WHERE THE WATER MEETS THE SKY:
THE EFFECTS OF ATMOSPHERIC OZONE POLLUTION
ON AQUATIC ALGAL AND
BACTERIAL COMMUNITIES.
By
Julianne Heinlein

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Zoology – Doctor of Philosophy
2013

ABSTRACT
WHERE THE WATER MEETS THE SKY:
THE EFFECTS OF ATMOSPHERIC OZONE POLLUTION
ON AQUATIC ALGAL AND
BACTERIAL COMMUNITIES.
By
Julianne Heinlein
Ozone is a highly reactive oxidant and is the primary constituent of photochemical smog.
Since the industrial revolution, ozone levels in the troposphere have been rising and, during
warm seasons, regions all over the world reach concentrations that exceed the US EPA’s national
health-based standard of 75 ppb. In the most polluted areas, maximum levels have reached over
400 ppb. Numerous studies have shown acute and chronic impacts of ozone pollution on the
health of terrestrial plants and animals. Ozone is readily soluble in water and is often
gravitationally deposited onto surface waters. This research is the first to examine the potential
of tropospheric ozone as an aquatic pollutant by focusing on the effects of atmospheric ozone
levels on algal and heterotrophic bacterial assemblages.
Algae and bacteria were grown in three different ozone atmospheres (0, 80, and 250
ppb) and biomass and assemblage composition were measured. Individual experiments focused
on 1) the community-level responses of natural, multi-division periphytic algal assemblages to
these different ozone levels and the interactive effect of dissolved organic carbon (DOC), 2) the
effect of elevated atmospheric ozone levels on heterotrophic bacteria within the above
periphyton matrices and in the absence of algal interactions, and 3) the independent responses of
diatom and cyanophyte communities to ozone pollution in the absence of interdivisional
interactions.

Ozone had both negative direct effects and positive indirect, interactive effects on algal
biomass and assemblage composition. Within the natural periphyton assemblages ozone effects
varied with algal division and DOC concentration. In the low DOC water, ozone effects were
minimal. However in the high DOC water, the interactive effects of ozone and DOC were great.
Diatom biomass was maintained at lower levels in ozone treatments but cyanophytes colonies
increased by two orders of magnitude. This DOC and ozone interaction therefore led to a shift of
assemblage dominance from diatoms to cyanophytes. Heterotrophic bacterial density in these
periphyton films was closely correlated with algal biomass.
Responses changed when each group was treated independently. The cyanophyte-only
assemblages were directly affected by the oxidative stress created by ozone treated environments
and biomass was significantly lower in the ozone treatments. Diatoms, in independent cultures,
were unaffected by ozone treatment and heterotrophic bacterial growth was facilitated.
This study indicates the potential of atmospheric ozone to cause ecologically significant
changes to aquatic systems and highlights the need to consider direct and indirect effects of any
potential ecosystem stressor, species interactions, and effects in different environments.
Integrating the results of my experiments indicates that ozone has greater effects on algae and
bacteria in high than low DOC waters, and that ozone may cause a shift toward cyanophytes in
high DOC waters. I hypothesize this is due to ozone oxidation of DOC and release of organic
and inorganic resources that stimulate growth, and diatom mucilages mediated oxidative stress of
ozone on bacteria in the periphyton mat. Because ozone effects differed among algal divisions
and heterotrophic bacteria, with different roles, atmospheric ozone may change microbial food
webs and biochemical cycling within ecosystem, and these effects are likely more important in
high than low DOC waters.

ACKNOWLEDGEMENTS

I’d like to express the deepest appreciation to many people who have helped me along
this road. I owe my first debt of gratitude to the Biosphere Atmosphere Research and Training
(BART) program for selecting this novel research for one of their fellowships and providing the
bulk of my financial and technical support.
I would like to sincerely thank my committee, as a whole, for providing invaluable
feedback and encouragement. In particular, I’d like to thank my advisor, Dr. Jan Stevenson, for
his continued guidance and support, for helping me “see the forest for the trees” when I got
caught up in the details, and for always providing quick review of my work (when he finally got
it) to ensure that I kept moving forward. Dr. Steve Bertman significantly supported my early
research at the University of Michigan Biological Station through project design, endless
question answering and problem solving. He also, knowingly or unknowingly, indefinitely lent
me the expensive ozone generator/analyzer that I used for my final data generation. Dr. Steve
Hamilton gave a very detailed review of my work and significant suggestions for publication
improvements that I truly appreciate. In addition to his academic support, Dr. Rich Merritt gave
me sincere encouragement when things were rough and I doubted my abilities. He also provided
the occasional “kick in the pants.” I truly appreciate it all.
In addition, I extend many “thank yous” to friends and family. A huge “thank you” to
Mary Wolf Staton for making it through those two summers of constant design modifications,
trips to Mike’s hardware store, ozone sniffing, and overall frustration. We didn’t get any data
but we worked our butts off, learned a lot, and cemented an already strong friendship.

iv

My lab mate and friend, Stephanie Miller, took the time to wade through my very complicated
data and provided some of the most insightful discussions of my analyses and results which
pulled me out of many a data mire. Kalina Manoylov, Sara LeSage and Scott and Krista Rollins
were my cheering section throughout my time at MSU, helping me to keep smiling and
remember the important things in life. I continue to go to them when I need some sunshine.
Thanks also to Jason Zalack and Seth Hunt for keeping me grounded and laughing, with their
sarcasm and general ridiculousness.
And last but certainly not least my family, my parents Tom and JoAnn and sister Shari
Heinlein who never gave up on me and always gave me full support even though they were not
allowed to ever ask how the dissertation was coming. I owe the greatest debt to my partner, Will
Rideoutt, for simultaneously being my rock, caretaker, stress release (to his detriment), and
primary guardian to our household zoo.

v

TABLE OF CONTENTS

LIST OF TABLES ....................................................................................................................... viii
LIST OF FIGURES ....................................................................................................................... ix
Chapter 1
Introduction to the tropospheric ozone pollution problem and potential effects on aquatic
communities ....................................................................................................................................1
Tropospheric Ozone: Chemistry, Transport and Distribution ......................................1
Tropospheric Ozone: Effects on Terrestrial Organisms ...............................................11
Tropospheric Ozone: Hypothesized Effects on Aquatic Ecosystems ...........................15
Synthesis.............................................................................................................................21
Chapter 2
Impacts of elevated atmospheric ozone and the interactive effects of dissolved
organic carbon content on benthic algal community composition and biomass. ......................24
Abstract ...............................................................................................................................24
Introduction ........................................................................................................................24
Methods ...............................................................................................................................28
Overall Experimental Set-up......................................................................................28
Sample Collection.......................................................................................................31
Laboratory Analyses...................................................................................................31
Statistical Analysis .....................................................................................................33
Results .................................................................................................................................34
Dissolved Organic Carbon ........................................................................................34
Total Algal Biomass ...................................................................................................35
Division-level Biomass ...............................................................................................38
Nutrients .....................................................................................................................44
Discussion............................................................................................................................46
Chapter 3
Mediation of atmospheric ozone effects on benthic heterotrophic bacterial biomass by
dissolved organic carbon and algae .............................................................................................53
Abstract ...............................................................................................................................53
Introduction ........................................................................................................................53
Methods ...............................................................................................................................57
Bacteria only experiment............................................................................................57
Periphyton and Bacteria Experiments .......................................................................61
Statistical Analysis .....................................................................................................63
Results .................................................................................................................................64
Bacteria only experiment............................................................................................64
Periphyton and bacteria experiment ..........................................................................66
Discussion............................................................................................................................72
vi

Acknowledgements ............................................................................................................76
Chapter 4
The effects of atmospheric ozone concentration on benthic cyanophyte and diatom
communities. .................................................................................................................................77
Abstract ...............................................................................................................................77
Introduction ........................................................................................................................77
Methods ...............................................................................................................................80
Experimental set-up common to both the cyanophyte and diatom experiments ........80
Cyanophyte experiment ..............................................................................................81
Diatom experiment .....................................................................................................82
Sample Collection.......................................................................................................82
Analysis of periphyton samples ..................................................................................83
Statistical Analysis .....................................................................................................84
Results .................................................................................................................................84
Cyanophyte experiment ..............................................................................................84
Diatom experiment .....................................................................................................87
Discussion............................................................................................................................90
Chapter 5
Summary........................................................................................................................................94
LITERATURE CITED ................................................................................................................100

vii

LIST OF TABLES

Table 2.1: ANOVA table for log-transformed values of total biomass (as measured by biovolume
and chlorophyll a) as a function of DOC, ozone, and week ..............................................35
Table 2.2: ANOVA table for log-transformed values of division-level biovolume as a function of
DOC, ozone, and week ......................................................................................................39
Table 4.1: Average nutrient concentrations (mg/L) for the cyanophyte experiment, +/- standard
deviation. Letters denote statistically significant differences ...........................................84
Table 4.2: Average nutrient concentrations (mg/L) in the diatom experiment +/- standard
deviation .............................................................................................................................88
Table 4.3: Relative abundance of diatom taxa by ozone treatment
(percentage of assemblage) ................................................................................................89

viii

LIST OF FIGURES

Figure 1.1: O3 production initialized by the photoreduction of NO2 ..................................................................... 1
Figure 1.2: Generalized reaction scheme illustrating the involvement of VOCs in ozone
production ............................................................................................................................5
Figure 1.3: Hypothesized direct (solid line) and indirect (dotted line) effects of atmospheric
ozone on aquatic algae .......................................................................................................18
Figure 2.1: Experimental set-up. Rectangles represent individual chambers and circles represent
microcosms within each chamber ......................................................................................29
Figure 2.2: Measured DOC concentration by ozone and week, for each DOC treatment .............34
Figure 2.3: Measures of total algal biomass by ozone and week, for each DOC treatment ..........37
Figure 2.4: Measures of division-level biomass by ozone and week, for each DOC
treatment ............................................................................................................................42
Figure 2.5: TDN, SRP, and silica concentration (mg/L) by ozone and week, for each DOC
treatment ............................................................................................................................45
Figure 3.1: Experimental set-up. Rectangles represent individual chambers and circles represent
microcosms within each chamber ......................................................................................59
Figure 3.2: Bacterial biomass as represented by cells/cm2 and DOC for each treatment in the
bacteria only experiment ....................................................................................................65
Figure 3.3: Measured DOC concentration by ozone and week, for each DOC treatment .............67
Figure 3.4: Measures of total algal biovolume, bacterial cell density, and ratio of bacterial
density to algal density by ozone and week, for each DOC water treatment. ...................70
Figure 3.5: TDN and SRP concentration (mg/L) by ozone and week, for each DOC
treatment ............................................................................................................................72
Figure 4.1: Total biomass by ozone treatment in the cyanophyte experiment ..............................85
Figure 4.2: Cyanophyte genus biovolume as a function of ozone .................................................86
Figure 4.3: Relative abundance of cyanophyte genera by biovolume for each ozone treatment.
Tolypothrix was less than 1% of all assemblages ..............................................................87

ix

Figure 4.4: Total biomass by ozone treatment in the diatom experiment ......................................88

x

Chapter 1: Introduction to the tropospheric ozone pollution problem and potential effects on
aquatic communities

Tropospheric Ozone: Chemistry, Transport and Distribution
Ozone (O3), the most important constituent of photochemical smog, is a highly reactive
oxidant linked to damage of animal and plant tissue as well as numerous inorganic materials.
Because of its strong potential to harm living organisms and cost property and crop owners, this
trace atmospheric gas has been regulated by the EPA. Being approximately 3000 times stronger
a greenhouse gas than CO2, tropospheric ozone also significantly affects the radiative balance of
the atmosphere. The Intergovernmental Panel on Climate Change now considers ozone to be
the third most important greenhouse gas after carbon dioxide and methane. It occurs naturally
in the troposphere at background levels of 10-20 ppb (Finlayson-Pitts and Pitts 1999) but has
increased greatly since the industrial revolution.
Some of this natural ozone is transported from the stratosphere where it is produced,
secondarily, from the photodissociation of molecular oxygen. It is also produced directly in the
troposphere through the photoreduction of nitrogen dioxide (NO2) and the subsequent
combination of a singlet O atom with molecular oxygen (Fig. 1.1) (Committee on Tropospheric
Ozone 1991).
•

NO2 +hν =NO + O

•

O + O2 = O3

•

O3 + NO = NO2 +O2

Fig. 1.1. O3 production initialized by the photoreduction of NO2
1

However, this ozone is quickly consumed by reacting with nitric oxide (NO) from the initial
photochemical reaction, so it does not accumulate to a great extent. Because of the limited
intrusion of stratospheric ozone, the quenching of newly produced ozone by NO, and the low
levels of naturally occurring NO2, the pre-industrial troposphere is thought to have contained
only 10% of all ozone in the atmosphere and to have varied very little in concentration (Chatfield
and Harrison 1977). Since ozone is a trace gas, small changes to ozone precursor molecules
may have profound effects on tropospheric concentrations. Since the industrial revolution, nearsurface ozone levels have been measured at much greater concentrations than can be accounted
for by the above sources. As an example, average pre-industrial tropospheric ozone
concentrations in Paris were about 10 ppb, 1988-89 averages in the least polluted parts of Europe
were between 20 and 45 ppb (Janach 1989; Volz and Kley 1988). Ambient ozone levels have
steadily increased and, today, maxima have been measured above 350 ppb in heavily polluted
areas (Kley et al. 1999).

In the U.S., the national ambient air quality standard (NAAQS) for

ozone, developed to protect human and ecosystem health, is 75 ppb for an eight hour average.
It was determined in the 1950s that this ozone accumulation in the near-surface
atmosphere requires volatile organic carbons (VOCs) in addition to NO2 and NO (the
combination of which is represented by “NOx”) (Haagensmit et al. 1953). The presence of
VOCs causes enhanced conversion from NO to NO2 and therefore, the accumulation of ozone to
concentrations above those found in the background troposphere, through the photolysis of this
additional NO2 (Committee on Tropospheric Ozone 1991). The chemical reactions involved are
numerous and complex with over a hundred different chemical species of VOCs, each with
varying reactivity and lifespan (Finlayson-Pitts and Pitts 1999). A generalized reaction scheme
2

is shown in Fig. 1.2, taken from Kley et al. (1999), in which the oxidation of an alkane (RH) is
used to illustrate the involvement of VOCs in ozone production and accumulation.
This relationship between NOx, VOCs, and ozone is non-linear and the control of ozone
production often depends on the specific VOC/NOx ratio. In general, for VOC/NOx ratios of
1

greater than 8:1 , lowering NOx concentrations results in lower peak concentrations of ozone
(Committee on Tropospheric Ozone 1991). This is called a “NOx–limited” system and is
characteristic of less polluted areas downwind of city centers where biogenic or industrial VOC
production outpaces NOx emissions from power plants and automobile exhaust. In such
situations, ozone concentrations are not sensitive to reduction in VOCs if it is not coupled with a
reduction in NOx. In some highly polluted urban areas with large NOx emissions, we see
VOC/NOx ratios less than 8:1. In these “VOC-limited” atmospheres, decreasing the VOC at
constant NOx or simultaneously lowering both VOC and NOx, proportionately, will result in
lower peak ozone concentrations. However, due to the complex photochemistry of this ozone
production, if only NOx is lowered, peak ozone concentrations will actually increase until the
critical ratio is reached. The sources, concentration, and ratio of VOC and NOx vary
tremendously geographically and temporally, but the underlying photochemistry remains the
same worldwide and must be considered when trying to control ozone accumulation.

1

The critical ratio may vary with the reactivity of the VOC mixture. Also, Kley et al.,1999 use a

VOC/NOx ratio of 4:1 as the ridge line between NOx-limited and VOC-limited regimes.
3

The sources of NOx emissions into the troposphere are both natural and anthropogenic.
Lightning, naturally-occurring vegetation fires and microbial metabolism in soils contribute the
majority of natural emissions. The contribution of lightning varies seasonally and may be more
important in the upper troposphere, particularly during summer months. However, globally,
these natural emissions contribute less than 20% of the total NOx emissions and in industrialized
nations may contribute a much smaller fraction (Zhang et al. 2003). The greater emissions are
all anthropogenic in origin and include fossil fuel combustion, biomass burning, and increased
microbial action due to fertilizer application (Kley et al. 1999). Estimates for total global
emission of NOx range from about 30-45 Tg/yr

(2)

, but most agree that over half of the total

tropospheric inventory of NOx is attributable to fossil fuel combustion by coal-fired electricity
production and gasoline/diesel use (Fowler et al. 1998) . The importance of the two sources
(and others) may vary regionally (http://www.epa.gov/air/emissions/), but over the Northern
Hemisphere, especially over the mid-latitudes, the majority of these emissions may be attributed
to gasoline and diesel engines (Kley et al. 1999).

2

Tg=Teragram=1,000,000 metric tons
4

RH + OH → H2O + R

(R1)

• reactive VOCs, represented as RH, react with hydroxyl radicals (OH) to form
organic radicals (R)
M
R + O2 → RO2
(R2)
• Organic radicals combine with molecular oxygen to form peroxy radicals
(RO2), a process that usually requires an inert third body, M (e.g., N2 or O2)
RO2 +NO → NO2 + RO
(R3)
• Peroxy radicals react with nitric oxide (NO) to form nitrogen dioxide (NO2)
NO2 + hν → NO + O(3P)
(R4)
• Nitrogen dioxide is photodissociated by solar radiation to release ground state
oxygen atoms, O(3P), and reform nitric oxide
M
O + O2 → O3 +M
(R5)
• Finally, oxygen atoms combine with molecular oxygen, in the presence of a
third body, to form ozone
O3 + hν → O2 + O(1D)
(R6)
• Ozone is photodissociated by near-ultraviolet solar radiation to form an excited
ultraviolet
oxygen atom, O(1D)
O(1D) + H2O → 2OH
(R7)
• which, in turn, can react with water vapor (H2O) to form two OH radicals.
The resulting OH radicals drive the chain process. Furthermore, reactions
initiated by the RO radicals formed in Reaction 1.3 can, in the presence of NO,
lead to further production of OH.

Fig. 1.2. Generalized reaction scheme illustrating the involvement of VOCs in ozone production
and accumulation.

The most recent U.S. EPA inventory (
(http://www.epa.gov/air/emissions/) lists gasoline
)
and diesel engines as supplying approximately 60% of the national emissions of NOx, 35% of
that being due to on-road vehicles. Electrical powerplants contributed approximately 20%.
road
contributed

5

Vehicle traffic has been statistically tied to ozone pollution in European and North American
cities (Committee on Tropospheric Ozone 1991; Granier and Brasseur 2003; Ibarra-Berastegi
and Madariaga 2003). With automobile and energy use on the rise in the U.S. as well as in
emerging economies, such as China and India, we expect to see greater overall emissions. In
regions where fossil fuel use is much less, such as the southern hemisphere, the other
anthropogenic sources of NOx may cause local pollution events. Burning biomass to clear
croplands and forests is a common practice in developing countries and has been shown to
contribute significantly to ozone production (Fiore et al. 2002; Kondo et al. 2003; Satsangi et al.
2004), especially over southern Africa (Thompson et al. 2002), the South Atlantic (Swap et al.
2003), and South America (Sanhueza et al. 2000). Even soil emissions of NOx, from the overfertilization of croplands, can contribute to local ozone highs in areas with a ready VOC source
(Simpson 1993). Due to newly introduced regulations (some voluntary) and demographic shifts,
there have been regional reductions in NOx emission worldwide but overall, they are on the rise
in the Northern Hemisphere, increasing over the US and especially over China and India where
energy and automobile use are increasing dramatically from year to year (Butler et al. 2011;
Fiore et al. 2002; Meng et al. 2010; Sheel et al. 2010; Vingarzan 2004). In fact, Xing et al.
(2011) estimated, based on 2005 emission measurements, that total NOx emissions in China will
increase 50% by 2020 if the current regulation and enforcement remains. Also, long range
transport of NOx emissions from other areas may keep concentrations high even where regional
emission reductions have occurred.
Volatile organic compounds are generated both anthropogenically and biogenically. In
urban and suburban areas, in the presence of NOx pollution, anthropogenic sources such as liquid
6

fuel combustion and evaporation, industrial use of solvents, and many household organic
chemicals, such as those in some paints and cleaners, are the fuel for ozone accumulation. As
with nitrogen oxides, the dominant VOC sources vary regionally
(http://www.epa.gov/air/emissions/). According to the EPA’s year 2008 measurements, the
largest U.S. source of anthropogenic VOCs was fossil fuel combustion by on-road vehicles and
non-road equipment. The use of solvents followed closely.
On average, biogenic emission of volatile organic carbons (BVOCs) greatly exceeds
anthropogenic sources. In 2005, the EPA estimated that approximately 74% of all VOCs
released in the U.S. were biogenic (EPA 2008). Most BVOCs are produced by plants, but
bacterial releases can be significant in certain areas. BVOC emission from forests alone is a
larger source of VOC than all the anthropogenic sources in the U.S. combined. Forest emissions
of BVOCs are important sources of ozone accumulating in rural areas. These compounds, in the
presence of NOx pollution downwind from urban or industrial areas, stimulate great ozone
accumulation. Although there are many different types of BVOCs with varying chemistry, most
are more highly reactive than anthropogenic VOCs. This fact, coupled with their high
concentrations, means that BVOCs are often the primary determinant of photochemistry in rural
and some suburban locations where vegetation is greatest, although others (Chameides et al.
1988; Chameides et al. 1997) suggest that BVOCs should be considered a potentially significant
contributor to the photochemistry above urban areas as well. This large source of biogenic
hydrocarbons sometimes leads to higher ozone levels over forested land than over large urban
centers (Bell and Ellis 2004). Even considering predictions of fewer trees due to future
deforestation, the levels of biogenic VOCs are hypothesized to rise due to increased production
by individual trees in response to increases in global CO2 and warming temperatures.

7

Because of this uneven emission of ozone precursors, their atmospheric transport, and
local weather conditions, the spatial and temporal variability of ozone concentration is great.
The levels of ozone recorded in the troposphere currently range from 10 to close to 400 ppb
(Eder et al., 1993; Bobbink, 1998; Fast and Heilman, 2003; Vingarzan, 2004). Spatial variability
is a function of geographic location, elevation, and meteorology, but more importantly of the
extent of anthropogenic emissions of nitrogen oxides as they are often the “limiting factor” in
ozone accumulation in rural/suburban areas. Ozone levels are often the highest above urban
centers that have both large emissions of nitrogen oxides and VOCs. Mexico City, for example,
has reached concentrations greater than 400 ppb (Zambrano and Nash 2000). However, ozone
concentrations in the atmosphere above the surrounding areas may often be higher than over the
city itself. This may be due to greater emissions of biogenic VOCs or in the following case, a
high production of industrial VOCs. Daum et al. (2003) measured ozone in and around Houston,
Texas. In this case Houston was measuring 40-90 ppb at the same time that air above a shipping
channel outside of the city was reaching 120-200 ppb. Nitrogen oxide concentrations were
equivalent but ozone production efficiency was much greater in this far less populated region
because of a ready source of industrial hydrocarbons that kept the VOC/NOx ratio high.
Rural and remote areas are subject to the same type of phenomenon (Brankov et al. 2003;
Chameides et al. 1997; Debaje et al. 2003; Kunhikrishnan e al. 2006; Metcalfe et al. 2002;
Rosenthal et al. 2003; Swap et al. 2003). Large emissions of biogenic hydrocarbons from
vegetation, especially from trees, create NOx sensitive zones in these less-populated areas so that
the transport of NOx to these areas from urban and suburban centers will increase ozone
concentrations (Bakwin et al. 1994; Utiyama et al. 2004). From 1993-1995, the air at 30-50% of

8

measuring stations in the rural eastern U.S. exceeded the national 8-hour standard of 80 ppb and
2-12% exceeded 120 ppb (Fiore et al. 2002). These trends are also seen in rural China (Cheung
and Wang 2001). Of primary ecological concern in the U.S. is ozone pollution of national parks.
Ozone levels in these protected areas, particularly in the southwest, regularly exceed the 8-hour
EPA standard for ground-level ozone (75 ppb) and in summer ozone concentrations often exceed
those over urban centers. From April through August of 2011, ozone levels exceeded the
NAAQS (75 ppb) 222 times in national parks with ozone monitors (NPS 2011) these maxima are
often higher than those in nearby metropolitan areas. Transport from rural areas may influence
even more remote areas and threaten some fairly pristine habitats that are largely shielded from
non-atmospheric pollutants. For example, ozone concentrations reaching 40 ppb were measured
in a remote Venezuelan cloud forest, where pre-industrial levels were 4-8 ppb (Sanhueza et al.
2000).
Local and regional transport is accompanied by long-range, intercontinental transport,
which occurs in the upper, free troposphere and contributes to global background ozone (Liu et
al. 2003; Thompson et al. 2002). For example ozone over the North Atlantic averages around 25
ppb in the summer and this is attributed to transport from North America (Parrish et al. 1993).
The footprint of North American pollution extends to Europe as well. The air over North
America is influenced by pollution from China and India (some estimates of influence range
from 3-10 ppb) (Liu et al. 2003; Parrish et al. 1993), while African biomass burning may
influence ozone production over the Southern Atlantic Ocean and Austral-Asia land masses
(Kondo et al. 2003). European emissions have been shown to greatly increase the background
pollution over Asia (Parrish et al., 1993). Because of this transport, it has been estimated that
global background ozone levels have doubled since the turn of the century (Vingarzan, 2004).

