HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (10) Citing Articles via Google Scholar Google Scholar Articles by Prévost, M. Search for Related Content PubMed Articles by Prévost, M. Agricola Articles by Prévost, M. Related Collections Soil Organic Matter Other Soil Management Forest Soils

DIVISION S-7—FOREST & RANGE SOILS

Predicting Soil Properties from Organic Matter Content following Mechanical Site Preparation of Forest Soils

Ministère des Ressources Naturelles, de la Faune et des Parcs, Forêt Québec, Direction de la Recherche Forestière, 2700, rue Einstein, Sainte-Foy, QC, Canada G1P 3W8

^* Corresponding author (marcel.prevost{at}mrnfp.gouv.qc.ca).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
The difficulties of sampling forest soils and their high spatial variability make estimation of soil physical properties following forest operations laborious. To develop prediction tools, soils were sampled from two sites located in the boreal forest of northern Québec, Canada. Soil organic matter (OM) content was found to be closely related to bulk density (D_b) and porosity after clearcutting and mechanical site preparation (MSP) on these sites. Reasonably good estimates of D_b, with an average error of 18 to 20%, can be made from the easily measurable OM concentration and the logarithmic relationships (R² = 0.731 and 0.847, respectively for the Alma and Chibougamau sites) developed in this study. The organic density approach, recently developed for forest soils in New England, was found to be less precise (R² = 0.637) than the logarithmic relationships following soil disturbance. For the two sandy till soils in northern Québec, the equation based on this concept best fit the data with a pure OM bulk density (D_bo) of 0.159 Mg m^–3 and a pure mineral matter bulk density (D_bm) of 1.561 Mg m^–3. The equations presented in this study also explain between 60 and 70% of the variation in porosity and C/N ratio from OM concentration, with prediction errors of 13 and 24%, respectively. In spite of soil surface disturbance associated with MSP, the easily measurable OM concentration can be used to predict D_b, porosity, and C/N ratio.

Abbreviations: D_b, bulk density • D_bo, pure organic matter bulk density • D_bm, pure mineral matter bulk density • EU, experimental unit • MBIA, mean bias • MSP, mechanical site preparation • OM, organic matter • RMSEP, Root mean square error of prediction

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
ORGANIC MATTER CONTENT and D_b are related soil properties that are important to characterize soil disturbance following harvesting and MSP. The pool of several nutrients is closely related to the OM concentration (Armson, 1977). Bulk density is related to other soil physical properties such as texture, structure, air-filled porosity, and water conductivity (Froehlich and McNabb, 1984). Although soil strength (resistance measurements) is presently the simplest and most practicable means of assessing soil physical properties associated with productivity (Powers et al., 1998), D_b is often used as an index of soil compaction to quantify harvesting impact on the stony soils of the boreal forest (Corns, 1988; McNabb et al., 2001; Block et al., 2002). In nutrient studies, estimates of D_b are also necessary to establish the nutrient pool of a soil layer (grams per square meter of land area, megagrams per hectare) from standard laboratory analyses of concentrations (mg kg^–1, %).

Theoretically, D_b is simple to estimate (oven-dry weight/total volume), but in practice, collection of volumetric samples from forest soils can be very difficult due to rocks, roots, and woody debris (Terry et al., 1981). Furthermore, spatial variability in forest soils is high (Grigal et al., 1991; Järvinen et al., 1993) and the sampling intensity that would be necessary for an adequate estimation frequently exceeds available resources. However, D_b and most of the properties affecting root development are closely related to OM concentration, which is determined by loss on ignition of easily collected soil samples. Curtis and Post (1964) developed an empirical relationship of log D_b on log OM concentration for till forest soils of Vermont. Federer (1983) and Huntington et al. (1989) obtained similar relationships for forest soils of New Hampshire, which supported the use of OM concentration measurement to estimate D_b throughout New England. More recently, Federer et al. (1993) developed a new theoretical expression based on the organic density concept to relate D_b and OM concentration. The expression supposes that (i) D_bo and D_bm remain constant in any mixture and (ii) the volumes occupied by the mineral and the organic fractions are additive. The relationship was tested satisfactorily for eight locations in New England, including three sites that had been recently affected by logging disturbance.

