HAL Id: hal-00885944
https://hal.science/hal-00885944
Submitted on 11 May 2020
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of sci-
entic research documents, whether they are pub-
lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diusion de documents
scientiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
The response of soil nitrogen and 15N natural
abundance to dierent amendments in a long-term
experiment at Ultuna, Sweden
Martin H. Gerzabek, Holger Kirchmann, Georg Haberhauer, Friedrich
Pichlmayer
To cite this version:
Martin H. Gerzabek, Holger Kirchmann, Georg Haberhauer, Friedrich Pichlmayer. The response
of soil nitrogen and 15N natural abundance to dierent amendments in a long-term experiment at
Ultuna, Sweden. Agronomie, 1999, 19 (6), pp.457-466. �hal-00885944�
Original
article
The
response
of
soil
nitrogen
and
15
N
natural
abundance
to
different
amendments
in
a
long-term
experiment
at
Ultuna,
Sweden
Martin
H.
Gerzabek
Holger
Kirchmann
b
Georg
Haberhauer
a
Friedrich
Pichlmayer
a
a
Austrian
Research
Centers,
2444
Seibersdorf,
Austria
b
Swedish
University
of
Agricultural
Sciences,
Department
of
Soil
Sciences,
Box
7014,
750 07
Uppsala,
Sweden
(Received
10
November
1998;
accepted
20
May
1999)
Abstract -
In
a
long-term
field
experiment
on
a
clay
loam
soil
at
Uppsala,
Sweden,
changes
in
nitrogen
contents,
natur-
al abundance
of
15
N
in
the
topsoils
receiving
various
organic
amendments
at
the
rate
of
2 000
kg
C
ha-1
year
-1
and
dif-
ferent
rates
of
nitrogen
mineralisation
were
studied.
Cropping
resulted
in
clearly
lower
N
losses
from
the
topsoil
(0-20
cm)
compared
to
the
bare
fallow
plots.
Green
manure,
animal
manure
and
sewage
sludge
increased
the
Nt
inven-
tory
significantly;
for the
latter
two
treatments
an
equilibrium
of
N
input
and
N
output
was
not
reached
within
35
years
of
application.
The
highest
N
efficiency
with
respect
to
plant
uptake
and
the
highest
N
losses
were
recorded
for
the
min-
eral
N-treated
plots.
15
N
abundances
were
clearly
affected
by
N
input,
differing
in
isotopic
signature
introduced
through
the
amendments.
This
effect
was
used
to
derive
the
relative
input
of
’stable’
N
from
green
manure
and
peat
in
compari-
son
with
the
difference
method.
(©
Inra/Elsevier,
Paris.)
fertiliser
/
isotopic
abundance
/
long-term
experiment
/
nitrogen
/
manure
Résumé -
Réponse
de
l’azote
du
sol
et
de
l’abondance
naturelle
de
15
N
à
différents
amendements
organiques
dans
une
expérience
à
long
terme
à
Ultuna,
Suède.
Une
expérience
à
long
terme
a
été
conduite
sur
un
sol
limono-
argileux
à
Uppsala,
Suède.
Les
modifications
de
la
teneur
en
azote
et
de
l’abondance
naturelle
de
15
N
dans
la
couche
superficielle
ont
été
étudiées
en
fonction
des
apports
de
différents
amendements
organiques
correspondant
à
2
000
kg
C
ha-1
·an
-1
avec
différents
taux
de
minéralisation
de
l’azote.
Sous
culture
il
apparaît
clairement
une
plus
faible
perte
en
azote
de
la
couche
superficielle
comparativement
à
un
sol
nu
non
cultivé.
Les
apports
d’engrais
vert,
de
fumier
et
de
boues
de
station
d’épuration
ont
accru
significativement
la
teneur
totale
en
azote
Nt
et
pour
les
deux
premiers
traite-
ments
l’équilibre
entre
les
apports
et
les
pertes
d’azote
n’a
pas
été
atteint
au
cours
de
35
ans
d’application.
La
meilleure
efficience
de
l’azote
en
ce
qui
concerne
l’extraction
par
la
plante
et
la
perte
la
plus
élevée
furent
enregistrées
pour
les
parcelles
qui
recevaient
un
apport
d’azote
minéral.
L’abondance
de
15
N
a
été
clairement
affectée
par
les
types
d’apport
Communicated
by
Marco
Trevisan
(Piacenza,
Italy)
*
Correspondence
and
reprints
martin.gerzabek @ arcs.ac.at
d’amendements
dont
les
signatures
isotopiques
étaient
différentes.
Cette
particularité
a
été
utilisée
pour
déduire
l’apport
relatif
d’azote
«stable»
à
partir
d’un
engrais
vert
et
de
la
tourbe
comparativement
à
la
méthode
différentielle.
