In order to follow the uptake and allocation of N in different plant functional types and microbes in two tundra ecosystems differing in nutrient availability, we performed a 15N-labeling experiment with three N forms and followed the partitioning of 15N label among plants, microorganisms and soil organic matter. At both sites the deciduous dwarf shrub Betula nana and the evergreen Empetrum hermaphroditum absorbed added 15N at rates in the order: NH4 > NO3− > glycine, in contrast to the graminoid Carex species which took up added 15N at rates in the order NO3− > NH4 > glycine. Carex transported a high proportion of 15N to aboveground parts, whereas the dwarf shrubs allocated most 15N to underground storage. Enhanced 13C in Betula nana roots represents the first field evidence of uptake of intact glycine by this important circumpolar plant. Plant and microbial uptake of label was complementary as plants took up more inorganic than organic N, while microbes preferred organic N. Microbes initially took up a large part of the added label, but over the following four weeks microbial 15N decreased by 50% and most 15N was recovered in soil organic matter, while a smaller but slowly increasing proportion was retained in plant biomass.
In arctic ecosystems, low nitrogen (N) availability often limits plant production (Chapin and Shaver, 1996; Jonasson and Shaver, 1999; Jonasson et al., 1999) and microbial turnover of soil organic matter (SOM) (Weintraub and Schimel, 2003; Mack et al., 2004). Microbes take up both inorganic and organic N forms, while the common assumption had been that plants take up N only in inorganic form. The plants therefore are dependent on either release of N to the soil through microbial mineralization or on supply of N via mycorrhizal symbionts. However, for some years, it has been known that mycorrhizal as well as nonmycorrhizal plants also take up organic N (Chapin et al., 1993; Kielland, 1994; Näsholm et al., 1998; Nordin et al., 2004). Thus, plants are able to compete with free-living microorganisms for organic N without need for prior mineralization.
At relatively nutrient-rich tundra sites, deciduous shrubs, graminoids, and forbs dominate the vegetation, whereas at relatively nutrient-poor tundra sites, evergreen shrubs typically are more abundant (Monk, 1966; Miller, 1982). The availability of different N forms may vary among sites with different vegetation types. Some studies have shown that sites with one dominant N form in the soil are dominated by species with a high uptake capacity for this particular N form (McKane et al., 2002; Weigelt et al., 2005). Consequently, uptake of different chemical N forms among co-existing species has been suggested as an important factor determining plant competition and dominance patterns and vegetation dynamics (McKane et al., 2002; Weigelt et al., 2005). However, other studies have reported that uptake preferences for either NH4+, NO3− or glycine are unrelated to N-form prevalence in the soil N pool (Nordin et al., 2001; Bardgett et al., 2003; Miller and Bowman, 2003; Nordin et al., 2004).
Plants take up N in competition with microorganisms (Schimel and Chapin, 1996; Kaye and Hart, 1997; Hodge et al., 2000). Plants and microbes may take up different N forms with complementary preferences, and thus reduce the competitive pressure between plants and microorganisms (Lipson and Näsholm, 2001).
Short-term 15N-labeling studies (hours to days) in arctic (Nordin et al., 2004; Grogan and Jonasson, 2005) and in temperate (McFarland et al., 2002; Bardgett et al., 2003) ecosystems show that the microorganisms absorb a large part of the added label, whereas the uptake rate by plants is much lower. However, only a few studies in arctic and alpine ecosystems have followed the partitioning of added label in plants and microorganisms during periods of weeks to years. The overall conclusion from these studies is that plant proportion of added N increases with time (Marion et al., 1982; McFarland et al., 2002), mainly because of plant sequestration of N that continuously is released from the microbes with much shorter turnover time (Kaye and Hart, 1997).
In this study, we used in situ injections of 15N-labeled NH4+ and NO3− and 15N-13C-labeled glycine in mixed solutions to follow the uptake and allocation of the N forms in plants and microorganisms. The 13C labeling of glycine was included to examine whether glycine was taken up as intact amino acid by plants in these ecosystems, as shown for a limited number of species in other field experiments (Näsholm et al., 1998; Henry and Jefferies, 2003b; Nordin et al., 2004).
The study had three objectives: The first was to test if the plant uptake of NH4+, NO3− and glycine varied among plant functional types and/or between two sites of different nutrient availability. Second, we wanted to follow the label in different plant compartments (fine roots, woody tissues and leaves) to test for effects of between-ecosystem differences in soil nutrient status and plant functional type on the allocation of N within the plants. The third objective was to follow the short-term (days) and longer-term (weeks) partitioning of label between plants and microorganisms.
Materials and Methods
The experiment was conducted near the Abisko Scientific Research Station in Northern Sweden (68°19′N, 18°51′E). The climate is subarctic with an average July temperature of 11°C, and an annual mean temperature of −0.5°C (Royal Swedish Academy of Sciences, Abisko Scientific Research Station, www.ans.kiruna.se/ans.htm). We worked at two communities with differences in vegetation height and species composition that indicated different levels of nutrient availability and plant productivity.
One community was a low-productive, dry heath dominated by prostrate-growing Empetrum hermaphroditum. Empetrum hermaphroditum grew together with the subdominant deciduous and low-growing Betula nana and Vaccinium uliginosum, the evergreen dwarf shrub Andromeda polifolia and the sedges Carex vaginata and Carex distica. The vegetation canopy was open, less than 15 cm high, and underlain by mosses. The organic horizon had a depth of about 10 cm and a well-defined transition to the underlying mineral soil.