Clearcutting and MSP followed by tree planting are widely applied in the boreal forest. In the Canadian boreal forest, the effect of MSP on seedling field performance (Sutton, 1987; Macdonald et al., 1998) and soil properties (Prévost, 1996; Schmidt et al., 1996) received some attention. Although it is recognized that D_b is closely related to OM concentration for a given soil, no relationships are available for organic matter-rich till soils in the boreal forest. In this paper, the logarithmic approach and the organic density concept of estimating D_b from OM concentration are validated following clearcutting and MSP on two sites of the Canadian boreal forest. Simple relationships between OM concentration and D_b, total porosity, and C/N ratio are also developed. A major soil surface disturbance as produced by MSP may greatly affect the physicochemical properties of the soil and change their interrelations. Thus, a further goal of this study was to assess the possibility of developing relationships that would be reliable for both harvest-only and MSP treated areas.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Experimental Sites
Data were collected from two experimental sites, where modified clearcut systems (strip or group seed-tree cuttings) were combined with MSP to promote the natural establishment of black spruce (Picea mariana [Mill.] BSP) seedlings. Both sites were in the black spruce-mosses domain (Saucier et al., 1998) included in the B.1b Chibougamau–Natashquan region of the boreal forest (Rowe, 1972); one site is located 120 km north of Alma (49° 48' N lat., 71° 27' W long.) and the other is 80 km southwest of Chibougamau (49° 23' N lat., 74° 52' W long.), Québec, Canada. The regional climate is classified as subpolar subhumid, continental (Robitaille and Saucier, 1998), with 90 frost-free days (Canadian Climate Program, 1982) and an average annual precipitation of 930 mm (one third as snow). Mean daily maximum temperatures are recorded in July (22.1°C) and minimum temperatures in January (–24.0°C; Canadian Climate Program, 1993). Soils of the region are developed from deep glacial till and, in both sites, the soil is classified as podzol (Canada Soil Survey Committee, 1992) or Spodosol (Soil Survey Staff, 1998). Both sites were black spruce stands (>90% of basal area) with an ericaceous understorey (Kalmia angustifolia [L.] and Ledum groenlandicum [Retzius]) and a ground cover of feathermoss (Pleurozium schreberi [BSG.} Mitt., Hylocomium splendens [Hedw.] BSG., Polytrichum sp.). The Alma site was described in detail by Prévost (1996) while the Chibougamau site was presented in Prévost (1997).

At Alma, the soil was a loamy sand covered with a Fibrimor layer that was 23 cm in average thickness. The soil moisture regime was classified as moderately to well-drained on 80% of the area, and as imperfectly drained in depressions (according to Saucier et al., 1994). The experimental design for this study was a complete randomized block design with four replications, each containing five soil treatments (50 x 100 m experimental units, EUs): an harvest-only control (no MSP) and four treatments combining two types of MSP (TTS-35H disc trencher [TTS Forest OY, Rajamäki, Finland] and Wadell cone trencher [Storebro International AB, Storebro, Sweden]) and two treatment intensities (single-pass and double-pass). The disc trencher consisted of two free-rotating toothed-discs and the cone trencher consisted of two toothed-cones with a contra-rotation to the direction of the travel. Both trenchers were hydraulically activated to apply downward pressure. These machines cut furrows in the soil and expose mineral soil in a trench bordered by a loosened berm that is elevated above the initial ground surface (Sutton, 1993). The scarification treatments were applied along stream buffer strips, to take advantage of seeding that would come from the strips which is similar to that occurring in strip clearcuts. Each EU was subdivided into two 25 x 100 m subunits, one being allowed to regenerate naturally and the other being planted with 2 + 0 black spruce seedlings.

At Chibougamau, the soil was an imperfectly to well-drained silty sand with a slightly undulating topography. The organic layer was a peaty Fibrimor 20 cm in depth. The MSP treatments were applied around circular groups of seed-trees (30, 40, 50, 60, and 70 m in diameter), which were left uncut. The experimental design consisted of 12 complete randomized blocks, each containing three soil treatments: a harvest-only control (no MSP), a single-pass, and a double-pass passive disc trenching (TTS-35). Each treatment was applied on a truncated pie slice shaped area that had a 40° angle and that extended 100 m from each seed-tree group.

