-
Title (dcterms:title)
-
Eng
Spatial variability of organic carbon within surface soils of an ombrotrophic bog in Western Washington
-
Date (dcterms:date)
-
2016
-
Creator (dcterms:creator)
-
Eng
Carter, Joshua
-
Subject (dcterms:subject)
-
Eng
Environmental Studies
-
extracted text (extracttext:extracted_text)
-
Spatial variability of organic carbon within surface soils of an ombrotrophic bog in
Western Washington
by
Joshua Carter
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June 2016
©2016 by Joshua Carter. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
Joshua Carter
has been approved for
The Evergreen State College
by
________________________
Erin Martin, Ph.D
Member of the Faculty
________________________
Date
ABSTRACT
Spatial variation of organic carbon within surface soils of an ombrotrophic bog in
Western Washington
Joshua Carter
Bogs are exceptional long-term carbon sinks, but have not been subject to the same level
of study as less quantitatively significant terrestrial carbon sinks. Although the role of
Sphagnum mosses as major drivers of carbon storage in bogs has been well-established,
the intrinsic variables that affect carbon content of surface soils within bogs tend to be
highly site-specific and are not well known in the Pacific Northwest, as they have not
been studied to the same extent as other peatland systems. Thus, several intra-bog
variables were selected in order to determine how they affect internal C cycling within
PNW bogs. Additionally, carnivorous plants were studied to determine their role in the C
cycle. To our knowledge, this is the first study to examine their role in C cycling within
peatland systems. Soil cores were taken and vegetative, chemical, and topographic
features were studied to identify associations within a bog on the coast of Western
Washington in Gray’s Harbor county. Results indicate the study site shares many
commonalities with bogs in other regions, particularly in Sphagnum’s preference for high
microtopography, its positive association with soil moisture, and its relationship with
higher soil carbon content. Carbon concentrations ranged from 396 mg/g to 445 mg/g by
dry weight, which is comparable to worldwide peatland averages as well as those of
neighboring regions. However, the C:N ratios and total nitrogen (N) content of the soils
were abnormally high (17:1 and 2.39% by dry weight, respectively) compared to similar
analyses in other regions of the world. This pattern appears to be driven by high organic
N content. These results have implications for wetland management and also for the
potential impacts of climate change as precipitation patterns shift and temperatures rise,
potentially shifting the carbon balance of peatlands from being net sinks to net sources of
atmospheric carbon. Further study is required to determine the causal relationships within
these results as well as whether these results apply to other bogs within the region.
Table of Contents
ABSTRACT
4
LIST OF FIGURES
V
LIST OF TABLES
VI
ACKNOWLEDGEMENTS
1. INTRODUCTION
VII
8
2. LITERATURE REVIEW
12
2.1 INTRODUCTION
2.2 OMBROTROPHIC BOGS
2.3. RELATIONSHIP BETWEEN THE VEGETATIVE COMMUNITY AND THE CARBON CYCLE
2.3.1 SPHAGNUM MOSS
2.3.2. PEAT
2.3.3. VASCULAR PLANT COMMUNITY
2.4. CARBON CYCLING IN OMBROTROPHIC BOGS
2.4.1 CARBON FLUXES
2.4.2 NITROGEN
2.4.3 C:N RATIOS
2.5. DISTRIBUTION OF CARBON STORAGE & EFFLUX WITHIN A BOG
2.6. CLIMATE CHANGE IMPLICATIONS FOR BOGS
2.6.1. VASCULAR PLANT DOMINANCE
2.6.2. DISAPPEARANCE AND REGRESSIVE SUCCESSION
2.6.3. GLOBAL IMPLICATIONS
2.7. PRESENT STUDY CONTRIBUTIONS
2.7.1. DETERMINE TOPOGRAPHIC PATTERNS OF CARBON DISTRIBUTION
2.7.2. PROVIDE AVENUES OF FUTURE RESEARCH
2.7.3. HELP UNDERSTAND THE CONSEQUENCES OF CLIMATE CHANGE AND OTHER HUMAN ACTIVITIES
2.8 CONCLUSION
12
12
14
14
16
18
20
21
22
24
25
26
26
27
27
29
29
29
30
30
3. METHODS AND MATERIALS
32
3.1 FIELD COLLECTION
3.1.1 WATER CHEMISTRY
3.1.2 VEGETATIVE SAMPLING
3.1.3 SOIL CORES
3.2 LAB ANALYSES
3.2.1 SOIL PH
3.2.2 NITRATE
3.2.3 AMMONIUM
3.2.4 ORGANIC CARBON & NITROGEN
3.3 STATISTICAL ANALYSES
3.3.1 TREATMENT OF DATA
32
34
34
35
36
37
37
38
39
39
40
4. RESULTS
41
4.1 VEGETATION
41
iv
4.1.1 PHYSICAL AND CHEMICAL SUMMARY DATA
4.1.2. SPHAGNUM
4.1.3. DROSERA
4.2 MICROTOPOGRAPHY
4.3 NITROGEN
4.3.1 INORGANIC NITROGEN
4.3.2. ORGANIC NITROGEN
4.4 ORGANIC CARBON CONCENTRATION
4.5 C:N RATIO
42
42
44
45
47
48
54
55
58
5. DISCUSSION
59
5.1 CARBON
5.2 NITROGEN
5.2.1 INORGANIC N
5.3 C:N RATIOS
5.4 VEGETATION
59
62
63
64
66
7. CONCLUSION
69
REFERENCES
71
APPENDICES
80
APPENDIX A. SITE VEGETATION DATA
APPENDIX B. MICROLANDFORM DATA
80
81
v
List of Figures
Figure 1: a scanning electron microscope image of hyaline cell structures within a Sphagnum leaf cell.
The pores in these cells allow for greater water uptake and retention in saturated soils than a vascular
root system would (https://sphagnumsem.wordpress.com).
14
Figure 2: Soil moisture is significantly greater in the presence of Sphagnum (p < 0.01).
43
Figure 3: Soil moisture increases as Sphagnum depth increases (p < 0.05)
44
Figure 4: The presence of Drosera is associated with lower soil moisture (p < 0.05).
45
Figure 5: Sphagnum presence is strongly associated with landform height (p < 0.01). See appendix B for
how average microtopography was determined.
46
Figure 6: Sphagnum depth is positively associated with landform height (p < 0.05). Graph shows 95%
confidence interval for each value.
47
Figure 7: While total inorganic N is not significantly correlated with water depth, both NH4+ and NO3are, but in opposite ways. NH4+ is negatively correlated with water depth (p < 0.01), while NO3- is
positively correlated with water depth (p < 0.01).
49
Figure 8: While total inorganic N is not significantly correlated with soil pH, NH 4+ is, and NO3- shows a
non-significant correlation in the opposite direction. NH4+ is positively correlated with soil pH (p =
0.0128), while NO3- is negatively correlated with soil pH (p = 0.0599).
50
Figure 9: NO3- is not significantly correlated with dissolved O2, but NH4+ is (p < 0.05), and total
inorganic N shows a non-significant correlation (p = 0.1031).
51
Figure 10: The presence of Drosera is negatively associated with inorganic N values (p < 0.05 +. NO3- and
NH4+ alone were not significantly different between plots. Graph shows 95% confidence intervals for
each value.
52
Figure 11: Total inorganic N concentrations are not affected by microtopography, but NH 4+
concentrations were higher in “low” topography areas than in the others, although this relationship was
not signifacant (p = 0.0884). NO3- concentrations were significantly higher in “high” topography than
they was in “low” topography (p < 0.05). Graph shows 95% confidence interval for each value.
53
Figure 12: Inorganic N concentrations are higher where Sphagnum is present, although not
significantly so (p = 0.2091). NO3- was not significantly correlated with the presence or absence of
Sphagnum, while NH4+ concentrations were significantly higher where Sphagnum was present (p <
0.05). Graph shows 95% confidence intervals for each value.
54
Figure 13: Organic C and N are positively correlated (p < 0.001), as are organic N and inorganic N (p <
0.01).
55
Figure 14: Organic C concentrations and water depth show a significant negative correlation (p < 0.05).
“Depth” refers to the maximum depth of standing water measured within the sample plot.
56
Figure 15: Soil C concentrations are non-signficiantly associated with the presence of Sphagnum when
the first five samples are excluded from the dataset (p= 0.0751).
57
Figure 16: C:N ratios are positively correlated with soil moisture (p < 0.05) and moss depth (p = 0.1012),
while they are negatively correlated with water depth (p = 0.0609). Only soil moisture shows a
significant correlation.
58
v
List of Tables
Table 1: Summary of vegetation data. The “Vascular” column excludes Drosera and graminoids. Actual
site data % cover was derived from can be found in Appendix 1. __________________________________________ 41
Table 2: Summary of physical and chemical data by the presence and absence of plant functional
groups. ± values indicate standard deviation. Note that dissolved O2 measurements were made in plots
with standing water only. ____________________________________________________________________________________ 42
Table 3: Summary microtopography data. Most plots contained more than one microlandform, while
one plot was almost completely inundated and did not contain any of the three microlandforms. Each
plot was standardized based on average landform data into one of three categories: “low,” “medium,”
and “high.” Landform data can be viewed in Appendix 2. __________________________________________________ 45
Table 4: Summary of N content in soil samples. Organic N makes up the vast bulk of the N present within
the soil samples, with ammonium making up the majority of the inorganic N present. __________________ 48
vi
Acknowledgements
First, I would like to thank Dr. Erin Martin for being an amazing mentor, advisor, and for
the wholehearted faith she put in my ability to perform my own research. Her door was
always open whenever I ran into trouble or required the benefit of her knowledge and
expertise.
I’d also like to thank everyone else who made this thesis possible:
David Wilderman (Natural Areas Ecologist, Washington DNR) for helping me gain
access to my research site and for assisting with site selection based on the criteria I
needed.
Sina Hill (Science Instructional Technician) for ensuring I knew how to use the scientific
instrumentation on campus and for helping me refine my methods.
Kaile Adney (Science Instructional Technician) for always being willing to help with
procurement of supplies and lab space.
Trisha Towanda (Science Instructional Technician) for working on Saturdays and saving
me that one time I ran out of filters in the middle of my lab work.
Dr. Kevin Francis (MES Director) for being a supportive member of the faculty and for
going above and beyond in terms of finding the funding and help I needed to complete
my thesis.
Dr. Dylan Fischer (Evergreen faculty) for allowing me to use his CHN analyzer and for
assisting with my methodology.
Rebekah Korenowsky for providing emotional and intellectual support, for helping me
learn how to use Evergreen’s instruments, and for assisting with my field work.
Allie Denzler for being an excellent friend and for assisting with my field work.
Heather May for being a supportive roommate and for providing me an sounding board to
bounce ideas off of.
And last, but not least, the entire 2014-2016 MES cohort! I couldn’t have done it without
you guys.
vii
1. Introduction
Bogs are a form of wetland characterized by highly acidic (pH < 5), peaty, and
oligotrophic soils; hydrologic isolation from surrounding landscapes resulting in
atmospheric deposition as the sole major water and nutrient input; vegetative
communities dominated by mosses of the Sphagnum genus, graminoids, and small shrubs
that tolerate poor soils well; and a high water table that is both highly acidic and anoxic.
These landforms contain a number of unique taxa, including the carnivorous plants,
which have evolved adaptations that allow them to live in soils with few available
nutrients, as well as numerous amphibians, macroinvertebrates, and insects that can not
live anywhere else (Ellison and Gotelli 2001, Matušíková et al. 2005). Despite only
encompassing 3% of the world’s land area, peatlands, of which bogs are the principle
type, represent over 1/3 of the world’s terrestrial carbon stores (Yu 2012, Gorham et al.
2012, Charman et al. 2013). Additionally, peatlands efficiently sequester carbon at an
estimated rate of 24-32 g C/m2/yr over the last 1000 years, indicating that they are
exceptional long-term carbon sinks (Loisel et al. 2014). However, these systems are
relatively understudied compared to other terrestrial habitats due to their scarcity and
relative inaccessibility, and so much is still unknown about them that they have been
traditionally excluded in most climate models, despite the large carbon pool they
represent (Limpens et al. 2008).
In particular, Pacific Northwest bogs have received very little attention compared
to bogs in other regions of the world. Most research has been focused on experimental
bogs in Eastern Canada, tropical bogs, and the vast peat bogs of Ireland and Northern
8
Europe (Draper et al. 2014, Sheppard et al. 2013, Ohlson & Okland 1998, Blodau et al.