9

Records of ozone measurements from the late 1800s over North America and Europe indicate
th

19 century background levels of 5-10 ppb (Vingarzan, 2004). Current estimates indicate
background levels of 10-45 ppb (Fiore et al., 2002).
Ozone concentrations also vary temporally on seasonal and diurnal scales. Seasonally,
ozone concentrations tend to be highest in the summer when sunlight is most intense and
biogenic VOCs are being produced in large quantities. During the day in urban areas, NOx and
VOCs tend to rise with the increase in industrial and automobile activity and, as a consequence,
ozone levels tend to be highest shortly after noon. Areas downwind of urban centers may peak
later in the afternoon or early evening. After sundown, photooxidation ceases and existing
atmospheric ozone is scavenged by other chemicals or is deposited on land or surface water.
Therefore, nighttime ozone levels near the ground can fall to near zero (Kley, et al., 1999)
Local production, regional and longer-distance transport, and high background levels lead
to a global ozone pollution problem. Models have estimated a 100-120% increase in ozone in
the northern mid-latitudes over the past 200 years and an increase of 80-100% in the tropics
(Committee on Tropospheric Ozone 1991). All but one scenario proposed by the
Intergovernmental Panel on Climate Change (IPCC) projects increases in tropospheric ozone
during the 21st century with projected background concentrations of 35-48 ppb by 2040, 38-71
ppb by 2060, and 42-84 by 2100 (Vingarzan, 2004). This means that background concentrations
may soon exceed internationally accepted maxima established to protect animal and plant health.
The increases are expected to be greatest in the tropics, specifically South Asia, and also in
Central America, South America, and southern Africa. However, the northern hemisphere will
also see a rise. Since dry deposition of this atmospheric ozone (i.e., deposition not associated
with precipitation) is the largest sink (Fowler et al., 1998), the potential for ozone damage to
10

ecosystems is very real and appears to pose an even bigger threat in the future. Much of this
ozone is and will be over surface waters. Because ozone is highly soluble in water some of this
atmospheric component may be expected, through dry deposition, to dissolve into and change
the oxidative state of the water.

Tropospheric Ozone: Effects on Terrestrial Organisms
Ozone, as a highly reactive oxidant, has the potential to react with unsaturated organic
chemical bonds, including proteins, lipids, peptidoglycans and enzymes, causing cellular damage
or death and inducing defensive responses that will change the chemical composition of cells
(Kley et al. 1999; Meehan et al. 2010). This cellular damage results from the oxidative stress
caused directly from the ozone molecules but also from other reactive oxygen species (ROS)
such as hydrogen peroxide, hydroperoxide, superoxide, hydroxyl, and singlet O. These oxygen
species result from ozone speciation, cellular production induced by the initial stress, or
oxidation of membranes (Oksanen 2003; Schraudner et al. 1997).
Inhalation is the primary mechanism of internal ozone exposure in humans and other
mammals so damage is predominantly to lung tissue. Acute respiratory damage and worsening
of asthma may occur at levels as low as 60 ppb and higher concentrations or chronic exposure
may lead to persistent effects on lung tissue and function (Lippmann 1991; Tilton 1989). Studies
of reptiles and amphibians have indicated that high O3 exposure may affect core body
temperature by changing thermal preferences and, potentially, water balance (Dohm et al. 2001;
Mautz and Dohm 2004). Atmospheric ozone may interfere directly with the olfactory senses of
insects or indirectly through ozone interaction with the signal chemicals. Several studies have

11

shown that the searching efficiency of parasitoid wasps and the proportion of hosts parasitized
decrease as a consequence of elevated ozone levels (Gate et al. 1995; Mondor et al. 2004).
In addition to the few studies showing direct effects, most studies of the effects of
ambient ozone levels on wildlife show indirect effects on herbivores mediated by changes in the
chemical composition or availability of their plant food sources (Augustaitis et al. 2007;
Ditchkoff et al. 2009). The response varies depending on species. The growth rate and behavior
of insects, in particular, have been shown to change in response to ozone-mediated increases in
leaf secondary metabolites or other physiological stress responses (Lindroth 2010; Valkama et al.
2007). The growth of many insect species, especially chewers that include potential crop pests,
has been positively affected by ozone treatment of their host plants, showing shortened larval
development time and greater pupal mass (Jones and Paine 2006). Feeding preference was also
affected by ozone treatment but was dependent on herbivore and tree species, however, many
insects prefer ozone treated leaves to those raised under ambient conditions (Agrell et al. 2005;
Holton et al. 2003). These positive effects have been attributed to the greater concentration of
secondary metabolites or the reduction of phenolic glycosides which have been shown to be
detrimental to larval performance (Holton et al. 2003; Valkama et al. 2007).
Photosynthetic cells are naturally high in reactive oxygen species (ROS) and therefore are
especially susceptible to oxidative overload; therefore, ozone-mediated changes to plants at the
individual and community level are common. The physiological changes associated with ozone
stress start with the loss of membrane integrity as these chemical species attack unsaturated
lipids and proteins leading to internal production of assorted ROS (Skarby et al. 2004). As ROS
increase, intracellular damage occurs (Kollner and Krause 2003; Oksanen et al. 2004; Skarby et
al. 2004; Yamaji et al. 2003). Organelles, especially chloroplasts, suffer injury (Manning and

12

Godzik 2004; Prozherina et al. 2003). Oxidative damage can be mitigated through the
production of a multitude of enzymatic and non-enzymatic antioxidants. This appears to be a
primary response of photosynthetic cells and serves to quench much of the increased oxidative
potential (Schraudner et al., 1997). The type and quantity of antioxidants produced varies
depending upon species and may be the basis for different tolerances observed. The metabolic
pathways associated with this physiological stress response leads to changes in leaf
phytochemistry, specifically an increase in secondary metabolites, such as phenolic acids and
flavonoids. At the same time, the production of primary metabolites, such as carbohydrates, is
often unaffected or only slightly negatively affected (Holton et al. 2003; Lindroth 2010; Valkama
et al. 2007). The response varies depending on species and even genotype within a species
((Furlan et al. 2004; Oksanen et al. 2004)). In terms of the above changes in leaf chemistry,
angiosperms have been shown to be much more susceptible to ozone than gymnosperms (Holton
et al. 2003).
Primary producers of all types have been shown to be sensitive to elevated levels of
atmospheric ozone (Bermejo et al. 2003; Booker et al. 2009; Braun et al. 2004; Fiscus et al.
2005; Fuhrer 2009; Heath 1994; Morgan et al. 2006). This ozone stress causes both acute and
chronic effects in plants. Acute responses resulting in foliar injury, pigment loss and premature
senescence occur when a plant’s antioxidant response is quickly overwhelmed by a large
concentration of ozone (Krupa et al. 1998). For example, necrotic lesions on tobacco plants have
developed within 15-72 hours after a single acute ozone dose of 150 ppb (Schraudner et al.
1997). Chronic damage is a cumulative effect of multiple responses to low and high ozone
intake by a plant through its life cycle. Responses to chronic exposure vary depending on age,
species, and developmental stage but generally include a reduction of photosynthetic rate, growth

13

rate and decreased root biomass due to efforts to maintain above-ground growth (Chappelka and
Samuelson 1998; Schraudner et al. 1997; Valkama et al. 2007).
Due to these effects on plants, tropospheric ozone has been identified as the most
damaging atmospheric pollutant to crops and forests (Heagle 1989). In fact, ozone is the only
major air pollutant that has been shown to induce plant damage at ambient concentrations
(Schraudner et al., 1997). Damage to potato, soybean, wheat, and tobacco crops due to ozone
has been reported (Eder et al. 1993; Ishii et al. 2004; Kollner and Krause 2003; Madden and
Hogsett 2001; Schraudner et al. 1997). Avnery et al. (2010) estimated “that year 2000 ozoneinduced global yield reductions ranged, depending on the metric used, from 8.5-14% for
soybean, 3.9-15% for wheat, and 2.2-5.5% for maize. Global crop production losses totaled 79121 million metric tons, worth $11-18 billion annually.” They and others (Bender and Weigel
2011) predict even greater losses in the future.
In addition to reductions in crop yield, damage to natural communities has been
documented, particularly to forest trees (Broadmeadow 1998; Chappelka and Samuelson 1998;
Eder et al. 1993; Oksanen et al. 2004) and grassland species (Franzaring et al. 2000), but also to
wetland plants such as Sphagnum mosses (Niemi et al. 2002). It is estimated that approximately
25% of the global forests are currently at risk during the growing season from ozone
concentrations above 60 ppb and estimates indicate that this will rise to 50% by 2100 (Oksanen,
2003). These damage responses are species dependent (Bobbink 1998; Franzaring et al. 2000;
Novak et al. 2003) and this differential species response may lead to compositional shifts in
natural communities, even if community biomass is not affected.
The study of the response of terrestrial algal cells to atmospheric ozone has been limited
to lichen cells and symbiotic cyanophytes. Nitrogen-fixing cyanophyte symbionts were shown

14

to have inhibited growth, nitrogen fixation, and heterocyst formation when exposed to ozone
concentrations as low as 30 ppb for one week (Hur and Wellburn 1994). Lichens have long been
used as indicators of air pollution and studies have indicated that a response to ozone stress has
led to precipitous declines in lichen species richness and cover in areas outside of Mexico City
and in coniferous forests in California (Canas and Pignata 2000; Garcia et al. 2000; Scheidegger
and Schroeter 1995; Sigal and Johnston 1986; Zambrano and Nash 2000; Zambrano et al. 1999).
This decline has been attributed to a loss of chlorophyll a by the algal symbiont and the
subsequent reduction of photosynthetic rate, a response similar to that of plants. Also, a primary
response of lichen algae appears to be an increase in the antioxidant, superoxide dismutase
(Calatayud et al. 2000).

Tropospheric Ozone: Hypothesized Effects on Aquatic Ecosystems
There has been almost no research on the effects of atmospheric ozone on aquatic
communities, but we may hypothesize similar effects. Changes to olfactory chemicals in the
water, behavioral changes as a response to oxidative stress (i.e. drift), and changing algal
community composition or chemistry may influence the abundance and types of invertebrates in
the affected area. Studies of wetland plants have indicated a greater sensitivity to ozone than
plants not associated with water-logged soils (Franzaring et al. 2000; Power and Ashmore 2002;
Williamson et al. 2010). This is attributed to the higher rates of growth, stomatal conductance,
and leaf area associated with wetland plants (Power and Ashmore 2002). Given the high
surface-to-volume ratio and thin cuticle of most submerged angiosperms and mosses we may
hypothesize a greater potential for response as well. Ozone effects on aquatic algae have been
studied only in regards to wastewater treatment. At sufficiently high ozone concentrations, algal

15

cells are “oxidized to death” and even their byproducts may be destroyed (e.g. cyanotoxins) (Yun
et al. 1997). There are differences in the way that certain algae respond to ozone, and different
concentrations and exposures times are necessary to eliminate different species (Hoeger et al.
2002; Lai et al. 2002; Paralkar and Edzwald 1996; Widrig et al. 1996). Although photosynthetic
rates of algae after ozonation were severely retarded, even by short ozone contact (Yun et al.,
1997), studies of the longer-term sub-lethal effects of ozone are few.
Studies of heavy metal and hydrogen peroxide (H2O2) induced oxidative stress may
serve as models of potential ozone damage. The oxidative stress created by these factors works
much the same way that ozone-induced stress does in terrestrial plants; a direct oxidation of
lipids and proteins and increased cellular production of reactive oxygen species which may lead
to a saturation of antioxidant capacity (Pinto et al. 2003). Algal cells show an immediate
increase in both enzymatic and non-enzymatic antioxidants, superoxide dismutase being primary
in most species (Leitao et al. 2003; Pinto et al. 2003). Where reactive oxygen species production
overwhelms the antioxidants, membrane damage (especially in chloroplasts), pigment loss, and
even cell death have occurred (Mallick and Mohn 2000). Algal species differ in their response to
H2O2-induced stress (Barroin and Feuillade 1986; Kay et al. 1984). Drabkova et al. (2007)
found that photosynthesis was inhibited by ozone to a greater extent in five cyanophyte species
than three green algal species and one diatom species. For example, Microcystis aeruginosa, a
cyanobacterium, was negatively affected at ten times lower H2O2 concentrations than that of the
green alga, Pseudokirchneriella subcapitata. Using these responses as a model, we may
hypothesize that the oxidative damage and defense response caused by ozone exposure may lead
to a reduction of algal photosynthetic rate and change of cellular chemistry. These may in turn

16

lead to ecologically relevant changes in biomass, community composition, as well as algal
nutritional value and palatability to herbivores.
Algal communities also may be indirectly affected by the response of other organisms
and/or dissolved substances to the increased oxidant levels (Fig. 1.3). Ozone stress may change
inter-species interactions within the biofilm by differentially affecting various algal groups or
heterotrophic bacteria, changing community dynamics and potentially changing algal community
composition or total biomass. Bacterial, viral, and fungal pathogens are common to all types of
algae and may have significant impacts on algal community composition, development, and
senescence (Hewson et al. 2001; Peterson et al. 1993; Pollard and Ducklow 2011; Proctor and
Fuhrman 1991; Vandonk and Ringelberg 1983; Vanetten et al. 1991). Viruses have been shown
to be vulnerable to light (Suttle and Chan 1993) and all three pathogens are inactivated by ozone
during wastewater treatment and also by gaseous ozone treatment of surfaces (Goncalves and
Gagnon 2011; Hudson et al. 2009; Lenes et al. 2010; Mahfoudh et al. 2010; Tseng and Li 2008;
Zhang et al. 2011).
Ozone and its dissolution species react with most organic molecules that they contact
and, in natural waters, an important component is dissolved organic carbon (DOC). Up to 20%
of terrestrial primary production can be exported as dissolved organic carbon, making it very
important to aquatic communities (Lennon 2004; Lennon and Pfaff 2005; Wetzel 2001a). DOC
molecules such as amino acids, carbohydrates, simple organic acids, and lipids are very labile
and assimilated readily by bacteria, and so have very high turnover rates (Wetzel, 2001a). Their
instantaneous concentration in the water column is very low. The remaining DOC, composing as
much as 80% of that dissolved in inland waters, is humic in nature; made up of complex organic
acids (humic and fulvic), that are mostly terrestrial in origin (Rosenstock et al. 2005). Humic

17

and fulvic acids are also the primary contributors to the tea like color in high DOC waters. Due
tea-like
wa
to their complex structure
Ozone and Other Reactive Oxygen

Competitors
(other algal
groups and
bacteria)

Pathogens
(viruses/bacteria)

Dissolved Organic Carbon
Release of N, P,
and labile
carbon and
trace nutrients

Water Color
Loss

Algae
Fig. 1.3. Hypothesized direct (solid line) and indirect (dotted line) effects of atmospheric ozone
on aquatic algae.

and aromatic carbon rings, humics are more recalcitrant than other DOC and thus are only very
slowly assimilated by bacteria. They tend to have long residence times in the water column
(Wetzel 2001a; Yu et al. 2002) and will be available for interactions with dissolved oxidants. In
with
fact, they are often the primary sink for ozone in water treatment (Cho et al. 2003). Because of
its primary and almost immediate reaction with dissolved oxidants, DOC may serve to protect
algal cells from direct oxidative damage by quenching the oxidative potential.
18

The oxidative breakdown of humic substances leads to the formation of more labile
organic compounds such as aldehydes, polysaccharides, and fatty acids that are more
bioavailable to bacterial communities (Fahmi et al. 2003; Freese et al. 1999; Goldstone et al.
2002; Lennon 2004; Levine et al. 2000; Romani et al. 2004; Swietlik and Sikorska 2004;
Westerhoff et al. 1999). Oxidation of humic substances can also lead to a release of iron and
other trace metals, ammonium, amino acids, and phosphates as they often form complexes with
these molecules (Ellis et al. 1999; Engelhaupt et al. 2003; Stewart and Wetzel 1982; Wetzel
2001d). Iron oxide forms a particularly tight bond with fulvic acids and there are data to suggest
that this may cause iron limitation in humic waters (Imai et al., 1999; Guildford et al. 1987;
Kerndorff and Schnizer 1980; Jackson & Hecky 1980; Giesy 1976).
Liberation of labile DOC and inorganic nutrients through the action of ozone on humic
substances should serve to stimulate heterotrophic bacterial growth if direct oxidative stress on
cells is not too great. There is ample evidence of an increase in bacterial growth rate after
breakdown of recalcitrant molecules in certain systems, via uptake of labile DOC (Anesio et al.
2004; Engelhaupt et al. 2003; Lennon 2004; Middelboe and Lundsgaard 2003; Romani et al.
2004). Natural ultraviolet light has been observed to partially oxidize humic substances
generating large amounts of simple fatty acids that will serve as ready metabolic substrates by
bacteria (Volk et al. 1997). Many studies have found that DOC exposed to natural UV radiation
leads to an immediate and sustained increase of bacterial growth (Lindell et al. 1995; Moran and
Zepp 1997; Rosenstock et al. 2005; Stewart and Wetzel 1981; Wetzel et al. 1995). Ozone
oxidation may do the same, changing algal/bacterial interactions.
The breakdown of aromatic bonds and depolymerization of molecules also leads to a
strong decrease in color and UV absorbance (Arvola 1984; Freese et al. 1999; Graham 1999).

19

Indeed, my preliminary research, in which algae was grown continuously in one of three ozone
atmospheres (0, 80, and 300 ppb) led to an unanticipated difference in water color between
ozone treatments, that was visible to the naked eye. The clearing of the water and resulting
increase in light penetration may have been the catalyst of an observed change from diatoms to
green algae in the ozone treatment. In humic rich systems, ozone may release algae from light
limitation through added oxidation of light-absorbing molecules leading to a potential increase in
benthic and phytoplankton production or a shift in algal species composition.
The hypothesized and observed effects of humic compounds on microorganisms are
complicated and may be positive or negative. The functional groups on many humic substances
may react with biomembranes leading to increased permeability (Visser 1985). There are some
data to indicate a unimodal response of cell membranes to humic effects. Lower amounts of
humics can stimulate growth with the consequent increased permeability to essential molecules,
but as concentrations increase, growth inhibition ensues (Steinberg et al. 2003; Wang et al.
1999). There is argument over whether or not humic substances are actually taken into cells;
however, smaller fractions of humics have been found inside cells and humic substances < 3.5
kDa are known to pass through plant cell membranes (Nardi et al. 2002; Wang et al. 1999).
Once inside, these molecules tend to have negative effects on algae, reducing photosynthetic
oxygen release and cell yield. The mechanism of effect has not been well discerned but there are
two accepted hypotheses: 1) that humic substances directly quench electrons or 2) that they bind
to the bioquinones in photosystem II and thereby block electron transfer. Either way, ozone
may affect photosynthetic processes.
Prokaryotic algae appear to be more sensitive to humic substances than other algae.
Studies have shown cyanophytes reacting to lower concentrations of humics than other algal

20

groups (Bahrs and Steinberg 2012; Prokhotskaya and Steinberg 2007).

Park et al. (2009) found

a 24% reduction of green algal cell density at the same concentration of rice hull extract that
reduced Microcystis densities by >98%. Prokhotskaya and Steinberg (2007) attribute this greater
sensitivity to the simpler cell structure of cyanophytes –lacking membrane-bound organelles may
leave internal apparatus susceptible to externally introduced substances-- and a smaller suite of
antioxidant enzymes. Other studies have indicated that the differential response to these
compounds may be species and molecule specific with the finding that genera within the
cyanophytes, green algae, diatoms and euglenoids could be “very susceptible” to humic plant
leachates (Martin and Ridge 1999). However, there have been studies that also indicate that
humic substances in low concentrations may actually stimulate growth of algae, including some
cyanophytes (Fagerberg et al. 2009; Sun et al. 2006). If dissolved humic substances do limit or
promote the growth of different algal taxa in natural systems, then ozone breakdown of these
substances may change competition dynamics and potentially cause a shift in algal community
composition.

Synthesis
Tropospheric ozone levels over extensive regions of the world are much higher than the
natural concentrations expected from stratospheric transfer and oxidation of natural NOx.
During warm seasons, regions all over the world reach concentrations that exceed the US EPA’s
national health-based standard of 80 ppb and maximum levels could reach over 400 ppb.
Perhaps more important is the prediction that ambient ozone concentrations will increase in the
future, as well as the extent of regions affected by high ozone days. These levels have and will
significantly change the oxidative potential of the atmosphere. Considering the relative ease of

21

dissolution into surface waters, we must turn our attention to the potential of tropospheric ozone
as an aquatic pollutant.
The myriad of documented effects of ozone pollution on terrestrial primary producers
through damage to membranes and proteins (especially those of the photosynthetic apparatus)
and pigment loss, suggest a potential effect on aquatic photosynthesizers and heterotrophic
microbes that depend on that production. My preliminary research on periphyton grown in a
continuous 300 ppb ozone atmosphere demonstrated a division-level community shift and the
many studies showing strong oxidative stress corroborate this hypothesis and indicate the need
for research on algal communities. Thus, I hypothesized a shift in algal taxonomic composition
and related ecosystem processes as a result of differential tolerances to oxidative stress by ozone.
Ozone and its dissolution species react with most organic molecules that they contact and
humic substances, which are not readily taken up by bacteria, are available for oxidative
breakdown and are often the primary sink for ozone in water treatment. As discussed, this
oxidative breakdown leads to the creation of low molecular weight, bioavailable molecules and a
loss of water color and hence reduced light attenuation. The effects of these changes on light
penetration and bacterial production and the potential to ameliorate the negative effects of
humics may be indirect factors affecting algal communities (Fig. 1.3). Thus I hypothesize
dissolved organic carbon will mediate ozone-algal interactions.
The objectives of my dissertation are to manipulate atmospheric ozone concentration and
DOC concentrations and determine their interactive effects on algal and bacterial communities.
Given the absence of detailed research on ozone effects on aquatic ecosystems in ranges
expected for tropospheric ozone pollution, my research will provide the foundation for future
studies of the effects of ozone pollution on aquatic and semi-aquatic ecosystems. Specifically,

22

Chapter 2 focuses on the community-level responses of natural, multi-division periphytic algal
assemblages to three different ozone levels and two different dissolved organic carbon
concentrations. Chapter 3 is a study of the effect of elevated atmospheric ozone levels on
heterotrophic bacteria within the original periphyton matrix and in the absence of algal
interactions. Finally, Chapter 4 explores the independent responses of diatom and cyanophyte
communities to ozone pollution without interdivisional interactions or resource competition. The
potential for further research on atmospheric ozone effects on algae, other aquatic communities,
and ecosystem functions is great. Invertebrates, vertebrates, and aquatic plants are all potentially
affected either physiologically or behaviorally.

23

Chapter 2: Impacts of elevated atmospheric ozone and the interactive effects of dissolved
organic carbon content on benthic algal community composition and biomass.

Abstract
Atmospheric ozone concentrations are increasing in many parts of the world. Ozone
dissolution into surface waters may directly influence algal growth through oxidative stress or
indirectly through changes to water chemistry. I used microcosms to address how atmospheric
ozone concentrations, independently and interacting with dissolved organic carbon (DOC, in the
form of humic substances) may change benthic algal assemblages. My results clearly indicate
that both DOC concentration and atmospheric ozone levels affect algal assemblage composition
and biomass. DOC concentration had a large independent effect on total algal biovolume but the
direction of effect was dependent on algal taxonomic division. Ozone effects were significant
but also varied with algal division and DOC treatment. I suggest that atmospheric ozone is a
potential aquatic pollutant, but that the degree to which it affects benthic algae will depend on
initial assemblage composition and the amount and quality of DOC.

