Citation
The photosynthetic response of a high-altitude spruce forest to nitrogen amendments with implications for carbon sequestration

Material Information

Title:
The photosynthetic response of a high-altitude spruce forest to nitrogen amendments with implications for carbon sequestration
Creator:
Tripodi, Amber Dawn
Publication Date:
Language:
English
Physical Description:
xi, 93 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Spruce ( lcsh )
Photosynthesis ( lcsh )
Atmospheric deposition ( lcsh )
Nitrogen compounds -- Environmental aspects ( lcsh )
Atmospheric deposition ( fast )
Nitrogen compounds -- Environmental aspects ( fast )
Photosynthesis ( fast )
Spruce ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 87-93).
General Note:
Department of Geography and Environmental Sciences
Statement of Responsibility:
by Amber Dawn Tripodi.

Record Information

Source Institution:
Auraria Library
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
436869400 ( OCLC )
ocn436869400
Classification:
LD1193.L547 2009m T74 ( lcc )

Full Text
THE PHOTOSYNTHETIC RESPONSE OF
A HIGH-ALTITUDE SPRUCE FOREST TO
NITROGEN AMENDMENTS WITH
IMPLICATIONS FOR CARBON SEQUESTRATION
by
Amber Dawn Tripodi
B.S., University of Arkansas, 2005
A thesis submitted to the
University of Colorado Denver
in partial fulfillment
of the requirements for the degree of
Master of Science
Environmental Sciences
2009


This thesis for the Master of Science
degree by
Amber Dawn Tripodi
has been approved
by
Philip Riggs
Casey Allen
u f o 2ctf\
Date


Tripodi, Amber Dawn (M.S., Environmental Sciences)
The Photosynthetic Response of a High-Altitude Spruce Forest to Nitrogen
Amendments with Implications for Carbon Sequestration
Thesis directed by Associate Professor Frederick Chambers
ABSTRACT
Characterizing the capacity of the terrestrial carbon sink to absorb CO2 is one
of the most pressing topics in current climate science. One of the key players in this
arena is also one of the most poorly understood: the role of nitrogen deposition.
Human activity has increased the concentrations of both CO2 and fixed nitrogen in
the atmosphere. While anthropogenically derived nitrogen deposition seems to
fertilize some forests, in others it seems to be responsible for forest decline and tree
mortality. In order to address the role of nitrogen deposition in carbon sequestration, a
nitrogen-amendment study was conducted in a high-nitrogen deposition region of the
CO Rocky Mountains over the course of three years.
Chlorophyll fluorometry is a non-invasive technique that allows for the
characterization of photosynthetic efficiency and photoinhibition in plants. Analysis
of chlorophyll fluorometry data showed that in this tree line forest, additional
nitrogen significantly lowered the photosynthetic efficiency as indicated by the
quantum yield ( individual branches when compared to control branches. Additionally, nitrogen
treatments significantly improved the non-photochemical dissipation of excess
absorbed energy as indicated by the non-photochemical quenching of variable
fluorescence (qN). In high-altitude spruce forests receiving high nitrogen inputs,
chronic additions of nitrogen may impair carbon sequestration. One potential cause of
this is an increase in the light harvesting capacity of nitrogen amended branches
without a concurrent increase in photosynthesis. In a one-year study of the differential
effects of nitrogen treatments on sun- and shade-adapted tissues, no significant
differences due to nitrogen were indicated, but longer-term studies may illuminate the
underlying mechanisms at work in this system.
Many studies attempting to characterize the role of nitrogen deposition in
carbon sequestration are only conducted over a growing season, severely limiting
their applicability to ecosystem-level changes. In addition to the work with nitrogen


and chlorophyll fluorometry, this work highlights the importance of long-term studies
in the analysis of climate-related data.
This abstract accurately represents the content of the candidates thesis^I recommend
its publication.
Signed.
lencic Chambers


ACKNOWLEDGEMENT
Funding was provided through a Long-term Ecological Research (LTER) Grant
(#DEB-0423662) from the National Science Foundation (NSF). Logistical support
and/or data were provided by the NSF supported Niwot Ridge LTER project and the
University of Colorado Mountain Research Station. My committee members,
Frederick Chambers, Casey Allen and Philip Riggs, provided invaluable support and
advice in the preparation of this document. Casey Flynn, Anobha Gurung, Christine
Seibold, Herman Sievering, Tim Tomaszewski and Mark Williams also provided
additional assistance.


TABLE OF CONTENTS
Figures......................................................................ix
Tables.......................................................................xi
Chapter
1. Introduction..............................................................1
2. Photosynthesis & Chlorophyll Fluorometry Overview.........................7
2.1 The Light Reactions of Photosynthesis...................................7
2.2 The Fate of Absorbed Energy............................................9
2.3 Using Chlorophyll Fluorescence to Evaluate Photosynthesis..............10
2.4 The Basic Parameters of Chlorophyll Fluorometry........................10
2.5 Special Considerations of the Parameter Fo............................16
2.6 Calculated Parameters Used in this Study...............................18
2.6.1 Quantum Yield of PS13, Opsh...........................................22
2.6.2 Maximum Efficiency of PSII, FJFm.......................................23
2.6.3 Effective Quantum Yield of PSII, FyFm................................23
2.6.4 Reduction State of QA, 1 -qP and Photochemical Quenching of
Fluorescence, qP.......................................................24
2.6.5 Non-photochemical Quenching, NPQ.......................................25
2.6.6 Relative Change of Minimum Fluorescence Yield, q0......................26
2.6.7 Complete Non-photochemical Quenching of Fluorescence Yield, qCN.......26
2.6.8 Non-photochemical Quenching of Variable Fluorescence Yield, qN.........27
vi


2.6.9 Total Quenching of Fluorescence Yield, qTQ.............................27
3. Photoinhibition............................................................28
3.1 Basics of Photoinhibition.................................................28
3.2 The Influence of Sun- and Shade-Adaptation on Photoinhibition.............30
3.3 The Role of Nitrogen in Photoinhibition..................................31
4. Nitrogen Deposition and Its Role in Carbon Sequestration...................35
4.1 Nitrogen Deposition.......................................................35
4.2 Nitrogen at the Ecosystem Level: Limitation and Saturation................36
4.3 Nitrogen and Carbon Sequestration.........................................40
5. Previous Fluorometry Work at Niwot Ridge Forests...........................44
6. Materials and Methods......................................................47
6.1 Site Description..........................................................47
6.2 Chlorophyll Fluorescence Measurements.....................................49
6.3 Amendment Solutions.......................................................50
6.4 Trial One: Nitrogen Amendment Effects over Three Years....................51
6.5 Trial Two: Sun- and Shade-Adapted Nitrogen Amendment Effects..............52
6.6 Trial Two: Sun- and Shade-Adapted Morphological Characteristics...........53
7. Statistical Analysis: Methods and Results..................................54
7.1 Data T reatment...........................................................54
7.2 Statistical Analysis and Results.........................................57
8. Discussion of Results......................................................74
vii


8.1 Trial One: Nitrogens Effects on Photosynthetic Efficiency.............74
8.2 Trial One: Nitrogens Effects on Non-Photochemical Processes...........77
8.3 Trial Two: Nitrogens Effects on Sun- and Shade-Adapted Tissues........78
8.3.1 Fluorometry...........................................................78
8.3.2 Morphological Characteristics.........................................79
9. Comparison between Two Niwot Ridge Sites................................82
10. Conclusions and Implications............................................84
References..................................................................87
viii


LIST OF FIGURES
Figure
2.1 The Light Reactions of Photosynthesis...............................8
6.1 The Location of the Study Site .....................................48
7.1 Values of 7.2 Values of F/Fm Observed in Trial One ...............................61
7.3 Values of FJFm Observed in Trial One .............................62
7.4 Values of 1 -qP Observed in Trial One ..............................62
7.5 Values of NPQ Observed in Trial One ................................62
7.6 Values of qO Observed in Trial One .................................62
7.7 Values of qCN Observed in Trial One.................................63
7.8 Values of qN Observed in Trial One..................................63
7.9 Values of qP Observed in Trial One..................................63
7.10 Values of qTQ Observed in Trial One ...............................63
7.11 Values of 0Psn Observed in Upper Branches in Trial Two.............64
7.12 Values of 0Psn Observed in Lower Branches in Trial Two ............64
7.13 Values of F/Fm Observed in Upper Branches in Trial Two ............64
7.14 Values of F^/Fm Observed in Lower Branches in Trial Two............64
7.15 Values of F VFm Observed in Upper Branches in Trial Two...........65
7.16 Values of Observed in Lower Branches in Trial Two ............65
IX


7.17 Values of 1 -qP Observed in Upper Branches in Trial Two...........65
7.18 Values of 1 -qP Observed in Lower Branches in Trial Two ..........65
7.19 Values of NPQ Observed in Upper Branches in Trial Two.............66
7.20 Values of NPQ Observed in Lower Branches in Trial Two ............66
7.21 Values of qO Observed in Upper Branches in Trial Two..............66
7.22 Values of q0 Observed in Lower Branches in Trial Two .............66
7.23 Values of qCN Observed in Upper Branches in Trial Two ............67
7.24 Values of qCN Observed in Lower Branches in Trial Two.............67
7.25 Values of qN Observed in Upper Branches in Trial Two..............67
7.26 Values of qN Observed in Lower Branches in Trial Two .............67
7.27 Values of qP Observed in Upper Branches in Trial Two .............68
7.28 Values of qP Observed in Lower Branches in Trial Two..............68
7.29 Values of qTQ Observed in Upper Branches in Trial Two ............68
7.30 Values of qTQ Observed in Lower Branches in Trial Two ............68
7.31 Needle Density in Trial Two.......................................72
7.32 Needle Length in Trial Two........................................72
7.33 Needle Area in Trial Two .........................................72
7.34 Needle Fresh Mass in Trial Two....................................72
7.35 Needle Dry Mass in Trial Two .....................................73
x


LIST OF TABLES
Table
2.1 Common parameters measured in chlorophyll fluorometry......................13
2.2 Commonly calculated parameters in chlorophyll fluorometry .................14
2.3 Explanatory descriptions of the fluorometry parameters used
within this study...........................................................19
7.1 Percent of fluorometry variables removed due to mathematical
or mechanical constraints in trials one and two.............................56
7.2 Percent of fluorometry outliers removed due to exceeding a four
standard deviation threshold ...............................................57
7.3 Results of the Levenes Tests for homoscedasticity on fluorometry data ....58
7.4 Results from the Z-tests for normality of fluorometry data, part one: skewness...59
7.5 Results from the Z-tests for normality of fluorometry data, part two: kurtosis ....60
7.6 Results from Kruskal-Wallis One-Way ANOVA testing of fluorometry data......70
7.7 Results from One-Way ANOVA testing of morphological data
against treatment ..........................................................71
7.8 Results from One-Way ANOVA testing of morphological data
against branch position ....................................................71
9.1 A comparison of fluorometric responses to nitrogen amendments
at two Niwot Ridge sites: Cl and Soddie.....................................83
xi


1. Introduction
With the growing recognition that humans have fundamentally altered the
atmosphere and climate, there has been an upsurge of interest in the capacity of forest
systems to absorb increasing amounts of anthropogenic carbon dioxide (CO2)
releases. Trees can fix carbon by utilizing CO2 in photosynthesis and converting the
carbon into organic tissues, effectively removing it from the atmosphere until the
tissues are degraded. The boreal and temperate forests of North America are
considered excellent candidates for carbon sequestration due to their large
investments in woody tissue, which contain high amounts of carbon, and their long-
lived coniferous inhabitants (Luyssaert, et al., 2008). A carbon atom from CO2
incorporated in woody tissue has a residence time of at least 40 years and can surpass
100 years (Gaudinski, Trumbore, Davidson, & Zheng, 2000). In terms of sequestering
carbon from the atmosphere into the biosphere, the longer it can remain fixed, the
better.
Carbon sequestration depends upon the capacity of plants to photosynthesize,
and nitrogen plays a substantial role in this process (Evans, 1989). In the natural
environment, biologically available nitrogen is often in short supply and its
abundance is dependent upon the nitrogen-fixing capacity of microorganisms. As
such, nitrogen has historically been viewed as one of the primary limiting factors in a
plants ability to photosynthesize (Xia & Wan, 2008). Most of the worlds
1


mountainous forests are assumed to be nitrogen-limited, with their ability to sequester
carbon in part restricted by the availability of nutrient nitrogen (Hyvonen, et al.,
2008). In recent years, however, anthropogenic sources of biologically available
nitrogen have met or surpassed the level naturally provided by microorganisms
(Vitousek, et al., 1997). This may improve photosynthesis in some areas, as
additional nitrogen inputs remove one of the most common nutritional constraints of
photosynthesis. On the other hand, in traditionally nitrogen-limited environments,
many native plants are adapted to functioning under nitrogen constraints (Bowman,
Gartner, Holland, & Wiedermann, 2006). An excess of nitrogen may have the
opposite effect, with photosynthesis declining under increased nitrogen conditions.
The response of a forest to nitrogen inputs depends on both the nutrient state of the
forest and the amount of nitrogen added (Hyvonen, et al., 2008). Additionally,
altering the carbon-nitrogen cycle through increased availability of both biologically
available nitrogen and CO2 may have more complex effects on other aspects of forest
ecosystems (e.g. soil processes or biodiversity) which may impact the future capacity
of these systems to sequester carbon (Xia & Wan, 2008). Quantifying the capacity of
the terrestrial carbon sink to absorb CO2 is one of the most pressing current topics in
climate science today, one that cannot be understood without addressing the role of
biologically available nitrogen in carbon sequestration (Reay, Dentener, Smith,
Grace, & Feely, 2008).
2


