éb

.53

I?

l».

‘P
$193!:

«:21: 2..
idnnu. .

§

 

é‘a

I
o.

 

5‘}; .

l
1

 

 

' 235;:
.
chi

I

.gé‘
a
«:27

‘E

as- ,
EL“;
3

‘

 

vii .
:5... . .
fared... :.....o.W0W.\)..
ӣ23.

 

1
.. .iv;
.2

J 1!.
x. Penna; .
fish. .

. 1"!

ppnn .

{:3}...

I. 1.4!“.

fat. I: “viisuu 3.
XIV . v ‘41?
‘ t. 1.! « or:
{IL 3':
33F .unuuuv (i.

an Iv!!! . 1‘ . .
“H.510... H >r .Vuumh-ih...

 

 

   

32.44.. run. I.» 1x. .

 

 
 
 
 

. .1 . ‘, an: :1. .
.‘ , 1.} 3.1.“ ._ £2..._...§§m7 fi.z§..€r 935...: “fiaflwfixfi..fuf

 

 

" 1mm

1
31004

SbS'S’l’JOJ”

 

LléfiARY
Michigan State
University

 

 

 

This is to certify that the
thesis entitled

The Recovery of Ecosystem Processes After Wildfire In
Michigan Jack Pine Forests

presented by

Zhanna Abramovsky

has been accepted towards fulfillment
of the requirements for the

Master of degree in Forestry & Ecology,
Science Evolutionary Biology and
Behavior

 

 

944%”ng

Major Professor’ 5 'Signature
/ ’L- I Z — 2003

Date

MSU is an Affirmative Action/Equal Opportunity Institution

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDuepGS-p. 1 5

 

THE RECOVERY OF ECOSYSTEM PROCESSES AFTER WILDFIRE
IN MICHIGAN JACK PINE FORESTS

By

Zhanna Abramovsky

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Departments of Forestry & Ecology, Evolutionary Biology and Behavior

2003

ABSTRACT

THE RECOVERY OF ECOSYSTEM PROCESSES AFTER WILDFIRE
IN MICHIGAN JACK PINE FORESTS

By

Zhanna Abramovsky

I investigated the recovery of ecosystem processes after wildfire in
Michigan jack pine (Pinus banksiana) forests using a chronosequence of 11
wildfire-regenerated stands spanning 72 years. The objective of this study was to
characterize recovery patterns of soil nutrients, soil carbon (C) and nitrogen (N)
mineralization, as well as determine the mechanisms that drive those patterns.

Total N mineralization spiked immediately following wildfire, decreased to
minimum values after about 10-15 years following wildfire, and then increased,
reaching a steady state after about 40 years. Soil respiration rates declined
immediately after wildfire, increased after about 10-15 years, followed by a
decreasing trend after 30 years.

In Michigan jack pine forests, accumulation of the forest floor largely
regulates the recovery of N mineralization. The successional pattern of annual N
mineralization was driven by rapid turnover of N in the mineral soil immediately
following wildfire, followed by a gradual accrual of a slow-cycling pool of N in
forest floor as stands matured. Soil respiration was most likely driven by
heterotrophic microbial respiration early in succession and later by both root
respiration and microbial respiration. All measured ecosystem processes

recovered to predisturbance levels within 40 years of wildfire.

ACKNOWLEDGEMENTS

My sincere gratitude to my advisor, David E. Rothstein, who always
provided me with guidance and freedom to explore many paths. I thank him for
his encouragement, commitment to the project, his patience and most
importantly, for his inspiration and creativity.

I thank Bridgette Gunther, Ali Buell, Mindy Woolman, Lynn Holcomb and
Gerald Milleski for their assistance in the field and laboratory. Without their help
and commitment, the extent of this project would never be realized.

I am very grateful to the Michigan Department of Natural Resources, the
US. Forest Service, the US. Fish and Wildlife Service and Camp Grayling for
allowing access to the sites. I also thank the Mio Ranger Station for providing
housing during fieldwork.

I thank my committee members, Mike Walters and Mike Klug, for their
intellectual contribution to the project, their patience and encouragement.

I thank my fellow graduate students in the Forestry department and the
EEBB program for their constant support and guidance. It has been a sheer joy
working with you.

I would like to thank my remarkable family for always encouraging me to
follow my heart and for being inquisitive about my interests. My sincere thanks to

Leonid for inspiring me to study this specific branch of ecology.

iii

TABLE OF CONTENTS

LIST OF TABLES ................................................................................. VI
LIST OF FIGURES .............................................................................. VII
CHAPTER 1: Literature Review
EFFECTS OF WILDFIRE ON FOREST ECOSYSTEMS ................................ 1
INTRODUCTION .......................................................................... 1
NUTRIENT CYCLING IN FOREST ECOSYSTEM ................................ 1
Nitrogen ............................................................................. 2
Carbon .............................................................................. 4
EFFECTS OF FIRE ON SOIL .......................................................... 6
Soil Physical System ............................................................ 7
Soil Chemical System ........................................................... 9
Soil Biological System ......................................................... 1O
DISTURBANCE RECOVERY AND FIRE RETURN INTERVAL ............13
JACK PINE ECOSYSTEM ............................................................ 14
MANAGEMENT IMPLICATIONS ...... , .............................................. 15
RATIONALE .............................................................................. 16
CHAPTER 2
THE RECOVERY OF ECOSYSTEM PROCESSES AFTER WILDFIRE IN
JACK PINE FORESTS ......................................................................... 17
ABSTRACT ............................................................................... 17
INTRODUCTION ........................................................................ 18
MATERIALS AND METHODS ....................................................... 23
Site description and Selection ............................................... 23
Net Nitrogen Mineralization .................................................. 24
Soil Respiration ................................................................. 26
Physical and Chemical Properties .......................................... 27
Soil Microclimate ................................................................ 28
Vegetation ........................................................................ 29
Statistical Analysis .............................................................. 30
RESULTS ................................................................................. 31
Soil Carbon and Nitrogen Pools ............................................. 31
Net Nitrogen Mineralization .................................................. 32
Soil Respiration .................................................................. 34
Other Soil Chemical Characteristics ....................................... 35
Soil Microclimate ................................................................ 35
Vegetation ........................................................................ 36
DISCUSSION ............................................................................. 36
Net Nitrogen Mineralization .................................................. 36
Soil Respiration .................................................................. 41

iv

Vegetation ........................................................................ 46

CONCLUSIONS ......................................................................... 47
CHAPTER 3
SUMMARY AND CONCLUSIONS ........................................................... 49
APPENDICES ..................................................................................... 50
Appendix A: Tables ...................................................................... 50
Appendix B: Figures ..................................................................... 56
Appendix B: Additional Methods ...................................................... 79
REFERENCES ................................................................................... 84

LIST OF TABLES

Table 1 .............................................................................................. 51
ANOVA results for monthly total N mineralization rates.

Table 2 .............................................................................................. 51
ANOVA results for monthly soil respiration rates.

Table 3 .............................................................................................. 51
ANOVA results for monthly soil moisture.

Table 4 .............................................................................................. 52
Age, location, vegetation and soil properties of chronosequence sites.

Table 5 .............................................................................................. 53
ANOVA results for potential nitrogen mineralization.

Table 6 ............................................................................................. 53
ANOVA results for potential carbon mineralization.

Table 7 ............................................. 54
ANOVA results for relative nitrogen mineralization.

Table 8 .............................................................................................. 54
ANOVA results for relative carbon mineralization.

Table 9 .............................................................................................. 55
ANOVA results for potential nitrogen immobilization.

vi

LIST OF FIGURES

Figure 1 ............................................................................................. 57
Accumulation of forest floor as a function of stand age after a clear-cut
(Covington 1981).

Figure 2 ............................................................................................. 57
Site locations and dates of stand origin of Pinus banksiana chronosequence.

Figure 3 ............................................................................................. 58
Forest floor (FF) C content (A), N content (B) and C:N ratio as a function of stand
age.

Figure 4 ............................................................................................. 59
Soil C content (A), N content (B) and C:N ratio (C) as a function of stand age.

Figure 5 ............................................................................................ 60
Total C content (A), N content (B) and C:N ratio (C) as a function of stand age.

Figure 6 ............................................................................................. 61
Total annual N mineralization as a function of stand age.

Figure 7 ............................................................................................. 62
Annual and relativized N mineralization in forest floor (A, B) and soil (C, D) as a
function of stand age.

Figure 8 ............................................................................................. 63
Percent of total nitrogen mineralized from forest floor as a function of stand age.

Figure 9 ............................................................................................. 64
Soil respiration for the growing season as a function of stand age.

Figure 10 ........................................................................................... 65
Monthly soil respiration rates as a function of stand age.

Figure 1 1 ........................................................................................... 66
Total exchangeable bases as a function of stand age.

Figure 12 ........................................................................................... 67
Soil (A) and forest floor (B) pH as a function of stand age.

Figure 13 ........................................................................................... 68
Total number of degree-days in the growing season as a function of stand age.

vii

Figure 14 ........................................................................................... 69
Seasonal gravimetric percent moisture in forest floor (A) and soil (B) as a
function of stand age.

Figure 15 ........................................................................................... 70
Average moisture tension data at selected stands at 5 different sampling periods.

Figure 16 ........................................................................................... 71
Jack pine biomass as a function of stand age.

Figure 17 ........................................................................................... 71
Live understory biomass as a function of stand age

Figure 18 .......................................................................................... 72
Biomass of selected understory vegetation categories as a function of stand age.

Figure 19 .......................................................................................... 73
Monthly microbial biomass in the soil as a function of stand age.

Figure 20 .......................................................................................... 74
Available phosphorus as a function of standage.

Figure 21 .......................................................................................... 75
Potential carbon mineralization as a function of temperature for four moisture
treatments and four stands.

Figure 22 .......................................................................................... 76
Potential nitrogen mineralization as a function of temperature for four moisture
treatments and four stands.

Figure 23 .......................................................................................... 77

Potential nitrogen immobilization as a function of temperature for four moisture
treatments and four stands.

viii

CHAPTER 1: Literature Review

EFFECTS OF WILDFIRE ON FOREST ECOSYSTEMS

INTRODUCTION

Understanding ecosystem dynamics is essential for predicting the response of
forests to climate change and sustaining production of commercial forests.
Particular attention has been given to nitrogen (N) and carbon (C) cycles
because N is often the most limiting nutrient to forest productivity and C plays a
vital role in global warming. Mechanisms controlling soil C cycling are of
particular interest because it has been estimated that, on a global basis, soils
contain 1.6 times as much C as the atmosphere (Adams et al. 1990). Because C
and N cycles are tightly linked, long-term changes in C storage have the potential
to influence soil fertility, forest productivity and atmospheric carbon dioxide
concentrations. Disturbances such as wildfire can greatly influence C and N
dynamics, yet their effects on C and N cycles are not well understood.
Understanding the effects of wildfire on ecosystem dynamics in forests where
wildfire is an inherent ecosystem property is essential in order to manage those

forests and determine their role in global climate change.

NUTRIENT CYCLING IN FOREST ECOSYSTEMS
Researchers know a great deal more about the distribution of nutrients in forest

ecosystems than how the nutrients are transferred from one nutrient pool to

another. Forest floor1 and mineral soil often contain the largest pools of C and N.
In temperate forests, mineral soil and forest floor account for approximately 60
and 95% of total ecosystem C (Cole and Rapp 1981; Birdsey 1990; Grigal and
Ohmann 1992) and N (Cole and Rapp 1981 ), respectively. Similar estimates
have been reported for boreal forests (Apps et al. 1993; Morris and Miller 1994).
The majority of soil organic matter (SOM) is recalcitrant and resistant to microbial
degradation (Smith and Paul 1990). On an annual basis, soil microorganisms
liberate only ~10% of N bound in organic matter (calculated from Nadelhoffer et
al. 1983, Zak and Grigal 1991). Net N mineralization is fundamental in regulating
forest productivity (Keeney 1980, Pastor et al. 1984, Zak et al. 1989) because N

is the most limiting nutrient to plant productivity.

Nitrogen

Nitrogen availability is determined by N mineralization, the conversion of
organically bound N to inorganic forms by soil microorganisms. The amount of N
mineralized in a given soil volume depends in part on the total amount of organic
N present. In temperate and boreal forests, only a small fraction of total organic
N is mineralized, suggesting that other factors are also responsible for N
mineralization, such as microbial activity. Microbial activity is controlled by soil
temperature and moisture, as well as the chemical composition of organic matter

(Swift et al. 1979).

 

‘ Forest Floor is composed of freshly cast (Oi), partly decomposed (Oe), and fully decomposed
(Oa) vegetative material on the soil surface.

The amount of soil organic matter is regulated by a balance between
aboveground and belowground production of plant litter and decomposition of
that material by soil microorganisms. When primary production exceeds
decomposition, organic matter accumulates in the soil and forest floor. Because
plant nutritional needs are supplied almost entirely by internal cycling rather than
external sources in most forested ecosystems (Bonnann and Likens 1967; Cole
and Rapp 1981; Johnson and Van Hook 1989), nutrients locked up in
undecomposed organic matter may have negative effects on forest productivity.
This is especially evident in boreal forest ecosystems (Bonnann and Sidle 1990;
Pastor et al. 1987; Van Cleve and Viereck 1981; Van Cleve et al. 1983) where
forest productivity declines as stands mature. In boreal forests of Alaska, forest
productivity and N mineralization decline with succession (Van Cleve and
Nooman 1975; Van Cleve and Viereck 1981). The decline in N mineralization is
attributed to establishment of coniferous species. Coniferous litter has been
found to reduce soil N availability because of its high lignin and low N content
(Pastor at al. 1987). Cold temperatures and high moisture may also contribute to
reduced N mineralization rates.

In boreal coniferous forests, a large proportion of N capital is contained in
the forest floor (Foster and Morrison 1976, Weetman and Algar 1983). Wildfire is
an integral ecosystem process in these forests because it directly mineralizes a
significant fraction of the forest floor and releases the bound nutrients. The
conditions that prevail after wildfire, such as warmer soil temperatures and higher

cation availability (Neary et al. 1999), favor higher rates of nutrient cycling.

Because N cycling and availability often increase in boreal forests immediately

following wildfire, it is considered a rejuvenating process (T amm 1991 ).

Carbon

In addition to N, C is also sequestered in organic matter. The major difference
between C and N is that the majority of C in the ecosystem is supplied externally
through primary production and returned to the atmosphere through soil
respiration. Soil respiration is defined as the soil surface flux of carbon dioxide
(00;) produced by soil microorganisms and roots. Soil respiration is a major flux
in the global C cycle, second in magnitude to gross primary productivity (GPP)
(Houghton and Woodwell 1989). Rates of soil respiration have been measured in
a variety of ecosystems to examine nutrient and C cycling, microbial activity and
root dynamics.

Soil respiration is primarily controlled by soil temperature (Edwards 1975, for
reviews see Singh and Gupta 1977, Raich and Schlesinger 1992) and moisture
(Edwards 1975, Linn and Doran 1984, Raich and Schlesinger 1992). However,
there is no consensus on the exact nature of the relationship between
temperature, moisture and soil respiration (Lloyd and Taylor 1994). Because
microclimate varies by vegetation type and time of year, C02 fluxes also differ
among biomes and vary during the growing season (Raich and Schlesinger
1992). Other factors that influence soil respiration include soil pH, substrate

(Katznelson and Stephenson 1956), rates of C inputs to soils (Trumbore et al.

1995), diffusivity (Davidson and Trumbore 1995) and organic matter (Gaarder
1957)

In recent decades, interest in the factors that control soil respiration has
grown dramatically. Global climate models predict temperature increases due to
the increasing CO2 emissions into the atmosphere, potentially leading to a
greenhouse effect in the near future (Houghton et al. 1990). Precipitation
patterns are also expected to change and therefore may alter soil respiration
rates. The relationship between temperature, moisture and soil respiration
suggests increasing rates of soil respiration, which will likely lead to a positive
feedback to the greenhouse effect. Land-use changes and different ecosystem
management techniques, such as fertilization and forest harvesting may also
lead to changes in soil respiration (Raich and Schlesinger 1992) and C storage.

