SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 23 May 2006
Published in Soil Sci Soc Am J 70:1189-1199 (2006)
DOI: 10.2136/sssaj2005.0186
© 2006 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
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 (5)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Belleau, A.
Right arrow Articles by Paré, D.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Belleau, A.
Right arrow Articles by Paré, D.
Agricola
Right arrow Articles by Belleau, A.
Right arrow Articles by Paré, D.
Related Collections
Right arrow Best Management Practices
Right arrow Nutrient Cycling
Right arrow Forest Soils

Forest, Range & Wildland Soils

Soil Nutrient Dynamics after Harvesting and Slash Treatments in Boreal Aspen Stands

A. Belleaua,*, S. Braisa and D. Paréb

a Chaire AFD, Dép. des Sciences Appliquées, Univ. du Québec en Abitibi-Témiscamingue, 445, Boulevard de l'Université, Rouyn-Noranda, QC, Canada, J9X 5E4
b Natural Resources Canada, Canadian Forest Service, Laurentian Forestry Centre, 1055 du P.E.P.S., P.O. Box 3800, Sainte-Foy, QC, Canada, G1V 4C7

* Corresponding author (annie.belleau{at}uqat.ca)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The effects of harvesting and slash treatments on soil nutrient dynamics were assessed in boreal aspen stands growing on mesic clayey sites. Stem-only harvested stands were compared with unharvested controls according to a complete block design with three replications. Within harvested areas, four slash treatments (stem-only harvesting [SOH], whole-tree harvesting [WHT], wood chip application, and slash burns) were compared. Treatments created a gradient of slash that ranged from 52.3 Mg ha–1 in stem-only to 13.8 Mg ha–1 in control stands. The amount of slash had no effect on wood decomposition rates but was strongly associated with higher forest floor organic C, Kjeldahl N and base cation concentrations (Cae and Mge), base saturation, pH, and effective cation exchange capacity (CEC) and lower microbial C/N when control stands were compared with stem-only harvested stands. Slash burn severity was too low to significantly reduce slash loads and induce base cation release but severe enough to reduce forest floor microbial C and N concentrations (–48 and –55%, respectively) for at least one complete growing season. Slash burn also induced increases in forest floor available P (54%) concentration compared with other slash treatments. Chipping reduced forest floor microbial N concentration by 25% and increased microbial C/N by 28% but had no impact on nutrient availability. Differences between WTH and SOH were linked to the abundance of slash. Finally, the results illustrate that whatever the treatment, the amount of slash left on the ground is the main factor found to affect soil microbial community characteristics and soil nutrient availability.

Abbreviations: CEC, effective cation exchange capacity • CHT, chipping harvesting treatment • PCA, principal component analysis • SB, slash-burn treatment • SOH, stem-only harvesting treatment • WTH, whole-tree harvesting treatment


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
FOREST HARVESTING of nutrient-demanding species such as trembling aspen (Populus tremuloides Michx.) may induce large reductions in ecosystem nutrient pools (Mann et al., 1988; Stevens et al., 1995; Paré et al., 2002) and changes in nutrient dynamics (Brais et al., 1995, 2002; Bélanger et al., 2003). Stem-only harvesting, where branches and tree tops are left on the ground, can limit nutrient export (Paré et al., 2002), but after harvesting, the contribution to nutrient dynamics of slash left on the ground still needs to be investigated. Slash has been shown to increase nitrogen net mineralization rates (O'Connell et al., 2004), but slash can also induce nutrient leaching by preventing vegetation growth (Fahey et al., 1991a) or may favor nutrient immobilization by microorganisms (Vitousek and Matson, 1985; Fahey et al., 1991b). Changes in substrate quality and soil moisture and temperature after harvesting have also been shown to affect microorganism biomass and activities (Pietikäinen and Fritze, 1993; Baldock and Nelson, 1999). In boreal forests of western Quebec, where wildfires are the major natural disturbance (Bergeron, 1991; Payette, 1992), control or slash burns could partially mimic the immediate effects of wildfires on soil (MacLean et al., 1983; Brais et al., 2000) and, as such, could induce processes that are of the same magnitude and direction as those observed after a natural disturbance. Because rate of slash decomposition is dependent on slash amount, size, composition, and proximity to soil (Harmon et al., 1986; Fahey et al., 1991b), chipping of slash should increase decomposition and present a safer alternative to slash burns. Proper slash management could contribute to maintaining productivity of forest ecosystems (Niemelä, 1999) by limiting nutrient losses and would allow for synchronism between nutrient release and stand requirements (Stevens et al., 1995).

