| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Soil and Crop Sciences, Colorado State University, Fort Collins, CO 80523
b Dep. of Crop and Soil Science, Oregon State Univ., Corvallis, OR 97331
* Corresponding author (Richard.Dick{at}orst.edu)
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
6c, varied among soil aggregates, and communities were classified according to aggregate size class by canonical discriminant analysis. Community Biolog substrate utilization patterns changed over time but were not affected strongly by aggregate size. Lack of community differentiation may be due to the frequent mixing of soil during cultivation and tillage events, whereby microbial communities become evenly distributed among soil aggregates.
Abbreviations: CDC, canonical discriminant analysis • FAME, fatty acid methyl ester • MBC, microbial biomass C • PCA, principal components analysis • TKN, total Kjeldahl N • TOC, total organic carbon
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Studies have shown that proportions of fungi and bacteria differ among aggregate sizes, with lower bacterial:fungal ratios associated with macroaggregates than microaggregates (Gupta and Germida, 1988; Singh and Singh, 1995; Monreal and Kodama, 1997). Fungal hyphae, but not spores, are restricted from the interparticle spaces of microaggregates (Monreal and Kodama, 1997), whereas bacteria may survive longer in the smaller pores of microaggregates by escaping predators (Heijnen et al., 1991). Alternatively, bacteria are more intimately associated with microaggregates as clay particles bind to polysaccharide-coated bacterial cells, whereas fungal hyphae enmesh microaggregates to form larger aggregates (Tisdall and Oades, 1982).
In addition to differences in fungal and bacterial proportions, overall biomass and activity of soil microorganisms can vary among aggregates. Gupta and Germida (1988) reported lower microbial biomass in microaggregates (<0.25 mm) compared with macroaggregates. Similar results were found by Miller and Dick (1995), who found lower biomass values in aggregates <0.25 mm compared with aggregates 0.5 to 1.0 mm in size. Stable macroaggregates (0.25–1.0 mm), in particular, favored greater assimilation of glucose by microorganisms, followed by an increase in microbial biomass, compared with microaggregates (0.053–0.25 mm) (Aoyama et al., 2000). Mendes et al. (1999) reported that not only do microbial mineralization and enzyme activities vary among different sizes of soil aggregates, but that activities among aggregates change over time. For example, in June, N mineralization in one Oregon soil was greatest in the 2.0- to 5.0- and <0.25-mm aggregate size classes, but in September the rate was highest in the 1.0- to 2.0-mm aggregates (Mendes et al., 1999).
Another factor that may affect microbial distribution patterns across soil aggregates is the presence of a winter cover crop. In Oregon's Willamette Valley, winter cover crops often are grown as a means for capturing leachable nutrients and preventing soil erosion during the rainy winter months (Burket et al., 1997). By providing greater root activity and C inputs, cover crops improve soil aggregation and maintain organic C pools compared with conventionally managed (winter fallow) soil (Miller and Dick, 1995). In addition, the distribution of microbial biomass and activity could be affected by winter cover crops, if microbes within a particular aggregate size class respond to cover crop residue inputs (Mendes et al., 1999).
Other than differences between fungal and bacterial biomass, there have been few published studies on microbial community structure and diversity among aggregate size classes. Limited work on specific organisms (denitrifiers, Seech and Beauchamp, 1988; and Rhizobium sp., Mendes and Bottomley, 1998) have demonstrated that populations can differ among aggregate sizes, which suggests community structure is heterogenously distributed. A study of the catabolic potential of microbial communities in a cultivated soil found no effect of aggregate size on bacterial diversity (Lupwayi et al., 2001). A better understanding of the community structure and functional diversity associated with aggregate size classes may help explain whether heterogenous activities are due to differences in nutrient availability, predation pressures, or different populations associating with different sizes of soil aggregates.
