Plant-microbial interactions in mangrove sediments under different nutrient conditions

PDF

Download

Open document

Flip pages

Summarize

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content PLANT-MICROBIAL INTERACTIONS IN MANGROVE SEDIMENTS UNDER
DIFFERENT NUTRIENT CONDITIONS

by

Isabel C. Romero

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
OF THE UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirement for the Degree
DOCTOR OF PHILOSOPHY
(OCEAN SCIENCES)

December 2009

Copyright 2009 Isabel Cristina Romero
ii
DEDICATION

To my loved ones,
Gerardo and Susana

iii
ACKNOWLEDGMENTS

I want to address my most sincere acknowledgments to all the people that helped
and supported this thesis work during its fulfillment. In particular to:
Dr. Myrna Jacobson, my thesis advisor for giving me the opportunity to join the
Biocomplexity Project, and for providing support and confidence in many times thorough
this work.
Dr. Doug Capone, my thesis coadvisor for welcoming me to his lab and for
offering countless support and advice during this work.
All the professors for giving me the opportunity to learn and analyze the samples
for my work: Dr. Marilyn Fogel for her numerous recommendations and advice during
and after field trips, and for opening her lab and home for me during the last stage of the
lab work; Dr. Jed Fuhrman for supporting my work and ideas, and also for letting me be
part of his team for about a year and for letting me participate in his weekly lab meetings
with great conversation and food; Dr. Susan Ziegler who welcomed me in her lab and
home, and for her support of my work.
All the institutions for their support that made possible the success of this work:
The National Science Foundation Grant “Biocomplexity of Mangroves”. The College of
Letters, Arts and Sciences – USC for the “Final Dissertation Fellowship”; the Graduate
School – USC for the “Oakley Fellowship”; and the Wrigley Institute for Environmental
studies and Rose Hill Program for the “Summer Fellowships”. The Smithsonian
Environmental Research Center, Washington D.C., for the one-year Fellowship. The
iv
Smithsonian Tropical Research Institute for a scholarship to participate in the “Tropical
Marine Ecology Graduate Course” Bocas del Toro, Panama. The WiSE “Travel Grant”
Fellowship for presenting at the 14
th
Ocean Sciences Meeting, Florida, and for their
amazing support with the “Childcare subsidy”.
To all the professors that were part of this work at different stages and gave me
advice on how to make it happen: many thanks to Dr. Michelle Riconscente, Dr. Doug
Hammond, Dr. Wiebke Ziebis, Dr. Lowell Stott, and Dr. Mattew Wooller.
To everyone in the Department of Biological Sciences who gave me support and
helped me many times, specially Linda Bazilian for being a super great student advisor
and Don Bingham for his prompt to help in any aspect.
Also, I want to thank Cyndi Waite for helping me with all the paperwork during
all these years.
Also a big thank you goes to all my USC friends for their time, caring and support
during this work.
Special thanks goes to my family for their constant support and affection.
My biggest thanks go to Gerardo, my husband, for his constant support during all
the different stages of this thesis work. A very special thanks go to Susana, my daughter,
who gave me so many reasons to laugh, dance and sing during the most difficult times!

v
TABLE OF CONTENTS

Dedication ii

Acknowledgments iii

List of Tables vii

List of Figures ix

Abstract xiii

Chapter 1: Introduction 1
Global significant of Mangroves 4
Plant-bacterial interaction 6
Nitrogen dynamics in mangroves 9
Carbon cycling in mangroves 12
Research questions 14

Chapter 2: Study area: Mangrove forests in Twin Cays, Belize 18

Chapter 3: Bacteria community and carbon source utilization patterns in
mangrove sediments: Effects of long-term experimental nutrient
enrichment 24

Chapter 4: Spatial and temporal variability of nitrogen-fixing microbial
populations in the rhizosphere of mangrove sediments 60

Chapter 5: Phylogenetic diversity of diazotrophs in the rhizosphere of
mangrove sediments 87
vi

Chapter 6: Synthesis 117

References 126

Appendix 1 147
Appendix 2 148
Appendix 3 149
Appendix 4 150
Appendix 5 151
Appendix 6 155
Appendix 7 157
Appendix 8 159

vii
LIST OF TABLES

Table 1. Pore-water and sediment parameters for each nutrient treatment (Control:
Ctrl, Nitrogen: N, Phosphorus: P) in the Fringe and Interior mangrove zones. The
data are: Avg. ± SE.

36
Table 2. Relative abundance (%) of PLFAs for the different treatments (Ctrl:
control; N: nitrogen-fertilized; P: phosphorus-fertilized) at each mangrove zone.
Values are Avg. ± SE.

43
Table 3. Calculated difference in bacterial PLFA concentration (µg PLFA / g dw;
Avg. ± SE) between the Ctrl- and nutrient treatments (N and P) in the mangrove
zones for each depth interval. * denotes significant differences.

44
Table 4. δ
13
C (‰) of bacteria (i+a15:0) compared to total organic carbon (TOC)
and all potential carbon sources found in the mangrove sediments (dead and live
roots, leaf litter, microbial mats, and seagrasses). Bacteria biomarkers were
corrected for fractionation using a known correction factor of -3.7±2.1‰ (Bouillon
and Boschker 2006). All values shown as Avg. ± SE.

49
Table 5. Pore-water parameters for each nutrient treatment and depth in the interior
mangrove zone. The data are: Avg. ± S.E.

71
Table 6. Distribution of mangrove sediment nifH clones among phylogenetic
groups.

98
Table 7. Presumptive microbial processes of TRFLP-OTUs related with clone-
OTUs. Similarity (%) based on amino acid sequences. Presumable microbial
processes based on phylogenetic affiliation at the group level from the phylogenetic
tree (Figure 1).

105
Table 8. Variability of environmental parameters at three depth intervals in each
nutrient treatment (Ctrl: control; N: Fertilized with nitrogen; P: Fertilized with
phosphorus). See text for details on the fertilization experiment. Data shown as
Avg. ± SE; N≥3.

107
viii
Table 9. Bray-Curtis similarity index (%) between the Ctrl-treatment and the
fertilized treatments (N and P) based on TRFLP-nifH OTUs relative abundance.

108
Table 10. pRDA results for the spatial variation in the TRFLP-nifH OTUs
explained by environmental parameters in the mangrove sediments.
108

ix
LIST OF FIGURES

Figure 1. Twin Cays is located 12 km off-shore from the coast of Belize (Top)
and is composed of two islands that are surrounded by seagrasses (bottom, aerial
photograph). Star: Boa Flats site.

19
Figure 2. Peat-based sediments in Twin Cays (A) are composed mainly by
mangrove roots (B).

20
Figure 3. Mangrove zones in Twin Cays, Boa Flats (A). Tree-height gradient
with tall trees in the fringe zone (B) followed by an interior zone with dwarf trees
and interior ponds (C).

21
Figure 4. Boa Flats sediment surfaces. In the fringe zone, sediment can be
covered by mangrove (A) or a mixture of mangrove and seagrass (B) leaf litter. In
the interior zone, sediment is partially covered by microbial mats and some
mangrove leaf litter.

23
Figure 5. Pore-water and total sediment parameters (Avg. ± SE) in the fringe and
interior mangrove zones for each depth interval (cm) at the Ctrl-treatment (natural
conditions). * Missing data.

37
Figure 6. Sediment components (Avg. ± SE) in the fringe and interior mangrove
zones for each depth interval (cm) at the Ctrl-treatment (natural conditions). Leaf
litter at 20-30 cm depth in the interior zone was not found in the cores collected.

38
Figure 7. Comparison of δ
13
C and C/N ratios between the nutrient (N, P) and
control treatments (Ctrl) denoted as ∆ from the fringe, and interior mangrove
zones. S: Total sediments; DR: Dead roots; LL: Leaf litter; and R: Live roots.
Data shown as Avg. ± SE.

40
Figure 8. Relative abundance (%) of live roots for the fringe and interior
mangrove zones in each nutrient treatment (Ctrl, N, P) at each depth interval
(cm). Data shown as Avg. ± SE.

41
x
Figure 9. Correlation of bacterial PLFA concentration (µg/g dw) between the
control treatment (Ctrl) and the nutrient treatments (N and P) at the fringe and
interior mangrove zones.

42
Figure 10. Relative abundance (%) of bacterial PLFAs in the Fringe and Interior
mangrove zones at different depth intervals

46
Figure 11. Ordination plot of RDA results showing sample clusters based on
bacterial PLFAs in response to environmental variables (70% explained variance;
p < 0.01). Environmental variables are indicated as triangles (categorical factors)
and arrows (numerical variables). The direction of the arrows denotes the steepest
increase in the variable, and the length indicates the strength in explaining PLFAs
variation relative to the other variables.

47
Figure 12. δ
13
C-PLFA (‰) of bacterial groups found in the fringe and interior
mangrove zones under natural conditions (A) and at different sediment depth
intervals (B). All values shown as averages ± SE. Actinomycetes: Actino.; Gram-
Negative: Gram-; Gram-Positive: Gram+.

49
Figure 13. Carbon isotope ratios of PLFA microbial groups (Gram-Negative and
Gram-Positive bacteria) and the general biomarker 16:0 in the sediments of the
fringe and interior mangrove zones. Data in each zone is shown under the
different nutrient treatments (Ctrl, N and P) at different depth intervals (0-1, 5-10,
20-30 cm). All values shown as Avg. ± SE.

50
Figure 14. δ
13
C values in the interior mangrove zone for PLFA biomarker 16:0
and the microbial group Gram-Negative bacteria, and for live roots. Data in each
nutrient treatments (Ctrl, N and P) is shown at different depth intervals (0-1, 5-10,
20-30 cm). All values shown as Avg. ± SE.

58
Figure 15. Comparisons of the relative abundance (%) of live and dead roots
among the nutrient treatments (Ctrl, N, P) at each depth interval in the interior
zone of Twin Cays mangrove forests. Data shown as Avg. ± SE.

73
Figure 16. Pore-water parameters at the different nutrient treatments (Ctrl, N, P)
during the wet and dry season. Data shown as Avg. ± SE.

74
xi
Figure 17. N
2
fixation rates diel variability under the different nutrient treatments
(Ctrl, N, P) in each sediment depth interval (0-5, 5-10, 20-30 cm). Ctrl: green
squares; N: blue circles; P: red triangles. Data shown as Avg. ± SE.

75
Figure 18. Nitrogen fixation rates under different nutrient treatments (Ctrl, N, P)
for a short period of time of 10 days. Day 1: T-1; Day 10: T-10. Data shown as
Avg. ± SE.

76
Figure 19. Ecological indices of community structure based on nifH OTUs in the
nutrient treatments (Ctrl, N, P) during the wet and dry seasons. Data shown as
Avg. ± SE.

77
Figure 20. Ordination plot of pCCA results showing the temporal variation in the
TRFLP-nifH fingerprints (42.3% explained variance, P=0.004). Sediment depth
intervals and substrate type were used as covariables. Triangles indicate OTUs at
their optima position in the plot. Arrows represent environmental variables and
fertilization treatment (Ctrl, N, P). The directions of the arrows denote the
steepest increase in the variable and the length indicates the strength in explaining
OTUs variation relative to the other variables.

78
Figure 21. Ordination plot of pCCA results showing the spatial variation in the
TRFLP-nifH fingerprints (78.4% explained variance, P=0.002). Temporal
variability between the seasons was used as covariable. Triangles indicate OTUs
at their optima position in the plot (only the best fir OTUs are shown). Arrows
represent environmental variables of the interaction of fertilization treatment
(Ctrl, N, P) with depth intervals (0-5, 5-10, 20-30 cm). The direction of an arrow
denotes the steepest direction in the variable, and the length indicates the strength
in explaining OTUs variation relative to the other variables. The size of the
circles indicates the relative abundance of live roots that range from 3% (smallest
circles) up to 50% (the biggest circles).

79
Figure 22. TRFLP-nifH fingerprints (OTUs) response to the relative abundance
of live roots using generalized additive models. A, N-treatment; B, P-treatment.

80
Figure 23. Phylogenetic tree of nifH gene based on amino acid sequences from
Belize mangrove sediments (in bold). Bootstrap values (≥40%) for 1000
replicates are indicated above the branches and Synechococcus elongatus was
used as outgroup.

99
xii
Figure 24. Accumulation curves for observed and estimated clone-OTUs (A),
and Chao1 index of richness estimated as a function of sample size (B) of
diazotrophs in mangrove sediments from Belize. Curves are averaged over 50
simulations using the computer program EstimateS. Error bars are 95% CIs.

103
Figure 25. Relative abundance of nifH clones sequenced from Belize mangrove
sediments among microbial processes (A) and natural environments (B).

103
Figure 26. Relative abundance (%) of each TRFLP-OTUs detected in mangrove
sediments of Belize. Numbers indicate the TRFLP-OTUs found to correspond
with clone-OTUs.

105
Figure 27. Variability of selected TRFLP-OTUs among nutrient treatments (A,
B), and depth intervals (C, D). Data shown as Avg. ± SE. N≥3.

106
Figure 28. Response curves of seven TRFLP-OTUs to individual environmental
parameters, fitted using generalized additive models (GAM) with a Poisson
distribution using log link function, df = 3.

109
Figure 29. Relative abundance of selected TRFLP-OTUs vs environmental
parameters in mangrove sediments.

110

Figure 30. Response curves of seven TRFLP-OTUs to Axis 1 generated by
pRDA analysis. Data fitted using generalized additive models (GAM) with a
Poisson distribution using log link function, df = 3. pRDA: Axis 1 Eigenvalue =
12.3%, P=0.01, model explained 56% of OTUs variability). Environmental
parameters with a significant effect (p<0.05) on the response of OTUs are shown
in the table (top). D.R.: Dead roots; L.R.: Live roots. OTU-317 not significant to
any parameter.

Figure 32. Conceptual model of principal factors affecting the
variability of bacterial populations in mangrove sediments.

111

119

xiii
ABSTRACT

The ecological interaction of plants and bacteria was studied in a peat-based
sediment subjected to long-term fertilization with nitrogen and phosphorus at Twin Cays,
Belize. The main purpose of this research was to better understand the functional
relationship among microorganisms, mangrove trees and sediment geochemistry in
mangrove sediments. Specifically, the variability of broad bacterial community structure,
the bacterial carbon source utilization patterns, the spatial and temporal dynamics of
nitrogen-fixing populations, and the molecular diversity of nitrogen fixers were studied.
A combination of molecular and chemical techniques combined with statistical tools was
used for the identification of key biological and environmental factors directly controlling
the community structure of microorganisms in mangrove sediments. Results showed that
mangrove trees strongly affect the activity and community composition of N
2
fixers, but
not the whole bacterial community (based on PLFAs) or taxonomic traits (based on
phylogenetic analysis). In most cases roots were inversely related to N
2
fixation rates and
N
2
fixers (community composition), primarily under fertilized conditions. The functional
relationship among microorganisms, plants and sediment geochemistry in Belize showed
that bacteria rely on degrading organic matter from mangroves as a primary source for
carbon, and that mangrove roots do not confer a stable microenvironment that promotes
stability and persistence in microbial populations. Effects of the long-term fertilization
with N or P on bacteria and N
2
fixers revealed that effects depend on the initial conditions
prior to disturbance, and that effects include changes in microbial metabolic pathways
and in community composition patterns of microbial functional groups. Remarkably,
xiv
variability of bacteria and N
2
fixers in mangrove sediments was observed in response to
natural environmental conditions and also to fertilization. A wide range of physiological
adaptations of N
2
fixers (primarily sulfate reducers) to a large heterogeneity of
microenvironments is a strategy ensuring important biogeochemical processes for the
whole ecosystem’s functioning in highly dynamic environments such as mangroves. This
research further explored the complexity in the ecology of marine microorganisms over
different temporal and spatial scales, an important step to better understand the link
between microbial communities and the biogeochemistry of sediments in coastal
environments.

1

Chapter 1

INTRODUCTION

2
1. Introduction

Mangrove forests are among the most productive ecosystems (Alongi et al. 2002)
playing a significant role in nutrient sequestration in coastal zones in tropical and
subtropical regions (Alongi et al. 2004, Holguin et al. 2001). Their high productivity is
based principally on leaf litter production and on standing biomass above- and below-
ground (Chen and Twilley 1999, Bouillon et al. 2008). Mangroves ability to conserve
nutrients as a whole ecosystem relies mainly on the efficiency of nutrient resorption prior
to leaf fall by the trees (Feller et al. 2003) and on the mineralization of organic matter by
active microorganisms, an important component for the productivity of the ecosystem
(Alongi 1996, Holguin et al. 2001). The bioavailability of nutrients in this ecosystem is
one of the most important factors controlling biomass production and nutrient use
efficiencies in trees (Feller et al. 1999). Consequently, eutrophication of mangrove
ecosystems has the potential to alter organic matter mineralization through changes in
mangrove tree physiology, microbial activity and plant-microbial interactions.
Trees and bacteria dominate the biomass and productivity of the mangrove
ecosystem (Alongi 1998) and therefore, the trees' photosynthetic activity linked with
bacterial activity may play a significant role in organic matter remineralization. Plant-
bacterial interactions through the root exudation of recently fixed carbon are an important
source of labile carbon for microorganisms (Butler et al. 2003), enhancing bacterial
activity, sediment mineralization and nutrient availability for plants (Kuzyakov et al.
2000) in terrestrial and coastal environments (for review see Welsh 2000 and Walker et
3
al. 2003). In mangroves, a similar ecological interaction has been suggested where root
exudates can be used by microorganisms present in sediments (Alongi et al. 1993,
Nedwell et al. 1994, Holguin et al. 2001) thus influencing the flow of carbon (Sherman et
al. 1998), nitrogen (Zuberer and Silver 1978) and other microbial processes (Alongi
1994, Alongi et al. 2002). In addition, mangrove roots can also affect microbial processes
by releasing oxygen (Alongi et al. 2001), however some mangrove species may have
limited capacity to oxidize sediments (Thibodeau and Nickerson 1986, Mckee et al. 1998,
Gleason et al. 2003). In general, below-ground processes in mangroves are poorly
understood, as well as the ecological role of plant-bacterial interactions on organic carbon
budgets and nutrient cycling.
The main purpose of this proposed research was to better understand the
functional relationship among microorganisms, mangrove trees and the sediment
geochemistry in mangrove sediments. Specifically, the variability of broad bacterial
community structures, bacterial carbon source utilization patterns, spatial and temporal
dynamics of nitrogen-fixing populations, and the molecular diversity of nitrogen fixers
were studied in a peat-based mangrove sediment subjected to a long-term fertilization
experiment with nitrogen and phosphorus. A combination of molecular and chemical
techniques combined with statistical tools were used for the identification of key
biological and environmental factors directly controlling the community structure of
microorganisms in mangrove sediments. Thus, this research further explored the
complexity in the ecology of marine microorganisms over different temporal and spatial
scales, an important step to better understand the link between microbial communities
and the biogeochemistry of sediments in coastal environments.
4
Global significance of mangroves

Mangroves cover a total area of 181,000 km
2
representing 60-75% of the tropical
and subtropical coastline (Spalding et al. 1997). Even though this ecosystem accounts for
less than 1% of the total forest area on the earth (Ayukai 1998), their global ecological
role is significant mainly because of their biogeochemical dynamics, their high
productivity and their ability to store organic carbon. This ecosystem is considered one of
the most productive systems supporting many adjacent marine ecosystems with
commercially valuable fish, crustaceans and mollusks (e.g. Alongi 2002). In addition,
mangrove forests contribute to the conservation of other coastal ecosystems by stabilizing
the shoreline and maintaining good water quality (Ewel et al. 1998, Twilley 1998).
However, the rapid disappearance of mangrove forests, mainly as a consequence of
human harvest activities (about 2% per year globally, Valiela et al. 2001) and the disposal
of wastewater (Wong et al. 1995, 1997) threaten the future of mangroves worldwide.
Although it is known that nutrient limitation is a common feature of mangrove
forests (Feller 1999) and that high efficiency in recycling and retaining nutrients occurs
in this ecosystem (Nedwell et al. 1994, Kristensen et al. 1995, Feller et al. 2002), there
are still many unanswered questions about the key factors controlling nutrient and carbon
dynamics in mangrove areas.
Mangrove forests can act as a sink for nitrogen (Nedwell et al. 1994, Rivera-
Monroy and Twilley 1996) by accumulating most of the nitrogen in tree biomass
(Wattayakorn et al. 2000), therefore nitrogen availability is strongly controlled by tree
uptake (Alongi et al. 2002). Additionally, N
2
fixation is one of the most important
5
microbial processes in mangrove sediments after sulfate reduction, which is responsible
for most of the organic matter decomposition within the forests (see review Holguin et al.
2001). However many sulfate reducing bacteria are also able to fix nitrogen (Zuberer and
Silver 1978). Nitrogen and carbon cycles may be tightly coupled during organic matter
mineralization (Chen and Twilley 1999) and thus the interaction between bacteria and
mangrove trees may be significant for nutrient conservation within the ecosystem
(Holguin et al. 2001). The coupling of sediment carbon efflux to plant photosynthetic
activity can be a major determinant of the ecosystem carbon budget, but a better
understanding of the role of this ecological interaction on nutrient cycling and carbon
storage is necessary for the conservation of mangrove forests.
Most studies of organic carbon balance in mangroves have indicated that these
forests are also sinks for carbon (e.g. Furukawa et al. 1997, Alongi et al.1999) with a
global carbon sequestration of 38 Tg y
-1
and high carbon storage in surface sediments of
about 10,000 Tg C (Chmura et al. 2003). Globally, mangroves can account for 11% of
the annual organic carbon inputs into the ocean and 15% of the total organic carbon
accumulating in modern marine sediments (Jennerjahn and Ittekkot 2002), therefore,
potentially driving the carbon budget in coastal zones (Alongi 2002). Some evidence
suggests that anthropogenic activities may negatively impact the mangrove ecosystem
with a reduction of carbon burial (Goneea et al. 2004), but the main concern with the
destruction of this ecosystem is the release to the atmosphere of organic carbon stored in
sediments (Gatusso et al. 1998, Alongi et al. 2001). Cebrian (2002) calculated a net loss
of 380 Tg C of stored mangrove biomass for a lost of approximate 35% of the world’s
mangroves.
6
Although these studies on carbon storage inventory in mangrove sediments
showed the main consequences of the destruction of this ecosystem, there is a lack of
understanding on the effects of other disturbances not only on carbon flow, but also on
overall nutrient cycling within the forests. It is widely known that eutrophication is one of
the major threats to coastal systems, but its effects on the mangrove ecosystem are not
well known. The goal of this proposed research is to better understand N
2
fixation and
carbon flow in mangrove sediments by determining the main environmental factors that
control the dynamics of these microbial-mediated processes under natural and nutrient
enriched conditions.

Plant-bacterial interactions

Plants can transfer up to ~30% of all photosynthetically fixed carbon to the
rhizosphere (the soil directly around the roots), but the regulatory processes are not well
understood (Kuzyakov and Domanski 2000, Hobbie et al. 2004). This mechanism of
exudation of carbon by plant roots is an important source of readily available carbon for
microorganisms (Butler et al. 2003) that potentially can affect microbial processes
(Karjalainen et al. 2001), and can enhance bacterial activity and mineralization rates in
soils; consequently, plants can acquire nutrients otherwise unavailable (Kuzyakov et al.
2000).
The acceleration of nutrient turnover due to root exudates is based on their
composition of rapidly utilizable organic compounds like sugars (50-60%), carbonic
7
acids (25-30%) and amino acids (20-35%) (Kuzyakov 2001). The composition of root
exudates varies among plant species, plant age and soils (with different characteristics
that change the permeability of root cells as pH, temperature, oxygen). However, the rate
of root exudation depends mainly on the moisture (Palta et al. 1998) and nutrient
conditions of soils (Saggar et al. 1997). Additionally, root exudation of organic
compounds occurs by active and passive processes in which plants select between
chemical (e.g. increase of carbonic acids in exudates to enhance solubility of phosphorus)
and biological mechanisms (e.g. increase of sugars in exudates to enhance microbial
activity) for altering nutrient availability. Farrar et al. (2003) showed that root exudation
occurs principally through a diffusional passive process accelerated by continual removal
of exudates by microorganisms, however, plants can increase the exudation of carbon
compounds to reach their nutrient demands.
The ecological interaction between plants and microorganisms also occurs
through the release of oxygen by roots, with a significant effect on microbial processes
involved in the carbon and nitrogen cycle (for review see Bodelier 2003). In flooded
environments, heterotrophic bacteria can use molecular oxygen as an electron acceptor,
and root exudates as an electron donor, both for the generation of energy. In coastal
environments, plants also may create ideal conditions for certain microbial groups by
providing an aerobic condition and carbon. For example, N
2
fixers are considered
important for primary production in salt marshes (e.g. Bagwell et al. 1998) and seagrasses
(e.g. Welsh 2000).
In salt marshes, N
2
fixation accounts for 91% of the fixed nitrogen in the
rhizosphere of Spartina maritima supplying most of the nitrogen required by sulfate
8
reducers (Nielsen et al. 2001). In seagrasses, N
2
fixers are also necessary for supplying
31% of the nitrogen in the rhizosphere of Zostera noltii (Nielsen et al. 2001) and
providing 65% of the nitrogen for Zostera capricorni growth (Hansen et al. 2000).
Similarly, high measured N
2
fixing activities are found associated with isolated roots and
in the rhizosphere of other salt marsh (Whiting et al. 1986) and seagrass species (Capone
et al. 1979, Capone and Budin 1982). These enhanced diazotrophic activities are thought
to be caused by root exudation of carbon compounds (like lactate and sucrose) and
oxygen released by roots into the rhizosphere (see Welsh 2000).
Specifically to the mangrove ecosystem, the below-ground compartment of
mangrove forests has not been extensively studied, but it is believed that the ecological
interaction between mangrove trees and microorganisms is important for maintaining
nutrients in this ecosystem (Holguin et al. 2001). The exudation of labile compounds by
roots may be an important energy source to microorganisms in mangrove sediments and
rhizosphere (Alongi et al. 1993, Nedwell et al. 1994, Sherman et al. 1993) enhancing N
2

fixation in Florida (Zubere and Silver 1978) and sulfate reduction in Vietnam (Alongi et
al. 2000b). This interaction between plant roots and microbial groups occurs through the
dissolved organic pool in which organic exudates released from roots partly enrich this
pool (Sherman et al 1998). Bacterial communities consume much of the carbon dissolved
in pore waters using it to process almost all carbon and nutrients in this ecosystem
(Holguin et al. 2001).
In addition, to enrich the labile pool in pore waters for the nourishment of
microorganisms, mangrove trees can also change physicochemical conditions of the
sediments by releasing oxygen through their roots (Sherman et al. 1998). Although some
9
mangrove species apparently lack the ability to oxidize the surrounding sediments
(Thibodeau and Nickerson 1986), it is believed that mangrove trees can significantly alter
the edaphic conditions of the sediments, thereby influencing the growth of certain
microbial groups and the biogeochemistry of the below-ground compartment (Holguin et
al. 2001). Therefore, a tight coupling between mangrove trees and microbial groups may
influence the carbon flow and nitrogen cycling in sediments, but the pathways and factors
controlling the biogeochemistry of carbon and nitrogen in this ecosystem are not well
known.

Nitrogen dynamics in mangroves

Nitrogen limitation in mangroves, caused by a high demand for nitrogen
necessary for sustaining mangroves primary production, is a common feature even
though atmospheric deposition, tidal water exchange and N
2
fixation are nitrogen sources
for this ecosystem (see Chen and Twilley 1999). Thus, nitrogen cycling in mangrove
sediments appears to be regulated not only by rates of nitrogen input and mineralization
but also by high rates of nitrogen uptake by the mangrove trees (Alongi et al. 2002). As a
consequence, mangrove forests are commonly recognized by their low levels of dissolved
and particulate nitrogen compounds (Boto and Wellington 1984). Most of the nitrogen
accumulated in this ecosystem is found in tree biomass due to the high efficiency of
mangrove trees to conserve nutrients (Nedwell et al. 1994, Kristensen et al. 1995). This
10
conservation of nutrients has been suggested to occur principally by the high resorption
of nitrogen prior to leaf fall (Feller et al. 2002).
In mangrove sediments, ammonium concentration in pore waters is the most
important pool of nitrogen for mangrove tree nutrition (Kristensen et al. 1995, Alongi
1998) via bacterial-mediated processes such as ammonification and N
2
fixation (Alongi
et al. 2002). Other microbial processes such as nitrification and denitrification appear to
have minor impact on overall tree nutrition (Kristensen et al. 1995). The large carbon
pool in mangrove sediments that decays slowly due to their refractory nature (Holmer
and Olsen 2002) acts as a long-term source of nutrients and carbon for these microbial
processes (Boto and Robertson 1990). The overall nitrogen cycling in mangrove forests
appears to be more similar to other forests than to other coastal ecosystems, such as salt
marshes (Alongi et al. 2002).
The importance of N
2
fixation in mangroves is poorly understood, as well as its
role as a source of “new” nitrogen for tropical mangrove trees. N
2
fixation in sediments
(Zuberer and Silver 1978), rhizosphere (Holguin et al. 1992), tree trunks (Sheridan 1991),
above-ground roots (Toledo et al. 1995a) and leaf-litter (Woitchik et al. 1997, Pelegri and
Twilley 1998) may provide the necessary nitrogen for the sustenance of mangrove
forests, supplying approximately 40% of the annual nitrogen requirement in south
Australia mangroves (13 g m
-2
y
-1
, van der Valk and Attiwill 1984) and up to 60% in
Florida (11 g m
-2
y
-1
, Zuberer and Silver 1978).
The native microflora of semi-arid and tropical mangroves is not well known.
Burns (2003) found that surface mangrove sediments in Belize host a highly diverse
diazotrophic community. Sengupta and Chaudhuri (1991) isolated the N
2
fixing bacteria
11
Azospirillum, Azotobacter, Rhizobium, Clostridium and Klebsiella from sediments,
rhizosphere and root surfaces of various mangrove species. Additionally, Holguin et al.
(1992) isolated from the rhizosphere of different mangrove species other N
2
fixers
(Vibrio campbelli, Listonella anguillarum, V. aestuarianus, and Phyllobacterium sp.).
The ecological role of N
2
fixers, besides being a source of nitrogen for mangroves, has
not been widely studied; although Ravikumar et al. (2004) found that three species of
Azobacter are important producers of hormones that promotes the growth of mangrove
seedlings. It is known that the distribution of N
2
fixers is affected by the oxidation state
of the sediments, however it is not well understood if mangrove trees constrain the
diversity and distribution of this microbial group.
In terrestrial environments, it is well known that certain trees use nitrogen from
their symbionts, while photosynthesis supplies organic carbon to the nodules (Tissue et
al. 1997). Nitrogen is an essential nutrient for trees because it is a primary component of
ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco), the enzyme that catalyzes
photosynthetic reduction of CO
2
to carbohydrates, while carbon in the root nodules is
important as a source of energy and reducing power to fix N
2
(Tissue et al. 1997). An
association between microbes and plants can also occur without developing nodules in
the roots. For example, Whiting et al. (1986) showed a tight coupling of root-associated
nitrogen fixers and plant photosynthesis in the salt marsh Spartina alterniflora. Evidence
for this type of association was also found in Avicennia germinans (black mangrove)
inoculated with the cyanobacteria Microcoleus (Toledo et al. 1995b, Bashan et al. 1998).
Mangrove trees may depend on bacteria for nutrient recycling, and at the same time
bacteria may benefit from an association with mangrove roots. Because N
2
fixation is a
12
high energy demanding process, it is likely to be limited by the availability of organic
substrates, which are refractory, even in sediments with high organic content.
In mangrove forests, available organic carbon for N
2
fixation can be derived from
the decomposition of leaves and roots by nondiazotrophic microflora and from
photosynthetic derived carbon exuded by mangrove roots (Zuberer and Silver 1978,
Alongi et al. 1993 and Nedwell et al. 1994) as occurs in terrestrial and other coastal
environments (e.g. seagrasses; Piceno and Lovell 2000b, Welsh 2000). Root exudation of
fixed carbon can be important in marine systems because the availability of labile carbon
sources is generally thought to be a major constraint for N
2
fixation in most marine
sediments (Capone 1988, Tibbles et al. 1994). Yet, little is still known about the ecology
of N
2
fixers and the factors that control their activity in the mangrove ecosystem.

Carbon cycling in mangrove sediments

Mangrove forests in estuarine systems are considered to release negligible
amounts of greenhouse gases relative to their high primary production (Chmura et al.
2003). While there is few data available for CO
2
fluxes in mangrove areas, it has been
determined that over-saturation of CO
2
with respect to atmospheric equilibrium in surface
surrounding waters is a general feature of mangroves with a global emission of 50 x 10
6

ton g C y
-1
, although the entire ecosystem (sediments, water, vegetation) is a sink for
atmospheric CO
2
(Borges et al. 2003). Based on measurements of δ
13
C
DIC
in surface
waters, Bouillon et al. (2003a) showed that mineralization of organic matter and the
13
subsequent efflux as CO
2
may represent a major pathway for mangrove carbon depending
on the extent in which organic matter is retained in the system (Bouillon et al. 2003b).
Mangrove organic matter has a small role as a carbon source for phytoplankton or for
benthic invertebrates (Bouillon et al. 2000, 2002).
Specifically in mangrove forests, carbon is mainly accumulated in above- and
below-ground biomass and stored in sediments. Sediments are important carbon sinks
(Kristensen et al. 1995, Alongi et al. 1999, 2000a,b) retaining 60% of the total input of
organic carbon even with carbon mineralization rates up to 197 g C m
-2
yr
-1
(Alongi et al.
2001). Although some specific studies have covered carbon fixation and storage in
mangroves and their relevance to global climate change (Alongi et al. 1998, Wolanski et
al. 1998), key processes controlling carbon dynamics in tropical and subtropical areas
with different mangrove forest types is far from being well understood.
Traditionally, the high productivity of mangroves has been estimated based on
litter fall measurements (e.g. Seelinger and Kjerfve 2001) overestimating its role in
sediment formation (organic matter accumulation) and in sustaining adjacent aquatic
secondary production (e.g. Dehairs et al. 2000, Bouillon et al. 2000, 2002). High
production and slow decomposition of roots (40% remained after 1.5 years) in the below-
ground compartment of mangrove sediments have demonstrated the importance of roots
in sediment formation and carbon sequestration in this ecosystem (Middleton and Mckee
2001). More recent, model sensitivity analysis has also showed that root production has a
more significant effect on sediment composition and nutrient biogeochemistry of this
ecosystem than litter fall (Chen and Twilley 1999). Roots may have a significant role in
sediment microbial processes by affecting sediment conditions (through the release of
14
oxygen, uptake of cations and root respiration) and by acting as a source of labile organic
carbon (from dead roots and live root exudates) (Alongi et al. 2001).
The coupling of photosynthetic activity to soil carbon efflux can be a major
determinant of the ecosystem carbon budget (Högberg et al. 2001) through plant-
microbial interactions in below-ground terrestrial (for review see Walker et al. 2003) and
coastal environments such as seagrasses (e.g. Welsh 2000) and mangroves. Alongi et al.
(2001, 2002) suggested that root-bacteria interactions might play an important role in
mineralization processes in mangrove sediments. Root exudation is an important source
of readily available carbon for microbes, on which in turn plants rely on the supply of
available nutrients from decomposition (Butler et al. 2003) and/or microbial processes
(e.g. N
2
fixation; Zuberer and Silver 1978, Whiting et al. 1986). Many uncertainties still
remain, principally concerning the role of the roots in the biogeochemistry of mangrove
sediments and the contribution to long term below-ground carbon storage.

