Download Microbial mobilization and immobilization of soil nitrogen

Microbial mobilization and immobilization
of soil nitrogen
Per Bengtson
Dissertation
Lund, 2004
A doctoral thesis at a university in Sweden is produced either as a monograph or a collection
of papers. In the latter case, the introductory part constitutes the formal thesis, which
summarises the accompanying papers. These have either already been published or are in
manuscripts at various stages (in press, submitted or in ms).
© Per Bengtson
ISBN 91-7105-210-0
SE-LUNBDS/NBKE-04/1034+104pp
This thesis is based on the following papers, which are referred to by their Roman
numerals.
I.
Bengtsson, G., Bengtson, P., and Månsson, K.F. 2003. Gross nitrogen
mineralization-, immobilization-, and nitrification rates as a function of soil C/N
ratio and microbial activity. Soil Biology and Biochemistry 35, 143-154.
II.
Bengtson, P., Bengtsson, G., and Falkengren-Grerup, U. Relieving substrate
limitation – soil moisture and temperature determine gross N transformation
rates. (Submitted)
III. Bengtson, P., Falkengren-Grerup, U., and Bengtsson, G. Spatial distribution of
gross N transformation rates and plants – A reciprocal relationship. (Submitted)
IV. Månsson, K.F., Bengtson, P., Bengtsson, G., and Falkengren-Grerup, U.
Competition for nitrogen – litter leachate favours microorganisms at the expense
of plants. (Manuscript)
V.
Bengtson, P. and Bengtsson, G. Rapid turnover of DOC in temperate forest soils
accounts for increased CO2 production at elevated temperature. (Manuscript)
VI. Bengtson, P. and Bengtsson, G. Bacterial immobilization and remineralization of
N at different growth rates and N concentrations. (Manuscript)
Paper I is reprinted with permission from the publisher.
Contents:
Introduction
7
N transformation processes
7
Aim of thesis
7
Microbial uptake and assimilation of N
8
The role of C and N in determining N mineralization and immobilization rates
9
Microbial activity and N mineralization and immobilization
11
Restrictions on the microbial biomass
15
The response of soil microorganisms to C and N additions
15
The influence of plants on N mineralization and immobilization rates
17
Abiotic immobilization of N
18
Abiotic vs. biotic immobilization of N
18
Mobilization of microbially bound N
19
Environmental concerns
20
Conclusions
22
References
22
Introduction
Microorganisms in forest ecosystems normally recycle nitrogen (N), such that
gaseous losses and leaching is quite limited. However, it has been hypothesized that
the high anthropogenic N input to forest ecosystems may cause N saturation (Aber,
1992). N saturation has been defined in a number of different ways but all include
increases in N availability over time that results in the removal of N limitation on all
biological processes in the ecosystem considered (Aber et al., 1998). This would open
the N cycle and lead to an increased mobility and leaching of N, soil acidification due
to elevated nitrification rates and subsequent aluminum toxicity and loss of base
cations.
The theory about N saturation and its consequences is widely accepted. However, a
number of N fertilization and removal experiments have been carried out over the last
decade or so (Wright and van Breemen, 1995; Wright and Rasmussen, 1998), and the
results from these experiments are not as clear-cut and uniform as expected. Many
forest ecosystems seem to have a much higher capacity to retain N than originally
thought (Emmett et al., 1995; Gundersen et al., 1998; Kjønaas et al., 1998), and the
response of an ecosystem to N fertilization is not always as expected. Therefore, the
interpretation of the effects of high chronic N deposition rates requires an improved
understanding of the biology of the microorganisms involved in N retention and
transformation in soils.
N transformation processes
N is the only essential nutrient that is not weathered from soil minerals. This means
that all habitats depend on bacterial fixation of molecular N from the atmosphere, or
alternatively, on input of NHX (NH4+/NH3) or NOX (NO, NO2, HNO2, HNO3, and
NO3-) mainly from anthropogenic sources. The ability to fix atmospheric N is limited
to a few species of bacteria, and the process requires a lot of energy. N availability in
most terrestrial ecosystems is therefore highly dependent on the tight recycling of N
carried out by microorganisms. NH4+ can be lost as NH3 (volatilization) or be taken up
by plants or microorganisms (NH4+ immobilization) and converted to organic N.
Alternatively, NH4+ is converted to NO3- via NO2- by ammonia-oxidizing and nitriteoxidizing bacteria, respectively (nitrification). NH4+ can also be produced by
degradation of organic matter (mineralization), dissimilatory reduction of NO3(DRNA), and maybe also leaked or excreted from bacterial cells (mobilization) and
fine roots. NO3- can either be immobilized and converted to organic N in
microorganisms (NO3- immobilization) or taken up by plants, or reduced to gaseous N
compounds such as N2O or N2 by denitrifying bacteria (denitrification).
Aim of thesis
In forest soils, mineralization of organic N, including microbial N, and
immobilization of inorganic N, especially NH4+, are the quantitatively most important
N transformation processes (Wessel and Tietema, 1992; Hart et al., 1997; Fisk and
Fahey, 2001; Paper I). The aim of this thesis was to elucidate the factors that
determine the rate of these processes.
7
Microbial uptake and assimilation of N
As all other living organisms, bacteria need N for the synthesis of compounds
necessary for growth, reproduction and survival, e.g. proteins, DNA and RNA. The
assimilation of N can be divided into two separate events: 1. the uptake of N from the
extracellular medium, and 2. the intracellular production of N containing compounds.
NH4+ solutions always contain NH3, and there is substantial evidence for rapid
diffusion of NH3 across cytoplasmic membranes (Montesinos et al., 1998). Diffusion
of NH3 followed by trapping of intracellular NH4+ could therefore potentially
represent a significant process for N acquisition. However, the high pKa of NH3 (9.25)
means that NH4+ is the predominant form in forest soils, which typically have a pH in
the range from 4-7, and diffusion of NH3 followed by intracellular trapping would be
a highly inefficient way to acquire N. Furthermore, growing bacteria can maintain an
intracellular ammonium-pool against a concentration gradient of at least 100-fold
(Jayakumar and Barnes, 1983; Jayakumar et al., 1985), suggesting that bacteria have
an active and efficient NH4+ transport system (Amt) that compensates for outward
diffusion of membrane permeable NH3. Such ammonium transport systems are found
in bacteria and fungi (Siewe et al., 1996; Marini et al., 1997; Meletzus et al., 1998;
Michel-Reydellet et al., 1998; Montesinos et al., 1998), as well as in plants
(Ninnemann et al., 1994), and they have a high degree of homology. There are
probably three different NH4+ transport proteins, each one with different affinity for
NH4+ (Marini et al., 1997; Montesinos et al., 1998; Gazzarrini et al., 1999) The three
NH4+ transporters possess distinct substrate affinities, ranging from less than
micromolar to millimolar concentrations, the range of NH4+ concentrations typically
found in soil (Marshner, 1995).
The Amt system is energy-dependent and activated when bacteria grow under N
limited conditions but repressed when NH4+ is abundant (Servín-Gonzáles and
Bastarrachea, 1984; Siewe et al., 1996). The protein is synthesized only under N
starvation and inhibited by internal glutamine (Siewe et al., 1996). Glutamine also
represses Amt, and the Amt system therefore seems to be regulated by N availability
in a manner similar to that of several amino acid transport systems (Servín-Gonzáles
and Bastarrachea, 1984; Merrick and Edwards, 1995). The level of glutamine
synthetase (GS) also affects the methylammonium uptake, with a higher uptake at
high GS levels (Michel-Reydellet et al., 1998). There are also indications that one of
the three Amt transporters is synthesized/activated in response to high intracellular
concentrations of C skeletons, possibly providing a link between C and N metabolism
(Gazzarrini et al., 1999).
The intracellular production of N containing compounds starts with the incorporation
of cellular N into glutamate (Glu) or glutamine (Gln). These compounds then serve as
N donors in different biosynthetic reactions. There are two primary pathways that
facilitate the synthesis of glutamate (Helling, 1998). It may be formed directly
through reductive amination of 2-oxoglutarate (2-OG, α-ketoglutarate) by glutamate
dehydrogenase (GDH, gdhA) (Equation 1), or indirectly via glutamine synthetase
(GS) which catalyses the addition of ammonia to glutamate to yield glutamine
(Equation 2). The amide group of glutamine is then reductively transferred to
oxoglutarate by glutamate synthetase (GOGAT) to yield two glutamate molecules
(Equation 3). The GS/GOGAT pathway is the most important, and it is ubiquitous in
8
all bacteria, while the GDH pathway only is present in some bacteria, such as the
enterics (Merick and Edwards, 1995).
NH3 + 2-oxoglutarate + NADPH GDH

→ glutamate + NADP+
GS
NH3 + glutatmate + ATP → glutamine + ADP + Pi
Glutamine + 2-oxoglutarate + NADPH GOGAT
 → 2 glutamate + NADP+
(1)
(2)
(3)
GDH has a relatively high Km of 1 to 4 mM, while both GS and GOGAT have low
Km values for their substrates (Merrick and Edwards, 1995; Vanoni and Curti, 1999).
Thus, at low ammonia concentrations the GS/GOGAT pathway is the only pathway
for utilization of ammonia. However, since glutamate dehydrogenase is functioning
when ammonium concentrations are high, ammonia assimilation is shifted from the
glutamine pathway to the glutamate pathway. This might be a way to prevent wasteful
consumption of ATP by glutamine synthetase (Helling, 1998).
The synthesis and/or activation of the proteins for uptake and utilization of different
nitrogenous compounds is tightly regulated in concert with the availability of their
substrates (Merrick and Edwards, 1995). A high intracellular N concentration is
reflected in the intracellular ratio of glutamine to 2-oxoglutarate, such that a high
amount of ammonium results in a high glutamine/2-oxoglutarate ratio, while
ammonium deficiency lowers the ratio. Under N sufficient conditions, the activity of
GS is repressed by adenylylation, while at low intracellular N concentrations, GS is
deadenylylated, resulting in an increase in the GS activity (Vanoni and Curti, 1999).
The expression of the gene encoding GS is also regulated by the intracellular N status.
The result of this is that the transcription is low under N sufficient conditions, while it
is markedly elevated (approximately 10-14 fold) under N deficient conditions
(Schreier et al., 1985; van Heeswijk et al., 2000). Downstream of the gene encoding
GS, a gene homologous to several ammonium transporters is found (van Heeswijk et
al., 1996; Meletzus et al., 1998; Michel-Reydellet et al., 1998). The lack of a clear
termination or promoter sequence between those genes indicates that they are cotranscribed, which is consistent with the finding that methylammonium uptake is
higher at high GS levels (Michel-Reydellet et al., 1998).
The role of C and N in determining N mineralization and immobilization rates
Soil microorganisms are generally considered to be C limited (Mikan et al., 2000;
Fisk and Fahey, 2001, Vance and Chapin, 2001) and heterotrophic decomposition and
nutrient cycling in soils have traditionally been described using first order kinetics
(Molina et al., 1983; Parton et al., 1987; Kersebaum and Richter, 1990; Sallih and
Pansu, 1993):
dC/dt = -kC
(4)
where C is the concentration of available C, and k the first order rate constant.
Integration of equation X yields:
Ct = C0exp(-kt)
(5)
9
where C0 is the initial concentration of C and Ct the concentration of available C at
time t. The C mineralised with time, Cm, can then be calculated by substituting Ct =
(C0 – Cm):
Cm = C0(1 – exp(-kt)
(6)
The rate of C assimilation, Ca, is linearly related to the rate of C mineralization and
depends on the C use efficiency (Ce), i.e. the fraction of the utilized C that is
incorporated into the biomass:
Ca = Cm/(1 - Ce) - Cm
(7)
The N immobilization, Nimm, can then be calculated with:
Nimm = Ca/C:Nmic
(8)
where C:Nmic is the C/N ratio of the microorganisms. If C limitation and first order
kinetics are assumed, N immobilization rates can be expected to be higher in soils
with high C content than in soils with low. Accordingly, Barret and Burke (2000)
found that the C content alone could explain more than 60% of the variation in N
immobilization at regional scales. The rate of N immobilization in soil profiles
decrease in the order litter layer > humic layer > mineral soil (Pulleman and Tietema,
1999; Fisk and Fahey, 2001; Puri and Ashman, 1998), and N immobilization rates are
higher in forest soils than in agricultural soils, possibly as a result of decreasing C
content (e.g. Recous et al., 1999; Fisk and Fahey, 2001) and incorporation of plant
material into soils poor in organic matter enhance mineralization and N
immobilization rates (Paper IV; Watkins and Barraclough, 1996; Recous et al., 1999;
Andersen and Jensen, 2001).
Soil inorganic N pools are small and may have a turnover time of less than a day
(Hart et al., 1994; Scott et al., 1998; Verchot et al., 2001; Paper I and II). The validity
of Equations 4-8 therefore depends on the assumption that there is enough
mineralizable N to support the calculated N immobilization. However, soil
microorganisms use the GS/GOGAT rather than the GDH pathway to assimilate NH4+
(Schimel and Firestone, 1989), some soils have net N immobilization (Paper I;
Nadelhoffer et al., 1984; Giblin et al., 1991; Wagener and Schimel, 1998), and gross
N immobilization rates sometimes increase when soils are fertilized with N (Tietema
and van Dam, 1996; Fisk and Fahey, 2001), all indicative of N limitation. The
coupling of C and N cycling in soils might therefore be controlled by the
concentration of available N rather than by the soil C content, with high N
concentrations resulting in high respiration and immobilization rates.
The soil C/N ratio may be important in determining if microorganisms are C or N
limited, and several models predicting N turnover and retention in soils include the
C/N ratio as an important factor in determining the rate of mineralization,
immobilization and nitrification (van Veen et al, 1984; Aber, 1992; Bradbury et al.,
1993; Janssen, 1996). The idea that the C/N ratio is important for the immobilization
rate rests on the assumption that microorganisms are C limited below a certain C/N
ratio and N limited above this ratio (Tate, 1995). The ratio can be calculated from the
10
C/N ratio of the microorganisms growing on the organic matter and their C use
efficiency:
C:Nlim = C:Nmic/Ce
(9)
where C:Nlim is the C/N ratio where the microorganisms become C or N limited. This
ratio is varying between 13 and 30 in the literature depending on the carbon use
efficiency and the biomass C/N ratio of the microorganisms, but fungi as well as
bacteria are considered to be N limited at substrate C/N ratios above 30 (Killham,
1994; Kaye and Hart, 1997).
Microbial activity and N mineralization and immobilization
Independent of C or N limitation, a number of problems exist with using first order
kinetics. C and N mineralization of soil organic matter is not spontaneous but
catalyzed by extracellular enzymes produced by microorganisms inhabiting the soil,
and as discussed above, the uptake and utilization of N is an active process. Catalyzed
reactions, as well as substrate utilization by microorganisms, can be described with
the Michaelis-Menten equation:
dS/dt = -KES/(Km + S)
(10)
where S is the concentration of the substrate, E the concentration of the calalyst or the
microbial biomass, K the fundamental rate constant that depends on the quality of the
substrate, and Km the half-saturation constant. If the concentration of the catalyst is
assumed to remain constant, K and E can be combined into a new constant, Vmax,
representing the maximum substrate utilization rate:
dS/dt = - Vmax S/(Km + S)
(11)
If substrate concentrations are low enough (S<<Km) Equation 11 can be simplified to
a linear first order approximation (Bekins et al., 1998):
dS/dt = - Vmax S/Km
(12)
First order kinetics may therefore be useful in describing mineralization and
immobilization rates, but it depends on the assumption that the maximum
concentration of the substrate is much less than the half-saturation constant. On the
other hand, if substrate concentrations are high (S>>Km), Equation 11 may be
approximated by a zero order equation (Bekins et al., 1998):
dS/dt = - Vmax
(13)
The choice of zero order, first order, or Michaelis-Menten kinetics in describing C
and N mineralization therefore depends on the concentration of the substrate in
relation to the half-saturation constant. In soils, the amount of SOM may be 50-100
times higher than the microbial biomass (Diaz-Ravina, et al., 1993). If a single pool of
C and N is assumed, substrate concentrations are likely to be much higher than the
half-saturation constant and zero-order kinetics the preferred choice when describing
11
C and N mineralization rates. On the other hand, soil organic matter is very complex
and consists of numerous different compounds with different availability to soil
microorganisms (Schulten et al., 1997, Schulten and Schnitzer, 1998) The
concentration of compounds such as organic acids are relatively low (Hongve et al.,
2000; Strobel, 2001), and if these are considered to be the substrate for
microorganisms, the use of first order kinetics might be justified. To my knowledge,
no studies have verified the assumption that the concentrations are much lower than
the half-saturation constant, probably since the combination of a high microbial
diversity, the ability of microorganisms to utilize several different substrates, and the
numerous possible substrates makes it virtually impossible.
In Paper I and II I tried to circumvent this problem by examining if N mineralization
and immobilization rates were more dependent on and variable with the microbial
biomass and activity than on soil C and N concentrations. If first order kinetics is
assumed and microorganisms C limited, the respiration rate (Cm) and N
immobilization rate (Na) should be correlated (Equation 16) and dependent on the
concentration of C (Equation 6). On the other hand, if microorganisms are N limited,
the N immobilization rate should be dependent on the concentration of inorganic N
(Equations 10-12). Since all known NH4+ transporters isolated from bacteria have a
Km in the micromolar range (Siewe et al., 1996; Montesinos et al., 1998; Soupene et
al., 1998), the NH4+ concentration is sufficiently high to fulfill the assumption that
S>>Km (Paper I), allowing NH4+ immobilization at a maximum rate and the use of
zero order approximations. However, N immobilization rates are often sufficiently
high to deplete the inorganic N pools in less than a day (Hart et al., 1994; Verchot et
al., 2001; Paper I and II). Thus, N mineralization may be the rate limiting step if
microorganisms are N limited.
In Paper I, I found that N mineralization and immobilization rates among soils were
more variable with and dependent on the microbial biomass and activity than on the
soil C/N ratio. The N immobilization was of the same magnitude in two of the soils,
but considerably lower in the third (Paper I, Fig. 1). The same pattern was found for
the ATP content, which reflects both biomass and metabolic state of the
microorganisms (Contin et al., 2000), and the respiration rate (Paper I, Fig. 2 and 3).
Several other studies have found gross immobilization and mineralization rates to be
correlated with respiration rates (Hart et al., 1994; Puri and Ashman, 1998; Tietema,
1998; Barret and Burke, 2000; Paper II). The slope of the relationship between
respiration and N immobilization rates varies between 0.032 (Barret and Burke, 2000)
and 0.20 (Hart et al., 1994) in the literature. From these slopes the C use efficiency
(Ce) and the critical soil C/N ratio for N limitation (C:Nmic, Equation 9) can be roughly
estimated.
Nimm = Cm × slope
(14)
Ca = Nimm × C:Nmic
(15)
Ce = Ca/(Ca + Cm)
(16)
If the soil microbial community is assumed to have a C/N ratio of 10, the C use
efficiencies will vary between 0.24 and 0.67, and the critical C/N ratio for N
limitation becomes 42 and 15, respectively. All the soils in Paper I had C/N ratios
12
above 15 but below 42, and it is therefore possible that that the same species (C or N)
was limiting in all three soils.
Equation 9 suggests the C use efficiency must remain constant C in limited
microorganisms, since variations would lead to the conclusion that the critical C/N
ratio for C or N limitation can vary from day to day. However, the C use efficiency
does not remain constant over time, but may vary several folds (Hart et al., 1994). The
C/N ratio of the substrate that is actually utilized is also variable and differs from that
of the whole soil (Hendrickson, 1985; van Veen et al., 1985; Hadas et al., 1987; Hart
et al., 1994). This may be the reason why there was no influence of the whole soil
C/N ratio on the N immobilization rate, and it also demonstrates that the amount of C
that is available to and utilized by the microorganisms is not necessarily correlated to
the total C content in a soil. Even though N mineralization and immobilization rates
were highest in soils with the highest microbial activity in Paper I, suggesting that
zero order or Michaelis-Menten kinetics should be used, those soils were also the ones
with the highest C and N content, which indicate that first order rates were applicable.
Hence, the results in Paper I were ambiguous with respect to the kinetic order of N
mineralization and immobilization, and to the influence of the microbial activity and
biomass on the processes. In Paper II and V, I addressed this ambiguity by estimating
the utilization of three different soil C fractions. Analysis of δ13C of the respired CO2
and the three C pools indicated that microorganisms mainly utilized dissolved organic
C (DOC; Paper V, Fig. 1). However, the respiration rate was not dependent on the
concentration of DOC (Paper II, Table 3), which would be expected if the
microorganisms were C limited and DOC was their C source (Equation 4), and the N
immobilization rate was independent of the DOC concentration (Equations 4-8; Paper
II, Table 3). The low DOC concentrations compared to the respiration rate meant that
the DOC pool would be depleted unless DOC was constantly produced. In addition to
microbial C recycled to the DOC pool, soil organic matter (SOM) seemed to be the
source of DOC (Paper V, Fig. 2), but the respiration rate was independent of the
amount of SOM as well (Paper II, Table 3). However, there was a strong connection
between temporal variations of temperature, which affect the rate of metabolic
processes, and the respiration rates, and between soil water content and respiration
rates (Paper II).
Those results suggested that the microorganisms were not C limited, and that first
order approximations, if applicable, should be based on N concentrations. However,
Paper I, II and IV gave no evidence for a relationship between in situ concentrations
of NH4+ and NH4+ immobilization rates but supported the suggestion above that the
NH4+ concentrations was sufficiently high fulfill the assumption that S>>Km, allowing
NH4+ to be immobilized at zero order rates. It can be argued that microorganism
mainly utilized other N sources than NH4+, hence the lack of association between
NH4+ concentrations and immobilization rates. NO3- would be a likely candidate as an
alternative N source because of its high mobility in soils compared to NH4+, but even
though NO3- immobilization commonly occurs in forest soils, the ratio between NO3and NH4+ immobilization is generally low (Recous and Mary, 1990; Zak et al., 1990;
Fisk and Fahey, 2001; Verchot et al., 2001; Paper I and IV). Similarly,
microorganisms have the capacity to take up a number of organic N compounds
(Merrick and Edwards, 1995), but direct uptake of organic N is small compared to the
NH4+ uptake (Molina et al., 1990; Hadas et al., 1992). Uptake of NO3- and organic N
13
therefore seem to make a minor contribution to the microbial N demand. Hence, N
mineralization seems to be the rate limiting step.
N mineralization rates were dependent on the soil water content, while neither
dissolved N concentrations nor total soil N content had any influence (Paper II). I did
however find a positive relationship between gross N mineralization rates, respiration
rates, and the concentration of phosphate buffer extractable N (PON), in agreement
with Matsumoto et al. (2000). PON is a homogeneous, bacterially derived group of
protein-like compounds with a size of 8-9 kDa (Matsumoto et al. 2000), which may
represent residues of exoenzymes that become abundant and released to degrade
organic N compounds during high microbial activity. Accordingly, positive
correlations between gross N mineralization rates and soil enzyme concentrations
have been found by e.g. Zaman et al. (1999).
Taken together, the lack of influence of the soil C/N ratio (Paper I) and soil C and N
pool sizes on the mineralization and immobilization of C and N (Paper II and V), and
the positive influence of temperature, soil water content, thymidine incorporation
rates (Paper IV), and ATP and exoenzyme concentrations (Paper I and II), suggests
that first order approximations (Equation 12) would be inappropriate to describe the
mineralization and immobilization of C and N. Turnover of C and N in forest soils
rather seems to be tightly controlled by the microbial biomass and activity alone. The
correlations between the production and consumption of the primary C and N sources,
DOC and NH4+, (Paper I, II and V; Davidson et al., 1992; Hart et al., 1997; Verchot et
al., 2001), and the correlation between respiration and N mineralization and
immobilization rates (Paper II; Hart et al., 1994; Puri and Ashman, 1998; Tietema,
1998; Barret and Burke, 2000) seem to confirm this.
The combination of K and E into a single term, Vmax, in Equation 13 rests on the
assumption that the concentration of the catalyzor (E) is constant. The production of
the enzymes necessary for the mineralization and uptake of C and N is however
dependent on the microbial activity, and because of its variability this assumption my
not be valid. Equation 13 can therefore be modified to:
dS/dt = -KE
(17)
which essentially is a first order equation, but dependent on the enzyme concentration
rather than the substrate concentration. Since the concentration of E is dependent on
the microbial activity Equation 17 can be rewritten to:
dS/dt = -KBmAm
(18)
where Bm is the microbial biomass and Am the microbial activity per biomass unit. The
microbial biomass in soils (Bm) is well correlated to the soil organic matter content
(Zak et al., 1990; Zhang and Zhang, 2003; Milne and Haynes, 2004), and under nonlimiting substrate and oxygen conditions, Am is dependent on the temperature and soil
water content.
Equation 18 can explain the discrepancy between Paper II, where C and N pool sizes
did not seem to have an influence on N mineralization and immobilization rates, and
e.g. the study by Barret and Burke (2000), in which the soil C content alone could
14
explain 60-70% of the variation of N mineralization and immobilization rates. Barret
and Burke (2000) used laboratory incubations with high and constant temperature and
soil water content to determine N transformation rates. In this way, the variation in Am
among soils was restrained and the N mineralization and immobilization rates
dependent on Bm alone. Since Bm is dependent on the soil organic matter content, their
results could have been predicted from Equation 18. In paper II, on the other hand, C
and N mineralization and N immobilization rates were determined in situ and Am was
probably more variable than Bm due to variations in soil water content and
temperature, which influence mineralization and immobilization rates (Puri and
Ashman, 1998; Zaman et al., 1999). Variations in Am may also explain the
dependence of N mineralization and immobilization rates on the microbial biomass in
some studies (Davidson et al., 1992; Hart et al., 1997), but not in others (Puri and
Ashman, 1998; Zaman et al., 1999). The commonly used chloroform fumigation
method extracts the total microbial biomass C rather than the active part of the soil
microbial community (Ross, 1991; van de Werf and Verstraete, 1987) and would fail
to discriminate between two bacterial populations with a similar biomass but varying
growth rates (Bloem et al., 1989).
Restrictions on the microbial biomass
Equation 18 includes the microbial biomass and activity as factors determining the
substrate utilization rate. Schimel and Weintraub (2003) demonstrated that a model
assuming exoenzyme catalysed SOM decomposition only had a knife-edge
equilibrium depending on the fundamental rate constant of the substrate (K). With
high K values, the microbial population grew out of control, and with low it crashed.
In reality though, the microbial biomass remains rather constant from day to day and
in some cases even from week to week (Witter, et al., 1993; Lovell and Hatch, 1998;
Puri and Ashman, 1998), while the microbial activity may vary considerably on the
same timescale (Bloem et al., 1989; Lovell and Hatch, 1998). There therefore have to
be mechanisms that restrain biomass variations. When conditions are unfavourable,
some soil bacteria form spores, and soil microorganisms can remain dormant for a
long period of time, resulting in a relatively small decrease in microbial biomass in
response to stress. Predation may be an important factor setting an upper limit for the
microbial biomass (Ekelund and Rønn, 1994; Zhang et al., 2000; Rønn et al., 2002).
The higher microbial biomass in soils with high SOM content might be a result of
protection of bacteria against predation by protozoa, since it is generally accepted that
fine textured soils offer greater protection than coarse textured soils (Alexander, 1981;
Rutherford and Juma, 1992).
The response of soil microorganisms to C and N additions
One objection to use zero order approximations to describe microbial C and N
mineralization and utilization may be that addition of C and N to a soil often enhances
either respiration or N immobilization rates or both (Mikan et al., 2000; Fisk and
Fahey, 2001; Vance and Chapin, 2001). Justus von Liebig's Law of the Minimum
states that yield is proportional to the amount of the most limiting nutrient, whichever
nutrient it may be. If a deficient nutrient is supplied, yields may be improved to the
point that some other nutrient is needed in greater quantity than the soil can provide,
15
and the Law of the Minimum would apply to that nutrient. Therefore, if
mineralization and immobilization rates are assumed to be independent of substrate
concentrations, microorganisms cannot be C or N limited and should not respond to C
or N additions. However, differences in substrate quality may induce an apparent C or
N limitation, although microorganisms are not strictly C or N limited according to the
definition. The response of soil microorganisms to C or N additions when zero order
kinetics are assumed can be elucidated by dividing Equation 18 into one C part and
one N part:
dC/dt = -KCBmAm
(19)
dN/dt = -KNBmAm
(20)
where KC is the fundamental rate constant for soil C, and KN the fundamental rate
constant for soil N. If KC > KN, microorganisms would appear to be N limited, and C
limited if KC < KN. The theoretical response of C and N mineralization and
immobilization rates to C and N additions in any of those cases is summarized in
Table 1.
Table 1. The theoretical response of C and N mineralization and immobilization rates to additions of C
and N with different quality if KC > KN (N limitation) and if KC < KN (C limitation). KAC is the
fundamental rate constant for the added C, and KAN the fundamental rate constant for the added N.
Assumption
1. KC > KN (N limitation)
2.
3.
4.
5. KC < KN (C limitation)
6.
7.
8.
Addition
C
N
C
N
Assumption
KAC > KC(1)
KAC < KC(2)
KAN > KN(3)
KAN < KN(2)
C ass.
→
→
↑
→
C min.
↑
→
↓
→
N ass.
→
→
↑
→
N min.
→
→
↑
→
KAC > KC(4)
KAC < KC(2)
KAN > KN(5)
KAN < KN(2)
↑
→
→
→
↑
→
→
→
↑
→
→
→
↑
→
→
→
(1) Microorganisms growing under nutrient limited conditions take up C compounds, resulting in
elevated respiration rates that are not associated with growth (overflow metabolism; Schimel and
Weintraub, 2003 and references therein).
(2) No response is detected since the fundamental rate constant is lower for the added C than for soil C
or N,
(3) Addition of N results in reduced overflow metabolism, resulting in a lower respiration rate and
higher C assimilation rate.
(4) A higher C utilization rate results in higher N utilization rates according to Equation 8.
(5) The synthesis and/or activation of proteins involved in the uptake and utilization of nitrogenous
compounds is closely regulated in concert with the availability of their substrates (Merrick and
Edwards, 1995 and references therein). That is, if the intracellular N content of the
microorganisms is sufficient, no additional N is assimilated, and no enzymes are produced to
mineralize organic N. Hence, no response to addition of N, even if KAN > KN, under C limited
conditions.
Table 1 show that addition of an easily available C source can always be expected to
increase respiration rates (Assumption 1 and 5). Such responses have been
documented numerous times. In fact, the substrate induced respiration method
(Anderson and Domsch, 1978), in which the magnitude of the increased CO2
production in response to glucose additions is used to determine microbial biomass in
soils, depends on this assumption. Addition of N to a N limited soil increases the C
use efficiency (Fisk and Fahey, 2001; Lovell and Hatch, 1998), resulting in decreased
16
respiration rates (Assumption 3). Increased respiration rates in response to C additions
(Mikan et al., 2000; Vance and Chapin, 2001) and decreased in response to N
additions (Fisk and Fahey, 2001) have commonly been interpreted as proof for C
limitation of the microbial biomass in soils. If zero order kinetics are assumed and the
theoretical responses in Table 1 to N additions valid this might not necessarily be the
case, and changes in respiration rates in response to C or N additions may not be a
reliable method to assess if microorganisms are C or N limited. Then again, increased
N immobilization rates as a response to N additions always seem to be indicative of N
limitation (Table 1, Assumption 3 and 7).
The influence of plants on N mineralization and immobilization rates
Theoretically plants may influence N mineralization and immobilization in several
ways. Plant litter is the major source of organic C and N in soils. Easily available C
sources in litter leachate, such as simple fatty acids, organic acids and carbohydrates
may stimulate the microbial activity (Henriksen & Breland, 1999, Hongve et al.,
2000). Root exudates also consist of readily available C and N sources, e.g. amino
acids, vitamins, enzymes and nucleotides (Graystone et al., 1996; Uren, 2001 in The
Rhizosphere eds. Pinton et al), which may have a stimulatory effect on the microbial
activity and C and N transformation rates.
Specific compounds in plant litter may also reduce the rate of N mineralization and
immobilization. Some polyphenolic compounds have an inhibitory effect on the
growth and activity of microorganisms, and the binding of proteins to polyphenolic
compounds might render the proteinacious N inaccessible to heterotrophic
microorganisms, thereby decreasing the N mineralization rate (Northup, 1995; Krauss
et al., 2004). Although plant N uptake generally is much lower than microbial N
uptake (reviewed by Kaye and Hart, 1997), it may sometimes even exceed microbial
uptake (Paper IV; Norton and Firestone, 1996). Plants may therefore compete with
microorganisms for N (Kaye and Hart, 1997), resulting in lower microbial activity C
and N transformation rates.
In Paper III and IV, I tried to elucidate the effect of plants on the microbial activity
and N mineralization and immobilization rates and vice versa. In Paper IV, the
bacterial activity seemed to be the main factor determining the microbial N uptake,
while plants had no effect. Plants stimulated the bacterial activity only when
fragmented litter had been added to a soil, resulting in low inorganic N
concentrations. However, microbial N uptake decreased in one of the three soils in
presence of plants but since the decreased microbial N uptake did not coincide with a
corresponding increase in plant N uptake or decreased bacterial activity it might be
that the plants specifically inhibited fungal activity and N uptake. Hence, the bacterial
activity was the main factor determining microbial N immobilization rates, while
plants only seemed to have minor effects on the microbial activity and N
immobilization rates. This is consistent with the hypothesis that C and N
mineralization and immobilization rates are independent of substrate concentrations.
In Paper III, I found that spatial patterns of N transformation rates were more related
to the spatial pattern of the cover of plants than to the species composition. Other
studies seem to confirm that plants promote N mineralization and immobilization
17
rates (Jackson and Caldwell, 1993; Schlesinger, 1996), supposedly by root exudates
that provide microorganisms with high quality C and N sources (high KC and KN
values). Similarly, different K values for oak and beech litter, reflecting the lignin
content which is higher in beech than in oak (Ferrari, 1999; Sariyildiz and Anderson,
2003a and 2003b) and inversely related to the decomposition rate (Sariyildiz and
Anderson, 2003a and 2003b), may explain the different influence of those tree species
in N mineralization and immobilization rates (Paper III).
Abiotic immobilization of N
N can be immobilized by fixation to soil minerals or direct incorporation into the soil
organic matter by chemical reactions. Fixation of NH4+ to clay minerals has been
shown to occur in a wide variety of soils. The fixation process is fast (<30 minutes),
but usually <10 % of the added NH4+ is fixed, even though cases of more than 25 %
NH4+ fixation have been reported (Drury and Beauchamp, 1991; Trehan, 1996). The
NH4+ fixation is positively correlated to the vermiculate clay content and to the
amount of added NH4+ (Keerthisinghe et al., 1984). Consumption of NH4+ by biotic
immobilization or nitrification seems to cause the release of fixed N (Drury and
Beauchamp, 1991; Trehan, 1996), and inhibition of the nitrification process has been
reported to enhance the NH4+ fixation (Juma and Paul, 1983). Therefore, at least a
part of the clay-fixed NH4+ seems to be reversibly bound, but the release of the fixed
NH4+ is much slower than the fixation (Drury and Beauchamp, 1991).
NH3 and other amines can react with soil organic matter to form stable organic N
complexes (Thorn and Mikita, 1992). Little is known about the mechanism involved
in NH3 fixation, but the reaction is favoured by high pH values and associated with
uptake of O2 (Stevenson, 1982). Several possible mechanisms have been proposed,
including formation of quinones capable of reacting with NH3 to form complex
polymers. NH3 also reacts with reducing sugars, and with a wide variety of ketones,
aldehydes and other carbonyl- containing compounds under alkaline conditions
(Stevenson, 1982). However, because of the high pKa of NH3 (9.25) and since NH3 is
the reactive form of N (Johnson et al., 2000), these reactions are unlikely to occur in
most natural soils.
Abiotic vs. biotic immobilization of N
Biotic N immobilization processes rather than chemical reactions seem to be
responsible for the majority of N incorporated into the soil organic matter fraction.
More 15N is recovered in the organic matter in 15NH4+ treated non-sterile soils
compared to autoclaved soils after incubation times as short as 30 minutes (Trehan,
1996). This difference is enhanced with longer incubation times. Nuclear magnetic
resonance (NMR) spectra of the organic matter in 15NH4+ or 15NO3- treated soils have
also revealed that the resulting organic 15N fraction consists almost exclusively of
peptides and proteins (>80%), nucleic acids, and aliphatic amine groups (Clinton et
al., 1995). Since the 15N is recovered in partly humified organic matter rather than in
poorly transformed plant fragments, it is possible to attribute the incorporation of 15N
into the organic matter to microbial activity, at least during incubations shorter than a
few months. The similarity of 15N spectra for soils treated with 15NH4+ and 15NO3-,
18
and the enhanced incorporation of 15N into the organic matter in non-sterile soils
compared to autoclaved soils also speaks against direct incorporation of 15N into
organic matter by chemical reactions.
The amount of NH4+ fixed in clay minerals compared to the amount incorporated in
organic matter varies widely depending on soil type. Trehan (1996) found the ratio
between clay fixed 15N and organic 15N 30 minutes after 15NH4+ addition to vary
between 0.15 and 2.5 in three agricultural soils. The distribution of 15N between the
clay and organic matter fractions in the three soils also varied widely over the
following 20 days, with no repeatable pattern among soils. Drury and Beauchamp
(1991) found the fixed and organic matter 15N fractions to be of equal size 6 hours
after 15NH4+ addition in two clay loams with different fixation capacity. However,
while the amount of 15N in the organic matter fraction remained fairly constant or
only increased slowly after this, the amount of 15N in the fixed fraction doubled
during the following 3 days. Johnson and Burke (2000) found an increase in the
relative contribution of abiotic N immobilization to the total N immobilization with
increasing N concentrations in a number of forest soils differing in N status. Abiotic
N immobilization accounted for 6-90% of the total N immobilization in these soils
after 12 hours incubation with 15N. Because of the high variability in experimental
data depending on the soil type used and the lack of theories able to explain this
variation, no model exists that can accurately describe the distribution and exchange
of N between the two pools.
Mobilization of microbially bound N
The soil microbial biomass often has short turnover times (Joergensen et al., 1990;
Hart et al., 1994; Bååth, 1998; Vance and Chapin, 2001). In Paper V we demonstrated
the flow of C between the microbial biomass, DOC and SOM was well balanced.