This study reports on short-term effects of clear-cut harvesting and four slash treatments on net mineralization rates, parameters of soil microorganism community, and soil nutrient status of boreal aspen stands growing on mesic clayey sites. We compared (i) unharvested control stands with stem-only harvested stands and (ii) three slash treatments after SOH (stem harvesting without further slash manipulation, slash chipping, and slash burn) and a whole-tree harvested treatment where branches with a diameter smaller than 7 cm were removed by hand after harvesting. An abundant aspen sprouting and rapid growth after disturbance (Peterson and Peterson, 1992; Brais et al., 2004) combined with high soil clay content should contribute to nutrient retention after harvesting. We also expected higher nutrient mineralization rates after slash burn and chipping than under untreated SOH and WTH treatments and higher soil nutrient inputs from slash decomposition under SOH than under WTH.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Study Area
The study area is located in the Lake Duparquet Research and Teaching Forest in the Abitibi region of northern Quebec, 45 km northwest of Rouyn-Noranda, Quebec (48°86'–48°32' N, 79°19'–79°30' W). The region is situated in the mixed-wood zone of the boreal shield. The climate is continental, with mean annual temperature of 0.6°C. Annual precipitation is 823 mm, of which 639 mm falls as rain from April to November. The average frost-free period is 64 d (Environment Canada, 1993). Soils are Gray Luvisols (Cryalfs) (Agriculture Canada Expert Committee on Soil Survey, 1987) originating from glaciolacustrine clay deposits left by proglacial Lake Ojibway (Vincent and Hardy, 1977). Soil texture is that of heavy clay (>75% clay), and the forest floor is a thin mor of 27 cm.

Experimental Design and Treatments
The study was conducted in aspen stands of fire origin dating from 1923 (Dansereau and Bergeron, 1993) and located on fresh clayey upland sites. Aspen represented 93% of stand basal area (see Brais et al. [2004] for a complete description of harvested stands). In the winter of 1998–1999, two levels of forest harvesting (no-harvest and SOH) were applied according to a complete-block design with three replications of each treatment (Fig. 1 ). Control stands and stem harvesting experimental plots were 1 and 2.5 ha, respectively (Fig. 1). In the spring of 1999, nine additional 50 x 50 m plots were located within stem-only harvested plots for slash treatments according to a complete–block, randomized design with three replications.


Figure 1
View larger version (27K):
[in this window]
[in a new window]
  		Fig. 1. Experimental design.

 
The slash treatments consisted of (i) SOH treatment with no additional slash manipulation, (ii) a WTH treatment where all debris between 3 and 7 cm in diameter was removed, (iii) a chipping harvesting treatment (CHT) where all debris between 3 and 7 cm in diameter was removed, and (iv) a slash-burn treatment (SB). The slash burn was conducted over the 50 x 80 m plots at the end of August 1999 by Quebec's Société de protection des forêts contre le feu. At the time of burning, air temperature was 26°C, relative air moisture was 37–42%, wind speed was 9–10 km h–1, and forest floor moisture content (w/w) varied between 166 and 220%. According to the Canadian forest fire weather index system, the fine fuel moisture code was 89, and the buildup index was 24. The rate of spread was 0.5 m m–1, and flame length was between 0.5 and 0.7 m. Similar conditions were observed for the three burns.

Field Methods
Five permanent circular (r = 11.28 m) sampling plots were located in all experimental units. Most nondestructive sampling (soil temperature and moisture, decomposition bags, and mineralization rates) was conducted within each permanent plot of each experimental unit. Destructive sampling (soil and vegetation biomass) was conducted close to but outside of each of the five permanent sampling plots. Slash inventory was conducted over the whole experimental unit.

