content (Fig. 4BM).
3.5. N2 O-N emissions
Median N2 O-N emissions varied between −1.6 in the Colombian
managed and 58.4 ␮g N2 O-N m−2 h−1 in the Canadian managed
sites (Fig. 3C1). In all other sites except the Canadian managed one,
median emissions remained close to zero and the emissions var-
ied from −8.4 in the Colombian managed to 9.8 ␮g N2 O-N m−2 h−1
in the Argentinean natural site. Even the minimum emission of
19.1 ␮g N2 O-N m−2 h−1 measured in the Canadian managed site
is double the maximum emission from other sites, and the max-
imum emission measured reached 475.4 ␮g N2 O-N m−2 h−1 . This

Please cite this article in press as: Veber, G., et al., Greenhouse gas emissions in natural and managed peatlands of America: Case studies
along a latitudinal gradient. Ecol. Eng. (2017), http://dx.doi.org/10.1016/j.ecoleng.2017.06.068
G Model
ECOENG-4874; No. of Pages 12 ARTICLE IN PRESS
G. Veber et al. / Ecological Engineering xxx (2017) xxx–xxx 9

Fig. 4. Predicted and observed results of stepwise forward multiple regression model shown on each figure for CO2 -C (A), CH4 -C (B) and N2 O-N (C) fluxes in natural (N) and
managed (M) areas. 95% confidence limits are shown by dashed lines.

was measured in a chamber in wet conditions, and the average of soil total nitrogen content, soil water pH, soil NO3 -N and Mg con-
six measurements conducted in this site over two consecutive days tent. The model predicted N2 O-N emissions with an accuracy of
was 279.4 ␮g N2 O-N m−2 h−1 . Mean emissions from three differ- 81% (Fig. 4CM). Nevertheless, model composition and prediction
ent sampling positions were as follows: wet (mean WT −13.7 cm) accuracy was clearly affected by the high emission values measured
149.0, interim (mean WT −34.0 cm) 62.9 and dry (WT > −100 cm) from the Canadian managed study site, which differed substantially
38.2 ␮g N2 O-N m−2 h−1 respectively. from other sites with lower emissions.
According to PCA conducted on both the natural and managed
sites, emissions of N2 O-N were most affected by environmen-
4. Discussion
tal conditions related to soil nitrogen content and pH (Fig. 2A).
There were few statistically significant correlations between N2 O-
The highest CO2 -C emissions, correlating primarily with soil
N emissions and environmental parameters in natural sites, except
temperature (explaining about 60% of the flux), have previously
for soil NO3 -N content and oxygen content. Correlations between
been measured in southwestern Quebec, in the warmest mid-
emissions and environmental parameters, water temperature, soil
summer period (Almaraz et al., 2009). In the same study the highest
temperatures at all measured depths, soil Ca and Mg content and
N2 O-N emissions were observed early in the season, more than 80%
C/N ratio (r = 0.50; p < 0.05) were significant under management
of which was explained by precipitation. Despite the dry summer
(Table 3 and Fig. 3C2).
period, our measurements yielded similarly high N2 O emissions in
Predicting N2 O-N emissions in natural sites, there were no
the Canadian managed site. There was substantially higher temper-
parameters closely related to the emissions, and the best obtained
ature and lower water table in the Canadian natural site compared
model (adjusted R2 = 0.19; p < 0.036) did not sufficiently describe
to the others. Even if methane was produced in deeper layers, a
emissions (Fig. 4CN). In the managed sites, the multiple regres-
thick aerated soil layer suitable for methane consumption (oxida-
sion analysis revealed statistically significant parameters such as
tion) was sufficient to prevent emitting CH4 -C into the atmosphere.

Please cite this article in press as: Veber, G., et al., Greenhouse gas emissions in natural and managed peatlands of America: Case studies
along a latitudinal gradient. Ecol. Eng. (2017), http://dx.doi.org/10.1016/j.ecoleng.2017.06.068
G Model
ECOENG-4874; No. of Pages 12 ARTICLE IN PRESS
10 G. Veber et al. / Ecological Engineering xxx (2017) xxx–xxx
Potential CH4 oxidation by methanotrophs is typically an order of
magnitude greater than potential CH4 production by methanogens
(Segers, 1998). As a result, methanotrophic bacteria can oxidize a
considerable portion of CH4 produced in the oxygen-depleted zone
and limit the amount of CH4 that can be released into the atmo-
sphere (Lai, 2009). For example, it has been demonstrated that CH4
emissions from an ombrotrophic bog in Quebec (9.4 ␮g m−2 h−1
CH4 -C) were significantly lower than from poor and rich fens
(247–844 ␮g m−2 h−1 CH4 -C) (Moore and Knowles, 1990).