The role of boreal forests in the global C cycle has been of particular
interest because this ecosystem stores approximately 800 Gt of C (Apps et al.
1993). This amount of C is large enough that even small changes in C fluxes
may have significant impacts on climate (Nalder and Wein 1999). Some
scientists have proposed that boreal forests are a net C sink (D’Arrigo et al.
1987; Myneni et al. 2001 ). Whether or not this ecosystem is a long-term sink for
C is uncertain. Because wildfire is an integral component of boreal forests,
understanding the effect of fire on C dynamics is vital in determining whether this
ecosystem can serve as a long-term sink for C.

The immediate effect of fire on the C cycle is the combustion of biomass

during the burning process, releasing C02 into the atmosphere. Fire also

converts part of the biomass into charcoal, which is an inert form of C that does
not break down via decomposition. Carbon is returned to the ecosystem through
the regrowth of vegetation. The net exchange of C between the atmosphere and
terrestrial ecosystems after wildfire is determined by examining changes in C

storage (live and dead biomass) and C losses (soil respiration) after wildfire.

EFFECTS OF FIRE 0N SOIL .
Soil is an integral component of ecosystem sustainability. Soil supplies
mineralizable nutrients for plant growth, serves as a habitat for soil
microorganisms, provides a protective layer against erosion, helps to regulate
soil temperature and moisture, and serves as a significant sink and source of C.
The overall affects of fire on the soil system are complex because all of the
components of the soil system interact to control the rate of recovery of
ecosystem processes (Dunn and DeBano 1977; DeBano et al. 1998). The
recovery of soil processes after a disturbance depends on SOM pools, physical
and chemical soil properties, SOM quality and soil microbial community
composition (Neary et al. 1999). Burning of SOM during a wildfire may result in a
decrease in total nutrient and C pools through leaching, runoff, volatilization and
erosion. However, burning of SOM can also lead to an increase in available
pools of N by enhancing mineralization and nitrification (Neary et al. 1999)
through a changing microclimate following a burn. Because temperature and

moisture regulate many ecosystem processes, such as microbial activity, soil

microclimate can be the most critical driving force in ecosystems (Swift et al.
1979)

In order to understand how fire-induced changes in the soil system affect soil
processes, we need to understand how the physical, chemical and biological soil
systems interact, which effects are transient and which are long-lasting, as well
as what mechanisms control the rate of recovery of the soil system. The physical
system includes soil microclimate, soil bulk density, and water-holding capacity.
Soil organic matter content and quality, pH, and nutrient availability are
components of the soil chemical system. The biological soil system includes

microorganisms, soil fauna and plant roots.

Soil Physical System
Fire affects soil by destroying organic matter, which leads to changes in soil
properties such as bulk density, porosity and water-holding capacity. The
magnitude of these changes largely depends on fire severity, proportions of
overstory and understory vegetation burned, forest floor consumption, heating of
the soil, proportion of area burned, and the fire return interval (Wells et al. 1979).
In general, soil temperatures can remain elevated for a few minutes to an
hour during prescribed fires, and for several days after a severe fire (DeBano et
al. 1998). If the entire organic horizon is consumed during a fire, soil
temperatures can remain elevated above normal for months to years due to solar

radiation heating exposed mineral soil (Neary et al. 1999).

Decrease or removal of the overstory following wildfire allows more solar
radiation to reach the soil surface and heat the soil. Blackening of the soil surface
as a result of incomplete combustion of organic matter produces additional heat
absorption by decreasing soil albedo (Christensen and Muller 1975). Removal of
the canopy and increased solar heating may also lead to greater evaporation
rates and decreased soil moisture availability.

Fire also affects soil moisture availability by changing the structure of the
soil. The structure of the soil in the upper horizons results from the aggregation of
mineral soil particles by organic matter. The aggregation of individual mineral
particles in the soil enhances porosity. Thus, when a wildfire burns soil organic
matter, soil structure can be destroyed, affecting total porosity and pore size
distribution in the surface horizons of the soil. This can lead to a reduction in
infiltration rates and an increase in overland flow and therefore reduced soil
water availability (DeBano et al. 1998).

Alteration or destruction of soil organic matter can further decrease
infiltration rates, and hence decrease soil moisture availability, by making
particles water-repellent. Hydrophobic organic compounds that form in high-
intensity fires cause water repellency by coating soil aggregates and preventing

soil wetting. This condition forms when soil temperatures rise above 176°C

during a wildfire and destroyed at temperatures above 288°C (DeBano 1981).

Soil Chemical System

Fire can reduce or completely destroy soil organic matter depending on fire
severity and soil heating. The destruction of organic matter by wildfire begins at
low temperatures, starting at 200°C, and is complete at 500°C (Debano et al.
1998). During combustion of organic matter, nutrients essential to plant and
microbial metabolism can be lost through volatilization, ash convection,
mineralization, erosion, runoff and leaching (Neary et al. 1999).

Direct loss of nutrients to the atmosphere through volatilization is
temperature dependant. Nitrogen losses begin at 200°C and over half the N in
organic matter can be volatilized at temperatures above 500°C. Other nutrients
such as sulfur (S), potassium (K), phosphorus (P), sodium (Na), magnesium (Mg)
and calcium (Ca) are lost at temperatures above 800°C, 760°C, 774°C, 880°C,
1107°C, and 1240°C, respectively (Weast 1988).

Total N decreases after wildfire, however, soil N can be made more
available after a low-intensity fire due to conversion of organic nitrogen to
inorganic forms such as ammonium (NHI) and nitrate (N03’) (Picket and White
1985). However, high-intensity fires can cause large losses of nitrogen through
consumption or volatilization of NH4". Furthermore, N03' can be lost through
leaching, denitrification, or overland flow (Neary et al. 1999). Nutrient-rich ash
material remaining after fire can be lost by convection in a smoke column or
surface wind transport (Neary et al. 1999). Cations from ash can also be easily
lost through leaching, especially in acidic soils such as Spodosols.

Wildfire produces a layer of charcoal from woody vegetation. Charcoal is

potentially important in forests where plants belong to the Ericaceae family are
present (Wardle et al. 1998, Pietikainen et al. 2000). Ericaceous plants produce
secondary metabolites, such as phenolics, which retard nutrient cycling, tree
seedling growth (Zackrisson et al. 1997), mycorrhizal functioning (Nilsson et al.
1993) and soil biological processes (Wardle and Lavelle 1997). Zackrisson et al.
(1996) demonstrated that charcoal could adsorb significant amounts of
phenolics, resulting in their deactivation. Furthermore, this study showed that
charcoal could maintain high sorptive capacity for about a century after wildfire.
The study also showed that experimental heating could reactivate charcoal at

temperatures above 450°C.

Soil Biological System

The least understood component of soil processes is the microbial community.
Soil microorganisms play a vital role in soil fertility and primary production by
driving organic matter turnover through the processes of mineralization and
immobilization. Soil microbes transform organic matter to inorganic forms,
making nutrients available for plant uptake. Microbial composition is related to
their function in soil because taxonomic groups of soil organisms vary in their
function in soil processes. Wildfire has direct effects on microbial community
composition (Vazquez et al. 1993) because heating tolerance differs between
fungi and bacteria (Wells 1979), but our understanding of the significance of
those changes on ecosystem processes is lacking. Sustainable forest

management will require an understanding of the function of different soil

10

taxonomic groups and their role in the recovery of soil and ecosystem processes
after disturbance.

Soil microbial activity can be limited by soil microclimate, as well as
substrate quality and quantity. Conifers produce poor-quality litter that has high
carbon: nitrogen ratios and is not easily decomposed (Pastor et al. 1987). Fire
can modify decomposition rates by removing organic substrate, changing the
chemical composition of the remaining substrate, and by altering the physical,
chemical and biological characteristics of the soil environment (Neary et al.
1997). High temperatures during fire can destroy or damage microbes, whereas
higher than normal temperatures associated with changing microsite climate may
temporarily enhance microbial activity. Soil desiccation as a result of decreased
infiltration rates may also inhibit the ability of microbes to recover after fire and
recolonize the site (Neary et al. 1999).

Soil microorganisms are most abundant in organic horizons and the top 1-
2 cm of mineral soil, where heating from wildfire will have the greatest effect. Soil
heating can have damaging or lethal effects on soil organisms depending on fire
severity. Fungi and bacteria vary in their tolerance of heating. The lethal
temperature for bacteria is 210°C in dry soil and 110°C in wet soil, whereas the
lethal temperature for fungi in dry and wet soil is 155°C and 100°C, respectively
(Dunn and DeBano 1977). Spores and other resting forms of microbes can
tolerate higher temperatures than actively growing microbes.

There are a wide variety of microbial responses to fire, even in the same

ecosystem. The composition of recolonizing soil microbes can be different after

11

wildfire than before wildfire due to differential responses of soil microbes to
heating temperatures. Vazquez et al. (1993) reported findings from an Atlantic
Pinus pinaster forest and observed a 16-fold increase in aerobic heterotrophic
bacteria, a 2-fold increase in acidophilic bacteria, and a decrease in
cyanobacteria, fungi and algae one month after wildfire. Overall, microbial
populations increased 25-fold, but returned to pre-fire abundance within one year
of burning. Increases in microbial populations may have resulted from improved
fire-induced substrate quality, more favorable environmental conditions or a
combination of the two factors. Zackrisson et al. (1996) demonstrated that
artificially made charcoal could enhance microbial biomass in humus. Therefore,
severely burned sites, where charcoal forms, may be an important habitat for
recolonizing microorganisms. Fonturbel et al. (1995) found no effect on microbial
abundance after a prescribed burning treatment in a Pinus pinaster forest. The
effect of fire may also depend on the frequency of burning. Hossain et al. (1995)
found that burning 2-3 times per year reduced microbial biomass, but biomass
increased at 7-year burning intervals.

The effects of fire on soil microorganisms vary widely. Overall, bacteria
seem to be favored over fungi, but the effect is temporary (Vazquez et al. 1993).
The response of soil microbes to fire depends on fire severity, charcoal presence

and activation, changes in microclimate, and fire frequency.

12

DISTURBANCE RECOVERY AND FIRE RETURN INTERVAL

Frequency of disturbance may play a significant role in whether that
disturbance has a positive or a negative effect on ecosystem processes. With
increasing disturbance frequency, a greater proportion of the forest is found in
younger age classes. The question to consider is whether those forests have
recovered from the previous disturbance before the next disturbance event. If
not, what are the possible consequences on ecosystem processes if the time
interval is too short to allow the ecosystem to recover? What factors control the
rate of recovery?

Covington (1981) documented the recovery of forest floor following clear
cutting in northern hardwoods. Figure 1 illustrates an initial decrease in forest
floor early in secondary succession, followed by an asymptotic increase.
Covington (1981) attributed the initial decrease in forest floor to a decline in plant
production and increase in decomposition rates, followed by accumulation of
forest floor when plant production exceeded the rate of decomposition. By year
64, the forest floor had recovered to within 5% of predisturbance levels. Can one
expect the same rate and pattern of forest floor recovery if a 50-year—old stand is
cut and what are the possible consequences on productivity of as subsequent
regenerating stand? These issues have important implications for forest
management strategies.

What about natural disturbances, such as wildfire? Climate change
models predict a change in climate in boreal forest ecosystems that will favor

greater fire activity (Flannigan and Van Wagner 1991). During the 1980s, the

13

warmest decade in recent history, the area of boreal forests burned increased
dramatically (in Stocks et al. 1996). This may result in a significant decrease in
the ability of these forests to store C, and raises the possibility of positive
feedbacks to global climate change from increased fire activity.

Rotation age and fire return interval are the keys to understanding the
benefits and damage a disturbance can cause to an ecosystem. Long
disturbance return intervals may lead to sequestration of nutrients essential to
plant growth, thereby limiting plant productivity. A disturbance return interval that
is too short may also lead to a decrease in plant productivity because the inputs
to the ecosystem between disturbance events are not high enough to replace

losses (Aber and Melillo 1991).

JACK PINE ECOSYSTEM

One of the most fire-prone ecosystems is one that is dominated by jack pine
(Pinus banksiana L.). Jack pine is an early successional, short-lived, shade-
intolerant tree species, and a significant component of North American boreal
forests. It grows on droughty, nutrient-poor, sandy soils (Cayford and McRae
1983), where most other species stnIggle to survive because of low nutrient
availability and drought stress. Jack pine forests are extremely prone to fire
because their foliage is highly combustible and the tree density is usually high
enough to spread the fire through the tree crowns (Rowe and Scotter 1973). One
characteristic that makes jack pine ideally suited for post-fire regeneration is

serotinous cones. Seeds are enclosed in the serotinous cones and the cone

14

scales are held together with a bonding resin that melts at about 50°C (Cameron
1953). Forest fires provide the necessary heat to melt the resin and open the
cones to begin seed germination. Jack pine stands are predominantly maintained
by wildfire, even-aged, and are typically less than 100 years old. Even though
jack pine is predominantly found in the boreal forest, the southern extent of this
species is in the temperate zone, in northern Lower Michigan.

Jack pine in Michigan is of extreme interest because it is home to the
Kirtland’s Warbler (Dendroica kirtlandii Baird), an endangered species. The
Kirtland’s warbler is an endangered songbird whose breeding range and nesting
habitat are very narrow and limited to young jack pine stands of northern Lower
Michigan (Walkinshaw 1983). Management of Kirtland’s warbler habitat requires
the creation of early-successional jack pine stands through clear-cutting and/or

prescribed burning.

MANAGEMENT IMPLICATIONS

Understanding the recovery patterns of ecosystem processes and the factors
controlling these patterns is fundamental to sustain the production of commercial
forests and manage for wildlife habitat. Management of forests for wood
production is often based on expected rates of productivity. However, productivity
depends on soil fertility and changes with stand age (Ryan et al. 1997).
Management activities that alter nutrient supply in the soil could affect forest
production, rotation ages, sustainability, and other aspects of forest

management. Sustainability problems are likely to occur when the post-harvest

15

replenishment of nutrients is less than what is required to reestablish pre-harvest
biomass production rates. Understanding the effects of natural disturbances on
ecosystems and comparisons with different management techniques may
provide great insight into the proper management of forests, as well as further

our knowledge of benefits and costs of ecosystem management.

RATIONALE

The overall affects of fire on ecosystem processes are complex because all the
components of the system interact to control nutrient and C cycling. Pools of
labile nutrients in soil organic matter are of vital importance not only because
they control ecosystem productivity, but also because these pools can respond to
changing soil moisture and temperature resulting from climate change. For this
reason, we need to understand which effects are transient and which are long-
lasting, as well as what mechanisms control the rate of ecosystem recovery.

The objective of this study is to examine how ecosystem processes, both
belowground and aboveground, recover after wildfire and determine what are the
main factors that limit and drive those processes. In this study I will investigate
the recovery of ecosystem process in the jack pine ecosystem of northern Lower
Michigan. The results of this project will provide important baseline information
on the maximum rotation age required for sustainable management of northern

Lower Michigan jack pine forests.

16

CHAPTER 2
THE RECOVERY OF ECOSYSTEM PROCESSES AFTER WILDFIRE IN
MICHIGAN JACK PINE FORESTS

ABSTRACT

I investigated the recovery of ecosystem processes after wildfire in
Michigan jack pine (Pinus banksiana) forests using a chronosequence of 11
wildfire-regenerated stands spanning 72 years. The objective of this study was to
characterize recovery patterns of soil nutrients, soil carbon (C) and nitrogen (N)
mineralization and aboveground biomass as well as determine the mechanisms
that drive those patterns.

In situ N and C mineralization were measured monthly during the 2002-
growing season. Multiple regression analysis was used to determine the
important factors controlling N and C mineralization. To determine seasonal
effects, N mineralization and soil respiration were analyzed using analysis of
vanance.

