OIKOS 97: 69–78, 2002
Reindeer grazing and soil microbial processes in two suboceanic
and two subcontinental tundra heaths
Sari Stark, Rauni Strömmer and Juha Tuomi
Stark, S., Strömmer, R. and Tuomi, J. 2002. Reindeer grazing and soil microbial
processes in two suboceanic and two subcontinental tundra heaths. – Oikos 97:
69–78.
In oceanic, nutrient-rich Fennoscandian arctic-alpine tundra heaths, grazing by
reindeer has been found to increase herbs and graminoids in relation to dwarf shrubs.
In continental lichen heaths in the inland with nutrient-poor conditions, however,
slowly decomposable dwarf shrubs are favoured by grazing. According to a hypothesis, by favouring easily decomposing plants in nutrient-rich conditions and slowly
decomposing plants in nutrient-poor conditions, herbivory enhances soil nutrient
cycling in nutrient-rich and retards it in nutrient-poor areas. We tested this hypothesis by comparing the impact of reindeer grazing on soil C and N mineralization
between two oceanic and two continental arctic-alpine tundra heaths.
Although soil respiration and microbial metabolic activity were enhanced by grazing
in the suboceanic but not in the subcontinental tundra heaths, gross N mineralization
rates were higher in the grazed areas in soils from all study sites, indicating that
reindeer grazing leads to increased rates of nutrient cycling in both nutrient-poor and
nutrient-rich tundra heaths. Thus, in the subcontinental tundra heaths, the increase in
soil N concentrations due to mammalian waste products enhances N mineralization
rates, even though the organic C quality is not improved by reindeer grazing. There
was some site-specific variation in the strength of the reindeer effects on various
microbial processes and soil properties, which can be related to spatial variation in
grazing intensity and timing, as these factors in turn affect the nutrient sink strength
of the vegetation.
S. Stark and J. Tuomi, Dept of Biology, Uni6. of Oulu, P.O. Box 3000, FIN-90014
Oulu, Finland. (Present address of SS: Finnish Forest Research Institute, Ro6aniemi
Research Station, P.O. Box 16, FIN-96300 Ro6aniemi, Finland [[email protected]].)
– R. Strömmer, Dept of Ecological and En6ironmental Sci., Uni6. of Helsinki,
Niemenkatu 73, FIN-15210 Lahti, Finland.
Grazing by reindeer (Rangifer tarandus L.) significantly
influences the plant community structure on northernmost Fennoscandian arctic-alpine tundra heaths (Oksanen 1978, Oksanen and Virtanen 1995, Oksanen et
al. 1995, Olofsson et al. 2001). Reindeer is the only
large mammalian herbivore species in the area, and has
been present throughout the post-glacial period (Oksanen et al. 1995). Mammalian herbivores frequently
have many effects on soil nutrient cycling and organic
matter decomposition. Firstly, herbivores alter the
quality of organic matter and litter by changing the
composition of plant communities and via their urine
and faecal N input, which has an impact on the mineralization rates (McKendrick et al. 1980, Pastor and
Naiman 1992, Pastor et al. 1993, Augustine and McNaughton 1998, Frank and Groffman 1998). Secondly,
soil microbial processes are affected by an impact of
herbivores on the soil microclimate (Zimov et al. 1995,
Insam et al. 1996). As productivity in arctic ecosystems
is strongly limited by soil nutrient availability (Chapin
and Shaver 1996, Jonasson et al. 1999a), any herbivore-mediated change in nutrient mineralization rates
Accepted 14 November 2001
Copyright © OIKOS 2002
ISSN 0030-1299
OIKOS 97:1 (2002)
69
has a potential to have a major feed-back effect on
the ecosystem productivity.
According to a hypothesis by Bryant et al. (1983),
Oksanen (1990) and Chapin (1991), herbivory often
selects for fast-growing and easily decomposing plants
– such as graminoids – in nutrient-rich ecosystems,
and slowly decomposing dwarf shrubs with high levels of defensive chemicals in nutrient-poor systems.
Consequently, herbivory may enhance soil nutrient
cycling in fertile soils and retard it in infertile soils
(Chapin 1991). In northernmost Fennoscandia, the
climatic differences between the Atlantic coast and
the inland create a natural gradient where the continental and oceanic sections differ in plant productivity and soil nutrient availability (Ahti et al. 1968,
Haapasaari 1988, Oksanen and Virtanen 1995). Reindeer annually migrate between these areas. In the
oceanic nutrient-rich coastland, summer pastures of
reindeer, vegetation consists of dwarf shrubs and
graminoids which are preferred in their summer diet,
and grazing has been found to increase herbs and
graminoids in relation to dwarf shrubs (Olofsson et
al. 2001). Covered by lichens, the most important
food resource for the reindeer during the winter, the
continental uplands with nutrient-poor conditions are
used as winter ranges (Warenberg et al. 1997). Lichen
biomass is considerably reduced by grazing, and grazing favours dwarf shrubs with high levels of defensive
chemicals (Oksanen and Virtanen 1995, Olofsson et
al. 2001). Therefore, the reindeer-mediated changes in
the decomposability of vegetation in the arctic-alpine
ecosystems would appear to support the hypothesis of
Chapin (1991).
We tested the hypothesis of differential responses
of microbial processes to herbivory in nutrient-rich
and nutrient-poor ecosystems, and compared the
long-term impact of reindeer grazing on mineralization processes between oceanic and continental arcticalpine tundra heaths. Our first-hand prediction was
that reindeer grazing enhances C and N mineralization in the oceanic summer ranges and retards them
in the continental lichen ranges. However, there is
some evidence that reindeer grazing enhances the net
N mineralization in lichen-dominated boreal forests
(Stark et al. 2000). The positive effect on N mineralization is probably due to decreased microbial immobilization, as litter decomposition and soil respiration
are retarded at the same time, or, alternatively, because the proportion of slowly decomposable lichen
litter is decreased (Stark et al. 2000). Consequently,
the effect of herbivory on soil microbial processes in
lichen-dominated ecosystems may differ from that in
other low-productivity ecosystems. We therefore predicted that reindeer grazing enhances soil C and N
mineralization in oceanic, and either enhances or retards them in continental arctic-alpine tundra heaths.