To gain insights to the spatial distribution patterns of microbial communities across aggregates, microbial communities were studied across a range of aggregate size classes. Aggregates were obtained during the growing season from winter-fallow and a winter cover-cropped soil to determine the impact of cover cropping on microbial distribution patterns among aggregates. Communities were characterized according to their FAME profiles as well as their ability to utilize a diverse range of C substrates as determined by the Biolog assay. The objectives of this study were to assess the impacts of winter cover cropping over time on the distribution of (i) aggregate sizes in a silt loam soil, (ii) microbial biomass and mineralization activities among aggregate size classes, and (iii) microbial community FAME structure and Biolog substrate utilization potential among aggregate size classes.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Soil was collected from each plot during the course of a growing season. The first sampling was in March 1998, 1 wk prior to the termination of the oat–vetch cover crop. Soil was collected again in July 1998, which corresponded to canopy closure of summer vegetable crop (green beans). For green bean, canopy closure was at the two-trifoliate leaf stage and occurred approximately 6 wk after planting. The third sampling occurred in August 1998, 1 wk prior to harvest of the green bean. The fourth sampling was in spring 1999, just prior to the termination of the winter wheat cover crop. Soil was collected from each plot with a shovel (0–15 cm) and transported in 3.8-L plastic bags to the laboratory.
Soil Sieving and Aggregate Distribution
After sampling, large soil clods were gently broken by hand, and then soils were laid out between sheets of brown paper to dry slowly for several days. This process was conducted at 4°C to minimize the impact of air drying on microbial communities and activities. Soils were allowed to dry until a gravimetric water content of about 80 g kg-1 was reached. Preliminary work indicated soil needed to be at this moisture level to facilitate aggregate fractionation sieving. Soils collected in spring were near field capacity 
, whereas soils at canopy closure and harvest soil were drier, ranging in water content from 180 to 280 g kg-1 at canopy closure and 140 to 210 g kg-1 at harvest.
The soil was fractionated into aggregates by a dry-sieving method. It was assumed that dry-sieving of soil would disrupt the physical habitat of microbial communities to lesser degrees than would wet-sieving, which would fully saturate the pores, and thus dry-sieving would better allow for the determination of potential habitat influences on soil microorganisms. Aggregates fractionation was done by placing 200-g soil portions on nested sieves (20-cm diam.) mounted on a Tyler Ro-Tap sieve shaker (Combustion Engineering Inc., Mentor, OH). Sieves were mechanically shaken (200–250 oscillations min-1) for 2 min to separate soil into the following aggregate size classes: 2.0 to 5.0, 1.0 to 2.0, 0.5 to 1.0, 0.25 to 0.5, 0.1 to 0.25, and <0.1 mm. Preliminary experiments showed that shaking for 2 min was adequate for separating soil aggregates without causing mechanical destruction of large aggregates (data not shown). Aggregate distribution was determined by weighing soil from each aggregate size class. Soil aggregates were stored at 4°C until analyzed for chemical, physical, and biological properties.
Soil Chemical and Physical Analyses
Samples of whole soils and aggregates from the cover-cropped and fallow plots were air-dried and ground to pass a 150-m (100-mesh) sieve for C and N analyses. Total organic carbon (TOC) of whole soils and aggregates was measured by combustion to CO2 with a Leco C analyzer (model WR12, St. Joseph, MI). Total Kjeldahl N (TKN) (organic N and NH+4) of aggregates and whole soils was determined following the method of Bremner and Mulvaney (1982). Particle size analysis was conducted on three aggregate fractions (1.0–2.0, 0.25–0.5, and <0.1 mm) by the Central Analytical Laboratory (Oregon State University, Corvallis, OR). The method of Gee and Bauder (1986) with pretreatment for organic matter removal was followed. Soil texture was found to be constant among aggregate size fractions. Clay contents were 279 g kg-1 for the 1.0- to 2.0-mm fraction, 284 g kg-1 for the 0.25- to 0.5-mm fraction, and 251 g kg-1 for aggregates <0.1 mm in size. Corresponding values for silt content were 626, 617, and 652 g kg-1. Because there were no significant differences in sand, silt, or clay content among aggregates, a sand-free aggregate weight was not used to normalize data from microbiological assays.