Introduction
Ground-level concentrations of the oxidizing pollutant, ozone, are on the rise as fossil
fuel consumption and vehicle use increase worldwide (Granier and Brasseur 2003; IbarraBerastegi and Madariaga 2003). The levels of ozone recorded in the troposphere currently range
from 10 to close to 400 ppb (Bobbink 1998; Eder et al. 1993; Fast and Heilman 2003; Vingarzan
2004; Zambrano and Nash 2000). During the warm seasons, regions all over the world regularly
reach tropospheric concentrations that exceed the US National Ambient Air Quality Standard

24

(NAAQS) of 75 ppb. Perhaps more important is the prediction that background ozone
concentrations and the extent of regions having high ozone days will continue to increase in the
future. In fact, some models predict that summer background levels, generally defined as the
concentration attributable to long-range transport, resuspension of previous emissions, and
nonanthropogenic sources, will regularly exceed the NAAQS standard within 100 years
(Vingarzan 2004).
Due to the long-range transport of ozone precursors from their source and the nature of
the chemical reactions that create ozone in the troposphere, emissions in one area may change
ozone levels long distances away (Kondo et al. 2003; Liu et al. 2003; Parrish et al. 1993;
Thompson et al. 2002) and concentrations may be as high or higher in rural areas downwind of
precursor sources than in urban areas (Chameides et al. 1997; Debaje et al. 2003; Metcalfe et al.
2002). In fact, higher ozone levels are routinely recorded in national parks downwind of urban
areas than within urban areas (N.P.S. 2011).
Considerable research has been done showing the negative effects of ozone pollution on
terrestrial systems, which can be great (Bermejo et al. 2003; Braun et al.; Heath 1994; Wonisch
et al. 2004). Photosynthetic cells are naturally high in reactive oxygen species (ROS) and very
sensitive to oxidative overload, thereby making them a sensitive receptor of ozone stress.
Primary producers of all types, in agricultural as well as natural settings, are sensitive to elevated
levels of atmospheric ozone, resulting in foliar injury, pigment loss, premature senescence, and
decreased photosynthetic/growth rates (Bermejo et al. 2003; Braun et al. 2004; Heath 1994;
Wonisch et al. 2003). In fact, ozone is the only major air pollutant that has been shown to induce
plant damage at commonly measured concentrations (Schraudner et al. 1997). These negative

25

effects are obvious at the organism scale but have also been shown to affect overall crop yield
and forest function (Avnery et al. 2010).
The study of the response of algae to atmospheric ozone has been limited to lichen cells
and symbiotic cyanobacteria. Lichens have long been used as indicators of air pollution. Recent
studies have indicated that a response to ozone stress led to precipitous declines in lichen species
richness and cover in areas outside of Mexico City and in coniferous forests in California
(Avnery et al. 2010; Canas and Pignata 2000; Garcia et al. 2000; Scheidegger and Schroeter
1995; Sigal and Johnston; Zambrano and Nash 2000; Zambrano et al. 1999). This decline has
been attributed to effects on the algal component due to a loss of chlorophyll a and the
subsequent reduction of photosynthetic rate. Nitrogen-fixing cyanobacterial symbionts have
shown inhibited growth, nitrogen fixation, and heterocyst formation when exposed to ozone
concentrations characterized as being “below ambient” for semi-urban conditions (Hur and
Wellburn 1994).
These negative responses of terrestrial algae to ozone pollution suggest a potential impact
on aquatic algae, as ozone may readily dissolve into surface waters. Dissolved ozone may
directly react with ecosystem constituents or the effect may be through secondary oxygen species
created during dissolution, such as hydrogen peroxide and the hydroxyl radical. Algal
communities may be directly affected by the oxidative stress caused by these ROS. The
response of aquatic algae to ozone has only been studied in regards to wastewater treatment.
These studies pumped ozone, in very large quantities, directly into the water as a means of
disinfection. At sufficient levels, algal cells are oxidized to death and even their byproducts
may be destroyed (e.g. cyanotoxins) (Yun et al. 1997). There are documented differences in
algal response to ozone and different concentrations and exposures times are necessary to

26

eliminate different species (Hoeger et al.; Lai et al. 2006; Lai et al. 2002; Paralkar and Edzwald
1996; Widrig et al. 1996). Although it has been noted that photosynthetic rates of algae after
ozonation were severely inhibited even by short ozone contact (Yun et al. 1997), studies of the
sub-lethal effects are virtually non-existent.
Algal communities may also be indirectly affected by the response of other organisms
and/or dissolved substances to the increased oxidant levels. Ozone stress may change interspecies interactions within the biofilm by differentially affecting various algal groups,
heterotrophic bacteria, and algal pathogens. Also, ozone and its dissolution species react with
most molecules that they contact and humic substances, which are not readily taken up by
bacteria, are available for oxidative breakdown and are often the primary sink for ozone in water
treatment (Cho et al. 2003). This breakdown of larger compounds of DOC changes the light
regime by reducing light attenuation, produces more labile carbon compounds and potentially
liberates inorganic nutrients that may be utilized by algal cells (Freese et al. 1999; Lindell et al.
1995; Tranvik 1988; Wetzel 2001a).
This study is the first to investigate the potential of atmospheric ozone as an aquatic
pollutant and the potential mitigating effects of humic substances. The experimental approach
consisted of a three (0, 80, 250 ppb ozone) by two (high and low DOC) factorial design testing
the independent and interactive effects of atmospheric ozone level and initial DOC concentration
on periphyton assemblages. My objectives were 1) to look for evidence of ozone or DOC effects
on total algal biomass and algal assemblage composition, and 2) to determine if ozone effects
were mediated by DOC concentration in the water column.

27

Methods
Overall Experimental Set-up
Three ten-gallon glass aquaria were used to create treatment chambers (Fig. 2.1), each having a
different atmospheric ozone concentration: a control treatment with 0 ppb O3; a medium
treatment with about 80 ppb O3; and a high treatment with about 250 ppb O3. Aquarium lids
were sealed with weather-stripping and lined with PTFE sheeting to maintain the treatment
atmospheres. Chemically non-reactive PTFE and PFA plastics were required for this experiment
to resist corrosion by the large ozone concentrations pumped into the experimental chambers.
Ozone was created and measured using a Thermo Scientific Model 49™ O3 analyzer. The air
used to generate the ozone was ambient air filtered through a carbon filtration unit prior to
entering the generator. PFA tubing conducted the ozone from the generator into two of the
aquaria and back from all three to the sample analyzer. Flow into each aquarium was controlled
using a PTFE ball valve and manipulated until the target ozone concentration was measured in
the chamber air. The control treatment was supplied with air from an aquarium bubbler filtered
through another carbon filtration device. Ozone in each chamber was monitored and
manipulated daily to insure an atmospheric concentration above the microcosms that was
consistently ± 10 ppb of the target concentration. Ozone levels in the control never measured
above 0 ppb. All three chambers were placed on an Eberbach EL600™ Orbital-Reciprocal
Shaker Table to create water motion, set to 96 rotations per minute.
Light was provided by fluorescent, natural-spectrum bulbs mounted above the chambers,
2

which generated 63.65 ± 2.16 µE/cm /sec. To simulate natural, summer, diurnal cycles, ozone
generation and light were on a timer, producing from 7:00 a.m. to 9:00 p.m. daily. All three
28

chambers were placed on an Eberbach EL600™ Orbital-Reciprocal Shaker Table to create
water motion, set to 96 rotations per minute.

Air
Ozone
Generator/Monitor

Carbon Filter

250 ppb Ozone

80 ppb Ozone

0 ppb Ozone

High DOC
Air Pump

Carbon Filter

Low DOC

Fig. 2.1. Experimental set-up. Rectangles represent individual chambers and circles represent
microcosms within each chamber.
Water, for use in the experiment, was taken from Vermillion Creek, Rose Lake Wildlife
Area, Clinton County, MI and an adjacent, partially formed oxbow wetland. The two water
sources are separated by sediment on one arm of the oxbow and a beaver dam on the other.
29

When water levels are low, exchange between the two sources is limited. Each source constituted
a different “DOC water treatment.” However, each source differed greatly in the concentrations
of dissolved organic carbon (DOC).

Other primary drivers of algal growth, such as pH and

temperature, did not differ greatly between the two. Macronutrients (silica, nitrogen, and
phosphorus) were directly measured and addressed. The creek water served as the growth
medium for the “low doc water” treatment and the wetland water served as growth medium for
the “high doc water” treatment. Initial DOC concentration for the high DOC water was 25.1 mg
C/L and that for the low doc water was 14.152 mg C/L, measured as nonpurgable organic carbon
(NPOC), and the high- and low-DOC sites had pH values of 7.4 and 8.2, respectively. These
DOC concentrations are both high by global standards but represent common levels for rivers in
our area. The high DOC water was noticeably stained with humic substances from leaf
decomposition, and the low DOC water was visibly clear.
Benthic and epiphytic algae were collected by scrubbing biofilms from numerous
substrates in each habitat. These samples were then combined into one suspension that was used
as the seed population for both DOC water treatments. Half of this composite algae sample was
added to the low DOC source water and the other have to the high DOC source water before
commencement of the experiment. Therefore the seed populations for both treatments were
very much the same, differing only in the suspended algal cells that may have been present
during initial water sampling.
Eight 4-inch diameter ceramic dishes were used as microcosms, each containing 150 mL
of source water and an inoculum of algae from the composite sample. Each dish contained five
2.5 x 2.5 cm unglazed ceramic tiles. Four of the eight microcosms in each of the three ozone
chambers (0, 80 ppb, and 250 ppb) were designated for either low or high doc water treatments.

30

th

th

st

Fresh source water was added to the microcosms on the 7 , 14 , and 21 days of the growth
period, to refresh nutrients and keep the water level constant. Water temperature stayed at a
constant 23-24°C for the four-week duration of the experiment.
Sample Collection
The first sampling date was determined by the appearance of a visible algal filmon the
tiles. Periphyton mats and water samples were collected after two and four weeks of treatment.
Two tiles were sampled from each replicate during the first sampling period and the remaining
three tiles were sampled at week 4. Periphyton was removed from each tile using a toothbrush
and deionized (DI) water from a squirt bottle. Periphyton samples from individual tiles were
pooled by replicate and date. Subsamples were taken from the homogenized, composite sample
for chlorophyll a measurement, algal biovolume estimate, and bacterial counts. Algal samples
were preserved with gluteraldehyde (2%).
Water samples were filtered through Whatman™ glass fiber filters, type GF/F, for
analysis of dissolved organic carbon, total dissolved nitrogen (TDN), and soluble reactive
phosphorus (SRP). All water samples were frozen until analysis.
Laboratory Analyses
Soft algal cells were identified and counted in a Palmer-Maloney counting cell at 400X
until 400 natural units were counted. This is a modification of a standard method for assessing
taxonomic composition and cell density of algal samples used in national ecological assessments
(Lowe and Laliberte 1996). I modified the counting method by increasing the number of natural
units counted to increase precision of taxonomic and cell density measurements. Natural units
are defined as the normal growth form of the alga, such as single cells, filaments, or colonies.
The number of cells composing each natural unit was also recorded. Measurements were made
31

of each taxon during identification for calculating total biovolume as an estimate of algal
biomass (Hillebrand 1999).
Diatom frustules were cleaned using nitric acid for cytoplasm digestion. A ratio of one
part sample to two parts nitric acid was added to a beaker and simmered on a hot plate for one
hour. Samples were rinsed several times with distilled water until the pH was neutral by
allowing frustules to settle for 24 hours, siphoning off the supernatant, and adding more water.
Neutral samples were concentrated, and then subsamples were dried onto coverslips and
subsequently mounted on microscope slides using Naphrax™ mounting medium. Diatom
identifications were made on clean frustules at 100X on a Leica compound microscope to a total
of 600 valves.
Chlorophyll a (chl a) concentrations were also used as a measure of total algal biomass.
Algae from each chlorophyll subsample were filtered onto Whatman™ GF/C glass fiber filters,
extracted overnight in ethanol, diluted if necessary, and analyzed fluorometrically before and
after acidification to correct for pheopigments (APHA 1998).
Total dissolved carbon (DOC), measured as non-purgeable organic carbon, and total
nitrogen concentrations were measured in filtered water samples by Dr. Stephen Hamilton’s lab
at Michigan State University using a Shimadzu TOC-VCPH™ Carbon Analyzer with a total
nitrogen module (TNM-1) and an ASI-V Autosampler. Because the water samples were filtered
prior to analyses, the TN readings did not include particulate nitrogen and were a measure of
total dissolved nitrogen (TDN).
Soluble reactive phosphorus (SRP) was analyzed on samples that had been frozen. Due
to a freezer malfunction, filtered water samples from week 2 were lost; therefore, SRP

32

measurements for the week 2 samples were done on a set of samples filtered after freezing. SRP
concentration was measured manually following the ascorbic acid method (APHA. 1998) on a
Spectronic® Genesys™ 2 Spectrophotometer by Spectronic Instruments.
Statistical Analysis
To test whether DOC, ozone, and sampling date (week) had an effect on total algal
biovolume, chl a, or division-level biovolume, I conducted three-way ANOVAs using the GLM
function in SYSTAT Version 12.0. To better understand the relationships between independent
and dependent variables, in the case of a significant three-way interaction, two-way ANOVAs
were conducted on the ozone/DOC interaction at each week. Two-way ANOVAs on ozone and
week at each DOC treatment were also done to indicate significant changes from week 2 to week
4 and to identify any significant relationships among the dependent and independent variables
that were not found in the former two-way ANOVA. All two-way ANOVA results may be
found in Appendix 1. Tukey’s HSD tests were used for pair-wise comparisons of independent
and interaction effects when ozone was a significant variable. All continuous variables were logtransformed prior to analysis. Because of potential issues with non-normality and some unequal
variances, the ANOVA results were corroborated by the non-parametric Kruskal-Wallis Oneway Analysis of Variance (KW). Results did not significantly differ between the parametric and
nonparametric tests.
To determine if nutrient concentration differed as a function of DOC, ozone, week, and
resulting responses of algae, I once again used a three-way ANOVA with subsequent Tukey’s
HSD pairwise comparisons. Correlations served to identify significant bivariate relationships
between measured nutrient concentrations and biomass. Nutrient data were log-transformed.

33

Results
Dissolved Organic Carbon
DOC treatment, ozone, and week each independently affected measured DOC
2

concentration (r =0.869) (Fig. 2.2). The concentration of total DOC increased from week 2 to
week 4 in all samples (F1,34=29.133; p=0.000), however, it was higher in the high DOC than in
the low DOC source water at both sampling dates (F1,34=170.498; p=0.000).

High DOC

Low DOC

50
50

DOC(mg/L)
DOC (mg/L)

43
43
36
36
29
29

a
a
b
b
b
b
OZONE

22
22

15
15

2
2

Week
WEEK

4
4

2
2

Week
WEEK

4
4

0 ppb
80 ppb
250 ppb

Fig. 2.2. Measured DOC concentration by ozone and week, for each DOC treatment. Lowercased letters indicate significant differences between ozone treatments at Week 2. Upper-case
letters, differences at Week 4. An asterisk denotes a significant change from Week 2 to Week 4.
Error bars represent one standard error above and below the mean.
*For interpretation of the references to color in this and all other figures, the reader is referred to
the electronic version of this dissertation.

Ozone had a negative effect on total DOC as DOC was lower in both medium and high ozone
treatments than in the zero ozone treatment (F2,34=8.099; p=0.002 and p=0.008, respectively).

34

These differences were also visible in the water color. Over the first two weeks of the
experiment, in the high DOC water, water color decreased greatly in the high and medium ozone
treatments but not in the zero ozone treatment.
Total Algal Biomass
Ozone, DOC, and week complexly affected total algal biomass, with all interaction
effects being significant (Table 2.1). Subsequent two-way ANOVAS on the interaction between
DOC and ozone at each week indicated that DOC effects were dependent on ozone treatment and
this effect varied with week.
Table 2.1. ANOVA table for log-transformed values of total biomass (as measured by
biovolume and chlorophyll a) as a function of DOC, ozone, and week.
Source

df

Mean sqr

F-ratio

P

Total Biovolume (r2=0.782)
Ozone

2

0.307

1.762

0.187

DOC

1

8.021

46.050

0.000

Week

1

0.073

0.419

0.522

Ozone*DOC

2

0.881

5.060

0.012

Ozone*Week

2

2.033

11.672

0.000

DOC*Week

1

2.462

14.136

0.001

3 way

2

2.162

12.412

0.000

Error

35

0.174

Chlorophyll a (r2=0.705)
Ozone

2

1.68

2.175

0.129

DOC

1

0.397

0.514

0.478

Week

1

1.601

2.072

0.159

Ozone*DOC

2

2.128

2.755

0.077

Ozone*Week

2

8.587

11.114

0.000

DOC*Week

1

26.151

33.848

0.000

3 way

2

5.910

7.649

0.002

Error

35

0.773

35

2

At week 2, the effect of DOC on total biovolume and chl a was dependent on ozone (r =0.883;
2

F2,18= 3.553, p=0.05 and r =0.907; F2,18= 4.337, p=0.029, respectively). At week 4, there was
once again a significant interaction of DOC and ozone on both total biovolume and chl a
2

2

(r =0.632; F2,18= 10.884, p=0.001 and r =0.627; F2,18= 5.141, p=0.018, respectively).
In the absence of ozone, DOC had a negative effect on total biomass over the entire study
period as total biovolume was three times greater at week 2 and five times greater at week 4
(p=0.002 for both weeks), Chl a was also significantly greater in low than high DOC water at
week 2 (p=0.004) (Fig. 2.3). In the medium and high ozone, DOC initially had a negative effect
on total algal biomass but a positive effect later. At week 2 in the high DOC water, total
biovolume and chlorophyll a were six times lower in high ozone (p=0.000 for both) and three
times lower in the medium ozone (p= 0.006 and p=0.000, respectively). However, due to a
three-fold increase in biomass in the high and medium ozone treatments in the high DOC water
but not in the low DOC water, by week 4 biovolume did not statistically differ between high and
low DOC water. Chlorophyll a was greater in high DOC water with the high ozone treatment
(p=0.016).

36

Low DOC

2

Chlorophyll a (mg/cm )
Chlorophyll a (mg/cm2)

3
2
Total Algal Biovolume (mm cm )

Total Algal Biovolume (mm /cm )
3
2

High DOC
0.65
0.65

0.56
0.56
0.47
0.47

a
a
a
a

0.38
0.38
0.29
0.29

0.11
0.11
0.02
0.02

OZONE

*B
*B
*B
*B

0.20
0.20

a
a
b
b
b
b

b
b
*
*

A
A

2

Week
WEEK

4

2
2

Week
WEEK

0 ppb
80 ppb
250 ppb

4
4

10.00
10.00

*B
*B
A,B
A,B

1.00
1.00

OZONE

a
a

A
A

*
*
*

*

0.10
0.10

b
b
2
2

Week
WEEK

4
4

2
2

Week
WEEK

0 ppb
80 ppb
250 ppb

4
4

Fig. 2.3. Measures of total algal biomass by ozone and week, for each DOC treatment. Lowercase letters indicate significant differences between ozone treatments at week 2. Upper-case
letters, differences at week 4. An asterisk denotes a significant biomass change from week 2 to
week 4. Error bars represent one standard error above and below the mean.

Ozone effect in the low DOC water was negative but limited to the medium ozone
treatment and was only significant at the first sampling date (Fig. 2.3). Biovolume was three
times lower in medium ozone than in either the zero or high ozone treatments (p=0.004 and
p=0.002, respectively) at week 2. There was no significant difference in chl a concentration at

37

week 2. Between sampling times, algal biomass decreased in zero and high ozone but not in
medium ozone. Biovolume significantly decreased with high ozone over this period (p=0.005)
but did not significantly change in the medium or zero ozone treatments. Chl a significantly
2

decreased in both zero and high ozone treatments (r =0.751; F1,18= 50.733, p=0.000 and
p=0.001, respectively).
In high DOC water, ozone had a negative effect on total algal biomass early in the study
period but a positive effect later (Fig. 2.3). At week 2, total biovolume was greater in zero ozone
than in medium ozone (p=0.011) and also high ozone, although not significantly (p=0.129). Chl
a was significantly greater in zero ozone than in either high or medium ozone treatments
(p=0.001 for both). As mentioned above, from week 2 to week 4, total biovolume increased
threefold in the high and medium ozone treatments (p=0.019 and p=0.011, respectively) but
decreased in zero ozone (p=0.019). Chl a changed in the same direction but the change was
significant only for the high ozone treatment (p=0.001). By week 4, the zero ozone assemblages
had less biovolume than those in high and medium ozone (p=0.001 and p=0.041, respectively).
Chl a shows the same trend but only the difference between the high and zero ozone treatment
was significant (p=0.003).
Division-level Biomass
Assemblages were composed of three major divisions of algae: diatoms, cyanophytes
and chlorophytes. The DOC and ozone treatments had very different effects on each division.
Early assemblages in all treatments were diatom-dominated by biovolume but, by week 4,
cyanophytes were dominant in the high and medium ozone in the high DOC water.
Chlorophytes were second by biovolume across treatments and sampling date.

38

As with total

algal biomass, all interactions among DOC, ozone, and week significantly affected both diatom
and cyanophyte biovolume (Table 2.2).

Table 2.2. ANOVA table for log-transformed values of division-level biovolume as a function
of DOC, ozone, and week.
Source

df

Mean sqr

F-ratio

P

Diatom Biovolume (r2=0.853)
Ozone

2

0.257

1.273

0.293

DOC

1

14.590

72.358

0.000

Week

1

10.928

54.196

0.000

Ozone*DOC

2

0.795

3.944

0.029

Ozone*Week

2

2.961

14.686

0.000

DOC*Week

1

1.237

6.134

0.018

3 way

2

2.595

12.412

0.000

Error

35

0.202

Cyanophyte Biovolume (r2=0.722)
Ozone

2

6.558

3.107

0.057

DOC

1

98.422

46.623

0.000

Week

1

52.478

24.859

0.000

Ozone*DOC

2

0.018

0.009

0.992

Ozone*Week

2

1.178

0.558

0.577

DOC*Week

1

14.281

6.765

0.014

3 way

2

11.133

5.274

0.010

Error

35

2.111

Chlorophyte Biovolume (r2=0.481)
Ozone

2

0.771

0.830

0.445

DOC

1

2.216

2.385

0.131

Week

1

14.890

16.026

0.000

Ozone*DOC

2

0.327

0.352

0.705

Ozone*Week

2

0.226

0.243

0.785

DOC*Week

1

9.291

9.999

0.003

3 way

2

0.200

0.215

0.808

Error

35

0.929

39

Subsequent two-way ANOVAs (DOC water and ozone by week) indicated independent
2

effects of DOC treatment and ozone on diatom biovolume at week 2 (r =0.894; F1,18= 107.064,
p=0.000 and F2,18= 19.656, p=0.000, respectively). In addition, there was a significant
2

interaction effect of ozone and DOC water at week 4 (r =0.708; F2,18= 10.446, p=0.001) (Fig.
2.4).

In the absence of ozone, DOC had a negative effect on diatom biomass over the entire

study period. Diatom biovolume was lower in high DOC water in zero ozone at both week 2 and
week 4 (p=0.001 for both). In the medium and high ozone treatments, DOC initially had a
negative effect on diatom biomass (high ozone: p=0.000 and medium ozone: p=0.001) but by
week 4, there was no significant difference between DOC treatments.
From week 2 to week 4, diatom biovolume either decreased or remained the same in both
2

high and low DOC water and this response was dependent on ozone (r =0.734; F2,18= 15.139,
2

p=0.000 and r =0.814; F2,18= 3.083, p=0.000, respectively). In the high DOC water, diatom
biovolume significantly declined in the zero ozone treatment (p=0.000) but was maintained in
high and medium ozone treatments. In the low DOC water, diatom biovolume significantly
declined in both the zero (p=0.023) and high ozone treatments (p=0.000) but was maintained
with medium ozone. By week 4, the negative effect of DOC was only significant in the zero
ozone treatment (p=0.001).
Medium ozone had a significantly negative effect on early diatom biovolume (F2,18=
19.656, p=0.000) in both high and low DOC water however this difference was greater in the
low DOC water.

By week 4, in high DOC water, the zero ozone assemblages had

approximately a quarter of the diatom biomass that was present in the high or medium ozone

40

treatments (p=0.013 and p=0.022, respectively). In low DOC water, diatom biomass was 70%
lower in the high ozone than in either the medium or zero ozone treatments (p=0.026 and
p=0.029, respectively).

41

2

Diatom Biovolume (mm/cm2) )
3 /cm
Diatom Biovolume (mm

High DOC

3

0.10000
0.10000

Low DOC
a
a
b
b

a
a
a,b
a,b
a
b

*B

A
A

*A

*B
*B

0.01000
0.01000
0.00100
0.00100

OZONE
0 ppb
80 ppb
250 ppb

0.00010
0.00010

2

Week
WEEK

4

2

Week
WEEK

4

2

ChlorophyteBiovolume (mm3/cm2)] ) Cyanophyte Biovolume (mm /cm )
3 2
Log[Chlorophyte Biovolume (mm /cm
Cyanophyte Biovolume (mm/cm )

0.00001
0.00001

*A
*A

0.01000
0.01000

3

0.10000
0.10000

B
B

0.00100
0.00100

OZONE
0 ppb
80 ppb
250 ppb

0.00010
0.00010

2

Week
WEEK

4
4

2

Week
WEEK

4
4

3

2

0.00001
0.00001

0.10000
0.10000
0.01000
0.01000
0.00100
0.00100

OZONE

0.00010
0.00010

0.00001
0.00001

2

Week
WEEK

4
4

2

Week
WEEK

4
4

0 ppb
80 ppb
250 ppb

Fig. 2.4. Measures of division-level biomass by ozone and week, for each DOC treatment.
Lower-case letters indicate significant differences between ozone treatments at week 2. Uppercase letters, differences at week 4. An asterisk denotes a significant biomass change from week
2 to week 4. Error bars represent one standard error above and below the mean.
42

Cyanophytes were also significantly affected by the interactions of DOC, ozone, and
week (Fig. 2.4). Subsequent two-way ANOVAS (ozone and DOC by week) indicated that DOC
had a positive effect on cyanophyte biomass at both sampling dates. At week 2, cyanophyte
2

biovolume was 3 times greater in the high DOC water than in the low DOC water (r =0.508;
F1,18= 10.579, p=0.004). By week 4, the difference had increased to two orders of magnitude
2

(r =0.731; F1,18= 37.976, p=0.000). This great difference between cyanophyte biovolume in the
high and low DOC waters at week 4 can be explained by a cyanophyte “bloom” that occurred
only in high DOC water in the high and medium ozone treatments.

A two-way ANOVA

(ozone and week by DOC) verified the significant interaction of ozone and week on cyanophyte
2

biomass in the high DOC water (r =0.821; F2,18= 9.025, p=0.002). Indeed, cyanophyte
biovolume increased by two orders of magnitude between week 2 and week 4 in the high and
medium ozone atmospheres (p=0.000 and p=0.002, respectively). Consequently, cyanophyte
biovolume at week 4, in the high DOC water, was approximately 10 times greater in the high and
medium ozone treatments than in zero ozone (p=0.012 and p=0.026, respectively). There was no
significant ozone effect on cyanophyte biomass in the low DOC water.
Chlorophytes were also a major constituent of all assemblages throughout the
experimental period. Chlorophytes were unaffected by ozone treatment but were positively
affected by DOC at week 4 (Table 2.2). At week 2 there was no difference between chlorophyte
biovolume in the low DOC or high DOC waters. Chlorophyte biovolume significantly increased
during that time in the low DOC water (p=0.000) but not in the high DOC water. Consequently,
at week 4 chlorophyte biomass was significantly greater in low DOC water than in the high DOC
2

water (r =0.481; F1,18= 9.999; p=0.003).

