SSSAJ Journal of Natural Resources and Life Sciences Education
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 HighWire
Right arrow Citing Articles via ISI Web of Science (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Plante, A. F.
Right arrow Articles by McGill, W. B.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Plante, A. F.
Right arrow Articles by McGill, W. B.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Plante, A. F.
Right arrow Articles by McGill, W. B.
Related Collections
Right arrow Structure and Properties
Right arrow Soil Conservation
Right arrow Soil Organic Matter
Soil Science Society of America Journal 66:1285-1295 (2002)
© 2002 Soil Science Society of America

DIVISION S-6—SOIL & WATER MANAGEMENT & CONSERVATION

Intraseasonal Soil Macroaggregate Dynamics in Two Contrasting Field Soils Using Labeled Tracer Spheres

A. F. Plante*,a and W. B. McGillb

a Unité de Science du Sol, INRA Versailles, Route St-Cyr, 78026 Versailles Cedex, France
b College of Science and Management, Univ. of Northern British Columbia, 3333 University Way, Prince George, BC Canada V2N 4Z9

* Corresponding author (alainfplante{at}hotmail.com)


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
Several studies have hypothesized that increased turnover of soil aggregates promotes soil organic matter losses under cultivation; while others suggest that organic matter protection requires occlusion into aggregates. However, few direct observations of aggregate dynamics are reported in the literature. A 2-yr field study was performed to observe active organic C dynamics and soil macroaggregate dynamics in two contrasting soils. Dysprosium-labelled tracer spheres were applied to field plots to observe soil macroaggregate dynamics, while CO2-evolution during 10-d laboratory incubations was used to measure active C dynamics. Results of biochemical analyses showed higher active C turnover in the low C soil, suggesting a lower proportion of incoming organic matter was protected when compared with the high C soil. No net aggregation or degradation was determined over the long-term, suggesting the soil was at steady-state. However, aggregation followed a cyclical pattern reset by the over-winter period and tillage. Tracer incorporation into large macroaggregates was observed within 9 d after tillage, reaching a maximum of 40 to 60% tracer incorporation into >1-mm aggregates after 72 d. A rapid approach to equilibrium within the study period reflected rapid dynamics of macroaggregates in both soils studied. Slower macroaggregate dynamics in the high C soil were attributed to sustained aggregate stability and resiliency at the end of the growing season. Based on observations of macroaggregate dynamics in soils with contrasting active organic C dynamics, we suggest that rapid macroaggregate turnover not only results in the exposure of labile organic matter but also provides a mechanism for the occlusion and physical protection of particulate organic matter.

Abbreviations: Dy, Dysprosium • INAA, instrumental neutron activation analysis • MWD, mean weight diameter


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
DYNAMICS OF SOIL AGGREGATION have gained increasing attention because of its potential role in the sequestration of organic C in soils. Several studies reported that water-stable macroaggregates are enriched with young particulate organic matter (e.g., Angers and Giroux, 1996; Puget et al., 1995) and proposed that the particulate organic matter can act as a nucleus for macroaggregate formation. Therefore, residue inputs become a significant factor in the feedback process of organic C input and aggregate formation and stabilization (Christensen, 1986). As organic materials become occluded in aggregates they enter an environment less conducive to decomposition (e.g., anaerobic), become physically inaccessible to degrading microorganisms, or interact with reactive clay surfaces thereby extending decomposition times and increasing the potential for long-term sequestration. Therefore, soil respiration measured in short-term incubations has been used to represent a pool of biologically active C in the soil (Dinwoodie and Juma, 1988; Paul et al., 2001), which is not sequestered or physically protected in aggregates. However, several researchers have proposed that the protection provided by the soil matrix should depend on the life expectancy of the protection sites themselves (Angers and Chenu, 1998; Besnard et al., 1996; Plante et al., 1999; Six et al., 1998, 1999, 2000) as illustrated by various crushing experiments which result in a flush of respiration (reviewed by Balesdent et al., 2000).