Several mechanisms can be responsible for the release of microbial C and N. During
bacterial growth there is a high production and breakdown of RNA (Mason and Egli,
1993) and proteins. The resulting compounds can either be reused by the bacteria or
excreted as organic nitrous compounds (e.g. urea), or after further degradation as
NH4+. Degradation of mRNA in particular is high during exponential growth (Norris
and Koch, 1972), and it is likely that the same is true for proteins. However, as long
as the bacteria are not C or N limited, a lot of the degradation products may be used to
synthesize new RNA or protein, resulting in a low excretion rate of nitrous
compounds. As carbon gets exhausted, the degradation rate of RNA gets even higher
(Mason and Egli, 1993), leading to a decrease in bacterial RNA content and a higher
excretion rate of nitrous compounds (Therkildsen et al., 1997). In Paper VI, I
demonstrated that bacteria have a high ability to recycle N intracellularly, especially
at low N concentrations and growth rates. Therefore, remineralization of microbial N
by other processes than predation or soil drying/rewetting (discussed below) is likely
to occur only when conditions promote high growth rates, i.e. when microorganisms
are not substrate limited and temperature and moisture conditions are favourable.
Soil drying and rewetting may be important in mobilizing micobially bound N. There
is no active transport mechanism for water in microorganisms, so water flows freely
in and out of the cell. In order to avoid lysis bacteria have to maintain turgor in
balance with the surrounding environment. During osmotic upshift this is
19
accomplished by accumulating osmotically active solutes in the cytoplasm, preferably
by active uptake or synthesis of so-called compatible solutes; compounds that are
highly congruous with cellular functions (Kempf and Bremer, 1998). These include
free amino acids and derivates thereof, as well as quaternary amines. If osmoprotected
cells are subjected to osmotic downshock, such as a heavy rainfall, water flows into
the cells because of the high concentration of solutes in the cytoplasm. This is a very
rapid process (<1 min), and to avoid swelling and bursting the microbes will have to
respond accordingly. Rapid and extensive solute efflux as a result of osmotic
downshift has been shown to occur in several bacterial species (Wood, 1999). This
gives a mechanistic explanation to the well-known phenomena that rewetting of an
air-dried soil results in a flush of C and N mineralization (Paper I; Birch, 1958;
Agarwal et al., 1971; van Gestel et al., 1991; Pulleman and Tietema, 1999; Fierer and
Schimel, 2003). Drying and rewetting release physically protected SOM, increasing
the amount of extractable C by up to 200% (Fierer and Schimel, 2003). Bacterial cells
with a high growth rate and increased efficiency of translation are favoured in rich
environments (Nyström, 2002). However, the payoff for this increased rate of
reproduction is a reduced ability to withstand starvation (Kurland and Mikkola, 1993),
and oxidation of proteins in aging growth-arrested cells may lead to proteolysis
(Nyström, 2002). Remineralization of a growth-arrested fast growing microbial
community may also explain the secondary peek in respiration and N mineralization
rates that is sometimes observed a few days after rewetting of an air-dried soil (Paper
I; Pulleman and Tietema, 1999)
Predation may also be an important factor controlling the recycling of microbially
bound N. High protozoan activity coincides with high mineralization and
immobilization rates and short turnover times of the microbial biomass (Christensen
et al., 1996; Rønn et al, 2002). Similarly, there is a positive correlation between the
abundance of nematodes and the microbial activity (Mamilov and Dilly, 2002), and
addition of earthworms to a soil decreases the total microbial biomass but increases its
activity (Zhang et al., 2000).
Environmental concerns
More N is fixed annually by human activities (≈140 Tg N yr-1) than by natural
processes (110-125 Tg N yr-1; Vitousek, 1994; Jefferies and Maron, 1997). In
addition, N have been mobilized from long term storage pools through biomass
burning, land clearing and conversion, and drainage of wetlands. Of particular
concern are the potential ecological effects of sustained atmospheric nitrogen
deposition in natural environments (Jefferies and Maron, 1997). Chronic N deposition
rates of up to 50% of the assimilatory N need by trees (van Oene et al., 2000)
combined with low in situ net mineralization rates have led to concerns that forest
ecosystems may get N saturated (Aber, 1992), resulting in increased mobility and
leaching of N.
Large-scale experiments such as NITREX and EXMAN, in which N has been added
or removed from the ambient deposition at eight sites spanning a gradient of N
deposition across Europe, were initiated during the 1990s to evaluate the effect of N
deposition on ecosystem processes (Wright and van Breemen, 1995; Wright and
Rasmussen, 1998). Unfortunately, none of 36 or so published NITREX studies
20
Box 1. A conceptual model describing the flow of C and N (solid arrows) in soil. The
main factors determining the rate of these processes are the size of the microbial biomass,
its activity, the quality of the substrate they are growing on, predation, and dryingrewetting cycles, as indicated for each process below. The size of the microbial biomass
(Bm) and its activity (Am) in turn seem to be dependent on soil SOM content (Bm) and
temperature and soil water content (Am).
CO2
1. Decomposition
Dissolved
organic carbon
(DOC)
2. C uptake
3.Respiration
4. C mobilization
Soil organic
matter
(SOM)
8. Death/loss
7. N mobilization
5. Decomposition
Dissolved
nitrogen
(DN)
Microbial
biomass
(Bm)
6. N immobilization
Factors affecting rate 1 to 8:
1. Microbial biomass (Bm), activity per biomass unit (Am), and the fundamental rate
constant for SOM-C (KSOM-C).
2. Microbial biomass (Bm), activity per biomass unit (Am), and the fundamental rate
constant for DOC (KDOC).
3. C uptake rate and microbial C use efficiency.
4. Drying-rewetting cycles, predation, and the microbial biomass (Bm) and activity per
biomass unit (Am)
5. Microbial biomass (Bm), activity per biomass unit (Am), the fundamental rate constant
for SOM-N (KSOM-N), and the intracellular N status of the microorganisms
6. Microbial biomass (Bm), activity per biomass unit (Am), the fundamental rate constant
for DN (KDN), and the intracellular N status of the microorganisms
7. Drying-rewetting cycles, predation, and the microbial biomass (Bm) and activity per
biomass unit (Am)
8. Predation, drought etc.
21
examined the difference in gross mineralization or immobilization rates between
treatment and control plots, but only among sites. Those rates vary between 9-18 mg
N kg soil-1 d-1 (Tietema, 1998). In Paper II and III, I found similar in situ N
mineralization rate with an average of 15.3 mg N kg-1 d-1. If those rates are
recalculated to g m-2 day (using a soil density of 1 g cm-3), the gross N mineralization
becomes 0.75 g m-2 day in the top 5 cm of the soil. The natural N deposition in southwestern Scania, 0.0054 g m-2 day (20 kg ha-1, Lagner et al., 1996), and the N addition
in the NITREX experiments, 0.0096 g m-2 day (Gundersen et al., 1998), only
represent 0.7-1.3% of the ammonium that was naturally produced. That is, a reduction
in the gross N mineralization rate of about 1% would be enough to compensate for the
addition of inorganic N. This decrease would hardly be detectable given the great
spatial and temporal variability of N transformation rates (Paper I and III). Chronic N
deposition may still be a problem. About 50% of the deposited N is in the form of
NOX, and although it is now well documented that soil microorganisms immobilize
NO3- even in the presence of NH4+, NO3- immobilization rates are low compared to
NH4+ immobilization rates. N leakage due to NOX deposition, especially during
periods of low microbial activity, may therefore still be a problem.
Conclusions
From the results in Paper I-VI and the cited literature I conclude that the main factors
determining microbial mobilization and immobilization of soil N are the size of the
microbial biomass, its activity, and the quality of the substrate they are growing on.
Those three factors in turn seem to be dependent on soil SOM content, predation,
drying-rewetting cycles, temperature, and soil water content. This conceptual view is
summarized in Box 1.
Acknowledgements I am grateful to Göran Bengtsson and Ursula Falkengren-Grerup for helpful
comments on a previous draft. The work included in the thesis was supported by grants from the
Foundation for Strategic Environmental Research in Sweden.
Refereces
Aber, J.D. 1992. Nitrogen cycling and nitrogen saturation in temperate forest
ecosystems. Trends in Ecology and Evolution 7, 220-223.
Aber, J., McDowell, W., Nadelhoffer, K., Magill, A., Berntson, G., Kamakea, M.,
McNulty, S., Currie, W., Rustad, L., Fernandez, I. 1998. Nitrogen saturation in
temperate forest ecosystems - Hypotheses revisited. Bioscience 48, 921-934
Alexander, M. 1981. Why microbial predators and parasites do not eliminate their
prey and hosts. Annual Reviews of Microbiology 35, 113-133.
Agarwal, A.S., Singh, B.R., Kanehiro, Y., 1971. Soil nitrogen and carbon
mineralization as affected by drying-rewetting cycles. Soil Science Society of
America Proceedings 35, 96-100.
22
Andersen, M.K., Jensen, L.S. 2001. Low soil temperature effects on short term gross
N mineralization-immobilization turnover after incorporation of green manure. Soil
Biology and Biochemistry 33, 511-521.
Anderson, J.P.E., Domsch, K.H. 1978. A physiological method for the quantitative
measurement of microbial biomass in soil. Soil Biology and Biochemistry 10, 215221.
Barret, J.E., Burke I.C. 2000. Potential nitrogen immobilization in grassland soils
across a soil organic matter gradient. Soil Biology and Biochemistry 32, 1707-1716.
Bekins, B.A., Warren, E., Godsy, M. 1998. A comparison of zero-order, first-order,
and Monod biotransformation models. Ground Water 36, 261-268.
Birch, H.F. 1958. The effect of soil drying on humus decomposition and nitrogen
availability. Plant and Soil 10, 9-31.
Bloem, J., Ellenbroek, F.M., Baer-Gilissen, M.J., Cappenberg, T.E. 1989. Protozoan
grazing and bacterial production in stratified Lake Vechten estimated with
fluorescently labeled bacteria and by thymidine incorporation. Applied and
Environmental Microbiology 55, 1787-1795.
Bloem, J., de Ruiter, P., Koopman, C.G., Lebbink, J.G., Brussaard, L. 1992.
Microbial numbers and activity in dried and rewetted arable soil under integrated and
conventional management. Soil Biology and Biochemistry 24, 655-665.
Bradbury, N.J., Whitmore, A.P., Hart, P.B.S., Jenkinson, D.S. 1993. Modelling the
fate of nitrogen in crop and soil in the years following application of 15N-labelled
fertilizer to winter wheat. Journal of Agricultural Science 121, 363-379.
Bååth, E. 1998. Growth rates of bacterial communities in soils at varying pH: A
comparison of the thymidine and leucine incorporation techniques. Microbial Ecology
36, 316-327.
Christensen, S., Rønn, R., Ekelund., F., Andersen, B., Damgaard, J., Friberg-Jensen,
U., Jensen, L., Kiil, H., Larsen, B., Larsen, J., Riis, C., Thingsgaard, K., Thirup, C.,
Petersen, T.A., Vesterdal, L. 1996. Soil respiration profiles and protozoan
enumeration agree as microbial growth indicators. Soil Biology and Biochemistry 28,
865-868.
Clinton, P.W., Newman, R.H., Allen, R.B. 1995. Immobilization of 15N in forest litter
studied by 15N CPMAS NMR spectroscopy. European Journal of Soil Science 46,
551-556.
Contin, M., Corcimaru, S., De Nobili, M., Brookes, P.C. 2000. Temperature changes
and the ATP concentration of the soil microbial biomass. Soil Biology and
Biochemistry 32, 1219-1225.
Davidson, E.A., Hart, S.C., Firestone, M.K. 1992. Internal cycling of nitrate in soils
of a mature coniferous forest. Ecology 73, 1148-1156.
23
Diaz-Ravina, M., Acea, M.J., Carballas, T. 1993. Microbial biomass and C and N
mineralization in forest soils. Bioresource Technology 43, 161-167.
Drury, C.F., Beauchamp, E.G. 1991. Ammonium fixation, release, nitrification, and
immobilization in high- and low-fixing soils. Soil Science Society of America Journal
55, 125-129.
Ekelund,F., Rønn, R. 1994. Notes on protozoa in agricultural soil with emphasis on
heterotrophic flagellates and naked amoebae and their ecology. FEMS Microbiology
Reviews 15, 321-353.
Emmett, B.A., Brittain, S.A., Hughes, S., Kennedy, V. 1995. Nitrogen additions
(NaNO3 and NH4NO3) at Aber forest, Wales: II. Response of trees and soil nitrogen
transformations. Forest Ecology and Management 71, 61-73.
Ferrari, J.B. 1999. Fine-scale patterns of leaf litterfall and nitrogen cycling in an oldgrowth forest. Canadian Journal of Forest Research 29, 291-302.
Fierer, N., Schimel, J.P. 2003. A proposed mechanism for the pulse in carbon dioxide
production commonly observed following the rapid rewetting of a dry soil. Soil
Science Society of America Journal 67, 798-805.
Fisk, M.C., Fahey, T.J. 2001. Microbial biomass and nitrogen cycling responses to
fertilization and litter removal in young northern hardwood forests. Biogeochemistry
53, 201-223.
Gazzarrini, S., Lejay, T., Gojon, A., Ninnemann, O., Frommer, W.B., von Wiren, N.
1999. Three functional transporters for constitutive, diurnally regulated, and
starvation-induced uptake of ammonium into arabidopsis roots. Plant Cell 11, 937947.
Giblin, A.E., Nadelhoffer, K.J., Shaver, G.R., Laundre, J.A., McKerrow, A.J. 1991.
Biogeochemical diversity along a riverside toposequence in arctic Alaska. Ecological
Monographs 61, 415-435.
Grayston, S.J., Vaughan, D., Jones, D. 1996. Rhizosphere carbon flow in trees, in
comparison with annual plants: the importance of root exudation and its impact on
microbial activity and nutrient availability. Applied Soil Ecology 5, 29-56.
Gundersen, P., Emmett, B.A., Kjønaas, O.J., Koopmans, C.J. and Tietema, A. 1998.
Impact of nitrogen deposition on nitrogen cycling in forests: a synthesis of NITREX
data. Forest Ecology and Management 101, 37-55.
Hadas, A., Molina, J.A.E., Feigenbaum, S., Clapp, C.E. 1987. Simulation of nitrogen15 immobilization by the model NCSOIL. Soil Science Society of America Journal
51, 451-455.
24
Hadas, A., Molina, J.A.E., Feigenbaum, S., Clapp, C.E. 1992. Factors affecting
nitrogen immobilization in soil as estimated by simulation models. Soil Science
Society of America Journal 56, 1481-1486.
Hart, S.C., Binkley, D., Perry, D.A. 1997. Influence on red alder on soil nitrogen
transformations in two conifer forests of contrasting productivity. Soil Biology
Biochemistry 29, 1111-1123.
Hart, S.C., Nason, G.E., Myrold, D.D., Perry, D.A. 1994. Dynamics of gross nitrogen
transformations in an old-growth forest: the carbon connection. Ecology 75, 880-891.
Helling, R.B. 1998. Pathway choice in glutamate synthesis in Escherichia coli.
Journal of Bacteriology 180, 4571-4575.
Hendrickson, O.Q. 1985. Variation in the C:N ratio of substrate mineralized during
forest humus decomposition. Soil Biology and Biochemistry 17, 435-440.
Henriksen, T.M, Breland, T.A. 1999. Evaluation of criteria for describing crop residue
degradability in a model of carbon and nitrogen turnover in soil. Soil Biology and
Biochemistry 31, 1135-1149.
Hongve, D., van Hees, P.A.W., Lundström, U.S. 2000. Dissolved components in
precipitation water percolated through forest litter. European Journal of Soil Science
51, 667-677.
Jackson, R.B., Caldwell, M.M. 1993. Geostatistical patterns of soil heterogeneity
around individual perennial plants. Journal of Ecology 81, 683-692.
Janssen, B.H. 1996. Nitrogen mineralization in relation to C:N ratio and
decomposability of organic materials. Plant and Soil 181, 39-45.
Jayakumar, A., Barnes, E.M. 1983. A filtration method for measuring cellular uptake
of [14C]methylamine and other highly permeant solutes. Analytical Biochemistry 135,
475-478.
Jayakumar, A., Epstein, W., Barnes, E.M. 1985. Characterization of ammonium
(methylammonium) potassium antiport in Escherichia coli. Journal of Biological
Chemistry 260, 7528-7532.
Jefferies, R.L., Maron J.L. 1997. The embarrassment of riches: Atmospheric
deposition of nitrogen and community and ecosystem processes. Trends in Ecology
and Evolution 12, 74-78.
Joergensen, R.G., Brookes, P.C., Jenkinson, D.S. 1990. Survival of the soil microbial
biomass at elevated temperatures. Soil Biology and Biochemistry 22, 1129-1136.
Johnson, D.W., Cheng, W., Burke, I.C. 2000. Biotic and abiotic nitrogen retention in
a variety of forest soils. Soil Science Society of America Journal 64, 1503-1514.
25
Juma, N.G., Paul, E.A. 1983. Effect of a nitrification inhibitor on n immobilization
and release of 15N from non-exchangeable ammonium and microbial biomass.
Canadian Journal of Soil Science 63, 167-175.
Kaye, J.P., Hart, S.C. 1997. Competition for nitrogen between plants and soil
microorganisms. Trends in Ecology and Evolution 12, 139-143.
Kempf, B., Bremer, E. 1998. Uptake and synthesis of compatible solutes as microbial
stress responses to high-osmolarity environments. Archives of Microbiology 170,
319-330.
Kersebaum, K.C., Richter, J. 1990. Modelling nitrogen dynamics in a plant-soil
system with a simple model for advisory purposes. Fertilizer Research 27, 273-291.
Keerthisinghe, G., Mengel, K., Dedatta, S.K. 1984. The release of nonexchangeable
ammonium (15N labeled) in wetland rice soils. Soil Science Society Of America
Journal 48, 291-294.
Kilham, K. 1994. Soil Ecology. Cambridge University Press, Cambridge.
Kjønaas, O.J., Stuanes, A.O., Huse, M. 1998. Effects of weekly nitrogen additions on
N cycling in a coniferous forest catchment, Gårdsjön, Sweden. Forest Ecology and
Management 101, 227-249.
Kraus, T.E.C., Zasoski, R.J., Dahlgren, R.A., Horwath, W.R., Preston, C.M. 2004.
Carbon and nitrogen dynamics in a forest soil amended with purified tannins from
different plant species. Soil Biology and Biochemistry 36, 309-321.
Kurland, C.G., Mikkola, R. 1993. The impact of nutritional states on the
microevolution of ribosomes. In: Kjellberg, S. (Editor). Starvation in Bacteria,
Phlenum Press, New York, pp.81-102.
Lagner, J., Persson, C., Robertson, L., Ullerstig, A. 1996. Air pollution assessment
study using the MATCH modelling system – Applications to sulphur and nitrogen
compounds over Sweden 1994. SMHI RMK Report, Norrköping.
Lovell, R.D., Hatch, D.J. 1998. Stimulation of microbial activity following spring
applications of nitrogen. Biology and Fertility of Soils 26, 28-30.
Lovell, R.D., Jarvis, S.C., Bardgett, R.S. 1995. Soil microbial biomass and activity in
long-term grassland: effects of management changes. Soil Biology and Biochemistry
27, 969-975.
Mamilov, A.Sh., Dilly, O.M. 2002. Soil microbial eco-physiology as affected by
short-term variations in environmental conditions. Soil Biology and Biochemistry 34,
1283-1290.
Marschner, H. 1995. Mineral Nutrition of Higher Plants, Academic Press, London.
26
Marini, A.M., Soussi-Boudekou, S., Vissers, S., Andre, B. 1997. A family of
ammonium transporters in Saccharomyces cerevisiae. Molecular and Cellular Biology
17, 4282-4293.
Mason, C.A., Egli, T. 1993. Dynamics of microbial growth in the decelerating and
stationary phase of batch culture. In: Kjellberg, S. (Editor). Starvation in Bacteria,
Phlenum Press, New York, pp.81-102.
Matsumoto, S., Noriharu, A., Yamagata, M. 2000. Extraction of mineralizable organic
matter from soils by a neutral phosphate buffer solution. Soil Biology and
Biochemistry 32, 2000.
Meletzus, D., Rudnick, P., Doetsch, N., Green, A., Kennedy, C. 1998.
Characterization of the glnK-amtB Operon of Azotobacter vinelandii. Journal of
Bacteriology 180, 3260-3264.
Merrick, M.J., Edwards, R.A. 1995. Nitrogen control in bacteria. Microbiological
Reviews 59, 604-622.
Michel-Reydellet, N., Desnoues, N., de Zamaroczy, M., Elmerich, C., Kaminski, P.A.
1998. Characterisation of the glnK-amtB operon and the involvement of AmtB in
methylammonium uptake in Azorhizobium caulinodans. Molecular and General
Genetics 258, 671-677.
Mikan, C.J., Zak, D.R., Kubiske, M.E., Pregitzer, K.S. 2000. Combined effects of
atmospheric CO2 and N availability on the belowground carbon and nitrogen
dynamics of aspen mesocosms. Oecologia 124, 432-445.
Milne, R.M., Haynes, R.J. 2004. Soil organic matter, microbial properties, and
aggregate stability under annual and perennial pastures. Biology and Fertility of Soils
39, 172-178.
Molina, J.A.E., Clapp, C.E., Shaffer, M.J., Chichester, F.W., Larson, W.E. 1983.
NCSOIL, a model of nitrogen and carbon transformations in the soil: Decomposition,
calibration and behaviour. Soil Science Society of America Journal 47, 85-91.
Molina, J.A.E., Hadas, A., Clapp, C.E. 1990. Computer simulation of nitrogen
turnover in soil and priming effect. Soil Biology and Biochemistry 22, 349-353.
Montesinos, M.L., Muro-Pastor, A.M., Herrero, A., Flores, E. 1998.
Ammonium/Methylammonium permeases of a cyanobacterium - Identification and
analysis of three nitrogen-regulated amt genes in Synechocystis sp. PCC 6803. Journal
of Biological Chemistry 273, 31463-31470.
Nadelhoffer, K.J., Aber, J.D., Melillo, J.M. 1984. Seasonal patterns of ammonium and
nitrate uptake in nine temperate forest ecosystems. Plant and Soil 80, 321-335.
Ninnemann, O., Jauniaux, J.C., Frommer, W.B. 1994. Identification of a high-affinity
NH4+ transporter from plants. EMBO Journal 13, 3464-3471.
27
Northup, R.R., Yu, Z.S., Dahlgren, R.A., Vogt, K.A. 1995. Polyphenol control of
nitrogen release from pine litter. Nature 377, 227-229.
Norton, J.M., Firestone, M.K. 1996. N dynamics in the rhizosphere of Pinus
ponderosa seedlings. Soil Biology and Biochemistry 28, 351-362.
Norris, T.E., Koch, A.L. 1972. Effect of growth rate on the relative rates of synthesis
of messenger, ribosomal and transfer RNS in Escherichia coli. Journal of Molecular
Biology 64, 633-649.
Nyström, T. 2002. Aging in bacteria. Current opinion in Microbiology 5, 596-601.
Parton, W.J., Schimel, D.S., Cole, C.V., Ojima, D.S. 1987. Analysis of factors
controlling soil organic matter levels in great plains grasslands. Soil Science Society
of America Journal 51, 1173-1179.
Pulleman, M., Tietema, A. 1999. Microbial C and N transformations during drying
and rewetting of coniferous forest floor material. Soil Biology and Biochemistry 31,
275-285.
Puri, G.M., Ashman, R. 1998. Relationship between soil microbial biomass and gross
N mineralization. Soil Biology and Biochemistry 30, 251-256.
Recous, S., Mary, B. 1990. Microbial immobilization of ammonium and nitrate in
cultivated soils. Soil Biology and Biochemistry 22, 913-922.
Recous, S., Aita, C., Mary, B. 1999. In situ changes in gross N transformations in bare
soil after addition of straw. Soil Biology and Biochemistry 31, 119-133.
Ross, D.J. 1991. Microbial biomass in a stored soil: A comparison of different
estimation procedures. Soil Biology and Biochemistry 23, 1005-1007.
Rutherford, P.M., Juma, N.G. 1992. Influence of texture on habitat pore space and
bacterial-protozoan populations in soil. Biology and Fertility of Soils 12, 221-227.
Rønn, R., Gavito, M., Larsen, J., Jakobsen, I., Fredriksen, H., Christensen, S. 2002.
Response of free-living soil protozoa and microorganisms to elevated atmospheric
CO2 and presence of mycorrhiza. Soil Biology and Biochemistry 34, 923-932.
Sariyildiz, T., Anderson, J.M. 2003a. Interactions between litter quality,
decomposition and soil fertility: a laboratory study. Soil Biology and Biochemistry
35, 391-399.
Sariyildiz, T., Anderson, J.M. 2003b. Decomposition of sun and shade leaves from
three deciduous tree species, as affected by their chemical composition. Biology and
Fertility of Soils 37, 137-146.
Sallih, Z., Pansu, M. 1993. Modelling of soil carbon forms after organic amendment
under controlled conditions. Soil Biology and Biochemistry 25, 1755-1762.
28
Schimel, J.P., Firestone, M.K. 1989. Inorganic nitrogen incorporation by coniferous
forest floor material. Soil Biology and Biochemistry 21, 41-46.
Schimel, J.P., Weintraub, M.N. 2003. The implications of exoenzyme activity on
microbial carbon and nitrogen limitation in soil: a theoretical model. Soil Biology and
Biochemistry 35, 549-563.
Schlesinger, W.H., Raikes, J.A., Hartley, A.E., Cross, A.F. 1996. On the spatial
pattern of soil nutrients in desert ecosystems. Ecology 77, 364-374.
Schreier, H.J., Sonenshein, A.L. 1986. Altered regulation of the glnA gene in
glutamine-synthetase mutants of Bacillus subtilis. Journal of Bacteriology 167, 35-43.
Schulten, H.-R., Schnitzer, M. The chemistry of soil organic nitrogen: a review.
Biology and Fertility of Soils 26, 1-15.
Schulten, H.-R., Sorge-Lewin, C., Schnitzer, M. 1997. Structure of “unknown” soil
nitrogen investigated by analytical pyrolysis. Biology and Fertility of Soils 24, 249254.
Scott, N.A., Parfitt, R.L., Ross, D.J., Salt, G.J. 1998. Carbon and nitrogen
transformations I New Zealand plantation forest soils from sites with different N
status. Canadian Journal of Forest Research 28, 967-976.
Servín-Gonzáles, L., Bastarrachea, F. 1984. Nitrogen regulation of synthesis of the
high affinity methylammonium transport system of Escherichia coli. Journal of
General Microbiology 130, 3071-3077.
Siewe, R.M., Weil, B., Burkovski, A., Eikmanns, B.J., Eikmanns, M., Kramer, R.
1996. Functional and genetic characterization of the (Methyl)ammonium uptake
carrier of Corynebacterium glutamicum. Journal of Biological Chemistry
271, 5398-5403.
Soupene, E., He, L.H., Yan, D.L., Kustu, S. 1998. Ammonia acquisition in enteric
bacteria: Physiological role of the ammonium/methylammonium transport B (AmtB)
protein. Proceedings of the National Academy of Sciences of the United States of
America 95, 7030-7034.
Stevenson, F.J. 1982. Humus Chemistry. John Wiley, New York.
Strobel, B. 2001. Influence of vegetation on low-molecular-weight carboxylic acids in
soil solution – a review. Geoderma 99, 169-198.
Tate, R.L. III. 1995. Soil microbiology. John Wiley and Sons, New York.
Therkildsen, M.S., Isaksen, M.F., Lomstein, B.Aa. Urea production in the marine
bacteria Delaya venusta and Pseudomonas stutzeri grown in a minimal medium.
Aquatic Microbial Ecology 13, 213-217.
29
Thorn, K.A., Mikita, M.A. 1992. Ammonia fixation by humic substances: a nitrogen15 and carbon-13 NMR study. Science of the Total Environment 113, 67-87.
Tietema, A., van Dam, D. 1996. Calculating microbial carbon and nitrogen
transformations in acid forest litter with 15N enrichment and dynamic simulation
modelling. Soil Biology and Biochemistry 28, 953-965.
Tietema, A., 1998. Microbial carbon and nitrogen dynamics in coniferous forest floor
material collected along a European nitrogen deposition gradient. Forest Ecology and
Management 101, 29-36.
Tietema, A., Wessel, W.W. 1992. Gross nitrogen transformations in the organic layer
of acid forest ecosystems subjected to increased atmospheric nitrogen input. Soil
Biology and Biochemistry 24, 943-950.
Trehan, S.P. 1996. Immobilization of 15NH4+ in three soils by chemical and biological
processes. Soil Biology and Biochemistry 28, 1021-1027.
Uren, N.C. 2001. Types, amounts, and possible functions of compounds released into
the rhizosphere by soil-grown plants. In: Pinton, R., Varanini, Z., Nannipieri, P.
(Editors). The Rhizosphere – Biochemistry and Organic Substances at the Soil-Plant
Interphase, Marcel Dekker, Ink., New York.
Vance, E.D., Chapin, F.S. III. 2001. Substrate limitations to microbial activity in taiga
forest floors. Soil Biology and Biochemistry 33, 173-188.
Vanoni, M.A., Curti, B. 1999. Glutamate synthase: a complex iron-sulfur
flavoprotein. Cellular and Molecular Life Sciences 55, 617-638.
van de Werf, H., Verstraete, W. 1987. Estimation of active soil microbial biomass by
mathematical analysis of respiration curves: Calibration of the test procedure. Soil
Biology and Biochemistry 19, 261-265.
van Gestel, M., Ladd, J.N., Amato, M. 1991. Carbon and nitrogen mineralization from
two soils of contrasting texture and microaggregate stability: influence of sequential
fumigation, drying and storage. Soil Biology and Biochemistry 23, 313-322.
van Heeswijk, W.C., Hoving, S., Molenaar, D., Stegeman, B., Kahn, D., Westerhoff,
H.V. 1996. An alternative P-II protein in the regulation of glutamine synthetase in
Escherichia coli. Molecular Microbiology 21, 133-146.
van Heeswijk, W.C., Wen, D.Y., Clancy, P., Jaggi. R., Ollis, D.L., Westerhoff, H.V.,
Vasudevan, S.G. 2000. The Escherichia coli signal transducers PII (GlnB) and GlnK
form heterotrimers in vivo: Fine tuning the nitrogen signal cascade. Proceedings of the
National Academy of Sciences of the United States of America 97, 3942-3947.
van Oene, H, Berendse, F., Persson, T., Harrison, A.F., Schultze, E.-D., Andersen,
B.R., Bauer, G.A., Dambrine, E., Högberg, P., Matteucci, G., Paces, T. 2000. Model
analysis of carbon and nitrogen cycling in Picea and Fagus forests. In: Schultze, E.-
30
D. (Editor). Ecological Studies, Vol. 142. Carbon and nitrogen cycling in European
forest ecosystems. Springer-Verlag, Berlin, Heidelberg, pp. 419-467.
van Veen, J.A., Ladd, J.N., Frissel, M.J. 1984. Modelling C and N turnover through
the microbial biomass in soil. Plant and Soil 76, 257-274.
van Veen, J.A., Ladd, J.N., Amato, M. 1985. Turnover of carbon and nitrogen trough
the microbial biomass in a sandy loam and a clay soil incubated with [14C(U)] glucose
and [15N](NH4)2SO4 under different moisture regimes. Soil Biology and Biochemistry
17, 747-756.
Verchot, L.V., Holmes, Z., Mulon, L., Groffman, P.M., Lovett, G.M. 2001. Gross vs
net rates of N mineralization and nitrification as indicators of functional differences
between forest types. Soil Biology and Biochemistry 33, 1889-1901.
Vitousek, P.M. 1994. Beyond global warming: Ecology and global change. Ecology
75, 1861-1876.
Wagener, S.M., Schimel, J.P. 1998. Stratification of soil ecological processes: a study
of the birch forest floor in the Alaskan taiga. Oikos 81, 63-74.
Wang, W.J., Chalk, P.M., Chen, D. Smith, C.J. 2001. Nitrogen mineralization,
immobilization and loss, and their role in determining differences in net nitrogen
production during waterlogged and aerobic incubation of soils. Soil Biology and
Biochemistry 33, 1305-1315.
Watkins, N., Barraclough, D. 1996. Gross rates of N mineralization associated with
the decomposition of plant residues. Soil Biology and Biochemistry 28, 169-175.
Wessel, W.W., Tietema, A. 1992. Calculating gross N transformation rates of 15N
pool dilution experiments with acid forest litter: analytical and numerical approaches.
Soil Biology and Biochemistry 24, 931-941.
Witter, E., Mårtensson, A.M., Garcia, F.V. 1993. Size of the soil microbial biomass in
a long-term field experiment as affected by different N-fertilizers and organic
manures. Soil Biology and Biochemistry 25, 659-669.
Wood, J.M. 1999. Osmosensing by bacteria: Signals and membrane-based sensors.
Microbiology and Molecular Biology Reviews 63, 230-262.
Wright, F.R., Rasmussen, L. 1998. Introduction to the NITREX and EXMAN
projects. Forest Ecology and Management 101:1-7.
Wright, R.F., van Breemen, N. 1995. The NITREX project: an introduction. Forest
Ecology and Management 71:1-5.
Wood, J.M. 1999. Osmosensing by bacteria: Signals and membrane-based sensors.
Microbiology and Molecular Biology Reviews 63, 230-262.
31
Zaman, M., Di, H.J., Cameron, K.C., Frampton, C.M., 1999. Gross nitrogen
mineralization and nitrification rates and their relationships to enzyme activities and
the soil microbial biomass in soils treated with dairy shed effluent and ammonium
fertilizer at different water potentials. Biology and Fertility of Soils 29, 178-186.
Zak, D.R., Groffman, P.M., Pregitzer, K.S., Christensen, S., Tiedje, J.M. 1990. The
vernal dam: Plant-microbe competition for nitrogen in northern hardwood forests.
Ecology 71, 651-656.
Zhang, B.-G., Li, G.-T., Shen, T.-S., Wang, J.-K., Sun, Z. 2000. Changes in microbial
biomass C, N, P and enzyme activities in soil incubated with the earthworms
Metaphire guillelmi or Eisenia fetida. Soil Biology and Biochemistry 32, 2055-2062.
Zhang, H., Zhang, G.L. 2003. Microbial biomass carbon and total organic carbon of
soils as affected by rubber cultivation. Pedosphere 13, 353-357.
32
Tack till familj, vänner, handledare och kollegor!
I
Soil Biology & Biochemistry 35 (2003) 143–154
www.elsevier.com/locate/soilbio
Gross nitrogen mineralization-, immobilization-, and nitrification
rates as a function of soil C/N ratio and microbial activity
G. Bengtsson*, P. Bengtson, Katarina F. Månsson
Department of Ecology, Lund University, Sölegatan 37, Lund SE-223 63, Sweden
Received 18 March 2002; received in revised form 6 September 2002; accepted 18 September 2002
Abstract
A laboratory experiment was designed to challenge the idea that the C/N ratio of forest soils may control gross N immobilization,
mineralization, and nitrification rates. Soils were collected from three deciduous forests sites varying in C/N ratio between 15 and 27. They
were air-dried and rewetted to induce a burst of microbial activity. The N transformation rates were calculated from an isotope dilution and
enrichment procedure, in which 15NH4Cl or Na15NO3 was repeatedly added to the soils during 7 days of incubation. The experiments
suggested that differences in gross nitrogen immobilization and mineralization rates between the soils were more related to the respiration
rate and ATP content than to the C/N ratio. Peaks of respiration and ATP content were followed by high rates of mineralization and
immobilization, with 1 – 2 days of delay. The gross immobilization of NHþ
4 was dependent on the gross mineralization and one to two orders
of magnitude larger than the gross NO2
3 immobilization. The gross nitrification rates were negatively related to the ATP content and the C/N
ratio and greatly exceeding the net nitrification rates. Taken together, the observations suggest that leaching of nitrate from forest soils may
be largely dependent on the density and activity of the microbial community.
q 2003 Elsevier Science Ltd. All rights reserved.
Keywords: Nitrogen; Ammonium; Nitrate; ATP; Isotope pool dilution; Forest; Rewetting
1. Introduction
The low concentration of NO2
3 found in forest soils has
often been attributed to low rates of nitrification (Vitousek
et al., 1982; Gosz and White, 1986). This interpretation is
supported by observations of low net nitrification rates
during incubation assays of soil samples. However,
occasional measurements of gross nitrification suggest a
rapid turnover of a small NO2
3 pool in forest soils, and that
þ
the dominant fate of NO2
3 , as well as NH4 , is immobilization in the soil organic matter pool (Davidson et al., 1992;
Groffman et al., 1993; Stark and Hart, 1997). In this way,
immobilization may prevent nitrogen leakage to ground and
surface waters.
The extent of immobilization varies between 35 and 95%
from one soil to another (Mead and Pritchett, 1975; Heilman
et al., 1982; Melin et al., 1983; Schimel and Firestone, 1989;
Hart and Firestone, 1991; Hart et al., 1993). Nitrogen is
immobilized either by abiotic or biotic processes. Fixation
* Corresponding author. Tel.: þ 46-46-2223777; fax: þ 46-46-2223790.
E-mail address: [email protected] (G. Bengtsson).
of NHþ
4 to clay minerals is fast (, 30 min) but usually less
than 10% of the added NHþ
4 is fixed (Drury and Beauchamp,
1991; Trehan, 1996). Most of the immobilization to the soil
organic matter is biotic. Early work estimated soil
microorganisms to be responsible for 10 – 50% of NHþ
4
immobilization (Brookes et al., 1985) and more recent
15
15
þ
2
analyses by NMR of the organic matter of NH4 or NO3
treated soils found the organic 15N in microbially derived
peptides and proteins (. 80%), nucleic acids, and aliphatic
amine groups after several months of in situ incubation
(Clinton et al., 1995). The average turnover time of N in
microbial biomass is estimated to 1 –2 months (Davidson
et al., 1992), but some of the immobilized N becomes very
stable (Kelley and Stevenson, 1985), as a result of fixation
by 2:1 clay minerals (van Veen et al., 1985; Breitenbeck and
Paramasivam, 1995), repeated cycles of immobilization and
mobilization (Jansson and Persson, 1982; He et al., 1988),
and accumulation of persistent residues of microbial cells
(Paul and Juma, 1981).
Despite the obvious importance of soil microorganisms
in the turnover and retention of nitrogen in forest soils, the
causes of variation of immobilization of N from one soil to
0038-0717/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 3 8 - 0 7 1 7 ( 0 2 ) 0 0 2 4 8 - 1
144
G. Bengtsson et al. / Soil Biology & Biochemistry 35 (2003) 143–154
another and between abiotic and biotic processes are not
well known. Some observations suggest that the immobilization is controlled by the concentration of available C
(Woodmansee and Duncan, 1980), and others that it is
inversely dependent on the amount of available inorganic N
(Priha and Smolander, 1995; Bengtsson and Bergwall,
2000).
