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 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 Web of Science (10)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hannam, K. D.
Right arrow Articles by Wasylishen, R. E.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Hannam, K. D.
Right arrow Articles by Wasylishen, R. E.
Agricola
Right arrow Articles by Hannam, K. D.
Right arrow Articles by Wasylishen, R. E.
Related Collections
Right arrow Humic Substances
Right arrow Soil Organic Matter
Right arrow Forest Soils
Published in Soil Sci. Soc. Am. J. 68:1735-1743 (2004).
© 2004 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA

DIVISION S-7—FOREST & RANGE SOILS

Forest Floor Composition in Aspen- and Spruce-Dominated Stands of the Boreal Mixedwood Forest

K. D. Hannama,*, S. A. Quideaua, S.-W. Ohb, B. E. Kishchukc and R. E. Wasylishenb

a Dep. of Renewable Resources, 442 Earth Sciences Bldg., Univ. of Alberta, Edmonton, AB Canada T6G 2E3
b Dep. of Chemistry, E344 Gunning/Lemieux Chemistry Centre, Univ. of Alberta, Edmonton, AB Canada T6G 2G2
c Natural Resources Canada, Canadian Forest Service, 5320-122 St., Edmonton, AB, Canada T6H 3S5

* Corresponding author (khannam{at}ualberta.ca)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The ability of high-resolution cross-polarization magic-angle spinning 13C nuclear magnetic resonance spectroscopy (CPMAS 13C NMR) to characterize soil organic matter (SOM) has been previously demonstrated, but rarely has this information been directly related to local environmental conditions that affect SOM formation. In this study, CPMAS 13C NMR was used to characterize the forest floor (Oe + Oa horizon) of stands dominated by trembling aspen (Populus tremuloides Michx.) or white spruce [Picea glauca (Moench) Voss] in the boreal mixedwood forest of Alberta, Canada. Aromatic C content was higher and carbonyl C content was lower in the forest floor of spruce stands than in aspen stands. Within stand types, correlation analyses indicated significant relationships between the composition of the forest floor and soil temperature, mass of the Oi horizon, and mass of the moss layer. However, these relationships could not explain observed differences in the chemical composition of the forest floor between stand types. Although forest floor from spruce stands was largely composed of moss, which is low in aromatic C, it had a greater aromatic C content than forest floor from aspen stands, where moss was rare. Furthermore, a lack of significant correlations across stand types suggests that there are different relationships between the chemical and environmental characteristics of forest floor from spruce and aspen stands.

Abbreviations: ALK, alkyl carbon region • AROM, aromatic carbon region • CARB, carbonyl carbon region • CPMAS 13C NMR, cross-polarization magic-angle spinning 13C nuclear magnetic resonance • CP, cross-polarization • DD, dipolar-dephased • EMEND, Ecosystem Management Emulating Natural Disturbance • NMR, nuclear magnetic resonance • O-ALK, O-alkyl carbon region • SOM, soil organic matter


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
HIGH-RESOLUTION solid-state CPMAS 13C nuclear magnetic resonance (NMR) spectroscopy is an exciting analytical tool for the characterization of SOM chemistry because, unlike more traditional techniques, it permits the direct study of whole-soil samples or SOM fractions without prior treatment (Knicker and Lüdemann, 1995; Faz Cano et al., 2002). The potential of CPMAS 13C NMR to detect differences in forest soil chemistry among vegetation communities has been demonstrated in numerous studies, and observed differences in soil chemistry have been attributed to a variety of soil-forming factors including climate, vegetation, and landscape position. For example, de Montigny et al. (1993) hypothesized that differences in the composition of forest floor from stands dominated by western hemlock [Tsuga heterophylla (Raf.) Sarg.] or by western redcedar (Thuja plicata Don ex D. Don) on northern Vancouver Island, Canada, were caused by variability in soil moisture and the abundance of tannin-rich shrubs. In a comparison of soils from a grassland and a recently afforested site in New Zealand, Condron and Newman (1998) attributed higher O-alkyl C and lower alkyl C levels in the grassland soils to greater rates of litter input in the grassland than in the forest. By comparing a biosequence to an elevational transect, Quideau et al. (2001) concluded that vegetation, rather than climate, was controlling SOM composition. Under oak (Quercus), SOM was dominated by carbonyl C, whereas under manzanita (Arctostaphylos), SOM was dominated by O-alkyl C, and under coniferous vegetation by alkyl C. Finally, Zech et al. (1989) reported that aromatic and alkyl C of uncultivated soils collected from Germany, Spain, and Liberia varied strongly with regional patterns of precipitation and temperature.