Slash volume and mass (load) was estimated by the triangular-transect method (McRae et al., 1979) in 1999 after harvesting and slash treatments. One 30-m side triangle was sampled in each experimental plot. Along each transect, the frequency of wood pieces was recorded by species, diameter class (0–0.49, 0.5–0.99, 1–2.99, 3–4.99, 5–6.99, >7 cm) and five decomposition classes (Daniels et al., 1997). Samples of slash were brought back to the laboratory for estimates of density for each combination of species/decomposition and diameter class. In SB treatment, sampling for slash or fuel load (McRae et al., 1979) was performed before and after fire and along two triangles (30-m sides). Total slash loads are the summation of all slash included in decomposition classes 1 to 3 and summed up over all species.

Soil Moisture and Temperature
Soil was sampled for moisture content once every 2 wk from the end of May to the end of September. Bulk samples were taken from the forest floor and the 0- to 10-cm mineral soil. Soil temperatures were measured at the same time and for the same soil layers. Air temperature 1 m above the ground was measured.

Decomposition Rates
To estimate wood decomposition rates, five litter bags containing aspen wood blocks of 10 by 5 by 4 cm, and, in the chipping treatment only, five litter bags containing chips from the chipping treatment were set on the ground (Trofymow, 1998) in the summer of 1999 in each of the permanent sampling plots. A bag of each litter type was collected 1 yr later in each sampling plot for a total of five in each experimental unit. To estimate loss of organic matter from slash, wood blocks and chip dry weight losses were applied to slash and chip loads.

Understorey Vegetation Biomass Increment
Annual increment of understorey vegetation biomass (above ground) was estimated from two 0.5-m2 quadrats located outside each sampling plot (10 per experimental unit). All vegetation within each quadrat was clipped at the base and grouped into the following categories: herbaceous plants, shrub annual growth (stem and branch yearly increments and leaves), and aspen annual growth (stem and branch yearly increments and leaves from aspen suckers <2 m tall). Different quadrats were sampled every year. Each category was weighed in the field, and a sample was kept for oven-dry weight conversion.

Soil Nutrient Concentration and Content
In the fall of 1999 and 2000 and close to each sampling plot, two forest floor samples were collected using a 25 by 25 cm template for ash-free dry weight estimation of forest floor and nutrient concentration. The forest floor samples did not include slashes or wood chips associated with slash treatments. In the chipping treatment, wood chips were collected in the same micro-quadrat and weighed separately from the rest of the forest floor. Three bulk soil samples were taken from the 0- to 10-cm mineral soil and pooled, leading to one sample per plot for mineral soil nutrient concentrations. Two undisturbed soil cores were taken from the 0- to 10-cm mineral soil for bulk density estimates.

Net Nitrogen Mineralization Rates and Microbial Parameters
Nitrogen mineralization rates were estimated by the closed top cores in situ incubation technique (Raison et al., 1987). Two cores were positioned in pairs in each sampling plot. Soils were incubated from mid-July to late October in 1999 (15 wk), from late October 1999 to late May 2000 (32 wk), and from late May 2000 to late October 2000 (23 wk). ABS cores, 30 cm long and 4.5 cm in diameter, were driven through the forest floor to a depth of 10 cm in the mineral soil. Tubes were covered with plastic to prevent rain from entering. At the beginning of each incubation period, forest floor and 0- to 10-cm mineral samples were collected beside each pair of cores for determination of initial NO3– and NH4+ concentrations and, in the fall, for additional laboratory measurements of soil microbial biomass and respiration. At the end of each period, the soil was removed from the core and forest floor material was separated from mineral soil. Samples from core pairs were pooled in the field.

Laboratory Analysis
Forest floor samples were ground (1.7 mm), and mineral soil samples were sieved (2 mm). Samples were analyzed for Kjeldahl N, including NO2– and NO3– (Bremner and Mulvaney, 1982) and for exchangeable cations (Cae, Mge, Ke, Nae) (inductively coupled plasma atomic emission) (PerkinElmer Plasma 40) and acidity (titration; Metler DL-40) after extraction with NH4Cl–BaCl2 (Amacher et al., 1990). Effective cation exchange capacity was estimated by summing base cations and exchangeable acidity. This method allows determination of CEC at field pH and was found preferable for use on acid forest soils (Pritchett and Fisher, 1987). Extractable phosphorus BRAY II (Pext) was estimated using a spectrophotometer (McKeague, 1976). Soil pH was determined in 0.01 M CaCl2 (Hendershot et al., 1993). Organic matter (organic C = 0.58 x OM) of forest floor samples was determined by loss of ignition and organic carbon of ground mineral samples (250 µm) by wet oxidation (Yeomans and Bremner, 1988). Forest floor samples were wet-digested (Parkinson and Allen, 1975), and total phosphorus and cation concentrations (Cat, Mgt, and Kt) were measured by inductively coupled plasma atomic emission. Samples from incubation cores were dried and extracted with 2M KCl. Mineral N concentrations in KCl extractions were determined using an autoanalyzer (Kalra and Maynard, 1991).