Mineralization of soil N depends on a wide range of
factors, including the C/N-ratio (Frankenberger and Abdelmagid, 1985), the N content (Iritani and Arnold, 1960),
lignin content (Frankenberger and Abdelmagid, 1985; De
Neve et al., 1994), water soluble N and cellulose content
(Iritani and Arnold, 1960; Bending et al., 1998) of the litter,
the light fraction organic matter of the soil (Janzen et al.,
1992; Sierra 1996), microbial respiration and ATP content
(Alef et al., 1988), microbial biomass (Dalal and Meyer,
1987), and microbial N content (Fisk and Schmidt, 1995).
The diversity of factors correlated with N mineralization
reflects the variation in substrates and microbial communities being used and temporal changes in substrate quality.
Models predicting nitrogen turnover and retention in
soils often include the C/N ratio as an important factor in
determining the rate of mineralization, immobilization and
nitrification (van Veen et al., 1984; Aber, 1992; Bradbury
et al., 1993; Janssen, 1996). The idea to use the soil C/N
ratio to predict variations in N mineralization and
immobilization rates among soils is based on the fact that
heterotrophic soil bacteria usually have a lower C/N ratio
than the soil they inhabit. If it is assumed that the cells have
a C/N ratio of 10 and respire about 50% of their C uptake,
they may be N limited above a soil C/N ratio of 20 and C
limited below (Tate, 1995). Soils with a high C/N ratio may
then be characterized by rapid immobilization of N and soils
with a low C/N ratio by slower N immobilization and a
surplus of available NHþ
4 , derived from deamination of
organic carbon sources.
The evidence for a correlation between the soil C/N ratio
and nitrogen immobilization and mineralization is equivocal, at least partly because early work was confined to
measurements of net rates of N transformation. Within-site
variations of the C/N ratios of different pools of organic
matter and between-site variations of the C and N
assimilation efficiency of the microbial biomass may also
confound the usefulness of the soil C/N ratio to predict gross
nitrogen transformation rates. Therefore, we designed a 15N
pool dilution experiment to simultaneously measure gross
immobilization, mineralization, and nitrification, in an effort
to challenge the hypotheses of a positive relationship
between the C/N ratio and N immobilization and a negative
relationship between the C/N ratio and N mineralization and
nitrification in three forest soils. Even if the soil C/N ratio
may prove useful to predict site-to-site variations of the N
transformations, temporal variation of soil temperature and
moisture will trigger dynamics of microbial biomass and
activity that influence N transformations more than the
spatial variation of the C/N ratio. Therefore, we developed
the sub-hypothesis that the variation in N transformation
rates is controlled by the variations in microbial biomass
and activity—a high biomass and activity would correspond
(1) to high rates of N immobilization and mineralization,
and, as nitrifiers tend to be competitively inferior to
heterotrophs (Verhagen and Laanbroek, 1991; Verhagen
et al., 1995), (2) to low nitrification rates.
It is well established from laboratory experiments that
soil drying and rewetting may bring about a flush of C and N
mineralization (Birch, 1958; Agarwal et al., 1971; van
Gestel et al., 1991) and an increase in microbial numbers
and activity (West et al., 1986). The origin of the
mineralized N may be both biomass lyzed by the drying,
and non-biomass organic matter that becomes more
accessible to microbial attack after drying and rewetting
(Jager and Bruins, 1975; Marumoto et al., 1982). We used
this evidence to test the sub-hypothesis and induced a
microbial growth phase by drying and rewetting the three
soils, and then evaluated the subsequent gross N transformations rates. Although short-time laboratory 15N pool
dilution experiments do not necessarily reflect N transformation rates in the field, they facilitate experimental
manipulations and help explaining mechanisms causing
variation in N transformation rates from one soil to another.
2. Material and Methods
2.1. Soil sampling
Soil material was collected from deciduous forests in
southwestern Sweden in July 1998. Soil I was found in the
vicinity of Maglö (56 830 N, 13 8120 E), and Soil II and III
were found in Torup (55 8330 N, 13 8370 E). After removal of
the litter, five samples of the top 5 cm of the soil were
collected randomly within a 20 by 20 m square. The
samples were sieved through a 4 mm mesh and then pooled.
Remaining roots and leaf pieces were removed by hand and
the soils were left to air-dry at room temperature for 1 week.
All three soils can be described as Dystric Cambisols (FAO
system). Properties for the three different soil materials used
in the experiment are given in Table 1.
2.2. Experimental design and
15
N additions
A total of 90 acid washed (10% HCl for 1 day) and burnt
(400 8C for 12 h) 100 ml serum bottles per soil received 10 g
of air-dried soil each. The soils were then rewetted to 60%
of the water holding capacity by adding autoclaved,
deionized water with an automatic pipette, and then gently
mixing the soil. The bottles were left over night on an orbital
shaker at a temperature of 15 8C, and the following day the
serum bottles were divided into two groups per soil, one
receiving 1.0 ml of 15NH4Cl (98% 15N, Cambridge Isotopic
Laboratories) at concentrations of 510, 86, and 17 mM þ
NaNO3 at concentrations of 110, 170, and 60 mM for Soil I,
G. Bengtsson et al. / Soil Biology & Biochemistry 35 (2003) 143–154
145
Table 1
Properties for the three different soil materials used in the experiment. All values are on dry weight basis
Soil I
Soil II
Soil III
21
NO2
3 -N (mg kg )
21
NHþ
4 -N (mg kg )
Kjeldahl N (g kg21)
Total C (g kg21)
C/N-ratio
pH
WHC (kg H2O kg21)
3.10
4.93
0.04
7.18
4.87
0.50
5.24
2.44
4.32
86.56
37.37
116.05
Low (17)
Low (15)
High (27)
4.1
3.7
3.1
1.29
0.64
1.54
II, and III, respectively, and the other 1.0 ml of Na15NO3
(99% 15N, Cambridge Isotopic Laboratories) þ NH4Cl at
the same concentrations as above. The two groups were
hence exposed to the same treatment, but labeled either on
2
NHþ
4 or on NO3 (‘inorganic paired’ treatment; Mary et al.,
1998). They were further divided into three subgroups of 15
bottles each, which received the labels immediately, or after
incubation for 3 or 6 days at 15 8C on the orbital shaker. The
resulting 15N enrichment varied between 6 and 25% for
2
NHþ
4 , and between 3 and 16% for NO3 . The bottles were
sampled destructively, in replicates of five, immediately,
and 1 and 2 days after the addition of the label. This made it
possible to estimate the N transformation rates in different
phases following the induced microbial growth. The soil
was used for ATP measurements, Kjeldahl digestion, and
2
extraction of NHþ
4 and NO3 . Microbial respiration was
measured 1 or 2 days after the addition of the label.
2.3. Kjeldahl digestion
Portions of 300 mg of soil were transferred to acid
washed 100 ml test tubes. About 1 g of K2SO4 and CuSO4
(10:1 by weight) and 3.0 ml of concentrated H2SO4 were
added to the test tubes. The tubes were placed on a heating
block at 400 8C for 3 h or until the content had turned bright
green. After digestion, the samples were diluted with
deionized water to 100 ml, and 1 ml of this dilution was
then subjected to NHþ
4 derivatization as described below.
2
2.4. Extraction and derivatization of NHþ
4 and NO3
Portions of 4.0 g of soil were transferred to acid washed
and burnt teflon lined 50 ml test tubes. Exchangeable soil
2
NHþ
4 and NO3 was extracted with 20 ml 0.5 M K2SO4 on a
home made rotary shaker (, 27 rpm) at 15 8C for 2 h.
Following extraction, the test tubes were centrifuged for
30 min at 700 £ g on a Hermle Z510. Aliquots of 5.0 ml of
the supernatant were transferred to test tubes and concentrated in a Savant Speed Vac vacuum centrifuge (Savant
Instruments, Inc.). The residue was re-dissolved in 1.0 ml of
Milli-Q water and then subjected to derivatization. Samples
from the Kjeldahl digestion were not vacuum centrifuged
prior to derivatization.
NO2
3 was derivatized according to Bengtsson and
Bergwall (2000) using 2-sec-butylphenol and NO-Bis(trimethylsilyl)acetamide (Sigma-Aldrich) to yield the trimetylsilyl ester of 4-NO2-2-sec-butylphenol. NHþ
4 was
derivatized by using pentafluorobenzoylchloride (SigmaAldrich) to yield pentafluorobenzamide (Bengtsson and
Bergwall, 2000). The samples were analyzed by a HewlettPackard 5889B MS-Engine mass spectrometer coupled to a
Hewlett-Packard 5890 Series II gas chromatograph as in
Bengtsson and Bergwall (2000).
2.5. Microbial respiration and ATP measurements
Microbial respiration was measured as accumulation of
CO2 in the headspace of the serum bottles over time. The
flasks were sampled 1 or 2 days after the addition of the
label, and the amount of CO2 measured as in Bengtsson and
Bergwall (1995) by injecting 1.0 ml of the headspace gas
into a Varian model 3700 gas chromatograph equipped with
a thermal conductivity detector. Helium was used as a
carrier gas at 20 ml min21 and the inert gases were
separated on a Porapak R column (2 m, O.D. 30 mm). The
injector temperature was 170 8C and the filament temperature 300 8C. Peak areas for carbon dioxide were standardized against known amounts of NaHCO3.
The amount of ATP was measured immediately, and 1 or
2 days after the addition of the label. ATP was extracted
from the soils using the method described by Eiland (1983)
but with some modifications. Soil (0.5 g wet wt) was
extracted with 20 ml of ice cold 500 mM H2SO4 and
250 mM Na2HPO4 for 30 min on the rotary shaker.
Following extraction, 25 ml of the soil solution were
transferred to 3 ml of 250 mM Tris with 4 mM EDTA and
a pH of 7.5. Of this buffered soil solution, 50 ml were
transferred to a scintillation vial and 50 ml of luciferase–
luciferin enzyme (Adenosine 50 -triphosphate (ATP) assay
mix, Sigma-Aldrich) added. The light output was immediately measured for 15 s in a Beckman LS6500 scintillator
equipped with a single photon meter. The amount of ATP in
the samples was calculated from internal standards prepared
by adding 19.5 ml of ice-cold extractant and 0.5 ml of ATP
standard (5.0 mg per 0.5 ml, disodium salt, Sigma-Aldrich)
to each sample prior to extraction.
2.6. Calculation of nitrogen transformation rates
Nitrogen transformation rates were calculated with the
FLUAZ-model (Mary et al., 1998). The calculations in this
model are based on the isotopic dilution and isotopic
enrichment principles (Monaghan and Barraclough, 1995).
It uses a numerical fourth order Runge– Kutta algorithm
146
G. Bengtsson et al. / Soil Biology & Biochemistry 35 (2003) 143–154
with a variable time-step to solve the differential system,
and a non-linear fitting program (based on Marquardt’s
algorithm) to calculate the unknown N transformation rates
2
between the NHþ
4 , NO3 , organic N and biomass N pools.
2
The gross mineralization, nitrification, and NHþ
4 and NO3
immobilization rates were calculated from the measurements of the amounts and 15N isotopic excesses of NH4-N,
NO3-N and organic N.
The FLUAZ model minimizes the quadratic weighted
error rather than the sum of squares and therefore requires
the average value plus the coefficient of variation for the
measured variables as input. This accounts for the variance
of the measurements, so that the variables with the highest
variability have the lowest weight, and yields the nitrogen
transformation rates for each treatment rather than for each
replicate. Prior to the modeling, outliers of the 95%
confidence intervals for the averages were removed. The
input data in the model are total organic 15N measurements,
but the model simulates the evolution of 14N and 15N in the
zymogenous (newly formed) microbial biomass. Biomass N
and 15N measurements remain difficult to make and
represent the total biomass rather than the zymogenous
biomass, which is incorporating the labeled
N. Measurements of total organic 15N are also considered
to be much more reliable in preparing 15N balances (Mary
et al., 1998).
2.7. Statistics
Stepwise multiple regression analysis (forward procedure, F to enter, 4.000, F to remove, 3.996) was used to
test the dependence of the gross mineralization-, immobilization-, and nitrification rates on different measured
factors. The gross mineralization rate and the ratio between
gross total immobilization and gross mineralization were
tested against the respiration rate, the net ATP production
rate, and the average of the amount of ATP present during 1
day (i.e. the average between the amount measured day 1
and 2, day 3 and 4, and so on). The soil C/N ratio (high or
low) was included in the analysis as a dummy variable.
The gross NHþ
4 immobilization rates were tested against
the same factors, plus the gross mineralization rate, while
þ
the ratio between gross NO2
3 immobilization and gross NH4
immobilization was tested against the gross nitrification rate
as well. Gross nitrification and NO2
3 immobilization rates
were tested against the respiration rate, the net ATP
production rate, the average of the amount of ATP present
during 1 day, the soil C/N ratio, the gross NH þ
4
immobilization rate, and the gross mineralization- or
nitrification rates. The respiration rate and the gross total
immobilization rate were log10 transformed, while the gross
þ
NO2
3 and NH4 immobilization rates, the gross mineralization rate, and the gross nitrification rates were log10(X þ 1)
transformed to meet the assumption of equality of variances.
All three soils were analyzed by multiple regression to test
whether the soil C/N ratio was responsible for differences in
N transformation rates between the three soils that were not
explained by the microbial biomass and activity. The soils
were then analyzed separately for the same factors except
the soil C/N ratio.
Differences between the three soils in respiration rate and
the average of the amount of ATP present at the beginning
and end of each experimental day were tested with repeated
measurement ANOVA followed by a Scheffé F-test, with
day as the within factor. Within each soil, the differences
between different days were tested with a one-way
ANOVA. All analyses were performed with the StatView
4.53 software (Abacus Concepts, Inc.).
3. Results
3.1. Gross nitrogen transformation rates
The high rates of mineralization and immobilization
during day 1– 2 and 3 – 4 were preceded by a burst of
microbial activity and then followed by lower rates as the
incubation of the soils proceeded (Figs. 1 and 2, Table 2).
This pattern coincided in general with the one hypothesized
when the experiment was designed, but no significant
relationships were found between the gross mineralization
and immobilization rates and the microbial biomass/activity
in the soils (Tables 3 and 4) due to 1 –2 days displacement of
peaks for microbial activity and N transformations. The
immobilization of NHþ
4 was dependent only on the gross
mineralization rate in the multiple regression analysis
(Table 3). When the soils were analyzed separately, the
amount of ATP could account for some of the variation of
the NHþ
4 immobilization in Soil II (Table 4). Since rapid
immobilization of nitrogen in Soil I occurred before the
mineralization rate reached its maximum, the dependence of
immobilization on mineralization was broken up in this soil.
In Soil III, a time lag of 1 or 2 days was observed between
the peaks of the mineralization and immobilization rates and
the ATP content (Fig. 3, Table 2).
The cumulative mineralization and NHþ
4 immobilization
were of the same magnitude in Soil I and III, but
considerably lower in Soil II (Fig. 1). The same pattern
was found for the ATP content and the respiration rate (Figs.
2 and 3), suggesting that differences in the mineralization
and immobilization potential between the soils were more
reflected by those variables than by the soil C/N ratio. The
gross NHþ
4 immobilization rate was lower than the gross
mineralization rate in Soil I and III, but the resulting net
mineralization was small (Fig. 1, Table 2). None of the
tested factors could explain the variation in the immobilization/mineralization ratio over time, neither when the soils
were analyzed together or separately (Tables 3 and 4).
Even though the immobilization of NO2
3 was in the order
2
of 1/10 – 1/100 of that of NHþ
4 , more NO3 was consumed
than produced in Soil I and III (Fig. 1, Table 2). When all
soils were analyzed together, the immobilization of NO2
3
G. Bengtsson et al. / Soil Biology & Biochemistry 35 (2003) 143–154
147
2
Fig. 1. The cumulative mineralization, NHþ
4 immobilization, nitrification, NO3 immobilization, the ratios between gross cumulative mineralization and gross
N immobilization, and gross cumulative nitrification and gross NO2
3 immobilization in the three soils during the experiment. All values are on dry weight basis.
was independent of all the tested factors (Table 3), but when
the soils were analyzed separately, the immobilization of
NO2
3 in Soil III was highly dependent on the gross
nitrification rate, the net ATP production, and the respiration
rate (Table 4). Two cases with gross NO2
3 immobilization
exceeding the gross NHþ
4 immobilization were also found in
þ
this soil (Table 2). The variation of the NO2
3 /NH4
immobilization ratio was explained by the variation in soil
C/N-ratio and ATP production (Table 3).
The gross nitrification rate was low in all three soils,
but especially in Soil III (Table 2). Nitrification took off
fastest in Soil I, but at the end of the experiment, it was
148
G. Bengtsson et al. / Soil Biology & Biochemistry 35 (2003) 143–154
3.2. Soil respiration rates and ATP content
The microbial respiration rates were significantly
different in the soils and during the course of the
experiment (Fig. 2). It was highest during the first day of
incubation, decreased during the second, and maintained
a low level during the rest of the incubation. Soil I had
the highest respiration rate and Soil II the lowest. Most
ATP was found in Soil III and least in Soil II (Fig. 3),
that is, the soil with the lowest biomass also had the
lowest respiration rate. Although there were irregular
variations from 1 day to another, the ATP content
seemed to remain constant throughout the experiment in
Soil I and decreased in Soil II and III.
4. Discussion
Fig. 2. The respiration rate (mg CO2 kg21 soil dry weight) in the three soils
during the experiment. The rate differed significantly among soils
( p , 0.001). Soil I had a higher rate than the two other soils at all times
during the experiment, and Soil II had the lowest rate (Scheffé F-test, p ,
0.001). There was also a significant difference in respiration rate among
days within each soil ( p , 0.001). Error bars represent the standard error of
the mean.
Rather than supporting the assumption that the soil C/N
ratio controls potential gross immobilization and mineralization and explains differences in their rates from one soil
to another, this study suggests that differences in respiration
rate and ATP content are more indicative of the magnitude
of the two processes. The soil with the high C/N ratio (Soil
III) complied with a number of criteria for high immobilization rates, such as low in situ concentrations of
þ
extractable NO2
3 and NH4 and essentially no net mineralization, but the latter criteria was true also for the soils with
the lower C/N ratio. Those soils clearly had similar C/N
ratios, although the total C and N concentrations were
different, and yet their ATP content varied by more than a
factor of three. Similar inconsistencies in the relationships
between the soil C/N ratio and immobilization were found
in the NITREX study, where the NHþ
4 immobilization was
highest at sites with a low C/N ratio (Tietema, 1998).
Gundersen et al. (1998a) suggested that the results were
influenced by fragmentation of mycorrhiza during sieving
of the soils. Mycorrhiza was assumed to be more abundant
in the high C/N ratio soils, so sieving would reduce the NHþ
4
immobilization rates more in those soils than in those with
low C/N ratios.
surpassed by nitrification in Soil II (Fig. 1). When
analyzed together, the nitrification rate in the three soils
was inversely dependent on the amount of ATP (Table
3). The same dependency of the nitrification rate on the
amount of ATP was also found when Soil I was
analyzed separately, and, in addition, it was dependent
on the gross mineralization rate and the gross NHþ
4
immobilization (Table 4). In Soil III, the ATP production, the mineralization rate, and the respiration rate
could explain a great deal of the variation of the
nitrification rate, whereas no significant relationships
between any of the tested variables and the nitrification
rate were found in Soil II (Table 4). Net nitrification, that
is the difference between gross nitrification and
immobilization, was negative in Soil I and III, and
about one third of gross nitrification in Soil II (Table 2).
Table 2
2
The gross mineralization, NHþ
4 immobilization, nitrification, and NO3 immobilization rate in the three soils during the experiment. Values within brackets
represent standard error of the mean. All rates are on dry weight basis
Day
0– 1
1– 2
3– 4
4– 5
6– 7
Gross mineralization
(mg N kg21 d21)
Gross NHþ
4 immobilization
(mg N kg21 d21)
Gross nitrification
(mg N kg21 d21)
Gross NO2
3 immobilization
(mg kg21 d21)
Soil I
Soil II
Soil III
Soil I
Soil II
Soil III
Soil I
Soil II
Soil III
Soil I
Soil II
Soil III
22.5 (14.3)
53.1 (9.8)
17.4 (5.9)
0.0 (13.9)
3.00 (39.5)
3.9 (1.3)
3.2 (2.0)
23.6 (3.5)
2.7 (1.0)
3.9 (1.0)
2.7 (17.6)
103 (27.6)
0.0 (41.2)
0.4 (6.8)
142 (13.0)
37.5 (15.9)
32.1 (10.1)
13.9 (9.6)
4.2 (15.1)
0.5 (51.0)
4.2 (1.3)
3.7 (2.0)
24.7 (4.1)
3.3 (1.4)
1.4 (1.1)
0.0 (17.6)
96.4 (27.4)
4.6 (42.8)
0.0 (21.0)
146 (12.0)
0.8
0.7
0.8
0.2
0.2
0.2
0.4
1.4
2.9
2.0
0.1
0.0
0.3
0.2
0.0
3.5 (1.0)
0.0 (0.7)
0.0 (1.6)
0.6 (3.0)
0.0 (19.8)
1.5 (0.5)
0.0 (1.2)
0.0 (3.4)
2.9 (1.6)
0.0 (0.5)
0.1 (0.0)
0.0 (0.1)
0.3 (0.3)
0.3 (0.5)
0.0 (0.0)
(0.0)
(0.0)
(0.7)
(1.4)
(1.2)
(0.1)
(0.1)
(2.6)
(0.9)
(0.8)
(0.0)
(0.0)
(0.4)
(0.5)
(0.0)
G. Bengtsson et al. / Soil Biology & Biochemistry 35 (2003) 143–154
149
Table 3
The dependency of the gross immobilization and nitrification rates, as well as the ratio between gross nitrate immobilization/gross ammonium immobilization,
on the different measured factors. The C/N ratio was included in the analyses as a dummy variable (Stepwise multiple regression analysis, forward procedure, F
to enter, 4.000, F to remove, 3.996, degrees of freedom ¼ 14). No factors were entered into the model when mineralization, NO2
3 -immobilization, or the ratio
between immobilization and mineralization were tested as dependent
Dependent
Factors not in model
Factors in model
NH4 immobilization
Respiration
ATP content
ATP production
C/N ratio
Respiration
ATP production
C/N ratio
Mineralization
NH4 immobilization
Respiration
ATP content
Mineralization
Nitrification
Mineralization
Nitrification
NO3 immobilization/ NH4
immobilization
ATP content
ATP production
C/N ratio
The ATP content, which reflects both biomass and
metabolic state of the microorganisms (Contin et al., 2000),
was responding to the tracer addition in a more systematic
and repeatable way than the nitrogen transformations,
generally with an increase 1 day after an addition (Fig. 3).
That pattern was superimposed upon the trend of a reduction
of the ATP content with the incubation time, but because of
the 1 –2 day displacement between peaks in ATP content
and mineralization and immobilization rates, the ATP
content could not explain the ceasing mineralization and
immobilization rates after 3 days of incubation. Tsai et al.
(1997) demonstrated that soils amended with glucose often
had the largest amounts of ATP 1– 2 days before the largest
amounts of biomass C. Similarly, 3– 5 days passed until the
peak in numbers of bacteria and the frequency of dividing
cells appeared (Bloem et al., 1992) after rewetting of an airdried soil, which may be analogous to glucose addition in
this respect (Diaz-Ravina et al., 1988). Hence, it is possible
that the delay of 1 – 2 days between the peaks of ATP
content and gross nitrogen mineralization and immobilization rates represents the time lag between maximum
energy production and the conversion of the organic carbon
substrate into a sufficiently large biomass for a maximum
impact on the nitrogen conversion.
The burst of respiration lasted for only 1 day (Fig. 2), but
the respiration pattern during the incubation overlapped
qualitatively with that of the cumulated mineralization and
immobilization, with 1 day of delay in Soil I and III.
Similarly, Hart et al. (1994) and Tietema (1998) found gross
mineralization and immobilization rates to be correlated
with respiration rates, implying that N and C cycling are
tightly coupled in forest soils, and Barrett and Burke (2000)
showed that soil carbon content alone can explain more than
60% of the variation in potential nitrogen immobilization at
regional scales. CO2 evolution might therefore be a valuable
indicator of the magnitude of gross N transformation rates.
R2
Significance
0.852
0.750
,0.001
20.052
0.362
,0.05
2.490
7.915
0.624
,0.01
Coefficient
The slow but more extended response of increased
mineralization and immobilization in Soil II reflected
other characteristics of the microbial community and the
environment, e.g. low concentrations of ATP and organic
carbon. Soil II also had several times lower NHþ
4 content
than Soil I and III throughout the experiment, but when
calculated on basis of soil water content, the concentrations
were in the same millimolar range in the three soils (Fig. 4).
Since all known NHþ
4 transporters isolated from bacteria
have a Km in the micromolar range (Siewe et al., 1996;
Montesinos et al., 1998; Soupene et al., 1998), the NHþ
4
concentration should be sufficiently high to allow NHþ
4
immobilization at a maximum rate in all three soils, even if
less than 5% of the extractable NHþ
4 is considered to be
present in solution.
Variations in gross mineralization/immobilization rates
could not be connected with corresponding variations in
microbial biomass in those few studies that addressed the
relationship by ‘snapshot’ observations during longer
experiments. The gross mineralization rate as well as
microbial biomass C and N were significantly affected by
the soil water potential in the 90 days study by Zaman et al.
(1999) but not correlated. Similarly, Puri and Ashman
(1998) found the gross mineralization rate to be dependent
on forest soil moisture and temperature, but not on soil
biomass N, which remained constant throughout a year. On
the other hand, mathematical simulation models have
predicted the size and activity of the microbial biomass to
have an impact on N mineralization and immobilization
rates (Hadas et al., 1992; Blagodatsky and Richter, 1998).
This discrepancy between theoretical model predictions and
results from incubations or long-term field experiments may
reflect methodological constraints in microbial biomass C
and N measurements. The commonly used chloroform
fumigation method does not produce reliable biomass
comparisons in situations of low spatial and temporal
Mineralization
Respiration
ATP content
ATP
production
ATP content
ATP
production
Nitrification
Respiration
ATP
production
Respiration
Factors
entered
in model
Soil I
Factors not
entered
in model
0.895
0.658
Sig.
Soil II
–
–
Factors
entered
in model
Respiration
ATP content
Respiration
Mineralization
Nitrification
1.903
0.664
ATP content
ATP
production
ATP content
ATP
production
0.559
0.592
20.651
0.242
0.428
0.598
1.153
Sig.
Soil III
Factors not
entered in
model
0.934 NS
0.936 NS
0.944 NS
Mineralization
Factors
entered
in model
0.906
Sig.
NS
0.999 ,0.05
0.859
1.000 ,0.05
0.999 ,0.05
0.804 ,0.05
Coefficient R 2
Respiration
0.031
ATP
0.002
production
0.990
Nitrification
ATP content
Respiration
0.001
NH4
ATP
0.064
immobilization production
0.046
Mineralization
ATP content
Respiration
1.203
ATP
production
ATP content
Respiration
217.1
ATP
Mineralization
5.044
production
Nitrification
0.991 ,0.01 Respiration
ATP content
ATP
production
0.629 NS
ATP
content
Coefficient R 2
Respiration
ATP
production
ATP
production
Respiration
ATP content
ATP production Mineralization
Factors not
entered
in model
0.999 ,0.05 ATP content
Mineralization
NH4
immobilization
0.966 NS
Respiration
0.666 NS
0.595 NS
Coefficient R 2
ATP content
0.081
Mineralization
0.020
NH4
0.166
immobilization
Immobilization/ ATP content Respiration
1.525
Mineralization
ATP
20.392
production
NO3 /NH4
Respiration None
–
immobilization ATP content
ATP
production
Mineralizaiton
Nitrification
Nitrification
NO3
immobilization
NH4
immobilization
Dependent
Table 4
The dependency of the gross immobilization and nitrification rates, and the ratios between gross total immobilization/gross mineralization and gross nitrate immobilization/gross ammonium immobilization, on
the different measured factors in each of the three soils (Stepwise multiple regression analysis, forward procedure, F to enter,4.000, F to remove, 3.996, degrees of freedom ¼ 4). No factors were entered into the
model when mineralization was tested as dependent
150
G. Bengtsson et al. / Soil Biology & Biochemistry 35 (2003) 143–154
G. Bengtsson et al. / Soil Biology & Biochemistry 35 (2003) 143–154
Fig. 3. The average content of ATP (mg kg21 soil dry weight) present
during a day in the three soils. The content differed significantly among
soils ( p , 0.001) and among days within each soil ( p , 0.001). Error bars
represent the standard error of the mean.
21
Fig. 4. The average concentration of NHþ
soil dry
4 – N in (top) mg kg
weight, and (bottom) mM in the three soils during the different days of the
experiment. Values are averages of the ‘inorganic paired’ treatments.
151
variation (Wardle and Ghani, 1995) and extracts the total
microbial biomass C rather than the active part of the soil
microbial community (van de Werf and Verstraete, 1987;
Ross, 1991). For instance, the chloroform fumigation
method would fail to discriminate between two bacterial
populations with a similar biomass but varying 10 fold in
growth rate (Bloem et al., 1989), and yet, the population
with a higher proportion of actively growing bacteria could
be expected to have a higher turnover rate of carbon and
nitrogen. Microbial biomass C and N measured by the
fumigation method may therefore be poorly related to
nitrogen mineralization rates.
The gross N immobilization rate was found to be highly
dependent on the mineralization rate in all three soils
(Tables 3 and 4). The high immobilization rates during the
first few days of incubation (Table 2) agree with those found
in other drying-rewetting experiments. For example, Pulleman and Tietema (1999) found immobilization rates of 18–
119 mg N kg soil21 d21 in dried and rewetted F horizon
material at respiration rates similar to those in our
experiment, and air dried and water logged mineral soils
immobilized between 10 and 120 mg N kg soil21 d21
(Wang et al., 2001). The rates are much lower at field
moisture conditions and similar to those we found in the late
part of the incubation. As a comparison, Puri and Ashman
(1999) found immobilization rates to vary between 0.5 and
1.7 mg N kg21 d21 in a sandy loam from a mature oak
woodland, and 7– 14 mg N kg21 d21 were immobilized in
the organic horizon of six northern hardwood forests (Fisk
and Fahey, 2001). The podzols in the NITREX study, which
are located in a nitrogen deposition gradient in northwestern
and central Europe, have immobilization rates of between 9
and 18 mg N kg soil21 d21 (Tietema, 1998). Immobilization rates as high as 70 – 500 mg N kg21 d21 at NHþ
4
concentrations of between 0.05 and 5.0 mM have been
found in soil slurries of organic horizon material from a
coniferous forest (Schimel and Firestone, 1989).
The pattern of the variations of the gross nitrification
rates was consistent with the hypotheses on negative
relationships between the rates and the C/N ratio (Fig. 1)
and the ATP content (Table 3). It is also in agreement
with earlier findings of negative relationships between
the net nitrification rates and the C/N ratio of the forest
floor (Gundersen et al., 1998b). This information taken
together with the observation that the amount of NHþ
4
immobilized was far exceeding the amount that was
nitrified suggests that high heterotrophic activity in a
forest soil coincides with low nitrification rates, possibly
because nitrifiers are poor competitors for NHþ
4 (Verhagen and Laanbroek, 1991; Verhagen et al., 1995).
However, while the immobilization rate ceased after a
few days of incubation, at least in Soil I and III, the
nitrification rate remained constant and even increased
(Table 2). This supports the idea that heterotrophs
assimilate most of the available NHþ
4 when heterotrophic
demand for N are high due to rapid growth (Hart et al.,
152
G. Bengtsson et al. / Soil Biology & Biochemistry 35 (2003) 143–154
1994), but it may also reflect a slow growth of nitrifiers
compared to heterotrophs, and a possibility that nitrification and NO2
3 immobilization may be underestimated in
short-term laboratory incubations. The nitrification may
also be underestimated because the FLUAZ model only
considers autotrophic nitrification, which declines markedly below pH 6 and becomes negligible below pH 4.5
(Paul and Clark, 1996). NO2
3 production in acid forest
soils has been attributed to heterotrophic nitrification
(Schimel et al., 1984; Duggin et al., 1991; Pedersen et al.,
1999), but the evidence is equivocal. Acidophilic
autotrophic nitrifiers may contribute with more than
90% of the NO2
3 production in acid forest soils (De Boer
et al., 1992; Barraclough and Puri, 1995; Stark and Hart,
1997), and Killham (1986) suggested that the character
and availability of substrates are more important than soil
pH to control the extent of heterotrophic nitrification.
The comparison of gross and net nitrification rates
adds another piece of evidence to the concern about net
nitrification measurements giving poor predictions of
gross nitrification rates in soils (Stark and Hart, 1997).
þ
Very low in situ concentrations of NO2
3 and NH4 in
combination with a high ATP content, such as in Soil III,
are a good point of departure for a tight coupling of
gross nitrification and immobilization, resulting in a
negligible or negative net nitrification. Soils with a low
C/N ratio, low ATP content, and high in situ concenþ
trations of NO2
3 and NH4 , such as Soil II, are good
candidates for a positive net nitrification and subsequent
leaching of excess NO2
3.
The correlation between the ATP content and net
nitrification rate is not the only environmental implication
arising from the attention given to the soil heterotrophic
microbial activity in this work. Generally, it demonstrates a
tight connection between nitrogen turnover and retention
and the microbial activity and a co-variation from one forest
soil to another between mineralization, immobilization,
ATP content, and respiration. The property of a forest soil as
a source of nitrogen for plants and of excess nitrogen ending
up in ground water and surface water may thus be more
variable with and dependent on its microbial biomass and
activity than on its C/N ratio.
References
Aber, J.D., 1992. Nitrogen cycling and nitrogen saturation in temperate
forest ecosystems. Trends in Ecology and Evolution 7, 220–223.
Agarwal, A.S., Singh, B.R., Kanehiro, Y., 1971. Soil nitrogen and carbon
mineralization as affected by drying-rewetting cycles. Soil Science
Society of America Proceedings 35, 96–100.
Alef, K., Beck, T.H., Zelles, L., Kleiner, D., 1988. A comparison of methods
to estimate microbial biomass and N-mineralization in agricultural and
grassland soils. Soil Biology & Biochemistry 20, 561 –565.
Barraclough, D., Puri, G., 1995. The use of 15N pool dilution and
enrichment to separate the heterotrophic and autotrophic pathways of
nitrification. Soil Biology & Biochemistry 27, 17–22.
Barrett, J.E., Burke, I.C., 2000. Potential nitrogen immobilization in
grassland soils across a soil organic matter gradient. Soil Biology &
Biochemistry 32, 1707–1716.
Bending, G.D., Turner, M.K., Burns, I.G., 1998. Fate of nitrogen from crop
residues as affected by biochemical quality and the microbial biomass.
Soil Biology & Biochemistry 30, 2055–2065.
Bengtsson, G., Bergwall, C., 1995. Heterotrophic denitrification potential
as an adaptive response in groundwater bacteria. FEMS Microbial
Ecology 16, 307–318.
Bengtsson, G., Bergwall, C., 2000. Fate of 15N labeled nitrate and
ammonium in a fertilized forest soil. Soil Biology & Biochemistry 32,
545 –557.
Birch, H.F., 1958. The effect of soil drying on humus decomposition and
nitrogen availability. Plant and Soil 10, 9–31.
Blagodatsky, S.A., Richter, O., 1998. Microbial growth in soil and nitrogen
turnover: a theoretical model considering the activity state of
microorganisms. Soil Biology & Biochemistry 30, 1743–1755.
Bloem, J., Ellenbroek, F.M., Baer-Gilissen, M.J., Cappenberg, T.E., 1989.
Protozoan grazing and bacterial production in stratified Lake Vechten
estimated with fluorescently labeled bacteria and by thymidine
incorporation. Applied and Environmental Microbiology 55,
1787–1795.
Bloem, J., de Ruiter, P., Koopman, C.G., Lebbink, J.G., Brussaard, L.,
1992. Microbial numbers and activity in dried and rewetted arable soil
under integrated and conventional management. Soil Biology &
Biochemistry 24, 655–665.
Bradbury, N.J., Whitmore, A.P., Hart, P.B.S., Jenkinson, D.S., 1993.
Modelling the fate of nitrogen in crop and soil in the years following
application of 15N-labelled fertilizer to winter wheat. Journal of
Agricultural Science 121, 363–379.
Breitenbeck, G.A., Paramasivam, S., 1995. Availability of 15N-labeled
nonexchangeable ammonium to soil microorganisms. Soil Science 159,
301 –310.
Brookes, P.C., Kragt, J.F., Powlson, D.S., Jenkinson, D.S., 1985. Chloroform fumigation and the release of soil nitrogen the effects of
fumigation time and temperature. Soil Biology & Biochemistry 17,
831 –836.
Clinton, P.W., Newman, R.H., Allen, R.B., 1995. Immobilization of 15N in
forest litter studied by 15N CPMAS NMR spectroscopy. European
Journal of Soil Science 46, 551–556.
Contin, M., Corcimaru, S., De Nobili, M., Brookes, P.C., 2000.
Temperature changes and the ATP concentration of the soil microbial
biomass. Soil Biology & Biochemistry 32, 1219– 1225.
Dalal, R.C., Meyer, R.J., 1987. Long-term trends in fertility of soils under
continuous cultivation and cereal cropping in Southern Queensland.
VII. Dynamics of nitrogen mineralization potentials and microbial
biomass. Australian Journal of Soil Research 25, 461 –472.
Davidson, E.A., Hart, S.C., Firestone, M.K., 1992. Internal cycling of
nitrate in soils of a mature coniferous forest. Ecology 73, 1148–1156.
De Boer, W., Tietema, A., Klein Gunnewick, P.J.A., 1992. The
chemolithotrophic ammonium-oxidizing community in a nitrogen
saturated acid forest in relation to pH-dependent nitrifying activity.
Soil Biology & Biochemistry 24, 229–234.
De Neve, S., Pannier, J., Hofman, G., 1994. Fractionation of vegetable crop
residues in relation to in situ N mineralization. European Journal of
Agronomy 3, 267 –272.
Diaz-Ravina, M., Carballas, T., Acea, M.J., 1988. Microbial biomass and
metabolic activity in four acid soils. Soil Biology & Biochemistry 20,
817 –824.
Drury, C.F., Beauchamp, E.G., 1991. Ammonium fixation, release,
nitrification, and immobilization in high- and low-fixing soils. Soil
Science Society of America Journal 55, 125–129.
Duggin, J.A., Voigt, G.K., Bormann, F.H., 1991. Autotrophic and
heterotrophic nitrification in response to clear-cutting a northern
hardwood forest. Soil Biology & Biochemistry 23, 779–787.
Eiland, F., 1983. A simple method for quantitative determination of ATP in
soil. Soil Biology & Biochemistry 15, 665–670.