Annual N mineralization rates spiked to 3564 mg N/m2 1 year following
wildfire, decreased to a minimum 325 mg N/m2 after about 8 years then began to
increase, reaching a steady state of 1284 after about 44 years. Rates of N
mineralization increased with increasing forest floor mass and decreasing C: N
ratios. Generally, soil respiration rates decreased after wildfire, began to
increase after about 10-15 years, and then declined after 30 years. Soil

respiration declined with increasing soil moisture and increased with increasing

17

temperature and forest floor moisture. Forest floor and soil N and C pools as well
as C: N ratio increased with succession, whereas temperature decreased. Soil
and forest floor became more acidic as stands matured. Understory vegetation
reached a peak 7 years after wildfire and then decreased.

In this jack pine chronosequence, accumulation of forest floor largely
regulates the recovery of N mineralization. The initial spike in N mineralization
immediately following wildfire was driven by rapid turnover of N in the mineral
soil. The recovery of N mineralization to predisturbance levels was driven by a
gradual accrual of forest floor. Soil respiration was most likely driven by microbial
heterotrophic respiration early in succession and later by both root respiration
and microbial respiration. All measured ecosystem processes recovered to

predisturbance levels within 40 years of wildfire.

INTRODUCTION

The response of forest ecosystems to wildfires has been studied extensively,
including studies of vegetation succession (Bergeron and Dubuc 1989; De
Grandpré et al. 1993), soil microbial activity (Bissett and Parkinson 1980;
Sharma 1981; Tiwari and Rai 1977; Vézquez et al. 1993; Widden and Parkinson
1975), soil chemistry (Austin and Baisinger 1955; Bauhus et al. 1993; Dunn et al.
1979; Dymess et al. 1989; Kutiel and Naveh 1987) and soil respiration (Weber
1990). However, most studies of soil processes have focused on the immediate
affects of fire on soil chemistry and microbial activity and long-term studies are

uncommon. Furthermore, numerous studies on the effects of fire on soil

18

chemistry are conducted using prescribed burns (Ahlgren and Ahlgren 1965;
Giovannini and Lucchesi 1997; Lynham et al. 1997). To study the recovery of soil
processes after wildfire requires following the same burn for several decades, a
method unlikely to be undertaken. Instead, scientists have used
chronosequences of stands regenerated at different times to study succession
(Borrnann and Sidle 1990; DeLuca et al. 2002; Krause 1998; Pare et al. 1993).
Because meeting the assumptions of a chronosequences is often problematic,
these studies are uncommon. At the same time, more studies on the long-term
effects of wildfire on ecosystem processes are needed to build better predictive
models of climate change and make more informed forest management
decisions.

Wildfire is the most common agent of disturbance in boreal forest
ecosystems. In boreal forests, decomposition rates decrease through succession
leading to low nutrient availability and accumulation of forest floor (Bormann and
Sidle 1990; Pastor at al. 1987; Van Cleve and Viereck 1981; Van Cleve et al.
1983). Nitrogen mineralization rates in boreal forests of Alaska decline during the
course of succession (Van Cleve and Viereck 1981; Van Cleve et al. 1983) and
net N immobilization rates increase in mature stands (Van Cleve and Nooman
1975; Van Cleve and Viereck 1981). These trends with succession are often
attributed to decreasing substrate quality and a changing soil microclimate that
slows down microbial activity. DeLuca et al. (2002) also observed a decrease in
N mineralization along a 352-year fire chronosequences in northern Sweden. In

contrast, Brais et al. (1995) reported no change in N mineralization over 231

19

years of succession in the southern boreal forest of Canada. In boreal forests, a
large proportion of the total ecosystem N capital is stored in the forest floor and
can range from 7.8 % in jack pine (Pinus banksiana) to as much as 72.4 % in
black spruce (Picea man'ana) (Morris and Miller 1994). Thus, wildfire plays a
positive role in these systems by rapidly mineralizing the nutrients sequestered in
the forest floor and creating post-fire conditions that stimulate microbial activity
and therefore increasing N mineralization rates (see review by Neary et al. 1999).

The accumulation of forest floor C and N depends on the balance between
primary production and decomposition. The source of forest floor is the littefall
and mortality of understory and overstory plants. The factors that control the rate
of production are complex (Ryan et al. 1997) and include nutrient, light and water
availability. Decomposition is controlled mainly by temperature, moisture, and
quality of the organic matter. Accumulation of forest floor occurs when the rate of
litter production exceeds the rate of decomposition. The balance between
production and decomposition may shift throughout succession and may lead to
loss of forest floor early in succession and accumulation of forest floor as stands
mature (Covington 1981).

The accumulation of forest floor in boreal forests plays an important role in
global carbon (C) cycling because it stores vast quantities of C. Nitrogen and C
cycles are tightly linked in terrestrial ecosystems (Bolin and Cook 1983; Melillo
and Gosz 1983), consequently, ecosystem C storage may be constrained by the
availability of N (Townsend and Rastetter1996). In boreal forest ecosystems, soil

respiration (CO2 flux) is a large component of total C flux and it has been

20

hypothesized that CO2 flux changes during forest succession following wildfire (in
Wang et al. 2002). Soil respiration is a cumulative flux of microbial respiration
and plant root respiration. The rate of soil respiration is controlled largely by soil
temperature (Edwards 1975, for reviews see Singh and Gupta 1977, Raich and
Schlesinger 1992), moisture (Edwards 1975, Linn and Doran 1984, Raich and
Schlesinger 1992) and substrate quantity (Trumbore et al. 1995). Because all
these factors change throughout the course of succession following wildfire, it is
reasonable to hypothesize that soil respiration rates change as well.

The objective of this study was to characterize the recovery of C and N
cycling after wildfire in Michigan jack pine forests and to determine the main
factors controlling those patterns. Specifically, the goal was to determine the
pattern in C and N mineralization, changes in microclimate and vegetation
accumulation. l hypothesized the following:

1) Soil temperature will decrease with succession.

Rationale: Soil temperature will be high after wildfire due to greater soil heating
associated with removal of the forest canopy and blackening of the soil surface
causing decreased soil albedo. Soil temperature will decrease as the canopy
closes.

2) Soil moisture will decrease with succession.

Rationale: Soil moisture depends primarily on plant transpiration and
evaporation. Immediately following wildfire, loss of vegetation will reduce
transpiration. Concurrently, higher soil temperature (Hypothesis 1)will lead to

greater evaporation rates. Decrease in transpiration will be more than increase in

21

evaporation, resulting in higher soil moisture. Soil moisture will decrease with
succession as plants reestablish and leaf area increases, resulting in greater
moisture loss via transpiration.

3) Organic matter quality will decrease with succession.

Rationale: Organic matter quality will decrease due to increasing inputs of poor-
quality coniferous litter from jack pine. As soil microorganisms exhaust more
labile sources of C and N, recalcitrance of organic matter will increase.

4) Substrate quantity will increase with succession.

Rationale: Litter inputs will increase as plant production increases with stand

age, whereas decomposition of organic matter will decrease due to less

favorable microclimate (Hypothesis 1 & 2) for heterotrophs and decreasing

organic matter quality (Hypothesis 3).

5) Net N mineralization rates will initially increase with succession and
then decrease.

Rationale: Nitrogen mineralization is partly dependant on the quantity of organic

N. Nitrogen mineralization will be low early in succession because of the limited

amount of organic N remaining after wildfire. Nitrogen mineralization will increase

as organic matter accumulates with succession (Hypothesis 4) and then

decrease as organic matter quality decreases (Hypothesis 3).

6) Soil respiration will increase with succession.

Rationale: Root respiration will increase with vegetation recovery. Concurrently,

microbial respiration will increase with greater accumulation of organic matter

(Hypothesis 4).

22

MATERIALS AND METHODS
Site Description and Selection
All study sites were located in the Highplains District (8) of Region II of northern
Lower Michigan (44°30’N, 84°30’W) (Albert et al. 1986). This region is
characterized by a cold climate (MAT = 67°C) and a short growing season (ca.
115 days) from May through September. The average temperature during the
growing season is 169°C (Albert et al. 1986). Soils in the Highplains District form
on glacial outwash and are dominated by excessively well-drained, acidic sands
of the Grayling series (Typic Udipsamments) (Werlein 1998). Jack pine (Pinus
banksiana) is the dominant tree species in the outwash plains often co-occurring
with northern pin oak (Quercus ellipsoidalis). The fire cycle for jack pine forests in
the in the Highplains District is ca. 30 years (Ryel 1981; Simard and Blank 1982).
In May of 2002 I established a chronosequence of 11 wildfire-regenerated
stands of jack pine, which burned in years 2001, 2000, 1998, 1995, 1990, 1988,
1980, 1975, 1966, 1950, and 1930. I used fire maps from the Michigan
Department of Natural Resources (DNR) and the United States Department of
Agriculture-Forest Service (USDA-FS), along with USDA-FS fire reports, to
identify fire locations for stands dating back to 1966. All of these stands
originated from recorded fires greater than 200 acres. The two stands older
than 36 years were selected using USDA-FS compartment maps followed by
field surveys to validate that stands were even-aged jack pine with no evidence
of planting (tree rows or furrows). I made the assumption that even-aged,

unplanted jack pine forests in this region must be of wildfire origin. Jack pine’s

23

extreme shade intolerance and dependence on wildfire for regeneration (Cayford
and McRae 1983), together with the high frequency of wildfires in this area,
suggests that any other scenario of stand establishment is unlikely. For all fires, I
used field observations to eliminate areas that showed evidence of planting,
salvage logging (where possible), or that were less than 90% jack pine basal
area. However, the presence of cut tree stumps at the stand that originated in
1975 does suggest that it was salvaged after the fire. I included this stand in the
chronosequence because it was the only stand in that age class. All of the study
sites were located within a 23 km radius (Figure 2).

At each site I identified an area (between 1 and 2.5 ha) of uniform terrain
with no evidence of human disturbance, hereafter referred to as a stand. Stand
borders were located at least 20 meters away from the nearest road. All soil and
vegetation sampling was conducted along 3 parallel, 60-m transects beginning at

stratified random points along the long axis of the stand.

Net Nitrogen Mineralization

Net N mineralization was measured using an in situ closed core method (Raison
et al., 1987). Soil sampling was conducted ca. monthly throughout the 2002
growing season (May 20-November 11). Monthly soil samples were collected
from 3 points along each transect. Points were randomly located within each 20-
m increment along the transect, and at a random distance between 1 and 5
meters from the transect. At each sample point, two 5.08-cm inside diameter,

PVC cores were driven 10 cm into the soil. Each soil core was sharpened at the

24

penetrating and of the pipe to facilitate insertion. One PVC core was removed
immediately for the time zero determination of soil NH..*- and N031 N
concentrations, whereas the second core of each pair was capped on top with
duct tape to prevent leaching and retrieved at the next sampling date. Each soil
sample was separated into mineral (A horizon) and forest floor (0,, +03 horizons)
components. The three cores from each transect were composited into one
mineral soil sample and one organic soil sample, yielding a final sample size of 3
for each stand.

All soil cores were kept on ice in coolers for transport to the laboratory,
stored in a refrigerator at 4°C, and analyzed within 48 hours. Mineral soil was
moist sieved to <4 mm and homogenized. The organic soil was homogenized
and roots were removed by hand. In instances where there was an inadequate
amount of sample for analysis, the organic samples from all three transects were
composited. A 20-g and a 5-g subsample from the mineral and organic soil,
respectively, were shaken with 50 mL of 2 M potassium chloride (KCI) for 1 hour
to extract NH]- and N031 N. The extracts were poured through Whatman #2
filter paper and stored frozen until analysis for inorganic nitrogen using an
Alpkem Flow Solution IV Auto-Analyzer (OI Analytical, College Station, TX). An
additional soil subsample was oven dried at 105°C to determine gravimetric soil
moisture. Net N mineralization rates were calculated on an oven-dry basis as the
increase in inorganic NHf- and N031 N between the two time periods, then
scaled to an areal basis using mineral soil bulk density and organic soil mass per

unit area (mg N/mz).

25

Soil Respiration

Soil respiration was measured monthly during the 2002 growing season using
the soda lime technique (Edwards 1982; Kabwe 2002). Two sample points were
randomly located along each transect (stratified within each 30 m increment).
Approximately 20 g of soda lime was added to a 5.08-cm diameter metal tin fitted
with a gastight lid. The containers were heated uncovered for 48 hr at 105°C to
remove moisture and transferred into the field in airtight tuppenivare containers
with mesh packets of Drierite desiccant (calcium sulfate). At time 0 in the field,
the open tin was sprayed with deionized water to activate the soda lime. The tin
was then placed on a plastic stand inside an inverted, white plastic bucket 20.3-
cm in diameter pushed 1 cm into the soil. The container dimensions and amount
of soda lime used adhere closely to Edward’s (1982) recommendations that the
surface area of the container with soda lime covers ca. 5% of the soil surface
area measured, and that 0.06 g of soda is used per cm2 of forest floor. A bag of
sand was placed on top of the inverted bucket to form a good seal and prevent
disruption from wind. All aboveground vegetation was clipped from each sample
area immediately prior to beginning respiration measurements. After 24 hours of
incubation, the lids were placed back on the tins and returned to the laboratory in
the airtight tupperware. The soda lime was again dried at 105°C for 48 hours. At
five randomly selected stands at each sample date, one tin of soda lime was
placed in closed chamber of similar volume to determine background CO2 uptake
in the absence of soil CO2 efflux. Respired CO2 was measured as the difference

between oven dry (105°C for 24 hours) weight before and after the soda lime

26

was placed in the field. The increase in weight is directly proportional to the
amount of CO2 adsorbed during the incubation period.

To estimate soil respiration for the growing season I extrapolated rates for
ca. 2 weeks before and after each sampling date based on the measured rate of
CO2 flux (Toland and Zak 1994, Vogt et al. 1980; Ewel et al. 1987a; 1987b).
Cumulative respiration was calculated by summing the respiration flux over the

five time periods.

Physical and Chemical Properties

Soil bulk density samples were collected in August using a soil bulk density
sampler (Soilmoisture Equipment Corp., CA) by inserted it into the A horizon and
sampling a volume of 615 cm3. Soil bulk density was determined as weight of the
soil per unit volume (g/mI) on an oven-dry (105° C) basis.

Soil pH and exchangeable bases were determined from air-dried, initial
subsamples of composite soils cores collected in May. Soil pH was determined in
water using a 2:1 ratio for mineral soil and 4:1 ratio for organic soil of deionized
water: air-dry soil. The samples were shaken for one hour and then pH was
measured with a glass electrode. Extractable base cations (Ca, Mg, K, Na) were
measured on ammonium chloride extracts using a Direct Current Plasma Atomic
Emission Spectrometer (SpectraMetrics/Beckaman, MA). Total exchangeable
cations ware determined by multiplying concentration by bulk density.

I used a 25.4 cm x 25.4 cm sampling square to measure the amount of

forest floor organic matter at each stand. In August, I collected 34 frames from

27

stratified random locations along each transect. The samples were oven dried at
65°C and weighed to determine the amount of forest floor at each stand on an
areal basis.

Carbon and N content for soil was determined using air-dried, initial
subsamples of composite soils cores collected in May. Forest floor C and N was
determined using subsamples from forest floor collected in August. Samples
were pulverized in a Kleco 4200 ball mill (Kinetic Laboratory Equipment
Company, CA) and analyzed for C and N on a Cano Erba NA1500 Nitrogen
Analyzer (Elantech, NJ). Carbon and N content for soil was determined by
multiplying soil C and N concentrations by bulk density. Forest floor C and N
content was determined by multiplying concentrations by amount of forest floor at

each stand.

Soil Microclimate
I measured soil temperature every 1 or 2 hours for the duration of the field study
using 1 WatchDog 100 data logger (Spectrum Technologies, Inc., IL) per transect
placed 5 cm below the soil surface. To summarize temperature differences
among stands, I calculated the number of growing season degree-days by
adding daily average number of degree hours above 0°C in the growing season
for each stand.