Microbiological Analyses
Prior to microbiological analyses, 50-g samples of whole soil and aggregates were moistened with distilled water until a water content of 205 g kg-1 (two-thirds field capacity) was obtained. Samples then were allowed to equilibrate for 4 d in the dark at 25°C.
MBC of whole soils and aggregates was determined by the fumigation-incubation method of Jenkinson and Powlson (1976). Premoistened soil (10 g) was fumigated with chloroform for 24 h. After fumigation, soils were incubated for 10 d at 25°C in acrylic tubes stoppered with rubber septa. The tube headspace was sampled for CO2, which was analyzed by gas chromatography (Carle Series 100 AGC, Loveland, CO). A kC of 0.41 (Voroney and Paul, 1984) was used to calculate MBC without the substraction of a nonfumigated control.
Microbial respiration was calculated from the amount of CO2-C evolved from 10 g of nonfumigated soil, which was incubated in septa-stoppered acrylic tubes for 10 d at 25°C. The headspace was sampled and analyzed for CO2 as described above. Nitrogen mineralization potential was determined by the anaerobic production of NH+4 as described by Bundy and Meisinger (1994).
Community FAME profiles of whole soils and aggregates were determined by the ester-linked method (Schutter and Dick, 2000). Three grams of soil were added to 35-mL glass centrifuge tubes and mixed with 15 mL of 0.2 M KOH in methanol. Soils were incubated at 37°C for 1 h with periodic vortexing. Next, 3 mL of 1.0 M acetic acid were added to neutralize the pH of the tube contents. Extracted FAMEs were partitioned into an organic phase by adding 10 mL of hexane. Tubes then were centrifuged for 10 min at 480 x g to separate organic matter from the hexane. The hexane layer was transferred to a clean glass tube and evaporated under a stream of N2. FAMEs were dissolved in 1:1 hexane: methyl-tert-butyl ether and were transferred to a GC vial for analysis by a Hewlett-Packard 5890 Series II gas chromatograph (Palo Alto, CA) equipped with 25-m by 0.2-mm fused silica capillary column (5% diphenyl-95% dimethylpolysiloxane) and a flame ionization detector.
FAMEs were identified and their relative peak areas determined by the MIS Aerobe method of the MIDI system (Microbial ID, Inc., Newark, DE). FAMEs are described by standard nomenclature. Numbering of carbons begins at the aliphatic () end of the molecule. The number of double bonds within the FAME is given after the colon. The cis and trans conformations are designated with suffixes "c" and "t", respectively. Other notations are "Me" for methyl, "OH" for hydroxy, "cy" for cyclopropane, and the prefixes "i" and "a" for iso- and anteiso-branched FAMEs, respectively.
Microbial community substrate utilization potential of whole soil and aggregates was determined with Gram-negative Biolog Microplates (Biolog Inc., Hayward, CA). Biolog plates were inoculated with 125 µL of soil suspension (10-3 strength) made by serially diluting 2-g soil samples in sterile saline. Plates were incubated at 25°C, and reactions were monitored every 24 h up to 96 h by measuring well absorbance values with an automated plate reader (595 nm; Bio-Tek Instruments, Winooski, VT). Biolog data from the 72-h reading were chosen for community analyses. Well absorbance values were adjusted by subtraction of a blank control; data were normalized by dividing the absorbance value of each well by the plate's maximum absorbance value.
Statistical Analyses
All statistical procedures were performed by the SAS statistical software package (SAS Institute, 1996). Aggregate distribution, TKN, MBC, microbial respiration, N mineralization, and individual FAME data were analyzed by repeated measures analysis of variance, with aggregate size as the repeated term. The repeated term consisted of seven levels, six corresponding to the aggregate size classes and the whole soil sample as the seventh. When the sphericity test was satisfied, the data were treated as a split-plot design, with aggregate size as the subplot. When the sphericity test failed, significance levels were determined by means of the Huynh and Feldt epsilon adjustment (Huynh and Feldt, 1976). To assess effects of sampling date, the data were analyzed by a two-factor repeated measures analysis of variance. Repeated terms were sampling date (four levels) and aggregate size (seven levels). Protected LSD values were calculated to separated means at the level of P < 0.05.