Microbial biomass was determined on fresh samples collected in the fall by the fumigation-extraction method (Voroney et al., 1993). Microbial C was determined on K2SO4 extracts acidified to pH 2 with a Shimadzu carbon analyzer. Microbial N was determined according to Cabrera and Beare (1993). Correction factors were kEC = 2.86 (Sparling et al., 1990) and kEN = 1.85 (Brookes et al., 1985; Joergensen and Mueller, 1996).

Soil basal respiration was estimated in the laboratory on fresh samples. Ten grams of forest floor and 50 g of mineral soil were incubated in closed jars at 21°C for 24 h. Respiration was estimated from the amount of CO2 absorbed in NaOH and determined by HCl titration (Skoog and West, 1965; Alef and Nannipieri, 1995). Jars containing NaOH but no soils were used as control.

Downed wood density was estimated from two sections from each of the 330 samples. The first section was weighed (fresh) and coated in hot paraffin before its volume was measured by water displacement. The second section was weighed (fresh) and oven dried for moisture content estimation. The density was estimated from the first section after its fresh weight was corrected for moisture content.

Statistical Analyses
Data analyses were done using GLM, MEANS and CORR, and PRINCOMP procedures of the SAS statistical package (SAS Institute Inc., 1988). Homogeneity of variance between treatments was tested using Bartlett's procedure (Steel and Torrie, 1980). Data that did not meet the requirements were log (Pext, mosses, herbaceous, and shrub stem biomass) or arc sinus (base saturation) transformed. In a parallel study, Brais et al. (2004) compared control and clear-cut stands with partially harvested stands. No direct comparison was made between control and clear-cut stands. To avoid duplication, only significant results of simple ANOVA between control and clear-cut stands are presented in the next section. To assess the effects of slash treatments, a second ANOVA was conducted according to a completely randomized design with four treatments and three replications per treatment. Orthogonal contrasts between treatments were used to compare slash treatments: (i) SB was compared with the three other treatments, and (ii) chipping treatment with WTH and SOH treatments and (iii) SOH and WTH were compared together. To summarize relationships between changes induced by treatments and soil characteristics, a principal component analysis (PCA) was computed from the correlation matrix between forest floor chemical and microbial characteristics (dependent variables). Correlations between principal component scores and independent variables (slash loads and decomposition, understorey biomass, soil humidity, and temperature) were assessed.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
As presented in Brais et al. (2004), the first growing season after harvesting (1999) was relatively hot and dry compared with the second one. In 1999, mean forest floor and mineral soil temperatures remained over 10°C from the end of May to September, and maximum soil temperatures were reached in June. In 2000, soil temperatures over 10°C were reached only in June for the forest floor and in July for the mineral layer. Maximum temperatures, observed in July, were 16.3 and 15°C for the forest floor and mineral soil, respectively. Soil moisture was higher in 2000 than in 1999.

Effects of Harvesting
In response to harvesting, we observed a significant increase (p = 0.04) in annual above-ground understorey vegetation biomass increments, which were 3210 ± 294 (mean ± SE) and 2520 ± 113 kg ha–1 in harvested stands in 1999 and 2000, respectively, and 450 and 554 kg ha–1 in control stands for the same years. Harvesting increased slash loads. Mean slash dry weights immediately after harvesting were 52.3 ± 4.1 and 13.8 Mg ha–1 for stem-only harvest and uncut control stands, respectively (p = 0.05). Wood block dry weight losses were 32% in harvested stands and 26% in control stands during the first growing season after harvesting. Applying wood block weight losses to initial slash loads led to estimated losses of slash dry matter of 16.2 ± 1.7 and 3.7 Mg ha–1 for stem-only harvested plots and control plots, respectively (p = 0.06).

