:E... . . £-
it:

:1
31 . i... a .5. 0.3....
1. L .2311}. .5

3 9.5.: ..

‘ .13....vlnfl 11“.”...10‘.
5.5.:

 

 

3 LIBRARY
21308 Michigan State
University

This is to certify that the
dissertation entitled

THE LONG-TERM EFFECTS OF LOW INTENSITY FIRES IN
A MATURE RED PINE (PINUS RESINOSA AIT.)
PLANTATION

presented by
SHAILENDRA N. ADHIKARY

has been accepted towards fulfillment
of the requirements for the

Doctoral degree in Forestry

5

/~

{I W A J .- _ \,
'L ““Q» I" C / (51-; ’1 1‘4 K&.\\__.‘\-- J”
Major Professor’s Signature

P or
\_ \(‘ivfiv , :50 \ ,LC‘O E7,

‘ a;

Date

MSU is an affirmative-action, equal-opportunity employer

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:/Proj/Acc&Pres/ClRC/DateDue.indd

 

 

THE LONG-TERM EFFECTS OF LOW INTENSITY FIRES IN A MATURE RED
PINE (PIN US RESINOSA AIT.) PLANTATION

By

Shailendra N. Adhikary

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Department of Forestry

2008

ABSTRACT

THE LONG-TERM EFFECTS OF LOW INTENSITY FIRES IN A MATURE RED
PINE (PIN US RESINOSA AIT.) PLANTATION

By

Shailendra N. Adhikary

This study focuses on the long-term effects of one- to two-decade-old prescribed
low-intensity tires at varying intervals on the vegetation dynamics of a red pine
plantation in northern Lower Michigan. One—bum, two-bums at a five-year interval, four-
bums at two-year intervals, and unburned control plots were organized in a randomized
complete block design during 1985 to 1991 and measured in the summers of 2001 and
2003. Overall, the effects of fire were apparent two decades after cessation of burning.

The fires had no effect on basal area, volume and mean annual increment (MAI)
of the overstory (>10 cm dbh); species richness, diversity and density of the woody
understory (>l.4m tall and < 10cm dbh); and percent cover of ground vegetation (< 1.4m
tall). However, when compared to unburned controls, burnings decreased overstory tree
density. The dominant overstory (>420m dbh) and small understory (0-2 cm dbh) woody
density were higher in burned plots, especially those repeatedly burned; whereas, sub-
canopy (10-32 cm dbh) and large understory (6.1 to 10 cm dbh) woody densities were
lowest in four-bum treatments. The more frequent the fires, the smaller were the
diameters of understory woody plants.

Ground vegetation (<1.4 m tall) species diversity was highest in repeatedly

burned plots. Bracken fern, Rubus spp., wild lily-of-the-valley, and red pine seedlings

had highest coverage. The consistent increase in importance values of overstory and the
understory red pine in burned plots as well as suppression of red maple and other
hardwoods indicated a shift in composition towards highly fire-adapted red pine, which
in the long run will dominate the future canopy. Red maple, beech and white ash will also
persist through new recruits.

The above data were projected 100 years into the future with and without a series
of thinnings via the Forest Vegetation Simulator (F VS). Compared to unburned controls
the key features of projected burned stands were extremely dense post-fire regeneration,
higher inequality in size-classes, higher mortality/self thinning, and a sharper decline in
density. At the end of no-management, unburned controls had grown slower and retained
the densest understory and lowest overstory density. High red pine sub-canopy density
produced by one- and two-bums indicates that low-intensity bums were sufficient to kill
competing low vegetation and stimulate red pine reproduction, which grew into the mid-
canopy layer of four-bum plots.

Burning, especially if repeated and followed by thinning, slowed down mortality,
enhanced recruitment, stimulated growth of larger trees, extended the growth culmination
age, and increased MAI and cumulative volume production. F our-bum treatments,
particularly after the second thinning, produced the highest cumulative volume and the
most uniform stand structure across tree size classes. Thus, frequent burning in mid-
rotation, especially if combined with subsequent thinning, could be the basis of long-term

sustainability, high productivity, and increases in biodiversity of red pine stands.

In the deep recollection of my ever loving father, Bodh N. Adhikary, a great source of
inspiration.

iv

ACKNOWLEDGEMENTS

I would like to acknowledge Dr. Donald Dickmann, my advisor, for his untiring
effort in helping me to accomplish this work, from its inception to this stage. He was
always available when I had a question or problem. His boundless optimism and
encouragement was inspiring.

I would like sincerely to thank Dr. Daniel Keathley who not only provided
funding for my graduate studies but also guided me throughout the program. He was a
great source of inspiration. I would also like to thank the other committee members, Dr.
Peter Murphy and Dr. Carolyn Malmstrom, who intellectually supported me and my
work. I thank the administrative staff at the Department of Forestry who was very helpful
during my study. I am thankful to my colleagues, especially Robert Bloye who was
always available and so helpful in the lab as well as in the field.

I would like to thank staff of the herbarium, Department of Plant Biology, MSU,
for the use of the herbarium and other resources available there. Also, I thank Mike
Penskar the Botany Program Leader at Michigan Natural Features Inventory for helping
me identify plant specimens. I would like to thank the Pigeon River Country State Forest
Management Unit staff, especially Joe Jerecky, Unit Manager, who provided
accommodations and other logistics and was supportive of my field work.

Last but not least, I am indebted to my spouse Maiya, son Omendra and daughter
Ambuja who accompanied me in the field and helped me collect data. Likewise, I am
thankful for the incredible support and encouragement I received from my family back

home in Nepal, especially from my mother, brothers and sisters.

TABLE OF CONTENTS

LIST OF TABLES ......................................................................... IX
LIST OF FIGURES ........................................................................ XI
Chapter 1: Literature Review ............................................................. l
1.1 Disturbance theory ....................................................................... 1
1.1.1 Definitions/disturbance regime. ........................................................ 1

1.1.2 Climatic climax or Monoclimax theory .................................... 4

1.1.3 Modern theories ................................................................ 6

1.1.3.1 Non-equilibrium .................................................... 6

1.1.3.2 Patch dynamics ..................................................... 7

1.1.3.3 Hierarchical patch dynamics ..................................... 10

1.2 Fire effects on ecosystems-general ..................................................... 12
1.2.1 Fire regimes .................................................................... 12

1.2.2 Forest fire suppression and effects of fire ................................. 15

1.2.3 Effects of fire on boreal forest ................................................ 19

1.2.4 Effects of fire on grassland and prairie ..................................... 19

1.2.5 Effects of fire on nitrogen fixing plant and nutrient cycling ............. 21
1.3 Fire effects in pine dominated ecosystems ............................................. 24
1.3.1 Ponderosa pine ................................................................. 24

1.3.2 Lodgepole pine ................................................................ 27

1.3.3 Southern pines ................................................................. 28

1.3.4 New Jersey Pine Barrens ..................................................... 30

1.3.5 Great Lake States pines ...................................................... 32

1.4 Disturbance, fire suppression and restoration management ........................ 35
1.4.1 Fire suppression and fuel reduction management ......................... 35

1.4.2 Low intensity prescribed fires ............................................... 37

1.4.3 Red pine density management 40

Chapter 2: Effects of Low-Intensity Fires on the Vegetation of Northern

Michigan Red Pine (Pinus resinosa Ait.) Plantation ............................................ 50
2.1 Introduction ............................................................................... 50
2.1.1 Global research hypothesis ................................................ 52

2.2 Materials and Method ................................................................. 53
2.2.1 Study sites .................................................................... 53

2.2.2 Data analysis ................................................................. 57

2.3 Results and Discussion ................................................................. 59
2.3.1 Overstory trees .............................................................. 59

2.3.2 Ground vegetation ........................................................... 69

2.3.2.1 Species diversity .................................................. 69

2.3.2.2 Percent cover ....................................................... 72

2.3.2.3 Tree species ........................................................ 75

2.3.2.4 Herbs and low shrubs ............................................. 78

2.3.3 Understory trees .............................................................. 83

vi

2.3.3.1 Understory density ................................................ 83

2.3.3.2 Diameter and age .................................................. 84

2.3.3.3 Density by size class ............................................... 89

2.3.3.4 Species diversity and abundance ................................ 97
2.4 Conclusion ................................................................................. 101
2.5 Recommendations ........................................................................ 105

Chapter 3: Using the Forest Vegetation Simulator (FVS) to Project the

Long-Term Effect of Prescribed Fires in a Red Pine Plantation .................. 108
3.1 Forest Simulation ......................................................................... 108
3.2 Materials and Method .................................................................... 111
3.3 Results ..................................................................................... 118
3.3.1 Tree density .................................................................... 118

3.3.1.1 Understory tree density ........................................... 119

3.3.1.2 Overstory tree density ............................................ 120

3.3.2 Stand Structure ................................................................ 123

3.3.2.1 Bimodal tree distribution ......................................... 123

3.3.2.2 Tri-modal tree distribution ........................................ 124

3.3.3 Species density ................................................................... 125

3.3.3.1 Understory woody and overstory species ....................... 129

3.3.4 Basal Area ..................................................................... 130

3.3.4.1 Understory basal Area ............................................. 130

3.3.4.2 Overstory basal Area .............................................. 133

3.3.5 Mean annual increment ...................................................... 139

3.3.6 Saw log volume ................................................................ 141

3.4 Discussion .................................................................................. 145
3.5 Summary .................................................................................... 155
3.6 Conclusion .................................................................................. 157
4. APPENDICES ............................................................................. 158
Appendix 2.1: Species Codes ............................................................... 158

Appendix 2.2: Density, Basal Area, Volume, MAI and Importance Value (IV) of
overstory trees under different burns ............................................. 159

Appendix 2.3: Basal Area, Volume and MAI of overstory trees under different
diameter classes ...................................................................... 160

Appendix 3.] : Average & relative live understory (0-1 1 cm dcl) tree density

under 100 years of no-management and thinning simulations in mature

red pine stands ........................................................................ 161
Appendix 3.2: Average and relative sub-canopy (>11-32 cm dcl), mid-canopy

(32-42 cm dcl) and dominant overstory (>42 cm dcl) tree density under

vii

100 years in mature red pine .......................................................... 162

Appendix 3.3a: Average tree density and their relative values by diameter
classes under different management simulations for 100 years ......... 163-164

Appendix 3.3b: Average basal area and their relative values by diameter
classes under different management simulations for 100 years ......... 165-166

Appendix 3.3c: Average sawlog volume and their relative values by diameter
classes under different management simulations for 100 years ......... 167-168

Appendix 3.4: Average relative sawlog volume of major species for 100 years
of no-management and thinning simulations in differently burned and

unburned mature red pine stands ................................................... 169

Appendix 3.5: Comparison of Forest Vegetation Simulation (F VS) projections with
that given in Benzie (1977) for standing merchantable cubic volume (cu m/ha)
and the predicted values in differently burned mature red pine stands at site
index 19.8 m in the beginning and at the end of no-management and
thinning ................................................................................ 170

5. REFERENCES ............................................................................. 171

viii

LIST OF TABLES

Table 1.1: A summary of fire return intervals in different locations/vegetation types ...14

Table 2.1: Stand characteristics of overstory trees (>10 cms in dbh) in experimental
red pine plantation under four burn treatments ........................................ 59

Table 2.2: Mean characteristics of overstory trees (>10 cm in dbh) in experimental
red pine plantation with four burn treatments .......................................... 60

Table 2.3 Effects of prescribed burning on dominant overstory, mid-canopy and sub-
canopy trees in experimental red pine plantation, summer 2001 .................... 64

Table 2.4: Species diversity (per sample plot) of ground vegetation (<14 m height)
in experimental red pine plantation 2003. .............................................. 71

Table 2.5: Average cover and relative cover of trees in descending order in the
understory ground vegetation (< 1.4 m) in a mature red pine plantation ........... 73

Table 2.6: Average and relative cover of herbs and low shrubs in descending order in
the understory ground vegetation (< 1.4 m) in a mature red pine plantation. ......74

Table 2.7:Analysis of number of three herb or shrub species per plot (sq m/plot). . ........82

Table 2.8:Woody understory stem density (trees per hectare) and relative density
in descending order under different size classes ....................................... 85

Table 2.9: Mean diameter at breast height (dbh) of understory woody vegetation in
experimental red pine plantation, 2003 ................................................. 88

Table 2.10: Age and diameter of understory woody seedlings in experimental red
pine plantation, 2004 ...................................................................... 90

Table 2.11: Woody understory species density (trees per hectare) of different
diameter size classes under different burning treatments, 2003 ........................ 93

Table 2.12: Indices of species diversity of understory woody vegetation in
experimental red pine plantation, 2003 .................................................... 99

ix

Table 2.13: Importance Value (IV) ofthe understory woody species. . . . . . . ....1.00

Table 3.1: Average and relative live understory and overstory tree density under
100 years of simulation in mature red pine stands .................................... 121

Table 3.2: Average and relative tree density of all live species under different
managements for 100 years of simulation in mature red pine stands ............ 131

Table 3.3: Average and relative live understory and overstory tree basal area under
100 years of simulation in mature red pine stands ................................... 136

Table 3.4: Average and relative sub-canopy, mid-canopy and dominant overstory
basal area under 100 years of simulation in mature red pine stands ............... 137

Table 3.5: Average and relative basal area of red pine and other species combined
under different management simulations run for 100 years ....................... 138

Table 3.6: Importance Value (IV) of all species under no-management and thinning
simulations run for 100 years in mature red pine. ..................................... 139

Table 3.7: Average and relative sawlog volume of sub-canopy, mid-canopy and
dominant overstory trees under 100 years of no-management and thinning
simulations in mature red pine stands ................................................. 141

LIST OF FIGURES

Figure 2.1: Location of Pigeon Country State Forest in the Lower Peninsula
of Michigan and Red Pine Study Site ......................................... 54

Figure 2.2: Average overstory tree density of different diameter size classes
in a mature red pine plantation ................................................. 68

Figure 2.3: Species area curves by blocks for herbaceous ground vegetation
in a 70-year-old red pine plantation under burn treatments ..................... 70

Figure 2.4a to d: Pronounced variation in the understory in unburned and
one-, two- and four-bumed treatments in a 70-year-old red pine
plantation, summer 2001 ....................................................... 86

Figure 2.5a: Density of understory woody vegetation of 0-4 cm size
classes under different burns .................................................. 91

Figure 2.5b: Density of understory woody vegetation in two size classes
under different burns ............................................................ 91

Figure 2.6a: A plot just after its second burning (two-year interval), with a
unburned plot in the background ............................................. 95

Figure 2.6b: A plot after the first fire in 1985 ....................................... 95

Figure 2.7: Species area curve (SPA) for understory woody vegetation in a
mature under red pine plantation under burning treatments, in
blocks A, B and C ............................................................... 98

Figure 3.1: Understory woody (0-11 cm dcl) tree density in differently
burned plots under no-management and thinning in mature red
pine stands ....................................................................... 122

Figure 3.2: Tree distribution by diameter size classes in differently burned
red pine stands under no-management and thinning simulations for
100 years ........................................................................ 126

Figure 3.3 (a to (1): Tree distribution by diameter size classes in year 2104
in differently burned red pine stands under no-management and
thinning at the end of simulations ........................................ 127-128

Figure 3.4: Woody understory (0-11 cm dcl) red pine and other species

combined density in different burn treatments and management
simulations for 100 years in mature red pine stands ........................ 132

xi

Figure 3.5: Average sub-canopy (>11-32 cm dcl) red pine and other species
combined density in different burn treatments and management
simulations for 100 years in mature red pine stands ........................ 134

Figure 3.6: Average basal area in different burns under no-management and
thinnings with residual basal area set at 22.9, 27.5 and 34.5 sq m/ha
at 30 years cycle for 100 years of simulation in mature red pine
stands .............................................................................. 135

Figure 3.7: Mean Annual Increment (MAI) in different burn treatments
under no-management and thinning for 100 years of simulation
in mature red pine stands ....................................................... 140

Figure 3.8: Average cumulative sawlog in different treatments under
no-management and thinning for 100 years of years of simulation
in mature red pine stands ....................................................... 143

Figure 3.9: Percent change of cumulative sawlog production after thinnings
over no-management in different burn treatments for 100 years
of simulations in mature red pine stands ...................................... 144

Figure 3.10: Percent change of cumulative sawlog production under different

burn treatments and management over control run for 100 years
of simulations in mature red pine stands ...................................... 144

xii

Chapter 1: Literature Review
1.1 Disturbance theory
1.1.1 Definitions/Disturbance regime

Traditionally, ecologists have considered disturbance to be a periodic event or a
force (such as hurricanes, fire, etc.) that originates from outside the system (exogenous or
allogenic) and causes sudden changes in species composition and the structure of
communities or ecosystems (Borman and Likens 1979, White 1979, White 1985, Rykiel
Jr 1985). Pickett and White (1985) define a disturbance as a relatively discrete event in
time that disrupts community or population structure and ecosystem processes, changing
resources, substrate availability or the physical environment. Such natural disturbances
are considered normal phenomena of community maintenance and repair (Marks and
Borrnann 1972). In addition to physical or abiotic disturbances (fire, wind, flooding etc.),
biological disturbances such as intense grazing, predation, insect-pathogen out breaks,
and the like can affect a system. Anthropogenic disturbances (construction, land
conversions etc.) often break up habitats into smaller fragments that lose connectivity and
diminish heterozygosity via loss of species (Baker 1989) and lower global biological
diversity (Wu and Levin 1994). Similarly, indirect effects of anthropogenic activities—
such as acid rain—affect ecosystems adversely (Likens and Lambert 1998). Disturbances
are called autogenic or endogenous when the changes originate from within the system;
for example, the death of bamboo forests after flowering, the decline and death of mature

aspen stands (White 1979, Gilliam and Turrill 1993).

Natural disturbances range in magnitude, occurrence (frequency), severity
(intensity), behavior (predictability) and duration, and their effects are relative to the size,
life-history characteristics and life span of organisms. Together, these constitute
disturbance regimes, which in turn are also affected by communities (Connell 1978,
White 1979, Rykiel Jr 1985, White and Pickett 1985, Reice 1994, Menges and Hawkes
1998). Fires are often distinct from other physical disturbances such as hurricanes,
volcano, flood and the like, which can happen in the absence of living beings or their
remains. However, barring man-made flammable materials, fire needs living beings or
their remains for its propagation; thus, fires are under partial or complete biotic control
(Pyne 2004). Disturbances open the canopy, alter the availability of resources for
recolonization and growth, and reduce the dominance of few species (Canham and Marks
1985). They create a pattern of spatio-temporal heterogeneity (Levin and Paine 1974).
Small-scale disturbances are more frequent than large-scale disturbances (Sousa 1984;
Bormnann and Likens 1979, Foster 1988, Merrens and Peart 1992, Heinselman 1973,
White 1979, Arno 1980, Rome 1982, Christensen 1985) and the magnitude of
disturbances vary according to site or location. For example, in southern Canada the
islands in Lake Duparquet are characterized by more frequent smaller-scale fires, than

those on adjacent mainland (Gauthier et a1. 1996).

Forest communities are exposed to one or more different types of disturbances
that sometimes act in synergism and may be correlated (White 1979). Such disturbances
set the stage for additional disturbances to prevail. The susceptibility of a tree to wind

storms depends on the tree attributes, site, surrounding forest structure, and the prevailing

storm systems (Canham and Loucks 1984). Once trees become exposed to some
exogenous disturbances, such as wind, they become more susceptible to other types of
disturbances, such as fire (Bormann and Likens 1979, Schowalter 1985, Foster 1988). For
example, in the Pisgaha old growth forest (New England), a series of different
disturbances—lightning, windstorms, pathogens, and fire—occurred during different
periods and had synergistic effects (Foster 1988). Chestnut blight entered the Pisgaha old
grth forest in about 1913 and by 1915 the mortality was already extensive. Later, small
windstorms hit the region a number of times, followed by an Armillaria mellea attack,
and then small fires during 1924 to 1930. Later the forest was logged extensively, and in

1938 it was hit by a hurricane, which devastated the whole forest.

The southern pine beetle’s attack on southeastern pines, which reestablish only
after fire, is another good example of synergistic effect of natural disturbances
(Schowalter 1985, Aber and Melillo 1991). The beetles slowly build up their population
and are without much effect if their number is less than 100,000. However, if the
population increases beyond 100,000 (threshold level) by June, then the attack is massive
with the beetles forming galleries, excavating and girdling live trees (Aber and Melillo
1991). Furthermore, through their root feeding, the beetles infest the pines with a blue
stain fungus that blocks their xylem, which causes the trees to further weaken and
predisposes them further to beetle attack. The overall result is the killing of pine trees,
which dry quickly during summer. Under these conditions, all stems including
hardwoods, are susceptible to lightning-caused fire, which consumes almost all the fuel

and makes the site suitable for southern pine regeneration again. Similarly, in the

Northern Rockies at Yellowstone there is a high mortality due to the mountain pine
beetles and such mortality favors stand replacing fires (Wellner 1970 in Arno 1980,

Loope and Gruell 1973, Arno 1980).

Benkman and Siepielski (2004) detail a good example of the disturbance effect of
animals. They found that squirrels in the Rocky Mountains have the potential to alter the
early stages of seedling establishment and succession. Whereas the forests without
squirrels had consistently a 100% frequency of serotiny, those with squirrels had only
about 50% frequency. Serotiny is the ability of conifers to produce seeds in closed cones
that release seeds after getting exposed to fire. Serotiny is also the life-history traits of

plants that experience stand replacing fires (Benkman and Siepielski 2004).

1.1.2 Climatic climax or Monoclimax theory

‘Climatic climax’ or ‘Clemential Monoclimax theory’ predicts that if there are no
human disturbances, ecological communities will eventually reach a climatic climax; i.e.,
species interactions with a stable environment lead to a relatively stable state of species
abundance and composition (Clements 1936 and 193 8). This idea supports the
equilibrium, balance of nature, steady state or homeostatic views of community
dynamics. Further, Clements (1916) viewed communities themselves as organisms and
the organisms within them as their parts. This led to the “Gaian hypothesis” in which the
Earth is seen as a ‘super organism.’ Later, ‘polyclimax theory’ (Tansley 1939 in Selleck
1960) and ‘Climax pattern hypothesis’ (Whittaker 1953 and 1956) evolved to point out

limitations of climatic climax. Others (Whittaker 1953, Selleck 1960) disagreed with the

‘Monoclimax concept,’ which was vague and lacked precision in terms of the geological

time scale that affects ecosystems.

Since disturbances are random and ecosystems are dynamic (Wu and Loucks
1995), the climatic climax must be referenced to some environmental conditions or
context, otherwise it loses its meaning (White 1979). Both the disturbances and
ecosystems are then described in terms of heterogeneity along soil, aspect, temperature,
moisture, light availability and other gradients that influence the disturbances and their
effects on species occurrences and community attributes (Whittaker 1953 and 1956,
Marks 1974, Bormnann and Likens 1979, White 1979, Sousa 1984, Foster 1988, Reice
1994). Grime (1977 and 1979) categorized the environment into productive, stressed and
disturbed sites, which harbor competitor (C), stress-tolerant (S) and ruderal (R) species,
respectively. However, Huston and Smith (1987) differ with Grime’s three discrete
strategies and envision a continuum across those three strategies and their combinations.
Additionally, the perturbations (effects) (Connell and Sousa 1983) and successions also
form continua across different communities (Selleck 1960, White 1979, Hollings 1973,
Wu and Loucks 1995). These gradients or continua across different systems/levels and
their interactions are the basis for the dynamic nature of ecosystems. This view is in line
with Sir Charles Elton’s claim in1930 that “the balance of nature does not exist and

perhaps never has existed.”

1.1.3 Modern theories
1.1.3.1 Non-equilibrium

The non-equilibrium model assumes frequent unpredictable disturbances (Connell
1978, Reice 1994) that cause random death of organisms/communities that never reach
equilibrium. Such disturbances create patches or openings (Pickett 1980). Thus,
ecosystems are always in a state of recovery from past disturbances (Merrens and Peart
1992, Reice 1994). The Intermediate Disturbance Hypothesis (IDH) as well as modern
theories of disturbances such as patch dynamics and hierarchical patch dynamics provide
important insight into mechanisms responsible for species migration and colonization,
species abundance, niche differentiation, species diversity and their maintenance (Watt
1947, MacArthur and Wilson 1967, Connell 1978, Grubb 1977, Sousa 1979, Huston

1979, Pickett and White 1985, Reice 1994) and will be discussed further.

IDH postulates that if the disturbance regime is intermediate (i.e., neither frequent
nor rare) both early and late successional species persist because of reduced competition
(Connell 1978). Superior competitors remain dominant if disturbances are not frequent,
whereas inferior competitors (good dispersers) persist and dominate the system if
disturbances are frequent enough to prevent competitive exclusion (Connell 1978, Huston
1979, Connell and Sousa 1983, Roxburgh et a1. 2004). However, species diversity will
decline if disturbances are too frequent. For any system, the problem lies in finding out
how intermediate is intermediate, which may vary in different systems as well. Therefore,
the parameters of responses in any system should be defined precisely (Pickett and White

1985). IDH does not explain the coexistence of species that are at the top of the trophic

level or mobile invertebrates (Huston 1979, Wooton 1998; Roxburgh et a1. 2004) and
may not apply to all terrestrial systems. For example, IDH does not explain increased
richness and diversity in the arid (low resource) Chamise shrub habitat where there is low

competition for resources (Christensen 1985).

1.1.3.2 Patch dynamics

Disturbances create patches of different shapes, sizes, ages and stages of
succession that vary in structure and composition of species in space and time.
Environmental gradients within and outside the patches create the heterogeneity that is
important in maintaining the mosaic structure of ecosystems (White and Pickett 1985,
Forman and Godron 1981, Collins et a1. 1985, Reice 1994). Thus, landscapes are mosaics
of patches (F orman and Godron 1981, Dunning et a1. 1992) and their widespread
occurrences and dynamism is called patch dynamics (Watt 1947). For example, about one
to two percent of old grth forests are under new patches or gaps at any time (Runkle

1981, Runkle 1985).

Gap size in forest ecosystems depends on the number and type of trees that fall
and how they fall (Brokaw 1985b). Patch size affects temperature fluctuations, nutrient
and water dynamics, timing and type of propagules reaching the patch, as well as
recolonization, overall species diversity, and productivity of the patch (F orman and
Godron 1981, Sousa 1984, Runkle 1985, Phillips and Shure 1990, Roberts and Gilliam
1995). Multiple tree fall forms large-sized gaps (Sousa 1984), which are brighter, hotter,

and drier than smaller size gaps formed from the fall of a single or few trees (Denslow

1980). Large patches favor large numbers of early successional species (ruderals or R),
whereas, small patches favor late successional species (competitors or C) (Marks 1974,
Grime 1977, Bormann and Likens 1979, Denslow 1980, Gilliam et al. 1995). The
intermediate size patches can harbor both types of species (Connell 1978). Mesic patches
exhibit highest species diversity as compared to xeric and hydric patches (Roberts and
Gilliam, 1995). Patch shapes and their edge-area, numbers, configuration and the
networking of patches, and connectivity or corridors between patches are important for
species migration and diversity (Forrnann and Gordon 1981, Sousa 1984, Dunning et a1.

1992)

Reproduction explains the ‘non-equilibrium theory’ through patch dynamism, i.e.,
patch creation and filling (Pickett 1980). Gap-phase reproduction or gap recovery occurs
through two types of models of vegetation recruitment (Egler 1954): initial floristic
(reaching the site first) or relay floristic (making the site suitable for others). The
recovery and succession of patches or success of species are determined by the following:
good seed years; seed migration (Marks 1974); species life-history traits such as, size,
tolerance to light, dispersability and growth (Grime 1977, Huston and Smith 1987);
environmental and resource gradient across patches (Watt 1947); time setting of patches
for recruitment and patch closure (Halpem and Spies 1995); and patch availability (Horn
and Mac Arthur 1972). Gaps or patches recruit through regrowth (vegetative), migration
from adjacent stock and recruitment from outside the proximate system (Reice 1994).
Gap-filling can be due to germination of seeds (buried and transported), growth of

preexisting seedlings and saplings, growth and expansion of clones and sprouts,

principally of pioneer species (Spurr 1956, Marks 1974, Likens et a1. 1978) or the growth
of epicormic branches (Runkle 1985) and occurs until resources become limiting
(Nicholson and Monk 1975). In the tropics, ruderals (R) grow in disturbed sites,
maximize seed production over vegetative growth and are density independent (I-
selection) (Grime 1977, Brokaw 1985a, Brokaw 1985b). On the other hand, competitors
(C) grow in productive sites, maximize vegetative growth over seed production, are
density dependent (K-selection), and are later replaced by stress tolerants (S). Further,
species are either competitor or ruderal but not both (Grime 1977, Roberts and Gilliam

1995).

Regeneration, the path of succession, future structure and composition of forests
depend on the availability of species and the temporal pattern of disturbances (Abugov
1982, Glitzenstein et a1. 1986, Peterson and Pickett 1995). Early arrivals are important
especially in the subsequent establishment of light-intolerant woody species in the gaps
(Bormann and Likens 1979, Canham and Marks 1985). In the Boundary Waters Canoe
Area Wilderness, canopy Openings are filled with one or more species. If more than one
species invade a gap, patches of each dominant species are usually formed in the gaps

(Frelich and Rich 1995).

If disturbances are too frequent, succession is shortened or takes a different course
(Cornell and Slatyer 1977, Frelich and Reich 1995), whereas if the disturbance is
catastrophic, then succession starts again (Reice 1994). However, all disturbances do not

start succession; they may rather alter successional changes (Glitzenstein et a1. 1986,

Abrams and Scott 1989). Sometimes even disturbances of the same magnitude and type

initiate different or multiple pathways of succession (Abrams et a1. 1985).

1.1.3.3 Hierarchical patch dynamics

In Wu and Loucks’s (1995) view, ‘disturbance theory’ is best explained by a
hierarchy of patch dynamics, non-equilibrium and stochastic patterns and processes.
Patch dynamics and hierarchical (‘hierarchical patch paradigm’) theory together consider
patches as structural and functional units of ecosystems working at multiple scales. After
disturbances, patch formation and its recovery across a landscape and the interactions of
patches with each other at multiple scales (spatial, temporal and organizational) create,
maintain and destroy patterns at various levels and form a hierarchical mosaics of patches
or patchiness. Changes in this mosaic occur due to interactions among vegetative,
disturbance and geomorphological processes (Watt 1947, Borman and Likens 1979,
Brokaw 1985b, Holling 1992, Wu and Levin 1994, Wu and Loucks 1995). Thus, there
are patches within patches within patches, creating a hierarchical system that absorbs
small disturbances when looked at through a larger level or resolution of a landscape. For
example, the disturbance at one level (death of a tree) may not be a disturbance at other
levels (stands, landscape or ecosystem); i.e., the effect is absorbed as we go from a lower

to a higher level (Rykiel Jr 1985).

Northern New Hampshire forests are characterized by such hierarchical mosaics
of different-aged patches of communities, which Bormann and Likens (1979) called a

‘shifting-mosaic-steady-state.’ Their system basically is a shifting of standing biomass of

10

even-aged stands produced by clearcutting. According to Wu and Loucks (1995), shifting
of such hierarchical mosaics of patches at lower level transforms to a shifting-mosaic-
steady-state at a higher level and constitutes a kind of equilibrium that has a range of
some lower and higher bounds or limits. After disturbances in old forests, trees fall and as
the canopy closes slowly, the standing biomass changes slightly over the mean; however,
the biomass on any small plot in the stands/watershed may vary at any time. For instance,
old growth forests of Douglas-fir show shifting mosaics of resources and environments
that support a diversity of species (Halpem and Spies 1995). However, Baker (1989) did
not find any such temporally stable patch-mosaic at any scale while studying a ca.
404,000 ha patch mosaic created by fire in the Boundry Waters Canoe Area in
Minnesota. He suggested that this situation could be due to heterogeneity between fire
regimes and environment or mismatching of the scale of patches and the environment.
Frelich and Lorimer (1991) found that in upper Michigan hemlock hard-wood forests the

age distribution in two fairly large areas (14500 and 6073 ha) were nearly in equilibrium.

In Wu and Loucks’s (1995) view, in a homeorhetic state, the stochastic non-
equilibrium patch processes at a smaller level are dynamic and translate to a quasi-
equilibrium state at higher levels; i.e., a shifting-mosaic steady state (Bormann and
Likens 1979). Such a state is congruent with the ‘hierarchical patch paradigm’ or
‘hierarchy theory.’ After perturbation, such a system returns back to the pre-perturbation
trajectory rate of change but not to the pre-perturbation steady state. For example, in
southern Wisconsin forests, fires with intervals of 50 to 200 years were found to

destabilize in the short-term but stabilize (steady state) on a long-term scale; i.e., a

ll

periodic wave of high and low diversity is maintained by periodic perturbations of 50-
200 years (Loucks 1970). Similarly, in a gopher mound study by Wu and Levin (1994),
local populations of certain species at the patch (or local population) level became extinct
frequently, but metapopulations showed little fluctuation. Thus, forest composition may
be relatively stable or in a dynamic equilibrium with local changes canceling out over the
whole under the influence of fire (Heinselman and Wright 1973). Further, several patches
at equilibria or multiple equilibrium patches and their shifting during disturbances have

also been suggested (Holling 1973 and 1992, Levin 1976).

1.2. Fire effects on ecosystems—general
1.2.1 Fire regimes

Fire plays an important role in establishing, shaping and modifying the structure
and composition and thus the successional pathways of north-temperate and boreal forest,
grassland and prairie ecosystems (Larsen 1929, Little and Moore 1949, Cooper 1960,
Bormann and Likens 1979, Vogl 1969, Van Wagner 1970,_Heinselman 1973,
Heinselman and Wright 1973, Rowe and Scotter 1973, Kilgore 1973, Wright 1976, Arno
1980, Pyne 1982, Rome 1982, Whitney 1986 and 1987, Hulbert 1988, Aber and Melilo
1991, Abrams 1992, Agee 1993, Arno 1993, Covington and Moore 1994b, Howe 1995,
F eeney et a1. 1998, Arthur et a1. 1998, Veblen et a1. 2000). Impacts of fire are better
understood in terms of ‘fire regime,’ which includes ignition source, its seasonality and
predictability; mosaic nature of landscapes; fire intensity and its spread; wind speed;
relative humidity; local weather conditions; climate; the past history of fire on vegetation;
and the impact of vegetation on the fire regime itself (Sousa 1984, White and Pickett

1985, Gill and Groves 1981 in Veblen 1985, Agee 1998). Depending on its intensity and

12

periodicity, fire alters availability of light, water, nutrients, microclimate and the
availability of propagules and their regeneration (Harmon 1984, Sousa 1984).
Flammability of vegetation or communities also affects the fire regime (Mutch 1970).
Fire affects the quality and quantity of fire] available (Reice 1994), which controls fire
frequency to some extent (Rome 1982). All of these factors vary over space and time
and their interactions create heterogeneity (patchiness) in ecosystems, which in turn
affects local fire regimes (Sousa 1984, Veblen 1985). Topography and location also are
important; e.g., mountain tops are more vulnerable to severe and frequent fires because of
low water and windy conditions (Reice 1994, Taylor and Solem 2001) than the bottom or
valley (Rome 1982). Frequent fires favor shrub and herb establishment rather than trees
(Runkle 1985). Fires are thus used in the managements of grasses and shrubs (Raison
1979). Fire regimes as they incorporate most aspects of ecosystem patterns and processes
have implications in maintaining forest health and managing forests (Cleland et a1. 2004).
Fire regimes of some important vegetation types and their locations are summarized in

Table 1.1.

13

.mvookuaa Panto: .332 682

 

mwm: comcflmtno

A82 mEaE< Ev 39 “5:82.

mom: .130 museum

32 SEE:

N2; SE20

$2 LoEoom ES :cmEom ”3co— ucw. Duo— 282 use 255

Sow Sue—om can 5:8...
Bow EBow can 873%
Sam EBom can high

Nam: .oEEom
one oE<

92 $5326:
ohm: 23.5 25 ccmctom

£2 ages
32 ._u we E88“. 553 mace—5 can agar—om ”ES SEES
32 $523

$2 @533

go— :Ommtm “Ea cohowcom

32 as: e5

Beau—mam— nooofi :Ommtm van .5533 63— 5:me =m>
£2 8:55

33 565ch v5 oMUoEUH

Avoomv .3 so 2220

9.89 a e 2206

$92
2 o:
2 an

2
2 am
822

on 9 on
Tn
n: can on

oov 8 8m

m2 8 mm

0mm 8 ow

ooom 8 03
82 9 a: 53

com
com 2 oov

8m 8 o:

$on mica—m2 923$ com 8 02
7.05 35:85 303

on :05 no on 3 m

Gee 5528 cm =o>o co co 8 mm
32 .o o

3323 80m 9 can 2: as on
Sex 8 8V soc Ems:

 

 

 

 

 

832$on
.m

 

 

 

3:8an .352 2E7. 3:550 23.
0595 3.9% was

“SE—Sow .38.“: via «8.56:8
ink :28an 58552 308m :35
385:5 05 5 38qu 2E GEES..—
mcotam 25 DE: 302

388m can BE? E03376 R: can E groin
20332 .05; Benomvo— 6:3 beta—.5.“ BE?
3839 cm 333-com

203 358033 5.6rosz 95% use vacuum

<0 £09830 Eur—Sow.
>3 Jinn 38:52 ozo~m30=o>£wocom oEa—nbsm ,

3:32 Ea S .0: v Essa: 38m E252

22 50an .82 ooch 285$ ransom
8580 £25.55 >52
E2 a 25 2.5 323-83%
3:2 3532 E 388% 2:285
3:2 EOE coda:
32¢. cam—m: E 388"—
QEBaE an vex

93:65 2cm com
uni x02.
82 Loam :2 .5058: 5 3.0.28 :8 nEom
52.. Swans Ea .5322 5
DE .80 wEwwo—EQ warn—u :2 E0552

 

Each 5 l EFCBE E32 8E

 

5:80 0385\5383

 

 

8:80 > use £5282 Eocowtw E can: mEZBE E32 05 mo buEEzm < 2 03a...

14

1.2.2 Forest fire suppression and effects of fire

In the US, human interventions have changed the fire regimes throughout major
ecosystems as a result of activities such as logging, clearcutting, burning slash, grazing,
land conversions and development activities including fire suppression that occurred
during the nineteenth and twentieth centuries (Heyward 1.939, Cooper 1960, White 1985,
Arno 1993, Fule et al. 1997, Abrams 1992, Covington and Moore 1994b, Heinselman
1973, Dickmann and Leefers 2003, Cleland et al. 2004). Heavy loads of slash left in
unburned stands after extensive logging operations during the nineteenth and early
twentieth centuries were followed by unprecedented catastrophic fires throughout this
country, including Great Lakes States (Dickmann and Leefers 2003). This situation led to
the implementation of a fire suppression policy that mandated fire exclusion. This policy
began during early 19005 and was strictly implemented within a couple of decades
throughout the country (Kilbum 1960, Whitney 1987, Arno 1993). The eventual result
was an accumulation of huge loads of forest fuel and many large-scale catastrophic wild
fires compared to the pre-logging and pre-settlement era (Larsen 1929, Vogl 1969, Van
Wagner 1970, Heinselman 1973, Wright 1976, Rome 1982, Whitney 1986 and 1987,
Abrams 1992, Arno 1993, Feeney et a1. 1998, Arthur et al. 1998, Veblen et al. 2000).
These modern fire regimes have changed the structure and composition of fire-adapted
forest and grassland ecosystems and reduced overall species diversity (Heyward 193 9,
Little and Moore 1945 and 1949, Buell and Cantlon 1953, Kilburn 1960, Mutch 1970,

White 1979, Romme 1982, Whitney 1986 and 1987, Taylor and Solem 2001).

15

Before fire suppression was common in Michigan, fires frequently occurred after
logging, consuming many of the remaining pine seed trees and seedlings and favoring
oak, maple, birch and aspen establishment, which, changed the composition and structure
of forests, as evident today (Heinselman 1973, Whitney 1987). Under current conditions,
however, oaks have not regenerated under oak shade (Hartman et a1. 2005) but have
yielded to more shade tolerant red maple and white pine that need only small gaps
(Arthur et al. 1998; Abrams 1992). Oaks are not adapted to low light and decline under
shade of other species (Abrams 1992; Lorimer et a1. 1994; Arthur et al., 1998). Due to
lack of fire and other disturbances needed for their perpetuation, the area occupied by

aspen-birch forest has also declined markedly since the 19303 (Cleland et al. 2001).

When burned, thick bark and sprouting ability of oaks favor them as compared to
non-oak species (Abrams 1992, Lorimer et al. 1994, Arthur et al. 1998). Oaks are
resistant to fire even in sapling stages. In the Cumberland Plateau northern hardwood
ecosystem, burning killed 67 to 100% of the black cherry, paper birch and large-toothed
aspen whereas none of the oak saplings were killed (Reich et al. 1990). Likewise, White
(1983) found that large oak trees (>=25cm) in Minnesota were affected very little by
annual burning of oak-savanna forest for 13 years and that the density and basal area of
overstory oak trees were significantly higher in burned than unburned areas. In the Great
Smoky Mountain National Park, vegetation changed in dominance from fast-growing,
thick-barked early successional trees to slow-growing thin-barked late successional trees

after fire suppression since 1940 (Harmon 1984).

16

In subalpine forests of Yellowstone National Park, as a result of fire suppression,
older trees at late successional stages dominated the landscape rather than younger trees
at post-fire early successional stages, and landscape diversity was less than that of natural
fire regime (Romme 1982). Arno (1993) suspected that “the wild fire problems will
worsen. . .in the Western North America.” Similarly, fire suppression has changed
vegetation from fire-dependent lodgepole pines to vegetation similar to those found in
lower montane forests (i.e., fir) in the southern Cascades (Caribou Wilderness),
California (Taylor and Solem 2001) and also in Alberta’s Rocky Mountains (Day 1972).
Douglas-fir thickets have increased in the sub-alpine forests of Yellowstone National
Park in the recent past because of fire suppression and grazing, and these conditions
render lodge pole pine forest prone to intense fire. In Colorado Front Ranges, both the fir

and spruce were replaced by lodgepole pine where fire was frequent (Moir 1969).