Research questions

Overall this research focused on (1) discerning spatial and/or temporal patterns of
microorganisms; (2) determining what characteristics of the mangrove ecosystem
primarily affect the variability of microorganisms; and (3) establishing how disturbances
long-term fertilization with nitrogen and phosphorus alter the distribution and metabolic
pathways of microorganisms in highly dynamic environments like mangrove forests.
15
Specifically, two main hypotheses were tested that looked at the ecological interaction
between plants and bacterial communities, and between plants and N
2
fixers:

Is mangrove-derived carbon an important source of labile carbon for bacteria?
It has been hypothesized that because of the refractory nature of the mangrove
organic matter (leaf litter) in surface sediments, microorganisms rely on mangrove trees
as a carbon source only in pristine conditions and when no other carbon source exists.
Deeper in the sediment, as well as in sediments enriched with nutrients, roots might play
an important role as the main source of carbon for microbes.
Traditionally carbon budgets have proposed that mangrove litter is the main carbon
source for in situ mineralization; however, bacterial communities were found to use other
carbon sources (e.g. microphytobenthos, seagrass-derived material, Bouillon et al.
2004a). Also, under disturbed conditions such as nutrient enrichment, bacteria may shift
to allochthonous carbon sources as happened in Mediterranean seagrass sediments
(Holmer et al. 2004). Additionally, due to the refractory nature of mangrove organic
matter, bacteria may rely on root exudates for labile organic compounds (Alongi et al.
2001, Alongi 2002). The analysis of
13
C-PLFA (phospholipids-derived fatty acids)
enables the study of the origin of the carbon supporting the living microbial communities
by comparing stable isotope signatures of specific biomarkers with the potential
substrates found in mangrove sediments. Specific bacteria biomarkers (PLFA) are widely
used to study broad microbial communities in sediments (e.g. Jones et al. 2003) due to
their properties like bacterial specificity and rapid turn over after cell death. The analysis
of
13
C-PLFA is considered a powerful tool to identify the main carbon sources for
16
microbes across different environments (Maier-Agustein 2002; Boschker and Middelburg
2002). However, sediment
13
C-PLFA measurements in coastal tropical ecosystems are
scarce, especially in mangroves. At present only two studies in riverine mangrove forests
have used this technique (Bouillon et al. 2004a,b; 2006) and both have shown that
mineralization cannot be directly incorporated in ecosystem carbon budgets without an
estimation of the contribution of various carbon sources for microorganisms.

To what extent are diazotrophs constrained by mangrove roots under different nutrient
conditions?
It has been hypothesized that because mangrove peat-based sediments are
composed mainly of refractory organic carbon, diazotrophs might rely on root exudation
for carbon. In addition, O
2
released from roots might constraint the distribution of
diazotrophs in the sediments. Nitrogen nutrification of sediments might impose a
competitive disadvantage for diazotrophs, affecting their diversity and activity.
N
2
fixation in sediments and roots has been recognized as an important microbial
process for the mangrove ecosystem (Zuberer and Silver 1978, van der Valk and Attiwill
1984, Alongi et. al. 2002). Specifically in Twin Cays, Burns (2003) found that surface
mangrove sediments host a highly diverse diazotrophic community, and Joye and Lee
(2004) found that N
2
fixation is an important source of nitrogen for this ecosystem.
However, little is yet known about the ecology of this microbial group in mangrove
sediments and the intra-forest factors that regulate its diversity and activity.
Nitrogen fixation rates were measured by the acetylene reduction technique
(Capone and Montoya 2001) in samples from sediment cores collected from different
17
depth intervals. Samples were measured at the University of Southern California on an
FID Shimadzu GC-9A gas chromatograph. To characterize the community composition
of the N-fixing microbial populations DNA was extracted from the sediment samples and
Terminal restriction length polymorphism (TRFLP) of PCR amplified nifH gene was
used. Nitrogen-fixing community composition patterns in response to the environmental
gradients were analyzed using a unimodal constrained model (CCA, Canonical
Correspondence Analysis) (Ter Braak and Prentice 1988). This combination of molecular
techniques and statistical analyses enable the ecological analysis of this microbial group
under different natural and nutrient induced condition in mangrove sediments.

18

Chapter 2

STUDY AREA:
MANGROVE FORESTS IN TWIN CAYS - BELIZE

19

2. Study area: Mangrove forests in Twin Cays

This study was conducted at Twin Cays
(16
o
50’N, 88
o
06’W), a 92 ha archipelago
located 12 km off-shore from the coast of
Belize (Central America) situated along the
MesoAmerican Barrier Reef system (Rützler
and Feller 1996) (Figure 1). Twin Cays is the
primary field site for mangrove research of the
Smithsonian Institution in Belize, which has
served as a model system for studying nutrient
dynamics in oligotrophic settings (e.g.,
McKee et al. 2002; Feller et al. 2003; Joye
and Lee 2004; Lee and Joye 2006).
Twin Cays is composed of two islands
separated by a 0.5 - 2.0 m deep channel and
surrounded by a shallow (~1 m) sand flat
vegetated by the seagrasses Thalassia
testudinum and Halodule wrightii. These
oceanic islands have limited terrestrial
influence and are constantly flushed by ocean
water (mixed semidiurnal tides with mean
Figure 1. Twin Cays is located 12 km
off-shore from the coast of Belize (Top)
and is composed of two islands that are
surrounded by seagrasses (bottom, aerial
photograph). Star: Boa Flats site.
20
amplitude of 15 cm, Kjerfve et al. 1982). There are two seasons, a wet season from July
to October with an average rainfall of 218 cm·yr
-1
and a dry season the rest of the year
(Rützler and Ferraris 1982).
The substrate is principally peat formed from the fine roots of Rhizophora mangle
(Figure 2) and calcareous algae (Mckee 1995) with a patchy covering of detritus mainly
from mangrove trees. These islands have adjusted to changes in sea level in the
Caribbean region mainly by accumulation of refractory mangrove roots as seen in
radiocarbon-dated cores which indicated a tight relation between peat sediment formation
and sea-level rise over the Holocene (Mckee et al. 2007). Mangrove communities in these
islands were established on a Pleistocene limestone substrate about 8,000 years ago,
forming a peat layer of almost 9 m thick (Macintyre et al. 2004); therefore, these are sites
of high organic carbon accumulation.

Vegetation is dominated principally by R. mangle (red mangrove), but Avicennia
germinans (black mangrove) and Laguncularia racemosa (white mangrove) are also
A B
Figure 2. Peat-based sediments in Twin Cays (A) are composed
mainly by mangrove roots (B).

21
present. From sea to landward, the forest is characterized by a tree-height gradient with
tall red mangrove trees in the seaward-most zone (5-6m tall), continuing to a transition
zone (2-4m) followed by an interior zone with dwarf red mangrove trees (≤ 1.5m tall
trees) and lagoons or shallow ponds of seawater (Figure 3). This tree-height gradient
follows stand litter production: fringe trees have more litter production (700 g m
2
year
-1
)
than transition (450 g m
2
year
-1
) and interior dwarf trees (280 g m
2
year
-1
) (Koltes et al
1998). The fringe zone is flooded and drained twice a day; the transition zone is flooded
only during spring tides and storms, and the interior zone is almost continually flooded.
Dwarf trees in the interior zone present the largest covertures of the mangrove ecosystem
(about 60%) followed by the fringe zone (~32%) (Rodriguez and Feller 2004).
A fertilization experiment was established in 1997 at each of the mangrove zones
and maintained by I.C. Feller and coworkers of the Smithsonian Institution (see details in
A B
C
Figure 3. Mangrove zones in Twin Cays, Boa Flats (A). Tree-height
gradient with tall trees in the fringe zone (B) followed by an interior zone
with dwarf trees and interior ponds (C).
22
Mckee et al 2007). In the mangrove forests, transects were established from the shoreline
to the island interior (distance = 30–40 m) and permanent walkways were installed in
2000 to minimize sediment disturbance. Each transect was randomly assigned to a
nutrient treatment to reduce cross-contamination among treatments due to tidal
movement across the zones. At six-month intervals, individual trees were fertilized with
nitrogen (urea) or phosphorus (P
2
O
5
) using dialysis tubing. At each time, 150 g of
fertilizer was added in two holes cored to ~ 30 cm depth into the sediment on opposing
sides of the tree and sealed with peat. A control treatment includes coring and plugging
with no fertilizer. Mangrove trees in this area have shown that mangrove shoot growth
varied spatially within the forests and was stimulated by N in the fringe zone and P in the
interior (McKee et al. 2002; Feller et al. 2003). Only addition of P caused changes in
physiological processes of dwarf trees (hydraulic conductivity, stomatal conductance and
photosynthetic assimilation rate) to levels similar to the fringe trees (Lovelock et al.
2006). In the belowground compartment (for details see McKee et al 2007), only P
additions stimulate fine root production in all zones and coarse root accumulation also in
all zones except in the fringe. High root mortality was found in all fertilized plots except
in the transition zone. A stronger effect of N on root mortality is found in dwarf trees and
P on fringe trees. Root decomposition rates are the same along the different zones and
nutrient treatments.
The specific study area for this research is located at Boa Flats fertilization site
(Figure 1). This mangrove area shows the typical height pattern from the fringe (tall
trees) to the interior (dwarf trees) of the island (Figure 3). An extensive shallow lagoon
separates the dwarf mangrove trees studied from tall interior trees. A dense canopy that
23
reduces direct sun exposure of the sediments characterizes the fringe zone, where surface
sediments are commonly found covered by seagrasses leaves from surrounding areas and
abundant invertebrates and algae on the subtidal and intertidal areas of the prop roots of
R. mangle. The dwarf trees in the interior zone are sparsely distributed with a limited
develop canopy and sediments are flooded over longer periods and extensively covered
by microbial mats (for details see Joye and Lee 2004). Although, during the dry season
sediments can be exposed for long periods.

Figure 4. Boa Flats sediment
surfaces. In the fringe zone,
sediment can be covered by
mangrove (A) or a mixture of
mangrove and seagrass (B)
leaf litter. In the interior zone,
sediment is partially covered
by microbial mats and some
mangrove leaf litter (C).
A
B
C
24

Chapter 3

BACTERIA COMMUNITY AND CARBON SOURCE UTILIZATION
PATTERNS IN MANGROVE SEDIMENTS:
EFFECTS OF LONG-TERM EXPERIMENTAL NUTRIENT ENRICHMENT

25
3. Bacterial Community and Carbon Source Utilization Patterns in Mangrove
Sediments: Effects of Long-Term Experimental Nutrient Enrichment

Abstract

Broad bacterial community structure and carbon source utilization patterns were
studied in peat-based mangrove sediments subjected to a long-term fertilization
experiment with nitrogen and phosphorus. Phospholipid-derived fatty acids and their
stable carbon isotopic composition (δ
13
C-PLFA) of specific bacterial biomarkers were
determined and compared to geochemical parameters analyzed from pore water and
sediments. δ
13
C-PLFAs were compared to potential carbon sources within the mangrove
sediments (i.e. seagrass leaves, mangrove leaf litter, dead roots, and living roots) to
examine plant-bacterial interactions under natural and induced nutrient conditions.
Bacterial PLFA community composition was constrained by reduced conditions in the
sediments, whereas bacterial biomass was determined primarily P- or N-limited
depending on the location within the mangrove forests. Comparisons of δ
13
C signatures
of bacterial PLFA biomarkers with potential carbon sources suggested that bacteria in
this mangrove system are primarily using in situ mangrove carbon sources for
mineralization rather than exogenous seagrass carbon. Bacterial groups based on PLFA
distributions had different responses to fertilization conditions at both fringing and
interior mangrove zones, with an isotopic shift indicating an increase in anaerobic
respiration where fertilization increased nutrient limitation conditions for plants and
26
bacteria. Therefore, fertilization in addition to being able to overcome nutrient limitations
has the potential to strongly affect carbon utilization patterns in bacteria. Overall, results
showed that, while the enhancement of nutrient limitation has a minor effect on bacterial
community structures, it can strongly influence metabolic processes of bacterial groups.
This study demonstrates the importance of functional relationship studies among trophic
groups for determining the effect of disturbance on natural marine environments.

Introduction

Mangroves are predominant in tropical and subtropical areas with an important
ecological role in coastal oceans. Their high primary production of organic matter (~149
mol C m
2
year
-1
, Kristensen et al. 2008) sustains a large diversity of organisms (e.g.
Sheaves and Molony 2000, Laegdsgaard and Johnson, 2001) and potentially drives the
carbon budgets in coastal areas (Alongi 2002, Dittmar et al. 2006). The overall
functioning of mangroves is largely dependent on sedimentary microorganisms through
the decomposition of organic matter and subsequent nutrient cycling (Duarte and Cebrian
1996, Holguin et al. 2001). Generally, carbon budgets in this ecosystem include
mangrove litter as the main carbon substrate for in situ mineralization, overestimating the
role of litter in sediment formation and in sustaining adjacent aquatic secondary
production (Dehairs et al. 2000, Bouillon et al. 2000, 2002). Mangrove sedimentary
bacterial communities may use imported (phytoplankton, seagrasses) and local
(microphytobenthos) carbon sources as well for respiration (Bouillon et al. 2004b). In
addition, other important local sources of organic carbon to the whole ecosystem have
27
also been recognized to potentially control the nutrient cycling in some forests, such as
microbial mats (Wooller et al. 2003) and root rhizodeposits (Nedwell et al. 1994, Alongi
et al. 2004). Because the role of mangrove roots as an important carbon source for
bacteria (e.g. refractory dead roots and labile rhizodeposits; Alongi et al. 2001), sediment
formation and carbon sequestration (Middleton and Mckee 2001, Kristensen et al 2008)
have been less studied, many uncertainties still remain on root-bacterial interactions
contributing to long-term sediment carbon storage. The decline in mangrove areas
worldwide (Valiela et al. 2001) and the increase of altered forests by human activities
such as nutrient inputs (Galloway et al. 2004, Vitousek et al. 1997) over the last decades,
may have significantly impacted the ecological processes related to carbon cycling in this
ecosystem. Therefore, a better understanding of the effects of nutrient enrichment in
ecological processes and the contribution of potential local and imported carbon sources
for mangrove microorganisms under different nutrient conditions is important to study
for improving carbon budgets in this ecosystem.
Analysis of specific microbial biomarkers such as phospholipid-derived fatty
acids (PLFA) enables the study of active broad microbial communities and the microbial
main carbon sources being utilized, owing to their specific properties such as microbial
group specificity (e.g. Zelles 1999) and their rapid turnover after cell death (White et al.
1979). In addition, PLFA have been shown to record variability in microbial communities
due to the environmental changes that have affected their main carbon source (Kaur et al.
2005, Billings and Ziegler 2005, Ramsey et al. 2006) and directly to perturbations such as
eutrophication (Pinturier-Geiss et al. 2002). Lipids in microorganisms are depleted in
13
C
relative to total biomass and substrate (Jones et al. 2003, Bouillon and Boschker 2006),
28
but the extent of this depletion depends on the carbon source being assimilated and the
environment during synthesis (Teece et al. 1999). Although this difference between
specific PLFAs and carbon sources limits their use quantitatively, the analysis of δ
13
C-
PLFA is still considered a powerful tool to identify the main carbon sources for microbes
across different environments (Meier-Augenstein 2002). In mangrove ecosystems, δ
13
C-
PLFA was important to determine that bacteria use carbon from various origins (Bouillon
et al. 2004a, Bouillon et al. 2004b) depending primarily on the availability of sedimentary
total organic carbon (TOC) (Bouillon and Boschker 2006). In this particular environment,
selectivity for labile carbon sources (potentially root exudates and macrophytobenthos)
only occurred in environments with TOC lower than 1%; whereas in areas with TOC
higher than 1% bacteria use the more available carbon source. Therefore, the δ
13
C-PLFA
from mangrove sediments have been proven to provide carbon source identification
important for establishing the fate of mangrove production in the overall carbon budget
of mangrove ecosystems.
Our study investigates the physical, geochemical, and biological factors of
sediments within a water-flooding gradient under a long-term fertilization experiment (8
years) in Belizean mangrove forests. We compared the information derived from bulk
analysis (C/N,
13
C and
15
N signatures) and fatty acid biomarker concentrations and
compound specific δ
13
C-PLFA to determine bacterial abundance and carbon source
utilization patterns. Our goal was to gain an improved understanding of bacterial
community dynamics within forests under natural and nutrient-enriched conditions.
Results from this study provide an initial understanding of the impacts of nutrient
loadings on mangrove sediment geochemistry and plant-bacterial interactions. Overall,
29
this study expands our knowledge of microbial dynamics in naturally complex
environments and the potential consequences of nutrient enrichment on bacterial
community structures and processes.

Methods

Field Collection

Samples were collected in mangrove forests located in Twin Cays, Belize, at the
experimental site known as Boa Flats (for a detailed description of the study area see
Chapter 2). In order to compare the broad bacterial community structures under natural
and long-term fertilization conditions, samples of pore water chemistry, bulk sediment
characterization, and δ
13
C-PLFA were taken from different depth intervals: 0-1 or 0-5 for
pore water measurements, 5-10, 20-30 cm. These samples were collected within two
mangrove zones (fringe and interior areas) and within three experimental sites in these
zones (Control, Ctrl; Phosphorus, P; and Nitrogen, N). Pore water samples were collected
using “sippers” constructed from 0.5 cm external diameter polycarbonate tubing sealed at
one end with two ~1 mm perforations on the side, which were inserted into the peat
sediment at three depth intervals (N=3 per depth). The sipper was connected to a filter
holder with pre-combusted glass-fiber filters (Whatman GF/F 0.7µm, 4.7cm diameter)
through a two-way stopcock valve to a luer-lock syringe, which serves as a reservoir for
30
the sample. Samples were filtered for a second time with an Acrodisc syringe filter (Pall
0.45 µm HT Tuffryn
®
membrane) and divided into subsamples for each pore-water
measurement. All glass vials used were acid-washed, rinsed with ultrapure deoinized
water, and combusted at 500ºC for 5 hours. Sediment cores were taken with a Russian
Peat corer, which is designed to avoid vertical compaction of samples, within ~0.5 m of
distance from the main trunk of a designated experimental mangrove tree, and divided by
sediment depth intervals and sample types.

Pore-water analysis

Temperature and pH were determined immediately after the collection of the pore
water samples (ORION with pH and ATC electrode; calibrated with certified standards,
reproducibility of ± 0.02 pH units and ± 1.0
o
C). Salinity was measured with a
refractometer (Fisher Scientific, calibrated with deionized water, with an accuracy of ±
1.0 ppt). Samples for determining the hydrogen sulfide (H
2
S) content (volume 0.1 ml)
were stored in glass vials containing 0.04 ml of 0.05 M zinc acetate and analyzed
following the method of Cline (1969). For ammonium (NH
4
+
) analysis, samples (volume
0.5 ml) were stored in vials containing 1 ml of phenol reagent and analyzed as in
Solorzano (1969). Dissolved inorganic phosphate (PO
4
3-
) was measured colorimetrically
using the method of Strickland and Parsons (1972). A spectrophotometer (Shimadzu UV-
1700) was used for pore-water analysis at the University of Southern California.

31

Bulk Carbon and nitrogen stable isotope analysis

Percent total organic carbon (%TOC), %N, δ
13
C, and δ
15
N were determined to
characterize the solid phase nutrient components of bulk sediment, leaf litter, and live and
dead roots at each depth interval. These subsamples were dried for 48 hours at 60
o
C and
analyzed at the Geophysical Laboratory Carnegie Institution of Washington. %TOC and
δ
13
C of the bulk sediment were analyzed on samples acidified to eliminate carbonates
(Harris et al 2001). Triplicates (600 to 1500µg) of each subsample were analyzed using
continuous-flow, stable isotope ratio mass spectrometry (Thermofisher Delta V Plus).
Nitrogen stable isotope ratios (δ
15
N) were expressed relative to air (δ
15
N = 0.0‰), and
carbon isotope ratios (δ
13
C) were expressed relative to Pee Dee Belemnite (δ
13
C = 0.0‰).
Acetanalide (C
8
H
9
NO) was analyzed as a check on the precision of the isotopic ratios
and the elemental compositions of carbon and nitrogen. Precision for δ
15
N was ± 0.28‰
and for δ
13
C was ± 0.21‰.
The relative abundances (%) of leaf litter and live and dead roots were determined
by washing, with freshwater, the sediment samples collected at each depth interval with
different sized sieves (4 mm, 2mm). Each of these sediment components was picked
manually, oven dried for 48 hours at 60
o
C, and weighed. Sediment components were
compared to total sediment weights for each sample to calculate relative abundance (%).

32
Identification and δ
13
C of sediment PLFAs

Frozen sediment samples were transported to the University of Arkansas,
lyophilized and extracted using the modified Bligh–Dyer method (Bligh and Dyer 1959).
Samples were fractionated into different lipid classes with a silic acid solid phase on a
vacuum manifold system to facilitate elution (White and Ringelberg 1998), where
phospholipids were separated from neutral lipids and glycolipids, and collected.
Phospholipids were saponified using KOH (mild alkaline hydrolysis), which were then
converted to their corresponding fatty acid methyl-esters (FAME) using BF
3
in methanol
(Dobbs and Findlay 1993).
FAMEs were quantified using a gas chromatograph with a flame ionization detector
(GC-FID; Agilent 6890), which was equipped with a 70% cyanopropyl polysilphenylene-
siloxane capillary column (SGE BPX-70; 50 m length, 0.22 mm i.d., and 0.25 mm film
thickness). Identification of each FAME was conducted by gas chromatography-mass
spectrometry (GC-MS) interfaced with a mass selective detector (Agilent 5973) based on
retention time and mass spectra of known standards, including individual FAMEs (Sigma
Aldrich Chemical Company, St Louis, MO, USA) and mixtures (Bacterial FAMEs and
37 component FAME standards, Supelco Co., Rockford, IL, USA) in addition to a direct
comparison of mass spectra to a NIST database.
General microbial groups associated to the PLFA detected were identified
according to previous studies (Guezennec and Fiala-Medioni 1996, White et al. 1996,
Ringelberg et al. 1997, Zelles et al. 1992, Zelles 1999, Fierer et al. 2003, Yeudokimov et
al. 2008). Total bacterial PLFA concentration (µg PLFA g dry weight
-1
) was based only
33
on known bacterial biomarkers (Table 1) to exclude PLFAs from plants. Fatty acid
standard nomenclature was used providing the number of C atoms followed by a colon
and the number of double bonds. The position of the first double bond from the aliphatic
end of the molecule is indicated by a “ω”. The prefixes “a” and “i” referred to anti- and
iso-branched fatty acids with a methyl group of 1 or 2 carbons from the aliphatic end,
respectively. Other positions of the methyl group were indicated by a number before the
prefix “Me”. Cyclopropane groups (cy) were designated by the number of C atoms from
the aliphatic end of the molecule.
The carbon isotopic composition of PLFA (δ
13
C-PLFA) was determined by GC–
isotope ratio mass spectrometry (GC-IRMS). The isotopic composition of each PLFA
was determined using a GC (HP6890) coupled to a stable isotope ratio mass spectrometer
(Finnegan Delta XP) via a combustion interface (Finnegan GC/CIII) at the University of
Arkansas Stable Isotope Facility. A correction for the addition of the methyl carbon from
BF
3
/methanol derivatization was calculated for each fatty acid by mass balance from the
analysis of free and methylated internal standard nonadecanoic acid (Abrajano et al.,
1994). The same capillary column, analysis conditions, and standards used for the GC–
MS and GC– FID were also used for the GC–IRMS. Stable isotopic ratios were measured
relative to high purity and calibrated reference gas standards, and were expressed relative
to PDB. In order to relate stable carbon isotope signatures of PLFAs to potential carbon
sources supporting bacterial communities, the dominant bacterial biomarkers were used
(i15:0 and a15:0) in which fractionation between substrate and PLFA is known to be
−3.7±2.1‰, an empirical value for fractionation in natural environments with a diverse
community of bacteria using complex organic substrates (Bouillon and Boschker 2006).
34
Statistical analysis

To determine the magnitude of the effect of long-term fertilization on carbon
stable isotopic signatures in sediment and sediment components, ∆
13
C was calculated as
the difference in δ
13
C between the nutrient treatment (N or P) and the control treatment.
Positive values indicated increase in δ
13
C in the nutrient treatment relative to the Ctrl-
treatment. All values were shown as the average ± SE (where N≈3).
All pore water, sediment parameters and PLFA data for each mangrove zone were
analyzed with an ANOVA with nutrient treatment, depth, and depth X nutrient treatment
interaction as factors (JMP Software, SAS Institute Inc., Cary, NC). To evaluate solely
the natural nutrient conditions, the Ctrl-treatment was analyzed with a two-way ANOVA
with zones and depths as factors. When significant differences were found, Tukey’s test
was used to compare treatment means. The level of statistical significance was set to P-
level <0.05. All data was tested to fulfill normality, and equal variance assumptions and
transformations were carried out when necessary. The T-test was performed to test for the
significance of linear correlations and differences between control (Ctrl) and nitrogen (N)
and phosphorus (P) treatments. Bacterial PLFA community patterns in response to
environmental gradients were analyzed using a linear constrained model (Redundancy
Analysis) due to the narrow range of environmental variation (Ter Braak and Prentice
1988). Redundancy Analysis and ordination plots were conducted with CANOCO
software (Microcomputer Power, Ithaca, N.Y.). Percent abundance of bacterial PLFAs
were transformed (log (x+1)) and environmental data was centered and standardized.
35
Results

Mangrove pore-water chemistry

High variability in pore water chemistry was found between mangrove zones and
depths in the control (Ctrl) treatment (Figure 5), whereas slight but significant differences
in pH and temperature were found only between the mangrove zones (p < 0.001). In
contrast, PO
4
3-
concentration changed substantially, with significant higher values in the
fringe zone (15.3 ± 3.6 µM) compared to the interior zone (2.0 ± 0.5 µM, p < 0.001).
Only NH
4
+
and H
2
S presented a trend with depth, with no differences between the
mangrove zones (Figure 5).
Long-term N or P additions into mangrove sediments changed pore-water
chemistry and temperature compared to the Ctrl-treatment in both mangrove zones (Table
1). In the fringe zone, salinity was lower in the P-treatment (~39 ± 0.4 ppt; p < 0.01)
compared to Ctrl- and N-treatment (both ~42 ± 0.7 ppt). Also in the fringe zone, PO
4
3-

was between 5 and 10 times higher in the P-treatment (Table 1; p < 0.001) relative to the
other treatments. NH
4
+
concentration was not significantly different in the N-treatment in
the fringe zone (P = 0.449; Table 1); however, in the interior zone it was significantly
greater (p < 0.01). Also in the interior zone, pH was higher in the N-treatment (pH: 6.5 ±
0.1; p < 0.01) and PO
4
3-
was higher in the P-treatment (112.1 ± 25.5 µM; p < 0.001),
showing an increase in these parameters due to N and P additions, respectively.
Interestingly, the pore-water temperature in this zone significantly decreased
36

Fringe Interior
Ctrl N P Ctrl N P
Pore Water

pH 6.55
(0.04)
6.53
(0.06)
6.47
(0.04)
6.30
(0.09)
6.53
(0.05)
6.25
(0.03)
Temperature
(°C)
31.37
(0.11)
31.15
(0.09)
31.59
(0.15)
34.39
(0.32)
32.92
(0.30)
31.94
(0.26)
Salinity (ppt) 42.12
(0.83)
42.68
(0.63)
39.83
(0.44)
40.93
(0.57)
40.82
(0.54)
41.68
(0.57)
NH
4
+
(µM) 56.72
(15.66)
108.39
(54.45)
67.95
(22.46)
61.28
(6.97)
587.43
(224.55)
61.48
(7.69)
PO
4
3-
(µM) 15.26
(3.61)
8.02
(2.10)
75.32
(26.61)
2.00
(0.49)
1.92
(0.68)
112.10
(25.47)
H
2
S (mM) 0.52
(0.12)
0.71
(0.34)
0.42
(0.06)
0.74
(0.13)
0.97
(0.32)
0.40
(0.04)
Sediment

%TOC 39.5
(1.1)
39.4
(0.8)
42.1
(2.5)
38.5
(0.6)
37.8
(0.9)
38.9
(0.9)
C/N (atoms) 59.1
(10.4)
61.0
(7.4)
28.5
(1.4)
34.9
(1.4)
30.9
(1.9)
26.2
(0.9)
δ
13
C (‰)
-26.3
(0.1)
-26.5
(0.1)
-26.4
(0.1)

-25.8
(0.1)
-25.9
(0.2)
-26.1
(0.1)
δ
15
N (‰)
-0.7
(0.2)
-0.6
(0.1)
-0.5
(0.1)

-0.6
(0.1)
-0.6
(0.1)
-0.8
(0.1)
Leaf litter (%)
2.2
(1.6)
3.6
(1.4)
0.2
(0.1)

3.3
(1.9)
3.1
(1.8)
1.3
(0.5)
Live roots (%)
51.5
(21.1)
16.5
(13.4)
17.8
(0.1)

15.0
(3.5)
11.9
(3.0)
29.0
(15.8)
Dead roots (%)
15.0
(2.9)
14.8
(2.3)
15.9
(3.9)

18.2
(7.1)
22.9
(2.5)
10.6
(1.8)

Table 1. Pore-water and sediment parameters for each nutrient treatment (Control: Ctrl,
Nitrogen: N, Phosphorus: P) in the Fringe and Interior mangrove zones. The data are: Avg. ±
SE.
37

Figure 5. Pore-water and total sediment parameters (Avg. ± SE) in the fringe and interior
mangrove zones for each depth interval (cm) at the Ctrl-treatment (natural conditions). * Missing
data.

pH Temperature (
o
C) Salinity (ppt)
H
2
S (mM) PO
4
3-
(µM) NH
4
+
(µM)
C/N (atoms) δ
13
C (‰) δ
15
N (‰)

*
38

Figure 6. Sediment components (Avg. ± SE) in the fringe and interior mangrove zones for each
depth interval (cm) at the Ctrl-treatment (natural conditions). Leaf litter at 20-30 cm depth in the
interior zone was not found in the cores collected.
Live Roots Dead Roots Leaf litter
39
from Ctrl- to N- and P-treatment (34 ºC, 33 ºC and 32 ºC, respectively; p < 0.0001) due to
a greater canopy development in the fertilized trees.

Mangrove peat solid phase chemistry

Under natural conditions (Ctrl-treatment), C/N in the fringe zone was similar (for live
roots and dead roots; p < 0.05) or higher (for leaf litter and total sediment; p < 0.05) than
in the interior zone (Figure 5, 6). Different patterns in C/N with depth between the
mangrove zones was found for dead roots, leaf litter and total sediment (p < 0.05). δ
13
C
values were higher in the interior zone (for live roots, leaf litter and total sediment; p <
0.05) except for dead roots with similar values between the mangrove zones (p > 0.05).
δ
15
N values were mostly similar between the zones and depth intervals except for live
roots and total sediments at 20-30 cm depth in the interior zone (both p < 0.05). The
effect of long-term fertilization on carbon stable isotopic signatures and C/N in total
sediments and sediment components was calculated as the difference in δ
13
C or C/N
between the nutrient treatment (N or P) and the control treatment (Ctrl), denoted as ∆
13
C
or ∆C/N (Figure 7). The ∆
13
C and ∆C/N of sediments, live and dead roots, and leaf litter
showed different patterns in both mangrove zones. In the fringe zone (Figure 7a), only
δ
13
C of the live roots was significantly more positive in the P-treatment relative to the
Ctrl-treatment (∆
13
C ≈ 1.5‰; p < 0.01). In this zone, C/N was significantly lower for leaf
litter in the N-treatment (∆C/N ≈ -60; p < 0.05), for dead roots in both N- and P-
40
treatments (both ∆C/N ≈ 50; p < 0.01), and for bulk sediments in the P-treatment (∆C/N ≈
30; p < 0.01). In contrast, in the interior zone (Figure 7b), the δ
13
C of the roots were
significantly more negative in the N-treatment (∆
13
C ≈ -1.0‰; p < 0.001) and P-treatment
(∆
13
C ≈ -0.5‰; p < 0.05), and more negative also for leaf litter but only in the N-
treatment (∆
13
C ≈ 0.6‰; p < 0.05) relative to the Ctrl-treatment. In this zone, C/N was
only significantly more negative for leaf litter in the N-treatment (∆C/N ≈ 40; p < 0.01)
and for bulk sediment in the P-treatment (∆C/N ≈ 8; p < 0.0001) relative to the Ctrl-
treatment.

A B
Figure 7. Comparison of δ
13
C and C/N ratios between the nutrient (N, P) and control
treatments (Ctrl) denoted as ∆ from the fringe (A), and interior (B) mangrove zones. S: Total
sediments; DR: Dead roots; LL: Leaf litter; and R: Live roots. Data shown as Avg. ± SE.

41
The total organic carbon (%TOC) in the mangrove sediments was found to be
high and relatively constant among the different zones, nutrient treatments, and depth
intervals studied (~39%; p > 0.05; Table 1). Moreover, the relative abundance (%) of the
dead roots and leaf litter was not significantly different among the nutrient treatments,
zones and depths (p > 0.05). In contrast, the relative abundance (%) of live roots showed
different distributional patterns in each mangrove zone (Figure 8). In the fringe zone, live
roots increased with depth (p < 0.05), and were significantly lower in the N-treatment (p
< 0.05). In the interior zone, live roots were higher in the upper depth intervals (p <
0.001), and were significantly higher in the P-treatment (p < 0.001) (Figure 8).

Figure 8. Relative abundance (%) of live roots for the fringe and interior mangrove
zones in each nutrient treatment (Ctrl, N, P) at each depth interval (cm). Data shown as
Avg. ± SE.