Recent meta-analytical and modeling studies suggest that globally nitrogen
deposition is fertilizing the terrestrial biome, and forests worldwide are sequestering
more carbon due to increasing nitrogen availability (e.g. Magnani, et al., 2007; Xia &
Wan, 2008). Carbon-nitrogen dynamics are complex, however, and this pattern is not
universal. There have been studies suggesting that increasing nitrogen inputs to forest
systems in the American West can increase the photosynthetic efficiency of the trees,
indicating that nitrogen deposition may improve the capacity of these forests to
sequester carbon (e.g. Tomaszewski & Sievering, 2007). On the other hand, long-
term nitrogen amendment studies in Eastern forests, where nitrogen deposition is
greater, have suggested that forest health, and thus photosynthetic efficiency, have
declined as these areas have become nitrogen saturated (e.g. Magill, et al., 2004).
Understanding the dynamics of forest carbon responses to nitrogen deposition will
require investigating a number of systems and understanding the underlying
mechanisms that drive the differential responses in many different types of systems.
Studies of photosynthesis and stress in plants at the extremes of their ranges
can yield novel ecological insights that would be difficult to discern if the same
species were in more favorable conditions in the center of their ranges. While
obscured in less extreme areas, minor differences in photosynthetic performance can
differentiate between the presence and absence of an individual under harsh
conditions (Osmond, 1987). This is another inspiration for centering this study near
tree line, a region that exemplifies the very definition of the margin of a species
3


range. Available moisture and low temperatures limit the growth of the high-altitude
forests of the Rocky Mountains. These factors have been implicated in 1) unusually
high respiration rates that return up to 60% of carbon sequestered over a growing
season back into the atmosphere and 2) significantly lower net ecosystem production
rates in the area (Monson, et al., 2002). These qualities suggest that tree-line
environments provide harsh growing conditions, and the effects of stressors are more
likely to be exaggerated and observable.
Another point raised in the photosynthesis literature is that the effects of
nitrogen amendments are more likely to be observed over a time span of several years
than within a single growing season (e.g. Pregitzer, Burton, Zak, & Talhelm, 2008).
In addition to the storage or lack of storage of nitrogen within an ecosystem (e.g. in
soils), long-lived plants can accumulate nitrogen within their tissues for later use.
Conifers, for example, have the capacity to store excess nitrogen within proteins such
as Rubisco and in amino acids such as arginine (Calanni, et al., 1999; Warren,
Dreyer, & Adams, 2003). This implies that the effects of nitrogen deposition may
occur on a time lag, or may not be apparent immediately. Parts of this study were
conducted over a period of three years in order to capture some of these long-term
effects.
Methodology complicates many of the experimental studies that seek to
improve predictions of the effects of nitrogen deposition on carbon sequestration
(Reay, et al., 2008). The easiest and most common means of amending natural
4


nitrogen deposition to a forest system is through applications of nitrogen to the soil.
Although trees do certainly uptake nitrogen through their roots, canopy uptake is also
a substantial nitrogen uptake pathway for trees (Boyce, Friedland, Chamberlain, &
Poulson, 1996). In one study, the tree canopy retained up to 80% of deposited
nitrogen (Sievering, Tomaszewski, & Torizzo, 2007). Applying nitrogen to soil
invites competition for the nitrogen from soil microbes and other soil processes,
which may confound the interpretation of results from a tree-centric perspective. As
nitrogen is deposited from the atmosphere to tree canopies (as well as to the soil), it
seems a more natural fit for experiments examining the effects of deposition to
manipulate nitrogen available through the canopy, rather than through the soil. This
study adds nitrogen directly to the foliage to mimic the natural processes that follow
ambient nitrogen deposition.
The main aim of this study is to examine the potential effects of nitrogen
deposition on carbon sequestration through foliar applications of biologically
available nitrogen to Engelmann spruce at a high-altitude, tree line site with higher
than average ambient nitrogen deposition (Holland, Braswell, Sulzman, & Lamarque,
2005). Photosynthetic efficiency is monitored using a chlorophyll fluorometer, a
simple hand-held instrument that can be used nondestructively in the field.
Fluorometry can be used a qualitative surrogate measure of photosynthetic efficiency
and capacity (Maxwell & Johnson, 2000). If nitrogen amendments can affect the
capacity for light harvesting and utilization, they are likely to affect gross
5


photosynthesis as well. Additionally, fluorometry provides information on the
ultimate fate of absorbed energy. If plant tissues absorb more energy than is useful for
photosynthesis, this excess energy has the potential to damage the photosynthetic
apparatus of plant tissue, leading to a reduced capacity for photosynthesis and carbon
sequestration. If nitrogen amendments can affect the non-photochemical dissipation
of energy, this may have implications for long-term carbon sequestration capacity as
well.
In order to investigate the effects of additional nitrogen on photosynthetic
efficiency at this site, two experiments were designed. In trial one, a comparison is
made between control and nitrogen-amended branches over a three year period. In
trial two, a single season of data is collected from application of nitrogen amendments
as compared to a control in both shade- and sun-adapted tissues in the hope of teasing
out the underlying, nitrogen-dependent mechanisms that may be at work.
6


2. Photosynthesis & Chlorophyll Fluorometry Overview
2.1 The Light Reactions of Photosynthesis
Photosynthesis is a two-step process, with each step composed of a series of
reactions. The light reactions take place within the thylakoid membranes, and are the
first series of processes leading to photosynthesis (Fig. 2.1). In this reaction series,
light energy is harvested and converted to chemical energy, creating oxygen (O2) as a
waste by-product. This chemical energy is then shunted to the stroma, where the
carbohydrate synthesis series, the Calvin cycle, takes place. Here the chemical energy
is used to fix carbon from CO2 to synthesize sugars (Campbell, Reece, Mitchell, &
Taylor, 2003). In chlorophyll fluorescence, it is the light reactions that are of interest.
The light reactions are initiated when the antennae molecules (pigments)
located in the chloroplasts absorb a photon (quantum of light energy). In photosystem
II (PSII), the antennae contain chlorophyll a pigments, which absorb light in the blue-
violet and red ranges, with 680 nm absorbed most efficiently. In photosystem I (PSI),
Chlorophyll b can absorb light energy in the blue and orange ranges, with light at 700
nm absorbed most efficiently. Rather than participating in the light reactions directly,
PSI will shunt the energy to PSII, expanding the range of light that can be utilized in
photosynthesis.
7


^ v-,
(Electron Acceptor)
Electron
Transport Chain
Pigments
Excited Electron
680 nm
Photon
Pigments
PSII
FIGURE 2 .1. The Light Reactions of Photosynthesis. Diagram illustrating the fate of absorbed
light. The dotted lines represent alternative energy qunenching pathways. Adapted from Campbell, et
al., 2003 and Ritchie, 2006.


This absorbed energy is passed from one pigment molecule to another, until it reaches
the reaction center of the photosystem. The reaction center is composed of
chlorophyll a pigment molecules and the primary quinone electron acceptor, QA. The
energy absorbed by the reaction center excites electrons within the pigment, and the
electrons then release energy as they return to their original ground state (Campbell,
et al., 2003).
2.2 The Fate of Absorbed Energy
This energy must be quenched (utilized or dissipated) to avoid damaging the
tissues. Absorbed energy can be quenched by being utilized in photosynthesis,
emitted as fluorescence or dissipated as heat. If QA is in an oxidized state, the excited
electrons will reduce it and photosynthesis will continue. If QA is already in a reduced
state, the energy will be dissipated as fluorescence or heat. These processes are
inversely related to one another, in that an increase in energy quenching by one
process will result in a decrease in the quenching carried out by the other two
processes (Maxwell & Johnson, 2000). This relationship is the basis of using
chlorophyll fluorometry to analyze photosynthesis and general plant performance.
Chlorophyll fluorescence, while a minor part of the overall photosynthetic process, is
an easily measured portion that can be used to discern properties of photosynthesis
and overall plant health.
9


2.3 Using Chlorophyll Fluorescence to Evaluate Photosynthesis
Chlorophyll fluorescence is a popular method for assessing plant performance
in the field due to the ease of its use. Still, one must be cautious of oversimplifying
the information garnered from the practice. Chlorophyll fluorometry may give an
indication of the photosynthetic efficiency and capacity of a plant in the field, but it
cannot be used as a direct estimator of photosynthesis between different plants or
treatments. The correlation between the electron transport in PSII and CO2 fixation
that is observed in controlled laboratory conditions will not hold true under most field
conditions. Additionally, rates of photosynthesis in different tissues are subject to a
number of uncontrollable metabolic conditions such as photorespiration that cannot
be captured by fluorometry. As such, one should be cautious in interpreting
differences in fluorometric performance directly as differences in photosynthesis.
Fluorometry can capture the efficiency and capacity for photosynthesis, as well as
photoinhibition and the effects of stressors on photosynthetic capacity. It is this
quality that gives fluorometry its power (Maxwell & Johnson, 2000).
2.4 The Basic Parameters of Chlorophyll Fluorometry
Chlorophyll fluorometry is conducted on both light- and dark-adapted plant
tissues. If the sample is darkened for around 15-20 minutes (Maxwell & Johnson,
2000), it reaches a steady-state characterized by a lack of electrons in transit within
10


the photosystem. In a dark-adapted sample, no photosynthesis is taking place. All of
the Qa molecules in PSII are thus in an oxidized state and the reaction centers are said
to be open. In this state of darkness, the non-photochemical means of energy
quenching, including fluorescence are at their minimums. Fluorescence measured in
this state is zero fluorescence and is recorded as F0 (Table 2.1). If the sample is
then exposed to a saturating pulse, a short-duration, high-intensity light, the PSII QA
molecules become fully reduced, and the reaction centers are said to be closed.
With this very brief pulse of light, no photochemical quenching is possible (Maxwell
& Johnson, 2000). The fluorescence emitted is then at the maximum magnitude and is
recorded as Fm (Table 2.1). If a darkened sample is then exposed to actinic light and
allowed to reach a steady state, photosynthesis resumes and fluorescence declines.
This parameter is recorded as F (Table 2.1). A saturating pulse applied to tissue in
this state will induce a maximum fluorescence lower than that of the dark-adapted
tissue and is recorded as Fm (Table 2.1). If the actinic light is then removed and a far-
red light is applied, the QA molecules within the reaction centers are quickly
reoxidized and the minimal fluorescence in the light-adapted state, F0 (Table 2.1),
can be recorded (Rohacek & Bartak, 1999). These are the basic parameters measured
in fluorometry (Table 2.1), and all other parameters can be calculated from these
initial measurements (Table 2.2). Although there have been attempts at synonymizing
the fluorometry parameters in the literature (e.g. Maxwell & Johnson, 2000),
contradictions and misnomers continue to confound workers in the field. To avoid
11


confusion, Tables 2.1 and 2.2 summarize the parameters used in this paper and cross-
reference them with others commonly encountered in the literature.
12


Table 2.1. Common parameters measured in chlorophyll fluorometry.
Parameter Represents Light Condition Pseudonyms Notes
F0 Fm Minimal Fluorescence Maximal Fluorescence Dark-Adapted Dark-Adapted F m
F Steady State Fluorescence Light-Adapted F F,
F 1 o Minimal Fluorescence Light-Adapted Directly measured
F9 1 m Fp Maximal Fluorescence Maximal Fluorescence Light-Adapted Light-Adapted If a saturating pulse is
PPFD1 Photosynthetic Photon Flux Light-Adapted PFD, PAR applied, Fp =Fm
i aL i.rr- Density Leaf Absorption Light-Adapted a
Not included in this study.


Table 2.2. Commonly calculated parameters in chlorophyll fluorometry.
Parameter Represents Light Condition Pseudonyms Calculation
Fo Minimal Fluorescence Light-Adapted F oOB F Fv/ +Fn/\ /F / F
Fv Variable Fluorescence Dark-Adapted Fm F
F9 1 V Variable Fluorescence Light-Adapted Fm-F0
F r ^PSII Quantum Yield of PS II Light-Adapted AF - 02, 0g, Yield Fm K Fm
17 Maximum Efficiency of PSII Dark-Adapted 0m, 0Po, 0PO,
* m K K 1 -qP Effective Quantum Yield of PSII Light-Adapted 4>m, r, Y
Reduction State of QA Light-Adapted
qP Photochemical Quenching of Fluorescence Light-Adapted AF ,qp, qo Fv Fv K Fv
ETR1 Electron Transfer Rate Light-Adapted J, JPSII Opsir PPFD i
NPQ Non-photochemical Quenching Comparative SVN F F m m


Table 2.2. (Cont.)
Parameter Represents Light Condition Pseudonyms Calculation
qO Relative Change of Minimum Fluorescence Yield Comparative q Fo-K F0
qCN Complete Non-photochemical Quenching of Fluorescence Yield Comparative qcN Fm~K Fm
qN Non-photochemical Quenching of Variable Fluorescence Yield Comparative qN> qsp fv-f: F
qTQ Total Quenching of Fluorescence Yield Comparative qrQ ^ l
Not included in this study.


2.5 Special Considerations of the Parameter Fo
Although modulated pulse fluorometers are equipped to measure F, the
results are not without some confounding difficulties. Prior to recording F, the leaf
tissue is exposed to far-red illumination which reoxidizes the reaction centers of PSII
and allows for the measurement of minimal fluorescence in a light-adapted state.
Unfortunately, far-red light also affects the reaction centers of PSI which can lead to
false readings of the oxidation state of PSII (Stefanov & Terashima, 2008) in which
the Qa molecules within the reaction centers of PSII may not be maximally oxidized
(Baker, 2008). Additionally, the magnitude of non-photochemical quenching is
assumed to remain constant between the determination of F and F0. In fact, the
changes in the proton gradient across the thylakoid membrane happen quite rapidly
and there is a good chance that the degree of non-photochemical quenching can
change between the two measurements (Oxborough & Baker, 1997). Rapid relaxation
of non-photochemical quenching can lead to an overestimate of F0 (Baker, 2008).
Oxborough and Baker (1997) devised a method of calculating the parameter from the
more robust measurements of F0, Fm and Fm to avoid the errors inherent in directly
measuring F.
Three assumptions govern the calculation of F0: 1) all PSII Qa molecules
within the reaction centers are fully oxidized, or open, when F is measured, 2) there
is no decrease in the amount of non-photochemical quenching between the
16


measurements of F0 and Fm and 3) there is no change in the number of
photochemically active PSD reaction centers between the measurements of Fm and
Fm (Baker & Oxborough, 2004). If these three assumptions are satisfied, the
calculation of F0 should be more accurate than the measurement of F0 under far-red
influence. The equation for calculating F0 is
F' =
Eq. 2.1
and has been included in the calculated parameter table (Table 2.2). As long as there
is no change in the photoinhibition state of the tissue during the period between
measuring Fm and Fm, the calculated parameter should be more accurate than the
measured one (Baker, 2008). In this experiment, the PPFD was monitored to ensure
that each light-adapted measurement was taken under similar light conditions, and
that dark-adapted measurements were conducted after equivalent dark-adaptation
times, so these conditions should be uniform across all measures. As such, the
calculated F0 parameter, rather than measured F0 values, will be used in subsequent
analysis.
17


2.6 Calculated Parameters Used in this Study
Although fluorometry can be a useful tool in discerning the health of plants,
some fundamental cautions must be recognized. The values of the basic parameters of
fluorescence that are directly measured, Fot Fm and Fm, are controlled by a number of
both physiological and environmental factors that are in constant flux and cannot
often be controlled in vivo. Comparing the F0 of two treatments, for example, would
not account for the potential differences in leaf tissues due to factors (e.g. leaf
hydration affects photon absorption), which could lead to convoluted interpretations.
While direct comparisons of basic measurements between treatment groups are
unsound, using calculated parameters that depend upon ratios to normalize the
differences that are associated with these fluctuations overcomes this problem (Baker,
2008). As such, much of the interpretive value of chlorophyll fluorescence depends
upon calculating additional parameters that are built upon ratios that negate the
inherent differences that might be seen in the basic, measured parameters. A summary
of the interpretations of the major calculated parameters of chlorophyll fluorometry
used within this study has been included to guide the reader (Table 2.3).
18


Table 2.3. Explanatory descriptions of the fluorometry parameters used within this study.
Parameter Represents Describes Constraints
F Minimal Fluorescence A dark-adapted measure of the amount of fluorescence when all Qa molecules in PSII are fully oxidized and PSII reaction centers are open, thus the potential for photochemistry is maximal
Fm Maximal Fluorescence A dark-adapted measure of the amount of fluorescence when all Qa molecules in PSII are fully reduced and PSII reaction centers are closed, thus the potential for photochemistry is minimal jc< 2.558*
F Steady State Fluorescence A light-adapted measure of the lower amount of fluorescence emitted in steady light as photosynthesis is occurring
F r o Minimal Fluorescence A light-adapted measure of the amount of fluorescence when all Qa molecules in PSII are fully oxidized and PSII reaction centers are open
F, Maximal Fluorescence A light-adapted measure of the amount of fluorescence when all QA molecules in PSII are fully reduced and PSII reaction centers are closed x < F,; jc < 2.558*
Fv Variable Fluorescence The difference between the maximal fluorescence and minimal fluorescence in dark- adapted tissue; an estimator of the capacity for Qa reduction and photosynthesis


