Citation
Impacts of mountain pine beetle (dendroctonus ponderosae) and fire disturbances on forest ecosystem carbon dynamics and species composition

Material Information

Title:
Impacts of mountain pine beetle (dendroctonus ponderosae) and fire disturbances on forest ecosystem carbon dynamics and species composition
Creator:
Caldwell, Megan K
Publication Date:
Language:
English
Physical Description:
1 electronic file : ;

Subjects

Subjects / Keywords:
Forest plants -- Carbon content ( lcsh )
Carbon sequestration ( lcsh )
Mountain pine beetle ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
ABSTRACT: Forests play an important role in storing and sequestering carbon, where conifer forests in particular, store more than 33% within the terrestrial carbon pool. Disturbances, such as fire and insects, impact the amount of carbon that can be stored over time in conifer forests. Stand composition and structure, which plays an important role in carbon storage over time, may be altered by these large disturbances. A mountain pine beetle (Dendroctonus ponderosae, MPB) epidemic has impacted lodgepole forests along the Rocky Mountains, and has potentially altered carbon storage and stand composition trajectories in the short and long-term. This research used the Forest Vegetation Simulator (FVS) to quantify the scope and magnitude of the impacts of MPB on carbon storage and stand composition in a 200 year simulation. FVS was initialized with forest inventory tree, advanced regeneration and fuels data collected in 2010 in Grand County, Colorado, where Grand County was the epicenter of the MPB outbreak in the Southern Rocky Mountains. This FVS simulation carbon and stand composition results were compared to a "control" FVS simulation, where all trees killed by MPB were recoded as live to represent the conditions before the major mortality years of the MPB epidemic. The "MPB" simulation and "control" simulation trajectories were also compared to the trajectories of a "Fire" simulation in FVS, to compare MPB disturbance to fire disturbance. Carbon and stand trajectories were altered between the three simulations, showing MPB has altered forest carbon storage and stand structure, which is different from how fire disturbances affect carbon and species composition. There were differences in the trajectories of carbon storage and stand composition between plots based on initial species composition as well. MPB impacts carbon storage on a relatively short temporal scale, and impacts stand composition on a longer time frame. Fire disturbance seems to affect carbon storage in lodgepole forests more drastically, and for a slightly longer time period. These results aid in management for optimal carbon storage while facing a greater potential for coarse-scale disturbances in a changing climate.
Bibliography:
Includes bibliographical references.
Statement of Responsibility:
by Megan K. Caldwell.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
856152595 ( OCLC )
ocn856152595

Auraria Membership

Aggregations:
Auraria Library
University of Colorado Denver

Downloads

This item has the following downloads:


Full Text
IMPACTS OF MOUNTAIN PINE BEETLE (DENDROCTONUS
PONDEROSAE) AND FIRE DISTURBANCES ON FOREST ECOSYSTEM
CARBON DYNAMICS AND SPECIES COMPOSITION
by
Megan K. Caldwell
B.A., University of Colorado Denver, 2009
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Masters of Science
Masters of Science Environmental Science
2012


This thesis for the Masters of Science Environmental Science degree by
Megan K. Caldwell
has been approved for the
Masters of Science Environmental Science
by
Casey Allen, Chair
Jon Barbour
Frederick Chambers
Date
11


Caldwell, Megan, K. (M.S., Environmental Science)
Impacts of Mountain Pine Beetle {Dendroctonus Ponderosae) and Fire
Disturbances on Forest Ecosystem Carbon Dynamics and Species Composition
Thesis directed by Casey Allen.
ABSTRACT
Forests play an important role in storing and sequestering carbon, where
conifer forests in particular, store more than 33% within the terrestrial carbon pool.
Disturbances, such as fire and insects, impact the amount of carbon that can be
stored over time in conifer forests. Stand composition and structure, which plays
an important role in carbon storage over time, may be altered by these large
disturbances. A mountain pine beetle (.Dendroctonus ponderosae, MPB) epidemic
has impacted lodgepole forests along the Rocky Mountains, and has potentially
altered carbon storage and stand composition trajectories in the short and long-
term. This research used the Forest Vegetation Simulator (FVS) to quantify the
scope and magnitude of the impacts of MPB on carbon storage and stand
composition in a 200 year simulation. FVS was initialized with forest inventory
tree, advanced regeneration and fuels data collected in 2010 in Grand County,
Colorado, where Grand County was the epicenter of the MPB outbreak in the
Southern Rocky Mountains. This FVS simulation carbon and stand composition
results were compared to a control FVS simulation, where all trees killed by
m


MPB were recoded as live to represent the conditions before the major mortality
years of the MPB epidemic. The MPB simulation and control simulation
trajectories were also compared to the trajectories of a Fire simulation in FVS, to
compare MPB disturbance to fire disturbance.
Carbon and stand trajectories were altered between the three simulations,
showing MPB has altered forest carbon storage and stand structure, which is
different from how fire disturbances affect carbon and species composition. There
were differences in the trajectories of carbon storage and stand composition
between plots based on initial species composition as well. MPB impacts carbon
storage on a relatively short temporal scale, and impacts stand composition on a
longer time frame. Fire disturbance seems to affect carbon storage in lodgepole
forests more drastically, and for a slightly longer time period. These results aid in
management for optimal carbon storage while facing a greater potential for coarse-
scale disturbances in a changing climate.
The form and content of this abstract are approved. I recommend its publication.
Approved: Casey Allen
IV


DEDICATION
I dedicate this thesis to my beautiful daughter, Aderyn Jade Caldwell.
v


ACKNOWLEDGMENTS
I would like to acknowledge the U.S. Geological Survey Rocky Mountain
Geographic Science Center, the U.S.G.S. Geographic Analysis and Monitoring
and Land-Remote Sensing programs, and thank my primary advisor at U.S.G.S.,
Todd Hawbaker, for providing the funding and guidance for the completion of this
research.
vi


TABLE OF CONTENTS
Figures...................................................................ix
Tables.....................................................................x
Chapter
1. Introduction............................................................1
2. Literature Review.......................................................5
2.1 Terrestrial Carbon Cycling in Forests...............................5
2.2 Forests and Climate Change.........................................11
2.3 Conifer Forests and Carbon Cycling.................................12
2.4 Disturbances and Lodgepole Forests.................................15
2.5 Succession, Carbon and Disturbances on Lodgepole Forests...........21
2.6 Lodgepole Pine Ecology.............................................23
3. Objectives and Uncertainties...........................................25
3.1 Objectives.........................................................25
3.2 Uncertainties and Limitations......................................27
4. Methods................................................................29
4.1 Field measurements and Modeling....................................29
4.2 Study Area.........................................................29
4.3 Field Methods......................................................31
4.4 Modeling Methods...................................................33
4.5 Model Scenarios....................................................34
4.6 Analysis Methods...................................................37
5. Results................................................................42
5.1 2010 Conditions....................................................42
vii


5.2 Impacts of MPB on Carbon Storage..................................46
5.3 Impacts of Simulated Fire Disturbance on Carbon Storage...........47
5.4 Impacts of MPB on Stand Structure.................................54
5.5 Impacts of Simulated Fire Disturbance on Stand Structure..........57
5.6 Variation in Plots for Carbon and Stand Structure Trajectories....57
5.6.1 Variation in Carbon Trajectories between Plot Types ........58
5.6.2 Variation in Stand Composition Trajectories between Plot Types ...64
6. Discussion Conclusions and Recommendations..........................65
6.1 Impacts of MPB on Carbon Storage ..................................66
6.2 Impacts of Simulated Fire Disturbance on Carbon Storage...........67
6.3 Impacts of MPB on Stand Structure.................................68
6.4 Impacts of Simulated Fire Disturbance on Stand Structure .........70
6.5 Conclusions and Recommendations...................................72
6.5.1 Conclusions.................................................72
6.5.2 Recommendations and Management Implications.................73
References ..............................................................74
viii


LIST OF TABLES
Table
5.1 Surveyed Lodgepole Pine Mortality
43
IX


LIST OF FIGURES
Figure
2.1 The Global Carbon Cycle (from CCSP 2007).................................8
2.2 Forest Acres with Tree Mortality (From the United States Forest Service
(USFS) Forest Health Damage Detection Surveys, 2009).........................20
4.1 Study Area...............................................................31
4.2.Field Plots by Dominant Composition Type.................................40
4.3 Plot Types...............................................................41
5.1 Total Stand Carbon........................................................50
5.2 Standing Live Carbon......................................................51
5.3 Standing Dead Carbon......................................................52
5.4 Downed Dead Carbon........................................................53
5.5 (a-c) Average Species Composition by Percent of Basal Area................55
5.6 Beginning and End Species Trajectory......................................56
5.7 Standing Live Carbon by Plot Type.........................................60
5.8 Standing Dead Carbon by Plot Type.........................................61
5.9 Downed Dead Carbon by Plot Type...........................................62
5.10 Total Stand Carbon by Plot Type........................................63
5.11 (a.-c) Species Trajectories between Plot Types.........................65
x


I.
INTRODUCTION
The thesis presented here is intended to address two main research
questions. Forest disturbances can impact carbon cycling and species composition
trajectories over time. First of all, it is important to be able to quantify the extent
and severity of these forest disturbances. The scope and magnitude of how
disturbances alter the carbon storage and species composition trajectories may
vary between forest types and disturbances, which could have important
implications for forest management. This thesis stands to first quantify the extent
and severity of the mountain pine beetle (MPB) epidemic in eastern Grand County,
Colorado. Next, the differences in carbon and species composition trajectories
over time that have occurred as a result of the MPB epidemic in the study area
were quantified out to 2210. Following, the carbon and species composition
trajectory alterations were compared to those resulting from a simulated fire. Fires
are the other common disturbance that typically affects a large expanse in the
study area. The purpose is to quantify how disturbances such as mountain pine
beetle and fire alter stand and carbon storage trajectories over time compared to
undisturbed forests. This research also compared three categories of forest type to
note differences in the trajectories for each scenario between plots, for insight at
landscape and local scales. Differences in carbon storage by the stand composition
in plots were also addressed. Managing for carbon sequestration and storage in
1


Windows-based computer. The two fonts print equally well on either platform so
this issue is only quality of display on the screen.
forests could be enhanced by managing for species composition that stores more
carbon.
This research is important because in order to most efficiently manage
forests, it will be important to know how disturbances alter the stand and carbon
storage trajectories over time. Forests, in particular, store much of the terrestrial
carbon sequestered from the atmosphere. Disturbances threaten this carbon storage
potential. This is an increasing threat as disturbances are projected to increase in
frequency and severity during modeled warming climate scenarios. Forests will
become increasingly threatened by disturbances, which could subsequently
threaten carbon storage, cycling and stand dynamics in both the long and short
term over multiple spatial scales.
Landscape-scale carbon flux and pool estimates reflect the dynamics of
local-scale measurements, as forest stand carbon dynamics are highly variable and
respond individually to differing management and disturbance regimes, as well as
local-scale site characteristics and vegetation (CCSP 2007). Thus, to be able to
understand the processes occurring at a landscape scale, local-scale processes must
first be quantified.
This study incorporated both a field survey in Grand County, Colorado and
modeled vegetation projections to quantify the impacts of an extensive mountain
2


pine beetle outbreak, compare the mountain pine beetle outbreak to undisturbed
conditions and to a simulated fire of similar extent on stand carbon and structure.
Stand carbon and species composition were tracked at the individual tree and plot
level, and averaged across plots as well for a more landscape-level insight.
The basic results of this survey determined that mountain pine beetle has
altered carbon and species trajectories from pre-disturbance conditions. Stand
carbon, however, recovers relatively quickly, where total stand carbon storage
recovers by 2040 and standing live carbon recovers to pre-disturbance storage by
2060. Stand composition trajectories have changed from pre-disturbance
trajectories, from a higher percentage of the basal area being lodgepole pine to
having a more mixed composition with a higher percentage of subalpine fir
especially. There is some variance in carbon storage and species composition
between plots by the type of overstory and understory species present in plots in
2010. Stand composition trajectories seem to rely heavily on remaining live
vegetation present immediately following disturbance. The impacts of insect
disturbance and how they alter stand composition and carbon trajectories differs
from how fire disturbances impact them as well.
This research is organized within this thesis by a review of literature on the
cycling of carbon terrestrially on a global scale, how forests and disturbances the
affect forests may be impacted by climate change, carbon storage in conifer forests
specifically, how disturbances impact lodgepole forests and how carbon storage,
3


species composition and disturbances interact in lodgepole forests. Then, the
objectives of this study and uncertainties within the research are addressed. The
methods are outlined by the study area, field methods and modeling, the specific
model scenarios simulated, and analysis methods. The results and discussion are
laid out by 2010 conditions, investigating the impacts of mountain pine beetle
disturbances on carbon storage, the impacts of fire disturbance on carbon and the
impacts of mountain pine beetle and fire on stand structure. Finally, the results and
discussion are concluded in a summary and references are included at the end.
Appendix I lists the raw plot data for two plots that were measured again in 2011
after an actual fire burned through them.
4


II. LITERATURE REVIEW
2.1 Terrestrial Carbon Cycling in Forests
Carbon is exchanged between and stored within the oceans,
atmosphere and terrestrial carbon systems, as shown in figure 2.1 (Wigley
and Schimel 2000). Carbon flux is the amount of carbon moved between
pools, and carbon storage refers to the amount within a pool that is not in
movement, but is steadily held. The terrestrial carbon cycle plays a
significant role in the global carbon budget, as it is one of the three main
reservoirs within the cycle. Forests cover about a third of earths terrestrial
land mass, fulfilling copious ecological roles (Winjum et al. 1992). One of
these ecological roles is terrestrial carbon storage and sequestration within
forest biomass and soils, where forests house an estimated 45-60% of the
global terrestrial carbon pool (CCSP 2007).Worldwide, forests store about
2.07 x 10 12 Mg (2280 gigatons (Gt)) of carbon, and coniferous forests
alone store more than 33% of terrestrial carbon (Smith et al. 1993, Kashian
2006, CCSP 2007). The main focus of this thesis will be specifically
subalpine conifer forest carbon storage. Terrestrially, 36% of the land area
of North America which accounts for approximately half of the carbon sink
in North America is provided by forest ecosystems which offset excess
atmospheric carbon inputs, such as greenhouse gas emissions (CCSP 2007,
5


Tallis et al. 2008). In 2003, carbon emissions in North America were about
2 billion Mg of carbon but 256 million Mg were offset by forests (Heath
and Smith 2004, Birdsey et al. 2006, CCSP 2007, Pacala et al. 2007,
Goward et al. 2008, Ryan et al. 2010).
In comparison, the global atmospheric carbon pool (estimated for
2003) stores about 7.05 x 1011 Mg (777 Gt) of carbon as carbon dioxide,
where 5.35 x 1011 Mg (590 Gt) of carbon are from non-anthropogenic
terrestrial sources of release, such as geologic features, fires and
decomposition, and 1.69 x 1011 Mg (187 Gt) are from anthropogenic
contributions (Lai et al. 2000, Lai 2004, CCSP 2007). The terrestrial and
atmospheric carbon pools are directly linked through forests, where forests
absorb carbon dioxide and release oxygen back to the atmosphere. An
imbalance between sources and sinks in the terrestrial carbon cycle may
cause a subsequent increase of carbon in the atmosphere, and changes in
Earths climate (CCSP 2007).
Oceanic carbon storage is approximately 50 times greater than the
atmospheric sink and cycling between the ocean and atmosphere occurs on
the time scale of hundreds of years (Prentice et al. 2001). Between the
ocean and atmosphere, there is about a 2 Mg/yr flux in balance overall
from the atmosphere to the ocean (where 90 Mg/yr is transferred to the
atmosphere to the ocean, and 92 Mg/yr is transferred from atmosphere to
6


ocean storage). The largest storage of carbon dioxide in the ocean is the
dissolved inorganic carbon (Siegenthaler and Sarmiento 1993). The
oceanic carbon sequestration capacity may be significantly reduced if the
global climate warms, as it has done in modeled GCM scenarios where
carbon dioxide concentrations were increased. This modeled reduction of
oceanic carbon uptake occurs from increased sea-surface temperature on
carbon dioxide solubility, as well as reduced vertical mixing on carbon
dioxide transport from the surface to deep ocean (Friedlingstein et al.
2001). This reduction in oceanic carbon dioxide uptake increases the need
for protection of terrestrial carbon sinks, which may be easier for
management.
Carbon stored terrestrially could incur changes from climate change
quickly. Increased atmospheric carbon dioxide could possible lead to
increased vegetation and soil carbon (Prentice et al. 2001). This could be
offset through increased disturbances however. This has direct implications
for management, and could possibly be mitigated.
7


;*>p


Plants & Soil [3800- *
O^anc Cmtaan t*
Figure 2.1 The Global Carbon Cycle (from CCSP 2007). The
global carbon cycle, shown with all three primary carbon pools with fluxes.
Black numbers show natural amounts and red numbers are for
anthropogenic mean amounts through the 1990s. Carbon pools are in
brackets, where fluxes are listed without brackets.
A primary sequestration method within the terrestrial carbon cycle
occurs through vegetation, where primary productivity removes carbon
dioxide from the atmosphere through photosynthesis and converts some of
it to biomass (as well as respiring some back to the atmosphere), allowing
for atmospheric carbon to accumulate and become stored in biomass in
8


what is referred to as the above-ground carbon pool (Vose 2006).
Sometimes carbon input into the terrestrial carbon cycle is measured as Net
Ecosystem Production (NEP), or the difference between Gross Primary
Productivity (GPP) and what the plant needs to use for maintenance over
time (Chapin III et al. 2002). The biomass accumulated through this
carbon sequestration eventually falls and decomposes in the detrital carbon
pool. And, over time, it decomposes enough to become part of the soil and
humus, or below-ground carbon pool (Vose 2006). Thus, the amount of
carbon stored in these pools in any ecosystem is ultimately a function of
the rates of primary productivity, respiration, mortality, decomposition
(Monson et al. 2002). Sources and sinks of carbon tend to determine how
much carbon a particular ecosystem can store. A source of carbon outputs
more carbon than it sequesters, and the opposite holds true for a carbon
sink, where inputs outweigh the outputs. Typically, live vegetation is a sink
for carbon. A source is an ecosystem component that decomposes and loses
carbon faster than it can be sequestered. The amount of carbon stored in
each component on an ecosystem can vary widely; however typically, the
most amount of carbon is stored in the aboveground live vegetation and the
belowground components (Chapin III et al. 2002).
Carbon sequestration is listed as an important ecosystem service,
which is to say that it is a societal benefit provided by natural ecosystems.
9


As an ecosystem service, carbon sequestration and storage is extremely
valuable to our society, and would be difficult to regain if lost (Heal 2000).
In order to preserve and protect this ecosystem service, resource managers
need information on how carbon is distributed and stored in specific
ecosystems, as well as which carbon stores are changing and why (Joyce
and Birdsey 2000, Tallis et al. 2008). Some ecosystem service models
quantify carbon sequestration by splitting up the amounts sequestered in
forests into four different pools: aboveground biomass, belowground
biomass, soil, and dead organic matter (Tallis et al. 2008), so this research
will talk about carbon storage in these four pools, generally. It is important
to be able to quantify carbon stores in each pool, and important to recall
that the amount of carbon stored as well as the fluxes within and between
ecosystems. For example, age and type of vegetation in an ecosystem
makes a difference in how much carbon can be stored in biomass. More
mature forests sequester carbon in the aboveground pool more slowly than
their younger counterparts (Chapin III et al. 2002, Monson et al. 2002).
Due to the wide variance of carbon sequestration capacity between
different ecosystems, it is important to investigate carbon storage on local
scale.
10


2.2 Forests and Climate Change
It is uncertain how forests will respond to a changing climate. This
is particularly true in the aboveground biomass pool, which may pressure
changes in the other main carbon pools (CCSP 2007). Forest species ranges
are projected to move northward and higher in elevation, where range
expansions may impact primary ecosystem processes like succession and
disturbance and subsequent changes in carbon cycling for many years into
the future (Joyce and Birdsey 2000). Disturbances may increase in severity
and frequency in a warming climate scenario (IPCC 2007, Bentz et al.
2010, Westerling et al. 2011). Increased drought in some areas due to
climate change can make some species more susceptible to disturbance and
also less productive (Dale et al. 2001). In some modeled warming climate
scenarios, terrestrial ecosystems become a source of carbon dioxide,
causing additional atmospheric carbon to accrue (Kurz et al. 2008, Sitch et
al. 2008). There is a pressing need to understand how the potential of
forests to sequester carbon may change over time in response to some of
these potential impacts of climate change (Newell and Stavins 2000).
Some of the uncertainties present when facing a changing climate and
forest carbon cycling will be addressed later in this paper.
The negative impacts of climate change may be offset by positive
changes for vegetation that could occur from increased atmospheric carbon
11


dioxide levels. For example, productivity of vegetation could occur.
Doubling atmospheric carbon dioxide concentration, under controlled
greenhouse conditions, has increased plant productivity and yield by more
than 30% on average, a 37% decrease in stomatal conductance, which
increases leaf temperature by 1 degree Celsius and decreases
evapotranspiration, even though these percentages may vary by species
(Kimball et al. 1993). However, in lodgepole forests, increased warming
and carbon dioxide could have negative impacts if precipitation does not
increase as well because lodgepole become moisture-stressed and their
ranges shrink (Barrow and Yu 2005, Hamann and Wang 2006, Monserud
2008). The impacts of climate change on vegetation, and lodgepole in
particular, are a function between the potential positive impacts on
productivity and evapotranspiration compared to the potential negative
impacts through disturbances and drought increases. The impacts of
drought and disturbances on forests may negate the potential positive
impacts of increased carbon dioxide.
2.3 Conifer Forests and Carbon Cycling
Coniferous forests cover approximately 15% of terrestrial land
mass, equivalent to 10 million km in area (Thorsell and Sigaty 1997).
Conifer forests contain 33% of all stored carbon in terrestrial ecosystems
and sequester much of the carbon in western North America (Smith et al.
12


1993, Kashian 2006) Of the conifer forests in North America, and in the
western U.S. in particular, one of the most prevalent forest types is the
lodgepole pine (Pinus contorta), covering 20 million hectares (ha) of area
and 6 million ha are dominated by lodgepole in the western United States
(Lotan and Critchfield 2004). Within Rocky Mountain forests, the below-
ground pool in lodgepole forests tends to store the greatest amount of
carbon (49%), followed by the above-ground vegetation (38%), detrital
(12%), and understory herbaceous cover (1%) vegetation pools
respectively (Birdsey 1992). Carbon storage in forested ecosystems varies
with forest age and stand composition (Bradford et al. 2008). Carbon flux
in a lodgepole forest has been shown to depend on climate and the spatial
distribution of trees (Kueppers and Harte 2005). Although the range of
carbon storage in lodgepole forests varies between sites, generally, the
aboveground live carbon pool in a lodgepole-dominated forest contains
around 67.9 Mg of carbon per hectare on average in the Rocky Mountain
region (Birdsey 1992). The concentrated carbon in plant tissues is about
50% for overstory conifer trees, 45% for herbaceous tissues often found in
a forest understory, but is highest in lipid-rich tissues from plant material,
such as seeds. Carbon storage in lodgepole forests is affected by several
factors, including stand density, stand age, species distribution, and is
essentially a product of the balance between carbon on the forest floor and
13


sequestration in biomass and the carbon lost through decomposition
(Kashian 2006).
Forest soils are one of the largest carbon storage components
(Birdsey 1992). Carbon is incorporated into soil in the subalpine forest
ecosystem from the flux of biomass from the live to the dead pool.
Decomposition, a major component in the carbon cycle, is mainly carried
out by fungi in lodgepole forests. Sap rot fungus is responsible for breaking
down much of lodgepole litter. Fungi is responsible for much of the
decomposition in subalpine forests because often soil microbes cannot
survive in the acidic soils produced under lodgepole forests or cold
temperatures at high elevations (Son 2010). Woody material in a forest is
comprised of mainly lignin. This high lignin content combined with
climatic factors and soil acidity cause decomposition in subalpine forest
soils to be very slow. The biological capacity of an ecosystem to
decompose organic matter is affected by organic inputs such as litterfall or
excretion by organisms in the ecosystem, which subsequently exerts
control on the rate of decomposition (Marschner and Rengel 2007).
Aboveground carbon storage in forest biomass is especially
threatened by climate change. Small changes in the delicate balance
between photosynthesis and respiration and decomposition could result in
increased emissions to the atmosphere (Pregitzer and Euskirchen 2004).
14


2.4 Disturbances and Lodgepole Forests
Disturbances, as above-mentioned, are one factor that is expected to
increase in severity and frequency under warming climate scenarios (Dale
et al. 2001, IPCC 2007). Forests are periodically affected by disturbance
(Roe and Amman 1970, Romme et al. 1986), or an episodic cause of
carbon loss from many ecosystems, defined by Chapin III in 2002.
Immediately following a disturbance, a forest can not only discontinue
sequestering carbon, but can even act as a source instead of a sink for
carbon, and the time it takes for the forest to once again return to being a
sink varies by forest type and stand age or disturbance history (Bradford et
al. 2008, Goward et al. 2008, Kurz et al. 2008). Disturbances can alter
carbon cycling dynamics considerably by causing extensive tree mortality,
reducing photosynthetic capacity and potentially altering carbon flux rates
between the various carbon pools (Kurz et al. 2008). Resulting changes in
biomass pools and environmental conditions may alter rates of
decomposition and further alter rates of other ecosystem processes, like
regeneration (Dale et al. 2001). The cumulative effect is that forests can
shift from being carbon sinks to carbon sources over long temporal scales
(Kurz et al. 2008, Raffa et al. 2008). Some disturbance events that affect
forests include fire (Romme 1982), wind blowdown (Veblen et al. 1989),
15


insect mortality (Roe and Amman 1970), or human management events,
such as thinning and harvesting (Franklin et al. 2002).
Wildfire and insect disturbances appear to impact the largest
acreage in forests in the western United States, and in the Rocky Mountain
coniferous lodgepole forests mentioned above (Dale et al. 2001, USFS
2009). Fire disturbances may vary in response to a warming climate as the
frequency, size and intensity of fire is directly dependent on weather and
precipitation patterns, as well as forest type (Dale et al. 2001, Westerling et
al. 2006). Fires are generally products of climate and available fuels
(Schmoldt et al. 1999). The most extreme fire events tend to burn the most
area and are controlled primarily by climate, where drought causes extreme
fire seasons (Bessie and Johnson 1995). The seasonal severity rating of
forest fires is modeled to increase by 10-15% depending on location by
2060, which increases forest fire activity, where fires are one of the most
rapid disturbances to respond to a warming climate (Flannigan et al. 2000).
Forest fires have burned increasing amounts of acreage in the last decade
which can be attributed to climatic factors, such as increased seasonal
temperatures and earlier snowmelt, suggesting management may not be
effective if the climate continues on a warming trend (Westerling et al.
2006).
16


Aboveground carbon storage, such as in the aboveground live
biomass and in the top layers of forest floor and downed-dead pools, could
be released and turned into a carbon source if consumed by a high-intensity
fire (Breshears and Allen 2002, Hurtt et al. 2002, Kashian 2006, Hurteau
and North 2008, 2009). Carbon is released directly to the atmospheric
carbon pool (Flannigan et al. 2000, Rapp 2004). Fire can disturb carbon
storage in soils both through acceleration of nutrient cycling and changes in
the top soil layer chemistry (Whelan 1995, Dale et al. 2001, Swift 2001).
Insect disturbances may be directly influenced by a warming
climate, where the spread and range of insects, as well as the susceptibility
of forests to insects typically increase under modeled warming climates
(Dale et al. 2001, Raffa et al. 2008, Klutsch et al. 2009). Among other
factors, forests become more susceptible to insects through the stress of
drought in a warming climate (Amman 1977). Insects that disturb forests
typically have been controlled by climatic conditions, so as climate
changes, these controls are relaxed. Bark beetles, for example, may expand
ranges northward, eastward, and toward higher elevations, generally
(Carroll et al. 2003, Safranyik et al. 2010). This would correspond with
forest species range expansions, possibly creating opportunities for greater
scale disturbances, and great impacts to carbon cycling and future species
succession trajectories (Joyce and Birdsey 2000).
17


Insect disturbances, such as the mountain pine beetle
(.Deondroctonusponderosae; MPB), typically impact forests differently
than other disturbances because infestation occurs in selectively larger-
diameter trees (Amman 1977), potentially having different impacts to
carbon storage and flux than other disturbances. MPB infests in areas
where lodgepole basal area is high (Klutsch et al. 2009, Pfeifer et al. 2011).
Other species of trees, as well as smaller DBH lodgepole persist in stands,
where large gaps are left once trees killed by MPB fall. Some level of live
biomass is maintained as compared to a stand replacing fire, which
eliminates most of the live biomass over the affected area. Wildfire
disturbance transfers carbon out of the ecosystem to the atmosphere and
within the ecosystem in the form of ash to the forest floor, while mountain
pine beetle transfers carbon to the standing dead and downed dead pools as
detritus and eventually carbon is transferred to the atmosphere through
decomposition on the forest floor.
MPB was the top forest mortality agent in the conterminous United
States for 2009, accounting for 73% of tree mortality in the conterminous
United States accounting for almost 3,500,000 hectares, as shown in figure
2.2 (USFS 2009). Historically, MPB has persisted at endemic levels in the
Southern Rockies, with periodic outbreaks (Amman 1977, Baker and
Veblen 1990, Raffa et al. 2008, Klutsch et al. 2009). Starting in 1996,
18


MPB populations have grown rapidly to epidemic levels that are
unprecedented in recorded history in the Rocky Mountains (USFS 2009).
The current epidemic has impacted millions of hectares (ha) of lodgepole
pine (Pinus contorta) across North America and 777,000 ha in Colorado
between 2000 and 2008 (USFS 2009). Pfeifer et al. in (2010), noted a
short term change in carbon stocks and fluxes after a mountain pine beetle
outbreak in an Idaho lodgepole forest, where there was immediately a
maximum 83% decrease in carbon stocks and 73% in carbon fluxes (or the
rate of carbon sequestration) that were recovered in 25 years or less. Pfeifer
et al. (2010) surveyed 12 plots in an Idaho forest consisting primarily of
lodgepole and Douglas fir (Pseudotsuga menziesii) and modeled the
trajectory of carbon pools and fluxes after MPB killed up to 52% of trees
within plots. Substantial variability of carbon stocks and fluxes resulted
from the size distribution of trees within the 12 plots.
19


Figure 2.2 Forest Acres with Tree Mortality (From the United States
Forest Service (USFS) Forest Health Damage Detection Surveys,
2009). Shows the acres of mortality, where 73% was from MPB in 2009.
Landscape-scale estimates of carbon storage change seem to
depend heavily on the time since any disturbance (CCSP 2007). Both
anthropogenic and natural disturbances can alter carbon cycling
considerably by removing biomass and altering flux rates among biomass
pools, but in spite of the recognized importance of disturbances, their
potential long-term impacts on carbon cycling has not been quantified
extensively and incorporated into land models that project carbon over time
(Running 2008). Not only does disturbance impact short term carbon
storage, it also could impact future stand trajectories and regeneration well
20


into the future. In turn, species shifts could cause a long-term shift in
carbon storage.
Disturbances like the MPB and fire can affect species succession
trajectories over long time periods. Forest species composition and tree age
are important in determining how quickly carbon storage and sequestration
recovers from MPB disturbances (Fahey and Knight 1986). Carbon content
stored within and between species in forests can vary significantly (Chapin
III et al. 2002, Lamlom and Savidge 2003, Kashian et al. 2004). Thus,
efforts to quantify the long-term impacts of MPB on carbon stocks and
fluxes should account for the potential changes in species composition that
may occur following insect outbreaks. However, it is difficult to quantify
or model future forest succession due to the many factors that must be
accounted for including seed dispersal, topography of the landscape,
moisture levels, competition and light availability, soil conditions and
future climate scenarios.
2.5 Succession, Carbon and Disturbances in Lodgepole Forests
Fire and succession have been studied fairly extensively in
lodgepole pine forests. Fires over large areas of vegetated space typically
initiate successions, but these succession events are dependent on seed
sources and number of advanced regeneration (Glenn-Lewin et al. 1992).
Fire is important in establishing new lodgepole forests, and most of the old
21


growth lodgepole forests in North America were established through fire,
especially in the Rockies (Lotan et al. 1985). Lodgepole sometimes
produces serotinous cones, where temperatures above 45 degrees Celsius
caused by fire or sometimes summer surface soil temperatures cause
serotinous cones to open. This varies greatly within the Rocky Mountains,
however. And lodgepole sometimes produces open cones in areas. (Lotan
1976). In the absence of fire, lodgepole can be replaced by more shade
tolerant species (Lotan 1976). Management and fire suppression tend to
cause fuel buildup, causing the potential for a high intensity fire that may
eliminate large amounts of biomass from forest stands if a fire were to
occur (Brown 1975). To recall from above, a high intensity fire is more
likely in a changing climate, which could have substantial impacts on
successional trajectories and carbon storage in lodgepole forests.
Impacts of insect disturbance on forest regeneration and succession
have not been quantified extensively. Lodgepole typically regenerates
abundantly where the mineral seedbed is adequate enough, but can be
hindered by a thick organic layer that tends to inhibit seedling recruitment,
where litter and fuels accumulate on the forest floor after MPB (Lotan and
Perry 1983, Collins et al. 2011). Remaining canopy after MPB may inhibit
lodgepole establishment, and favor the growth of more shade tolerant
species, such as Engelmann spruce and Subalpine fir (Claveau et al. 2002,
22


Collins et al. 2011). Post-MPB lodgepole regeneration is not limited by the
viable seed availability in the serotinous cones left behind (Aoki et al.
2011). Forest recovery after MPB may rely more on seedlings, saplings and
residual live tree biomass, collectively known as advance regeneration,
rather than new seedling recruitment, especially with the influence of a
deep litter layer on the forest floor (Klutsch et al. 2009, Collins et al. 2011).
Pre-epidemic forest conditions are a large determinant of MPB post-
epidemic forest trajectories (Diskin et al. 2011). Modeling advance
regeneration could provide greater understanding of future forest species
composition and carbon storage.
Post disturbance stand species composition is dependent on time
since stand initiation and severity of disturbance as well, where secondary
non-fire disturbances, such as MPB, that are high in severity, in young
stands that have recently incurred fire disturbance, seem to favor lodgepole
pine re-establishment (Sibold et al. 2007).
2.6 Lodgepole Pine Ecology
Lodgepole pine is shade intolerant, and grows in three different
ecological roles: serai, persistent and climax (Roe and Amman 1970).
Lodgepole is able to colonize after disturbances because they have easily
dispersed seeds, can grow where the canopy is open and can grow on non-
ideal sites, such as nutrient-poor soils or steep slopes (Parker and Parker
23


1983). Persistence of lodgepole dominance in a stand is driven by
topographic variables, fire frequency, seed sources, as well as an insect
disturbance. Endemic MPB kills larger DBH lodgepole in 20-40 year
cycles until lodgepole is eliminated from the stand (Amman 1977, Romme
and Knight 1981).
Changes in the dominant canopy of forest stands where gaps are
introduced drive species composition and seedling establishment after a
disturbance (Klutsch et al. 2009), and consequently impacts carbon storage.
Specifically in Rocky Mountain National Park, Colorado, lodgepole pine
distribution is defined by elevation and moisture, and to a lesser extent,
summer soil moisture and sand content, where the transition from
lodgepole forests to subalpine fir forests correlated with summer soil
moisture (Stohlgren and Bachand 1997) The impacts of MPB disturbance
on stand trajectories in Rocky Mountain National Park vary based on pre-
epidemic stand structure and composition, where there was high variance
in future stand trajectories between overstory and understory species types
present in plots before a MPB outbreak (Diskin et al. 2011).
24


III. OBJECTIVES AND UNCERTAINTIES
3.1 Objectives
It is possible that carbon storage potential in live biomass has been
reduced due to the extensive tree mortality in the Southern Rocky
Mountains as a result of an extensive MPB outbreak, and potentially
impact carbon storage over longer time periods through changes in stand
structure and species composition. Ecosystems encompass numerous
complex interactions that define and drive carbon storage and flux, and
future stand composition. Quantifying the long-term impacts of insect
outbreaks can be difficult because multiple processes must be accounted
for including mortality, vegetation regrowth and succession, changes in
primary productivity and decomposition rates, and potential for future
disturbances. One of the best ways to incorporate these multiple processes
is to utilize a modeling approach. Rates of mortality, drivers of growth and
regrowth, potential for future disturbances and sometimes climatic
variables can all be included with the use of ecosystem models. Also, when
there is no historical analogy to events, such as massive epidemic-scale
outbreaks, models can be used to project the potential trajectory after such
event more efficiently than from observation of past and current events
alone.
25


The objectives of this study were to quantify the amount of time
required for aboveground forest carbon storage to recover to pre-MPB
outbreak levels for the actual epidemic that has occurred in a Rocky
Mountain lodgepole forest, as well as to investigate species composition
response to MPB disturbance in order to better understand how carbon
storage will change over various temporal scales in insect-disturbed forests.
Then, the carbon and species succession data corresponding to the MPB
epidemic will be compared to the carbon storage change and species
composition after a simulated fire, to compare the impacts of these two
common natural disturbances. A combination of in-situ field data
collection and vegetation simulation modeling was used to quantify carbon
and stand structure in the Southern Rocky Mountains. It could possibly
take a very long time for carbon storage to regain the carbon storage
potential they would have had if not for the MPB due to high tree
mortality, decomposition of increased litterfall from dead trees, and even
potential species composition change. We compared the trajectories of
vegetation growth and mortality on plots that had been impacted by MPB,
where peak mortality occurred 2006-2008, to the trajectories of the same
plots had there not been epidemic-level beetle activity to measure the
amount of carbon and species composition response through basal area
26


over time to the MPB disturbance. Carbon was tracked in standing live,
standing dead and downed dead pools over time.
3.2 Uncertainties and limitations
There are several uncertainties with this research that must be
addressed. Only aboveground carbon is quantified, where much of the
carbon in forest ecosystems is stored in the soil and root system. Lodgepole
biomass is 20% belowground within the root system, so only part of the
entire system is accounted for (Comeau and Kimmins 1989).
Not only is carbon storage and sequestration affected by stand
density, composition and age, it is also a function of litter decomposition
and fall rates (Kashian et al. 2004). This research did not address
decomposition, fall rates or belowground soil processes. Also, FVS, the
model used for these simulations, does not account for soil type or texture,
which is important in succession and carbon storage.
This research addresses conservative estimates of stand structure
trajectories but does not address future disturbances that could potentially
go along with those future trajectories, such as an increase in other
disturbance insects or disease, such as spruce beetle or mistletoe. It could
be difficult to track disturbances with the uncertainty that occurs with
climate change.
27


Improvements to this modeling method would be to incorporate
soils and belowground processes and future disturbances for an integrated
approach. Also, better, less conservative, regeneration estimates would
have aided in projecting stand trajectories over time, such as utilizing better
predictors from wind dispersal, mineral seed bank data and typical stand
trajectories after MPB disturbance.
28


IV. METHODS
4.1 Field Measurements and Modeling
In order to quantify the impacts of a MPB outbreak, and further the study
of disturbance impacts on carbon storage and succession in the study area, I
combined field data collection with vegetation simulation modeling. Model
simulations can track large amounts of carbon and species composition over time,
and make projections based on current trajectories (Bazzaz 1996). This
methodology was also chosen because there is no historical analogy to the current
mountain pine beetle outbreak, where modeling allows me to initialize the
simulations with current conditions established with field data and then quantify
potential changes in forest vegetation and carbon under a series of scenarios.
4.2 Study Area
The study area was located in eastern Grand County Colorado (105 43'
32" 106 O' 47" W and 3954' 58" 4018' 2" N). Forests in the study area are
even-aged stands of lodgepole pine (Pinus Contorta) with subalpine fir {Abies
Lasiocarpa) seedlings and saplings. Average stand age is approximately 70 years.
A large percentage of the study area is public land where most wildfires are
suppressed. Until recently, the disturbance history of the area consisted of fires of
mixed severity and periodic bark beetle outbreaks at endemic levels. Beginning in
1996, an extensive and severe MPB outbreak started in the Southern Rocky
Mountains- with peak MPB mortality occurring between 2005 and 2008. The
29


epicenter of this outbreak was in Grand County, making it an ideal area for
studying the impacts of MPB and other disturbances on forest vegetation and
carbon storage.
30


106WW
lOeWW
Figure 4.1 Study Area. This map shows the study area within eastern Grand
County, Colorado where 119 field plots were placed in a randomly stratified
sampling scheme and forest inventory data was collected.
4.3 Field methods
Field measurements were collected to characterize forest vegetation present
in 2010, following the peak of MPB mortality. Plot locations were selected using
stratified random sampling. The strata used included a gradient of years since peak
MPB mortality derived from the Forest Flealth and Monitoring Aerial Surveys (1
year, 2-3 years, 4-5 years, and 5+ years), elevation (elevation quartiles), and aspect
(north, south, east, west, and flat), for a total of 80 different strata. Plot locations
were restricted to public lands and areas classified as Rocky Mountain Lodgepole
31


Pine Forest, Southern Rocky Mountain Ponderosa Pine Woodland, Rocky
Mountain Subalpine Dry-Mesic Spruce-Fir Forest and Woodland, Rocky
Mountain Subalpine Mesic-Wet Spruce-Fir Forest and Woodland, Southern Rocky
Mountain Dry-Mesic Montane Mixed Conifer Forest and Woodland, or Southern
Rocky Mountain Mesic Montane Mixed Conifer Forest and Woodland in the
LANDFIRE existing vegetation type layer (Zhu et al. 2006, Rollins 2009). This
sampling scheme was selected because it ensured that the field data captured the
range of variability in biophysical gradients and MPB mortality present in the
study area. 2-3 plots were placed in each stratum randomly, but revised plot
locations based on accessibility; plots in potentially dangerous and inaccessible
locations were manually moved within strata. Ultimately, data was collected at 119
plot locations.
A field crew measured trees, seedling and saplings as well as surface and
canopy fuel loads using the Fire Effects Monitoring and Inventory Protocol:
FIREMON (Lutes et al. 2006). Each plot had a fixed radius of 8 meters. Trees
were defined as woody vegetation with a diameter at breast height (DBH) > 12 cm
and total height > 1.4 m. Plots were overall categorized based on a primary
overstory species, understory species and soil type at plot center. The tree data
included a bearing distance measurement from plot center, as well as DBH, total
height (measured using a Haglof Vertex Laser), status (live or dead), number of
years dead, and cause of mortality. Saplings were defined as any woody vegetation
32


with DBH < 12 cm and height > 1.4 m. Seedlings had DBH < 12 cm and height <
1.4 m Seedlings and saplings were counted within a 3.6 m radius subplot and
classified according to species and diameter (saplings) or height (seedlings).
Down-woody debris were measured along three transects between 5 and 25m from
plot center. Data on tree and herbaceous cover were also collected along each
transect at 15 m and 25 m from plot center.
4.4 Modeling methods
I initialized the Forest Vegetation Simulator (FVS) model (Stage 1973,
Dixon 2003) Central Rockies Variant, with plot data from the 119 plots to simulate
potential changes in vegetation composition and carbon storage as a result of the
MPB outbreak. FVS is a growth and yield model, created and used by the USDA
Forest Service to help manage forests across the United States. FVS is aspatial and
simulates individual tree growth in stands using multiple forestry growth, diameter
and height algorithms. FVS estimates changes in carbon storage over time with
carbon pools split into aboveground total live, belowground live, belowground
dead, standing dead, downed dead wood, forest floor and herbaceous cover. First,
FVS estimates growth of trees for each cycle by diameter and height. Then, it
estimates mortality based on the individual tree variables like diameter, as well as
on stand variables like basal area or stand density index. Crown ratio (provided as
part of the initialization dataset collected in the field), crown competition factor
and total basal area, and stand density index calculations in FVS help the model
33


project stand trajectories over time. Mortality is primarily projected using stand
density index (SDI) measurements per plot in an effort to model stand densities
realistically where they do not exceed natural stand densities seen in the Central
Rockies, however, background tree mortality was suppressed so as to project the
trajectories of species and carbon immediately following a disturbance and not
introduce new or multiple independent variables. Each growth cycle for the
scenarios lasted 10 years and the output tree list from FVS reads stand and
individual tree level projections for each growth cycle.
FVS was chosen over other models for several reasons. It incorporates
forest growth at the tree and stand level, as compared to other models that operate
over landscape scales, such as LANDIS or BIOME-BGC (White et al. 2000,
Mladenoff 2004). FVS was designed to track carbon storage pools, as well as
forest biological variables, like seedling and sapling growth and tree death
(Crookston and Dixon 2005). It also incorporates many kinds of disturbances, and
has been tested and validated specifically for the Rocky Mountain region in the
Central Rockies variant that I used in this research (Dixon 2003). It focuses on
aboveground carbon storage and biomass, instead of being focused specifically on
belowground processes, such as CENTURY (Parton 1996).
4.5 Model Scenarios
Field measurements collected in 2010 were used to quantify the initial
impacts of the MPB outbreak. Then, future carbon stocks were simulated for 200
34


years. Three simulations were initialized in FVS. For the first simulation, referred
to as the MPB scenario hereafter, I parameterized FVS using field data from all
119 plots exactly as they were collected, where an average 72% of basal area was
classified as dead because of MPB activity. The second simulation, referred to
as the control scenario, was intended to represent near pre-outbreak conditions.
In the control scenario, the input tree list was altered to reclassify all trees that had
recently suffered MPB mortality to live status. For the third simulation, known as
the fire scenario, the control scenario input tree list was burned in 2010 for 72%
of basal area was killed. This matched the average area killed by mountain pine
beetle so as to compare carbon storage and succession trajectories between the
disturbances. The simulated fire maintained default parameters for the study area
in climatic variables. Each simulation was run forward to 2210.
For each of the scenarios, I used the FVS partial regeneration model
using seedlings and saplings present in plots in 2010. Regeneration was modeled
conservatively where future disturbances, factors behind regeneration (such as
wind or seed dispersal), soil processes and decomposition were not taken into
account. It is difficult to model these factors. Regeneration is difficult to predict
because the impacts of light, moisture, climate change, soil changes and overall
disturbance have differing impacts on regeneration. Soil is difficult to model
because it is highly variable across spatial scales, especially at a landscape level.
Also, decomposition rates are also difficult to predict. FVS simulations for both
35


scenarios produced annual data on the growth and attributes of individual trees,
including new seedlings added each cycle, as well as plot level characteristics such
as trees per hectare (TPH) and basal area (BA) per species, and carbon storage and
fluxes among carbon pools. Simulation cycles were set to 10 year growth cycles
and individual trees as well as plot-level summaries were given as output for each
growth cycle.
This methodology to couple field data with the FVS model to track carbon
storage and fluxes was used in Pfeifer et al. (2010), where FVS was initialized for
12 plots in an Idaho lodgepole forest after a MPB outbreak had occurred. They did
not simulate fire, and their plots consisted of slightly different composition, nor
did they track stand composition trajectories.
Collins et al. (2011) modeled species succession trajectories over time in
the Southern Rocky Mountains using advanced regeneration after a MPB
disturbance in lodgepole stands using gridded field data and FVS, but did not
include an assessment for stand carbon.
36


4.6 Analysis Methods
To compare near- and long- term impacts of MPB outbreak on carbon
storage and species composition, I first compared the differences in the vegetation
conditions used to initialize FVS. This comparison also provided the baseline
measurements of carbon storage and species composition to evaluate simulated
potential changes against. Specifically, I compared total stand carbon overall as
well as differences in standing live, downed dead and standing dead carbon pools
between the three scenarios using Welch 2-sided t-tests assuming unequal
variance. To address the research questions investigating how carbon storage
changed over the years following a MPB outbreak, the MPB disturbance trajectory
was compared to the control scenario. Then, to investigate differences in how fire
disturbances compare to MPB disturbances, a simulated fire scenario was
compared to the control scenario as well. The same approach was used to compare
the baseline initial and simulated potential changes in TPH and BA by species for
each scenario over time to address how species composition changed as a result of
the MPB outbreak or fire.
Plots were further categorized for analysis into dominant overstory and
understory combinations, modified from the method used to categorize plot types
in Diskin et al. (2011), where plots were split up by 5 different forest types in
order to track species trajectories based on the advanced regeneration present
37


before a MPB epidemic. For this research, I categorized plots based on the overall
dominant overstory and understory as well, however, there were only three main
types of plots. Of the 119 plots, 103 were dominated by an overstory of lodgepole
pine. The other 16 plots were dominated by subalpine fir or Engelmann spruce.
The plots could further be categorized by the understory level, or advanced
regeneration, of trees, where there was a split between those plots that were
predominately subalpine fir in the understory, and those that were predominately
lodgepole pine in the understory. The majority of plots could be split into three
type categories, locations illustrated in figure 4.2 and examples of each plot type in
figure 4.3. The first type of plot had a stand structure with an overstory dominated
by lodgepole pine, and an understory consisting of primarily subalpine fir; there
were 17 plots in this category. 79 plots could be characterized by the second type,
which had both an overstory and understory dominated by lodgepole pine. The
third type of plot was dominated in the overstory canopy and understory by
subalpine fir; there were 7 plots in this category. The rest of the plots contained
aspen, Engelmann spruce or other species. There didnt seem to be any apparent
pattern to the aspect, slope or elevation when compared to vegetation composition.
This was possibly due to our stratified sampling design. The plots dominated in the
overstory and/or understory by subalpine fir were in very specific soil texture
types, where subalpine fir plots grew in sandy clay loam, clay loam and fine sandy
38


loam. The lodgepole plots were in these soil textures, as well as a broad spectrum
of other soil texture types.
39