G. Bengtsson et al. / Soil Biology & Biochemistry 35 (2003) 143–154
Fisk, M.C., Fahey, T.J., 2001. Microbial biomass and nitrogen cycling
responses to fertilization and litter removal in young northern hardwood
forests. Biogeochemistry 53, 201 –223.
Fisk, M.C., Schmidt, S.K., 1995. Nitrogen mineralization and microbial
biomass nitrogen dynamics in three alpine tundra communities. Soil
Science Society of America Journal 59, 1036– 1043.
Frankenberger, W.T., Abdelmagid, H.M., 1985. Kinetic parameters of
nitrogen mineralization rate of leguminous crops incorporated into soil.
Plant and Soil 87, 257 –271.
Gosz, J.R., White, S.C., 1986. Seasonal and annual variation in nitrogen
mineralization and nitrification along an elevational gradient in New
Mexico. Biogeochemistry 2, 281–297.
Groffman, P.M., Zak, D.R., Christensen, S., Mosier, A., Tiedje, J.M., 1993.
Early spring nitrogen dynamics in a temperate forest landscape.
Ecology 74, 1579–1585.
Gundersen, P., Emmett, B.A., Kjonaas, O.J., Koopmans, C.J., Tietema, A.,
1998a. Impact of nitrogen deposition on nitrogen cycling in forests: a
synthesis of NITREX data. Forest Ecology and Management 101,
37–55.
Gundersen, P., Callesen, I., De-Vreis, W., 1998b. Nitrate leaching in forest
ecosystems is related to forest floor C/N ratios. Environmental Pollution
102, 403 –407.
Hadas, A., Molina, J.A.E., Feigenbaum, S., Clapp, C.E., 1992. Factors
affecting nitrogen immobilization in soil as estimated by simulation
models. Soil Science Society of America Journal 56, 1481–1486.
Hart, S.C., Firestone, M.K., 1991. Forest floor-mineral soil interactions in
the internal nitrogen cycle of an old-growth forest. Biogeochemistry 12,
103–128.
Hart, S.C., Firestone, M.K., Paul, E.A., Smith, J.L., 1993. Flow and fate of
soil nitrogen in an annual grassland and a young mixed-conifer forest.
Soil Biology & Biochemistry 25, 431 –442.
Hart, S.C., Nason, G.E., Myrold, D.D., Perry, D.A., 1994. Dynamics of
gross nitrogen transformations in an old-growth forest: the carbon
connection. Ecology 75, 880– 891.
He, X.T., Stevenson, F.J., Mulvaney, R.L., Kelley, K.R., 1988. Extraction
of newly immobilized nitrogen-15 from an Illinois millisol using
aqueous phenol. Soil Biology & Biochemistry 20, 857 –862.
Heilman, P.E., Dao, T.H., Cheng, H.H., Webster, S.R., Christensen, L.,
1982. Comparison of fall and spring applications of nitrogen-15 labeled
urea to douglas-fir 2. Fertilizer nitrogen recovery in trees and soil after 2
years. Soil Science Society of America Journal 46, 1300–1304.
Iritani, W.M., Arnold, C.Y., 1960. Nitrogen release of vegetable crop
residues during incubation as related to their chemical composition.
Soil Science 9, 74–82.
Jager, G., Bruins, E.H., 1975. Effects of repeated drying at different
temperatures on soil organic matter decomposition and characteristics, and on the soil microflora. Soil Biology & Biochemistry 7,
153–159.
Janssen, B.H., 1996. Nitrogen mineralization in relation to C/N ratio and
decomposability of organic materials. Plant and Soil 181, 39–45.
Jansson, S.L., Persson, J., 1982. Mineralization and immobilization of soil
nitrogen. In: Stevenson, F.J., (Ed.), Nitrogen in Agricultural Soils.
Agronomy: a Series of Monographs, No. 22, American Society of
Agronomy, Inc. Crop Science Society of America, Inc. Soil Science
Society of America, Inc, Madison, WI, pp. 229–252.
Janzen, H.H., Campbell, C.A., Brandt, S.A., Lafond, G.P., Townley-Smith,
L., 1992. Light fraction organic matter in soils from long term crop
rotations. Soil Science Society of America Journal 56, 1799–1806.
Kelley, K.R., Stevenson, F.J., 1985. Characterization and extractability of
immobilized nitrogen-15 from the soil microbial biomass. Soil Biology
& Biochemistry 17, 517–524.
Killham, K., 1986. Heterotrophic nitrification. In: Prosser, J.I., (Ed.),
Nitrification, Special publication for Society of General Microbiology,
vol. 20. IRL Press, Oxford, pp. 117–126.
Marumoto, T., Anderson, J.P.E., Domsch, K.H., 1982. Decomposition of
14
C- and 15N-labelled microbial cells in soil. Soil Biology &
Biochemistry 14, 461 –467.
153
Mary, B., Recous, S., Robin, D., 1998. A model for calculating nitrogen
fluxes in soil using 15N tracing. Soil Biology & Biochemistry 30,
1963–1979.
Mead, D.J., Pritchett, W.L., 1975. Fertilizer movement in a slash pine
ecosystem part 1: uptake of nitrogen phosphorus and nitrogen
movement in the soil. Plant and Soil 43, 451 –465.
Melin, J., Nommik, H., Lohm, U., Flower-Ellis, J., 1983. Fertilizer nitrogen
budget in a Scotch pine Pinus sylvestris ecosystem attained by using
root isolated plots and nitrogen-15 tracer technique. Plant and Soil 74,
249– 264.
Monaghan, R., Barraclough, D., 1995. Contributions to gross N
mineralization from 15N-labelled soil macroorganic matter fractions
during laboratory incubation. Soil Biology & Biochemistry 27,
1623–1628.
Montesinos, M.L., Muro-Pastor, A.M., Herrero, A., Flores, E., 1998.
Ammonium/methylammonium permeases of a cyanobacterium. Identification and analysis of three nitrogen-regulated amt genes in
Synechocystis sp. PCC 6803. Journal of Biological Chemistry 273,
31463–31470.
Paul, E.A., Clark, F.E., 1996. Soil Microbiology and Biochemistry, second
ed., Academic Press, San Diego.
Paul, E.A., Juma, N.G., 1981. Mineralization and immobilization of soil
nitrogen by microorganisms. In: Clark, F.E., Rosswall, T. (Eds.),
Terrestrial Nitrogen Cycles: Processes, Ecosystem Strategies and
Management Impacts, Proceedings of an International Workshop,
Osterfarnebo, Sweden, September 16–22, 1979. Ecological Bulletins
NFR No. 33, 33. Swedish Natural Science Research Council, Stockholm, pp. 179–195.
Pedersen, H., Dunkin, K.A., Firestone, M.K., 1999. The relative importance
of autotrophic and heterotrophic nitrification in a conifer forest soil as
measured by 15N tracer and pool dilution techniques. Biogeochemistry
44, 135–150.
Priha, O., Smolander, A., 1995. Nitrification, denitrification and microbial
biomass N in soil from two N-fertilized and limed Norway spruce
forests. Soil Biology & Biochemistry 27, 305–310.
Pulleman, M., Tietema, A., 1999. Microbial C and N transformations
during drying and rewetting of coniferous forest floor material. Soil
Biology & Biochemistry 31, 275– 285.
Puri, G.M., Ashman, R., 1998. Relationship between soil microbial
biomass and gross N mineralization. Soil Biology & Biochemistry
30, 251–256.
Puri, G.M., Ashman, R., 1999. Microbial immobilization of 15N-labelled
ammonium and nitrate in a temperate woodland soil. Soil Biology &
Biochemistry 31, 929–931.
Ross, D.J., 1991. Microbial biomass in a stored soil: a comparison of
different estimation procedures. Soil Biology & Biochemistry 23,
1005–1007.
Schimel, J.P., Firestone, M.K., Killham, K.S., 1984. Identification of
heterotrophic nitrification in a Sierran forest soil. Applied and
Environmental Microbiology 48, 802– 806.
Schimel, J.P., Firestone, M.K., 1989. Inorganic N incorporation by
coniferous forest floor material. Soil Biology & Biochemistry 21,
41– 46.
Sierra, J., 1996. Nitrogen mineralization and its error of estimation under
field conditions related to the light fraction soil organic matter.
Australian Journal of Soil Research 34, 755–767.
Siewe, R.M., Weil, B., Burkovski, A., Eikmanns, B.J., Eikmanns, M.,
Krämer, R., 1996. Functional and genetic characterization of the
(methyl)ammonium uptake carrier of Corynebacterium glutamicum.
Journal of Biological Chemistry 271, 5398–5403.
Soupene, E., He, L., Yan, D., Kustu, S., 1998. Ammonia acquisition in
enteric bacteria: physiological role of the ammonium/methylammonium transport B (AmtB) protein. Proceedings of the National Academy
of Sciences of the United States of America 95, 7030– 7034.
Stark, J.M., Hart, S.C., 1997. High rates of nitrification and nitrate turnover
in undisturbed coniferous forests. Nature 385, 61– 64.
Tate, R.L. III, 1995. Soil Microbiology, Wiley, New York.
154
G. Bengtsson et al. / Soil Biology & Biochemistry 35 (2003) 143–154
Tietema, A., 1998. Microbial carbon and nitrogen dynamics in coniferous
forest floor material collected along a European nitrogen deposition
gradient. Forest Ecology and Management 101, 29–36.
Trehan, S.P., 1996. Immobilization of 15NHþ
4 in three soils by chemical and
biological processes. Soil Biology & Biochemistry 28, 1021– 1027.
Tsai, C.S., Killham, K., Cresser, M.S., 1997. Dynamic response of
microbial biomass, respiration rate and ATP to glucose additions. Soil
Biology & Biochemistry 29, 1249– 1256.
van de Werf, H., Verstraete, W., 1987. Estimation of active soil microbial
biomass by mathematical analysis of respiration curves: calibration of
the test procedure. Soil Biology & Biochemistry 19, 261 –265.
van Gestel, M., Ladd, J.N., Amato, M., 1991. Carbon and nitrogen
mineralization from two soils of contrasting texture and microaggregate
stability: influence of sequential fumigation, drying and storage. Soil
Biology & Biochemistry 23, 313– 322.
van Veen, J.A., Ladd, J.N., Amato, M., 1985. Turnover of carbon and
nitrogen through the microbial biomass in a sandy loam and a clay soil
incubated with uniformly carbon-14-labeled glucose and nitrogen-15
ammonium sulfate under different moisture regimes. Soil Biology &
Biochemistry 17, 747–756.
van Veen, J.A., Ladd, J.N., Frissel, M.J., 1984. Modelling C and N turnover
through the microbial biomass in soil. Plant and Soil 76, 257– 274.
Verhagen, F.J.M., Laanbroek, H.J., 1991. Competition for ammonium
between nitrifying and heterotrophic bacteria in dual energy-limited
chemostats. Applied and Environmental Microbiology 57, 3255–3263.
Verhagen, F.J.M., Laanbroek, H.J., Woldendorp, J.W., 1995. Competition
for ammonium between plant roots and nitrifying and heterotrophic
bacteria and the effects of protozoan grazing. Plant and Soil 170,
241 –250.
Vitousek, P.M., Gosz, J.R., Grier, C.C., Melillo, J.M., Reiners, W.A., 1982.
A comparative analysis of potential nitrification and nitrate mobility in
forest ecosystems. Ecological Monographs 52, 155– 177.
Wang, W.J., Chalk, P.M., Chen, D., Smith, C.J., 2001. Nitrogen
mineralization, immobilization and loss, and their role in determining
differences in net nitrogen production during waterlogged and aerobic
incubation of soils. Soil Biology & Biochemistry 33, 1305–1315.
Wardle, D.A., Ghani, A., 1995. Why is the strength of relationships
between pairs of methods for estimating soil microbial biomass often so
variable? Soil Biology & Biochemistry 27, 821–828.
West, A.W., Ross, D.J., Cowling, J.C., 1986. Changes in microbial C, N, P
and ATP contents, numbers and respiration on storage of soil. Soil
Biology & Biochemistry 16, 9– 14.
Woodmansee, R.G., Duncan, D.A., 1980. Nitrogen and phosphorus
dynamics and budgets in annual grasslands. Ecology 61, 893–904.
Zaman, M., Di, H.J., Cameron, K.C., Frampton, C.M., 1999. Gross nitrogen
mineralization and nitrification rates and their relationships to enzyme
activities and the soil microbial biomass in soils treated with dairy shed
effluent and ammonium fertilizer at different water potentials. Biology
and Fertility of Soils 29, 178–186.
II
Relieving substrate limitation - soil moisture and temperature
determine gross N transformation rates
Per Bengtson, G. Bengtsson, U. Falkengren-Grerup
Department of Ecology, Lund University, Sölvegatan 37, Lund SE-223 62, Sweden
Abstract
A field experiment was designed with the objective to reveal the interactions between soil moisture,
temperature, total, dissolved, and phosphate buffer extractable C and N, and microbial activity in the control
of in situ gross N mineralization and immobilization rates in a deciduous forest. We set up three alternative
hypotheses to explain variations of the gross N transformations: 1. microorganisms are C limited, 2.
microorganisms are N limited, or 3. neither C nor N limit the microorganisms but moisture and temperature
conditions. Each hypothesis had specific criteria to be fulfilled for its acceptance. The results demonstrated
that neither C nor N were limiting respiration and gross N transformation rates, but the rates were more
dependent on and variable with soil moisture and temperature than the size of the different C and N pools.
The immobilization of N was dependent on the gross mineralization rate, suggesting that if the intracellular N
content of the microorganisms is sufficiently high no additional N is immobilized from the surrounding soil
and no enzymes are produced to mineralize organic N. If the microorganisms are starved for N, enzyme
systems involved in both the assimilation and mineralization of N are activated. The concentrations of C and
N in different forms therefore seem to be a result of microbial activity rather than causing it.
1. Introduction
Mineralization and immobilization by
heterotrophic
microorganisms
are
the
quantitatively most important N transformation
processes in many forest soils (Wessel and
Tietema, 1992; Hart et al., 1997; Fisk and
Fahey, 2001; Bengtsson et al., 2003). The
mineralization process produces substrate for
nitrifiers and plants, but heterotrophic
microorganisms are also successful competitors
for NH4+, and considerable amounts of NH4+
are immobilized in the heterotrophic biomass
(Zak et al., 1990; Verhagen and Laanbroek,
1991; Verhagen et al., 1995). A number of soil
characteristics, including various C and N
pools (Iritani and Arnold, 1960; De Neve et al.,
1994; Fisk and Schmidt, 1995; Bending et al.,
1998), influence the N mineralization rate. The
same characteristics, such as available C
(Jackson et al., 1988, 1990; Barrett and Burke,
2000), the amount of inorganic N (Bengtsson
and Bergwall, 2000), and the soil C/N ratio
(Tietema, 1998), seem to control the rate of
immobilization. Apart from abiotic soil
characteristics, the microbial biomass and
activity, viz., respiration rate, ATP content, and
microbial C and N content, also influence
mineralization and immobilization rates (Alef
et al., 1988; Hart et al., 1994; Tietema, 1998;
Barrett and Burke, 2000; Bengtsson et al.,
2003). However, it seems counter-productive to
consider any of the factors above in isolation if
the aim is to understand the mechanisms that
determine the rates of N mineralization and
immobilization in situ, since the amount of C
and N in different pools of organic matter not
only influence the microbial activity but are
also a result of it. Therefore, we designed a
field experiment in which immobilization and
mineralization rates, microbial activity, and a
number of abiotic factors were simultaneously
quantified.
Our previous observation (Bengtsson et al.,
2003), that variations in the microbial biomass
and activity among three different forest soils
were reflected in corresponding variations in N
transformation rates, was the basis for the
conceptual idea that microorganisms act as
catalytic
converters
of
environmental
conditions
into
mineralization
and
immobilization of N. The dependence of this
conversion
activity
on
individual
environmental conditions has been frequently
1
studied, and there is a general agreement that
the soil moisture content and temperature have
a profound influence on the microbial activity
(Swift et al, 1979; Paul and Clark, 1989;
Killham, 1994). Drying and rewetting of soils
induce significant microbial growth and
activity (Lund and Goksøyr, 1980; Bottner,
1985; West et al., 1986; Bloem et al., 1992),
but the effect of small day-to-day fluctuations
in temperature and moisture content under field
conditions is largely unknown. The amount of
C is also thought to have a large impact on the
microbial activity, and microorganisms in
temperate forest soil are often considered to be
C limited, in spite of the fact that they are
surrounded by large amounts of C. However,
the total C and N content in a soil are not
necessarily correlated to the amount of C and N
that is available to and utilized by the
microorganisms. For instance, Matsumoto et al.
(2000) showed that the activity of N
mineralizing
microorganisms
was
quantitatively
related
to
protein-like
compounds that make up most of a phosphate
buffer extractable N fraction (PON). Likewise,
root exudates, which consist mainly of watersoluble organic acids (Graystone et al., 1996;
Uren, 2001), also have stimulatory effects on
the microbial activity (Graystone et al., 1996).
Our objective was to reveal the interaction
between
moisture,
temperature,
total,
dissolved, and phosphate buffer extractable C
and N, and microbial activity in the control of
mineralization and immobilization of N under
natural conditions in a deciduous forest soil.
The study was conducted in the field for five
days. We set up three alternative hypotheses,
each one with specific criteria to be fulfilled for
its acceptance. Our first hypothesis, that
microorganisms are C limited, had three
criteria: 1) The microbial activity (respiration
rate) is positively dependent on the amount of
C in at least one of the three quantified C
pools; 2) The N mineralization rate is
independent of the amount of N; and 3) The N
immobilization rate is not dependent on the
amount of N but positively dependent on the
amount of C in the pool that has a positive
impact on the microbial activity. The second
hypothesis, that microorganisms are N limited,
had three criteria: 1) The microbial activity
(respiration rate) is independent of the amount
of C; 2) The N mineralization rate is positively
dependent on the amount of N in at least one of
the three quantified organic N pools; 3) The N
immobilization rate is positively dependent on
the gross N mineralization rate and on the
amount of N in the pool that has a positive
impact on the mineralization rate, but not on
the amount of C. Finally, for the third
hypothesis, that the microbial activity
determine the N mineralization and
immobilization rates, and that neither C nor N
limit the microorganisms but rather moisture
and temperature conditions two criteria were
defined: 1) Small daily fluctuations in
temperature and moisture content are translated
into corresponding fluctuations in the microbial
activity; 2) The microbial respiration rate is
independent of the amount of C, and N
immobilization and mineralization rates are
independent of the concentration of N.
2. Material and methods
The experiment was conducted in a mixed
beech/oak stand in Torup in southwestern
Scania, the southernmost province in Sweden
(55°33’N, 13°12’E), on 5-9 May 2003. One
hundred and ten positions, separated by 1 m
and forming a 1×1m grid superimposed on a
10×11 m square, were marked. During day 1-3
and day 5 of the experiment, the microbial
respiration rate was measured in soil samples
taken at 20 of those positions. The same 20
positions were sampled each day. Samples
were taken by inserting a 6 ml scintillation vial
into the soil and then carefully withdrawing the
vial. The respiration rate was estimated by
adding the samples to 20 ml headspace vials,
which were then sealed with rubber septa and
kept at ambient temperature in the field. After
24 hours, the vials were transported to the
laboratory and weighed. The amount of CO2 in
the headspace was measured in a Hewlett
Packard 6890 gas chromatograph, equipped
with a thermal conductivity detector, and
connected to a Perkin Elmer HS 40XL
headspace sampler. Helium was used as a
2
carrier gas at 6 ml min-1 and CO2 was separated
from N2/O2 at 30°C on a HP-PLOT Q column
(15 m, I.D. 0.53 mm). The injector temperature
was 250°C and the filament temperature
300°C. Samples were injected at a split ratio of
3:1. The amount of water in the samples was
determined by weighing the vials before and
after drying at 105°C.
The microbial respiration rate during the
fourth day was estimated at 108 of the marked
positions (two positions were excluded due to
obstacles) using the same sampling procedures
as above. However, after the incubation of the
headspace vials, 12 ml of gas (atmospheric
pressure) from each vial were transferred with
a gastight syringe to a set of evacuated 12 ml
exetainers. Exetainers that did not retrieve at
least 11 ml of gas from the syringe were
discarded. The amount of CO2 in the headspace
of these exetainers was measured in a PDZ
Europa 20-20 continuous flow isotope ratio
mass spectrometer (CF-IRMS) connected to a
gas/solid/liquid preparation module equipped
with a Europa Scientific gas autosampler
(ANCA-GSL,
PDZ
Europa
Scientific
Instruments, Crewe, UK). The exetainers were
flushed with a double holed needle so that all
the headspace gas was transferred first through
a water trap containing magnesium perchlorate,
and then onto a GC column kept at 120°C
where the inert gases were separated. The CO2
was then quantified by IRMS.
On the fourth day of the experiment, 108
PVC tubes (10 cm in length and 11 cm in
diameter) were inserted into the soil at the
marked positions (two positions were excluded
due to obstacles) after removal of the litter.
Plants present within the periphery of the
cylinders were cut at the soil surface with a pair
of scissors and removed. Gross N
mineralization and immobilization rates were
estimated by the pool dilution/enrichment
technique. Twenty ml of a solution containing
2.4 µg 15NH4+-N ml-1 (as 15NH4Cl, 98% 15N,
Cambridge Isotopic Laboratories) were added
to each cylinder with a syringe. To ensure an
even distribution of the label, the needle was
inserted at least at 10 positions in the cylinder
to a depth of five centimetres and slowly drawn
towards the surface while the 15NH4Cl solution
was injected. Within two hours, two soil
samples were taken from each cylinder as
described above. The two soil samples (totally
~10 g dry weight) were combined in a preweighed 50 ml test tube and put on ice. In the
lab, the tubes were weighed a second time and
the soil extracted with 35 ml 1.0 M KCl over
night on an orbital shaker (Tabulator Teknik
AB) kept at 11 °C. The tubes were then
centrifuged at 1200 rpm for 20 min on a
Hermle Z510 centrifuge and the supernatant
transferred to a 100 ml serum flask. Two more
soil samples were taken from the cylinders 24 h
after the first sampling, combined and extracted
as described above.
NH4+ and NO3- were isolated from the soil
KCl extract by using standard IAEA diffusion
procedures (IAEA, 2001), with minor
modifications. Briefly, a standard office paper
punch was used to cut quartz filter discs that
were then placed on a strip of PTFE tape and
prepared with 10 µl of 2.5 M KHSO4. A
second strip of tape was placed on top of the
filter and the two tape strips were sealed by
pressing the open end of a test tube, in a
rocking circular motion around the filter,
against the tape. The trap was added to the
serum bottle containing the KCl extract, 0.4 g
of MgO was added, the serum bottle sealed and
left at ambient temperature with periodic
shaking by hand. After five days, the trap was
removed and the filter was placed in a 5×8 mm
tin cup and left to dry in a dessicator. The
serum bottle was left open for 48 hours to
release residues of ammonia. A new trap was
added followed by 0.4 g of Devarda’s alloy and
0.4 g of MgO, and the incubation was repeated.
15
N and 14N in the filters were determined
using CF-IRMS. The filters were oxidized in
an ANCA-GSL elemental analyzer and NOX
reduced to N2, which was passed to a 20-20
IRMS (PDZ Europa UK). The amount of
NH4+-N and 15NH4+-N, NO3--N and 15NO3--N
was quantified after subtraction of blank values
and calibration against filter discs which had
received 50, 100, or 150 µg N as glycine
(calibrated against an IAEA KNO3 standard)
dissolved in 5 µl of water. The accuracy of the
quantification was determined by calculating
the average value and confidence interval of
3
( g t − g )Z t − ( g 0 − g )Z 0
r=
( f − k )W0 ( f t − k )Wt
(k − g )t + 0
−
c
c
f −k
W
× ln t
t
f0 − k
(4)
(3)
(2)
ft − k 

 ln fo − k  W 0 − Wt
×
c = 1 +
Wt 
t

ln

W 0 
p=c=−
(1)
ft − k
ln
f 0 − k W 0 − Wt
p=
×
Wt
t
ln
W0
Equation
mg N kg –1 day -1
atom %
atom %
atom %
mg N kg -1 day -1
mg N kg -1 day -1
Gross NH4+ consumption
N abundance of NH4+
N abundance of NO3-
15
15
Average 15N abundance of NO3Natural 15N abundance
Gross NH4+ production
Gross NO3- production
c
f
g
g
k
p
r
Concentration of 14N plus 15N in NO3-
mg N kg –1
mg N kg –1
Concentration of 14N plus 15N in NH4+
W
Z
day
Time
t
atom %
Unit
Meaning
Symbol
Table 1. The equations used to calculate the gross N mineralization, immobilization, and nitrification rates, and the definition of the symbols in the equations.
4
eight filter discs (for each concentration of N),
prepared with glycine as above. At all three N
concentrations, the 95% confidence interval
equaled the average value ±3%. The precision
of the isotopic determinations was 0.2‰.
The gross N transformation rates were
calculated by the 15N pool dilution/pool
enrichment technique according to Blackburn
(1979), Nishio et al. (1985), and Wessel and
Tietema (1992). The equations are given in
Table 1. For the calculations of gross NH4+
production (mineralization) and consumption
rates we distinguished between two different
situations, where the NH4+ concentration either
increased (Equations 1 and 2) or remained
constant (Equation 3). The NH4+ concentration
was said to be constant if the values at the
beginning and end of the 24-hour period
differed by less than the internal precision of
the instrument. This was determined by
calculating the average value, W , of W0 and Wt
and forming the 95% CI of it. If the NH4+
concentration was outside the 95% CI,
equations 1 and 2 were used, and if it was
inside equation 3 was used and W was set to
W.
The gross NO3- production (nitrification)
was calculated by Equation 4 (Nishio et al.,
1985; Wessel and Tietema, 1992). The validity
of equation 4 was justified by re-calculating it
once with g0 and once with gt substituted for g
( g = (g0 + gt) / 2). An r calculated for g had to
take a value between r’s calculated for go and
gt (Wessel and Tietema, 1992). The gross
NH4+-N immobilization was then calculated by
subtracting the gross NO3- production (r) from
the gross NH4+ consumption (c). These
calculations
assume
that
the
gross
transformation rates remained constant and that
no 15N was recycled to the enriched pool
during the measurement period. The short (24
h) assay period was an effort to meet this
assumption.
Field moist soil was transferred to 0.2 µm
Z-spin filter units and centrifuged at 26000×g
for 20 min at 4°C on a Hermle Z252MK. This
recovered essentially all soil water since there
was no additional weight loss after drying. The
soil water in the receiver tube was transferred
to a 5×8 mm tin cup, evaporated to dryness at
35°C, and analyzed by IRMS for dissolved
organic C (DOC) and dissolved N (DN).
Phosphate buffer extractable C and N (POC
and PON) were determined according to
Matzumoto et al. (2000), with some
modifications. Field moist soil (3.75 g) and 15
ml of 1/15 M phosphate buffer (pH 7),
consisting of Na2HPO4 · 2H2O (7.3 g l-1) and
KH2PO4 (3.5 g l-1), was added to a 50 ml test
tube. The soil was extracted on a homemade
rotary shaker (~27 rpm) for one hour and
centrifuged at 700×g for 15 min on a Hermle
Z510. Then 50 µl of the supernatant were
transferred to a 5×8 mm tin cup and evaporated
to dryness in a Termaks oven at 35°C before
analysis by IRMS. The total amount of C and
N per gram soil was determined by grinding
dried soil samples in a Retch MM200 ball mill,
transferring a portion of the soil to a 5×8 mm
tin cup and analyzing it in the IRMS.
The day-to-day variation in soil moisture
(∆H2O) at each of the 20 positions used for
respiration measurements was calculated by
subtracting the soil water content day t from
the soil water content day t +1. Changes in the
respiration rate (∆CO2) from day-to-day were
calculated in the same way. Temperature and
precipitation data for the five days of the
experiment were taken from a weather station
in Lund, ~20 km north of the site (SMHI,
2003).
3. Results
The temporal variation of the average
microbial respiration rate was strongly
dependent on the average daily temperature
(Figure 1). The 4.0°C temperature difference
between day 1 and 3 resulted in a difference of
15.8% in respiration rates. Similarly, the rate of
moisture changes in each individual sampling
point corresponded to a decrease/increase in
the microbial respiration rate (Figure 2). The
slightly overestimated ∆CO2 between day 5
and 4, and the slight underestimation between
day 4 and 3 supports the suspicion that a small
loss of CO2 occurred in the transfer from the
5
185
Respiration (mg C kg-1 d-1)
180
Day 1
Y = 69.1 + 7.4 * X
R2 = 0.99, p < 0.001
175
Day 2
170
165
160
Day 5
155 Day 3
Y = 57.7 + 8.1 * X
R2 = 0.87, p < 0.01
Day 4*
150
145
11
12
13
14
15
16
Temperature (°C)
Fig. 1. Relationship between the daily average air
temperature and the microbial respiration rate. The low
respiration rate day 4 can be explained by small
systematic losses of CO2 when it was transferred from
the headspace vials to the exetainers. Two different
regressions were therefore made, one including day 4
(dotted line), and one excluding it (solid line).
800
∆CO2 (mg C kg -1 d-1 )
600
1
Y = 14.3 + 26.7 * X
R2 = 0.50, p < 0.001
1
2
2 2
4
2 1 3 1
2 24 24 22222 2 133
11
4
4 4 224 11 31 2 113 33
1 2 11 1
1
1
400
200
0
-200
-400
-600
-800
-1 5
1
1
2
2
-10
-5
0
5
10
15
∆H2O (%)
Fig. 2. Relationship between the rate of change in soil
moisture and the rate of change in microbial respiration
from one day to another. Number one represents the
difference between day 2 and day 1, number two
represents the difference between day 3 and day 2, and
so on. See Material and Methods section for details.
headspace vials to the exetainers during day 4.
Had the gas losses not occurred, the
relationship between soil drying/wetting rates
and microbial respiration rates would have
been stronger. The rate of change in respiration
from one day to another was in some cases
very high (Figure 2). The spatial variation of
the respiration rates was also associated with
the water content of the soil, with higher
respiration rates at spatial locations with high
soil water content (Figure 3). Each cylinder
received 20 ml of water day 4 when the
15
NH4Cl solution was injected, corresponding
to a rainfall of 2.1 mm. With average soil water
content of 25 % this corresponds to an increase
in water content of approximately six
percentage points. Also day 1 and 2 of the
experiment had incidents of light rainfall (<1
mm).
The C concentrations varied substantially
among the soil samples (Table 2), with
coefficients of variation in the order of 2555%. However, the respiration rate was not
dependent on the concentration of DOC, POC,
or total C (Table 3), which would be sufficient
to reject the first hypothesis that the soil
microorganisms were C limited. Moisture
(Table 3, Figure 2 and 3) and temperature
(Figure 1) conditions alone therefore seemed to
be able to explain both spatial and temporal
variations of the respiration rate.
The gross N mineralization rate was
positively related to the soil water content
(Table 3), with high mineralization rates in
soils with high water content, but also weakly
to the microbial activity as determined by the
respiration rate (Figure 4). Thus, it was
difficult to distinguish a direct effect of soil
water content on N mineralization rates from
an indirect via the microbial activity. In
addition to the microbial activity and the soil
water content, the concentration of PON also
had a positive influence on the gross
mineralization
rate
(Table
3).
The
concentration of total N, DN, and NH4+ had no
significant
influence
on
the
gross
mineralization rate (Table 3). Thus, the first
two criteria of the second hypothesis, that the
microbial activity (respiration rate) was not
dependent on the amount of C, and that the N
mineralization rate was positively dependent on
the amount of N in at least one of the
quantified
N
pools,
were
satisfied.
6
Table 2. Summary statistics for C and N pools. All weights are on dry weight basis.
Statistic
NH4+-N
NO3--N
Total C
Total N C/Nratio
POC
(mg kg-1)
(mg kg-1)
(g kg-1)
(g kg-1)
(soil)
(g kg-1)
Mean
10.9
8.8
40.4
2.9
17.3
3.0
SD
6.7
3.2
12.0
2.1
7.3
0.8
CV (%)
61.1
36.7
29.8
71.3
42.3
26.9
Skewness
1.2
0.8
1.1
3.1
1.1
0.4
10th Percentile
4.3
4.7
27.7
1.4
9.5
2.1
Median
9.6
8.6
38.8
2.4
16.5
2.9
90th Percentile
20.4
12.6
57.3
4.8
28.6
4.0
PON
(mg kg-1)
251.5
72.7
28.9
0.8
159.2
238.9
355.9
C/N ratio
(POC/PON)
12.3
2.0
16.3
0.4
9.6
12.1
15.2
DOC
(mg kg-1)
37.1
20.2
54.5
1.2
16.2
30.8
62.0
DN
(mg kg-1)
12.6
6.6
52.8
0.8
5.2
12.0
20.3
C/N ratio
(DOC/DN)
4.2
4.9
115.7
4.0
1.2
2.5
8.5
7
Table 3. The dependency of the gross mineralization, immobilization, and respiration rates on the different abiotic soil
properties. (Stepwise multiple regression analysis, forward procedure, F to enter = 4.000, F to remove = 3.996). All
factors except mg DOC kg-1, mg DN kg-1, mg DN l-1, and POC/PON were ln-transformed before analysis.
Dependent
Factors not in model
-1
Factors in model
Coefficient
Model R2
Significance
Respiration
mg DOC l
mg DOC kg-1
mg POC kg-1
g C kg -1
H2O (%)
3.22
0.51
<0.001
Mineralization
mg NH4-N kg-1
mg DN l-1
mg DN kg-1
g N kg-1
H2O (%)
mg PON kg-1
2.99
0.99
0.22
0.30
<0.001
NH4+ immobilization
mg NH4-N kg-1
mg DOC kg-1
mg DN kg-1
mg DOC kg-1
DOC/DON
mg PON kg-1
mg POC kg-1
POC/PON
g N kg-1
g C kg-1
C/N ratio
H2O (%)
Mineralization
mg DN l-1
0.99
0.02
0.44
0.58
<0.001
On average, 11.2 mg NH4+-N kg-1 d-1 were
immobilized, slightly less than the amount that
was mineralized (Table 4). The immobilization
rate was not dependent on the concentration of
DOC, POC, or total soil C (Table 3),
suggesting, again, that microorganisms were
not C limited. This conclusion was reinforced
by the lack of influence of varying C/N ratios
on the gross immobilization rate (Table 3).
Neither was there any dependence of the N
immobilization rate on the concentration of
NH4+, PON or the total N content of the soil,
ruling out N limitation of microorganisms as a
controlling factor of immobilization (Table 3).
On the other hand, there was a positive
relationship between gross mineralization and
immobilization rates (Table 3). High DN
concentrations also had an additional positive
influence on the immobilization rate (Table 3),
but the main factor controlling immobilization
was the gross mineralization rate. A doubling
of this rate would also double the
immobilization rate (Table 3). However, since
the immobilization rate, in contrast to the
mineralization rate, was not dependent on the
amount of PON, which was the third criteria of
the hypothesis that microorganisms were N
limited, we had to reject this hypothesis as
well. The third hypothesis, that moisture and
temperature conditions alone determine the
microbial activity and hence the N
mineralization and immobilization rates, was
the only one that could not be rejected on the
basis of not fulfilling the defined criteria.
4. Discussion
The close connection between temporal
variations of temperature and respiration rates
and the high degree of explanation of spatial
and temporal variation in respiration by the soil
water content indicate that temperature and
water have the greatest influence on the
microbial activity. The water content and its
co-variable microbial activity also explained
most of the variation of the gross N
mineralization rate, reinforcing observations by
8
ln Respiration (mg C kg -1 d-1)
7.0
6.5
Y = -5.9 + 3.3 * X
R2 = 0.58, p < 0.001
6.0
5.5
5.0
4.5
4.0
3.5
3.0
2. 9 3.0 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3. 8
ln H2O (%)
Fig. 3. The relationship between the gravimetric soil
water content and the microbial respiration rate during
day 4 of the experiment.
ln Mineralization (mg N kg-1 d-1)
5
4
3
2
1
0
Y = - 0.97 + 0.70 * X
R2 = 0.20, p < 0.01
-1
-2
3.0
3.5
4.0
4.5
5.0
5.5
-1
6.0
6. 5
-1
ln Respiration (mg C kg d )
Fig. 4. The relationship between the microbial
respiration rate and the gross mineralization rate. The
maximum and minimum derivative of the function
describing
back-transformed
respiration
and
mineralization values were 0.09 and 0.05.
Puri and Ashman (1998) and Thomsen et al.
(1999). Several studies have observed a covariation in microbial respiration and N
mineralization rates with a slope that was
similar to ours (Alef et al., 1988; Puri and
Ashman, 1998; Scott et al., 1998; Tietema,
1998; Barret and Burke, 2000), while the slope
that Hart et al. (1994) found in an Oregon
forest soil was 3-4 times steeper, suggesting
that a certain relative change of the respiration
rate corresponded to a smaller change in the
gross mineralization rate in the Torup soil than
in the Oregon soil. This variation from one soil
to another of the slope of the relationship
between mineralization and respiration rates is
also evident from the soils survey of Schimel
(1986). The difference among soils may be
related to several soil and microbial community
characteristics: 1. Microorganisms with a high
C/N ratio assimilate less N per assimilated C
than microorganisms with a low C/N ratio,
resulting in a shallow slope. 2. In soils with
high
competition
for
NH4+
among
heterotrophic microorganisms, nitrifiers, and
plants, the heterotrophs would need to
mineralize more N to meet their metabolic
demands since plants and nitrifiers would use
some of the produced NH4+, resulting in a steep
slope. 3. In non- or slow growing microbial
populations, essentially all utilized C is
diverted towards catabolic needs, resulting in a
shallow slope since little or no N is assimilated.
4. Differences in microbial growth efficiency
among soils may arise from differences in the
chemical composition of the soil organic C.
We tried to address the chemical
heterogeneity and bioavailability of the organic
C by dividing it into three fractions, total C,
POC, and DOC. DOC was assumed to be best
correlated with respiration for a number of
reasons: 1. DOC is by definition dissolved in
the soil water and can easily diffuse to the
cells, 2. The C containing molecule must pass
the cell membrane, and 3. Smaller molecules
(DOC) are more likely to pass the cell
membrane than larger (POC). The absence of a
significant relationship between the respiration
rate and the DOC concentration, or any other C
pool, shows that organic C was not a main
factor determining the microbial activity and
the rate of turnover. In contrast, Hart et al.
(1994) found the respiration rate and the
concentration of K2SO4 extractable organic C
to be inversely related to chloroform fumigated
microbial C during an initial phase of high
microbial growth in soil samples that were
incubated under constant temperature and soil
moisture conditions, suggesting that extractable
organic C was a major C source for
microorganisms. Soil organic carbon and
microbial biomass are in general well
9
Table 4. Summary statistics for gross N mineralization, immobilization, and microbial respiration
rate.