The Colombian natural site, with moderate temperature and
high water table, has relatively high CH4 -C emissions. Meanwhile,
the Argentinean natural site, which has a high water table, also has
low CH4 -C emissions, probably limited by the low temperature.
Any management system that involves large N inputs into the soil,
like the intensively grazed pasture in Colombia, may have a signif-
icant effect on the production of CH4 in the soil (Gregorich et al.,
2005). Mean total carbon content in our Colombian natural site was
39.1%, whereas the C content of peat in an earlier study of six natural
Ecuadorian páramos (cushion plant peatlands, >3900 m a.s.l.) had
a mean value of 23.1%, ranging between 19 and 30% (Hribljan et al.,
2016). Similarly, we observed lower mean soil carbon values with
higher spatial variation in a grazed páramo site (26.3%). Mean soil
organic matter content in the Colombian natural peatland (70.7%)
was significantly higher than in the managed peatland (44.5%) but
still lower than that measured previously in 13 undrained Colom-
bian páramo peatlands (84%) by J.C. Benavides (2014).
In the natural sites N2 O-N emissions are small and have high
variability. After intensification or introduction of unsuitable man-
agement practices, the emissions increase substantially and links
between the emission and environmental factors are established.
The Canadian continental summer with high temperatures creates
suitable conditions for fast mineralization and N2 O-N emission.
Other studies on N2 O emissions from farmed peat soils give simi-
larly high values as in our Canadian drained peatland site. Regina
et al. (2004) report that mean annual N2 O fluxes in a farmed peat-
land in southern Finland during a 3-year study were 7.3, 15, 10 and
25 kg N2 O-N ha−1 for grass, barley, potato and fallow plots respec-
tively. Disturbances such as the ploughing of permanent grassland
on organic soils can significantly increase N2 O and CO2 effluxes,
whereas CH4 emission may also appear (Merbold et al., 2014).
N2 O-N emission is often localized in hotspots, the occurrence of
which may be related to the distribution of anaerobic microsites
and carbon availability (Gregorich et al., 2005). Higher rates of
denitrification have been linked to higher levels of mineralizable
organic carbon (Tenuta et al., 2000). In Colombia and Argentina
temperatures are more stable and lower, resulting in minor N2 O-N
emissions.
Unlike our study areas in peatland pastures with moderate to
low grazing intensity (the páramo in Colombia and the fen in Patag-
onia), intensive grazing can initiate significant emission of GHGs.
Teh et al. (2011) reported that an intensively grazed peatland in Cal-
ifornia was a net source of CH4 (25.8 mg CH4 -C m−2 d−1 ) and N2 O
(6.4 mg N2 O-N m−2 d−1 ). The N2 O fluxes were comparable to those
from heavily fertilized agroecosystems and tropical forests. Urbina
and Benavides investigated the effects of small-scale disturbances
such as increased fertility and physical disturbances of the soil
as expected from the expansion of agricultural practices into the
páramo on key functions of peatland ecosystems, and concluded
that extremely rapid decomposition rates in disturbed treatments
have the potential to produce an exponential effect on the emis-
sions of greenhouse gases when compared to undisturbed controls
(Urbina and Benavides, 2015).
Several studies pay attention to the mitigation of GHGs from
cultivated peatlands. Water table depth, soil and peat water tem-
perature, SOM/N or C/N ratio and NO3 − or C/NO3 − ratio are well
known factors describing GHG emissions in a particular peatland.
Please cite this article in press as: Veber, G., et al., Greenhouse gas emissions in natural and managed peatlands of America: Case studies
along a latitudinal gradient. Ecol. Eng. (2017), http://dx.doi.org/10.1016/j.ecoleng.2017.06.068
Regina et al. (2015) show that increasing water table depth from
70 to 30 cm in cultivated organic soils decreased both CO2 and N2 O
emissions, whereas net consumption of CH4 changed to net pro-
duction as the water table rose. The authors see that promoting
drainage systems for raising the ground water table and cultivat-
ing crops capable of producing good yields also in wet conditions
would be beneficial for GHG mitigation in cultivated organic soils.
The implementation of more sustainable management practices is
one of the best methods of reducing GHG emissions in agricultural
(managed) soils. The conversion of cultivated land to permanent
grassland would result in the greatest reduction in N2 O emissions,
particularly in eastern Canada, where the estimated reduction is
about 60% less N2 O emissions for this conversion (Grant et al.,
2004). Poyda et al. (2016) show that the lowest yield-related GHG
emissions from cultivated fen soil were achieved by a three-cut
system of productive grassland swards in combination with high
groundwater level (long-term mean ≤20 cm below the surface).