Principal components analysis (PCA) was performed on community FAME or Biolog data to characterize microbial communities from different sampling dates, cover crop treatments, and aggregate size classes. Canonical discriminant analysis (CDA) also was performed on FAME and Biolog community data for the purpose of classifying community structure or substrate utilization potential according to aggregate size class. A stepwise selection procedure was performed first to identify the best FAME or Biolog substrate discriminators of aggregate size, followed by canonical discriminant analysis to identify linear functions that provide for the maximum separation among aggregate size classes. In addition, simple correlation analysis was used to calculate correlation coefficients between soil properties, activities, and soil community positions on axes from PCA and canonical discriminant plots.
|
RESULTS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Microbial Biomass Carbon
Microbial biomass C in whole soil was significantly affected by winter cover cropping, with higher biomass levels in cover-cropped soil compared with fallow soil in July and August 1998 (Table 1). Sampling date and the interaction between date and cover cropping treatment also were significant factors affecting MBC. Although MBC tended to decline after March 1998, biomass levels were maintained and even stimulated in cover-cropped soil in July and August 1998. In addition, MBC differed significantly among aggregate size classes for all sampling dates except March 1999 (Table 1). In March 1998, MBC was highest in the 0.5- to 1.0-mm aggregate size class for both cover-cropped and fallow treatments. In July 1998, MBC increased in the smallest aggregate size class for the fallow soil, but did not differ significantly among aggregates in the cover-cropped soil. For each aggregate size class in July 1998, MBC was significantly higher in cover-cropped soil compared with fallow soil. A similar trend occurred in August 1998, with higher MBC in cover-cropped soil in each aggregate size class. The effect of aggregate size was significant at this sampling date as well, with the lowest levels of MBC occurring in 2.0- to 5.0-mm aggregate size class for both treatments.
|
|
Microbial Community FAME Profiles
A total of 46 different FAMEs were extracted from the soil aggregates over the course of the study; multivariate statistical analyses were performed on the 34 FAMEs that were present in soil at all sampling dates. Principal components analysis was performed on community FAME data, expressed as relative percentages, from all sampling dates and aggregates to determine the overall effects of sampling date, winter cover cropping, and aggregate size on microbial community structure (Fig. 2)
. For simplification, community PC scores were averaged across aggregates within each cover crop treatment x sampling date combination in Fig. 2A, and across sampling date within each cover crop treatment x aggregate size combination in Fig. 2B. According to the analysis, community FAME profiles were influenced more strongly by sampling date and cover crop treatment rather than aggregate size. Community FAME profiles are separated according to sampling date on PC 1 of Fig. 2A, while shifts in communities in response to cover crops are distinguished on PC 2 (Fig. 2A and 2B). On the basis of eigenvector loadings, microbial communities in March contained greater amounts of saturated (15:0 and 17:0) and unsaturated FAMEs (16:19c, 18:36c, and 20:46c) compared with communities in August, which were elevated in branched FAMEs (i15:0, a15:0, i17:0, and a17:0), 10Me18:0, 17:0 cy, and 19:0 cy. Differences between communities from winter-fallow and cover-cropped soil were mainly due to elevated amounts of 16:0, 18:0, 18:19c, and 18:26c in the fallow treatment versus greater amounts of i14:0, i15:1, i16:1, 16:15c, 16:19c, 18:17c/9t/12t, and a17:19c in cover-cropped soil.
