BLACK SPRUCE {Picea mariana [Mill.] B.S.P.) BOREAL ECOSYSTEMS: 
HOW TREE-LENGTH AND FULL-TREE HARVESTING AFFECTS 
SOIL MICROBIAL POPULATIONS 
by 
ANNA ARVONIO 
A graduate thesis submitted in partial fulfilment 
of the requirements for the degree ofr 
Master of Science in Forestry 
Faculty of Forestry and the Forest Environment 
Lakehead University 
Thunder Bay, Ontario 
ProQuest Number: 10611925 
All rights reserved 
INFORMATION TO ALL USERS 
The quality of this reproduction is dependent upon the quality of the copy submitted. 
In the unlikely event that the author did not send a complete manuscript 
and there are missing pages, these will be noted. Also, if material had to be removed, 
a note will indicate the deletion. 
Pro 
ProQuest 10611925 
Published by ProQuest LLC (2017). Copyright of the Dissertation is held by the Author. 
All rights reserved. 
This work is protected against unauthorized copying under Title 17, United States Code 
Microform Edition © ProQuest LLC. 
ProQuest LLC. 
789 East Eisenhower Parkway 
P.O.Box 1346 
Ann Arbor, M! 48106 - 1346 
11 
ACKNOWLEDGEMENTS 
I would like to thank Nancy Luckai for acting as my main advisor, as well 
as the great academic and moral support that she has provided for me. It was 
greatly appreciated. I extend my thanks and appreciation to Dr. D. Morris for his 
advice, as well as for residing on my committee. Further, I would like to express 
my gratitude to Dr. Morris for providing me with generous financial support. I 
would also like to thank Dr. W. L. Meyer for residing on my committee, and for 
lending me his advice on soil science. A special thank-you goes out to Dr. O. 
Hendrickson for acting as my external reader, and for responding in a timely 
fashion. In addition, I extend my thanks to those who helped me along the way by 
providing assistance in the field as well as in the laboratory. 
I would like to express my deepest thanks and gratitude to all my fiiends 
and family for providing me with great moral support. Without them, I would 
never have come this far, thank-you all. 
June 27, 2000 
A. A. 
Ill 
ABSTRACT 
Arvonio, A. 2000. Black spruce (Picea mariana [Mill.] B.S.P.) Boreal 
ecosystems: how tree-length and full-tree harvesting affect soil microbial 
populations. 106 pp + append. Advisor: N. Luckai 
Key Words: microbial biomass, soil organic matter, soil respiration, API 20E, 
substrate utilization, chloroform fumigation extraction. 
To evaluate the hypothesis that microbial populations are affected by 
organic biomass removals, a study was designed to complement ongoing work in a 
black spruce {Picea mariana [Mill.] B.S.P.) Boreal ecosystem. The treatments 
included a control (uncut) and two harvest levels, tree-length (TL) and full-tree 
(FT). Soil samples from the organic and mineral horizons were taken from nine 
plots, representing three each of the treatments. Samples were taken once a month 
for four consecutive months; May through August during 1998. Soil respiration 
on two dates in September was estimated using the soda-lime technique. Bacterial 
cultures were prepared from the soil samples and pure strains identified using 
morphological and substrate utilization characteristics (specifically API 20E). Soil 
descriptors, including pH, total nitrogen, total phosphorus, organic matter content, 
and moisture content, were measured to investigate relationships with microbial 
biomass. Microbial biomass carbon (MBc) and nitrogen (MBN) were estimated 
using chloroform fumigation extraction. The data were statistically analyzed, using 
ANOVA, Pearson and Spearman correlations and, in the case of the MBc and 
MBN, ANCOVA, to determine if there were any treatment or seasonal effects. 
Soil respiration demonstrated a significant treatment effect where the efflux was 
significantly greater on the control treatment compared to the harvest treatments. 
Five bacterial cultures were identified from the soil samples, Chryseomonas 
luteola, Aeoromonas salmonicida, Serratia marcescens, Syntrophomonas 
multifilia, and Pseudomonas fluorescens. MBc and MBN values measured were in 
agreement with other published values for boreal coniferous soils. The MBc in the 
organic horizon was significantly affected by the interaction of the treatment and 
month factors. There was a significant treatment effect on the MBc in the mineral 
horizon, with the control mean significantly higher than those of the harvest 
treatments. The MBN revealed no significant effects in either the organic or the 
mineral horizons. The author concludes that soil moisture and temperature did 
affect the values for microbial biomass and that these environmental conditions 
were likely impacted by the level of harvest. 
IV 
CONTENTS 
Page 
ACKNOWLEDGEMENTS ii 
ABSTRACT iii 
TABLES iv 
FIGURES viii 
INTRODUCTION 1 
LITERATURE REVIEW 4 
METHODOLOGY 29 
Sample Site 29 
Field Sampling Techniques 32 
Laboratory Analysis 3 5 
Statistical Analysis 43 
RESULTS 50 
Bacteriology 50 
Soil Respiration 52 
Soil Descriptors 55 
Microbial Biomass 59 
V 
DISCUSSION 77 
Bacteriology 77 
Soil Respiration 80 
Variability in Soil Descriptors 82 
Relationship Between Soil Descriptors & Microbial Biomass 84 
Variability in Microbial Biomass 89 
CONCLUSIONS 92 
RECOMMENDATIONS 95 
LITERATURE CITED 96 
APPENDIX I: COMPLETE DATA SET FOR THE ORGANIC 
AND MINERAL SOIL 108 
APPENDIX II: COMPLETE DATA SET FOR THE SOIL RESPIRATION 110 
VI 
TABLES 
Table Page 
1. Physical soil characteristics and classification of soil profiles associated 
with the black spruce study site. 31 
2. Expected Mean Squares table associated with Eq. 12. 46 
3. Expected Mean Squares table associated with Eq. 13. 49 
4. A description and summary of the bacteria species isolated from organic 
and mineral soil samples. Summer 1998. 51 
5. Summary of the amount of carbon respired (g hr'^ m‘^) from the three 
treatments associated with the black spruce study site, Summer 1998 52 
6. Analysis of variance table associated with CO2 evolved (g hr'^ m^). 52 
7. Summary of the soil parameters measured in the organic and mineral 
soil horizons, summer 1998. 56 
8. One-way Analysis of variance table associated with the harvest treatment 
factor tested against the environmental parameters in the organic and 
mineral soil horizons. 57 
9. Summary of Microbial Biomass C and N for the organic and mineral soil 
horizon, summer 1998. 60 
10. Analysis of variance table associated with the microbial biomass carbon 
in the organic soil horizon. 61 
11. Analysis of variance table associated with the microbial biomass carbon 
in the mineral soil horizon. 61 
12. Analysis of variance table associated with the microbial biomass nitrogen 
in the organic soil horizon. 62 
13. Analysis of variance table associated with the microbial biomass nitrogen 
in the mineral soil horizon. 62 
14. Results of the Pearson and Spearman Correlation analyses for the organic 
and mineral horizons. 71 
Vll 
15. Analysis of covariance table associated with the microbial biomass carbon 
in the organic soil horizon. 72 
16. Analysis of covariance table associated with the microbial biomass carbon 
in the mineral soil horizon. 73 
17. Analysis of covariance table associated with the microbial biomass nitrogen 
in the organic soil horizon. 75 
18. Analysis of covariance table associated with the microbial biomass nitrogen 
in the mineral soil horizon. 76 
Vlll 
FIGURES 
Figure Page 
1. Map of northern Ontario identifying the black spruce study site location. 30 
2. Map of the black spruce study site and treatment plot location. 33 
3. The average temperature and precipitation values associated with the 
black spruce study site for 1998. 34 
4. Results of the LSD Post Hoc test demonstrating similar and dissimilar 
harvest treatments, using mean CO2 efflux (g hr'^ m‘^) as the response 
variable. 54 
5. LSD Post Hoc test demonstrating similar and dissimilar harvest 
treatments in the organic soil horizon, using mean pH as the response 
variable 58 
6. Mean MBc (Og/ g soil) throughout the sampling season, demonstrating 
interactions between harvest treatments. 64 
7. Mean MBN (Dg/ g soil) throughout the sampling season, demonstrating 
interactions between harvest treatments. 65 
8. Scattergrams of the microbial biomass carbon with the soil description 
parameters in the organic soil horizon. 67 
9. Scattergrams of the microbial biomass carbon with the soil description 
parameters in the mineral soil horizon. 68 
10. Scattergrams of the microbial biomass nitrogen with the soil description 
parameters in the organic soil horizon. 69 
11. Scattergrams of the microbial biomass nitrogen with the soil description 
parameters in the mineral soil horizon. 70 
12. LSD Post Hoc tests demonstrating similar and dissimilar harvest 
treatments, using mean MBc (Dg/ g soil) as the response variable 
in the mineral soil horizon. 74 
1 
INTRODUCTION 
The microbial biomass, defined as the functional component of the micro-biota 
primarily responsible for decomposition, organic matter turnover, and nutrient 
transformation, is an important component of soil (Haron et al. 1998). Microbial 
biomass includes organisms such as bacteria, fungi, actinomycetes, microfauna and algae 
(Killham 1994). Acting as both a source and a sink of nutrients in soils, the microbial 
biomass, which depends mainly on organic matter for energy, influences the amount of 
available nutrients to plants via the mineralization-immobilization process (Gallardo and 
Schlesinger 1994). Microorganisms, therefore, play a major role in determining the 
relative quality of forest soil within the limitations of climate and topography (Entry et al. 
1986). 
Microbial biomass has been used to indicate and predict changes in various 
ecosystems caused by natural and anthropogenic disturbances (Weber 1990). Microbial 
biomass has been used as an indicator of global warming (Anderson and Joergensen 
1997), sustainable development (Cairns and Meganck 1994), and biodiversity (Staddon et 
al 1996). Microbial biomass has served many research applications in forestry. For 
example, it has been used in monitoring the effects of: municipal solid waste leachate on 
forest soils (Gordon et al. 1988); xenobiotic substances in soil (Dictor et al 1998); fire 
(Diaz-Ravina et al 1992); tillage induced changes (Carter 1986); pesticide application 
(Wardle and Parkinson 1990); response to thinning (Thibodeau et al. 2000) and microbial 
stress (Anderson and Joergensen 1997). 
2 
In addition, some components of the soil microbial biomass may act as natural 
biological control agents (Staddon et al. 1996). For example, the hdiCX&Lidi Pseudomonas 
fluorescens is antagonistic to soil-borne root pathogens that inhabit the soil rhizosphere 
(Killham 1994). This bacterial strain restricts a wide range of pathogens including: 
Erwinia carotovora, black leg in potatoes; Pythium spp., damping-off fungi; Rhizoctonia 
solani, cereal bare patch; and Thielaviopsis basicola, tobacco black root rot (Killham 
1994). Continued study of these microorganisms and analysis of the molecular genetics 
of these antagonistic responses may lead to new biocontrols for other plant pathogens 
through genetic manipulation (Killham 1994). 
Since little is known about the diversity of Boreal (northern) forest soil 
microorganisms, species loss or modification may deprive the ecosystem of processes, 
products, or genetic material provided by undescribed microbes (Staddon et al 1996). 
Many studies have reported that the microbial biomass is affected by various timber 
harvesting treatments (Hendrickson et al 1985; Entry etal 1986; Foster and Morrison 
1987; Weber 1990; Smolander et al 1998). For example, timber harvesting may 
indirectly alter microbial biomass and activity by changing the amount and type of organic 
matter inputs, soil pH, soil temperature, and soil moisture (Entry et al 1986). 
The author was fortunate to have access to an established trial investigating the 
effects of various levels of biomass removal on a black spruce {Picea mariana [Mill.] 
B.S.P.) Boreal ecosystem. This study will attempt to augment the existing trial by asking 
the following questions; 
3 
1. Do increasing levels of biomass removal result in quantitative differences in soil 
microbial populations? 
2. Do increasing levels of biomass removal result in differences in the bacterial 
species isolated from treated plots? 
3. Do soil microbial populations exhibit seasonal responses and are these responses 
influenced by biomass removal? 
4. Are there correlations between observed/measured fluctuations in soil microbial 
populations and soil parameters, such as pH, organic matter, total nitrogen, total 
phosphorous, and moisture content? 
These questions will be addressed as follows: 
1. Using the fumigation-extraction technique (Voroney et al. 1993), estimate the 
microbial biomass C and N found in representative samples from the study area, 
2. Following the traditional taxonomic approach, isolate, culture and describe (Gram 
stain, cell and colony appearance, etc.) representative and unique bacteria, 
3. Using the soda-lime respiration technique (Edwards 1982), estimate the soil 
respiration, 
4. Using a functional or substrate utilization approach, specifically API 20E 
(Anonymous 1996), attempt to identify/characterize the isolated bacteria, and. 
5. Using appropriate statistics, attempt to identify relationships between the 
microbial populations, treatment regimes and certain environmental parameters. 
4 
LITERATURE REVIEW 
The microbial biomass, which acts as both a source and a sink for nutrients, 
influences the availability of nutrients in the soil through the complementary processes of 
mineralization and immobilization. Mineralization is the process of releasing organically 
bound nutrients into mineral forms, which are then available for plant uptake, while 
immobilization refers to the process of microbial uptake of mineral nutrients, thus 
rendering them unavailable to plants (Killham 1994). The rates of these two processes 
depend, in part, upon the size of the microbial biomass, the composition of the microbial 
community (e.g., fiingi versus bacteria, nitrifiers versus denitrifiers), the quality of organic 
matter inputs, physical and chemical characteristics of the soil (e.g, temperature, water 
content, % clay, pH) as well as the vegetation and climate above the ground. Each 
component will influence the others ultimately affecting the condition of the soil. Using 
forest canopy as an example, the presence of closely spaced spruce trees may cause the 
cycling of nutrients to slow down due to lower soil temperatures and acidic foliage 
inputs, whereas the presence of alder, with less acidic foliage and nitrogen fixing root 
nodules, may improve soil fertility (Bradley and Fyles 1995). Since microorganisms 
influence decay processes, mineral conversions, and plant root activity in the soil 
environment, plant growth conditions are therefore affected (Entry et al 1986). 
The elements affecting nutrient cycling and availability are numerous, complicated 
and interconnected. A change in any one may lead to subsequent changes in some or all 
of the other elements. The concept of “stability” or steady state in the microbial biomass 
5 
must, therefore, be used advisedly under such dynamic conditions. Nonetheless, some 
trends have been reported in the literature and will be summarized here. 
In the literature review, the author will attempt to cover the major factors 
affecting microbial biomass quantity and composition: organic matter (OM) inputs, soil 
pH, soil temperature, soil moisture content, and seasonal conditions. Soil microbial 
activity measurements, specifically respiration, are also included due to their potential as 
ecosystem monitors following disturbance events. A discussion on microorganism 
culturing techniques follows, providing insights to the advantages and disadvantages of 
the various methodological approaches. 
ORGANIC MATTER INPUTS AND DECOMPOSITION 
Approximately half of OM is carbon (C) (Paul and Clark 1989). Carbon 
compounds incorporated in the organic matter of soil are required primarily by the soil 
micro-biota for energy and nutrition (Paul and Clark 1989). Carbon consumption by 
microbes generates energy required to synthesize and break down other chemical 
compounds, such as organically bound nitrogen (N) and phosphorus (P), which are used 
for nutrition (Killham 1994). 
Carbon and nitrogen are utilized by the micro-biota for cell wall formation and 
maintenance (Paul and Clark 1989). The cell walls of bacteria are made up of sugar 
derivatives, N-acteylglucosamine and N-acetylmuramic acid, that are linked by amino 
acids via peptide bonds (Paul and Clark 1989). Organic matter containing both these 
elements is therefore necessary for microbial biomass growth. 
6 
Carbon is present in the soil organic matter (SOM) in a variety of inorganic and 
organic forms. The latter, however, predominate and may include other elements like N 
and P. These organic compounds could occur as biomass C, root exudates, cellulose, 
lignin, chitin, or humus (Killham 1994). The biomass C is made up of soil microbes and 
animals (Killham 1994). This pool of C represents only 1-2% of the total organic C in 
soil. However, it is the backbone of the soil C-cycle and all other nutrient cycles (Killham 
1994). Root exudates, a highly decomposable form of organic C (Paul and Clark 1989) 
represent less than 1% of total soil organic carbon, but provide an immediate substrate 
for many species of soil microbes (Killham 1994). 
Cellulose and lignin (remnants of plant residues), chitin (remnants of soil animals 
and fungal residues), and soil humus (previously decomposed material) comprise 90% of 
the total soil organic C (Killham 1994). These compounds are highly resistant to 
decomposition due to their complex chemical nature. Decomposition follows a process 
whereby each compound is slowly broken down into smaller units, and then eventually 
into a form available for assimilation by the microbial biomass. Cellulose, for example, is 
broken down into glucose by specialized cellulytic saprophytes, such as the fungal species 
Fusarium and Aspergillus, or the bacterial species Bacillus and Pseudomonas (Killham 
1994). 
Lignins, having random structures and strong linkages, are even more resistant to 
decomposition than cellulose (Killham 1994). White rot fungi, such as Coriolus 
\ersicolor and Phanerochaete chrysosporium mediate the degradation of lignin into CO2 
and H2O (Paul and Clark 1989). Humus, the product of decomposition of fresher 
7 
material, has three constituents, humic acid, flilvic acid, and humin (Anderson and 
Schoenau 1993). Because each constituent is recalcitrant, humus is constantly present in 
the SOM and is continually being formed at the same time that it is being degraded 
(Alexander 1961). 
The quality and availability of the substrate, as defined by the chemical 
composition of decomposing material, influences the microbial growth rate (Bosatta and 
Agren 1994). Thus, properties of the original litter together with soil physical factors are 
important in determining the amount of microbial biomass in the soil (Bosatta and Agren 
1994). Bauhus et al. (1998) state that microbial biomass amounts are sensitive to 
changes in soil physical and chemical composition. For example, soils high in lignin 
residues can only support a low level of microbial activity. This is because lignin is 
relatively high in N and is a poor quality resource for soil microbes that require C (Brady 
and Weil 1996). Furthermore, Killham (1994) states that resource quality is described by 
C/N ratios; low C/N ratios indicate high litter quality and rapid rates of decomposition; 
high C/N ratios indicate poor litter quality and slow decomposition rates. 
Litter inputs are basically from two sources - plant and animal residues. Plant 
litter may include any part of the plant from seed to root, and the decomposition rates of 
the various tissues vary according to the chemical structure involved (Haynes 1986). For 
example, the leaves and stems of a plant could contain up to 60% of N occurring as 
enzymes or proteins and 40% as free amino acid-N. The former require much time and 
energy for degradation, while the latter are water soluble and easily broken down 
(Haynes 1986). 
8 
Animal residues vary in components and turnover rates, just as plants do. Some 
important differences between animal and plant residues do exist, such as the presence of 
chitin in animal residues (Haynes 1986). Chitin, a contribution from the exoskeletons of 
arthropods and eggs of nematodes, is stable in soil (Haynes 1986). Furthermore, animals 
contribute urine and feces to the soil. Urine, containing 50-80% of N in the form of urea, 
is easily degraded following the soil process of hydrolysis to form ammonium (NH4^) 
which is readily immobilized by microbes and plants (Haynes 1986). Feces, in 
comparison, has a slower turnover rate due to its composition of highly resistant organic 
forms ofN (Haynes 1986). 