70
Methods
Study sites
Two subcontinental arctic-alpine tundra heaths in
northern Finland (Jesnalvaara and Nuorttitunturi) and
one suboceanic tundra heath in northernmost Norway
(Lagisduoddar) were chosen for study. Soil data from
another suboceanic tundra heath (Raisduoddar), collected one year previously and reported by Olofsson et
al. (2001), were included. The study sites were established just above the forest line to the low oroarctic
zone (Ahti et al. 1968, Haapasaari 1988, Oksanen and
Virtanen 1995).
The suboceanic tundra heaths, Lagisduoddar
(70°30%N, 27°30%E) and Raisduoddar (69°30%N,
27°30%E), in northernmost Norway are both divided
into a section of ungrazed and grazed parts by a fence
built in the 1960’s to reduce the risk of reindeer illegaly
entering the winter ranges during the summer. The
winter range side of the fence will here be referred to as
ungrazed control, because the area has deep snow cover
during reindeer spring migration, and the animals pass
through the area within a short time during their
autumn migration (Olofsson et al. 2001). In Lagisduoddar, the vegetation consists of dwarf shrubs (Betula
nana, Vaccinium uliginosum, V. myrtillus, V. 6itisidaea), graminoids (Deschampsia flexuosa, Juncus
trifidus, Carex sp.). Grazing by reindeer has somewhat
increased not only the proportion of graminoids in the
vegetation, but also the bare ground area (Olofsson et
al. 2001). In Raisduoddar, the vegetation on the ungrazed part consists of the same species as in Lagisduoddar, but on the grazed part the abundance of
graminoids (e.g. Carex sp., J. trifidus, D. flexuosa) and
herbs have dramatically increased (Olofsson et al.
2001).
In Jesnalvaara (69°46%N, 26°57%E) and Nuorttitunturi
(67°48%N, 29°40%E) in northern Finland, the lichen
grounds are grazed by reindeer during the winter.
Jesnalvaara is situated in the hemiarctic zone where the
forests mainly consist of birch woodland (Ahti et al.
1968, Haapasaari 1988, Oksanen and Virtanen 1995).
There are a few bush-like Betula pubescens ssp. czerepano6ii on the summit of Jesnalvaara, but the dominant higher plants are Empetrum nigrum, V. uliginosum,
Arctostaphylos alpina and Loiseleuria procumbens (Kytöviita 1988). A reindeer exclosure (size of 4000 m2) was
built on the summit in 1968. The exclosed area is
dominated by Cladonia stellaris and Cetraria ni6alis,
but in the grazed area their abundance and the amount
of lichen biomass in general are significantly smaller
(Kytöviita 1988, Broll 2000). E. hermaphroditum is decreased by grazing, while J. trifidus is increased. Nuorttitunturi is a treeless hill situated in the eastern border
of Finland in the northern boreal zone in the Saariselkä
uplands, where the forests are mainly dominated by
OIKOS 97:1 (2002)
Norway spruce (Picea abies). Nuorttitunturi is divided
into ungrazed and grazed sections by the fence built
close to the borderline to Russia to prevent the Finnish
reindeer from entering there. The ungrazed part of
Nuorttitunturi has not been grazed for approximately
80 years, because there are hardly any reindeer on that
Russian territory. The ground layer is dominated by
several species of Cladonia and dwarf shrubs (V. uliginosum, V. 6itis-idaea, B. nana), and the effect of grazing
on the species composition and biomass is similar to
that in Jesnalvaara (S. Stark, pers. obs.).
Soil sampling and analysis
Eight sampling plots were chosen along the reindeer
fence on both the grazed and ungrazed sides, the plots
being situated 12 m from the fence and 24 m from the
nearest replicate on the same side of the fence. The
distances in Jesnalvaara were 6 m and 12 m, respectively. Composite soil samples consisting of 3 soil cores
of 10 cm diameter and reaching through the whole
organic layer ( \5 cm in thickness) were taken from
each sampling plot. The samples were kept cool during
transportation to the laboratory, and then frozen until
analysis.
Live plant material and roots were removed and the
samples were carefully homogenized. Bulk density,
gravimetric moisture (105°C, 12 h) and organic matter
content (475°C, 4 h) were determined. A subsample of
ca 6 g was extracted with 50 ml of 0.5M K2SO4 for a
colorimetric analysis of the extractable N and P contents. For analysing NH4-N content, 10 ml of extractant was made alkaline by addition of MgO and the
resulting NH3 diffused into 0.01M H2SO4 (Crooke and
Simpson 1971) and was analysed colorimetrically
(Bremner 1965). NO3-N content was analysed colorimetrically using automated flow injection (QuikChem
8000, Lachat Instruments, Inc., Milwaukee, USA). Total dissolved N was analysed from the extracts by
oxidizing all extractable N to NO3− (Williams et al.
1995), analysing it as NO3− , and dissolved organic N
was calculated as the difference between total dissolved
N and the NH4-N and NO3-N concentrations. Total
dissolved P was analysed by oxidizing all total extractable P to PO4− (Williams et al. 1995), and assessing
PO4-P with automated flow injection (QuikChem 8000,
Lachat Instruments, Inc.). As the total dissolved P
contents were very low, we pooled the dissolved organic
P and PO4 contents into a common pool of soluble P.
Microbial N and P were extracted from soils using
0.5M K2SO4 after chloroform fumigation (18 h)
(Brookes et al. 1985) and analysed as total dissolved N
and P after oxidation as above. Microbial N and P
were calculated by subtracting the total dissolved N
and P in the unfumigated extracts from those in the
fumigated ones. Soil total N content was analysed by
OIKOS 97:1 (2002)
an automated analyzer (EA 1110, CHNS-O, CE
Instruments).
