• Carex curvula;
  • CO2 enrichment;
  • δ15N;
  • ecosystem N inputs;
  • native grassland communities;
  • nitrogen fixation;
  • %Ndfa:percentage of nitrogen derived from atmosphere;
  • Trifolium alpinum


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

1. Increasing carbon dioxide concentration (E: 680 μl CO2 litre–1vs ambient, A: 355 μl CO2 litre–1) around late-successional Alpine sedge communities of the Swiss Central Alps (2450 m) for four growing seasons (1992–1995) had no detectable effect on symbiotic N2 fixation in Trifolium alpinum—the sole N2-fixing plant species in these communities (74 ± 30 mg N m–2 year–1, A and E plots pooled).

2. This result is based on data collected in the fourth growing season showing that elevated CO2 had no effect on Trifolium above-ground biomass (4·4 ± 1·7 g m–2, A and E plots pooled, n = 24) or N content per unit land area (124 ± 51 mg N m–2, A and E pooled), or on the percentage of N Trifolium derived from the atmosphere through symbiotic N2 fixation (%Ndfa: 61·0 ± 4·1 across A and E plots) estimated using the 15N dilution method.

3. Thus, it appears that N inputs to this ecosystem via symbiotic N2 fixation will not be dramatically affected in the foreseeable future even as atmospheric CO2 continues to rise.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

Exposure of nitrogen-fixing (N2-fixing) plants and plant communities containing N2-fixing species to elevated CO2 often stimulates symbiotic N2 fixation as a result of increased carbon availability for nodule formation (e.g. Phillips et al. 1976; Norby 1987; Arnone & Gordon 1990; Thomas et al. 1991) and for fuelling nitrogenase activity (Arnone & Gordon 1990; Vogel & Curtis 1995; Tissue, Megonigal & Thomas 1997). At the level of the plant community, CO2-enhanced N2 fixation (Zanetti et al. 1996) has been shown to stimulate the growth of N2-fixers (Newton et al. 1994; Soussana & Hartwig 1996; Hebeisen et al. 1997), increase their representation in the community and may even reduce the growth of neighbouring non-fixing plant species (in highly productive agro-ecosystems: e.g. Overdieck & Reining 1986; Lüscher et al. 1996). However, the longer-term consequences of increased inputs of symbiotically fixed N to ecosystems under elevated CO2 are as yet uncertain. Data from native ecosystems containing N2-fixing plants are sparse, but do suggest that CO2 effects are smaller, occur more slowly and are more variable (Leadley & Stöcklin 1997; Leadley et al. 1998) than those observed in agro-ecosystems. These relatively small responses may owe in part to significant phosphorus limitation in native soils (Stöcklin, Schweizer & Körner 1998).

In high Alpine ecosystems N inputs from symbiotic N2 fixation are relatively low (Wojciechowski & Heimbrook 1984; Johnson & Rumbaugh 1986; Bowman, Schardt & Schmidt 1996). However, they still may be significant compared to other sources of N (Bowman et al. 1996) and may have important local effects on neighbouring non-fixing plant species (as occurs in other native ecosystems, see Vitousek & Walker 1989). Thus, CO2-induced changes in symbiotic N2 fixation could eventually lead to noticeable shifts in the composition of Alpine plant communities. The objectives of this study were to: (1) estimate the amount of symbiotically fixed N entering high Alpine grassland ecosystems via the root nodules of Trifolium alpinum, the only N2 fixing species in this community; (2) assess the potential effects of elevated atmospheric CO2 on this in the fourth growing season under CO2 enrichment; and (3) determine whether potential CO2-induced changes in N2 fixation result from changes in the biomass of Trifolium in the plant communities (and consequently nodule biomass and possibly nitrogenase activity) or to shifts in the percentage of its N it derives from the atmosphere.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