(©
Inra/Elsevier,
Paris.)
amendement
/
abondance
isotopique
/
expérience
à
long
terme
/
azote
/
fumure
organique
1.
Introduction
Nitrogen
is
one
of
the
most
important
plant
nutrients.
The
major
sources
of
N
supply
to
plants
are
i)
the
mineralisation
of
soil
nitrogen,
ii)
the
application
of
mineral
or
organic
fertilisers,
iii)
the
wet
and
dry
deposition
of
N
compounds,
and
iv)
the
release
of
nitrogen
from
residues
of
N
fixing
plants.
The
mineralisation
of
soil
nitrogen
is
influ-
enced
by
the
same
factors
which
govern
soil
organ-
ic
matter
build-up
and
turnover.
Thus,
climate,
soil
texture,
application
of
fertilisers
and
manures
[16]
and
different
management
practices
[2]
are
impor-
tant
factors
in this
respect.
Quantification
of
N
release
from
manures
and
the
soil
N
pool
remains
difficult,
although
isotopic
methods
offer
some
interesting
options
[3].
Another
question,
which
is
today
considered
to
be
of
high
relevance,
concerns
the
losses
of
nitrogen
from
fertiliser
and
manure.
These
losses
can
affect
both
the
groundwater
by
increasing
the
nitrate
concentration
and
the
atmos-
phere
by
the
release
of
greenhouse
gases
[20].
Short-term
studies
on
SOM
turnover
and
build-
up
are
not
feasible
owing
to
the
high
variability
introduced
by
environmental
factors.
Quantification
of
input
and
output
needs
longer
observation
periods;
long-term
experiments,
there-
fore,
seem
to
be
ideal
for
evaluating
effects
of
agri-
cultural
practice
on
soil
carbon
and
nutrients
[15].
The
natural
abundance
of
15
N
in
soil
can
be
altered
by
several
factors.
Besides
changes
due
to
microbial
activity
in soil
[3],
differences
in
the
veg-
etation
cover
and
tillage
[19]
and
the
long-term
application
of
N
fertilisers
and
organic
manures
can
influence
15
N
abundances.
The
latter
effect
may
be used
to
quantify
the
stable
portion
of
nitro-
gen
after
a
certain
time
period
of
continuous
manure
application
[21].
The
Ultuna
organic
matter
experiment,
started
in
1956,
was
designed
to
study
soil
organic
matter
and
structural
changes
in
soils
under
a
range
of
manurial
treatments
and
inorganic
fertilisers.
The
same
amount
of
carbon
(on
average
2
000
kg
C
ha-1
year
-1
)
has
been
applied
through
a
range
of
organic
amendments.
The
experiment,
therefore,
enables
us
to
compare
changes
in
both
amount
and
composition
of
soil
humus
as
a
result
of
applica-
tion
of
different
organic
manures.
In
previous
pub-
lications
the
impact
of
organic
manures
on
soil
aggregate
stability
[4],
the
sulphur
balance
[11]
and
on
soil
organic
matter
including
humification
[5]
were
reported.
In
this
paper,
changes
in
the
nitrogen
contents
of
topsoils
due
to
long-term
application
of
different
manures
and
their
effect
on
the
natural
abundance
of
15
N
are
described.
Nitrogen
contents
in
the
topsoil
are
explained
by
a
nitrogen
balance.
Half-lives
of
fertiliser
N
derived
from
the
difference
method
are
compared
with
the
results
of
the
isotopic
method.
2.
Materials
and
methods
2.1.
Site
and
treatments
The
field
experiment
is
located
in
central
Sweden,
Uppsala,
on
a
Eutric
Cambisol
(FAO)
with
37 %
clay
and
41
%
silt.
The
parent
material
consists
of
postglacial
clay
with
illite
as
the
main
clay
mineral.
A
complete
documentation
of
the
experiment
and
compilation
of
data
can
be found
in
Kirchmann
et
al.
[10].
In
1956
the
soil
(0-20
cm
deep)
had
15
g
kg-1
of
organic
carbon,
1.7
g
kg-1
of
nitrogen
and
a
pH
of
6.6.
The
experimental
design
consists
of
14
treatments,
laid
out
in
duplicate
in
a
randomised
block
design.
The
individual
plots
(2
m
x
2
m),
were
separated
by
pressure-treated
wooden
frames.
Five
of
the
treatments,
fallow,
no-N,
green
manure,
animal
manure
(well
decomposed)
and
peat
(sphagnum)
were
selected
for intensive
measurements
in
this
study.
Additional
treatments
(sewage
sludge,
Ca(NO
3)2
and
(NH
4)2
SO
4)
were
evaluated
only
in
1991.