During the first year after harvesting, a decrease in forest floor Cmic/Nmic from 8.06 ± 0.14 to 6.88 (p = 0.05) was observed in harvested stands. Net in situ N immobilization was observed in the forest floor of both treatments during the first growing season (control stands = –7.7 ± 30.1; harvested stands = –74.0 mg N kg–1 soil wk–1) and in the 0- to 10-cm mineral soil (average = –7.11 ± 0.9 mg N kg–1 soil wk–1). Net immobilization was observed again during the winter season (results not shown) and in the forest floor during the second growing season with somewhat higher values in the harvest treatment (control stands = –27 ± 0.3; harvested stands = –7.2 mg N kg–1 soil wk–1). Low rates of net mineralization were measured in the 0- to 10-cm mineral soil. Little nitrification was observed in any of the treatments during the study period.

In the first year after harvesting, a small but significant decrease in KT (from 1.21 ± 0.01 to 1.11 g kg–1, p = 0.05) and a large increase in extractable P (from 60.5 ± 0.1 to 80.5 mg kg–1, p < 0.01) were observed in harvested stands. During the second growing season, harvesting increased forest floor organic C, Kjeldahl N and base cation concentrations (Cae and Mge), base saturation, pH, and effective CEC (Table 1). Two years after harvesting, no significant changes in mineral soil characteristics were observed.


View this table:
[in this window]
[in a new window]
  		Table 1. Significant changes in forest floor nutrient characteristics 2 yr after stem-only harvesting of aspen stands on clayey soils.

 
Effects of Slash Treatments
Overall, burns affected 70% of SB area. Compared with pre-burn values, burns decreased forest floor dry weight (ash free) by 6.74 ± 1.24 Mg ha–1 (P > T = 0.161) or 15% and fine debris (0–0.99 cm diameter) loads by 12% but had no effect on medium and coarse slash loads (Belleau, 2002). Overall, compared with pre-burn values, total slash loads were reduced by 4 Mg ha–1. Nonetheless, slash burn had on average higher medium (3.0–6.99 cm diameter) slash loads than the three other treatments (Table 2). Decreases in medium slash loads after chipping were compensated for by chips returned to the sites (Table 2). Total slash load after WTH was 34% lower than under SOH.


View this table:
[in this window]
[in a new window]
  		Table 2. Forest floor and fresh slash loads after slash treatments in harvested aspen stands. A priori comparisons by means of orthogonal contrasts (p > F).

 
No significant differences in soil temperature and moisture content were observed among slash treatments, and no significant differences in decomposition rates (wood block dry weight losses) were observed the first year after implementation of the treatments (Table 2). No significant differences in slash dry weight losses were observed among any of the slash treatments.

In the second year after treatments, understorey vegetation biomass increment amounted to 3230, 2520, 2459, and 2425 kg ha–1 under WTH, SOH, CHT, and SB, respectively. No significant differences between treatments were observed. Slash burn biomass increment was delayed in 2000, the first growing season after fire, with decreases of 73 ± 86.4% (p = 0.07) and 92 ± 0.25% (p < 0.01) respectively, in shrub and moss biomass relative to other treatments. However, herbaceous biomass was 30% higher in SB than in the other treatments.

Slash burn induced immediate and significant decreases in forest floor organic concentrations of C (–14%, Fig. 2 ) and microbial C and N (–48 and –55%, respectively) (Fig. 3 ) and in forest floor basal respiration (–37%), whereas the microbial biomass respiratory quotient increased by 20% (Fig. 3). Differences in microbial C and N concentrations and basal respiration between SB and other slash treatments remained significant throughout the next growing season, and SB also resulted in positive net N mineralization in the forest floor (Table 3). Immediately after SB, forest floor–extractable P, total P, and total Mg concentrations increased by 54, 27, and 40%, respectively, and forest floor pH increased by 0.6 units (Fig. 2). Extractable P forest floor concentrations remained higher throughout the year after burning. Slash burn had no significant effects on extractable base cation concentrations (Fig. 2).