In addition, the type of organic matter in the soil is also dependent on the quality 
of foliage found in the ecosystem. For example, the soils of mixed-wood stands more 
often contain simple sugars, starches and cellulose (Rowell 1994) which are easily broken 
down and assimilated by microbes (Bradley and Fyles 1995). In contrast, the soil of 
coniferous stands includes more complex organic residues exhibiting slower turnover 
rates which in turn restricts microbial numbers (because of the poor quality and acidic soil 
conditions) thus slowing decomposition (Bradley and Fyles 1995). Furthermore, Bradley 
and Fyles (1995) state that previous studies have found that environmental factors, such 
as extremely low pH and reduced organic matter (OM) residues in coniferous stands, 
negatively influenced microbial activity and litter turnover rates. Also, as stand age 
increases the quality of organic matter inputs decreases, and this was reflected in declines 
in the microbial biomass carbon-to-soil organic carbon ratio and the microbial biomass 
nitrogen-to-soil organic nitrogen ratio found by Bauhus et al (1998). 
9 
Assuming that sufficient C is available in the OM, it is appropriate to consider the 
presence and availability of macronutrients such as N and P. It is generally accepted that 
soil microbial populations under steady state conditions maintain an average C/N ratio of 
8:1 (Rowell 1994) with the range being from 15:1 for some fungi to 3:1 for some bacteria 
(Paul and Clark 1989). Since only one-third of the carbon metabolized by 
microorganisms is assimilated (the rest is lost to respiration), substrates with a 24:1 C/N 
ratio could be considered ideal. Paul and Clark (1989) noted that OM inputs at 25:1 
ratios result in no net mineralization or immobilization. Organic matter inputs in forest 
systems usually exceed this relatively low ratio. For example, leaves of oak trees have 
ratios of 65:1 while pine needles have ratios of 225:1 (Haynes 1986). When high C/N 
ratio residues are added to the soil, competition among microorganisms for scarce N 
resources restricts the rate of decomposition and limits the amount of inorganic N 
available for plant uptake. Complex chemical structures, such as lignins found in conifer 
needles, decompose very slowly as evidenced by low C oxidation and N mineralization 
rates (Brady and Weil 1996). Therefore, N may be limiting to microbial activity and 
numbers even when adequate C supplies are present (Brady and Weil 1996). 
Gallardo and Schlesinger (1994) reported that P also has a limiting effect on 
microbial biomass. They found that P was less frequently immobilized than N by soil 
microorganisms in litter, causing an increase of N/P ratio in the litter. This ratio increase 
suggests that P and not N may be limiting in the F-H horizon. Phosphorous availability 
could also become limiting in the lower horizons through adsorption by A1 and Fe 
sesquioxides (Gallardo and Schlesinger 1994). 
10 
The rate of OM decomposition is also dependent upon the various species of 
microbes present in the soil, and the timing of the OM inputs. Rapid turnover of biotic 
residues, for example, is mediated by zymogenous (bacilli and spore forming bacteria) 
soil microbes, which do not occur as a numerically significant component of the soil 
community under “normal” conditions, but proliferate when large amounts of organic 
residues are added into the soil (Killham 1994). The on-going, low level cycling of 
organic C is a fianction of autochthonous (coccoid bacteria) soil microbes (Killham 1994). 
These bacteria are most competitive at low substrate concentrations and persist actively 
in soil for long periods of time (Killham 1994). However, while bacteria alone may 
“stall” on complex substrates such as lignins, net mineralization has been noted when 
fungi are present (Paul and Clark 1989). 
Timber harvesting can affect above- and below-ground aspects of the forest 
ecosystem. For example, the removal of a forest canopy will immediately affect annual 
organic matter inputs in the form of foliage and fine root mortality. One-time additions 
of coarse woody debris, such as large roots and branches, will drastically change the 
organic matter pools in the soil and at the forest floor (Chatarpaul et at 1984; 
Hendrickson et al 1985; Entry et al. 1986; Foster and Morrison 1987). Measured 
characteristics which support reported reductions in fungal and bacterial biomass include 
slowed rates of leaf and woody litter decay, excessive drying and high temperatures at the 
forest floor surface, increased CO2 efflux and changes in soil N levels (Hendrickson et al 
1985). While the act of harvesting itself results in significant changes in the soil 
environment, there is evidence that harvesting intensity further influences the magnitude 
11 
of the change (Chatarpaul et al 1984; Hendrickson et al 1985; Entry et al 1986; Foster 
and Morrison 1987). Thibodeau et al. (2000) investigated the effect of pre-commercial 
thinning on microbial populations in balsam fir stands in Quebec. They hypothesized that 
both the change in the soil temperatures and the input of foliage and branches would 
positively affect the microbial biomass immediately following treatment. In fact, no 
significant changes were noted in either MBc nor MBN in the mineral layer but a strong 
relationship was found between MBN and soil temperature. 
In the Boreal forest of Canada, clear-cutting predominates. Most commonly, 
either a stems-only (also known as conventional) or full-tree technique is employed. In 
the former, the tree is processed at the stump leaving foliage and branches scattered 
throughout the site. In the latter, the tree is processed at roadside with foliage and 
branches being removed from the site. The chief difference in the two techniques lies in 
the amount and type of organic residue left in the stand. As of 1987, 65% of harvesting 
in Ontario’s boreal forest was full-tree (Wiensczyk 1992). 
Many authors have considered the question of nutrient capital depletion due to 
the removal of relatively nutrient rich foliage and small branches (for a summary see 
Wiensczyk 1992). The project upon which this thesis is based was established in 
response to concerns expressed about full-tree harvesting in the Class Environmental 
Assessment (Ontario Ministry of Environment 1994). In addition to the “balance-sheet” 
approach employed by many researchers, it is also necessary to consider potential 
changes in the microbiological system responsible for nutrient transformations. 
12 
Entry et al (1986) observed that the bacterial biomass was significantly higher in 
a cutover where the organic residue was left on-site {i.e. stems-only) compared to 
treatments where organic residues were removed. Hendrickson et al. (1985) found that 
whole-tree harvesting resulted in decreased microbial biomass when compared to 
conventional harvest methods. The difference was attributed to a loss of residual OM in 
the former. Hendrickson et al. (1985) also reported that due to greater on-site traffic 
during the whole tree harvest, increased mixing of the forest floor with mineral soil 
resulted in a reduction of water-holding capacity and OM content of the soil, which 
implied a reduction in microbial activity. 
After harvest, silvicultural treatments intended to promote the growth of crop tree 
species may also affect microbial populations. Ohtonen et al. (1992) found that intensive 
silvicultural activities {e.g., scarification, fertilization and herbicide application) generally 
reduced microbial biomass on coniferous sites found in Central Ontario (Petawawa 
Research Forest). The modified scarification treatment alone, which removed the humus 
layer, caused a nutrient limitation as evidenced by a widening of the C/N ratio in the 
mineral soil (Ohtonen et al. 1992). 
Bauhus et al. (1998) suggested that microbial biomass was influenced by soil 
texture and SOM quality. Microbial populations were found to be sensitive to changes in 
the soil chemistry and physical environment, and were negatively influenced by forest 
management practices. Management practices, such as harvesting and silviculture, could 
cause the microbial population to decline, and ultimately have a negative affect on OM 
turnover rates. Ohtonen et al. (1992) noted that the reduction of microbial biomass per 
13 
unit OM and the reduction of microbial biomass C in surface organic and mineral layers 
was indicative of a reduced capacity of the ecosystem to maintain its nutrient reservoir. 
This reduced nutrient reservoir results in slower decay rates because substrates containing 
readily mineralizable N were less available (Hendrickson et al 1985). 
SOIL MOISTURE AND TEMPERATURE 
The general consensus throughout the literature seems to be that moisture and 
temperature levels are key components in determining the nature and size of soil 
microbial populations. Soil bacteria require a water film for movement and can only 
remain active if there is suitable moisture in the soil, so that nutrients and waste products 
can diffuse in and out of the bacterial cell (Wong and Griffen 1976a & b). Thus, bacterial 
activity may increase or decrease as the water potential rises and falls, respectively. 
Changes in microbial activity can be estimated by soil respiration rates. According to 
Schlentner and Van Cleve (1985), soil respiration fluctuates as moisture or temperature 
changes, depending on which of the two parameters is most limiting at the time of 
measurement. At moisture contents less than 75%, by weight of soil, temperature 
increases had little effect on respiration, whereas at moisture contents of 100-250%, 
respiration increased with temperature. Alternatively, moisture levels had little influence 
on respiration when temperatures were below 5”C; however, at temperatures of 10-20”C, 
respiration increased with moisture changes. 
Lundgren and Soderstom (1983) reported that in podzolic soils, seasonal changes 
in precipitation and moisture content exerted strong influences on soil bacteria numbers. 
Precipitation provides moisture and available nutrients to the soil, thus, immediately 
altering the soil environment. This increases the soil moisture content and the soil 
microbial biomass (Lundgren and Soderstom 1983). Evaporation, which is dependent on 
air temperature and humidity, may rapidly decrease this moisture content so that no 
detectable changes in soil bacteria populations can be found a few days after rainfall. 
Alternately, Weber (1990) reported that rainfall events did not result in higher 
water contents or altered soil respiration rates of cut and burned aspen ecosystems at the 
Petawawa Research Forest. Furthermore, he indicated that temperature is more closely 
correlated to soil respiration, with a strong temperature control exerted over respiration 
patterns in both cut and burned treatments of aspen. Howard and Howard (1979) found 
that in hardwood stands, microbial numbers were not significantly correlated with 
moisture content. Moisture and temperature accounted for 5% and 64%, respectively, of 
the variation in soil respiration. Orchard and Cook’s (1983) results correspond with 
those of Lundgren and Soderstom (1983), where they found a correlation between soil 
respiration (an indicator of microbial activity) with soil moisture content; as the soil 
moisture content decreased, respiration decreased, reflecting an estimated 10% reduction 
in microbial activity. Rapid increases in respiration rate immediately following rewetting 
of the soil indicated 1) the death of some organisms, and 2) that many microorganisms 
are capable of surviving water stress and resuming activity quickly in response to 
favourable changes in their environment. Orchard and Cook (1983) also suggested that it 
15 
was likely that an increase in activity, rather than in biomass, was responsible for 
increased respiration rates. 
Berg et al (1998) suggested that seasonal conditions might have a direct 
influence on the microbial biomass by inducing specific microbial community responses to 
soil moisture and temperature. For example, microbial biomass declines during periods 
with extreme climate conditions. They further speculated that seasonal effects on plant 
productivity and organic matter release also indirectly influence densities of soil fauna 
populations, and interactions between grazers and microflora. 
Salonius (1983b) suggested that air drying of soils may lower species diversity, 
resulting in a significant reduction of the metabolic activity of the population (using a soil 
suspension method), as compared to that of an undried soil. Damage due to drying was 
found to be less in the H horizon than in the L and F horizons. Salonius (1983b) further 
recommended that if a soil is not to be studied immediately after sampling then it should 
be stored moist to maximize the amount of living microbial biomass. 
Mixing of humus with mineral soil, as Salonius (1983 a) reported, may also lead 
to increased soil temperatures, and enhanced organic matter decomposition. Clay 
colloids may act to buffer the soil environment against toxic accumulation of metabolic 
end products of the developing microbial population, thus allowing activity levels to be 
enhanced. However, this buffering effect may simply be more obvious in populations at 
temperatures of 20-40X, as compared to less active populations at 10“C (Salonius 
1983a). Anderson (1978) suggested that enhanced decomposition may be attributed to 
16 
increased microbial activity and species diversity resulting from the mixing of soil layers, 
which has subsequently created a greater diversity of micro-habitats. 
SOIL pH 
Any removal of forest vegetation may result in a change of soil pH, with 
subsequent consequences for the integrity of microbial functional groups, as well as for 
microbial processes (Gallardo and Schlesinger 1994). Gallardo and Schlesinger (1994) 
speculated that P limitations in soils with highly basic or acidic pH levels will affect the 
activity and nature of microorganisms present. Baath et ah (1995) suggested that pH 
may have the ability to alter other soil properties, such as the C/N ratio, which indirectly 
affect microbial community composition by restricting available nutrients. Staddon et al. 
(1996) reported that a change in soil pH results in a loss of species in microbial functional 
groups. They stated that cellulose decomposition, for example, is predominately 
mediated by filamentous fungi at conditions below pH 5.5, whereas other species of fungi 
and bacteria dominate at neutral to alkaline pH. The change in microbial functional 
groups, as environmental habitats change in soil pH and chemistry, may also limit the 
activity of the remaining members. Hendrickson et al (1985) reported a significant pH 
increase (4.7 to 5.2) in the forest floor in a mixed-wood stand, after whole-tree 
harvesting. Foster and Morrison (1987) showed that forest removal using the fiill-tree 
method resulted in an acidifying effect of the forest soil and suggested that the 
incorporation of forest phytomass into the soil would have a neutralizing effect, thereby 
reducing the limiting effect of pH on microbial organisms. 
17 
Fungi tend to dominate the microbial community in acid forest soils (Anderson 
and Domsch 1975; Bewley and Parkinson 1985; Scheu and Parkinson 1994; Matthies et 
al 1997). Matthies etal (1997) reported that culturable fungi predominated over 
bacteria at a pH range of 2.2-6.5. Not surprisingly, they found that the bacterial 
populations in acidic forest soils were more tolerant of the ambient conditions than were 
bacteria from less acidic forest soils. It has also been reported that bacteria have 
predominated over fungi in acidic Boreal forest soils (Frostegard et al 1993; Baath et al 
1995; Berg et al 1998). After many pH-raising treatments, Baath et al (1995) 
concluded that the fungal-to-bacterial biomass ratio remained fairly constant across a 
range of pH, suggesting that fungi and bacteria may not have varying pH optima. 
Further, Berg et al (1998) noted that high N levels in soil may influence the shift of 
dominance from fungi to bacteria. They also hypothesized that high atmospheric N 
deposition may eventually lead to N saturation of the soil, which would impose a stress 
on fungal communities, and cause a decrease in their abundance and activity. 
METHODS OF DETERMINING MICROBIAL BIOMASS 
Many methods have been developed and applied in the estimation of microbial 
biomass and activity. These approaches are: 1) direct and 2) indirect. Direct methods 
involve assays and measurements of the actual microbial biomass (Hartmann et al 1997) 
Indirect methods estimate the size of the microbial biomass by measuring the metabolic 
activities of microbes (Hartmann et al 1997). Direct techniques include; 1) chloroform 
fiimigation incubation (CFI) (Jenkinson and Powlson 1976), and 2) chloroform 
18 
fumigation extraction (CFE) (Voroney et al. 1993). Indirect techniques include: 1) 
substrate induced respiration (SIR) (Sparling 1985), and 2) soil CO2 evolution methods 
(Edwards 1982). The soil fumigation methods, as described by Jenkinson and Powlson 
(1976), assume that: 
1. Carbon in dead organisms is more rapidly mineralized than that in living 
organisms. 
2. Fumigation leads to a complete kill. 
3. Death of organisms in the unfumigated soil is negligible compared with that in 
fumigated soil. 
4. The only effect of soil fumigation is to kill the microbial biomass. 
5. The fraction of dead biomass C mineralized over a given time period does not 
differ in different soils. 
This method was originally developed for soils with a water holding capacity of 50-55% 
fJenkinson and Powlson 1976T which mav be limitine and nroblematic for soils outside 
19 
West 1988). Furthermore, the question of a correct “control” has plagued the 
methodology (Voroney 1985). 
The CFE method, on the other hand, is reported to provide stable estimates of 
soil microbial biomass (Tate et al. 1988; Sparling and West 1988; Merckx etal. 1988; 
Vance et al. 1987). Chloroform fumigation extraction may be used on soils with low pH, 
high organic matter, and excessive water content (Inubushi et al. 1991). Further, CFE is 
not dependent on the physiological state of the soil microflora, suggesting that the 
dormant population may also be captured (Martens 1995). Feigl et al. (1995) concluded 
that the CFE method is more convenient and suitable for estimating both microbial C and 
N in the same extract, as compared to the CFI and SIR methods. Beck et al. (1997) also 
recommended CFE over CFI and SIR techniques with respect to forest soils. 
Martikainen and Palojarvi (1990) compared CFE to microscopic counting in ten forest 
soils with a range of pH (3.6-6.8) and organic C (2.6-36%), and concluded that CFE was 
better suited for all ten soils. 
The SIR technique utilizes the physiological respiration response of soil organisms 
to substrate amendment to provide an estimate of soil microbial biomass C (Sparling 
1985). Like the soil fumigation techniques, the SIR method also follows assumptions 
implicit in the estimation of microbial biomass. The SIR assumes that (Sparling 1985): 
1. The response of different organisms to the method is reasonably constant. 
2. The majority of the soil micro-biota will respond during the period of 
measurement. 
20 
3. Glucose is a suitable substrate to induce the maximal response of respiration. 
4. The contribution to microbial C from non-glucose metabolizing organisms is 
insignificant or consistently low. 
Unlike CFI, SIR can be applied to soils of low pH, and to leaf and forest floor 
materials (Sparling 1985). Also, very small soil samples can be analyzed and the relative 
contributions by bacteria and fungi to rhizosphere populations can be distinguished 
through the incorporation of inhibitors (Sparling 1985). A limitation is that SIR, like 
most other methods, requires calibration using another estimate of the microbial biomass 
(Feigl el al 1995). Furthermore, SIR relies on the active soil population showing a 
respiratory response within a few hours after the addition of substrate. Therefore, the 
dormant population will not be captured by the assay (Feigl et al. 1995). Feigl et al. 
(1995) found SIR to be inappropriate for acidic soils with high clay content, in 
comparison to CFI and CFE methods. Ocio and Brookes (1990) suggested that the SIR 
response per unit of microbial biomass C may not be constant throughout the whole 
biomass range. It has also been reported that if the microorganisms are actively growing, 
SIR will overestimate the microbial biomass C (Sparling 1985). 
Soil respiration, an indirect method, is defined as the sum total of all soil 
metabolic fiinctions in which carbon dioxide is produced (Singh and Gupta 1977). It 
includes microbial, microfaunal, mycorrhizal, rhizospheral, and root respiration (Weber 
1985). These components may be measured using dynamic or static procedures, both of 
which are forms of indirect sampling methods. Dynamic methods use an infrared gas 
analyzer (IRGA), whereby a sample of air of known composition is drawn over a known 
21 
area and the increase in CO2 concentration is measured (Schlentner and Van Cleve 1985). 
Static techniques use alkali absorbents, like soda-lime, whereby an air-tight chamber 
covered with aluminum foil is inverted over an open tin can containing previously dried 
and weighed absorbent. After a measured period of time, the absorbent is removed and 
the amount of CO2 absorbed is calculated (Edwards 1982). Pongracic et al. (1997) 
reported that estimates of soil CO2 efflux using an IRGA were consistently greater than 
those calculated using soda-lime. However, explanations for the discrepancy between the 
two methods were not given. 