Soil Properties
To evaluate soil properties before MSP, two soil sampling plots were established in each EU. In each plot, two types of samples were collected at three depths, 0 to 10, 10 to 20, and 20 to 30 cm, under the litter layer (L and F horizons) thus including the decomposed humus layer (H horizon, 0–1 cm). A 130 cm³ undisturbed soil core (5.25 cm diameter, 6.0 cm long) was first collected at each depth with a double-cylinder, hammer-driven core sampler. Bulk samples were collected afterward from the side of the created hole and transported into plastic bags for chemical analyses. Both types of sampling were repeated at each of the two sites at different time intervals following scarification (2, 14, 26, and 38 mo in Alma and 12, 36, and 60 mo in Chibougamau). In scarified EUs, samples were collected at elevated position on the trench profile, while avoiding the bottom of the furrow and undisturbed areas to estimate soil properties in the best planting microsite (Örlander et al., 1990).

Undisturbed core samples were first water saturated from the bottom and weighed. Samples were oven-dried at 105°C for 24 h and weighed again to determine dry weight. Total pore space was estimated by the difference between the saturated weight and the dry weight, removing the density of water as 1.0 g cm^–3. The dry weight and total pore space were expressed on a volume basis to determine D_b and total porosity (Armson, 1977). Values of D_b were not corrected for stone content, that is, for particles >2 mm, but cores found to contain stones >10 mm were discarded. Bulk samples were air dried and sieved (2 mm). The Bouyoucos method was used to determine soil textural classes. The OM concentration was obtained by loss on ignition (Nelson and Sommers, 1982) and total C was estimated by dividing this concentration by a factor of 1.724, assuming that inorganic C is insignificant in these acidic soils (pH around 4). The Kjeldahl method (after grinding to 0.5 mm) was used to mineralize the organic matter (Thomas et al., 1967) to determine organic N + NH₄–N by colorimetry with a Lachat Flow Injection Analyzer (model 8000, Lachat Instruments, Milwaukee, WI), while NO₃–N was also estimated by colorimetry on 2 M KCl extracts (Kalra and Maynard, 1991). The pH was determined on a 0.01 M CaCl₂ solution, while extractable P was measured by the Bray2 method. Exchangeable Ca was measured by plasma-atomic emission spectrometry on 1 M CH₃COONH₄ extracts (McKeague, 1978).

Data Analysis
All statistical analyses were performed with the SAS software (SAS Institute, 1996). The CORR procedure was used to obtain Pearson correlation coefficients between OM concentration and other parameters. For each site, data of all treatments and all sampling dates were pooled for the regression analyses. The STEPWISE procedure, with the stepwise selection method applying a same significance level = 0.05, for entry and for staying into the model, was used to select among three expressions of the independent variable (OM, ln OM, and [ln OM]²) and obtain the regression coefficients (a, b, c, and d) for the best fit of the model:

[1]

where OM is the organic matter concentration (g OM g^–1 of soil). The response variable Y was first analyzed without transformation and thereafter using a logarithmic transformation (ln Y). The model having the higher R² was retained. The accuracy of equations was evaluated with the statistical terms MBIA (mean bias) and RMSEP (root mean square error of prediction):

[2]

[3]

where, Y_ip is the predicted value, Y_io, the observed value, N, the number of observations and p, the number of adjusted regression coefficients. These statistical terms are recognized as being valuable to estimate model performance from data that were used to obtain regression coefficients. The nonlinear regression (MODEL procedure), with the Marquardt iteration method, was used to estimate parameters of the Federer et al. (1993) equation:

[4]

where D_b is the soil bulk density, D_bm is the bulk density of the mineral fraction, and D_bo is the bulk density of the organic fraction.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Relationships between Organic Matter Concentration and Other Variables
The two soils were of coarse texture with only small differences in particle-size distribution (Table 1). This result reflects the crystalline nature of the bedrock from which the till was derived. The ranges of D_b values were also comparable for the two sites (0.10 to 1.90 and 0.11 to 1.96 Mg m^–3, respectively for Alma and Chibougamau), as were ranges of porosity values (32 to 91 and 33 to 96%). Both sites had a wide range of OM concentration values (0.003 to 0.903 and 0.004 to 0.894 g g^–1), the organic matter content of these forest soils decreasing with depth of sampling (not shown). As a whole, the ranges of C/N ratios were also comparable for the two sites (12 to 81 and 12 to 67), but the average C/N ratio was higher in unscarified EUs at Alma (42) than in other EUs (29). For both sites, pH, extractable P, and exchangeable Ca appeared to be in the normal ranges for till soils of the boreal forest (Schmidt et al., 1996).