Figure 2
View larger version (55K):
[in this window]
[in a new window]
  		Fig. 2. Slash treatment effects on forest floor characteristics. SB, slash burn; CHT, chipping; SE, population standard error; SOH, stem-only; WHT; whole-tree harvesting. Lines = orthogonal contrast between treatments. See Methods section for contrast descriptions. Contrast significance levels: *p = 0.1; **p = 0.05; ***p = 0.001 (n = 12).

 

Figure 3
View larger version (45K):
[in this window]
[in a new window]
  		Fig. 3. Slash treatment effects on forest floor microbial community. SB, slash burn; SE, population standard error; CHT, chipping; SOH, stem-only; WHT, whole-tree harvesting. Lines = orthogonal contrast between treatments. See Methods section for contrast descriptions. Contrast significance levels: *p = 0.1; **p = 0.05; ***p = 0.001 (n = 12).

 

View this table:
[in this window]
[in a new window]
  		Table 3. Slash treatment effects on in situ net N mineralization rates 1 and 2 yr after harvesting of aspen stands. A priori comparisons by means of orthogonal contrasts (p > F).{dagger}

 
Chipping induced temporary changes in forest floor microbial N and C/N ratio (Fig. 3). At the end of the first growing season, forest floor microbial N concentration was 25% lower and microbial C/N 28% higher than average values observed under SOH and WTH. A small decrease in organic C was also observed at the end of the second growing season (Fig. 2). Chipping had no effect on forest floor N net mineralization rates (Table 3) or on extractable base cation and P concentrations (Fig. 2).

Besides a slight but significant higher basal respiration rate under WTH, no differences in forest floor microbial community characteristics (Fig. 3) and nutrient concentrations (Fig. 2) were observed between SOH and WTH during the first growing season (Fig. 2). In the following year, forest floor microbial C/N increased while extractable base cation concentrations decreased under WTH, although the reduction was only significant for exchangeable Mg (Fig. 2).

Slash treatments had little immediate effect on nutrient dynamics in the 0- to 10-cm mineral soil. Nonetheless, higher exchangeable acidity values were observed for both growing seasons (p = 0.06; p = 0.02) under WTH (1.65 and 1.78 cmol[+] kg–1 for 1999 and 2000) than under SOH (0.76 and 0.72 cmol[+] kg–1 for 1999 and 2000).

Overall Effects of Slash Treatments on Forest Floor Nutrient Status and Microbial Community
The first component of PCA (Component 1; Fig. 4 and Table 4) explained 32% of forest floor characteristic variance and was correlated to forest floor pH; effective CEC; and concentrations of organic C, Kjeldahl N, exchangeable Ca, and exchangeable Mg. The second component explained 29% of forest floor variance and was positively correlated to microbial respiratory coefficient, extractable P, net N mineralization rate, and pH and was negatively correlated to forest floor–exchangeable K, microbial N, and microbial C concentrations. Ordination of experimental units on the two first principal components (Fig. 4) showed a clear differentiation between control unharvested plots (negative values on Component 1, lower nutrient and C concentrations and pH) and stem-only harvested units (positive values on Component 1, higher nutrient concentrations and pH). Slash burn and chipping experimental plots were found at the center or to the left of Component 1. Component 2 differentiated between SB (low microbial N and C and high respiratory coefficient, high extractable P, net N mineralization) and all of the other treatments. The first component was correlated to slash loads, slash dry weight loss, and forest floor average water content. The second component was not correlated to any of the independent variables but was clearly linked to slash combustion. No significant correlations were found between soil temperature or biomass increment and any of the components.


Figure 4
View larger version (19K):
[in this window]
[in a new window]
  		Fig. 4. Principal component analysis of forest floor properties after harvesting and four slash treatments. Gray ellipses show the influence zone of segregate treatment types. SB, slash burn; CHT, chipping; SOH, stem-only; WHT, whole-tree harvesting.