Table 2.3. (Cont.)
Parameter Represents
F\ Variable Fluorescence
Fluorescence Quenching
^PSII Quantum Yield of PSII
20 Maximum Efficiency of PSII
K K Effective Quantum Yield of PSII
1 -qP Reduction State of QA
qP Photochemical Quenching of Fluorescence
Constraints
Describes___________________________________
The difference between the maximal
fluorescence and minimal fluorescence in light-
adapted tissue; an estimator of the capacity for
Qa reduction and photosynthesis
The difference between the maximal
fluorescence and steady state fluorescence in
light-adapted tissue; an estimator of
photochemical quenching of fluorescence
Estimates the operational efficiency at which
absorbed photons reduce QA within PS II
A relative estimate of the maximum quantum
yield of PSII photochemistry based upon the
efficiency of QA reduction
A relative estimate of the yield of PSII in the
light once non-photochemical quenching has
been initiated
Quantifies the proportion of reduced (closed)
Qa molecules within the reaction centers of
PSII
Estimates the reduction of fluorescence by
photochemistry through a quantification of the
proportion of oxidized (open) QA molecules
within the reaction centers of PSII; describes
the relationship between the maximum
efficiency of PSII and the operational efficiency
of PSII
0 < x < 1
0 0 x < Fv/Fm
0 < x < 1
0 < x < 1


Table 2.3. (Cont.)
Parameter Represents Describes Constraints
NPQ Non-photochemical Quenching Estimates the reduction in fluorescence not due to photosynthetic quenching (i.e. heat) between dark-adapted maximal fluorescence and light- adapted maximal fluorescence, often equated with ApH-dependent, non-damaging processes 0 qO Relative Change of Minimum Fluorescence Yield Describes the change in minimal fluorescence between dark- and light- adaptation due to non- photochemical processes -0.2 < x < 1
qCN Complete Non-photochemical Quenching of Fluorescence Yield Quantifies all of the thermal dissipation processes within PSII based on both changes in variable fluorescence and minimal fluorescence between the dark- and light- adapted states 0 < jc < 1
qN Non-photochemical Quenching of Variable Fluorescence Yield Based on the relative change in variable fluorescence from dark-adaptation to light adaptation, estimates the reduction in fluorescence not due to photosynthetic quenching (i.e. heat), often equated with ApH- dependent, non-damaging processes 0 qTQ Total Quenching of Fluorescence Yield Quantifies the total quenching of fluorescence by both photochemical and non-photochemical processes 0 < jc < 1
indicates a mechanical constraint limited to the PAM-2100 fluorometer. Table compiled from Adams & Demmig
Adams, 2004; Baker, 2008; Maxwell and Johnson, 2000; Osmond, 1994; Rohacek, 2002 and Rohacek & Bartak,
1999.


2.6.1 Quantum Yield of PSII, Opsh
Many parameters used in chlorophyll fluorometry quantify the quantum yield
or efficiency of processes involving photons within the leaf. In general terms, the
magnitude of a quantum yield is determined by dividing the number of molecules
involved in the process by the number of photons that have been absorbed (Baker,
2008). For PSII photochemistry, the parameter Opsh (Table 2.3) describes the
proportion of absorbed photons (energy) that are used in photosynthesis (Maxwell &
Johnson, 2000). The parameter PSn is sometimes termed as the operating efficiency
of PSII and is linearly associated with rates of CO2 assimilation and O2 evolution in
empirical studies. Through this association, OpSn is seen as a surrogate measurement
of photosynthetic activity (Baker, 2008). Its popularity in the fluorometry literature is
linked to the fact that it can be calculated without dark adaptation or the problematic
Fo measurement, making it ideal for field conditions. Mathematically, Opsiican also
be expressed by multiplying qP and F/Fm, which illustrates that amount of energy absorbed in PSD (as limited by the proportion of open reaction
centers, qP) that is actually used in effective photochemical processes (Rohacek,
2002). Strictly speaking, a reduction in Opsn indicates a reduction in energy use
efficiency (Osmond, 1987).
22


2.6.2 Maximum Efficiency of PSII, F/Fm
The parameter F/Fm (Table 2.3) describes the maximum efficiency of the
intrinsic capacity of photosynthesis in PSII if all reaction centers were open (Maxwell
& Johnson, 2000). Although F/Fm should not be taken as a quantitative estimate of
the maximum efficiency of photosynthesis, it is a useful relative measure of
photosynthetic potential. There is a standard maximum value reported for unstressed
plants of 0.83 (or, more exactly 0.832 0.004) across the fluorometry literature, and
values that drop below this are usually indicative of stress (Rohacek, 2002).
Discerning the exact stressor causing a decrease of FJFm in a particular system is,
however, much more difficult (Baker, 2008). Changes in F/Fm are typically
indicative of a change in the efficiency of non-photochemical quenching. Sustained
reductions in F/Fm indicate that photoinhibition is occurring (Maxwell & Johnson,
2000).
2.6.3 Effective Quantum Yield of PSII, FyJFm
Whereas FJFm speaks to the photosynthetic potential of plant tissue, the
parameter FVFm (Table 2.3) captures the proportion of OPSii that is actually realized
(Baker & Oxborough, 2004). Along with F/F F VFm is considered one of the most
robust and reliable fluorescence parameters (Adams & Demmig-Adams, 2004).
F/Fm is never greater than F/Fm, and the reduction between the dark-adaptation
23


and the light-adaptation is due to non-photochemical processes (Rohacek & Bartak,
1999). Experimentally, FJFm is usually inversely related to NPQ or qN, with
decreases in FyFm accompanied by increases in NPQ and qN (Adams & Demmig-
Adams, 2004).
2.6.4 Reduction State of QA, 1 -qP and Photochemical Quenching of
Fluorescence, qP
The parameter qP (Table 2.3) indicates the proportion of open reaction centers
in PSII (Maxwell & Johnson, 2000), and can be used to quantify the reduction in
photosynthetic capacity due to the reduction state of QA within PSII (Oxborough &
Baker, 1997). In effect, qP quantifies the proportion QA molecules that are open
(oxidized) within the reaction centers of PSII and thus prepared to conduct
photochemical processes (Rohacek, 2002). As such, some authors term qP as the PSII
efficiency factor (Baker, 2008). The reduction state of QA has been linked to dynamic
photoinhibition, with increased qP (or decreased 1 -qP) indicative of a greater
capacity for protective, dynamic photoinhibition (Posch, Warren, Kruse,
Guttenberger, & Adams, 2008). Because some authors in the field favor \-qP (Table
2.3) over qP, both parameters are included here, although they capture the same
information.
24


2.6.5 Non-photochemical Quenching, NPQ
One of the most widely used parameters in fluorometry research is NPQ
(Table 2.3), which estimates non-photochemical quenching. In the presence of
photons in excess of those necessary for photochemistry, plants will release energy in
an effort to avoid damage from any excess. The parameter NPQ estimates the
quenching of fluorescence emission due to heat dissipation and is associated with the
non-damaging, ApH-dependent process of dynamic photoinhibition (Osmond, 1994),
which is discussed further below. Experimentally, increases in NPQ are often
dynamic and associated with rapid changes in the xanthophyll cycle across many
studies (Adams & Demmig-Adams, 2004). Non-photochemical quenching is often
seen as a regulatory process that protects the chloroplasts from damage. It should be
noted, however, that the NPQ parameter alone cannot distinguish between dynamic
and chronic photoinhibition (Stefanov & Terashima, 2008).
Like qN, the parameter NPQ estimates changes in heat dissipation in light-
adapted tissue relative to the dark-adapted state (Table 2.3). NPQ is calculated on the
basis of changes in the maximal fluorescence between dark- and light- adapted states,
while qN is calculated from the changes in variable fluorescence (Rohacek, 2002).
NPQ and qN both capture non-photochemical quenching relatively well, but on
different scales. Because NPQ is measured on a scale from 0 to infinity, differences in
25


NPQ values are often easier to detect than differences in qN values, which are
measured on a scale between 0 and 1 (Maxwell & Johnson, 2000).
2.6.6 Relative Change of Minimum Fluorescence Yield, qO
The parameter qO (Table 2.3) describes the changes in minimal fluorescence
that take place upon exposing dark-adapted tissue to actinic light. It is associated with
the regulation of electron flow, PSEI reaction center inactivation and conformational
changes within the thylakoid membranes (Rohacek & Bartak, 1999). Thus, qO
describes changes in fluorescence minimums that are due to non-photochemical
processes (Rohacek, 2002).
2.6.7 Complete Non-photochemical Quenching of Fluorescence Yield, qCN
The complete non-photochemical quenching of fluorescence parameter, qCN
(Table 2.3), quantifies the energy dissipation associated with both the non-
photochemical quenching of variable fluorescence captured by qN and the changes
associated with minimal fluorescence between dark- and light-adaptation described
by qO. Thus, qCN captures the complete non-photochemical quenching taking place
within PSD (Rohacek & Bartak, 1999).
26


2.6.8 Non-photochemical Quenching of Variable Fluorescence Yield, qN
The parameter qN (Table 2.3) predates NPQ as a measure of the non-
photochemical quenching of fluorescence, or heat dissipation. Because qN is built
upon variable fluorescence alone, it has a smaller range of potential values than NPQ.
It may be more difficult to detect changes in non-photochemical quenching using qN.
Because of the differences in scale between NPQ and qN, many workers in the field
have favored NPQ as a measure of non-photochemical quenching (Maxwell &
Johnson, 2000). Like NPQ, qN is a sensitive indicator of stress, and unlike qP, tends
to vary greatly throughout a growing season as it adjusts rapidly to changing
environmental conditions (Ritchie, 2006). A variety of factors can alter qN, including
changes in ApH, photoinhibition of PSII reaction centers and damage to the antennae
of PSII (Rohacek & Bartak, 1999).
2.6.9 Total Quenching of Fluorescence Yield, qTQ
The parameter qTQ quantifies the total fluorescence quenching by all
quenching processes, both photochemical and non-photochemical (Table 2.3). It
provides a picture of the total excitation energy moving through PSII in light-adapted
samples, regardless of the mechanism for its dissipation or use (Rohacek & Bartak,
1999).
27


3. Photoinhibition
3.1 Basics of Photoinhibition
Photoinhibition is a term used to describe the reduction in photosynthetic
efficiency characterized by a reduction in the fluorometry parameter FJFm (Skillman
& Osmond, 1998). Photoinhibition illustrates a disconnection between the processes
of light absorption and CO2 fixation. An optimally functioning system can utilize
absorbed light to fix carbon, either reductively through photosynthesis or oxidatively
through respiration. When the light absorbed by the antennae of PSII is in excess,
which is often the case, the energy absorbed must be rerouted or it can permanently
damage the tissues. Photoinhibition can either be a regulated, protective process that
is reversible (dynamic photoinhibition), with the effects lasting only minutes to hours,
or an irreversible process (chronic photoinhibition) reflecting damage to the PSII
reaction centers, with the effects lasting from hours to days. Although their time
scales overlap, fluorometry can distinguish the two types by monitoring changes in
the dark-adapted parameter F0, or zero fluorescence. If a reduction in the parameter
FJFm suggests photoinhibition, a reduction in F0 will characterize dynamic
photoinhibition, while an increase in F will characterize chronic photoinhibition
(Skillman & Osmond, 1998). The harmless means by which excess energy is
dissipated include a minor amount as fluorescence and a larger proportion as heat
(Alves, Magalhaes, & Barja, 2002). The destructive means of excess energy
28


dissipation include the creation of free radical oxygen molecules and promoting
oxygen molecules within the reaction center to an excited state. Both of these
processes will degrade PSII reaction centers, physiologically reducing a plants
photosynthetic capacity (Hendrickson, Furbank, & Chow, 2004), and both will
generally cause an increase in the chlorophyll parameter F0 (Baker, 2008).
Dynamic photoinhibition is thought to be regulated by the antennae and
reaction centers of PSII. The antennae absorb light, but in the presence of excess light
or the absence of carbon fixing, the typical transport of electrons is halted and the
energy is dissipated as heat. Thermal dissipation is a major process. In full sunlight,
plants lose 50 70% of the energy from absorbed light to thermal dissipation. This
process is regulated by a proton gradient across the thylakoid membrane formed
during the process of electron transport, frequently referred to as ApH in the literature
(Posch, et al., 2008). While this gradients main function is to drive ATP synthesis, it
also acts as a means of regulating the fate of absorbed photons via biochemical
feedback. As the pH in the thylakoid lumen drops below 5.5, two processes occur: 1)
the chlorophyll pigments in the light harvesting centers aggregate and 2) the
conversion of violaxanthin to zeaxanthin initiates the xanthophyll cycle, both of
which can be responsible for the quenching of excited electrons and the non-
destructive dissipation of their energy as heat (Pospisil, 1997). As the difference in
pH between the interior and exterior of the thylakoid (ApH) increases, the
xanthophyll cycle is initiated and violaxanthin is converted to zeaxanthin (Niyogi,
29


2000). The excess protons in the lumen may also protonate some carboxylic proteins
in the antennae of PSII (Baker, 2008). Together, these events stimulate a
conformational change in PSII, leading to the harmless dissipation of excess energy
as heat (Niyogi, 2000). As the light stress is decreased, the ApH is reduced,
zeaxanthin is converted back to violaxanthin, the carboxyl groups on associated
proteins are deprotonated and the energy from photon absorption is once again routed
to photosynthesis (Baker, 2008). The protein D1 has also been shown to play a crucial
role in regulating photoinhibition, as its absence will cause a reduction in the FJFm
parameter. The turnover capacity of D1 is tied to the xanthophyll cycle in that
inhibition of the cycle leads to a greater loss in the D1 protein and greater expressed
photoinhibition (Alves, et al., 2002). Environmental stressors, such as nutritional
deficiencies, drought or low temperatures can reduce a plants capacity to carry out
photosynthesis and elevate photoinhibition (Niyogi, 2000).
3.2 The Influence of Sun- and Shade-Adaptation on Photoinhibition
High light situations can often amplify the physiological effects of various
stressors. Shade and sun adaptation is environmental and takes place at the leaf level,
with varying distributions of photosynthetic machinery (Osmond, 1987). For
example, plants environmentally adapted to high light conditions will exhibit grana,
stacks of thylakoids, while those adapted to shade arrange their thylakoids lamellarly.
This can effect the creation of ApH, with shade plants less capable of dynamic non-
30


photochemical quenching than sun plants (Stefanov & Terashima, 2008). Sun-
adapted tissues also show a substantial reduction in the proportions of the tissue
functioning as antennae and an increase in the amount of tissue dedicated to reaction
centers when compared to shade-acclimated leaf tissues (Osmond, 1987). Often
shade-adapted plants show a maximum photosynthesis rate comparable to that of
light-adapted plants, but if placed in full sunlight, a reduction of quantum yield can be
observed (Osmond, 1987). This differential acclimation allows the investigator to
amplify the effects of a stressor by placing shade-adapted tissues into full sunlight
and measuring the quantum yield of photosynthesis.
3.3 The Role of Nitrogen in Photoinhibition
Although the mechanisms are still unclear, empirical analyses suggest that
nitrogen is indispensable to dynamic photoinhibition and the process of acclimating
to differing light conditions. In studies comparing high and low nitrogen treatments,
plants grown with lower nitrogen regimes exhibit more photoinhibition and a greater
reluctance to adapt to high light conditions (Osmond, 1987). The influence of
nitrogen becomes more obvious in studies comparing photoinhibition in shade- and
sun-adapted plants. While nitrogen regimens had no effect on the quantum yield
(Opsii) of shade-adapted plants kept under low light conditions, when transferred to
bright light the low nitrogen plants exhibited a greater depression of Opsii and a
slower recovery time than the high nitrogen plants (Ferrar & Osmond, 1986).
31