View larger version (30K): [in this window] [in a new window]   Fig. 1. Bulk density (Db, Mg m–3), total porosity (%), and C/N ratio related to organic matter (OM) concentration (g g–1), following harvesting (dots) and mechanical site preparation (MSP) (circles), at the Alma and Chibougamau sites. The equations adjusted for the present study are presented in Table 3 and illustrated by a solid line (Curve A). Broken lines illustrate the equations reported in Table 4 and originally given in papers by (B) Huntington et al. (1989), (C) Federer (1983), (D) Federer et al. (1993), and (E) Tremblay et al. (2002).

The equation based on the organic density concept fit quite well to the pooled data of the two sites in the present study (R² = 0.637, Table 4). However, this model explains 13% less of the D_b variance than the logarithmic relationship (R² = 0.767, Table 4). Federer et al. (1993) report that in spite of soil disturbance during harvesting at one site, the D_b–OM concentration relationship was similar to that obtained in a nearby mature spruce-fir forest and fit the assumption of additive mineral and organic volumes. In the present study, the severe soil disturbance produced during MSP represents a source of error in the organic density model. This approach resulted in a higher D_bo value (D_bo = 0.159 Mg m^–3 and D_bm = 1.561 Mg m^–3) for the sandy till soils of northern Québec than for the sandy till loam soils of New Hampshire (D_bo = 0.111 Mg m^–3 and D_bm = 1.450 Mg m^–3). This latter curve, taken from Federer et al. (1993), corresponds well to the logarithmic curves that were developed in the present study (Fig. 1). Similar to Federer et al. (1993), Tremblay et al. (2002) obtained 0.12 Mg m^–3 for D_bo and 1.40 Mg m^–3 for D_bm, from soils supporting sugar maple (Acer saccharum Marsh) stands and some spruce and fir stands in southern Québec. The D_bo in the present study corresponds to the 0.155 Mg m^–3 value obtained by Federer et al. (1993) for the data collected by Curtis and Post (1964) from Vermont till soils. It is possible that the relatively high D_bo values observed in the present study could be in part attributed to the use of a core sampler which may have compressed organic particles more definitely.

Similarly to D_b, porosity is clearly related to OM concentration (Fig. 1), particularly at the Chibougamau site where the R² of the fitted model reaches 0.752, compared with 0.561 at Alma (Table 3). For both sites, ln OM was first included into the model and thereafter (ln OM)² added 2% to the explained variation. The better fit in Chibougamau could be explained by a less severe disturbance as compared with Alma, where the treatment removed 80% of the organic matter content of the microsites without significantly changing the porosity (Prévost, 1996). Furthermore, similarly to D_b, prediction of porosity is subject to the intrinsic heterogeneity of forest soils.

The C/N ratio–OM concentration relationships are more variable than the D_b–OM concentration and porosity–OM concentration relationships (Table 3 and Fig. 1). The C/N ratio is best explained by ln OM_, but contrary to the relationships with D_b and porosity, a second term is introduced into the model only for the Alma site ([ln OM]², partial R² = 0.071). Surprisingly, the C/N ratio is better predicted for each of the two sites taken separately (R² = 0.611 and 0.649) rather than for the pooled data (R² = 0.603). By disturbing surface horizons, MSP produced immediate site-specific changes in C and N concentrations. For example, MSP decreased the C/N ratio by exposing mineral horizons to the surface in Alma (Prévost, 1996), while the treatment little modified this ratio in Chibougamau. Furthermore, subsequent dynamic processes such as OM decomposition and N mineralization may add to data variability by modifying these concentrations in different ways. The accelerated OM decomposition following MSP is likely to decrease the C/N ratio with time, as observed in Alberta by Schmidt et al. (1996). At the Alma site, the buried bag method has indicated that organic matter mineralization was not accelerated during the first year following MSP (Prévost, 1996). Unfortunately, N mineralization was not studied afterward and it is thus impossible to relate the mineralization rate to the observed C/N ratio values.

Adequacy of the Regression Models
In their current practice, both managers and researchers have to characterize forest site condition and thus they need a good estimate of the mean of the principal soil properties. The MBIA statistic shows that the logarithmic equations presented in this study allow to predict D_b, porosity and the C/N ratio from OM concentration practically without bias (Table 5). This is an indication that the accuracy of prediction will increase with the number of soil samples and that the estimation of the mean should be reliable.