The other community was a shrub tundra with a canopy height of 40 to 50 cm dominated by Betula nana. Carex paralella made up the understorey together with mosses. Other common species were E. hermaphroditum, V. uliginosum, Equisetum arvense, A. polifolia, and Salix myrsinites. The organic horizon was 15 to 50 cm deep, moist and more mixed with the underlying mineral soil than at the heath. Hereafter the two communities are designated the heath tundra and the shrub tundra, respectively.
In each community, 48 0.2 × 0.2 m plots were laid out in six replicate blocks, i.e. with eight plots per block. The plots were selected so that there were at least one rooted individual of E. hermaphroditum and B. nana plus either C. vaginata (heath) or C. paralella (shrub tundra) in each of them. The labeling took place on 18 July 2003 at the heath tundra and on 22 July at the shrub tundra. NH4+, NO3− and glycine were mixed so that the three N-forms made up equal proportions of N in the solution. In each solution, one N-form was labeled, either as 15N-NH4Cl, 15N-KNO3− or, 15N-2(13C)-glycine (15N 96–99 atom%, 13C 98 atom%, Cambridge Isotope Laboratory). Within each community, we dispensed 100 mL of each solution evenly into the top 5 cm of two plots in each block, leaving two plots as untreated controls. The added amount corresponded to 136 mg 15N m−2.
Soil Chemical Analysis
The content of NH4+-N, NO3−-N, microbial extractable N, dissolved organic N (DON), total amino acid N, and glycine N was analyzed in soils of the control plots sampled at both sites on 18 July and at the heath and shrub tundra on 21 and 23 August, respectively. In the same samples, we determined the soil water content gravimetrically and the soil organic matter content (SOM) by combustion of dry soil at 660°C for 6 h.
Microbial extractable N was determined after CHCl3 fumigation of 10 g of fresh soil per sample for 24 h followed by extraction for 1 h in 50 mL 0.1 M K2SO4 (Anderson and Domsch, 1977; Brookes et al., 1985; Hobbie, 1992; Cheng and Virginia, 1992; Greenfield, 1995). Another 10 g fresh soil was treated as above, but without fumigation to extract soil NH4+, NO3−, and DON. Amino acids were extracted with 50 mL demineralized water in an additional 10 g soil. The extracts were filtered through Whatman GF-D filters and frozen until analysis.
NO3−-N and NH4+-N were analyzed in the extracts from the unfumigated soils using an auto analyzer (Fiastar 5000, Foss Tecator, SE). K2SO4-extracts of fumigated and unfumigated soils were digested with persulfate to oxidise the organic N fractions to inorganic N (Zhou et al., 2003) followed by analyses as above. DON was determined as the N content in digested, unfumigated extracts after subtraction of the inorganic N fraction, and the extractable microbial N was calculated by subtracting the N in the digested, unfumigated extracts from the content in the digested, fumigated extracts (Brookes et al., 1985).
The analyses of amino acids were done by HPLC using a gold amperometry cell and the analytical column AminoPac PA10, with separation of arginine, alanine, asparagine, glutamic acid, glutamine, glycine, histidine, isoleucine, leucine, lysine, methionine, phenylalanine, proline, serine, valine, tryptophan, and tyrosine. Their concentrations were summed to a measure of total soil amino acid N concentration.
15N uptake by micro-organisms was determined using the acid-trap diffusion technique (Stark and Hart, 1996). Five milliliters of digested, fumigated extract and 10 ml of unfumigated extracts spiked with 70 µg N (as NaNO3 with δ15N = −0.38 ‰) were placed in 60 ml HDPE flasks. The spiking of the unfumigated extracts was necessary to obtain enough N of 50 to 150 µg for optimal sensitivity of the isotope-ratio measurements. Devardas alloy was added to reduce any NO3−/NO2 to NH4+, and KCl was added to increase the ionic strength of the solution. Immediately before sealing the flasks with lids, 1 mL 5 M NaOH was added in order to raise the pH to >13 and convert the NH4+ to NH3. An acidified (15 µL 1.5 M H2SO4) Millipore quartz filter of 0.8 cm diameter was attached inside the lids to trap the NH3 as (NH4)2SO4. The flasks were gently shaken for four days on a shaking table at 50 rpm, after which the filters with the trapped N were removed, dried and analyzed for their 15N/14N isotope ratios. In order to calculate the microbial 15N recovery, we used the microbial N pools measured with the auto analyzer, because the N spiking of the unfumigated samples decreased the precision of the total N analysis in diffused samples. Recovery of 15N in dissolved total N (DTN) is reported as the total recovery of 15N in digested unfumigated extracts, i.e. soluble organic 15N plus inorganic 15N.
Sampling and Preparation of Plants and Soil for Isotope Analysis
Two and 26 days after labeling, we harvested six plots from each treatment by excavating the whole labeled 20 × 20 cm soil column, including attached plants, to a depth of 10 cm on the heath and to 15 cm on the shrub tundra. The samples were brought to the Abisko Scientific Research Station, kept at 2°C and sorted in order to detect differences in uptake patterns among plants of different functional type. Three “target species” from each site were selected for analysis of isotope ratios viz. the deciduous B. nana, the evergreen E. hermaphroditum, and the sedges C. vaginata (heath tundra) or C. paralella (shrub tundra), while all other species were pooled into one fraction of “mixed species.” The target species were carefully separated from the soil with as much as possible of the root system attached. The two sedges were subsequently separated into roots and leaves + rhizomes, while Betula and Empetrum were separated into leaves, fine roots (<0.5 mm diameter) and woody tissues (coarse roots + above- and belowground stems). The mixed species were separated into roots (i.e. all belowground) and leaves + stems (i.e. all aboveground). The soil was cut into cubes of about 2 × 2 × 2 cm and mixed, after which roots and rhizomes were sorted from randomly chosen cubes and processed further as for aboveground plant parts before drying, and the remaining soil was used for soil chemical analysis.
