SSSAJ Grow Your Career with SSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (12)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Barrett, J. E.
Right arrow Articles by Burke, I. C.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Barrett, J. E.
Right arrow Articles by Burke, I. C.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Barrett, J. E.
Right arrow Articles by Burke, I. C.
Related Collections
Right arrow Nitrogen
Right arrow Soil Organic Matter
Right arrow Range Soils
Soil Science Society of America Journal 66:979-987 (2002)
© 2002 Soil Science Society of America

DIVISION S-7—FOREST & RANGE SOILS

Abiotic Nitrogen Uptake in Semiarid Grassland Soils of the U.S. Great Plains

J. E. Barrett*,a, D. W. Johnsonb and I. C. Burkec

a Environmental Studies Program, 6182 Steele Hall, Dartmouth College, Hanover, NH 03755
b Desert Research Institute and Environmental and Resource Sciences, Univ. of Nevada, Reno, NV 89512
c Department of Forest Sciences and the Natural Resource Ecology Laboratory, Colorado State University Fort Collins, CO 80523

* Corresponding author (John.E.Barrett{at}Dartmouth.edu)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Soils represent the largest terrestrial sink for N, yet current understanding of nutrient cycling cannot account for some of the mechanisms and sinks that stabilize anthropogenic N. We assessed the influence of soil properties, particularly soil organic C, pH, and clay content on potential biotic and abiotic N assimilation in soils collected across a temperature gradient in the U.S. Great Plains. We pulse labeled HgCl2-sterilized and unsterilized soils with 15N to examine the relative importance of abiotic and biotic N assimilation in short-term laboratory incubations. Estimates of total N assimilation in unsterilized soils ranged from 1.21 to 2.40 mg N kg-1 soil. Soil C content accounted for 50 and 60% of the variance in estimates of biotic immobilization and total N assimilation, respectively. Estimates of abiotic N assimilation ranged from 0.089 to 0.80 mg N kg-1 soil. Abiotic N uptake represented a large proportion of total N assimilation (mean equals 20%) in short-term laboratory incubations. In contrast to previous reports, abiotic N uptake was negatively correlated to soil clay content and pH, perhaps because of differences in mineralogy and soil organic matter composition across the gradient. These results emphasize the importance of nonbiological N uptake in semiarid soils and suggest that abiotic pools could be an important sink for N.

Abbreviations: {delta}, natural abundance


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
SOIL ORGANIC MATTER accounts for the majority of N in terrestrial ecosystems and most of the N retained in elevated N deposition and fertilizer studies (Aber et al., 1998; Fenn et al., 1998; Nadelhoffer et al., 1999), yet it has been difficult to identify the specific mechanisms or pools responsible for N stabilization. Most biogeochemical models assume that N inputs are incorporated into detrital, microbial, and plant pools by microbes and plants with concomitant fluxes of C reflective of their respective structural requirements (i.e., their C/N ratios). However, recent studies indicate that significant amounts of added N may be stabilized without observed increases in microbial biomass, soil respiration, or plant productivity (Aber et al., 1998; Fenn et al., 1998). For example, Aber et al. (1998) reported high rates of N recovery in northeastern forest soils without observed increases in microbial biomass, soil respiration, or root biomass. Agronomic studies have long recognized the importance of abiotic stabilization of fertilizer-N additions (Burge and Broadbent, 1961; Mortland and Wolcott, 1965; Nömmik 1965), and rapid abiotic uptake has been noted in pool dilution studies (Davidson et al., 1991; Bernston and Aber 2000); yet abiotic mechanisms of N uptake have received little attention in discussions of natural N cycling or N saturation until quite recently (Johnson et al., 2000; Dail et al., 2001).