In the Southwest, fire suppression for decades has changed the open park-like
ponderosa pine forests with clusters of old pines and lush green grassy areas to dense
thickets of saplings and shrubs, which suppress grass growth and increase vulnerability to
stand-replacing wildfires and insect outbreaks (Dickman 1978, Cooper 1960, Moir 1966,
Feeney et al. 1998, Allen et al. 2002). Yeaton (1983) have suggested that ponderosa pine
and sugar pine co-exist because of disturbance that opens up space and microsites for

them to establish in Sierra Nevada, California.

The absence of fire that used to occur every two to three years before fire

exclusions caused long-leaf pine dominated forests (sandhill vegetation) to convert to

17

oak-hickory forests in the southeastern coastal plain from Florida to North Carolina
(Quarterman and Keever 1962; Abrams 1992). Also, grass-sedge bogs are changed to
forest areas when protected from fires (Garren 1943). In the absence of fire, long leaf
pine forests are taken over by loblolly pines in the drier regions and by slash pines along
the coastal areas (Walker 1980). These southern pines are shade intolerant and they
germinate and establish in exposed mineral soil and in the absence of fire are encroached
by hardwood species, mainly oaks (Walker 1980). In the Cumberland Plateau and other
forests, oaks declined, whereas other fire sensitive species became abundant due to fire
suppression during the latter part of the twentieth century (Arthur et al. 1998). However,
in the New Jersey Pine Barrens, fire suppression increased fire-sensitive hardwood
species, including oaks which replaced swamp cedar. The area burned each year was

reduced to about 10% of the total area (Forman and Boemer 1981).

Oaks invaded and dominated the tall grass prairies after fire exclusion as well as
during European settlement, during which time construction, land conversion and
development activities acted as barriers for fire to expand (Abrams 1992). In the
Midwest, the total area of oak savanna, which used to be 11 million ha, has now been
reduced by >98% (Abella et al. 2004). In an oak-savanna study in Wisconsin, openings
were found to be filled with woody trees and shrubs in 10 years and completely
converted to a dense oak forest in 30 years of fire exclusion (Curtis 1959 in White 1983).
In Minnesota oak-savanna, fire suppression was found to double the amount of carbon
storage (220 mg/ha) in woody biomass and the forest floor as compared to carbon amount

of presettlement; however, there was no effect in soil (Tilman et al. 2000).

18

1.2.3 Effects of fire on Boreal forest

The boreal forest is well adapted to frequent fires (Rowe 1961, Heinselman 1973,
Rowe and Scotter 1973). Frequent fires account for the occurrence of young communities
that determine future structure and composition of boreal forests (Zoladeski and Maycock
1990). In Taiga (Interior Alaska), intense fire is needed to release the system by getting
rid of feather and Sphagnum mosses covering the forest floor and starting succession. Fire
opens up space, consumes the thick litter layer, exposes mineral soil completely and
melts the permafrost and thus basically determines the type of forest (Aber and Melillo
1991). Black spruce forest inhabits the colder northern and less steep slopes with poor
drainage that develop thicker permafrost layer unlike the southern slopes which are
inhabited by white spruce. Poplar forests may develop in the less steep areas that lack
permafrost. Post-fire recruitment and species retained depends on time lapsed since fire
(J ohnstone et al. 2004). In Yukon, British Columbia, the post-fire spruce population
remained constant after the first decade, whereas both aspen and lodgepole pine declined
due to density-dependent mortality (self-thinning process). The highest post-fire
recruitment occurred in the first five years. Also, forests undisturbed for a long time act
as refugia from where plants and animals can spread and recolonize the burned areas

(Rowe and Scotter 1973).

1.2.4 Effects of fire on grassland and prairie
Fire stimulates grass coverage and flowering guilds and suppresses woody
vegetation and forbs. If the prairies are not burned, cut or grazed, they convert to forest

slowly in the long run (Niering et al. 1970, Bragg and Hulbert 1976, Hulbert 1988, Howe

19

1995). Tall bluestem prairie grasses grow earlier and faster in burned than unburned areas
(Hulbert 1969 and 1988, Niering et al. 1970, D’ Antonio and Vitousek 1992). For
example, after 17 years of burning, standing biomass of Andropogon spp. increased to
364 g/m2 as compared to 252 g/m2 in unburned prairie (N iering and Dreyer 1989). All
fire regimes (not just a single burn) are critical in determining the community response in
long-grass prairies (Hobbs and Huenekke 1992). High surface/volume ratio, dead
standing biomass that can easily catch fire and spread, and highly photosynthetic new
tissues with luxuriant growth are all fire adaptive characteristicsof grasses (D’Antonio
and Vitousek 1992). Following high-intensity burns, closed canopy forests, because of
their high fuel loading, had a tendency to change to prairies (Anderson and Brown 1986).
Thus burning destabilizes the closed canopy system. In the case of prairies, however,
fires help to stabilize the system. Vogl (1969) found alternations of floods during wet
periods and fires during drought maintained pristine open marsh, sedge meadow, and wet
prairie in lowland areas in southeastern Wisconsin (1836 to 1966). Recurring fires
perpetuated the sucker-sprouting aspen, whereas burning decadent aspen forests could
originate true prairie. Also, if burned with high intensity fires, pure even-aged aspen
forests could originate from seeds. Alien grasses in the seasonal submontane zone of
Hawaii are due to fire which favors more exotic grasses thus changing the overall
vegetation dynamics in ecosystems that did not have fire adapted species (Hughes et al.
1991, D’ Antonio and Vitousek 1992). This is also true in many prairie ecosystems in the

western US. states; e. g., the exotic cheat grass (Covington et al. 1994).

20

1.2.5 Effects of fire on nitrogen fixing plants and nutrient cycling

Fire volatilizes carbon, nitrogen and sulfur but mineralizes phosphorous,
potassium, calcium, magnesium and certain micronutrients (Ahlgren 1960, Alban 1977,
Vitousek 1985, DeBano et a1. 1998). Fire redistributes nutrients in the soil (Ahlgren and
Ahlgren 1960, Raison 1979) and most sodium, potassirun and calcium leach during the
first three months after burning (Smith 1970, Raison 1979). Microbial activities as well
as nitrogen fixation decline immediately after burning but they resume soon because of
increased ash, nutrients and soil pH (Alban 1977, Raison 1979). Such increases in
nutrients, particularly phosphorous and pH, also occur in peat burns (V ogl 1969) and
favor soil microorganisms and nitrogen-fixing bacteria which increase soil nitrogen
(Christensen and Muller 1975, Spurr and Barnes 1980, Barbour et al. 1999). For example,
in grasslands in Connecticut, Baptisia tictoria, a fire increaser and nitrogen fixer
increased by 2- and 3.5-fold after 17 years of annual and biennial prescribed burning
(N iering and Dreyer 1989). In the frequently burned (every 1-3 years) plots of longleaf
pine, burning increased establishment of legumes in the understory (Hainds et al. 1999).
However, ectomycorrhizal hyphal densities decreased following fire in the New Jersey

Plains in contrast to the expected increase (Buchholz and Gallagher 1982).

Among different nutrients, nitrogen and phosphorous are most limiting to plant
growth in most forests (Vitousek 1985). However, the loss of nitrogen during fires is
usually low, although losses may be severe with intense burns (Binkley et a1. 1992). Fire
consumes surface litter, reduces C/N ratio, reduces microbial immobilization and releases

nutrients. Fire may either increase or decrease nitrogen (Binkley et a1. 1992). Nitrogen

21

uptake and immobilization by growing vegetation are important processes that are
responsible for nitrogen retention after harvest and decomposition (Vitousek et al. 1979).
Nitrogen and phosphorous availability increases immediately after fire due to decreased
immobilization and increased soil temperature and moisture (Raison 1979, Vitousek
1985), despite the losses due to volatilization (Christensen 1985, Christensen and Miller
1975). Such increases have been found in the pine-oak forest in the nutrient poor
spodosols of the New Jersey Pine Barren, where oaks showed rapid growth after burning
(Burns 1952 in Boemer et al. 1988, Boemer et al. 1988). However, because of the
nutrient poor sandy soil the nutrient availability was higher only for a period of about
four months after burning. Also, there was a very small loss of phosphorous and calcium

(Boemer et al. 1988).

In reviewing 87 studies carried out between 1955-1999, Wan et al. (2001) found
nitrogen in fuel wood to decrease significantly (by 58%) and soil ammonium and soil
nitrate pools to increase significantly by 94% and 152%, respectively, after fire, but such
changes depended on fuel type as well as fire time. None of the variables—vegetation
type, fire type, fuel type, fuel consumption type, fuel consumption percent, time after fire,
and soil sampling depth—affected soil nitrogen amount. Soil ammonium pool increased
by two folds immediately after fire and then gradually declined to pre-fire level after one
year. Similarly, the soil nitrate pool increased less (only 24%) immediately after fire but

increased to three fold after about 0.5- to 1-year after fire and then decreased.

22

In a disturbed forest, nutrient increase is apparent until the forest grows back and
the canopy closes and then nutrient decreases later (Vitousek 1985). Resources in
disturbed gaps are under-utilized as they are released from decaying plant organic matter
and there is low uptake during the recruitment process (Marks 1974). In the absence of
fire, dense thickets of trees are formed in the understory of ponderosa forests and such
thickets intercept more light, which increases the soil organic matter. Increase in soil
organic matter leads to nitrogen immobilization by microbes causing nitrogen deficiency,
which affects herbaceous growth (Moir 1966). In loblolly pines in low productive soil (SI
27 m at 50 years), net nitrogen immobilization was highest in unburned than in two- and
four-year burn interval plots. Carbon and nitrogen were very high in the forest floor in
unburned than in once-burned areas in loblolly and long—leaf pine forests in North
Carolina (Binkley et al. 1992). Covington and Sackett (1986) also reported an increase in

nitrogen after burning in ponderosa pine forests.

In Chaparral, only 146 kg ha“1 and 49 kg ha'1 of nitrogen and potassium,
respectively, were lost after burning, but there were additional nutrient losses due to
erosion during the first rainy season (Debano and Conrad 1978). This nitrogen loss
represented about 11% of the nitrogen in the plants, litter, and upper 10cm of soil before
burning. Two years after the Little Sioux forest fire in BWCA in northern Minnesota,
potassium and phosphorous increased by 265% and 93%, respectively, in Meander Lake
that received down-stream flow from burned plots and Dogfish Lake that was down-

stream from unburned plots (Wright 1976).

23

In general, the loss from the forest due to burning has been considered to be small
when compared to the total nutrient balance of forest ecosystems, which retain most of

their nutrients even after a maj or fire.

1.3 Fire effects on pine dominated ecosystems

Pine forests are maintained by fire and adapted to low and high intensity fires
(Burdon 2002, Richardson 1998). In the US, high-valued pines such as, longleaf and
ponderosa pine are adapted to frequent surface fires (Chapman 1932, Cooper 1961, Moir
1966, Burdon 2002). Red pine and white pine are resistant to low intensity fires; but they
establish readily by seed following stand-repacing fires (Ahlgren 1976, Van Wagner
1970, Dickmann 1993). Jack pine and lodgepole pine have thin bark and have little fire
resistance; however, they are prolific seed producers with serotinous cones that open
when burned by a stand-replacing crown fire (Arno 1980, Burdon 2002, Schoennagel et
al. 2003). Pitch pine sprouts after burning and is found along with shortleaf pine in the
New Jersey Pine Barrens (Little and Moore 1949). Loblolly pine and slash pine are two
other high-valued and fire-adapted pines of the South. In general, pines with thick bark,
protected buds, serotinous cones and sprouting are considered fire-adapted (Little and

Moore 1949, Van Wagner 1970, Dickmann 1993).

1.3.1 Ponderosa pine
Historically, mature ponderosa pine forests were open savannah-like
monocultures with some cover of grass and herbs and were maintained by periodic low

intensity natural surface fires. This regime, however, was disrupted during Euro-

24

American settlement (Cooper 1960, Cooper 1961, Moir 1966, Arno 1980, Covington and
Moore 1994b). Historically, low intensity fires occurred between 3 to 30 years (Cooper
1960, Covington et al. 1997, Arno 1980, Feeney et al. 1998) and varied from place to
place. Ponderosa pine is a shade intolerant but drought tolerant high-valued timber
adapted to fire, with thick bark, insulated buds, and tolerance to crown scorch (Agee
1998). Periodic low intensity natural fires used to kill groups of trees here and there and
formed gaps that transformed into patches of young even-aged pine trees, forming a
mosaic of patches of different ages over the landscape (Cooper 1961). At high altitude ( ~
2800 m) in the Colorado Front Range, ponderosa pine mixes with Douglas-fir and
lodgepole pine forming stands that are not open and park-like. As the altitude increases
aspen and limber pine increase (Veblen et al. 2000). In the absence of fire, stands shift to

less fire resistant trees such as white fir, Douglas-fir and junipers (Allen et a1. 2002).

One century of human activities—fire suppression, overgrazing, logging and
introduction of exotic species—have changed the structure and function of ponderosa
pine forests from park-like savannas to dense thickets of sapling and shrub grth
(Cooper 1961, White 1985, Covington and Moore 1994b, Fule et al. 1997, Feeney et al.
1998, Kaye and Hart 1998, Mast et al. 1999, Allen et a1. 2002). Large ponderosa pine
density used to be 47 to 49 trees ha'l, in Flagstaff, Arizona (Covington and Moore
1994a). Tree density was 148 trees ha'1 in 1883 and after fire exclusion it increased to
1265 tpha in Navajo, Arizona (Fule et a1. 1997). Similarly, fire exclusion increased
ponderosa pine density to >3000 tpha in Arizona in 1992 (Mast et al. 1999). Currently,

the conditions of many ponderosa pine forests are considered unhealthy because of the

25

thickets of white fir and Douglas—fir (Mutch et al. 1993 in Covington et a1. 1994,
Covington et al. 1997,_Fule et al. 1997, Covington 2000). Douglas-fir used to be kept at
minimum before the fire exclusion policy (Dickman 1978). The forest fuel loads are well
above the presettlement level and are prone to stand replacing wildfires and mountain
beetle outbreaks (Startwell and Stevens 1975). It is likely that frequency and magnitude
of wildfire will increase for many years (Covington 2000). Climatic oscillations
(Covington and Moore 1994b) and elevated carbon dioxide have added to all these
changes in ponderosa pine forests (Covington and Moore 1994a). Currently, high-
intensity fire hazards have been linked with the Southern Oscillation events (ENSO)——E1
Nino and La Nino—every 2-5 years in the Colorado Front Range (Veblen et al. 2000).
Warm-phase El Nino- Southern Oscillations events cause high fuel production, whereas

dry phase La Nino events cause widespread fires.

Currently, scientists and resource managers have realized the problems related to
a century-long fire exclusion in western ponderosa pine forests (Covington et a1. 1997,
Fule et al. 1997, Veblen et al. 2000, Allen et al. 2002). The major problems have been
identified as the huge accumulation of fuel wood, dense thickets of seedlings and shrubs,
and potential insect attack, all of which increase fire hazards. Therefore, to restore and
improve the condition of these forests, wood reduction via thinning and burning has been
recommended (Feeney et al. 1998; Veblen et al. 2000, Allen et al. 2002). The shade-
intolerant ponderosa pines need not only open space but also exposed mineral soil for the
seeds to germinate and develop. Thus if both the thinning and underburning is carried

out, it is possible to maintain the healthy and low-fire hazard ponderosa forests. In a 80-

26

year-old forest at Bitterroot National Forest in Montana, F iedler et a1. (1992) found that
different combinations of burning and thinning were able to reduce 60 to 65 % of litter
and woody fuels (out of about 5 tons per ac). The main objective was to consume fuel by
burning and let trees grow to an old growth condition under a shelterwood. In the mean
time, burning was able to thin out 60% of the seedlings and saplings. Fule et al. (2001)
found thinning followed by burning to lower density, lower crown basal area and enhance

crown-fire resistance.

1.3.2 Lodgepole pine

Lodgepole pine is a shade intolerant species that bears serotinous cones (Baker
1949) and is a good example of pine adapted to stand replacing fire. It maintains a climax
forest where there is sufficient light for them to grow (Despain 1983). It dominates both
pioneer and climax forests in the sub-alpine zone of Yellowstone National Park, and it is
mixed with sub-alpine fir, Engelmann spruce, and whitebark pine in the climax (Rome
1982). In Yellowstone, the vulnerability to fire depends on whether or not spruce and fir
occupy the understory. If yes, the forest is susceptible to fire (Despain 1983) and if not,
they become resistant to fire (Rome 1982), which gives lodgepole pine sufficient time
(200-300 years) to establish as a climax (Despain 1983). Fire regimes vary as the habitats
vary. Fire return intervals are around 300 years at higher elevations with stand replacing
fires in the Northern Rockies at Yellowstone (Rome 1979 in Arno 1980, Schoennagel
et al. 2003). Similarly, the fire return intervals are 25-50 years in the upper limit of
Douglas-fir at Yellowstone and in the Bitterroot Mountains of Northern Idaho with low

intensity fires (Rome 1982, Larsen 1929). Rugged terrain in the Northern Rockies

27

makes the fire regime more complex, resulting in establishment of one to many aged
stands differing in composition and structure (Rome 1979 in Arno 1980, Arno 1980,
Arno 1993). Similar fire regimes have been found in Oregon and Washington. In the
Front Range of the Canadian Rockies, fire ensures regeneration and the long-term

population dynamics of lodgepole pine and Englemann spruce (Johnson and Fryer 1989).

1.3.3 Southern pines - Longleaf pine (Pinus palustris)

The longleaf pine of the southern coastal plains forms monocultures with a natural
ground cover of Andropogon or Carex spps. (Chapman 1932). Such vast tracts of forests
(millions of ha) were possible only because of the fire resistance of longleaf pine and
vulnerability of other species to fire (Heyward 1939). Longleaf pines produce thick, stout
root (with a lot of food reserve) for the first five years, but they do not grow much in
height and then shoot off when conditions are favorable. About 90% of longleaf pine
used to be burned every 3-4 years and that maintained its open park-like nature with rare
catastrophic fires (Heyward 193 9). Longleaf pines are not damaged by low-intensity fires
which usually occur every 3-10 years due to lightning during spring and summer months
(Chapman 1932, Wahlenberg 1946 in Campbell 1955). Long leaves (20-46 cm) and
scales covering the buds protect them from fire at the young or the grassy stage
(Chapman 1947, Burdon 2002). Bumings kill seedlings; however, the ones that survive
periodic winter burns become established (Garren 1943). Annual fires, however, are
deleterious to pine seedlings in the long-run and give rise to persistently sprouting oaks
and shrubs. One-year-old longleaf pine seedlings survive low-intensity fires but their

height growths are reduced by 5 years compared to height growths of unburned seedlings

28

(Bruce 1947). However, after the seedlings from burned plots reach 3m tall, their height
growths become similar to those of unburned ones. Effects of burning on seedling
mortality depend on the season of burn and summer bums should be avoided (Garren
1943). Stands under long-term fire exclusion as well as annual burning in North Carolina
showed an increase in hardwood sprouts in gaps between trees and decreased pine
reproduction (Heyward 1939). Even 5-6 years of fire protection affected the seedlings
negatively because of the competition with sedges and grasses (Chapman 1932).
Longleaf pine seedlings failed to establish because of the buildup of organic matter if fire
was excluded for >10 years, and stems up to 30 cm or so in diameter were killed if

burning was re-introduced during the hot summer (Chapman 1932).

After extensive eighteenth and early nineteenth century logging, most of the
longleaf pines failed to reestablish in the areas they once occupied; however, longleaf
pines were replaced mostly by different types of oaks, slash pine and loblolly pines
(Campbell 1955). The commercial southern pines, except longleaf pines, need to be about
3 to 4m high to resist even low intensity burns. Among its associates, slash pine is
resistant, whereas loblolly pine is more susceptible to fire (Heyward 1939). Loblolly,
slash and shortleaf pines grow quickly up to 10 years and develop thick bark to resist
most low intensity fires; however, they cannot survive frequent fires unlike the longleaf
pines (Chapman 1932). Loblolly, slash and shortleaf pines are similar to red pine in their
fire ecology and are shade intolerant and need mineral soils for their germination and
establishment. Shortleaf pines also can sprout after fire (Walker 1980). Among the two

associates, longleaf pine is taken over by slash pine, and loblolly pine by shortleaf pine

29

during succession. Hardwoods (predominantly oaks and hickories) dominate the
community in the absence of fire or other disturbances for a long time, as the pines
cannot establish in the shade (Walker 1980). Also, intense burning, cutting or bark beetle
attack that remove the canopy pine trees result growth of understory hardwoods to the

canopy (Walker 1980).

Prescribed burning is an essential tool for natural regeneration of pines and also
helps in decreasing brown spot needle blight (Westveld 1949). Prescribed bunting was
found to check hardwoods like scrubby oaks and shrubs like gallberry (Campbell 1955,
Lemon 1949). However, shrubs sprouted more following dormant season fire than
growing season fire (Drewa et al. 2002). Burning also increased grass forage—slender
bluestem and native legume species such as Lespedeza and Desmodium spp. (Bruce 1947,
Walker 1980). In a 32-year study, unburned plots produced only 37 kg ha], whereas
annually burned but ungrazed plots produced as much as 1120 kg ha'1 of grass (Bruce
1947). Earlier studies suggested that the regeneration of longleaf pine and understory
vegetation may be maximized by increasing gap size. Gap sizes as small as 0.1 ha are
strongly correlated with seedling and understory vegetation growth (Me Guire et al.

2001).

1.3.4 New Jersey Pine Barrens
The New Jersey Pine Barrens is the nations’ first National Reserve. The highly
drained upland forests are dependent on frequent fires and cover about 30% of New

Jersey’s land area and have scrub oaks and ericaceous shrubs— huckleberries and

30

lowbush blueberry—in the understory (Buell and Cantlon 1953, Boemer 1981, Forman
and Boemer 1981). The Pine Barren occupies 550,000 ha in southern New Jersey with
pitch pine or shortleaf pine, or both mixed with black, white, scarlet and chestnut oaks,
hickory and red maple on sandy gravelly acidic coastal areas or abandoned fields (Little
and Moore 1949, Forman and Boemer 1981, Good and Good 1984). Pitch pine is the
most dominant canopy tree, constituting most of the volume in pine-oak upland forests.
An average point today in the Barrens burns every 65 years which used to burn every 20
years. Such reduction in fire frequency favors non-fire adapted species. For example,
hardwood swamps (red maple) are being replaced by cedar swamp (Forman and Boemer
1981, Good and Good 1984). Extensive areas of very dense dwarf pine-oak forests (<3m
height) called ‘Plains’ also occur (Lutz 1934, Boemer 1981, Buchholz and Gallagher
1982, Buchholz and Good 1982, Forman 1998). These ‘Plains’ are pigmy forests formed
by fire in large areas which lack fire barriers (Givnish 1980). Dwarf trees are serotinous,
precocious and predominant in vegetative reproduction. Among the two pines, pitch
pines are of high importance as compared to shortleaf pines and are usually succeeded by
hardwood (Westveld 1949). Because of their thick bark and sprouting capabilities, pines
are persistent (Lutz 1934). Hardwoods can never replace pines completely as fires are so
frequent, formerly about 1100 fires per year (Lutz 1934, Forman and Boemer 1981,
Forman 1998, Barbour et al. 1999). The most frequent fires produced pigmy vegetation
of scrub oaks and pitch pines, whereas less frequent fires produced oak-pine forests
which shift to oaks after some time (Boemer 1981, Little 1998). Currently, the total area

burned every year has been reduced from 22,000 ha to 8,000 ha.

31

Cutting and burning suppresses the growth of mixed hardwoods and promotes
pine seedlings and are profitable in such sandy soils. Both pine species sprout when
damaged by fire (Westveld 1949, Little 1974). Because of frequent fires occurring every
15-40 years (Little and Moore 1945), all the hardwoods are of sprout origin as they are
killed back if they are <06 m; however, light winter fires do not kill well established
pines >5 cm dbh (Buell and Cantlon 1953). The oaks also sprout from underground buds
and grow faster and have a large root crown development after a series of frequent fires.
The sprouting is so dense that germination and establishment from seeds would be
difficult for new species (Westveld 1949, Little 1998). Such high production (density,
biomass, net annual above ground productivity) increases flammability by increasing fire
frequency and severity in the Barrens (Buchholz and Good 1982). Old oak pine stands
(40-60 years old) are easy to convert to pine-dominant stands by periodic burning every 4

or 5 years because of the decreased sprouting capacity of old trees (Westveld 1949).

1.3.5 Great Lake States pines (red, white and jack pine)

Red pine is often associated with jack and white pine in the Lake States. Red pine
is distributed widely in the US. from Maine to Minnesota and the southern parts of
Canada (Engstrom and Mann 1991, Van Wagner 1970). The range of white pine is
similar but extends farther south than red pine. Both red pine and white pine are
important timber species of the Great Lakes region; however, the importance of red pine
as a timber resource has increased because of damage to the jack pine by the jack pine
budworrn and to the white pine by white pine weevil and blister rust (Burns and Honkala

1990)

32

White and jack pine are more frequent seed producers than red pine. Seed
production starts later (20-25 years) in red pine than in jack pine (5-10 years) and white
pine (<20 years) (Van Wagner 1970, Lancaster and Leak 1978, Rouse 1988).
Additionally, jack pine produces abundant seeds (>4 million seeds ha'l) that have higher
viability than those of red pine. Red pine cones are destroyed immediately if burned
(Rouse 1988), whereas serotinous jack pine cones need fire to open. Jack pine cone
temperatures have been found to reach as high as 300 0C for 60 seconds while burning;
however, the seeds remain viable in soil until favorable conditions set in (Van Wagner
1970, Whitney 1987). The serotinous cones of jack pine sometimes open even with
intense summer heat if they are near ground surface. Thus, jack pine regenerates and
establishes better than red pine. White pine is a more frequent seed producer than red
pine (Ahlgren 197 6) with a good seed year every 3 to 5 years, and it is much more
capable of establishing in stands after fires, under the canopy in undisturbed stands (Van
Wagner 1970; Lancaster and Leak 1978), or in steep eroded slopes and along low river
banks (Maissurow 193 5). Red pine seeds are also highly self-fertile and thus even a
single mature tree left in the stand can provide sufficient seeds for reestablishment (Spurr
and Barnes 1980). Red pine has sporadic seed production, and a good seed year (seed
mast) occurs every 5 to 10 years (Van Wagner 1970). Both jack pine and white pine can
tolerate a layer of ash during germination whereas red pine germination is inhibited by a

thick ash layer (Ahlgren 1976).

33

Red pine seedlings have to compete with hardwood associates, such as aspen, red
maple and white birch, which are prolific annual seeders and successful sprouters (Van
Wagner, 1970). Red pine seedlings cannot survive competition with the quick flush of
herb growth that follows fire (Ahlgren 1976). Because of the competing associates and
the organic matter accumulated on the soil, red pine seeds cannot reach mineral soil that
is essential for regeneration and, thus, they have difficulty establishing in new stand
(Metheven and Murray 1974, Ahlgren and Ahlgren 1981, Rouse 1988). Red pine is a
species that cannot regenerate and establish in the forest floor with thick duff. Thus, fire
is needed to consume duff, expose mineral soils, clear the understory shrubs, and prepare
a suitable seed bed (Ahlgren 1976, Bergeron and Brisson 1990). In a simulation study
using ‘LANDIS’ to look at the effect of presence or absence of fire, Schellar et al. (2005),
found that red pine regenerates in fire rotation periods of 50, 100 and 300 years, and large

fires were needed for red pine regeneration.

Among the two differently fire adapted species, red pine survives low-intensity
fires but jack pine is usually killed; however, they co-exist (Clark 1991). Similarly,
despite the fire adaptability, fire kills almost all red and white pine <6 m in height or 15
cm in dbh immediately and larger trees may also die; however, a crown fire is less likely
after they reach >18 m in height (McConkey and Gedeny 1951, Van Wagner 1970). In
red pines, branches in mature trees are well above the ground and such branchless tall
boles also save trees from crown scorching and crown fire. Mature pine trees, however
may be affected badly or even die if crown scorch is >75% (Van Wagner 1970,

McConkey and Gedeny 1951), and intense fires might enhance bark beetle attack as well

34

(Dickmann 1993). Survival of red pine stands is due to thick bark which at the same age

is considered to be more resistant to heat transfer than white pine (Van Wagner 1970).

Both red and white pines do not regenerate well in the heavy shade of natural
stands or in severely burnt, extensively logged and fire-suppressed areas (Reich et al.
2001, Van Wagner 1970, Heinselman 1973, Lancaster and Leak 1978). However, red
pine stands are maintained by localized fires that are not so severe (Bergeron and Gagnon
1987). High intensity fires have been found to create scattered and mixed white pine
stands along Alleghany Plateau, Pensylvannia (Runkle 1985). Metheven and Murray
(1974) concluded that mature red and white pines are not damaged by low intensity
prescribed burns and excellent regeneration can be obtained but periodic fires may be
detrimental to established regeneration (Van Wagner 1970). Inability of red and white
pine to regenerate in their own stands is well known and Kittredge (1934), in Star Island
in Minnesota, found jack pine, red pine and white pine to fully occupy a site for 100, 300
and 250 years, respectively, with succession ultimately to maple-basswood in 650 years,

but only in the absence of fire.

1.4 Disturbance, fire suppression and restoration management
1.4.1 Fire suppression and fuel reduction management

As discussed earlier, natural disturbances occur in forest ecosystems and create
heterogeneity (mosaic structure) and a state of non-equilibrium with patches at different
stages of succession in space and time. To manage an ecosystem, it is necessary to realize

its ever-changing and dynamic nature and to understand the role of disturbance regimes,

35

patchiness, heterogeneity, and its patterns and processes. Such disturbances may also
alter or degrade natural communities by invasion and thus is to be considered as an
important problem to conservation management (Hobbs and Huenekke 1992). Therefore,
active management is needed and must be informed by a better knowledge of disturbance
regimes and the species to be discarded or favored (Hobbs and Huenneke 1992). Since
recolonization is the outcome of interactions between heterogeneity and disturbances,
disturbances should get proper appreciation from managers and policy makers as they are
beneficial to biodiversity (Reice 1994). Nature reserves should be designed on the basis
of minimum dynamic area and such areas must be guided by disturbance theory and
patch dynamism regulated by gap formation to gap filling (Pickett and Thompson 1978).
The smallest area in nature reserve should have natural disturbance regime that maintains
internal recolonization and hence minimizes extinctions. Traditional successional and
climax concepts may not be advantageous to sound vegetation management (N iering
1987). Such considerations are also needed because almost all management plans involve

decisions that alter landscapes (Turner 1989).

In the US, fire suppression carried out for almost a century has caused unique
ecological problems across landscapes and ecosystems, including the buildup of fuels that
cause more intense and uncontrollable fires (Mutch, 1970); vulnerability to insects, pests
and diseases; loss of fire dependent species from sites that are not allowed to burn (White
1979); loss of biodiversity (Christensen 1985); and ultimately an alteration of the natural
disturbance regime (Sousa 1984). Such problems have led to increased use of low

intensity fire or prescribed burning in forest, natural areas, and national park management

36

(Kilgore 1973, Kilgore 1976, Mutch 1976 in Rome 1982, Christensen et al. 1989,
Feeney et a1. 1998, Veblen et al. 2000). For example, Yellowstone National Park has
adopted a let-bum policy for some lightning-caused fire since 1975, but only if human
life, property, or other values are not threatened by such fires (Romme 1982). Landscape
diversity was found to increase following two small fires and a mountain pine beetle
outbreak in Yellowstone National Park, a non-steady state subalpine ecosystem, and such
disturbances may significantly influence wildlife habitat, stream flow, nutrient cycling,
and other ecological processes and characteristics within the park (Rome 1982). The
USDA Forest Service has also increased prescribed burning since 1989 (Arthur et al.
1998). Christensen et a1. (1989) suggested that different forms of fire suppression as well
as prescribed burning should be included in management of wilderness areas, but there is
not a single cook-book recipe to manage ecosystems that are so complex. As discussed
earlier, to restore and improve the condition of ponderosa pine forests, reduction of wood
as well as thinning and burning have been recommended (Feeney et al. 1998, Arno and

Fiedler 2005, Veblen et al. 2000, Allen et al. 2002).

1.4.2 Low intensity prescribed fires

Lemon (1949) described prescribed burning as a “purposeful and planwise use of
fire under strict control.” It is used as a silvicultural tool to manage forests and wildlife.
Fire prepares a suitable seed bed by consuming litter/duff, clearing understory shrubs and
competing associates, opening the canopy, and exposing mineral soil for the
establishment of pines in the Great Lakes as well as other regions (Lemon 1949, Buell

and Cantlon 1953, Van Wagner 1970; Metheven and Murray 1974, Kilgore 1973,

37

Ahlgren 1976, Alban 1977, Lancaster and Leak 1978, Thomas and Wein 1985;
Dickmann 1993). The reduction in depth of organic matter with fire increases
establishment because seedlings are closer to the constant water supply of more
decomposed organic layers and mineral soils (Thomas and Wein, 1985). Also, the
mineral soils have higher moisture content and do not dry quickly (unlike the litter).
Mallik and Roberts (1994) found red pine regeneration to increase with a decrease in duff
thickness. The openings created by burning allow sunlight to reach to the forest floor.
Burning decrease competition by killing the existing vegetation; however the extent of
killing depends on the type, duration and intensity of the burn and also on the vegetation
itself. Such killings modify the soil micro-environment (Raison 1979). Fire affects
species diversity (Reich et al. 2001) as well as soil organic matter, partially or wholly
(Debano and Conrad 1978) and re-establishes the pioneer condition. Regeneration of red
and white pine has failed in pine and hardwood mixed stands under different
management systems, and thus there is alarming loss of areas containing these species
(McRae 1994). Therefore, fire is needed to perpetuate natural pine ecosystems

(Maissurow 1935, Van Wagner 1970, Dickmann 1993).

Low-intensity (200-500 Btu/sec-ft) prescribed burns are sufficient to control
understory herbs and shrubs in red and white pine stands (Van Wagner 1970). In the New
Jersey Pine Barrens the density of pine seedlings was found to increase as much as three
times after burning (Little and Moore 1949, Buell and Cantlon 1953) but was unfavorable
to oaks, as their seeds were exposed to sun and bird and mammal predators (Boemer et

al. 1988). Additionally, in the case of longleaf pine, fire consumes the mat of dead wire

38

grass, which otherwise would not decay, increasing fire hazard. Fire also reduces damage
from pests and diseases, especially brown spot needle blight (Chapman 1932, F eeney et
al. 1998) and red pine cone beetle (see ref. in Dickmann 1993). Such reduction in pest
and diseases following prescribed fire has been found in ponderosa pine (Covington et al.

1997)

Prescribed fire also reduces the heights of understory vegetation and provides
browse for wildlife, improving habitat and thus increasing hunting and recreation scope
in the Great Lakes States (Little and Moore 1949, Rouse 1988, Dickmann 1993, Bender
et al. 1997). Niering and Dreyer (1989) have highlighted fire’s potential in the
management of wildlife and restoration of grass-dominating areas within oak forests.
Further, the reduction of hazardous forest fuels by prescribed burning helps to prevent
forest fires and makes suppression of them easier. For example, prescribed fire is used to
remove slash from about 10% of all clear cuts in Canada and 20-25% in British Columbia
(Sullivan et al. 1999). Prescribed fire is also useful in managing grasslands, prairies and
wild flowers and maintaining park-like vistas with high visual qualities in stands; e. g.,

Sequoia National Park (Biswell 1989).

Burning forests may have adverse effects as well. Fire is not selective in killing
vegetation and their size classes, and thus it may be difficult to save desired plants or the
desired size classes. Moreover, the fire controls vegetation only for a short time. Further,
negative effects of prescribed burning on the physiological condition, growth, and

survival of ponderosa pines have also been reported in studies where fuel loads were not

39

reduced prior to burning (Sutherland et al. 1991). Also, small mammal diversity and
abundance was reduced after burning clearcuts occupied by spruce-fir (Sullivan et al.
1999). Prescribed burning also has risks of escape and produces blackened charred stems,
smoke and other particulate matter and several radiatively active gases (C02, NOX, SO x
and others), which add to pollution and the greenhouse effect. Thus, prescribed fire may
prove unacceptable in some environments. In any case, it needs to be done correctly

(Dickmann 1993) and its use depends on clearly defined objectives.

Fire risks should be understood for conflict resolution and sustainable
management (Cleland et al. 2004). Risks of fire escape are also because of lack of proper
training in the use of prescribed burning (McRae et al. 1994). But there will be fire
regardless of whether we burn or not. Prescribed fire should be highly desirable as long
as its goal is the betterment of human beings without irreversibly changing the ecosystem

or vegetation.

1.4.3 Red pine density management

Red pine is one of the most widely planted and intensively managed species in the
Great Lake States (Michigan, Minnesota and Wisconsin) with a very high economic as
well as ecological importance in the region (Lundren 1965, Benzie 1977, Bassett 1984).
Because of its faster growth, higher productivity and ability to grow well in infertile soil,
compared to its hardwood and other conifer associates in the region, it has been planted
throughout these states since the early part of the last century (Gilmore and Palik 2006).

Currently, it occupies about 0.4 million ha in the Great Lakes states alone (Benzie 1977,

40

Lundgren 1983, Parker et a1 2001). Red pine grows well in clayey, loamy and sandy
acidic soils (Bassett 1984) and site indices ranging between 13.7 to 22.9 m at 50 years
(Gilmore and Palik 2006). The products obtained vary with management or the
silvicultural technique applied to red pine plantations and its growth and yield depends on
stand age, site quality and stand density. For example, high-density management
produces small poles and sawlogs, whereas low density management produces large
utility poles and sawlogs (Buckman 1962, Lundgren 1965). Red pine grows fast, provides
better yield than other native pines, has good form and is relatively free from defects and
diseases (Buckman 1962, Lundgren 1965, Burgess and Robinson 1998, Penner et al.

2001).

Because they had such a high value, most of the pristine natural red pine forests in
the region were deforested by the turn of the last century. This situation demanded that
extensive research and management be carried out, and various aspects of research
started as early as the beginning of the last century in both the US and Canada (Burgess
and Robinson 1998). Numerous studies carried out in the Lake States as well as in
adjacent provinces of Canada have examined density management, thinning, grth and
yield and other aspects of management including succession, wildlife habitat and browse
availability (Rouse 1988, Bassett 1984, Lundgren 1981, Dickmann et al. 1987, Dickmann
1993, Bender et al. 1997, Burgess and Robinson 1998, Penner et al. 2001). In the last 50
years or so, US and Canadian research has explored the effects of current and historical
wildfire on vegetation composition, structure and succession and their role in the

management of red pine forests (Mutch 1970, Heinselman 1973, Benzie 1977, Whitney

41

1986, Bergeron and Gagnon 1987, Engstrom and Mann 1991, Dickmann 1993, McRae et
a1 1994, Henning and Dickmann 1996, Nevvrnann and Dickmann 2001, Cleland et al.

2004).

Red pine is a shade intolerant (light demanding) species (Benzie, 1977) and is
recommended for even-aged management. Red pine can also be managed as uneven-
aged stands with a selection system; however, this becomes expensive because thinning
must be carried out across different size-classes (Benzie 1977, Parker et a1. 2001). If the
management objective is related to some research questions or obtaining sawtimber, then
uneven-aged management could be justified and applied. For maximtun productivity of
red pine it is necessary to thin it periodically (Day and Rudolph 1972). Red pine response
to thinning is very quick. Stephens and Spurr (1947), for example, observed daily and
weekly increases in growth rates of red pine after thinning and pruning. Thus, thinning is

one of the most important management tools for this species.

Thinning has a variety of well-understood effects on tree stands, so the type of
thinning undertaken depends upon the management objectives. Artificial thinning
usually involves cutting or removal of suppressed or overtopped trees, which cannot
perform well and would eventually die if left unattended during stand development. In a
study at Petawawa Research Station in Ontario, Canada, Burgess and Robinson (1998)
found that unthinned natural stands of white and red pine had mortality as high as 10
times that in thinned stands. Thinning can enhance the diameter growth but not the gross

(overall) productivity (Haberland and Wilde 1961). The growth potential of the

42

remaining trees can be increased by removing suppressed, diseased, deformed and dying
trees and concentrating the growth in fewer larger trees that are retained (Spur et al. 1957,
Althen and Stiell 1990). Such removal does not decrease the potential unit-area volume
growth of the stands (N yland 1996). Even if released, suppressed trees do not grow well
due to their physiological deficiencies and are less productive of unit area biomass than
the unifome distributed large ones that are retained (Smith 2003). According to
Kozlowski and Peterson (1962), dominant red pines have early growth initiation, faster
growth and longer growing season than the intermediate and suppressed trees. Crown
touching and friction leads to deterioration and decreased average grth of diameter in
unthinned stands (Spurr et al. 1957). All of these factors result in growth inhibition in
suppressed trees. Further, if the stands are too dense and grown for a long time without
any treatment, there is loss of volume grth due to suppression (Buckman 1962). Thus,

thinning is needed to maximize the net productivity of the stands.

In the Lake States in 15 to 20 years-old stands, if the tree density was more than
2500 trees ha", there was stagnation; yield and MAI varied with the intensity of
management (Rudolph 1957). Coffmann (1976) also found that the gross as well as net
volume grth in red pine plantations decreases if unthinned for a longer period and it is
caused by mortality due to self-thinning. Stiell et al. (1994) in an overstory release
experiment in a 80-year old white pine mixed wood, found 80% increase in saw log
volume after 20 years of thinning by bringing back the stands to residual BA of 16 m2 ha'
1. The greatest growth was 159 m3 M”. Smith (2003) found that the larger trees reserved

during the thinning improved production of board foot volume.