Samples for analyses of 15N and 13C natural abundance were collected from the control plots on 21 August (heath tundra) and on 23 August (shrub tundra) and fractionated as the labeled samples. The aboveground plant parts were sorted into the target species, and the mixed species were sorted into growth forms of deciduous shrubs, evergreen shrubs, graminoids, forbs, mosses, and lichens. The root biomass of each growth form was estimated by assuming the same proportions of root mass among growth forms as measured in the aboveground biomass.
All plant material was washed several times in 0.5 mM CaCl2 to remove any label adhering to the surface. All plant and soil samples were dried at 70°C for 48 h, weighed and milled, and stored in darkness until further processing.
The 15N/14N and 13C/12C isotope ratios, total N and C contents of about 5 mg of the plant and soil samples were analyzed with a Eurovector CN analyzer coupled to an Isoprime isotope ratio mass spectrometer (IRMS) (Micromass-GV Instruments). Natural abundance of isotopes is expressed in the δ notation relative to international standards: δXsample (‰) = 1000 × [(Rsample/Rstandard) − 1] where R is the molar ratio of heavyX/lightX. The C standard Vienna Pee Dee Belemnite has a 13C/12C ratio of 0.011237 and the N standard atmospheric N2 has a 15N/14N ratio of 0.003676. Atom-percentages of 15N in samples from control plots (natural abundance) were subtracted from atom-percentages of the treated samples. The recovery of 15N in bulk soil (soil + microorganisms) was determined as the average recoveries in the two separately processed samples from each plot, i.e. the dried subsample of sorted soil used in the extractions and the dried bulk soil sample remaining after sorting. The amount of 15N in the extractable microbial biomass plus 15N in the DTN pool was subtracted from the bulk soil to give the reported 15N recovery in the SOM fraction.
The statistical analyses were performed using SAS statistical package (SAS Institute v8.02, 2001). Soil variables were tested separately for the July and August harvests using one-factor ANOVAs with site as main factor. Three-factor ANOVAs were used to test for differences in 15N recovery in the different plant compartments separately including site, treatment and harvest time as main factors and with biomass included as covariate. Since there were significant differences between sites in the 15N recovery in many plant species and compartments, we increased the resolution of the test by analyzing the treatment and harvest effects by two-factor ANOVAs for each plant species and each site separately, followed by Tukey's test to compare means of each treatment. Total 15N recoveries in microorganisms were tested with two-factor ANOVAs as above, but with the N pool included as covariate. Block was included in the models whenever the block effect was significant at P < 0.10. Prior to the ANOVA tests, all data were tested for homogeneity of variances with Levene's test and, if necessary, transformed.
Plant Biomass, Soil Water and Ecosystem Nitrogen Pools
The total plant biomass (Table 1) was about 1800 g m−2 at the heath tundra, with deciduous and evergreen shrubs each making up about one-third of the biomass. B. nana, E. hermaphroditum, and C. vaginata, i.e. the “target species”, made up 9, 14, and 2%, respectively, of the total plant biomass. At the shrub tundra, with an estimated biomass of about 2400 g m−2, deciduous shrubs dominated and made up 63% of the plant biomass. B. nana, E. hermaphroditum, and C. paralella made up 31, 6, and 5%, respectively, of the biomass.
Biomass (means ± 1 SE, n = 6) of different growth forms at the heath and shrub tundra.
There were no differences in plant N pools among treatments or harvest dates. We therefore pooled the data over the different treatments and present the N distribution among plant groups and plant fractions at the July harvest where the biomass peaked (Fig. 1). The N pools of Betula, Carex, and the mixed species were larger at the shrub tundra than at the heath tundra (Fig. 1a, c, d), while the N pools of Empetrum, particularly in stems and roots, were largest at the heath tundra (Fig. 1b). Generally, the root biomass made up a higher proportion of the total biomass at the heath than at the shrub tundra. For instance, the pool in roots of Carex made up 37% of the total N pool at the heath but only 15% at the shrub tundra (Fig. 1a), and N in fine roots of Empetrum made up 12% at the heath and 8% at the shrub tundra. The pools in fine roots of Betula, however, made up an equal proportion of about 9% of the total Betula N pool at both sites.
In July, the soil water content and the concentrations of NH4+, DON, extractable microbial N, and amino acid N were significantly higher at the shrub tundra than at the heath tundra (Table 2), while in August only the soil water content and DON concentration were higher at the shrub tundra (Table 2). In contrast, the NO3− concentration was significantly higher at the heath tundra in July but nonsignificantly lower in August. At both sites, the NH4+ concentration increased significantly from July to August and the extractable microbial N concentration at the shrub tundra decreased.
Soil characteristics of heath and shrub tundra (means ± 1 SE, n = 6). Monthly means between sites that do not share the same letters are significantly different (P ≤ 0.05; Tukey's test).
Among the three N forms used in the labeling experiment, NH4+ dominated at both sites. At the heath tundra, the concentration was 30 (July) to 250 (August) times higher than the concentration of NO3− and about 3 to 4 times higher than the concentration of amino acid N (Table 2). At the shrub tundra, the NH4+ concentration was 150 to 230 times higher than the concentration of NO3− and 3.5 times higher than the amino acid N concentration. Amino acid N made up about 1% of the DON at both sites (Table 2).