Mineralization of N and C
For assessing the effect of reindeer grazing on the
organic matter (OM) quality at the contrasting study
sites, laboratory measurements for potential gross and
net N mineralization and soil respiration rates were
conducted. Gross N mineralization was determined using the isotope dilution method (Hart et al. 1994). After
2-day preincubation, 10 ml of dilute (0.04 g ammonium
sulphate 15N-atom% 97.5 in 1 dm3 distilled water)
(15NH4)2SO4 solution was added to 200 cm3 of soil, and
a subsample was immediately extracted with 0.5M
K2SO4 for the recovery of 15NH4+ and the initial
15
NH4/14NH4 isotope ratio. The rest of the soil was
incubated at +20°C for three days, and after the
incubation a subsample was extracted with 0.5M
K2SO4. The NH4+ contents of the extracts were
analysed as described earlier. The samples for 15N
isotope ratio analysis were prepared by vaporizing
NH4+ from the extracts and diffusing into 0.01M
H2SO4, which was then dried at + 45°C. Diffusion and
drying were repeated until a sufficient amount of NH4+
for the 15N analysis was obtained. The 15N/14N ratio
was analysed by mass spectrometry (Europa Scientific
ANCA system) and the rates of gross N mineralization
and immobilization were calculated by the equation of
Kirkham and Bartholomew (1954). Potential net N
mineralization was measured by incubating a subsample in field moisture for 12 weeks at + 20°C, and the
initial and final NH4+ -N contents were determined as
above.
Soil respiration (carbon mineralization) was analysed
using a respirometer (Nordgren 1988, Nordgren et al.
1988). Soil moisture was adjusted to 250% of the weight
of OM (Nordgren 1988). Basal respiration was analysed
as the rate of CO2-C efflux for 40 h of incubation at
+20°C. Substrate-induced respiration (SIR), an index
of microbial biomass C, was analysed by adding 313.8
mg of substrate (200 mg of glucose, 103.7 mg of
(NH4)2SO4 and 10.1 mg of KH2PO4) to fresh weight
equivalent to 1 g OM (Anderson and Domsch 1978,
Ohtonen 1994). The metabolic quotient of the soil
microflora (qCO2) was calculated as a ratio of basal
respiration to SIR. The lag time between the addition
of substrate and the start of the exponential increase in
the respiration rate was also measured as an indicator
of microbial activity (Nordgren 1988, Ohtonen 1994).
Data analysis
The results were tested in two different ways, first after
calculating values per g soil OM, and second, after
71
calculating values per dm3, because these two ways of
presenting the results give different information in
some cases. Measurements expressed per soil OM
reflect the quality of the substrate for the soil organisms, and represent the reindeer-mediated chemical
changes in soil quality, but values per soil volume
also include the integrated effects of grazer-mediated
changes in OM content and bulk density, and are
often more appropriate to use when comparing different ecosystems (Cheng et al. 1998).
The assumption of homogeneity of variances was
tested by Levene’s test, and in the case of unequal
variances, logarithmic and reciprocal square root
transformations were used. The data were analysed
statistically by split-plot ANOVA with grazing, vegetation type and site as the main factors. Study site, a
random factor, was nested with the vegetation type.
The error d.f. for evaluating the main effect of graz-
ing and the interaction between grazing and vegetation type was calculated as MS [grazing × site
(vegetation type)]. The error d.f. for evaluating the
main effect of forest type was calculated as MS [site
(vegetation type)]. The error d.f. for evaluating the
main effect of study site was calculated as MS [grazing ×site (vegetation type)], and error d.f. for calculating the interaction between grazing and study site
was calculated as MS (error). In parameters where
ANOVA showed a significant interaction between
grazing and site (vegetation type), t-test for each site
individually was conducted. A non-parametric test
(Mann-Whitney U-test) for each site individually was
also used for net N mineralization that contained
both positive and negative values and equal variances
could not be obtained with any transformation. Data
analysis was performed using the SPSS for Windows
8.0.
Fig. 1. Soil NH4,
extractable organic N,
extractable microbial N and
total N contents per soil
volume (means 9S.E., n=8,
except in Raisduoddar,
where n =6).
J= Jesnalvaara,
N= Nuorttitunturi
(subcontinental study sites)
and L=Lagisduoddar and
R= Raisduoddar
(suboceanic study sites).
72
OIKOS 97:1 (2002)
Results
N mineralization and immobilization
Grazing increased gross N mineralization and immobilization rates on both OM and volume basis irrespective
of vegetation type (Fig. 2, Tables 1 and 2). Due to the
small soil NH4+ -N content, we were unable to obtain
reliable measures on every replicate in Jesnalvaara.
The microbial N uptake relative to supply was high
in all laboratory incubations, resulting in negative net
N mineralization rates (net immobilization) in soils
from all study sites except Jesnalvaara, being 9.4 91.3,
and 3.5 91.5 mg dm − 3 d − 1 in grazed and ungrazed
areas, respectively. In Jesnalvaara, grazing increased
potential net N mineralization on a volume basis
(Mann-Whitney test, P =0.023). Otherwise, due to a
very large variation among replicates, there were no
significant differences in net N mineralization in relation to grazing. Net N mineralization was −18.99
13.5 and −12.9918.0 mg dm − 3 d − 1 in grazed and
ungrazed areas in Nuorttitunturi, −46.9940.1 and
−75.2 916.6 mg dm − 3 d − 1 in grazed and ungrazed
areas in Lagisduoddar, and −1.7590.57 and
−1.1690.39 mg dm − 3 d − 1 in grazed and ungrazed
areas in Raisduoddar, respectively.
(P=0.049, t-test). The NO3− -N and total dissolved P
concentrations were negligible or below the detection
limit (values not shown).