Figure 4.2.Field Plots by Dominant Composition Type. Map shows study area
with spatial distribution of plots in each of the three plot species type categories.
40


Plot Type 1:
Lodgepole
Overstory,
Subalpine Fir
Understory
17 Plots
Plot Type 2:
Lodgepole
Overstory,
Lodgepole
Understory
79 Plots
Plot Type 3:
Subalpine Fir
Overstory,
Subalpine Fir
Understory
7 Plots
Figure 4.3 Plot Types. Photos illustrate an example of each category of plot type.
41


V. RESULTS
5.1 2010 Conditions
Of 6257 trees and advanced regeneration surveyed in 119 plots, 86% of
trees per hectare were lodgepole pine representing 85% of the total tree basal area
across plots; 74% of the total number of trees inventoried had MPB mortality
(72% of total tree basal area) and 12% were healthy lodgepole (13% of total tree
basal area). The MPB outbreak in our study area began in 1996, and peak
mortality occurred 2006-2008, where the majority died in 2007, as shown in table
5.1. Lodgepole pines killed by MPB tended to have a larger DBH than live
lodgepole pines (mean DBH was 16.2 cm for live lodgepole and 22.6 cm for dead
lodgepole p-value for 2-sided t-test of means assuming unequal variance was
0.0001). 7% of trees in plots were subalpine fir (6% of total tree basal area) and
6% of trees were healthy subalpine fir (representing 5% of total tree basal area as
well) and 1% was sick or dead subalpine fir (1% of basal area). About 4.6% of
trees were Engelmann spruce (representing 5.8% of total tree basal area), with
4.2% of trees being healthy (4.7% of basal area) and 0.4% sick or dying (1.1%
basal area).
Across all plots, seedlings (measured in trees per hectare (TPH)) were
dominated by lodgepole pine, where TPH was 336 averaged across plots.
Subalpine Fir and Engelmann spruce were the next most common species with an
average 252 and 261 TPH across plots, respectively. Saplings were dominated by
42


lodgepole as well, with 359 TPH on average, and subalpine fir representing an
average 244 TPH and Engelmann spruce averaging 218 TPH. Lodgepole was
more plentiful in the control scenario saplings, representing 370 TPH.
Table 5.1 Surveyed Lodgepole Pine Mortality.Table showing the number of
lodgepole pine surveyed in 2010 in each MPB mortality category, where mortality
ranged from 2004-2009.
MPB Mortality Count of surveyed trees in 2010 field data
1 year ago: full crown of fading needles 53
2 years ago: > 50% orange needles remaining 226
3 years ago: < 50% needles remaining 566
4 years ago: no needles remaining but small and large twigs present 269
5 years ago: only large twigs remaining 37
6+ years ago: both small and large twigs not remaining 14
Unknown 15
Grand Total of Trees 1180
Carbon is calculated in FVS by standard forestry procedures, where
amount of carbon is half the biomass measurement (Avery and Burkhart 2002).
Dry tons/hectare of biomass (where biomass is a measurement of the amount of
living material calculated in FVS by the DBH and height of each tree) multiplied
43


by .5 for all categories except the litter and duff calculations calculates all
aboveground measurements listed for carbon. It is standard to calculate litter and
duff carbon by multiplying the dry biomass in these categories by .37 (Dixon
2003). These carbon estimates were part of the FVS output, and were listed for
each 10-year cycle, beginning in 2010. In 2010, the average amount of total
carbon per plot was 94.7 Mg/ha, but would have been 107 Mg/Ha if there was no
MPB-caused mortality (p-value of .01). The standing live carbon pool averaged at
21.5 Mg/ha, but would have stored 57.2 Mg/ha had the outbreak not occurred,
where pre-outbreak conditions stored 2.67 times more carbon in the standing live
pool (p-value 0.0001). Standing dead pools on average had 28.3 Mg/ha versus
the 1.2 Mg/ha they would have contained without the MPB disturbance (p-value
0.0001). In 2010, the downed dead pool averaged 19.5 Mg/ha of carbon, where
it was not possible to estimate pre-outbreak downed dead wood (p-value of
0.8796).
In August 2011, two plots in the study area actually burned in a fire (plot
22 and plot 6111). Both of these plots were type 3 plots. The plots were
reexamined in September 2011. The fire resulted in every tree in each plot being
transferred to the standing dead or downed dead pool, some even classified as
gone. The plot data taken for these two plots in September 2011 is given in
Appendix A, where the raw plot data serves as an example of the field data
collected, and as information on the fire.
44


5.2 Impacts of MPB on Carbon Storage
After a 200 year simulation, the total stand carbon (which includes the sum
of the standing live, belowground live, belowground dead, standing dead, downed
dead, herbaceous cover and forest floor measurements), was similar with 271
Mg/ha in the control scenario, and 267 Mg/ha for the MPB scenario (p-value for 2-
sided t-test of means assuming unequal variance between the two datasets was
0.4). The slight, but statistically insignificant, differences were because of an
increase in live biomass when MPB mortality was reclassified as live trees in the
control scenario. Carbon between the simulations is listed in figures 5.1-5.4.
Without additional disturbances added to the simulation throughout time and
mortality turned off, total stand carbon (figure 5.1) increased steadily in both the
MPB and control simulations, but was consistently greater in the control scenario
than the MPB scenario. However, by 2040, total stand carbon reached 115 Mg/ha,
which was not significantly different (p-value of. 15) than total carbon stored in
the control scenario in 2010, representing pre-outbreak conditions (107 Mg/ha).
The amount of standing live carbon was greater in the control scenario than
the MPB scenario throughout the 200 year simulations (Figure 5.2). Average
standing live carbon was 136.1 Mg/ha in the MPB scenario, and was 145.7 Mg/ha
in the control scenario by 2210 (with a p-value of 0.0009). Standing dead carbon
was higher in the MPB scenario, but not significantly when compared to the
45


control simulation in 2210 (Figure 5.3) with 4.3 Mg/ha and 3.7 Mg/ha respectively
(p-value of 0.13). However, there was a rapid decline in standing dead biomass
during the first 30 years of the MPB simulation as dead trees fell and were
transferred to the downed dead carbon pool. Consequently, the downed dead
carbon pool increased dramatically during the first 30 years of the MPB scenario
(Figure 5.4). For the remaining years of the simulation, there were negligible
differences in standing dead carbon between the two scenarios (p-value of 0.15).
Carbon in the downed dead pool increased steadily over time for both simulations,
but was generally greater in the MPB scenario, with 58 Mg/ha in the control
scenario and 63 Mg/ha in the MPB scenario. However, after 120 years of
simulation the differences between the two scenarios were less pronounced.
5.3 Impacts of Simulated Fire Disturbance on Carbon Storage
In order to compare the impacts of MPB disturbances to fire disturbance on
carbon and species composition trajectories, a fire was simulated to burn 72% of
basal area (same as the average basal area affected by MPB disturbance in the
plots) on the control scenario input table, to be referred to as the fire scenario.
Carbon storage and stand structure trajectories were compared between the fire
scenario and the control pre-disturbance scenario, and subsequently to the MPB
disturbance scenario. The fire scenario showed the lowest amount of carbon stored
in all four pools of carbon, which is also shown in figure 5.1-5.4. In 2010, after the
fire was simulated (in late summer for default climatic conditions for the central
46


Rockies variant), the standing live carbon pool stored 79.4 Mg/ha (p-value of
0.0001 between the fire scenario and the control scenario). The standing dead pool
stored 0.52 Mg/ha (p-value of 0.0001), downed dead stored 19.91 Mg/ ha (p-
value of 0.0001) and the total stand carbon pool stored 142.6 Mg/ha (p-value of
0.0001).
In 2010, the simulated fire reduced standing live carbon (figure 5.2) to
about 19 Mg/ha from 57.2 Mg/ha in the control (p-value of 0.0001). Standing
live carbon returned to pre-fire levels also by about 2070, to about 57 Mg/ha.
Standing dead carbon (figure 5.3), however, is projected to increase from 1.2
Mg/ha in the control to about 33 Mg/ha in the fire scenario (p-value of 0.0001).
The standing dead carbon pool in the fire scenario stored almost 3 Mg/ha, and the
downed dead carbon pool stored 42 Mg/ha (p-value of 0.0001). It was not
possible to project pre-fire downed dead carbon for 2010 in the fire scenario
because the control scenario tree input data was burned and it was not possible to
validly predict pre-disturbance downed dead wood. After the fire, however, 10.9
Mg/ha remained in the downed dead carbon pool in 2010. The average amount of
carbon released from the fire from all carbon pools was 35.7 Mg/ ha in 2010.
The simulated fire results (figure 5.1) for 2210 showed total stand carbon
to be 228.43 Mg/Ha (p-value of 0.0001), which was slightly lower than the
total stand carbon for either the MPB or the control scenarios. Total stand carbon
rebounded from around 81 Mg/ha to about 107 Mg/ha (pre-disturbance total stand
47


carbon) by 2070 (p-value of 0.0001) as well, representing a return to pre-fire
conditions within 60 years following fire.
48


Total Stand Carbon
O
O -
CO
O
in -
CN
O
in
o -
2050
Legend
* MPB Scenario
o Control Scenario
x Fire Scenario
2100 2150 2200
Year
Year Carbon pool Control scenario (Mg/ha) MPB scenario (Mg/ha) Fire scenario (Mg/ha)
2010 Total 107 6.8 97 6.4 81 5.3
2210 Total 271 6.7 267 10.2 228 11.2
Figure 5.1 Total Stand Carbon. Scatter plot shows the trajectory of total stand
carbon with confidence intervals, which is a compilation of all carbon pools, for
each of the three scenarios: MPB, Control and Fire, where the table lists starting
2010 carbon storage and ending 2210 storage for each trajectory.
49


Standing Live Carbon
L_
ro
o
CD
x
1/i
o
_Q
TO
o
2050
2100
2150
2200
Year
Year Carbon pool Control scenario (Mg/ha) MPB scenario (Mg/ha) Fire scenario (Mg/ha)
2010 Stand. Live 57.2 5.0 21.5 4.5 19.0 2.2
2210 Stand. Live 145.7 3.7 136 4.2 127.4 5.0
Figure 5.2 Standing Live Carbon. Scatter plot shows the trajectory of carbon
the standing live carbon pool with confidence intervals for each of the three
scenarios: MPB, Control and Fire, where the table lists starting 2010 carbon
storage and ending 2210 storage for each trajectory.


Standing Dead Carbon
CD
3
o
<1)
X
To
c
o
H
o
.Q
co
o
2050
2100
Year
2150
2200
Year Carbon pool Control scenario (Mg/ha) MPB scenario (Mg/ha) Fire scenario (Mg/ha)
2010 Stand. dead 1.2 0.6 28.3 3.0 33.0 2.7
2210 Stand. Dead 3.7 0.5 4.3 0.5 2.9 0.4
Figure 5.3 Standing Dead Carbon. Scatter plot shows the trajectory of carbon
the standing dead carbon pool with confidence intervals for each of the three
scenarios: MPB, Control and Fire, where the table lists starting 2010 carbon
storage and ending 2210 storage for each trajectory.


Downed Dead Carbon
O
00
O
CD
O
0)
X
1/5
c
o
o
JO
U_
03
o
o
o
CM
Him
.. O
-- O
1 J | I 1 1
i 1
o -
2050
I----
2100
Year
Legend
MPB Scenario
o Control Scenario
x Fire Scenario
2150
---T
2200
Year Carbon pool Control scenario (Mg/ha) MPB scenario (Mg/ha) Fire scenario (Mg/ha)
2010 Down dead 19.5 2.3 19.8 2.3 4.62 0.6
2210 Down dead 58.0 4.1 63.0 5.6 42.54 4.8
Figure 5.4 Downed Dead Carbon. Scatter plot shows the trajectory of carbon
the downed dead carbon pool with confidence intervals for each of the three
scenarios: MPB, Control and Fire, where the table lists starting 2010 carbon
storage and ending 2210 storage for each trajectory.


5.4 Impacts of MPB on Stand Structure
Four primary species were present in stands in 2010: lodgepole pine,
Engelmann spruce, subalpine fir and trembling aspen. There were a few other
species, such as white fir {Abies concolof) and Douglas fir (Pseudotsuga
menziesii), present in limited quantities in plots, which were grouped and
referenced as other in figures.
In 2210, as shown in figure 5.5a, the control scenario showed a dominance
of lodgepole pine in the overstory canopy, where lodgepole accounted for 58% of
basal area in 2210, but subalpine fir only accounted for 19% of basal area.
Engelmann spruce and aspen were subdominant species in both scenarios;
however, the MPB scenario had greater proportion of aspen and spruce than in the
control scenario in 2210. The MPB scenario (figure 5.5b) shows a much more
even mixture of tree species in the overstory vegetation layer, where there was
about 30% of basal area of each subalpine fir and lodgepole, 12.6% was
Engelmann spruce, and 17.7% was aspen. For an example plot in the study area,
figure 5.6 shows the FVS 2010 and 2210 results for the MPB and control scenario,
where there has been a shift in the species composition trajectory following MPB
mortality.
53


(a) Control Scenario: BA percent by species
ro
.c
CO
CD
o
*
CO
JZ
CD
<*-
o
Year
(b) MPB Scenario: BA percent by species
mt mi Tut m lm 211* 21 si 21M 2121 21M 221
Other
Aspen
Lodpepole
Engelmann Spruce
Subilprne Fir
Figure 5.5 (a-c) Average Species Composition by Percent of Basal Area.
Percent of basal area (in m2/Ha) species trajectories for each simulation averaged
across all plots where a.) is the percent basal area for each species in the MPB
scenario, b.)is the Control scenario and c.) is the Fire scenario
54


Control Scenario
2010
2210
Mountain Pine Beetle Scenario
2010 2210
Figure 5.6 Beginning and End Species Trajectory. Output images from FVS
showing the differences in stand composition trajectories over time for each
scenario for field plots initially dominated by a lodgepole overstory and an
understory of subalpine fir the MPB and Control model scenarios.
55


5.5 Impacts of Simulated Fire Disturbance on Stand Structure
In 2010, the fire scenario was primarily lodgepole pine in the overstory,
with an average 87.9% of the remaining basal area represented by lodgepole after
the simulated fire. The simulated fire stand structure looked somewhat different
from the MPB or control scenario simulations in 2210 overall averaged across
plots. As shown in figure 5.5c., the live overstory was dominated by 71.6%
lodgepole pine, 13% of the basal area consisted of subalpine fir, 5.1% was aspen
and 8.4% was Engelmann spruce, leaving 1.5% of basal area as other species. The
fire scenario had the most percentage of basal area represented by lodgepole out of
the three scenarios in 2210.
In 2010, a real fire burned through two plots. Every tree was killed, and no
regeneration was present during the 2011 resurvey. There was some herbaceous
cover present along the fuels transects. Much of the standing live carbon was
transferred to the standing or downed dead carbon pools, which changes the stand
structure. This transfer of carbon to the ground level pools may have impacts on
soil and regeneration processes in the near and long term. The fire I simulated to
consume 72% of basal area was a conservative estimate, as the real fire consumed
almost all of the basal area in the two plots.
5.6 Variation in Plots for Carbon and Stand Structure Trajectories
The 119 field plots were split up between primary overstory and understory
species composition, as described in the methods section, to investigate variation
56


in carbon and species composition between plot types as opposed to averages of
carbon and stand composition trajectories across all plots. The plots were split up
based on the species composition observed in plots in 2010, and thus included
MPB mortality. Figures 5.7-5.10 show an example of the differences in the four
carbon pool trajectories for the three plot types for the MPB scenario. These stand
types were investigated for the MPB scenario to compare the individual plot
species compositions and carbon trajectories between scenarios by primary
overstory and understory species composition. Only the MPB scenario was split up
into plot types because the MPB scenario represented reality, where stand
composition types were qualified in the field after MPB disturbance, and thus
would not reflect valid species composition necessarily for the other two simulated
scenarios.
5.6.1 Variation in Carbon Trajectories between Plot Types
There were differences in carbon storage in some of the plots within the
MPB scenario due to changes in stand structure, which is shown in figure 5.7-5.11.
Of the three main types of plots, the third type of plot, which consisted of an
overstory and understory dominated by subalpine fir, stored more carbon than the
average across all plots in the standing live carbon pool (where the calculated
confidence interval for type 3 plots did not intersect the other values for 2010,
validating a statistically significant difference). The average for these 7 plots in
2210 was 136.9 Mg/ha standing live carbon. The first type of plot, with a
57


lodgepole-dominated overstory and a subalpine fir-dominated understory, the
standing live pool stored less than the average stored across all plots at 130.5
Mg/ha. The second type of plot, where lodgepole dominated both the overstory
canopy and the understory, there was a majority of 79 plots in this category, was
about the same as the average across all plots.
58


MPB Scenario: Standing Live Carbon
Figure 5.7 Standing Live Carbon by Plot Type. Shows variability in carbon
stored within the standing live carbon pool between the different plot types in the
MPB scenario.
59


MPB Scenario: Standing Dead Carbon
Figure 5.8 Standing Dead Carbon by Plot Type Variability within the standing
dead carbon pool between the different plot types in the MPB scenario.
60


Carbon (Tons/Hectare)
MPB Scenario: Downed Dead Carbon
Figure 5.9 Downed Dead Carbon by Plot Type. Variability in carbon stored
within the downed dead carbon pool between the different plot types in the MPB
scenario.
61


Carbon (Tons/Hectare)
MPB Scenario: Total Stand Carbon
Figure 5.10 Total Stand Carbon by Plot Type. Variability in total stand carbon
storage between the different plot types in the MPB scenario.
62


5.6.2 Variation in Stand Composition Trajectories between Plot Types
For the MPB scenario (as shown in figure 5.11 a-c), when the plots are
split into the three species type categories, the stand trajectories for species varied
somewhat. Type 1 plots initially were dominated in the overstory vegetation by
lodgepole (45%) and subalpine fir (31.4%), and in 2210, are dominated by 62.3%
of the basal area by subalpine fir. Engelmann spruce occupy 19.5% of the basal
area in type 1 plots in 2210, and 11.8% is lodgepole on average. Type 2 plots were
initially dominated by lodgepole in 2010 accounting for 65% of basal area, which
was reduced to 43.8% by 2210. This was offset by an increase in subalpine fir,
which began at 6.3% in 2010 and increased to representing 19.2% in 2210, as well
as aspen, which increased from less than 1% in 2010 to almost 23% in 2210. Type
3 plots were initially dominated by subalpine fir, which only increased from 70%
to 79% of basal area between 2010 and 2210. These plots experienced a decrease
in lodgepole from 11.9% to 8.5% basal area from 2010 to 2210. Engelmann spruce
was reduced over time from 2010 to 2210 by representing 18.2% to 11.8% of basal
area.
63


(a) MP8 Scenario: Percent of Overstory Beta I Area Type 1 Plots
2010 2030 20*0 2070 2000 2110 2130 2160 2170 2190 2210
> e*
(b) MPB Scenario:Perceni of Overstory Basal Area Type 2 Plots
2010 2030 20*0 2070 2090 2110 2130 21*0 2170 2190 2210
Ytsr
Other
A
Lodpep 4e
Ervgelmann spruce
Subdipine Fir
(c) MPB $cenario:Percent of Overstory Basal Area Type 3 Plots
2010 2030 20*0 2070 20K 2H0 2130 21*0 2170 219C 2210
>etr
Figure 5.11 (a.-c) Species Trajectories between Plot Types. Species trajectory
projections by type 1, 2 or 3 species plots over time for the MPB scenario.
64


VI. DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS
This research stands to track and compare carbon and species succession
trajectories without disturbance, after a real MPB disturbance, and after a
simulated fire, in order to better quantify the impacts of large disturbances on
carbon and species, in order to provide better information to forest managers. I
used forest inventory data collected in field plots to initialize the FVS model to
quantify initial and future carbon and species composition trajectory projections.
This data provided information on how differing disturbances, which may increase
with global climate change, may change carbon and species trajectories and
provides insight into how to manage forests better. The MPB outbreak in the
Southern Rocky Mountains has altered the trajectory of carbon storage and species
composition in Southern Rocky Mountain forests. The surveyed field plots were
dominated by lodgepole, where most of the larger-diameter lodgepole were killed
by MPB (unlike fire disturbances, where size is not typically selected). As
disturbances are predicted to increase in frequency and severity in a warming
climate, the intricate relationship between carbon storage and species composition,
and how trajectories change due to disturbance, will become ever-important. These
results showed carbon storage and species composition are altered after insect
disturbances, and are different from the impacts on stand structure and carbon
storage from fire disturbances. Further, carbon storage may differ between species
types present in stands.
65


6.1 Impacts of MPB on Carbon Storage
Stand carbon storage trajectories were altered short-term as a result of the
MPB disturbance. The immediate impacts of MPB outbreaks on carbon storage are
substantial, causing a rapid transfer of 27.1 Mg/Ha, or 28.6% of the total stand
carbon from the standing live pool to the standing dead pool and subsequently to
the downed dead pool. However, these pools recover at varying rates following
MPB outbreaks. Klutsch et al. (2009) found a 69% reduction in overall basal area,
and transfers from the standing live carbon pool to the standing dead and downed
dead/litter pools for northern Colorado lodgepole forests experiencing MPB
epidemic conditions. This differed from the field plots collected for my research,
where an average 72% of basal area was lost. I found that the impacts on the
standing live pool were relatively short-lived and that standing live carbon
recovered to pre-outbreak conditions by 2060 (roughly 60 Mg/ha). The total stand
stored carbon returns to pre-outbreak conditions by 2040. After about 90 years,
differences in carbon storage in all the carbon pools were negligible between the
two scenarios.
In contrast to this study, Pfeifer et al. (2010) examined stand carbon
changes in a northern Idaho forest as a result to a beetle outbreak. They observed a
rebound in carbon to pre-outbreak conditions in 25 years. Mortality of basal area
was not as high for their region and the forest had slightly different stand
characteristics, such as the presence of whitebark pine (Pinus albicaulis) and more
66


Douglas fir (Pseudotsuga menziesii var. menziesii). Pfeifer et al. validates the
notion that forest recovery of carbon storage to pre-outbreak conditions is not only
possible, but occurs over a relatively short time span.
The trajectories of the field plots by the dominant species present in the
overstory and understory varied somewhat in carbon content and species
composition. This study only split plots based on their dominant overstory and
understory vegetation, and not any other factors, such as stand densities or
diameter distributions of trees. As modeled for actual conditions in the MPB
scenario, the 7 plots dominated by subalpine fir in both the overstory and
understory, or the type 3 plots, stored above the average amount of carbon stored
in all plots. This may indicate that subalpine fir has a greater carbon storage
potential in this ecosystem. Initially, type 1 plots had higher standing dead carbon
storage, indicating potentially higher percent basal area impacted by MPB. Type 3
plots had the lowest storage in the standing dead carbon pool, which is intuitive in
that they primarily consist of non-host species for MPB. Type 3 plots stored more
carbon on average in the downed dead pool than the other plots.
6.2 Impacts of Simulated Fire Disturbance on Carbon Storage
Both MPB and fire disturbances altered the trajectories of carbon storage,
however a fire that burns the same area MPB affects seems to impact carbon pools
more severely and for a longer time period. For all 200 years of the simulation, the
fire scenario stored less carbon in the standing live, standing dead and total stand
67


carbon pools compared to the other scenarios. This could have been due to the fire
consuming surface fuels in the downed-dead wood biomass pools and smaller
seedlings and saplings that remained in place during the MPB simulation. The fire
scenario showed higher carbon storage in the standing dead pool than the other
scenarios. This shows how the fire impacts stands differently, where doesnt
impacts selective species or size classes. Downed dead carbon was significantly
lower in the fire scenario by 2210. Initially, downed dead carbon was projected in
2020 to be similar to that stored in the downed dead pool in the MPB scenario. By
2030, however, downed dead carbon stored was lower in the fire scenario than in
the MPB scenario. It seems that more carbon, likely from the standing dead pool,
is transferred to the downed dead pool over a longer time period in MPB
disturbances and more of the carbon from the standing dead pool is transferred
over time to the other pools in fire disturbances over time.
The total carbon storage recovered to pre-fire levels more slowly than the
MPB scenario recovered to pre-MPB storage, where carbon returned to pre-fire
carbon storage in the standing live pool and total stand carbon by 2080 which was
twenty years later than recovery of the MPB disturbance trajectory.
6.3 Impacts of MPB on Stand Structure
MPB disturbances impact stand structure differently than other major
disturbances, such as fire or wind, because only select trees, specifically older
lodgepole with a greater DBH, are taken out of the live biomass pool. Insect
68


disturbances allow seed dispersal, light and water to interact differently within the
forest compared to an even-aged, undisturbed forest. This allows for differences in
stand structure and species composition between disturbed and undisturbed forests
of the same type. Changes in stand composition also can subsequently change the
rate at which carbon is sequestered and stored, as different species sequester
carbon at different rates and ages of development. Initially, an average 72% of
basal area was killed by mountain pine beetle in this study area. This corresponds
to Klutsch et al. (2009) where the live basal area declined 71% in their beetle
infested study plots in the Colorado Arapaho National forest.
Succession was modeled conservatively by projecting the current
trajectories with only advanced regeneration, out 200 years, but there was still a
shift in species composition over time in the MPB scenario as compared to the
control scenario. If the MPB outbreak had not occurred, species composition
would be dominated by primarily lodgepole pine. The MPB scenario however,
shows a species composition shift towards a greater percentage of the basal area
being subalpine fir. A mixed vegetation composition in the future could be
beneficial for ecological sustainability as it encourages biodiversity, which will
make stands more resilient.
This observation is supported through the study carried out by Collins et al.
(2011) that showed a shift towards a subalpine-fir dominated canopy in Colorado
in untreated stands affected by MPB through the use of vegetation modeling.
69


Collins et al (2011) found that tree regeneration in MPB impacted lodgepole
stands were more evenly distributed with subalpine fir and lodgepole pine as well.
This validates a landscape level increase in basal area represented by subalpine fir
after a MPB outbreak, for a more mixed stand composition. Klutsch et al. (2009)
also concluded that stands impacted by the mountain pine beetle epidemic will
result in an uneven aged stand structure consisting of older, existing lodgepole and
younger subalpine fir and Engelmann spruce. Their work along with this study
suggest that future stand composition depends heavily on advance regeneration
present at the time of disturbance.
6.4 Impacts of Simulated Fire Disturbance on Stand Structure
Based on the FVS model results, fire seems to encourage a stand structure
trajectory that becomes dominated by primarily lodgepole pine, which is different
from the trajectories of species composition in the absence of disturbance and after
MPB disturbance. These findings correspond with known successional trends
where pure lodgepole stands typically regenerate after stand-replacing fires in
forests that were dominated by lodgepole (Lotan et al. 1985). Even in places where
forests are dominated by spruce and fir, lodgepole can increase following fires. In
the absence of fire though, subalpine fir and Englemann spruce increase in
dominance where there have been fewer fires (Romme and Knight 1981).
In the fire scenario, the dominant species in plots makes a difference in the
stand composition trajectories over time.
70


6.5 Conclusions and Recommendations
6.5.1 Conclusions
Changes to carbon storage in forests affected by MPB were relatively
short-lived under scenarios that did not incorporate further disturbances. Standing
live carbon rebounded within 50 years in most stands. However, species
composition changed over a longer time period, and there were substantial
differences in forest stand structure and species composition that persisted after
200 years of simulation. MPB disturbances impact stand structure differently than
other major disturbances, such as fire. MPB targets older, large diameter
lodgepole pines, leaving behind advanced regeneration that grow rapidly in the
years after disturbance. This resulted in standing-live carbon stocks recovering
quickly after MPB. This was accompanied by changes in species composition
trajectories, with a slight shift toward more aspen, fir, and spruce. Succession
modeling was conservatively, however there was still a shift in species
composition over time in the MPB simulation, compared to the control simulation.
The findings of this study corroborate those found by Collins et al. (2011), who
showed a shift towards a subalpine-fir dominated canopy in Colorado in untreated
stands affected by MPB through the use of vegetation modeling. Lodgepole pine
forests affected by MPB appear to enter a self-correcting mode, where a more
diverse forest develops after MPB disturbance, making it less susceptible to future
MPB disturbances. Carbon storage differs within each scenario as well, based on
71


dominant overstory and understory vegetation, where plots dominated in the over
and understory by subalpine fir trajectories store more carbon than other plots.
In lodgepole pine forests, fire affects the larger trees favored by beetles, but
also causes mortality in advanced regeneration. Carbon loss during a fire is a
function of fuel loads and fire severity and recovery of carbon after a fire is a
function of seed availability and regeneration rates. The results from this study
and those found in the literature, suggest that burned lodgepole pine forests will
regenerate as lodgepole, and the rate of carbon recovery is somewhat slower after
fire than after the MPB outbreak. Fires perpetuating lodgepole pine in stands could
make the ecosystem more vulnerable to disturbances in the future.
These changes in both carbon and species composition may have additional
effects on carbon cycling by altering primary productivity, soil biogeochemistry,
and regeneration over time; processes that warrant further investigation.
6.5.2 Recommendations and Management Implications
This research suggests that disturbances change forest composition
and carbon trajectories over time, and that different disturbances alter these
trajectories differently. Also, this research concluded that there are differences in
the trajectories between individual stands within forest types that are impacted by
disturbance. The local scale impacts of MPB outbreaks are substantial and could
have implications for how forests offset greenhouse gas emissions and global
climate change at broader scales. It would be beneficial for stakeholders managing
72


for carbon storage and sequestration in forests to account for the varying carbon
impacts and the length of recovery time in carbon pools resulting from the various
disturbances that affect forests. This is especially important as disturbances are
projected to increase in a warming climate. Managers should also account for the
differences that may occur in between stands across a landscape scale, and that a
uniform management design may not be optimal across an entire landscape, as
there are variances between stands and species composition types.
73


REFERENCES
Amman, G. D. 1977. The role of mountain pine beetle in lodgepole ecosystems:
Impact on succession Pages 3-18 in W. J. Mattson, editor. The Role of
Arthropods in Forest Ecosystem. Springer-Verlag, New York.
Aoki, C. F., W. H. Romme, and M. E. Rocca. 2011. Lodgepole Pine Seed
Germination Following Tree Death from Mountain Pine Beetle Attack in
Colorado, USA. The American Midland Naturalist 165:446-451.
Avery, T. E. and H. E. Burkhart. 2002. Forest Measurements. 5 edition. McGraw
Hill, NY, NY.
Baker, W. L. and T. T. Veblen. 1990. Spruce Beetles and Fires in the Nineteenth-
Century Subalpine Forests of Western Colorado, U.S.A. Arctic and Alpine
Research 22:65-80.
Barrow, E. and G. Yu. 2005. Climate Scenarios for Alberta. Prairie Adaptation
Research Collaborative (PARC).
Bazzaz, F. A. 1996. Plants in Changing Environments. Cambridge University
Press, U K.
Bentz, B. J., J. Regniere, C. J. Fettig, E. M. Hansen, J. L. Hayes, J. A. Hicke, R. G.
Kelsey, J. F. Negron, and S. J. Seybold. 2010. Climate Change and Bark
74


Beetles of the Western United States and Canada: Direct and Indirect
Effects. BioScience 60:602-613.
Bessie, W. C. and E. A. Johnson. 1995. The relative importance of fuels and
weather on fire behavior in subalpine forests. Ecology 76:747-762.
Birdsey, R., K. Pregitzer, and A. Lucier. 2006. Forest Carbon Management in the
United States. Journal of Environment Quality 35:1461.
Birdsey, R. A. 1992. Carbon storage and accumulation in United States forest
ecosystems. Gen. Tech. Rep. WO-59, Washington D.C.: U.S. Department
of Agriculture, Forest Service, Washington Office.
Bradford, J. B., R. A. Birdsey, L. A. Joyce, and M. G. Ryan. 2008. Tree age,
disturbance history, and carbon stocks and fluxes in subalpine Rocky
Mountain forests. Global Change Biology 14:2882-2897.
Breshears, D. and C. D. Allen. 2002. The importance of rapid, disturbance-induced
losses in carbon management and sequestration. Global Ecology &
Biogeography 11:1-5.
Brown, J., editor. 1975. Fire cycles and community dynamics in lodgepole pine
forests.
Carroll, A. L., S. W. Taylor, J. Regniere, and L. Safranyik. 2003. Effect of climate
change on range expansion by the mountain pine beetle in British
Columbia. Page Paper 195.
75


CCSP. 2007. The First State of the Carbon Cycle Report (SOCCR): The North
American Carbon Budget and Implications for the Global Carbon Cycle.
National Oceanic and Atmospheric Administration, National Climatic Data
Center, Asheville, NC.
Chapin III, F. S., P. A. Matson, and H. Mooney. 2002. Principles of terrestrial
ecosystem ecology. Springer-Verlag, New York.
Claveau, Y., C. Messier, P. G. Comeau, and K. D. Coates. 2002. Growth and
crown morphological responses of boreal conifer seedlings and saplings
with contrasting shade tolerance to a gradient of light and height. Canadian
Journal of Forest Research 32:458-468.
Collins, B. J., C. C. Rhoades, R. M. Hubbard, and M. A. Battaglia. 2011. Tree
regeneration and future stand development after bark beetle infestation and
harvesting in Colorado lodgepole pine stands. Forest Ecology and
Management 261:2168-2175.
Comeau, P. and J. P. Kimmins. 1989. Above- and below-ground biomass and
production of lodgepole pine on sites with differing soil moisture regimes.
Canadian Journal of Forest Research 19:447-454.
Crookston, N. and G. Dixon. 2005. The forest vegetation simulator: A review of
its structure, content, and applications. Computers and Electronics in
Agriculture 49:60-80.
76


Dale, V. H., L. A. Joyce, S. McNulty, R. P. Neilson, M. P. Ayres, M. D.
Flannigan, P. J. Hanson, L. C. Irland, A. E. Lugo, C. J. Peterson, D.
Simberloff, F. J. Swanson, B. J. Stocks, and B. Michael Wotton. 2001.
Climate Change and Forest Disturbances. BioScience 51:723-734.
Diskin, M., M. E. Rocca, K. N. Nelson, C. F. Aoki, and W. H. Romme. 2011.
Forest developmental trajectories in mountain pine beetle disturbed forests
of Rocky Mountain National Park, Colorado. Canadian Journal of Forest
Research 41:782-792.
Dixon, G. E. 2003. Essential FVS: a users guide to the Forest Vegetation
Simulator., USDA-Forest Service, Forest Management Service Center, Fort
Collins, CO.
Fahey, T. J. and D. H. Knight. 1986. Lodgepole Pine Ecosystems. BioScience
36:610-617.
Flannigan, M. D., B. J. Stocks, and B. M. Wotton. 2000. Climate change and
forest fires. The Science of the Total Environment 262:221-229.
Franklin, J. F., T. A. Spies, R. V. Pelt, A. B. Carey, D. A. Thornburgh, D. R. Berg,
D. B. Lindenmayer, M. E. Harmon, W. S. Keeton, D. C. Shaw, K. Bible,
and J. Chen. 2002. Disturbances and structural development of natural
forest ecosystems with silvicultural implications, using Douglas-fir forests
as an example. Forest Ecology and Management 155:399-423.
77


Friedlingstein, P., L. Bopp, P. Ciais, J. Dufresne, Louis, L. Fairhead, H. LeTreut,
P. Monfray, and J. Orr. 2001. Positive feedback between future climate
change and the carbon cycle. Geophys. Res. Lett. 28:1543-1546.
Glenn-Lewin, D. C., R. K. Peet, and T. T. Veblen, editors. 1992. Plant Succession:
Theory and Prediction. Chapman and Hall, London, UK.
Goward, S. N., J. G. Masek, W. Cohen, G. Moisen, G. J. Collatz, S. Healey, R. A.
Houghton, C. Huang, R. Kennedy, B. Law, S. Powell, D. Turner, and M.
A. Wulder. 2008. Forest Disturbance and North American Carbon Flux.
EOS Transactions American Geophysical Union 89:105.
Hamann, A. and T. Wang. 2006. Potential effects of climate change on ecosystem
and tree species distribution in British Columbia Ecology 87:2773-2786.
Heal, G. 2000. Valuing Ecosystem Services. Ecosystems 3:24-30.
Heath, L. S. and J. E. Smith. 2004. 2003 Report on Sustainable Forests:
Contribution of forest ecosystems fo the total global carbon budget,
including absorption and release of carbon USDA Forest Service.
Hurteau, M. and M. North. 2008. Fuel treatment effects on tree-based forest
carbon storage and emissions under modeled wildfire scenarios. Frontiers
in Ecology and the Environment 7:409-414.
Hurteau, M. and M. North. 2009. Fuel treatment effects on tree-based forest
carbon storage and emissions under modeled wildfire scenarios. Frontiers
in Ecology and the Environment 7:409-414.
78


Hurtt, G. C., S. W. Pacala, P. R. Moorcroft, J. Caspersen, E. Shevliakova, R. A.
Houghton, and B. Moore. 2002. Projecting the future of the U.S. carbon
sink. Proceedings of the National Academy of Sciences 99:1389-1394.
IPCC. 2007. Intergovernmental Panel on Climate Change Assessment Report.
Cambridge University Press, Cambridge, United Kingdom and New York,
NY, USA.
Joyce, L. A. and R. Birdsey. 2000. The impact of climate change on America's
forests: a technical document supporting the 2000 USDA Forest Service
RPA Assessment. U.S. Department of Agriculture, Forest Service, Rocky
Mountain Research Station, Fort Collins, CO:.
Kashian, D. M., D. B. Tinker, M. G. Turner, and F. L. Scarpace. 2004. Spatial
heterogeneity of lodgepole pine sapling densities following the 1988 fires
in Yellowstone National Park, Wyoming, U.S.A. Canadian Journal of
Forest Research:2263-2276.
Kashian, D. M., W.H. Romme, D.B. Tinker, M.G. Turner, and M.G. Ryan. 2006.
Carbon storage on landscapes with stand-replacing fires. BioScience 56:
598-606.
Kimball, B. A., J. R. Mauney, F. S. Nakayama, and S. B. Idso. 1993. Effects of
increasing atmospheric CO2 on vegetation. Plant Ecology 104-105:65-75.
Klutsch, J. G., J. F. Negron, S. L. Costello, C. C. Rhoades, D. R. West, J. Popp,
and R. Caissie. 2009. Stand characteristics and downed woody debris
79


accumulations associated with a mountain pine beetle (Dendroctonus
ponderosae Hopkins) outbreak in Colorado. Forest Ecology and
Management 258:641-649.
Kueppers, L. and J. Harte. 2005. Subalpine Forest Carbon Cycling: Short- and
Long-Term Influence of Climate and Species. Ecological Applications
15:1984-1999.
Kurz, W. A., C. C. Dymond, G. Stinson, G. J. Rampley, E. T. Neilson, A. L.
Carroll, T. Ebata, and L. Safranyik. 2008. Mountain pine beetle and forest
carbon feedback to climate change. Nature 452:987-990.
Lai, R. 2004. Soil Carbon Sequestration Impacts on Global Climate Change and
Food Security. Science: 1623-1627.
Lai, R., J. M. Kimble, R. F. Follett, and B. A. Stewart, editors. 2000. Assessment
Methods for Soil Carbon. CRC/Lewis Publishers, Boca Raton, FL.
Lamlom, S. H. and R. A. Savidge. 2003. A reassessment of carbon content in
wood: variation within and between 41 North American species. Biomass
and Bioenergy 25:381-388.
Lotan, J. 1976. Cone Serotiny Fire Relationships in Lodgepole Pine./// Tall
Timbers Fire Ecology Conference Proceedings, Tall Timbers Research
Center, Tallahassee, FL.
Lotan, J., J. Brown, and L. Neuenschwander., editors. 1985. Role of fire in
lodgepole pine forests, Washington State University, Pullman.
80


Lotan, J. E. and W. B. Critchfield. 2004. Lodgepole Pine. USFS.
Lotan, J. E. and D. A. Perry. 1983. Ecology and regeneration of lodgepole pine. ,
Washington D C. .
Lutes, D. C., R. E. Keane, J. F. Caratti, C. H. Key, N. C. Benson, S. Sutherland,
and L. J. Gangi. 2006. FIREMON: Fire Effects Monitoring and Inventory
System. Technical RMRS-GTR-164-CD.
Marschner, P. and Z. Rengel, editors. 2007. Nutrient Cycling in Terrestrial
Ecosystems. Springer-Verlag, Berlin, Heidelberg.
Mladenoff, D. 2004. LANDIS and forest landscape models. Ecological Modelling
180:7-19.
Monserud, R. A. Y., Yuqing; Huang, Shongming; Tchebakova, Nadja. 2008.
Potential change in lodgepole pine site index and distribution under
climatic change in Alberta Canadian Journal of Forest Research 38:343-
352.
Monson, R. K., A. A. Tumipseed, J. P. Sparks, P. C. Harley, L. E. Scott-Denton,
K. Sparks, and T. E. Huxman. 2002. Carbon sequestration in a high-
elevation, subalpine forest. Global Change Biology 8:459-478.
Newell, R. G. and R. N. Stavins. 2000. Climate Change and Forest Sinks: Factors
Affecting the Costs of Carbon Sequestration. Journal of Environmental
Economics and Management 40:211-235.
81


Pacala, S. W., R. A. Birdsey, S. D. Bridgham, R. T. Conant, K. Davis, B. Hales, R.
A. Houghton, J. C. Jenkins, M. Johnston, G.Marland, and K. Paustian.
2007. The North American Carbon Budget Past and Present. In: The First
State of the Carbon Cycle Report (SOCCR): The North American Carbon
Budget and Implications for the Global Carbon Cycle. U.S. Climate
Change Science Program and the Subcommittee on Global Change
Research. National Oceanic and Atmospheric Administration, National
Climatic Data Center, Asheville, NC, USA.
Parker, A. J. and K. C. Parker. 1983. Comparative successional roles of trembling
aspen and lodgepole pine in the Southern Rocky Mountains.
Parton, W. J. 1996. The CENTURY model. Pages 283-293 in D. S. Powlson, P.
Smith, and J. U. Smith, editors. Evaluation of Soil Organic Matter Models
Using Existing Long-Term Datasets. Springer-Verlag, Heidelberg
Pfeifer, E. M., J. A. Hicke, and A. J. H. Meddens. 2011. Observations and
modeling of aboveground tree carbon stocks and fluxes following a bark
beetle outbreak in the western United States. Global Change Biology
17:339-350.
Prentice, I. C., G.D. Farquhar, M.J.R. Fasham, M.L. Goulden, M. Heimann, V.J.
Jaramillo, H.S. Kheshgi, C. Le Quere,, D. W. R. W. R.J. Scholes, D.
Archer, M.R. Ashmore, O. Aumont, D. Baker, M. Battle, M. Bender, L.P.
Bopp, P. Bousquet, K. Caldeira,, P. M. C. P. Ciais, W. Cramer, F.
82


Dentener, I.G. Enting, C.B. Field, P. Friedlingstein, E.A. Holland,, J. I. H.
R.A. Houghton, A. Ishida, A.K. Jain, FA. Janssens, F. Joos, T. Kaminski,
C D. Keeling,, D. W. K. R.F. Keeling, K.E. Kohfeld,W. Knorr, R. Law, T.
Lenton, K. Lindsay, E. Maier-Reimer,, R. J. M. A.C. Manning, A.D.
McGuire, J.M. Melillo, R. Meyer, M. Mund, J.C. Orr, S. Piper, K.
Plattner,, and S. S. P.J. Rayner, R. Slater, S. Taguchi, P.P. Tans, H.Q. Tian,
M.F. Weirig, T. Whorf, A. Yool. 2001. The Carbon Cycle and
Atmospheric Carbon Dioxide. Pages 183-285 in J. T. Houghton, editor.
Cambridge University Press, Cambridge.
Raffa, K. F., B. H. Aukema, B. J. Bentz, A. L. Carroll, J. A. Hicke, M. G. Turner,
and W. H. Romme. 2008. Cross-scale Drivers of Natural Disturbances
Prone to Anthropogenic Amplification: The Dynamics of Bark Beetle
Eruptions. BioScience 58:501-517.
Rapp, V. 2004. Western Forests, Fire Risk, and Climate Change. Page 12. Forest
Service, U.S.Department of Agriculture, Portland, OR: Pacific Northwest
Research Station.
Roe, A. L. and G. D. Amman. 1970. Mountain Pine Beetle in Lodgepole Pine
Forests. Research Paper INT-71, USDA Forest Service, Intermountain
Research Station.
Rollins, M. G. 2009. LANDFIRE-a nationally consistent vegetation, wildland fire,
and fuel assessment. International Journal of Wildland Fire 18.
83


Romme, W. H. 1982. Fire and Landscape Diversity in Subalpine Forests of
Yellowstone National Park. Ecological Monographs 52:199-221.
Romme, W. H. and D. H. Knight. 1981. Fire Frequency and Subalpine Forest
Succession Along a Topographic Gradient in Wyoming. Ecology 62:319-
326.
Romme, W. H., D. H. Knight, and J. B. Yavitt. 1986. Mountain Pine Beetle
Outbreaks in the Rocky Mountains: Regulators of Primary Productivity?
The American Naturalist 127:484-494.
Running, S. W. 2008. Ecosystem Disturbance, Carbon, and Climate. Science
321:652-653.
Ryan, M., M. Harmon, R. Birdsey, C. Giardina, L. Heath, R. Houghton, R.
Jackson, D. McKinley, J. Morrison, BC Murray, D. Pataki, and K. Skog.
2010. A Synthesis of the Science on Forests and Carbon for U.S. Forests.
Ecological Society of America, Issues In Ecology 13:1-16.
Safranyik, L., A. L. Carroll, J. Regniere, D. W. Langor, W. G. Riel, T. L. Shore, B.
Peter, B. J. Cooke, V. G. Nealis, and S. W. Taylor. 2010. Potential for
Range Expansion of Mountain Pine Beetle into the Boreal Forest of North
America. The Canadian Entomologist 142:415-442.
Schmoldt, D. L., David L. Peterson, R. E. Keane, J. M. Lenihan, D. McKenzie, D.
R. Weise, and D. V. Sandberg. 1999. Assessing the Effects of Fire
84


Disturbance on Ecosystems: A Scientific Agenda for Research and
Management. USFS, Pacific Northwest
Research Station.
Sibold, J. S., T. T. Veblen, K. Chipko, L. Lawson, E. Mathis, and J. Scott. 2007.
Influences of Secondary Disturbances on Lodgepole Pine Stand
Development In Rocky Mountain National Park Ecological Applications
17:1638-1655.
Siegenthaler, U. and J. L. Sarmiento. 1993. Atmospheric carbon dioxide and the
ocean. Nature 365:119-125.
Sitch, S., C. Huntingford, N. Gedney, P. E. Levy, M. Lomas, S. L. Piao, R. Betts,
P. Ciais, P. Cox, P. Friedlingstein, C. D. Jones, I. C. Prentice, and F. I.
Woodward. 2008. Evaluation of the terrestrial carbon cycle, future plant
geography and climate-carbon cycle feedbacks using five Dynamic Global
Vegetation Models (DGVMs). Global Change Biology 14:2015-2039.
Smith, T. M., W. P. Cramer, R. K. Dixon, R. Leemans, R. P. Neilson, and A. M.
Solomon. 1993. The Global Terrestrial Carbon Cycle. Water, Air and Soil
Pollution 70:19-37.
Son, E., a J.-J. Kim,b Y.W. Lim,c T.T. Au-Yeung,a C.Y.H. Yang,a C. Breuila
2010. Diversity and decay ability of basidiomycetes isolated from
lodgepole pines killed by the mountain pine beetle. Canadian Journal of
Microbiology 57:33-41.
85