Abiotic N stabilization can account for a large proportion of NH+4, NO-2, and NO-3 retention in studies from a broad survey of temperate forests utilizing a variety of sterilization techniques (Davidson et al., 1991; Johnson et al., 2000; Dail et al., 2001). Davidson et al. (1991) reported that rapid, presumably abiotic mechanisms could account for up to 30% of the N assimilation during 15N-pool dilution assays. Johnson et al. (2000) reported that abiotic assimilation of NH+4 can be a significant process in a variety of soil types (from 6 to 90% of total immobilization) and may be indicative of N deposition history and degree of saturation. Dail et al. (2001) showed that NO-2 and NO-3 were stabilized directly onto organic matter in sterilized forest soils, and that a relatively large fraction of N was abiotically fixed into an unidentified soluble-organic fraction. Indirect evidence from grassland ecosystems suggests that a large proportion (up to ~50%) of NH+4 additions retained in soils is not actively cycled in plant or microbial pools but becomes rapidly (1 yr) incorporated into slowly cycling and unavailable forms of soil organic matter (Barrett and Burke, 2002, in press; Kaye et al., 2002, in press), suggesting that abiotic mechanisms may be generally important in stabilizing different forms of inorganic N over a variety of temperate ecosystem.

Known mechanisms of abiotic N assimilation include the following types of reactions: ionic substitution of NH+4 in 2:1 clay minerals (Young and Aldag, 1982; Stevenson 1994); reduction of NO-2 by humic substances at low soil pH (Stevenson, 1994); and condensation of amino acids or NH3 with phenolic compounds, contributing to humification (Mattson and Koulter-Andersson, 1942, 1943; Burge and Broadbent, 1961; Mortland and Wolcott, 1965; Nömmik, 1965, 1970; Nömmik and Vahtras, 1982; He et al., 1988; Thorn and Mikita, 1992). To test these influences of soil organic matter and clay content on potential abiotic N immobilization, we performed laboratory incubations on 15N-labeled sterilized and nonsterilized soils of varying clay and organic matter content. Our objective was to assess the relative importance of biotic and abiotic assimilation of an 15NH+4 tracer in soils collected over an environmental gradient. We hypothesized that abiotic NH+4 assimilation would increase as a function of organic matter and clay content over the gradient.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
Sites
We selected a series of ten sites in the semiarid region of the U.S. Great Plains extending from the Panhandle of Texas northward into southeastern Montana (Table 1). The six southern sites are located in shortgrass steppe and the four northern sites are in the northern mixed prairie according to Dodd (1979) and Epstein et al. (1996). Sites were selected for a range of textures to address the influence of clay on N assimilation. Soil textures ranged from sandy loams to clay over the sites sampled. General soil classifications are given in Table 1.


View this table:
[in this window]
[in a new window]
  		Table 1. Location and soil characteristics of study sites. Mean annual temperature (MAT) and mean annual precipitation (MAP) are 30-yr averages from the closest meteorological stations. Soil organic C (SOC), sand and clay content of the top 10 cm are means of six replicates for each soil types (n = 60).

 
Climate data for each site are from the closest available meteorological station with continuous climate records of at least 30 yr. We attempted to pair fine- and coarse-textured soils with one representative meteorological station. Where this was not possible (where coarse- and fine-textured soils could not be located within 5 miles of one another) we used separate meteorological stations for each site. Mean annual temperature decreased from 14.3°C in the south to 7.12°C in the north and mean annual precipitation ranged from 45 cm in the south to 35 cm in the north. Reported estimates of evapotranspiration for this region are highest in the south, corresponding to the highest mean annual temperature and precipitation (Sims et al., 1978; Webb et al., 1983).

Estimates of aboveground net primary production based upon both simulation modeling and remote sensing range from 150 to 200 g C m-2, depending upon interannual and seasonal variability in precipitation, but do not vary markedly across the temperature gradient (Parton et al., 1989; Sala et al., 1988; Paruelo et al., 1997; Burke et al., 1997). Soil C content and composition vary widely throughout this region across soils and landscapes, but are broadly influenced by mean annual temperature and precipitation resulting in latitudinal gradients in soil C content and composition (Table 1) (Parton et al., 1987; Burke et al., 1989a; Amelung et al., 1997; Amelung et al., 1998).

