全 文 ：农田和森林土壤中氧化亚氮的产生与还原 3
于克伟 3 3 　陈冠雄　(中国科学院沈阳应用生态研究所 , 沈阳 110015)
Sten Struwe 　Annelise KjФller 　(哥本哈根大学基础微生物系 , D K21307 ,丹麦)
【摘要】　采用土壤淤浆方法对丹麦农田和山毛榉森林土壤反硝化过程中 N2O 的产生与还原进行了研究 , 同时
考察了硝酸根和铵离子对反硝化作用的影响. 结果表明 , 森林土壤反硝化活性大于农田土壤 , 但农田土壤中
N2O 还原活性大于森林土壤 , 表现在农田和森林土壤中 N2O/ N2 的产生比率分别为 0. 11 和 3. 65. 硝酸根和铵
离子能促进两种土壤中的 N2O 产生 , 但可降低农田土壤中的 N2O 还原速率. 与农田土壤相比 , 硝酸根可降低
森林土壤 N2O 的还原能力 , 而铵离子却表现出对 N2O 还原的促进作用. 培养 1 周内尽管农田土壤中硝酸根始
终存在 , 但几乎所有产生的 N2O 都被还原成 N2 . 然而 , 只有当硝酸根几乎耗尽 , 森林土壤中 N2O 的还原才开
始进行. 由于两种土壤 p H 值的不同造成硝酸根对 N2O 还原抑制作用的差异 , 因为酸性条件下硝酸根对 N2O
还原的抑制较强. 因此 , 土壤消耗大气 N2O 的适宜条件为厌氧 , p H 中性和硝酸根含量低.
关键词 　反硝化作用 　温室气体 　N2O 还原 　N2O/ N2 比率
Production and reduction of nitrous oxide in agricultural and forest soils. YU Kewei ,CHEN Guanxiong ( Institute of
A pplied Ecology , Chinese Academy of Sciences , S henyang 110015 , China) , Sten Struwe , Annelise Kj«ller ( Depart2
ment of General Microbiology , U niversity of Copenhagen , S «lvgade 83 H , D K21307 Copenhagen K , Denm ark) . 2
Chin. J . A ppl . Ecol . ,2000 ,11 (3) :385～389.
A soil2water slurry experiment was conducted to study the potentials of N2O production and reduction in denitrification
of agricultural and beech forest soils in Denmark. The effects of nitrate and ammonium additions on denitrification
were also investigated. The forest soil showed a higher denitrification potential than the agricultural soil. However ,
N2O reduction potential of the agricultural soil was higher than the beech forest soil , shown by the ratio of N2O/ N2
approximately 0. 11 and 3. 65 in the agricultural and the beech forest soils , respectively. Both nitrate and ammonium
additions stimulated the N2O production in the two soils , but reduced the N2O reduction rates in the agricultural soil
slurries. In contrast to the effect on the agricultural soil , nitrate reduced the N2O reduction rate in the beech forest
soil , while ammonium showed a stimulating effect on the N2O reduction activity. After one week incubation , all of the
N2O produced was reduced to N2 in the agricultural soil when nitrate was still present . Nitrous oxide reduction in the
beech forest soil occurred only when nitrate almost disappeared. The different nitrate inhibitory effect on the N2O re2
duction activity in the two soils was due to the difference in soil p H. Inhibition of nitrate on N2O reduction was signifi2
cant under acidic condition. Consequently , soil could serve as a sink of atmospheric N2O under the conditions of anaer2
obic , p H near neutral and low nitrate content .
Key words 　Denitrification , N2O reduction , Greenhouse gases , N2O/ N2 ratio.
　　3 欧洲联盟资助项目 The research was supported by an European U2
nion project ,grant CHN GNRL/ J RP/ 1130. L .
　　3 3 通讯联系人. Corresponding author.
　　Received 1999 - 10 - 08 ,Accepted 1999 - 11 - 18.