Stage, A. R. 1973. Prognosis model for stand developement. USDA Forest
Service, Ogden, UT Intermountain Forest and Range Experiment Station.
Stohlgren, T. J. and R. R. Bachand. 1997. Lodgepole Pine (Pinus contorta)
Ecotones in Rocky Mountain National Park, Colorado, USA Ecology
78:632-641.
Swift, R. S. 2001. Sequestration of Carbon By Soil. Soil Science 166:858-871.
Tallis, H., T. Ricketts, E. N. Driss Ennaanay, Kari Vigerstol, Guillermo Mendoza,,
N. O. Stacie Wolny, Juliann Aukema, John Foster, Jessica Forrest, Dick
Cameron,, andK. Arkema. 2008. In VEST 1.003 beta Users Guide.
Stanford.
Thorsell, J. and T. Sigaty. 1997. A Global Overview of Forest Protected Areas on
the World Hertitage List Natural Heritage Programme, IUCN, Gland,
Switzerland.
USFS. 2009. Forest Health Damage Detection Surveys Acres with Mortality.
USDA Forest Service.
Veblen, T. T., K. S. Hadley, M. S. Reid, and A. J. Rebertus. 1989. Blowdown and
stand development in a Colorado subalpine forest. Canadian Journal of
Forest Research 19:1218-1225.
Vose, J. 2006. Forest Carbon Cycling. Encyclopedia of Environmetrics. John
Wiley & Sons, Ltd.
86


Westerling, A. L., H. G. Hidalgo, D. R. Cayan, and T. W. Swetnam. 2006.
Warming and Earlier Spring Increase Western U.S. Forest Wildfire
Activity. Science 313:940-943.
Westerling, A. L., M. G. Turner, E. A. H. Smithwick, W. H. Romme, and M. G.
Ryan. 2011. Continued warming could transform Greater Yellowstone fire
regimes by mid-21st century. Proceedings of the National Academy of
Sciences 108:13165-13170.
Whelan, R. J. 1995. The Ecology of Fire. Cambridge University Press, UK.
White, M. A., P. E. Thornton, S. W. Running, and R. R. Nemani. 2000.
Parameterization and Sensitivity Analysis of the BIOME-BGC Terrestrial
Ecosystem Model: Net Primary Production Controls. Earth Interactions
4:1-85.
Wigley, T. M. L. and D. S. Schimel, editors. 2000. Constraints on the atmospheric
carbon budget from spatial distributions of C02. Cambridge University
Press, Cambridge, UK.
Winjum, J. K., R. K. Dixon, and P. E. Schroeder. 1992. Estimating the global
potential of forest and agroforest management practices to sequester
carbon. Water, Air, & Soil Pollution 64:213-227.
Zhu, Z., C. Key, D. Ohlen, and N. Benson. 2006. Evaluate Sensitivities of Bum-
Severity Mapping Algorithms for Different Ecosystems and Fire Histories
87


in the United States. USGS, National Center for Earth Resources
Observation and Science, Sioux Falls, SD.
88


Appendix I. Raw field data from 2 plots impacted by fire
89


fine Woody Debris (< 8 cm) and duff/lltter depths 15-m (Pt 1.) and 25-m (Pt. 2) from plot center
^"Va^sect 1 ft Slope (%) 1-hr (0-0.6) cm 10-hr (0.6 -2.5) cm 100-hr (2.5 8.0) cm Utter & duff depth (cm) Utter depth (cm)
Pt. 1 Pt. 2 Pt. 1 Pt. 2
' r -o-
2 120
3 240* i 2. -2,<
4-60*
5 -180*
6 300*
Vegetation Cover- 1-m radius plot at 15-m (Pt. 1} and 25-m (Pt. 2) from plot center
Live tree / shrub cover (%) Dead tree / shrub cover (%> Live herb cover (%) Dead herb cover (%) Avg. herb height (m) Rock cover <*> Soil cover m
Pt. 1 Pt. 2 Pt. 1 Pt. 2 Pt. 1 Pt. 2 Pt.l Pt. 2 Pt.l Pt. 2 Pt.l Pt. 2 Pt. 1 Pt. 2
l-0
2 120*
3-240* n , O A' 4/ Ao n (?.i Os< O.i
4-60* r
5 -180*
6-c00
>3 in. or >N cm
sampling r> llmirr
Coarse Woody Debris (> 8 cm) If the
central axis of the piece lies in or below
the duff layer then it should not be
Transect n Log # Diameter (cm) Decay Class
i ~h 3
1 r
3
4 rz- -> _v
*J
3 6 5
7
3 n


























Full Text

PAGE 1

IMPACTS OF MOUNTAIN PINE BEETLE ( DENDROCTON US PONDEROSAE ) AND FIRE DISTURBANCES ON FOREST ECOSYSTEM CARBON DYNAMICS AND SPECIES COMPOSITION by Megan K. Caldwell B.A., University of Colorado Denver, 2009 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Masters of Science Masters of Science Environmental Science 2012

PAGE 2

ii This thesis for the Masters of Science Environmental Science degree by Megan K. Caldwell has been approved for the Masters of Science Environmental Science by Casey Allen, Ch air Jon Barbour Frederick Chambers Date _____________________

PAGE 3

iii Caldwell, Megan, K. (M.S., Environment al Science) Impacts of Mountain Pine Beetle ( Dendroctonus Ponderosae ) and Fire Disturbances on Forest Ecosystem Carbon Dynamics and Species Composition Thesis directed by Casey Allen. ABSTRACT Forests play an important role in storing and sequestering carbon, where conifer forests in particular, store more than 33% within the terrestrial carbon pool. Disturbances, such as fire and insects, impact the amount of carbon that can be stored over time in conifer forests. Stand composition and structure, which plays an important role in carbon storage over time, may be altered by these large disturbances. A mountain pine beetle ( Dendroctonus ponderosae, MPB) epidemic has impacted lodgepole forests along the Rocky Mountains, and has potentially altered carbon storage and stand composition trajectories in the short and long term. This research used the Forest Vegetation Simulator (FVS) to quantify the scope and magnitude of the impacts of MPB on carbon storage and stand composition in a 200 year simulation. FVS was initialized with forest inventory tree, advanced regeneration and fuels data collected in 2010 in Grand County, Colorado, where Grand County was the epicenter of the MPB outbreak i n the Southern Rocky Mountains. This FVS simulation carbon and stand composition

PAGE 4

iv MPB were recoded as live to represent the conditions before the major mortality years of the MPB compare MPB disturbance to fire disturbance. Carbon and stand trajectories were altered between the three simula tions, showing MPB has altered forest carbon storage and stand structure, which is different from how fire disturbances affect carbon and species composition. There were differences in the trajectories of carbon storage and stand composition between plots based on initial species composition as well. MPB impacts carbon storage on a relatively short temporal scale, and impacts stand composition on a longer time frame. Fire disturbance seems to affect carbon storage in lodgepole forests more drastically, and for a slightly longer time period. These results aid in management for optimal carbon storage while facing a greater potential for coarse scale disturbances in a changing climate. The form and content of this abstract are approved. I recommend its public ation. Approved: Casey Allen

PAGE 5

v DEDICATION I dedicate this thesis to my beautiful daughter, Aderyn Jade Caldwell

PAGE 6

vi ACKNOWLEDGMENTS I would like to acknowledge the U.S. Geological Survey Rocky Mountain Geographic Science Center, the U.S.G.S. Geographic Analysis and Monitoring and Land Remote Sensing programs, and thank my primary advisor at U.S.G.S., Todd Hawbaker, for providing the fu nding and guidance for the completion of this research.

PAGE 7

vii TABLE OF CONTENTS Figures ................................ ................................ ................................ ..................... ix Tables ................................ ................................ ................................ ....................... x Chapter 1. Introduction ................................ ................................ ................................ ......... 1 2. Literature Review ................................ ................................ ............................... 5 2.1 Terrestrial Carbon Cycling in Forests ................................ ............................. 5 2.2 Forests and Climate Change ................................ ................................ .......... 11 2.3 Conifer Forests and Carbon Cycling ................................ ............................. 12 2.4 Disturbances and Lodgepole Forests ................................ ............................. 1 5 2.5 Succession, Carbon and Disturbances on Lodgepole Forests ....................... 21 2.6 Lodgepole Pine Ecology ................................ ................................ ............... 23 3. Objectives and Uncertainties ................................ ................................ ........... 2 5 3.1 Objectives ................................ ................................ ................................ ...... 25 3.2 Uncertainties and Limitations ................................ ................................ ....... 2 7 4. Methods ................................ ................................ ................................ .............. 2 9 4.1 Field measurements and Modeling ................................ ............................... 2 9 4.2 Study Area ................................ ................................ ................................ ..... 2 9 4.3 Field Methods ................................ ................................ ................................ 31 4.4 Modeling Methods ................................ ................................ ........................ 33 4.5 Model Scenarios ................................ ................................ ............................ 34 4.6 Analysis Methods ................................ ................................ .......................... 37 5. Results ................................ ................................ ................................ ................ 42 5.1 2010 Conditions ................................ ................................ ............................ 42

PAGE 8

viii 5.2 Impacts of MPB on Carbon Storage ................................ ............................. 4 6 5.3 Impacts of Simulated Fire Disturbance on Carbon Storage .......................... 4 7 5.4 Impacts of MPB on Stand Structure ................................ .............................. 54 5.5 Impacts of Simulated Fire Disturbance on Stand Structure .......................... 57 5.6 Variation in Plots for Carbon and Stand Structure Trajectories ................... 57 5.6.1 Variation in Carbon Trajectories between Plot Types ...................... 58 5.6.2 Variation in Stand Composition Trajectories between Plot Types ... 64 6. Discussion Conclusions and Recommendations ................................ .......... 6 5 6.1 Impacts of MPB on Carbon Storage ................................ ............................ 6 6 6.2 Impacts of Simulated Fire Disturbance on Carbon Storage ......................... 6 7 6.3 Impacts of MPB on Stand Structure ................................ ............................. 68 6.4 Impacts of Simulated Fire Disturbance on Stand Structure ......................... 7 0 6.5 Conclusions and Recommendations ................................ ............................. 72 6 5 .1 Conclusions ................................ ................................ ........................ 72 6 5 2 Recommendations and Management Implications ............................ 73 References ................................ ................................ ................................ ............. 74

PAGE 9

ix LIST OF TABLES T able 5.1 Surveyed Lodgepole Pine Mortality ................................ ................................ ......... 4 3

PAGE 10

x LIST OF FIGURES Figure 2.1 The Global Carbon Cycle (from CCSP 2007) ................................ .............................. 8 2.2 Forest Acres with Tree Mortality (From the United States Forest Service (USFS) Forest Health Damage Detection Surveys, 2009). ................................ ............... 20 4.1 Study Area ................................ ................................ ................................ ................. 31 4.2.Field Plots by Dominant Composition Type ................................ ............................. 4 0 4.3 Plot Types ................................ ................................ ................................ ................... 41 5.1 Total Stand Carbon. ................................ ................................ ................................ .... 50 5.2 Standing Live Carbon ................................ ................................ ................................ 51 5.3 Standing Dead Carbon ................................ ................................ ................................ 52 5.4 Downed Dead Carbon ................................ ................................ ................................ 53 5.5 (a c) Average Species Comp osition by Percent of Basal Area ................................ .. 55 5.6 Beginning and End Species Trajectory ................................ ................................ ...... 56 5.7 Standing Live Carbon by Plot Type ................................ ................................ ............ 60 5.8 Standing Dead Carbon by Plot Type ................................ ................................ ......... 6 1 5.9 Downed Dead Carbon by Plot Type ................................ ................................ .......... 6 2 5.10 Total Stand Carbon by Plot Type ................................ ................................ ............. 6 3 5.11 (a. c) Species Trajectories between Plot Types ................................ ...................... 6 5

PAGE 11

1 I. INTRODUCTION The thesis presented here is intended to address two main research questions. Forest disturbances can impact carbon cycling and species composition trajectories over time. First of all, it is important to be a ble to quantify the extent and severity of these forest disturbances. The scope and magnitude of how disturbances alter the carbon storage and species composition trajectories may vary between forest types and disturbances, which could have important impli cations for forest management. This thesis stands to first quantify the extent and severity of the mountain pine beetle (MPB) epidemic in eastern Grand County, Colorado. Next, the differences in carbon and species composition trajectories over time that ha ve occurred as a result of the MPB epidemic in the study area were quantified out to 2210. Following, the carbon and species composition trajectory alterations were compared to those resulting from a simulated fire. Fires are the other common disturbance t hat typically affects a large expanse in the study area. The purpose is to quantify how disturbances such as mountain pine beetle and fire alter stand and carbon storage trajectories over time compared to undisturbed forests. This research also compared th ree categories of forest type to note differences in the trajectories for each scenario between plots, for insight at landscape and local scales. Differences in carbon storage by the stand composition in plots were also addressed. Managing for carbon seque stration and storage in

PAGE 12

2 Windows based computer. The two fonts print equally well on either platform so this issue is only quality of display on the screen. forests could be enhanced by managing for species composition that stores more carbon. This research is important because in order to most efficiently manage forests, it will be important to know how disturbances alter the stand and carbon storage trajectories over time. Forests, in particular, store much of the terrestrial carbon sequestere d from the atmosphere. Disturbances threaten this carbon storage potential. This is an increasing threat as disturbances are projected to increase in frequency and severity during modeled warming climate scenarios. Forests will become increasingly threaten ed by disturbances, which could subsequently threaten carbon storage, cycling and stand dynamics in both the long and short term over multiple spatial scales. Landscape scale carbon flux and pool estimates reflect the dynamics of local scale measurements, as forest stand carbon dynamics are highly variable and respond individually to differing management and disturbance regimes, as well as local scale site characteristics and vegetation ( CCSP 2007 ) Thus, to be able to understand the processes occurring at a landscape scale, local scale processes must first be quantified. This study incorporated both a field survey in Grand County, Colorado and modeled vegetation projections to quantify the impacts of an extensive mountain

PAGE 13

3 pine beetle outbreak, compare the mountain pine beetle outbreak to undisturbed condi tions and to a simulated fire of similar extent on stand carbon and structure. Stand carbon and species composition were tracked at the individual tree and plot level, and averaged across plots as well for a more landscape level insight. The basic resul ts of this survey determined that mountain pine beetle has altered carbon and species trajectories from pre disturbance conditions. Stand carbon, however, recovers relatively quickly, where total stand carbon storage recovers by 2040 and standing live carb on recovers to pre disturbance storage by 2060. Stand composition trajectories have changed from pre disturbance trajectories, from a higher percentage of the basal area being lodgepole pine to having a more mixed composition with a higher percentage of su balpine fir especially. There is some variance in carbon storage and species composition between plots by the type of overstory and understory species present in plots in 2010. Stand composition trajectories seem to rely heavily on remaining live vegetatio n present immediately following disturbance. The impacts of insect disturbance and how they alter stand composition and carbon trajectories differs from how fire disturbances impact them as well. This research is organized within this thesis by a review of literature on the cycling of carbon terrestrially on a global scale, how forests and disturbances the affect forests may be impacted by climate change, carbon storage in conifer forests specifically, how disturbances impact lodgepole forests and how car bon storage,

PAGE 14

4 species composition and disturbances interact in lodgepole forests. Then, the objectives of this study and uncertainties within the research are addressed. The methods are outlined by the study area, field methods and modeling, the specific mo del scenarios simulated, and analysis methods. The results and discussion are laid out by 2010 conditions, investigating the impacts of mountain pine beetle disturbances on carbon storage, the impacts of fire disturbance on carbon and the impacts of mounta in pine beetle and fire on stand structure. Finally, the results and discussion are concluded in a summary and references are included at the end. Appendix I lists the raw plot data for two plots that were measured again in 2011 after an actual fire burned through them.

PAGE 15

5 II. LITERATURE REVIEW 2.1 Terrestrial Carbon Cycling in Forests Carbon is exchanged between and stored within the oceans, atmosphere and terrestrial carbon systems, as shown in figure 2.1 ( Wigley and Schimel 2000 ) Carbon flux is the amount of carbon moved between pools, and carbon storage refers to the amount within a pool that is not in movement, but is steadily held. The te rrestrial carbon cycle plays a significant role in the global carbon budget, as it is one of the three main land mass, fulfilling copious ecological roles ( Winjum et al. 1992 ) One of these ecological roles is terrestrial carbon storage and seques tration within forest biomass and soils, where forests house an estimated 45 60% of the global terrestrial carbon pool ( CCSP 2007 ) .Worldwide, forests store about 2.07 x 10 12 Mg (2280 gigatons (Gt)) of carbon, and coniferous forests alone store more than 33% of terrestrial carbon ( Smith et al. 1993 Kashian 2006 CCSP 2007 ) The main focus of this thesis will be specifically subalpine conifer forest carbon storage. Terrest rially, 36% of the land area of North America which accounts for approximately half of the carbon sink in North America is provided by forest ecosystems which offset excess atmospheric carbon inputs, such as greenhouse gas emissions ( CCSP 2007

PAGE 16

6 Tallis et al. 2008 ) In 2003, carbon emissions in North America were about 2 billion Mg of carbon but 256 million Mg were offset by forests ( Heath and Smith 2004 Birdsey et al. 2006 CCSP 2007 Pacala et al. 2007 Goward et al. 2008 Ryan et al. 2010 ) In comparison, the global atmospheric carbon pool (estimated for 2003) stores about 7.05 x 10 11 M g (777 Gt) of carbon as carbon dioxide, where 5.35 x 10 11 Mg (590 Gt) of carbon are from non anthropogenic terrestrial sources of release, such as geologic features, fires and decomposition, and 1.69 x 10 11 Mg (187 Gt) are from anthropogenic contributions ( Lal et al. 2000 Lal 2004 CCSP 2007 ) The terrestrial and atmospheric carbon pools are directly linked through forests, where forest s absorb carbon dioxide and release oxygen back to the atmosphere. An imbalance between sources and sinks in the terrestrial carbon cycle may cause a subsequent increase of carbon in the atmosphere, and changes in ( CCSP 2007 ) Oceanic carbon storage is approximately 50 times greater than th e atmospheric sink and cycling between the ocean and atmosphere occurs on the time scale of hundreds of years ( Prentice et al. 2001 ) Between the ocean and atmosphere, there is about a 2 Mg/yr flux in balance overall from the atmosphere to the ocean (wher e 90 Mg/yr is transferred to the atmosphere to the ocean, and 92 Mg/yr is transferred from atmosphere to

PAGE 17

7 ocean storage). The largest storage of carbon dioxide in the ocean is the dissolved inorganic carbon ( Siegenthaler and Sarmiento 1993 ) The oceanic carbon sequestration capacity may be significantly reduced if the global climate warms, as it has done in modeled GCM scenarios where ca rbon dioxide concentrations were increased. This modeled reduction of oceanic carbon uptake occurs from increased sea surface temperature on carbon dioxide solubility, as well as reduced vertical mixing on carbon dioxide transport from the surface to deep ocean ( Friedlingstein et al. 2001 ) This reduction in oceanic carbon dioxide uptake increases the need for protection of terrestrial carbon sinks, which may be easier for management. Carbon stored terrestrially could incur changes from climate change quickly. Increased atmospheric carbon dioxide could possible lead to increased vegetation and soil carbon ( Prentice et al. 2001 ) This could be offset through increased disturbances howev er. This has direct implications for management, and could possibly be mitigated.

PAGE 18

8 Figure 2.1 The Global Carbon Cycle (from CCSP 2007). The global carbon cycle, shown with all three primary carbon pools with fluxes. Black numbers show natural amounts and red numbers are for brackets, where fluxes are listed without brackets. A primary sequestration method within the terrestrial carbon cycle occurs through vegetation, where primary pr oductivity removes carbon dioxide from the atmosphere through photosynthesis and converts some of it to biomass (as well as respiring some back to the atmosphere), allowing for atmospheric carbon to accumulate and become stored in biomass in

PAGE 19

9 what is referr ed to as the above ground carbon pool (Vose 2006). Sometimes carbon input into the terrestrial carbon cycle is measured as Net Ecosystem Production (NEP), or the difference between Gross Primary Productivity (GPP) and what the plant needs to use for mainte nance over time ( Chapin III et al. 2002 ) The biomass accumulated through this carbon sequestration eventually falls and decomposes in the detrital carbon pool. And, over time, it decomposes enough to become part of the soil and humus, or below ground c arbon pool ( Vose 2006 ) Thus, the amount of carbon stored in these pools in any ecosystem is ultimately a function of the rates of primary productivity, respiration, mortality, decomposition ( Monson et al. 2002 ) Sources and sinks of carbon tend to determine how much carbon a particular ecos ystem can store. A source of carbon outputs more carbon than it sequesters, and the opposite holds true for a carbon sink, where inputs outweigh the outputs. Typically, live vegetation is a sink for carbon. A source is an ecosystem component that decompose s and loses carbon faster than it can be sequestered. The amount of carbon stored in each component on an ecosystem can vary widely; however typically, the most amount of carbon is stored in the aboveground live vegetation and the belowground components ( Ch apin III et al. 2002 ) Carbon sequestration is listed as an important ecosystem service, which is to say that it is a societal benefit provided by natural ecosystems.

PAGE 20

10 As an ecosystem service, carbon sequestration and storage is extremely valuable to our society, and would be difficult to regain if lost ( Heal 2000 ) In order to preserve and protect this ecosystem service, resource managers need information on how carbon is distributed and stored in specific ecosystems, as well as which carbon stores are changing and why ( Joyce and Birdsey 2000 Tallis et al. 2008 ) Some ecosystem service models quantify carbon sequestration by splitting up the amounts sequestered in forests into four different pools: aboveground biomass, belowground biomass, soil, and dead organic matter ( Tallis et al. 2008 ) so this research will talk about carbon storage in these four pools, generally. It is important to be able to quantify carbon s tores in each pool, and important to recall that the amount of carbon stored as well as the fluxes within and between ecosystems. For example, age and type of vegetation in an ecosystem makes a difference in how much carbon can be stored in biomass. More m ature forests sequester carbon in the aboveground pool more slowly than their younger counterparts ( Chapin III et al. 2002 Monson et al. 2002 ) Due to the wide variance of carbon sequestration capacity between different ecosystems, it is important to investigate carbon storage on local scale.

PAGE 21

11 2.2 Forests and Climate Change It is uncertain how forests will respond to a changing climate. This is particularly true in the aboveground biomass pool, which may pressure changes in the other main carbon pools ( CCSP 2007 ) Forest species ranges are projected to move northward and higher in elevation, where range expansions may impact primary ecosystem processes like succession and disturbance and subsequent changes in carbon cycling for many years into the future ( Joyce and Birdsey 2000 ) Disturbances may increase in severity and frequency in a warming climate scenario ( IPCC 2007 Bentz et al. 2010 Westerling et al. 2011 ) Increased drought in some areas due to climate change can make some species more susceptible to distu rbance and also less productive ( Dale et al. 2001 ) In some modeled warming climate scenarios, terrestrial ecosystems become a source of carbon dioxide, causing additional atmospheric carbon to accrue ( Kurz et al. 2008 Sitch et al. 2008 ) There is a pressing need to understand how the potential of forests to sequester carbon may change over time in response to some of these potential impacts of climate change ( Newell and Stavins 2000 ) Some of the uncertainties present when facing a changing climate and forest carbon cycling will be addressed later in this paper. The negative impacts of climate change may be offset by positive changes for vegetation that could occur from increased atmospheric carbon

PAGE 22

12 dioxide levels. For example, productivity of vegetation could occur. Doubling atmospheric carbon dioxide concentration, under controlled greenhouse conditions, has increased plant productivity a nd yield by more than 30% on average, a 37% decrease in stomatal conductance, which increases leaf temperature by 1 degree Celsius and decreases evapotranspiration, even though these percentages may vary by species ( Kim ball et al. 1993 ) However, in lodgepole forests, increased warming and carbon dioxide could have negative impacts if precipitation does not increase as well because lodgepole become moisture stressed and their ranges shrink ( Barrow and Yu 2005 Hamann and Wang 2006 Monserud 2008 ) T he impacts of climate change on vegetation, and lodgepole in particular, are a function between the potential positive impacts on productivity and evapotranspiration compared to the potential negative impacts through disturbances and drought increases. The impacts of drought and disturbances on forests may negate the potential positive impacts of increased carbon dioxide. 2.3 Conifer Forests and Carbon Cycling Coniferous forests cover approximately 15% of terrestrial land mass, equivalent to 10 million km 2 in area ( Thorsell and Sigaty 1997 ) Conifer forests contain 33% of all stored carbon in terrestrial ecosystems and sequester much of the carbon in western North America ( Smith et al.