The total N content of the soil at the heath was high, exceeding the plant N pool 7- to 10-fold, while the content was lower at the shrub tundra, but yet exceeded the plant N pool 2- to 4-fold. The N content in the extractable microbial biomass made up between 2 and 7% of the bulk soil N pool corresponding to about 1/5–1/8 of the plant N pool (Fig. 2).
15n Uptake and Allocation in Plants
Due to the low concentrations of inorganic and glycine N in the soil solution, the added 15N-NH4+, 15N-glycine, and 15N-NO3− made up 92.3, 99.7, and 99.7% of the respective soil solution pools at the heath and 89.6, 99.6, and 99.9 % at the shrub tundra immediately after labeling. However, at both harvests, 15N recovery in DTN was low (0.08–0.24% of the added label) at the heath tundra and slightly, but significantly (P < 0.001) higher (0.30–0.46%) at the shrub tundra, indicating that virtually all added 15N was incorporated into either microbes, plants or the insoluble soil organic matter pool.
The plants took up 15N label from all added N forms. The 15N concentration in plants was generally higher in Carex than in Betula and Empetrum, and higher at the shrub than at the heath tundra except in Carex, which had significantly higher concentrations at the heath.
Across the sites, the 15N recovery in plants generally increased from day 2 to day 26 (Table 3). Also, the 15N plant recovery differed between sites and between the different 15N treatments (Table 3). At the shrub tundra, the 15N recovery in both roots and leaves of Carex were significantly higher in the 15NO3− treatment than in the other treatments (P < 0.001 and P < 0.0001 for roots and leaves, respectively; one-way ANOVA), and at the heath tundra we found the same treatment effect in both Carex roots (P < 0.001) and leaves (P < 0.0001) separately, and also for total Carex recovery (Fig. 3a). In Empetrum at the shrub site after 2 d, the 15N recovery was similar in the NH4+ and NO3− treatment and near significantly lower in the glycine treatment (Fig. 3b) (P = 0.052, total Empetrum recovery). Betula took up more 15NH4+ than 15NO3− and 15N-glycine at the heath tundra, and this preference pattern was also evident at the shrub tundra, but only after 26 d. At the shrub tundra there was a proportionally higher uptake of 15NO3− than 15N-glycine, compared to the heath site, although this was not significant (Fig. 3c). In the mixed species, tested across the harvest occasions, there were significant effects of treatments at both sites, with lower uptake of 15N-glycine than of the inorganic N-forms and lower recovery of 15N-glycine than of 15NH4+ at the shrub tundra (Table 3, Fig. 3d).
P-values from three-factor ANOVA of recovery of 15N in different plant fractions sampled at the heath tundra and the shrub tundra 2 (July) and 26 (August) days after labeling. P-values are for main factor effects: Site (S), Treatment (T) and Harvest (H) and interactions between main effects. Only significant (P ≤ 0.05) or near-significant (P ≤ 0.1) values are included; n.s. not significant (P > 0.1).
The different species and growth forms allocated 15N into the various plant compartments in different manners. However, there were no significant differences in the relative 15N allocation patterns of the three N types into plant compartments within species (two-way ANOVAs). Already 2 d after labeling, Carex had allocated a large part of the absorbed 15N to the aboveground parts (Fig. 3a), and 26 d after labelling, the distribution of 15N had approached that of the total Carex N pool (Fig. 1a). In Empetrum and Betula, the fine roots contained 30 to 70% of the total absorbed 15N 2 d after labeling and this proportion decreased only slightly with time (Fig. 3b-c). At both sites, the fraction of 15N allocated to Betula leaves had doubled after 26 d, although it made up only about 12% of the total 15N uptake, while the N fraction in Betula leaves was about 25% of the total N pool (Fig. 1c). Empetrum shows a similar pattern with only about 20% of the absorbed 15N allocated to the leaves (Fig. 3b), while 40% of the total Empetrum N pool was in the leaves (Fig. 1b).
Based on the data on 15N-glycine uptake 2 d after labeling, we calculated the δ13C values that the plant root theoretically would obtain if uptake of glycine was in intact form. These δ13C values were compared with the natural δ13C values from the control plots and the δ13C values measured in the 15N-2(13C)-glycine treated plots (Table 4). Only in fine roots of Betula at the heath tundra we identified a significantly higher δ13C value in the 15N-2(13C)-glycine treatment than in the control plots. Here we found a 13C:15N-ratio of 1.78 ± 0.38. However, there was a clear tendency for all plants to be more enriched with 13C in treated than in control plots.
δ13C-values in roots of control plots, of glycine treated plots two days after labeling and the theoretically δ13C-values calculated under the assumption of 100% uptake of intact glycine. Species means that do not share the same letters are significantly different (P ≤ 0.05; Tukey's test).
Recovery of 15n Among Plants, Microbes and Som
At both the heath and the shrub tundra, there were significant effects of treatment on incorporation of label into plants (Fig. 4). Two days after labeling, the plants had incorporated about 6% of the added 15NO3− and 15NH4+ at the heath and about 10% at the shrub tundra but only about 3 and 4% of the 15N-glycine at the two tundra types. After 26 d, the proportion recovered in the plants at the heath had increased significantly to 9, 7, and 4% in the 15NH4+, 15NO3− and 15N-glycine treated plots while there was no significant change at the shrub tundra.