Studies such as these have been useful for detecting patterns in SOM chemistry among different ecosystems. However, because of time constraints and limited access to instrumentation, samples are typically composited before NMR analysis, thereby obscuring within-sample variability, and precluding statistical analysis of treatment differences or correlations between NMR results and environmental variables. Only very rarely has statistical analysis of NMR results been specifically related to surrounding environmental conditions. In one of these studies, Faz Cano et al. (2002) suggested that statistical differences in O-alkyl and aromatic C in A-horizon soil among three vegetation communities and two climatic zones in Spain were due to variability in soil temperature. Preston et al. (2002) determined that forest floor from west-facing slopes on southern Vancouver Island, Canada, possessed more lignin features and fewer charcoal features than forest floor from east-facing slopes, which they hypothesized was the result of differences in the rates of blowdown and wildfire. Unfortunately, environmental variables were not measured in these studies, so the proposed explanations could not be verified. The use of statistical analysis to relate measured environmental characteristics, such as soil temperature, to NMR analyses could prove valuable in differentiating the effects of various soil-forming factors on SOM chemical composition.

The Ecosystem Management Emulating Natural Disturbance (EMEND) experiment, in northwestern Alberta, is a long-term research study covering 1000 ha of boreal mixedwood forest that includes stands dominated by trembling aspen or white spruce. This controlled and replicated experiment was established, in part, to examine differences in ecosystem-level processes among stand types of the boreal mixedwood forest. A recent study of forest soils at EMEND revealed higher forest floor C contents (kg ha–1), total C (%), and C/N ratios in white-spruce-dominated (SPRUCE) stands than in trembling aspen-dominated (ASPEN) stands (Kishchuk, 2002). Stand-type differences in the organic matter composition of the forest floor warrant further investigation because they may have implications for nutrient cycling processes and for the productivity of regenerating vegetation following logging. Indeed, the results of previous studies in the boreal mixedwood forest have suggested that forest floors in ASPEN stands are of higher quality (i.e., they contain more labile organic material) and support more rapid nutrient cycling than forest floors in SPRUCE stands (Flanagan and Van Cleve, 1983; Paré and Bergeron, 1996; Ste-Marie and Paré, 1999; Coté et al., 2000).

The objectives of this study were:

  1. To compare the chemical and local environmental characteristics (e.g., temperature, moisture, mass of surface materials) of forest floor material from stands dominated by white spruce or trembling aspen in the boreal mixedwood forest.
  2. To explore relationships among measured chemical and environmental characteristics of these forest floors to gain insight into why these differences might occur.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Study Site and Sampling
Samples were collected in June 2002 from the EMEND experiment (56° 46' 13'' N, 118° 22' 28'' W). The EMEND site is located on the boreal mixedwood plain in the Clear Hills Upland EcoRegion within the Boreal Plains EcoZone (Wiken, 1986; EcoRegions Working Group, 1989). The area is characterized by cold winters (mean temperature –14.0°C), warm summers (mean temperature 11.6°C), and an average of 433 mm of precipitation, 2/3 of which usually falls during the summer (Environment Canada, 2002). The EMEND site is characterized by a rolling topography ranging in elevation from 677 to 880 m above sea level. Soils are usually haplocryalfs developed on fine-textured glacio-lacustrine parent material (Soil Survey Staff, 1998; Kishchuk, 2002).

Samples were collected from three 10-ha replicates each of undisturbed SPRUCE and ASPEN stands (six experimental units in total), which range in age from 80 to 140 yr old. The SPRUCE stands consisted of >70% white spruce, with some trembling aspen, balsam poplar (Populus balsamifera L.), paper birch (Betula papyrifera Marshall), balsam fir [Abies balsamea (L.) Mill], or lodgepole pine (Pinus contorta Douglas ex Loudon). The understory of SPRUCE stands includes Rosa acicularis Lindl., Sheperdia canadensis (L.) Nutt, and a dense ground cover of moss, especially step moss (Hylocomium splendens). Forest floors in SPRUCE stands are typically Humimors. Humimors are Mors whose profile is dominated by an Oa (H) horizon that contains few recognizable plant residues (Green et al., 1993). Selected chemical data for the Oe + Oa horizon in SPRUCE stands are shown in Table 1. The ASPEN stands consisted of >70% trembling aspen, with some of the tree species listed above for SPRUCE stands. The understory of ASPEN stands includes Rosa acicularis Lindl., Viburnum edule (Michx.) Raf., and Alnus spp., with an herb layer of Calamagrostis canadensis, Epilobium angustifolium L., and Cornus canadensis L.. Forest floors in ASPEN stands are typically Mormoders. Mormoders are Moders whose diagnostic Oe (F) layer possesses evidence of both faunal activity and fungal hyphae; mormoders are considered an intergrade between moders and mors (Green et al., 1993). Selected chemical data for the Oe + Oa horizon in ASPEN stands are provided in Table 1. Six sampling sites were randomly selected within each experimental unit. At each sampling site, the Oi horizon and moss were collected from within a 15 by 15 cm template that was placed on the forest floor surface. Leaves, needles, twigs, bark, seeds, and cones were included in the Oi horizon. This material was placed in plastic bags and kept on ice until it was transported to the laboratory where it was stored at ±5°C for a maximum of 90 d. Below the Oi horizon and moss layer, the Oe + Oa horizon of the forest floor within the 15 by 15 cm template was excavated to the depth of the mineral soil surface. All of the Oe + Oa material was placed in a plastic bag and kept on ice until it was transported to the laboratory, where it was stored at ± 5°C for a maximum of 30 d. The thickness of the Oe + Oa horizon (from the surface of the Oe horizon to the surface of the mineral soil) was measured at the four corners of the 15 by 15 cm cavity. Immediately after sample collection, temperature measurements were taken 5 cm below the surface of the Oe horizon and 5 cm below the mineral soil surface using a temperature probe.