The effect of grazing on microbial N was limited to
only some of the study sites with no consistent impact
within the vegetation type [grazing ×site (vegetation
type) interaction; Tables 1 and 3]. Grazing significantly
increased microbial N in Jesnalvaara (P= 0.034, t-test)
and in Lagisduoddar (PB 0.001, t-test). Extractable
microbial P was 27.0 915.9 mg dm − 3 in grazed and
8.992.4 in ungrazed plots in Jesnalvaara, 2.7 9 0.6 in
grazed and 8.6 95.4 in ungrazed plots in Nuorttitunturi, and 27.9 94.7 in grazed and 10.9 9 1.7 in ungrazed side in Lagisduoddar. Grazing significantly
increased microbial P in Lagisduoddar (P =0.024, ttest).
Total N concentrations on the volume basis were
higher on the grazed than on the ungrazed plots in the
suboceanic but not in the subcontinental tundra heaths
(Fig. 1, grazing × vegetation type interaction in Table
1). However, there was no significant influence on total
N when the values were calculated per OM (Tables 1
and 2).
Soil properties
Soil respiration, SIR and qCO2
On the volume basis, grazing significantly enhanced soil
respiration in the suboceanic tundra heaths but not in
the subcontinental tundra heaths (grazing × vegetation
type interaction; Tables 1 and 3). There were no effects
on soil respiration on the OM basis, or on SIR, but
basal respiration to SIR values (qCO2) were significantly increased by grazing in the suboceanic tundra
heaths but not in the subcontinental tundra heaths
(grazing ×vegetation type interaction; Tables 1 and 3).
The effect on time lag before exponential growth of
respiration after substrate addition was limited to only
some of the study sites [grazing ×site (vegetation type)
interaction], and was significantly shorter on the grazed
than the ungrazed plots in Raisduoddar (P= 0.019;
t-test).
Soil and microbial N and P contents
Soil NH4+ -N concentrations were affected by grazing in
only some of the sites [grazing × site (vegetation type)
interaction; Fig. 1, Tables 1 and 3], and was significantly increased by grazing in Raisduoddar (P= 0.004;
t-test). Similar trend was found in extractable organic
N concentrations (Norg; [grazing× site (vegetation type)
interaction, P B0.10], which was significantly higher on
the grazed than on the ungrazed plots in Raisduoddar
OIKOS 97:1 (2002)
Soil OM and moisture contents were significantly
higher in the suboceanic than in the subcontinental
tundra heaths, and there were no effects in relation to
reindeer grazing (Tables 1 and 4). Bulk density on a dry
weight basis, however, was significantly increased by
grazing irrespective of the vegetation type.
Discussion
Grazing significantly increased organic matter decomposition rates in the suboceanic but not in the subcontinental tundra heaths. Grazing also increased soil
respiration to SIR ratio, a carbon availability index
(CAI) (Cheng et al. 1998) or microbial metabolic activity (qCO2) (Nordgren 1988, Ohtonen 1994), in the
suboceanic tundra heaths. The results indicate that
reindeer-mediated changes in soil C quality differ between the suboceanic and the subcontinental tundra
heaths, and thus conform to our study hypothesis. The
shorter lag upon grazing in Raisduoddar indicates especially improved availability of labile C for soil microbes, because they respond quickly to added glucose
if they are adjusted to using it (Nordgren 1988). However, grazing enhanced the rate of gross N mineralization in both nutrient-poor and nutrient-rich tundra
heaths. Thus, the direction of the impact on soil nutrient availability was the same in both vegetation types,
which was not predicted by the hypothesis.
73
74
Table 1. ANOVA table for reindeer effects on microbial and soil properties tested on soil OM and volume basis. The transformations to meet the assumptions of ANOVA are
indicated by numbers: 1) logarithm 2) reciprocal square root. F and P values obtained by nested ANOVA (grazing hypothesis df =1, error df= 2; vegetation type hypothesis df = 1,
error df = 2; site (vegetation type) df = 2, error df =2; grazing×vegetation type hypothesis df =1, error df =2; grazing×site (vegetation type) hypothesis df = 2, error df = 51).
Response
Grazing
F
Microbial properties
Gross N min
OM
volume
Gross N imm
OM
volume
Soil resp
OM
volume
SIR
OM
volume
qCO2
Lag
Soil properties
OM
NH+
4
volume
Norg
OM
volume
OM
Nmic
volume
Ntot
OM
volume
OMC
moisture
bulk density
P
Veg. Type
F
Site (Veg. type)
P
F
P
Grazing×Veg. type
F
Grazing×Site (Veg. type)
P
F
P
1)
1)
1)
1)
9.085
17.564
9.358
18.825
5.383
24.220
5.693
3.587
18.702
5.194
0.033
0.013
0.052
0.017
0.143
0.035
0.109
0.194
0.028
0.150
1.520
1.648
1.342
1.469
0.114
0.315
0.007
0.057
0.300
0.460
0.342
0.327
0.366
0.349
0.768
0.631
0.940
0.833
0.638
0.567
198.447
185.715
87.900
117.661
34.103
73.078
320.553
16.647
133.839
22.177
0.005
0.005
0.011
0.008
0.028
0.013
0.003
0.057
0.007
0.043
0.005
1.748
2.848
0.958
4.580
25.971
0.001
2.164
42.617
0.266
0.946
0.255
0.186
0.391
0.163
0.033
0.977
0.274
0.010
0.657
0.087
0.102
0.179
0.140
0.958
0.444
0.073
0.529
0.068
3.643
0.917
0.903
0.837
0.869
0.390
0.644
0.930
0.592
0.935
0.033
2)
2.433
1.232
4.409
2.649
1.798
2.646
4.342
101.863
5.080
2.815
20.187
0.259
0.382
0.170
0.245
0.311
0.245
0.170
0.006
0.145
0.233
0.038
4.441
2.716
1.938
1.136
0.953
2.921
0.122
1.308
58.076
456.880
57.257
0.169
0.241
0.298
0.398
0.432
0.229
0.760
0.371
0.016
0.002
0.016
1.139
1.048
4.256
9.031
2.440
0.751
7.883
28.780
6.892
0.282
7.878
0.468
0.488
0.190
0.100
0.291
0.571
0.113
0.034
0.127
0.780
0.113
2.688
1.439
2.337
0.979
0.448
1.147
2.887
113.470
0.447
0.392
4.675
0.242
0.353
0.265
0.426
0.572
0.396
0.229
0.005
0.569
0.594
0.151
4.378
8.593
2.842
3.126
4.986
5.404
0.989
0.137
0.355
1.255
0.217
0.018
0.001
0.068
0.053
0.011
0.008
0.379
0.872
0.703
0.294
0.806
2)
1)
1)
OIKOS 97:1 (2002)
Table 2. Soil NH4, extractable organic N, extractable microbial N and total N contents, and gross N mineralization and
immobilization rates per soil OM (1 = subcontinental tundra site, 2 =suboceanic tundra site). Values are means 9 S.E., n= 8,
except in Raisduoddar, where n =6.