Twenty-four 37 cm-diameter plots were established in 1992 in a high Alpine (2450 m) grassland in the Swiss Alps over a c. 0·2 ha area of natural grassland as described in Körner et al. (1996). This grassland is dominated by the sedge Carex curvula (50–60% cover) but contains a number of subdominant species (at least 13: Schäppi & Körner 1996), including the N2-fixing species T. alpinum (< 5% cover, B. Schäppi, unpublished data) and the forb Leontodon helveticus (20–30% cover). From June 1992 to August 1995, half of the plots were exposed to elevated atmospheric CO2 concentrations (mean: 680 μl litre–1, ‘E’) during the growing seasons (early July to mid-September) using 40 cm tall Plexiglas open-top chambers (OTCs). The other half of the plots were also equipped with OTCs but were maintained at current ambient CO2 concentrations (mean: 355 μl litre–1, ‘A’) as described in Körner et al. (1996).

The 15N-dilution method (e.g. Weaver 1986) was used to estimate the percentage of N in T. alpinum above-ground biomass which derived from the atmosphere (%Ndfa). To do this a total of 390 mg 15NH415NO3 m–2 (99%15N-enriched) was applied to each of the 24 plots twice during the 1995 growing season (equivalent to 1·4 kg N ha–1 year–1). This amount of N added to label the plots with 15N (140 mg N m–2) represented about 2% of the N taken up by the plant communities maintained at both ambient and elevated CO2 (i.e. plant-available N: see Arnone 1997) and can thus be considered negligible in terms of its potential effects as a fertilizer or an inhibitor of symbiotic N2 fixation. Half was applied at the beginning of shoot growth (13 July, when all plots were free of snow) and half at mid-season (26 July). On 13 July 15NH415NO3 was dissolved in spring water and applied as ‘rain’ (0·5 mm) using a pressurized spray. Each plot was then watered again (as spray) with an additional 1·0 mm spring water to wash any 15N label from the foliage into the soil. Because soils were drier on 26 July, label was applied in 4 mm ‘rain’, followed by a 4 mm wash. Leontodon helveticus, a co-occurring non-fixing forb, was used as the reference species because its root distribution and phenology are similar to those of Trifolium, and because it was the only forb species present on all plots where Trifolium occurred. The sedge C. curvula was also present on all plots containing Trifolium but its phenology, rooting depth and potential to harbour rhizospheric N2 fixing bacteria (Karagatzides, Lewis & Schulman 1985) differed markedly from those of Trifolium and disqualified its use as a suitable reference species.

In August 1995, all plots were harvested and above-ground biomass of each species in each plot measured (see Körner et al. 1997). Trifolium root systems could not be separated effectively from those of the co-occurring species at harvest (Körner et al. 1997). Peak Trifolium above-ground N content for each plot was calculated by multiplying peak above-ground biomass (g m–2) by its N concentration (mg N g–1; measured with a Leco Model 900 CHN analyser, LECO, St. Joseph, MI, USA, on dried and powdered tissue pooled from each plot). The amount of 15N label in Trifolium and Leontodon foliage was measured on pooled leaf samples from each of the plots where species were present together using a Delta-S Finnigan isotope ratio mass spectrometer coupled on-line to a Carlo Erba element analyser. Samples were combusted to N2 and the 29N2/28N2 ratios determined. Organic N standards provided by the International Atomic Energy Agency were used to calibrate the mass spectrometer and element analyser. These standards were calibrated against atmospheric N2 (M. Saurer, personal communication). Percentage Ndfa in Trifolium was calculated for each plot as follows:

%Ndfa = 100 × (atm%15Nref– atm%15Nfixer)/

(atm%15Nref– atm%15Natmos),

where atm%15N is atom percentage of 15N present in a pooled leaf sample, ref is a non-fixing reference species (Leontodon), fixer is a N2-fixing species (Trifolium) and atmos is an atmosphere with an atom%15N of 0·3663%. Symbiotic N2 fixation on a land area basis was calculated for each plot containing Trifolium by multiplying its peak above-ground N content (mg N m–2) in 1995 by the %Ndfa. Trifolium neither appeared nor disappeared from any plot over the 4 years of the experiment. Leontodon was present in all 24 plots.