|
Because PCA revealed small effects of aggregate size on community structure relative to sampling date and cover crop treatment, CDA was performed to discriminate among microbial community FAME structure by aggregate size classes. To simplify and limit the variables included in the canonical functions, a stepwise selection procedure was conducted to identify FAMEs that best discriminated among aggregate size classes. These FAMEs were then used as variables for canonical discriminant analysis of soil aggregates. The data set for the stepwise selection procedure and CDA consisted of soil from all aggregate sizes of both cover crop treatments for all sampling dates. The stepwise selection procedure identified 12 FAMEs (listed in Table 3) whose concentrations varied among aggregate size classes. The discriminant procedure identified two significant canonical functions that classified communities according to aggregate size class in a two-dimensional space (Fig. 3)
. The first function discriminated communities across a gradient of aggregate sizes (Can 1, P < 0.0001), whereas communities of intermediate aggregates (0.1–2.0 mm) were distinguishable from communities of the largest (2.0–5.0 mm) and smallest (<0.1 mm) aggregate size classes by the second canonical function (Can 2, P = 0.02). Analysis of variance tests were performed on the twelve FAMEs to determine if their relative amounts varied significantly across aggregate size classes (Table 3). Relative amounts of four FAMEs (16:1 2OH, 10Me16:0, 19:0 cy, and 18:26c) increased significantly as aggregate size increased, whereas three FAMEs were present in significantly greater amounts in the smallest aggregate size fraction (14:0, 16:0, and 16:17c/i15:0 2OH). Significantly less 18:0 was extracted from intermediate aggregate size classes, particularly the 0.5- to 1.0-mm size fraction. A similar, although not statistically significant, trend was observed for the FAME 20:46c.
|
|
|
Correlations among Aggregate Properties
Correlations among aggregate size median, soil properties, microbial activities, and community positions along PCA and CDA plots are shown in Table 4. Microbial biomass C and N mineralization potential were positively correlated with TOC and TKN (P < 0.0001). Microbial respiration also was correlated with TKN (r = 0.21, P < 0.01). Although TOC and TKN were not correlated with aggregate size, aggregate size was correlated with microbial respiration (r = -0.21, P < 0.01) and N mineralization potential r = -0.27, P < 0.01). Positions of community FAME structure on PC 1 and 2 were correlated with several properties. Positions on PC 1 were most strongly correlated with aggregate size (r = 0.23, P < 0.01), microbial respiration (r = -0.39, P < 0.0001), and N mineralization potential (r = -0.25, P < 0.01). Positions on PC 2, which separated community FAME structure according to cover cropping practice, were significantly correlated with TOC, TKN, MBC, respiration, and N mineralization potential. For the CDA of community FAME profiles, community positions on Canonical Function 1 were most strongly correlated with aggregate size r = -0.59, P < 0.0001), whereas community positions on Canonical Function 2, with separated intermediate aggregates from the largest and smallest size classes, were correlated with MBC, respiration, and N mineralization (P < 0.0001). Correlations among aggregate properties and Biolog substrate utilization patterns were fewer than with community FAME profiles. Positions of community Biolog substrate utilization patterns on PC 1 of Fig. 3 were not correlated with aggregate size, TOC, or TKN, but were positively associated with MBC, respiration, and N mineralization potential (P < 0.0001).
|
|
DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Distribution patterns of MBC and microbial activity among soil aggregates changed over time. For example, in March 1998, MBC and N mineralization potential were greatest in intermediate size aggregates (0.25–0.5 and 0.5–1.0 mm), but in July and August, levels in fractions <0.1 mm were as great, if not greater, than in the intermediate aggregates. Differences over time may be due to seasonal changes in nutrient supply associated with different sizes of aggregates. In this study, the highest concentration of TKN was observed in the <0.1-mm aggregate size fraction July 1998; an increase in TKN content in this size fraction may have contributed to greater microbial biomass and activity at this sampling date. This is consistent with results from Miller and Dick (1995), who also detected greater MBC in aggregates with greater total N content. It also is possible that differences among sampling dates were due to spatial or random variation. Because this study did not include a second vegetation period, it is difficult to conclude if the trends presented here were strictly seasonal. However, certain patterns are similar to those reported in other studies. Ndiaye et al. (2000) consistently found greater MBC and enzyme activities in Oregon winter-fallow soils in spring compared with summer and harvest sampling dates (1997 and 1998 growing seasons); in our study, MBC, respiration, and N mineralization were greater in fallow aggregates and whole soils in March 1998 compared with July and August 1998. In addition, changes in community FAME profiles over time were similar to seasonal trends observed in another study, where whole soils collected in March 1998 from the same research station were enriched in 18:36c but depleted in i15:0, i17:0 a17:0, and 10Me16:0 compared with summer and fall sampling dates (Schutter and Dick, 2002).
