Sol1 Bioi. Biochem. Vol. 16. No. I. pp. 63-67. 198-l
Printed in Great Britain. All rights ressrwd
0038-0717 S1 S3.00 + 0.00
Copyright r” 1981 Pergamon Press Ltd
ENERGY OR NUTRIENT REGULATION OF
DECOMPOSITION: IMPLICATIONS FOR
THE MINERALIZATION-IMMOBILIZATION
RESPONSE TO PERTURBATIONS
ERNESTO BOSATTA and FRASK BEREsDsE
Department of Ecology and Environmental Research. Swedish University of Agricultural Sciences.
S-750 07 Uppsala. Sweden and Department of Landscape Ecology and Nature Management.
University of Utrecht, Opaalweg 20. SL-3523 RP Utrecht. The Netherlands
Summary-A model developed previously to describe the turnover of forest soil nitrogen is modified here
to explain the effects of carbon and nitrogen additions on their dynamics. The model. which is structurally
very simple, seems to explain correctly, among other phenomena. the negative correlation between N
mineralization and CO1 evolution observed in many experimental situations. An important variable used
to explain this behaviour is the deficiency factor, which is related to the critical C-to-nutrient ratio and
which gives a measure of the C or nutrient deficiency in the substrate with respect to the needs of the
decomposers. Ways are discussed in which the model output can be used to explain the observed retention
in the soil of fertilizer 9 added to mature forest soils.
INTRODUCTION
Mineralization of plant nutrients is a key process
controlling primary production in many ecosystems.
Even if understanding of the mechanisms behind the
mineralization-immobilization process is far from
complete, an acceptable level of description of its
dynamics under unperturbed conditions is available
today (Swift et al., 1979).
For N, Aber and Melillo (1979) Staaf and Berg
(1982) and others have found that, in a great number
of decomposing litter materials, two distinct phases
can be distinguished in the turnover. A period of N
accumulation, the duration and magnitude of which
depends on the substrate, is followed by a period of
net N release. An explanation of this behaviour has
already been provided by Waksman and Tenney
(1927). During the accumulation phase the N in the
decomposing material is in short supply with respect
to the needs of the microorganisms which, con-
sequently, must import nitrogen from the sur-
roundings: the material is N-deficient (immo-
bilization predominates over mineralization). On the
other hand, during the release phase, the material is
C- or energy-deficient (mineralization predominates
over immobilization).
According to Swift et al. (1979) the same general
trends of the mineralization-immobilization dynam-
ics should also apply to other nutrient elements, and
they claim that the carbon-to-element ratio in
resource and in decomposer biomass are main factors
determining the presence or the lack of an accumu-
lation phase of the nutrient in the material. Very little
is understood of the processses involved in perturbed
situations but they may be more complex.
It has often been found that N mineralization from
soils is stimulated by the additions of nitrogen fertil-
izers (Johnson and Guenzi. 1963; Broadbent and
Nakashima, 1971). The same effect has been observed
in the laboratory by Johnson et al. (1980) following
the addition of urea to control soils. The interesting
point, however, is that C mineralization seems to be
negatively correlated to N mineralization. viz., both
in laboratory and field experiments, the CO2 evo-
lution decreases following N fertilization (Johnson er
al., 1980: BUth er al., 1981). The common idea that
the priming effect of N additions on N mineralization
is due to enhanced microbial activity (e.g. Westerman
and Tucker, 1974) seems to be challenged by this
observed negative correlation. Another perturbation
phenomenon that is not yet understood is the reten-
tion of N in the soil organic matter following N-
fertilization of forest soils (Ingestad et a/., 1981).
The aim of our paper is two-fold. First. a model
previously developed (Bosatta and Staaf, 1982) is
modified here in order to study the effects of C and
N additions to the mineralization-immobilization
dynamics. Second, to show that the deficiency factor
(a variable measuring the status of the decomposing
material as being energy or nutrient deficient) is of
primary importance in explaining the responses of
the mineralization-immobilization dynamics to
perturbations.
A GENERAL MODEL FOR .MINERALtZATION-
IMMOB1LIZATION. THE DEFICIENCY
FACTOR
The mineralization-immobilization process is in-
timately associated with the activity of the decom-
posers. Suppose that an amount, B, of living micro-
bial biomass is growing in a certain amount Of a
homogeneous decomposing substrate producing new
biomass at a rate P and with a given concentration
of C, f,. and N, fn. (Throughout this chapter, the
word “nitrogen” could equally well be substituted by
the word “nutrient”).