The visual analysis of residuals confirmed the absence of a bias in D_b prediction from OM concentration (not shown), although it indicated that the absolute error slightly increases with D_b, particularly for samples above 0.8 Mg m^–3. Such a result was noted by Huntington et al. (1989) for low OM concentrations (<0.08 g g^–1), primarily in the 10- to 20- and 20+-cm depth strata. In the present study, the lack of fit to high D_b values (Fig. 1) can be explained by the high variability of D_b (0.45 to 1.96 Mg m^–3) at low OM concentrations (<0.10 g g^–1). Other soil factors such as texture, structure, and rock content greatly affect D_b, particularly at low OM concentration. However, the relative error of estimate (percentage of observed D_b) is around 20% for the entire OM concentration range, since D_b decreases with increasing OM concentration. These results, together with similarities between the presented relationships and those found in the literature, support the use of OM concentration to obtain reliable D_b estimates following MSP on sandy till soils in the boreal forest.

Between 60 and 70% of porosity variation can be explained using OM concentration and the logarithmic relationships developed in this study (Table 3). For both sites, the RMSEP is around 7% and corresponds to mean errors between 12 and 14% of the observed values (Table 5). The MBIA statistic and residuals (not shown) indicate that porosity estimation is generally without bias. Data from the two sites indicate that the porosity–OM concentration and D_b–OM concentration relationships were clearly defined and changed little during the 5-yr sampling period, suggesting that the soil surface layers rapidly stabilized after treatment. Mechanical site preparation created a variety of seedbeds, from compact mineral soil in the trench to loosened OM-rich seedbeds in adjacent berm (Prévost, 1996). According to Örlander et al. (1990), trenching does not loosen the soil significantly below the surface and the soil loosening that occurs in elevated spots does not last for a long time. In the present study, the rapid soil surface settling could be attributed to the compression by the winter snowpack and to the frequent freeze–thaw cycles occurring in the shallow depths of the cold soils in the boreal forest. The presented equations are thus generally reliable for any post-cut condition including MSP treated areas.

Similar to D_b and porosity, the prediction of the C/N ratio is slightly better for the Chibougamau site than for the Alma site. The RMSEP statistics, around 6 in Chibougamau and 7 in Alma, correspond to mean errors of 20 and 24%, respectively (Table 5). The residuals are independent of the predicted C/N ratio (not shown), soil treatment and site, although MSP modified the C/N ratio at Alma. Following MSP in west-central Alberta, Schmidt et al. (1996) observed that the treatment effect on the C/N ratio was closely related to its effect on the OM content. In the present study, the C/N ratio–OM concentration relationship was not affected by MSP.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Among the studied soil properties, D_b, total porosity, and C/N ratio were the most closely related to OM concentration following clearcutting and MSP at two sites in the Canadian boreal forest. In spite of soil surface disturbance, a close relationship was found between the easily measurable OM concentration and D_b (which is laborious to estimate). The logarithmic relationships presented for the two northern black spruce stands explain nearly 80% of the variation in D_b and are valid on both harvest-only and MSP treated areas. The prediction error of 20% for D_b is acceptable considering the inherent difficulties of measurement and their associated error in forest soils. An approach using organic density was shown to be less precise (R² = 0.637) than logarithmic relationships (R² = 0.767) in this study. Indeed, it appears difficult to totally reconcile the concept of additive mineral and organic densities to the well established effect of soil loosening that is attributed to MSP. Relationships using OM concentration as independent variable and explaining between 60 and 70% of the porosity and C/N ratio variations suggest that they can also be useful for a simple and rapid estimation of these variables.

	ACKNOWLEDGMENTS

 
The author thanks Rock Ouimet, Daniel Kneeshaw, and André Plamondon for revising an earlier version of this manuscript. The helpful suggestions of the associate editor and two anonymous reviewers are also acknowledged for improving the manuscript. Finally, the author acknowledges Gilles Audy, Louis Faucher, and Pascal Gagnon for soil sample collection; Carol Deblois and his team for laboratory analyses; and Louis Blais for statistical advice.