The
application
of
organic
amendments,
analysed
before
use,
was
based
on
equal
amounts
of
ash-free
organic
matter
amounting
to
2 000
kg
C
ha-1
year
-1
on
average.
Organic
matter
was
added
in
1956,
1960,
1963
and
thereafter
every
second
year
by
hand.
All
plots
received
a
dressing
of
20
kg
P
ha-1
in
the
form
of
super-
phosphate
and
35-38
kg
K
ha-1
in
the
form
of
potassi-
um
chloride
annually
in
spring.
Cereals
(70
%),
rape
crops
(25
%)
and
fodder
beet
(5
%)
were
grown
alter-
nately.
No
N-fixing
plants
were
included
in
the
crop
rotation.
At
harvest
the
above-ground
portion
of
the
crop
was
completely
removed.
Topsoil
samples
(0-20
cm)
were
taken
in
autumn
before
the
organic
mat-
ter
was
added.
They
were
air-dried,
passed
through
a
2-mm
sieve
and
then
stored.
2.2.
Analysis
Total
nitrogen,
organic
carbon
and
δ
15
N
were
mea-
sured
in
soil
samples
(1956,
1967,
1975,
1977, 1979,
1981,
1983,
1985,
1987,
1989,
1991,
1993,
1994),
and
in
the
green
manure,
animal
manure,
peat
(1975,
1979)
and
spring
wheat
or
rye
(1987,
1989,
1991)
using
an
elemental
analyser
coupled
to
a
mass
spectrometer
as
described
by
Pichlmayer
et
al.
[14].
Finely
powdered
samples
were
burned
in
an
elemental
analyser
(Carlo
Erba
Nitrogen
Analyser
1500).
The
released
N2
was
separated
by
a
gas-chromatographic
column
from
the
other
gases
and
introduced
via
a
gas-splitting
interface
into
the
mass
spectrometer
(Finnigan
MAT
251).
The
relative
abundance
of
the
stable
isotopes
of
nitrogen
(
14N,
15N)
was
measured
against
the
air
as
standard.
The
results
were
calculated
as
δ
values
as
follows:
where
R
samp
is
15N/14
N -
ratio
of
sample
and
R
standard
is
15N/14
N -
ratio
of
standard
(air).
2.3.
Calculation
of
N-amounts
in
the
topsoil
and
the
N-balance
Nitrogen
concentrations
in
the
soil
were
converted
to
soil
N
amounts
by
the
following
steps.
Bulk
density
of
the
soil
was
determined
for
each
treatment
in
1956,
1975
and
1991,
and
values
were
interpolated
using
lin-
ear
regression
to
give
yearly
values.
The
bulk
densities
were
used
to
calculate
the
mass
of
soil
in
the
top
0-20
cm.
The
inorganic
portion
of
the
soil,
i.e.
the
whole
soil
minus
the
organic
matter
of
the
soil,
was
assumed
to
be
constant
during
the
experiment,
and
the
amount
of
inorganic
soil
in
the
0-20-cm
layer
in
1956
was
used
as
a
reference
value.
If
the
bulk
density
decreased,
i.e.
the
soil
volume
increased,
soil
from
below
a
depth
of
20
cm
had
to
be
added
to
the
0-20-cm
layer
to
give
the
same
amount
of
inorganic
soil
as
in
1956.
Soils
between
20
and
25
cm
were
sampled
in
1991
and
N
concentrations
were
analysed.
Because
con-
centration
of
N
in
the
subsoil
(20-25
cm)
of
the
differ-
ent
treatments
was
similar,
we
assume
that
the
N
con-
centration
in
this
layer
did
not
change
significantly
over
time.
The
amount
of
N
in
the
quantity
of
subsoil
required
to
give
the
same
mass
of
inorganic
soil
as
in
1956
was
added
to
the
amount
of
N
present
in
the
0-20-
cm
layer.
The
nitrogen
balance
was
calculated
using
the
following
formula:
where
N1
is
the
annual
loss
of
nitrogen
from
the
topsoil
(kg
ha-1),
Nf
is
the
fertiliser
N
input
(kg
ha-1
year
-1),
Nd
is
the
annual
wet
and
dry
deposition
(kg
ha-1
,
according
to
[12]),
Nb
is
the
non-symbiotic
biologically
fixed
N
(kg
ha-1
year
-1),
ΔN
s
is
the
annual
change
in
the
nitro-
gen
inventory
in
the
topsoil
(kg
ha-1
)
and
N
is
the
plant
uptake
(kg
ha-1).
The
form
of
N
storage
in
soil
does
not
at
all
affect the
overall
N
balance,
which
can
only
describe
input
and
output.
Interlayer
ammonium,
there-
fore,
as
well
as
microbial
biomass
has
not
been
consid-
ered.