43

The removal of suppressed or overtopped trees provides extra space and resources
to the trees retained, increases rates of diameter and volume growth, shortens the time to
grow to a given diameter, shortens rotation time, and increases yield (net production) by
removing trees through early harvests that otherwise would die (Spurr et al. 1957, Smith
1986, Nyland 1996, Lundgren 1981, Parker et a1 2001). So long as the gaps created by
thinning are fully occupied or covered by green vegetation, gross productivity does not
change whether thinned or not (Bassett 1984). However, if there is any vacancy in
growing space, productivity is reduced and will take some time to recover (Day and
Rudolf 1972, Gilmore et a1. 2005). If there are fewer trees they tend to become larger and
if there are many trees they tend to be smaller; however, the total basal area growth rate
does not change. Such increase in diameter affects net increase in volume grth
(Assman 1970, Nyland 1996). Burgess and Robinson (1998) found an increase in
diameter grth in both white and red pine stands after thinning. Residual saw log
volumes were concentrated in the fewer and larger natural white and red pine trees (40
years old) after a series of six thinnings over 71 years at Petawawa Research Forest,
Ontario Canada. Such volume accumulation in large sized trees makes them more

valuable as well.

Thinning has some disadvantages too. For example, if red pine stands are thinned
heavily, wind susceptibility might increase, although red pine is considered to be a
relatively wind firm species (Buckman 1962). If only the inferior and suppressed
understory or overtopped co-dominants are removed from below, there could be very

small or even no increase in net growth as their removal might not stimulate additional

44

growth in larger trees (Smith 1986, Nyland 1996). Other disadvantages of thinning are: i)
the high cost involved during harvesting; ii) formation of large openings that stimulate
hardwood invasion (Burgess and Robinson, 1998); iii) damage to residuals during felling

and skidding; iv) firngal (rot) invasion through cut stumps; and v) increased fuel loading.

If the stands are too dense, they might also need a precommercial thinning
(Benzie 1977). Buckman (1962) advises that thinning start early at a stand age of 25
years. Further, early thinning to a lower residual density than recommended for
maximum volume production will promote understory succession of woody and
herbaceous plant species thus increasing the flora and fauna diversity and improving the
visual appearance of red pine monocultures (Dickmann et a1. 1987, Dickmann 1993).
Again, whether it is advantageous or not to do so depends upon the objective of

management.

Lundgren (1965) showed that a 25-year-old red pine stand regularly thinned (from
below and above) every 10 years to residual BA of 20.6 m2 ha'l had the highest
investment returns when compared to other regimes. Similarly, Benzie (1977) also
recommended for the basal area to be retained as low as 20.6 m2 ha'l for better yield. In
the growth projection study by Buckman (1962), red pine growth and yield with three
schedules of thinning to 20.6, 27.5, 34.4 m2 ha'l were not significantly different;

however, he found out that the radial growth could be increased at lower residual density.

45

In a study using a grth and yield simulator program called REDPINE,
Lundgren (1981) found that either thinning to low densities or planting fewer trees would
accelerate diameter growth and provide larger trees at a given age. Lundgren
recommended the best tree density to be 494 tpha for sawtimber (with initial density 123
to 3954 tpha); below this density the mean annual increment (MAI) declined. Similarly at
the higher densities (2965 to 3954 trees ha”), the total cubic volume production was
found to level off. Therefore, both the initial tree density (trees ha”) and residual stand
density following thinning strongly influenced diameter growth, although there were
fewer larger trees at the time of harvesting when fewer trees were retained. This effect
was more pronounced in the better sites. He also showed that site index, timing of
thinning, and rotation length affect volume of wood produced. Total cubic volume MAI
was found to be affected more by site index than length of rotation. In a productivity
study of red pine in Lake States using another growth model (STEMS), Lundgren (1983)
found that the merchantable volume from unthinned stands to be 18% higher and that
from the thinned stands to be 32% higher than current estimates in the Lake States.
Average yield in thinned stands was found to increase from 6.4 to 7.1 m3 ha"l per year
after thinning and the culmination of mean annual increment increased from 55 years (in

unthinned stands at 18.3 site index) to 80 years.

Gilmore et al. (2005) in their simulation study of a 46-year-old red pine plantation
tried to test the Langsaeter Hypothesis, which stated that “the total production of cubic
volume by a stand of given age and composition on a given site is, for all practical

purposes, constant and optimum for a wide range of density stocking. It can be decreased

46

but not increased by altering the amount of growing stock to levels outside this range.” In
their study with stand level production equations and growth projection models, there
was a difference of only about 10% of actual volumes on average. They did not have
strong evidence to support this hypothesis. The low thinning with 32.1 m2 ha'1 of BA
retained had 30% more volume than unthinned. This volume was lower than the
geometric and crown thinning set at 28.7 m2 ha'l BA, which had a volume increase of
35% over the unthinned control. The unthinned plantation grew in volume by only about

22% at the end of the 10-year projection.

Well-spaced stands use more resources (sunlight, water and nutrients) as they
occupy the site more uniformly (Buckman 1962). In a 27-year-old spacing study, Marty
(1988) found that 39% of red pine trees in the thinned stands with 494 tpha reached their
threshold of sawtimber size (23 cm dbh) much earlier than in densely planted stands,
showing that wide spacing in red pine plantation lowers the per unit area planting cost.
Penner et al. (2001) in an initial spacing trial of red pine established in 1953, thinned in
1982 and sampled in 1998, found that both the initial lower spacing (that is, higher
densities) of 1.2 and 1.5 m had higher mortality rates than the higher (i.e., less dense)
initial spacing of 1.8 to 2.4 m. In terms of total volume production, thinning was found to
reduce the total time needed to reach utility pole size with little mortality. Although the
lower initial spacing helped to produce more poles, it required thinning to prevent

mortality and maintain good diameter growth.

47

Following an improvement cut (set BA to 6.9, 11.5 and 16.1 m2 ha") in a 80-year-
old mixed hardwood forest in Canada, Stiell et al. (1994) found highly significant grth
responses in white and red pine 20 years after release. There was about an 80% increase
in sawlog volume in the treated stands, and also mortality and rotation age for red pine

was significantly reduced, increasing the natural rate of succession.

To summarize, disturbances are sources of heterogeneity that create patchiness
(mosaics of patches) in the environment of plant/forest communities in time and space.
At any time such disturbances not only create patches of different sizes, shapes and ages
but also different stages of patch development or successional stages in forest
ecosystems. Looking at the fire regimes throughout the temperate North America, it is
revealed that fire could play an important role in the restoration and management of
forest ecosystems. Further, learning from the past experience of fire suppression in the
US, prescribed fire or controlled burning is being used more and more as a tool in the
management and restoration of forests, national parks and reserves. Fire could also be
important in restoring pine forests in the Lake States, as red and white pines sometimes
have problems in regenerating and establishing in natural stands, as these pines need
exposed mineral soils with clear forest floor and reduction in competing hardwoods for

their successful establishment.
I studied long-term effects of fire in a red pine plantation in northern Lower

Michigan, since conventional thinning or other silvicultural systems alone may not be

sufficient to create a suitable environment for these species. Also, there is lack of

48

information on long-term effects of low-intensity fires on red pine forests. Effects of
single and multiple fires at different intervals after one to two decades of burning on the
structure, composition and overall vegetation dynamics of a red pine plantation were
studied and are presented in Chapter 2. Further, longer term effects of such burnings were
studied in relation to no-management and different thinning regimes projected for 100
years into the fixture with the use of the Forest Vegetation Simulator and are presented in

Chapter 3.

49

Chapter 2: Effects of Low-Intensity Fires on the Vegetation of Northern Michigan

Red Pine (Pinus resinosa Ait.) Plantation.

2.1 Introduction

Red pine was a prominent species in the forests and barrens of the Lake States
before European settlement. However, it declined in natural stands in the nineteenth and
early twentieth centuries due to harvesting and slash burning (Heinselman 1973). Before
European settlement, low to medium intensity fires occurred every 5-50 years or even 70-
80 years in natural red pine stands (Bergeron and Brisson 1990, Englestrom and Mann
1991, Van Wagner 1970) and occasional intense stand replacing fires occurred at
intervals of 150-200 years (Bergeron and Brisson 1990). Such occasional intense fires
were ideal for red pine regeneration and establishment because they consume litter,
decrease duff thickness, expose mineral soil and prepare seed beds (Van Wagner 1970,
Alban 1977, Engstrom and Mann 1991, Dickmann 1993, Mallik and Roberts 1994).
However, the exclusion or suppression of fire after the 19203 has disrupted the natural
regeneration and establishment of red pine, which does not regenerate well in fire

suppressed areas (Reich et al. 2001).

Prescribed burning is being used as a potential tool for red pine regeneration
(Alban 1977). Mature red pine trees are tolerant of low-intensity understory fires, and the
species is considered to be fire adapted because of its very thick and insulated bark which
protects the cambium from fire damage (Van Wagner 1970, Metheven and Murray 1974,

Whitney 1986 and 1987, Rouse 1988, Dickmann, 1993). Its bark is considered to be even

50

more resistant than of white pine (Van Wagner 1970). Branch free boles up to 5 m also

help red pines withstand fire.

Multiple prescribed fires remove unwanted vegetation such as maples, hazelnut
and balsam fir and are useful in red pine regeneration (Alban 1977; Metheven and
Murray 1974; Van Wagner 1970; Mallik and Roberts 1994) and also alter post-fire
vegetation cover by consuming some buried seeds and rhizomes (Thomas and Wein,
1985). However, burning at frequent intervals also reduces the number of red pine
seedlings or even kills them (Henning and Dickmann 1996; Van Wagner 1970), so

burning should be done only as needed.

Short-term effects of single and periodic prescribed burns in red pine ecosystems
in Michigan have been studied (Dickmann 1993; Henning and Dickmann 1996;
Neumann and Dickmann 2001). However, the long-term effects (>10 years), as well as
the effects of different growing periods (or length of time) after the last fire, have not
been documented. The overall goal of my research is to broaden the understanding of the
long-term effects of prescribed burning on species diversity, regeneration and
establishment of red pine in northern Lower Michigan. The specific objectives are to: a)
examine the long-term effects of both short and long interval low intensity prescribed
fires on the vegetation dynamics of overstory red pine trees, herbaceous ground flora and
understory woody vegetation in mature red pine ecosystems; b) explore the relationship
between fire and red pine ecosystems by quantifying plant species diversity of various

sites burned with low intensity prescribed fires; and c) determine the effects of low

51

intensity prescribed fires on regeneration and age distribution of saplings of red pine and

its associates.

2.1.1 Global research hypothesis

A disturbance is any relatively discrete event in time that disrupts ecosystem,
community, or population structure and changes resources, substrate availability, or the
physical environment (Pickett and White, 1985). Also, disturbances are different in their
magnitude, frequency and size and their physical and biological effects are different at
different layers, species and size classes (White 1985). A low-intensity fire affects
ecosystems in different ways. First, it affects the vegetation dynamics of a stand by
consuming duff, reducing competition, changing resource availability and altering the
seedbed and seedling establishment. Second, it alters the understory vegetation by
selectively killing woody stems of smaller diameter classes, especially conifers. Third, it
affects small-sized sub-canopy trees by selectively killing them and promoting the
growth of large-sized mature overstory trees. Thus, fire ultimately changes a red pine
ecosystem as a whole, by killing or altering the growth and development of community

components.

This study focused on how red pine forest changes during the course of succession
following different burning regimes. The following working hypotheses were tested:
a. Low intensity fires cause changes in community structure, species composition,
and diversity of a red pine ecosystem and such changes are evident even 10 or

more years following burning.

52

b. Compared to total fire exclusion, red pine ecosystems with a history of one or
more low intensity fires (discontinued for more than 10 years) will have:

i) higher species richness, diversity and coverage of understory vegetation,
especially with more frequent burning.

ii) a decrease in density and diameter of larger understory woody vegetation
but an increase in density of smaller understory woody vegetation,
especially with more frequent burning.

iii) a decrease in density but an increase in growth and yield of mature
overstory trees.

iv) shifted from a hardwood dominated to a red pine dominated understory.

2.2 Materials and Methods
2.2.1 Study site

The study was carried out on experimental plots established in the 19803 in the
Pigeon River Country State Forest Management Unit in northern Lower Michigan
(Figure 2.1). My study gave continuity to established on-going research (Henning and

Dickmann 1996), which was nearing two decades in duration when I began.

The study area, located along Clark Bridge Road, is a 61 ha-pine plantation
established in 1931 with red, white and jack pines. In 1970, all the white and jack pines
were harvested, leaving a variable residual basal area (mean 10 m2 ha"), much of it in an

under-stocked condition. The site is very productive; the Emmett Sandy loam soil

53

Figure 2.1: Location of Pigeon River Country State Forest in the Lower Peninsula of
Michigan and Red Pine Study Site experimental plots along Clark Bridge Road. Unused

plots in each block were designated for fall burns, which never occurred because of lack
of suitable weather.

C= Unbumed control
1 = One spring burn (1985)
2 = Two spring burns (1985 and 1990; S-year burn interval)
4 = Four spring burns (1985, 1987, 1989 and 1991; 2-year burn intervals)

   
      
   
   
  

Clark Bridge Road Red Pine Study Site
T 33N. R 1w, Section 25

 
 

Pigeon River
Country State

 

     

  
   
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Forest 1 L
Lower Peninsula
of Michigan | '/
I
'I
* Cl k 8 id e R d ‘/"l
Lansing \ ar r 9 oa : :|
BIO??? l C 4 2 i n
ca. . a H, H
II
Block C C 1 '1, 1' i'
N ca. 8.1 ha 2 4 ,3 l'
IL .
I i
W E '~ Block B
\t‘ ca. 7.9 ha I I C 2 4 I
u ’ I
S -:.“_- =- -:-_.-_-..-_=..-’_ _ _. .- _ .. _ -_.
I

 

54

(Coarse-loamy mixed frigid, Alfic haplorthods) produces an average site index for red

pine of 20m (65 ft) at 50 years.

The experiment I used was set up in a randomized complete block design
(RCBD). The study site had three blocks (A, B & C), each of which had four treatments
(Figure 2.1). The experimental plots (0.86 to 1.0 ha) were burned beginning in 1985
during spring with low-intensity prescribed fires with planned intervals between
successive burns of two (four fires completed), five (two fires completed), and 10 (one
fire completed) years. Thus, the study site provided plots with a mix of fire frequencies

and intervals and different growing periods after the last fire ranging from 10 to 18 years.

The study plots were measured in the summer of 2001 and 2003 and analyzed to
test the long-term effects of different prescribed fire regimes. The blocks (A, B & C)
were within few minutes walking distance but not adjacent to each other. Each plot in the
three blocks was divided by two north-south running transects. Four points along each
transect were chosen randomly and served as the center of subplots. Thus, there were a
total of eight subplots in each plot within a particular treatment. However, only five sub-
plots were taken in the one-bum plot of block A because an intense flare up during the
first burning killed many of the overstory trees and the understory woody and ground
vegetation looked quite different than other plots. The center of each subplot was marked
and also a nearby large tree was selected as a witness tree for each of the points, marked
with paint, the direction and distance from the point was recorded and marked, and GPS

coordinates logged to relocate the points.

55

Vegetation data were collected as follows:

Overstory tree measurements were taken in variable radii plots using a 20 BAF
(English) prism at each of the sample points along each north-south running
transect. Diameter at breast height (DBH) was measured using a diameter tape
and tree height using a clinometer. Tree density, basal area, volume and mean
annual increment were computed from these data and used for further analysis.
All trees and shrubs >l.4m tall and < 10cm DBH were categorized as woody
understory and measured at the same sample points as above using 100 m2
rectangular sub-plots. DBH of individual plants of each species was measured.
The age of red pine, red maple, back cherry, beech and white ash seedlings >1 .m
in height and <10cm in dbh were determined from a total of 218 sample cross-
sections cut at the base of the understory woody vegetation to establish the
relationship of their establishment with the fires.

All plants < 1.4m tall were categorized as herbaceous ground vegetation and were
measured in 1 m2 quadrats. Three such quadrats were taken in each 100 m2 plot,
one at the center and the other two on the outer sides (i.e., 24 subplots per plot).
Percent cover of all species was estimated in each of the 1 m2 subplots. Also,
number of individuals of each species except grasses, mosses and lichens were
counted in those plots.

Species area curves for the woody understory and herbaceous ground vegetation

were constructed in all of the three blocks to see if the sampling was adequate.

56

2.2.2 Data analysis

Species diversity; vegetation composition of the tree, understory woody and
herbaceous ground vegetation components and their density, frequency, growth and yield
were evaluated for each of the burned and unburned study treatments. Different indices of
species diversityw—species richness, Simpson’s Index, Reciprocal of Simpson’s Index,
Shanon’s index, and Evenness index—were used to examine biodiversity in the
understory woody and herbaceous ground vegetation (Magurran 1988). Relative Value
Index or Importance Value (IV) also was determined for overstory trees and woody
understory trees as the magnitude of this value gives a better indication of species
importance as compared to just density, frequency or basal area alone (Curtis and
McIntosh 1951). IV is the sum of relative density, relative frequency and relative basal

areas.

The Mixed Linear Model (ProcMix in SAS) was used to analyze the data for
treatment effects. Normality of the data was checked and natural log or other appropriate
transformations were done. Arcsine transformation was done in case of ground vegetation
cover. A binomial test (Proc Glimmix in SAS) was used to test if the proportions of one
species and the remaining others were significantly different across treatments. Such tests
were repeated with all of the species or groups of species in the overstory tree, understory
woody and the herbaceous ground vegetation groups. Similarly, binomial tests were also
done to examine the proportions of different size classes of overstory and understory
trees. Goodness of fit or chi-square tests were used to compare the observed and expected

number of species in different treatments as well as to see if the diameter classes and

57

species under different treatments had some associations. Least Square Difference (LSD)
was carried out to compare among treatments. Because of the low number of replicates
and heterogeneous nature of the experimental blocks, a standard p-value (<0.1) was set

for examining the rejection or failure of hypotheses.

If an F -test was significant (P<0.1), means were contrasted using a two sided t-
test in three different ways: 1) unburned plot against all of the burns, 2) unburned and
one-bum plots against two and four-bum plots, 3) one-burn against two and four burn
plots. Biodiversity indices were tested in the Mixed Linear Model using ProcMIX.
Microsoft SAS was run for all of the analysis. JMP (a SAS product) and Excel were also
used for summarizing data. Images in this dissertation are presented in color and used for

discussion.

58

2.3 Results and Discussion
2.3.1 Overstory trees (>10 cm dbh)

The overstory trees in the Clark Bridge Road plantation consisted mostly of red
pine, with Acer rubrum, A. sachharum and Pinus banksiana as minor tree species
(relative frequencies respectively of 94, 2, 2 and 1 percent of a total of 414 trees
inventoried in all of the plots). The remaining one percent consisted of F raxinus
americana, Pinus strobus, and T ilia americana. The dominance of red pine in the plots is
clearly shown by the Importance Value (IV) (Table 2.1). IV of red pine overstory trees
was highest in the four-bum plots and lowest in unburned control plots. Block A had the
highest density of red pine trees followed by block C and block B. Block B contained

only red pine trees with no other overstory tree species in the sample.

Table 2.1: Stand characteristics of the overstory (>10 cm dbh) in experimental red pine plantation
under four burn treatments (n=12; df=6).

 

 

Number of Tree Density Basal Area Volume MAI Red pine

Treatments observations tree ha.I m2 ha.l m3 ha'l m3 ha'I yr '1 IV

Control 3 281 a (90.2) 19.3 a (3.2) 203.9 a (38.9) 2.9 a (.5) 186

One Burns 3 128 b (42.5) 17.9 a (1.6) 193.5 a (39.61: 2.7 a (.6) 267

Two Burns 3 146 b (48.3) 21.4 a (3.2) 227.0 a (38.9) 3.2 a (.5) 278

Four Burns 3 148 b (49.1) 23.3 a (3.2) 261.8 a (38.9) 3.7 a (.5) 285

Contrasts Tree Density
Control vs All Burns 5
C & 13 vs 23 & 4B 5
13 vs 28 & 48 ns

 

The tree density are the log back transformed values.

MAI = Mean Annual Increment

IV = Importance Value (sum of relative frequencies, density and basal areas)

The means used above are the Least Square Means (LSM).

LSM in each column followed by the same letters are not significantly different at p S 0.1
s and ns indicate significant and non-significant respectively

Standard errors of means (i) are given in parentheses

C, Control; 18, One Burn; ZB, Two Burns; 4B, Four Burns

 

 

 

59

The effects of the prescribed burning treatments on stand and tree characteristics
of the 70—year-old red pine plantation were analyzed statistically for both the sub-plot and
plot level means, and significant treatment effects were detected only for overstory tree
density (Table 2.1 and 2.2). Among different treatments in my study, the overstory tree
density was significantly higher in unburned controls than in burned plots and
approximately double in control (281 trees ha'l) than in burn treatments (Table 2.1).
Unbumed controls were significantly different in overstory density than all of the burned
plots together. Also, overstory tree density in unburned controls combined with one-bum
treatments were significantly different than in both of the repeatedly burned plots
combined. Here, the higher density in unburned controls as compared to burned plots was
both due to ingrowth in the sub-canopy layer of the unburned controls as well as removal
of substantial number of sub-canopy trees during repeated burnings. However, since the
prescribed burns were of low intensity, mature overstory red pine growth and yield were

not affected adversely (cf. Mc Conkey and Gedney 1951, Van Wagner 1970, Dickmann

Table 2.2: Mean characteristics of overstory trees (>10 cm dbh)
in experimental red pine plantation under four burn treatments
(n= 414; df=6).

 

DBH Height Crown Ratio
Treatments cm m %
Control 41.2 a (1.8) 25.6 a (2.0) 4.5 a (0.5)
One Burn 44.7 a (1.8) 26.5 a (2.0) 4.5 a (0.5)
Two Burns 45.7 a (1.8) 25.9 a (2.0) 4.8 a (0.5)
Four Burns 45.2 a (1.8) 27.1 a (2.0) 4.0 a (0.5)

 

dbh = Diameter at breast height;

The means used above are the Least Square means.

LS Means of parameters followed by the same letters are not
significantly different at p301.

Standard errors of means (:h) are given in paranthesis

Note: Crown ratio 1 equals 10% of tree heigfl

 

 

 

6O

1993, Henning and Dickmann, 1996, Neuman and Dickmann 2001). In previous studies,
a single low-intensity surface fire < 200 Btu/S/ft. (692 Kw/m) or repeated fires at two and
five-year intervals neither caused significant pine overstory mortality (Henning and
Dickmann, 1996) nor affected the basal area in a mature red pine stand (N euman and
Dickmann, 2001). There was a negative correlation between the density and the average
diameter of the overstory trees in my study (11 = 93; p-value 5 0.0001, r = -0.7827). The
control plots had the highest density of overstory trees and the lowest average diameter
(41.2 cm); whereas, the burned plots had the lowest density of trees and larger average
diameters (Table 2.2). Average tree height also was slightly lower in the control plots.
The reduced average diameter and height of overstory trees in unburned controls was due
to the presence of greater numbers of smaller trees compared to those in the burned plots,
where most of the small—sized trees were killed or severely scorched during burning.
Earlier, Lunt (1950 in Dickmann 1993) found 20 years of annual burning to

increase height and volume of red pine in Connecticut and the increases were thought to
be due to the increased pH as well as available phosphorous and exchangeable calcium.
There was not much difference in crown ratios among the burned and unburned plots in

my study, although they were lower in the four-bum plots.

There were no significant treatment effects on basal area, volume and mean
annual increment (MAI) of the overstory trees (Table 2.1). Nonetheless, there was a trend
of increasing stand values with more burning. Basal area, volume and MAI all increased
as the munber of burns increased from one- to two- to four-burns, with the highest growth

occurring in the four-burn plots despite having less than half the density of trees than in

61

the unburned controls. Among the two and four-burn plots, the latter had the highest

basal area, volume and MAI, even though both had the same tree density.

Fire as a disturbance agent not only works at a stand or community level, but also
at different levels of species and their diameter classes (Pickett and White 1985). I tested
the effects of burning on overstory tree characteristics and their size classes to see if the
effects of fire were different at different levels. However, the effects on species were not
tested as red pine was so dominant in the overstory. Overall, the contribution of red pine
to total density, basal area, volume and MAI in four-burn plots were 95, 98, 99 and 98%,
respectively, and were the lowest in one-burned plots among burned plots (Appendix
2.2). Unbumed control plots, however, were clearly different than burned plots; red pine
density, volume, basal area and MAI dropped to 35, 82, 89 and 95%, respectively, of the
total. In unburned control plots, red maple (32% of total density), followed by sugar
maple (22% of total density) and jack pine (8% of total density) were important

members.

To examine the effects of fire on the overstory composition and structure, trees
were grouped into three different diameter size classes: large (42 to 62 cm), medium (32
to 42 cm) and small (10 to 32 cm), which will be called dominant overstory, mid-canopy
and sub—canopy trees, respectively, hereafter. Species richness across plots was very low
among the dominant overstory trees, which had a total of only four red maple, sugar
maple and jack pine trees besides red pine. Similarly, there was only one basswood tree

besides red pine in the mid-canopy trees. However, the sub-canopy trees had relatively

62

higher species richness and abundance than other size classes, with a total of six
species—four jack pines, seven red pines, eight red maple, six sugar maples, and one
each of white ash and white pine. A binomial comparison revealed that there were no
significant treatment effects on the proportions of any of the three size classes versus the
remaining two classes across treatments. However, further statistical analysis with Mixed
Linear Model (Proc MIX) revealed that there was a significant treatment effect on
dominant overstory tree density (Table 2.3). The dominant overstory tree density in four-
burn plots (103 trees ha") was significantly higher than in both the unburned control and
one-burn plots. When compared with two-burn plots, the dominant overstory density in
one-bum were significantly lower. Further, when control and one-bum plots were
contrasted with two- and four-burn plots, they also showed significant difference on
dominant overstory density. Similarly, one-burn treatments were significantly different
from two- and four-burn treatments together. However, control versus all burn treatments
when contrasted did not show a significant effect. Thus, the analysis shows that there is
some difference in the density of dominant overstory trees in repeatedly burned plots,

although the effect was lesser in the two-bum than in four-burn plots.

The increase in density of dominant overstory trees in four-burn plots than in
other treatments was probably due to faster diameter growth and shifting of some of the
mid-canopy trees to dominant overstory class during the post-fire growing period. Such
shifts to larger size classes have been reported in the New Jersey pine region by
Stephenson (1965). Radial growth of western larch (Larix occidentalis) increased up to

eight years after a single burn (Reinhardt and Ryan 1988b in Dickmann 1993).

63

Table 2.3: Effects of prescribed burning on dominant overstory (>36 cm dbh) , mid-canopy

(32 to 36 cm dbh) and sub-canopy (10 to 32 cm dbh) trees1 in experimental red pine
plantation, summer 2001 (n=93).

 

 

 

 

 

 

Tree Density Basal Area Volume
Treatment effects trees ha.l m2 ha"l m3 ha-l
Dominant overstory
F-tests, p-value 0.0448 0.1208 0.1493
Chi-square tests, p-value 0.0055 0.0325 0.0726
Control 66 ac (8.4) 12.8 a (1.9) 138.5 a (19.6)
One Burn 62 a (8.6) 12.0 a (1.9) 122.5 a (19.2)
Two Burns 89 be (9.8) 16.4 a (1.9) 170.7 a (21.8)
Four Burns 103 b (10.5) 18.2 a (1.9) 194.5 a (23.3)
Contrasts
Control vs All Burns ns ns ns
C&lesZB&4B 5 ns ns
18 vs 28 & 4B 5 ns ns
Mid-canopy“
Control 27 (19.8) 3.1 (2.2) 34.3 (25.3)
One Burn 42 (20.0) 4.9 (2.2) 56.2 (25.7)
Two Burns 40 (19.8) 4.4 (2.2) 47.0 (25.3)
Four Burns 42 (19.8) 5.0 (2.2) 55.7 (25.3)
Sub-campy"
Control 235 (65.5) 3.4 (1.3) 19.8 (5.3)
One Burn 30 (67.4) 0.96 (1.4) 6.1 (5.5)
Two Burns 25 (65.5) 0.6 (1.3) 3.7 (5.3)
Four Burns 5 (65.5) 0.2 (1.3) 2.0 (5.3)
Dominant overstory tree densities are back transformed values of square root.
MAI = Mean Annual Increment
The means used above are the Least Square means (LSM).
LSM of parameters followed by the same letter are not significantly different at S 0.1.
s and ns indicate significant and non-significant respectively.
Standard errors of means (i) are in parentheses.
C, Control; 18, One Burn; 23, Two Burns; 48, Four Burns
*Comparisons were not made for mid-canopy and sub-canopy trees because of insufficient data
' Sub-canopy has approximately 25% red pine trees and the remaining are mainly red maple, sugar
maple and jack pine, whereas, dominant overstory and mid-canopy layers have red pine trees only.

 

 

There were no detectable treatment effects on basal area, volume or MAI of the

dominant overstory trees, although values for the four-bum treatments were larger (Table

64

2.3). However, chi-square tests showed that there was some relationship between fire and

the basal area of dominant overstory trees.

Unlike in the dominant overstory, sub-canopy tree density decreased due to
burning and the decrease was more prominent as the number of burns increased. Even a
single burn reduced the sub-canopy tree density drastically, agreeing with Van Wagner
(1970) who stated that young recruitrnents cannot withstand even a low intensity fire.
Consequently, the basal area, volume and MAI of sub-canopy trees in the unburned plots
were all much higher than those in the burned plots and they decreased gradually as the
frequency of burns increased. It seems that there was a lot of killing and scorching of
sub-canopy trees due to the burning. Thus, there was an enhanced overall growth of
dominant overstory red pine trees in the two- and four-burn plots due to the killing of
competing bushy vegetation and the sub-canopy trees by repeated burnings. Further, the
prescribed burning must have altered competition among the dominant overstory trees
and mid—canopy trees, resulting in greater overall growth of mature trees. The removal of
bushy vegetation and the sub-canopy trees create openings (extra-space) that reduced
competition. The overall effect of burning on larger trees is similar to thinning from
below, which helps to enhance diameter growth of fewer, larger and unifome
distributed trees that are retained during the process (Kozlowski and Peterson 1962,
Lundgren 1981, Smith 1986, Nyland 1996, Burgess and Robinson 1998, Parker et al.
2001, Smith 2003). Such volume accumulation in large—sized trees makes them more

valuable as well.

65

Higher growth in repeatedly burned plots must also be due to the increased
availability of nutrients. Nutrients such as potassium, phosphorus, calcium and
magnesium are released during burning and nitrogen fixation is enhanced. Release of
nutrients is advantageous, as nitrogen and phosphorous are usually limiting in most
forests (V itousek 1985). Marks (1974) also suggested that immediately after disturbance
nutrients are released from decomposition of organic matter and are in a state of under-
utilization. Some of the nutrients such as nitrogen and potassium may even leach during
the first few months of their release and get lost to the system (Ahlgren 1960, Debano
and Conrad 1978, Wright 1976); however, they may become available to the trees

(Ahlgren 1960) as competition gets reduced due to burning.

Sub-canopy trees were absent or very few in most of the burned plots and thus
ANOVA could not be carried out (for example, the 10-11 cm diameter class was present
only in the unburned control). No significant effect was found during binomial test of
sub-canopy trees with others using Proc Glimmix in SAS. The reduction of species
richness and abundance of the sub-canopy trees agrees with earlier studies (Dickmann
1993). Furthermore, there was no adverse effect of low intensity prescribed burnings on
the overstory red pine trees as they are tolerant of low intensity understory fires and
considered to be fire adapted because of their thick bark (V anWagner 1970). From the
tree distribution pattern (Figure 2.2a—d), it seems that among the sub-canopy trees, the 22-
27 and 27-32 cm classes might have escaped severe scorching during burning and grown
into the mid-canopy stratum during the 10-16 year gap since the last fire. Sub-canopy

trees in the 10-22 cm diameter class must have either been killed or severely scorched

66

during the low intensity burning. The species that were killed or affected mostly by the
fire were saplings of red maple, sugar maple, jack pine and white pine. In unburned plots,
overstory tree distribution was clearly a bimodal, but the distribution became increasingly
unimodal as burning frequency increased (Figure 2.2; Appendix 2.3). In the repeatedly
burned plots there were almost no sub-canopy trees at all and, thus, the mid-canopy and
dominant overstory trees formed a unimodal or bell shaped structure. A similar shift from
bimodal to unimodal was also observed for basal area, volumes and MAI of different
diameter trees (Appendix 2.3), although the overall effect on them was not significantly

different.

Therefore, my analysis reveals that low-intensity fires have a profound effect
on tree density and species richness of sub-canopy trees. Repeated fires at an interval of
two or five years essentially wiped out the sub-canopy and enhanced the diameter growth
and yield of dominant overstory trees. The enhanced diameter growth was probably due

to reduced competition and enhanced resource availability.

67

Figure 2.2: Average overstory tree density of different diameter size classes in a mature red
pine plantation (n=3).

a. Unbumed Control

 

    

150
{U
E 100
”-3
g 50
3
ZOI'I VDTDI—HEIDIDIDfiDI |
10—11 12-17 17-22 22-27 27-32 32-37 37-42 42-47 47-52 52-57 57-62
Diameter Classes (cm)
b. One Burn
«1 150
fi
if 100
«‘5
g 50
20 m r ..... m.‘- I _
10-11 12-17 17-22 22-27 27-32 32-37 37-42 42-47 47-52 52-57 57-62
Diameter Classes (cm)
'1; 150 C. Two Burns
0
it; 100
«‘3'
E 50
z 0 Mpfi
10-11 12-17 17-22 22-27 27-32 32-37 37-42 42-47 47-52 52-57 57-62
Diameter Classes (cm)
N 150 D. Four Burns
i
3;: 100
a
E 50 I
:3
Z
0 I 1 1 _ r 1 r ‘1 l‘J—T—fi

 

10-11 12—17 17-22 22-27 27-32 32-37 37-42 42-47 47-52 52-57 57-62

Diameter Classes (cm)

68

2.3.2 Ground vegetation
2.3.2.1 Species diversity

Species richness in all of the three blocks (A, B & C) tended to stabilize after
sampling about 17 to 20 one-sq m quadrats out of a total of 24 quadrats sampled for each
of the treatments in each block (Figure 2.3). Although not always universal, local species
richness is positively correlated with the area over which they are recorded (Gaston and
Lawton 1990). Total species richness of ground vegetation (all herbs, shrubs and trees
<1.4 m in height) was always higher (32 to 49 species) in four-bum plots than in others.
Conversely, the unburned plots always had the lowest species richness (16-26 species),
except in block C where both unburned and two-burn plots were almost the same (25
species). Thus, burning increased total species richness which increased gradually from
control to one- to two- to four-bum plots. All of the treatment plots in block C were
higher in ground vegetation species richness than in other blocks (Figure 2.3). Among
three blocks, block C had the lowest woody understory tree density and was thus more
open; the openness, among other reasons, might have caused more herbaceous species to
grow in this block than others. The openness in block C was also reported in earlier study
by Henning and Dickmann (1996). In my study, generally, the species richness declined
in all of the plots compared to Henning and Dickmann (1996), who measured the plots
immediately after burning in 1991. However, both Henning and Dickmann (1996) and
Neumann and Dickmann (2001) found species richness of herbaceous and woody ground
flora (<1m in height) to be greater in periodically or once-burned areas compared to

unburned areas.

69

Figure 2. 3: Species area curves by blocks for herbaceous ground vegetation in the 70-year-old
red pine plantation under burning treatments.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

/
BlockA
5 50
E
a 40 7~A 77* Vii 7, 7 , ,
3
g 30-
%
g 20
"E
'3 10
E
:3
U 01"IITTIIl‘I1IT‘IIT'I'I'III'I'I‘I'I‘IIITI‘I
1 3 5 7 9 11 13 15 17 19 21 23
Total sample plots
K —X— One burn +Four burns +Two burns +Control
/
BlockB
50 —
§
0
é a
0
e
g E
E
:1
U
O11'1‘|11'1'1'1lI'I'I'T'TlllI'I'I‘I'I‘I'llllllfrj
1 3 5 7 9 11 13 15 17 19 21 23
Total sample plots
L [—I— Four burns +Two burns —X— One burn +Control]
/
BlockC
50 —
:3; 40 _
:2 30-
.2 E 20
o—v 3 ..
g r.-
E 10 ~
0
0 . r . . . .
1 3 5 7 9 11 13 15 17 19 21 23
Total sample plots
K l-I—Four burns —X-- One bum +Two burns +ControlJ

 

 

Note: only 15 quadrats were used for one burn plot block A.

70

There were no significant effects of treatments on species richness on a per

sample plot basis. However, the F-tests and chi-square tests showed significant treatment

effect on Simpson’s Index, Simpson’s reciprocal and Shannon’s Indices for ground

vegetation (Table 2.4). Simpsons’ reciprocal and Shanon’s indices were significantly

lower in unburned controls whereas Simpson’s diversity was significantly higher in

unburned controls than in four-bum treatments. Simpson and Shanon’s indices in

unburned controls were all significantly different from one-bum treatments as well.

These differences indicate that unburned controls had lower diversity of ground flora than

four-times burned plots. The higher Simpson’s diversity index in unburned controls

Table 2.4: Species diversity (per sample plot) of ground vegetation (<1.4 m height) in experimental

red pine plantation, 2003 (n=12).

 

Indices of Diversity

 

 

Treatments Species richness Simpson Simpsons' Shanon Evennessl
S D Reciprocal, l/D H‘ J
Treatment Effects
p>F-va1ue 0.2544 0.0725 0.0983 0.081
p>Chi square 0.1524 0.0082 0.0189 0.01 12
Control 2.3 a (0.9) 0.6 a (0.1) 2.3 a (0.3) 0.8 a (0.1) 0.7 (0.1)
One Burn 4.7 a (1.0) 0.4 b (0.1) 2.8 ab (0.3) 1.1 b (0.1) 0.7 (0.1)
Two Burn 6.7 a (0.9) 0.5 ab (0.1) 2.6 ab (0.3) 1.0 ab (0.1) 0.6 (0.1)
Four Burn 2.3 150.9) 0.4 b (0.1) 3.1 b (0.3) 1.2 b (0.1) 0.7 (0.1)
Contrasts
Control vs All Burns s s s
C&les2B&4B ns ns 5
les23&4B ns ns ns

 

 

F-tests p-values are given along with the chi-square values as treatment effects.

C, Control; 18, One Burn; 28, Two Burns; 43, Four Burns

Values in the columns are the Least Square Means. Values with the same letter are not
significantly different from one another at p S 0.1.

s and ns signifies significance and non-significance, respectively, at p S 0.1.

Standard error of means (2t) are in parenthesis.

I . . .
Evenness index was not analyzed as it was not normal even afier transformations

 

71

 

suggests that the unburned plots had dominance (or abundance) of fewer species than in
burned treatments. Evenness indices were not analyzed as they were not normal but their
values were almost the same in all of the treatments. There was no difference in species
diversity among the burned treatments. Unbumed controls were significantly different
than all of the burned plots combined for all of the Simpson, Simpsons’s reciprocal and
Shanon’s indices. Unbumed controls and one-burn treatments when contrasted with two-
and four-burn plots were significantly different for Shanon Indices whereas similar
contrasts for Simpson and Simpson’s Reciprocal were not different. Control, one- and
four-bum plots had high evenness (0.7), which means that 70% of the species had the
same number of individuals in them. Two-bum plots had a somewhat lower evenness

(0.6), suggesting that 60% of the species had the same number of individuals in them.

2.3.2.2 Percent cover

The response of ground vegetation to the burn treatments varied and
depended on the type or form of the vegetation (tree, herb or shrub species). The overall
average percent cover of ground vegetation increased fi'om unburned control (27%) to
burn (34 to 36 %) plots; although, there was no significant treatment effect (Tables 2.5
and 2.6). Similarly, Henning and Dickmann (1996) detected no differences in total
percent cover in measurement of this same study in 1991. l categorized and analyzed the
ground vegetation in two groups: 1) trees and 2) herbs and shrubs. A binomial test
comparing the two groups shows that there was no significant difference between the

proportions of

72

 

380mm

=e Eat 35:28 83 B>8 :33 05 3223 oES £5 E 35:5 soon 02“: .x. m _.e 323 530 owns: wiifi 36on =< ”802
.35 £88 hoe mfiofieoam E 2m 3e £82: we hobo Ewe—“Sm
4.er «a 8505 28 Sec Engage begoEwEm 8: one $58 wEeeommotoQ 05 wee—m
$532 253. 05 6:5 mos—g 05 e5 Emd :32 osscm “was testament x03 ammo; 2: 8w 5>oo owfio>< 13o h “8 8:15 2:.
.530 033—8” .>oU mom £38 owflo>w on“ 2 u.>oO.w><

 

 

 

 

 

 

 

 

 

 

.Aoemnscousag 05 e2 unease

 

3.8 a 8 See a 3m $.on a 3m Se a m: as: a. 32% .35:
Gem 3 2.33% sea 2 2.3“: 3:38;
:58 owfio>< .88.
we N.e EzeEeéAEmm. know We N.e mam $3.65:me

e; We mam LmEutngV We N.e 3:233. 255% he Ne 93»: magnum We N.e £3.55 Leo—x
e._ We EBEENAEMRN Sow. We N.e mam BESEEEV e.e We SENSE? 993.0 we N.e etaotmfia :55
N._ we 3:: axocmam We N.e SSNEmEa $.55 ~.— We EESEQE. Lee—x e.e N.e Expieébfmk Bow
m._ We Eiexuuen know ee me 565%; SEE W_ We areutmfie “SEHEK W_ we 3:553? .323an
e; We £3.55 L8... v; We c2553? 35sz YN e.e ethane“. 355% m: we 53.5503. goo—x
5m W— entoxwm “352% W_ We 3:31.053 §EB¢~ 5N e._ E335 SUV NM— ve eteuimfiu “Student
ed h; erg—ewe: 3.52% 9N e.e 553%an xmoV Wm m._ 6:553? QEEQ e; We mam xmiotEmEV
v.” QN 305.3: SEE e. N. WN SEQ: but ma v; 36:33 SEE WN We 56:23 wast

500 .>o0 .>oo .>oU .>oU .>oU .>oU >00
43. .w>< 36on mod .w>< 860mm ._oM .m>< 88on .Em .w>< 863m

25m Sou gm 9:. E=m 0:0 6250

 

E e: v. E 535%? 95on bounce—5 05 E 52o wedeeoomoe E 02a 2953 83 805 we S>8 03:22 was .38 owfiu>< ”WN 2an

73

 

360%

:a 9:3 353mg was b>oo :38 2t 320:3 038 25 E 838$“. :03 05; .x. 2 .e 323 :38 owns; 9:32 860% :< ”802
u..w:mfi3::u E=ENS::.EV< - new 53:65:95: ,

.bco £88 Sm 38553:“ E was A3 23:. mo Stu nae—sum ,

.fieWm 8 5:85 25 Eob “BEBE 328536 8: 05 $58 wEvEEmotoQ on» mac?

wrote. 053 05 5:: $3? 2: can CZmd :82 Bang ~33 unaccemnub M23 swag? 05 Ba B>oo uwfio>< 38% :8 mos—g 2F
S>oo gnu—2” .>oU god £950 awash,“ 05 m_ n.>oU.m><

 

 

 

 

 

 

 

 

 

8.3 a om 86: a ”.2 $3: a Sm 8.3 a 3m 85 a. 32% .35:
Gem EN :6“ EN 3:32 2.32.2 :8 33% 2295:
:98 owfio>< 130%
We Ne mam 55:80::
We Ne ::E:.~M:.§ EEwEK
he Ne EtnSNm SEAR we We 3:553? SEMEE
oe We v.83»; Sinemo§$u ee We 93ng Eaiomo§xb We me 333:: E:.€e&o:m:b
2 2 Q; 5:332: I to 8&5» SEE 3 do gm omanm
.1 We 3:30: EEFxé n; We “583.5. .3qu m._ We mam 5:33me
m; We “Swami: 3:an YN We 33.8: 35:8.fi W“ We 6:553? EEMEK
e.N he 5:05 e.~ : mam $3.932: Nu— We 9.23.6: 38:92: We Ne 55:
Wm e._ figmfiyfifitu 353% _.w M: 33:83.: ENE—x Wm e._ .2? :S:MEEM N; me new 52.532:
m6 N.N 8%ka N6 Nd 8320 Wu 5N mam Exsmifivsxx Wm We 338.83: 52::
we Wm naaaofix: SEQ. v6 Wm mam 5:53:33: m.» e.m 33:83:: 23:: Wm W_ 3320
~23 ad 65. EEMSEW fie m.m figmsgfimza “SSE W: 3. 3.83:: 253% WE Ev mam Ezsm£§e§<
:2 an 25533 553:2: 2: 3 .am Eaxwuim 92 an 335 SN 3 .am sfimfinm,
We. «am 9% Eafiwigfi: hem v.9 =53..ng SENENE v.3 06 53:23? 553.225: Nem We £55239» EEE§E
Axe exp A5 exp 3.; 3.; Se axe
.>oU .>oU $00 .>oU $00 .>oU .>oU .>oU
Aux .m>< $6on 43— .m>< 36on 43” .w>< $6on .31 .w>< museum
8.5m Sow mEzm 03% gm 0:0 85:00

 

 

 

 

.Ampmncv cots—.2 2: v2 822: a E E vgvv
:ofifiow? v.58» begets: 05 E EEO wcgcoomuu E 92a 29:8 home 33% 32 EB 35: mo 830 3:22 can owflo>< Wm 2an

74

trees or herbs and shrubs across the treatments. Further statistical analysis (F -tests)
showed that there were no significant differences among treatments for either of the two
groups (Tables 2.5 and 2.6). Nonetheless, for both groups the average coverage tended to

be higher in all of the burned plots than in the control.

Thus, despite no significant differences, it is apparent that there are some
distinct effects of burning on ground vegetation coverage. First, all the burn treatments
had higher overall percent cover than the control. Second, the most frequently burned
plots had lower overall species coverage than the least frequently burned plots. Third,
burned plots had relatively more tree coverage than unburned plots, and among the
burned plots the highest was in the two-burn plots and the lowest in the four-bum plots.
Such increase in woody species in burned plots was suggested by Ahlgren (1960). Once-
bumed plots also were found to contain the greatest composition of woody ground flora
than periodically burned plots by Neumann and Dickmann (2001) who measured the
following season after burning. Fourth, herb and shrub coverage was the lowest in
unburned control and generally decreased gradually as the number of burns increased. In
central Appalachian forests, herbaceous species were found to play an important role in
the recruitment, establishment and development of woody seedlings, which are always

more as the stand becomes mature (Gilliam et al. 1995).

2.3.2.3 Tree species

Red pine, red maple, juneberry (Amelanchier spps.), choke cherry (Prunus

virginiana), sugar maple, white ash and black cherry (Prunus serotina) were the seven

75

tree species that were most abundant in all of the treatments (Table 2.5). None of these
species were statistically different in their coverage across treatments. However, red
maple, black cherry, choke cherry, red pine and white ash showed substantial differences
over control in their coverage after burning. Differences in post-fire establishment of red
maple and black and choke cherry densities were also reported earlier (Henning and
Dickmann 1996). These species were most favored by repeated fires in my site. Increase
in these species improves wildlife habitat as forage and soft mast availability increases
(Rouse 1988, Bender et al. 1997, Dickmann 1993). Black cherry had been found to
densely seed in the openings after 1938 Hurricane in New England (Spurr 1956). The
seeds of black and choke cherries can remain dormant in soil for long periods (Canham
and Marks 1985) and they sprout vigorously from the root collar, which makes them
aggressive post fire colonizers (Henning and Dickmann 1996). Despite the increase in
tree species in the burned plots, the species coverage in our study was lower when
compared with that of Henning and Dickmann (1996). Such a decrease of tree seedlings
with time since disturbance was also found in Pennsylvannia by Peterson and Pickett
(1995). Likewise, in a boreal system in the Yukon in British Columbia, Canada, new

regeneration occurred only in first five years after fire (Johnstone et al. 2004).

Red pine and red maple were the two most abundant tree species found in the
ground vegetation for all the treatments. Red pine, a species considered to be shade
intolerant, had the highest relative cover in four-bum plots, 334% higher than in the
control. Apparently repeated burning prepared the seedbed by opening up space and

exposing mineral soil, allowing seeds to germinate and seedlings to establish. Prescribed

76

burning improves seedbed for pines (Buell and Cantlon 1953). Red pine is highly adapted
to such fire-prepared seed beds (Ahlgren 1976, VanWagner 1970). Also, since good seed
years in red pine are every 5-10 years (VanWagner 1970, Rouse 1988), the four-bum
plots provided a much longer window of optimum seedbed conditions. In repeatedly
burned plots competing vegetation is killed or at least suppressed (Alban 1977, Metheven
and Murray 1974, Van Wagner 1970) increasing the chance that red pine seedlings will
become established. Henning and Dickmann (1996) found that red pine seedlings
occurred less commonly in four-bum plots; however, they measured the plots
immediately after the last burn. Therefore, my re-measurement showed that the 10—years
gap since the last fire provided an appropriate environment for further recruitment of red
pine seedlings in the ground vegetation layer. My results also show that shade tolerance
is a relative concept; even in a shaded understory, an “intolerant” species like red pine

can recruit if seedbeds are optimum and competition is suppressed.

Red maple, a moderately tolerant species with long lasting dormant seeds, had
higher coverage in burned plots than in controls (Table 2.5). Sugar maple, the third
highest ranking and a very shade tolerant species, showed highest coverage in the two-

burn plots. Juneberry and bass wood coverage was less in the burned plots.

Red oak (Quercus rubra), black oak (Quercus velutina), white oak (Quercus
alba), and American elm (Ulmus americana) were all absent in unburned plots but
showed up in at least one burned plot. The relatively tolerant species American beech

(F agus grandifolia) and shade intolerant red oaks were very minor, with a relative cover

77

less than 0.1 percent. White pine and American elm also showed up in the burned plots
even though their average coverage was very low. The sub-plot in block A which was
burned once with an intense, hot fire causing complete overstory mortality were omitted
from analysis in this study. They had mostly American elm, red raspberry and blackberry

(Rubus spps.) in the ground vegetation layer.

2.3.2.4 Herbs and low shrubs

Bracken fern (Pteridium aquilinium), moss (Sphagnum spps.), wild—lily-of the
valley (Mianthemum canadense), red raspberry, blackberry, sarsaparilla (Aralia
nudicaulis) and grasses were the herbs and shrubs that occupied the top five positions or
ranks based on their average percent cover and relative cover for all of the treatments
(Table 2.6). Shrubs are killed back and are reduced in number and biomass (Little and
Moore 1949). The overall species coverage was not significantly different across
treatments and was consistent with the results of Henning and Dickmann (1996).
However, in general, the herbaceous layer increases with increase in fire frequency (Buell
and Cantlon 1953). There was an increase of herb and shrub species richness with a cover
>O.15% from unburned (7) to once burned (11) to twice burned (12) to burned four times
(14). Henning and Dickmann (1996) and Neumann and Dickmann (2001) also found that
the species richness of herbaceous and woody ground flora (<1m in height) was greater in
periodically or once-burned compared to unburned areas. Increased soil moisture and
nutrients due to burning may stimulate some herbaceous plants, while increases in
temperature in gaps that form may affect both herb growth and production (Collins et al.

1985). Chabot (1978) found variation in light to produce greatest change in growth of

78

strawberry plants as compared to other enviromnental factors. Growth of strawberries
increased with an increase in light intensity, with a daily fluctuation of 10 0C, in gap

edges. However, Bakelaar and Odum (1978), in an nutrient perturbation (added NPK
fertilizer) experiment, found that a few species dominated others, which decreased in

their importance value and thus decreased diversity or niche width.

The highest relative cover in my study was of bracken fern, which ranked top
in all of the treatments except in the four-bum plots where wild-lily—of the valley had
almost the same percent cover (Table 2.6). Bracken fern showed the least cover in the
four-burn plots. Ahlgren (1960) found that bracken fern was more abundant in burned
areas than in unburned areas. But it was found to have higher cover values in less
frequently burned plots (Buell and Cantlon 1953). Bracken fern was found both in control
and older burn plots but was highest in the most recent burns in New Jersey Pine Barrens
(Boemer 1981). The rhizomes of bracken ferns are vigorous and sprout back after fire; its
spores also germinate better in burned soil (Olinonen 1967a and 1967b in Henning and
Dickmann 1996, Henning and Dickmann 1996). The decrease of bracken fern cover in
my four-burn plots could be due to the repeated fires every two years or the increase in
competing species. Besides killing unwanted vegetation, repeated fires also create more
openness; bracken ferns are also susceptible to frost killing in more open conditions

(Cody and Crompton 1975 in Abella et al., 2004).

Generally, the moss and lichen decreased as the frequency of burning

increased, and this result is consistent with that of Henning and Dickmann (1996).

79

However, it is not consistent with findings of Buell and Cantlon (1953) in New Jersey

pine region in which burning increased moss and lichen.

Rubus species coverage was negligible or absent in unburned control plots but
higher in burned plots, especially those btu'ned once or twice (Table 2.6). F-tests and Chi-
square tests revealed that the number or count per square meter of red raspberry and red
pine showed significant treatment effects (Table 2.7). Red raspberry was significantly
lower in unburned controls than in four-burn plots, which were also higher than one-bum
plots. Red pine counts were significantly higher in two-bum than in other burned and
unburned plots. Both species in unburned control and one-burn treatments when
contrasted with two- and four-burn treatments were significantly different. This shows
that red pine regeneration is low in unburned controls and also four-bum plots because
repeated fires every two years kill already regenerated and established red pine seedlings.
Increases with burning were also observed by Henning & Dickmann (1996) for Rubus
species in an earlier analysis of this study. Vigorous growth of Rubus spp. (after burning)
in northeastern Minnesota was reported by Ahlgren (1960).The ability of Rubus seeds to
remain in soil for a long period before they germinate (Marks 1974) and their sprouting
and heavy fruiting after fire (Abrahamson 1984, Ahlgren 1979), were two main reasons
for their success and increased growth in the burned sites. In northern hardwoods, pin
cherry (Prunus pensylvam’cum), a colonizing and exploitative species replaced Rubus

spp. after three years from seeds that existed in the soil (Marks 1974; Bormann and

Likens 1979).

80

Other species that were favored by fire were grasses, sarsaparilla and shinleaf,
which did not show treatment effects. Stimulatory effects of burns on grass cover have
also been reported by others (Buell and Cantlon 1953, Ahlgren 1979, White 1983, Howe
1995, Henning and Dickmann 1996). In tall prairies, vigorous clones of the dominant
grass (Andropogon gerardii), dominant forb (Solidago altissima) and a less common forb

(Aster simplex) increased steadily in all burn treatments (Howe 1995).

Common strawberry (F ragaria virginiana), fragrant bedstraw (Galium
triflorum), wild lettuce (Lactuca canadense), shin leaf (Pyrola elliptica), common wood
sorrel (Rumex aeetosella), lady’s slipper (Cypripedium spps.), daisy fleabane (Erigeron
spps.), cow wheat (Melampyrum lineare), wild or prickly gooseberry (Ribes cynosbatti),
starflower (Trientalis borealis), Achillea millefolium, carrot (Daucus carrota),
Mushroom, shin leaf (Pyrola rotundifolia), Lonicera, and wild rose (Rosa spps) were
very scarce or absent in unburned controls but found in all or at least in one or more types
of burns. Common dandelion (T araxacum oflicinale), Solidago spps, Aster sagittifolius,

and Chenopodium vulgare were among the other species found in very small numbers.

Species that were present only in the four times (repeatedly) burned plots
were staghom sumac (Rhus typhina), Solidago spps., blueberry (Vaccinium spps), bunch
berry (Camus canadensis), pin or fire cherry (Prunus pensylvanica) and clammy
cudweed (Gnaphalium macounii Greene). If disturbance frequency or intensity is high,

communities may have greater numbers of fire adapted species. Without disturbances

81

 

 

mEsm Son— .mv ”2.5m 03:. .mm 6:5 0:0 .9 28:50 .0
._.o w a a: £33838: .858E:w_m-:o: U5 858535 moc_:w_m m: :5 m
“.o w a E “55$: m::8c_:m_m Ho: 80>» 2m;
.82"; UoE:£m:mb xuwn-wo_ 2: 8: 3E: ou:onc:ou 2: :5 32%: 232 083cm 284 2: Br. 82:5 2:.

£55 85280 “d

 

 

 

 

 

 

 

 

 

 

 

m m: 91:21:
m m mvemmsm: do
w: m: 35m =< m> 65:00
oEm uom gonmmw: cox 322.50

we...“ :3 31 Rod 3.? f: 53 fit: «2 :3 8S? 2: 28: 25.0 2&8:

£3 0:: ZN am; :8 2:3..— wwfi $33 22 23.. was: a: 55.0 NS: 93:33;

Sum: 533 3:8 Emu: .533 3:8 him: $30: “:38 bum: .0304 E28 02:55 mfluomm

d :82 so :82 ,6 :82 so :82 as: 3-368
wEDm EDGE mph—3m OBP Ehsm 0:0 Wyn—=00 «unto uEDEumob :NLO>O

 

 

 

.N_n: .ANEV SE E2585 :um 860% :ozfiow? n::o:w 95 mo 53:5: of Co mix—mix EN 033.

82

these species decline. For example, in northern hardwoods, large scale disturbance
dependent species like pin cherry may be locally extirpated except for seeds in soil

(Marks 1974).

Partridge berry (Mitchelia repens) was scarce and found only in unburned
control and absent in other burn treatments. Moss, false Solomon seal (Smilacina spps.),
wild fern (Dryopteris carthusiana) and lycopodium generally decreased with burning.
Both moss and lichen were found to decrease in burned plots in their average and relative
percent cover and a similar decrease for these species was reported by Henning and

Dickmann (1996).

2.3.3 Understory trees (>1.4 m height and < 10 cm dbh)

The understory trees were mainly composed of red pine, red maple, beech, black
cherry, white ash, juneberry and sugar maple (relative frequencies 47, 16, 9, 9, 4, 3 and 3
percent). Remaining species — choke cherry, eastern hophombeam (Ostrya virginiana,
white pines, and others were — each were less than three percent of total. The binomial
comparison of proportions carried out between two groups, i.e., each of the understory
woody species versus the remaining species grouped together, did not show any

significant treatment effects.
2.3.3.1 Understory density

The density of understory woody species of all size classes combined together

ranged from 2738 to 6377 trees ha"1 and the lowest was in unburned and the highest in

83

two-bum plots (Table 2.8). There was no significant treatment effect on the overall stem
density of understory woody vegetation; however, there were some remarkable increases
in densities across different burns. For example, the overall density for two, one and four-
burn plots increased over control by about 133, 124 and 61%, respectively. The increase
in the overall density of understory woody vegetation in all of the burn plots was due to
post-fire regeneration and establishment, and the magnitude of this effect varied with the
number of burns. The lack of fire for a long period (10 to 18 years) allowed seeds to
germinate and root suckers and stump sprouts to grow. The changes brought about in the
understory woody species composition by burning are visually prominent among the
treatments (Figure 2.4). Similarly, understory trees also increased by 52% after burning

in loblolly pine in the Piedmont, North Carolina (Oosting 1944).

Understory tree density was negatively correlated with parameters such as
overstory tree density (n = 93; r = -0.2314; p = 0.0256); basal area (11 = 93; F -0.3614; p
= 0.0004) and volume and MAI (n = 93; r = -0.28605; p = 0.0053). A high overstory tree
density certainly will cause a decrease in understory woody density, but fire, especially if

repeated, has a much more profound negative effect on the woody understory.

2.3.3.2 Diameter and age

Treatments had a significant effect on the average diameter at breast height (dbh)
of understory woody trees (>1.4m height and < 10cm dbh) (Table 2.9). Average dbh
ranged from 0.9 to 3.0 cm, with the lowest in four-burn plots and the highest in the

unburned plots. All of the burned plots had significantly lower average diameter than

84

Table 2.8: Woody understory stem density (trees ha!) and relative density in descending
order under different size classes

 

 

 

 

 

 

 

 

 

 

 

 

 

Treatment Sps 0-4 % of Sps 4.1-6 % of Sps 6.1-10 % of Sps Total % of
Code cm total Code cm total Code cm total Code total
Contro RP 87 33 RM 158 49 RM 200 60 RM 746 27
RM 388 19 SM 47 14 SM 47 14 RP 704 26
SM 267 13 AB 42 13 WA 30 9 SM 358 13
BC 168 8 WA 35 10 HH 20 6 AB 208 8
AB 151 7 WP 25 8 AB 17 5 WA 179 7
WA 1 16 6 RP 12 4 WP 8 3 BC 167 6
AM 104 5 HH 4 1 RP 4 1 AM 104 4
CC 72 3 Oaks 4 1 Others 4 1 CC 71 3
Others 44 2 BC 0 0 Oaks 4 1 WP 71 3
WP 37 2 AM 0 0 BC 0 0 HH 58 2
HH 35 2 CC 0 0 AM 0 0 Others 50 2
Oaks 12 1 Others 0 0 CC 0 0 Oaks 21 1
Total 2080 ($577.7) 325 (i388) 334 (i371) 2738
On'e'Burn RP 2868 48 RP 48 BC 19 50 RP 2915 4‘81
RM 795 13 AB 38 24 WA 10 25 RM 838 14
AB 714 12 RM 38 24 RM 5 13 AB 753 12
BC 477 8 BC 14 9 SM 5 13 BC 510 8
WA 324 5 Oaks 10 6 AB 0 0 WA 33 8 6
AM 296 5 CC 5 3 AM 0 0 AM 295 5
CC 200 3 WA 3 CC 0 0 CC 205 3
Others 1 14 2 AM 0 0 HH 0 0 Others 1 l4 2
SM 86 1 HH 0 0 Oaks 0 0 SM 91 1
Oaks 47 1 Others 0 0 Others 0 0 Oaks 57 1
HH 20 0 SM 0 0 0 0 HH 19 0
WP 0 0 WP 0 0 0 0 WP 0 0
Total 5940 (i632.8) 157 (i388) 38 @388) 6135
Two Burns RP 3418 54 Oaks 21 45 HH 8 40 RP 3430 54
RM 884 14 Others 8 18 Oaks 4 20 RM 884 14
AB 604 10 RP 8 18 RP 4 20 AB 604 9
BC 588 9 HH 4 9 WA 4 20 BC 588 9
AM 154 2 WA 4 9 AB 0 0 Oaks 171 3
WA 150 2 AB 0 0 AM 0 0 WA 158 2
Oaks 146 2 AM 0 0 BC 0 0 AM 154 2
SM 104 2 BC 0 0 CC 0 0 Others 154 2
Others 88 1 CC 0 0 Others 0 0 SM 104 2
CC 46 1 RM 0 0 RM 0 0 HH 58 1
HH 46 1 SM 0 0 SM 0 0 CC 46 1
WP 25 0 WP 0 0 WP O 0 WP 25 0
Total 6310 (i577.7) 46 (:38 8) 20 (i373) 6377
Four Burns RP 2305 53 Oaks 21 50 HH 4 50 RP 2322 53
RM 709 16 RP 17 40 WA 4 50 RM 709 16
BC 471 11 HH 4 10 AB 0 0 BC 471 11
AB 217 5 AB 0 0 AM 0 0 AB 217 5
HH 1 13 3 AM 0 0 BC 0 0 HH 121 3
Others 158 4 BC 0 0 CC 0 0 Others 158 4
AM 96 2 CC 0 0 Oaks 0 0 WA 100 2
WA 96 2 Others 0 0 Others 0 0 AM 96 2
CC 75 2 RM 0 0 RM 0 O Oaks 83 2
Oaks 63 1 SM 0 0 RP 0 0 CC 75 2
SM 33 1 WA 0 0 SM 0 0 SM 33 1
WP 17 WP 0 0 WP 0 0 WP 17 0
Total 4352 @6328) 42 (i388) 7 (i373) 4402
[Standard errors of mean are given in paranthesis (i).
Note: The common and the Latin names ofthe species codes are explained in Appendix-2.1

 

85

 

     
   

u . . I... a . .. . . 3...... .
A33» 0_ “mum of :8 EB 05 A 28% _ _ “man 05

:8 EB 05
32-me .255 So”— .u

092-me .mEnn 95% .o

 
 

A25» 2 an: 65 :8 E3 05
mmm. .:.=5 0:0 .3

35:8 853:3 .m

damn—32D in 3 88:3 .08 SEE; 50:95:: 65: to:

20.3%-? a E 35:53: 3E3¢=£ «E: -93 :80 v5 BEBE. :_ 59:26:: 05 E Stat? 36:50:85“ ”in 2sz

86

the control; however, there were no significant differences among different burns in their
average diameters. Also, when contrasted unburned control versus burned plots, control
and one-burn versus repeatedly burned plots, and one—bum versus repeatedly burned plots
were all significantly different. The overall analysis revealed that multiple burning
lowered the average diameter of understory woody stems significantly. The mean
understory woody diameter was positively correlated with the overstory tree density (n =
93; r = 0.4295; p = 0.0001) because the advanced large-sized understory woody
regeneration (seedlings and saplings) in the unburned plots were killed by the fires in the
burned plots. However, the average understory woody diameter was negatively correlated
with other parameters of the overstory trees: heights (n = 93; r = -0.2568; p = 0.013),
mean diameter (n= 93; r = -0.21825; p = 0.0356), and basal area (n = 93; r = -0.23027; p
= 0.0264). Also the new post-fire regeneration was very young and small in average

diameter.

Treatment effects were significant only on the average diameter of beech and red
maple (Table 2.9); the more frequent the fires, the smaller the diameter of those two
species. These two species, which comprise most of the large-size class understory trees,
were reduced the most in the repeatedly (two and four) burned plots. DBH of both red
maple and American beech in one-burn plots also were significantly different from four-
burned plots. In beech DBH was significantly higher in one-burn than in two-burn plots.
In case of red maple, DBH in one-burn treatments was significantly lower than in

unburned control. Contrast of unburned control versus burn plots, one-bum and unburned

87

Table 2.9: Mean Diameter at breast height (dbh) of understory woody vegetation
in experimental red pine plantation, 2003. (n=93; All species df=6, Beech dfi5.81,
Red maple df=8.01).

 

All species Beech Red Maple Red pine

 

dbh dbh dbh dbh
cm cm cm cm
F-test, p-value 0.0142 0.0257 0.0069 ns
Chi-square value 0.0002 0.0002 0.0001 ns
Treatments
Control 3.0 a (0.6) 4.1 a (0.4) 5.3 a (0.6) 0.33 (0.13)
One Burn 1.2 b (0.2) 4.3 a (0.4) 3.8 b (0.4) 0.15 (0.1)
Two Burns 1.7 b (0.3) 3.5 b (0.3) 3.0 be (0.3) 0.26 (0.1)
Four Burns 09 b (0.2) 3.0 b (0.3) 2.8 c (0.3) 0.21 (0.1)
Contrast
Control vs All Burns 5 s 5
4B & 28 vs 18 & Control 5 s 5

18 vs 28 & 4B 5 s s
dbh is back transformed values of Natural log, for which the Least Square Means
(LSM) are given above, except for red pine.

Values with the same letter/s not significantly different from one another at p S 0.1.
s and ns indicate significance and non-significance, respectively, at p S 0.1.
Standard errors of means (i) are in parenthesis.

18, 2B & 4B are one-, two- & four-burn treatments

Differences among treatments for average diameter of other species were non-
significant

 

 

 

plots versus the two- and four-bum plots together, and one- with two- and four-bum

treatments together showed significant differences for both species. This also indicates

that two— and four-bum lower DBH more than one-burn. There were differences seen in

other species across the treatments also, but they were not significantly different.

The average age of understory woody vegetation, determined by counting rings of

red pine, red maple, black cherry, white ash and beech cross-sections was 13.0 years

(S.E. i 2.69) in the burned plots and 15.0 years (S.E. i 5.3) in the unburned plots, which

88

shows that the new regeneration in the burned plots was established right after the fire
i.e., within the past 10 to 18 years and had smaller average diameter than in control
(Table 2.10). Average diameter and age of all the species were correlated (n = 218; r =
0.6767; p = 0.0001). Only beech was >18 years in unburned controls and most variable in
age among others. All of the species were older and of larger-diameter in unburned
controls than in burned treatments. Of the total of 21 8 cross sections studied, 192 were of

0-4 cm and the remaining 25 of 4.1 to 6.0 cm and only one of 6.1-10 cm dbh.

2.3.3.3 Density by size class

Understory trees of different diameter classes showed different responses to the
prescribed burns (Table 2.8, Figures 2.5a and b). The frequency of the 0-4, 4.1-6.0 and
6.1-10 cm diameter classes were 95, 3 and 2%, respectively. Seedlings up to 4 cm dbh
were extremely dense in all of the plots compared to the other two larger diameter classes
(Figure 2.5), with unburned plots having the lowest and two-burn plots having the highest
density. In contrast, understory tree density of 4.1 to 6.0 cm and 6.1 to 10 cm diameter

saplings decreased sharply as the frequency of burns increased (Table 2.8; Figure 2.5b).

The largest (6.1 to 10 cm) diameter understory trees had the lowest density and
the highest decrease over control in the four-burn plots. As the response of different
diameter seedlings to fire was different, for convenience, I created several new groups of
diameter classes and analyzed binomial proportions. The understory woody stems were
divided into two groups: large diameter (6.1-10 cm) and remaining others (0—6.0 cm); the

second grouping is 2.1 to 10 cm and 0—2.0 cm. The frequency analysis showed that the

89

Table 2.10Aggmd diameter of understory woody seedlings in experimental red pine plantation, 2004

 

 

 

 

 

 

Species Treatments Number Mean (iStandard Error) Correlation
of cross- DBH iSE Age iSE Age by DBH Correlation
sections years p-value <r Coefiicient r
Beech Control 12 1.28 0.7 18.6 7 0.0004 0.85
All burns 41 0.85 0.5 14.2 2.9 0.0010 0.71
Black cherry Control 9 0.53 0.3 13.55 4.8 0.0260 0.54
All burns 25 0.52 0.3 11.7 1.8 0.0060 0.53
Red maple Control 16 0.97 0.6 13.93 2.2 0.0010 0.85
A11 burns 36 0.74 0.5 13.8 3 0.0001 0.88
Red pine Control 14 0.91 0.7 14.8 4.2 0.0009 0.78
All burns 47 1 0.6 12.13 2.2 0.0001 0.61
ALL Total Control 58 0.92 0.7 15.1 5.3 0.0001 0.78
All burns 160 0.79 0.5 13 2.7 0.0001 0.62
Note: Species (ALL Total) also includes white ash which had a total of only 18 cross-sections.

 

large-sized (6.1-10 cm) understory trees had a relative density of 2% whereas the smaller
seedlings (0-6.0 cm) were 98% of the total. Binomial comparisons for the proportions of
both of the above size classes were tested using General Linear Mixed Model (Proc
GLIMMIX). There was a significant treatment effect (11 = 4530; df = 6; p = 0.0157) on
the proportions of the large (6.1 to 10 cm) and others. Also, the control was significantly
different from all of the bums; however, there were no proportional differences among
any of the three different burns. Another similar binomial test between the small (0-2.0
cm) and others (2.1-10.0 cm) also showed that there was a significant treatment effect of
burning on the proportions of the diameter classes (n = 4530; df = 6; p = 0.0383);
however, no significant differences were detected among any of the four treatments.

F Luther analyses also showed that the differences between unburned control and four-

burn plots for both large and small understory trees were greater than one- and both

90

Figure 2.5a : Density of understory woody vegetation in
the 0-4 cm size class under different burn treatments

8000 ~

7000 -

6000 —

Number of trees per ha
in A 'J!
O O O
O O O
O O O
l I

2000 ~

1000 4

 

with standard error bards (n=93)

   
  
  

6311

0-4 cm

Diameter size class

 

 

1:] Control
B One Burn
3 Two Bums

I Four Burns

 

Figure 2.5b: Density of understory woody vegetation in
two size classes under different burn treatments with

400 -

350 ~

300 -

Number of trees per ha
-- N N
UI O U1
0 O O
l | l

100 :

 

standard error bars (n=93)

325

   
  
  
  

 

 

 

4.1-6 cm 6.1-10 cm

Diameter size classes

91

 

 

Cl Control
E1 One Burn
E Two Burns

I Four Burns

 

 

 

frequently burned plots and two and four-bum plots. Thus, both binomial tests suggest
that the burning in the red pine stand substantially altered the structure of the woody

understory that was evident even 10 to 18 years after the prescribed fires.

Since the above proportions were significant, further analysis was done. The chi-
square tests showed significant effects of burning on small (0-2 cm), medium (2.1-6.0
cm) and large (6.1-10 cm) size classes of understory trees (Table 2.11). However, Mixed
Linear Model (F-tests) for plot level means showed significant treatment effects only on
the density of small (002 cm) and large diameter (6.1-10 em) but not the medium
understory trees. Small diameter understory trees were significantly lower in unburned
controls and four-bum than in one- and two-burn treatments. Further, the contrast
comparing the control against all of the bums was significantly different for density of
small-sized understory trees. Unlike small-sized understory trees, large-sized understory
trees in unburned controls were significantly higher than in burned treatments; however,
one-bum and two-burns and also two-btuns and four-bums were not significantly
different from each other. Further, the contrasts comparing the unburend control against
all of the bums, one-bum and unburned controls against two- and four-burn treatments as
well as one-burn plots against two- and four-burn plots were significantly different.
These analyses clearly show that unburned controls had the highest density of large-
diameter understory trees than the burned treatments, the size goes on decreasing as the
number of burns increases and such effects remain in the forest one to two decades after

low-intensity prescribed burns.

92

 

005m 50:: .9: ”3:3.“— 03._- .mm m:::m 0:0 .9
._.o w 0 :0 ::00:-::w_m-:0: 88:00 0: 0:0 ::00_-::w_m 88:00 0 .:<
._.o w 0 :0 050:0 0:0 E00 E0000 »_::00_-:_:w_m :0: 0:0 0:2 2:00 :3 0032—0: 030: 08:00 m0:_0> 0::-

.:00: 0:0:00. :0 m0:_0> 005002000000 0:: 0:0 :0E3 000.0 00800 53002
:0: 508:0 .wo. 0:30: :0 m0:_0> 00:50:20: :30: 0:: 0:0 SEE 00:00::00 0:: 0:0 55:00 . AEm-d 0:002 20:5 :0004 0:: 0:0 €200 00-h
.::0E:00:: £000 :0: 0:02: 0:: :000 :0>_w 0:0 :25: 0:0 :032 .m:_E: 00:000000n40

 

 

 

 

 

 

 

m m: m: m? 0% mm m> m:
m m: 0: m0 0% mm m> 0:50 a. m:
m m: m .0:::00 m> 2:3 =<
10% g g 0:005:00
0N N 0 0 N.» N 0...— 2 m0: 2 a 0m m2: m0 0 03 800.0 «90.: 8.0: .8 mmflqli
00: M: a mom am: _ No 0 000 new: Nvm 0 0m:- mvfl 0mm 0 03- 003.0 00.3.0 8.0-: .6 E3002
gm: .3 0 mm: 03:: on: a $00 :33: Z: a 3.3 Smm Em 0 £9 omood 09.0.0 SN-B =0Em
70: 0.00:: T0: 000:: T0; 000:: T0: 000:: 00-50 0:_0>-.:Am E0
:0003 030-: .0650 :0003 :0304 b_m:00 00:3 030-: €200 0003 :0304 €200 :00t0 ::0E:00:-_- 000020 00:05
40 40 40 40
mE:m ::0--: mE:m 0.5-r 5:5 0:0 35:00

 

 

 

 

.51:
moom .m::0E:00:: wEEB E00050 :00:: 0030—0 0E0 :0:0E0_0 ::0:0.c_0 :0 A70; m00::v 03200 006000 09200:: .0003 H_ _ .m 030-0

93

Killing of stems smaller than 10 cm diameter by low intensity burns has also been
reported by Niering et al. (1970) and Hodgkins (1958). A decrease in density of large
understory trees (6 to 10 cm) was reported by Henning and Dickmann (1996), who
surveyed in the growing season immediately after the prescribed burns. However,
recruitment of small-diameter seedlings was enhanced in all the burned plots. In a study
of a red and white pine forest in Michigan, 0 to 6 cm diameter seedlings and saplings
decreased due to burning (Neumann and Dickmann, 2001), but the sampling was done

right after the fire.

One of the obvious reasons for a low density of seedlings in the absence of fire is
that the seedbed was not suitable for the germination and establishment of new seedlings
since the understory low herb and shrub canopy, and thick duff and litter layer in the
forest floor prevented seeds from reaching the mineral soil (Figure 2.6a and b). The
reduction in depth of organic matter by fire increases establishment directly because the
seedlings are near to a more constant water supply and more decomposed organic layers
and mineral soil (Thomas and Wein 1985). In my study, two-burn plots had the highest
frequency and density of seedlings; thus the preparation of seedbed must have been much
better than in unburned control and one-burn plots (Figure 2.6a and b). In the four-bum
plots newly recruited seedlings were killed again and again, suppressing their
establishment and development until after the last burn. However, repeated burns must
have consumed more duff and litter, allowing more seeds to reach the soil and delayed
the growth of understory shrubs. The exposed mineral soil (Figure 2.6a) favored red pine

more than its associates (Table 2.8, Figure 2.4a to (1), because it is more adapted to

94

Figure 2.6a: A plot just after its second burning (two-year interval), with an
unburned plot in the background. Most of the soil organic layer has been
consumed, leaving a layer of ash over mineral soil, and all understory vegetation
has been killed back to ground level (photo by D.I. Dickmann).

.1

. L . I ‘0 0 _ -
.~ ‘ .- : ; >‘ .43, ' _
. ~ ‘ 7 .'-_ P

    

        

Figure 2.6b: A plot after the first fire in 1985, which was relatively “cool.” A
patchy mosaic of unburned forest floor and ash-covered mineral soil was created.
This plot was burned again with a hotter fire five years later (photo by D.l.
Dickmann).

 

95

post-fire establishment. Average age (13 years) of 0 to 6 cm seedlings of red pine and
other important species from the burned plots (Table 10) also suggest that the new
regeneration was from seeds that germinated and established right after the fires. In a
long-term study in a boreal forest in Yukon, British Columbia and in interior Alaska,
recruitment of dominant tree species were highest in the first five years (J ohnstone et a1.

2004).

The introduction of fire has brought changes in the overall structure and
composition of the forest I studied. Burned plots have basically two layers; one of
overstory canopy red pine trees and the other an understory of small-diameter woody
vegetation. Additionally, there is a low ground vegetation layer at the forest floor. The
burned stands in due course of time would develop a different composition and structure
than the unburned plots (see Chapter 3). The extent of changes that occur in the forest
structure and composition depends on the number of burns carried out, the burn intervals,
and the time period lapsed after the last burns. Since large hardwoods such as red maple,
sugar maple and basswood were eliminated by fire, the repeatedly burned forest will
develop into two layers: an overstory of red pine trees without any hardwood species and
a lower, red pine - dominated understory layer with other hardwood species mixed in.
Once-burned plots also will have red pine - dominated understory; however, it will have
some sub-canopy overstory trees of red maple, sugar maple, beech, and white ash. The
species densities measured a few years after fire and two to three decades later show that

future stand structure can be predicted from observations made early in succession and

96

that early post-fire colonizers usually will dominate the canopy of mature stands

(J ohnstone et al. 2004).

2.3.3.4 Species diversity and abundance

The species area curves (SPA) shows that the number of species in all of the
blocks tended to stabilize after sampling about five 100 sq m plots out of a total of 8
(Figure 2.7). Statistical analysis of the species richness and diversity indices revealed that
there was no significant overall treatment effect on species richness and diversity indices
of the understory trees (Table 2.12). However, once and twice-burned plots tended
towards highest species richness than four-bum and unburned plots. A similar significant
decrease of species richness in four-bum and an increased species richness in two-burn
plots were reported earlier by Henning and Dickmann (1996). Simpson’s Index was the
highest in four-bum plots and lowest in control and one-bum, which indicated that the
four-burn plots showed dominance of fewer species, in this case red pine. This finding is
very important if the goal of management is the establishment and growth of red pine
trees. The Simpson’s Reciprocal (l/D), Shanon and Evenness indices indicate that the
diversity of species tended to be highest in control and once burned plots and the lowest

in the four-bum plots.

The frequency for all size classes of woody understory species increased due to

burning and was the highest (34 %) in two-burn plots and the lowest (14%) in control.

Among the different size classes, the 0-4 cm size class was most frequent, with the red

97

Figure 2.7: Species area curves by block for understory woody vegetation in a mature red pine
plantation under burning treatments.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

20 4 Block A
g 15 ‘ 14
"’ 3 11
o .9 _
g g ‘0 —x— One Burn
"'2‘ : +Two Burns
5 5 _ +Control
0 + Four Burns
1 2 3 4 5 6 7 8
Total number of sample plots (10 sq m)
Block B
E 20 . 20
E
a 17
33 15 0
§ 10 q ’ A. ‘ 11 +Four Burns
g + Two Burns
‘t‘é 5 7 X + Control
6 O I I ‘1' I I 1 V fl —x- one Burn
1 2 3 4 5 6 7 8
Total number of sample plots (10 sq m)
Block C
3 20 -
.D
E
3
C 15 :
.§ 11
O
at. 10 4 /x 8: X4 10 +Control
.12: / 8 —X- One Burn
g 5 " +Two Burns
6 0 I I I l I T l 1 + Four Burns
1 2 3 4 5 6 7 8

Total number of sample plots (10 sq m)

98

Table 2.12: Indices of species diversity of understory woody vegetation in experimental red pine
plantation, 2003. (n=12)

 

 

Species Simpsons'
richness Simpson Reciprocal Shanon Evenness
S D ND 11' J
Treatments ns ns ns ns

Control 5.4 a (1.2) 0.4 a (0.07) 3.2 a (0.5) 1.2 a (0.17) 0.7 (0.08)
OneBum 6.1 a (1.2) 0.4 a (0.07) 3.1 a (0.5) 1.2 a (0.18) 0.7 (0.09)
TwoBum 6.0 a (1.2) 0.5 a (0.07) 2.8 a (0.5) 1.1 a (0.17) 0.6 (0.08)

FourBurn 4.4 a (1.2) 0.6 a (0.07) 2.0 a (0.5) 0.8 a (0.17) 0.6 (0.08)
Species diversity under Simpson column are backtransforrned natural log vlaues.

 

For others, the Least Square Means (LSM) are given and the values with the same letters are not
significantly different from one another at p S 0.1.
ns signifies non-significance, respectively, at p S 0.1.

Standard errors of means (i) are in parenthesis.

 

 

Evenness index was not analyzed as it was not normal even afier transformations

 

pine (49%) the most frequent species followed by red maple, beech and black cherry.
However, red maple was the highest occurring species for both the 4.1-6 cm and 6.1-10
cm classes with 35% and 52% relative frequency, respectively. For each of the above
diameter classes beech, choke cherry, black cherry, non-commercial trees, and oaks were
below 5% of total. There was no significant treatment effect (F-test) on any of the major
species density. Also binomial comparison of proportions among each of the important
species and the remaining species were not different across the treatments and, therefore,

only important species will be discussed.

There was a remarkable overall increase of understory red pine density in one-,

two- and four-bum plots——3l4, 387 and 230 %, respectively, over unburned controls,

99

mainly in the 0-4 cm class (Table 2.8). The importance value (IV) of red pine was the
highest in two-burn plots and the lowest in unburned controls (Table 2.13). Red pine
obviously will be dominant in the canopy of future stands, especially after burning, but

few conifers—with the exception of an occasional white pine—will accompany it.

Red maple was second to red pine in density and relative density in burned plots
(Tables 2.8) but its IV was highest in unburned control (Table 2.13). Relative density of
red maple was remarkably similar in all of the burned plots. Vigorous sprouting and
enhanced seed germination on the exposed mineral soil after burning allowed red maple
to successfully recruit into the 0-4 cm class following fire (Table 2.8). The canopy of
future stands also will be occupied by red maple. Other, hardwood tree species with
significant densities and IV in burned plots were beech, black cherry and white ash.
Sugar maple, on the other hand, was much reduced following burning. Very few
hardwoods survived burning to recruit into the sapling and small pole classes (4.1 — 10

cm dbh), especially if the burning was repeated (Table 2.8).

Table 2.13: Importance Value (IV) of the understory woody species.