1 　Introduction
Nitrous oxide (N2O) is not only one of the green2
house gases , but also one of the gases which react with
ozone and may contribute to depletion of the ozone layer
in the st ratosphere[10 ,27 ] . Since preindustrial times the
amount of N2O present in the atmosphere has increased
from about 275 ppb (v) to about 310 ppb (v) with a con2
tinuously increasing rate about 0. 2 to 0. 3 % yr - 1 . The
atmospheric lifetime of N2O is about 120 years , much
longer than some other important greenhouse gases ,
such as CO2 and CH4 , indicating that the present in2
crease will have a long term effect on the global cli2 mate[15 ] .There is a very broad range of values in the esti2mated global sources and sinks of N2O from Intergov2ernmental Panel on Climate Change ( IPCC) , which in2dicates that there still are considerable uncertainties inthe estimation due to the temporal and spatial variationsin N2O production and consumption. Denit rification isone of the major sources of N2O production and emissioninfluenced by different environmental factors. Denit rifi2
应 用 生 态 学 报 　2000 年 6 月 　第 11 卷 　第 3 期 　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　
CHIN ESE JOURNAL OF APPL IED ECOLO GY ,J un. 2000 ,11 (3)∶385～389
cation represents the only known biological process for
consumption of N2O and the rate of this last step in the
reduction of NO -3 determines the ratio between N2O and
N2 [13 ] . There is some evidences in field measurements
that soils consume ambient N2O [1 ,5 ,6 ,11 ,23 ] , but there
are not enough data to evaluate the global capacity of
N2O removal by soil and consequently there is no esti2
mated value in the IPCC report on this issue[14 , 15 ] . In
previous research carried out in this field , the effect of
nit rate on the reduction of N2O to N2 was stud2
ied[3 , 4 , 26 ] . In this study , we included both nit rate and
ammonium treatments in anaerobic incubation with an
agricultural and a beech forest soils to investigate the po2
tentials of N2O reduction to N2 and the effects of nit rate
and ammonium.
2 　Materials and Methods
211 　Sample soils and treatments
Surface soils (0～20cm) from an agricultural field and a beech
forest in Denmark were collected in March. The agricultural field
is planted with barley and fertilized once a year during sowing of
the crop in May at a rate of 80 kg N hm - 2 with NO -3 2N ¬NH42N
5. 5 ¬7. 0. The soils were air2dried at room temperature , sieved
(2 mm) and stored at 4 ℃before the use. Basic soil characteristics
were analyzed , and the results are shown in table 1.
Table 1 Characteristics of the agricultural and the beech forest soils
Agricultural soil Beech forest soil
Total C ( %) 1. 30 1. 69
Total N ( %) 0. 19 0. 17
Ammonium (μg N g - 1 dry soil) 0. 08 1. 26
Nitrate (μg N g - 1 dry soil ) 7. 42 6. 64
p H 6160 3180
Water content ( % ,w/ w) 4. 22 9. 84
　　The soil2water slurry experiment was carried out by the fol2
lowing procedure : 15g (fresh weight) air2dried soil was weighted
into a 50 ml bottle with 10 ml distilled water. Each of the two
soils has six treatments , control , nitrate and ammonium additions
treated with and without acetylene. KNO32N or NH4Cl2N solution
was added to make the final addition of 10μg N g - 1 in the slurry.
Each bottle was sealed with a rubber stopper and evacuated and
then refilled with pure nitrogen to ensure an anaerobic incubation
environment. Afterwards , acetylene was injected to 10 % of the
headspace of the corresponding bottles to inhibit the reduction of
N2O to N2 . Each treatment has six replicates. Samples were kept
at 15 ℃after the initial gas sampling. The incubation lasted for 7
days and gas samples were drawn every 24 hours for analysis of
N2O (in triplicate) and CO2 (in duplicate) concentrations in the
headspace of the bottles. Immediately after the gas sampling ,one
bottle of each treatment was used to analyze the nitrate , nitrite and
ammonium contents in the slurry.
212 　Sample analysis and calculation
Each soil slurry was extracted with 25ml 0. 1M KCl. After
shaking for 2 hrs , soil solution was filtered to remove soil particles.
An aquatic flow injection system was used to analyze NH+4 , NO -2
and NO -3 contents. N2O concentration was measured on a
Hewlett2Packard GC 5890 with ECD. The oven , injector and de2
tector temperatures were 40 ℃, 120 ℃ and 325 ℃, respectively ,
and the carrier gas flow rate was 20 ml N2 min - 1 . A Hewlett2
Packard GC 5890 with TCD was used for CO2 measurement with
oven , injector and detector temperatures being 90 ℃, 120 ℃ and
85 ℃, respectively , and carrier gas flow rate being 20 ml H2
min - 1 . A p H meter 28 ( Radiometer Copenhagen) was used to
measure p H value in the slurry.