Plots were arranged in a randomized complete block (RCB) design with 12 blocks. However, because Trifolium was present in only 10 of the 24 plots (i.e. five plots each in ambient and elevated CO2 treatments), the effects of elevated CO2 on Trifolium leaf δ15N and %Ndfa were examined with an unpaired Student’s t-test. The effects of elevated CO2 on peak-season Trifolium above-ground biomass (g m–2) and N content (g N m–2), and on land-area-based N2 fixation (mg N m–2 year–1) were analysed in two ways: first using only those plots containing Trifolium (n = 5 with the Student’s t-test); second using the RCB design and all plots [n = 12: two-way analysis of variance, ANOVA, with CO2 (1 df) and Block (11 df) as factors]. To determine whether Trifolium and Leontodon leaf δ15N labels differed significantly, and whether elevated CO2 affected leaf δ15N, a two-way ANOVA was used on just those plots containing Trifolium, with Species (1 df) and CO2 (1 df) as factors. A significant treatment effect was designated as P < 0·05, however, non-significant effects reported below were always associated with P > 0·40.

Results and discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

Elevated CO2 had no detectable effect on Trifolium peak above-ground biomass or N content in the final growing season (1995), regardless of whether all plots (n = 12; Table 1), or just those with Trifolium (data not shown), were used for the analysis. The relatively high variability observed in these two measures at the ecosystem level (n = 12) was owing to the fact that seven plots in each CO2 treatment contained no Trifolium and to the fact that Trifolium abundance on plots where it occurred varied dramatically. Peak Trifolium above-ground biomass estimated using all 24 plots (4·4 g m–2) was relatively low, representing about 2% of total above-ground vascular plant biomass, but similar to estimates from pre- (2·6 g m–2; Körner et al. 1996) and post- (3·2 g m–2; S. Schneiter, unpublished data) study surveys conducted over a larger area (1 ha) than the one used for the CO2 experiment. The N content of Trifolium above-ground biomass at peak season viewed across all plots averaged 124 mg N m–2 (Table 1). The lack of a CO2 effect was not particularly surprising because net ecosystem carbon gain, and thus carbon availability, under elevated CO2 was only slightly greater than that measured under ambient CO2 (Diemer 1997).

Table 1.  . Effects of four growing seasons’ exposure to elevated CO2 (E: 680 μl litre–1vs ambient CO2, A: 355 μl litre–1) on Trifolium alpinum above-ground biomass and N content, and symbiotic N inputs to the ecosystem (all plots). Percentage Ndfa data are from plots in which Trifolium occurred: five A and five E plots. Symbiotic N2 fixation in Trifolium nodules was assumed to take place only during the growing season, but is expressed on an annual basis here. Values are mean ± SE. No significant (P < 0·05) CO2 effects were observed Thumbnail image of

Elevated CO2 also had no detectable effect on either Trifolium or Leontodon leaf δ15N values but the two species differed significantly from each other (Fig. 1). Trifolium leaf δ15N averaged across all 10 plots in which that species occurred (819‰) was about 60% (i.e. Trifolium%Ndfa) lower than the mean measured in Leontodon leaves (2184‰). This dilution of the 15N signal in Trifolium (relative to Leontodon), along with the fact that root nodules were present on plants harvested, indicate that Trifolium at this site fixes atmospheric N2, but that its reliance upon N2 (%Ndfa) was not affected by high CO2 (Table 1). Percentage Ndfa values calculated for Trifolium were at the bottom of the range reported for Trifolium species growing in Rocky Mountain Alpine meadows (70–100%; Bowman et al. 1996). Even when Carex was used as a reference species, no CO2 effect was apparent on Trifolium%Ndfa (data not shown). Percentage Ndfa calculated in this way averaged about 85% across both CO2 treatments. The similar results obtained using the two different reference species increases confidence that elevated CO2 had no real effect on symbiotic N2 fixation in Trifolium on plots where it was present.