Most of the recovered label was incorporated in the soil organic matter, with no significant change of the recovery between the early and late harvest. There was, however, a significant treatment effect at the shrub tundra of more recovered 15N in SOM of 15NH4+ than 15NO3− and 15N-glycine treated plots 26 d after labeling. Two days after label addition, a large part of the 15N was recovered in the extractable microbial biomass, but the proportion had decreased significantly by about 50% at day 26. While there was no significant treatment effect at the heath, less 15N was recovered in the extractable microbial biomass in the 15NO3− than in the 15NH4+ and 15N-glycine treatments at the shrub tundra. Despite the reduced proportion recovered in the extractable microbial biomass between day 2 and 26, the 15N recovery in microbes in 15NH4+ and 15N-glycine treated plots exceeded the 15N recovery in plants also at day 26, while the microbial recovery of 15N in 15NO3− treated plots did not exceed the 15N recovery in plants.
N Uptake, Allocation, and Uptake Preferences Related to Plant Species and Site Conditions
Within 2 d after labeling, the plants had absorbed N from all added sources, i.e. also from the organic glycine source as direct glycine uptake and/or as mineralized N. The difficulties in detecting uptake of organic N is well known and is due to large dilution of an expected and proportionally small uptake in comparison with the large 13C pool already present in the tissues (Näsholm and Persson, 2001), as also observed in the present experiment (Table 4). Despite the dilution, we yet detected a significant difference between δ13C in Betula roots from controls and glycine treated plots at the heath tundra (Table 4), which is the first field evidence of uptake of intact glycine by the important circumpolar arctic dwarf shrub Betula nana. The 13C:15N-ratio of 1.78 ± 0.38 in Betula roots from the heath site 2 d after glycine addition is close to the 13C:15N-ratio of 2 in the added glycine, which points to a high degree of intact glycine uptake. In addition, the trend towards higher δ13C-values in roots from glycine treated than from control plots for all species at both sites indicates uptake of intact glycine also among the other species. The uptake of glycine-derived 15N 2 d after labeling, and the fast exhaustion of 15N from the dissolved soil N pools in all treatments, also shows that the plants are able to take up glycine-N either directly in competition with free-living microorganisms or after fast microbial mineralization taking place within a short time after the glycine is supplied. However, at both sites the plants generally had lower initial recovery of 15N in the glycine treatment than in the inorganic 15N treatments (Fig. 3). In the only other field study of N form preferences after injections of mixed N forms to an arctic ecosystem that we are aware of, the uptake of amino acid-N sources also was lower than uptake of added inorganic N sources (Nordin et al., 2004)
There was a tendency for Carex to take up more NO3− than NH4+, while both Betula and the mixed species generally showed increased preference in the order NH4+ ≥ NO3− ≥ glycine. Other labeling studies in northern ecosystems also have shown preference for uptake of NO3− over uptake of NH4+ and glycine among graminoids, e.g. Carex bigelowii in an Alaskan tussock tundra (McKane et al., 2002), Eriophorum vaginatum in an acidic tussock tundra (Nordin et al., 2004) and Deschampsia flexuosa in a boreal forest (Persson et al., 2003). One explanation to the observed high graminoid NO3− uptake could be that at our study sites as in the studies cited above, the biomass of graminoids was low compared to the biomass of dominant deciduous and evergreen shrubs. The graminoids may therefore have taken up NO3− because the dominant shrubs tend to acquire NO3− at slower rates resulting in a relative increase of the NO3− available for uptake (McKane et al., 2002). However, in our experiment, virtually all the added label was taken up within the first 2 d after labeling, and the availability of all N forms therefore should be equally high even shortly after labeling. Furthermore, graminoids are most often dominant in relatively nutrient-rich sites where availability of NO3− is high, and observations of high nitrate reductase activity in arctic graminoid species compared to other plant growth forms (Michelsen et al., 1996) point to high uptake capacity for NO3−.
Compared to the initial uptake rate of label into the plant pool during the 2 d following the addition, the rate during the following 24 d declined (Figs. 3, 4). In Alaskan tussock tundra, Schimel and Chapin (1996) found the highest uptake rate within 5 d after labeling, followed by slight increases in the plant 15N recovery during the following 25 d. Likewise, Grogan and Jonasson (2003) found some incorporation of label into evergreen plants 4 to 7 d after labeling in October, followed by no uptake during the winter, but a subsequent high uptake by all plant growth forms from May to July the following year. There can be at least two explanations for this pattern of uptake rates. First, the uptake rate in our study was high immediately after label addition because of the pulse of available N, but decreased rapidly as most of the added N was efficiently incorporated in the extractable microbial biomass and the SOM pool within a few days. Thus, the 15N released from microbial biomass over the following 24 d was diluted by the N pool already contained in microbial biomass and by the native N pool processed by the microbes over this period. This would lead to a substantial pool dilution of the available soil 15N pool and to underestimated plant N uptake based on 15N uptake alone. Second, the plants take up N mainly in the beginning and middle of the growing season, and not late in the season (August). For instance, McKane et al. (2002) observed that most plant species took up more N after injection of label in June than after injection of label in August. However, Betula nana took up more label in August than in June (McKane et al., 2002), a pattern also found in our study, as the 15N recovery in Betula at the shrub tundra, but not at the heath tundra, had doubled from the first to the second harvest.