Soil aggregate dynamics have been expressed as changes in water-stable aggregate mean weight diameter (MWD) related to differences in management (e.g., Perfect et al., 1990a). Other studies have inferred aggregate dynamics from the accrual or loss of organic matter in aggregates with time (Angers and Giroux, 1996; Jastrow et al., 1996; Monreal et al., 1997; Puget et al., 1995; Six et al., 1999; Skjemstad et al., 1990) by investigating the inputs and turnover of organic C in water-stable aggregates of different sizes. Turnover times for micro- and macroaggregates were estimated from first-order rate constants calculated from {delta}13C data. Generally, aggregate turnover rates increased with increasing aggregate size, suggesting that aggregate turnover was more rapid for larger aggregates. However, soil macroaggregate formation occurs faster than total soil organic C, mineral-associated C, and macroaggregate-associated C accumulations (Gale et al., 2000; Jastrow, 1996; Six et al., 1999; Wander and Yang, 2000). This suggests that organic C accumulation or loss may not be a precise tracer for soil aggregate dynamics. The use of a dynamic tracer substance (e.g., decomposing organic materials) to study an equally dynamic system (e.g., mainly inorganic soil macroaggregates) may limit the interpretation of results. Staricka et al. (1992) avoided this problem by using a physical tracer approach to examine aggregate dynamics, but the tracers used and the aggregate sizes measured were large and therefore the study was pertinent to only a small fraction of the soil mass. The observation of tracer sphere incorporation into soil aggregates provides direct observations of soil aggregate formation and can be used to estimate aggregate turnover rates under steady-state conditions. Rates of tracer incorporation become equivalent to aggregate turnover rates when soil aggregation is at steady-state, with no net aggregation or degradation, because formation is balanced by breakdown.

While many of the studies cited have concluded that soil aggregate turnover is a significant control on organic C turnover, few have made direct links between the observed organic matter dynamics and the dynamics of soil aggregates. The direct evidence currently available to test such hypotheses is insufficient because studies using rates of organic matter accumulation or loss, or the measurement of net changes in aggregation with time provide only indirect evidence for the relationship between aggregate and organic matter dynamics. We contend that determining how soil aggregate turnover controls organic matter turnover requires determining the turnover of the aggregates themselves. To this end, the objectives of the work reported here were: (i) to observe the incorporation of tracer spheres into soil aggregate fractions, (ii) to use incorporation of spheres into aggregate fractions to describe patterns in aggregation, (iii) to observe the dynamics of the active organic C as measured by relative rates of respiration in two contrasting soils, and (iv) to attempt to link the observed organic matter dynamics to soil aggregate dynamics. The hypothesis to be tested is that a soil with rapid organic matter dynamics will also demonstrate rapid soil aggregate dynamics.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
Experimental Field Sites
Experiments were conducted on long-term field plots established in 1983 at the Ellerslie and Breton research stations near Edmonton, Alberta (Solberg et al., 1997). The plots are currently managed by the Agronomy Unit of Alberta Agriculture, Food and Rural Development, and are located on two contrasting soils. Soils at Ellerslie (53° 25'N lat., 113° 33'W long.) are from the Malmo series, a Typic Cryoboroll, while soils at Breton (53° 07'N lat., 114° 28'W long.) are from the Breton series, a Typic Cryoboralf. Soil properties reported in Table 1 show that the sampling locations and treatments permitted comparisons between soils under similar climate, but with contrasting organic C contents. Contrasting organic C dynamics between the sites was previously reported (Solberg et al., 1997; Dinwoodie and Juma, 1988).


View this table:
[in this window]
[in a new window]
  		Table 1. Selected site characteristics and surface layer (0–15 cm) properties of field soils used (from Solberg et al. 1997).

 
At each site, four replicates of two of the established treatments consisting of annual N additions of 56 kg ha-1, with or without straw addition were used to provide differing amounts of C input at each site. Original plots sizes were 2.75 by 6.86 m, but were split into three to form subplots of 2.75 by 2.29 m. In addition to the existing four replicates, the subdivision of the field plots allowed us to repeat the experiments in consecutive years (1998 and 1999), maintain ongoing experiments during the same years, and reduce the amount of tracer spheres required for each application. Hence, subplots are referred to as follows: 1998(I), the first year of sampling on subplots where tracers were applied in 1998; 1998(II), the second year of sampling (1999) on subplots where tracers were applied in 1998; and 1999(I), the first year of sampling on subplots where tracers were applied in 1999. Results reported in this study are from the 1998(I) and 1998(II) subplots only.