43

Nutrients
DOC, ozone, and week were all independently significant factors affecting total dissolved
2

nitrogen (r =0.696) (Fig. 2.5). TDN was lower in the low DOC water (F1,34=21.710, p=0.000)
than in the high DOC water across all ozone treatments. It was also lower in the medium ozone
than in the high or zero ozone treatments (F2,34=13.099, p=0.000 both) in both the low and high
DOC waters. From week 2 to week 4, TDN increased across DOC and ozone treatments
(F1,34=8.377, p=0.007).
SRP concentrations were independently affected by DOC and ozone treatment but the
2

ozone effects varied with week (r =0.724) (Fig. 2.5). SRP concentrations were higher in the
high DOC water throughout the study (F2,31=14.96; p=0.001). At week 2, SRP was higher in
the high ozone treatment (F2,17=5.504; p=0.014). From week 2 to week 4, SRP in the high and
medium ozone treatments decreased significantly (F1,31=4.580; p=0.001 and p=0.006,
respectively) and at week 4 SRP in the zero ozone treatment was higher than in both the high and
medium ozone (p=0.018 and p=0.05, respectively).
Silica concentrations did not differ significantly between the two DOC waters but
significantly varied with ozone treatment (Fig. 2.5). Silica was higher in the zero ozone than in
the high or medium ozone treatments (F2,30=8.068; p=0.013 and p=0.002, respectively) and
increased significantly from week 2 to week 4 across treatments.

44

High DOC

Low DOC

(mg/L)

TDN (mg/L)
TDN

3.0
3.0

2.3
2.3

aa
aa

1.6
1.6

OZONE
*
*

0.9
0.9

b
b

2
2

Week
WEEK

4

2

Week
WEEK

0 ppb
80 ppb
250 ppb

4

1.000
1.000

A

SRP (mg/L)
SRP (mg/L)

A

0.100
0.100

A,B

A ,B

0.001
0.001

OZONE

B

B

0.010
0.010

*

*

2
2

Week
WEEK

4
4

2

Week
WEEK

0 ppb
80 ppb
250 ppb

4

Silica (mg/L)
Silica (mg/L)

10.000
10.000

1.000
1.000

OZONE

0.100
0.100

*

*

0.010
0.010
2
2

Week
WEEK

2

4
4

Week
WEEK

0 ppb
80 ppb
250 ppb

4

Fig. 2.5. TDN, SRP, and silica concentration (mg/L) by ozone and week, for each DOC
treatment. Lower-cased letters indicate significant differences between ozone treatments at week
2; upper-case letters, differences at week 4. An asterisk denotes a significant biovolume change
from week 2 to week 4. Error bars represent one standard error above and below the mean.
45

Total algal biomass (as biovolume or chl a) was not correlated with nutrients across
treatments and sampling periods. Cyanophyte biovolume was positively correlated with TDN
(r=0.390; p=0.001) and diatom biovolume was negatively correlated with silica concentration
(r=-0.341; p=0.027).

However, when analyzed separately at each week, there were no

correlations between nutrient concentrations and total algal biomass or division-level biomass.

Discussion
Our results clearly indicate that both DOC concentration and atmospheric ozone levels
affect algal assemblage composition and biomass. This study presents the first evidence that
atmospheric ozone, at levels measured in the troposphere and allowed to dissolve naturally into
the water column, may cause changes to aquatic assemblages. Our results also indicate that algal
divisions (diatoms, cyanophytes, and chlorophytes) respond differently to the independent and
interactive effects of DOC concentration and elevated ozone.
After 2 weeks of exposure, total algal biomass, which was dominated by diatoms, was
negatively affected by DOC. This appeared to be driven by the diatom response, as DOC had a
negative effect on diatom biomass, no effect on chlorophytes, and a positive effect on
cyanophytes. This negative effect of DOC on diatoms was mostly likely indirect as few studies
have shown direct negative effects of DOC on diatom growth. TDN did vary statistically with
DOC treatment but there was no significant correlation with total or diatom biomass at this
sampling date. Also nitrogen levels were high enough to not be a limiting factor for diatom
growth (Rier and Stevenson 2002). Light attenuation may have limited total diatom growth in
the colored, high DOC water, although diatoms have been shown to be very tolerant of low light
conditions (Steinman et al. 1989; Pillsbury and Lowe 1999).

46

Although different source waters were used in the DOC water manipulation, DOC
concentration itself, most likely regulated the observed effects of water treatment. The two
treatment water sources were from a connected system separated only by sediment and a beaver
dam so some water flow was maintained throughout the year and catchment geology was the
same. The primary difference between the two areas was retention of leaf litter which led to a
higher DOC concentration in the wetland area. pH differed slightly but, based on autecological
similarities of the diatom taxa in samples, the difference in pH was most likely not great enough
to cause a significant shift in the division level taxonomy. In fact, cyanophytes, which are
generally more sensitive to low pH (Wetzel 2001c), were in greater abundance in the lower pH
water. Also, any pH differences between the two treatments would most likely have been a
secondary effect of DOC concentration since DOC is often a primary source of acidity in aquatic
ecosystems. The algal seed populations for experiments were the same for each treatment so
initial source water communities should not be the cause of any differences among treatments.
During the experiment, microcosms warmed to room temperature and were maintained at that
temperature, in both treatments, throughout the experiment. Macronutrient concentrations were
measured directly during the experiment and accounted for during analysis. Light attenuation
differed in a small amount, initially, between the two treatments but this difference would have
been a secondary effect of humic DOC content in the water. Thus I argue experimental effects in
DOC water treatments were likely the direct or indirect effects of DOC because the most
important factors ecological factors that could affect algal biomass and composition were
measured directly for comparison or held constant and do not provide plausible explanations for
observed effects of DOC water treatments.

47

I did not measure iron concentration, but humic substances are known to chelate iron,
potentially reducing its availability to photosynthesizers (Imai et al. 1999; Wetzel 2001a).
Although speculative, iron limitation would explain both the lower diatom and greater
cyanophyte biomass at week 2 in the high doc water. Many cyanophyte species will produce
soluble, iron-chelating compounds (siderophores) in response to iron limitation which have been
shown to give them a competitive advantage over other algal groups (Murphy 1976; Mahasneh
and Tiwari 1992; Imai et al. 1999). Simultaneously, iron limitation of diatom biomass would
reduce competition for other nutrients, particularly phosphorus, explaining the greater
cyanophyte biomass in the high DOC than in the low doc water. However, cyanophytes have
been shown to be inhibited by the light attenuating effects of humic acids (Jackson and Hecky
1980; Sun et al. 2006). Although the water depth was quite shallow, the intensity of water color
at the beginning of the experiment may influenced the light regime enough to give diatoms the
overall competitive advantage at week 2.
Ozone effects at week 2 were influenced by DOC concentrations. In the high DOC water
there was more total algal biomass with zero ozone, indicating a direct negative effect on algal
growth. However, in the low DOC water, only the medium ozone treatment showed a
statistically significant growth effect. This may have been an anomaly, however, but evidence
indicates it was not. If we look at the division level results, diatoms were the only algal group
significantly affected by ozone at this sampling date and only medium ozone had a negative
effect on their growth. It is possible that the oxidative stress caused by medium ozone levels was
enough to slow diatom growth but was not a strong enough stressor to induce cellular defense
mechanisms that would then ameliorate the negative effects at higher levels of ozone.

48

Alternatively, there was an indirect effect caused at medium ozone that was mitigated at high
ozone levels.
Algal assemblage changes from week 2 to week 4 appear to show both senescence and
succession. By week 4, the high biomass diatom assemblages (high and zero ozone treatments in
the low DOC water and zero ozone in the high DOC water) precipitously declined while the
lower biomass diatom assemblages were maintained at close to the initial biomass. These
diatom assemblages most likely reached peak biomass and were experiencing density-dependent
stressors that led to autogenic community succession. Density-dependent periphyton senescence
is common and may be precipitated by low nutrient levels, toxin build-up within the algal film,
or an increase of pathogen/cell interactions (Borchardt 1996; Hewson et al. 2001; Peterson et al.
1993; Wetzel 2001c).

The diatom communities that did not reach that “peak” did not crash and

were maintained at close to their original biomass.
From week 2 to week 4, in the high DOC water only, there was an ozone correlated
succession from diatom dominance to cyanophyte dominance. Chlorophyte and diatom biomass
did not significantly change, while cyanophyte biomass increased greatly in ozone-treated
microcosms. That this is a direct stimulation of cyanophyte growth by ozone or humic DOC is
unlikely, since there was no such stimulation at week 2 and since previous studies indicate
negative direct effects of humic substances on cyanophyte growth (Bahrs and Steinberg 2012;
Kosakowska 2007; Pflugmacher 2001).
A clearly visible reduction of water color coincided with the cyanophyte increase. This
loss of water color can be considered indicative of the breakdown of DOC as many studies have
shown the strong relationship between color units and dissolved organic matter (Wetzel 2001b).
Since color is produced by chromophoric compounds, which are comprised primarily of high to

49

medium molecular-weight dissolved humic and fulvic acids (Wetzel 2001b), we may assume
that an oxidative breakdown of complex humic substances took place in the ozone treatments
during the course of the experiment. It is this breakdown of humic substances that may have
been the indirect mechanism of ozone effect on the cyanophyte populations in the high DOC
water.
The color reduction in the ozone-treated water would have increased available light and
may have released cyanophytes from light limitation. At the same time, this breakdown of
humic substances may have released chelated iron, increasing bioavailability. Cyanophytes have
been shown to quickly respond to iron additions by facilitated production of siderophores (Imai
et al. 1999) and so the coupling of a more favorable light regime with an increase in nutrients
may have been sufficient to stimulate the cyanophyte “bloom” that occurred by week 4.
Competition for iron would also explain why chlorophytes did not take off, as well, when light
increased.
Finally, humic substances are complex molecules with greatly varying molecular weights
and chemical characteristics and studies have shown both stimulation and reduction of algal
growth as a response to added humic compounds (Fagerberg et al. 2009; Steinberg et al. 2003;
Sun et al. 2006; Wang et al. 1999). The bloom may be indicative of a stimulation of
cyanophytes by heterotrophy. Studies have shown that many types of algae exhibit heterotrophic
use of photolytic degradation products of recalcitrant humic substances and even aromatics
(Larson et al. 2002; Semple et al. 1999; Tittel and Kamjunke 2004) as a survival mechanism in
extremely light-limited situations. These organic carbon substrates may be utilized by algae at
natural concentrations and in the presence of bacteria, and it has been suggested that
heterotrophy may also be an efficient mode of metabolism in systems with high levels of labile

50

dissolved organic compounds even in the presence of light (Tittel and Kamjunke 2004; Tittel et
al. 2009). It is possible, then, that the abundance of labile organic compounds being produced
through the oxidation of humic compounds in the ozone treatments may have stimulated
facultative heterotrophy in the cyanobacteria.
Chlorophytes appear to be unaffected by ozone treatment and the significantly lower
chlorophyte biovolume in the high DOC water at week 4 may have more to do with a reduction
of competition with diatoms after the diatom communities declined in the low doc water. This
hypothesis is corroborated by the fact that this difference between DOC treatments was only
significant in the zero and high ozone treatments, the treatments that witnessed the diatom crash.
These complex responses of algae to ozone and their interactions with DOC indicate
many direct, indirect, and interactive causal pathways. In the absence of ozone interactions,
assumed humic DOC generally had a negative effect on algal biomass with the exception of
cyanophytes that were unique in their positive response to higher DOC concentration. These
results support previous research that showed humic control of algal biomass through altering the
light and/nutrient regime in aquatic systems. It is difficult to specifically isolate the role of
ozone when comparing the effects of ozone in both the high and low DOC waters but its effect
was generally greater in high DOC water, negative in the short-term, and positive in the long–
term, and was dependent on algal taxonomic division.
In conclusion, my results indicate that ozone will likely have a greater effect in high DOC
waters, by breaking down complex molecules which will change the light and nutrient regime
and, at least for cyanophytes, this interactive effect appears to be positive. However, the
response of phytoplankton assemblages, which are less physically complex, may be very
different. The interaction of ozone and DOC will cause changes in algal function in ecosystems

51

because their effects differ across divisions of algae with varying nutritional content for
herbivores, nutrient and light requirements, and roles in biochemical cycling. This study
indicates the potential of atmospheric ozone as an agent of change to aquatic algal assemblages,
especially in humic waters, and underscores the importance of considering species interactions
and other complex indirect responses when determining the potential impact of a given stressor.

52

Chapter 3: Mediation of atmospheric ozone effects on benthic heterotrophic bacterial
biomass by dissolved organic carbon and algae

Abstract
Ozone is an EPA regulated atmospheric pollutant that has been shown to greatly damage
human and ecosystem health. Few studies have addressed the potential of atmospheric ozone as
an aquatic pollutant through dissolution into surface waters. The objective of this study was to
examine the direct and indirect effects of dissolved atmospheric ozone on aquatic heterotrophic
bacteria. Ozone effects on bacteria-only assemblages and bacteria/periphyton assemblages were
assessed using microcosms. In the absence of algal constituents, heterotrophic bacteria were
stimulated by ozone treatment. When bacteria grew in the presence of algae, ozone also had a
significant effect on bacterial biomass but these changes were better explained by total algal
biomass than ozone treatment. These results indicate an indirect effect of ozone on heterotrophic
bacteria in benthic periphyton, likely through a change in the availability of organic carbon
and/or altered interspecific interactions.

Introduction
Ozone (O3) is a primary constituent of photochemical smog and is often considered a
pollutant of serious concern because of the great potential for damage to human and ecosystem
health. During summer, regions all over the world reach concentrations that exceed the U.S.
national health standard of 75 ppb, with recorded maximum levels reaching 400 ppb (Vingarzan
2004, Sanchez and Ayala 2008). Tropospheric ozone is spatially and temporally heterogeneous
and while maximum levels have decreased in some areas, such as North America and Europe,
global background ozone concentrations are increasing steadily. Background O3, generally
53

defined as the concentration attributable to long-range transport, resuspension of historical
emissions, and non-anthropogenic sources is predicted to increase by 0.5–2% per year over the
next century, and some models predict that background concentrations will reach 75 ppb by the
end of the this century (Vingarzan 2004).
Ozone damage to terrestrial ecosystems has been extensively studied and damage to
crops, forests, and other natural systems is widely reported (Broadmeadow 1998, Madden and
Hogsett 2001, Niemi et al. 2002, Oksanen 2003, Avnery et al. 2010). Considering the extent of
surface waters globally and that ozone readily dissolves in water, we should consider the
potential of atmospheric ozone as an aquatic pollutant.
Heterotrophic bacteria are primary constituents of freshwater plankton and benthos,
where they contribute to the base of the food web and interact complexly with algae (Wetzel
2001a). They are integral to the aquatic carbon cycle. Changes in the quantity and health of
aquatic bacteria may greatly affect ecological functions from the biofilm to the ecosystem scale.
Wastewater treatment and food preservation studies have repeatedly shown the antibacterial
potential of high concentrations of dissolved ozone (far exceeding atmospheric levels) to both
Gram positive and Gram negative bacteria (Tanner et al. 2004, Voidarou et al. 2007, de
Velasquez et al. 2008, Gerrity et al. 2011, Izdebski et al. 2011). Primary damage results from
degradation of lipids in the cell wall and membrane, particularly the double bonds of unsaturated
lipids, resulting in membrane disruption, leakage of cytoplasm and eventually cell lysis (GuzelSeydim et al. 2004). However, we have little knowledge of the direct or indirect effects of dissolved
ozone on aquatic bacteria in the ranges observed and expected under ambient and elevated atmospheric
ozone concentrations, and the degree to which observations from ozonation treatment of water can be
extrapolated to lower exposures is unknown.

54

Dissolved ozone may also cause indirect effects on bacterial biomass if the increased

oxidative stress affects other biota in the ecosystem. Heterotrophic bacteria and algae form tight
relationships in the plankton and benthos that may be either facilitative or competitive (Wetzel
2001d). Whether the relationship is negative or positive largely depends upon the supply of
inorganic nutrients and organic carbon available for bacterial use. Natural bacterial communities
are commonly limited by phosphorus and/or organic carbon concentrations and, since they are
much better competitors for phosphate than algae, they are expected to outcompete algae in low
phosphorus situations leading to a decline in algal biomass as bacterial biomass increases
(Jansson et al. 1996). However, algae and bacteria commonly coexist and their relationship is
often found to be mutually positive. It is widely accepted that algae provide, through exudates,
labile organic carbon for use by bacteria and will stimulate bacterial growth when phosphorus is
not limiting (Klug 2005, Rier, 2002). Benthic algal mats may also provide greater physical
substrate for bacterial colonization, facilitate internal nutrient cycling (Rier and Stevenson 2002),
and act as protection for the bacterial communities. These bacterial/algal interactions will be
affected by any environmental variables that alter algal biomass and composition or the
availability of organic carbon. My research in Chapter 2 showed a change in both biomass and
species composition of algal communities after exposure to high ozone atmospheres
Dissolved organic carbon (DOC) may also mediate the effect of ozone on bacterial
communities. Most DOC in the water column is composed of large, complex, and more
recalcitrant substances, such as humic and fulvic acids as well as lignocelluloses (Wetzel 2001a).
These are utilized slowly by bacteria and therefore accumulate in the water column; whereas
small and simple organic molecules, like amino and fatty acids and simple carbohydrates, have a
very high turnover rate because they are preferentially taken up by bacteria (Moran and Hodson

55

1990, Ellis et al. 1999, Wetzel 2001a, Rosenstock and Simon 2003). Oxidation of large,
complex DOC molecules by UV-B and ozone result in production of smaller more labile
molecules such as aldehydes, ketones, fatty acids, and amino acids (Mann and Wetzel 1995,
Moran and Zepp 1997, Volk et al. 1997, Escobar et al. 2001, Judd et al. 2007). An increase in
bioavailable molecules may lead to greater bacterial metabolism and biomass. Indeed, many
studies have shown the stimulation of bacterial growth after photooxidation of humic substances
(Engelhaupt et al. 2003, Middelboe and Lundsgaard 2003, Anesio et al. 2004, Lennon 2004,
Romani et al. 2004). Oxidation of humic substances can be an important sink for ozone (Cho et
al. 2003). If the concentration of humic substances is sufficiently high, oxidation of humic
substances may ameliorate the oxidative potential of ozone and protect aquatic biota from much
of the direct oxidative stress.
The objective of this study was to examine the direct and indirect effects of dissolved
atmospheric ozone on aquatic heterotrophic bacteria. My study consisted of two experiments,
the first being a “bacteria only” study in which bacteria were grown in the absence of algae, in
three different ozone atmospheres. This experiment was intended to discern direct effects of
ozone concentration. The second was a “periphyton and bacteria experiment” in which natural
algal/bacterial assemblages were grown in the same experimental ozone conditions as the
previous experiment with the additional variable of DOC concentration. This experiment was
meant to simulate a more natural system in which competition with or facilitation by algae may
influence bacterial response to ozone treatment. I hypothesized that high and medium ozone
levels would reduce bacterial biomass in the absence of algae due to oxidative stress. In the
bacterial and algal assemblages, I expect indirect ozone effects, such as a facilitation of internal

56

organic matter and nutrient recycling through the ozone-mediated breakdown of periphyton
mucilage or altered algal/bacterial interactions.

Methods
Bacteria only experiment
Three covered, ten-gallon glass aquaria were used to create the treatment chambers, each
having a different atmospheric ozone concentration: a control treatment with 0 ppb O3; a
medium treatment with about 80 ppb O3; and a high treatment with about 250 ppb O3 (Fig. 3.1) .
Each chamber was completely covered in reflective paper to exclude ambient light and thus
prevent algal growth. Aquarium lids were sealed with weather-stripping and lined with PTFE
sheeting to maintain the treatment atmospheres. Chemically non-reactive PTFE and PFA
plastics were required for this experiment to resist corrosion by the large ozone concentrations
pumped into the experimental chambers. Ozone was created and measured using a Thermo
Scientific Model 49™ O3 analyzer. The air used to generate the ozone was ambient air filtered
through a carbon filtration unit prior to entering the generator. PFA tubing conducted the ozone
from the generator into two of the aquaria and back from all three to the sample analyzer. Flow
into each aquarium was controlled using a PTFE ball valve and manipulated until the target
ozone concentration was measured in the chamber air. The control treatment was supplied with
air from an aquarium bubbler filtered through another carbon filtration device. Ozone in each
chamber was monitored and manipulated daily to insure an atmospheric concentration above the
microcosms that was consistently ± 10 ppb of the target concentration. Ozone levels in the

57

control never measured above 0 ppb. All three chambers were placed on an Eberbach EL600™
Orbital-Reciprocal Shaker Table to create water motion, set to 96 rotations per minute.
Water samples were taken from Vermillion Creek, Rose Lake Wildlife Area, Clinton
County, MI. This supply water served as the growth medium and provided the bacterial seed
populations for the ozone experiments. The DOC of the source water was 19.7 mg C/L. Before
ozone treatment, the inoculum of bacteria was created by filtering the supply water through a
three µm mesh plankton net to remove zooplankton, protozoans, and most of the existing algal
community.
Eight 4 inch diameter ceramic dishes were used as microcosms. Each dish contained five
clean 2.5 x 2.5 cm unglazed ceramic tiles. 150 mL of the filtered source water were added to
each microcosm and bacteria grew for two weeks before sampling. . To simulate natural,
summer, diurnal shifts, ozone generation was on a timer, producing ozone from 7:00 a.m. to 9:00
p.m. daily. Water temperature stayed at a constant 23-24°C for the duration of the experiment.

58

Air
Ozone
Generator/Monitor

Carbon Filter

250 ppb Ozone

80 ppb Ozone

0 ppb Ozone

High DOC
Air Pump

Carbon Filter

Low DOC

Fig. 3.1. Experimental set-up. Rectangles represent individual chambers and circles represent
microcosms within each chamber. *The high and low DOC water treatments only pertain to the
periphyton/bacteria experiment.
At the end of the experiment, all tiles and the entire water volume were sampled.
Bacteria were removed from each tile using a razor blade and rinsed into 20 mL of DI water.
Samples from individual tiles from a microcosm were pooled. Bacterial films were

59

®

homogenized using a Biospec M 133/1281-0 2-speed biohomogenizer, preserved with
gluteraldehyde (2%), and refrigerated until counted.
Water samples were filtered through Whatman™ glass fiber filters, type GF/F, for
analysis of dissolved organic carbon, total dissolved nitrogen (TDN), and soluble reactive
phosphorus (SRP). All water samples were frozen until analysis.
Non-purgeable organic carbon (NPOC) and total nitrogen concentrations were measured
in filtered water samples by Dr. Stephen Hamilton’s lab at Michigan State University using a
Shimadzu TOC-VCPH™ Carbon Analyzer with a total nitrogen module (TNM-1) and an ASIV Autosampler. Because the water samples were filtered prior to analyses, the TN reading did
not include particulate nitrogen and were a measure of total dissolved nitrogen (TDN).
Soluble reactive phosphorus (SRP) was analyzed on frozen samples. Due to a freezer
malfunction, filtered water samples from week 2 were lost; therefore, SRP measurements for the
week 2 samples were done on a set of samples filtered post-freezing. SRP concentration was
measured manually following the ascorbic acid method (APHA 1998) on a Spectronic®
Genesys™ 2 Spectrophotometer by Spectronic Instruments.
Bacterial abundance was estimated by direct cell counts at 1000x magnification on a
Nikon Eclipse™ E800 epifluorescent microscope using 4’, 6’ –diamidino-2-phenylindole
(DAPI) dye. Samples were variously diluted, to a density of 30-300 bacteria in the viewing grid,
in a phosphate buffer solution (PBS) to a final volume of 5 mL. Forty microliters of 0.25 mg/mL
DAPI stock solution was then added to each sample/PBS mixture and allowed to stain cell DNA
for 6 minutes. Using a Millipore™ filtration system 2 mL of PBS and then 1 mL of stained

60

sample were filtered onto a 0.2µm black, membrane filter with a white grid. The membrane
filter was then mounted on a microscope slide using non-fluorescent immersion oil. All bacteria
in a viewing grid were enumerated to a total of 20 grids, dispersed over 5 transects.
Periphyton and Bacteria Experiments
The physical set-up was the same as the previous experiment with the exception that
chambers were not covered with reflective paper and light was provided by fluorescent, natural2

spectrum bulbs mounted above the chambers. These lights provided 63.65 ± 2.16 µE/cm /sec,
which is similar to common conditions for periphyton in streams. To simulate natural, summer,
diurnal shifts, ozone generation and light were on a timer, producing ozone and light from 7:00
a.m. to 9:00 p.m. daily.
Water was taken from Vermillion Creek, Rose Lake Wildlife Area, Clinton County, MI
and an adjacent, backwater wetland for use in the experiment. Each source constituted a
different “DOC water treatment.” The two water bodies were physically connected by a beaver
dam that allowed continuous water exchange. However, each source differed greatly in the
concentrations of dissolved organic carbon (DOC).