2007). While bogs are less abundant in Western North America than they are in these
parts of the world, Lacourse and Davies (2015) found significantly lower C deposition
and higher N deposition within a bog in Western Canada compared to the aforementioned
sites. Additionally, temperate bogs sequester much more carbon per unit area than do
tropical bogs and are much more numerous, making their study more important in terms
of understanding global carbon cycling and potential feedbacks from climate change
(Page et al. 2011, Holden 2005). The near future may bring with it severe declines in
wetland habitat due to rising temperatures, disturbances in precipitation regimes,
development, agriculture, and other human-caused disturbances, so it is important to
understand these systems and the underlying processes that control carbon flux (Charman
et al. 2013, Bu et al. 2011).
The intra-wetland variables that affect carbon cycling within bogs are highly sitespecific, and bogs found within the PNW have already been found to differ from those
elsewhere (Loisel et al. 2014). While many factors, Sphagnum cover, water table depth,
and precipitation are positively correlated with carbon concentrations in a predictable
way across many different regions, many other variables, such as nitrogen addition,
vascular plant cover, and air temperature have inconsistent relationships to carbon
concentrations (Basiliko et al. 2006, Wang et al. 2014, Charman et al. 2015). As such,
based on published results so far, it is hard to predict how carbon cycling is affected by
these latter variables in this region. Additionally, while these factors have not been
studied adequately in Pacific Northwest bogs, the role of carnivorous plants in carbon
cycling has remained unexplored at the time of writing this paper. Due to their presence
9
in many peatland systems, and the additional mechanism of carbon capture through
insectivory they represent, I hypothesize that insect capture may represent a significant
carbon flux in peatland systems.
In this paper, I investigate several of these intra-wetland variables within a bog
that is part of the Washington Department of Natural Resources North Bay Natural Area
Preserve on the coast of Western Washington, just south of the Olympia Peninsula.
Specifically, the role of bog microtopography, the presence of two plant functional
groups (Sphagnum and carnivorous plants), and various soil components (organic and
inorganic nitrogen, pH) were sampled and compared to levels of organic carbon within
the surface soil (2-5 cm deep) to determine relationships between carbon content and
these variables. In keeping with the literature, carbon concentrations were expected to be
positively correlated with the thickness of Sphagnum cover, soil acidity, water depth, and
the presence of high microtopography (hummocks), and negatively associated with the
presence of low topography (hollows) (Ohlson and Okland 1998, Laine et al. 2006).
Sundews are a unique area of study for this paper, as their role in C cycling has
not been studied. There are many species of carnivorous plants, but the most common in
this region is Drosera rotundifolia, the round-leafed sundew. This species is one of the
most widely distributed carnivorous plants, occurring in northern latitudes throughout
North America, Europe, and Asia. Additionally, its physiology is well understood and has
been studied for well over a century, making it a model plant for this type of study.
Because bogs have such acidic and nutrient poor soils, sundews have evolved the
capability to ensnare and digest insects in order to acquire the nutrients they need (mostly
nitrogen, phosphorous, and calcium, but also many trace elements) (Givnish et al. 1984).
10
But like all plants, they photosynthesize using CO2 taken up from the atmosphere. This
means that the fate of the carbon contained within the insects they ensnare remains
unknown and has not been studied at the time of writing this thesis. I hypothesize the C
from such insects to be either consumed by microbes or remain in the bog itself, and thus
I expect to find that soils supporting sundews will have higher concentrations of organic
carbon than those without. Additionally, because sundews are not 100% efficient at
extracting the nitrogen from the insects they capture, higher levels of organic nitrogen
may also be found in soils where they are present.
I expect this research to elucidate the intra-wetland variables that affect carbon
cycling in Pacific Northwest bogs and to also further our understanding of carbon cycling
in bogs in general. If the results obtained by this study can be replicated on a larger scale
and in a broader array of sites, the processes that control carbon cycling in PNW bogs
may be better elucidated, allowing us to understand the unique characteristics of
peatlands in this region. These results may allow wetland restoration efforts to be
customized to the specific needs of the region, to create more accurate climate change
models, and may increase the precision of global carbon budget models.
In the following paper I will review the current literature significant to this topic
and report the findings of this study. While there were a number of challenges in
collecting and analyzing the data, several interesting trends revealed themselves. Many of
these results are in agreement with prior studies or are inconclusive, but some, such as the
C:N ratio values reported in soil samples were surprisingly low, which has greater
implications in terms of the study site.
11
2. Literature Review
2.1 Introduction
In the following text I briefly explain what an ombrotrophic bog is in relation to
other peatlands, the relationships between bog vegetation and carbon cycling, the
important carbon fluxes within bog systems and how nutrient limitation affects them,
how carbon is distributed within bogs and why, how climate change may impact bogs in
the future, and what gaps in the literature this research fills.
2.2 Ombrotrophic bogs
Bogs are one of the principle types of wetlands (the others being fens, marshes,
and swamps), and are characterized by highly acidic soils, dominance of the vegetative
community by Sphagnum spp., and the accumulation of peat, which is largely composed
of the partially decayed remains of Sphagnum mosses and vascular bog plants. There are
several types of bogs, but ombrotrophic bogs are characterized by being hydrologically
separate from the surrounding landscape and by being entirely precipitation fed (Vitt,
2013). While many non-bog wetlands also produce peat (primarily fens), ombrotrophic
bogs make up the majority of the world’s peatlands, contain the majority of its peat, and
contain peat that is less labile than that of other peatlands, making them the single most
important type of wetland in terms of carbon storage (Bridgham et al. 2006, Corbett et al.
2009).
Ombrotrophic bogs are also distinct from other wetlands in their water and soil
chemistry. Rain is typically more acidic and contains fewer nutrients than surface or
12
groundwater, and because these bogs receive all of their moisture from precipitation, this
results in very low nutrient inputs and highly acidic conditions. The lack of water
movement also tends to create anoxic conditions, as microbial activity quickly depletes
the water column of its available oxygen. Because of these factors, ombrotrophic bogs
are inhospitable to most plants, as there is very little nitrogen or phosphorous available,
and the acidic soil makes nutrients, even when present, difficult to absorb by a plant’s
roots due to a much decreased cation exchange capacity (Vitt, 2013). Additionally,
Sphagnum mosses produce a number of phenolic compounds that have been shown to
inhibit microbial and fungal growth and their ability to decompose organic matter
(Bragazza et al. 2006, Turetsky et al. 2008). These properties of bogs are what allow
ancient bodies, wood, and other organic material to remain largely preserved after being
submerged for thousands of years (Purdy 2002). Ancient peoples even used bogs to
preserve substances such as butter and tallow which are still recognizable to
archaeologists after thousands of years (Ball, 2004)!
Sphagnum mosses are ecosystem engineers and arguably the most important
biotic component of an ombrotrophic bog. Their presence facilitates the formation of
peat, contributes to the water and soil chemical properties that prevent decomposition,
and prevents most vascular plants, which are typically more labile sources of carbon,
from growing within the bog (Nichols et al. 2014). When Sphagnum cover declines, it is
often the first indication that a bog is shifting from a net carbon sink and is becoming a
net carbon source, as microbial respiration rates and vascular plant cover both increase as
Sphagnum cover decreases (Larmola et al. 2013).
13
2.3. Relationship between the vegetative community and the carbon cycle
2.3.1 Sphagnum moss
Unlike most ecosystems, bogs are dominated by bryophytes, specifically the
Sphagnum mosses. Because bogs typically have a high water table year-round,
Sphagnum’s lack of vascular tissue is advantageous compared to vascular plants, as it
absorbs water and nutrients directly without the need to grow roots, enabling it to quickly
spread across the surface of the soil (Adkinson and Humphreys 2011). These mosses
typically blanket the surface of a bog, which prevents most other plants from growing.
Sphagnum mosses are characterized by having alternating layers of living green
cells and large, dead hyaline cells within their leaves (Fig. 1) (Vitt 2013).
Figure 1: a scanning electron microscope image of hyaline cell structures within a Sphagnum leaf cell.
The pores in these cells allow for greater water uptake and retention in saturated soils than a vascular
root system would (https://sphagnumsem.wordpress.com).
14
These hyaline cells are also found outside stems and branches, and are perforated
throughout, providing numerous tiny reservoirs that allow for capillary transport of water
well above the water table (Vitt 2013).
Additionally, Sphagnum produces uronic acid, which increases its cation
exchange capacity when pH is low, allowing it to extract nutrients from acidic soils much
more efficiently than vascular plants (Spearing 1972, Richter & Dainty 1990). When
combined with its hyaline structure, Sphagnum is exceptionally efficient at taking up
nutrients, essentially acting as a sponge for water and the nutrients contained within it
(Fritz et al. 2014). This also prevents other plants from encroaching, as Sphagnum is able
to take up nutrients as they become available much faster than vascular plants. However,
these adaptations work against Sphagnum when conditions are dry or less acidic,
respectively, as without vascular tissue, the plant is unable to control its water loss and is
much less efficient at extracting nutrients from less acidic soils (Adkinson and
Humphreys 2011).
Interestingly, Sphagnum also produces a unique phenolic compound, Sphagnum
acid, found in no other taxa. This compound prevents predation, preserves remains within
the bog site, and contributes to the ability of Sphagnum to act as ecosystem engineers, as
Sphagnum acid contributes to the acidification of their habitat, inhibits microbial growth,
is unpalatable to animals, and creates adverse growing conditions for other plants (van
Breeman 1995). In turn, because vascular plant growth is reduced, less evapotranspiration
occurs within the bog, resulting in higher moisture retention, which allows the Sphagnum
to flourish while excluding plants that are less tolerant of inundation.
15
While Sphagnum is a less efficient photosynthesizer than vascular plants, its
phenolic content and the resulting acidity of the peat produced from its remains means
that carbon derived from Sphagnum has a much longer residence time than that of
vascular plants or other bryophytes. Importantly, while Sphagnum’s capacity to act as an
ecosystem engineer decreases the lability of all carbon within the peat column, carbon
derived from Sphagnum is still up to ten times less labile than carbon from other sources
(Dorrepaal et al. 2005).
2.3.2. Peat
As previously mentioned, Sphagnum mosses are extremely important in the
formation of peat, the primary form of carbon stored within bogs. Peat is composed of the
partially decayed remains of Sphagnum that have sunk below the living Sphagnum layer.
As new Sphagnum growth occurs, this dead layer compacts and sinks deeper into the bog.
After it sinks below the water table, where anoxic conditions prevail, decomposition
almost comes to a stop, and only small amounts of anaerobic respiration occur, allowing
this peat to remain essentially preserved for long periods of time (Vitt 2013). However,
this is such a slow process that peat is not considered a renewable resource; peak peat
accumulation in most bogs is less than 1 mm yr-1, and it can take well over a millennium
to form a layer of peat one meter deep (Holden 2005).
If components of water and soil chemistry aren’t right, mossy remains will
decompose rather than form peat. Acidic soils, anoxic water, and oligotrophic conditions
are essential to the formation of peat, as they inhibit microbial growth and respiration and
enable the preservation of mossy remains (Vitt 2013). Also important to the formation of
16
peat are the aforementioned phenolic compounds Sphagnum produces within its leaves,
which, upon death of the plant, inhibit microbial and fungal decomposition, either by
direct action on the organism or by inhibition of the enzymatic processes that break down
organic matter (Bragazza et al. 2006, Turetsky et al. 2008).
Generally, as long as the water table is high and there is no significant
disturbance, a bog will sustain the correct soil and water chemistry on its own. However,
significant disturbances in the form of peat mining, development, or drainage can kill the
Sphagnum layer, and when bare peat is exposed to the atmosphere, it undergoes
accelerated decay due to increased oxygen levels and temperature, increased availability
of nutrients, and decreased moisture and acidity (Leifeld 2011). Additionally, climatic
changes can affect soil and water chemistry in bogs, primarily by changing temperature
and precipitation patterns. Generally, colder temperatures and higher levels of
precipitation contribute to a bog’s ability to act as a carbon sink, while the opposite
increases the rate of decomposition and may create habitat unsuitable for Sphagnum (Ise
et al. 2008, Charman et al. 2013). Unfortunately, human disturbance on local and global
scales (climate change) will likely lead to increased loss of peatlands in the foreseeable
future.
The formation of peat also constitutes a positive feedback in forming more peat.
Because so little decomposition occurs, peat retains most of the nutrients the living plant
contained at the time of death. As a result, Sphagnum peat can contain up to 0.5%
nitrogen, while graminoid peat can contain up to 3% nitrogen, as well as significant
amounts of other macro and micronutrients, but because all of it occurs below the water
table where little respiration occurs, these nutrients are essentially unavailable (Crum and
17
Planisk, 1992). In a seemingly contradictory fashion, oligotrophic bogs are very rich
sources of nutrients, but those nutrients are virtually unavailable so long as the bog
maintains a positive carbon balance, as they are stored within the anoxic peat layer where
very little respiration can occur.