Although microaggregates (<0.25 mm) may offer protection from predators, the habitats provided by microaggregates generally have been described as harsh, containing low levels of biologically recalcitrant organic matter with turnover rates of hundreds of years (Elliott, 1986; Angers and Giroux, 1996; Jastrow et al., 1996; Monreal et al., 1997). A harsh environment with poor substrate quality may explain why microbial biomass and activity can be lower in microaggregates compared with macroaggregates (Elliott, 1986; Gupta and Germida, 1988). However, several studies that report such trends employed wet-sieving techniques to separate soils into aggregate fractions (Elliot, 1986; Beauchamp and Seech, 1990; Franzluebbers and Arshad, 1997). In other studies, high microbial biomass levels and activities have been observed in microaggregates when soils were dry sieved to separate aggregate size classes (Seech and Beauchamp, 1988: Mendes and Bottomley,1998; Mendes et al., 1999). Seech and Beauchamp (1988) suggested that low levels of activity were found in wet-sieved microaggregates because water-soluble C may have been removed from microaggregates during the wet-sieving process. In a later study, they found that microbial denitrification activity decreased as the dry-sieved aggregate size increased but that the rate increased as the wet-sieved aggregate size increased. In both cases, however, denitrification activity was correlated with mineralizable C, and presumably thus labile C substrate (Beauchamp and Seech, 1990). Although labile C was not measured in our study, high respiration and TKN levels in the smallest aggregates are indicative of the presence of labile substrates in microaggregates during the July sampling period.
Greater biomass in microaggregates also has been attributed to absorption of bacterial cells to clays in microaggregate size fraction (Van Gestel et al., 1996), but this seems unlikely in our study as clay content did not differ significantly between aggregate size classes. It also is possible that the porosity and texture of microaggregates may protect microorganisms from dessication, but such protection did not occur in several studies (Van Gestel et al., 1991; West et al., 1992). Alternatively, it is possible that in our study, some of the <0.1-mm microaggregates were loosely held to the surfaces of macroaggregates and disassociated from the surface during the sieving process. A substantial portion of the soil microbial biomass is thought to live on or near the aggregate surface (Oades, 1984), which may explain the high MBC in microaggregates <0.1 mm if they originated on macroaggregate surfaces.
Despite differences in microbial biomass and activity associated with aggregate size, aggregate size did not appear to have a strong influence on microbial community FAME structure or substrate utilization potential. When analyzed by PCA, microbial community substrate utilization potential was not strongly affected by aggregate size, and four canonical functions were necessary to discriminate Biolog patterns according to aggregate size. Similar results were found by Lupwayi et al. (2001), who found no effects of aggregate size on the diversity of Biolog substrates utilized by microbial communities. Differences in Biolog substrate utilization potential have been reported, however. Winding (1994) found distinct patterns between micro- (<0.002 mm) and macroaggregate (>0.25 mm) size fractions from a sandy loam soil agricultural soil. Specifically, the well absorbance values were higher in plates inoculated with microaggregate suspensions compared with macroaggregate suspensions. These patterns could partly be explained by the method employed by Winding (1994), who used different dilution factors for different sizes of aggregates.