Despite the differences in vegetation type and plant biomass, and therefore assumed differences in nutrient supply rates between the two sites, we did not find any differences between the relative availability of three N forms at the two sites (Table 2), contrary to reported differences along a nutrient gradient of a boreal forest (Nordin et al., 2001). However, reports on fast turnover of soil nutrients in arctic soils (hours to a few days) (Jackson et al., 1989; Jones and Kielland 2002) suggest that nutrient concentrations are poorly related to nutrient availability, which may depend more on nutrient fluxes. Because of the apparent lack of difference in relative supply of the different N forms between the two sites, we assumed that the uptake preferences would not differ between sites. However, we did find a tendency of higher preference for NO3− uptake in Betula at the shrub tundra than at the heath tundra. Also, at our sites, the leaf natural abundance of δ15N in Betula was −6.24 ± 0.56 ‰ at the heath and higher, −4.10 ± 0.52 ‰, at the shrub tundra. Hobbie et al. (2000) found that plant foliar δ15N was lower at sites with low N availability than at sites with high N availability over a range of sites at Glacier Bay, Alaska, mainly owing to higher dependency on N supply from mycorrhiza at sites with low N availability. Furthermore, ectomycorrhizal fungi seem to discriminate against uptake of NO3− (Finlay et al., 1992; Gessler et al., 2005). Therefore, the higher δ15N values and higher Betula preference for NO3− at the shrub than at the heath tundra could be caused by a combination of higher inorganic nutrient availability and lower relative N supply via mycorrhiza at the wet shrub tundra than at the drier heath, as the nutrient availability was higher at this site.
The two Carex species translocated a much larger part of the acquired 15N to the leaves than the dwarf shrubs already two days after label addition, and after 26 d, more than half of the 15N in Carex was recovered in the leaves (Fig. 3a). Other field (Schimel and Chapin, 1996; Grogan and Jonasson, 2003) and mesocosm (Grogan et al., 2004) studies of 15N uptake and allocation also showed high allocation of recently acquired 15N to leaves in graminoids and low allocation to leaves of shrubs. High growth rates and thereby a demand for high photosynthetic rate in Carex may drive the fast allocation to leaves, while Betula and Empetrum give priority to long-term N storage.
15n Partitioning among Microbes, Plants, and Soil Organic Matter
At the shrub tundra, the microbes showed a preference for glycine over the other N forms, opposed to the plants, which took up more of the inorganic N-forms. These results support the theory on complementary N-form preferences by microbes and plants (Lipson and Näsholm, 2001). The microbes probably take up glycine at higher rates than the inorganic N-forms as it acts as a source of C for the generally substrate-limited microbes (Jonasson et al., 1996; Schmidt et al., 1997).
The proportion of added 15N taken up by plants was low compared with the microbial uptake, and of the same order of magnitude as in reports from other arctic ecosystems (Schimel and Chapin, 1996; McFarland et al., 2002; Grogan and Jonasson, 2003). In the extractable microbial biomass, we observed a high initial 15N uptake (Fig. 4), which confirms that retention of nutrients in microbes play a crucial role in the control of N cycling in arctic ecosystems (Schimel and Chapin, 1996; McFarland et al., 2002; Grogan and Jonasson, 2003; Henry and Jefferies, 2003a; Nordin et al., 2004). However, after 26 d, the extractable microbial 15N pool was reduced to half the size it had after two days. Furthermore, the proportion of 15N incorporated in the soil organic matter pool strongly increased (Fig. 4), a pattern that agrees with findings in other labelling studies (Marion et al., 1982; McFarland et al., 2002; Grogan et al., 2004). Some of the decline in the microbial 15N pool could be a result of methodological overestimation of initial 15N uptake in microbes, as the 15N incorporated in the microorganisms after 2 d may be more chloroform labile and, hence, with higher extractability than the microbial 15N 26 d after label addition. Also, a high proportion of fungal derived microbial biomass in the soil may lead to an underestimation of the microbial biomass, as fungi in some cases seem to be less extractable by the chloroform fumigation than bacteria (Ingham et al., 1991; Wallander et al., 2003). This could explain part of the high incorporation of 15N into the SOM pool. However, most of the change in the microbial 15N pool is probably a consequence of rapid microbial turnover of 15N and subsequent incorporation into a more recalcitrant SOM pool, eventually approaching the same proportion of a few percent of the bulk soil N incorporated in the microbes (Fig. 2). Because we observed little plant 15N uptake between day two and 26 after label addition, it appears that the label once it is incorporated in the bulk soil is only slowly released to plant available forms.
Our study has shown, first, that all co-existing plant species did take up all added N-forms but all species generally showed higher preference for the inorganic forms, either NH4+ or NO3−, than for N from the glycine source. The shrub species Betula preferentially absorbed NH4+ while Carex preferred NO3−. Second, we found a tendency for higher preference for NO3− than for NH4+ at the shrub tundra compared to the heath in Betula, probably caused by lower dependency of uptake via mycorrhiza than at the nutrient-poor heath tundra. Third, the Carex species translocated a higher proportion of the absorbed 15N to the aboveground parts than the shrubs, supposedly for immediate photosynthetic use, while the shrubs gave priority to longer-term N storage. Fourth, the microbes showed a preference for the organic N form while the plants preferentially took up inorganic N, which could release the competition for resources between plants and microbes. Fifth, we found that microbial 15N immobilization was high immediately after the label addition, whereas after 26 d, the 15N recovered in the microbes was reduced to half of the amount found 2 d after label addition, and most of the label was recovered in the nonmicrobial part of the soil. During the same period, the plant 15N uptake rate decreased strongly compared with the rate within the 2 d following label addition, and the 15N recovery was, hence, similar (NO3−) or lower (NH4+ and glycine) in plants than in microbes 4 wk after N addition.
Abisko Scientific Research Station (ANS) provided excellent facilities and logistic support. Pia Lund Nielsen, Jane Kongstad Pedersen, Jette Brandt, Anders Tesgaard, Kasper Andersen, and Peter Lindberg Jannerup assisted in the field, and Karin Larsen and Esben Vedel Nielsen assisted with the laboratory analyses. Inger Kappel Schmidt kindly provided advice on the 15N-diffusion technique. We would like to thank two anonymous reviewers for constructive comments to the manuscript. The research was funded by ANS, The Faculty of Science, University of Copenhagen and The Danish Natural Science Research Council.
- J. P. E. Anderson and K. H. Domsch . 1977. Mineralization of bacteria and fungi in chloroform-fumigated soils. Soil Biology and Biochemistry 10:207–213. Google Scholar
- R. D. Bardgett, T. Streeter, and R. A. Bol . 2003. Soil microbes compete effectively with plants for organic-nitrogen inputs to temperate grasslands. Ecology 84:1277–1287. Google Scholar
- P. C. Brookes, A. Landman, G. Pruden, and D. S. Jenkinson . 1985. Chloroform 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. Google Scholar
- F. S. Chapin III, L. Moilanen, and K. Kielland . 1993. Preferential use of organic nitrogen for growth by a non-mycorrhizal arctic sedge. Nature 361:150–153. Google Scholar
- F. S. Chapin III and G. R. Shaver . 1996. Physiological and growth responses of arctic plants to a field experiment simulating climatic change. Ecology 77:822–840. Google Scholar
- W. Cheng and R. A. Virginia . 1992. Measurement of microbial biomass in arctic tundra soils using fumigation-extraction and substrate-induced respiration procedures. Soil Biology and Biochemistry 25:135–141. Google Scholar
- R. D. Finlay, Å Frostegård, and A-M. Sonnerfeldt . 1992. Utilization of organic and inorganic nitrogen sources by myccorhizal and nonmycorrhizal fungi in pure culture and in symbiosis with Pinus contorta Dougl. ex Loud. New Phytologist 120:105–115. Google Scholar
- A. K. Gessler, K. Jung, R. Gasche, H. Papen, A. Heidenfeler, E. Borner, B. Metzler, S. Augustin, E. Hildebrand, and H. Rennenberg . 2005. Cilmate and forest management influence nitrogen balance of European beech forests: microbial N transformations and inorganic N net uptake capacity of mycorrhizal roots. European Journal of Forest Research 124:95–111. Google Scholar
- L. G. Greenfield 1995. Release of microbial cell N during chloroform fumigation. Soil Biology and Biochemistry 27:1235–1236. Google Scholar
- P. Grogan and S. Jonasson . 2003. Controls on annual nitrogen cycling in the understory of a subarctic birch forest. Ecology 84:202–218. Google Scholar
- P. Grogan, A. Michelsen, P. Ambus, and S. Jonasson . 2004. Freeze-thaw regime effects on carbon and nitrogen dynamics in sub-arctic heath tundra mesocosms. Soil Biology and Biochemistry 36:641–654. Google Scholar
- P. Grogan and S. Jonasson . 2005. Temperature and substrate controls on intra-annual variation in ecosystem respiration in two subarctic vegetation types. Global Change Biology 11:465–475. Google Scholar
- H. A. L. Henry and R. L. Jefferies . 2003a. Interactions in the uptake of amino acids, ammonium and nitrate ions in the Arctic salt-marsh grass, Pucinella phryganodes. Plant, Cell and Environment 26:419–428. Google Scholar
- H. A. L. Henry and R. L. Jefferies . 2003b. Plant amino acid uptake, soluble N turnover and microbial N capture in soils of a grazed Arctic salt marsh. Journal of Ecology 91:627–636. Google Scholar
- S. E. Hobbie 1992. Effects of plant species on nutrient cycling. Trends in Ecology and Evolution 7:336–339. Google Scholar
- E. A. Hobbie, S. A. Macko, and M. Williams . 2000. Correlations between foliar δ15N and nitrogen concentrations may indicate plant-mycorrhizal interactions. Oecologia 122:273–283. Google Scholar
- A. Hodge, D. Robinson, and A. Fitter . 2000. Are microorganisms more effective than plants for competing nitrogen. Trends in Plant Science 5:304–308. Google Scholar
- E. R. Ingham, R. P. Griffiths, K. Cromack, and J. A. Entry . 1991. Comparison of direct vs. fumigation incubation microbial biomass estimates from ectomycorrhizal mat and non-mat soils. Soil Biology and Biochemistry 23:465–471. Google Scholar
- R. B. Jackson, J. P. Schimel, and M. K. Firestone . 1989. Short-term partitioning of ammonium and nitrate between plants and microbes in an annual grassland. Soil Biology and Biochemistry 21:409–415. Google Scholar
- S. Jonasson, A. Michelsen, I. K. Schmidt, E. V. Nielsen, and T. V. Callaghan . 1996. Microbial biomass C, N and P in two arctic soils and responses to addition of NPK fertilizer and sugar: implications for plant nutrient uptake. Oecologia 106:507–515. Google Scholar
- S. Jonasson and G. R. Shaver . 1999. Within-stand nutrient cycling in arctic and boreal wetlands. Ecology 80:2139–2150. Google Scholar
- S. Jonasson, A. Michelsen, I. K. Schmidt, and E. V. Nielsen . 1999. Responses in microbes and plants to changed temperature, nutrient and light regimes in the Arctic. Ecology 80:1828–1843. Google Scholar
- D. L. Jones and K. Kielland . 2002. Soil amino acid turnover dominates the nitrogen flux in permafrost-dominated taiga forest soils. Soil Biology and Biochemistry 34:209–219. Google Scholar
- J. P. Kaye and S. C. Hart . 1997. Competition for nitrogen between plants and soil microorganisms. Tree 12:139–143. Google Scholar
- K. Kielland 1994. Amino acid absorption by arctic plants: implications for plant nutrition and nitrogen cycling. Ecology 75:2373–2383. Google Scholar
- D. A. Lipson and T. Näsholm . 2001. The unexpected versatility of plants: Organic nitrogen use and availability in terrestrial ecosystems. Oecologia 128:305–316. Google Scholar
- M. C. Mack, E. A. G. Schuur, M. S. Bret-Harte, G. R. Shaver, and F. S. Chapin III . 2004. Ecosystem carbon storage in arctic tundra reduced by long-term nutrient fertilization. Nature 431:440–443. Google Scholar
- G. M. Marion, P. C. Miller, J. Kummerow, and W. C. Oechel . 1982. Competition for nitrogen in a tussock tundra ecosystem. Plant and Soil 66:317–327. Google Scholar
- J. W. McFarland, R. W. Ruess, K. Kielland, and A. P. Doyle . 2002. Cycling dynamics of NH4+ and amino acid nitrogen in soils of a deciduous forest ecosystem. Ecosystems 5:775–788. Google Scholar
- R. B. McKane, L. C. Johnson, G. R. Shaver, K. J. Nadelhoffer, E. B. Rastetter, B. Fry, A. E. Giblin, K. Kielland, B. L. Kwiatkowski, J. A. Laundre, and G. L. Murray . 2002. Resource-based niches provide a basis for plant species diversity and dominance in arctic tundra. Nature 415:68–71. Google Scholar
- A. Michelsen, I. K. Schmidt, S. Jonasson, C. Quarmby, and D. Sleep . 1996. Leaf 15N abundance of subarctic plants provides field evidence that ericoid, ectomycorrhizal and non- and arbuscular mycorrhizal species access different sources of soil nitrogen. Oecologia 105:53–63. Google Scholar
- P. C. Miller 1982. Environmental and vegetational variation across a snow accumulation area in montane tundra in central Alaska. Holoarctic Ecology 5:85–98. Google Scholar
- A. E. Miller and W. D. Bowman . 2003. Alpine plants show species-level differences in the uptake of organic and inorganic nitrogen. Plant and Soil 250:283–292. Google Scholar
- C. D. Monk 1966. An ecological significance of evergreenness. Ecology 47:504–505. Google Scholar
- T. Näsholm, A. Ekblad, A. Nordin, R. Giesler, M. N. Högberg, and P. Högberg . 1998. Boreal forest plants take up organic nitrogen. Nature 392:914–916. Google Scholar
- T. Näsholm and J. Persson . 2001. Plant acquisition of organic nitrogen in boreal forests. Physiologia Plantarum 111:419–426. Google Scholar
- A. Nordin, P. Högberg, and T. Näsholm . 2001. Soil nitrogen form and plant nitrogen uptake along a boreal forest productivity gradient. Oecologia 129:125–132. Google Scholar
- A. Nordin, I. K. Schmidt, and G. R. Shaver . 2004. Nitrogen uptake by arctic soil microbes and plants in relation to soil nitrogen supply. Ecology 85:955–962. Google Scholar
- J. Persson, P. Högberg, A. Ekblad, M. N. Högberg, A. Nordgren, and T. Näsholm . 2003. Nitrogen acquisition from inorganic and organic sources by boreal forest plants in the field. Oecologia 137:252–257. Google Scholar
- J. P. Schimel and F. S. Chapin III . 1996. Tundra plant uptake of amino acid and NH4+ nitrogen in situ: plants compete well for amino acid N. Ecology 77:2142–2147. Google Scholar
- I. K. Schmidt, A. Michelsen, and S. Jonasson . 1997. Effects of labile soil carbon on nutrient partitioning between an arctic graminoid and microbes. Oecologia 112:557–565. Google Scholar
- J. M. Stark and S. C. Hart . 1996. Diffusion technique for preparing salt solutions, Kjeldahl digests, and persulfate digests for Nitrogen-15 analysis. Soil Science Society of America 60:1846–1855. Google Scholar
- H. Wallander, L. O. Nilsson, D. Hagerberg, and U. Rosengreen . 2003. Direct estimates of C:N ratios of ectomycorrhizal mycelia collected from Norway spruce forest soils. Soil Biology and Biochemistry 35:997–999. Google Scholar
- A. Weigelt, R. Bol, and R. D. Bardgett . 2005. Preferential uptake of soil nitrogen forms by grassland plant species. Oecologia 142:627–635. Google Scholar
- M. N. Weintraub and J. P. Schimel . 2003. Interactions between carbon and nitrogen mineralization and soil organic matter chemistry in arctic tundra soils. Ecosystems 6:129–143. Google Scholar
- J. B. Zhou, Z. J. Chen, and S. X. Li . 2003. Oxidation efficiency of different oxidants of the persulfate method to determine total nitrogen and phosphorus in solutions. Communications in Soil Science and Plant Analysis 34:725–734. Google Scholar