2.3.3. Vascular plant community
Vascular plants that exist in bog ecosystems have developed a number of novel
adaptations allowing them to survive in conditions of semi-permanent inundation,
chronic oligotrophic conditions, and a Sphagnum community that occupies a majority of
the bog’s substrate. These are typically small, herbaceous plants or slow-growing,
oligotrophic conifers (such as Pinus contorta, Shore Pine). In general, they play a much
smaller role in carbon cycling as their abundance is very low and their growth is very
slow compared to Sphagnum (Ward et al. 2009). Vascular plants do represent a more
labile source of carbon than Sphagnum, and Nichols et al. (2014) reconstructed historical
carbon accumulation rates in a bog near Cordova, Alaska using hydrogen isotope ratios
and found that during the periods of time in which the bog’s vegetative community was
dominated by Sphagnum, its carbon accumulation rate was 60% higher than during the
periods of time in which the community was dominated by sedges. They also found that
in addition to accumulating less carbon than Sphagnum dominated communities, sedgedominated communities also increased the rate at which the underlying peat lost carbon.
Of special consideration are the carnivorous plants, the most common of which in
the Pacific Northwest is Drosera rotundifolia, the Roundleaf Sundew. Sundews have
evolved a unique adaptation that allows them to grow directly on beds of Sphagnum,
18
using their roots primarily for anchoring and support rather than nutrient absorption
(although their roots are still capable of absorbing nutrients if they are present within the
substrate they grow upon) (Chandler and Anderson 1976). Instead, they are covered in
trichomes (glandular protrusions) that secrete sweet, sticky mucilage which act to attract
and then ensnare insects. When an insect becomes stuck, the sundew reacts by curling its
stalk around the insect to trap it further and by secreting digestive enzymes that break
down the insect’s chitinous exoskeleton and internal organs (Matušíková et al. 2005). It
then absorbs the nutrients released from these tissues (primarily nitrogen and
phosphorous, but also calcium and a number of micronutrients) (Ellison and Gotelli
2001). In this way, a sundew is able to attain the nutrients it needs without competing
with Sphagnum for those resources. Indeed, experiments have shown that even when
grown on a nutrient-rich substrate, sundews display inhibited growth compared to when
they get their nutrients from insects (Chandler and Anderson 1976). Additionally, while
sundews have vascular tissue and grow taller than mosses, they do not shade out the
Sphagnum they grow upon, implying at least a commensal relationship between the two
taxa (Svensson 1995). Given that sundews don’t appear to stifle photosynthesis in
Sphagnum and that they capture insects to acquire the nutrients they need, one might
expect carbon concentrations to be higher in areas where sundews occur, as they do not
utilize the carbon contained within their prey. Conversely, it is also possible that insectderived carbon is decomposed or otherwise removed via abiotic/biotic processes prior to
being preserved within the peat column. However, while Millett et al. (2003) attempted to
determine if insect-derived C was taken up by sundews, they were unable to determine
the isotopic carbon signature of the insects they were utilizing as prey and were thus
19
unable to analyze this aspect of sundew carnivory. No further studies have been
performed that investigate the relationship between Drosera spp. and carbon distribution
within bogs as of yet, so the fate of this carbon is unknown at this time.
Few studies have examined the effects of vascular plants in bogs on carbon
cycling. Ward et al. (2009) performed a plant-removal experiment on an ombrotrophic
bog in northern England and found that the presence of dwarf-shrubs actually inhibited
the ability of the bog to take up carbon in the short-term by impairing photosynthesis in
graminoid plants (grasses, sedges, and rushes) that were shaded by the shrubs. However,
this experiment did not examine the long-term impacts of the presence of these plants,
and while it established a change in gross carbon flux, there was no change in net carbon
flux.
2.4. Carbon cycling in ombrotrophic bogs
Peatlands1 contain up to one-third of the world’s terrestrial soil carbon stores,
between 400–600 Pg, while only representing less than 3% of its total land surface area,
making them the single largest component of the terrestrial carbon sink (Gorham, 2012,
Yu 2012, Charman et al. 2013). Russia currently holds the largest peatland reserves of
any region (214 Pg), with North America containing the second largest (178 Pg), the
majority of which is found in Canada (Botch et al. 1995, Bridgham et al. 2006). Despite
this, carbon cycling in peatlands is still poorly understood.
1
Peatlands are peat-forming wetlands, of which there are two types: fens and bogs. If left undisturbed, fens
will eventually become bogs as layers of peat form and decrease the hydrologic interaction of the wetland
from its surroundings. Bogs represent the bulk of the world’s peatlands (Blodau 2002).
20
Bogs are composed of two main layers: the acrotelm and the catotelm. The
acrotelm is composed of the shallow section of the bog where living Sphagnum and other
plants reside and typically is above the minimum depth of the water table, while the
catotelm includes the partially decomposed column of peat below the acrotelm (Holden
2005). Virtually all aerobic respiration occurs within the acrotelm, as well as a significant
portion of anaerobic respiration (which varies depending upon temperature and water
table depth), while the catotelm is permanently saturated, anoxic, and undergoes a very
slow rate of anaerobic respiration.
As a result of interactions between these two zones, carbon cycling in bogs is
mainly affected by three processes: photosynthesis, aerobic decomposition within the
acrotelm layer, and anaerobic processes within the catotelm layer (Vitt, 2013).
2.4.1 Carbon fluxes
In general, carbon flux and storage are driven by the flow of water in
ombrotrophic bogs. Because of their anoxic environments, most carbon mineralization
occurs anaerobically, producing primarily methane (CH4), the second most important
greenhouse gas as well as one of the most potent, and hydrogen sulfide (H2S) (Keller &
Bridgham, 2007). CO2 is also produced, but mostly as a product of plant and microbial
respiration within the acrotelm layer (Holden 2005). CO2 production tends to increase
when the water table drops and peat is exposed to the air, as well as during desiccation
events (Adkinson and Humphreys 2011).
Ohlson and Okland (1998) found that it takes approximately 40 years for a
statistically significant amount of carbon to have been released from peat by decay. This
21
extremely slow process of decay is the main reason why bogs are net carbon sinks.
Bridgham et al. (2006) calculated a global sequestration of 137 Tg C yr-1 into peatlands,
although this number comes with an uncertainty of over 100% (the actual number may be
>±100%). Nichols et al. also calculated a historical carbon sequestration rate of 7.5
g/m2/yr to 27 g/m2/yr in sedge dominated and Sphagnum dominated bos, respectively.
The vast majority of carbon flux to the atmosphere comes from the acrotelm. The
catotelm is responsible for less than 1% of total ecosystem respiration and is not
considered a significant flux, while 90-97% of all the carbon fixed by plants is respired
before and during transfer to the acrotelm through plant respiration and decomposition
(Blodau et al. 2007, Ohlson and Okland 1998). An experiment performed by Blodau et al.
(2004) determined that microbial respiration decreases as depth increases, but that water
table position can greatly influence this process. When water tables were low,
photosynthesis decreased by 24–42% due to desiccation, while aerobic respiration within
the catotelm and acrotelm increased significantly; in addition, while overall methane
production decreased when the water table was low, CO2 production increased by a factor
of 13–103!
2.4.2 Nitrogen
Nitrogen mineralization is generally extremely limited in bog habitats due to their
acidic soils, and in ombrotrophic bogs nitrogen inputs are limited entirely to atmospheric
deposition (Keller and Bridgham 2007). Thus, bogs are generally nitrogen-poor habitats
with plant communities that are highly adapted to these conditions.
22
A number of experiments have been performed in bogs to determine the
relationship between nitrogen deposition and carbon accumulation. This is of special
importance because peat contains large amounts of nitrogen that may become available as
the climate warms. Most of these experiments confirm that inorganic nitrogen addition
(whether through artificial means or by monitoring atmospheric deposition) increases
vascular plant growth and decreases Sphagnum growth and cover, resulting in increased
CO2 efflux from bogs and a loss of carbon storage capacity (Basiliko et al. 2006,
Bragazza et al. 2006, Bubier et al. 2007, Kivimäki et al. 2013, Larmola et al. 2013).
However, some studies have shown a positive correlation between nitrogen
addition and carbon accumulation. Basiliko et al. (2006) found that increased nutrient
inputs (nitrogen, phosphorous, and potassium) also caused the soil microbial community
to become a larger, more efficient carbon sink. Charman et al. (2015) also found a
slightly positive effect of nitrogen addition on carbon accumulation at one of their sites,
but only at lower levels of nitrogen addition. This is in agreement with Wang et al.
(2014), who reconstructed the historical records of a bog in Eastern Canada and found
that carbon accumulation was positively correlated with nitrogen accumulation over the
last ~8000 years, although these records only look at total nitrogen, not inorganic and
organic nitrogen separately.
It thus appears that the effects of nitrogen addition to bog systems are variable,
with smaller additions potentially increasing a bog’s carbon uptake efficiency, while
larger amounts decrease that efficiency. This is in keeping with Sphagnum’s established
role as a very efficient scavenger of nitrogen, as it can effectively prevent other plants
23
from taking up excess nitrogen to a certain point, after which vascular plants begin to
benefit from excess nitrogen as well (Fritz et al. 2014).
It is also important to note that these experimental studies are, by necessity, shortterm when compared to the length of time it takes carbon to move throughout the peat
column, so it is not clear whether large amounts of N addition permanently decreases a
bog’s carbon sequestration or if it only does in the short-term.
2.4.3 C:N Ratios
C:N ratios generally decrease along the peat column as decomposition occurs,
preferentially decomposing carbon and releasing CO2 and CH4 while retaining much of
the N present within the vegetative tissues (Wang et al. 2014). The result is a soil column
relatively high in nitrogen, albeit mostly unavailable to microbes and plants, as
mentioned in section 2.2.
Wang et al. (2014) report an average C:N ratio of 47:1 in living tissues at Mer
Bleue Bog, and a steady decrease of 42:1 to 31:1 down the gradient of the peat column.
Kuhry and Vitt (1996) reported C:N ratios of 55:1 to 76:1 in living Sphagnum fuscum,
with a similar decrease in C:N ratios in older, deeper peat. However, Kurhy and Vitt
found preferential decomposition of nitrogen in the acrotelm resulting in C:N ratios as
high as 300:1, in contrast to Wang et al.’s findings, indicating that these processes may
differ from site to site depending upon variables such as community composition, soil
pH, water table depth, and so forth. Loisel et al. (2014) supports this notion in a metaanalysis of the available literature, showing average regional C:N ratios of 34.2–77.2 for
most of the Northern Hemisphere’s temperate peatlands.
24
2.5. Distribution of carbon storage & efflux within a bog
The spatial relationships between carbon accumulation and bog components are
highly variable from site to site, and are affected by factors such as vegetative
community, water table depth, nitrogen content of soils, temperature, topography, and
seasonality.
Charman et al. (2015) looked at historical carbon deposition within three different
bogs on the East Coast of North America and found that while nitrogen and vegetation
were weakly associated with carbon accumulation in one or two of the sites, water table
depth and temperature were the only variables that were significant in all three bogs.
Additionally, water table depth was always positively correlated with higher soil carbon
levels, while temperature was always positively correlated with methane and CO2
production, whereas the other variables often showed contradictory relationships with
carbon accumulation and/or greenhouse gas production, likely due to individual
differences in vegetation, climate, and location between the sites.
In another study, Blodau et al. (2007) found that by artificially lowering the water
table in an ombrotrophic bog, the total ecosystem respiration rate increased by 61%,
indicating that water table depth is perhaps one of the most important controls on total
carbon efflux.
Landscape features also play a prominent role in the spatial distribution of carbon
in bogs. There are three main landforms within bogs: hummocks, or elevated sections;
hollows, which are the lower elevation spaces between hummocks; and lawns, which are
flat expanses or plateau-like features (Laine et al. 2006). Hummocks tend to have deeper
25
Sphagnum cover and higher rates of photosynthesis and respiration, while hollows are
closer to the water table and engage in less respiration, although Laine et al. (2006) found
that even though respiration is higher in hummocks, they are still more effective carbon
sinks than hollows due to their increased rate of photosynthesis (Ohlson and Okland
1998).
2.6. Climate change implications for bogs
Climate change is likely to lead to a number of changes in bogs, but those specific
changes depend upon regional climatic changes. Overall, evapotranspiration is predicted
to increase in most bogs, which may result in desiccation events that dry out Sphagnum
and result in a higher carbon flux to the atmosphere (Bu et al. 2011). If
evapotranspiration surpasses precipitation, the hydrology of the bog will shift, resulting
in the disappearance of bog plant communities and ultimately the loss of bog function
(Nichols et al. 2014).