Statistic
Gross mineralization Gross NH4+ immobilization
Microbial respiration
(mg N kg-1 d-1)
(mg N kg-1 d-1)
(mg C kg-1 d-1)
Mean
15.3
11.2
149.2
SD
11.6
8.6
118.5
CV (%)
76.0
76.9
79.4
Skewness
1.1
0.9
2.5
10th Percentile
2.8
1.4
56.3
Median
13.8
9.8
115.7
90th Percentile
33.6
23.5
306.0
correlated (Zak et al., 1990; Zhang and Zhang,
2003; Milne and Haynes, 2004), but as Bloem
et al. (1989) have demonstrated, the microbial
activity can vary up to tenfold in soils with the
same microbial biomass.
Since the majority of laboratory
experiments on C and N turnover have been
made at constant temperature and constant and
non-limiting soil water conditions (references),
the variations in activity per biomass unit are
restrained and the importance of soil organic
matter in controlling the C and N
transformation is overestimated. Heterogeneity
in soil physical properties cause varying spatiotemporal patterns in soil water status at the
metre scale (Dekker et al., 1999; Wendroth et
al., 1999), and species-specific transpiration
behavior of trees results in a high spatiotemporal variation in drying rates in mixed
forest stands (Schume et al. 2003). When this
spatial and temporal variability of soil moisture
content and temperature is included in N
transformation assays, the dependence on soil
organic matter seems to be relieved.
The change in respiration rates from one
day to another in response to soil
drying/wetting was sometimes large but of the
same magnitude as observed by Pulleman and
Tietema (1999). They found respiration rates to
increase by 250% four hours after wetting field
moist forest floor material, and by ~ 4300%
when the same material had been air dried for
11 days before rewetting. Similarly, Fierer and
Schimel (2003) observed respiration rates in
dried forest and grassland soils to increase by
400-500% compared with field moist soils
when the soils were rewetted.
In addition to the microbial activity, the
concentration of PON was positively related to
the gross N mineralization rate, in agreement
with Matsumoto et al. (2000). PON is a
homogeneous, bacterially derived group of
protein-like compounds with a size of 8-9 kDa,
which seems to be the source of the
mineralized N (Matsumoto et al. 2000).
However, it is also possible that PON is not
mainly a source of N in itself, but that it
represents residues of exoenzymes that become
abundant and released to degrade other organic
N compounds during high microbial activity.
Indeed, Zaman et al. (1999) found correlations
between extracellular enzyme activities and the
gross mineralization rate. This interpretation is
also supported by the results of Matsumoto et
al. (2000), who observed the concentration of
the phosphate buffer extractable protein-like
compounds to increase in response to addition
of various C sources known to increase the
microbial activity and N mineralization, and
possibly also the production and release of
PON. Such a positive relationship between the
concentration of PON and the respiration rate
was found in our study (R2 = 0.16, p < 0.01).
On the other hand, the concentration of POC
had no influence on the respiration rate,
possibly for the same reason as for DOC. In
addition, the phosphate buffer also extracts
some more refractory humic substances
(Matsumoto et al., 2000), which may obscure a
quantitative relationship between the amount of
POC and the respiration rate.
The
dependence
of
the
gross
immobilization rate on the gross mineralization
rate (Table 3) is in agreement with several
laboratory studies with both sieved soils and
intact soil cores (Davidson et al., 1992; Hart et
al., 1997; Verchot et al., 2001; Bengtsson et al.,
2003). This seems to be a consistent
10
relationship, based on a diversity of soils and
incubation conditions, probably because the
same microorganisms participate in both
processes. The synthesis and/or activation of
proteins involved in the uptake and utilization
of nitrogenous compounds is closely regulated
in concert with the availability of their
substrates (Merrick and Edwards, 1995). That
is, if the intracellular N content of the
microorganisms is sufficient, no additional N is
immobilized from the surrounding soil, and no
enzymes are produced to mineralize organic N.
On the other hand, if the microorganisms are
starved for N, a number of enzyme systems
involved in both the assimilation of inorganic
N and mineralization of organic N are
activated.
The small but positive influence of the DN
concentration on the immobilization rate
should not be interpreted as an evidence for N
limitation, for several reasons. First, if
microorganisms were N limited the
immobilization rate should be more dependent
on the total amount of DN per gram of soil than
on the concentration in the soil water. Second,
the concentration of ammonium and the gross
immobilization rate were not correlated. Third,
if the concentration of DN was limiting the
immobilization rate, it should also limit the
mineralization rate, since the DN would have
to be mineralized to ammonium before
immobilized. However, that dependence was
not observed (Table 3). Alternatively, the small
increase in immobilization rates with
increasing DN concentrations may be an effect
of uptake of N for osmoprotection. During
osmotic up-shift caused by increased
concentration of solutes in the soil water as a
soil dries, microorganisms maintain turgor by
accumulating osmotically active solutes in the
cytoplasm, of which many contain N, e.g., free
amino acids and derivates thereof, the dipeptide
N-acetylglutaminylglutamine, and quaternary
amines (Kempf and Bremer, 1998).
One objection to the statement that
microorganisms are not N limited is the limited
N leakage from N fertilization experiments
(Emmett et al., 1995; Gundersen et al., 1998;
Kjønaas et al., 1998). This is difficult to
explain unless microorganisms are N limited,
since there is no intracellular storage
compounds for N and it is most likely actively
taken up by microorganisms. However, if we
recalculate our mean gross N mineralization
rate to g m-2 day (using a soil density of 1 g cm3
), we end up with a gross N mineralization of
0.75 g m-2 day in the top 5 cm of the soil. The
natural N deposition in southwestern Scania,
0.0054 g m-2 day (20 kg ha-1, Lagner et al.,
1996), and the N addition in the NITREX
experiments, 0.0096 g m-2 day (Gundersen et
al., 1998), represents 0.7-1.3% of the
ammonium that was naturally produced. That
is, a reduction in the gross N mineralization
rate of about 1% would be enough to
compensate for the addition of inorganic N.
This decrease would hardly be detectable given
the great spatial and temporal variability of N
transformation rates (Paper III).
Taken together, the positive relation
between
microbial
activity
and
N
transformation rates and the negligible
importance of the size of different C and N
pools suggest that, at least in temperate forest
soils rich in organic matter, neither C nor N is
limiting the microbial activity. Other factors,
such as moisture content, drying rate, and
temperature may be more important in
determining the microbial activity and the N
transformation rates. Numerous laboratory
studies have also shown that both respiration
and N mineralization rates increase with
increasing moisture and temperature (e.g. Puri
and Ashman, 1998; Thomsen et al., 1999;
Subke et al., 2003; Templer et al., 2003). Our
study demonstrates that those results are
applicable under field conditions, although the
range of variation in temperature and soil
moisture content from day to day is several
times lower than in most laboratory
experiments. By relieving the limitation of
water and temperature on the microbial
activity, as is often done in laboratory
experiments, conditions are created at which
the rates of C and N assimilation and
mineralization are determined by the quality
and bioavailabilty of the soil organic matter.
11
Acknowledgements
This work was sponsored by grants from
the Foundation for Strategic Environmental
Research in Sweden. Niklas Törneman and
Therese Nilsson are greatly acknowledged for
their assistance in the field and laboratory.
References
Alef, K., Beck, T.H., Zelles, L., Kleiner, D.
1988. A comparison of methods to estimate
microbial biomass and N-mineralization in
agricultural and grassland soils Soil Biology
and Biochemistry 20, 561-565.
Barret, J.E., Burke I.C. 2000. Potential nitrogen
immobilization in grassland soils across a
soil organic matter gradient. Soil Biology
and Biochemistry 32:1707-1716.
Bending, G.D., Turner, M.K., Burns, I.G. 1998.
Fate of nitrogen from crop residues as
affected by biochemical quality and the
microbial biomass. Soil Biology and
Biochemistry 30, 2055-2065.
Bengtsson, G., Bengtson, P., Månsson, K.F.
2003. Gross nitrogen mineralization-,
immobilization-, and nitrification rates as a
function of soil C/N ratio and microbial
activity. Soil Biology and Biochemistry 35,
143-154.
Bengtsson, G., Bergwall, C. 2000. Fate of 15N
labeled nitrate and ammonium in a fertilized
forest soil. Soil Biology and Biochemistry
32, 545-557.
Blackburn, T.H. 1979. Method for measuring
rates of NH4+ turnover in anoxic sediments,
using a 15N-NH4+ dilution technique.
Applied and Environmental Microbiology
37, 760-765.
Bloem, J., Ellenbroek, F.M., Baer-Gilissen,
M.J., Cappenberg, T.E. 1989. Protozoan
grazing and bacterial production in stratified
Lake Vechten estimated with fluorescently
labeled bacteria and by thymidine
incorporation. Applied and Environmental
Microbiology 55, 1787-1795.
Bloem, J., de Ruiter, P., Koopman, C.G.,
Lebbink, J.G., Brussaard, L. 1992. Microbial
numbers and activity in dried and rewetted
arable soil under integrated and conventional
management. Soil Biology and Biochemistry
24, 655-665.
Bottner, P. 1985. Response of microbial
biomass to alternate moist and dry conditions
in a soil incubated with 14C- and 15N-labelled
plant
material.
Soil
Biology
and
Biochemistry 17, 329-337.
Davidson, E.A., Hart, S.C., Firestone, M.K.
1992. Internal cycling of nitrate in soils of a
mature coniferous forest. Ecology 73, 11481156.
Dekker, L.W., Ritsema, C.J., Wendroth, O.,
Jarvis, N., Oostindie, K., Pohl, W., Larsson,
M., Gaudet, J.-P. Moisture distributions and
wetting rates of soils at experimental fields
in the Netherlands, France, Sweden and
Germany. Journal of Hydrology 215, 4-22.
De Neve, S., Pannier, J., Hofman, G. 1994.
Fractionation of vegetable crop residues in
relation to in situ N mineralization. European
Journal of Agronomy 3, 267-272.
Emmett, B.A., Brittain, S.A., Hughes, S.,
Kennedy, V. 1995. Nitrogen additions
(NaNO3 and NH4NO3) at Aber forest, Wales:
II. Response of trees and soil nitrogen
transformations. Forest Ecology and
Management 71:61-73.
Fierer, N; Schimel, J.P. 2003. A proposed
mechanism for the pulse in carbon dioxide
production commonly observed following
the rapid rewetting of a dry soil. Soil Science
Society of America Journal 67, 798-805.
Fisk, M.C., Fahey, T.J. 2001. Microbial
biomass and nitrogen cycling responses to
fertilization and litter removal in young
northern hardwood forests. Biogeochemistry
53, 201-223.
Fisk, M.C., Schmidt, S.K. 1995. Nitrogen
mineralization and microbial biomass
nitrogen dynamics in three alpine tundra
communities. Soil Science Society of
America Journal 59, 1036-1043.
Grayston, S.J., Vaughan, D., Jones, D. 1996.
Rhizosphere carbon flow in trees, in
comparison with annual plants: the
importance of root exudation and its impact
on microbial activity and nutrient
availability. Applied Soil Ecology 5, 29-56.
Gundersen, P., Emmett, B.A., Kjønaas, O.J.,
Koopmans, C.J., Tietema, A. 1998. Impact
12
of nitrogen deposition on nitrogen cycling in
forests: a synthesis of NITREX data. Forest
Ecology and Management 101, 37-55.
Hart, S.C., Nason, G.E., Myrold, D.D., Perry,
D.A. 1994. Dynamics of gross nitrogen
transformations in an old-growth forest: the
carbon connection. Ecology 75, 880-891.
Hart, S.C., Binkley, D., Perry, D.A. 1997.
Influence on red alder on soil nitrogen
transformations in two conifer forests of
contrasting productivity. Soil Biology and
Biochemistry 29, 1111-1123.
IAEA Training Course Series No. 14. Use of
Isotope and Radiation Methods in Soil and
Water Management and Crop Nutrition,
IAEA, Vienna, 2001.
Iritani, W.M., Arnold, C.Y. 1960. Nitrogen
release of vegetable crop residues during
incubation as related to their chemical
composition. Soil Science 9, 74-82.
Jackson, L.E., Schimel, J.P., Firestone, M.K.
1989. Short-term partitioning of ammonium
and
nitrate
between
plants
and
microorganisms in an annual grassland. Soil
Biology and Biochemistry 21, 409-415.
Jackson, L.E., Strauss, R.B., Firestone, M.K.,
Bartolome, J.W. 1988. Plant and soilnitrogen dynamics in California annual
grassland. Plant and Soil 110, 9-17.
Jackson, L.E., Strauss, R.B., Firestone, M.K.,
Bartolome, J.W. 1990. Influence of tree
canopies on grassland productivity and
nitrogen dynamics in deciduous oak savanna.
Agriculture, Ecosystems and Environment
32, 89-105.
Kempf, B., Bremer, E. 1998. Uptake and
synthesis of compatible solutes as microbial
stress
responses
to
high-osmolarity
environments. Archives of Microbiology
170, 319-330.
Kilham, K. 1994. Soil Ecology. Cambridge
University Press, Cambridge.
Kjønaas, O.J., Stuanes, A.O., Huse, M. 1998.
Effects of weekly nitrogen additions on N
cycling in a coniferous forest catchment,
Gårdsjön, Sweden. Forest Ecology and
Management 101:227-249.
Lagner, J., Persson, C., Robertson, L.,
Ullerstig, A. 1996. Air pollution assessment
study using the MATCH modelling system –
Applications to sulphur and nitrogen
compounds over Sweden 1994. SMHI RMK
Report, Norrköping.
Lund, V., Goksøyr, J. 1980. Effects of water
fluctuations on microbial biomass in soil.
Microbial Ecology 6, 115-123.
Matsumoto, S., Noriharu, A., Yamagata, M.
2000. Extraction of mineralizable organic
matter from soils by a neutral phosphate
buffer solution. Soil Biology and
Biochemistry 32, 2000.
Merrick, M.J., Edwards, R.A. 1995. Nitrogen
control in bacteria. Microbiological Reviews
59, 604-622.
Milne, R.M., Haynes, R.J. 2004. Soil organic
matter, microbial properties, and aggregate
stability under annual and perennial pastures.
Biology and Fertility of Soils 39, 172-178.
Nishio, T., Kanamori, T., Fujimoto, T. 1985.
Nitrogen transformations in an aerobic soil
as determined by a 15N-NH4+ dilution
technique. Soil Biology and Biochemistry
17, 149-154.
Paul, E.A., Clark, F.E. 1989. Soil Microbiology
and Biochemistry. AP, San Diego.
Pulleman, M., Tietema, A. 1999. Microbial C
and N transformations during drying and
rewetting of coniferous forest floor material.
Soil Biology and Biochemistry 31, 275-285.
Puri, G.M., Ashman, R. 1998. Relationship
between soil microbial biomass and gross N
mineralization.
Soil
Biology
and
Biochemistry 30, 251-256.
Schimel, D.S. 1986. Carbon and nitrogen
turnover in adjacent grassland and cropland
ecosystems. Biogeochemistry 2, 345-357.
Schimel, J.P, Jackson, L.E., Firestone, M.K.
1989. Spatial and temporal effects on plantmicrobe competition for inorganic nitrogen
in a California annual grassland. Soil
Biology and Biochemistry 21, 1059-1066.
Schume, H., Jost, G., Katzensteiner, K. 2003.
Spatio-temporal analysis of the soil water
content in a mixed Norway spruce (Picea
abies (L.) Karst.)-European beech (Fagus
sylvatica L.) stand. Geoderma 112, 273-287.
Scott, N.A., Parfitt, R.L., Ross, D.J., Salt, G.J.
1998. Carbon and nitrogen transformations I
New Zealand plantation forest soils from
13
sites with different N status. Canadian
Journal of Forest Research 28, 967-976.
Sierra, J. 1997. Temperature and soil moisture
dependence of N mineralization in intact soil
cores. Soil Biology and Biochemistry 29,
1557-1563.
SMHI. 2003. Väder och Vatten 5, 15.
Subke, J.-A., Reichstein, M., Tenhunen, J.D.
2003. Explaining temporal variation in soil
CO2 efflux in a mature spruce forest in
Southern Germany. Soil Biology and
Biochemistry 35, 1467-1483.
Swift, M.E., Heal, O.W., Anderson, J.M. 1979.
Decomposition in Terrestrial Ecosystems.
Blackwell Scientific Publications, Oxford.
Templer, P., Findlay, S., Lovett, G. 2003. Soil
microbial
biomass
and
nitrogen
transformations among five tree species of
the Catskill Mountains, New York, USA.
Soil Biology and Biochemistry 35, 607-613.
Thomsen, I.K., Schjonning, P., Jensen, B.,
Kristensen, K., Christensen, B.T. 1999.
Turnover of organic matter in differently
textured soils - II. Microbial activity as
influenced by soil water regimes. Geoderma
89, 199-218.
Tietema, A., 1998. Microbial carbon and
nitrogen dynamics in coniferous forest floor
material collected along a European nitrogen
deposition gradient. Forest Ecology and
Management 101, 29-36.
Tietema, A., Wessel, W.W. 1992. Gross
nitrogen transformations in the organic layer
of acid forest ecosystems subjected to
increased atmospheric nitrogen input. Soil
Biology and Biochemistry 24, 943-950.
Uren, N.C. 2001. Types, amounts, and possible
functions of compounds released into the
rhizosphere by soil-grown plants. In: Pinton,
R., Varanini, Z., Nannipieri, P. (Editors).
The Rhizosphere – Biochemistry and
Organic Substances at the Soil-Plant
Interphase, Marcel Dekker, Ink., New York.
Verchot, L.V., Holmes, Z., Mulon, L.,
Groffman, P.M., Lovett, G.M. 2001. Gross
vs net rates of N mineralization and
nitrification as indicators of functional
differences between forest types. Soil
Biology and Biochemistry 33, 1889-1901.
Verhagen, F.J.M., Laanbroek, H.J. 1991.
Competition for ammonium between
nitrifying and heterotrophic bacteria in dual
energy-limited chemostats. Applied and
Environmental Microbiology 57, 3255-3263.
Verhagen,
F.J.M.,
Laanbroek,
H.J.,
Woldendorp, J.W. 1995. Competition for
ammonium between plant roots and
nitrifying and heterotrophic bacteria and the
effects of protozoan grazing. Plant and Soil
170, 241-250.
Wendroth, O., Pohl, W., Koszinski, S.,
Rogasik, H., Ritsema, C.J., Nielsen, D.R.
1999.
Spatio-temporal
patterns
and
covariance structures of soil water status in
two Northeast-German field sites. Journal of
Hydrology 215, 38-58.
Wessel, W.W., Tietema, A. 1992. Calculating
gross N transformation rates of 15N pool
dilution experiments with acid forest litter:
analytical and numerical approaches. Soil
Biology and Biochemistry 24, 931-941.
West, A.W., Ross, D.J., Cowling, J.C. 1986.
Changes in microbial C, N, P and ATP
contents, numbers and respiration on storage
of soil. Soil Biology and Biochemistry 16, 914.
Zak, D.R., Grigal, D.F., Gleeson, S., Tilman,
D. 1990. Carbon and nitrogen cycling during
old-field succession - constraints on plant
and microbial biomass. Biogeochemistry 11,
111-129.
Zak, D.R., Groffman, P.M., Pregitzer, K.S.,
Christensen, S., Tiedje, J.M. 1990. The
vernal dam: Plant-microbe competition for
nitrogen in northern hardwood forests.
Ecology 71, 651-656.
Zaman, M., Di, H.J., Cameron, K.C.,
Frampton, C.M. 1999. Gross nitrogen
mineralization and nitrification rates and
their relationship to enzyme activities and
the soil microbial biomass in soils treated
with dairy shed effluent and ammonium
fertilizer at different water potentials.
Biology and Fertility of Soils 29, 178-186.
Zhang, H., Zhang, G.L. 2003. Microbial
biomass carbon and total organic carbon of
soils as affected by rubber cultivation.
Pedosphere 13, 353-357.
14
III
Spatial distribution of gross N transformation rates and plants
– A reciprocal relationship
Per Bengtson, U. Falkengren-Grerup, G. Bengtsson
Department of Ecology, Lund University, Sölvegatan 37, Lund SE-223 62, Sweden
Abstract
A geostatistical analysis was performed to reveal the reciprocal dependence between the spatial variation of
in situ gross N mineralization, NH4+ immobilization, and nitrification rates and the abundance, distribution,
and species composition of understory plants in a mixed beach-oak forest. Variations in N transformation
rates explained the spatial variation of understory plants within the forest, even at scales as small as a few
metres. NO3- preferring plants were more common in areas with low heterotrophic activity, while the total
cover was higher in areas with high N transformation rates. Beech and oak trees also had an effect on the
spatial variation of the understory. Beech trees created conditions that were suitable for NO3- preferring
plants, whereas the plant cover was higher under oak trees, probably because of less light interception. Oak
generally had a positive impact on the N transformation rates and beech a negative, probably in response to
the litter quality of the two species. The influence of trees alone could not explain the full magnitude of the
variation, or the presence of overlapping hotspots with high mineralization and immobilization rates. These
were probably caused by other factors, such as soil moisture content.
1. Introduction
Large-scale spatial heterogeneity in
topography, bedrock geology, and hydrology
has a major influence on variations in habitat
characteristics, such as water and nutrient
availability, light exposure, and temperature in
a forest (Hutchinson et al., 1999; Hook and
Burke, 2000). These variations are reflected in
the species composition of plant communities.
At smaller scales, within-habitat spatial
heterogeneity
maintains
diversity
and
facilitates the co-existence of species (Chesson,
1986; Tilman and Kareiva, 1997). For instance,
the heterogeneity of ammonium and nitrate
concentrations at a scale of 2 m (Lechowicz
and Bell, 1991; Jackson and Caldwell, 1993) is
likely to be recognized by roots of trees and
shrubs and affect species distributions
differently depending on differences in uptake
affinity. Spatial variations at a scale of 0.2 m
(Schlesinger et al., 1996; Farley and Fitter,
1999) appear to be relevant for herbaceous
plants. The plants themselves are active in
creating and maintaining this mosaic of fertility
due to e.g. differences in transpiration,
throughfall, and litter deposition (Ferrari, 1999;
Schume et al., 2003).
Microbial biomass and composition exhibit
spatial heterogeneity at the same scales as
plants (Ettema and Wardle, 2002). Their patch
sizes are structured by topography, soil carbon,
and moisture at scales from tens to hundreds of
metres (Fromm et al., 1993; Robertson et al.,
1997), but by plant size and spacing at
centimetres to metres scale (Bruckner et al.,
1999; Klironomos et al., 1999; Saetre, 1999;
Saetre and Bååth; 2000). Nitrifying bacteria
can perceive heterogeneity even at a scale of
some few millimetres and below, perhaps in
response to micro-sized pores and soil
aggregates and fine roots (Grundmann and
Debouzie, 2000; Grundmann et al. 2001).
Microbial activity, such as net N
transformations and respiration, also seems to
be structured by the spatial patterning of plants
(Robertson et al., 1988; Saetre, 1999; Stoyan et
al., 2000).
Primary production in forest ecosystems is
often considered to be N limited (Tamm, 1991;
Knops and Tilman, 2000), and N availability is
an important factor in determining the
distribution and species composition of plants
(Diekmann and Falkengren-Grerup, 1998;
Hutchinson et al., 1999; Falkengren-Grerup
and Diekmann, 2003). N availability cannot
1
simply be defined as the concentration of
inorganic N, since this N pool often is very
small and may have a turnover time of less than
a day (Hart et al., 1994; Scott et al., 1998;
Verchot et al., 2001; Bengtson et al., 2003).
The use of net mineralization and nitrification
rates in defining N availability to plants may be
misleading, since it is not uncommon to find
high gross mineralization rates in soils with
low or negative net mineralization (Davidson et
al., 1992; Hart et al., 1994; Verchot et al.,
2001; Bengtsson et al., 2003). Similarly,
several cases of rapid NO3- turnover in soils
without net nitrification have been documented
(Davidson et al., 1992; Stark and Hart, 1994;
Scott et al., 1998; Verchot et al. 2001).
However, floristic variation among
deciduous forests can partly be explained by
differences in net N mineralization and
nitrification
rates
(Diekmann
and
Falkengren-Grerup, 1998; Hutchinson et al.,
1999; Falkengren-Grerup and Diekmann,
2003). This does not necessarily reflect a link
between plants and microbial activities but may
be an indirect effect of the same soil properties
affecting
both
the
plant
community
composition and N transformation rates. On the
other hand, plant litter provides the necessary
carbon for supporting N immobilization by
microorganisms (Watkins and Barraclough,
1996; Andersen and Jensen, 2001; Knops et al.,
2002). It is also a source of N for heterotrophic
microorganisms in undisturbed soils (Muller
and Bormann, 1976; Jandl, 1997; Tardiff,
1998). Spatial heterogeneity in plant litter
distribution and composition may cause the
observed differences in net N transformation
rates since the quality of the litter differs
between plants (Ferrari, 1999; Sariyildiz and
Anderson, 2003a and 2003b; Chapman et al.,
2003; Schweitzer et al., 2004), and since
different compounds in plant litter might either
increase or decrease the microbial activity
(Northup, 1995; Henriksen & Breland, 1999;
Hongve et al., 2000; Kraus et al., 2004).
Furthermore, some plant species may be better
than others at competing with microorganisms
for N due to differences in N demand and
uptake capacity. This may also explain the
observed differences in net mineralization and
nitrification rates among plant communities.
We therefore designed a field experiment to
reveal the reciprocal dependence between the
spatial variation of in situ gross N
transformation rates and understory plant
abundance,
distribution,
and
species
composition, and the influence of individual
beach and oak trees on the same variation.
2. Material and Methods
2.1. Site description
The experiment was conducted in a mixed
beech/oak stand (Fagus sylvatica L. and
Querus robur L.) in Torup in southwestern
Scania, the southernmost province in Sweden
(55°33’N, 13°12’E) on May 5-9, 2003. It had a
relatively varying understory layer of plant
species with different N demand (Table 1). No
bottom layer was present. The soil can be
described as a Dystric Cambisol (FAO system)
with a pH of 4.3.
2.2. 15N pool dilution experiment
Gross N mineralization, immobilization and
nitrification rates were estimated by the pool
dilution/enrichment
technique.
At
108
positions, separated by 1 m and forming a
1×1m grid superimposed on a 10×11 m square
(two positions were excluded due to obstacles),
PVC cylinders (10 cm in length and 11 cm in
diameter) were inserted into the soil after
removal of the litter. Plants present within the
periphery of the cylinders were cut at the soil
surface with a pair of scissors and removed.
Twenty ml of a solution containing 2.4 µg
15
NH4+-N ml-1 (as 15NH4Cl, 98% 15N,
Cambridge Isotopic Laboratories) were added
to each cylinder with a syringe. The needle was
inserted to a depth of five centimetres in at
least 10 positions in the cylinder and slowly
withdrawn to the surface while the 15NH4Cl
solution was injected. Within two hours, two
soil samples were taken from the top 5 cm of
the soil in each cylinder by inserting a 6 ml
scintillation vial into the soil and then carefully
withdrawing
it.
The
two
samples
2
(approximately 10 g dry weight) were
combined in a pre-weighed 50 ml test tube and
put on ice. In the lab, the tubes were weighed a
second time and the soil extracted with 35 ml
1.0 M KCl overnight on an orbital shaker
(Tabulator Teknik AB) at 11 °C. The tubes
were then centrifuged at 1200 rpm for 20 min
on a Hermle Z510, and the supernatant
transferred to a 100 ml serum flask. Two more
soil samples were taken from the cylinders 24 h
after the first sampling, combined, and
extracted as described above.
NH4+ and NO3- were isolated from the soil
KCl extract by using standard IAEA diffusion
procedures (IAEA, 2001) with minor
modifications. Briefly, a standard office paper
punch was used to cut out quartz filter discs
that were then placed on a strip of PTFE tape
and prepared with 10 µl of 2.5 M KHSO4. A
second strip of tape was placed on top of the
filter and the two tape strips were sealed by
pressing the open end of a test tube, in a
rocking circular motion around the filter,
against the tape to create a NH3-trap. The trap
was added to the serum bottle containing the
KCl extract, 0.4 g of MgO was added, the
serum bottle sealed and left at ambient
temperature with periodic shaking by hand.
After five days, the trap was removed and the
filter was placed in a 5×8 mm tin cup and left
to dry in a dessicator. The serum bottle was left
open for 48 hours to release residues of
ammonia. A new trap was added to the bottle
followed by 0.4 g of Devarda’s alloy and 0.4 g
of MgO, and the incubation was repeated.
15
N and 14N in the filters were determined
using continuous flow isotope ratio mass
spectrometry (CF-IRMS). The filters were
oxidized in an ANCA-GSL elemental analyzer
and NOX reduced to N2, which was passed to a
20-20 IRMS (PDZ Europa UK). The amount of
NH4+-N and 15NH4+-N, NO3--N and 15NO3--N
was quantified after subtraction of blank values
and calibration against filter discs which had
received 50, 100, or 150 µg N as glycine
(calibrated against an IAEA KNO3 standard)
dissolved in 5 µl of water. The accuracy of the
quantification was determined by calculating
the average value and confidence interval of
eight filter discs (for each concentration of N),
prepared with glycine as above. At all three N
concentrations, the 95% confidence interval
was within ±3% of the average value. The
precision of the isotopic determinations was
0.2‰.
2.3. Survey of field layer
The abundance of each vascular plant
species in the 90 1x1 m plots was estimated as
the horizontal cover and expressed in percent
as <1%, 1-3%, 3-5%, 5-10%, 10-15%, 15-20%,
20-30%, 30-40%, 40-50%, 50-60%, and 6070%. No species had a higher cover than 70%
in any of the plots. A principal component
analysis (PCA) was made to assess differences
in community composition among plots. The
cover of the different plant species in each of
the 90 plots was used as input data. The
analysis was performed with the MVSP 3.01
statistical package (Kovach computing
services, UK). The effect of dominating species
was reduced by transforming the cover
percentages to classes ranging from 1 (<1%) to
11 (60-70%) and using the software’s
standardize function. The score on the first
principal axis for each plot was recorded and
used in the geostatistical analysis.
An index of nitrogen availability was
calculated for each plot by combining the
functional N index (FNIS, Equation 1, from
Diekmann and Falkengren-Grerup, 1998) and
cover of the species in that plot (equations 2
and 3). The FNIS indices are derived from
detrended correspondence analysis of the
distributions of plant species in hundreds of
deciduous forests in relation to the rates of net
ammonification and nitrification and depend on
the concept that the presence of a particular
species is a function of these processes
(Diekmann and Falkengren-Grerup, 1998).
FNIS=3.40-0.065NH4+ index+0.019NO3- index (1)
A high NH4+ index means that the species
tends to occur at locations with high net
ammonification rates, and a high NO3- index
that it tends to occur at locations with high net
nitrification rates. Two different N availability
indices were calculated for each plot; in the
3
Table 1. Understory plant species found in the experimental plot in the beech/oak forest, their
FNIS values (from Diekmann and Falkengren-Grerup, 1998), the number of 1×1 m squares they
were found in, and the average cover in these squares.
Number of
occurrences
Species
FNIS value
Average cover (%)
22.8
Anemone nemorosa1
1.9
86
2.0
-0.46
1
Carex pilulifera
1.4
2.82
5
Dactylis glomerata
4.5
3.47
3
Deschampsia cespitosa
1.1
2.83
31
Lamiastrumum galeobdolon
1.2
1.71
9
Maianthemum bifolium
2.4
2.86
68
Melica uniflora
1.6
3.44
56
Milium effusum
4.3
2.56
83
Oxalis acetosella
1.4
2.75
12
Poa nemoralis
1.5
1.8
25
Rubus idaeus
2.8
2.65
70
Stellaria holostea
0.8
4.1
6
Stellaria nemorum
1
Since Anemone nemorosa do not have any preference for NH4+ or NO3- it was considered to
belong to FNIS class 5 (on a scale from 1-9). It was given the average FNIS value of the plants in
FNIS class 5, 1.9 (calculated from Diekmann and Falkengren-Grerup, 1998).
first case we assumed that the community
composition and the relative cover would be
affected by the gross N transformation rates
(Weighted Average FNIS):
WA FNIS
(x y + x 2 y 2 + ⋅ ⋅ ⋅ + x n y n )
= 1 1
(x 1 + x 2 + ⋅ ⋅ ⋅ + x n )
(2)
where x1, x2,..., xn are the fractions of the plot
that the species cover (class means), and y1,
y2,..., yn, represent FNIS values of those
species. In the second case, we assumed that
both the community composition and the total
cover would be affected by the gross N
transformation rates (Total Weighted FNIS):
TWFNIS = x 1 y1 + x 2 y 2 + ⋅ ⋅ ⋅ + x n y n
(3)
2.4. Calculations of tree influence
The influence potential of beech and oak on
the gross N mineralization and immobilization
rates was determined according to Saetre
(1999). The influence of an individual tree was
defined to be largest at its base and then
decrease exponentially from it. Large trees
were considered to have a larger influence
potential than small trees and the influence was
additive so that at a spatial location, p, the
combined influence (IP) of all trees was
expressed as:
IP(p) = ∑ DBH k ⋅ exp(-d k )
(4)
k
where DBHk is the diameter at breast height of
tree k and dk is the distance between tree k and
point p. The IP values were calculated at each
node on the 9x10 m grid and used in the
geostatistical analysis. IP at distances beyond 5
m was considered to be negligible. In total, 3
oaks and 5 beeches in and around the
experimental plot were included in the
calculations of IP. The coefficient of influence
(CoI) for each tree species on a certain
dependent variable, Y, could then be calculated
by means of multiple regression (Saetre 1999):
Y = a ⋅ IPbeech + b ⋅ IPoak + c
(5)
where a is the regression coefficient for beech
influence (CoIbeech), b is the regression
coefficient for oak influence (CoIoak), and c is
the
model
intercept.
4
r=
( g t − g )Z t − ( g 0 − g )Z 0
( f − k )W0 ( f t − k )Wt
(k − g )t + 0
−
c
c
f −k
W
× ln t
t
f0 − k
(9)
(8)
(7)
ft − k 

 ln fo − k  W 0 − Wt
×
c = 1 +
Wt 
t

ln

W 0 
p=c=−
(6)
ft − k
ln
f 0 − k W 0 − Wt
p=
×
Wt
t
ln
W0
Equation
mg N kg –1 day -1
atom %
atom %
atom %
mg N kg -1 day -1
mg N kg -1 day -1
Gross NH4+ consumption
N abundance of NH4+
N abundance of NO3-
15
15
Average 15N abundance of NO3Natural 15N abundance
Gross NH4+ production
Gross NO3- production
c
f
g
g
k
p
r
Concentration of 14N plus 15N in NO3-
mg N kg –1
mg N kg –1
Concentration of 14N plus 15N in NH4+
W
Z
day
Time
t
atom %
Unit
Meaning
Symbol
Table 2. The equations used to calculate the gross N mineralization, immobilization, and nitrification rates, and the definition of the symbols in those equations.
5
2.5. Calculation of gross N transformation
rates
The gross N transformation rates were
calculated by the 15N pool dilution/pool
enrichment technique according to Blackburn
(1979), Nishio et al. (1985), and Wessel and
Tietema (1992). The equations are in Table 2.
For the calculations of gross NH4+ production
(mineralization) and consumption rates we
distinguished between two different situations,
where the NH4+ concentration either increased
(Equations 6 and 7) or remained constant
(Equation 8). The NH4+ concentration was said
to be constant if the values at the beginning and
end of the 24-hour period differed by less than
the internal precision of the instrument. This
was determined by calculating the average
value, W , of W0 and Wt and forming the 95%
CI of it. If the NH4+ concentration was outside
the 95% CI, equations 6 and 7 were used, and
if it was inside, equation 8 was used and W was
set to W .
The gross NO3- production (nitrification)
was calculated by equation 9 (Nishio et al.,
1985; Wessel and Tietema, 1992). The validity
of equation 9 was justified by re-calculating it
once with g0 and once with gt substituted for g
( g = (g0 + gt) / 2). An r calculated for g had to
take a value between r’s calculated for go and
gt (Wessel and Tietema, 1992). The gross
NH4+-N immobilization was then calculated by
subtracting the gross NO3- production (r) from
the gross NH4+ consumption (c). These
calculations
assume
that
the
gross
transformation rates remained constant and that
no 15N was recycled to the enriched pool
during the measurement period. The short (24
h) assay period was an effort to meet this
assumption.
2.6. Geostatistics
Variogram model fitting, kriging, and
mapping were performed with the geostatistical
software GS+TM Version 5 (Gamma Design
Software, USA). The active lag distance (the
range over which the semivariance was
calculated) was set to 7.64 m for gross N
mineralization, immobilization, TWFNIS, and
PC1 since the variograms tended to be erratic
at greater lags. However, the active lag distance
was set to 8.64 m for IPbeech and IPoak, 9.64 m
for WAFNIS, and 10.64 m for gross nitrification
since the semivariance did not reach the sill at
shorter distances. The semivariance was
determined for lag classes (maximum distance)
1, 1.42, 2, 2.25, 3.17, 4.25, 5.4, 6.41, 7.63,
8.61, and 9.50 m. The lag classes were chosen
to be separated by approximately 1 m, except at
1-2 m distances, where the number of pairs was
sufficiently high for a finer resolution.
Exponential, spherical or gaussian model
variograms were fitted to the calculated
semivariances using a least squares technique.
Ordinary kriging was used for interpolation of
values between sampling points. The
interpolation estimates were placed on a
uniformly spaced grid with 0.25 m intervals
and used in the different statistical analyses.
3. Results
The N transformation rates alone could not
explain the spatial pattern of the distribution of
different plant species or vice versa. The gross
mineralization and immobilization rates were
spatially autocorrelated within a range of 3.5
and 2.7 m, respectively (Fig. 1, Table 3), and
varied between 0.4 and 52.8 mg NH4+-N kg-1
day-1 (median 13.8), and 0.3 and 34.6 mg
NH4+-N kg-1 day-1 (median 10.9). The
frequency distributions were positively skewed
due to the presence of hotspots with high
mineralization and immobilization rates (Fig.
2). Those hotspots overlapped, indicating a
tight
connection
between
gross
N
mineralization and immobilization rates.
WAFNIS was spatially autocorrelated at a
different and much larger scale (Fig. 1, Table
3), just as the nitrification rate, but its spatial
pattern differed from that of WAFNIS (Fig. 1
and 2, Table 3). On the other hand, TWFNIS
exhibited spatial dependence at a similar scale
as the mineralization and immobilization rates,
with an effective range of 4.2 m (Fig. 1, Table
3). Similarly, analysis of the spatial variation of
the plant community composition by the PCA
6
Fig. 1. Semivariograms for the gross N mineralization, immobilization, and nitrification rates, as well as the different
estimates of the species composition of the understory plant community (WAFNIS, TWFNIS, and PC1). The lines represent
the models fitted to the estimated semivariance (open squares); see Table 5 for parameter descriptions.
revealed that it was autocorrelated within a
spatial scale of <3.7 m (PC1, Fig. 1, Table 3).