View this table:
[in this window]
[in a new window]
  		Table 1. pH, total C (%), total N (%), and C/N ratio of Oe + Oa horizon material from white-spruce-dominated (SPRUCE) and trembling-aspen-dominated (ASPEN) stand (n = 3). Values are means with standard deviation in parentheses. pH was determined using a 1:10 ratio of forest floor (fresh weight in grams) to 0.01 M CaCl2 (milliliters).

 
Sample Preparation and Nuclear Magnetic Resonance Analyses
Samples of the Oi horizon and moss were separated in the lab, dried at 65°C for 48 h, and weighed. After the moss and the Oi horizon materials were separated, the proportion of green moss in each moss sample was visually estimated to the nearest 5% and used to calculate the mass of green moss per sample. Samples of the Oe + Oa horizon were sieved (6.3 mm) to remove roots and twigs, thoroughly mixed, and dried at 65°C for 48 h. Dried samples of the Oe + Oa horizon were finely ground using a ball mill. To compare the NMR spectra of the Oe + Oa horizon with specific litter types found in the Oi horizon, composite samples of white spruce needle litter, trembling aspen leaf litter (the most common materials in the litterfall of SPRUCE and ASPEN stands, respectively; Lindo and Visser, 2003) and step moss were also dried at 65°C for 48 h and ground using a ball mill.

For two of the three replicates of SPRUCE and ASPEN stands, samples of the Oe + Oa horizon collected from each of the six sampling sites within a replicate were composited before NMR analysis. For the remaining replicate of each stand type, NMR analysis was performed separately on the six subsamples collected within the replicate, to examine variability among samples.

Solid-state 13C CPMAS NMR experiments were performed on a Varian Chemagnetics CMX Infinity 200 [B0 = 4.7 T, {nu}L(13C) = 50.3 MHz] NMR spectrometer using a 7.5-mm double-resonance MAS probe with high-power 1H decoupling. All samples were packed into 7.5-mm (o. d.) rotors with Zirconia (ZrO2) sleeves, drive tips made of Kel-F, and end caps and spacers made of Teflon (DuPont, Circleville, OH). All 13C NMR spectra were acquired using cross-polarization (CP), and were referenced to TMS ({delta}iso = 0.0 ppm) by setting the high-frequency isotropic peak of solid adamantane to 38.56 ppm (Earl and VanderHart, 1982; Bryce et al., 2001). The 1H 90° pulse and Hartmann-Hahn matching conditions were also determined using this sample. All 13C CP NMR spectra were acquired using a 1H 90° pulse width of 4.5 µs, a pulse delay of 5.0 s, a contact time of 1.0 ms, an acquisition time of 17.1 ms, and a spinning frequency of 6.5 kHz. One thousand transients were collected for each sample of Oe + Oa horizon material, litter, and moss that was analyzed. Two-pulse phase modulation (Bennett et al., 1995) with a 1H decoupling field of 56 kHz was employed during the acquisition of all spectra. A Gaussian line broadening of 100 Hz was used to process all spectra. Contribution of the background signal to the spectra was determined by acquiring a spectrum of an empty rotor set under identical conditions as for the Oe + Oa horizon material, litter, and moss samples. This contribution was subtracted from all Fourier-transformed 13C NMR spectra before analysis.

Bruker's WIN-NMR package was used to estimate the relative integrated areas of various regions between 0 and 194 ppm. Many different spectral regions have been reported for the integration of 13C NMR spectra (e.g., Skjemstad et al., 1997; Mao et al., 2000; Preston et al., 2000). In this study the spectral divisions were assigned based on local minima of the spectra. The following regions were used for integration: 0 to 45 ppm, attributed to alkyl C (ALK); 45 to 112 ppm, attributed to O-alkyl C (O-ALK); 112 to 165 ppm, attributed to aromatic C (AROM), and 165 to 194 ppm, attributed to carbonyl C (CARB).