2.6.1. Vascular plant dominance
Climate change is likely to cause a shift in the community composition of bogs
towards vascular plant dominance. Increased evapotranspiration and altered precipitation
patterns may decrease the water table, and as the water table decreases, vascular plants
become increasingly common and Sphagnum disappears (Moore et al. 2002, Bubier et al.
2007). Because Sphagnum is instrumental in the accumulation and preservation of carbon
within bogs, and because it is easily outcompeted by vascular plants in drier, warmer
conditions, this will result in increased contribution of carbon to the atmosphere not just
26
because vascular plants represent a more labile source of carbon than Sphagnum, but also
because the disappearance of Sphagnum will result in increased peat decomposition (Ise
et al. 2008).
2.6.2. Disappearance and regressive succession
Ombrotrophic bogs are often an “end stage” in terms of succession; those that
exist today were formed, on average, 7000 years ago (Gorham et al., 2012). Unless
climate and/or precipitation patterns change, a bog will continue to form layers of peat
and sequester more and more carbon (Bu et al. 2011). However, if evapotranspiration
exceeds precipitation (either through higher temperatures, less precipitation, or some
combination of both), the hydrology of the bog will no longer support Sphagnum and the
bog will undergo regressive succession, meaning it will transition to a forested or grassy
ecosystem (Bu et al. 2011). As this happens, stored peat undergoes increased
decomposition, releasing more carbon into the atmosphere than the plant community
takes up via photosynthesis.
2.6.3. Global implications
Despite being the single largest terrestrial carbon store, peatlands are often
excluded from climate models and emissions scenarios, which may be problematic, as
Holden (2005) estimates that carbon sequestration by peatlands over the past 10,000
years has reduced global temperatures by 1.5–2º C (Limpens et al. 2008).
Northern latitudes are also disproportionately warming compared to lower
latitudes, and 80% of the world’s peatlands exist in these locales, containing an estimated
27
547 Pg of carbon (Holden, 2005). This suggests a possible positive feedback loop as
warming causes the loss of bog habitat, which then results in the decomposition and
release into the atmosphere of the carbon stored within them, which will in turn cause
global temperatures to rise even further. Ise et al. (2008) projects losses of 40% of the
carbon in shallow peat layers and 90% losses in deep layers if the global climate warms
4º C, suggesting a catastrophic positive feedback to global warming.
Alternatively, some studies have shown that an increase in temperature may lead
to negative climate feedbacks in bogs. As temperatures rise to about 27º C, the net
photosynthetic rate of Sphagnum increases so long as sufficient water is present (Bu et al.
2011). In fact, a number of studies have shown increases in Sphagnum abundance as
temperatures rise, even as the abundances of vascular species are reduced, although too
much warming leads to the death of Sphagnum communities and the dominance of
vascular plants (Bu et al. 2011). Additionally, high-latitude bogs may expand as the
climate warms and their rate of carbon sequestration may even increase as well (Charman
et al. 2013)
There are also studies showing no correlation between temperature and peat
formation. Novak et al. (2008) studied five bogs in the UK and the Czech Republic and
used pollen records to construct regional climate models. They found that while
temperature was related to carbon storage in the short-term (<50 years) in all five sites,
there was no relationship between temperature and carbon storage over the past 150
years. Wieder et al. (1994) found similar results in their study of five Sphagnumdominated peatlands in North America.
28
2.7. Present study contributions
2.7.1. Determine topographic patterns of carbon distribution
Topographic carbon distribution has only been addressed in a handful of studies,
none of which were performed in the Pacific Northwest. This research will increase our
understanding of where carbon is deposited in an ombrotrophic bog by looking at various
factors other than Sphagnum cover. In particular, it will attempt to show the relationships
between carbon concentrations within surface soils and vascular plant communities,
water table depth, microtopography, dissolved oxygen concentrations, and nitrogen levels
(which may, in turn, be influenced by these other factors). Additionally, it will further our
understanding of PNW bogs, potentially elucidating their similarities and differences
from those in other regions.
2.7.2. Provide avenues of future research
This research may reveal relationships in two general categories that stimulate
future research: the spatial relationships found within bogs in general, and the
relationships within Pacific Northwestern bogs.
Ombrotrophic bogs are still poorly understood, and while research has been
performed on spatial relationships within several bog systems, few of these studies have
looked at these relationships on as fine a scale as I intend to. Additionally, all of the
studies I have found have either been manipulative experiments looking at a single
relationship (e.g. nitrogen input vs. carbon efflux), or have reconstructed the climatic and
vegetative history of a site using proxy data, whereas I intend to look at carbon and
nutrient content within the acrotelm layer and associated vegetative data.
29
Most research has also focused on bogs outside of the Pacific Northwest. The
ecology and carbon balance of bogs is highly variable depending on regional climatic and
soil conditions, and what we know about bogs in one part of the world does not
necessarily apply to bogs elsewhere. By looking at a bog in the Pacific Northwest, I may
find relationships unique to the region.
2.7.3. Help understand the consequences of climate change and other human activities
Bogs are currently under threat from a variety of human activities, including peat
mining (for peat products and for fuel), bog drainage (for different land use), and
pollution (particularly N and P). Understanding the processes in bogs may allow us to
better mitigate the deleterious effects of human activities on them, inform policy relating
to bogs, devise improved restoration strategies for bogs that have already been degraded,
promote carbon sequestration in existing bogs, and possibly inform a strategy towards
preventing a massive CO2 flux into the atmosphere as peat trapped within permafrost at
northern latitudes thaws.
2.8 Conclusion
Ombrotrophic bogs are highly variable systems that can display a range of
responses in carbon flux and storage when faced with environmental changes. In general,
lower temperatures, higher water tables, and nutrient limitation promote carbon
sequestration, but regional processes are more important than global ones and individual
bogs may deviate from these patterns depending on their specific soil chemistry,
vegetative community, nutrient inputs, and landscape structure. Because of this high
30
variability, research on ombrotrophic bogs is needed on the regional level within the
Pacific Northwest in order to predict the effects of climate change, reveal previously
unknown relationships, and to determine what course of action, if any, should be taken to
protect our bog systems.
31
3. Methods and Materials
3.1 Field collection
The study site consisted of a bog within the North Bay Natural Area Preserve
wetland system, which is located in Gray’s Harbor county in Washington state, west of
Hoquiam. This wetland system consists of 1,215 acres of various wetland systems,
including salt marshes and a Sphagnum bog. It is surrounded by a temperate rainforest
system averaging ~70 inches of precipitation a year and is dominated by Douglas-Fir
(Pseudotsuga menziesii), Western Hemlock (Tsuga heterophylla), and Western Redcedar
(Thuja plicata). Also located nearby are a number of privately owned wetlands, including
several cranberry bogs.
This site was chosen for three reasons: ease of access, the presence of vegetation
of interest (Sphagnum spp., Drosera rotundifolia, graminoids) and relative lack of human
disturbance. This particular site fit all three criteria better than other sites considered in
Western Washington, although it was located adjacent to a rural road and trash was
observed within a buffer of woody shrubs separating the road from the bog complex.
Sampling occurred between February 6th, 2016 and February 10th, 2016.
The study site consisted of a 100 m baseline with ten transects ranging from 61–
100 m in length. Because of the trash observed within the buffer, the baseline was set 50
m away from the buffer to minimize the impact of human disturbance on the samples
taken. Transects were spaced 10 m from each other and perpendicular to the baseline,
with the first one being placed randomly along the baseline using a random number
32
generator. Each was checked to ensure heterogeneity of vegetation, microtopography,
and inundation, and to ensure there were no obvious disturbances present.
There were a total of 25 sample units (SU) placed along the transects. Two
random SUs were placed on each transect, while the remaining five SUs were randomly
assigned a transect and a point along the transect using the same process as the first
twenty SUs. A new random number was generated if an SU fell within 10 m of another
SU, and if an SU was completely barren (>95% free of vegetation) or inundated, a new
location for the SU was chosen. The number of sample units was agreed upon in
conjunction with the State of Washington Department of Natural Resources to prevent
unnecessary damage to the wetland complex.
Each SU consisted of a 0.5 m2 quadrat placed so that the right side of the frame
overlapped with the meter tape and the upper right hand corner of the frame was
underneath the meter number assigned to the SU. Within these quadrats, the following
measurements were made in this order so as not to disturb the SUs before data were
collected: water chemistry, vegetation functional groups, microtopography, and soil
cores.
Additionally, living Sphagnum thickness was measured using a meter stick on
both sides of the quadrat, with a final value consisting of the average of both values.
Water depth was also measured at the deepest point in each SU using a meter stick. All of
these measurements will be elaborated upon within the following sections.
33
3.1.1 Water chemistry
Temperature, dissolved O2, and pH were characterized in each SU using a Yellow
Springs Instrument Pro2030 meter and an Oakton pH meter with an Oakton 35641-51
electrode, respectively. These values were taken before any other sampling so as to
minimize disturbance within the SU, and they were placed either near the center of the
SU or where the water was deep enough to cover the electrodes. Values were given time
to equilibrate (~10 minutes) before final values were recorded as data. The electrodes
were thoroughly rinsed with DI water after each measurement was taken, and the pH
meter was calibrated before each sampling event with three standards: pH 3, 5, and 7.
3.1.2 Vegetative sampling
Vegetation functional groups (Sphagnum spp., carnivorous plants, noncarnivorous vascular plants, graminoids) were marked as either present or absent in each
SU. These functional groups were chosen because Sphagnum is considered an ecosystem
engineer and has dramatic, well-defined impacts on wetland systems, graminoids are also
common wetland plants that have well-defined impacts on wetland systems, and nongraminoid vascular plants are also common, and their presence may have differing
impacts than the graminoids.
If cover for a functional group within each SU was qualitatively estimated to be
less than 5%, it was not counted as being present with the exception of carnivorous
plants. Because Drosera rotundifolia, the carnivorous plant present at this site, forms a
tiny, difficult to see hibernaculum in winter, it was counted as present if observed within
34
an SU regardless of its estimated cover, as it is impossible to estimate vegetative cover of
this plant unless it is its active stage.
Vegetation for the entire study site was also characterized using the pointintercept method using the same functional groups, save for the carnivorous plants. Using
the same SU numbers as for the sample plots, points every 0.5 m for 10 m were recorded
as either hits or misses for each functional group, for a total of 20 points in each SU and
500 points in the entire site, following the methods of Canfield (1941). Cover estimates
for each group were then obtained via the following equation, where a “hit” and “miss”
are defined as the presence or absence of vegetation directly below the point along the
tape, respectively:
# ℎ𝑖𝑡𝑠 𝑔𝑟𝑜𝑢𝑝 𝐴
𝐶𝑜𝑣𝑒𝑟 𝑜𝑓 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑎𝑙 𝑔𝑟𝑜𝑢𝑝 𝐴 = (
) 𝑥 100
𝑇𝑜𝑡𝑎𝑙 # 𝑝𝑜𝑖𝑛𝑡𝑠
This data was not used for actual analysis, but was taken to ensure the site contained a
heterogeneous mix of vegetative groups, to ensure the vegetative community was of the
correct type for an ombrotrophic bog, and to ensure the site chosen was representative of
the bog complex as a whole.
3.1.3 Soil cores
Soil cores were taken in each SU after all other sampling was finished. A 1.27 cm
(0.5 inch) diameter stainless steel tubular soil corer was used for each soil core. Two soil
cores were taken 2-5 cm in depth (due to difficulty in coring peaty soils with narrow
corers) from opposite sides of the quadrat in each SU, just below the litter where the
acrotelm begins. These cores were then combined in a plastic bag, and placed in a cooler
35
filled with ice until it could be brought back to the lab for freezing at the end of the day.
The soil corer was thoroughly rinsed with DI water in between SUs.
Some inconsistency is present with the first five soil cores. First, 5 shallow soil
cores were taken from each SU and combined instead of two. The method was modified
due to challenges in using the soil corer within a peaty bog. Second, the samples were
placed in a refrigerator/incubator upon return to the lab at the end of the day. These
samples were moved to a freezer two days later and stored over the next two months, but
the temperature in the refrigerator was raised to just above room temperature sometime
during these two days. However, soil N and C analyses did not reveal any obvious
abnormalities compared to the other 20 samples that did not undergo this inconsistency,
and they were thus included in all results except where stated otherwise.
3.2 Lab analyses
Except for the first five samples collected, all of the soil samples were placed in
the freezer upon being homogenized the same day they were collected. Samples were
homogenized via a milkshake mixer and placed inside 500 mL HDPE bottles before
being frozen. Any living plant material present within the soil samples was also removed
by hand at this point. Living plant material was defined as green vegetation, as well as
red in the case of Sphagnum.