Other primary drivers of algal growth, such

as pH and temperature, did not differ greatly between the two. Macronutrients (silica, nitrogen,
and phosphorus) were directly measured and addressed. The creek water served as the growth
medium for the “low DOC water” treatment and the wetland water served as growth medium for
the “high DOC water” treatment. Initial DOC concentration for the high DOC water was 25.1
mg C/L and that for the low DOC water was 14.152 mg C/L as nonpurgable organic carbon
(NPOC) with pH 7.4 and 8.2, respectively. The high DOC water was noticeably stained with
humic substances from leaf decomposition, and the low DOC water was visibly clear. These
DOC concentrations are both high by global standards but represent common levels for rivers in

61

our area. The DOC treatments of this experiment were approximately 5 mg C/L less or greater
than the 19.661 mg C/L in the bacteria only experiment. This difference in DOC between the
two experiments was due to seasonal changes in the source water, making direct replication
impossible
Benthic and epiphytic algae were collected by scrubbing biofilms from numerous
substrates in each habitat. These samples were then combined into one suspension that was used
as the seed population for both DOC water treatments. Half of this composite algae sample was
added to the low DOC source water and the other have to the high DOC source water before
commencement of the experiment. Therefore the seed populations for both experiments were
very much the same, differing only in the suspended algal cells that may have been present
during water sampling.
150 mL of each source water and an inoculum of algae and bacteria from the composite
sample were added to each microcosm. Four of the eight microcosms in each of the three ozone
chambers (0, 80 ppb, and 250 ppb) were designated for either low or high DOC water treatments
th

th

st

(Fig. 3.1). Fresh source water was added to the microcosms on the 7 , 14 , and 21 days of the
growth period, to refresh nutrients and keep the water level constant. Water temperature was
maintained at a constant 23-24°C for the four-week duration of the experiment.
Periphyton and water samples were collected after two and four weeks of treatment. The
first sampling date was determined by the appearance of a visible algal film on the tiles.
Periphyton mats and water samples were collected after two and four weeks of treatment. Two
tiles were sampled from each microcosm during the first sampling period and the remaining
three tiles were sampled at week 4. Periphyton were removed from each tile using a toothbrush
and deionized (DI) water from a squirt bottle. Periphyton samples from individual tiles were

62

pooled by microcosm and date. Subsamples were taken from the homogenized, composite
sample for algal biovolume estimation and bacterial counts. Algal samples were preserved with
gluteraldehyde (2%).
Algal cells were identified and counted in a Palmer-Maloney counting chamber at 400X
until 400 natural units were counted, which is a modification of a standard method for assessing
taxonomic composition and cell density of algal samples used in national ecological assessments
(Lowe and LaLiberte 1996). I modified the counting method by increasing the number of natural
units counted to increase precision of taxonomic and cell density measurements. Natural units
are defined as the normal growth form of the alga, such as single cells, filaments, or colonies.
The number of cells composing each natural unit was also recorded. Measurements were made
of each taxon during identification for calculating total biovolume as an estimate of algal
biomass.
Bacterial cell enumeration and water sample processing for this experiment were done as
described in the previous experiment.
Statistical Analysis
A one-way ANOVA was used to determine if ozone treatment had an effect on the
density of the bacteria-only assemblages or measured nutrient concentrations. To test whether
DOC, ozone, and sampling date (week) had an effect on total algal biovolume, bacterial density,
or the ratio of bacteria biomass to algal biomass in the periphyton/bacteria experiment, I
conducted three-way ANOVAs using the GLM function in SYSTAT Version 12.0. To better
understand the relationships between independent and dependent variables, two-way ANOVAs
were conducted on the ozone/DOC interaction at each week. Two-way ANOVAs on ozone and
week at each DOC water treatment were also done to indicate significant changes from week 2 to

63

week 4 and to identify any significant relationships among the dependent and independent
variables that were not found in the former two-way ANOVAs.
Tukey’s HSD tests were used for pair-wise comparisons of independent and interaction
effects. All continuous variables were log-transformed prior to analysis. Because of issues with
non-normality and some unequal variances, the ANOVA results from the periphyton/bacteria
experiment were corroborated by the non-parametric Kruskal-Wallis One-way Analysis of
Variance (KW). Results did not significantly differ between the parametric and nonparametric
tests.
To determine if nutrient concentrations differed as a function of ozone, DOC water, or
week, I once again used a three-way ANOVA with subsequent Tukey’s HSD pairwise
comparisons. Nutrient data were log-transformed.
Correlations were used to identify significant bivariate relationships between measured
nutrient concentrations and algal or bacterial biomass. Correlations were also used to determine
if bacterial density was correlated with algal biomass.

Results
Bacteria only experiment
Ozone had a positive effect on bacterial biomass in the absence of algae. Bacterial
2

density (cells/cm ) was greater in the high and medium ozone than in the zero ozone treatment
2

(F2,21=5.367, p=0.071, p=0.012 respectively) (Fig. 3.2). Although the mean (20509 cells/cm )
2

and maximum (36569 cells/cm ) bacterial densities in the high ozone treatment were lower than
2

in the medium ozone treatment (Mean: 26613; Max: 44839 cells/cm ), this difference was not
statistically significant.
64

Bacterial films on many tiles consisted of a strong, rubbery, polysaccharide matrix that
maintained its integrity after removal from the tile. Although not statistically verifiable, this
thick biofilm appeared to be more prevalent in the ozone treatments than the control. Five of
eight microcosms in the high and medium ozone treatments contained this matrix while only
three of eight had it in the zero ozone treatment. Also, the number of tiles within each
microcosm containing this rubbery biofilm was greater in the high and medium ozone treatments
than in the zero ozone treatment.
60

Dissolved Organic Carbon (mg/L)

2
Bacterial Density (cells/cm )

50000
40000

b
30000

b
20000

a
10000
0

0

80
Ozone (ppb)

50

40

b
b

30

20

250

a

0

80
Ozone (ppb)

250

2

Fig. 3.2. Bacterial biomass as represented by cells/cm and DOC for each treatment in the
bacteria-only experiment. Lower-case letters denote significant differences. Bars are means of
eight replicates ± SE.
After two-week exposure to the ozone treatments, total DOC was greater in all three
treatments than in the inoculum. However, the DOC concentration was significantly lower in
high and medium ozone than in the zero ozone treatment (F2,21=11.345, p=0.000, p=0.011

65

respectively) (Fig. 3.2). There was no observable color difference between the treatment water
and the control at the end of the experiment.
Soluble reactive phosphorus (control=0.016 ± 0.016; medium=0.012± 0.012; high=0.014
± 0.007 mg/L) did not statistically differ among treatments. The variance was very high for
these measurements. Total dissolved nitrogen was greater in the control than in the high and
medium ozone treatments (F2,21=5.147, p=0.018, p=0.055 respectively), although the difference
was slight (control: 3.592 ± 0.667 mg/L; medium: 3.023 ± 0.322 mg/L, and high: 2.899 ± 0.300
mg/L). Bacterial abundance was not correlated with either SRP or TDN.
Periphyton and bacteria experiment
DOC treatment, ozone, and week each independently affected measured DOC
2

concentration (r =0.869) (Fig. 3.3). The concentration of total DOC increased from week 2 to
week 4 in all samples (F1,34=29.133; p=0.000), however, it was higher in the high DOC than in
the low DOC source water at both sampling dates (F1,34=170.498; p=0.000). Ozone had a
negative effect on total DOC as DOC was lower in both medium and high ozone treatments than
in the zero ozone treatment (F2,34=8.099; p=0.002 and p=0.008, respectively). These differences
were also visible in the water color. Over the first two weeks of the experiment, in the high
DOC water, water color decreased greatly in the high and medium ozone treatments but not in
the zero ozone treatment.
Ozone, DOC water, and week complexly affected total algal biomass, with all interaction
effects being significant. At week 2 and week 4, the effect of ozone on total biovolume was

66

2

2

dependent on DOC water treatment (r =0.883; F2,18= 3.553, p=0.05 and r =0.632; F2,18=
10.884, p=0.001, respectively).

High DOC

Low DOC

50
50

DOC (mg/L)
DOC (mg/L)

43
36
36

29

a
a
b
b
b
b

OZONE
0 ppb
80 ppb
250 ppb

22
22

15
15

2

4

2

Week
WEEK

4
4
WEEK
Week

Fig. 3.3 Measured DOC concentration by ozone and week, for each DOC water treatment.
Lower-cased letters indicate significant differences between ozone treatments at Week 2. Uppercase letters, differences at Week 4. An asterisk denotes a significant change from Week 2 to
Week 4. Error bars represent one standard error above and below the mean.
In the absence of ozone, DOC had a negative effect on total algal biomass over the entire
study period as total biovolume was three times greater in low DOC water at week 2 and five
times greater at week 4 (p=0.002 for both weeks) (Fig. 3.4). In the medium and high ozone,
DOC initially had a negative effect on total algal biovolume but a positive effect later. At week
2 in the high DOC water, total biovolume was six times lower in high ozone (p=0.000) and three
times lower in the medium ozone (p= 0.006). However, due to a three-fold increase in biomass
in the high and medium ozone treatments in the high DOC water but not in the low DOC water,
by week 4 biovolume did not statistically differ between high and low DOC water.

67

Ozone effect on algal biomass in the low DOC water was negative but limited to the
medium ozone treatment and was only significant at the first sampling date (Fig. 3.4).
Biovolume was three times lower in medium ozone than in either the zero or high ozone
treatments (p=0.004 and p=0.002, respectively) at week 2. Between sampling times, algal
biomass decreased in zero and high ozone but not in medium ozone. Biovolume significantly
decreased with high ozone over this period (p=0.005), but did not significantly change in the
medium or zero ozone treatments.
In high DOC water, ozone had a negative effect on total algal biomass early in the study
period but a positive effect later (Fig. 3.4). At week 2, total biovolume was greater in zero ozone
than in medium ozone (p=0.011) and also high ozone, although not significantly (p=0.129). As
mentioned above, from week 2 to week 4, total biovolume increased threefold in the high and
medium ozone treatments (p=0.019 and p=0.011, respectively) but decreased in zero ozone
(p=0.019). By week 4, the zero ozone assemblages had less biovolume than those in high and
medium ozone (p=0.001 and p=0.041, respectively).
Bacterial biomass was not significantly affected by DOC water treatment. Ozone had a
significant effect on bacterial biomass, as measured by cell density, but the direction of effect
2

was dependent on sampling date (r =0.502; F2,18= 9.367, p=0.001) (Fig. 3.4). In high DOC
water, medium ozone initially had a negative effect on bacterial biomass. At week 2, bacterial
density in the medium ozone treatment was roughly half that in either the high or zero ozone
treatments (p=0.002 and p=0.04, respectively). During the next two weeks, bacteria in the high
2

and medium ozone increased, but only significantly in the medium ozone (r =0.385;
F2,18=13.844, p=0.014), and by week 4 bacterial density did not significantly differ among the

68

ozone treatments. The bacteria:algae ratio was greater in the ozone treatments at week 2
(F2,18=6.916, p=0.015) and lower at week 4 (F2,18=7.964, p=0.012). This decrease in
bacteria:algae ratio coincided with a change in dominant algal taxa from diatoms to cyanophytes
in the ozone treatments (Chapter 2).
In the low DOC water, at week 2, ozone had a positive effect on bacterial biomass as
bacterial density was approximately twice as great in the high ozone treatment than either the
2

medium or zero ozone treatments (r =0.724; F2,18=29.044, p=0.004 and p=0.000, respectively).
Bacteria numbers then declined precipitously in high ozone (F2,18=22.357, p=0.002), but did not
significantly change in zero or medium ozone. Consequently, at week 4, bacterial biomass did
not statistically differ among treatments and the control. The bacteria:algae ratio in the high and
medium ozone was roughly twice that in the zero ozone treatment at week 2 (F2,18=3.925,
p=0.05) and at 2-3 times greater at week 4 (F2,18=6.832, p=0.016). Total algal biovolume and
bacterial density were significantly positively correlated in both high and low DOC water (r=
0.0.647, p=0.001 and r=0.454, p=0.026, respectively).

69

Low DOC

0.65
0.65
0.56
0.56
0.47
0.47

a
a
a
a

0.38
0.38
0.29
0.29

OZONE

0.20
0.20
0.11
0.11
0.02
0.02

a
a
b
b
b
b

b
b
*

*

A
A
2
2

Week
WEEK

4
4

2
2

Week
WEEK

0 ppb
80 ppb
250 ppb

4
4

*
*

95000
95000

aa

75000
75000
55000
55000
35000
35000

aa
aa

15000
15000

2
2

1000000
1000000

Week
WEEK

2

4

Week
WEEK

OZONE
0 ppb
80 ppb
250 ppb

4

*B
*B

aa
aa
bb

A
A

*A
*A

A
A

100000
100000

*
*

bb
bb

bb

3

Bacterial cells/Algal Biovolume (mmmm
biovolume3)
Bacterial Cells / Algal

*B
*B
*B
*B

115000
115000

Bacterial Density (cells/cm)

2

Bacterial Density (cells/cm )
2

3

2

Total Total Algal Biovolume (mmcm2) /cm )
Algal Biovolume (mm
3

High DOC

A
A

aa
aa

BB

bb
2

Week
WEEK

4

2

Week
WEEK

OZONE
0 ppb
80 ppb
250 ppb

4

Fig. 3.4 Measures of total algal biovolume, bacterial cell density, and ratio of bacterial density to
algal density by ozone and week, for each DOC water treatment. Lower-case letters indicate
significant differences between ozone treatments at week 2. Upper-case letters, differences at
week 4. An asterisk denotes a significant biomass change from week 2 to week 4. Error bars
represent one standard error above and below the mean.

70

DOC water, ozone, and week were all independently significant factors affecting TDN
2

(r =0.696) (Fig. 3.5). Total dissolved nitrogen was lower in the low DOC water treatment
(F1,34=21.710, p=0.000) than in the high DOC across ozone treatments. It was also lower in the
medium ozone than in the high or zero ozone treatments (F2,34=13.099, p=0.000 both) across
DOC water treatments. From week 2 to week 4, TDN increased across DOC water and ozone
treatments (F1,34=8.377, p=0.007).
Soluble reactive phosphorus concentrations were affected by DOC water and ozone
2

treatment but the effects varied with week (r =0.724) (Fig. 3.5). At week 2 there was no
statistical difference in SRP concentration among the treatments. However from week 2 to week
4, SRP in the high and medium ozone decreased significantly (F2,31=4.580; p=0.001 and
p=0.006, respectively) and at week 4 SRP in the zero ozone was higher than both (p=0.018 and
p=0.05, respectively). Also, at week 4, SRP was lower in the high DOC than in the low DOC
water (F2,31=3.973; p=0.055).
Total algal biovolume was not correlated with nutrients across treatments and sampling
periods. Bacterial density was positively correlated with TDN (r=0.36; p=0.014) across
treatment variables. However, when examined within DOC treatment, this correlation was not
significant.

71

High DOC

Low DOC

TDN (mg/L)

TDN (mg/L)

3.0
3.0

2.3
2.3
aa
aa

1.6
1.6

OZONE
*
*

0.9
0.9

bb
2
2

Week
WEEK

4
4

2
2

Week
WEEK

4
4

0 ppb
80 ppb
250 ppb

1.000
1.000

A

SRP (mg/L)

SRP (mg/L)

A

0.100
0.100

A,B

A ,B

0.001
0.001

OZONE

B

B

0.010
0.010

*

*

2
2

Week
WEEK

4
4

2
2

Week
WEEK

0 ppb
80 ppb
250 ppb

4
4

Fig. 3.5 TDN and SRP concentrations (mg/L) by ozone and week, for each DOC treatment. Lowercased letters indicate significant differences between ozone treatments at week 2; upper-case letters,
differences at week 4. An asterisk denotes a significant biovolume change from week 2 to week 4. Error

bars represent one standard error above and below the mean.

Discussion
My experimental manipulation of ozone, DOC water, and light indicated that
atmospheric ozone, in concentrations measured in nature, has a significant effect on benthic
bacterial biomass, but the effect is primarily indirect and that bacterial response may depend on
community interactions. I did not see the hypothesized negative effects of oxidative stress. In
the dark experiment, in the absence of algae, the ozone treatment stimulated bacterial production,
72

resulting in higher densities in both the medium and high ozone treatments than the control. No
published evidence provides a mechanism for sustained stimulation of bacterial metabolism due
to direct oxidative exposure, so I infer that the recorded response is due to indirect effects.
Bacterial production is most often limited by phosphorus, organic carbon, or a
combination of the two and less often, by nitrogen (Tranvik 1988, Balogh and Voros 1997,
Thomas 1997, Bergstrom and Jansson 2000, Drakare et al. 2003, Lennon 2004). In this study,
inorganic nutrients were not likely to be secondary factors leading to the higher bacterial density
in the ozone treatments. Nitrogen was not limiting as TDN concentrations were above an
accepted limiting value of 00.05-.06 mg/L (Wetzel 2001e) and TDN was not statistically
correlated with bacterial density. Also, the higher concentrations were in the control samples
which had the lowest bacterial density. Phosphorus concentrations in this experiment were at or
near limiting for bacterial growth. Wetzel (2001e) lists the upper end of phosphorus limitation as
< 0.015 mg/L, however, SRP concentrations did not statistically differ among treatments, were
not correlated with bacterial density and were greater in the control.
Organic carbon limitation of these bacterial communities is likely. As mentioned in the
introduction, the oxidation of humic substances generally leads to the creation of smaller, more
labile compounds. This increase in preferred carbon sources may lead to increased bacterial
production. The greater numbers of bacteria in the ozone treatments agree with the many studies
that have shown a stimulation of carbon uptake by bacteria, leading to a higher growth rate, after
UV breakdown of recalcitrant humic molecules (Stewart and Wetzel 1981, Lindell et al. 1995,
Moran and Zepp 1997, Engelhaupt et al. 2003, Middelboe and Lundsgaard 2003, Rosenstock and
Simon 2003, Anesio et al. 2004, Lennon 2004, Rosenstock et al. 2005). Even though the source
water was not particularly rich in humic substances, the total DOC concentration at the end of

73

the experimental run was lower in the ozone-treated waters, indicating a breakdown and loss of
organic carbon. This supports the hypothesis that ozone treatment in our experiment stimulated
bacterial growth by oxidizing humic substances to more bioavailable compounds. However,
there was no statistical correlation between DOC concentration and bacterial density.
Internal biofilm recycling of phosphorus or carbon may be a more important
consideration within these communities. Some of these bacterial communities formed very
thick, gelatinous matrices that remained intact even when removed from the tiles. The existence
of these thick matrices was generally associated with higher bacterial densities and, although not
statistically testable, these biofilms were also thicker in the ozone treatments. It is possible that
greater mucilage production per cell was a defensive response to oxidative stress occurring in the
treatments (Reynolds 2007). These more tightly-associated communities formed by greater
production of extracellular polysaccharides would lead to greater internal recycling of carbon
and inorganic nutrients and also provide greater surface area for bacterial growth.
Although not statistically significant, bacterial numbers were lower in the high ozone
treatment than the medium ozone treatment. This may suggest a unimodal response of bacterial
growth to atmospheric ozone. I predict that as ozone increases past a certain point, some of the
benefits of increased production of labile organic carbon are lost. As the concentration of
oxidants increases there is a greater potential for direct interaction with bacterial cell walls
leading to changes in permeability and reducing cell function.
In the periphyton and bacteria experiment, in both DOC water treatments, bacterial
biomass was highly correlated with algal biomass. Bacterial density rose and fell with algal
biovolume over the course of the experiments. As mentioned previously, heterotrophic bacteria
preferentially use smaller, simpler organic molecules. In benthic and plankton communities,

74

many of these molecules are produced extracellularly by algae or are released from algal cells
due to membrane damage or death. These are ready sources of carbon substrates for bacterial
metabolism and are almost immediately assimilated. Indeed, many studies have shown a
positive correlation between algal and bacterial production when inorganic nutrients are not
limiting. Algae are hypothesized to be a very important source of organic compounds for
heterotrophic bacterial metabolism in many aquatic systems (Bird and Kalff 1984, Chrzanowski
1985, Sundh and Bell 1992, Wetzel 2001a, Romani et al. 2004).
The relationship between bacteria and algae was not equal among treatments, however.
At week 2, when significant differences in bacterial numbers existed among treatments, the
bacteria to algae biomass ratio was significantly greater in the ozone treatments. This
differential effect of algae on bacterial growth explains why, in the low DOC water experiment,
there was significantly higher bacterial biomass in high ozone than in the control even though
algal biomass was not greater and why, in the high DOC water experiment, there was
comparable bacterial production in the high ozone and the control even though algal biomass
was significantly higher in the control. More bacterial biomass may be supported per volume of
algae if each algal cell is contributing more nutrients to the bacterial community. As with
bacterial cells, reactive oxygen forms will interact with algal cell wall constituents potentially
resulting in increased mucilage production as a defense (Reynolds 2007) or a change of cell
permeability (Mallick and Mohn 2000) and leakage of labile organic carbon and other nutrients.
Either of these scenarios would provide greater resources for heterotrophic bacterial growth.
The greater number of bacteria in high ozone and high DOC water may also be explained by a
higher concentration of SRP, although there was no significant correlation between SRP and
bacterial density. In the periphyton mat, algal interactions are of greater importance to bacterial

75

numbers than external stressors and ozone had an indirect effect on bacterial biomass through
effects on algal biomass.
Direct negative effects of oxidative stress caused by ozone dissolved from the atmosphere
appear to be minimal for benthic, heterotrophic bacteria. In fact, ozone treatments stimulated the
growth of bacterial communities whether associated with algae or not. Most ozone effects on
benthic bacteria are likely indirect through effects on carbon cycling either by breaking down
recalcitrant molecules, making them more bioavailable, or by causing increased autochthonous
carbon production which is then used by the heterotrophic bacteria. This carbon mediation was
indicated in my previous study of the interactive effects of ozone and DOC water on different
algal taxa. Ozone initiated changes to the aquatic carbon cycle may have wide-ranging effects.
Further studies are needed to verify the mechanism of ozone effects and whether other aquatic
groups, particularly the plankton, will respond differently than the benthos.

Acknowledgements
I would like to thank Angeline Kosnik for her work on the bacterial counts.

76

Chapter 4: The effects of atmospheric ozone concentration on benthic cyanophyte and
diatom communities.
Abstract.
High concentrations of ozone have been used to kill algal constituents in water used for
drinking and recreation. In such treatments, algal taxa have been shown to respond differentially
to ozone treatment. No studies published to date have investigated the effects of atmospheric
levels of ozone, allowed to dissolve naturally into the water column, on algal groups. My
previous studies have indicated that ozone effects in this situation are highly dependent on algal
taxonomic division and complicated by interspecies interactions and water chemistry. To test for
direct effects of atmospheric ozone on algal groups, I grew cyanophytes and diatoms,
independently, and with an abundance of nutrients, in three atmospheric ozone concentrations.
In the absence of interactive effects, cyanophytes were negatively affected by ozone
concentration, as indicated by reduced total biomass, and diatoms were unaffected. This study
indicates that, in the absence of strong indirect effects, algal assemblages may undergo a shift in
composition caused by a greater suppression of cyanobacterial than diatom growth.

Introduction
Local production, regional and longer distance transport, and high background levels lead
to a global ozone pollution problem (Chameides et al. 1997; Granier and Brasseur 2003).
Models have estimated a 100-120% increase in ozone in the northern mid-latitudes over the past
200 years and an increase of 80-100% in the tropics (Committee on Tropospheric Ozone 1991).
All but one scenario proposed by the Intergovernmental Panel on Climate Change (IPCC) project
st

increases in tropospheric ozone during the 21 century with projected background

77

concentrations of 35-48 ppb by 2040, 38-71 ppb by 2060, and 42-84 ppb by 2100 (Vingarzan
2004). These models predict that background concentrations may exceed internationally
accepted criteria established to protect animal and plant health.
Increases in atmospheric ozone have led to significant acute and chronic effects on
terrestrial primary producers. A chronic response to ozone pollution appears to be a general
reduction of growth rate due to a decrease in photosynthesis. The acute, physiological changes
associated with ozone stress start with a loss of membrane integrity as unsaturated lipids and
proteins are attacked (Skarby et al. 2004). As reactive oxygen species (ROS) increase and
overwhelm the cell’s ability to compensate, intracellular damage occurs (Kollner and Krause
2003; Oksanen 2003; Skarby et al. 2004; Yamaji et al. 2003). This cellular damage results from
the oxidative stress caused directly by ozone molecules but also from other reactive oxygen
species. These oxygen compounds result from ozone speciation into other reactive oxygen
molecules, cellular production induced by the initial stress, or are secondary products resulting
from the oxidation of membranes (Oksanen et al. 2004; Schraudner et al. 1997).

As

photosynthetic cells are naturally high in ROS they are especially susceptible to oxidative
overload and the photosynthetic apparatus can readily suffer injury (Prozherina et al. 2003).
Aquatic primary producers may be susceptible to similar ozone-induced damage.
Numerous wastewater treatment studies have shown the lethal effects of large amounts of
dissolved ozone on algal and bacterial cells when pumped directly into the water. At large
enough concentrations, no microbial cells survive, but lethal levels vary widely among taxa
(Betzer et al. 1980; Gavand et al. 2007; Huang et al. 2006; Widrig et al. 1996; Yun et al. 1997).
These studies have all been done with the aim of sanitizing waters and have utilized ozone
concentrations in the parts-per-million range. They do little to address atmospheric ozone as an

78

aquatic pollutant in natural settings. Other reactive oxygen species have been examined at lower
concentrations, in ecological and physiological studies (Abd El-Baky et al. 2009; Choo et al.
2005; Gorbi et al. 2006; Ledford and Niyogi 2005; Mallick and Mohn 2000). Research has
shown that externally introduced hydrogen peroxide (H2O2), a potential secondary ROS created
from dissolved ozone, can cause negative effects on algal populations at natural or near natural
levels and different algal taxa show widely varying sensitivities (Barroin and Feuillade 1986;
Drabkova et al. 2007; Kay et al. 1984). Growth rates and pigment concentrations of
cyanobacterial cells are negatively affected at much lower (up to ten times lower) H2O2
concentrations than other algal groups, particularly green algae (Drabkova et al. 2007b). This
may give us insight into the effects of dissolved ozone at natural levels. However, due to the
specificity of antioxidant enzymes we cannot assume the same outcome for oxidation by ozone
and its breakdown products.
In addition to the potential for direct oxidative stress, there are several possible indirect
effects of ozone on algal communities. Changes in competition, disease vectors, light, and
nutrients may all be facilitated by ozone. The research done for Chapter 2, studying the effects
of atmospheric ozone and DOC concentrations (selected to represent real-world exposures to
aquatic ecosystems) on multi-division periphyton assemblages, found complex independent and
interactive effects on algal biomass that varied by algal division. These results indicated that
inter-taxa interactions and the chemical constituents of the environment may mitigate or change
the effects of ozone stress on algal populations.
The following study was devised to resolve some of these complex interactive effects of
ozone on algal communities by looking at single division assemblages in the absence of
inorganic nutrient limitation and high concentrations of humic substances. Cyanophytes and

79

diatoms were grown independently in three different ozone atmospheres and high levels of
nutrients. The results add to the general body of knowledge on the effects of oxidants on benthic
communities by using sub-lethal concentrations of ozone. I hypothesized that, in the absence of
these interactions, algal biomass would be lower in the high ozone treatments and that
cyanophytes would be more sensitive to the oxidative stress than diatoms.