 

 

Species*
Treatments AB AM BC CC I-IH Oaks Others RM RP SM WA
Control 33 14 14 11 17 6 14 79 44 31 31

Onean 54 17 41 22 3 7 17 52 52 12 29
Two Burns 44 12 42 12 8 23 13 29 92 6 14
Four Burns 27 12 58 15 21 16 10 20 68 11 31
I V=Importance Value (sum of relative frequencies, density and basal area).
*AB=American beech, AM=Amalanchier, BC=black cherry, CC=choke cherry,
HH=Hophornbeam, RM=red maple, Others=non-commercial species, RP=red pine,
SM=sugar maple, WA=white ash, WP=white pine.

‘S
d-bON

 

 

 

 

 

100

2.4 Conclusion

The effects of prescribed fires conducted during 1985 to 1991 were still evident in
2003, even after such a long gap. Fires affected differently the overstory, mid- and sub-
canopy trees, understory woody seedlings and saplings, and herbaceous ground
vegetation. Such differential effects on size classes and species as well as dense
recruitment after fire create distinctive stand structures (Peet and Christensen 1987).
Henning and Dickmann (1996) and Neumann and Dickmann (2001) reported that there
was no adverse effect on mature red pine overstory trees when measured immediately
after fire. In my study, overstory tree density was affected, where as there was no effect
on other features of overstory red pine trees. Also, the structure and composition beneath
the red pines, from the sub-canopy to the ground layer, was altered by fire, and the
alteration was substantial after repeated burning. In the southern Appalachians, Philips
and Shure (1990) also found a considerable shifting in tree composition following fire.
The effects of fire as a disturbance agent vary at different levels because of the
hierarchical organization and patchy nature of ecosystems (Pickett and White 1985,
Rykiel 1985, Holling 1992, Wu and Levin 1994). Whereas the response to burning shown
by different species and their size classes varied in my study and depended on the number
of low-intensity burns applied, the burn intervals, and the number of growing seasons
passed since the last fire, recovery was nonetheless rapid. In fact, forests or biological

communities are always recovering from the last disturbances (Spurr 1956, Reice 1994).

Prescribed low-intensity burns, particularly if repeated, reduced the sub-canopy

tree density sharply and wiped out most sub-canopy conifer and hardwood trees.

101

However, post-fire recruitment of most species—especially red pine, red maple, beech
and black cherry—did occur via seedlings or sprouts. Consequently, these burnings
altered the intra-specific competition among the mid-canopy and dominant overstory red
pine trees. The fires eventually must have also changed the availability of nutrients and
other resources reaching the remaining mid-canopy and large-sized overstory trees. Lunt
(1950 in Dickmann 1993) found red pine to increase in height and volume after 20 years
of annual burning in Connecticut, and such increase was thought to be due to release of
phosphorous and exchangeable calcium and increased pH due to burning. Disturbances
release nutrients tied in pre-disturbed vegetation and also those utilized by vegetation
before disturbance (Bormann and Likens 1979). Usually, phosphorous, calcium and
potassitun are released after fire and utilized quickly after their release by existing or new
vegetation (Alban 1977, Boemer 1988). Vitousek (1985) suggested that such nutrient

availability declines after the canopy closes.

Size of openings created by fire also affects micro-environment, species
composition and increases productivity (Philips and Sure 1990), and if the openings
become larger, productivity increases further. Three to four times increases in net primary
productivity of above ground biomass in larger gaps were shown by Philips and Shure
(1990). Ahlgren (1960) reported the increase of early plant invaders or pioneers in
burned areas for several years, which declines as the time from burning increases. The
nutrients leached from the system after burning gets trapped by larger mature trees.
Resources in disturbed gaps are under-utilized as the gaps partially or completely lack

vegetation (Marks 1974). These resources will be available to dominant overstory trees

102

and also to fast growing pioneers that occupy the open space first after disturbances and
nutrients are tied up in the biomass of those early invaders (Boring et al. 1981). In spite
of the losses due to volatilization and leaching during the rainy season, nitrogen and
phosphorus availability increases immediately after fire due to increased soil temperature
and moisture, nitrogen fixation, mineralization and release (Christensen 1985, Vitousek
1985, Raison 1979). In my study, enhanced growth in the dominant overstory trees in all

of the burned plots and particularly in four-burn plots, confirms such effects of fire.

Additionally, the highest IV of the overstory red pine trees in the burned
treatments and the gradual increase of IV for overstory red pine trees from control to one-
to two- to four-bum treatments indicates that red pine becomes more dominant with more
frequent burning. Prediction of canopy dominance of trees in future stands based on
higher IV of younger age/size classes has been suggested (Flaccus and Ohmann 1964).
The greater IV of red pine of different ages and size classes in burned treatments

indicates that red pine will be important in the canopy of future stands (see Chapter 3).

One- to almost two-decade-old prescribed burns have enhanced overall
vegetation diversity and cover in the understory and favored the reproduction of red pine.
Enhanced immediate post-fire red pine recruitment was reported by Henning and
Dickmann (1996), and this cohort has developed over time. These effects probably will
be carried over, but as the overstory trees, understory woody seedlings and saplings and
the herbaceous ground vegetation grow in time and space, diversity will decline. Species

richness and diversity increase quickly after fire (disturbance) and reach a peak but

103

decrease later as the succession progresses (Loucks 1970; Denslow 1985). In boreal
forests in British Columbia and interior Alaska, early recruitment was highest in the first
five years and then some of the species declined after 10 years in mixed stands

(J ohnstone et al. 2004). In my study, the higher relative densities and IV but lower overall
densities of red pine reproduction in the four-bum plots than in once- and twice-bumed
plots show clearly how repeated burning can shift composition toward a highly fire-
adapted species (Van Wagner 1970, Dickmann 1993). Absence of fire caused highly fire
adapted longleaf pines to convert to loblolly and slash pines along Southeast coastal areas
(Walker 1980) and increased fire sensitive species (Arthur et al. 1998). Even infrequent
disturbances can have long lasting effects in forest structure and composition (Henry and

Swan 1974).

The most remarkable changes in structure in the red pine forests in my study
was the shift from a bimodal diameter distribution in unburned plots to a unimodal
distribution in two- and four-bum plots. A single burn created a structure roughly in
between bi- and unimodal because it eliminated only a few large-sized sub-canopy trees.
Frequent burning virtually eliminated the sub-canopy stratum. These overall changes in
the species composition and structure of these stands can reflect future stand

development.

Burning decreased the height of shrubs and sub-canopy trees (Little and

Moore 1949; Ahlgren 1960; Dickmann 1993). Early stages of red pine plantations were

considered as unsuitable for white tailed deer, snowshoe hare, and ruffed grouse (Gysel

104

1966). However, after burning in mature red pine plantations, herbs and shrubs increase '
and the heights of the sub-canopy stratum is reduced (Little and Moore 1949; Dickmann
1993). The increase in herb and shrub richness and the sprouting back of hardwood
species provide a good source of browse for animals; thus, thinning and burning could be
very useful for wildlife management (Dickmann 1993). Bender et al. (1997) found small
and large mammals such as elk and deer increased after overstory thinning, again due to
increased quantity and quality of forage. Although thinning and prescribed burning are
different, they have some common benefits because they both create openings, enhance
sprouting and seedling recruitment, and decrease heights of hardwoods, all of which

enhance wildlife habitat in red pine stands.

2.5 Recommendations

Prescribed burning should be considered a tool for the long-term management
of red pine forests. The number of burns carried out, the burn intervals, and the period of
gap to be left after prescribed burning should be chosen judiciously and will depend

entirely on the management goals and objectives.

A single fire with a long growing period (one to two decades) without fire
results in a forest with an overstory of red pine and a few small-sized sub-canopy
overstory trees of other species below them. Red pine seedlings and saplings will be
mixed with hardwoods in the understory woody layer. If the focus of management is the
enhancement of growth and yield of the red pine only, one-burn alone may not be

sufficient since the effects of single low-intensity fire declines with time. The

105

determination of the length of the growing period after last fire or other silvicultural
practices to be adopted in combination with burning have to be chosen judiciously so that
forest growth is revitalized and stimulated. Burning such forests every 15 to 20 years
favors red pine grth as well as some browse in a sustained basis for wildlife

management.

A red pine forest with a history of two-bums at a five-year interval and left to
grow for a long gap of about 11 years will lead to a unimodal distribution of overstory
trees. Such repeated burnings increase the importance of red pine in both the overstory
and understory tree layers, increase overall density and diversity of understory woody
seedlings, enhance sprouting of red maple and other hardwood species, increase tree
density in the herbaceous ground vegetation layer, increase openness in the forest, and
kill back small hardwood stems, many of which resprout. This is a better option for

management of red pine as well as wild animals, compared to a single or no burning.

Multiple fires at two- to five- year intervals, and then a period of fire
exclusion seems to be the best option to obtain vigorous red pine reproduction. This fire
regime will clear everything fi'om the sub-canopy and understory leading to a unimodal
overstory tree distribution with the highest annual growth in the overstory trees. It also
creates an ideal seedbed for red pine. Red maple, black cherry, beech, oaks,
hophombearn, white ash and other species also get favored due to repeated burnings,
although their relative density will be low. Thus, with multiple burns we get a dominance

of red pine in all the three strata: the overstory, the woody understory, and the ground

106

vegetation. So this regime is the best option if the objective is the long-term management
of red pine. Multiple prescribed burns combined with some other silvicultural practices,
such as thinning, would also give increased growth and yield of red pine. The low
understory vegetation created by this regime would be very favorable to browsing
wildlife and low-nesting birds, although the lack of mid- and sub-canopy strata would not
provide habitat for other mammals and birds. Understory vegetation, which has low
commercial value, should be given due importance as it has a very important ecological

role in occupying disturbed sites and yielding to other late successionals.

Additionally, extremely dense red pine regeneration established after one to
two decades of repeated prescribed burning in mature red pine stands on nutrient rich
sites might need release from suppression by pre-commercial thinning so as to maximize
growth and yield (Benzie 1977). On nutrient poor sites repeated burning should be
applied to the extent that it may suppress hardwood species sprouting and enhance red
pine establishment and growth; however, burning should not retard or interfere with red
pine regeneration and establishment. On these less fertile and often droughty sites red
pine recruitment following fire may not be as vigorous or predictable as occurred on my

site, so early pre-commercial thinning may not be necessary.

107

Chapter 3: Using the Forest Vegetation Simulator (FVS) to Project the Long-Term

Effect of Prescribed Fires in a Red Pine Plantation.

3.1 Forest Simulation

The Forest Vegetation Simulator (F VS) is the USDA Forest Service’s nationally
supported framework for forest growth and yield modeling (Dixon 2002). The Growth
and Yield Unit of the Forest Management Service Center (FMSC) in Fort Collins,
Colorado develops, maintains, supports and transfers F VS and related technology. F VS
developed from Prognosis, a modular structure developed by Stage (1973 in Dixon 2002)
and is an individual-tree, distance-independent growth and yield model. It can simulate a
wide range of silvicultural treatments for most major forest tree species, forest types, and
stand conditions, ranging from even-aged to uneven-aged. There are different variants
developed for different geographic regions. For example, the Lake States (LS) Variant
used in this study is an adaptation of the LS-TWIGS model and applied to the Lake State
Regions (MI, WI, and MN). However, since it is a distance-independent model, it does
not do well with thinning by group selection. F VS treats a stand as the population unit,
using forest inventories or stand examination data. FVS works through the Suppose
interface (platform), which is a graphical user-interface that accesses FVS software
through Windows 95 operating system 9. Simulations are reproducible and deterministic
by default; however, they can also be run stochastically. While simulating, post-
processors (other programs) are used for further reporting and analysis of FVS output

(discussed further under Materials and Method section of this chapter).

108

Canavan and Ramm (2000) compared the actual measured growth, BA and
mortality for seven upland hardwood species in northern Lower Michigan with TWIGS
and FVS_LS five and 10-year projections. They found that FVS-LS produced more
accurate results for diameter and BA growth than LS_TWIGS, whereas neither of the two
simulators produced as accurate results in case of ten-year mortality (tree density). The

diameter prediction was within :t 3.2 cm across all projections, species and size classes.

Similarly, Smith-Mateja (2003) used RPAL (REDPIN E) to project stand level
attributes and FVS-LS to project tree and stand level attributes of large trees of red pine
(Pinus resinosa) at Kellogg Forest in Lower Michigan. Overall FVS was more robust
than RPAL. F VS was found to be a good predictor of diameter growth and was even
better when diameter growth data from increment cores was used to calibrate it. She
suggested that validation should be carried out in small-sized red pines and minor species

and different site conditions as well.

The red pine plantations that were the basis for my simulations were mature
(established in 1931) and partially harvested in 1970, leaving an understocked condition
with a variable residual basal area (BA) averaging 10 m2 ha'l. The stands were burned
using low-intensity single and multiple prescribed fires during 1985 to 1991 and were
left to grow without further intervention (Henning and Dickmann 1996). The re-
measurement of the stands was done in 2001 and 2003 (Chapter 2). The data in Chapter 2
revealed that the average diameter of the burned and unburned stands was about 43.2 cm

and the overstory trees were mature, occupying between 17.9 to 23.3 m2 ha'1 in

109

differently burned plots (Table 2.1). Moreover, there was a very dense advance post-fire
regeneration of red pine and hardwoods established after different prescribed burnings
were carried out, but the stands needed some kind of silvicultural operation or
management to enhance growth and productivity. Re-burning was not an option as the
burned plots already had a well-established one- to two-decade-old advance regeneration
which needed protection; young red pine recruitments cannot withstand even a low-
intensity fire (Van Wagner 1970). So, several obvious questions emerged. First, how will
current stand structure and composition change during 100 years of growth and
development? Second, which management system could be applied for maximum and
sustained yield from the burned and unburned forests and when could such silvicultural
management be started? Third, how much volume of wood or BA could be removed

during thinning to maximize saw log production of such stands?

The uniqueness of this study is that since the plots were affected differently by
prescribed fires, the data from burned and unburned plots could be projected 100 years
into the future using FVS with both no-management as well as thinning to examine the
long-term effects. The results could be useful for managers and planners as they

formulate desired future conditions for red pine dominated ecosystems.

My hypotheses are as follows: Burned plots, if allowed to grow for a long period
of time (longer than one to two decades), will be different from unburned plots in their
species composition and structures, and the changes or the differences will depend on the

frequency of burns. The successional trajectory of the red pine stands will be different

110

among differently burned and unburned plots and also the management applied (no-
management and thinning). The net volume yield of thinned stands will be higher than

those of unthinned stands whether they are burned or unburned.

The objectives of this study are:

0 To examine compositional and structural changes in previously burned and
unburned plots of mature red pine stands simulated 100 years into the future with
and without thinning.

0 To summarize growth and yield in previously burned and unburned mature red

pine stands managed differently.

3.2 Materials and Method

The Forest Vegetation Simulator (FVS) was used for forest growth and yield
modeling. The data input, interpretation of output, simulating management scenarios and
regeneration were carried out as suggested in the “Essential F VS: A User’s Guide to the
Forest Vegetation Simulator” (Dixon 2002) and the use of keywords from the “Keyword
Reference Guide for the Forest Vegetation Simulator” (Van Dyck 2003). Data on mature
overstory trees (>10cm in dbh) and understory trees (>1.4m in height and 0-10 cm dbh)
(Chapter 2) were entered in F VS tree data files. All the understory trees >30 cm to 1.4 m
in height were counted from all of the respective burned and unburned treatments and
were entered with an assigned diameter of 0.254 cm (Dixon 2002). This manipulation

was required by F VS, otherwise it would not grow anything <14 m during simulations.

111

F VS works through three sets of file systems that work in coordination through the
Suppose interface: the location file (*.loc), the stand list file (*.slf) and the tree data file
(*.fvs). As a first step, all the tree data (from 2001 and 2003 measurements) in Microsoft
Excel were changed to comma delimited files and then reformatted into FVS input format
(*. fvs) which had settings id, design purpose code and measurement numbers. Later,
stand-level information such as variant description, group codes and tree data files were
entered to the stand list file (*.slf). Finally, specific information about each setting, such
as sample design, was entered in the Suppose locations file (*.loc), which also contained
the tree list file name and a grouping code. Suppose version 1.18 and F VS Version 6.21
of Lake States Twigs (Revised 03.07.06) were used to project the data into different
management simulations which were manipulated through keywords. Simulations were
used deterrninistically by default (although they can also be run stochastically) and were
reproducible. While simulating, the post-processors (other programs in FVS) were used
for further reporting and analysis of F VS output, which was later brought back to Excel

for further study and analysis.

To test the study hypotheses and address study objectives these steps were
followed. First, the stands were projected without any silvicultural management. A Mean
Annual Increment (MAI) was obtained which showed that the stand was still growing up
to the year 2044. It was suspected that this observed MAI could have been due to the
actively growing advance post-fire recruitments. Rapid growth of MAI in red pine occurs
at 15 to 45 years of age and later the growth slows down before it culminates (Lundgren

1983). Therefore, to see if the mature trees were still growing or not, the seedlings or

112

saplings below 10 cm were all removed by using “thinning from below with dbh removal
from 0-4 inch” option from the F VS model and then the stands were projected again.
Once the fast—growing small seedling/saplings (0-10 cm dbh) were removed, growth
started declining much earlier than observed previously, after about the year 2004 in
some of the stands. This demonstrated that the growth observed in previous run was
mainly due to the growth of understory trees and the large overstory trees were mostly
growing at a very slow rate. Thus, some silvicultural intervention could be performed
immediately or within a few years in both the burned and unburned stands. The
simulation was started in 2003, the year when the inventory of trees and the understory
vegetation for this study was completed. Since the MAI started decreasing after 2004 (at
least in some of the stands), the silvicultural operations were always scheduled in the year

2004.

The following two options of management were chosen to be simulated for 100 or
more years into the future in all of the burned and unburned plots using FVS. The no-
management projection provides a longer-term perspective of different fire regimes and
their effects without any external disturbance, whereas thinning, as an additional
disturbance, provides an opportunity to study the effects of multiple silvicultural
disturbances. The stands were 78-year-old when simulations began and l78-year—old at

the end, when they exhibited some of the characteristics of old growth forests.

Thinning (partial cutting) is one of the best silviculutral operations for the

management of even-aged red pine (Benzie 1977) and is recommended to be done every

113

10 years or more (Lundgren 1983). I chose thinning at different residual BA targets for
different cuttings through the FVS simulator. Although there are some limitations, fixed
target BA seemed to suit this site, as it had lower understory trees waiting to be released.
Thinning from below does not seem to be appropriate as the advance post-fire
regeneration needs protection and release, and the slow-growing tolerant smaller woody
understory species would all be removed from the stands. Rather, healthy and vigorous
intermediate and co-dominant trees need to be released from competition by dominant
trees for better growth in diameter. In other words, the removal of mature dominant trees

can rejuvenate the vigor of residual trees as well as the stand as a whole.

As per the stocking guide for red pine (Benzie 1977), stands having an average
diameter of 43 cm dbh (my study) need to retain a residual BA following thinning
between 24.1 and 27.5 m2 ha'l. However, Lundgren (1965) advised for high investment
returns to go as low as 20.6 m2 ha'l thinned every 10 years, regardless of the density and
stocking of the stands. My stands were highly variable in their BA, with some carrying a
BA smaller than 22.9 m2 ha". To determine the residual BA of subsequent thinnings, the
number of thinnings to be carried out and the interval periods between them, several runs
were made with various target residual BA. First, the residual BA of two different sets,
25.2, 27.5 and 34.4 m2 ha" and 25.2, 29.8 and 34.4 m2 ha", both at intervals of 20 as well
as 30 years were run for one-, two- and three-times thinning over the next 100 years. For
both the sets of residual BA used, the final grth in BA did not reach as high as 57.3 m2
M”. Later, as I kept lowering the target BA, the first thinning lowered to 22.9 m2 ha’l

basal area, resulting in a final BA as high as 57.3 m2 ha'l after 100 years. Thus 22.9 m2

114

ha'l BA was decided to be retained for the first thinning for all of the regimes. After
analyzing several runs I tested the following three thinning regimes that provided the
highest final basal area and yield:
0 Single Thinning: retaining BA at 22.9 m2 ha'l after a single cut in year
2004.
0 Two Thinnings: retaining BA at 22.9 m2 ha" in 2004 and 27.5 m2 ha'l
after the second thinning in 2034.
0 Three Thinnings: retaining BA at 22.9 m2 ha'l in 2004, 27.5 m2 ha'1 in

2034 and 34.4 m2 ha'l after the last thinning in 2064.

As I needed to retain and release the advance post-fire red pine and hardwood
establishment as well as competing co-dominants, I chose the ‘THINDBH’ key word
which considers thinning of all diameter class trees; however, a calculated proportion of
trees is removed until target basal area is met. This algorithm begins thinning the largest
diameter trees first, so the target BA may be reached before any smaller trees are

removed.

The tree density after the ‘single’ thinning was almost the same as under no-
management but much lower than that under the ‘two’ and ‘three’ thinning because
adequate or sufficient number of trees were not removed from the understory. Further,
the high BA (22.9 m2 ha") chosen to be retained during the first cut in 2004 was not even
available in some of the stands, particularly the one-burn ones. Thus, only the three times

thinning was used for this study as it retained highest density, species richness and

115

cumulative volume at the end of simulation compared to one and two times thinnings. It

will be referred to as thinnings or thinning, hereafter.

Because of the unusually high density of post-fire recruitment, the FVS model
was not performing well in the beginning; this problem was fixed with the help of Dr.
Don Dixon, biometrician with the USDA Forest Service, Fort Collins, Colorado.
Normally, regeneration is assumed to be around 988 seedlings per ha, whereas my stands
had a greater post-fire understory seedling density. There was virtually no space for new
seedlings to emerge as the ground was either covered with seedlings (in case of burned
plots), litter, or in open areas with thickets of Rubus and grasses. In unburned controls
there was a dense understory and sub-canopy of mainly the hardwoods and the duff was
so thick that very little regeneration was occurring. After discussing the matter with Dr.
Dixon and Prof. Don Dickmann, I decided to keep the new regeneration to only 2% of
total seedlings in each of the burned and unburned controls. For example, in unburned
controls the seedlings/saplings to be regenerated were controlled by creating regeneration
files (*.kcp) and regeneration was limited to 208 trees ha". Similarly, regeneration files
(*.kcp) were created for other treatments: one-, two- and four-bum plots with 388, 395
and 319 trees ha", respectively. Thus four sets of regeneration files were created by
averaging among the replicates of each treatment for use in the FVS simulations. Those
files helped to control the erratic behavior of FVS and the model ran perfectly fine after
the adjustments. I observed that the low rate of regeneration chosen to run the simulation
did not matter much as there was very little new regeneration taking place under the

dense advance regeneration thickets in the study sites anyway.

116

The FVS model averaged all the data in the output and the results were obtained
on a per stand basis, so I averaged the output for the three replicates of each treatment
(n=3) and then summarized and discussed. As the model result was based on so many
assumptions, 1 did not use statistical analysis. Consultants at the MSU Statistical
Consulting services recommended that such analyses are inappropriate because of the

multiple assumptions that are inherent in simulation models, including F VS.

I summarized tree density, BA, MAI and total cubic merchantable and sawlog
volumes. For detailed study, the trees in two strata or horizontal layers were categorized
by breast height diameter class (dcl) into the understory (0-11 cm dcl) and the overstory
(>11 cm dcl) trees. The understory trees were further categorized into small (0-6 cm dcl)
and large (6-11 cm dcl) trees. Similarly, the overstory trees (>11 cm dcl) were
categorized into sub-canopy (>11-32 cm dcl), mid-canopy (32-42 cm dcl) and dominant

(>42cm dcl) overstory trees.

The cumulative total cubic and saw log volumes were calculated from the output
data. Relative densities and relative BAs for different treatment plots as well as species
also were calculated. Relative density is the number of individuals of one species as a
percentage of the total number of individuals of all species. Similarly, relative BA is the
BA of species as a percent of the total BA of all species. Relative value index or
importance value (IV) then was calculated from the relative densities and relative BA.

Percent increase over unburned treatments was calculated for each period to look at the

117

long-term effects of fire. Also, the percent change over no-management was calculated

for each period to assess the effects of post-fire silvicultural management.

MAI was taken directly from the output of the model, which by default added all
the previous removals of cubic volume (through thinning, if any) to the total cubic
volume before thinning at any age and divided the sum by the age of the stand. To
calculate the cumulative cubic volume for the year in which the cutting occurs, the
removals of the same year were not added. The previously removed total cubic volume
by thinning was added to the volume before thinning to compare the yield between
thinned versus unthinned stands but not for total growth (Spurr et al. 1957). Thus, yield

has been used as the additions of all the net annual increments as in Buckman (1962).

3.3 Results
3.3.1 Tree density

Tree density at the beginning of the simulation ranged from 10,067 to 19,934
trees ha'l with the highest density in two-burn and the lowest in the unburned treatments
(Table 3.1). The trees were categorized into two groups; the woody understory (0-11 cm
dcl) and the overstory (>11 cm dcl) trees. Understory trees constituted 98% (unburned
plots) to 99% (burned plots) of the total trees. Overall tree density dropped sharply in all

of the treatment plots by the end of both of the no-management and thinning simulations.

118

3.3.1.1 Understory tree density

At the end of no-management, the relative density of understory trees had
declined sharply, especially in the burned treatments. At the end of thinning, relative
density declined less and was still quite high in the frequently burned treatments (Table
3.1). After no-management, the unburned controls retained the highest understory tree
density, despite their lowest density in the beginning. However, at the end of thinning, the
densities of understory trees retained in the one-bum and unburned controls were 6 to 7
times lower than that in the two- and four-burn plots. When compared with no-
management, understory tree density after thinning was more than double in unburned
controls and one—burn treatments and more than 20 fold in two-burn treatments and 90

fold in the four-burn treatments.

Small understory tree (0-6 cm dcl) density in the unburned controls was almost
one-half that in the burned plots in the beginning (Figure 3.1a, Appendix 3.1). At the end
of the no-management, the density of small understory trees retained was the highest in
the unburned controls, whereas in all of the burned treatments, small trees were nearly
absent (Figure 3.1b). After thinning, the densities of small understory trees retained were
much higher in the two- and four-bum treatments than the other treatments (Figure
3.1c)—two, six, 45 and 90 times more in control, one-, two- and four-burn plots,
respectively, than after no-management. At the end of no-management, the relative
density of the small understory trees declined to 7% and 2% of total trees in unburned
controls and all burned treatments, respectively, whereas the relative percentage was

much higher after thinning, particularly in the burned treatments.

119

In the beginning, large understory trees (6-11 cm dcl) were absent or <0.5 % of
total trees in all of the burned treatments, whereas they were >6 fold higher in unburned
controls (Figure 3.1a). Unlike the small trees, the large understory trees increased in
density and relative density in all of the burned plots during early periods of both the no-
managements and thinning simulations and then declined gradually (Figures 3.1b and
3.1c, Appendix 3.1). At the end of no-management, few large understory trees remained,
with an especially large drop in density in the control (Figure 3.1b). However, at the end
of thinning, trees in this size class also were lost in unburned treatments but big gains
were shown in frequently burned treatments. At the end of thinning, the increase in
density of large understory trees was more than three times in unburned controls and

more than five times in all of the burned treatments than in no-management.

3.3.1.2 Overstory tree density

Initially, overstory tree (>11 cm dcl) density was greatest in the control plot
followed by that in the four, two and one-bum plots (Table 3.1). As the no-management
and thinning simulations progressed, overstory tree densities in all burning treatments
increased and reached their highest value in the year 2044, then declined slightly;
however, densities were still much higher (about 3 to 5 times) at the end of the run than
in the beginning. At the end of both no-management and thinning, the burned stands had
a higher density of overstory trees than the unburned controls, which had the highest
density in the beginning (Table 3.1). Relative densities after thinning were one-half those
after no-management in two and four-bum treatments; however, they were only slightly

less in one—burn and control plots.

120

 

Table 3.1: Average and relative live understory (0-1 1 cms dcl) and overstory (>11 cm
dcl) tree density" under 100 years of simulation" in mature red pine stands.

 

 

 

 

 

Understory-density Overstory-density
trees ha.l trees ha-l
Burn treatments
Year C 18 28 48 C 18 ZB 48
Beginning
2003 9869 18797 19793 16725 198 1 15 141 154
% 98 99 99 99 2 1 l 1
No-management
2014 8418 16143 15902 13977 385 129 164 159
% 96 99 99 99 4 1 l 1
2044 778 3413 2192 3059 741 1063 911 704
% 51 76 71 81 49 24 29 19
2074 126 133 91 71 549 767 641 509
% 19 15 12 12 81 85 88 88
2104 45 18 30 8 437 593 513 445
% 9 3 6 2 91 97 94 98
Thinnings
2014 8226 16143 15600 13325 370 129 150 139
% 96 99 99 99 4 1 1 l
2044 1783 3260 3042 3109 557 824 639 491
% 76 80 83 86 24 20 17 14
2074 349 624 1017 1720 469 568 475 330
% 43 52 68 84 57 48 32 16
2104 106 101 610 711 378 483 427 417
% 22 17 59 63 78 83 41 37

 

Note: Some of the 'zeros' in % column have values below 0.5% and have been rounded

IO ZCTO.

Burn treatments in columns are: C, Unbumed Control; 18, One Burn; ZB, Two Burns;

43, Four Burns

% (in rows) is the percent of total tree density.

dcl=diameter size class.

*Stand density does not include removals in thinnings.
"Simulation started on year-2004.

The first, second and third thinnings occurred in 2004, 2034, and 2064.

121

 

Figure 3.1: Understory (0-11 cm dcl) tree density in differently burned plots under
no—management and thinnings“ simulations for 100 years in mature red pine stand.

 

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a: Beginning-Year 2003
25000 " DControl

2 20000 ~ 18752 19756 EIOne Bur”
,7, flTwo Burns
§ 15000 . IFour Burns_
in

I—

; 10000 —

U)

5

o 5000 «

324 46 37 0
0 ’ 1
(0—6 cm) (6-11 cm)
Diameter Size class
b: No-management—end of simulation-Year 2104
700 — 1:] Control

N 600 — EOne Burn

.5

a, 500 — BTwo Burns

3 400 < I Four Burns

2 .1

'T 300 -

E

2 200 —

8

10° ‘ 31 9 1o 7 14 9 20 o
0 I ' 1
(0—6 cm) (6-11 cm)
Diameter Size classes
c. Thinnings-end of simulation-Year 2104
700 . 633
a 111 Control
.C
5 BOne Burn
3 ETwo Burns
8 I Four Burns
*7
2‘
“a
C
(D
D
(06 cm) (6-11 cm)
Diameter Size Class

 

 

 

*First, second and third thinnings were done in years-2004, 2034 and 2064, respectively.

122

3.3.2 Stand structure

Trees were distributed in two distinct age or size-class strata at the beginning of
the simulation (Figures 3.1a and 3.2a; Appendix 3.3a). The understory was formed of
extremely dense post-fire regeneration as well as sub-canopy trees, the latter especially in
unburned treatments. The upper stratum contained mature mid-canopy and dominant
overstory trees and formed a uniform bell shape, with most trees in the 37 to 52 cm dcl.
The overstory tree distributions were similar in the beginning and during early
simulations in both the no-management and the thinning; however, they became different
towards the end of the simulations. As both the simulations progressed, a shift of trees
from smaller to larger dcl occurred, which led to the formation of different vertical strata
in the stands at the end of the simulations than at the beginning (Figures 3.2a to c; 3.3a to
d). In general, the burned plots had higher average densities in the entire range of

diameter classes than in the unburned controls.

3.3.2.1 Bimodal tree distribution

After no-management, the trees formed a bimodal distribution of which the lower
stratum formed a normal bell shape, with understory trees and overstory trees to 47 cm,
whereas the upper stratum formed a J -shape with large trees from 47 to >62 cm (Figures
3. l b, 3.2b; 3.3a to (1). There was a shift from ‘inverse-J’ shape in the understory and the
sub-canopy tree layers in the beginning to a more uniform and normally distributed (bell)
shape at the end. Second, the bell shaped distribution in overstory trees at the start shifted

to a J -shape at the end of the simulation (Figure 3.2b). The bell shaped distributions

123

showed distinct peaks for unburned, one-bum, and two-burn treatments but was much

flatter for four-bum treatments.

3.3.2.2 Tri-modal tree distribution

After thinning, the trees formed a tri-modal distribution, which was most
pronounced in the burned treatments, with an inverse-J shape in the understory layer, a
uniform bell shape in middle layer (sub-canopy and mid-canopy overstory trees) and a J -
shape in the dominant overstory layer (Figures 3.1c and 3.2c; 3.3 a to d). The bell shaped
distributions again showed distinct peaks for unburned controls and both one- and two-

burn treatments, but was essentially flat for four-burn treatments.

Total tree density was always higher in the thinned than in the unthinned plots
(Appendix 3.3). More trees were found in the 32 to 47 dcl as the management changed
from no-management to thinning, in both unburned and burned treatments. Thinning thus
enhanced more diameter growth in sub-dominant trees, resulting in the shifting of smaller
to larger size classes, than no—management. Because the thinning was from above, the
density of dominant overstory trees (> 57 cm) was lower in the thinned stands than in the
unthinned ones in all of the treatment plots (Figure 3.3a to d). The highest density of
large trees was in four-bum treatment followed by the two, one and unburned controls in

both managements.

124

3.3.3 Species density

In the beginning, red pine, red maple (Acer rubrum), sugar maple (Acer
sacchharum), choke cherry (Prunus virginiana) and black cherry (Prunus serotina) had
high densities and relative densities, followed by the least represented species white ash
(F raxinus americana), beech (F agus grandifolia), and others (Table 3.2). Red pine
density was higher than any other species, with highest density in two-burn and lowest in
unburned controls, which also were lowest for all other species except sugar maple. Red

maple was next to the red pine in its density and relative density.

The density and relative density of red pine was highest in all of the treatments
throughout the simulations (Table 3.2). They were higher in the burned than in the
unburned controls at the end of no-management, whereas, they were much lower in the
burned treatments (except in the once burned) than in unburned controls at the end of
thinnings. After thinning, most species except red maple and beech had their highest

densities in the four-bum plots.

Despite an overall decline, red maple was next to red pine in species dominance
until the end for both of the managements, and it was the lowest in four-burn treatments
among different burn treatments (Table 3.2). Red maples were higher in all of the thinned
treatments than the corresponding no-management treatments. The density of red maple
was the lowest in unburned controls than in other treatments until the year 2044;
however, later it increased and became the highest in control at the end of no-

management, while it remained higher in one and four-bum treatments after thinning.

125

Figure 32: Tree distribution by diameter size classes in differently burned red pine
stands under no-management and thinning" simulations for 100 years.

 

   

 

 

 

 

 

 

 

    

 
 

Density-Trees per ha
0)
O

  

"2
' r v
‘ 4':
it
. . ‘.'.
.z. 1.;
. . _.

11-17 17-22 22-27 27-32

Diameter Size Classes (cm)

 

z: .1.
I 3 >2 I 3 m I

         

T

a: Beginning - Year 2003
180
DControl

m 160i
'5 140 j BOne Burn
3. 120 . ETwo Burns
in
8 100 ‘ IFour Burns
1‘—j 80 -
g 60 —
3 4° “ ..

20 - ,2 £53; £33 32;

0 , ..... I... _, “:3: lat? lair? '38: .1...
11-17 17-22 22-27 27-32 32-37 37-42 42-47 47-52 52-57 57-62 >62
Diameter Size Classes (cm)
b: No—management-end of simulation-Year 2104

180 —

160 UControI

140 ‘ EIOne Burn

120 ‘ BTwo Burns

- .3

32-37 37-42 42-47 47-52 52-57 57-62 >62

 

I Four Burns

    

.;.

.0.‘

.3

0.1

. ~.‘

. . . .

. .
In...

 

 

 

c. Thinnings-end of simulation—Year 2104

_I
0)
O

160 i

 

  

.a .I

N A

O O
l I

Density-Trees per ha
a)
O

  

 

Diameter Size Classes (cm)

 

.- ;.,
0V
1‘.
.
..
.
”e. n o
'u‘ 0'.
5' .3 ...
.;. .z.
. 1 c '.‘ . . c 5.
a“ '0‘ 5'. .‘ ‘o‘ y'. ’0'
. .- -z- I a to: I -

11-17 17-22 22-27 27—32 32-37 37-42 42-47 47-52 52-57 57-62 >62

 

 

[:1 Control

13 One Burn
BTwo Burns
I Four Burns

 

 

 

*First, second and third thinnings were done in years - 2004, 2034 and 2064, respectively.

126

Figure 3.3: Tree distribution by diameter class in year 2014 in differently burned mature
red pine stands under no-management and thinning simulations.

 

700 T

 

a: Unbumed Control

 

—e—— ueur
- - x- - -UBTH

 

 

 

 

 

 

 

 

.
v 1 r" 1

     

I

0-6 6-11 11-17 17-22 22-27 27-32 32-37 37-42 42-47 47-52 52-57 57-62 >62
Diameter Size Class (cm)

 

 

700 .
m 600 -
.C
a 500 ~
0.
§400 «
1? 300 -
.E‘
g 200 «
0)
0100 .

 

 

b: One Burn

 

—o—1BUT
- - ax- --1BTH

 

 

 

   

 

 

 

06 6-11 11-17 17-22 22-27 27-32 32-37 37-42 42-47 47-52 52-57 57-62 >62
Diameter Size Class (cm)

 

Thinned and unthinned (TH and UT respectivelY): Unbumed Control (UB), One Burn (1 8),
Two Burns (28), Four Burns (48).

127

 

Figure 3.3 (continued) : Tree distribution by diameter class in year 2014 in differently
burned mature red pine stands under no—management and thinning simulations.

 

c: Two Burns
700 w

“,1500 _
i: l +ZBUT
$5001 --ai<---2BTH
9- X

8400 i '.
9

17300 - ‘.
g. 1
2200 «
0

     
   

 

 

7 7 I I 7 — "I“ 7'

0-6 6-11 11-17 17-22 22-27 27-32 32-37 37-42 42-47 47-52 52-57 57-62 >62
Diameter Size Class (cm)

 

 

 

 

d: Four Burns

.2 ‘. +4BUT
2500 - ‘l‘ - - X— - '4BTH
a) .

  
     

 

,j , Era—.—

0—6 6-11 11-17 17-22 22-27 27-32 32-37 37-42 42-47 47-52 52-57 57—62 >62
Diameter Size Class (cm)

 

 

 

Thinned and unthinned (TH and UT respectively); Unburned Control (UB), One Burn (18),
Two Burns (2B), Four Burns (48).

128

Sugar maple density was highest in unburned than in other treatments throughout the
simulations and it persisted until the end of no-management, at least in the unburned
treatments. It density was highest in the four-burn treatments at the end of thinning. Black
and choke cherry, white ash, beech and other trees disappeared or almost disappeared
after the year 2044 under no-management (Table 3.2), but they persisted at relatively low

densities after thinning.

3.3.3.1 Understory woody and overstory species

In the beginning of the simulation, the density of understory red pine was lower
than other understory species (Figure 3.4a) but all species had much higher densities in
burned plots than in unburned controls. Understory red pine density declined to almost
nothing—except in the unburned controls—at the end of no-management (Figure3.4b)
due to ingrowth to higher dcl and mortality. A similar sharp decline occurred in all other

species over the course of the simulation.

The higher dominance of other understory species over red pine was apparent in
the frequently burned treatments compared to the unburned controls at the end of the
thinning simulation, despite their overall decline after 2044. At the end of thinning, the
four-bum plots had the highest density of woody understory red pines, but the decline
from initial conditions again was abrupt (Figure 3.40). Understory red pine densities in
two and four-bum plots were much higher than that in similar plots under no-

management.

129

There was not much influence on mid-canopy and dominant overstory species
density during both management regimes. Ingrowth into the sub-canopy by red pine trees
occurred in all of the treatments during both management simulations (Figure 3.5a &
3.5b). Sub-canopy red pine density was highest in unburned controls in the beginning;
however, at the end of no-management it was highest in the bumed plots. Growth into the
sub-canopy by red pine in the four-burn treatments was remarkable as there were no trees
of this size class at the beginning of the simulation. Other species were highest in
unburned treatments throughout. Sub-canopy hardwood species increased somewhat in

all of the burned plots but their relative density actually decreased.

3.3.4 Basal area (BA)

The standing total BA at the outset ranged from 17.1 to 24.1 m2 ha" with the
lowest in one-burn and the highest in four-bum treatments (Appendix 3.3b). During no-
management, the overall BA increased rapidly until about the year 2044, and thereafter
grew slowly and formed a plateau until the end (Figure 3.6; Appendix 3.3b). However,
BA formed the classical zigzag, saw-tooth shape during successive thinnings and was

always lower than no-management.

3.3.4.1 Understory basal area
Although their density was high (Table 3.1) understory trees formed only 19, 12, 6 and 3
% of the total understory BA in unburned, one-, two- and four-burn treatments,

respectively, and the remaining portion constituted the overstory trees (Table 3.3).