Figure 1. . Leaf δ15N values for Leontodon helveticus (reference species) and Trifolium alpinum (N2-fixing species) measured at the peak of the fourth growing season (1995) of exposure to ambient (A: 355 μl litre–1) or elevated (E: 680 μl litre–1) atmospheric CO2 concentrations. Leontodon was present in all 12 plots in each CO2 treatment, Trifolium was present in five A and five E plots. Each bar represents the mean ± SE for each species and treatment (n = 12 and 5, respectively). Leontodon leaf δ15N values measured in plots where Trifolium was present did not differ (P > 0·40) from values measured in plots where Trifolium was not present.

Download figure to PowerPoint

As a consequence of the apparent lack of a CO2 effect on Trifolium peak above-ground N content and Trifolium%Ndfa, estimated N inputs to the ecosystem from symbiotic N2 fixation were also not affected by elevated CO2. Annual N input from Trifolium N2 fixation viewed across the entire ecosystem (i.e. A and E plots pooled) was 74 ± 30 mg N m–2 year–1 (Table 1), less than 10% of the amount of N estimated to be entering the system annually via atmospheric deposition (Rihm 1996). This annual input from symbiotic N2 fixation is significantly higher than that reported for Alpine tundra in the Rocky Mountains containing Trifolium spp. (estimated using acetylene reduction assays and calculated using data presented in Wojciechowski & Heimbrook 1984; 1 mg N m–2 year–1) but significantly lower than values obtained later from the same area (using 15N natural abundance; Bowman et al. 1996; 450 mg N m–2 year–1). The apparent differences in symbiotic N inputs among these sites appear to be due to the use of different methods to estimate N2 fixation and to actual differences in abundance of N2-fixing plant species in the communities. The relatively low inputs of symbiotically fixed N measured in this study do suggest that non-symbiotic N2 fixation, by cyanobacteria or by free-living rhizoplane bacteria (Karagatzides et al. 1985), and N from atmospheric deposition (see Holzmann & Haselwandter 1988) may be important sources of N in this Swiss Alpine grassland. Of course estimates of N2 fixation per unit land area may also differ from those presented in Table 1, and would be slightly higher, if Carex had been used as a reference species.

Although elevated CO2 had no apparent effect on Trifolium N2 fixation, either via altered %Ndfa or changes in above-ground biomass, the slight increase in community root biomass (12%, P = 0·09; Körner et al. 1997; and root N, unpublished data) observed in 1995 in plots maintained at elevated CO2 suggests that land-area-based estimates of Trifolium N content and N2 fixation presented in Table 1 may be slightly low, perhaps more so under elevated CO2. However, conclusions about the effects of elevated CO2 drawn, based on above-ground data, would probably not be affected even if Trifolium roots had been included in the analyses. This is because mean N content and N2 fixation at high CO2 tended to be lower than means at ambient CO2 (Table 1).

Thus, the results from this study indicate that present N inputs from Trifolium symbiotic N2 fixation to this late-successional Alpine system are relatively small (and spatially quite variable on scales of one to tens of meters), and suggest that rising atmospheric CO2 will probably not dramatically change this situation in the foreseeable future. These results, taken together with those from two previous studies conducted at this site (Niklaus & Körner 1996; Arnone 1997), also suggest that rising atmospheric CO2 will not greatly affect the nitrogen economy in this high Alpine grassland. These findings are also in marked contrast to the large effects of elevated CO2 on symbiotic N2 fixation observed in fertile lowland grasslands (e.g. Zanetti et al. 1996) and underline the importance of resource co-limitation in constraining the response of native ecosystems to elevated CO2.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

I thank: Bernd Schäppi for providing plant N data; Matthias Diemer, Lukas Zimmermann and Fritz Ehrsam for maintaining the remote Alpine CO2-enrichment facility; and Rolf Siegwolf and Matthias Saurer for their expertise with mass spectrometry. I am especially grateful to Christian Körner for valuable comments on an earlier version of the manuscript. The research was supported by a grant from the Swiss National Science Foundation (NFP-31: 31–30048·90). I am also grateful to the Treubel-Fonds of Basel, Switzerland for financial support.