Ctrl N P
42
PLFA Bacteria Broad Community Structure

We detected and identified 13 different PLFAs unique to bacteria (Table 2). The
most common bacterial PLFAs found were from Gram-positive bacteria with the highest
relative abundances (0.7–10.4%) across all mangrove zones. Actinomycetes (10Me18:0)
are the least common bacteria (<1%) present in all treatments. Fungi, plants and
ubiquitous PLFAs are highly variable among zones with 16:0 as the most abundant
(~30%).
Mangrove areas subjected to long-term N or P additions were found to have
distinctly different sediment bacterial PLFA concentrations when compared to the Ctrl-
treatment (Figure 9). In the interior zone, a 3-fold increase of bacterial PLFA

Figure 9. Correlation of bacterial PLFA concentration (µg/g dw) between the control treatment
(Ctrl) and the nutrient treatments (N and P) at the fringe and interior mangrove zones.

Fringe-P
y = -1.02 + 1.39*x
r
2
= 0.78; p < 0.001
Fringe-N
y = 5.35 + 1.88*x
r
2
= 0.74; p < 0.001
Interior-N
y = -2.37 + 2.10*x
r
2
= 0.97; p < 0.001
Interior-P
y = -4.43 + 2.96*x
r
2
= 0.92; p < 0.001
43

Table 2. Relative abundance (%) of PLFAs for the different treatments (Ctrl: control; N:
nitrogen-fertilized; P: phosphorus-fertilized) at each mangrove zone. Values are Avg. ± SE.

44
concentration in the P-treatment and a 2-fold increase in the N-treatment relative to the
Ctrl-treatment were observed. In contrast, in the fringe zone, higher bacterial PLFA
concentrations were found in the N-treatment relative to the Ctrl-treatment (about a 2-
fold increase, Figure 9). Fertilization also affected the bacterial PLFA concentration at
different depths (Table 3). In the interior zone, only the surface sediment in the N- and P-
treatment showed an increase in bacterial PLFA concentrations, which were found to be
50% higher than in the Ctrl-treatment. In the fringe zone, an increase in bacterial PLFA
concentration relative to the Ctrl-treatment occurred in all depths in the N-treatment
(especially in the upper depth intervals with ~60% increase, p < 0.001), whereas in the P-
treatment, increases in bacterial PLFA concentration were only seen in the surface
sediment (~40% increase, p < 0.01).

Table 3. Calculated difference in bacterial PLFA concentration (µg PLFA / g dw; Avg.
± SE) between the Ctrl- and nutrient treatments (N and P) in the mangrove zones for
each depth interval. * denotes significant differences.
Difference
Zone Treatment
Depth
(cm) % Avg.
T-
test
df p
Fringe Ctrl vs N 0-1 52.3 20.1 ± 4.1 4.9 12 < 0.001*

5-10
69.0 23.2
±
4.8 4.8 12 < 0.001*

20-30
14.3 0.4
±
0.1 2.8 12 < 0.05*
Ctrl vs P
0-1
39.9 12.2
±
3.1 3.9 12 < 0.01*

5-10
-46.6 -3.5
±
1.3 -2.7 12 < 0.05*

20-30
7.0 0.2
±
0.2 0.7 12 > 0.05
Interior Ctrl vs N
0-1
51.1 17.7
±
3.5 5.0 12 < 0.001*

5-10
2.7 0.1
±
0.1 0.8 12 > 0.05

20-30
-22.1 -0.3
±
0.1 -2.5 12 < 0.05*
Ctrl vs P
0-1
63.5 29.4
±
7.6 3.9 12 < 0.01*

5-10
22.1 1.1
±
0.6 1.7 12 > 0.05

20-30
15.3 0.3
±
0.1 2.5 12 < 0.05*

45
Environmental parameters in both mangrove zones had a strong effect on broad
bacterial PLFA community composition. The relative abundance of bacterial PLFAs was
different between mangrove zones and changed with depth (Figure 10). Contrasting
distributions between zones along the depth intervals were found in specific PLFAs such
as i17:0, 14:0, and 6cy18:0. RDA analysis showed that bacterial PLFAs vary spatially
and account for 70% of the variance in the fitted data (p < 0.01; Figure 11).
Environmental parameters varying among the depth intervals in each mangrove zone had
a greater effect on bacterial PLFA composition (axes 1: 47%, p < 0.01) than the nutrient
treatments (Figure 11). Based on the parameters measured in this study, the increase of
H
2
S concentration with depth had the strongest influence on bacteria PLFA composition
primarily in the interior zone. NH
4
+
and pH showed a stronger effect on bacterial PLFA
composition in the fringe zone than in the interior zone, whereas PO
4
3-
, pore-water
temperature and the relative abundance of dead roots seemed more important for the
RDA clustering in the interior zone.

δ
13
C PLFA biomarkers

Overall, δ
13
C of all bacterial PLFA biomarkers showed slightly lower values in the
fringe zone (-31.1 ± 0.9 ‰) than in the interior zone (-29.3 ± 0.8 ‰) under natural
conditions (Figure 12a). Although the values for each PLFA biomarker were highly
variable and statistically the same between the mangrove zones (p > 0.05), the pattern in
46

Figure 10. Relative abundance (%) of bacterial PLFAs in the fringe and interior mangrove zones
at different depth intervals.

Fringe-Ctrl
(0-1 cm)
i14:0
14:0
i15:0
a15:0
i16:0
i17:0
a17:0
16:1w7
16:1
17:0
10 Me18:0
8 cyclo18:0
6 cy18:0
Interior-Ctrl
(0-1 cm)
6 cy18:0
i14:0
14:0
i15:0
a15:0
i16:0
i17:0
a17:0
16:1w7
16:1
17:0
10 Me18:0
8 cyclo18:0
i14:0
14:0
Interior-Ctrl
(5-10 cm)
6 cy18:0
i15:0
a15:0
i16:0
i17:0
a17:0
16:1w7
16:1
17:0
10 Me18:0
8 cyclo18:0
Fringe-Ctrl
(20-30 cm)
6 cy18:0
i14:0
14:0
i15:0
a15:0
i16:0
i17:0
a17:0
16:1w7
16:1
17:0
10 Me18:0
8 cyclo18:0
Interior-Ctrl
(20-30 cm)
6 cy18:0
i14:0
14:0
i15:0
a15:0
i16:0
i17:0
a17:0
16:1w7
16:1
17:0
10 Me18:0
8 cyclo18:0
Fringe-Ctrl
(5-10 cm)
i14:0
14:0
i15:0
a15:0
i16:0
i17:0
a17:0
16:1w7
16:1
17:0
10 Me18:0
8 cyclo18:0
6 cy18:0
47

the δ
13
C
PLFA
data follows the trend in the isotopic signature of the trees, and slightly that
of the PLFAs of higher plants (24:0 and 26:0; Table 2) between the mangrove zones. The
comparison of δ
13
C
PLFA
from specific biomarkers such as i+a15:0, using a known
correction of -3.7‰ as the average fractionation between the carbon source utilized and
the PLFA biomarkers (Bouillon and Boschker 2006), to all the possible carbon sources
found in this area, also indicated that bacteria are using mostly in situ carbon sources from
mangrove trees (Figure 13). The δ
13
C in the deeper sediment intervals did not have more
negative values relative to the surface and mid-depth samples (Figure 12b).
Under natural nutrient conditions (Ctrl-treatment), significant differences between
mangrove zones were observed only in the Gram-Positive bacterial group (p<0.01; Figure
13). In the fringe zone, all bacterial groups presented significant differences among the
depth intervals (lower values at 5-10 cm depth; p<0.05), and only Gram-Positive bacteria
-1.0 1.0
-0.8 0.8
Axis 1 (47.1%)
Axis 2 (12.1%)
F-P
Temp Dead roots
pH
Roots
H
2
S
PO
4
3-

Salinity
NH
4
+

I-N
I-Ctrl
I-P
F-N
F-Ctrl
Figure 11. Ordination plot of RDA
results showing sample clusters
based on bacterial PLFAs in
response to environmental variables
(70% explained variance; p < 0.01).
Environmental variables are
indicated as triangles (categorical
factors) and arrows (numerical
variables). Samples are indicated as
crosses. The direction of the arrows
denotes the steepest increase in the
variable, and the length indicates the
strength in explaining PLFAs
variation relative to the other
variables.

48
presented significant differences among the nutrient treatments (NutrientXDepth
interaction; p<0.05). Gram-Positive bacteria in this zone showed more positive δ
13
C
PLFA

values at 0-1 and 5-10 cm depth, and more negative at 20-30 cm depth in the nutrient
treatments than at the respective depths in the Ctrl-treatment. In the interior zone,
significant differences in δ
13
C
PLFA
were observed only in Gram-Negative bacteria
(NutrientXDepth interaction, p<0.001) with isotopic shifts relative to the Ctrl-treatment at
5-10 and 20-30 cm depth in the N-treatment and at 0-1 cm in the P-treatment.
The δ
13
C
PLFA
of the general biomarker 16:0 correlated with 24:0 and 26:0 (higher
plants biomarkers; r
2
=0.604 and r
2
=0.458, respectively; p<0.01) and with 18:3w3
(cyanobacteria and fungi biomarker; r
2
=0.617; p<0.05). 16:0 also correlated with bacteria
biomarkers such as Gram-Positive bacteria (r
2
=0.140; p< 0.0001) and bacterial general
biomarkers (r
2
=0.116%; p<0.0001) but with a low coefficient of determination (r
2
).
Therefore, the 16:0 biomarker is primarily indicative of the δ
13
C
PLFA
of higher plants,
cyanobacteria and fungi. Comparison of δ
13
C
16:0
with δ
13
C of Gram-Positive bacteria in
the fringe zone (Figure 13) showed that only at 20-30 cm depth under the P-treatment
bacterial biomarkers were strongly depleted relative to the 16:0 values. Similarly, δ
13
C of
Gram-Negative bacteria in the interior zone at 20-30 cm depth under the N-treatment
showed strongly depleted values relative to the δ
13
C
16:0
values. Conversely, δ
13
C of Gram-
Negative bacteria at 5-10 cm depth in the fringe zone under N-treatment presented
enriched values relative to 16:0. Conversely, the δ
13
C of Gram-Negative bacteria in the
interior zone under Ctrl- and N-treatment conditions were enriched relative to the 16:0 at
the surface sediment.
49

Table 4. δ
13
C (‰) of bacteria (i+a15:0) compared to total organic
carbon (TOC) and all potential carbon sources found in the
mangrove sediments (dead and live roots, leaf litter, microbial
mats, and seagrasses). Bacteria biomarkers were corrected for
fractionation using a known correction factor of -3.7±2.1‰
(Bouillon and Boschker 2006). All values shown as Avg. ± SE.

N δ
13
C (‰)
Bacteria 36 -27.4 ± 0.4
TOC 160 -26.2 ± 0.1
Dead roots 125 -27.0 ± 0.1
Live roots 136 -26.2 ± 0.1
Leaf litter 95 -27.6 ± 0.2
Microbial mats 90 -18.5 ± 0.3
Seagrasses 6 -7.6 ± 0.5
i14:0
14:0
17:0
10Me18:0
16:1
16:iw7
6cy18:0
8cyclo18:0
i15:0
a15:0
i16:0
i17:0
B
A
-25

-27

-29

-31

-33

-35

-37
-25

-27

-29

-31

-33

-35

-37
δ
13
C-PLFA δ
13
C-PLFA
General Gram +
Actino. Gram -
Figure 12. δ
13
C-PLFA (‰) of
bacterial groups found in the
fringe and interior mangrove
zones under natural conditions
(A) and at different sediment
depth intervals (B). All values
shown as averages ± SE.
Actinomycetes: Actino.; Gram-
Negative: Gram-; Gram-Positive:
Gram+.

50

Figure 13. Carbon isotope ratios of PLFA microbial groups (Gram-Negative and Gram-Positive bacteria) and the general biomarker
16:0 in the sediments of the fringe and interior mangrove zones. Data in each zone is shown under the different nutrient treatments (Ctrl,
N and P) at different depth intervals (0-1, 5-10, 20-30 cm). All values shown as Avg. ± SE.
51
Discussion

The sedimentary environment

Nutrient concentrations in the pore-water of the Belizean mangrove forest studied
showed that N is more available and P is less available in the fringe zone than in the
interior zone. Previously, long-term fertilization experiments in the same area showed
that mangrove trees experience N and P limitation across a gradient from the fringe to the
interior mangrove zones (Feller et al. 1999, 2002). In this study, we found the effect of
long-term nutrient enrichment on mangrove sediment geochemistry and bacteria
communities (based on PLFA) depended both on the forest zone and on the nutrient
added (N or P). In the interior zone, sediment and pore water parameters changed towards
a more deficient P condition when N was added, and increased P availability when P was
added. In the fringe zone, only fertilization with P had a strong effect on sediment
geochemistry, indicating an increase in N deficiency. The similar spatial pattern between
the nutrient pore-water pools, and the nutrient limitation of the mangrove trees and
bacteria is indicative of a strong interaction between the biotic and the abiotic fraction of
the sediments.
The δ
13
C of the R. mangle sedimentary components (i.e. leaf litter, dead and live
roots) were in the same range as the mangrove leaf tissue from Twin Cays (from -30‰
to -24‰) (Mckee et al. 2002, Wooller et al. 2003, Fogel et al. 2008) and elsewhere
(Schwamborn et al. 2002, Bouillon 2004a). The difference in the carbon stable isotopic
52
signature of mangrove leaves in the fringe versus the interior zone (-28.3‰ and -25.3‰,
respectively; Smallwood et al. 2003) was also observed in the sediment of these zones.
This difference between the mangrove zones was explained previously by a decrease in
stomatal conductance (photosynthetic rate) in the interior zone due to higher irradiance
(Cheeseman and Lovelock 2004) and water temperature, both affected by a greater
canopy openness in this zone (McKee et al. 2002). This pattern of increased δ
13
C towards
the interior forest area was also observed in the leaf litter analyzed in our study (Figure
6). The δ
13
C of the leaf litter remains relatively unchanged during microbial colonization
and decomposition (Werry and Lee 2005), even after a prolonged period of time (Wooller
et al. 2003) suggesting that any variation in δ
13
C occurred prior to leaf fall. In contrast,
roots may undergo different diagenetic processes due to differences in δ
13
C between live
and dead roots (Figure 6). Although the stable carbon isotopic composition of these local
mangrove carbon sources only differed by ~3‰, the patterns observed in δ
13
C between
the zones and sediment depths were contrasting, allowing the interpretation of general
trends in bacterial carbon utilization patterns relative to other in situ non-mangrove
carbon sources.

Bacterial communities in mangrove sediments

The analysis of bacterial PLFAs in Belize R. mangle peat sediments showed that
the community composition and biomass of bacteria are strongly controlled by
environmental parameters like H
2
S pore-water concentrations that largely change with
53
depth. Therefore, long-term fertilization with N or P is less important in structuring
bacterial community composition than environmental parameters like H
2
S concentrations.
Results also indicated that a switch from only N limitation to N and P limitations in
bacteria occurs across a short spatial scale from the fringe to the interior forest zone.
Specifically, under natural conditions in Belize (Ctrl-treatment), bacterial PLFA
biomass was high (26 - 228 ug/g dw) compared to the biomass found in other mangrove
forest types (fringe and estuarine) with low content of TOC (in Sri Lanka and India up to
80 ug/g dw; Bouillon et al. 2004a). At our study site, the bacterial PLFA biomass dropped
an average of about 92% from the surface (0-1 cm) to the deepest depth interval (20-30
cm) (Table 3), a much higher variability with depth than found in other mangrove forests
(Bouillon et al. 2004b). This decrease in bacterial PLFA biomass in Belize is more likely
to occur due to an increase in H
2
S concentration with depth and not because of changes in
%TOC. Alongi (1988) and Bouillon et al. (2004a) found that bacterial densities in
mangrove sediments are related mostly to organic C and N content. Peat sediments in
Belize are consistently high in %TOC, so bacteria may not be constrained by the
availability of carbon relative to other nutrients. Moreover, bacterial C composition
(calculated using 0.056 g of carbon PLFA per gram of carbon biomass; Alongi 1988 and
Bouillon et al. 2004a) relative to sediment TOC content at our study site ranged from
~0.3% in the deepest depth up to ~3% in the surface sediment, similar to the observed
variability of bacteria biomass with depth. Bacterial C at the surface sediment is close to
the highest bacterial carbon stocks found in benthic habitats calculated from mangrove
sediments (~5%; Alongi 1988). Our calculation of bacterial C composition relative to the
sediment TOC content is in the low range in this mangrove forest because microbial mats
54
growing on top of the sediments were not included. Microbial mats in Twin Cays are
known to have a large distribution (Joye and Lee 2004) and therefore will increase the
values of bacterial C composition calculated.
Although several studies using phospholipid analyses have indicated an increase in
microbial biomass by increased levels of eutrophication (e.g. Koster et al. 1997) and other
sources of contamination (Cotano and Villate 2006, Polymenakou et al. 2006, Mchenga
and Tsuchiya 2007), here we present for the first time the relationship between bacterial
biomass (based on PLFA concentrations) and long-term nutrient additions (N and P) in
mangrove sediments. Both N and P stimulated the bacterial biomass in the fringe and
interior zones, but the highest increase in bacterial biomass occurred in the interior zone
fertilized with P (up to ~3 fold), and in the fringe zone fertilized with N (~2 fold),
following the major trend in nutrient limitation of mangrove trees seen in previous studies
(Feller et al. 1999, 2002). Different bacterial groups may be limited by different nutrients
in the sediments but the overall bacterial community may experience P and N limitations
depending on the location within the mangrove forest.
Further study needs to be conducted for understanding how nutrient limitations of
bacteria and trees regulate the dynamics of N and P processes between mangrove forest
zones. For example, P-limited bacteria and plants in the interior zone may strongly impact
N availability, which is known to influence carbon fixation by cyanobacteria in microbial
mats (Joye and Lee 2004) and mangrove tree growth in Belize (e.g. Feller et al. 2002).
The poor relationship between nutrient demands and sediment geochemistry in the fringe
zone makes it difficult to predict the effects of nutrient enrichment in this zone, although
55
stimulation of nitrogen fixation may occur when P is added (as observed in the interior
zone, Chapter 4). It seems that nutrient limitation of mangrove trees and bacteria in Belize
is likely a consequence of nutrient source availability and not differential strategies among
plants and microorganisms to maximize overall resource utilization within an ecosystem
as observed in other coastal environments (Sundareshwar et al. 2003).

Bacterial carbon source utilization patterns

The δ
13
C
PLFA
values observed in Twin Cays indicate that the bacteria were
primarily using mangrove carbon sources even when the sediment surface was covered
with seagrass (at the fringe Ctrl-treatment, δ
13
C

of seagrass is about -9.5‰) or when the
sediments have been treated with N or P fertilizers. This trend to use local carbon sources
in TOC rich environments was observed as well in other mangrove forests (Bouillon et
al. 2004a, Bouillon et al. 2005, Bouillon and Boschker 2006). In our study, we went
further and analyzed the δ
13
C
PLFA
within the mangrove peat sediments to elucidate carbon
source utilization patterns under natural and induced nutrient conditions. High variability
observed in δ
13
C
PLFA
indicates bacteria use different carbon sources under variable
anoxic/oxic conditions.
Twin Cays peat sediment in the Boa Flats area is a mixture of different locally
produced R. mangle tissues (dead and live roots and leaf litter) and microbial mats that
act as different carbon sources for bacteria. Under natural conditions in Twin Cays,
neither strongly depleted values nor a decrease in δ
13
C
PLFA
with depth was observed,
56
indicating that reducing conditions in these mangrove sediments had little effect on the
bacteria δ
13
C
PLFA
values, although high concentrations of H
2
S at 20-30 cm depth
significantly affected the bacterial PLFA community composition. Low δ
13
C
PLFA
values
in the fringe zone at 5-10 cm depth (-36‰) corresponded with low δ
13
C of the leaf litter,
suggesting that leaf litter at this mid-depth is the primary carbon source being utilized by
bacteria. Sulfate reduction is responsible for most of the organic matter mineralization in
this mangrove forest (Lee et al. 2008), and the fact that the δ
13
C
PLFA
values did not reflect
this important microbial process suggests that these sediments are aerated by semidiurnal
tides (Lee et al. 2008 and in this study) and by oxygen leakage from the mangrove roots
(McKee 1993). In addition, the lack of highly δ
13
C values in bacterial PLFAs indicates
that methane and CO
2
produced during anaerobic respiration are not being used by
methanotrophs and chemoautotrophs, respectively, at least to a degree to greatly decrease
the δ
13
C values of lipids, as has been seen previously elsewhere (e.g. Hollander and
Smith 2001, Cifuentes and Salata 2001). Uncommon for mangroves, Belize mangrove
sediments have high concentrations of CH
4
(up to 80 µM) and persistent methanogenesis
rates over time (Lee et al. 2008). Most of the CH
4
may thereby be lost through the water-
sediment interface.
Variability in δ
13
C
PLFA
between Gram-Negative and Gram-Positive bacteria
across the mangrove zones indicates that carbon source utilization patterns were specific
to microbial groups and dependent on the environmental gradients observed within the
mangrove forests. The significant differences observed among the nutrient treatments in
the Gram-Negative bacteria in the interior zone and the Gram-Positive bacteria in the
57
fringe zone (Figure 13) have three possible explanations: 1) the δ
13
C of the primary
source utilized by bacteria changed with nutrient treatment; 2) the bacteria shifted to a
different carbon source; and/or 3) the bacterial community changed to different metabolic
strategies. In the fringe zone, the shift in the δ
13
C of Gram-Positive bacteria under N- and
P-treatment corresponded well with the shift in δ
13
C
16:0,
except at 20-30 cm depth in the
P-treatment. This trend may indicate bacteria relied principally on roots and degrading
organic matter (e.g. dead roots and leaf litter) as carbon sources and switched to
anaerobic metabolism at 20-30 cm depth in the P-treatment. In contrast, in the interior
zone, the δ
13
C of Gram-Negative bacteria corresponded well only with δ
13
C
16:0
in the P-
treatment, indicating bacteria may rely principally on live roots due to a similar pattern in
the δ
13
C of live roots with 16:0 with depth (Figure 14). Also in the interior zone,
enriched δ
13
C
PLFA
on the surface of the sediments indicates that microbial mats in Twin
Cays are an important carbon source, except in the P-treatment where an increase in the
canopy of the trees reduces light on the surface of the sediments, thereby reducing the
occurrence of microbial mats. Hence, microbial mats in the interior of the forests are
important not only for importing new nitrogen into the sediments through N
2
fixation
(Joye and Lee 2004) but also for providing labile carbon to heterotrophic bacteria.
Some studies have shown the effects of anthropogenic disturbance on bacteria
carbon source utilization patterns, such as land use in terrestrial environments (Burke et
al. 2003) and nutrient loadings in seagrasses (Holmer et al. 2004) due to changes in the

58

Figure 14. δ
13
C values in the interior mangrove zone for PLFA biomarker 16:0 and the
microbial group Gram-Negative bacteria, and for live roots. Data in each nutrient
treatments (Ctrl, N and P) is shown at different depth intervals (0-1, 5-10, 20-30 cm). All
values shown as Avg. ± SE.

availability of the carbon sources with different δ
13
C signatures. In addition, Hollander
and Smith (2001) show that long-term eutrophication of a fresh water lake depleted the
isotopic signatures of microorganisms by increasing the use of secondary products from
anaerobic metabolism. Our study, however, shows for the first time, that fertilization with
N or P can enhance anaerobic metabolism deeper in the mangrove sediments only when
fertilization increased nutrient limitation conditions for bacteria.
Few studies have attempted to better understand the functional relationships
among plants, bacteria, and sediment geochemistry in a tidally dynamic, oligotrophic
system. In Twin Cays mangrove forests, we found that bacteria experienced a similar
nutrient limitation pattern as R. mangle trees, indicating a limited nutrient supply within
the forests. Additionally, results show that fertilization with N or P, while having a minor
16:0 Gram-Negative Live roots
δ
13
C
PLFA
(‰) δ
13
C
PLFA
(‰) δ
13
C

(‰)
59
effect on broad bacteria community structures, can strongly influence not only the pore-
water and sediment pools of mangrove forests, but also metabolic processes in specific
microbial groups when nutrient limitation is enhanced through fertilization. In sum, our
study shows that Belizean mangrove sediments are characterized by pronounced
gradients of chemical and biological parameters, resulting in multiple factors interacting
in short spatial scales that have an important role in the overall ecosystem function.

60

Chapter 4

SPATIAL AND TEMPORAL VARIABILITY OF NITROGEN-FIXING
MICROBIAL POPULATIONS IN THE RHIZOSPHERE OF
MANGROVE SEDIMENTS

61
4. Spatial and Temporal Variability of Nitrogen-Fixing Microbial Populations in the
Rhizosphere of Mangrove Sediments

Abstract

The ecological interaction of nitrogen-fixing populations and mangrove roots was studied
in a peat-based sediment subjected to a long-term fertilization with nitrogen and
phosphorus in Twin Cays, Belize. Spatial and temporal variability of N
2
fixation rates
was measured and the molecular diversity and community structure of N
2
fixers were
determined in the rhizosphere and bulk sediments. N
2
fixation rates under natural
conditions showed a distinct pattern compared to the fertilized treatments, with rates
following the distribution of live roots in the sediment and remaining relatively
unchanged over the different temporal scales studied (diel, 10 day period, and seasons).
N
2
fixation rates in the P-treatment also followed the distribution of live roots but
presented strong temporal changes over short (diel) and 10 day time periods. In contrast,
low N
2
fixation rates in the N-treatment were relatively constant over the spatial and
temporal scales studied. The community composition of N
2
fixers showed a large
temporal variability primarily explained by the changes in nutrients and H
2
S
concentrations between the dry and wet season. A large spatial variation was also
observed but a higher number of OTUs (Operational taxonomic units) were found in
sediments with fewer live roots. Overall, nitrogen-fixing microbial populations showed
high spatial and temporal variability not always explained by variability in the abundance
62
of live roots in the sediments. Fertilization induced patterns not observed under natural
conditions, indicating that a link between community composition and function may be
relevant in communities displaced from natural conditions.

Introduction

The bioavailability of nutrients in sediments is one of the most important factors
controlling ecological processes in the mangrove ecosystem (Alongi et. al. 2002, Feller et
al. 2003, Cheeseman and Lovelock 2004), but the extent and degree to which nutrients
affect the pathways for nitrogen cycling in sediments is still not well understood. This is
in part due to the multiple pathways in the nitrogen cycle that are mediated principally by
microbial rather than by chemical processes and that are dependent both on the forms of
the nitrogen pools and the physical environment (Purvaja et al. 2008).
Within the mangrove forests, N
2
fixation has been detected in sediments (Zuberer
and Silver 1978) including the rhizosphere (Holguin et al. 1992), on tree trunks (Sheridan
1991) and aerial roots (Toledo et al. 1995a) as well as in cyanobacterial mats (Lee and
Joye 2006) and in leaf litter (Gotto and Taylor 1976, Woitchik et al. 1997, Pelegri and
Twilley 1998). Therefore, the biological fixation of N
2
is an important source for “new”
nitrogen in this marine environment. Although still little is known about the ecology of
N
2
fixers in mangrove sediments and the intra-forest factors that regulate its diversity and
activity, this process is estimated to contribute about ~538 nmol N / m
2
· h in marine
benthic environments with 10% in mangrove systems (Capenter and Capone 1983). A
63
close root-bacterial interaction may sustain the high activity (Zuberer and Silver 1978,
Sengupta and Chaudhuri 1991, Ravikumar et al. 2004) and diversity (Flores-Mireles et al.
2007) of N
2
fixers compared to bare sediments due to high availability of labile organic
carbon, low nitrogen concentrations and microaerophilic conditions in the rhizosphere.
Free-living N
2
fixers in sediments (Zhang et al 2008) and in association with
mangrove roots (Flores-Mireles et al. 2007) have shown to be strongly influenced by
organic carbon concentration than by other sediment parameters (e.g. nitrogen). Despite
several studies that claim microbial processes are not primarily driven by the availability
of dissolved organic carbon from roots (e.g. Sengupta and Chaudhuri 1991), other studies
have shown that plant-microbial interactions are important in supplying through N
2

fixation up to 60% of the annual nitrogen requirement of mangrove trees (e.g. Zuberer
and Silver 1978). In turn, mangrove trees can enrich the organic carbon pool and change
the redox conditions of the sediments, thereby influencing the growth of certain microbial
groups (see references in Holguin et al. 2001). Studies in other ecosystems such as salt
marshes (e.g. Piceno et al 1999, Bagwell and Lovell 2000, Piceno and Lowell 2000b),
seagrasses (see reviews by Welsh 2000 and McGlathery 2008) and terrestrial
environments (for review see Singh et al. 2004) have also indicated that plants strongly
influence N
2
fixers. Overall, it has been hypothesized that the rhizosphere harbors greater
diversity, abundance and activity of microorganisms that promote the stability of the
microbial community over different spatial and temporal scales.
Despite the number of studies on N
2
fixation in mangrove ecosystems (see
reviews by Holguin et al. 2001 and Purvaja et al. 2008), the activity and diversity of free-
living (in sediments) and associative N
2
fixers (in the rhizosphere) under different
64
environmental conditions have not been thoroughly studied. This study explores the
variability of nitrogen-fixing populations in peat-based mangrove sediments over
different spatial and temporal scales and whether long-term fertilization with nitrogen
and phosphorus influences the community structure and activity of this functional group.
Here we provide the first characterization of how the community structure and
composition of free living and root-associative N
2
fixers in mangrove sediments of
Rhizophora mangle change along natural and induced gradients of physical and chemical
parameters. We used a culture-independent molecular tool to determine the variability in
the community structure and composition of N
2
fixers, and direct multivariate methods to
evaluate the covariation of N
2
fixers with the sediment environmental parameters. This
combination of tools allowed us to identify key factors that influence the activity and
community structure of N
2
fixers. This study advances our understanding of both
microbial ecology in coastal tropical areas and how changes in natural and nutrient-
induced conditions affect the dynamics of the microbial population inhabiting benthic
environments, with potential negative consequences for the overall ecosystem.

Methods

Field Experiment and sample collection

The study area is located in the interior zone of Boa Flats forest (for a detailed
description of the study area see Chapter 2). To study the variability of nitrogen-fixing
65
bacteria in the rhizosphere of mangrove sediments subjected to long-term fertilization,
samples where taken from discrete depth intervals (0-5, 5-10 and 20-30 cm) at the
experimental sites in the interior zone (Control, Ctrl; Phosphorus, P; and Nitrogen, N
treatments) during two different seasons (wet and dry). Also, the activity of the N
2
fixers
was determined during short periods of time (diel and 10 days). In addition, in parallel we
collected roots and bulk sediments in the rhizosphere of the mangroves for comparison of
the N-fixing community structure and composition in these two substrates over the spatial
(depth) and temporal (seasons) range covered in this study.
Pore-water samples were collected using sippers inserted into the peat sediment at
the three depth intervals (N=3 per depth). The sippers were attached to a filter holder with
pre-combusted glass-fiber filters (Whatman GF/F 0.7µm, 4.7cm diameter) and connected
through a two-way stopcock valve to a luer-lock syringe (which serves as a reservoir for
the sample). Samples were filtered for a second time with an Acrodisc syringe filter (Pall
0.45 µm HT Tuffryn
®
membrane) and divided into subsamples for each pore-water
measurement. All glass vials used were acid-washed, rinsed with ultrapure deoinized
water and combusted at 500ºC for 5h. Sediment cores were taken with a Russian Peat
corer which is designed to avoid vertical compaction of the sediment samples, within
~0.5 m of the main trunk of the mangrove trees, and divided by sediment depth intervals
and sample types for further analysis.

66
Pore-water analysis

Temperature and pH were determined immediately after the collection of the pore
water samples (ORION with pH and ATC electrode; calibrated with certified standards,
reproducibility of ± 0.02 pH units and ± 1.0
o
C). Salinity was measured with a
refractometer (Fisher Scientific, calibrated with deionized water, with an accuracy of ±
1.0 ppt). Samples for determining the hydrogen sulfide (H
2
S) content (volume 0.1 ml)
were stored in glass vials containing 0.04 ml of 0.05 M zinc acetate and analyzed
following the method of Cline (1969). For ammonium (NH
4
+
) analysis, samples (volume
0.5 ml) were stored in vials containing 1 ml of phenol reagent and analyzed as in
Solorzano (1969). Dissolved inorganic phosphate (PO
4
3-
) was measured colorimetrically
using the method of Strickland and Parsons (1972). A spectrophotometer (Shimadzu UV-
1700) was used for pore-water analysis at the University of Southern California.

Sediment characterization

The relative abundance (%) of live and dead roots was determined by washing the
sediment samples collected at each depth interval with different sized sieves (4 mm,
2mm). Each of these sediment components was picked manually, oven dried for 48 h at
60
o
C, and weighed. The weight of the live and dead roots was compared to the weight of
the total sediment in each sample to calculate the relative abundance (%).

67
N
2
Fixation rates

Nitrogen fixation rates were measured by the acetylene reduction technique
(Capone and Montoya 2001). The sediment cores were divided into subsamples from
each depth interval in a glove bag filled with N
2
. The subsamples were placed
immediately into a serum bottle, sealed, flushed with N
2
, filled with 10% acetylene, and
incubated in ambient water for 24h. For each incubation experiment, gas samples were
taken every 4-6 hours and stored in vacutainers. The samples were measured at the
University of Southern California on an FID Shimadzu GC-9A gas chromatograph. The
acetylene reduction rates were calculated and converted to N
2
fixation rates assuming the
conversion factor of 4:1 (C
2
H
2
:N
2
) (Postgate 1982), which had been previously used in
the same study area (Lee and Joye 2006). The depth integrated rates within each nutrient
treatment were calculated by averaging the depth integrated rates for each depth interval.

Community structure and composition

The roots and bulk sediment samples (~1.5g) were taken at each sediment depth,
fixed with TE (10mM Tris HCL plus 1mM EDTA, pH of 7.4) and frozen. We used
Terminal restriction length polymorphism (TRFLP) of PCR amplified nifH gene to study
the community composition of the N-fixing microbial populations (Widmer et al. 1999,
Hewson and Fuhrman 2006). The DNA was extracted using the FastDNA SPIN kit
(Biogene). Reactions for the nested PCR for amplification of the nifH gene consisted of
1XPCR Buffer, 2.5 mM MgCl
2
, 0.2 mM dNTPs, 0.8 µM for each primer (nifH3 and
68
nifH4 for the first reaction, nifH1-TET and nifH2 for the second reaction), 400 ng/ul
BSA, and 5.0 U Taq.
The PCR products were run in an agarose gel (2%) for 2 h and the desired product
size (~360-380 bp) was excised, purified (QIAGEN), quantified (Pico Green
Fluorescence) and digested with the restriction enzyme HaeIII following the
manufacture’s instructions. This enzyme was chosen because previous tests showed
better consistency of peak numbers over a series of replicates than other enzymes. Clean
digested products were run in duplicates in an ABI 3730 Automated Sequencer and the
outputs were aligned against all possible fragment lengths (100-360bp). The area under
each peak was used to calculate the relative abundance (%) of each OTU (Operational
Taxonomic Unit) in each sample. The bacterial community structure was determined
using three different ecological indices: the Shannon index of diversity (H’; Shannon and
Weaver 1963), the Pielou index of community evenness (J’; Pielou 1966), and the
number of OTUs for community richness.