Methods
Experimental set-up common to both the cyanophyte and diatom experiments
Three ten-gallon glass aquaria were used to create treatment chambers, each having a different
atmospheric ozone concentration: a control treatment with 0 ppb O3; a medium treatment with
about 80 ppb O3; and a high treatment with about 250 ppb O3. Aquarium lids were sealed with
weather-stripping and lined with PTFE sheeting to maintain the treatment atmospheres.
Chemically non-reactive PTFE and PFA plastics were required for this experiment to resist
corrosion by the large ozone concentrations pumped into the experimental chambers. Ozone was
created and measured using a Thermo Scientific Model 49™ O3 analyzer. Ozone in each
chamber was monitored and manipulated daily to insure an atmospheric concentration above the
microcosms that was consistently ± 10 ppb of the target concentration. Ozone levels in the
control never measured above 0 ppb.
The air used to generate the ozone was ambient air filtered through a carbon filtration
unit prior to entering the generator. PFA tubing conducted the ozone from the generator into
two of the aquaria and back from all three to the sample analyzer. Flow into each aquarium was
controlled using a PTFE ball valve and manipulated until the target ozone concentration was

80

measured in the chamber air. The control treatment was supplied with air from an aquarium
bubbler filtered through another carbon filtration device. All three chambers were placed on an
Eberbach EL600™ Orbital-Reciprocal Shaker Table to create water motion, set to 96 rotations
per minute.
Light was provided by fluorescent, natural-spectrum bulbs mounted above the chambers,
2

which generated 63.65 ± 2.16 µE/cm /sec. To simulate natural, summer, diurnal cycles, ozone
generation and light were on a timer, producing from 7:00 a.m. to 9:00 p.m. daily. All three
chambers were placed on an Eberbach EL600™ Orbital-Reciprocal Shaker Table to create
water motion, set to 96 rotations per minute.
Cyanophyte experiment
Replicates were filled with 146 mL of water from Vermillion Creek, Rose Lake Wildlife
Area, Clinton County, MI that had been filtered through a 3 µm mesh plankton net to remove
most of the algae. Nutrients were supplemented by adding 2 mL of 5 mL/L Bold Modified Basal
Freshwater Nutrient Solution. The following cyanophytes, including two nitrogen-fixing taxa,
were purchased from Carolina Biological Supply Co.: Tolypothrix distorta cultured in a soilwater medium; and Oscillatoria sp., Lyngbya sp., and Anabaena affinis Lemmerman cultured in
Alga-Gro® Freshwater Medium. I combined half the volume from each specimen vial into a
composite seed population. Each replicate was inoculated with 2 mL of the homogenized seed
sample. Algae grew in the treatments for a total of 12 days and additional creek water was added
once during the incubation at day six to refresh nutrient supplies and maintain water levels.

81

Diatom experiment
A diatom inoculant was created by sampling from several diatom-dominated
communities growing in the lab. In each replicate, two mL of this inoculant was added to 148
mL of a 20 mL/L sterile medium of Bold Modified Basal Nutrient Solution supplemented with
0.06 g/L NaSiO3. A sterile medium was used instead of stream water to decrease the chance of a
cyanophyte or chlorophyte bloom during the duration of the study. Also, since diatoms are more
tolerant of low-light situations than other algal groups (Pillsbury and Lowe 1999), shade cloth
was used to encourage diatom growth over that of other algal groups. The communities
developed on the blank tiles for nine days, after which diatom dominance was verified under the
microscope, and then the ozone treatment was started. The diatom experiment was run for 11
days.
Sample Collection
At the end of each experiment, a significant growth was observed on the bottom of the
microcosms as well as on the tiles, so all periphyton and the total water volume from each cup
was sampled. Algae were scraped from individual tiles, using a toothbrush and DI water from a
squirt bottle, and added into the corresponding water sample which also contained the loose algal
biomass that grew off the tiles. Subsamples were taken from this homogenized, composite
sample for chlorophyll a measurement and analysis of soft algal and diatom community
composition. Algal samples were preserved with gluteraldehyde (2%). Water samples were
filtered through Whatman™ glass fiber filters, type GF/F, for analysis of TDN, soluble reactive
phosphorous (SRP), and silicate concentration. All water samples were frozen until analysis.

82

Analysis of periphyton samples
Cyanophyte cells were identified and counted in a Palmer-Maloney counting chamber at
400X until 300 natural units were counted (Lowe and Laliberte 1996). Natural units are defined
as the normal growth form of the alga, such as single cells, filaments, or colonies. The number
of cells composing each natural unit was also recorded. Measurements were made of each taxon
during identification for calculating total biovolume as an estimate of algal biomass.
Live and dead diatom species were identified from subsamples mounted on microscope
slides using the modification by Stevenson (1984) of Taft’s (1978) glucose mounting technique
at 1000X on a Leica compound microscope to a total of 600 live and dead cells.
Chlorophyll a concentrations were also used as a measure of total algal biomass. Algae
from each chlorophyll subsample were filtered onto Whatman™ GF/C glass fiber filters,
extracted overnight in ethanol, diluted if necessary, and analyzed fluorometrically before and
after acidification to correct for pheopigments (APHA 1998).
Non-purgeable organic carbon (NPOC) and total nitrogen concentrations were measured
in filtered water samples by Dr. Stephen Hamilton’s lab at Michigan State University using a
Shimadzu TOC-VCPH™ Carbon Analyzer with a total nitrogen module (TNM-1) and an ASIV Autosampler. Because the water samples were filtered prior to analyses, the TN reading did
not include particulate nitrogen and is thus representative of TDN.
Soluble reactive phosphorous (SRP) was analyzed on frozen samples. SRP concentration
was measured manually following the ascorbic acid method (APHA 1998) on a Spectronic®
Genesys™ 2 Spectrophotometer by Spectronic Instruments. Silicate concentration was
measured on a Skalar Auto Analyzer following the molybdosilicate method (APHA 1998).

83

Statistical Analysis
A one-way between-subjects ANOVA was conducted to compare the effect of ozone
treatment on algal biovolumes and DOC or nutrient concentrations. If the effect of ozone was
significant, a Tukey’s HSD post-hoc comparison was used to evaluate differences among ozone
treatments. All continuous variables were log-transformed prior to analysis. Relative abundance
data were square-root transformed. I used an alpha level of 0.05 as the significance criterion for
all tests.

Results
Cyanophyte experiment
TDN and silica concentrations did not vary significantly between control and the ozone
treatments (Table 4.1). SRP concentrations were significantly lower in the control than in either
of the ozone treatments (F2,21=9.244, p=0.002). TDN and SRP concentrations were very high in
all samples due to the growth medium.
Table 4.1. Average nutrient concentrations (mg/L) for the cyanophyte experiment, +/- standard
deviation. Letters denote statistically significant differences.
Nutrient (mg/L)

Control

Medium Ozone

High Ozone

TDN
SRP

9.001 ± 1.097
a
0.873 ± 0.195
4.935 ± 0.506

8.026 ± 0.986
b
1.220 ± 0.296
5.073 ± 0.380

8.039 ± 0.868
b
1.408 ± 0.244
4.651 ± 0.224

SiO2

Ozone treatment had a significantly negative effect on cyanobacterial growth, as
3

2

2

represented by both biovolume (µm /cm ) (F2,21=7.316, p=0.004) and chlorophyll a (mg/cm )
(F2,21=7.085, p=0.005) (Fig. 4.1). Tukey’s Post Hoc comparisons indicated that biovolume and

84

chlorophyll a content in the communities exposed to the high ozone treatment were almost half
that in either the medium treatment or the control (biovolume: p=0.030, p=0.004; chlorophyll a:
p=0.013and p=0.002, respectively). The communities in the medium (80 ppb) treatment and
control did not differ significantly from each other in either chlorophyll a or total algal

5

200000
Chlorophyll a (mg/cm2)

Total Cyanophyte Biovolume (cm3/cm2)

biovolume although biomass in the medium ozone treatment was lower.

150000

a

a

b
b

100000

b
b
50000

0

4

80
250
Ozone (ppb)

a
b

3

b
b

2
1
0

0

a
a

0

80
250
Ozone (ppb)

Fig. 4.1. Total biomass by ozone treatment in the cyanophyte experiment. Lowercase letters
denote statistically significant differences. Bars are means of eight replicates ± SE.
Biovolume of all cyanophyte genera, except for Tolypothrix, which was a limited
constituent of all samples, was lower in the high ozone treatment (Fig. 4.2). The biovolumes of
Anabaena and Oscillatoria were lower in the high ozone treatment than in both the medium
ozone treatment and the control (Anabaena: F2,21=8.220, p=0.02 and p=0.002, respectively;
Oscillatoria: F2,21=7.697, p=0.01 and p=0.006 ). Lyngbya biovolume in the high ozone
treatment was also lower than in the medium ozone treatment and the control, however, only

85

significantly for the control (F2,21=7.780, p=0.093 and p=0.003, respectively). Biovolume in
the medium ozone treatment was lower, though never significantly, than in the control for
Anabaena, Lyngbya, and Oscillatoria.

0.3

3
Lyngbya Biovolume (mm /cm2)

3
Anabaena Biovolume (mm /cm 2)

0.020

0.015

a
0.010

b
0.005

0.000

b

0

0

80

250

0.0004

a

3
Tolypothrix Biovolume (mm /cm 2)

3
Oscillatoria Biovolume (mm /cm2 )

b
0.1

Ozone(ppb)
Ozone (ppb)

a
b

0.05

0.00

a
a,b

0.0

80
250
Ozone(ppb)
(ppb)
Ozone

0.15

0.10

0.2

0

0.0003

0.0002

0.0001

0.0000

80
250
Ozone(ppb)
Ozone (ppb)

0

80

250

Ozone(ppb)
Ozone (ppb)

Fig. 4.2 Cyanophyte biovolume by genus as a function of ozone. Lowercase letters denote
statistically significant differences. Bars are means of eight replicates ± SE.

86

While total biovolume of the various cyanophyte genera differed significantly as a result
of ozone exposure, relative abundances of these taxa did not, keeping the community
composition consistent among ozone levels (Fig. 4.3). Lyngbya was the largest constituent of all
communities by biovolume, followed by Oscillatoria, and Anabaena. Tolypothrix was less than
0.05% of any sample.

Percent of assemblage

120
100
80
60
40
20
0

0

80
250
Ozone (ppb)

ANABAENA
LYNGBYA
OSCILLATORIA

Fig. 4.3. Relative abundance of cyanophyte genera by biovolume for each ozone treatment.
Tolypothrix was less than 1% of all assemblages. Bars are means of eight replicates.
Diatom experiment
TDN and SRP were very high in all samples, due to the growth medium, and did not
statistically differ among treatments (Table 4.2). SiO2 was drawn down to moderate levels but
did not reach limiting concentrations for diatom growth, and was not statistically different
between control and treatments.
87

Table 4.2 Average nutrient concentrations (mg/L) in the diatom experiment +/- standard
deviation.
Nutrient (mg/L)

Zero Ozone

Medium Ozone

High Ozone

TDN
SRP
SiO2

76.428 ± 4.699
72.609 ±10.952
5.541 ± 0.209

79.016 ± 10.683
65.157 ± 14.379
5.686 ± 0.299

75.755 ± 7.068
65.571 ± 11.433
5.597 ± 0.314

Diatom biomass did not significantly differ as a function of ozone treatment (Fig. 4.4).
2

Chlorophyll a ranged from 0.5 and 1.7 mg/cm for all samples. Chlorophyll averages for the
2

high, medium and control treatments were, respectively, 1.09, 1.06, and 1.25 µg/cm . Total
diatom biovolume did not statistically change with ozone treatment. Total biovolume ranged
7

8

3

from 1.85x10 - 1.02x10 µm /cm

2

.
2.0

0.10
0.09

C hlorophyll a (m g/cm2 )

3
T otal D iatom B iovolum e (m m /cm 2 )

0.11

0.08
0.07
0.06
0.05
0.04
0.03

1.5

1.0

0.5

0.02
0.01

0

80
Ozone (ppb)

0.0

250

0

80
Ozone (ppb)

250

Fig. 4.4. Total biomass by ozone treatment in the diatom experiment. There are no statistically
significant differences among treatments. Bars are means of eight replicates ± SE.

88

Individual species biovolumes also did not differ among the control and treatments. The
ratio of live:dead diatoms did not differ among treatments and control.
Nineteen diatom taxa were observed in the experiment (Table 4.3). No differences were
observed in relative or absolute abundances of diatom species. The largest community
constituents in all samples, by biovolume, were D. confervacea (average 64.2%±126), F. cf.
pelliculosa (average 11.6%±5.3), G. parvulum (average 8.9%±4.8), P. elginensis (average
6.2%±5.0), and P. frequentissimum (average 3.9%±2.2).

The live:dead diatom ratio did not

differ among treatments for any individual species.
Table 4.3 . Relative abundances of diatom taxa by ozone treatment (percentage of assemblage).
Taxon

0 ppb O3

80 ppb O3

Diadesmis confervacea Kützing
Fistulifera cf. pelliculosa (Brebissón ex Kützing)
Lange-Bertalot
Planothidium frequentissimum (Lange-Bertalot) LangeBertalot
Gomphonema parvulum Kützing

63.9 ± 14.5
53.1 ± 4.6

65.4 ± 14.8
53.8 ± 8.3

250 ppb
O3
62.5 ± 8.3
59.7 ± 7.9

7.0 ± 3.6

6.2 ± 3.3

6.6 ± 1.7

5.0 ± 2.5

4.9 ± 2.3

4.8 ± 2.0

Achnanthidium minutissimum (Kützing) Czarnecki

2.0 ± 0.7

5.0 ± 3.3

3.5 ± 1.7

Nitzschia inconspicua Grunow

2.4 ± 0.9

1.4 ± 0.8

1.7 ± 1.1

Nitzschia palea (Kützing) W. Smith

2.2 ± 1.0

1.3 ± 0.7

1.7 ± 1.4

Achnanthidium exiguum (Grunow) Czarnecki
Nitzschia frustulum (Kützing) Grunow
Placoneis elginensis (Gregory) Ralfs
Amphora pediculus (Kützing) Grunow
Fragilaria capucina Desmazieres
Navicula cryptotenella Krammer & Lange-Bertalot
Nitzschia linearis (Agardh ex W. Smith) W. Smith
Navicula schadei Krasske
Rhoicosphenia abbreviata (Agardh) Lange-Bertalot
Staurosira construens (Ehrenberg) Williams & Round
Synedra parasitica var. subconstricta (Grunow)
Hustedt

1.8 ± 0.9
1.6 ± 1.1
1.2 ± 0.7
<1
<1
<1
<1
<1
<1
<1
<1

1.5 ± 0.9
1.3 ± 0.6
1.0 ± 0.5
<1
<1
<1
<1
<1
<1
<1
<1

1.7 ± 1.2
1.5 ± 0.9
1.2 ± 0.7
<1
<1
<1
<1
<1
<1
<1
<1

89

Discussion
Few studies have assessed the effects of ozone at atmospheric concentrations, let alone,
the effects of the oxidant entering the water naturally through diffusion from the atmosphere.
The current study shows a significant but taxon-selective negative effect of atmospheric ozone
on algal communities, in the absence of competing algal groups and high concentration of humic
substances.
Cyanophytes were negatively affected by the ozone treatment. Total cyanophyte
biovolume and chlorophyll a were significantly lower in the high ozone than in the medium
ozone treatment and the control. Community composition did not differ, however, as all major
constituents, Lyngbya, Oscillatoria, and Anabaena, responded in the same way. Although
statistically not significant, there is an apparent linear cyanophyte biovolume response to ozone
in the medium treatment that was intermediate between that in the high ozone and the control for
all major community constituents.
Diatom biomass, unlike that of the cyanophytes, was not affected by ozone treatment at
any taxonomic level. Total biovolume, chlorophyll a, and species biovolumes were statistically
the same among both treatments and the control.
The negative cyanophyte response is likely due to the direct oxidative stress caused by
the dissolution of atmospheric ozone. Light and temperature did not differ among treatments, no
other algal groups were competing for resources, and nutrients were not significant factors in the
algal response to treatment because they were above saturating concentrations (Rier and
Stevenson, 2002). Phosphorus was the only macro-nutrient that significantly varied with
treatment and it was higher in the high ozone treatment, which had the lower cyanophyte
biomass. This negative relationship between SRP and biomass indicates that it is biomass that is

90

influencing the nutrient concentration, and not the other way around. This SRP/biomass
relationship, the lack of statistical difference in SiO2 and TDN levels, and the biologically nonlimiting concentrations of all three (Bothwell 1988; Horner et al. 1990; Lohman et al. 1991;
Rosemarin 1982; Wong and Clark 1976), provide evidence that differences in these inorganic
nutrients are likely not an indirect cause of ozone-related biomass differences in this experiment.
Heterotrophic bacteria were not enumerated for these samples, but the high level of available
nutrients makes competition between algae and bacteria an unlikely indirect factor.
These experiments indicate that cyanophytes are more sensitive to ozone-induced
oxidative stress than diatoms, corroborating other studies that show cyanophyte species
responding to lower concentrations of oxidants than other algal groups (Apostolova et al.;
Drabkova et al. 2007). These different tolerances may be a result of cell morphology. The
simple prokaryotic cell structure of cyanobacteria leaves these cells more susceptible to external
stress factors. Light-harvesting complexes are on the outside of the thylakoidal membrane and
the thylakoidal membrane itself is not contained in a membrane-bound chloroplast. This may
cause the photosynthetic apparatus to be more susceptible to molecules entering the cytosol.
However, oxidative damage can be mitigated through the production of a multitude of enzymatic
and non-enzymatic antioxidants, which appears to be a primary response of photosynthetic cells
to ameliorate much of the increased oxidative potential (Schraudner et al. 1997). The type and
quantity of antioxidants produced varies depending upon taxon and may be the basis for differing
tolerances to oxidative stress (Choo et al. 2004; Contreras et al. 2009; Gorbi et al. 2006; Knauert
and Knauer 2008; Sabatini et al. 2009). Taxa within the Cyanophyta typically have a smaller
suite of molecules available to decompose reactive oxygen species (Drabkova et al. 2007). The

91

simpler set of antioxidants and less-protected cellular components may leave prokaryotic cells
more susceptible to damage from externally introduced oxidants.
The cyanophyte response in the current research is different than that of cyanophytes in
the high DOC water in the Chapter 2 experiment. In that study, cyanophyte growth rate
increased significantly in the latter half of the experiments in the ozone treatments but not in the
control. In fact, community dominance shifted from diatoms to cyanophytes in the ozone-treated
samples. As our current study indicates the cyanophyte sensitivity to ozone-induced oxidative
stress, it is unlikely that the cyanophyte bloom seen in the Chapter 2 experiment was due to
direct stimulation of cyanophyte cell growth by ozone, but, instead, to ozone effects on other
ecological factors. The success of cyanophytes in those systems, in spite of the high ozone load,
suggests that other system components may have mitigated negative effects. The higher DOC
concentration may have protected algal cells from direct oxidation by reacting with the reactive
oxygen molecules before they reacted with the algal cells themselves. It is known that DOC is a
sink for ozone in wastewater treatment (Huang et al. 2006; Rositano et al. 2001) and that humic
substances, in particular, mitigate the negative effects of other stressors such as herbicides and
heavy metals by reacting with the offending molecules (Kulikova et al. 2005; Steinberg et al.
2008; Wetzel 2001b). In addition, the cyanophytes may be insulated by the mucilage layer
associated with the periphyton mat. In the Chapter 2 assemblages, a diatom mat was already
established before the cyanophyte bloom and may have provided a stable biofilm with associated
mucilage that protected cyanophytes from water column oxidants.
The significant results of the current research and, those seen in Chapters and 3, indicate
that atmospheric ozone, at levels seen in nature, is a potential aquatic pollutant and illustrates its
complex direct and indirect effects. The direct oxidative stress caused by ozone dissolution into

92

freshwaters may create an environment that differentially affects groups of algae within each
community, causing a shift in relative abundance. This study indicates that, in the absence of
strong indirect effects, this shift may be caused by a greater suppression of cyanobacterial than
diatom growth. However, other dissolved molecules and complex species interactions may alter
the effect greatly, as was evidenced by the cyanophyte “bloom” in the ozone treatments in the
Chapter 2 experiment.
Ozone levels continue to rise with increasing worldwide industrialization and fossil fuel
combustion. When modeling aquatic ecosystem health, it may be important to consider the
effects of this pollutant and its oxidative potential on aquatic as well as terrestrial ecosystems.
Algal as well as bacterial communities can be changed at a very fundamental level. Further
research to untangle the complex, direct and indirect effects of atmospheric ozone on aquatic
biota is warranted.

93

Chapter 5: Summary

Background levels of tropospheric ozone and local maxima have been increasing since
the industrial revolution causing ozone pollution problems in both rural and urban areas around
the globe. There is a tremendous amount of data that show significant negative effects of ozone
pollution on human and ecosystem health. Terrestrial primary producers are particularly
susceptible to ozone-mediated oxidative stress and responses to both acute exposure to high
concentrations and chronic exposure to lower concentrations may cause tissue damage and slow
the growth rate of plants. The results of laboratory experiments in my dissertation research
indicate that there is also a potential for atmospheric ozone to act as an aquatic pollutant. My
examination of the effects of ozone concentration and DOC on both multi-divisional periphyton
assemblages and independent cultures of cyanophytes, diatoms, and heterotrophic bacteria have
shown that atmospheric ozone may significantly affect aquatic ecosystems through changes to
algae and bacteria in periphyton films. Periphyton responses to elevated atmospheric ozone were
complex and varied with algal division, water chemistry, and time. In general the hypothesized
negative effects of oxidative stress were minimal in natural periphyton assemblages and occurred
early in assemblage development. Indirect effects on biomass, brought about by interactions
with other periphyton and water chemistry, were primarily positive and varied with algal
division. These effects were seemingly time-dependent as the direct effects appeared to be of
greater importance early in film development and indirect effects more important as the
assemblage grew and changed and as ozone effects in the water column increased.
The effects of atmospheric ozone and DOC on natural periphyton assemblages, those
consisting of several algal divisions and heterotrophic bacteria, indicated many direct, indirect,
94

and interactive causal pathways. High DOC water, in general, had a negative effect on diatom
biomass, a positive effect on cyanophytes, and no effect on chlorophytes and heterotrophic
bacteria. Responses of diatoms and cyanophytes to high DOC water seem to be due to DOC
concentration rather than other covarying factors. The response of algal biomass to ozone
treatment appeared to be primarily indirect and largely influenced by the DOC concentration of
the water. Statistical significance of ozone effect was seen in both the low and high DOC
waters, but ozone effects were more distinct in the high DOC water with great differences
between the cyanophyte and diatom response.
As constituents of these complex periphyton films, cyanophytes were stimulated by the
interaction between ozone and DOC. The greatest assemblage change that occurred in this
experiment was a cyanophtye bloom in the high DOC water in both the medium and high ozone
treatments between weeks 2 and 4. Cyanophyte biomass increased two orders of magnitude in
those microcosms. The mechanism of this ozone-DOC interaction effect cannot be directly
discerned from the experiment but may be related to the breakdown of complex humic
molecules, by ozone and other reactive oxygen species, and the subsequent changes in the light
and/or nutrient regime. Cyanophytes were not affected by ozone in the low DOC water.
Diatoms also responded to ozone treatment, but their response was negative and of a
lesser magnitude than that of the cyanophytes. This negative response was only statistically
significant in the medium ozone treatments in both DOC waters and only at week 2. This may
indicate a non-linear effect of ozone on these diatom assemblages, but further studies would be
needed to support this hypothesis.
Chlorophyte biomass did not significantly respond to ozone treatment in either DOC
water. Ozone effects on heterotrophic bacterial biomass within these periphyton films were

95

variable and echoed those of total algal biomass. In fact, bacterial density was strongly
correlated with algal biovolume, indicating that it was an interaction with these algal constituents
that most heavily influenced bacterial growth. The results indicate that ozone will likely have a
greater effect in high DOC waters and, at least for cyanophytes, this interactive effect appears to
be positive.
Cyanophytes, diatoms and heterotrophic bacteria responded differently to ozone
treatment when grown as independent cultures with sufficient nutrients and low humic carbon
content. The results indicated differential sensitivity of each taxonomic group to the direct
effects of atmospheric ozone. Cyanophytes were the only taxonomic group studied that
responded negatively to ozone treatment. Cyanophyte biomass, when grown in the absence of
other algal groups and provided with saturating nutrients, was reduced in both the medium and
high ozone environments, indicating a direct negative effect of oxidative stress. All cyanophyte
genera used responded similarly. Independent diatom assemblages, also grown with saturating
nutrients, were not sensitive to ozone treatment. Neither diatom biomass nor assemblage
composition was affected by oxidative stress. Heterotrophic bacteria, when grown in the
absence of algae, were not negatively affected by ozone treatment indicating no direct effects of
ozone treatment. However, positive indirect effects were apparent as bacterial biomass was
stimulated by both the medium and high ozone treatments.
The differences in response between the single group experiments and that of the same
groups in the mixed periphyton experiment suggest that the role of ozone, whether it stresses or
benefits a population, depends on several ecosystem factors. This illustrates the importance of
considering inter-taxa interactions and other environmental variables when considering organism
responses to stressors. The complex effects of ozone on algae and bacteria reveal a strong

96

interactive effect of community composition, pre-exposure water chemistry, and oxidative stress.
This is not unexpected when considering the intimate relationship between individual algal or
bacterial cells with the dissolved substances in the water column and the complex nature of a
periphyton community. It is most likely indicative of periphyton responses to other types of
stress as well. Within a periphyton film, inter-organismal interactions are great due the
proximity of individual cells, making competition and facilitation very important factors
affecting growth of the different groups. Secondly, the physical structure of periphyton films
creates micro-climates within the mats that may vary in nutrient, chemical, and light regime.
Finally, the upper layers of cells may act as a buffer between water column stressors and those
assemblage constituents deeper in the film.
In spite of these ambiguous results, it’s clear that the dissolution of atmospheric ozone
into surface waters may have an ecologically significant effect on aquatic communities,
particularly the interaction of ozone with dissolved organic carbon. The differential responses of
algal divisions and the apparent stimulation of heterotrophic bacterial communities may greatly
change the functional status of periphyton assemblages, as algal taxa and heterotrophic bacteria
differ in their roles in food webs and geochemical cycles and have different metabolic rates and
requirements. My results also indicate that systems high in humic carbon will be most likely to
be affected by high atmospheric ozone levels.
This research was an initial look into the effects of atmospheric ozone as an aquatic
pollutant. The results beg for further research to discern the mechanisms of ozone effects and
explore the potential differential responses of other algal assemblages and habitats. For
example, phytoplankton communities lack the physical and chemical complexity of periphyton
films and species interactions are reduced due to a greater distance between cells. Subaerial

97

algae on tree bark or wet walls, for example, are less protected then either periphyton or
phytoplankton assemblages since they are in direct contact with atmospheric ozone. Ozone
effects on these assemblages may be very different than those on algae and bacteria in
periphyton films. And certainly, due to the great potential of high DOC waters to mediate ozone
effects on algal communities, the study of humic heavy systems such as wetland habitats and
black water lakes would be beneficial.
As we were only able to speculate on the potential mechanisms of indirect ozone effects,
further detailed studies are warranted to support our current hypotheses or generate new ones.
For example, experimental designs with more detailed analysis and direct manipulation of DOC
may support the hypothesized role of humic substances in the response of periphyton to ozone
concentrations. Fractionation of the DOC samples before and after ozone treatment may be used
to verify a breakdown of larger molecules and also identify the production and quantity of algalderived polysaccharides. Isotope-labeled DOC molecules could be used to test the nutrient
addition hypotheses by tracking the breakdown of humic substances, potential release of
inorganic minerals, and their subsequent uptake by algae and bacteria.
This dissertation has provided the foundation for a novel area of ecotoxicological
research within the field of aquatic ecology. My results indicate that atmospheric ozone
pollution may have a significant impact on aquatic communities through functional changes to
algal and bacterial communities. These changes may impact carbon and nutrient cycling as well
as the quality of algal communities as a food source for zooplankton and invertebrates. In
addition, the photo breakdown of DOC molecules, which was shown to indirectly affect
periphyton assemblages, may also have direct effects on water chemistry and light penetration.