The spatial patterns of TWFNIS and PC1
loadings were remarkably similar (Fig. 2;
Spearman Rank Correlation, Rho = 0.92,
p<0.0001), and both were correlated to the total
cover of the understory (Spearman Rank
Correlation, Rho = 0.99 (TWFNIS) and 0.87
(PC1), p<0.0001). Thus, the spatial patterns of
N transformation rates seemed to be more
related to the spatial pattern of the cover of
plants than to the species composition.
WAFNIS was negatively related to the gross
nitrification as well as the gross mineralization
rate (Table 4), indicating that NO3- preferring
species were less common at sites with high
gross mineralization and nitrification rates.
Although there was a slight difference in
regression coefficients, similar results were
produced by the principal component analysis.
7
Table 3. Model parameters for the semivariograms in Figure 1.
Parameter
Model
Nugget variance
(C0)
Range (m)
R2
Gross N mineralization
Exponential
20.9
138.1
C/C0 + C
(%)
84.9
3.5
0.70
Gross N immobilization
Spherical
9.9
74.9
86.8
2.7
0.71
33.0
56.7
12.0
0.89
Gross nitrification
Exponential
14.3
Sill variance
(C0 + C)
PC 1
Spherical
0.0039
0.0193
79.8
3.7
0.82
WAFNIS
Gaussian
0.018
0.237
92.4
30.9
0.99
TWFNIS
Exponential
0.0001
0.1522
99.9
4.2
0.75
The first principal component (PC1), which
explained 59.1% of the floristic variation, was
negatively related to both the gross
mineralization and the gross nitrification rate
(Table 4). If the total instead of the relative
cover was used in the calculation of the N
index (TWFNIS), it was positively related to the
gross N mineralization rate and negatively to
the gross nitrification rate (Table 4).
There was also a weak positive correlation
between the gross mineralization and
nitrification rates (Spearman Rank Correlation,
Rho = 0.187, p<0.0001), which made the
relationships between the plant community
composition and gross mineralization and
nitrification rates hard to interpret. We
therefore introduced the ratio between gross
nitrification and gross mineralization as an
interaction term in the multiple regression
analysis. This nitrification ratio had a higher
explanatory power than the individual rates,
and was positively related to WAFNIS and
negatively to TWFNIS and PC1 (Table 5). It was
also negatively correlated to the gross N
mineralization rate, such that a smaller fraction
of the mineralized N was nitrified at sites with
high mineralization and nitrification rates
(results not shown). The species composition
of the understory therefore seemed to be more
dependent on the relative fraction of the
mineralized N that is nitrified than on the rates
themselves, while the total cover seemed to be
dependent on the gross mineralization rate.
Since the tree influence potential was
defined to decrease exponentially with the
distance from a tree and be negligible at
distances above 5 m, the range of spatial
autocorrelation of N mineralization and
immobilization rates fitted well into the range
of tree influence (Fig. 1, Table 3), as did
variations in the cover and species composition
of the understory (TWFNIS and PC1, Fig. 1,
Table 3). Oak had a positive influence on the
gross mineralization rate, while the opposite
was true for beech (Table 6). Given that the
total cover of the understory was the major
source of variation of TWFNIS and PC1
(Spearman Rank Correlation, Rho = 0.99
(TWFNIS) and 0.87 (PC1), p<0.0001), the
negative influence of beech on them (Table 6)
was probably caused by reduced light input,
resulting in a low cover under the canopy.
Beech trees seemed to create favourable
conditions to NO3- preferring plants, since they
had a positive influence on WAFNIS (Table 6).
This was opposite to the finding that both the
gross nitrification and mineralization rate were
lower when beech influence was high (Table
6), indicating that the gross nitrification rate
was poorly related to NO3- availability.
4. Discussion
A possible explanation to the low
prevalence of NO3- preferring plants, i.e. plants
with high FNIS values, at sites with high gross
nitrification rates may be a lack of correlation
between gross and net rates (Davidson et al.,
1992; Hart et al., 1994; Stark and Hart, 1997;
Verchot et al.; 2001), especially in soils with a
high heterotrophic activity, where high gross
mineralization and immobilization rates
coincide with a tight microbial recycling of
NH4+ and NO3- (Davidson et al., 1992; Hart et
al., 1994; Verchot et al., 2001). Plants are
8
Fig. 2. Krieged contour plots for the influence potential of oak and beech (IPoak and IPbeech), the gross N mineralization,
immobilization, and nitrification rates (mg kg-1 d.w.-1), and the different estimates of the species composition of the
understory plant community (WAFNIS, TWFNIS, and PC1)
9
Table 4. The dependency of the species composition of the understory plant community on the gross N transformation rates.
The gross mineralization was ln-transformed and the gross nitrification (ln+1)-transformed before analysis to ensure normal
distribution of the residuals. Since the input values of the dependent variables were interpolated from spatially autocorrelated
data (n = 1221), a conservative value of p=0.001 was chosen as the significance level.
Dependent
WAFNIS
Independent
Intercept
Gross mineralization
Gross nitrification
Coefficient
2.523
-0.144
-0.043
Significance
<0.0001
<0.0001
<0.0001
Model R2
0.33
TWFNIS
Intercept
Gross mineralization
Gross nitrification
0.317
0.263
-0.194
<0.0001
<0.0001
<0.0001
0.22
PC1
Intercept
Gross mineralization
Gross nitrification
0.366
-0.077
-0.075
<0.0001
<0.0001
<0.0001
0.20
generally considered to be inferior competitors
to heterotrophic microorganisms for N
(Jackson et al., 1989; Schimel et al., 1989, Zak
et al., 1990; Kaye and Hart, 1997), and may
take up less NH4+ and NO3- in soils with high
microbial activity than in soils with low
microbial activity (Paper IV). Similarly,
nitrifiers are inferior competitors to
heterotrophic microorganisms for NH4+ (Zak et
al., 1990; Verhagen and Laanbroek, 1991;
Verhagen et al., 1995), which may explain the
decreased ratio of nitrification to mineralization
with increasing mineralization rates. A larger
fraction of the mineralized N became available
to nitrifiers when mineralization rates and
heterotrophic activity were low, as in Hart et al
(1994) and Zak et al. (1990), and high residual
NO3- concentrations are typically found in soils
with low heterotrophic activity and gross N
mineralization rates (Hart et al., 1994; Zaman
et al., 1999; Bengtsson et al., 2003). Since
FNIS values are based on potential net
ammonification and nitrification rates, plant
species that occur at sites with a large fraction
of inorganic N present as NO3- get a high FNIS
value (Diekmann and Falkengren-Grerup,
1998). This may explain why NO3- preferring
plants were less abundant at sites with high
gross nitrification rates and more dependent on
the relative fraction of the mineralized N that
was nitrified than on the rates themselves.
The positive influence of beech on WAFNIS
indicates that the trees create conditions that
are more favorable to NO3- preferring plants
than to others, e.g. by decreasing the
heterotrophic activity and gross mineralization
rates. This interpretation is also supported by
the observation of Verchot et al. (2001) that
beech stands have higher net nitrification rates
and in situ concentrations of NO3-compared to
oak stands. The question whether trees had a
direct effect on the mineralization rate or
indirect via the understory was not intended to
be resolved by the experimental design.
However, beech litter has a higher lignin
content compared to oak (Sariyildiz and
Anderson, 2003a and 2003b), and the lignin
concentration and lignin/N ratio of the litter has
a negative influence on the decomposition rate
(Sariyildiz and Anderson, 2003a and 2003b)
and net N transformation rates (Ferrari, 1999).
Polyphenolic compounds also have a negative
effect on N mineralization rates (Schweitzer et
al., 2004), probably by forming proteinpolyphenol complexes that are inaccessible to
soil bacteria (Northup, 1995). Thus, the
evidence, so far, are inclined towards a direct
effect of canopy trees on the N mineralization
rate, caused by differences in litter quality.
The positive influence of oak on the gross
N mineralization and immobilization rates may
be a result of oak litter providing carbon to the
soil microorganisms, supporting high N
transformation rates (Watkins and Barraclough,
1996; Andersen and Jensen, 2001; Knops et al.,
2002). In addition, more light penetrates
10
Table 5. The dependency of the species composition of the understory plant community on the gross nitrification and
gross mineralization rates and on the ratio between those rates. The gross mineralization was ln-transformed and the
gross nitrification (ln+1)-transformed before analysis to ensure normal distribution of the residuals. Since the input
values of the dependent variables were interpolated from spatially autocorrelated data (n = 1221), a conservative value
of p = 0.001 was chosen as the significance level.
Dependent
WAFNIS
Independent
Intercept
Gross mineralization
Gross nitrification
Nitrification/Mineralization
Coefficient
2.420
-0.088
-0.022
0.220
Significance
<0.0001
<0.0001
N.S.
<0.0001
Model R2
0.35
TWFNIS
Intercept
Gross mineralization
Gross nitrification
Nitrification/Mineralization
0.707
0.054
0.050
-0.827
<0.0001
N.S.
N.S.
<0.0001
0.25
PC1
Intercept
Gross mineralization
Gross nitrification
Nitrification/Mineralization
0.485
-0.004
-0.016
-0.315
<0.0001
N.S.
N.S.
<0.0001
0.24
through the oak canopy than through the beech
canopy, facilitating a higher cover of vascular
plants under oaks. The “vernal dam
hypothesis” (Muller and Bormann, 1976)
suggests that those plants may act as a
temporary sink for N and be the source of
seasonal variations in N flows (Jandl, 1997;
Tardiff, 1998; Olsson and Falkengren-Grerup
2003). The importance of understory plants in
determining N flows has been questioned on
the grounds that soil microorganisms
immobilize up to an order of magnitude more
N even during spring (Zak et al., 1990;
Rothstein, 2000), and that neither removal of
spring ephemeral plants nor addition of fresh
litter of those plants has any effect on soil NO3concentrations
during
spring
or
N
mineralization in the summer (Rothstein,
2000). The contrasting influence of oak and
beech on the N transformation rates may add
additional uncertainty to the analysis of the tree
influence, since the positive influence of oak
on the gross mineralization rate may follow
from the absence of beech influence in the
vicinity of oaks, and vice versa.
The
two
independent
ways
of
characterizing the spatial variation of the
understory, TWFNIS and PC1, ranked the total
cover of the understory as more variable than
the species composition. This is reasonable
since most of the species have a vegetative
reproduction and individuals may extend their
underground range and appear in many of the
plots with one or more shoots, independent of
plot-to-plot variations of N transformation
rates. Unlike the presence/absence of
aboveground parts, their frequency (cover) and
the gross N transformation rates should covary
from one plot to another if plant available N is
dependent on gross rates of N transformations.
The presence of individuals that cover large
areas would also explain the longer range of
spatial autocorrelation for WAFNIS compared
with TWFNIS, since samples are almost certain
to be spatially autocorrelated at least over the
area that the same individuals cover. The
understory cover therefore seems to give a
better estimation of gross N transformation
rates than the species composition, at least in
sites with a relatively narrow FNIS-range.
Many soil properties, including net N
transformation rates, have been shown to vary
by one to several orders of magnitude at scales
<1-40 m in a number of ecosystems (Robertson
et al., 1988; Smith et al., 1994; Gallardo et al.,
2000). As in this study, more ore less
pronounced “hotspots” are often found,
resulting in positively skewed frequency
11
Table 6. Influence of beech and oak (Equation 5) on the gross N mineralization and immobilization rates, and on the
species composition of the understory plant community (see Material and Methods section for details). The gross N
mineralization and immobilization rates were ln-transformed to ensure normal distribution of the residuals.
Dependent (Y)
Gross mineralization
Beech CoI (a)
-1.54**
Oak CoI (b)
0.82**
Full model (r)
0.30**
Gross immobilization
PC 1
WAFNIS
TWFNIS
0.53
-0.57**
0.77**
-1.57**
0.94**
-0.09**
-0.20**
0.04
0.22**
0.39**
0.47**
0.34**
* significant at the 0.001 level, ** significant at the 0.0001 level
distributions (Robertson et al., 1988; Smith et
al., 1994). Although trees influenced gross
mineralization and immobilization rates, there
was no evidence from the geostatistical
analysis for their contribution to the N
transformation hotspots. Instead, other factors
affecting the microbial activity, such as soil
moisture and small-scale variations in
temperature, probably caused these variations
(Paper II).
The extent of spatial variation in gross N
mineralization and immobilization rates found
here has implications for the interpretation of
experiments such as NITREX and EXMAN,
where nitrogen was added to or removed from
the ambient deposition at eight sites spanning a
gradient of N deposition across Europe (Wright
and van Breemen, 1995; Wright and
Rasmussen, 1998). Unfortunately, none of 36
NITREX studies available in the BIOSIS BIOL
database examined the difference in gross
mineralization or immobilization rates between
treatment and control plots, but only among
sites. The data on net N transformation rates
and NO3- leakage are inconclusive (e.g.
Emmett et al., 1995; Gundersen et al., 1998;
Kjønaas et al., 1998), perhaps because several
samples from each treatment were pooled in
the laboratory incubations, and because sample
independence has been assumed in field
incubations. It is possible that treatment effects
get hidden if samples from treatment plots are
pooled and when ordinary parametric statistics
are used to evaluate the data. Furthermore, the
use of average values may underestimate rates
expressed per area unit if N transformation
rates are positively skewed and appear in
“hotspots”. This emphasizes the importance of
considering the scale of spatial heterogeneity
and dependence when field experiments are
designed (Jonsson and Moen, 1998; Aubry and
Debouzie, 2000).
In conclusion, our results suggest that
variations in N transformation rates not only
explain differences in the floristic variation
between ecosystems and regions (Hutchinson
et al., 1999; Falkengren-Grerup and Diekmann,
2003), but also the spatial variation of the
ground vegetation within ecosystems, even at
scales as small as a few metres. The influence
of oak and beech on the N transformation rates
supports previous observations that individual
trees affect the magnitude and spatial patterns
of different soil properties and processes
(Bruckner et al., 1999; Saetre, 1999; Stoyan et
al., 2000) Influence of trees alone could not
explain the full magnitude of the variation or
the presence of hotspots. These were probably
caused by other factors such as soil moisture
content and small-scale variations in
temperature (Puri and Ashman, 1998; Thomsen
et al., 1999; Subke et al., 2003; Templer et al.,
2003). Taken together, the observations imply
that a better knowledge about the causes of
spatial variations in gross N transformation
rates may hold the key to understand not only
the factors affecting the below ground N
dynamics, but also small-scale variations in the
distribution and composition of above ground
plant biota.
Acknowledgements
This work was sponsored by grants from
the Foundation for Strategic Environmental
Research in Sweden. Niklas Törneman and
12
Therese Nilsson are greatly acknowledged for
their assistance in the field and laboratory.
References
Andersen, M.K., Jensen, L.S. 2001. Low soil
temperature effects on short term gross N
mineralization-immobilization turnover after
incorporation of green manure. Soil Biology
and Biochemistry 33, 511-521.
Aubry, P., Debouzie, D. 2000. Geostatistical
estimation of variance for the spatial mean in
two dimensional systematic sampling.
Ecology 81, 543-553.
Bengtsson, G., Bengtson, P., Månsson, K.F.
2003. Gross nitrogen mineralization-,
immobilization-, and nitrification rates as a
function of soil C/N ratio and microbial
activity. Soil Biology and Biochemistry 35,
143-154.
Blackburn, T.H. 1979. Method for measuring
rates of NH4+ turnover in anoxic sediments,
using a 15N-NH4+ dilution technique.
Applied and Environmental Microbiology
37, 760-765.
Bruckner, A., Kandeler, E., Kampichler, C.
1999. Plot-scale spatial patterns of soil water
content, pH, substrate induced respiration
and N mineralization in a temperate
coniferous forest. Geoderma 93, 207-223.
Chapman, S.K., Hart, S.C., Cobb, N.S.,
Whitham, T.G., Koch, G.W. 2003. Insect
herbivory increases litter quality and
decomposition: An extension of the
acceleration hypothesis. Ecology 84, 28672876.
Chesson, P. 1986. Environmental variation and
the coexistence of species. In: Diamond, J.,
Case, T.J. (Editors). Community Ecology,
Harper & Row, pp. 240–254.
Davidson, E.A., Hart, S.C., Firestone, M.K.
1992. Internal cycling of nitrate in soils of a
mature coniferous forest. Ecology 73, 11481156.
Diekmann, M., Falkengren-Grerup, U. 1998. A
new index for forest vascular plants:
development of functional indices based on
mineralization rates of various forms of soil
nitrogen. Journal of Ecology 86, 269-283.
Dise, N.B., Matzner, E., Gundersen, P. 1998.
Synthesis of nitrogen pools and fluxes from
European forest ecosystems. Water Air and
Soil Pollution 105, 143-154.
Emmett, B.A., Brittain, S.A., Hughes, S.,
Kennedy, V. 1995. Nitrogen additions
(NaNO3 and NH4NO3) at Aber forest, Wales:
II. Response of trees and soil nitrogen
transformations. Forest Ecology and
Management 71, 61-73.
Ettema, C.H., Wardle, D.A. 2002. Spatial soil
ecology. Trends in Ecology and Evolution
17, 177-183.
Falkengren-Grerup, U., Diekmann, M. 2003.
Use of a gradient of N-deposition to
calculate effect-related soil and vegetation
measures in deciduous forests. Forest
Ecology and Management 180, 113-124.
Farley, R.A., Fitter, A.H. 1999. Temporal and
spatial variation in soil resources in a
deciduous woodland. Journal of Ecology,
688-696.
Ferrari, J.B. 1999. Fine-scale patterns of leaf
litterfall and nitrogen cycling in an oldgrowth forest. Canadian Journal of Forest
Research 29, 291-302.
Fromm, H., Winter, K., Filser, J., Hantschel,
R., Beese, F. 1993. The influence of soil type
and cultivation system on the spatial
distributions of the soil fauna and
microorganisms and their interactions.
Geoderma 60, 109-118.
Gallardo, A., Rodríguez-Saucedo, J.J., Covelo,
F., Fernández-Alés, R. 2000. Soil nitrogen
heterogeneity in a Dehesa ecosystem. Plant
and Soil 222, 71-82.
Grundmann, G.L., Debouzie, D. 2000.
Geostatistical analysis of the distribution of
NH4+ and NO2—oxidizing bacteria and
serotypes at the millimeter scale along a soil
transect. FEMS Microbiology Ecology 34,
57-62.
Grundmann, G.L., Dechesne, A., Bartoli, F.,
Flandois, J.P., Chassé, J.L., Kizungu, R.
2001. Spatial modeling of nitrifier
microhabitats in soil. Soil Science Society of
America Journal 65, 1709-1716.
13
Gundersen, P., Emmett, B.A., Kjønaas, O.J.,
Koopmans, C.J., Tietema, A. 1998. Impact
of nitrogen deposition on nitrogen cycling in
forests: a synthesis of NITREX data. Forest
Ecology and Management 101, 37-55.
Hart, S.C., Nason, G.E., Myrold, D.D., Perry,
D.A. 1994. Dynamics of gross nitrogen
transformations in an old-growth forest: the
carbon connection. Ecology 75, 880-891.
Henriksen, T.M, Breland, T.A. 1999.
Evaluation of criteria for describing crop
residue degradability in a model of carbon
and nitrogen turnover in soil. Soil Biology
and Biochemistry 31, 1135-1149.
Hongve, D., van Hees, P.A.W., Lundström,
U.S. 2000. Dissolved components in
precipitation water percolated through forest
litter. European Journal of Soil Science 51,
667-677.
Hook, P.B., Burke, I.C. 2000. Biogeochemistry
in a shortgrass landscape: Control by
topography, soil texture, and microclimate.
Ecology 81, 2686-2703.
Hutchinson, T.F., Boerner, R.E.J., Iverson,
L.R., Sutherland, S., Sutherland, E.K. 1999.
Landscape
patterns
of
understory
composition and richness across a moisture
and nitrogen mineralization gradient in Ohio
(U.S.A.) Quercus forests. Plant Ecology 144,
177-189.
IAEA Training Course Series No. 14. Use of
Isotope and Radiation Methods in Soil and
Water Management and Crop Nutrition,
IAEA, Vienna, 2001.
Jackson, L.E., Schimel, J.P., Firestone, M.K.
1989. Short-term partitioning of ammonium
and
nitrate
between
plants
and
microorganisms in an annual grassland. Soil
Biology and Biochemistry 21, 409-415.
Jackson, R.B., Caldwell, M.M. 1993.
Geostatistical patterns of soil heterogeneity
around individual perennial plants. Journal
of Ecology 81, 683-692.
Jandl, R., Kopeszki, H., Glatzel, G. 1997.
Effect of a dense Allium ursinum (L.) ground
cover on nutrient dynamics and mesofauna
of a Fagus sylvatica (L.) woodland. Plant
and Soil 189, 245-255.
Jonsson, B., Moen, J. 1998. Patterns in species
association in plant communities: the
importance of scale. Journal of Vegetation
Science 9, 327-332.
Kaye, J.P., Hart, S.C. 1997. Competition for
nitrogen
between
plants
and
soil
microorganisms. Trends in Ecology and
Evolution 12, 139-143.
Kjønaas, O.J., Stuanes, A.O., Huse, M.1998.
Effects of weekly nitrogen additions on N
cycling in a coniferous forest catchment,
Gårdsjön, Sweden. Forest Ecology and
Management 101, 227-249.
Klironomos, J.N., Rillig, M.C., Allen M.F.
1999.
Designing
belowground
field
experiments with the help of semi-variance
and power analyses. Applied Soil Ecology
12, 227-238.
Knops, J.M.H., Tilman, D. 2000. Dynamics of
soil carbon and nitrogen accumulation for 61
years after agricultural abandonment.
Ecology 81, 88-98.
Knops, J.M.H., Bradley, K.L., Wedin, D.A.
2002. Mechanisms of plant species impact
on ecosystem nitrogen cycling. Ecology
Letters 5, 454-466.
Kraus, T.E.C., Zasoski, R.J., Dahlgren, R.A.,
Horwath, W.R., Preston, C.M. 2004. Carbon
and nitrogen dynamics in a forest soil
amended with purified tannins from different
plant species. Soil Biology and Biochemistry
36, 309-321.
Lechowicz, M.J., Bell, G. 1991. The ecology
and genetics of fitness in forest plants .2.
Microspatial heterogeneity of the edaphic
environment. Journal of Ecology 79, 687696.
Muller, R.N., Bormann, F.H. 1976. Role of
Erythronium americanum Ker. in energy
flow and nutrient dynamics of a northern
hardwood ecosystem. Science 193, 11261128.
Nishio, T., Kanamori, T., Fujimoto, T. 1985.
Nitrogen transformations in an aerobic soil
as determined by a 15N-NH4+ dilution
technique. Soil Biology and Biochemistry
17, 149-154.
Northup, R.R., Yu, Z.S., Dahlgren, R.A., Vogt,
K.A. 1995. Polyphenol control of nitrogen
release from pine litter. Nature 377, 227-229.
Robertson, G.P., Klingensmith, K.M., Klug,
M.J., Paul, E.A., Crum, J.R., Ellis, B.G.
14
1997. Soil resources, microbial activity, and
primary production across an agricultural
ecosystem. Ecological Applications 7, 158170.
Robertson, G.P., Hutson, M.A., Evans, F.C.,
Tiedje, J.M. 1988. Spatial variability in a
successional plant community: patterns of
nitrogen availability. Ecology 69, 15171524.
Rothstein, D.E. 2000. Spring ephermal herbs
and nitrogen cycling in a northern hardwood
forest: an experimental test of the venal dam
hypothesis. Oecologia 124, 446-453.
Saetre, P. 1999. Spatial patterns of ground
vegetation, soil microbial biomass and
activity in a mixed spruce-birch stand.
Ecography 22, 183-192.
Saetre, P., Bååth, E. 2000. Spatial variation and
patterns of soil microbial community
structure in a mixed spruce-birch stand. Soil
Biology and Biochemistry 32, 909-917.
Sariyildiz, T., Anderson, J.M. 2003a.
Interactions
between
litter
quality,
decomposition and soil fertility: a laboratory
study. Soil Biology and Biochemistry 35,
391-399.
Sariyildiz, T., Anderson, J.M. 2003b.
Decomposition of sun and shade leaves from
three deciduous tree species, as affected by
their chemical composition. Biology and
Fertility of Soils 37, 137-146.
Schimel, J.P, Jackson, L.E., Firestone, M.K.
1989. Spatial and temporal effects on plantmicrobe competition for inorganic nitrogen
in a California annual grassland. Soil
Biology and Biochemistry 21, 1059-1066.
Schlesinger, W.H., Raikes, J.A., Hartley, A.E.,
Cross, A.F. 1996. On the spatial pattern of
soil nutrients in desert ecosystems. Ecology
77, 364-374.
Schume, H., Jost, G., Katzensteiner, K. 2003.
Spatio-temporal analysis of the soil water
content in a mixed Norway spruce (Picea
abies (L.) Karst.)-European beech (Fagus
sylvatica L.) stand. Geoderma 112, 273.287.
Schweitzer,
J.A.,
Bailey,
J.K., Rehill,
B.J., Martinsen, G.D., Hart, S.C., Lindroth,
R.L., Keim, P., Whitham, T.G. 2004.
Genetically based trait in a dominant tree
affects ecosystem processes. Ecology Letters
7, 127-134.
Scott, N.A., Parfitt, R.L., Ross, D.J., Salt, G.J.
1998. Carbon and nitrogen transformations I
New Zealand plantation forest soils from
sites with different N status. Canadian
Journal of Forest Research 28, 967-976.
Smith, J.L., Halvorson, J.J., Bolton Jr, H. 1994.
Spatial relationships of soil microbial
biomass and C and N mineralization in a
semi-arid shrub-steppe ecosystem. Soil
Biology and Biochemistry 26, 1151-1159.
Stoyan, H., De-Polli, H. Böhm, S., Robertson,
G.P., Paul, E.A. 2000. Spatial heterogeneity
of soil respiration and related properties at
the plant scale. Plant and Soil 222, 203-214.
Subke, J.-A., Reichstein, M., Tenhunen, J.D.
2003. Explaining temporal variation in soil
CO2 efflux in a mature spruce forest in
Southern Germany. Soil Biology and
Biochemistry 35, 1467-1483.
Tamm, C.O. Nitrogen in terrestrial ecosystems.
Ecological Studies 81, 1-115.
Tardiff, S.E., Stanford, J.A. 1998. Grizzly bear
digging: effects on subalpine meadow plants
in relation to mineral nitrogen availability.
Ecology 79, 2219-2228.
Templer, P., Findlay, S., Lovett, G. 2003. Soil
microbial
biomass
and
nitrogen
transformations among five tree species of
the Catskill Mountains, New York, USA.
Soil Biology and Biochemistry 35, 607-613.
Thomsen IK, Schjonning P, Jensen B,
Kristensen K, Christensen BT. 1999.
Turnover of organic matter in differently
textured soils - II. Microbial activity as
influenced by soil water regimes. Geoderma
89, 199-218.
Tilman, D., Kareiva, P. (Editors). 1997. Spatial
Ecology. The Role of Space in Population
Dynamics and Interspecific Interactions,
Princeton University Press, Princeton.
Verchot, L.V., Holmes, Z., Mulon, L.,
Groffman, P.M., Lovett, G.M. 2001. Gross
vs net rates of N mineralization and
nitrification as indicators of functional
differences between forest types. Soil
Biology and Biochemistry 33, 1889-1901.
15
Verhagen, F.J.M., Laanbroek, H.J. 1991.
Competition for ammonium between
nitrifying and heterotrophic bacteria in dual
energy-limited chemostats. Applied and
Environmental Microbiology 57, 3255-3263.
Verhagen,
F.J.M.,
Laanbroek,
H.J.,
Woldendorp, J.W. 1995. Competition for
ammonium between plant roots and
nitrifying and heterotrophic bacteria and the
effects of protozoan grazing. Plant and Soil
170, 241-250.
Watkins, N., Barraclough, D. 1996. Gross rates
of N mineralization associated with the
decomposition of plant residues. Soil
Biology and Biochemistry 28, 169-175.
Wessel, W.W., Tietema, A. 1992. Calculating
gross N transformation rates of 15N pool
dilution experiments with acid forest litter:
analytical and numerical approaches. Soil
Biology and Biochemistry 24, 931-941.
Wright, F.R., Rasmussen, L. 1998. Introduction
to the NITREX and EXMAN projects.
Forest Ecology and Management 101, 1-7.
Wright, R.F., van Breemen, N. 1995. The
NITREX project: an introduction. Forest
Ecology and Management 71, 1-5.
Zak, D.R., Groffman, P.M., Pregitzer, K.S.,
Christensen, S., Tiedje, J.M. 1990. The
vernal dam: Plant-microbe competition for
nitrogen in northern hardwood forests.
Ecology 71, 651-656.
Zaman, M., Di, H.J., Cameron, K.C.,
Frampton, C.M. 1999. Gross nitrogen
mineralization and nitrification rates and
their relationship to enzyme activities and
the soil microbial biomass in soils treated
with dairy shed effluent and ammonium
fertilizer at different water potentials.
Biology and Fertility of Soils 29, 178-186.
16
IV
Competition for nitrogen – litter leachate favours microorganisms
at the expense of plants
Katarina F. Månsson, P. Bengtson, G. Bengtsson, U. Falkengren-Grerup
Department of Ecology, Lund University, Sölvegatan 37, Lund SE-223 62, Sweden
Abstract
A green house experiment was designed to test the idea that the C:N ratio of the soil determines if the
heterotrophic soil microorganisms are N or C limited and consequently whether they compete with plants for
N or not. The short-term (24 hours) 15N-uptake uptake by plants and microorganisms in planted and
unplanted soils was determined and bacterial activity was measured by the 3H-thymidine incorporation
technique. Two deciduous forest soils with C:N-ratios of 20 and 31were used and one litter amended soil with
a C:N-ratio of 34. A novel and important part of the experimental design was to use an unplanted control soil
that had been planted (with plants having the roots enclosed in nylon bags in the pots) until 15N was added
and was therefore similar to the planted soil (these plants were also grown in nylon bags until 3 days before
15
N-additions when they were allowed to grow free in the soil). Our data suggest that plants and soil
microorganisms compete for inorganic N but under influence of other factors than the C:N ratio. In contrast
to previous studies, we found differences between planted and unplanted control soils in both microbial 15N
uptake (soil with C:N 20) and bacterial activity (litter amended soil, C:N 34), which emphasizes the
significance of the character of the unplanted control soil. The results also illustrate the importance of
considering the different characteristics of e.g. fungi and bacteria when discussing plant-microbial
competition.
Plant growth can be limited by N even in
relatively fertile soils, with high N turn-over
rates, (Tamm, 1991), while carbon availability
and decomposability is often considered to be
the major factor limiting microbial growth and
N assimilation (Tate, 1995). However, as
discussed by e.g. Kaye and Hart (1997), this
generalisation depends on the C:N ratio of the
substrate
used
by
the
heterotrophic
microorganisms. Fungi and bacteria are
considered to be N limited at substrate C:N
ratios above 30, but the ratio varies in the
literature between 13 and 30 depending on the
carbon use efficiency and the biomass C:N
ratio of the microorganisms, (e.g. Kaye and
Hart, 1997; Killham, 1994). Thus, C-rich
substrates with C:N ratios exceeding 30 are
likely to be N-limited to microorganisms and
result in a net microbial immobilization of N
and potential competition between plants and
microorganisms. The main aim of the present
study was to test the importance of the C:N
ratio of the soil for competition for N between
plants and soil microorganisms in two forest
soils that were similar in pH, dominating tree
species, soil type and climate.
1. Introduction
For a long time the superiority of plants or
soil microorganisms in competition for
nitrogen (N) has been debated, and the
heterotrophic microorganisms have often been
considered as winners (Jackson et.al., 1989;
Schimel et al., 1989; Zak et al., 1990).
Unequivocal demonstration of competition is,
however, difficult if competition is defined as
the simultaneous demand by two or more
organisms for a common limited resource so
that both are inhibited by the interaction
(Tilman, 1982). Whereas the first part of the
definition calls for reasonably straightforward
experiments, the second part may be a
challenge for very asymmetric interactions in
which the dominant competitor may be so little
affected by the subordinate that negative
effects cannot be detected (Keddy, 1989).
Uptake of 15N labelled inorganic nitrogen was
measured to test the potential competition
between plants and microorganisms. If the
microorganisms were limited by and
competing for N with plants the microbial 15N
uptake and activity should increase when the
plants were removed.
1
show that microorganisms take up a greater
part of added 15NH4+ and 15NO3- than plants
(Jackson et al., 1989; Schimel et al., 1989; Zak
et al., 1990; Norton and Firestone, 1996).
However, this do not prove plant-microbial
competition for N and so far increased
metabolic activity and N uptake in
microorganisms in absence of plants has not
been demonstrated.
Plants are the major source of soil organic
matter and the high C:N ratios of plant litter,
often above 30, suggests that litter
decomposition is N limited and a potential
source of plant-microbial competition. Indeed,
microorganisms are known to immobilise large
amounts of inorganic N from surrounding soil
during early stages of plant litter
decomposition (Downs et al., 1996; Bremer
and Kuikman, 1997; Jingguo and Bakken,
1997). Adding fresh litter with high C content
to the soil may temporarily increase amount of
C with high turn-over rate and C:N ratio of that
organic matter fraction and shift the balance
between N mineralization/immobilization
towards immobilization. To test the importance
of C quality in relation to soil C:N ratio we
used a soil-litter mixture with the same Ccontent and similar C:N ratio and pH as a
natural soil with high C:N ratio.
Heterotrophic microorganisms are known
to use both NH4+ and NO3-, with a preference
for NH4+ (Jackson et al., 1989; Schimel et al.,
1989; Zak et al., 1990; Norton and Firestone,
1996). However, recent studies have shown
that soil microorganisms can also immobilise
NO3- when N is limiting (Zak et al., 1990;
Davidson et al. 1992; Stark and Hart, 1997).
Since the assimilative reduction of NO3- to
NH4+ requires energy, which is often limiting
to heterotrophic microorganisms (Tate, 1995),
and since it is repressed by NH4+ (Paul and
Clark, 1996), microbial assimilation of NO3- is
thought to occur in places with NH4+ deficit
(Schimel et al., 1989). Plants seem to grow best
on a combination of NH4+ and NO3(Falkengren-Grerup
and
LakkenborgKristensen, 1994), but various species have
different preferences both within and between
habitats (Högbom and Ohlson, 1991;
Falkengren-Grerup, 1995; Falkengren-Grerup
et al., in press). Thus, plants and soil
microorganisms are able to use the same forms
of N and may compete for them as long as N is
the growth limiting resource.
Short-term (hours to a few days)
experiments are most appropriate to
demonstrate specific plant-microbe competition
events because of the short turn-over times of
soil microorganisms. Most of these studies
We hypothesise that
(1) soil micro-organisms and plants
compete for NH4+ and NO3-, at higher
soil C:N ratios but not at lower because
theoretically, N is more likely to be a
limiting element for heterotrophic
microorganisms at high soil C:N ratios,
(2) microorganisms take up more NH4+
and NO3- than plants in all soils but the
magnitude of the difference will be
inversely related to the soil C:N ratio,
(3) the difference between microbial and
plant uptake will be greater for NH4+
than NO3- since NO3- is more readily
mobile in soil and microorganisms seem
to prefer NH4+ over NO3-.
2. Materials and Methods
2.1. Soil description
The two soils with high (31) and low (20)
C:N ratios, respectively, were classified as
Dystric Cambisol (FAO system) and located
500 m apart at Torup, southern Sweden
(55°33,N, 13°37,E). Oak and beech were the
dominating tree species and the understory
vegetation was sparse. At the end of October
2002 litter was removed within a 20 by 20 m
square and five samples of the top five cm of
the soil were collected, sieved through a 4 mm
mesh and then analysed to determine the C:N
ratio. The heterogeneity of the soil limited us to
pool only two of the five samples from each
site to get appropriate C:N. After determining
pH and water holding capacity (WHC), the soil
was stored at 2ºC until planting two weeks
after sampling. The soil-litter mixture (20L)
was produced by mixing the oak-beech litter
2
weeks after planting and the soils were kept at
60% of WHC by addition of deionised water.
Pots used for the treatment with plants had the
grasses replanted without nylon bags three days
before 15N additions so that the roots were free
to explore the soil. In the treatment without
plants the grasses were kept in the root bags
until 1 hour before 15N additions. In this way
unplanted and planted soils were similar during
the competition assay. When replanting, the
plants with root bags were removed with
minimal disturbance to the soil and the plants
were carefully taken out of the bags. The soil in
the root bags were emptied in the hole left from
the bag and the grass were replanted in the
same hole.
The pots received either distilled water
(control), or distilled water spiked with 50
nmol 15N per g dw soil 15NH4Cl or Na15NO3
(Cambridge Isotopic Laboratories, >98% 15N).
The 15N additions resulted in < 5% of added
15
NH4+ to intrinsic NH4+ (extractable amounts
in 0.4M KCl measured by flow injection
analysis at the start of the experiment) in the
two natural soils and 52% in the soil-litter
mixture, and 82 and 13 % of added 15NO3- to
intrinsic NO3- in soil 20 and 31 respectively.
The 15NO3- addition to the soil-litter mixture
was 100 times the intrinsic. Four 0.5 mL
injections per pot were made with a syringe,
which was gradually withdrawn from the
bottom to the top of the pot to distribute the
solution evenly. The soils were dried by ~2 g
(this was achieved by avoid watering the day
before 15N additions) before injections to avoid
over-watering. One hour after injection, half of
the pots were harvested to determine the
concentration of extractable 14N/15N and NO3and NH4+. Plants were gently removed, the soil
was thoroughly mixed and 10 g was extracted
with 50 ml of 0.4 M KCl for one hour by
shaking. The extracts were filtered through acid
washed Munktell analytical filter papers (00K).
The extracts were then analysed by flow
injection analysis (FIAstar, Tecator ASN 5005/90, 110-01/92) to determine the NH4+ and
NO3- concentrations. Five mL of the extracts
were vacuum centrifuged without heat in a
Savant AES 1000 and the residue analysed to
determine 15N and total N concentration. As the
fragments (≤ 1 mm, C:N ratio 42) with the soil
with C:N ratio 20 to obtain a soil with the same
C content and a similar C:N ratio as the soil
with C:N ratio 31 (0.21 litter g-1 dw soil). The
litter was collected at the site with low C:N
ratio and at the same time as the soils.
