 

KEV
Eu

.3,
A

5%
£516.?»

{2
g:
.-

‘i
ii
35

a. a 1.; $23..
a; 2:353)...

. n.2,": 13......
f9”

:3! ,
in.
.l

1":
i

..
.

E: .
t .
. .4
I:
u . (trio-l
nu... _ in.
t. .9. hi
In“
s .3 3.5.2
1‘ :ﬂ iii
3.
5
9...“:
.3...
.3...
3

'3
2’5»

I31... .
.. .3 .

:3

. : 353w"... .
3m: 7 a»

ME... ..

g2

 

THESIS

.2 LIBRARY
.10 o 4 Michigan State
University

This is to certify that the
dissertation entitled

ECOLOGICAL STRATEGIES OF BENTHIC DIATOMS FOR
NUTRIENT COMPETITION

_ _ —-_.--n-o-I-l-l-l-I-I-

presented by
KALINA M. MANOYLOV

has been accepted towards fulﬁllment
of the requirements for the

Ph.D. degree in Zoology and Program in

Ecology, Evolutionary Biology
and Behavior

(7?.489,

Major Professor’s Signature
We) 8 4’20 or
C]

Date

 

 

 

MSU Is an Afﬂnmtiw Action/Equal Opportunity Institution

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 ch'Iﬁcmatm.m-p.1s

 

ECOLOGICAL STRATEGIES OF BENTHIC DIATOMS FOR NUTRIENT
COMPETITION

By

Kalina M. Manoylov

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Zoology
Program in Ecology, Evolutionary Biology & Behavior

2005

ABSTRACT

ECOLOGICAL STRATEGIES OF BENTHIC DIATOMS FOR NUTRIENT
COMPETITION

By

Kalina M. Manoylov

Diatoms are an important component of stream benthic communities and
depend on nutrients and light, which are often limiting resources. The effects of
different nutrient concentrations on benthic diatoms were examined using ﬁeld
and laboratory experiments and observations in natural habitats. The ﬁeld
mesocosm experiment was conducted to determine growth rates of benthic
diatoms in limiting versus saturated nutrient conditions. In this experiment,
pristine stream water was supplied continuously to artiﬁcial streams, and nutrients
were manipulated to assess effects of nitrogen (N) and phosphorus (P) additions
on algae in near-natural stream conditions. Many diatoms were able to grow at
very low N and P concentrations. Of the 14 most common species the same
species grew fastest in low N and low P conditions. No evidence indicated that
there were tradeoffs in species abilities (measured as growth rate) to compete for
N and P at low resource levels, or that there were any tradeoﬁ‘s for highest
performance in low and high resource conditions.

Intra- and inter-speciﬁc competition of clonal populations of
Achncmthidium minutissimum and two potential competitors (Cocconeis
placentula var. Iineata and Cymbella cistula) were examined in laboratory

experiments. In circular artiﬁcial channels, four treatment combinations of

nutrients and light were created. When light was limiting, nutrient addition did
not increase density. Biomass (as total cell surface area) of A. minutissimum
increased when both nutrients and light were added. Intraspeciﬁc competitive
effect rather than interspeciﬁc competitive effects regulated A. minutissimum
growth rates. Cocconeis placentula var. Iineata grew very slowly in cultures.
Cymbella cistula grew well in culture, formed stalks, and overgrew A.
minutiss'imum. Facilitation by C. cistula stalks, providing substratum for
colonization, may have countered interspeciﬁc competition on A. minutissimum.
Diatom species abundance depends on suitability of ecological conditions
and species’ ability to disperse. A positive correlation was observed between
diatom abundance and distribution of diatom species. Achnanthidr’um
minutissimum occurred at all sites and had the highest relative abundance.
Relative abundance of the most common species in pristine streams, such as A.
minutissr'mum, Achnanthes deﬂexa, C ymbella aﬂinis, and Gomphonema
angustatum decreased with increasing nutrients and chloride concentrations,
however they persisted in habitats with high human disturbance. Taxa requiring
high nutrients were more diverse than were taxa from pristine streams, but even
the high-nutrient taxa with widest distributions occurred at low abundance. Some
taxa requiring high nutrients grew faster than taxa from pristine streams when
streams were anthropogenically enriched, which caused, partial turnover of the
taxa in the natural diatom assemblage that was adapted to low nutrients.
Evolutionary strategies of persistence of species adapted to low nutrients were

evident in their wide-spread regional distribution.

Tomyhusbcmd

iv

ACKNOWLEDGEMENTS

This work is dedicated to my husband Vladislav I. Manoylov, who as a scientist
always believed in me, and helped me with encouragement and ideas. I am in
debt to my children, son Ivan and daughter Mariah, for their understanding and
unconditional love. I am grateful to my parents and in-laws who helped me
tremendously coming from Bulgaria every time I needed them. Completion of
this research would not have been possible without the support and assistance
from my adviser R Jan Stevenson. Dr. Stevenson ensured my full time
employment over the years, ﬁnanced the experimental designs and allotted time in
his busy schedule for discussions of my ideas and research. I am very grateﬁrl to
my committee members Thomas Burton, Donald Hall and Gary Mittelbach for the
advice and scientiﬁc discussions, which helped with the development and
realization of this project. I am grateful and proud knowing the many current and
former students, technicians and postdocs in the Stevenson Algae Lab: Andrew
Alverson, Cathleen Craft, Carrie DeJaco, Susanna DeCeles, Julianne Heinlein,
Vanessa Lougheed, Yangdong Pan, Lara Panayotoff, Christian Parker, Agnieszka
Pinowska, Steve Rier, Scott Rollins, Rich Schultz, Jennifer Slate, Tom Smith,
Roger P. Sweets, Mark Vanderborgh, Brian Wade, Yr-Kuang Wang, Sarah Wolf,
Major Waltman and Lei Zheng. Many of those individuals helped me shape and

clear my ideas, and remain my true friends.

TABLE OF CONTENTS

LIST OF TABLES ...................................................................... .viii
LIST OF FIGURES ........................................................................ xi
INTRODUCTION ............................................................................. 1
Theoretical background and historic perspective ........................................ 2
Experimental work ........................................................................... 7
Taxonomic perspective ...................................................................... 8
Statement of research goals ................................................................ 11
Literature Cited .............................................................................. 14
CHAPTER 1
ALGAL RESPONSES IN N AND P NUTRIENT GRADIENTS IN ARTIFICIAL
STREAMS ................................................................................... 21
Introduction ................................................................................. 21
Material and methods ...................................................................... 24
Study area .......................................................................... 24
Experimental design ............................................................... 25
Sample collection ................................................................. 27
Data analysis ....................................................................... 29
Results ........................................................................................ 33
Physicochemical characteristics ................................................. 33
Community response in N and P gradient ...................................... 34
Diatom species responses and evidence for tradeoffs ........................ 36
Discussion ................................................................................... 39
Literature Cited ............................................................................. 46
CHAPTER 2
INTRA AND INTERSPECIFIC COMPETITION TO NUTRIENT
MANIPULATIONS IN CULTURES .................................................... 73
Introduction ................................................................................. 73
Material and methods ...................................................................... 77
Experimental design ............................................................... 78
Competitive effect ................................................................. 82
Results ........................................................................................ 84
Discussion ................................................................................... 87
Literature Cited ............................................................................. 92
CHAPTER 3
DIATOM COMMUNITY STRUCTURE FROM NATURAL STREAMS IN
KENTUCKY .............................................................................. 115
Introduction ................................................................................ 1 15
Material and methods ..................................................................... 120

Laboratory assays ................................................................. 121

Experiment ........................................................................ 123
Data analysis ...................................................................... 126
Results ...................................................................................... 129
Species abundance-distribution relationship, sampling error, aggregation,
and dispersal ...................................................................... 129
Relation of core-satellite species to ecological conditions ................ 130
Discussion ................................................................................. 134
Literature Cited ........................................................................... 141
GENERAL CONCLUSIONS ........................................................... 164
Literature Cited ............................................................................ 171
APPENDICES ............................................................................ 172

vii

LIST OF TABLES
CHAPTER 1

Table 1.1: Summary of repeated measures (ANOVA) results of individual taxa
growth rates in N treatment when P saturated at 50 pg L'l. S-nutrient effect, T-
time eﬂ‘ect, and SxT interaction ........................................................... 54

Table 1.2: Summary of repeated measures (ANOVA) results of individual taxa
growth rates in P treatment when N saturated at 500 pg L". S-nutrient effect, T-
time effect and S x T interaction ......................................................... 55

Table 1.3: Growth parameters estimated with Monod model along N gradient. . .56
Table 1.4: Growth parameters estimated with Monod model along P gradient. . .57
CHAPTER 2

Table 2.1: Medium used for culture growth ............................................ 98

Table 2.2: Culture combination of the initial communities (n=15). Species cell
biovolume (pm’) (mean :1: SE) and cell biosurface mm2 (mean i SE) ............ 100

Table 2.3: Repeated measures ANOVA for differences in growth rate of
Achnanthidr‘um minutissimum when grown in different case combinations: case A
- A. minutissimum only; AA- A. minutissimum in double initial density; with
Cocconeis placentula var. Iineata (AC); and in combination with C. cistula
(ACM). Between treatment effect — treatment combination and time and
treatment interaction tested as within channel effect. For each effect: F statistics,
df-degrees of ﬁ'eedom, and p- value adjusted for circularity ....................... 101

Table 2.4: Competitive effects of diatoms on Achnanthidium minutissr’mum
(target). Intraspeciﬁc competition (column 0,, ). Intraspeciﬁc competition input
from the neighbors are given under other (competitive indices estimations of the
neighbor species in mixed cultures, 00c ac"). Values are the intercept (maximum
growth rate, GR“) and regression coeﬁicient ‘a’s from the linear regression
analysis, Growth rate as a function of B- total neighbor biosurface. Growth rate of
Acimanthidr‘um minutissimum when grown in different case combinations: case A
- A. minutissimum only, without competition; AA - A. minutissimum in double
initial density; AC -with Cocconeis placentula var. Iineata); and ACM with
Cymbella cistula. Indices estimates given with standard error given (SE), slopes
different ﬁ'om O are given in bold ...................................................... 102

CHAPTER 3

viii

Table 3.1: Taxa names and abbreviation for diatoms found in the survey of all 69
sites and 8 reference sites; N-ascending number of sites; mean RA-mean relative
abundance ................................................................................. 148

Table 3.2: Test of the inﬂuence of sampling on the abundance—distribution
relationship for Kentucky stream diatoms. Species-speciﬁc coefﬁcient of
variation (CV) and average abundance (xm) were calculated across all sites. SE-
standard error, p — signiﬁcance level ................................................... 152

Table 3.3: Effect of species dispersal abilities on abundance and distribution
relationship for Kentucky stream diatoms ............................................. 153

Table 3.4: Physicochemical measurements in reference and non-reference sites.
p= signiﬁcance level ..................................................................... 154

Table 3.5: Correlations between physicochemical variables in reference and non-
reference streams; (*) signiﬁcance level p<0.05 ...................................... 155

Table 3.6: Community attributes in reference and human disturbance sites. p=
signiﬁcance level ......................................................................... 156

Table 3.7: Diatom species average similarity (aveS) in reference sites,
contribution from each species; species are ordered in decreasing contribution
with 90 % cumulative contribution (Cum %) as a cut-off percentage; (SD)S —
standard deviation of the S (ith species) values; aveS/SD(S ) - discriminating
ratio, highest in bold; {(aveS) % - average percent contribution to similarity by
the (ith species) ........................................................................... 157

Table 3.8: Diatom species average similarity (aveS) in non-reference sites,
contribution from each species; species are ordered in decreasing contribution
with 90 cumulative contribution (Cum %) as a cut-off percentage; (SD)S —
standard deviation of the S (ith species) values; aveS/SD(S ) - discriminating
ratio, highest in bold; 2(aveS) % — average percent contribution to similarity by
the (ith species) ............................................................................ 158

Table 3.9: Diatom species average dissimilarity (ave 6) between reference and
non-reference sites, contribution ﬁ'omeach species; species are ordered in
decreasing contribution with 90 % cumulative contribution (Cum %) as a cut-off
percentage; meanRAnr- mean relative abundance in non-reference sites, compare
with mean RA in reference sites Table 3.4; (SD)6 - standard deviation of the 6
(ith species) values; ave6/SD(6 ) - discriminating ratio, highest in bold; 2(ave5)
% - average percent contribution to similarity by the (ith species) ................. 159

Table 3.10: Effect of ecological specialization on the average abundance and

distribution of Kentucky stream diatoms. The table gives means i standard
error ...................................................................................................................... 160

ix

APPENDICES

APPENDD( I: Achnanthidr’um minutissimum distribution. Habitat and locality
ﬁom selected records (more than 10% reported abundance) in USA Source:
General Collection (GC) at the Academy of Natural Sciences, Philadelphia,

PA ........................................................................................... 173

APPENDIX 1.1: Nutrient concentrations pg L" along N gradient (mean :1: SD,
n=3) ........................................................................................ 174

APPENDD( 1.2: Nutrient concentrations pg L'1 along P gradient (mean :t SD,
n=3) ......................................................................................... 175

APPENDD( 3.1: Site locality and identiﬁcation sampled for 1996 and 1997, sites
in bold are considered reference ........................................................ 17 6

LIST OF FIGURES
CHAPTER 1
Figure 1.1: Schematic view ofeach partially re-circulating channel. .. . .. ...........58

Figure 1.2: Algal biomass (Chlorophyll a) change at the end of the experiment
along the a. N gradient; b. P gradient Bars indicate standard error of the treatment
means for all times ......................................................................... 59

Figure 1.3: Total diatom community density (cells cmz). a, c. Nitrogen gradient;
b, (1. Phosphorus gradient; a, b mean diatom density per treatment; Bars indicate
standard error per treatment. c, (1. Density with nutrient and time; D3-D15 days of
sampling, similar symbols represent replicates; lines connect the means .......... 60

Figure 1.4: Whole community growth rates (cells.cell".d”). N and P gradients;
Ro.3-change in growth between day 0 and day 3, small dotted line and circles; R3.
6-dashed line and x; Ram-solid line and plus; R1043 dash, dotted line and up
triangles; R1345 larger dotted line and down triangles; lines connect the means...62

Figure 1.5: Mean growth rates Low N and Low P. a Time Ro.3 r = 0.70 (p=0.01);
b. Time R34, r = 0.19 (n.s.); c. Time R540, r = 0.53 (p=0.044); d. Time R1043 r =
0.16 (n.s.); e. Time R1346 r = -0.04 (n.s.); r - Spearman correlation coefﬁcient;
given p values represent signiﬁcant correlation at 0.05 level (2-tailed), n=14 for
all time periods. Taxa abbreviations as in table 1.5 ................................... 63

Figure 1.6: Correlation values for performance in limiting nutrient concentration
for all 5 sampling periods. a. Low N and Low P. b. Ratios normalized Low N and
Low P values with grth at preferable condition. Dotted line separates positive
and negative values ........................................................................ 68

Figure 1.7: Correlation values for maximum growth estimated with the Monod
equation in both gradients for all the 3 sampling times prior to peak biomass.
Tradeoffs measure for N (o) and P (x). Dotted line separates positive and negative
values ........................................................................................ 69

Figure 1.8: Relationships between maximum observed grth rates in high N and
in low N conditions. a. Time Ro.3, r = 0.66 (p=0.022); b. Time R36, r = 0.40 (as);
c. Time R640, r = -0.22 (n.s.); (1. Time R1043 r = 0.32 (n.s.); e. Time R1346 r = 0.21
(n.s.); r - Spearman correlation coefﬁcient; given p values represent signiﬁcant
correlation at 0.05 level (2-tailed), n=14 for all time periods ........................ 70

Figure 1.9: Relationships between maximum observed grth rates in high P and

in low P conditions. a. Time Ro-3, r = 0.29 (n.s.); b. Time R345, r = - 0.28 (n.s.); c.
Time R640, 1:052 (p=0..044) (1. Time R1043 I' = 0.50 (p=0.05); e. Time R1345 I’ =

xi

0.36 (n.s.); r - Spearman correlation coefficient; given p values represent
signiﬁcant correlation at 0.05 level (2-tailed), n=l4 for all time periods ........... 71

Figure 1.10: Correlation values for maximum observed growth along the two
gradients for all the 5 sampling times with increase in density. Tradeoffs measure
for N (o) and P (x). Dotted line separates positive and negative values ............ 72

CHAPTER 2

Figure 2.1: Schematic representation of the closed experimental circular channels;
a. whole view; b. side view, circular channel diameter = 3 cm; c. view from top,
where tile arrangement can be seen. Arrow indicates flow direction (ﬂow = 5
cm.s’1) ....................................................................................... 104

Figure 2.2: Nutrient concentrations measured (pg/L); (a.) N and (b.) P values in
the control and NP treatments (n=4). Error bars around the means represent
standard error .............................................................................. 105

Figure 2.3: Total community density (Ln cells mmz) in shaded (D) and full light
(L) regimes. a. in low nutrients (C); b. in high nutrients (NP). Error bars around
the means represent standard error ..................................................... 106

Figure 2.4: Achnanthidium minutissimum biosurface (106 total cell surface area,
mm'z) in each treatment combination. Bars (means :1: standard error) with the
same letters are not signiﬁcantly different ............................................. 107

Figure 2.5: Growth rate of Achnanthidium minulissimum in culture combination.
a. DC; b. DNP; c. LC; d. LNP; open bars- initial growth (difference in days of
community development 0 to 3); shaded bars high biomass grth (difference in
days of community development 6 to 9; RM ANOVA polynomial test result,
signiﬁcant time period change); Bars (means i standard error) with the same *
(for the initial growth) or same letter (for the high biomass growth) not
signiﬁcantly different ..................................................................... 108

Figure 2.6: Growth rate relationships of Achnanthidium minutissimunr per
biomass estimate in culture combination (Case); A- dashed line, AA-dashed &
dotted line, AC - dotted line, and ACM-solid line in each treatment combination;
a. DC; b. DNP; c. LC; (1. LNP .......................................................... 110

Figure 2.7: Intraspeciﬁc competition indices (or) estimates in high biomass of the
target species Achnamhidium minutiss'imum in each culture and treatment
combination. Bars (model slope estimates, means :h standard error) .............. 112

Figure 2.8: Interspeciﬁc competition indices (0.) estimates in high biomass of
neighbor species a. Cocconeis placentula var. Iineata; b. Cymber cistula in each

xii

culture and treatment combination. Bars (model slope estimates, means i standard
error) ....................................................................................... l 14

CHAPTER3

Figure 3.1: Relationship between diatom species abundance and distribution (as
percent of sites occupied) in the survey for all sites (n=69); a. A scatter plot of
mean relative abundance for the sites occupied; b. A histogram number of species
and percent sites occupied (second y axes represents the proportion of taxa in the
bars) against their distribution .......................................................... 161

Figure 3.2: Species growth rates in ambient and high NP experimental channels.

Dashed line is the 1:1 line; dotted lines separates positive and negative values.
Abbreviations as in Table 3.1 ........................................................... 163

xiii

INTRODUCTION

As suggested by Darwin (1859), complexity of biological systems can be
observed in many habitats and conditions. Multispecies benthic algal
communities present a great opportunity to understand the role individual species
play in structuring complex communities. Benthic algae contribute a large
portion of primary production in aquatic systems (Parsson er al. 1977), attain high
biomass (Moss 1968, Hansson 1992) and are a major food source for grazers
(Lamberti et al. 1989). Since they are the most abundant and diverse algae in
periphyton (Round et al. 1990), more research is needed on diatom species and
periphyton community natural history. My research contributes to the knowledge
of algal diversity and species performance as shaped by their response to different
nutrient conditions. This species-speciﬁc information is needed since complete,
unidirectional species turnover is possible with global nutrient contamination
(National Research Council 2000). My goal is to understand if morphological
characters and species traits are the result of evolutionary adaptations, ecological
strategies or both.

Algae have been used in testing many important ecological theories
(Patrick 1967, 1968, Porter 1973, Tilman 1976, 1977, Stevenson 1983, Pringle
1990, McCormick and Stevenson 1991). Lotic algae in particular are very diverse
and sensitive to environmental changes (Pan et al. 1996). The ﬁrndamental
questions in algal benthic ecology in streams are similar to ﬁrndamental questions

in general ecology. What abiotic conditions drive benthic community

development and species diversity? Are there species-speciﬁc interactions that
play a major role in structuring benthic communities? How do some species have
great ecological success and dominate a community in one environment, while
remain rare in another for a long time? Ultimately, similar to the question posed
by Hutchinson (1961): why are there so many taxa in the benthos, when there are
only a few limiting resources?

Theoretical background and historic perspective

Environmental heterogeneity leads to the evolution of species ecological
strategies that are important for survival under different conditions (Southwood
1977). Evolutionary trade—offs in species’ ability to maximize performance under
diﬁ‘erent ecological conditions are a part of those ecological strategies
(McCormick 1996). A trade-off according to Tilman (1976, 1977, 1982) is the
inverse correlation between species’ competitive ability and its capacity to
maintain populations under suboptimal conditions (Grime 1979, Keddy 1990).
Understanding ecological and evolutionary strategies in diatoms will allow '
predictions in population dynamics and community development at different
environmental scales.

Algae depend on nutrients and light for survival. Nitrogen (N) and
phosphonrs (P) are important nutrients in development of algal communities
(Schindler 1974, Hecky and Kilham 1988, Elser er al. 1990) and have unique
biochemical roles in cell structure and function (Borchardt 1996). Algae increase
in density and abundance with an increase of both nutrients (Stevenson and

Glover 1993). At very low concentrations of N and P, non-equilibrium

competition results in high species diversity and biomass (Botlrwell 1988, 1989,
Grover 1991). During periods of intense competition for a limiting resource,
tradeoffs in species-speciﬁc traits like nutrient uptake ability, reproduction rate,
grazer resistance, or dispersal ability are hypothesized because the species that are
evolutionarily successful will have allocated most resources and energy to the
ﬁtness trait that enables species survival. Species that allocate energy and
resources to several traits would, theoretically, be unsuccessful (competitively
excluded) because of lack of specialization (Tilman 1976, 1977, Stevenson and
Pan 1994). Therefore, competition for resources may restrict a species’ allocation
of resources to sets of coadapted traits that confer ﬁtness during periods of intense
competition for a limiting resource (e.g. nutrient, space, or light).

Lotic algae compete for nutrients and there is compelling evidence that
ﬂowing water modiﬁes nutrient delivery, perhaps differentially among algal
species (Peterson and Stevenson 1992). Algae in streams have access to more
nutrients and have greater growth rates than algae in slow moving water
(Stevenson and Peterson 1989). Borchardt (1996) reported that aspects of the
benthic grth habitat, especially exposure to greater water motion and the
formation of mats have ramiﬁcations for nutrient uptake, utilization, recycling and
competition that are unique to benthic algae. Therefore, accounting for and
controlling these conditions may provide a basis for elucidating the mechanisms
and consequences of ﬂowing water to periphytic algal species performance.
Stream benthic algae are adapted to a punctuated supply of resources (Biggs

1996).

Algal resource theory and trade-offs (Tilman 1982), have been developed
for phytoplankton and have been successful in predicting competitive interactions
and successional trajectories among phytoplankton species in lakes (Sommer
1985). Resource trade-offs were also proposed for individual species
competition for low nutrients in benthic communities (McConnick 1996) and
have been tested with diatom periphyton communities in ﬂasks (Stevenson and
Pan 1994), green algae in streams (Borchardt 1996) and among marine benthic
species (Sommer 1996). Research with benthic algae demands consideration of
factors like attachment, scour resistance, irregular nutrient ﬂow, and intensive
grazing. Therefore, in many cases results ﬁ'om phytoplankton and chemostat
research cannot be applied directly to benthos. Flow acts as a disturbance agent
(Biggs 1995), affects algal colonization, dynamics and immigration patterns
(Stevenson and Peterson 1989), and enhances metabolic processes (Stevenson
1984). The ecological and evolutionary role of resource competition in benthic
algal assemblages has yet to be critically explored (McCormick 1996).

Diatom communities are dynamic structures, where time is an important
component in their development. Patrick (1976) related the succession of benthic
diatoms on substrata to Clementsian forest community dynamics in which low
prostrate species conditioned substrata and were followed by apically attached
and stalked species. On the other hand, Borchardt (1996) suggested that benthic
communities were more complex, with exposure to greater water motion and the
formation of mats having ramiﬁcations for nutrient uptake, utilization, recycling,

and competition that were unique to benthic algae. Successional community

structure and functions inferred by individual species requirements and tolerance
to environmental conditions have not been addressed to date. As in plant
communities, changes in the quality and quantity of periphytic algae are known to
be inﬂuenced by autecological and biological conditions (Pickett et al. 1987,
Stevenson 1996).

The concept of ecological strategies is based on the assumptions that
speciﬁc traits confer different survival advantages, traits have a signiﬁcant energy
and resource costs, and these traits require evolutionary choices. Thus, species
evolve different sets of traits and become specialized to either live in different
habitats or use different energy and material resources within a habitat. Two
resource-based strategies for survival are often referred to as r- and K- selection
strategies (Biggs et al. 1998). With K-selection strategies, species specialize to
use resources in very low supply, persist after resources have been exploited, and
grow slowly but efficiently in low resource environments. Species could
theoretically coexist in low resource environments if they reﬁned traits to use one
of several potential resources more effectively in a habitat than another species.
With r-selection, species grow rapidly, exploit resources, and produce many
offspring when resources are in great supply, and then disperse to new
environments where resources are abundant as resource supplies at “home”
become limiting. These strategies for benthic algae in streams have been
hypothesized during the frequent disturbance and re-colonization stages following
rains and ﬂoods that scour algae off the stream bottom to restart the colonization

process with renewed supplies of nutrient resources (McCormick 1996).

Insight into species preferences and tolerances can be gained with
assessment of ecological preferences and distribution of algae over geographic
space (Kociolek and Stoermer 2001). Individual traits and ecological strategies
determine species regional abundance and distribution (Collins and Glenn 1991).
Traditionally, ecological inferences have been constrained to dominant algae in
streams (Pan et al. 1996), while rare taxa ecology and taxonomy remains unclear.
In his classiﬁcation of species as dominant, subordinate, and transient, Grime
suggests that ecosystem functions are determined by the traits of the dominant
populations (Grime 1998). The core-satellite hypothesis (Hanski 1982) evaluates
rare and dominant taxa simultaneously. It addresses similar ecosystem functions
as Grime classiﬁcation, and relates species abundance to regional distribution.
Species are divided into core, intermediate and satellite, where the abundance of
core (dominant and common) and satellite (rare and sparse) species relates to the
environmental characteristics in each habitat. In both classiﬁcations the
intermediate (subordinate) species are spatially dynamic (Hanski 1982) and
function as biological ﬁlters during community assembly (Grime 1998) with a
potentially important, but unclear function. Ifspecies abundance is related to
species growth in speciﬁc nutrient conditions, species abundant in low nutrient
conditions (reference, native conditions) will be replaced at high nutrient
condition (anthropogenically altered).

Experimental work
In nature, environmental variability is composed of numerous

physicochemical factors and species interactions (Tilman 1982). Many times

partitioning only a few variables that contribute to the community response is
inconclusive. Experimental research allows the creation of simpler systems,
where different factors can be manipulated and environmental components can be
partitioned. Results from experimental studies can provide mechanistic
explanation of the events in nature. Experimental studies sometimes are hard to
generalize (MacArthur 1972), but the beneﬁt of relating speciﬁc environmental
conditions to populations and communities is invaluable in understanding natural
systems. In periphyton, results from experimental manipulation were comparable
to results of periphyton growth in natural streams (Kevem et al. 1966).
Additionally, experiments allow study of the dynamic properties of communities
in a timely manner, i.e. processes that take months and years can be shortened in
experimental studies.

In phycological studies, experimental manipulation of phosphorous and
nitrogen are often chosen (F airchild et al. 1985, Stevenson 1995) because either
one or both nutrients are considered limiting in natural streams (Grimm and
Fisher 1986, Borchardt 1996). Moreover, the effect of a given nutrient
concentration, as expressed through growth rate and abundance, can be a measure
of species performance at those conditions (Stevenson 1995). Artiﬁcial streams
and substrata are used widely, because they help standardize environmental
variables such as light and current, when nutrients are manipulated (Stevenson et
al. 1991, Tuchman 1996).

Working in a simpliﬁed system could be beneﬁcial in unveiling and

understanding species interactions in diatoms with controlled variables. Algal

biodiversity has been shaped by competition for nutrients, and there is concern
that global nutrient contamination and eutrophication will change algal species
membership. With a closer look at community grth within a nutrient gradient,
at optimal light and temperature and without grazers, one can evaluate the
dynamics within a population. Size, growth rates and density-dependence are
evolutionary attributes that might give advantage in changing nutrient conditions.
Experimental evidence (Austin and Austin 1980) has shown that varying
intensities of competition modify the ecological response curves of species in
vegetation.
Taxonomic perspective

Taxonomic identiﬁcation of algae presents opportunities to understand
systematic entities like species (Mayr 1942) and to relate them to evolutionary
and ecological processes. Correct taxonomic identiﬁcation provides consistency
and transferability of ecological inferences (Kociolek and Stoermer 2001).
Species size and distribution limits are important in addressing ecological
questions. In order to understand periphyton ecology, diversity attributes, like
identity and number of species, growth form, and reproduction, have to be taken
into consideration. In streams, rapid growth and dispersal abilities have been
reported as advantageous strategies in heterogeneous environments (Patrick
1976)

Morphologically outlining species identity requires careful research on
diatom species biology. For example, a very common diatom Achnanthidr‘um

minutissimum (Kiitzing) Czarnecki has been reported as widespread and abundant

in North America (Patrick and Reimer 1966, Villeneuve and Pienitz 1998, Laing
et al. 1999). In the General Collection at the Academy of Natural Sciences in
Philadelphia, there are records ﬁom more than 2500 localities in the United
States, where at the time of collection, abundance of Achnanthes minutissima
Kiitzing was more than 10%. In the subset of slides that I evaluated, the types of
water bodies, different habitats, environmental conditions and times of collection
varied (Appendix I), but all slides contained A. minutissimum. Similarly, wide
distribution in variable environments was reported in a quantitative study from
1109 rivers in the US. (U .8. Geological Survey, National Water-Quality
Assessment Program NAWQA). A. minutissimum was reported from 79% of the
2674 samples counted (Potapova and Charles 2003).

In the literature, A. minutissimum is reported as indifferent to nutrients in
Europe (van Dam et al. 1994). This wide distribution and no indicator value may
potentially be due to the fact that this morphological entity potentially contains
several different ecodemes able to grow in diverse conditions. Ecophenotypical
variation in diatoms has been recognized and reported in seasonal sampling
(Kociolek and Stoermer 1988), as well as within genus variation in response to
nutrient variability (Geissler 1982). Species-speciﬁc phylogenetic variation in
response to resource gradients has also been reported (Kilham et al. 1996).

Taxonomically, research in diatoms has been difﬁcult because of historical
approaches with different objectives. Originally, species were described to ﬁrlﬁll
species inventory objectives (Ehrenberg 1843, Hustedt 1967, Patrick and Reimer

1966, 197 5) rather than as a part of population studies where ecological optima

could be differentiated and linked to morphological entity. Part of the uncertainty
comes from the original description. Originally, Kl'itzing (1833, 8:578, pl. 16, ﬁg.
54) stated only that Achnanthes mintutissima could be four times smaller than
Achnanthes einis Kiitzing. In the line drawing, all Kutzing showed were girdle
views of curved valves on stalks. Later, Lange-Bertalot (in Krammer and Lange-
Bertalot 1991), based on observations of Kiitzing’s type material, added
characters to the vague original description such as rounded ends, a straight distal
raphe and striae that became ﬁner towards the end. Adding more conﬁrsion,
Lange-Bertalot stated that the lectotypes he had chosen for A. minutissima
Kiitzing and A. microcephala Kﬁtzing (a completely different diatom until that
point) were the same. As a result, morphometric data for routine identiﬁcation
are either not strictly deﬁned (Achnanthes minutissima ‘ Sippen’-complex in
Krammer and Lange-Bertalot 1991, size length 5-25 pm usually less than 20,
width 2.5 to 4 pm) or well deﬁned, but with wider size ranges (Hustedt 1930,
Patrick and Reimer 1966, length 5-40 pm).

Achmmthidium minutissimum (the current name) is an example of the need
for interdisciplinary research that combines natural history knowledge,
observations and experiments to better elucidate taxonomy. Therefore, diatom
taxonomy is evolving and changing rapidly as types and original descriptions are
tested with currently existing populations from North America (Morales 2001,

2002, 2003 a, b, Manoylov et al. 2003, Potapova and Ponader 2004).

10

Statement of research goals

The goals of my research were to examine the effect of nutrients on
benthic diatoms in controlled experimental settings and in nature. Benthic
diatoms were identiﬁed to the lowest taxonomic level. Diatom abundances and
morphologies were hypothesized to be a result of ecological and evolutionary
processes. Species-speciﬁc evolutionary adaptations that could help ecological
success of one algal species over another were characterized. Those strategies
were determined in a ﬂow-through, strearn-side, nutrient experiment, tested in the
laboratory with a culture-competition study, and conﬁrmed with a study of
regional distribution of diatoms in natural streams. In each of the three studies, I
concentrated on the species that performed well in the ﬂow-through experiment,
which included many algal species ﬁ'om the region.

In my ﬁrst study (Chapter 1), different species were expected to be best
adapted to live in low N and low P concentrations because of specializations for
sequestering those nutrients, when N and P are in low supply. I also intended to
determine whether trade-off strategies existed among different adaptive traits and
whether species traits were complementary in competition for nutrients. The most
important variables in the development of the benthic community are still unclear.
Finding ﬂowing water with known nutrients, constant light, and no grazers is
impossible in nature. I independently studied either N or P gradients, when the
other nutrient was at saturated concentration. This approach is a novel one
because conditions are usually characterized broadly as oligotrophic, mesotrophic

or eutrophic.