 

View this table:
[in this window]
[in a new window]
  		Table 4. Correlation between forest floor characteristics and principal components (see Fig. 4) derived from these characteristics and between principal components and independent variables.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Effects of Harvesting
Short-term changes in forest floor nutrient dynamics after harvesting can be attributed to a number of factors, including increases in decomposition and mineralization rates, decreases in plant and microbial uptake, or nutrient leaching from woody debris (Abbott and Crossley, 1982; Burger and Pritchett, 1984; Keenan and Kimmins, 1993; Prescott, 1997). Despite increases in forest floor temperature and water content (Brais et al., 2004), decomposition rates were not affected by harvesting, and, contrary to what would be expected after disturbance (Mann et al., 1988), net N immobilization was observed in harvested stands. Net nitrogen immobilization was consistently higher after harvesting, although not significantly so. Nitrogen leaching from fresh logging slash and subsequent forest floor immobilization could have induced the observed increase in forest floor total organic (Kjeldahl) N. The latter process is plausible because aspen branches release N during the first years of decomposition (Miller, 1983). Such increases have also been reported by Adams and Boyle (1982), Hornbeck and Kropelin (1982), Silkworth and Grigal (1982), and Titus et al. (1997). Leaching of water-soluble organic C from fresh slash, an easily accessible source of energy for soil microorganisms (Huang and Schoenau, 1996), coupled with a reduction in tree root exudation (Bååth, 1980) and root capacity to support ectomycorrhizae (Harvey et al., 1980) and changes in forest floor acidity, could have induced the observed changes in the microbial C/N ratio. The decrease in microbial C/N ratio that we observed after harvesting is believed to reflect a change toward a community where the ratio of bacteria to fungi is higher (Marumoto et al., 1982). Hassett and Zak (2005) observed similar changes in community structure 8 yr after harvesting of aspen stands in Michigan. Changes in community structure did not affect the community's ability to function because no changes in basal respiration or respiratory coefficient were observed after harvesting.

Increases in forest floor Mg and Ca concentrations observed after harvesting were also linked to the amount of slash left on the ground (Fig. 4). No mechanism can be provided for the short-term increase in extractable phosphorus. Higher forest floor effective CEC and understorey vegetation biomass increment after harvesting contributed to nutrient immobilization, and, with the absence of forest floor disturbance, may account for the lack of changes in mineral soil properties. On the other hand, slash decomposition in harvested stands is of short duration, whereas a continuous input of organic matter from coarse woody debris is expected to take place in unharvested stands. Long-term reductions in soil C and N content were observed after harvesting of stands of similar composition growing on fine-textured soils in western Canada (Pennock and van Kessel, 1997).

Slash Treatment Effects
Slash burn severity was low, and burns did not significantly reduce the amount of slash when compared with amounts left by other slash treatments. Nonetheless, combustion induced changes in soil microbial community as shown by the distinct position of slash burns on the second PCA axis. Forest floor microbial biomass was reduced by combustion and presented a higher respiratory coefficient. Similar decreases in microbial biomass after burning have been reported for boreal forests (Pietikäinen and Fritze, 1993). Short-term changes in the forest floor microbial community resulted in a reduced net N immobilization rate after fire, although no significant differences have been observed between slash burns and other slash treatments in forest floor Kjeldahl N concentrations. Increases in Ca and Mg availability were observed when forest floor pre-burn values were compared with post-burn values (Belleau, 2002), but post-burn values were within the range of those observed under other slash treatments. Sudden increases in phosphorus after fire have been reported by Romanyà et al. (1994), Cade-Menun et al. (2000), Giardina and Rhoades (2001), and Simard et al. (2001). High P concentrations after fire may be linked to high concentrations of P in fine ashes that are rapidly returned to the soil surface (Carter and Foster, 2004). Response of herbaceous species after fire, notably pyrophilous Geranium bicknellii (Britton), was swift, as indicated by the high herbaceous biomasses observed in slash burns. Higher P availability and reduced microbial competition for N may have contributed to herbaceous vegetation response.

The effects of slash chipping or of residual understorey vegetation on soil have been studied by Brais et al. (1996), Corns and Maynard (1998), Scherer et al. (2000), and Zabowski et al. (2000). Differences in techniques and amount of chips spread on the ground make comparisons difficult. In this study, slash fragmentation did not result in increased decomposition rates and nutrient release. However, a higher Cmic/Nmic ratio under the chipping treatment than under WHO and SOH suggested changes in soil microbial community composition shifting toward a higher proportion of fungi. Fungi colonizing chip surfaces could have acted as a sink for soluble organic C (Harmon et al., 1986), explaining the lower forest floor C concentration observed under this treatment. Chip C and N concentrations were not measured in this study, but Corns and Maynard (1998) have observed N immobilization within aspen chips in a similar study.