Denitrification potential was determined by the rate of N2O
accumulation in C2 H22inhibited treatment . Some of the N2O pro2
duced can be reduced to N2 when there is no inhibition of N2O re2
duction by C2 H2 . Therefore N2O reduction potential was deter2
mined by subtracting the amount of N2O produced in C2 H22inhibit2
ed treatment from the N2O accumulated without C2 H2 . The
amount of N2O dissolved in water phase of the slurry was consid2
ered in the calculation[21 ] . All results were calculated based on dry
weight of the soil.
3 　Results and Discussion
311 　Soil general metabolism
Soil anaerobic respiration is experessed by CO2 pro2
duction in the experiment . The forest soil showed much
higher CO2 production activity than the agricultural
soil , and the ratio between CO2 and N2O evolution in
the forest soil was much higher than in the agricultural
soil ( Table 2 and 3) , indicating a larger portion of deni2
t rifiers in the microbial community of the agricultural
soil. There was no significant difference in CO2 produc2
tion between different t reatments of each soil , indicat2
ing that the general microbial metabolism did not differ ,
and therefore the acetylene inhibition technique was ac2
Table 2 CO2 productions (μg C g soil - 1 d - 1 , n = 8) under different treat2
ments
Agricultural soil
With C2H2 Without C2H2
Beech forest soil
With C2H2 Without C2H2
Control 3. 34 3. 29 22. 85 22198
NH +4 3. 67 3. 51 21. 23 23. 83
NO -3 3. 35 3. 47 20. 98 21178
Data are calculated by the linear regression of CO2 production during the in2
cubation. Values in column are not significantly different at 0. 05 level ( T2
test , Microsoft Excel 5. 0) .
683 应 　用 　生 　态 　学 　报 　　　　　　　　　　　　　　　　　　　11 卷
ceptable and the effects of different additions on N2O
production and reduction were comparable.
312 　N2O production and reduction potentials
Research to identify sources of N2O in soils has in2
dicated that most , if not all , of the N2O evolved from
soils is produced by biological processes and that little , if
any , is produced by chemical processes such as chemod2
enit rification[6 ] . In most cases , chemodenit rification
may be responsible for NO but not N2O production[1 ] .
Biological denit rification is considered as the only source
of N2O production in this anaerobic incubation experi2
ment. N2O production and reduction potentials were
calculated based on the first 24 hrs measurements. Ni2
t rous oxide production potential was estimated by the
rate of N2O accumulation with acetylene blockage , and
N2O reduction potential by the rate of N2O reduction to
N2 (the difference between N2O accumulations with and
without acetylene) .
Stronger denit rification activity was found in the
forest soil illust rated by higher N2O production potential
( Table 3) and faster decrease in nit rate content ( Fig. 1
(c) and 2 (c) ) . Actually the acidic environment of the
forest soil is not favorable for denit rification , because
denit rification activity is generally st ronger at neutral
conditions , and the initial nit rate content in the forest
soil slurry was lower than the agricultural soil slurry due
to the difference in original nit rate and water contents of
the sample soils. Larger denit rifier population in the
forest soil could be responsible for the st ronger denit rifi2
cation activity , and higher organic matter content in the
forest soil might partially account for the result .
Most of the N2O produced was reduced to N2 in the
agricultural soil , leading to only 10 % of the end prod2
ucts of denit rification being N2O (N2O/ N2 0. 11) , but
in the forest soil about 78 % of the end products was
N2O (N2O/ N2 3. 65) ( Table 3) . The results were ob2
tained from soils with different physical and chemical
characteristics. The p H values were 6. 64 and 3. 80 in
the agricultural and the forest soil , respectively ( Table
1) . The difference in N2O emission corresponded with
other observations that high N2O emission occurred at
low p H conditions[3 , 12 ] . In soil slurry experiments
Struwe and Kj«ler [25 ] found that the ratio of N2O/ N2
was inversely related to p H , and increased from 0. 5～
2. 8 with decreasing p H from 7. 2～4. 3. Blackmer and
Bremner [3 ] found a strong inhibitory effect of nit rate on
Table 3 Effects of nitrate and ammonium on N2O production and reduction potentials (μg N2O2N g soil - 1 d - 1)
Agricultural soil
Production Reduction N2O/ N2
Beech forest soil
Production Reduction N2O/ N2
Control 1. 63 ±0. 15a 1. 46 ±0. 14a 0. 11 ±0. 01a 1. 90 ±0. 07a 0. 41 ±0. 02a 3. 65 ±0. 02a
NH +4 2. 04 ±0. 01b 1. 30 ±0. 06a 0. 57 ±0. 07b 1. 99 ±0. 12a 0. 48 ±0. 07a 3. 15 ±0. 47a
NO -3 1. 75 ±0. 06a 1. 44 ±0. 01a 0. 22 ±0. 02c 1. 99 ±0. 13a 0. 17 ±0. 02b 10. 86 ±0. 97b
Values are expressed as means ±standard deviations of three replications. Values with the same character in column are not significantly different at 0. 05 level
( T2test , Microsoft Excel 5. 0) .
the N2O reduction under acidic conditions.
　　Both nit rate and ammonium additions slightly en2
hanced N2O productions in the agricultural soil , while
N2O reduction potentials decreased ( Table 3) . The in2
hibitory effect of nit rate on N2O reduction was an ex2
pected result . It was also found that ammonium had a
stronger stimulating effect on N2O production and
stronger inhibitory effect on N2O reduction than ni2
t rate . There was about 36 % N2O ( N2O/ N2 0. 57) in
the gaseous end products when ammonium was added
compared with only 10 % N2O (N2O/ N2 0. 11) in con2
t rol . If this finding is also true under natural conditions ,
it may have great implication for agricultural practice ,
since large areas of cultivated soil are fertilized with ure2
a. The capacity of agricultural soil to consume N2O may
be limited by the N H +4 released from urea hydrolysis.
Blackmer and Bremner[2 ] emphasized that a small reduc2
tion of the N2O consumption activity may lead to a large
increase in N2O emission. It could be expected that a
small change in the N2O/ N2 ratio would greatly influ2
ence the N2O emission from cultivated soil and also the
source inventory of N2O.
The results obtained from forest soil were differ2
ent . The ammonium addition did not show an inhibitory
effect on N2O reduction , but showed a slightly stimulat2
ing effect , which resulted in a lower N2O/ N2 ratio than
control . Both ammonium and nit rate additions enhanced
the N2O productions in the forest soil , but not as much
as in the agricultural soil. The inhibitory effect of ni2
t rate on N2O reduction was found , corresponding to the
7833 期 　　　　　　　　　　　　　　于克伟等 :农田和森林土壤中氧化亚氮的产生与还原 　　　　　　　　　
result in the agricultural soil ( Table 3) .
313 　Dynamic accumulation of N2O during the incuba2
tion
The dynamic patterns of N2O accumulation and
change of nit rate content during the incubation were il2
lust rated in Fig. 1 and 2. We didn’t find significant
amount of nit rite accumulation and ammonium content
kept constant during the anaerobic incubation. The re2
sults f rom the agricultural soil without acetylene showed
that virtually all of the N2O produced in denit rification
was reduced to N2 in one week , and this pattern was not
significantly affected even when nit rate was added. Due
to the st ronger inhibitory effect of ammonium on N2O
reduction , largest N2O accumulation was found in the
ammonium added treatment ( Fig. 1 ( a ) ) . Nit rate
st rongly inhibited N2O reduction , and this inhibition
was stronger under acidic conditions of the forest
soil [24 ]. Only when nit rate was almost undetectable in
the forest soil slurry and denit rification nearly stopped
after incubation for 3 days , the consumption of N2O in
the headspace occurred ( Fig. 2) . The result indicates
that the capacity of the forest soil to consume N2O
greatly depended on nit rate content . It has been pro2
posed that the development of nit rous oxide reductase
requires more time[19 ] . The duration of this lag increas2
es with NO -3 concentration[9 , 17 ] .
　　The agricultural soil showed higher N2O reduction
potential than the forest soil , and the capacity depended
on several environmental parameters. The microorgan2
isms responsible for N2O consumption are widely pre2
sent in different ecosystems. Kromka et al . [18 ] observed
that 80～90 % of the denit rifiers were able to reduce
N2O in their study. Okereke[22 ] reported that 59 out of
71 denit rifiers isolated from eight different countries uti2
lized N2O as terminal electron acceptor. A suitable re2
duced environment can be found in flooded areas or
deeper layers of soil. Observations on consumption of
atmospheric N2O have been reported from field mea2
surements of cultivated soils[5 , 28 ] , grasslands[7 , 11 , 23 ]
and tropical soils[16 , 20 ] . Ryden[23 ] reported that an un2
fertilized control grassland plot in U K served as a sink of
atmospheric N2O during the period from August to
November when nit rate content in the soil was low and
soil water content exceeded 2 0 % , which corresponded
Fig. 1 Accumulation of nitrous oxide and change of nitrate content in the a2
gricultural soil slurry.
Vertical bars represent the standard deviations of means and can not be seen
when they are smaller than the symbols of points , n = 3 for (a) and (b) , n
= 2 for (c) . Points in (c) represent means of nitrate contents in the slurries
with and without acetylene inhibition , based on the assumption that there is
no effect of acetylene on denitrification in slurries except the inhibition of
N2O reduction to N2. Ⅰ. Control , Ⅱ. NH +4 , Ⅲ. NO -3 The same below.
to the results of this experiment . Similar results were
found in a bare field ( cultivated soil with no vegeta2
tion) , which showed a continuous consumption of ambi2
ent N2O from September to November , and this con2
sumption was close to the amount of N2O emitted in the
remaining months of the year [28 ] .
Reduction of N2O in soil is probably only a minor
sink , but may still play an important role on a global
scale . The elimination of N2 O in the stratosphere is so
slow that even a small sink could contribute significant2
ly to reduction of the atmospheric residence time of
N2O [8 ,14 ] . Further studies are needed to evaluate the
importance of this sink on a global scale. The poten2
tialof soil as a sink for atmospheric N2O deserves atten2
883 应 　用 　生 　态 　学 　报 　　　　　　　　　　　　　　　　　　　11 卷
Fig. 2 Accumulation of nitrous oxide and change of nitrate content in the
forest soil slurry.
Explanations are the same as in Fig. 1 , but the scales of y2axis in (a) and
(b) are different from Fig. 1.
tion in future attempts to estimate the atmospheric N2O
budget .
Acknowledgments
The authors thank Karin Vestberg , Department of General
Microbiology , University of Copenhagen , for her skillful technical
assistance.
References
1 　Anderson IC and Poth MA. 1998. Controls on fluxes of trace gases
from Brazilian cerrado soils. J Envi ron Qual ,27 (5) : 1117～1124
2 　Blackmer AM and Bremner J M. 1976. Potential of soil as a sink for
atmospheric nitrous oxide. Geophys Res Lett ,3 : 739～742
3 　Blackmer AM and Bremner J M. 1978. Inhibitory effect of nitrate on
reduction of N2O to N2 by soil micro2organisms. Soil Biol Biochem ,
10 : 187～191
4 　Blackmer AM and Bremner J M. 1978. Stimulatory effect of nitrate on
reduction of N2O to N2 by soil micro2organisms. Soil Biol Biochem ,
10 : 313～315
5 　Bremner J M , Robbins SG and Blackmer AM. 1980. Seasonal variabil2
ity in emission of nitrous oxide from soil. Geophys Res Lett , 7 : 641～
644
6 　Bremner J M. 1997. Sources of nitrous oxide in soils. N ut r Cycl A2
groecosys ,49 : 7～16
7 　Cicerone RJ , Shetter JD , Stedman DH et al . 1978. Atmospheric
N2O : Measurements to determine its sources , sinks , and variations. J
Geophys Res , 83 : 3042～3050
8 　Cicerone RJ . 1989. Analysis of sources and sinks of atmospheric ni2
trous oxide (N2O) . J Geophys Res ,94 : 18265～18271
9 　Colbourn P and Harper IW. 1987. Denitrification in drained and
undrained arable clay soil. J Soil Sci ,38 : 531～540
10 　Crutzen PJ . 1981. Atmospheric chemical processes of the oxides of ni2
trogen including nitrous oxide. In : Delwiche CC ed. Denitrification ,
Nitrification and Nitrous Oxide . Wiley , New York. pp . 17～44
11 　Donoso L , Santana R and Sanhueza E. 1993. Seasonal variation of
N2O fluxes at a tropical savannah site : soil consumption of N2O during
the dry season. Geophys Res Lett ,20 : 1379～1382
12 　Firestone M K , Firestone RB and Tiedje J M. 1980. Nitrous oxide
from soil denitrification : Factors controlling its biological production.