11

Diﬂ‘erent species were expected to have highest performance (high
immigration, growth and reproduction rates, low death rates, and high resistance
to sloughing) in low N and low P environments. These species-speciﬁc traits
were hypothesized to lead to signiﬁcant differences in community structure in low
N environment than in low P environment. Where nutrients in low supply are the
prime determinant of algal species selection, interspeciﬁc competition is the most
signiﬁcant factor shaping algal communities (Stevenson and Pan 1994). The
goals of the ﬁrst study were to determine whether trade-off strategies existed
among different adaptive traits and whether species traits were complementary in
competition for nutrients, i.e. to assess growth rates at low N and low P conditions
compared to growth rates in low vs. high conditions of the same nutrient.

During the artiﬁcial stream experiment, I developed an additional study
non-clonal single species populations were compared in different environments
and morphological differences were observed and the results have been published
(Manoylov and Stevenson 2004). As a result, I decided to study clonal diatom
population growth in different environmental conditions and to address biotic
interactions. In the second project (Chapter 2), the effects of different nutrient
conditions on grth rates of genetically identical populations were studied.
Intra- and interspeciﬁc competition in four culture combinations under different
conditions of variable nutrients and light were assessed. Reduction in competitive
interactions was expected with resource availability. For benthic algae, intra- and
interspeciﬁc competition indices have not been determined in cultures under

different nutrient conditions.

12

Diatom community composition in natural streams with different nutrient
conditions were evaluated and related to their abundance and distribution within
an ecoregion. In the survey study (Chapter 3), algae growing in pristine streams
were compared with algal communities growing where human inﬂuence was
large. The ﬂow-through, culture and natural streams were limited to the algal

species from Kentucky (Mid-Atlantic Eco-region).

13

LITERATURE CITED

Austin, M. P. & Austin, B. O. 1980. Behavior of experimental plant communities
along a nutrient gradient. J. Ecol. 68: 891-918.

Biggs, B. J. F. 1995. The contribution of disturbance, catchment geology and land

use to the habitat template of periphyton in stream ecosystems. Freshwat.
Biol. 33: 419-438.

Biggs, B. J. F. 1996. Patterns in benthic algae of streams. In Stevenson, R J.,
Bothwell, M. L., Lowe R. L. [Eds] Algal Ecology. Academic Press, San
Diego, pp. 31-56.

Biggs, B. J. F., Stevenson, R. J. & Lowe, R L. 1998. A habitat matrix conceptual
model for stream periphyton. Arch Hydrobiol. 143(1): 21-56.

Borchardt, M. A. 1996. Nutrients. In Stevenson, R. J ., Bothwell, M. L., Lowe R.
L. [Eds] Algal Ecology. Academic Press, San Diego, pp. 183-227.

Bothwell, M. L. 1988. Growth rate responses of lotic periphytic diatoms to
experimental phosphorus enrichment: the inﬂuence of temperature and
light. Can. J. Fish. Aquat. Sci. 45: 261-270.

Bothwell, M. L. 1989. Phosphorus-limited growth dynamics of lotic periphytic
diatom communities: Areal biomass and cellular growth rate responses.
Can. J. Fish. Aquat. Sci. 46: 1293-1301.

Collins, S. L. & Glenn, S. M. 1991. Importance of spatial and temporal dynamics
in species regional abundance and distribution. Ecol. 72: 654-664.

Darwin, C. 1859. On the origin of species by means of natural selection, or the
preservation of favored races in the struggle for 111%. John Murray.
London.

Elser, J. J., Marzolf, E. R & Goldman, C. H. 1990. Phosphorus and nitrogen
limitation of phytoplankton growth in the freshwaters of North America: a

review and critique of experimental enrichments. Can. J. Fish. Aquat. Sci.
47: 1468-1477.

Ehrenberg, C. G. 1843. Verbreitng und Einﬂufs des mikroskopischen Labens in

Sﬁd- und Nord-Amerika. Abhandlungen der Koniglichen Akademie der
Wissenschqﬂen zu Berlin 1: 291-445.

14

Fairchild, G. W., Lowe, R L. & Richardson, W. T. 1985. Algal periphyton rowth
on nutﬁent-diﬁrsing substrates: An in situ bioassay. Ecol. 66: 465-472.

Geissler, U. 1982. Experimentelle untersuchungen zur variabilitat der
schalenmerkmale bei einigen zentrischen Susswasser-Diatomeen. 1. Der
Einﬂuss underschiedlicher Salzkonzentrationen auf den Valva-
Durchmesser von Stephanodiscus hantzschii Grunow. Nova Hedw. 73:
21 1-247.

Grime, J. P. 1979. Plant strategies and vegetation processes. Wiley, New York.

Grime, J. P. 1998. Beneﬁts of plant diversity to ecosystems: immediate, ﬁlter and
founder effects. J. Ecol. 86: 902-910.

Grimm, N. B. & Fisher, S. G. 1986. Nitrogen limitation in a Sonoran Desert
stream. J. N. Am. Benthol. Soc. 5: 2-15.

Grover, J. P. 1991. Resource competition in a variable environment:
Phytoplankton growing according to the variable-intemal-stores model.
Am. Nat. 138: 811-835.

Hanski, I. 1982. Dynamics of regional distribution: the core and satellite species
hypothesis. Oikos 38: 210-221.

Hansson, L.-A. 1992. Factors regulating periphytic algal biomass. Limnol.
Ocewrogr. 37: 322-328.

Hecky, R. E. & Kilham, P. 1988. Nutrient limitation of phytoplankton in
fieshwater and marine environments: a review of recent evidence on the
effects of enrichment. Limnol. Oceanogr. 33: 796-822.

Hustedt, F. 1930. Bacillariophﬂa (Diatomeae). In: Pascher, A [Ed] Die
Susswasser-Flora Mitteleuropas. 466 pp. G. Fischer, Jena.

Hutchinson, G. E. 1961. The paradox of the plankton. Am. Nat. 95: 137-145.

Keddy, P. A. 1990. Competitive hierarchies and centrifugal organization in plant
communities. In: Grace, J. B. and Tilman, D. [Eds]. Perspectives on plant
competition, Academic Press, San Diego, CA, pp. 266-290.

Kevern, N. R, Wilhnr, J. L. & Van Dyne, G. M. 1966. Use of artiﬁcial substrata

to estimate the productivity of periphyton. Limnol. Oceanogr. 11: 499-
502.

15

Kilham, S. S., Theriot, E. C. & Fritz, S. C. 1996. Linking planktonic diatoms and
climate change in the large lakes of the Yellowstone ecosystem using
resource theory. Limnol. Oceanogr. 41: 1052-1062.

Kociolek, J. P. & Stoermer, E. F. 1988. Taxonomy, ultrastructure and distribution

of Gomphoneis herculeana, G. eriense, and closely related species. Proc.
Acad. Nat. Sci. Philadelphia 140: 24-97.

Kociolek, J. P. & Stoermer, E. F. 2001. Taxonomy and ecology: a marriage of
necessity. Opinion. Diat. Res. 16: 433-442.

Krammer, K. & Lange-Bertalot, H. 1991. Bacillariophyceae. 2/4. Teil:
Achnanthaceae. In: Siisswasserﬂora von Mitteleuropa (H. Ettl, J. Gerloff,
H. Heynig & D. Mollenhauer, eds.) 876 pp. Gustav Fisher Verlag,
Germany.

Kiitzing, F. T. 1833. Synopsis Diatomacearum oder Versuch einer systematischen
Zusammenstellung der Diatomeen. Linnaea 8: 529-620, Taf. 13-19.

Laing, T. E, Pienitz, R & Smol, J. P. 1999. Freshwater diatom assemblages from
23 lakes located near Norilsk, Siberia: a comparison with assemblages
from other circumpolar tree line regions. Diat. Res. 14: 285-305.

Lamberti G. A., Gregory, S. V., Ashkenas, L. R., Steinman, A. D. & McIntire, C.
D. 1989. Productive capacity of periphyton as a determinant of plant-
herbivore interactions in streams. Ecol. 70: 1840-1856.

Mann, D. G. 1999. The species concept in diatoms. Phycol. 38: 437-495.
Manoylov, K. M. & Stevenson, R. J. 2004. Cymbella excisa Kﬁtz. in different

nutrient conditions. In: Poulin, M. [Ed] Proceedings of the 17‘”
International Diatom Symposium. Biopress Ltd., Bristol, pp. 223-233.

Manoylov, K. M., Morales, E. & Stoermer, E. 2003. Staurosira stevensonii
(Bacillariophyta) a new species from acidic environments in Florida,
USA. Europ. J. Phycol. 38: 65-71.

Mayr, E. 1942. Systematics and the origin of Species. Columbia University Press,
New York, 334pp.

MacArthur, R H. 1972. Geographical ecology. Harper & Row, New York.

16

McCormick, P. V. 1996. Resource competition and species coexistence. In
Stevenson, R J ., Bothwell, M. L., Lowe R L. [Eds] Algal Ecology.
Academic Press, San Diego, pp. 229-252.

McCormick, P. V. & Stevenson, R. J. 1991. Mechanisms of benthic algal
succession in lotic environments. Ecol. 72: 1835-1848.

Morales, E. A 2001. Morphological studies in selected fragilarioid diatoms
(Bacillwiophyceae) from Connecticut waters, USA Proc. Acad. Nat. Sc.
Philadelphia 151 : 39-54.

Morales, E. A 2002. Studies in selected fragilarioid diatoms of potential
indicator value ﬁom Florida (USA) with notes on the genus Opephora
Petit (Bacillariophyceae). Limnologica 32: 102-113.

Morales, E. A 2003a. On the taxonomic position of the Belonastrum and
Synedrella, two new fragilarioid genera described by Round and Maidana
(2001). Cnptog. Algal. 24: 277-288.

Morales, E. A. 2003b. Fragilaria pennsylvanica, a new diatom
(Bacillariophyceae) species from the United States of America with
comments on the taxonomy of the genus Synedra Ehrenberg. Proc. Acad.
Nat. Sc. Philadelphia 153: 155-166.

Moss, B. 1968. The chlorophyll a content of some benthic algal communities.
Arch. Hydrobiol. 65: 51-62.

National Research Council 2000. Global change ecosystem research. Ecosystems
panel. National Academy Press, Washington DC.

Pan, Y., Stevenson, R J., Hill, B. H., Herlihy, A T. & Collins, G. B. 1996. Using
diatoms as indicators of ecological conditions in lotic systems: a regional
assessment. J. N. Am. Benthol. Soc. 15: 481-495.

Parsson, G., Holmgren, S. K., Jansson M., Lundgren, A., Nyman, B., Solander, D.
& Annel, C. 1977. Phosphorus and nitrogen and the regulation of lake
ecosystems: experimental approaches in subarctic Sweden. Proc.
Circumpolar Conference N. Ecol. Ottawa, Canada, p. 1-19.

Patrick, R 1967. The effect of invasion rate, species pool, and size of area on the
structure of the diatom community. Proc. Natl. Acad. Sci. USA. 58:
1335-1342.

17

Patrick, R 1968. The structure of diatom communities in similar ecological
conditions. Am. Nat. 102: 173-183.

Patrick, R 1976. The formation and maintenance of benthic diatom communities.
Proc. Am. Philos. Soc. 120: 475-84.

Patrick, R & Reimer, C. W. 1966. The diatoms of the United States, Exclusive of
Alaska and Hawaii. Monogr. 13. Vol. 1, 688 pp. The Academy of Natural
Sciences of Philadelphia, Philadelphia.

Patrick, R & Reimer, C. W. 1975. The diatoms of the United States, Exclusive of
Alaska and Hawaii. Monogr. 13. Vol. 2, 213 pp. The Academy of Natural
Sciences of Philadelphia, Philadelphia.

Peterson, C. G. & Stevenson, R. J. 1992. Resistance and recovery of lotic algal

communities: importance of disturbance timing, disturbance history, and
current. Ecol. 73: 1225-1461.

Pickett, S. T. A, Collins, S. L. & Armesto, J. J. 1987. Models, mechanisms and
pathways of succession. Bot. Rev. 53: 33 5-371.

Porter, K. G. 1976. Enhancement of algal growth and productivity by grazing
zooplankton. Science 192: 1332-1334.

Potapova, M. G. & Charles, D. F. 2003. Distribution of benthic diatoms in US.
Rivers in relation to conductivity and ionic composition. Fresh. Biol. 48:
13 1 1-1328.

Potapova, M. G. & Ponader, K. C. 2004. Two common North American diatoms,
Achnanthidium rivulare sp. nov. and A. deﬂexum (Reimer) Kingston:
morphology, ecology and comparison with related species. Diat. Res. 19:
33-57.

Pringle, C. M. 1990. Nutrient spatial heterogeneity: effects on community
structure, physiognomy, and diversity of stream algae. Ecol. 71: 905-920.

Round, F. E., Crawford, R M. & Mann, D. G. 1990. The Diatoms. Biology and
morphology of the genera. Cambridge University Press, Cambridge 747

PP-

18

Schindler, D.W. 1974. Eutrophication and recovery in experimental lakes:
implication for lake management. Science 184: 897-899.

Sommer, U. 1985. Comparison between steady state and non-steady state
competition: experiments with natural phytoplankton. Limnol. Oceanogr.
30: 335-346.

Sommer, U. 1996. Nutrient competition experiments with periphyton from the
Baltic Sea. Mar. Ecol. Prog. Series 140: 161-167.

Southwood, T. R. E. 1977. Habitat, the template for ecological strategies? J.
Animal. Ecol. 46: 337-365.

Steinman, A. D. 1996. Effects of grazers on freshwater benthic algae. In
Stevenson, R J., Bothwell, M. L., Lowe R L. [Eds] Algal Ecology.
Academic Press, San Diego, pp. 341-3 73.

Stevenson, R J. 1983. Effects of current and conditions simulating autogenically
changing microhabitats on benthic algal immigration. Ecol. 64: 1514-
1524.

Stevenson, R J. 1984. How currents on different sides of substrates in streams
affect mechanisms of benthic algal accumulation. Int. Rev. Gesamten.
Hydrobiol. 69(2): 241-262.

Stevenson, R. J. 1995. Community dynamics and differential species performance
of benthic diatoms along a nutrient gradient. In Kociolek, J. P. & Sullivan,
M. J [Eds] A Century of Diatom research in North America. A tribute to
the Distinguished Careers of Charles W. Reimer and Ruth Patrick. Koeltz
Scientiﬁc Books USA, Champaign, pp. 29-45.

Stevenson, R. J. 1996. An introduction to algal ecology in freshwater benthic
habitats. In Stevenson, R. J ., Bothwell, M. L., Lowe R. L. [Eds] Algal
Ecology. Academic Press, San Diego, pp. 3-30.

Stevenson, R J. & Glover, R. 1993. Effects of algal density and current on ion

transport through periphyton communities. Limnol. Oceanogr. 38: 1276-
1281.

Stevenson, R. J. & Pan, Y. 1994. Are evolutionary tradeoffs evident in responses
of benthic diatoms to nutrients? In Marino, D. [Ed] Proceedings of the
13‘" International Diatom Symposium, Bio Press Ltd., Bristol, pp.71-81.

19

Stevenson, R J. & Peterson, C. G. 1989. Variation in Benthic Diatom
(Bacillariophyceae) immigration with habitat characteristics and cell
morphology J. Phycol. 25: 120-129.

Stevenson, R. J ., Peterson, C. G., Kirschtel, D. B., King, C. C. & Tuchman, N. C.
1991. Succession and ecological strategies of benthic diatoms
(Bacillariophyceae): density-dependent grth and effects of nutrients
and shading. J. Phycol. 27: 59-69.

Tilman, D. 1976. Ecological competition between algae: experimental
conﬁrmation of resource-based competition theory. Science 192: 463-465.

Tilman, D. 197 7. Resource competition between planktonic algae: an
experimental and theoretical approach. Ecol. 58: 338-348.

Tilman, D. 1980. Resources: a graphical-mechanistic approach to competition
and predation. Am. Nat. 116: 362-393.

Tilman, D. 1982. Resource competition and community structure. Princeton
University Press, Princeton, 296 pp.

Tuchman, N. C. 1996. The role of heterotrophy in algae. In Stevenson, R J .,
Bothwell, M. L., Lowe R L. [Eds] Algal Ecology. Academic Press, San
Diego, pp. 299-319.

van Dam, H., Mertens, A. & Sinkeldam, J. 1994. A coded checklist and ecological
indicator values of freshwater diatoms from the Netherlands. Neth. J.
Aquat. Ecol. 28(1): 117-133.

Villeneuve, V. & Pienitz, R. 1998. Composition diatomifere de quatre sources
thermales au Canada, en islande et au Japon. Diat. Res. 13: 149-175.

20

CHAPTER 1
RESOURCE TRADEOFFS AND ADAPTIVE STRATEGIES OF
DIATOMS IN DIFFERENT NUTRIENT CONDITIONS
INTRODUCTION

Competition for nutrients is widely considered an important process
structuring phytoplankton communities (Tilman 1982, Sommer 1990) and has
been generalized to attached communities in the periphyton (McCormick 1996).
Even though the factors controlling development of benthic algal communities
have been studied for many years (Patrick 1976, Hill and Knight 1988, Blinn
1993, Stevenson 1996, 1997), little experimental work has been conducted to
evaluate the role of competition for nutrients in the evolution of benthic diatoms
and their growth in streams (Biggs 1996).

Phosphorous and nitrogen are essential nutrients for algal cell growth
(Borchardt 1996). Physiologically, algal cells require ﬁxed amounts of nutrients
to reproduce (Hecky and Kilham 1988). Resource limitation can constrain algal
growth in streams (Grover 1990, 2000, McCormick 1996) and algal competition
for nitrogen and phosphorus can increase with algal density on substrata
(Stevenson 1990). Concentrations greater than 28 [lg P/L (Bothwell 1989) and 55
pg N03-N/L (Grimm and Fisher 1986, Stevenson 1995) have been assumed to be
saturating concentrations for achieving and maintaining maximum growth.
Therefore, in nutrient limited habitats, co-adaptive traits such as species growth
rates, nutrient uptake, growth forms, dispersal rates and motility on substrata may

be evolutionary important.

21

Species ﬁtness is conferred in part by “species performance” (sensu
Pickett et al. 1987). Algal performance in low nutrients has been related to
diﬂ‘erent characters, such as size (Round et al. 1990), growth forms (Hudon and
Legendre 1987, Stevenson et al. 1991, Stevenson and Pan 1994), reproduction
rate, and immigration rate (Stevenson and Peterson 1991). During periods of
intense competition for a limiting resource, tradeoffs are hypothesized because
diatom species that allocate energy and resources to several traits would,
theoretically, be competitively excluded (Tilman 1982, Grace 1990, Stevenson
and Pan 1994, Stevenson 1995). Species grth rates decrease as benthic diatom
density increases, which may be caused by competition for nutrients (Stevenson
1983, 1990, Humphrey and Stevenson 1992). Thus, competition for nutrients
may be greatest during later stages of benthic community development
(Stevenson et al. 1991), which coincides with development of morphological
adaptations, such as motility and stalk formation.

The concept of ecological strategies is based on the assumptions that
speciﬁc traits confer different survival advantages, traits have signiﬁcant energy
and resource costs, and these traits require evolutionary choices (Schluter and
Ricklefs 1993). Thus, species could evolve different sets of evolutionary traits
and specialization. For algae as for other organisms, two resource-based
strategies for survival are often identiﬁed and referred to as r- and K- selection
strategies (MacArthur and Wilson 1967). These strategies have been called:
competitive (C-), stress tolerant (S-), and ruderal (R-) strategies by Grime (1979)

and for aquatic organisms Categories 1, II and III (Reynolds 1984), but regardless

22

of the terminology, those evolutionary strategies are consistent in the prediction of
tradeoffs between fast and slow reproduction with high and low nutrient
requirements, respectively (Grace 1995). Adaptive strategies have been directly
applied to algae (Margalef 1978, Kilham and Kilham 1980, Reynolds 1988,
Stevenson et al. 1991, Biggs et al. 1998). With r-selection, species grow rapidly,
exploit resources, produce many offspring, and then disperse to new environments
where resources are abundant as resource supplies become limiting at the original
habitat (Biggs et al. 1998). These r-selection strategies for benthic algae in
streams have been hypothesized during frequent disturbance and re-colonization
stages (Reynolds 1988, Biggs et al. 1998). K-selection strategies allow species to
utilize resources in very low supply (Pianka 1970). These species persist after
resources have been exploited and grow slowly but efficiently in very low
resource environments (McCormick 1989). Species could theoretically coexist in
low resource environments if they possess traits to use one of several potential
resources more effectively in a habitat than another species (Whittaker 1975,
Stevenson and Pan 1994). Experimental conﬁrmation of those relationships with
periphyton is needed.

In this study, resource tradeoffs and adaptive strategies of benthic diatoms
were addressed by relating diatom species performance in different nutrient
conditions. Multispecies competition experiments conducted with diatoms to date
(Humphrey and Stevenson 1992, Lamberti and Steinman 1993, McCormick 1996)
have not manipulated nutrients along two gradients and removed grazers. In this

study, natural stream-like conditions were provided, a diverse species pool was

23

introduced, and nutrient supply was reduced to limiting concentrations of both
nitrogen and phosphorus through partial recirculation in stream-side channels
with a novel design. The ﬁrst goal of this experiment was to relate responses of
individual diatom populations to nitrogen and phosphorous limitation, while the
other nutrient was in saturating supply. I hypothesized that different species
would be best adapted to live in low N or low P concentrations because of
specializations for sequestering one nutrient or the other, when in low supply
(Stevenson and Pan 1994). The second goal of this research was to test for
tradeoffs for grth in low and high concentrations of the same nutrient.
Maximum growth at very low concentrations theoretically is possible only for the
good competitors. Species-speciﬁc growth rates were compared in low and high
nutrient concentrations and in both low and high density.
MATERIAL AND METHODS
Study area

Artiﬁcial streams in the University of Louisville research Stream facility
at Bemheim Forest, Kentucky were used for the experiment. Bemheim Forest is
a 10,000 acre nature preserve in which the entire watershed of Hart's Run, a
pristine third order stream, is located. The relatively small 7 .5-km2 catchment of
this stream is 97 % deciduous forested and shaded for most of its length. No
fertilizers are used in the open ﬁelds (Bemheim Forest Foundation public records,
2002). Hart's Run at base ﬂow has 40 L/s discharge and an average channel
width of 6 m. Current velocities are stable, when there are no storms and vary

ﬁ'om <5 cm/s in pools to >30cm/s in riffles (Humphrey and Stevenson 1992).

24

The upper reaches of the stream drain predominantly Devonian oil shale, with the
lower reaches cutting into dolomitic limestone. The dominant substrate in Hart’s
Run is siltstone cobble (Bemheim Forest Foundation public records, 2002).
There are no public trails close to the stream, and all access roads are gated.
Experimental design

Manipulations were carried out between May 17 and June 3, 1999 in 36
partially re-circulating streams that consisted of closed 3.5 x 1.5 m loops of 10 cm
diameter PVC with the t0p piece cut lengthwise to allow placement of substrata
and algal growth (Fig. 1.1). These loops were ﬁlled with stream water from
Hart’s Run. Six water baths were used to control water temperature. Bottoms of
each channel were immersed in water baths. Current was generated in each loop,
by forcing air into the bottom of one end of the loop through a hose with a Sweet
WaterTM model S45 Blower. Air bubbles lifted water and produced current
velocities of 25 cm/s. Turbulence was minimized by placing ﬂow stabilizers,
made ﬁom 15, 1.5 cm diameter plastic tubes, parallel to the ﬂow at the head of
each channel. Stream water was continuously exchanged in each bath at a high
enough rate to maintain temperatures in the recirculating channels similar to that
of the stream.

Fresh steam water was slowly added to each channel at a rate of 7 ml/s
Water was allowed to overﬂow through a hole drilled in the ends of each channel.
This caused complete turnover of water every 4 hrs. Partial recirculation allowed

microorganisms inhabiting unenriched channels to draw nitrogen and phosphorus

25

below those found in the ambient stream water. This method was shown to be
effective by Mulholland et al. (1991).

Twenty unglazed ceramic tiles (29.2 cm2) were placed at the downstream
end of each channel where turbulence was minimal. Light levels were reduced by
placing two layers of greenhouse shade cloth, supported by 1 cm2 wire mesh
screen over the stream channels. This simulated the light regimes in the nearby
streams. The shade allowed 12% of incident photosynthetically active radiation
(measured with LicorTM Model Li 189 light meter) to reach the tiles.

A large species pool of benthic algae ﬁ’om a variety of nutrient conditions
were added to each stream on May 17 (one day before the start of nutrient
addition). This inoculum was produced by scraping rocks from three streams
with different nutrient concentrations: Hart’s Run was the low nutrient stream
where the experiment was run; Wilson Creek located 5 km north of Hart’s Run
had moderate concentrations of nitrogen and phosphorus (R. J. Stevenson
unpublished data); and the Middle Fork of Beargrass Creek, with an urban
watershed had high nutrient concentrations (R. J. Stevenson unpublished data).
Scrapings from the three streams were combined in a single container and
homogenized before 500 ml aliquots were added to each stream. Algal
community in the inoculum was evaluated to ensure large species pool
availability.

Stock solutions with different concentrations of NaN03 and KHzPoa were
delivered ﬁ'om 1 L containers to each channel using peristaltic pumps

(Manostatm model STD) set at a drip rate of 0.3 ml/min. Nitrogen and

26

phosphorus were added throughout the experiment at a constant rate that would
elevate ambient nutrient concentrations by the following amounts: 0, 16, 64, 256,
512 and 1024 (pg N/L) with 50 pg P/L; and 0, 2, 8, 32, 50 and 128 (pg P/L) with
500 pg N/L prior to uptake within the channels. Three replicate channels were
used for each treatment. Twelve nutrient treatments were assigned randomly to
the 36 channels. Nutrient addition was started on May 18“, with one day of
colonization, because I did not want to affect initial immigration and
establishment of a diverse community with nutrient treatments. Benthic algae on
tiles in channels were sampled on 6 dates approximately 72 hours apart.
Sample collection

Water samples were collected from the natural stream and ﬁ'om each
channel on days 1, 3, 6, 8, 12, and 14 with acid-washed 125 ml polypropylene
bottles. Water samples were immediately packed on ice and returned to the lab
for analysis. Nitrate + nitrite nitrogen, ammonium nitrogen (NIL-N), and silicate
(Si) concentrations were estimated on a Skalar" auto-analyzer according to
standard methods (APHA 1998). Since nitrate + nitrite nitrogen consists mainly
of nitrate in this stream (R. Schultz unpublished data), I considered this a measure
of nitrate nitrogen (NO3-N). Soluble reactive phosphorus (SRP) was also
determined according to standard methods using a Hitachi“ U-200 1
spectrophotometer (APHA 1998). Since nutrient uptake within each channel
likely modiﬁed treatment conditions, NO3-N and SRP were measured in the

analyses against the targeted valued. After determining that the nutrient

27

manipulation produced the targeted concentrations, those concentrations were
used as the independent variable in regression models.

Two tiles (not neighboring and not from the upstream end) were collected
ﬁom each channel on days 3, 6, 10, 13, and 16 of community development and
frozen (APHA 1998). Sampling stopped after day 16 because sloughing was
possible. Algae on day one (May 18“) were collected from 6 channels and
represented the initial assemblage. Samples were immediately frozen and
transported to the laboratory and stored in freezer till processing. One channel in
the P gradient of the 8P/500N treatment leaked and lost water so was not used in
the analyses. For the experiment, there were 181 samples collected and counted.

At the laboratory, algal material was scraped from the upper surface of the
thawed tiles into a beaker with a toothbrush and razor blade. Samples were
diluted to known volumes with deionized water, homogenized with a Biospec"
model m133\1281-O biohomogenizer, and then sub-sampled with a mechanical
pipette while continuously mixing on a magnetic stir plate (subsamples for
chlorophyll a and algal enumeration). Chlorophyll a was determined
spectrophotometerically with Spectronic Genesysmz spectrophotometer after 24
hours extraction in 90% buffered acetone and correction for phaeophytin (APHA
1998). Subsamples for counts were preserved with M3 for algae counts (APHA
1998). For algal enumeration, semi-permanent syrup mounts were prepared with
0.2 to 2.0 ml sample amount on the cover slip and 4 ml Tuft Mounting syrup

(Stevenson 1984).

28

Initially, a minimum of 300 natural algal units (live unicellular,
ﬁlamentous and colonial algae) were identiﬁed and enumerated to the lowest
taxonomic level. For diatoms (Bacillariophyta) only frustules with a visible,
healthy looking chloroplast, assumed to be physiologically active, were
enumerated. Counts were performed at IOOOX on a Leica DMLB light
microscope. The main diatom ﬂoras used for identiﬁcation were Patrick and
Reimer (1966, 1975) and Krammer and Lange-Bertalot (1986, 1988, 1991 a, b).
Common taxa in these counts were identiﬁed. Then counts were continued until
cell densities of the 14 common taxa were estimated precisely (Alverson et al.
2003). Common taxa for this experiment were deﬁned as taxa found in all
treatments. At least 15 live cells were observed for precise estimate of density.
Total community and individual taxa densities were calculated (D, cells cm'z).
Log-transformed densities were used to calculate growth rates for the whole algal
community and for individual taxa. Densities were log-transformed in order to
stabilize the variance for correlation and regression analysis.

Data analysis

The experiment was created to evaluate limiting concentrations of N and
P, as well as to evaluate tradeoffs for N and P competitive abilities during
successive stages of community development. Tradeoffs for growth at different
treatments (nutrient concentrations) were assessed. Treatments usually had 3
statistically independent replicates. Treatment eﬂ’ects (between-subject factor)
were tested using repeated measures analyses of variance (RM ANOVA). The

univariate approach to RM AN OVA takes into account correlation among

29

sampling dates, which can violate the assumption of independence between
samples for regular analysis of variance (ANOVA). In RM ANOVA, there is
potential variance violation where a circular variancecovariance matrix might not
have the same constant value for the variance of the difference between any two
levels of the within-subject factor (see Winer et al. 1991). In RM ANOVA, all
results were adjusted for circularity with Geisser-Greenhouse epsilon (G-G, von
Ende 2001) which is very conservative, measures sphericity and ensures
adjustment of the F-statistic degrees of ﬁ'eedom (a circular variancecovariance
matrix that is transformed to its normalized orthogonal form is spherical). Time
effects on nutrients were evaluated as a within-subject factor in RM ANOVA.
Interactions of nutrient and time were reported also. Particular time intervals in
which the treatment effects were different were evaluated with polynomial proﬁle
analysis (univariate ANOVAs for each contrast: linear, quadratic, etc.). The
higher order term was considered ﬁrst and was dropped from the model if not
signiﬁcant.

Tilman (1982) deﬁnes successful competitors as species able to utilize
effectively the lowest possible concentration of certain nutrient. Based on this
deﬁnition, N and P species were deﬁned as the fastest growing in low N and P
conditions, respectively.

Grth rates (R) were estimated as daily per capita changes in cell density
in each channel and for all time combinations using the following equation:

R = (In D(t+1) — In D (.))/( TM -Tt), where D is algal biomass (total

density) and Tr, Tm are consecutive times. Grth rates (cellscell’ld’l) were

30

calculated for populations of each common species in each channel for each time
period. Growth rates for each period of time were designated as Ros for the
growth change between day 0 and day 3 and similarly for R343, R510, R1043, and
R1345. Densities and growth rates were quite variable among channels and
through time; the results are presented without transformation or exclusion of
outliers. The question of interest for this part of the study was species population
statistics, so the patterns of variability were not considered an obstacle, but were
checked for statistical violations.

Maximum growth rate (Rm) was estimated using two methods. First,
growth rates for each time period were independently related to nitrogen and
phosphorus gradients with the Monod (1950) equation: R=Rmf S/(K.+S), where
R is growth rate for each period, S is nutrient concentration and the parameters
were maximum growth rate (Monod estimate, Rm (M)) and the half saturation
constant (K.). The Monod curve was ﬁtted using a nonlinear regression
procedure (Wilkinson 1989, in Systat" 10 for Windows‘D XP). Rm(M) and K.
estimates for each taxon were determined. Estimates did not account for density
dependence. Estimations of Rm and K, with Monod were not possible after day
10, potentially because the inﬂow nutrient concentrations and the increase in
biomass were not at equilibrium. Second, observed Rm at high NP (R..m O), was
estimated for each treatment as the average grth rate in the two highest nutrient
concentrations and could be calculated for all 5 time periods. The results from the
estimation of Rm (M) and Rm, (0) were used to assess species competitive

abilities.

31

Growth rate in low nutrient concentrations was calculated 3 ways. K,
ﬁ'om the Monod model was one indicator of species performance in low nutrient
conditions, but were too variable. In many cases, statistically stable estimates of
K. and its error variance could not be estimated. Average growth rates in the low
N and low P control treatments were also used as estimates of species
performance in low nutrients. The third method accounted for expected variation
in species abilities to perform in the environments that I created. Species growth
rates vary greatly with non-nutrient abiotic factors, so accounting for those effects
may provide a better estimate of multiple species performances in one speciﬁc
habitat. For method 3, average grth rates in low N and low P were divided by
the maximum growth rate (Rm (0)) for that species in high N and P treatments.
Bothwell (1989) used a similar method for normalizing species performances to
their optimal performance so that he could compare results of multiple
experiments conducted throughout the year, when seasonal variation in
performance had to be taken into account. Results of tradeoff analyses were
compared using the multiple estimates of performance in low and high nutrient
concentrations.