Jesnalvaara1
grazed
Gross Nmin.
(mg g−1
OM d−1)
Gross Nimm.
(mg g−1
OM d−1)
NH+
4 -N (mg
g−1 OM)
Norg (mg g−1
OM)
Nmic (mg g−1
OM)
Ntot (mg g−1
OM)
ungrazed
Nuorttitunturi1
grazed
Lagisduoddar2
ungrazed
grazed
ungrazed
Raisduoddar2
grazed
ungrazed
1.9 90.5
1.9 9 0.5
25.0 9 4.7
23.1 9 5.9
23.7 93.7
24.1 96.9
27.5 96.0
22.1 9 6.4
2.49 0.1
2.09 0.9
27.2 9 6.5
23.1 95.9
26.6 9 6.4
29.1 9 7.4
32.3 96.5
26.2 97.8
4.390.5
4.7 90.3
50.0 913.1
55.4 9 14.4
64.3 921.1
57.1 9 13.4 180.2 9 30.0
33.6 9 7.7
27.6 9 5.0
44.6 95.7
29.2 9 9.6
39.4 911.5
30.2 915.7 178.5 929.8 113.5 917.4
30.5 917.8
501.9958.2 349.9 9 28.6 820.69381.8 496.3 9 40.8 312.3 927.3 116.5 936.9 541.3 963.5 486.1 959.8
1.829 0.07
1.73 9 0.07
2.00 9 0.07
2.02 90.09
1.83 9 0.20
1.66 90.12
2.49 9 0.20
1.96 90.17
Table 3. Soil microbial properties at the study sites in relation to reindeer grazing (1 = subcontinental tundra site, 2 =suboceanic tundra site). Values are mean+S.E.
Microbial
property
Jesnalvaara1
grazed
Soil respiration
(mg CO2
dm−3 h−1)
(mg CO2 g−1
OM h−1)
SIR
(mg CO2
dm−3 h−1)
(mg CO2 g−1
OM h−1)
qCO2
Lag (h)
ungrazed
Nuorttitunturi1
grazed
ungrazed
grazed
ungrazed
Raisduoddar2
grazed
ungrazed
3.7 90.6
3.3 9 0.2
7.9 9 0.5
8.3 91.1
11.3 90.7
7.2 9 0.4
7.15 9 1.0
6.0 9 1.9
29.893.2
26.9 9 1.3
65.492.9
67.0 97.0
60.6 9 4.3
51.2 92.4
62.8 9 9.7
47.7 9 9.0
11.4 92.6
9.8 9 1.4
21.7 9 2.7
10.5 9 3.7
21.6 94.2
13.8 90.8
18.7 9 1.3
19.2 93.6
89.1 915
75.7 9 10
176.49 16.4 173.7 921
110.4 918
99.6 96.0
164.4 917
0.3690.02 0.40 9 0.06
37.5 9 1.9 42.4 92.3
0.38 90.02
17.39 3.0
0.40 90.03
23.19 0.8
Suboceanic tundra heaths
Reindeer grazing often increases the proportion of
graminoids in the tundra heaths (Oksanen and Virtanen
1995, Kumpula et al. 1996, Manseau et al. 1996, Post
and Klein 1996, Olofsson et al. 2001). Graminoids –
although a food resource preferred by reindeer – gain
from grazing because of their capability of fast regrowth and large investment of biomass in the root
system, which makes them less sensitive to the loss of
biomass than dwarf shrubs (Chapin 1980, Chapin et al.
1986). Thus, changes in vegetation promoted by grazing
are results of both food selection by herbivores and
differences in grazing tolerance among plant species
and growth forms (Bryant et al. 1983, Chapin et al.
1986, Augustine and McNaughton 1998). In suboceanic
tundra heaths, soil C transformations were higher in
the grazed relative to ungrazed areas, presumably because graminoid litter decomposes faster than litter of
dwarf shrubs (Hobbie 1996), and because graminoids
have more fine roots that probably exudate more low
OIKOS 97:1 (2002)
Lagisduoddar2
0.69 9 0.15 0.53 90.03
9.2 91.3 11.3 90.9
0.37 9 0.03
23.3 9 3.0
163.8916.3
0.28 90.04
37.4 93.7
molecular weight organic substances, which is easily
available substrate for soil microbes (Ruess and Seagle
1994, Bardgett et al. 1998).
The urine and faeces produced by mammalian herbivores also affect the soil microbial processes, because
slowly decomposable plant material is transformed into
labile and N-rich organic substances (McKendrick et al.
1980, Ruess and McNaughton 1987, Frank and Groffman 1998). Interestingly, graminoids gain from increased nutrient availability more than dwarf shrubs
(McKendrick et al. 1980, Jonasson 1992, Chapin and
Shaver 1996, Grellmann 2001), and the increase in
graminoids by grazing may be a result of both a better
tolerance for biomass removal and efficient absorption
of increased amount of nutrients (Olofsson et al. 2001).
This view is supported by the result of Grellmann
(2001), who found that in the absence of mammalian
waste products as an underlying mechanism, herbivores
decreased the abundance of graminoids in the arctic
plant community and had a decelerating effect on soil
nutrient cycling.
75
Fig. 2. Gross N
mineralization and
immobilization per soil
volume (means9S.E.).