Soil moisture was measured as moisture tension using WatchDog 400
data loggers and Watermark soil moisture sensors (Spectrum Technologies, Inc.,

IL) at stand ages 2, 12, 22, 36 and 72. Each soil moisture datalogger

28

accommodated three sensors that were placed at a distance of 6.1 m from the
logger and at an azimuth of 0°, 120°, and 240° from each point. Soil moisture

was also measured gravimetrically from subsamples of soil collected for initial N

pool measurements.

Vegetation
Live understory biomass was sampled in August using a 0.25 m2 sampling
square at two random points along each transect. Plants were clipped at the
base and placed in one of the following categories: mosses, lichens, ericaceous
shrubs (primarily Vaccinium spp.), sand cherry (Prunus pumila), graminoids and
forbs, sweet fern (Comptonia peregrina), bracken fern (Pten'dium aquilinum) and
a miscellaneous category. Samples were composited by transect, dried in a
forced air convection oven at 55°C, and then weighed. Total understory biomass
was calculated as the total sum of all categories for a particular stand age.
To measure overstory biomass, l established one 7x14 m plot parallel to each
transect at stand ages 12 and older. I measured diameter at breast height (dbh)
of every live tree (over 1 cm dbh). Overstory biomass was estimated using the
following equation:

Biomass=0.1054 (D)2'3‘"
(Green and Grigal 1978; Grigal and Ohmann 1992), where D is dbh (p<0.001;
r2=0.983). At stand age 7, none of the tress reached breast height. Therefore, I
built the following regression:

Biomass= 0.0014*EXP(0.039*Helght)

29

predicting aboveground biomass from height based on 5 harvested trees

(p<0.001; r2 = 0.994).

Statistical Analysis
I used linear and nonlinear regressions to analyze changes with stand age in N
mineralization, C and N content, C:N ratio, exchangeable cations, pH, number of
degree-days and overstory biomass. The function that best fit the relationship
was based on the combination of minimum sum of squares of error and lowest p-
values. A modified gamma function was fit to the annual total N mineralization
data using estimated parameters calculated in CurveExpert (Microsoft, Inc.). The
gamma function was modified from the original [y=a+bx°exp(dx')] to
[y=a+bx°exp(dx°)-fx], allowing a lower asymptotic value. A linear [y=mx-I-b]
function was fit to forest floor C:N ratio, total C, total N, total C:N ratio, forest floor
pH, as well as number of degree-days. Quadratic [y=ax2+bx+c] and power
[y=ax"] functions were fit to forest floor pH and soil C:N ratio, respectively. I
used hyperbolic [y=(a*b)l(b+x)] and carrying capacity [y=a*(1-exp(bx))°]
functions to fit curves to soil N and overstory biomass, respectively. I used the
exponential decay function [y=a*exp(-b*x)] to fit curves to soil C and total
exchangeable base cations. Stand age 7 was excluded from the regression of N
mineralization, total C and N content against stand age. All regression analyses
were performed on stand average data to yield an n of 1 for each stand.

I used multiple regression analyses to determine the main variables

affecting N mineralization and soil respiration. Soil respiration data were

30

analyzed using all stands (11) and sampling dates (5) yielding an n of 55.
Backward elimination procedures were used in multiple regression analysis to
select the model with the conceptually strongest parameters and highest
adjusted and partial r2. Two-way analysis of variance was used to determine
differences in monthly soil moisture, soil respiration and total N mineralization for
all stands.

In multiple regression analyses, N mineralization, soil respiration, C and N
content, C: N ratio, exchangeable bases, as well as forest floor mass were In-
transformed to normalize the data. Multiple regression analyses were performed
using Statistical Analysis System (SAS) for personal computers (SAS Institute
1987). Analysis of variance was performed using SYSTAT software (Systat
Software, Inc). Linear and nonlinear regressions were performed using
SigmaPlot software (SPSS, Inc.). Significance was accepted at minimum alpha

of 0.05.

RESULTS

Soil Carbon and Nitrogen Pools

Forest floor C and N content initially declined and then began to increase after
ca. 10 years following wildfire, finally reaching a steady state after almost 40
years (Figure 3A-B). The exception to this trend is stand age of 7, which
exhibited the highest amount of forest floor C and N. Forest floor C: N ratio
(Figure 30) did not change with stand age (p=0.1853). Soil C (Figure 4A) also did

not change with stand age (p=0.0990). Soil N content (Figure 43) decreased with

31

stand age (p<0.0189, adjusted r2=0.42), however, stand age was not a strong
predictor of change in soil N content. Soil C: N ratio (Figure 4C) did not change
with stand age (p=0.2043). Total C (Figure 5A) content increased with
succession (p<0.0138, adjusted r2=0.50, whereas total N (Figure 58) did not
change (p=0.0818). Total C: N ratio (Figure 5C) increased with stand age
(p<0.0391; adjusted r2=0.325). Stand age 7 was excluded from the linear
regression analysis of total C and N because the high contribution of C and N

from the forest floor at that stand made it an outlier.

Net Nitrogen Mineralization
The annual amount of total N mineralized decreased rapidly in the first ten years
following wildfire to rates less than what was observed in the mature stands and
then slowly increased until reaching a steady state after almost 40 years
following wildfire, with rates less than half the initial amount (p<0.001; adjusted
r2=0.80) (Figure 6). Stand age 7 again was an outlier having the highest amount
of total N mineralization (6348.511893 mg N/mz). Total N mineralization
increased with increasing forest floor mass (FFM) and decreasing C: N ratio
(p=0.004; adjusted r2=0.69) resulting in the following equation:
In (Nminm.)=22.67 + 0.511 ln(FFM) - 6.01 ln(C: NW)

The percentage of total variance explained by forest floor mass and C: N ratio
individually were 39 and 11 percent, respectively.

The amount of N mineralized in the forest floor initially decreased and then

increased approximately ten years following wildfire to an amount double the

32

initial value (Figure 7A). Stand age 7 was again an outlier and had the highest
amount of N mineralized in the forest floor (14411518). Forest floor N
mineralization was best explained by forest floor mass (FFM) (p<0.001; adjusted
r2=0.86) resulting in the following equation:

In (Nminrr)= - 0.85 + 0.92 ln(FFM)

The amount of N mineralized in the soil decreased by an order of
magnitude by 15 years after wildfire (Figure 7C). Stand age 7 did not follow this
pattern and had the highest amount of N mineralized in the soil (4907.511 586).
Soil N mineralization was explained by soil C: N ratio, total exchangeable bases
and forest floor mass (FFM)(p<0.001; adjusted r2=0.85), resulting in the following
equafion:

In (Nmin.o..)= 22.61- 5.38 ln(C:N.ou) + 0.53 ln(bases) + 0.20 ln(FFM)
The percentage of total variance explained by soil C: N ratio, total exchangeable
bases and forest floor mass individually was 62, 44 and 11, respectively.

To gain further insight into the controlling factors of N availability and how
it varies with stand age, N mineralization was relativized for N content.

The increase in forest floor N mineralization with succession disappeared when
relativized to account for changes in N content (Figure 7B). The trend in soil N
mineralization did not change (Figure7D) and therefore soil N pool size did not
explain the change in soil N mineralization with stand age. Forest floor N pool at
stand age 7 accounted for the high N mineralization rates in the forest floor.

However, soil N pool size failed to account for high N mineralization rates in the

33

soil. The percent of total N mineralization coming from the forest floor (Figure 8)
increased linearly with stand age (p=0.001; adjusted r2=0.676).

A two-way analysis of variance with stand and month as factors showed
no significant seasonal differences in total N mineralization between stands
(p=0.446) (Table 1). However, the significance of stand age as a factor (p<0.001)

indicated that N mineralization was different between stands.

Soil Respiration
Annual CO2 flux decreases in the first decade following wildfire (Figure 9),
increased to rates double the initial values in the next decade and finally declined
slowly for the next 50 years. A two-way analysis of variance with stand and
month as factors showed a significant effect of season on soil respiration
(p=0.001) (T able 2) and a significant interaction between stand age and sampling
month (p=0.001). Soil respiration was highest in July in the three youngest
stands and stand age of twenty-two, whereas it was higher in August in the four
oldest stands (Figure 10). Multiple regression analysis revealed that mean
temperature (T) for day of measurement (a), gravimetric moisture from forest
floor (MFF) and soil (Mson) best predicted the variance in soil respiration (p< 0.001,
r2 = 0.44) , resulting in the following equation:

In (Respiration)=2.95 + 0.379 ln(T) + 0.0074(MFF) - 0.036(Mm.)
The percentage of total variance in soil respiration explained by temperature,
forest floor moisture and soil moisture individually was 34, 5 and 2 percent,

respectively.

34

Other Soil Chemical Characteristics

Total exchangeable bases (Figure11) decreased exponentially with increasing
stand age (p<0.0001; adjusted r2=0.885). Calcium (Ca) made up most of the
bases, whereas sodium (Na) the least. Both soil and forest floor had an acidic pH
at all the stands and decreased with stand age (Figure 12). Soil pH was best
described by a quadratic function (p-value=0.0043; adjusted r2=0.680) with little
change in pH for the first two decades followed by a steady decline. Forest floor
pH decreased linearly with increasing stand age (p-value<0.001; adjusted

=o.752).

Soil Microclimate

The total number of degree-days above 0°C at 5 cm depth in soil (Figure 13)
decreased with increasing stand age (p-value=0.0346), however, a large
proportion of the variability was not accounted for by stand age (adjusted
3:0.342). Gravimetric moisture in the forest floor (Figure 14A) was lowest in
September at all stands. With the exception of stand age of 2, percent forest floor
moisture was higher in the older stands than the younger stands. Gravimetric
percent moisture in the mineral soil was lowest in the July and September
sampling days (Figure 148) and was much lower compared to forest floor. A two-
way analysis of variance of gravimetric moisture in the mineral soil (Table 3) with
month of sampling and stand age as factors indicated that soil moisture varied
seasonally (p<0.001) and differences were significant between stands (p=0.017).

There was no interaction between stand age and sampling month (p=0.175).

35

Analysis of variance was not performed on forest floor moisture data because the
data for youngest stands were composited, yielding an n of 1. Soil moisture
measurements using dataloggers (Figure 15) showed that the months of June
and August were the driest. Youngest and oldest stands were the driest, whereas

stand age of 22 was wettest.

Vegetation
In the first decade following wildfire, live jack pine biomass (Figure 16) increased
slowly to 6.96 Mg/ha, followed by a logarithmic increase in biomass for the next
30 years (p<0.001; adjusted r2=0.983). By 50 years following wildfire, jack pine
biomass increased to 84.73 Mg/ha, followed by a period of slower biomass
accretion. ~

Understory biomass increased rapidly following wildfire and began to
decrease rapidly in less than 10 years (Figure 17). By 50 years, understory
biomass began to increase. Stand age of 27 was the exception to this trend

having the highest understory biomass.

DISCUSSION

Net Nitrogen Mineralization

In the jack pine chronosequence, forest floor accumulation largely regulates N
mineralization. Changes in N mineralization can result from changes in N capital,
N turnover, or both. The observed pattern in N mineralization with succession in

jack pine forests occurs as a consequence of changes in the distribution (Figures

36

3-5) and turnover of N in the soil (Figure 7D) and forest floor (Figure 7B) as these
forests recovered after wildfire.

ln mature stands, N turnover was slow (Figure 7B,D) and about 40% of
total N capital was tied up in the forest floor. Wildfire had an immediate effect on
both the distribution and turnover of N. Wildfire removed most of the forest floor
and only about 13% of the total N capital remained there. Even though soil N
declined after wildfire, N turnover in the soil increased, leading to a spike in N
mineralization. The increase in N turnover can be partially attributed to greater
exchangeable cation availability and warmer soil temperatures following wildfire.
Soil temperature can increase due to the presence of charcoal and through
destruction of forest canopy (Christensen and Muller 1975). Incomplete
combustion of organic matter during wildfire results in the formation of charcoal.
Because of its low albedo, charcoal can absorb more radiation and heat the soil.
Furthermore, the destruction of the forest canopy allows more radiation to reach
the soil surface. Because soil temperature can limit microbial metabolic rates
(Paul and Clark 1996; Swift et al. 1979), higher soil temperature can lead to
higher rates of N mineralization. Even though wildfire causes large losses of total
N (see review by Neary et al. 1999; Raison et al. 1985; Grier 1975; Hanivood and
Jackson 1975), it can also lead to increased availability of less volatile nutrients,
such as phosphorus and base cations through the production of ash. Burning of
organic matter releases these nutrients and makes them available for plant and

microbial uptake. It appears that warmer soil temperatures and the spike in

37

exchangeable cations provide more favorable conditions for increased microbial
activity, resulting in higher turnover rate of N remaining after the fire.

However, the stimulating effect of fire was transient. In the first decade
following wildfire, a large proportion of exchangeable cations were either taken
up by soil microorganisms and recovering vegetation, or lost via leaching. Soil
microorganisms have most likely exhausted any sources of labile C and N.
Furthermore, forest floor mass was at its lowest levels and contributed to less
than 6% of total N capital. Soil N turnover rate rapidly declined in the first decade
to levels measured in mature stands. Within 15 years following wildfire, N
mineralization was at its lowest levels in the chronosequence.

Nitrogen mineralization rates began ~to gradually recover ca. 15 years
following wildfire. As jack pine biomass entered its exponential growth phase,
forest floor began to accumulate owing to jack pine needle litter and tree
mortality. The increasing forest floor provided fresh substrate for soil
microorganisms. The increase in N mineralization can be attributed to a gradual
accumulation of the forest floor N pool. The fact that N turnover rate in forest floor
(Figure 7B) did not change with stand age further suggests that higher N
mineralization rates resulted from more N in the forest floor, not from higher N
turnover. As forest floor reached a steady state at ca. 40 years, N mineralization
rates also stabilized.

Although stand age of 7 was consistently an outlier in terms of N cycling
parameters, this stand further illustrated the importance of forest floor as a

driving force in the recovery of N mineralization. Stand age of 7 exhibited not only

38

the highest N mineralization rates, but also had the highest forest floor mass and
lowest C: N ratio. In the forest floor, N mineralization at this stand was almost an
order of magnitude higher than stands of similar age. However, N turnover in the
forest floor was similar to the other stands, suggesting that higher N
mineralization rates were higher as a result of more forest floor mass, not more
rapid N cycling.

The greater accumulation of forest floor at stand age of 7 is most likely a
consequence of greater moisture availability. The soil profile (Table 4) revealed
the presence of a gravel band, which can slow down percolation of water through
the soil profile. In addition, high clay content (Table 4) at this site can also lead to
higher moisture content. Greater moisture availability may have reduced the loss
of forest floor to volatilization during wildfire. Thus, this stand may have had more
forest floor in the beginning of succession than the other stands in the
chronosequence. Furthermore, greater water availability may allow for more
rapid understory vegetation growth and thus to greater forest floor accumulation.

The results of the multiple regression for soil N mineralization suggest that
C: N ratio and mass of forest floor are also important in driving the pattern of N
cycling in the mineral soil as stands mature. Forest floor content can influence N
mineralization via leaching of organic compounds into the soil and by affecting
soil microclimate. Translocation of organic compounds into the mineral soil has
been documented (Froberg et al. 2002; Park et al. 2002; Solinger et al. 2001).
Dissolved organic matter can include labile compounds such as peptides and

amino acids, as well as recalcitrant compounds. Recalcitrant compounds such as

39

fulvic and humic acids may have a negative effect on N mineralization because
fulvic acids contain little N and humic acids are not very labile (Paul and Clark
1996). This mechanism is an unlikely scenario in this system because the model
produced by the multiple regression predicts a positive effect of forest floor
content on N mineralization. However, leaching of labile compounds from forest
floor into the soil may have a positive effect on N mineralization. Forest floor can
also protect mineral soil from drying out during the hottest months.