The use and interpretation of soil respiration has historically been complicated by 
the difficulty of separating root respiration from microbial respiration. Attempts by 
various researchers, such as Singh and Gupta (1977) and Schlentner and Van Cleve 
(1985), have not resolved this problem. However, soil respiration remains a widely used 
method for assessing biological and metabolic activity (Reiners 1968; Schlentner and Van 
Cleve 1985; Weber 1985; Gordon et al. 1987; Weber 1990; Pongracic etal 1997). As 
Weber (1990) stated, soil respiration measurements help to assess the metabolic activity 
of a site in relation to forest practices, thereby determining the degree of impact imposed 
by these practices on site productivity and recovery rates of ecosystem processes. 
The usefulness of respiration measurements may be limited because of the 
occurrence of soil atmosphere alterations during sampling. This may affect the level of 
microbial activity (Prosser 1997) in that the gaseous phase in which the soil normally 
exists may be altered if there is a passage of gas over or through the soil. This could 
increase the mixing of the gases and ultimately change the concentration of the O2 and 
22 
CO2 , thereby resulting in a false measurement of microbial activity (Prosser 1997). 
Prosser (1997) suggested that since the microbial biomass is quite sensitive to the 
chemical and physical environment, and because the forest soil environment is 
considerably heterogeneous, it is difficult to measure microbial activity without error or 
bias. 
Reiners (1968) speculated that soil respiration measurements did not capture the 
transfer of carbon compounds other than CO2. While these losses may cause an under- 
estimation of the rate of energy release, the inclusion of tree root respiration may cause 
an over-estimation of activity rates. Schlentner and Van Cleve (1985), however, reported 
that static methods, such as soda-lime, are feasible approaches for estimating total soil 
respiration at remote field locations, such as forest sites. 
MICROORGANISM CULTURE AND IDENTIFICATION 
Even though the fungal biomass exceeds that of bacteria, the latter will be the 
focus of this aspect of the study because bacteria act as excellent indicators of 
physicochemical conditions in the soil (Killham 1994) exhibiting higher sensitivity, over 
fungi, to changes in environmental parameters. Berg et al. (1998) reported that fungi 
were less susceptible to drying than bacteria, suggesting that spatial distribution patterns 
of bacteria may be predicted or determined by soil moisture. In addition, bacterial 
growth and activity are strictly limited by substrate quality (Killham 1994), and as such 
are indicators of how organic matter biomass removal influences the microbial biomass as 
a whole. 
23 
Microbial diversity encompasses a large array of taxonomic, physiological, and 
genetic characteristics, as well as the diversity of functional groups (Staddon et al. 1996). 
Difficulties occur in describing these attributes because current cultural methods for 
bacteria isolate no more than a small fraction of species (Staddon et al 1996). This 
limitation is due to a variety of factors. For example, Staddon et al (1996) described the 
effect of the association of r-strategist microbes (which grow well in vitro) with the K- 
strategist microbes (which are non-culturable) occurring in the same habitat. Drawbacks 
may also lie with the type of species of bacteria being cultured. For example, nitrifying 
bacteria are quite sensitive to techniques such as direct plating because organic materials 
introduced with the inoculum permit growth of heterotrophic contaminants (Schmidt and 
Belser 1982). Initial isolation of nitrifiers is difficult and, once isolated, these bacteria are 
slow growing in culture, sparse in yield, and susceptible to contamination (Schmidt and 
Belser 1982). Furthermore, bacteria occur in patches, which may be only a few cubic 
micrometres in volume, throughout soil (Coleman and Crossley 1996). Because bacteria 
are passive, they depend on episodic events such as rainfall or root growth for movement. 
Thus, bacterial distribution and abundance are difficult to estimate without a high 
variance about the mean (Coleman and Crossley 1996). 
In order to isolate highly oxygen-sensitive anaerobic populations, specific 
cultivation techniques are required to avoid exposing the microbes to oxygen (Casida 
1968). Bacteria which are parasitic on other bacteria may be present in the soil, but may 
not be demonstrated in isolations due to the lack of the presence of a suitable host 
(Casida 1968). Casida (1968) also suggested that much of the soil bacterial population 
24 
may exist in such a manner that antibiotics or other inhibitors in the soil stabilize their 
growth. Alexander (1961) advised that since the soil atmosphere contains such high 
levels of CO2, the CO2 level in the laboratory should be adjusted to reflect that found in 
the soil so that growth of soil microorganisms may occur under more natural conditions. 
For these reasons, cultural methods tend to be biased in that only a minor fraction 
of the bacterial population may be available for characterization in pure cultures (Bakken 
1997). Thus, the cultured isolates may or may not be representative of the bacterial 
species inhabiting the soil. Due to the diversity and variability of microbial communities 
found in the soil, classic taxonomic methodology does not always yield a clear 
identification (Bakken 1997). Sorheim et al (1989) compared cultures of microbes 
growing on various nutrient media, and found only a partial overlap between populations 
growing on similar media. The Gram staining technique is used to identify between gram 
positive and gram negative bacteria based on cell wall characteristics. Gram negative 
bacteria belong to the family Enterobacteriaceae, which includes most bacteria that occur 
in the soil (Holt et al. 1994). The traditional method used in the identification of Gram 
negative bacteria is C substrate utilization (Palmieri et al. 1988). Such C sources may 
include glucose, xylose, mannitol, lactose, sucrose, maltose, fructose, galactose, 
mannose, rhamose, lysine, ornithine, arginine, phenylalanine, esculin, and gelatin (Palmieri 
et al. 1988). These bacteria will yield either a positive or a negative result in a pattern 
characteristic of a particular bacterial species. However, this approach may be 
impractical for assessments on diverse populations due to it being very labour intensive, 
time consuming, and expensive. 
25 
Modern automated systems, which are much less labour intensive and less 
expensive, were introduced originally and developed primarily for the identification of 
clinical isolates (Palmieri et al 1988). Many of these commercial test systems have been 
modified to suit other situations, such as the identification of microorganisms isolated 
fi-om soil. For example, the Biolog system has been used to determine activity patterns 
for assessing functional diversity of soil microorganisms (Zak et al. 1994). But, this 
system has limitations in that reactions are sensitive to inoculum densities, the selection of 
C sources is biased to clinical isolates, and it is unable to determine fungal activity (Zak et 
al 1994). 
Washington et al. (1971) evaluated the accuracy of another multitest 
micromethod system, Analytab, which is also used for the identification of soil 
Enterobacteriaceae. They found this system to be about 93% accurate after repeat 
testing with a heavier inoculum of those strains failing to ferment glucose initially 
(Washington et al 1971). At the time, this system was the most complete commercially 
available test series for Enterobacteriaceae identification that provided an initial testing 
accuracy of 90%. 
A study conducted by Robertson et al (1976), revealed that identification kits, 
such as API 20E, have generally demonstrated satisfactory performance identifying 
clinical isolates when compared with traditional culturing methods. The authors 
concluded that the identification kits offered savings in both time and material costs, 
while allowing about the same rate of accuracy (exact percentage not stated). 
Furthermore, the study concluded that the kits offered, 1) improved quality control; 2) a 
26 
standardization of methods, which allowed the use of interpretative pattern directories; 
and 3) the application of these tests by less sophisticated operators (Robertson et al. 
1976). 
Palmieri et al (1988), however, concluded that the API 20E system had 
limitations in the identification of various bacterial species in clinical trials. They stated 
that the API 20E system has deficient characterization of some test organisms, especially 
at the species level. This identification system was unable to distinguish, for example, 
between various Pseudomonas spp., and simply grouped them under the category of 
other Pseudomonas species (Palmieri et al 1988). It was suggested that the API 20E 
system may be useful as a rapid screening method for preliminary characterization of 
various bacteria groups. The API 20E system has been employed successfully in the 
characterization and identification of bacteria isolates in other forest ecology studies, such 
as Mireku (1981) and Roy (1984). Nevertheless, whether traditional technique, or kit 
systems are used, there is no one technique ensures an accurate representation of all soil 
bacteria. 
More recent techniques for assessing soil microbial biomass include phospholipid 
fatty acid (PLEA) profiles (Baath et al 1995), fatty acid methyl ester (FAME) assays 
(Bailey et al 1997), and deoxyribonucleic acid (DNA) extractions (Liesack etal 1997). 
These methods are often used to determine the microbial population and community 
structure and activity via chemical means. The PLFA profiles use phospholipids, found in 
membranes of all living cells, as biomarkers (Morgan and Winstanley 1997). 
Phospholipids, comprising a constant proportion of the bacterial biomass, emit different 
27 
patterns for different subsets of microbial communities (Baath et al 1995). This method 
is applicable to the study of mixed populations of varying degrees and complexity 
(Morgan and Winstanley 1997). The PLFA profiles can also be used to detect 
environmental changes, stress responses or periods of activity by tracking differences in 
lipid profiles of microbes (Morgan and Winstanley 1997). 
The FAME assays are similar to PLFA's in that fatty acids are used to identify 
bacteria species rather than to estimate biomass (Bailey et al. 1997). The profiles are 
generated using extracted cellular fatty acid methyl esters that are assayed via gas/liquid 
chromatography (Bailey et al 1997). A third technique uses DNA extraction. This 
entails direct cell lysis once separated from the soil matrix. The method retrieves up to 
35% of the microbial biomass from the soil (Liesack et al 1997). This technique, 
however, only selects for those microbes that have less affinity for soil particles (Liesack 
et al 1997). Direct cell lysis, where the cells are first lysed in the soil then the DNA is 
extracted and purified, detects the overall genetic potential of the sample (Liesack et al 
1997). However, the method of lysis chosen is critical since this may influence the 
microbial biomass potential detected in the soil (Liesack et al 1997). 
Microbial diversity is not only demonstrated in the number of species contained 
within a community and in the number of functions they provide, but also in the number 
of rare habitats that they occupy (Staddon et al 1996). Therefore, as Staddon et al 
(1996) proposed, there may be a vast number of habitats yet unexplored, containing new 
species still unknown. Unless current culturing techniques are modified to include 
28 
methods that allow for these unknown species to be sampled, a clear understanding of 
microbial biodiversity may not be possible. 
29 
METHODOLOGY 
SAMPLE SITE 
The study site for the microbial biomass investigation is located about 80 km 
north of Thunder Bay, ON, to the west of highway 527 (Figure 1). The sample site 
consists of a 110 year old black spruce (Picea mariana [Mill.] B.S.P.) stand. The 
topography consists of gentle rolling hills, based on bedrock formed by morainal parent 
material. The site represents an ES20, spruce/pine feathermoss ecosite (Racey et al. 
1996). The soil is classified as gleyed dystric brunisol with a very fresh moisture regime 
(3) (Sims et al 1989). A description of the soil profile is provided in Table 1. 
Harvested in the winter of 1994, four different biomass removal treatments were 
applied (Gordon et al 1993), which included: 
1. Tree-length harvesting system, where the slash is left near the stump. 
2. Full-tree harvesting system, where the slash is removed to the roadside. 
3. Whole-tree harvesting system, where the slash, duff, and stumps are removed to 
roadside. 
4. Full-tree chipping system, where slash is removed to roadside, chipped and 
redistributed on cut blocks 
5. Control, where the block remains uncut. 
Only the tree-length, full-tree, and control treatments were considered for the purposes of 
the study reported here. 
Figure 1. Map of northern Ontario identifying the black spruce study site location. (Not drawn to scale) 
31 
Table 1. Physical soil characteristics and classification of soil profiles associated with the 
black spruce study site. 
Particle Distribution (%) 
Horizon Thickness Soil Soil Soil 
(cm) Texture Moisture Classification 
sand silt clay 
F 7 n/a n/a n/a 
H Discontinuous Very Gleyed Dystric 
(<1 cm) Fresh (3) Brunisol 
Bm 15 Silty loam 42 52 6 
Bcgj 1 Sandy loam 46 49 5 
Note: Coarse fragment content represented slightly over 32%, by volume, of the mineral soil profile. 
32 
FIELD SAMPLING TECHNIQUES 
Figure 2 demonstrates the layout of the treatment plots found on the study site. 
Originally, each plot (30mx30m) was allocated a number from 1 to 12, and each plot 
was randomly allocated 1 of 4 different harvesting treatments, as specified earlier. The 
control plots (50 m x 50 m) were left untreated. Sampling for this study occurred once a 
month (at the end of the month), for four consecutive months. May, June, July, and 
August, in 1998. 
Figure 3 presents the soil temperature and precipitation values for May through 
September of 1998. The temperature is averaged weekly, and the precipitation is event- 
based. Soil temperature, at a depth of 15 cm, was measured in the treatment plots. 
Precipitation was measured at a centrally located (on site) weather station. During the 
measurement period, the full-tree treatment displays consistently higher soil temperatures 
than do the tree-length or control treatments, with the control having the lowest 
temperatures. During the month of May, only 30 mm of rainfall was recorded. However, 
soil moisture content at this time was augmented by spring thaw and snow melt. June and 
August were relatively dry with only 42 mm and 55 mm of rain, respectively. July 
experienced the most amount of rain during the year of sampling (103 mm), with 
September following closely behind at 90 mm. 
Samples were taken from nine plots, representing three each of the tree-length, 
full-tree and control treatments. These two harvest treatments were chosen because they 
are the most commonly used in the forest industry and represent a range of biomass 
removal options. 
Fieure 2. Map of the black spruce study site and treatment plot location. Plots sampled are marked as C (control, only 1 to 3), TL (tree-length, 4,8, 12), and 
FT (full-tree, 1,5,10). The respiration experiment demonstrates the random allocation of the soda-lime tins on the plo s. 
* Sampling date 
  Control 
 Tree-length 
15 20 25 30 35 40 
Time of the year (weeks) 
Figure 3. Average temperature and cumulative precipitation values associated with 
the black spruce study site for 1998. 
Soil Temperature (°C) 
Cumulative Precipitation (mm) 
35 
Three sample points were obtained at random locations, in each plot. At each sample 
point, the litter (feather moss) layer was peeled back and two soil samples were taken 
using a trowel. One sample was taken from the upper organic layer, the F/H horizon, and 
the second sample was taken from the upper mineral layer, the Bm horizon. Samples from 
each plot were bulked for each of the organic and mineral layers, so that there were nine 
organic and nine mineral samples taken each month. The samples were bulked in order to 
minimize the variability present within each plot. High spatial heterogeneity of chemical 
and physical properties of both the forest floor and soils has been recognized (Arp and 
Krause 1984) and bulking of samples is an attempt to reduce this variability. The samples 
were sealed in plastic bags and placed in frozen storage (-15°C) pending analysis. 
LABORATORY ANALYSIS 
Soil pH 
The pH of each soil sample was measured following procedures outlined by 
Forster (1995a). The distilled water method was used in order to conserve soil which was 
dried and reused in a different test. Ten grams of air-dried and sieved (2mm) mineral soil 
was weighed into a glass beaker with 25 mL of distilled water (1:2.5 v/v). Five grams of 
organic soil was weighed into a glass beaker with 50 mL of distilled water added to the 
beaker (1:10 v/v). Each sample was stirred immediately for one minute and then allowed 
to stand for 30 minutes. After a second short stirring, the pH was measured with a pH 
meter (Orion 520A) and a glass electrode (Orion 91-57). 
36 
Organic Matter Content 
The loss-on-ignition method described by Meyer and Vanson (1997) was used to 
determine the organic matter content of the soil sampled. The organic matter % of each 
sample was calculated as follows (Meyer and Vanson 1997); 1) In a crucible of known 
weight oven dried at 110“C, a 10 g sample of 2mm sieved soil was placed; 2) the sample 
was placed into a muffle furnace and the temperature was gradually raised to 600°C with 
the sample ignited for three hours; 3) samples were cooled in a desiccator, and reweighed 
to determine weight loss; 4) the organic matter content was calculated using the following 
formula: 
organic matter content (%) = [(soil weight before - soil weight after) / soil weight before] * 100 Eq. 1 
Water Content 
The gravimetric water content was measured following Forster (1995b). The 
procedure included weighing out 10 g of field-moist soil for each sample into an 
aluminium weighing tin, and the weight was recorded to the nearest O.OOlg. The tins of 
soil were placed into a drying oven at 105°C for 24 hours. Upon removal, the tins were 
cooled in a desiccator, then reweighed. The per cent water content was calculated using 
the following formula: 
% water content = [(moist weight - diy weight) / moist weight] x 100 Eq.2 
37 
Total Nitrogen and Phosphorous 
A modified Micro-Kjedahl method was used to measure the total N and P in the 
soil samples. Original methodology follows steps outlined by Bremner and Mulvaney 
(1982). Soil was air dried at 65°C for 48 hours, ground by grinder, and sieved through a 
2mm screen. The soil was placed in a container with an air-tight lid and labelled. For 
digestion, 1.0 g of soil was weighed and placed into a digestion tube. Two replications 
per sample were weighed out. One blank was included at the beginning and at the end of 
the digestion rack. Two or three boiling chips were added to each tube including the 
blanks and controls. Four mL of digestion solution (mixture of H2SO4 and salicylic acid) 
was added, and the tube was turned to wash down sides, then shaken gently. A rubber 
stopper was placed on all digestion tubes which were then left over night. 
Stoppers were removed under the fumehood and one level scoop of crushed 
sodium thiosulphate was added to each tube. The digestion tubes were placed on a rack in 
the digester and placed along the sides were side plates. The digester was set to 350“C. 
After the sample tubes finished frothing, they were removed from the digester to a 
wooden stand and were allowed to cool for 5 minutes. One scoop of catalyst mixture 
(potassium sulfate, cupric sulfate, and selenium) was added to each tube and the tubes 
were returned to the digester at 400°C for 1 hour. Tubes were then removed from the 
digester and cooled for 10 minutes. 
When cooled, 10 mL of distilled water was added. The samples were placed into 
the sonicator for several hours until the digest was dissolved. The rubber stoppers were 
removed from the tubes and distilled water was added to each sample until about 1 cm 
38 
from the top of the tube. The tubes were rubber stoppered again and inverted gently once 
or twice to mix the digest with the distilled water. An Autoanalyzer was used to measure 
the total nitrogen and phosphorus. Standards were used to establish a linear relationship 
for each element. Actual readings for each sample were then transformed using the line 
generated for the standards. This value, in ppm, could be calculated as a per cent using 
the following formula: 
X = (y - a) / b, where a and b are calculated from a regression based on a series of standards; 
y = reading in ppm from the Autoanalyzer; 
X = ppm of nitrogen. 
% N or P = [(ppm N (or P) X 0.075 L ) / weight of sample (mg)] x 100 Eq.3 
Values were multiplied through by 1000 to reduce the number of decimal places, 
allowing for easier reporting and reading of figures. 
Culturing 
Culturing of bacteria followed methods outlined in Phillips et al. (1986). Twenty- 
five grams of fresh soil was placed in a 1 L graduated cylinder and distilled water was 
added so that the total volume was 250 mL. The suspension was then stirred and poured 
into a 1 L Erlenmeyr flask and shaken for 30 minutes. Using a sterilized pipette, 10 mL of 
the shaken suspension was drawn and transferred into a 90 mL sterile water blank. One 
mL of this diluted suspension was immediately transferred through successive 9 mL sterile 
water blanks, totalling six separate suspensions, each at various dilutions starting at 1:10 
and ending with 1:1 000 000. Each dilution was shaken for a few seconds and was kept in 
motion while a sample was drawn into the pipette. Using the Eppendorf tip-ejector 
pipette 4700, 0.5 mL of each dilution was transferred aseptically to previously prepared 
39 
plates of Tryptic Soy agar medium (Anonymous 1996). Each plate was sealed with 
parafiRn wax strips and incubated at 24”C for 48 hours. The bacteria colonies were then 
counted and described. Representative cultures were transferred to a separate plate of 
Tryptic Soy agar using the streak-plate method. The inoculated plates were incubated 
again at 24”C for an additional 36-48 hours. 