5. Conclusions
Three of our hypotheses were supported by the results. Inten-
sive peatland management alters the soil C/N balance, increases
emissions of CO2 -C, N2 O-N, and leads to higher variability of GHG
emissions.
More specifically, intensive management of peatlands has a neg-
ative effect on carbon balance and N2 O-N emissions. In natural
areas the effect of any single environmental variable on GHG fluxes
is smaller and emissions are lower than in managed areas where
the ecosystem balance is affected and several key factors e.g. water
table level, increased soil temperature, oxygen content, nitrogen
availability contribute to higher emissions.
Agricultural activities, especially crop production in lowlands
and intensive grazing in mountain pastures where soil structure
is severely disturbed are the main causes of GHG emissions in the
studied peatlands.
One of the hypotheses was not supported by our observa-
tions: methane flux was not lower in the managed areas. That
might indicate increased ecosystem instability due to human
impact where added nitrogen (manure, fertilizers) and soil struc-
ture disturbance (animal trampling) support methanogenesis in
anaerobic microsites and reveal the complexity of processes related
to methane emissions in peatlands.
Acknowledgments
This research was supported by the Estonian Research Coun-
cil (the IUT2-16, IUT2-17 and PUTJD619 grants); and the EU
through the European Regional Development Fund (ENVIRON and
EcolChange Centres of Excellence, Estonia), and the 7th Framework
People programme (the PIRSES-GA-2009-269227 grant).
References
Almaraz, J.J., Mabood, F., Xiaomin, Zhou, Brace, M., Rochette, P., Bao-Luo, M., Smith,
D.L., 2009. Carbon dioxide and nitrous oxide fluxes in corn grown under two
tillage systems in southwestern Quebec. Soil Sci. Soc. Am. J. 73, 113–119,
http://dx.doi.org/10.2136/sssaj2006.0371.
Bellisario, L.M., Bubier, J.L., Moore, T.R., 1999. Controls on CH4 emissions from a
northern peatland. Glob. Biogeochem. Cycles 13 (1), 81–91, http://dx.doi.org/
10.1029/1998GB900021.
Benavides, J.C., 2014. The effect of drainage on organic matter accumulation and
plant communities of high-altitude peatlands in the Colombian tropical Andes.
Mires Peat 15 (1), 1–15.
Bridgham, S.D., Pastor, J., Janssens, J.A., Chapin, C., Malterer, T.J., 1996. Multiple
limiting gradients in peatlands: a call for a new paradigm. Wetlands 16, 45–65,
http://dx.doi.org/10.1007/BF03160645.
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, http://dx.doi.
org/10.1672/0277-5212.
G Model
ECOENG-4874; No. of Pages 12 ARTICLE IN PRESS
G. Veber et al. / Ecological Engineering xxx (2017) xxx–xxx
Bridgham, S.D., Pastor, J., Dewey, B., Weltzin, J.F., Updegraff, K., 2008. Rapid carbon
response of peatlands to climate change. Ecology 89, 3041–3048.
Broder, T., Blodau, C., Biester, H., Knorr, K.H., 2015. Sea spray, trace elements, and
decomposition patterns as possible constraints on the evolution of CH4 and
CO2 concentrations and isotopic signatures in oceanic ombrotrophic bogs.
Biogeochemistry 122, 327–342, http://dx.doi.org/10.1007/s10533-014-0044-5.
Bubier, J., Costello, A., Moore, T.R., Roulet, N.T., Savage, K., 1993. Microtopography
and methane flux in boreal peatlands, northern Ontario, Canada. Can. J. Bot. 71,
1056–1063, http://dx.doi.org/10.1139/b93-122.
Burton, D.L., Beauchamp, E.G., 1994. Profile nitrous oxide and carbon dioxide
concentrations in a soil subject to freezing. Soil Sci. Soc. Am. J. 58, 115–122,
http://dx.doi.org/10.2136/sssaj1994.03615995005800010016x.
Butterbach-Bahl, K., Baggs, E.M., Dannenmann, M., Kiese, R.,
Zechmeister-Boltenstern, S., 2013. Nitrous oxide emissions from soils: how
well do we understand the processes and their controls? Philos. Trans. R. Soc. B
368, 20130122.
Buytaert, W., Célleri, R., De Bièvre, B., Cisneros, F., Wyseure, G., Deckers, J.,
Hofstede, R., 2006. Human impact on the hydrology of the Andean páramos.
Earth-Sci. Rev. 79, 53–72, http://dx.doi.org/10.1016/j.earscirev.2006.06.002.
Buytaert, W., Cuesta-Camacho, F., Tobon, C., 2011. Potential impacts of climate
change on the environmental services of humid tropical alpine regions. Glob.
Ecol. Biogeogr. 20 (1), 19–33, http://dx.doi.org/10.1111/j.1466-8238.2010.
00585.x.
Carter, M.S., Larsen, K.S., Emmett, B., Estiarte, M., Field, C., Leith, I.D., Lund, M.,
Meijide, A., Mills, R.T.E., Niinemets, Ü., Penuelas, J., Portillo-Estrada, M.,
Schmidt, I.K., Selsted, M.B., Sheppard, L.J., Sowerby, A., Tietema, A., Beier, C.,
2012. Synthesizing greenhouse gas fluxes across nine European peatlands and
shrublands − responses to climatic and environmental changes.
Biogeosciences 9 (10), 3739–3755.
Cleary, J., Roulet, N.T., Moore, T.R., 2005. Greenhouse gas emissions from Canadian
peat extraction, 1990–2000: a life-cycle analysis. Ambio 34 (6), 456–461.
Comas, X., Terry, N., Hribljan, J., Chimner, R., Lilleskov, E., Suarez, E., Kolka, R., 2017.
Estimating belowground carbon stocks in peatlands of the Ecuadorian páramo
using ground-penetrating radar (GPR). J. Geophys. Res. Biogeosci. 122,
370–386, http://dx.doi.org/10.1002/2016JG003550.
Diem, T., Morley, N.J., Ccahuana, A.J., Huaraca Quispe, L.P., Baggs, E.M., Meir, P.,
Richards, M.I.A., Smith, P., Teh, Y.A., 2017. Complex controls on nitrous oxide
flux across a long elevation gradient in the tropical Peruvian Andes. Biogeosci.
Discuss., 1–44, http://dx.doi.org/10.5194/bg-2017-107.
Dise, N.B., Gorham, E., Verry, E.S., 1993. Environmental factors controlling methane
emissions from peatlands in northern Minnesota. J. Geophys. Res. 98,
10583–10594.
Dunfield, P., Knowles, R., Dumont, R., Moore, T.R., 1993. Methane production and
consumption in temperate and subarctic peat soils: response to temperature
and pH. Soil Biol. Biochem. 25, 321–326, http://dx.doi.org/10.1016/0038-
0717(93)90130-4.
Euskirchen, E.S., Edgar, C.W., Turetsky, M.R., Waldrop, M.P., Harden, J.W., 2014.
Differential response of carbon fluxes to climate in three peatland ecosystems
that vary in the presence and stability of permafrost. J. Geophys. Res. Biogeosci.
119 (8), 1576–1595, http://dx.doi.org/10.1002/2014JG002683.
Freeman, C., Lock, M.A., Reynolds, B., 1993. Fluxes of CO2 , CH4 and N2 O from a
Welsh peatland following simulation of water-table draw-down − potential
feedback to climatic change. Biogeochemistry 19 (1), 51–60.
Fritz, C., Pancotto, V.A., Elzenga, J.T.M., Visser, E.J.W., Grootjans, A.P., Pol, A.,
Iturraspe, R., Roelofs, J.G.M., Smolders, A.J.P., 2011. Zero methane emission
bogs: extreme rhizosphere oxygenation by cushion plants in Patagonia. New
Phytol. 190, 398–408, http://dx.doi.org/10.1111/j.1469-8137.2010.03604.x.
Frolking, S., Talbot, J., Jones, M.C., Treat, C., Kauffman, Boone, Tuittila, J., Roulet,
E.-S., 2011. Peatlands in the Earth’s 21 st century climate system. Environ. Rev.
19, 371–396, http://dx.doi.org/10.1139/a11-014.
Graham, J.A., Vitt, D.H., 2016. The limiting roles of nitrogen and moisture on
Sphagnum angustifolium growth over a depth to water table gradient. Plant
Soil 404, 427–439, http://dx.doi.org/10.1007/s11104-016-2906-1.
Grant, B., Smith, W.N., Desjardins, R., Lemke, R., Li, C., 2004. Estimated N2 O and CO2
emissions as influenced by agricultural practices in Canada. Clim. Change 65,
315–332, http://dx.doi.org/10.1023/B:CLIM.0000038226.60317.35.
Gregorich, E.G., Rochette, P., VandenBygaart, A.J., Angers, D.A., 2005. Greenhouse
gas contributions of agricultural soils and potential mitigation practices in
Eastern Canada. Soil Tillage Res. 83, 53–72, http://dx.doi.org/10.1016/j.still.
2005.02.009.
Groffman, P.M., Brumme, R., Butterbach-Bahl, K., Dobbie, K.E., Mosier, A.R., Ojima,
D., Papen, H., Parton, W.J., Smith, K.A., Wagner-Riddle, C., 2000. Evaluating
annual nitrous oxide fluxes at the ecosystem scale. Glob. Biogeochem. Cycles
14 (4), 1061–1070.
Grootjans, A., Iturraspe, R., Lanting, A., Fritz, C., Joosten, H., 2010. Ecohydrological
features of some contrasting mires in Tierra del Fuego, Argentina. Mires Peat 6
(1), 1–15.
Grootjans, A., Iturraspe, R., Fritz, C., Moen, A., Joosten, H., 2014. Mires and mire
types of Peninsula Mitre, Tierra del Fuego, Argentina. Mires Peat 14 (1), 1–20.
Haddaway, N.R., Burden, A., Evans, C.D., Healey, J.R., Jones, D.L., Dalrymple, S.E.,
Pullin, A.S., 2014. Evaluating effects of land management on greenhouse gas
fluxes and carbon balances in boreo-temperate lowland peatland systems.
Environ. Evid. 3 (5), 1–30, http://dx.doi.org/10.1186/2047-2382-3-5.
Henneberg, A., Elsgaard, L., Sorrell, B.K., Brix, H., Petersen, S.O., 2015. Does Juncus
effusus enhance methane emissions from grazed pastures on peat.
Biogeosciences 12, 5667–5676, http://dx.doi.org/10.5194/bg-12-5667-2015.
Please cite this article in press as: Veber, G., et al., Greenhouse gas emissions in natural and managed peatlands of America: Case studies
along a latitudinal gradient. Ecol. Eng. (2017), http://dx.doi.org/10.1016/j.ecoleng.2017.06.068
11
Heusser, C.J., 1995. Palaeoecology of a Donatia-Astelia cushion bog, Magellanic
Moorland-Subantarctic evergreen forest transition, southern Tierra del Fuego,
Argentina. Rev. Palaeobot. Palynol. 89 (3-4), 429–440, http://dx.doi.org/10.
1016/0034-6667(95)00004-2.
Houghton, R.A., House, J.I., Pongratz, J., van der Werf, G.R., DeFries, R.S., Hansen,
M.C., Le Quéré, C., Ramankutty, N., 2012. Carbon emissions from land use and
land-cover change. Biogeosciences 9, 5125–5142, http://dx.doi.org/10.5194/
bg-9-5125-2012.
Hribljan, J.A., Suárez, E., Heckman, K.A., Lilleskov, E.A., Chimner, R.A., 2016. Peatland
carbon stocks and accumulation rates in the Ecuadorian páramo. Wetl. Ecol.
Manage. 24, 113–127, http://dx.doi.org/10.1007/s11273-016-9482-2.
Hutchinson, G.L., Livingston, G.P., 1993. Use of chamber systems to measure trace
gas fluxes. Agric. Ecosyst. Eff. Trace Gases Glob. Clim. Change, 63–78, http://dx.
doi.org/10.2134/asaspecpub55.c4 (ASAE special public).
Jones, S., Diem, T., Huaraca Quispe, L. Priscila, Quispe Ccahuana, A.J., Meir, P., Teh, Y.
Arn, 2014. Methane fluxes from a wet puna ecosystem in the Peruvian Andes.
Geophysical Research Abstracts. Presented at the EGU General Assembly
Conference Abstracts, p. 681.
Jones, S.P., Diem, T., Huaraca Quispe, L.P., Cahuana, A.J., Reay, D.S., Meir, P., Teh,
Y.A., 2016. Drivers of atmospheric methane uptake by montane forest soils in
the southern Peruvian Andes. Biogeosciences 13, 4151–4165, http://dx.doi.org/
10.5194/bg-13-4151-2016.
Kasimir-Klemedtsson, A., Klemedtsson, L., Berglund, K., Martikainen, P., Silvola, J.,
Oenema, O., 1997. Greenhouse gas emissions from farmed organic soils: a
review. Soil Use Manage. 13 (4), 245–250 (S).
Kip, N., Fritz, C., Langelaan, E.S., Pan, Y., Bodrossy, L., Pancotto, V., Jetten, M.S.M.,
Smolders, A.J.P., Op den Camp, H.J.M., 2012. Methanotrophic activity and
diversity in different Sphagnum magellanicum dominated habitats in the
southernmost peat bogs of Patagonia. Biogeosciences 9, 47–55, http://dx.doi.
org/10.5194/bg-9-47-2012.
Klinger, L.F., Zimmerman, P.R., Greenberg, J.P., Heidt, Guenther, L.E., 1994. Carbon
trace gas fluxes along a successional gradient in the Hudson Bay lowland. J.
Geophys. Res. 99 (D1), 1469–1494, http://dx.doi.org/10.1029/93jd00312.
Kroetsch, D.J., Geng, X., Chang, S.X., Saurette, D.D., 2011. Organic soils of Canada:
part 1. Wetland organic soils. Can. J. Soil Sci. 91, 807–822.
Lai, D.Y.F., 2009. Methane dynamics in northern peatlands: a review. Pedosphere
19, 409–421, http://dx.doi.org/10.1016/s1002-0160(09)00003-4.
Lehmann, J.R.K., Münchberger, W., Knoth, C., Blodau, C., Nieberding, F., Prinz, T.,
Pancotto, V.A., Kleinebecker, T., 2016. High-resolution classification of South
Patagonian peat bog microforms reveals potential gaps in up-scaled CH4 fluxes
by use of Unmanned Aerial System (UAS) and CIR imagery. Remote Sens. 8,
173, http://dx.doi.org/10.3390/rs8030173.
Levy, P.E., Burden, A., Mark, D.A., Cooper, M.D.A., Kerry, J., Dinsmore, K.J., Drewer, J.,
Evans, C., Fowler, D., Gaiawyn, J., Gray, A., Jones, S.K., Jones, T., McNamara, N.P.,
Mills, R., Ostle, N., Sheppard, L.J., Skiba, U., Sowerby, A., Ward, S.E., Zielinı́ski, P.,
2012. Methane emissions from soils: synthesis and analysis of a large UK data
set. Glob. Change Biol. 18, 1657–1669, http://dx.doi.org/10.1111/j.1365-2486.
2011.02616.
Liblik, L.K., Moore, T.R., Bubier, J.L., Robinson, S.D., 1997. Methane emissions from
wetlands in the zone of discontinuous permafrost: fort Simpson, Northwest
Territories, Canada. Glob. Biogeochem. Cycles 11, 484–494.
Limpens, J., Heijmans, M.M.P.D., Berendse, F., 2006. The nitrogen cycle in boreal
peatlands. In: Wieder, R.K., Vitt, D.H. (Eds.), Boreal Peatland Ecosystems,
Ecological Studies. Springer, Berlin Heidelberg, pp. 195–230.
Maldonado Fonkén, M.S., 2014. An introduction to the bofedales of the Peruvian
High Andes. Mires Peat 15 (5), 1–13.
–
Maljanen, M., Sigurdsson, B.D., Gudmundsson, H., Óskarsson, H., Huttunen, J.T.,
Martikainen, P.J., 2010. Greenhouse gas balances of managed peatlands in the
Nordic countries - present knowledge and gaps. Biogeosciences 7, 2711–2738.
Mander, Ü., Kuusemets, V., Lõhmus, K., Mauring, T., Teiter, S., Augustin, J., 2003.
Nitrous oxide, dinitrogen and methane emission in a subsurface flow
constructed wetland. Water Sci. Technol. 48 (5), 135–142.
McKnight, J.Y., Harden, C.P., Schaeffer, S.M., 2017. Soil CO2 flux trends with
differences in soil moisture among four types of land use in an Ecuadorian
páramo landscape. Phys. Geogr. 38, 51–61, http://dx.doi.org/10.1080/
02723646.2016.1256101.
Mena-Vásconez, P., Farley, K.A., 2016. Andean Páramo Grasslands. Salem Press
(Encycl. Sci.).
Merbold, L., Eugster, W., Stieger, J., Zahniser, M., Nelson, D., Buchmann, N., 2014.
Greenhouse gas budget (CO2 , CH4 and N2 O) of intensively managed grassland
following restoration. Glob. Change Biol. 20, 1913–1928.
Minkkinen, K., Laine, J., 1998. Long-term effect of forest drainage on the peat
carbon stores of pine mires in Finland. Can. J. For. Res. 28, 1267–1275, http://
dx.doi.org/10.1139/x98-104.
Moore, T.R., Knowles, R., 1990. Methane emissions from fen, bog and swamp
peatlands in Quebec. Biogeochemistry 11, 45–61, http://dx.doi.org/10.1007/
bf00000851.
Moore, T.R., Roulet, N.T., 1993. Methane flux: water table relations in northern
wetlands. Geophys. Res. Lett. 20, 587–590, http://dx.doi.org/10.1029/
93gl00208.
Munir, T., Strack, M., 2014. Methane flux influenced by experimental water table
drawdown and soil warming in a dry boreal continental bog. Ecosystems 17,
1271–1285.
Ochoa-Tocac, B., Buytaert, W., De Bièvre, B., Célleri, R., Crespo, P., Villacís, M.,
Llerena, C.A., Acosta, L., Villazón, M., Guallpa, M., Gil-Ríos, J., Fuentes, P., Olaya,
D., Viñas, P., Rojas, G., Arias, S., 2016. Impacts of land use on the hydrological
G Model
ECOENG-4874; No. of Pages 12 ARTICLE IN PRESS
12 G. Veber et al. / Ecological Engineering xxx (2017) xxx–xxx
response of tropical Andean catchments. Hydrol. Process. 30, 4074–4089,
http://dx.doi.org/10.1002/hyp.109.
Oertel, C., Matschullat, J., Zurba, K., Zimmermann, F., Erasmi, S., 2016. Greenhouse
gas emissions from soils - a review. Chemie der Erde − Geochemistry 76 (3),
327–352, http://dx.doi.org/10.1016/j.chemer.2016.04.002.
Oliver, V., Oliveras, I., Kala, J., Lever, R., Teh, Y.A., 2017. No long-term effect of
land-use activities on soil carbon dynamics in tropical montane grasslands.
Biogeosci. Discuss., 1–25, http://dx.doi.org/10.5194/bg-2017-113.
Olson, D.M., Griffis, T.J., Noormets, A., Kolka, R., Chen, J., 2013. Interannual,
seasonal, and retrospective analysis of the methane and carbon dioxide
budgets of a temperate peatland. J. Geophys. Res.: Biogeosci. 118 (1), 226–238,
http://dx.doi.org/10.1002/jgrg.20031.
Pärn, J., Aasa, A., Egorov, S., Filippov, I., Gabiri, G., Gheorghe, I., Järveoja, J., Kasak, K.,
Laggoun-Défarge, F., Luswata, C., Maddison, M., Mitsch, W., Óskarsson, H.,
Pellerin, S., Salm, J.-O., Sohar, K., Soosaar, K., Teemusk, A., Tenywa, M., Villa, J.,
Vohla, C., Mander, Ü., 2015. Global boundary lines of N2 O and CH4 emission in
peatlands. In: Vymazal, J. (Ed.), The Role of Natural and Constructed Wetlands
in Nutrient Cycling and Retention on the Landscape. Springer International
Publishing, Třeboň, pp. 87–102.
Pelletier, L., Moore, T.R., Roulet, N.T., Garneau, M., Beaulieu-Audy, V., 2007.
Methane fluxes from three peatlands in the La Grande Rivière watershed,
James Bay lowland, Canada. J. Geophys. Res. Biogeosci. 112, G01018, http://dx.
doi.org/10.1029/2006jg000216.
Peria, P.L., Bahamonde, H., Christiansen, R., 2015. Soil respiration in Patagonian
semiarid grasslands under contrasting environmental and use conditions. J.
Arid Environ. 119, 1–8, http://dx.doi.org/10.1016/j.jaridenv.2015.03.008.
Petrescu, A.M.R., van Beek, L.P.H., van Huissteden, J., Prigent, C., Sachs, T., Corradi,
C.A.R., Parmentier, F.J.W., Dolman, A.J., 2010. Modeling regional to global CH4
emissions of boreal and arctic wetlands. Glob. Biogeochem. Cycles 24, GB4009.
Poyda, A., Reinsch, T., Kluss, C., Loges, R., Taube, F., 2016. Greenhouse gas emissions
from fen soils used for forage production in northern Germany. Biogeosciences
13, 5221–5244.
Preuss, I., Knoblauch, C., Gebert, J., Pfeiffer, E.-M., 2013. Improved quantification of
microbial CH4 oxidation efficiency in arctic wetland soils using carbon isotope
fractionation. Biogeosciences 10, 2539–2552, http://dx.doi.org/10.5194/bg-10-
2539-2013.
Regina, K., Syväsalo, E., Hannukkala, A., Esala, M., 2004. Fluxes of N2 O from farmed
peat soils in Finland. Eur. J. Soil Sci. 55, 591–599.
Regina, K., Sheehu, J., Myllys, M., 2015. Mitigating greenhouse gas fluxes from
cultivated organic soils with raised water table. Mitig. Adapt. Strateg. Glob.
Change 20, 1529–1544.
Roulet, N.T., 2000. Peatlands, carbon storage, greenhouse gases and the Kyoto
Protocol: prospects and significance for Canada. Wetlands 20, 605–615.
Salm, J.-O., Maddison, M., Tammik, S., Soosaar, K., Truu, J., Mander, Ü., 2012.
Emissions of CO2 , CH4 and N2 O from undisturbed, drained and mined
peatlands in Estonia. Hydrobiologia 692, 41–55, http://dx.doi.org/10.1007/
s10750-011-0934-7.
Please cite this article in press as: Veber, G., et al., Greenhouse gas emissions in natural and managed peatlands of America: Case studies
along a latitudinal gradient. Ecol. Eng. (2017), http://dx.doi.org/10.1016/j.ecoleng.2017.06.068
Salvador, F., Monerris, J., Rochefort, L., 2014. Peatlands of the Peruvian Puna
ecoregion: types, characteristics and disturbance. Mires and Peat 15 (3), 1–17.
Segers, R., 1998. Methane production and methane consumption: a review of
processes underlying wetland methane fluxes. Biogeochemistry 41, 23–51,
http://dx.doi.org/10.1023/A:1005929032764.
Teh, Y.A., Silver, W.L., Sonnentag, O., Detto, M., Kelly, M., Baldocchi, D.D., 2011.
Large greenhouse gas emissions from a temperate peatland pasture.
Ecosystems 14 (2), 311–325.
Teh, Y.A., Diem, T., Jones, S., Huaraca Quispe, L.P., Baggs, E., Morley, N., Richards, M.,
Smith, P., Meir, P., 2014. Methane and nitrous oxide fluxes across an elevation
gradient in the tropical Peruvian Andes. Biogeosciences 11, 2325–2339, http://
dx.doi.org/10.5194/bg-11-2325-2014.
Tenuta, M., Bergstrom, D.W., Beauchamp, E.G., 2000. Denitrifying enzyme activity
and carbon availability for denitrification following manure application.
Commun. Soil Sci. Plant Anal. 31, 861–876, http://dx.doi.org/10.1080/
00103620009370483.
Turetsky, M.R., Wieder, R.K., Vitt, D.H., 2002. Boreal peatland C fluxes under
varying permafrost regimes. Soil Biol. Biochem. 34, 907–912.
Turetsky, M.R., Kotowska, A., Bubier, J., Dise, N.B., Crill, P., Hornibrook, E.R.C.,
Minkkinen, K., Moore, T.R., Myers-Smith, I.H., Nykänen, H., Olefeldt, D., Rinne, J.,
Saarnio, S., Shurpali, N., Tuittila, E.-S., Waddington, J.M., White, J.R., Wickland,
K.P., Wilmking, M., 2014. A synthesis of methane emissions from 71 northern,
temperate, and subtropical wetlands. Glob. Change Biol. 20, 2183–2197.
Urbina, J.C., Benavides, J.C., 2015. Simulated small scale disturbances increase
decomposition rates and facilitates invasive species encroachment in a high
elevation tropical Andean peatland. Biotropica 47, 143–151, http://dx.doi.org/
10.1111/btp.12191.
Wagner-Riddle, C., Thurtell, G.W., Kidd, G.K., Beauchamp, E.G., Sweetman, R., 1997.
Estimates of nitrous oxide emissions from agricultural fields over 28 months.
Can. J. Soil Sci. 77, 135–144, http://dx.doi.org/10.4141/s96-103.
Wagner-Riddle, C., Congreves, K.A., Abalos, D., Berg, A.A., Brown, S.E., Ambadan,
J.T., Gao, X., Tenuta, M., 2017. Globally important nitrous oxide emissions from
croplands induced by freeze-thaw cycles. Nat. Geosci. 10, 279–283.
Webster, K.L., McLaughlin, J.W., Kim, Y., Packalen, M.S., Li, C.S., 2013. Modelling
carbon dynamics and response to environmental change along a boreal fen
nutrient gradient. Ecol. Model. 248, 148–164, http://dx.doi.org/10.1016/j.
ecolmodel.2012.10.004.
Wolf, K., Veldkamp, E., Homeier, J., Martinson, G.O., 2011. Nitrogen availability
links forest productivity, soil nitrous oxide and nitric oxide fluxes of a tropical
montane forest in southern Ecuador. Glob. Biogeochem. Cycles 25, 1–12,
http://dx.doi.org/10.1029/2010GB003876.
Wolf, K., Flessa, H., Veldkamp, E., 2012. Atmospheric methane uptake by tropical
montane forest soils and the contribution of organic layers. Biogeochemistry
111, 469–483, http://dx.doi.org/10.1007/s10533-011-9681-0.
Young, K.R., León, B., 2007. Tree-line changes along the Andes: implications of
spatial patterns and dynamics. Philos. Trans. R. Soc. B Biol. Sci. 362, 263–272,
http://dx.doi.org/10.1098/rstb.2006.1986.