The results partially support the hypothesis that N mineralization will
increase as stands begin to recover and decrease with subsequent succession.
My original hypothesis did not incorporate the transient stimulating effect of fire
on N mineralization. I also hypothesized a decrease in N mineralization in mature
stands. Even though forest floor mass has reached a steady state at the end of
the chronosequence, N mineralization rates may decline as a result of
decreasing substrate quality as evidenced by the widening total C: N ratio.

The observed pattern is different from what is found in the boreal forest of
Alaska (Van Cleve and Viereck 1981; Van Cleve et al. 1983) where forest floor
accumulation is greater, but N mineralization decreases with succession and N
immobilization occurs in mature stands (Van Cleve and Nooman 1975; Van
Cleve and Viereck 1981). The three major differences between jack pine forests
of Michigan and more northern forests are number of days in the growing
season, the composition of the understory and succession of the overstory. The
lower number of days in the growing season of boreal forests limits the amount of

N mineralized annually. Furthermore, the development of a moss layer in mature

40

stands may inhibit N mineralization (Wardle and Lavelle 1997). Lastly, overstory
succession from deciduous to conifer species may explain the decrease in N
availability. The deciduous litter decomposes more rapidly than litter from
conifers due to the acidic nature and low chemical quality of coniferous litter
(Flannagan and Van Cleve 1983; Miles 1985). Furthermore, the deciduous litter
prevents the formation of a moss layer early in succession (Van Cleve and
Viereck 1981). Paré et al. (1993) also did not observe a decrease in N
mineralization with succession after wildfire in a southern boreal forest and
partially attribute the maintenance of N availability to persistence of some
deciduous species in late successional stages.

In the jack pine forests of Michigan, (N mineralization began to recover
after ca. 15 years following wildfire (Figure 6) and reached a steady state by ca.
40 years. The rate of N turnover declined to predisturbance levels by 20 years
and remained constant. Measurements of N mineralization rates suggest that
rotation age of greater than 40 years should sustain yields in Michigan jack pine

forests.

Soil Respiration

The pattern of annual soil respiration is most likely controlled by different factors
throughout succession. The decline in soil respiration early in succession can be
attributed to a decline in heterotrophic respiration. The increase in soil respiration
10 years after wildfire is most likely a factor of accumulation of forest floor and

overstory production, whereas the decline as stands mature is a consequence of

41

decreasing substrate quality and decline in jack pine root respiration as
productivity declines.

Dead roots can serve as a primary source of C for heterotrophic microbes
after wildfire (Klopatek and Klopatek 1987), but the source can be quickly
exhausted, causing the decline in soil respiration. In the first decade following
wildfire, the decline in soil respiration coincided with a decline in substrate
availability (Le. forest floor mass). Other studies have also shown an initial
decrease in soil respiration following wildfire (Amiro et al. 2003; Wang et al.
2002). Amiro et al. (2003) documented recovery of soil respiration to the same
rate as that of a mature site in Canadian boreal forests between 10-30 years
after fire. In the jack pine forests of Michigan, soil respiration also begins to
increase in the same time frame. I attribute the increase in soil respiration to
increasing jack pine biomass and forest floor accumulation. This time period
coincides with both the highest rate of jack pine growth and a rapid increase in
forest floor carbon (Figure 3A). Rapid jack pine growth aboveground likely
corresponds with increased belowground production and root respiration (Ryan
et al. 1997). Because the forest floor in boreal forests represents the zone of
greatest microbial activity (Van Cleve and Moore 1978), increasing forest floor
accumulation may lead to greater heterotrophic respiration. Thus, both root and
heterotrophic respiration probably contribute to the increase in soil respiration.
Soil respiration rates declined after 30 years onward. The decline in soil
respiration can result from decreasing substrate quality, decreasing gas

diffusivity through the soil profile or less favorable microclimate for microbial

42

respiration. The decrease in substrate quality can lead to a decline in
heterotrophic respiration. Even though total C: N ratio widened in matured
stands, the relationship between C: N ratio and age was not strong and therefore
may not be a contributing factor to declining soil respiration. However, C: N ratio
may not have been the best predictor of substrate quality in these stands. Not
only may the acidic nature of the coniferous litter contribute to high C:N ratio of
the soil, this litter may also have high lignin content, which also may retard soil
microbial activity (Oades 1988). Furthermore, the presence of ericaceous shrubs
(Figure 18) at some sites may indicate an input of polyphenol compounds into
the soil, which has been shown to affect litter quality (Hattenschwiler and
Vitousek 2000). Thus, in these soils, a different measure such as % lignin or
polyphenols content could be a better indicator of substrate quality. A decline in
diffusivity, associated with greater moisture content, can contribute to declining
soil respiration (Davidson and Trumbore 1995; Risk et al. 2002). Decreasing
diffusivity is an unlikely explanation for declining soil respiration because the data
indicates decreasing soil moisture at older stands. Lastly, decreasing soil
temperature and moisture at older stands may have caused a decline in
heterotrophic respiration.

Soil respiration also varied seasonally (Figure 10) and the variation was
most likely a factor of changing soil temperature and moisture through the
season. Many studies have illustrated the influence of temperature (Edwards
1975, for reviews see Singh and Gupta 1977; Reich and Schlesinger 1992) and

soil moisture (Edwards 1975, Linn and Doran 1984, Raich and Schlesinger

43

1992) on soil respiration. Temperature appears to regulate a general level of
respiratory activity, whereas moisture is a secondary factor dictating the
maximum level of CO2 evolution (Cowling and MacLean 1981). Power,
exponential (010) and Arrhenius equations have often been used to relate soil
respiration to temperature (see Lloyd and Taylor 1994 for review), whereas
quadratic (Howard and Howard 1993) and linear relationships (Davidson et al.
1998) have been used to correlate soil respiration to moisture content or matric
potential. Linear relationships are especially utilized when data on both
temperature and moisture are included in the analysis (eg. Gupta and Singh
1981)

Mean temperature during the incubation, as well as soil and forest floor
moisture were significant predictors of soil respiration but explained only 44
percent of the variation. Poor prediction by these parameters may be due to
several reasons. (1) Gravimetric soil moisture was measured one day after the
end of the respiration measurement and therefore may not represent the
conditions while soil respiration was measured; (2) moisture and temperature
were not measured at the same point as the soil respiration;(3) other factors,
such as 0 quantity (Trumbore et al. 1995) and quality, as well as diffusivity
(Davidson and Trumbore 1995) affect soil respiration but were not included in the
analysis because those factors were not measured for every soil respiration
sampling point; (4) gravimetric soil moisture might be a poor predictor in soils
that are prone to desiccation, such as the sandy soil in these stands, and matric

potential is a more appropriate expression of soil water status (Davidson et al.

44

1998). Some of the variation in soil respiration could have been due to changes
in seasonal root turnover and root respiration, but these parameters were not
measured.

In a review of soil respiration rates from a variety of terrestrial habitats,
Raich and Schlesinger (1992) found a significant linear trend between soil
respiration and annual temperature, but the models explained only 49 and 31
percent of the variation in temperate coniferous and temperate deciduous
forests, respectively. Depending on the season, volumetric water content
explained between 22 and 48 percent of variation in soil respiration in a mixed
hardwood forest (Davidson et al. 1998). Thus, temperature or soil moisture alone
do not always correlate well with soil respiration. Studies of soil respiration rates
in jack pine stands of eastern Ontario (Weber 1985) found similar CO2 flux rates
ranging from 45-56 mg C hr'1 m'z. The stands ranged from 6-63 years of age and
regenerated from either wildfire or experimental burns. Some differences
between stands were documented and attributed mostly to differences in N
pools.

My prediction that soil respiration continues to increase after wildfire is not
supported. These data suggest a different trend in recovery controlled by
substrate quantity for heterotrophic respiration in early succession, substrate
quantity and biomass accumulation at middle stages and substrate quality and
root respiration in mature stands. Interestingly, the recovery of soil respiration is
similar to N mineralization, suggesting that labile substrate is being exhausted for

both C and N mineralization.

45

Vegetation

Changes in live understory biomass appear to be driven by an increase in
resource availability immediately after fire, and then by decreased solar radiation
as jack pine stands develop. The immediate response to wildfire is an increase in
live understory vegetation biomass (Figure 17). This recovery period is likely
enhanced by post-fire nutrient and light availability. The decrease in N
mineralization and nutrients in the first decade most likely limits further
understory biomass accumulation. Furthermore, jack pine biomass begins to
increase (Figure 16) and therefore limits light penetration to the forest floor. The
increase in the understory biomass in mature stands may be caused by
increased light penetration to the forest floor as jack pine mortality increases and
stands start to break up. Even though I did not measure light penetration directly,
I did observe that stand age of 27 is much more open than the other older stands
allowing higher understory vegetation growth. This is illustrated by the
abundance of reindeer moss (Cladonia spp.), which requires high light and
thrives in open canopy and tundra habitats (Ahmadjian and Hale 1973; Cringan
1957)

Jack pine biomass accumulation conforms to a classic pattern of slow
increase, then rapid increase, and followed again by a period of slow increase
(Figure 16) (Ryan et al. 1997). Understory biomass serves as a source of forest
floor in the first decade following wildfire. As jack pine growth entered a period of
exponential growth, forest floor accumulation was at its highest rate. Jack pine

needle litter and tree mortality became the main source of forest floor inputs at

46

this time allowing for increased N availability. It appears that understory
vegetation litter may serve as source of available N for initial jack pine
exponential growth, but the cycling of the needle litter sustains further jack pine
growth.

Jack pine productivity at maturity may be partially limited by N availability.
Single applications of N fertilizer in jack pine have resulted in greater total volume
growth (Morrison and Foster 1990; Weetman et al. 1987). Vogel and Gower
(1998) also demonstrated greater overstory C in boreal jack pine forest growing
with green alder compared to stands growing without alder. They partially
attributed higher growth in the former to greater N availability, associated with N-
fixing ability of green alder. However, Foster and Morrison (2002) showed that C
accumulation in a productive semi-mature stand in Ontario was weakly
constrained by N. Therefore, jack pine productivity is site dependant and may

depend on inherent soil fertility and N turnover.

CONCLUSIONS

In the jack pine forests of Michigan, rapid loss followed by gradual
accumulation of forest floor after wildfire drives N mineralization and soil
respiration dynamics. The forest floor contains a large pool of nutrients and a
source of N and C for soil microorganisms, which make inorganic N available for
plant uptake. This pool is rapidly volatilized by wildfire and takes approximately
40 years to recover. If Michigan jack pine productivity is limited by N availability,

a disturbance frequency of more than 40 years is required to sustain maximum

47

yield without fertilization. If the time interval between disturbance events is less
than 40 years, loss of nutrients during a wildfire or removal of nutrients in a clear-
cut may lead to a remaining nutrient supply that is insufficient to sustain growth

or biomass accumulation levels previously observed.

48

CHAPTER 3
SUMMARY 8: CONCLUSIONS

In Michigan jack pine forests, the recovery of nitrogen mineralization and
soil respiration is largely driven by accumulation of forest floor and occurs
approximately 40 years following wildfire.

Understanding changes in soil microbial community composition may
provide further insights into the mechanisms that drive the observed
recovery patterns.

Even though understory vegetation was measured in this study, a more
intense survey of understory species composition may also prove helpful
in understanding the role of certain plants, such as mosses and
ericaceous shrubs, in N and C cycling.

Jack pine in Michigan grows on sandy outwash plains. Because these
soils are inherently infertile, jack pine productivity may largely be
dependant on the accumulation of forest floor for a continuous supply of
nutrients. However, whether or not jack pine productivity in Michigan is
actually limited by N availability is uncertain. Fertilization trials at different
stand ages should provide useful information regarding if and when N
limitation to productivity is greatest.

In order to make better forest management decisions, comparisons need
to be made between forest recovery patterns after wildfire and after

harvesting.

49

APPENDIX A:

Tables

50

Table 1. ANOVA results for monthly total N mineralization rates.

Adjusted R2= 0.465

 

 

 

 

 

Source Sum-of-Squares df Mean-Square F-ratio p-value
AGE 17071.606 10 1707.161 5.139 0.000
MONTH 1242.849 4 310.712 0.935 0.446
AGE*MONTH 13457.850 40 336.446 1 .013 0.464
Error 36542.1 96 1 10 332.202

Table 2. ANOVA results for monthly soil respiration rates.

Adjusted R2=0.888

Source Sum-of-Squares df Mean-Square F-ratio p-value
AGE 4909365939 1 0 490936.594 1 6.021 0.000
MONTH 1 .4621 1 E+07 4 3655286721 1 19.282 0.000
AGE*MONTH 7225093773 40 1 80627.344 5.894 0.000
Error 3370851 .926 1 10 30644.108

Table 3. ANOVA results for monthly soil moisture.

Adjusted R2=0.549

Source Sum-of-Squares df Mean-Square F-ratio p-value
AGE 137.533 10 13.753 2.300 0.017
MONTH 361 .808 4 90.452 15.128 0.000
AGE*MONTH 301.134 40 7.528 1.259 0.175
Error 657.703 110 5.979

 

51

Table 4. Age, location, vegetation and soil properties of chronosequence sites.

Ammo? 3039308 350:. LoumEoLEE 803258 B nos—Ewan .coutoc m econ: Em
.mmmE .3 mucoEmmc omomoo .x. or A 5? 556: coatomoom coo «mum. on Co 8583 65 3529: we?

.3me Lcumiuzom .3. En

Anni moo: b226>o .92..N
do? .. moon Emma ucmfi .maotmoi .88 wow. xom?

 

 

e oz to 82 8 3.3 .303 as
o ez ems em? 8 .53 .803 mm
m 8> :4 es: 8? .303 .803 on
e oz Rs we? 8? .803 .803 R
e. oz 3.4 ex: 2: .808 .803 «N
2 we, 8.4 88 SF .203 .53. 3
k me, me... SR so ones .93 NP
2 8> 43 E2 3 .203 .803 k
s oz 8.4 oz oz .803 .803. v
2 oz 2.4 oz oz .803 .803 N
: ez ewe oz oz .203 .503 F
es 390 + so .23 .996 are Sea; 4. Nasceo 5% .25 some 3. 9a 95.. g 8.. E em<

52

Table 5. ANOVA results for potential nitrogen mineralization.
(T=temperature W=matric potential)

 

 

 

R2= 0.812

Source Sum-of-Squares df Mean-Square F-ratio p-value
STAND 14.526 3 4.842 19.773 0.000
T 150.080 3 50.027 204.302 0.000
‘4’ 18.982 3 6.327 25.840 0.000
STAND*T 2.229 9 0.248 1.01 1 0.432
STAND* 9’ 7.360 9 0.818 3.340 0.001
T* 9’ 6.514 9 0.724 2.956 0.003
STAND*T* 9’ 3.058 27 0.113 0.462 0.990
Error 47.014 192 0.245

Table 6. ANOVA results for potential carbon mineralization.

(T=temperature W=matric potential)

R2: 0.821

Source Sum-of-Squares df Mean-Square F-ratio p-value
STAND 52.636 3 17.545 35.468 0.000
T 232.184 3 77.395 156.456 0.000
9’ 112.262 3 37.421 75.647 0.000
STAND*T 3.886 9 0.432 0.873 0.551
STAND* 9’ 9.102 9 1.011 2.044 0.037
T* ‘P 13.701 9 1.522 3.077 0.002
STAND*T” 9’ 11.981 27 0.444 0.897 0.616
Error 94.977 192 0.495

 

53

Table 7. ANOVA results for relative nitrogen mineralization.

(T=temperature LIJ=matric potential)

 

 

 

R2: 0.801

Source Sum-of-Squares df Mean-Square F-ratio p-value
STAND 16.569 3 5.523 20.902 0.000
T 150.080 3 50.027 189.332 0.000
‘1” 18.982 3 6.327 23.947 0.000
STAND*T 2.229 9 0.248 0.937 0.494
STAND* \P 7.360 9 0.818 3.095 0.002
T* ‘1’ 6.514 9 0.724 2.739 0.005
STAND*T* ‘41 3.058 27 0.113 0.429 0.994
Error 50.732 192 0.264

Table 8. ANOVA results for relative carbon mineralization.