Properties of the three soil materials are given
in Table 1.
2.2. Plant material
Festuca gigantea (L) was used in the
experiment because it is relatively common in
the type of oak-beech forest where the soils
were sampled. F. gigantea has a rather high N
demand and NO3- uptake with FNIS value of 8
(Diekmann
and
Falkengren-Grerup,
unpublished) and Ellenberg N-value of 6
(Ellenberg, 1991). The plants were grown from
seeds in > 99% pure silica sand supplied with a
nutrient solution (mineral N concentration was
150 µM NO3- and 100 µM NH4+) for one to
two weeks after germination and then planted
in the experimental soils. Seedlings that
qualified for the experiment had a shoot height
of 7-10 cm and root length of 5- 8 cm.
2.3. Experimental design and 15N additions
The soils were wetted with deionized water
to obtain 60% WHC before planting.
Transparent plastic pots with a diameter of 27
mm and a height of 85 mm were wrapped in
tin-foil to exclude sunlight from the soil core.
They were filled with soil from the samples
with low and high C:N ratios and with the soillitter mixture to obtain a weight of 40, 35 and
30 g fw soil respectively to get the same
volume. Each pot was planted with three
specimens of F. gigantea. The roots of each
plant were enclosed in nylon bags (width 15
mm, height 50 mm and mesh size 25 µm) to
keep the roots out of the bulk soil but allow N
uptake from the surrounding soil. The pots
were placed in a greenhouse with 16°C
night/20°C day temperature and with 16 hours
of day light radiation (with additional light of
160 µmol m-2 s-1). Sprinklers adjusted the
relative air humidity to about 50%. The pots
were used in the 15N uptake experiment three
3
Table 1. The pH (0.2 M KCl), soil-C and -N (mg g-1 dw soil), soil C:N ratio, microbial biomass N (µg N g-1 dw
soil).
Soil
pH
Soil C
Soil N
Soil C:N
Microbial biomass N
20
31
20L
3.5
3.8
3.6
57
164
164
2.9
5.2
4.0
20
31
34
78.5
147.6
82.0
50, or 100 µg N as glycine (calibrated against
and IAEA KNO3 standard) dissolved in 5 µl of
water. The precision of the isotopic
determinations was 0.2‰. The plant and
microbial uptake of NH4+ or NO3- was
estimated by multiplying the fraction of 15N
immobilised by plants and microorganisms
(data from fumigation-extraction) by the
14
N/15N ratio in the NH4+ and NO3- pools one
hour after 15N additions, assuming that the
uptake of 15N and 14N was proportional to their
relative abundance:
pH of the soil water extracts was below 4.5,
loss of ammonia through volatization was <
2% (data not shown). The 15N, NO3- and NH4+
concentrations one hour after 15N addition were
then used as starting values for the calculation
of the N-uptake. The remaining pots were
placed for 24 h in a climate chamber
(20°C/15°C day-night temperature) with 16
hours per day of radiation of 400 µmol m-2 s-1
and an air humidity of 50%. When harvested,
the plants were gently removed from the soil
and rinsed in 1.0 mM KCl and 0.5 mM CaCl2
for 5 minutes to remove residual label. Soil
samples were taken for mass spectrometry
analysis. Samples of 10 g of soil where taken
for chloroform fumigation-extraction (Brookes
et al., 1985; our samples was extracted by 0.4
M KCl instead of 0.5 M K2SO4) to determine
the microbial 15N uptake and biomass-N. Plants
were dried at 70°C for 24 hours. Five mL of
the KCl soil extracts from fumigated and
unfumigated samples were vacuum centrifuged
without heat in a Savant AES 1000 to remove
water and the residue analysed to determine
15
N and total N concentration. Dried plant
material was weighed and ground to a fine
powder using a ball mill (Retsch, Mixer Mill
200), while salt from dried soil extracts were
ground by using a mortar and pestle.
Ni=(15Nb·Nt0/15Nt0)
where Ni is the gross immobilization of NH4+
or NO3-, Nb is the 15N taken up by the
microorganisms or the plants and Nt0/15Nt0 is
the ratio of total N and the added 15N in the
NH4+- and NO3--pools 1 hour after 15N
additions. The Nb values are calculated by
subtracting the natural abundance of 15N from
the data we received from the 15N treatments.
2.5. 3H-thymidine incorporation
The 3H-thymidine incorporation technique
(Bååth et al., 2001) estimates the bacterial
growth rate as the rate of DNA synthesis. A 50
mL centrifuge tube containing two g of soil and
30 ml of distilled water was put in a shaker for
15 minutes and then centrifuged at 3000 rpm.
Methyl[3H]thymidine (Amersham, 25 Ci
mmol-1, 925 GBq mmol-1) and 1.5 mL bacterial
suspension (the supernatant) were added to 2
mL eppendorf-vials that were incubated at
20ºC for 2 hours (final 3H-thymidine
concentration of 100nM). Incorporation was
stopped by adding 75µL cold 100%
trichloroacetic acid (TCA) to each vial and the
vials were shaken on a vortex and then stored
in the refrigerator for 30 minutes. The vials
2.4. Analysis of 15N by mass spectrometry
Plant tissues, soil and salt from dried soil
extracts were weighed in tin capsules and
analysed for 15N by continuous flow isotope
mass spectrometry. The samples were oxidized
in an ANCA-GSL elemental analyzer, NOX
reduced to N2, which was passed to a 20-20
IRMS (PDZ Europa UK). The amount of 15N
and total N were quantified after subtraction of
blank values and calibration against tin
capsules with absorbent which had received 5,
4
plants and microorganisms. The plant NH4+
uptake was similar in the two natural soils but
significantly lower in the soil-litter mixture
(ANOVA; p<0.05), while there was no
difference in uptake of NO3- (Fig 1).
The plants took up 1.3 – 2 times more NH4+
and NO3- than the microorganisms in the two
natural soils, while microorganisms took up an
order of magnitude more N in the litter
amended soil (Table 2). The ratio between
microbial and plant N uptake increased with
increasing C:N ratio, especially in comparison
between the natural soils and the litter amended
and the ratio was lower for NO3- than for NH4+
(Table 2).
were then centrifuged at 15 000 g for 10
minutes and the radioactive pellets were
washed with 1.5 mL cold 5% TCA (to
precipitate macro molecules) and 1.5 mL 80%
ice cold ethanol (to remove thymidine bound
lipids). Then 0.2 mL 1 M NaOH was added
and heated to 90ºC for 1 hour to solubilise the
macromolecules. After cooling, 1.25 mL
scintillation cocktail (Ultima Gold, Packard)
was added and the radioactivity was measured
on a Beckman LS 1801 liquid scintillation
spectrometer.
2.6. Statistical analyses
The effect of plants on bacterial activity
(3H-thymidine incorporation rate) was tested in
a 3-way ANOVA showing a significant effect
of plants, and it was followed by multiple
comparisons (LSD test) testing all treatment
separately. The effect of plants on microbial
NH4+ and NO3- uptake was tested using a 3way ANOVA followed by multiple
comparisons (LSD test) testing all treatment
separately. Plant 15NH4+ and 15NO3- uptake was
analysed in a 2-way ANOVA followed by
multiple comparisons testing all treatments
separately. The relationship between bacterial
activity (3H-thymidine incorporation) and
microbial N uptake was tested for all soils
together in an ANCOVA with 15N-form and
with-without plant as fixed factors. All
statistical analyses were performed in SPSS
10.0.5 and Statistica 6.1).
3.2. Bacterial activity
The bacterial activity (3H-thymidine
incorporation rates measured in DPM) and the
microbial N uptake were positively related
across the soils (ANCOVA; p< 0.001,
R2=0.764). The ratio of the activity to 15N
uptake was calculated with the plant uptake
included or excluded in order to get an
additional estimate of the effect of plant 15N
uptake on microbial uptake. In the litter
amended soil the ratio of the activity to
microbial 15N uptake was higher in the planted
soil (9.68⋅106 and 17.7⋅106 for NH4+ and NO3respectively) than in the unplanted (8.42⋅106
and 16.6⋅106 for NH4+ and NO3- respectively).
When plant uptake of 15N was included in the
ratio (8.70⋅106 and 16.0⋅106 for NH4+ and NO3respectively) the planted and unplanted
treatments got similar ratios.
The activity was significantly lower in soil
20 compared to the other soils on a soil dry
weight basis (Fig. 2, ANOVA; p<0.001, n=5),
but significantly higher on a soil C basis
(ANOVA; p<0.001, n=5), implying a higher C
quality in soil 20.
The plants had an over-all positive effect on
the bacterial activity (ANOVA p<0.01, n=5),
but multiple comparisons between treatments
showed that the significance was attributed to
the litter amended soil (Fig. 2). Plants seemed
to stimulate the bacterial activity in the natural
soils as well, especially soil 31 (p=0.068, n=5),
but only when 15NH4+ was added. The impact
3. Result
3.1. Microbial and plant N uptake
The presence of the plant reduced microbial
uptake of NH4+ and NO3- (ANOVA; p<0.05,
n=5) in the soil with the lowest C:N ratio (20)
but had no significant effect in the other two
soils (Fig. 1). The plant N uptake was
significantly reduced by the litter amendment
of the soil with low C:N ratio (20L) but
independent of the C:N ratio of the natural
soils. The NH4+ uptake was about one order of
magnitude larger than the NO3- uptake in both
5
slightly higher (ANOVA; p=0.118, n=15) and
the NO3- concentrations lower (p<0.05, n=15)
in the planted soils than in the unplanted (Table
3).
3.4. Soil effects on F. gigantean
The root dry weight was similar in the soils
(ANOVA; p=0.985, n=20), with means ranging
from 19.07 to 19.39 mg, while the shoot dry
weight varied with the soil type. Soil 20 and
20L had the lowest shoot weights, 26.44 and
23.64 mg, respectively, and soil 31 the highest
of 35.03 mg, significantly higher than the other
soils (ANOVA; p<0.01, n=20). The shoots in
soil 20 were short and dark green while the
shoots in soil 20L showed signs of chlorosis.
After NH4NO3 (six times 5 ml 500µM) was
added to the litter amended soil the plants
shifted in colour toward darker green. The
plants in soil 31 showed no signs of nutrient
deficiency.
Figure 1. Uptake of (a) NH4+- and (b) NO3-- (µg N per
pot) by plants and soils microorganisms in the three
soils, for the planted and unplanted treatments. The bars
represent means and standard error (n=5).
4. Discussion
The data on the uptake of NH4+ and NO3by microorganisms and plants (Fig. 1) in the
planted and unplanted pots can be used to
reject the first hypothesis, that competition for
N is expected in soils with high rather than low
C:N ratios. The microorganisms were less
successful in acquiring inorganic N from soil
20 in presence of active plant roots than in their
absence but unaffected by the plants in the soils
with the higher C:N ratio. However, only 50%
of the decreased microbial NH4+ uptake in soil
20 could be explained by plant uptake
indicating that it may be other factors limiting
microbial uptake. One of them may be plant
control of microbial phosphorous (P) uptake,
subsequently limiting the N uptake. Soil 20 had
a high fraction of mineral soil and a low pH
(3.5 in KCl), at which inorganic P tend to be
fixed in unavailable forms that involves Al-,
Fe- or Mn-ions, oxides or hydrous oxides
(Schlesinger, 1991; Gallardo and Schlesinger,
1994). Indeed, the plants in soil 20 did show
symptoms of P limitation, such as short shoots
with a dark-green colour. Phosphate supply has
of the plants in the litter amended soil was
probably associated with the root morphology.
The roots were thinner, more branched, and
with a higher frequency of rot hairs than in the
other soils (observed during harvest). That
would enhance the surface to volume ratio and
most
likely
increase
rhizodeposition,
stimulating rhizosphere bacteria.
3.3. Ammonium and nitrate concentrations
The NH4+ and NO3- concentrations in the
natural soils increased by a factor two or
slightly more during the 24 hour assay (Table
3). The effect may be related to the enclosure
of roots in nylon bags, indicating that the soil
microorganisms were not N limited, at least not
in the bulk soil, or to the exclusion of
ectomycorrhiza.
Soil 20 and 31 had significantly higher
NH4+ concentrations than soil 20L and the
NO3- concentration was significantly higher in
soil 31 compared to the other two soils (Fig. 3
and Table 3). The NH4+ concentrations were
6
Table 2. Microbial and plant uptake of 15N in percent of added 15N.
15
Soils
N-form
Ratio
Microbial 15N uptake (%)
Plant 15N uptake
15
Micr:Plant N-uptake
(%)
Planted
Unplanted
20
NO30
10.03
0
9.73
NH4+
0.42
9.02
3.27
23.7
31
NO3NH4+
0.49
0.74
5.58
8.90
2.74
8.82
6.06
10.09
20L
NO3NH4+
9.5
13.4
2.57
4.59
24.29
40.10
19.21
38.5
The data on the microbial relative plant
uptake of 15N labelled NH4+ and NO3- seem to
be more in favour of acceptance than rejection
of the second hypothesis suggesting that plants
will be more efficient at low C:N ratios and
microorganisms at high. Whereas the plant
uptake was inversely related to the C:N ratio,
the microorganisms increased the uptake of
added ions in the planted soil as the C:N ratio
became larger (Table 2). One is given the
impression that the plants acquire a higher
proportion of added N in soil 20 and that
microorganisms acquire a higher proportion in
soil 20L, as expressed by the uptake ratio in
Table 2. The difference between plant and
microbial N uptake was largest in soil 20L, and
the low plant 15N uptake compared to
microbial, despite the apparent N deficiency in
the plants, suggests a one-sided competitive
interaction were the microorganisms were
superior.
Apart from soil 20, the bacterial activity
was the main factor determining microbial N
uptake, while plants had no effect. Plants
stimulated the bacterial activity only when
fragmented litter had been added to soil 20,
resulting in low inorganic N concentrations. A
fungal fraction in soil 20L as well as in soil 20
may explain both the discrepancies between
microbial N uptake and bacterial activity in
response to plants in soil 20, and the higher
microbial N uptake in soil 20L compared to
soil 31 despite the similar bacterial activity. It
is possible that the structural litter components
in soil 20L promoted N uptake by fungi
decomposing these compounds, adding to the
N taken up by bacteria growing on litter
leachates and plant exudates (Suberkropp and
also been shown to control nitrification in
different forest soils (Pastor et al., 1984), which
may explain the low nitrate concentrations in
soil 20 despite high NH4+ supply.
However, the plants had no significant
effect on the bacterial activity in soil 20,
suggesting that they specifically inhibited
fungal N uptake and not bacterial uptake. It
seems reasonable to assume that bacteria with
their higher surface to volume ratio and high
growth rates will be superior to the plants at
taking up NH4+ at low concentrations.
However, mycelia forming fungi may compete
with plants on more equal terms than bacteria
because of their lower turn-over rates and
higher surface to volume ratios as compared to
bacteria and unicellular fast-growing fungi.
One of the key assumptions of the first
hypothesis was the limited availability of
inorganic N that follows with a high C:N ratio.
It is true that the NH4+ concentration was lower
in soil 31 than in soil 20, but the NO3concentration was more than 5 times as high
(Fig. 3, Table 3). However, NH4+ and NO3concentrations and plant uptake of NH4+ were
significantly lower in soil 20L than in the
natural soils, suggesting the litter amended soil
as the best candidate to fulfil the assumption on
N limitation in the experiment. The plants in
this soil had characteristics of nutrient
limitations, such as development of a high
frequency of finer roots (Badalucco &
Kuikman, 2001) and signs of chlorosis in the
shoots, which was reversed when the soil was
fertilised with NH4NO3. The insensitivity of
microbial N uptake to the presence of plant
roots in that soil (Fig. 1) is a strong argument
to reject the first hypothesis.
7
Figure 3. The (a) NH4+ and (b) NO3- concentrations in
the three soils, for the planted and unplanted treatments.
Bars represent means and standard error (n=5).
Figure 2. The bacterial activity measured by 3Hthymidine incorporation [Disintegration’s Per Minute
(DPM) g-1 dw soil] in the three soils, for the planted and
unplanted treatments. Bars represent means and standard
error (n=5). (ANOVA; ns=p>0.05, * p<0.05, ** p<0.01,
*** p<0,001)
the increase in bacterial activity in planted
soils, an increase that most likely is a response
to high C:N rhizodeposition which is difficult
to explain in a soil that in many respects seem
to be N limited. That is, bacterial acquisition of
N may be efficient enough for bacterial growth
not to be N limited, while plants and fungi are
potentially N limited and compete for this
nutrient. An alternative explanation could be
that N limited microorganisms are stimulated
by N containing compounds, such as, NH4+,
NO3-, amino acids, vitamins, enzymes and
nucleotides
(Uren,
2001)
in
the
rhizodeposition.. The extra N input from roots,
though small in relation to C, could be
important to N limited microorganisms
(Brimecombe et al., 2001).
The first assumption in the second
hypothesis, suggesting that microbial N uptake
would be higher than plant N uptake in all
soils, had to be rejected since the plant uptake
of NH4+ and NO3- exceeded the uptake by
microorganisms, except in soil 20L. This was
unexpected since microbial uptake that was 2
to 10 times as high as plant uptake in previous
short-term competition studies (Jackson et al.,
1989; Schimel et al., 1989; Zak et al., 1990).
Klug, 1976; Swift et al., 1979; Henriksen and
Breland, 1999). It is interesting to notice that
the microorganisms had a higher ratio of
activity to 15N uptake in the planted 20L soil
than the unplanted. With everything else in the
planted and unplanted soil being equal, the
difference should be attributable to the
presence of plants, that is, an expression of
competition. Indeed, if the number for the 15N
uptake by the plants is added to the number for
the 15N uptake by the microorganisms in the
planted soil, the ratios of microbial activity to
15
N uptake in the planted and unplanted soils
become similar both for NH4+ and for NO3-.
Thus, the bacterial activity was positively
affected by plants but the microbial uptake of
NH4+ and NO3- did not increase in proportion
to the activity, which might be explained by
competition between plants and fungi for N
while bacterial N uptake was relatively
unaffected. This explanation could account for
8
Nannipieri, P. (Editors). The Rhizosphere –
Biochemistry and organic substances at the
soil-plant interface. Marcel Dekker, Inc.,
New York.
Bremer, E., Kuikman, P. 1997. Influence of
competition for nitrogen in soil on net
mineralization of nitrogen. Plant and Soil
190, 119-126.
Brimecombe, M.J., De Leij, F.A., Lynch, J.M.
2001. The effect of root exudates on
rhizosphere microbial populations. In:
Pinton, R., Varanini, Z., Nannipieri, P.
(Editors). The Rhizosphere – Biochemistry
and organic substances at the soil-plant
interface. Marcel Dekker, Inc., New York.
Brookes, P.C., Landman, A., Pruden, G.,
Jenkinson, D.S. 1985. Cloroform fumigation
and the release of soil nitrogen: A rapid
direct extraction method to measure
microbial biomass nitrogen in soil. Soil
Biology and Biochemistry 17, 837-842.
Bååth, E., Pettersson, M., Söderberg, K.H.
2001. Adaptation of a rapid and economical
microcentrifugation method to measure
thymidine and leucine incorporation by soil
bacteria. Soil Biology and Biochemistry 33,
1571-1574.
Davidson, E.A., Hart, S.C., Firestone, M.K.
1992. Internal cycling of nitrate in soils of a
mature coniferous forest. Ecology 73, 11481156.
Downs, M.R., Nadelhoffer, K.J., Melillo, J.M.,
Aber, J.D. 1996. Immobilization of a 15Nlabeled nitrate addition by decomposing
forest litter. Oecologia 105, 141-150.
Ellenberg, H., Weber, H.E., Düll, R., Wirth, V.,
Werner,
W.,
Paulißen,
D.
1991.
Zeigerwertevon Pflanzen in Mitteleuropa.
Scripta Geobotanica 18, 1-248.
Falkengren-Grerup,
U.,
LakkenborgKristensen, H. 1994. Importance of
ammonium and nitrate to the performance of
herb layer species from deciduous forests in
southern Sweden. Environmental and
Experimental Botany 34, 31-38.
Falkengren-Grerup, U. 1995. Interspecies
differences in the preference of ammonium
and nitrate in vascular plants. Oecologia 102,
305-311.
Norton and Firestone (1996) compared
microbial and plant (Pinus ponderosa seedlings
with ectomycorrhiza) uptake of NH4+ and NO3in a sandy loam soil from a mixed conifer site.
They found that plants accounted for 30% of
the total NH4+ consumption and 70% of the
NO3- consumption, which is more consistent
with our results. The studies that reported a
higher microbial to plant 15N uptake were field
studies (Jackson et al., 1989; Schimel et al.,
1989; Zak et al., 1990) or laboratory
experiment (Norton and Firestone, 1996) where
mycorrhizal N uptake probably was included in
the microbial uptake. There was no
ectomycorrhiza present in our experiment and
the arbuscular mycorrhizal N uptake is not
likely to have been significant because the
roots were enclosed in root bags, thus giving a
possible explanation to the lower microbial N
uptake in the natural soils. The third hypothesis
was supported by the ratio of microbial to plant
uptake, which was lower for NO3- than for
NH4+This was expected from previous studies
(Jackson et al., 1989; Schimel et al., 1989; Zak
et al., 1990; Norton and Firestone, 1996) and
from the higher diffusion rates of NO3compared to NH4+, which would favour plant
roots with their more limited cover of the soil
volume.
Our data suggest that plants and soil
microorganisms compete for inorganic N but
under influence of other factors than the C:N
ratio. In contrast to previous studies (Schimel
et al., 1989; Zak et al. 1990; Norton and
Firestone, 1996), we found differences between
planted and unplanted control soils in both
microbial 15N uptake (soil with C:N 20) and
bacterial activity (litter amended soil, C:N 34),
which emphasizes the significance of the
character of the unplanted control soil. The
results also illustrate the importance of
considering the different characteristics of e.g.
fungi and bacteria when discussing plantmicrobial competition.
References
Badalucco, L., Kuikman, P.J. 2001.
Mineralization and Immobilization in the
rhizosphere. In: Pinton, R., Varanini, Z.,
9
Suberkropp, K., Klug, M.J. 1976. Fungi and
bacteria associated with leaves during
processing in a woodland stream. Ecology
57, 707-719.
Stark, J.M., Hart, S.C. 1997. High rates of
nitrification and nitrate turnover in
undisturbed coniferous forests. Nature 385,
61-64.
Swift, M.J., Heal, O.W., Anderson, J.M. 1979.
Decomposition in terrestrial ecosystems.
Blackwell, Oxford.
Tamm, C.O. 1991. Nitrogen in terrestrial
ecosystems. Ecological Studies 81, 1-115.
Tate, R.L. 1995, Soil Microbiology. John
Wiley and Sons Inc, New York.
Tilman, D. 1982. Resource competition and
community structure. Princeton University
Press, Princeton.
Uren, N.C. 2001. Types, amounts, and possible
functions of compounds released into the
rhizosphere by soil-grown plants. In: Pinton,
R., Varanini, Z., Nannipieri, P. (Editors) The
Rhizosphere – Biochemistry and organic
substances at the soil-plant interface. Marcel
Dekker, Inc., New York.
Zak, D.R., Groffman, P.M., Pregitzer, K.S.,
Christensen, S., Tiedje, J.M. 1990. The
vernal dam: Plant-microbe competition for
nitrogen in northern hardwood forests.
Ecology 71, 651-656.
Gallardo, A., Schlesinger, W.H. 1994, Factors
limiting microbial biomass in the mineral
soil and forest floor of a warm-temperate
forest. Soil Biology and Biochemistry 26,
1409-1415.
Henriksen, T.M., Breland, T.A. 1999. Nitrogen
availability effects on carbon mineralization,
fungal and bacterial growth, and enzyme
activities during decomposition of wheat
straw in soil. Soil Biology and Biochemistry
31, 1121-1134.
Högbom, L., Ohlson, M. 1991. Nitrate
assimilation in coexisting vascular plants and
swamp forest habitats in central Sweden.
Oecologia 87, 495-499.
Jingguo, W., Bakken, L.R. 1997. Competition
for nitrogen during mineralization of plant
residues in soil: Microbial response to C and
N
availability.
Soil
Biology
and
Biochemistry 29, 163-170.
Kaye, J.P., Hart, S.C. 1997, Competition for
nitrogen
between
plants
and
soil
microorganisms. Trends in Evolution and
Ecology 12, 139-143.
Keddy, P.A. 1989. In: Rosenzweig, M.L.,
Kitching, R.L. (Editors). Competition.
Chapman and Hall, New York.
Killham, K. 1994. Soil Ecology. Cambridge
University Press, Cambridge
Norton, J.M., Firestone, M.K. 1996. N
dynamics in the rhizosphere of Pinus
ponderosa seedlings. Soil Biology and
Biochemistry 28, 351-362.
Pastor, J., Aber, J.D., McClaugherty, C.A.
1984. Aboveground production and N and P
cycling along a nitrogen mineralization
gradient on Blackhawk Island, Wisconsin.
Ecology 65, 256-268.
Paul, E.A., Clark, F.E. 1996, Soil Biology and
Biochemistry. Academic Press Inc, San
Diego
Schimel, J.P., Jackson, L.E., Firestone, M.K.
1989. Spatial and temporal effects on plantmicrobial competition for inorganic nitrogen
in a California annual grassland. Soil
Biology and Biochemistry 21, 1059-1066.
Schlesinger, W.H. 1991. Biogeochemistry. An
analysis of global change. Academic Press,
San Diego.
10
V
Rapid turnover of DOC in
temperate forests accounts
for increased CO2 production
at elevated temperature
Per Bengtson & Göran Bengtsson
Department of Ecology, Lund University, Sölvegatan 37,
Lund SE-223 62, Sweden
Climatic models predict a global-mean
temperature rise of 1.2-6.9 °C by the end of
the century1. The effect of the temperature
increase on the mineralization of organic
carbon is debated and has implications for
the interpretation of temperate forest soils
as carbon sinks2-4. Whereas CO2 enrichment
experiments have failed to show an effect of
elevated CO2 levels on the mass of carbon in
the mineral soil5-7, carbon is accumulated in
soil organic matter fractions with relatively
short turnover times, suggesting a limited
potential of long-term carbon sequestration
in forest soils5. Here we present a simple
model, supported by experimental data, that
accounts for temperature dependence of the
CO2 production through a tight microbial
control of the flow of carbon between soil
organic matter (SOM), DOC, and the
microbial biomass.
Increased tree growth in response to
elevated CO2 levels and global warming are
thought to mitigate the current rise in
atmospheric CO2 concentrations8-9. But then,
increased temperature can also be expected to
accelerate the decomposition of soil organic
matter and the release of CO2 to the
atmosphere. This assumption is supported by
short-term degradation experiments10-11, but
results from forest warming experiments and
analysis of decomposition data from soils
spanning the global range of mean annual
temperature have called it into question12-13.
One of the reasons for the opposing
conclusions may be that turnover times of soil
C are sometimes estimated with the assumption
that C exists as a single homogenous pool12,
possibly masking the effect of temperature on a
small, active soil C fraction with temperature
dependent turnover rates3. Additional problems
to accurately predict C mineralization rates
may arise from the ubiquitous use of first order
decomposition kinetics when modelling soil
organic matter breakdown14. First order
approximations may be useful in describing C
mineralization rates but depend on the
assumption that the maximum concentration of
the substrate is much lower than the halfsaturation constant. To our knowledge, the
assumption has not been validated, probably
since the combination of high microbial
diversity, the ability of microorganisms to
utilize several different substrates, and the
presence of numerous possible substrates
makes it virtually impossible.
To circumvent these problems we designed
an experiment in which the concentration and
δ13C of total C (SOM), phosphate buffer
extractable C (POC)15, and DOC in a temperate
forest soil were estimated. DOC was enriched
in 13C by 1.5‰ compared to CO2, and the other
two C pools were depleted by 1.4‰ (Fig. 1).
The comparison suggested that DOC was the
source of the respired CO2, since CO2 is
normally depleted in the order of 1-2‰
compared to the substrate16-17. Given that C
mineralization is described by first order
kinetics, and that DOC was the source of the
respired CO2, the respiration rate should be
dependent on the concentration of DOC.
However, the CO2 production during a 24 hour
incubation of 108 soil samples from the top 5
cm of the mineral soil at field temperature was
not related to the concentration of C in any of
the three C pools, indicating that first order
approximations would be inappropriate to
estimate C mineralization rates. Furthermore,
the quantity of the respired C exceeded the
DOC pool 3-4 times, suggesting a turnover of
DOC several times per day (Figure 2). The
validity of this observation was elucidated by
an independent calculation of the flow of C
between the microbial biomass and the DOC
pool.
Since gross C assimilation in soil
microorganisms is difficult to estimate directly,
it was calculated from the gross NH4+-N
assimilation found in a 15N pool dilution
experiment18. The NH4+-N assimilation in the
1
-24.0
-24.5
δ 13CPDB (‰)
-25.0
-25.5
-26.0
-26.5
-27.0
-27.5
-28.0
CO2
DOC
POC
SOM
Figure 1 δ13C values relative to PDB of CO2, DOC,
POC, and SOM.
108 soil samples was 11.2 ± 1.2 mg N kg-1 d-1
(Mean ± SE). This would correspond to an
assimilation of 112 mg C kg-1 d-1, given a
microbial C/N ratio of 10 (Cassimilated =
Nassimilated × C/Nmicroorganisms). With a respiration
rate of 149.2 ± 11.4 mg C kg-1 d-1 (Fig. 2), the
calculated C use efficiency becomame 42.9%
(Cuse efficiency = (Cassimilated / (Cassmilated + Crespired))
× 100), which is within the range normally
found19.
In situ variations in microbial biomass or
numbers in soils are hardly detected from day
to day or even week to week, although the
microbial activity can vary ten-fold20-22.
Therefore, we assumed a constant microbial
biomass, and that 112 mg C kg-1 d-1 was
mobilized and returned to the DOC pool. The
13
C enrichment of the microorganisms utilizing
DOC, corresponding to the depletion of 13C in
CO2, was calculated as: frespired × (δ13CCO2 δ13CDOC) – fassimilated × (δ13CDOC δ13Cmicroorganisms) = 0, where frespired and fassimilated
represent the fraction of the substrate that was
respired and assimilated, respectively. With a C
use efficiency of 42.9%, δ13C of the
microorganisms was - 22.7‰, equivalent to a
13
C enrichment by 2.0‰ compared to their C
source (DOC). This magnitude of enrichment
is in agreement with literature data23-24.
Two insights emerge from the calculations.
First, with an import of 112 µg C kg-1 and an
export of 261 µg C kg-1 d-1 (Cassimilated +
Crespired) the DOC pool would rapidly be
depleted. Second, with a 13C enrichment by
2.0‰ of the microorganisms and their waste
products, δ13C of DOC would rapidly increase.
The ambiguity can be resolved by considering
soil organic matter (SOM) as another source of
DOC, since neither litter nor roots could supply
enough DOC. We calculated the fraction of
DOC that would originate from SOM (fSOM) in
order to balance the import of enriched C from
the bacteria and maintain δ13CDOC at -24.7‰:
fSOM = (δ13CDOC - δ13CSOM) / (δ13Cmicroorganisms δ13CSOM), where δ13Cmicroorganisms was -22.7‰
(calculated above) and δ13CSOM -27.6‰ (Fig.
1). SOM was found to contribute with 59.1%
of the DOC, leaving a 40.9% contribution with
the microorganisms, and the total DOC
production became 112 / 0.409 = 274 mg C kg1 -1
d , or almost exactly the same quantity as
was exported. The calculations reaffirm the
rapid turnover of DOC by microorganisms, in
accordance with the observation that DOC
leached from soil is old because of extensive
recycling in the microbial biomass25. The
findings emphasize the importance of microbial
activity in controlling the flow of C to and
from the DOC pool.
The enzymatic breakdown of SOM to DOC
is obviously a rate-limiting step for the CO2
production, and since enzymatic activity is
dependent
on
temperature,
increased
temperatures would tend to enhance the
breakdown of SOM and the release of CO2.
The mechanism for the control of CO2
production suggested is independent of the
extent to which some of the DOC is
refractory26, since the flow in and out of the
DOC pool is dependent on the microbial
activity rather than on the amount of DOC. Our
analysis of the transient fate of DOC draws the
attention to microbial activity as the link
between increased CO2 production from
temperate forest soils and increased
temperature and precipitation, causing what has
been referred to as a “runaway greenhouse
effect”2.
Methods
The experiment was conducted in a mixed
beech/oak stand in Torup in southwestern
2
CO2
DOC release
(162 mg C kg-1 d-1)
SOM
(40.4 ± 1.2 g C
kg-1)
DOC
(37.1 ± 2.0 mg
2. C uptake
C kg-1)
DOC uptake
(261 mg C kg-1 d-1)
C mobilization
(112 mg C kg-1 d-1)
Respiration
(149 mg C kg-1 d-1)
Microorganisms
Microbial breakdown of SOM
Figure 2 The amount of DOC and SOM (Mean ± SE) in the 108 soil samples, and the measured and calculated flow of
C between those pools and the microbial biomass.
Sweden (55°33’N, 13°12’E) at the beginning
of May 2003. Soil samples (approx. 5 g) were
put in 20 ml headspace vials, which were then
sealed with rubber septa and kept at ambient
temperature in the field. After 24 hours, 12 ml
of gas (atmospheric pressure) from each vial
were transferred to a set of evacuated 12 ml
exetainers with a gastight syringe. Exetainers
that did not retrieve at least 11 ml of gas from
the syringe were discarded. The amount and
δ13CPDB of CO2 in the headspace of these
exetainers was measured in a PDZ Europa 2020 isotope ratio mass spectrometer (IRMS)
coupled to an ANCA-GSL elemental analyser
(PDZ Europa, UK).
The gross NH4+ assimilation was estimated
by the pool dilution/enrichment technique.
PVC cylinders (length 10 cm, diameter 11 cm)
were inserted into the soil after removal of the
litter. 20 ml of a solution containing 2.4 µg
15
NH4+-N ml-1 (as 15NH4Cl, 98% 15N,
Cambridge Isotopic Laboratories) were added
with a syringe to the soil in the cylinders.
Within two hours, two samples of the top 5 cm
of the soil were taken from each cylinder. The
two samples (approximately 10 g) were
combined and extracted with 1.0 M KCl. The
amount of 14NH4+-N and 15NH4+-N, and 14NO3-N and 15NO3--N in the extract was determined
in the ANCA-GSL 20-20 IRMS after diffusion
according to standard IAEA procedures27.
Another set of samples was taken after 24
hours and treated as above.
The amount of DOC and its δ13CPDB was
determined by transferring field moist soil
samples to 0.2 µm Z-spin filter units, which
were centrifuged at 26000×g for 20 min at 4°C.
The soil water in the receiver tube was
transferred to a 5×8 mm tin cup, evaporated to
dryness at 35°C, and analysed by IRMS as
described above. The total amount of C in
SOM and its δ13CPDB was determined by
grinding dried (70°C) soil samples in a Retch
MM200 ball mill, transferring a portion of the
soil to a 5×8 mm tin cup and analysing by
IRMS.
1. Stott, P.A. & Kettleborough, J.A. Origins
and estimates of uncertainty in predictions
of twenty-first century temperature rise.
Nature 416, 723-726 (2002).
2. Grace, M. & Raymen, M. Respiration in
the balance. Nature 404, 819-820 (2000).
3. Davidson, E.A., Trumbore, S.E. &
Amundson, R. Biogeochemistry: Soil
warming and organic carbon content
Nature 408, 789-790 (2000).
4. Trumbore, S.E., Chadwick, O.A. &
Amundson, R. Science 272, 393-396
(2000).
5. Schlesinger, W.H. & Lichter, J. Limited
carbon storage in soil and litter of
experimental forest plots under increased
atmospheric CO2. Nature 411, 466-469
(2001).
3
6. Hungate, B.A. et al. The fate of carbon in
grasslands
under
carbon
dioxide
enrichment. Nature 388, 576-579 (1997).
7. van Kessel, C., Horwath, W.R., Hartwig,
U., Harris, D. & Andreas Lüscher, A. Net
soil carbon input under ambient and
elevated CO2 concentrations: isotopic
evidence after 4 years. Glob. Change Biol.
6, 435-444 (2000).
8. Ciais, P., Peylin, P. & Bousquet, P.
Regional biospheric fluxes as inferred from
atmospheric CO2 measurments. Ecol. Appl.
10, 1574-1589 (2000)
9. Houghton, R.A., Hackler, J.L. &
Lawrence, K.T. The U.S. carbon budget:
contributions from land use change.
Science 285, 574-578 (1999).
10. Kirchbaum, M.U.F. The temperature
dependence of soil organic matter
decomposition, and the effect of global
warming on soil organic C starage. Soil
Biol. Biochem. 27, 753-760 (1995).
11. Winkler, J.P., Cherry, R.S. & Schlesinger,
W.H. The Q10 relationship of microbial
respiration in a temperate forest soil. Soil
Biol. Biochem. 28, 1067-1072 (1996).
12. Giardina, C.P. & Ryan, MG. Evidence that
decomposition rates of organic carbon in
mineral soil do not vary with temperature.
Nature 404, 858-861 (2000).
13. Valentini, R. et al. Respiration as the main
determinant of carbon balance in European
forests. Nature 404, 861-865 (2000).
14. Schimel, J.P. & Weintraub, M.N. The
implications of exoenzyme activity on
microbial carbon and nitrogen limitation in
soil: a theoretical model. Soil Boil.
Biochem. 35, 549-563 (2003).
15. Matsumoto, S., Noriharu, A. & Yamagata,
M. Extraction of mineralizable organic
matter from soils by a neutral phosphate
buffer solution. Soil Biol. Biochem. 32,
(2000).
16. Schweizer, M., Fear, J. & Cadisch, G.
Isotopic (13C) fractionation during plant
residue decomposition and its implications
for soil organic matter studies. Rapid
Commun. Mass Spectrom. 13, 1284-1290
(1999).
17. Aggarawal, P.K., Fuller, M.E., Gurgas,
M.M., Manning, J.F. & Dillon, M.A. Use
of stable oxygen and carbon isotope
analyses for monitoring the pathways and
rates of intrinsic and enhanced in situ
biodegradation. Environ. Sci. Technol. 31,
590-596 (1997).
18. Wessel and Tietema
19. Holland, E.A. & Coleman, D.C. Litter
Placement Effects on Microbial and
Organic Matter Dynamics in an
Agroecosystem. Ecology 68, 425-433.
20. Lovell, R.D., Hatch, D.J. 1998. Stimulation
of microbial activity following spring
applications of nitrogen. Biol. Fertil. Soils
26, 28-30.
21. Puri, G.M. & Ashman, R. Relationship
between soil microbial biomass and gross
N mineralization. Soil Biol. Biochem. 30,
251-256 (1998).
