Content uploaded by Lauren Parry
Author content
All content in this area was uploaded by Lauren Parry
Content may be subject to copyright.
CLIMATE RESEARCH
Clim Res
Vol. 45: 13–29, 2010
doi: 10.3354/cr00903 Published online December 30
1. INTRODUCTION
Peatlands cover ~15% of the land area of the UK
and, while estimates of carbon stored in peatland have
varied in the past (Milne & Brown 1997, Bradley et al.
2005), the best current value is 2302 Mt C (R. Milne
pers. comm.). This is based on the estimate for Scot-
land from Chapman et al. (2009) combined with data
from Bradley et al. (2005) and Smith et al. (2007), with
the addition of pro rata estimates for England and
Northern Ireland, as peat stocks below 1 m are not well
characterised (Table 1).
Although UK peatlands form a relatively small part
of the peatlands of the northern boreal and temperate
regions, they are of national and international impor-
tance for a number of reasons. Many of them are bio-
© Inter-Research 2010 · www.int-res.com*Email: mbill@ceh.ac.uk
Carbon balance of UK peatlands: current state of
knowledge and future research challenges
M. F. Billett1,*, D. J. Charman2, J. M. Clark3, 8, C. D. Evans4, M. G. Evans5, N. J. Ostle6,
F. Worrall7, A. Burden4, K. J. Dinsmore1, T. Jones4, N. P. McNamara6, L. Parry2,
J. G. Rowson7, R. Rose6
1Centre for Ecology & Hydrology, Bush Estate, Penicuik, Midlothian EH26 0QB, UK
2School of Geography, University of Exeter, Exeter EX4 4QJ, UK
3Wolfson Carbon Capture Laboratory, School of Biological Sciences, Bangor University, Deiniol Road, Bangor,
Gwynedd LL57 2UW, UK
5Department of Geography, School of Environment & Development, University of Manchester, Oxford Road,
Manchester M13 9PL, UK
6Centre for Ecology & Hydrology, Library Avenue, Bailrigg, Lancaster LA1 4AP, UK
7Department of Earth Sciences, Science Laboratories, South Road, Durham DH1 3LE, UK
8Present address: Walker Institute for Climate Systems Research and Soils Research Centre, Department of Geography and
Environmental Science, School of Human and Environmental Sciences, University of Reading, Whiteknights, PO Box 233,
Reading RG6 6DW, UK
ABSTRACT: The retention of peatland carbon (C) and the ability to continue to draw down and store
C from the atmosphere is not only important for the UK terrestrial carbon inventory, but also for a
range of ecosystem services, the landscape value and the ecology and hydrology of ~15% of the land
area of the UK. Here we review the current state of knowledge on the C balance of UK peatlands
using several studies which highlight not only the importance of making good flux measurements,
but also the spatial and temporal variability of different flux terms that characterise a landscape
affected by a range of natural and anthropogenic processes and threats. Our data emphasise the
importance of measuring (or accurately estimating) all components of the peatland C budget. We
highlight the role of the aquatic pathway and suggest that fluxes are higher than previously thought.
We also compare the contemporary C balance of several UK peatlands with historical rates of C accu-
mulation measured using peat cores, thus providing a long-term context for present-day measure-
ments and their natural year-on-year variability. Contemporary measurements from 2 sites suggest
that current accumulation rates (–56 to –72 g C m–2 yr–1) are at the lower end of those seen over the
last 150 yr in peat cores (–35 to –209 g C m–2 yr–1). Finally, we highlight significant current gaps in
knowledge and identify where levels of uncertainty are high, as well as emphasise the research chal-
lenges that need to be addressed if we are to improve the measurement and prediction of change in
the peatland C balance over future decades.
KEY WORDS: Carbon · Peatland · Peat · DOC · Flux · Aquatic
Resale or republication not permitted without written consent of the publisher
Contribution to CR Special 24 ‘Climate change and the British Uplands’
O
PEN
PEN
A
CCESS
CCESS
Clim Res 45: 13–29, 2010
sphere or nature reserves, they are archives of change
and provide significant water resources to large and
often densely populated parts of the country. They
comprise a significant proportion of the total UK soil
carbon (C) pool, with the most extensive areas of peat
development in the cooler, wetter (mainly upland)
areas of the UK. The long-term viability of peatlands is
based upon their ability to continue to draw down and
store more C from the atmosphere than they lose via
a number of different export pathways. Whilst re-
searchers now recognise the range of flux pathways
(inputs and outputs) and the various forms of C in-
volved (gases, solutes and particulates), there is a con-
tinuous need to improve measurement and scaling
techniques and reduce levels of uncertainty in the
overall C balance.
Carbon accumulation in UK peatlands started at the
end of the last ice age and over the last 8000 to
10 000 yr rates has fluctuated due to changes in exter-
nal factors such as temperature and precipitation
(Charman 2002). The current consensus is that Gor-
ham’s (1991) value of –23 g C m–2 yr–1 (negative fluxes
represent net uptake from the atmosphere) is still a
reasonably accurate estimate of contemporary peat-
land C accumulation rates in the Northern Hemi-
sphere. This has been reaffirmed by a number of recent
measurements in Canada and Sweden (Roulet et al.
2007, Nilsson et al. 2008). However, there is some
suggestion that net ecosystem exchange (NEE) values,
defined as CO2exchange between the land surface
and the atmosphere (the balance between photosyn-
thesis and total ecosystem respiration commonly mea-
sured using eddy covariance), for British and UK peats
may be higher than Gorham’s (1991) estimate. Values
of –40 to – 70 g C m–2 yr–1 (Cannell et al. 1993) and
–102 ± 47 g C m–2 yr–1 (Janssens et al. 2005) have been
reported in the literature. Other estimates suggest a
lower range of values (–32.5 and –13.1 g C m–2 yr–1)
that vary with respect to micro-scale changes in sur-
face peatland topography (Clymo & Pearce 1995).
One of the fundamental differences in temperate UK
peatlands compared to more northern boreal and sub-
arctic peatlands is climate; the UK has a more variable
maritime climate in contrast to the continental climate
associated with areas such as the Canadian and Siber-
ian peatlands. Consequently, UK peatlands are domi-
nated by extensive areas of blanket bog, which is at
the warmer and wetter end of the climate envelope
associated with northern peatlands (Wieder & Vitt
2006). This implies that variables such as precipitation,
frequency of storm events and drought are more likely
to have a significant year-on-year and seasonal effect
on flux rates in UK peatlands compared to those devel-
oped under more continental climates. Blanket peats
in the UK are therefore unusual in a global context as
the climate is generally wetter and they are often
developed in upland areas (with steeper hydrological
gradients); peatland development in the UK during the
Holocene is also thought to be strongly linked to past
human impact as a direct result of early forest clear-
ance leading to wetter soil conditions (Charman 2002,
Simmons 2003).
UK peatlands are often actively managed in ways
which can alter the net ecosystem C balance (NECB)
(Chapin et al. 2006). Land management includes the
widespread use of drainage, prescribed fire, peat
extraction, grazing, fertilisation and liming (Holden et
al. 2007). The effects of these management practices
on plant community productivity and composition are
significant, potentially resulting in a net increase in C
losses to the atmosphere and the fluvial system. Cur-
rently, one of the most widespread peatland manage-
ment interventions in the UK is drain blocking (Holden
et al. 2007); this aims to reverse the drainage process
by restoring the water table to natural levels, with pos-
itive effects for biodiversity, habitat, water and C stor-
age. UK peatlands and their associated vegetation and
surface waters have been significantly impacted by
changes in nitrogen (N) and sulphur (S) deposition
(e.g. Ferguson et al. 1978, Skiba et al. 1989, C. D.
Evans et al. 2006). There is currently widespread con-
cern about the impact of global warming in the north-
ern regions and the potential impact on the C pool (e.g.
Frey & Smith 2005). These effects are closely linked to
changes in seasonal snow cover, temperature, precipi-
tation patterns and the length of the growing season,
all of which will have consequences for C storage in
peatlands.
The combination of climatic variability and the effects
of various direct and indirect anthropogenic threats
make UK peatlands particularly vulnerable to change.
However, whether these effects are having an impact
on the peatland C balance is unclear. A major UK
research challenge is to measure, attribute and predict
rates of change in C budgets and to provide the scien-
14
Country Soil depth
Area 0–100 cm >100 cm Total
(km2) (Mt C) (Mt C) (Mt C)
Scotland 17789 1104a516a1620
England 4246 296b123d419
Wales 732 67b52c119
Northern Ireland 1873 90b54d144
UK 24640 1557a745a2302
aChapman et al. (2009); bBradley et al. (2005); cSmith et al.
(2007); dPro rata to Scottish stocks below 1 m
Table 1. UK peatland area and carbon storage based on
information from a number of sources
Billett et al.: Carbon balance of UK peatlands
tific basis for optimally managing C sequestration in
peatlands for climate mitigation alongside other essen-
tial ecosystem services. In the present study, we review
the current understanding of the contemporary peat-
land C balance from: (1) short-term high resolution
monitoring and (2) long-term paleoecological studies.
We also identify key research areas in need of future de-
velopment required to provide the necessary evidence
for optimally managing long-term peatland C pools.
2. APPROACHES AND METHODOLOGY
The C balance of any terrestrial ecosystem is mea-
sured by quantifying the fluxes (amount of C lost or
gained) of all known C species into and out of a spe-
cific landscape unit of known size (e.g. Billett et al.
2004). For C, this is particularly challenging because of
the range of dissolved (dissolved organic C [DOC], dis-
solved inorganic C [DIC]), gaseous (CO2, CH4) and
particulate (particulate organic C [POC]) species as
well as the number of different flux pathways (land –
atmosphere exchange, water–atmosphere exchange,
precipitation, runoff) involved (see Hope et al. 1994).
Gas exchange with the atmosphere also varies spatially
within a landscape unit (McNamara et al. 2008, Dins-
more et al. 2009b). All fluxes vary spatially and tempo-
rally on a daily, seasonal and annual basis. This sug-
gests that 5+ yr data sets are needed to reduce un-
certainty, obtain reliable long-term
trends and unpick more subtle deca-
dal changes in the NECB.