Identification 
Pure cultures were tested for Gram positive and Gram negative properties 
following the staining procedures outlined in Anonymous (1992). The identification of the 
isolated bacteria was done using a carbon substrate utilization system called API 20E 
(bioMerieux Vitek 1996). This system involved the following steps: 1) preparation of the 
inoculum, where a sterile loop of culture was mixed into a 5% saline solution; 2) 
preparation of the strips, where the holder was filled with sterile water and labelled; 3) 
inoculation of the strips, where each well was filled with the inoculation using a 5 mL 
pipette, and anaerobic tests were sealed with mineral oil; 4) incubation of the strips, where 
the strips were put into a dark chamber at 36 '’C for 24 to 48 hours; 5) reading of the 
strips, where each well was identified by colour and designated a set number for that 
colour, leading to a nine digit number code; and 6) identification of microorganisms, 
where the number code is registered in a profile index, and each code is linked with a 
bacterial species {i.e. 0040024-10 Pasteurella multocidd) and identified on a scale of poor 
to excellent probability. 
40 
Soil Respiration 
Two respiration measurements were taken approximately one week apart, in 
September, using the soda-lime technique as described by Edwards (1982). One plot for 
each treatment was randomly selected. Each plot was further sub-divided into three 
sections (Figure 2); five tins of soda-lime were randomly placed on each sub-division (15 
tins of soda-lime on each plot), plus one blank per plot. Within each pre-weighed tin, 35- 
40 grams of soda-lime was placed. The tins each have a diameter of 6.5 cm and a surface 
area of 33.18 cm^, thus meeting the requirement for a minimum surface area (Edwards 
1982). The tins of soda-lime were dried in an oven for 24 hours at 100°C and weighed 
before being placed on the sample sites. The tins were placed on the sites uncovered, and 
then a plastic tub (158.36 cm^) covered in aluminium foil was placed over each tin. The 
foil helped to reflect the sunlight so that photo synthetic respiration was not falsely 
elevated. Also, the tins were placed such that a minimal amount of green material was in 
the area covered by the tub. The tins were allowed to stand for 24 hours, retrieved and 
capped. The blanks of soda-lime remained covered for this entire period. The soda-lime 
was again oven-dried for 24 hours at 100“C and reweighed to measure the amount of 
carbon dioxide evolved. Calculations followed the formula: 
CO2 evolved = (weight of si. at time 2 - weight of si. at time 1 )/(§ hours * area m^) Eq.4 
Chloroform Fumigation-Extraction 
The method for determining microbial biomass carbon and nitrogen is described by 
Voroney et al (1993). Ten grams of wet soil were oven-dried at 100°C for 24 hours to 
41 
calculate the moisture content. Twenty-five grams of fresh soil was weighed into a glass 
jar; four replicates per soil sample. Two replicates were fumigated with chloroform 
(CHCI3) for 24 hours. The other two replicates of each set of four were immediately 
saturated with either 60 mL (organic samples) or 40 mL (mineral samples) of 0.5 M 
K2SO4. The replicates were shaken for one hour to allow for complete mixing of soil and 
K2SO4. The mixture was filtered through VWR glass fibre filter papers (grade. 696), and 
the extracts were frozen at -10°C in plastic vials for later analysis. The fumigated samples, 
after repeated evacuation of the chloroform, were subject to the same treatment. 
Biomass N and Biomass C 
The extracts were analyzed for total N and C, using a colorimetric method 
(Anonymous 1978; 1984). The N containing compounds in the soil extracts were 
oxidized to nitrate by digestion in acidic and basic conditions with ultraviolet light. In 
order to avoid a suppression of N by C in the sample, the concentration of the potassium 
persulfate solution was increased from 1 to 3% (Luckai et al in prep.). The nitrate was 
further reduced to nitrite by copper-hydrazine solution. The nitrite ion then reacted with 
sulfanilamide under acidic conditions to form a diazo compound, which then was coupled 
with N-(l-naphthyl)-ethylenediamine to form a reddish-purple colour. The intensity of the 
colour was measured at 520 nm. 
In order to measure the total organic C, inorganic C (carbonate) was removed by 
entraining the acidified stream with a high velocity stream of N or C free air. The sample 
was transformed into a thin turbulent liquid film that was transported rapidly through a 
42 
large bore coil providing the necessary surface area for efficient CO2 removal. At a purge 
rate of 500 mL per minute, up to 500 mg of inorganic C can be removed with minimal loss 
of volatiles. An aliquot of the carbonate free sample was air segmented, mixed with a 
stream of acid and potassium persulfate (4%) and subjected to UV radiation. The 
resultant CO2 was dialyzed through a silicone rubber membrane and reacted with a weakly 
buffered phenolphthalein indicator. The decrease in colour of the indicator was 
proportional to the original C concentration. Calculations for biomass C and N followed 
the formulae outlined in Voroney et al. (1993); 
1. Soil water content: 
WS (%) = [ soil wet weight (g) - soil oven-dry weight (g) / soil oven-dry weight (g)]*100 Eq.5 
2. Weight of soil sample (oven-dry weight equivalent) taken for microbial biomass measurements (MS): 
MS (g) = [soil wet weight (g) * 100] / [100 + WS (%)] Eq.6 
3. Total volume of solution in the extracted soil (VS): 
VS (mL) = soil wet weight (g) - soil oven-diy weight (g) + extractant volume (mL) Eq.7 
4. Total weight of extractable C and N in fumigated (Op) and unfiimigated (Oup) soil samples: 
OCp, OCuF (g/ g soil) = extractable C (g/ mL) * [VS (mL) / MS (g)] Eq.8 
ONp, ONup (g/ g soil) = extractable N (g/ mL) * [VS (mL) / MS (g)] Eq.9 
5. Microbial biomass C and N in the soil (MB-C, MB-N): 
a) MB-C (g/ g soil) = ( OCp - OCyp) / kpc Eq.lO 
where: kpc = 0.25 ± 0.05 and represents the efficiency of extraction of microbial biomass C. 
b) MB-N (g/ g soil) = ( ONp - ONup) / kpN Eq.l 1 
where: kpN = 0.18 ± 0.04 and represents the efficiency of extraction of microbial biomass N. 
43 
STATISTICAL ANALYSIS 
Overview 
The variables of interest in this study which are subject to statistical analysis are 
microbial biomass carbon (MBc) and nitrogen (MBN), soil organic matter (OM), soil total 
nitrogen (N), soil total phosphorus (P), soil pH (pH), soil water content (WC), and soil 
respiration. The terminology and approach to the statistical analysis follows that of Zar 
(1996). Full data-sets can be found in Appendices I and II. The objective of the statistical 
analysis was to determine if the independent or fixed variables {i.e. harvesting 
treatment and time of sampling) affected the dependant or response 
variables. In all cases “a”, or the probability of committing a Type I error, 
was set at 0.05. 
The data were investigated carefully for outliers. Side-by-side dot plots were 
constructed to aid in the identification of any outliers. However, after a thorough search 
all datum points were left in the analysis because there was no just cause for elimination. 
For example, the pH value of one of the control plots in the organic soil horizon was 
measured at 3.12, which was much lower than all the other points measured. However, a 
pH of 3.12 is a reasonable value on a control plot that has a high input of acidic foliage. 
Therefore, this datum, as well as others like it, remain in the data sets. 
In the case of the respiration data, an analysis of variance (ANOVA) was based on 
a two-way factorial, completely randomized design (see subsequent section for details). 
One-way analysis of variance (ANOVA) was employed to determine treatment effects, if 
any, on the soil descriptors (OM, N, P, pH, and WC). . When appropriate, the LSD Post 
44 
Hoc (Zar 1996) test for differences between treatment means was applied where the main 
effect (biomass removal) was found to be significant, and bar charts were constructed to 
demonstrate any differences. In an initial survey for variable response, MBc and MBN 
were also subject to two-way ANOVA; line graphs of MBc and MBN over time were used 
to identify trends and shed light on interaction effects. 
As stated in the Literature Review, soil characteristics, such as pH, OM content 
and nutrient levels, are reported to 1) change in response to disturbance events and 2) 
appear to affect the structure and function of soil microbial communities. The 
investigative approach taken in this study therefore included a number of these variables in 
anticipation that the application of covariance analysis might assist in the interpretation of 
results through either 1) adjusting for sources of bias on the response variable or 2) 
throwing light on the nature of treatment effects in randomized experiments (Snedecor and 
Cochran 1967). 
In order to explore and identify relationships between MBc and MBN, and the soil 
descriptors, scatter plots were constructed and Pearson and Spearman correlation 
coefficients were computed. Both correlation tests compute a value between +1 and -1. 
The closer the value is to either extreme, the more highly correlated are the two 
parameters. The Spearman correlation differs from the Pearson in that it uses ranked data 
rather than absolute numbers. This reduces the effect of extreme data points (Zar 1996). 
Given that some soil descriptors exhibited correlation with either or both MBc and MBN, 
the investigation of the data was then conducted using analysis of covariance (ANCOVA). 
Microbial biomass data are treated as a split-plot design with covariates, with the mineral 
45 
and organic layers separate. One difficulty associated with this approach and this data set 
is the resultant limited number of degrees of freedom, particularly when cases with unfilled 
cells were excluded from the ANCOVA. 
All data were analyzed using DataDesk 6.0 software (Velleman 1997). Tests of 
normal variance were run on the data and they met a normal variance of distribution of 
means, as assumed by the Central Limit Theorem (Zar 1996). Expected mean square 
(EMS) tables were constructed to determine the appropriate tests for the null hypotheses, 
as well as to confirm the significance of the various response variables measured by the 
analysis of variance. Bar charts present treatment means and illustrate the results of the 
LSD Post Hoc tests. 
46 
Soil Respiration 
As stated previously, the soil respiration data were subjected to a two-way analysis 
of variance with a completely randomized design (CRD). This design may be represented 
by the following equation; 
Yijki = p+ Ti + Dj + TDij + E(ij)k Eq 12. 
i= 1,2,3; j = l,2; k=l,2,...,15 
where 
Yijk = the carbon evolution of the tin of the j* sample date within the i^ treatment, 
p = the overall mean. 
Ti = the fixed effect of the i^ treatment. 
Dj = the fixed effect of the sample date. 
TDij = the fixed interaction effect of the i^ treatment with the j* sample date. 
C(ij)k = the random effect of the k^ tin in the ij* treatment combination, 
assumed IID N(0, o^). 
The treatment (T) factor represents three levels of biomass removal: control, tree-length, 
and full-tree. The sample date (D) is represented by two different sample dates on which 
the respired carbon was measured at different locations within the selected plots. 
The expected means square (EMS) table (Table 2) will help to determine the test statistic 
for the following null hypotheses; 
i. Ho; (T) = 0; 
ii. Ho:(D) = 0; 
iii. Ho: (TD) = 0. 
47 
Table 2. EMS table associated with Eq 12. 
3 2 15 
F F R 
Source i j k df EMS 
T, 0 2 15 2 o^ + 30(T) 
D, 3 0 15 1 o2 + 45 (D) 
TD.J 0 0 15 2 a2 + 15 (TD) 
84 o2 
Microbial Biomass 
The analyses of the microbial biomass carbon and nitrogen in the organic and 
mineral soil horizons are represented by a two-factor analysis of covariance 
(ANCOVA) with the possible inclusion of up to five covariate variables 
(note: covariates were included only if the Pearson correlations were 
significant at a = 0.05). This ANCOVA followed a split-plot design, whereby 
the harvest treatments were allocated to the main plots with 3 replicates and the month 
factor was allocated to the subplots. This experiment may be represented by the following 
linear model: 
Yijki - p+ Ti+ Q(i)j + + Mk + TMik+ coM(i)jk + C(ijk) Eq 13. 
where 
Yijk = the microbial biomass of the soil sample for the month in the j* plot within the 
• th 
1 treatment. 
= the overall mean. 
48 
Ti = the fixed effect of the harvest treatment. 
0(i) = the random effect of the j* plot within the i* harvest treatment. The 0(i)j's 
are assumed to be IID N (0, o^). 
??6(ij) = the restriction error due to the restriction on the randomization of the 4 months 
within the plot within the i^ harvest treatment. The 6(y)'s are assumed to 
be IID N (0, o^). 
Mk = the fixed effect of the month. 
TMik = the interaction effect of the i^^ harvest treatment with the month. 
coM(i)jk= the interaction effect of the k‘‘'month with the plot within the i^ harvest 
treatment. 
C(ijk) = due to bulking, there is no between sample variability of the k* month in the j* 
plot within the i^** treatment. 
The month (M) factor represents four points in time: May, June, July, and August. 
The treatment (T) factor is described by three levels of biomass removal: control, tree- 
length, and full-tree. Also, theoretically up to five covariates (pH, N, P, OM, and WC) 
could be included depending on the strength of their relationship to the dependent variable 
(measured with the Pearson Correlation test). The expected means square (EMS) table 
(Table 3) is used to determine the test statistic for the following null hypotheses: 
i. Ho: (T) = 0; 
ii. Ho.- (M)=^0; 
iii. Ho: (TM) = 0. 
49 
The table would include the covariate factors where appropriate. 
Table 3. EMS table associated with Eq. 13. 
F F R F 
Source 1 J k 1 df EMS 
T. 10 3 4 0^ + 4 8^+ 4 12 cp (T) 
"(i)i 113 4 0^ + 4 6^+ 4 
11 14 <j2 + 4 5^ 
Mk 13 3 0 3 + COM^ + 9 9 (M) 
TMac 10 3 0 6 + 3 <p (TM) 
2 2 
G)M(i)jk 11 10 18* O + G)M 
^qjk) 1111 0 
*The degrees of freedom will vary depending on the number of covariate factors included in the analysis. 
50 
RESULTS 
BACTERIOLOGY 
The objective of the bacterial culturing was to determine if increasing levels of 
biomass removal would result in differences in the bacterial species isolated from the 
treated plots. Throughout the sampling season, the amount of cultures produced on the 
plates of solid media varied. Quantitatively speaking, in May a very high number of 
bacterial cultures were produced in each of the six dilutions up to 1:1 000 000. As the 
season of study progressed, the number of cultures gradually declined, and in August only 
a few cultures were produced in each of the first three dilutions, up to 1:1000. 
Five different bacterial species were isolated from the soil collected from the 
sample sites; Chryseomonas luteola, Pseudomonas fluorescens, Aeromonas salmonicida, 
Serratia marcescens, and Syntrophomonas multifilia (Table 4). These species each 
occurred on all three treatment plots (control, tree-length, and full-tree), except for P. 
fluorescens which only occurred on the harvested plots. The number of colonies counted 
varied, however, a greater amount was generally observed from samples of the organic 
horizon compared to the mineral layer. Furthermore, species occurrence varied from 
month to month. For example, S. multifilia was found during June and July in the organic 
horizon, and during August in the mineral horizon. However, C. luteola was present 
throughout the entire season in both the organic and mineral horizons. 
T3 t3 
O O 
O O 
O O 
> > 
Q -iS Q T3 
O o
O o o
  O O o 
O -4 O O HJ 
H 
(U 
w 
2 H ^ H 3 H 33 H 33 
H 00 ... UH .3-r • PH -3P' 33 
< 2 "o O S-. I !-i 
a c
H O 
 ^c!  o r2 C go <U O C o> O 33 <15 
CJ ^ L) ^ U i3 U 03 
C^ C3^ ON O' >02 C3N 
r- o o o (N 04 ro <50 O O O O O 
rn ot <N —^ NO ro CNt O'! 
O 
CO 
O 
X 
< (N CO) o >02 O i/~) _ ^ oro  o -H <50 o 0^00 
O NO ^ ^ ' cn (N (N fNl O') 
O 
X 
H 
o cd C 00 '^0  3 23 ^ 33 < S < /=0  I—335  I—33>  < 
"cO  
g a 3J SP g 
S u S 102 ’o .. >02 ^ 3- 
>02 O CD >02 
I/^ o o o) .3 
.b O 23 Mg' CO C/5 
bfl C3 C/2 O bB O cd >—I o 00 
o § -s" — -3d SO - _33 O o 
H o B o o 
S T 
=S- gN 
& =L S 102 H2  2 s 
>02 .33 g 
:d- c fH s /2 O- .—I c/2 
u i-T 
rn 
c/T c/2 c3 c/2-' 42 c/2 c/i' CO 
W T3 C TD "23 P "O 
cc3 X3 
2 O :3 2 O 33 O Q c/2 X) NH O !-. X) <o (U O <D O > > •S; "w) ^ w) > 00 
■§ J cd ■'0§ 0> WD  .Z3 OJ0  -o
cc3 
-g  0CD0  '.igd  00 CD 
C >N s=l PH C3 ^ 33 cu 33 
s03  ^ti-0  S (U £ cu S CD" 
g 
cc3 |t3 cc3 X) 03 x3 cc3 
S-I 
o ^ O o o 'i O 
CO 0 
W 
o 133 , P« w S 
cx, 
CO 
u t>0 CP. 
Table 4. A description and summary of the bacteria species isolated from organic and mineral soil samples, Summer 1998. 
fliiorescens wide; beige, shiny, concentric circle pattern. June 0 0 Full-tree. 
52 
SOIL RESPIRATION 
Mean values for CO2 respired (g hr'^ m'^) at locations in the control, tree-length 
and full-tree plots are summarized in Table 5. The values recorded range from 0.0069 to 
0.0165 g hr'^ m'^. The control treatment had the highest mean for both sampling dates. 
Table 6 presents the results of the ANOVA for the respiration data. 
Table 5. Summary of the amount of carbon respired (g hr'^ m'^) from the three 
treatments associated with the black spruce study site. Summer 1998. 
C Efflux (Date 1) C Efflux (Date 2) 
Treatment mean max min mean max min 
Control 0.0123 0.0159 0.0094 0.0131 0.0165 0.0103 
Tree-Length 0.0109 0.0139 0.0069 0.0114 0.0137 0.0074 
Full-Tree 0.0101 0.0126 0.0069 0.0107 0.0133 0.0078 
Date 1= September 5, 1998; Date 2= September 12, 1998. 
Table 6. ANOVA table associated with CO2 evolved (g hr‘^ m'^). 
Sums of 
Source df Squares F-ratio Prob 
Treatment (T) 2 8.5 E-05 12.867 0.000* 
Sample Date (D) 1 9.3 E - 06 2.8188 0.097 
TD 2 2.7 E - 07 0.0412 0.959 
Error 84 2.8 E-04 
* Significant at a = 0.05. 
53 
The fixed effect of treatment (T) significantly affected the amount of carbon evolved. 
The raw data revealed a normal distribution. The LSD Post Hoc test revealed that the 
microbial activity (indexed by soil respiration) of plots in the uncut, mature stand differed 
significantly from those of the harvest treatments, even though soil temperatures were 
lower on the uncut areas (Figure 4). Neither the fixed effect of the sample date (D) nor 
the interaction (T x D) appeared to significantly affect the amount of carbon evolved. 