130

Table 3. 2: Average tree density“ (trees ha'l) and relative density (%) of all species under different
management simulations” for 100 years in mature red pine stand.
Treatments Year AB % BC % CC % Others % RM % RP % SM % WA % Totafl

 

 

 

Beginning

Control 2003 292 3 583 6 896 9 1735 17 1290 13 3930 39 940 9 400 4 10067
OneBum 788 4 1013 5 2381 13 2598 14 4007 21 5948 31 827 4 1351 7 18913
Two Burns 701 4 1551 8 2673 13 2235 11 2886 14 9163 46 430 2 296 1 19934
Four Burns 321 2 2097 12 2247 13 2721 16 2143 13 5726 34 896 5 727 4 16879
No—Management

Control 2014 268 3 401 5 708 8 1396 16 1058 12 3794 43 864 10 314 4 8803
One Burn 695 4 771 5 1962 12 2026 12 3269 20 5611 34 767 5 1173 7 16273
Two Bums 599 4 526 3 2023 13 1649 10 2087 13 8630 54 377 2 176 1 16067
FourBurns 267 2 1475 10 1774 13 2148 15 1531 11 5526 39 794 6 621 4 14136
Control 2044 66 4 0 0 23 2 62 4 184 12 880 58 292 19 12 1 1519
One Burn 306 7 157 4 901 20 631 14 624 14 1286 29 385 9 185 4 4476
TwoBums 220 7 48 2 138 4 410 13 357 12 1711 55 183 6 35 1 3103
Four Burns 54 1 273 7 634 17 580 15 196 5 1488 40 415 11 123 3 3763
Control 2074 19 3 0 0 0 0 5 1 77 11 487 72 87 13 0 0 675
One Burn 41 5 8 1 0 0 6 l 64 7 774 86 5 1 2 0 899
Two Burns 28 4 0 0 0 0 l3 2 6 l 677 93 7 1 0 0 731
Four Burns 0 0 0 0 0 0 4 1 0 0 573 99 0 0 1 0 579
Control 2104 11 2 0 0 0 0 3 1 40 8 380 79 49 10 0 O 483
One Burn 13 2 2 0 0 0 4 1 17 3 575 94 1 0 0 0 612
Two Burns 18 3 O 0 0 0 11 2 4 l 508 93 4 l 0 0 544
Four Burns 0 0 0 0 0 0 3 1 0 0 449 99 0 0 0 0 453
Thinnings

Control 2014 251 3 397 5 700 8 1345 16 1037 12 3778 44 791 9 298 3 8596
One Burn 695 4 771 5 1962 12 2026 12 3269 20 5611 34 767 5 1173 7 16273
Two Burns 549 3 514 3 2000 13 1569 10 2002 13 8611 55 337 2 167 1 15750
Four Burns 251 2 1321 10 1700 13 1982 15 1504 11 5429 40 676 5 600 4 13463
Control 2044 84 4 6 0 76 3 168 7 195 8 1482 63 291 12 37 2 2340
One Burn 247 6 190 5 779 19 620 15 765 19 915 22 319 8 250 6 4084
Two Burns 210 6 85 2 507 14 514 14 594 16 1546 42 167 5 58 2 3681
Four Burns 48 l 500 14 539 15 637 18 277 8 1047 29 378 10 174 5 3600
Control 2074 41 5 2 0 28 3 63 8 129 16 356 44 182 22 17 2 819
OneBum 113 9 36 3 171 14 114 10 155 13 462 39 110 9 30 3 1192
Two Burns 133 9 46 3 79 5 257 17 334 22 500 33 111 7 33 2 1492
Four Burns 33 2 293 14 284 14 371 18 112 5 634 31 252 12 72 4 2051
Control 2104 23 5 0 0 0 0 5 1 84 17 262 54 109 22 2 0 484
OneBurn 65 11 12 2 0 0 9 1 61 11 390 67 42 7 4 1 583
Two Burns 114 11 25 2 4O 4 144 14 187 18 418 40 91 9 18 2 1037
Four Burris 22 2 148 13 67 6 172 15 20 2 515 46 167 15 16 1 1128

 

 

 

Note: % Tot (in columns) are the percent of total density.

Some of the 'zeros' in % Tot, column have values below 0.5% and have been rounded to zero.
*Stand density does not include removals in thinnings.

”Simulation started on year-2004.

The first, second and third thinnings occurred in 2004, 2034, and 2064.

131

Figure 3.4: Woody understory (0-11 cm dcl) red pine and other species combined
density in different burn treatments and management simulations‘ for 100 years in a
mature red pine stand.

 

   
 
 
 

 

 

 

 

 

 

 

     
 
    

 

 

 

 

 

 
 
   

 

 

 

 

 

a: Beginning - Year 2003
14000 ~ 12964
12000 « 11146
2
2 10000 . UControl
g 8000 7 BOne Burn
E 6000 - aTwo Burns
2 4000 4 IFour Burns
G 2000 4
O 5
Red pine Others
Species
b: No-management-Year 2104
700 -
m 600 <
S
3 500 I DControI
g 400 . E-lOne Burn
E 300 ; BTwo Burns
g 200 0 IFour Burns
0
100 ~ 31 9 1 7 14 10 30 0
0 [—L— 1 m 1
Red pine Others
Species
c: Thinnings - Year 2104
700 —
594 601
600 4
N
’5 500 ~
3 El Control
g 400 “ EOne Burn
"'1" 300 - aTwo Burns
§
0, I Four Burns
5 200 -
D
100 -
0 _ .........
Red pine Others
Species

 

 

 

"First, second and third thinnings were done in years - 2004, 2034 and 2064, respectively.

132

Initially all burned treatments had about one-half or less understory BA than in unburned
controls; however, later understory BA decreased almost to nothing in all of the

treatments at the end of both management regimes.

3.3.4.2 Overstory basal area

Unlike the understory, the overstory BA doubled or tripled at the end of both
management regimes (Table 3.3). Burned plots were higher in their average BA than the
unburned treatments after both managements. Relative basal area (RBA) of the overstory
reached 100 % in all of the treatments after no-management, whereas it was only slightly

lower after thinning.

Among the sub-canopy, mid-canopy and dominant overstory trees the latter
constituted the highest BA and relative BA of total overstory throughout the simulations
(Table 3.4). Although considerable growth in the BA of the dominants occurred, their
relative BA changed little or declined. The dynamics of the sub- and mid-canopy were
different than the dominants. Under no-management major BA ingrth and grth
occurred in the sub-canopy, with less in the mid-canopy. When thinning was applied
major growth and ingrth occurred in both the sub— and mid-canopy. Burning

treatments had little effect on BA.

Red pine had a very high dominance over other species. Its relative BA ranged

from 76% in unburned treatments to 90 to 97% in burn treatments, with the highest value

133

Figure 3.5: Average sub-canopy (>1 1-32 cm dcl) red pine and other species combined
denisty in different burn treatments and management simulaitions for 100 years in
mature red pine stands.

 

a: Beginning - Year 2003

 

 

 

N 62

.c:

I—

8- E1 Control

§ 13 One Burn
E I Two Burns
"2‘ 6 I Four Burns
a 0 0 0

. L

 

 

Species

 

 

b: No—management-end of simulation-Year 2104

   

 

 

 

 

 

 

 

 

 

 

 

 

 

424
450
.E: 400 372
8. 338 13 Control
g 250 E3 One Burn
E; figg I Two Burns
:2 100 64 27 I Four Burns
50 6 3
c3 0 . | |.-—1 1
Red pine Others
Species
c: Thinnings-end of simulation-Year 2104
N
J:
a 13 Control
g One Burn
1; I Two Burns
g I Four Burns
3
'50 Red pine Others
Species

 

 

 

‘First, second and third thinnings were done in years - 2004, 2034 and 2064, respectively.

134

Figure 3.6: Basal area in different treatments under no—management and thinnings

with residual basal area set at 22.9, 27.5 & 34.4 m2 ha'l at 30 years cycle for 100
of years of similation in mature red pine stand.

 

55 1 _ ’- — :1. i
1” -
50 ~ 3’ o
0; 5°, A
' ' o ,A
45 — :13;
40 ~ 8:
0

BA - sq m/ ha
b)
M

O.)
O
1

25~ ,1. ‘0‘

 

 

~.,
/:g

20~

1. CE. .- . . . . . . ..

2003 2004 2014 2024 2034 2044 2054 2064 2074 2084 2094 2104

Simulation running periods (Year)

—I— Four Burns Unthin —0— Two Burns Unthin + One Burn Unthin
+ Unbumed Unthin - - O - - Four Burns Thin - - 0 - - Two Burns Thin
--0--OneBurnThin ---A---UnburnedThin

Note: 1, 2 and 3 arrows indicate first, second and third thinnings, respectively,
scheduled on years-2004, 2034 and 2064. Note that the "one burn thin" treatment

was below the target basal area for the first thinning (22.9 m2 ha'l), so it could
not be thinned.

135

Table 3.3: Average and relative live understory (0-11 cms dcl) and overstory (>1 1 cm
dcl) basal area under 100 years of simulation“ in mature red pine stands.

 

 

 

 

 

 

Understory-BA Overstory-BA
m2 ha_l m2 ha—l
Burn treatments
L Year C 1B 28 48 C 18 ZB 4B
Beginning
2003 4.2 2.0 1.2 0.8 18.0 15.1 21.2 23.3
% 19 12 6 3 81 88 94 97
No-management
2014 5.4 8.5 5.5 4.5 25.8 19.4 27.1 28.2
% 17 31 17 14 83 69 83 86
2044 2.6 5.3 3.3 2.2 46.3 44.7 50.0 51.0
% 5 11 6 4 95 89 94 96
2074 0.5 0.8 0.5 0.4 50.8 52.8 54.0 54.3
% 1 1 l l 99 99 99 99
2104 0.1 0.1 0.2 0.0 52.3 54.3 54.6 54.9
% 0 0 0 0 100 100 100 100
Thinnings
2014 5.3 8.5 5.5 4.5 24.8 19.4 25.1 24.8
% 18 31 18 15 82 69 82 85
2044 2.7 4.0 3.0 1.7 32.7 33.1 32.3 32.3
% 8 1 1 8 5 92 89 92 95
2074 0.7 1.1 1.7 2.7 38.9 39.3 37.4 37.2
% 2 3 4 7 98 97 96 93
2104 0.4 0.4 2.0 1.8 46.7 51.7 48.2 50.3
% 1 1 4 4 99 99 96 96

 

 

Note: Some of the 'zeros' in % column have values below 0.5% and have been rounded
to zero.

Burn treatments in columns are: C, Unbumed Control; 13, One Burn; 28. Two Burns;
48, Four Burns

% (in rows) is the percent of total basal area.

dcl=diameter size class.

*Stand basal area does not include removals in thinnings.

I""Simulation started on year-2004.

The first, second and third thinnings occurred in 2004, 2034, and 2064.

in four-bum treatments (Table 3.5). Other species contribution to BA was low and
relative BA was highest in unburned treatments and consisted mainly of red maple (13%
RBA) and sugar maple (8%). Red pine BA more than doubled in all of the treatments
after both managements (Table 3.5). Red pine BA always was higher after burning, and

increased with the frequency of burning.

136

Table 3.4: Average and relative sub—campy (>11-32 cm dcl), mid-campy (32 8. 42
cm dcl) and dominant overstory (>42 cm dcl) basal area" under 100 years of
simulations" in mature red pine stands.

 

 

 

Sub-canopy Mid-canopy Dominant overstory

2 -1 2 -1 2 -1

m ha m ha m ha

Burn treatments
Year C 18 28 4B C 18 28 48 C 18 28 48
Beginning

2003 2.3 0.8 0.4 0.2 4.0 4.4 5.9 7.3 11.7 9914.9 15.9
°/o 1O 4 2 1 18 26 26 30 53 58 66 66

 

No-management
2014 6.7 1.3 0.7 0.3 1.9 2.0 2.9 2.1 17.216.1234 25.8

 

 

% 21 5 2 1 6 7 9 6 55 58 72 79
2044 18.0 20.6 15.9 14.3 2.9 0.1 0.7 0.9 25.4 24.0 33.4 35.8
°/o 37 41 3O 27 6 O 1 2 52 48 63 67
2074 16.9 23.1 19.9 16.7 4.0 4.4 2.2 2.1 29.8 25.3 31.9 35.5
°/o 33 43 37 31 8 8 4 4 58 47 59 65
2104 14.1 20.5 20.1 11.3 6.6 7.5 5.2 11.2 31.6 26.3 29.3 32.5
% 27 38 37 21 13 14 9 20 60 48 54 59
Thinnings
2014 6.4 1.3 0.7 0.3 1.8 2.0 2.5 1.7 16.6 16.1 22.0 22.8
% 21 5 2 1 6 7 8 6 55 58 72 78
2044 14.9 17.2 12.1 10.3 1.9 0.1 0.4 0.5 15.9 15.8 19.8 21.4
°/o 42 46 34 30 5 0 1 1 45 43 56 63
2074 16.3 22.1 16.9 11.0 6.4 3.5 2.9 4.7 16.2 13.8 17.7 21.5
°/o 41 55 43 28 16 9 7 12 41 34 45 54
2104 7.8 14.6 13.0 8.4 13.2 18.5 13.8 11.7 25.8 18.6 21.4 30.1
% 16 28 26 16 28 36 28 22 55 36 43 58

 

Note: Some of the 'zeros' in % column have values below 0.5% and have been
rounded up to zero.

Burn Treatments columns are: C, Unbumed Control; 18, One Burn; 28, Two Burns;
48, Four Burns.

°/o (in rows) is the percent of total basal area.

dcl=diameter size class.

*Stand basal area does not include removals in thinnings.

"Simulation started on year-2004.

The first, second and third thinnings occurred in 2004, 2034, and 2064.

137

 

Table 3.5: Average and relative basal area of red pine and other species" combined

under different management simulations“ for 100 years years in mature red pine stand.

 

 

 

 

 

Red pine % Others* % Total
Treatments Year m2 ha.l Tot m2 ha.l Tot m2 ha.l
Beginning
Control 2003 16.9 76 5 24 22.2
One Burn 15.3 90 2 10 17.1
Two Burns 21.5 96 1 4 22.5
Four Burns 23.4 97 1 3 24.1
No-management
Control 2014 22.2 71 9 29 31.2
One Burn 24.2 87 4 13 27.9
Two Burns 31.1 95 l 5 32.6
Four Burns 31.6 97 1 3 32.7
Control 2044 38.4 79 10 21 48.9
One Burn 43.2 86 7 14 50.]
Two Burns 51.3 96 2 4 53.3
Four Burns 51.4 97 2 3 53.2
Control 2074 43.7 85 8 15 51.3
One Burn 50.8 95 3 5 53.6
Two Burns 53.9 99 l 1 54.5
Four Burns 54.5 100 0 0 54.7
Control 2104 46.7 89 6 l 1 52.4
One Burn 53.2 98 l 2 54.3
Two Burns 54.3 99 l 1 54.8
Four Burns 54.8 100 0 0 54.9
Thinnings
Control 2014 21.4 71 9 29 30.1
One Burn 24.2 87 4 13 27.9
Two Burns 29.2 95 l 5 30.6
Four Burns 28.3 97 l 3 29.3
Control 2044 25.5 72 10 28 35.4
One Burn 31.1 84 6 16 37.1
Two Burns 33.1 94 2 6 35.3
Four Burns 32.4 95 2 5 34.0
Control 2074 27.7 70 12 30 39.6
One Burn 34.3 85 6 15 40.4
Two Burns 36.1 92 3 8 39.2
Four Burns 37.3 94 3 6 39.8
Control 2104 34.2 73 13 27 47.2
One Burn 46.8 90 5 10 52.1
Two Burns 46.0 92 4 8 50.2
Four Burns 49.2 94 3 6 52.1

 

 

Note: Some of the 'zeros' in % column have values below 0.5% and have been

rounded to zero.
*Consists of red maple, sugar maple, American beech, black cherry, choke cherry,

white ash, non-commercial species, and other minor species.
“Stand basal area does not include removals after thinnings.
% Tot (in columns) is the percent of total basal area.

The first, second and third thinnings occurred in 2004, 2034, and 2064.

138

 

Importance values (IV) also reflect the dominance of red pine in my stand, both at
the beginning and end of the simulations (Table 3.6). The importance of hardwoods
declined under no-management but changed little after thinning. Burning increased IV of

red pine, especially under no-management.

Table 3.6: Importance ('IV) of all species under no—mangement and thinning
simulations“ run for 100 ears in differently burned ma_ture red pine.

 

 

 

 

 

Species'“

War AB BC CC RM RP SM WA
Control 2003 2 3 4 13 58 9 3
One Burn 3 3 6 13 61 3 4
Two Burn 2 4 7 8 71 1 1
Four Burn 1 6 7 6 65 3 3
No-management
Control 2104 1 0 0 7 84 7 0
One Burn 1 0 0 2 96 0 0
Two Burn 2 0 0 1 96 0 0
Four Burn 0 0 0 O 99 0 0
Thinnings
Control 2104 3 0 0 17 63 16 0
One Burn 6 2 0 8 78 4 1
Two Burn 6 1 2 11 66 5 2
Four Burn 1 7 1 70 2

 

 

 

 

8
*IV is the average of relative density and relative basal area of all trees but does not
include removals in thinnings.
“Simulations were started on year 2004.
Non-commercial and other minor species which form a total of <10% are not shown
in the column.
Note: Some of the 'zeros' in columns are values 50.5% and have been rounded to
zero.
***AB=American beech, BC=black cherry, CC=choke cherry, RM=red maple, RP=red
pine, SM=sugar maple, WA=white ash.

3.3.5 Mean annual increment (MAI)
In the beginning of the simulation, mean annual increment (MAI) was highest in
four-bum followed by two-bum, unburned controls and one-burn treatments (Figure 3.7).

Without any management, MAI increased slowly until 2044 and then declined.

139

Differences in MAI among treatments were less at the end of the simulation than at the
beginning. At the end of thinning, MAI was higher than under no-management and
peaked between 2084 and 2094, much later than the unthinned regime (Figure 3.7). As

under no-management, MAI was highest in the four-burn treatments than all others

throughout the simulation period.

Figure 3.7: Mean annual increment (MAI) in different burn treatments under
no—management and thinnings‘ for 100 years of simulation in mature red pine stand

4.5. ”DMD

 

MAI - cu m/ha/yr

 

 

2.5 .
2.0 T I I I I T T 7 T T T 1
2003 2004 2014 2024 2034 2044 2054 2064 2074 2084 2094 2104
Simulation mnning periods
- - D - - Four Burns Thin - - o - - Two Burns Thin - - a - - Unbumed Thin
- - 0 - - One Burn Thin + Four Burns Unthin +Two Burns Unthin

+ Unbumed Unthin +One Burn Unthin

' F irst, second and third thinnings were scheduled on years - 2004, 2034 and 2064,
respectively.

140

3.3.6 Sawlog volume

The total standing sawlog volume at the beginning of the simulation was highest
in the four-burn plots followed by the two-burn, unburned, and one-bum treatments in
decreasing order (Figure 3.8; Appendix 3.3c). In the beginning, most of the sawlog
volume was contributed by dominant overstory trees, with less in the mid-canopy and

virtually nothing in the sub-canopy (Table 3.7).

Table 3.7: Average and relative standing“ sawlog volume of sub-canopy (>11-32 cm
dcl), mid-canopy (32-42 cm dcl) and dominant overstory (>42 cm dcl) trees under
100 years of no-management and thinning simulation“ in mature red pine stands.

 

 

 

 

 

 

 

Subocanopy Mid-canopy Dominant overstory
m3 ha'1 1113 ha'1 rn3 ha'1
Burn treatments
Year C 18 28 48 C 18 28 48 C 18 28 48
Beginning
2003 1 1 38 39 51 66 114 95 140 154
% 1 1 O 0 25 29 26 30 75 70 73 7O
No-management
2014 3 5 1 0 17 19 26 19 181 169 243 267
% 2 3 0 0 8 10 1O 7 90 88 90 93
2044 8 10 5 9 22 1 4 8 299 281 392 418
% 3 3 1 2 7 0 1 2 91 96 98 96
2074 22 17 20 36 21 27 14 11 352 310 385 428
% 5 5 5 8 5 8 3 2 89 88 92 90
2104 29 22 38 20 33 41 28 58 362 313 350 396
% 7 6 9 4 8 11 7 12 85 83 84 84
Thinnings
2014 3 5 1 O 16 19 23 15 175 169 228 237
°/o 2 3 0 0 8 10 9 6 90 88 91 94
2044 12 9 3 8 14 1 3 4 188 186 233 251
% 6 5 1 3 7 0 1 2 88 95 97 95
2074 30 28 29 26 34 22 17 27 189 170 218 265
% 12 13 11 8 13 10 7 9 75 77 82 83
2104 12 29 26 14 66 81 72 66 266 216 257 348
% 4 9 7 3 19 25 20 15 77 66 73 81

 

 

Burn Treatments columns are: C, Unbumed Control; 18, One Burn; 28, Two Burns;
48, Four Burns

°/o (in rows) is the percent of total sawlog volume.

*Standing volume does not include removals in thinning.

“Simulation started on year-2004.

The first, second and third thinnings occurred in 2004, 2034, and 2064.

141

During no-management and thinning, standing sawlog volume increased
gradually towards the end; however, with a flattening curve in all treatments after year
2054. Total standing sawlog volume was slightly lower after thinning by 45 to 80 m3 ha'l
in all of the treatments compared to no-management (Table 3.7; Appendix 3.3c).
However, total cumulative volume (including thinning removals) was much higher in the
thinned than unthinned regimes at the end of the simulation, with no decline in
production (Figures 3.8 and 3.9). Among the burn treatments, the four-burn plots
continually produced the highest cumulative sawlog volume, whereas the once-burned
treatment produced the least (Figure 3.10). The relative effects of burns on sawlog
production declined as simulation progressed for both managements. Differences among
different burns in cumulative sawlog production were higher after thinning than no-

management, especially in four-bum treatments.

At the end of both managements, the proportional change in relative volume was
substantial in sub-canopy trees but they never constituted more than 13% of the standing
volume. Even the four-burn treatments, which did not have any live sawlog in sub-
canopy stratum in the beginning increased to 3 to 4% of total sawlog by the end of both
managements. The proportion of total volume in mid-canopy overstory trees dropped
somewhat during the simulation, especially under no-management where 84% of the
volume was in the dominant overstory. Almost no change occurred in the relative
standing volume of the dominant overstory in the thinning regimes but it increased under

no-management.

142

Red pine was the major contributor to sawlog volume (96% to 100%) both in the

beginning and at the end of simulation in most of the burning management regimes.

However, at the end of thinning, the volume in red pine in the unburned controls had

declined to 84% (Appendix 3.4), with the remaining volume made up by red and sugar

maple.

Figure 3.8: Average cumulative' sawlog volume in different treatments under
no—management and thinnings" for 100 years of simulation in mature red pine stand.

Sawlog-cu m/ha

700

600

500

400

300

200

100

- - D - - Four Bums Thin
- - O - ~One Burn Thin
+ Unbumed Unthin

,13
Do
Do
a '0
13‘ “0
.' ,- ,A
x0 ,4"
fl 0' '0' o
.9 .5 o—"
o .‘A' O
'12! “‘
o 0' A
an .‘o A o
o " 0 '0'
I‘ "o '0'
. 'o'
I I 1 I I I I T r I I

 

 

I

2003 2004 2014 2024 2034 2044 2054 2064 2074 2084 2094 2104

Simulation running periods (Year)

- - 'A- - - Unbumed Thin
—O—Two Burns Unthin

- - o - -Two Burns Thin
—I—— Four Bums Unthin
+One Bum Unthin

‘Includes the volume removed in thinnings.
“ First, second and third thinnings were scheduled on years - 2004, 2034 and 2064,

respectively.

143

Figure 3.9: Percent change of cumulative" sawlog production after
thinnings over no-management in different burn treatments for 100
years of simulation in mature red pine.

29

 

Percent Change-(%)
(a) 45 0|
0 O O

 

 

 

 

 

 

 

 

lCIControl 1:] One Burn EmTwo Burns I Four Burn_s

 

 

‘Includes volume removed in thinnings.

Figure 3.10: Percent change of cumulative“ sawlog production under different burn treatments
and management over unburned controls run for 100 years in mature red pine stands.

 

 

501 44

—l N (A) A
O O O O
‘ 1

Percent Change-(%)

 

 

 

lQOne Burn ITwo Burns I Four Burnfl

 

 

 

 

*Includes volume removed in thinnings.

144

 

3.4 Discussion

The inequality in size classes in the beginning of the simulation due to dense
recruitment is reflected in the stratified distribution of trees in all of the burned and
unburned treatments and is typical of mature and productive forests that are regenerating
and growing aggressively (Feet and Christensen 1987, Leak et al. 1995). The extremely
high tree density observed in all of the burned plots was due to newly established post-
fire regeneration, which caused variation in size classes and a highly left-skewed

distribution.

At the start of and during no-management simulations, self-thinning and
mortality, occurring mainly in the smaller understory layer, resulted in the decline in size
class inequality (Feet and Christensen 1987). These processes in my stands match with
the stand initiation and stem exclusion phases of stand development (Bormann and Liken
1979, Feet and Christensen 1987). Mortality occurs after crown closure, as the
suppressed trees cannot compete for light, water, growing space or soil nutrients (Peet
and Christensen 1987). Johnstone et al. (2004), found aspen and lodgepole pine to decline
due to self-thinning and mortality a decade after burning in a boreal system in Yukon
British Columbia, Canada. The canopy closure timing matches with my study site,
inventoried 10 to 18 years after prescribed burns were carried out; thus self-thinning and

mortality were already occurring.

Mortality caused a sharp decline in small-sized understory trees in all treatments

under no-management. At the end of no-management, slower understory growth and

145

reduced ingrowth to sub-canopy trees resulted in the lowest overstory tree density in
unburned controls. One- or two- burn treatments are sufficient to kill unwanted
vegetation and decrease competition, resulting in faster growth as well as higher ingrowth
(shifting) of large understory to sub-canopy trees compared to unburned and four-burn
treatments, both of which had lower initial sub-canopy densities. Four-bum treatments
completely lacked large-sized understory trees. Single and multiple fires reduced 60 to
80% of 2 to 10 cm size class (i.e., large-sized understory) fire sensitive species in oak
forests of the Cumberland Plateau (Arthur et al. 1998). Species richness of large woody
understory species in pine understories was reduced immediately by even a single burn
(Henning and Dickmann 1996, Neumann and Dickmann 2001). At the end of no-
management, destruction of the sub-canopy trees during repeated burnings lowered the
sub-canopy density and BA the most in the four-bum treatments. Low-intensity fires
instantly kill red and white pines below 15 cm diameter class or 6 m height (McConkey
and Gedney 1951, Van Wagner1970). Whereas the pines <18 m tall are susceptible to
fires (Henning and Dickmann 1996). Red pine trees with up to 75% of the crown
scorched survive fires (Van Wagner 1970), although some with >75% crown scorching
may die later. Mature red pines with >95% crown scorching had up to 50% mortality
(Van Wagner1970). In my four-burn treatments, repeated burnings every two years may
have increased exposure of already scorched mature but small stature (low height) trees,
leading to death of some of them. In the Pigeon River area (other than the study plots),
some mature red pine trees with >90% crown scorch even if not killed instantly, were

killed later by bark beetles.

146

As no-management simulations progressed, diameter distribution became more
normal with a shift to older age classes forming a more positively skewed distribution
than in the beginnings (Mohler et a1. 1978, Feet and Christensen 1987). All of the
treatments formed a bi-modal stand structure with higher densities in dominant overstory
classes than in the beginning, because burning, at the cost of smaller understory trees,
stimulated faster grth in larger size classes. This effect was more pronounced in

progressively burned treatments.

All treatments lost a remarkably high proportion of understory trees by the end of
no-management and according to Cullen and Leak (1988 in Leak et al. 1995) such
decline to <10% relative understory tree density would not sustain productivity. In
contrast, frequent burning plus thinning produced a relatively dense understory, which
could sustain productivity and could be the basis for a succeeding stand if the overstory
was removed. The gaps created during thinnings reduced understory mortality and
stimulated faster grth of dominant trees while burning created a favorable seed bed for
seedling recruitment. In 63-year-old unthinned pine stands mortality was ten times higher
than in similar but thinned stands (Burgess and Robinson 1998). Thinning increases
diameter growth which is greater in progressively heavier thinned plots (Buckman 1962,
Day and Rudolph 1972). Increased light level enhanced seedling survival and also their
height, diameter and volume growth after thinning young red and white pine plantations
and hardwood stands (V olk and F ahey 1994, Stiell et al. 1994, Burgess and Robinson
1998, Parker et al. 2001). They also suggested that such gaps enhance succession in

young red pine plantations.

147

The repeatedly burned treatments (two— and four-bum) responded quickly to
thinnings and they produced a higher density of large-sized understory trees than in other
treatments. Large-sized understory trees must have developed from post-fire recruitments
in repeatedly burned plots, since they lacked them in the beginning. Larger understory
trees are stronger and more vigorous (Smith 1986, Nyland 1996) and they will enhance
future productivity of forests (Leak et al. 1995) and create a more complex long-term

stand structure.

The occurrence of dense and aggressively growing understory seedlings and
saplings and their shifting to larger diameter classes, particularly in the two- and four-
bum treatments, created a tri-modal stand structure in which the overstory (>11cm dcl)
shifted to the right in progressively higher size classes after thinnings. Increase in
understory trees after thinning, however, produced less sub-canopy but more mid-canopy
trees as compared to no-management. Among different burn treatments after thinning,
four-bum treatments had little size or age class differentiation in 17-47 cm dcl as
compared to two- and one-bum treatments. This shows that four-times burning, in the
long run, created more uniformity in overstory size classes or reduced the variation
among the overstory size classes as compared to two- and one-burn treatments. One-burn
treatments had the highest sub-canopy tree density and variation in overstory tree size

classes.

Burning and thinning interactions did not produce much difference in BA among

different treatments, although four-bum plots always maintained the highest overall BA

148

for both simulations. BAs in my study plots, in the beginning, were one-half to one-third
of what has been predicted for a red pine plantation of the same age and site index by
Benzie (1977). Such a low residual BA after thinning increases the future diameter
growth of retained trees (Haberland and Wilde 1961; Lundgren 1981) but reduces the
overall BA, MAI and the ultimate volume production of red pine stand as there is vacant
space that is not utilized by the overstory (Gilmore 2005). At the end of no-management,
the overall BA produced was similar to that suggested by Burgess and Robinson (1998),
who found the BA in their 101-year-old unthinned white and red pine stands to be 51 m2
ha". Similarly, in another study the BA in unthinned stands was about 52.7 m2 ha'l,
varying, however, with mortality (Spurr et al. 1957). Yet, my study plots were much
older (173-year-old) at the end than the stands in these examples. The F VS predicted BAs
were low because a low residual BA (~ 10 m2 ha") was retained during earlier thinnings
carried out in the year 1970 (Henning and Dickmann 1996), i.e., long before the current
prescribed bum study started. At a given age and SI, BA growth will be about the same
(Buckman 1962), whereas at mature age grth is affected by SI, tree age and density

(Lundgren 1981 ).

Partial or complete killing of understory vegetation and sub-canopy trees due to
repeated burnings reduced competition and enhanced diameter growth of mature
overstory trees, which increased MAI and sawlog volume. However, the effects were not
great. During no management, the differences both in MAI and sawlog volume between
treatments started narrowing after the decline of overall tree productivity at age 113

years, which matches closely to the grth culmination age predicted by Lundgren

149

(1981). He showed that production levels off when plantations exceeded 3954 tpha. If
stands are not thinned for a long period both the net and gross volume production decline

due to self-thinning and mortality (Coffmann 1976).

MAI after thinning in four-burned treatments started picking up and increased up
to year 2094 (age 163) before it declined after thinning. F our-burn treatments always had
highest growth; however, they started higher also. Despite starting higher, there was still
somewhat higher grth in four—bum treatments as compared to other treatments,
especially after the second thinning. In addition to the effects of burning, the gaps created
by removal of large overstory trees and the cutting cycle during thinning provided
additional space and resources (light, moisture, nutrients, and space) to the residual trees
increasing their growth and vigor. The result is consistent with one of the major
objectives of thinning, which shortens the rotation age (Lundgren 1983); i.e., intended
volume can be obtained earlier than in unthinned stands. In my study, the continued
growth of red pine stands up to 163 years was because of the actively growing young
post-fire recruitments that were released after thinning. Thus, the effects of multiple
(frequent) burnings lasted long for MAI and sawlog production after thinning. These

results have important implications for the management of red pine forests.

At the end of my study, percent change of cumulative sawlog volume over
unburned control declined as simulation progressed during both managements in
repeatedly burned plots; however, this effect was not pronounced if thinning occurred. A

13% higher percent change of cumulative volume over no-management in four- and two-

150

burn treatments than in unburned controls after thinning show that there was a
remarkable increase in cumulative sawlog volume in repeatedly burned treatments, even
though they had higher values to start with. The multiple burned plots acquired higher
growth vigor from earlier burnings carried out one to almost two decades before the
measurements were taken for this study. Removal of dominant overstory trees increases
radial increment of the retained overstory trees, resulting in the increase of net volume
and yield (Assman 1970, Nyland 1996). Removal also lowered the total standing
overstory sawlog volume compared to no-management; however, when the volume
removed during thinnings was included there was a net increase in cumulative sawlog
volume (cf. Coffmann 1976). The slight decrease in relative volume of red pine in burned
plots after thinning as compared to no-management, as well as the higher decrease in
unburned controls when thinned were due to an increase of hardwood volume, especially

red and sugar maple.

Red pine dominance increased in burned treatments as the shade intolerant
species—black cherry, choke cherry, white ash and other minor hardwood species—
declined gradually and disappeared during no-management simulations. These findings
are backed up by the increase in IV of red pine in all treatments, especially the burned
ones. Bormann and Likens (1979) also suggested such a shift in relative importance (IV)
among species due to mortality of those suppressed ones during self-thinning. Thus, if
unmanaged for a long time, mature red pine stands lose hardwood species, especially
shade intolerants, and become low in species richness. There was virtually no new

regeneration in the no-management treatments, as post-fire recruitment depends on time

151

elapsed after fire. For example, it occurred in first five years after fire in boreal forests in
British Columbia, Canada (Johnstone et al. 2004). Mineral soil gradually becomes
covered with re-growth and litter deposits that prevent seeds from germinating and
establishing (Metheven and Murray 1974). Red and white pines and hardwood species
(except birches and black cherry) failed to establish under different managements,
including clearcutting (McRae 1994, Smith and Ashton 1993). However, burning and
thinning reduced depth of organic matter, exposed mineral soil and increased pine
seedling establishment (Little and Moore 1949, Buel and Cantlon 1953, Thomas and

Wein 1985, Mallik and Roberts 1994, Duvall and Grigall 1999).

Despite lack of new recruitments in most treatments, new red pine and hardwood
recruits established in the understory in four-burn plots after thinning. Other hardwood
species recruits were also very high compared to the red pine in two-burn plots.
Relatively more shade tolerant red and sugar maple recruit in smaller gaps (Abrams 1982,
Arthur et al. 1998). Smaller gaps are covered within two to three years of their formation,
whereas larger gaps increase species richness and biodiversity (Parker et al. 2001) and
such gaps were created during successive thinnings of burned treatments in this study.
The new recruits reveal that the forest floor in two- and four-times burned treatments
remained relatively more suitable for germination and establishment for a longer period
than other treatments. Single and multiple burns have been suggested for successful
eradication or control of unwanted understory vegetation such as balsam fir and hazel—nut
in red pine plantations as well as pine seedling germination and establishment (Van

Wagner 1970, Alban 1977). Further, reduction in mortality after thinning reduces the

152

coarse woody debris in the young stands and might facilitate seeds of some species to
reach soil (Duvall and Grigall 1999). Three times thinning of red and white pine stands
increased recruitment of understory shrubs and other species (Bender et al. 1997, Burgess
and Robinson 1998). Besides increase in IV of red maple and sugar maple, beech and
cherries also persisted in my study. Volk and Fahey (1994) also found that black cherry
(a short lived shade intolerant species) had higher mortality in the unthinned than in the

thinned Alleghany hardwood forest.

Dense understory seedlings and their mortality resulted in higher red pine sub-
canopy density, relative density and faster growth of residual trees, especially in one- and
two-burn treatments after no-management compared to thinning. However, removal of
trees after thinning reduced competition and enabled residual large-sized trees of other
sub-canopy species to grow by more than two-fold in burned plots, compared to no-
management. Increase in IV of other hardwood species after thinning compared to no—
management has important ecological and management implications. Increase in
hardwood species provides better habitat, cover, mast and browse for wild animals, as
well as hunting and recreational scope (Little and Moore 1949, Rouse 1988, Dickmann

1993, Bender et al. 1997), in addition to biodiversity benefits.

Comparison between the total merchantable volume production by FVS in this
study and that predicted by Benzie (1977) for similar age and SI shows that FVS
predicted volume was slightly lower than that of Benzie in the beginning of the

simulation (Appendix 5.5). At age 143 the standing merchantable volume change

153

obtained for no-management was within a range of -12 to +12% compared to Benzie’s
values in all treatments, with the highest in four-burn and lowest in one-bum treatments.
However, the cumulative volume predicted was much higher than Benzie for similar age
and SI after thinning, with the highest in four-burn and the lowest in one-bum treatments.
These predictions would actually be even higher if trees removed during thinning in 1970
(Hennings and Dickmann 1987) had been included in the F VS projections. In this study,
the high variation in volume predicted by FVS could also be because the FVS model was
developed for young red pine plantations or natural stands without disturbance, whereas I
used it to simulate differently burned plots. However, considering such a very long
projection period of FVS simulation (100 years), the result could still be considered

useful in red pine management.

154

3.5 Summary

Mature red pine stands re-measured one to two decades after prescribed burnings

(one-, two- and four-times burned) and projected through FVS for 100 years with and

without a series of thinnings showed unique differences in vegetative and growth

responses. The long-term effects depended on number of burnings and thinnings and their

combinations and also the past history of disturbances. Stand structure, species

composition, density, BA and volume changed differently among burned and unburned

treatments as simulations progressed.

1.

Extremely dense post-fire regeneration and inequality in size-classes at the outset
showed increased productivity of the burned stands, but the regeneration
experienced high mortality compared to unburned controls.

Decline in density started at the time of re-measurement of the stands, i.e., 10-18
years after burning, and was due to self-thinning mortality; however, a sharp
decline occurred five to six decades after burnings.

Lower mortality, slower grth and reduced ingrowth occurred in unburned
controls. However, unburned plots retained the highest small understory and
lowest overstory density.

High red pine sub-canopy density produced by one- and two-times burnings at the
end of no-management indicate that fires were sufficient to kill competing low
vegetation, including some of the sub-canopy red pines in the beginning, and

stimulated growth of small diameter red pines for a long period.

155

10.

11.

Low sub-canopy density due to fire-caused mortality and increase in mid-canopy
trees due to ingrth in four-bum treatments indicate that fire effects were long-
term and still apparent after a century.

Thinnings slowed down mortality and enhanced recruitment of understory trees in
burned plots, resulting in an increase in an inequality in size classes, in contrast to
decrease in inequality during no-management.

Two- and four-burn treatments produced dense, aggressively growing understory
seedling and saplings after thinning, and such repeated burns were more
productive.

A tri-modal stand structure developed in progressively bumed plots due to grth
of larger trees as compared to the bimodal stand structure after no—management.

F our-times burning created a more uniform structure across size classes in sub-
and mid-canopy layers (17-47 cm dcl) as compared to two- and one-burn
treatments.

MAI in repeatedly burned treatments, particularly the four-burn, was greater and
faster for both managements, indicating stimulated growth both in the understory
and overstory. MAI declined at age 113 in unthinned simulations but not until age
163 in thinned simulations. In both cases, the growth culmination age was much
greater than normal for red pine and was due to both fast-growing, dense post-fire
understory after burning and their release after thinning.

Four-bum treatments, particularly after the second thinning, produced highest

cumulative volume although they started higher in the beginning. Also, higher

156

percent change of volume production in burned treatments over no burning after
thinning reveals that the effects of burning can be extended for a long period.

12. Red pine dominance was highest in burned plots throughout both simulations,
followed by red and sugar maple. The latter two species were better represented in
unburned controls compared to burned treatments at the end of no-management.

13. Disappearance of most hardwood species (except red maple, sugar maple and
American beech) and occurrence of few new recruits regardless of burnings by
the end of no-management indicates that species diversity declined as succession
progressed. In contrast, repeated burnings followed by multiple thinnings
increased the establishment, growth and sustenance of red pine as well as other
hardwood species for period as long as a century; consequently, species diversity
increased as succession progressed.

14. The highest ingrowth of red pine density into the sub-canopy in burned plots, as
well as new red pine and hardwood recruits in the understory show increased

sustainability of red pine, along with other species, when thinned.

3.6 Conclusion
Frequent burning in mid-rotation, especially if combined with subsequent
thinning, could be the basis of long-term sustainability, high productivity, and increases

in biodiversity of red pine stands.