All common taxa were used in the tradeoff analyses. A trade-oﬁ’ relates
competitive performance for different traits, because traits require energy and
resource allocation. Trade-off hypotheses predict negative correlations between
species traits. Therefore, I expected a negative correlation between the growth
rates of species in low N and low P conditions if tradeoﬂ‘s exist for gleaning N

and P at low concentrations. A negative correlation between species growth rates

32

in low N and low P conditions would indicate that overall, different species were
best at growing in low N and low P conditions. I also expected a negative
correlation between species growth rates in low and high N or in low and high P
conditions if different species were r- and K-selected (i.e. gleaners and
opportunists sensu Grover 1990). The questions concerning diatom competitive
abilities do require identifying the species, but correlations between species
growth, rather than the species themselves imply tradeoffs. Correlations were
calculated with Spearman rank correlation analyses. All statistical analyses were
performed with SYSTAT" 10 (spss Inc., Wilkinson 1989). Each growth rate
was calculated for the following successional stages: initial colonization (Rita);
accrual phase (RM); peak biomass (R640) and persistence phases (R1043 and R13-
16). Peak biomass was deﬁned as the maximum areal biomass attained by a
periphyton community during development (Bothwell 1989).
RESULTS
Physkochemical chwacteristics

Water temperatures in the recirculating channels averaged approximately
20°C which was close to temperatures recorded in Hart’s Run. In ambient stream
water, NO3-N ranged from 15 to 38 pg N/L and SRP ranged from 2 to 7 pg P/L.
Silicate concentrations were consistently high in Hart’s Run ranging ﬁ'om 11 to
12 mg Si/L.

Along the N gradient nitrate concentration in the control treatment was
from 0 to 33 pg N/L, SRP was always above 25 pg P/L (Appendix 1.1). Nitrate

concentration signiﬁcantly increased with treatment. Ammonium concentrations

33

ranged ti'om 6 to 98 pg N/L. Total nitrogen varied from 91 to 1113 pg N/L.
Soluble reactive phosphorus, after uptake, ranged from 18 to 59 pg P/L. The
lowest measure of TP was around 30 pg P/L (TP ranged from 29-168 pg P/L).
Silicate was high along the N gradient ranging fi'om 8 to 12 mg Si/L.

Along the P gradient, POa-P concentration in the control treatment ranged
2 to 5 pg P/L nitrate was always above 200 pg N/L (Appendix 1.2). Nitrate
(NO3-N) concentration, after uptake, stayed high ranging from 112 to 789 pg N/L.
Ammonium concentrations ranged from 7 to 98 pg N/L. Total nitrogen was also
high, ranging from 228 to 815 pg N/L. SRP ranged fiom 2 to 110 pg P/L. TP

ranged ﬁ'om 4 to 169 pg P/L. Silicate was also high along the P gradient ranging
from 7 to 11mg Si/L.

Appendices 1.1 and 1.2 present detailed changes in nutrient chemistry in
the artiﬁcial streams with nutrient treatment and time during the experiment.

Community response in N and P gradient

Total community biomass measured as chlorophyll a increased along the
two gradients (ANOVA, p<0.0001, Fig. 1.2). Chlorophyll 0 increased linearly
ﬁom 1.0 to 4.9 pg chlorophyll a cm'2 along the N gradient (Fig. 1.2). Along the P
gradient, chlorophyll a increased asymptotically from 0.9 to about 4.3 pg
chlorophyll a cm’2 (Fig. 1.2).

Diatoms were the dominant algae in all treatments. Visual differences
were observed macroscopically in communities among the nutrient treatments.

The algal mat was thinner in the low N compared to thicker golden mat in the low

34

P treatment. During the later stages of community development, periphyton
became a darker color in higher nutrient levels of both N and P gradients.

Cell density of the whole community increased with both nutrients (RM
ANOVA, for N and P, p<0.0001). Density of green algae (Chlorophyta) did not
diﬁ‘er signiﬁcantly with increasing N or P (RM ANOVA, F5 ,233=1.328, p=0.253
and F4,154=1.655, p=0. 163 respectively). Bluegreen algae (Cyanophyta) did not
differ signiﬁcantly in density with increasing N or P also (RM ANOVA, F5,
36=1.089, p=0.366 and F2,12=0.054, p=0.948 respectively).

Total diatom density increased with N concentration and time (RM
ANOVA, F 5,12=4.11, p=0.0209 and F 5, 60=328.58, p<0.0001, respectively, Fig.
1.3). Maximum algal density along the N gradient was 2.1 x 106 (:l: 0.9 x 106 SE)
cells cm'2 and occurred in the 1024N/50P treatment (Fig. 1.3). Adjusted for
circularity, the treatment x time interaction was marginally signiﬁcant (RM
ANOVA, F 25, 50=2.02, p=0.053 7) and the positive effect of nutrients on density
seemed to increase during the colonization period (Fig. 1.3). Density increased
linearly with N concentration (polynomial proﬁle analyses, F 5,12=6.69,
p=0.0034).

Total diatom density increased with P concentration and time (RM
ANOVA, F5, 11:5.41, p=0.0094 and F5, 55=3 17.64, p<0.0001, respectively, Fig.
1.3). Maximum algal density along the P gradient was 1.98 x 106 (:l: 1.1 x 106 SE)
cells cm'2 and occurred in the SOON/SOP treatment (Fig. 1.3). The treatment x
time interaction after adjustment for circularity was signiﬁcant (RM ANOVA, F25,

55=2.45, p=0.0254) indicating that diatom density increased faster in high than in

35

low P concentrations. Density increased nonlinearly along the P gradient
(polynomial proﬁle analyses: quadratic, F 5, ”=7 .29, p=0.0031).

Diatom community grth rate (Fig. 1.4 a) increased signiﬁcantly with
increasing N concentration (RM ANOVA, F 5,12=6.56, p=0.0037) and decreased
signiﬁcantly with time (RM ANOVA, F4,4g=24.52, p<0.0001). Growth rates
ranged from -0.3 to 0.9 (cells cell'lday'l). The treatment x time interaction was
not signiﬁcant (RM ANOVA, F20,4g=0.85, p=0.5955). After day 6, growth rates
dropped signiﬁcantly (polynomial proﬁle analysis quadratic, time R55, F 5,
12=4.21, p=0.0193) and thereafter did not differ with time (Fig. 1.4 a).

Diatom community growth rate (Fig. 1.4 b) increased signiﬁcantly with
increasing P concentration (RM ANOVA, F 5, 11:17.64, p<0.0001) and decreased
signiﬁcantly with time (RM ANOVA, F4, 44:30.90, p<0.0001). Growth rates
ranged from -0.4 to above 1 (cells cell'1 day!) The treatment x time interaction
was not signiﬁcant (RM ANOVA, F 20, “=1 .96, p=0.0649). Growth rates
changed signiﬁcantly along the P gradient also (polynomial proﬁle analysis
quadratic, time R345, F 5,11=6.24, p=0.0056, Fig. 1.4 b).

Diatom species responses and evidence for tradeoﬁ's

From 138 taxa observed in the artiﬁcial streams, 14 taxa had sufﬁcient
densities to allow accurate estimation of growth rates for all days and treatment.
Those taxa appeared in all treatments in all channels. Of the 14 common taxa
assessed, grth rates of 6 were changed signiﬁcantly (p<0.05) by increasing N
concentration (Table 1.1). These taxa were Achnanthes deﬂexa, Achnanthidium

minutissimum, Encyonema minutum, Meridion circulare, Nitzschia linearis, and

36

N. palea. Grth rates of 8 of the common diatom taxa changed signiﬁcantly
(p<0.05) with increasing P concentration: Achnanthidium minutissimum,
Encyonema minutum, Meridion circulare, Nitzschia acicularis, N. dissipata, N.
linearis, N. palea and Synedra ulna (Table 1.2). Growth rates of all taxa, except
Achnanthes deﬂexa along the P gradient, decreased signiﬁcantly with time
(Tables 1.1 and 1.2). Along both gradients (treatment x time) interactions were
not signiﬁcant.

Evidence for trade-offs in species ability to compete in low N and low P
conditions was not observed (Fig. 1.5 a-e), most correlations were positive rather
than the expected negative correlations. During the early stages of colonization
Ros, the same species with highest growth rates in low N also had highest growth
rates in low P: Fragilaria crotonensis, F. nanana, and F. vaucheriae (growth rates
close to 1, Fig. 1.5 a). Species growth rates in low N and P during initial
colonization (Rm) were positively (Spearman r = 0.70, p=0.01) correlated.
During the accrual phase, there was no signiﬁcant correlation between species
growth in low N or low P (R35). Grth rates ﬁ'om day 6 to 9 in low N ranged
from 0 to 0.9 (attached taxa like Achnanthidium minutissimum and Achnanthes
deﬂexa, together with motile Nitzschia acicularis and N. dissipata) grew at
similar high growth (0.9), while growth rates in low P concentrations were less
than 0.3 for most diatoms (Fig. 1.5 b). As communities developed toward peak
biomass, stalk forming Cymbella aﬂinis grew as well as or better than all other
taxa in both low N and low P (Spearman r = 0.55, p=0.044, Fig. 1.5 c). Grth

rates after peak biomass phase and during the persistence phase (R1043, R1345)

37

were lower and were similar for all taxa (correlation not signiﬁcant, Fig. 1.5 d-e).
For most periods, small attached achnanthoid taxa, rosette forming Fragilaria
species, and Nitzschia acicularis grew faster than the larger stalk-forming
Cymbella species.

Evidence for tradeoffs did not increase signiﬁcantly during community
development when density and exploitative depletion of nutrients should increase.
The highest positive correlation was observed for the initial colonization period
(Ron), and lowest was usually during later stages of colonization (Fig. 1.6 a). In
addition, statistically signiﬁcant negative correlations were not observed at any
day. Correlation among ﬁtness traits remained positive for most of the
successional periods rather than dropping to signiﬁcantly negative close to peak
biomass; the same species were able to grow fastest at low N and low P
concentration. Similar results were observed with correlations of normalized
growth rates (Fig. 1.6 b). Diatoms were able to utilize minimum available
resources and grow close to their maximum.

Maximum species growth rates Rm (M) estimated with the Monod
equation decreased with time and estimates did not converge for all taxa after Rag
growth period. Along the N gradient, the model estimated lowest maximum
growth rate for Achncmthes deﬂexa (0.339) and highest for Nitzschia palea
(0.848). Along the P gradient, A. deﬂexa and Meridion circulare, had the lowest
maximum growth rate (0.302) and N. palea had highest growth rate (0.879). The
same taxa that had the highest grth rates in low N also had the highest growth

rates in low P through peak biomass development (Tables 1.3 and 1.4). The

38

correlation between species maximum growth rates in low and low P with the
Monod model was either positive or non-signiﬁcant (Fig. 1.7).

Observed maximum grth rates R..m (O) in the highest nutrient
treatments were calculated for all successional phases along each of the 2
gradients (Figs. 1.8 and 1.9). Along the N gradient most taxa growing fast in low
N grew fast in high N, but the patterns changed with time. A non-signiﬁcant
negative correlation was observed again only before peak biomass (Fig. 1.8 c).
As with the Monod estimation tradeoﬁ‘s were not indicated by different species
growing fastest in low P versus high P conditions dining any phase of
colonization (Figs. 1.9 b-c). During the persistence phases, the patterns along the
N and P gradients were very similar (Figs. 1.8 d-e and 1.9 d-e). Negative
correlations, when observed were not signiﬁcant when the correlations
coefficients were plotted for all sampling time periods. More often the observed

correlations were positive, so no tradeoffs were observed (Fig 1.10).

DISCUSSION

The experimental conditions created in the recirculating streams were
optimal for testing relationships among species performance traits in different
nutrients regimes. Along the two nutrient gradients, diatom communities
developed under similar light, temperature and ﬂow conditions. No grazers were
observed and no disturbance in the nearby stream was recorded. Nutrients were
consistemly at the desired concentrations in the water column. The planned low

nutrient concentrations were achieved. Study of a benthic community dominated

39

by diatoms created in these experiments allowed estimation of diatom grth
parameters in response to known variables. Biomass and densities increased with
nutiient addition, saturating around 265 pg N L" along the N gradient and at 32
pg L" P along the P gradient. These concentrations are close to saturating levels
reported in the periphyton literature (Bothwell 1989, Stevenson 1995).
Chlorophyll a values increased to similar levels along the two gradients. The
negative density effect on growth increased with time and density as seen by
others (Stevenson 1983, Bothwell 1989, Stevenson et al. 1991).

The ﬁrst goal of this study was to determine diatom population responses
.to increasing N and P. The estimated growth parameters were very similar to
values reported in the literature (Sommer 1991, Stevenson 1995). The use of
observed maximum grth rates provided estimates of individual taxa
performances for all time periods, while maximum growth estimated with the
Monod model were possible for three sampling periods only. Species like
Fragilaria vaucheriae, F. crotonensis and Nitzschia acicularis together with well
established eutrophic species (N. dissipata and N. palea; Van Dam et al. 1994)
had the highest observed grth rates. The former three taxa are known as early
colonizers and fast immigrators in benthic diatom communities (Hoagland et al.
1982, Stevenson 1990, Stevenson and Peterson 1989, 1991), but they do not
compete well when biomass increases and nutrient supplies are depleted within
periphyton assemblages during late succession (Stevenson and Glover 1993). In
the literature, Fragilaria crotonensis and the motile species Nitzschia acicularis

and N. palea are reported in nutrient-rich, eutrophic conditions (Kelly and

40

Whitton 1995, van Dam et al. 1994). Contrary to this ﬁnding, in another
experiment (Stevenson and Pan 1994), Fragilaria crotonensis grew faster at any
treatment, except in saturated P conditions. In my study, fast growing species like
F ragilaria crotonensis and F. vaucheriae were able to grow at rates higher than
any reported for diatoms in both low and high nutrients (Stevenson et al. 1991,
Stevenson and Pan 1994, Stevenson 1995).

Late succession algae like Achnanthes deﬂexa, Achnanthidium
minutissimum, Cymbella aﬁinis, and C. cistula grew faster than other taxa after
peak biomass (around day 9), when resources were depleted due to shading and
uptake within the mat. Those taxa vertically elevated their position in the benthos
by attaching to other diatoms or growing on stalks. In variable nutrient
conditions, achnanthoid diatoms have been reported to attach to bare substrate
(after disturbance, Biggs et al. 1998), to other diatoms (Burkholder 1996), or to
macroalgal and plant ﬁlaments (Burkholder and Wetzel 1989, Burkholder 1996).
Elevating strategies have been reported for stalked diatoms (Stevenson and
Peterson 1989, McCormick and Stevenson 1991, Stevenson et al. 1991). In this
study, population performances within the diatom community were due to
species—speciﬁc abilities to colonize independent of competitive interactions
1 during the initial stages. With increasing periphyton thickness, density-dependent
interactions caused changes in community composition.

Competitive abilities have been hypothesized to be related to size in
diatom studies without agreement of which size is evolutionary advantageous

(Grover 1989). Small centric diatoms were reported to have higher maximal

41

grth rates than larger elongate species in limiting nutrients (Sommer 1981,
1983). In experiments with phytoplankton, larger diatoms performed as better
competitors in competition for phosphorus than smaller cells (Tilman et al. 1986),
but when small cells were excluded in another experiment, differences in
competitive performance were observed for two large diatoms Synedra sp. and
Fragilaria crotonensis (Grover 1988). Growth rates and species performances in
this experiment were not a function of size. For example: Fragilaria vaucheriae
(small), Nitzschia palea (medium), and F. crotonensis (large) and N. acicularis
(large) grew fast from day 0 to day 3, while the large Cymbella aﬁ‘inis and the
small Achnanthes deﬂexa grew slower.

The second objective of this study was to test for tradeoffs in competition
for low N and low P, or low and high concentration of the same nutrient. No
tradeoffs were observed between species abilities to grow fastest either in low N
or low P or grow fastest in low or high concentrations of a nutrient. Theory
predicts good competitors will be specialists for gathering one resource or
another. In addition, good competitors will have faster growth rates than other
taxa at low nutrient concentrations (Tilman 1982). Tradeoﬂ‘s research is oriented
towards ﬁnding organisms that are superior performers in sequestering sparse
resources and who fare poorly when resources are abundant (V elicer et al. 1999).
The observed positive relationships between species competitive performances in
low versus high concentrations of the same nutrient indicate that same species
were able to grow fast at very low concentrations of each nutrient. Regardless of

the method of estimation, species tradeoffs for nutrient uptake of limiting

42

resources were not observed, due probably to co-adaptive traits for persistence in
the stream periphyton.

Many adaptive strategies may enable coexistence of the great diversity of
species and morphological adaptations in stream diatom communities. Diatom
community composition may be determined by individual species’ ability to
immigrate (Stevenson 1983), colonize (Korte and Blinn 1989, McCormick and
Stevenson 1991, Tilman and Wedin 1991), utilize unsteady supply of nutrients
(Bothwell 1988, Tuchman 1996) and to reproduce rapidly (Sommer 1985), while
resisting ﬂow. Changes in species composition of benthic diatom assemblages
during community development have been hypothesized to result from succession
of r- to K-selected taxa (Biggs et al. 1998). However, benthic diatom succession
may result ﬁ'om non-interactive processes, such as a shift ﬁom fast immigrators to
fast reproducers (Patrick 1976; Stevenson and Peterson 1989, 1991; McCormick
and Stevenson 1991). Immigration and emigration were not tested in this
experiment, but were probably not very important during late stages of succession
after settling of the initial inoculum. Fresh stream water was dripped into all
channels and recirculated for few hours and Hart’s Run water typically has low
density of suspended algae.

Numerous reasons may explain the lack of evidence for competitive
tradeoffs in benthic diatoms when compared with results ﬁom many ﬁndings in
phytoplankton ecology, which seem to indicate tradeoffs in nutrient uptake
abilities (Tilman 1982, Sommer 1990). In streams, disequilibrium conditions

rather than equilibrium conditions (as in lakes) might lead to strong species

43

selection. Competition for nutrients, on an evolutionary time scale, increases the
efﬁciency of population performance by development of co-adaptive traits
(Tilman 1982, Stevenson and Pan 1994). Perhaps competition for nutrients was
sufﬁciently great historically and generated characteristics such as size
differences, stalks and motility, but disequilibrium conditions have relaxed
competitive interactions and constraints on periphyton phenotypes. In streams,
periphyton adaptation to simultaneous N and P limitation (and/or any other
resource limitation) might functionally constrain the expected nutrient tradeoffs
because timing of N and P depletion is the same and rapid. Thus, no evidence
was found for tradeoffs because species allocate resources to compete for both N
or P simultaneously. Alternatively, good competitors for low N may not be
common and therefore may not have occurred in the initial algal inoculum for this
experiment if streams in this region are predominantly P-limited. Although empty
frustules of Epithemia and Rhopalodia were found in the initial community, these
low-N adapted taxa that can have cyanobacteria endosymbionts (Der et al.
1992) were not observed alive in the channels.

Relationships between competition for low nutrients and mechanisms of
species coexistence are still not resolved in complex benthic algal communities.
Although trade-off responses to limiting factors were not observed, it is evident
that diatoms have adopted differential and versatile functional approaches for
long-term survival in the complex conditions of the benthos. More research is
needed to resolve diatom abilities to grow in the understory and/or overstory of

the periphyton mat and the evolutionary adaptations that generated those

44

differences. Within mat changes in nutrients during succession might have been a
strong evolutionary force in streams. Few traits that may confer adaptive ﬁtness
have been investigated extensively and included in rigorously tested models at a
community scale. The value and costs of traits like motility and stalks should be
included in research, along with other factors like the genetic regulation of
physiological rates through rRNA operons (as found in bacteria by Klappenbach
et al. 2000).

In conclusion, different nutrient concentrations were added to study effects
of nitrogen and phosphorus independently by lowering only one nutrient at a time
to a limiting concentration, while controlling for light, temperature and grazers.
The rate of biomass production (growth rate) in benthic diatoms depends on the
supply of limiting nutrients, but maximum grth rates of common diatom taxa
were reached at very low nutrient concentrations. Growth rates decrease greatly
during community development and with increasing density on substrata. The
same diatom species were able to grow at very low N and low P concentrations
and differences in performances in response to different concentrations of the

same nutrient were not observed.

45

LITERATURE CITED

Alverson, A. J ., Manoylov, K. M. & Stevenson, R J. 2003. Laboratory sources of
error for algal community attributes during sample preparing and
counting. J. Appl. Phycol. 15: 357-369.

APHA 1998. Standard methods for examination of water and wastewater.
Washington DC, American Public Health Association.

Biggs, B. J. F. 1996. Patterns in benthic algae of streams. In Stevenson, R J .,
Bothwell, M. L., Lowe R. L. [Eds] Algal Ecology. Academic Press, San
Diego, pp. 31-56.

Biggs, B. J. F., Stevenson, R. J. & Lowe, R. L. 1998. A habitat matrix conceptual
model for stream periphyton. Arch. Hydrobiol. 143(1): 21-56.

Blinn, D. W. 1993. Diatom community structure along physicochemical gradient
in saline lakes. Ecol. 74(4): 1246-1263.

Borchardt, M. A 1996. Nutrients. In Stevenson, R. J ., Bothwell, M. L., Lowe R
L. [Eds] Algal Ecology. Academic Press, San Diego, pp. 183-227.

Bothwell, M. L. 1988. Growth rate responses of lotic periphytic diatoms to
experimental phosphorus enrichment: the inﬂuence of temperature and
light. Can. J. Fish. Aquat. Sci. 45: 261-270.

Bothwell, M. L. 1989. Phosphorus-limited growth dynamics of lotic periphytic
diatom communities: Areal biomass and cellular grth rate responses.
Can. J. Fish. Aquat. Sci. 46:1293-1301.

Burkholder, J. M. 1996. Interactions of benthic algae with their substrata. In
Stevenson, R J., Bothwell, M. L., Lowe R L. [Eds] Algal Ecology.
Academic Press, San Diego, pp. 253-297.

Burkholder, J. M. & Wetzel, R. G. 1989. Microbial colonization of natural and
artiﬁcial macrophytes in a phosphorus-limited, hardwater lake. J. Phycol.
25: 55-65.

46

DeYoe, H. R, Lowe, R L. & Marks, J. C. 1992. The effect of nitrogen and
phosphorus on the endosymbiont load of Rhopalodia gibba and Epithemia
turgida (Bacillariophyceae). J. Phycol. 28: 773-777.

Grace, J. B. 1990. On the relationship between plant traits and competitive ability.
In Grace, J. B., Tilman, D. [Eds] Perspectives of Pltmt Competition.
Academic Press, San Diego, CA, pp. 51-66.

Grace, J. B. 1995. A clariﬁcation of the debate between Grime and Tilman. Func.
Ecol. 5: 583-587.

Grime, J. P. 1979. Plant strategies and vegetation processes. Chichester, New
York, Wiley, 222 pp.

Grimm, N. B. & Fisher, S. G. 1986. Nitrogen limitation in a Sonoran Desert
stream. J. N. Am. Benthol. Soc. 5: 2-15.

Grover, J. P. 1988. Dynamics of competition in a variable environment:
experiments with two diatom species. Ecol. 69: 408-417.

Grover, J. P. 1989. Inﬂuence of cell shape and size on algal competitive ability.
J. Phycol. 25: 402-405.

Grover, J. P. 1990. Resource competition in a variable environment:
phytoplankton growing according to Monod’s model. Am. Nat. 136: 771-
7 89.

Grover, J. P. 2000. Resource competition in experimental communities of algae
and bacteria along gradient of organic carbon to inorganic phosphorus
supply. J. Planldon Res. 22: 1591-1610.

Hecky, RE. & Kilham, P. 1988. Nutrient limitation of phytoplankton in
ﬁ'eshwater and marine environments: A review of recent evidence on the
effects of enrichment. Limnol. Oceanogr. 33(4): 796-822.

Hoagland, K. D., Roemer S. C. & Rosowski, J. K. 1982. Colonization and
community structure of two periphyton assemblages, with emphasis on the
diatoms (Bacillariophyceae). Am. J. Bot. 69: 188-213.

47

Hill, W. R & Knight, A. W. 1988. Nutrient and light limitation of algae in two
northern California streams. J. Phycol. 24: 25-32.

Hudon, C. & Legendre, P. 1987. The ecological implications of growth forms in
epibenthic diatoms. J. Phycol. 23: 434-441.

Humphrey, K. P. & Stevenson, R J. 1992. Responses of benthic algae to pulses
in current and nutrients during simulations of subscouring spates. J. N.
Am. Benthol. Soc. 11: 37-48.

Kelly, M. G. & Whitton, B. A. 1995. The trophic diatom index: a new index for
monitoring eutrophication in rivers. J. Appl. Phycol. 7: 4232-4244.

Kilham, P. & Kilham, S. S. 1980. The evolutionary ecology of phytoplankton. In
Morris, 1. [Ed] The physiological ecology of phytoplankton. pp 571-597,
Oxford, Blackwell.

Klappenbach, J. A, Dunbar, J. M. & Schmidt, T. M 2000. rRNA operon copy
number reﬂects ecological strategies of bacteria. Appl. Envir. Microb.
66(4): 1328-1333.

Korte, V. L. & Blinn, D. W. 1989. Diatom colonization of artiﬁcial substrata in

pool and riffle zones studied by light and scanning electron microscopy. J.
Phycol. 19: 332-341.

Krammer, K. & LangeBertalot, H 1986. Bacillariophyceae. 1. Teil:
Naviculaceae. In Ettl, H., Gerloff, J ., Heynig, H., & Mollenhauer, D.
[Eds]. Siisswasserflora von Mitteleuropa. 2(1): 1-876. Gustav Fisher
Verlag, Germany.

Krammer, K. & LangeBertalot, H. 1988. Bacillariophyceae. 2. Teil:
Bacillariaceae, Epithemiaceae, Surirellaceae. In Ettl, H., Gerloff, H.,
Heynig, H. & Mollenhauer, D. [Eds]. Srisswasserflora von Mitteleuropa.
2(2): 1-596. Gustav Fisher Verlag, Germany.

Krammer, K. & Lange-Bertalot, H 1991a Bacillariophyceae. 3. Teil: Centrales,
Fragilariaceae, Eunotiaceae. In Ettl, H., Gerloff, J., Heynig, H. &
Mollenhauer, D. [Eds]. Susswasserﬂora von [Mitteleuropa. 2(3): 1-576.
Gustav Fisher Verlag, Germany.

48

Krammer, K. & Lange-Bertalot, H 1991b. Bacillariophyceae. 4. Teil:
Achnanthaceae. Kritische Erganzungen zu Navicula (Lineolatae) und
Gomphonema. In Ettl, H., Gartner, G., Gerloff, J ., Heynig, H. &
Mollenhauer, D. [Eds]. Siisswasserﬂora von Mitteleuropa. 2(4): 1-43 7.
Gustav Fisher Verlag, Germany.

Lamberti, G. A & Steinman, A. D. 1993. Research in artiﬁcial streams:
applications, uses and abuses. J. N. Am. Benthol. Soc. 12: 313-384.

Margalef, R. 1978. Life-forms of phytoplankton as survival alternatives in an
unstable environment. Oceanologica Acta 1: 493-509.

MacArthur, R. H. & Wilson, E. O. 1967. The Theory of Island Biography.
Prinston University Press, 203 pp.

McCormick, P. V. 1989. Mechanisms of benthic algal succession. Ph.D. thesis,
University of Louisville, Louisville, KY, 158 pp.

McCormick, P. V. 1996. Resource competition and species coexistence. In
Stevenson, R J ., Bothwell, M. L., Lowe R L. [Eds] Algal Ecology.
Academic Press, San Diego, pp. 229-252.

McCormick, P. V. & Stevenson, R J. 1991. Mechanisms of benthic algal
succession in lotic environments. Ecol. 72: 183 5-1848.

Monod, J. 1950. La technique de culture continue, théorie et applications.
Annules Institute Pasteur, Paris 79: 390-410.

Mulholland, P. J ., Steinman, A. D., Palumbo, A V., Elwood, J. W. & Kirschtel,
D. B. 1991. Role of nutrient cycling and herbivory in regulating
periphyton communities in laboratory streams. Ecol. 72: 966-982.

Patrick, R 1976. The formation and maintenance of benthic diatom communities.
Proceedings of the Am. Philosophical Soc. 120: 47 5-484.

Patrick, R & Reimer, C. W. 1966. The Diatoms of the United States. Vol. 1.
Monogr. Acad. Nat. Sci. Philadelphia 13: 1-688.

49

Patrick, R & Reimer, C. W. 1975. The Diatoms of the United States. Vol. 2,
Part 1. Monogr. Awd. Nat. Sci. Philadelphia 13: 1-213 pp. '

Pianka, E. 1970. On r- and K- Selection. Am. Nat. 104: 592-597.

Pickett, S. T. A, Collins, 8. L. & Armesto, J. J. 1987. Models, mechanisms, and
pathways of successions. Bot. Rev. 53: 335-371.

Reynolds, C. S. 1984. Phytoplankton periodicity: the interactions of form,
function and environmental variability. Freshwat. Biol. 14: 111-142.

Reynolds, C. S. 1988. Functional morphology and the adaptive strategies of
ﬁ'eshwater phytoplankton. In Sangren, C.D. [Ed]. Growth and
reproductive strategies of ﬁeshwater phytoplankton. Cambridge
University Press, Cambridge, pp. 1-442.

Round, F. E., Crawford, R M. & Mann, D. G. 1990. The Diatoms. Biology and
morphology of the genera. Cambridge University Press, Cambridge, pp. 1-
747.

Schluter, D. & Ricklefs, R. E. 1993. Species diversity. An introduction to the
problem. In Ricklefs, R. E. & Schluter, D. [Eds]. Species diversity in
ecological communities. Historical and geographical perspectives. The
Chicago University Press, Chicago, pp. 1-10.

Sommer, U. 1981. The role of r-nd K-selection in the succession of
phytoplankton in Lake Constance. Acta. 0ecol./0ecol. Gener. 2: 327-342.

Sommer, U. 1983. Nutrient competition between phytoplankton species in
multispecies chemostat experiment. Arch. Hydrobiol. 96: 399-416.

Sommer, U. 1985. Comparison between steady state and non-steady state

competition: Experiments with natural phytoplankton. Limnol Oceanogr.
30: 335-346.

Sommer, U. 1990. Phytoplankton Nutrient Competition- From laboratory to lake.
In Grace, J.B. & Tilman, D. [Eds] Perspectives on Plant Competition.
Academic Press, San Diego, pp. 193-213.

50

Sommer, U. 1991. A comparison of the Droop and the Monod models of nutrient
limited growth applied to natural populations of phytoplankton. F unct.
Ecol. 5: 535-544.

Stevenson, R. J. 1983. Effects of current and conditions simulating autogenically
changing microhabitats on benthic algal immigration. Ecol. 64: 1514-
1524.

Stevenson, R J. 1984. Procedures for mounting algae in a syrup medium. Trans.
Am. Microsc. Soc. 103: 320-321.

Stevenson, R. J. 1990. Benthic algal community dynamics in a stream during and
after a spate. J. N. Am. Benthol. Soc. 9: 277-288.

Stevenson, R. J. 1995. Community dynamics and differential species performance
of benthic diatoms along a nutrient gradient. In Kociolek, J. P. & Sullivan,
M. J. [Eds] A Century of Diatom research in North America A tribute to
the Distinguished Careers of Charles W. Reimer and Ruth Patrick. Koeltz
Scientiﬁc Books, USA, Champaigrr, pp. 29-45.

Stevenson, R. J. 1996. An introduction to algal ecology in freshwater benthic
habitat. In Stevenson, R J ., Bothwell, M. L., Lowe R L. [Eds] Algal
Ecology. Academic Press, San Diego, pp. 3-30.

Stevenson, R. J. 1997. Scale-dependent determinants and consequences of
benthic algal heterogeneity. J. N. Am. Benthol. Soc. 16(1): 248-262.

Stevenson, R J. & Glover, R. 1993. Effects of algal density and current on ion
transport through periphyton communities. Limnol. Oceanogr. 38: 1276-
1281.

Stevenson, R J., & Pan, Y. 1994. Are evolutionary tradeoffs evident in responses
of benthic diatoms to nutrients? In Marino, D. [Ed] Proceedings of the
13‘" International Diatom Symposium, Biopress Ltd, Bristol, 71-81.

51

Stevenson, R J. & Peterson, C. G. 1989. Variation in benthic diatom
(Bacillariophycea) immigration with habitat characteristics and cell
morphology. J. Phycol. 25: 120—129.

Stevenson, R. J. & Peterson, C. G. 1991. Emigration and immigration can be
important determinants of benthic diatom assemblages in streams.
Freshwat. Biol. 26: 279-294.

Stevenson, R J ., Peterson, C. G., Kirschtel, D. B., King, C. C. & Tuchman, N. C.
1991. Succession and ecological strategies of benthic diatoms

(Bacillariophyceae): density-dependent grth and effects of nutrients and
shading. J. Phycol. 27 : 59-69.

Tilman, D. 1982. Resource competition and community structure. Princeton
University Press, Princeton, 296 pp.

Timan, D., Kiesling, R. L., Sterner, R, Kilham, S. S. & Johnson, F. A 1986.
Green, bluegreen, and diatom algae: taxonomic differences in competitive
ability for phosphorus, silicon, and nitrogen. Arch. Hydrobiol. 106: 473-
485.

Timan, D. & Wedin, D. 1991. Dynamics of nitrogen competition between
successional grasses Ecol. 3: 215-219.

Tuchman, N. C. 1996. The role of heterotrophy in algae. In Stevenson, R. J.,
Bothwell, M. L., Lowe R. L. [Eds] Algal Ecology. Academic Press, San
Diego, pp. 299-319.

van Dam, H., Mertens, A. & Sinkeldam, J. 1994. A coded checklist and ecological
indicator values of ﬁeshwater diatoms from the Netherlands. Neth. J.
Aquat. Ecol. 28(1): 117-133.