Sampling Protocol
We collected soil samples from each site in August, 1998. We collected six cores to a depth of 10 cm from randomly determined points on a 10-m transect at each site (n = 60). Soil samples were placed into polyethylene bags and stored at 4°C for transport back to the Grassland Biogeochemistry Laboratory at Colorado State University in Fort Collins, CO. Soils were processed in the laboratory using a 2-mm sieve to remove litter, large roots, and rocks prior to soil analyses and incubation. We weighed a subsample of each soil sample and dried it at 50°C and then reweighed the subsample for gravimetric soil moisture determination. We measured soil pH in a saturated paste extract (1:2 soil in deionized water) with an Orion model 720A pH meter(Orion Research Inc., Boston, MA). A 10-g subsample was extracted in 50 mL of 2.0 M KCl-Phenyl mercuric acetate (5 ppm) for 30 min on an orbital shaker, filtered through Whatman #40 paper (Whatman Ltd, Maidstone, UK) and analyzed on an Alpkem Autoanalylzer (Perstorp Analytical, Silver Springs, MD) to determine inorganic concentrations. We estimated soil texture on a subsample of each sieved soil sample using a hydrometer method (Gee and Bauder, 1986). A subsample was ground in a ball mill and analyzed for total C and N content using a LECO CHN-1000 analyzer (LECO, St. Joseph, MI). We tested soils for the presence of carbonates with 50% HCl. Soils with a significant carbonate content were acidified with 50% HCl, and oven-dried prior to elemental analysis. All values reported for C are for soil organic C. Subsamples of ground, preacidified soils were sent to the Stable Isotope Laboratory at Utah State University, Logan, UT for determination of 15N natural abundance ({delta}15N) of the bulk soil.

Incubations
We used a subset of these soils (n = 30) in series of incubations in June of 1999 to estimate 15N assimilation in sterilized and unsterilized soils at the Desert Research Institute Laboratory in Reno, NV following a technique described by Johnson et al. (2000). To distinguish biological processes from abiotic mechanisms we used a 5% (wt/wt) HgCl2 solution to sterilize soils. This constituted our abiotic treatment. We selected this method because HgCl2 is persistent in the soil and maintains its effectiveness throughout the course of an experiment (whereas radiation may not), and HgCl2 produces little change in the properties of organic matter, as autoclaving might (Wolf and Skipper, 1994; Johnson et al., 2000). In testing several methods of soil sterilization, Johnson et al. (2000) found that HgCl2 was completely effective in preventing soil respiration throughout the course of a laboratory incubation, without altering the properties of organic matter.

We pulse labeled sterilized and unsterilized soils with 25 mL of a 8 x 10-5 M solution of 73.6 atom % 15N-(NH4)2SO4 to estimate rapid 15N assimilation. This level of N addition averaged 10 mg N g-1 total soil N and increased the total inorganic N pool (N-NO-3 + N-NH+4) by 140% and tripled the N-NH+4 pool on average across the sites. Applications of 15N were pipetted onto triplicate 2.0-g subsamples of each soil in 25-mL aliquots and incubated at 21°C for 12 h under low tension in a mechanical vacuum extractor (Centurion International, Lincoln, NE) to slowly percolate the 15N cocktail through the soil and prevent accumulation of mineralized 14NH+4. The incubation procedure was identical for sterilized soils except that the 15N cocktail contained 5% (wt/wt) HgCl2 to preclude any biological activity or N uptake. Molarity of the 15N cocktail was adjusted for the presence of the HgCl2 to ensure an equal application of 15N to the sterilized and unsterilized soils.