54 
September 5/1998 
September 12/1998 
Harvest Treatments 
Figure 4. Results of the LSD Post Hoc test demonstrating similar and dissimilar harvest 
treatments, using mean CO, efflux (g hr'm'") as the response variable. The 
letters on the bars represent which groups are similar to or different from each 
other. 
55 
SOIL DESCRIPTORS 
The soil descriptors measured in this experiment were pH, organic matter (OM), 
moisture content (WC), total P and total N. These parameters were measured in order to 
provide further insight into possible relationships with the microbial biomass. Mean 
treatment values within either the organic or mineral layers were similar (Table 7) 
however, one pattern did arise. Within the organic horizon, means for all descriptors 
except pH were highest for the tree-length treatment. With respect to pH, the range was 
from 3.12 (in a control sample) to 4.18 (in a full-tree sample). Ranges in the other 
parameters were relatively consistent from one treatment group to another with the 
exception of total N, where that of the full-tree treatment exceeded that of tree-length by a 
factor of 0.5 and that of the control by a factor of nearly three. 
Within the mineral layer, no single treatment group exhibited consistently high or 
low values and ranges were, once again, very comparable with the exception of total N in 
the control group. Interestingly, with regard to pH, the lowest and highest values 
mirrored those found in the organic layer. 
56 
Table 7. Summary of the soil parameters measured in the organic and mineral soil 
horizons, summer 1998. 
Soil Parameter Control Tree-Length Full-Tree 
mean max mm mean max mm mean max mm 
Organic Horizon 
pH 3.47 3.93 3.12 3.92 4.02 3.84 4.09 4.18 3.98 
Organic Matter (%) 82 95 69 84 91 77 78 93 59 
Moisture Content (%) 65 70 55 74 77 71 71 76 61 
Total Phosphorus (mg/kg) 151 161 144 159 181 150 149 161 133 
Total Nitrogen (mg/kg) nil 1138 1072 1178 1268 1139 1 152 1232 1057 
Mineral Horizon 
pH 3.73 4.34 3.38 3.83 4.18 3.7 4.04 4.26 3.8 
Organic Matter (%) 13 20 8 11 13 9 12 16 8 
Moisture Content (%) 31 42 20 36 38 34 34 40 28 
Total Phosphorus (mg/kg) 32 36 30 32 38 26 33 36 28 
Total Nitrogen (mg/kg) 167 225 118 135 141 121 136 142 121 
One-way analysis of variance, where the treatment factor was tested against each 
of the five soil descriptors (pH, P, N, WC, and OM), was employed to determine if there 
were any effects due to the increasing levels of biomass removal. Table 8 presents a 
summary of the results. Only pH in the organic layer was significantly affected. The 
LSD test revealed that the control treatment mean was significantly lower than those of 
the harvest treatments (Figure 5). The one-way ANOVA employed on the mineral data 
(Table 8) revealed no significant effects of the treatment factor on any of the soil 
descriptors. 
57 
Table 8. One-way ANOVA table associated with the harvest treatment factor tested 
against the environmental parameters in the organic and mineral soil horizons. 
Sums of 
Source df Squares F-ratio Prob. 
Organic Horizon 
pH Treatment 2 1.3 E-7 4.110 0.029* 
Error 25 3.9 E-7 
H itrogen Treatment 2 54520 .843 0.175 
Error 32 473222 
Phosphorus Treatment 2 788 0.662 0.52: 
Error 31 18462 
Moisture Content Treatment 2 509 2.908 0.069 
Error o o JO 2886 
Organic Matter Treatment 2 119 0.387 0.68: 
Error 24 3685 
Mineral Horizon 
pH Treatment 2 6.2 E-9 1.115 0.340 
Error 33 9.2 E-8 
Nitrogen Treatment 2 6058 .016 0.373 
Error 33 98399 
Phosphorus Treatment 2 1 1 0.112 0.895 
Error 33 1649 
Moisture Content Treatment 2 129 0.232 0.794 
Error 2o) Jn)  9146 
Organic Matter Treatment 2 34 0.606 0.552 
Error o o 919 
Significant at a= 0.05. 
58 
5 
X 
(X 
c 
CJ 
(O 
1 
0 
Harvest Treatments 
Figure 5. LSD Post Hoc test demonstrating similar and dissimilar harvest treatments in 
the organic soil horizon, using mean pH as the response variable. The letters on 
the bars represent which groups are similar to or different from each other. 
59 
MICROBIAL BIOMASS 
Variability in Original Observations 
Table 9 displays values for soil microbial biomass carbon and nitrogen (pg/g soil) 
in the organic and mineral soil horizons, respectively. The full data set can be found in 
Appendix I. A considerable numeric difference was observed between the microbial 
biomass carbon (MB^) values of the organic and mineral soil layers regardless of the 
month. Biomass values for the organic layer range in the thousands (pg/g soil), whereas 
those of the mineral layer range in the hundreds (pg/g soil). Microbial biomass nitrogen 
(MBN) follows a similar pattern, where the organic layer samples had much higher values 
than did those of the mineral layer. The C:N ratios (Table 9) are all generally within 
normal ranges for microbial biomass, that is 15:1 to 3:1 depending on the relative amount 
of fungi and bacteria (Paul and Clark 1989). 
Results of the analyses of variance for the microbial biomass C are presented in 
Tables 10 and 11 (organic and mineral horizons, respectively). The ANOVA for the MB^ 
in the organic layer revealed a significant treatment (T) factor, as well as the Treatment x 
Month interaction factor. In the mineral layer, the month and treatment main effects were 
found to significantly affect microbial biomass C. 
OO vq 
u vd 
p r~- VO O') ov (N 
bJ) PQ o VO 
< 
rn (N 0 VO 0 
Ov CN VO 01 rn 
OQ VO rn ol rn rn 01 
VO VO VO 
cli 
z VO Ov o OV VO O ov oo VO rn VO VO ov •o Ov rn rv) 
3 PQ ON (N rn (N rN (N 
ov O rn o 
<N o ov (N VO O O 
PP oo o VO VO VO rn 
VO r^i 
:z; vq rvi vq ov 
Li >o vd oo 
z >o rn 
PQ ov oi ov 
ov o o CN 
oo (N o rn o o 
PQ VO VO o r' rn 
»o 
rn VO oo 
o 
cli vd 
0\ Ov oo o 
o oo VO VO 
PQ VO rn rn 
o Ov oo ^ ov 
PQ (N 
rn ooov  oo VO 
rn lo 
Q Q Q Q Q Q 
GO GO CO CO GO CO 
a a 
o o 
_N _N 
'E 'C 
o O 
ffi 
00 00 
CJ ol 
s«  
e>x) ol d 
o o 
H U H 
G3 
(X, O H d to 
Table 9: Summary of Microbial Biomass C and N for the Organic and Mineral Soil Horizon, summer 1998. 
61 
Table 10. ANOVA table associated with the microbial biomass carbon in the 
organic soil horizon. 
Source df Sums of F-ratio Prob. 
Squares  
Treatment (T) 2 1.8E + 7 6.0625 0.036* 
Whole Plot Error (co) 6 9.1 E + 6 no test 
Restriction Error (5) 0 no est. no test 
Month (M) 3 1.3 E + 7 1.9697 0.156 
TM (TM) 6 3.8 E+ 7 2.9315 0.037* 
oM (oM) 17 3.7 E + 7 no test 
Error 0 0 
Significant at a = 0.05. 
Table 11. ANOVA table associated with the microbial biomass carbon in the 
mineral soil horizon. 
Source df Sums of F-ratio Prob. 
Squares  
Treatment (T) 2 1.3 E+ 6 20.011 0.002* 
Whole Plot Error (CD) 6 194645 no test 
Restriction Error (6) 0 no est. no test 
Month (M) 3 1.3 E+ 6 5.0855 0.010* 
TM (TM) 6 1.4 E+ 6 2.6236 0.053 
G)M (COM) 15 1.6 E+ 6 no test 
Error 0 0 
Significant at a = 0.05. 
62 
The ANOVA for the MB^ in the organic layer (Table 12) found no factors to be 
significant. In the mineral horizon (Table 13), however, the month factor was found to 
have a significant effect on the MB^ • 
Table 12. ANOVA table associated with the microbial biomass nitrogen in the organic 
soil horizon. 
Source df Sums of F-ratio Prob 
Squares 
Treatment (T) 2 35454 0.0986 0.907 
Whole Plot Error (co) 6 1.1 E + 6 no test 
Restriction Error (6) 0 no est. no test 
Month (M) 3 175221 0.3441 0.793 
TM (TM) 6 1.3 E + 6 1.2561 0.327 
0)M (coM) 15 2.9 E + 6 no test 
Error 0 0 
Table 13. ANOVA table associated with the microbial biomass nitrogen in the mineral 
soil horizon. 
Source df Sums of F-ratio Prob 
Squares 
Treatment (T) 2 8093 3.5687 0.095 
Whole Plot Error (co) 6 6803 no test 
Restriction Error (5) 0 no est. no test 
Month (M) 41727 4.6240 0.014* 
TM (TM) 6 16248 0.9002 0.516 
ooM (coM) 15 54144 no test 
Error 0 0 
* at a = 0.0,5. 
63 
Line graphs of the MB^ (Figure 6) and MB^ (Figure 7) were used to identify 
trends in the treatment means over time. These figures represent the unadjusted treatment 
means and any pattern illustrated is indicative of the sum of all environmental influences 
on the dependant variable. In Figure 6, it appears that biomass (as estimated by MB^) 
declines from May through August and that these declines are most evident in the control 
samples. In both layers, control samples are clearly separated from harvest treatment 
samples in May and June, but achieve similar values in July and August. The pattern of 
increases and declines for tree-length is exactly the same in both the organic and mineral 
layers, while that for full-tree differs only in August. 
In Figure 7, the MB^ estimate of biomass varies widely in the organic layer as the 
season progresses. In the mineral layer, however, a general pattern of decline over the 
study period is evident. Initially, the full-tree treatment means are lower than the others, 
however, by July all three are displaying very similar amounts. 
With the exception of a general decline from May to June, there is little similarity 
between the two estimators of microbial biomass. 
64 
Control 
Tree-length 
Full-tree 
1400 -1 
• Control 
S 1200 - ■ Tree-length 
bJO 
"bi) ^ Full-tree 
O 1000 - 
OC/3 ) 
B 800 - 
o 
m 
600 
400 - 
200 - 
May June July August 
Months 
Figure 6. Mean MB^ (|.ig/ g soil) throughout the sampling season, demonstrating 
interactions between harvest treatments influenced by soil descriptors: 
a) organic soil layer; b) mineral soil layer. 
65 
• Control 
■ Tree-length 
^ Full-tree 
ISO-. 
• Control 
160 - 
■ Tree-length 
140 - ^ Full-tree 
120 -| 
100 - 
80 - 
60 - 
40 - 
20 -- 
May June July August 
Months 
Figure 7. Mean MBN (gg/ g soil) throughout the sampling season, demonstrating 
interactions between harvest treatments influenced by soil descriptors: 
a) organic soil layer; b) mineral soil layer. 
Mean Microbial Biomass N (|ig/g soil) Mean Microbial Biomass N (gg/g soil) 
66 
Relationship Between Soil Descriptors and Microbial Biomass 
Scattergrams of the MB^ and MB^ against the soil descriptors demonstrate the 
association with each parameter (Figures 8 through 11). Comparisons between the 
organic and mineral layers must take scale into account. With the exception of pH, 
positive correlations between MB^ and MB^ and the soil descriptors are more apparent in 
the mineral layer. 
Pearson and Spearman correlations were done to discover statistical relationships 
between the measured variables (Table 14). The pH and moisture content factors were 
found, by the Pearson correlation and not the Spearman, to have a significant association 
with the MBc in the organic soil layer. Total N, moisture content, and the OM content 
factors had a significant association with both the MBc and the MB^ in the mineral soil 
horizon, as found by both the Pearson and Spearman correlations. Total N was found to 
be correlated with the MB^ in the organic soil horizon. The Spearman correlation also 
found total P to be significant with MB^ in the organic horizon. 
Interestingly, and rather unexpectedly, the Pearson analysis revealed no 
correlation between MB^ and MB^ in the organic layer. However, Spearman’s, which 
uses ranked, rather than absolute, values did identify this relationship as significant. The 
anticipated MB^ to MB^ relationship was identified by both tests in the mineral layer. 
67 
Total Phosphorus (mg/Kg) 
d) 
Total Nitrogen (mg/Kg) 
e) 
Figure 8. Scattergrams of the microbial biomass carbon with the soil description 
parameters in the organic soil horizon; a) pH; b) total P; c) total N; 
d) moisture content; and e) OM content. 
a) 68 b) 
Total Phosphorus (mg/Kg) 
c) d) 
Total Nitrogen (mg/Kg) 
e) 
Organic matter (%) 
Figure 9. Scattergrams of the microbial biomass carbon with the soil description 
parameters in the mineral soil horizon: a) pH; b) total P; c) total N; 
d) moisture content; and e) OM content. 
a) 69 b) 
Total Phosphorus (mg/Kg) 
c) d) 
Total Nitrogen (mg/Kg) Moisture content (%) 
e) 
Figure 10. Scattergrams of the microbial biomass nitrogen with the soil description 
parameters in the Organic Soil Horizon; a) pH; b) total P; c) total N; 
d) moisture content; and e) OM content. 
70 
a) b) 
Total Phosphorus (mg/Kg) 
C) d) 
Total Nitrogen (mg/Kg) Moisture content (%) 
e) 
Figure 11. Scattergrams of the microbial biomass nitrogen with the soil description 
parameters in the Mineral Soil Horizon: a) pH; b) total P; c) total N; 
d) moisture content; and e) OM content. 
71 
Table 14. Results of the Pearson (lower triangle) and Spearman (upper triangle) 
Correlation analyses for the organic and mineral soil horizons. 
MB, MB^ WC pH* N OM 
Organic Horizon 
MBe 1.000 .353* -.044 .165 .157 -.065 .288 
MBN .258 1.000 .193 -.023 .349* .407* -.067 
WC -.481* -.017 1.000 -.641* .040 .243 .055 
pH** .536* -.263 -.250 1.000 -.079 -.501* .031 
P .247 .323 -.073 -.225 1.000 .411* .004 
N .078 .367* .101 -.403* .448* 1.000 .022 
OM .367 -.018 -.032 .223 .074 -.178 1.000 
IVIineral Horizon 
MBc 1.000 .584* .468* -.017 .037 .424* .367* 
MBN .686* 1.000 .524* -.021 .227 .598* .582* 
WC .396* .362* 1.000 -.019 .323 .476* .470* 
pH** .209 .150 .329 1.000 .098 .114 .088 
P .022 .163 .348* .258 1.000 .557* .427* 
N .554* .525* .389* .415* .519* 1.000 .913* 
OM .494* .553* .400* .484* .428* .915* 1.000 
* Correlation is significant at the 0.05 level (2-taiIed). 
**pH was calculated as hydronium ion concentration (M). 
Note; Some data in the Organic Horizon was missing. 
Total N was significantly correlated with many soil descriptors in both the organic 
and mineral soil horizons. In the case of total N and % OM, the relationship is so strong 
as to suggest the inclusion of only one or the other as a covariate. 
72 
Variability in Microbial Biomass Adjusted for Covariates 
Summarized in Table 15 is the ANCOVA for the MB^ in the organic soil horizon, 
revealing that the interaction of the treatment and month significantly affected the 
response variable. The ANCOVA results for the MB^ in the mineral horizon are 
summarized in Table 16. This analysis showed that the treatment (T) significantly 
affected MB,;-. The LSD Post Hoc test found the control treatment to be significantly 
different from both of the harvest treatments, tree-length and full-tree (Figure 12). 
Because total N was independently correlated with OM content, an ANCOVA was run 
without the total nitrogen covariate factor. This did not result in any significant 
differences from the original ANCOVA, therefore, only the original analysis is presented. 
Table 15. ANCOVA table associated with the microbial biomass carbon in the 
organic soil horizon. 
Source df Sums of F-ratio Prob. 
Squares 
pH (pH) 1 3924 no test 
Moisture Content (WC) 1 155827 no test 
Treatment (T) 2 76222 0.0405 0.961 
Whole Plot Error (o)) 6 5.6 E + 6 no test 
Restriction Error (5) 0 no est. no test 
Month (M) 3 207285 0.1303 0.939 
TM (TM) 6 1.1 E +7 3.6322 0.048* 
toM (coM) 8 4.2 E + 6 no test 
Error 0 0 
* Significant at a = 0.05. 
73 
Table 16. ANCOVA table associated with the microbial biomass carbon in the 
mineral soil horizon. 
Source df Sums of F-ratio Prob. 
Squares 
Total Nitrogen (N) 89050 no test 
Moisture Content (WC) 33172 no test 
Organic Matter (OM) 96716 no test 
Treatment (T) 527640 5.5325 0.043* 
Whole Plot Error (co) 286111 no test 
Restriction Error (5) no est. no test 
Month (M) 3 16141 0.0608 0.979 
TM (TM) 6 606084 1.1416 0.386 
(oM (coM) 15 1.3 E + 6 no test 
Error 0 0 
Significant at a = 0.05. 
74 
a 
Harvest Treatments 
Figure 12. LSD Post Hoc tests demonstrating similar and dissimilar harvest treatments, 
using mean MBc (gg/ g soil) as the response variable in the mineral soil 
horizon. The letters on the bars represent which groups are similar to or 
different from each other. 
75 
The ANCOVA results for microbial biomass N in the organic soil horizon, 
summarized in Table 17, show that neither the factors nor their interaction influence the 
response. The ANCOVA was run with only the total nitrogen covariate factor in attempt 
to increase the degrees of freedom. This was done because the total N and the organic 
matter were highly correlated with one another according to the Pearson and Spearman 
correlation tests. The ANCOVA results for MB^ in the mineral layer, summarized in 
Table 18, again revealed that no factors were found to be significant. 
Table 17. ANCOVA table associated with the microbial biomass nitrogen in the organic 
soil horizon. 
Source df Sums of F-ratio Prob 
Squares 
Total Nitrogen (N) I 39762 no test 
Treatment (T) 2 11242] 0.6634 0.549 
Whole Plot Error (OD) 6 508400 no test 
Restriction Error (5) 0 no est. no test 
Month (M) 3 100170 0.2351 0.870 
TM(TM) 6 1.3 E+ 6 1.4930 0.246 
OJM (coM) 15 2.1 E + 6 no test 
Error 0 0 
76 
Table 18. ANCOVA table associated with the microbial biomass nitrogen in the mineral 
soil horizon. 
Source df Sums of F-ratio Prob 
Squares 
Total Nitrogen (N) 1 4133 no test 
Moisture Content (WC) I 2470 no test 
Organic Matter (OM) 1 4496 no test 
Treatment (T) 2 1326 0.5267 0.615 
Whole Plot Error (co) 6 7556 no test 
Restriction Error (6) 0 no est. no test 
Month (M) 1932 0.1988 0.895 
TM (TM) 6 5348 0.2751 0.940 
wM (o)M) 15 48604 no test 
Error 0 0 
77 
DISCUSSION 
BACTERIOLOGY 
The culturing, isolation and identification of bacteria was undertaken to determine 
if different levels of biomass removal affected the number and type of bacterial species 
found in the treatment plots. In this investigation, the number of bacteria cultured 
declined as the sampling season progressed. This observation may be attributed to the soil 
temperature and moisture differences measured throughout the season. In the spring, 
warm temperatures and high amounts of moisture due to snow-melt (see Figure 3), 
appeared to stimulate microbial activity. Lundgren and Soderstom (1983) and Schlentner 
and VanCleve (1985) also attributed increased microbial activity to increased temperatures 
and moisture. As the climate continued to become warmer and drier, the number of 
cultures declined. A summary and description for each of the five bacterial species 
identified follows below. 