Velicer, G. J ., Schmidt, T. M. & Lenski, R E. 1999. Application of traditional and
phylogenetically based comparative methods to test for a trade-off in
bacterial growth rate at low versus high substrate concentration. Microb.
Ecol. 38: 191-200.

52

von Ende, C. N. 2001. Repeated-measures analysis. In Scheiner, S. M. &
Gurevitch, J. [Eds]. Design and Analysis of Ecological Experiments.
Oxford University Press, pp 134-157.

Winer, B. J., Brown, D. R. & Michaels, K. M. 1991. Statistical principles in
experimental design. McGraw-Hill, New York

Whittaker, R. H. 1975. Communities and ecosystems, 2“" edn. MacMillan, New
York, pp.1-162.

Wilkinson, L. 1989. Systat: The System for Statistics. Evanston, 822 pp.

53

Table 1.1: Summary of repeated measures (ANOVA) results of individual taxa
growth rates in N treatment when P saturated at 50 pg L". S-nutrient effect, T-
time effect, and S x T interaction. *-p value signiﬁcant at p<0.05,**-p value

signiﬁcant at p<0.001, ***-p value signiﬁcant at p<0.0001, ns- not signiﬁcant; all

p-values corrected with GreenhouseGeisser Epsilon for circularity.

 

 

 

S T L SxT
Taxa F5,11 E,“ F20,44
Achnanthes deﬂexa Reimer 3.15 * 3.51 * 1.23 ns
Achnanthidium minutissirnum (Kﬁtzing) Czar. 9.25 ** 4.21 " 1.86 ns
Cymbella aﬂinis Kutzing 0.44 ns 14.29 "* 1.18 ns
Cymbella cistula (Ehrenberg) Kirchner 1.09 ns 11.67 ”’ 0.72 ns
Encyonema minutum (Hilse) Mann 8.56 * 6.24 * 1.60 ns
Fragilaria crotonensis Kitton 3.03 ns 22.23 "W 1.11 ns
Fragilaria nanana Lange-Bertalot 1.22 as 20.01 "* 1.64 ns
Fragilaria vaucheriae (Ki‘rtzing) Petersen 2.38 as 33.68 *“ 0.85 ns
Meridion circulate (Grev.) Ag. 6.02 “ 23.70 "* 1.46 ns
Nitzschia acicularis (Ki'rtz) W. Sm. 1.62 ns 39.36 ”* 1.51 ns
Nitzschia dissipata (Kiitz) Grun. 1.02 ns 10.50 ** 0.96 ns
Nitzschia linearis (Ag. ex W. Sm.) W. Sm. 3.82 * 16.71 "t 1.01 ns
Nitzschia palea (Ki'rtz) W. Sm. 5.27 * 33.46 *" 1.24 ns
ﬁnedra ulna (Nitz) Ehr. 2.79 as 14.18 t" 0.52 ns

 

54

Table 1.2: Summary of repeated measures (ANOVA) results of individual taxa
growth rates in P treatment when N saturated at 500 pg L". S-nutrient effect, T-
time effect and S x T interaction. *-p value signiﬁcant at p<0.05, **-p value
signiﬁcant at p<0.001, ***-p value signiﬁcant at p<0.0001, ns — not signiﬁcant;
all p-values corrected with Greenhouse-Geisser Epsilon for circularity.

 

effects slr [SxT

 

 

Taxa Tsar F4,“ F20,44

Achnanthes deﬂexa Reimer 2.67 ns 0.84 ns 1.62 ns
Achnanthidium minutissirnum (Kutzing) Czar. 13.91 " 6.94 t 1.56 ns
Cymbella aﬂinis Kutzing 3.12ns 16.87 *‘* 1.64ns
Cymbella cistula (Ehrenberg) Kirchner 2.05 ns 8.45 ** 0.76 ns
Encyonema minutum (Hilse) Mann 4.03 t 7.17 * 1.02 ns
Fragilaria crotonensis Kitton 2.41 as 19.79 *** 1.02 ns
Fragilaria nanana LangeBertalot 1.07 ns 19.44 *" 1.00 ns
Fragilaria vaucheriae (Ki‘rtzing) Petersen 0.96 as 21.44 ‘** 0.7 ns
Meridion circulare (Grev.) Ag. 7.05 * 13.28 "* 1.12 ns
Nitzschia acicularis (Kﬁtz) W. Sm. 4.98 ‘ 26.66 ” 0.88 ns
Nitzschia dissipata (Kﬁtz) Grun. 6.41 * 5.47 * 0.8 ns
Nitzschia linearis (Ag. ex W. Sm.) W. Sm. 8.98 * 10.82 " 1.04 ns
Nitzschia palea (Kﬁtz) W. Sm. 9.84 *** 23.03 **"‘ 2.06 ns
Synedra ulna (Nitz) Ehr. 7.4 * 16.00 "* 1.32 ns

 

55

2 v 2 descends see I we

scomcﬁw 2 mac? Levee coco—2 55 codename @888qu 5380 ”mg 2an

 

 

 

me «new med- m: wood id and o. T 8d :w 35: cﬁoakm
md tum med sod dad- and mod and ad dz SEQ 853.2
and odd end on ad. 3d 5d «.9. mad ._2 “fauna c3852
and ad Ed and ad and mod n. F. 5d 02 Seneca» 3388.2
m: «dc. 3d 8d dd 3 Ed 3d- NF <2 £33083 3388.2
m: «.5- mod- and o.» «d end «.0 med 0.2 8333.8 scents:
m: odm- m d m: ..d. mud 3d 3d v; >n_ carotene; SsSeMER
mmd mde. m Pd m... dam 3d and ad _. z“. 385»: ELSREK
o: 2.. 8d- md omd 3d 5d ed and on. essence cremweak
m: mdn- «d me es. and ed edp mod 2m $833: geecbsm
o: N. 31 Ed on o; vmd 2d 3.? ed 00 Sago SEQEAU
o: F. 5. mod ed m... and Ed Em and <0 Eek? SEEAU
Nod mdv end we dd- dud mod 06 md 2< 5:25.832? EEVEEeEoV
3d tum ed as dd- ..md o: wood- «Pd o< 3333 woSSVSBV
«m 3. new. am 3. seem am .1 seem coco axe. coEEoo
051 may... mom

56

md v «M .ccmowawwm do: I m: 638%“? do: I m:

#:2de m wee? Loves coco—2 ﬁes “codename 958888 £380 ”v4 2an

 

 

 

en ea 8 so 8 8.0 «so to can use seesaw
we a: a: mod a: med Ed and 3 33m c3882
a: a: a: m: a: N_ d 3.0 S .0 3d @825 3385.2
8 ea 8 as 8 and 8 Rd mod Sesame SEES
a: m: me mm d mm m._ wed a: _ Mc§ao~oc SE32
a: a: we we 2 d dud ed odd mod 263.28 coasts:
no cd mod- wed ea 8.0 end we mg 3.2325: ELSEEK
8 ed mod and a: 2 d wed mod and SEE»: EEEMEK
an em; edd- we end 3 d 3.0 mod Nwd esteemed 35%.:ch
ed v; med add we end ed ﬁd em d £3355 goegem
m: a: 8d- med Wan LN md a: emd 38.4.8 3?ng
me a: o- dmd a: med vod we ed @3ch SRQEAD
wed wed dud mud and med 8 a: and SEEESSE EEEEEQEV
me 8 mad 8 a: vod- 8 E and SS? coaceeeﬁw
mm .M seam R ...M 58M am pm seem 823 88:80
Sex arm 8M

57

358:0 meaning??— >=Etmq some do 30.5 oumaoaom n: onE

 

 

 

 

seesaw ocean?

 

58

.88: as com mason— Eogmob 05 .«o coho wavesm 38de mam “:06me m d
356me 2 .m oﬁ wee—e omega Ac =Ed€oEUV $383 Rd? ”N; onE

9-4 was macadmonm

o

 

 

wmﬁ om mm m N

_

Soodvd

l

 

 

I\\Omﬁ’MNv-lo

i we assess

Ln 03 we 2 o

 

was N

UJ

1 _ _

 

 

 

bO'ﬁﬂ’MNr—O

(z-uro 811) n [[Aqdoromg)

59

“connect cod coho damage.
8863 mam vacuumed Hod bacon 8036 538 anodes» 38:98am .9
scone—aw comp—dz d .ANEo mzoov bade—c 5:58an 803% :33. ”m4 2:me

 

 

 

 

 

 

 

 

 

 

 

 

 

 

9-4 wad mEondmonm 9-4 wad cowoﬁz
E on mm m N o E: an RN ec 2 o
_ _ . .1 _ - INN _ q _ _ _ n1
.. . O
* ﬂ * I. l l Nﬁ am
.. .. - . L - .2 m
d t I 1 VA .1.
. . m.
. . - . . L in
3 l I W l 0.." H
3 S
- - L. l . o
e ed w.
.. r . n w; (cc
. _ . _ _ . _ L _ _ e _ X
.9 m 09

6O

.388 25 8088c
mod: 6355 ETSD .2ng ESPQQ .cwﬁ mﬁddﬁm . area .Bobormn ”main—mm mo 99%
882%: 38052 $88an 3:86 68: cam 305:: 5? bacon d .o .3568» macs—emcee

d Macadam cowoﬁz .o .345 £33 Emcee 3:88:80 Scene :88. .9583 m4 oEwE

 

 

 

9-4 wzv mEondmonm A74 mad 5on Z
m2 on mm m N o £2 Sm RN E 2 o
u A - - u _ OH H - q T H 1 Oﬁ
.w m
.m.
r - D
E u
0
w.
t - 2 .81
m
r i S ...,t.

 

 

 

 

 

 

2

 

61

.8808 05 8288 88— $2»qu 838 88 08. 3:8

392 2-2m $298.5 a: 88 28— 88 88% 2.2M win was on: 3852M 8 88 0::
358-3% $286 88 on: 386 =m8m .m >3 85 o mac 8853 £308 8 omgokém
”8883 m .2 .8238 2 .m .C-c.:_oo.m=oov $88 83on 5.88888 2839 a: 088E

2-4 wzv mEonmmonm A; wnv 5on Z
88 8 mm m m o . 82 NE 02 E 8 o

 

 

 

V".
c?

;
o.
o

1
V".
O

 

 

 

 

 

h
P
p
L
"2
H
r-
O
F—q

 

([-p°[-s[[ao'snao) 9112.1 mag

62

.mA 038 8 3 88338an 88,—. .8883 08: =a 8% En:
Agavmv _o>o_ mod “a 883—088 88m8wﬁ 808.88 889» a 803» 820880 nova—9:8
8882mm .. 8 Racks and u a 8M 083 d .m Bed 88 Z 33 888 £38» 882 ”m; 088m

233
2 Ac No We owe memo-
- -3
.. -3
19o
-3
_ _ _ _ S

 

d MO’I

 

 

 

 

63

.2 033 a a 208.9058 95 3&2. 2% 8 é in: .823
.$ $32 mod 8 80:30.88 885888 88082 mos—g a 83% ”820508 80:88.80 88.8on

- H is a; n h .38 2:: A m 33 Ea z 33 88 £38» a»: .8283 3 2:5

230A
NA ad 9o md od m6...
. mo-

02m
920 AZ 0 .
$4002... «0 oo

 

- d u

l

   
  

Lmd T
m

. L00 :0

.. -md

. . F . JWN.._

 

 

 

.2 039 a a 80.8.5058 98. $88 28 8 8 in: .628
.3 _o>o_ mod 8 noun—088 880886 8828.. 82g a 888 388508 838088 88825

- H ”@815 mg n . 528 ea: .0 m 33 2a 2 33 88 £32» 50: €880 2 88:

2301—

NA 0.0 0.0 m0 0.0 m.0..

4 1 . + m—m CI
0.0
m0
I <2 1 0.0
I -00
F r _ . N.~

 

 

1
0
m
.d

 

 

 

65

.mA 038 8 00 08830800 88H .0880 080 =0 8.0 Eu: .0808
-m0 ~0>0_ 3.0 00 802088 880807. 8000808 0038, a 820 808808 888088 88.80%

- 0 85 8.0 n 0 3:8 28. .0 m 33 9a 2 33 8.8 £320 802 .8080 2 28E

2300
NA 0.0 0.0 m0 0.0 m.0..
. _ 4 1 m0-

. m2 -8

 

8.98 2m
2882 .
.. :1 m 0 T
80 “.0 m
I 0m 0 Aw? I 0.0 d

.. . 0.0
_ _ _ _ N.H

 

 

 

66

.mg 038. E 00 0020030500 008,—. 002.000 0E: =0 8.0 En: .0025
.6 _0>0_ no.0 00 nous—00.80 0:80.803 8000500 0038, a 00020 uaosﬁooo 8:20:00 58.30%
- 0 00.8 00.0- n 0 2-2M 08:. .0 .m 304 0:0 Z 264 00000 5320 502 3.880 m; 0503

Z >54

NA 0.0 0.0 m0 00 m0-
_ _Uoomm MoOI

- 5.2100. 0.0
o>m
I can m.o
- . 0.0
. - 0.0

L h _ _ N.ﬁ

 

(1 M0’]

 

 

 

67

00:03 030000:

0:0 020000 00000000 0:: 00:09 .82008 03000005 00 £320 £05 0029, A >54
000 2 >64 00.02088: 800m .0 .m 304 0:0 2 33 .0 002000 0:09:00 m =0 8.0
8000000008 0005:: 0000:: 3 008880000 8.0 m030> 80200.80 ”04 050E

 

 

 

 

 

 

 

 

 

 

30500.05 3.200%:
Sadnem 08% 00% 8d 0._ Sadnem 08m 00m 8d 0.7
m
L m 0.. u u . m 0-
«10..
./ m...
0.0 m .r 0.0
m
. m. u .. .
1 m 0 1 m 0
0
M
p n F b p o.# w F n b h - o.~

mo'I uoywpuog)

68

.003? 000.8000: 0:0 0300000 000000000 08— 0000a .00 00
0:0 000 2 00.0 080008 0.0000000. 000803 8000 00 00000 0080 08—0800 m 0:0 :0 00.0 08000000
500 8 :000000 00:02 08 805 0008000 8.3000 808808 00.0 00800, 8000000000 H 0.0 0000000

 

 

 

 

 

$0500 2000
cam on“ mom
mx 0 q . 04-
O
m
- 0.0-0
M
-1 H
I/ / m
I

i / .. 0.0 m
J ..u
0
D.
0 0 0 CA W.
M

69

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2.5 i I I I
g 1.8 ' . 1
M 1.1 L /
0.4 " . "
.03. J
'1.0 J: . ‘ ‘ 2.5 j u I I i
2.5 ' T 18 . J
1.8 ' 3
g 1 l I- It
a: 1.1 ' ° "3‘ .
° ° 04' LA” *
04 ' / b ' d
' -0.3
-0.3 T 4.0 . .
2.5 I 2.5 i f
g 1.8 ' ' g 1.8 ' ‘
. . . J
m 1.1 M 1.]
04 ' N: i 04' .....- ‘
-0.3 ‘ l -0.3 ' - ‘
-1.o ‘ - L - -1.o - - L -
-1.0-0.3 0.4 1.1 1.8 2.5 -1.0-0.3 0.4 1.1 1.8 2.5
LowN LowN

Figure 1.8: Relationships between maximum observed growth rates
in high N and in low N conditions. a. Time R03, r = 0.66 (p=0.022);
b. Time R345, r = 0.40 (n.s); c. Time R610, r = -0.22 (n.s.); (1. Time
R1043 r = 0.32 (n.s.); e. Time R1}16 r = 0.21 (n.s.); r - Spearman
correlation coefficient; given p values represent signiﬁcant
correlation at 0.05 level (2-tailed), n=14 for all time periods

70

 

 

 

 

 

 

 

 

 

 

 

 

 

2.5 . r ..
g 1.8' J
1.1“
a: 04' >¢
-o.3+ «
. . . d
-1.0
b 25 - r .
2.5 . T -
1.8' r
1.8' g l
11
E 1.1* ‘94 , .9
0.4* T‘ 0'4 /
J .0.3 -
-o.3
10 4 . -1.0 + ‘
C e
25 . a 25 ' ' ' '
1.8L - 18 ‘
a; 1.1- -mg 11
. . .o . o.4* . i
0.4 f ,3:
-03 I W '0.3 I

 

 

 

 

 

-1 A - -

 

1.1.0-03, 0.4 1.1 1’3 2‘5 '-1.o-o.3 0.4 1.1 1.8 2.5
LowP LowP

Figure 1.9: Relationships between maximum observed growth rates
in high P and in low P conditions. a. Time RM, r = 0.29 (n.s.); b.
Time R345, r = - 0.28 (n.s.); c. Time R610, r=0.52 (p=0.044). (1. Time
R1043 r = 0.50 (p=0.05); e. Time R1346 r = 0.36 (n.s.); r - Spearman
correlation coefﬁcient; given p values represent signiﬁcant
correlation at 0.05 level (2-tailed), n=14 for all time periods

71

00205 0>00w00 000 030000 00080000 08.
00:09 do m 000 A3 Z 00.0 080008 £0000? 50800 8 0000.88 60$ 0080 850800 m 05
:0 00% 080.00% 030 0.0 mac—0 £38m 002030 808088 How 0030> 00:03:00 A: ._ 083m

 

 

 

 

 

 

$0500 25

22 m2: 2 0m m
1.01.0 «M0 2:- m
m w...
x . .02- m.
20 m.
8.0 m1
. .03 W.
M
m.
. . . . . OO.— m.
9
m

72

CHAPTER II
INTRA- AND INTERSPECIFIC COMPETITION FOR NUTRIENTS AND
LIGHT IN DIATOM CULTURES
INTRODUCTION

Few studies on species interactions within periphyton mats have been
reported in the literature (Patrick 1968, Hoagland et al. 1982, Stevenson et a1.
1991, Stevenson 1995) and there is a need for experimental validation of negative
(competitive) or possible positive interactions between algal species (McCormick
and Stevenson 1991). In diatoms, as in plant communities, patterns of abundance
could be shaped by competitive interactions. Interspeciﬁc competition for
nutrients was reported in the literature as a major determinant of algal community
structure (Kilham 1971). Theory predicts that under stable conditions (no stress
or disturbance) species with similar requirements can coexist when intraspeciﬁc
competition is stronger than interspecifrc competition (MacArthur and Levins
1967, Tilman and Pacala 1993). How this applies to benthic diatoms is still
unclear (McCormick 1996).

Benthic diatoms attach to the substrates available in streams and limitation
or abundance of shared resources like nutrients and light affect their growth.
There is a large amount of literature on the constraints on algal grth in streams
caused by limiting nutrients (Bothwell 1985, Grimm and Fisher 1986, Pringle
1987), light (Triska et al. 1983, Steinman and McIntire 1987, Hill 1996) or both
variables (Hill and Knight 1988, Hill and Harvey 1990, Rosemond 1993, 1994).

Most studies show diatoms as the dominant algal group in streams, therefore

73

concentrating on diatom populations can lead to important insights to periphyton
dynamics in naturally complex systems. Nutrient concentrations had different
effects on algal growth in shaded (< 40 umol m’1 5", Fisher and Likens 1973, Hill
and Knight 1988, Hill et al. 1995) and unshaded streams (ﬁrll sun light around
100 umol m'1 s", Bothwell 1985, Grimm and Fisher 1986).

Light requirements by individual benthic species are rarely studied (Hill
1996). Light attenuation by water, inorganic particles (Kuhl and Jorgensen 1994)
and sediment can leave little light available for the periphytic communities (5-6
umol m'2 5", Richardson et al. 1983). On a broad community scale, the outcome
of diatom population interactions for limiting nutrients measured as growth in
different light regimes is unknown. In natural communities, nonmotile adnate
algae (e.g., Achnwrthidium, Cocconeis and other monoraphid genera) are
hypothesized to be at a disadvantage when an overstory develops vertically
(Hoagland et a1. 1982, Steinmen 1996). Species from these genera are expected
to decrease in relative abundance (Hill 1996), while biraphid diatoms (e. g.,
C ymbella and Nitzschia) are expected to increase in abundance. Biraphid diatoms
can regulate their position along a light gradient through phototaxis (either gliding
or stalk formation, Fame-Fremiet 1951, Hoagland et al. 1982). Light effects have
been related to algal size (Laws 1975), where in low light conditions, larger
diatom cells grew faster than small cells due to lower respiration loss per unit
biomass (but see Michelson et al. 1979). Simultaneous nutrient and light

dependence has also been reported (Hannson 1992).

74

Clonal diatom populations have been used to deﬁne phytoplankton grth
patterns in controlled environments (Harrison et al. 1976, Tilman 1977, Grover
1988, 1989, 1990). Diatom experimental work has been performed in chemostats,
where limitation to nutrients lead to competitive exclusion (Taylor and Williams
1975) or coexistence in planktonic diatoms (Epply et al. 1971). Diatom growth
rate and competition for nutrients have been addressed with water column algae in
chemostats (Michelson et al. 1979, Tilman 1982), with benthic algae in culture
chambers (Stevenson 1995), and with marine benthic algae in nature (Sommer
1996). Genetically identical populations of periphyton diatoms in controlled
environments are excellent systems to address similar growth constraints in
benthic diatoms (Mann 1999). In natural and experimental studies with
multispecies membership, morphologically identical cryptic species might
obscure ecological differentiation among species. Manipulating nutrients and
light in a controlled environment allows isolation and determination of biomass
responses of diatom populations to common environmental variables (Manoylov
and Stevenson 2004).

When resources are limited, populations interact in ways leading to
decline in their performance measured as growth, reproduction, or survival
(Begon et al. 1996). The complexity of species competitive ability (ability to
reduce resources for others and tolerate reduction of resources by others) has been
related to evolutionary theory of coexistence when resources are limited (Aarssen
1983). The ability of a diatom to reduce the performance (growth, reproduction

or survival) of other diatoms is termed competitive effect (following Goldberg

75

and Fleetwood 1987, Goldberg 1996). Competitive effect examines the role
diﬂ‘erent neighbor species have on the same target species. Competitive effect
can be assessed with estimation of competitive indices (reviewed by Weigelt and
Jolliffe 2003). Competitive indices incorporate the effect of neighbor’s growth on
a target species (Goldberg and Barton 1992). The ability of a diatom species to
grow well with competition from other diatoms is termed competitive response
(i.e. response of different target species to the same neighbor). Competitive
response requires several target species to grow in mono and mixed cultures.
Competitive effect and competitive response can be unique or complementary
proxies in assessing species-speciﬁc competitive ability (Goldberg and Werner
1983, Goldberg 1996).

Intra- and interspeciﬁc competition can be studied in many ways
(Goldberg and Barton 1992). In plant competition, several studies compared
species performance in low and high densities of the same species (’community
density series’ Goldberg et a1. 1995). Alternatively, species performance in
monocultures has been compared with species performance in mixtures with other
species (Campbell and Grime 1992, Goldberg and Fleetwood 1987). In this
experiment, I used a combination of the two models: ﬁrst, a target species was
grown at low and high initial density in limiting or luxurious resource (nutrient
and light) levels; second, the same target species was grown with another diatom
species at varying resource levels.

This study addressed performance of clonal populations of A clmanthidium

minutissimum (Kt‘rtzing) Czarnecki in different densities and environmental

76

conditions and with other diatom species. The ﬁrst objective of this study was to
assess growth rate of A. minutissimum in four environmental conditions created
with low and high levels of nutrients and light. Higher growth rates were
expected when nutrients and light were high. Strong intraspeciﬁc competition
was expected at limiting levels of nutrients and light, and those negative
interactions should be reduced with higher resource availability (Tilman 1982).
The second objective of the study was to assess the competitive effect of
contraspeciﬁc diatom on A. minutissimum. Interspeciﬁc competition was
examined with another adnate diatom, Cocconeis placentula var. Iineata Krasske
and with a larger, stalk-forming diatom, Cymber cistula (Ehrenberg) Kirchner.
These species are abundant and widely distributed in North America (Patrick and
Reimer 1966) and occur together in natural assemblages (personal observations).
MATERIAL AND METHODS

Diatom cultures were purchased ﬁ'om Loras College Freshwater Diatom
Culture Collection (maintained and operated by Dr. D. Czarnecki).
Achnanthidium minutissimum (strain identiﬁcation LI468), Cocconeis placentula
var. Iineata (023), and Cymbella cistula (LISSO) were used in the experiment.
Taxa in the mixed treatments were selected because of their morphological and
growth traits. Achnanthidium minutissimum and Cocconeis placenlula var.
Iineata were assumed to have similar modes of attachment to the substratum
(adnate), but to differ in shape and size. The smaller, A. minutissimum, is
lanceolate and the larger, Cocconeis placentula var. Iineata, is oval. The second

mixture of A. minutissimwn and Cymbella cistula was chosen because those two

77

species have different modes of attachment, shape and size. Cymber cistula
grows on stalks. It is wedge shaped and larger.
Experimental design

The experiment was conducted May 16 to June 6, 2004. The cultures
were grown in modiﬁed Bold‘s medium (Sigma-Aldrich Scientiﬁc, Inc) for two
weeks prior to inoculation of the experimental channels. The medium was diluted
10X and 28.4 mg/L sodium silicate (NazsiO3.9H20) was added. The experiment
was conducted in a controlled environmental chamber (Percival Scientiﬁc, Inc.
Model I-36LL). Temperature was kept at 18°C, 14 h light to 10 h dark cycle, and
humidity was 90 :1: 5%. Full light in the chamber was 96-100 umol m'2 s".
Shaded (low light) treatments were created with the use of 0.9 neutral density
ﬁlter (LEE Filters, Burbank CA, item # 211), which reduced the full light to 10%.
A remote alarm system signaled any deviation ﬁ'om the expected conditions.

Custom circular channels (Plas Labs, Inc., Lansing, MI) were constructed
with 49 cylinders (diameter 10 cm, height 6 cm) of clear plastic (3 mm thick),
glued onto a 70 x 70 cm clear plastic (7 mm thick) platform (Figure 2.1 a, b).
Within each cylinder, a smaller cylinder 4 cm in diameter was centered and glued
to create a 3 cm wide circular channel between the cylinders. Six unglazed tiles
(2 x 2 cm) were scrubbed, washed with distilled water, and arranged ﬂat in the
channels (Figure 2.1 c). Each circular channel was a closed system. A clear
plastic lid (3 mm thick) covered the whole structure to slow evaporation. The
platform was placed on a rotary shaker (45 rpm, EL600 Heavy-Duty Orbital-

Reciprocal Shaker, Eberbach, Inc.) to move water in the channels. The current

78

velocity in the channels was 5 cm s". The shaker and the platform were
positioned within the incubator. The peripheral 24 channels on the artiﬁcial
structure (Figure 2.1) were selected for the shading. Within each of the 24 shaded
or ﬁrll light channels, nutrient treatments and species combinations were
randomly assigned in triplicate. F orty-eight of the available 49 channels were
used.

Bold’s algal medium (Bolds, Sigma-Aldrich Scientiﬁc) was modiﬁed by
adding silica (as sodium silicate) and introducing nitrogen (N, as cobalt nitrate
and sodium nitrate) and phosphorus (P, as potassium phosphate monobasic and
potassium phosphate dibasic) in different concentrations (Table 2.1) to create two
nutrient treatments: control (C) and high nutrient (NP). In the C treatment,
nutrient concentrations were roughly equivalent to those considered to be growth-
limiting for periphytic diatoms (Bothwell 1989), 25 pg N/L and 3 pg P/L. 250 pg
N/L and 35 pg P/L were used in the NP treatment, which was designed to satisfy
nutrient demand in the cultures. Before each use, the culture medium was ﬁlter-
sterilized using 0.45 pm ﬁlters (APHA 1998). Each circular channel was ﬁlled
with 80 ml modiﬁed medium, which was 2 cm deep over the tiles. Twenty ml of
fresh medium was added every other day to replenish nutrients depleted by algal
uptake.

Four species combinations were used (Table 2.2): 1. a monoculture of
Achnanthidium minutissimum (A); 2. monoculture with a double initial density of
A. minutissimum (AA); 3. mixture of A. mimtissimum and Cocconeis placentula

var. Iineata (AC); and 4. mixture of A. minuﬁssimum and Cymbella cistula

79

(ACM). In the later two cases, species were inoculated in approximately equal
biosurface (density x surface area) at the beginning of the experiment.

To monitor nutrient concentrations, water samples were taken before
periphyton sampling every 3 days ﬁom randomly selected channels (2 from
shaded and 2 ﬁom full light, n = 4). NO3-N and PO4-P concentrations were
measured by automated cadmium reduction and automated ascorbic methods,
respectively (Scalar, Inc). Periphyton was sampled by collecting 1 tile at random
from the same position in each channel every 3 days for 6 consecutive sampling
dates. After sampling, the diatoms were preserved with M3 (APHA 1998). The
removed tile was replaced with a clean tile to preserve ﬂow conditions. Diatoms
were scraped from the tiles with razor blades and toothbrushes. Cell densities
were enumerated and measured using a Palmer-Maloney counting chamber at
400X magniﬁcation (Leica DMLB microscope). This method allowed detection
of the cytoplasm in the cells and good resolution of cell wall structure of the 3
diatom species. At least 300 cells with an intact chloroplast were identiﬁed and
enumerated in each channel for each sampling date. Standard diatom
morphological measurements were taken from individuals in each treatment. For
all cultures the following measurements were performed: cell apical length,
transapical width, frustule depth in the middle and for the purposes of taxonomic
evaluation striae in 10 pm (mainly in Cymbella cistula). At least 15 specimens
were measured at random in each channel.

For all inoculum and sampled populations, cell volumes (V, pm3) and cell

surface area (S, pmz) in each treatment were calculated using formulae from

80

Hillebrand et al. (1999). Volume and surface area for A. minutissimum and
Cocconeis placentula var. Iineata were calculated with the following formulae:
V=(1r/4)*a*b*c and S=(1r/4)"'(a*b+[a+b]*c), where a is the apical length, b is
transapical width and c is the depth of the mantle. The volume of Cymbella
cistula was calculated using the formula: V=4/6*1t"b2*a*(B/360) and cell surface
area was calculated using s= ((u‘b*a)/2)+c*((b+((a/2)2/((sqrt (a/2) 2)~b2))*sin'
WWW 2W"), where a is the apical length, b is transapical width, c is the
pervalvar axis on dorsal side and B is the angle between the two transapical sides,
calculated as sin Bl2=c/(2*b). Cell surface area calculation and use in biosurface
calculation (total cell surface area, mmz) was necessary to offset the potential
overestimation of the larger C. cistula and underestimation of the smaller A.
minutissimum that could result from using biovolume.

Absolute cell densities (D, cells mm'z) were calculated for each treatment.
Total biosurface (B) was calculated as the average cell surface area x cell density
in each treatment. Growth rates were calculated based on species density for each
treatment. Grth rates (R) were estimated as daily per capita changes in cell
density (in each channel): R = (ln D (m) — 1n D (1))l3, where D (t) and D ((+1) are
total densities on consecutive times and 3 was the uniform number of days
between sampling days. Growth rates (In cells cell'l d'l) were calculated for each
species in each channel, for each time period. Growth rates for each period were
designated as Rxy for the growth change between community development day x

and day y.

81

Treatment effects on growth rates were tested using repeated measures
analysis of variance (RM ANOVA) in each culture combination. The univariate
approach to RM ANOVA takes into account temporal autocorrelation within each
channel, which can violate the assumption of independence among samples. All
results were adjusted for circularity with Geisser-Greenhouse epsilon (G-G, von
Ende 2001) which is the most conservative correction of degrees of ﬁeedom for
independence among samples within a channel. Ifthe treatment effect was
signiﬁcant, time effects were evaluated with polynomial, proﬁle analyses in RM
ANOVA, in which differences are constructed ﬁom adjacent levels of the time
factor.

I assessed growth of clonal A. minutissimum in limiting concentrations of
nutrient and very low light. Addition of nutrients in low light was expected to
increase growth rate and competition for light between taxa in this treatment,
where ability to produce stalks by C. cistula should be advantageous. Nutrient
enrichment in low light was expected to similarly stimulate A. minutissimum and
Cocconer's placentula var. Iineata. With the addition of light in low nutrients, A.
minutissimum was expected to increase growth rate, as a good competitor for
limiting nutrients (reported as indifferent in Van Dam, 1994), while grth rate of
Cocconeis placentula var. Iineata should decline (reported as eutraphentic,
Butcher 1947, Van Dam 1994). Throughout the time periods of community
development I hypothesized that addition of both nutrients and light would
alleviate all competitive interactions.

Competitive eﬁ'ect

82

A regression based method for measuring the competitive eﬂ‘ect of
conspeciﬁc and/or contraspeciﬁc individuals on a ‘target’ species was developed
by Goldberg and Werner (1983). In my modiﬁed model, the slope of a regression
line (relating growth rate as a function of biosurface) is a measure of the
performance of the target species as a ﬁlnction of the biosurface of all
neighboring individuals. Competitive effects were estimated with competitive
indices assessing the proportional competitive effects of neighbors (other diatoms
either conspeciﬁc or contraspeciﬁc) on the grth rate of the target species, A.
minutissimum (GRA). The following model was used:

GRA = C + (GA* BA) + (00c * Bcc) + (0cm * BCM),

Here, C is an estimate of maximum growth rate and BA, Bee, BCM are
biosurface of Achnanthidium minutissimum, Cocconeis placentula var. lineata
and Cymbella cistula, respectively. The following were estimates of competitive
indices: all is a measure of the competitive performance of A. minutissimum in
each culture combination; ace performance of C. placentula var. lineata in the AC
culture combination; and com performance of Cymbella cistula in the ACM
culture combination. In the mixed cultures, together with conspeciﬁc inﬂuence,
performance of A. minutissimum proportionally measures the interspeciﬁc
inﬂuence of the other taxa. If (M, ace and «CM are equal to 0, there was no
competition. In each species combination, intra- and interspeciﬁc competition
respectively was low when slopes were close to 0 and not signiﬁcamly different
ﬁom each other (slope comparison, Sokal and Rohlf 1995). Any (1 >0 indicated

facilitation and a<0 indicated competition. All coefﬁcients have variance

83

associated with them, which allowed comparison of the competitive response of
the target species in each culture combination. a estimates were graphically
presented for high biosurface gradient only in each treatment, where stronger
competition was expected. All statistical analyses were performed with
SYSTAT® 10 (Wilkinson 1989).