  1. Present address: Biological Sciences Center, Desert Research Institute, 2215 Raggio Parkway, Reno, NV 89512, USA.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References
  • 1
    Arnone III, J.A. (1997) Indices of plant N availability in an alpine grassland under elevated CO2. Plant and Soil 13, 6166.
  • 2
    Arnone III, J.A. & Gordon, J.C. (1990) Effect of nodulation, nitrogen fixation and CO2 enrichment on the physiology, growth and dry mass allocation of seedlings of Alnus rubra Bong. The New Phytologist 13, 5566.
  • 3
    Bowman, W.D., Schardt, J.C., Schmidt, S.K. (1996) Symbiotic N2-fixation in alpine tundra: ecosystem input and variation in fixation rates among communities. Oecologia 13, 345350.
  • 4
    Diemer, M. (1997) Effects of elevated CO2 on gas exchange characteristics of alpine grassland. Acta Oecologica 13, 177182.
  • 5
    Hebeisen, T., Lüscher, A., Zanetti, S., Fischer, B., Hartwig, U.A., Frehner, M., Hendrey, G.R., Blum, H., Nösberger, J. (1997) Growth response of Trifolium repens L. & Lolium perenne L. as monocultures and bi-species mixture to free air CO2 enrichment and management. Global Change Biology 13, 149160.
  • 6
    Holzmann, H.-P. & Haselwandter, K. (1988) Contribution of nitrogen fixation to nitrogen nutrition in an alpine sedge community (Caricetum curvulae). Oecologia 13, 298302.
  • 7
    Johnson, D.A. & Rumbaugh, M.D. (1986) Field nodulation and acetylene reduction activity of high-altitude legumes in the western United States. Arctic and Alpine Research 13, 171179.
  • 8
    Karagatzides, J.D., Lewis, M.C., Schulman, H.M. (1985) Nitrogen fixation in high arctic tundra at Sarcpa Lake, Northwest Territories. Canadian Journal of Botany 13, 974979.
  • 9
    Körner, Ch., Diemer, M., Schäppi, B., Niklaus, P., Arnone III, J.A. (1997) The responses of alpine grassland to four seasons of CO2 enrichment: a synthesis. Acta Oecologica 13, 165175.
  • 10
    Körner, Ch., Diemer, M.W., Schäppi, B., Zimmermann, L. (1996) Response of alpine vegetation to elevated CO2. Carbon Dioxide and Terrestrial Ecosystems (eds H. A. Mooney & G. W. Koch), pp. 177–196. Academic Press, San Diego.
  • 11
    Leadley, P.W., Niklaus, P.A., Stocker, R., Körner, Ch. (1999) A field study of the effects of elevated CO2 on plant biomass and community structure in a calcareous grassland. Oecologia 13, 3949.
  • 12
    Leadley, P.W. & Stöcklin, J. (1997) Effects of elevated CO2 on model calcareous grasslands: community, species, and genotype level responses. Global Change Biology 13, 389397.
  • 13
    Lüscher, A., Hebeisen, T., Zanetti, S., Hartwig, U.A., Blum, H., Hendrey, G.R., Nösberger, J. (1996) Differences between legumes and nonlegumes of permanent grassland in their responses to free-air carbon dioxide enrichment: its effect on competition in a multispecies mixture. Carbon Dioxide, Populations, and Communities (eds Ch. Körner & F. A. Bazzaz), pp.287–300. Academic Press, San Diego.
  • 14
    Newton, P.C.D., Clark, H., Bell, C.C., Glasgow, E.M., Campbell, B.D. (1994) Effects of elevated CO2 and simulated seasonal changes in temperature on the species composition and growth rates of pasture turves. Annals of Botany 13, 5359.