157

APPENDICES

Appendix 2.1: Species codes
The following species codes have been used in the tables and text

 

 

Latin names Species codes Common names
Acer rubrum RM Red maple

Acer saccharum SM Sugar maple

A melanchier spps AM Juneberry

F agus grandrfolia AB Beech

Fraxinus americana WA White ash
Ostrya virginiana HH Eastern hophombeam
Pinus resinosa RP Red pine

Pinus strobus WP White pine
Prunus serotina BC Black cherry
Prunus virginiana CC Choke cherry

 

Following species are grouped under the different species codes and each of the
species constitutes less than 1% of total density in my study.
NC (Non commercial trees) and others:

A cer pensylvam'cum ST Stripped Maple

Betula alleghanensis YB Yellow birch

Hamamelis virginiana WH Witch hazel

Malus baccata SC Siberian crab apple

Populus balsamifera BP Balsam poplar

Populus grandidentata BT Bigtooth aspen

Populus tremuloides QA Quaking aspen

Prunus pensylvanicum PR Pin cherry

Salix spps W1 Willow

Tilia americana BW Bass wood

Tsuga canadensis EH Hemlock

Ulmus americana AE American elm

Viburnum acerifolium MV Maple-leaf Viburnum
Oaks: Quercus rubra RO Red oak

Quercus alba WO White oak

Quercus velulina BO Black oak

 

158

0.02:0 0000 303 :5 6:032.“ .4820“. 038.00% E3 05 fl 2 ”00oz

 

 

 

 

 

 

 

 

o m v SN 9 o w 33m Bow
0 o c ”N 3 o o 2:5 2C.
o o 0 SN o N o Esm 0:0
v o 9 S: on S o 8:80
l 3% >—
5m 0 o _ o _ o wo v o o o o _ o 0.59m Sow
NM o o o o o o mm m _ o o o o 0 mFSm 95:.
wN o o o o c o 3 m o o _ o o o Esm 0:0
EN o o o o o o 3.. n v o _ o o o 85:00
ATS Tm: mEv
10.2
m._oN o o _ N o o om me o o o o fl N 05—5 Son
oNNN o o o o o c we MNN _ m o ~ 0 o 2.5m 93.
:3 o o o o N m am :2 o o _ _ o o Esm 0:0
o.moN o _ o o m _ _ ow NM: v w _ m o o 38:00
:2 we
0E=_o>
MAN o o _ o o 0 we MN 0 o o o _ o 2:5 Son
v. _ N o o o o o o no _N N o _ o o o 3:5 03,:
PM: o o o o N o no M: o o N o o o :::m 0:0
No— _ o o o N. _ Na 2 n — m _ o o 68:60
:2 we
«00¢. .0021:
of o o v m. o o 3 w: o o o o _ N 2.5m Son
92 o o o o o c on 02 S E _ _ o o 35m 03,—.
v: o o o o _ N N» M: _ o o E N o 0 E3. 0:0
Nmm m 2 o o NN Q. mm 2 _ Nm 2: w N o o _ob:oU
$2 .85
30:00
:88 we m3 220:0 <3 :30va Em 138.8 mm E0016 SE :38 Co E. _lSoCo 3m 3:258; ,
.86.: E089: 0:080: 0:080: E080: E080: E080: E080:

 

 

.0153 220%: 60:: 0005 b00830? 9: 0=_w> 00:86 :: 0:8 22 .0E:_o> .005 imam . ESQ “NN 56:0 <

 

  

 

 

159

 

 

 

 

 

 

Z 3 3 2 2 IA 2 Ev 2 n: no 3 me o: a: o: o: 2 9o 3 3 9o 3 magmas:
mm 3 o: 0.2 to 3m 3 3m 2 a: to 3 2 9o 3 3 3 no 2. .2 9o 90 o: ”Esme:
Z o.o 3 i: 2 SM 3 :8 no 92 9o 3 No 3 3 3 9o a: 9o 3 3 o: o: 5325
3 3 o: a: 2 0.3 a: as 0.: S: no 3 3 3 9o 2 3 mo 3 3 3 Q” 3 3:80
A a 2 :5

T _- m

22
380°b®_§$$o__8$¢:oooo:NoocoEsmso:
o.RN_N3:$8E§~_R¢o~oo _ _ o_:_oc “Esme:
$200___m£$omomm~9o:_m____:_ooE325
380°zomzmomfieommm_va_Nmo~v .228
pins

25.75

:8 o 3 b E a 3 S 3 mm ow v 3 o 3 o 0.0 _ S o 3 o 3 mleamso:
2N _ S 2 3 R 3 E 2 2 mm o S o 3 _ No _ mo _ 2 o 3 E593
E: o 90 : 3 mm 2 a E a ”a w I N no _ we _ no N 2 o 3 E525
2: o 0.0 3 Z ”N E a 3 2 S m 2. _ S m 0.0 N :0 o f m 3 3:80
ATENEV

:0:<_am:m

EB £8 2 .x. 2% g. a: .x. S S x S a .x. :8 $28 x R m g. N 2 .x. t 2 $: 2 25:58;

 

 

6.3V 280:0 E020: 96:88:80 05 :5 E0 5 momma—U 535:5
$3.20 0N_m 55:35 ~:0:0uc_u 5:5. 80: ESE0>O .3 3:2 a. 0:|:::_o> .a0:< Emwm ”ON 0220 <

 

 

 

 

160

Appendix 3.1: Average relative understory (0-11 cms dcl) tree density“ under 100
years of no-management and thinning simulations" in mature red pine stands.

 

Small understory (0-6 cm)

trees ha'1

 

Year

Large understory (6-11 cm)

 

Burn tr Burn treatments

C 18 2B 4B

 

 

Beginning
2003
%

 

9545 18752 19756 16725
95 99 99 99

 

 

No-management
2014
%
2044
%
2074
%
2104
°/o

 

 

8007 15719 15755 13735
91 97 98 97
477 3000 1949 2941
31 67 63 78
72 58 48 20

1 1 6 7 3
31 9 10 7

7 2 2 2

 

 

Thinnings
2014
%
2044
°/o
2074
%
21 O4
%

7824 15719 15455 13086
91 97 98 97
1527 2970 2897 3052
65 73 79 85
297 573 960 1591
36 48 64 78
66 56 450 633
14 10 43 56

 

trees ha'1

C 1 B 28 4B
324 46 37 0
3 0 0 0
41 1 424 147 242
5 3 1 2
300 412 243 1 18
20 9 8 3
54 75 43 51
8 8 6 9
14 9 20 0
3 1 4 0
402 424 145 239
5 3 1 2
256 290 145 57
1 1 7 4 2
53 51 57 129
6 4 4 6
40 45 159 78
8 8 15 7

 

Note: Some of the 'zeros' in % column of BA have values below 0.5% and have been

rounded to zero.

Burn Treatments columns are: C, Control; 18, One Burn; 28, Two Burns; 48,

Four Burns

% (in rows) is the percent of total woody understory tree density.

dcl=diameter size class.

*Stand density does not include removals in thinnings.

"Simulation started on year-2004.
The first, second and third thinnings occurred in 2004, 2034, and 2064.

161

 

Appendix 3.2: Average and relative sub—canopy (>11-32 cm dcl), mid-canopy (32 & 42

cm dcl) and dominant overstory (>42 cm dcl) tree density“ under 100 years of
simulations" in mature red pine stands.

 

 

 

 

 

 

 

Sub-canopy Mid-canopy Dominant overstory
trees ha'1 trees ha'1 trees ha'1
Burn treatments

Year C 18 2B 48 C 18 28 48 C 18 28 48
Beginning

2003 1 03 26 1 0 6 33 36 50 58 63 53 82 90

% 1 0 0 0 O O O O 1 0 0 1
No-management

2014 287 37 30 10 16 16 24 17 82 76 111 132

% 3 0 O O 0 O O 0 1 0 1 1

2044 624 980 790 564 28 1 7 8 90 82 114 133

°/o 41 22 25 15 2 0 O 0 6 2 4 4

2074 414 653 526 372 37 42 20 23 98 71 95 114

"/0 61 73 72 64 6 5 3 4 14 8 13 20

2104 277 451 378 245 62 67 48104 99 75 87 96

% 57 74 7O 54 13 11 9 23 20 12 16 21
Thinnings

2014 276 37 26 9 15 16 21 14 79 76 103 116

% 3 0 0 0 0 0 0 0 1 0 1 1

2044 483 770 569 410 17 0 4 4 56 54 66 77

% 21 19 15 11 1 0 O 0 2 1 2 2

2074 359 497 398 222 59 34 28 46 52 37 49 62

°/o 44 42 27 11 7 3 2 2 6 3 3 3

2104 161 252 240 223 123 174 127 99 94 57 59 96

% 33 43 23 20 25 30 12 9 19 10 6 8

 

Note: Some of the 'zeros' in % column have values below 0.5% and have been rounded

up to zero.
Burn Treatments columns are: C, Unbumed Control; 1B, One Burn; 28, Two Burns;
48, Four Burns.

°/o (in rows) is the percent of total tree density.
dcl=diameter size class.

*Stand density does not include removals in thinnings.
"Simulation started on year—2004.

The first, second and third thinnings occurred in 2004, 2034. and 2064.

162

 

.38 65 .6ch .38 s 60:88 55:55 2:: Ba 6:83 .5: 2:

.666N -500» :0 605550 650.58%: .mw:_::_£ :_ 200.50: 06205 50: 0006 A5306 6:89. 53:06 6505 6 5:00:00 05 0:0 3:560 :_ o\o
.000N 05 60656.. :006 020: 6:: $6.6 30.06 00:65 08: 5:060 o\.. E .0050N. 05 6 060m “0502

 

 

 

 

NN mm6 N6 6 ON N n _ m _ 6 m an m. 66 6. M6 3 mm N_ 66 o_ 66 o o N h 0:55 0:0,.—
6N 666 mm N I _ w o N N 2 m E 6 .m 6N 62 0m 06— 6 mm 0 66 6 0N N 2 3.5m 03...
0N N6 66 _ 6 o _ o _ N 6_ 6 6N n .6 M: o: 6N :3 _N 62 2 66 _ a N a :55 0:0
mN 36 _m N I _ h m 6N _ 6 m 6N w :m ON 3 NN 66— 2 66 6 6m m 6. n .m 6o_N 65:00
ON 006 6 :6 6 mm N N_ _ 6 _ w o o 6 NN 2 ow S 3 6N an. w :6 o E m 0N 055m :00...
6N 3:. w 06 N M: N Z _ 6 o N _ m N N_ 6 0N 6N NM: 6m 66N o 06 6 m6 N 6 0:55 03:.
5m 00» 6 G _ 2 _ w o N o o _ o. 6 Nm 6 m6 6_ 6N_ 0m 66m 6_ oN_ w m: 6 mm :55 0:0
MN 66 w E N 2 N N_ o m m 2 N 6. m MN 6 N6 _N Nnm _N 66: Z 2. w 66 _ N: 63N 65:00
0N M6: _ 6N _ m6 _ _6 o 2 o m o w o N o _ N mm 6 S: a 0Nm m w: w: _60N 0:55 5.0“—
6N no; _ _6 _ om _ «N o 2 o 6 o o o N. o m _ 66 N m: NN 066 w M6N M6 062 0:55 03,—.
mm 666 _ MN 0 a: o NN o S o m o _ o o o o. N 66 m 6N_ 2 6R 0 N6 :6 606m :55 0:0
N_ 26— N wN N 6N _ 2 _ 2 o N o m N mN N 6N m E. w. NNN 6N Nom 6N 66m 3. 2.6 660N 65:00
6N 626— o o o m o 6N 0 6m 6 6m 0 o. o N o o o m o o o 6 N N6N no 6min— mfism 5.0”.
0N 0606— o N o o. o 6N o 66 6 Nm 6 ON 6 6 o o o 6 o N o _N _ :6. 60 62.6— 855 03...
0N m>N6_ o _ o N. o NN o 6N o _N o N_ o 6 o 6. o o o m o 6N m 6N6 3 0:2 :55 0:0
6_ now: 6 o o 6 o wN o mN o 0N o m_ o m o 0 o I _ mm N 6M: 6 :6 a Son 6SN 65:00
5:0E0w::0:_.02

6N 032 o o o o o m o wN o 06 6 66 o 6_ o o o o o 6 o o o o 00 696— 0:55 50:
on 693— o o o _ o w 6 MN 0 66 c on 0 ON 6 o o 6 o 6 o o o um 00 663— 855 03,—.
0N Saw— 6 0 o o o h o mN 0 MN o N o o_ o m o 6 o 6 o 2 o 66 00 N32 :55 0:0
6. >606. o o o o o n o wN 0 MN 6 mN o o. o m o N_ o 6 _ mm m 6Nm no 6660 moON 65:00
w:_::_w0m

..\o 650:. N6A o\o N6-Nm .x. 59% ..\° NW: o\o 56-920 N6-R o\o 5N3; NMNN o\o NN-N~o\o NN-:o\° 2-: ..\o :6 o\.. 6-0 60> 55:585.:

 

 

.500» 66— :0 0:052:85 5:0E0w0::E 5:00 _6 06:: 00000 0 050:5: > 000 0: :> 0250 0: :05 6:: A 0: 0.0056 0:07:06 005 0w80>< ”00m 56:00.3.
6 . t _ .6 6 _ . _ . _- .

163

 

.608 0:0 .680 .686 5 00:88 “05:55 2:: 050 0:88 .E: 2:
.600N L006 :0 60:80 5.06.56: 6655:: 5. m.0>0E0: 0630:. .0: 0006 65.206 6:056... 65.206 .0596 5:00:00 0.6 00 255.00 :. o\..

.800 0. 606::0: :006 0>0.. 6:0 $6.6 30.06 60:.0> 0>0.. :E:60 o\o :. 60.0.0. 0... 6 0:66 ”0.02

 

 

 

66 6N: 6 66 o 6 o N . 6 6 66 6 66 6 66 6 N6 6 66 6 66 6N No b 6b 66 666 0:56. Son.
N6 N6... 6 66 o N o . . 6 . N. 6 .6 6 66 o. No. 6 on 6 N6 6 6. 6. 66. 66 066 0:55 03.6
6. 666 6 66 o N o N . N 6 6. 6 66 .N 6N. NN 6N. N. .N. 6 6N 6. 6N 6 66 o. 66 :55 0:0
6. 666 6 «N . N. N o. N o. 6 06 o .6 0. N6 o. 06 N. 66 6. .6 o. 6. 6 66 6. 66 6o_N 65:00
06 .6ON N 66 o 6 o 6 o 6 o 6 o o N N6 6 N6 6 NN. 6 .6 N N. 6 oN. 6N. .66. 0:56. 50“.
N.N No6. N 66 o 6 o 6 o o o 6 o 6 N 6N 6 on o 66. 6 6N. 6. 66 6 6 66 06a mE=m 03.6
.N No: N O6 6 6 o N o . o N . .. N NN 6 .N. o. 6NN N. 66. 6. N6 6 .6 66 666 Eam 0:0
6. 6.6 6 oN o 6 o 6 . 6 . 6 6 6N 6 66 .. 6o 6. S. o. .6 6N E. 6 66 66 SN 6N.0N 65:00
6N 866 o N.. . 6N . NN o 6 o N o 6 o . o N N .6 N co .N 66N N 6 66 N666 0:56. 50“.
6N .666 . 6N o N.. o 6. o a o N o o o 6 o 6 . N6 N NN. 66 .66 6 66. on N06N 6:56. 03..
06 6606 o o. o 6. o 6. o 6 o N o o o o o 6 . N6 6 6.. 66 666 N. RN 6» onoN :55 0:0
N.. 666N . 0N . 6. o N. o o o N o 6 . 6. N 66 N 66 0 66. 6N ONN .. 66N 66 6N6. 660N 65:00
6N 6666. o o o 6 o .N o 06 6 N6 6 o o 6 o o o 6 o o o 6 N 06N no 6606. 0:56. 50..
oN 6666. o N o o. o 6N o 66 o oN o n. o 6 o o o 6 o N o o. . 66. 60 6666. 0:56. 03...
o6 6N.N6. o . o n o NN o 6N o _N o N. o 6 o o. o o o 6 o 6N 6 6N6 no «0.66. :56. 0:0
6. 66.66 c o o 6 o N.N o NN c 6N o N. o 6 o 6 o o. . .6 6 2.. 6 N06 .0 6N6N. 6.0N 65:00

06:65.:
.x. .80.. .x. N6A o\° N6-N.6 ..\o N.6-N6 ..\o N6-N.6 o\o N.6-N6.x. N6.N.m.x. n6-Nmo\o N6-N.N.x. NN-N~°\° NN-:..\o 6.-.. o\.. ...6 ..\o 6.0 00> 85:502..

 

 

 

.6000» oo. :06

0:0.50.:E.m.:0E0w0:0E 5:0:0 _ :0 :5 60060005050. z ..\o 60: 0> 03.0.0: 5056:0A 0 6005 35:0 005 06000><u 0::5:00 06.6 x. :0&.<
06.66 . .66 . . 7.. 6. 6 6. .6

164

.666N 6:0 .666N .666N :. 60:50:00 06:656. 6:3: 6:0 6:000m .52... 06...
.666N L006 :0 63:05: 5:06:66: 06:6:6: :. 0.0680: 0620:. 50: 0006 00:0 .0006 6:86.. 00:0 .0006 .6 5:00:06 06: 0:0 2:560 :. o6
.0:0~ 0. 6065.0: :006 0>06 6:0 066.6 30.06 m0:.0> 0>06 :::=60 .x. :. .00:0N. 06: 60 0:56 ”0:07.

 

 

 

 

6N 6.66 66 6.6N .. 6.6 6 6.. . 6.6 . 6.6 6 6.6 N. 6.6 6 6.6 6 N6 6 6.. . 6.6 6 6.6 6 6.6 0:56. 58:.
6N 6.66 66 ..NN6 ..6 6 6.. . 6.6 6 6.. 6 N.N 6 6.N 6. 6.6 6. 6.6 N ... N 6.6 6 ..6 6 6.6 0:56. 03...
6N 6.66 .6 N.NN N N.. . 6.6 . 6.6 6 N.N 6 6.6 6 ..6 6. 6.6 6. 6.6 6 6.6 N N.. 6 ..6 6 6.6 :::m 0:0
6N 6.N6 66 N.N 6 ..6 6 6.. 6 6.6 N 6.6 6 ..6 6 6.6 6. 6.6 6. 6.6 6 6.. . 6.6 6 ..6 6 6.6 66.N 65:00
6N 6.66 66 6.6N 6. N6. 6 6.6 N 6.. . 6.6 6 6.6 6 ..N N. 6.6 6 6.6 6 6.6 N 6.. . 6.6 6 6.6 0:56. :00...
6N 6.66 N6 6.NN 6. 6.6 6 6.N N 6.6 . 6.6 N 6.. N N.. 6 ..N 6. 6.6 6. ..6 N N.. . 6.6 6 ..6 0:56. 03...
6N 6.66 66 6.6.6 6.6 N N.. . 6.6 6 6.6 N N.. 6 ..6 6 ..6 .. 6.6 NN 6... 6 6.N . 6.6 6 ..6 :56. 0:0
6N 6.6 66 6.6. 6 6.6 6 6.6 . 6.6 6 6.6 6 6.. 6 6.N 6 6.6 6. 6.6 6 6.6 6 6.. . 6.6 6 6.6 666N 65:00
6N N66 6. 6.6 6N 6.N. 6. 6.6. 6 6.6 N 6.6 . 6.6 6 N6 6 ..6 6 ..6 6 6.6 .. 6.6 N N.. N 6.. mEsm :00“.
6N 6.66 6N 6.6.6. 6.6 .. 6.6 6 ..6 N 6.6 6 6.6 . 6.6 . 6.6 6 6.. 6 6.N .N 6... 6 6.N 6 6.. 0:56. 03...
6N ..66 .N 6.6. .. 6.6 .. 6.6 6 6.N . 6.6 6 ..6 6 6.6 . 6.6 6 6.6 6 6.6 6N 6.N. 6 6.6 6 6.. :56. 0:0
6N 6.66 6N 6.6 6. 6.6 6 6.6 6 6.6 . 6.6 . 6.6 6 6.N 6 6.. 6 6.6 6. 6.6 N. 6.6 6 ..N . 6.6 666N 65:00
6N 6.N6 6 6.6 6 6.6 6. 6.6 66 6.6. 6N 6.6 6 6.. N 6.6 6 6.6 . 6.6 6 6.6 6 ..6 6 6.. 6 6.N mEzm :00“.
6N 6.N6 N 6.6 6 6.N 6N 6.6 6N N6 6. 6.6 6 6.N . 6.6 6 6.6 . 6.6 6 ..6 . 6.6 6 6.. 6. 6.6 mEzm 03.6
NN 6.6N N 6.6 6 6.. 6. 6.6 6. 6.6 6. 6.6 6 6.. N 6.6 6 6.6 6 6.6 6 ..6 N 6.6 .. ..6 6N 6.6 :55 0:0
6N N..6 6 6.6 6 6.. NN 6.6 6. 6.6 6. 6.6 6 6.. . 6.6 N 6.6 N 6.6 6 6.N 6. 6.6 6. ..6 6 6.N 6.6N 65:00
::0::0m0:0::.02

6N ..6N 6 6.6 6 6.6 6 6.6 6N 6.6 66 6.6 6N 6.6 6 6.. 6 6.6 6 6.6 . N6 6 6.6 6 6.6 6 6.6 mEsm :30“.
6N 6.NN 6 6.6 . N6 6 6.. 6N 6.6 N6 6.6 6. 6.6 6 6.. 6 6.6 . N.6 . N6 6 6.6 . 6.6 6 6.6 0:56. 03...
6N ..6. 6 6.6 6 6.6 6 6.. 6N 6.6 NN 6.6 6N 6.6 6 6.. . N6 . N6 . N.6 . N.6 . N6 6. 6.. :56. 0:0
6N N.NN 6 6.6 6 6.6 6 6.. 6N 6.6 .N 6.6 6. ..6 6 6.. . N6 6 6.6 . N6 6 6.. .. 6.N 6 6.. 666N 65:00
655.606.

06 .000... .6 N6A o\o N666 .x. 66-N6 .x. N6-66 o\o 66.N6o\o N6-6mc\e 66-Nmo\° N6-6No\o 6N-N~c\o NN.6...\° 6.-.. .x. ..-6 o\° 6-6 :00> 2:05.086

 

 

 

 

0:006 66. :06 :m:0.50.=::.m 60806050.: 5:20.66 06:: 60000.0 00:06 66 6x; m0:.0> 03.0.0: :66: 6:0 $-06 NEV .00:0 .0006 060:0>< ”66.6 x.6:006<

165

.666N 6:0 .666N .666N :. 60::0000 06:65.6 6:6. 6:0 6:0000 00:6 06...
.666N L006 :0 605050 5:06:66: 06:65.6 :. 0.0>0::0: 0606:. :0: 0006 00:0 .0006 6:056... 00:0 .0006 60 5:00:06 06: 0:0 0:50.00 :. 06

0:00 0: 606:00: :006 0>06 6:0 066.6 30.06 000.0> 0>06 5:060 o6 :. .800. 065 6 0806 ”0:02

 

 

 

 

6N ..N6 N6 6..N N 6.. . 6.6 N N.. .. 6.6 6. 6.6 6 N6 6 ..6 6 6.N N N.. 6 6.. . 6.6 N 6.. 0506. :00...
6N N66 66 6.6. . 6.6 6 N6 6 6.. 6 6.. .. 6.6 6. 6.6 6. 6.6 6 6.6 6 6.. . 6.6 N 6.. N 6.6 050m 03...
6N ..N6 6N 6.6. . 6.6 . 6.6 6 6.. 6 6.N N. 6.6 6N ..N. 6. 6.6 6 6.6 . 6.6 . 6.6 . 6.6 6 ..6 E06. 0:0
6N N.66 6N ..6. 6 6.. 6 6.N 6 6.. 6. 6.6 .. N6 6. 6.6 6 6.6 6 6.. 6 ..N . 6.6 . 6.6 6 ..6 66.N 65:00
6N 6.66 66 6.6. 6 6.. N 6.6 N 6.6 . 6.6 6 ... 6 6.6 N. 6.6 6 6.6 6 6.N . N6 6 6.. 6 6.. 0::06. :00...
6N N66 66 6.6. 6 6.. N 6.6 6 6.6 N 6.6 N 6.6 6 N.N N. 6.6 6. 6.6 6. ..6 6 6.. . 6.6 6 6.. 0::06. 03...
6N 6.66 6N 6... N 6.6 . 6.6 6 N6 . 6.6 6 6.. 6 ..N N. 6.6 6N 6... N. 6.6 6 6.. . 6.6 N 6.6 :05 0:0
6N 6.66 6N 6... 6 6.. N 6.. 6 6.. 6 6.. 6 6.N 6 6.6 6. 6.6 6. 6.6 6 6.N 6 6.. . 6.6 . 6.6 666N 65:00
6N 6.66 6. 6.6 6N ..6 6. 6.6 6 6.. . 6.6 . 6.6 6 ..6 6 N6 6 6.6 6 6.N 6. 6.6 N 6.6 6 N.. 3.5.. :00".
6N 6.66 6N 6.6 6. 6.6 6 ..6 6 6.. . 6.6 6 6.6 . 6.6 . N6 6 6.. 6 ..N 6N N6 6 6.. 6 6.. 0::06. 03...
6N ..66 6. N6 6. 6.6 6 6.6 6 6.. . 6.6 6 ..6 6 6.6 N 6.6 6 6.N 6 6.6 6N 6.6. 6 6.N 6 6.. 50m 0:0
6N 6.66 6N 6.6 .. 6.6 6 6.N 6 6.. . 6.6 . 6.6 6 6.. 6 6.6 6 6.N 6. 6.6 N. 6.6 6 6.N N 6.6 666N 65:00
6N 6.6N 6 6.6 6 6.6 6. 6.6 66 6.6. 6N 6.6 6 N.. . 6.6 6 6.6 . N6 6 6.6 6 ..6 6 6.. 6. 6.N 0::06. 500“.
6N 6.66 N 6.6 6 6.N 6N ..6 6N 6.6 6. 6.6 6 ..N . 6.6 6 6.6 . N6 6 ..6 . 6.6 6 6.. 6. 6.6 0::0m 03...
6N 6.6N N 6.6 6 6.. 6. 6.6 6. 6.6 6. 6.6 6 6.. N 6.6 6 6.6 6 6.6 6 ..6 N 6.6 .. ..6 6N 6.6 0:06. 0:0
6N ..66 6 6.6 6 N.. NN 6.6 6. 6.6 6. 6.6 6 6.. . 6.6 N 6.6 N 6.6 6 6.N 6. 6.N 6. ..6 6 N.N 6.6N 65:00

06:65:...
06 .050... 96 N6A o6 N6-66 e6 66-N6 o6 N6-66 o6 66-N6o\o N6-6mo\° 66-Nmo\o N6-6No\o 6N-Nmo\o NN-6.o\o 6.-.. 06 ..-6 .x. 6-6 :00> 05:08:08..-

 

 

.0:§ 66. :06

10:0.50.0E.0 .:0::060:0E 55:06.6 :06:0 00000.0 :0.0::0.6 66 A060 000.0> 0350.0: :66: 6:0 $-06 NEV 000:0 .0006 060:0>< ”605:5:000 66.6 66:00.66~

166

.666N 6:0 .666N .666N :. 60::0000 06:.::.6: 6:6... 6:0 6:800 60:66 06...
666N :006 :0 68:80 8.0.08.6: 0655.6: :. 0.0>0::0: 0630:. :0: 0006 60.300 6:006... 080.? 60.300 .0:0:..0 200:2. 06: 0:0 0:::0.00 :. ..\o

0:00 0: 606:00: :006 0>06 6:0 0666 30.06 000.0> 0>06 :E0.00 c\o :. .00:0~. 06:..0 0:56 “0:02

 

 

 

 

 

6N 666 .6 66N 6. N6 6 6. N 6 N 6 6 6N 6 .6 6 6. . 6 6 6 6 6 6 6 6 6 0::06. :00...
6N 6.6 66 N6N 6 66 6 .N . 6 6 6. 6 6. 6 6. 6 6N 6 .. 6 6 6 6 6 6 6 6 0506.03...
NN 666 66 66N 6 6. . 6 . N 6 6. 6 6N 6 6. 6 6N . 6 6 6 6 6 6 6 6 6 E06. 0:0
6N 6N6 .6 66N 6 66 6 6. 6. .6 . 6 6 6. 6 6. 6 .N N 6 6 6 6 6 6 6 6 6 66.N .0:::00
6N 666 66 66N 6N 6N. 6 66 N N. N 6 6 6 N .. 6 6N N 6 6 6 6 6 6 6 6 6 0::06. :00...
6N 6.6 66 66N 6. 66 6 N6 N 6. . N N 6 N 6 N 6 6 .. 6 6 6 6 6 6 6 6 0E0m036
NN 666 66 66N 6. 66 6 6. . 6 6 6 N 6 6 6. 6 N. N 6 6 6 6 6 6 6 6 6 E06. 0:0
6N 666 66 66N N. 66 6 66 N 6 6 6N N 6 6 N. 6 6. N 6 6 6 6 6 6 6 6 6 666N .0:::00
66 666 6N 66. 66 66. 6N 6.. 6. 66 N 6 N 6 6 N 6 6 N 6 6 6 6 6 6 6 6 6 053630...
6N .66 66 66. 6N 66. 6. 66 6 66 N 6. 6 6 . 6 6 N . 6 6 6 6 6 6 6 6 6 0506.036
6N 66N 66 6N. 6N 66 .N N6 6 NN N 6 6 . 6 6 . 6 N 6 6 6 6 6 6 6 6 6 50m 0:0
6N 6N6 66 6.. 6N 66 6. 66 6. 66 N 6 . 6 6 6. N 6 . N 6 6 6 6 6 6 6 6 666N .00:00
66 66N 6 6 6 6 .N 66 66 6.. 66 66 6 N. N 6 6 6 6 6 6 6 6 6 6 6 6 6 0506.50,...
6N .6N 6 6 .. 66 6N 66 .6 66 6N 66 6 6N . 6 6 6 6 . 6 6 6 6 6 6 6 6 00:06. 03...
6N 66. N 6 6. 6N 66 66 6N .6 6. 66 6 6. N 6 6 6 6 6 6 6 6 6 6 6 6 6 506.25
.N .6N 6 6 6 6. 66 .6 6N 66 6N 66 6 6. . N . N . N 6 6 6 6 6 6 6 6 6.6N 30:00

::0E060:0E.02
.6 6NN 6 6 6 6 6 6 6N 66 .6 66 6N 66 6 .. 6 6 6 6 6 6 6 6 6 6 6 6 0806.50“.
6N N6. 6 6 . N 6. 6. 6N 66 66 66 6. 66 6 6. 6 6 6 . 6 6 6 6 6 6 6 6 0506. 03...
6. 66. 6 6 6 6 .. 6. 66 66 6N 66 NN 66 6 6 . . . . 6 6 6 6 6 6 6 6 ::0m0:O
NN N6. 6 6 6 6 6. 6. 66 66 66 66 6. 6N 6 6 . . 6 6 6 6 6 6 6 6 6 6 666N .0:::00
..\o .806 o\.. N6A o\° N6-66 o\o 66-N6 o\° N6-66 c\o 66-N6o\o N6-6mo\o 6?ng N6-6N o\o 6N-N~°\° NN-6.°\° 6.-..0\o ..-6 .x. 6-6 :00> 8:08:00;

 

 

0:006

66. :06 :0:0_:0_0::.0 ::0::060:0:: 0:0:0 _ :0 :0 00000 0 0060. 6 ..\o 000 0> 02:0 0: :0 : :0 0 E 080 0> 60 300 060:0>< ”06.6 x: :06“;
. 6.66 . .66 . .. .66 .-66 _.. . . .6

167

.666N 6:0 .666N .666N 0. 60::0000 06:65.6: 6.66: 6:0 6:800 60:: 06...
666N :006 :0 60:00 60.20.0666: 066.656: :. 0.0>0E0: 060.06. :0: 0006 60.300 6:0:6... .0E0.0> 60.300 .80: .0 E0806 06: 0:0 0:80.00 6. :6
0:00 0: 606:00: :006 0>06 6:0 5666 30.06 000_0> 0>06 6.60.00 o\o :. .0800. 06:60 06:06 ”0:02

 

 

 

 

6N 6.6N6 66 66N 6 .. . 6 6 .. 6. .6 N. N6 6 66N N 6. . 6 6 6 6 6 6 6 6 6 06:06. :00...
6N 666 66 6NN N 6 . N 6 .. 6 6. 6 66 .. 66 6 .N . 6 6 6 6 6 6 6 6 6 06:06. 036
NN 6N6 66 66. N 6 N 6 6 N. 6 6. .. 66 6. 66 6 6N . 6 6 6 6 6 6 6 6 6 6:06. 0:0
6N 6.666 66 66. 6 6. 6 6N. 6 6. N. .6 6 6N .. 66 6 .. 6 . 6 6 o 6 6 6 6 6 66.N .0:::00
66 6.6.6 66 .NN 6 NN 6 6. N 6 N 6 N 6 6 ON 6 ON N 6 6 6 6 6 6 6 6 6 0E06. :00“.
6N ..66N .6 66. 6 6. 6 6 o 6 N 6 N 6 6 6. 6 6. 6 .. 6 6 6 6 6 6 6 6 0:56.03...
.N N6NN 66 66. 6 .. 6 6 . N . N 6 6 6 N. 6 6. 6 6. 6 6 6 6 6 6 6 6 E06. 0:0
6N N.N6N 66 66. 6 N. 6 N. 6 6. 6 6 6 6. 6 6. 6. 6N N 6 6 6 6 6 6 6 6 6 666N .06600
6N 6.N6N 6N 66 66 66 6N 66 6 6. . 6 . 6 6 . 6 . 6 6 6 6 6 6 6 6 6 6 06:06. :00...
6N 6.66N 66 6.. 6N 66 6. 66 6 .N . N 6 6 . 6 6 . . N 6 6 6 6 6 6 6 6 06:06. 03...
NN 66. 66 66 NN 66 6N 66 6 6. N 6 6 . o 6 . 6 6 6 6 6 6 6 6 6 6 6 E06. 0:0
6N 66.N 66 66 NN 66 6. 66 6 ON N 6 N 6 6 6. 6 6 . 6 6 6 6 6 6 6 6 6 666N .0:::00
6N 6.N6N 6 6 6 6 .N 66 .6 66. 6N .6 6 .. N 6 6 6 6 6 6 6 6 6 6 6 6 6 0506. :00...
6N ..N6N 6 6 .. 6N 6N 66 .6 66 6. 66 6 6N . 6 6 6 6 . 6 6 6 6 6 6 6 o 06:06. 03...
NN 6.N6. N 6 6. ON 66 66 6N .6 6. 66 6 6. N 6 6 6 6 6 6 6 6 6 o 6 6 6 6:06.0:O
NN N66. 6 6 6 6. 66 66 6N 66 6N 66 6 6. . . . N . . 6 6 6 6 6 6 6 6 6.6N .0:::00

0656566
06 .08... ..\o N6A o\.. N6-66 o\° 66-N6 .x. N6-66 o\o 66.N6o\e N6-6mo\o 6.6-ng N6-6No\e 6N-N~o\o NN-6.o\.. 6......\.. ..-6 o\o 66 :00> 3:05:00;

 

 

0:00.. 66. :06

308603660 ::0::060:0E E80666 :06:0 00000.0 0:0—60.6 66 6.6. 000.0> 0>.:0.0: 606: 6:0 ...06 6:6 ..0E0.0> 60.300 060:0>< A600:.::00. 06.6 56:00.36

168

 

Appendix 3.4: Average and relative standing* sawlog volume of major species under 100 years
of no-management and thinning simulations“ in differently burned and unburned mature red

 

 

 

 

 

pinestands.
Red maple Red pine Sugar maple Others Total
m3 ha'1 rn3 ha'1 rn3 ha.1 m3 ha'1 m3 ha'1
Burn treatments
Year C18284B C 18 28 48 C182848 0182848 C 18 28 48
Beginning
2003 0010150134190219 2100 0001 152136192220
% 0010 9899991002100 000
No-management
2014 0 0 2 0 197191 269 285 4 2 0 0 0 0 0 1 201193271287”
% 0010 9899991002100 0000
2044 2020320290398429 8200 0007329293401435
°/o 1010 979999982100 0002
2074 9 0 1 0 378352 418474 8 2 0 O 0 0 0 2 395354419475
% 2000 96991001002100 0000
2104 9 0 2 0 407 376 414474 8 1 0 0 0 0 0 0 424377415474
% 20009610010010020000000
Thinnings
2014 0 0 2 0 190191250251 4 2 0 0 O O 0 1 194193252252
% 0010 9899991002100 0000
2044 6 0 2 0 202194 238 258 6 2 0 0 0 O 0 4 213196239263
% 3010 959999983100 0002
2074 20 1 6 0 222214 257311 9 1 0 0 1 3 1 8 252220264318
% 8120 889797984100 0212
2104 34 6 9 0 2913153414151910 0 14 513 344326355429
% 10230 849796975000 0113

 

* consists of American beech, black cherry, choke cherry, non-commercial species, others

and white ash.

Burn Treatments columns are: C, Unbumed Control; 18, One Burn; 28, Two Burns; 48, Four

Burns

% (in rows) is the percent of total species sawlog volume.
*Stand volume does not include removals in thinning.

“Simulation started on year-2004.
The first, second and third thinnings occurred in 2004, 2034, and 2064.

169

 

Appendix 3.5: Comparison of Forest Vegetation Simulator (FVS) projection with that given in

Benzie (1977) for *Standing merchantable cubic volume (m3 ha'l) and the predicted values
in differently burned mature red pine stands at site index l9.8m in the beginning and at the
end ofno-mangement and thinning.

A. Stand age 72:years in the beginning ofthe simulation (year 2003*")

 

Standing volume

 

FVS predicted volume Benzie predicted volume
C 1B 2B 48 C 1B 2B 4B
Basal Area (m3 ha'l) 22.2 17.1 22.5 24.] 22.2 17.1 22.5 24.]
Volume (m3 ha'l) 180 155 218 248 227 172.3 227 245
% change over -21 -10 -4 1.2

Benzie

B. Stand age 143’” and BA set to 34.4 sq m/ha in year 2074

 

 

FVS predicted F VS predicted Benzie

(No-management) (Thinnings) Predicted

SE-nding volume Cumulative volume Volume
Volume (m3 ha'l) 470 420 480 536 586 504 586 641 480
% change over -2 -12 0 12 22 5 22 34

 

Benzie

*Standing volume does not include removals in thinning.

"Predicted volumes were chosen for the closest stand ages 70 and 140 years in Benzie's
Table.

The average values in Benzie (1977) for corresponding age, site index and basal area (BA)
were converted to metric.

"“” The average standing volume from the beginning (year 2003) was used instead of the
year 2004 when the first cutting was done as they were in the same age range in Benzie's
table (1977).

Treatments in the columns were C, Unbumed Control, 18, One Burn; ZB, Two Burns and
4B, Four Burns.

170

 

REFERENCES

Abella, S. R., J. F. Jaeger, et al. (2004). "Fifteen Years of Plant Community Dynamics
During a Northwest Ohio Oak Savanna Restoration." The Michigan Botanist 43.

Aber, J. D. and J. M. Melillo (1991). Terrestrial Ecosystems. Orlando, Saunders College
Publishing, 429 p.

 

Abrahamson, W. G. (1984). "Species Responses to Fire on the Florida Lake Wales
Ridge." American Journal of Botany 71(1): 35—43.

 

Abrams, M. D. (1982). Early plant succession on clearcut and burned jack pine sites in
northern lower Michigan. Department of Forestry. East Lansing, Michigan State
University. Ph.D.: 191 p.

Abrams, M. D. (1992). "Fire and the Development of Oak Forests." BioScience 42(5):
346-353.

Abrams, M. D. and M. L. Scott (1989). "Disturbance-Mediated Accelerated Succession
in Two Michigan Forest Types." Forest Science 35(1): 42-49.

 

Abrams, M. D., D. G. Sprugel, et al. (1985). "Multiple successional pathways on recently
disturbed jack pine sites in Michigan." Forest Ecology and Management 10: 31-
48.

Abugov, R. (1982). "Species Diversity and Phasing of Disturbance." Ecology 63(2): 289-
293.

Agee, J. K. (1993). Fire ecology of Pacific Northwest forests. Washington, DC, Island
Press.

Agee, J. K. (1998). Fire and pine ecosystems. Ecology and Biogeography of Pinus. D. M.
Richardson. New York, NY, USA, University Press. Cambridge: 193-218 p.

Ahlgren, C. E. (1960). "Some Effects of Fire on Reproduction and Growth of Vegetation
in Northeastern Minnesota." Ecology 41(3): 431-445.

Ahlgren, C. E. (1976). "Regeneration of Red Pine and White Pine following wildfire and
logging in northeastern Minnesota." Journal of Forestry. 74: 135-140.

Ahlgren, C. E. (1979). "Emergent seedlings on soil from burned and unburned red pine
forest." Minnesota For. Res. Notes 273.

 

Ahlgren, C. E. and I. F. Ahlgren (1981). "Some effects of different forest litters on seed
germination and growth." Canadian Journal of Forest Research.

171

Ahlgren, I. and C. Ahlgren (1960). "Ecological Effects of F ires." The Botanical Review
26(4): 483-533.

Alban, D. H. (1977). "Influence on soil properties of prescribed burning under mature red
pine." North Central Forest Experiment Station, USDA Forest Service Research

Paper (NC-139.).

Allen, C. D., M. Savage, et al. (2002). "Ecological Restoration of Southwestern

Ponderosa Pine Ecosystems: A Broad Perspective." Ecological Applications
12(5): 1418-1433.

Althen, F. W. v. and W. M. Stiell (1990). "A red pine case history: development of the
Rockland plantation from 1914 to 1986." Forestry Chronicle 66(6): 606-610.

 

Anderson, R. C. and L. E. Brown (1986). "Stability and instability in plant communities
following fire." American Journal of Botany 73(3): 364-368.

Arno, S. F. (1980). "Forest Fire History in the Northern Rockies." Journal of Forestry
78(8): 460-465.

Arno, S. F. (1993). "History of Fire Occurrence in Western North America." Renewable
Resources Journal 11(1).

Arno, S. F. and C. E. Fiedler (2005). Mimicking nature's fire: restoring fire-prone forests
in the West. Washington DC, Island Press, 242p.

Arthur, M. A., R. D. Paratley, et a1. (1998). "Single and Repeated Fires Affect Survival
and Regeneration of Woody and Herbaceous Species in an Oak-Pine Forest."
Journal of the Torrey Botanical Sociefl 125(3): 225-236.

Assman, E. (1970). The principles of forest yield study; studies in the organic production,
structure, increment, and yield of forest stands. Oxford, New York, Permagon

Press, 506 p.

Bakelaar, R. G. and E. P. Odum (1978). "Community and Population Level Responses to
Fertilization in an Old-Field Ecosystem." Ecology 59(4): 660-665.

Baker, F. S. (1949). "A Revised Tolerance Table." J. of Forestry 47(3): 179-181.

 

Baker, W. L. (1989). "Landscape Ecology and Nature Reserve Design in the Boundary
Waters Canoe Area, Minnesota." Ecology 70(1): 23-35.

Barbour, M. G., J. H. Burk, et al. (1999). Terrestrial Plant Ecology. Menlo Park,
Benjamin/Cummings, an imprint of Addison Wesley Longman, Inc., 649 p.

Bassett, J. R. (1984). "Red Pine Plantation Management in the Lake States: A Review."

172

Intensive Forestry Systems Project, IFSIM Publication No. 3.

Bender, L. C., D. L. Minnis, et al. (1997). "Wildlife Responses to Thinning Red Pine."
Northern Journal of Applied Forestry 14(3): 141-146.

 

Benkman, C. W. and A. M. Siepielski (2004). "A keystone selective agent? Pine squirrels
and the frequency of serotiny in lodgepole pine." Ecology 85(8): 2082-2087.

Benzie, J. W. (1977). "Manager's handbook for red pine in the north central states."
USDA Forest Service, North Central Forest Experiment Station (General
Technical Issue Rep. NC-33).

Bergeron, Y. and J. Brisson (1990). "Fire Regime in Red Pine Stands at the Northern
Limit of the Species' Range." Ecology 71(4): 1352-1364.

Bergeron, Y. and D. Gagnon (1987). "Age structure of red pine (Pinus resinosa Ait.) at
its northern limit in Quebec." Canadian Journal of Forest Research 17 : 129-137.