Sample sizes and
abbreviations as in Fig. 1.
Table 4. Mean and S.E. of soil moisture (g H2O g−1 fresh weight), bulk density (kg dm−3 dry weight) and OM content (g OM
g−1 dry weight) in relation to reindeer grazing and vegetation type (1 =subcontinental tundra heath, 2 = suboceanic tundra
heath).
Study site
Jesnalvaara
1
Nuorttitunturi1
Lagisduoddar2
Raisduoddar2
grazed
ungrazed
grazed
ungrazed
grazed
ungrazed
grazed
ungrazed
Moisture
Bulk density
OM content
0.46 9 0.04
0.47 90.03
0.45 90.01
0.48 90.03
0.66 9 0.02
0.74 90.01
0.50 90.01
0.66 90.03
0.34 9 0.03
0.31 90.02
0.31 9 0.02
0.30 9 0.03
0.21 90.15
0.15 9 0.01
0.16 9 0.01
0.12 9 0.01
0.42 9 0.08
0.43 9 0.04
0.40 9 0.03
0.45 9 0.07
0.90 9 0.02
0.92 9 0.02
0.74 9 0.05
0.85 9 0.04
There was some site-specific variation in the strength
of the reindeer effects on various microbial processes,
which can be related to spatial variation in grazing
intensity and timing, as these factors in turn affect the
nutrient sink strength of the vegetation. Microbial
biomass N and P were strongly increased by reindeer
grazing in Lagisduoddar, indicating increased microbial
nutrient immobilization. As a comparison, there was a
non-significant elevation in microbial biomass due to
grazing in Raisduoddar, but the soil NH4+ -N and
organic N contents were markedly increased. The contrast in the reindeer effects coincides with the differences in the impacts of reindeer grazing on vegetation,
discussed by Olofsson et al. (2001). They suggested
that, in Lagisduoddar, the dominance of graminoids is
suppressed by the continuously high grazing pressure,
while graminoids have established better in Raisduoddar due to the intensive but periodic grazing. As
changes in plant communities are dependent on grazing
pressure and temporal patterns, local variation in the
changes in soil processes is also expected. In Lagisduoddar, plant production is limited by severe grazing and is
probably indifferent to N availability. A decline in the
sink strength of nutrients in plants may result in a
76
considerable increase in the microbial biomass N and P
(Jonasson et al. 1999b), which was the case in Lagisduoddar. By contrast, in Raisduoddar, plant productivity was significantly higher in the heavily grazed relative
to ungrazed area (Olofsson et al. 2001), and soil nutrients were efficiently assimilated into plant rather than
microbial biomass. Thus, soil nutrient availability was
increased by grazing at both of the suboceanic study
sites, but targeted to different components of biota.
Subcontinental tundra heaths
In nutrient-deficient lichen heaths, reindeer grazing
changes the overall composition of material entering
the decomposer foodweb by the decrease in the lichen
biomass. In heavily grazed areas, lichens form very little
necromass (Ahti 1977, Helle and Aspi 1982), and thus
produce little dead organic matter for the soil decomposer community. As grazing did not change soil respiration in subcontinental tundra heaths, this suggests
that, even though lichens produce litter that decomposes extremely slowly (Moore 1984), a reduction in the
relative amount of lichen litter does not necessarily
OIKOS 97:1 (2002)
improve litter or soil quality. Gross N mineralization
rates were, yet, increased by reindeer grazing. Thus, in
the subcontinental tundra heaths, increase in soil N
contents due to the fertilization by urine and faeces
probably enhance soil N mineralization rates, despite
organic matter decomposition rates are not affected.
In arctic-alpine ecosystems, soil microbes contain a
large proportion of the soil nutrient pool, particularly
in nutrient-poor soils (Jonasson et al. 1999b). At the
lichen-dominated sites, vegetation was dominated by
slow-growing lichens that acquire their nutrients from
atmospheric deposition, while, at the same time, large
amounts of nutrients are stored in the microbial
biomass. However, Jesnalvaara was the only area where
measurable potential net mineralization took place during the laboratory incubation. This agrees with the field
measurements of Giblin et al. (1991) and Cheng et al.
(1998), who measured net N mineralization during the
growing season in lichen heaths by contrast to net
immobilization in some other vegetation types. Mineralization measurements by incubation and in the absense of living plant roots have proved to be
problematic in the arctic, as the measurements often
lead to net immobilization (Jonasson et al. 1999a).
Assuming a likely control of microbial activity by easily
available C, the low respiration rates in Jesnalvaara
indicate that only a small proportion of total organic C
was readily available to soil microbes (Nadelhoffer et
al. 1991), resulting in low microbial N uptake and
mobilization rather than immobilization of the mineralized N.
There are some differences in the grazing effects on
soil microbial processes between the lichen heaths and
the lichen-dominated boreal forests, and these may
provide insight into some of the underlying mechanisms. Reindeer grazing decreases soil respiration in the
humus layer of lichen-dominated boreal forests (Väre et
al. 1996, Stark et al. 2000), while net N mineralization
is simultaneously enhanced (Stark et al. 2000). Thus, in
lichen-dominated forests with a very thin humus layer,
reindeer grazing causes a C limitation for the soil
microbes, but does not seem to do so in arctic-alpine
lichen heaths. In arctic-alpine ecosystems, considerable
storage of organic matter takes place through accumulation, and, as C availability does not limit microbial
growth, N is immobilized in microbial biomass to a
much greater extent than in the boreal forest (Giblin et
al. 1991, Jonasson et al. 1996, 1999a). Consequently,
increased nutrient availability by reindeer grazing may
lead to increased microbial N immobilization in arcticalpine areas, but not in the boreal forests, where soil
microbes face strong C limitation due to grazing. The
view that microbes at high altitudes are not only C-limited but limited by both C and nutrients (Körner 1999)
is well supported by the differences in the soil microbial
responses to reindeer grazing between boreal and arctic-alpine ecosystems. Whether soil microbes are limited
OIKOS 97:1 (2002)
by C or nutrients may therefore be one of the key
determinants in the effects of herbivory on soil microbial processes, especially on the balance between microbial nutrient mobilization and immobilization.