(T=temperature W=matric potential)

R2= 0.811

Source Sum-of-Squares df Mean-Square F-ratio p-value
STAND 1 .272 3 0.424 23.937 0.000
T 8.082 3 2.694 152.141 0.000
9’ 3.915 3 1.305 73.701 0.000
STAND*T 0.109 9 0.012 0.685 0.722
STAND* 4’ 0.335 9 0.037 2.100 0.031
T* \P 0.480 9 0.053 3.015 0.002
STAND*T* 91 0.413 27 0.015 0.864 0.662
Error 3.400 192 0.018

 

54

Table 9. ANOVA results for potential nitrogen immobilization.

(T=temperature LIJ=matric potential)

 

R2: 0.482

Source Sum-of-Squares df Mean-Square F-ratio p-value
STAND 25762.202 3 8587.401 4.444 0.005

T 144856.465 3 48285.488 24.988 0.000

W 63286.043 3 21095.348 10.917 0.000

STAND*T 18728.059 9 2080.895 1 .077 0.382

STAND* 9’ 3280.379 9 364.487 0.189 0.995

T* 9’ 60640.643 9 6737.849 3.487 0.001

STAND*T* 9’ 28005.662 27 1037.247 0.537 0.971

Error 371010.549 192 1932.3

 

55

APPENDIX B:

Figures

56

Figure 1. Accumulation of forest floor as a function of stand age after a clear-cut
(Covington 1981).

 

 

g. :T II ...................
35:11.) I H i
gg .: ...... 11

AG E (YEARS)

Figure 2. Site locations and dates of stand origin of Pinus banksiana
chronosequence.

8 0 8 16 24 Kilometers
m

 

N
Bald Hill Fire
+ / I975
/

set... aids:
l9”

 

 

momma.
0110mm I980 ' No Polo Fr:
2000

0 Old M
1930 I 1950

Jacob: Fire
“"1115”. m

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Paul-loll]

MecII Flt 19.“
19% .
Dana: Flt:

1966

 

 

 

 

 

 

 

 

 

 

 

 

 

 

!_—-——'—-"-‘

 

 

 

57

Figure 3. Forest floor (FF) C content (A), N content (B) and C:N ratio as a
function of stand age. Note differences in scale between panels.

4000 —

I

3000 -l

I

FF Carbon (glm‘Z)
N
8
O

FF Nitrogen (glm‘Z)
§

271

 

 

 

 

 

 

 

' i 1
I O

Gigi- #9: _._°_____ . , , ____+_ ____

10 20 30 40 50 60 70

Stand Age (years)
.L
e i f
I
OI—A ‘9. .. . 7 I _-_T --—.
10 20 30 4O 50 60 70
StandAae(years)
0

FFC:NRItIo

24«

214

 

 

T1

y = 0.0405x + 25.418
Adjusted R"2 = 0.096
p=0.1853

 

18%

10

20 30

 

 

40 50 60 70

Stand Age (years)

58

 

Figure 4. Soil C content (A), N content (B) and C:N ratio (C) as a function of
stand age. Note differences in scale between panels.

SoIl Nitrogen (glm‘Z) SoII Carbon (glm‘Z)

Soil C: N Ratlo (glm‘Z)

 

 

 

j 4

 

 

 

 

 

y=2381'EXP(-0.0025‘x)
Adjusted R"2=0.19
‘ p=0.0990
600 i
|
o H3 ”2, fi_. .. . __-- __ -_-_., -.
0 10 20 30 40 50 60 70 80
Stand Age (years)
160 a
a B
120 I
. T .
..I I t + 4
I
I y=(108.0‘239.4)l(239.4+x)
40 2, Adjusted R*2=0.42
p=0.0189
I
o +————— . ms»-.-— . . T . ———~ 2 4*
0 10 20 30 40 50 60 70 80
Stand Age (years)
33 —
C
30 ~
y=21.62'x*0.0232
. Adjusted R"2=0.08
27 . p=0.2043
2‘ I T H
21
I
tel . - _,_ ....__e. _ ___.__ 2 ___ - _ .
0 10 20 30 40 50 60 70 80

Stand Age (years)

59

Figure 5. Total C content (A), N content (B) and C:N ratio (C) as a function of
stand age. Note differences in scale between panels. Unfilled data points were
excluded from the analysis.

Total Nitrogen (glm‘Z) Total Carbon (glm‘2)

Total C: N Ratio

7000 1
6000 1
5000 1
4000 j

 

y=2524+11.26*x
Adjusted R“2=0.50
<5 p=0.0138

 

 

 

3000 W 9 + 42
2000 -+

 

o + _i.. ‘__'—T—_ _ 1 'MT— 1 __ ‘1' r T " "—1 "‘l

10 20 30 4O 50 60 70 80
Stand Age (years)

300 --

250

200

150

100

50

33

30

27

24

21

18

I y=111.1+0.2759*x 3
Adjusted R"2=0.25
1 . p=0.0818

t "' “ 1 i T r fir '—

0 10 20 30 40 50 60 70 80

 

 

 

 

Stand Age (years)

7
I
) y=22.62+0.0445*x)
Adjusted 1242:0325
p=0.0391

«i t i T 7*

 

 

 

 

T7 . : T 7

0 10 20 30 40 50 60 70 80
Stand Age (years)

60

Figure 6. Total annual N mineralization as a function of stand age. Value for
stand age 7 is off the scale (6348.5 11893).

 

 

 

 

 

.. O
‘ 00
if. j
/ i E
X
I: B,
vs. ,1
‘— i
F '1}; 0
Rk I, i ‘0
‘-. J:
.3 ,
8 8. Hi4
0 l ’— o
e E ,_ . .0
V
£1 3: 8. I *5
T ‘8 ‘1? I , 8
9 I; Q I o g
i g I I §
F—Y-“fl' 5
l0 .\ U)
‘7 \.
R \ - ,9,
IO .
O) . -1
00
>1
\ r- 8
m
,2 O
‘_
F4- -—4 II
Q.— —-——-—— —--——~— —”“/—. H—-——l I
J7; T, l . ————,— —____..__ __ I O
V (‘0 N s-
(szm 6w)
“II“ N IBIOJ.

61

Figure 7. Annual and relativized N mineralization in forest floor (A, B) and soil (C,
D) as a function of stand age. Note that scales are different between panels.
Values for forest floor N mineralization, soil N mineralization and soil relative N
mineralization for stand age 7 are off scale and are 1441.0 (1:518), 4907.5
(11586) and 0.046 ($00132), respectively.

 

 

 

 

 

 

m ‘Fg o taco
'0 o 0* F o
l " l "
!
l 8 l8
W i
l3 E . L 8 ‘E
. Q Q
. 5 5
l- 3 5 ls 5.
0‘ '3 O . '3
L8 3 is g
0' l0 {
O Lo *4 o
“‘ r N
l 3’
. - e
l . g i
. f , v 4 o
3. 8. 8. 5. °
C O O 0
(Moms) (MIND)
u! m N null-u u m In N uni-Ion uos
- g 8
< 0 l
w 2 0 l2
~ 8 l8
L ‘ A o .4 L8 ....
I
5 .5
8 3. 9 §
H‘d E ’ l ‘8
r 8 g r 8 g
w W
. h o b. l— 8
N
z, o z - a
. wit—.4 i
. ,‘-—,j%4. o :9—5 . , . o
8 O
.8. s a s ° 3 § 2 9.
(szmbw) (szmflwl
um N Joouuuog um N nos

62

Figure 8. Percent of total nitrogen mineralized from forest floor as a function of
stand age.

 

 

 

O
l 00
4 a (D l o
\ F. {B "
\ N o'
\\ ‘— II ‘- ,
\ + N o
\ x 3: O- l
\ 00 o H o
*2 8 a 0
\\ O a I
§ " 2 I
\ " L 8 ~
\ l (D
\ ' L.
\ 1 cu
\ o
\ >
\ V
I o 0’
\ l V O)
, <
w \ 13
C
\ :3
O (D
\\ (‘0
\\ l———. l l
\ l
\
\\
\x —I o
\ l N
l, _____ \, ‘4‘.
l— ——-—\“ Q- ,fi l
. O
F—O—~ V“
+— _ .__ . M _ _. l‘
____ e _- ., __ __;—k°\‘+. __. o
I’— —T 4— T
O O O O O O O
(0 ID V 0') N ‘-

63

Figure 9. Soil respiration for the growing season as a function of stand age.

 

 

 

 

F O
(I)
W
l o
‘ l\
i
l 8
W
o
w
h
l u
0
5 >
v
o a:
V U)
l <
** 2
8
(I)
L O
- (O
F*-—l
w 3
l. o
i N
. :
l 52
m
0|
W
F f T . I T l I o
O O O O O O O
O to O to 0 L0
or) N N ‘- ‘-
(szIB)

uogmudseu nos

Figure 10. Monthly soil respiration rates as a function of stand age.

 

Stand Age (years)

 

 

L-
03 N
.0
6%
o 3...;
gczgag
a:
2fifi<¢0 ‘-
III--1 -
N
‘—

 

(sz/Ju/o 6w)
uogeudseu nos

65

Figure 1 1. Total exchangeable bases as a function of stand age.

3.805 ou< 33m
cm on cm on ow om om

 

 

 

 

58.90.
agony». 8332
Axfiifimiamvmdl

 

06
f md
T o._.
.. m...

__ 4 em

1 gm
1 o.m
I m.m
I 0*

 

-1 mé

on

(nos :0 Snow) sesea

emeefiueuoxg mo;

66

Figure 12. Soil (A) and forest floor (B) pH as a function of stand age. Errors bars
are missing where samples were composited to yield n of 1.

4.50 l
4.40
4.30
4.20
4.10

4.00
y = -0.0001x2 + 0.0014x + 4.2349

3.90
Adjusted R"2=0.68 {
3-30 p=0.0043

3.70
3.60
3.50 «»—e—— r- . .-_..

41

     
 

J_..__.L_t_4 _4__i

Soil pH

i

 

; .

l
‘1'".

 

 

 

 

0 10 20 30 40 50 60 70 80
Sta nd Age (years)

 

   
 

y = -0.0111x + 4.2301
3.6 l Adjusted R"2=0.752
p<0.001

Forest Floor pH

0)

A
g #0,...
O

9’
N

 

 

 

(A)
o fvv ...z_‘.L.

10 20 30 40 50 60 70 80
Stand age (years)

67

 

Figure 13. Total number of degree-days in the growing season as a function of

stand age.

 

 

 

A335 om< ucflm
om cm. 3 on cm 2 o
.!._l} lL‘ll. l- z , F . IlliL|| . , Ill 9
08
$8.03 . 82
voouem Baa?
03.3 + xmmmmh- u > com?
z.ooo~
4;;;:lsll4l-lsllvll;l . W
. -l---ll+wll-l.+- 8mm
4 . coon

68

silep eeJBep to #

Figure 14. Seasonal gravimetric percent moisture in forest floor (A) and soil (B)
as a function of stand age. Errors bars are missing where samples were
composited to yield n of 1.

 

mil
1
J

72

I!

’1

:1

2211;“

Stand Age (years)
1
Stand Age (years)

tlilml

1 June
Y
ugust
a September
' . [E

I May

 

5%
1
1

 

é 8 3.2—9 E; a 2 o a a $2 2 m o
NmSIOW y. “Milo“! “I.
we” new; "OS

69

Figure 15. Average moisture tension data at selected stands at 5 different
sampling periods. May data were not available for stand age 22. The higher the
value, the lower the moisture content.

I Age 2

I Age 12
I Age 22
[:1 Age 36
I Age 72

i—
o
.o
E
9.9.
o.
m
(I)

 

Sampling Month

 

 

 

 

 

 

T T T l I

O O O O O
m (D V N

120
1
100 a

lad») uogsue: efiuenv

70

Figure 16. Jack pine biomass as a function of stand age.

120 -.

l
100 l

     
 

y = 97.04*(1-exp(-0.0661*x))3.627
Adjusted R"2=0.983

Jack Pine Biomass (Mglha)
O)
O

 

401 P<0.001

20 l

0 l . » . _. . .
0 10 20 30 40 50 60 7o 80

Stand Age (years)

Figure 17. Live understory biomass as a‘function of stand age.

0.7 1

.0
U"
_L

Understory Biomass
(kgIM‘Z)
O O
on «b
l—e—l

 

on; 1 f i 1

 

 

 

0 ‘1—” _"" 0"“ ' 'r T 4V " — l d 1
0 10 20 30 40 50 60 70 80
Stand Age (years)

71

 

Figure 18. Biomass of selected understory vegetation categories as a function of

stand age.

 

 

sz/B

 

 

 

 

 

Emma 0?. 28m Adamo dm< 20m
ow on cm on 0v on ON 9. cm on ow om ov on on or o
_1-I $21 71% 1%:114L-1W1.“ o _I- Z.» .1d-1HrI1-L101 W 101.- .T O
- w --
H kw - ON e 4;
- 4 - ov
a 4 A - cc 3. W
_ 4 m - 8
v 00 Z .
.-. - om % - our
Eotoozw I oo_. EmeuBEEEO . 8..
08¢ om< 29w . Ewes 0?. 23m
cm 2. 8 on ow on cm or o ow on cm on ov om ow or o
” T1 1.‘L11L. ...1.- 01.111, 11WT O _1III-,|r 1-11 1: o
4 w w I. - ov
F .._ MI 00—. _ W
s. w a . m . . 8 m
H - com 4
N z A mom? z
A - 8m . m.- cm.
325 3088.5. .- oov . 009.22 _I CON

72

Figure 19. Monthly microbial biomass in the soil as a function of stand age.

N
[\

  

 

til

I May
I June
1*
1,.
52

7 —
_} i7 , o
(‘0
, ,, N
__ N
, , , ., -
. N A
,, N 2
-- . a
; 0
— >'
, ' V
%*—- , 7,, ' v a
‘_
4 <
- U
, ,,,,,,-_ =
11 N 8
‘— w
777777 -
_ “
, v
’:' ' N
1
a _
#"t ,g_
‘_
,,,, a,,_,
O O O O O C) O O
l0 0 '4') O to O
00 (‘0 N N ‘- ‘-
szmfiw

Figure 20. Available phosphorus as a function of stand age.

 

 

 

 

1'
. $
T
, i
[
W+
o .
e
;
0 i
'0‘ |
J
‘_’V____ ___fi- I
". I ”T r —T—.‘.—‘:—_4 T ‘
0. *0. 0. l0 0. l0 0.
(O N N ‘- ‘- O O
(zvuufi) smoquoqd

74

O
a)

30 4O 50 60 70
Stand Age (years)

20

10

6 $293 9283th 6 $283 58858.

mm on on cm 9 2 m o mm on mm 8 2 2 m o

L I'll} a. 0 Pl .l... P}‘ it”. -zlLl LIL'- lLlilIliLIl ‘ C

"‘

\iuuqflJu......

o ||\|| ||I§.\\..\w _

\.w ........ e
|I\\‘I‘. \\ .TNF

x9 \ . \

OOON I'll .. mp
89 . ...I
89 -III , vu
89 . I. .

 

CD

 

coom III
82 . .I
. 82 IT
Y on 82 . t .
.2 P.

Q
\
\
mos 6/3 6n) mg :03
‘ \
\ s
\
\\'.
Z \
I. .0
N 1-
(IIOS 51:) 6n) xnli ZOO

 

 

8 8

han m F-

8

8 $233 2.5858 . G 8283 «5.8828

1.0
O

 

 

i

 

 

 

O
O
z
éw
x
25m
Soul-II O
82.7 #«W
©8w+ Tommi. m .\ r .l.
82.... e ( 1 .33 82.? (
8 . -8

Figure 21. Potential carbon mineralization as a function of temperature for four

moisture treatments and four stands.