RESULTS

Nutrient concentrations in the control channels were low, 19-25 pg N03-
N/L and 1.5-3 pg POa-P/L. Adding enriched medium to the high nutrient
treatments resulted in an eight to ten-fold increase in N and P concentrations (Fig.
2.2).

Observations of cells on tiles showed cells attached to the tile surface and
covered it completely by the end of the experiment. Achnanthidium
minutissimum did not grow in a uniform pattern in any treatment, it was observed
in single and multiple cells, within and around mucilage globules. Those globular
structures protruded above the tile surface, remained adhered to the tiles and kept
growing in size. The protruded in the water column ﬂocks of cells, attached at
one end to the tile, potentially were a source for immigrating cells. Judging ﬁ'om
the healthy appearance of the chloroplasts of the cells in all those structures,
growth was equally successful in the different growth structures. This diversity in
extracellular mucilage production was not observed for Cocconeis placentula var.
lineata. Cells remained attached close to the surface without elevating over the
tiles and without visible aggregation of cells. Healthy looking single cells of

Cocconeis placentula var. Iineata were observed within the mucilage globules

84

created by A. minutissimum. Cells of Cymber cistula initially remained close to
the substrate, but with increasing time and density, short and long stalks formed
and elevated the cells above and around the mucilage aggregates of A.
minutissimum.

Total density increased for all species combinations with nutrients
addition (p<0.0001, Fig. 2.3). The increase in density was greatest at the early
time periods when density was low. Total density increased to 8000-9000 cells’l
mrn'1 after day 9 of community development. Total density was consistently
higher in the shaded treatment (Fig. 2.3). Total biosurface of Achrwtthidium
minutissimum did not diﬁ‘er signiﬁcantly with addition of nutrients in low light,
decreased with higher light availability in low nutrient treatments and increased
with nutrient addition in high light treatments (Fig. 2.4).

Growth rates of A. minutissimum decreased with density and time in
almost all treatments. Grth rate of A. minutissimum varied signiﬁcantly with
treatment in most culture combinations (Table 2.3, Figs 2.5, 2.6). There were no
signiﬁcant treatment x time effects for any culture combination (Table 2.3).
Polynomial analysis of the time period that contributed to the signiﬁcant
treatment or time effect revealed that growth rate from day 6 to 9 contributed
(cubic trend signiﬁcant p=0.0175 ANOVA, Fig. 2.5) to the signiﬁcance when all
treatments and consecutive times were analyzed together.

Nutrient or light addition did not have a signiﬁcant effect on
Achnanthidium growth rate when A. minutissimum grew alone or with Cocconeis

placentula var. Iineata (Fig. 2.5). During the initial growth period (Fig. 2.5 open

85

bars) monocultures of A. minutissimum grew at the same rate regardless of the
initial density (Fig. 2.5). Grth rates in monocultures were always higher than
growth rate of A. minutissimum in mixture with another diatom. Addition of
nutrients signiﬁcantly increased A. minutissimum grth at higher density only
when A. minutissimum was mixed with Cymbella cistula and in low light (Fig. 2.5
dark bars). Growth rate at higher density was independent of nutrient addition
with excess light (Fig. 2.5).

Most regression coefﬁcients (competition coefﬁcients), whether measured
using the whole biomass gradient or just high biomass conditions were negative
when they were signiﬁcantly different from 0 (Table 2.4, Fig. 2.6). Within each
culture combination, the steepest slopes were observed in the low light-low
nutrient treatment (T able 2.4). Because of the similarities in the direction of the
regression coeﬁicient estimates along the whole biomass gradient and at high
biomass, competition coefficients were studied only in high biomass conditions.
With addition of nutrients and light, biosurface tripled (compare range in
biosurface values in Figs.2.6 a. with 2.6 d.).

Low nutrients and light had strong effects on intraspeciﬁc competition in
three of the culture combinations (Fig. 2.7), with coefﬁcients of -2.4, -O.95, and -
0.59 in cases: A, AC and ACM respectively. When nutrients were abundant,
competitive interactions in populations of clonal A. minutissimum were not
detected, regardless of the neighbor identity (Fig. 2.7).

Cocconeis placentula var. Iineata grew slower than A. minutissimum when

nutrients were limiting, (Fig. 2.8 a and Table 2.4), the strongest competitive index

86

(-2.14) was estimated in low nutrient and high light conditions. No interspeciﬁc
competitive interactions between A. minutissimum and Cymbella cistula were
observed in comparison with the very little evidence observed with Cocconeis
placentula var. Iineata. Most slope estimates were not different from 0 and
positive (Fig. 2.8 b and Table 2.4).
DISCUSSION

Intraspeciﬁc competitive eﬂ‘ects in all culture combinations regulated
grth of Achnanthidium minutissimum. Grth rate of Achnanthidium
minutissimum decreased with increasing biosurface of algae in all treatments.
Initial grth rate of Achnanthidium minutissimum were reduced by presence on
interspeciﬁc competitors, reduction of grth rate was independent of treatment
Intraspeciﬁc competition was strongest when nutrients and light were low and A.
minutissimum was in monocultures. Intraspeciﬁc competition was relieved with
addition of nutrients or both nutrients and light. Light availability relieved
constraint on Cocconeis placentula var. Iineata growth and enabled strong
interspeciﬁc competitive effects on A. minutissimum. Addition of nutrients
signiﬁcantly increased A. minutissimum grth at higher density only when A.
minutissimum was mixed with Cymbella cistula. The observed potential
facilitation of the presence of C. cistula on the growth of the target species in low
resources suggests potential coexistence in nature. The observed cell densities
were comparable to other studies where competitive interactions were observed in

nature (Butcher 1947, Rosemond 1993).

87

When initial densities of Aclmcmthidz'um minutissimum were higher
intraspeciﬁc competition at low resources was not as strong as observed in
treatments with lower initial density. Growth rates of A. minutissimum in
monocultures were always higher than its grth rate in mixture with another
diatom. These results need further research with the same diatom and different
diatoms in monocultures with varied densities. For example, chemical signals in
the denser initial monoculture might have inﬂuenced diatom growth.

Achnanthidium minutissimum has been reported from many ecologically
diverse environments (Potapova and Charles 2003). In in situ nutrient enrichment
experiments, Cholnoky (1968) and Kawecka (1981) reported high growth rates
for A. minutissimum in both low and high nutrient conditions. Consistently, this
taxon has been reported as indifferent to nutrient concentrations (Van Dam et al.
1994). Clones from the same species have shown great afﬁnity for nutrients,
depending on the natural conditions. Diatoms potentially adapted to growth in
low nutrients were able to grow at very low nutrient concentrations (Chapter 1 of
this dissertation, Carpenter and Guillard 1971). In this study, using genetically
identical populations, I characterized response of one phenotype that might diﬁ‘er
from the response of other populations of the same species isolated ﬁom different
natural habitats. However, many of the above observations for A. minutissimum
indicate it is a good competitor for nutrients when they are in low supply.

Species-speciﬁc diatom responses to light has not been studied in
relationship to competition, but in ﬁeshwater benthic systems total community

response to light has been reported (DeNicola et al. 1992, Hill 1996, Wellnitz and

88

Rinne 1999). One possible mechanism explaining the observed higher biomass in
shaded streams in this study compared to ﬁrll light is that the species I
manipulated may have low photosynthetic optima. Shade adaptation increases
carbon uptake at low light, and it is common response of phytoplankton
(Richardson et al. 1983). This has been reported in single culture studies
(Dubinsky et al. 1986, Falkowski and LaRoche 1991) and multispecies
communities (Hill and Boston 1991, Hill et al. 1995). At high light,
photoinhibition might have been possible among ‘ shade-tolerant’ diatom species
(Hill et a1. 1995) manipulated in this experiment.

In the experiment, Achnanthidium minutissimum in limited light
treatments produced mucilage layers, with very healthy growth within them.
Taxa from Achnanthidium and Cocconeis usually attach close to the substrate and
do not produce long chains, but the ability to produce mucilage might have been
advantageous for A. minutissimum in the low light and low nutrient environment.
C. placentula var. Iineata grew slower than A. minutissimum in this experiment
potentially due to inability to elevate itself when it was light and nutrient limited.
In the other interspeciﬁc competition treatment, Cymbella cistula had the ability
to extend well above the surface in the shaded treatments (DC and DNP),
potentially reducing shading effects and increasing surface area for attachment.
However, because A. minutissimum was able to attach to the stalks of C. cistula or
aggregate in mucilage globules, A. minutissimum was elevated over the substrate
also. These results suggest a potential for A. minutissimum to competitively

exclude C. placentula var. Iineata, but coexistence with C. cistula was possible.

89

Intra- and interspeciﬁc response to competition was relieved when
nutrients and light were available. It had been hypothesized that rrricroalgae
would not be successful competitors at both low and high resource concentrations,
but instead they would pursue one strategy or the other (McCormick 1996). This
experimental research presented a few successful ecological strategies of one
common diatom in different taxa combinations and nutrient concentrations. This
experiment is the ﬁrst reported in the literature where competitive effects of other
diatoms was evaluated on clonal populations of Achmmthidium minutissimum.
Evolutionary traits like small size might be more important in interspeciﬁc
competition than nutrient uptake efﬁciency. Competitive abilities are restricted
by the morphological differences of the competing species (e. g. size and bio-
surface).

In conclusion, this was the ﬁrst attempt to characterize the competitive
ability of Achnanthidium minutissimum in a controlled setting. I was able to
create periphyton communities in culture conditions and to evaluate the ecological
effects of other diatoms on clonal populations of A. minutissimum. Light had a
strong negative effect on A. minutissimum grth when nutrients were limiting
and had a positive effect when nutrients were available. The prediction of
competitive effects being strongest at limiting resources was conﬁrmed in this
experiment. The mechanisms to compete for low nutrients and low light of these
morphologically diverse diatoms may be different. Achmthidium minutissimum
and Cocconeis placentula var. lineata are monoraphid diatoms, but the former

had potentially lower nutrient requirements than the latter. The morphologically

90

diﬂ‘erent pair: A. minutissimum and Cymbella cistula were both good competitors
for limiting resources. The stalks produced by C. cistula provided substratum for

A. minutissr’mum colonization.

91

LITERATURE CITED

Aarssen, L. W. 1983. Ecological combining ability and competitive aombining
ability of plants: toward a general evolutionary theory of coexistence in
systems of competition. Am. Nat. 122: 707-731.

APHA 1998. Standard Methods for the Examination of Water and Wastewater.
Part 10300. American Public Health Association. Washington, D. C.

Begon, M., Harper, J. L. & Townsend, C. R 1996. Ecolog. Individuals,
populations and communities. Blackwell Science.

Bothwell, M. L. 1985. Phosphorus limitation of lotic periphyton grth rate: an
intersite comparison using continuous-ﬂow troughs (Thompson River
system, British Columbia). Limnol. Oceanogr. 36: 527-542.

Bothwell, M. L. 1989. Phosphorus—limited growth dynamics of lotic periphytic

diatom communities: areal biomass and cellular growth rate responses.
Can. J. Fish. Aquat. Sci. 46: 1293-1300.

Butcher, R W. 1947. Studies in the ecology of rivers. VII. The algae of
organically enriched waters. J. Ecol. 35: 186-191.

Carpenter, E. J. & Guillard, R. R. 1971. Intraspeciﬁc differenced in nitrate half-
saturation constants for three species of marine phytoplankton. Ecol. 52:
183-185.

Campbell, B. D. & Grime, J. P. 1992. An experimental test of plant strategy
theory. Ecol. 73: 15—29.

Cholnoky, B. J. 1968. Die Okologie der Diatomeen in Binnengewd‘ssem. 699 pp.
J. Cramer Verlag, Weinheim.

DeNicola, D. M., Hoagland, K. D. & Roemer, S. C. 1992. Inﬂuence of canopy
cover on spectral irradiance and periphyton assemblages in a prairie
stream. J. N. Am. Benthol. Soc. 11: 391-404.

Dubinsky, A., Falkowski, P. G. & Wyman, K. 1986. Light harvesting and
utilization by phytoplankton. Plant Cell Plgisiol. 27: 1335-1349.

Epply, R. W., Rogers, J. N., McCarthy, J. J. & Soumia, D. 1971. Light/dark
periodicity in nitrogen assimilation of the marine phytoplankters

92

Skeletonema costalum and Coccolithus lualeyi in N-limited chemostat
cultures. J. Phycol. 7: 150—154.

Falkowski, P. G. & LaRoche, J. 1991. Acclimation to spectral irradiance in algae.
J. Phycol. 27: 8-14.

Faure-Fremiet, E. 1951. The tidal rhythm of the diatom Hantzschia amphioxys.
Biol. Bull. (Woods Hole, Mass.) 100: 173-177.

Fisher, S. G. & Likens, G. E. 1973. Energy ﬂow in Bear Brook, New Hampshire:
an integrative approach to stream ecosystem metabolism. Ecol. Monogr.
43: 421-439.

Goldberg, D. E. 1996. Competitive ability: deﬁnitions, contingency and
correlated traits. Phil. Trans. R. Soc. Land. 351: 1377-1385.

Goldberg, D. E. & Barton, A. M. 1992. Patterns and consequences of
interspeciﬁc competition in natural communities: a review of ﬁeld
experiments with plants. Am. Nat. 139: 771-801.

Goldberg, D. E. & Fleetwood, L. 1987. Competitive eﬂ‘ect and response in four
annual plants. J. Ecol. 75: 1131-1143.

Goldberg, D. E. & Werner, P. A. 1983. Equivalence of competitors in plant

communities: a null hypothesis and an experimental approach. Am. J. Bot.
70: 1089-1104.

Goldberg, I). E., Turkington, R. & Olsvig—Whittaker, L. 1995. Quantifying the
community-level consequenced of competition. Folia Geobot. Plzytotaxon.
30: 231-242.

Grimm, N. B. & Fisher, S. G. 1986. Nitrogen limitation in a Sonoran desert
stream. J. of N. Am. Benthol. Soc. 5: 2-15.

Grover, J. P. 1988. Dynamics of competition in a variable environment:
experiments with two diatom species. Ecol. 69: 408-417.

Grover, J. P. 1989. Inﬂuence of cell shape and size on algal competitive ability.
J. Phycol. 25: 402-405.

Grover, J. P. 1990. Resource competition in a variable environment:
phytoplankton growing according to Monod’s model. Am. Nat. 136: 771-
789.

Hansson, L.-A 1992. Factors regulating periphytic algal biomass. Limnol.
Oceanogr. 37: 322-328.

93

Harrison, P. J., Conway, H. L. & Dugdale, R C. 1976. Marine diatom grth in
chemostats under silicate or ammonium limitation. 1. Cellular chemical
composition and steady-state growth kinetics of Skeletonema costatum.
Mar. Biol. (Bed) 35: 177-186.

Hill, W. R 1996. Effects of light. In: Stevenson, R J ., Bothwell, M. L and Lowe,
R L. [Eds] Algal Ecology: Freshwater Benthic Ecosystems, Academic
Press, San Diego, pp.121-148.

Hill, W. R & Boston, H. L. 1991. Community development alters
photosynthesis-irradiance relations in stream periphyton. Limnol.
Oceanogr. 36: 1375-1389.

Hill, W. R. & Harvey, B. C. 1990. Periphyton responses to higher trophic levels
and light in a shaded stream. Can. J. Fish. Aq. Sc. 47: 2307-2314.

Hill, W. R & Knight, A W. 1988. Nutrient and light limitation of algae in two
northern California streams. J. Phyc. 24: 125-132.

Hill, W. R, Ryon, M. G., & Schilling, E. M. 1995. Light limitation in a stream
ecosystem: responses by primary producers and consumers. Ecol. 76:
1297-1309.

Hillebrand, H., Durselen, C-D., Kirschtel, D., Pollingher, U. & Zohary, T. 1999.
Biovolume calculation for pelagic and benthic microalgae. J. Phycol. 35:
403-424.

Hoagland, K. D., Roemer, S. C. & Roemer, J. K. 1982. Colonization and
community structure of two periphyton assemblages, with emphasis on the
diatoms (Bacillariophyceae). Am. J. Bot. 69: 188-213.

Kawecka, B. 1981. Sessile algae in European mountain streams. 2. Taxonomy
and autecology. Acta Hydrobiol. 23: 17-46.

Kilhanr, P. 1971. A hypothesis concerning silica and the freshwater planktonic
diatoms. Limnol. Oceanogr. 16: 10-18.

Kuhl, M. & Jorgensen, B. B. 1994. The light ﬁeld of microbenthic communities:
Radiance distribution and microscale optics of sandy coastal sediments.
Limnol. Oceanogr. 39: 1368-1398.

Laws, E. A 1975. The importance of respiration losses in controlling the size
distribution of marine phytoplankton. Ecol. 56: 273-294.

94

MacArthur, R & Levins, R 1967. The limiting similarity, convergence, and
divergence of coexisting species. Am. Nat. 101: 377-385.

Mann, D. G. 1999. The species concept in diatoms. Phycologia 38: 437-495.

Mann, D. G., Chepumov, V. A. & Droop, S. J. M. 1999. Sexuality,
incompatibility, size variation and preferential polyandry in natural
populations and clones of Sellaphora pupula (Bacillariophyta). J. Phycol.
35: 152-170.

Manoylov, K. M. & Stevenson, R. J. 2004. Cymbella excisa Ki’rtzing in different
nutrient conditions. In Poulin, M. [Ed] Preceedings of the 1 7th
International Diatom Symposium. Ottawa, Canada, Biopress Ltd., pp. 223-
233.

McCormick, P. V. 1996. Resource competition and species coexistence. In:
Stevenson, R. J ., Bothwell, M. L and Lowe, R. L. [Eds] Algal Ecology:
Freshwater Benthic Ecosystems, Academic Press, San Diego, pp. 229-252.

McCormick, P. V. & Stevenson, R J. 1991. Mechanisms of benthic algal
succession in different ﬂow environments. Ecol. 72: 1835-1848.

Michelson, M. J ., Maske, H. & Dugdale, R. C. 1979. Nutrient-deternrined
dominance in multispecies chemostat cultures in diatoms. Limnol.
Oceanogr. 24(2): 298-3 15.

Patrick, R 1968. The structure of diatom communities in similar ecological
conditions. Am. Nat. 102: 173-183.

Patrick, R & Reimer, C. W. 1966. ﬁre diatoms of the United States, Exclusive of
Alaska and Hawaii. Monogr. 13. Vol. 1, 688 pp. The Academy of Natural
Sciences of Philadelphia, Philadelphia.

Potapova , M. & Charles, D. F. 2003. Distribution of benthic diatoms in US.
Rivers in relation to conductivity and ionic composition. Fresh. Biol. 48:
131 1-1328.

Pringle, C. M. 1987. Effects of water and substratum nutrient supplies on lotic
periphyton growth: An integrated bioassay. Can. J. Fish. Aquat. Sci. 44:
619-629.

Richardson, K., Beardall, J. & Raven, J. A 1983. Adaptation of unicellular algae
to irradiance: An analysis of strategies. New Phytol. 93: 61-72.

Rosemond, A D. 1993. Interactions among irradiance, nutrients, and herbivores
constrain a stream algal community. 0ecologia 94: 585-594.

95

Rosemond, A D. 1994. Multiple factors limit seasonal variation in periphyton in
a forest stream. J. N. Am. Benthol. Soc. 13: 333-344.

Sokal, R R & Rohlf, F. J. 1995. Biometry. The principles and practice of
statistics in biological research. Third Edition, pp. 887. W. H. Freeman &
Co. New York.

Sommer, U. 1996. Nutrient competition experiments with periphyton ﬁom the
Baltic Sea. Marine Ecol. Progress Ser. 140: 161-167.

Steinman, A. D. 1996. Effects of grazers on freshwater benthic algae. In:
Stevenson, R J ., Bothwell, M. L and Lowe, R L. [Eds] Algal Ecology:
Freshwater Benthic Ecosystems, pp. 341-3 66. Academic Press, San Diego.

Steinman, A D. & McIntire, C. D. 1986. Enrichment of autotrophic production
by nutrient addition in a coastal rainforest stream on Vancouver Island. J.
Fish. Res. Board Can. 35: 28-34.

Stevenson, R J. 1984. Procedures for mounting algae in a syrup medium. Tran.
Am. Microsc. Soc. 103: 320—321.

Stevenson, R. J. 1995. Community dynamics and differential species
performance of benthic diatoms along a nutrient gradient. In: Kociolek, J.
P. and Sullivan, M. J. [Eds] A Century of Diatom research in North
America. A tribute to the Distinguished Careers of Charles W. Reimer and
Ruth Patrick, Koeltz Scientiﬁc Books USA, Champaign, pp. 29-45.

Stevenson, R J ., Peterson, C. G., Kirschtel, D. B., King, C. C. & Tuchman, N. C.
1991. Succession and ecological strategies of benthic diatoms
(Bacillariophyceae): density-dependent growth and effects of nutrients and
shading. J. Phycol. 27: 59-69.

Taylor, P. A. & Williams, J. L. 1975. Theoretical studies on the co-existence of

competitive species under continuous-ﬂow conditions. Can. J. Microbiol.
2 1 : 90-98.

Tilman, D. 1977 . Resource competition between planktonic algae: an
experimental and theoretical approach. Ecol. 58: 338-348.

Tilman, D. 1982. Resource competition and community structure. 296 pp.
Princeton University Press, Princeton.

Tilman, D. & Pacala, S. 1993. The maintenance of species richness in plant
communities. In: Ricklefs, R E. and Schluter, D. [Eds] Species Diversity

96

in Ecological Communities. Historical and Geographical Perspectives,
pp. 13-21. University of Chicago Press, Chicago.

Triska, F. J., Kennedy, V. C., Avanzino, R. J. & Reilly, B. N. 1983. Effects of
simulated canopy cover on regulation of nutrient uptake and primary
production by natural periphyton assemblages. In: Fountaine 111, T. D. and
S. M. Bartell [Eds] Dynamic of Lotic Ecosystems. Ann Arbor Science,
Ann Arbor. pp. 129-159.

van Dam, H., Mertens, A. & Sinkeldam, J. 1994. A coded checklist and ecological
indicator values of freshwater diatoms from the Netherlands. Neth. J.
Aquat. Ecol. 28(1): 117-133.

von Ende, C. N. 2001. Repeated-measures analysis. In Scheiner, S. M. &
Gurevitch, J. [Eds]. Design and Analysis of Ecological Experiments.
Oxford University Press, pp. 134-157.

Weigelt, A. & Jolliffe, P. 2003. Indices of plant competition. J. Ecol. 91: 707-
720.

Wellnitz, T. & Rinne, B. 1999. Photosynthetic response of stream periphyton to
ﬂuctuating light regimes. J. Phycol. 35: 667-672.

Wilkinson, L. 1989. Systat: The System for Statistics. Evanston, 822 pp.

97

630% 08:00 com 000: 880004 ”0N 030B

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

$880.0 $880.0 I01 000800»: gmmmﬁom
«00800.0 «0000000 00:10 .052 Game . 000830 30082
mommmﬁv mommmﬁv m002 000800 88000202
0088.0 0088.0 -1 Game .0602 ONE. . 000200 000000002
000.0 000.01 ONE. ...OmmE ONE. . 000.000 88008w0§
0088.0 0088.0 ONE ...omcm ONE. . 3003 8080.0
0000080 0000080 7002320 1 l €00 02.0 «team
$088.0 53500.0 003m .595 003m . 080.200 otqau
mwwod Z Snood 2
3068.0 00-0800 00mm .. 267000 00% . 0.08.8 8300
50030.0 30060.0 0003,0500 08080800 000820 880000
90800.0 05800.0 mommm 060 otom
._ az c_ lwcl :0an 401 o 2 Iall Eagle 2252 l .58..ch
Eo. 8200:. E 80.0 E2008 E

 

 

 

 

 

 

 

 

 

 

98

 

0.0000 AN 0300.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a

 

 

 

 

 

 

 

$00.0 500.0 00.00.6562 M810 008:0 eauoﬂ
$08.0 $08.0 ONE ...Omﬁ ONE . 30:3 88
8.082 8.082 owmm .m> nﬂmm . 000.03 §€08>
3-0800 00-0800 1 0.60m. 00.2030 20080
8-00000 8-0800 m00%02 11 008000 83000
08 z 0.8 2 1
5.0 03.0 00202 0000.8 8203
000.0 «00.0 1 502 000020 53000
0.00 n. 08.0 n.
80.0 6%; 0030 320800 0303.00
F. 8 1 0.0 n. 1
88.0 $000 8.0 1 dafx
000000008 000000000 82000qu
«0000000 «0000000 .11 E 0202 Eimmﬁom
jg do: 0 c_ 1 2252 05000500

 

 

Table 2.2: Culture combination of the initial communities (n=15). Species cell

biovolume (um3) (mean :t SE) and cell biosurface mm2 (mean :t SE).

 

 

 

 

 

 

 

 

 

 

 

 

Biosudace
Density cells'1ml" Biovolume (urns) (pmz)
Case Taxon
A Ac. minutissimum
mean 100213 68.715 11116
AA Ac. minutissimum
mean 2000i5 68.715 11116
AC Ac. minutissimum
mean 100203 68.715 11116
C. placentula var. Iineata
mean 2003:] 408:k66 466:52
ACM Ac. minutissimum
mean 1002i3 68.715 11116
Cymbella cistula
mean 180 d: 5 505 :t 45 1459 1149

 

 

100

Table 2.3: Repeated measures ANOVA for differences in growth rate of
Achnanthidium minutissimum when grown in different case combinations: case A
- A. minutissimum only; AA- A. minutissimum in double initial density; with
Cocconeis placentula var. lineata (AC); and in combination with C. cistula
(ACM). Between treatment eﬂ‘ect — treatment combination and time and
treatment interaction tested as within channel effect. For each effect: F statistics,
df-degrees of ﬁ'eedom, and p- value adjusted for circularity.

 

 

 

 

 

 

 

 

 

 

 

 

 

Case Taxon Sourceofvarlatlon :11 Mean Square F ratio Pp “J
A Ac. minutissimum Treatment 3 0.029 4.86 0.0328
Error 8 0.006
Time 5 1.328 47.91 <0.0001
Treatrnent'Time 15 0.02 1.00 ns
Error 40 0.03
AA Ac. minutissimum Treatment 3 0.029 10.06 0.0043
Error 8 0.003
Time 5 1.19 35.35 <0.0001
Treatment‘TIme 15 0.03 0.75 ns
Error 40 0.034
AC Ac. minutissimum Treatment 3 0.007 1.21 ns
Error 8 0.006
Time 5 0.95 43 <0.0001
Treatment'l’ime 15 0.019 0.87 ns
Error 40 0.022
C. placentula var. lineata Treatment 3 o . 003 o . 75 n5
Error 8 0.004
Time 5 2.13 1.58 ns
Treatment‘Time 15 0.05 2.3 ns
Error 40 0.025
ACM Ac. minutissimum Treatment 3 0.033 21.2 0.0004
Error 8 0.001
Time 5 1.01 49.06 <0.0001
Treatment‘Time 15 0.046 2.23 ns
Error 40 0.02
Cymbella cistula Treatment 3 0.004 1.09 ns
Error 8 0.003
Time 5 0.065 4.16 0.0273
Treatment'Time 15 0.029 1.92 ns
Error 40 0.015

 

Pp - probability that F ratio would be equal to or greater than what is due to
chance alone.

10]

Table 2.4: Competitive effects of diatoms on Achncmthidium minutissimum
(target). Intraspeciﬁc competition (column 0,. ). Intraspeciﬁc competition input
from the neighbors are given under other (competitive indices estimations of the
neighbor species in mixed cultures, doc ac"). Values are the intercept (maximum
growth rate, GRmx) and regression coefﬁcient ‘a’s from the linear regression
analysis, Grth rate as a function of B- total neighbor biosurface. Growth rate of
Achnwrthidium minutissimum when grown in different case combinations: case A
- A. minutissimum only, without competition; AA - A. minutissimum in double
initial density; AC -with Cocconeis placenlula var. Iineata); and ACM with
Cymbella cistula. Indices estimates given with standard error given (SE), slopes
different from 0 are given in bold. Estimates of intercept and slope compared (Pp
-— probability that F ratio would be equal to or greater than what is due to chance
alone); ns- non signiﬁcantly different, different at p<0.05). Designated same letter
indicated non signiﬁcant means.

102

Table 2.4:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Wholemebiomase Hg) biomass
GR... (SE) GA (SE) GR"... (SE) aA (SE)
Case/Treatment
A target other target other
DC 0.48 (10.35) -1.95(1o.70)' 0.47 (10.06)‘ 22.40 (10.51)‘
DNP 0.39 (10.17) -0.32(10.18)" 0.17(10.03)" -0.06(10.06)"
LC 0.20 (10.13) 0.00 (10.13)” 0.13(10.05)" -0.01(10.09)°
LNP 0.31(10.08) 0.14 (10.06)" 0.18(10.05)b -0.07 (1005)”
ns p<0.001 p<0.001 p<0.001
AA
DC 0.30 (10.09) -0.36 (10.59) 0.13(10.05) -0.07(10.30)
DNP 0.29 (10.08) -0.16 (10.27) 0.17 (10.06) -0.08 (10.18)
LC 0.19(10.08) .003 (10.14) 0.23 (10.06) 0.25(1012)
LNP 0.33 (1cm -o.24(10.09) 0.11(10.05) 010410.05)
ns ns _ ns ns
AC aoc(SE) aoc(§E)
DC 0.39 (10.08) -1.20(1o.45) 0.1 (10.07) 0.27 (10.05)‘ .0.ss(1o.27)' -0.20 (10.85)
DNP 0.48 (10.18) 091(1056) 020 (10.10) 0.12 (10.10)“ -0.05(10.33)" 155(1088)
LC 0.41 (10.12) -0.57 (10.31) 0.06 (10.07) 0.36(10.10)"° -o.51(1o.25)' 214(1016)
LNP 0.42 (10.09) -0.38 (10.13) 0.00 (10.06) 0.32 (10.08)“ 025410.11)" -0.12(10.64)
ne no ns- p=0.04 p=0.01 ns
ACM aCM(SE) 004(35)
oc 0.31 (10.12) -0.27 (10.44) 0.24(10.os) 0.25 (10.04) «0.50(1o.1o)' 0.20 (10.18)
DNP 0.28(10.12) -0.07 (10.29) -O.18 (10.09) 0.17 (10.11) 0.11 (10.25)b 0.05 (10.32)
LC 0.39 (10.08) 41.28 (10.10) 0.13 (10.07) 0.28 (10.08) -o.17(10.09)° 0.48 (10.39)
LNP 0.44 (1 0.08L -o.23 ($9.08) 0.13(10.09) 0.32 (10.07) -o.1c(10.05)" 0.05 (10.45)
no ns ns ns p<0.001 ns

 

103

Flow

 

 

. :\\
\ <\\§(\\\ (\
‘\ ‘.\_\,\§‘\1\\\‘§§‘\1 §

3 \\\\-.- w? ‘2‘ ‘v \ - \‘y \
\\\\\\\\ ;\ \ \\\ \ \\§ \ \
2 \\=.\\

\ ‘\\\\§\\‘
W‘\

i\

/

\
.\ ‘\\ \
5‘ » ‘ 88‘
.\\\\ \\:( {\‘\\\ ‘1

\ \

//////

‘\\
k

\\
\" ' \ \\

/

/
///

 

5 cm.s—1)

Figure 2.1: Schematic representation of the closed experimental circular channels; a. whole view;

b. side view, circular channel diameter = 3 cm; c. view from top, where tile arrangement can be

seen. Arrow indicates ﬂow direction (ﬂow

 

N
U}

 

 

 

 

 

 

<20
3315 i i
z i i
1001 1
5°." .
0 1 t t t t r
b.
50 r ' '
<1 40'
2301: ‘
CL.
20'
1 1 1 1
10' 1
ob ' 9 !_.v

0 3 6 9121518
Day

 

Treatment
0 C

A

Figure 2. 2: Nutrient concentrations measured (pg/L); (a. ) N and (b. ) P
values in the control and NP treatments (n=4). Error bars around the
means represent standard error

105

 

p—1
O
—1

 

 

 

 

bUIChﬂOOC

 

 

Ln total density (cells.mm'2)

 

 

 

 

LIGHT

‘ -—D

40 3 6 9121518 L
Day

Figure 2.3: Total community density (Ln cells mm2) in shaded
(D) and full light (L) regimes. a. in low nutrients (C); b. in high
nutrients (NP). Error bars around the means represent standard
error

106

Bogota buqmomawmm 8: 2m 98:2 2:8 05 FEB Coho 23:86 h” 2825 mam dosaqﬁaoo
Eugene“ some E 3.88 £on 03.35 :8 .58 co: 08.3.85 385888 :38. Him onE

3253;

A75 04 AZQ 0Q

1111.