A portion of each sample was weighed and then dried by placing it in an oven set
to 110º C (±5º C) for 16 hours, weighed again, and then placed within a desiccator for
future C:N ratio analysis.
36
Soil moisture percentage was calculated at this point via the following equation,
where W1 and W2 equal the weights of the soil before and after drying, respectively
(Delaune et al. 2013):
𝑀𝑜𝑖𝑠𝑡𝑢𝑟𝑒 𝑐𝑜𝑛𝑡𝑒𝑛𝑡 (%) =
(𝑊1− 𝑊2 )100
𝑊1
3.2.1 Soil pH
Soil pH was determined in duplicate for each sample using an Orion 601A
Ionalyzer pH meter with an Oakton 35641-51 electrode using a method adapted from
Hendershot et al. (2008). Briefly, approximately 2.00 g of dried sample was placed
within a 50 mL beaker. 20 mL of deionized water was then pipetted into the beaker and it
was then agitated for half an hour, left to sit for an hour, and the pH was then measured.
The pH meter’s calibration was checked every ten samples and recalibrated as necessary.
Per Hendershot et al.’s method (2008), if duplicates registered a difference in pH by more
than 0.20 new subsamples were measured. The mean of both readings was used as the
final pH value for statistical purposes.
3.2.2 Nitrate
Nitrate was extracted from soil samples using DI water per EPA Method 300.0
(1993). Approximately 10 g of field wet sample was measured into a beaker, and then 50
mL of DI water was pipetted into it. Samples were mixed for 15 minutes and then filtered
using a 0.45 μm filter (Whatman no. 42) into 60 mL polypropylene bottles. These
37
samples were then frozen until analysis, which was approximately one week. Samples
were then thawed and analyzed via ion chromatography using a Dionex 25A ion
chromatograph. A standard curve was constructed using 0.15 mg/L, 0.25 mg/L, and 0.50
mg/L NO3- standard solutions with an R2 value of 0.999. Samples 3, 8, 10, and 20 were
determined to be above the range of the high standard and were diluted with a 4:1 ratio of
DI water and filtrate and re-run. NO3- concentrations were then standardized to mg/gdw
(gram dry weight) sample.
3.2.3 Ammonium
NH4+ was extracted from soil samples by weight approximately 2 g of wet sample
into an Erlenmeyer flask, pipetting 20 mL of a 2 M KCl solution, and agitating the flask
for 30 minutes (Maynard et al. 1993). The sample was then filtered using a 0.45 μm filter
(Whatman no. 42) into a 60 mL polypropylene bottle and frozen. Samples were then
thawed and analyzed by UV-visible spectroscopy using an Agilent 8453 diode array at a
wavelength of 640 nm. This analysis was adapted from a colorimetric analysis method
using sodium salicylate in place of phenate by Le and Boyd (2012), with the addition of
the EPA’s extraction method. A standard curve was constructed using 0.1 mg/L, 0.5
mg/L, and 1.0 mg/L NH4+ standard solutions with an R2 value of 0.997. Actual NH4+
concentrations were computed via the following equation, where C = NH4+ concentration
(mg/L), A = absorbance, and s and u are the standards and unknowns, respectively:
𝐶𝑠 𝐴𝑠
=
𝐶𝑢 𝐴𝑢
38
NH4+ concentrations were then standardized to mg/gdw of sample.
3.2.4 Organic carbon & nitrogen
Total organic carbon and nitrogen content of the soil samples were determined by
dry combustion using using a Perkins Elmer Series II 2400 CHNS/O Analyzer with a
Perkins Elmer AD6 Autobalance (Batjes 1996). 8x5 mm tin capsules were weighed to the
nearest 0.001 mg, packed with 1.5–2.5 mg of dry sample, weighed again, and combusted
within the CHN analyzer to determine the C:N ratio, total nitrogen, and total carbon of
each sample. NIST SRM 1547 Peach Leaf standard was used as a conditioner, and
acetanilide was used as a K-factor. Because these samples were so acidic, inorganic
carbon was assumed to be absent, and thus the total carbon value reported by the
instrument was assumed to be entirely organic.
Organic nitrogen content was determined by subtracting the NH4+ and NO3- from
the total nitrogen content of each sample. Inorganic nitrogen content was determined by
summing the NH4+ and NO3- content of each sample. NO2- is generally not detectable in
peatlands and was thus not included in this analysis (Sheppard et al. 2013).
3.3 Statistical analyses
JMP Pro 12 and Microsoft Excel were used for all statistical analyses. Categorical
variables paired with continuous variables were analyzed via one-way ANOVA, while
continuous variables paired with continuous variables were analyzed via linear
regression. Data were checked for normal distribution and analyzed via the nonparametric Kruskal-Wallis test and Spearman rank correlation in addition to ANOVA and
39
linear regression, respectively, to determine if significance differed between statistical
tests. Significance was set at α = .05 for all tests.
3.3.1 Treatment of data
The ion chromatograph used to analyze NO3- samples was unable to differentiate
rinse blanks from sample concentrations below 0.10 mg/L. Sample numbers 2 and 5
registered NO3- values below this concentration and were treated as if concentrations half
the detection limit, or 0.05 mg/L. Per Croghan & Egeghy (2003), this method of data
substitution should not introduce significant bias into the results, but statistical tests
involving NO3- and total inorganic N were performed with and without these two samples
to ensure that this was the case.
Because multiple microlandforms were present within each quadrat,
microtopograhy was characterized as either low, medium, or high based on the average of
the landforms. Hummocks were considered high, lawns medium, hollows low, and the
single plot that was almost completely inundated was also treated as low. Qualitative
observations were also used to characterize terrain, particularly when a specific
microlandform made up an overwhelming majority of the plot.
While four vegetative functional groups were surveyed, only Drosera and
Sphagnum were present and absent in enough plots to perform statistical analysis. All but
one plot contained vascular plants, and all but three contained graminoids. Thus, they
were not included in analyses.
40
4. Results
4.1 Vegetation
Graminoids and other vascular plants were present in nearly all plots, but
qualitative estimates of cover indicated that graminoids were dominant overall and were
most common in the more inundated portions of the site, while Sphagnum tended to be
dominant near the less inundated Northern and Southern portions as well as the raised
section near the Western boundary of the site. Additionally, Sphagnum tended to be more
clumped, with a smaller overall distribution but a tendency to dominate the areas it was
present in. Vascular plants, while found in the most plots, did not contribute much overall
cover, as they mostly consisted of small shrubs such as Bog Cranberry (Vaccinium
oxycoccos), Labrador Tea (Rhododendron groenlandicum), Bog Rosemary (Andromeda
polifolia), etc. Overall, the plots’ vegetative composition largely matched that of the site
at large. Vegetative data in study plots and for the site as a whole are summarized in table
1.
Vegetation Type
# Sample plots present
% of plots present
Actual site % Cover
Graminoids
22
88%
71.6% ±16.2%
Vascular
24
96%
40.2% ±19.3%
Sphagnum
18
72%
35.4% ±19.1%
Drosera
9
36%
NA
Table 1: Summary of vegetation data. The “Vascular” column includes all vascular plants except for
Drosera and graminoids. Actual site data % cover was derived from can be found in Appendix 1. ________
Graminoids and vascular plants were nearly ubiquitous to the study plots. As a
result of this, their interactions with the other two functional groups and the other
physical and chemical measurements were unable to be determined.
41
4.1.1 Physical and chemical summary data
Sphagnum’s impact on physical and chemical properties such as soil pH and soil
moisture are well known, but the presence of Drosera also appears to vary with some of
these factors (table 2). Additionally, across all plots studied, dissolved O2 values ranged
from 2.7–9.5 mg/l, water depth from 0–15 cm, Sphagnum depth from 0–9.5 cm, soil
moisture from 90.96–95.69%, and soil pH from 4.71–5.79.
Physical/chemical Factor
Site Total
Dissolved O2 (mg/l)
Water depth (cm)
Sphagnum depth (cm)
Soil moisture %
Soil pH
7.49±1.87
5.16±4.88
3.62±3.23
94.04±1.06
5.46±0.26
Sphagnum
Absent
Present
7.87±0.78 7.17±0.72
8.71±1.79 3.39±1.05
N/A
5.03±0.55
93.11±0.34 94.41±0.21
5.33±0.10 5.50±0.06
Drosera
Absent
Present
6.74±066 8.37±0.71
4.31±1.30 5.89±1.74
4.00±0.81 2.94±1.09
94.37±0.24 93.46±0.33
5.44±0.06 5.48±0.10
Table 2: Summary of physical and chemical data by the presence and absence of plant functional groups. ±
values indicate standard error for each value. Note that dissolved O2 measurements were made in plots
with standing water only. _________________________________________________________________
4.1.2. Sphagnum
Even though soil moisture differences were slight, soil moisture in plots where
Sphagnum was present was higher (94.41% ±0.69%2) than in plots where it was absent
(93.11% ±1.30%) (Fig. 2; t(23) = 3.27; p < 0.01), with soil moisture increasing as moss
depth increased (Fig. 3; R = 0.4750). Soil pH, by contrast, did not differ in plots where
Sphagnum was present from those where it was absent.
2
± Values within text indicate one standard deviation
42
Figure 2: Soil moisture is significantly greater in the presence of Sphagnum (p < 0.01).
43
Figure 3: Soil moisture increases as Sphagnum depth increases (p < 0.05)
4.1.3. Drosera
The presence of Drosera was not affected by the depth of Sphagnum or standing
water. Drosera’s presence was affected by the presence of Sphagnum, with Drosera
showing a preference for plots without Sphagnum (X2 (1, N = 25) = 5.226; p < 0.05).
Presence of Drosera also indicated lower soil moisture (93.46% ±1.40%) than plots
without Drosera (94.37% ±0.65%) (Fig. 4; F(1, 23) = 4.94 ;p < 0.05)). There was no
association between soil pH and the presence of Drosera nor was there a significant
correlation between Drosera and dissolved O2, although there was a modest, nonsignificant positive association (t(11) = 1.67 ; p > 0.10).
44
Figure 4: The presence of Drosera is associated with lower soil moisture (p < 0.05).
4.2 Microtopography
Most plots contained multiple microlandforms, and all plots contained at least one
of the microlandforms, except for plot #10, which was almost completely inundated and
was classed as “low” topography (table 3).
Microlandform
# Plots present
% of plots
Hollow
11
44%
Lawn
13
52%
Hummock
14
56%
Table 3: Summary microtopography data. Most plots contained more than one microlandform, while one
plot was almost completely inundated and did not contain any of the three microlandforms. Each plot was
45
standardized based on average landform data into one of three categories: “low,” “medium,” and “high.”
Landform data can be viewed in Appendix 2.
Sphagnum presence (Fig. 5; p < 0.01) and thickness (Fig. 6; F(2, 22) = 3.62; p <
0.05) showed a strong positive association with average microtopography height. The
presence of Drosera showed no such association with average microtopography height (p
> 0.05).
Figure 5: Sphagnum presence is strongly associated with landform height (p < 0.01). See appendix B for
how average microtopography was determined.
46
Figure 6: Sphagnum increases with landform height (p < 0.05). Graph shows 95% confidence interval
for each value.
Average microtopography height was not associated with either soil pH or
dissolved O2 in standing water.
4.3 Nitrogen
Total N content was largely dictated by organic N, which made up greater than
99% of the total N present in soils (table 4). Nitrogen results will be discussed beginning
47
with inorganic nitrogen content (NH4+ and NO3-) and ending with an examination of
variation in organic N content.
N Type
Total N
Organic
Inorganic (NH4+ + NO3-)
Ammonium
Nitrate
Mean (mg/g)
24.08
23.91
0.17
0.127
0.043
Std. Deviation (mg/g)
3.21
3.18
0.06
0.04
0.06
% Total N
100.00
99.30
0.70
0.53
0.17
Table 4: Summary of N content in soil samples. Organic N makes up the vast bulk of the N present within
the soil samples, with ammonium making up the majority of the inorganic N present.
4.3.1 Inorganic nitrogen
Total inorganic N (NH4+ + NO3-) made up less than 1% of total soil nitrogen
content, with a mean of 0.17 mg/g and a range of 0.10 mg/g to 0.34 mg/g. It was not
associated with soil moisture, Sphagnum depth, water depth, or soil pH. However, NH4+
concentration was negatively correlated with water depth (R = -0.57; p < 0.01) and NO3concentration was positively correlated with water depth (R = 0.54; p < 0.01) (Fig. 7).