75

Figure 22. Potential nitrogen mineralization as a function of temperature for four

moisture treatments and four stands.

6 000503 0.30.033...
cu m w o F

mm on 0N

6 000503 0.380QE0H

*. NH

.50 06.

ON

 

ill l III

mr

Ill .lol [I |.

or

 

 

 

 

m o
Lrll a o
I w Nd
e to
OOON o .
89 4 T o o
comp O T 0.0
. N?
m 0
IF‘ A o
w
to
._ 9o
cm? 4 e F
009 O
.. 3
one I

(nos 5N 6n)

”OMB-BU!“ N

(nos BIN 6n)

@2218me N

3 000.603 93903.0...

 

 

 

 

 

mm on ma cm 9 o. m o
_ 3 l . sLI .Lla- 1!».IrlllLll -L ..-.I|_t o
w u m _,. No N
. I . ) m.
coon . co m m
0 0mm—. 4 , 00 m m.
4_ 08? O I 0.0 0 w.
m 82 n ,e P
.3 F- __- 3
6 30603 05.93th
mm on mm ow 9 9 m o
r|l||llb I H.l . . y w . _F o
I
a a a .
W i No \I N
+ 08: -vd .m m.
w 82. . mm
. no ( m
89-
V. P
.3 v? ,, we

76

Figure 23. Potential nitrogen immobilization as a function of temperature for four

moisture treatments and four stands.

 

6 000.003 0.20.380...

 

 

GOON III-
ommw .. II
08.. llll

ommr u .I. ..
.5an

8 000503 0.30.0080...

 

 

.2 mac-

#0
o

 

uogezmqouw|

(N 61:) 6) lenueaod

 

-ov

 

l
O
N
F

0
co
(N Bro 6) lenuewd
ummuqouwu

O
(D
P

—. CON

8 80.503 0.30.8 808

mm on mm

_ F F

   

urn-urn

OOON III:
000—. u II
89. +
ommv u .I. u

ON

or

0..

 

l. . . l
\
\e
l x

    

.09 T

A0 000.603 0.392.808

 

J.
If

.00 v.0-

 

 

o
no
(N 5K) 5) amend
uoqezmqouiwl

77

APPENDIX C:

Additional Methods

78

 

Microbial Biomass

Soil microbial biomass N was measured five times on mineral and organic soil
separately. Soil microbial biomass N was determined using the chloroform
fumigation-extraction technique (Beck et al. 1997; Brookes et al. 1985). Two
subsamples of the soil cores collected for determination of initial extractable
NH..*- and N031 N (see above) were used to measure microbial biomass N. One
subsample of 20 g mineral and 5 9 organic fresh weight soil was shaken for 1
hour with 50 ml 0.5 M potassium sulfate and then extracted using Whatman #1
filter paper. The second subsample of 20 g mineral and 5 9 organic soil was
fumigated with chloroform in a desiccator for 5 days followed by extraction with
0.5 M K2804 (Cabrera and Beare 1993).‘ The extracts were analyzed for total N
by persulfate digestion, followed by colorimetric analysis for N03' on the Alpkem
Flow Solution lV Auto-Analyzer. Results are presented on an oven dry (105°C for

48 hours) basis.

Understory Composition

Live biomass was sampled and separated into eight different categories:
mosses, lichens, ericaceous shrubs (primarily Vaccinium spp.), sand cherry
(Prunus pumila), graminoids and forbs, sweet fern (Comptonia peregrine),
bracken fern (Pten'dium aquilinum) and a miscellaneous category. Samples were
composited by transect, dried in a forced air convection oven at 55°C, and then

weighed.

79

 

Phosphorus

Available phosphorus was determined using the Mehlich 3 method (Mehlich
1984). Two grams of air-dry soil was extracted with Mehlich 3 solution (acetic
acid, ammonium nitrate nitric acid, ammonium fluoride,
ethylenediaminetetraacetic acid). A working solution (ammonium molybdate,
antimony, potassium tartrate, sulfuric acid) was added to the filtrate for color
development, which was analyzed at a wavelength of 882 nm on a

ThermoSpectronic spectrophotometer.

Incubation Experiment
To determine the relative importance of microclimate in driving N and C
dynamics of a forest recovering from a wildfire disturbance, I collected soil from 4
jack pine forests of the following ages: 2, 12, 36, 72. Soil was collected in
September 2002. I used stratified random design to sample 3 points along each
transect. Soil along each transect was composited. I collected soil to a 10-cm
depth using a 5.08-cm diameter core. Additional three samples were collected
and composited at each stand for an additional replicate. The soils were kept in a
cooler for transport and thereafter stored in the refrigerator until processed
through a 4 mm sieve within 48 hours. The roots were removed, the soil
homogenized by hand and air-dried at room temperature.

I incubated 10-g soil subsamples from each transect in plastic sample cups
at 16 different temperature (6°C, 15°C, 24°C and 33°C) and matric potential (-

0.1, -0.5, -1 and —1 5 bars) combinations that encompass field conditions. Prior to

80

 

the start of the incubation, I used a ceramic-membrane pressure extractor (Soil
Moisture Corp., Santa Barbara, CA) to determine the volumetric moisture content
of the soils samples that correspond to the matric potentials used in the study. I
adjusted each sample gravimetrically to the desired matric potential by misting

the soil with deionized water and then mixing the soil to evenly distribute

moisture.
E;
The soil samples were then placed into 924-mL glass Mason jars sealed
l
with lids containing rubber septa for gas sampling. The jars were then placed in ',_
incubators and maintained at the different temperature treatments. Carbon L.

 

dioxide evolution was measured 9, 19 and 26 days after the start of the
incubation using a Tracor 540 gas chromatograph (T racor Instrument, Austin,
TX). Soil moisture was adjusted gravimetrically after each gas sampling. Carbon
flux was determined as the cumulative C02 divided by days of incubation per
gram of soil.

To measure potential N mineralization, l extracted a 10-g fresh weight
subsample of each soil at the start of the incubation and each incubated soil
sample at the end of the incubation. The samples were shaken with 50 mL of 2
M potassium chloride (KCL) for 1 hour to extract NHJ- and N031 N. The extracts
were poured through Whatman #2 filter paper and stored frozen until analysis for
inorganic nitrogen using the Alpkem Flow Solution IV Auto-Analyzer. An
additional soil subsample was used to determine gravimetric soil moisture.
Potential N mineralization rates were calculated on an oven-dry (105°C for 48

hours) mass basis as soil increase in inorganic NH..*- and N031 N over the

81

incubation period. Carbon and nitrogen content was determined on an additional
subsample of the soil. Carbon and nitrogen content was determined on air-dry
ground samples and analyzed on the Carlo Erba CN analyzer. Percent C and %
N results are expressed on a 105°C oven-dry soil mass basis.

The effect of stand age, soil moisture and soil temperature on potential
and relative N and C mineralization, as well as N immobilization was analyzed
using three-way analysis of variance. Potential and relative nitrogen
mineralization, and N immobilization data were ln-transformed, whereas potential
and relative C mineralization data were transformed using square root to
normalize the data. Significance was accepted at minimum alpha of 0.05. All

analyses were analyzed with statistical package SYSTAT (SPSS, Inc.)

82

REFERENCES

Ahmadjian, V., and Hale, M. E. 1973. The lichens. New York: Academic Press.
697 pp.

Albert, D.A., Denton, SR, and Barnes, B.V. 1986. Regional landscape
ecosystems of Michigan. School of Natural Resources. University of
Michigan, Ann Arbor, MI. 32 pp.

Amiro, B.D., MacPherson, J.l., Desjardins, R.L., Chen, J.M. and Liu, J. 2003.
Post-fire carbon dioxide fluxes in the western Canadian boreal forest:
evidence from towers, aircraft and remote sensing. Agricultural and Forest
Meteorology 1 15: 91 -1 07.

Apps, M.J., Kurz, W.A. Luxmore, R,J., Nilsson, L.O., Sedjo, R.J., Schmidt, R.,
Simpson, LG. and Vinson, T. 1993. The changing role of circumpolar
Boreal forests and tundra in global C cycle. Water Air Soil Pollut. 70: 39-
53.

Austin, R. C. 8. Baisinger, D. H. 1955. Seme effects of burning on forest soils of
Western Oregon and Washington. Journal of Forestry 53: 275-280.

Bauhus, J., Khanna, P. K. & Raison, R. J. 1993. The effect of fire on carbon and
nitrogen mineralization and nitrification in an Australian forest soil.
Australian Journal of Soil Research 31: 621-639.

Beck, T., Joergensen, G., Kandeler, E., Makeschin, F., Nuss, E., Oberholzer,
HR, and Scheu, S. 1997. An inter-laboratory comparison of ten different
ways of measuring soil microbial biomass C. Soil Biology and
Biochemistry 29: 1023-1032.

Bergeron, Y. and Dubuc, M. 1989. Succession in the southern part of the
Canadian boreal forest. Vegetatio 79: 51-63.

Birdsey, RA. 1990. Inventory of carbon storage and accumulation in US. forest
ecosystems. Pages 24-31 in HE. Burkhart et al. (ed). Research in forest
inventory, monitoring, growth and yield. Proc. IUFRO World Cong.,
Montreal. 5-11 August 1990. School of For. and Wildl. Resour. Publ. FSW-
3090. Virg. Polytech. Inst. and State Univ., Blacksburg.

Bissett, J. & Parkinson, D. 1980. Long-term effects of fire on the composition and

activity of the soil microflora of a subalpine, coniferous forest. Canadian
Journal of Botany 58: 1704-1721.

83

Bolin, B. and Cook, R.B (eds.). 1983. The major biogeochemical cycles and their
interactions. Wiley, Chichester, UK. 532 pp.

Bormann, B.T. and Sidle, RC. 1990. Changes in productivity and distribution of
nutrients in a chronosequence at a Glacier Bay National Park, Alaska.
Journal of Ecology 78: 561-578.

Bormann, PH. and Likens, GE. 1967. N cycling. Science 155: 424-429.

Bouyoucos, G.J. 1962. Hydrometer method improved for making particle size
analysis of soils. Agric. J. 54: 464-465.

Brais, S., Camiré, C., Bergeron, Y. and Pare, D. 1995. Changes in nutrient
availability and forest floor characteristics in relation to stand age and
forest composition in the southern part of the boreal forest of northwestern
Quebec. Forest Ecology and Management 76: 181-195.

Brookes, P.C., Landman, A., Panen, G., and Jenkinson, BS. 1985. Chloroforrn
fumigation and the release of soil nitrogen: a rapid direct extraction
method to measure microbial biomass nitrogen in soil. Soil Biology and
Biochemistry 17: 837-842.

Cabrera, ML, and Beare, M.H. 1993. Alkaline Persulfate Oxidation for
Determining Total Nitrogen in Microbail Biomass Extracts. Soil Science
Society of America Journal 57: 1007-1012.

Cameron, H. 1953. Melting point of the bonding material in lodgepole and jack
pine cones. Can. Dep. Resour. Develop., For. Branch., Silvic. Leafl. 86. 33
PP-

Cayford, J.H. and McRae, DJ. 1983. The ecological role of fire in jack pine
forests. Pages 183-199 in: Wein, R.W. and Maclean, D.A. (eds). The role
of fire in northern Circumpolar ecosystems. Wiley and Sons, New York.

Christensen, ML. and Muller, CH. 1975. Effects of fire factors controlling plant
growth in Adenostoma Chaparral. Ecological Monographs 45: 29-55.

Cole, D., and Rapp, M. 1981. Elemental cycling of forest ecosystems. Pages
341-411 in DE. Reichle, editor. Dynamic properties of forest ecosystems.
Cambridge University Press, New York.

Covington, WW. 1981. Changes in forest floor organic matter and nutrient
content following clear-cutting in northern hardwoods. Ecology 62: 41-48.

Cowling, J.E. and MacLean, Jr., SF. 1981. Forest floor respiration in a black
spruce taiga forest ecosystem in Alaska. Holarct. Ecol. 4: 229-237.

84

Cringan, AT. 1957. History, food habits and range requirements of the woodland
caribou of continental North America. Transactions, North American
Wildlife Conference. 22: 485-501.

D’Arrigo, R., Jacoby, GO. and Fung, |.Y. 1987. Boreal forests and atmosphere-
biosphere exchange of carbon dioxide. Nature 329: 321-323.

Davidson, E.A., Belk, E., and Boone, R. 1998. Soil water content and
temperature as independent or confounded factors controlling soil
respiration in a temperate mixed forest. Global Change Biology 4: 217-
227.

Davidson, E.A. and Tmmbore, SE. 1995. Gas diffusivity and production of C02
in deep sols of the eastern Amazon. Tellus 478: 550-565.

DeBano, L.F., Neary, D.G., and Ffolliott, RF. 1998. Fire Effects on Ecosystems.
John Wiley & Sons, New York.

DeBano, L.F. 1981. Water repellent soils: a state-of-the-art. USDA Forest
Service General Technical Report. PSW-46. pp. 21.

De Grandpré, D., Gagnon, D., and Yves, B. 1993. Changes in the understory of
Canadian southern boreal forest after fire. Journal of Vegetation Science
4: 803-810.

DeLuca, T.H., Nilsson, M.-C. and Zackrisson, O. 2002. Nitrogen mineralization
and phenol accumulation along a fire chronosequences in northern
Sweden. Oecologia 133: 206-214.

Dunn, PH. and Debano, L.F. 1977. Fire’s effect on the biological properties of
Chaparral soils. In: lntemational Symposium on the Environmental
Consequences of Fire and Fuel Management in Mediterranean-climate
Ecosystems (Forests and Scrublands).

Dunn, P. H., DeBano, L. F. 8. Eberlein, G. E. 1979. Effects of burning on
Chaparral soils: ll Soil microbes and nitrogen mineralization. Soil Science
Society of America Journal 43: 509-514.

85

Dymess, C.T., Van Cleve, K., and Levison, JD. 1989. The effect of wildfire on
soil chemistry in four forest types in interior Alaska. Canadian Journal of
Forest Research 19: 1389-1396.

Edwards, NT 1975. Effect of temperature and moisture on carbon dioxide
evolution in a mixed deciduous forest floor. Soil Sci. Amer. Proc. 39: 361-
365.

Edwards, N.T. 1982. The use of soda-lime for measuring respiration rates in
terrestrial systems. Pedobiologia 23: 321-330.

Ewel, K.C., Kropper, WP. and Gholz, H.J. 1987a. Soil COz evolution in Florida
slash pine plantations. l. Changes through time. Canadian Journal of
Forest Research 17: 325-329.

Ewel, K.C., Kropper, WP. and Gholz, H.J. 1987b. Soil C02 evolution in Florida
slash pine plantations. ll. Importance of root respiration. Canadian Journal
of Forest Research 17: 330-333.

Flanagan, P.W., and Van Cleve, K. 1983. Nutrient cycling in relation to
decomposition and organic matter quality in taiga ecosystems. Canadian
Journal of Forest Research 13: 795-817.

Flannigan, MD. and Van Wanger, CE. 1991. Climate change and wildfire in
Canada. Canadian Journal of Forest Research 21: 66-72.

Fonturbel, M.T., Vega, J.A., Bara, S., and Bemardez, I. 1995. Influence of
prescribed burning of pine stands in NW Spain on soil microorganisms.
European Journal of Soil Biology 31: 13-20.

Foster, NW. and Morrison, |.K. 2002. Carbon sequestration by a jack pine stand
following urea application. Forest Ecology and Management 169: 45-52.

Foster, NW. and Morrison, I.K. 1976. Distribution and cycling of nutrients in a
natural Pinus banksiana ecosystem. Ecology 57: 110-120.

Froberg, M., Berggren, D., Bergkvist, 8., Bryant, C. and Knicker, H. 2002.