 

1
l
v—t

l
N
z.urur (901) oonpnsorg

 

 

M

 

 

107

Figure 2.5: Growth rate of A chnanthidium minutissimum in culture combination.
a. DC; b. DNP; c. LC; (1. LNP; open bars- initial growth (difference in days of
community development 0 to 3); shaded bars high biomass growth (difference in
days of community development 6 to 9; RM ANOVA polynomial test result,
signiﬁcant time period change); Bars (means :1: standard error) with the same *
(for the initial growth) or same letter (for the high biomass growth) not
signiﬁcamly different

108

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

20< 0< << <
, (m m .
e t
L .1
Q2.—
\
m m m
I n I.
.... ... +
n_Zn_

 

 

 

no-
3
no
3
3
2-
3
no
3
2

20<

0<

 

 

0..

 

ma 2:9".

md-

 

 

 

 

 

.. o.o
- m6
- o4
m4

 

 

 

 

 

 

00

 

m6-

 

 

 

 

..o.o
Jmeo
-oA

 

m4

 

(Map 81190 111) 918.1 qmorﬁ zunwgssynugul 'V

109

Figure 2.6: Growth rate relationships of Achnanthidium minutissimum per
biomass estimate in culture combination (Case); A- dashed line, AA-dashed &

dotted line, AC - dotted line, and ACM-solid line in each treatment combination;
a. DC; b. DNP; c. LC; (1. LNP

110

2

NEE Roam oomtsm :8 $202.10: 835QO

mm 6.59“.

 

 

 

 

 

 

 

 

 

 

mzo

 

 

one- m

.o .

.22 .

.2 . .2
1-4 0.. .0 m4
no- no-

.22 . .o

.o . no

.2 . .2
-.2 (2.3.0.1 m2

 

 

 

(1.1(8p sues 111) 818.1 (1148018 umwgssynugw 'V

111

Figure 2.7 : Intraspeciﬁc competition indices (or) estimates in high biomass of the
target species Achnanthidium minutissimum in each culture and treatment
combination. Bars (model slope estimates, means :t standard error)

112

AZAQ
Ode
BAD-
DOD

22225822.

2822222222228 282220

20< 0< << <

 

 

 

 

_

q

 

 

 

 

 

Nm 8:9“.

MI
NI
.2.
2- 2
W...
o m.

113

 

O N
1 r

(ICC estimate

 

 

 

 

.5"
A1515

 

 

 

a) 1 _
4g 2
E 0 — 5&1 4 Treatments
2 D DC
3’ -2 ' ' ' DNP
LC
-4 ” LNP

f" V f9 PIC)”
Culture combination
Figure 2.8: Interspeciﬁc competition indices (0.) estimates in high
biomass of neighbor species a. Cocconeis placentula var. lineata; b.

Cymbella cistula in each culture and treatment combination. Bars
(model slope estimates, means :1: standard error)

114

CHAPTER 3
COMMUNITY ORGANIZATION AND DISTRIBUTIONAL PATTERNS
OF BENTHIC DIATOMS FROM STREAMS IN KENTUCKY
INTRODUCTION

The relationship between species abundance (relative, absolute or percent
cover) and distribution has been studied for many years (Adrewartha and Birch
1954, Brown 1984, Krebs 1978, Hengeveld and Haeck 1982, Gotelli and
Simberloff 1987, Gaston and Blackburn 2000) on the premise that dynamics of
resource exploitation by species in local communities are connected with their
regional distribution. Almost without exception (Hanski et al. 1993), species with
wide distribution, i.e. present at a large proportion of sites in a deﬁned region,
tend to be more abundant than species with restricted distribution (McNaughton
and Wolf 1970, Hanski 19828). In this context, abundance is the average
abundance at sites, where the species population is observed. Either positive
(Hanski 1982a) or no (Levins 1969) correlation between abundance and
distribution of species has been observed in many groups of organisms (Hanski er
a1. 1993 ). A positive correlation between species abundance and distribution
contradicts the presumed relation expected if ecological specialization was
important. Specialization is the ability of a species to exploit and maintain
populations in unique environmental conditions (McNaughton and Wolf 1970).
Specialists will have higher mean abundance in unique sites, which will result in

no or negative correlations between species abundance and distribution in a

region.

115

Study of the abundance-distribution relationship in periphyton may be
valuable because species abundance changes greatly along environmental
gradients and relative abundance is oﬁen reported in large-scale surveys (Pan et
al. 1996). Most survey studies report algal species (distributional) patterns, but
distribution here refers to correspondence between abundance and environmental
characteristics. In those studies, usually dominant species are weighted more than
species with lower abundance in an ecoregion (Patrick 1961, Lewis 1977,
Peterson et al. 1993). Algal community structure is a result of complex
interactions with the habitat condition (Stevenson 1997) and addressing
relationships between species abundance and distribution (as ﬁ'equency of
occurrence) could be valuable. Periphyton abundance and distribution are
inﬂuenced by disturbance (Peterson et al. 1990, Humphrey and Stevenson 1992),
hydrology (Biggs and Stokseth 1996), substrate conditions (Peterson and
Stevenson 1989), grazing (Hill and Knight 1988, Steinman 1996), and nutrient
concentrations and ratios (Peterson and Grim 1992).

Diatoms are an excellent group for testing the abundance-distribution
relationships, because they are mostly microscopic, widely distributed and every
species has the potential of establishing populations under a given set of
environmental conditions. Diatom species composition is sensitive to
environmental changes (van Dam 1982, Winter and Duthie 2000) and diatom
community attributes, like species abundance, are used in ecological assessment
(Pan et al. 1996, Stevenson and Smol 2003). Consideration of both ecological

preferences and historical occurrences could provide the accurate understanding

116

of diatom abundance and distribution (Stoermer and Julius 2003). The abundance
and distribution relationship in periphyton diatoms within an ecoregion have not
been addressed.

The shape of abundance—distribution relationship for diatoms as other
organisms might be regulated by one of the following paradigms of community
organization: ﬁrndamental niche differentiation (FND, Whittaker 1975), shifting
competitive hierarchy (Keddy 1989), or balanced continuum (Austin 1985). The
FND predicts niche diﬂ‘erentiation along a gradient according to limits in species
physiological optima (for temperature, pH, alkalinity, and conductivity for
example). In the shifting competitive hierarchy (SCH), all taxa have
physiological optima at resource availability, but different competitive abilities at
resource limitation. Tradeoffs are expected, and poor competitors are displaced to
undesirable habitats. The continuum concept (CC) incorporates both theories and
predicts gradual changes in variables and gradual changes in communities
depending on the nature of the gradients.

According to the FND hypothesis, diatoms that are most abundant at
limiting nutrient concentrations will not perform at relatively high rates when
nutrients are abundant. Distinguishing between species performance as a
limitation due to poor physiological functioning at a given resource level or
performance limitation due to resource depletion by neighboring populations in
the communities is needed in SCH and CC (Bigelow and Canham 2002). It is
unclear which one or combination of the three paradigms is most applicable to

diatoms. Ifthe gradient in question provides the full spectrum from limiting

117

conditions to saturation, diﬂ‘erent species are expected to perform best at the
different concentrations (Tilman 1982). Abundance-distribution patterns along
those gradients will be easy to distinguish. Species separation according to
nutrient preference in streams is expected (Biggs 1996, Stevenson 1995).

The above mentioned patterns and paradigms provide hypotheses that can
be tested with several models. The abundance-distribution patterns have been
addressed with static niche-based models (Brown 1984, Collins and Glenn 1991,
Hanski and Gyllenberg 1993, Sheiner and Rey-Benayas 1997), with mechanistic
models (Collins and Glenn 1997) and with metapopulation models (Levins 1969,
Hanski 19828., Gotelli 1992, Tokeshi 1992, van Rensburg et al. 2000). The static
niche-based model tests whether the positive correlation between abundance and
distribution is a sampling artifact. It is assumed that rare species are more
difficult to detect than abundant species (McArdle 1990). The abundance-
distribution patterns have been addressed with mechanistic models like the model
for ecological specialization (Brown 1984). This model is based on the testable
hypothesis that ecological generalists are both widely distributed and locally
abundant, while specialized species are restricted in distribution and with low or
high average abundance. The metapopulation model relates dispersal abilities of
species, for a given average abundance, to distribution. For the given average
abundance, it could be tested whether species with high ability to disperse have
wider distribution (Hanski et al. 1993).

Species in streams with minimal human impact can be deﬁned as

reference species (Stevenson et al. in press). Diatoms from those streams are

118

likely the reference diatoms of the area or region. Those diatoms presumably
existed in streams prior to anthropogenic alterations. Reference species are
expected to have wide distribution under reference conditions, while specialists
(potentially introduced to a region) will have narrow distribution, according to the
FND paradigm. In recent years, reference diatom communities have been
recognized as important in regional river and stream assessment (the European
Union Document 2000). Diatoms are a very diverse group of algae with
estimated species of tens to hundreds of thousands (Gordon and Drum 1994,
Mann 1999). It is unrealistic to expect a few diatom species to be declared ‘most
valuable’ for protection, rather in current bioassessment, the desire to protect and
restore reference communities, regardless of their diversity, is considered an
important ecological goal (Karr and Dudley 1981, Stevenson, personal
communication).

The goals of this study were to describe and understand the relationship
between relative abundance and distribution of diatom species in streams.
Diatoms with high abundance were expected to have a narrow distribution at a
small number of sites due to specialization. As nutrient loading, conductivity and
chloride increase with human disturbance, a loss of reference taxa was expected,
according to the niche differentiation hypothesis. Sampling error, specialization
and dispersal ability were tested as the possible explanation of the abundance-
distribution pattern in diatoms. Diatom communities in reference streams were
predicted to be similar, while diatom communities between reference and non-

reference streams were expected to differ. Ultimately, I hypothesized that relative

119

abundance and distribution differences could be explained by responses of
species-speciﬁc growth rate to nutrient enrichment. Therefore, results from the
survey, were compared with experimental growth rate results in the same
ecoregion.
MATERIAL AND METHODS

Samples from 69 samples from the Knobs ecoregion (northwestern
Kentucky and southeastern Indiana) were collected during the April — June
periods in 1996 and 1997 (Appendix 3.1). Streams of this region have rapid
runoff due to steep watershed topography, which is related to the underlying
limestone bedrock formation (McGrain 1983). Streams sampled ranged from 1st
to 4'“ order (Stralrler 1957) with most streams being 2'” and 3”d order (Stevenson
et al. in press). Streams in this region have minimal invertebrate grazing
compared to other regions (Riseng et al. 2004). Four streams were designated as
reference sites and were sampled during both years. Those sites were completely
enclosed in nature preserves or human disturbance in the watersheds was low.

Each stream was visited weekly for 8 weeks. At each visit, pH and
conductivity (CON) were measured with a YSI meter (Y SI Incorporated, Yellow
Springs, Ohio, USA); water temperature was determined with a thermometer; and
canopy cover (CAN) was assessed with a spherical canopy densiometer. Samples
for chloride and nutrients were collected in 2 125-mL acid-rinsed polyethylene
bottles. Water in one sample was ﬁltered in the ﬁeld through 0.45 urn pore-size
ﬁlters to measure dissolved nutrients. Nutrient samples were stored on ice until

returning to the lab where they were frozen until analysis.

120

Benthic algae were sampled during 1 of the 8 weeks of sampling, which
was timed so that algae had greater than 7 days to recover after the last storm
event. Benthic algae were sampled from 15 rocks in riffles (3 rocks at 5 random
locations) with a spoon and toothbrush into a container. The composite sample
was split for different analysis. In this project, subsamples for chlorophyll a and
enumeration were analysed. Estimates of rock surface areas from which benthic
algae were scraped were made in the ﬁeld by measuring the upper surface of
rocks (an exact outline of the cleared from algae surface was estimated with a
ruler in the ﬁeld). Subsamples for chlorophyll a were stored on ice until returning
to the lab, where they were frozen until processing. Algal enumeration samples
were preserved with M3 in the ﬁeld (APHA 1998).

Laboratory assays

Water samples were analyzed for chloride, nitrate (N03-N), and
ammonium (NH4-N) using a Skalarg auto-analyzer. Soluble reactive phosphorus
(SRP) was measured using a HitachiG U-2001 spectrophotometer. Alkalinity
(ALK) was assessed according to standard methods (APHA 1998). Silica was
determined with ascorbic acid method on a Skalar° auto-analyzer (APHA 1998).
To determine total phosphorus (TP) and total nitrogen (TN) concentrations,
particulate matters in water samples were oxidized with persulfate and analyzed
for SRP and NO3-N. Chlorophyll a was measured spectrophotometrically aﬁer
extraction from the periphyton samples with 90 % buffered acetone (APHA

1998).

121

Algal subsamples were analyzed for diatom species composition.
Permanent diatom slides were prepared by acid-cleaning to increase the clarity of
observing diatoms (Patrick and Reimer 1966) and mounting in Naphrax resin (RI
1.74, Northern Biological Supplies L., Ipswich, UK). A minimum of 600 diatom
valves were identiﬁed and counted on each slide. The main diatom ﬂoras used
for identiﬁcation were Patrick and Reimer (1966, 1975), Krammer and Lange-
Bertalot (1986, 1988, 1991 a, b). Additional taxonomic sources like Hustedt
(1930 a, b, 1959 — 1966), Lange-Bertalot and Metzeltin (1996) were available at
the R J. Stevenson algal laboratory library (at Michigan State University) and
were consulted when needed. Correct taxonomic identiﬁcation in the dataset was
ensured following a quality assurance/quality control (QAQC) protocol on the
original counts. A second analyst independently counted samples from the
reference streams and four random streams in each year. Relative abundances of
diatom taxa in each sample were determined and mean relative abundance for the
sites a species occurred was calculated.

The following community characteristics were calculated for each sample:
Richness (total number of taxa observed in a site); Shannon-Weaver diversity
index H= - Zia-opt log(pi), where pi is the proportion of the total count arising ﬁ'om
the ith species (Shannon and Weaver 1949); evenness
(equitability)=H’/ln(richness) (Pielou 1969) and Simpson’s dominance index
(Simpson 1949), which measures the likelihood of two randomly chosen
individuals in a sample to be the same species. If Simpson’s dominance index is

0, there is complete evenness and low dominance in a site. For simplicity, diatom

122

species that appeared in more than 10 % of the samples were deﬁned as common.
In this study, diatom species that occurred in less than 7 samples were deﬁned as
rare. Taxa that appeared as auxospores or a single broken valve without central
area were not included in the analyses.

In the reference sites, human disturbance is considered negligible
(Stevenson et al. in press), so diatoms ﬁom those sites were deﬁned as reference
taxa. Those species are the species that occur naturally in Kentucky (or not
introduced) and in many water bodies, fossil records and habitats. Extensive
counts were performed to account for as many rare taxa as possible in the four
reference streams. For all reference sites and 8 random non-reference sites, after
the original 600 count and the QAQC count, a third count continued until at least
ﬁve taxa were enumerated with 100 valves (which in diverse sites reached a count
of >2500 valves). This protocol ensured accounting for all dominant taxa (more
than 5 % relative abundance in each count), allowed observation of more rare taxa
in the count and was independent of a predetermined number count.

Experiment

Species-speciﬁc grth rates were estimated in different nutrient
conditions with an experiment in artiﬁcial streams and then compared to survey
results. I tested the hypothesis that reference species were growing at higher rates
in low nutrient conditions than in high nutrient sites.

Artiﬁcial streams at the University of Louisville Stream Facility in
Bemheim Forest, Kentucky, were used to manipulate nutrient concentrations

(Manoylov, Chapter 1, Figure 1.1). Those streams were built alongside a pristine

123

3rd order stream, Hart’s Run, which was completely enclosed within Bemheim
Nature Preserve. Three artiﬁcial channels had ambient N and P from Hart’s Run
(had 17.5-38 pg Nos-NIL, 6.8-51 pg NH4-N/L, 1.5-4.4 pg SRP-P/L, and 10.5-
11.9 mg SiOz-Si/L). Three channels had 1024 pg N/L with 50 pg P/L and three
channels had 500 pg N/L with 128 pg P/L. Stock solutions with different
concentrations of N8NO3 and KHzPO4 were delivered from 1 L containers to each
channel using peristaltic pumps (ManostatTM model STD). The pumps were set at
a different drip rate to achieve the desired nutrient concentration. Nutrient supply
was kept at a constant rate throughout the experiment. Light in the experiment
was reduced by covering the channels with greenhouse shade cloth that lowered
the full sunlight to 10 % incident light, which was comparable to light levels in
many habitats of Hart’s Run. The experiment was carried out in May 1999, with
stream water temperature around 20°C.

Periphyton communities were created by algal colonization of 29.2 cm’2
unglazed ceramic tiles, which provided uniform habitat surface simulating rocks
in natural streams. Tiles were placed at the lower end of each channel where
water ﬂow was most stable. To ensure algal diversity, the initial algal inoculum
was gathered from three streams in the Mid-Atlantic Ecoregion. The streams
were collected in the survey and had different trophic status: urban Beargrass
Creek (with nuisance growths of Cladophora), agriculturally impacted Wilson
Creek, and the pristine Hart’s Run (reference site). Equal amounts of algal
suspension from the three streams were mixed, and 250 ml aliquots were placed

in each re—circulating experimental stream channel.

124

A sample of two random tiles (non-adjacent and not from the upstream
end, where disturbance from ﬂow was possible) were collected for 16 days of
community development in the controlled conditions. The sample was
immediately frozen after collection. At the laboratory, algae were scraped from
the tiles with a scalpel and toothbrushes. A lO-ml subsample was preserved with
M3 (APHA 1998) for algae counts. Semi-permanent syrup mounts were prepared
with Taﬁs Mounting Medium (Stevenson 1984), which allows good resolution of
cell wall structure of diatoms without cleaning. Total algal community counts
were performed. At least 300 live algal cells (natural units as colonies, ﬁlaments
etc. in soft algae and diatom frustules) were identiﬁed to the lowest taxonomic
level and counted. Diatoms dominated the community. Only frustules with a
visible, healthy-looking chloroplast were counted. Counts were performed at
IOOOX on a LEICA DMLB light microscope. Species relative abundances were
averaged per treatment.

Careful monitoring and consecutive sampling every 3 days allowed a good
estimate of peak biomass. Absolute cell densities (D, cells mmz) were estimated.
Growth rates were calculated based on species density in response to each nutrient
treatment. Grth rates (R) were estimated as daily per capita changes in cell
density in each circular channel using the following equation: R = (ln D (10) - In D
(3))/( T10 —T;), where D is total density, T3 and T10 are days of community
development prior to peak biomass, this corresponded to the time of sampling in
the survey. Growth rates (ln cells cell'1d") were calculated for each species in

control and the NP channels.

125

Data analysis

The relationship between species abundance and distribution was tested
with regression analysis. Following the terminology of the core-satellite
hypothesis (Hanski 19828), a core species was a common diatom that appeared in
more than 90 % of the sites, while a satellite species appeared in less than 10 % of
the sites (rare). Species in the middle (found in more than 10 % and less than 90
% of the sites) were called intermediate. Based on the number of species in each
category, different distribution patterns are possible (skewed to the left or right,
unimodal, or bimodal). All categories can vary in abundance.

Several analyses: sampling error, ecological specialization and dispersal
abilities were performed to explain the observed patterns of abundance and
distribution. In sampling microorganisms, there is a need to test for possible
aggregation of species at a subset of sampling sites. The role of sampling error
due to diatom aggregation was tested with regression analyses. Parameters of the
sampling model were expected to differ from zero (Hanski et a1. 1993). The
probability (p0) for a taxon to occupy all available sites (69 sites) was calculated
with the following equation:

Ln(-ln(po)) z ln(ln x.n)-2 * ln (CV),

where x.n is the mean species relative abundance for all available sites;
CV is species speciﬁc coefﬁcient of variation (CV), when measuring mean
species relative abundance for all available sites with the observed mean relative
abundance (i.e. mean relative abundance for the actual sites observed). With the

decrease of a species average abundance (Km) for all available sites and increase

126

of spatial variation (CV for all taxa), the fraction of empty sites in a region will
increase (hence the regional distribution (p) will decrease).

For a given abundance, more dispersive species should have wider
distribution than less dispersive species (Hanski et al. 1993). Phylogenetic
relationships between diatom species were not considered in this study.
Taxonomic identiﬁcation of diatoms is usually based on morphological characters
(size, shape, attachment etc.) and smaller size has been assigned to good dispersal
abilities (Biggs et al. 1998). A metapopulation model was tested with multiple
regression analysis with two dependent variables, average abundance (for the sites
the taxon was found, loge-transformed) and size, which I assumed was related to
dispersal ability. Size was assigned to each diatom species in the following rank
order ﬁ'om highest dispersal ability: small length (5-15 pm), medium (15-40 pm)
and large (>40 pm, for some Nitzschia and Synedra species even >100 pm). The
independent variable was number of sites (arcsin-sqrt-transformed, method
follows Hanski et al. 1993). Using ANOVA, the dispersal effect tests whether
residuals from the abundance-distribution relationship can be explained with
dispersal ability (size).

The physicochemical differences between reference and non-reference
sites were tested with a Mann-Whitney nonparametric test. Spearman rank
correlations were used to identify relationships between physico-chemical
variables in reference and non-reference sites. Only signiﬁcant relations (p<0.05)

were reported.

127

Average species similarity (aveS) within reference and non-reference sites
or dissimilarity (aveS) between the two groups of sites was evaluated with
SIMPER procedure (Primer-E Ltd., Clarke and Gorley 2001). This allowed
examining the contribution of each diatom species to either similarity within or
dissimilarity between reference and non-reference sites. Standard deviation of the
means were estimated for S and 6, respectively SD(S) and SD(S). For both
measures, how consistently a diatom species contributes to either aveS or aveS
was measured with the ratios aveS/SD(S) and aveS/SD(S) respectively. When
SD( S) and SD(S) were small, the ratios were large and the respective diatom
species not only contributed much for the similarity or dissimilarity, but was also
consistent in doing so. Diatom species with high discriminating ratio aveS/SD(S)
were called discriminating species (Clarke and Warwick 2001) and are reported.

The effect of ecological specialization (for all taxa observed) in the
abundance—distribution relationship was tested with ANOVA, where ecological
specialization was deﬁned in three categories. All common diatoms were
included in the analyses. Common reference taxa were hypothesized to be
generalists for the region. If there were no human alterations of the habitat, the
reference taxa are expected to be widely distributed and with high mean relative
abundance. Taxa ﬁom non-reference sites (where human inﬂuence is expected)
are expected to be more specialized. Being interested in the identity of taxa
occurring in reference sites not by chance, I took the following approach: if a
taxon was found in more than 3 reference sites and some non-reference sites,

regardless of its abundance, it was not included in the specialists’ number from

128

non-reference sites. Non-reference taxa were expected to have a narrower
distribution with high abundance at a few sites. The third category was termed
random taxa (appeared in less than 0.5 % of the sites, either reference or non-
reference). Those taxa were considered in the sites by chance and no eco-
specialization was assigned. Species distribution (proportion of sites in which a
species occurred) was arcsin-sqrt-transformed and mean abundance was log,(x +
1)-transformed.

Presence-absence of all taxa observed in the experiment was compared
with the survey. Ifa diatom species was present in a reference site, but not
necessarily in a non-reference site, grth rate was evaluated in the experiment.
In the experiment, grth rates were assessed in control (ambient low nutrients)
and enriched channels. The potential of higher grth rates of taxa with increase
in nitrogen and phosphorus concentration was tested in the experiment. The
Spearman correlation coefﬁcient for the relationship between species growth rates
in low and high nutrients was determined. If reference taxa grew best at low
nutrient concentrations, grth rates of reference taxa were expected to decrease
in high nutrients. Statistical analyses were performed with SYSTAT® version 10
(Wilkinson 1990) unless stated otherwise.

RESULTS
Species abundance-distribution relationship, sampling error, aggregation, and
dispersal
A positive relationship was observed between diatom species abundance

and distribution in the survey, i.e. more abundant species occupied a larger

129

percent of the sites than less abundant species (regression analysis: r2 = 0.68,
p<0.01; Figure 3.1 a). A total of 184 diatom species were recorded. Three taxa
were observed in more than 90 % of the sites sampled. Achnanthidium
minutissimum was observed at all sites (Table 3.1). This diatom had the highest
mean relative abundance for the survey (20.8 %). The other two core species in
this study, Nitzschia dissipata and Gomphonema angustatum, had 6.5 and 15.5 %
mean relative abundances, respectively. Sixty-nine percent of the observed taxa
were deﬁned as rare because they appeared in less than 10 % of all sites. The
highest observed mean relative abundance for a rare taxon was 4.2 % (Figure 3.1
b, Table 3.1).

The observed positive relationship between abundance and distribution
was not due to sampling error or aggregation. The coefﬁcient of variation was
negatively related to species distributions (Table 3.2). Diatoms with high
abundance were found in more sites and had smaller coefficient of variation. The
analysis showed that the parameters differed from zero.

In this study, diatom dispersal abilities were not dependent on diatom size
(p=0.75). High, medium and low dispersal abilities (based on species size) were
evenly distributed between sites. Abundance was size dependent (p<0.0001,
Table 3.3), the most abundant diatom was small, but small diatoms were observed
in all categories.

Relation of core-satellite species to ecological conditions

Temperature, canopy cover (measure of light) and pH were not different

between the reference and non-reference streams. Silica was higher in the

130

reference streams. All other physico-chemical variables in the reference sites
were signiﬁcantly lower compared to sites with human inﬂuence (Table 3.4).
Conductivity and chloride were 1.5 times higher in the non-reference sites
compared to the reference sites. Nutrients were several orders of magnitude
higher also.

In reference sites, TN was positively correlated with N:P ratio and
chloride, while TP was negatively correlated with silica and conductivity. N:P
ratios were positively related to chloride concentration. Silica was positively
correlated to alkalinity and conductivity (Table 3.5). Different correlations
between the measured physico—chemical variables were observed in the non-
reference streams. TN was positively correlated with TP and chloride. Alkalinity
and conductivity increased in the same direction also (Table 3.5). The lower
nutrient availability in the reference sites corresponded to lower chlorophyll a
compared to the non-reference streams (Table 3.6). With the increase in
nutrients, chlorophyll a increased 3-fold in non-reference sites.

Shannon diversity (Table 3.6) and Simpson’s dominance index were lower
in the reference sites also (mean 1.6 and mean 0.67 respectively) compared to
non-reference sites, where the mean value of Shannon diversity was 2.2 and the
mean value of Simpson’s index was 0.81. In reference streams 15 to 26 taxa were
observed (evenness 0.41 to 0.66), while in sites with human inﬂuence, 14 to 45
diatom taxa were observed (evenness 0.36 to 0.82). For the 8 available reference

sites, diatom species richness was less than at all sites (Table 3.1).

131

Average similarity among reference streams was 48 %. Four taxa
contributed to 90 % of the similarity, with over 60 % contributed by A.
minutissimum alone. Achrxmthrdiran minutissimum and Gomphonema
angustatum were the two discriminating species in reference sites, because of
their high similarity and low standard deviation measures (discriminating ratios
16.44 and 8.38 respectively, Table 3.7).

Within non-reference streams, average similarity among sites was lower,
34 % (Table 3.8). Thirteen taxa accounted for 90 % of the accounted similarity.
Achnanthidium minutiss'imum accounted for only 30.6 % of similarity among the
non-reference sites. Three discriminating species in non-reference sites had high
similarities and low standard deviations: A. minutim'imwn, G. angustatum and
Cymbella aﬁm‘s (discriminating ratios 9.15, 6.75 and 5.60 respectively, Table
3.8).

The observed differences in species richness between reference and non-
reference sites contributed to the high 68 % dissimilarity between the two groups.
Twenty-seven taxa (15 % of all identiﬁed) contributed for 90 % of the
dissimilarity, with 50 °/o of it contributed by four taxa (Table 3.9). The highest
discriminating species in the between sites dissimilarity was Cymber aﬁinis
(ratio 8.67). Additional dissimilarity was contributed by Gomphonema
olivaceum, Nitzschia inconspicua, Rhoicosphenia abbreviata and Navicula
gregaria (Table 3.9). All four discriminating taxa appeared in both reference and

non-reference taxa.

132

Cymber aﬁinis occurred in 80 % of the samples, but had mean relative
abundance of 14.9 % and 8.8 % in reference and non-reference sites, respectively
(Tables 3.1 and 3 .9). Achnanthes deﬂexa and Achmmthidium minutissimum were
the most abundant taxa in reference streams. Achnanthes deflexa appeared in 62
% of non-reference and 88 % of reference sites (mean relative abundance 5.9 %
and 14.9 % respectively). Only A. minutissimum and Cymbella sp. K appeared in
all reference and non-reference samples, but differed in mean relative abundance
(in reference sites 38 % to 2.8 % and in non-reference sites 18.6 % and 1.8 %
respectively for the 2 species).

For the whole dataset, reference taxa had greater average abundance and
wider distribution than non-reference taxa (Table 3.10). Non-reference taxa had
low average abundance regardless of the distribution pattern. For example:
Rhoicosphenia abbreviata had 2.8 % mean relative abundance and was found in
54 % of the non-reference sites; Amphora pediculus was distributed in 78 % of
the sites and also had 2.8 % mean relative abundance, Navicula lanceolata, N.
subminuscula, and Sellaphora seminulum appeared in about 17 % of the sites with
similar mean relative abundance (2.1-3.2 % mean relative abundance). The
abundance of Achnanthes deﬂexa and Achnanthidium minutissimum was reduced
in half in non-reference sites (Tables 3.1 and 3 .9).

In the experiment, grth rates were calculated for 11 taxa. In the survey,
those taxa were mostly reference taxa, only Nitzschia acicularis and Fragilaria
nanana were found in the non-reference sites only. The hypothesis of reference

taxa growing faster at low nutrients compared to high nutrient conditions was

133

rejected. The potential of higher grth rates of non-reference taxa with
increasing nitrogen and phosphorus concentration was conﬁrmed in the
experiment, where grth rates of the reference taxa decreased in general or did
not change with enrichment (taxa like: Achnanthidium minutissimum, Achncmthes
deﬂexa, Encyonema minutum, Fragilaria nanmra, F. vaucheriae, and C ymbella
cistula). In low nutrient concentrations, representatives of the genus Nitzschia (N.
dissipata, N. palea, and N. acicularis) had lower grth rates than the other taxa
from reference sites in the survey, but their growth rate increased more than
reference taxa with increasing nutrients (Figure 3.2).
DISCUSSION

In this study, a positive abundance—distribution relationship was observed
for benthic diatoms. Achnanthidium minutissimum had highest mean relative
abundance in the survey and was found at all sites. Core diatom species
comprised only 2 % of the diatom community, while the majority of diatoms
observed were satellite species. The large number of rare taxa observed did not
change the positive correlation between abundance and distribution. Diatom
communities from reference streams had lower diversity and higher similarity
than non-reference stream communities. Reference and non-reference diatom
communities were highly dissimilar. A relatively small number of diatom taxa
contributed to the observed dissimilarity. Diatoms observed in the reference sites
were observed in many non-reference sites, but in signiﬁcantly lower abundance.
All non-reference diatoms were in lower abundance regardless of the number of

sites in which they appeared.

134

With an increase in nutrients, eutrophic species grow faster when
compared with diatom species from the reference community. The robustness of
the pattern was conﬁrmed with the experimental results. In the experiment,
conductivity, temperature and chloride were constant; only nutrient
concentrations were manipulated. Reference species like Achnanthidium
minutissimum and Achnanthes deﬂexa had relatively high growth rates in low
nutrients and continued to grow with increasing nutrients, but growth of non-
reference taxa, like representatives of Nitzschia, were stimulated more than the
achnanthoid diatoms in higher nutrients. Thus, reference taxa could occur in non-
reference conditions, but non-reference taxa were not likely to occur in reference
conditions because they could not grow well in low nutrient conditions. This
ﬁnding implies nestedness of the diatom taxa from reference streams within
communities in non-reference streams, as patterns of nested subsets of species can
be related in this study to nested physiological tolerances and colonization
ability, but not to motility as found in other organisms (Greve et al. 2005).

In the abundant literature addressing the abundance-distribution
relationships, only one relates to algae. Lewis (197 7) found an asymmetric
relationship for phytoplankton, which was skewed to the right with most taxa in a
few sites. That distribution was independent of mean abundance and resource
supply. The relationship was speciﬁc for green algae, bluegreen algae, and
diatoms and varied with grazer conditions. In the present study, I found a strong
positive relationship between abundance and distribution, common taxa were

most abundant and had wider distribution.

135

Biologically, the appearance of Achnanthidium minutissimum at all sites
and with high mean relative abundance can be explained with better colonization,
the relatively fast growth of this diatom and high tolerance for low nutrient
concentrations. Tradeoff theory suggests that there should be optimal conditions,
so if A. minutissimum was at optimal grth in the reference sites, different taxa
should have high abundance at the non-reference sites. The mean abundance of
A. minutissimum in the reference sites was signiﬁcamly higher compared to all
sites. In the literature, Achnanthidium minutissimum was found in 79 % of
samples throughout the United States, which include great variability in
environmental conditions (Potapova and Charles 2003). In areas with minimum
human disturbance like Alaska, A. minutissimum was the most abundant species
in a four year study and community composition did not change signiﬁcantly with
nutrient increase (Peterson et al. 1993).

The signiﬁcant decrease in relative abundance of reference taxa with
increasing human inﬂuence was due to increases in abundance of non-reference
species like, Nitzschia dissrpata and N. inconspicua. None of the observed
diatoms had high abundance and low distribution. Several species were observed
only in non-reference sites: Amphora pediculus, Diatoma vulgaris, Naviucla
copiotenella, and Rhoicosphenia abbreviata. Those species were distributed in
up to 79 % of the sites, but their mean abundance was never above 7 % for any
site. None of the rare diatoms had more than 5 % mean abundance for any site

observed.