Binkley, D., D. Richter, et al. (1992). "Soil Chemistry in a Loblolly/Longleaf Pine Forest
with Interval Burning." Ecological Applications 2(2): 157-164.

Biswell, H. (1989). Prescribed Burning in California wildlands vegetation management,
Berkeley: University of California Press.

Boemer, R. E. J. (1981). "Forest Structure Dynamics Following Wildfire and Prescribed
Burning in the New Jersey Pine Barrens." American Midland Naturalist 105(2):
321-333.

Boemer, R. E. J ., T. R. Lord, et al. (1988). "Prescribed Burning in the Oak-pine Forest of
the New Jersey Pine Barrens: Effects on Growth and Nutrient Dynamics of Two

Quercus Species." The American Midland Naturalist 120(1): 108-119.

Boring, L. R., C. D. Monk, et al. (1981). "Early Regeneration of a Clear-Cut Southern
Appalachian Forest." Ecology 62(5): 1244-1253.

Bormann, F. H. and G. E. Likens (1979). "Catastrophic Disturbance and the Steady-State
in Northern Hardwood Forests." American Scientist 67(6): 660-669.

Bragg, T. B. and L. C. Hulbert (1976). "Woody Plant Invasion of Unbumed Kansas
Bluestem Prairie." Journal of Range Management 29(1): 19-24.

Brokaw, N. V. L. (1985a). "Gap-Phase Regeneration in a Tropical Forest." Ecology
66(3): 682-687.

Brokaw, N. V. L. (1985b). Treefalls, Regrowth, and Community Structure in Tropical
Forests. The ecology of natural disturbance and patch dynamics. S. T. A. Pickett

173

and P. S. White. San Diego, California, Academic Press: 53-68.

Bruce, D. (1947). "Thirty-two years of Annual Burning in Longleaf Pine." Journal of
Forestry 45: 809-814.

Buchholz, K. and M. Gallagher (1982). "Initial Ectomycorrhizal Density Response to
Wildfire in the New Jersey Pine Barren Plains." Bulletin of the Torrey Botanical
Club 109(3): 396-400.

Buchholz, K. and R. E. Good (1982). "Density, Age Structure, Biomass and Net Annual
Aboveground Productivity of Dwarfed Pinus rigida Mill. from the New Jersey
Pine Barren Plains." Bulletin of the Torrey Botanical Club 109(1): 24-34.

Buckman, R. E. (1962). "Growth and yield of Red Pine in Minnesota." Lake States Forest
Experiment Station, Forest Service Tech Bull US Dep Agric (1272): 50 p.

Buell, M. F. and J. E. Cantlon (1953). "Effects of Prescribed Burning on Ground Cover in
the New Jersey Pine Region." Ecology 34(3): 520-528.

Burdon, R. D. (2002). Introduction to Pines. Pines of Silvicultural Importance
CAB International. New York, NY and Wallingford, UK., Compiled by CAB
International, CABI Publishing.

Burgess, D. and C. Robinson (1998). "Canada's oldest permanent sample plots-thinning
in white and red pine." Forestry Chronicle 74(4): 606-616.

Burns, R. and B. Honkala (1990). Silvics of North America. Washington DC, USDA
Agrculture Handbook: 654.

 

Campbell, R. S. (1955). "Vegetational Changes and Management in the Cutover Longleaf
Pine-Slash Pine Area of the Gulf Coast." Ecology 36(1): 29-34.

Canavan, S. J. and C. W. Ramm (2000). "Accuracy and precision of 10 year predictions
for Forest Vegetation Simulator - Lake States." Northern Journal of Applied
Forestgy 17(2): 62-70.

Canham, C. and P. L. Marks (1985). The Response of woody Plants to Disturbance
Patterns of Establishment and Growth. The ecology of natural disturbance and
patch dynamics. S. T. A. Pickett and P. S. White. San Diego, California,
Academic Press: 197-216 p.

Canham, C. D. and O. L. Loucks (1984). "Catastrophic Windthrow in the Presettlement
Forests of Wisconsin." Ecology 65(3): 803-809.

Chabot, B. F. (1978). "Environmental Influences on Photosynthesis and Growth in
Fragaria vesca." New Phflolgist 80(1): 87-98.

174

Chapman, H. H. (1932). "Is the longleaf type a climax?" Ecology 13: 328-334.

Chapman, H. H. (1947). "Results of a Prescribed Fire at Urania, La., on Longleaf Pine
Land." Journal of Forestry 45(2): 121-123.

Christensen, N. L. (1985). Shrubland Fire Regimes and Their Evolutionary
Consequences. The ecology of natural disturbance and patch dynamics. S. T. A.
Pickett and P. S. White. San Diego, California, Academic Press: 86-97 p.

Christensen, N. L., W. Cronon, et al. (1989). "Landscape history and ecological change."
Journal of Forest Histog 33(3): 116-125.

Christensen, N. L. and C. H. Muller (1975). "Effects of fire on factors controlling plant
growth in Adenostoma Chaparral." Ecological Monographs 45(1): 29-55.

Clark, J. S. (1991). "Disturbance and Tree Life History on the Shifting Mosaic
Landscape." Ecology 72(3): 1102-1118.

Cleland, D. T., T. R. Crow, et al. (2004). "Characterizing historical and modern fire
regimes in Michigan (USA): a landscape ecosystem approach." Landscape

Ecology.

Cleland, D. T., L. A. Larry, et al. (2001). "Ecology and Management of Aspen: A Lake
States Perspective." USDA Forest Service Proceedings RMRS-P-18.

Clements, F. E. (1938). Plant Ecology. NY, USA, McGraw Hill, New York, 601 p.

 

Clements, F. E. (1916). Plant Succession: An Analysis of the Develgment of
Vegetation. Washington, Carnegie Institute of Washington, Publication no. 242,
p. 512

Clements, F. E. (1936). "Nature and Structure of the Climax." The Journal of Ecology
24(1): 252-284.

 

Coffrnan, M. S. (1976). "Thinning highly productive Red Pine plantations." Research
Notes, No. 18, Ford Forestry Center, Michigan Technological University: 10 p.

Collins, B. S., K. P. Dunne, et a1. (1985). Responses of forest herbs to canopy gaps. The
ecology of natural disturbance and patch dynamics. S. T. A. Pickett and A. S.
White. San Diego, Academic Press: 218-234 p.

Connell, J. H. (1978). "Diversity in Tropical Rain Forests and Coral Reefs - High
Diversity of Trees and Corals Is Maintained Only in a Non-Equilibrium State."
Science 199(4335): 1302-1310.

Connell, J. H. and R. O. Slatyer (1977). "Mechanisms of Succession in Natural

175

Communities and Their Role in Community Stability and Organization."
American Naturalist 111(982): 1119-1144.

Connell, J. H. and W. P. Sousa (1983). "On the Evidence Needed to Judge Ecological
Stability or Persistence." The American Naturalist 121(6): 789-824.

 

Cooper, C. F. (1960). "Changes in vegetation, structure, and growth of South-Western
Pine forests since white settlement." Ecological Monograph 30(2): 129-64.

 

Cooper, C. F. (1961). "Pattern in Ponderosa Pine forests." Ecology 42(3): 493-9.

Covington, W. W. (2000). "Helping Western forests heal. The prognosis is poor for US
forest ecosystems." Nature 408.

Covington, W. W., R. L. Everett, et al. (1994). "Historical and Anticipated Changes in
Forest Ecosystems of the Inland West of the United States." Journal of
Sustainable Forestg 2: 13-63.

 

Covington, W. W., P. Z. Fule, et al. (1997). "Restoring ecosystem health in ponderosa
pine forests of the southwest." Journal of Forestry 95(4): 23-29.

Covington, W. W. and M. M. Moore (1994a). "Southwestern Ponderosa Forest Structure
- Changes since Euro-American Settlement." Journal of Forestry 92(1): 39-47.

 

Covington, W. W. and M. M. Moore (1994b). "Postsettlement changes in natural fire
regimes and forest structure: ecological restoration of old-growth ponderosa pine
forests." Journal of Sustainable Forestry 2(2): 153-181.

 

Covington, W. W. and S. S. Sackett (1986). "Effect of Periodic Burning on Soil-Nitrogen
Concentrations in Ponderosa Pine." Soil Science Society of America Journal
50(2): 452-457.

Curtis, J. T. and R. P. McIntosh (1951). "An Upland Forest Continuum in the Prairie-
Forest Border Region of Wisconsin." Ecology 32(3): 476-496.

D'Antonio, C. M. and P. M. Vitousek (1992). "Biological Invasions by Exotic Grasses,
the Grass/Fire Cycle, and Global Change." Annual Review of Ecology and
Systematics 23: 63-87.

Day, M. W. and V. J. Rudolph (1972). "Thinning Plantation Red Pine." Research Report,
151, Michigan State University, Agricultural Experiment Station East Lansing: 10

13.

Day, R. J. (1972). "Stand Structure, Succession, and Use of Southern Alberta's Rocky
Mountain Forest. " Ecology 53(3): 472-478.

176

Debano, L. F. and C. E. Conrad (1978). "The Effect of Fire on Nutrients in a Chaparral
Ecosystem." Ecology 59(3): 489-497.

Denslow, J. S. (1980). "Gap Partitioning among Tropical Rainforest Trees." Biotropica
12(2): 47-55.

Denslow, J. S. (1985). Disturbance-mediated Coexistence of Species. The ecology of
natural disturbance and patch dynamics. S. T. A. Pickett and P. S. White. San
Diego, California, Academic Press: 307-321 p.

Despain, D. G. (1983). "Nonpyrogenous Climax Lodgepole Pine Communities in
Yellowstone-National-Park." Ecology 64(2): 231-234.

Dickman, A. (1978). "Reduced Fire Frequency Changes Species Composition of a
Ponderosa Pine Stand." Journal of Forestry 76: 24-25.

 

Dickmann, D. I. (1993). "Management of red pine for multiple benefits using prescribed
fire." Northern Journal of Applied Forestm 10(2): 53-62.

Dickmann, D. I. and L. A. Leefers (2003). The Forests of Michigan. Ann Arbor,
University of Michigan Press, 297 p.

 

Dixon, G. E. (2002). Essential FVS: A User's Guide to the Forest Vegetation Simulator.
Fort Collins, CO US. Department of Agriculture, Forest Service, Forest
Management Service Center (Last revised: March 2004), 194 p.

Drewa, P.-B., W.-J. Platt, et al. (2002). "F ire effects on resprouting of shrubs in
headwaters of southeastern longleaf pine savannas." Ecology (Washington D Q
83(3): 755-767.

 

Dunning, J. B., B. J. Danielson, et al. (1992). "Ecological Processes That Effect
Populations in Complex Landscapes." Oikos 65(1).

Duvall, M. D. and D. F. Grigal (1999). "Effects of timber harvesting on coarse woody

debris in red pine forests across the Great Lakes states, USA." Canadian Journal
of Forest Research. 29: 1926-1 934.

 

Egler, F. E. (1954). "Vegetation Science Concepts 1. Initial Floristic Composition, A
Factor in Old-field Vegetation Development." Vegetation 4: 412-417.

Elton, C. (1930). Animal Ecology and Evolution. Oxford. The Claredon Press; London.
96 p.

Engstrom, F. B. and D. H. Mann (1991). "Fire ecology of red pine Pinus resinosa in
northern Vermont, USA." Canadian Journal of Forest Research 21: 882-889.

 

177

Feeney, S. R., T. E. Kolb, et al. (1998). "Influence of Thinning and Burning Restoration
Treatments on Presettlement Ponderosa Pines at the Gus Pearson Natural Area."
Canadian Journal of Forest Research 28(9).

Fiedler, C. E., S. F. Arno, et al. (1992). "Management Prescriptions for Restoring
Biodiversity in Inland Northwest Ponderosa Pine-Fir Forests." Northwest
Environmental Journal 8(1): 21 1-213.

F laccus, E. and L. F. Ohmann (1964). "Old-growth Northern Hardwood Forests in
Northeastern Minnesota." Ecology 45(3).

Forman, R. T. T., Ed. (1998). The Pine Barrens of New Jersey: An Ecological Mosaic.
Pine barrens: ecosystem and landscape. New Brunswick, New Jersey, and
London, Rutgers University Press.

Forman, R. T. T. and R. E. Boemer (1981). "Fire Frequency and the Pine Barrens of
New-Jersey." Bulletin of the Torrey Botanical Club 108(1): 34-50.

Forman, R. T. T. and M. Gordon (1981). "Patches and Structural Components for a
Landscape Ecology." BioScience 31(10): 733-739.

Foster, D. R. (1988). "Disturbance History, Community Organization and Vegetation
Dynamics of the Old-Growth Pisgah Forest, South-Westem New Hampshire,
USA." The Journal of Ecology 76(1): 105-134.

 

Frelich, L. E. and C. G. Lorimer (1991). "Natural Disturbance Regimes in Hemlock-
Hardwood Forests of the Upper Great Lakes Region." Ecological Monographs
61(2): 145-164.

F relich, L. E. and P. B. Reich (1995). "Spatial patterns and succession in a Minnesota
southern-boreal forest." Ecological Monographs.

F ule, P. Z., W. W. Covington, et al. (1997). "Determining Reference Conditions for
Ecosystem Management of Southwestern Ponderosa Pine Forests." Ecological
Applications 7(3): 895-908.

Fule, P. Z., A. E. M. Waltz, et al. (2001). "Measuring forest restoration effectiveness in
reducing hazardous fuels." Journ_al of Forestry 99(11): 24-29.

Garren, K. H. (1943). "Effects of Fire on Vegetation of the Southeastern United States."
The Botanical Review 9: 617—654.

Gaston, K. J. and J. H. Lawton (1990). "Effects of Scale and Habitat on the Relationship
Between Regional Distribution and Local Abundance." Oikos 58(3): 329-335.

178

Gauthier, S., Y. Bergeron, et al. (1996). "Effects of fire regime on the serotiny level of
jack pine." Journal of Ecology 84(4): 539-548.

Gilliam, F. S. and N. L. Turrill (1993). "Herbaceous Layer Cover and Biomass in a
Young Versus a Mature Stand of a Central Appalachian Hardwood Forest."
Bulletin of the Torrey Botanical Club 120(4): 445-450.

Gilliam, F. S., N. L. Turrill, et al. (1995). "Herbaceous-Layer and Overstory Species in
Clear-cut and Mature Central Appalachian Hardwood Forests." Ecological
Applications 5(4): 947-955.

 

Gilmore, D. W., B. T. C. O, et al. (2005). "Thinning red pine plantations and the
Langsaeter hypothesis: a Northern Minnesota case study." Northern Journal of
Applied Forestq 22(1): 19-26.

Gilmore, D. W. and B. J. Palik (2006). "A revised Managers handbook for red pine in the
North Central Region. ." USDA Forest Service, North Central Forest Experiment
Station General Technical Report-NC-264: 55 p.

Givnish, T. J. (1980). "Ecological Constraints on the Evolution of Breeding Systems in
Seed Plants: Dioecy and Dispersal in Gymnosperms." Evolution 34(5): 959-972.

Glitzenstein, J. S., P. A. Harcombe, et al. (1986). "Disturbance, Succession, and

Maintenance of Species Diversity in an East Texas Forest. " Ecological
Monographs 56(3): 243—258.

Good, R. E. and N. F. Good (1984). "The Pinelands National Reserve: An Ecosystem
Approach to Management." BioScience 34(3): 169-173.

Grime, J. P. (1977). "Evidence for the Existence of Three Primary Strategies in Plants
and its Relevance to Ecological and Evolutionary Theory." The American
Naturalist 3(982): 1 169-1 194.

Grime, J. P. (1979). Plant Strategies and Vegetation Processes. New York:Wiley.

 

Grubb, P. J. (1977). "The Maintenance of species-richness in plant Communities: The
importance of the regeneration niche." Biol. Rev. 52: 107-145.

Gysel, L. W. (1966). "Ecology of a Red Pine (Pinus Resinosa) Plantation in Michigan."
Ecology 47(3): 465-472.

Haberland, F. P. and S. A. Wilde (1961). "Influence of thinning of Red Pine plantation on
soil." Ecology 42(3): 584-6.

179

Hainds, M. J ., R. J. Mitchell, et al. (1999). "Distribution of Native Legumes
(Legurninosae) in Frequently Burned Longleaf pine (Pinaceae)-wiregrass
(Poaceae) Ecosystems." American Journal of Botany 86(11): 1606-1614.

Halpem, C. B. and T. A. Spies (1995). "Plant species diversity in natural and managed
forests of the Pacific Northwest." Ecological Applications 5(4): 913-934.

Harmon, M. E. (1984). "Survival of Trees After Low-Intensity Surface Fires In Great
Smoky Mountains National Park." Ecology 65(3): 796-802.

Hartman, J. P., D. S. Buckley, et al. (2005). "Differential success of oak and red maple
regeneration in oak and pine stands on intermediate-quality sites in northern
Lower Michigan." Forest Ecology and Management 216: 77-90.

Heinselman, M. L. (1973). "Fire in the Virgin forests of the Boundary Waters Canoe
Area, Minnesota. ." Ouatemary Research 3: 329-3 82.

 

Heinselman, M. L. and H. E. J. Wright (1973). "The Ecological Role of Fire in Natural
Conifer Forests of Western and Northern North America." Quaternary Research
3(3): 317-318.

Henning, S. J. and D. I. Dickmann (1996). "Vegetative Responses to Prescribed Burning
in a Mature Red Pine Stand." Northern Journal of Applied Foresg 13: 140-146.

Henry, J. D. and J. M. A. Swan (1974). "Reconstructing Forest History from Live and
Dead Plant Material--An Approach to the Study of Forest Succession in
Southwest New Hampshire." Ecology 55(4): 772-783.

Heyward, F. (1939). "The relation of fire to stand composition of longleaf pine forests."
Ecology 20(2): 287-304.

Hobbs, R. J. and L. F. Huenneke (1992). "Disturbance, Diversity, and Invasion:
Implications for Conservation." Conservation Biology 6(3): 324-337.

Hodgkins, E. J. (195 8). "Effects of Fire on Undergrowth Vegetation in Upland Southern
Pine Forests." Ecology 39(1): 36-46.

Holling, C. S. (1973). "Resilience and Stability of Ecological Systems." Annual Review
of Ecology and Systematics 4: 1-23.

 

Holling, C. S. (1992). "Cross-Scale Morphology, Geometry, and Dynamics of
Ecosystems." Ecological Monographs 62(4): 447-502.

Horn, H. S. and R. H. M. Arthur (1972). "Competition among Fugitive Species in a
Harlequin Environment." Ecology 53(4): 749-752.

180

Howe, H. F. (1995). "Succession and Fire Season In Experimental Prairie Plantings."
Ecology 76(6): 1917-1925.

Hughes, F., P. M. Vitousek, et al. (1991). "Alien Grass Invasion and Fire In the Seasonal
Submontane Zone of Hawai’i." Ecology 72(2): 743-747.

Hulbert, L. C. (1969). "Fire and Litter Effects in Undisturbed Bluestem Prairie in
Kansas." Ecology 50(5): 874-877.

Hulbert, L. C. (1988). "Causes of Fire Effects in Tallgrass Prairie." Ecology 69(1): 46-58.

Huston, M. (1979). "A General Hypothesis of Species Diversity." The American
Naturalist 113(1): 81-101

Huston, M. and T. Smith (1987). "Plant Succession: Life History and Competition." The
American Naturalist 130(2): 168-198.

Johnson, E. A. and G. I. Fryer (1989). "Population-Dynamics in Lodgepole Pine-
Engelmann Spruce Forests." Ecology 70(5): 1335-1345.

Johnstone, J. F ., F. S. r. Chapin, et al. (2004). "Decadal Observations of Tree
Regeneration Following Fire in Boreal Forests." Canadian Journal of Forest
Research 34(2).

Kaye, J. P. and S. C. Hart (1998). "Ecological Restoration Alters Nitrogen

Transformations in a Ponderosa Pine-bunchgrass Ecosystem." Ecological
Applications 8(4): 1052- 1060.

 

Kilbum, P. D. (1960). "Effects of Logging and Fire on Xerophytic Forests in Northern
Michigan." Bulletin of the Torrey Botanical Club 87(6): 402-405.

Kilgore, B. M. (1973). "The Ecological Role of Fire in Sierran Conifer Forests."
Quartemary Research 3(3): 496-513.

Kilgore, B. M. (1976). "From fire control to fire management: an ecological basis for
policies." Transactions, Forty first North American Wildlife and Natural

Resources Conference, March 21 25, 1976, Washington, DC.

Kittredge, J ., Jr (1934). "Evidence of the rate of forest succession on Star Island,
Minnesota." Ecology 15(1): 24-35.

Kittredge, J. and A. K. Chittenden (1929). "Oak forests of northern Michigan." Michigan
Agr.Exp.Sta. Spec. Bull. (190): 47 p.

181

Kozlowski, T. T. and T. A. Peterson (1962). "Seasonal Growth of Dominant,
Intermediate, and Suppressed Red Pine Trees." Botanical Gazette 124(2): 146-
154.

Lancaster, K. F. and W. B. Leak (1978). "A Silvicultural Guide for White Pine in the
Northeast." Forest Service General Technical Report NE-4ly Forest Service. US.
Department of Agriculture, Northeastern Forest Experiment Station: 1-13.

Larsen, J. A. (1929). "Fires and Forest Succession in the Bitterroot Mountains of
Northern Idaho." Ecology 10(1).

Leak, W. B., J. B. Cullen, et al. (1995). "Dynamics of White Fine in New England."
Research Paper, Northeastern Forest Experiment Station, USDA Forest Service.

 

Lemon, P. C. (1949). "Successional Responses of Herbs in the Longleaf-slash Pine Forest
After Fire." Ecology 30(2).

Levin, S. A. (1976). "Population Dynamic Models in Heterogeneous Environments."
Annual Review of Ecology and Systematics 7: 287-310.

Levin, S. A. and R. T. Paine (1974). "Disturbance, Patch Formation, and Community
Structure." Proceedings of the National Academy of Sciences of the United States
of America 71(7): 2744-2747.

Likens, G. E., F. H. Bormann, et al. (1978). "Recovery of a Deforested Ecosystem."
Science 199(4328): 492-496.

Likens, G. E. and K. F. Lambert (1998). "The Importance of Long-Terrn Data in
Addressing Regional Environmental Issues." Northeastern Naturalist 5(2): 127-
136.

 

Little, S. (1974). Effects of fire on temperate forests: Northeastern United States. Fire and

Ecosystems. T. Kozlowski and C. Ahlgren. New York, New York Academic
Press: 225-250.

Little, S. (1998). Fire and Plant Succession in the New Jersey Pine Barrens. Pine Barrens:
Ecosystem and Landscape. R. T. T. F orman. New Brunswick, New Jersey, and
London, Rutgers University Press: 601 p.

Little, S. and E. B. Moore (1945). "Controlled Burning in South Jersey's Oak-Pine
Stands." Ecology 43: 499-506.

Little, S. and E. B. Moore (1949). "The Ecological Role of Prescribed Burns in the Pine-
Oak Forests of Southern New Jersey." Ecology 30(2): 223-233.

182

Loope, L. L. and G. E. Gruell (1973). "The Ecological Role of Fire in the Jackson Hole
Area, Northwestern Wyoming." Quaternary Research 3(3): 425-443.

Lorimer, C. G. (1977). "The Presettlement Forest and Natural Disturbance Cycle of
Northeastern Maine." Ecology 58(1): 139-148.

Lorimer, C. G., J. W. Chapman, et al. (1994). "Tall Understorey Vegetation as a Factor in
the Poor Development of Oak Seedlings beneath Mature Stands." Journal of
Ecology 82(2): 227-237.

Loucks, O. L. (1970). "Evolution of Diversity, Efficiency, and Community Stability."
American Zoologist 10(1): 17-25.

Lundgren, A. L. (1965). "Alignment chart for numbers of trees-diameters-basal areas."
US For Serv Res Note Lake St For Exp Sta.

Lundgren, A. L. (1965). "Thinning Red Pine for high investment returns." US For Serv
Res Pap Lake St For Exp Sta.

Lundgren, A. L. (1981). "The effect of initial number of trees per acre and thinning
densities on timber yields from red pine plantations in the Lake States." USDA
Forest Service Research Paper, NC-l93, USDA North Central Forest Experiment
Station: 25p.

Lundgren, A. L. (1983). "New site productivity estimates for red pine in the Lake States."
Journal of Forestry.

Lutz, H. J. (1934). "Ecological relations in the pitch pine plains of southern New Jersey."
Yale School of Forestry Bulletin-38: 80 p.

Mac Arthur, R. H. and E. O. Wilson (1967). The Theory of Island Biogeography.
Princeton, New Jersey, Princeton University Press, 203.

Magurran, A. E. (1988). Ecological Diversity and Its Measurement. Princeton, New
Jersey, Princeton University Press, 179.

Maissurow, D. K. (1935). "Fire as a necessary factor in the perpetuation of white pine."
Journal of Forestfl 33(4): 373-378.

Mallik, A. U. and B. A. Roberts (1994). "Natural Regeneration of Pinus resinosa on
Burned and Unbumed Sites in Newfoundland." Journal of Vegetation Science
5(2): 179-186.

Marks, P. L. (1974). "The Role of Pin Cherry (Prunus pensylvam'ca L.) in the

Maintenance of Stability in Northern Hardwood Ecosystems." Ecological
Monographs 44(1): 73-88.

183

Marks, P. L. and F. H. Bormann (1972). "Revegetation following Forest Cutting:
Mechanisms for Return to Steady-State Nutrient Cycling." Science 176(403 7):
914-915.

Marty, R. (1988). "Red pine spacing and diameter growth." Northern Journal of Applied
Forestgy 5: 224-225.

McConkey, T. W. and D. R. Gedney (1951). "A guide for salvaging white pine injured by
forest fires." Norhteastem Research Notes, Northeastern Forest Experiment
Station, Forest Service, USDA.

 

McGuire, J. P., R. J. Mitchell, et al. (2001). "Gaps in a gappy forest: plant resources,
longleaf pine regeneration, and understory response to tree removal in longleaf
pine savannas." Canadian Journal of Forest Research 31: 765-778.

 

McRae, D. J ., T. J. Lynham, et al. (1994). "Understory prescribed burning in red pine and
white pine." Forestry Chronicle 70(4): 395-401.

Menges, E. S. and C. V. Hawkes (1998). "Interactive Effects of Fire and Microhabitat on
Plants of Florida Scrub." Ecological Applications 8(4): 935-946.

 

Merrens, E. J. and D. R. Peart (1992). "Effects of Hurricane Damage on Individual
Growth and Stand Structure in a Hardwood Forest in New Hampshire, USA." I_h;e_
Journal of Ecology 80(4): 787-795.

 

Metheven, I. R. and W. G. Murray (1974). "Using fire to eliminate understory balsam fir
in pine management." F orestfl Chronicle 50: 77-79.

Mohler, C. L., P. L. Marks, et al. (1978). "Stand Structure and Allometry of Trees during
Self-Thinning of Pure Stands." Journal of Ecology 66(2): 599-614.

Moir, W. H. (1966). "Influence of Ponderosa Pine on Herbaceous Vegetation." Ecology
47(6): 1045-1048.

Moir, W. H. (1969). "The Lodgepole Pine Zone in Colorado." American Midland
Naturalist 81(1): 87-98.

Mutch, R. W. (1970). "Wildland Fires and Ecosystems--A Hypothesis." Ecology 51(6):
1046-1051.

Neumann, D. D. and D. I. Dickmann (2001). "Surface burning in a mature stand of Pinus
resinosa and Pinus strobus in Michigan: effects on understory vegetation."
International Journal of Wildland Fire 10: 91-101.

 

184

Nicholson, S. A. and C. D. Monk (1975). "Changes in several community characteristics
associated with forest formation in secondary succession." American Midland
Naturalist 93(2): 302-310.

Niering, W. A. (1987). "Vegetation Dynamics (Succession and Climax) in Relation to
Plant Community Management." Conservation Biology 1(4): 287-295.

Niering, W. A. and G. D. Dreyer (1989). "Effects of Prescribed Burning on Andropogon
scoparius in Postagricultural Grasslands in Connecticut." American Midland
Naturalist 122(1): 88-102.

Niering, W. A., R. H. Goodwin, et al. (1970). Prescribed Burning in Sourthem New
England: Introduction to Long-Range Studies. 10th Annual Tall Timbers Fire
Ecology Conference, Tallahasse, Florida. .

Nyland, R. D. (1996). Silviculture: concepts and applications. New York, McGraw-Hill.
633 p.

 

Oosting, H. J. (1944). "The Comparative Effect of Surface and Crown Fire on the
Composition of a Lob-lolly Pine Community." Ecology 25(1): 61-69.

Parker, W. C., K. A. Elliott, et al. (2001). "Managing succession in conifer plantations:
converting young red pine (Pinus resinosa Ait.) plantations to native forest types
by thinning and underplanting." Forestry Chronicle 77(1): 129-139.

 

Peet, R. K. and N. L. Christensen (1987). "Competition and Tree Death." BioScience
37(8): 586-595.

Penner, M., C. Robinson, et al. (2001). "Pinus resinosa product potential following initial
spacing and subsequent thinning." Forestry Chronicle. 77(1): 129-139.

Peterson, C. J. and S. T. A. Pickett (1995). "Forest Reorganization: A Case Study in an
Old-Growth Forest Catastrophic Blowdown." Ecology 76(3): 763-774.

Phillips, D. L. and D. J. Shure (1990). "Patch-Size Effects on Early Succession in
Southern Appalachian F orests." Ecology 71(1): 204-212.

Pickett, S. T. A. (1980). "Non-Equilibrium Coexistence of Plants." Bulletin of the Torrey
Botanical Club 107(2): 238-248.

 

Pickett, S. T. A. and J. N. Thompson (1978). "Patch dynamics and the design of nature
reserves." Biological Conservation 13(1): 27-37.

 

Pickett, S. T. A. and P. S. White (1985). Patch Dynamics: A Synthesis. The ecology of
natural disturbance and patch dynamics. S. T. A. Pickett and P. S. White. San
Diego, California, Academic Press: 371-384

185

Pyne, S. J. (1982). Fire in America Princeton, New Jersey, Princeton University Press,
654 p.

Pyne, S. J. (2004). Tending Fire. Washington, Covelo, London, Island Press/Shear Water
Books.

 

Quarterman, E. and C. Keever (1962). "Southern Mixed Hardwood Forest - Climax in
Southeastern Coastal Plain, USA." Ecological Monographs 32(2): 167-185.

Raison, R. J. (1979). "Modification of the Soil Environment by Vegetation Fires, with
Particular Reference to Nitrogen Transformations: A review." Plant and Soil 51:
73-108.

Reice, S. R. (1994). "Nonequilibrium Determinants of Biological Community Structure -
Biological Communities Are Always Recovering from the Last Disturbance -
Disturbance and Heterogeneity, Not Equilibrium, Generate Biodiversity."
American Scientist 82(5): 424-435.

 

Reich, P. B., M. D. Abrams, et al. (1990). "F ire Affects Ecophysiology and Community
Dynamics of Central Wisconsin Oak Forest Regeneration." Ecology 71(6): 2179-
2190.

Reich, P. B., P. Bakken, et al. (2001). "Influence of Logging, Fire, and Forest Type on
Biodiversity and Productivity in Southern Boreal F orests." Ecology 82(10): 2731-
2748.

Roberts, B. A. and A. U. Mallik (1994). "Responses of Pinus resinosa in Newfoundland
to Wildfire." Journal of Vegetation Science 5(2): 187-196.

Roberts, M. R. and F. S. Gilliam (1995). "Disturbance Effects on Herbaceous Layer
Vegetation and Soil Nutrients in Populus Forests of Northern Lower Michigan."
Journal of Vegetation Science 6(6): 903-912.

Roberts, M. R. and F. S. Gilliam (1995). "Patterns and Mechanisms of Plant Diversity in
Forested Ecosystems: Implications for Forest Management." Ecological
Applications 5(4): 969-977.

 

Romme, W. H. (1982). "Fire and Landscape Diversity in Subalpine Forests of
Yellowstone National Park." Ecological Monographs 52(2): 199-221.

 

Rouse, C. (1988). "Fire effects in northeastern forests: red pine." General Technical
Report North Central Forest Experiment Station, USDA Forest Service: 9.

Rowe, J. S. (1961). "Critique of some vegetational concepts as applied to forests of
northwestern Alberta." Can. J. Botany 3: 1007-1017.

186

Rowe, J. S. and G. W. Scotter (1973). "Fire in the Boreal Forest." Quartemary Research
3: 444-464.

Roxburgh, S. H., K. Shea, et a1. (2004). "The Intermediate Disturbance Hypothesis: Patch
Dynamics and Mechanisms of Species Coexistence." Ecology 85(2): 359-371.

Rudolph, V. J. (1957). "Silvical characteristics of Red Pine [Pinus resinosa Ait.] " Lake
States Forest Experiment Station paper number 44, USDA Forest Service 1-32.

Runkle, J. R. (1981). "Gap Regeneration in Some Old-Growth Forests of the Eastem-
United-States." Ecology 62(4): 1041-1051.

Runkle, J. R. (1985). Disturbance Regimes in Temperate Forests. The ecology of natural
disturbance and patch dynamics. S. T. A. Pickett and P. S. White. San Diego,
California, Academic Press.

Rykiel, E. J. (1985). "Towards a Definition of Ecological Disturbance." Australian
Journal of Ecology 10(3): 361-365.

Scheller, R. M., D. J. Mladenoff, et al. (2005). "Simulating the effects of fire
reintroduction versus continued fire absence on forest composition and landscape
structure in the Boundary Waters Canoe Area, Northern Minnesota, USA."
Ecosystems. 81396-411.

Schoennagel, T., M. G. Turner, et al. (2003). "The influence of fire interval and serotiny

on postfire lodgepole pine density in Yellowstone National Park." Ecology
84(11): 2967-2978.

Schowalter, T. D. (1985). Adaptations of Insects to Disturbance. The ecology of natural
disturbance and patch dynamics. S. T. A. Pickett and P. S. White. San Diego,
California, Academic Press.

 

Selleck, G. W. (1960). "The Climax Concept." The Botanical Review 26(4): 434-545.
Smith, D. M. (1986). The practice of silviculture. New York, John Wiley & Sons, 527 p.

Smith, D. M. (2003). "Effect of method of thinning on wood production in a red pine
plantation." Northern Journal of Applied Forestry 20(1): 39-42.

Smith, D. M. and P. M. S. Ashton (1993). "Early dominance of pioneer hardwood after
clearcutting and removal of advanced regeneration." Northern Journal of Applied
Forestgy 10(1): 14-19.

 

Smith, D. W. (1970). "Concentrations of Soil, Nutrients Before and After Fire." Canadian
Journal of Soil Science 50: 17-29.

187

 

Smith-Mateja, E. E. (2003). Validation of the forest vegetation simulator growth and
mortality predictions on red pine in Michigan. Forestgy. East Lansing, Michigan
State University. M.S.: 55 p.

Sousa, W. P. (1979). "Disturbance in Marine Inter-Tidal Boulder Fields - the Non-
Equilibrium Maintenance of Species-Diversity." Ecology 60(6): 1225-1239.

Sousa, W. P. (1984). "The Role of Disturbance in Natural Communities." Annual Review
of Ecology and Systematics 15: 353-391.

Spurr, S. H. (1956). "Natural Restocking of Forests following the 1938 Hurricane in
Central New England." Ecology 37(3): 443-451.

Spurr, S. H. and B. V. Barnes (1980). Forest Ecology. John Wiley & Sons, Inc. 687 p.

Spurr, S. H., L. J. Young, et al. (1957). "Nine Successive Thinnings in a Michigan White
Pine Plantation." J For 55(1): 7-13.

Startwell, C. and R. E. Stevens (1975). "Mountain Pone Beetle In Ponderosa pine:
prospects for silvicultural control in second-growth stands." Journ_al of Forestry:
136-141.

Stephens, E. P. and S. H. Spurr (1947). "The immediate response of red pine to thinning
and pruning." Proc. Society of American Foresters, Annual Meeting,
Minneapolis, MN: 353-369.

Stephenson, S. N. (1965). "Vegetation Change in the Pine Barrens of New Jersey."
Bulletin of the Torrey Botanical Club 92(2): 102-114.

Stiell, W. M., C. F. Robinson, et al. (1994). "20-year growth of white pine following
commercial improvement cut in pine mixedwoods." Forestry Chronicle 70(11):
385-394.

Sullivan, T. P., R. A. Lautenschlager, et al. (1999). "Clearcutting and Burning of
Northern Spruce-Fir Forests: Implications for Small Mammal Communities." T_he
Journal of Applied Ecology 36(3): 327-344.

Sutherland, E. K., W. W. Covington, et al. (1991). "A Model of Ponderosa Pine Growth-
Response to Prescribed Burning." Forest Ecology and Management 44(2-4): 161-
173.

Taylor, A. H. and M. N. Solem (2001). "F ire regimes and stand dynamics in an upper

montane forest landscape in the southern Cascades, Caribou Wilderness,
California." Journal of the Torrey Botanical Society 128(4): 350-361.

188

Thomas, P. A. and R. W. Wein (1985). "The influence of shelter and the hypothetical
effect of fire severity on the postfire establishment of conifers from seed."
Canadian Journal of Forest Research 15: 148-155.

Tilman, D., P. Reich, et al. (2000). "Fire suppression and ecosystem carbon storage."
Ecology 81(10): 2680-2685.

Turner, M. G. (1989). "Landscape Ecology: The Effect of Pattern on Process." Annual
Review of Ecology and Systematics 20: 171-197.

Van Dyck, M. G. (2002). Keyword Reference Guide for the Forest Vegetation Simulator.
Fort Collins, CO US. Department of Agriculture, Forest Service, Forest
Management Service Center.

Van Wagner, C. E. (1970). "Fire and Red Pine." Proc Tall Timbers Fire Ecol Conf.

Veblen, T. T. (1985). Stand dynamics in Chilean Nothofagus Forests. The ecology of
natural disturbance and patch dynamics. S. T. A. Pickett and P. S. White. San
Diego, California, Academic Press: 35-52.

Veblen, T. T., T. Kitzberger, et al.(2000). Climatic and Human Influences on Fire

Regimes in Ponderosa Pine Forests in the Colorado Front Range. Ecological
Applications 10 (4): 1178-1195.

Vitousek, P. M. (1985). Community Turnover and Ecosystem Nutrient Dynamics. 11$
ecology of natural disturbance and patch dynamics. S. T. A. Pickett and P. S.
White. San Diego, California, Academic Press: 325-332

Vitousek, P. M., J. R. Gosz, et al. (1979). "Nitrate Losses from Disturbed Ecosystems."
Science 204(4392): 469-474.

Vogl, R. J. (1969). "One Hundred and Thirty Years of Plant Succession in a Southeastern
Wisconsin Lowland." Ecology 50(2): 248-255.

Volk, T. A. and T. J. Fahey (1994). "Fifty-Three Years of Change in an Upland Forest in
South-Central New York: Growth, Mortality and Recruitment." Bulletin of the
Torrey Botanical Club 121(2): 140-147.

Walker, L. C. (1980). The Southern Pine Region. Regional Silviculture of The United
States. J. W. Barrett. New York. Chichester. Brisbane. Toronto, A Wiley-
Interscience Publications: 231-276.

Wan, S., D. Hui, et al. (2001). "Fire Effects on Nitrogen Pools and Dynamics in

Terrestrial Ecosystems: A Meta-Analysis." Ecological Applications 11(5): 1349-
1365.

189

Watt, A. S. (1947). "Pattern and process in the plant community." J Ecol 35(1/2): 1-22.

Westveld, R. H. (1949). Applied Silviculture in the United States, John Wiley & Sons,
Inc.,590 p.

White, A. S. (1983). "The Effects of Thirteen Years of Annual Prescribed Burning on a
Quercus ellipsoidalis Community in Minnesota." Ecology 64(5).

White, A. S. (1985). "Presettlement Regeneration Patterns in a Southwestern Ponderosa
Pine Stand." Ecology 66(2): 589-594.

White, P. S. (1979). "Pattern, Process, and Natural Disturbance in Vegetation." Botanical
Review 45(3): 229-299.

White, P. S. and S. T. A. Pickett (1985). Natural disturbance and patch dynamics: An
Introduction. The ecology of natural disturbance and patch dynamics. S. T. A.
Pickett and P. S. White. San Diego, California, Academic Press: 3-9.

Whitney, G. G. (1986). "Relation of Michigan's Presettlement Pine Forests to Substrate
and Disturbance History." Ecology 67(6): 1548-1559.

Whitney, G. G. (1987). "An Ecological History of the Great Lakes Forest of Michigan."
The Journal of Ecology 75(3): 667-684.

Whittaker, R. H. (1953). "A Consideration of Climax Theory: The Climax as a
Population and Pattern." Ecological Monographs 23(1): 41-78.

Whittaker, R. H. (1956). "Vegetation of the Great Smoky Mountains." Ecological
Monographs 26(1): 1-80.

Wooton, J. T. (1998). "Effects of Disturbance on Species Diversity: A Multitropic
Perspective." The American Naturalist 152(6): 803-805.

Wright, R. F. (1976). "The Impact of Forest Fire on the Nutrient Influxes to Small Lakes
in Northeastern Minnesota." Ecology 57(4): 649-663.

Wu, J. and S. A. Levin (1994). "A Spatial Patch Dynamic Modeling Approach to Pattern
and Process in an Annual Grassland." Ecological Monographs 64(4): 447-464.

Wu, J. and O. L. Loucks (1995). "From Balance of Nature to Hierarchical Patch
Dynamics: A Paradigm Shift in Ecology." The Quarterly Review of Biology
70(4): 439-466.

Yeaton, R. I. (1983). "The Successional Replacement of Ponderosa Pine by Sugar Pine in
Sierra Nevada." Bulletin of the Torrey Botanical Club 110(3): 292-297.

190

Zoladeski, C. A. and P. F. Maycock (1990). "Dynamics of the Boreal Forest in
Northwestern Ontario." American Midland Naturalist 124(2): 289-300.

191

111111111

6852

     

lllllllllifliliiflliflijr