The ability to combine flux mea-
surements, which in the past have
often been made separately by dif-
ferent groups of researchers, is key
to determining the NECB (Chapin
et al. 2006). Whilst it is well beyond
the scope of this review to describe
all the methods in detail, Fig. 1 sum-
marises the approaches involved
which are described in the paper.
NEE is the largest and most variable
flux term and is commonly measured
using an eddy covariance system
mounted on a flux tower with a CO2
analyser measuring concentrations
at short time intervals (see Foken &
Nappo 2008 for a full description of
the method). The tower samples a
time-variant flux footprint, the orien-
tation and size of which is controlled
by the height of the flux tower, wind
velocity and direction (e.g. Har-
greaves et al. 2003, Roulet et al.
2007). Most flux towers typically sample areas of
~1 km2over distances of 0.25 to 1.5 km. The eddy co-
variance system has also been used to measure CH4
fluxes continuously (Hargreaves & Fowler 1998); recent
advances in instrumentation mean that it will now be
possible to make these measurements more routinely
in the field. Chambers can also be used to measure
land–atmosphere gas exchange of both CO2and CH4
within fixed areas of <0.5 m2. The method, which has
been used and reviewed extensively in the literature
(e.g. Livingstone & Hutchinson 1995), is attractive
because it is cheap, flexible and can be used to exam-
ine the role of specific vegetation communities and
microtopographic features on greenhouse gas fluxes,
capturing the smaller scale variability missed by the
flux towers (McNamara et al. 2008, Dinsmore et al.
2009b). However, chambers can introduce signifi-
cant methodological artefacts into measured flux
rates due to physical and biological disturbance of
the microenvironment; natural small-scale variability
also means that upscaling to large catchment areas is
problematic. The best approach to measuring land –
atmosphere CO2and CH4exchange is therefore to
use both chambers and flux towers; this allows inde-
pendently measured fluxes at different scales to be
combined. Alternatively, there are a number of model-
ling approaches used to estimate NEE, examples of
which are described in some of the following case
studies.
15
DOC input in
precipitation
Evasion of
CO2 and CH4
from stream
surface
Downstream export
of DOC, DIC, POC,
CO2, CH4
Land surface
exchange of
gaseous C Floating chambers
or tracer gas
injections
Outlet measurement
station: discharge
and aquatic C
concentrations
Array of static
chambers Eddy covariance
measurement system
(flux footprint)
Peat cores for
measuring C
accumulation rate
Gas exchange
from bog-pool
system
Fig. 1. Approaches and methods used to measure catchment C balance and accumu-
lation rates in peatlands. The size of the arrows represents the relative magnitude of
the various C fluxes. DOC: dissolved organic carbon; DIC: dissolved inorganic
carbon; POC: particulate organic carbon
Clim Res 45: 13–29, 2010
Measuring fluxes of C leaving the catchment via the
aquatic pathway is generally easier since the flux is
unidirectional (export) and, provided the catchment is
watertight and the boundaries accurately delineated,
the lateral losses occur at one point, the catchment out-
let. This is a critical but often overlooked point; water-
tightness is evaluated by: (1) a judgement based upon
detailed knowledge of the underlying geology and/or
(2) comparison of annual catchment evapotranspiration
rate and water balance (precipitation input minus
runoff) (Cerny et al. 1994) A sampling regime that cap-
tures short-term temporal changes in streamwater hy-
drochemistry and continuous discharge measurement
(often through a control structure such as a flume or
weir) provides the basis for measuring year-on-year
fluxes in DOC, POC, DIC and dissolved gases (CO2and
CH4). Since peatland streams are often supersaturated
with respect to CO2and CH4(Dawson et al. 1995, Billett
& Moore 2008), the aquatic system also provides a con-
duit for exchange of dissolved gases between the water
surface and the atmosphere (the degassing or evasion
flux). Although relatively few measurements have been
made, it is known that the flux varies both spatially and
temporally (Hope et al. 2001), with the greatest fluxes
associated with a high degree of supersaturation and
water turbulence (Billett & Moore 2008); hence it is
difficult to quantify accurately at a landscape scale.
This has led to a range of often indirect approaches to
estimating what appears to be a significant unknown
flux term in the context of the overall peatland C bal-
ance. Inputs of dissolved C also occur in precipitation.
Although the amounts are generally small compared to
the magnitude of land–atmosphere gaseous exchange
and aquatic losses, they should be considered and in-
cluded in a complete catchment C balance.
Few studies in the UK (and globally) include a com-
plete set of directly measured flux C terms, a notable
exception being that from the Auchencorth Moss peat-
land in SE Scotland (Dinsmore et al. 2010). Some of the
longest year-on-year studies are from outside the
UK, most notably in Canada (Roulet et al. 2007) and
Sweden (Nilsson et al. 2008). Whilst this situation is
changing in the UK with new initiatives like the
Centre for Ecology and Hydrology (CEH) carbon
catchments (www.ceh.ac.uk/sci_programmes/Carbon
ExchangeattheCatchmentScale.htm), which will help
improve full quantification of the peatland C balance,
most studies to date still rely on a combination of mea-
sured and estimated values in the absence of a full
suite of data. A number of studies estimate fluxes (e.g.
respiration or decomposition) from established rela-
tionships with water table or temperature (e.g. Worrall
et al. 2003). Others gap-fill the NECB by using litera-
ture values for some of the smaller flux terms like DOC
inputs in precipitation or CH4emissions (Billett et al.
2004, Worrall et al. 2009). Until these best estimate
approaches can be compared to fully quantified bud-
gets, the uncertainty associated with these methods
will remain unknown.
Longer term peat (and C) accumulation rates can be
measured throughout the Holocene period using peat
cores. These paleoenvironmental methods allow con-
temporary accumulation rates to be compared to those
of the recent past and provide context for direct
measurement of C budgets via NEE and monitoring of
aquatic fluxes. Peatlands are one of the few ecosys-
tems that preserve a record of their own development
with the successive accumulation of material over
decades to millennia. Estimates of long-term C accu-
mulation rates have been made for a number of areas
of the world, focusing primarily on millennial-scale C
accumulation rates and changes (e.g. Mäkila 1997,
Mäkila & Moisanen 2007, Beilman et al. 2009). The
data required for such estimates are C density and age
measurements from different depths of peat, the latter
normally based on a series of radiocarbon ages. The
same approach can also be used to estimate relative
rates of C accumulation in near-surface peats by
dating the last few decades, although the cost of radio-
carbon dating this period is often prohibitive (see
Hardie et al. 2007). Another technique for estimating
recent peat ages is the use of short-lived radioisotopes
such as 210Pb; however, these are often problematic in
peat and require validation with other markers (Old-
field et al. 1995). A relatively inexpensive technique
that has been used in a number of palaeoenvironmen-
tal studies are counts of spheroidal carbonaceous par-
ticles (SCPs), which are markers of major changes in
the level of pollution from high-temperature combus-
tion of oil and coal over the past 150 yr. Three time-
markers are available for the UK (Rose et al. 1995),
relating to: (1) the start of SCP deposition in the mid-
19th century, (2) the rapid take-off following wide-
spread expansion of oil and coal-fired power stations in
the 1950s and 1960s, and (3) a peak in SCPs in the mid-
1970s before the introduction of clean burn techno-
logies. The precise dates for these markers vary
regionally, but the northwest of the UK, where most
blanket peatlands occur, shows the same changes
across the region (Rose & Appleby 2005). The rapid
take-off and peak levels are easier to detect than the
start of the curve, and Rose & Appleby (2005) use
cumulative SCP curves prior to the peak instead of
markers to improve comparisons between sites. Since
raw SCP data were not available to us here, we use
markers as the main reference points. The following
set of 5 case studies provides a measure of the current
state of knowledge of the C balance of UK peatlands
and highlights future research challenges. Four are
from sites with continuous monitoring data (Auchen-
16
Billett et al.: Carbon balance of UK peatlands
corth Moss, Moor House, Bleaklow, Conwy) (Fig. 2)
and one is a collection of paleoenvironmental data
from 3 locations across the UK (Laxford Bridge, Loch-
nagar, Butterburn Flow). Collectively they demonstrate
the range of methods and approaches used, as well
as highlighting the importance of individual fluxes in
specific peatlands.
3. PEATLAND C BALANCE AND FLUXES: CASE
STUDIES
3.1. C balance of Auchencorth Moss (SE Scotland)
Auchencorth Moss was the first UK peatland to com-
bine measurements of land–atmosphere C exchange
and aquatic C fluxes (Billett et al. 2004) to produce an
assessment of whether the catchment was acting as a
sink or source of C. The measurements were made for
the period 1996–1998, and the overall conclusion was
that the peatland was either acting as a C source or
was C-neutral. Continuous long-term measurements
of NEE have now been made since 2002 from an eddy
covariance system (prior to 2002, NEE measurements
were made discontinuously). In addition, aquatic C
flux measurements for particulate, dissolved and
gaseous C species have been made continuously since
2006 and the site is building a long-term record of C
import and export terms. Auchencorth Moss is one of
4 CEH carbon catchments which have recently been
set up to quantify the long-term changes in the year-
on-year C balance in peatlands by measuring all the
important C flux terms. The other sites are located in
Forsinard (N Scotland), Moor House (N England) and
Conwy (N Wales).
Auchencorth Moss (55° 47’ N, 03° 14’ W; altitude
range 249 to 300 m) is a 3.4 km2low-lying ombrotrophic
peatland (85% peat) developed on glacial till which lies
above an Upper Carboniferous/Lower Devonian se-
quence of sandstones and shaly sandstones with minor
limestone, mudstone, coal and clay layers (Billett et al.
2004). The majority of the catchment is used for low-
intensity sheep grazing with a small area of peat extrac-
tion in the southwest corner. The vegetation consists of
a patchy mix of grasses and sedges covering a primarily
Sphagnum base layer on a typical peatland hummock/
hollow microtopography; mean annual precipitation
(2002– 2008) and temperature (1971–2000) are 1155 mm
and 10°C, respectively (Drewer et al. 2010) The catch-
ment drains through a series of natural tributaries and
overgrown (>100 yr old) drainage ditches into the main
stream, the Black Burn.