Contributions of Oi, Oe and Oa horizons to dissolved organic matter in
forest floor leachates. Geoderrna 113: 31 1-322.

86

Gaarder, T. 1957. Studies in soil respiration in western Norway, the Bergen
District (Naturvitenskapelig rekke Nr. 3) Univeritetetl Bergen Arbok.

Giovannini, G. & Lucchesi, S. 1997. Modifications induced in soil physico—
chemical parameters by experimental fires at different intensities. Soil
Science 162: 479-486.

Green, DC. and Grigal, D.F. 1978. Generalized biomass estimation equations
for jack pine. Univ. Minn. For. Res. Note No. 268, 4pp.

Grier, CC. 1975. Wildfire effects on nutrient distribution and leaching in a
coniferous ecosystem. Canadian Journal of Forest Research 5: 599-607.

Grigal, DR, and Ohmann, L.F. 1992. Carbon storage in upland forests of the
Lake States. Soil Science Society of America Journal 56: 935-943.

Gupta, SR. and Singh, J.S. 1981. Soil respiration in a tropical grassland. Soil
Biology and Biochemistry 13: 261-268.

Harwood, CE. and Jackson, W.D. 1975. Atmospheric losses of four plant
nutrients during a forest fire. Australian Forestry 38: 92-99.

Hossain, A.K.M.A., Raison, R.J., and Khanna, PK. 1995. Effects of fertilizer
application and fire regime on soil microbial biomass carbon and nitrogen,
and nitrogen mineralization in an Australian subalpine eucalypt forest.
Biology and Fertility of Soils 19: 246-252.

Houghton, J.T., Jenkins, G.L., and Ephraums, J.J. 1990. Climate Change: The
IPCC assessment. Cambridge University Press, Cambridge, UK.

Houghton, RA. and Woodwell, GM. 1989. Global climatic change. Sci. Am. 260:
36-44.

Howard, D.M. and Howard, P.J.A. 1993. Relationships between C02 evolution,
moisture content and temperature for a range of soil types. Soil Biology
and Biochemistry 25: 1537-1546.

Johnson, D.W. and Van Hook, RI. 1989. Analysis of biogeochemical cycling
processes in Walker Branch Watershed. Springer-Verlag, New York, 401

PP-
Kabwe, L.K., Hendry, M.J., Wilson, G.W., and Lawrence, JR. 2002. Quantifying

C02 fluxes from soil surfaces to the atmosphere. Journal of Hydrology
260: 1-14.

87

Katznelson, B.H. and Stephenson, LL. 1956. Observations on metabolic activity
of soil microflora. Can. J. Microbiol. 2: 611-622.

Keeney, DR. 1980. Prediction of soil nitrogen availability in forest ecosystems: A
literature review. For. Sci. 26: 159-171.

Klopatek, CC. and Klopatek, J.M. 1987. Mycorrhizae, microbes and nutrient
cycling processes in pinyon-juniper systems. Pages 360-367 in Everett,
R.L. (compiler). Proceedings of the Pinyon-Juniper Conference, January
13-16, 1986, Reno, Nevada. USDA For. Serv. Gen. Tech. Rep. INT-215.

Krause, H.H. 1998. Forest floor mass and nutrients in two chronosequences of
plantations: Jack pine vs. black spruce. Canadian Journal of Soil Science
78: 77-83.

Kutiel, P. and Naveh, Z. 1987. The effect of fire on nutrients in a pine forest soil.
Plant and Soil 104: 269-274.

Linn, D.M. and Doran, J.W. 1984. The effect of water-filled pore space on carbon
dioxide and nitrous oxide production in tilled and nontilled soils. Soil
Science Society of America Journal 48: 1267-1272.

Lloyd, J. and Taylor, J.A. 1994. On the temperature dependence of soil
respiration. Functional Ecology 8: 315-323.

Lynham, T.J., Wickware, GM. and Mason, J.A. 1997. Soil chemical changes and
plant succession following experimental burning in immature jack pine.
Canadian Journal of Soil Science 78: 93-104.

Mehlich, A. 1984. Mehlich 3 soil test extractant: A modification of the Mehlich 2
extractant. Comm. Soil Sci. Plant Anal. 15: 1409-1416.

Melillo, J.M. and Gosz, JR. 1983. Interactions of the biogeochemical cycles in
forest ecosystems. Pages 177-222 in Bolin, B. and Cook, R.B (eds.). The
major biogeochemical cycles and their interactions. Wiley, New York.

Miles, J. 1985. The pedogenic effects of different species and vegetat6ion types
and the implications of succession. Journal of Soil Science 36: 571-584.

Morris, LA. and Miller, RE. 1994. Evidence of long-term productivity change as
provided by field trials. Pages 41-80 in Dyck, W.J., Cole, D.W. and
Camerford, N.B. (eds.). Impacts of Forest Harvesting on Long-tenn site
productivity. Chapman and Hall, London.

Morrison, I.K. and Foster, NW. 1990. On fertilizing semimature jack pine stands
in the boreal forest of central Canada. Pages 416-430 in Proceedings of

88

 

the Seventh North American Forest Soils Conference. University of British
Columbia, Vancouver, BC.

Myneni, R.B. Dong, J., Tucker, C.J. Kaufmann, R.K. Kauppi, P.E. Liski, J., Zhou,
L., Alexeyev, V. and Highes, MK. 2001. A large carbon sink in the woody
biomass of northern forests. Proc. Natl. Acad. Sci. USA 98: 14784-14789.

Nadelhoffer, K.J., Aber, JD, and Melillo, J.M. 1983. Leaf-litter production and
soil organic matter dynamics along a nitrogen-availability gradient in
southern Wisconsin (USA). Canadian Journal of Forest Research 13: 12-
21.

Nalder, LA. and Wein, R.W. 1999. Long-term forest floor carbon dynamics after
fire in upland boreal forests of western Canada. Global Biogeochemical
Cycles 13: 951-968.

Neary, D.G., Klopatek, C.C., DeBano, L.F., and Ffolliott, RF. 1999. Fire effects
on belowground sustainability: a review and synthesis. Forest Ecology and
Management 122: 51-71.

Nilsson, M.-C., Hogberg, P., Zackrisson, O., and Fengyou, W. 1993. Allelopathic
effects by Empetrum hermaphroditum on development and nitrogen
uptake by roots and myccorhizae of Pinus sylvestris. Canadian Journal of
Botany 71: 620-628.

Pare, D., Bergeron, Y., and Camiré, C. 1993. Changes in the forest floor of
Canadian southern boreal forest after disturbance. J. Veg. Sci. 4: 811-818.

Park, J.H., Kalbiz, K. and Matzner, E. 2002. Resource control on the production
of dissolved organic carbon and nitrogen in a deciduous forest floor. Soil
Biology and Biochemistry 34: 813-822.

Pastor, J., Aber, J.D., McClaugherty, C.A., and Melillo, J.M. 1984. Aboveground
production and N and P cycling along a nitrogen mineralization gradient
on Blackhawk Island, Wisconsin. Ecology 65: 256-268.

Pastor, J., Gardner, R.H., Dale, V.H., and Post, WM. 1987. Successional
changes in nitrogen availability as a potential factor contributing to spruce
declines in boreal North America. Canadian Journal of Forest Research
17: 1394-1400.

Paul, EA. and Clark, F8. 1996. Soil Microbiology and Biochemistry. 2"d edition.
Academic Press, CA.

Picket, S. and White, PS. 1985. The Ecology of Natural Disturbance and Patch

89

Dynamics. Academic Press. San Diego, CA. pp.472.

Pietikainen, J., Kiikkila, O., and Fritze, H. 2000. Charcoal as a habitat for
microbes and its effect on the microbial community of the underlying
humus. Oikos 89: 231-242.

Raich, J.W. and Schlesinger, W.H. 1992. The global carbon dioxide flux in soil
respiration and its relationship with to vegetation and climate. Tellus 44B:
81-99.

Raison, R.J., Connell, M.J., and Khanna, PK. 1987. Methodology for studying
fluxes of soil mineral-N insitu. Soil Biology and Biochemistry 19: 521-530.

Raison, R.J, Khanna, PK and Woods, P.V. 1985. Mechanisms of element
transfer to the atmosphere during vegetation fires. Canadian Journal of
Forest Research 15: 132-140.

Risk D., Kellman. L. and Beltrami, H. 2002. Soil C02 production and surface flux
at four climate observatories in eastern Canada. Global Biogeochemical
Cycles 16: art.no. 1122.

Rowe, J.S., and Scotter, G.W. 1973. Fire in the boreal forest. Quat. Res. 3: 444-
464.

Ryan, M.G., Binkley, D., and Fownes, J.H. 1997. Age-related decline in forest
productivity: pattern and process. Advances in Ecological Research 27:
213-262.

Ryel, LA. 1981. Population change in the Kirtland’s warbler. Jack-Pine Warbler
59:76-91.

SAS Institute. 1987. SAS/STAT Guide for personal computers, 6th edition, SAS
Institute, Cary NC, 1028 pp.

Schlesinger, W.H. 1977. Carbon balance in terrestrial detritus. Annu. Rev. Ecol.
Syst. 8: 51-58.

Sharma, G. D. 1981. Effect of fire on soil microorganisms in a Meghalaya pine
forest. Folia Microbiologica 26: 321-327.

Simard, A.J. and Blank, R.W. 1982. Fire history of Michigan jack pine forest.
Michigan Academician 15: 59-71.

Singh, J.S. and Gupta, SR. 1977. Plant decomposition and soil respiration in
terrestrial ecosystems. Botanical Review 43: 449-528.

90

 

Smith, J.L., and Paul, A8. 1990. The significance of soil microbial biomass
estimations. Pages 357-393 in J. Bolagg and G. Stotsky (ed). Soil
biochemistry. Marcel Dekker, New York.

Solinger, S., Kalbitz, K. and Matzner, E. 2001. Controls on the dynamics of
dissolved organic carbon and nitrogen in a Central European deciduous
forest. Biogeochemistry 55: 327-349.

Stocks, B.J., Lee, BS. and Martell, D.L. 1996. Some potential carbon budget
implications of fire management in the boreal forest. Pages 89-96 in Apps,
M.J. and Price, D.T. (eds.). Forest Ecosystems, Forest Management and
the Global carbon Cycle. Springer-Verlag, Berlin. j

Swift, M.J., Heal. O.W., and Anderson, J.M. 1979. Decomposition in Terrestrial ;
Ecosystems. University of California Press, Berkley, CA. 3

 

Tamm, CO. 1991. Nitrogen in terrestrial ecosystems. Springler, NY. 115 pp.

Tiwari, V. K. & Rai, B. 1977. Effect of soil burning on microfungi. Plant and Soil
47: 693-697.

Toland, DE. and Zak, DR. 1994. Seasonal patterns of soil respiration in intact
and clear-cut northern hardweood forests. Canadian Journal of Forest
Research 24: 1711-1716.

Townsend, AR. and Rastetter, EB. 1996. Nutrient constraints on carbon storage
in forested ecosystems. Pages 35-45 in Apps, M.J. and Price, D.T. (eds.).
Forest Ecosystems, Forest Management and the Global carbon Cycle.
Springer-Verlag, Berlin.

Trumbore, S.E., Davidson, E.A., Camargo, P.B., Nepstad, DC. and Martinelli,
LA. 1995. Below ground cycling of carbon in forests and pastures of
eastern Amazonia. Tellus B 47: 550-565.

Van Cleve, K. and Moore, TA. 1978. Cumulative effects of nitrogen, phosphorus
and potassium fertilizer additions on soil respiration, pH, and organic
matter content. Soil Science Society of America Journal 42: 121-124.

Van Cleve, K., and Nooman, LL. 1975. Litterfall and nutrient cycling in the forest
floor of birch and aspen stands in interior Alaska. Canadian Journal of
Forest Research 5: 626-639.

91

Van Cleve, K., Oliver, L.K., Schlentner, R., Viereck, LA. and Dymess, CT. 1983.
Productivity and nutrient cycling in taiga forest ecosystems. Canadian
Journal of Forest Research 13: 747-766.

Van Cleve, K. and Viereck, LA. 1981. Forest succession in relation to nutrient
cycling in the boreal forest of Alaska. Pages 184-211 in West, D.C.,
Shugart, H.H. and Botkin, B.B. (ed.). Forest succession: Concepts and
Applications. Springer, NY.

Vazquez, F.J., Acea, M.J. and Carballas, T. 1993. Soil microbial populations after
wildfire. FEMS Microbiology Ecology 13293-104.

 

Vogel, J.G. and Gower, ST. 1998. Carbon and Nitrogen Dynamics of Boreal
Jack Pine Stand With and Without a Green Alder Understory. Ecosystems
1: 386-400.

Vogt, K.A., Edmonds, R.L., Antos, GO. and Vogt, D.J. 1980. Relationships
between C02 evolution, ATP concentrations and decomposition in four
forest ecosystems in western Washington. Oikos, 35: 72-79.

Walkinshaw, L.H. 1983. Kirtland Warbler: The natural history of an endangered
species. Cranbrook Inst. Science, Bloomfield Hills, MI.

Wang, C.K., Bond-Lamberty, B., and Gower, ST. 2002. Soil Surface C02 flux in
a boreal black spruce fire chronosequence. Journal of Geophysical
Research-Atmospheres 108 (D3): art. no. 8224.

Wardle, D.A., Zackrisson, 0., and Nilsson, M.-C. 1998. The charcoal effect in

Boreal forests: mechanisms and ecological consequences. Oecologia
1 1 5:419-426.

Wardle, D.A., Lavelle, P. 1997. Linkages between soil biota, plant litter quality
and decomposition. Pages 107-124 in Cadisch, G., and K.E. Giller, eds.
Driven by nature. Plant litter quality and decomposition. CAB lntemational,
Wallingford.

Weast, RC. 1988. Handbook of Chemistry and Physics. CRC Press. Boca
Raton, FL.

Weber, M. G. 1990. Forest soil respiration after cutting and burning in immature
aspen ecosystems. Forest Ecology and Management 31: 1-14.

92

Weber, MG. 1985. Forest soil respiration in eastern Ontario jack pine
ecosystems. Canadian Journal of Forest Research 15: 1069-1073.

Weetman, GR and Algar, D. 1983. Low-site black spruce and jack pine nutrient
removals after full-tree and tree-length logging. Canadian Journal of
Forest Research 13: 1030-1036.

Weetman, G.F., Krause, H.H., Koller, E. and Veilleux, J.-M. 1987. Interprovincial
forest fertilization trials: 5- and 10-year results. For. Chron. 63: 184-192.

Wells, 06., Campbell, R.E. DeBano, L.F., Lewis, 08., Fredricksen, R.L.,
Franklin, 80, Froelich, RC. and Dunn, PH 1979. Effects of Fire on Soil:
A State-of-Knowledge Review. USDA Forest Service. General Technical
Report WO-7.

Werlein, J.O. 1998. Soil survey of Crawford County, Michigan, US. Department
of Agriculture, Natural Resources Conservation Service and US. Forest
Service. 274 pp and 96 sheets. »

Widden, P. & Parkinson, D. 1975. The effects of a forest fire on soil microfungi.
Soil Biology and Biochemistry 7: 125-138.

Zackrisson, O, Nilsson, M.-C., Wardle, A. 1996. Key ecological function of
charcoal from wildfire in the Boreal forest. Oikos 77: 10-19.

Zackrisson, O, Nilsson, M.-C., Dahlberg, A, Jaderlund, A. 1997. Interference

mechanisms in conifer — Ericaceae - feathermoss communities. Oikos 78:
9-20.

Zak, DR. and Grigal, D.F. 1991. Nitrogen mineralization, nitrification, and
denitrification in upland and wetland ecosystems. Oecologia 88: 189-196.

Zak, D.R., Host, GE. and Pregitzer, KS. 1989. Regional variability in nitrogen

mineralization, nitrification, and overstory biomass in northern Lower
Michigan. Canadian Journal of Forest Research 19: 1521-1526.

93