In this study, sampling date and winter-cover cropping were more important determinants of community structure than aggregate size when the community FAME data set was analyzed by PCA. Additional statistical analyses were necessary to identify a subset of community FAMEs that did vary among aggregate size fractions. For example, relative amounts of 16:1 2OH, 10Me16:0, and 19:0 cy increased with increasing aggregate size, and the <0.1 mm fraction was enriched in 14:0, 16:0, and 16:17c/i15:0 2OH but depleted in 18:26c. Several of these FAMEs are reported to be markers for specific groups of microorganisms. 10Me16:0 has been described as a biomarker for sulfate-reducing bacteria, specifically Desulfobacter, and 19:0 cy is one marker for Gram-negative bacteria (Ringelberg et al., 1997; Petersen et al. 1997). The FAME 18:26c is found in diverse organisms, including plants, fungi, and cyanobacteria (Harwood and Russell, 1984; Vestal and White, 1989). This FAME often is used as a biomarker for fungi (Frostegrd et al., 1997; Ringelberg et al., 1997; Petersen et al., 1997) because its concentration has been correlated with ergosterol, a fungal-specific membrane component, (Frostegrd and Bth, 1996). In our study, 18:26c was depleted in the <0.1-mm fraction, suggesting that amounts of fungal hyphae, relative to other community components, were greater in larger aggregates. This is in agreement with studies that have found fungi to be an important stabilizing agent in soil aggregates (Tisdall and Oades, 1982), especially when one considers that the <0.1-mm-size fraction may partly consists of soil particles freed from unstable aggregates. In a separate study, Petersen et al. (1997) found very few differences in fatty acid profiles among wet-sieved aggregates from conventionally and organically-managed soils. One exception was the increase in the relative concentration of 18:26c with decreasing aggregate size, which was partly attributed to the preferential growth of fungi near the surface of aggregates. A negative relationship between 18:26c and aggregate size does not necessarily conflict with the positive trend observed in our study, however, because the smallest aggregate in the study by Petersen et al. (1997) was only 0.25 to 0.50 mm in diameter. Overall, their findings contribute little evidence for community differentiation based on aggregate size. Some differences were found when FAMEs were analyzed individually, but not when multivariate statistical procedures were performed on community data sets. One possible explanation for the lack of community differentiation offered by Petersen et al. (1997) was that some of the fatty acid material was lost from the surface of aggregates during the wet-sieving processes. In our study, soils were dry-sieved, and fatty acid material lost from aggregate surfaces should have been recovered in the <0.1-mm fraction. Rather, lack of community differentiation may be due to the frequent mixing of soil during cultivation and tillage events. The continuous process of aggregate destruction and reformation may result in an equilibrium whereby microbial communities become evenly distributed among soil aggregates (Petersen et al., 1997). Another possible explanation is provided by the hierarchical model of soil aggregate structure, meaning that larger aggregates are made up of smaller aggregates (Tisdall and Oades, 1982), and thus communities from larger aggregates are the sum of communities from smaller aggregates. One exception to this hierarchical model is that fungal biomass should be higher in larger aggregates, which is supported by Gupta and Germida (1988) and by the depletion of 18:26c in the smallest fraction of our study.
In general, winter cover crops increased MBC, respiration, and N mineralization potential in aggregates and whole soils relative to the fallow treatment, although these effects usually were not observed until July or August 1998, after cover crops were incorporated into soil. In some cases, July and August levels of microbial activity in aggregates of cover-cropped soil were equal to those observed in March 1998, whereas in the winter-fallow soil, levels tended to decline from March to July. Positive responses of MBC and activities to cover cropping, as observed in this study, are indicative of a substrate limitation in whole soils and aggregates which was ameliorated by the addition of labile cover crop residues. Similar findings were reported by Miller and Dick (1995) at a different experimental research station in Oregon, where establishment of a legume cover crop shortly was followed by increases in MBC and labile organic matter pools (particulate and dissolved organic C). Ndiaye et al. (2000) also observed enhanced MBC and soil enzyme activities in several Oregon soils in response to cover crop residues during the 1997 and 1998 growing seasons.