Dipolar-dephased (DD) spectra of Oe + Oa horizon material from SPRUCE and ASPEN stands were produced by inserting a delay period of 40 µs (in the absence of 1H decoupling) between the CP and the acquisition portions of the CPMAS pulse sequence (Hatcher, 1987). Peaks in spectra generated by DD correspond either to quaternary C or to C capable of some motion in the solid state (e.g., acetate or CH2 in long chains; Lorenz et al., 2000). As a consequence, features of lignins and tannins can be more easily distinguished using DD than using CPMAS (Hatcher, 1987; Wilson and Hatcher, 1988; Lorenz et al., 2000). Peaks in CPMAS and DD spectra were compared to determine the relative importance of lignins and tannins in the Oe + Oa horizon material from SPRUCE and ASPEN stands.

Statistical Analyses
Data collected using solid-state CPMAS 13C NMR experiments are considered ‘semi-quantitative’, primarily because of the variability in CP efficiencies and rates of relaxation among C atoms in different functional groups (Preston et al., 1997; Ussiri and Johnson, 2003; Smernik and Oades, 2003). As a result of this phenomenon, C in some chemical environments may be under-represented. This problem is of particular concern for C atoms that are not directly bonded to H atoms, such as those in highly condensed aromatic structures (Preston et al., 1997). However, the results of CPMAS 13C NMR experiments are reproducible (Peuravuori et al., 2003). As a result, this technique can be used with confidence when the ultimate goal of the analyses is the comparison of trends and patterns among similar types of samples that have been analyzed under identical conditions (Kinchesh et al., 1995; Preston et al., 1997; Peuravuori et al., 2003), as was the case in our study.

The chemical and physical characteristics of the Oe + Oa horizon from SPRUCE and ASPEN stands were compared using analysis of variance for a completely randomized design. Before analysis, the temperatures of the Oe + Oa horizon and mineral soil, and the moisture contents of the Oe + Oa horizon were transformed (–1/x2) to meet the assumptions of normality and homogeneity of variance. Differences were considered statistically significant if P < 0.05. Data transformations were not effective in obtaining a normal distribution for the mass of moss on the forest floor surface. Therefore, differences in the mass of moss from SPRUCE and ASPEN stands were compared using a Kruskal–Wallis nonparametric test after ranking the moss data (Zar, 1984). To examine relationships between the chemical and environmental characteristics of these forest floors, Pearson correlations were calculated across stand types (i.e., data from ASPEN and SPRUCE stands were analyzed together) and within stand types (i.e., data from ASPEN and SPRUCE stands were analyzed separately). Correlation analysis was performed only on the data for the six subsamples from one SPRUCE stand and six subsamples from one ASPEN stand that were not composited before NMR analysis. For the moss data, correlations were performed only on the data from the SPRUCE stand because there was no moss present at several sampling sites in the ASPEN stand. Correlations were considered statistically significant if P < 0.05. Bonferroni corrections were used in the correlation analyses of data from SPRUCE and ASPEN stands to test for overall significance of the correlation matrices, which were considered not significant if the adjusted P-value was lower than the smallest P-value in the correlation matrix (Legendre and Legendre, 1998). All statistical analyses were performed using SAS (version 8.01, SAS Institute Inc., Cary, NC).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Forest Floor Chemical Characteristics
Major signals in the Oe + Oa horizon spectra were found around 73 ppm, and were characteristic of the C-2, C-3, and C-5 carbons of cellulose and hemicelluloses (Fig. 1) . The shoulder at 63 ppm was assigned to the C-6 carbon in carbohydrates, while anomeric carbons were noticeable around 105 ppm (Teeaar and Lippmaa, 1984). In the ALK region of the spectra, the main peaks occurred around 30 ppm, suggesting that alkyl carbons present in the Oe + Oa horizons of the two stands were mainly of the polymethylene type (Keeler and Maciel, 2000). The methoxyl C signal characteristic of lignins was apparent as a shoulder at 56 ppm in all spectra, although it was less well-resolved in spectra derived from Oe + Oa horizons of SPRUCE stands. In the AROM region of the spectra, the small peaks at 130 to 131 ppm probably originated from C-substituted aromatic carbons, such as the C-1 carbon of guaiacyl and syringyl units, or the C-1, C-2, and C-6 carbons of p-hydroxyphenyl lignin moieties (Fig. 1). The C-2 and C-6 carbons of syringyl lignin units likely contributed to the peak centered around 105 ppm (Preston et al., 2000), while the peak at 117 ppm may be derived from the C-2, C-5 and C-6 carbons of lignin guaiacyl units (Landucci et al., 1998; Preston et al., 2000). The C-3 carbons of guaiacyl units and the C-3 and C-5 carbons of syringyl units typically contribute a broad signal at 151 to 154 ppm (Landucci et al., 1998; Lorenz et al., 2000; Preston et al., 2000). This peak was apparent on all spectra, but the Oe + Oa horizon material from SPRUCE stands showed an additional peak at 145 ppm, which was absent in Oe + Oa horizons from ASPEN stands (Fig. 1). Methoxylated C-3 carbons of the guaiacyl moieties were observed around 145 to 148 ppm (Landucci et al., 1998; Preston et al., 2000), although the occurrence of well-resolved maxima at 145 and around 154 ppm in the SPRUCE spectra are a characteristic marker for condensed tannins (Preston et al., 2000). In addition, tannins and tannin-like structures may have contributed to the signals at 105, 117, and 130 ppm. Finally, the peak at 175 ppm was indicative of the carbonyl C in acetyl and ester moieties (Skjemstad et al., 1997).