At the end of the 12-h incubation period, we extracted the incubating soils with 25 mL of 2 M KCl in a mechanical vacuum extractor for 4 h to remove readily exchangeable NH+4 and NO-3. Extraction of soil solution under tension is effective at flushing the unstabilized inorganic N without extracting fixed NH+4 (Johnson et al., 2000). Potassium chloride extracts were discarded and the solid soil samples immediately processed in preparation for 15N analysis. The incubated soils were immediately recovered, oven dried at 60°C, and ground on a ball mill in preparation for 15N analyses. Analyses of 15N were conducted at the University of California, Davis, CA using a Europa Scientific Integra analyzer (Crewe, UK).

We calculated recovery of the 15N application using the following formula developed by Hauck and Bremner (1976) and modified by Powlson and Barraclough (1993):

where N content is the total N content of the sample in grams of N per kilogram of soil dry weight equivalent, 15Nsample is the measured percentage of 15N content of the sample, 15Nnatural abundance is the natural abundance percentage of 15N content of the soil, and 15Nlabel is the percentage of 15N content of the N application (0.736). Biotic N immobilization was calculated as the difference in 15N assimilation between unsterilized (samples treated with 15N-[NH4]2SO4 only) and sterilized soils (those treated with 15N-[NH4]2SO4 plus HgCl2).

Statistical Analyses
We used analyses of variance (ANOVA) and a post hoc comparison of the least significant differences among means to test for differences in soil properties and 15N assimilation among the sites. We calculated Pearson correlation coefficients for total, biotic, and abiotic N assimilation, soil-clay content, soil organic matter content, soil pH, and KCl-extractable NH+4 content to test for relationships between potential 15N assimilation and soil properties in sterilized and unsterilized soils. Based upon the Pearson correlation matrix, we chose the soil properties that accounted for the largest proportion of variance in the dependent variables and performed a series of simple linear regressions. All analyses were done using the SAS statistical software package v 8.0. Significance for all tests was accepted at P < 0.05.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
Soil Properties
In general soil organic matter increases across this gradient (Table 1). Northern and fine-textured soils generally have the highest levels of soil organic matter in this region (Burke et al., 1989a; Amelung et al., 1997, 1998; Barrett and Burke, 2000, 2002 in press). Soil N content varied significantly among the sites (Fig. 1a) and was significantly correlated with clay content of the soil (Table 2). Soil extractable-NH+4 concentrations also increased northward (Fig. 1b) and were strongly correlated with soil organic C (Table 2). The correlation between NH+4 and total soil N while significant, was weak (Table 2) suggesting that soil NH+4 concentration is not controlled solely by N availability and first-order kinetics.



View larger version (53K):
[in this window]
[in a new window]
  		Fig. 1. (a) Total soil N, (b) KCl-extractable N-NH+4 content, and (c)15N natural abundance for bulk soils collected from across an environmental gradient in semiarid grasslands of the U.S. Great Plains. Bars with the same letter are not significantly different at P = 0.05. Error bars are one standard error of the mean. Site codes are as follows: TX-1 and TX-2, Muleshoe National Wildlife Refuge, TX; SECO-1 and SECO-2, Comanche National Grasslands, CO; NECO-1 and NECO-2, Pawnee National Grasslands, CO; WY-1 and WY-2, Thunder Basin National Grasslands, WY; MT-1 and MT-2, Fort Keogh Livestock and Range Research Laboratory, MT.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Pearson correlation matrix for soil properties and estimates of N assimilation in sterilized and unsterilized soils (n = 30).

 
There were no statistically significant trends in bulk soil {delta}15N values across the gradient (Fig. 1c). Soils from the Texas Panhandle sites and northeastern Colorado (NECO) were the most depleted in 15N, and all other soils had {delta}15N ratios above 4 {per thousand} (Fig. 1c). These estimates are enriched relative to {delta}15N values reported for mineral soils in humid grasslands and forest ecosystems, but higher {delta}15N values are typical for dry ecosystems (Nadelhoffer and Fry, 1988; Fry, 1991; Frank et al., 2000; Johnson et al., 2000).