PAGE 23

13 1993 Kashian 2006 ) Of the conifer forests in North America, and in the western U.S. in particular, one of the most prevalent forest types is the lodgepole pine ( Pinus contorta ), covering 20 million hectares (ha) of area and 6 million ha are dominated by lodgepole in the western United States ( Lotan and Critchfield 2004 ) Within Rocky Mountain forests, the below ground pool in lodgepole forests tends to store the greatest amount of carbon (49%), followed by the above ground vegetation (38%), detrital (12%), and understory herbaceous cover (1%) vegetation pools respectively ( Birdsey 1992 ) Carbon storage in forested ecosystems varies with forest age and sta nd composition ( Bradford et al. 2008 ) Carbon flux in a lodgepole forest has been shown to depend on climate and the spatial distribution of trees ( Kueppers and Harte 2005 ) Although the range of carbon storage in lodgepole forests varies between sites, generally, the aboveground live carbon pool in a lodgepole dominated fores t contains around 67.9 Mg of carbon per hectare on average in the Rocky Mountain region ( Birdsey 1992 ) The concentrated carbon in plant tissues is about 50% for overstory conifer trees, 45% for herbaceous tissues often found in a forest understory, but is h ighest in lipid rich tissues from plant material, such as seeds. Carbon storage in lodgepole forests is affected by several factors, including stand density, stand age, species distribution, and is essentially a product of the balance between carbon on the forest floor and

PAGE 24

14 sequestration in biomass and the carbon lost through decomposition ( Kashian 2006 ) Forest soils are one of the largest carbon storage components ( Birdsey 1992 ) Carbon is incorporated into soil in the subalpine forest ecosystem from the flux of biomass from the live to the dead pool. Decomposition, a major component in the carbon cycle, is mainly carried out by fungi in lodgepole forests. Sap rot fungus is responsible for breaking down much of lodgepole litter. Fungi is responsible for much of the decomposition in subalpine forests because often soil microbes cannot survive in the acidic soils produced under lodgepole forests or cold temperatures at high elevations ( Son 2010 ) Woody material in a forest is comprised of mainly lignin. This high lignin content combined with climatic factors and soil acidity cause decomposition in subalpine forest soils to be very slow. The biological capacity of an ecosystem to decompose organic matter is affected by organic inputs such as litterfall or excretion by organisms in the ecosystem, which subsequently exerts control on the rate of decomposition ( Marschner and Rengel 2007 ) Aboveground carbon storage in forest biomass is especially threatened by climate change. Small changes in the delicate balance between photosynthesis and respiration and decomposition could result in increased emissions to the atmosphere (Pregitzer and Euskirchen 2004).

PAGE 25

15 2.4 Disturbances and Lodgepole Forests Disturbances, as above mentioned, are on e factor that is expected to increase in severity and frequency under warming climate scenarios ( Dale et al. 2001 IPCC 2007 ) Forests are periodically affected by disturbance ( Roe and Amman 1970 Romme et al. 1986 ) Immediately following a disturbance, a forest can not only discontinue sequestering carbon, but can even act as a source instead of a sink for carbon, and the tim e it takes for the forest to once again return to being a sink varies by forest type and stand age or disturbance history ( Bradford et al. 2008 Goward et al. 2008 Kurz et al. 2008 ) Disturbances can alter carbo n cycling dynamics considerably by causing extensive tree mortality, reducing photosynthetic capacity and potentially altering carbon flux rates between the various carbon pools ( Kurz et al. 2008 ) Resulting changes in biomass pools and environmental conditions may alter rates of decomposition and further alter rates of other ecosystem processes, like regeneration ( Dale et al. 2001 ) The cumulative effect is that forests can shift from being carbon sinks to carbon sources over long temporal scales ( Kurz et al. 2008 Raffa et al. 2008 ) Some disturbance events that affect forests include fire ( Romme 1982 ) wind blowdown ( Veblen et al. 1989 )

PAGE 26

16 insect mortality ( Roe and Amman 1970 ) or human management events, such as thinning and harvesting ( Franklin et al. 2002 ) Wildfire and insect disturbances appear to impact the largest acreage in forests in the western United States, and in the Rocky Mountain coniferous lodgepole forests mentioned above ( Dale et al. 2001 USFS 2009 ) Fire disturbances may vary in response to a warming climate as the frequency, size and intensity of fire is directly depende nt on weather and precipitation patterns, as well as forest type ( Dale et al. 2001 Westerling et al. 2006 ) Fires are generally products of climate and available fuels ( Schmoldt et al. 1999 ) The most extreme fire events tend to burn the most area and are controlled primarily by climate, where drought causes extr eme fire seasons ( Bessie and Johnson 1995 ) The seasonal severity rating of forest fires is modeled to increase by 10 15% depending on location by 2060, which increases forest fire activity, where fires are one of the most rapid disturbances to respond to a warming climate ( Flannigan et al. 2000 ) Forest fires have burned increasing amounts of acreage in the last decade which can be attrib uted to climatic factors, such as increased seasonal temperatures and earlier snowmelt, suggesting management may not be effective if the climate continues on a warming trend ( Westerling et al. 2006 )

PAGE 27

17 Aboveground carbon storage, such as in the aboveground live biomass and in the top layers of forest floor and downed dead pools, could be released and turned into a carbon source if consumed by a high intensity fire ( Breshears and Allen 2002 Hurtt et al. 2002 Kashian 2006 Hurteau and North 2008 2009 ) C arbon is released directly to the atmospheric carbon pool ( Flannigan et al. 2000 Rapp 2004 ) Fire can disturb carbon storage in soils both through acceleration of nutrient cycling and changes in the top soil layer chemistry ( Whelan 1995 Dale et al. 2001 Swift 2001 ) Insect disturbances may be directly influenced by a warming climate, where the spread and range of insects, as well as the susceptibility of forests to insects typically increase under modeled warming climates ( Dale et al. 2001 Raffa et al. 2008 Klutsch et al. 2009 ) Among other fac tors, forests become more susceptible to insects through the stress of drought in a warming climate ( Amman 1977 ) Insects that disturb forests typically have been controlled by climatic conditions, so as climate changes, these controls are relaxed. Bark beetles, for example, may expand ranges northward eastward, and toward higher elevations, generally ( Carroll et al. 2003 Safranyik et al. 2010 ) This would correspond with forest species range expansions, possibly creating opportunities for gr eater scale disturbances, and great impacts to carbon cycling and future species succession trajectories ( Joyce and Birdsey 2000 )

PAGE 28

18 Insect disturbances, such as the mountain pine beetle ( Deondroctonus ponderosae ; MPB), typically impact forests differently than other disturbances because infestation occurs in selectively larger diameter trees ( Amman 1977 ) potentially having different imp acts to carbon storage and flux than other disturbances. MPB infests in areas where lodgepole basal area is high ( Klutsch et al. 2009 Pfeifer et al. 2011 ) Other species of trees, as well as smaller DBH lodgepole persist in stands, where large gaps are left once trees killed by MPB fall. Some level of live biomass is maintained as compared to a stand replacing fire, which eliminates most of the live biomass over the affected area. Wildfire disturbance transfers carbon out of the ecosystem to the atm osphere and within the ecosystem in the form of ash to the forest floor, while mountain pine beetle transfers carbon to the standing dead and downed dead pools as detritus and eventually carbon is transferred to the atmosphere through decomposition on the forest floor. MPB was the top forest mortality agent in the conterminous United States for 2009, accounting for 73% of tree mortality in the conterminous United States accounting for almost 3,500,000 hectares, as shown in figure 2.2 ( USFS 2009 ) Historically, MPB has persisted at endemi c levels in the Southern Rockies, with periodic outbreaks ( Amman 1977 Baker and Veblen 1990 Raffa et al. 2008 Klutsch et al. 2009 ) Starting in 1996,

PAGE 29

19 MPB populations have grown rapidly to epidemic levels that are unprecedented in recorded history in the Rocky Mountains ( USFS 2009 ) The current epidemic has impacted millions of hectares (ha) of lodgepole pine ( Pinus contorta ) across North America and 777,000 ha in Col orado between 2000 and 2008 ( USFS 20 09 ) Pfeifer et al. in (2010), noted a short term change in carbon stocks and fluxes after a mountain pine beetle outbreak in an Idaho lodgepole forest, where there was immediately a maximum 83% decrease in carbon stocks and 73% in carbon fluxes (or th e rate of carbon sequestration) that were recovered in 25 years or less. Pfeifer et al. (2010) surveyed 12 plots in an Idaho forest consisting primarily of lodgepole and Douglas fir ( Pseudotsuga menziesii ) and modeled the trajectory of carbon pools and fluxes after MPB killed up to 52% of trees within plots. Substantial variability of carbon stocks and fluxes resulted from the size distribution of trees within the 12 plots.

PAGE 30

20 Figure 2.2 Forest Acres with Tree Mortality (From the United States Forest Service (USFS) Forest Health Damage Detection Surveys, 2009 ) Shows the acres of mortality, where 73% was from MPB in 2009. Landscape scale estimates of carbon storage change seem to depend heavily on the tim e since any disturbance ( CCSP 2007 ) Both anthropoge nic and natural disturbances can alter carbon cycling considerably by removing biomass and altering flux rates among biomass pools, but in spite of the recognized importance of disturbances, their potential long term impacts on carbon cycling has not been quantified extensively and incorporated into land models that project carbon over time ( Running 2008 ) Not only does disturbance impact short term carbon storage, it also could impact future stand trajectories and regeneration well

PAGE 31

21 into the future. In turn, species shifts could cause a long term shift in carbo n storage. Disturbances like the MPB and fire can affect species succession trajectories over long time periods. Forest species composition and tree age are important in determining how quickly carbon storage and sequestration recovers from MPB disturbanc es ( Fahey and Knight 1986 ) Carbon conte nt stored within and between species in forests can vary significantly ( Chapin III et al. 2002 Lamlom and Savidge 2003 Kashian et al. 2004 ) Thus, efforts to quantify the long term impacts of MPB on carbon stocks and fluxes should account for the potential changes in species composition that may occur following insect outbreaks. However, it is difficult to quantify or model future forest succession due to the many factors that must be accounted for including seed dispersal, topography of t he landscape, moisture levels, competition and light availability, soil conditions and future climate scenarios. 2.5 Succession, Carbon and Disturbances in Lodgepole Forests Fire and succession have been studied fairly extensively in lodgepole pine forest s. Fires over large areas of vegetated space typically initiate successions, but these succession events are dependent on seed sources and number of advanced regeneration ( Glenn Lewin et al. 1992 ) Fire is important in establishing new lodgepole forests, and most of the old

PAGE 32

22 growth lodgepole forests in North America were established through fire, especially in the Rockies ( Lotan et al. 1985 ) Lodgepole sometimes produces serotinous cones, where temperatures above 45 degrees Celsius caused by fire or sometimes summer surface soil temperatures cause serotinous cones to open. This varies greatly within the Rocky Mountains, however. And lodgepole s ometimes produces open cones in areas. ( Lotan 1976 ) In the absence of fire, lodgepole can be replaced by more shade tolerant species ( Lotan 1976 ) Management and fire suppression tend to cause fuel buildup, causing the potential for a high intens ity fire that may eliminate large amounts of biomass from forest stands if a fire were to occur ( Brown 1975 ) To recall from above, a high intensity fire is more likely in a changing climate, which could have substantial impacts on successio nal trajectories and carbon storage in lodgepole forests. Impacts of insect disturbance on forest regeneration and succession have not been quantified extensively. Lodgepole typically regenerates abundantly where the mineral seedbed is adequate enough, bu t can be hindered by a thick organic layer that tends to inhibit seedling recruitment, where litter and fuels accumulate on the forest floor after MPB ( Lotan and Perry 1983 Collins et al. 2011 ) Rem aining canopy after MPB may inhibit lodgepole establishment, and favor the growth of more shade tolerant species, such as Engelmann spruce and Subalpine fir ( Claveau et al. 2002

PAGE 33

23 Collins et al. 2011 ) Post MPB lodgepole regeneration is not limited by the viable seed availability in the serotinous cones left behind ( Aoki et al. 2011 ) Forest recovery after MPB may rely more on seedlings, saplings and residual l ive tree biomass, collectively known as advance regeneration, rather than new seedling recruitment, especially with the influence of a deep litter layer on the forest floor ( Klutsch et al. 2009 Collins et al. 2011 ) Pre epidemic forest conditions are a large determinant of MPB post epidemic forest trajectories ( Diskin et al. 2011 ) Modeling advance regeneration could pro vide greater understanding of future forest species composition and carbon storage. Post disturbance stand species composition is dependent on time non as MPB, that are high in severity, in young stands that have recently incurred fire disturbance, seem to favor lodgepole pine re establishment ( Sibold et al. 2007 ) 2.6 Lodgepole Pine Ecology Lodgepole pine is shade intolerant, and grows in three different ecological roles: seral, persistent and climax ( Roe and Amman 1970 ) Lodgepole is able to colonize after disturbances because they have easily dispersed seeds, can grow where the canopy is open and can grow on non ideal sites, such as nutrient poor soils or steep slopes ( Parker and Parker

PAGE 34

24 1983 ) Persist ence of lodgepole dominance in a stand is driven by topographic variables, fire frequency, seed sources, as well as an insect disturbance. Endemic MPB kills larger DBH lodgepole in 20 40 year cycles until lodgepole is eliminated from the stand ( Amman 1977 Romme and Knight 1981 ) Changes in the dominant canopy of forest stands where gaps are introduced drive species composition and seedling establishment after a disturbance ( Klutsch et al. 2009 ) and consequently impacts carbon storage. Specifically in Rocky Mountain National Pa rk, Colorado, lodgepole pine distribution is defined by elevation and moisture, and to a lesser extent, summer soil moisture and sand content, where the transition from lodgepole forests to subalpine fir forests correlated with summer soil moisture ( Stohlgren and Bachand 1997 ) The impacts of MPB disturbance on stand trajectories in Rocky Mountain National Park vary based on pre epidemic stand structure a nd composition, where there was high variance in future stand trajectories between overstory and understory species types present in plots before a MPB outbreak ( Diskin et al. 2011 )

PAGE 35

25 III. OBJECTIVES AND UNCERTAINTIES 3.1 Objectives It is possible that carbon storage potential in live biomass has been reduced due to the extensive tree mortality in the Southern Rocky Mountains as a result of an extensive MPB outbreak, and potentially impact carbon storage over longer time pe riods through changes in stand structure and species composition. Ecosystems encompass numerous complex interactions that define and drive carbon storage and flux, and future stand composition. Quantifying the long term impacts of insect outbreaks can be d ifficult because multiple processes must be accounted for including mortality, vegetation regrowth and succession, changes in primary productivity and decomposition rates, and potential for future disturbances. One of the best ways to incorporate these mul tiple processes is to utilize a modeling approach. Rates of mortality, drivers of growth and regrowth, potential for future disturbances and sometimes climatic variables can all be included with the use of ecosystem models. Also, when there is no historica l analogy to events, such as massive epidemic scale outbreaks, models can be used to project the potential trajectory after such event more efficiently than from observation of past and current events alone.

PAGE 36

26 The objectives of this study were to quantify t he amount of time required for aboveground forest carbon storage to recover to pre MPB outbreak levels for the actual epidemic that has occurred in a Rocky Mountain lodgepole forest, as well as to investigate species composition response to MPB disturbance in order to better understand how carbon storage will change over various temporal scales in insect disturbed forests. Then, the carbon and species succession data corresponding to the MPB epidemic will be compared to the carbon storage change and specie s composition after a simulated fire, to compare the impacts of these two common natural disturbances. A combination of in situ field data collection and vegetation simulation modeling was used to quantify carbon and stand structure in the Southern Rocky M ountains. It could possibly take a very long time for carbon storage to regain the carbon storage potential they would have had if not for the MPB due to high tree mortality, decomposition of increased litterfall from dead trees, and even potential species composition change. We compared the trajectories of vegetation growth and mortality on plots that had been impacted by MPB, where peak mortality occurred 2006 2008, to the trajectories of the same plots had there not been epidemic level beetle activity to measure the amount of carbon and species composition response through basal area

PAGE 37

27 over time to the MPB disturbance. Carbon was tracked in standing live, standing dead and downed dead pools over time. 3.2 Uncertainties and limitations There are several unc ertainties with this research that must be addressed. Only aboveground carbon is quantified, where much of the carbon in forest ecosystems is stored in the soil and root system. Lodgepole biomass is 20% belowground within the root system, so only part of the entire system is accounted for ( Comeau and Kimmins 1989 ) Not only is carbon storage and sequestration affected by stand de nsity, composition and age, it is also a function of litter decomposition and fall rates ( Kashian et al. 2004 ) This research did not address decomposition, fall rates or belowground soil processes. Also, FVS, the model used for these simulations, does not account for soil type or texture, which is important in s uccession and carbon storage. This research addresses conservative estimates of stand structure trajectories but does not address future disturbances that could potentially go along with those future trajectories, such as an increase in other disturbance insects or disease, such as spruce beetle or mistletoe. It could be difficult to track disturbances with the uncertainty that occurs with climate change.

PAGE 38

28 Improvements to this modeling method would be to incorporate soils and belowground processes and futu re disturbances for an integrated approach. Also, better, less conservative, regeneration estimates would have aided in projecting stand trajectories over time, such as utilizing better predictors from wind dispersal, mineral seed bank data and typical sta nd trajectories after MPB disturbance.

PAGE 39

29 IV. METHODS 4.1 Field Measurements and Modeling In order to quantify the impacts of a MPB outbreak, and further the study of disturbance impacts on carbon storage and succession in the study area, I combined field data collection with vegetation simulation modeling. Model simulations can track large amo unts of carbon and species composition over time, and make projections based on current trajectories ( Bazz az 1996 ) This methodology was also chosen because there is no historical analogy to the current mountain pine beetle outbreak, where modeling allows me to initialize the simulations with current conditions established with field data and then quantify potential changes in forest vegetation and carbon under a series of scenarios. 4.2 Study Area The study area was located in eastern Grand County Colorado (105 43' 32" 106 0' 47" W and 3954' 58" 4018' 2" N). Forests in the study area are even aged stands of lodgepole pine ( Pinus Contorta ) with subalpine fir ( Abies Lasiocarpa ) seedlings and saplings. Average stand age is approximately 70 years. A large percentage of the study area is public land where most wildfires are suppressed. Until recently, the disturbance history of the area consisted of fires of mixed severity and per iodic bark beetle outbreaks at endemic levels. Beginning in 1996, an extensive and severe MPB outbreak started in the Southern Rocky Mountains with peak MPB mortality occurring between 2005 and 2008. The

PAGE 40

30 epicenter of this outbreak was in Grand County, mak ing it an ideal area for studying the impacts of MPB and other disturbances on forest vegetation and carbon storage.

PAGE 41

31 Figure 4.1 Study Area. This map shows the study area within eastern Grand County, Colorado where 119 field plots were placed in a ran domly stratified sampling scheme and forest inventory data was collected. 4.3 Field methods Field measurements were collected to characterize forest vegetation present in 2010, following the peak of MPB mortality. Plot locations were selected using strati fied random sampling. The strata used included a gradient of years since peak MPB mortality derived from the Forest Health and Monitoring Aerial Surveys (1 year, 2 3 years, 4 5 years, and 5+ years), elevation (elevation quartiles), and aspect (north, south east, west, and flat), for a total of 80 different strata. Plot locations were restricted to public lands and areas classified as Rocky Mountain Lodgepole

PAGE 42

32 Pine Forest, Southern Rocky Mountain Ponderosa Pine Woodland, Rocky Mountain Subalpine Dry Mesic Sp ruce Fir Forest and Woodland, Rocky Mountain Subalpine Mesic Wet Spruce Fir Forest and Woodland, Southern Rocky Mountain Dry Mesic Montane Mixed Conifer Forest and Woodland, or Southern Rocky Mountain Mesic Montane Mixed Conifer Forest and Woodland in the LANDFIRE existing vegetation type layer ( Zhu et al. 2006 Rollins 2009 ) This sampling scheme was selected because it ensured that the field data captured the range of variability in biophysical gradients and MPB mortality present in the study area. 2 3 plots were placed in each stratum randomly, but revised plot locations based on accessibility; plots in potentially dangerous and inaccessible locations were manually moved within strata. Ultimately, data was collected at 119 plot locations. A field crew measured trees, seedling and saplings as well as surface and canopy fuel loads using the Fire Effects Monitoring and Inventory Protocol: FIREMON ( Lutes et al. 2006 ) Each plot had a fixed radius of 8 meters. Trees overstory species, understory species and soil type at plot center. The tree data included a bearing distance measurement from plot center, as well as DBH, total height (measured using a Haglf Vertex Laser), status (live or dead), number of years dead, and cause of mortality. Saplings were defined as any woody vegetation

PAGE 43

33 1.4 m Seedlings and saplings were counted within a 3.6 m radius subplot and classified according to species and diameter (saplings) or height (seedlings). Down woo dy debris were measured along three transects between 5 and 25m from plot center. Data on tree and herbaceous cover were also collected along each transect at 15 m and 25 m from plot center. 4.4 Modeling methods I initialized the Forest Vegetation Simulato r (FVS) model ( Stage 1973 Dixon 2003 ) Central Rockies Variant, with plot data from the 119 plots to simulate potential changes in vegetation composition and carbon storage as a result of the MPB outbreak. FVS is a growth and yield model, created and used by the USDA Forest Service to help man age forests across the United States. FVS is aspatial and simulates individual tree growth in stands using multiple forestry growth, diameter and height algorithms. FVS estimates changes in carbon storage over time with carbon pools split into aboveground total live, belowground live, belowground dead, standing dead, downed dead wood, forest floor and herbaceous cover. First, FVS estimates growth of trees for each cycle by diameter and height. Then, it estimates mortality based on the individual tree varia bles like diameter, as well as on stand variables like basal area or stand density index. Crown ratio (provided as part of the initialization dataset collected in the field), crown competition factor and total basal area, and stand density index calculatio ns in FVS help the model

PAGE 44

34 project stand trajectories over time. Mortality is primarily projected using stand density index (SDI) measurements per plot in an effort to model stand densities realistically where they do not exceed natural stand densities seen in the Central Rockies, however, background tree mortality was suppressed so as to project the trajectories of species and carbon immediately following a disturbance and not introduce new or multiple independent variables. Each growth cycle for the scenari os lasted 10 years and the output tree list from FVS reads stand and individual tree level projections for each growth cycle. FVS was chosen over other models for several reasons. It incorporates forest growth at the tree and stand level, as compared to o ther models that operate over landscape scales, such as LANDIS or BIOME BGC ( White et al. 2000 Mladenoff 2004 ) FVS was designed to track carbon storage pools, as well as forest biological variables, like seedling and sapling growth and tree death ( Crookston and Dixon 2005 ) It also incorporates many kinds of disturbances, and has been tested and validated specifically for the Rocky Mountain region in the Central Rockies variant that I used i n this research ( Dixon 2003 ) It focuses on above ground carbon storage and biomass, instead of being focused specifically on belowground processes, such as CENTURY ( Parton 1996 ) 4.5 Model Scenarios Field measurements collected in 2010 were used to quantify the initial impacts of the MPB outbreak. Then, future carbon stocks were simulated for 200

PAGE 45

35 years. Three simulations were initialized in FVS. For the first simulation, referred 119 plots exactly as they were collected where an average 72% of basal area was classif outbreak conditions. In the control scenario, the input tree list was altered to reclassify all trees that had recently suffered MPB mortality to live status. For the third simulation, known as of basal area was killed. This matched the average area killed by mountain pine beetle so as to compare carbon storage and succession trajectories between the disturbances. The simulated fire maintained default parameters for the study area in climatic variables. Each simulation was run forward to 2210. using seedlings and saplings present in plots in 2010. Regeneration was modeled conservatively where future disturbances, factors behind regeneration (such as wind or seed dispersal), soil processes and decomposition were not taken into account. It is difficult to model these factors. Regeneration is difficult to predict because the impacts of light, moisture, climate change, soil changes and overall disturbance have differing impacts on regeneration. Soil is difficult to model beca use it is highly variable across spatial scales, especially at a landscape level. Also, decomposition rates are also difficult to predict. FVS simulations for both

PAGE 46

36 scenarios produced annual data on the growth and attributes of individual trees, including n ew seedlings added each cycle, as well as plot level characteristics such as trees per hectare (TPH) and basal area (BA) per species, and carbon storage and fluxes among carbon pools. Simulation cycles were set to 10 year growth cycles and individual trees as well as plot level summaries were given as output for each growth cycle. This methodology to couple field data with the FVS model to track carbon storage and fluxes was used in Pfeifer et al. (2010), where FVS was initialized for 12 plots in an Idaho lodgepole forest after a MPB outbreak had occurred. They did not simulate fire and their plots consisted of slightly different composition, nor did they track stand composition trajectories. Collins et al. (2011) modeled species succession trajectories over time in the Southern Rocky Mountains using advanced regeneration after a MPB disturbance in lodgepole stands using gridded field data and FVS, but did not include an assessment for stand carbon.