The tracer spheres used in the field experiments are made by prilling a ceramic material (nepheline syenite) along with dysprosium oxide (Dy2O3) powder, which was added during the manufacturing process by Kinetico Inc. (Nashwauk, MN). The spheres were previously characterized and results reported in detail in Plante et al. (1999). Table 2 provides a summary of tracer sphere properties used in the field experiment. The application rate in 1998 was operationally set to 0.65 kg spheres plot-1. This rate is ~0.63 and 0.49 mg spheres g-1 soil to a depth of 15 cm for the Ellerslie and Breton sites, respectively. Differences in the rates are because of differences in soil bulk density values used in application rate calculations; 1.1 Mg m-3 for Malmo and 1.4 Mg m3 for Breton (Dinwoodie and Juma, 1988). These rates are two to three times that used by Staricka et al. (1992), which was required to improve the power of the experiment, especially considering the use of smaller spheres. Measurement of tracer content by instrumental neutron activation analysis (INAA) is reported in micrograms of dysprosium (Dy) rather than mass of spheres. Therefore, the calculated Dy application rate was 98.9 µg Dy g-1 soil at Ellerslie, and 77.6 µg Dy g-1 soil at Breton.


View this table:
[in this window]
[in a new window]
  		Table 2. Selected properties of tracer particles applied to field plots.{dagger}

 
In 1998, the ceramic tracers were hand broadcast to the plots on 30 May (at Ellerslie) and 31 May (at Breton). Plots were rototilled and seeded to wheat (Triticum aestivum L.) on 2 June 1998. In 1999, plots were tilled on 7 and 12 May at Ellerslie and Breton respectively, and seeded to glyphosate-resistant canola (Brassica napus L.) to control weed infestation. Personnel from Alberta Agriculture, Food and Rural Development performed all plot seeding and tillage. Four 3-cm soil core samples were taken at random within each plot to a depth of 15 cm every 3 wk beginning ~7 d after tillage. The cores from each plot were composited and returned to the laboratory for analyses. Composited samples were gently broken by hand along natural failure plans to pass an 8-mm sieve. Separate subsamples were taken and used for physical and biochemical analyses. Subsamples for physical analyses were air-dried, while subsamples for biological analyses were kept field-moist. Gravimetric field moisture content was determined by oven drying ~10 g samples at 105°C overnight.

Soil Physical Analyses
Water-Stable Aggregate-Size Distribution
The water-stable aggregate-size distribution was determined using a standard wet sieving method (Angers and Mehuys, 1993). Briefly, a 40 g air-dry subsample was placed on the top of a sieve stack (20-cm diam.) for 2 min of rapid wetting and slaking, then wet sieved for 10 min. The sieve stack consisted of a geometric series of 4-, 2-, 1-, 0.5-, 0.25-, and 0.125-mm sieves. Wet sieving was performed using a tank-design apparatus operating at 0.55 Hz with a 4-cm stroke length. The soil retained on each sieve was oven dried at 105°C overnight and weighed. When reported, the <0.125-mm fraction was determined by subtraction.

The MWD of the water-stable soil aggregates was calculated as an integrated representation of the degree of aggregation. The formula used to calculate the MWD is:

[1]
where xi is the mean diameter of the size fraction and Wi is the weight proportion of soil retained in the fraction.

Tracer Content
The tracer content of individual soil aggregate fractions was determined by INAA at the University of Alberta SLOWPOKE facility using the method described in Plante et al. (1999) and Duke et al. (2000). Briefly, materials collected on each sieve were transferred to 7-mL plastic vials and, if necessary, mixed with granular sugar to fill the vial for a consistent counting geometry. The aggregate-sphere mixtures were irradiated in the SLOWPOKE reactor for 240 s at a nominal neutron flux of 0.2 x 1011 n cm-2 s-1, allowed to decay for ~30 min and counted for 300 s live-time. Malmo soil samples were counted at a sample-to-detector geometry of 3 cm, while samples containing Breton soil were counted at 6 cm because of an increased concentration of interfering 56Mn in the Breton soil. The Dy content of each sample was determined by measuring the 94.7 keV {gamma}-ray emission of the longer-lived Dy radionuclide, 165Dy (T1/2 = 2.334 h), with appropriate decay time, pulser, and soil content corrections applied. Recovery rates of applied tracer spheres were calculated using the rate of application of tracer sphere-associated Dy and the total amount of sphere Dy detected in all sieve fractions of a sample.

Considering the size of the tracer spheres and if no aggregation of the tracer were to occur, the largest proportion of the spheres would be expected to be found in the 0.25- to 0.5-mm sieve fraction. The next largest sieve fraction might also contain some free tracer spheres because ~15% of the spheres were >500-µm in diameter (Plante et al., 1999). However, an estimation of the free versus incorporated amounts of tracer spheres in the 0.5- to 1-mm fraction was not possible without tracer extraction and inspection, and was not critical for our purposes. Instead, individual aggregate fractions were combined to form a >1-mm fraction that represents the incorporated, or aggregated, tracer sphere pool. No tracer Dy should be observed in this fraction unless it has been incorporated into larger aggregates.

A method of expressing the tracer Dy distribution in aggregate size fractions is using the MWD calculation. Dysprosium mean weight diameter can be calculated in a manner similar to soil aggregate MWD:

[2]
where xi is the mean diameter of the aggregate size fraction and Wi is the weight proportion of tracer Dy retained in the size fraction rather than that of soil. Dy MWD data were calculated without the size fractions smaller than 250 µm because INAA detected no tracer-associated Dy in these fractions.

Soil Biochemical Analyses
Total Soil Organic Carbon
Total soil organic C was determined on air-dried, ground samples by dry combustion using a Carlo-Erba NA 1500 elemental analyzer (Milan, Italy) after removal of carbonates with the addition of HCl.

Soil Respiration
Twenty-five grams (dry basis) of field-moist soil were brought to 60% of water-filled pore space and incubated for 10 d in 100-mL beakers contained in 2-L canning jars. The lids of the jars were drilled and a septum inserted for gas sampling. On alternate days, CO2 was measured every 6 h by a Varian 3400 gas chromatograph equipped with a methanizer and flame ionization detector (FID) (Varian Inc., Mississauga, ON), and a stream selector valve for handling multiple samples. The GC was controlled by a PC running Star Workstation v.5 control software (Varian Inc., Mississauga, ON).

Soil Microbial Biomass Carbon
Microbial biomass C was determined by the chloroform fumigation direct-extraction method (Voroney et al., 1993). Field-moist subsamples (25 g) were fumigated overnight with chloroform and subsequently extracted with 50 mL of 0.5M K2SO4. Nonfumigated control samples were also extracted. Soluble C in the K2SO4 extracts was measured using an Astro 2001 System II soluble carbon analyzer (Astro, League City, TX). Microbial biomass C was determined using the equation:

[3]
where Cf and Cnf are the C measured in the fumigated and nonfumigated samples, respectively and kec is the extraction coefficient (0.25; Voroney et al., 1993).

Statistical Analyses
Data were analyzed by two-way repeated measures ANOVA using PROC MIXED in SAS/STAT (SAS Institute Inc., Cary, NC) to properly account for within-subject error because of the multiple sampling points in time. Calculated least-square means are reported as treatment means. Pairwise comparisons of means were performed using Tukey tests when treatment comparisons were found statistically significant, and a value of 5% was selected as the level of significance for all tests unless otherwise noted.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSION
 REFERENCES
 
Soil Biochemical Analyses
During the two growing seasons, total 10-d CO2 respiration was greater in the Malmo soil than the Breton soil (P < 0.001, data not shown). Straw residue application provided a fresh supply of labile organic matter to the microbial biomass and significantly increased soil respiration in both Malmo (P = 0.04) and Breton (P = 0.02) samples. The input of residue represented different proportions of total substrate available because total organic C content was higher in Malmo (60.9 g C kg-1 soil) than in Breton (13.1 g C kg-1 soil). Therefore, when expressed as a proportion of total soil organic C, Breton soil respired a much larger proportion of the organic C in the soil (Fig. 1) . These relative rates of soil respiration were slightly higher with straw residue amendment, but the difference was not strongly significant (P = 0.08 for Malmo, P = 0.09 for Breton). Microbial biomass C was significantly higher (P < 0.001) in Malmo soil compared with Breton (Table 3). While straw addition produced an increase in soil respiration, it did not produce a significant difference in microbial biomass C in either soil (P = 0.12 for Malmo, and P = 0.72 for Breton). When 10-d respiration was normalized to microbial biomass C (Fig. 2) , the Breton samples tended to have a more active biomass, especially in 1998, but the difference was not significant (P = 0.12), suggesting the soils may have contained similar microbial activities. Both the low total soil C measured and the high relative rate of respiration suggest that the Breton soil retains only a small proportion of its annual C inputs. Based on the hypothesis that a more rapid aggregate turnover contributes to organic matter losses, we might expect to observe more rapid aggregate turnover in the Breton soil compared with the Malmo soil.



View larger version (43K):
[in this window]
[in a new window]
  		Fig. 1. Carbon dioxide evolution in 10-d laboratory incubations on a per gram of organic C basis. Error bars are standard deviations (n = 4).

 

View this table:
[in this window]
[in a new window]
  		Table 3. Microbial biomass C from 1998 subplots (mg C kg-1 soil).{dagger}

 


View larger version (37K):
[in this window]
[in a new window]
  		Fig. 2. Carbon dioxide evolution in 10-d laboratory incubations on a per gram of microbial biomass C (MBC) basis. Error bars are standard deviations (n = 4).

 
Patterns of Soil Aggregation
The Malmo soil demonstrated a higher degree of aggregation, with more soil retained in the upper sieve fractions, when compared with the Breton soil as illustrated by the MWD values calculated for each soil at each sampling period (Fig. 3) . Soil MWD values averaged over the duration of the experiment were 2.42 ± 0.54 mm for Malmo soil, which was significantly different (P < 0.001) from 1.46 ± 0.37 mm for Breton soil. Straw addition significantly increased values of soil MWD in Malmo samples (P = 0.03), but not in Breton samples (P = 0.26). Straw amendment increased mean soil MWD by ~0.25 mm in both soils, which represents a larger proportional increase in the Breton soil. Therefore, we suspect the lack of statistical significance in the Breton samples is because of higher variability. Generally, the observed increase in soil MWD with straw amendment supports the hypothesis that particulate organic matter acts as a nucleus for macroaggregate formation. In addition, aggregate enrichment by relatively young organic matter has been observed in several studies (Angers et al., 1997; Jastrow et al., 1996; Puget et al., 1995; Six et al., 1999) and has led to the conclusion that young organic matter contributed to aggregate stability such that only aggregates enriched in young organic matter can resist slaking (Elliot, 1986; Gale et al., 2000; Puget et al., 1996; Six et al., 1998).



View larger version (42K):
[in this window]
[in a new window]
  		Fig. 3. Water-stable soil aggregate mean weight diameter (MWD). Error bars are standard deviations (n = 4).

 
While the soils displayed significant short-term dynamics, they appear to be at steady-state over the medium- to long-term. Soil MWD values were not significantly different between growing seasons (P = 0.39 for Malmo, P = 0.53 for Breton). In addition, previous studies reported similar results for soil aggregate MWD for the same soils, but under slightly different wet sieving protocols. Singh et al. (1994) used a wet sieving time of 30 min compared with our 10 min. They reported values of 2.3 ± 0.4 mm for Malmo soil under tillage and straw addition, and 1.3 ± 0.3 mm for samples under tillage without straw addition. Toogood and Lynch (1959) reported an average aggregate MWD value of 0.86 mm for the rotation cropping system on the Breton classic plots, which are adjacent to the plots used at Breton in this work. In their wet sieving procedure, Toogood and Lynch forced the soil sample through the 4-mm sieve resulting in a lower value. Calculating aggregate MWD values from data in this work without the mass retained on the 4-mm sieve yields an average value of 0.83 mm. The close correlation between previously reported soil aggregate MWD values and those reported in the current study suggest that the soils are at steady-state in reference to aggregation, and are not aggregating or degrading over the span of several years to several decades. Therefore measurements of aggregate formation or destruction within a growing season may reflect the intraseasonal turnover of macroaggregates.

Soil aggregation generally followed a cyclical saw-toothed pattern (Fig. 3). This pattern is likely controlled, directly or indirectly, by microbial activity, climate and plant growth changes during the growing season. The soil MWD data showed some correlation with antecedent soil water content (data not shown), similar to that reported by Perfect et al. (1990a)(1990b). Water content generally decreased during the growing season, which may help to increase soil aggregate stability. Variability in measured microbial biomass within the season is likely linked to soil water content, and increases in microbial biomass can also increase macroaggregate stability. The growth of plants increased labile organic C inputs through root exudates, which could also stimulate the biomass and stabilize macroaggregates, leading to increases in aggregate MWD. The saw-toothed pattern suggests that the over-winter period, when several freeze-thaw cycles occurred, along with spring tillage may reset the state of aggregation resulting in no net aggregation or degradation over medium to longer time scales.

Tracer Sphere Incorporation
Mean recovery rates of Dy are summarized in Table 4. Recovery rates of individual samples ranged from 22 to 264%, reflecting hot spots of localization because of uneven application and imperfect homogenisation during tillage. After the first tillage, the mean recovery rates (averaged over the whole season) were 81% for Malmo and 54% for Breton. Recovery improved after 1 yr in the field and a second tillage in 1999 with mean values of 98% for Malmo and 66% for Breton and a narrowing of the range of values. The general trend for <100% recovery suggests some loss of the tracer during tillage. A significant amount of soil translocation occurred during tillage. We found elevated levels of Dy in random samples taken outside the border of the plots before tillage in the spring of 2000. We believe that tillage translocation removed some of the tracer spheres from the plots to which they were applied. Because of the varying amounts of total spheres detected, all sphere contents of aggregate fractions were reported as values normalized to the total amount detected in all aggregate fractions of the sample (µg Dy [{sum}µg Dy]-1).