Nb
was
obtained
using
data
on
the
relative
poten-
tial
N
fixing
activity
of
the
respective
treatment
in
a
previous
experiment
[13]
related
to
the
maximum
activi-
ty
observed
in
the
animal
manure
treatment.
Absolute
amounts
were
calculated
by
multiplying
the
relative
activities
with
15
kg
N
ha-1
year
-1
,
which
was
assumed
to
be
the
maximum
amount
of
N
fixed
under
the
prevail-
ing
climatic
conditions
[6].
This
is
slightly
lower
than
reported
by
Jenkinson
[9]
from
the
Broadbalk
experi-
ment
in
Rothamsted.
2.4.
Calculation
of
N
fractions
and
N
turnover
To
describe
turnover
rates
of
nitrogen
in
the
experi-
mental
plots
a
simple
first-order
kinetic
model
was
applied:
where
Nr
is
the
remaining
nitrogen
in
soil
of
the
yearly
N
input
(N
i
=
Nf
+
Nd
+
Nb
).
Integration
of
equation
(3)
gives
the
exponential
rate
of
decomposition
for
a
single
fertiliser
application,
where λ
is
ln2
divided
by
the
turnover
half-life:
A
continuous
N
fertiliser
input
can
be
described
by
equation
(5):
Integration
of
equation
(5)
from
0
to
time
t
yields
equation (6):
Equation
(6)
was
used
to
calculate
turnover
half-lives
of
the
fertiliser
nitrogen
fractions
in
soil.
Assuming
that
Nr
is
approximated
by
the
difference
(N
diff
)
of
the
N
content
of
the
amended
soil
minus
the
N
content
of
the
no-N
soil,
Ni
is
approximated
by
the
difference
in
the
N
input
of
the
amended
soil
minus
the
N
input
of
the
no-N
soil
and
t
is
35
years,
all
parameters
are
known
except
the
decomposition
rate
constant
(λ). λ
can
be
deter-
mined
using
an
iterative
method.
Assuming
equilibration
of
the
system
(t
→
∞,
e
-λt
→
0),
that
means
N
input
equals
N
output,
the
turnover
half-lives
can
be
directly
calculated:
The
nitrogen
content
of
the
fertiliser
in
soil
derived
from
isotopic
data
was
calculated
adopting
the
proce-
dure
described
by
Puget
et
al.
[17]:
where
δ
s
is
δ
15
N
value
(‰)
of
the
fertiliser-treated
soil,
δ
Ref.
is
δ
15
N
value
(%c)
of
the
no-N
soil,
δ
Fert
is
δ
15
N
value
(%o)
of
the
applied
fertiliser,
NT
is
the
total
nitro-
gen
content
in
the
topsoil
and
N
fr
is
the
nitrogen
fraction
originating
from
the
fertiliser
nitrogen.
A
slightly
over-
estimated
uncertainty
of
N
fr
,
e
fr
was
obtained
using
the
following
equation
proposed
by
Puget
et
al.
[17]:
where
es,
e
Ref
,
e
Fert
are
the
standard
deviations
of
the
respective
δ
values
and
eT
is
the
standard
deviation
of
the
NT
measurement.
Turnover
half-lives
of
nitrogen
derived
from
fertiliser
were
calculated
from
isotopic
data
using
equations
(6)
(35
years)
and
(7)
(assuming
equilibration),
where
N
diff
is
the
amount
of
nitrogen
in
the
soil
derived
from
equa-
tion
(8)
(=
N
fr
)
and
Ni
=
Nf.
3.
Results
and
discussion
3.1.
Nitrogen
in
topsoils
and
N
balance
Soil
characteristics
in
the
experimental
plots
were
distinctly
altered
by
the
different
treatments
(table
I).
The
application
of
peat,
sewage
sludge
and
ammonium
sulphate
resulted
in
a
significant
decrease
in
the
pH
values,
which
has
already
been
discussed
in
detail
in
Kirchmann
et
al.
[11].
Both
organic
carbon
and
Nt
contents
were
clearly
affect-
ed
by
the
fertiliser
and
manure
applications
(table I,
figure
1).
The
bare
fallow
plot
exhibited
a
remark-
able
decrease
in
nitrogen
contents
as
compared
with
1956
(0.16 %
Nt)
ending
up
with
0.11
%
in
1994.
Statistical
analysis
indicated
that
the
highly
significant
decrease
follows
a
linear
regression
(N
t
=
-0.0014
year
+
2.8081;
R2
=
0.969
***),
equilibrium
between
N
mineralisation
of
the
soil
N
pool
and
N
deposition
has,
therefore,
not
yet
been
reached
in
the
bare
fallow
plots.