Chryseomonas luteola was isolated on a number of occasions from the organic and 
mineral soil collected from all three treatment groups in this study, therefore, occurring in 
the general environment. Chryseomonas luteola is described in Bergey’s Manual (Holt et 
al 1994) as not being known to be present in the general environment but are saprophytes 
or commensals of humans. This discrepancy may be due to the limitations of the API 20E 
system to identify genetically similar strains (Palmieri et al 1988). In a study by Anzai et 
al (1997), it was found that C luteola has a 93.9% sequence homology of 16S rRNA 
with Pseudomonas spp. Thus, Anzai et al (1997) concluded that Chryseomonas is a 
junior subjective synonym of Pseudomonas spp. which are widely distributed in nature 
78 
(Holt et al. 1994). It is possible that what API 20E identified as C luteola, is a subspecies 
of Pseudomonas thus explaining its high occurrence of in our soil. 
Identified with a rating of “low discrimination”, meaning that it may have been 
misidentified, P. fluorescens was cultured from only the harvested plots during the month 
of July. P. fluorescens is an aerobic microorganism with a respiratory type of metabolism 
(Holt et al. 1994) that functions in the synthesis of urease and in the process of 
denitrification (Killham 1994). Furthermore, P. fluorescens decomposes pure proteins 
with the formation of end products such as ammonia (Holt et al. 1994). P. fluorescens 
also functions as an antagonist, providing effective biocontrol of pathogens, such as 
Gaeeumannomyces graminis (take-all root disease) and Pythium ultimum (damping-off 
fungi) (Whipps 1997). Since it is commonly found in a number of niches, the low 
frequency of P. fluorescens observed here may be due to the limitations of culturing 
techniques as well as the possibility for misidentification by the API 20E system (Palmieri 
etal. 1988). 
Aeromonas salmonicida was also identified with a “low discrimination” rating. 
Occurring on all three treatment plots throughout the entire sampling season, A. 
salmonicida is a facultative anaerobe that reduces nitrates and nitrites in the environment 
(Holt et al. 1994). This bacteria is chemolithotrophic (has both respiratory and 
fermentative types of metabolism); grows well at 22-28“C; occurs in fresh water and 
sewage; and may be pathogenic to frogs, fish, and humans (Holt et al. 1994). 
79 
Isolated throughout the entire sampling season from all three treatment plots, 
Serratia marcescens was identified with a confidence rating of “good to very good”. 
Serratia marcescens, like A. salmonicida, is facultatively anaerobic, chemolithotrophic, 
and reduces nitrates (Holt et al. 1994). This bacteria is capable of growth at 30-37 ”C 
(Holt et al 1994). Serratia marcescens usually appears in soils enriched with chitin, and 
decomposes these proteins into various end products, such as ammonia (Holt et al 1994). 
Further, this species functions as an antagonist, providing effective biocontrol of 
pathogens for plants (Whipps 1997). 
Syntrophomonas multifilia was identified with a rating of “good”, and was 
cultured only a few times during the sampling season; however, it did occur on all three 
treatment plots. Capable of growing at temperatures of 30-37 °C, S. multifilia functions 
as a fermenter (Holt et al 1994). Mostly occurring in anoxic (without O2) mud, this 
microbe obtains energy via p-oxidation of fatty acids, degrading the acids 
primarily to acetate and H2 (Holt et al 1994). This species was extracted from only those 
samples that happened to be from very wet pockets of soil. Having such a restricted 
niche, the low frequency of S. multifilia throughout the sampling season is therefore 
understandable. 
The occurrence of the various bacterial isolates seems to respond to moisture and 
temperature levels rather than levels of organic residue. However, residues may, over 
time, influence the moisture and temperature of the soil (Hendrickson et al 1985; Entry et 
al 1986), which will eventually influence the type and amount of bacteria inhabiting the 
soil. Further, considering that the bacterial species identified were primarily nitrifying and 
80 
denitrifying species, there is the potential for short and long-term effects on nitrogen 
presence and availability on this site. 
SOIL RESPIRATION 
The objective of measuring the soil respiration was to determine if increasing levels 
of biomass removal would result in measurable differences. The ANOVA and subsequent 
LSD test revealed that CO2 evolved in the control (uncut) plots was significantly higher 
than that evolved in the harvested plots. Since soil respiration consists of both root and 
microbial contributions, it is difficult to separate the two. Killham (1994) states that roots 
may contribute up to 30% of total soil respiration. In the harvest plots, the measuring 
apparati were intentionally placed so as to minimize the effect of living plant roots. The 
difference in CO2 evolved (approximately 18%) could therefore be attributed to the 
presence or absence of roots. However, soil temperatures in the control plots were much 
lower than in the cut plots and, in general, respiration rates are reported to decline with 
temperature (Entry et al 1986). For example, the Qio for root respiration has been 
measured at 2.5 (Singh and Gupta 1977) and at 2.4 (Uchida et al. 1998). Weber (1990) 
also reported a decline in CO2 evolution when aspen stands were cut and burned. He 
attributed the difference to the loss of vegetation and biomass on site, contributing to 
lower metabolic activity. Furthermore, pH may be a factor affecting microbial respiration 
rates. Anderson and Domsch (1993) found that microbial communities released more 
CO2-C (per unit MB) under acidic soil conditions than those with a more neutral pH. 
81 
Since the control plots had a more acidic pH than the harvested plots, the difference in 
CO2 evolved may be related to this characteristic. 
Given the opposing influences of root presence, soil pH and soil temperature, it 
seems reasonable to also consider that some of the difference was due to a change in 
either the quantity or activity level of the microbial component of the soil. From the 
statistical analyses of the estimates of microbial biomass (Table 11) it appears that 
treatment alone did not contribute significantly to quantitative differences in the organic 
layer although ANCOVA did reveal a statistically significant effect in the mineral horizon 
(Table 12). However, looking at the treatment means for the month of August, just before 
the respiration trials, it is apparent that differences between treatments at this time were 
minimal. It could therefore be surmised that quantity is not driving the differences in 
respiration. The level of microbial activity, on the other hand, may be a factor. 
Orchard and Cook (1983) stated that an increase in respiration of a silt loam used 
for pastoral and crop farming, was most likely due to an increase in microbial activity. 
Weber (1990) attributed the decrease in soil respiration from uncut to harvested plots in 
his study to a decline in microbial activity. Metabolic activity of microorganisms is 
thought to depend on a number of factors including temperature, moisture and substrate 
(Orchard and Cook 1983; Lundgren and Soderstom 1983; Schlentner and Van Cleve 
1985; Bosatta and Agren 1994; Berg et al 1998; Bauhus et ah 1998). It is possible that 
the soil environment of the control plots is more conducive to higher rates of activity than 
is that of the cutover. Exposed areas may experience lower moisture levels, higher soil 
temperatures and lower quality organic matter inputs {i.e. essentially no root exudates or 
82 
foliar litterfall) all of which could depress microbial activity. Interestingly, chloroform- 
flimigation-extraction has as one of its advantages, the inclusion of all microbial organisms 
regardless of their metabolic state. Dormant organisms are therefore just as likely to be 
included in the quantitative estimate as are active organisms (Martens 1995). 
Although this one time measurement must be considered cautiously, however, it 
appears to be an indicator of a change in state, if not quantity, of the microbial biomass 
after a forest removal event. Given that microbial populations may respond to crop 
removal events, there are implications for nutrient cycling, nutrient availability and, 
ultimately, the success and growth of vegetative regeneration. This change in state implies 
a reduction of microbial activity, which in turn may lead to a reduction in nutrient cycling 
and availability. Further investigation into the nature of change in microbial communities 
must be conducted before we can be sure of the direction of the trajectory. 
VARIABILITY IN SOIL DESCRIPTORS 
One-way ANOVA for the soil descriptors revealed no significant treatment effects 
except for pH in the organic layer. It has been four years since the harvest treatments 
were imposed and it is interesting that, despite the obvious and dramatic change in the 
immediate environment, key soil characteristics such as % OM, moisture content, total P, 
total N and pH in the mineral layer do not differ significantly from one another. This 
observation must be made with caution however because other factors may have 
prevented the assignation of statistical significance. For example, there may be so much 
variability in the samples that trends due to treatment effect were obscured. Spatial 
83 
heterogeneity of the forest floor has been noted by other authors (Arp and Krause 1984). 
It is possible that random sampling may have captured a wide variety of combinations of 
mineral and organic elements. In this study, bulking of the samples was intended to 
reduce variation but may not have been sufficient given the number of samples actually 
incorporated. One means of determining the minimum sample size required to identify 
treatment differences is by calculating the required number of samples based on observed 
variance. Thus, the approximate minimum sample size required would be 1143. This is 
such a large number of samples that it is not practical to undertake. As always, scientific 
method must find a compromise between the resources available and the level of certainty 
associated with results. 
In the case of pH in the organic layer, ANOVA and subsequent application of the 
LSD test for differences between treatment means revealed that control treatment means 
were significantly lower than those of the harvested areas. This maintenance of low 
acidity in the uncut, mature forest plots may be attributed to the continual inputs of acidic 
foliage (Brady and Weil 1986) and to higher cation uptake by roots. Hendrickson et al. 
(1985) and others (Baath et al 1995; Smolander et al 1998) have reported increases in 
soil pH after forest cover removal. However, this change may be temporary. The removal 
of forest vegetation has been shown to accelerate leaching of nitrates and weathering of 
pedogenic minerals, which release hydronium ions, thus promoting a return to the soil’s 
original acidic level 
(Rowell 1994). 
84 
RELATIONSHIP BETWEEN SOIL DESCRIPTORS AND MICROBIAL 
BIOMASS 
Correlation analysis was used to identify relationships between two independent 
variables; it is not expected that change in one variable is dependant upon change in the 
other {i.e. regression) but rather that the magnitudes of change are somewhat similar (Zar 
1996). As stated previously, some of the results of the Pearson correlation test confirmed 
hypotheses about the samples while other results called our understanding of the system 
into question. A discussion of key findings follows. 
In the organic soil layer, MBc was found not to be correlated with MBN. This was 
somewhat unexpected considering that throughout the literature, the two are often linked 
with C/N ratios being used to confirm or estimate quantities of one or the other 
(Marumoto et al. 1982; Marumoto 1984; Ohtonen etal. 1992; Raubuch and Beese 1998; 
Haron et al. 1998). However, the result obtained here may be explained using two 
theories. For example, Alef (1993) suggested that fumigation of soil with chloroform 
(CHCI3) liberates C and N that is chemically bound in humic fractions. These liberated 
minerals may inflate the results of the MBc and MBN calculations and lead to false 
conclusions fi’om the correlation tests. 
Another theory suggests that even a “reasonable” range of soil C:N ratios {i.e. 
15; 1 to 3:1) might not reveal a correlation. Specifically, the MBC;MBN ratios for the full 
data set of the organic soil (Appendix I) ranges from 2.6:1 to 13.9:1. Even though these 
ratios are reasonable according to standards listed in the literature (Paul and Clark 1989), 
85 
the actual range of MBc and MBN values is so large that statistical relationships between 
the two may be difficult to determine. 
The result that MBc and MBN were correlated in the mineral horizon adds weight 
to Alef s (1993) proposed argument that fumigation with CHCI3 liberates C and N from 
humic fractions in addition to that from the MB. In samples from the mineral layer, mean 
% organic matter varied from 11 to 13 for the three treatments with a maximum of 30% 
measured in a control sample (high values are due to inexperienced sampling). These 
values are on the order of a fifth to a tenth that of the organic layer samples and are 
somewhat higher than others found in the literature. For example, Hendrickson et al 
(1985) reported mean % OM values of 4.11% to 4.99% in the mineral soil and 57 to 76% 
in the organic layer of a mixed-wood forest in Central Ontario treated with conventional 
and whole-tree harvest methods. Chatarpaul (1987) studied conventional and whole-tree 
harvest methods on a mixed-wood forest of Central Ontario, where mean % OM values of 
the mineral soil ranged from 3.13% to 7.59%, with mean values of 54% to 69% in the 
organic soil. 
The MBc and MBN were significantly correlated to % OM in the mineral layer but 
not in the organic layer. At first glance, these results may seem contradictory. However, 
two possible explanations for this apparent discrepancy present themselves. First, recall 
that samples from the organic layer averaged 80% (range 59 to 96%) organic matter while 
those of the mineral layer averaged 12% (range of 7 to 30%). Furthermore, estimates of 
MBc in the organic layer were at least an order of magnitude greater than those of the 
mineral layer and there were no apparent differences between treatment means for either 
86 
OM or MBc in the organic layer. It may be that variability in the measures of the organic 
layer was too great to reveal either similarities or differences in patterns of change. 
Second, the microbial biomass requires substrates. One might conclude that more OM in 
the soil matrix would be related to larger, or perhaps more active, populations of 
microorganisms (Hendrickson et ah 1985; Entry et al 1986; Ohtonen et al. 1992; Bauhus 
et al 1998). The range of OM in the mineral samples might have been sufficient to reveal 
a similar range in the microbial biomass while the much higher OM presence in the organic 
samples would be associated with a “saturation” effect. Treatment effects, either due 
directly to biomass removal or indirectly to changes in micro-environment, might be more 
easily detected in the mineral layer where numbers are naturally lower and populations are 
more likely to respond to moisture, temperature or substrate gradients. 
Total N was significantly correlated with many factors in both the organic and 
mineral soil horizons. Specifically, with MBN, pH and total P in the organic samples and 
with MBc, WC, OM, pH and total P in the mineral layer. The correlation to OM in the 
mineral samples is understandable as this material, particularly after some decomposition, 
may be nitrogen rich (Haynes 1986; Killham 1994), while that of the more surficial layer 
may be less decomposed and therefore contain materials with wider C/N ratios. For 
example, compounds such as chitin, cellulose and lignin represent highly resistant pools of 
organic C (Killham 1994). These compounds, which are relatively N poor, may persist in 
the soil for long periods of time, requiring much time, energy, and import of N for 
degradation to occur (Killham 1994). 
The correlation of total N to MBN in both layers may provide further evidence of 
the liberation of N due to CFE. With respect to the mineral samples, it may also be 
explainable in terms of larger microbial populations found in association with more 
nitrogen rich material. Nitrogen levels in the soil may be influenced by factors such as P 
levels, pH, and moisture content. For example, Gallardo and Schlesinger (1994) reported 
that P is less frequently immobilized than N by soil microbes, causing an increase of N/P 
ratios in the litter, and subsequently causing a limiting effect on the MB. They further 
speculated that P limitations in soils with extreme pH, where pH levels are highly acidic or 
highly basic, will affect the activity and nature of microorganisms present. In addition, 
Baath et al (1995) suggest that pH may have the ability to alter other soil properties, such 
as C/N ratios, which indirectly affect microbial biomass by restricting available nutrients. 
The level of moisture content may also affect the nitrogen levels, and subsequently 
C/N/P ratios. Soil bacteria, require a water film for movement and can only remain active 
if there is suitable moisture in the soil, so that nutrients and waste products can diffuse in 
and out of the cell (Wong and Griffen 1976a & b). Bacterial activity will increase or 
decrease as the water potential rises and falls, respectively. If moisture levels are 
inadequate, the mineralization-immobilization of nutrients may not occur, thus limiting the 
growth and activity of the microbial biomass. 
In the organic layer samples, pH was correlated positively to MBc and negatively 
to WC and total N. Bosatta and Agren (1994) speculated that properties of the original 
litter and soil physical factors partly determine the amount of MB in the soil. Bauhus et 
al (1998) stated that MB amounts are sensitive to changes in soil physical and chemical 
88 
composition. Therefore, in the organic samples, the pH was positively correlated to the 
MBc possibly because the organic residues remaining were less acidic, as compared to the 
continual inputs of more acidic spruce needles in the control plots. The less acidic sites 
(cut areas) did not have a continual input of acidic residues, making the environment 
tolerable for a greater variety of soil microbes. Although the site experienced a large 
deposit of needles during the harvest, this input may have been quickly assimilated, 
creating a surge in MB activity and growth, followed by a drop in biomass due to the lack 
of continual inputs. 
Moisture content and total N were negatively correlated with pH, suggesting that 
the moisture content may be implicated in changes to pH. On the harvested sites, where 
the soil is less acidic, the moisture will run off, evaporate or drain through the soil because 
there is little vegetation uptake. In addition, the total N decreases on these harvested sites 
because there is little vegetation to take up and retain the nutrients. As water moves down 
through the soil, N as NOs' ( as suggested by the bacteriology results) may be leached out 
and deposited into the water table (Haynes 1986). Also, the level of N-containing organic 
residues is decreased on the harvested sites due to the lack of continual inputs by 
vegetation. 
However, in the mineral samples, pH is positively correlated with OM and total N 
because more nutrients and OM may be translocated into the mineral horizon from the 
organic layer. Since mineral turnover rates and chemical processes are much slower in 
mineral soil compared to organic soil it stands to reason that as the pH increases, the level 
of OM and N increases in the mineral layer, due to the lack of assimilation by plants. 
89 
VARIABILITY IN MICROBIAL BIOMASS 
With respect to actual measurements of biomass reported elsewhere for boreal 
forest soils, the values obtained here are reasonable. For example, Thibodeau et al. 
(2000) reported very similar values for both MBc and MBN in the soil of thinned and 
unthinned balsam fir stands. Furthermore, the values reported here account for 1-3% of 
the total soil organic C which is the standard accepted by most soil ecologists (ICillham 
1994). For these reasons, as well as the C:N ratios, it appears that the actual values 
measured are credible. This is an important point as the estimates were arrived at using 
CFE without calibration against another method (usually CFI). The latter is used to 
determine a suitable conversion factor however, it appears that the conversion factors 
applied in this study were appropriate. 
Microbial Biomass Carbon 
Statistical analysis for the MBc in the organic horizon (ANCOVA, Table 15) found 
the interaction of the treatment and month factors to be significant. This suggests that a 
combination of seasonal conditions along with the amount of the organic residues present 
influences the amount of MBc found in the organic soil horizon. One way to decipher 
these influences is to consider the pattern of treatment means as presented in Figure 6a 
with the climatic information presented in Figure 3. As temperatures increased and 
moisture levels declined, the microbial biomass generally decreased. Specific evidence for 
the interplay between soil temperature and soil moisture is found in the values for the 
90 
control, tree-length and full tree treatments in July. Conditions on the full-tree plots, 
where all above-ground portions were removed, resulted in the lowest MBc value 
(approximately 5600 ug/g). It can be expected that the lack of coarse woody debris on 
the surface and/ vegetative regeneration led to high soil temperatures and evaporation. At 
this same time, values for the control and tree-length plots lie at 7000 ug/g 7800 ug/g, 
respectively. In the uncut plots, soil temperature is indeed lower and it could be predicted 
from a general knowledge of forested systems that soil water contents were reduced due 
to interception and uptake by the forest canopy. The latter effect would be exacerbated 
by a relative lack of precipitation over the growing season. In the tree-length harvested 
plots, soil temperatures would be higher but mitigated by surface debris while bulk 
precipitation would not be intercepted by standing vegetation. Thus, as indicated in the 
literature, microbial biomass is quite sensitive to unfavourable changes in temperature 
(Howard and Howard 1979; Weber 1990) and moisture content (Orchard and Cook 1983; 
Lundgren and Soderstom 1983). In this study, while microbial biomass numbers generally 
declined it is not clear which of treatment or environmental conditions were responsible. 