Net ecosystem C balances, which have now been
produced on 2 separate occasions for Auchencorth
Moss covering the 2-yr periods 1996–1998 and 2006–
2008, show that the 2 largest flux terms are NEE and
aquatic C loss (Table 2). The data show significant
variability in the NEE value, which can only partially
be explained by differences and improvements in pro-
tocols used for the calculation of NEE. The second most
significant flux term, loss of C in the aquatic system, is
dominated by DOC export in the Black Burn. In both
the 1996–1998 and 2006–2008 studies, the DOC loss
term is relatively consistent, with POC, DIC and free
CO2making relatively minor contributions to total
aquatic C loss. On both occasions, evasion of CO2from
the water surface to the atmosphere was the second
most important flux involving the aquatic system.
Quantifying the flux is complex since evasion is known
to vary spatially and temporally within catchments
(Billett & Moore 2008, Dinsmore et al. 2010). The larger
evasion flux value in 2006–2008 reflects recent devel-
opments and improvements in upscaling approaches,
which allow a better catchment-scale estimate of CO2
17
Fig. 2. Locations of the main study sites described in the text.
Palaoenvironmental data is also cited from Moor House
Clim Res 45: 13–29, 2010
exchange between the water surface and the atmo-
sphere (Dinsmore et al. 2010). Methane emissions from
the land surface, measured using chambers in 2006–
2008, are unusually low at Auchencorth Moss com-
pared to other peatland sites (Dinsmore et al. 2009b,
Drewer et al. 2010). This is likely to be because site
conditions (low water table, absence of surface bog-
pool systems) do not encourage methane production.
Peat mesocosm experiments in peats derived from the
site also confirm the low methane emissions from
Auchencorth Moss (Dinsmore et al. 2009a).
At this early stage in the long-term measurement of
the overall C balance at Auchencorth Moss it can be
concluded that significant year-on-year variability in
NEE has the largest single impact on the NECB. The
aquatic system is a significant and less variable C
export term. The most recent measurements suggest
the peatland is acting as a major sink for C, with an
accumulation rate of ~72 g C m–2 yr–1 during the period
2006–2008.
3.2. C budgets and fluxes at Moor House (N England)
Moor House is the most extensively studied of all UK
peatland sites, with research on upland ecology dating
back to the 1950s. The mass balance approach was first
used to quantify fluxes of several key macronutrients
(Ca, Na, K, P and N) in Rough Sike as early as the
1960s (Crisp 1966). Measurement of aquatic C con-
centrations and fluxes is a more recent development,
although long-term records still date back to 1993.
Routine measurements now take place on 3 catch-
ments: Rough Sike (0.83 km2) and Cottage Hill
Sike (0.20 km2), both subcatchments of Trout Beck
(11.4 km2). There have also been 2 attempts to estimate
complete C budgets for the Moor House catchment
(Worrall et al. 2003, 2009). Here we present data
on several of the important fluxes from the Moor
House area and demonstrate the importance of within-
catchment hotspots of gaseous C emissions and the
impact of upland land management on C budget, as
well as describe some of the indirect techniques which
can be used to support direct measurements in the
development of the NECB.
The Moor House National Nature Reserve is located
in the North Pennines of England (54° 65’ N, 2° 45 W;
altitude 450 to 893 m). It includes an extensive area of
blanket peat ranging in thickness from 0.4 to 5 m (Heal
& Smith 1978) developed on a thick layer of imper-
meable glacial boulder clay overlying a sequence of
Carboniferous limestone, sandstone and shale. The
most common vegetation species are Calluna vulgaris
(L.), Eriophorum vaginatum (L.), E. angustifolium (L.),
Rubus chamaemorus (L.), Pleurozium schreberi (Brid.)
and Sphagnum capillifolium (Ehrh). Mean annual tem-
perature is 5.3°C; mean annual rainfall (at 550 m) is
1982 mm (Holden & Adamson 2001).
The overall C balance of the largest catchment at Moor
House (Trout Beck; 11.4 km2) has been the focus of a
recent paper by Worrall et al. (2009). Using both direct
measurements (e.g. aquatic DOC fluxes) and indirect
methods (e.g. estimating net ecosystem respiration
and primary productivity using empirical relationships
established between chamber measurements, soil tem-
perature and water table), the authors have compiled
a complete peatland C budget over the period 1993 to
2005. During the 13 yr study period the peatland acted as
a net C sink (–20 to –91 g C m–2 yr–1) with an average of
–56 g C m–2 yr–1. The DOC flux at the outlet of catchment
varied from +10.3 to + 25.3 g C m–2 yr–1. These values are
consistent with those described for Auchencorth Moss,
but are quite distinct from the early values reported for
Moor House of a net sink of –15.4 ± 11.9 g C m–2 yr–1
(Worrall et al. 2003). These differences can be ascribed
to developments in and understanding of methods for
estimating C budgets.
The smallest of the study catchments, Cottage Hill
Sike, developed wholly on blanket peat, is the site of
one of the CEH carbon catchments. Stream water has
been sampled weekly by the Environmental Change
Network at the outlet of Cottage Hill Sike from late
1992 onwards; these data have been used to develop a
15 yr record of DOC fluxes (Fig. 3). Water samples
were analysed for DOC following published protocols
(Sykes & Lane 1996). As continuous long-term dis-
charge monitoring only began at Cottage Hill Sike in
2007, proxy estimates of discharge were obtained by
proportional downscaling of catchment area using
18
Flux term 1996–1998 2007–2008
NEE –27.8 –115
DOC (precipitation) –3.1a–1.4a
CH4emission +4.1a+0.3
Aquatic CO2evasion +4.6 +12.7
Aquatic DOC +26.9 +25.4
Aquatic POC +1.4a+ 3.6
Aquatic DIC +1.2 +0.7
Aquatic CO2+0.9 +1.3
Aquatic CH4+<0.01 +<0.01
Net C balance + 8.3 – 72.4
aEstimated value
Table 2. Carbon fluxes into (positive) and out (negative) of
Auchencorth Moss peatland summed to produce the net
ecosystem C balance (all values in g C m–2 yr–1). 1996–1998
and 2006–2008 data are from Billett et al. (2004) and Dins-
more et al. (2010), respectively; both publications contain a
full description of the methods used. NEE: net ecosystem
exchange; DOC: dissolved organic carbon; POC: particulate
organic carbon; DIC: dissolved inorganic carbon
Billett et al.: Carbon balance of UK peatlands
continuous 15 min discharge measurements made
since 1992 at the nearby Trout Beck weir. Gaps in the
discharge record were filled using empirical relation-
ships between rainfall and discharge (Clark 2005).
Annual load was estimated using Method 5 (Walling
& Webb 1985, Littlewood 1992), which uses long-
term flow data and (in this case) weakly concentration
measurements; confidence limits for fluxes were deter-
mined using methods described in Hope et al. (1997a).
Annual DOC fluxes varied from 14.3 (1995) to 32.7
(2006) with an overall mean of 23.4 g C m–2 yr–1 (Fig. 3).
Although there appears to be no clear statistical trend,
the 4 most recent years (2004–2007) are the years with
the highest annual downstream fluxes and also the
years with the highest recorded and flow-weighted
DOC concentrations. Previous analysis of the flux data
from Cottage Hill Sike has shown that the greatest
DOC fluxes were observed during years with greatest
rainfall (Clark et al. 2007). Although increased rainfall
and discharge is often associated with decreased DOC
concentrations in peat drainage waters (e.g. Bishop
& Pettersson 1996), the decline in the annual flow-
weighted mean concentration in years of high rainfall
was small compared to the magnitude of the increase
in rainfall and overall C export (Clark et al. 2007). Inte-
grated analysis of soil and stream water concentrations
in Cottage Hill Sike have shown a close coupling
between the supply of DOC from near surface soil
water (–1 to –10 cm depth) and DOC transported
through the stream water network (Clark et al. 2005,
Clark et al. 2008). On the basis of these data alone, it is
difficult to determine whether increased DOC fluxes
during years with the greatest rainfall were simply due
to increased leaching and transport of DOC from soil to
stream and/or increased supply of DOC during wet
years due to a decline in consumption and respiration
of DOC within the soil. The degree to which aquatic
and gaseous peatland C fluxes are coupled or discon-
nected has important implications for understanding
the future response of the whole C budget to changes
in climate, especially rainfall.
Numerous gullies intersect the blanket peats at Moor
House, characterised by shallower peat depths, in-
creased water percolation and a distinctive vegetation
cover. A study by McNamara et al. (2008) showed sig-
nificant variation in ecosystem CO2respiration and net
CH4fluxes from typical plant-peat systems at Moor
House located in dendritic drainage gullies and adjacent
blanket peat during the growing season (June–August).
Typically, Eriophorum spp., Sphagnum spp. and mixed
grasses occupied gullies, while Calluna vulgaris oc-
curred in adjacent blanket peat. Gross ecosystem (i.e.
plant and soil) CO2respiration was highest in the areas
of Eriophorum spp. (0.18 ± 0.04 g CO2-C m–2 h–1) com-
pared to those with Sphagnum spp. (0.09 ± 0.01 g CO2-C
m–2 h–1), mixed grasses (0.09 ± 0.02 g CO2-C m–2 h–1) and
C. vulgaris (0.05 ± 0.02 mg CO2-C m–2 h–1). Measure-
ments of the net CH4flux showed higher fluxes when
Eriophorum spp were present (0.002 ± 0.001 g CH4-C
m–2 h–1) compared to Sphagnum spp. (0.0006 ± 0.0004 g
CH4-C m–2 h–1), mixed grasses (0.0001 ± 0.0001 g CH4-C
m–2 h–1) and C. vulgaris (negligible fluxes). A GIS
approach was applied to calculate the contribution of
gullies to landscape-scale greenhouse gas fluxes and it
was estimated that, although gullies occupied only 9.3%
of the total peatland surface, they accounted for 95.8 and
21.6% of the net CH4and gross CO2respiratory fluxes,
respectively. These findings confirmed that landscape
features such as gullies were hotspots of greenhouse gas
emissions and emphasised the importance of including
flux measurements from spatially less significant areas of
the catchment in the overall assessment of the C balance
(e.g. Dinsmore et al. 2009b). In another study of vegeta-
tion management, Ward et al. (2007) found that both
peatland burning and grazing increased net CO2fluxes.
The greatest effects were seen in the burning treatment,
where the observed NEE increased by up to 40 % in the
summer months (May– August), with NEE equal to –38 g
CO2-C m–2 in controls compared with –85 g CO2-C m–2
in burned treatments. Collectively, these recent studies
emphasise the importance of plant–soil interactions as
key regulators of peatland ecosystem greenhouse gas
fluxes at Moor House, and show that changes in plant
community composition and/or productivity play an
important role in determining the magnitude of gaseous
C fluxes.
The extensive gully network at Moor House is also a
reflection of the degree of erosion that the site has
experienced in the past. Unlike the more actively
eroding systems of the southern Pennines, the Moor
House gullies have experienced significant natural
revegetation of gully floors. M. Evans & Warburton
(2005) demonstrated that gully floor revegetation
causes a significant reduction in the particulate flux
from the gullies to the stream system. For the Rough
Sike sub-catchment this was equivalent to a reduction
19
0
10
20
30
40
1993 1995 1997 1999 2001 2003 2005 2007
DOC flux (g C m–2 yr–1)
Fig. 3. Annual variation in dissolved organic carbon (DOC)
flux in Cottage Hill Sike. Error bars represent 95 % confi-
dence limits
Clim Res 45: 13–29, 2010
in POC flux from 45 to 18 g C m–2 yr–1 over the period
1960–1998 (M. Evans et al. 2006). This example of
change in the peatland landscape demonstrates the
effect it has on gully land–atmosphere C (McNamara
et al. 2008) and aquatic (POC) fluxes.
3.3. C fluxes at Conwy (N Wales)
The Migneint is one of the largest areas of blanket
bog in Wales, designated as a Special Area of Conser-
vation (SAC). The northern part of the SAC forms the
headwaters of the River Conwy, and has been inten-
sively studied as one of the 4 CEH carbon catchments.
Annual precipitation and temperature are 2200 mm
and 5.6°C, respectively, whilst the underlying geology
comprises Ordovician mudstones with areas of basic
and acid volcanic tuffs. Within a 1 km2peat-dominated
subcatchment (415 to 487 m), the Nant y Brwyn
(55° 47’ N, 03° 14’ W) fluvial DOC, POC and inorganic
solute concentrations have been monitored fortnightly
since 2006, augmented by a continuous stream moni-
toring system measuring conductivity, pH, coloured
dissolved organic matter and turbidity by fluorescence.
Static chamber gas fluxes have been measured on a
monthly basis in areas of blanket bog and acid grass-
land on mineral soil since 2007. An eddy covariance
system for CO2flux measurement was installed in
2009, along with an expanded static chamber network
covering a wider range of soil/vegetation types within
the catchment. Chemical measurements are supported
by an automatic weather station, discharge and water
table gauges.
The Conwy study is at an early stage, and a full
catchment C budget cannot yet be reliably constructed.
A year of monthly static chamber data suggest mean
CH4emissions of the order of 6 mg C m–2 yr–1 from an
area of Calluna–Sphagnum blanket bog, with a negli-
gible net flux from the acid grassland. This is consider-
ably higher than the value for Auchencorth Moss and
Moor House. At the catchment scale (~80% blanket
bog), this equates to a net CH4emission of around 5 mg
C m–2 yr–1. However, this estimate does not yet include
measured data from potential hotspots such as gullies
and riparian wetlands (McNamara et al. 2008, Dins-
more et al. 2009b), which might be expected to lead to
an increase in the overall catchment flux.
Fluvial C fluxes have been estimated based on 2-yr
volume-weighted mean concentrations (October 2006
to September 2008) and annual runoff. DOC was mea-
sured directly, POC by loss-on-ignition (3 h at 550°C)
and DIC (HCO3–plus dissolved CO2) was estimated
from stream pH and Gran alkalinity, following the
method of Neal (1988). Estimated annual DOC flux for
the 2-yr period was 19.3 g C m–2 yr–1, POC 0.9 g C m–2
yr–1 and DIC 0.6 g C m–2 yr–1. These data suggest that
(as in other catchments) DOC forms the major fluvial C
flux from the Conwy site, and that fluvial export com-
prises a significant component of the overall catchment
C budget.
Conwy is the only study site for which parallel DOC
measurements exist for a lake. Llyn Conwy drains a
small, peat-dominated catchment adjacent to the
Nant y Brwyn, and has a water residence time of 1 yr.
DOC concentrations have been measured for peats and
podzol O horizons, based on monthly sampling at 0 to
10 cm depth using multiple Rhizon™ samplers. Annual
DOC flux estimates (Fig. 4) provide an insight into the
transport and fate of DOC through the soil–water con-
tinuum. They suggest that DOC production from the
peat acrotelm and podzol O horizons are similar, but
that around 50% of this DOC is lost during transport to
the catchment outlet. Considerable retention (and sub-
sequent mineralisation) of DOC is likely within podzol
mineral horizons, but given the small areal extent of
podzols within the Nant y Brywn, it appears that DOC
retention and/or mineralisation may also be occurring
either during lateral transport or immediately following
emergence of DOC-rich water into the stream channel.
Rapid in-stream DOC removal was also inferred from
downstream concentration decreases at Moor House
(Gibson et al. 2009). Based on a comparison of the
Conwy stream and lake output fluxes, it appears that a
further 40% of DOC may be removed by in-lake pro-
cessing. The fate of this DOC (i.e. oxidation to CO2
versus burial in sediments) represents a significant
uncertainty in the C budget of catchments containing
lakes (e.g. Jonsson et al. 2007, Battin et al. 2009).
20
0
5
10
15
20
25
30
35
40
Peat
Podzol
Nant y Brywn
Afon Ddu
Llyn Conwy
DOC (g m–2 yr –1)
(acrotelm)
(O horizon)
Soil
Peat stream
Peat lake
Fig. 4. Estimated dissolved organic carbon (DOC) fluxes (g C
m–2 yr–1) from the surface horizons of 2 soil types, 2 peat
streams and from a lake draining a peat catchment, from
October 2007 to September 2008. Note that the estimate
for Llyn Conwy is per unit land area within the catchment
(i.e. allowing for no DOC production within the lake itself)
Billett et al.: Carbon balance of UK peatlands
3.4. C fluxes on Bleaklow (N England)
A long history of land management, atmospheric de-
position and climate change characterises the Bleak-
low Plateau (S. Pennines) as one of the most impacted
peatland areas in the UK, affected by significant C loss,
particularly by erosion. The plateau (53°27’ N,1° 51’ W)
is an area of extensive deep peat with altitude ranging
from 468 to 630 m above sea level; mean annual rain-
fall is 1554 mm (Daniels et al. 2008) and mean annual
temperature 7.1°C. The area has suffered from exten-
sive erosion and dissection due to heavy grazing, visi-
tor pressure, wildfire and a legacy of atmospheric
deposition of metals and acidifying pollutants. In April
2003 the plateau experienced a wildfire that lead to the
devegetation of 5.5 km2of the peat soils in the area. As
a result of this and other impacts, the whole Bleaklow
area has become the focus of extensive restoration
activity (drain blocking and deliberate revegetation).
The extensive dissection of the plateau by gully ero-
sion and consequent exposure of bare peat on gully
walls and by wildfire scars means that rates of POC
production by surface erosion are high. Measured ero-
sion rates on gully walls are ~0.034 m yr–1 (M. Evans et
al. 2006). Assuming a typical peat density of 0.1 g cm–3
and a C content of 50% ,this equates to a particulate C
loss of 136 g C m–2 yr–1 based on a realistic drainage
density of 20 km km–2 and an average gully depth of
2 m (M. Evans & Lindsay 2010). This POC production
rate does not, however, equate to POC flux at the
catchment scale, because of in-catchment deposition
of POC in gully floors and on floodplains, and because
of the potential transformation of POC to DOC in the
fluvial system (M. Evans et al. 2006, Pawson 2008,
Pawson et al. 2008). Measured catchment POC fluxes
available from a range of small catchments across the
plateau are summarised in Table 3.
Measured POC fluxes vary from 3.4 g C m–2 yr–1 in a
small uneroded catchment to over 90 g C m–2 yr–1 in
severely eroding systems. Flux is controlled not only
by sediment production from bare surfaces, but also by
the connectivity of those surfaces to the stream
drainage network. The lower flux values reported for
Torside Clough (Table 3) are consistent with observa-
tions by Rothwell et al. (2007) that the eroded areas of
this catchment are poorly linked to the stream system.
Connectivity in eroding peatland systems is strongly
controlled by the balance between erosion and re-
vegetation (M. Evans & Warburton 2005, 2007). Al-
though there is evidence of some active revegetation
of the Bleaklow Plateau (Crowe et al. 2008), gully ero-
sion is widespread and covers over 25% of the area
(M. Evans & Lindsay 2010). Values of POC flux in the
higher range of the reported catchments are therefore
more typical of the system. This implies that, unlike
intact systems discussed by Hope et al. (1997b), the flu-
vial C flux of the Bleaklow Plateau is dominated by C
in the solid phase. For comparison, Pawson et al. (2008)
reported a DOC flux from the Upper North Grain
catchment of 18.5 g C m–2 yr–1 and O’Brien et al. (2008)
estimated fluxes ranging from 6 to 18 g C m–2 yr–1 from
3 catchments on Bleaklow where there had been recent
management change.
The dominance of the POC flux within the fluvial C
budget of eroding systems emphasises the importance
of a major knowledge gap in the understanding of
peatland C cycling, namely the fate of POC in the
fluvial system. There is some evidence that a propor-
tion of POC is converted to both the gaseous and dis-
solved phases during transport (M. Evans et al. 2006,
Pawson et al. 2008, Pawson 2008), suggesting that con-
sideration of the POC flux is pertinent to assessments
of atmospherically active C flux. The magnitude of this
effect and the transformation processes require further
research.
Other research on the Bleaklow Plateau has focussed
not just on the fluvial system, but on producing whole
C budgets from specific catchment areas using a com-
bination of direct and indirect methods. The 2 sites
were chosen as controls against which success of re-
21
Catchment Area POC flux Source Notes
(km2) (g C m–2 yr–1)
Upper North Grain 0.38 95.7 M. Evans et al. (2006) Severe gully erosion
Upper North Grain (Snake Pass) 0.85 74.0 Pawson et al. (2008) Severe gully erosion
Torside Clough 3.31 11.3 Rothwell (2006) Calculated from data in Rothwell (2006)
based on 70% organic content
Unnamed micro- catchment 0.0007 92.5 M. G. Evans unpubl. data Bare peat catchment
near Bleaklow summit
Unnamed micro- catchment 0.005 3.4 M. G. Evans unpubl. data Uneroded catchment
near Snake summit
Lady Clough 1.33 44.8 Pawson (2008)
Table 3. Measured particulate organic carbon (POC) fluxes from Bleaklow Plateau catchments
Clim Res 45: 13–29, 2010
vegetation could be judged thought to be typical of the
vegetation on the Bleaklow Plateau. One site (PD) was
dominated by Eriophorum spp., the second control
(WC) was dominated by the shrubs Vaccinium spp.
and Empetrum spp. The catchment areas for PD and
WC were 0.005 and 0.02 km2, respectively. The meth-
ods used to develop overall catchment C budgets of PD
and WC were:
(1) Plot-scale measurements using piezometers
(water table), chambers (gas fluxes) and erosion pins.
Sites were sampled from December 2006; data pre-
sented here are for the first 2 yr of the study.
(2) Catchment-scale measurements of fluvial DOC
fluxes based on weekly samples collected in 2008. The
method used for estimating excess dissolved CO2in
runoff from fresh soil water samples drawn from
piezometers in described in Worrall et al. (2009). POC
fluxes losses (surface lowering) are measured from
annual measurements of nests of erosion pins (12 pins
per plot). At the catchment scale, POC flux was mea-
sured directly through runoff sampling using rising
stage and automatic water samplers.
(3) Surface exchange of CO2and CH4was measured
using a closed chamber method (see Worrall et al.
2009).
(4) Interpolation and extrapolation methods are used
to upscale and estimate C budgets. For example, NEE
is estimated using calibrated equations for soil respira-
tion and primary productivity. For a full description of
the approach and the calculated errors see Worrall et
al. (2009).
The combined flux measurement shows strong differ-
ences between the 2 sites, with one appearing as a sink
and the other a source (Table 4). The main differences
in the flux terms, which appear to produce this shift in
the NECB, are a weaker drawdown of CO2from the
atmosphere (relatively lower primary productivity and
higher respiration rates) and much higher fluvial (POC
and DOC) fluxes. This could be due to the difference in
vegetation type between the 2 control sites or it could
represent the greater proportion of bare soil under
shrub vegetation. In conclusion, research on the Bleak-
low Plateau suggests that this part of the UK is charac-
terised by unusually high POC fluxes. Clearly manage-
ment can have a significant effect on peatland C
balance in both the short and the long term.
3.5. Historical rates of C accumulation
We compiled the published SCP data from England
and Scotland to provide estimates of peatland C accu-
mulation for different time periods (Table 5). Although
there are many other sites with SCP data available in
the literature, most of these only used SCPs to provide
dates for palaeoenvironmental reconstructions (e.g.
Chambers et al. 1999), and relatively few sites have
bulk density and/or C content data available, which
are needed to transform depths and dated layers to net
C accumulation rates. There are sufficient data to pro-
vide some insights into C accumulation rates over the
last 150 yr on peatlands in the UK.
To assemble these data we took published depth
estimates for the 3 major changes in SCPs (start, take-
off and peak) and bulk density data to estimate peat
accumulation as dry mass per year since the mid-19th
century, the mid-20th century and the late-1970s,
together with accumulation rates between these time
intervals. Regional curves from Rose & Appleby (2005)
were used to estimate ages from unpublished SCP data
(Table 5). Dry mass accumulation was converted into
C accumulation using loss-on-ignition (LOI) and % C
data. Where only bulk density data were available, we
assumed 98% LOI and C content of 50 % organic mat-
ter. LOI was measured using methods based on those
of Allen (1989).
Calculated rates of C accumulation varied between
35.1 and 209.1 g C m–2 yr–1. The highest rates generally
occur in the uppermost peat layers, as would be
expected due to the incomplete decay of organic mat-
ter in the acrotelm. However, temporal trends within
individual sites suggest that despite incomplete decay
in near-surface material, rates may have slowed
(Lochnagar) or remained the same (Laxford Bridge)
since the 1970s. Consistent with likely trends from
incomplete decay, rates are generally lower for the
mid-19th to mid-20th century period (39.6 to 73.1 g C
m–2 yr–1), but again rates at Lochnagar are comparable
with the late 20th century, which could be interpreted
as a recent reduction in productivity or more rapid
decay of recent organic matter. Differences between
sites might be related to local context, especially vege-
tation composition and hydrology. The highest rates of
C accumulation occur in the raised mire sites (Butter-
22
Site Fluvial Fluvial NEE CH4Dissolved Carbon
DOCaPOC CO2balance
Year 1
PD 13.0a1.9 –92 0.7 1.3 –75
WC 95.6and – 44 0.7 1.3 nd
Year 2
PD 13.0a3.4 –12000.7 1.3 –102
WC 95.6a37.7 – 65 0.7 1.3 70
aMean 2-yr value
Table 4. Summary of the individual fluxes and the overall esti-
mated C budgets (g C m–2 yr–1) for the 2 study catchments (PD
and WC) on Bleaklow Plateau. DOC: dissolved organic car-
bon; POC: particulate organic carbon; NEE: net ecosystem
exchange; nd: not determined
Billett et al.: Carbon balance of UK peatlands
burn Flow and Laxford Bridge) with rates >100 g C
cm–2 yr–1 since the mid-20th century. These sites are
generally wetter with Sphagnum-dominated vegeta-
tion. However, the Moor House blanket mire has simi-
lar rates in areas unaffected by burning, although it
might be expected to have lower Sphagnum presence
and a drier surface. A wider range of accumulation
rates was found for this site by Hardie et al. (2007) who
reported a range of 20 to 125 g C m–2 yr–1 since ca. 1950
based on the presence of post-bomb 14C in peat. How-
ever, the precision of the estimates by Hardie et al.
(2007) was low because sample depth resolution was
only 4 cm, leading to wide variations in the range of
estimated C accumulation rates for individual cores
(e.g. 19.6 to 60 g C m–2 yr–1 for a single core). This
emphasises the need for high resolution (samples of
1 cm thickness or less) for stratigraphic estimates of C
accumulation, to enable differences between sites to
be identified. The lowest rates are at Lochnagar, a high
altitude, low net primary productivity, sloping blanket
mire site more likely dominated by vascular plants and
with a relatively dry surface due to more rapid runoff.
The estimates of C accumulation presented here are
subject to a number of potential errors and uncertain-
ties, principally related to dating. Despite extensive
work on SCP profiles in lakes dated independently
with 210Pb, the absolute ages of changes in SCPs may
only be accurate to within ±15 yr (Rose & Appleby
2005), although relative age estimates may be much
better within individual regions. Furthermore, the pre-
cision of the age estimate also depends on the sam-
pling precision down core. However, the scale of the
uncertainty is unlikely to alter the main trends and
differences between site types, and the longer-term
absolute rates of C accumulation from the mid-19th
century. Further data are clearly required to under-
stand the relationship between C accumulation, site
type and management regime (e.g. Garnett et al.
2000), as well as to evaluate any possible regional pat-
terns. For example, it might be expected that C accu-
mulation is lower in regions at the limit of peat distrib-
ution or in less favourable topographic situations.
These regions are also those most likely to be vulnera-
ble to future change. It should be remembered that
apparent rates of C accumulation in recent peat do not
reflect the C balance of the entire peat column. The
deeper peat continues to decay, albeit at a much
slower rate (Clymo 1984). The total C balance is thus
partly dependent on losses at depth as well as gains in
the near surface. Large changes in hydrological or
temperature regime may lead to significant shifts in C
balance of deeper as well as near-surface peat. This
highlights an important difference between the catch-
ment-based NECB approach and the use of peat cores
to measure recent changes in C accumulation rates.
23
Location Site type Sample cores Average rates Rates between periods Sampling Source
1970s Mid-20th C Mid-19th C Mid-20th C– Mid-19th C– date
1970s mid-20th C
Butterburn Flow Raised mire BFA 122.5 99.5 81.8 82.2 73.1 1999 Charman (2007)
BFB 160.6 136.8 71.6 119.0 39.7 1999
Mean 141.5 118.1 76.7 100.6 56.4
Lochnagar Sloping NAG51 35.1 45.7 41.3 56.9 39.6 1997 Yang et al. (2001)
blanket mire Mean (n = 10) 27.2
Moor House Blanket Grazed (n = 4) 115.4 1997 Garnett et al. (2000)
mire Grazed, burned (n = 4) 64.9
Unburned, ungrazed (n = 4) 98.4
Mean (n = 12) 92.9
Laxford Bridge Raised LAB-A 209.1 206.2 203.4 1995 D. J. Charman &
mire Coovrey unpubl. data
Table 5. Rates of C accumulation (g C m–2 yr–1) from peatlands in England and Scotland where spheroidal carbonaceous particle (SCP) and bulk density data are available.
For the Scottish sites, start of SCP = 1865, mid-20th C = 1960. For England, mid-20th C = 1950. The mid-19th century age estimate for Butterburn Flow is based on
wiggle-matched 14C ages and a pollen marker (Charman 2007). Gaps indicate no data
Clim Res 45: 13–29, 2010
4. DISCUSSION
4.1. UK peatland C balance: uncertainty and
measurement techniques
The question ‘is the C balance (or rate of C accumu-
lation) in UK peatlands changing?’ has provided focus
for this review. It highlights the importance of reducing
levels of uncertainty in flux measurements and apply-
ing the best methods and approaches. Contemporary
flux data gives a snapshot in time with annual report-
ing of UK peatland C budgets at a relatively early
stage compared to sites in Canada (Roulet et al. 2007)
and Sweden (Nilsson et al. 2008). Generally, all sites
showed NEE to be the largest and most variable flux
term both temporally (from continuous measurements)
and spatially with respect to vegetation cover, gullying
and surface wetness. For example, year-on-year NEE
fluxes measured at Mer Bleue (Canada) using eddy
covariance varied from –2 to –112 g C m–2 yr–1, with a
6 yr mean (±SE) of –40.2 ± 40.5 g C m–2 yr–1 (Roulet et
al. 2007). The long-term (2003–2008) mean NEE for
Auchencorth Moss in SE Scotland is –74 ± 22 g C
m–2 yr–1, with year-on-year variability ranging from a
small source of CO2in 2003 (+20) to a significant sink
in in 2008 (–136) (Dinsmore et al. 2010). In all cases the
potential errors in the annual calculation of NEE val-
ues are large (30 to 100% of the overall flux). Clearly
as more data become available the controls on long-
term temporal variability in NEE will become easier to
interpret and model, leading to a reduction in uncer-
tainty. Studies have shown that it is possible to esti-
mate NEE from other factors (Worrall et al. 2009),
although no comparison of direct versus indirect meth-
ods of measuring and/or estimating this key flux term
have, to date, been made.
In the UK, as well as in other countries, there has
been greater emphasis on aquatic C fluxes compared
to continuous land–atmosphere exchange monitoring,
which is more expensive and technically more chal-
lenging. Whilst aquatic fluxes are important, they are
generally smaller. The 15 yr DOC flux record at Cot-
tage Hill Sike shows that the flux is annually less vari-
able (Fig. 3) than NEE and levels of uncertainty are
significantly lower. The DOC flux data from all of the
study sites (catchment sizes 0.2 to 3.35 km2) also point
to a level of consistency, with fluxes ranging from
18.5 to 26.9 g C m–2 yr–1. Fluxes from smaller (0.005 to
0.05 km2) catchments are more variable (Table 4),
hence our data suggest that levels of uncertainty in
aquatic flux measurements can be significantly re-
duced by working at scales >0.1 km2. The level of
uncertainty in aquatic flux measurements will also be
reduced by the increased use of sensors to continu-
ously measure concentrations of C species like DOC
and CO2(Dinsmore & Billett 2008, Koehler et al. 2009,
Johnson et al. 2010).
All of these fluxes are mediated spatially with
respect to vegetation type and land management prac-
tices. Processes affecting POC export differ, as these
exports are greatest in disturbed landscapes affected
by erosion. An optimal monitoring strategy for measur-
ing the annual peatland C balance requires continuous
flux tower measurements of CO2and CH4exchange;
continuous, weekly or biweekly spot sampling for
DOC (associated with continuous discharge), a method
for estimating the evasion flux; and, ideally, event-
based POC sampling. The importance of individual
flux terms (and the need to measure them) will vary
from catchment to catchment in response to the degree
of disturbance and the amount and type of precipitation.
Other methods have also been used to address the
question of peatland stability. Isotopic evidence (14C)
can be used to identify whether older C is being
released into the drainage system, a potential indicator
of mobilisation of deep peat C which has been stored
and remained immobile for 100s or 1000s of years.
Measurements of DOC-14C from UK peat catchments
suggest that the majority is derived from recently fixed
plant material rather than older peat (C. D. Evans et al.
2007, Palmer et al. 2001). On the other hand, 14C mea-
surements of evaded CO2suggest that significantly
older C is being released in some peatland streams,
some of which may be geogenic rather than biogenic
in origin (Billett et al. 2007).
4.2. Role of the aquatic flux pathway
This group of case studies highlights the importance
of including aquatic C fluxes in peatland C budgets.
DOC fluxes in the range of 19 to 27 g C m–2 yr–1 are
typical of UK peatlands and significant in the context
of the overall budget; POC losses are more variable,
but can reach up to 100 g C m–2 yr–1 in eroding sys-
tems. In addition, recent measurements of CO2evasion
from the water surface of peatlands to the atmosphere
(Hope et al. 2001, Dinsmore et al. 2010) show that at
the catchment scale, fluxes are significant (up to 14.1 g
C m–2 yr–1) and comparable to those of DOC. The com-
bined DOC, POC and CO2evasion fluxes represent a
significant loss term in the peatland C budget and one
that is clearly sensitive to change. In UK blanket peats,
the rapid rate of hydrological flushing clearly favours
fluvial C export over other C loss pathways. This
review highlights the significant rates of POC removal
in the aquatic system from the eroding peatlands on
Bleaklow, and evidence of past high rates from Moor
House (M. Evans & Warburton 2005) emphasises the
importance of maintaining physical stability in peat-
24
Billett et al.: Carbon balance of UK peatlands
lands. As UK peatlands become increasingly climate-
stressed over the next century (Clarke et al. this Spe-
cial), the risk of enhanced erosion is likely to extend to
other peatland systems. A full assessment of the poten-
tial impact of reported high POC fluxes also requires
further research on the fate of POC in the fluvial envi-
ronment. Transformation of POC to DOC and oxidation
of POC from river floodplains has the potential to im-
pact the greenhouse gas budgets of peatlands as well
as individual peatland C budgets (Pawson 2008). The
rates and controls of these transformations is an impor-
tant gap in our current knowledge of fluvial C fluxes.
Whilst contemporary C budgeting involves measur-
ing all the flux terms, a more selective approach would
be to identify a key flux pathway that shows less short-
term (annual or seasonal) variability and, therefore, is
more likely to show long-term underlying change.
Since the aquatic system and the export rates of vari-
ous forms of C are controlled by a variety of (often
interconnected) within-catchment processes, stream
export of C may be a better integrator (or signal) of
change within the catchment. Interestingly, the typical
annual export of DOC for peatland streams (19 to 27 g
C m–2 yr–1) is comparable to Gorham’s (1991) estimate
of the contemporary C accumulation rate for northern
peatlands (23 g C m–2 yr–1). We could therefore hypoth-
esise that change in accumulation rate might be con-
nected to and be reflected in changes to the release of
C into the peatland aquatic system. As long-term data
on DOC fluxes are more readily available and easier to
monitor, this approach may be more pragmatic. How-
ever, before this assumption can be made, we need to
establish to what extent the aquatic (DOC) and
gaseous C fluxes (NEE) are coupled. Does an increase
in DOC flux equate to an increase in C storage? Or
does an increase in DOC flux result in a lower gaseous
flux, as less C is mineralised in situ? Detailed long-
term monitoring data over a range of climatic condi-
tions will be required to unravel these factors.
There are, however, many uncertainties in relating
DOC flux to other components of the C cycle. In par-
ticular, 14C data, indicating a predominantly recent
origin for DOC, imply that DOC may be more closely
related to primary production than to peat decomposi-
tion. Additionally, there is evidence that other factors,
notably decreases in acid deposition, have altered
the mobility of DOC within upland soils, leading to
increasing aqueous DOC losses (Monteith et al. 2007).
For peat catchments specifically, fewer long-term data
are available from which to assess whether fluxes
have increased. Data for Moor House are somewhat
equivocal (Fig. 3), while the other sites considered
here have insufficiently long records to evaluate
trends. However, longer records (1988–2008) for the
4 peat-dominated catchments in the UK Acid Waters
Monitoring Network all show clear increases in DOC
concentration (C. D. Evans et al. 2005). In the absence
of clear hydrological changes over the same period
(C. D. Evans et al. 2006), an approximate rate of flux
increase can be obtained by multiplying the rate of
annual DOC increase by mean runoff. These calcula-
tions (Table 6), which also include the Cottage Hill
Sike data, suggest that DOC fluxes have risen by
between 0.3 and 0.6 g C m–2 yr–1 in the period
2003–2007. If this was applied to a 20 yr period
(1988–2008) it would be equivalent to a total increase
in the annual flux of 5.5 to 10.5 g m–2. One of the
rivers (Etherow) drains the northern part of the Bleak-
low Plateau and the increase here can be compared to
the high POC flux estimates given earlier. Overall, it
appears that DOC losses from UK peatlands may have
increased considerably over the last 20 yr. However,
the controls on DOC leaching are complex, and the
nature of inter-linkages between DOC flux and other
components of the C balance remains obscure. There-
fore, we do not (at present) believe that fluvial DOC
flux can be used as a reliable indicator of changes in
the overall peatland C balance.
25
Site Period Mean runoff Mean Rate of concen- Mean Rate of flux
(m yr–1) concentration tration change flux change
(mg l–1) (mg l–1 yr–1) (g C m–2 yr–1) (g C m–2 yr–1)
Cottage Hill Sike 1993–2007 1.55 18.93 0.32 29.2 0.54
(N Pennines)
River Etherow 1988–2007 1.44 9.23 0.42 10.5 0.48
(S Pennines)
Loch Tinker 1988–2007 1.86 6.81 0.20 12.7 0.37
(C Scotland)
Beaghs Burn 1988–2007 1.22 15.44 0.43 18.8 0.52
(N Ireland)
Coneyglen Burn 1988–2007 1.10 10.08 0.25 11.1 0.27
(N Ireland)
Table 6. Estimated dissolved organic carbon concentrations, fluxes and rate of change at 5 UK peat-dominated catchments
for which long-term data are available. Mean concentration and mean flux refer to the period 2003 –2007
Clim Res 45: 13–29, 2010
4.3. Peatland C management
Optimal management of soil C and its losses has re-
cently been extensively reviewed (Dawson & Smith
2007). The case studies presented here demonstrate that
peatland management and change clearly affects annual
fluxes, but how do these changes affect the overall C bal-
ance or pool? To put Holocene rates of C accumulation
and contemporary flux rates into context, Billett et al.
(2006) estimated the length of time it would take to
release all the soil organic C from a small peatland catch-
ment in NE Scotland based on rates of annual export of
organic C in the aquatic system alone. Assuming no
further C sequestration from the atmosphere, it would
take ~4300 yr for the catchment to be depleted of its store
of soil organic C. This demonstrates the magnitude of the
stored C pool in relation to contemporary flux rates. A
similar calculation for the impacted peatland catchments
on the Bleaklow Plateau based on POC losses alone
gives a figure of ~1150 yr. Therefore, considering the
changes in climate predicted (Clark et al. this Special),
there is a millennial-scale period for managing declining
peatlands, even if annual drawdown of CO2from the
atmosphere ceased. This has important implications
for peatland management in the UK.