View larger version (36K):
[in this window]
[in a new window]
  		Fig. 1. Representative cross-polarization magic-angle spinning (CPMAS) and dipolar-dephased (DD) 13C nuclear magnetic resonance (NMR) spectra of the Oe + Oa horizon of a white-spruce-dominated (SPRUCE) and a trembling-aspen-dominated (ASPEN) stand.

 
The distribution of peaks in the DD spectra confirmed that the content of condensed tannins is different in the Oe + Oa horizon of SPRUCE and ASPEN stands. In DD spectra of Oe + Oa horizon material from SPRUCE stands, the methoxyl signal of lignin, at 56 ppm, was not evident as a clear peak (Fig. 1), while the strong peak around 130 ppm, and the presence of two peaks at 145 and 152 ppm are characteristic of materials with high tannin contents (Preston et al., 1997; Lorenz et al., 2000). In contrast, DD spectra of Oe + Oa horizon material from ASPEN stands exhibited a clear peak at 56 ppm (Fig. 1), consistent with a high lignin content. There was no strong peak at 130 ppm in DD spectra of Oe + Oa horizon material from ASPEN stands, and only a single peak at 151 ppm. These observations indicate that lignins are relatively less abundant, and condensed tannins are relatively more abundant, in the Oe + Oa material from SPRUCE stands than from ASPEN stands (Preston et al., 1997; Lorenz et al., 2000).

The reproducibility of the NMR analysis was tested by performing the same analysis six times on a single sample on separate days, and integrating the resulting spectra (Table 2). The confidence intervals for the ALK, O-ALK, AROM, and CARB regions of the six NMR spectra obtained from this one sample ranged from ±0.6 to ±0.9. These were small compared with the confidence intervals obtained from the analysis of six separate subsamples of Oe + Oa horizon material from an ASPEN or SPRUCE stand, which ranged from ±0.9 to ±2.5 (Table 2). Therefore, differences in the NMR spectra of the Oe + Oa horizon from the two stand types can be confidently assigned to differences in forest floor composition and not to random errors, which might occur from the NMR analysis and integration of the spectra.


View this table:
[in this window]
[in a new window]
  		Table 2. Confidence intervals around alkyl (ALK), O-alkyl (O-ALK), aromatic (AROM), and carbonyl (CARB) regions of 13C nuclear magnetic resonance (NMR) spectra obtained from (i) one sample of Oe + Oa horizon material analyzed repeatedly (n = 6), and (ii) uncomposited subsamples of Oe + Oa horizon material collected from a white-spruce-dominated (SPRUCE) and a trembling aspen-dominated (ASPEN) stand and analyzed separately (n = 6). Confidence intervals were calculated at {alpha} = 0.05.

 
All 13C NMR spectra were dominated by the O-ALK region, followed by the ALK, the AROM, and finally the CARB region (Fig. 2) . However, integration of the 13C NMR spectra indicated differences in the composition of the Oe + Oa horizon from SPRUCE and ASPEN stands. The Oe + Oa horizon from SPRUCE stands consisted of 21.4 ± 1.8% alkyl, 52.9 ± 2.1% O-alkyl, 17.4 ± 0.1% aromatic, and 8.3 ± 0.4% carbonyl C while the Oe + Oa horizon from ASPEN stands consisted of 22.8 ± 1.6% ALK, 52.1 ± 1.9% O-ALK, 14.5 ± 1.5% aromatic, and 10.6 ± 0.4% carbonyl C. Aromatic C was significantly more abundant in the Oe + Oa horizons of SPRUCE stands than in those of ASPEN stands (P = 0.027). Carbonyl C was significantly more abundant in the Oe + Oa horizons of ASPEN stands than in those of SPRUCE stands (P = 0.002).



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 2. Distribution of C in the alkyl, O-alkyl, aromatic, and carbonyl regions of 13C NMR spectra obtained from the Oe + Oa horizon of white-spruce-dominated (SPRUCE) and trembling-aspen-dominated (ASPEN) stands. Error bars indicate one standard deviation (n = 3). Different letters indicate significance at the 0.05 probability level.