Science ,208 : 749～751
13 　Hutchinson GL and Davidson EA. 1993. Processes for production and
consumption of gaseous nitrogen oxides in soil. In : Agricultural E2
cosystem Effect on Trace Gases and Global Climate Change. ASA Spe2
cial Publication 55. 79～93
14 　Intergovernmental Panel on Climate Change ( IPCC) . 1992. Climate
Change 1992 : The supplementary report to the IPCC scientific assess2
ment ( Houghton J T , Callander BA and Varney SK eds. ) , Cam2
bridge , U K.
15 　Intergovernmental Panel on Climate Change ( IPCC) . 1994. Climate
Change 1994 : Radiative forcing of climate change and an evaluation of
the IPCC IS92 emission scenarios ( Houghton J T , Meira Filho L G ,
Bruce J eds. Cambridge , U K.
16 　Keller M , Kaplan WA and Wofsy SC. 1986. Emissions of N2O , CH4
and CO2 from tropical forest soils. J Geophys Res ,91 : 11791～11802
17 　Kroeze C , Van Faassen HG and De Ruiter PC. 1989. Potential deni2
trification rates in acid soils under pine forest . Neth J A gric Sci , 37 :
345～354
18 　Kromka M , Stepanov AL and Umarov MM. 1992. Reduction of ni2
trous oxide by the microbial biomass in soils. Eurasian Soil Sci , 24 :
68～74
19 　Letey J , Valoras N , Hadas A et al . 1980. Effect of air filled porosi2
ty , nitrate concentration , and time on the ratio of N2O/ N2 evolution
during denitrification. J Envi ron Qual ,9 : 227～231
20 　Matson P and Vitousek PM. 1987. Cross2system comparisons of soil
nitrogen transformation and nitrous oxide flux in tropical forest ecosys2
tems. Global Biogeochem Cycles ,1 : 163～170
21 　Moraghan J T and Buresh R. 1977. Correction for dissolved nitrous
oxide in nitrogen studies. Soil Sci Soc A m Proc ,41 : 1201～1202
22 　Okereke GU. 1993. Growth yield of denitrifiers using nitrous oxide as
a terminal electron acceptor. World J Microbiol Biotech ,9 : 59～62
23 　Ryden J C. 1981. N2O exchange between a grassland soil and the at2
mosphere. Nat ure ,292 : 235～237
24 　Smith CJ , Wright MF and Patrick J r WH. 1983. The effect of soil
redox potential and p H on the reduction and production of nitrous ox2
ide. J Envi ron Qual , 12 : 186～188
25 　Struwe S and Kj«ler A. 1994. Potential for N2O production from
beech ( Fagus silvaticus ) forest soils with varying p H. Soil Biol
Biochem ,26 : 1003～1009
26 　Terry RE and Tate III RL . 1980. The effect of nitrate on nitrous ox2
ide reduction in organic soils and sediments. Soil Sci Soc A m J , 44 :
744～746
27 　Weiss RF. 1981. The temporal and spatial distribution of tropospheric
nitrous oxide. J Geophy Res ,86 : 7185～7195
28 　Yu K2W (于克伟) , Chen G2X (陈冠雄) , Yang S2H (杨思河) et
al . 1995. Role of several upland crops in N2O emissions from farm2
lands and its response to environmental factors. Chin J A ppl Ecol (应
用生态学报) ,6 (4) : 387～391 (in Chinese)
Author introduction 　Kewei Yu , born in 1964 , associate research
professor , major research in biogenic greenhouse gases and nutrient
cycling in terrestrial ecosystem. He has more than 30 publications
in Chinese and English. He is currently studying at Louisiana State
University , USA for Ph. D , E2mail : kewei. yu @mailcity. com
9833 期 　　　　　　　　　　　　　　于克伟等 :农田和森林土壤中氧化亚氮的产生与还原