View this table:
[in this window]
[in a new window]
  		Table 4. Mean{dagger} tracer sphere recovery based on applied rates and detected Dysprosium (Dy){ddagger} for the 1998 subplots (%).

 
Rapid incorporation of the tracer spheres occurred shortly after the initial tillage events in both 1998 and 1999, where ~25% of the tracers appeared in the aggregated pool (Fig. 4) . Staricka et al. (1992), found that <5% of tracers were incorporated after initial tillage. In part, the differences could be attributed to differences in experimental designs. Staricka et al. (1992) used much larger spheres (1–3 mm compared with ~0.4 mm) and examined incorporation into much larger aggregates (3–40 mm compared with 1–4 mm). In addition, the field plots in the Staricka et al. (1992) experiments were tilled by mouldboard or chisel, whereas our plots were rototilled. We expect that the rototillage produced much greater soil mixing, thus increasing the likelihood of tracer sphere incorporation. While long-term soil cultivation decreases in aggregate MWD (Elliot, 1986; Beare et al., 1994a, 1994b; Six et al., 1998), the large amount of tracer detected in macroaggregates suggests that a large number of new macroaggregates were formed and stabilized during and shortly after tillage.



View larger version (43K):
[in this window]
[in a new window]
  		Fig. 4. Normalized tracer sphere Dysprosium(Dy) content of the pooled "aggregated" fraction (1 to >4 mm). Error bars are standard deviations (n = 4).

 
Distribution of Incorporated Tracer Spheres
Having observed that a significant proportion of the tracer spheres were rapidly incorporated after an initial tillage event, it is interesting to note their distribution in varying sizes of soil aggregates. A large proportion of tracer spheres were incorporated in the largest aggregate fraction (>4 mm) in Malmo soil samples, but less so for Breton (data not shown). The Dy-MWD data for time zero are from wet-sieving trials performed immediately after tracer sphere addition to soil samples without mixing. We found ~90% of the spheres in the 0.25 to 0.5 mm and very little in any aggregate-size fraction >1 mm (data not shown). Dysprosium MWD values during the two field seasons for each soil and treatment are illustrated in Fig. 5 . The increasing Dy-MWD data reflect the increasing amount of tracer incorporated in macroaggregates during the growing season.



View larger version (46K):
[in this window]
[in a new window]
  		Fig. 5. Tracer sphere Dy mean weight diameter (MWD). Error bars are standard deviations (n = 4).

 
For proper comparison with Dy MWD, the soil-aggregate MWD data were recalculated excluding the size fractions <250 µm. Figure 6 compares the soil versus Dy MWD over the two growing seasons. Comparing the Dy MWD with the soil MWD values provides a better representation of the distribution of the tracer spheres. During the first growing season, Dy-MWD values were generally higher than soil MWD values in the Breton samples, while in Malmo samples the Dy-MWD values began lower then increased beyond the soil MWD values. A Dy-MWD value larger than soil MWD would indicate a preferential enrichment of the largest aggregate size classes.



View larger version (35K):
[in this window]
[in a new window]
  		Fig. 6. Comparison of soil aggregate versus Dy MWD in straw amended plots from (a) Malmo and (b) Breton soils. Error bars are standard deviations (n = 4).

 
These observations may be because of tracer spheres skipping the intermediate-size fractions and becoming incorporated into large macroaggregates because these larger aggregates are formed first. Indeed, as the free tracer pool is depleted over the course of the growing season, increases are observed most clearly in the >4-mm fraction (data not shown). These results may provide evidence to support a model of aggregate formation where macroaggregates are formed first and smaller aggregates are subsequently created when the larger become unstable and are disrupted (Angers et al., 1997; Gale et al., 2000; Golchin et al., 1994; Oades, 1984; Six et al., 2000). However, it is possible that because of the architecture of soil macroaggregates, the 400-µm tracer spheres can be incorporated in only the largest aggregate fraction because aggregates <4 mm in diameter cannot accommodate structural units of ~400 µm. However, in studies using 60Co-tagged aggregates of various sizes, Toth and Alderfer (1960) noted that as the size of tagged water-stable aggregates decreases, its contribution to the formation of larger aggregates also diminishes, suggesting that macroaggregates are made of few large constituents rather than many smaller ones. Observations of similar incorporation patterns using much smaller tracer spheres may provide the evidence necessary to determine whether the current observations are artefacts.

Dynamics of Tracer Sphere Incorporation
Since the soil was at steady-state in terms of aggregate MWD over the longer-term, incorporation of the tracer spheres was not simply because of net aggregate formation or breakdown, but was mostly because of the internal cycling of aggregates within a short-term cyclical dynamic. The initially rapid rise followed by flattening or decrease in normalized tracer contents of the aggregated pool (Fig. 4) and Dy MWD (Fig. 5) data suggest that tracer incorporation was approaching equilibrium with soil aggregate dynamics. The apparent rapid approach to equilibrium indicates that the turnover time of macroaggregates in the tilled soils studied is short and that the internal turnover of macroaggregates is more rapid in the Breton soil than in the Malmo soil. However, turnover rate can only be estimated by kinetic analysis of the rate of tracer incorporation or release. As a first approximation of macroaggregate turnover rates, the normalized tracer Dy content of the incorporated fraction (>1 mm) during the 1998 growing season can be fit to the first-order kinetic model y = A (1 - ekt). Results of the analyses generated macroaggregate mean residence times of 33 and 26 d in the Malmo soil with and without straw, and 6.5 and 5.8 d in the Breton soil with and without straw (Table 5). However, the model fit for the Breton soil was quite poor (R2 = 0.04). A more detailed analysis quantifying aggregate dynamics and turnover rates was achieved by the development of a compartmental model, which is reported in Plante et al. (2002). The model generated mean residence times for soil macroaggregate components >1 mm ranging from 33 to 95 d, and were generally two to three times shorter for Breton than for Malmo. Our results indicate that while a soil may not express changes in net aggregate formation or disruption from year to year significant turnover may occur within a season as ~40 to 60% of the tracer spheres were incorporated into large macroaggregates in a period of ~72 d.


View this table:
[in this window]
[in a new window]
  		Table 5. Kinetic parameters{dagger} from model fitting of normalized tracer Dysprosium (Dy) content of aggregate pool (>1 mm).

 
Linking Organic Matter Dynamics and Macroaggregate Dynamics
The supply of fresh organic matter in the straw amendment treatment was thought to supply binding materials for stable aggregates. While the treatment generally increased soil aggregate MWD, no significant differences were observed for tracer incorporation into >1-mm aggregates (P = 0.12 for Malmo, P = 0.07 for Breton) or for tracer Dy MWD (P = 0.31 for Malmo, P = 0.13 for Breton). Allmaras et al. (1996) found high colocation of oat residues and tracer spheres, but our results suggest that tracer incorporation was independent of the presence or absence of fresh residue particulate organic matter. It is important to note that our observations do not contradict the role particulate organic matter may play in the formation and stabilization of aggregates.

The original hypothesis of this study was that a rapid organic C turnover and low C content would be indicative of a high rate of aggregate turnover. The Breton soil respired a greater proportion of soil organic C and generally showed more rapid aggregate turnover, when compared with the Malmo soil. Physical and tracer analyses indicate that large macroaggregates in the Breton soil are easily slaked and turnover rapidly, which would not provide a stable environment for the physical protection of incoming organic materials. This phenomenon may be attributable to both soil organic C and clay contents, which are both contributors to stable soil aggregation (Kay, 1998). While the macroaggregates in the Malmo soil were more stable, the rate of tracer sphere incorporation suggests that a significant portion of macroaggregates in the Malmo soil may turnover within a growing season. This coupled with the high stability of the formed aggregates in the Malmo soil suggests that aggregate turnover provides the opportunity for the physical protection of incoming organic materials and its subsequent stabilization by chemical or physical sorption mechanisms. The relationship between soil aggregate turnover and organic C dynamics in the two soils observed in this experiment is perhaps evidence for the self-protection of organic matter proposed by Angers and Chenu (1998) and Balesdent et al. (2000), which suggests that the protection of organic matter in soil is enhanced in soils through a feedback process involving protection of new organic matter in aggregates stabilized by native organic matter.

Differences in aggregate stability and tracer incorporation are most evident at the end of the growing season, when much of the supply of new labile organic matter has been depleted. In the Breton soil, which has a lower organic C content, macroaggregates may lack resiliency such that the recovery from the disruption caused by a rainfall event and reformation of aggregates is insufficient to continue to occlude organic materials, resulting in higher relative rates of respiration. In a soil with higher organic matter content, aggregates are sufficiently resilient that after disruption events aggregates are reformed and stabilized, and therefore can continue to protect organic materials.

Soil macroaggregate formation models propose that particulate organic matter can act as a nucleus for aggregate formation (e.g., Golchin et al., 1994). This suggests that straw residue additions may increase rates of aggregation formation. However, observations that straw addition marginally increased macroaggregate residence times may reflect the increased stability of the aggregate formed around the particulate organic matter. Conversely, the lack of straw addition will result in the generation of aggregates of lower stability, which would breakdown more easily and sooner, and would result in higher values for aggregate turnover rates.

Physical protection of particulate organic matter in macroaggregates requires a balance between opposing mechanisms during aggregate turnover, occlusion, and exposure. Rapid aggregate turnover may increase the probability of exposure of previously protected organic matter. However, some aggregate turnover is necessary for physical protection to occur. If the process of organic C stabilization requires the occlusion of particulate organic matter in aggregates as an initial step, there must be a means by which the organic matter enters the aggregate. Without the breakdown and reformation of macroaggregates, there is little opportunity for the creation of new occlusion sites and recent inputs of particulate organic matter may remain exposed. Therefore, the periodic formation of new aggregates from the products of previous aggregate breakdown; i.e., aggregate turnover, likely provides an opportunity for not only the loss, but also the protection of organic C. The physical protection of organic matter within soil macroaggregates may be ephemeral, but it promotes the long-term retention of organic matter in soil by increasing the contact time with reactive surfaces such as clay or other organic matter, which increases the opportunity for chemical or physical sorption. However, the threshold between the exposure of organic C to microbial decomposition, and occlusion and physical protection during soil aggregate turnover remains unclear.


   CONCLUSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
REFERENCES
 
A field experiment was conducted to observe macroaggregate dynamics in two soils with contrasting organic matter dynamics. Results showed that the Breton soil contained less organic matter when compared with the Malmo soil, and respired a greater proportion of the soil organic C. Over the course of the field study, tracer spheres were detected in larger aggregate fractions than allowed by their size, thus providing a direct observation of gross rates of macroaggregate formation. The observed tracer incorporation reflects the internal cycling or turnover of soil macroaggregates occurring within the growing season since net soil aggregation was under steady-state conditions over the longer term. The incorporation of the tracers into macroaggregates appeared to approach equilibrium with soil aggregates within two growing seasons, reflecting a rapid turnover of macroaggregates in both soils studied. While lower soil organic matter protection in the Breton soil may been attributed to the increased turnover of macroaggregates, observations of rapid macroaggregate dynamics in the Malmo soil, with high organic matter content, suggests that macroaggregate turnover may also provide the opportunity for the occlusion of organic materials. Just as experiments relying on the accumulation of organic C in aggregates cannot be used to determine to aggregate turnover rates with certainty, our experiments using only a physical tracer were insufficient to provide a direct link between organic matter dynamics and aggregate turnover. We propose that experiments coupling an organic C tracer (e.g., 13C) and a physical tracer such as smaller spheres as those used here may generate the clear link sought.


   ACKNOWLEDGMENTS
 
The authors thank Mr. J.C. Stensrud of Kinetico Inc. for the kind contribution of Macrolite material and addition of Dy-label during manufacture, Dr. M.J.M. Duke for assistance with instrumental neutron activation analysis, and an anonymous reviewer for comments and new insights. This work has been supported by NSERC operating funds (to WBM) and an Agriculture & Agri-Food Canada scholarship (to AFP).


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
The Research for this paper was performed at the Department of Renewable Resources, University of Alberta.

Received for publication June 21, 2001.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 Soil Sci.Home page
G. Yoo and M. M. Wander
Tillage Effects on Aggregate Turnover and Sequestration of Particulate and Humified Soil Organic Carbon
Soil Sci. Soc. Am. J., May 1, 2008; 72(3): 670 - 676.
[Abstract] [Full Text] [PDF]

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 HighWire
Right arrow Citing Articles via ISI Web of Science (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Plante, A. F.
Right arrow Articles by McGill, W. B.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Plante, A. F.
Right arrow Articles by McGill, W. B.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Plante, A. F.
Right arrow Articles by McGill, W. B.
Related Collections
Right arrow Structure and Properties
Right arrow Soil Conservation
Right arrow Soil Organic Matter

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