98

The ecologically significant effects of these alterations will most likely be greatest in humic
waterbodies.

99

LITERATURE CITED

100

LITERATURE CITED

ABD EL-BAKY, H. H., F. K. EL BAZ, and G. S. EL-BAROTY. 2009. Enhancement of
antioxidant production in Spirulina platensis under oxidative stress. Acta Physiologiae
Plantarum 31: 623-631.
AGENCY, U. S. E. P. 2008. Volatile Organic Compounds Emissions, Report on the
Environment.
AGRELL, J., B. KOPPER, E. P. MCDONALD, and R. L. LINDROTH. 2005. CO2 and O-3
effects on host plant preferences of the forest tent caterpillar (Malacosoma disstria).
Global Change Biology 11: 588-599.
ANESIO, A. M., C. HOLLAS, W. GRANELI, and J. LAYBOURN-PARRY. 2004. Influence of
humic substances on bacterial and viral dynamics in freshwaters. Applied and
Environmental Microbiology 70: 4848-4854.
APHA. 1998. Standard methods for the examination of water and wastewater. American Public
Health Association.
APOSTOLOVA, E., I. POUNEVA, I. GRIGOROVA, K. MINKOVA, N. NIKOLAEVA, and G.
RASHKOV. Differential Response of the Photosynthetic Apparatus of Antarctic Algae
Synechocystis Salina (Cyanophyta) and Chlorella Vulgaris (Chlorophyta) to Uv-B
Radiation. Comptes Rendus De L Academie Bulgare Des Sciences 63: 1009-1016.
ARVOLA, L. 1984. Diel Variation in Primary Production and the Vertical-Distribution of
Phytoplankton in a Polyhumic Lake. Archiv Fur Hydrobiologie 101: 503-519.
AUGUSTAITIS, A. and others 2007. Trend in ambient ozone and an attempt to detect its effect
on biota in forest ecosystem. Step I of Lithuanian studies. Thescientificworldjournal 7:
37-46.
AVNERY, S., D. L. MAUZERALL, J. F. LIU, and L. W. HOROWITZ. 2010. Global crop yield
reductions due to surface ozone exposure: 1. Year 2000 crop production losses and
economic damage. Atmospheric Environment 45: 2284-2296.
BAHRS, H., and C. E. W. STEINBERG. 2012. Impact of two different humic substances on
selected coccal green algae and cyanobacteria-changes in growth and photosynthetic
performance. Environmental Science and Pollution Research 19: 335-346.
BAKWIN, P. S. and others 1994. Reactive Nitrogen-Oxides and Ozone above a Taiga
Woodland. Journal of Geophysical Research-Atmospheres 99: 1927-1936.

101

BALOGH, K. V., and L. VOROS. 1997. High bacterial production in hypertrophic shallow
reservoirs rich in humic substances. Hydrobiologia 342: 63-70.
BARROIN, G., and M. FEUILLADE. 1986. Hydrogen-Peroxide as a Potential Algicide for
Oscillatoria-Rubescens Dc. Water Research 20: 619-623.
BELL, M., and H. ELLIS. 2004. Sensitivity analysis of tropospheric ozone to modified biogenic
emissions for the Mid-Atlantic region. Atmospheric Environment 38: 1879-1889.
BENDER, J., and H. J. WEIGEL. 2011. Changes in atmospheric chemistry and crop health: A
review. Agronomy for Sustainable Development 31: 81-89.
BERGSTROM, A. K., and M. JANSSON. 2000. Bacterioplankton production in humic Lake
Ortrasket in relation to input of bacterial cells and input of allochthonous organic carbon.
Microbial Ecology 39: 101-115.
BERMEJO, V., B. S. GIMENO, J. SANZ, D. DE LA TORRE, and J. M. GIL. 2003. Assessment
of the ozone sensitivity of 22 native plant species from Mediterranean annual pastures
based on visible injury. Atmospheric Environment 37: 4667-4677.
BETZER, N., Y. ARGAMAN, and Y. KOTT. 1980. Effluent Treatment and Algae Recovery by
Ozone-Induced Flotation. Water Research 14: 1003-1009.
BIRD, D. F., and J. KALFF. 1984. Empirical Relationships between Bacterial Abundance and
Chlorophyll Concentration in Fresh and Marine Waters. Canadian Journal of Fisheries
and Aquatic Sciences 41: 1015-1023.
BOBBINK, R. 1998. Impacts of tropospheric ozone and airborne nitrogenous pollutants on
natural and seminatural ecosystems: a commentary. New Phytologist 139: 161-168.
BOOKER, F. and others 2009. The Ozone Component of Global Change: Potential Effects on
Agricultural and Horticultural Plant Yield, Product Quality and Interactions with Invasive
Species. Journal of Integrative Plant Biology 51: 337-351.
BORCHARDT, M. A. 1996. Nutrients. In R. J. Stevenson, M. L. Bothwell and R. L. Lowe
[eds.], Algal Ecology: Freshwater Benthic Ecosystems. Academic Press.
BOTHWELL, M. L. 1988. Growth rate responses of lotic periphytic diatoms to experimental
phosphorus enrichment: The influence of temperature and light. Canadian Journal of
Fisheries and Aquatic Sciences 45: 261-270.
BRANKOV, E. and others 2003. Assessing the effects of transboundary ozone pollution between
Ontario, Canada and New York, USA. Environmental Pollution 123: 403-411.
BRAUN, S., U. ZUGMAIER, V. THOMAS, and W. FLUCKIGER. 2004. Carbohydrate
concentrations in different plant parts of young beech and spruce along a gradient of
ozone pollution. Atmospheric Environment 38: 2399-2407.
102

BROADMEADOW, M. 1998. Ozone and forest trees. New Phytologist 139: 123-125.
BUTLER, T. J. and others 2011. Response of ozone and nitrate to stationary source NOx
emission reductions in the eastern USA. Atmospheric Environment 45: 1084-1094.
CALATAYUD, A., P. J. TEMPLE, and E. BARRENO. 2000. Chlorophyll a fluorescence
emission, xanthophyll cycle activity, and net photosynthetic rate responses to ozone in
some foliose and fruticose lichen species. Photosynthetica 38: 281-286.
CANAS, M. S., and M. L. PIGNATA. 2000. Temporal variation of pigments and peroxidation
products in the lichen Parmotrema uruguense (Kremplh.) Hale transplanted to urban and
non-polluted environments. Symbiosis
1998 24: 147-161.
CHAMEIDES, W. L., R. W. LINDSAY, J. RICHARDSON, and C. S. KIANG. 1988. The Role
of Biogenic Hydrocarbons in Urban Photochemical Smog - Atlanta as a Case-Study.
Science 241: 1473-1475.
CHAMEIDES, W. L., R. D. SAYLOR, and E. B. COWLING. 1997. Ozone pollution in the rural
United States and the new NAAQS. Science 276: 916-916.
CHAPPELKA, A. H., and L. J. SAMUELSON. 1998. Ambient ozone effects on forest trees of
the eastern United States: a review. New Phytologist 139: 91-108.
CHATFIELD, R., and H. HARRISON. 1977. Tropospheric Ozone .1. Evidence for Higher
Background Values. Journal of Geophysical Research-Oceans and Atmospheres 82:
5965-5968.
CHEUNG, V. T. F., and T. WANG. 2001. Observational study of ozone pollution at a rural site
in the Yangtze Delta of China. Atmospheric Environment 35: 4947-4958.
CHO, M., H. KIM, S. H. CHO, and J. YOON. 2003. Investigation of ozone reaction in river
waters causing instantaneous ozone demand. Ozone-Science & Engineering 25: 251-259.
CHOO, K. S., J. NILSSON, M. PEDERSEN, and P. SNOEIJS. 2005. Photosynthesis, carbon
uptake and antioxidant defence in two coexisting filamentous green algae under different
stress conditions. Marine Ecology-Progress Series 292: 127-138.
CHOO, K. S., P. SNOEIJS, and M. PEDERSEN. 2004. Oxidative stress tolerance in the
filamentous green algae Cladophora glomerata and Enteromorpha ahlneriana. Journal of
Experimental Marine Biology and Ecology 298: 111-123.
CHRZANOWSKI, T. H. 1985. Seasonality, Abundance, and Biomass of Bacteria in a
Southwestern Reservoir. Hydrobiologia 127: 117-123.
COMMITTEE ON TROPOSPHERIC OZONE, N. R. C. 1991. Rethinking the Ozone Problem in
Urban and Regional Air Pollution. National Academy Press.
103

CONTRERAS, L., D. MELLA, A. MOENNE, and J. A. CORREA. 2009. Differential responses
to copper-induced oxidative stress in the marine macroalgae Lessonia nigrescens and
Scytosiphon lomentaria (Phaeophyceae). Aquatic Toxicology 94: 94-102.
DAUM, P. H. and others 2003. A comparative study of O-3 formation in the Houston urban and
industrial plumes during the 2000 Texas Air Quality Study. Journal of Geophysical
Research-Atmospheres 108.
DE VELASQUEZ, M. T. O., M. N. ROJAS-VALENCIA, and A. AYALA. 2008. Wastewater
disinfection using ozone to remove free-living, highly pathogenic bacteria and amoebae.
Ozone-Science & Engineering 30: 367-375.
DEBAJE, S. B., S. J. JEYAKUMAR, K. GANESAN, D. B. JADHAV, and P.
SEETARAMAYYA. 2003. Surface ozone measurements at tropical rural coastal station
Tranquebar, India. Atmospheric Environment 37: 4911-4916.
DITCHKOFF, S. S., J. S. LEWIS, J. C. LIN, R. B. MUNTIFERING, and A. H. CHAPPELKA.
2009. Nutritive Quality of Highbush Blackberry (Rubus argutus) Exposed to
Tropospheric Ozone. Rangeland Ecology & Management 62: 364-370.
DOHM, M. R., W. J. MAUTZ, P. G. LOOBY, K. S. GELLERT, and J. A. ANDRADE. 2001.
Effects of ozone on evaporative water loss and thermoregulatory behavior of marine
toads (Bufo marinus). Environmental Research 86: 274-286.
DRABKOVA, M., H. C. P. MATTHIJS, W. ADMIRAAL, and B. MARSALEK. 2007. Selective
effects of H2O2 on cyanobacterial photosynthesis. Photosynthetica 45: 363-369.
DRABKOVA, M., W. ADMIRAAL, and B. MARSALEK. 2007. Combined exposure to
hydrogen peroxide and light - Selective effects on cyanobacteria, green algae, and
diatoms. Environmental Science & Technology 41: 309-314.
DRAKARE, S., P. BLOMQVIST, A. K. BERGSTROM, and M. JANSSON. 2003.
Relationships between picophytoplankton and environmental variables in lakes along a
gradient of water colour and nutrient content. Freshwater Biology 48: 729-740.
EDER, B. K., J. M. DAVIS, and P. BLOOMFIELD. 1993. A Characterization of the
Spatiotemporal Variability of Nonurban Ozone Concentrations over the Eastern UnitedStates. Atmospheric Environment Part a-General Topics 27: 2645-2668.
ELLIS, B. D., P. BUTTERFIELD, W. L. JONES, G. A. MCFETERS, and A. K. CAMPER.
1999. Effects of carbon source, carbon concentration, and chlorination on growth related
parameters of heterotrophic biofilm bacteria. Microbial Ecology 38: 330-347.
ENGELHAUPT, E., T. S. BIANCHI, R. G. WETZEL, and M. A. TARR. 2003. Photochemical
transformations and bacterial utilization of high-molecular-weight dissolved organic
104

carbon in a southern Louisiana tidal stream (Bayou Trepagnier). Biogeochemistry 62: 3958.
ESCOBAR, I. C., A. A. RANDALL, and J. S. TAYLOR. 2001. Bacterial growth in distribution
systems: Effect of assimilable organic carbon and biodegradable dissolved organic
carbon. Environmental Science & Technology 35: 3442-3447.
FAGERBERG, T., P. CARLSSON, and M. LUNDGREN. 2009. A large molecular size fraction
of riverine high molecular weight dissolved organic matter (HMW DOM) stimulates
growth of the harmful dinoflagellate Alexandrium minutum. Harmful Algae 8: 823-831.
FAHMI, W. NISHIJIMA, and M. OKADA. 2003. Improvement of DOC removal by multi-stage
AOP-biological treatment. Chemosphere 50: 1043-1048.
FAST, J. D., and W. E. HEILMAN. 2003. The effect of lake temperatures and emissions on
ozone exposure in the western Great Lakes region. Journal of Applied Meteorology 42:
1197-1217.
FINLAYSON-PITTS, B. J., and J. N. J. PITTS. 1999. Chemistry of the Upper and Lower
Atmosphere Theory, Experiments, and Applications. Academic Press.
FIORE, A. M. and others 2002. Background ozone over the United States in summer: Origin,
trend, and contribution to pollution episodes. Journal of Geophysical ResearchAtmospheres 107: art. no.-4275.
FISCUS, E. L., F. L. BOOKER, and K. O. BURKEY. 2005. Crop responses to ozone: uptake,
modes of action, carbon assimilation and partitioning. Plant Cell and Environment 28:
997-1011.
FOWLER, D., C. FLECHARD, U. SKIBA, M. COYLE, and J. N. CAPE. 1998. The
atmospheric budget of oxidized nitrogen and its role in ozone formation and deposition.
New Phytologist 139: 11-23.
FRANZARING, J., A. E. G. TONNEIJCK, A. W. N. KOOIJMAN, and T. A. DUECK. 2000.
Growth responses to ozone in plant species from wetlands. Environmental and
Experimental Botany 44: 39-48.
FREESE, S. D., D. NOZAIC, M. J. PRYOR, D. L. TROLLIP, and R. A. SMITH. 1999.
Comparison of ozone and hydrogen peroxide/ozone for the treatment of eutrophic water.
Water Science and Technology 39: 325-328.
FUHRER, J. 2009. Ozone risk for crops and pastures in present and future climates.
Naturwissenschaften 96: 173-194.

105

FURLAN, C. M., A. SALATINO, and M. DOMINGOS. 2004. Influence of air pollution on leaf
chemistry, herbivore feeding and gall frequency on Tibouchina pulchra leaves in Cubatao
(Brazil). Biochemical Systematics and Ecology 32: 253-263.
GARCIA, A. Z., T. H. NASH, and M. A. HERRERA-CAMPOS. 2000. Lichen decline in
Desierto de los Leones (Mexico City). Bryologist 103: 428-441.
GATE, I. M., S. MCNEILL, and M. R. ASHMORE. 1995. Effects of air pollution on the
searching behaviour of an insect parasitoid. Water Air and Soil Pollution 85: 1425-1430.
GAVAND, M. R., J. B. MCCLINTOCK, C. D. AMSLER, R. W. PETERS, and R. A. ANGUS.
2007. Effects of sonication and advanced chemical oxidants on the unicellular green alga
Dunaliella tertiolecta and cysts, larvae and adults of the brine shrimp Artemia salina: A
prospective treatment to eradicate invasive organisms from ballast water. Marine
Pollution Bulletin 54: 1777-1788.
GERRITY, D. and others 2011. Pilot-scale evaluation of ozone and biological activated carbon
for trace organic contaminant mitigation and disinfection. Water Research 45: 21552165.
GOLDSTONE, J. V., M. J. PULLIN, S. BERTILSSON, and B. M. VOELKER. 2002. Reactions
of hydroxyl radical with humic substances: Bleaching, mineralization, and production of
bioavailable carbon substrates. Environmental Science & Technology 36: 364-372.
GONCALVES, A. A., and G. A. GAGNON. 2011. Ozone Application in Recirculating
Aquaculture System: An Overview. Ozone-Science & Engineering 33: 345-367.
GORBI, G., E. TORRICELLI, B. PAWLIK-SKOWROTISKA, L. S. DI TOPPI, C. ZANNI, and
M. G. CORRADI. 2006. Differential responses to Cr(VI)-induced oxidative stress
between Cr-tolerant and wild-type strains of Scenedesmus acutus (Chlorophyceae).
Aquatic Toxicology 79: 132-139.
GRAHAM, N. J. D. 1999. Removal of humic substances by oxidation/biofiltration processes - A
review. Water Science and Technology 40: 141-148.
GRANIER, C., and G. P. BRASSEUR. 2003. The impact of road traffic on global tropospheric
ozone. Geophysical Research Letters 30.
GUZEL-SEYDIM, Z. B., A. K. GREENE, and A. C. SEYDIM. 2004. Use of ozone in the food
industry. Lebensmittel-Wissenschaft Und-Technologie-Food Science and Technology
37: 453-460.
HAAGENSMIT, A. J., C. E. BRADLEY, and M. M. FOX. 1953. Ozone Formation in
Photochemical Oxidation of Organic Substances. Industrial and Engineering Chemistry
45: 2086-2089.

106

HEAGLE, A. S. 1989. Ozone and Crop Yield. Annual Review of Phytopathology 27: 397-423.
HEATH, R. L. 1994. Possible Mechanisms for the Inhibition of Photosynthesis by Ozone.
Photosynthesis Research 39: 439-451.
HEINLEIN, J. 2012. Impacts of elevated atmospheric ozone and dissolved organic carbon
content on benthic algal community composition and biomass. Zoology. Michigan State
University.
HEWSON, I., J. M. O'NEIL, C. A. HEIL, G. BRATBAK, and W. C. DENNISON. 2001. Effects
of concentrated viral communities on photosynthesis and community composition of cooccurring benthic microalgae and phytoplankton. Aquatic Microbial Ecology 25: 1-10.
HOEGER, S. J., D. R. DIETRICH, and B. C. HITZFELD. 2002. Effect of ozonation on the
removal of cyanobacterial toxins during drinking water treatment. Environmental Health
Perspectives 110: 1127-1132.
HOLTON, M. K., R. L. LINDROTH, and E. V. NORDHEIM. 2003. Foliar quality influences
tree-herbivore-parasitoid interactions: effects of elevated CO2, O-3, and plant genotype.
Oecologia 137: 233-244.
HORNER, R. R., E. B. WELCH, M. R. SEELEY, and J. M. JACOBY. 1990. Responses of
periphyton to changes in current velocity, suspended sediment and phosphorus
concentration. Freshwater Biology 24: 215-232.
HUANG, W. J., B. L. CHENG, S. K. HU, and C. W. CHU. 2006. Ozonation of algae and odor
causing substances in eutrophic waters. Journal of Environmental Science and Health
Part a-Toxic/Hazardous Substances & Environmental Engineering 41: 1587-1605.
HUDSON, J. B., M. SHARMA, and S. VIMALANATHAN. 2009. Development of a Practical
Method for Using Ozone Gas as a Virus Decontaminating Agent. Ozone-Science &
Engineering 31: 216-223.
HUR, J. S., and A. R. WELLBURN. 1994. Effects of Atmospheric O-3, on Azolla-Anabaena
Symbiosis. Annals of Botany 73: 205-209.
IBARRA-BERASTEGI, G., and I. MADARIAGA. 2003. Traffic congestion and ozone
precursor emissions in Bilbao (Spain). Environmental Science and Pollution Research
10: 361-367.
IMAI, A., T. FUKUSHIMA, and K. MATSUSHIGE. 1999. Effects of iron limitation and aquatic
humic substances on the growth of Microcystis aeruginosa. Canadian Journal of Fisheries
and Aquatic Sciences 56: 1929-1937.
ISHII, S., F. M. MARSHALL, J. N. B. BELL, and A. M. ABDULLAH. 2004. Impact of ambient
air pollution on locally grown rice cultivars (Oryza sativa L.) in Malaysia. Water Air and
Soil Pollution 154: 187-201.
107

IZDEBSKI, T., M. DORS, and J. MIZERACZYK. 2011. River Water Remediation Using
Electrohydraulic Discharges or Ozonation. Ieee Transactions on Plasma Science 39: 953959.
JANACH, W. E. 1989. Surface Ozone - Trend Details, Seasonal-Variations, and Interpretation.
Journal of Geophysical Research-Atmospheres 94: 18289-18295.
JANSSON, M., P. BLOMQVIST, A. JONSSON, and A. K. BERGSTROM. 1996. Nutrient
limitation of bacterioplankton, autotrophic and mixotrophic phytoplankton, and
heterotrophic nanoflagellates in Lake Ortrasket. Limnology and Oceanography 41: 15521559.
JONES, M. E., and T. D. PAINE. 2006. Detecting changes in insect herbivore communities
along a pollution gradient. Environmental Pollution 143: 377-387.
JUDD, K. E., B. C. CRUMP, and G. W. KLING. 2007. Bacterial responses in activity and
community composition to photo-oxidation of dissolved organic matter from soil and
surface waters. Aquatic Sciences 69: 96-107.
KAY, S. H., P. C. QUIMBY, and J. D. OUZTS. 1984. Photo-Enhancement of HydrogenPeroxide Toxicity to Submersed Vascular Plants and Algae. Journal of Aquatic Plant
Management 22: 25-34.
KLEY, D., M. KLEINMANN, H. SANDERMAN, and S. S. KRUPA. 1999. Photochemical
oxidants: state of the science. Environmental Pollution 100: 19-42.
KLUG, J. L. 2005. Bacterial response to dissolved organic matter affects resource availability for
algae. Canadian Journal of Fisheries and Aquatic Sciences 62: 472-481.
KNAUERT, S., and K. KNAUER. 2008. The role of reactive oxygen species in copper toxicity
to two freshwater green algae. Journal of Phycology 44: 311-319.
KOLLNER, B., and G. H. M. KRAUSE. 2003. Effects of two different ozone exposure regimes
on chlorophyll and sucrose content of leaves and yield parameters of sugar beet (Beta
vulgaris L.) and rape (Brassica napus L.). Water Air and Soil Pollution 144: 317-332.
KONDO, Y. and others 2003. Effects of biomass burning and lightning on atmospheric
chemistry over Australia and South-east Asia. International Journal of Wildland Fire 12:
271-281.
KOSAKOWSKA, A., M. NEDZI, and J. PEMPKOWIAK. 2007. Responses of the toxic
cyanobacterium Microcystis aeruginosa to iron and humic substances. Plant Physiology
and Biochemistry 45: 365-370.

108

KRUPA, S. V., M. NOSAL, and A. H. LEGGE. 1998. A numerical analysis of the combined
open-top chamber data from the USA and Europe on ambient ozone and negative crop
responses. Environmental Pollution 101: 157-160.