Similarly, the level of available nitrogen has been implicated in the ability of shade-
adapted plants to acclimate to high light conditions, suggesting that nitrogen plays a
significant role in favoring protective, dynamic photoinhibition over damaging,
chronic photoinhibition (Skillman & Osmond, 1998). These results have been
confirmed in conifer species as well (Grassi, Colom, & Minotta, 2001; Posch, et al.,
2008). Interestingly, Grassi et al. (2001) found that in conifers kept under low light
regimes, plants with a reduced nitrogen supply exhibited an elevated rate of
photochemical quenching of fluorescence (qP) over those plants with high levels of
available nitrogen despite similar rates of photosynthesis as measured by gas-
exchange. This suggests that in conifers adapted to shade conditions, increased
nitrogen supplies elevate levels of dynamic photoinhibition (Grassi, et al., 2001).
These results are mirrored in a 2008 study analyzing the differences under high and
low nitrogen regimes using shade- and sun- adapted portions of a single tree, rather
than different plants. The shade-adapted limbs given low nitrogen treatments
exhibited lower values for 1 -qP as well as higher values for NPQ when exposed to
high light, again suggesting an increased capacity for protective photoinhibition
within nitrogen-limited, shade-adapted plants (Posch, et al., 2008).
These results suggest that nitrogen enhances the capacity of plant tissue to
absorb light. Under non-light-stressed conditions, nitrogen will encourage
photosynthesis by increasing the amount of photons absorbed by the tissue. Under
conditions of high light stress, this extra absorption capacity can lead to damage to
32


PSD by increasing the amount of energy in excess of what can be utilized in
photosynthesis. Other studies have shown that nitrogen amendments can lead to an
increased capacity to absorb light in other conifers, including Scots pine (Wang &
Kellomaki, 1997), Monterey pine (Posch, et al., 2008) and Engelmann spruce
(McKinnon & Mitchell, 2003).
This hypothesis is also borne out by studies analyzing the allocation of
nitrogen to various plant compounds, most notably ribulose-1,5-biphosphate
carboxylase/oxygenase (Rubisco) and chlorophyll within the leaf. Nitrogen is
allocated to chlorophyll and Rubisco differently in sun- and shade-adapted tissues
when nitrogen is limiting, with shade-adapted plants allocating more nitrogen to
chlorophyll than Rubsico (Makoto & Koike, 2007). The synthesis of chlorophyll, the
light-harvesting pigment within plant tissues, is heavily dependent upon the
availability of nitrogen. Increases in available nitrogen are often accompanied by
increases in the concentration of chlorophyll within leaves and needles (Makoto &
Koike, 2007; McKinnon & Mitchell, 2003; Posch, et al., 2008). Studies analyzing the
concentration of Rubisco in plant leaves under differing nitrogen regimes have
generally concluded that increasing nitrogen availability leads to greater levels of the
compound within the leaves (Makoto & Koike, 2007; Millard, Sommerkom, &
Grelet, 2007; Posch, et al., 2008; Sievering, et al., 2007). Rubisco is arguably one of
the most important enzymes on earth as it is the most abundant and is responsible for
the initial catalytic reduction of CO2 in photosynthesis. In addition, Rubisco is
33


thought to serve as a nitrogen storage compound. (Millard, et al., 2007) and Rubisco
synthesis accompanies the acclimation of shade-adapted plant tissue to high light
conditions (Henley, Levavasseur, Franklin, Osmond, & Ramus, 1991). Nitrogens
effect on photosynthesis is most closely tied to these two compounds.
34


4. Nitrogen Deposition and Its Role in Carbon Sequestration
4.1 Nitrogen Deposition
Nitrogen in the form of N2, while composing about 78% by volume of air, is
primarily unavailable to the biological organisms that depend on it. In order to be
utilized biologically, nitrogen must be fixed into available forms, which was until
recently, a process primarily carried out by numerous microorganisms. Human
activities such as agricultural fertilizer production, biomass burning and fossil fuel
combustion also fix nitrogen in surprisingly large amounts. Within the last 30 years,
anthropogenic nitrogen fixation has surpassed that carried out by natural systems,
effectively doubling the amount of nitrogen available to organisms (Vitousek, et al.,
1997). This additional fixed nitrogen can be transported over large distances in the
atmosphere and is returned to terrestrial and aquatic systems through deposition
(Tamay, Gertler, & Taylor, 2002). Fixed nitrogen can be deposited wet, as in
precipitation, dry, as vapor or particulate matter, or occult, as clouds and fog. Wet
deposition is fairly easy to measure and analyze, and as a result, many deposition
datasets only include this measurement for nitrogen (Erisman, Beier, Draaijers, &
Lindberg, 1994). Occult deposition is considered the most difficult to directly
measure and is often ignored, as only fog-prone and some high altitude areas are
thought to be substantially affected by this type of deposition
35


(Sievering, Rusch, & Marquez, 1996). Measuring deposition of dry nitrogen is often a
cumbersome project, and as a result, many of the so-called nitrogen budgets leave out
this element entirely. However, in some ecosystems, dry deposition may account for
over 50% of the total nitrogen input during the growing season (Sievering, et al.,
1996). In light of the difficulties that arise from direct measurements of nitrogen
deposition, models often serve to estimate deposition over large expanses. For the
conterminous United States, total (wet + dry) deposition of nitrogen (as NH4+, NO3
and HNO3) has been estimated to average 0.48 gNm^yr'1, but may exceed 1.7 gNm'
2yr_1 in some areas (Holland, et al., 2005). Nitrogen deposition can provide valuable
nutrient nitrogen to forest ecosystems, but too much may be detrimental to some
ecosystems.
4.2 Nitrogen at the Ecosystem Level: Limitation and Saturation
The increase of anthropogenic nitrogen deposition to ecosystems that has been
observed has led to speculations of the effects of chronic nitrogen inputs to forest
systems over the long term. One of the factors controlling plant growth, and thus
carbon sequestration by forest systems, is the availability of fixed nitrogen as a
nutrient. In most forested areas worldwide, nitrogen is in short supply, and additions
of biologically available nitrogen are expected to enhance plant growth. Conversely,
other areas have an overabundance of biologically available nitrogen and are
considered nitrogen-saturated. Additional inputs of nitrogen to these systems will not
36


result in increases in primary productivity and may lead to the decline of forest health
in such regions (Fenn, et al., 1998).
Nitrogen saturation of a forest system is usually characterized by the removal
of nitrogen as a limiting factor in biological processes and an increase in leaching, the
export of nitrogen through soils to ground and stream waters within the system (Aber,
et al., 1998). Determining the nitrogen status of a forest is not a simple process. Many
factors can contribute to nitrogen leaching, and this character alone cannot be used to
determine the nitrogen status of a watershed. For example, mountainous regions such
as the Colorado Front Range exhibit several inherent characteristics that can influence
rate of nitrogen loss and lead to an incorrect assessment of the system as being
nitrogen saturated. Steep grades encourage nitrogen losses from the system, shorter
growing seasons in alpine regions limit the biotic assimilation of nitrogen and fleeting
plugs of nitrogen available in snowmelt may move quickly through the system, all of
which increase the apparent leaching from the system. These factors should be taken
into consideration before an ecosystem is classified as nitrogen saturated on the basis
of leaching alone (Fenn, et al., 1998). Similarly, the response of a forest to
experimental increases in nitrogen can lead to incorrect diagnoses of nitrogen status.
Although increased amounts of foliar nitrogen in plant tissue have been linked to
increases in primary productivity (Reich, Kloeppel, Ellsworth, & Walters, 1995),
foliar nitrogen content can continue to increase without a concurrent increase in
primary productivity. Thus, this character has recently been decoupled from the
37


determination of forest nitrogen limitation or saturation (Aber, et al., 1998; Bauer, et
al., 2004).
Broad ecosystem changes can be difficult to observe experimentally, since
their effects are cumulative and only become apparent after a number of years. For
example in an attempt to stimulate nitrogen saturation experimentally, Kahl, et al.,
(1993) only observed leaching in the second and third years of a nitrogen amendment
study. Similarly, forest decline and increased mortality were observed as the result of
nitrogen amendment studies after a six year period in two different evergreen stands
(reviewed in Aber, et al., 1998). Nitrogen-driven increases in net primary productivity
have also had a delayed measurable response. In a study of sugar maple-dominated
forests across Michigan, Pregitzer et al. (2008) could not observe a measurable
difference in woody biomass production between nitrogen plots and control plots
until the fourth year of the study. This trend then continued for the next six years until
the end of the study (Pregitzer, et al., 2008). Hypothetically, there is a nonlinear
relationship between nitrogen inputs to forest systems and biomass production over
the long term, with critical tipping points signaling changes as extreme as forest
decline and mortality (Aber, et al., 1998) or dominant species shifts (Fenn, et al.,
1998).
Viewed as chronic additions of nitrogen over an extended period of time, a
forest can be classified by stages on a gradient from nitrogen-limited to nitrogen-
saturated with respect to the use of nitrogen by trees within the forest and nitrogen
38


cycling within the system (Aber, Nadelhoffer, Steudler, & Melillo, 1989). Stage 0 is
the typical state, with nitrogen as a limiting nutritional agent. As nitrogen is added
to the system, primary productivity will initially increase, and the forest would be
considered to be in stage 1. After years of chronic additions of nitrogen to the system
the forest moves into stage 2. Primary productivity peaks and nitrogen leaching
increases. Continued nitrogen inputs will trigger a decline in biomass production, and
by stage 3, the primary production of the forest is lower than it had been in the
nitrogen-limited condition in stage 0. Although much of the literature in North
American systems describing nitrogen-saturated forests refers to those in stage 2,
there have been at least four long-term experiments illustrating nitrogen-driven, stage
3 forest decline through increased mortality or declining primary productivity
(reviewed in Fenn, et al., 1998).
Although many forests and ecosystems in the West are presumed to be
nitrogen-limited, there are hotspots of high nitrogen deposition to be found as well
(Fenn, et al., 2003). Proximity to urban areas as well as geological and climactic
controls on nitrogen transport can influence nitrogen deposition rates in a given
locality, leading to a variability that can be quite localized. The Niwot Ridge in
Colorados Rocky Mountains is a high-elevation site situated east of the continental
divide and near the metropolitan Front Range region, which predispose it to higher
nitrogen deposition (4 8 kgN ha'1 yr'1) than average for the Western region as a
whole (Baron, et al., 2000). Based on the export of nitrogen to surface waters in the
39


midst of the growing season, some workers in the area have concluded that the Niwot
Ridge region of the Colorado Rocky Mountains is approaching stage 2 of nitrogen
saturation (Williams, Baron, Caine, Sommerfeld, & Sanford, 1996). Although
leaching alone may not be strong enough evidence to conclude nitrogen saturation,
examinations of the local biota have certainly supported the notion that the chronic
deposition of nitrogen has begun to change the ecosystems at Niwot. Nitrogen
deposition and amendment studies at Niwot have shown that nitrogen can influence
the metabolic properties of many native plant species, including Engelmann spruce,
as well as effect community composition in the area (Bowman, et al., 2006; Calanni,
et al., 1999; Rueth & Baron, 2002; Tomaszewski & Sievering, 2007). Whether the
forests at the Niwot Ridge (or elsewhere in the Rocky Mountains) are nitrogen
saturated or not remains to be seen.
4.3 Nitrogen and Carbon Sequestration
Carbon sequestration is a metabolic end point in biomass production. A tree
can utilize carbon that is stored in compounds such as starch, but cannot utilize
carbon stored in woody tissue (Millard, et al., 2007). In terms of CCb budgets
worldwide, carbon sequestration is of great interest to those looking to lessen the
climatic impacts of increasing levels of CO2 in the atmosphere. Will nitrogen
deposition influence the capacity of high-altitude forests to sequester carbon?
40


Increased nitrogen supply via deposition in forest systems has resulted in both
increased growth and decreased growth, depending on the site-specific nitrogen
availability, tree species present, as well as the availability of other nutrient
compounds. For example, conifer forests in Norway are thought to be limited by the
availability of nitrogen, and deposition has been linked to an estimated 25% increase
in growth in the region (Solberg, et al., 2004). Contrastingly, spruce forests in
Germany exhibited an initial growth increase in response to nitrogen deposition, but
this was soon counteracted by a general decline in the forest health (e.g. yellow
needles, increased mortality) thought to be due to an imbalance between the
availability of nitrogen and magnesium (Schulze, 1989).
Globally, it seems that nitrogen amendments, which are used as a proxy for
nitrogen deposition, do contribute to increases in forest biomass. In a meta-analysis of
over 300 studies, nitrogen additions increased the biomass production of broadleaved
trees 73.1% and 37.5% in conifers (Xia & Wan, 2008). In another analysis of
temperate and boreal forests in Europe and North America, net ecosystem production
was positively and very strongly (R2 = 0.97) correlated with the amount of nitrogen
available through wet deposition, suggesting that anthropogenic nitrogen is increasing
the capacity of forests worldwide to operate as carbon sinks (Magnani, et al., 2007).
Although these results have been controversial, most authors tend to agree that
nitrogen deposition does play an important role in carbon sequestration in North
America and Europe (e.g. De Schrijver, et al., 2008; de Vries, et al., 2008; Solberg, et
41


al., 2004; Sutton, et al., 2008). While these literature trends may suggest that
additional nitrogen inputs into forest systems will increase their capacity to sequester
carbon, it is important to note that these studies did not capture the effects of chronic
nitrogen inputs over a long time series or the possibility of other nutritional
limitations, such as phosphorus, that forests may encounter under increased growth
scenarios. Still, it is interesting to note that the trend worldwide is still an increase in
growth in response to additional nitrogen inputs.
There have been some long-term studies analyzing the responses of forested
areas to varying levels of nitrogen amendments, and these have shown that nitrogens
role in carbon sequestration is complex (e.g. Hogberg, Fan, Quist, Binkley, & Tamm,
2006; Magill, et al., 2004; McNulty, Boggs, Aber, Rustad, & Magill, 2005; Pregitzer,
et al., 2008). A study analyzing the growth over 30 years of a Scots pine-dominated
forest in Sweden under three nitrogen regimes plus a control plot shows that the
fertilizing effects of nitrogen over time do not always match those observed initially.
All three nitrogen regimes drove rates of biomass accumulation higher than the
control, but without a linear relationship between the amounts of nitrogen applied and
growth. The highest nitrogen regime actually resulted in the lowest increase in
biomass. Another conclusion drawn from this data indicates that the rate of nitrogen
applied over time, rather than the total cumulative amount added to the system, has a
greater effect on growth (Hogberg, et al., 2006). Other studies have shown that
chronic amendments of nitrogen can have a negative effect on biomass production
42


over the long term. In a 14 year study with red spruce in Vermont, both high and low
concentrations of nitrogen resulted in lower biomass production (measured as live
basal area- 40% lower for high nitrogen and 18% lower for low nitrogen sites) than
control sites receiving only ambient nitrogen deposition (McNulty, et al., 2005).
Clearly, more long-term studies at various sites and ecosystems will help to answer
the question of whether or not anthropogenic nitrogen is increasing or decreasing the
global carbon sink.
43