 
Surface Material Chemical Characteristics
As with the Oe + Oa horizon, the greatest amount of organic C in white spruce needle litter, trembling aspen leaf litter, or step moss was detected in the O-ALK region (Fig. 3) . The smallest amount of organic C in white spruce needle litter and trembling aspen leaf litter was detected in the CARB region. In contrast, the smallest amount of organic C in step moss was detected in the AROM region (Table 3) . Lower levels of aromatic and carbonyl C and higher levels of O-alkyl C were found in white spruce needle litter, trembling aspen leaf litter, and step moss than in the Oe + Oa horizon from either stand type. This was especially true for step moss, which consisted of more than 75% O-alkyl C and only 4% aromatic C.



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 3. Representative 13C NMR spectra of white spruce needle litter, trembling aspen leaf litter and step moss.

 

View this table:
[in this window]
[in a new window]
  		Table 3. Distribution of C in alkyl (ALK), O-alkyl (O-ALK), aromatic (AROM), and carbonyl (CARB) regions of 13C NMR spectra obtained from composite samples of trembling aspen leaf litter, white spruce needle litter, and step moss collected from the forest floor surface.

 
In the ALK region, spectra from step moss and white spruce needle litter showed a peak around 20 ppm, characteristic of acetate CH3 (Preston et al., 2000). Trembling aspen leaf litter exhibited a peak around 33 ppm, which may be due to C in long chains of CH2 that are more rigid than those resonating at 30 ppm (Lorenz et al., 2000). Peaks at 130 to 131 and 151 to 157 ppm, associated with lignins and condensed tannins, were notably absent in the spectrum from step moss (Fig. 3). The split peak at 169 and 174 ppm in the spectrum from white spruce needle litter may reflect the presence of cutins (Preston et al., 2000).

Relation to Environmental Variables
There were strong differences between SPRUCE and ASPEN stands in the quantity and type of materials on the forest floor surface, and in the characteristics of the Oe + Oa horizon itself (Table 4). The mass of moss on the surface of the forest floor was greater in SPRUCE stands than in ASPEN stands (P = 0.046), and the mass of the Oi horizon was greater in ASPEN stands than in SPRUCE stands (although differences were not statistically significant; P = 0.057). The physical structure of the forest floors appeared to reflect these differences. The Oe + Oa horizon from SPRUCE stands was thicker (P = 0.009) with a lower bulk density (P = 0.049) than that from ASPEN stands. This is consistent with the observation that decomposing moss, which is fluffy and fibrous, appeared to form the bulk of the forest floor in SPRUCE stands. On the other hand, there were no significant differences between SPRUCE and ASPEN stands in the moisture content or temperature of the Oe + Oa horizon, or the temperature of the mineral soil (Table 4). However, two of the 18 sampling sites in SPRUCE stands were saturated and/or frozen about 20 cm below the forest floor surface at the time of sample collection; this was never observed in ASPEN stands. Furthermore, there was a trend toward lower mineral soil temperatures in SPRUCE than in ASPEN stands (P = 0.095). However, it must be stressed that forest floor temperature and moisture was measured on only one date, at the time of forest floor sample collection.


View this table:
[in this window]
[in a new window]
  		Table 4. Differences in environmental characteristics associated with the Oe + Oa horizon of white-spruce-dominated (SPRUCE) and trembling-aspen-dominated (ASPEN) stands. Values are means with standard deviation in parentheses (n = 3).

 
Correlation analyses between forest floor chemical and environmental characteristics were performed both within and across SPRUCE and ASPEN stands (Table 5). Significant relationships between chemical and environmental variables were only found within stand types and not across stand types. The lack of significant correlations between the composition of the Oe + Oa horizon and measured environmental characteristics across stand types suggests that environmental characteristics affect the composition of the Oe + Oa horizon differently in the two stand types. In SPRUCE stands, aromatic C in the Oe + Oa horizon was negatively correlated with the mass (g m–2) of green moss (P = 0.030, although the overall correlation matrix for SPRUCE stands was not statistically significant). In ASPEN stands, alkyl C in the Oe + Oa horizon was negatively correlated with the mass (g m–2) of the Oi horizon (P = 0.001), and O-alkyl C was positively correlated with mineral soil temperature (P = 0.017).