Slash loads under WTH were intermediary between that of SOH and control stands, and differences between WTH and SOH in forest floor microbial community structure and basic cation concentrations were linked to the abundance of slash loads and forest floor water content (Fig. 4), with higher forest floor concentrations under higher loads. Higher forest floor organic C and base cation concentrations are often reported for SOH when compared with WTH (Bélanger et al., 2003), whereas WTH harvesting may show little difference in nutrient dynamics when compared with unharvested stands (Hendrickson et al., 1989). Results from Bélanger et al. (2003) were obtained in black spruce stands, indicating that rapid nutrient leaching from slash is not limited to faster decomposing species. Increases in vegetation biomass and plant nutrient uptake under WTH could account for differences in nutrient availability between WTH and SOH (Stevens et al., 1995; Titus et al., 1997), but this was not the case here.

The rapid increase in mineral soil exchangeable acidity under WTH could not be linked solely to slash decomposition dynamics or nutrient uptake. Mineral soil changes in acidity after WTH may reflect complex interactions between slash loads and decomposition, vegetation uptake, and disruption of root networks induced by harvesting.

In our experimental design, the level of replication was low, which is inherent to large experimental designs that are made to study the effect of forest practices on ecosystem processes (e.g., Emend: Volney et al., 1999; North American Long-Term Soil Productivity: Powers, 1999; Hubbard Brook Experimental Forest: Likens and Bormann, 1995). The low level of replication generated a low power and a high probability for type II error. Therefore, the effects that were significant in the present study were strong, and we could have missed some moderate effects of treatment. On the other hand, because the study was conducted immediately after treatment, it is likely that weaker effects vanished with increasing time since disturbance. A longer-term follow-up is required to determine the effects that are persistent throughout a forest rotation cycle.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
Slash burns had distinct effects on nutrient dynamics that none of the other treatments could mimic. We do not know if these effects are critical for long-term soil fertility. Slash burn application in aspen stands on rich sites may be limited by slash humidity immediately after harvesting and vegetation growth early in the growing season. Burning also killed most aspen suckers, and its applicability to naturally regenerating aspen stands is limited. Independently from combustion, the amount of slash left on the ground was the main factor driving forest floor base nutrient dynamics in the years immediately after harvesting. Slash reduced forest floor acidity and increased basic cation availability. Although SOH may leave large amounts of slash on these upland mesic sites, slash did not reduce soil temperature and limit aspen growth. Rapid understorey biomass response and effective CEC limited nutrient leaching after SOH. Although it is not clear if base cation availability is a limiting factor in the early phase of aspen stand establishment, in the longer term, SOH could alleviate nutrient losses induced by harvesting (Paré et al., 2002). Chipping did not improve nutrient availability or reduce vegetation competition, and its application did not result in any tangible immediate advantage. On the other hand, with SOH, slash is often piled in skidding trails, and chipping might improve its distribution in cutovers.


   ACKNOWLEDGMENTS
 
This work was supported by the Division de la recherche forestière of the ministère des Ressources naturelles et de la Faune du Québec (Project No. 0905-4510), by the CFI (Project No. 1818), by the NSERC Partnership program with the CFS (Project No. CFS 215528-98), by the Sustainable Forest Management Network and by the NSERC-UQAT-UQÀM Industrial Chair in Sustainable Forest Management. We are grateful to our technical staff and to NEXFOR and Tembec for their support and to Thuy Nguyen, Marie-Hélène Longpré, and Sonia Légaré for discussions and statistical and technical advice.

Received for publication June 15, 2005.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 




This article has been cited by other articles:


Home page
 Soil Sci.Home page
H. S. Jones, L. G. Garrett, P. N. Beets, M. O. Kimberley, and G. R. Oliver
Impacts of Harvest Residue Management on Soil Carbon Stocks in a Plantation Forest
Soil Sci. Soc. Am. J., September 30, 2008; 72(6): 1621 - 1627.
[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 (5)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Belleau, A.
Right arrow Articles by Paré, D.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Belleau, A.
Right arrow Articles by Paré, D.
Agricola
Right arrow Articles by Belleau, A.
Right arrow Articles by Paré, D.
Related Collections
Right arrow Best Management Practices
Right arrow Nutrient Cycling
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