  • 15
    Niklaus, P.A. & Körner, Ch. (1996) Responses of soil microbiota of a late successional alpine grassland to long-term CO2 enrichment. Plant and Soil 13, 219229.
  • 16
    Norby, R.J. (1987) Nodulation and nitrogenase activity in nitrogen-fixing woody plants stimulated by CO2 enrichment of the atmosphere. Physiologia Plantarum 13, 7782.
  • 17
    Overdieck, D. & Reining, E. (1986) Effect of atmospheric CO2 enrichment on perennial ryegrass (Lolium perenne) and white clover (Trifolium repens L.) competing in managed model ecosystems. I. Phytomass production. Acta Oecologica 13, 2635.
  • 18
    Phillips, D.A., Newell, K.D., Hassell, S.A., Felling, C.E. (1976) The effect of CO2 enrichment on root nodule development and symbiotic N2 reduction in Pisum sativum L. American Journal of Botany 13, 356362.
  • 19
    Rihm, B. (1996) Critical Loads of Nitrogen and their Exceedances. Environmental Series no. 275. Federal Office of Environment, Forests and Landscape, Bern, Switzerland.
  • 20
    Schäppi, B. & Körner, Ch. (1996) Growth responses in an alpine grassland to elevated CO2. Oecologia 13, 4352.
  • 21
    Soussana, J.F. & Hartwig, U.A. (1996) The effects of elevated CO2 on symbiotic N2 fixation: a link between the carbon and nitrogen cycles in grassland ecosystems. Plant and Soil 13, 321332.
  • 22
    Stöcklin, J., Schweizer, K., Körner, Ch. (1998) Effects of elevated CO2 and phosphorus addition on productivity and community composition of intact monoliths from calcareous grassland. Oecologia 13, 5056.
  • 23
    Thomas, R.B., Richter, D.D., Ye, H., Heine, P.R., Strain, B.R. (1991) Nitrogen dynamics and growth of seedlings of an N-fixing tree (Gliricidia sepium (Jacq. Walp.)) exposed to elevated atmospheric carbon dioxide. Oecologia 13, 415412.
  • 24
    Tissue, D.T., Megonigal, J.P., Thomas, R.B. (1997) Nitrogenase activity and N2 fixation are stimulated by elevated CO2 in a tropical N2-fixing tree. Oecologia 13, 2833.
  • 25
    Vitousek, P.M. & Walker, L.R. (1989) Biological invasion by Myrica faya in Hawai’i: plant demography, nitrogen fixation, ecosystem effects. Ecological Monographs 13, 247265.
  • 26
    Vogel, C.S. & Curtis, P.S. (1995) Leaf gas exchange and nitrogen dynamics o f N2-fixing, field-grown Alnus glutinosa under elevated atmospheric CO2. Global Change Biology 13, 5561.
  • 27
    Weaver, R.W. (1986) Measurement of biological dinitrogen fixation in the field. Field Measurement of Dinitrogen Fixation and Denitrification (eds R. D. Hauck & R. W. Weaver), pp. 1–10. Soil Science Society of America, Madison, WI.
  • 28
    Wojciechowski, M.F. & Heimbrook, M.E. (1984) Dinitrogen fixation in alpine tundra, Niwot Ridge, Front Range, Colorado, U.S.A. Arctic and Alpine Research 13, 110.
  • 29
    Zanetti, S., Hartwig, U.A., Lüscher, A., Hebeisen, T., Frehner, M., Fisher, B.U., Hendrey, G.R., Blum, H., Nösberger, J. (1996) Stimuation of symbiotic N2 fixation in Trifolium repens L. under elevated atmospheric pCO2 in a grassland ecosystem. Plant Physiology 13, 575583.