Received for publication April 4, 2002.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 

• Armson, K.A. 1977. The architecture of soil: Texture, structure, and porosity. p. 15–29. In Forest soils: Properties and processes. University of Toronto Press, Toronto.
• Block, R., K.C.J. Van Rees, and D.J. Pennock. 2002. Quantifying harvesting impacts using soil compaction and disturbance regimes at a landscape scale. Soil Sci. Soc. Am. J. 66:1669–1676.[Abstract/Free Full Text]
• Canada Soil Survey Committee. 1992. The Canadian system of soil classification. Publ. 1646. Agriculture Canada, Ottawa, ON.
• Canadian Climate Program. 1982. Canadian climate normals. 1951–1980. vol. 6. Frost. Environment Canada, Atmospheric Environment Service, Downsview, ON.
• Canadian Climate Program. 1993. Canadian climate normals 1961–1990. Québec. Environment Canada, Atmospheric Environment Service, Downsview, ON.
• Corns, I.G.W. 1988. Compaction by forestry equipment and effects on coniferous seedling growth on four soils in the Alberta foothills. Can. J. For. Res. 18:75–84.
• Curtis, R.O., and B.D. Post. 1964. Estimating bulk density from organic-matter content in some Vermont forest soils. Soil Sci. Soc. Am. Proc. 28:285–286.
• Federer, C.A. 1983. Nitrogen mineralization and nitrification: Depth variation in four New England forest soils. Soil Sci. Soc. Am. J. 47:1008–1014.[Abstract/Free Full Text]
• Federer, C.A., D.E. Turcotte, and C.T. Smith. 1993. The organic fraction—Bulk density relationship and the expression of nutrient content in forest soils. Can. J. For. Res. 23:1026–1032.
• Froehlich, H.A., and D.H. McNabb. 1984. Minimizing soil compaction in Pacific northwest forests. p. 159–192. In E.L. Stone (ed.) Forest soils and treatment impacts. Proceedings of the 6th North American Forest Soil Conference. Knoxville, TN. Soc. of Am. Foresters. Publication 84-10. Dep. of Forestry, Wildlife, and Fisheries, University of Tennessee, Knoxville.
• Grigal, D.F., R.E. McRoberts, and L.F. Ohmann. 1991. Spatial variation in chemical properties of forest floor and surface mineral soil in the north central United States. Soil Sci. 151:282–290.
• Huntington, T.G., C.E. Johnson, A.H. Johnson, T.G. Siccama, and D.F. Ryan. 1989. Carbon, organic matter, and bulk density relationships in a forested spodosol. Soil Sci. 148:380–386.
• Järvinen, E., T.J. Hokkanen, and T. Kuuluvainen. 1993. Spatial heterogeneity and relationships of mineral soil properties in a boreal Pinus sylvestris stand. Scand. J. For. Res. 8:435–445.
• Kalra, Y.P., and D.G. Maynard. 1991. Methods' manual for forest soil and plant analysis. Inf. Rep. NOR-X-319. Canadian Forest Service, Northern Forestry Centre, Edmonton, AB.
• Macdonald, S.E., M.G. Schmidt, and R.L. Rothwell. 1998. Impacts of mechanical site preparation on foliar nutrients of planted white spruce seedlings on mixed-wood boreal forest sites in Alberta. For. Ecol. Manage. 110:35–48.
• McKeague, J.A. 1978. Manual of sampling methods and soil analysis. 2nd ed. Canadian Society of Soil Science, Ottawa, ON.
• McNabb, D.H., A.D. Startsev, and H. Nguyen. 2001. Soil wetness and traffic level effects on bulk density and air-filled porosity of compacted boreal forest soils. Soil Sci. Soc. Am. J. 65:1238–1247.[Abstract/Free Full Text]
• Nelson, D.W., and L.E. Sommers. 1982. Total carbon, organic carbon, and organic matter. p. 539–579. In A.L. Page et al. (ed.) Methods of soil analysis. Part 2. Agron. Monogr. 9. ASA and SSSA, Madison, WI.
• Örlander, G., P. Gemmel, and J. Hunt. 1990. Site preparation: A Swedish overview. FRDA Rep. 105. Canadian Forest Service, Pacific Forestry Centre, Victoria, BC.