The
same
result
was
obtained
for
the
no-N
plots,
although
the
N
content
of
the
topsoil
remained
on
a
higher
level
as
compared
to
the
fallow
plots
(%N
t
=
-0.0008
year
+
1.7914;
R2
=
0.8091
***),
despite
the
significant
removal
of
N
with
the
harvest
(table
II).
Green
manure-treated
plots
did
not
show
any
significant
changes
in
Nt
contents
in
the
topsoil.
This
is
sup-
ported
by
the
identical
behaviour
of
C
org
contents
in
this
treatment
[5].
Green
manure was
obviously
able
to
keep
the
Nt
concentrations
in
the
topsoil
constant.
Peat,
animal
manure
and
sewage
sludge
increased
the
Nt
contents
of
topsoils
significantly
(table
I,
figure
1).
The
effect
of
peat -
due
to
the
small
amount
of
N
introduced
through
peat
amend-
ments -
was
quite
small.
None
of
the
five
inten-
sively
investigated
treatments -
except
green
manure,
where
no
trend
was
observed -
showed
a
non-linear
trend
of
N
contents
in
the
topsoil.
This
indicates
that
a
much
longer
time-period
is
needed
for
equilibration
than
an
observation
period
of
35-38
years.
This
is
supported
by
the
results
of
the
Broadbalk
experiment
in
Rothamsted,
where
the
increase
in
Nt
contents
in
farmyard
manure-treated
plots
started
to
level
off
approximately
50-60
years
after
the
start
of the
experiment
[15].
The
mineral
fertilisers,
Ca(NO
3)2
and
(NH
4)2
SO
4
maintained
the
Nt
concentrations
in
topsoils,
but
not
the
total
N
inventory
when
taking
into
account
soil
bulk
density
changes
(table
II).
In
general,
only
green
manure,
animal
manure
and
sewage
sludge
applications
resulted
in
an
increase
in
total
nitrogen
present
in
the
mineral
soil
mass
(0-20
cm)
on
the
basis
of
the
year
1956.
The
high-
est
N
losses
calculated
according
to
equation
(2)
were
recorded
in
the
bare
fallow
plots
(table
II).
N
losses
were
smallest
for
the
no-N
plots.
The
miner-
alisation
of
soil
N
(34
kg
N
ha-1
year
-1
)
and
the
N
input
by
deposition
and
fixation
of
approximately
17
kg
N
ha-1
year
-1
were
only
somewhat
exceeding
N
removal
by
harvest.
N
losses
from
the
other
treatments
showed
the
following
ranking:
green
manure >
calcium
nitrate >
ammonium
sul-
phate
>
animal
manure >
peat >
sewage
sludge.
Taking
into
account
total
N
input
alters
the
ranking
(table
II),
peat
being
first
followed
by
green
manure
and
calcium
nitrate.
Losses
from
sewage
sludge
plots
were
remarkably
low,
although
the
N
input
was
highest
amongst
all
treatments.
Considering
again
the
no-N
treatment
as
refer-
ence,
it
was
feasible
to
estimate
the
average
fertili-
sation
effect
of
the
different
treatments
(N
f
+
Nb)
on
agricultural
crops
and
removed
with
the
harvest.
Leaving
out
peat,
which
had
a
negative
effect
on
N
removal
by
plants,
the
following
ranking
was
observed:
ammonium
sulphate
(57
%)
>
calcium
nitrate
(55
%)
>
sewage
sludge
(40
%)
>
green
manure
(37
%)
>
animal
manure
(24
%).
The
avail-
ability
of
N
introduced
by
peat,
or
even
the
induced
N
immobilisation
by
this
treatment
makes
it
clear
that
peat
adds
stable
N
forms
to
soil
but
cannot
be
considered
as
a
N
fertiliser.
The
N
recov-
ery
from
mineral
fertilisers
plus
fixation
reached
the
same
magnitude
as
reported
for
fertiliser
utili-
sation
by
several
authors
from
15
N
labelling
experi-
ments
although
these
are
not
directly
comparable
with
results
obtained
by
the
difference
method.
A
literature
overview
by
Bhogal
et
al.
[1]
showed
a
range
of
46.6-61
%
N
from
fertiliser
recovered
by
wheat
or
barley
(median:
58.9
%,
n
=
14).
The
lower
average
N
recovery
from
organic
amend-
ments
in
our
experiment
was
not
due
to
larger
N
losses
from
mineral
fertilisers
but
to
the
supply
of
more
stable
N
fractions
by
the
manure,
which
sub-
stituted
the
N
amount
mineralised
from
native
soil
N
and
on
top
of
that
increased
the
N
inventory.
3.2.
15
N
abundances
The
natural
abundance
of
15
N
varied
distinctly
with
time
(figure
2),
while
changes
seemed
to
have
a
general
pattern
and
were
followed
by
most
treat-
ments.