63
6-l
ERXESTO BOSATTA ud FRASK BERENDSE
Let S, and S, denote the amounts of C and N in
A series of relevant results can be deduced from
the substrate respectively and B, and B, the corre-
this model without making any assumption what-
sponding amounts of the same elements in the micro-
soever about the nature of the microbial production
bial biomass. By definition, C = S, + B; and
P (except, of course. that P > 0). Using this. the
N = S, i- B, will be the total amounts of C and N in
definition of e and equation (7a) vve see that C
the substrate-microbe complex. To SC: up a general
behaves as a monotonously decreasing function of
model for the dynamics of C and h. the rates of
time.
change of C and N are given by the rates of respira- To find the behaviour of N. we first take the time
tion, R and mineralization. m, respectively.
derivative of the C-to-N ratio and introduce equation
The rate of respiration is the difference between the
(7) into it to obtain:
rate at which C is taken up from the substrate,f;P e.
and the rate at which it is incorporated into biomass,
r: = k cf, -f)).
(8)
fcP, i.e.
R =A P -f.p
Let C,,, N, and r,, = C&N, be the amounts of C, N
e L
(1)
and the C-to-N ratio at some time t = 0. Since P,‘N
is a positive quantity, r will either decrease or increase
where e is the production-to-assimilation ratio. In
in time depending on whether r. >_/: j, or r,, <i,‘f,
this model, e, L and f, are constant magnitudes
until it reaches the final value fcjfn.
(parameters).
Two situations are of practical interest. Assume
N uptake is equal to the product of C uptake and first that the substrate is initially C deficient. i.e.
the substrate N-to-C ratio and so, the rate of net
rc > r, >fi/fi. Under these conditions. m > 0 all the
mineralization is
time, and N will decay steadily in time. The other
m=!ip?!l-fp
situation is that for which r. > r, >f;/fn. i.e. the
e 5, n
(2)
substrate is N-deficient. From equation (5). m < 0
and N will increase from its starting value N,. But,
where fnP is the rate at which the N is incorporated
from equation (8) r decays in time and when it
into biomass.
reaches the value r,, m equals zero and N reaches a
In our previous model (Bosatta and Staaf, 1982) it
maximum. Thereafter, the substrate becomes C-
was assumed that living and dead microbial biomass
deficient and a net release of N begins.
had the same C-to-N ratio. The present derivation is
The conclusion is that, regardless of the definition
more general, since that assumption is not used here.
adopted for P, the equations (7a) and (7b) predict the
The amount of living microbial biomass is usually
accumulation and release phases of 3 in the decom-
a small fraction of the whole decomposing litter.
posing substrate.
Berg and Soderstrom (1979) found that the total
amount of fungal biomass after I yr was about
CARBON OR NITROCES Ll.WT.ATIOSS TO
13,; of the weight of the total litter at that time. Thus,
MICROBIAL GROWTH. THE EFFECTS OF
C ASD N ADDITIOKS
we assume that the C-to-N ratio of the substrate
(SJS,) can be approximated by the C-to-N ratio, r,
To analyze the effects of C and S additions on
of the substrate-microbe complex, i.e.
mineralization and respiration we must define P in
r = C N 2 SJS,.
(3
some explicit way.
If p (day-‘) is the relative growth rate of the
The critical C-to-N ratio, r,, is defined as (Parnas.
microbial biomass B when no element is limiting
1975):
growth, we define productivity as:
r, =_Q(f,e).
(4)
P=pB.
(9)
Introducing equations (3) and (4) into (2) we
If some element is limiting growth. vve introduce a
obtain a net mineralization rate:
reduction factor, p,
mZ
( >
2 - 1
f”P.
P=ppB
(10)
r such that 0 <p < 1.
We now define the deficiency factor as:
In N-poor systems, i.e. systems in which N is
limiting growth (r > rc), the reduction factor coin-
rc
cides with the deficiency factor defined in equation (6)
.,.’ = -,
r
(Berendse and Bosatta, 1984, to be published), i.e.
If the substrate is C deficient r, > r (i.e. 7 > I), then
px = rJr.
(11)
m > 0, i.e. mineralization will predominate. On the
other hand, for a N deficient substrate r > r,(;t < 1)
Now, we assume that if C is limiting growth
then m < 0 and immobilization will predominate.
(r < r,), the reduction factor is:
From equations (I) and (5), we can now obtain
pc = r/rc.