• Powers, R.F., A.E. Tiarks, and J.R. Boyle. 1998. Assessing soil quality: Practicable standards for sustainable forest productivity in the United States. p. 53–80. In J.M. Bigham et al. (ed.) The contribution of soil science to the development and implementation of criteria and indicators of sustainable forest management. SSSA Spec. Publ. 53. SSSA, Madison, WI.
• Prévost, M. 1996. Effets du scarifiage sur les propriétés du sol et l'ensemencement naturel dans une pessière noire à mousses de la forêt boréale Québécoise. (In French with an English abstract.) Can. J. For. Res. 26:72–86.
• Prévost, M. 1997. Effects of scarification on seedbed coverage and natural regeneration after a group seed-tree cutting in a black spruce (Picea mariana) stand. For. Ecol. Manage. 94:219–231.
• Robitaille, A., and J.P. Saucier. 1998. Paysages régionaux du Québec méridional. Ministère des Ressources naturelles du Québec. (In French.) Direction de la gestion des stocks forestiers et Direction des relations publiques, Québec, QC.
• Rowe, J.S. 1972. Forest regions of Canada. Publ. 1300. Canada Forest Service, Ottawa, ON.
• SAS Institute. 1996. SAS user's guide: Statistics. Ver. 6.12 SAS Inst. Inc., Cary, NC.
• Saucier, J.P., J.P. Berger, H. D'Avignon, and P. Racine. 1994. Le point d'observation écologique: Normes techniques. (In French.) Ministère des Ressources naturelles du Québec, Direction de la gestion des stocks forestiers, Québec, QC.
• Saucier, J.P., J.F. Bergeron, P. Grondin, and A. Robitaille. 1998.The land regions of Southern Québec (3rd Version): One element in the hierarchical classification system developed by Québec's Ministère des Ressources naturelles. Ministère des Ressources naturelles du Québec, Direction de la gestion des stocks forestiers. Québec,QC.
• Schmidt, M.G., S.E. Macdonald, and R.L. Rothwell. 1996. Impacts of harvesting and mechanical site preparation on soil chemical properties of mixed-wood boreal forest sites in Alberta. Can. J. Soil Sci. 76:531–540.
• Soil Survey Staff. 1998. Keys to soil taxonomy. 8th ed. USDA-NRCS, Washington, DC.
• Sutton, R.F. 1987. Root growth capacity and field performance of jack pine and black spruce in boreal stand establishment in Ontario. Can. J. For. Res. 17:794–804.
• Sutton, R.F. 1993. Mounding site preparation: A review of European and North American experience. New For. 7:151–192.
• Terry, T.A., D.K. Cassel, and A.G. Wollum. 1981. Effects of soil sample size and included root and wood on bulk density in forested soils. Soil Sci. Soc. Am. J. 45:135–138.[Abstract/Free Full Text]
• Thomas, R.L., R.W. Sheard, and J.R. Moyer. 1967. Comparison of conventional and automated procedures for nitrogen, phosphorus, and potassium analysis of plant material using a single digestion. Agron. J. 59:240–243.[Abstract/Free Full Text]
• Tremblay, S., R. Ouimet, and D. Houle. 2002. Prediction of organic carbon content in upland forest soils of Québec, Canada. Can. J. For. Res. 32:903–914.

This article has been cited by other articles:

				S. Zacharias and G. Wessolek Excluding Organic Matter Content from Pedotransfer Predictors of Soil Water Retention Soil Sci. Soc. Am. J., January 1, 2007; 71(1): 43 - 50. [Abstract] [Full Text] [PDF]

				N. Volland-Tuduri, A. Bruand, M. Brossard, L. C. Balbino, M. I. L. de Oliveira, and E. de Souza Martins Mass Proportion of Microaggregates and Bulk Density in a Brazilian Clayey Oxisol Soil Sci. Soc. Am. J., September 1, 2005; 69(5): 1559 - 1564. [Abstract] [Full Text] [PDF]

				B. De Vos, M. Van Meirvenne, P. Quataert, J. Deckers, and B. Muys Predictive Quality of Pedotransfer Functions for Estimating Bulk Density of Forest Soils Soil Sci. Soc. Am. J., March 1, 2005; 69(2): 500 - 510. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (10) Citing Articles via Google Scholar Google Scholar Articles by Prévost, M. Search for Related Content PubMed Articles by Prévost, M. Agricola Articles by Prévost, M. Related Collections Soil Organic Matter Other Soil Management Forest Soils