One
reason
for
that
could
be
the
generally
considerable
variation
of
δ
15
N
in
deposited
N
with
time
and
space
[8].
Compared
with
the
δ
15
N
values
at
the
beginning
of
the
experiment
in
1956
(8.4-8.9
‰)
there
was
a
clear
change,
which
could
be
related
to
the
amendments
(figure
2).
All
miner-
al
fertilisers
and
organic
materials
except
animal
manure
exhibited
significantly
smaller
δ
15
N
values
than
the
initial
soil
N
pool.
This
was
well
reflected
in
soil
samples
after
different
time
periods
(table
I,
figure
2).
Quantification
of
this
quite
obvious
effect,
however,
was
only
feasible
for
green
manure
and
peat
because
of
the
larger
data
set
available.
Using
equation
(8)
and
averaging
over
5-7
years
it
was
possible
to
estimate
the
contribu-
tion
of
fertiliser
N
to
Nt
in
the
topsoil
using
the
no-
N
treatment
as
reference
(table
III).
A
reference
treatment
is
necessary
in
such
cases
to
take
into
account
changes
in
the
natural
abundance
of
15
N
due
to
other
factors
such
as
changes
in
the
isotopic
composition
of
the
atmospheric
input
or
isotopic
shifts
caused
by
microbial
activity
[3].
Uncertainties,
nevertheless
were
quite
high
owing
to
the
small
isotopic
differences
between
reference
plots
and
treatments
and
the
variation
in
values
from
sampling
to
sampling.
On
the
other
hand,
it
was
possible
to
support
the
consistency
of
the
data
by
a
highly
significant
linear
correlation
between
experimental
years
and
contribution
of
peat
N and
manure
N
to
Nt
in
the
topsoil.
The
calculated
N
portion
from
green
manure
or
peat
can
be
consid-
ered
as
an
already
stable
N
fraction.
The
stable
N
input
of
green
manure was
approximately
22
kg
ha-1
year
-1
(versus
39
kg
N
ha-1
year
-1
according
to
the
difference
method),
which
is
20 %
of
the
total
N
input
through
this
treatment.
The
respective
values
for
peat
were
31
kg
ha-1
year
-1
(versus
27
kg
N
ha-1
year
-1
)
and 102
%.
The
latter
result
indicated
that
N
contained
in
peat
was
just
stored
in
the
soil
and
did
not
contribute
to
the
plant
uptake
or
losses,
both
of
which
were
therefore
derived
from
mineralisation
of
native
soil
N and
wet
and
dry
deposition
of
nitrogen.
This
result
is
corroborated
by
the
high
C/N
ratio
of
peat
of
approximately
70.
This
low
decomposability
was
already
mentioned
as
one
of
the
main
reasons
for
organic
matter
piling
up
in
the
peat
plots
[5]
result-
ing
in
a
seven
times
lower
microbial
biomass
relat-
ed
to
the
C
org
content
of
the
soil
as
compared
to
the
other
treatments
[22].
δ
15
N
values
of
selected
plant
samples
were
determined
and
ranged
from
1.8
to
10.4
‰.
The
well-known
effect
of
isotopic
shifts
due
to
selective
N
uptake
of
N
compounds
of
different
isotopic
sig-
nature
by
plant
roots
from
soil
[8]
did
not
allow
quantification
of
the
contribution
of
fertiliser
or
manure
N
to
N
uptake
into
crops.
3.3.
Turnover
half-lives
Table
IV
shows
the
increase
in
soil
N
inventory
in
relation
to
the
no-N
reference
plots.
According
to
equations
(6)
and
(7)
it
is
possible
to
use
the
dif-
ference
between
reference
plots
and
treatments
to
estimate
the
turnover
half-lives
of
the
additional
N
input
through
treatments
(mainly
applied
fertiliser
N).
The
shortest
half-lives
(HL)
were
derived
for
the
mineral
fertilisers,
the
longest
HL
for
the
hard-
ly
decomposable
organic
substance
peat.
Half-lives
of
the
two
mineral
fertilisers
did
not
differ.
Therefore,
processes
which
might
influence
the
N
turnover
of
differing
N
sources
(NO
3,
NH
4
),
such
as
ammonium
fixation
in
interlayers
of
clay
miner-
als
for
example,
were
of
no
significant
impact.
This
result
is
in
line
with
other
field
experiments
conducted
in
Sweden
[18].
Green
manure
exhibited
the
shortest
half-life
of
nitrogen
input
amongst
all
organic
amendments.
In
some
cases
(peat,
animal
manure,
sewage
sludge)
a
distinct
difference
between
the
iterative
calculation
of
the
HL
(equa-
tion
(6))
and
the
method
assuming
equilibrium
(equation
(7))
occurred,
the
latter
values
being
shorter
than
the
former.