48
Figure 7: While total inorganic N is not significantly correlated with water depth, both NH4+ and NO3are, but in opposite ways. NH4+ is negatively correlated with water depth (p < 0.01), while NO3- is
positively correlated with water depth (p < 0.01).
Similarly, NH4+ and NO3- were both significantly correlated with soil pH, but in
opposite ways. NH4+ appears to increase with soil pH (R = 0.49; p < 0.05), while NO3appears to decrease as water depth increases (R = -0.38; p = 0.0599) (Fig. 8). Total
inorganic N, however, was not significantly associated with soil pH.
49
Figure 8: While total inorganic N is not significantly correlated with soil pH, NH4+ is, and NO3- shows a
non-significant correlation in the opposite direction. NH4+ is positively correlated with soil pH (p =
0.0128), while NO3- is negatively correlated with soil pH (p = 0.0599).
Inorganic N and NO3- were not significantly associated with the dissolved O2
content in plots containing standing water, but ammonium was (R = -0.62; p < 0.05)
(Fig. 9).
50
Figure 9: NO3- is not significantly correlated with dissolved O2, but NH4+ is (p < 0.05), and total
inorganic N shows a non-significant correlation (p = 0.1031).
The presence of Drosera was negatively associated with inorganic N
concentrations (0.14 mg/g dry weight ± .02 mg/g in the presence of Drosera and 0.18
mg/g ±0.01 mg/g in its absence; (t(23) = 2.52; p < 0.05) (Fig. 10).
51
Figure 10: The presence of Drosera indicates lower inorganic N values than in plots where it was absent
(p < 0.05). NO3- and NH4+ alone were not significantly different between plots. Graph shows 95%
confidence intervals for each value.
Microtopography also did not show any association with total inorganic N levels,
although “low” topography was associated with significantly higher concentrations of
NO3- (F(2, 22) = 3.69; p < 0.05), and “high” topography was non-significantly
associated with higher concentrations of NH4+ (F(2, 22) = 2.71; p = 0.0884) (Fig. 11).
52
Figure 11: Total inorganic N concentrations are not affected by microtopography, but NH4+
concentrations were higher in “low” topography areas than in the others, although this relationship was
not signifacant (p = 0.0884). NO3- concentrations were significantly higher in “high” topography than
they was in “low” topography (p < 0.05). Graph shows 95% confidence interval for each value.
Inorganic N concentrations were also higher in the presence of Sphagnum (0.19
mg/g dry weight ±0.02 mg/g) (t(23) = 1.29; p = 0.2091) than in its absence (0.14 mg/g
dry weight ±0.01 mg/g) (t(23) = 1.29; p = 0.2091), but there was no relationship between
Sphagnum depth and inorganic N. Further analysis showed that NH4+ was significantly
associated with the presence of Sphagnum but not its depth (t(23) = 2.23; p < 0.05).
Results are summarized in Fig. 12.
53
Figure 12: Inorganic N concentrations are higher where Sphagnum is present, although not
significantly so (p = 0.2091). NO3- was not significantly correlated with the presence or absence of
Sphagnum, while NH4+ concentrations were significantly higher where Sphagnum was present (p <
0.05). Graph shows 95% confidence intervals for each value.
4.3.2. Organic nitrogen
Mean organic N concentration was 2.39% of dry soil weight, with a range of
1.81% to 3.14%. Organic N was not found to be significantly associated with any
chemical variables except organic C (R = 0.79; p < 0.001) and inorganic N (R = 0.51; p
< 0.01), where it was positively correlated with both (Fig. 13).
54
Figure 13: Organic C and N are positively correlated (p < 0.001), as are organic N and inorganic N (p <
0.01).
Though there was a modest negative association between the presence of Drosera
and organic N levels, it was not significant (t(23) =1.71; p = 0.1007).
Microtopography was not associated with organic N concentrations.
4.4 Organic Carbon Concentrations
The concentration of organic carbon averaged 420 mg/g ± 59 mg/g of dry weight,
with a range of 310 mg/g to 509 mg/g. Concentrations were not significantly affected by
soil pH or dissolved O2, although a positive non-significant relationship was found
55
between organic C concentrations and dissolved O2 concentrations (R = -0.59; p =
0.1607), as well as between organic C and soil moisture % (R = 0.38; p = 0.0589). Water
depth was negatively correlated with organic C concentrations (R = -0.44; p < 0.05) (Fig.
14).
Figure 14: Organic C concentrations and water depth show a significant negative correlation (p < 0.05).
“Depth” refers to the maximum depth of standing water measured within the sample plot.
The presence of Sphagnum as well as its depth were not significantly associated
with organic C concentrations. However, when samples 1-5 were excluded from the
dataset, a non-significant association between Sphagnum’s presence and carbon
56
concentration appears (t(18) = 1.89; p = 0.0751) (Fig. 15). These samples were excluded
due to the additional incubation they underwent prior to analysis (see section 3.1.3).
Figure 15: Soil C concentrations are non-signficiantly associated with the presence of Sphagnum when
the first five samples are excluded from the dataset (p = 0.0751).
The presence of Drosera showed a modest negative association with organic C
concentrations, but this was not significant (t(23) = 1.46; p = 0.1572).
Microtopography height was not significantly associated with organic C
concentrations.
57
4.5 C:N Ratio
The mean C:N ratio was 17:1, with a range of 21:1 to 15:1. Soil moisture was
positively correlated with the C:N ratios of the soil samples (R = 0.48; p < 0.05), while
water depth (R = 0.38; p = 0.0609) and Sphagnum depth (R = 0.34; p = 0.1012) were
negatively correlated with them, but only soil moisture significantly so (Fig. 16). There
was no relationship between the presence of Drosera and C:N ratios.
Figure 16: C:N ratios are positively correlated with soil moisture (p < 0.05) and moss depth (p = 0.1012),
while they are negatively correlated with water depth (p = 0.0609). Only soil moisture shows a
significant correlation.
58
5. Discussion
Overall, many of the relationships between internal variables found within the
study site are consistent with relationships found within bogs in other regions of the
world, while a few inconsistencies may be explained by unique soil chemistry or
vegetative characteristics. However, the site was less acidic than is typical in bogs with
most plots registering a pH above 5, where most bogs fall within a pH range of 3-5 (Vitt
2013). In addition, organic nitrogen content is abnormally high, causing the C:N ratios to
be unusually low, which is possibly due to unique vegetative and/or soil characteristics of
the region, as discussed further below. In aggregate these results may indicate that this
site is atypical or that the soil chemistry within the region is atypical compared to sites in
regions that have undergone more study.
5.1 Carbon
The average carbon concentration of the site (420 ± 59 mg C/gdw) fell within
both the worldwide average for peatlands (492 ± 24 mg C/gdw)3 and the averages of
nearby regions (Loisel et al. 2014). This suggests that the study site’s ability to
accumulate organic carbon within the acrotelm is similar to that of peatlands in other
regions, although these results do not reflect the long-term storage potential of this
carbon, as only the top half of the acrotelm was sampled, and the vast bulk of
decomposition ( > 99%) occurs within this layer (Blodau et al. 2007).
Carbon concentrations decreased as dissolved O2 concentrations increased, but
this was statistically insignificant (p > 0.15). Significance may be achieved with a higher
3
Note that this value excludes tropical peatlands, which undergo very different carbon cycling regimes
than do northern peatlands and are also much less efficient carbon sinks (Holden 2005)
59
sample size, as a lack of dissolved O2 indicates anoxic soil conditions, which should
result in less respiration and more carbon retention, although it is important to note that
dissolved O2 in soils was not actually looked at, so the dissolved O2 values observed for
standing water may be completely unrelated to O2 concentrations within the soil column
(Vitt 2013). Additionally, one of the sampling events occurred during a rain event, and
one occurred just after a rain event. The physical agitation that precipitation causes and
the higher dissolved O2 content of rainwater may have temporarily elevated dissolved O2
and water levels in plots sampled at these times, potentially reducing the ability to detect
this expected relationship. Wind speeds ranged from 12 mph to 23 mph during the
sampling period as well, which may have resulted in significant mixing and aeration of
the waters on site.
Additionally, while not statistically significant, there was a positive correlation
between percent soil moisture and organic carbon concentrations. The actual variation in
carbon concentrations is probably due to the presence of Sphagnum, which is also
associated with higher soil moisture (see Fig. 15). When the presence of Sphagnum was
controlled for between soil moisture and organic C concentrations, a partial correlation of
r = 0.6074 (p = 0.1480) was found between the two variables in plots where Sphagnum
was absent, and both the strength and certainty of this correlation drops dramatically in
plots where Sphagnum was present (r = 0.1027; p = 0.6850). This likely indicates that
Sphagnum largely controls organic C concentrations, as Sphagnum’s positive association
with soil carbon levels is well-documented (see section 2.3.1). This is in agreement with
the literature, which shows Sphagnum reduces the decomposition of existing carbon
within the soil column, represents a less labile source of carbon than other plants, and
60
promotes the acidic conditions required for the formation of peat (van Breeman 1995,
Adkinson and Humphreys 2011, Leifeld 2011).
Interestingly, no relationship was found between microtopography and carbon
concentrations. This is in contrast to the findings by Laine et al. (2006), who found that
higher microtopography represented a larger net carbon sink than lower microtopography
due to increased levels of photosynthesis, leading to increased carbon sequestration. This
discrepancy may be attributed to differences in site-specific variables between this study
site and Laine et al.’s or an inability to detect differences between microtopographic
features in this study due to a small sample size.
Likewise, a negative trend between water depth and carbon content was detected.
This was not expected, as standing water reduces the rate of respiration and CO2 efflux,
resulting in larger amounts of carbon retention within soils (Charman 2015, Blodau et al.
2007, Laine 2006). As previously mentioned, much of the sampling was done during and
after rain events, so it is possible that water depth was temporarily increased throughout
the bog or that standing water existed where it doesn’t normally, potentially skewing the
results. Sampling was also more difficult in inundated areas, leaving more room for
researcher error. Additionally, many of the inundated areas had little or no vegetative
cover, which, particularly in the case of Sphagnum, could lead to higher rates of peat
decomposition (Ise et al. 2008). The vegetation that did exist in these areas were typically
graminoids and vascular plants, which represent a much more labile source of carbon
than does Sphagnum (Dorrepaal et al. 2005).
Carbon concentrations within the surface soils were positively correlated with
both inorganic and organic N content (see Fig. 13). As reported in Basiliko et al. (2006)
61
and Wang et al. (2014), increases in available inorganic N tends to lead to increases in the
organic C content of soils. However, Basiliko et al. attributes this increase to the soil’s
microbial community becoming larger but also more efficient at respiration, leading to an
increase in total soil organic C, which is beyond the scope of this study. Wang et al., by
contrast, looked at historical records within a bog and found that the increases in N likely
led to more vegetative growth, resulting in larger C additions to the underlying peat
structure.
5.2 Nitrogen
Total N content in all soil samples were high, exceeding 2% in most cases. This
may indicate that a larger proportion of the peat is made up of graminoid and vascular
plant matter rather than Sphagnum, as Sphagnum peat tends to contain less than 1% total
N, while graminoid peat can contain up to 3% total N (Crum and Planisk 1992). This
agrees with Lacourse and Davies (2015), who also found elevated N concentrations in
comparison to previously studied sites, although the N values found here are even higher.
This conclusion is supported by the graminoid dominance of the site, although
interestingly, there was no significant difference in total N between plots where
Sphagnum was present than in those where it was absent.
The presence of Drosera also showed a modest, non-significant negative
correlation to organic N concentrations. This result was expected, as Drosera utilizes the
N trapped within an insect’s tissues, but it is also capable of extracting N from soils with
its roots (Millett et al. 2003). Likewise, inorganic N concentrations were significantly
lower in plots containing Drosera than in those without, suggesting that Drosera are
62
utilizing soil nutrients. However, because associations between graminoids and other
vascular plants with inorganic N could not be determined, this difference may be
associated with the absence of Sphagnum rather than the presence of other functional
groups, or with the relative abundance of the functional groups, which was not studied.
5.2.1 Inorganic N
Total inorganic N did not show a significant relationship with water depth.
However, when NH4+ and NO3- were looked at individually, both displayed a significant
relationship with water depth, but in opposite ways (see Fig. 7). Interestingly, NH4+
decreased as water depth increased, while NO3- increased along with water depth, which
is the opposite of what one would expect, as NO3- requires the presence of oxygen to
form, the availability of which tends to be lower in inundated areas compared to those
exposed to the atmosphere. This may be a result of differences in preferential uptake of
NH4+ and NO3- between species, which unfortunately has not been thoroughly studied, so
it is unknown whether this is the case for the species found at the study site, although
Wang and Macko (2011) found that plants in wetter environments preferred NH4+ over
NO3-, which is consistent with the relationship between water depth and inorganic N.