KULIKOVA, N. A., E. V. STEPANOVA, and O. V. KOROLEVA. 2005. Mitigating activity of
humic substances: direct influence on biota., p. 285-309. In I. V. e. a. Perminova [ed.],
Use of Humic Substances to Remediate Polluted Environments: From Theory to Practice.
Springer.
KUNHIKRISHNAN, T. and others 2006. Regional NOx emission strength for the Indian
subcontinent and the impact of emissions from India and neighboring countries on
regional O-3 chemistry. New Journal of Physics 8.
LAI, W. L., H. H. YEH, I. C. TSENG, T. F. LIN, J. J. CHEN, and G. T. WANG. 2002.
Conventional versus advanced treatment for eutrophic source water. Journal American
Water Works Association 94: 96-108.
LAI, W. L., H. H. YEH, and I. C. TSENG. 2006. The effect of ozonation and filtration on AOC
(assimilable orgainic carbon) value of water from eutrophic lake. Ozone-Science &
Engineering 28: 29-35.
LARSON, J. L., D. R. ZAK, and R. L. SINSABAUGH. 2002. Extracellular enzyme activity
beneath temperate trees growing under elevated carbon dioxide and ozone. Soil Science
Society of America Journal 66: 1848-1856.
LEDFORD, H. K., and K. K. NIYOGI. 2005. Singlet oxygen and photo-oxidative stress
management in plants and algae. Plant Cell and Environment 28: 1037-1045.
LEITAO, M. A. D., K. H. M. CARDOZO, E. PINTO, and P. COLEPICOLO. 2003. PCBinduced oxidative stress in the unicellular marine dinoflagellate Lingulodinium
polyedrum. Archives of Environmental Contamination and Toxicology 45: 59-65.
LENES, D. and others 2010. Assessment of the removal and inactivation of influenza viruses
H5N1 and H1N1 by drinking water treatment. Water Research 44: 2473-2486.
LENNON, J. T. 2004. Experimental evidence that terrestrial carbon subsidies increase CO2 flux
from lake ecosystems. Oecologia 138: 584-591.
LENNON, J. T., and L. E. PFAFF. 2005. Source and supply of terrestrial organic matter affects
aquatic microbial metabolism. Aquatic Microbial Ecology 39: 107-119.
LEVINE, B. B., K. MADIREDDI, V. LAZAROVA, M. K. STENSTROM, and I. H. SUFFET.
2000. Treatment of trace organic compounds by ozone-biological activated carbon for
wastewater reuse: The Lake Arrowhead pilot plant. Water Environment Research 72:
388-396.
109

LINDELL, M. J., W. GRANELI, and L. J. TRANVIK. 1995. Enhanced Bacterial-Growth in
Response to Photochemical Transformation of Dissolved Organic-Matter. Limnology and
Oceanography 40: 195-199.
LINDROTH, R. L. 2010. Impacts of Elevated Atmospheric CO2 and O-3 on Forests:
Phytochemistry, Trophic Interactions, and Ecosystem Dynamics. Journal of Chemical
Ecology 36: 2-21.
LIPPMANN, M. 1991. Health-Effects of Tropospheric Ozone. Environmental Science &
Technology 25: 1954-1962.
LIU, X. H. and others 2003. Understanding of regional air pollution over China using CMAQ,
part I performance evaluation and seasonal variation. Atmospheric Environment 44:
2415-2426.
LOHMAN, K., J. R. JONES, and C. BAYSINGER-DANIEL. 1991. Experimental evidence for
nitrogen limitation in a northern Ozark stream. Journal of the North American
Benthological Society 19: 14-23.
LOWE, R. L., and G. D. LALIBERTE. 1996. Benthic Stream Algae: Distribution and Structure,
p. 269-293. In F. R. Hauer and G. A. Lamberti [eds.], Methods in Stream Ecology.
Academic Press, Inc.
MADDEN, M. C., and W. E. HOGSETT. 2001. A historical overview of the ozone exposure
problem. Human and Ecological Risk Assessment 7: 1121-1131.
MAHASNEH, I. A., and D. N. TIWARI. 1992. The use of biofertilizer of Calothrix sp. M103,
enhanced by addition of iron and siderophore production. Journal of Applied
Microbiology 73: 286-289.
MAHFOUDH, A., M. MOISAN, J. SEGUIN, J. BARBEAU, Y. KABOUZI, and D.
KEROACK. 2010. Inactivation of Vegetative and Sporulated Bacteria by Dry Gaseous
Ozone. Ozone-Science & Engineering 32: 180-198.
MALLICK, N., and F. H. MOHN. 2000. Reactive oxygen species: response of algal cells.
Journal of Plant Physiology 157: 183-193.
MANN, C. J., and R. G. WETZEL. 1995. Dissolved organic carbon and its utilization in a
riverine wetland ecosystem. Biogeochemistry 31: 99-120.
MANNING, W. J., and B. GODZIK. 2004. Bioindicator plants for ambient ozone in Central and
Eastern Europe. Environmental Pollution 130: 33-39.
MARTIN, D., and I. RIDGE. 1999. The relative sensitivity of algae to decomposing barley
straw. Journal of Applied Phycology 11: 285-291.
110

MAUTZ, W. J., and M. R. DOHM. 2004. Respiratory and behavioral effects of ozone on a lizard
and a frog. Comparative Biochemistry and Physiology a-Molecular & Integrative
Physiology 139: 371-377.
MEEHAN, T. D., M. S. CROSSLEY, and R. L. LINDROTH. 2010. Impacts of elevated CO2
and O-3 on aspen leaf litter chemistry and earthworm and springtail productivity. Soil
Biology & Biochemistry 42: 1132-1137.
MENG, Z. Y. and others 2010. Ambient sulfur dioxide, nitrogen dioxide, and ammonia at ten
background and rural sites in China during 2007-2008. Atmospheric Environment 44:
2625-2631.
METCALFE, S. E., J. D. WHYATT, R. G. DERWENT, and M. O'DONOGHUE. 2002. The
regional distribution of ozone across the British Isles and its response to control
strategies. Atmospheric Environment 36: 4045-4055.
MIDDELBOE, M., and C. LUNDSGAARD. 2003. Microbial activity in the Greenland Sea: role
of DOC lability, mineral nutrients and temperature. Aquatic Microbial Ecology 32: 151163.
MONDOR, E. B., M. N. TREMBLAY, C. S. AWMACK, and R. L. LINDROTH. 2004.
Divergent pheromone-mediated insect behaviour under global atmospheric change.
Global Change Biology 10: 1820-1824.
MORAN, M. A., and R. E. HODSON. 1990. Bacterial Production on Humic and Nonhumic
Components of Dissolved Organic-Carbon. Limnology and Oceanography 35: 17441756.
MORAN, M. A., and R. G. ZEPP. 1997. Role of photoreactions in the formation of biologically
labile compounds from dissolved organic matter. Limnology and Oceanography 42:
1307-1316.
MORGAN, P. B., T. A. MIES, G. A. BOLLERO, R. L. NELSON, and S. P. LONG. 2006.
Season-long elevation of ozone concentration to projected 2050 levels under fully openair conditions substantially decreases the growth and production of soybean. New
Phytologist 170: 333-343.
MURPHY, T. P., D. R. S. LEAN, and C. NALEWAJKO. 1976. Blue-Green-Algae - Their
Excretion of Iron-Selective Chelators Enables Them to Dominate Other Algae.
Science 192: 900-902.
NARDI, S., D. PIZZEGHELLO, A. MUSCOLO, and A. VIANELLO. 2002. Physiological
effects of humic substances on higher plants. Soil Biology & Biochemistry 34: 15271536.

111

NIEMI, R., P. J. MARTIKAINEN, J. SILVOLA, and T. HOLOPAINEN. 2002. Ozone effects on
Sphagnum mosses, carbon dioxide exchange and methane emission in boreal peatland
microcosms. Science of the Total Environment 289: 1-12.
NOVAK, K. and others 2003. Ozone air pollution and foliar injury development on native plants
of Switzerland. Environmental Pollution 125: 41-52.
OKSANEN, E. 2003. Responses of selected birch (Betula pendula Roth) clones to ozone change
over time. Plant Cell and Environment 26: 875-886.
OKSANEN, E., E. HAIKIO, J. SOBER, and D. F. KARNOSKY. 2004. Ozone-induced H2O2
accumulation in field-grown aspen and birch is linked to foliar ultrastructure and
peroxisomal activity. New Phytologist 161: 791-799.
PARALKAR, A., and J. K. EDZWALD. 1996. Effect of ozone on EOM and coagulation.
Journal American Water Works Association 88: 143-154.
PARK, M. H., B. H. KIM, I. M. CHUNG, and S. J. HWANG. 2009. Selective Bactericidal
Potential of Rice (Oryza sativa L. var. japonica) Hull Extract on Microcystis Strains in
Comparison with Green Algae and Zooplankton. Bulletin of Environmental
Contamination and Toxicology 83: 97-101.
PARRISH, D. D., J. S. HOLLOWAY, M. TRAINER, P. C. MURPHY, G. L. FORBES, and F.
C. FEHSENFELD. 1993. Export of North-American Ozone Pollution to the NorthAtlantic Ocean. Science 259: 1436-1439.
PETERSON, C. G., T. L. DUDLEY, K. D. HOAGLAND, and L. M. JOHNSON. 1993.
Infection, Growth, and Community-Level Consequences of a Diatom Pathogen in a
Sonoran Desert Stream. Journal of Phycology 29: 442-452.
PFLUGMACHER, S., L. F. TIDWELL, and C. E. W. STEINBERG. 2001. Dissolved humic
substances can directly affect freshwater organisms. Acta Hydrochimica Et
Hydrobiologica 29: 34-40.
PINTO, E., T. C. S. SIGAUD-KUTNER, M. A. S. LEITAO, O. K. OKAMOTO, D. MORSE,
and P. COLEPICOLO. 2003. Heavy metal-induced oxidative stress in algae. Journal of
Phycology 39: 1008-1018.
PILLSBURY, R. W., and R. L. LOWE. 1999. The response of benthic algae to manipulations of
light in four acidic lakes in northern Michigan. Hydrobiologia 394: 69-81.
POLLARD, P. C., and H. DUCKLOW. 2011. Ultrahigh bacterial production in a eutrophic
subtropical Australian river: Does viral lysis short-circuit the microbial loop? Limnology
and Oceanography 56: 1115-1129.

112

POWER, S. A., and M. R. ASHMORE. 2002. Responses of fen and fen-meadow communities to
ozone. New Phytologist 156: 399-408.
PROCTOR, L. M., and J. A. FUHRMAN. 1991. Roles of Viral-Infection in Organic ParticleFlux. Marine Ecology-Progress Series 69: 133-142.
PROKHOTSKAYA, V. Y., and C. E. W. STEINBERG. 2007. Differential sensitivity of a coccal
green algal and a cyanobacterial species to dissolved natural organic matter (NOM).
Environmental Science and Pollution Research 14: 11-18.
PROZHERINA, N., V. FREIWALD, M. ROUSI, and E. OKSANEN. 2003. Interactive effect of
springtime frost and elevated ozone on early growth, foliar injuries and leaf structure of
birch (Betula pendula). New Phytologist 159: 623-636.
REYNOLDS, C. S. 2007. Variability in the provision and function of mucilage in
phytoplankton: facultative responses to the environment. Hydrobiologia 578: 37-45.
RIER, S. T., and R. J. STEVENSON. 2002. Effects of light, dissolved organic carbon, and
inorganic nutrients on the relationship between algae and heterotrophic bacteria in stream
periphyton. Hydrobiologia 489: 179-184.
ROMANI, A. M. and others 2004. Biofilm structure and function and possible implications for
riverine DOC dynamics. Microbial Ecology 47: 316-328.
ROSEMARIN, A. S. 1982. Phosphorus nutrition of two potentially competing filamentous algae,
Cladophora glomerata and Stigeoclonium tenue from Lake Ontario. Journal of Great
Lakes Research 8: 66-72.
ROSENSTOCK, B., and M. SIMON. 2003. Consumption of dissolved amino acids and
carbohydrates by limnetic bacterioplankton according to molecular weight fractions and
proportions bound to humic matter. Microbial Ecology 45: 433-443.
ROSENSTOCK, B., W. ZWISLER, and M. SIMON. 2005. Bacterial consumption of humic and
non-humic low and high molecular weight DOM and the effect of solar irradiation on the
turnover of labile DOM in the Southern Ocean. Microbial Ecology 50: 90-101.
ROSENTHAL, J. S., R. A. HELVEY, T. E. BATTALINO, C. FISK, and P. W. GREIMAN.
2003. Ozone transport by mesoscale and diurnal wind circulations across southern
California. Atmospheric Environment 2003 37: S51-S71.
ROSITANO, J., G. NEWCOMBE, B. NICHOLSON, and P. SZTAJNBOK. 2001. Ozonation of
NOM and algal toxins in four treated waters. Water Research
2001 35: 23-32.
SABATINI, S. E., A. B. JUAREZ, M. R. EPPIS, L. BIANCHI, C. M. LUQUET, and M. D. R.
DE MOLINA. 2009. Oxidative stress and antioxidant defenses in two green microalgae
exposed to copper. Ecotoxicology and Environmental Safety 72: 1200-1206.
113

SANCHEZ, J. A., and F. J. G. AYALA. 2008. Recent Trend in Ozone Levels in the
Metropolitan zone of Mexico City. Journal of the Mexican Chemical Society 52: 256262.
SANHUEZA, E., E. FERNANDEZ, L. DONOSO, and J. ROMERO. 2000. Boundary layer
ozone in the tropical America northern hemisphere region. Journal of Atmospheric
Chemistry 35: 249-272.
SATSANGI, G. S., A. LAKHANI, P. R. KULSHRESTHA, and A. TANEJA. 2004. Seasonal
and diurnal variation of surface ozone and a preliminary analysis of exceedance of its
critical levels at a semi-arid site in India. Journal of Atmospheric Chemistry 47: 271-286.
SCHEIDEGGER, C., and B. SCHROETER. 1995. Effects of Ozone Fumigation on Epiphytic
Macrolichens - Ultrastructure, Co-2 Gas-Exchange and Chlorophyll Fluorescence.
Environmental Pollution 88: 345-354.
SCHRAUDNER, M., C. LANGEBARTELS, and H. SANDERMANN. 1997. Changes in the
biochemical status of plant cells induced by the environmental pollutant ozone.
Physiologia Plantarum 100: 274-280.
SERVICE, U. S. N. P. 2011. Ozone Standard Exceedances in National Parks.
SEMPLE, K. T., R. B. CAIN, and S. SCHMIDT. 1999. Biodegradation of aromatic compounds
by microalgae. Fems Microbiology Letters 170: 291-300.
SHEEL, V., S. LAL, A. RICHTER, and J. P. BURROWS. 2010. Comparison of satellite
observed tropospheric NO2 over India with model simulations. Atmospheric
Environment 44: 3314-3321.
SIGAL, L. L., and J. W. JOHNSTON. 1986. Effects of Acidic Rain and Ozone on NitrogenFixation and Photosynthesis in the Lichen Lobaria-Pulmonaria (L) Hoffm. Environmental
and Experimental Botany 26: 59-64.
SIMPSON, D. 1993. Photochemical Model-Calculations over Europe for 2 Extended Summer
Periods - 1985 and 1989 - Model Results and Comparison with Observations.
Atmospheric Environment Part a-General Topics 27: 921-943.
SKARBY, L. and others 2004. Growth of Norway spruce (Picea abies) in relation to different
ozone exposure indices: a synthesis. Atmospheric Environment 38: 2225-2236.
STEINBERG, C. E. W., A. PAUL, S. PFLUGMACHER, T. MEINELT, R. KLOCKING, and C.
WIEGAND. 2003. Pure humic substances have the potential to act as xenobiotic
chemicals - A review. Fresenius Environmental Bulletin 12: 391-401.

114

STEINMAN, A. D., C. D. MCINTIRE, S. V. GREGORY, and G. A. LAMBERTI. 1989. Effects
of Irradiance and Grazing on Lotic Algal Assemblages. Journal of Phycology 25: 478485.
STEVENSON, R. J. 1984. Procedures for Mounting Algae in a Syrup Medium. Transactions of
the American Microscopical Society 103: 320-321.
STEWART, A. J., and R. G. WETZEL. 1981. Dissolved Humic Materials - Photodegradation,
Sediment Effects, and Reactivity with Phosphate and Calcium-Carbonate Precipitation.
Archiv Fur Hydrobiologie 92: 265-286.
---. 1982. Influence of Dissolved Humic Materials on Carbon Assimilation and AlkalinePhosphatase Activity in Natural Algal Bacterial Assemblages. Freshwater Biology 12:
369-380.
SUN, B. K., Y. TANJI, and H. UNNO. 2006. Extinction of cells of cyanobacterium Anabaena
circinalis in the presence of humic acid under illumination. Applied Microbiology and
Biotechnology 72: 823-828.
SUNDH, I., and R. T. BELL. 1992. Extracellular Dissolved Organic-Carbon Released from
Phytoplankton as a Source of Carbon for Heterotrophic Bacteria in Lakes of Different
Humic Content. Hydrobiologia 229: 93-106.
SUTTLE, C. A., and A. M. CHAN. 1993. Marine Cyanophages Infecting Oceanic and Coastal
Strains of Synechococcus - Abundance, Morphology, Cross-Infectivity and GrowthCharacteristics. Marine Ecology-Progress Series 92: 99-109.
SWAP, R. J., T. A. SZUBA, M. GARSTANG, H. J. ANNEGARN, L. MARUFU, and S. J.
PIKETH. 2003. Spatial and temporal assessment of sources contributing to the annual
austral spring mid-tropospheric ozone maxima over the tropical South Atlantic. Global
Change Biology 9: 336-345.
SWIETLIK, J., and E. SIKORSKA. 2004. Application of fluorescence spectroscopy in the
studies of natural organic matter fractions reactivity with chlorine dioxide and ozone.
Water Research 38: 3791-3799.
TAFT, C. E. 1978. Mounting Medium for Freshwater Plankton. Transactions of the American
Microscopical Society 97: 263-264.
TANNER, B. D., S. KUWAHARA, C. P. GERBA, and K. A. REYNOLDS. 2004. Evaluation of
electrochemically generated ozone for the disinfection of water and wastewater. Water
Science and Technology 50: 19-25.
THOMAS, J. D. 1997. The role of dissolved organic matter, particularly free amino acids and
humic substances, in freshwater ecosystems. Freshwater Biology 38: 1-36.

115

THOMPSON, A. M., J. C. WITTE, M. T. FREIMAN, N. A. PHAHLANE, and G. J. R.
COETZEE. 2002. Lusaka, Zambia, during SAFARI-2000: Convergence of local and
imported ozone pollution. Geophysical Research Letters 29.
TILTON, B. E. 1989. Health-Effects of Tropospheric Ozone - Major Effects and Related
Scientific Questions. Environmental Science & Technology 23: 257-263.
TITTEL, J., and N. KAMJUNKE. 2004. Metabolism of dissolved organic carbon by planktonic
bacteria and mixotrophic algae in lake neutralisation experiments. Freshwater Biology
49: 1062-1071.
TITTEL, J. and others 2009. Utilisation of terrestrial carbon by osmotrophic algae. Aquatic
Sciences 71: 46-54.
TRANVIK, L. J. 1988. Availability of Dissolved Organic-Carbon for Planktonic Bacteria in
Oligotrophic Lakes of Differing Humic Content. Microbial Ecology 16: 311-322.
TSENG, C., and C. LI. 2008. Inactivation of surface viruses by gaseous Ozone. Journal of
Environmental Health 70: 56-62.
UTIYAMA, M. and others 2004. Formation and deposition of ozone in a red pine forest. Water
Air and Soil Pollution 151: 53-70.
VALKAMA, E., J. KORICHEVA, and E. OKSANEN. 2007. Effects of elevated O(3), alone and
in combination with elevated CO(2), on tree leaf chemistry and insect herbivore
performance: a meta-analysis. Global Change Biology 13: 184-201.
VANDONK, E., and J. RINGELBERG. 1983. The Effect of Fungal Parasitism on the
Succession of Diatoms in Lake Maarsseveen-I (the Netherlands). Freshwater Biology 13:
241-251.
VANETTEN, J. L., L. C. LANE, and R. H. MEINTS. 1991. Viruses and Virus-Like Particles of
Eukaryotic Algae. Microbiological Reviews 55: 586-620.
VINGARZAN, R. 2004. A review of surface ozone background levels and trends. Atmospheric
Environment 38: 3431-3442.
VISSER, S. A. 1985. Physiological Action of Humic Substances on Microbial-Cells. Soil
Biology & Biochemistry 17: 457-462.
VOIDAROU, C. and others 2007. Experimental effect of ozone upon some indicator bacteria for
preservation of an ecologically protected watery system. Water Air and Soil Pollution
181: 161-171.

116

VOLK, C., P. ROCHE, J. C. JORET, and H. PAILLARD. 1997. Comparison of the effect of
ozone, ozone-hydrogen peroxide system and catalytic ozone on the biodegradable
organic matter of a fulvic acid solution. Water Research 31: 650-656.
VOLZ, A., and D. KLEY. 1988. Evaluation of the Montsouris Series of Ozone Measurements
Made in the 19th-Century. Nature 332: 240-242.
WANG, W. H., C. M. BRAY, and M. N. JONES. 1999. The fate of C-14-labelled humic
substances in rice cells in culture. Journal of Plant Physiology 154: 203-211.
WEED, D. 2006. Hamilton Lab NPOC/DOC Protocol. Laboratory of Dr. Steve Hamilton,
Kellogg Biological Station, Michigan State University.
WESTERHOFF, P., G. AIKEN, G. AMY, and J. DEBROUX. 1999. Relationships between the
structure of natural organic matter and its reactivity towards molecular ozone and
hydroxyl radicals. Water Research 33: 2265-2276.
WETZEL, R. G. 2001a. Bacterioplankton, p. 490-526, Limnology: Lake and River Ecosystems.
Elsevier Science (USA).
---. 2001b. Light in Inland Waters: Color of Natural Waters, p. 64, Limnology: Lake and River
Ecosystems. Elsevier Science (USA).
---. 2001c. Plankton Communities: Algae and Cyanobacteria, Limnology: Lake and River
Ecosystems. Elsevier Science (USA).
---. 2001d. The Phosphorous Cycle, p. 239-286, Limnology: Lake and River Ecosystems.
Elsevier Science (USA).
---. 2001e. The Nitrogen Cycle, p. 232, Limnology: Lake and River Ecosystems. Elsevier
Science (USA).
---. 2001f. Detritus: Organic Carbon Cycling and Ecosystem Metabolism, p. 731-780,
Limnology: Lake and River Ecosystems. Elsevier Science (USA).
WETZEL, R. G., P. G. HATCHER, and T. S. BIANCHI. 1995. Natural photolysis by ultraviolet
irradiance of recalcitrant dissolved organic matter to simple substrates for rapid bacterial
metabolism. Limnology and Oceanography 40: 1369-1380.
WIDRIG, D. L., K. A. GRAY, and K. S. MCAULIFFE. 1996. Removal of algal-derived organic
material by preozonation and coagulation: Monitoring changes in organic quality by
pyrolysis-GC-MS. Water Research 30: 2621-2632.
WILLIAMSON, J., G. MILLS, and C. FREEMAN. 2010. Species-specific effects of elevated
ozone on wetland plants and decomposition processes. Environmental Pollution 158:
1197-1206.

117

WONG, S. L., and B. CLARK. 1976. Field determination of hte critical nutrient concentrations
for Cladophora in streams. J. Fish. Res. Board Can. 33: 85-92.
WONISCH, A., M. TAUSZ, M. MULLER, G. SOJA, and D. GRILL. 2003. Ozone-induced
long-term effects on chromosomal aberration rates in root-tip meristems of spruce trees
do not correspond to changes in tissue antioxidant status. Phyton-Annales Rei Botanicae
43: 147-160.
XING, J. and others 2011. Projections of air pollutant emissions and its impacts on regional air
quality in China in 2020. Atmospheric Chemistry and Physics 11: 3119-3136.
YAMAJI, K., R. JULKUNEN-TIITTO, M. ROUSI, V. FREIWALD, and E. OKSANEN. 2003.
Ozone exposure over two growing seasons alters root-to-shoot ratio and chemical
composition of birch (Betula pendula Roth). Global Change Biology 9: 1363-1377.
YU, M. J., Y. H. KIM, I. HAN, and H. C. KIM. 2002. Ozonation of Han River humic
substances. Water Science and Technology 46: 21-26.
YUN, Y. S., S. R. LIM, K. K. CHO, and J. M. PARK. 1997. Variations of photosynthetic
activity and growth of freshwater algae according to ozone contact time in ozone
treatment. Biotechnology Letters 19: 831-833.
ZAMBRANO, A., and T. H. NASH. 2000. Lichen responses to short-term transplantation in
Desierto de los Leones, Mexico City. Environmental Pollution 107: 407-412.
ZAMBRANO, A., T. H. NASH, and C. GRIES. 1999. Physiological effects of the Mexico city
atmosphere on lichen transplants on oaks. Journal of Environmental Quality 28: 15481555.
ZHANG, R., X. TIE, and D. W. BOND. 2003 Impacts of anthropogenic and natural NOx
sources over the U.S. on tropospheric chemistry. Proceedings of the National Academy
of Sciences of the United States of America 100 1505-1509
ZHANG, Y. Q., Q. P. WU, J. M. ZHANG, and X. H. YANG. 2011. Effects of ozone on
membrane permeability and ultrastructure in Pseudomonas aeruginosa. Journal of
Applied Microbiology 111: 1006-1015.

118