This
is
a
strong
indication
that
equilibrium
has
not
yet
been
reached
for
these
treatments.
A
very
large
difference
was
obtained
for
peat
N,
which
leads
to
the
conclusion
that
the
peat
N,
in
particular,
is
highly
stable
and
exhibited
a
more
or
less
inert
behaviour
during
the
time
span
of
the
experiment.
Taking
into
account
the
isotopic
changes
the
peat
and
green
manure
plots
led
to
deviating
results
concerning
half-lives.
In
the
case
of
peat
we
observed
an
even
longer
half-life
than
with
the
difference
method.
This
matches
with
the
fact
that
peat
N
was
not
lost
at
all
during
the
exper-
imental
period
(table
III).
The
green
manure
N
behaved
exactly
the
reverse.
The
isotopic
method
showed
a
considerably
shorter
HL
for
green
manure
nitrogen
already
in
equilibrium.
The
calcu-
lated
HL
of
5
years
is
much
closer
to
the
value
observed
for
15
N-labelled
grass/clover
residues
of
≈2
years
after
a
single
application
in
New
Zealand
(calculated
from
[7]).
This
implies
that
the
observed
positive
difference
in
the
N
inventory
in
the
green
manure-treated
plots
compared
to
the
no-
N
reference
plots
is
only
partly
due
to
input
of
sta-
ble
green
manure
N.
In
addition
it
seems
that
applying
green
manure
considerably
decreased
the
mineralisation
of
the
original
soil
N.
A
similar
effect
can
be
assumed
for
the
mineral
fertiliser
nitrogen
because
the
half-life
values
obtained
by
the
difference
method
for
these
treatments
seem
to
be
too
high.
4.
Conclusion
Green
manure,
animal
manure
and
sewage
sludge
increased
the
amount
of
N
present
in
the
0-20-cm
topsoil
layer
on
the
basis
of
constant
min-
eral
soil
mass
since
1956.
N
losses
were
highest
for
mineral
fertilisers
and
green
manure
and
very
low
for
the
no-N
plots.
Mineral
N
fertilisers
exhibited
the
highest
N
use
efficiency.
The
15
N
abundance
of
soil
was
altered
by
fer-
tiliser
and
manure
N
input.
The
’stable’
N
portion
could
be
deduced
for
green
manure
(20
%
of
input)
and
peat
(≈
100 %).
Mineral
fertilisers
showed
relatively
short
half-
lives
for
applied
N;
N
input
and
N
output
were
already
in
equilibrium
after
the
time
elapsed
from
the
start
of
the
experiment.
Equilibration
between
N
input
and
N
output
was
not
observed
in
plots
receiving
organic
amend-
ments.
Peat
N
showed
a
more
or
less
inert
behav-
iour
during
the
time
span
of
the
experiment
owing
to
its
high
stability.
The
15
N
abundance
method
added
valuable
information
to
the
results
obtained
by
the
tradition-
al
difference
method
using
the
no-N
treatment
as
reference.
Acknowledgements:
We
thank
the
Austrian
Science
Foundation
(Fonds
zur
Förderung
der
wis-
senschaftlichen
Forschung)
for
financing
this
bilateral
project.
The
authors
thank
Pär
Hilsström
for
the
careful
and
responsible
maintenance
of
the
long-term
experi-
ment.
References
[1]
Bhogal
A.,
Young
S.D.,
Sylvester-Bradley
R.,
Fate
of
15N-labelled
fertilizer
in
a
long-term
field
trial
at
Ropsley,
UK,
J.
Agric.
Sci.,
Cambridge
129
(1997)
49-63.
[2]
Ellert
B.H.,
Bettany
J.R.,
Calculation
of
organic
matter
and
nutrients
stored
in
soils
under
contrasting
management
regimes,
Can.
J.
Soil
Sci.
75
( 1995)
529-538.
[3]
Gerzabek
M.H.,
Soil
organic
matter
(SOM)
dynamics
determined
by
stable
isotope
techniques,
Mitteilungen
der
Deutschen
Bodenkundlichen
Gesellschaft
87
(1998)
161-174.
[4]
Gerzabek
M.H.,
Kirchmann
H.,
Pichlmayer
F.,
Response
of
soil
aggregate
stability
to
manure
amend-
ments
in
the
Ultuna
long-term
soil
organic
matter
exper-
iment,
Zeitschrift
für
Pflanzenernährung
und
Bodenkunde
158
(1995)
257-260.
[5]
Gerzabek
M.H.,
Pichlmayer
F.,
Kirchmann
H.,
Haberhauer
G.,
The
response
of
soil
organic
matter
to
manure
amendments
in
a
long-term
experiment
at
Ultuna,
Sweden,
Eur.