Statistical analysis

The spatial and temporal variability of all the pore-water parameters, the N
2

fixation rate data, and the community structure indices were analyzed with an ANOVA
with nutrient treatment, depth, season, season X nutrient treatment interaction and depth
X nutrient treatment interaction as factors (JMP Software, SAS Institute Inc., Cary, NC).
When significant differences were found, Tukey’s test was used to compare the means of
the treatments. The level of statistical significance was set to P-level <0.05. All data were
69
tested to fulfill normality and equal variance assumptions, and transformations were
carried out when necessary.
Nitrogen-fixing community composition patterns in response to the environmental
gradients were analyzed using a unimodal constrained model (CCA, Canonical
Correspondence Analysis) due to the range of variation of the environmental variables
(Ter Braak and Prentice 1988). In CCA plots, each canonical ordination axis corresponds
to a direction in the multivariate scatter of the TRFLP-nifH (OTUs) data that is
maximally related to a weighted linear combination of the environmental variables
(Legendre and Legendre, 1998). How well the data are displaced in the CCA result plot
was expressed by the percentage variance accounted for the fitted OTUs data. The
temporal variability of the nitrogen-fixing community composition was analyzed using
nutrient treatments (Ctrl, N, P) and seasons (wet and dry) as interaction factors, and
sediment depth and substrate types were used as covariables. The spatial variability of the
nitrogen-fixing community composition was analyzed using the nutrient treatments and
the depth intervals as interaction factors, and seasons as covariables. Two assumptions
were made when running the temporal and spatial analyses: First, the relative abundance
of dead and live mangrove roots does not change between the wet and the dry seasons
sampled. Second, the physical and chemical parameters of the pore-waters are the same
for the roots and sediment samples at each depth interval. The Canonical Correspondence
Analysis (CCA) and ordination plots were conducted with CANOCO software
(Microcomputer Power, Ithaca, N.Y.). The percent abundance of OTUs was transformed
(log (x+1)), and the environmental data was centered and standardized. The Generalized
70
Additive Model was used to look at the OTUs’ response curves to the environmental
parameters. All values are shown as the average ± SE (where N=3).

Results

Environmental parameters

The depth variability of the pore-water parameters measured was different among
the nutrient treatments studied. The increases in salinity with depth in all the nutrient
treatments indicated poor flooding in this interior mangrove zone (Table 5). In contrast,
the pH was lower at 5-10 cm depth (p<0.01) with the highest values within the N-
treatment compared to the Ctrl- and P-treatments (p<0.001). The pore-water temperature
was relatively constant with depth, but a gradient from high to low was observed from the
Ctrl- to the N- and P-treatments (p<0.001) as an effect of greater canopy development
caused by fertilization. Although NH
4
+
was consistently higher at all depths in the N-
treatment, it was not significantly different from the other nutrient treatments (p>0.01).
PO
4
3-
concentration was significantly higher in the P-treatment (p<0.01) and variable with
depth (p<0.01). Lower PO
4
3-
values were observed at 20-30 cm depth in the Ctrl- and N-
treatments, while higher values at this depth were observed in the P-treatment. H
2
S
concentration increased with depth in all the nutrient treatments (p<0.01, Table 5),
although the P-treatment had lower values (p<0.01).
71
Table 5. Pore-water parameters for each nutrient treatment and depth in
the interior mangrove zone. The data are: Avg. ± S.E.

Nutrient treatments
Parameter
Depth
(cm)
Ctrl N P
pH 0-5 6.31 ± 0.14 6.51 ± 0.04 6.32 ± 0.06
5-10 6.12 ± 0.14 6.34 ± 0.04 6.13 ± 0.03
20-30 6.17 ± 0.11 6.52 ± 0.08 6.24 ± 0.03
Salinity (‰) 0-5 34.67 ± 1.30 38.36 ± 1.07 40.54 ± 1.41
5-10 40.42 ± 1.13 40.27 ± 0.75 41.42 ± 1.07
20-30 42.20 ± 0.55 43.6 ± 0.33 42.67 ± 0.44
T (°C) 0-5 34.49 ± 0.79 32.48 ± 0.63 32.07 ± 0.59
5-10 34.20 ± 0.55 32.82 ± 0.36 31.57 ± 0.44
20-30 34.35 ± 0.40 33.23 ± 0.47 32.32 ± 0.38
NH
4
+
(µM) 0-5 38.29 ± 6.08 74.15 ± 23.62 46.68 ± 11.41
5-10 65.95 ± 6.78 69.95 ± 23.70 48.81 ± 4.16
20-30 68.95 ± 10.15 88.78 ± 23.83 62.81 ± 4.98
PO
4
3-
(µM) 0-5 1.54 ± 0.29 1.45 ± 0.46 30.12 ± 8.82
5-10 1.18 ± 0.29 0.73 ± 0.06 110.79 ± 13.23
20-30 0.95 ± 0.07 0.85 ± 0.20 248.62 ± 49.52
H
2
S (mM) 0-5 0.29 ± 0.08 0.29 ± 0.04 0.35 ± 0.06
5-10 0.72 ± 0.18 0.38 ± 0.07 0.39 ± 0.08
20-30 1.34 ± 0.17 2.42 ± 0.06 0.41 ± 0.03

72
The percentage of dead roots was higher but not significantly different (p>0.05)
towards the deeper depths at the Ctrl and N-treatment, whereas the abundance of live
roots was significantly higher at 5-10 cm depth in the Ctrl- and P-treatment and at 0-1 cm
depth in the P-treatment (p<0.001). The highest percentage of live roots was observed in
the P-treatment (p < 0.001) (Figure 15).
All the parameters measured, except H
2
S, displayed significant differences
between the seasons in each nutrient treatment (p<0.05) (Figure 16). Salinity and NH
4
+
were higher and temperature was lower in the dry season. The pH was also lower in the
dry season, but only in the Ctrl- and P-treatments. PO
4
3-
concentration was significantly
higher in the dry season but only in the P-treatment (p<0.05).
For shorter periods of time (1-10 days) the pore-water temperature was
significantly lower after 10 days in the Ctrl- and N-treatments (Ctrl: from 32.7 ± 0.1 °C to
31.0 ± 0.2 °C; N: from 31.3 ± 0.2 °C to 29.6 ± 0.2 °C; both p<0.01), as well as the PO
4
3-
concentration but only in the P-treatment (from 130.1 ± 27.7 µM down to 66.2 ± 13.4
µM; p<0.001).

N
2
fixation rates

Under natural nutrient conditions, the N
2
fixation rates range from 0.01 up to 3.6 nmol
N/m
2
· h with no differences between seasons (p>0.05), but with significant differences
among the depth intervals (0-5 cm: 0.6 ± 0.1; 5-10 cm: 1.3 ± 0.2; 20-30 cm: 0.4 ± 0.1
nmol N/g dw · h; p<0.05). Also, significant differences were observed among the nutrient
73

Figure 15. Comparisons of the relative abundance (%) of live and dead roots
among the nutrient treatments (Ctrl, N, P) at each depth interval in the
interior zone of Twin Cays mangrove forests. Data shown as Avg. ± SE.

Ctrl
N
P
74

Figure 16. Pore-water parameters at the different nutrient treatments (Ctrl, N, P)
during the wet and dry season. Data shown as Avg. ± SE.

75
treatments (Crtl: 0.8 ± 0.1, N: 0.4 ± 0.1, P: 4.2 ± 0.5 nmol N/g dw · h; p<0.001), with
decreasing N
2
fixation rates with depth only in the N- and Ctrl-treatments. In addition,
diel differences were observed only in the P-treatment (p<0.0001), from 1.8 ± 0.4
nmol N/g dw · h in the day, and up to 4.9 ± 0.5 nmol N/g dw · h at night (Figure 17).
Significant positive correlations of the N
2
fixation rates were observed only with
PO
4
3-
concentration (r
2
= 0.458%, p<0.0001) and with the percentage of root abundance
(r
2
= 0.473%, p<0.001). For the 1-10 day period, N
2
fixation rates were significantly
lower at 10 days in the P-treatment (p<0.01) relative to day 1, indicating changes over
short periods of time possibly due to the variability in the environmental parameters
observed during the experiment (Figure 18).

Figure 17. N
2
fixation rates diel variability under the different nutrient
treatments (Ctrl, N, P) in each sediment depth interval (0-5, 5-10, 20-30 cm).
Ctrl: green squares; N: blue circles; P: red triangles. Data shown as Avg. ± SE.

5-10
20-30
0-5
Depth intervals (cm)
Day Night
N
2
fixation (nmol N/g dw · h)
0.0 2.0 4.0 6.0 8.0 0.0 2.0 4.0 6.0 8.0
76

Figure 18. Nitrogen fixation rates under different nutrient
treatments (Ctrl, N, P) for a short period of time of 10 days.
Day 1: T-1; Day 10: T-10. Data shown as Avg. ± SE.

Nitrogen-fixing community structure

The ecological index of diversity based on nifH OTUs (H’) was not significantly
different between the roots and sediment samples, sediment depths, nutrient treatments,
and seasons (p>0.1). The index of evenness (J’) showed only significant differences
between the seasons in the N-treatment of both live root and bulk sediment samples
(p<0.05). The number of OTUs was also not different among the sediment depths and
nutrient treatments (p>0.1), but showed a significant increase in the number of OTUs
from the wet to the dry season only in the root samples of the fertilized treatments
(p<0.05; Figure 19). In addition, the total number of OTUs found in this environment was
77

Figure 19. Ecological indices of community structure based on nifH OTUs for live
root and bulk sediment samples in the nutrient treatments (Ctrl, N, P) during the
wet and dry seasons. Data shown as Avg. ± SE.
Live roots Bulk sediments
78
similar between the root (with 27% unique OTUs) and bulk sediment (with 22% unique
OTUs) samples, and nearly 51% of the OTUs were found at both microsites.

Effect of environmental parameters on Nitrogen-fixing community

Environmental parameters and fertilization conditions strongly influenced the
nitrogen-fixing community composition. Analysis of TRFLP-nifH (OTUs) showed that
pCCA accounts for 42.3% (p<0.01) of the total temporal variability in the community
composition of N
2
fixers (24.7% on axis 1 and 11.8% on axis 2; Figure 20). PO
4
3-

concentration seemed to be primarily driving the community composition in the

-1.0 1.0
-0.6 1.0
Axis 1 (24.7%)
Axis 2 (11.8%)
Temp
P-Treatment
pH
H
2
S
PO
4
3-

Salinity
NH
4
+

N-Treatment
Ctrl-Treatment
Figure 20. Ordination plot of pCCA results
showing the temporal variation in the
TRFLP-nifH fingerprints (42.3% explained
variance, P=0.004). Sediment depth
intervals and substrate type were used as
covariables. Triangles indicate OTUs at
their optimal position in the plot. Arrows
represent environmental variables and
fertilization treatment (Ctrl, N, P). The
directions of the arrows denote the steepest
increase in the variable and the length
indicates the strength in explaining OTUs
variation relative to the other variables.

79
P-treatment. In contrast, NH
4
+
and H
2
S concentrations had a strong influence on the
community composition in the Ctrl-treatment. Parameters such as salinity, pH and
temperature were important for the community composition of N
2
fixers only in the Ctrl-
and P-treatments. The community of N
2
fixers in the N-treatment seemed to be less
variable between the seasons and less influenced by the temporal changes in the
environmental parameters studied.
Spatial analysis of the nitrogen-fixing community composition using pCCA
showed a strong change in composition among the sediment depth intervals but the
strongest variability was observed under fertilized conditions (78.4% of the total
variability on the first two axes; p<0.01; Figure 21). The percentage of live root
abundance

Figure 21. Ordination plot of pCCA results
showing the spatial variation in the
TRFLP-nifH fingerprints (78.4% explained
variance, P=0.002). Temporal variability
between the seasons was used as
covariable. Triangles indicate OTUs at
their optimal position in the plot (only the
best fit OTUs are shown). Arrows
represent environmental variables of the
interaction of fertilization treatment (Ctrl,
N, P) with depth intervals (0-5, 5-10, 20-30
cm). The direction of an arrow denotes the
steepest direction in the variable, and the
length indicates the strength in explaining
OTUs variation relative to the other
variables. The size of the circles indicates
the relative abundance of live roots that
range from 3% (smallest circles) up to 50%
(the biggest circles).

N x Depth
P x Depth
Ctrl x Depth
-3 4
-2 3
Axis 1 (19.7%)
Axis 2 (12.3%)
80

was observed to decrease with depth (Figure 15), influencing the composition of
nitrogen-fixers, yet the niche optima for the majority of the OTUs (as the position in the
pCCA plot) were found at low abundance of live roots (<20%) in each nutrient treatment
(N and P). Response curves of the OTUs in each nutrient treatment confirmed that
nitrogen-fixers are preferentially found where roots are less abundant in the sediments
(Figure 22).
Overall, the community composition of nitrogen-fixing bacteria in mangrove
sediments responds to natural changes in the environment (temporal between seasons and
spatial among depths) and is strongly affected by fertilization with N and P.

Figure 22. TRFLP-nifH fingerprints (OTUs) response to the relative abundance of
live roots using generalized additive models. A, OTUs curves in the N-treatment; B,
OTUs curves in the P-treatment.

Live roots (%)
0 50 0 50
Live roots (%)
0 Response 0.1
0 Response 0.3
A B
81

Discussion

N
2
fixation rates in the mangrove sediments in Belize showed moderate
variability under natural nutrient conditions. Depth integrated rates (54.2 ± 4.2 nmol
N/m
2
· h; range: 0 – 204 nmol N/m
2
· h) are comparable but higher in most cases than
rates in other mangrove forests worldwide (see summary for mangroves in Lee and Joye
2006). The N
2
fixation rates in this study showed a wider range but were similar on
average to N
2
fixation rates calculated from microbial mats in the same mangrove area (0
– 50 nmol N/m
2
· h; Lee and Joye 2006).
Under fertilized conditions, depth integrated N
2
fixation rates were lower in the
N-treatment (20.8 ± 4.2 nmol N/m
2
· h; range: 0 – 154 nmol N/m
2
· h) and higher in the
P-treatment (279.2 ± 29.2 nmol N/m
2
· h; range: 42 – 687 nmol N/m
2
· h) compared to the
N
2
fixation rates under natural conditions (Ctrl-treatment). These changes in the N
2

fixation rates with fertilization are expected as NH
4
+
additions inhibit the synthesis of the
nitrogenase enzyme responsible for N
2
fixation, while PO
4
3-
additions increase the
demand for nitrogen (Capone 1988). However, N
2
fixers were still active in the N-
treatment due to the NH
4
+
pore-water concentrations lower than the known inhibiting
ammonium value for N
2
fixers of ~100-500µM (Capone 1988, Yoch and Whiting 1986).
Interestingly, the large amount of urea fertilizer added to the N-treatment (600 g
Urea/year) increased pore-water NH
4
+
concentration by only 43% (which was not
significantly different from the other treatments) and decreased N
2
fixation rates by 60%
relative to the Ctrl-treatment. Moreover, in the N-treatment mangrove trees showed no
82
increase in biomass and had a ~100% increase in root mortality (McKee et al. 2007)
indicating that the excess nitrogen from fertilization with urea is not promoting tree
growth. In this interior mangrove zone, P and not N is the critical limiting nutrient for
plants (Feller et al. 2003, Mckee et al. 2007) and bacteria (see Chapter 3). The fact that
fertilization with urea does not substantially enrich the pore-water pool when it is not the
limiting nutrient for plants, supports the idea that mangroves are good at buffering the
effects of nutrient perturbation by maintaining good water quality (e.g. Ewel et al. 1998,
Twilley 1998). At the same time, however, excess nitrogen may be lost primarily through
the sediment-water interface, possibly via ammonia volatilization, denitrification and/or
during high tidal water exchange (mainly around June, Lee and Joye 2006). It has been
shown that ammonia fluxes can be significant (Fogel et al. 2008) and that denitrification
rates are lower or close to N
2
fixation rates (Lee and Joye 2006) in this mangrove area.
Nevertheless, the excess nitrogen leaving the sediment system may impact other
ecosystems surrounding the mangrove forests and may increase the production of gases
with climatic relevance (e.g. nitrous oxide, ammonia). Fertilizers in coastal areas are
considered one of most important pollutants in coastal environments with deleterious
effects on the nitrogen cycle (Galloway et al. 2004, Vitousek et al. 1997). Recently,
studies in terrestrial environments have shown an increase in the emissions of N
2
O (Lund
et al. 2009) and NH
3
(Milford et al. 2009) from fertilized soils. Therefore, future studies
in mangroves should focus on the export pathways for nitrogen to better quantify the
effect of excess nutrients on the nitrogen cycle in adjacent ecosystems and in the
atmospheric nitrogen chemistry.
83
High N
2
fixation rates in mangroves have been found in sediments associated with
roots, in direct contact with roots and on decomposing leaf litter (Holguin et al 2001),
similar to other systems like seagrasses (Welsh 2000). Plant-microbial interaction in
sediments occurs in the rhizosphere (root-sediment interface), where roots are able to
affect redox conditions and biological activity by supplying large amounts of organic
compounds while depleting inorganic compounds (Bertin et al 2003). This rich labile
carbon transfer from roots to the soil/sediment environment (rhizodeposition) is
composed of secretions, dead cells, mucilage and exudates (Graystone et al. 2007) which
influence microbial activity and therefore, nutrient cycling (e.g. Baudoing et al. 2003).
Previous studies in mangrove forests suggest that the activity of N
2
fixers is largely
influenced by the availability of dissolved organic carbon from roots (Zuberer and Silver
1978, Nedwell et al. 1994, Joye and Lee 2004). In our study, however, N
2
fixation rates
in sediments under natural and P-fertilized conditions may rely on rhizodeposits for
energy source, but a strong dependence on other sources for carbon, for example
decomposing organic matter and not root exudates, primarily occur. Two observations
from our results support this statement.
First, the diel differences of N
2
fixation rates in the P-treatment suggested that
only a fraction of the community is active during the day when the trees are
photosyntetically active. Mangrove trees oxidize their rhizosphere by releasing the
oxygen from the roots into the sediment (Mckee 1993, Mckee et al. 1988). Therefore, the
oxygen released by the roots will have the capacity to oxidize the sediments and affect
microbial processes. The nitrogenase enzyme responsible for N
2
fixation is known to be
inhibited by O
2
, imposing temporal separation of day vs. night activity in non-
84
heterocysteous N
2
fixers (e.g. Gallon 1992). Lee and Joye (2006) previously found in
Belize that mats dominated by heterocysteous cyanobacteria were active during the day
while non-heterocysteous containing mats fixed N
2
at night. In our study a similar diel
pattern was observed only in the P-treatment where nitrogenase activity was the highest,
H
2
S concentrations were lower, and the percentage of live roots was higher than in the
other nutrient treatments. Results indicated that coupling between root activity and N
2

fixation largely depends on the nutrient condition of the sediments, which impose or
overcome plant nutrient limitation. Specifically in the P-treatment, the long-term addition
of phosphorus increased plant biomass and demands for nitrogen, therefore N
2
activity
increased.
Second, in our study the community structure of the N-fixing microbial
populations was remarkably stable across the spatial range studied despite high
variability in environmental parameters. In contrast, 78% of the variability in the
community composition of N
2
fixers was found to be explained by live roots and nutrient
treatments (Figure 21), although with an inverse response of N
2
fixers to live roots
abundance (Figure 22). These results suggest that N
2
fixation in Belize sediments
depends more on decomposing organic matter than on rhizodeposits for energy source. A
positive influence of plants on N
2
fixation rates has been observed in non-peaty
sediments in mangroves (Zuberer and Silver 1978, Sengupta and Chaudhuri 1991,
Nedwell 1994, Sjöling et al. 2005) and seagrasses (e.g. Welsh 2000) where sediments are
typically mineral based and/or low in organic carbon content. In Belize, on the other
hand, the peat sediments have high organic content that decompose slowly independent
of nutrient conditions (McKee et al. 2007). In addition, bacterial communities in the same
85
mangrove zone were mostly limited by nutrients such as P (see Chapter 3). Peat sediment
stability in Belize is known to be controlled largely by subsurface accumulation of
refractory mangrove roots, which depends on the availability of nutrients like phosphorus
in the interior mangrove zone (McKee et al. 2007). If mangrove root production is low,
microorganisms relying mostly on carbon from dead organic matter can potentially affect
the sediment stability of this oceanic mangrove system.
Overall, we presented in this study a more comprehensive ecological analysis of
N
2
fixers in sediments than has previously been conducted in tropical mangrove systems,
indicating that observed differences are largely explained by the availability of nutrients,
H
2
S concentration, and abundance of mangrove roots. In other mangrove sediments, the
community composition of N
2
fixers is influenced positively by organic carbon content
(Zhang et al. 2008) or a combination of high organic matter and low O
2
concentration
(Flores-Mireles et al. 2007). In contrast, we found that in Twin Cays N-fixing
microorganisms, depending on the nutrient conditions of the sediment, can be strongly
influenced by specific environmental parameters (e.g. NH
4
+
, PO
4
3-
, H
2
S) and/or plant
roots. However, a large unexplained variation in the data (about 60% in the temporal and
30% in the spatial variability) suggests other factors may be important in determining the
community composition of this microbial group. We hypothesized that biological factors
(not included in this study) may also affect the activity and community composition of
this microbial group. Interactions between the N
2
fixing bacteria (Holguin and Bashan
1996) and the non-fixing bacteria (Holguin et al. 1992) observed in the rhizosphere of
mangrove sediments, indicate that ecological interactions among different microbial
86
groups and plants may also be an important factor controlling biogeochemical processes
in mangrove sediments.
We conclude that the availability of nutrients is an important factor shaping the
community composition of nitrogen fixers and their activity, providing evidence that the
link between community composition and function (the functionality significance of
community structure of N
2
fixers) may be relevant for communities perturbed from
natural conditions. In addition, fertilization can cause differences in plant-microbial
interactions. As we observed under N and P fertilized conditions, the rhizosphere neither
necessarily harbors a greater diversity and activity nor promotes community stability and
persistence of N
2
fixers over different temporal scales. Further work should test if these
results are particular to N
2
fixers in mangrove peat sediments. This study demonstrates
the importance of evaluating the long-term effects of nutrients on the composition and
activity of microbial communities and their interaction with plants to better understand
the ecological role of microorganisms in dynamic environments such as mangrove
sediments.

87

Chapter 5

PHYLOGENETIC DIVERSITY OF DIAZOTROPHS IN THE
RHIZOSPHERE OF MANGROVE SEDIMENTS

88
5. Phylogenetic Diversity of Diazotrophs in the Rhizosphere of Mangrove Sediments

Abstract

Diazotroph diversity was studied using Terminal Restriction Fragment Length
Polymorphism (TRFLP) and cloning of PCR-amplified nifH gene in the sediment of
Belizean mangrove forests in two microenvironments (roots and bare sediments) located
at different depths in the sediments and subjected to a long-term fertilization with
nitrogen and phosphorus. Also, via in silico analysis we assigned sequenced clones to
TRFLP- nifH phylotypes, and modelled the occurrence of phylotypes in response to
environmental conditions using Generalized Additive Models (GAMs). Phylogenetic
analysis grouped the majority of the clones from yet uncultivated diazotrophs (62%),
some of which are presumably novel (26%). These nifH groups were mostly related to
sulfur cycle bacteria (35%) and poorly related to nifH sequences from other mangrove
areas (12%). In addition, nifH groups were more closely related to those found in
microbial mats (26%) and seagrasses (21%) from distant geographic areas. GAM models
were fitted to 7 TRFLP-OTUs that corresponded with 16 sequenced clones mostly from
presumable sulfate reducers showing specific optimal microenvironmental conditions for
most of the OTUs. Overall, it was shown that mangrove sediments from Belize harbor a
unique diazotrophic community not seen before in mangrove sediments with a low
diversity in metabolic traits and is dominated by sulfate reducers. Nitrogen-fixing sulfate
reducers presented a narrow distribution explained by environmental parameters,
89
primarily by the abundance of dead roots, H
2
S and/or PO
4
-3
concentrations. This study
demonstrated the multifunction importance of diazotrophs in mangrove sediments and the
existence of a variety of microenvironments in sediments. Also, was demonstrated the
significance of identifying the diversity and main factors controlling microbial
populations in marine systems to better understand ecological processes and community
variability in highly dynamic natural systems.

Introduction

Nitrogen fixation is considered one of the most important microbial processes in
mangrove sediments after sulfate reduction, the latter responsible for most of the organic
matter decomposition within the forests (Holguin et al. 2001). Many sulfate-reducing
bacteria are also able to fix nitrogen (e.g. Zuberer and Silver 1978). Therefore the
nitrogen, sulfur, and carbon cycles are tightly coupled during organic matter
mineralization in mangrove sediments, an important interaction of biogeochemical
processes that has not been thoroughly studied in the mangrove ecosystem.
The fixation of N
2
by microorganisms is considered an important process in
bringing nitrogen into mangrove ecosystems. However, in comparison with the large
information obtained from marine aquatic environments (Zehr et al. 2003) and salt
marshes (e.g. Brown et al. 2003, Lovell et al. 2008), the knowledge on the diversity of
diazotrophs through phylogenetic analysis of the nifH gene in mangrove sediments is
limited to a few studies in Mexico (Flores-Mireles et al. 2007) and China (Zhang et al.
90
2008). A more comprehensive understanding of the composition of diazotrophs in coastal
sediments and particularly in mangrove sediments will enhance our knowledge of how
the biodiversity of diazotrophs contributes to microbial community structure and stability
and overall ecosystem functioning.
Most of the information on the diversity of diazotrophs in mangroves comes from
isolation studies indicating diazotrophs are mostly affiliated to Proteobacteria (alpha,
gamma) and Firmicutes groups. Molecular diversity of diazotrophs has shown a similar
diversity of groups but in addition has also showed as well sequences related to the delta
and beta Proteobacteria (Zhang et al. 2008), and the Actinobacteria group (Flores-Mireles
et al. 2007). Nitrogen-fixing bacteria isolated from sediments, rhizosphere and root
surfaces did not show any specificity for a mangrove tree species and were identified as
Azospirillum, Azotobacter, Rhizobium, Clostridium and Klebsiella from various
mangrove species (Sengupta and Chaudhuri 1991). Additionally, Holguin et al. (1992)
isolated from the rhizosphere of different mangrove species other diazotrophs (Vibrio
campbelli, Listonella anguillarum, V. aestuarianus, and Phyllobacterium sp.). The
ecological role of diazotrophs, besides being a source of nitrogen for mangroves, has not
been widely studied, although Ravikumar et al. (2004) found that three species of
Azobacter are important producers of hormones that promote the growth of mangrove
seedlings.
Not only there is little information on the diversity of diazotrophs in mangrove
sediments, little is also known about the environmental factors that regulate their
diversity. This is important especially in coastal areas where fertilizers are considered one
of the most important pollutants in coastal environments, with deleterious effects on the
91
nitrogen cycle (Galloway et al. 2004, Vitousek et al. 1997). Anthropogenic activities have
the potential to increase (e.g. nitrogen fertilizers) or reduce (e.g. organic rich wastes) the
availability of nutrients in coastal and terrestrial environment (e.g. Bürgmann et al. 2003).
The spatial niches for optimal activity of diazotrophs in natural environments remain
largely unknown, and the role of diazotrophs in the biogeochemistry of marine sediments
is still poorly understood.
The present study aimed to characterize the diversity of diazotrophs in mangrove
sediments and to better understand their ecological role in the overall sediment nutrient
cycling. To accomplish this, we compared a nifH clone library with nifH-TRFLP
phylotypes (Terminal restriction fragment length polymorphism), and we modeled
phylotype occurrence in response to environmental conditions to test whether
biogeochemical parameters of the mangrove sediments under natural and fertilization
conditions influence the relative abundance of diazotroph groups. Naturally coastal
environments experience changes over time, and understanding of how microbial
communities respond to natural and disturbance conditions can reveal important
relationships between community composition and whole ecosystem function.

Methods

Environmental samples

92
Root and sediment samples for extraction of the nifH gene were collected from
mangrove areas subjected to a long-term fertilization with nitrogen and phosphorus for
the last 5 years. Samples were collected from the interior of a mangrove forest at Twin
Cays, Belize. A detailed analysis of the spatial and temporal variability of the nifH gene
was previously done (Chapter 4). Samples were collected from the top 30 cm from the
fertilized areas as well as at a control treatment without fertilizer. The roots and sediment
samples (~1.5g) were taken at each sediment depth, fixed with TE (10 mM Tris HCL
plus 1 mM EDTA, pH of 7.4) and frozen (-20
o
C).
Pore-water samples were collected using sippers inserted into the peat sediment at
three depth intervals (0-5, 5-10, 20-30 cm; N=3 per depth interval). Temperature and pH
were determined immediately after the collection of the pore water samples (ORION with
pH and ATC electrode; calibrated with certified standards, reproducibility of ± 0.02 pH
units and ± 1.0
o
C). Salinity was measured with a refractometer (Fisher Scientific,
calibrated with deionized water, with an accuracy of ± 1.0 ppt). Samples for determining
the hydrogen sulfide (H
2
S) content (volume 0.1 ml) were stored in glass vials containing
0.04 ml of 0.05 M zinc acetate and analyzed following the method of Cline (1969). For
ammonium (NH
4
+
) analysis, samples (volume 0.5 ml) were stored in vials containing 1
ml of phenol reagent and analyzed as in Solorzano (1969). Dissolved inorganic phosphate
(PO
4
3-
) was measured colorimetrically using the method of Strickland and Parsons
(1972). A spectrophotometer (Shimadzu UV-1700) was used for pore-water analysis at
the University of Southern California.

93

DNA extraction

DNA was extracted in the root and sediment samples by applying approximately 500 mg
of sample to the FastDNA® SPIN Kit for Soil (MB Biomedicals Laboratories, Solo,
Ohio, USA) according to the manufacturer’s protocol. DNA was eluted in 50ul DNA free
water. The quantity of the extracted DNA was analyzed by PicoGreen (Molecular
Probes, Eugene, OR, USA) following the manufacturer’s instructions, using a
fluorometer (Statagene). DNA elutions were stored at -20
o
C.

PCR amplification

All nested PCR reactions using degenerate primers (Zehr and McReynolds 1989)
were conducted in 50 ul aliquots containing 1 ng template DNA, 1XPCR Buffer, 2.5 mM
MgCl
2
, 0.2 mM dNTPs, 400 ng/ul BSA, 5.0 U Taq, and 0.8 µM for each primer: nifH3
(5’-ATRTTRTTNGCNGCRTA-3’) and nifH4 (5’-TTYTAYGGNAARGGNGG-3’) for
the first reaction, nifH1 (5’-TET-TGYGAYCCNAARGCNGA-3’) and nifH2 (5’-
ANDGCCATCATYTCNCC-3’) for the second reaction. The PCR reactions were carried
out with a denaturation step of 3 minutes at 95
o
C, followed by 30 cycles of 30 s
denaturation at 95
o
C, 30 s annealing at 55
o
C, and 45 s extension at 72
o
C, followed by a
final extension step of 7 min at 72
o
C. All amplification reactions were performed in a
Thermocycler (Stratagene RoboCycler 96 PCR). Extreme care was taken to avoid any
DNA contamination and controls (PCR reaction free of DNA template) were run with the
94
DNA samples. The quality of the PCR products were evaluated by gel electrophoresis,
and the quantity of the extracted DNA was analyzed by PicoGreen (Molecular Probes,
Eugene, OR) following the manufacturer’s instructions, using a fluorometer (Statagene).

Phylogenetic analysis

The nifH PCR products were cloned using the TOPO TA Cloning® Kit for
sequencing following the manufacturer’s instructions (Invitrogen, Carlsbad, CA). Clones
were isolated and sequenced with vector-specific primers by MCLAB (San Francisco,
CA). All nifH sequences obtained were edited to remove non-coding sequences up- and
down-stream from the primers, and primers were removed. The sequences obtained were
compared with nifH sequences from GenBank (October 2008) using blastx. Sequences
were translated into amino acids (Geneious Pro 4.0 software), and aligned together with
the reference sequences from Genbank with ClustalW (MEGA 4.0 Pennsylvania State
University). The reliability of the alignment was estimated as well using MEGA 4.0 (Hall
2008). Neighbor-joining phylogenetic trees were constructed (MEGA 4.0) using the
Poisson distance correction, pairwise deletion of gaps, and 1000 bootstrapping replicates
to estimate the reliability of the tree. Sequences with >98% similarity were not included
in the phylogenetic tree.

95
TRFLP analysis

We used Terminal restriction fragment length polymorphism (TRFLP) of PCR
amplified nifH gene to study the relative composition of nitrogen-fixing microbial
populations (Widmer et al. 1999, Hewson and Fuhrman 2006). The PCR products were
run in an agarose gel (2%) for 2 h and the desired product size (~360-380 bp) was
excised manually with a clean, sharp scalpel. Then, the nifH gene was extracted and
purified with the MinElute Gel Extraction Kit (QIAGEN Laboratories, USA) following
the manufacturer’s protocol. The extracted and purified nifH gene was eluted in 10 µl
DNA free water, quantified (PicoGreen Molecular Probes, Eugene, OR), and digested
with the restriction enzyme HaeIII following the manufacture’s instructions (in
comparison with other enzymes, this enzyme was chosen because previous tests showed
better consistency of peak numbers over a series of replicates). Clean digested products
were run in duplicates in an ABI 3730 Automated Sequencer and the outputs were
aligned against all possible fragment lengths found in both replicates (100-360bp). The
area under each peak was used to calculate the relative abundance (%) of each TRFLP-
nifH phylotype (TRFLP-OTUs; OTU: Operational Taxonomic Unit) in each sample. In
addition, the theoretical sequence length cleaved with HaeIII from the clones obtained
through sequencing was calculated (MEGA 4.0 software) to assign sequenced clones
(98% cutoff) to TRFLP-OTUs.

96
Statistical analysis

The clones obtained were analyzed with the software EstimateS at 92% and 98%
cutoff to know how well the data represent the genetic diversity of the nifH gene in the
Belize mangrove sediments. Accumulation curves of number of sequences observed
(clone-OTUs) and of Chao1 (total richness index) were generated by randomization and
averaging from 50 runs.
The distribution of selected TRFLP-OTUs was analyzed with a One-way
ANOVA with nutrient treatment, and depth treatment factors (JMP Software, SAS
Institute Inc., Cary, NC). When significant differences were found, Tukey’s test was used
to compare the means of the treatments. The level of statistical significance was set to P-
level <0.05. All data were tested to fulfill normality and equal variance assumptions, and
transformations were carried out when necessary.
Modelling of TRFLP-OTU distributions in response to particular environmental
variables was conducted with the CANOCO software (Microcomputer Power, Ithaca,
N.Y.) by applying Generalized Additive Models (GAMs, Hastie and Tibshirani 1986).
The percent abundance of TRFLP-OTUs was transformed (log (x+1)), and the
environmental data was centered and standardized. A partial Redundancy Analysis
(RDA, Ter Braak and Prentice 1988) was conducted and the scores of individual samples
on the first RDA axis were used as the explanatory variable in the final GAM analysis,
for modelling the TRFLP-OTUs relative to the sum of all independent environmental
variables analyzed in this study. A Monte Carlo permutation full model test was
97
conducted to evaluate the significance of the first axis generated by the RDA analysis
with 499 unrestricted permutations.

Results

The total number of nifH clones sequenced from Belize mangrove sediments and
the number of different nifH amino acid sequences within each phylogenetic group are
shown in Table 6. A total of 154 nifH clones were sequenced and are related to sequences
from Delta-, Gamma- Proteobacteria, Firmicutes, and Green sulfur phylogenetic groups.
Subgroup clones showed no specificity for a nutrient treatment, except D4-subgroup and
D10 subgroup (with 8 and 10 total number of sequences, respectively) only found in the
nitrogen-fertilized treatment (Table 6). The comparison between the clones obtained from
this study and GenBank clones revealed that Belize mangrove sediments host multiple
yet unidentified nifH sequences (49%). After taking into account identical amino acid
sequences (98% cutoff), there were 72 different sequences obtained in this study. A
Neighbor-joining tree containing the 72 nifH different sequences and the GenBank
sequences is shown in Figure 23. The phylogenetic tree was divided into four groups and
is described in detail below.
Group A: Only one clone falls into group A of the nifH phylogenetic tree
(Figure 23). Clone RN26 is most closely related (99% amino acid similarity) to an
isololate from Italian soils (AAS47809) and largely differed from the Archaea group
(only 87% amino acid similarity).
98
Table 6. Distribution of mangrove sediment nifH clones among phylogenetic groups.

99

Unidentified-N053 ABD74331
RN212
SP28
RSC15
RN12
RN28
SP29
Chlorobium phaeov ibrioides YP 001130857
Chlorobium limicola YP 001942747
Chloroherpeton thalassium YP 001995946
Prosthecochloris aestuarii YP 002016292
Chlorobium phaeobacteroides YP 001960150
SP25
Unidentified-B3 AAF61027
Unidentified-SY7 ABM67091
SP31
RN36
RN18
Unidentified-SB7 AAL07952
RN33
RN314
RN37
RN313
Unidentified-J7 AAA65422
RC36
SP26
RN211
RN15
RC31
RN310
RN27
RN22
RC21
RC35
RN39
RN111
RN23
RN13
RN14
RN110
Unidentified-H05 DNA E4 ABQ50824
RN213
SP27
RN32
SP23
RC34
RC23
Desulfatibacillum alkeniv orans YP 002430688
RN25
RN312
RN112
RN38
RN24
Unidentified-nif1003301L04 AAZ77144
RC13
SRP13
RC11
RN19
SRP11
Unidentified-F59FN ACI32206
SP34
RN21
Unidentified-55 AAA65425
RC33
SP210
RC32
Unidentified-OilUp10330 AAY85458
Unidentified-S 1A1 AAS57673
SP21
SP15
RN35
RC14
Desulfov ibrio v ulgaris YP 002437020
Desulfov ibrio v ulgaris YP 009055
Desulfov ibrio salexigens ZP 03466626
Desulfomicrobium baculatum ZP 04344086
SP22
Unidentified-47 AAA65429
RN17
RC22
Crocosphaera watsonii ZP 00516386
Nostoc punctiforme YP 001864137
Nodularia spumigena ZP 01629115
Myxosarcina sp. MSU73133
Lyngbya lagerheimii UTEX 1930
Rhodopseudomonas palustris NP 946726
Azospirillum brasilense AZSFEFEMO
Methylobacterium sp. 4-46 YP 001770351
Beijerinckia indica YP 001831615
Sphingomonas azotifigens BAE71134
X anthobacter autotrophicus YP 001415059
Bradyrhizobium sp. ORS278 YP 001207288
Unidentified-F30 CF2 ACI32162
Unidentified-c1-HW6 AAO48684
SRP32
Unidentified-pCPS211 ACC95201
Unidentified-T3t047 ACI26001
SP14
RN311
Desulfitobacterium hafniense YP 520504
Heliobacterium modesticaldum YP 001679706
Heliobacterium chlorum BAD80878
Vibrio natriegens AAD55588
Vibrio diazotrophicus AAF17302
Klebsiella pneumoniae AAO85881
RC12
Pelobacter carbinolicus YP 357508
SRP211
RN31
SP24
Geobacter uraniireducens YP 001229952
Unidentified-SY25 ABM92894
RN210
SP33
SP12
Geobacter sulfurreducens NP 953865
Ectothiorhodospira shaposhnikov ii ABN10974
Thiocapsa roseopersicina ACC95826
Methylomonas methanica AAK97417
Unidentified-PriUp10313 AAY85422
RN29
RN34
RN16
Unidentified-A37 AAS47809
RN26
Methanococcus maripaludis MMU75887
Methanothermobacter thermautotrophicu...
Synechococcus elongatus X 67694
99
98
53
42
91
76
85
59
80
73
43
40
58
42
66
41
52
48
62
97
79
81
77
98
93
60
83
69
52
45
47
94
47
82
56
53
94
48
40
61
65
83
58
56
0.2
D
C
B
A
E
α –
Proteobacter
δ- and γ- Proteobacter
Archaea
Green
Sulfur
δ-Proteobacter
Cyano-
bacteria
Firmicutes
δ-Proteobacter
Figure 23. Phylogenetic tree
of nifH gene based on amino
acid sequences from Belize
mangrove sediments (in
bold). Bootstrap values
(≥40%) for 1000 replicates
are indicated above the
branches and Synechococcus
elongatus was used as
outgroup.
100
Group B: Thirteen different mangrove sediment nifH clones fall within group B,
which includes gamma, delta proteobacteria, and Firmicutes. Clones RN31 and SRP211
are most closely related (97% and 96% amino acid similarity, respectively) to
Paleobacter carbinolicus DSM 2380, an iron and sulfur-reducing anaerobic organism
found in marine and freshwater environments (Mussmann et al. 2005). Clone RC12 and
SP24 are most closely related (both 94% amino acid similarity) to facultative anaerobes
such as Vibrio natriegens and Klebsiella pneumoniae. Clones SP12 and SP33 are closely
related (93% and 94% amino acid similarity, respectively) to a facultative anaerobe
Geobacter sulfurreducens (Methe et al. 2003), a metal and sulfur reducer. Clones RN34
and RN29 are similar (both 99%) to a clone (AAY85422) from a pristine marine
environment (Musat et al. 2006). Clone RN16 is closely related (93% amino acid
similarity) to Thiocapsa roseopersicina an anaerobic phototrophic sulfur reducer
bacterium. A subgroup containing three mangrove sediment clones (SRP32, SP14,
RN311) is closely related to soil clones with unknown phylogenetic affiliation, that are
close (~86%) as well to the Firmicutes group in the nifH phylogenetic tree. Clone SP14 is
close (94%) to a clone from a Malaysian soil (ACC95201). Clone SRP32 is most closely
related (93% amino acid similarity) to a clone from forest soils in the Colombian Amazon
region (ACI32162). Clone RN311 is close (91%) to a clone obtained from soils in New
York (ACI26001).
Group C: Seventeen clones fall into group C including bacteria with
phylogenetic affiliation to Delta Proteobacteria, most related (14 clones, 87-91% amino
acid similarity) to a group of sulfate reducers. Clones SRP11, RC13, RC11, SRP13 and
RN19 are 93-97% close to a clone (AAZ77144) from microbial mats located in the
101
Bahamas (Yannarell et al. 2006). Clones SP34, RN21, and SP210 are 90-92% close to a
clone (AC132206) from forest soils in the Colombian Amazon region. Clone RC33 is
88% close to a clone from a microbial mat in North Carolina (Zehr at al. 1995). Clones
RC32 and SP21 are 90-93% closely related to a clone (AAY85458) from a microbial mat
(Musat et al. 2006) and to a clone (AAS57673) from standing dead stems of Spartina
alterniflora (Moissander et al. 2005). Clone RC14 is most closely related (90% amino
acid similarity) to Desulfovibrio vulgaris, a sulfate and metal reducer (Heidelberg et al.
2004). Clone RN17 is closely related (99% amino acid similarity) to a clone
(AAA65429) from a microbial mat in North Carolina (Zehr at al. 1995). Clone SP22 is
close (92%) to Desulfomicrobium baculatum a sulfate reducer. Clones SP15, RN35 and
RC22 grouped separately and may represent novel sequences.
Group D: Thirty-five clones fall into group D, which includes Delta
proteobacteria. Clones SP31 and SP25 are closely related (90 and 92%, respectively) to a
clone from mangrove sediments in China (Zhang et al. 2008). Clones RN36, RN18,
RN33, RN314, RN37, and RN313 are close (89-96%) to a clone from seagrass sediments
in the Bahamas (Bagwell et al. 2002). Clones SP26 and RC36 are close (97% amino acid
similarity) to a clone (AAA65422) from a microbial mat in North Carolina (Zehr et al.
1995). Clones SP27, RN32, and RN213 are 93-94% close to a clone (ABQ50824) from
the eastern Mediterranean Sea. Clones RC23, RN312, SP23, RN112, RN25, and RC34
are most closely related (90-97% amino acid similarity) to Desulfatibacillum
alkenivorans AK-01, a sulfate reducer capable of utilizing C13-C18 alkanes as growth
substrates. Clones RN211, RN15, RC31, RN310, RN27, RN22, RC21, RC35, RN39,
102
RN111, RN23, RN13, RN14 and RN110, as well as clones RN38 and RN24 grouped
together and may represent novel sequences.
Group E: Six clones fall into group E, all related (90-92% amino acid similarity)
to a group of green sulfur bacteria. Clones SP28, RSC15, RN12 and RN212 are most
closely related (96-97%) to a clone (ABD74331) from the rhizosphere of the smooth
cordgrass Spartina alterniflora (Lovell et al. 2008). Clones RN28 and SP29 are more
closely related (87-89%) to clones from the genus Chlorobium, a phototrophic sulfur
oxidizer group.
The accumulation curves of clone-OTUs observed versus sampling effort at two
different cutoff levels (92% and 98%) were clearly different (Figure 24a), indicating
differences in the relative abundance of defined OTUs between the two cutoff levels.
However, the Chao1 index for total species richness showed no significant difference
between the two cutoff levels (Figure 24b). It is important to note that although the two
cutoff levels are similar using the Chao1 index, the curves tend to separate at the highest
number of clones sampled indicating that some rare species may be detected only at the
98% cutoff. In Figure 24b the 98% cutoff curve flattens out at 100 clone-OTUs estimated
and completely stabilizes at 120 clone-OTUs estimated. Using the Chao1 index, this
study covers approximately 75-88% of the total species richness of the nifH gene in
mangrove sediments from Belize. Overall, the phylogenetic analysis in Belize mangrove
sediments showed multiple different nifH sequences mostly related to clones potentially
involved in the sulfur cycle (34%); only 12% of the nifH sequences from this study are
closely related to unidentified nifH sequences found in other mangrove areas, and 26% of
the nifH sequences are novel (Figure 25).
103

Figure 25. Relative abundance of nifH clones sequenced from Belize mangrove sediments
among microbial processes (A) and natural environments (B).

Figure 24. Accumulation curves
for observed and estimated
clone-OTUs (A), and Chao1
index of richness estimated as a
function of sample size (B) of
diazotrophs in mangrove
sediments from Belize. Curves
are averaged over 50 simulations
using the computer program
EstimateS. Error bars are 95%
CIs.
B
0 20 40 60 80 100 120 140 160
400

300

200

100

0
Number Clones Sampled
OTUs
estimated
A • cutoff 98%
 cutoff 92%
100

80

60

40

20

0
OTUs
observed
B
Saltmarsh
12%
Mangroves
12%
Seagrass
21%
Unknown
9%
Soils
12%
Microbial
mats
26%
Open
ocean
8%
A
Unknown
62%
Sulfate
reducers
28%
Phototrophic sulfur
oxidizers
1%
Sulfur and
metal
reducers
5%
Phototrophic sulfur
reducers
1%
Phototrophic
3%
104
We found 21 sequenced clones (clone-OTUs, 98% cutoff) that corresponded with
11 TRFLP-OTUs by in silico calculation of the TRFLP theoretical length of sequenced
clones. Only in some cases a specific clone corresponded to a specific TRFLP-OTU
(Table 7), and just two clones corresponded to dominant TRFLP-OTUs in the nifH
community (Figure 26). Four of these TRFLP-OTUs (153, 321, 227, 117) were found
only once in all the samples collected and therefore were not included in the statistical
analysis. Most of the TRFLP-OTUs that corresponded with the sequences are related to
putative sulfate reducers, except for OTU-155 (corresponded sequences related to sulfate
reducers and sulfur oxidizers), and OTU-162 (corresponded sequences related to sulfate
reducers and anaerobic phototrophic bacteria).
The distribution of 7 of the TRFLP-OTUs corresponding to the sequenced clones
showed that only the most dominant in the community presented significant differences
between the nutrient treatments (N, P, Ctrl) and depth intervals (0-5, 5-10, 20-30 cm)
from where the samples were collected. The whole nifH community (all OTUs) also
presented differences among the nutrient treatments, with ~50% similarity between the
Ctrl-treatment and the fertilized treatments (Table 9). OTU-154 peak is higher (p<0.05)
in the Ctrl-treatment and at the surface of the sediments, while the OTU-317 is only
significantly higher at 5-10 cm depth (Figure 27). Environmental parameters among and
within the nutrient treatments are variable (Table 8), indicating possible
microenvironments for the microorganisms inhabiting these mangrove sediments. A more
detailed distribution of the seven TRFLP- OTUs by GAM analysis in response to
particular environmental variables was conducted, showing that the dominant and rare
OTUs have significant non-linear correlations with some of the environmental variables
105
Table 7. Presumptive microbial processes of TRFLP-OTUs related with clone-OTUs. Similarity
(%) based on amino acid sequences. Presumable microbial processes based on phylogenetic
affiliation at the group level from the phylogenetic tree (Figure 23).

Figure 26. Relative abundance (%) of each TRFLP-OTUs detected in mangrove sediments of
Belize. Numbers indicate the TRFLP-OTUs found to correspond with clone-OTUs.

TRFLP-
OTUs
Clone
ID
Total
clones
Presumable
microbial process
Match
Accesion
number
%
similarity
181 SP22 2 Sulfate reducer Desulfomicrobium bacalatum ZP04344086 92
155 RN12 1 Sulfur oxidizer Unidentified from S. alterniflora ABD74331 89
RN28 1 Sulfur oxidizer Chlorobium thalassium YP001995946 89
SRP13 2 Sulfate reducer Unidentified from microbial mats AAZ77144 97
RC11 3 Sulfate reducer Unidentified from microbial mats AAZ77144 97
RN211 1 Sulfate reducer
154 RN21 1 Sulfate reducer Unidentified from soils AC132206 90
RC14 1 Sulfate reducer Desulfovibrio vulgaris YP002437020 90
RN111 1 Sulfate reducer
RN18 7 Sulfate reducer Unidentified from seagrass AAL07952 89
SRP11 2 Sulfate reducer Unidentified from microbial mats AAZ77144 93
153 RC35 2 Sulfate reducer
321 RN213 2 Sulfate reducer Unidentified from Eastern Med. Sea ABQ50824 97
RN110 2 Sulfate reducer
227 SP33 1 Sulfate reducer Geobacter sulfurreducens NP953865 94
117 RN17 1 Sulfate reducer Unidentified from microbial mats AAA65429 99
315 SP210 1 Sulfate reducer Unidentified from soils AC132206 88
317 RC31 1 Sulfate reducer
162 RC32 1 Sulfate reducer Unidentified from microbial mats AAY85458 92
SRP32 2 Phototrophic Unidentified from soils ACI32162 93
165 RN36 2 Sulfate reducer Unidentified from seagrass AAL07952 90
clone-OTUs Closest match in GenBank
TRFLP-OTUs
154
317
181
155
315
165 162

15

10

5

0
Abundance (%)
106
measured (Figure 28). The relative abundance of the selected OTUs vs the environmental
parameters is also shown in Figure 29. Figure 30 shows the response curves of the seven
TRFLP-OTUs fitted using the values of Axis 1 from the results of the pRDA analysis
(Table 10). OTUs 154 and 181 were found to have the highest response in
microenvironments where high abundance of live roots and low concentrations of H
2
S
are found, although these two OTUs were found in root and sediment (OTU-154) or only
in sediment samples (OTU-181). In contrast, OTUs 165 and 162 have the highest
response under high concentrations of PO
4
-3
and H
2
S, both OTUs observed in all nutrient
treatments. OTU-155 has the highest response at high abundance of dead roots, high H
2
S
concentration and low NH
4
+
concentration, although it was only observed in samples
from the N-fertilized treatment. OTU-315 presented a different distribution, with the
highest response at high pH and dead roots. OTU-317 presented a relatively constant
distribution along Axis 1 with no significant response to any of the environmental
parameters measured in this study.

Figure 27. Variability of
selected TRFLP-OTUs
among nutrient treatments
(A, B), and depth intervals
(C, D). Data shown as Avg.
± SE. N≥3.
N
154 (%) 317 (%)
20

15

10

5

0
Ctrl P
20

15

10

5

0
0-5 20-30 5-10
A
B
C
D
Nutrient treatments Depth intervals (cm)
107

Table 8. Variability of environmental parameters at three depth intervals in each nutrient treatment (Ctrl: control; N: Fertilized
with nitrogen; P: Fertilized with phosphorus). See text for details on the fertilization experiment. Data shown as Avg. ± SE; N≥3.

108

Fertilized Treatment
Depth
(cm)
N P
0-1 53.5 55.5
5-10 63.6 58.3
20-30 49.1 50.2
Parameters 1 2
Eigenvalues 0.135 0.12
Species-environment
correlations
0.881 0.848
Correlation coefficients
Dead roots 0.51 -0.23
Live roots -0.47 0.02
Temperature 0.56 -0.46
pH 0.56 -0.48
Salinity 0.54 -0.52
NH4 0.44 -0.44
H2S 0.5 -0.4
PO4 0.07 -0.36
N:P (molar) 0.18 -0.16
Axes
Table 10. pRDA results for the spatial variation in
the TRFLP-nifH OTUs explained by environmental
parameters in the mangrove sediments.
Table 9. Bray-Curtis similarity index (%)
between the Ctrl-treatment and the fertilized
treatments (N and P) based on TRFLP-nifH
OTUs relative abundance.
109

Figure 28. Response curves of seven TRFLP-OTUs to individual environmental parameters,
fitted using generalized additive models (GAM) with a Poisson distribution using log link
function, df = 3.

110

Abundance (%) Abundance (%) Abundance (%) Abundance (%) Abundance (%)
Salinity (ppt)
pH Temperature (
o
C)
N/P Phosphate (µM)
Ammonium (µM) Hydrogen sulfide (µM)
Live roots (%) Dead roots (%)
Figure 29. Relative abundance
of selected TRFLP-OTUs vs
environmental parameters in
mangrove sediments.
111

Figure 30. Response curves of seven TRFLP-OTUs to Axis 1 generated by
pRDA analysis. Data fitted using generalized additive models (GAM) with a
Poisson distribution using log link function, df = 3. pRDA: Axis 1 Eigenvalue =
12.3%, P=0.01, model explained 56% of OTUs variability). Environmental
parameters with a significant effect (p<0.05) on the response of OTUs are
shown in the table (top). D.R.: Dead roots; L.R.: Live roots. OTU-317 not
significant to any parameter.

154 162 181 315 165 155
0.1mM H
2
S D.R. 5% 38.5 ppt 36.0 ppt 32.0
o
C 38.5
o
C
D.R. 10% 250uM PO
4
-3
pH ~6.4 pH ~6.8 D.R. 6% 42.5 ppt
N/P ~15 D.R. 38% N/P ~12 pH ~6.6
20uM NH
4
L.R. 15% 1.0mM H
2
S D.R. 30%
0.1mM H
2
S N/P ~45 260uM PO
4
-3
L.R. 10%
40uM NH
4
4.0mM H
2
S
~0.0uM NH
4

154 162 181 315 165 155
0.1mM H
2
S D.R. 5% 38.5 ppt 36.0 ppt 32.0
o
C 38.5
o
C
D.R. 10% 250uM PO
4
-3
pH ~6.4 pH ~6.8 D.R. 6% 42.5 ppt
N/P ~15 D.R. 38% N/P ~12 pH ~6.6
20uM NH
4
L.R. 15% 1.0mM H
2
S D.R. 30%
0.1mM H
2
S N/P ~45 260uM PO
4
-3
L.R. 10%
40uM NH
4
4.0mM H
2
S
~0.0uM NH
4

154
Axis 1
Response
0
0.2 -
-2 2
181
317
315
165
155
162
112
Discussion

The phylogenetic analysis from this study shows that we have little information
about the identity and ecology of the microorganisms potentially responsible for N
2

fixation in mangrove sediments and that the molecular diversity of N
2
fixers in this
environment is high. In addition, the results suggest that the community of diazotrophs is
composed of microorganisms with a low diversity in metabolic traits dominated by
sulfate reducers, in which microorganisms present narrow spatial distribution explained
by the variability of certain environmental parameters.
The nifH sequences from this study are closely related with sequences from
different environments from distinct geographic areas such as mangrove sediments,
terrestrial soils, microbial mats, open ocean, seagrasses, saltmarsh sediments, and marine
mudflats. Only 12% of the sequenced clones obtained in this study are closely related to
clones from other mangrove areas located in China (Sanya Mangrove Natural Reserve;
Zhang et al. 2008). None of the clones sequenced in this study were close to clones from
a second study in mangroves located in Mexico (Balandra Bay, Pacific ocean; Flores-
Mireles et al. 2007). These two mangrove areas are very different in nifH composition
from the mangroves in Belize. Belize mangrove sediments seem to lack representatives
from the alpha-preoteobacteria group, and the overall nifH community composition is
different with a high representation of N
2
fixers involved in the sedimentary sulfur cycle.
Based on the total number of sequenced clones, only 17% are ≥92% similar to other
identified nifH sequences, 54% are ≥92% similar to unidentified clones in the GenBank
113

database, 26% of the clones are suspected to be novel, and 28% are potentially sulfate
reducers. These results demonstrate an even higher diversity of potentially diazotrophs
than reported previously in mangrove sediments (Zhang et al. 2008) mostly related to
bacteria with the potential to reduce sulfate.
In coastal marine sediments, many sulfate reducers have the ability to fix N
2
and
their importance in the sedimentary cycling of nitrogen, carbon and sulfur in coastal
marine sediments is widely recognized (Nedwell and Azis 1980, see review by Welsh
2000). In mangroves, it has been observed that sulfate reducers are not only important
players for the degradation of the organic matter but also for production of soluble iron
and phosphorus (Sherman et al. 1998) and in bringing new N into the system through N
2

fixation (Zuberer and Silver 1978). In Belize mangroves, the dominant microbial-
mediated process for organic matter oxidation in sediments is sulfate reduction (Lee et al.
2008) as also reported for other mangrove areas (Alongi et al. 1999, Sherman et al.
1998). It appears that the biogeochemical signatures observed previously in Belize are
unique for mangroves (Lee et al. 2008) with high rates of N
2
fixation (Lee and Joye 2006,
and Chapter 2 of this thesis) and sulfate reduction (Lee et al. 2008). The distinct potential
community of N
2
fixers observed in this study through the phylogenetic analysis of the
nifH gene, supports the notion of unique biogeochemical and metabolic signatures in
Belize mangrove sediments.
The statistical analysis of the relationship between the identified groups of
nitrogen-fixing sulfate reducers and environmental parameters in mangrove sediments
was shown using GAM models. This study reports for the first time the application of
114
such models to elucidate the microbial ecology of specific functional traits in marine
sediments. Niche-based species distribution models (Guisan and Zimmermann, 2000)
have been applied to a wide range of plants and macroorganisms (see references in
Mcgill et al. 2006) and consist of observations of species over a gradient of
environmental parameters (or “predictors”) that can have a direct or indirect effect on the
establishment or survival of the species at a limited time period. GAM models assume
pseudo-equilibrium between the environmental parameters and the species spatial pattern,
therefore underlies the realized niche for the species studied (Guisan and Zimmerman
2000). GAM models from Belize demonstrated how nitrogen-fixing sulfate reducers
(from a group of OTUs) are affected directly by specific environmental parameters
(unimodal distribution, 56% explained variance of the fitted data, p<0.05). In contrast,
specific OTUs like 317 showed indirect or no relationship with the parameters measured
in this study. Overall, these results show the optimal microenvironment conditions for the
OTUs representing nitrogen-fixing sulfate reducers.
The fact that the OTUs were observed in a wider range of conditions than their
ideal microenvironment condition indicates that other factors (e.g. biotic factors such as
competition) shift microorganisms from their optima microenvironment and determines
their distribution in sediments. As an example, OTUs 154 and 181 are relatively more
abundant in conditions with higher abundance of live roots and lower concentrations of
H
2
S, but were found in live root samples as well as in sediment samples (Figure 28).
Therefore, it is important to determine the distribution of microorganisms in natural
environments and to examine the interaction between microorganisms and their
environment to better understand the optimal environmental conditions that enhance their
115
presence in nature and that determines their temporal variability over seasonal or sporadic
events. Also, this study demonstrates for the first time that sulfate reducers in natural
environments can be constrained by pore-water parameters and not necessary primarily
by the availability of organic compounds (e.g. Alongi et al. 2000b). In Belize, no
correlation was observed between SO
4
2-
depletion and dissolved organic carbon
concentrations (Lee et al. 2008) supporting our findings that sulfate reducers are not only
affected by the availability of organic carbon but as well by specific pore-water
parameters.
In summary, we found that Belize harbors a high molecular diversity of nifH
genes with a low diversity in metabolic traits influenced by specific environmental
parameters. The high number of microenvironments or niches observed for the nitrogen-
fixing sulfate reducers reveals the complexity of microbial communities even at specific
functional groups with varied roles in sediments. The fact that even rare OTUs presented
a high response to specific environmental parameters (Figure 28) suggests that rare
species often observed in microbial communities (e.g. Hughes et al. 2001) may play an
important role in the environment when disturbed conditions approach their optimal
environmental conditions. Therefore, the role of rare species on the stability and
performance of the community may be important, as observed previously in other
ecosystems (Lyons et al. 2005). The “species” redundancy observed at the OTUs level in
Belize suggests a high biodiversity of microorganisms ensuring the activity of important
biogeochemical processes for the well being of the whole ecosystem (e.g. nitrogen
fixation, organic matter oxidation mediated by sulfate reduction). Overall, this study
demonstrates the importance of identifying the diversity and the principal environmental
116
factors controlling microbial populations in marine systems to better understand the role
of microbial diversity and multifunctionality in highly dynamic natural systems.

117

Chapter 6

SYNTHESIS

118
6. Synthesis

The bioavailability of nutrients is widely recognized as one of the most important
factors controlling ecological processes in mangrove ecosystems (e.g. Feller et al. 1999,
2003; Cheeseman and Lovelock 2004). The extent to which nutrients affect the diversity
and community composition of microbial populations and their interaction with
mangrove trees and the geochemistry of the sediment has not been fully studied and was
the main goal of this research. Specifically, this research focused on (1) discerning spatial
and/or temporal patterns of microorganisms; (2) determining what primarily affects the
variability of microorganisms; and (3) establishing how long-term fertilization with
nitrogen and phosphorus alter the distribution and metabolic pathways of microorganisms
in highly dynamic environments like mangrove forests (conceptual model; Figure 32).
Under natural conditions in the mangroves of Belize, bacterial community
composition based on PLFA was found primarily constrained by reduced conditions in
the sediments of each mangrove zone and not strongly by live roots (RDA analysis
explained 70% of the variance of PLFA data; Chapter 3; Figure 11a). Contrasting
conditions between the mangrove zones were observed with higher H
2
S concentrations in
the interior zone than in the fringe zone (Chapter 3; Figure 5). These two mangrove zones
also differed with higher pore-water temperatures, greater relative abundance of dead
roots, lower pH, and lower relative abundance of live roots in the interior zone than in the
fringe zone. Fertilization with N and P was less important in explaining the variability of
the bacterial PLFA composition (Chapter 3; Figure 11b). In contrast, fertilization
119
treatments were found to enhance bacterial biomass, indicating nutrient limitation
primarily by N in the fringe zone and by P in the interior zone (Chapter 3; Figure 9).
These results suggest that bacteria follow the major trend in nutrient limitation of
mangrove trees (Feller et al. 1999, 2002) and that this similar spatial pattern in nutrient
limitation is likely a consequence of nutrient source availability. Comparisons of stable
carbon δ
13
C signatures

Figure 32. Conceptual model of principal factors affecting the variability
of bacterial populations in mangrove sediments.

120
of bacterial PLFA biomarkers with potential carbon sources suggested that bacteria in
this mangrove system primarily use in situ mangrove tissues as carbon sources for
mineralization. However, under fertilization conditions at both mangrove zones a shift in
the isotopic composition of δ
13
C-PLFA toward more negative values was observed,
indicating an increase in anaerobic respiration but only where fertilization increased
nutrient limitation conditions for plants and bacteria (Chapter 3; Figure 11). Plant-
bacterial interactions were not found to affect the composition, biomass and carbon
utilization patterns of bacteria in sediments. Overall, it was shown that fertilization of
mangrove sediments has a minor effect on bacterial community composition, but can
strongly influence metabolic processes of bacterial groups.
Distinct patterns of N
2
fixation rates in the interior mangrove zone were observed
in the unfertilized areas where rates followed the distribution of live roots in the
sediments and remained relatively unchanged over short (diel, 10 days period) and long
(seasons) periods of time. In contrast, high N
2
fixation rates in the P-treatment followed
the distribution of live roots but with strong short temporal changes (diel). Low N
2

fixation rates in the N-treatment were relatively constant over the spatial and temporal
scales studied (Chapter 4; Figure 17 and 18). Seasonal variability of TRFLP-nifH (OTUs)
was primarily found at the Ctrl- and P-treatment, and was explained by changes in
nutrient and H
2
S concentrations (pCCA analysis explained 42% of the variance of the
fitted OTUs data, p<0.05; Chapter 4; Figure 20). In contrast, spatial variability of OTUs
was principally found at the N- and P-treatment, and was explained primarily by the
distribution of mangrove live roots (pCCA analysis explained 78% of the variance of the
fitted OTUs data, p<0.05; Chapter 4; Figure 21). Response curves of the OTUs in the N-
121
and P-treatment showed that N
2
fixers are preferentially found where roots are less
abundant in the sediments (Chapter 4; Figure 22). In both cases spatial and temporal
scales a distinct composition of OTUs was observed for each nutrient treatment. Overall,
nitrogen-fixing microbial populations in the rhizosphere of Belizean mangrove sediments
showed changes over different spatial and temporal scales not observed before in marine
environments (e.g. Piceno and Lovell 2000a, b). This indicates that this microbial group
is dynamic and distributed in the sediment by specific environmental parameters. The
interaction of plant and N
2
fixers in the rhizosphere of mangrove sediments is strong and
more likely to be constrained by the release of oxygen through the roots into the
surrounding sediment than by root exudation of organic carbon as seen previously (e.g.
Zuberer and Silver 1978, Ravikumar et al. 2004). Fertilization induced patterns that were
not observed under natural conditions, indicating that a link between community
composition and function may be relevant in microbial communities disturbed from
natural conditions.
The phylogenetic diversity of diazotrophs in the rhizosphere of mangrove
sediments showed that the majority of the clones sequenced from Belize are from yet
uncultivated (62%) and some presumably novel (26%) diazotrophs (Chapter 5; Figure
23). These nifH groups were mostly related to the sulfur cycle (35%) and poorly related
(12%) to nifH sequences from other mangrove areas (Flores-Mireles et al. 2007, Zhang et
al. 2008). In addition, nifH groups were more closely related to those found in microbial
mats (26%) and seagrasses (21%) from distant geographic areas. Therefore, mangrove
sediments from Belize harbor a unique diazotrophic community not seen before in
mangrove sediments and is composed by a low metabolic diversity dominated by sulfate
122
reducers. In addition, nitrogen-fixing sulfate reducers in Belize were found to present a
distribution explained by environmental parameters, primarily by the abundance of dead
roots, H
2
S and/or PO
4
-3
concentrations (Chapter 5; Figure 28, 29). Most of the clones
were not specific either to a nutrient treatment or to a substrate for colonization (e.g.
roots). Therefore, plant-microbial interactions do not strongly shape the phylogenetic
diversity of N
2
fixers. The highly functional redundancy of microorganisms in mangrove
sediments is likely an ecological strategy to maintain important biogeochemical processes
such as N
2
fixation (this study, Lee and Joye 2006) and sulfate reduction (Lee et al. 2008)
in the mangrove forest. Nitrogen-fixing sulfate reducers presented optimal conditions for
their distribution under distinct environmental conditions. These results support the
“Insurance hypothesis” (Yachi and Loreau 1999) that multiple microorganisms with the
same function maintain the activity of that function in an ecosystem despite spatial and
temporal variability. This study demonstrates the importance of diazotroph multifunction
in mangrove sediments and the existence of a variety of microenvironments in sediments
that supports the high diversity of diazotrophs. Also, it demonstrated the significance of
identifying the diversity and main factors controlling microbial populations in marine
systems to better understand ecological processes and community variability in highly
dynamic natural systems.
In summary, the functional relationship among microorganisms, plants and
sediment geochemistry covered in this research supported the hypothesis that in carbon-
rich sediments bacteria rely primarily on the predominant primary producer for carbon
(Bouillon et al. 2005, Bouillon and Boschker 2006), but did not support the hypothesis
that roots confer a microenvironment that promotes stability and persistence in microbial
123
populations (e.g. Piceno et al. 1999, Piceno and Lovell 2000b). We were also able to
show that bacteria and plants can be limited by the same nutrients. For the first time, we
were able to demonstrate that in mangrove sediments fertilization with N or P can affect
bacterial metabolic processes. Particularly for nitrogen-fixing bacteria, this functional
group was found to be important not only in providing nitrogen and carbon (through
photosynthesis; Joye and Lee 2004) to the ecosystem, but also in potentially having a
significant role through sulfate reduction in the sedimentary anaerobic respiration.
General patterns of plant-bacterial interactions were different in Belize than in other
mangrove forests (Zuberer and Silver 1978, Holguin et al. 1992, Holguin et al. 2001) and
in other ecosystems like seagrasses (Welsh 2000, McGlathery 2008) and saltmarshes
(Bagwell and Lovell 2000, Piceno and Lowell 2000b). In Belize, mangrove trees strongly
affect the activity and community composition of N
2
fixers, but not the whole bacterial
community (based on PLFAs) or taxonomic traits (based on phylogenetic analysis),
depending on the nutrient condition of the sediments (natural or fertilized with N or P). In
most cases the abundance of roots was inversely related to N
2
fixation rates and nitrogen-
fixers, primarily under fertilized conditions.
The patterns observed in this study revealed important trends in the ecology of
microorganisms in shallow sediment environments. Our findings are in contrast to
theories on stable microbial populations over temporal and spatial scales. Microbial
communities in Belizean mangroves are extremely dynamic which helps support a
functionally stable ecosystem. Population redundancy (e.g. nitrogen-fixing sulfate
reducers) indicates a large variety of microniches determined primarily by a combination
of environmental factors and possible less by competitive interactions. Therefore,
124
functional stability does not necessary imply microbial community stability (Fernandez et
al. 1999, Tilman et al. 2006). The large diversity of rare populations (less abundant
OTUs) and its distribution (microniche) contributed largely to the dynamics observed
indicating the importance of rare species as potential keystone species (Lyons et al. 2005)
under changing environmental conditions. Phylogenetic analysis indicated that potential
physiologic changes in the microbial community of N
2
fixers were not as remarkable as
were the genetic changes. This is fundamental for incorporating microorganisms in
ecosystem models, when only coarse information about microbial composition is viable
to use due to the overwhelming diversity of microorgnaims (Allison and Martiny 2008).
Overall, we found that microorganisms biodiversity have the potential to affect
ecosystem processes and influence ecosystem response to disturbance such as N and P
additions. It was demonstrated that the composition and richness of microorganisms as
indicators of microbial biodiversity are important to ecosystem functioning. Processes
like N
2
fixation are important for the whole ecosystem. It is a critical source for new
nitrogen into the mangrove ecosystem and was found to varied depending on a
combination of factors including presence of live mangrove roots and nutrient availability
which also affect nitrogen fixers community composition. Also, anaerobic respiration in
sediments was found to be affected by the availability of nutrients but the effects largely
depends on the broad microbial group studied. Therefore, is reiterated the need to better
understanding microbial ecological patterns and processes in interaction with the
environment and plants to improve predictive models of how ecosystem processes
respond to global change (Allison and Martiny 2008).
125
In conclusion, this research involves new approaches to understanding microbial
ecology in marine sediments by including the dynamics of plant-microbe-environment
interactions. The studies of the effects of long-term fertilization with N or P revealed that
(1) the effects depend on the initial conditions prior to disturbance; and (2) the effects
include changes in pathways for microbial metabolism and in community composition
patterns of microbial functional groups. Remarkably, variability of bacteria and N
2
fixers
in mangrove sediments was observed in response to natural conditions and fertilization
treatments. A wide range of physiological adaptations of N
2
fixers (composed of high
molecular diversity of nifH genes with low diversity in metabolic traits) to a large
heterogeneity of microenvironments is a strategy ensuring important biogeochemical
processes for the ecosystem in highly dynamic environments such as mangroves. Also,
was shown the importance of including biological interactions in ecological studies,
specifically plant-microbial interactions. Ecological studies have broadened our
understanding of the relationship between biological diversity and ecosystem function but
largely have excluded the role of plant-microbial interactions. Also, we have
demonstrated the importance of disturbance by nutrient additions to not only ecological
patterns and processes of bacteria and microbial functional groups but as well to the
relationship between ecosystem function and microbial diversity. We demonstrated the
relationship between plant-microbial interactions, microbial diversity-function and
disturbance by nutrient additions in sediments from mangrove environments.

126
References

Abrajano, T. A., D. E. Murphy, J. Fang, P. Comet, and J. M. Brooks. 1994.
13
C/
12
C
ratios in individual fatty acids of marine mytilids with and without bacterial
symbionts. Organic Geochemistry 21:561-827.
Allison, S. D., and J. B. Martiny. 2008. Resistance, resilience, and redundancy in
microbial communities. Proc. Natl. Acad. Sci. USA 105 (1):11512-11519.
Alongi, D. M. 1988. Bacterial productivity and microbial biomass in tropical mangrove
sediments. Microb. Ecol. 15:59-79.
Alongi, D. M. 1994. The role of bacteria in nutrient cycling n tropical mangrove and
other coastal benthic ecosystems. Hydrobiologia 285:19-32.
Alongi, D. M. 1996. The dynamics of benthic nutrient pools and fluxes in tropical
mangrove forests. J. Mar. Res. 54:123-148.
Alongi, D. M. 1998. Coastal ecosystem processes. CRC Press, Marine Science series,
Boca Raton, FL, USA, 448 pp.
Alongi, D. M. 2002. Present State and future of the worlds mangrove forests. Environ.
Conserv. 29(3):331-349.
Alongi, D. M., F. Tirendi, and B. F. Clough. 2000a. Below-ground decomposition of
organic matter in forests of the mangroves Rhizophora stylosa and Avicennia
marina along the arid coast of Western Australia. Aquat. Bot. 68:97-122.
Alongi, D. M., F. Tirendi, L. A. Trott, T.T. Xuan. 2000b. Benthic decomposition rates
and pathways in plantations of the mangrove Rhizophora apiculata in the Mekong
delta, Vietman. Mar. Ecol. Prog. Ser. 194:87-101.
Alongi, D. M., F. Tirendi, P. Dixon, L. A. Trott, and G. J. Brunskill. 1999.
Mineralization of organic matter in intertidal sediments of a tropical semi-
enclosed delta. Estuar. Coast. Shelf Sci. 48:451-467.
Alongi, D. M., G. Waittayakorn, S. Boyle, F. Tirendi, C. Payn, and P. Dixon. 2004.
Influence of roots and climate on mineral and trace storage and flux in tropical
mangrove soils. Biogeochemistry 69:105-123.
127
Alongi, D. M., G. Wattayakor, J. Pfitzner, F. Tirendi, I. Zarorskis, G. J. Brunskill,
A Davidson, and B. F. Clough. 2001. Organic carbon accumulation and
metabolic pathways in sediments of mangrove forest in southern Thailand. Mar.
Geol. 179:85-103.
Alongi, D. M., L. A Trott, G. Wattayakorn, and B.F. Clough. 2002. Below-ground
nitrogen cycling in relation to net canopy production in mangrove forest of
southern Thailand. Mar. Biol. 140:855-864.
Alongi, D. M., P. Christoffersen, and F. Tirendi. 1993. The influence of forest type on
microbial-nutirent relationships in tropical mangrove sediments. J. Exp. Mar.
Biol. Ecol. 171:201-223.
Alongi, D. M., T. Ayukai, G. J. Brunskill, B. F. Clough, and E. Wolanski. 1998.
Sources, sinks and export of organic carbon through a tropical, semi-enclosed
delta (Hinchinbrook Channel, Australia). Mangroves Salt Marshes 2:237-242.
Ayukai, T. 1998. Introduction: carbon fixation and storage in mangroves and their
relevance to the global climate change - a case in Hinchinbrook Channel in
northeastern Australia. Mangroves and Salt Marshes 2:189-190.
Bagwell, C. E., and C. R. Lovell. 2000. Persistence of selected Spartina alterniflora
rhizoplane diazotrophs exposed to natural and manipulated environmental
variability. Appl. Environ. Microbiol. 66 (11):4625-4633.
Bagwell, C. E., Y. M. Piceno, and A. Ashburne-Lucas. 1998. Physiological diversity
of the rhizosphere diazotroph assemblages of elected salt marsh grasses. Appl.
Environ. Microbiol. 64:4276-4282.
Bagwell C, Rocque J, Smith G, Polson S, Friez M, Longshore J, Lovell C. 2002.
Molecular diversity of diazotrophs in oligotrophic tropical seagrass bed
communities. FEMS Microb. Ecology 39: 113-119.

Bashan, Y., M. E. Puente, D. D. Myrold, and G. Toledo. 1998. In vitro transfer of
fixed nitrogen from diazotrophic filamentous cyanobacteria to black mangrove
seedlings. FEMS Microbiol. Ecol. 26:165-170.
Baudoin, E., E. Benizri, and A. Guckert. 2003. Impact of artificial root exudates on the
bacterial community structure in bulk soil and maize rhizosphere. Soil Biol.
Biochem. 35(9):1183-1192.

128
Bertin, C., X. Yang, and L. A. Weston. 2003. The role of root exudates and
allelochemicals in the rhizosphere. Plant and soil 256:67-83.
Billings, S., and S. Ziegler. 2005. Linking microbial activity and soil organic matter
transformations in forest soils under elevated co2. Global Change Biol 11:203-
212.
Bligh, E. G., and W. J. Dyer. 1959. A rapid method of total lipid extraction and
purification. Canadian Journal of Biochemical Physiology, 37, 911–917.
Bodelier, P. L. E. 2003. Interactions between oxygen-releasing roots and microbial
processes in flooded soils and sediments. In H. de Kroon, and E. J. W. Visser
(ed.). Root Ecology (Ecological Studies v. 168), Springer-Verlag, Berlin
Heidelberg, pp 331-362.
Borges, A. V., S. Djenidi, G. Lacroix, J. Theate, B. Delilli, and M. Frankignoulle.
2003. Atmospheric CO
2
flux from mangrove surrounding waters. Geophys. Res.
Lett. 30:11-14.
Boto, K. G., and A. I. Robertson. 1990. The relationship between nitrogen fixation and
tidal exports of nitrogen in a tropical mangrove ecosystem. Estuar. Coast. Shelf
Sci. 31:531-540.
Bouillon, S., A. V. Borges, E. Castañeda-Moya, K. Diele, T. Dittmar, N. C. Duke, E.
Kristensen, S. Y. Lee, C. Marchand, J. J. Middelburg, V. H. Rivera-Monroy,
T. J. Smith, and R. R. Twilley. 2008. Mangrove production and carbon sinks: A
revision of global budget estimates. Global Biogeochem. Cycles 22(2):1-12.
Bouillon, S., and H. Boschker. 2006. Bacterial carbon sources in coastal sediments: A
cross-system analysis based on stable isotope data of biomarkers. Biogeosciences
3:175-185.
Bouillon, S., F. Dahdouh-Guebas, A. V. V. S. Rao, N. Koedam, and F. Dehairs.
2003b. Sources of organic carbon in mangrove sediments: variability and possible
ecological implications. Hydrobiologia 495:33-39.
Bouillon, S., M. Chandra, N. Sreenivas, and F. Dehairs. 2000. Sources of suspenden
matter and selective feeding by zooplankton in an estuarine mangrove ecosystem,
as traced by stable isotopes. Mar. Ecol. Prog. Ser. 208:79-92.

129
Bouillon, S., M. Frankignoulle, F. Dehairs, B. Velimirov, A. Eiler, G. Abril, H.
Etcheber, and A. Vieira Borges. 2003a. Inorganic and organic carbon
biogeochemistry in the Gautami Godavari estuary (Andhra Pradesh, India) during
pre-monsoon: the local impact of extensive mangrove forest. Global Biogeochem.
Cycles 17(4): 25-1 – 25-12.
Bouillon, S., N. Koedam, A. V. Raman, and F. Dehairs. 2002. Primary producers
sustaining macro-invertebrate communities in intertidal mangrove forests.
Oecologia 130:441-448.
Bouillon, S., T. Moens, and F. Dheairs. 2004b. Carbon sources supporting benthic
mineralization in mangrove and adjacent seagrass sediments (Gazi Bay, Kenya).
Biogeosciences 1:71-78.
Bouillon, S., T. Moens, N. Koedam, F. Dahdouh-Guebas, W. Baeyens, and F
Dehairs. 2004a. Variability in the origin of carbon substrates for bacterial
communities in mangrove sediments. FEMS Microbiology Ecology 49: 171–179.
Brown, M. M., M. J. Friez, and C. R. Lovell. 2003. Expression of nifH genes by
diazotrophic bacteria in the rhizosphere of short form Spartina alterniflora FEMS
microbiology ecology. vol. 43, no3, pp. 411-417.
Bürgmann, H., F. Widmer, W. V. Sigler, and J. Zeyer. 2003. mRNA Extraction and
Reverse Transcription-PCR Protocol for Detection of nifH Gene Expression by
Azotobacter vinelandii in Soil. Appl. Environ. Microbiol., 69 (4): 1928-1935.
Burke, R. A., M. Molina, J. E. Cox, L. J. Osher, and M. C. Piccolo. 2003. Stable
carbon isotope ratio and composition of microbial fatty acids in tropical soils. Am
Soc Agronom. 32: 198-206.
Burns, J. 2003. Molecular Diversity of diazotrophs in contrasting shallow marine
sediments. Ph.D. thesis, University of Maryland, College Park, Maryland.
Butler, J. L., M. A. Williams, P. J. Bottomley, and M. D. Myrold. 2003. Microbial
community associated with rhizosphere carbon flux. Appl. Environ. Microbiol.
69:6793-6800.
Capone, D. G. 1988. Benthic nitrogen fixation, p. 85-123. In T. H. Blackburn, J.
Sørensen (ed.), Nitrogen Cycling in Coastal Marine Environments, SCOPE
Series, J. Wiley, New York, NY.
130
Capone, D. G. 1988. Benthic nitrogen fixation, p. 85-123. In T. H. Blackburn, J.
Sørensen (ed.), Nitrogen Cycling in Coastal Marine Environments, SCOPE
Series, J. Wiley, New York, NY.
Capone, D. G., and J. M. Budin. 1982. Nitrogen fixation associated with rinsed roots
and rhizomes of the eelgrass Zostera marina. Plant Physiol. 70:1601-1604.
Capone, D. G., and J. P. Montoya. 2001. Nitrogen fixation and denitrification. Methods
in Microbiology 30:501-515.
Capone, D. G., P. H. Penhale, R. S. Oremland, and B. F. Taylor. 1979. Relationship
between productivity and N
2
(C
2
H
2
) fixation in a Thalassia testudinum
community. Limnol. Oceanogr. 24(1):117-125.
Carpenter, E. J., and D. G. Capone. 1983. (ed.) Nitrogen in the Marine Environment.
Academic Press, New York, NY. 900 pp.
Cebrian, J. 2002. Variability and control of carbon consumption export, and
accumulation in marine communities. Limnol. Oceanogr. 47:11-22.
Cerna, B., E. Rejmankova, J. M. Snyder, and H. Santruckova. 2009. Heterotrophic
nitrogen fixation in oligotrophic tropical marshes: changes after phosphorus
addition. Hydrobiologia 627: 55–65.
Cheeseman, J. M., and C. E. Lovelock. 2004. Photosynthetic characteristics of dwarf
and fringe Rhizophora mangle L. in a Belizean mangrove. Plant Cell Environ.
27(6):769-780.
Chen, R., and R. R. Twilley. 1999. A simulation model of organic matter and nutrient
accumulation in mangrove wetland soils. Biogeochemistry 44:93-118.
Chmura, G. L., S. C. Anisfeld, D. R. Cahoon, and J. C. Lynch. 2003. Global Carbon
sequestration in tidal, saline wetland soils. Global Biogeochem. Cycles 17(4):22-
12.
Cifuentes, L. A., and G. G. Salata. 2001. Significance of carbon isotope discrimination
between bulk carbon and extracted phospholipid fatty acids in selected terrestrial
and marine environments. Organic Geochemistry 32: 613-621.
Cline, J. D. 1969. Spectrophotometric determination of hydrogen sulfide in natural
waters. Limnol. Oceanogr. 14:454-458.
131
Cotano, U., and F. Villate. 2006. Anthropogenic influence on the organic fraction of
sediments in two contrasting estuaries: A biochemical approach. Mar Pollut Bull
52: 404-414.
Dehairs, F., R. G. Rao, M. P. Chandra, V. Raman, S. Marguillier, and L. Hellings.
2000. Tracing mangrove carbon in suspended matter and aquatic fauna of the
Gautami-Godavari Delta, Bay of Bengal (India). Hydrobiologia 431: 225-241.
Dittmar, T., N. Hertkorn, G. Kattner, and R. J. Lara. 2006. Mangroves, a major
source of dissolved organic carbon to the oceans. Global Biogeochem. Cycles 20,
GB1012, doi:10.1029/2005GB002570.
Dobbs, F. C., and R. H. Findlay. 1993. Analysis of microbial lipids to determine
biomass and detect the response of sedimentary microorganisms to disturbance.
In: Handbook of Methods in Aquatic Microbial Ecology (ed. Kemp PF), Lewis
Publishers, Florida, pp. 347–358.
Duarte, C. M., and J. Cebrian. 1996. The fate of marine autotrophic production.
Limnology and oceanography 41:1758-1766.
Duarte, C. M., J. J. Middelburg, and N. Caraco. 2005. Major role of marine vegetation
on the oceanic carbon cycle. Biogeosciences 2: 1-8.
Ewel, K. C., R. R. Twilley, and J. E. Ong. 1998. Different kinds of mangrove forests
provide different goods and services. Global Ecol. Biogeogr. Lett. 7:83-94.
Farrar, J., M. Hawes, D. Jones, and S. Lindow. 2003. How roots control the flux of
carbon to the rhizosphere. Ecology 84(4):827-837.
Feller, I. C., D. F. Whigham, J. P. O´Neil, and K. L. McKee. 1999. Effects of nutrient
enrichment on within-stand cycling in a mangrove forest. Ecology 80(7):2193-
2205.
Feller, I. C., D. F. Whigham, K. L. McKee, and C.E. Lovelock. 2003. Nitrogen
limitation of growth and nutrient dynamics in a disturbed mangrove forest, Indian
River Lagoon, Florida. Oecologia 134:405-414.
Feller, I. C., K. L. McKee, D. F. Whigham, and J. P. O´Neil. 2002. Nitrogen vs.
phosphorus limitation across an ecotonal gradient in a mangrove forest.
Biogeochemistry 62:145-175.
132
Fernández, A., S. Huang, S. Seston, J. Xing, R. Hickey, C. Criddle, and J. Tiedje.
1999. How stable is stable? Function versus community composition. Appl.
Environ. Microbiol. 65:3697-3704
Fierer, N., J. P. Schimel, and P. A. Holden. 2003. Variations in microbial community
composition through two soil depth profiles. Soil Biology and Biochemistry
35:167-176.
Flores-Mireles, A., S. Winans, and G. Holguin. 2007. Molecular Characterization of
Diazotrophic and Denitrifying Bacteria Associated with Mangrove Roots. Appl.
Environ. Microbiol. 73(22):7308-7321.
Fogel, M., M. Wooller, J. Cheeseman, B. Smallwood, Q. Roberts, I. Romero, and M.
Meyers. 2008. Unusually negative nitrogen isotopic compositions (δ 15 n) of
mangroves and lichens in an oligotrophic, microbially-influenced ecosystem.
Biogeosciences 5:1693-1704.
Furukawa, K., E. Wolanski, and H. Mueller. 1997. Currents and sediment transport in
mangrove swamps. Estuar. Coast. Shelf Sci. 44:301-310.
Gallon, J. R. 1992. Reconciling the incompatible: N
2
fixation and O
2
. New Phytol.
122(4): 571-609.
Galloway, J. N., F. J. Dentener, D. G. Capone, E. W. Boyer, R. W. Howarth, S. P.
Seitzinger, G. P. Asner, C. C. Cleveland, P. A. Green, E. A. Holland, D. M.
Karl, A. F. Michaels, J. H. Porter, A. R. Townsend, and C. J. Vöosmarty.
2004. Nitrogen cycles: past, present, and future. Biogeochemistry 70:153-226.
Gatusso, J. P., M. Frankignoulle, and R. Wollast. 1998. Carbon and Carbonate
metabolism in coastal aquatic ecosystems. Annu. Rev. Ecol. Syst. 29:405-434.
Gleason, S. M., K. C. Ewel, and N. Hue. 2003. Soil redox conditions and plant-soil
relationships in a Micronesian mangrove forest. Estuar. Coast. Shelf Sci. 56:1065-
1074.
Goneea, M. E., A. Paytan, and J. A. Herrera-Silveira. 2004. Tracing organic matter
sources and carbon burial in mangrove sediments over the past 160years. . Estuar.
Coast. Shelf Sci. 61:211-227.
Gotto, J. W., and B. F. Taylor. 1976. N
2
fixation associated with decaying leaves of the
red mangrove (Rhizophora mangle). Appl. Environ. Microbiol. 31(5):781-783.
133
Grayston, S. J. et al. 1997. Rhizosphere carbon flow in trees, in comparison with annual
plants: the importance of root exudation and its impact on microbial activity and
nutrient availability. Appl. Soil Ecol. 5, 29-56.
Guisan, A., and N. Zimmermann. 2000. Predictive habitat distribution models in
ecology. Ecological modelling 135:147-186.
Hall, B. G. 2008. Phylogenetic trees made easy: a how-to manual. Sunderland, Mass. 2nd
ed. 221 p.
Hansen, J. W., J. W. Udy, C. J. Perry, W. C. Dennison, and B. A. Lomstein. 2000.
Effect of the seagrass Zostera Capricorni on sediment microbial processes. Mar.
Ecol. Prog. Ser. 199:83-96.
Harris, D., W. Horwath, and C. van Kessel. 2001. Acid fumigation of soils to remove
carbonates prior to total organic carbon or carbon-13 isotopic analysis. Soil
Science Society of America Journal 65:1853-1856.
Hastie, T., and R. Tibshirani. 1986. Generalized additive models. Statistical Science
1:297-310.
Heidelberg, J., R. Seshadri, S. Haveman, C. Hemme, I. Paulsen, J. Kolonay, J.
Eisen, N. Ward, B. A. Methe, L. Brinkac, et al. 2004. The genome sequence of
the anaerobic, sulfate-reducing bacterium desulfovibrio vulgaris hildenborough.
Nature biotechnology 22:554-559.
Hewson, I., and J. A. Fuhrman. 2006. Spatial and vertical biogeography of coral reef
sediment bacterial and diazotroph communities. Mar Ecol. Prog. Ser. 306:79-86.
Hobbie, E. A., M. G. Johnson, P. T. Rygiewicz, D. T. Tingey, and D. M. Olszyk.
2004. Isotopic estimates of new carbon inputs into litter and soils in a four-year
climate change experiment with Douglar-fir. Plant soil 331:331-343.
Högberg, P., A. Nordgren, N. Buchmann, A. F. S. Taylor, A. Ekblad, M. N.
Högberg, G. Nyberg, M. Ottosson-Lo fvenius, and D. J. Read. 2001. Large-
scale forest girdling shows that current photosynthesis drives soil respiration.
Nature 411:789-792.
Holguin, G., M. A. Guzman, and Y. Bashan. 1992. Two new nitrogen-fixing bacteria
from the rhizosphere of mangrove trees: Their isolation, identification and in vitro
interaction with rhizosphere Staphylococcus sp. FEMS Microbiol. Lett. 101:207-
216.
134
Holguin, G., P. Vazquez, and Y. Bashan. 2001. The role of sediment microorganisms
in the productivity, conservation, and rehabilitation of mangrove ecosystems: an
overview. Biol. Fertil. Soils 33:265-278.
Hollander, D., and M. Smith. 2001. Microbially mediated carbon cycling as a control
on the δ
13
C of sedimentary carbon in eutrophic Lake Mendota (USA): New
models for interpreting isotopic excursions in the sedimentary record. Geochimica
et Cosmochimica Acta 65:4321-4337.
Holmer, M., and A. B. Olsen. 2002. Role of decomposition of mangrove and seagrass
detritus in sediment carbon and nitrogen cycling in a tropical mangrove forest.
Mar. Ecol. Prog. Ser. 230:87-101.
Holmer, M., C. M. Duarte, H. T. S. Boschker, and C. Barron. 2004. Carbon cycling
and bacterial carbon sources in pristine and impacted Mediterranean seagrass
sediments. Aquat. Microb. Ecol. 36:227-237.
Hughes, J. B., J. J. Hellmann, T. H. Ricketts, and B. J. M. Bohannan. 2001. Counting
the uncountable: Statistical approaches to estimating microbial diversity. Appl.
Environ. Microb. 67:4399-4406.
Jennerjahn, T. C., and V. Ittekkot. 2002. Relevance of mangroves for the production
and deposition of organic matter along tropical continental margins.
Naturwissenschaften 89:23-30.
Jones, W. B., L. A. Cifuentes, and J. E. Kaldy. 2003. Stable isotope evidence for
coupling between sedimentary bacteria and seagrasses in a sub-tropical lagoon.
Mar. Ecol. Prog. Ser. 255: 15-25.
Joye, S., and R. Lee. 2004. Benthic microbial mats: important sources of fixed nitrogen
and carbon to the Twin Cays, Belize Ecosystem. Atoll. Res. Bull. 528:1-24.
Karjalainen, H., G. Stefansdottir, L. Tuominen, and T. Kairesalo. 2001. Do
submersed plants enhance microbial activity in sediments?. Aquat. Bot. 69:1-13.
Kaur, A., Chaudhary A., A. Kaur, R. Choudhary, and R. Kaushik. 2005.
Phospholipid fatty acid – a bioindicator of environmental monitoring and
assessment in soil ecosystem. Current Science 89 (7): 1103-1112.

135
Kjerfve, B., K. Rützler, and G. H. Kierspe. 1982. Tides at Carrie Bow Cay, Belize.
Terrestrial environment and climate, Carrie Bow Cay, Belize. In: The Atlantic
barrier reef ecosystem at Carrie Bow Cay, Belize I: Structure and Communities
(eds. K. Rützler and I. G. Macintyre), pp. 47-51. Washington D.C.: Smithsonian
Institution Press.
Koltes, K. H., J. J. Tschirkya, and I. C. Feller. 1998. Carrie Bow Cay, Belize.
Caribbean coastal marine productivity (CARICOMP): coral reef, seagrass, and
mangrove site characteristics (ed. by B. Kjerfve), pp. 79–94. UNESCO, Paris.
Koop, K., D. Booth, A. D. Broadbent, J. Brodie, D. Bucher, D. Capone, J. Coll, W.
Denison, M. Erdmann, P. Harrison, O. Hutchings, G. B. Jones, A. W. Larkum,
J. O'Neil, A. Steven, E. Tentori, S. Ward, J. Williamson, D. Yellowlees. 2001.
ENCORE: the effect of nutrient enrichment on coral reefs, synthesis of results and
conclusions. Mar. Pollut. Bull. 42(2): 91-120.
Koster, M., S. Dahlke, and L. A. MeyerReil. 1997. Microbiological studies along a
gradient of eutrophication in a shallow coastal inlet in the southern baltic sea
(nordrugensche bodden). Mar Ecol-Prog Ser 152: 27-39.
Kristensen, E., M. Holmer, and N. Bussarawit. 1991. Benthic Metabolism and Sulfate
Reduction in a Southeast Asian Mangrove Swamp. Marine Ecology Progress
Series 73 (1): 93-103.
Kristensen, E., M. Holmer, G. T. Banta, M. H. Jensen, and K. Hansen. 1995. Carbon,
nitrogen and sulfur cycling in sediments of the Ao Nam Bor mangrove forest,
Phuket, Thailand: a review. Res. Bull. Phuket Mar. Biol. Cent. 60:37-64.
Kristensen, E., S. Bouillon, T. Dittmar, and C. Marchand. 2008. Organic carbon
dynamics in mangrove ecosystems: A review. Aquatic Botany. 89: 201-219.
Kuzyakov, Y. 2001. Tracer studies of carbon translocation by plants from the
atmosphere into the soil (a review). Eurasian Soil Sci. 34(1):28-42.
Kuzyakov, Y., and G. Domanski. 2000. Carbon inputs by plants into the soil. Review. J.
Plant Nutr. Soil Sci. 163:421-431.
Kuzyakov, Y., and W. Cheng. 2001. Photosynthesis controls of rhizosphere respiration
and organic matter decomposition. Soil Biol. Biochem. 33(14):1915-1925.
Kuzyakov, Y., J. K. Friedel, and K. Stahr. 2000. Review of mechanisms and
quantification of priming effects. Soil Biol. Biochem. 32:1485-1498.
136
Laegdsgaard, P., and C. Johnson. 2001. Why do juvenile fish utilise mangrove
habitats? J. Exp. Mar. Biol. Ecol. 257, 229–253.
Lee, R. Y., W. P. Porubsky, I. C. Feller, K. L. McKee, and S. B. Joye. 2008.
Porewater biogeochemistry and soil metabolism in dwarf red mangrove habitats
(Twin Cays, Belize). Biogeochemistry 87: 181-198.
Lee, R., and S. Joye. 2006. Seasonal patterns of nitrogen fixation and denitrification in
oceanic mangrove habitats. Mar. Ecol. Prog. Ser. 307:127-141.
Lee, R., W. Porubsky, I. Feller, K. McKee, and S. Joye. 2008. Porewater
biogeochemistry and soil metabolism in dwarf red mangrove habitats (Twin Cays,
Belize). Biogeochemistry 87:181-198
Legendre, P., and L. Legendre. 1998. Numerical Ecology. 2
nd
English edition, Elsevier
Science, Amsterdam.
Lovell, C., P. Decker, C. Bagwell, S. Thompson, and G. Matsui. 2008. Analysis of a
diverse assemblage of diazotrophic bacteria from spartina alterniflora using dgge
and clone library screening. Journal of Microbiological Methods 73:160-171.
Lovelock, C. E., M. C. Ball, B. Choat, B. M. J. Engelbrecht, N. M. Holbrook, and
I.C. Feller. 2006. Linking physiological processes with mangrove forest
structure: phosphorus deficiency limits canopy development, hydraulic
conductivity and photosynthetic carbon gain in dwarf Rhizophora mangle. Plant
Cell Environ. 29 (5): 793-802.
Lyons, K. G., C. A. Brigham, B. H. Traut, and M. W. Schwartz. 2005. Rare species
and ecosystem functioning. Conservation Biology 19:1019-1024.
Macintyre, I. G., M. A. Toscazo, R. G. Lightly, and G. B. Bond. 2004. Holocene
history of the mangrove islands of Twin Cays, Belice, Central America. Atoll
Research Bull. 510: 1- 16.
Maier-Agustein, W. 2002. Stable isotopes analysis of fatty acids by gas
chromatography-isotope ratio mass spectrometry. Analytica Chimica Acta 465:
63-79.
Marcarelli, A. M., and W. A. Wurtsbaugh. 2006. Temperature and nutrient supply
interact to control nitrogen fixation in oligotrophic streams: An experimental
examination. Limnol. Oceanogr., 51(5): 2278–2289.
137
McGill, B. J., B. J. Enquist, E. Weiher, and M. Westoby. 2006. Rebuilding
community ecology from functional traits. Trends in Ecology & Evolution 21.
McGlathery, K. J., N. Risgaard-Petersen, and P. B. Christensen. 1998. Temporal
and spatial variation in nitrogen fixation activity in the eelgrass (Zostera marina
L.) rhizosphere. Mar. Ecol. Prog. Ser. 168:245-258.
Mchenga, I. S. S., and M. Tsuchiya. 2007. Nutrient dynamics in mangrove crab burrow
sediments subjected to anthropogenic input. Journal of Sea Research 59:103-113.
McKee, K. L. 1993. Soil Physicochemical Patterns and Mangrove Species Distribution--
Reciprocal Effects? J. Ecol. 81(3):477-487.
Mckee, K. L. 1995. Interspecific variation in growth, biomass partitioning, and defensive
characteristics of neotropical mangrove seedlings: response to light and nutrient
availability. American Journal of Botany 82 (3): 299-307.
Mckee, K. L., D. R. Cahoon, and I. C. Feller. 2007. Caribbean mangroves adjust to
rising sea level through biotic controls on change in soil elevation. Global Ecol.
Biogeogr. 16(5): 545-556.
McKee, K. L., I. A. Mendelssohn, and M. W. Hester. 1988. Reexamination of Pore
Water Sulfide Concentrations and Redox Potentials Near the Aerial Roots of
Rhizophora mangle and Avicennia germinans. Am. J. Bot. 75(9):1352-1359.
McKee, K. L., I. C. Feller, M. Popp, and W. Wanek. 2002. Mangrove isotopic (δ
15
N
and δ
13
C) fractionation across a nitrogen vs. phosphorous limitation gradient.
Ecology, 83, 1065–1075.
McKee, K.L. 1993, Soil physicochemical patterns and mangrove species distribution
reciprocal effects?: Journal of Ecology 81: 477-487.
Meier-Augenstein, W. 2002. Stable isotope analysis of fatty acids by gas
chromatography–isotope ratio mass spectrometry. Analytica Chimica Acta. 465:
63-79.

138
Methe, B., K. Nelson, J. Eisen, I. Paulsen, W. Nelson, J. Heidelberg, D. Wu, M. Wu,
N. Ward, M. Beanan, R. J. Dodson, R. Madupu, L. M. Brinkac, S. C.
Daugherty, R. T. DeBoy, A. S. Durkin, M. Gwinn, J. F. Kolonay, S. A.
Sullivan, D. H. Haft, J. Selengut, T. M. Davidsen, N. Zafar, O. White, B.
Tran, C. Romero, H. A. Forberger, J. Weidman, H. Khouri, T. V.
Feldblyum, T. R. Utterback, E. V. Aken, D. R. Lovley, and C. M. Fraser.
2003. Genome of geobacter sulfurreducens: Metal reduction in subsurface
environments. Science 302:1967-1969.
Middleton, B. A., and K. L. Mckee. 2001. Degradation of mangrove tissues and
implications for peat formation in Belizean island forest. J. Ecol. 89:818-828.
Musat, F., J. Harder, and F. Widdel. 2006. Study of nitrogen fixation in microbial
communities of oil-contaminated marine sediment microcosms. Environ.
Microbiol. 8:1834-1843.
Mussmann, M., K. Ishii, R. Rabus, and R. Amann. 2005. Diversity and vertical
distribution of cultured and uncultured Deltaproteobacteria in an intertidal mud
flat of the Wadden Sea. Environ. Microbiol. 7 (3): 405-18.
Nedwell, D. and S. Azis. 1980. Heterotrophic nitrogen fixation in an intertidal salt marsh
sediment. Estuarine Coastal Mar. Sci. 10: 699-702.
Nedwell, D. B., T. H. Blackburn, and W. J. Wiebe. 1994. Dynamic nature of the
turnover of organic carbon, nitrogen and sulphur in the sediments of Jamaican
mangrove forest. Mar. Ecol. Prog. Ser. 110: 223-231.
Nielsen, L. B., K. Finster, D. T. Welsh, A. Donnely, R. A. Herbert, R. DeWit, and B.
A. Lomstein. 2001. Sulphate reduction and nitrogen fixation rates with roots,
rhizomes and sediments from Zostera noltii and Spartina maritima meadows.
Environ. Microbiol. 3:63-71.
Palta, J. A., and P. O. J. Gregory. 1998. Drought affects the fluxes of carbon to roots
and soil in 13C pools-labeled plants of wheat. Soil Biol. Biochem. 29(9-10):1395-
1403.
Pelegri, S. P., and R. R. Twilley. 1998. Heterotrophic nitrogen fixation (acetylene
reduction) during leaf-litter decomposition of two mangrove species from South
Florida, USA. Mar. Biol. 131: 53-61.
139
Piceno, Y. M., and C. R. Lovell. 2000a. Stability in natural bacterial communities: I.
Nutrient addition effects on rhizosphere diazotroph assemblage composition.
Microb. Ecol. 39:32-40.
Piceno, Y. M., and C. R. Lovell. 2000b. Stability in natural bacterial communities: II.
Plant resource allocation effects on rhizosphere diazotroph assemblage
composition. Microb. Ecol. 39:41-48.
Piceno, Y. M., P. A. Noble, and C. R. Lovell. 1999. Spatial and Temporal assessment of
Diazotroph Assemblage composition in vegetated Salt Marsh sediments using
Denaturing Gradient Gel Electrophoresis Analysis. Microb. Ecol. 38:157-167.
Pielou, E. C. 1966. Species-diversity and pattern-diversity in the study of ecological
succession. J. Theor. Biol. 10:370–383.
Pinturier-Geiss, L., L. Méjanelle, B. Dale, and D. Karlsen. 2002. Lipids as indicators
of eutrophication in marine coastal sediments. Journal of Microbiological
Methods 48:239-257.
Polymenakou, P. N., A. Tselepides, E. G. Stephanou, and S. Bertilsson. 2006. Carbon
speciation and composition of natural microbial communities in polluted and
pristine sediments of the eastern mediterranean sea. Mar Pollut Bull 52:1396-
1405.
Postgate, J. R. 1982. The fundamentals of nitrogen fixation. Cambridge University
Press, Cambridge, GB; New York, NY, USA, 252 pp.
Purvaja, R., R. Ramesh, A. Ray, and T. Rixen. 2008. Nitrogen cycling: A review of
the processes, transformations and fluxes in coastal ecosystems. Curr. Sci.
94(11):1419-1438.
Ramsey P. W., M. C. Rillig, K. P. Feris, W. E. Holben, and J. E. Gannon. 2006.
Choice of methods for soil microbial community analysis: PLFA maximizes
power compared to CLPP and PCR-based approaches. Pedobiologia 50: 275-280.
Ravikumar, S., K. Kathiresan, S. Ignatiammal, M. Babu Selvam, and S. Shanthy.
2004. Nitrogen-fixing azotobacters from mangrove habitat and their utility as
marine biofertilizers. Journal of Experimental Marine Biology and Ecology
312:5-17.

140
Ringelberg, D., S. Sutton, and D. White. 1997. Biomass, bioactivity and biodiversity:
Microbial ecology of the deep subsurface: Analysis of ester-linked phospholipid
fatty acids. FEMS Microbiology Reviews 20:371-377.
Rivera-Monroy, V. H., and R. R. Twilley. 1996. The relative role of denitrification and
immobilization in the fate of inorganic nitrogen in mangrove sediments (Terminos
Lagoon, Mexico). Limnol. Oceanogr. 41:284-296.
Rodriguez, W., and I. C. Feller. 2004. Mangrove landscape characterization and change
in Twin Cays, Belize using aerial photography and Ikonos satellite data. Atoll
Res. Bull. 513: 1-24.
Rützler, K., and I. C. Feller. 1996. Caribbean mangrove swamps. Sci. Am. 274: 94–99.
Rützler, K., and J. D. Ferraris. 1982. Terrestrial environment and climate, Carrie Bow
Cay, Belize. In The Atlantic barrier reef ecosystem at Carrie Bow Cay, Belize I:
Structure and Communities (eds. K. Rützler and I. G. Macintyre), pp. 77-92.
Washington D.C.: Smithsonian Institution Press.
Saggar, S., C. Hedley, and A. D. Mackay. 1997. Partitioning and translocation of
photosynthetically fixed
14
C in grazed hill pastures. Biol. Fertility Soils 25:152-
158.
Schwamborn, R., M. Voss, W. Ekau, and U. Saint-Paul. 2002. How important are
mangroves as carbon sources for decapod crustacean larvae in a tropical estuary?
Mar. Ecol. Progr. Ser. 229: 195-205.
Seelinger, U., and B. Kjerfve. 2001. Coastal marine ecosystems of Latin American.
Springer, New York, NY, 362 pp.
Sengupta, A. and S. Chaudhuri. 1991. Ecology of heterotrophic dinitrogen fixation in
the rhizosphere of mangrove plant community at the Ganges river estuary in
India. Oecologia 87: 560-564.
Shannon, C. E., and W. Weaver. 1963. The Mathematical Theory of Communication.
University of Illinois Press, Urbana, IL, USA, 144 pp.
Sheaves, M., and B. Molony. 2000. Short-circuit in the mangrove food chain. Mar. Ecol.
Prog. Ser. 199: 97–109.
Sheridan, R. P. 1991. Epicaulous, Nitrogen-fixing microepiphytes in a Tropical Mangal
community, Guadalupe, French West Indies. Biotropica 23(4b):530-541.
141
Sherman, R. E., T. J. Fahey, and R. W. Howarth. 1998. Soil-plant interactions in a
neotropical mangrove forest: iron, phosphorus and sulfur dynamics. Oecologia
115:553-563.
Short, F. T., W. C. Dennison, and D. G. Capone. 1990. Phosphorus-limited growth of
the tropical seagrass Syringodium filiforme in carbonate sediments. Mar. Ecol.
Prog. Ser. 62: 169-174.
Singh, B., P. Millard, A. Whiteley, and J. Murrell. 2004. Unravelling rhizosphere–
microbial interactions: opportunities and limitations. Trends Microbiol.
12(8):386-393.
Sjoling, S., S. M. Mohammed, T. J. Lyimo, and J. J. Kyaruzi. 2005. Benthic bacterial
diversity and nutrient processes in mangroves: impact of deforestation. Est. Coast.
Shelf Sci. 63:397-406.
Smallwood, B., M. Wooller, M. Jacobson, and M. Fogel. 2003. Isotopic and molecular
distributions of biochemicals from fresh and buried rhizophora mangle leaves†.
Geochemical Transactions 4:38.
Solorzano, L. 1969. Determination of ammonia in natural waters by the phenol
hypochlorite method. Limnol. Oceanogr. 14:799-801.
Spalding, M. B., F. Blasco, C. D. Field. 1997. (ed.) World Mangrove Atlas.
International Society of Mangrove Ecosystems, Okinawa, Japan, 178 pp.
Strickland, J. D. H., and T. R. Parsons. 1972. A practical handbook of sea-water
analysis. Fisheries reseach board of Canada, Ottawa. CA, 311 p.
Sundareshwar, P. V., J. T. Morris, E. K. Koepfler, and B. Fornwalt. 2003.
Phosphorus limitation of coastal ecosystem processes. Science 299:563-565.
Teece, M. A., M. L. Fogel, M. E. Dollhoph, and K. H. Nealson. 1999. Isotopic
fractionation associated with biosynthesis of fatty acids by a marine bacterium
under oxic and anoxic conditions. Org. Geochem. 30(12): 1571-1579.
Ter Braak, C. J. F., and I. C. Prentice. 1988. A Theory of Gradient Analysis. Adv.
Ecol. Res. 18:271-339.
Thibodeau, F. R., and N. H. Nickerson. 1986. Differential oxidation of mangrove
substrate by Avicennia germinans and Rhizophora mangle. Am. J. Bot. 73(4):512-
516.
142
Tibbles, B. J., M. I. Lucas, V. E. Coyne, and T. S. Newton. 1994. Nitrogenase activity
in marine sediments from a temperate salt marsh lagoon: modulation by complex
polysaccharides, ammonium and oxygen. J. Exp. Mar. Biol. Ecol. 184:1-2.
Tissue, D. T., J. P. Megonigal, and R. B. Thomas. 1997. Nitrogenase activity and N
2

fixation are stimulated by elevated CO
2
in a tropical N
2
-fixing tree. Oecologia
109:28-33.
Toledo, G., Y. Bashan, and A. Soeldner. 1995. Cyanobacteria and Black mangroves in
Northwestern Mexico: colonization, and diurnal and seasonal nitrogen fixation on
aerial roots. Can. J. Microbiol. 41:999-1011.
Toledo, G., Y. Bashan, and A. Soeldner. 1995b. In vitro colonization and increase in
nitrogen fixation of seedling roots of black mangrove inoculated by a filamentous
cyanobacteria. Can. J. Microbiol. 41:1012-1020.
Twilley, R. R. 1998. Mangrove wetlands. In Southern forested wetlands: Ecology and
Management. Messina M. G., and W. H. Conner (ed.). CRC Press, Boca Raton,
FL, USA, pp. 445-475.
Valiela, I., J. L. Bowen, and J. K. York. 2001. Mangrove forests: One of the world´s
threatened major tropical environments. BioScience 51(10):807-815.
van der Valk, A. G., and P. M. Attiwill. 1984. Acetylene reduction in an Avicennia
marina community in Southern Australia. Aust. J. Bot. 32:157-164.
Vazquez, P., G. Holguin, M. Puente, A. Lopez-Cortes, and Y. Bashan. 2000.
Phosphate-solubilizing microorganisms associated with the rhizosphere of
mangroves in a semiarid coastal lagoon. Biol Fertil Soils 30:460-468.
Vitousek, P. M., J. D. Aber, R. W. Howarth, G. E. Likens, P. A. Matson, D. W.
Schindler, W. H. Schlesinger, and D. G. Tilman. 1997. Human Alteration of
the Global Nitrogen Cycle: Sources And Consequences. Ecol. Appl. 7(3):737-
750.
Walker, T. S., H. P. Bais, E. Grotewold, and J. M. Vivanco. 2003. Root exudation and
rhizosphere biology. Plant Physiol. 132:44-51.
Wattayakorn, G., T. Ayukai, and P. Sojisuporn. 2000. Material transport and
biogeochemical processes in Sawi Bay, southern Thailand. Spec. Publ. Phuket
Mar. Biol. Cent. 22:63-78.
143
Welsh, D.T. 2000. Nitrogen fixation in seagrass meadows: regulation, plant-bacteria
interactions and significance to primary production. Ecol. Lett. 3:58-71.
Werry, J., and S. Y. Lee. 2005. Grapsid crabs mediate link between mangrove litter
production and estuarine planktonic food chains. Marine Ecology Progress Series
293:165-176.
White, D. C., and D. B. Ringelberg. 1998. Signature lipid biomarker analysis. In:
Techniques in Microbial Ecology (eds Burlage RD, Atlas R, Stahl D, Geesey G,
Sayler G), Oxford University Press, New York, pp. 255–272.
White, D., W. M. Davis, J. S. Nickels, J. D. King, and R. J. Bobbie. 1979.
Determination of the sedimentary microbial biomass by extractible lipid
phosphate. Oecologia 40:51-62.
Whiting, G. J., E. L. Gandy, and D. C. Yoch. 1986. Tight coupling of root-associated
nitrogen fixation and plant photosynthesis in the salt marsh grass Spartina
alterniflora and carbon dioxide enhancement of nitrogenase activity. Appl.
Environ. Microbiol. 52:108-113.
Widmer, F., B. T. Schafer, L. A. Porteous, R. J. Seidler. 1999. Analysis of nifH gene
pool in soil and litter at a Douglas fir forest site in the Oregon Cascade Mountain
Range. Appl. Environ. Microbiol. 65:374–380.
Woitchik, A. F., B. Ohowa, J. M. Kazungu, R. G. Rao, L. Goeyens, and F. Dehairs.
1997. Nitrogen enrichments during decomposition of mangrove leaf litter in an
east African coastal lagoon (Kenya): Relative importance of biological nitrogen
fixation. Biogeochemistry 39:15-35.
Wolanski, E., S. Spagnol, and T. Ayukai. 1998. Field and model studies of the fate of
particulate carbon in mangrove-fringed Hinchinbrook Channel, Australia.
Mangroves Salt Marshes 2:205-221.
Wong, Y. S., C. Y. Lan, G. Z. Chen, S. H. Li, X. R. Chen, Z. P. Liu, and N. F. Y.
Tam. 1995. Effect of wastewater discharge on nutrient contamination of
mangrove soils and plants. Hydrobiologia 295:243-254.
Wong, Y. S., N. F. Y. Tam, and C. Y. Lan. 1997. Mangrove wetlands as wastewater
treatment facility: a field trial. Hydrobiologia 352:49-59.

144
Wooller, M., U. Scharler, B. Smallwood, M. Jacobson, and M. Fogel. 2003. A
taphonomic study of δ
13
C and δ
15
N values in Rhizophora mangle leaves for a
multiproxy approach to mangrove paleoecology, Organic Geochemistry 34: 1259–
1275.
Yachi, S., and M. Loreau. 1999. Biodiversity and ecosystem productivity in a
fluctuating environment: The insurance hypothesis. Proc. Natl . Acad. Sci. 96:
1463–1468.
Yannarell, A., T. Steppe, and H. Paerl. 2006. Genetic variance in the composition of
two functional groups (diazotrophs and cyanobacteria) from a hypersaline
microbial mat. Appl Environ Microb 72:1207-1217.
Yevdokimov, I., A. Gattinger, F. Buegger, J. C. Munch, and M. Schloter. 2008.
Changes in microbial community structure in soil as a result of different amounts
of nitrogen fertilization. Biol Fertil Soils 44:1103-1106.
Yoch, D. C., and G. J. Whiting. 1986. Evidence for NH
4
+
switch-off regulation of
nitrogenase activity by bacteria in salt marsh sediments and roots of the grass
Spartina alterniflora. Appl. Environ. Microbiol. 51:143-149.
Zehr, J. P., M. Mellon, S. Braun, W. Litaker, T. F. Steppe, and H. W. PAERL. 1995.
Diversity of heterotrophic nitrogen fixation genes in a marine cyanobacterial mat.
Appl Environ Microb 61:2527-2532.
Zehr, J., B. Jenkins, S. Short, and G. Steward. 2003. Nitrogenase gene diversity and
microbial community structure: A cross-system comparison. Environ Microbiol
5:539-554.
Zehr, J. P., and L. A. McReynolds. 1989. Use of degenerate oligonucleotides for
amplification of the nifH gene from the marine cyanobacterium Trichodesmiun
thiebautii. Appl. Environ. Microbiol. 55(10):2522-2526.
Zelles, L. 1999. Fatty acid patterns of phospholipids and lipopolysaccharides in the
characterisation of microbial communities in soil: A review. Biol Fertil Soils
29:111-129.
Zelles, L., Q. Y. Bai, T. Beck, and F. Beese. 1992. Signature fatty acids in
phospholipids and lipopolysaccharides as indicators of microbial biomass and
community structure in agricultural soils. Soil Biol Biochem. 24: 317–323.
145
Zhang, Y. Y., J. J. dong, Z. H. Yang, S. Zhang, and Y.S. Wang. 2008. Phylogenetic
diversity of nitrogen-fixing bacteria in mangrove sediments assessed by PCR-
denaturing gradient gel electrophoresis. Arch. Microbiol. 190:19-28.
Zuberer, D. A., and W. S. Silver. 1978. Biological dinitrogen fixation (Acetylene
reduction) associated with Florida mangroves. Appl. Environ. Microbiol. 35:567-
575.

146

APPENDIX

147
Appendix 1. Dead roots measurements.

Zone Treatment Depth (cm) C/N (atoms)
I C 0-1 56.2 -28.0 -3.3
I C 0-1 38.4 -28.4 -3.1
I C 0-1 37.9 -26.9 -2.0
I C 0-1 49.3 -26.8 -3.7
I C 0-1 57.9 -26.4 -2.7
I C 0-1 52.8 -28.0 -2.0
I C 5-10 147.0 -28.9 -1.7
I C 5-10 174.9 -28.6 -2.0
I C 20-30 56.7 -26.0 -0.8
I C 20-30 62.7 -25.8 -0.8
I C 20-30 52.6 -25.9 -0.8
I C 20-30 59.1 -26.6 -1.7
I C 20-30 49.0 -24.6 -1.9
I C 20-30 52.0 -25.3 -2.1
I N 0-1 49.9 -30.0 -2.0
I N 0-1 55.5 -30.8 -2.2
I N 0-1 48.2 -30.0 -2.0
I N 5-10 67.1 -26.3 -7.3
I N 5-10 72.5 -26.1 -7.2
I N 5-10 46.4 -25.9 -8.5
I N 20-30 214.6 -26.6 -1.6
I N 20-30 195.7 -26.7 -2.3
I N 20-30 194.4 -26.1 -1.4
I P 5-10 69.0 -26.8 -1.2
I P 5-10 65.1 -26.7 -1.5
I P 5-10 59.4 -26.9 -1.3
I P 20-30 94.1 -25.8 -1.4
I P 20-30 84.4 -26.1 -1.0
I P 20-30 86.3 -25.9 -0.8
F C 0-1 166.5 -27.0 -1.7
F C 0-1 102.5 -26.9 -2.8
F C 0-1 70.7 -27.0 -2.3
F C 5-10 70.9 -26.6 -2.5
F C 5-10 72.1 -26.4 -1.9
F C 5-10 72.3 -26.4 -2.2
F C 20-30 113.3 -27.1 -1.3
F C 20-30 124.3 -27.2 -1.6
F C 20-30 139.8 -27.5 -1.7
F N 0-1 55.7 -26.7 -0.2
F N 0-1 50.5 -26.8 -0.8
F N 0-1 55.5 -27.6 -0.9
F N 5-10 47.3 -26.5 -2.8
F N 5-10 42.5 -26.3 -2.1
F N 5-10 36.5 -26.0 -2.1
F N 20-30 89.7 -26.8 -0.7
F N 20-30 78.4 -27.0 -0.7
F N 20-30 51.9 -26.7 -0.6
F P 0-1 57.8 -26.9 -2.0
F P 0-1 50.3 -26.8 -1.2
F P 0-1 61.0 -26.9 -2.6
F P 5-10 42.8 -27.4 -1.1
F P 5-10 50.9 -27.6 -1.2
F P 5-10 47.7 -27.0 -0.4
F P 20-30 68.1 -27.2 -0.5
F P 20-30 65.9 -27.3 -1.6
F P 20-30 87.3 -26.6 -1.0
148

Appendix 2. Leaf litter measurements.

Zone Treatment Depth (cm) C/N (atoms)
I C 0-1 30.7 -28.2 -1.6
I C 0-1 54.5 -27.7 -1.7
I C 0-1 51.7 -27.3 -1.8
I C 5-10 79.1 -27.9 -1.7
I C 5-10 57.2 -26.6 -4.0
I C 5-10 112.9 -27.5 -2.4
I N 0-1 31.6 -27.2 -2.6
I N 0-1 32.5 -26.8 -3.1
I N 0-1 29.6 -27.2 -2.3
I N 5-10 19.9 -26.8 -4.2
I N 5-10 24.2 -26.6 -3.1
I N 5-10 24.1 -26.8 -3.7
I P 0-1 92.6 -29.3 -0.1
I P 0-1 85.8 -28.4 -0.4
I P 0-1 86.9 -28.7 -1.4
I P 5-10 67.8 -27.3 -1.4
I P 5-10 66.3 -25.9 -1.4
I P 5-10 79.8 -26.9 -3.7
F C 0-1 169.1 -29.4 0.6
F C 0-1 149.9 -29.2 0.1
F C 5-10 90.4 -29.8 -0.5
F C 5-10 68.6 -29.4 0.0
F C 5-10 98.1 -29.9 0.0
F C 20-30 165.4 -26.8 -1.0
F C 20-30 156.1 -26.9 -1.0
F C 20-30 158.7 -26.5 -0.1
F N 0-1 124.6 -29.0 -1.3
F N 0-1 68.4 -29.1 -0.5
F N 0-1 139.9 -29.2 -0.3
F N 20-30 47.4 -28.1 -1.0
F N 20-30 43.1 -27.9 -0.8
F N 20-30 43.3 -28.0 -0.9
F P 0-1 128.3 -28.4 -1.5
F P 0-1 106.9 -28.3 -1.1
F P 0-1 110.7 -28.4 -1.3
F P 5-10 150.2 -27.4 -0.7
F P 5-10 121.9 -27.4 -0.6
F P 5-10 227.8 -28.2 -0.9
149
Appendix 3. Live roots measurements.

Zone Treatment Depth (cm) C/N (atoms)
I C 5-10 49.7 -25.5 -1.9
I C 5-10 55.3 -25.3 -1.8
I C 5-10 51.9 -25.7 -1.7
I C 0-1 51.9 -25.7 -2.1
I C 0-1 78.0 -25.9 -3.4
I C 0-1 58.3 -25.9 -2.8
I C 20-30 110.5 -25.1 -8.7
I C 20-30 103.2 -24.7 -8.4
I C 20-30 124.5 -24.9 -10.8
I N 5-10 126.4 -26.1 -2.4
I N 5-10 113.1 -26.2 -2.5
I N 5-10 124.7 -25.8 -3.0
I N 0-1 70.6 -26.6 -5.5
I N 0-1 81.3 -26.7 -5.3
I N 0-1 44.5 -26.0 -5.3
I N 20-30 88.4 -26.0 -0.4
I N 20-30 111.0 -25.6 -2.0
I N 20-30 98.4 -25.7 -1.5
I P 5-10 57.1 -25.7 -1.3
I P 5-10 61.9 -25.5 -1.7
I P 5-10 64.6 -25.8 -1.7
I P 0-1 51.3 -25.9 -1.5
I P 0-1 43.9 -26.3 0.4
I P 0-1 53.7 -25.8 -1.2
I P 20-30 110.9 -25.4 -4.4
I P 20-30 65.6 -25.7 -2.7
I P 20-30 80.1 -25.3 -2.4
F C 5-10 93.3 -26.9 -1.9
F C 5-10 110.2 -26.7 -1.7
F C 5-10 98.3 -27.1 -2.3
F C 0-1 60.4 -27.2 -1.8
F C 0-1 59.3 -27.2 -1.4
F C 0-1 66.2 -27.1 -1.5
F C 20-30 95.2 -26.6 -0.2
F C 20-30 92.7 -26.6 -0.4
F C 20-30 145.7 -27.3 -1.2
F N 5-10 67.3 -27.0 0.5
F N 5-10 75.0 -27.0 0.3
F N 5-10 70.5 -26.9 0.6
F N 0-1 49.9 -26.0 0.3
F N 0-1 47.9 -26.0 -0.1
F N 0-1 38.6 -25.7 -0.1
F N 20-30 134.6 -26.4 -1.4
F N 20-30 118.4 -26.7 -0.4
F N 20-30 90.5 -26.7 -0.5
F P 5-10 148.1 -26.6 -1.1
F P 5-10 146.9 -27.3 -1.4
F P 5-10 144.2 -27.1 -1.9
F P 0-1 57.3 -26.7 -1.3
F P 0-1 54.2 -26.8 -0.6
F P 0-1 40.2 -26.5 -0.4
F P 20-30 119.4 -25.1 -0.7
F P 20-30 114.6 -25.1 -0.8
F P 20-30 135.6 -25.7 -1.4
150
Appendix 4. Total sediment measurements.

Zone Treatment Depth (cm) C/N (atoms)
I C 5-10 27.5 -25.9 -0.7
I C 5-10 28.2 -26.1 -0.6
I C 0-1 35.3 -25.5 -0.8
I C 20-30 42.3 -25.9 -0.4
I C 20-30 45.1 -25.9 -0.4
I C 20-30 38.4 -25.5 -0.7
I N 5-10 48.5 -26.9 -0.7
I N 5-10 44.6 -26.1 -0.4
I N 0-1 26.6 -27.2 -0.9
I N 0-1 19.0 -26.5 -0.8
I N 0-1 20.2 -26.0 -1.0
I N 20-30 25.1 -24.5 0.2
I N 20-30 24.2 -24.5 0.3
I N 20-30 25.1 -24.6 0.0
I P 5-10 30.2 -25.9 -0.7
I P 5-10 30.4 -25.8 -0.8
I P 5-10 32.5 -25.8 -0.9
I P 0-1 21.7 -26.2 -0.8
I P 0-1 21.8 -26.3 -0.9
I P 0-1 21.5 -26.3 -1.0
I P 20-30 27.3 -25.5 -0.1
I P 20-30 26.7 -25.5 -0.4
I P 20-30 26.5 -25.5 -0.3
F C 5-10 152.4 -26.4 -1.0
F C 5-10 39.4 -26.7 -0.5
F C 5-10 33.4 -26.6 -0.2
F C 0-1 34.5 -25.7 -1.2
F C 0-1 37.2 -25.8 -0.7
F C 0-1 36.6 -26.1 -0.5
F C 20-30 38.6 -26.9 -0.3
F C 20-30 37.2 -26.3 -0.1
F N 5-10 24.9 -26.3 -1.1
F N 5-10 21.9 -26.0 -0.5
F N 5-10 23.2 -26.0 -0.4
F N 0-1 30.8 -26.5 -1.0
F N 0-1 29.2 -26.8 -0.7
F N 20-30 29.0 -25.5 -0.4
F N 20-30 29.9 -25.5 -0.6
F N 20-30 32.4 -25.6 -0.7
F P 5-10 26.7 -26.9 0.0
F P 5-10 27.8 -26.6 -0.2
F P 5-10 33.3 -26.8 -1.2
F P 0-1 25.4 -26.2 -0.7
F P 0-1 29.0 -26.7 -0.7
F P 0-1 28.5 -26.1 -0.6
F P 20-30 36.9 -26.6 -1.7
F P 20-30 37.3 -26.5 -0.2
F P 20-30 32.8 -26.6 -1.1
151
Appendix 5. Dead roots, leaf litter and live roots distribution.

Category Zone Treatment Depth Relative abundance (%)
Dead roots Fringe Ctrl 0-5 0.2
Dead roots Fringe Ctrl 0-5 3.0
Dead roots Fringe Ctrl 0-5 17.0
Dead roots Fringe Ctrl 20-30 0.0
Dead roots Fringe Ctrl 20-30 4.9
Dead roots Fringe Ctrl 20-30 5.1
Dead roots Fringe Ctrl 5-10 0.1
Dead roots Fringe Ctrl 5-10 5.0
Dead roots Fringe Ctrl 5-10 9.6
Dead roots Fringe N 0-5 0.1
Dead roots Fringe N 0-5 1.3
Dead roots Fringe N 0-5 18.1
Dead roots Fringe N 20-30 0.9
Dead roots Fringe N 20-30 3.1
Dead roots Fringe N 20-30 8.4
Dead roots Fringe N 5-10 1.2
Dead roots Fringe N 5-10 3.9
Dead roots Fringe N 5-10 7.5
Dead roots Fringe P 0-5 0.1
Dead roots Fringe P 0-5 3.5
Dead roots Fringe P 0-5 16.6
Dead roots Fringe P 20-30 0.1
Dead roots Fringe P 20-30 3.0
Dead roots Fringe P 20-30 5.1
Dead roots Fringe P 5-10 0.0
Dead roots Fringe P 5-10 8.1
Dead roots Fringe P 5-10 11.3
Dead roots Fringe P 5-10 9.3
Dead roots Interior Ctrl 0-5 6.8
Dead roots Interior Ctrl 0-5 9.6
Dead roots Interior Ctrl 0-5 7.1
Dead roots Interior Ctrl 0-5 2.5
Dead roots Interior Ctrl 0-5 0.1
Dead roots Interior Ctrl 0-5 7.4
Dead roots Interior Ctrl 0-5 12.4
Dead roots Interior Ctrl 20-30 0.6
Dead roots Interior Ctrl 20-30 39.1
Dead roots Interior Ctrl 20-30 7.4
Dead roots Interior Ctrl 20-30 8.5
Dead roots Interior Ctrl 20-30 6.0
Dead roots Interior Ctrl 20-30 33.8
Dead roots Interior Ctrl 5-10 1.6
Dead roots Interior Ctrl 5-10 6.0
Dead roots Interior Ctrl 5-10 0.4
Dead roots Interior Ctrl 5-10 9.3
Dead roots Interior Ctrl 5-10 5.4
Dead roots Interior N 0-5 0.7
Dead roots Interior N 0-5 37.5
Dead roots Interior N 0-5 1.9
Dead roots Interior N 0-5 12.5
Dead roots Interior N 20-30 2.8
Dead roots Interior N 20-30 22.9
Dead roots Interior N 20-30 7.6
Dead roots Interior N 20-30 0.5
Dead roots Interior N 20-30 14.9
Dead roots Interior N 5-10 6.6
Dead roots Interior N 5-10 17.8
Dead roots Interior N 5-10 2.3
152
Appendix 5, continue….

Dead roots Interior N 5-10 9.3
Dead roots Interior P 0-5 0.5
Dead roots Interior P 0-5 16.5
Dead roots Interior P 0-5 0.0
Dead roots Interior P 0-5 0.0
Dead roots Interior P 0-5 5.4
Dead roots Interior P 20-30 0.3
Dead roots Interior P 20-30 4.5
Dead roots Interior P 20-30 0.0
Dead roots Interior P 20-30 0.0
Dead roots Interior P 20-30 21.6
Dead roots Interior P 5-10 3.9
Dead roots Interior P 5-10 6.5
Dead roots Interior P 5-10 0.0
Dead roots Interior P 5-10 0.6
Dead roots Interior P 5-10 3.5
Leaf litter Fringe Ctrl 0-5 0.0
Leaf litter Fringe Ctrl 0-5 4.9
Leaf litter Fringe Ctrl 0-5 0.3
Leaf litter Fringe Ctrl 20-30 0.0
Leaf litter Fringe Ctrl 20-30 0.0
Leaf litter Fringe Ctrl 5-10 0.4
Leaf litter Fringe Ctrl 5-10 0.9
Leaf litter Fringe N 0-5 0.0
Leaf litter Fringe N 0-5 1.0
Leaf litter Fringe N 0-5 4.4
Leaf litter Fringe N 20-30 0.5
Leaf litter Fringe N 20-30 0.2
Leaf litter Fringe N 5-10 3.8
Leaf litter Fringe N 5-10 0.8
Leaf litter Fringe P 0-5 0.0
Leaf litter Fringe P 0-5 0.1
Leaf litter Fringe P 0-5 0.1
Leaf litter Fringe P 20-30 0.2
Leaf litter Fringe P 20-30 0.0
Leaf litter Fringe P 5-10 0.0
Leaf litter Fringe P 5-10 0.0
Leaf litter Fringe P 5-10 0.0
Leaf litter Fringe P 5-10 0.3
Leaf litter Interior Ctrl 0-5 0.0
Leaf litter Interior Ctrl 0-5 0.0
Leaf litter Interior Ctrl 0-5 0.0
Leaf litter Interior Ctrl 0-5 0.0
Leaf litter Interior Ctrl 0-5 1.0
Leaf litter Interior Ctrl 0-5 1.5
Leaf litter Interior Ctrl 0-5 0.2
Leaf litter Interior Ctrl 0-5 0.0
Leaf litter Interior Ctrl 0-5 0.0
Leaf litter Interior Ctrl 20-30 0.0
Leaf litter Interior Ctrl 20-30 0.0
Leaf litter Interior Ctrl 20-30 0.0
Leaf litter Interior Ctrl 20-30 0.0
Leaf litter Interior Ctrl 20-30 0.0
Leaf litter Interior Ctrl 20-30 0.1
Leaf litter Interior Ctrl 20-30 0.0
Leaf litter Interior Ctrl 20-30 0.0
Leaf litter Interior Ctrl 20-30 0.0
Leaf litter Interior Ctrl 5-10 0.2
Leaf litter Interior Ctrl 5-10 0.0
Leaf litter Interior Ctrl 5-10 0.7
153
Appendix 5, continue….

Leaf litter Interior Ctrl 5-10 13.7
Leaf litter Interior Ctrl 5-10 8.2
Leaf litter Interior Ctrl 5-10 4.7
Leaf litter Interior N 0-5 0.0
Leaf litter Interior N 0-5 5.4
Leaf litter Interior N 0-5 0.0
Leaf litter Interior N 0-5 0.0
Leaf litter Interior N 0-5 8.5
Leaf litter Interior N 0-5 4.4
Leaf litter Interior N 20-30 0.0
Leaf litter Interior N 20-30 0.0
Leaf litter Interior N 20-30 0.0
Leaf litter Interior N 20-30 0.0
Leaf litter Interior N 20-30 0.0
Leaf litter Interior N 20-30 0.0
Leaf litter Interior N 5-10 0.0
Leaf litter Interior N 5-10 0.0
Leaf litter Interior N 5-10 0.0
Leaf litter Interior N 5-10 0.0
Leaf litter Interior P 0-5 0.0
Leaf litter Interior P 0-5 1.3
Leaf litter Interior P 0-5 0.1
Leaf litter Interior P 0-5 0.0
Leaf litter Interior P 0-5 2.3
Leaf litter Interior P 0-5 0.0
Leaf litter Interior P 20-30 0.0
Leaf litter Interior P 20-30 0.0
Leaf litter Interior P 20-30 0.0
Leaf litter Interior P 20-30 0.0
Leaf litter Interior P 20-30 0.0
Leaf litter Interior P 20-30 0.0
Leaf litter Interior P 5-10 0.0
Leaf litter Interior P 5-10 0.3
Leaf litter Interior P 5-10 3.7
Leaf litter Interior P 5-10 0.0
Live roots Fringe Ctrl 0-5 0.1
Live roots Fringe Ctrl 0-5 0.8
Live roots Fringe Ctrl 0-5 16.5
Live roots Fringe Ctrl 20-30 0.0
Live roots Fringe Ctrl 20-30 32.3
Live roots Fringe Ctrl 20-30 57.7
Live roots Fringe Ctrl 5-10 31.7
Live roots Fringe Ctrl 5-10 2.0
Live roots Fringe Ctrl 5-10 13.3
Live roots Fringe N 0-5 0.1
Live roots Fringe N 0-5 0.3
Live roots Fringe N 0-5 2.3
Live roots Fringe N 20-30 11.1
Live roots Fringe N 20-30 7.7
Live roots Fringe N 20-30 24.5
Live roots Fringe N 5-10 0.2
Live roots Fringe N 5-10 0.8
Live roots Fringe P 0-5 0.1
Live roots Fringe P 0-5 0.3
Live roots Fringe P 0-5 3.0
Live roots Fringe P 20-30 3.2
Live roots Fringe P 20-30 5.8
Live roots Fringe P 20-30 25.3
Live roots Fringe P 5-10 2.5
Live roots Fringe P 5-10 7.2
154
Appendix 5, continue….