5. Previous Fluorometry Work at Niwot Ridge Forests
The Niwot Ridge is an east-west formation approximately 8 km long located
in the southern edge of the Colorado Rocky Mountains. Since 1921, the University of
Colorado has conducted alpine and subalpine research at Niwot, and long-term
climatic and vegetation monitoring have been ongoing since 1950. Since 1980, the
area has been a part of the Long-Term Ecological Research program of the National
Science Foundation. Although the entire research station is located above 3000 m in
elevation, topography is quite variable in the area, leading to the formation of
numerous microclimates with large differences in deposition rates, climate and
vegetation contained within a relatively small area (Bowman & Seastedt, 2001). The
average elevation at which tree line occurs on the ridge is 3406 m (Komarkova &
Webber, 1978). One of the subalpine (3000 m) climate monitoring sites at Niwot, Cl,
has been continually collecting climatic data since 1952. The total fixed nitrogen (wet
+ dry) deposition at Cl is estimated to be 0.4 0.8 gN m2 yr'1 (Sievering, Kelly,
McConville, Seibold, & Tumipseed, 2001). In contrast, the Soddie site is located at
3345 m elevation, slightly below tree line. Wet deposition is continually monitored at
the site through the National Atmospheric Deposition Program and a series of snow
lysometers capture the chemistry and dynamics of snowmelt. To date, no attempt has
been made to characterize the dry deposition of nitrogen at Soddie. An estimate of the
wet only nitrogen deposition at Soddie (0.401 gN m2yr'') is in accord with the lower
44


end of the estimate at Cl. Although snowmelt plays an important role at Cl in the
early spring, it is assumed that the quantity of snowmelt and constituent nitrogen is
higher at Soddie than at C1 due to the early season plug of nitrogen available in
snowmelt there.
Tomaszewski and Sievering (2007) conducted a study similar to the one
designated as trial one in this document at the Niwot Cl site over a single growing
season in 2004, which was used as a model for this study. Engelmann spruce foliage
was subjected to three treatments: 1) Background- no spray, 2) Control- non-nitrogen
ion matrix spray applied to match precipitation in the area (comparable to C used in
this study) and 3) Nitrogen- a spray comprised of the ion matrix used as a control and
10 mgN l'1 as NH/NCV (comparable to N1 in this study). Chlorophyll fluorometry
measurements and gas exchange measurements were recorded. In shoots over one
year old, the nitrogen treatment increased F\/Fm by 11.5% (p < 0.05), FJFm by 2.8%
(not significant) and decreased NPQ by 11.5% (not significant) over the control and
background treatments. The fluorometry data was supported by gas exchange
measurements which showed that V^, the maximum rate of carboxylation by
Rubisco, was 14.6% greater in nitrogen treated branches (p < 0.05) and JnmXl the
maximum rate of light-saturated electron transport, was 11.2% greater in the nitrogen
treatment (not significant). These data suggest that nitrogen positively affected
photosynthesis at the C1 site, improving photosynthetic efficiency, quantum yield and
actual CO2 incorporation rate while reducing the non-photochemical dissipation of
45


energy as heat (Tomaszewski & Sievering, 2007). In the context of carbon
sequestration, this study would suggest that increasing nitrogen deposited to the Cl
forest may result in increased photosynthesis and biomass production. In fact, the
authors extrapolate this increase to estimate that canopy uptake of deposited nitrogen
may account for an increase in the rate of carbon sequestration on the order of 25 -
135 gC m'2 in eastern forests (Sievering, et al., 2007). A comparative study was
devised to investigate the influence of nitrogen amendments at the Soddie site, which
has more early-season nitrogen available than the lower-elevation Cl site.
46


6. Materials and Methods
6.1 Site Description
Experimental trees were located adjacent to the Soddie meteorological station
of the University of Colorados Mountain Research Station on the Niwot Ridge (40
02 52" N, 105 34' 15" W). Mean elevation of the site is 3345 m on a 10 south-facing
slope (Fig. 6.1). This sub-alpine forest is located just below tree line and is dominated
by Engelmann spruce (Picea engelmannii), the species chosen for this experiment.
Precipitation at the site is monitored by the University of Colorado Mountain
Research Station and averaged 502.2 L yr'1 for 2005 2007. The NO3' and NH/
components of wet deposition are also monitored at the site, and for the period
between 2000 and 2006 these components amounted to an estimated annual input of
0.401 gN m'2yr_l via precipitation at the site. Another source of NO3" and NH4+ rather
unique to tree line sites is seasonal snowmelt. Snow usually covers the Soddie site
from October to June, with the remaining months considered the growing season.
Each spring, the frozen precipitation that has collected over the winter begins to melt,
supplying the site with a seasonal plug of nitrogen at the start of the growing season.
This additional input of NO3' and NH4+ averaged 0.221 gN m2yr'' between 2006 and
2007 (Williams, Seibold, & Chowanski, 2009).
47


FIGURE 6.1 The Location of the Study Site. A map of the elevation of the area surrounding the
study site. The physical location of the study trees is marked with a star. The map was created with
the internet-based custom map tool from the United States Geological Surveys National Map by
choosing the location and the Elevation layer option 1/3 (10 meter) ArcSecond Ned CONUS.
Accessed from http://nmviewogc.cr.usgs.gov on October 12, 2008. The inset map in the upper right
shows the state of Colorado with the location of the study site marked with a star. This map was
created using ESRIs ArcMap 9.3 software.
48
N.CULOr


6.2 Chlorophyll Fluorescence Measurements
Chlorophyll fluorescence was measured and recorded with a PAM-2100
portable fluorometer (Heinz Walz GmbH, Effeltrich, Germany) fitted with the
manufacturers leaf-clip holder (2030-B). The leaf-clip holder maintained a 60 angle
and consistent distance between any given sample and the fiber optic sensor.
Measurements of both light- and dark-adapted samples were conducted in the first 2
hours of direct sunlight in the area (09:00 11:00 hrs). Samples were considered
light-adapted when they had been exposed to full sunlight (wavelengths in the range
of 400 700 nm > 1000 pmol mV) as measured at the shoot with a handheld light
meter (Field Scout Quantum Light Meter, Spectrum Technologies, Plainfield, Dlinois)
for a period greater than 15 minutes. Measurements were then conducted in saturation
pulse mode and the parameters Fm, F0 and F were recorded. In this mode, the
actinic (650 nm) light was switched on and F' was recorded. Next, a saturating pulse
of light (5000 pmol m'V1, 710 nm) was applied for 0.8 s and Fm was determined.
The sample and leaf-clip holder were then covered with an opaque black cloth. After
3 s, the actinic light was switched off and a far-red (735 nm) light was applied for 5.5
s during which F0 was recorded. Prior to the dark-adapted measurements, samples
were covered in opaque cloths for 30 min to fully oxidize the PSII reaction centers,
and darkness was maintained throughout the measurements. In this mode, Fm was
recorded, following a saturating pulse. F0 was determined from the measurements of
49


F0, Fm and Fm, as described in chapter 2.5. From these basic measurements, all other
fluorometry parameters were determined (Table 2.2).
6.3 Amendment Solutions
Three treatment amendment solutions were employed and are referred to as
Control (C), Nitrogen One (Nl) and Nitrogen Three (N3). The C and N1 treatments
were applied in 2005, 2006 and 2008, while the N3 treatment was only added in the
2008 study period. All three treatments consisted of an array of ions at concentrations
found in natural precipitation in the area diluted in deionized water: Ca2+ (0.46 mg 1'
'), Mg2+ (0.05 mg l1), Na+ (0.54 mg l1), K+ (0.07 mg l1), CF (0.09 mg 1') and S042
(0.31 mg l"1) (Tomaszewski & Sievering, 2007). The nitrogen treatments (Nl and N3)
included these ions as well as added nitrogen in the form of NH4NO3 at two different
concentrations: 10 mg f1 in the Nl treatment and 30 mg l"1 in the N3 treatment.
Meteorological records show that this area receives about 0.401 gN m'2 yr'1 through
wet deposition. Dry deposition has not been estimated at Soddie, and estimates of dry
deposition are too site-specific (e.g. relying on tree canopy height) to rely on
measurements obtained in nearby forests with much different canopy characteristics.
As such, dry deposition estimates have been excluded from estimates of nitrogen
availability at the site, resulting in an underestimate of deposited nitrogen. The
treatment concentrations increased the nitrogen load by 10 20% for the Nl
treatment and 30 50%% for the N3 treatment over the nitrogen available from wet
50


deposition plus snowmelt. Each treatment solution was applied directly to the branch
by spraying from a point about 10 cm from the trunk to the apex of the branch.
During the treatments, each branch was enclosed such that the spray was excluded
from the surrounding branches. The treatments were applied during the growing
season at an average rate of one 250 ml application per branch per week. For
example, in 2008, treatments were administered for the period June 24 August 22,
resulting in a total of seven applications (no spray was administered in the last week,
and weather prevented application during another). This corresponds to an additional
17.5 mg of nitrogen per branch in the N1 branch treatments and an additional 52.5 mg
per branch in the N3 branch treatments over the course of the 2008 growing season.
6.4 Trial One: Nitrogen Amendment Effects over Three Years
Three branches on each of three trees were selected that had similar density
and eastward light orientation and marked as C, N1 or N3. In 2005 and 2006, C and
N1 treatments were applied as described, and the trees and branches used in this study
are the same for those two treatments. The N3 treatment was added in the 2008
growing season, following the same procedures. Additionally, three apical shoots of
last years growth were chosen from each treatment branch and marked to ensure
consistent fluorometry measurements throughout the season. Replicating the
51


measurements on three trees reduced the possibility that tree-specific effects would
confound the data.
6.5 Trial Two: Sun- and Shade-Adapted Nitrogen Amendment Effects
In order to assess the differing effect of nitrogen treatments on sun- and
shade- adapted shoots under field conditions, it was necessary to utilize excised
shoots. Conifers have been shown to be resilient to such methods and show no
measurable changes in chlorophyll fluorometry parameters until 12 hours after
excision (Richardson & Berlyn, 2002). Comparable branches were chosen on the
upper, sunlit (eastern) side and lower, shaded (western) side of each of three trees.
Although efforts were made to select sunlit and shaded branches of similar growth,
the shaded branches had noticeably smaller needles than the sunlit branches in all
cases. Periodic samples were trimmed from each treatment branch and three clusters
of needles per excised shoot were subjected to chlorophyll fluorometry. For each
sample, a shoot 6 10 cm in length was moistened with deionized water and cut from
the branch at an angle. Needles were removed from the lower 1 2 cm and the shoot
was immediately placed in an individual tube of deionized water. This tube was then
placed on an ice pack in a dark location for fluorometric analysis in the field within a
few (< 3) hours. After taking dark-adapted measurements, excised shoots were placed
in full sunlight (> 1000 pmol m'V) for at least 15 min prior to taking light-adapted
measurements. During fluorescence measurements, the temperature was monitored to
52


ensure that the shoots were maintained at temperatures similar to those of non-excised
shoots (approximately 20 37 C), but this proved difficult on a number of occasions
under field conditions. Although temperatures of trimmed shoots reached a maximum
of 41.1 C, the higher temperatures did not seem to affect the fluorescence
measurements. This being in line with the findings of Richardson and Berlyn (2002),
the few measurements made under these conditions were retained in the analysis.
Otherwise, fluorometry and amendments were applied as described above.
6.6 Trial Two: Sun- and Shade-Adapted Morphological Characteristics
The morphological differences between shade- and sun-adapted tissues were
characterized by measuring needle length, fresh and dry mass, needle area and needle
density along the shoot. Mass was measured to the closest 0.001 g with a VIC-123
Electronic Precision Scale (Acculab, Edgewood, NY). Needles were dried in a
standard oven at approximately 66C for 48 hours prior to assessment of dry mass.
Needle length and area were assessed by scanning the needles and analyzing their
images with ImageJ software (version 1.38, National Institute of Mental Health,
Bethesda, MD). Needle density along each shoot was assessed by trimming a 3 cm
portion of last years grown of an apical shoot and counting the needles to find a
density of needles per cm branch length.
53


7. Statistical Analysis: Methods and Results
7.1 Data Treatment
For each date, three shoots were recorded from each treatment branch, and
three trees were used for each trial. This gave a total of nine replicated measurements
for each treatment on each date. For trial one, the three years of fluorometry data
were grouped into a single set of 782 individual sets of measurements. For trial two,
there was only a single year of fluorometry data recorded, resulting in 270 individual
sets of measurements. The trial two data set was split into two groups based on
measurement position (upper = sunlit, lower = shaded) for the remainder of statistical
analyses, leaving each group with 135 individual sets of measurements.
Morphological data was only recorded for trial two, with three replicates measured
for each tree and treatment, leading to a total of 54 sets of measurements. All of these
data sets were considered large (n > 30) for statistical purposes. All statistical analysis
was conducted with the Statgraphics Plus 5.0, Professional Edition software package
(StatPoint Technologies, Warrenton, VA). A significance level, a, of 0.05 was a
priori and used in all tests.
For the fluorometry data in both trials, the raw data for parameters F0, Fm, F and F'm
(Table 2.1) were recorded in the field and used to calculate the parameters used in this
study (Table 2.2). After computing the calculated parameters, mathematical and
instrumental constraints (Table 2.3) were applied to all parameters
54


in both trials, and both measured and calculated variables that were outside of these
constraints were removed. The percentage of data points removed for each parameter
in both trials is reported in Table 7.1.
Although non-parametric analytical methods are not sensitive to outliers, a
small number of data points were well outside of the spread of the majority of the
data. Rather than applying an arbitrary technique for outlier removal, the outliers
were removed on a conservative basis by visually examining the data and removing
any data points outside of a threshold set at four standard deviations from the mean.
By conservatively choosing a threshold at four standard deviations, the natural
skewness exhibited by most of the data was preserved, while anomalous
measurements were excluded. The percentage of data points removed as outliers by
this method was low. The percentages are reported in Table 7.2. In the morphological
data sets from trial two, no data points were removed as outliers.
55


Table 7.1. Percent of fluorometry variables removed due to
mathematical or instrumental constraints in trials one and two.
Trial One Trial Two
Parameter Raw N Percent Removed Raw N Percent Removed
Fm 760 3.3% 270 0.0%
Fm 702 18.5% 255 5.6%
Fo 680 15.9% 255 5.6%
Fv 760 3.3% 270 0.0%
F\ 680 12.9% 255 5.6%
F r ^PSII 702 15.7% 255 5.6%
Fv
Fm p' 760 3.3% 270 0.0%
rv K 680 16.0% 254 5.9%
1 -qP 680 18.4% 249 7.8%
qP 680 18.4% 249 7.8%
NPQ 680 15.9% 255 5.6%
qO 680 15.9% 255 5.6%
qCN 680 15.9% 255 5.6%
qN 680 15.9% 255 5.6%
qTQ 680 6.3% 266 1.5%
56


Table 7.2. Percent of fluorometry outliers removed due to exceeding a
four standard deviation threshold.
Trial One Trial Two-Upper Trial Two-Lower
Parameter Raw Percent Raw Percent Raw Percent
N Removed N Removed N Removed
Dpsn 592 0.0% 124 0.0% 131 0.0%