The ANCOVA for the MBc in the mineral layer found the treatment factor to have 
a significant effect on the MBc, suggesting direct influences of the harvest treatments. 
Figure 6b presents the treatment means by month. As noted earlier, the LSD test (based 
on ANOVA) confirms that mean MBc in the control plots is greater than that of the two 
harvest treatments. There are several possible reasons for this significant (at a = 
0.05) difference: 1) the removal of standing biomass and annual inputs has been 
documented to result in quantitative differences in the soil microbial populations 
91 
(Hendrickson et al 1985; Entry et al 1986; Chatarpaul 1987; Foster and Morrison 1987); 
2) a change in the microclimate to one less favourable for indigenous 
populations/communities may have contributed to declines in microbial numbers (Howard 
and Howard 1979; Orchard and Cook 1983; Lundgren and Soderstom 1983; Weber 1990; 
Berg et al 1998); 3) a change in the soil chemistry to one less favourable; and, 4) loss of 
root exudates. For example, less organic matter inputs and change in pH due to loss of 
acidic foliar inputs from the previous forest stand causes the soil properties to change 
(Brady and Weil 1996). 
However, it is important to note that qualitative changes in the microbial biomass 
are not captured by this method of measurement. It has been indicated in the literature 
that the removal of forest vegetation results in a change in the integrity of microbial 
functional groups (Gallardo and Schlesinger 1994; Baath et al 1995; Staddon etal 
1996). A possible shift in microbial populations is further substantiated by the results for 
the soil respiration data, where higher levels of CO2 efflux were found on the control 
treatment, indicating a higher level of microbial biomass/activity. 
Microbial Biomass Nitrogen 
The ANCOVA results for the MBN in both the organic and mineral horizons 
showed that the treatment and month factors, as well as their interaction, had no 
significant effects. Difficulties with the estimation of microbial biomass using nitrogen 
have been discussed elsewhere in this paper. Thibodeau et al. (2000) also reported a lack 
of change in the microbial biomass using both nitrogen and carbon as reference elements. 
92 
In general, Wardle and Ghani (1995) suggest that microbial biomass may not be 
reliably estimated when calibrating methods (CFE, CFI, and/or SIR) are employed on a 
spatially heterogeneous soil. Furthermore, they found that only comparatively large 
relative differences in the MB can be estimated reliably, suggesting that MB as a 
bioindicator of soil quality is limited. Therefore, as is the case with all experimental 
techniques, caution should be exercised when interpreting and attributing changes in soil 
microbial biomass to treatment or seasonal effects (Wardle and Ghani 1995). 
93 
CONCLUSIONS 
With respect to the isolations, only five bacteria that were cultured were in fact 
identified - two with low discrimination and three with good to very good discrimination. 
The species fulfilled a variety of roles in the soil ecosystem including N transformations. 
The API 20E system performed adequately on the organisms isolated. The author 
suspects that many more organisms were present than were cultured. 
With respect to soil respiration measurements, it was evident that CO2 evolved on 
control plots in the uncut, mature forest was significantly greater than that of the 
harvested plots. The author suggests that activity levels, and in turn, available C as root 
exudates, rather than biomass quantities, were responsible for the difference. 
With respect to the soil descriptors, one-way ANOVA revealed no differences 
among the treatments for %OM, WC, total N and total P. Soil pH did, however, vary 
significantly with the mean of the control plots lower than those of the harvested plots. 
Despite obvious and dramatic changes in the environment above the soil, the soil itself 
does not exhibit statistically significant modification four years after the treatments were 
imposed. 
Several of the soil descriptors exhibited correlation to each other as well as to the 
measures of microbial biomass. In the organic layer, it was noted that MBc and MBN did 
not in fact exhibit correlation with one another. A correlation between the two was 
determined in the mineral layer. The author presents several reasons for this apparent 
contradiction. Use of the chloroform fumigation extraction method without calibration by 
a second method gave reliable estimates of microbial biomass for boreal coniferous soils. 
94 
Application of ANCOVA to the measures of MBc and MBN produced different 
results for the two soil layers. Neither main effects nor the interaction term significantly 
affected MBN in either layer. The interaction of treatment and month was significant for 
MBc in the organic layer. Treatment alone significantly affected MBc in the mineral layer. 
In the latter, mean MBc in the control plots was nearly double that of the full-tree and 
tree-length plots. Throughout the data set, MBN appeared to fluctuate more so than MBc; 
some reasons are suggested. The most sensitive indicator of change in the system appears 
to be microbial biomass C in the mineral layer. This may be because gradients in 
temperature, moisture and substrate quality are more readily apparent and indigenous 
microbial populations more likely to respond when changes do occur. 
In general, descriptors of soil characteristics thought to reflect nutritional quality 
and quantity do not demonstrate statistically significant responses to harvest treatment 
although microbial response to soil moisture and temperature is evident. This particular 
black spruce system, four years after harvest, does not appear to have been adversely 
affected by the treatments imposed. It would be interesting to repeat this study in the 
future to determine if it is moving away from or closer to its pre-harvest set of conditions. 
95 
RECOMMENDATIONS 
Having had the opportunity to consider alternative approaches, the author would 
recommend the following as options for future work in this area; 
1. Increase the number of samples analyzed in order to better capture and apply 
statistically the effect of spatial heterogeneity, 
2. Avoid missing data by ensuring that generously sized soil samples are collected, 
3. Consider soil texture and modify soil characteristics, such as moisture content, 
which may have affected the results of chloroform fumigation extraction (CFE), 
4. Take respiration measurements much more frequently to account for rapid shifts in 
environmental conditions. 
5. Use chloroform fumigation incubation or substrate induced respiration as a means 
to calibrate a site specific conversion factor for CFE, 
6. Compare the impacts of anthropogenic disturbances to natural disturbances, such 
as wildfire, on soil microbial populations, 
7. Using a chronosequence, assess microbial changes through successional stages of 
development. 
8. Artificially change some of the soil descriptors in a contf oiled-factorial experiment 
to better isolate their effects. 
96 
LITERATURE CITED 
Alef, K. 1993. Bestunmieng mikrobieller biomasses im Boden; Eine kirtische 
Betrachtung. Z. Pflanzenernahr. Bodenkd. 156:109-114. (Original not seen) 
Alexander, M. 1961. Introduction to soil microbiology. John Wiley & Sons, Inc., New 
York. 472 pp. 
Anderson, D.W. and J.J. Schoenau. 1993. Soil humus fractions pp.391-395 in Carter, 
M.R. (ed.) Soil Sampling and Methods of Analysis, 1993. 
Anderson, J.M. 1978. A method to qunatify soil-microhabitat complexity and its 
application to a study of soil animal species diversity. Soil Biol. Biochem. 
10:77-78. 
Anderson, J.P.E. and K.H. Domsch. 1975. Measurements of bacterial and fungal 
contributions to respiration of selected agricultural and forest soils. Can. J. 
Microbiol. 21:314-322. 
Anderson, J.P.E. and K.H. Domsch. 1978. A physiological method for the quantitative 
measurement of microbial biomass in soils. Soil Biol. Biochem. 10;215-221. 
Anderson, T.H., and K.H. Domsch. 1993. The metabolic quotient for CO2 (q CO2) as a 
specific activity parameter to assess the effects of environmental conditions, such 
as pH on the microbial biomass of forest soil. Soil Biol. Biochem. 25:393-395. 
Anderson, T.H., and R.G. Joergensen. 1997. Relationship between SIR and FE estimates 
of microbial biomass C in deciduous forest soils at different pH. Soil Biol. 
Biochem. 29:1033-1042. 
Anonymous. 1978. Method for analysis of TOC in solution. Land Resource Science, 
University of Guelph, ON 
Anonymous. 1984. Method for analysis of TSN in solution. Land Resource Science, 
University of Guelph, ON 
Anonymous. 1992. Differential Gram stain. Becton Dickinson & Co. 
Anonymous. 1996. API 20E system. bioMerieux Vitek, Inc. 
Anzai, Y., Y. Kudo, and H. Oyaizu. 1997. The phylogeny of the genera Chryseomonas, 
Flavimonas, and Pseudomonas supports synonymy of these three genera. 
Internat. J. Sys. Bacteriol. 47:249-251. 
97 
Arp, P.A., and H.H. Krause. 1984. The forest floor: lateral variability as revealed by 
systematic sampling. Can. J. Soil Sci. 64:423-437. 
Baath, E., A. Frostegard, T. Pennanen, and H. Fritze. 1995. Microbial community 
structure and pH response in relation to soil organic matter quality in wood-ash 
fertilized, clear-cut or burned coniferous forest soils. Soil Biol. Biochem. 27:229- 
240. 
Bailey, M.J., A.K. Lilley, R.J. Ellis, P.A. Bramwell, and I.P. Thompson. 1997. Microbial 
ecology, inoculant distribution, and gene flux within populations of bacteria 
colonizing the surface of plants: case study of a GMM field release in the UK. 
pp.479-500, in van Elsas, J.D., J.T. Trevors, E.M.H. Wellington (eds.) Modem 
Soil Microbiology, Marcel Dekker, New York, 1997. 683pp. 
Bakken, L.R. 1997. Culturable and nonculturable bacteria in soil pp.47-62 in van Elsas, 
J.D., J.T. Trevors, E.M.H. Wellington (eds.) Modem Soil Microbiology, Marcel 
Dekker, New York, 1997. 683pp. 
Bauhus, J., D. Pare, and L. Cote. 1998. Effects of tree species, stand age and soil type on 
soil microbial biomass and its activity in a southern Boreal forest. Soil Biol. 
Biochem. 30:1077-1089. 
Beck, T., R.G. Joergensen, E. Kandeler, F. Makeschin, E. Nuss, H.R. Oberholzer, and S. 
Scheu. 1997. An inter-lab comparison of ten different ways of measuring soil 
microbial biomass C. Soil Biol. Biochem. 29:1023-1032. 
Berg, M.P., J.P. Kniese, and H.A. Verhoef 1998. Dynamics and stratification of bacteria 
and fungi in the organic layers of a Scots pine forest soil. Biol. Fertil. Soils 
26:313-322. 
Bewley, R.J.F., and D. Parkinson. 1985. Bacterial and fungal activity in sulphur dioxide 
polluted soils. Can. J. Microbiol. 31:13-15. 
Bloem, J., P.R. Bolhuis, M.R. Veninga, and J. Wierinza. 1995. Microscopic methods for 
counting bacteria and fungi in soil. pp. 162-173 in Alef, K., and P. Nannipieri 
(ed.s) Methods in applied soil microbiology and biochemistry, 1995. 576 pp. 
Bosatta, E. and G. Agren. 1994. Theoretical analysis of microbial biomass dynamics in 
soils. Soil Biol. Biochem. 26:143-148. 
98 
Bradley, R.L., and J.W. Fyles. 1995. Growth of paper birch {Betula papyriferd) 
seedlings increases soil available C and microbial acquisition of soil nutrients. Soil 
Biol. Biochem. 27:1565-1571.Brady, N.C., and R.R. Weil. 1996. The nature and 
properties of soils, XII. Prentice-Hall, Inc., New Jersey. 881 pp. 
Bremner, J.M., and C.S. Mulvaney. 1982. Nitrogen -total, pp. 595-624 in Page, A.L., 
R.H. Miller, and D.R. Keeney (eds.) Methods of soil analysis, chemical and 
microbiological properties, II. 1982. American Society of Agronomy, Inc., 
New York. 1159 pp. 
Cairns, M.A., and R.A. Meganck. 1994. Carbon sequestration, biological diversity and 
sustainable development; integrated forest management. Environ. Mgmt. 18:13- 
22. 
Carter, M.R. 1986. Microbial biomass as an index for tillage-induced changes in soil 
biologicl properties. Soil and Tillage Res. 7:29-40. 
Casida, L.E. 1968. Methods for the isolation and estimation of activity of soil bacteria, 
pp.97-122 in Gray, T.R.G., and D. Parkinson (eds.) The ecology of soil bacteria. 
1968. University of Toronto Press, Toronto. 681pp. 
Chatarpaul, L., D. Burgess, and O.Q. Hendrickson,. 1984. A comparison of biomass and 
nutrients removed in whole tree harvesting and conventional (stem only) 
harvestings, pp. 121-125 in Hasnain, S. (ed.) Proc. Fifth Cdn. Bioenergy R and D 
Seminar. Elsevier, London. 
Chatarpaul, L. 1987. Biological and nutritional implications of harvesting biomass; a 
Central Ontario perspective, pp. 139-145 in Stiasny, Z.Z. (ed.) 1987. Proc. Sixth 
Cdn. Bioenergy R and D Seminar. Richmond, B.C. 
Coleman, D.C., and D.A. Crossley, Jr. 1996. Fundamentals of soil ecology. Academic 
Press, Toronto. 205 pp. 
Cowling, J.E., and S.F. MacLean, Jr. 1981. Forest floor respiration in a black spruce 
taiga forest ecosystem in Alaska. Holarctic Ecology 4:229-237. 
Diaz-Ravina, M., A. Prieto, M.J. Acea, and T. Carballas. 1992. Fumigation-extraction 
method to estimate microbial biomass in heated soils. Soil Biol. Biochem. 24:259- 
264. 
Dictor, M.C., L. Tessier, and G. Soulas. 1998. Reassessment of the kec coefficient of the 
fumigation-extraction method in a soil profile. Soil Biol. Biochem. 30:119-127. 
99 
Edwards, N.T. 1982. The use of soda-lime for measuring respiration rates in terrestrial 
systems. Pedobiologia 28:35-40. 
Entry, J.A., N.M. Stark, and H. Loewenstein. 1986. Effect of timber harvesting on 
microbial biomass fluxes in a northern Rocky Mountain forest soil. Can. J. For. 
Res. 16:1076-1081. 
Feigl, B.J. G.P. Sparling, D.J. Ross, and C.C. Cerri. 1995. Soil microbial biomass in 
Amazonian soils: evaluation of methods and estimates of pool sizes. Soil Biol. 
Biochem. 27:1467-1472. 
Forster, J.C. 1995a. Determination of soil pH. pp. 55 in Alef, K. and P. Nannipieri 
(eds.) Methods in applied soil microbiology and biochemistry. 1995. Academic 
Press, Toronto. 576 pp. 
Forster, J.C. 1995b. Soil physical analysis: determination of the gravimetric water 
content and soil dry mass. pp. 105-121 in Alef, K. and P. Nannipieri (eds.) 
Methods in applied soil microbiology and biochemistry. 1995. Academic Press, 
Toronto. 576 pp. 
Foster, N.W., and I.K. Morrison. 1987. Alternate strip clearcutting in upland black 
spruce IV. Projected nutrient removals associated with harvesting. For. Chron. 
:451-456. 
Frostegard, A., A. Tunlid, and E. Baath. 1993. Phospholipid fatty acid composition, 
biomass, and activity of microbial communities from two soil types experimentally 
exposed to different heavy metals. App. Environ. Microbiol. 59:3605-3617. 
Gallardo, A., and W.H. Schlesinger. 1994. Factors limiting microbial biomass in the 
mineral soil and forest floor of a warm-temperate forest. Soil Biol. Biochem. 
26:1409-1415. 
Gordon, A.G., D.M. Morris, and N. Balakrishnam. 1993. Impacts of various levels of 
biomass removals on the structure, frinction, and productivity of black spruce 
ecosystems: research protocols. Min. Nat. Res., For. Res. Info. Paper 109. 21 pp. 
Gordon, AM., R.A. McBride, A.J. Fisken, and R.P. Voroney. 1988. Effect of landfill 
leachate spraying on soil respiration microbial biomass in a northern hardwood 
forest ecosystem. Waste Mgmt. Res. 6:141-148. 
Gordon, AM., R.E. Schlentner, and K. Van Cleve. 1987. Seasonal patterns of soil 
respiration and CO2 evolution following harvesting in the white spruce forests of 
interior Alaska. Can. J. For. Res. 17:304-310. 
100 
Haron, K., P.C. Brookes, J.M. Anderson, and Z.Z. Zakaria. 1998. Microbial biomass and 
soil organic matter dynamics in oil palm (Elaeis guineensis Jacq.) plantations. 
West Malaysia. Soil Biol. Biochem. 30:547-552. 
Hartmann, A., B. Apmus, G. Kirchhof, andM. Scholoter. 1997. Direct 
approaches for studying soil microbes, pp.279-310 in van Elsas, J.D., J.T. 
Trevors, E.M.H. Wellington (eds.) Modern Soil Microbiology, Marcel Dekker, 
New York, 1997. 683pp. 
Haynes, R.J. 1986. Mineral nitrogen in the plant-soil system. Academic Press, Toronto. 483 pp. 
Hendrickson, O.Q., L. Chatarpaul, and J.B. Robinson. 1985. Division S-7 - forest and 
range soils. Soil Sci. Soc. Am. J. 49:739-746. 
Holt, J.G., N.R. Krieg, P.A.A. Sneath, J.T. Staley, and S.T. Williams. 1994. Bergey’s 
manual of determinative bacteriology, IX. Williams & Wilkins, Baltimore. 787 
PP- 
Howard, P.J.A., and D M. Howard. 1979. Respiration of decomposing litter in relation 
" to temperature and moisture. Oikos 33:457-465. 
Inubushi, K., P.C. Brookes, and D.S. Jenkinson. 1991. Soil microbial biomass C, N and 
ninhydrin-N in aerobic and anaerobic soils measured by the fumigation-extraction 
method. Soil Biol. Biochem. 23:737-741. 
Jenkinson, D.S., and J.N. Ladd. 1981. Microbial biomass in soil: measurement and 
turnover. Soil Biol. Biochem. 5:415-471. 
Jenkinson, D.S., and D.S. Powlson. 1976. The effects of biocidal treatments on 
metabolism in soil V.A. method for measuring soil biomass. Soil Biol. Biochem. 
8:209-213. 
Kaiser, E.A., T. Mueller, R.G. Joergensen, H. Insam, and O. Heinemeyer. 1992. 
Evaluation of methods to estimate the soil microbial biomass and the relationships 
with soil texture and organic matter. Soil Biol. Biochem. 24:675-683. 
Killham, K. 1994. Soil ecology. Cambridge University Press, Great Britain. 242 pp. 
Liesack, W., P.H. Janssen, F.A. Rainey, N.L. Ward-Rainey, and E. Stackebrandt. 1997. 
Microbial diversity in soil: the need for a combined approach using molecular and 
cultivation technology, pp.375-439 in van Elsas, J.D., J.T. Trevors, E.M.H. 
Wellington (eds.) Modern Soil Microbiology, Marcel Dekker, New York, 1997. 
683pp. 
Luckai et al 2000. Eliminating negative nitrogen numbers in the estimation of microbial 
biomass using chloroform fiamigation extraction. In prep. 
Lundgren, B., and B. Soderstrom. 1983. Bactrerial numbers in a pine forest soil in 
relation to environmental factors. Soil Biol. Biochem. 15:625-630. 