Live roots Fringe P 5-10 8.4
Live roots Fringe P 5-10 9.3
Live roots Interior Ctrl 0-5 2.8
Live roots Interior Ctrl 0-5 14.6
Live roots Interior Ctrl 0-5 5.4
Live roots Interior Ctrl 0-5 5.6
Live roots Interior Ctrl 0-5 0.1
Live roots Interior Ctrl 0-5 0.5
Live roots Interior Ctrl 0-5 8.9
Live roots Interior Ctrl 20-30 0.0
Live roots Interior Ctrl 20-30 0.4
Live roots Interior Ctrl 20-30 3.5
Live roots Interior Ctrl 20-30 0.0
Live roots Interior Ctrl 20-30 10.0
Live roots Interior Ctrl 20-30 0.0
Live roots Interior Ctrl 20-30 4.5
Live roots Interior Ctrl 20-30 13.1
Live roots Interior Ctrl 5-10 0.0
Live roots Interior Ctrl 5-10 0.5
Live roots Interior Ctrl 5-10 20.4
Live roots Interior Ctrl 5-10 0.0
Live roots Interior Ctrl 5-10 2.6
Live roots Interior Ctrl 5-10 19.2
Live roots Interior Ctrl 5-10 0.0
Live roots Interior Ctrl 5-10 0.1
Live roots Interior Ctrl 5-10 23.0
Live roots Interior N 0-5 0.0
Live roots Interior N 0-5 0.2
Live roots Interior N 0-5 21.2
Live roots Interior N 0-5 0.0
Live roots Interior N 0-5 6.1
Live roots Interior N 20-30 0.0
Live roots Interior N 20-30 0.4
Live roots Interior N 20-30 0.8
Live roots Interior N 20-30 0.0
Live roots Interior N 20-30 0.1
Live roots Interior N 20-30 10.3
Live roots Interior N 5-10 0.0
Live roots Interior N 5-10 3.8
Live roots Interior N 5-10 9.8
Live roots Interior N 5-10 0.1
Live roots Interior N 5-10 16.6
Live roots Interior P 0-5 25.6
Live roots Interior P 0-5 5.8
Live roots Interior P 0-5 28.8
Live roots Interior P 0-5 0.0
Live roots Interior P 0-5 3.9
Live roots Interior P 0-5 55.6
Live roots Interior P 20-30 0.0
Live roots Interior P 20-30 0.0
Live roots Interior P 20-30 11.4
Live roots Interior P 20-30 0.1
Live roots Interior P 20-30 2.8
Live roots Interior P 5-10 0.0
Live roots Interior P 5-10 1.2
Live roots Interior P 5-10 23.1
Live roots Interior P 5-10 0.1
Live roots Interior P 5-10 3.0
Live roots Interior P 5-10 12.4
155
Appendix 6. PLFAs abundance (ug/g/dw) in the mangrove zones (Fringe: F; Interior: I) at each
nutrient treatment (Control: C; Nitrogen: N; Phosphorus: P).

PLFA Depth (cm) FC FN FP IC IN IP
i14:0 0-1 6.1 17.9 9.3 8.7 10.3 16.8
14:0 0-1 14.6 40.1 21.4 28.1 30.3 72.9
i15:0 0-1 24.8 87.7 44.4 41.2 44.6 79.4
a15:0 0-1 7.1 71.1 38.6 38.0 43.4 62.6
i16:0 0-1 13.7 36.0 19.7 22.9 25.3 37.3
i17:0 0-1 36.7 80.1 57.0 31.0 38.5 66.5
a17:0 0-1 1.3 3.8 0.0 2.5 2.3 8.1
16:1w7 0-1 0.0 58.2 26.7 31.7 36.8 97.9
16:1 0-1 9.7 24.3 18.4 7.7 8.3 17.6
17:0 0-1 6.9 17.9 12.7 15.6 14.4 51.7
10 Me18:0 0-1 8.1 19.7 3.7 3.6 3.9 9.7
6cy18:0 0-1 13.6 36.3 19.0 9.0 11.6 32.2
8cyclo18:0 0-1 46.4 104.1 60.8 61.0 65.7 168.8
i14:0 0-1 7.0 11.9 13.8 3.0 12.7 9.2
14:0 0-1 17.9 29.7 36.4 13.2 45.3 43.3
i15:0 0-1 38.0 63.9 71.9 17.1 63.1 56.6
a15:0 0-1 27.6 49.5 53.4 13.5 64.3 45.9
i16:0 0-1 15.4 23.8 31.0 10.6 42.7 20.6
i17:0 0-1 50.3 50.8 82.6 18.9 78.3 41.3
a17:0 0-1 3.8 7.3 8.4 2.6 11.1 12.6
16:1w7 0-1 39.0 63.2 0.0 21.1 80.7 73.2
16:1 0-1 13.7 14.4 26.0 4.4 10.0 12.4
17:0 0-1 7.1 12.6 15.7 6.2 21.3 25.1
10 Me18:0 0-1 1.8 3.5 5.3 1.7 9.8 2.3
6cy18:0 0-1 18.6 21.9 35.0 6.7 30.8 19.4
8cyclo18:0 0-1 46.7 49.5 81.6 19.3 93.2 119.4
i14:0 0-1 6.6 14.9 11.6 5.8 11.5 13.0
14:0 0-1 16.3 34.9 28.9 20.7 37.8 58.1
i15:0 0-1 31.4 75.8 58.1 29.2 53.8 68.0
a15:0 0-1 17.4 60.3 46.0 25.7 53.9 54.3
i16:0 0-1 14.6 29.9 25.4 16.8 34.0 28.9
i17:0 0-1 43.5 65.5 69.8 24.9 58.4 53.9
a17:0 0-1 2.6 5.5 4.2 2.6 6.7 10.3
16:1w7 0-1 19.5 60.7 13.4 26.4 58.7 85.6
16:1 0-1 11.7 19.4 22.2 6.1 9.2 15.0
17:0 0-1 7.0 15.2 14.2 10.9 17.8 38.4
10 Me18:0 0-1 5.0 11.6 4.5 2.6 6.9 6.0
6cy18:0 0-1 16.1 29.1 27.0 7.9 21.2 25.8
8cyclo18:0 0-1 46.6 76.8 71.2 40.2 79.5 144.1
i14:0 5-10 1.5 14.1 0.8 2.0 1.3 1.6
14:0 5-10 3.8 29.2 2.7 7.8 5.1 6.6
i15:0 5-10 5.5 61.6 3.1 8.3 5.4 6.9
a15:0 5-10 5.0 57.1 2.9 9.6 5.8 7.1
i16:0 5-10 3.4 40.7 1.8 7.0 4.7 5.4
i17:0 5-10 15.4 67.8 8.7 6.9 5.3 11.9
a17:0 5-10 0.3 5.2 0.6 1.2 0.7 1.1
16:1w7 5-10 1.6 51.1 3.2 6.3 3.1 4.4
16:1 5-10 0.0 9.0 0.9 2.9 1.9 2.1
17:0 5-10 1.3 9.4 0.8 2.9 1.7 1.9
10 Me18:0 5-10 0.4 3.8 1.1 1.4 0.3 0.4
6cy18:0 5-10 4.2 28.5 2.5 1.8 1.2 0.5
8cyclo18:0 5-10 7.9 60.2 5.6 6.1 4.6 0.8
i14:0 5-10 3.9 3.9 1.1 1.9 1.6
14:0 5-10 14.4 10.5 5.4 7.3 8.0
i15:0 5-10 21.1 16.3 4.5 9.3 8.7
a15:0 5-10 16.1 13.6 4.8 9.1 8.8
i16:0 5-10 11.3 9.1 4.4 6.4 6.7
i17:0 5-10 61.4 34.6 4.9 6.6 17.1
156
Appendix 6, continue….

a17:0 5-10 3.4 2.0 0.8 1.3 2.1
16:1w7 5-10 20.7 18.5 5.0 10.8 9.8
16:1 5-10 9.7 5.9 0.0 0.7 0.9
17:0 5-10 5.3 4.9 2.3 2.8 3.6
10 Me18:0 5-10 2.0 1.1 0.0 0.8 1.0
6cy18:0 5-10 23.0 8.7 1.4 2.1 3.9
8cyclo18:0 5-10 28.9 17.4 2.7 4.3 7.4
i14:0 5-10 2.7 2.3 1.5 1.6 1.6
14:0 5-10 9.1 6.6 6.6 6.2 7.3
i15:0 5-10 13.3 9.7 6.4 7.3 7.8
a15:0 5-10 10.6 8.3 7.2 7.4 8.0
i16:0 5-10 7.4 5.5 5.7 5.6 6.1
i17:0 5-10 38.4 21.7 5.9 5.9 14.5
a17:0 5-10 1.9 1.3 1.0 1.0 1.6
16:1w7 5-10 11.2 10.9 5.7 7.0 7.1
16:1 5-10 4.8 3.4 1.5 1.3 1.5
17:0 5-10 3.3 2.9 2.6 2.3 2.8
10 Me18:0 5-10 1.2 1.1 0.7 0.5 0.7
6cy18:0 5-10 13.6 5.6 1.6 1.7 2.2
8cyclo18:0 5-10 18.4 11.5 4.4 4.5 4.1
i14:0 20-30 0.3 0.1 0.3 1.6 0.0 0.2
14:0 20-30 1.4 0.9 1.2 5.6 0.6 1.3
i15:0 20-30 1.7 1.0 1.2 4.7 0.6 0.9
a15:0 20-30 1.7 1.1 1.5 4.4 0.6 0.9
i16:0 20-30 1.8 1.0 1.1 4.7 0.6 0.8
i17:0 20-30 2.1 0.9 0.8 1.6 0.3 0.2
a17:0 20-30 0.3 0.4 0.2 0.0 0.0 0.7
16:1w7 20-30 1.4 1.1 1.1 3.1 0.4 0.3
16:1 20-30 0.3 0.2 0.3 1.5 0.2 0.0
17:0 20-30 0.7 0.5 0.5 1.6 0.2 0.5
10 Me18:0 20-30 0.0 0.0 0.1 0.4 0.0 0.0
6cy18:0 20-30 0.6 0.3 0.4 0.7 0.1 0.2
8cyclo18:0 20-30 2.7 0.8 0.4 1.3 0.4 0.3
i14:0 20-30 1.9 2.9 2.7 1.3 1.8 2.1
14:0 20-30 6.0 8.2 6.9 2.5 5.1 7.1
i15:0 20-30 7.0 9.5 10.2 2.1 4.7 6.9
a15:0 20-30 7.2 9.2 4.4 1.7 4.4 6.6
i16:0 20-30 6.0 8.8 7.7 1.9 4.7 6.4
i17:0 20-30 4.1 5.1 4.8 1.0 2.4 3.5
a17:0 20-30 1.2 1.4 1.1 0.0 0.8 1.2
16:1w7 20-30 6.4 9.1 10.5 1.8 4.3 6.3
16:1 20-30 0.7 0.9 1.3 0.0 0.0 0.4
17:0 20-30 2.5 3.1 3.3 0.0 1.9 2.4
10 Me18:0 20-30 0.5 0.8 0.6 0.0 0.4 0.6
6cy18:0 20-30 1.7 1.9 1.9 0.4 0.7 1.3
8cyclo18:0 20-30 2.2 3.2 2.2 0.6 1.2 1.7
i14:0 20-30 1.1 1.5 1.5 1.4 0.9 1.1
14:0 20-30 3.7 4.6 4.1 4.0 2.9 4.2
i15:0 20-30 4.4 5.2 5.7 3.4 2.6 3.9
a15:0 20-30 4.5 5.2 3.0 3.1 2.5 3.8
i16:0 20-30 3.9 4.9 4.4 3.3 2.6 3.6
i17:0 20-30 3.1 3.0 2.8 1.3 1.3 1.9
a17:0 20-30 0.7 0.9 0.6 0.0 0.4 1.0
16:1w7 20-30 3.9 5.1 5.8 2.5 2.3 3.3
16:1 20-30 0.5 0.5 0.8 0.8 0.1 0.2
17:0 20-30 1.6 1.8 1.9 0.8 1.1 1.5
10 Me18:0 20-30 0.2 0.4 0.4 0.2 0.2 0.3
6cy18:0 20-30 1.2 1.1 1.1 0.6 0.4 0.7
8cyclo18:0 20-30 2.5 2.0 1.3 0.9 0.8 1.0
157
Appendix 7. δ
13
C of sediment PLFAs (‰) in the mangrove zones (Fringe: F; Interior: I) at each
nutrient treatment (Control: C; Nitrogen: N; Phosphorus: P).

PLFA Depth (cm) FC FN FP IC IN IP
i15:0 0-1 -31.2 -28.8 -29 -32.1 -28.7 -31.5
a15:0 0-1 -32.2 -29.1 -29.5 -35.7 -29.7 -32.1
i16:0 0-1 -29.6 -29.1 -30.4 -27.8 -29.1 -30.8
i17:0 0-1 -34.8 -32.4 -31.5 -30.3 -29.8 -32
10 Me 18:0 0-1 -34.8 -32.4 -31.5 -30.3 -29.8 -32
16:iw7 0-1 -32 -32.1 -28.1 -27.5 -29.8 -33
16:01 0-1 -32 -32.1 -39.6 -27.5 -29.8 -33
6cy18:0 0-1 -37 -32.8 -32.1 -30.1 -24.7 -35.3
8cyclo18:0 0-1 -29.7 -28.9 -30.8 -25 -29.5 -34.2
i14:0 0-1 -29.2 -26.8 -25 -25.6 -28.4 -30.7
14:00 0-1 -31 -31.1 -31.1 -29.3 -31.1 -34.9
17:00 0-1 -27.3 -31.7 -30.2 -26.7 -24.7 -31.8
15:00 0-1 -36.3 -28.8 -31.4 -30.1 -27.3 -31.3
16:00 0-1 -34 -35.4 -31.6 -32 -33.2 -36.1
18:00 0-1 -26.8 -30.3 -28.2 -28.5 -29.7 -32.3
i15:0 0-1 -30.8 -25.9 -28.2 -30.4 -29.3 -29.3
a15:0 0-1 -32.2 -26.6 -29.8 -30.6 -29.8 -30.3
i16:0 0-1 -25.6 -27.2 -23.8 -29.9 -27.4 -28.1
i17:0 0-1 -29.4 -30.4 -23.8 -29.9 -27.4 -28.1
10 Me 18:0 0-1 -30.8 -27.7 . . . .
16:iw7 0-1 -27.2 -27.7 -25.5 -28.6 -27.2 .
16:01 0-1 -27.2 -27.7 -27.5 -28.6 -27.2 .
6cy18:0 0-1 -28.8 -32.4 -27.1 . . -23.8
8cyclo18:0 0-1 -28.8 -30.5 -29.8 . -27.7 -27.4
i14:0 0-1 -25.7 -30.4 -25 -28.9 -27.7 -24.3
14:00 0-1 -27.6 -30.3 -27.6 -32.9 -30 -31.1
17:00 0-1 -27.6 -26.5 -24.9 -29.5 -23.1 -26.1
15:00 0-1 -29 -28.4 -24.3 -30.3 -30 -30.3
16:00 0-1 -29.2 -28.6 -27.7 -32.2 -36 -32.4
18:00 0-1 -24.3 -26.6 -23.1 -30.7 -28.4 -29
i15:0 5-10 -38 -30.7 -34.5 -28 -31.1 -32.6
a15:0 5-10 -39.5 -31.8 -35.5 -30.6 -32.4 -33.3
i16:0 5-10 -32.6 -28.7 -33.4 -26.2 -29 -30.9
i17:0 5-10 -40.7 -36 -38 -24.9 -28.8 -33.3
10 Me 18:0 5-10 -40.7 -36 -38 -24.9 -28.8 -33.3
16:iw7 5-10 . -34.4 -35.4 -28.8 -29.1 -30.7
16:01 5-10 . -34.4 -35.4 -28.8 -29.1 -30.7
6cy18:0 5-10 -38.4 -34.1 -39.1 . -21.9 -33.3
8cyclo18:0 5-10 -30.1 -31.2 -33.2 . -21.9 -28.4
i14:0 5-10 -31 -30 -34.1 . -30.6 -28.4
14:00 5-10 -33.4 -31.7 -34.5 -28.8 -31.6 .
17:00 5-10 -28.3 -32 -30.1 -25.4 -22 -26.3
15:00 5-10 -30.7 -29.9 -32.2 -27.3 -28.7 -28.7
16:00 5-10 -36 -33 -34.5 -28.7 -30 -32.1
18:00 5-10 -33.4 -27.4 -29.5 -24 -27.5 -31.9
i15:0 5-10 -30.5 -27.4 -28.3 -31.8 -27.1 -28
a15:0 5-10 -32.1 . -33 -33.4 -28.3 -29.4
i16:0 5-10 -32.1 . -27.5 -31.6 -26.3 -27.6
i17:0 5-10 -37.6 . -35.1 -31.6 -26.3 -26.8
10 Me 18:0 5-10 -37.6 . -28.3 . . -26.8
16:iw7 5-10 . . -28.3 . . .
6cy18:0 5-10 -32.9 . -32.2 . . .
8cyclo18:0 5-10 -28.8 . -33.5 -37.1 . .
i14:0 5-10 -28.8 . -33.5 -37.1 . -27
14:00 5-10 -33.6 . -29.6 -33.1 -28.5 -28.3
17:00 5-10 -30.8 . -29.6 -33.1 -28.5 -28.3
158
Appendix 7, continue….

15:00 5-10 -33.5 . . . . -25.3
16:00 5-10 -35 . -31 -31.2 -28.1 -29.1
18:00 5-10 -27.1 . -25 -30 -28.1 -24.2
i15:0 20-30 -33.2 -30.2 -31.3 -31.6 -29.9 -32.7
a15:0 20-30 -34.9 -33.5 -33.9 -33 -31.3 -34.6
i16:0 20-30 -33 -30.8 -33.9 -22.8 -30.9 -31.6
i17:0 20-30 -33 -30.8 -35.2 -22.8 -30.9 -31.6
10 Me 18:0 20-30 -25.6 . -35.2 . . -26.3
16:iw7 20-30 -25.6 -27.4 . -29.3 -35.6 -26.3
16:01 20-30 . -27.4 . -29.3 -35.6 .
6cy18:0 20-30 -35.5 . . . . -26.1
8cyclo18:0 20-30 -35.5 . . . . -26.1
i14:0 20-30 -29.2 -26.1 -28.9 -27.5 -28.8 -30.1
14:00 20-30 -30.9 -33.5 -30.2 -29.9 -28.5 -30.6
17:00 20-30 -24.6 -33.5 -25.2 -29.9 -26.1 -25.6
15:00 20-30 -29.1 -28.2 -27.3 -31.8 -28.7 -29.7
16:00 20-30 -31.4 -31.2 -26.7 -26.1 -27.7 -30.5
18:00 20-30 -28.7 -31.2 -26.6 -27 -25.8 -26.8
i15:0 20-30 -27.4 . -26.6 -30.4 -25.8 -27.6
a15:0 20-30 -29 . . -30.8 . -30.4
i16:0 20-30 -29.3 . . -30.6 . -30.4
i17:0 20-30 -29.3 . . -30.6 . -30.4
8cyclo18:0 20-30 -26.6 . . . . .
i14:0 20-30 -26.6 . . -28.3 . .
14:00 20-30 -27.9 . . -30.6 . -27.2
17:00 20-30 -27.9 . . -30.6 . -27.2
15:00 20-30 . . . -29.8 . .
16:00 20-30 . . . -29.6 . -25.3
18:00 20-30 -23.9 . . -29.2 . -23.6
159
Appendix 8. Pore water parameters in the mangrove zones (Fringe: F; Interior: I) in each nutrient
treatment (Control: C; Nitrogen: N; Phosphorus: P) and depth interval (0-5, 5-10, 20-30 cm).

Zone Treatment
Depth
(cm) pH
Salinity
(ppt)
NH
4
+

(µM)
H
2
S
(mM)
PO
4
-

(µM)
Temp
(
o
C)
Interior Ctrl 0-5 6.56 35 9.41 . 2.94 .
Interior Ctrl 0-5 6.55 35 28.73 0.09 1.99 .
Interior Ctrl 0-5 6.64 34 19.18 . 1.09 .
Interior Ctrl 0-5 6.64 36 22.59 0.14 2.97 .
Interior Ctrl 0-5 6.64 36 45.36 0.14 2.97 33.1
Interior Ctrl 0-5 5.68 46 41.07 . . 34.0
Interior Ctrl 0-5 5.65 45 64.95 0.24 0.51 33.4
Interior Ctrl 0-5 5.68 46 39.45 0.25 0.58 33.3
Interior Ctrl 0-5 5.84 41 46.45 0.24 0.86 32.4
Interior Ctrl 0-5 5.70 44 59.95 0.45 0.68 32.0
Interior Ctrl 0-5 5.73 42 72.95 0.80 0.48 37.7
Interior Ctrl 0-5 7.09 36 . . 1.00 38.0
Interior Ctrl 0-5 6.83 48 . . 1.00 38.4
Interior Ctrl 5-10 7.24 40 77.16 . 7.62 32.4
Interior Ctrl 5-10 5.75 46 65.95 0.42 2.79 33.4
Interior Ctrl 5-10 5.73 45 47.95 0.70 0.64 32.9
Interior Ctrl 5-10 5.71 45 63.45 0.46 0.76 32.7
Interior Ctrl 5-10 5.77 40 62.45 0.60 1.13 31.8
Interior Ctrl 5-10 5.58 45 89.95 1.43 1.44 32.3
Interior Ctrl 5-10 5.69 41 . . 0.76 36.6
Interior Ctrl 5-10 6.51 37 . . 0.76 36.0
Interior Ctrl 5-10 7.40 42 47.56 . 4.76 36.2
Interior Ctrl 5-10 7.01 37 . . . 35.1
Interior Ctrl 5-10 6.47 36 . . . 33.9
Interior Ctrl 20-30 5.73 44 45.45 0.85 1.06 32.8
Interior Ctrl 20-30 5.66 45 40.95 0.92 0.77 32.5
Interior Ctrl 20-30 5.70 42 38.95 0.82 0.79 33.3
Interior Ctrl 20-30 5.87 44 82.45 1.57 0.95 32.8
Interior Ctrl 20-30 5.68 45 75.95 1.59 1.16 32.2
Interior Ctrl 20-30 5.67 45 99.45 1.81 . 32.1
Interior Ctrl 20-30 6.48 40 99.45 1.81 . 35.8
Interior Ctrl 20-30 6.31 40 . . . 35.1
Interior Ctrl 20-30 6.43 39 . . . 35.5
Interior Ctrl 20-30 6.41 43 . . . 35.2
Interior Ctrl 20-30 7.11 39 . . . 36.3
Interior N 0-5 7.21 42 370.36 . 2.67 30.6
Interior N 0-5 6.49 36 26.00 . 2.37 .
Interior N 0-5 6.48 36 4.86 0.07 3.65 .
Interior N 0-5 6.74 31 3.73 0.47 0.15 .
Interior N 0-5 6.49 36 3.73 0.48 3.60 .
Interior N 0-5 6.49 36 96.79 0.48 3.60 .
Interior N 0-5 6.39 45 65.95 0.47 0.54 31.2
Interior N 0-5 6.25 44 48.95 0.23 0.46 30.8
160
Interior N 0-5 6.62 40 296.45 0.33 0.52 30.7
Interior N 0-5 6.57 40 100.95 0.13 0.88 31.0
Interior N 0-5 6.81 40 109.95 0.17 0.12 31.1
Interior N 0-5 6.69 36 109.95 0.17 0.12 35.0
Interior N 0-5 6.99 41 163.16 . 15.41 34.7
Interior N 5-10 7.01 40 289.56 . 2.71 .
Interior N 5-10 6.47 43 69.95 0.21 0.89 32.7
Interior N 5-10 6.47 43 69.95 0.21 . 32.7
Interior N 5-10 6.42 41 751.45 0.40 0.58 30.5
Interior N 5-10 6.28 42 660.95 0.60 0.76 31.4
Interior N 5-10 6.61 41 434.95 0.46 0.70 31.6
Interior N 5-10 6.14 40 434.95 0.46 0.70 33.8
Interior N 5-10 6.27 37 . . . 34.0
Interior N 5-10 6.36 36 . . . 34.4
Interior N 5-10 6.24 38 . . . 33.6
Interior N 20-30 7.26 44 103.96 . 1.47 33.2
Interior N 20-30 6.16 45 136.45 0.59 1.15 29.7
Interior N 20-30 6.74 45 64.95 0.73 1.45 32.4
Interior N 20-30 6.74 45 64.95 0.73 . 32.4
Interior N 20-30 6.55 43 3191.70 3.50 0.80 32.2
Interior N 20-30 6.65 43 3773.70 3.78 1.16 30.9
Interior N 20-30 6.64 41 2723.70 3.81 0.26 30.3
Interior N 20-30 5.93 45 2723.70 3.81 0.26 33.5
Interior N 20-30 6.57 42 . . . 34.2
Interior N 20-30 6.53 43 . . . 34.2
Interior N 20-30 6.29 44 . . . 34.8
Interior N 20-30 6.43 43 . . . 34.1
Interior N 20-30 6.84 42 . . . 34.9
Interior P 0-5 6.55 44 137.16 . . 31.2
Interior P 0-5 6.68 35 5.09 0.58 . .
Interior P 0-5 6.45 36 3.50 0.03 1.15 .
Interior P 0-5 6.61 38 3.50 0.49 11.47 .
Interior P 0-5 6.61 38 12.50 0.49 11.47 .
Interior P 0-5 . . 14.64 . . .
Interior P 0-5 . . 152.50 . . .
Interior P 0-5 6.20 45 47.45 0.57 51.83 31.5
Interior P 0-5 6.02 45 57.95 0.22 20.84 31.4
Interior P 0-5 6.14 44 55.45 0.60 63.82 31.2
Interior P 0-5 6.16 44 54.45 0.37 87.99 30.0
Interior P 0-5 6.16 45 63.95 0.15 54.88 30.0
Interior P 0-5 5.98 50 67.95 0.19 5.84 30.2
Interior P 0-5 6.49 35 67.95 0.19 5.84 33.6
Interior P 5-10 6.30 47 74.36 . . 33.8
Interior P 5-10 . 41 79.56 . . .
Interior P 5-10 6.05 46 38.45 0.64 89.92 31.3
Interior P 5-10 6.19 43 43.45 0.72 95.12 30.6
Interior P 5-10 6.10 45 56.45 0.52 132.21 31.4
Interior P 5-10 6.08 45 45.95 0.28 57.26 29.0
161
Interior P 5-10 5.99 46 70.45 0.24 99.28 29.5
Interior P 5-10 6.06 44 43.45 0.18 150.86 29.8
Interior P 5-10 6.31 38 43.45 0.18 150.86 33.2
Interior P 5-10 6.14 37 . . . 32.6
Interior P 5-10 6.27 37 . . . 32.6
Interior P 20-30 6.80 40 161.56 . . 32.5
Interior P 20-30 6.16 45 55.45 0.39 391.35 30.8
Interior P 20-30 6.34 45 57.45 0.46 72.15 33.1
Interior P 20-30 6.34 45 57.45 0.46 . 33.1
Interior P 20-30 6.07 45 84.95 0.54 287.27 29.6
Interior P 20-30 6.13 43 44.45 0.41 364.06 31.2
Interior P 20-30 6.15 44 69.95 0.32 188.45 29.1
Interior P 20-30 6.61 42 69.95 0.32 188.45 34.4
Interior P 20-30 6.18 40 . . . 33.2
Interior P 20-30 6.21 41 . . . 32.6
Interior P 20-30 6.39 42 . . . 33.0
Interior P 20-30 6.15 41 . . . 32.8
Interior P 20-30 6.21 41 . . . 33.1
Fringe Ctrl 0-5 6.36 50 75.96 . 36.19 35.0
Fringe Ctrl 0-5 6.53 32 10.77 0.19 6.59 .
Fringe Ctrl 0-5 6.35 38 4.41 0.73 5.50 .
Fringe Ctrl 0-5 6.48 35 3.95 0.52 5.86 .
Fringe Ctrl 0-5 6.40 35 158.21 0.62 18.40
Fringe Ctrl 0-5 6.40 41 42.50 0.62 18.40 31.7
Fringe Ctrl 0-5 6.54 41 38.21 . . 32.1
Fringe Ctrl 0-5 6.54 40 38.21 . . 31.8
Fringe Ctrl 5-10 6.66 48 97.96 0.49 16.62 31.8
Fringe Ctrl 5-10 6.98 50 103.96 1.03 19.58 31.6
Fringe Ctrl 5-10 6.23 44 103.96 0.82 19.58 31.6
Fringe Ctrl 5-10 6.31 44 . 1.02 . 32.1
Fringe Ctrl 5-10 6.66 40 . 1.03 . 31.3
Fringe Ctrl 5-10 6.31 42 . . . 31.4
Fringe Ctrl 20-30 7.28 45 87.96 . 13.38 31.5
Fringe Ctrl 20-30 6.54 42 87.96 . 13.38 31.1
Fringe Ctrl 20-30 6.81 39 . . . 30.8
Fringe Ctrl 20-30 6.41 43 . . . 30.7
Fringe Ctrl 20-30 6.71 43 . . . 31.0
Fringe Ctrl 20-30 6.52 42 . . . 31.4
Fringe Ctrl 20-30 6.49 42 . . . 31.1
Fringe N 0-5 6.90 45 . . . 35.1
Fringe N 0-5 6.44 45 11.23 0.67 6.65 .
Fringe N 0-5 6.45 36 80.09 1.65 9.94 .
Fringe N 0-5 6.75 35 4.86 0.45 9.80 .
Fringe N 0-5 6.63 42 0.00 0.07 2.20 .
Fringe N 0-5 6.63 42 41.79 0.07 2.20 31.4
Fringe N 0-5 6.46 42 583.93 . . 31.0
Fringe N 0-5 6.64 40 52.50 . . 30.5
Fringe N 5-10 6.56 45 110.36 . 16.12 30.5
162
Fringe N 5-10 6.51 43 88.76 . 11.48 31.4
Fringe N 5-10 6.60 43 88.76 . 11.48 31.1
Fringe N 5-10 6.37 43 . . . 30.8
Fringe N 5-10 6.41 41 . . . 30.9
Fringe N 5-10 6.53 44 . . . 30.8
Fringe N 20-30 7.69 45 110.36 . 0.00 31.1
Fringe N 20-30 6.37 44 110.36 . 0.00 31.8
Fringe N 20-30 6.54 43 . . . 31.7
Fringe N 20-30 6.11 48 . . . 31.9
Fringe N 20-30 6.63 41 . . . 30.7
Fringe N 20-30 6.46 42 . . . 31.0
Fringe N 20-30 6.36 45 . . . 30.7
Fringe P 0-5 6.62 38 10.77 0.53 91.17 32.6
Fringe P 0-5 6.41 37 5.77 0.34 23.39 .
Fringe P 0-5 6.48 36 4.64 0.39 111.40 .
Fringe P 0-5 6.35 37 43.93 0.39 111.40 32.5
Fringe P 0-5 6.35 37 45.36 32.4
Fringe P 0-5 6.78 37 45.36 . . 31.3
Fringe P 5-10 6.14 40 110.76 . . 31.3
Fringe P 5-10 6.81 40 134.76 . . 31.1
Fringe P 5-10 6.32 40 134.76 . . 32.3
Fringe P 5-10 6.31 40 . . . 31.8
Fringe P 5-10 6.29 41 . . . 32.4
Fringe P 5-10 6.21 41 . . . 32.1
Fringe P 20-30 7.11 40 113.56 . . 32.0
Fringe P 20-30 6.39 45 113.56 . . 31.7
Fringe P 20-30 6.28 39 . . . 30.5
Fringe P 20-30 6.24 41 . . . 31.0
Fringe P 20-30 6.63 40 . . . 31.6
Fringe P 20-30 6.21 43 . . . 31.2
Fringe P 20-30 6.64 39 . . . 30.8

Abstract (if available)

Abstract The ecological interaction of plants and bacteria was studied in a peat-based sediment subjected to long-term fertilization with nitrogen and phosphorus at Twin Cays, Belize. The main purpose of this research was to better understand the functional relationship among microorganisms, mangrove trees and sediment geochemistry in mangrove sediments. Specifically, the variability of broad bacterial community structure, the bacterial carbon source utilization patterns, the spatial and temporal dynamics of nitrogen-fixing populations, and the molecular diversity of nitrogen fixers were studied. A combination of molecular and chemical techniques combined with statistical tools was used for the identification of key biological and environmental factors directly controlling the community structure of microorganisms in mangrove sediments. Results showed that mangrove trees strongly affect the activity and community composition of N2 fixers, but not the whole bacterial community (based on PLFAs) or taxonomic traits (based on phylogenetic analysis). In most cases roots were inversely related to N2 fixation rates and N2 fixers (community composition), primarily under fertilized conditions. The functional relationship among microorganisms, plants and sediment geochemistry in Belize showed that bacteria rely on degrading organic matter from mangroves as a primary source for carbon, and that mangrove roots do not confer a stable microenvironment that promotes stability and persistence in microbial populations. Effects of the long-term fertilization with N or P on bacteria and N2 fixers revealed that effects depend on the initial conditions prior to disturbance, and that effects include changes in microbial metabolic pathways and in community composition patterns of microbial functional groups. Remarkably, variability of bacteria and N2 fixers in mangrove sediments was observed in response to natural environmental conditions and also to fertilization.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Biological nitrogen fixation associated with living and decomposing macroalgae

Asset Metadata

Creator Romero, Isabel C. (author)

Core Title Plant-microbial interactions in mangrove sediments under different nutrient conditions

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Ocean Sciences

Publication Date 08/17/2009

Defense Date 07/14/2009

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Belize,diazotrophs,generalized additive models,mangrove ecosystem,marine sediments,microbial ecology,multivariate analysis,nitrogen fixation,OAI-PMH Harvest,phospholipid fatty acids,phylogenetic analysis,stable isotopes,statistical modelling

Place Name Belize (countries), islands: Twin Cays (geographic subject)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Capone, Douglas (committee chair), Jacobson, Myrna (committee chair), Fogel, Marilyn (committee member), Fuhrman, Jed Alan (committee member), Rinconscente, Michelle (committee member)

Creator Email iromero@usc.edu,isaromero12@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2567

Unique identifier UC1147744

Identifier etd-Romero-3205 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-254153 (legacy record id),usctheses-m2567 (legacy record id)

Legacy Identifier etd-Romero-3205.pdf

Dmrecord 254153

Document Type Dissertation

Rights Romero, Isabel C.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu