11a2007

Greenhouse Impacts
of the Use of Peat and
Peatlands in Finland





1

Ministry of Agriculture and Forestry 11a/2007

Greenhouse Impacts of the Use of
Peat and Peatlands in Finland

Research Programme Final Report



2

Title of publication:
Greenhouse Impacts of the Use of Peat and Peatlands in Finland
Research Programme Final Report

Publisher:
Ministry of Agriculture and Forestry

Editor:
Sakari Sarkkola / Metla

Photos:
Sakari Sarkkola, Jyrki Hytönen, Sanna Saarnio, Mikulas Cernota, Mika Yli-Petäys

Photos on the cover:
Sakari Sarkkola

Translation of the original Finnish version:
Jaakko Mäntyjärvi and Graham Whitfield, the English Centre Helsinki Oy

ISBN 978-952-453-394-2
ISSN 1238-2531

Layout:
Paavo Ojanen / University of Helsinki

Printers:
Vammalan Kirjapaino 2008



3

Foreword
Peat is a domestic fuel of national importance to Finland and, considering its employment impact, it also has
regional policy significance. On the basis of the National Climate Strategy (VNS 1/2001 vp), the Ministry of Trade
and Industry commissioned in January 2001 a survey of the needs for further research into the life cycle analysis of
peat. The purpose of the survey was to estimate what information will be needed to scientifically motivate, if
possible, the introduction of calculation methods which better take into account the life cycle of peat in calculating
emissions from the use of peat under the principles of the Kyoto Protocol. Another motivating factor was Finland’s
obligation to report on emissions from peat and peatlands under the United Nations Framework Convention on
Climate Change (UNFCCC), the aim being to further specify these values.

As a result of the survey, an extensive research programme entitled Greenhouse Impacts of the Use of Peat and
Peatlands in Finland was launched. Since the four-year programme, by its nature, required extensive resources for
conducting measurements in the field, it was jointly funded and steered by the Ministry of Trade and Industry, the
Ministry of Agriculture and Forestry and the Ministry of the Environment. The programme consisted of several
research projects whose purpose was to establish the greenhouse gas (GHG) balances of peatlands in various
types of land use. The practical work involved was carried out by research teams at three universities and four
research institutions: the Universities of Helsinki, Joensuu and Kuopio, the Finnish Forest Research Institute
(Metla), the Finnish Meteorological Institute, the Geological Survey of Finland and VTT Technical Research Centre
of Finland.

The results of the programme have already contributed to a significant specification of the emission factors of the
greenhouse impacts of peatland land use for the national GHG inventory, and life cycle analyses have been
employed to establish peat utilization models that would minimize the greenhouse impact of peat use.

It was partly due to the findings of this programme that the classification of peat in the 2006 IPCC Guidelines for
National Greenhouse Gas Inventories was changed to recognize peat as a class of its own between fossil fuels and
biomass. In the greenhouse gas inventory, however, the emission calculations in the energy sector are only based
on emissions generated through combustion. Life cycle emissions analyses cannot be applied in the reporting of
greenhouse gas emissions under the Kyoto Protocol.

Parkano, 28 September 2007

Jukka Laine
Professor
Research Programme Coordinator



4

Contents

Summary
Background and needs for information
Research programme: Greenhouse Impacts of the Use of
    Peat and Peatlands in Finland 2002–2005
Use of peatlands and greenhouse gas balances
Greenhouse impacts of using peat for energy,
    from the life cycle perspective
Conclusions

1. Background and aims of the research programme

2. Implementing the research programme

3. Use of peat and peatlands as a source of
greenhouse gas emissions
3.1. Pristine peatlands as greenhouse gas
    sinks and sources

3.1.1. Background
3.1.2. Aim of the study
3.1.3. Research methods and principal results
3.1.4. Conclusions

3.2. Greenhouse gas emissions from forestry-drained
     peatlands and contributing environmental factors

3.2.1. Background
3.2.2. Research methods and aims
3.2.3. Results
3.2.4. Conclusions

3.3. Greenhouse gas emissions from cultivated
    and abandoned organic croplands

3.3.1. Background and aim
3.3.2. Materials and methods
3.3.3. Results
3.3.4. Conclusions

3.4. The effect of afforestation of organic croplands and
    cutaway peatlands on greenhouse gas balance

3.4.1. Background
3.4.2. Material and methods
3.4.3. Results
3.4.4. Discussion
3.4.5. Conclusions

6
6
6

7
11

13

14

16

17

17

17
17
17
21
22

22
22
23
26
27

27
28
29
30
31

31
31
33
35
37



5

3.5. Ecosystem-level carbon sink measurements on
    forested peatlands

3.5.1. Background and aim
3.5.2. Material and methods
3.5.3. Results
3.5.4. Conclusions

3.6. Garbon gas exchange of re-vegetated cutaway
peatland
3.6.1. Background and aim
3.6.2. Material
3.6.3. Results
3.6.4. Conclusions

4. Emission factors and their uncertainty in Finnish
    managed peatlands, and need for further research

4.1. Background
4.2. Material and methods
4.3. Factors affecting emission factors in different
    types of peatland
4.4. Gas emissions from peat harvesting and after-use
    of peat harvesting areas
4.5. Outlook
4.6. Conclusions

5. Greenhouse impact due to different peat fuel utilization
    chains in Finland – a life cycle approach

5.1. Background and aim
5.2. Assessment of greenhouse impact for
    the peat fuel life cycle
5.3. Peat utilization chains examined
5.4. Input data for calculations
5.5. Results
5.6. Sensitivity analysis
5.7. Discussion
5.8. Conclusions

6. Final conclusions of the research programme

Scientific articles published under the reasearch programme
Background literature

APPENDIX
Greenhouse Impact of the Use of Peat and Peatland
    in Finland – Research programme component
    projects and their researchers

38

38
38
39
40
41

41
41
42
46

47

47
47
48

50

51
52

53

53
53

54
56
57
60
62
63

64

65
65

68



6

Summary

Background and needs for information

Pristine peatlands are carbon accumulating ecosystems. Since the last Ice Age, the Finnish

peatlands are estimated to have accumulated some 5.4 billion tonnes of carbon, forming the

largest soil carbon stock. From the original 10 million peatland hectares ~5.7 million ha have

been drained for forestry and ~0.7 million ha for agriculture. The area used for harvesting of

energy and environmental peat is ~60,000 ha. Circa 40% of the original peatland area still re-

mains in a natural state.

Peat combustion covers 5 to 6 per cent of Finland’s total energy requirements. Peat is a nationally

important fuel for Finland; with the employment aspects, it also has regional political importance. In the

international statistics (OECD/IEA/Eurostat) peat is paralleled with fossil fuels. In Finland peat is clas-

sified as slowly renewing biomass fuel, the time required for the peat layer to rebuild being very long.

In the National Climate Strategy published in 2001 (VNS 1/2001 vp), it was noted that “peat is left for

the national authorities to decide upon and thus outside the scope of the EC Directives on energy

taxation. In international statistical practice, it is supported that peat would be separated from

the fossil fuel category to form a category of its own.” It was also noted: “As far as the [United

Nations Framework] Convention on Climate Change and other international cooperation are concerned,

the following action will be taken:

The need for supplementary studies will be mapped out in a survey and a research programme will

be launched based on this survey for a life cycle analysis of the energy use of peat.

If it is justified on the basis of the research findings, measures will be undertaken to influence the

rules and definitions of the methodology used for calculating greenhouse gases by virtue of the

international Convention on Climate Change. In this, the aim will be that the calculation methods

subject to the Convention would take the greenhouse gas balance of peat into account during the

entire life cycle and not just the emissions from combustion.

To be able to influence the Convention on Climate Change, clear criteria, besides new research

data, will be required for the energy use of peat. The criteria shall state e.g. the definition of those

peatlands on which the production of energy peat will be directed, as well as the requirements for

subsequent use of the peat production areas.”

Related to the preparation of the Climate Strategy, the Ministry of Trade and Industry commissioned in

January 2001 a survey of the need for further research into the life cycle analysis on peat. The purpose

of the survey was to estimate what information will be needed to scientifically motivate, if possible, the

introduction of calculation methods better taking into account the life cycle of peat in calculating emis-

sions from the use of peat under the principles of the Kyoto Protocol.

Another motivating factor was Finland’s obligation to report on emissions from peat and peatlands

under the United Nations Framework Convention on Climate Change (UNFCCC), the aim being to

further specify these values.

As a result of the survey (Minkkinen & Laine 2001), a four-year research programme (2002–2005) was

set up, jointly funded by the Ministry of Trade and Industry, the Ministry of Agriculture and Forestry and

the Ministry of the Environment. Its principal purpose was to assess the levels of greenhouse gas

emissions from the use of peat and peatlands in Finland.

Research programme: Greenhouse Impacts of the Use of Peat and Peatlands in

Finland 2002–2005

The programme consisted of nine separate research projects under a coordination project; some were

‘method projects’ and others involved a specific sector of land use.

•

•

•



7

The purpose of the programme projects was to draw up models (dynamic emission factors) for the

greenhouse gas (GHG) balances of peatlands subject to different kinds of land use and to quantify the

underlying ecosystem processes. The scientific findings of these projects have been reported in a

special issue of the journal Boreal Environment Research, which can be found at:

http://www.borenv.net.

The emission factors developed in the programme have already been used in Finland’s new national

GHG emission calculations, the annual inventory reports submitted under the United Nations Frame-

work Convention on Climate Change (UNFCCC), and the Kyoto Test Report in 2007.

The newest information produced by the programme was also used as source material for the

2006 IPCC Guidelines for National Greenhouse Gas Inventories, adopted at the IPCC session

on Mauritius in April 2006. It was partly due to the findings of this programme that the classifi-

cation of peat in the IPCC 2006 guidelines in the emission source class ‘energy’ was changed in

order that peat is now classified as a class of its own between fossil energy sources and bio-

mass. In this reporting, however, the emission calculations are only based on emissions generated

through combustion. Life cycle emissions analyses cannot be used for reporting under the Kyoto Pro-

tocol.

The emission factors calculated for various classes of land use are based on measurements conduct-

ed in different parts of Finland over several years.

Use of peatlands and greenhouse gas balances

Out of Finland’s original 10 million hectares of peatlands circa 5.4 million ha has been drained for

forestry and circa 0.7 million ha for agriculture. This leaves some 4 million hectares in a pristine state.

The largest part of the area used for fuel peat production has already been drained for forestry, but also

pristine peatlands have been used for peat production. By contrast, peatlands formerly drained for

agriculture are rarely used for harvesting peat because of the problems involved in the traditional

production method. From the perspective of climate impact, it naturally makes sense to locate

peat production in areas where anthropogenic greenhouse gas emissions are high, so that in

the overall greenhouse gas balance the cessation of those emissions because of fuel peat

harvesting will compensate for some of the greenhouse impact created by the combustion of

that peat. Similarly, further use of areas no longer used for peat production must be planned so

that at the life cycle level the greenhouse impact of the use of those areas is as low as possible.

Greenhouse gas balances of pristine peatlands as a background to the impact of the

use of peatlands and peat

This research programme focused on the carbon balance of pristine, i.e. undrained, peatlands by

measuring carbon dioxide (CO
2
) and methane (CH

4
) exchange at the ecosystem-atmosphere bounda-

ry of both raised bogs and sedge fens. The balance for these two carbon-containing gases constitutes

the majority of the carbon balance of a peatland. In photosynthesis, CO
2
 is bound into plant biomass.

Decomposition releases most of the bound CO
2
 back into the atmosphere. Anaerobic decomposition

also generates CH
4
, some of which is emitted into the atmosphere. CH

4
 has a GWP factor (Global

Warming Potential, IPCC) about 20 times higher than that of CO
2
 when emissions are considered over

a 100-year period.

The gas exchange and the resulting annual carbon balance are sensitive to varying weather

conditions. The annual carbon dioxide balance of a sedge fen can vary from a net release of

over 1,000 kg per hectare to nearly an equal net sink. Similarly, on raised bogs the annual CO
2

balance was found to fluctuate between an emission of 850 kg/ha per year to an accumulation of 670

kg/ha per year. CH
4
 emissions vary between <15 and 530 kg/ha per year on sedge fens (minerotrophic

peatlands) and between <15 and 200 kg/ha per year on raised bogs (ombrotrophic peatlands). Wet-

ness slows down decomposition so much that over a long period of time some of the dead plant

biomass turns into peat. On average, Finland’s sedge fens and raised bogs have accumulated 170 kg/

ha and 210 kg/ha of carbon per year, respectively, since the last Ice Age. The increased occurrence of



8

Table 1. An example of the composition of the soil carbon dioxide balance in forestry-drained peatlands with

different nutrient status. Negative values indicate net release from the ecosystem into the atmosphere.

Litter production Decomposition Soil

g C m–2 a–1 of peat and litter C balance

Intermediate peatland 500 –572 –72

[1,834 g CO
2
] [2,099 g CO

2
] [264 g CO

2
]

Nutrient-poor peatland 496 –491 +5

[1,818 g CO
2
] [1,799 g CO

2
] [+18 g CO

2
]

The conversion factor C => CO
2
 is 3.667

summer droughts causes water level lowering also in peatlands. In these conditions the organic matter

that has been stored during the previous decades is exposed for aerated conditions above the water

level and the rate of oxidation increases, causing an annual net loss of peat. Many observations in

drought situations support such losses of peat in both fens and bogs.

Forestry is the most important form of land use for peatlands

Previous research results show that the soil of forestry-drained peatlands is a significant car-

bon sink, but the results from modelling simulations carried out in this research programme

indicate the opposite. According to these results, the soil carbon balance of Finnish forestry-

drained peatlands is, on average, negative. Instead, the carbon uptake of the growing tree stand

may exceed the soil emissions. The simulations so far contain high uncertainty and the recently

initiated supplementary studies will confirm the results in the coming years.

In the forestry-drained areas the aerobic decomposition of the old organic matter of the soil

(peat and humus) to carbon dioxide constitutes the largest release/removal in the carbon bal-

ance of peatland forests. Based on the estimation, this carbon dioxide (CO
2
) emission is annually

6,050–16,900 kg of carbon per hectare, depending on the nutrient status of the peatland. The emission

is the largest in nutrient-rich sites and smallest in the nutrient-poor drainage areas. Most of our peat-

land forests grow in the nutrient-poor and intermediate peatland site types. The simulated average

annual emission from these site types is 7,700–10,600 kg/ha of carbon. In addition, carbon is released

in leaching as dissolved organic carbon, which counts annually some 60 to 100 kg/ha, and in the

possible forest/peat fires. Part of the released carbon is replaced by the litter production of the tree

stand and forest vegetation, which feeds new organic matter above the ground and into the soil (peat).

For this reason the total carbon balance of a certain site may be positive or negative (Table 1).

When water level is high, methane (CH
4
) is produced in drained sites, as well; however, the measured

annual methane emissions in sparsely forested nutrient-poor sites are, at most, less than 30 kg of

methane per hectare. The effectively drained areas are commonly weak methane sinks, where the

annual methane consumption from the atmosphere remains less than 7.5 kg per hectare. Significant

amount of nitrous oxide (N
2
O) is released only from the nutrient-rich or nitrogen fertilized peat-

land forests. Nitrous oxide is released in small amounts (on average ~2 kg per hectare per year),

however, due to its large greenhouse impact (~300 times CO
2
) this emission source may be

significant for the total emission of drained peatlands. The ongoing supplementary studies will

improve the estimate in the near future.

More reliable emission factors for peat harvesting areas through new measurements

in the present study

Most of the gas emissions measured were of the same magnitude as the few earlier findings, but there

were some surprises, too. In especially warm and humid conditions, a milled peat field may emit up to

five times the normal amount of carbon dioxide (CO
2
) (40,300 kg/ha per year as opposed to the aver-

age of 9,400 kg/ha per year). The storage of harvested peat in the stockpiles may double the carbon



9

Table 2. Annual greenhouse gas (GHG) emissions from organic cropland. The minimum and maximum values

represent actual observed values for each cultivation type. Negative values indicate a net influx of the substance

from the atmosphere into the ecosystem.

Average

GHG cropland Grass Cereal Fallow Abandoned

CO
2
, g m–2 a–1

Average 2,072 1,485 1,760 2,971 1,188

Min–Max 290–4,033 2,167–4,033 –330–3,300

CH
4
, g m–2 a–1

Average 0.42 1.27 –0.43 0.41 –0.22

Min–Max –0.49–0.91 0.11–0.91 –0.49–0.51 –0.35–4.00

N
2
O, g m–2 a–1

Average 1.74 0.85 1.74 2.63 1.29

Min–Max 0.17–5.81 0.17–1.56 0.85–3.79 0.60–5.81

dioxide emissions of milled peat fields, if the annual area of the stockpiles is assumed to be 10% of the

peat harvesting area. The peat harvesting fields release methane at moderate rates (3–90 kg/ha per

year) and nitrous oxide at weak rates (2–5 kg/ha per year).

New research confirms and specifies data on significant carbon dioxide and nitrous

oxide emissions from peatlands used in agriculture

The emission ranges shown below come from new, measurement-based studies. The nitrous oxide

(N
2
O) emissions are due not only to peat being naturally rich in carbon and nitrogen but also to the

nitrogen content of fertilizers. Although the emission levels are sensitive to weather conditions, only a

small portion of the emissions can actually be predicted on the basis of the weather. An uncertainty factor

must be added to any emission estimates, and this broadens the range of variation considerably, partic-

ularly for N
2
O (Table 2).

CO
2
 emissions from organic cropland, whether ploughed land or covered by vegetation, are roughly the

same, being 25,300 to 40,300 kg/ha per year on ploughed land, 2900 to 27500 kg/ha per year in grass

cultivation and 7700 to 30400 kg/ha per year on a grain field. N
2
O emissions from organic cropland that

is ploughed are 6 to 58 kg/ha per year, typically higher than for organic cropland covered by vegetation

(2 to 37 kg/ha per year in measurements in recent years). Recent studies show that anything from 25%

up to 60% of all N
2
O emissions occur in the winter. Organic cropland is a weak methane (CH

4
) sink,

because the water table is quite far from the surface. When the peat gets wet and its oxygen content

decreases, organic cropland only generates slight emissions. On organic cropland where cultivation has

been abandoned, gas emissions would seem to stay at approximately the same level for decades after-

wards.

Afforestation of organic cropland and cutaway peatland has been estimated to reduce

greenhouse impact, and new findings seem to support this projection

The findings do show, however, that emissions from the soil do not decrease so much that the

amount of carbon bound by the trees at the site would be enough to make the overall carbon

balance positive. Decomposition of peat and old litter caused annual emissions of carbon dioxide (CO
2
)

of between 10,100 and 17,600 kg per hectare per year in afforested cutaway peatland and between

7,600 and 19,800 kg per hectare per year in afforested organic cropland (Table 3), which is similar to or

greater than emissions in forestry-drained peatland but clearly less than emissions in organic cropland

that is not afforested. When the CO
2
 absorbed annually by trees is subtracted from the CO

2
 emissions

from the soil, assuming the CO
2
 absorption due to annual growth to be between 1,650 and 12,100 kg/ha

per year, the CO
2
 balance for the entire ecosystem probably remains negative in the majority of cases.

These figures do not include the carbon bound in the surface vegetation and accumulating litter, but



10

to anaerobic decomposition. The most important factor for the success of restoration is sufficient mois-

ture.

CH
4
 emissions follow the binding of new organic matter with a delay. At the beginning of the restoration

process, for several years, CH
4
 emissions from the former cutaway peatland can be lower than those

from pristine sedge fens. However, an area accumulating new plant biomass exceptionally fast can

under suitable circumstances generate CH
4
 emissions up to over 400 kg/ha per year. Over time, car-

bon dioxide (CO
2
) sequestration may slow down, and the CH

4
 release processes become stable so

that the carbon gas fluxes in restored peatlands come to match those of pristine peatlands.

The findings concerning the gas balances in pristine peatlands on the land uplift coast in Siikajoki (on

the coast of the Gulf of Bothnia) support the conclusions drawn regarding restoration sites. Studies of

wetlands of different ages in Siikajoki, from shore swamps 100 years old to Sphagnum bogs 2,500

years old, show that photosynthesis was at its most efficient in the younger areas and slowed down as

the age of the peatland increased. CH
4
 dynamics were also the most unstable in younger areas, where-

as the oldest peatlands were steady sources of CH
4.
 The paludification process in pristine peatlands

was much slower than in human-regulated restored cutaway peatlands, where the occasional dry spell

did not hinder development. After the re-establishment of stabilised peatland vegetation, restored

cutaway peatland functions just like a pristine peatland. Because there is a strong correlation

between carbon absorption and peatland vegetation, vegetation could be used as a simple indicator for

evaluating the carbon balance and CH
4
 emissions of restored cutaway peatland. Unlike pristine peat-

lands, restored cutaway peatland forms part of the land use categories to be reported in greenhouse

gas inventories, even though the purpose of land use in these areas is to remove the anthropogenic

impact and to return them to their natural state. Indeed, there is a case for considering restoration

as a temporary form of land use for returning an area to the state in which it was before human

intervention, which means that a restored peatland could be excluded from the greenhouse

gas inventory after a specified period of time, just like pristine, undrained peatlands.

Table 3. Annual greenhouse gas (GHG) emissions due

to decomposition of organic matter in the soil in affor-

ested organic cropland and cutaway peatland. Nega-

tive figures indicate a net influx of the substance in ques-

tion from the atmosphere into the ecosystem. NB: In

order to calculate the carbon dioxide (CO
2
) balance of

the soil and the ecosystem as a whole, the carbon bound

in the trees and litter in the area must be subtracted

from the emissions. Tree species and site type affect

the amount of uptake.

Afforested Afforested

GHG organic cropland cutaway peatland

CO
2
, g m–2 a–1

Average 1,354 1,397

Min–Max 759–1,976 1,008–1,756

CH
4
, g m–2 a–1

Average –0.15 –0.05

Min–Max –0.43–0.81 –0.03–0.09

N
2
O, g m–2 a–1

Average 1.02 0.15

Min–Max 0.16–4.71 0.02–0.75

micro-meteorological measurements on an affor-

ested cropland confirm the estimate that afforest-

ed cropland remains a small source of CO
2
 (about

500 kg/ha per year). Afforestation does slow down

CO
2
 emissions for a few decades, i.e. for the peri-

od during which the trees and underground bio-

mass are growing. However, both afforested

cutaway peatland and afforested organic cro-

pland would seem to continue emitting nitrous

oxide (N
2
O). Even when afforested, organic cro-

pland was found to emit more N
2
O (2 to 47 kg/ha

per year) than cutaway peatland (0.2 to 7.5 kg/ha

per year).

Restoration of cutaway peatland as an

after-use of peat harvesting binds carbon

dioxide into long-term stock

Restoration also causes methane (CH
4
) emis-

sions to recur as the peatland evolves. Restored

areas begin to show a net influx of carbon within a

few years of the return of peatland vegetation.

Under favourable conditions, carbon may be ab-

sorbed very quickly, particularly because photo-

synthesis is efficient and little CH
4
 is emitted due



11

Greenhouse impacts of using peat for energy, from the life cycle perspective

Peat is a fuel of national importance in Finland, but using peat for energy causes greenhouse

gas emissions. These emissions have grown with the increasing use of fuel peat. All Parties are

required to make an inventory of greenhouse gas emissions and to report on it to the UNFCCC. Fin-

land’s greenhouse gas inventory and emissions trading equate peat with fossil fuels for the purpose of

calculating emissions, in accordance with the IPCC Guidelines, although under the 2006 Guidelines

peat is to be reported in a class of its own, separate from fossil fuels. With the progress of emissions

trading, peat production is expected to decrease, especially in the generation of condensing electricity.

The purpose of the greenhouse gas inventory under the UNFCCC is to present anthropogenic

greenhouse gas emissions and sinks during the report year as accurately as possible. This

enables monitoring of actual trends in greenhouse gases and assessment of meeting the commit-

ments under the Kyoto Protocol, among other things.

Life cycle assessment of greenhouse impacts differs from the greenhouse gas inventory in that

it takes into account all the significant emissions and sinks caused by the product in question.

For an ordinary product, emissions are generated within a relatively short period of time. In the case of

peat fuel, however, the time dimension introduced by the after-use of the cutaway peatland (e.g. resto-

ration, afforestation or cultivation) may be defined in anything up to centuries in terms of emissions and

sinks. Because life cycle assessment takes into account emissions and sinks that may exist in the

remote future, its results are not compatible with the inventory approach, which only takes into account

the actual emissions and sinks of the report year. In addition, in the greenhouse gas inventory the

emissions are reported by land use sectors and emission classes. Here the emissions during the life

cycle of one function, such as the energy use of peat, are placed in several emission classes (for

example, combustion, harvesting machines and subsequent use of the peat harvesting area).

This present research programme considered the greenhouse impacts of the use of peat for energy

from the life cycle assessment perspective. The greenhouse impact is estimated using radiative forc-

ing, which describes the perturbation to the Earth’s radiative energy balance caused by greenhouse

gases and leading to climate change.

The peat-energy production chain consists of fuel peat harvesting in various production areas, gener-

ating energy by peat combustion, and after-use of the production areas. The studies considered vari-

ous types of peat harvesting area: pristine peatland (fens), forestry-drained peatland and organic cro-

pland. After the peat harvesting, the subsequent land use forms of the cutaway areas were peatland

restoration, afforestation and cultivation of reed canary grass (Phalaris arundinacea). Several different

peat fuel production chains were created using these options of initial states and after-use. In this study,

the use of peatland was examined from two perspectives: the production chain was restricted to peat

only, or secondly, renewable energy (wood biomass, reed canary grass) grown in the harvesting area

was also considered.

In the research project, the greenhouse impacts of the various fuel peat production chains were simu-

lated over a period of 300 years. The production chain is delimited to apply to peat energy only. In order

to limit the increase of the average temperature of the Earth to two or three degrees (the European

Union has proposed 2 °C), the world’s greenhouse gas emissions must be reduced significantly in the

course of the present century. Therefore, a time horizon of 100 years may be considered significant in

assessing the greenhouse impacts of various fuels. The greenhouse impact of using a pristine

sedge fen for fuel peat production is similar or even greater than that of coal (chains 1–2). This

is due, among other things, to the ceasing of carbon sequestration when the peatland is prepared for

harvesting. A uniform method is used for assessing the greenhouse impacts of coal and peat, and the

results are comparable. Using forestry-drained peatland for fuel peat production (chain 3) also

causes a slightly greater greenhouse impact than coal, unless the peat is harvested with suffi-

cient precision. Harvesting residual peat as precisely as possible to recover its energy potential

brings the greenhouse impact of this chain to the same level with that of coal.



12

With modern technology, the most climate-friendly peat energy production chain is the one

using peatland which is or has been in agricultural use (organic cropland) and which is affor-

ested once peat harvesting is finished (chain 4). Taking organic cropland over for peat harvesting

discontinues the major emissions associated with its agricultural use, making this chain beneficial for

the climate.

Using high-emission peatlands (such as organic cropland) for peat harvesting and

employing new technology can keep the greenhouse impact of the energy use of

peat somewhat lower than that of coal when assessed over a 100-year time horizon.

In this study also the so called ‘vision chains’ were examined, to show the lowest possible level of

greenhouse impact from peat production. This can be achieved by minimizing emissions in different

steps of the chain by using modern technology. In vision chain A peat is harvested from forestry-drained

peatland. In vision chain B peat harvesting is located in areas that are major sources of greenhouse

gas emissions prior to this harvesting (organic cropland). The modern technology employed includes

improved combustion especially with regard to nitrous oxide (N
2
O) emissions and new peat harvesting

technology aiming at shorter production times and lower emissions from the peat field and stockpiles.

In vision chain A, where the harvesting area is forestry-drained peatland, the greenhouse impact is

lower than that of coal. In vision chain B, the greenhouse impact begins to decline after only 100 years

and almost achieves neutrality in 300 years.

Using cutaway peatland for producing renewable bioenergy and partial mixed fuel

combustion can significantly reduce greenhouse impact

Using peatland for both peat harvesting and producing renewable bioenergy has also been studied.

Included in the study were the production chains where the peat harvesting area is used not only for

producing fuel peat but also for producing renewable energy in the long term (reed canary grass or

wood biomass) after peat harvesting is ended. The production areas studied are cultivated peatland

(organic cropland) and forestry-drained peatland. The study time horizons are 100 and 300 years from

the start of the production. Renewable energy production brings down the relative impact of the use of

peat for energy. In the long term, especially using peat from organic cropland together with biomass

decreases the greenhouse impact to a fraction of that of coal. New technology in fuel peat production

further decreases the greenhouse impact.

The greenhouse impact of peat has been studied in Sweden as well. A review of earlier Finnish and

Swedish studies showed that their findings were largely in agreement but that there were also differ-

ences. Because of this, a study was conducted on the similarities and differences between the most

recent Finnish and Swedish research. The greenhouse impact of peat has not been studied in any

other country, and for this reason it is important to find out what causes the differences in the assess-

ments and baseline values. A comparative study showed that the scientific approach and calculation

method are very similar in both the Finnish and Swedish studies. The main difference was in the input

values, particularly in the case of emissions from forestry-drained peatlands.



13

-

-

•

•

•

•

•

•

•

•

Conclusions

The research programme contributed substantially to the knowledge about the impact of land use

on the greenhouse gas balance of Finland’s peatlands, and in some respects existing concep-

tions about greenhouse gas emissions were significantly revised.

The carbon dioxide balance of forestry-drained peatland used to be considered positive, but

the new findings show that on average it is negative (forestry drained peatland looses carbon).

By contrast, existing data on substantial emissions of carbon dioxide and nitrous oxide from

current or former organic cropland were confirmed by the new findings.

It was somewhat surprising to note that afforestation of organic cropland is not enough to

render the overall greenhouse gas balance positive, although it does reduce emissions.

The purpose of the greenhouse gas inventory under the UNFCCC is to report as accurately as

possible the actual anthropogenic greenhouse gas emissions and sinks during the report year.

This enables, inter alia, monitoring of true development of greenhouse gases and assessment of

meeting the commitments under the Kyoto Protocol.

Life cycle assessment of greenhouse impacts differs from the greenhouse gas inventory in that it

takes into account all the significant emissions and sinks caused by the product in question. The

present research programme concerned the greenhouse impacts of the use of peat for energy

from the life cycle assessment perspective.

With present methods, the use of peat for energy causes a greenhouse impact of similar magni-

tude as the use of coal. The results, however, include uncertainty partly due to the long integration

periods considered, and partly due to the unknown distribution of the initial state emissions of the

former forestry-drained peatlands which currently are under peat production. Taking into account

the use of peatland for renewable bioenergy production after peat harvesting is finished, the

greenhouse impact of the overall energy use of peatland is less than that of coal.

The greenhouse impact of peat can further be significantly reduced by directing peat harvesting

to current or former organic cropland and to those forestry-drained areas which have high emis-

sions in their current state. In these cases, the greenhouse impact decreases significantly in the

long term.

The greenhouse impact of peat energy can be decreased by thorough utilization of residual peat,

improvement of combustion techniques, and with new peat harvesting methods. The production

of renewable bioenergy in the areas available after peat harvesting will decrease the greenhouse

impact per total produced energy unit. Afforestation is slightly more climate-friendly as an after-

use measure for cutaway peatland than restoration. Cultivation of reed canary grass has roughly

the same effect as afforestation with regard to the greenhouse impact.

Peatland restoration aims at removing human impact and restoring the natural condition. Indeed,

there is a case for considering restoration as a temporary form of land use for returning an area to

the state in which it was before human intervention. Accordingly, restored peatlands could be

removed from greenhouse gas inventories, corresponding to pristine peatlands, after a certain

time period.

Knowledge of the greenhouse impact of land use in peatlands is still fragmented and largely

deficient. More research is needed particularly on the net exchange of carbon dioxide (soil carbon

balance) of forested peatlands in different peatland site types from the southern and northern

parts of Finland and on nitrous oxide balances on nutrient-rich peatlands, especially organic cro-

plands. The knowledge on greenhouse gas balances of cut-away peatlands after peat harvesting

and from the different after-use options is still meagre, since there are only few such areas at

present. The proportion of these areas will increase considerably in the future, however, and

more research should be conducted.

-



14

Pristine peatlands are ecosystems which act as

carbon sinks. It is estimated that Finland’s peat-

lands have absorbed some 5.7 billion tonnes of

carbon since the last Ice Age, constituting the larg-

est natural carbon stock in Finnish soil. Finland

originally had nearly 10 million hectares of peat-

lands, of which c. 5.4 million ha have been drained

for forestry and c. 0.7 million ha for agriculture.

Some 60,000 ha are used for harvesting fuel peat

and environmental peat. This leaves c. 40% of the

original peatland area still in a natural state. Most

of the land now used for harvesting fuel peat (peat

burned to generate energy) was formerly forest-

ry-drained, but some pristine peatlands have also

been taken over for peat production. By contrast,

peatlands formerly drained for agriculture are rarely

used for harvesting peat because of the problems

involved in the traditional production method.

Peat is a domestic fuel of national importance to

Finland and, considering its employment impact,

it also has regional policy significance. Some 5 to

6% of Finland’s overall energy demand is met by

peat combustion. The energy yield of peat harvest-

ed in Finland is about 18 TWh per year, and its

carbon content is about 1.35 million tonnes.

In international statistics (OECD/IEA/Eurostat),

peat is equated with fossil fuels. In Finland, peat

is classified as a slowly renewing biomass fuel,

the time required for the peat layer to rebuild be-

ing very long. In the National Climate Strategy

published in 2001 (VNS 1/2001 vp), it was noted

that “the basic principle underlying the comments

[of the Finnish Government] on Community ener-

gy taxation is that peat is left for the national au-

thorities to decide upon and thus outside the scope

of the EC Directives on energy taxation. [...] In in-

ternational statistical practice, [...] it is supported

that peat would be separated from the fossil fuel

category to form a category of its own.” It was also

noted: “As far as the [United Nations Framework]

Convention on Climate Change and other interna-

tional cooperation are concerned, the following

action will be taken:

Greenhouse Impacts of

the Use of Peat and Peatland

in Finland

1. Background and aims of

the research programme

The need for supplementary studies will be mapped

out and a research programme will be launched

based on this mapping for a life cycle analysis of

the energy use of peat.

If it is justified on the basis of the research find-

ings, measures will be undertaken to influence the

rules and definitions of the methodology used for

calculating greenhouse gases by virtue of the in-

ternational Convention on Climate Change. In this,

the aim will be that the calculation methods sub-

ject to the Convention would take the greenhouse

gas balance of peat into account during the entire

life cycle and not just the emissions from combus-

tion.

To be able to influence the Convention on Climate

Change, clear criteria, besides new research data,

will be required for the energy use of peat. The

criteria shall state e.g. the definition of those peat

lands on which the production of energy peat will

be directed, as well as the requirements for sub-

sequent use of the peat production areas.”

Related to the preparation of the Climate Strate-

gy, the Ministry of Trade and Industry commis-

sioned in January 2001 a survey of the need for

further research into the life cycle analysis. The

purpose of the survey was “to estimate what infor-

mation will be needed to scientifically motivate, if

possible, the introduction of calculation methods

better taking into account the life cycle of peat in

calculating emissions from the use of peat under

the principles of the Kyoto Protocol.” Another mo-

tivating factor was Finland’s obligation to report

on emissions from peat and peatlands under the

United Nations Framework Convention on Climate

Change (UNFCCC), the aim being to further spec-

ify these values.

In terms of climate impact, it would make sense to

direct peat harvesting to areas where anthropo-

genic greenhouse gas (GHG) emissions are high,

meaning that the cessation of these because of

the area is taken over for fuel peat harvesting would

compensate for some of the greenhouse impact

of the burning of that peat. Similarly, the after-use

of areas where peat harvesting is discontinued

should be planned so that the life cycle greenhouse

impact of the use of those areas would be as low

as possible.

As a result of the information needs outlined above,

a four-year research programme (2002–2005) was

set up, jointly funded by the Ministry of Trade and

Industry, the Ministry of Agriculture and Forestry

and the Ministry of the Environment. Its principal



15

purpose was to establish the levels of GHG emis-

sions from the use of peat and peatlands in Fin-

land. The research programme also drew up mod-

els for the GHG balances of peatlands in various

types of land use and quantified the underlying

ecosystem processes. The purpose of the com-

ponent research projects was to establish the GHG

balances of pristine peatlands and peatlands in

various types of land use and using them as a basis

for undertaking a life cycle analysis of various types

of production chains in the harvesting and use of

fuel peat. The purpose of a life cycle analysis is to

assess the overall climate impact of peat, taking

into account all the significant emissions and sinks

involved in the product. As an approach this dif-

fers from the normal GHG inventory, which aims

to present actual GHG emissions and sinks as

accurately as possible for the report year only.

This publication contains a brief description of the

aims, methods and principal findings of the stud-

ies carried out in the research programme, togeth-

er with a synthesis of the new findings and earlier

data. More detailed descriptions of each study are

available in the respective research articles on

them.



16

The research programme consisted of nine re-

search projects under a coordinating project. Some

of these were ‘method projects’, while others con-

cerned a specific sector of land use. The practical

work involved was carried out in cooperation with

three universities and four research institutions:

the Universities of Helsinki, Joensuu and Kuopio,

the Finnish Forest Research Institute (Metla), the

Finnish Meteorological Institute, the Geological

Survey of Finland and VTT Technical Research

Centre of Finland.

Steering group:

The research programme projects, their relation-

ships and their principal researchers are shown in

Figure 1. All researchers who took part in the re-

search projects are listed in the Appendix.

2. Implementig the research

programme

Figure 1. Structure of the re-

search programme and principal

researchers. The green boxes

represent ‘method projects’,

while the yellow boxes represent

projects involving greenhouse

gas balances in specific sectors

of land use.

Funding:

The funding and human resource input, analysed

by principal funding provider, for the period 2002–

2005 were as follows:

Funding Person- Funding

provider years (EUR)

Ministry of Trade 10.2 562,611

and Industry

Ministry of Agriculture 9.8 447,899

and Forestry

Ministry of the Environment 2.9 119,819

Other *) 35.9 1,909,081

Total 58.8 3,039,410

*) Ministries of Labour and Justice, Metla, VTT, Finnish

Meteorological Institute, Universities of Helsinki and Joen-

suu, Vapo Oy, and funding from the Academy of Finland.



17

3.1.1. Background

A pristine peatland (mire) is an ecosystem where

the decomposition of dead vegetation is slower in

the long term than the formation of litter from the

primary production of plants. This results in part of

the plant residue stratifying into peat. In primary

production, carbon dioxide (CO
2
) from the atmos-

phere is absorbed by vegetation and returned into

the atmosphere through respiration of heterotroph-

ic organisms. In a wet environment, the lack of

oxygen slows down decomposition, and dead veg-

etation may be stored in the ecosystem for very

long periods of time. However, some of the ab-

sorbed carbon returns into the atmosphere in the

form of methane (CH
4
), the end product of anaer-

obic decomposition chains. Also, carbon enters the

peatland ecosystem with rainwater and ground-

water and is carried out with runoff water. Howev-

er, CH
4
 and dissolved carbon only account for a

few per cent of the annual carbon balance com-

pared with CO
2
 absorption and respiration.

Peatlands thus have dual significance for the

greenhouse effect. It is estimated that the peat-

lands of the boreal and sub-arctic zones have ab-

sorbed more than one fifth of all the carbon in the

soil. While pristine peatlands have decreased the

greenhouse impact by absorbing carbon from the

atmosphere, the release of CH
4
 from anaerobic

decomposition chains into the atmosphere has

helped maintain the natural greenhouse effect.

However, human activity is increasing the level of

many greenhouse gases in the atmosphere, which

will probably lead to an acceleration of the green-

house impact and global climate change. Warm-

ing climate and redistribution of rain patterns geo-

graphically and temporally may change current

habitats and communities and also the function-

ing of societies. Also, CO
2
 and CH

4
 fluxes between

peatlands and the atmosphere change as circum-

stances change.

3. Use of peat and peatland

as a source of greenhouse

gas emissions

3.1. Pristine peatlands as green-
house gas sinks and sources
Sanna Saarnio, Micaela Morero,

Markku Mäkilä, Jukka Alm

3.1.2. Aim of the study

Carbon stored in peatlands is one of Finland’s most

important sources of domestic fuel. Taking a peat-

land into energy production discontinues the natu-

ral functioning of its ecosystem. The removal of

living vegetation prevents CO
2
 absorption, and

drying oxidizes the peat layers, thus increasing CO
2

release, although CH
4
 release is minimized. The

peat is harvested and burned, and as a result the

carbon stored in it over thousands of years is re-

leased back into the atmosphere in a matter of

decades. After peat harvesting, the cutaway peat-

land is afforested, re-wetted, left to regenerate

naturally or planted, for example, with reed canary

grass. As the area re-acquires vegetation, it grad-

ually begins to absorb CO
2
 again. Depending on

the form of after-use and the climate conditions,

the new ecosystem evolving in the area may be-

gin to function as a carbon sink. This study focused

on the annual greenhouse gas exchange and car-

bon balance of different types of pristine peatlands

under current conditions and on the carbon accu-

mulation of the past centuries and decades. The

ultimate aim was to draft average annual balanc-

es for CO
2
 and CH

4
 in pristine peatlands.

3.1.3. Research methods and principal

results

Field measurements

Literature was reviewed to find information on the

annual CO
2
 and CH

4
 balances of peatlands in the

coniferous forest belt, in Finland and abroad. The

data collated cover the entire boreal belt of North

America, Russia and Scandinavia, reflecting how

peatlands in general perform in this climate zone.

In addition to the literature review, carbon gas ex-

change was measured on two different types of

peatland in the municipality of Anjalankoski in the

region of Kymenlaakso (in southern Finland). The

two study sites were a sparsely forested ombro-

trophic raised bog (Haukkasuo, 60°49’N 26°57’E)

and the minerotrophic fen lagg of another raised

bog (Hangassuo, 60°47’N 26°54’E), chosen so as

to complement the range of sites in much-studied

areas and other component projects. In life cycle

analyses, peatlands are divided into two main

groups: nutrient-poor peatlands sustained mainly

by rainwater (ombrotrophic peatlands) and peat-

lands which besides rainwater also receive nutri-

ents from groundwater and runoff water (minero-

trophic peatlands). This division was employed in

the present study too. In practice, both these



18

Figure 2. Tuulia Tanttu

measuring net carbon di-

oxide flux at Hangassuo in

summer 2003; Micaela

Morero taking gas sam-

ples for measuring carbon

dioxide and methane ex-

change in winter 2003.

groups contain a wide range of peatland types dif-

fering greatly in terms of wetness and vegetation.

Primary production and decomposition were mon-

itored at different times of year, using the general-

ly employed closed chamber method (Figure 2).

Environmental factors known to be important were

also measured at the study sites: temperature,

photosynthetically active radiation, water table lev-

el, and the quantity and quality of vegetation. Us-

ing non-linear regression models representing the

relationship between environmental factors and

gas exchange and environmental time series, the

carbon gas fluxes for both sites over the study pe-

riod, 2002 to 2004, were reconstructed. Carbon

accumulation at the same sites over the past dec-

ades and centuries was examined by taking peat

core samples and analysing their dry matter den-

sity, carbon content and age (13C and 14C dating).

Carbon gas fluxes in peatlands are dynamic

Primary production and decomposition vary ac-

cording to light, temperature and humidity, daily

and annually. The differences in the quantity and

quality of vegetation between peatland types and

even between different types of surface structure

in the same peatland have a substantial effect on

CO
2
 and CH

4
 fluxes. Depending on what the

weather is like, the annual CO
2
 balance may rep-

resent a net efflux (emission)(–) or net influx (ac-

cumulation)(+). The CO
2
 balance of pristine om-

brotrophic peatlands in the boreal belt has been

found to vary between –85 and +67 g C m–2 a–1.

The corresponding estimate for minerotrophic

peatlands is between –101 and +98 g C m–2 a–1.

Thus, even in pristine peatlands in some years

decomposition exceeds carbon absorption, and the

carbon storage sink decreases. A negative annual

balance is created when the water table level sinks

lower than usual during the growing season. A dry

spell of only a few weeks is enough to speed up

aerobic decomposition in the exposed layer so that

the annual balance turns negative. In the long term,

however, the wet years have outnumbered the dry

ones, as witnessed by the accumulation of peat.

In wet and therefore anoxic peat layers, anaero-

bic decomposition produces CH
4
 as its end prod-

uct. The annual CH
4
 emission is the greatest in

wet peatlands with large numbers of vascular

plants. The differences in CH
4
 emissions may be

huge between different types of peatland and even

between different types of microsite surface in the

same peatland. Measurements show that the lev-

el of CH
4
 emissions is less than 1 to 16 g C m–2 a–

1 for ombrotrophic peatlands and less than 1 to 42

g C m–2 a–1 for minerotrophic peatlands. In a wet

and warm year, the local CH
4
 emission level may

be double that of a cool and dry year. Because

studies on CH
4
 have been conducted on many dif-

ferent types of peatland in years with different

weather, they probably reflect well the actual var-

iations in the level of CH
4
 emissions in boreal peat-

lands.

From measurements to models and predictions

Regression models were built to illustrate the re-

lationship between the measured carbon gas flux-

es and environmental factors, enabling the study

of the annual variation in the CO
2
 and CH

4
 bal-

ance at the same sites over a period longer than

the measurement period of three years. Typical

weather for the region over a 30-year period was

simulated using the Finnfor weather simulator. The



19

Figure 3. The simulated CO
2
 exchange at Hangassuo and Haukkasuo was the higher the greater the annual

temperature sum was. The temperature sum is the accumulation of the daily mean air temperature exceeding 5

°C. The variation in weather conditions in the simulated years correlated with the actual variation in weather condi-

tions in Anjalankoski between 1961 and 1990.

Figure 4. Carbon gas balances at

Haukkasuo and Hangassuo study

sites in Anjalankoski in the study years

2002 to 2004. The Haukkasuo site

was an ombrotrophic raised bog,

while the Hangassuo site was a min-

erotrophic fen lagg.

Photosynthesis
Ecosystem respiration

 Hangassuo, minerotrophic Haukkasuo, ombrotrophic

yr
.

yr.

Temperature sum (daily mean temp Temperature sum (daily mean temp

Year

yr
.

(minerotrophic)
(ombrotrophic)

weather variables (temperature, rainfall) were

used to calculate ambient temperature and mois-

ture for the study sites over the period simulated.

For annual vegetation development, models were

built using field observations. These time series

were used as input data for the regression mod-

els, and the end result was an hour-by-hour car-

bon gas exchange pattern for both study sites over

a period of 30 years.

The coverage of the regression models was good.

Development needs were found in the accurate

modelling of temperature and water table level at

the study sites. However, the models do yield find-

ings for this one peatland similar to those conclud-

ed on the basis of field observations at several

sites for previous individual years. As long as cli-

mate change does not lead to vegetation change,

photosynthesis rate fluctuations from one year to

the next will remain slighter than decomposition

rate fluctuations (Figure 3).

Dryness and heat favour aerobic decomposition,

which may lead to the peatland emitting more CO
2

than it absorbs. Although increased oxidization of

the peat reduces the amount of carbon emitted

into the atmosphere with CH
4
, the total carbon sink

of the peatland is reduced in such years. On the

basis of carbon gas balances calculated for the

years of study, both peatlands loosed carbon in

2002 and in the exceptionally dry year of 2003 (Fig-

ure 4).



20

Figure 5. Water table level of peat compared with sea level at the official Finnish Environment Institute measuring

point nearest to the study sites (Valkeala 60°55’N, 27°02’E). The values for the period most critical for carbon

absorption (June–August) are given in the time series in red.

Figure 6. Peat age, virtual peat accumulation and remaining dry matter relative to the age of the sample, at various

depths at the Hangassuo and Haukkasuo study sites. Hummocks are dry surfaces in a raised bog and hollows are

wet surfaces. About 40–50% of the dry matter is carbon.

In the wettest year 2004, the minerotrophic fen lagg

at the Hangassuo site would seem to have acted

as a carbon sink, but the ombrotrophic Haukkas-

uo site would still seem to have released slightly

more carbon than it absorbed. Because the water

table level has been lower than average in three

of the six years that have so far elapsed of this

decade, there is reason to believe that the peat

stock at the study sites has not increased at all in

the present millennium (Figure 5).

Recent carbon accumulation

Samples taken from the top peat layers were ana-

lysed for age, dry matter density and carbon con-

tent. This showed the carbon accumulation rate of

slightly decomposed litter and the carbon it con-

tains. The fen lagg seemed to acquire much more

litter than the raised bog, but decomposition re-

duced this amount so that over time the remain-

ing amount is roughly the same as in the raised

bog (Figure 6). On the top of the peatland, most of

the vegetation a few years old is still not decom-

posed, but at a depth of only 20–40 cm only a frac-

tion remains of vegetation decades or centuries

old. It is only deeper, in the permanently anoxic

layer under the water table level, that the peat and

the carbon it contains accurately reflect the car-

bon accumulation of the peatland. However, de-

composition continues in anaerobic conditions, too,

so the carbon accumulation rate is in any case

relative, and the accumulation rate for carbon ab-



21

•

•

•

•

sorbed at a particular time decreases over time

as the decomposition of the dead vegetation

progresses. For example, the current litter accu-

mulation rate in the top peat layer at Hangassuo is

almost 200 g C m–2 a–1, whereas the long-term

average carbon accumulation, based on the thick-

ness of the entire peat layer and the ages of its

various strata, is 22 g m–2 a–1 at both Hangassuo

and Haukkasuo, according to the Geological Sur-

vey of Finland (GTK). In Finland as a whole, the

average carbon accumulation rate since the lat-

est Ice Age has been 17 g C m–2 a–1 in minero-

trophic peatlands and 21 g C m–2 a–1 in ombro-

trophic peatlands. Although the long-term carbon

accumulation at the study sites of the present study

concur with these figures, carbon accumulation

has varied greatly over comparable periods of

time, depending on the evolution stage of the peat-

lands, the climate and local circumstances. Wheth-

er the carbon stock in peatlands will increase or

decrease and what the long-term accumulation

rate will be in the future for the peat layer as a

whole depends on the weather or, more specifi-

cally, on how common dry spells during the grow-

ing season will be in the future.

Outlook

The functioning of peatlands depends on local cir-

cumstances. Changes in the atmosphere and in

land use will thus continue to have a direct impact

on the functioning of peatland ecosystems. For

example, anthropogenic increase of CO
2
 in the

atmosphere will increase the amount of carbon

both absorbed and released by peatland ecosys-

tems. Thus, depending on the weather, peatlands

will be even greater carbon sinks or an even great-

er source of net emission in the future, assuming

that CO2 level is the only parameter to change.

However, it is estimated that Finland will have a

warmer climate in the future, particularly in the

autumn and winter, and that a higher rainfall is to

be expected than at present. Dry spells during the

growing season, which have become common-

place in recent years, may lead to an annual net

emission of carbon from peatlands. Warm au-

tumns will increase decomposition but, on the other

hand, the snow melting earlier in spring will bring

forward the beginning of carbon absorption at least

by plants which stay green in the winter.

Longer-term changes in the weather and thus in

the circumstances of ecosystems may cause ma-

jor changes in flora and fauna, too. Other factors

such as increased UV radiation, increased ozone

levels in the lower atmosphere or chemicalization

of our environment make it even more difficult to

predict how primary production, decomposition

and peat accumulation will evolve in peatlands.

3.1.4. Conclusions

Only a fraction of the vegetation which accumulates on the top of pristine undrained peatlands

ends up being stored as peat. The occurrence of dry spells during the growing season, which

recently has become more frequent, has slowed down the accumulation rate of peat in peatlands

in Finland. Dry spells are thus a highly significant factor for peat as a carbon sink.

The study shows that the carbon dioxide and methane fluxes in pristine peatlands vary widely

depending on the weather and the type of vegetation. This natural variation causes great diversity

in average annual balances. This could be reduced by categorizing peatlands into more groups

than just the main two, ombrotrophic and minerotrophic, and by conducting further research.

Current knowledge does not provide a sufficient overall view of the fluctuation over time of the

carbon dioxide balance in peatlands of different types. The information on methane emissions is

more reliable.

On the basis of the problems which emerged in the modelling of carbon balances, the following

points should be addressed: 1) developing models based on various processes in the carbon

cycle rather than blanket regression models; 2) taking into account the effect of water table vari-

ation in the surrounding areas on peatland water balance models; 3) developing modelling of the

annual variation in vegetation to account for the weather; and 4) securing that diverse field meas-

urements are obtained over the long term so that reliable models for predicting carbon gas ex-

change and thereby carbon accumulation under changing circumstances can be developed.



22

3.2.1. Background

Covering some 5 million hectares, forestry drain-

age is a potentially significant contributor to green-

house gas emissions in Finland. The drainage was

mostly carried out between the 1960s and 1980s

and today contributes an extra growth amounting

to more than 13 million cubic metres of wood per

year. Much of the forests growing in these areas

is due for thinning. But how do drainage and for-

estry affect greenhouse gas emissions on forest-

ry-drained peatland?

Drainage lowers the water table level of a peat-

land, oxygenates the top strata of the peat and

3.2. Greenhouse gas emissions
from forestry-drained peatlands
and contributing environmental
factors
Kari Minkkinen, Jukka Laine, Timo Penttilä

A pristine sedge fen (left) and the same fen 40 years

after drainage. The drainage has considerably increased

tree growth while also radically changing the surface

vegetation species. Peatland vegetation has been re-

placed by species typical of a heath forest. (Photo: Sakari

Sarkkola).

thereby enables the growing of trees on the peat-

land. The increased oxygen level in the top stra-

tum of the peat accelerates decomposition, lead-

ing to increased emissions of carbon dioxide (CO
2
)

and, in nutrient-rich peatlands, possibly also in-

creased emissions of nitrous oxide (N
2
O). On the

other hand, this decreases and often completely

prevents emissions of methane (CH
4
), a product

of anaerobic decomposition as the top strata of

the peat become oxidized and the deep-rooted

peatland vegetation disappears.

After drainage, the carbon bound in the biomass

in the area (mainly the trees) increases strongly.

Nevertheless, as a whole litter production of the

peatland (i.e. the carbon flow at ground level) will

not necessarily increase, since the biomass re-

mains bound in the above-ground portions of the

trees growing in the area until they are felled/cut

down. By contrast, the changes in underground

production (root growth and death) can be very

substantial, but this process is poorly known as

yet. However, the litter becomes woodier and de-

composes more slowly compared with the situa-

tion before drainage. The temperature and pH of

the peat decrease as the trees grow, and both fac-

tors slow down the decomposition of organic mat-

ter. Cuttings in forestry-drained areas typically raise

the water table level, whereas the soil preparation

involved in restoration, e.g. mounding, leads to the

peat mounds becoming oxidized, which may reac-

celerate decomposition. On the other hand, excess

dryness may limit decomposition in the summer

months.

Thus, whether a peatland turns into a carbon

source after drainage or whether it continues to

accumulate organic matter depends on changes

in the relationship between decomposition and

production. Earlier measurements indicate that

both trends are possible and that the outcome is

influenced by the geographical location of the

drained peatland and its ecohydrological status

before drainage.

3.2.2. Research methods and aims

There are two principal methods for measuring

greenhouse gas emissions: 1) the closed cham-

ber method, which enables measurement of gas

exchange of the soil and surface vegetation, and

2) the micrometeorological covariance method

(eddy covariance method), which enables meas-

urement of the gas exchange (usually CO
2
) of the

entire ecosystem above the treetops (see Laurila



23

Figure 7. Conceptual model used for calculating green-

house gas balances.

et al. 2008 p. 38 in this report). Soil gas fluxes in

forestry-drained peatland has previously been

measured at a few locations in central and east-

ern Finland, but the CO
2
 exchange of an entire

peatland has not been measured earlier. The tow-

er method gives the CO
2
 balance for the entire

ecosystem under measurement, but dividing this

balance into components (trees, surface vegeta-

tion, soil) is not possible without further measure-

ments. The closed chamber methods can be used

to measure the contributions of soil components

(peat, litter, roots) and surface vegetation to gas

exchange, leaving the contribution of the trees to

be modelled through growth and litterfall models.

In this research project, the closed chamber meth-

od was used to study gas fluxes of the soil. Spe-

cifically, 1) the effect of fellings on greenhouse gas

emissions from the soil was studied through fell-

ing studies in southern and northern Finland; 2)

greenhouse gas fluxes were measured in various

drained areas under different climate conditions;

and 3) gas fluxes were statistically modelled us-

ing statistically measured environment factors.

Using statistical models and weather simulations,

4) the amount of greenhouse gases released from

peat in various drained areas were estimated; and

5) the overall amount of greenhouse gases re-

leased from drained peatland in Finland annually

was estimated. Carbon fluxes into the soil were

also estimated by modelling tree yield and litterfall

using the Motti model (Hynynen et al. 2005) and

litter decomposition in the topsoil using the Yasso

model (Liski et al. 2005). The overall carbon bal-

ance of forestry-drained peatland was estimated

by combining the results of the production and de-

composition models (Figure 7).

3.2.3. Results

Carbon dioxide (CO
2
)

Emissions of carbon dioxide from forestry-drained

peatland were measured on sample plots where

ground vegetation and litter had been removed,

and roots had been cut one year before measure-

ments were begun. The CO
2
 emissions measured

were therefore due only to the decomposition of

organic matter in the peat (including the cut roots).

Temperature is the most important factor affect-

ing the decomposition of organic matter and hence

the volume of CO
2
 released from peatland in for-

estry-drained peatland. About 90% of the varia-

tion in CO
2
 emissions at the same location over

time can be explained by soil temperature. How-

ever, there is great spatial variation. Emissions are

affected not only by temperature but also by the

composition of the organic matter in the peat and

its microbe populations, but so far it has been dif-

ficult to model these. Unlike in pristine peatlands,

in forestry-drained peatland the water table level

is in most cases in so deep that its variations usu-

ally have only a minor effect on CO
2
 emissions,

since most of the CO
2
 released through the de-

composition of organic matter comes from new

litter and the top strata of the peat.

There were significant differences in CO
2
 emis-

sions between new test sites on forestry-drained

peatland studied. There was a clear increasing

trend in emissions from nutrient-poor peatland to

nutrient-rich peatland, which was expected. What

was unexpected was the result of comparing

southern and northern sites: emissions in compa-

rable peatland areas were highest in the north,

even though the temperatures were lower than in

the south. As CO
2
 emissions correlate to a great

degree with temperature, the effect of different

years was tested by simulating emissions in the

measurement areas using weather data for a pe-

riod of 30 years. The annual carbon emissions from

organic matter decomposition for the southern

Vaccinium myrtillus peatland type were about 350

g m–2, while the figure in the north was more than

470 g m–2 (Figure 8). A significant difference was

caused by winter emissions being substantially

smaller in the south than in the north, since in the

north the thick snow cover insulates the soil, where-

by decomposition can continue in the topsoil al-

most through the winter. Nutrient status of the soil

or water table levels of the peat provided no clear

indicator for these differences. However, the study



24

Figure 8. Simulated CO
2

emissions from decompo-

sition in peatland at various

measurement sites, rela-

tive to soil temperature. In

the simulation, the emis-

sion regression models

were run with input data

consisting of hour-per-hour

weather data from a peri-

od of 30 years. In the fig-

ure, each dot represents

the total emissions for one

year.

material is so limited that no conclusions can be

drawn regarding whether these findings can be

generalized to apply to all forestry-drained peat-

land or whether these are just properties of the

specific sites studied. The results are being tested

and their general validity examined in a new project

financed by the Ministry of Agriculture and Forest-

ry (Carbon balance of forested peatlands – pre-

dictions and monitoring in changing conditions),

where CO
2
 measurements are being carried out

at some 70 sites all around Finland.

The soil respiration measurements only account

for CO
2
 released from the soil. The soil CO

2
 bal-

ance of forestry-drained peatlands was estimated

by combining the soil respiration measurement re-

sults with the modelled carbon binding figures. The

model calculations show that in most cases the

amount of CO
2
 released from the soil exceeded

the amount of CO
2
 bound, leading to a negative

CO
2
 balance. The soil of such forestry-drained

peatlands was thus shown to be a source of car-

bon emissions into the atmosphere.

Based on earlier studies, we have suggested that

the soil carbon balance of forestry-drained peat-

land is, on average, positive (i.e. that forestry-

drained peatland is a carbon sink). Positive bal-

ances were observed on nutrient-poor peatland

and negative balances on nutrient-rich peatland.

The soil respiration results obtained now show the

same trend: the modelled carbon absorption com-

pletely or almost completely compensates for the

measured decomposition in the more nutrient-poor

site types, but in nutrient-rich site types and in

northern Finland, in particular, the soil carbon bal-

ance was decisively negative (Figure 9). Howev-

er, when the carbon absorbed by the trees over

the long term (1st cycle) was included in the cal-

culation (including an estimate of the carbon re-

moved in connection with cutting), the carbon bal-

ance for the ecosystem as a whole usually turned

out to be positive (Figure 9).

In this research programme, the CO
2
 balance was

measured directly for the first time using the Eddy

covariance method, at two forested locations in

southern Finland – afforested organic cropland and

a nutrient-poor forestry-drained peatland site. The

former turned out to be a source of carbon and the

latter a carbon sink. When the carbon absorbed

by the trees during the measurement interval was

subtracted from the total net exchange of carbon,

the net emission from the soil (and surface vege-

tation) of the afforested organic soil cropland was

c. 250 g C m–2 a–1, while the forestry-drained peat-

land showed a net accumulation of more than 100

g C m–2 a–1. These results support earlier findings

that there is great variation in carbon dynamics

between different types of drained peatland, and

that even drained peatlands can function as sig-

nificant carbon sinks.

Naturally, there is much scope for error in both the

modelling and the measurements. The production

models are based on limited information regard-

ing, for example, underground production, and the

decomposition models have not yet been calibrat-

ed for peatland. There may be systematic errors

in the measurements due to site processing (root



25

Figure 9. Sample calculation

of carbon fluxes in drained

peatlands of different types

(Vatkg–Mtkg) in southern Fin-

land (1200 dd) and northern

Finland (<900 dd). Negative

figures refer to net carbon

emission and positive figures

to net carbon accumulation.

AG = above ground, BG =

below ground.

cutting and vegetation clearing). Further research

will be undertaken to estimate the effects of these

sources of error.

Finland’s first greenhouse gas inventory report

drew on the findings of the modelling method,

which showed that the overall carbon balance of

forestry-drained peatland in Finland was negative.

However, because of the great variation observed

and the partly conflicting results obtained through

different research methods, we cannot yet assess

the reliability of this modelling calculation. Esti-

mates will improve as research data with better

regional coverage will be obtained in the near fu-

ture.

Methane (CH
4
)

The situation with regard to methane is consider-

ably simpler than that of carbon dioxide. Emissions

of CH
4
 decrease as time passes from the drain-

age, with the progression of drying and flora suc-

cession. The rate of drying succession depends

on the changes that occur in the site’s ecohydrolo-

gy: in peatland types which are originally nutrient-

rich and wet, the vegetation changes quickly, while

in nutrient-poor peatland types the process is slow-

er. The rate of change also correlates closely with

the growth of the trees, and the CH
4 

emissions

correlate negatively with tree stand volume – as

tree stand volume increases, the CH
4
 emissions

decrease. Indeed, it is quite possible to estimate

CH
4
 emissions on the basis of the tree stand vol-

ume in the area. Data gathered from Finnish peat-

lands suggests that, on average, CH
4
 emissions

cease when the tree stand volume exceeds 140

m3 per hectare (Figure 10). This never happens

on nutrient-poor forestry-drained peatland, but the

more nutrient-rich peatlands become CH
4
 sinks

20 to 30 years after drainage. Because all drained

peatlands are still relatively young, and many of

them are nutrient-poor, Finland’s forestry-drained

peatland is still a small source of CH
4
 (0.048 Tg

per year), albeit this figure has decreased consid-

erably compared with the natural state.

Nitrous oxide (N
2
O)

Generation of nitrous oxide is possible under con-

ditions where nitrification and denitrification proc-

esses are active. In pristine peatlands, nitrifica-

tion is prevented because of lack of oxygen, and

in forestry-drained peatland the low pH inhibits the

process. However, in nutrient-rich peatlands or

fertilized drained peatland, N
2
O can be generat-

ed, and substantial isolated emissions have been

observed, for instance, in connection with the

ground freezing. These emissions cannot yet be

modelled at the process level, but there is a strong

statistical correlation between the carbon-nitrogen

ratio of the soil (CN) and its annual emissions: if

there is a lot of nitrogen in relation to carbon (a

low CN ratio), N
2
O emissions increase. The de-

pendency is a non-linear one, and a significant

change occurs when the CN ratio falls from 40 to

20 (Figure 11). In drained spruce peatlands, CN

ratios are considerably lower than in pine peat-

lands, which can also be seen in their bigger than



26

Figure 10. CH
4
 emissions from forestry-drained peat-

land as a function of tree stand volume.

Figure 11. N
2
O emissions from forestry-drained peat-

land as a function of the CN ratio.

•

•

•

average emission levels. We used the regression

model and the CN ratio distributions of forestry-

drained peatland together with peatland-type-spe-

cific CN ratios to predict N
2
O emissions from Fin-

land’s forestry-drained peatland. The current emis-

sion prediction is, depending on the approach,

between 0.010 and 0.015 Tg N
2
O a–1.

Impact of forest fellings on emissions

Felling of tree stand raises the water table level

and the soil temperature slightly. Overall, CO
2

emissions from the soil decreased by 35 to 45%,

corresponding to the volume of root respiration,

but the volume of CO
2
 released by the decompo-

sition of old peat did not change. Felling thus had

little impact on the soil carbon balance. The rise

of the water table level in the felling area caused

a slight decrease in the absorption rate of CH
4
.

N
2
O emissions rose at the locations of felling res-

idue piles. Obviously, felling residue releases ni-

trogen faster than the sparse local vegetation can

use it, which leads to nitrification and the release

of N
2
O, an intermediate product in this process.

On the whole, however, cutting has a minor im-

pact on soil GHG balance.

3.2.4. Conclusions

The results of the research project showed that the carbon dioxide balances of forestry-drained

peatland vary greatly; an area may be a carbon source or a carbon sink, depending on the site

and the climatic conditions. Carbon balances can be estimated using the existing models, but it is

not yet possible to assess the reliability of the average values obtained due to the multitude of

factors affecting emission dynamics.

The results show that drying substantially reduces methane emissions if the drying succession is

sufficient to cause a clear change in the flora and tree growth. Because in Finland many peat-

lands have been drained that are too nutrient-poor for profitable wood production, such peatlands

continue to release methane into the atmosphere.

According to new calculations made during the project, forestry drainage increases nitrous oxide

emissions in nutrient-rich locations more than previously estimated. Drained spruce peatlands, in

particular, are significant sources of nitrous oxide.



27

3.3.1. Background and aim

About 0.7 million ha of Finland’s peatlands have

been drained for agriculture. About half of these

areas have been abandoned or afforested, but

about 300,000 ha are still in agricultural use (Myl-

lys & Sinkkonen 2004). In agriculture, a lowering

of the water table level, repeated tillage, fertiliza-

tion, liming and mineral soil addition change the

properties of peatland. These also enhance the

formation and fluxes of greenhouse gases (GHGs).

Drained organic cropland is always a net emitter

of carbon dioxide (CO
2
) and (N

2
O), but it may be a

weak methane (CH
4
) sink.

The global warming potential of nitrous oxide is

296 times and methane 23 times than that of car-

bon oxide, in a 100-year time horizon (IPCC 2001).

Agricultural soils are responsible for most of the

global N
2
O emissions from soils. A pristine peat-

land emits hardly any N
2
O at all, but drainage in-

creases emissions of N
2
O from organic soils and

organic croplands are particularly important as

regards the atmospheric N
2
O load. As much as

25% (4 Tg annually) of the N
2
O emissions in Fin-

land may originate from organic croplands

(Kasimir-Klemedtsson et al. 1997), although these

soils represent only 13.6% of the total agricultural

land in the country (Myllys & Sinkkonen 2004). N
2
O

is produced in soil microbial activities, with nitrifi-

cation and denitrification as the key processes, with

contributions from environmental factors such as

temperature, soil moisture, soil pH and vegeta-

tion. Cultivation practices such as tillage, fertiliza-

tion, irrigation and compaction of soil by machin-

ery can also affect the N
2
O production. Winter

emissions should be given particular attention in

boreal areas, as they can account for more than

half of the annual N
2
O emissions.

Methane from agriculture originates mainly in the

digestion of ruminants, manure management and

from waterlogged soils such as rice paddies. Well-

drained organic croplands are generally sinks for

CH
4
 because CH

4
 is oxidized by methanotrophic

bacteria. The water table level, temperature, ni-

trogen fertilization, liming, tillage and other similar

measures can affect CH
4
 fluxes.

After the cultivation practices have ceased, grad-

ual secondary vegetation succession starts. Grass-

es and herbs dominate in field vegetation for 15

years, and open field ditches are the first habitat

for the pioneer tree species (birch and willow). The

gradual deterioration of the ditches leads to high-

er water table levels, which may increase CH
4

emissions. On the other hand, ending cultivation

practices may reduce the decomposition rate of

peat and thus also CO
2
 and N

2
O emissions. The

annual GHG emissions from Finnish organic cro-

plands in active use were measured at five sites.

In this paper we summarize the results of these

studies. We also report the annual and seasonal

GHG emissions from five abandoned organic cro-

plands and discuss how GHG emissions change

after cultivation practices have ceased.

3.3. Greenhouse gas emissions
from cultivated and abandoned
organic croplands
Marja Maljanen, Jyrki Hytönen, Päivi

Mäkiranta, Jukka Alm, Kari Minkkinen,

Jukka Laine, Pertti Martikainen

A summery bar-

ley field on or-

ganic cropland in

southern Finland
(Photo: Sakari

Sarkkola).



28

3.3.2. Materials and methods

GHG fluxes on organic croplands

Annual N
2
O and CH

4
 emission measurements

were carried out at five different sites between

1991 and 2002. N
2
O and CH

4
 fluxes were meas-

ured using a chamber method. CO
2
 balances were

measured at two sites using a chamber method

and at two sites using the eddy covariance (EC)

method. The measurements were made on soils

under barley or grass. Sites with no vegetation (fal-

low) were also studied (Table 1).

GHG fluxes on abandoned organic croplands

N
2
O and CH

4
 fluxes were measured between 2002

and 2005 on five abandoned organic croplands in

Kannus, western Finland (Table 1). There were

altogether 30 gas sampling plots on these fields.

These sites were drained and used for cultivation

for decades before being abandoned (no fertiliza-

tion or ploughing) 20 to 30 years ago.

CO
2
 exchange between the soil-vegetation sys-

tem and the atmosphere was measured using a

chamber method between 2002 and 2004. The

results given here are preliminary results for 2003.

The net ecosystem CO
2
 exchange (NEE) was

measured with a transparent climate-controlled

chamber (60 x 60 cm, height 30 cm) that was

placed on an aluminium collar set in the ground

throughout the measurement. The collar had a

groove in the upper edge which was filled with

water to ensure a gas-tight seal. A CO
2
 analyser

(EGM-4 Environmental Gas Monitor for CO
2
, PP

Systems, UK) monitored the change in the CO
2

concentration in the chamber during the chamber

closure period of 180 seconds. Simultaneously, the

light intensity (PAR) and air temperature (T
a
) in-

side the chamber were recorded. The measure-

ments were made in full light and then in reduced

light under mosquito netting. After that, the gross

respiration rate (R
TOT

) of the plant-soil system was

measured in the same way by covering the collar

with an opaque cover.

NEE and R
TOT

 were calculated from the linear in-

crease or decrease in the CO
2
 concentration in the

chamber. CO
2
 uptake from the atmosphere into

the soil is designated with a positive, and release

of CO
2
 from the soil into the atmosphere with a

negative sign. NEE can be positive or negative,

while R
TOT

 is always negative. An estimate of gross

photosynthesis (P
G
) was calculated using the for-

mula: NEE
 
= P

G
 – R

TOT
 (1).

For calculation of the diurnal cycles of NEE, the

values for P
G
 and R

TOT
 were needed for every hour.

Statistical response functions were constructed

separately for each site in order to predict P
G
 and

R
TOT

 (e.g. Alm et al. 1997). The model included

gross photosynthesis, photosynthetically active

radiation, leaf area index, soil temperature at a

depth of 5 cm, and water table level.

The diurnal cycles of P
G
 and R

TOT
 were then recon-

structed using the above model and continuous

data for the environment variables. The net CO
2

emission for the growing season was calculated

from the hourly values of NEE. Outside the grow-

ing season, the net CO
2
 flux was measured simi-

larly to the N
2
O and CH

4
 fluxes (see next section).

N
2
O and CH

4
 flux measurements were made on

abandoned organic croplands over three years.

The results reported here are for 2003 and 2004.

During the snow-free periods, fluxes of N
2
O and

CH
4
 were measured every second week with the

static chamber method. Gas concentrations were

analysed within 24 hours of sampling with a gas

chromatograph. During the winter snow cover, the

gas fluxes were determined using the gas gradi-

ent technique, where gas samples for concentra-

Table 1. Test sites: Cultivated sites: 1. Jokioinen (Regina et al. 2004, Regina et al. 2004, Regina et al. 2006), 2.

Liperi (Maljanen et al. 2001, 2003a, b), 3. Ilomantsi (Nykänen et al. 1995), 4. Rovaniemi (Regina et al. 2004,

Regina et al. 2006), 5a. Kannus, (two subsites with different peat depths) (Maljanen et al. 2004); Abandoned soils:

5b. (5 sites).

Site C/N Bulk density pH (H
2
O) Peat depth Drainage

(g cm–3) (m) (years ago)

1. 21 0.49–0.51 5.8 nd 100

2. 16 0.33 6.0 0.2 40

3. 19 nd 5.3 1.4 60

4. 18 0.24–0.29 5.6 nd 50

5a. 31–32 0.32–0.50 4.8 0.3–0.7 nd

5b. 16–19 0.33–0.47 4.3–5.9 0.2– >1.0 50–100

nd = not determined



29

Figure 12. CO
2
, N

2
O and CH

4
 flux averages for cultivat-

ed organic cropland, fallow soil with no vegetation and

abandoned organic croplands. The error lines show the

standard error of the average. n = number of annual

emissions measurements.

Table 2. Annual measured GHG emissions on cultivated organic croplands in Finland. Negative figures indicate

gas consumption on the ground. CO
2
-C, CH

4
-C and N

2
O-N emissions are given as g m–2.

Plant Site CO
2
-C CH

4
-C N

2
O-N Reference

Barley 1.   210 –0.01– –0.04 0.62–2.41 Lohila et al. 2004, Regina et al. 2004,  2006

Barley 2. 400 –0.37– –0.01 0.83–0.84 Maljanen et al. 2001, 2003a, 2003b

Barley 4. nd –0.02–0.38 0.73–1.88 Regina et al. 2004, Regina 2006

Barley 5a. 830 –0.13– –0.06 0.54–1.13 Maljanen et al. 2004

Grass 1. 80 –0.05– –0.01 0.50–0.99 Lohila et al. 2004, Regina et al. 2004, 2006

Grass 2. 750 –0.08 1.10 Maljanen et al. 2001, 2003a, 2003b

Grass 3. nd 0.10–0.20 0.78–0.93 Nykänen et al. 1995

Grass 4. nd 0.27–0.68 0.26–0.53 Regina et al. 2004,  2006

Grass 5a. 330–460 –0.07– –0.18 0.17–0.38 Maljanen et al. 2004

Fallow 1. nd –0.03– –0.01 1.34–3.70 Regina et al. 2004

Fallow 2. 880–1100 –0.26– –0.13 0.65–0.71 Maljanen et al. 2001, 2003a, 2003b

Fallow 4. nd 0.04–3.00 0.38–0.50 Regina et al. 2004,  2006

Fallow 5a 690–790 –0.14– –0.01 0.40–3.70 Maljanen et al. 2004

nd = not determined

tion analyses were drawn from the snow pack us-

ing a stainless steel probe 3 mm in diameter. Gas

fluxes from the soil through the snow pack into

the atmosphere were calculated using Fick’s law.

3.3.3. Results

At the annual level the cultivated organic croplands

were all net sources of CO
2
 (Table 2, Figure 12).

The CO
2
 losses measured with the chamber meth-

od from soil under grass varied from 79 to 750 g

CO
2
-C m–2 and CO

2
 losses from under barley were

from 210 to 830 g CO
2
-C m–2. The net CO

2
 emis-

sions from fallow soils (without vegetation) had a

similar CO
2
 net loss, from 690 to 1,100 g CO

2
-C

m–2, as the barley fields.

The EC measurements showed lower CO
2
 emis-

sions than the chamber measurements. The CO
2

balance is very sensitive to climatic conditions,

and therefore this difference can be partly ex-

plained by the variation in the environmental fac-

tors controlling CO
2
 flux, because the EC and

chamber measurements were conducted in dif-

ferent years, in different climatic conditions. How-

ever, the emissions are close to the net emissions,

400 to 550 g CO
2
-C m–2, estimated earlier for bo-

real organic croplands.

The abandoned organic croplands were either

small net sinks of CO
2
 (max. 90 g CO

2
-C m–2) or

sources of CO
2
 (max. net emission 900 g CO

2
-C

m–2). All abandoned sites showed some net up-

take of CO
2 
during the growing season, but out-

side that period all abandoned croplands were

sources of CO
2
. The mean annual CO

2 
emission,

324 g CO
2
-C m–2, is close to the net CO

2 
emis-

sions from cultivated croplands. It seems that



30

•

•

•

abandoned organic croplands do not generally turn

into CO
2 
sinks quickly after the agricultural prac-

tices have ceased, although their net CO
2 
emis-

sions may be slightly reduced compared to culti-

vated croplands (Figure 12). As abandoned crop-

lands slowly become afforested, the amount of

carbon assimilated by the trees increases, de-

creasing emissions. Natural afforestation is rath-

er slow, though (Hytönen 1999). Fallow soils, with-

out vegetation, showed an almost similar CO
2 
net

loss as the barley fields.

Cultivated organic croplands were either small

sinks or sources of CH
4
 (Figure 12), depending

on the water table level at the site. The fallow soils

without vegetation had a lower CH
4
 uptake than

the cultivated soils, or the corresponding emission

was greater (Figure 12). Although CH
4
 fluxes

seemed to depend on the weather at all the study

sites, there was no correlation between the mean

annual CH
4
 flux and the mean water table level,

pH, the soil C or N level or the CN ratio of soil.

Annually, all the abandoned cropland soils were

net sinks for atmospheric CH
4
, though some peri-

ods of low CH
4
 emissions were measured during

the summer. Despite the deterioration of the ditch

network after 20–30 years of abandonment, the

water table level remained reasonably deep. The

mean soil CH
4
 uptake rate was even higher at the

abandoned sites than at the cultivated sites. In

other words, the ability of abandoned organic crop-

lands to withdraw CH
4
 from the atmosphere may

increase gradually over time (Figure 12).

The mean N
2
O emissions from croplands under

barley were higher than those from croplands un-

der grass (Figure 12), but N
2
O emissions from bare

soils were generally higher still. This may be relat-

ed to higher nitrogen availability for denitrification

in the absence of plants. Surprisingly, the N
2
O

emissions from the abandoned cropland soils were

similar to or even higher than those from the culti-

vated croplands (Figure 12). The time since the

end of cultivation practices did not correlate with

the annual N
2
O emission rates from the abandoned

croplands. There is thus no evidence that the end-

ing of cultivation activities would reduce N
2
O emis-

sions from organic croplands.

There was high variation in the seasonal and an-

nual N
2
O emissions, and only a part of this could

be explained by weather conditions, e.g. temper-

ature and precipitation. Generally, the annual N
2
O

emissions from the croplands and abandoned cro-

plands did not correlate with the mean water table

level, even though there was some correlation

during the growing season. There was, however,

a weak correlation between the carbon-nitrogen

(CN) ratio of both cultivated and abandoned crop-

lands and annual N
2
O emissions. Klemedtsson

et al. (2005) have recently reported that the CN

ratio of afforested organic soils may predict annu-

al N
2
O emissions. However, in croplands other

factors, e.g. fertilization and tillage, may overcome

the CN dependence, which reflects the availabili-

ty of mineral nitrogen for microbial processes im-

portant in N
2
O emissions. The winter emissions

(from October to May) of N
2
O accounted for a sig-

nificant portion of the annual N
2
O emissions: 25%

for grass, 50% for abandoned cropland and 60%

for barley. It should also be noted that winter emis-

sions of N
2
O varied greatly from one year to

another.

CO
2
 was the most important gas in terms of the

atmospheric impact of these organic croplands,

being responsible, on average, for 78% of the to-

tal global warming potential (GWP), i.e. the sum

of CO
2
, N

2
O and CH

4
 with a 100-year time horizon

(IPCC 2001). N
2
O was responsible for about 22%

of the GWP, and the effect of CH
4
 was insignifi-

cant, less than 1%.

3.3.4. Conclusions

The results of the research project indicate that emissions of carbon dioxide from abandoned

organic croplands do not seem to decrease significantly with time after agricultural practices

have ceased – the decomposition of drained peat evidently continues.

The results further indicate that emissions of nitrous oxide from organic croplands can still be

high even after 20–30 years of abandonment. Winter emissions account for a significant portion

of the annual N
2
O balance.

Methane fluxes from the atmosphere into the soil may gradually increase after cultivation has

ceased, but the slow deterioration in the ditch network may raise the water table level, which in

turn may increase methane emissions. It should be noted, however, that methane emissions from

abandoned organic cropland are of a marginal significance compared with the global warming

potential of carbon dioxide and nitrous oxide emissions.



31

3.4. The effect of afforestation of
organic croplands and cutaway
peatlands on greenhouse gas
balance
Jyrki Hytönen, Lasse Aro, Marja Maljanen,

Päivi Mäkiranta, Hannamaria Potila,

Jukka Laine, Annalea Lohila,

Pertti Martikainen, Kari Minkkinen,

Mari Pihlatie, Narasinha Shurpali

3.4.1. Background

In Finland the afforestation of organic croplands

and cutaway peatlands constitutes one of the most

important means for creating carbon sinks set out

in the Kyoto Protocol. Article 3.3 of the Kyoto Pro-

tocol allows for the use of land-use, land-use

change and forestry measures (LULUCF) as car-

bon sinks. The most obvious effect of afforesta-

tion is the absorption of carbon by the growing

trees. The changes in soil greenhouse gas fluxes

are more difficult to estimate, and there are little

research data on this.

The area of cultivated organic soils (soil organic

matter content >20% of solid matter in soil) in Fin-

land is nowadays about 300,000 hectares, which

is less than half of the peatland area originally tak-

en over for agricultural use (Myllys & Sinkkonen

2004). Large-scale afforestation of agricultural

land, aimed at reducing the area under cultivation

in the country, began in the late 1960s. Of the more

than 240,000 ha of afforested agricultural land

(Finnish Statistical Yearbook of Forestry 2004), the

area on peatlands is estimated to be more than

80,000 ha. From the total area of 1,200,000 ha of

peatland suitable for peat production in Finland

(Virtanen & Hänninen 2004), annual peat harvest-

ing nowadays covers 42,000 to 59,000 ha. Peat

harvesting has already ceased on an area cover-

ing more than 20,000 ha (Selin 1999).

During agricultural use or fuel peat production,

peatlands are efficiently drained. Continuous cul-

tivation measures such as tillage and harrowing,

fertilization, liming and addition of mineral soil

change the properties of the peat soil substantial-

ly. In the afforestation of cutaway peatland, the

nutrient imbalance is corrected by mixing mineral

soil into the peat or by using fertilization. In both

organic cropland and cutaway peatland the peat

is well decomposed, and it has a high bulk densi-

ty and nitrogen content. Organic cropland and cut-

away peatland are significant sources of carbon

dioxide (CO
2
), as studies show.

Due to efficient drainage, organic croplands are

minor methane (CH
4
) sinks. The drainage may

begin to deteriorate gradually after afforestation.

After cultivation measures are ended, the aera-

tion of the topmost peat layer may decrease, lim-

iting tree growth and aerobic decomposition, es-

pecially if efficient drying is not provided for. Pro-

duction of CH
4
 may also increase in these areas.

In general, agricultural soils account for most of

the global nitrous oxide (N
2
O). Despite the fact that

the area of organic croplands is small, they are

estimated to account for 25% of the total anthro-

pogenic N
2
O emissions in Finland (Kasimir-

Klemedtsson et al. 1997). Only scattered data on

the effects of afforestation on N
2
O fluxes from soil

exist (cf. Maljanen 2001b).

Not much is known about the impact of afforesta-

tion on the greenhouse gas (GHG) balances of

organic cropland and cutaway peatland. After af-

forestation, gradual changes in the soil structure,

chemistry and biology may change the peat de-

composition rate. Though the soil respiration rate

is mainly regulated by soil temperature and mois-

ture, also substrate properties can have substan-

tial impacts on microbial activity in peat. Affores-

tation implies that the annual cycle of cultivating

and harvesting crops is replaced by a much long-

er forest tree rotation. After afforestation, repeat-

ed soil amelioration measures such as tillage, fer-

tilization and liming cease. These factors may

change the soil properties into less favourable ones

for the microbes and thereby lead to a slower de-

composition rate of the organic matter and to re-

duced CO
2
 and N

2
O emissions. T