136

Niche differentiation was potentially inﬂuenced by nutrients, but was not
inﬂuenced by light, temperature, and pH, because those were the same in the
reference and non-reference sites. Effects of grazing on taxa abundance and
distribution were assumed to be low for all natural streams in Kentucky (Riseng et
al. 2004) and were not directly tested in this survey. Achnanthidium
minutissimwn and Achnanthes deﬂexa are understory species, resistant to grazing
(Steinman 1996). If grazing was important, differences in abundance of overstory
taxa like Gomphonema angustatum, G. olivaceum, Cymbella aﬁ‘inis and other
stalked diatoms would be expected. Those taxa were abundant in the reference
and non-reference sites in similar pattern with the experimental study where
grazers were excluded. The same relationship of high abundance of the grazer-
resistant Achnanthidium minutissimum and grazer-sensitive C. aﬂinis in all
experimental channels was observed.

The observed pattern did not follow completely the core-satellite
hypothesis (Hanski 1982a), which predicts a positive abundance-distribution
relationship and a bimodal pattern in species numbers in the core and satellite
categories. This pattern has been observed in different organisms and along
multiple gradients (Gibson et al. 1999, McGeoch and Gaston 2002), but it was not
evident for benthic diatoms. Diatoms with high abundance were found in more
sites and had a smaller coefficient of variation. Thus, the prediction of ecological
specialization was not supported.

Sample collection, sampling area, and the number of samples were

appropriate to study the abundance-distribution relationship (McGeogh and

137

Gaston 2002). However, the degree of environmental heterogeneity was
minimized by sampling only rifﬂes, rocks, and streams within an ecoregion.
Environmental heterogeneity among sampled habitats may not have been
sufﬁcient to observe ecological specialization of diatom taxa. Different species
composition would be expected if, for example, depositional sampled ﬁ'om
streams and wetlands were sampled at different temporal and spatial scales.
Diatom specialization has been observed in wetlands, springs and streams with
metal or acid contamination (Stevenson et al. 1996). In diatom studies,
specialization has been based on single species (Raschke 1993), on groups of
indicator species (Schoeman 1976) or on the whole community (Pan et al. 1996).
Depending on the scale of the ecological assessment, diatoms respond rapidly to
environmental changes and provide useful assessment of the biotic integrity of
streams.

Distribution based on dispersal ability was not different for the different
diatom size groups. Diatom species differ in size (from 2 pm to 2 mm in some
marine forms, Round et al. 1990) and all have potentially exceptional powers of
dispersal (Round 1981). Air, insects and birds are considered effective dispersal
agents. Direct research on diatom dispersal is old and was not directed to streams
(Schlichting 1960, 1964, 1969). Circumstantial evidence of the limitless dispersal
abilities of diatoms can be obtained fi'om the literature (Behre and Schwabe 1970
in Mann 1999). Diverse freshwater diatom communities were observed on an
island only 5 years after volcanic eruption in the ocean. Diatom dispersal is

poorly understood, but in this research, dispersal was not related to size. In this

138

study, size was related to mean abundance, as the most abundant and distributed
diatom was small, but several small diatoms were found in the rare category.
Smaller diatoms are known to have faster immigration (r-selection), but A.
minutissimum (a small diatom) was a K-selection species because of the
persistence pattern observed in this study.

This research did conﬁrm the prediction of the fundamental niche
differentiation hypothesis, that diatom that were most abundant in limiting
nutrient concentrations would not perform as well with increasing nutrients.
Reference species persisted in sites with human inﬂuence, but were less abundant.
Changes in species membership followed the shifting competitive hierarchy
hypothesis: non-reference species grew faster, compared to reference taxa with
increasing nutrient supplies. The relationship predicted by the continuum concept
with gradual changes in variables and gradual changes in communities would also
explain the observed relationships. The direction of negative change (predicted
by the continuum concept hypothesis) as mean relative abundance of reference
taxa gets lower with increase of human inﬂuence, does not relate to distribution
yet. Brown (1984) introduced the ecological specialization hypothesis, where one
group of species utilizes more abundant and widespread resources than a second
group of species, suggesting low specialization for group one. For the non-
reference taxa, lower observed abundance might still warrant future dominance in
a site, because the transition period between competitive exclusion of high

nutrient species in diatoms is unknown.

139

In conclusion, diatom species with high mean relative abundance had the
widest distribution. Species from non-reference sites had narrower distribution
and lower abundance. Mean relative abundance of reference species decreased
with increasing nutrients and chloride. Non-reference sites had higher diversity,
lower evenness and lower richness than reference sites. Grth rates of non-
reference taxa increased faster than reference taxa with increasing nutrient
enrichment related to anthropogenic inﬂuence. Complete taxa turnover and
competitive exclusion or physical loss of the reference taxa was not observed,
however partial turnover in community composition was observed. These
patterns were due to the species-speciﬁc abundance and distribution relationships

of a relatively small number of diatom species.

140

LITERATURE CITED

APHA 1998. Standard methods for examination of water cmd wastewater.
Washington DC, American Public Health Association.

Austin, M. P. 1985. Continuum concept, ordination methods, and niche theory.
Ann. Rev. Ecol. Syst. 16: 39-61.

Adrewartha, H. G. & Birch, L. C. 1954. The distribution and abundance of
animals. University of Chicago Press, Chicago.

Behre, K. & Schwabe, C. H. 1970. Auf surtsey/Island im Sommer 1968
nachgewiesene nicht marine Algen. (Uber die nati'rrliche Fn'ihbesiedling
postvulkanischer Substrate oberhalb des Litorals.) Schriften des
Naturwissenschaftlichen Vereins ﬁir Schleswig-Holstein (supplement):
31-100.

Bigelow, S. W. & Canham, C. D. 2002. Community organization of tree species
along soil gradients in a north-eastem USA forest. J. Ecol. 90: 188-200.

Biggs, B. J. F. 1996. Patterns in benthic algae of streams. In Stevenson, R J.,
Bothwell, M. L. & Lowe R. L. {Eds} Algal Ecology. Academic Press,

San Diego, pp. 31-56.

Biggs, B. J. F., Stevenson, R. J. & Lowe, R. L. 1998. A habitat matrix conceptual
model for stream periphyton. Archiv Hydrobiol. 143: 21-56.

Biggs, B. J. F. & Stokseth, S. 1996. Hydraulic habitat suitability for periphyton in
rivers. Regul. Rev. 12: 251-261.

Brown, J. H. 1984. On the relationship between abundance and distribution of
species. Am. Nat. 124: 253-279.

Brown, J. H, Mehlman, D. W. & Stevens, G. C. 1995. Spatial variation in
abundance. Ecol. 76: 2028-2043.

Clarke, K. & Gorley, R. N. 2001. Primer v5 . Primer Ltd, Plymouth Marine
Laboratory UK.

Clarke, K. & Warwick, R. M. 1998. A taxonomic distinctness and its statistical
properties. J. Appl. Ecol. 35: 523-531.

Collins, S. L. & Glenn, S. M. 1991. Importance of spatial and temporal dynamics
in species regional abundance and distribution. Ecol. 72(2): 654-664.

141

Collins, S. L. & Glenn, S. M. 1997. Eﬁ‘ects of organismal and distance scaling on
analysis of species distribution and abundance. Ecol. Applic. 7: 543-551.

Gaston, K. J. & Blackburn, T. M. 2000. Pattern and process in macroecology.
Blackwell Science, Oxford.

Gibson, D. J ., Ely, J. S. & Collins, S. L. 1999. The core-satellite species
hypothesis provides a theoretical basis for Grime’s classiﬁcation of
dominant, subordinate, and transient species. J. Ecol. 87: 1064-1067.

Gordon, R & Drum, R. W. 1994. The chemical bvasis of diatom morphogenesis.
Intern. Rev. Cytol. 150: 243-272.

Gotelli, N. J. 1992. Metapopulation models: The rescue effect, the propagule
rain, and the core-satellite hypothesis. Am. Nat. 138: 768-776.

Gotelli, N. J. & Simberloff, D. 1987. The distribution and abundance of tallgrass
prairie plants: a test of the core-satellite hypothesis. Am. Nat. 130: 18-35.

Greve, M., Gremmen, N. J. M., Gaston, K. J. & Chown, S. L. 2005. Nestedness
of Southern Ocean island biotas: ecological perspectives on a
biogeographical conundrum. J. Biogeogr. 32: 155-168.

Hanski, I. 19828. Dynamics of regional distribution: the core and satellite species
hypothesis. Oikos 38: 210-221.

Hanski, I. & Gyllenberg, M. 1993. Two general metapopulation models and the
core-satellite hypothesis. Am. Nat. 142: 17-41.

Hanski, I., Kouki, J. & Halkka, A 1993. Three explanations of the positive
relationship between distribution and abundance of species. In Ricklefs, R.
E. and Schluter, D. [Eds] Species diversity in ecological communities.
The University of Chicago Press. Chicago, I]. pp. 108-116.

Hengeveld, R, & Haeck, J. 1982. The distribution of abundance. 1.
Measurements. J. Biogeogr. 9: 303-316.

Hill, W. R & Knight, A W. 1988. Concurrent grazing effects of two stream
insects on periphyton. Limnol. Oceanogr. 33: 15-26.

Humphrey, K. P. & Stevenson, R. J. 1992. Responses of benthic algae to pulses

in current and nutrients during simulations of subscouring spates. J. N.
Am. Benthol. Soc. 11(1): 37- 48.

142

Hustedt, F. 1930a. Bacillariophyta (Diatomae). In Pascher, A (Ed). Die
Sr'rﬁwasser-Flora Mitteleuropas 10: 1-466. Verlag von Gustav Fischer.
Jena, Germany.

Hustedt, F. 1930b. Kryptogamen-Flora von Deutschland. Osterreich und der
Schweiz. Band 7 Kieselalgen Teil 1. Akademische Verlagsgesellschaft.
Leipzig. 920 pp.

Hustedt, F. 195 9. Kryptogamen-Flora von Deutschland Osterreich und der
Schweiz. Band 7 Kieselalgen Teil 2 Lieferung 6. Akademische
Verlagsgesellschaft. Leipzig. 109 pp.

Hustedt, F. 1961. Kryptogamen-Flora von Deutschland. Osterreich und der
Schweiz. Band 7. Kieselalgen Teil 3. Lieferung 1. Akadenrische
Verlagsgesellschaﬁ. Leipzig. 160 pp.

Hustedt, F. 1962. Kryptogamen-Flora von Deutschland Osterreich und der
Schweiz. Band 7 Kieselalgen Teil 3 Lieferung 2. Akademische
Verlagsgesellschaft. Leipzig. 188 pp.

Hustedt, F. 1964. Kryptogamen-Flora von Deutschland, Osterreich und der
Schweiz. Band 7 Kieselalgen Teil 3 Lieferung 3. Akademische
Verlagsgesellschaft. Leipzig. 208 pp.

Hustedt, F. 1966. Koptogamen—Flora von Deutschland Osterreich und der
Schweiz. Band 7. Kieselalgen Teil 3. Akademische Verlagsgesellschaft.
Verlag von J Cramer, Leipzig. 816 pp.

Karr, J. R & Dudley, D. R. 1981. Ecological perspective on water quality goals.
Envir. Manag. 5: 55-68

Keddy, P. A 1989. Competition. Chapmen & Hall, London, pp. 552.

Krammer, K. & Lange-Bertalot, H. 1986. Bacillariophyceae. Teil: Naviculaceae.
In Ettl, H., Gerloff, J ., Heynig, H., & Mollenhauer, D. [Eds].
St‘isswasserﬂora von Mitteleuropa. 2(1): 1-876. Gustav Fisher Verlag,
Germany.

Krammer, K. & Lange-Bertalot, H. 1988. Bacillariophyceae. Teil: Bacillariaceae,
Epithemiaceae, Surirellaceae. In: Ettl, H., Gerloff, H., Heynig, H. &
Mollenhauer, D. [Eds]. Sr'isswasserflora von Mitteleuropa. 2(2): 1-596.
Gustav Fisher Verlag, Germany.

Krammer, K. & Lange-Bertalot, H. 1991a Bacillariophyceae. Teil: Centrales,
Fragilariaceae, Eunotiaceae. In Ettl, H., Gerloff, J ., Heynig, H. &

143

Mollenhauer, D. [Eds]. Susswasserﬂora von Mitteleuropa. 2(3): 1-576.
Gustav Fisher Verlag, Germany.

Krammer, K. & Lange-Bertalot, H 1991b. Bacillariophyceae. Teil:
Achnanthaceae. Kritische Erganzungen zu Navicula (Lineolatae) und
Gomphonema. In Ettl, H., Gartner, G., Gerloff, J., Heynig, H. &
Mollenhauer, D. (Eds). Susswasserﬂora von Mitteleuropa. 2(4): 1-43 7.
Gustav Fisher Verlag, Germany.

Krebs, C. J. 1978. Ecology: the experimental catalysis of distribution and
abundance. Harper and Row, New York.

Lange-Bertalot, H & Metzeltin, D. 1996. Ecology-Diversity—Taxonomy.
Indicators of oligotrophy - 800 taxa representative of three ecologically
distinct lake types. Iconogr. Diatomol. 2: 1-390.

Levins, R 1969. Some demographic and genetic consequences of environmental
heterogeneity for biological control. Bull. Entomol. Soc. Am. 15: 23 7-240.

Lewis, W. M. 1977. Ecological signiﬁcance of the shapes of abundance-
frequency distributions for coexisting phytoplankton species. Ecol. 58:
850-859.

Mann, D. G. 1999. Phycological reviews. The species concept in diatoms.
Phycologia 38(6): 43 7-495.

McArdle, B. H. 1990. When are rare species not there? Oikos 57: 275-277.

McGeogh, M. A. & Gaston, K. J. 2002. Occupancy frequency distributions:
patterns, artifacts and mechanisms. Biol. Rev. 77: 311-331.

McGrain, P. 1983. The Geologic Story of Kentucky. Kentucky Geological
Survey, University of Kentucky, Lexington, Kentucky.

McNaughton, S. J. & Wolf, L. L. 1970. Dominance and the niche in ecological
systems. Science 167: 131-139.

Pan, Y., Stevenson, R J ., Hill, B. H., Herlihy, A T. & Collins, G. B. 1996. Using
diatoms as indicators of ecological conditions in lotic systems: a regional
assessment. J. N. Am. Benthol. Soc. 15: 481-495.

Patrick, R. 1961. A study of the numbers and kinds of species found in rivers of
the Eastern United States. Proc. Ac. Nat. Sci. Philadelphia 113: 215-258.

Patrick, R. & Reimer, C. W. 1966. The Diatoms of the United States. Vol. 1.
Monogr. Acad. Nat. Sci. Philadelphia 13: 1-688.

144

Patrick, R & Reimer, C. W. 1975. The Diatoms of the United States. Vol. 2,
Part 1. Monogr. Acad. Nat. Sci. Philadelphia 13(2): 1-213 pp.

Peterson, C. G. & Grimm, N. B. 1992. Temporal variation in enrichment effects
during periphyton succession in a nitrogen-lirnited desert stream
ecosystems. J. N. Am. Benthol. Soc. 11: 20-36.

Peterson, C. G. & Stevenson, R. J. 1989. Substratum conditioning and diatom
colonization in different current regimes. J. Phycol. 25: 790-793.

Peterson, C. G., Hoagland, K. D. & Stevenson, R. J. 1990. Timing of wave
disturbance and the resistance and recovery of a ﬁeshwater epilithic
microalgal community. J. N. Am. Benthol. Soc. 9: 54-67.

Peterson, B. J., Deegan, L., Helﬁich, J ., Hobbie, J. E., Hullar, M., Moller, B.,
Ford, T. E., Hershey, A., Hiltner, A., Kipphut, G., Lock, M. A., Fiebig, D.
M., McKinley, V., Miller, M., Vestal, J. R, Ventullo, R & Volk, G. 1993.
Biological responses of a tundra river to fertilization. Ecol. 74(3): 653-
672.

Pielou, E. C. 1969. An introduction to mathematical ecology. New York: John
Wiley & Sons

Potapova , M. & Charles, D. F. 2003. Distribution of benthic diatoms in US.
Rivers in relation to conductivity and ionic composition. Fresh. Biol. 48:
131 1-1328.

Raschke, R. L. 1993. Diatom (Bacillariophyta) community respone to
phosphorus in the Everglades National Park, USA Phycol. 32: 48-58.

Riseng, C. M., Wiley, M. J. & Stevenson, R J. 2004. Hydrologic disturbance and
nutrient effects on the structure of benthic communities in Midwestern US
streams: a covariance structure analysis. J. N. Am. Benthol. Soc. 23:309-
26.

Round, F . E. 1981. The ecology of algae. Cambridge University Press, pp. 1-653.

Round, F. E., Crawford, R. M. & Mann, D. G. 1990. The Diatoms. Biology and
morphology of the genera. Cambridge University Press, Cambridge, pp. 1-
747 .

Schoeman, F. R 1976. Diatom indicator groups in the assessment of water
quality in the Jukskei-Crocodile River System (Transvaal, Republic of
South Aﬁica). J. Limnol. Soc. S. Africa. 2: 21-24.

145

Shannon, C. E. & Weaver, W. 1949. The mathematical theory of communication.
University of Illinois Press, Urbana IL, pp] 17.

Simpson, E. H 1949. Measurement of diversity. Nature 163: 688.

Schlichting, H. E. 1960. The role of waterfowl in the dispersal of algae. Trans.
Amer. Microsc. Soc. 79: 160-166.

Schlichting, H. E. 1964. Meteorological conditions affecting the dispersal of
airborne algae and protozoa. Lloydia 27: 64-7 8.

Schlichting, H. E. 1969. The importance of airborne algae and protozoa. J. Air
Pollut. Control Ass. 19: 946-951.

Stahler, A N. 1957. Quantitative analysis of watershed geomorphology. Trans.
Am. Geophys. Union. 38: 913-920.

Steinman, A. D. 1996. Effects of grazers on freshwater benthic algae. In
Stevenson, R J ., Bothwell, M. L. & Lowe R L. [Eds] Algal Ecology.
Academic Press, San Diego, pp. 341-3 74.

Scheiner, S. M. & Rey-Benayas, J. M. 1997. Placing empirical limits on
metapopulation models for terrestrial plants. EvoI. Ecol. 11: 275-288.

Stevenson, R J. 1984. Procedures for mounting algae in a syrup medium. Trans.
American Microsc. Soc. 103: 320-321.

Stevenson, R J. 1995. Community dynamics and differential species performance
of benthic diatoms along a nutrient gradient. In Kociolek, J. P. & Sullivan,
M. J. [Eds] A Century of Diatom research in North America. A tribute to
the Distinguished Careers of Charles W. Reimer and Ruth Patrick. Koeltz
Scientiﬁc Books, USA, Champaign, pp. 29-45.

Stevenson, R. J. 1997. Scale-dependent determinants and consequences of
benthic algal heterogeneity. J. N. Am. Benthol. Soc. 16(1): 248-262.

Stevenson, R J. & Smol, J. 2003. Use of algae in environmental assessment. In
Wehr, J. D. & Sheath, R G. [Eds] Freshwater algae of North America.
pp 775-797.

Stevenson, R J ., Bothwell, M. L. & Lowe R. L. 1996. Algal Ecology:
Freshwater Benthic Assemblages. Academic Press, San Diego. pp 1-750.

146

Stevenson, R J ., Rier, S. T., Riseng, C. M., Schultz, R & Wiley, M. J. (in press).
Comparing effects of nutrients on algal biomass in streams in 2 ecoregions
with different disturbance regions. Hydrob.

Stoermer, E. F. & Julius, M. W. 2003. Centric diatoms. In Wehr, J. D. & Sheath,
R G. [Eds] Freshwater algae of North America. pp. 559-594.

Tilman, D. 1982. Resource competition and community structure. 296 pp.
Princeton University Press, Princeton

Tokesi, M. 1992. Dynamics of distribution in animal communities: theory and
analysis. Res. Pop. Ecol. 34: 249-273.

The European Parliament and the Council of the European Union. 2000.
Directive 200/60/EC of the European Parliament and the Council of 23
October 2000 establishing a framework for Community action in the ﬁeld
of water policy. Oﬂicial J. of the European Communities L3 27/1, 1-72.

Van Dam, H. 1982. On the use of measures of structure and diversity in applied
diatom ecology. Nova Hedwigia 73: 97-1 15.

Van Rensburg, B. J., McGeoch, M. A., Matthews, W., Chown, S. L. & Van

J aarsveld, A S. 2000. Testing generalities in the shape of patch
occupancy frequency distributions. Ecol. 81: 3163-3177.

Wilkinson, L. 1989. Systat: The System for Statistics. Evanston, 822 pp.

Winter, J. G. & Duthie, H. C. 2000. Export coefﬁcient modeling to assess
phosphorus loading in an urban watershed. J. Am. Water Resour. Assoc.
36 (5): 1053-1061.

Whittaker, R H. 1975. Communities and ecosystems, 2"“ edn. MacMillan, New
York, pp.1-162.

147

Table 3.1: Taxa names and abbreviation for diatoms found in the survey of all 69
sites and 8 reference sites; N-ascending number of sites; mean RA-mean relative
abundance.

 

 

 

All sites Reference

Taxon ABBR N mean RA N mean RA
Achnanthidirnn minutissimum (K012) Cam. ACM 69 20.8 8 38.0
Nitzschia dissipata (sz) Gran N11) 64 6.5 5 1.5
Gomphonerna angustatum (K012) Rabh. GOA 62 15.5 7 19.0
Navicula minima Grun NAM 59 2.6 7 0.4
Gomphonema olivaceum (Lyngb.) K1112. 58 7.6 5 3.2
Cymbella afﬁnis sz. CMA 57 10.5 6 11.2
Navicula veneta K012. 56 4.0 5 1.8
Surirella minuta Bréb. SUM 52 2.2 2 1.5
Achnanthes deflexa Reimer ACD 50 9.3 7 14.9
Cymbella sp.K CMK 50 2.7 8 2.8
Amphora pediculus (Kutzing) Grun. AMP 48 2.8

Planothidium lanceolatum (Bréb. ex Kutz.) L.-B. 47 0.9 2 0.1
Diatoma vulgaris Bory 45 1.8

Navicula cryptotenella Lange-Bert NAC 43 3.5

Reimeria sinuata (Greg) Kocio. & Stoer. 43 1.6 2 0.5
Navicula gregaria Donk 39 3.0 2 0.2
Nitzschia sociabilis Hust. 39 1.2

Caloneis bacillum (Gran) Cl. 38 1.0 2 0.6
Encyonema silesiacum (Bleisch) Mann 37 1.5 7 2.6
Gomphonema minutum (Ag.)Ag. 36 1.8 4 1.6
Meridion circulate (Grev.) Ag. 35 1.0 6 1.7
Synedra ulna (Nitz.) Ehr. SYU 35 1.2 3 1.2
Rhoicosphenia abbreviata (K012) Mann ROC 33 6.5

Nitzschia inconspicua Grun N11 32 8.5 2 0.5
Navicula triprmctata (O. F. Mull.) Bory 31 0.8

Nitzschia palea (Kutz) W. Sm. NIP 31 1.3 2 0.7
Cocconeis pediculus Ehr. 24 1.9 2 0.2
Nitzschia frustulmn (K012) Grun 23 2.6 2 0.2
Navicula arvensis Bust 22 0.4 3 0.5
Navicula mmiscus Schumann 22 1.0

Nitzschia perminuta (Grun) Peragallo 22 0.5

Gomphonema parvulurn (K012) Km. 21 7.7 2 0.7
Nitzschia palea var. tenuirostris Grun. 21 0.9

Nitzschia palea var. debilis (K012) Grurr 20 0.9 1 0.2
Nitzschia amphibia Grun. 18 0.7

Fragilaria vauchaiae (Kutzing) Petasen FRV 17 1.3 1 6.0
Cymbella delicatula K012. 15 1.7 5 2.9
Navicula capitatoradiata Germain 15 0.6

Nitzschia linearis (Ag. ex W. Sm.) W. Sm. 14 0.5

Achnanthes conspicua Mayer 12 1.0

Encyonopsis microcephala (Grtmow) Krammer 12 1.3 2 2.4
Navicula cryptocephala Km. 12 0.3 1 0.3

 

148

Table 3.1: cont’d

 

 

 

All sites Reference

Taxon ABBR N mean RA N mean RA
Navicula lanceolata (Ag) Ehr. NAL 12 3.2

Navicula subminuscula Mang. NASM 12 2.1

Nitzschia solita Hust. 12 0.6

Cocconeis placentula var. lineata (Ehr.) V. H 11 0.8

Sellaphora seminulum (Grun) Mann NASL 11 3.1

Nitzschia acicularis (Kntz.) W. Sm. NIA 11 1.2

Achnanthes ventralis (Krasske) Lange-Bert. 9 2.5 2 6.6
Eucocconeis laevis (Oestrup) Lange-Bertalot 9 0.7 4 1.0
Craticula molestiformis (Hustedt) Lange-Bert. 9 0.4

Nitzschia sinuata var. tahellaria (Gum) G.inV.H. 9 0.2

Surirella angusta Kutz. 9 0.3

Mayamaea atomus (Kutzing) Lange-Bertalot 8 0.9

Cocconeis placentula Ehr. 7 0.2

Cyclotella meneghiniana Kutz. 7 0.2

Fragilaria capucina var. gracilis (03m) Hust 7 0.7

Navicula exilis Km. 7 0.8 2 0.2
Aulacoseira granulata (Ehr.) Simonsen 6 0.3 1 0.1
Cocconeis placentula var. euglypta (Ehr.) C1. 6 0.3

Encyonema minutum (Hilse) Mann ENM 6 1.4 1 5.3
Fragilaria capucina var. nnnpens (Kntz.) L.-B. FR 6 4.2 3 7.4
Nitzschia heuﬂeriana Grunow 6 0.3

Cymbella cistula (Ehr.) Kirchn. CMC 5 0.9 3 1.0
Fragilaria capucina Desrn. 5 0.2 2 0.2
Synedra delicatissima W. Sm 5 0.2

Navicula cincta (Ehr.) Ralfs 5 0.4

Navicula recens Lange—Bert 5 0.2

Cyclostephanos invisitatos (H.& H.)'I'h.,Sto.& H. 4 0.2

Melosira varians Ag. 4 0.3

Mayamaea atomus var. permitis (Hustedt) L.-B. 4 0.7

Fistulifera pelliculosa (Brébisson ex Kutz.) L.-B. 4 0.2

Nitzschia pusilla Grun. 4 0.3

Synedra acus Kutz. 4 1 .2

Amphora montana Krasske 3 0.1

Encyonopsis falaisensis (Gnmow) Krammer 3 0.5

Cymbella hebridica Grunow ex Cleve 3 0.1

Fragilaria nanana Lange-Bert. FRN 3 1 .1

Psammodictyon constrictum (Gregory) Mann 3 0.2

Psammothidium chlidanos (H. et Heller.) L.-B. 2 1.2

Achnanthidium exilis (Kutzing) Round et Bukh. 2 1.6

Achnanthidium affrne (Grunow) Czarnecki 2 1 .0

Achnanthes gracillima Hustedt 2 0.3 2 0.3
Cocconeis placentula var. pseudolineata Geitl. 2 0.2

Cyclotella stelligera (Cl. & Grun.) V. H. 2 0.1

 

149

Table 3.1: cont’d

 

 

 

All sites Reference

Taxon ABBR N mean RA N mean RA
Cymbella mesiana Choln. 2 0.2
Denticula lmetzingii Guru. 2 0.4
Eunotia bilunaris var. mucophila L.-B. & Net. 2 0.3
Fragilaria bidens Heiberg 2 0.5
Fragilaria capucina var. mesolepta Rabh. 2 0.6
Tabularia fasciarlata (A3.) Williams and Round 2 1.1
Gomphonema affine K012. 2 1.1
Gomphonema truncatum Ehr. 2 0.2
Navicula harderii Hust. 2 0.3
Fallacia monoculata (Hustedt) Mann 2 0.2
Navicula subrotundata Hust. 2 0.2
Nitzschia capitellata Hust. 2 0.3
Nitzschia fonticola Grun. 2 0.2
Nitzschia perspicua Choln. 2 0.5
F ragilaria tenera (Smith) Lange-Bertalot 2 0.3
Karayevia clevei (Grunow) Kingston 1 0.2
Eucocconeis flexella (Katzing) Cleve 1 0.3
Planothidium hauckianum (Grunow) Ret Bukht 1 1.5
Psammothidium lauenburgianum (Hust.) Ret Bukht. 1 0.2
Psammothidium marginulatum (Grun) Ret Bukht. 1 0.2
Achnanthes minutissima var. jackii (Rab) L.-B. 1 1.0
Achnanthidium saprophila (Kob.et May.)Ret Bukht. 1 0.3
Rossithidium pusillum (Grunow) Ret Bukht. 1 0.5
Achnanthidium kranzii (L.-B.) Ret Bukht. 1 1.8
Amphora ovalis (K012) Katz. 1 0.2
Asterionella formosa Hassal 1 0.2
Aulacoseira alpigena (Grun.) Krammer 1 0.2
Caloneis tenuis (Greg) Kram. 1 0.2
Cymbella amphioxys (Kata) Cl. 1 0.8
Cymbella obscure Krasske 1 0.3
Encyonema prostratum (Berkeley) Ktltzing 1 0.3
Cymbella simmsenii Krammer 1 0.1 1 0.1
Denticula tenuis K012. 1 0.2
Diploneis ovalis (Hilse ex Rabh.) Cl. 1 0.1
Epithemia adnata (K011) Bréb. 1 0.2
Eunotia bilunaris (Ehr.) Mills 1 0.1
Eunotia exigua (Bréb. ex K1112.) Rabh. 1 0.2
Eunotia subarcuatoides Alles et al. 1 0.1
Fragilaria capucina var. distans (Grun.) Lange—Bert. 1 0.2
Fragilaria parasitica var. suboonstricta Grun. 1 0.2
Fragilaria capucina var. radians (sz.)1.ange-Bert. 1 0.5
F_ragilaria crotonensis Kitten 1 0.7

 

150

Table 3.1: cont’d

 

 

 

All sites Reference
Taxon ABBR N mean RA N mean RA
Gomphonema sarcophagus Greg. 1 1.0
Gomphonema augur var. sphaerophorurn (Ehr.) L.-B. 1 0.1
Gomphonema gracile Ehr. emend. V. H l 0.3
Gomphonema olivaeeoides Hustedt 1 0.9
Gyrosigrna eximium (Thwaites) Boyer 1 02
Melosira undulata (Ehr.) K012. 1 0.3
Craticula accomoda (Hustedt) Mann 1 0.2
Diadesmis eonfervaeea Kutzing 1 0.2
Navicula ingenua Hust. 1 1.0
Navicula lenzii Hust. l 0.3
Navicula gerrminii Wallace 1 0.3
Navicula salinarum Grun. l 0.2
Geissleria schoenfeldii (Hustedt) L.-B. et Metzeltin l 0.3
Navicula stroernii Hust. 1 0.3
Navicula tantula Hust. 1 0.3
Navicula tenelloides Hust. 1 0.3
Navicula trivialis Lungs-Bert l 0.3
Nitzschia amphi bioides Hust. 1 0.1
Nitzschia bacillum Hust. 1 0.3
Nitzschia clausii Hantz. 1 0.2
Tryblionella hungarica (Grun) Mann 1 0.3
Nitzschia recta Hantz. ex Rabh. l 0.3
Nimchia subconstricta Grunow l 0.3
Nitzschia tropics Hust. 1 0.2
Pinnularia gibba var. mesogongyla (Ehr.) Hust. 1 0.2
Stauroneis thermicola (Peters) Lund 1 0.2
Stephanodiscus nrinutus H. L. Sm. l 1.1
Stephanodiscus tenuis Hust. 1 0.5
Surirella brebissonii var. kuetzingii Kr. & L.-Bert l 0.3
Thalassiosira pseudonana Hasle & Heimdal 1 0.2

 

151

Table 3.2: Test of the inﬂuence of sampling on the abundance-distribution
relationship for Kentucky stream diatoms. Species-speciﬁc coefficient of

variation (CV) and average abundance (xm) were calculated across all sites. SE-
standard error, p — signiﬁcance level.

 

 

Variable Estimate sr‘s p
Intercept 0.04 0.09 0.66
ln(ln(x....)) 0.99 0.03 <0.0001
-2*ln(CV) 1.99 0.03 <0.0001

 

Regression analysis: Model: R2=0.98, F2,1gg=5095.22, p<0.0001

152

Table 3.3: Effect of species dispersal abilities on abundance and distribution
relationship for Kentucky stream diatoms.

 

 

Variable Estimate SE p
Intercept 4.62 2.04 0.02
Abundance 0.14 0.01 <0.0001
Dispersal ability 0.28 0.87 0.75

 

Multiple regression analysis: Model R2=0.43, F2,133=71.93, p<0.0001

153

2962
oonaoﬁewa HQ .826 858.2280: 2228 32220.28 222 8220822280222 2828022806322 ”Tm 0322.2.

 

 

 

e: 22.228 3 «.2 22.82 22.228 no new «.8 «228858;:
5.0 0.: 22.2 no 3 22$ 3 we we 23 95 _m
5.22 8.22 on $2 2.3 Er. em 3 3 32a 223 n:
28.9 8.08 od :2 0.5 eon 22.2 4.2 we :a meemw
8.22 22.82 22.8 0.52 22.5: 2283 ed mama 2.228 :22 2235
28.9 0.08 on on 8.8 0.8 2.2.. an R2 222.128.222-412
5.22 «.88 32 8.82 38 pews QR ago Sam :2 223 2-62
5.22 3m 3 no as E 22.22 2 Wm 22-.. 8&2 88.8

8.: m s o 2.22 m m o w Ia
No.22 8m «2 we 3: 82 5 mg a? 23 95 22228.6
28.9 2.8 «.82 2 24.8 «8 82 at 22mm 22-56 8:22.828

.2. 8 o 3 8.8 E o 8.22 on: 2.2.2 52,8 28:8

6.: mew o no 3.2 32 we 2 0.2 0. 238852 226;

we 8 m 3 22.2 8 o E 32 0. 2258562 26

5E mm“ 22on me EE lﬁu cmoE , 8.8:;

 

|lj

8% 85288.52 mote. 8:826”.