J.
Soil
Sci.
48
(1997)
273-282.
[6]
Granhall
U.,
Kvävefixering
av
cyanobakterier
i
åkermark -
ett
förbisett
kvävetillskott,
Slu
Fakta
Mark/växter
no.
12,
Swedish
University
of
Agricultural
Sciences,
Uppsala,
1990.
[7]
Haynes
R.J.,
Fate
and
recovery
of
15
N
derived
from
grass/clover
residues
when
incorporated
into
a
soil
and
cropped
with
spring
or
winter
wheat
for
two
suc-
ceeding
seasons,
Biol.
Fertil.
Soils
25
(1997)
130-135.
[8]
Högberg
P.,
Tansley
Review
no.
95:
15
N
natural
abundance
in
soil-plant
systems,
New
Phytologist
137
(1997) 179-203.
[9]
Jenkinson
D.S.,
Rothamstedt
Exp.
Sta.
Report,
Part
2, 1977,
pp.
103-109.
[10]
Kirchmann
H.,
Persson
J.,
Carlgren
K.,
The
Ultuna
long-term
soil
organic
matter
experiment,
1956-1991.
Reports
and
Disseration
17,
Swedish
University
of
Agricultural
Sciences,
Uppsala,
1994.
[11]
Kirchmann
H.,
Pichlmayer
F.,
Gerzabek,
M.H.,
Sulfur
balances
and
sulfur-34
abundance
in
a
long-term
fertilizer
experiment,
Soil
Sci.
Soc.
Am.
J.
59
(1996)
174-178.
[ 12]
Lövblad
G.,
Amann
M.,
Andersen
B.,
Hovmand
M.,
Joffre
S.,
Pedersen
U.,
Deposition
of
sulfur
and
nitrogen
in
the
nordic
countries:
present
and
future,
Ambio 21
(1992) 339-347.
[13]
Mårtensson
A.M.,
Witter
E.,
Influence
of
vari-
ous
soil
amendments
on
nitrogen-fixing
soil
microor-
ganisms
in
a
long-term
field
experiment,
with
special
reference
to
sewage
sludge,
Soil
Biol.
Biochem.
22
(1990) 977-982.
[14]
Pichlmayer
F.,
Blochberger
K.,
Isotopenhäufig-
keitsanalyse
von
Kohlenstoff,
Stickstoff
und
Schwefel
mittels
Gerätekopplung
Elementaranalysator-
Massenspektrometer,
Zeitschrift
für
Analytische
Chemie 331
(1988) 196-201.
[15]
Powlson
D.S.,
Quantification
of
nutrient
cycles
using
long-term
experiments,
in:
Leigh
R.A.,
Johnston
A.E.
(Eds.),
Long-term
Experiments
in
Agricultural
and
Ecological
Sciences,
CAB
International,
Wallingford,
1994, pp. 97-115.
[16]
Prasad
R.,
Power
J.E.,
Soil
Fertility
Management
for
Sustainable
Agriculture,
CRC
Press,
Boca
Raton,
FL,1997.
[17]
Puget
P.,
Chenu
C.,
Baldesdent
J.,
Total
and
young
organic
matter
distributions
in
aggregates
of
cul-
tivated
soils,
Eur.
J.
Soil
Sci.
46
(1995)
449-459.
[18]
Scheffer
F.,
Schachtschabel
P.,
Lehrbuch
der
Bodenkunde,
Enke,
Stuttgart,
1998.
[19]
Selles
F.,
Karamanos
E.,
Bowren
K.E.,
Changes
in
natural
15
N
abundance
of
soils
associated
with
tillage
practices,
Can.
J.
Soil
Sci.
64
(1984)
345-354.
[20]
Van
Breemen
N.,
Feijtel
T.C.J.,
Soil
processes
and
properties
involved
in
the
production
of
greenhouse
gases,
with
special
relevance
to
soil
taxonomic
systems,
in:
Bouwman
A.F.
(Ed.),
Soils
and
the
Greenhouse
Effect,
John
Wiley,
Chichester,
1990,
pp.
195-223.
[21]
Wagner
G.H.,
Using
the
natural
abundance
of
13
C
and
15
N
to
examine
soil
organic
matter
accumulated
during
100
years
of
cropping,
in:
Stable
Isotopes
in
Plant
Nutrition,
Soil
Fertility
and
Environmental
Studies,
International
Atomic
Energy
Agency,
Vienna,
1991, pp. 261-268.
[22]
Witter
E.,
Mårtensson
A.M.,
Garcia
F.V.,
Size
of
the
soil
microbial
biomass
in
a
long-term
field
exper-
iment
as
affected
by
different
N-fertilizers
and
organic
manures,
Soil
Biol.
Biochem.
25
(1993)
659-669.