View this table:
[in this window]
[in a new window]
  		Table 5. Selected correlation coefficients (Pearson's r) describing the relationship between the chemical and environmental characteristics of the Oe + Oa horizon in a white-spruce-dominated (SPRUCE) and a trembling aspen-dominated (ASPEN) stand (n = 6). P-values were adjusted using Bonferroni corrections to test for overall significance of the correlation matrices: P < 0.004 (SPRUCE) and P < 0.006 (ASPEN).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Nuclear magnetic resonance analysis revealed statistically significant differences in the aromatic and carbonyl C contents of the Oe + Oa horizon from SPRUCE and ASPEN stands. There are several factors that could have contributed to these differences, including (i) differences in the types and amounts of litter inputs to the forest floor, (ii) variations in the processes of decomposition and humification, as affected by litter chemistry, the presence/absence of certain groups of decomposer organisms, and environmental conditions (Baldock and Preston, 1995), or (iii) the age of the forest floor.

On the surface of the forest floor in SPRUCE stands, the mass of moss was double that of the Oi horizon, while in ASPEN stands it was about 10% that of the Oi horizon (Table 4). This is consistent with Rowe's (1956) observation that moss cover in the boreal mixedwood forest tends to increase with white spruce cover. Moss tissue is low in aromatic C because, as a nonvascular plant, it contains some phenols but no lignin or tannins (Williams et al., 1998). Thus, it is not surprising that within SPRUCE stands, Oe + Oa horizons with high moss cover were associated with lower levels of aromatic C (Table 5).

The significantly higher levels of aromatic C in the Oe + Oa horizon of SPRUCE stands compared with that in ASPEN stands was surprising, given that moss was nearly ubiquitous in SPRUCE stands (Fig. 2). Levels of aromatic C in white spruce needle litter were also lower than in the Oe + Oa horizon of SPRUCE stands (Fig. 2 and Table 3). Therefore, neither moss nor white spruce needle litter are a particularly rich source of aromatic C, and levels of aromatic C in moss and white spruce needle litter do not explain the higher levels of aromatic C in the Oe + Oa horizon of SPRUCE stands compared with ASPEN stands.

In addition to litterfall, the contribution of belowground C inputs through fine root turnover must be considered (Zech et al., 1989). Strong and La Roi (1983) found that about 24% of white spruce biomass is allocated below ground. Unlike trembling aspen, whose lateral roots are typically concentrated in the mineral soil, most white spruce lateral roots are found in the forest floor (Strong and La Roi, 1983). Therefore, fine roots probably contribute a larger amount of C to the Oe + Oa horizons of SPRUCE stands than to ASPEN stands at EMEND. Fine roots were not analyzed in this study, but aromatic C in needles and fine roots of mature Norway spruce in southern Germany were similar to each other and to that found in white spruce needle litter in this study (11–13%; Rosenberg et al., 2003). Consequently, inputs of C from the fine roots of white spruce cannot fully explain the higher concentrations of aromatic C in the Oe + Oa horizon of SPRUCE stands.

Woody litter inputs could also cause higher levels of aromatic C in the Oe + Oa horizon of SPRUCE stands compared with that in ASPEN stands. However, the CPMAS and DD spectra of Oe + Oa horizon material indicated that relatively more condensed tannins had accumulated in the forest floors of SPRUCE stands while relatively more lignin had accumulated in the forest floors of ASPEN stands (Fig. 1). A buildup of condensed tannins in SPRUCE forest floor was unexpected because, unlike lignin, condensed tannins are not considered particularly resistant to degradation (Lorenz et al., 2000). Nonetheless, an accumulation of condensed tannins was also observed in the humus of a northern Ontario black spruce forest, and was hypothesized to be the result of environmental conditions that inhibited decomposition in the forest floor (Lorenz et al., 2000). Spruce stands in the Alaskan taiga tend to have higher soil moisture contents than ASPEN stands (Van Cleve and Powers, 1995). The dense rich moss cover on the forest floor surface in SPRUCE stands suggests that this is also the case at the EMEND sites. Differences in the temperature and moisture content of the Oe + Oa horizons of SPRUCE and ASPEN stands at EMEND may not have been detected because measurements were taken only once, after several weeks of hot, dry weather. Therefore, the higher levels of aromatic C in the Oe + Oa horizon of SPRUCE stands may be due to microclimatic conditions, such as high moisture contents, that hinder decomposition and favor the accumulation of condensed tannins in these forest floors.

Lower levels of carbonyl C in the Oe + Oa horizon of SPRUCE stands than in ASPEN stands support the hypothesis that higher levels of aromatic C in the Oe + Oa horizon of SPRUCE stands are the result of differences in the pattern of decomposition in these forest floors. Carbonyl C includes carboxylic acids, aldehydes, and esters (i.e., relatively oxidized forms of C; Baldock and Preston, 1995). As organic material is aerobically decomposed, carbonyl C tends to increase (Kögel et al., 1987; Zech et al., 1987; Baldock and Preston, 1995). Therefore, lower carbonyl C levels in the Oe + Oa horizon of SPRUCE stands than in ASPEN stands may be caused by the inhibition of oxidative degradation, perhaps during periods of saturation.