Acknowledgements – This study was funded by Emil Aaltonen
Foundation, Maj and Tor Nessling Foundation, and the
Academy of Finland (project c 40951). We thank Ismo
Sipola and Eeva Rönkä for assistance with the field work. The
15
N analysis was conducted by Martti Esala and Leena Seppänen in the Finnish Agricultural Research Center, Jokioinen.
Marko Hyvärinen and Jari Oksanen gave valuable advice with
the statistical problems, and Sirkka-Liisa Leinonen corrected
the English. John Pastor and an anonymous reviewer are
thanked for their valuable comments on the manuscript.
References
Ahti, T. 1977. Lichens in the boreal coniferous zone. – In:
Seaward, M. R. D. (ed.), Lichen ecology. Academic Press,
pp. 145 – 181.
Ahti, T., Hämet-Ahti, L and Jalas, J. 1968. Vegetation zones
and their sections in northwestern Europe. – Ann. Bot.
Fenn. 5: 169 – 211.
Anderson, J. P. E. and Domsch, K. H. 1978. A physiologically
active method for the quantification of microbial biomass
in soil. – Soil Biol. Biochem. 10: 215 – 221.
Augustine, D. J. and McNaughton, S. J. 1998. Ungulate
effects on the functional species composition of plant communities: herbivore selectivity and plant tolerance. – J.
Wildlife Manage. 62: 1165 – 1183.
Bardgett, R. D., Wardle, D. A. and Yeates, G. 1998. Linking
above-ground and below-ground interactions: how plant
responses to foliar herbivory influence soil organisms. –
Soil Biol. Biochem. 14: 1867 – 1878.
Bremner, J. M. 1965. Inorganic forms of nitrogen. – In: Black,
C. A. (ed.), Methods of soil analysis, part 2. American
Society of Agronomy, Madison, WI, pp. 1179 – 1237.
Broll, G. 2000. Influence of overgrazing by reindeer on soil
organic matter and soil microclimate of well-drained soils
in the Finnish Subarctic. – In: Lal, R., Kimble, J. M. and
Stewart, B. A. (eds), Global climate change: cold regions
ecosystems. Advances in Soil Science. Lewis Publishers,
CRC Press, Boca Raton, FL, pp. 163 – 172.
Brookes, P. C., Landman, A., Pruden, G. and Jenkinson, D.
S. 1985. Chloroform fumigation and the release of soil
nitrogen: a rapid direct extraction method to measure
microbial biomass nitrogen in soils. – Soil Biol. Biochem.
17: 837 – 842.
Bryant, J. P., Chapin, F. S. III and Klein, D. R. 1983.
Carbon/nutrient balance of boreal plants in relation to
vertebrate herbivory. – Oikos 40: 357 – 368.
Chapin, F. S. III 1980. Nutrient allocation and response to
defoliation in tundra plants. – Arct. Alp. Res. 12: 553 –
563.
Chapin, F. S. III 1991. Effects of stresses on nutrient availability. – In: Mooney, H. A., Winner, W. E. and Pell, E. J.
(eds), Response of plants to multiple stresses. Academic
Press, pp. 66 – 67.
Chapin, F. S. III and Shaver, G. S. 1996. Physiological and
growth responses of arctic plants to a field experiment
simulating climatic changes. – Ecology 77: 822 – 840.
Chapin, F. S. III, McKendrick, J. D. and Johnson, D. A.
1986. Seasonal changes in carbon fractions in Alaskan
tundra plants of differing growth forms: implications for
herbivores. – J. Ecol. 74: 707 – 731.
Cheng, W., Virginia, R. A., Oberbauer, S. F. et al. 1998. Soil
nitrogen, microbial biomass, respiration along an arctic
toposequence. – Soil Sci. Soc. Am. J. 62: 654 – 662.
77
Crooke, W. M. and Simpson, W. E. 1971. Determination of
ammonium in Kjeldahl digests of crops by an automated
procedure. – J. Sci. Food Agric. 22: 9 –10.
Frank, D. A. and Groffman, P. M. 1998. Ungulate versus
topographic control of soil carbon and nitrogen processes
in grasslands of Yellowstone National Park. – Ecology 79:
2229 – 2241.
Giblin, A. E., Nadelhoffer, K. J., Shaver, G. R. et al. 1991.
Biogeochemical diversity along a toposequence in arctic
Alaska. – Ecol. Monogr. 61: 415 – 435.
Grellmann, D. 2001. Top-down and bottom-up effects in a
Fennoscandian tundra plant community. – Doctoral dissertation, Umeå Univ.
Haapasaari, M. 1988. The oligotrophic heath vegetation of
northern Fennoscandia and its zonation. – Acta Bot.
Fenn. 135: 1 –219.
Hart, S. C., Stark, J. M., Davidson, E. A. and Firestone, M.
K. 1994. Nitrogen mineralization, immobilization, and nitrification. – In: Mickelson, S. H. (ed.), Methods of soil
analysis, part 2. Microbiological and biochemical properties. Soil Science Society of America, pp. 985 –1018.
Helle, T. and Aspi, J. 1982. Effects of winter grazing by
reindeer on vegetation. – Oikos 40: 337 – 343.
Hobbie, S. E. 1996. Temperature and plant species control
over litter decomposition in Alaskan tundra. – Ecol.
Monogr. 66: 503 –522.
Insam, H., Rangger, A., Henrich, M. and Hitzl, W. 1996. The
effect of grazing on soil microbial biomass and community
on alpine pastures. – Phyton 36: 205 –216.
Jonasson, S. 1992. Plant responses to fertilization and species
removal in tundra related to community structure and
clonality. – Oikos 63: 420 –429.
Jonasson, S., Michelsen, A., Schmidt, I. K. et al. 1996. Microbial biomass C, N and P in two arctic soils and responses
to addition of NKP fertilizer and sugar: implications for
plant nutrient uptake. – Oecologia 106: 507 –515.