PAGE 47

37 4.6 Analysis Methods To compare near and long term impacts of MPB outbreak on carbon storage and species composition, I first compared the differences in the vegetation conditions used to initialize FVS. This comparison also provided the baseline measurements of carbon storage and species composition to evaluate simulated potential changes against. Specifically, I compared total stand carbon overall as well as differences in standing live, downed dead and standing dead carbon pools between the three scenarios using Welch 2 sided t tests assuming unequal variance. To address the research questions investigating how carbon storage c hanged over the years following a MPB outbreak, the MPB disturbance trajectory was compared to the control scenario. Then, to investigate differences in how fire disturbances compare to MPB disturbances, a simulated fire scenario was compared to the contro l scenario as well. The same approach was used to compare the baseline initial and simulated potential changes in TPH and BA by species for each scenario over time to address how species composition changed as a result of the MPB outbreak or fire. Plots w ere further categorized for analysis into dominant overstory and understory combinations, modified from the method used to categorize plot types in Diskin et al. (2011), where plots were split up by 5 different forest types in order to track species trajec tories based on the advanced regeneration present

PAGE 48

38 before a MPB epidemic. For this research, I categorized plots based on the overall dominant overstory and understory as well, however, there were only three main types of plots. Of the 119 plots, 103 were d ominated by an overstory of lodgepole pine. The other 16 plots were dominated by subalpine fir or Engelmann spruce. The plots could further be categorized by the understory level, or advanced regeneration, of trees, where there was a split between those pl ots that were predominately subalpine fir in the understory, and those that were predominately lodgepole pine in the understory. The majority of plots could be split into three type categories, locations illustrated in figure 4.2 and examples of each plot type in figure 4.3. The first type of plot had a stand structure with an overstory dominated by lodgepole pine, and an understory consisting of primarily subalpine fir; there were 17 plots in this category. 79 plots could be characterized by the second typ e, which had both an overstory and understory dominated by lodgepole pine. The third type of plot was dominated in the overstory canopy and understory by subalpine fir; there were 7 plots in this category. The rest of the plots contained aspen, Engelmann s pattern to the aspect, slope or elevation when compared to vegetation composition. This was possibly due to our stratified sampling design. The plots dominated in the overstory and/or understory by subalpine fir were in very specific soil texture types, where subalpine fir plots grew in sandy clay loam, clay loam and fine sandy

PAGE 49

39 loam. The lodgepole plots were in these soil textures, as well as a broad spectrum of other soil texture types.

PAGE 50

40 Figu re 4.2. Field Plots by Dominant Composition Type. Map shows study area with spatial distribution of plots in each of the three plot species type categories.

PAGE 51

41 Figure 4.3 Plot Types. Photos illustrate an example of each category of plot type.

PAGE 52

42 V. RESULTS 5.1 2010 Conditions Of 6257 trees and advanced regeneration surveyed in 119 plots, 86% of trees per hectare were lodgepole pine representing 85% of the total tree basal area across plots; 74% of the total number of trees inventoried had MPB mortality (72% of total tree basal area) and 12% were healthy lodgepole (13% of total tree basal area). The MPB outbreak in our study area began in 1996, and peak mortality occurred 2006 2008, where the majority died in 2007, as shown in table 5.1. Lodgepole pines kille d by MPB tended to have a larger DBH than live lodgepole pines (mean DBH was 16.2 cm for live lodgepole and 22.6 cm for dead lodgepole p value for 2 sided t test of means assuming unequal variance was <<0.0001). 7% of trees in plots were subalpine fir (6% of total tree basal area) and 6% of trees were healthy subalpine fir (representing 5% of total tree basal area as well) and 1% was sick or dead subalpine fir (1% of basal area). About 4.6% of trees were Engelmann spruce (representing 5.8% of total tre e basal area), with 4.2% of trees being healthy (4.7% of basal area) and 0.4% sick or dying (1.1% basal area). Across all plots, seedlings (measured in trees per hectare (TPH)) were dominated by lodgepole pine, where TPH was 336 averaged across plots. Suba lpine Fir and Engelmann spruce were the next most common species with an average 252 and 261 TPH across plots, respectively. Saplings were dominated by

PAGE 53

43 lodgepole as well, with 359 TPH on average, and subalpine fir representing an average 244 TPH and Engelm ann spruce averaging 218 TPH. Lodgepole was more plentiful in the control scenario saplings, representing 370 TPH. Table 5.1 Surveyed Lodgepole Pine Mortality. Table showing the number of lodgepole pine surveyed in 2010 in each MPB mortality category, wh ere mortality ranged from 2004 2009. MPB Mortality Count of surveyed trees in 2010 field data 1 year ago: full crown of fading needles 53 needles remaining 226 3 years ago: < 50% needles remaining 566 4 years ago: no needles remaining but small and large twigs present 269 5 years ago: only large twigs remaining 37 6+ years ago: both small and large twigs not remaining 14 Unknown 15 Grand Total of Trees 1180 Carbon is calculated in FVS by standard forestry procedures, where amount of carbon is half the biomass measurement ( Avery and Burkhart 2002 ) Dry tons/hectare of biomass (where biomass is a measurement of the amount of living material calculated in FVS by the DBH and height of each tree) multiplied

PAGE 54

44 by .5 for all categories except the litter and duff calculations calculates all aboveground measurements listed for carbon. It is standard to calculate litter and duff carbon by multiplying the dry biomass in these categories by .37 ( Dixon 2003 ) These carbon estimates were part of the FVS output, and we re listed for each 10 year cycle, beginning in 2010. In 2010, the average amount of total carbon per plot was 94.7 Mg/ha, but would have been 107 Mg/Ha if there was no MPB caused mortality (p value of .01). The standing live carbon pool averaged at 21.5 M g/ha, but would have stored 57.2 Mg/ha had the outbreak not occurred, where pre outbreak conditions stored 2.67 times more carbon in the standing live pool (p value << 0.0001). Standing dead pools on average had 28.3 Mg/ha versus the 1.2 Mg/ha they would h ave contained without the MPB disturbance (p value <<0.0001). In 2010, the downed dead pool averaged 19.5 Mg/ha of carbon, where it was not possible to estimate pre outbreak downed dead wood (p value of 0.8796). In August 2011, two plots in the study area actually burned in a fire (plot reexamined in September 2011. The fire resulted in every tree in each plot being transferred to the standing dead or downed dead pool, some even clas sified as Appendix A, where the raw plot data serves as an example of the field data collected, and as information on the fire.

PAGE 55

45 5.2 Impacts of MPB on Carbon Storage After a 200 year simulation, the total stand carbon (which includes the sum of the standing live, belowground live, belowground dead, standing dead, downed dead, herbaceous cover and forest floor measurements), was similar with 271 Mg/ha in the control sce nario, and 267 Mg/ha for the MPB scenario (p value for 2 sided t test of means assuming unequal variance between the two datasets was 0.4). The slight, but statistically insignificant, differences were because of an increase in live biomass when MPB mortal ity was reclassified as live trees in the control scenario. Carbon between the simulations is listed in figures 5.1 5.4. Without additional disturbances added to the simulation throughout time and mortality turned off, total stand carbon (figure 5.1) incre ased steadily in both the MPB and control simulations, but was consistently greater in the control scenario than the MPB scenario. However, by 2040, total stand carbon reached 115 Mg/ha, which was not significantly different (p value of .15) than total car bon stored in the control scenario in 2010, representing pre outbreak conditions (107 Mg/ha). The amount of standing live carbon was greater in the control scenario than the MPB scenario throughout the 200 year simulations (Figure 5.2). Average standing live carbon was 136.1 Mg/ha in the MPB scenario, and was 145.7 Mg/ha in the control scenario by 2210 (with a p value of 0.0009). Standing dead carbon was higher in the MPB scenario, but not significantly when compared to the

PAGE 56

46 control simulation in 2210 (Fi gure 5.3) with 4.3 Mg/ha and 3.7 Mg/ha respectively (p value of 0.13). However, there was a rapid decline in standing dead biomass during the first 30 years of the MPB simulation as dead trees fell and were transferred to the downed dead carbon pool. Conse quently, the downed dead carbon pool increased dramatically during the first 30 years of the MPB scenario (Figure 5.4). For the remaining years of the simulation, there were negligible differences in standing dead carbon between the two scenarios (p value of 0 .15). Carbon in the downed dead pool increased steadily over time for both simulations, but was generally greater in the MPB scenario, with 58 Mg/ha in the control scenario and 63 Mg/ha in the MPB scenario. However, after 120 years of simulation the d ifferences between the two scenarios were less pronounced. 5.3 Impacts of Simulated Fire Disturbance on Carbon Storage In order to compare the impacts of MPB disturbances to fire disturbance on carbon and species composition trajectories, a fire was simul ated to burn 72% of basal area (same as the average basal area affected by MPB disturbance in the Carbon storage and stand structure trajectories were compared between th e fire scenario and the control pre disturbance scenario, and subsequently to the MPB disturbance scenario. The fire scenario showed the lowest amount of carbon stored in all four pools of carbon, which is also shown in figure 5.1 5.4. In 2010, after the f ire was simulated (in late summer for default climatic conditions for the central

PAGE 57

47 Rockies variant), the standing live carbon pool stored 79.4 Mg/ha (p value of << 0.0001 between the fire scenario and the control scenario). The standing dead pool stored 0.5 2 Mg/ha (p value of << 0.0001), downed dead stored 19.91 Mg/ ha (p value of << 0.0001) and the total stand carbon pool stored 142.6 Mg/ha (p value of << 0.0001). In 2010, the simulated fire reduced standing live carbon (figure 5.2) to about 19 Mg/ha from 57.2 Mg/ha in the control (p value of << 0.0001). Standing live carbon returned to pre fire levels also by about 2070, to about 57 Mg/ha. Standing dead carbon (figure 5.3), however, is projected to increase from 1.2 Mg/ha in the control to about 33 Mg/ha i n the fire scenario (p value of << 0.0001). The standing dead carbon pool in the fire scenario stored almost 3 Mg/ha, and the downed dead carbon pool stored 42 Mg/ha (p value of << 0.0001). It was not possible to project pre fire downed dead carbon for 201 0 in the fire scenario because the control scenario tree input data was burned and it was not possible to validly predict pre disturbance downed dead wood. After the fire, however, 10.9 Mg/ha remained in the downed dead carbon pool in 2010. The average amo unt of carbon released from the fire from all carbon pools was 35.7 Mg/ ha in 2010. The simulated fire results (figure 5.1) for 2210 showed total stand carbon to be 228.43 Mg/Ha (p value of << 0.0001), which was slightly lower than the total stand carbon for either the MPB or the control scenarios. Total stand carbon rebounded from around 81 Mg/ha to about 107 Mg/ha (pre disturbance total stand

PAGE 58

48 carbon) by 2070 (p value of << 0.0001) as well, representing a return to pre fire conditions within 60 years fol lowing fire.

PAGE 59

49 Figure 5.1 Total Stand Carbon. Scatter plot shows the trajectory of total stand carbon with confidence intervals, which is a compilation of all carbon pools, for each of the three scenarios: MPB, Control and Fire, where the table lists starting 2010 carbon storage and ending 2210 storage for each trajectory.

PAGE 60

50 Figure 5.2 Standing Live Carbon. Scatter plot shows the trajectory of carbon in the standing live carbon pool with confidence intervals for each of the three scenarios: MPB, Control and Fire, where the table lists starting 2010 carbon storage and ending 2210 storage for each trajectory.

PAGE 61

51 Figure 5.3 Standing Dead Carbon. Scatter plot shows the trajectory of carbon in the standing dead carbon pool with confidence intervals for each of the three scenarios: MPB, Control and Fire, where the table lists starting 2010 carbon storage and ending 221 0 storage for each trajectory.

PAGE 62

52 Figure 5.4 Downed Dead Carbon. Scatter plot shows the trajectory of carbon in the downed dead carbon pool with confidence intervals for each of the three scenarios: MPB, Control and Fire, where the table lists starting 2010 carbon storage and ending 2210 storage for each trajectory.

PAGE 63

53 5.4 Impacts of MPB on Stand Structure Four primary species were present in stands in 2010: lodgepole pine, Engelmann spruce, subalpine fir and trembling aspen. There were a few other species, such as white fir ( Abies concolor ) and Douglas fir ( Pseudots uga menziesii ), present in limited quantities in plots, which were grouped and In 2210, as shown in figure 5.5a, the control scenario showed a dominance of lodgepole pine in the overstory canopy, where lodgepole accounted for 58% of basal area in 2210, but subalpine fir only accounted for 19% of basal area. Engelmann spruce and aspen were subdominant species in both scenarios; however, the MPB scenario had greater proportion of aspen and spruce than in the control scenario in 2210. The MPB scenario (figure 5.5b) shows a much more even mixture of tree species in the overstory vegetation layer, where there was about 30% of basal area of each subalpine fir and lodgepole, 12.6% was Engelmann spruce, and 17.7% was aspen. For an example plot in the study area, figure 5.6 shows the FVS 2010 and 2210 results for the MPB and control scenario, where there has been a shift in the species composition trajectory following MPB mortality.

PAGE 64

54 Figure 5.5 ( a c ) Average Species Composition by Percent of Basal Area. Percent of basal area (in m 2 /Ha) species trajectories for each simulation averaged across all plots where a.) is the percent basal area for each species in the MPB scenario, b.)is the Control scenario and c.) is the Fire scenario

PAGE 65

55 Figure 5.6 Beginning and End Species Trajectory. Output images from FVS showing the differences in stand composition trajectories over time for each scenario for field plots initially dominated by a lodgepole overstory and an understory of subalpine fir the MPB and Control model scenarios.

PAGE 66

56 5.5 Impacts of Simulated Fire Disturbance on Stand Structure In 2010, the fire scenario was primarily lodgepole pine in the overstory, with an average 87.9% of the remaining basal area represented by lodgepole after the simulated fire. The simulated fire stand structure looked somewhat different from the MPB or control scenario simulations in 2210 overall averaged across plots. As shown in figure 5.5c., the live overstory was domin ated by 71.6% lodgepole pine, 13% of the basal area consisted of subalpine fir, 5.1% was aspen and 8.4% was Engelmann spruce, leaving 1.5% of basal area as other species. The fire scenario had the most percentage of basal area represented by lodgepole out of the three scenarios in 2210. In 2010, a real fire burned through two plots. Every tree was killed, and no regeneration was present during the 2011 resurvey. There was some herbaceous cover present along the fuels transects. Much of the standing live c arbon was transferred to the standing or downed dead carbon pools, which changes the stand structure. This transfer of carbon to the ground level pools may have impacts on soil and regeneration processes in the near and long term. The fire I simulated to c onsume 72% of basal area was a conservative estimate, as the real fire consumed almost all of the basal area in the two plots. 5.6 Variation in Plots for Carbon and Stand Structure Trajectories The 119 field plots were split up between primary overstory a nd understory species composition, as described in the methods section, to investigate variation

PAGE 67

57 in carbon and species composition between plot types as opposed to averages of carbon and stand composition trajectories across all plots. The plots were split up based on the species composition observed in plots in 2010, and thus included MPB mortality. Figures 5.7 5.10 show an example of the differences in the four carbon pool trajectories for the three plot types for the MPB scenario. These stand types were investigated for the MPB scenario to compare the individual plot species compositions and carbon trajectories between scenarios by primary overstory and understory species composition. Only the MPB scenario was split up into plot types because the MPB scen ario represented reality, where stand composition types were qualified in the field after MPB disturbance, and thus would not reflect valid species composition necessarily for the other two simulated scenarios. 5.6.1 Variation in Carbon Trajectories betwee n Plot Types There were differences in carbon storage in some of the plots within the MPB scenario due to changes in stand structure, which is shown in figure 5.7 5.11. Of the three main types of plots, the third type of plot, which consisted of an overstory and understory dominated by subalpine fir, stored more carbon than the average across all plots in the standing live carbon pool (where the calculated confidence interval for type 3 plots did not intersect the other values for 2010, validating a statistically significant difference) The average for these 7 plots in 2210 was 136.9 Mg/h a standing live carbon The first type of plot, with a

PAGE 68

58 lodgepole dominated overstory and a subalpine fir dominated understory, the standing live pool stored less tha n the average stored across all plots at 130.5 Mg/ha. The second type of plot, where lodgepole dominated both the overstory canopy and the understory, there was a majority of 79 plots in this category, was about the same as the average across all plots.

PAGE 69

59 Figure 5.7 Standing Live Carbon by Plot Type Shows v ariability in carbon stored within the standing live carbon pool between the different plot types in the MPB scenario.

PAGE 70

60 Figure 5.8 Standing Dead Carbon by Plot Type Variability within the stand ing dead carbon pool between the different plot types in the MPB scenario.

PAGE 71

61 Figure 5.9 Downed Dead Carbon by Plot Type Variability in carbon stored within the downed dead carbon pool between the different plot types in the MPB scenario.

PAGE 72

62 Figure 5.10 Total Stand Carbon by Plot Type Variability in total stand carbon storage between the different plot types in the MPB scenario.

PAGE 73

63 5.6.2 Variation in Stand Composition Trajectories between Plot Types For the MPB scenario (as shown in figure 5.11 a c), when the plots are split into the three species type categories, the stand trajectories for species varied somewhat. Type 1 plots initially were dominated in the overstory vegetation by lodgepole (45%) an d subalpine fir (31.4%), and in 2210, are dominated by 62.3% of the basal area by subalpine fir. Engelmann spruce occupy 19.5% of the basal area in type 1 plots in 2210, and 11.8% is lodgepole on average. Type 2 plots were initially dominated by lodgepole in 2010 accounting for 65% of basal area, which was reduced to 43.8% by 2210. This was offset by an increase in subalpine fir, which began at 6.3% in 2010 and increased to representing 19.2% in 2210, as well as aspen, which increased from less than 1% in 2 010 to almost 23% in 2210. Type 3 plots were initially dominated by subalpine fir, which only increased from 70% to 79% of basal area between 2010 and 2210. These plots experienced a decrease in lodgepole from 11.9% to 8.5% basal area from 2010 to 2210. En gelmann spruce was reduced over time from 2010 to 2210 by representing 18.2% to 11.8% of basal area.

PAGE 74

64 Figure 5.11 (a. c) Species Trajectories between Plot Types Species trajectory projections by type 1, 2 or 3 species plots over time for the MPB scen ario

PAGE 75

65 VI DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS This research stands to track and compare carbon and species succession trajectories without disturbance, after a real MPB disturbance, and after a simulated fire, in order to better quantify the imp acts of large disturbances on carbon and species, in order to provide better information to forest managers I used forest inventory data collected in field plots to initialize the FVS model to quantify initial and future carbon and species composition trajectory projections. This data provided information on how differing disturbances, which may increase with global climate change, may change carbon and species trajectories and provides insight into how to manage forests better. The MPB outbreak in the Southern Rocky Mountains has altered the trajectory of carbon storage and species composition in Southern Roc ky Mountain forests. The surveyed field plots were dominated by lodgepole, where most of the larger diameter lodgepole were killed by MPB (unlike fire disturbances, where size is not typically selected). As disturbances are predicted to increase in frequen cy and severity in a warming climate, the intricate relationship between carbon storage and species composition, and how trajectories change due to disturbance, will become ever important. These results showed carbon storage and species composition are alt ered after insect disturbances, and are different from the impacts on stand structure and carbon storage from fire disturbances. Further, carbon storage may differ between species types present in stands.

PAGE 76

66 6.1 Impacts of MPB on Carbon Storage Stand carbon storage trajectories were altered short term as a result of the MPB disturbance. The immediate impacts of MPB outbreaks on carbon storage are substantial, causing a rapid transfer of 27.1 Mg/Ha, or 28.6% of the total stand carbon from the stan ding live pool to the standing dead pool and subsequently to the downed dead pool. However, these pools recover at varying rates following MPB outbreaks. Klutsch et al. (2009) found a 69% reduction in overall basal area, and transfers from the standing liv e carbon pool to the standing dead and downed dead/litter pools for northern Colorado lodgepole forests experiencing MPB epidemic conditions. This differed from the field plots collected for my research, where an average 72% of basal area was lost. I foun d that the impacts on the standing live pool were relatively short lived and that standing live carbon recovered to pre outbreak conditions by 2060 (roughly 60 Mg/ha). The total stand stored carbon returns to pre outbreak conditions by 2040. After about 90 years, differences in carbon storage in all the carbon pools were negligible between the two scenarios. In contrast to this study, Pfeifer et al. (2010) examined stand carbon changes in a northern Idaho forest as a result to a beetle outbreak. They observ ed a rebound in carbon to pre outbreak conditions in 25 years. Mortality of basal area was not as high for their region and the forest had slightly different stand characteristics, such as the presence of whitebark pine ( Pinus albicaulis ) and more

PAGE 77

67 Douglas fir ( Pseudotsuga menziesii var. menziesii ). Pfeifer et al. validates the notion that forest recovery of carbon storage to pre outbreak conditions is not only possible, but occurs over a relatively short time span. The trajectories of the field plots by the dominant species present in the overstory and understory varied somewhat in carbon content and species composition. This study only split plots based on their dominant overstory and understory vegetation, and not any other factors, such as stand densit ies or scenario, the 7 plots dominated by subalpine fir in both the overstory and understory, or the type 3 plots, stored above the average amount of carbon stored in all plots. This may indicate that subalpine fir has a greater carbon storage potential in this ecosystem. Initially, type 1 plots had higher standing dead carbon storage, indicating potentially higher percent basal area impacted by MPB. Type 3 plots had the lowest storage in the standing dead carbon pool, which is intuitive in that they primarily consist of non host species for MPB. Type 3 plots stored more carbon on average in the downed dead pool than the other plots. 6.2 Impacts of Simulated Fire Disturbance on Car bon Storage Both MPB and fire disturbances altered the trajectories of carbon storage, however a fire that burns the same area MPB affects seems to impact carbon pools more severely and for a longer time period. For all 200 years of the simulation, the fir e scenario stored less carbon in the standing live, standing dead and total stand

PAGE 78

68 carbon pools compared to the other scenarios. This could have been due to the fire consuming surface fuels in the downed dead wood biomass pools and smaller seedlings and sap lings that remained in place during the MPB simulation. The fire scenario showed higher carbon storage in the standing dead pool than the other impacts selective species or size c lasses. Downed dead carbon was significantly lower in the fire scenario by 2210. Initially, downed dead carbon was projected in 2020 to be similar to that stored in the downed dead pool in the MPB scenario. By 2030, however, downed dead carbon stored was l ower in the fire scenario than in the MPB scenario. It seems that more carbon, likely from the standing dead pool, is transferred to the downed dead pool over a longer time period in MPB disturbances and more of the carbon from the standing dead pool is tr ansferred over time to the other pools in fire disturbances over time. The total carbon storage recovered to pre fire levels more slowly than the MPB scenario recovered to pre MPB storage, where carbon returned to pre fire carbon storage in the standing l ive pool and total stand carbon by 2080 which was twenty years later than recovery of the MPB disturbance trajectory. 6.3 Impacts of MPB on Stand Structure MPB disturbances impact stand structure differently than other major disturbances, such as fire or wind because only select trees, specifically older lodgepole with a greater DBH, are taken out of the live biomass pool. Insect

PAGE 79

69 disturbances allow seed dispersal, light and water to interact differently within the forest compared to an even aged, undisturbed forest. This allows for differences in stand structure and species composition between disturbed and undisturbed forests of the same type. Changes in stand composition also can subsequently change the rate at which carbon is sequestered and stored, as diff erent species sequester carbon at different rates and ages of development. Initially, an average 72% of basal area was killed by mountain pine beetle in this study area. This corresponds to Klutsch et al. (2009) where the live basal area declined 71% in th eir beetle infested study plots in the Colorado Arapaho National forest. Succession was modeled conservatively by projecting the current trajectories with only advanced regeneration, out 200 years, but there was still a shift in species composition over t ime in the MPB scenario as compared to the control scenario. If the MPB outbreak had not occurred, species composition would be dominated by primarily lodgepole pine. The MPB scenario however, shows a species composition shift towards a greater percentage of the basal area being subalpine fir. A mixed vegetation composition in the future could be beneficial for ecological sustainability as it encourages biodiversity, which will make stands more resilient. This observation is supported through the study car ried out by Collins et al. (2011) that showed a shift towards a subalpine fir dominated canopy in Colorado in untreated stands affected by MPB through the use of vegetation modeling.