(12)
dynamic equations for the total amounts of C and N.
namely:
From equations (11) and (12), growth will be at a
maximum when r = r,, a result already observed by
C = - R = - (I - e)AP/e;
(7a)
Pamas (1976) in culture experiments. The next
fi=-m=- :-I fnp.
i J
assumption is that B is a constant fraction, r. of the
(7b)
total C in the substrate (Bosatta and Staaf, 1982). i.e.
\ / B=rC.
(13)
Energy or nutrient regulation of decomposition
65
Finally, the effect of perturbations must be com-
pared to some reference state, This is generated here
by driving the system to a steady state (ss) by means
of an external input of carbon at a rate I (mg day-‘)
having a N-to-C ratio w.
With this, and introducing (10) and (13) into
equations (7) we get
t= -kpC+I
(I4a)
ti=-~k,pN+-bpC+wf,
f l4b)
where
K_=
$%M
(15)
k, =$pl
(I@
and
b =fnp~.
(17)
If p is set to pE = C/N/r, the system (14) reaches ss
in which
(N/C)* = r;’ + ~v(I -e)
(18)
and so r* < rc. This means that the ss is carbon
deficient and a net release of nitrogen is produced
from the substrate in this state, namely
m* = wr.
(19)
For the N-limited case, p in equations (14) must be
substituted with pN (equation 11). Trivially, no N-
deficient ss can be generated from equations (14a and
b) if a net input of N is given to the system.
A N-deficient ss can be achieved, on the other
hand, if an output of N (such that output > wl) is
subtracted from equation (14b). If we use the litter
layer as an image of a N-deficient system, we could
regard this output as a transfer of nitrogen reiated to
the humificat~on process. The easiest way to mimic N
C-limited
C is added
N - Ltmited
N is odded
losses is by giving !c a negative value, and, in this way.
equation (18) can be used again to calculate the
(N/C)*. So, putting w to a negative value, we get
r* > rc and the substrate has a net immobilization.
An analysis of the characteristic roots of equations
(14) in the neighborhood of both ss points shows
that:
-the C-limited ss is stable For any choice of param-
eter values;
-the N-limited ss will be stable only if:
l~vl.L(l -e)<L
%
1’
(20)
Furthermore, when the following relationship is
fulfilled
I )v I&t 1 - e)
1
ef,
+ 2[1 - (1 - e)‘j*] ’ ’
(21)
the variables in the N-limited system will, as a
response to the perturbation, oscillate in their return
to equilibrium.
As a conclusion, a qualitatively different response
to same perturbation can be expected depending on
whether the system is C- or N-deficient. This is
depicted in Fig. 1.
The values off, andf, agree with values found by
BHBth and Siiderstram (1979) in fungal biomass from
different forest soils. The values given to e has been
used before (Bosatta et al., 1980) as an estimate of the
production efficiency in the same kind of systems.
As can be seen from equations (20) and (21) the
qualitative pattern of b&aviour is not affected by the
choice of ~1: this choice can only speed or sIow down
the whole dynamics. The value given to z is based in
the observation of Berg and S~derstr~m (1979) pre-
viously mentioned (see discussion preceeding equa-
tion 3). With this election of parameter values, the
critical C-to-N ratio is r, = 62.5.
C-limited
N is added
t
I I
1 I
I
N-limited
C is added
Time (days)
Fig. 1. The changes (in %) with respect to steady state of rates of respiration (----) and net nitK@n
mineralization (-----
), In all cases, the amounts added are 10% of the steady state values. The curves are
calculated assuming that e = 0.2, I, = 0.5, f, = 0.04, p = 2. day-‘, x = 0.01 and I = I mg day-‘.
S.f.8. Ib’l-E
66
ERNESTO BOSATN and
FRASK BERESUSE
40 - C -limlred
C 1s added
---_
-20 - !/
-40
I
I I
I
I
80 - N-l~rmted
N IS added
40 - \
\
ob ’ ___/
t
/----.
/
-40 -
-80 -
I I
I I
I
0 100 200 300 400 500
/
0
I00 200
300 400 500
N-lImIted
C is added
Time (days)
Fig. 2. Changes in the rates of respiration and net nitrogen mineralization when the production-to-
assimilation ratio, e, is set to 0.4. The remaining parameters are the same as in Fig. 1.
The C-to-N ratio of Scats pine litter is about 100
(Staaf and Berg, 1982). So. the C-limited steady state
is generated by choosing IV = 0.01, and, in this way,
r * = 41.7. The N-deficient ss is generated by placing
w = - 0.006 and, in this way, r* = 91.2. The
efficiency in producing new biomass, e, is probably
the most uncertain parameter. According to McGill
er al. (1981). its value can lie between 0.2 and 0.6.
Also, as can be seen from equations (20) and (21).
changes on e can drastically change the qualitative
properties of the system.
In the example in Fig. 2, e has been set to 0.4.
Comparing this figure with Fig. 1, the conclusion can
be drawn that N is more sensitive than C to changes
in this parameter. It should be noticed that the
concentration of N in the ss of Fig. 2 is higher than
in the corresponding ss of Fig. 1.
DISCUSSION
From the results of Fig. 1, C-limited or nutrient-
limited systems will react in very different ways when
perturbed. This indicates that the “primary effect”
explanation must be used with care, and that consid-
eration must be made of the deficiency status of the
system.
The priming effect is introduced here in equation
(10) through the reduction factor p. As can be seen
in equation (12), for example, this factor will increase
if C is added to a C-deficient system, thus increasing
the productivity of microorganisms. The status of a
system is given here by the deficiency factor (equation
6), which has previously also been found to be a
variable of major importance in explaining the
dynamics of co-existence between plants and decom-
posers (Bosatta, 1981). The easiest way to define the
deficiency factor in an experimental way is, perhaps,
to measure whether the substrate is releasing or
accumulating nutrient before the perturbation is
applied. Also, if the specific decomposition rate of the
material is known, the critical C-to-N ratio can be
estimated with the regression model of Bosatta and
Staaf (1982).
The negative correlation between respiration and
mineralization for the C-limited case supplied with N
in Fig. 1 agrees well with the experimental results of
SGderstrGm et al. (unpublished), Johnson et al. (1980)
for forest soils, and Kowalenko er nl. (1975) for
agricultural soils. Foster et al. (1980) also observed a
decrease in respiration when they added N to soil
sample consisting of a mixture of L and F horizons
from a pine stand, and they attributed this decrease
to toxic effects of the added N. They claimed that the
high C-to-N ratio (_ 43) will make this substrate
N-limited with respect to microbial growth. Instead.
an alternative explanation could be that their mate-
rial was C-limited, a reasonable assumption for this
range of C-to-N (e.g. Berg and Staaf, 1981).
Figure I shows that a C-limited system amended
with C will increase respiration and immobilization
(net mineralization decreases in respect to ss). This is
in agreement with experimental results of Johnson
and Edwards (1979) on N mineralization and CO,
evolution from a girdled stand when sucrose was
added to it.
Niimmik and Mijller (1981) had. among others,
observed an increased retention of N in forest soils
after fertilization. The litter layer of a forest soil can
be considered N-limited and thus will be steadily
immobilizing N (Staaf and Berg, 1977). It can be
speculated that the addition of nitrogen to the N-
limited L-layer stimulates microbial immobilization
and this, combined with the humification process,
will make the N less mobile in the soil. Some support
for increased immobilization speculated upon here
can be found in Fig. I; after a certain period immo-
bilization increases with respect to steady state if N
is added to a N-limited system.
Energy or nutrient regulation of decomposition
67
dcknowvledgements-We thank 6. &ren for discussions
and suggestions. We also thank F. Andersson. 8. Berg, Ch.
McClaugherty, K. Paustian, T. Penson and D. Waring
for comments. This work was supported by the Swedish
Natural Science Research Council and the Netherlands
Organization for the Advancement of Pure Research.
REFERENCES
Aber J. D. and Melillo J. M. (1979) Litter d~omposit~on:
measuring relative contributions of organic matter and
nitrogen to forest soils. Canadian Journal of Eororr_s 58,
416-421.
BUth E. and S&derstriim B. (1979) Fungal biomass and
fungal immobilization of plant nutrients m Swedish coni-
ferous forest soils. Revue d‘Ecoiogie et de Biologic du Soi
$6, 477159.
Baith E., Lundgren B. and Soderstrom B. (198 I) Effects of
nitrogen fertilization on the activity and biomass of fungi
and bacteria in a podzolic soil. Zentralblarr fir
Bakreriologie, Parasireikurtde, Infektionskrankheiten &d
Hygiene. Abt. I.. Orig. C 2, 90-48.
Bere B. and Siiderstrijm B. (1979) Fumtal biomass and
&rogen in decomposing Scats pine needle litter. Soil
Biology & Biochemistry 11, 339-341.