NH4+ concentrations were significantly higher in plots containing Sphagnum, while NO3concentrations in plots containing Drosera were non-significantly higher, suggesting that
Drosera and Sphagnum have different N uptake preferences (see figs. 10 and 12,
respectively). However, this experiment was performed hydroponically with European
deciduous forest species, so it is unknown if its findings apply to Pacific Northwest
63
wetland species. Additionally, once again, it is important to note that the influence of
other functional groups could not be established.
Similarly, total N was unrelated to soil pH, but NH4+ concentrations were
positively correlated with higher soil pH, and NO3- concentrations showed a nonsignificant negative correlation with soil pH (see Fig. 8). Olsson and Falkengren-Grerup
(2000) found that acid-tolerant plants preferentially uptake NH4+ and that less acidtolerant plants preferentially uptake NO3-, which may possibly explain why NH4+
concentrations increase with pH and NO3- concentrations decrease. Alternatively, N
mineralization has been shown to favor NH4+ when pH is lower and NO3- when pH is
higher, so this relationship may largely be a result of microbial activity (Cornfield 1952).
5.3 C:N Ratios
The C:N ratios of surface soils within the site were surprisingly low. Wang et al.
(2014) reported an average C:N ratio of 42:1 for peat in the acrotelm layer of their study
site, while Kuhry and Vitt (1996) reported C:N ratios of up to 300:1 within the acrotelm.
Lacourse and Davies (2015) reported C:N ratios as low as 20:1 in a bog in British
Columbia. This latter result is closest to those found in this study, but still higher, and
also falls within the low end of Loisel et al.’s (2014) meta-analysis of peatlands
worldwide. Even more surprisingly, the C:N ratios gathered by this study were even
lower than the most decomposed layers of peat in all of the aforementioned studies!
These results are not driven by low levels of organic C, however, as the site had relatively
normal concentrations of C, as discussed in section 5.1. Instead, it is being driven by high
N concentrations. A low C:N ratio can indicate excessive decomposition within a bog,
64
but in order to determine if this is the case here, soil cores looking at both the acrotelm
and catotelm would need to be taken (Clymo 1996).
Alternatively, because the bog is situated near a number of privately-owned
cranberry bogs, fertilizer runoff may have contaminated the site, contributing to
decomposition and/or artificially elevating N concentrations. This explanation is also
troubling, as Sphagnum loses its competitive advantage and produces fewer
decomposition-inhibiting polyphenols in the presence of elevated N concentrations,
potentially increasing CO2 efflux (Bragazza and Freeman 2007). However, this wouldn’t
increase organic N concentrations, which drove the high total N concentrations within the
site, making this explanation unlikely. Future studies should look at C:N ratios at various
depths within the peat column to determine whether the surface soils have abnormally
low C:N ratios compared to deeper soils and to determine if N signatures match those of
adjacent privately-owned wetlands, and whether or not these landowners apply fertilizers
with high levels of organic nitrogen (such as manure).
Wang et al. (2014) did, however, differentiate between soil microbes and the peat
itself, as the microbial community’s C:N composition was 10:1, significantly lower than
the peat itself. Because this study did not take into account soil microbe abundance, C:N
ratios may have been affected by higher than normal microbial growth, either due to
environmental factors or due to the period of unexpected thawing the samples were
exposed to prior to analysis. All of the samples were thawed and exposed to room
temperature conditions for up to five days due to a freezer power failure, so it is possible
that a significant level of decomposition and/or increase in soil microbes may have
occurred during this time. However, samples 1-5 were accidentally incubated for several
65
days prior to being frozen, and there was no statistical difference in C:N ratios between
these samples and the 20 that were not incubated.
Additionally, Loisel et al.’s 2014 meta-analysis, while not including this region,
reported average C:N values of 62:1 (±38) and 44:1 (±33) in Western Canada and
Alaska, respectively. All values were highly variable, but the results presented in this
study fall outside the range of the former and within the latter, suggesting that bogs in
coastal Washington may have more in common with Alaskan bogs than Western
Canadian bogs. It is worth noting that the study site’s total N content were well above the
ranges reported for both regions. However, these results may also simply be a result of
PNW soil variability as a whole. Carpenter et al. (2014) reports mean C:N ratios of 20:1
to 22:1 in temperate rainforest systems within Gray’s Harbor County, WA, the same area
as this study’s research site. As this study site was surrounded by temperate rainforest
typical for the area (Douglas-Fir and Western Hemlock dominant), the nitrogen values
for the soils may, in fact, be driven by processes outside of the bog itself.
5.4 Vegetation
Carbon concentrations were not associated with the presence or depth of
Sphagnum cover, which was not expected (see Fig. 15). However, samples 1-5 had been
accidentally incubated for several days, which may have artificially lowered their carbon
concentrations. When excluded, Sphagnum presence was positively associated with
carbon concentrations, albeit non-significantly. This is interesting, because Sphagnum is
the most important biotic component that influences carbon retention in bogs, but in this
study, it is not as strongly associated with carbon concentrations as one would expect
66
(Nichols et al. 2014). This may be due to small sample size, or it may indicate that other
plants (such as the graminoids) are the overarching control of carbon concentrations at
this site.
The presence of Drosera was not significantly associated with soil carbon
concentrations, and in fact showed a modest non-significant negative association. This
was unexpected as Drosera doesn’t utilize the carbon contained within the insects it
captures, as its digestive enzymes function solely to release the nutrients (largely N and
P, but also Ca, K, and other micronutrients) contained within the insect’s tissues (Ellison
and Gotelli 2001, Matušíková et al. 2005). Therefore, the carbon within the insects
should end up in the bog where normal bog processes would reduce the lability of that
carbon. This hypothesis was not borne out by the data, but there several possibilities as to
why this would be the case. First, Drosera was simply counted as “present” or “absent”
in each plot, regardless of the actual number of plants present, so a larger number of
Drosera may be required to detect a signal in the data. Second, Drosera were observed
during the winter when their leaves form a small, furled hibernaculum that is very
difficult to detect amongst the other foliage present, so it may be that Drosera were
missed in several plots. Third, Drosera has been shown to rely less upon carnivory to
attain nutrients when soil nutrients are high, and as discussed in section 5.2, N content in
soils at this site was high (Millett et al. 2015). Fourth, because Drosera are vascular
plants, it may be that the increased lability of their tissues overpowers the signal from the
carbon they collect. Fifth, microbial communities, other macroinvertebrates, or small
animals may consume the partially digested insect before it can be affected by the
biogeochemical processes within the bog that act to preserve carbon. And sixth, it is
67
possible that carbon derived from insects may represent an insignificant flux relative to
other carbon fluxes within the bog. Microbial community size and composition should be
evaluated in future research, as was the original intent in this study. Additionally, soils
should be tested for the presence and concentration of chitinase, as an absence of this
substance may suggest that Drosera is not, in fact, capturing insects (Matušíková et al.
2005).
68
7. Conclusion
In this study several variables were analyzed that contribute to carbon cycling in
bogs. Several of these variables were related to soil carbon concentrations in expected
ways, but some were not. Particularly, the C:N ratios of the soil samples were
exceptionally low when compared to the values cited in existing literature, and this is
potentially troubling in that it may indicate a higher rate of decomposition than is normal
for bogs in other regions, and thus it may be a weaker carbon sink or even a net carbon
source. However, because organic carbon concentrations are not abnormally low, it is
more likely that the soil chemistry at this site is simply significantly different from that of
sites in other regions, or that the peat at this site is largely made up of plant matter from
sedges and other vascular plants that tend to have lower C:N ratios than Sphagnum does.
Another surprising result was the negative correlation between soil carbon and
water depth, as the literature indicates that these should be positively, although only
surface soils were sampled, so this correlation may not hold up as peat depth increases.
Likewise, dissolved O2 was not significantly associated with soil carbon or inorganic
nitrogen content, when, again, it should have shown a positive and negative correlation,
respectively, based upon the literature. However, confounding factors may have affected
the data, casting doubt on the veracity of these results. Should this study be replicated,
care should be taken to sample during a year with a normal amount of precipitation and
during a time when the weather is unlikely to change.
Still, the results show that the study site is unique and differs substantially from
other sites that have been studied, even within the Pacific Northwest. While they can’t be
69
generalized to other bogs, or even the entire bog that the study site was part of, they
provide direction for future studies of this sort, and may provide hints at the kinds of
variables that uniquely affect bogs in our region.
70
References
Adkinson, A. C., and E. R. Humphreys. 2011. The response of carbon dioxide exchange
to manipulations of Sphagnum water content in an ombrotrophic bog. ECO
Ecohydrology 4:733–743.
Ball, P. 2004. Mysteries of bog butter uncovered. Nature News.
Basiliko, N., T. R. Moore, R. Jeannotte, and J. L. Bubier. 2006. Nutrient Input and
Carbon and Microbial Dynamics in an Ombrotrophic Bog. Geomicrobiology
Journal 23:531–543.
Batjes, N. h. 1996. Total carbon and nitrogen in the soils of the world. European Journal
of Soil Science 47:151–163.
Blodau, C., N. Basiliko, and T. R. Moore. 2004. Carbon turnover in peatland mesocosms
exposed to different water table levels. Biogeochemistry 67:331–351.
Blodau, C., N. T. Roulet, T. Heitmann, H. Stewart, J. Beer, P. Lafleur, and T. R. Moore.
2007. Belowground carbon turnover in a temperate ombrotrophic bog. Global
Biogeochemical Cycles 21:1–12.
Botch, M. S., K. I. Kobak, T. S. Vinson, and T. P. Kolchugina. 1995. Carbon pools and
accumulation in peatlands of the former Soviet Union. Global Biogeochemical
Cycles 9:37–46.
Bragazza, L., and C. Freeman. 2007. High nitrogen availability reduces polyphenol
content in Sphagnum peat. Science of The Total Environment 377:439–443.
71
Bragazza, L., C. Freeman, T. Jones, H. Rydin, J. Limpens, N. Fenner, T. Ellis, R. Gerdol,
M. Hájek, T. Hájek, P. Iacumin, L. Kutnar, T. Tahvanainen, and H. Toberman.
2006. Atmospheric nitrogen deposition promotes carbon loss from peat bogs.
Proceedings of the National Academy of Sciences 103:19386–19389.
van Breemen, N. 1995. How Sphagnum bogs down other plants. Trends in Ecology &
Evolution 10:270–275.
Bridgham, S. D., J. P. Megonigal, J. K. Keller, N. B. Bliss, and C. Trettin. 2006. The
carbon balance of North American. Wetlands 26:889–916.
Bubier, J. L., T. R. Moore, and L. A. Bledzki. 2007. Effects of nutrient addition on
vegetation and carbon cycling in an ombrotrophic bog. Global Change Biology
13:1168–1186.
Bu, Z., J. Hans, H. Li, G. Zhao, X. Zheng, J. Ma, and J. Zeng. 2011. The response of
peatlands to climate warming: A review. Acta Ecologica Sinica 31:157–162.
Canfield, R. H. 1941. Application of the Line Interception Method in Sampling Range
Vegetation. Journal of Forestry 39:388–394.
Carpenter, D. N., J. G. Bockheim, and P. F. Reich. 2014. Soils of temperate rainforests of
the North American Pacific Coast. Geoderma 230–231:250–264.
Chandler, G. E., and J. W. Anderson. 1976. Studies on the Nutrition and Growth of
Drosera Species with Reference to the Carnivorous Habit. New Phytologist
76:129–141.
Charman, D. J., M. J. Amesbury, W. Hinchliffe, P. D. M. Hughes, G. Mallon, W. H.
Blake, T. J. Daley, A. V. Gallego-Sala, and D. Mauquoy. 2015. Drivers of
72
Holocene peatland carbon accumulation across a climate gradient in northeastern
North America. Quaternary Science Reviews 121:110–119.
Charman, D. J., D. W. Beilman, M. Blaauw, R. K. Booth, S. Brewer, F. M. Chambers, J.
A. Christen, A. Gallego-Sala, S. P. Harrison, P. D. M. Hughes, S. T. Jackson, A.
Korhola, D. Mauquoy, F. J. G. Mitchell, I. C. Prentice, M. van der Linden, F. De
Vleeschouwer, Z. C. Yu, J. Alm, and I. E. Bauer. 2013. Climate-related changes
in peatland carbon accumulation during the last millennium. Biogeosciences
10:929–944.
Corbett, J., J. Chanton, D. Burdige, P. H. Glaser, D. I. Siegel, S. S. Dasgupta, M. M.