PAGE 80

70 Collins et al (2011) found that tree regeneration in MPB impacted lodgepo le stands were more evenly distributed with subalpine fir and lodgepole pine as well. This validates a landscape level increase in basal area represented by subalpine fir after a MPB outbreak, for a more mixed stand composition. Klutsch et al. (2009) also concluded that stands impacted by the mountain pine beetle epidemic will result in an uneven aged stand structure consisting of older, existing lodgepole and younger subalpine fir and Engelmann spruce. Their work along with this study suggest that future s tand composition depends heavily on advance regeneration present at the time of disturbance. 6.4 Impacts of Simulated Fire Disturbance on Stand Structure Based on the FVS model results, fire seems to encourage a stand structure trajectory that becomes dom inated by primarily lodgepole pine, which is different from the trajectories of species composition in the absence of disturbance and after MPB disturbance. These findings correspond with known successional trends where pure lodgepole stands typically reg enerate after stand replacing fires in forests that were dominated by lodgepole ( Lotan et al. 1985 ) Even in places w here forests are dominated by spruce and fir, lodgepole can increase following fires. In the absence of fire though, subalpine fir and Englemann spruce increase in dominance where there have been fewer fires ( Romme and Knight 1981 ) In the fire scenario, the dominant species in plots makes a difference in the stand composition trajectories over time.

PAGE 81

71 6.5 Conclusions and Recommendations 6.5.1 Conclusions Changes to carbon storage in forests affected by MPB were relatively short lived under scenarios that did not incorporate further disturbances. Standing live carbon rebounded within 50 years in most stands. However, species composition changed over a longer time period, and there were substantial differences in forest stand structure a nd species composition that persisted after 200 years of simulation MPB disturbances impact stand structure differently than other major disturbances, such as fire. MPB targets older, large diameter lodgepole pines, leaving behind advanced regeneration t hat grow rapidly in the years after disturbance. This resulted in standing live carbon stocks recovering quickly after MPB T his was accompanied by changes in species composition trajectories with a slight shift toward more aspen, fir, and spruce. Succes sion modeling was conservatively, however there was still a shift in species composition over time in the MPB simulation, compared to the control simulation. The findings of this study corroborate those found by Collins et al. (2011), who showed a shift towards a subalpine fir dominated canopy in Colorado in untreated stands affected by MPB through the use of vegetation modeling. Lodgepole pine forests affected by MPB appear to enter a self correcting mode, where a more diverse forest develops after MPB d isturbance, making it less susceptible to future MPB disturbances. Carbon storage differs within each scenario as well, based on

PAGE 82

72 dominant overstory and understory vegetation, where plots dominated in the over and understory by subalpine fir trajectories s tore more carbon than other plots. In lodgepole pine forests, fire affects the larger trees favored by beetles, but also causes mortality in advanced regeneration Carbon loss during a fire is a function of fuel loads and fire severity and recovery of ca rbon after a fire is a function of seed availability and regeneration rates. The results from this study and those found in the literature, suggest that burned lodgepole pine forests will regenerate as lodgepole, and the rate of carbon recovery is somewha t slower after fire than after the MPB outbreak. F ires perpetuat ing lodgepole pine in stands could make the ecosystem more vulnerable to disturbances in the future. These changes in both carbon and species composition may have additional effects on carbon cycling by altering primary productivity, soil biogeochemistry, and regeneration over time; processes that warrant further investigation. 6.5.2 Recommendations and Management Implications This research suggests that disturbances change forest composition and carbon trajectories over time, and that different disturbances alter these trajectories differently. Also, this research concluded that there are differences in the trajectories between individual stands within forest types that are impacted by distur bance. The local scale impacts of MPB outbreaks are substantial and could have implications for how forests offset greenhouse gas emissions and global climate change at broader scales. It would be beneficial for stakeholders managing

PAGE 83

73 for carbon storage and sequestration in forests to account for the varying carbon impacts and the length of recovery time in carbon pools resulting from the various disturbances that affect forests. This is especially important as disturbances are projected to increase in a war ming climate. Managers should also account for the differences that may occur in between stands across a landscape scale, and that a uniform management design may not be optimal across an entire landscape, as there are variances between stands and species composition types.

PAGE 84

74 REFERENCES Amman, G. D. 1977. The role of mountain pine beetle in lodgepole ecosystems: Impact on succession Pages 3 18 in W. J. Mattson, editor. The Role of Arthropods in Forest Ecosystem. Springer Verlag, New York. Aoki, C. F., W. H. Romme, and M. E. Rocca. 2011. Lodgepole Pine Seed Germination Following Tree Death from Mountain Pine Beetle Attack in Colorado, USA. The Amer ican Midland Naturalist 165 :446 451. Avery, T. E. and H. E. Burkhart. 2002. Forest Measurements. 5 edition. McGraw Hill, NY, NY. Baker, W. L. and T. T. Veblen. 1990. Spruce Beetles and Fires in the Nineteenth Century Subalpine Forests of Western Colorado, U.S.A. Arctic and Alpine Research 22 :65 80. Barrow, E. and G. Yu. 2005. Climate Scenarios for Alberta. Prairie Adaptation Research Collaborative (PARC). Bazzaz, F. A. 1996. Plants in Changing Environments. Cambridge University Press, U.K. Bentz, B. J., J. Rgnire, C. J. Fettig, E. M. Hansen, J. L. Hayes, J. A. Hicke, R. G. Kelsey, J. F. Negrn, and S. J. Seybold. 2010. Climate Change and Bark

PAGE 85

75 Beetles of the Western United States and Canada: Direct and Indirect Effects. BioScience 60 :602 613. Bessie, W. C. and E. A. Johnson. 1995. The relative importance of fuels and weather on fire behavior in subalpine forests. Ecology 76 :747 762. Birdsey, R., K. Pregitzer, and A. Lucier. 2006. Forest Carbon Management in the United States. Journal of Environment Quality 3 5 :1461. Birdsey, R. A. 1992. Carbon storage and accumulation in United States forest ecosystems. Gen. Tech. Rep. WO 59, Washington D.C.: U.S. Department of Agriculture, Forest Service, Washington Office. Bradford, J. B., R. A. Birdsey, L. A. Joyce, and M. G. Ryan. 2008. Tree age, disturbance history, and carbon stocks and fluxes in subalpine Rocky Mountain forests. Global Change Biology 14 :2882 2897. Breshears, D. and C. D. Allen. 2002. The importance of rapid, disturbance induced losses in carbon managemen t and sequestration. Global Ecology & Biogeography 11 :1 5. Brown, J., editor. 1975. Fire cycles and community dynamics in lodgepole pine forests. Carroll, A. L., S. W. Taylor, J. Regniere, and L. Safranyik. 2003. Effect of climate change on range expansion by the mountain pine beetle in British Columbia. Page Paper 195.

PAGE 86

76 CCSP. 2007. The First State of the Carbon Cycle Report (SOCCR): The North American Carbon Budget and Implications for the Global Carbon Cycle. National Oceanic and Atmospheric Administration National Climatic Data Center, Asheville, NC. Chapin III, F. S., P. A. Matson, and H. Mooney. 2002. Principles of terrestrial ecosystem ecology. Springer Verlag, New York. Claveau, Y., C. Messier, P. G. Comeau, and K. D. Coates. 2002. Growth and crown mo rphological responses of boreal conifer seedlings and saplings with contrasting shade tolerance to a gradient of light and height. Canadian Journal of Forest Research 32 :458 468. Collins, B. J., C. C. Rhoades, R. M. Hubbard, and M. A. Battaglia. 2011. Tree regeneration and future stand development after bark beetle infestation and harvesting in Colorado lodgepole pine stands. Forest Ecology and Management 261 :2168 2175. Comeau, P. and J. P. Kimmins. 1989. Above and below ground biomass and production of lo dgepole pine on sites with differing soil moisture regimes. Canadian Journal of Forest Research 19 :447 454. Crookston, N. and G. Dixon. 2005. The forest vegetation simulator: A review of its structure, content, and applications. Computers and Electronics i n Agriculture 49 :60 80.

PAGE 87

77 Dale, V. H., L. A. Joyce, S. McNulty, R. P. Neilson, M. P. Ayres, M. D. Flannigan, P. J. Hanson, L. C. Irland, A. E. Lugo, C. J. Peterson, D. Simberloff, F. J. Swanson, B. J. Stocks, and B. Michael Wotton. 2001. Climate Change and F orest Disturbances. BioScience 51 :723 734. Diskin, M., M. E. Rocca, K. N. Nelson, C. F. Aoki, and W. H. Romme. 2011. Forest developmental trajectories in mountain pine beetle disturbed forests of Rocky Mountain National Park, Colorado. Canadian Journal of Forest Research 41 :782 792. Dixon, G. E. 20 Simulator., USDA Forest Service, Forest Management Service Center, Fort Collins, CO. Fahey, T. J. and D. H. Knight. 1986. Lodgepole Pine Ecosystems. BioScience 36 :610 617. Flannigan, M. D., B. J. S tocks, and B. M. Wotton. 2000. Climate change and forest fires. The Science of the Total Environment 262 :221 229. Franklin, J. F., T. A. Spies, R. V. Pelt, A. B. Carey, D. A. Thornburgh, D. R. Berg, D. B. Lindenmayer, M. E. Harmon, W. S. Keeton, D. C. Shaw K. Bible, and J. Chen. 2002. Disturbances and structural development of natural forest ecosystems with silvicultural implications, using Douglas fir forests as an example. Forest Ecology and Management 155 :399 423.

PAGE 88

78 Friedlingstein, P., L. Bopp, P. Ciais, J. Dufresne, Louis, L. Fairhead, H. LeTreut, P. Monfray, and J. Orr. 2001. Positive feedback between future climate change and the carbon cycle. Geophys. Res. Lett. 28 :1543 1546. Glenn Lewin, D. C., R. K. Peet, and T. T. Veblen, editors. 1992. Plant Succes sion: Theory and Prediction. Chapman and Hall, London, UK. Goward, S. N., J. G. Masek, W. Cohen, G. Moisen, G. J. Collatz, S. Healey, R. A. Houghton, C. Huang, R. Kennedy, B. Law, S. Powell, D. Turner, and M. A. Wulder. 2008. Forest Disturbance and North A merican Carbon Flux. EOS Transactions American Geophysical Union 89 :105. Hamann, A. and T. Wang. 2006. Potential effects of climate change on ecosystem and tree species distribution in British Columbia Ecology 87 :2773 2786. Heal, G. 2000. Valuing Ecosystem Services. Ecosystems 3 :24 30. Heath, L. S. and J. E. Smith. 2004. 2003 Report on Sustainable Forests: Contribution of forest ecosystems fo the total global carbon budget, including absorption and release of carbon USDA Forest Service. Hurteau, M. and M. N orth. 2008. Fuel treatment effects on tree based forest carbon storage and emissions under modeled wildfire scenarios. Frontiers in Ecology and the Environment 7 :409 414. Hurteau, M. and M. North. 2009. Fuel treatment effects on tree based forest carbon st orage and emissions under modeled wildfire scenarios. Frontiers in Ecology and the Environment 7 :409 414.

PAGE 89

79 Hurtt, G. C., S. W. Pacala, P. R. Moorcroft, J. Caspersen, E. Shevliakova, R. A. Houghton, and B. Moore. 2002. Projecting the future of the U.S. carbo n sink. Proceedings of the National Academy of Sciences 99 :1389 1394. IPCC. 2007. Intergovernmental Panel on Climate Change Assessment Report. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Joyce, L. A. and R. Birdsey. 2000. T he impact of climate change on America's forests: a technical document supporting the 2000 USDA Forest Service RPA Assessment. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fort Collins, CO:. Kashian, D. M., D. B. Tinke r, M. G. Turner, and F. L. Scarpace. 2004. Spatial heterogeneity of lodgepole pine sapling densities following the 1988 fires in Yellowstone National Park, Wyoming, U.S.A. Canadian Journal of Forest Research:2263 2276. Kashian, D. M., W.H. Romme, D.B. T inker, M.G. Turner, and M.G. Ryan. 2006. Carbon storage on landscapes with stand replacing fires. BioScience 56 : 598 606. Kimball, B. A., J. R. Mauney, F. S. Nakayama, and S. B. Idso. 1993. Effects of increasing atmospheric CO 2 on vegetation. Plant Ecology 104 105 :65 75. Klutsch, J. G., J. F. Negrn, S. L. Costello, C. C. Rhoades, D. R. West, J. Popp, and R. Caissie. 2009. Stand characteristics and downed woody debris

PAGE 90

80 accumulations associated with a mountain pine beetle (Dendroctonus ponderosae Hopkins) out break in Colorado. Forest Ecology and Management 258 :641 649. Kueppers, L. and J. Harte. 2005. Subalpine Forest Carbon Cycling: Short and Long Term Influence of Climate and Species. Ecological Applications 15 :1984 1999. Kurz, W. A., C. C. Dymond, G. Stins on, G. J. Rampley, E. T. Neilson, A. L. Carroll, T. Ebata, and L. Safranyik. 2008. Mountain pine beetle and forest carbon feedback to climate change. Nature 452 :987 990. Lal, R. 2004. Soil Carbon Sequestration Impacts on Global Climate Change and Food Secu rity. Science:1623 1627. Lal, R., J. M. Kimble, R. F. Follett, and B. A. Stewart, editors. 2000. Assessment Methods for Soil Carbon. CRC/Lewis Publishers, Boca Raton, FL. Lamlom, S. H. and R. A. Savidge. 2003. A reassessment of carbon content in wood: vari ation within and between 41 North American species. Biomass and Bioenergy 25 :381 388. Lotan, J. 1976. Cone Serotiny Fire Relationships in Lodgepole Pine. in Tall Timbers Fire Ecology Conference Proceedings, Tall Timbers Research Center, Tallahassee, FL. Lotan, J., J. Brown, and L. Neuenschwander., editors. 1985. Role of fire in lodgepole pine forests, Washington State University, Pullman.

PAGE 91

81 Lotan, J. E. and W. B. Critchfield. 2004. Lodgepole Pine. USFS. Lotan, J. E. and D. A. Perry. 1983. Ecology and regeneration of lodgepole pine. Washington D.C. Lutes, D. C., R. E. Keane, J. F. Caratti, C. H. Key, N. C. Benson, S. Sutherland, and L. J. Gangi. 2006. FIREMON: Fire Effects Monitoring and Inventory System. Technical RMRS GTR 164 CD. Marschner, P. and Z. Rengel, editors. 2007. Nutrient Cycling in Terrestrial Ecosystems. Springer Verlag, Berlin, Heidelberg. Mladenoff, D. 2004. LANDIS and forest landscape models. Ecological Modelling 180 :7 19. Monserud, R. A. Y., Yuqing; Huang, Shongming; Tchebakova, Nad ja. 2008. Potential change in lodgepole pine site index and distribution under climatic change in Alberta Canadian Journal of Forest Research 38 :343 352. Monson, R. K., A. A. Turnipseed, J. P. Sparks, P. C. Harley, L. E. Scott Denton, K. Sparks, and T. E. Huxman. 2002. Carbon sequestration in a high elevation, subalpine forest. Global Change Biology 8 :459 478. Newell, R. G. and R. N. Stavins. 2000. Climate Change and Forest Sinks: Factors Affecting the Costs of Carbon Sequestration. Journal of Environmental Economics and Management 40 :211 235.

PAGE 92

82 Pacala, S. W., R. A. Birdsey, S. D. Bridgham, R. T. Conant, K. Davis, B. Hales, R. A. Houghton, J. C. Jenkins, M. Johnston, G.Marland, and K. Paustian. 2007. The North American Carbon Budget Past and Present. In: The F irst State of the Carbon Cycle Report (SOCCR): The North American Carbon Budget and Implications for the Global Carbon Cycle. U.S. Climate Change Science Program and the Subcommittee on Global Change Research. National Oceanic and Atmospheric Administratio n, National Climatic Data Center, Asheville, NC, USA. Parker, A. J. and K. C. Parker. 1983. Comparative successional roles of trembling aspen and lodgepole pine in the Southern Rocky Mountains. Parton, W. J. 1996. The CENTURY model. Pages 283 293 in D. S. Powlson, P. Smith, and J. U. Smith, editors. Evaluation of Soil Organic Matter Models Using Existing Long Term Datasets. Springer Verlag, Heidelberg Pfeifer, E. M., J. A. Hicke, and A. J. H. Meddens. 2011. Observations and modeling of aboveground tree car bon stocks and fluxes following a bark beetle outbreak in the western United States. Global Change Biology 17 :339 350. Prentice, I. C., G.D. Farquhar, M.J.R. Fasham, M.L. Goulden, M. Heimann, V.J. Jaramillo, H.S. Kheshgi, C. Le Qur,, D. W. R. W. R.J. Sch oles, D. Archer, M.R. Ashmore, O. Aumont, D. Baker, M. Battle, M. Bender, L.P. Bopp, P. Bousquet, K. Caldeira,, P. M. C. P. Ciais, W. Cramer, F.

PAGE 93

83 Dentener, I.G. Enting, C.B. Field, P. Friedlingstein, E.A. Holland,, J. I. H. R.A. Houghton, A. Ishida, A.K. Ja in, I.A. Janssens, F. Joos, T. Kaminski, C.D. Keeling,, D. W. K. R.F. Keeling, K.E. Kohfeld,W. Knorr, R. Law, T. Lenton, K. Lindsay, E. Maier Reimer,, R. J. M. A.C. Manning, A.D. McGuire, J.M. Melillo, R. Meyer, M. Mund, J.C. Orr, S. Piper, K. Plattner,, a nd S. S. P.J. Rayner, R. Slater, S. Taguchi, P.P. Tans, H.Q. Tian, M.F. Weirig, T. Whorf, A. Yool. 2001. The Carbon Cycle and Atmospheric Carbon Dioxide. Pages 183 285 in J. T. Houghton, editor. Cambridge University Press, Cambridge. Raffa, K. F., B. H. Au kema, B. J. Bentz, A. L. Carroll, J. A. Hicke, M. G. Turner, and W. H. Romme. 2008. Cross scale Drivers of Natural Disturbances Prone to Anthropogenic Amplification: The Dynamics of Bark Beetle Eruptions. BioScience 58 :501 517. Rapp, V. 2004. Western Fores ts, Fire Risk, and Climate Change. Page 12. Forest Service, U.S.Department of Agriculture, Portland, OR: Pacific Northwest Research Station. Roe, A. L. and G. D. Amman. 1970. Mountain Pine Beetle in Lodgepole Pine Forests. Research Paper INT 71, USDA Fores t Service, Intermountain Research Station. Rollins, M. G. 2009. LANDFIRE a nationally consistent vegetation, wildland fire, and fuel assessment. International Journal of Wildland Fire 18

PAGE 94

84 Romme, W. H. 1982. Fire and Landscape Diversity in Subalpine Forests of Yellowstone National Park. Ecological Monographs 52 :199 221. Romme, W. H. and D. H. Knight. 1981. Fire Frequency and Subalpine Forest Succession Along a Topographic Gradient in Wyoming. Ecology 62 :319 326. Romme, W. H., D. H. Knight, and J. B. Yavitt. 1986. Mountain Pine Beetle Outbreaks in the Rocky Mountains: Regulators of Primary Productivity? The American Naturalist 127 :484 494. Running, S. W. 2008. Ecosystem Disturbance, Carbon, and Climate. Science 321 :652 653. Ryan, M., M. Harmon, R. Birdsey, C. Giardina, L. Heath, R. Houghton, R. Jackson, D. McKinley, J. Morrison, BC Murray, D. Pataki, and K. Skog. 2010. A Synthesis of the Science on Forests and Carbon for U.S. Forests. Ecological Society of America, Issues In Ecology 13 :1 16. Safranyik, L., A. L Carroll, J. Rgnire, D. W. Langor, W. G. Riel, T. L. Shore, B. Peter, B. J. Cooke, V. G. Nealis, and S. W. Taylor. 2010. Potential for Range Expansion of Mountain Pine Beetle into the Boreal Forest of North America. The Canadian Entomologist 142 :415 442 Schmoldt, D. L., David L. Peterson, R. E. Keane, J. M. Lenihan, D. McKenzie, D. R. Weise, and D. V. Sandberg. 1999. Assessing the Effects of Fire

PAGE 95

85 Disturbance on Ecosystems: A Scientific Agenda for Research and Management. USFS, Pacific Northwest Research Station. Sibold, J. S., T. T. Veblen, K. Chipko, L. Lawson, E. Mathis, and J. Scott. 2007. Influences of Secondary Disturbances on Lodgepole Pine Stand Development In Rocky Mountain National Park Ecological Applications 17 :1638 1655. Siegenthaler, U. and J. L. Sarmiento. 1993. Atmospheric carbon dioxide and the ocean. Nature 365 :119 125. Sitch, S., C. Huntingford, N. Gedney, P. E. Levy, M. Lomas, S. L. Piao, R. Betts, P. Ciais, P. Cox, P. Friedlingstein, C. D. Jones, I. C. Prentice, and F. I. Woodward. 2008. Evaluation of the terrestrial carbon cycle, future plant geography and climate car bon cycle feedbacks using five Dynamic Global Vegetation Models (DGVMs). Global Change Biology 14 :2015 2039. Smith, T. M., W. P. Cramer, R. K. Dixon, R. Leemans, R. P. Neilson, and A. M. Solomon. 1993. The Global Terrestrial Carbon Cycle. Water, Air and So il Pollution 70 :19 37. Son, E., a J. J. Kim,b Y.W. Lim,c T.T. Au Yeung,a C.Y.H. Yang,a C. Breuila 2010. Diversity and decay ability of basidiomycetes isolated from lodgepole pines killed by the mountain pine beetle. Canadian Journal of Microbiology 57 :33 4 1.

PAGE 96

86 Stage, A. R. 1973. Prognosis model for stand developement. USDA Forest Service, Ogden, UT Intermountain Forest and Range Experiment Station. Stohlgren, T. J. and R. R. Bachand. 1997. Lodgepole Pine ( Pinus contorta ) Ecotones in Rocky Mountain National Pa rk, Colorado, USA Ecology 78 :632 641. Swift, R. S. 2001. Sequestration of Carbon By Soil. Soil Science 166 :858 871. Tallis, H., T. Ricketts, E. N. Driss Ennaanay, Kari Vigerstol, Guillermo Mendoza,, N. O. Stacie Wolny, Juliann Aukema, John Foster, Jessica Forrest, Dick Stanford. Thorsell, J. and T. Sigaty. 1997. A Global Overview of Forest Protected Areas on the World Hertitage List Natural Heritage Programme, IUCN, Gland, Switzerland. USFS. 200 9. Forest Health Damage Detection Surveys Acres with Mortality. USDA Forest Service. Veblen, T. T., K. S. Hadley, M. S. Reid, and A. J. Rebertus. 1989. Blowdown and stand development in a Colorado subalpine forest. Canadian Journal of Forest Research 19 : 1218 1225. Vose, J. 2006. Forest Carbon Cycling. Encyclopedia of Environmetrics. John Wiley & Sons, Ltd.

PAGE 97

87 Westerling, A. L., H. G. Hidalgo, D. R. Cayan, and T. W. Swetnam. 2006. Warming and Earlier Spring Increase Western U.S. Forest Wildfire Activity. Scie nce 313 :940 943. Westerling, A. L., M. G. Turner, E. A. H. Smithwick, W. H. Romme, and M. G. Ryan. 2011. Continued warming could transform Greater Yellowstone fire regimes by mid 21st century. Proceedings of the National Academy of Sciences 108 :13165 13170 Whelan, R. J. 1995. The Ecology of Fire. Cambridge University Press, UK. White, M. A., P. E. Thornton, S. W. Running, and R. R. Nemani. 2000. Parameterization and Sensitivity Analysis of the BIOME BGC Terrestrial Ecosystem Model: Net Primary Production C ontrols. Earth Interactions 4 :1 85. Wigley, T. M. L. and D. S. Schimel, editors. 2000. Constraints on the atmospheric carbon budget from spatial distributions of CO2. Cambridge University Press, Cambridge, UK. Winjum, J. K., R. K. Dixon, and P. E. Schroede r. 1992. Estimating the global potential of forest and agroforest management practices to sequester carbon. Water, Air, & Soil Pollution 64 :213 227. Zhu, Z., C. Key, D. Ohlen, and N. Benson. 2006. Evaluate Sensitivities of Burn Severity Mapping Algorit hms for Different Ecosystems and Fire Histories

PAGE 98

88 in the United States. USGS, National Center for Earth Resources Observation and Science, Sioux Falls, SD.

PAGE 99

89 Appendix I. Raw field data from 2 plots impacted by fire

PAGE 100

90

PAGE 101

91

PAGE 102

92

PAGE 103

93

PAGE 104

94

PAGE 105

95