Fm F 735 1.4% 135 2.2% 135 0.0%
rv K 571 0.0% 124 0.0% 130 0.0%
1 qP 554 120 0.0% 129 0.0%
qP 554 0.0% 124 0.0% 129 0.0%
NPQ 572 0.7% 124 0.0% 131 0.8%
qO 572 0.0% 124 0.0% 131 0.0%
qCN 572 0.0% 124 0.0% 131 0.0%
qN 572 0.0% 124 0.0% 131 0.0%
qTQ. 637 0.0% 132 0.8% 134 0.0%
7.2 Statistical Analysis and Results
In this study, each parameter of interest (e.g. F/Fm) is the response variable
(T), while the treatment factor (e.g. control, N2 or N3) is the explanatory variable (X).
Homoscedasticity was tested using Levenes Test, which is considered robust to data
that is not normally distributed as long as there is a large data set (Charway & Bailer,
2007). Table 7.3 exhibits the results of the Levenes Test. In trial one, the parameters
NPQ and qO were found to be heteroscedastic and were excluded from further
analysis. In trial two, qP and 1 -qP were found to be heteroscedastic and were
57


excluded from further analysis (Table 7.3). All morphological data for trial two were
found to be homoscedastic (data not shown).
Table 7.3. Results of the Levenes Tests for homoscedasticity
on fluorometry data.
Parameter Trial One p-value Trial Two-Uoper p-value Trial Two-Lower p-value
^PSII 0.1215 0.2621 0.6457

Fm 0.4301 0.1601 0.7387
K
K 0.2352 0.3820 0.4888
1 qP 0.4250 0.0145* 0.9332
qP 0.4250 0.0145* 0.9332
NPQ 0.0017* 0.2708 0.5973
qO 0.0040* 0.8314 0.8856
qCN 0.1065 0.3826 0.7913
qN 0.1800 0.4537 0.6309
qTQ 0.4422 0.9290 0.9915
indicates p-values < 0.05. These parameters did not pass the lest and were
excluded from subsequent analysis of variance.
To test for normality, skewness and kurtosis were analyzed with Z-tests.
Results of the skewness test for normality are shown in Table 7.4. Of the parameters
analyzed for trial one, only qP and 1 -qP were not significantly skewed. The skewness
results in trial two were more mixed (Table 7.4). The results of the kurtosis test for
normality are shown in Table 7.5. None of the fluorometry data in trial one passed
58


both tests, while Opsn, qP, 1-qP and qCN in trial two did pass in both positions. Due
to the general non-normality of the fluorometry data, the Kruskal-Wallis non-
parametric test was chosen for analysis. The morphological data in trial two passed all
tests for normality (data not shown), and thus subsequent statistical analysis was
carried out in the usual fashion using One-Way ANOVA F-tests.
Table 7.4. Results from the Z-tests for normality of fluorometry data, part
one: skewness.
Parameter Trial One Z-Score p-value Trial Two-Upper Z-Score p-value Trial Two-Lower Z- p-value Score
^PSII 3.982 <0.0005* -1.722 0.0851 -0.580 0.5620
K
Fm p -2.915 0.0036* -3.429 0.0006* -3.215 0.0013*
rv K -2.915 0.0036* -2.774 0.0055* -1.950 0.0512
1 qP -0.0386 0.9692 -1.837 0.0663 -0.270 0.7879
qP -0.0386 0.9692 -1.837 0.0663 -0.2690 0.7879
NPQ 6.492 <0.0005* 5.013 <0.0005* 3.433 0.0006*
qO 3.167 0.0015* 2.508 0.0121* 1.992 0.0461*
qCN -3.032 0.0024* -0.771 0.4406 -1.774 0.0760
qN -3.966 <0.0005* -1.229 0.2190 -2.1814 0.0292*
-7.498 <0.0005* -3.654 <0.0005* -2.264 0.0424*
Positive Z-score values indicate distributions that are skewed to the right, while a negative
Z-score value indicates a distribution that is skewed to the left. indicates p-values < 0.05.
These parameters did not pass the first portion of normality assessment.
59


Table 7.5. Results from the Z-tests for normality of fluorometry data, part two:
kurtosis.
Parameter Trial One Z-Score p-value Trial Two-Upper Z-Score p-value Trial Two-Lower Z-Score p-value
^PSII -1.299 0.1941 0.0475 0.9621 -2.134 0.3287
F,
Fm -0.9643 0.3349 2.544 0.0110* 1.882 0.0598
K
K -0.9643 0.3349 1.995 0.0460* 0.1661 0.8681
1 -qP -7.549 <0.0005* 1.035 0.3005 -1.907 0.0565
qP -7.549 <0.0005* 1.035 0.3005 -1.907 0.0565
NPQ 3.851 <0.0005* 5.609 <0.0005* 2.445 0.0145*
qO -2.425 0.0153* 1.337 0.1813 0.2618 0.7935
qCN -5.503 <0.0005* -1.287 0.1979 -0.5857 0.5581
qN -3.751 <0.0005* -1.249 0.2117 -0.1775 0.8591
QTQ 4.866 <0.0005* 3.016 0.0026* 0.3412 0.7330
Positive Z-score values indicate distributions that are leptokurtic, while a negative Z-score
value indicates a distribution that is platykurtic. indicates p-values < 0.05. These parameters
did not pass the second portion of normality assessment.___________________________________________
For data that are non-normally distributed, the median may be a better
estimator of the central tendency of the data than the mean (Norman & Streiner,
2008). The box and whisker plots1 for trial one are shown in Figures 7.1-7.10, while
those for trial two are shown in Figures 7.11-7.30. Trends apparent in these graphs
1 In these plots, the grey boxes represent the data contained within the upper and lower quartiles. The
whiskers extending from each end represent the minimum and maximum end points within the typical
range of the data. This typical range was calculated by multiplying 1.5 times the interquartile distance,
which is found by subtracting the value of the lower quartile from that for the upper quartile. The
points designated outside of this range are technically outliers, but they did not exceed the limits in
place for outlier removal. The line drawn between the upper and lower quartile within the box
designates the median.
60


will be discussed below. These box and whisker plots allow for a visual comparison
of both the spread and central tendency of the fluorometry data without relying on
distribution assumptions.

PSII
F / F
v m
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
FIGURE 7.1 Values of PS[[ Observed in
Trial One. The median value of the control
(0.3617) was higher than those of the nitrogen
treatments (N1 = 0.2943, N3 = 0.3318),
indicating that control branches were more
efficient at carrying out photosynthesis than
nitrogen-treated branches. The median values
were found to be significantly different
(Kruskal-Wallis p < 0.0005).
FIGURE 7.2 Values of Fx/Fm Observed in
Trial One. The median value of the control
(0.7720) was higher than those of the
nitrogen treatments (N1 = 0.7694, N3 =
0.7420), indicating that photosynthetic
potential decreases with increasing nitrogen.
The median values were found to be
significantly different (Kruskal-Wallis p =
0.0487).
61


F' IF
1-qP
0.9
0.8
Control N1 N3
FIGURE 7.3 Values of FJFm Observed in
Trial One. The median value of the control
(0.6033) was higher than those of the nitrogen
treatments (N1 = 0.5758, N3 = 0.5510),
indicating that effective photosynthetic
efficiency decreases with increasing nitrogen.
The median values were found to be
significantly different (Kruskal-Wallis p <
0.0005).
NPQ
7
FIGURE 7.5 Values of NPQ Observed in
Trial One. The median values of the nitrogen
treatments (N3 = 1.2603, N1 = 1.2424) were
higher than that of the control (1.0624).
Although the data were hetcroscedastic and
significance could not be assessed, the pattern
of N3 > N1 > C indicates that non-
photochemical quenching may increase with
increasing nitrogen.
1
Control N1 N3
FIGURE 7.4 Values of 1 -qP Observed in
Trial One. The median value of the N1
treatment (0.4694) was higher than those of
the control (0.3902) and the higher nitrogen
treatment (N3 = 0.3872). The pattern
exhibited by the data is of uncertain meaning.
The median values were found to be
significantly different (Kruskal-Wallis p =
0.0051).
qO
Control N1 N3
FIGURE 7.6 Values of qO Observed in Trial
One. The median values of the nitrogen
treatments (N3 = 0.2468, N1 = 0.2296) were
higher than that of the control (0.1949).
Although the data were heteroscedastic and
significance could not be assessed, the pattern
of N3 > N1 > C indicates that the relative
change of minimal fluorescence may increase
with increasing nitrogen.
62


qCN
qN
FIGURE 7.7 Values of qCN Observed in
Trial One. The median values of the nitrogen
treatments (N3 = 0.5614, N1 = 0.5560) were
higher than that of the control (0.5151).
Although this suggests that complete non-
photochemical quenching increases with
increasing nitrogen, the median values were
not found to be significantly different
(Kruskal-Wallis p = 0.0614).
qP
FIGURE 7.8 Values of qN Observed in Trial
One. The median values of the nitrogen
treatments (N3 = 0.6725, N1 = 0.6625) were
higher than that of the control (0.6076),
indicating that non-photochemical quenching
increases with increasing nitrogen. The
median values were found to be significantly
different (Kruskal-Wallis p = 0.0452).
qTQ
FIGURE 7.9 Values of qP Observed in Trial
One. The median value of the N3 treatment
(0.6128) was higher than that of the control
(0.6098) and the lower nitrogen treatment (N1
= 0.5306). The pattern exhibited by the data is
of uncertain meaning. The median values
were found to be significantly different
(Kruskal-Wallis p = 0.0051).
FIGURE 7.10 Values of qTQ Observed in
Trial One. The median values of all three
treatments were roughly equivalent (C =
0.6991, N1 = 0.6904, N3 = 0.6956),
indicating no apparent trend. The median
values were not found to be significantly
different (Kruskal-Wallis p = 0.8668).
63


Upper d>
Lower ' i
FIGURE 7.11 Values of PSIi Observed in
Upper Branches in Trial Two. The median
values of the nitrogen treatments (N1 =
0.4584, N3 = 0.4455) were higher than that of
the control (0.4302). Although this suggests
that nitrogen-treated branches were more
efficient at carrying out photosynthesis than
control branches, the median values were not
found to be significantly different (Kruskal-
Wallisp = 0.5179).
Upper F /F
v m
FIGURE 7.12 Values of Lower Branches in Trial Two. The median
value of the control treatment (0.3913) was
higher than those of the nitrogen treatments
(N1 = 0.3879, N3 = 0.3709). Although this
suggests that nitrogen-treated branches were
less efficient at carrying out photosynthesis
than control branches, the median values were
not found to be significantly different
(Kruskal-Wallis p = 0.7721).
Lower F /F
09
0.85
0.8
0.75
07
0.65
0.6
0.55
0.5
O
Control N1 N3
FIGURE 7.13 Values of F/Fm Observed in
Upper Branches in Trial Two. The median
values of the nitrogen treatments (N1 =
0.7782, N3 = 0.7772) were higher than that of
the control treatment (0.7733). Although this
suggests that photosynthetic potential
increases with added nitrogen, the median
values were not found to be significantly
different (Kruskal-Wallis p = 0.7981).
FIGURE 7.14 Values of F,/Fm Observed in
Lower Branches in Trial Two. The median
value of the control treatment (0.7803) was
higher than that of the nitrogen treatments
(N3 = 0.7795, N1 = 0.7788). Although this
suggests that photosynthetic potential
decreases with added nitrogen, the median
values were not found to be significantly
different (Kruskal-Wallis p = 0.6435).
64


FIGURE 7.15 Values of F\/Fm Observed in
Upper Branches in Trial Two. The median
values were, in decreasing order: N1 =
0.6682, C = 0.6395 and N3 = 0.6212. The
pattern exhibited by the data is of uncertain
meaning. The median values were not found
to be significantly different (Kruskal-Wallis p
= 0.3724).
Upper 1-qP
Control N1 N3
FIGURE 7.17 Values of 1-qP Observed in
Upper Branches in Trial Two. The median
value of the control treatment (0.3515) was
higher than that of the nitrogen treatments
(N1 = 0.3004, N3 = 0.2895). Although the
data were heteroscedastic and significance
could not be assessed, the pattern of C > N1 >
N3 suggests that the reduction state of QA
may decrease with increasing nitrogen.
FIGURE 7.16 Values of F/Fm Observed in
Lower Branches in Trial Two. The median
value of the control treatment (0.6325) was
higher than that of the nitrogen treatments
(N1 = 0.6271, N3 = 0.5950. Although this
suggests that effective photosynthelic
efficiency decreases with added nitrogen, the
median values were not found to be
significantly different (Kruskal-Wallis p =
0.4147).
Lower 1-qP
0.8 ( ....
0 2 -
Control N1 N3
FIGURE 7.18 Values of 1 -qP Observed in
Lower Branches in Trial Two. The median
values were, in decreasing order: N1 =
0.4090, C = 0.3841 and N3 = 0.3639. The
pattern exhibited by the data is of uncertain
meaning. The median values were not found
to be significantly different (Kruskal-Wallis p
= 0.5337).
65


Upper NPQ
Lower NPQ
FIGURE 7.19 Values of NPQ Observed in
Upper Branches in Trial Two. The median
values were, in decreasing order: N3 =
0.9417, C = 0.8811 and N1 = 0.7446. The
pattern exhibited by the data is of uncertain
meaning. The median values were not found
to be significantly different (Kruskal-Wallis p
= 0.4997).
Upper qO
0
Control N1 N3
FIGURE 7.20 Values of NPQ Observed in
Lower Branches in Trial Two. The median
values of the nitrogen treatments (N3 =
1.3535, N1 = 1.1372) were higher than that
of the control (1.0609), suggesting that non-
photochemical quenching may increase with
increasing nitrogen. The median values were
not found to be significantly different
(Kruskal-Wallis p = 0.2249).
Lower qO
08--------------,------------,-----------
Control N1 N3
FIGURE 7.21 Values of qO Observed in
Upper Branches in Trial Two. The median
values were, in decreasing order: N3 =
0.1697, C = 0.1505 and N1 =0.1424. The
pattern exhibited by the data is of uncertain
meaning. The median values were not found
to be significantly different (Kruskal-Wallis p
= 0.5164).
FIGURE 7.22 Values of qO Observed in
Lower Branches in Trial Two. The median
values of the nitrogen treatments (N3 =
0.2353, N1 = 0.2036) were higher than that of
the control (0.2008), suggesting that the
relative change of minimal fluorescence may
increase with increasing nitrogen. The median
values were not found to be significantly
different (Kruskal-Wallis p = 0.2740).
66


Upper qCN
Lower qCN
Control N1 N3
FIGURE 7.23 Values of qCN Observed in
Upper Branches in Trial Two. The median
values were, in decreasing order: N3 =
0.4850, C = 0.4684 and N1 = 0.4268. The
pattern exhibited by the data is of uncertain
meaning. The median values were not found
to be significantly different (Kruskal-Wallis p
= 0.4979).
Upper qN
0.8 -...
Control N1 N3
FIGURE 7.25 Values of qN Observed in
Upper Branches in Trial Two. The median
values were, in decreasing order: N3 =
0.5715, C = 0.5495 and N1 = 0.5132. The
pattern exhibited by the data is of uncertain
meaning. The median values were not found
to be significantly different (Kruskal-Wallis p
= 0.5732).
Control N1 N3
FIGURE 7.24 Values of qCN Observed in
Lower Branches in Trial Two. The median
values of the nitrogen treatments (N3 =
0.5867, N1 = 0.5321) were higher than that of
the control (0.5148), suggesting that complete
non-photochemical quenching may increase
with increasing nitrogen. The median values
were not found to be significantly different
(Kruskal-Wallis p 0.1606).
Lower qN
0 8
Control N1 N3
FIGURE 7.26 Values of qN Observed in
Lower Branches in Trial Two. The median
values of the nitrogen treatments (N3 =
0.6986, Nl = 0.6333) were higher than that of
the control (0.6164), suggesting that non-
photochemical quenching may increase with
increasing nitrogen. The median values were
not found to be significantly different
(Kruskal-Wallis p = 0.1907).
67