5. CONCLUSIONS AND FUTURE RESEARCH
CHALLENGES
We are at a relatively early stage in making year-
on-year integrated C fluxes measurements in UK peat-
lands, with few complete data sets to explore long-term
trends and drivers of change. The contemporary mea-
surements that exist from 2 large catchments suggest
that current accumulation rates (– 56 to –72 g C m–2 yr–1)
may be lower than those seen over the last 150 yr in peat
cores (–35 to –209 g C m–2 yr–1). This conclusion may well
change in the future as longer-term data sets become
available. There are also considerable between-site dif-
ferences in rates derived from peat cores and a number
of fundamental differences in approaches. Our current
stage of knowledge therefore suggests that UK peat-
lands continue to operate as a long-term sink for C
(as they have done since the end of the last ice age) un-
less they are affected by high management or climatic
pressures. Although this may change in the future, the
ability to manage peatland C stores is evident and
should be a high priority for stakeholders.
Currently we have a rough understanding of the
time scale over which we expect flux terms to vary, and
the environmental factors that influence these, but our
knowledge of the magnitude of these changes is in-
complete. UK peatlands are managed, and manage-
ment intensity varies from location to location. Site-
based evidence suggests that management, like burn-
ing, can reduce the C sink strength, but we are unable
to quantify the resulting flux changes at a national
scale. Alternatively, interventions like drain blocking
are able to increase the sink strength of peatlands. We
also know that large areas of blanket peats have been
exposed to a long-history of atmospheric pollution
(particularly S and N); however, we do not know what
the impact of this has been on long-term C turnover
(e.g. by lowering the C:N ratio) or the individual flux
terms. More detailed process-level research suggests
that the role of plant community composition is impor-
tant as a determinant of ecosystem scale C fluxes. The
potential to use this understanding in the future devel-
opment of mechanistic models of peatland C dynamics
holds considerable promise as a means to reduce
uncertainty in model estimates.
Our understanding of interactions between local and
regional pressures on the contemporary C balance is
incomplete, as is our understanding of how these pres-
sures may interact with future climate change to
increase the net fluxes of C from these systems. High
quality, reliable flux data, with empirical understand-
ing of the factors controlling these fluxes is essential to
underpin the modelling work aimed at upscaling con-
temporary flux measurements spatially and tempo-
rally, and under future climate change scenarios.
There have been a number of recent demands for im-
proved understanding and assessment of the source–
sink status of UK peatlands to inform policy, as envi-
ronmental stewardship moves from a single resource
focussed on exploitation for food production to a more
holistic ecosystem services approach (Everard 2009).
Attempts to quantify the C budget of UK peatlands are
in their infancy and the short-term nature of these data
sets is currently a weakness. The growing interest in
managing C in the context of ecosystem services as a
viable climate mitigation strategy suggests that there is
a real need for a coherent, unified approach under-
pinned by scientific research based on long-term, site-
based monitoring networks.
Acknowledgements. The research presented here has been
supported by a range of sources, in particular the UK Natural
Environment Research Council. J.M.C. acknowledges the
support of the Environment Agency (Science Project
SC070036). D.J.C. thanks H. Yang for providing the raw data
for the published plots in Yang et al. (2001) to enable cal-
culation of C accumulation rates for Lochnagar.
LITERATURE CITED
Allen SE (1989) Chemical analysis of ecological materials.
Blackwell Scientific Publications, Oxford
Battin TJ, Luyssaert S, Kaplan LA, Aufdenkampe AK, Richter
A, Tranvik LJ (2009) The boundless carbon cycle. Nat
26
➤
Billett et al.: Carbon balance of UK peatlands
Geosci 2:598– 600
Beilman DW, MacDonald GM, Smith LC, Reimer PJ (2009)
Carbon accumulation in peatlands of West Siberia over
the last 2000 years. Global Biogeochem Cycles 23:GB1012
doi:10.1029/2007GB003112
Billett MF, Moore TR (2008) Supersaturation and evasion of
CO2and CH4in surface waters at Mer Bleue Peatland,
Canada. Hydrol Process 22:2044–2054
Billett MF, Palmer SM, Hope D, Deacon C and others (2004)
Linking land-atmosphere-stream carbon fluxes. Global
Biogeochem Cycles 18:GB1024 doi:10.1029/2003GB002058
Billett MF, Deacon C, Palmer SM, Dawson JJC, Hope D (2006)
Connecting organic carbon in streamwater and soils in
a peatland catchment. J Geophys Res 111:G02010 doi:
10.1029/2005JG000065
Billett MF, Garnett MH, Harvey F (2007) UK peatland streams
release old carbon dioxide to the atmosphere and young
dissolved organic carbon to rivers. Geophys Res Lett
34:L23401 doi:10.1029/2007GL031797
Bishop K, Pettersson C (1996) Organic carbon in the boreal
spring flood from adjacent subcatchments. Environ Int
22:535– 540
Bradley RI, Milne R, Bell J, Lilly A, Jordan C, Higgins A (2005)
A soil carbon and land use database for the United King-
dom. Soil Use Manag 21:363– 369
Cannell MGR, Dewar RC, Pyatt DG (1993) Conifer planta-
tions on drained peatlands in Britain: A net gain or loss of
carbon? Forestry 66:353–369
Cerny J, Billett MF, Cresser MS (1994) Element budgets. In:
Moldan B, Cerny J (eds) Biogeochemistry of small catch-
ments: a tool for environmental research. Wiley, London,
p 189–205
Chambers FM, Mauquoy D, Todd PA (1999) Recent rise to
dominance of Molinia caerulea in environmentally sensi-
tive areas: new perspectives from palaeoecological data.
J Appl Ecol 36:719– 733
Chapin FS III, Woodwell GM, Randerson JT, Rastetter EB and
others (2006) Reconciling carbon-cycle concepts termino-
logy and methods. Ecosystems 9:1041–1050
Chapman SJ, Bell J, Donnelly D, Lilly A (2009) Carbon stocks
in Scottish peatlands. Soil Use Manag 25:105–112
Charman DJ (2002) Peatlands and environmental change.
John Wiley & Sons, New York, NY
Charman DJ (2007) Summer water deficit variability controls
on peatland water-table changes: implications for Holo-
cene palaeoclimate reconstructions. Holocene 17:217–227
Clark JM (2005) Environmental controls on the production
and export of dissolved organic carbon from an upland
peat catchment. PhD thesis, University of Leeds, Leeds
Clark JM, Chapman PJ, Adamson JK, Lane SN (2005) Influ-
ence of drought-induced acidification on the mobility of
dissolved organic carbon in peat soils. Global Change Biol
11:791–809 doi:10.1111/j.1365-2486.2005.00937.x
Clark JM, Lane SN, Chapman PJ, Adamson JK (2007) Exports
of dissolved organic carbon from an upland peatland dur-
ing storm events: implications for flux estimates. J Hydrol
347:438– 447
Clark JM, Lane SN, Chapman PJ, Adamson JK (2008) Link
between DOC in near surface peat and stream water in an
upland catchment. Sci Total Environ 404:308– 315
Clymo RS (1984) The limits to peat bog growth. Proc R Soc
Lond B 303:605– 654
Clymo RS, Pearce DME (1995) Methane and carbon dioxide
production in, transport through, and efflux from a peat-
land. Philos Trans R Soc Lond A 351:249–259
Crisp DT (1966) Input and output of minerals for an area of
Pennine moorland: the importance of precipitation, drainage,
peat erosion and animals. J Appl Ecol 3:327– 348
Crowe SK, Evans MG, Allott THE (2008) Geomorphological
controls on the re-vegetation of erosion gullies in blanket
peat: implications for bog restoration. Mires Peat 3:1–14
Daniels SM, Agnew CT, Allott TEH, Evans MG (2008) Water
table variability and runoff generation in an eroded peat-
land, South Pennines, UK. J Hydrol 361:214–226
Dawson JJC, Smith P (2007) Carbon losses from soil and its
consequences for land-use management. Sci Total Envi-
ron 382:165–190
Dawson JJC, Hope D, Cresser MS, Billett MF (1995) Down-
stream changes in free CO2in an upland catchment in
northeastern Scotland. J Environ Qual 24:699 –706
Dinsmore KJ, Billett MF (2008) Continuous measurement and
modeling of CO2losses from a peatland stream during
stormflow events. Water Resour Res 44:W12417doi:10.1029/
2008WR007284
Dinsmore KJ, Skiba UM, Billett MF, Rees RM (2009a) Effect of
water table on greenhouse gas emissions from peatland
mesocosms. Plant Soil 318:229–242
Dinsmore KJ, Skiba U, Billett MF, Rees RM, Drewer J (2009b)
Spatial and temporal variability in CH4and N2O fluxes
from a Scottish ombrotrophic peatland; implications for
modelling and upscaling. Soil Biol Biochem 41:1315–1323
Dinsmore KJ, Billett MF, Skiba UM, Rees RM, Helfter C
(2010) Role of the aquatic pathway in the carbon and
greenhouse gas budgets of a peatland catchment. Glob
Change Biol 16:2750–2762
Drewer J, Lohila A, Aurela M, Laurila T and others (2010)
Comparison of greenhouse gas fluxes and nitrogen bud-
gets from an ombrotrophic bog in Scotland and a pristine
mire in Finland. Eur J Soil Sci 61:640–650
Evans CD, Monteith DT, Cooper DM (2005) Long-term
increases in surface water dissolved organic carbon: ob-
servations, possible causes and environmental impacts.
Environ Pollut 137:55–71
Evans CD, Chapman PJ, Clark JM, Monteith DT, Cresser MS
(2006) Alternative explanations for rising dissolved
organic carbon export from organic soils. Global Change
Biol 12:2044–2053
Evans CD, Freeman C, Cork LG, Thomas DN and others (2007)
Evidence against recent climate-induced destabilisation
of soil carbon from 14C analysis of riverine dissolved
organic matter. Geophys Res Lett 34:L07407 doi:10.1029/
2007GL029431
Evans M, Lindsay J (2010) High resolution quantification of
gully erosion in upland peatlands at the landscape scale.