Tfaily, and W. T. Cooper. 2009. Using C/N ratios to investigate DOM reactivity
in northern peatlands. ResearchGate:0296.
Cornfield, A. H. 1952. The mineralization of the nitrogen of soils during incubation:
Influence of pH, total nitrogen, and organic carbon contents. Journal of the
Science of Food and Agriculture 3:343–349.
Croghan, C. W., and P. P. Egeghy. 2003. Methods of Dealing with Values Below the
Limit of Detection using SAS. EPA.
Crum, H. A., and S. Planisek. 1992. A Focus on Peatlands and Peat Mosses. University
of Michigan Press.
Dorrepaal, E., J. H. c. Cornelissen, R. Aerts, B. Wallén, and R. S. p. Van Logtestijn.
2005. Are growth forms consistent predictors of leaf litter quality and
decomposability across peatlands along a latitudinal gradient? Journal of Ecology
93:817–828.
73
Ellison, A. M., and N. J. Gotelli. 2001. Evolutionary ecology of carnivorous plants.
Trends in Ecology & Evolution 16:623–629.
Fritz, C., L. P. M. Lamers, M. Riaz, L. J. L. van den Berg, and T. J. T. M. Elzenga. 2014.
Sphagnum Mosses - Masters of Efficient N-Uptake while Avoiding Intoxication.
PLoS ONE 9:e79991.
Givnish, T. J., E. L. Burkhardt, R. E. Happel, and J. D. Weintraub. 1984. Carnivory in the
bromeliad Brocchinia reducta, with a cost/benefit model for the general restriction
of carnivorous plants to sunny, moist, nutrient-poor habitats. American
Naturalist:479–497.
Gorham, E., C. Lehman, A. Dyke, D. Clymo, and J. Janssens. 2012. Long-term carbon
sequestration in North American peatlands. Quaternary Science Reviews 58:77–
82.
Hendershot, W. H., H. Lalande, and M. Duquette. 2007. Soil Sampling and Methods of
Analysis, Second Edition. (M. R. Carter and E. G. Gregorich, Eds.). CRC Press.
Holden, J. 2005. Peatland hydrology and carbon release: why small-scale process
matters. Philosophical Transactions of the Royal Society of London A:
Mathematical, Physical and Engineering Sciences 363:2891–2913.
Ise, T., A. L. Dunn, S. C. Wofsy, and P. R. Moorcroft. 2008. High sensitivity of peat
decomposition to climate change through water-table feedback. Nature
Geoscience 1:763–766.
Keller, J. K., and S. D. Bridgham. 2007. Pathways of anaerobic carbon cycling across an
ombrotrophic-minerotrophic peatland gradient. Limnology & Oceanography
52:42–42.
74
Kivimäki, S. K., L. J. Sheppard, I. D. Leith, and J. Grace. 2013. Long-term enhanced
nitrogen deposition increases ecosystem respiration and carbon loss from a
Sphagnum bog in the Scottish Borders. Environmental and Experimental Botany
90:53–61.
Kuhry, P., and D. H. Vitt. 1996. Fossil Carbon/Nitrogen Ratios as a Measure of Peat
Decomposition. Ecology 77:271–275.
Lacourse, T., and M. A. Davies. 2015. A multi-proxy peat study of Holocene vegetation
history, bog development, and carbon accumulation on northern Vancouver
Island, Pacific coast of Canada. The Holocene:0959683615580201.
Laine, A., M. Sottocornola, G. Kiely, K. A. Byrne, D. Wilson, and E.-S. Tuittila. 2006.
Estimating net ecosystem exchange in a patterned ecosystem: Example from
blanket bog. Agricultural and Forest Meteorology 138:231–243.
Larmola, T., J. L. Bubier, C. Kobyljanec, N. Basiliko, S. Juutinen, E. Humphreys, M.
Preston, and T. R. Moore. 2013. Vegetation feedbacks of nutrient addition lead to
a weaker carbon sink in an ombrotrophic bog. Global Change Biology 19:3729–
3739.
Leifeld, L. G. J. 2011. Organic matter losses from temperate ombrotrophic peatlands: An
evaluation of the ash residue method. Plant and Soil 341:349–361.
Le, P. T., and C. E. Boyd. 2012. Comparison of Phenate and Salicylate Methods for
Determination of Total Ammonia Nitrogen in Freshwater and Saline Water.
Journal of the World Aquaculture Society 43.
75
Limpens, J., F. Berendse, C. Blodau, J. G. Canadell, C. Freeman, J. Holden, N. Roulet, H.
Rydin, and G. Schaepman-Strub. 2008. Peatlands and the carbon cycle: from local
processes to global implications – a synthesis. Biogeosciences 5:1475–1491.
Loisel, J., and M. Garneau. 2010. Late Holocene paleoecohydrology and carbon
accumulation estimates from two boreal peat bogs in eastern Canada: Potential
and limits of multi-proxy archives. Palaeogeography, Palaeoclimatology,
Palaeoecology 291:493–533.
Loisel, J., Z. Yu, D. W. Beilman, P. Camill, J. Alm, M. J. Amesbury, D. Anderson, S.
Andersson, C. Bochicchio, K. Barber, L. R. Belyea, J. Bunbury, F. M. Chambers,
D. J. Charman, F. D. Vleeschouwer, B. Fiałkiewicz-Kozieł, S. A. Finkelstein, M.
Gałka, M. Garneau, D. Hammarlund, W. Hinchcliffe, J. Holmquist, P. Hughes,
M. C. Jones, E. S. Klein, U. Kokfelt, A. Korhola, P. Kuhry, A. Lamarre, M.
Lamentowicz, D. Large, M. Lavoie, G. MacDonald, G. Magnan, M. Mäkilä, G.
Mallon, P. Mathijssen, D. Mauquoy, J. McCarroll, T. R. Moore, J. Nichols, B.
O’Reilly, P. Oksanen, M. Packalen, D. Peteet, P. J. Richard, S. Robinson, T.
Ronkainen, M. Rundgren, A. B. K. Sannel, C. Tarnocai, T. Thom, E.-S. Tuittila,
M. Turetsky, M. Väliranta, M. van der Linden, B. van Geel, S. van Bellen, D.
Vitt, Y. Zhao, and W. Zhou. 2014. A database and synthesis of northern peatland
soil properties and Holocene carbon and nitrogen accumulation. The Holocene
24:1028–1042.
Matušíková, I., J. Salaj, J. Moravčíková, L. Mlynárová, J.-P. Nap, and J. Libantová. 2005.
Tentacles of in vitro-grown round-leaf sundew ( Drosera rotundifoliaL.) show
76
induction of chitinase activity upon mimicking the presence of prey. Planta
222:1020–1027.
Maynard, D. G., Y. P. Kalra, and J. A. Crumbaugh. 2013. Methods in Biogeochemistry
of Wetlands. (R. D. DeLaune, K. R. Reddy, C. J. Richardson, and J. P.
Megonigal, Eds.). Soil Science Society of America.
Millett, J., G. W. Foot, and B. M. Svensson. 2015. Nitrogen deposition and prey nitrogen
uptake control the nutrition of the carnivorous plant Drosera rotundifolia. Science
of The Total Environment 512–513:631–636.
Millett, J., R. I. Jones, and S. Waldron. 2003. The contribution of insect prey to the total
nitrogen content of sundews (Drosera spp.) determined in situ by stable isotope
analysis. New Phytologist 158:527–534.
Moore, T. R., J. L. Bubier, S. E. Frolking, P. M. Lafleur, and N. T. Roulet. 2002. Plant
biomass and production and CO2 exchange in an ombrotrophic bog. Journal of
Ecology 90:25–36.
Nichols, J. E., D. M. Peteet, C. M. Moy, I. S. Castañeda, A. McGeachy, and M. Perez.
2014. Impacts of climate and vegetation change on carbon accumulation in a
south-central Alaskan peatland assessed with novel organic geochemical
techniques. The Holocene 24:1146–1155.
Novak, M., E. Brizova, M. Adamova, L. Erbanova, and S. H. Bottrell. 2008.
Accumulation of organic carbon over the past 150 years in five freshwater
peatlands in western and central Europe. Science of The Total Environment
390:425–436.
77
Olsson, M. O., and U. Falkengren-Grerup. 2000. Potential Nitrification as an Indicator of
Preferential Uptake of Ammonium or Nitrate by Plants in an Oak Woodland
Understorey. Annals of Botany 85:299–305.
Page, S. E., J. O. Rieley, and C. J. Banks. 2011. Global and regional importance of the
tropical peatland carbon pool. Global Change Biology 17:798–818.
Pfaff, J. D. 1993. Method 300.0 Determination of Inorganic Anions by Ion
Chromatography, Rev. 2.1. EPA.
Purdy, A. 2002. Unearthing the past: the archaeology of bog bodies in Glob, Atwood,
Hébert and Drabble. Textual Practice 16:443–458.
Richter, C., and J. Dainty. 1990. Ion behaviour in plant cell walls. IV. Selective cation
binding by Sphagnum russowii cell walls. Canadian Journal of Botany 68:773–
781.
Sheppard, L. J., I. D. Leith, S. R. Leeson, N. van Dijk, C. Field, and P. Levy. 2013. Fate
of N in a peatland, Whim bog: immobilisation in the vegetation and peat, leakage
into pore water and losses as N2O depend on the form of N. Biogeosciences
10:149–160.
Spearing, A. M. 1972. Cation-Exchange Capacity and Galacturonic Acid Content of
Several Species of Sphagnum in Sandy Ridge Bog, Central New York State. The
Bryologist 75:154–158.
sphagnumsem | A look into scanning electron microscopy….. (n.d.). .
https://sphagnumsem.wordpress.com/.
Svensson, B. M. 1995. Competition between Sphagnum fuscum and Drosera rotundifolia:
a caseof ecosystem engineering. Oikos 74:205–212.
78
Turetsky, M. R., S. E. Crow, R. J. Evans, D. H. Vitt, and R. K. Wieder. 2008. Trade-offs
in resource allocation among moss species control decomposition in boreal
peatlands. Journal of Ecology 96:1297–1305.
Vitt, D. H. 2013. Peatlands☆. Reference Module in Earth Systems and Environmental
Sciences. Elsevier.
Wang, L., and S. A. Macko. 2011. Constrained preferences in nitrogen uptake across
plant species and environments. Plant, Cell & Environment 34:525–534.
Wang, M., T. R. Moore, J. Talbot, and P. J. H. Richard. 2014. The cascade of C:N:P
stoichiometry in an ombrotrophic peatland: from plants to peat. Environmental
Research Letters 9:024003.
Ward, S. E., R. D. Bardgett, N. P. McNamara, and N. J. Ostle. 2009. Plant functional
group identity influences short-term peatland ecosystem carbon flux: evidence
from a plant removal experiment. Functional Ecology 23:454–462.
Wieder, R. K., J. B. Yavitt, M. Novak, R. K. Wieder, J. B. Yavitt, and M. Novak. 1994.
Global climate change and peatland carbon balance: an overview. Journal of
GEOsciences 39:330–334.
Yu, Z. C. 2012. Northern peatland carbon stocks and dynamics: a review. Biogeosciences
9:4071–4085.
79
Appendices
Appendix A. Site vegetation Data
Each “hit” represents a point where the vegetation was present. Each sample unit had a
total of 20 possible points over a 10 m length along the transect, and multiple vegetative
types could be “hit” at each point.
SU
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Total #
hits
# hits Sphagnum
10
11
11
13
3
7
7
1
6
2
5
5
5
13
6
8
11
13
8
1
3
2
7
8
11
# hits Vascular
5
10
8
6
11
15
14
13
10
11
12
3
2
6
8
9
8
4
3
2
2
11
10
9
9
# hits Graminoid
13
11
13
13
12
16
20
18
17
15
13
17
16
11
15
14
14
17
15
6
10
19
19
12
12
# hits Bare ground
2
2
2
1
4
1
0
2
2
2
5
2
4
2
2
3
0
0
4
14
10
0
1
1
3
177
201
358
69
80
Appendix B. Microlandform data
Sample plots were characterized as average “high,” “medium,” or “low” topography to
account for the presence of multiple microlandforms within each plot.
Hummock
SU
(high)
1 x
2
3
4
5
6 x
7 x
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Lawn
(medium)
x
x
x
x
x
x
Hollow
(low)
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
Average
topography
High
Low
Low
Medium
Medium
Medium
Medium
Low
Low
Low
Low
High
Medium
High
High
High
High
High
Medium
High
High
Medium
High
Low
Medium
Notes
No landforms present, plot
almost completely inundated
Hollow less than 5% of total
area; hummock dominates
81