Upper qP
Lower qP
FIGURE 7.27 Values of qP Observed in
Upper Branches in Trial Two. The median
values of the nitrogen treatments (N3 =
0.7105, N1 = 0.6996) were higher than that of
the control treatment (0.6485). Although the
data were heteroscedastic and significance
could not be assessed, the pattern of N3 > N1
> C suggests that the photochemical
quenching may increase with increasing
nitrogen.
Upper qTQ
Control N1 N3
FIGURE 7.28 Values of qP Observed in
Lower Branches in Trial Two. The median
values were, in decreasing order: N3 =
0.6361,0 = 0.6159 andNl =0.5910. The
pattern exhibited by the data is of uncertain
meaning. The median values were not found
to be significantly different (Kruskal-Wallis p
= 0.5337).
Lower qTQ
1 -------------,------------
02 -..............-----........
Control N1 N3
FIGURE 7.29 Values of qTQ Observed in
Upper Branches in Trial Two. The median
values of the nitrogen treatments (N3 =
0.7136, N1 = 0.7008) were higher than that of
the control (0.6864), suggesting that total
quenching may increase with increasing
nitrogen. The median values were not found
to be significantly different (Kruskal-Wallis p
= 0.6829).
FIGURE 7.30 Values of qTQ Observed in
Lower Branches in Trial Two. The median
values of the nitrogen treatments (N3 =
0.7435, N1 = 0.7195) were higher than that of
the control (0.7037), suggesting that total
quenching may increase with increasing
nitrogen. The median values were not found
to be significantly different (Kruskal-Wallis p
= 0.3461).
68


In order to assess whether or not each fluorometry parameter expressed
differences between treatment groups, a distribution-free test, the Kruskal-Wallis test,
was used. Because the Kruskal-Wallis test is sensitive to departures from
homoscedasticity (Bradley, 1968), any parameters that did not pass the tests for
homoscedasticity were excluded from analysis. The Kruskal-Wallis tests were
conducted in Statgraphics with the parameter (e.g. /\/Fm) as the dependent variable
and treatment as the factor. The results of the Kruskal-Wallis tests for each applicable
fluorometry parameter in trials one and two are summarized in Table 7.6. While the
three-year compiled data set in trial one showed that a number of parameters were
sensitive to treatments, none of these same parameters exhibited significant
differences due to treatment in the single-year position data sets in trial two.
The analysis of the morphological data set for trial two was conducted with
One-Way ANOVA F-Tests with character (e.g. needle density) as the response
variable and either treatment or position as the explanatory factor. No characteristic
exhibited any statistical significance due to treatment (Table 7.7), however there were
differences due to position in all characters tested, with the exception of needle
density (Table 7.8). To test for the interaction between the treatment and position
factors, Two-Way ANOVAQ F-Tests were conducted using character as the
dependent variable and both treatment and position as the independent variables. The
results of these analyses did not show any significant interactions (data not shown).
69


The means for each character by treatment and position are shown in Figures 7.31-
7.35.
Table 7.6. Results from Kruskal-Wallis One-Way ANOVA
testing of fluorometry data.
Parameter Trial One p-value Trial Two-Upper p-value Trial Two-Lower p-value
^PSII <0.0005* 0.5179 0.7721
Fv
Fm 0.0487* 0.7981 0.6435
K
Fm <0.0005* 0.3724 0.4147
1 -qP 0.0051* NA 0.5337
qP 0.0051* NA 0.5337
NPQ NA 0.4997 0.2249
qO NA 0.5164 0.2740
qCN 0.0614 0.4979 0.1606
qN 0.0452* 0.5732 0.1907
QTQ 0.8668 0.6829 0.3461
indicates p-values < 0.05, indicating a significant difference between the
treatment groups. NA indicates that these parameters did not pass the
previous tests for homoscedasticity, invalidating the Kruskal-Wallis test.
70


Table 7.7. Results from One-Way ANOVA
testing of morphological data against treatment.
Character Trial Two-UDDer p-value Trial Two-Lower p-value
Needles cm'1 0.8929 0.5131
Length (mm) 0.3298 0.6718
Area (mm2) 0.8400 0.8723
Fresh Mass (g) 0.1086 0.3964
Dry Mass (g) 0.3660 0.4068
Values for the dependent variables, shown in the character
column, were analyzed for variance due to the experimental
treatment classes: C, N1 and N3. No character exhibited
statistical significance due to treatment at an a = 0.05.
Table 7.8. Results from One-Way ANOVA testing of morphological
data against branch position.
Character Trial Two-Control p-value Trial Two-Nl p-value Trial Two-N3 p-value
Needles cm'1 0.9349 0.6623 0.4039
Length (mm) 0.0088* 0.01623* 0.0002*
Area (mm'2) 0.0038* 0.0013* <0.00005*
Fresh Mass (g) 0.0005* 0.0004* <0.00005*
Dry Mass (g) 0.0003* 0.0039* <0.00005*
Values for the dependent variables, shown in the character column, were analyzed for
variance due to branch position: Upper (sun-adapted) and Lower (shade-adapted). An
* indicates p-values < 0.05, indicating a significant difference between the upper and
lower braches within each treatment group.
71


E
£
.c
c
14
12
10
8
6
4
Upper
| | Lower
Needle Length
Control
N1
N3
FIGURE 7.31 Needle Density in Trial Two.
The mean number of needles on a shoot per
centimeter + 1 SE is shown for each position
and treatment group. The data showed no
significant differences due to treatment
(Upper: p = 0.8929, Lower: p = 0.5131), or to
position (C: p = 0.9349, N1: p = 0.6623, N3:
p = 0.4039).
FIGURE 7.32 Needle Length in Trial Two.
The mean needle length + 1 SE is shown for
each position and treatment group. While not
significant (p = 0.3298), length in the upper
branches increases with increasing nitrogen.
No such pattern was observed in the lower
branches (p = 0.6718). All treatments were
significantly different due to position (C: p =
0.0088, Nl: p = 0.01623, N3: p = 0.0002).
0.012
0.01
0.008
TO
"> 0.006

0.004
0.002
0
upper
| | Lower
Needle Fresh Mass
Control
N1
N3
FIGURE 7.33 Needle Area in Trial Two. The
mean needle area + 1 SE is shown for each
position and treatment group. While not
significant (p = 0.8400), area in the upper
branches decreases with increasing nitrogen.
No such pattern was observed in the lower
branches (p = 0.8723). All treatments were
significantly different due to position (C: p =
0.0038, Nl: p = 0.0013, N3: p < 0.00005).
FIGURE 7.34 Needle Fresh Mass in Trial
Two. The mean fresh needle mass + 1 SE is
shown for each position and treatment group.
While not significant (p = 0.3964), mass in
the lower branches increases with increasing
nitrogen. No such pattern was observed in the
upper branches (p = 0.1086). All treatments
were significantly different due to position
(C: p = 0.0005, Nl: p = 0.0004, N3: p <
0.00005).
72


0.007
0.006
0.005
m 0.004
M
§ 0.003
0.002
0.001
0
upper
| | Lower
Needle Dry Mass
Control N1
N3
FIGURE 7.35 Needle Dry Mass in Trial
Two. The mean dry needle mass + 1 SE is
shown for each position and treatment group.
The data showed no significant differences
due to treatment (Upper: p = 0.3660, Lower: p
= 0.4068). All treatments were significantly
different due to position (C: p = 0.0003, N1: p
= 0.0039, N3: p < 0.00005).
73


8. Discussion of Results
8.1 Trial One: Nitrogens Effects on Photosynthetic Efficiency
The three most common parameters used to report photosynthetic efficiency
are Opsn, F^/Fm and FJFm, all of which were significantly (p < 0.0005, p = 0.0487
and p < 0.0005, respectively) different due to nitrogen treatments in this study (Table
7.6). Plant tissues with higher values in these three parameters are more capable of
efficient photosynthesis than plant tissues with lower values. Higher values for Opsn,
F-JFm and F'JF'm would be expected if nitrogen were to increase carbon
sequestration by improving photosynthetic efficiency at this site. In this case, both the
N1 and N3 nitrogen treatments resulted in tissues that had values of OPSh, FJFm and
Fv/Fm lower than control tissues (Figs. 7.1-7.3). This suggests that the control
branches were more capable of photosynthesis than branches exposed to nitrogen
amendment treatments.
Since values of OPSii have been linearly correlated with rates of CO2
assimilation experimentally, OpSn has frequently been used as an assessment of the
photosynthetic capacity of plant tissue (Baker, 2008). In this experiment, the two
nitrogen treatments seemed to depress d>psn when compared to the level seen in the
control group (Fig. 7.1). Although it would be rash to conclude that the nitrogen
74


branches were carrying out less photosynthesis than the control branches, results from
this experiment show that they were less efficient at utilizing absorbed energy for
photosynthesis than the branches that did not receive nitrogen amendments. Other
fluorometry data within this study support these results.
The parameter FJFm captures the maximum photosynthetic capacity of plant
tissue based on the entire suite of reaction centers being in the oxidized state and thus
prepared to receive electrons that can be used to carry out photosynthesis. In trial one,
the median values for FJFm were 0.772, 0.769 and 0.742 for the control, N1 and N3
treatments, respectively (Fig 7.2). Clearly, none of the values for FJFm approached
the commonly reported maximum value of 0.83, indicating that the trees in this
region are under most likely some sort of stress (Rohacek, 2002). This is not
unexpected as the harsh environmental conditions of an even lower elevation site
(3050 m) at the Niwot Ridge have been well-documented and may be responsible for
a lower capacity to sequester carbon than predicted by models (Monson, et al., 2002).
In this study, the F,/Fm values exhibited a pattern of C > N1 > N3. This trend of
decreasing F/Fm with increasing nitrogen dose suggests that nitrogen may suppress
the maximum photosynthetic capacity in these trees. Often, lower values of F/Fm
sustained over a long period of time are indicative of either chronic or dynamic
photoinhibition and suggest an increase in non-photochemical processes (Maxwell &
Johnson, 2000). It is possible that increasing the amount of biologically available
nitrogen at this site caused increased photoinhibition. Although the design of the
75


experiment does not allow for differentiating between chronic and dynamic
photoinhibition, the sum of the results and the nature of nitrogens effect on light
harvesting suggest that the nitrogen amended tissues at this site may be chronically
inhibited.
In this study, values of F/Pm followed the same pattern of C > N1 > N3 as
seen in the FJFm values (Fig. 7.3). The actual capacity for photosynthesis as
measured in actinic light was lower in branches receiving nitrogen amendments than
in control branches. These results also support the conclusion that nitrogen
amendments negatively affected the photosynthetic capacity of the trees at this site.
Fv/Fm captures the actual efficiency of photosynthesis in actinic light, and is
typically inversely related to NPQ and qN (Adams & Demmig-Adams, 2004).
Interpreting the results of the parameters qP and 1-qP based on nitrogen
effects is more complicated. Although there were significant differences between
treatments (p = 0.0051), there is no easily discemable trend. The control treatment
exhibited qP on par with N3, but much greater than N1 (Fig. 7.9). It may be that the
effects of chronic nitrogen applications are not apparent within the first season, and
the N3 branches would follow the lower trend seen in the N1 branches if these
treatments were to continue over two more seasons.
76


8.2 Trial One: Nitrogens Effects on Non-Photochemical Processes
The capacity for photosynthesis and thus carbon sequestration can be greatly
influenced by photoinhibition and the non-photochemical processes by which a plant
deals with energy that is excess of that which can be used in photosynthesis. It has
been suggested that nitrogen amendments can increase the light-harvesting capacity
of plant tissue (Millard, et al., 2007; Posch, et al., 2008). If this increased capacity to
absorb energy is not coupled with an increased capacity for photosynthesis, the
excess energy must be dealt with in some manner. If it can be dissipated harmlessly
as heat, no damage will result and photosynthesis can continue. If not, PS II may be
damaged and photosynthesis will be reduced. The parameters NPQ, qO, qCN, qN and
qTQ all speak to these non-photochemical dissipation processes.
The results of trial one indicate that nitrogen increased non-photochemical
processes at this site, although this experiment could not differentiate between
dynamic and chronic photoinhibition. Values for the parameter qN in trial one
significantly (p = 0.0452) varied by treatment, increasing in value with increasing
nitrogen (N3 > N1 > C) (Fig. 7.8). Two of the non-photochemical process parameters,
NPQ and qO were not homoscedastic and were therefore excluded from ANOVA
tests. Still, they exhibit the same N3 > N1 > C pattern seen in other non-
photochemical process parameters within this study (Figs. 7.5 and 7.6). This suggests
that nitrogen amendments may aid the tissues in dissipating energy in excess of that
77


which can be utilized in photosynthesis. The parameter qCN captures the total non-
photochemical energy-dissipation taking place in PSII. Although there was no
significant difference due to nitrogen treatments in qCN, the non-significant (p =
0.0614) trend of N3 > N1 > C also supports this conclusion. The parameter qTQ
captures the reduction in fluorescence from the dark-adapted maximum (Fm) to that
observed at steady state in the light (F) due to both the photochemical and non-
photochemical quenching. In trial one, there was no significant difference in qTQ due
to nitrogen treatment (p = 0.8668). Like the qP and 1 -qP values, there was no
discemable pattern in qTQ between nitrogen treatments (Fig. 7.10).
8.3 Trial Two: Nitrogens Effects on Sun- and Shade-Adapted Tissues
8.3.1 Fluorometry
Although none of the fluorometry parameters measured in trial two exhibited
any significant differences due to nitrogen treatments (Table 7.6), some interesting
trends can be addressed. The potential effects of nitrogen on photosynthetic
performance appear to be different for sun- and shade-adapted tissues. In the
parameters that capture information about photosynthetic efficiency, the sun-adapted,
nitrogen amended branches (N1 and N3) had higher values for Opsn and FJFm than
the sun-adapted control branches (Figs. 7.11 and 7.13). In the shade-adapted tissues,
this pattern was reversed, with the control branches higher than the N1 and N3
branches (Figs. 7.12 and 7.14). Interestingly, this differential is not seen in the
78


performance of the non-photochemical parameters. In both the sun- and shade-
adapted branches, the values for the highest nitrogen (N3) treatment were greater than
those for the control for NPQ, qO, qCN and qN. In the sun-adapted tissues, N1 scored
lower than either the control or the N3 treatments in all cases (Figs. 7.19-7.26). Due
to the limitations of the experimental design, this anomaly was not addressed within
this study. The pattern of N3 > N1 > C for total quenching of fluorescence, qTQ, was
the same for sun- and shade-adapted branches (Figs. 7.29 and 7.30). Overall, all
values were higher for NPQ, qO, qCN, qN and qTQ in the shade-adapted branches
than in the sun-adapted branches, suggesting an increased capacity for dissipating
energy in the shade-adapted branches. The fluormetric results of trial two were not
significant, but certain patterns seemed consistent. Perhaps through a long-term
application of this experimental design or an increased sample size, the true pattern
could be discerned.
8.3.2 Morphological Characteristics
Although none of the morphological characters measured exhibited any
significant differences due to nitrogen treatments (Table 7.7), a number of these
properties were significantly different between the sun-adapted (upper) branches and
the shade-adapted (lower) branches (Table 7.8). Some of the properties that exhibited
differences due to position also showed trends in their responses to nitrogen. For
example, needle length in the upper branches increased with increasing nitrogen (N3
79