Jonasson, S., Michelsen, A. and Schmidt, I. K. 1999a. Coupling of nutrient cycling and carbon dynamics in the
Arctic, integration of soil microbial and plant processes. –
Appl. Soil Ecol. 11: 135 –146.
Jonasson, S., Michelsen, A., Schmidt, I. K. and Nielsen, E. V.
1999b. Responses in microbes and plants to changes in
temperature, nutrient, and light regimes in the arctic. –
Ecology 80: 1828 –1843.
Kirkham, D. and Bartholomew, W. V. 1954. Equations for
following nutrient transformations in soil, utilizing tracer
data. – Soil Sci. Soc. Am. Proc. 18: 33 –34.
Kytöviita, M.-M. 1988. The growth of reindeer-lichens and the
effect of reindeer grazing at two altitudes at Kevo, Finnish
Lapland. – MSc thesis, University of Turku (in Finnish).
Körner, C. 1999. Alpine plant life. – Springer-Verlag, Berlin
Heidelberg, 338 pages.
Kumpula, J., Colpaert, A., Kumpula, T. and Nieminen, M.
1996. Poronhoitoalueen pohjoisosan porolaidunten inventointi (in Finnish). – Riista – ja kalatalouden tutkimuslaitos. Porotutkimusaserna, Kaamanen.
Manseau, M., Huot, J. and Crete, M. 1996. Effect of summer
grazing by caribou on composition and productivity of
vegetation: community and landscape level. – J. Ecol. 84:
151 – 170.
McKendrick, J. D., Batzli, G. O., Everett, K. R. and Swanson,
J. C. 1980. Some effects of mammalian herbivores and
fertilization of tundra soils and vegetation. – Arct. Alp.
Res. 12: 565 –578.
Moore, T. R. 1984. Litter decomposition in a subarctic sprucelichen woodland, eastern Canada. – Ecology 65: 229 – 308.
78
Nadelhoffer, K. J., Giblin, A. E., Shaver, G. R. and Laundre,
J. L. 1991. Effects of temperature and substrate quality on
element mineralization in six arctic soils. – Ecology 72:
242 – 253.
Nordgren, A. 1988. An apparatus for the continuous, longterm monitoring of soil respiration rate in large number of
samples. – Soil Biol. Biochem. 20: 955 – 957.
Nordgren, A., Bååth, E. and Söderström, B. 1988. Evaluation
of soil respiration characteristics to assess heavy metal
effects on soil microorganisms using glutamic acid as a
substrate. – Soil Biol. Biochem. 20: 949 – 954.
Ohtonen, R. 1994. Accumulation of organic matter along a
pollution gradient: application of Odum’s theory of ecosystem energetics. – Microb. Ecol. 27: 43 – 55.
Oksanen, L. 1978. Lichen grounds of Finnmarksvidda, northern Norway, in relation to summer and winter grazing by
reindeer. – Reports from the Kevo subarctic Research
Station 14: 64 – 71.
Oksanen, L. 1990. Predation, herbivory, and plant strategies
along gradients of primary productivity. – In: Grace, J. B.
and Tilman, D. (eds), Perspectives on plant competition.
Academic Press, pp. 445 – 474.
Oksanen, L. and Virtanen, R. 1995. Topographic, altitudinal
and regional patterns in continental and suboceanic heath
vegetation of northern Fennoscandia. – Acta Bot. Fenn.
153: 1 – 80.
Oksanen, L., Moen, J. and Helle, T. 1995. Timberline patterns
in northernmost Fennoscandia. Relative importance of
climate and grazing. – Acta Bot. Fenn. 153: 93 – 105.
Olofsson, J., Kitti, H., Rautiainen, P., Stark, S. and Oksanen,
L. 2001. Effects of summer grazing by reindeer on composition of vegetation, productivity and nitrogen cycling. –
Ecography 24: 13 – 24.
Pastor, J. and Naiman, R. J. 1992. Selective foraging and
ecosystem processes in boreal forests. – Am. Nat. 139:
690 – 705.
Pastor, J., Dewey, B., Naiman, R. J. et al. 1993. Moose
browsing and soil fertility in the boreal forests of Isle
Royale National Park. – Ecology 74: 467 – 480.
Post, E. S. and Klein, D. R. 1996. Relationships between
graminoid growth form and levels of grazing by caribou
(Rangifer tarandus) in Alaska. – Oecologia 107: 364 – 372.
Ruess, R. W. and McNaughton, S. J. 1987. Grazing and the
dynamics of nutrient and energy regulated microbial processes in the Serengeti grasslands. – Oikos 49: 101 – 110.
Ruess, R. W. and Seagle, S. W. 1994. Landscape patterns in
soil microbial processes in the Serengeti National Park,
Tansania. – Ecology 75: 892 – 904.
Stark, S., Wardle, D. A., Ohtonen, R. et al. 2000. The effect of
reindeer grazing on decomposition, mineralization and soil
biota in a dry oligotrophic Scots pine forest. – Oikos 90:
301 – 310.
Väre, H., Ohtonen, R. and Mikkola, K. 1996. The effects and
extent of heavy grazing by reindeer in oligotrophic pine
heaths in northern eastern Fennoscandia. – Ecography 19:
245 – 253.
Warenberg, K., Danell, O8 ., Gaare, E. and Nieminen, M. 1997.
Porolaidunten kasvillisuus (in Finnish). – Landbruksforlaget, WSOY Helsinki, Finland, 112 pages.
Williams, B. L., Shand, C. A., Hill, M. et al. 1995. A procedure for the simultaneous oxidation of total soluble N and
P in extracts of fresh and fumigated soils and litters. –
Comm. Soil Sci. Plant Anal. 26: 91 – 106.
Zimov, S. A., Chuprynin, V. I., Oreshko, P. et al. 1995.
Steppe-tundra transition: a herbivore-driven biome shift at
the end of the pleistocene. – Am. Nat. 146: 765 – 794.
OIKOS 97:1 (2002)