Berg B. and Staaf H. (1981) Leaching. accumulation and
release of nitrogen in decomposing forest litter. In Rrres-
trial Nitrogen Cycles (Clark F. E. and Rosswall T., Eds).
Ecological Bulletins (Stockholm) 33, 163-178.
Bosatta E. (1981) A qualitative analysis of the stability of
the root-microorganisms soil system. I. Carbon-nitrogen
status and nitrogen mineralization. Ecological .\/odelling
13, 223-236.
Bosatta E. and Staaf H. (1982) The control of nitrogen
turnover in forest litter. Oikos 39, 143-15 I.
Bosatta E., Bringmark L. and Staaf I-I. (1980) Nitrogen
transformations in a Scats pine forest mor-model analysis
of mineralization, uptake by roots and leaching. In
Structure and Function of ~arthern Corriji?rous Forests-
An Ecasvstem Study (T. Persson, Ed.). Ecolo&at Bulle-
tins (Sr~ckhalm~ 3i, %5-589.
Broadbent F. E. and Nakashima T. 119711 E&cts of added
salts on nitrogen mineralization in‘ three Califomia soils.
Sail Science Society of America Proceedings 35,457-460.
Foster N. W., ~auchamp E. G. and Corke C. T. (1980)
Microbial activity in a Pinus banksiana Lamb. forest Roor
amended with nitrogen and carbon. Canadian Journal
Soil Science 60, 199-5209.
Ingestad T., Aronsson A. and Agren G. I. (1981) Nutrient
Rux density model of mineral nutrition in conifer eco-
systems. Studia Foresialia Suecia 160. 61-71.
Johnson D. D. and Guenzi W. D. (1963) Influence of salts
on ammonium oxidation and carbon dioxide evolution
from soil. Soil Science Saciet_s of America Proceedings 27,
663-666.
Johnson D. W. and Edwards N. T. (1979) The effects of
stem girdiing on biogeochemical cycles within a mixed
deciduous forest in Eastern Tennessee II. Soil nitrogen
mineralization and nitrification rates. Oecotogia 4%
259-271.
Johnson D. W.. Edwards N. T. and Todd D. E. (1980)
Nitrogen mineralization immobilization and nit~~cation
following urea fertilization of a forest soil under field and
laboratory conditions. Soil Science Society of America
Journal 44, 6 I O-6 16.
Kowalenko C. G.. Ivarson K. C. and Cameron D. R. (1978)
Effect of moisture content. temperature and nitrogen
fertilization on carbon dioxide evolution from field soils.
Soil Biology & Biochemistry 10, 417423.
McGill W. B., Hunt H. W., Woodmansee R. G. and Reuss
J. 0. (1981) PHOENIX, a model of the dynamics of
carbon and nitrogen in grassland soil. In Terrestrial
Nitrogen C,vcles. (F. E. Clark and T. Rosswall, Eds).
Ecological Bulletins (Stockholm) 33, 49-i 15.
Niimmik H. and Mliller Go (1981) Nitrogen recovery in
soils and needle biomass after fertilization of a Scats pine
stand. and growth responses obtained. Studia Forestalia
Suecia 159, 5-37.
Parnas H. (1975) Model for decomposition of organic
material by microorganisms. Soil Biology & Biochemistry
7, 161-169.
Pamas H. (1976) A theoretical explanation of the priming
effect based on microbial growth with two limiting sub-
strates. Soil Biolog_r & Biochemistry 8, 139-144.
Staaf H. and Berg B. (1977) Mobilization of plant nutrients
in a Scats pine forest mor in central Sweden. Sika Fennica
11, 210-217.
Staaf H. and Bere B. (1982) Accumulation and release of
plant nutrients% decomposing Scats pine needle litter.
Long-term decomposition in a Scats pine forest II. Cnna-
dian Journal 01 Bofany 60, 1561-I 568.
Swift M. J., Heai 0. W. and Anderson J. M. (1979)
~ecampositian in Terrestrial Ecosystems. Studies in Ecol-
ogy. Vol. 5. Blackwell. Oxford.
Waksman S. A. and Tenney F. G. (1927) The compositjon
of natural organic materials and their d~ompo~ition in
the soil: II influence of age of plant upon the rapidity and
nature of its decomposition-Rye plants. Soil Science 24,
317-333.
Westerman R. L. and Tucker T. C. (1974) Etfects of salts
plus nitrogen- 1 Mabelled ammonium chloride on mineral-
ization of soil nitrogen, nitrification and immobihzation.
Soil Science Society of America Proceedings 38, 602-605.