Winter cover cropping did affect the distribution patterns of MBC and microbial activity among soil aggregates. In July 1998, the cover crop treatment distributed MBC more evenly among aggregates compared with the fallow treatment. At the same time, microbial respiration patterns shifted, with more respiration occurring in 0.1- to 1.0-mm aggregates relative to other aggregate size fractions of cover-cropped soil. In August, the impacts of cover cropping on microorganisms appeared to peak, and differences in MBC, respiration, and N mineralization became more distinct among aggregate size fractions of cover-cropped soil compared with fallow soil. In general, the impacts of winter cover crops on microbial distribution patterns can be complex. Mendes et al. (1999) found that microbial responses to cover cropping not only depends on aggregate size, but on the type of winter cover crop. In September 1996, microbial respiration increased in 1.0- to 2.0-mm aggregates of legume cover-cropped soil relative to fallow, but an increase in the same aggregate size fraction was not observed in the triticale cover-cropped soil. Distributions of Rhizobium serotypes among aggregates also were affected differently by different cover crop treatments (Mendes and Bottomley, 1998). Such results indicate that soil microorganisms and their activities can be influenced in as-of-yet unpredictable ways because of the complex interactions among environmental factors, substrate quality, and time that occur in aggregate microhabitats.
In this study, winter cover cropping did not significantly affect the size distribution of soil aggregates, which prevented a more-detailed understanding of the practical impacts of aggregate size shifts on microbial communities and activities. Moreover, the majority of the soil was represented by aggregates >2.0 mm, and even though microbial biomass and activity increased in microaggregates over time, much of the overall biomass and activity was still associated with the two largest aggregate size fractions. It is possible that in this particular soil, microaggregate habitats did not contribute significantly to the overall biological activity of the whole soil. However, in other Oregon soils, a large proportion of the soil can be represented by microaggregates <0.25 mm (Mendes et al., 1999), and microbial community or functional changes in these microaggregates in response management or time will affect significantly the overall activity of the whole soil. Thus, understanding the heterogeneity of soil biological properties contributed by microhabitats will be beneficial, particularly when the distribution of soil aggregate sizes is altered by alternative management practices or soil degradation.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Received for publication August 16, 2000.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. N. Miller, B. J. Zebarth, C. E. Dandie, D. L. Burton, C. Goyer, and J. T. Trevors Denitrifier Community Dynamics in Soil Aggregates under Permanent Grassland and Arable Cropping Systems Soil Sci. Soc. Am. J., September 11, 2009; 73(6): 1843 - 1851. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. M. Sainju, T. Caesar-TonThat, and J. D. Jabro Carbon and Nitrogen Fractions in Dryland Soil Aggregates Affected by Long-term Tillage and Cropping Sequence Soil Sci. Soc. Am. J., July 14, 2009; 73(5): 1488 - 1495. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Muruganandam, D. W. Israel, and W. P. Robarge Activities of Nitrogen-Mineralization Enzymes Associated with Soil Aggregate Size Fractions of Three Tillage Systems Soil Sci. Soc. Am. J., May 1, 2009; 73(3): 751 - 759. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Six, S. D. Frey, R. K. Thiet, and K. M. Batten Bacterial and Fungal Contributions to Carbon Sequestration in Agroecosystems Soil Sci. Soc. Am. J., February 27, 2006; 70(2): 555 - 569. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. M. Banowetz, G. W. Whittaker, K. P. Dierksen, M. D. Azevedo, A. C. Kennedy, S. M. Griffith, and J. J. Steiner Fatty acid methyl ester analysis to identify sources of soil in surface water. J. Environ. Qual., January 1, 2006; 35(1): 133 - 140. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. J. Allison and R. M. Miller SOIL GRINDING INCREASES THE RELATIVE ABUNDANCE OF EUKARYOTIC PHOSPHOLIPID FATTY ACIDS Soil Sci. Soc. Am. J., March 1, 2005; 69(2): 423 - 426. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. M. Sainju, T. H. Terrill, S. Gelaye, and B. P. Singh Soil Aggregation and Carbon and Nitrogen Pools under Rhizoma Peanut and Perennial Weeds Soil Sci. Soc. Am. J., January 1, 2003; 67(1): 146 - 155. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||