In ASPEN stands, alkyl C in the Oe + Oa horizon was strongly negatively correlated with the mass of the Oi horizon (Table 5). This relationship was unexpected, because a greater mass of Oi horizon material in ASPEN stands was anticipated to indicate greater inputs of alkyl C-rich trembling aspen litter (Table 3 and Preston et al., 2000). Comminution and ingestion of litter by macrofauna, such as insects cause minimal changes to the chemical composition of litter (Fox et al., 1994). Therefore, the inverse relationship between alkyl C in the Oe + Oa horizon and the mass of the Oi horizon in ASPEN stands may reflect: (i) faunal mixing of trembling aspen leaf litter into the forest floor, or (ii) reduction of trembling aspen foliage in the overhead canopy by defoliating insects, and its deposition, as frass, in the forest floor. Both possibilities merit further investigation.

In ASPEN stands, the significant positive relationship between O-alkyl C in the Oe + Oa horizon and the temperature of the mineral soil contrasts with the results of a Spanish study, where O-alkyl C levels increased with decreasing soil temperature, presumably because SOM degradation was inhibited at low temperatures (Faz Cano et al., 2002). Although O-alkyl C levels generally decline with the degradation of litter material, there are some situations (e.g., more rapid production of O-alkyl C than alkyl or carbonyl C by soil microbes; Baldock et al., 1992), in which the relative abundance of O-alkyl C may increase with decomposition. Therefore, a positive relationship between mineral soil temperature and O-alkyl C levels in the Oe + Oa horizon may still be consistent with enhanced microbial activity under higher soil temperatures, although this clearly requires further study.

The boreal mixedwood forest is a mosaic of stand ages and tree species whose structure and composition is believed to be driven by time since stand-initiating fire (Rowe, 1961). In northern Alberta, trembling aspen dominates the canopy of early seral mixedwood stands; with increasing stand age, white spruce becomes more abundant (Rowe, 1961). Therefore, the Oe + Oa horizon of SPRUCE stands could differ from that of ASPEN stands simply because it has been allowed to develop for a longer period of time. Ratios of aromatic/O-alkyl or alkyl/O-alkyl C are commonly used as indices of the extent of organic matter decomposition (Baldock and Preston, 1995). In the Oe + Oa horizons of both SPRUCE and ASPEN stands, ratios of aromatic/O-alkyl C and alkyl/O-alkyl C were 0.3 and 0.4, respectively. The lack of a difference in either decomposition index between stand types suggests that differences in the composition of the Oe + Oa horizons are the result of variability in humification pathways, rather than a factor of soil age. The importance of different humification pathways in distinguishing the Oe + Oa horizons of the two stands types is supported by the correlation analyses. While these analyses showed significant correlations within stand types between the composition of the Oe + Oa horizon and measured environmental characteristics, they revealed no clear relationships across stand types.


   ACKNOWLEDGMENTS
 
Soils research at the EMEND study is funded by the Canadian Forest Service and Weyerhaeuser Canada Ltd. The EMEND study is funded by Canadian Forest Products Ltd. (CanFor), Daishowa-Marubeni International (DMI), the Sustainable Forest Management Network (SFMN) and the University of Alberta. The authors thank Martin Blank, Lucie Jerabkova, Candis Staley, and Roshini Nair for field and laboratory assistance, Jason Edwards for organization of the EMEND field camp, and Dr. Guy Bernard for his interest in the 13C NMR aspects of this project. This work was supported in part by a Sustainable Forest Management Network grant to SAQ and BEK, and Natural Science and Engineering Research Council grants to SAQ and REW. In addition, REW acknowledges the Government of Canada for a Canada Research Chair in Physical Chemistry, and SWO is grateful to Mokpo National University, Republic of Korea, for an award under the Professors' training program (2001).

Received for publication October 20, 2003.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 Soil Sci.Home page
A. Balaria, C. E. Johnson, and Z. Xu
Molecular-Scale Characterization of Hot-Water-Extractable Organic Matter in Organic Horizons of a Forest Soil
Soil Sci. Soc. Am. J., May 1, 2009; 73(3): 812 - 821.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
T. E. Redding, K. D. Hannam, S. A. Quideau, and K. J. Devito
Particle Density of Aspen, Spruce, and Pine Forest Floors in Alberta, Canada
Soil Sci. Soc. Am. J., August 4, 2005; 69(5): 1503 - 1506.
[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 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 Web of Science (10)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hannam, K. D.
Right arrow Articles by Wasylishen, R. E.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Hannam, K. D.
Right arrow Articles by Wasylishen, R. E.
Agricola
Right arrow Articles by Hannam, K. D.
Right arrow Articles by Wasylishen, R. E.
Related Collections
Right arrow Humic Substances
Right arrow Soil Organic Matter
Right arrow Forest 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