Earth Surf Proc Landf 35:876 –886
Evans M, Warburton J (2005) Sediment budget for an eroding
peat moorland catchment in northern England. Earth Surf
Process Landf 30:557– 577
Evans M, Warburton J (2007) The geomorphology of upland
peat: pattern, process, form. Blackwell, Oxford
Evans M, Warburton J, Yang J (2006) Eroding blanket peat
catchments: global and local implications of upland organic
sediment budgets. Geomorphology 79:45– 57
Everard M (2009) The business of biodiversity. WIT Press,
Ashurst
Ferguson P, Lee JA, Bell JNB (1978) Effects of sulphur pol-
lutants on the growth of Sphagnum species. Environ Pollut
16:151–162
Foken T, Nappo CJ (2008) Micrometeorology. Springer Verlag,
Berlin
Frey KE, Smith LC (2005) Amplified carbon release from
vast West Siberian peatlands by 2100. Geophys Res Lett
32:L09401 doi:10.1029/2004GL022025
Garnett MH, Ineson P, Stevenson AC (2000) Effects of burn-
27
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤➤
➤
➤
➤
➤➤
➤
➤
➤
➤
➤
➤➤
➤
➤
➤
➤
Clim Res 45: 13–29, 2010
ing and grazing on carbon sequestration in a Pennine
blanket bog, UK. Holocene 10:729– 736
Gibson HS, Worrall F, Burt TP, Adamson JK (2009) DOC bud-
gets of drained peat catchments: implications for DOC
production in peat soils. Hydrol Process 23:1901–1911
Gorham E (1991) Northern peatlands: role in the carbon cycle
and probable responses to climatic warming. Ecol Appl
1:182–195
Hardie SML, Garnett MH, Fallick AE, Rowland AP, Ostle NJ
(2007) Spatial variability of bomb C-14 in an upland peat
bog. Radiocarbon 49:1055–1063
Hargreaves KJ, Fowler D (1998) Quantifying the effects of
water table and soil temperature on the emission of
methane from peat wetland at the field scale. Atmos
Environ 32:3275–3282
Hargreaves KJ, Milne R, Cannell MGR (2003) Carbon bal-
ance of afforested peatland in Scotland. Forestry 76:
299– 317
Heal OW, Smith R (1978) The Moor House Program: introduc-
tion and site description. In: Heal OW, Perkins DF (eds)
Production ecology of British moors and montane grass-
lands. Springer, Berlin, p 304–331
Holden J, Adamson JK (2001) Gordon Manley and the north
Pennines. J Meteorol 26:329–333
Holden J, Shotbolt L, Bonn A, Burt TP and others (2007) Envi-
ronmental change in moorland landscapes. Earth Sci Rev
82:75–100
Hope D, Billett MF, Cresser MS (1994) A review of the export
of carbon in river water: fluxes and processes. Environ
Pollut 84:301–324
Hope D, Billett MF, Cresser MS (1997a) Exports of organic
carbon from two river systems in NE Scotland. J Hydrol
193:61–82
Hope D, Billett MF, Milne R, Brown TAW (1997b) Exports
of organic carbon in British rivers. Hydrol Process 11:
325– 344
Hope D, Palmer SM, Billett MF, Dawson JJC (2001) Carbon
dioxide and methane oxidation evasion from a temperate
peatland stream. Limnol Oceanogr 46:847– 857
Janssens IA, Freibauer A, Schlamadinger B, Ceulemans R
and others (2005) The carbon budget of terrestrial eco-
systems at country scale: a European case study. Biogeo-
sciences 2:15–26
Johnson MS, Billett MF, Dinsmore KJ, Wallin M, Dyson K
(2010) Direct and continuous measurement of dissolved
carbon dioxide in freshwater aquatic systems— method
and applications. Ecohydrology 3:68–78
Jonsson A, Algesten G, Bergstrom AK, Bishop K, Sobek S,
Tranvik LJ, Jansson M (2007) Integrating aquatic carbon
fluxes in a boreal catchment carbon budget. J Hydrol 334:
141–150
Koehler AK, Murphy K, Kiely G, Sottocornola M (2009) Sea-
sonal variation of DOC concentration and annual loss of
DOC from an Atlantic blanket bog in south western
Ireland. Biogeochemistry 95:231–242
Littlewood IG (1992) Estimating contaminant loads in rivers:
a review. Institute of Hydrology, Wallingford
Livingstone GP, Hutchinson GL (1995) Enclosure-based mea-
surement of trace gas exchange: applications and sources
of error. In: Matson PA, Harriss RC (eds) Biogenic trace
gases: measuring emissions from soil and water. Blackwell
Science, Cambridge, MA, p 14– 51
Mäkila M (1997) Holocene lateral expansion, peat growth
and carbon accumulation on Haukkasuo, a raised bog in
southeastern Finland. Boreas 26:1–14
Mäkila M, Moisanen M (2007) Holocene lateral expansion
and carbon accumulation of Luovuoma, a northern fen
in Finnish Lapland. Boreas 36:198–210
McNamara NP, Plant T, Oakley S, Ostle NJ (2008) Gully
hotspot contribution to landscape methane and carbon
dioxide fluxes in a northern peatland. Sci Total Environ
404:354– 360
Milne R, Brown TAW (1997) Carbon in the vegetation and
soils of Great Britain. J Environ Manag 49:413–433
Monteith DT, Stoddard JL, Evans CD, de Wit HA and others
(2007) Dissolved organic carbon trends resulting from
changes in atmospheric deposition chemistry. Nature 450:
537–540
Neal C (1988) Determination of dissolved CO2in upland
stream water. J Hydrol 99:127–142
Nilsson M, Sagerfors J, Buffam I, Laudon H and others (2008)
Contemporary carbon accumulation in a boreal oligo-
trophic minerogenic mire: a significant sink after account-
ing for all C-fluxes. Global Change Biol 14:2317–2332
doi:10.1111/j.1365-2486.2008.01654.x
O’Brien HE, Labadz JC, Butcher DP, Billett MF, Midgley NG
(2008) Impact of catchment management upon dissolved
organic carbon and stream flows in the Peak District,
Derbyshire, UK. In: Sustainable Hydrology for the 21st
Century, Proc 10th BHS Nat Hydrol Symp, Exeter, 15–17
Sep 2008. British Hydrological Society, London, p 178–185
Oldfield F, Richardson N, Appleby PG (1995) Radiometric
dating (210Pb, 137Cs, 241Am) of recent ombrotrophic
peat accumulation and evidence for changes in mass-
balance. Holocene 5:141–148
Palmer SM, Hope D, Billett MF, Dawson JJC, Bryant CL
(2001) Sources of organic and inorganic carbon in a
headwater stream: evidence from carbon isotope studies.
Biogeochemistry 52:321– 338
Pawson RR (2008) Assessing the role of particulates in the
fluvial organic carbon flux from eroding peatland systems.
PhD thesis, University of Manchester, Manchester
Pawson RR, Lord DR, Evans MG, Allott TEH (2008) Fluvial
organic carbon flux from an eroding peatland catchment,
southern Pennines, UK. Hydrol Earth Syst Sci 12:625 –634
Rose NL, Appleby PG (2005) Regional applications of lake
sediment dating by spheroidal carbonaceous particle
analysis. I: United Kingdom. J Paleolimnol 34:349– 361
Rose NL, Harlock S, Appleby PG, Battarbee RW (1995) Dating
of recent lake sediments in the United Kingdom and
Ireland using spheroidal carbonaceous particle (SCP) con-
centration profiles. Holocene 5:328–335
Rothwell JJ (2006) Fluvial export of heavy metals from conta-
minated and eroding peatlands, southern Pennines, UK.
PhD thesis, University of Manchester, Manchester
Rothwell JJ, Evans MG, Liddaman LC, Allott TEH (2007) The
role of wildfire and gully erosion in particulate lead export
from contaminated peatland catchments in the southern
Pennines. Geomorphology 88:276–284
Roulet N, Lafleur PM, Richard PJH, Moore TR, Humphreys
ER, Bubier J (2007) Contemporary carbon balance and
late Holocene carbon accumulation in a northern peat-
land. Global Change Biol 13:397–411 doi:10.1111/j.1365-
2486.2006.01292.x
Simmons IG (2003) The moorlands of England and Wales: an
environmental history 8000 BC to AD 2000. Edinburgh
University Press, Edinburgh
Skiba U, Cresser MS, Derwent RG, Futty DW (1989) Peatland
acifidification in Scotland. Nature 337:68–69
Smith P, Smith J, Flynn H, Killham K and others (2007)
ECOSSE: estimating carbon in organic soils— sequestra-
tion and emissions. Scottish Executive Environment and
Rural Affairs Department, Edinburgh, www.scotland.gov.
uk/Publications/2007/03/16170508/0
28
➤
➤
➤
➤
➤
➤➤
➤
➤
➤
➤
➤
➤
➤
➤➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
Billett et al.: Carbon balance of UK peatlands
Sykes JM, Lane AMJ (1996) The United Kingdom Environ-
mental Change Network: protocols for standard measure-
ment at terrestrial sites. Her Majesty’s Stationery Office,
London
Walling DE, Webb BW (1985) Estimating the discharge of
contaminants to coastal waters by rivers: some cautionary
comments. Mar Pollut Bull 16:488– 492
Ward SE, Bardgett RD, McNamara NP, Adamson JK, Ostle NJ
(2007) Long-term consequences of grazing and burning
on northern peatland carbon dynamics. Ecosystems 10:
1069–1083
Wieder RK, Vitt DH (2006) Boreal peatland ecosystems.
Springer, Berlin
Worrall F, Reed M, Warburton J, Burt TP (2003) Carbon bud-
get for a British upland peat catchment. Sci Total Environ
312:133–146
Worrall F, Burt TP, Rowson JG, Warburton J, Adamson JK
(2009) The multi-annual carbon budget of a peat-covered
catchment. Sci Total Environ 407:4084– 4094
Yang H, Rose NL, Boyle JF, Battarbee RW (2001) Storage and
distribution of trace metals and spheroidal carbonaceous
particles (SCPs) from atmospheric deposition in the catch-
ment peats of Lochnagar, Scotland. Environ Pollut 115:
231–238
29
Submitted: November 2, 2009; Accepted: May July 5, 2010 Proofs received from author(s): December 17, 2010
➤
➤➤
➤➤