22. Bloem, J., Ellenbroek, F.M., Baer-Gilissen,
M.J., Cappenberg, T.E. Protozoan grazing
and bacterial production in stratified Lake
Vechten estimated with fluorescently
labeled bacteria and by thymidine
incorporation. Appl. Environ. Microbiol.
55, 1787-1795 (1989).
23. Gleixner, G., Danier, H.-J., Werner, R.A.
& Schmidt, H.-L. Correlations between the
13
C content of primary and secondary plant
products in different cell compartments and
that in decomposing Basidiomycetes. Plant
Phys. 102, 1287-1290 (1993)
24. Santruckova, H., Bird, M.I. & Lloyd, J.
Microbial processes and carbon-isotope
fractionation in tropical and temperate
grassland soils. Func. Ecol. 14, 108-104
(2000).
25. Schiff, S.L., Aravena, R., Trumbmore,
S.E., Hinton, M.J., Elgood, R. & Dillon,
P.J. Export of DOC from forested
catchment on the Precambrian Shield of
Central Ohio: Clues from 13C and 14C.
Biogeochem. 36, 43-65 (1997).
26. Hongve, D. Production of dissolved
organic carbon in forested catchments. J.
Hydrol. 224, 91-99 (1999).
27. IAEA Training Course Series No. 2. Use of
Isotope and Radiation Methods in Soil and
4
Water Management and Crop Nutrition,
IAEA, Vienna (2001).
Acknowledgements. This work was sponsored
by grants from the Foundation for Strategic
Environmental Research in Sweden. Niklas
Törneman and Therese Nilsson are greatly
acknowledged for their assistance in the field
and laboratory, respectively.
5
VI
Bacterial immobilization and remineralization of N at different
growth rates and N concentrations
Per Bengtson and Göran Bengtsson
Department of Ecology, Lund University, Sölvegatan 37, Lund SE-223 62, Sweden
Abstract
An experiment was designed with the ambition to resolve two largely unadressed questions about the
turnover of N in soils. One is the influence of microbial growth rate on mobilization and remineralization of
cellular N. The other is to what extent heterotrophic immobilization of NO3- is controlled by the soil
concentration of NH4+. Bacteria were extracted from soil and inoculated into an aqueous medium at a low
density. Various pool dilution/enrichment experiments were then carried out to [1] calculate the gross N
immobilization and remineralization rates, [2] their dependence on NH4+ and NO3- concentrations, and [3] the
microbial preference for NH4+ and NO3- depending on the NH4+/NO3- concentration ratio. Remineralization of
microbial N occurred mainly at high growth rates and NH4+ concentrations, while intracellular recycling of N
seemed to be an efficient way for bacteria to withstand low inorganic N concentrations. Thus, mobilization
and remineralization of microbial N is likely to occur only when conditions promote high growth rates, i.e.
when microorganisms are not substrate limited and temperature and moisture conditions are favourable. The
results supported previous observations of high NO3- immobilization rates, especially at low NH4+
concentrations, but NO3- was also immobilized at high NH4 concentrations. The latter can be understood if
part of the microbial community has a preference for NO3- over NH4+.
1. Introduction
Soil
microorganisms
are
efficient
scavengers of inorganic N, with immobilization
rates sometimes even exceeding mineralization
rates (Nadelhoffer et al., 1984; Giblin et al.,
1991; Wagener and Schimel, 1998). Bacteria
have the capacity to utilize both NH4+ and NO3, but NO3- needs to be reduced to NH4+ before
it can be incorporated into amino acids
(Jansson, 1958; Broadbent and Tyler, 1962;
Myrold and Tiedje, 1986). The reduction of
NO3- to NH4+ requires energy, which is often
limiting heterotrophic microorganisms in soils
(Tate, 1995), making NH4+ the preferred N
source for heterotrophic bacteria. However, the
use of isotopic techniques to estimate gross
rates of N turnover has illustrated the capacity
of soil microorganisms to immobilize NO3even in the presence of a measurable NH4+
pool (Zak et al., 1990; Davidson et al. 1992;
Stark and Hart, 1997; Kaye and Hart, 1998).
Depletion of NH4+ at micro-sites within a
heterogeneous soil has been proposed as an
explanation (Schimel et al., 1989), but the
conditions under which NO3- immobilization
occurs remain uncertain.
Whereas the quantitative importance of the
immobilization process is well recognized, the
factors determining the rate of remineralization
of microbial N are largely unknown. Since the
bacterial biomass often has short turnover
times (Joergensen et al., 1990; Hart et al.,
1994; Bååth, 1998; Vance and Chapin, 2001),
it is reasonable to expect that mobilization of N
contained within the bacterial biomass is
quantitatively as important as immobilization
of inorganic N. Several mechanisms may be
responsible for the mobilization. Bacteria
accumulate solutes, many of which contain N,
in the cytoplasm during osmotic upshift due to
soil drying. When the soil is rewetted, rapid
and extensive solute efflux occurs as a result of
osmotic downshift (Wood, 1999) This gives a
mechanistic explanation to the well-known
phenomenon that rewetting of an air-dried soil
results in a flush of C and N mineralization
(Birch, 1958; Agarwal et al., 1971; van Gestel
et al., 1991; Pulleman and Tietema, 1999;
Fierer and Schimel, 2003). Predation may also
be important in controlling remineralization of
microbially bound N. For instance, high
protozoan activity coincides with high
mineralization and immobilization rates and
1
short turnover times of the microbial biomass
(Christensen et al., 1996, Rønn et al., 2002).
Similarly, the abundance of nematodes and the
microbial activity are correlated (Mamilov and
Dilly, 2002), and addition of earthworms to a
soil decreases the total microbial biomass but
increases its activity (Zhang et al., 2000).
The immobilization of N is largely
dependent on the microbial biomass and
activity. Accordingly, a high production and
breakdown of RNA (Mason and Egli, 1993)
and proteins accompanies bacterial growth,
especially during exponential growth (Norris
and Koch, 1972), and Therkildsen et al. (1997)
showed that urea production in marine bacteria
depended on the bacterial state of growth.
Therefore, one may anticipate that the growth
rate control of the intracellular turnover of N
compounds is extended to the mobilization
rates, either as a result of leakage of NH3 or by
excretion of metabolic waste products, such as
urea.
We made a series of bioassays with soil
bacteria suspended in an aqueous medium to
test the influence of microbial growth rate on
the remineralization of cellular N (hypothesis:
remineralization positively dependent on
growth rate), and the influence of different
NH4+ concentrations on NO3- immobilization
(hypothesis: no NO3- is immobilized at high
NH4+ concentrations). The experiments were
made in an aqueous environment to eliminate
the influence of predation and osmotic
variations on the mobilization rate, and the
effect of micro-sites free of NH4+ on the NO3immobilization rate, which may have occurred
in a soil.
with 200 ml of distilled water in a Sorvall
Omnimixer at 80 % of max speed for 1 min.
The soil suspension was then centrifuged at
750 ×g for 10 min at 5°C, after which the
supernatant containing the bacteria was
decanted through glass wool and then filtered
through a 0.8 µm filter to remove predators.
The filtrate was transferred to centrifuge tubes
and centrifuged at 10000 ×g for 10 min at 4°C
on a Sorvall RC5B Plus. The supernatant was
discarded and the pellet washed twice with the
culture medium (see below).
2.2. Isolation and
bacteria
15
N enrichment of the
A bacterial culture was established in 1000
ml of artificial lake water (Lehman 1980). The
lake water was enriched with glucose (3.0 µg C
ml-1, final concentration), phosphate (KH2PO4,
1.7 µg ml-1, final concentration), and 15Nnitrogen (15NH4Cl, 0.3 µg N ml-1, final
concentration), and the extracted bacteria
inoculated to a density of 1.0*105 bacteria ml-1.
The bacteria were then left to grow in the dark
at 20°C on an orbital shaker. At the same time,
another bacterial culture was prepared in the
same way as above, but with 15N replaced by
14
N. When the bacterial density in the cultures
had reached 1.0*107 bacteria ml-1, 25 ml
aliquots from the 14N and 15N cultures were
transferred to centrifuge tubes and centrifuged
at 10000 ×g for 10 min at 4°C. The supernatant
was discarded, the pellet washed twice with the
lake water and resuspended in 25 ml of lake
water.
2.3. Incubation and sampling
2. Material and Methods
2.1. Extraction of bacteria
Bacteria were extracted from a deciduous
forest soil (Dystric Cambisol) collected at
Torup
in
southwestern
Scania,
the
southernmost province in Sweden (55°33’N,
13°12’E), using the homogenization-extraction
method described by Bååth (1992). A portion
of soil (2.5 g wet weight) was homogenized
The suspended bacteria were inoculated to a
density of 6.6*103 bacteria ml-1 into 500 ml
polypropylene flasks containing 225 ml of lake
water. One third of the flasks received 15Nenriched bacteria and the others bacteria grown
on 14N. The cultures growing on 14N were
divided into two subgroups, one receiving
15
NH4+ and 14NO3-, and the other 14NH4+ and
15
NO3-. The cultures with 15N-enriched bacteria
received 14NH4+ and 14NO3-. Each of the three
treatments (15N-enriched bacteria (87.7 atom%
2
15
N excess), 15N-enriched NH4+ (10% 15N
excess), and 15N-enriched NO3- (10%15N
excess)) was further divided into three
subgroups, receiving 50, 100, or 250 ng NH4+N ml-1, respectively. Nitrate was added as 100
ng NO3--N ml-1, carbon as glucose at 3.0 µg C
ml-1, and phosphate as KH2PO4 at 1.7 µg ml-1
to all groups. The cultures were left to grow in
the dark at room temperature for 4 days.
Samples for measurements of bacterial density
were taken at the start of the experiment and
every 24 hours. Four ml samples were fixed
with formaldehyde at a final concentration of 2
%.
2.4. Determination of 14N + 15N in NH4+, NO3-,
and bacteria
Samples for measurements of 14N + 15N in
NH4+ and NO3- were taken at the start of the
experiment and after 48 and 96 hours. Bacteria
were collected after 96 hours for determination
of bacterial 15N enrichment. Aliquots of 20 ml
were centrifuged at 10000 ×g for 10 min at
4°C. The supernatant was transferred to 20 ml
scintillation vials and frozen, the pellet washed
once with sterile media and dissolved in 10 ml
of MilliQ water. The samples were kept frozen
until analyzed.
The frozen supernatant was thawed in a
refrigerator over night, and then 18 ml were
filtered through a 0.2 µm polycarbonate filter
to remove bacteria still remaining in the
supernatant. The filtrate was collected in a 20
ml headspace vial. NH4+ and NO3- were
diffused according standard IAEA procedures
(IAEA, 2001) with minor modificaitons.
Briefly, a standard office paper punch was used
to cut out quartz filter discs that were then
placed on a strip of PTFE tape and prepared
with 10 µl of 2.5 M HCl. A second strip of tape
was placed on top of the filter and the two tape
strips were sealed by pressing the open end of a
test tube, in a rocking circular motion around
the filter, against the tape to create an NH3trap. The trap was then added to the headspace
vial containing the supernatant. To prevent
swelling of the trap due to inward diffusion of
water vapor, KCl was added to a final
concentration of 1.0 M, followed by 0.7 g of
MgO. The headspace vials were sealed and left
at 35 °C with periodic shaking by hand. After
48 hours, the trap was removed, and the filter
was placed in a 5×8 mm tin cup and left to dry
in a dessicator. The serum bottle was left open
for 48 hours to release residues of ammonia. A
new trap was added to the bottle followed by
0.7 g of Devarda’s alloy and 0.7 g of MgO, and
the incubation was repeated. The 15N isotopic
excess in bacteria after 96 h was determined by
filtering the dissolved bacteria through a GF/F
filter (diameter 25 mm). Five smaller discs
were then cut out from the filter with a punch
(diameter 5 mm), placed in a 5×8 mm tin cup,
and left to dry in the dessicator.
15
N and 14N in the filters were determined
using continuous flow isotope ratio mass
spectrometry (CF-IRMS). The filters were
oxidized in an ANCA-GSL elemental analyzer
and NOX reduced to N2, which was passed to a
20-20 IRMS (PDZ Europa UK). The amount of
NH4+-N and 15NH4+-N, NO3--N and 15NO3--N
was quantified after subtraction of blank values
and calibration against filter discs which had
received known amounts of N as NH4Cl
(calibrated against an IAEA KNO3 standard)
dissolved in 5 µl of water.
2.5. Estimation of N transformation rates
Nitrogen transformation rates were
calculated with the FLUAZ-model (Mary et al.,
1998). The calculations are based on the
isotopic dilution and isotopic enrichment
principles (Monaghan and Barraclough, 1995).
It uses a numerical 4th order Runge-Kutta
algorithm with a variable time-step to solve the
differential system, and a non-linear fitting
program (based on Marquardt´s algorithm) to
calculate the unknown N transformation rates
between the NH4+, NO3-, organic N and
biomass N pools. The gross remineralization
rate and the gross NH4+ and NO3immobilization rates were calculated from the
measurements of the amounts and 15N isotopic
excesses of NH4+ and NO3-. The FLUAZ
model minimizes the quadratic weighted error
rather than the sum of squares and therefore
requires the average value plus the coefficient
of variation for the measured variables as
3
Table 1. The generation time (days), NH4+ and NO3- immobilization, remineralization, and nitrification rates during the
experiment at the three different initial NH4+ concentrations (ng N ml-1). All rates are expressed as ng N ml-1 d-1.
Days
NH4+-N
Generation time NH4+ immobilization NO3- immobilization
Remineralization
Nitrification
1-2
3-4
250
0.247
1.266
95.7
65.6
4.28
33.3
14.3
7.8
20.15
3.53
1-2
3-4
100
0.251
1.417
34.6
0.11
3.22
37.4
0.31
0.02
12.26
0.7
1-2
3-4
50
0.263
2.129
18.87
0.75
0.00
67.0
0.06
0.24
11.46
3.57
inputs. This accounts for the variance of the
measurements, so that the variables with the
highest variability have the lowest weight
(Mary et al., 1998).
The estimated rates from the 15NH4+ and
15
NO3- enrichment experiments were validated
by simulating the 15N isotopic excess in NH4+,
NO3-, and biomass pools for given N
transformation rates (Mary et al., 1998). The
initial 15N isotopic excess in NH4+, NO3-, and
bacteria was taken from the treatment with 15N
labelled bacteria, and the dilution of bacterial
15
N during the experiment was simulated from
the estimated gross remineralization rate and
the gross NH4+ and NO3- immobilization rates
in the 15NH4+ and 15NO3- enrichment
experiments. The simulated 15N isotopic excess
in bacteria day 4 was then compared to the
measured 15N isotopic excess day 4.
2.6.Measurements of bacterial numbers and
calculations of growth rates
The bacterial density was measured using a
flow-cytometric method described by del
Giorgio et al. (1996). One ml subsamples
received 50 µl SYTO 13 stain (50 µM,
Molecular Probes) and 50 µl Fluoresbrite
Carboxy YG microspheres (1.58 µm diameter,
ca 3*105 ml-1) and were analyzed with a
Beckton Dickinson FacSort flow-cytometer
running at low sample flow rate (12 µl min-1).
The cytometer was controlled with the
CellQuest 1.2 software. Bacterial cells and
microspheres were separated in a log-log
scattergram of green fluorescence intensity
(FL1) and side light scattering (SSC). Voltages
for these parameters were set to 560 and 400,
respectively. Samples were run for one minute
or until 10000 events had occurred. Bacterial
density in the subsamples was calculated by
using the microspheres as an internal standard.
The density of microspheres in the stock
solution was counted by epifluorescence
microscopy.
The bacterial growth rate was expressed as
generation time (G) and calculated from:
G = t/N
(1)
where t is time in days, and N is the number
of generations. N was calculated from:
Bt+1 = Bt2N
(2)
where Bt is the number of bacteria at time t,
and Bt+1 the number of bacteria at time t+1. If
solved for N, equation 2 becomes:
N = (logBt+1 - logBt)/log2
(3)
3. Results
The remineralization rates of immobilized
N were low, especially at the two lowest initial
NH4+ concentrations (Table 1), indicating an
efficient intracellular recycling of N in the
bacteria.
Remineralization
and
NH4+
immobilization were positively correlated
(Table 2), supporting our hypothesis, but the
fraction of the immobilized N that was
remineralized was independent on the NH4+
immobilization rate (Table 2), probably
4
Table 2. Spearman rank order correlations between NH4+ immobilization, NO3- immobilization, remineralization,
nitrification, the ration between remineralization and NH4+ immobilization, the ratio between NO3- and NH4+
immobilization, and the ratio between nitrification and NH4+ immobilization. The significance level is indicated by
asterisks (* p<0.05, ** p<0.01)
NH4+ imm
NO3- imm
Remineralization
Nitrification
Remin/NH4+ imm
NO3- imm/NH4+ imm
Nitrification/NH4+ imm.
NH4+ imm
1.00
-0.43
0.94**
0.66
-0.37
-0.60
-0.94**
RemineNO3- imm ralization
-0.43
0.94**
1.00
-0.14
-0.14
1.00
-0.60
0.60
0.94**
-0.09
0.89*
-0.37
0.37
-0.89*
because of the low ratio of remineralization to
NH4+ immobilization day 1-2 at the lowest
initial
NH4+
concentration.
The
remineralization rate was positively related to
the growth rate day 1-2 (Table 1). It was also
highest in the culture with the highest growth
rate day 3-4. The combination of a positive
correlation between NH4+ immobilization and
remineralization and the influence of the
growth rate on remineralization day 1-2
supports our first hypothesis, that the
remineralization rate would be dependent on
the growth rate.
The immobilization of NH4+ and NO3followed the same general pattern at all three
initial NH4+ concentrations, with high rates of
NH4+ immobilization between the first two
days of the incubation, followed by increased
NO3- immobilization between day 3 and 4 as
the NH4+ concentrations decreased (Table 1).
The NH4+ immobilization rate was linearly
related to the initial NH4+ concentration during
the exponential growth phase (day 1-2, Fig. 1),
with the highest rates at the highest initial
concentration (Fig. 2). Similarly, the NO3immobilization rate during day 1-2 was also
highest at the highest initial NH4+
concentration (Table 1), contradictory to our
hypothesis that NO3- immobilization would not
occur at high NH4+ concentrations. However,
the NO3- immobilization rate day 1-2 was low
compared to day 3-4, when NH4+ became
depleted.
Two main patterns of nitrification emerged
from the data. First, the highest nitrification
rates coincided with the highest initial NH4+
Nitrification
0.66
-0.60
0.60
1.00
-0.37
-0.77
-0.43
Remin/
NH4+ imm
-0.37
0.94**
-0.09
-0.37
1.00
0.77
0.43
NO3- imm/
NH4+ imm
-0.60
0.89*
-0.37
-0.77
0.77
1.00
0.49
Nitrification/
NH4+ imm
-0.94**
0.37
-0.89*
-0.43
0.43
0.49
1.00
concentrations and the exponential growth
phase (Fig. 1, Table 1). The nitrification rates
decreased during day 3-4 (Table 1), at the same
time as the rate of NO3- immobilization became
quantitatively more important than the NH4+
immobilization, that is, the bacteria seemed to
replace NH4+ with NO3- as a N source (Table
1). Second, the ratio of nitrification to NH4+
immobilization was negatively correlated to the
NH4+ immobilization (Table 2), indicating that
the relative fate of NH4+ shifted from
nitrification to NH4+ immobilization with
increasing bacterial growth and NH4+
immobilization rates.
The FLUAZ model accurately simulated
the dilution of 15N in labelled cells during the
experiment for the given NH4+ immobilization,
NO3- immobilization, and remineralization
rates from the 15NH4+ and 15NO3- treatments.
The predicted 15N in the labelled cells day 4
was inside the 95% confidence interval of the
measured 15N enrichment in all treatments,
indicating that the FLUAZ model calculated
the rates accurately (Fig. 3). The simulations
were less precise at low initial NH4+
concentrations, probably since the precision of
the quantification and isotopic determination
by IRMS decreased at low N concentrations.
4. Discussion
Remineralization of microbial N appears to
occur mainly at high growth rates and NH4+
concentrations. Two different mechanisms may
account for the high remineralization rates on
5
1E+7
H
H
1E+6
Bacteria ml
-1
H
H
1E+5
+
-1
+
-1
250 ng NH4 -N ml
100 ng NH4 -N ml
1E+4
H
+
-1
50 ng NH4 -N ml
H
1E+3
0
1
2
3
4
Day
Fig. 1. The growth of bacteria at the three different
initial NH4+ concentrations. Error bars represent the
standard error of the mean.
+
-1
-1
NH4 immobilization (ng ml d )
100
90
80
70
60
50
40
30
20
10
0
0
50
100
+
150
200
250
-1
Initial NH4 concentration (ng ml )
Fig. 2. The NH4+ immobilization rates day 1-2 at the
three different initial NH4+ concentrations.
day 1-2 and day 3-4, respectively, in the
treatment with the highest initial NH4+
concentration. The turnover rate of intracellular
mRNA is high during exponential growth
(Norris and Koch 1972), but most of the C and
N are used to synthesize new RNA and
proteins and the excretion and remineralization
of N compounds become low, as in the cultures
with low initial NH4+ concentrations. An
unbalanced C to N supply enhances mRNA
degradation (Mason and Egli, 1993), and while
mRNA C is recycled, the excess N is excreted.
The dependence of remineralization on the
carbon supply was evident from the substrate
C:N ratio in the treatments. The treatment with
the highest initial NH4+ concentrations and the
high remineralization rate had the lowest initial
C:N ratio, <10, while the two treatments with
lower remineralization had initial C:N ratios of
15 and 20. Bacteria with an average C/N ratio
of 6 and a C use efficiency of 50% are assumed
to be C limited at C/N ratios <12 (Tate, 1995).
The mechanism responsible for the high
remineralization rate day 3-4 may be that
bacterial strains with a high growth rate and
increased efficiency of translation proliferated
in the rich environment. The payoff for high
rates of reproduction is a reduced ability to
withstand starvation (Kurland and Mikkola,
1993). When substrates became depleted and
the fast growing strains growth-arrested,
oxidation of proteins in may have lead to
proteolysis (Nyström, 2002).
Our results add another piece of evidence to
the observations that soil microorganisms
immobilize NO3- even in the presence of a
measurable NH4+ pool (Zak et al., 1990;
Davidson et al. 1992; Stark and Hart, 1997),
and that NH4+ depletion at micro-sites within a
soil is not a prerequisite for NO3immobilization (Schimel et al., 1989). These
results may also have implications for the
interpretation of a potential competition for N
between plants and microorganisms (Kaye and
Hart, 1997). One requirement for competition
is that the competitors use the same sources of
N. Plants seem to have different preferences for
NH4+ and NO3- both within and between
habitats (Högbom and Ohlson, 1991;
Falkengren-Grerup
and
LakkenborgKristensen, 1994; Falkengren-Grerup, 1995),
while heterotrophic microorganisms are
considered to prefer NH4+ and immobilize
NO3- only under N limited conditions (Jansson,
1958; Broadbent and Tyler, 1962; Myrold and
Tiedje, 1986). Our results suggest that this may
not be the case. Some strains seem to express
phenotypic plasticity in the uptake of NH4+ and
NO3-, depending on the concentration, but
others immobilize NO3- even at high NH4+
concentrations. This may follow from a
selection for strains with high capacity and
affinity for NO3-, independent of NH4+
6
2.5
B
Measured
15
J
Simulated
15
Atom %
2
N excess
N excess
1.5
B
1
B
J
B
J
J
0.5
0
+
250 ng NH4 -N ml
-1
+
100 ng NH4 -N ml
-1
+
50 ng NH4 -N ml
-1
Fig 3. The measured and simulated 15N isotopic excess in 15N enriched bacteria at the end of the experiment (day 4).
The 15N isotopic excess in the bacteria was 87.7 atom% at the start of the experiment. Error bars represent the 95%
confidence interval of the measured 15N isotopic excess.
concentrations, and those strains may compete
with plants preferring NO3- even at relatively
high NH4+ concentrations.
The observation that the ratio between
nitrification
and
NH4+
immobilization
decreased with increasing NH4+ immobilization
rates supports our results from in situ
estimations of nitrification and NH4+
immobilization rates (Paper III). A larger
fraction of NH4+ was available to nitrifiers
when the heterotrophic activity was low, as in
Hart et al. (1994) and Zak et al. (1990),
probably since nitrifiers are inferior
competitors to heterotrophic microorganisms
for NH4+ (Zak et al., 1990; Verhagen and
Laanbroek, 1991; Verhagen et al., 1995).
Substantial amounts of NH4+ were nitrified at
the beginning of the incubation at all three
NH4+ concentrations. The efficient intracellular
recycling of N in the treatments with low initial
NH4+ concentrations meant that nitrification as
well as NH4+ immobilization ceased as NH4+
was depleted. However, while NH4+ was still
immobilized during day 3 and 4 in the
treatment with the highest initial NH4+
concentration,
nitrification
decreased,
reflecting higher heterotrophic biomass yields
compared with autotrophic. Considerable
amounts of NH4+ were nitrified as long as the
heterotrophic biomass was low and the NH4+
concentrations sufficiently high to support the
N demand of both heterotrophs and the
nitrifiers, but as the biomass of the heterotrophs
increased and the NH4+ concentrations
dropped, nitrification became insignificant.
In conclusion, bacteria seem to have a high
ability to recycle N intracellularly, especially at
low N concentrations and growth rates.
Remineralization of microbial N by other
processes
than
predation
or
soil
drying/rewetting is likely to occur only when
conditions promote high growth rates, i.e.
when microorganisms are not substrate limited
and temperature and moisture conditions are
favourable. The results support previous
observations of high NO3- immobilization rates
by soil microbial communities, especially at
low NH4+ concentrations, and add new
evidence for NO3- immobilization carried out
by a fraction of the microbial community even
at high NH4+ concentrations.
Acknowledgements
This work was sponsored by grants from
the Foundation for Strategic Research in
Sweden.
7
References
Agarwal, A.S., Singh, B.R., Kanehiro, Y. 1971.
Soil nitrogen and carbon mineralization as
affected by drying-rewetting cycles. Soil
Science Society of America Proceedings 35,
96-100.
Birch, H.F. 1958. The effect of soil drying on
humus
decomposition
and
nitrogen
availability. Plant Soil 10, 9-31.
Broadbent, F.E., Tyler, K.B. 1962. Laboratory
and greenhouse investigations of nitrogen
immobilization. Proceeding of the Soil
Science Society of America 26, 459-462.
Bååth, E. 1992. Thymidine incorporation into
macromolecules of bacteria extracted from
soil by homogenization-centrifugation. Soil
Biology and Biochemistry 24, 1157-1165.
Bååth, E. 1998. Growth rates of bacterial
communities in soils at varying pH: A
comparison of the thymidine and leucine
incorporation techniques. Microbial Ecology
36, 316-327.
Christensen, S., Rønn, R, Ekelund., F,
Andersen, B., Damgaard, J., Friberg-Jensen,
U., Jensen, L., Kiil, H., Larsen, B., Larsen,
J., Riis, C., Thingsgaard, K., Thirup, C.,
Petersen, T.A., Vesterdal, L. 1996. Soil
respiration
profiles
and
protozoan
enumeration agree as microbial growth
indicators. Soil Biology and Biochemistry
28, 865-868.
Davidson, E.A., Hart, S.C., Firestone, M.K.
1992. Internal cycling of nitrate in soils of a
mature coniferous forest. Ecology 73, 11481156.
del Giorgio, P.A., Bird, D.F., Prairie, Y.T.,
Planas, D. 1996. The flow-cytometric
determination of bacterial abundance in lake
plankton with green nucleic acid stain SYTO
13. Limnology and Oceanography 41, 783789.
Falkengren-Grerup,
U.,
LakkenborgKristensen, H. 1994. Importance of
ammonium and nitrate to the performance of
herb layer species from deciduous forests in
southern Sweden. Experimental Botany 34,
31-38.
Falkengren-Grerup, U. 1995. Interspecies
difference in the prefference of ammonium
and nitrate in vascular plants. Oecologia 102,
305-311.
Fierer, N; Schimel, J.P. 2003. A proposed
mechanism for the pulse in carbon dioxide
production commonly observed following
the rapid rewetting of a dry soil. Soil Science
Society of America Journal 67, 798-805.
Giblin, A.E., Nadelhoffer, K.J., Shaver, G.R.,
Laundre, J.A., McKerrow, A.J. 1991.
Biogeochemical diversity along a riverside
toposequence in arctic Alaska. Ecological
Monographs 61, 415-435.
Hart, S.C., Nason, G.E., Myrold, D.D., Perry,
D.A. 1994. Dynamics of gross nitrogen
transformations in an old-growth forest: the
carbon connection. Ecology 75, 880-891.
Högbom, L., Ohlson, M. 1991. Nitrate
assimilation in coexisting vascular plants and
swamp forest habitats in central Sweden.
Oecologia 87, 495-499.
IAEA Training Course Series No. 14. Use of
Isotope and Radiation Methods in Soil and
Water Management and Crop Nutrition,
IAEA, Vienna, 2001.
Jansson, S.L. 1958. Tracer studies on nitrogen
transformation in soil with special attention
to
mineralization-immobilization
relationships. Annuals of the Royal
Agricultural College of Sweden 24, 101-361.
Joergensen, R.G., Brookes, P.C., Jenkinson,
D.S. 1990. Survival of the soil microbial
biomass at elevated temperatures. Soil
Biology and Biochemistry 22, 1129-1136.
Kaye, J.P., Hart, S.C. 1997. Competition for
nitrogen
between
plants
and
soil
microorganisms. Trends in Ecology and
Evolution 12, 139-143.
Kaye, J.P., Hart, S.C. 1998. Ecological
restoration alters nitrogen transformations in
a ponderosa Pine-Bunchgrass ecosystem.
Ecological Applications 8, 1052-1060.
Kurland, C.G., Mikkola, R. 1993. The impact
of nutritional states on the microevolution of
ribosomes. In: Kjellberg, S. (Editor).
Starvation in Bacteria, Phlenum Press, New
York, pp.81-102.
Lehman, P.A. 1980. Release and the cycling of
nutrient between planktonic algae and
herbivores. Limnology and Oceanography
25, 620-632.
8
Mary, B., Recous, S., Robin, D., 1998. A
model for calculating nitrogen fluxes in soil
using 15N tracing. Soil Biology and
Biochemistry 30, 1963-1979.
Mamilov, A.Sh., Dilly, O.M. 2002. Soil
microbial eco-physiology as affected by
short-term variations in environmental
conditions. Soil Biology and Biochemistry
34, 1283-1290.
Mason, C.A., Egli, T. 1993. Dynamics of
microbial growth in the decelerating and
stationary phase of batch culture. In:
Kjellberg, S. (Editor). Starvation in Bacteria,
Phlenum Press, New York, pp.81-102.
Monaghan, R., Barraclough, D. 1995.
Contributions to gross N mineralization from
15
N-labelled soil macroorganic matter
fractions during laboratory incubation. Soil
Biology and Biochemistry 27, 1623-1628.
Myrold, D.D., Tiedje, J.M. 1986. Simultaneous
estimation of several nitrogen cycling rates
using 15N: theory and application. Soil
Biology and Biochemistry 18, 559-568.
Nadelhoffer, K.J., Aber, J.D., Melillo, J.M.
1984. Seasonal patterns of ammonium and
nitrate uptake in nine temperate forest
ecosystems. Plant and Soil 80, 321-335.
Norris, T.E., Koch, A.L. 1972. Effect of
growth rate on the relative rates of synthesis
of messenger, ribosomal and transfer RNS in
Escherichia coli. Journal of Molecular
Biology 64, 633-649.
Nyström, T. 2002. Aging in bacteria. Current
opinion in Microbiology 5, 596-601.
Pulleman, M., Tietema, A., 1999. Microbial C
and N transformations during drying and
rewetting of coniferous forest floor material.
Soil Biology and Biochemistry 31, 275-285.
Rønn, R., Gavito, M., Larsen, J., Jakobsen, I.,
Fredriksen, H., Christensen, S. 2002.
Response of free-living soil protozoa and
microorganisms to elevated atmospheric CO2
and presence of mycorrhiza. Soil Biol.
Biochem. 34, 923-932.
Schimel, J.P, Jackson, L.E., Firestone, M.K.
1989. Spatial and temporal effects on plantmicrobe competition for inorganic nitrogen
in a California annual grassland. Soil
Biology and Biochemistry 21, 1059-1066.
Stark, J.M., Hart, S.C. 1997. High rates of
nitrification and nitrate turnover in
undisturbed coniferous forests. Nature 385,
61-64.
Tate, R.L. III. 1995. Soil microbiology. John
Wiley and Sons, New York.
Therkildsen, M.S., Isaksen, M.F., Lomstein,
B.Aa. Urea production in the marine bacteria
Delaya venusta and Pseudomonas stutzeri
grown in a minimal medium. Aquatic
Microbial Ecology 13, 213-217.
Vance, E.D., Chapin, F.S. III. 2001. Substrate
limitations to microbial activity in taiga
forest floors. Soil Biology and Biochemistry
33, 173-188.
van Gestel, M., Ladd, J.N., Amato, M. 1991.
Carbon and nitrogen mineralization from
two soils of contrasting texture and
microaggregate stability: influence of
sequential fumigation, drying and storage.
Soil Biology and Biochemistry 23, 313-322.
Verhagen, F.J.M., Laanbroek, H.J. 1991.
Competition for ammonium between
nitrifying and heterotrophic bacteria in dual
energy-limited chemostats. Applied and
Environmental Microbiology 57, 3255-3263.
Verhagen,
F.J.M.,
Laanbroek,
H.J.,
Woldendorp, J.W. 1995. Competition for
ammonium between plant roots and
nitrifying and heterotrophic bacteria and the
effects of protozoan grazing. Plant and Soil
170, 241-250.
Wagener, S.M., Schimel, J.P. 1998.
Stratification of soil ecological processes: a
study of the birch forest floor in the Alaskan
taiga. Oikos 81, 63-74.
Wood, J.M. 1999. Osmosensing by bacteria:
Signals and membrane-based sensors.
Microbiology and Molecular Biology
Reviews 63, 230-262.
Zak, D.R., Groffman, P.M., Pregitzer, K.S.,
Christensen, S., Tiedje, J.M. 1990. The
vernal dam: Plant-microbe competition for
nitrogen in northern hardwood forests.
Ecology 71, 651-656.
Zhang, B-G., Li, G.-T., Shen, T.-S., Wang, J.K., Sun, Z. 2000. Changes in microbial
biomass C, N, P and enzyme activities in soil
incubated with the earthworms Metaphire
guillelmi or Eisenia fetida. Soil Biology and
Biochemistry 32, 2055-2062.
9
The following is a list of Doctoral theses from the Department of
Chemical Ecology and Ecotoxicology, Lund University, Sweden
1. ANDERS TUNLID, Chemical signatures in studies of bacterial communities.
Highly sensitive analyses by gas chromatography and mass spectrometry. October 3,
1986.
2. ANDERS THURÉN, Phthalate esters in the environment: analytical methods,
occurrence, distribution and biological effects. November 4, 1988.
3. PETER SUNDIN, Plant root exudates in interactions between plants and soil
microorganisms. A gnobiotic approach. March 16, 1990.
4. ANDERS VALEUR, Utilization of chromatography and mass spectrometry for the
estimation of microbial dynamics. October 16, 1992.
5. HANS EK, Nitrogen acquisition, transport and metabolisation in intact
ectomycorrhizial assocoations studied by 15N stable isotope techniques. May 14,
1993.
6. ROLAND LINDQUIST, Dispersal of bacteria in ground water-mechanism, kinetics
and consequences for facilitated transport. December 3, 1993.
7. ALMUT GERHARDT, Effect of metals on stream invertebrates. February 17, 1995.
8. OLOF REGNELL, Methyl mercury in lakes: factors affecting its production and
partitioning between water and sediment. April 21, 1995.
9. PER WOIN, Xenobiotics in aquatic ecosystems: effects at different levels of
organization. December 15, 1995.
10. GÖRAN EWALD, Role of lipids in the fate of organochlorine compounds in
aquatic ecosystems. October 18, 1996.
11. JOHAN KNULST, Interfaces in aquatic ecosystems: Implications for transport
and impact of anthropogenic compounds. December 13, 1996.
12. GUDRUN BREMLE, Polychlorinated biphenyls (PCB) in a river ecosystem.
April 25, 1997.
13. CHRISTER BERGWALL, Denitrification as an adaptive trait in soil and
groundwater bacteria. November 14, 1997.
14. ANNA WALLSTEDT, Temporal variation and phytotoxicity of Batatasin-III
produced by Empetrum hermaphrodytum. November 27, 1998.
15. DARIUS SABALIUNAS, Semipermeable membrane devices in monitoring of
organic pollutants in the aquatic environment. April 28, 1999.
16. CECILIA AGRELL, Atmospheric transport of persistent organic pollutants to
aquatic ecosystems. May 21, 1999.
17. OLOF BERGLUND, The influence of ecological processes on the accumulation
of persistent organochlorines in aquatic ecosystems. September 17, 1999.
18. HELENA BJÖRN, Uptake, turn-over and distribution of chlorinated fatty acids in
aquatic biota. October 1, 1999.
19. RUEY-JANE FAN, Learning and memory in moths Plasticity in behaviour and
neurophysiology. December 1, 2000.
20. FREDRIK ÖSTRAND, Behaviour of pine sawflies in relation to pheromonebased pest management. January 19, 2001.
21. MATTIAS LARSSON, Neural interfaces to the odour world of scarab beetles.
March 2, 2001
22. CECILIA BACKE, Persistent organic pollutants in the atmosphere — spatial and
temporal variations. May 4, 2001.
23. RICKARD IGNELL, Olfaction in desert locusts — Anatomy, function and
plasticity of the central olfactory system. May 11, 2001.
24. CAMILLA RYNE, Pyralid moth reproduction: Communication, Constraints &
Control. November 2, 2001.
25. DAINIUS PLEPYS, Odour-mediated nectar foraging in the silver Y moth,
Autographa gamma.November 30, 2001.
26. DAVID ABRAHAM, Molecular aspects of pheromone evolution in moths. May 7,
2002.
27. GLENN SVENSSON, Disruption of moth mating behaviour using olfactory and
acoustic cues. October 25, 2002.
28. LINA WENDT-RASCH, Ecological effects of pesticides in freshwater model
ecosystems February 28, 2003.
29. ARNOUT TER SCHURE, Polybrominated diphenylethers in the environment
October 3, 2003.
30. ELNA NILSSON, Movement patterns and displacement of a soil insect. May 28,
2004.
31. PERBENGTSON, Microbial mobilization and immobilization of soil nitrogen.
June 4, 2004.