 

154

Table 3.5: Correlations between physicochemical variables in reference and non-
reference streams; (*) signiﬁcance level p<0.05.

 

 

 

 

 

Reference _

_ TN TP NP Si Chloride ALC CON CAN
TN 1

TP 0.19 1

NP 0.91‘ -0.17 1

Si -0.31 -0.97‘ 0.02 1

Chloride 0.96‘ 0.06 0.90“ -0.19 1

ALC -0.58 -0.62 -0.33 0.75‘ -0.55 1

CON -0.41 -0.88‘ -0.16 0.90' -0.24 0.53 1

CAN -0.05 0.06 0.07 -0.07 -0.27 0.22 -0.35 1
_ Nonreterenee

TN 1

TP 065‘ 1

NP 0.13 -0.3 1

Si -0.22 -0.17 -0.05 1

Chloride 0.73‘ 0.55 -0.05 -0.22 1

ALC 0.27 0.2 -0.05 -0.55 0.29 1

CON 0.49 0.5 -0.12 -0.49 0.55 0.83‘ 1

CAN 0.18 0.07 -0.2 0.17 0.1 0.21 0.03 1

 

155

2962 822285298 um .8228 82232228228 2282:2222 2828 82280.22 222 8232.228 2322222222800 66 £an

 

roodv
voodv
..oodv
woodv

mod

 

 

 

8.22 8.0 3.22 8.0 8.0 8.22 8.22 8.22 88528 8.8888

ed 2.2 8.0 «a «a 3 9.22 3 2225

8.22 8.22 5.22 8.22 88 3.22 8.0 8.0 885$

8 3 8.22 22.8 8 «2 88 on 8858

8.2.8 8.2. m: 8.2 «.8 o 2.2. 3&5 a... 2 a __Eeeozo

28E 5E mm“ 22on er EE mwu cuoE «23:23
mega oochCeaéFoz mega. ooceolwolm i

 

156

.8828on 225 85 .3 292282222228 02

2202225222228 2223222 omega - 2x. Amo>8vw 623 222 2828222 .0228 32282228228228 - A mvnmhozw
82228.» @828on 225 m 222 .20 2202282822 2882222828 1 mamv 883228222 mode 8 8 98 E228
2202282228228 8328222222228 2x. 3 22223 2822822222228 9228828822 222 28228.8 88 828228 8828on

22028 222222 220223.588 .8228 8288.22 222 8982 5282222228 8898 828% 80285 2 Em 038.2.

 

 

888 E {.2 88 88 x .8 8:85.20
8.8 R2 88 Q8 88 22.252 2.538 83885 88822888
8.8 a: 8.2 88 88 8.58 88.28 85822
8.8 8.8 3.2 8.2 8.8 .280 285832 58528855 522518522
.858 $2852 2. $28288 6288 88 88*

 

157

Table 3.8: Diatom species average similarity (aveS) in non-reference sites,
contribution from each species; species are ordered in decreasing contribution
with 90 cumulative contribution (Cum %) as a cut-off percentage; (SD)S —
standard deviation of the S (ith species) values; aveS/SD(S ) - discriminating
ratio, highest in bold; {(aveS) % - average percent contribution to similarity by
the (ith species).

 

 

Taxon aveS SD(S) aveS/SD(S) ?aveS% Cameo
Achnanthidium minutissirnum (Ki‘rtzing) Cza 10.43 1.14 9.15 30.62 30.6
Gomphonerm angustaturn (Ki'rtz) Rabh. 4.86 0.72 6.75 14.28 44.9
Nitzschia dissipata (Kiitz.) Grutt 3.21 0.93 3.45 9.44 54.3
Cymbella afﬁnis Krttzing 2.91 0.52 5.50 8.55 62.9
Gomphonema olivaceum (Lyngb) Kiitz. 2.57 0.68 3.78 7.53 70.4
Achnanthes deﬂexa Reimer 1.79 0.49 3.65 5.26 75.7
Navicula veneta Kiitz. 1.14 0.6 1.90 3.36 79.1
Navicula cryptotenella Lange-Bert. 0.71 0.52 - 1.37 2.09 81.1
Navicula minima Grun. 0.66 0.49 1.35 1.92 83.1
Cymbella sp. K 0.62 0.54 1.15 1.81 84.9
Surirelh minuta Bréb. 0.59 0.62 0.95 1.74 $8
Nitzschia inconspicua (3an 0.59 0.22 2.68 1.73 88.3
Amphora perpusilla (Grim) Grun. 0.59 0.46 1.28 1.73 90.1

 

158

Table 3.9: Diatom species average dissimilarity (ave 0) between reference and
non-reference sites, contribution from each species; species are ordered in
decreasing contribution with 90 % cumulative contribution (Cum %) as a cut-off
percentage; meanRAnr- mean relative abundance in non-reference sites, compare
with mean RA in reference sites Table 3.4; (SD)6 - standard deviation of the 6
(ith species) values; aveO/SD(6 ) - discriminating ratio, highest in bold; [(aveb)
% - average percent contribution to similarity by the (ith species).

 

 

Taxon meanRAnr ave 6 ave 6 (SD) 6/6SD gave 6% Cum%
Achnanthidium minutissimum 18.6 12.67 1.79 7.08 18.7 18.7
Gomphonema angustatum 11.6 8.42 1.08 7.80 12.5 31.2
Cymbella afﬁnis 8.8 6.85 0.79 8.67 10.1 41.3
Achnanthes deflexa 5.9 5.97 1.36 4.39 8.8 50.1
Gomphonema olivaceum 7 3.27 0.75 4.36 4.8 54.9
Nitzschia dissipata 6.6 3.04 0.93 3.27 4.5 59.5
Nimchia inconspicua 4.5 2.25 0.41 5.49 3.3 62.8
Rhoicosphenia abbreviata 6.5 1.75 0.47 3.72 2.6 65.4
Navicula veneta 3.5 1.75 0.67 2.61 2.6 68
Cymbella sp. K 1.8 1.57 l 1.57 2.3 70.3
Fragilaria capucina var. rumpens 0.1 1.41 0.59 2.39 2.1 72.4
Navicula cryptotenella 3.5 1.22 0.6 2.03 1.8 74.2
Navicula minima 2.5 1.17 0.52 2.25 1.7 75.9
Amphora pediculus 2.8 1.13 0.53 2.13 1.7 77.6
Navicula gregaria 1.9 0.95 0.25 3.80 1.4 79
Encyonema silesiacum 0.6 0.94 1 0.94 1.4 80.4
Surirella minuta Bréb. 1.8 0.92 0.66 1.39 1.4 81.8
Achnanthes ventralis 0.2 0.86 0.56 1.54 1.3 83
Cymbella delicatula 0.2 0.77 0.92 0.84 1.1 84.2
Gomphonema minutum l 0.69 0.48 1.44 1 85.2
Meridian circulare 0.4 0.68 0.84 0.81 1 86.2
Diatoma vulgaris 1.8 0.64 0.45 1.42 1 87.1
Reimeria sinuata 1 0.52 0.73 0.71 0.8 87.9
Nitzschia frustulum l 0.49 0.31 1.58 0.7 88.6
Fragilaria vaucheriae 0.3 0.48 0.47 1.02 0.7 89.3
Synedra ulna 0.7 0.45 0.51 0.88 0.7 89.9
Nitzschia sociabilis 1.2 0.37 0.64 0.58 0.6 90.6

 

159

Table 3.10: Effect of ecological specialization on the average abundance and
distribution of Kentucky stream diatoms. The table gives means 3: standard error.

 

 

Average
Category loge-abundance Distribution n
reference (gen) 1.18 1 0.12 0.74 1 0.06 37
random 0.33 :1: 0.02 0.14 :1: 0.00 113
nonreference (sp) 0.61 :I: 0.07 0.45 t 0.03 39
ANOVA:
R2 0.38 0.65
F 57.66 175.26
p <0.001 <0.001
dfmodel/dftotal 2/188 21188

 

160

 

Figure 3.1: Relationship between diatom species abundance and distribution (as
percent of sites occupied) in the survey for all sites (n=69); a. A scatter plot of
mean relative abundance for the sites occupied; b. A histogram number of species
and percent sites occupied (second y axes represents the proportion of taxa in the
bars) against their distribution

161

.28 828822

 

 

 

 

 

 

 

 

 

 

2282222288 8228 .20 222882822 2282222280 8222 .20 22282822
82 28 e8 28 N8 88 8.2 28 e8 28 N8 88

o.ole|.2L|rrrl_.l o o W
.w 2.8. N - m m
a N O .- LOW m e e .0 I.
Mme . m . . . . 82 m.
.m 2. 8 - m. - . - m2 .1.
mm 8 - 82 w m.
6.“ e8 . m ... . 88 m.
28. w
2 2 2 P mﬁ 2 2 2 2 mm m

.D
(U

162

2.8. 0322.2. 222 8 8220228582232.
8222228228 282222222on8 m2 awE 2228 222852228 222 82222 22230.2» 828on ”NM. 8222mm

822282.228: 222852228 :2 88.2 22230.20

 

 

 

 

 

8.2 28 88 m8-
2 2 , .. . ..
428 \xwo
.. .2228 .88 m
020.3572 m.
5222827222 .52 m.
t 22272 (no N
N222 8.2....
2 2 ho;

 

163

GENERAL CONCLUSIONS

The goals of my research were to evaluate ecological and evolutionary
strategies of benthic diatoms and relate them to nutrient conditions. Diatom
abundances and morphologies observed in nature are hypothesized to be a result
of ecological constraints and evolutionary processes. Many ecological factors
affect benthic algae, and competition for nutrients is likely one of the most
important. Species-speciﬁc evolutionary strategies involving physiological and
morphological traits develop to ensure advantages of individual algal species at
different environmental conditions. I tested the likelihood that sets of co-adaptive
traits do exist and that they improve performance of taxa with a combination of
experiments and surveys in the laboratory and in natural stream settings. The
results of these studies broaden our understanding of how periphyton
communities are structured.

In a near-natural setting (Chapter 1), stream water from a pristine
watershed was supplied to experimental streams in which nutrients were
manipulated. Recirculating experimental streams were used in a novel approach
when controlling for effects of nitrogen and phosphorus, providing constant
stream conditions, broad species pool, and nutrient supply of nitrogen and
phosphorus at limiting concentrations. Different species were expected to grow
best at low N and low P or at low and high concentrations of one nutrient. A
broad species pool was inoculated and a gradient of resource conditions among

channels was established in which only one nutrient was limiting at a time.

164

Responses of diatom species to each nutrient condition were measured as
differences in growth rates and density.

Diatoms were able to grow at very low P and N concentrations. The same
species were adapted to growing fastest in low N and low P conditions, so no
evidence was observed for tradeoffs Positive rather than negative correlations
were observed among species growth rates in low and high P concentrations, or
low and high N concentrations. Thus, the same species grew fastest in low and
high concentrations. The same diatom species may be co-adapted to live in both
low N and low P concentrations due to common co-occurrence of low
concentrations of both nutrients in streams.

Even though species grth rates were positively correlated in low and
high nutrient conditions (for both nutrients) the magnitude of those relations
decreased with increasing algal density on substrata. Some evidence for density-
dependent growth in competitive hierarchies was evident. Differences in N and P
utilization and storage during colonization of substrata may explain the diﬂ‘erent
strategies for N and P competition. Maximum growth rate of most taxa was
signiﬁcantly inﬂuenced by the growth of other taxa, especially initial colonizers
like Acnanthes deﬂexa. Functional response of diatoms was species speciﬁc and
time dependent. Abilities to utilize low nutrient concentrations and grow rapidly
seemed to be co-adaptive traits for benthic diatoms in streams. This may be due
to the likelihood that both low N and P supply constrain benthic diatom growth in

many habitats.

165

The rate of biomass production (growth rate) in benthic diatoms does
depend on the supply of limiting nutrients, but maximum grth rates of common
diatom taxa were reached at very low nutrient concentrations. Little evidence for
tradeoffs in resource specialization was observed. These results are inconsistent
with the resource competition theory (Tilman 1982) in which a species able to
grow at low concentration of one nutrient was not a good competitors at low
concentrations of another nutrient. This work demonstrates that in stream benthic
communities when diatoms are the dominant algal group, some diatom species are
adapted to grow rapidly and to utilize very low resources.

Intra- and interspeciﬁc competition of clonal populations of
Achnanthidium minutissimum were examined in low and high levels of nutrients
and light (Chapter 2). Growth rate of Achnanthidium minutissimum was
compared at low and high densities and when in competition with another
monoraphid diatom (Cocconeis placentula var. lineata) or with a large stalk-
forming Cymbella cistula species. Circular artiﬁcial channels were designed as a
novel experimental apparatus for use within an incubator. In this study, I was
able to create periphyton communities of genetically identical populations to
evaluate the ecological effects of other diatoms on clonal populations of A.
minutissimum. Nutrients limited growth of A. rnirmtissimum. Achnanthidium
minutissimum growth decreased with increase of diatom density and that decrease
was dependent on the culture combinations. Light had a strong negative effect
when nutrients were limiting but had a positive effect when nutrients were

available. The mechanisms to compete for low nutrients and low light of those

166

morphologically diverse diatoms may be different. A. minutissimum and C.
placentula var. lineata are both monoraphid diatoms, but the latter had very slow
growth rates in cultures. Although C. cistula grew well in culture, formed stalks,
and overgrew A. minutissimum, the competitive effect on A. minutissimum may
have been reduced by stalks providing substratum for it’s colonization. The
prediction of strong intraspeciﬁc competition in limiting resources was conﬁrmed
in this experiment. The importance of intraspeciﬁc competition regulating A.
minutissimum, the most common diatom in North American streams, is an
important contribution for the knowledge about competition regulating
composition of stream periphyton. In addition, this study presents evidence of
possible evolutionary adaptation to low light by common diatoms in natural
streams.

Evolutionary strategies of several diatom species were evident in the large
scale pattern observed in nature (Chapter 3). Species would be expected to be
most abundant in relatively few habitats in a region if species specialized with
optimal performance in speciﬁc environmental conditions. Contrary to this
prediction, a positive correlation was observed between diatom abundance and
breadth of their distribution in the Knobs ecoregion of Kentucky and southern
Indiana. A positive relationship between relative abundance and breadth of
species distribution was observed. Few taxa were abundant and they had broad
distributions, the majority (79%) of the observed diatom species had low
abundance and appeared in few sites (less that 10% of sites). Findings on relative

abundance changes with nutrients in experimental work were successfully linked

167

with the patterns from the survey. Mean relative abundance of species ﬁom
pristine streams decreased with increasing nutrient and chloride concentrations,
but distribution remained wide. These results do warrant concern in regards to
conservation of the native low nutrient, low diversity sites in the Knobs
ecoregion.

To evaluate ecological constraints on abundant and rare taxa, species
abundance and frequency of occurrence were related to variation in ecological
factors, which was dominated by variability in nutrients and chloride associated
with an human disturbance gradient. Species that were most common and most
abundant were species that occurred in streams in low disturbance watersheds.
Relative abundance of these species decreased with increasing nutrient and
chloride concentrations, however they persisted in habitats with high human
disturbance. Taxa requiring high nutrients were more diverse than taxa from
pristine streams, but even those high-nutrient taxa with widest distributions had
low abundance. Some taxa requiring high nutrients grew faster than taxa ﬁom
pristine streams when streams were anthropogenically enriched, which caused,
partial turnover of taxa in the natural diatom assemblage. Evolutionary strategies
of several diatom species were evident in the large scale pattern observed in
nature. Integration of results provided a more certain description of the ecological
strategies and adaptations employed by diatoms when multiple research

approaches were used rather than any one approach alone.

168

The observed patterns were due to the species-speciﬁc abundance and
distribution relationship of relatively small number of diatom species. Increase in
nutrients cannot have a direct negative effect on diatoms, because algae depend
on nutrients. One of the important ﬁndings in my work was that higher nutrients
have species speciﬁc eﬂ’ects. Diatom species that were able to grow at very low
nutrient concentrations grew equally well when nutrients were not limiting, but
the high nutrient species grew faster with nutrient increases and were limited from
low nutrient habitats. Correspondence between ob served natural patterns on
regional scales and experimental results in my dissertation indicate that diatoms
have adapted to nutrient-related environmental conditions. On a larger scale in
streams, potential tradeoffs for slow and fast growth in different environmental
conditions have enabled specialization of taxa.

In the three studies, one diatom Achnanthidium minutissimum was an
important species in the periphyton communities. My experimental study
conﬁrmed the ability of this diatom to grow at very low nutrients and low light,
both common conditions in natural steams. Achnanthidium minutissimum
employed different strategies in different conditions. This diatom grew fast at
low nutrient concentrations. It was strongly regulated by intraspeciﬁc
competition but not interspeciﬁc competition. Achnanthidium minutissimum was
able to attach to other diatoms, and to produce mucilaginous support structures
under some conditions. The observed results suggest that diatoms have great
abilities to adapt to environmental conditions and that in the process they adopt

evolutionary strategies, like changes in growth rate, changes in grth form or

169

both, that can explain community and regional patterns commonly observed in

nature.

170

LITERATURE CITED

Tilman, D. 1982. Resource competition and community structure. Princeton
University Press, Princeton, 296 pp.

171

APPENDICES

172

APPENDIX I :Achnanthidium minutissimum distribution. Habitat and locality

from selected records (more than 10% reported abundance) in USA. Source:

General Collection (GC) at the Academy of Natural Sciences, Philadelphia, PA

 

 

 

Habitat type W Ilectionlocali State Collected by Year GC number
cave tee , oro 0
Clear mountain stream National Forest Arizona C. Reimer 1944 6736a
Stagmnt pool near Beaver Dam Arizona 1934 6728a
Clear mountain stream White River Arkansas 616168
Plankton San Joaquin River California RR Grant Jr 1962 620128
Waterfall California R Patrick 1952 6778
Plankton Sacremerrto River California 1956 8013
Plankton Lake Tahoe California 6222011
Warm sluggish stream South Boulder Creek California 8848b
Periphyton North fork of Colorado River Colorado 89748
Epiphytes on grass & weeds Roudre River Colorado 8986b
Stagnant pool Thompson River Colorado 8843b
Cold muddy stream bed Tonahutu creek Colorado 9001b
Rock scrub East River Colorado 62521
hot spring above Copper lake Colorado J. Lennox 1967 62584
Surface scum Coal Creek Colorado 62570
Periphyton Red Clay Creek Delaware M.H. Hohn 1956 44090
periphyton Delaware River Delaware M.I-L Hohn 1957 44226b
epiphytes Silver springs Florida 44479
Plankton Lake Okechobee Florida 551608
Periphyton Escambia River Florida 43850
Periphyton Savannah River Georgia 43929
Periphyton Loon Creek Idaho 64306a
Plankton Lake Mchigan Illinois 46919
Raised bog Cabin Creek Indiana 45669
Plankton Kansas River Kansas 8052
Periphyton Green River Kentucky 55859a
Plankton Lake Maurepas Lousiana 48526a
Periphyton Kennebec River Maine 56723a
Periphyton Potomak River Maryland 45426a
Periphyton Center creek Missuri 4514711
Plankton Flathhead Lake Montana 62224
Periphyton Assunpink River New Jersey 44888
Periphyton Auglaize River Ohio 45184a
Periphyton Burpee Station Oregon J. L. Parker 1969 62744
Periphyton Slippery Rock Creek Pemrsylvania RR. Grant Jr 1967 50884
Periphyton Brandywine Creek Pennsylvania C. Reimer 1964 48716
Periphyton Susquehanna River Pennsylvania RR. Grant Jr 1970 52169
Diatoms on mud ﬂat Savannah River South Carolina CReimer 1965 48869a-b
Mat, scraped from stone Savannah River South Carolina Joy Morrill 1972 528823
Periphyton Holston River Tennessee R. R Grant J 1965 48925a
(hr rock near shore Cumberland River Tennessee C. Reimer 1973 536613
On cypress log Sabine River Texas R R Grant J 1962 62036a
Periphyton - composit Guadalupe River Texas RR. Grant Jr 1962 62l93a
Periphyton Providence Creek Utah G. Piranian 1936 6775a
Floating in stream from hot Lincoln Beach Utah Edna Snow 1930 604288
Diatom tuft in rifﬂe Holston River Vrrginia R R Grant I 1974 56409a
Periphyton - composit Potomac River Virginia CReimer 1973 533658
Periphyton Quinault River Washington DC C.D. McIntirt 1967 60990

 

173

APPENDIX 1.1: Nutrient concentrations pg L'1 along N gradient (mean 4: SD,

 

 

 

 

 

 

 

 

 

 

n=3).

day cone WEN—WWW
1 0 33 (:1) 8 (:3) 137 (:3) 25 (:2) 213 (:3) 11 (:1)
1 16 38 (:2) 9 (:3) 122 (:5) 26 (:1) 30 (:2) 11 (:1)
1 64 67 (:3) 16 (:10) 148 (:5) 23 (:2) 27 (:3) 11 (:1)
1 256 167 (:5) 16 (:12) 233 (:7) 25 (:1) 30 (:2) 11 (:1)
1 512 219 (:4) 8 (:1) 340 (:3) 23 (:1) 31 (:05) 11 (:0)
1 1024 500 (:33) 15 (:14) 578 (:38) 25 (:3) 29 Q3) 11 (:1)
3 0 12 (:2) 33 (:2) 139 (:3) 29 (:6) 33 (:8) 11 (:0)
3 16 22 (:3) 20 (:4) 143 (:3) 24 (:3) 31 (:3) 11 (:0)
3 64 74 (:26) 29 (:2) 184 (:18) 59 (:35) 64 (:37) 11 (:0)
3 256 188 (:24) 24 (:6) 273 (:24) 32 (:12) 35 (:12) 11 (:0)
3 512 242 (:28) 34 (:8) 332 (:29) 26 (:2) 33 (:2) 11 (:0)
3 1024 503 (:71) 34 (:5) 636 (:56) 22 (:2) 27 (:2) 1 1 (:0)
6 0 0 (:0) 9 (:1) 126 (:5) 25 (:1) 49 (:1) 12 (:0)
6 16 2 (:1) 9 (:1) 126 (:12) 26 (:2) 46 (:2) 12 (:0)
6 64 30 (:16) 10 (:1) 152 (:9) 26 (:10) 46 (:12) 12 (:0)
6 256 289 (:130) 14 (:3) 423 (:174) 49 (:29) 68 (:33) 10 (:1)
6 512 401 (:36) 11 (:1) 552 (:53) 24 (:5) 43 (:5) 10 (:0)
6 1024 898 (:153) 13 (:1) 1045 (:121) 27 (:7) 41 (:10) 10 (:0)
8 0 5 (:0) 13 (:2) 91 (:21) 26 (:7) 50 (:3) 10 (:0)
8 16 9 (:1) 12 (:1) 106 (:36) 21 (:4) 46 (:1) 10 (:1)
8 64 21 (:3) 16 (:5) 121 (:20) 20 (:63) 44 (:2) 9 (:0)
8 256 185 (:129) 16 (:3) 312 (:131) 26 (:11) 56 (:32) 9 (:0)
8 512 280 (:122) 17 (:1) 437 (:93) 18 (:6) 40 (:9) 9 (:0)
8 1024 668 (:504) 18 (:1) 1113 (:336) 27 (:8) 46 (:15) 9 (:0)
10 0 4 (:1) 6 (:0) 106 (:3) 47 (:24) 55 (:25) 11 (:0)
10 16 6 (:2) 8 (:5) 107 (:7) 32 (:8) 39 (:8) 10 (:0)
10 64 11 (:8) 9 (:2) 128 (:16) 36 (:4) 39 (:8) 9 (:2)
10 256 85 (:3) 58 (:10) 254 (:46) 37 (:113) 45 (:13) 10 (:0)
10 512 229 (:34) 62 (:8) 473 (:76) 41 (:10) 52 (:14) 1o (:0)
10 1024 802 (:331) 49 (:15) 997 (:397) 30 (:5) 50 (:9) 9 (:1)
12 0 3 (:1) 14 (:2) 162 (:5) 40 (:2) 52 (:5) 10 (:0)
12 16 6 (:2) 20 (:8) 142 (:12) 37 (:3) 50 (:1) 10 (:0)
12 64 16 (:2) 31 (:3) 189 (:30) 42 (:3) 60 (:7) 9 (:2)
12 256 87 (:23) 42 (:13) 273 (:21) 44 (:6) 54 (:8) 9 (:0)
12 512 220 (:37) 98 (:10) 502 (:75) 46 (:12) 54 (:13) 8 (:0)
12 1024 760 (:55) 51 (:17) 931 (:64) 36 (:3) 44 (:4) 8 (:0)
14 0 3 (:1) 12 (:1) 129 (:7) 58 (:60) 88 (:66) 10 (:1)
14 16 9 (:2) 12 (:1) 99 (:2) 21 (:7) 42 (:4) 10 (:0)
14 64 15 (:4) 12 (:1) 152 (:49) 50 (:46) 51 (:5) 9 (:1)
14 256 244 (:255) 15 (:2) 442 (:311) 41 (:35) 79 (:49) 9 (:1)
14 512 283 (:56) 14 (:2) 477 (:79) 14 (:6) 41 (:9) 9 (:0)
14 1024 723 (:19) 14 (:1) 928 (:53) 20 (:6) 168 (:227) 9 (:0)

 

174

APPENDIX 1.2: Nutrient concentrations pg L'1 along P gradient (mean i SD,

 

 

 

 

 

 

 

 

 

 

 

n=3).
day Poonc “Us N734 TN SRP TP 3T
1 0 234 (:0) 77:0) 254 (:4) 5 (:1) 9 (:0) 11 (:0)
1 2 245 (:39) 11 (:7) 321 (:13) 9 (:1) 1o (:1) 11 (:0)
1 9 227 (:9) 19 (:19) 352 (:w) 9 (:0) 12 (:1) 11 (:1)
1 32 219 (:5) 12 (:7) 323 (:22) 17 (:3) 20 (:3) 11 (:1)
1 50 214 (:4) 9 (:1) 340 (:3) 23 (:1) 31 (:1) 11 (:0)
1 1439 125 (:109) 9 (:3) 229 (:94) 30 (:30) 35 (:31) 11 (:_0)_
3 0 306 (:69) 27 (:4) 410 (:99) 5 (:1) 9 (:3) 11 (:0)
3 2 159 (:116) 21 (:2) 299 (:100) 5 (:2) 7 (:2) 11 (:0)
3 9 442 (:399) 32 (:7) 610 (:339) 9 (:4) 11 (:3) 11 (:0)
3 32 329 (:197) 21 (:1) 436 (:222) 21 (:9) 26 (:11) 11 (:0)
3 50 242 (:29) 34 (:9) 392 (:29) 26 (:2) 33 (:2) 11 (:0)
3 129 205621) 30(:6) 321 (:35) 524:1) 59(:3) 11 (:_0)_
6 0 423 (:79) 11 (:1) 915 (:404) 4 (:0) 5 (:1) 11 (:0)
6 2 451 (:79) 11 (:1) 579 (:100) 5 (:1) 10 (:0) 10 (:0)
6 9 426 (:39) 13 (:4) 493 (:109) 10 (:7) 20 (:9) 9 (:3)
6 32 391 (:64) 13 (:1) 499 (:49) 17 (:4) 24 (:3) 1o (:0)
6 50 401 (:36) 11 (:1) 551 (:53) 24 (:5) 43 (:5) 10 (:0)
6 129 347 (:11) 12(:1) 456(:19) 97 :12 119 :19 10 :0
9 0 499 (:47) 19 (:2) 620 (:43) 3 (:2) 4 (:1) 1o (:0)
9 2 409 (:9) 12 (:1) 549 (:25) 3 (:1) 4 (:1) 9 (:0)
9 9 221 (:139) 12 (:0) 314 (:142) 5 (:0) 7 (:0) 9 (:0)
9 32 300 (:70) 14 (:0) 471 (:59) 12 (:4) 29 (:4) 9 (:0)
9 50 290 (:122) 17 (:1) 437 (:99) 19 (:6) 40 (:9) 9 (:0)
9 139 209 (:99) 16 (:1 ) 399 (:96) 62 (:43) 90 (:41) 9 (:_0)_
1o 0 497 (:64) 16 (:9) 590 (:51) 3 (:2) 7 (:2) 10 (:0)
10 2 301 (:105) 15 (:3) 447 (:74) 2 (:1) 6 (:1) 9 (:0)
10 9 560 (:401) 22 (:2) 674 (:394) 4 (:0) 9 (:0) 9 (:1)
1o 32 529 (:562) 23 (:4) 724 (:640) 21 (:14) 32 (:21) 9 (:1)
10 50 229 (:34) 62 (:9) 473 (:76) 41 (:10) 52 (:14) 10 (:0)
10 129 11g(:99) 39(:10) 309 (:191) 51 (:43) 65(:51) 9&1)
12 0 459 (:31) 19 (:6) 613 (:23) 3 (:1) 4 (:1) 9 (:0)
12 2 496 (:159) 40 (:15) 690 (:199) 2 (:1) 5 (:2) 9 (:0)
12 9 370 (:65) 40 (:7) 599 (:93) 4 (:0) 7 (:1) 9 (:1)
12 32 297 (:46) 39 (:19) 457 (:23) 21 (:4) 29 (:3) 9 (:1)
12 50 220 (:37) 99 (:10) 502 (:75) 46 (:12) 54 (:13) 9 (:0)
12 129 120 :102 40 :31 299 :159 69 :64 79 :72 9 :2
14 0 514 (:104) 15 (:4) 591 (:101) 2 (:1) 4 (:1) 10 (:0) ~
14 2 519 (:169) 12 (:0) 992 (:160) 4 (:1) 6 (:1) 9 (:3)
14 9 323 (:29) 11 (:0) 469 (:6) 5 (:0) 9 (:0) 9 (:0)
14 32 799 (:720) 14 (:2) 901 (:991) 10 (:4) 44 (:16) 6 (:3)
14 50 293 (:56) 14 (:2) 477 (:79) 14 (:6) 41 (:9) 9 (:0)
14 129 479 (:401) 16 (:2) 660 (:437) 110 (:72) 169 (:79) 7 (:3)

 

175

APPENDD( 3.1: Site locality and identiﬁcation sampled for 1996 and 1997, sites

in bold are considered reference.

 

 

19% 1997

 

site
I Barebone Creek Mo River Wu 6 . 5.4
2 Bird Hollow BIHO Little Blue River Crawford 38.36 86.77
3 Bluelick Creek BLCR Ohio River Bullitt 38.03 85.69
4 Bogard Creek BOCR Little Blue River Crawford 38.29 86.43
5 Brushy Creek BRCR Little Blue River Crawford 38.31 86.44
6 Brooks Run BRRU Ohio River Bullitt 38.05 85.69
7 Brownstown Creek BTCR Little Blue River Crawford 38.35 86.48
8 Buck Creek BUCR Buck Creek Harrison 38.14 86.04
9 Camp Fork Creek CAFO Little Blue River Crawford 38.34 86.41
10 Cain Run CARU Ohio River Jefferson 38.07 85.87
11 Cedar Creek #1 CBC] Salt River Bullitt 38.06 85.65
12 Cedar Creek #2 CEC2 Salt River Bullitt 38.42 85.45
13 Com Creek COCR Indian River Harrison 38.33 86.04
14 Crandall Branch CRBR Indian River Harrison 38.29 86.07
15 Crooked Creek CRCR Salt River Bullitt 37.89 85.72
16 Dog Creek DOCR Little Blue River Crawford 38.36 86.45
17 Goose Creek GOCR Ohio River Jefferson 38.3 85.62
18 Hardy Creek I-IRCR Ohio River Trimble 38.61 85.27
19 Harrison Fork HAFO Ohio River Nelson 37.87 85.6
20 Harts Run HARU Salt River Bullitt 37.86 85.61
21 Hite Creek HTCR Ohio River
22 Jersey Park Creek JPCR Indian River Floyd 38.37 85.96
23 Knob Creek KNCR Ohio River Bullitt 38.04 85.79
24 Little Indian Creek LIIN Indian River Floyd 38.31 85.91
25 Long Lick Creek LLCR Salt River Bullitt 37.93 85.71
26 Lower Wilson Creek WICR Salt River Nelson 37 .87 85.61
27 Lickskillet Creek LSCR Salt River Bullitt 37 .93 85.65
28 MB Harrods Creek HACR Ohio River Oldham 38.45 85.42
29 MF Beargrass Creek MFBG Ohio River Jefferson 38.24 85.7
30 MF Blue River MFBL Blue River Washington 38.59 85.98
31 MF Buck Creek MFBU Buck Creek Harrison 38.14 86.04
32 MF Indian Creek MFIN Indian River Floyd 38.36 85.96
33 NF Harrods Creek NFHA Ohio River Oldlnm 38.42 85.45
34 Otter Creek OTCR Little Blue River Crawford 38.3 86.53
35 Overalls Creek OVCR Salt River Bullitt 37.87 85.6
36 Pryors Fork PRFO Ohio River Trimble 38.06 85.41
37 Racoon Branch RABR Indian River Harrison 38.28 86.1
38 Richland Creek RICR Indian River Floyd 38.32 86.01
39 SF Blue River SFBR Blue River Washington 38.53 85.92
40 Stinking Fork S'I'FO Little Blue River Crawford 38.23 85.53
41 Turkey Fork TUFO Little Blue River Crawford 38.2 86.4
42 Upper Wilson Creek UWIL Salt River Nelson 37.87 85.6
43 Whiskey Run WHRU Ohio River Washington 38.52 85.92
44 Yellow Fork YEFU 1mm River Floyd 38.35 85.91

 

176

uuuuu