Martens, R. 1987; Estimation of microbial biomass in soil by respiration method: 
importance of soil pH and flushing methods for the measurement of respired 
carbon dioxide. Soil Biol. Biochem. 19:77-81. 
Martens, R. 1995. Current methods for measuring microbial biomass C in soil: potentials 
and limitations. Biol. Fertil. Soils. 19:87-99. 
Martikainen, P.J., and A. Palojarvi. 1990. Evaluation of the fumigation-extraction 
method for the determination of microbial C an N in a range of forest soils. Soil 
Biol. Biochem. 22:797-802. 
Marumoto, T., J.P.E. Anderson, and K.H. Domsch. 1982. Decomposition of and 
^^N-labelled microbial cells in soil. Soil Biol. Biochem. 14:461-467. 
Marumoto, T. 1984. Mineralization of C and N from microbial biomass in paddy soil. 
Plant and Soil 76:165-173. 
Matthies, C., H.P. Erhard, and H.L. Drake. 1997. Effect of pH on the comparative 
cuturability of fungi and bacteria from acidic and less acidic forest soils. J. Basic 
Microbiol. 37:335-343. 
Merckx, R., A.M.A. Van der Linden, and K.V. Leuren. 1988. The extraction of 
microbial biomass components from soils, pp. 327-339 in Jenkinson, D.S., and 
K.A. Smith (eds.) Nitrogen efficiency in agricultural soils. 1988. Elsevier Applied 
Science, London. 
Meyer, W.L., and G. Vanson. 1997. Field guide to forest soils: Forestry 1537/2110. 
Course Manual. Lakehead University, Faculty of Forestry and the Forest 
Environment. 
Mireku, E. 1981. Characterization, distribution and pathogenicity of bacteria isolated 
from chlorotic white spruce. MScF. Lakehead U. 102pp. 
Morgan, J.A.W., and C. Winstanley. 1997. Microbial biomarkers, pp.331-352, in van 
Elsas, J.D., J.T. Trevors, E.M.H. Wellington (eds.) Modem Soil Microbiology, 
Marcel Dekker, New York, 1997. 683pp. 
102 
Ocio, J.A., and P.C. Brookes. 1990. An evaluation of methods for measuring the 
microbial biomass in soil following recent additions of wheat straw and the 
characterization of the biomass that develops. Soil Biol. Biochem. 22:685-694. 
Ohtonen, R., A. Munson, and D. Brand. 1992. Soil microbial community response to 
silvicultural intervention in coniferous plantation ecosystems. Ecol. Appli. 2: 363- 
375. 
Ontario Ministry of Environment. 1994. Reasons for discussion and decision-class 
environmental assessement by the MNR for timber management on crown lands in 
Ontario. Queen’s Printer for Ontario, Toronto, Ontario. 
Orchard, V.A., and F.J. Cook. 1983. Relationship between soil respiration and soil 
moisture. Soil Biol. Biochem. 15:447-453. 
Palmieri, M.J., S.L. Carito, and R.F. Meyer. 1988. Comparison of rapid NFT and API 
20E with conventional methods for identification of Gram-negative 
nonfermentative Bacilli from pharmaceuticals and cosmetics. Appl. Environ. 
Microbiol. 54:2838-2841. 
Paul, E.A., and F.E. Clark. 1989. Soil microbiology and biochemistry. Academic Press, 
Toronto. 273 pp. 
Phillips, N.J., S. Navratil, and L. Sevean. 1986. For. 3037 Forest pathology laboratory 
manual. 167 pp. Faculty of Forestry, Lakehead University. 
Pongracic, S., M.U.F. Kirschbaum, and R.J. Raison. 1997. Comparison of soda lime and 
infrared gas analysis technique for in situ measurement of forest soil respiration. 
Can. J. For. Res. 27:1890-1895. 
Prosser, J.I. 1997. Microbial processes within the soil. pp. 183-213, in van Elsas, J.D., 
J.T. Trevors, E.M.H. Wellington (eds.) Modern Soil Microbiology, Marcel 
Dekker, New York, 1997. 683pp. 
Racey, G.D., A.G. Harris, J.K. Jeglum, R.F. Foster and G.M. Wickware. 1996. 
Terrestrial and wetland ecosites of northwestern Ontario. Ont. Min. Natur. 
Resourc. Northwest Sci. and Technol. Field Guide FG-02. 94 pp. 
Raubuch, M., and F. Beese. 1998. Interaction between microbial biomass and activity 
and the soil chemical conditions and the processes of acid load in coniferous forest 
soils. Z. Pflanzenemahr. Bodenk., 161:59-65. 
103 
Reiners, W. A. 1968. Carbon dioxide from the floor of three Minnesota forests. Ecology 
49:471-483. 
Robertson, A.E., G.C. Macks, and J.D. MacLowry. 1976. Analysis of cost and accuracy 
of alternative strategies for Enterohacteriaceae identification. J. Clinical 
Microbiol. 3:421-424. 
Rowell, D.L. 1994. Soil science: methods and applications. Addison Wesley Longman 
Ltd. Essex, England. 350 pp. 
Roy, P. 1984. Bacteria in fresh and stored white spruce 2+2 nursery seedlings. BSc. 
Lakehead U. 84 pp. 
Salonius, P.O. 1983a. Effects of organic-mineral soil mixtures and increasing 
temperatures on the respiration of coniferous raw humus material. Can. J. For. 
Res. 13:102-107. 
Salonius, P.O. 1983b. Effects of air drying on the respiration of forest soil microbial 
populations. Soil Biol. Biochem. 15:199-203. 
Scheu, S., and D. Parkinson. 1994. Changes in bacterial and fungal biomass C, bacterial 
and fungal biovolume and ergosterol content after drying, remoistening and 
incubation of different layers of cool temperate forest soil. Soil Biol. Biochem. 
26:1515-1525. 
Schlentner, R.E., and K. Van Cleve. 1985. Relationships between CO2 evolution from 
soil, substrate temperature, and substrate moisture in four mature forest types in 
interior Alaska. Can. J. For. Res. 15:97-106. 
Schmidt, E.L., and L.W. Belser. 1982. Nitrifying bacteria, pp. 1027-1042 in Page, A.L., 
R.H. Miller, and D.R. Keeney (eds.) Methods of soil analysis: chemical and 
microbiological properties, II. 1982. American Society of Agronomy, Inc., New 
York. 1159 pp. 
Sims, R.A., W.D. Towill, K.A. Baldwin, and G.M. Wickware. 1989. Field guide to the 
forest ecosystem classification for Northwestern Ontario. Ont. Min. Nat. Res. 
Queens Printer of Ontario, Toronto. 191pp. 
Singh, J.S., and S.R. Gupta. 1977. Plant decomposition and soil respiration in terrestrial 
ecosystems. Botanical Rev. 43:449-528. 
Smolander, A., O. Priha, L. Paavolainen, J. Steer, and E. Malkonen. 1998. Nitrogen and 
carbon transformations before and after clear-cutting in repeatedly N-fertilized and 
limed forest soil. Soil Biol. Biochem. 30:477-490. 
Snedecor, G.W., and W.G. Cochran. 1967. Statistical method, VI. Iowa State University 
Press, Ames, Iowa. 
Sorheim, R, V.L. Torsvik, and J. Goksoyr. 1989. Phenotypical divergences between 
populations of soil bacteria isolated on different media. Microbial Ecol. 17:181. 
Sparling, G.P. 1985. The soil biomass, pp. 223-262 in Vaughn, D., and R.E. Malcolm 
(eds.) Soil organic matter and biological activity. 1985. Developments in Plant 
and Soil Science 16:223-262. 
Sparling, G.P. and A.W. West. 1988. A direct extraction method to estimate soil 
microbial C: calibration in situ using microbial respiration and labeled cells. 
Soil Biol. Biochem. 20:337-343. 
Staddon, W.J., L.C. Duchesne, and J.T. Trevors. 1996. Conservation of forest soil 
microbial diversity: the impact of fire and research needs. Environ. Rev. 4: 267- 
275. 
Tate, K.R., D.J. Ross, and C.W. Feltham. 1988. A direct extraction method to estimate 
soil microbial C: effects of experimental variables and some different calibration 
procedures. Soil Biol. Biochem. 20:329-335. 
Tate, R.L. 1991. Microbial biomass measurement in acidic soil: effect of fungal :bacterial 
activity ratios and of soil amendment. Soil Sci. 152:220-225. 
Thibodeau, L., P. Raymond, C. Camire and A.D. Munson. 2000. Impact of pre- 
commercial thinning in balsam fir stands on soil nitrogen dynamics, microbial 
biomass, decomposition and foliar nutrition. Cdn. J. For. Res. 30: 229-238. 
Uchida, M., T. Nakatsubo, T. Horikoshi, and K. Nakane. 1998. Contribution of micro- 
organisms to the carbon dynamics in black spruce (Picea mariana) forest soil in 
Canada. Ecological Res. 13:17-26. 
Vance, E.D., P.C. Brookes, and D.S. Jenkinson. 1987. An extraction method for 
measuring soil microbial biomass C. Soil Biol. Biochem. 19:703-707. 
Velleman, P.F. 1997. DataDesk, VI, Statistics Guide: learning and using statistics 
procedures. Data Description, Inc., New York. 
105 
Voroney, R.P. 1985. Measurement of the flows of carbon and nitrogen through the soil 
microbial biomass during decomposition. Cdn. Soc. Microbiol. 35^ Ann. Meet., 
Halifax, N.S. Conference Abstracts, p. 58. 
Voroney, R.P., J.P.Winter, and R.P. Beyaert. 1993. Soil microbial biomass C and N. 
pp.277-285 in Carter, M.R. (ed.) Soil sampling and methods of analysis, Lewis 
Publishing, 1993. 
Wardle, P.A. and D. Parkinson. 1990. Response of the soil microbial biomass to glucose, 
and selective inhibitors, across a soil moisture gradient. Soil Biol. Biochem. 
22:825-834. 
Wardle, D.A., and A. Ghani. 1995. Why is the strength of relationships between pairs of 
methods for estimating soil microbial biomass often so variable? Soil Biol. 
Biochem. 27:821-828. 
Washington, J.A., P.K.W. Yu, and W.J. Martin. 1971. Evaluation of accuracy of 
multitest micromethod system for identification of Enterobacteriaceae. Appl. 
Microbiol. 22:267-269. 
Weber, M.G. 1985. Forest soil respiration in eastern Ontario jack pine ecosystems. Can. 
J. For. Res. 15:1069-1073. 
Weber, M.G. 1990. Forest soil respiration after cutting and burning in immature aspen 
ecosystems. For. Ecol. Manage. 31:1-14. 
Wiensczyk, A. 1992. A brief review of the issues surrounding full-tree harvesting. Ont. 
Min. Nat. Resour., N.W. On. For. Tech. Dev. Unit, ThBay, On, Tech. Note TN- 
13. 
Whipps, J.M. 1997. Ecological considerations involved in commercial development of 
biological control agents for soil-borne diseases, pp. 525-546 in van Elsas, J.D., 
J.T. Trevors, E.M.H. Wellington (eds.) Modem Soil Microbiology, Marcel 
Dekker, New York, 1997. 683pp. 
Wong, P.T.W., and D.M. Griffin. 1976a. Bacterial movement at high matric potentials: 
in artificial and natural soils. Soil Biol. Biochem. 8:215-218. 
Wong, P.T.W., and D.M. Griffin. 1976b. Bacterial movement at high matric potentials: 
in fungal colonies. Soil Biol. Biochem. 8:219-223. 
106 
Zak, J.C., M.R. Willig, D.L. Moorhead, and H.G. Wildman. 1994. Functional diversity 
of microbial communities: a quantitative approach. Soil Biol. Biochem. 26: 1101- 
1103. 
Zar, J.H. 1996. Biostatistical analysis, III. Prentice Hall, New Jersey. 620 pp. 
APPENDIX I 
FULL DATA SET FOR THE ORGANIC AND MINERAL SOIL HORIZONS 
Appendix la. Complete data set for the Organic soil horizon. 
Harvest Month of MBc MBN CVN H' Concentration Total Phosphorus Total Nitrogen Water Organic Matter 
Treatment Sampling Oig/ g soil) O^g/ g soil) Ratio (M) (mg/Kg) (mg/Kg) Content (%) (%) 
Control May 14889 1790 8.3 187 1399 46 
Control May 8385 981 8.5 123 1034 49 
Control May 10087 724 13.9 0.0007585 125 981 69 95 
Tree-length May 5256 1702 3.1 0.0000229 141 1151 68 91 
Tree-length May 5618 1115 5.0 0.0000316 137 1157 71 90 
Tree-length May 7094 1326 5.3 0.0000407 179 nil 75 78 
Full-tree May 8086 1782 4.5 0.0000537 156 1151 64 93 
Full-tree May 164 1224 35 
Full-tree May 6693 1136 5.9 0.0000257 164 1136 80 93 
Control June 7041 1025 6.9 0.0001862 127 1032 65 79 
Control June 8383 1425 5.9 0.0000912 220 1168 62 89 
Control June 7932 1963 4.0 0.0000309 137 1200 73 85 
Tree-length June 5416 753 7.2 0.0000380 137 1157 72 88 
Tree-length June 4647 845 5.5 0.0000575 166 1106 69 79 
Tree-length June 6797 1239 5.5 0.0000257 148 1209 76 77 
Full-tree June 6211 1013 6.1 0.0000157 117 1369 80 87 
Full-tree June 4896 691 7.1 0.0000478 129 1128 69 76 
Full-tree June 6114 1118 5.5 0.0000117 153 1200 78 86 
Control July 4956 734 6.8 0.0001000 158 1139 68 80 
Control July 6240 1075 5.8 0.0000776 123 962 63 59 
Control July 9268 1148 8.1 151 1117 70 
Tree-length July 7229 1213 6.0 0.0000251 179 1471 78 
Tree-length July 7179 1218 5.9 0.0000218 214 1330 74 
Tree-length July 8389 1559 5.4 0.0000724 151 1139 78 
Full-tree July 6696 1321 5.1 80 
Full-tree July 6217 1029 6.0 0.0000537 1057 63 
Full-tree July 4081 1441 2.8 0.0000117 1471 80 
Control August 5347 1644 3.3 0.0001174 169 1202 68 
Control August 6004 667 9.0 137 1018 73 
Control August 5582 1233 4.5 150 1078 70 
Tree-length August 6270 1373 4.6 0.0000870 133 1180 74 
Tree-length August 6218 945 6.6 0.000575 160 1174 76 
Tree-length August 6483 974 6.7 169 1089 73 
Full-tree August 6666 2610 2.6 0.0000288 171 1197 78 
Full-tree August 5066 1364 3.7 0.0000524 143 1253 63 
Full-tree August 6603 955 6.9 0.0000229 134 994 77 
Appendix Ib. Complete data set for the Mineral soil horizon. 
Harvest Month of MBc MB„ C/N H* Concentration Total Phosphorus Total Nitrogen Water Organic 
Treatment Sampling Cug/ g soil) Cug/ g soil) Ratio (M) (mg/Kg) (mg/Kg) Content (%) Matter (%) 
Control May 1449 212 6.8 0.0000436 35 134 59 9 
Control May 1616 203 7.9 0.0002691 39 312 70 30 
Control May 414 77 5.4 0.0001071 35 231 65 21 
Tree-length May 528 76 7.0 0.0000085 32 153 40 11 
Tree-length May 358 87 4.1 0.0000177 43 142 68 10 
Tree-length May 437 91 4.8 0.0001479 40 . 126 74 11 
Full-tree May 851 59 14.5 0.0000288 34 168 74 13 
Full-tree May 1087 298 3.6 0.0000457 35 192 47 19 
Full-tree May 647 84 7.6 0.0000177 40 196 62 16 
Control June 1118 113 9.9 0.0000338 30 123 49 9 
Control June 1033 172 6.0 0.0000025 26 161 41 14 
Control June 1861 190 9.8 0.0000089 34 288 35 21 
Tree-length June 371 53 6.9 0.0000354 25 108 34 10 
Tree-length Jime 458 105 4.3 0.0000575 34 135 40 8 
Tree-length Jime 237 75 3.2 0.0000602 21 121 29 10 
Full-tree June 287 188 1.5 0.0000138 35 152 45 10 
Full-tree June 300 114 2.6 0.0000407 29 160 19 19 
Full-tree June 626 121 5.2 0.0000977 37 165 30 13 
Control July 482 40 11.9 0.0000208 37 121 19 9 
Control July 748 75 10.0 0.0000549 30 113 22 7 
Control July 605 48 12.5 0.0000275 25 121 19 9 
Tree-length July 981 99 9.9 0.0000213 24 143 47 10 
Tree-length July 443 123 3.6 0.0000165 52 257 36 22 
Tree-length July 378 67 5.6 0.0000281 27 27 31 7 
Full-tree July 486 89 5.5 0.0000060 31 114 43 8 
Full-tree July 362 48 7.5 0.0000323 36 124 27 8 
Full-tree July 383 111 3.4 0.0000213 35 158 34 11 
Control August 239 26 9.2 0.0000177 35 111 22 8 
Control August 296 76 3.9 0.0000549 41 170 27 13 
Control August 424 57 7.4 0.0001023 21 120 20 8 
Tree-length August 502 52 9.6 0.0000199 28 123 40 10 
Tree-length Ai^^ust 291 40 7.4 0.0001513 48 182 35 13 
Tree-length August 215 33 6.5 0.0000251 30 115 30 7 
Full-tree August 304 82 3.7 0.0000097 30 115 33 8 
Full-tree August 148 34 4.3 0.0000218 30 90 24 7 
Full-tree August 270 48 5.6 0.0000229 27 116 28 9 
APPENDIX n 
COMPLETE DATA SET FOR THE SOIL RESPIRATION STUDY 
Appendix II: Complete data set for the Soil Respiration study. 
Treatment C Efflux (09/05/98) C Efflux (09/12/98) 
(gCOj/hr/m^) (gCOj/hr/m") 
Control 0.0118 0.0121 
0.0159 0.0165 
0.0139 0.0141 
0.0138 0.0143 
0.0138 0.0141 
0.0127 0.0131 
0.0128 0.0136 
0.0094 0.0103 
0.0111 0.0116 
0.0117 0.0126 
0.0118 0.0123 
0.0107 0.0112 
0.0122 0.0165 
0.0121 0.0132 
0.0112 0.0114 
Mean 0.0123 0.0131 
Tree-length 0.0128 0.0133 
0.0100 0.0137 
0.0117 0.0128 
0.0126 0.0133 
0.0109 0.0152 
0.0139 0.0116 
0.0128 0.0084 
0.0129 0.0117 
0.0090 0.0099 
0.0095 0.0113 
0.0095 0.0105 
0.0088 0.0094 
0.0108 0.0109 
0.0069 0.0074 
0.0107 0.0118 
Mean 0.0109 0.0114 
Full-tree 0.0103 0.0103 
0.0119 0.0117 
0.0126 0.0133 
0.0094 0.0100 
0.0112 0.0111 
0.0114 0.0121 
0.0082 0.0078 
0.0069 0.0108 
0.0105 0.0116 
0.0092 0.0093 
0.0117 0.0123 
0.0107 0.0121 
0.0117 0.0117 
0.0084 0.0082 
0.0075 0.0079 
Mean 0.0101 0.0107