> N1 > C), while no discemable pattern was seen in the needle length in the lower
branches (Fig. 7.32). Similarly, the upper branches exhibited a greater needle area in
the control group than in either nitrogen treatment, but area was similar across all
treatments in the lower branches (Fig. 7.33). In the mass of fresh needles, the lower
branches were heavier with increasing nitrogen (N3 > N1 > C), while the upper
branches showed no such pattern (Fig. 7.34). In the dry mass measurements, the
pattern seen in the lower branches breaks down, with the heaviest needles seen in the
N3 treatment, and the C and N1 masses approximately equal (Fig 7.35). There was no
pattern observed due to nitrogen treatments within the upper branches (Fig. 7.35).
The morphological characters were only measured in the final year of the study,
2008. These patterns might become more obvious if carried out over a longer period
of time.
These results are both in agreement with and contrast to a similar 2008 study
that examined the differential effects of nitrogen amendments on the morphological
characteristics of upper, mid and lower branches of Monterey pines. In needle length,
this study is in agreement with the findings of Posch et al. (2008), who found that
nitrogen-amended branches were longer only in the more sun-adapted upper and mid
regions of the trees, with no effect observed in the lower, shade-adapted branches. In
their dry mass measurements, however, they found that, while nitrogen again
increased the mass of needles in the upper and mid regions, it seemed to reduce the
dry mass in lower tissues (Posch, et al., 2008). In this present study, the highest
80


nitrogen treatment correlated with the heaviest needles in both the upper and lower
branches. Although their study produced no significant results, Millard et al. (2007)
also had similar results, with the highest nitrogen supplement correlated with the
heaviest needles. Other studies have shown that the effects of nitrogen on
morphological characteristics of conifers can be complex and contradictory. For
example, nitrogen amendments can increase leaf and total biomass production in both
shade- and sun-adapted Engelmann spruce (McKinnon & Mitchell, 2003) or it can
decrease the specific leaf area (area per unit mass) in sun-adapted tissue, but not in
shade-adapted tissue of Korean pine (Makoto & Koike, 2007). Most of the studies on
morphological characteristics of evergreens are short-term studies capturing
measurements over a single season. In order to ascertain the effects that nitrogen will
have on conifers, more long-term investigations are needed.
81


9. Comparison between Two Niwot Ridge Sites
As described in Chapter 5, Tomaszewski and Sievering (2007) conducted a
similar study at a subalpine site, C1 on the Niwot Ridge in 2004. Compared to the
tree line site (Soddie) in this study, Cl is at a lower elevation and receives less
nitrogen as snowmelt. Table 9.1 displays a subset of the subalpine Cl fluorometry
data alongside the comparable numbers from the tree line Soddie study site,
demonstrating a more direct comparison between the two. The Soddie data have been
reorganized so that the values for the two nitrogen treatments (N1 and N3) are
combined and compared to the control treatment as a percent difference. Following
the data available from the Cl study, only the parameters FJFm, FyFm and NPQ are
included. Note specifically that NPQ was not homoscedastic and therefore not
analyzed for significant treatment differences due in the Soddie study. To match the
experimental design utilized at Soddie, only the active treatment period data for old
growth are used from the Cl study. At Cl, only FyFm was significantly different (p
< 0.05) due to treatment. Comparing the data in this manner, nitrogen obviously
exhibited the opposite effects on fluorometry parameters at these two sites despite
their close geographic proximity. This further emphasizes that the effects of nitrogen
deposition are complicated in large part by local ecosystem dynamics, and
extrapolations from single study areas should be made with caution. One hypothesis
82


for why nitrogen should affect fluorometric parameters so differently at each site is
that the early growing season snowmelt is heavily nitrogen-laden at Soddie. This
situation provides an overall larger amount of nitrogen to the Soddie spruce, as
compared to the lower elevation Cl spruce. Characterizing the availability and use of
snowmelt nitrogen by trees at Soddie, as well as estimating the dry deposition rates of
nitrogen at the site, would further illuminate this issue. Another potential explanation
for the differential between Soddie and C1 results is the possibility of an unusual
season at Cl coinciding with the data collection period in 2004.
Table 9.1. A comparison of fluorometric responses
to nitrogen amendments at two Niwot Ridge sites:
C1 and Soddie.
Site F/Fm Percent Difference FVF'm NPQ
Cl 2.8% 11.5% -11.5%
Soddie -2.1% -6.9% 15.9%
Cl data from Tomaszweski & Sievering, 2007; Soddie
calculated as 100% x [(Nl+N3)/2] C / [(N1+N3 + C)/3],
83


10. Conclusions and Implications
The aim of this investigation was to examine the possible impacts of nitrogen
deposition on carbon sequestration through foliar applications of biologically
available nitrogen to a high-altitude site with a history of higher than average ambient
nitrogen deposition (Holland, et al., 2005). Although fluorometry cannot be used as a
direct quantification of photosynthetic activity or carbon fixation, it can be used as a
qualitative surrogate measure of photosynthetic efficiency and capacity. Logically, if
nitrogen amendments can affect the capacity for light harvesting and utilization, gross
photosynthesis will likely be affected as well. Additionally, fluorometry can provide
insight into photoinhibition and the fate of excess absorbed energy. If not dealt with
properly, excess energy has the potential to harm the photosynthetic apparatus of
plant tissue, leading to a reduced capacity for photosynthesis and carbon
sequestration. If nitrogen amendments can affect the non-photochemical dissipation
of energy, this may have implications for long-term carbon sequestration capacity as
well.
In order to address these issues, two nitrogen amendment experiments were
conducted at a spruce forest located just below tree line, using chlorophyll
fluorometry to measure photosynthetic capacity and non-photochemical energy
dissipation. Trial one benefits from multiple years of data, allowing the potential
examination of trends over the long term. At the trial one site, results indicate that
84


increased nitrogen deposition could negatively affect carbon sequestration in this
forest. Trial two revealed insight into nitrogens effect on photoinhibition by
comparing sun- and shade-adapted tissue responses to nitrogen amendments.
Although differences were observed in fluorescence parameters due to light
adaptation, no significant trends due to nitrogen amendments were observed in the
single season for which data was collected. Similarly, morphological characteristics
showed differences due to light-adaptation state, and trends due to nitrogen were
suggested, but none exhibited significant differences.
In trial one, chlorophyll parameters that speak to photosynthetic capacity and
efficiency (Opsn, Fv/Fm and F/Fm) were significantly (p < 0.05) depressed by
nitrogen treatments, while non-photochemical dissipation processes (qN) were
significantly (p < 0.05) elevated when compared to controls and viewed over a three
year period. If these parameters accurately reflect the potential for photosynthesis at
the site, and the site is representative of other tree line sites, then these results imply
that chronic additions of nitrogen could be detrimental to carbon sequestration in
similar high-altitude sites with large amounts of nitrogen deposition. This sharply
contrasts with a previous short-term study indicating an increase in potential carbon
sequestration with increasing nitrogen deposition at a nearby site with less nitrogen
available through snowmelt (Tomaszewki and Sievering, 2007).
That nitrogen often has a fertilizing effect on tree growth is well documented
(Evans, 1989), but less clear is whether or not this relationship can be sustained in
85


light of chronic nitrogen amendments over the long-term. Many studies attempting to
characterize the role of nitrogen deposition in carbon sequestration are only
conducted over a single growing season, severely limiting their applicability to
ecosystem-level changes. In addition to the work with nitrogen and chlorophyll
fluorometry, this thesis highlights the importance of long-term studies in the analysis
of climate-related data. Although degree completion time restricted the length of the
study to only three years, the results of trial one show the benefit of a multiple-season
data set. In trial ones three-year data set, chlorophyll fluorescence parameters
indicate that increased nitrogen amendments do not stimulate further photosynthesis,
thus leading to greater carbon sequestration. In fact, photosynthetic efficiency
actually decreases with increasing nitrogen at this site.
86


REFERENCES
Aber, J. D., McDowell, W., Nadelhoffer, K., Magill, A., Bemtson, G., Kamakea, M.,
et al. (1998). Nitrogen saturation in temperate forest ecosystems Hypotheses
revisited. Bioscience, 48(11), 921-934.
Aber, J. D., Nadelhoffer, K. J., Steudler, P., & Melillo, J. M. (1989). Nitrogen
saturation in northern forest ecosystems. Bioscience, 39(6), 378-386.
Adams, W. W., & Demmig-Adams, B. (2004). Chlorophyll fluorescence as a tool to
monitor plant response to the environment. In G. C. Papageorgiou &
Govindjee (Eds.), Chlorophyll a fluorescence: A signature of photosynthesis
(pp. 583-604). Dordrecht: Springer.
Alves, P., Magalhaes, A. C. N., & Barja, P. R. (2002). The phenomenon of
photoinhibition of photosynthesis and its importance in reforestation.
Botanical Review, 68(2), 193-208.
Baker, N. R. (2008). Chlorophyll fluorescence: A probe of photosynthesis in vivo.
Annual Review of Plant Biology, 59, 89-113.
Baker, N. R., 8c Oxborough, K. (2004). Chlorophyll fluorescence as a probe of
photosynthetic productivity. In G. C. Papageorgiou & Govindjee (Eds.),
Chlorophyll a Fluorescence: A Signature of Photosynthesis (pp. 65-82).
Dordrecht: Springer.
Baron, J. S., Rueth, H. M., Wolfe, A. M., Nydick, K. R., Allstott, E. J., Minear, J. T.,
et al. (2000). Ecosystem responses to nitrogen deposition in the Colorado
Front Range. Ecosystems, 3(4), 352-368.
Bauer, G. A., Bazzaz, F. A., Minocha, R., Long, S., Magill, A., Aber, J., et al. (2004).
Effects of chronic N additions on tissue chemistry, photosynthetic capacity,
and carbon sequestration potential of a red pine (Pinus resinosa Ait.) stand in
the NE United States. Forest Ecology and Management, 796(1), 173-186.
Bowman, W. D., Gartner, J. R., Holland, K., & Wiedermann, M. (2006). Nitrogen
critical loads for alpine vegetation and terrestrial ecosystem response: Are we
there yet 1 Ecological Applications, 16(3), 1183-1193.
87


Bowman, W. D., & Seastedt, T. R. (Eds.). (2001). Structure and function of an alpine
ecosystem: Niwot Ridge, Colorado. New York, New York: Oxford University
Press.
Boyce, R. L., Friedland, A. J., Chamberlain, C. P., & Poulson, S. R. (1996). Direct
canopy nitrogen uptake from N-15-labeled wet deposition by mature red
spruce. Canadian Journal of Forest Research-Revue Canadienne De
Recherche Forestiere, 26(9), 1539-1547.
Bradley, J. V. (1968). Distribution-free statistical tests. Englewood Cliffs, NJ:
Prentice-Hall, Inc.
Calanni, J., Berg, E., Wood, M., Mangis, D., Boyce, R., Weathers, W., et al. (1999).
Atmospheric nitrogen deposition at a conifer forest: response of free amino
acids in Engelmann spruce needles. Environmental Pollution, 705(1), 79-89.
Campbell, N. A., Reece, J. B., Mitchell, L. G., & Taylor, M. R. (2003). Biology
concepts & connections (4th ed.). San Francisco: Benjamin Cummings.
Charway, H., & Bailer, A. J. (2007). Testing multiple-group variance equality with
randomization procedures. Journal of Statistical Computation and Simulation,
77(9), 797-803.
De Schrijver, A., Verheyen, K., Mertens, J., Staelens, J., Wuyts, K., & Muys, B.
(2008). Nitrogen saturation and net ecosystem production. Nature, 457(7180),
El-El.
de Vries, W., Solberg, S., Dobbertin, M., Sterba, H., Laubhahn, D., Reinds, G. J., et
al. (2008). Ecologically implausible carbon response? Nature, 457(7180), El-
E3.
Erisman, J. W., Beier, C., Draaijers, G., & Lindberg, S. (1994). Review of deposition
monitoring methods. Tellus Series B-Chemical and Physical Meteorology,
46(2), 79-93.
Evans, J. R. (1989). Photosynthesis and nitrogen relationships in leaves of c-3 plants.
Oecologia, 78(1), 9-19.
Fenn, M. E., Baron, J. S., Allen, E. B., Rueth, H. M., Nydick, K. R., Geiser, L., et al.
(2003). Ecological effects of nitrogen deposition in the western United States.
Bioscience, 55(4), 404-420.
88


Fenn, M. E., Poth, M. A., Aber, J. D., Baron, J. S., Bormann, B. T., Johnson, D. W.,
et al. (1998). Nitrogen excess in North American ecosystems: Predisposing
factors, ecosystem responses, and management strategies. Ecological
Applications, 8(3), 706-733.
Ferrar, P. J., & Osmond, C. B. (1986). Nitrogen supply as a factor influencing
photoinhibition and photosynthetic acclimation after transfer of shade-grown
solanum-dulcamara to bright light. Planta, 168(4), 563-570.
Gaudinski, J. B., Trumbore, S. E., Davidson, E. A., & Zheng, S. H. (2000). Soil
carbon cycling in a temperate forest: radiocarbon-based estimates of residence
times, sequestration rates and partitioning of fluxes. Biogeochemistry, 57(1),
33-69.
Grassi, G., Colom, M. R., & Minotta, G. (2001). Effects of nutrient supply on
photosynthetic acclimation and photoinhibition of one-year-old foliage of
Picea abies. Physiologia Plantarum, 111(2), 245-254.
Hendrickson, L., Furbank, R. T., & Chow, W. S. (2004). A simple alternative
approach to assessing the fate of absorbed light energy using chlorophyll
fluorescence. Photosynthesis Research, 82(1), 73-81.
Henley, W. J., Levavasseur, G., Franklin, L. A., Osmond, C. B., 8c Ramus, J. (1991).
Photoacclimation and photoinhibition in ulva-rotundata as influenced by
nitrogen availability. Planta, 184(2), 235-243.
Hogberg, P., Fan, H. B., Quist, M., Binkley, D., & Tamm, C. O. (2006). Tree growth
and soil acidification in response to 30 years of experimental nitrogen loading
on boreal forest. Global Change Biology, 12(3), 489-499.
Holland, E. A., Braswell, B. H., Sulzman, J., & Lamarque, J. F. (2005). Nitrogen
deposition onto the United States and western Europe: Synthesis of
observations and models. Ecological Applications, 75(1), 38-57.
Hyvonen, R., Persson, T., Andersson, S., Olsson, B., Agren, G. I., & Linder, S.
(2008). Impact of long-term nitrogen addition on carbon stocks in trees and
soils in northern Europe. Biogeochemistry, 89(1), 121-137.
Kahl, J. S., Norton, S. A., Fernandez, I. J., Nadelhoffer, K. J., Driscoll, C. T., & Aber,
J. D. (1993). Experimental inducement of nitrogen saturation at the watershed
scale. Environmental Science & Technology, 27(3), 565-568.
89