Total and Biotic Nitrogen Immobilization
Estimates of total N assimilation ranged from 1.21 to 2.40 mg N kg-1 soil and were at the low end of estimates reported in previous studies for semiarid ecosystems (Schimel, 1986; Burke et al., 1989b) but compared well with estimates of gross N immobilization using pool-dilution techniques for soils collected across the same gradient (Barrett and Burke, 2000). Total N assimilation tended to be highest in northern, high C soils (Fig. 2a) . Estimates of biotic N immobilization (unsterilized - sterilized) varied significantly across the gradient and showed similar patterns to total N assimilation (Fig. 2b).



View larger version (53K):
[in this window]
[in a new window]
  		Fig. 2. (a) Total N immobilization in unsterilized soils, (b) biological N immobilization in unsterilized soils (unsterilized -sterilized), and (c) abiotic N immobilization in sterilized soils collected from across a temperature gradient in the US Great Plains. Bars with the same letter are not significantly different at P = 0.05. Error bars are one standard error of the mean.Site codes are as follows: TX-1 and TX-2, Muleshoe National Wildlife Refuge, TX; SECO-1 and SECO-2, Comanche National Grasslands, CO; NECO-1 and NECO-2, Pawnee National Grasslands, CO; WY-1 and WY-2, Thunder Basin National Grasslands, WY; MT-1 and MT-2, Fort Keogh Livestock and Range Research Laboratory, MT.

 
Total N assimilation and biotic N immobilization were both strongly correlated with soil KCl-extractable NH+4 content (Table 2). Large standing pools of KCl-extractable NH+4 may be indicative of high rates of N mineralization in the soil, or slow nitrification. Since previous work has shown that potential N mineralization increases across this gradient (Barrett and Burke, 2000), concomitantly with the KCl-extractable NH+4 concentrations reported here, we suggest that the correlation between NH+4 and N assimilation reflects faster rates of potential N cycling in the northern soils under laboratory incubations.

Soil organic C content was significantly correlated with both total N assimilation and biotic N immobilization (Table 2). Soil organic C content accounted for 64% of the variance in estimates of N immobilization in unsterilized soils and 52% of the variance in estimates of biotic N immobilization (Fig. 3a,b) . This influence of soil C upon N immobilization is consistent with results from previous studies conducted over a broad survey of ecosystems and soil types in which high rates of N immobilization have been linked to C availability (Monreal et al., 1981; Schimel, 1986; Strickland et al., 1992; Hart et al., 1994; Nadelhoffer et al., 1995; Downs et al., 1996; Zink and Allen, 1998; Barrett and Burke, 2000). In forested ecosystems, availability of labile C to microbes can limit N immobilization, and has been shown to be particularly important in reducing losses of mobile N from disturbed forests (Vitousek and Matson, 1984, 1985; Nadelhoffer et al., 1995; Downs et al., 1996). In grassland ecosystems, C content, and N immobilization are positively correlated at both landscape and regional scales (Schimel, 1986; Burke et al., 1989b; Delgado et al., 1996; Burke et al., 1997; Barrett and Burke, 2000).



View larger version (29K):
[in this window]
[in a new window]
  		Fig. 3. Scatter plot and linear regression of (a) total N assimilation and (b) biological N immobilization vs. soil organic C content in soils collected from across a temperature gradient in the US Great Plains. Error bars are one standard error of the mean.

 
In contrast, using the same 15N tracer methodology Johnson et al. (2000) reported that total N assimilation (abiotic + biotic) declined as a function of total soil N (and presumably total soil organic matter) over a variety of forest ecosystems, reasoning that total soil N was indicative of N saturation status; the more N saturated a particular ecosystem (sensu Aber et al., 1989), the greater the total soil N and the lower the capacity of the soil to immobilize N inputs. Similarly, Magill et al. (1996) showed that the capacity of a forest ecosystem to retain anthropogenic inputs was contingent upon the history of N inputs and management. Johnson et al. (2000) found that in such cases, abiotic mechanisms of N stabilization contributed proportionally more to total N immobilization suggesting that abiotic mechanisms of retention may represent important N sinks in soils receiving high rates of N deposition. These results raise an interesting question: if abiotic mechanisms of N retention are proportionally more important in N saturated ecosystems, to what extent is the influence of soil organic matter over N retention influenced by the history and severity of N deposition? Perhaps in pristine systems under low N deposition, soil organic matter may constitute a strong sink for excess N, relative to N saturated systems where the capacity of soils to retain N inputs has been exceeded.

Other differences in soil properties are also likely to have influenced the differences between forest and grassland soils considered in these studies. For example, forests soils typically have higher organic matter content, lower pH, and lower nitrification potentials than the grassland soils considered here (Fenn et al., 1998; Johnson et al., 2000). Moreover, forest soils typically have higher concentrations of humic acids and high molecular weight compounds comprising their soil organic matter, which may influence both biological and abiotic N assimilation mechanisms (Stevenson, 1994).

Abiotic Nitrogen Immobilization
Estimates of abiotic N uptake averaged 19% of total N immobilization across all sites along the organic matter gradient. These results are consistent with respect to the proportion of fixed NH+4 found in grassland and agricultural soils (Young and Aldag, 1982; Stevenson, 1994) and to the rates of rapid abiotic uptake reported in pool-dilution assays (Davidson et al., 1991). Estimates of abiotic N assimilation ranged from 0.089 to 0.80 mg N kg-1 soil (Fig. 2c). Soils collected from northeastern Colorado and Wyoming exhibited the highest rates of abiotic uptake, expressed as both an absolute estimate and as a proportion of the total N assimilation (Fig. 4) . Johnson et al. (2000) and Dail et al. (2001) reported higher absolute, but proportionally similar rates of abiotic N immobilization in a variety of temperate forest soils. Such results emphasize the importance and generality of nonbiological mechanisms of N retention. If actual in situ rates of abiotic N immobilization are as high as the potentials reported here, and if they remain similarly high above tracer level application, then abiotic stabilization of fertilizer inputs and atmospheric deposition may be a substantial sink for anthropogenic sources of biologically available N, especially if abiotic mechanisms of N retention become proportionally more important with the history and severity of N inputs as Johnson et al. (2000) suggest.



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 4. Proportion of total N assimilation attributed to abiotic uptake (N assimilation in sterilized soils/N assimilation in unsterilized soils) for soils collected across a temperature gradient in the US Great Plains. Error bars are one standard error of the mean.Site codes are as follows: TX-1 and TX-2, Muleshoe National Wildlife Refuge, TX; SECO-1 and SECO-2, Comanche National Grasslands, CO; NECO-1 and NECO-2, Pawnee National Grasslands, CO; WY-1 and WY-2, Thunder Basin National Grasslands, WY; MT-1 and MT-2, Fort Keogh Livestock and Range Research Laboratory, MT.

 
While previous studies have reported that abiotic mechanisms of N retention are largely facilitated by substitution in clay minerals and high molecular weight organic compounds (Burge and Broadbent, 1961; Young and Aldag, 1982; He et al., 1988; Stevenson, 1994), we found no relationship between abiotic N assimilation and soil organic C content, and only a weak relationship between abiotic N assimilation and clay content (Table 2). Perhaps the effect of clay on abiotic N assimilation has been confounded by differences in mineralogy and parent material among the study sites. Nevertheless, this lack of clear relationship between abiotic N assimilation and soil organic matter or clay is surprising, particularly given the strong correlations between soil organic matter and total and biotic N uptake. Perhaps differences in soil organic matter composition over the gradient (Amelung et al. 1997) are more important influences over abiotic N uptake than soil organic matter content alone, particularly given the importance of humic compounds of previous reports of abiotic N uptake (Stevenson, 1994).

Soil pH and abiotic N assimilation were negatively correlated (Table 2). This too is in contrast to previously published results where reactions facilitating abiotic stabilization of NH3 and NH+4 by humic substances and substitution in clay minerals are favored at high soil pH (Mattson and Koulter-Andersson, 1942, 1943; Nömmik, 1965; Nömmik and Vahtras, 1982; He et al., 1988; Stevenson, 1994). However, there is evidence of abiotic N uptake mechanisms facilitated at low pH (Stevenson, 1994). Axelsson and Berg (1988) reported that abiotic NH+4 retention occurred at pH 4 on decomposing Scots pine (Pinus sylvestris L.) litter. Schimel and Firestone (1989a)(b) found that nonbiological reactions accounted for 20% of the N retained in an acid (pH 4.3–4.5) forest soils. However, these mechanisms are poorly understood in general and all these examples occurred in acidic forest soils with pH considerably lower than the soils we describe (Table 1). Differences in mineralogy and organic matter composition across the gradient could be an additional source of variability and influence the relationship between abiotic N uptake and pH. Future study should address the influence of organic matter composition, clay mineralogy, and the influence of pH more directly to identify the specific mechanisms responsible for abiotic N assimilation.

The sterilization procedure and 15N tracer application may have contributed to error in our estimates of N assimilation in both sterilized and unsterilized soils. For example, the use of HgCl2 may have inadvertently suppressed abiotic N uptake as well as biological N immobilization because the presence of a divalent cation (Hg2+) could have competed with NH+4 for exchange sites and for abiotic mechanisms that stabilize N. While Hg2+ would have successfully competed with NH+4 for cation-exchange sites, differences in the charge/mass ratio make it unlikely that Hg2+ would compete with NH+4 for substitution in interlayers of 2:1 clays or in phenolic rings. A more likely source of error is from dilution of the 15N tracer by native NH+4 resulting in underestimates in our calculations of total N assimilation in unsterilized soils. Since mineralization NH+4 would not have occurred in the sterilized soils, this would have tended to underestimate the differences between sterilized and unsterilized soils resulting in an underestimate of biological 15N uptake. Regional variability in soil NH+4 content may have contributed to errors in estimating 15N recovery since native 14NH+4 would have competed with the 15N label. If this had occurred one would expect a negative correlation between the KCl-extractable N-NH+4 content of the soil and our estimates of N assimilation. The positive correlation between total assimilation and KCl-extractable N-NH+4 content, and the insignificant correlation between abiotic N assimilation and KCl-extractable N-NH+4 content suggests that there was no discernable effect of preexisting soil NH+4 content on estimates of 15N uptake over the gradient (Table 2).

We attempted to minimize these sources of error by limiting the assays to 12 h and by incubating the soils under low tension to maximize the diffusion of the label through the soils and to minimize the accumulation of native 14NH+4. Moreover, these potential biases would mainly affect the calculation of biological immobilization, as all N assimilated by sterilized soils must necessarily occur by abiotic processes. We conclude that abiotic N uptake can account for a significant proportion (up to 40%) of total assimilation in laboratory incubations of semiarid grassland soils and may represent a significant sink for N in the field. Additional study is necessary to elucidate the mechanism or mechanisms responsible for abiotic N uptake.


   ACKNOWLEDGMENTS
 
We thank C. Supples, T. Hare, I. Hesse, and P. Lowe for assistance in the laboratory and R. Virginia for discussions on 15N natural abundance in dry ecosystems. Numerous representatives of the USDA and U.S. Fish and Wildlife Service granted permission and assistance at these field sites. T. Boutton and three anonymous reviewers helped to clarify and strengthen the presentation of this manuscript. This work was supported by National Science Foundation grants DEB #9350273 and DEB #9707296 and the Shortgrass Steppe Long Term Ecological Research Program.

Received for publication December 1, 2000.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (12)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Barrett, J. E.
Right arrow Articles by Burke, I. C.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Barrett, J. E.
Right arrow Articles by Burke, I. C.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Barrett, J. E.
Right arrow Articles by Burke, I. C.
Related Collections
Right arrow Nitrogen
Right arrow Soil Organic Matter
Right arrow Range Soils

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome