ABSTRACT 
 
 
 
 
Title of Document: NITROGEN CYCLING AND CONTROLS ON 
DENITRIFICATION IN MESOHALINE 
SEDIMENTS OF CHESAPEAKE BAY   
 
 Michael S. Owens, Master of Science, 2009 
Directed By: Dr.  Jeffrey C. Cornwell, Marine, Estuarine and 
Environmental Science Program 
 
 
Nitrogen is a key nutrient in the eutrophication of coastal and estuarine systems.  In 
shallow water systems, sediment recycling can be an important source of nutrients for 
phytoplankton growth.  The balance between nitrogen recycling and denitrification 
regulates the importance of sediments as a nitrogen source.  To assess controls on 
denitrification, we conducted intensive seasonal measurements of sediment water 
exchange and denitrification using sediment core incubations.   Peak rates of 
denitrification were observed in fall and spring (>100 ?mol N
2
-N m
-2
 h
-1
) followed by 
a decrease to 10 ?mol N m
-2
 h
-1
 in summer.  Although denitrification rates were 
stimulated by labile organic carbon additions from the water column, the overall 
efficiency of the process sharply declined as temperature increased and bottom water 
O
2
 declined.   Macrofauna activity was shown to enhance sediment transport of O
2
 by 
>5 fold, increase organic matter decomposition and maintain a high rate of 
denitrification efficiency. 
  
 
 
 
 
NITROGEN CYCLING AND CONTROLS ON DENITRIFICATION IN 
MESOHALINE SEDIMENTS OF CHESAPEAKE BAY      
 
 
 
By 
 
 
Michael S. Owens 
 
 
 
 
 
Thesis submitted to the Faculty of the Graduate School of the  
University of Maryland, College Park, in partial fulfillment 
of the requirements for the degree of 
Master of Science 
2009 
 
 
 
 
 
 
 
 
 
 
Advisory Committee: 
Professor Jeffrey C. Cornwell, Chair 
Professor Todd M. Kana 
Professor Louis A. Codispoti 
 
 
 
 
 
  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
? Copyright by 
Michael S. Owens 
2009 
 
 
 
 
 
 
 
 
 
 
 
 
 
  
 
Preface 
 
Nitrogen has been identified as key nutrient in the eutrophication of coastal and 
estuarine systems.  These shallow water environments are impacted by many sources 
of nitrogen including groundwater discharge, land runoff, atmospheric deposition, 
and sewage.  Once nitrogen enters the water column, given sufficient light and 
nutrients, growth of phytoplankton is stimulated.   Many coastal and estuarine 
systems have been shown to be N limited in terms of phytoplankton production, and 
the over fertilization of these systems has caused a decline in ecosystem health.  
Phytoplankton blooms decrease water column light transparency and inhibit the 
growth of benthic algae and submerged aquatic grass through light limitation.  In 
shallow water systems, dead phytoplankton cells rapidly sink and are decomposed in 
the sediment.  This decomposition can be an important process that promotes and 
maintains hypoxic/anoxic bottom water conditions in many shallow water systems. 
 Dead phytoplankton cells that settle to the sediment decay and nutrients are 
regenerated through heterotrophic processes.  The remineralized N that is formed 
during this decomposition can either be recycled back to the water column as 
dissolved inorganic nitrogen DIN (NO
3
-
+NO
2
-
+ NH
4
+
), dissolved organic nitrogen 
DON, or be denitrified to N
2
 gas (Figure P-1).   The NH
4
+
 that is produced through 
ammonification can be released back to the water column or be transformed into NO
3
-
 
through the aerobic chemoautotrophic process of nitrification. Heterotrophic 
denitrifying bacteria use NO
3
- 
as a terminal electron acceptor during the breakdown of 
organic matter and produce N
2
 gas as an end product.   Nitrification and 
 ii 
 
 
denitrification are often closely coupled, providing efficient removal of remineralized 
N.   Nitrate that is not denitrified can also be transported back to the water column.  
Some DON can be lost directly to the water column or decomposed in the sediment.  
Measurements of DON flux from the sediment often do not show consistent trends 
and in many cases DON does not appear to be a significant source or sink for N. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
PON
DON
PON Deposition
(Phytodetritus)
Ammonification
Nitrification
Denitrification
DON
Flux Flux
WATER
SEDIMENT
Burial
Bioirrigation
O
2
O
2
O
2
A
d
v
e
c
t
io
n
D
i
ffu
sion
DNRA
Aerobic
Anerobic
NH
4
+
Flux
NO
3
-
N
2
Flux
N
2
NH
4
+
NO
3
-
NO
2
-
ANAMMOX
D
i
ffu
sion
 
 
 
   Fig. P-1. Conceptual diagram of the nitrogen cycle.  Particulate organic 
matter settles to the sediment surface and is decomposed.  Remineralized N can 
be buried, recycled back to the water column, or denitrified.  Oxygen 
dependant nitrification can be closely coupled to denitrification leading to 
highly efficient removal of N as N
2
 gas. In sulfidic sediments nitrification can 
be inhibited and DNRA may out compete denitrifiers for NO
3
- 
.  Bioirrigation 
can play an important role in the nitrogen cycle by oxygenating the sediment, 
preventing the buildup of H
2
S, enhancing denitrification and ANAMMOX. 
Herein we define denitrification as conventional heterotrophic denitrification, 
anammox and all other biogenic process that convert fixed-N to N
2. 
 iii 
 
 
Denitrification has been identified as the most important process removing nitrogen 
from many estuaries.  The recycling of DIN from sediments can be an important 
source of N fueling phytoplankton blooms.  Many factors can influence the balance 
between DIN release from the sediment and N removal via denitrification including 
the depth penetration of O
2
 , organic carbon loading, bioirrigation, and pore water 
H
2
S concentrations.   Past studies of denitrification have been hampered by poor 
techniques that were time consuming or involved indirect measures of the process 
using chemical blocks.  For this reason, few intensive studies of denitrification have 
been conducted and most studies report data from a number of locations in a system 
sampled a single time.  In this study, we use membrane inlet mass spectrometry 
(MIMS) to obtain a direct measure of denitrification on flux incubations.  The MIMS 
technique has the advantage of measuring all pathways to N
2
 production including 
anammox (Thamdrup & Dalsgaard 2002) and oxidation of NH
4
+
 by MnO
2 
(Luther et 
al. 1997).    The logistic efficiency of our incubation methods allows us to examine 
seasonal changes in the nitrogen cycle at multiple locations.   
The progress of eutrophication often leads to reduced bottom water O
2
 and an 
increased organic carbon load to the sediment.   These two factors can cause 
biogeochemical changes in the sediment leading to; 1) a decrease in the depth 
penetration of O
2
; 2) lower rates of coupled nitrification/denitrification and 
denitrification efficiency;  and  3)  a greater proportion of carbon metabolism via 
sulfate reduction.    As O
2
 penetration decreases there is a shift from oxic/suboxic 
metabolism to sulfate reduction and often an increase in pore water H
2
S 
concentrations.  Pore water sulfides have been shown to limit nitrification and 
 iv 
 
 
promote dissimilatory NO
3
-
 reduction to NH
4
+ 
(Figure P-1). The overall impact of 
these biogeochemical changes is a shift from sediments being a sink for fixed N 
through denitrification to a source of N as DIN is recycled back to the water column.   
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Diffusion
Advection
Hypoxic
Aerobic
 
 
  Fig. P-2 Conceptual diagram showing the influence of macrofauna on the 
nitrogen cycle. Under aerobic conditions (left panel) O
2
 is transported into the 
sediment by both molecular diffusion and advection via bioirrigation.   Under 
aerobic conditions sediments are more oxidized and organic matter is rapidly 
decomposed.  Under low water column O
2
 conditions (right panel) advection of 
O
2
 may be lost and the depth of sediment O
2
 penetration will decrease.  This 
decrease in supply of O
2
 will result in a greater proportion of DIN recycling. 
Another consequence of reduced bottom water O
2
 is the loss of habitat  
suitable for macrofauna (Figure P-2).  Many species of macrofauna irrigate the 
sediment by flushing O
2
 rich water through their burrows.  The overall effect of 
bioirrigation is to increase the surface area of sediment where oxic and suboxic 
process can occur and to diminish the buildup of pore water sulfide.  Bioirrigation 
 v 
 
 
should have a positive impact on the removal of fixed N from the sediments by 
promoting higher rates of nitrification.   Polychaetes can greatly enhance the rate of 
sediment organic matter decomposition and decrease the amount of carbon available 
for sulfate reduction.   
In this study we will follow seasonal changes in nutrient flux and 
denitrification with emphasis on factors controlling the rate of denitrification and 
shifts in denitrification efficiency.  Intensive sampling during the spring bloom will 
provide insight into how the sediments respond to labile organic matter settling to the 
bottom.  We will also evaluate the rate of bioirrigation and the depth penetration of 
O
2
 to determine how these factors may influence the nitrogen cycle.   
 
 
 
 
 
 
 
 
 
 
 
 
 vi 
 
 
Dedication 
Kabrena Owens 
My best friend and the love of my life  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 vii 
 
 
Acknowledgements 
 I would like thank my graduate committee members; Dr. Jeffrey C. Cornwell, 
Todd M. Kana, and Louis A. Codispoti for their guidance and support.  I greatly 
appreciate my advisor, Jeff Cornwell, who has always encouraged me to pursue a 
graduate degree and provided the opportunity for me develop not only my technical 
expertise but also my aptitude as a scientist.   I would like to thank Todd Kana for his  
mass spectrometer wizardry and taking the time to teach me his craft.  This work 
would not have been possible without Membrane Inlet Mass Spectrometry.  I would 
like to thank Lou Codispodti for agreeing to serve on my committee and his 
enthusiasm for the nitrogen cycle.   
 I thank Pete Sampou  for believing in my abilities and teaching me the basics 
of sediment biogeochemistry during my first years at Horn Point.  I would like to 
express appreciation for those who assisted me in the laboratory and field work: Lora 
Pride, Erica Kiss, Jeff Alexander, Chris Chick and Teresa Coley.  I especially thank 
Jack Seabrease for his help in the design and construction of flux chambers and 
sampling devices.  I appreciate Jack?s positive can do attitude that seems to be 
contagious. I thank my mother and father for unconditional love and encouragement 
in all my endeavors.   Lastly, I would like to thank my wife and daughters who have 
been supportive and understanding when this work consumed nights and weekends 
and took priority over them.   
 Financial support for this work was provided by the Maryland Sea Grant 
program and the National Science Foundation Biocomplexity program. 
 viii 
 
 
Table of Contents 
Preface........................................................................................................................... ii 
Dedication................................................................................................................... vii 
Acknowledgements....................................................................................................viii 
Table of Contents......................................................................................................... ix 
List of Figures............................................................................................................... x 
Chapter 1: Denitrification Efficiency in Chesapeake Bay Sediments .......................... 1 
Abstract................................................................................................................... 2 
Introduction............................................................................................................. 3 
Methods................................................................................................................... 6 
Sampling Location............................................................................................ 6 
Sample Collection............................................................................................. 7 
Sediment-Water Exchange................................................................................ 8 
Gas, Solute and Sediment Analysis .................................................................. 9 
Statistical Analysis............................................................................................ 9 
Results................................................................................................................... 11 
Multiple Linear Regression Analysis.............................................................. 13 
Denitrification Efficiency ............................................................................... 16 
Temporal versus Spatial Variability ............................................................... 19 
Water Column and Sediment Chlorophyll a................................................... 20 
Conclusions........................................................................................................... 23 
Chapter 2: Seasonal Changes in Sedimentary Denitrification Efficiency: The effects 
of Bioirrigation and Reduced Bottom Water Oxygen ................................................ 40 
Abstract................................................................................................................. 41 
Introduction........................................................................................................... 42 
Methods................................................................................................................. 44 
Sampling Location.......................................................................................... 44 
Sample Collection........................................................................................... 45 
Sediment-Water Exchange.............................................................................. 46 
Microelectode oxygen measurements............................................................. 47 
Bioirrigation measurements............................................................................ 47 
Gas, Solute and Sediment Analysis ................................................................ 49 
Infaunal sampling............................................................................................ 50 
Results................................................................................................................... 50 
Sediment water exchange ............................................................................... 50 
Bottom water O
2
 and Polychaete Abundance................................................. 51 
Sediment microelectrode O
2
 measurements ................................................... 52 
Denitrification Efficiency ............................................................................... 53 
Discussion............................................................................................................. 54 
Seasonal changes in bioirrigation ................................................................... 54 
Bioirrigation and Nutrient Cycling ................................................................. 56 
Denitrification Efficiency ............................................................................... 58 
Implications........................................................................................................... 59 
Bibliography ......................................................................................................... 75 
 
 ix 
 
 
List of Figures 
 
Figure P-1 Conceptual diagram of the nitrogen cycle. 
 
Figure P-2 Conceptual diagram showing the influence of macrofauna on the 
nitrogen cycle. 
 
Figure 1.1 Map of the Choptank River - a subestuary of the Chesapeake Bay.    
 
Figure 1.2 Sample data of time course changes in N
2
 measured with the MIMS 
system.   
 
Figure 1.3  Sediment-water exchange data for July 2000 through May 2001 for 
SOD.   
 
Figure 1.4 Sediment-water exchange data for July 2000 through May 2001 for 
N
2
.   
 
Figure 1.5 Sediment-water exchange data for July 2000 through May 2001 for 
NO
3
-
+NO
2
-
.  
 
Figure 1.6 Sediment-water exchange data for July 2000 through May 2001 for 
NH
4
+
.  
 
Figure 1.7 Bottom water temperature and O
2
 concentrations for all three transect 
sites measured monthly. 
 
Figure 1.8 Sediment Cholorphyll a from the top 1.0 cm of sediment.  Error bars 
represent standard deviation of 4 replicate samples.  
 
Figure 1.9 Multiple linear regression model of sediment oxygen demand for 
station HP.    
 
Figure 1.10 Multiple linear regression model of denitrification for station HP.    
 
Figure 1.11 Multiple linear regression model of denitrification efficiency for 
station HP.    
 
Figure 1.12 Percentage of the total nitrogen efflux from the sediment as NH
4
+
 or 
N
2
.  
 
Figure 1.13 Correlation between N
2
 and SOD for individual flux cores.   
  
Figure 1.14 Pooled N
2
 flux data from each transect plotted over an annual cycle.   
 
 x 
 
 
Figure 1.15 Correlation between nitrification and denitrification for individual core 
fluxes.       
 
Figure 2.1 Map of the Choptank River - a subestuary of  the Chesapeake Bay. 
Figure 2.2 Sediment O
2
 flux from March through October 2006.     
 
Figure 2.3 Sediment NH
4
+
 flux from March through October 2006.   
 
Figure 2.4 Sediment NO
2
-
 + NO
3
-
 flux from March through October 2006.   
 
Figure 2.5 Sediment N
2
-N flux from March through October 2006.   
 
Figure 2.6 Polychaete abundance from March through October 2006.  
 
Figure 2.7 Sample microelectrode O
2
 profiles.  
 
Figure 2.8 Sediment O
2
 penetration depth.  
 
Figure 2.9 Diffusive and advective flux of O
2 
 across the sediment water 
interface. 
 
Figure 2.10 Pore water Br
-
 profiles.  
 
Figure 2.11 Comparison of two methods for measuring bioirrigation.   
 
Figure 2.12 Sediment O
2
 demand and denitrification efficiency for the period 
March through October 2006.  
 
Figure 2.13 Denitrification efficiency plotted against bottom water O
2
. 
 
Figure 2.14 Denitrification efficiency plotted against the estimated in situ O
2
 
sediment depth penetration.  
 
 
 
 
 
 
 
 
 
 
 
 xi 
 
 
 
 
 
 
 
 
Chapter 1: Denitrification Efficiency in Chesapeake Bay Sediments 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
Abstract 
 
 
Spatial and temporal variations in sediment-water flux of dissolved inorganic nitrogen 
(DIN), dissolved gases N
2
 and O
2
 were measured for the Choptank River, a 
subestuary of the Chesapeake Bay USA.   Membrane Inlet Mass Spectrometry was 
used as a high precision direct measure of N
2
 production in sediment core 
incubations.  Peak rates of denitrification were observed in fall and spring (>100 
?mol N
2
-N m
-2
 h
-1
) followed by a sharp decline to 10 ?mol N m
-2
 h
-1
 in summer.   
Coupled nitrification/denitrification was identified as the main pathway for the 
production of N
2
 gas in the sediment.     Although denitrification rates were 
stimulated by labile organic carbon additions from the water column, the overall 
efficiency of the process sharply declined as temperature increased and bottom water 
O
2
 declined.  Consequently, there was a transition from almost complete 
remineralization of nitrogen as N
2
 gas in winter to nearly 100% recycling of NH
4
+
 
back to the water column in summer.   Denitrification efficiency was shown to be 
sensitive to bottom water oxygen concentrations even in the absence of hypoxia or 
anoxia. 
 
 
 
 
 
 
 
 
 2 
 
 
 
Introduction 
 
 
Increased nitrogen loading to coastal ecosystems has been identified as a key factor in 
the eutrophication of coastal and estuarine ecosystems (Taylor et al 1995).  Nitrogen 
is often cited as the limiting nutrient in coastal ecosystems (Fisher 2006) and 
considerable efforts have been dedicated to understanding the role of internal cycling 
processes in the fate of nitrogen.  In ecosystems such as the Chesapeake Bay, the 
major N sinks include burial in sediments, denitrification, and export to the ocean 
(Boynton et al. 1995).  Both fish harvest (Boynton et al. 1995) and the efflux of 
ammonium are much smaller terms in such budgets.   Despite considerable effort 
studying the role of denitrification in coastal ecosystems, estimates of denitrification 
in many ecosystems suffer from 1) relatively poor spatial or temporal sampling 
regimes and 2) inadequate techniques for measurement (Cornwell et al. 1999).  While 
new techniques and measurement approaches (i.e. Nielsen 1992; Kana et al. 1998) 
have provided better site-specific measurements, high quality measurements that 
encompass appropriate temporal and spatial scales are generally lacking.  
 The NO
3
-
 for denitrification can be supplied from the water column (D
W
) via 
diffusion into the sediment or produced in the sediment via nitrification (D
N
).  Studies 
have suggested that a close coupling of these two processes must occur even though 
nitrification only occurs to the depth of O
2
 penetration and denitrification is 
predominately a suboxic process (Cornwell et al. 1999, Codispoti et al. 2005).  The 
spatial separation of these two processes suggests that (D
N
) should be inefficient with 
some loss of nitrate to the overlying water.  However, the presence of anoxic 
 3 
 
 
microsites of denitrification may provide a close coupling of nitrification and 
denitrification allowing for a highly efficient removal of remineralized nitrogen 
(Jenkins and Kemp 1984, Kana et al. 1998). 
The importance of bacterially mediated sediment nitrification and 
denitrification as an ecological process in coastal and estuarine sediments has been 
studied extensively with varying results.  In some estuaries denitrification is 
responsible for removal of over 40% of DIN inputs yet in other estuaries 
denitrification may remove only minor amounts of nitrogen (Seitzinger et al. 1988).   
This variability stems from the observation that nitrogen recycling from sediments to 
the estuarine water column is highly dependent on nutrient input rates, sediment 
deposition rates, and sediment redox conditions; low oxygen leads to a decrease in the 
importance of D
N
 (Seitzinger 1988, Rysgaard et al 1994). Under low O
2  
concentrations, high NH
4
+
 efflux is often observed (Kemp et al. 1990).  This seasonal 
increase in NH
4
+
 flux is important in providing sufficient DIN to sustain summer 
planktonic production in many coastal systems (e.g. Fisher et al. 1978; Cowan and 
Boynton 1996).  Controls on nitrification/denitrification also include rates of 
deposition of labile carbon on the sediment surface (Seitzinger 1985, Caffrey et 
al.1993), activity of macrofauna (Svensson et al. 1996), temperature (Henriksen and 
Kemp 1988) and water column O
2
 concentrations (McCarthy et al. 1990).  In a single 
estuarine system, all of these controlling factors can have considerable spatial and 
temporal variability (Boynton and Kemp 1985).   
Denitrification efficiency has been applied on a system wide scale as a 
measure of ecosystem function and has been shown to be dependent on the water 
 4 
 
 
residence time of a system (Nixon et al. 1996). The efficiency of denitrification 
defines the balance of recycling versus loss of nitrogen as N
2
 and is a critical measure 
of ecosystem function for shallow water systems. 
 A given atom of nitrogen may cycle from the water column to the sediment 
and back again numerous times and each pass through the sediment represents an 
opportunity for that atom of N to be denitrified.  The longer a given atom of fixed 
nitrogen resides in the system the greater chance that it will be lost to the atmosphere 
as N
2
.  In this paper, we will define the efficiency of denitrification as the probability 
of an atom of nitrogen being denitrified during one pass through the sediment.   
Studies of sediment nitrification/denitrification using indirect measurements 
(acetylene block, N-serve, denitrification potentials) in the mesohaline Chesapeake 
Bay have shown that 1) hypoxic/anoxic conditions inhibit (D
N
) and promote NH
4
+
 
recycling (Kemp et al. 1990), 2) denitrification potentials are higher in the tributaries 
than the mainstem and 3) denitrification rates exhibit high temporal and spatial 
variability (Twilley and Kemp 1986).  This study presents a detailed examination of 
denitrification rates using a direct measure of N
2
 flux from the sediment of an 
estuarine river and provides evidence for seasonal patterning in both the 
denitrification rate and the efficiency of coupled nitrification/denitrification relative 
to NH
4
+
 regeneration. 
 
 
 
 
 5 
 
 
Methods 
 
Sampling Location 
 
The Choptank River is a large subestuary of the Chesapeake Bay and its 
primarily agricultural watershed covers 251.7 km
2
.  Although nitrate inputs from the 
watershed (upriver direction) are large, tidally mixed nitrate and ammonium inputs 
from the mainstem Chesapeake Bay may be substantial in the lower river (Fisher et 
al. 1992).  Our area of study included three paired stations in different basins of the 
mesohaline portion of the Choptank River (Figure 1.1). The stations in each basin 
were chosen to be similar in water depths near the axis of the channel.   This portion 
of the Choptank River does not experience bottom water anoxia or hypoxia (Fisher et 
al. 2006) and oxidized surface sediments were observed at all sites and at all times 
throughout this study.  The HP paired sites were in 12 m of water and consisted of 
very fine-grained sediment.  The G13 paired sites were located in 5 m of water in an 
area of the estuary with no well defined channel with mixed fine grained and sandy 
sediment.  The R10 sites were located in the channel in 13 m of water near the mouth 
of the estuary.  R10 sediments were very fine grained and black at depth. Sampling 
sites were located using GPS and are listed in Table1.  We observed macrofauna at all 
sites; the benthic community was predominately composed of small polychaetes with 
the highest densities at the up river sites (G13, HP).  
 
 
 
 6 
 
 
 
 
Table 1.  Station locations 
 
Site   Latitude                   Longitude
HPA 38? 37.197N 76?08.061W 
HPB 38?36.253N 76?07.553W 
G13A 38?39.149N 76? 13.354W 
G13B 38? 40.539N 76? 13.995W 
R10A 38? 38.842N 76? 18.406W 
R10B 38?39.212N 76?18.745W 
 
 
 
 
 
 
 
 
 
 
Sample Collection 
 
Sampling commenced in July 2000 and proceeded monthly through the fall.  
In 2001, sampling was less consistent but included the winter and spring seasons.  
Our sampling schedule covered a range of bottom water temperatures ranging from 
5
o
C to 26
o
C.  Cores were collected using an acrylic/PVC Soutar-design box corer that 
collects cores with minimal exposure of sediment to metal.  This device is particularly 
good at collecting samples from areas with soft sediment.  Only cores with an 
undisturbed sediment water interface were used for flux incubations.  The box corer 
was sub-cored with 6.35 cm inner diameter acrylic liners to a depth of ~15 cm.  The 
subcores were topped off with ~15 cm of bottom water, stoppered, and transported to 
the lab within 2 h of collection.  Bottom water from each site was collected with a 
diaphragm pump.  Sediment cores were held submersed in a bath of bottom water 
overnight with continual aeration and circulation of the overlying water with bath 
 7 
 
 
water.  Holding cores overnight promoted 1) the equilibration of temperature, oxygen, 
nitrogen, and argon in overlying and near-surface pore waters and 2) provided a better 
equilibration of gases in the acrylic used for the chambers.   
Sediment-Water Exchange 
 
 Sediment cores were incubated under dark conditions in a temperature-
controlled chamber at in situ (? 2? C) temperatures.  Cores were capped with buna O-
ring fitted stirring tops prior to incubation, taking care to exclude bubbles.  During 
sample withdrawal, replacement bottom water was supplied though a port in the 
stirring top, with gravity feed providing the head pressure required to fill vials and 
syringes.  A suspended magnetic stir bar was used to mix the overlying water in each 
core.    Solute samples were filtered using a 25 mm diameter, 0.45 ?m cellulose 
acetate syringe filter (Nalgene #191-2045).  Typically, 20 ml was filtered into vials 
for analyses of NH
4
+
, and NO
3
-
. Dissolved gas samples were collected in ~7 ml 
ground glass stoppered test tubes that were filled with a dip tube; samples were 
preserved with 10 ?l of 50% saturated HgCl
2
 solution.  Dissolved gas samples were 
held under water at ambient bottom water temperature until analyses the following 
day.  Samples were collected from the top 1 cm of sediment from each flux core at 
the end of the incubations for chlorophyll a analyses and overlying water volumes 
were determined from water column height above the sediment.     
 Time courses of solute and gas concentration were determined over the course 
of 4-8 hours, with 4 samples collected for each analyte.  Oxygen concentrations were 
monitored during the incubations to insure that oxygen did not fall below 50 % of 
saturation.   Linear regression of the data was used to determine the slope of the 
 8 
 
 
time-concentration relationship and the flux rates were calculated on an areal basis.  
In a limited number of slope calculations a single data point was excluded from the 
regression.  Blank cores had only minor changes in concentration at most sites.  
Corrections were made to account for water column effects when linear changes in 
the blank were observed for an analyte.   
Gas, Solute and Sediment Analysis 
 
 Membrane inlet mass spectrometry (MIMS; Kana et al. 1994; 1998) was used 
for the analysis of dissolved N
2
 and O
2 
and Ar.  Air saturated water was used for 
calibration of the N
2
:Ar and O
2
:Ar gas ratios which were used to estimate gas 
concentration.     The overnight equilibration of sediments at in situ bottom water 
temperatures ensured that the overlying water and the pore water were close to 
thermal equilibrium, minimizing potential artifacts associated with the sensitivity of 
N
2
:Ar gas ratios to temperature . 
 Nitrate + nitrite was analyzed via segmented flow analysis after Cd reduction 
and ammonium was manually analyzed with a phenylhypochlorite colorimetric 
technique (Parsons et al. 1994).  Chlorophyll a was determined via HPLC (Van 
Heukelem et al. 1994) after extraction with acetone.  
Statistical Analysis 
 
We used multiple linear regression analysis to determine factors that might be 
important in controlling sediment oxygen demand (SOD), denitrification, and 
denitrification efficiency at the HP site.   This site was chosen for statistical analysis 
 9 
 
 
because of its close proximity to a monthly water quality monitoring site which 
provided a good proxy for algal biomass (water column chlorophyll a) and bottom 
water oxygen conditions in this part of the river.  The HP site was sampled 3 
additional times in August, October, and December of 2001 for a total of 11 flux and 
denitrification measurements.   Since we sampled monthly, we used the water column 
chlorophyll a from the previous month for the regression analysis to compensate for 
settling of the bloom and incorporation of the algae into the sediment.   All four 
replicate cores were entered into the model separately.   The best fit model was 
chosen by examining combinations of different variables that gave the best adjusted 
R-squared, Mallows? C(p), Akaike?s Information Criteria, and Bayesian Information 
Criteria  for the smallest number of variables.  Stepwise linear regression was used to 
remove variables that were not significant and exceeded an alpha value of 0.05.   The 
residuals were examined on each analysis and found to be normally distributed 
without transformation.  Bottom water properties (salinity, dissolved oxygen, nitrate, 
temperature, chlorophyll a) were entered as independent variables.  We found that 
temperature and oxygen varied collinearly and could not be entered into the model 
simultaneously.  These variables were found not to be significant for denitrification 
but were significant components for SOD and denitrification efficiency.  We ran the 
model including temperature and bottom water oxygen separately to determine which 
model best predicted denitrification efficiency and found that oxygen explained most 
of the variability.  Temperature has been shown to have an effect on sediment 
metabolism and was included in the SOD regression instead of bottom water O
2
. 
 
 
 10 
 
 
 
Results 
 
 
 Linear changes in N
2
 greater than the blank were observed in most of our 
incubations but in 38% of incubations no change in N
2
 was detected.  Sample time 
course data showing replicate cores and blanks for N
2 
and O
2
 are shown in figure 1.2.  
The O
2
 correction for our MIMS data was very small and difficult to measure with 
changes in the N
2
/Ar ratio close to the analytical precision of the method (Kana and 
Weiss 2004).  The flux data for SOD, N
2
, NH
4
+
, and NO
3
-
+NO
2
-
 at our paired stations 
within a basin were similar with less than 5% of fluxes having a statistical difference 
(analysis of variance ANOVA). Our data indicates that the sediments were fairly 
homogeneous within each basin at least over the scale of several hundred meters and 
at similar water depth. (Figure1.3-1.6).  The HP basin stations had the highest 
sediment oxygen demand (SOD) ranging from 900 to 4800 ?mol m
-2
 h
-1
; G13 and 
R10 had ranges of 400 to 3800 ?mol m
-2
 h
-1
 and 400 to 3000 ?mol m
-2
 h
-1
 
respectively (Figure 1.3).  SOD rates peaked in early May at all sites well before the 
temperature maximum in July.  A second peak in SOD occurred in late summer early 
fall at the HP and G13 sites.  In November SOD rates were low in all three basins.      
The highest N
2
 flux rates coincide with the cooler months of the year and rates 
drop off sharply in the summer (Figure 1.4). The highest rates observed were in the 
HP basin at ca. 160 ?mol ?
2
?? m
-2
 h
-1 
with the other basins exhibiting highest rates 
near 100  ?mol ?
2
?? m
-2
 h
-1
.     G13 was the only basin to have denitrification rates 
above 50 ?mol ?
2
?? m
-2
 h
-1
 during July and August (Figure 1.4).  No measurable N
2
 
 11 
 
 
flux was detected in July or October at sites R10 or G13 and rates were < 25 ?mol 
?
2
?? m
-2
 h
-1
 during July and August at all sites. 
 Ammonium fluxes followed the same pattern as temperature at the HP and 
R10 paired stations with a maximum in August and minimum in February and 
November (Figure 1.6).  The flux of NH
4
+
 increased in May along with SOD at the 
HP and R10 basins stations.  The R10 transect had the highest NH
4
+
 fluxes at ~ 
450 ?mol m
-2
 h
-1
 followed by HP with ~ 250  ?mol m
-2
 h
-1
.  The pattern of NH
4
+
 flux 
at the G13 basin sites showed minimum rates in February and November but there 
was no clear seasonal pattern from April through October, averaging about 100  ?mol 
m
-2
 h
-1
.   Bottom water reached a temperature maximum in June and began to decline 
by late August while dissolved oxygen showed an inverse pattern with a minimum of 
3-6 mg l
-1
 in July (Figure 1.7).   
 The mesohaline portion of the Choptank River typically has low bottom water 
NO
3
-
 concentrations year round of <8 ?mol L
-1
 (Malone et al. 2003).  High sediment 
NO
3
-
 uptake occurred in May when bottom water NO
3
-
 concentrations peaked (Figure 
1.5).   Station G13 and HP had low sediment uptake rates of NO
3
-
 (<60 ?mol m
-2
 h
-1
) 
between February and May.  R10 had the highest NO
3
-
 uptake and the highest bottom 
water concentration in May when NO
3
-
 levels were unusually high (16 ?M).  Stations 
HP and R10 had NO
3
-
 fluxes of about 40 ?mol m
-2
 h
-1 
directed out of the sediment in 
October/November.   
 Sediment chlorophyll a has been used as an indicator of labile carbon loading 
to the sediment (Cowan et al. 1996).  This indicator works best in sediments where 
the major source of labile organic carbon is from phytoplankton and benthic primary 
 12 
 
 
production is light inhibited.  No seasonal changes in sediment chlorophyll a were 
found during this study despite high temporal variability in spring and fall water 
column chl a concentrations (Malone et al. 2003).  The sediment chlorophyll a 
averaged 15.6?4.7 mg m
-2
, 13.9?6.4 mg m
-2
 and 35.6?18 mg m
-2
 for stations HP, 
G13, and R10 respectively (Figure 1.8).  Station R10 had sediment chlorophyll a 
concentrations that were two to three fold higher than the other sites, suggesting that 
organic carbon loading may be higher at this location (Figure 1.8). 
 
Multiple Linear Regression Analysis 
 
 The initial independent variables used in all our multiple linear regressions 
included bottom water parameters; salinity, temperature, dissolved oxygen, nitrate, 
and chlorophyll a.  Oxygen and temperature covaried and could not be used in the 
analysis simultaneously. The best model fit for SOD included 4 variables, water 
column nitrate, temperature, sediment chlorophyll a, and bottom water chlorophyll a.  
Stepwise regression showed that bottom water chlorophyll a and temperature were 
significant at the 0.05 level.  Bottom water chlorophyll a had a positive linear 
relationship with SOD and explains 75% of the variation in SOD (Figure 1.9) over 
this 1 year study (O
2
 flux ?mol m
-2
 h
-1
 = 109.55*(chl a ?g l
-1
) +28.94*Temperature 
+404.84).    
The best fit model for denitrification rate included the same four variables as 
SOD.   Interestingly, bottom water O
2
 and temperature were not found to be 
significant in the model. After stepwise regression only salinity, chlorophyll a, and 
 13 
 
 
nitrate remained in the model as significant.  Our model results represent the 
denitrification data very well explaining 78% over the variability in the rate (Figure 
1.10).   The linear relationship can be expressed as: denitrification = 3.28 * (chl a ?g 
l
-1
) + 17.59*salinity + 1.53* nitrate -188.8 
We calculated denitrification efficiency (DE) as the fraction of sediment 
nitrogen flux lost to denitrification versus being returned to the water column as a 
nutrient source for further use by algae.    DE was calculated by dividing the N
2
 flux 
by the sum of the efflux of NH
4
+
, ?
2
??
 
 and the NO
3
-
+NO
2
-
flux : 
DE=?
2
?? flux/ (?
2
?? flux + NH
4
+
flux + NO
3
-
+NO
2
-
flux) 
   Denitrification efficiency was found to be linearly related to the single variable 
bottom water oxygen (Figure 1.11). Temperature covaried with oxygen and was not 
included in the regression.   The equation for this relationship is DE = -
0.23164+0.0857 * O
2 
mg L
-1
; explaining 73% of the variability in DE. 
 
 
Discussion 
 
Temporal Patterns and the influence of oxygen 
 
 
 The maximum rates of denitrification measured were considerably higher in 
this study compared to previous measurements in Chesapeake Bay.  Kemp et al.?s 
(1990) measured maximum rates in spring  in the mesohaline portion of the bay, 
measured using the acetylene block method, were only ~ 20% of the average spring 
 14 
 
 
rates reported here.   An underestimate would be expected considering the inhibitory 
effects of acetylene on nitrification and lower bottom water O
2
 concentrations in this 
part of the bay.  Ambient spring time rates (77-89 N
2
-N ?mol m
-2
 h
-1
;Jenkins and 
Kemp 1984) estimated for the mesohaline Patuxent River using 
15
N isotopic tracer 
techniques were very close to our peak rates in spring.  Our study, together with the 
two previous studies, shows a dramatic suppression of denitrification rates during 
summer.    The near absence of summer denitrification in previous studies was 
attributed to bottom water anoxia/hypoxia inhibiting sediment nitrification.   In this 
study a similar pattern of denitrification inhibition is shown in summer for the 
Choptank River despite the bottom water being oxic (Figure 1.7).  In our study 
denitrification rates are much more dynamic than would be predicted by changes in 
bottom water O
2
 concentrations alone and O
2
 was not a significant variable 
explaining denitrification rates in our regression analysis.  Denitrification rates 
decreased to very low levels in summer despite having well oxygenated bottom water 
> 4 mg L
-1
.  The decline in summer rates in our study indicates that other factors are 
responsible for driving denitrification under oxic conditions.   Our model results for 
denitrification efficiency suggest that the process of coupled 
nitrification/denitrification at our sites is sensitive to small changes in bottom water 
O
2
.  The deposition of the spring bloom along with rising temperature drives an 
increase in SOD, decreases the depth of O
2
 penetration, and decreases the zone where 
nitrification can occur.  This decrease in nitrification shifts the sediment nitrogen 
efflux from N
2
 to NH
4
+
 and decreases denitrification efficiency.  The eutrophication 
 15 
 
 
feedback associated with the inhibition of denitrification is likely to begin before 
hypoxic/anoxic conditions develop in the water column. 
 
Denitrification Efficiency 
 
 One of the advantages of the MIMS technique is the ability to simultaneously 
measure N
2
 fluxes combined with NO
3
-
+ NO
2
-
+ NH
4
+
 fluxes and account for all the 
inorganic forms of nitrogen on the same flux core.  Ignoring fluxes of organic 
nitrogen, we can construct a budget of the nitrogen fluxes and calculate the overall 
efficiency of the denitrification process.  To determine denitrification efficiency we 
summed the DIN effluxes and calculated the percentage that was lost as N
2
 versus 
NH
4
+
 (Figure 1.12).   Despite the fact that these sites were always exposed to aerobic 
bottom water there is a distinct seasonal pattern in denitrification efficiency.  In the 
winter and fall, denitrification efficiency was high (50% to 75%) , declining sharply 
through the spring and having negligible rates  during the summer.  The proportion of 
NH
4
+
 flux (Figure 1.12) showed the inverse pattern accounting for nearly all the 
inorganic fixed nitrogen released from the sediment.  The process of denitrification at 
these sites is heavily dependent on sediment nitrification since water column nitrate 
concentrations were low most of the year.  An exception to this pattern was found at 
station R10 where there was sufficient water column NO
3
-   
delivered
 
primarily
 
from 
the Susquehanna River in April and May to supply the majority of the denitrification 
based on fluxes of NO
3
-
 +NO
2
-
 into the sediment.  During the winter and fall when 
temperature drives sediment metabolism to a minimum, and dissolved O
2 
to an annual 
maximum, O
2
 penetration into the sediment would be maximized.  Deeper O
2
 
 16 
 
 
penetration translates into more sediment volume for nitrification.  As temperature 
rises in spring and organic matter supply increases, nitrification is reduced due to 
lower O
2
 availability.   Figure 1.13 shows a plot of the individual core fluxes of O
2
 vs 
N
2
 with a line representing the Redfield ratio of O
2
/N (6.6:1) for remineralized 
phytoplankton biomass.  Assuming that the main source of labile carbon to the 
sediment was derived from algae, we would expect decomposition to yield 
approximately 1 mole of N for 6.6 moles of O
2
 consumed or CO
2
 produced.    Points 
that fall on the line represent cores where all the remineralized nitrogen was 
denitrified while points to the right of the line represent fluxes where nitrogen was 
either sequestered in the sediment or released to the water column as NH
4
+
.  The 
general pattern shows a pronounced divergence from the 100% line as SOD 
increases, indicative of a transition from denitrification to NH
4
+
 release.  There 
appears to be 3 different groupings of the data: 1) during fall and winter when 
metabolism is low nearly all remineralized nitrogen was being lost via denitrification, 
2) during the spring bloom deposition when metabolism reached maximum rates the 
denitrification yield was 16.8-38.5%, and 3) during the summer the denitrification 
efficiency approached 0%.  The sediments appear to be the most efficient in 
denitrifying during the late fall and winter when organic carbon supply from the 
water column is low and NO
3
-
+NO
2
-
 is high.  
  The addition of labile organic matter in the spring tends to stimulate 
denitrification but drive efficiency down allowing a greater proportion of the 
remineralized nitrogen to be returned to the water column as NH
4
+
.  This trend has 
been shown in other studies (Newell et al. 2002, Enoksson 1993) where increasing 
 17 
 
 
organic carbon additions result in a greater proportion of remineralized N being 
returned to the water column as NH
4
+
.  A similar trend was also found in warm-
temperate Australian lagoons where declining denitrification efficiency was observed 
across an increasing eutrophication gradient (Eyre et al. 2002).    High denitrification 
efficiencies were found for Port Phillip Bay in the range of 75-85% for sediments 
with organic matter degradation rates of 625-1,041 ?mol O
2
 m
-2
 h
-1
 but at other sites 
efficiencies decreased to 0 at degradation rates of 4167 ?mol O
2
 m
-2
 h
-1
 during 
summer (Heggie et al. 1999).  The temporal range in organic matter degradation at 
our sites (400-4800 ?mol O
2
 m
-2
 h
-1
) was similar to the Port Phillip Bay study but our 
denitrification efficiencies were much higher (16.8-38.5%) during periods of highest 
organic loading.  High loading sites in the Port Phillip Bay study had sulfide in the 
porewaters which has been shown to inhibit nitrification (Joye & Hollibaugh 1995) 
and stimulate dissimilatory NO
3
-
 reduction to NH
4
+
 (An & Gardener 2002).   Sulfate 
reduction may still be an important respiratory pathway but reduced metabolites like 
H
2
S must oxidized rapidly before they can buildup in the pore water.   The absence of 
pore water H
2
S at our sites even under the highest loading conditions could explain 
the higher denitrification efficiencies.  The presence of pore water H
2
S could be an 
important control of sedimentary N-cycling, tipping the balance toward NH
4
+
 
recycling. 
Denitrification efficiency is a measure of how well a system eliminates 
nitrogen nutrients.  High rates of sediment denitrification can be sustained by 
eutrophic systems with adequate nitrate supply but in the absence of coupled 
nitrification/denitrification a high proportion of the remineralized nitrogen is returned 
 18 
 
 
to the water column as NH
4
+
.  Our linear regression results show that bottom water 
oxygen concentration was a strong indicator of the proportion of nitrogen removed 
via denitrification; higher O
2
 levels enhanced the fraction of nitrogen removed via 
denitrification.   Low water column O
2
 in summer lead to nearly complete recycling 
of remineralized nitrogen back to the water column as NH
4
+
.  These results suggest 
denitrification efficiency can be sensitive to even small decreases in water column O
2
. 
This result is somewhat surprising when we consider that bottom water oxygen 
concentrations remain above hypoxic levels for the study period.    Water column O
2
 
levels were close to saturation during the majority of the year (except summer) yet the 
model suggests that DE is sensitive to changes in oxygen saturation induced by 
seasonal temperature increases.  The efficiency of denitrification could decline under 
a warmer climate scenario, enhancing the recycling of sediment nitrogen back to the 
water column. 
   
Temporal versus Spatial Variability 
 
The stimulation of denitrification and declining denitrification efficiency caused by 
settling water column blooms has implications for the frequency at which such 
processes must be studied.  The short half-life of sedimented fresh algal derived 
organic carbon (4 weeks - Enoksson 1993; 2-3 weeks - Garber 1984)) causes rapid 
biogeochemical changes in the sediment altering nitrogen recycling rates and 
denitrification.  These rapid recycling rates represent large sources of recycled 
nitrogen that might not be included in nutrient budgets if sampling frequencies are 
less than monthly.  The cross channel sites showed very little spatial variability 
 19 
 
 
indicating relatively uniform N cycling processes over spatial scales of 10-100 
meters, at least at a given depth.    Depth of water was different at each of our paired 
sites and flux rates did vary considerably between sites (Figure 1.3-1.6).  Due to 
funding constraints, many studies prioritize spatial over temporal coverage hoping to 
gain as much system wide information as possible.  In this study spatial variability 
within a single basin was much less important than the temporal variability at a given 
site.  Future studies may benefit from making a one-time assessment of spatial 
variability to determine the initial site locations and with a more directed focus of 
remaining resources on temporal coverage.   
 
Water Column and Sediment Chlorophyll a 
 
    The seasonal rates of denitrification at the three sites have a similar pattern 
with peaks in spring and late summer to early fall (Figure 1.4).   These peaks were 
associated with corresponding peaks in SOD and NH
4
+
 flux and were likely the result 
of water column bloom deposition.  The multiple linear regression results for SOD 
suggest that fresh organic matter in the form of algae was responsible for most of the 
seasonal change in SOD and that freshly deposited material to the sediment was 
processed rapidly with little storage.    The highest flux of organic matter to sediment 
occurred during the spring bloom but the elevated SOD that resulted only persisted 
for several weeks.   This rapid turnover of organic matter is beneficial with respect to 
nitrification because more of the organic matter is processed directly through aerobic 
rather than anaerobic pathways. There was no smell of H
2
S in these sediments even 
during the lowest bottom water oxygen concentration in summer and almost no 
 20 
 
 
measurable phosphate flux (data not shown).  This rapid turnover of organic matter 
likely promotes the retention of porewater P through Fe oxide binding near the 
sediment/water interface.   
Figure 1.14 shows the annual average bottom water chlorophyll a along with 
the average denitrification rates for our 3 transects.  The peaks in denitrification 
occurred about a month after the peaks in bottom water chlorophyll a suggesting a lag 
time for material to settle and become incorporated into the sediment.    A study on 
the degradation of sinking particulate organic carbon (POC) in the oxygen deficient 
waters of the eastern tropical North Pacific has shown that denitrifiers preferentially 
consume N-rich amino acids but would not degrade refractory POC (Van Mooy et al. 
2002).  This suggests that denitrifiers may be limited by the quality of the organic 
substrate.   The denitrifiers in our sediments appear to be stimulated by highly labile 
organic substrate derived from both the spring and fall phytoplankton blooms.  Our 
multiple linear regression results suggest that water column clorophyll a, salinity, and 
nitrate are the most important factors for denitrification.  Since the lower Choptank 
River does not have a large salinity range and nitrate concentrations were low for 
most of the year, the most important factor controlling the overall rate of 
denitrification was the supply of fresh organic matter from the water column.  
Interestingly, the model suggests that an increase in salinity would promote higher 
rates of denitrification; the low salinity range would suggest that salinity-driven 
differences in adsorption are very low.  Water column O
2
 was not found to be a 
significant factor controlling the overall rate of denitrification.        
 21 
 
 
The process of nitrification appears to be stimulated as a result of bloom 
deposition as well, probably by providing higher pore water NH
4
+
 concentrations to 
the nitrifiers.  Nitrification and denitrification were closely coupled in our sediments 
with nitrification rates explaining 66% of the variability in denitrification rates 
(Figure 1.15).  Nitrification rates were estimated by subtracting our nitrate flux rate 
from the N
2
 flux for individual cores. 
Surface sediment chlorophyll a showed no seasonal trends despite a very 
pronounced spring bloom (Figure 1.8).  Trimmer et al. (1999) also found enhanced 
denitrification and SOD resulting from spring bloom deposition in Irish Sea 
sediments with no change in sediment chlorophyll a concentrations.   In our study this 
could be attributed to our frequency of sampling since chlorophyll a has been shown 
to degrade very quickly under oxic conditions with a half-life <7 days (Sun et al. 
1991).   This would suggest that our sampling scheme might not have captured peak 
bloom deposition.  Sediment chlorophyll a has been used as a surrogate for labile 
organic supply to the sediment below the photic zone and generally correlates well 
with SOD and NH
4
+
 release.  The lack of a seasonal pattern in this study suggests that 
sediment chlorophyll a may not preserve a monthly-integrated record of 
phytoplankton deposition in oxic sediments and inter site differences may only 
represent focusing of material at a given site.    The longer half-life of sediment 
chlorophyll a in anoxic sediments of 50-60 days (Sun et al. 1993) should preserve a 
longer-term record of planktonic deposition and may account for its greater utility in 
deep mid-bay sediments (Cowan and Boynton 1996).   
 22 
 
 
The NH
4
+
 fluxes at R10 were nearly twice that of the other sites in May, 
August and September and were out of proportion with SOD (Figures 1.4, 1.6).  
Using a Redfield ratio of  (O
2
/N) 6.6 we would predict a summer NH
4
+
 efflux rate of  
~230 ?mol m
-2
 h
-1 
which is about half of the rate we measured.  R10 appears to have 
the highest carbon loading and may have a higher proportion of anaerobic metabolism 
than the other sites.    Stations HP and G13 both had maximum NH
4
+
 fluxes in 
summer when almost 100% of the remineralized nitrogen was returned to the water 
column (Figure 1.13).  The determination of nutrient remineralization rates from 
stoichiometric relationships to O
2
 may result in an under estimate because the 
summer burial of solid phase sulfide minerals derived from anaerobic 
remineralization would not be accounted for.  Moreover, the fall reoxidation of these 
minerals would result in an overestimation of metabolism (Giblin 1984; Sampou and 
Oviatt 1991; Cornwell and Sampou 1995). 
 
Conclusions 
 
 High rates of denitrification were found in the mesohaline portion of the 
Choptank River in late spring as remnants of the spring bloom settled to the sediment 
fueling high rates of metabolism.    The supply of labile organic carbon 
(phytoplankton) to the sediment stimulated rates of denitrification, SOD, and NH
4
+
 
release.  The supply of NO
3
-
 via nitrification appears to be the main controlling factor 
limiting the rate of denitrification in these sediments.  Labile carbon substrate may 
also limit denitrification especially during summer.  The correlation between bottom 
 23 
 
 
water O
2
 and denitrification efficiency is likely the result of a depression in the 
sediment O
2 
penetration depth at higher metabolic rates and a subsequent depression 
in nitrification.  These sediments become less efficient at removing nitrogen via 
denitrification with higher carbon loading and decreasing water column O
2
.  This 
study demonstrates how dynamic the nitrogen cycle can be even in sediments with 
well-oxygenated overlying water and how sensitive denitrification is to changes in 
labile carbon loading.  The intensity of phytoplankton blooms may have an impact on 
the proportion of remineralized nitrogen that is denitrified.  Increased algal bloom 
intensity brought on by eutrophication could reduce sediment oxygen penetration 
depths thereby reducing nitrification and hence driving down sediment denitrification 
efficiency.  Our study suggests that a positive eutrophication feed back mechanism 
from the sediments may be invoked prior to the onset of anoxic or hypoxic water 
column conditions. 
 
 
 
 
 
 
 
 
 
 
 24 
 
 
 
 
R10
G13
HP
Chesapeake Bay
Choptank River
38
o
40?
38
o
50?
7
6
o
1
0
?
7
6
o
0
0
?
7
6
o
1
0
?
7
6
o
1
0
?
Figure 1.1 Map of the Choptank River - a sub estuary of Chesapeake Bay.   
 
 
 
 
 
 
 
 
 
 25 
 
 
 
 
 
N
2
 
?
mo
l
 l
-1
512
514
516
518
520
522
N
2
 
?
mol l
-1
526
528
530
532
534
536
Replicate Core I
Replicate Core II
Blank
Time h
02468
N
2
 
?
mo
l
 l
-1
512
514
516
518
520
522
R10A
G13A
HPB
O
2
 
?
mo
l
 l
-1
0
50
100
150
200
250
300
O
2
 
?
mol
 l
-1
0
50
100
150
200
250
300
Time h
8 9 10 11 12 13 14 15 16
O
2
 
?
mol l
-1
0
50
100
150
200
250
300
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 1.2 Sample data of time course changes in N
2
 and O
2
 measured with the 
MIMS system.  Cores I and II are replicates from one site in May. 
 
 
 26 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
HPA HPB
-6000
-5000
-4000
-3000
-2000
-1000
0
July   00
Aug   00
Sept  00
Oct    00
Nov   00
Feb   01
Apr    01
May   01
O
2
 
?
mo
l m
-2
 h
-1
G13A G13B
-6000
-5000
-4000
-3000
-2000
-1000
0
R10A R10B
-6000
-5000
-4000
-3000
-2000
-1000
0
O
2
 
?
mo
l m
-2
 h
-1
n.d
.
n.d.
 Figure 1.3  Sediment-water exchange data for July 2000 through May 
 
 
 
2001 for SOD.  Each bar represents the average of replicate
cores collected at each station.  A and B represent distinct 
stations sampled within a given basin.  Missing bars indicate
no measurable flux.  No flux data (n.d.) was collected in July 
at site R10. 
 
 
 
 
 
 
 
 
 
 
 
 
 27 
 
 
HPA HPB
0
50
100
150
200
N
2
-N 
?
mol m
-2
 h
-1
G13A G13B
0
50
100
150
200
R10A R10B
0
50
100
150
200
n.
d.
n.d.
July   00
Aug   00
Sept  00
Oct    00
Nov   00
Feb   01
Apr    01
May   01
N
2
-N 
?
mol m
-2
 h
-1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 1.4 Sediment-water exchange data for July 2000 through May 
 
2001 for N
2
.  Each bar represents the average of replicate 
cores collected at each station.  A and B represent distinct 
stations sampled within a given basin.  Missing bars 
indicate no measurable flux.  No flux data (n.d.) was 
collected in July at site R10. 
 
 
 
 
 
 
 28 
 
 
 
 
 
 
 
 
 
HPA HPB
-100
-50
0
50
100
NO
3
-
+ NO
2
- 
?
mo
l m
-2
 h
-1
G13A G13B
-100
-50
0
50
100
R10A R10B
-200
-150
-100
-50
0
50
100
NO
3
-
+ NO
2
- 
?
mol m
-2
 h
-1
July   00
Aug   00
Sept  00
Oct    00
Nov   00
Feb   01
Apr    01
May   01
n.
d.
n.
d.
 
Figure 1.5 Sediment-water exchange data for July 2000 through May 2001 
for NO
3
-
+NO
2
-
.  Each bar represents the average of replicate 
cores collected at each station A and B represent distinct 
stations sampled within a given basin.  Missing bars indicate no 
measurable flux.  No flux data (n.d.) was collected in July at 
site R10. 
 
 
 
 
 
 
 29 
 
 
 
 
 
 
 
 
 
 
Figure 1.6 Sediment-water exchange data for July 2000 through May 2001 for 
NH
4
+
.  Each bar represents the average of replicate cores collected 
at each station.  A and B represent distinct stations sampled within a 
given basin.  Missing bars indicate no measurable flux.  No flux 
data (n.d.) was collected in July at site R10. 
 
 
HPA HPB
0
100
200
300
400
500
NH
4
+
 
?
mol
 m
-2
 h
-1
G13A G13B
0
100
200
300
400
500
R10A R10B
0
100
200
300
400
500
NH
4
+
 
?
mol m
-2
 h
-1
July   00
Aug   00
Sept  00
Oct    00
Nov   00
Feb   01
Apr    01
May   01
n.d
.
n.
d
.
 
 
 
 
 
 
 
 
 
 30 
 
 
 
Bott
om
 W
a
t
e
r
 Oxy
g
e
n
 m
g
 l
-1
0
2
4
6
8
10
12
T
e
mp
erat
ure 
o
C
0
5
10
15
20
25
30
O
2
Temperature
Jan March MayJuly Sept Nov
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 1.7 Monthly bottom water temperature and O
2
 concentrations for site 
HP. 
 
 
 
 
 
 
 
 
 31 
 
 
Sedi
ment
 Chlorophy
l
l
 
a
 mg m
-2
0
20
40
60
80
100
R10
G13
HP
Jul        Sep      Nov        Jan        Mar        May
Date
 
 
 
 
 
 
 
Figure 1.8 Sediment chlorophyll a from the top 1.0 cm of sediment.  Error 
bars represent standard deviation of 4 replicate samples.  
 
 
 
 
 
 
 
 
 32 
 
 
Date
 O
2
 
?
mol
 m
-2
 h
-1
0
1000
2000
3000
4000
5000
6000
SOD
Model
Aug  Oct    Dec   Feb  Apr   Jun   Aug   Oct    Dec
 
 
 
 
 
Figure 1.9 Multiple linear regression model of sediment oxygen demand for 
station HP.   Average data (solid line) and model results (dashed 
line) plotted by month from July 2000 through December 2001.  
Error bars represent standard deviation of replicate core fluxes. 
 
 
 
 
 
 
 
 
 
 
 
 
 33 
 
 
Aug  Oct    Dec   Feb  Apr   Jun   Aug   Oct    Dec
Date
  N
2
-N
 
?
mol
 m
-2
 h
-1
-40
0
40
80
120
160
200
Denitrification
Model
 
 
 
 
 
 
 
 
Figure 1.10 Multiple linear regression model of denitrification for station HP.   
Average data (solid line) and model results (dashed line) plotted 
by month from July 2000 through December 2001.  Error bars 
represent standard deviation of replicate core fluxes. 
 
 
 34 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Date
 
D
e
nitr
if
ic
ati
on Ef
f
i
c
i
enc
y
0.00
0.25
0.50
0.75
1.00
1.25
1.50
DE
Model
Aug   Oct   Dec   Feb  Apr   Jun   Aug   Oct    Dec
Figure 1.11 Multiple linear regression model of denitrification efficiency for 
station HP.   Average data (solid line) and model (dashed line) 
plotted by month from July 2000 through December 2001.  Error 
bars represent standard deviation of replicate core fluxes. 
 35 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
% Total
 N Fl
ux
-20
0
20
40
60
80
100
120
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 1.12 Percentage of the total nitrogen efflux from the sediment as NH
4
+
 
(filled circles) or N
2
 (open circles).  All cores for a given transect 
have been pooled (n=4).  Error bars represent standard deviation 
of replicate cores. 
-20
0
20
40
60
80
100
120
NH
4
+
N
2
G13
Jul       Sep       Nov        Jan       Mar       May
HP
-20
0
20
40
60
80
100
120
R10
Jul       Sep       Nov        Jan       Mar       May
Jul       Sep       Nov        Jan       Mar       May
Date
Date
 36 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 1.13  Correlation between nitrification and denitrification for individual 
core fluxes.      The 1:1 line represents cores where sediment 
nitrification supplies all the NO
3
-
 for denitrification.  Points below 
the line represent cores where some NO
3
-
 was supplied from the 
water column.  
 
Denitrification ?mol N m
-2
 h
-1
0 30 60 90 120 150 180
0
30
60
90
120
150
180
Nitrification
 
?
mol N m
-2
 h
-1
HP
G13
R10
1:1
r
2
= 0.66
P< 0.01
 37 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 1.14 Pooled N
2
 flux data from each transect plotted over an annual cycle.  
Bottom right panel shows the average bottom water chloroph
concentration over the same time period.  Error bars for N
2 
represents the standard deviation of all cores sampled on a given 
date and transect (N=4).  Chlorophyll a error bars represent the 
standard deviation of all 3 sites.  
yll a 
0
50
100
150
200
N
2
-N 
?
mo
l m
-2
 h
-1
0
50
100
150
200
Chlorophy
ll
 
a
 
?
g L
-1
0
10
20
30
40
50
HP R10
G13
Jul      Sep   Nov    Jan   Mar    May
Date
Jul      Sep   Nov    Jan   Mar    May
 38 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 1.15 Correlation between N
2
 and SOD for individual flux cores.  The 
100% line represents a Redfield ratio of O
2
:N of 6.6.  Points 
falling near this line represent cores where all remineralized N 
was denitrified.  The 20% line represents core fluxes where most 
of the remineralized N was returned to the water column as NH
4
+
. 
 
O
2
 ?mol m
-2
 h
-1
-6000-5000-4000-3000-2000-10000
0
50
100
150
200
N
2
-N 
?
mo
l m
-2
 h
-1
HP
G13
R10
100%
20%
 39 
 
 
 
 
 
 
 
 
 
 
 
 
Chapter 2: Seasonal Changes in Sedimentary Denitrification Efficiency: 
he effects of Bioirrigation and Reduced Bottom Water Oxygen T
 40 
 
 
 
Abstract 
 
 
Estuaries are constantly in a state of change, often leading to shifts in water 
column oxygen on short time scales that can alter the fate of remineralized nitrogen. 
Intensive temporal scale studies of sediment denitrification are often not conducted 
due to the expense and effort required. Consequently, sampling strategies may not 
capture the effects that these perturbations may have on nitrogen cycling.  We 
conducted a seasonally intensive study of nitrogen fluxes and denitrification at one 
site in the Choptank River estuary.  Bioirrigation was measured immediately 
following core incubations using a bromide tracer.  The diffusive flux of O
2
 was 
calculated from microelectrode O
2
 profiles and compared with core O
2
 fluxes as a 
second measure of irrigation. Bioirrigation rates were ranged from ~2 to 15 times 
molecular diffusion in Spring and early Summer to negligible rates by August.     
Following the spring bloom, bioirrigation increased, indirectly accounting for a large 
proportion of the sediment oxygen demand.  Low bottom water oxygen (~3 mg L
-1
) 
in summer was associated with a loss of bioirrigation, polychaete abundance, and 
very low denitrification efficiency.  Denitrification efficiency decreased from ~ 90% 
in early Spring to < 5% in August. The depth penetration of O
2
 ranged from ~4 mm to 
0.5 mm and was found to be a good indicator of denitrification efficiency under both 
high and very low irrigation rates.    
 
 
 41 
 
 
Introduction 
 
 
 Eutrophication of coastal systems has led to increased intensity of algal 
blooms (Kemp et al. 2005), loss of submersed vegetation (Orth and Moore 1984, 
Stevenson et al. 1993), toxic algal blooms (Anderson et al. 2008), and water column 
hypoxia/anoxia (Hagy et al. 2004).  Nitrogen has been identified as a key nutrient 
limiting phytoplankton growth in many estuarine systems (Gobler et al. 2006, 
Bernhard and Peele 1997, Fisher et al. 1992).  In shallow water systems benthic and 
pelagic processes can be closely coupled (Nixon 1981) with sediments an important 
source of nutrients to the water column (Cowan & Boynton 1996).   Sediments can be 
a source or a sink of dissolved inorganic nitrogen (DIN) to the water column 
depending on the efficiency of denitrification.  Denitrification is a heterotrophic 
anaerobic process that transforms NO
3
-
 into N
2
 gas removing fixed nitrogen from the 
system.  The NO
3
-
 supply for denitrification can be supplied from the water column 
(Dw) or from the process of nitrification (Dn) in sediments.  Denitrification has been 
identified as the most important nitrogen removal process in many estuaries (Nixon et 
al. 1996).  Several factors can alter denitrification efficiency: 1) low bottom water O
2
 
concentrations can inhibit nitrification (Rysgaard et al. 1994);  2)  bioirrigation 
transport of O
2
 into the sediment can enhance nitrification (Mayer et al. 1995; Nielsen 
et al. 2004);  3)  dissimilatory nitrate reduction to ammonia (DNRA) can short circuit 
Dw and convert NO
3
-
 into NH
4
+
 (An & Gardener 2002); and  4)  organic carbon 
loading to the sediment can decrease the O
2
 penetration and limit nitrification 
(Caffrey et al. 1993).   Denitrification efficiency has been applied on a system wide 
 42 
 
 
scale as a measure of ecosystem function and has been shown to be dependant on the 
water residence time of a system (Nixon et al. 1996).  Nitrogen may cycle from the 
water column to the sediment and back again numerous times and with each pass 
through the sediment there is a chance for N to be denitrified.  The longer the N 
residence time in the system, the greater the probability that N will be lost to the 
atmosphere as N
2
 gas.  In this study, we focus on the efficiency of denitrification as 
an instantaneous rate in the sediment.  Denitrification efficiency defined in this 
manner defines the balance between recycling versus denitrification and may be a 
critical measure of ecosystem function for many shallow water systems. 
 Many species of macrofauna irrigate the sediment by flushing O
2
 rich water 
through their burrows.  The overall effect of bioirrigation is to increase the volume of 
sediment where oxic and suboxic process can occur and to diminish the importance of 
sulfate reduction and the buildup of pore water sulfide (Heilskov & Holmer 2001).  
Bioirrigation has been shown to have a positive impact on the removal of fixed N 
from the sediments by promoting higher rates of nitrification (Mayer et al. 1995) and 
denitrification (Gilbert et al. 2003, Tuominen et al 1999). The degree to which 
macrofauna alter the nitrogen cycle has been linked to both density and species 
composition (Karlson et al. 2007).   The presence of macrofauna can fundamentally 
change the diagenetic processing in sediments not only through irrigation and 
reworking of sediment but also by increasing the turnover rate of organic matter 
(Levin 1986, Grassle & Grassle 1974).  Bottom water hypoxia or anoxia can result in 
decreased macrofaunal abundance and a loss denitrifying capacity.  This loss of 
 43 
 
 
denitrification may result in increased internal N loading to the system (Karlson et al. 
2007). 
 In this paper, we examine seasonal nutrient fluxes, denitrification rates and the 
efficiency of denitrification in the Choptank River estuary.   The main objectives of 
this paper are to:  1) present evidence for seasonal shifts in bioirrigation rates; 2) 
provide insight into the effects of bioirrigation on the nitrogen cycle; and 3) 
investigate factors controlling denitrification efficiency, (i.e., sediment O
2
 penetration 
depth).   
Methods 
 
Sampling Location 
 
The Choptank River watershed covers 252 km
2
 and is heavily dominated by 
agriculture.  Primary production is nitrogen limited in the river except in early spring 
when high nitrate concentrations from upstream sources persist over much of the river 
(Fisher et al. 2006).   Our study site was located in the mesohaline portion of the 
Choptank River (Figure 2.1).  This study site was chosen for several reasons: 1) 
previous studies had shown this site to be representative of the sediment type and 
bottom water oxygen conditions of a large portion of the lower Choptank River; 2) 
we had records of nutrient exchange and denitrification data sampled at varying 
intervals since July 2000 (Owens Chapter 1); and 3) this site was in close proximity 
to a monthly water column monitoring site maintained by the Maryland Department 
of the Environment.  This portion of the Choptank River does not experience bottom 
water anoxia or hypoxia but historical data shows that bottom water oxygen 
 44 
 
 
concentrations are declining as a result of continued eutrophication (Fisher et al. 
2006).  Oxidized surface sediments were observed in all cores throughout this study.  
The site (HPA) chosen for this study was located near the channel in the mesohaline 
portion of the river (38? 37.197N 76?08.061W) in ~9 m of water and consisted of 
very fine-grained sediments.  Brown oxidized surface sediments were observed at all 
sampling times and there was no evidence of free porewater H
2
S at any depths in our 
sediment cores.  We observed only small infauna at this site predominately in the 
form of small polychaetes.  
 
Sample Collection 
 
Sampling was initiated in March 2006 with monthly sampling through June 
and two additional samplings in August and October 2006.    Our sampling schedule 
covered a range of bottom water temperatures between 5
o
C and 29
o
C.  Cores were 
collected using an acrylic/PVC Soutar-design box corer that collects cores with 
minimal exposure of sediment to metal.  This device worked well collecting intact 
sediments with minimal resuspension from the very fine grained bottom deposits 
found at our sampling site.  Only cores with an undisturbed sediment water interface 
were used for our measurements.  The box corer was sub-cored with 6.35 cm inner 
diameter acrylic liners to a depth of ~15 cm.  The subcores were topped off with ~15 
cm of bottom water, stoppered, and transported to the lab within 2 h of collection.  
Bottom water from each site was collected with a diaphragm pump.  Sediment cores 
were held submersed in a bath of bottom water overnight with continual aeration and 
circulation of the overlying water with bath water.  Holding cores overnight 
 45 
 
 
promoted: 1) the equilibration of temperature, oxygen, nitrogen, and argon in 
overlying and near-surface pore waters; and 2) provided a better equilibration of 
gases in the acrylic used for the chambers.   
 
Sediment-Water Exchange 
 
 Sediment cores were incubated under dark conditions in a temperature-
controlled chamber at in situ (? 2? C) temperatures.  Cores were capped with buna O-
ring-fitted stirring tops prior to incubation, taking care to exclude bubbles.  During 
sample withdrawal replacement bottom water was supplied though a port in the 
stirring top, with gravity feed providing the head pressure required to fill vials and 
syringes.  A carboy of bottom water elevated above the cores supplied bottom water 
through manifolds which distributed water to each core.   A suspended magnetic stir 
bar was used to mix the overlying water in each core.  Solute samples were filtered 
using a 25 mm diameter, 0.45 mm cellulose acetate syringe filter.  Typically, 20 mL 
were filtered into vials for analyses of NH
4
+
, and NO
3
-
. Dissolved gas samples were 
collected in ~7 ml ground glass stoppered test tubes that were filled with a dip tube; 
samples were preserved with 10 ?l of 50% saturated HgCl
2
 solution.  Dissolved gas 
samples were held under water at ambient bottom water temperature until analysis the 
following day.   
 Time courses of solute and gas concentration were determined over the course 
of 4-8 hours with 4 samples collected for each analyte.  Oxygen concentrations were 
monitored during the incubations to insure that oxygen did not fall below 50 % of 
 46 
 
 
saturation.   Linear regression of the data was used to determine the slope of the 
time-concentration relationship and the flux rates were determined on an areal basis.  
In a limited number of cases, single data points were excluded from the regression.  
Blank cores had only minor changes in concentration at most sites.  Corrections were 
made to account for water column effects when linear changes in the blank were 
observed for an analyte.    
 
 
 
Microelectode oxygen measurements 
 
Dissolved pore water O
2
 concentrations were determined using a Clark type  
O
2
 microsensor (www.unisense.com ) with 50?m tip (Revsbech 1989).   Triplicate 
microelectrode profiles of O
2
 were made on each of the three replicate cores.  
Overlying water in the core was homogenized with a low flow of air through an air 
stone. Measurements were made at 100 ?m intervals starting ~0.4 mm above the 
sediment water interface and continued down to the depth where O
2
 was depleted.  
All microelectrode measurements were made at ambient bottom water temperatures.  
 
Bioirrigation measurements 
 
Estimates of infaunal irrigation rates were made by comparing discrepancies in 
modeled and actual measured profiles of an added conservative bromide tracer 
(Mackin et al. 1988; Martin & Banta 1992).  Immediately following the flux core 
incubations water lift pumps were placed in each core.  The cores were kept 
 47 
 
 
submerged with lift pumps providing both aeration of the water and complete mixing 
of the water overlying the cores with the bottom water in the tank.  A concentrated 
solution of NaBr was added to the tank water to reach a final concentration of ~5 
mM.  The Br
-
 tracer was allowed to diffuse into the sediment overnight (17-24 hours) 
before cores were removed and sectioned for pore water.  Surface sections down to 2 
cm were sectioned at 0.5 cm intervals and deeper strata were sectioned at coarser 
intervals down to a total depth of ~ 8 cm.  Pore water was extracted by centrifugation 
and analyzed with a Dionex ICS 2000 ion chromatograph. Theoretical bromide tracer 
profiles were modeled using Fick?s first law of diffusion and compared with the 
actual concentrations of bromide found in the pore water.  Modeled pore water 
concentrations of Br
-
 were calculated from Fick?s first law of diffusion J
s
= - 
?D
s
(dc/dx) where J
s
 is the flux mol cm
-2
 s
-1
, ? is porosity, D
s
 is the molecular 
diffusion coefficient cm
-2
 s
-1
, and dc/dx the concentration gradient in mol cm
-3
 cm
-1
. 
Diffusion coefficients within sediments were estimated from diffusion coefficients in 
a particle free solution at in situ temperature and the conversions of porosity to 
tortuosity squared using 
Boudreau?s law (?2 = 1 ? ln (?2), (Boudreau 1997) 
 The pore water O
2
 measured with microelectrodes was used in conjunction 
with our measured O
2
 flux (core incubations) as a second measure of bioirrigation. 
We used our microelectrode profiles of pore water dissolved oxygen to calculate the 
diffusive flux of O
2
 across the sediment water interface.  This was calculated using 
the equation: Flux = ??Ds(dC/dx)x = 0  where ? is porosity, C  pore water 
concentration, x depth and Ds molecular diffusivity corrected for tortuosity (Berg et 
 48 
 
 
al. 2001).  A porosity of 0.88 was used for all calculations of diffusive O
2
 flux.  The 
calculated flux O
2
 from the microelectrode profiles was compared to actual flux 
measured during our flux core incubations to quantify any enhancement of solute 
transport beyond that which could be explained by simple molecular diffusion.  
 
   
 
 
 
Gas, Solute and Sediment Analysis 
 
 A "dissolved gas analyzer" (DGA; Kana et al. 1994; 1998) was used for the 
analysis of N
2
 and O
2
 in flux samples.  This analyzer consisted of a quadrupole mass 
spectrometer with a silicone membrane inlet, which exchanged the gases between the 
water and the instrument?s vacuum.   Saturated water standards were used for 
calibration of the N
2
:Ar and O
2
:Ar gas ratios which were used as gas concentration 
estimates.    The overnight equilibration of sediments at in situ bottom water 
temperatures ensured that the overlying water and the pore water were close to 
thermal equilibrium, minimizing potential artifacts associated with the sensitivity of 
N
2
:Ar gas ratios to temperature. Nitrate + nitrite was analyzed via segmented flow 
analysis after Cd reduction and ammonium was manually analyzed with a 
phenylhypochlorite colorimetric technique (Parsons et al. 1994).   
 
 
 
 49 
 
 
Infaunal sampling 
 
Previous work at the HPA site had shown that replicate core fluxes were very 
similar over spatial scales of 10 -100m for a given water depth.  All the measurements 
made at a given sampling date including the infaunal sampling were made using the 
same cores.   Sediment cores were collected (day 1), conditioned overnight and flux 
measurements completed the next day. Once the flux incubations were complete, 
bromide was added and allowed to diffuse into the sediment overnight (day 2).  The 
following morning (day 3) microelectrode pore water O
2
 profiles were measured on 
all cores and a small sub core was sectioned from each flux core for pore water Br
- 
.
   
The remaining sediment in each core was then sieved through a 100?m mesh for 
infaunal sampling.   This method allowed us to measure all parameters on the same 
cores and minimized problems caused by core to core variability.  
 
Results 
 
Sediment water exchange 
 
Our lowest rates of sediment oxygen demand (SOD), 508 ?34 ?mol O
2
 m
-2
 h
-1
  
were measured in March when water temperatures were 5 ? C ( Figure 2.2).  The SOD 
rates increased through the spring with a peak rate of 4198?72 ?mol O
2
 m
-2
 h
-1
 
measured in June. SOD decreased 3 fold between June and August and remained low 
into the fall.  The flux of NH
4
+
 was directed out of the sediment and followed a 
 50 
 
 
similar pattern to SOD with peak rates in June of 490?95 ?mol N m
-2
 h
-1 
(Figure 2.3).  
The peak in NH
4
+
 flux was more pronounced than for SOD; NH
4
+
 flux increased 6 
fold from May to June and then decreased nearly 9 fold from June to August.  The 
peak bottom water temperatures of 29
o
C were not reached until August.  Measured 
NO
3
-
+NO
2
- 
fluxes were variable ranging from 60 ? 9 ?mol N m
-2
 h
-1 
directed out of 
the sediment to -34?9 ?mol N m
-2
 h
-1
 into the sediment (Figure 2.4).   Fluxes of NO
3
-
+NO
2
-
 were directed out of the sediment only in early spring and fall.  The flux of 
dissolved N
2
 across the sediment water interface was directed out of the sediment on 
all individual core measurements made during this study (Figure 2.5).  The rates of N
2 
flux were variable with the highest rates of efflux of ~170 ?mol N
2
-N m
-2
 h
-1
 
measured in April and June.  In August, the rate of dissolved N
2
 efflux had declined 
to very low levels and was only detectable in one of the triplicate flux cores.  
Bottom water O
2
 and Polychaete Abundance 
 
Bottom water O
2
 concentrations at our sampling site remained near 100 % 
saturation from March to April and declined from May through August (Figure 2.6).   
The lowest bottom water O
2
 concentration of 3.05 mg l
-1
 (43% saturation) was 
measured in August.   The dominant infaunal species found at our sampling site was a 
small spionid polychaete ~0.5 mm in width and ranged from ~1-1.5 cm in length.  
Polychaete densities were high in the spring ranging from 2.8x 10
4
 to 1.1x10
4
 
individuals m
-2
 but declined by an order of magnitude in summer and remained low 
into the fall (Figure 2.6).  No data on polychaete densities was collected in June.   
 
 51 
 
 
Sediment microelectrode O
2
 measurements 
 
Large changes in the sediment depth penetration of O
2
 were found during the 
course of this study (Figure 2.7, 2.8).  Sediment pore water O
2 
penetration reached 
nearly 4 mm into the sediment in March when both temperature and SOD were low.  
The penetration depth of O
2
 decreased in April and May and reached the lowest level 
in August.  Our microelectrode profiles of pore water O
2 
were measured with O
2
 
saturated overlying water.  Under O
2
 saturating conditions, the depth penetration 
decreased from 4 mm to about 2 mm from March to August (Figure 2.7).  The actual 
depth penetration of O
2
 would have been much lower in situ as the bottom waters 
became undersaturated with O
2
 from May through October (Figure 2.8).   We 
estimated the in situ O
2
 penetration depth by taking the difference in the overlying 
water between O
2
 saturating conditions and actual bottom water O
2
 and applied this 
difference to the measured profiles.   The in situ depth penetration of O
2
 was then 
estimated from our adjusted profiles. During the course of our study, the in situ depth 
penetration of O
2
 changed from a 4 mm max in March to less than a 0.5mm in August 
(Figure 2.8).   
The microelectrode O
2
 profile data was used to calculate the diffusive flux of 
O
2
 through the sediment water interface (Figure 2.9).  The diffusive flux of O
2
 
increased with temperature but only increased ~2 fold during this study.  The core O
2
 
flux and the diffusive flux were similar in March and August but the core flux 
increased more than 8 fold from March to April as a result of advective transport 
induced by bioirrigation.    
 52 
 
 
  Bioirrigation was also measured by modeling the transport of Br
-
 into the 
sediment.  We adjusted the theoretical (diffusion only) profiles to match our 
measured Br
-
 profiles and estimated an advection factor.  Plots of Br
-
 concentration 
with depth are shown in figure 2.10 for June and August.  Highest rates of 
bioirrigation were measured in June and were reflected in the large discrepancy 
between theoretical and measured profiles of Br
-
.    An advection factor > 1 would 
represent enhanced solute transport via bioirrigation.  The advection factors 
calculated from the Br
-
 tracer and the comparison of SOD and diffusive flux from O
2
 
sediment profiles were similar for most of the months sampled (Figure 2.11).  The 
June data showed a large disparity between the 2 approaches with Br
-
 tracer giving a 
3 fold higher advection factor.  The June sampling also had by far the highest rates of 
bioirrigation; rates were 5 to 15 times that of diffusion depending on which method 
was used.   
Denitrification Efficiency 
 
 Denitrification efficiency is a good measure of how effective the sediments 
are in removing nitrogen as N
2
 gas through the process of denitrification.  We 
calculated denitrification efficiency by dividing the denitrification rate by the sum of 
the dissolved inorganic nitrogen (NO
3
- 
+ NO
2
- 
+ NH
4
+ 
+ N
2
) efflux.  Water column 
nitrate concentrations were in the range of 2-20 ?M during our study and at times the 
water column was an important source of nitrate for denitrification.  The sediments at 
our study site were very efficient at denitrification in March and April with nearly all 
DIN flux in the form of N
2
 gas (Figure 2.12).    Efficiency decreased after April and 
 53 
 
 
reached a low of < 3% by August and then increased to 36% by October.  In June and 
August NH
4
+
 was the dominant form of DIN flux from the sediment.   
 
Discussion 
 
Seasonal changes in bioirrigation 
 
 The two methods used in this study to measure bioirrigation have been shown 
to give comparable results (Berg et al. 2001).  In our study, the March and June Br
-
 
method gave statistically (analysis of variance ANOVA) lower values for enhanced 
transport of solutes via bioirrigation compared to the microelectrode/core flux 
method.  The largest disparity was measured in June when rates of bioirrigation 
peaked for the year.  Small opportunistic polychaetes have been shown to take 
advantage of fresh organic material by increasing growth rate and population 
densities over short time scales (Levin 1986, Grassle & Grassle 1974). Spring peaks 
in small polychaete abundance have also been observed in other studies and thought 
to be a response to fresh algal detritus settling to the bottom as a result of the spring 
bloom (Cheng et al. 1993, Rossi & Lardicci 2002).     
 Our previous work at this site (2000-2001) had shown a tight benthic pelagic 
coupling with the spring bloom stimulating high rates of SOD.  If we apply our linear 
regression SOD model from our previous study, we find that the model predicts the 
general pattern of SOD except for the winter rates (Figure 2.12).  Our previous SOD 
model (O
2 
Flux =131.2*Chl a ?g l
-1
 +682.5) did not include temperature and as a 
consequence over predicted SOD during the intense 2006 winter bloom.  These 
 54 
 
 
results suggest that temperature is a component driving SOD and that there was 
probably some sediment storage of the winter bloom that was decomposed later in the 
spring as temperature increased.  These model results also indicate that the activity of 
macrofauna were responsible for the tight benthic-pelagic coupling.      
 In our study, we observed a 10 fold decrease in polychaete abundance from 
May to August.  This type of boom and bust cycle is common for opportunistic 
species of polychaetes and the summer decline in abundance may result from a lack 
of nutrition in summer (Rossi & Lardicci 2002, Linton & Taghon 2000).  Spionid 
polychaetes have been shown to have good survival during short term hypoxic events 
lasting up to 2 weeks but longer term effects of reduced oxygen are likely to have an 
adverse effect on survival (Llanso 1991).  The lack of bioirrigation in our study 
during summer is likely due to both decreased abundance of polychaetes and 
modified feeding behavior in response to lower bottom water oxygen concentrations 
(Llanso 1991).    The highest abundance of polychaetes in March did not correspond 
to high rates of irrigation, suggesting that worm activity may have been suppressed 
due to temperature (~5
o
 C).  In the absence of abundance data, we can only speculate 
as to why irrigation rates were high in June.  Increased feeding activity, increased 
abundance, or increased irrigation due to lower O
2
 concentrations could be possible 
explanations for the June spike in irrigation rates.  Visual inspection of the cores 
during sectioning for Br
-
 in June suggested that small polychaetes were still the 
dominant species. 
  
 55 
 
 
Bioirrigation and Nutrient Cycling 
 
 The cycling of fresh organic matter in the sediments has been shown to be 
greatly enhanced through the activity of macrofauna (Heisklov & Holmer 2001, 
Bianchi et al. 2000).   The exact mechanism of organic matter degradation 
enhancement by macrofauna is under debate but there is some evidence that increased 
exposure to O
2
 alone can enhance decomposition (Harvey et al. 1995).   In our study, 
the large spike in SOD rates in June could be the result of enhanced polychaete 
feeding and resulting transport of O
2
 into the sediment.   Through the activity of these 
animals the sediment was rapidly depleted of fresh organic matter and SOD rates 
decline within a month and remain low until fall.  Studies have shown that sediment 
chlorophyll a can preserve a short term record of phytoplankton deposition to the 
sediment (Cowan & Boynton 1996).  Our previous work at the HP site showed low 
sediment chlorophyll a through spring and summer after the spring bloom.  The 
efficient feeding activity of macrofauna may have enhanced the decomposition of 
chlorophyll a. 
   High rates of irrigation can short circuit the process of coupled 
nitrification/denitrification through rapid transport of NH
4
+
 from burrows into the 
water column (Mermillod-Blondin et al. 2004).  This irrigation effect is demonstrated 
in our study; under moderate rates of irrigation in spring denitrification efficiencies 
remain high but under very high rates of irrigation in June denitrification efficiency  
decreased to ~30% with most of the DIN returning the water column as NH
4
+
.  The 
high denitrification efficiencies found in Port Phillip Bay sediments over a wide range 
of bioirrigation rates suggests that irrigation has little effect on efficiency (Berelson et 
 56 
 
 
al, 1998).   In our study the decline in denitrification efficiency in spring was also 
associated with the settling of the spring bloom and this increase in fresh organic 
matter may have contributed to less efficient denitrification due to a decreased 
penetration of O
2
 and a shrinking of the zone where nitrification could occur   
(Caffrey et al. 1993).    The summer decline in both the abundance of polychaetes and 
irrigation was also associated with the lowest denitrification efficiency measured in 
our study.  These results suggest that when present in sufficient abundance, these 
small opportunistic polychaetes promote the removal of nitrogen via denitrification. 
 Irrigation and rapid organic matter turnover via macrofauna promote 
denitrification for the following reasons: 1) There is little long term labile organic 
matter storage in the sediment that could drive a bottom up enhancement of hypoxia 
later in the summer; 2) The increased transport of O
2
 into the sediment promotes 
higher rates of denitrification through both coupled nitrification/denitrification and 
enhanced transport of nitrate from the water column ; and  3)  Sediments are more 
oxidized inhibiting the buildup of H
2
S and promoting a deeper diffusional depth of 
O
2
.   Studies of fish-farm sediment with added Capitella sp and Nereis diversicolor 
have shown both an enhancement of aerobic metabolism and a decrease in sulfate 
reduction compared to azoic sediments (Heilskov & Holmer 2001).    The sediments 
below these fish pens were very reducing with H
2
S present near the sediment-water 
interface.  In sediments that are more oxidized, macrofauna could have a greater 
impact on the balance between aerobic versus anaerobic metabolism and may actually 
prevent free sulfide from building up in pore water by advecting O
2
 and promoting 
the reoxidation of reduced metabolites.   A tipping point or threshold is likely to exist 
 57 
 
 
driven by decreasing bottom water O
2
 or increasing organic matter load that could 
drive the sediments into a state where free sulfide could persist in pore water.  The 
result of crossing this threshold would be a substantial increase in the recycling of 
remineralized N through the inhibition of nitrification (Joye and Hollinbaugh 1995), 
dissimilatory NO
3
-
 reduction to NH
4
+ 
(An & Gardner 2002), and a decrease in habitat 
for marcofauna due to H
2
S poisoning.   
 
Denitrification Efficiency 
 
 Our study indicates that DE was strongly influenced by both bottom water O
2
 
concentrations (Figure 2.13) and the depth penetration of O
2
 (L) into the sediment 
(Figure 2.14).   The multiple regression analysis from our previous work had shown 
that the main factor controlling the balance between NH
4
+
 release and denitrification 
was the concentration of bottom water O
2
.  When applied to the 2006 data, our 
regression predicted ~75% of the variability in DE (Figure 2.12).    These model 
results indicate that the relationship between the depth of O
2
 penetration and DE is 
robust and might be used as a predictive tool in other coastal systems.    
 The majority of the change in L in our study can be attributed to changes in 
bottom water O
2
.   The in situ estimated L explains ~67% of the variability in DE 
even though most of time the transport of O
2  
through bioirrigation was at least 
equivalent to the diffusional flux of O
2
.   There are several possibilities that could 
explain the strong dependence of DE on L: 1) deeper O
2
 penetration leads to greater 
surface area for nitrification; 2) the rapid turnover of labile organic matter by 
macrofauna maintains a low level of labile organic carbon in the sediment promoting 
 58 
 
 
greater L year round; and 3) decreasing L may lead to less irrigation through modified 
macrofaunal behavior or population decline.  This study gives some strong evidence 
that O
2
 penetration depth could be a useful tool in predicting the efficiency of 
denitrification in other systems. 
 
Implications 
 
The results of this study have implications for nitrogen cycling in coastal 
systems.  The loss of bioirrigation through sediment disturbance or low bottom water 
O
2
 could have major impacts on the nitrogen cycle leading to higher rates of NH
4
+
 
recycling back to the water column.  Our results indicate that macrofauna can; 1) 
enhance O
2
 advection, 2) enhance the decomposition of labile organic matter, and 3) 
maintain sediment conditions that promote a high level of denitrification efficiency.   
We cannot definitively conclude that the 10 fold decline in polychaete abundance 
during summer in this study was caused by low bottom water O
2
 but the data show 
that under low O
2
 (~3 mg l
-1
) conditions and in the absence of irrigation, coupled 
nitirification/denitrification was greatly inhibited.  The progression of eutrophication 
often leads to reduced bottom water O
2
 concentrations and at some point a loss of 
ecosystem function.  Thresholds of hypoxia (~2 mg l
-1
) used for management 
purposes are currently being revised. New information suggests that macrofauna 
behavior and abundance begin to change above the 2 mg l
-1
 (hypoxia) threshold 
(Vaquer-Sunyer and Duarte 2008). 
 
 Our study demonstrates the importance of 
defining O
2
 thresholds that will maintain a high level of denitrification efficiency 
 59 
 
 
through maintaining healthy infaunal communities and a deeper aerobic zone within 
sediments.    
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  
 
 
 60 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 2.1 Map of the Choptank River a sub estuary of  the Chesapeake Bay. 
 
 
 
 
 
 
 
 
 
 61 
 
 
O
2
 
?
mo
l
 m
-2
 h
-1
-6000
-5000
-4000
-3000
-2000
-1000
0
March     April    May     June             August              October
Date
 
 
 
 
Figure 2.2 Sediment O
2
 flux from March through October 2006.  Error bars 
  represent standard deviation of 3 replicate cores. 
 
 
 
 
 
 
 
 
 
 62 
 
 
 
 
NH
4
+
 
?
mol
 m
-2
 h
-1
0
100
200
300
400
500
600
Tem
per
at
u
r
e  
o
C
0
5
10
15
20
25
30
March     April    May     June             August              October
Date
 
 
 
 
Figure 2.3 Sediment NH
4
+
 flux from March through October 2006 plotted on 
left axis.  Bottom water temperature plotted on the right axis. Error 
bars represent standard deviation of 3 replicate cores.     
 
 
 
 
 
 
 63 
 
 
Date
NO
3
-
+NO
2
-
 
?
mol
 m
-2
 h
-1
-100
-50
0
50
100
March     April    May     June             August              October
 
 
 
 
 
 
Figure 2.4 Sediment NO
2
-
 + NO
3
-
 flux from March through October 2006.  Error      
bars represent standard deviation of 3 replicate cores. 
 
 
 
 
 
 
 
 
 64 
 
 
 
 
 
 
 
Date
N
2
-N 
?
mol
 m
-2
 h
-1
0
50
100
150
200
250
300
March     April    May     June             August              October
 
 
Figure 2.5 Sediment N
2
-N flux from March through October 2006.  Error bars 
  represent standard deviation of 3 replicate cores. 
 
 
 
 
 
 
 65 
 
 
 
 
 
 
 
0
5000
10000
15000
20000
25000
30000
0
20
40
60
80
100
120
Worm abundance
Bottom water O
2
O
2
 %
 s
a
t
u
r
a
ti
on
I
n
d
i
vidu
als m
-2
n.d.
March     April    May     June             August              October
Figure 2.6 Polychaete abundance from March through October 2006.  No 
abundance data was collected in June.  In situ bottom water O
2
 
percent saturation plotted on the right hand axis.  Error bars 
represent standard deviation of 3 replicate cores. March data n=1 
 
 
 
 
 
 
 
 66 
 
 
O
2
 ?mol l
-1
0 100 200 300 400
Dept
h mm
0
1
2
3
4
5
March Core A
March Core B
March Core C
June Core A
June Core B
June Core C
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 2.7 Microelectrode O
2
 profiles from March and June.  One profile for 
each replicate core measured in March (blue) and June (red)  
 
 
 
 
 
 67 
 
 
 
 
 
 
O
2
 Pen
etr
atio
n de
pth
 m
m
0
1
2
3
4
5
100% O
2
 saturation
Ambient bottom water O
2
March     April    May     June             August              October
 
 
Figure 2.8 Solid line represents the O
2
 penetration depth in mm under 
100% saturated O
2
 conditions in the laboratory.  The dashed 
line represents the estimated depth of O
2
 under actual water 
column O
2
 concentrations. Error bars represent standard 
deviation of 3 replicate cores. 
 
 
 
 
 
 68 
 
 
 
  O
2
 
?
mol
 m
-2
 h
-1
-5000
-4000
-3000
-2000
-1000
0
Core flux
Diffusive flux
March     April    May     June             August              October
 
 
 
 
Figure 2.9 The black bars represent the flux of O
2 
from our core incubations 
(diffusive and advective).  The grey bars show the flux of O
2
 from 
molecular diffusion only.  The difference between the bars is the 
advective transport of O
2
 via bioirrigation. Error bars represent 
standard deviation of 3 replicate cores. 
 
 
 
 
 
 
 
 
 69 
 
 
 
Br
-
 ?mol l
-1
0123456
Depth cm
0
2
4
6
8
Theoretical
Measured
0123456
Depth
 cm
0
2
4
6
8
August
June
Theoretical
Measured
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 2.10 Bromide profiles from June (top panel) and August (bottom 
panel).  The theoretical Br
-
 diffusion profile (diamonds) plotted 
with the measured profile (squares). 
 
 70 
 
 
 
 
 
March April May June August October
A
d
v
e
ction Facto
r
0
2
4
6
8
14
16
18
Oxygen
Bromide
Figure 2.11 Comparison of our 2 methods for measuring bioirrigation.  The 
black bars represent the advection factor for the Br
-
 tracer.  The 
white bars show the advection factor calculated from the 
comparison of whole flux core incubations and diffusional 
transport of O
2
 measured with microelectrodes.  Only March 
and June data show statistically different advection numbers 
between methods.  An advection factor of 1 represents transport 
only by diffusion.  Error bars represent standard deviation of 3 
replicate cores. 
 
 
 
 
 
 
 71 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 2.12 
odel 
odel developed 
 
Sediment O
2
 demand and denitrification efficiency for the period 
March through October 2006.  Error bars represent standard 
deviation of 3 replicate cores.  Closed circles represent m
predictions based on a multiple linear regression m
from seasonal data collected at the same site in 2000-2001. 
O
2
 
?
mol m
-2
 h
-1
-5000
-4000
-3000
-2000
-1000
0
Den
i
tri
f
icat
ion
 E
f
f
i
cien
cy
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Measured
Model
Mar  Apr   May  Jun          Aug           Oct
 72 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 2.13 Denitrification efficiency plotted against bottom water O
2
 in mg l
-1
 Bottom water O
2
 is a strong indicator of efficiency explaining 75% 
of the variability in DE. 
.  
 
Bottom water O
2  
mg l
-1
 
024681012
0.0
0.2
0.4
0.6
0.8
1.0
1.2
r
2
=0.75
p<0.01
ci
enc
y
D
e
nitr
if
i
c
ation
 Ef
f
i
 73 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 2.14 Denitrification efficiency plotted against the estimated in situ O
2
 
sediment depth penetration (L) in mm (diamonds) and the measured 
L under saturated O
2
 conditions (closed circles).     
 
 
O
2
 Penetration Depth m
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
m
O
2
 100% saturation
Ambient bottom water O
2
Denit
r
if
icat
ion Ef
f
i
ciency
r
2
=0.43
p<0.01
r
2
=0.70
p<0.01
012345
 74 
 
 
Bibliography 
 
An, S. M., and W. S. Gardner. 2002. Dissimilatory nitrate reduction to ammonium 
(DNRA) as a nitrogen link, versus denitrification as a sink in a shallow 
estuary (Laguna Madre/Baffin Bay, Texas). Marine Ecology-Progress Series 
237: 41-50. 
nderson, D. M. and others 2008. Harmful algal blooms and eutrophication: 
Examining linkages from selected coastal regions of the United States. 
Harmful Algae 8: 39-53. 
Berg, P., S. Rysgaard, P. Funch, and M. K. Sejr. 2001. Effects of bioturbation on 
solutes and solids in marine sediments. Aquatic Microbial Ecology 26: 81-94. 
Bernhard, A. E., and E. R. Peele. 1997. Nitrogen limitation of phytoplankton in a 
shallow embayment in northern Puget Sound. Estuaries 20: 759-769. 
Bianchi, T. S., B. Johansson, and R. Elmgren. 2000. Breakdown of phytoplankton 
pigments in Baltic sediments: effects of anoxia and loss of deposit-feeding 
macrofauna. Journal of Experimental Marine Biology and Ecology 251: 161-
183. 
Boudreau, B. P. 1997. Diagenetic Models and Their Implementation. Springer. 
Boynton, W. R., J. H. Garber, R. Summers, and W. M. Kemp. 1995. Inputs, 
Transformations, and Transport of Nitrogen and Phosphorus in Chesapeake 
Bay and Selected Tributaries. Estuaries 18: 285-314. 
Boynton, W. R., and W. M. Kemp. 1985. Nutrient regeneration and oxygen 
consumption by sediments along an estuarine salinity gradient. Mar. Ecol. 
Prog. Ser. 23: 45-55. 
 
A
 75 
 
 
Boynton, W. R., W. M. Kemp, and C. W. Keefe. 1982. A comparative analys
nutrients and other factors influencing estuarine phytoplankton production, p
69-90. Estuarine Comparisons. Academic Press. 
is of 
. 
f 
crobiol. Ecol. 12: 159-167. 
. 
 
scale. 
Geochemical Transformations 
Cowan
ongitudinal axis of Chesapeake Bay:  Seasonal patterns, 
Enokss ed algal 
Caffrey, J. M., N. P. Sloth, H. F. Kaspar, and T. H. Blackburn. 1993. Effect o
organic loading on nitrification and denitrification in a marine sediment 
microcosm. FEMS Mi
Cheng, I. J., J. S. Levinton, M. Mccartney, D. Martinez, and M. J. Weissburg. 1993
A bioassay approach to seasonal-variation in the nutritional-value of sediment.
Marine Ecology-Progress Series 94: 275-285. 
Cornwell, J. C., W. M. Kemp, and T. M. Kana. 1999. Denitrification in coastal 
ecosystems:  environmental controls and aspects of spatial and temporal 
Aq. Ecol. 33: 41-54. 
Cornwell, J. C., and P. A. Sampou. 1995. Environmental controls on iron sulfide 
mineral formation in a coastal plain estuary, p. 224-242. In M. A. 
Vairavamurthy and M. A. A. Schoonen [eds.], 
of Sedimentary Sulfur. American Chemical Society. 
, J. L. W., and W. R. Boynton. 1996. Sediment-water oxygen and nutrient 
exchanges along the l
controlling factors and ecological significance. Estuaries 19: 562-580. 
on, V. 1993. Nutrient recycling by coastal sediments - effets of add
material Marine Ecology-Progress Series 92: 245-254. 
 76 
 
 
Eyre, B. D., S. Rysgaard, T. Dalsgaard, and P. B. Christensen. 2002. Comparison 
isotope pairing and N-2 : Ar methods for measuring sediment-denitrification-
assumptions, modifications, and implications. Estuaries 25: 1077-1087.
of 
 
f Chesapeake Bay. 
Fisher,
 
Fisher,  
y. Mar. Ecol. Prog. Ser. 82: 51-
Garber
 Marine Ecology-Progress 
Garber
merjohn. 1989. Ecosystems Processes Component and Benthic Exchange 
Giblin,
cle in salt marshes. Limnology and Oceanography 29: 47-
Fisher, T. R., J. D. Hagy, W. R. Boynton, and M. R. Williams. 2006. Cultural 
eutrophication in the Choptank and Patuxent estuaries o
Limnology and Oceanography 51: 435-447. 
 T. R., L. W. Harding, Jr., D. W. Stanley, and L. G. Ward. 1988. 
Phytoplankton, nutrients, and turbidity in the Chesapeake, Delaware, and
Hudson estuaries. Est. Coast. Shelf Sci. 27: 61-93. 
 T. R., E. R. Peele, J. W. Ammerman, and L. W. Harding, Jr. 1992. Nutrient
limitation of phytoplankton in Chesapeake Ba
63. 
, J. H. 1984. N-15 tracer study of the short-term fate of particulate organic 
nitrogen at the surface of coastal marine - sedimets.
Series 16: 89-104. 
, J. H., W. R. Boynton, J. M. Barnes, L. L. Matteson, J. L. Watts, and S. 
Stam
and Sediment Transformation. 
 A. E., and R. W. Howarth. 1984. Porewater evidence for a dynamic 
sedimentary iron cy
63. 
 77 
 
 
Gilbert, F., R. C. Aller, and S. Hulth. 2003. The influence of macrofaunal bu
spacing and diffusive scaling on sedimentary nitrification and denitrification: 
An e
rrow 
xperimental simulation and model approach. J. Mar. Res. 61: 101-125. 
gen 
Grassle  
Hagy, J. D., W. R. Boynton, C. W. Keefe, and K. V. Wood. 2004. Hypoxia in 
ay 
c conditions. Geochimica et 
Heggie
ents fo 
s of bigenic N2 and N-metabolite fluxes 
Heilsko
Gobler, C. J., N. J. Buck, M. E. Sieracki, and S. A. Sanudo-Wilhelmy. 2006. Nitro
and silicon limitation of phytoplankton communities across an urban estuary: 
The East River-Long Island Sound system. Estuarine Coastal and Shelf 
Science 68: 127-138. 
, J. F., and J. P. Grassle. 1974. Opportunistic life histories and genetic systems
in marine benthic polychaetes J. Mar. Res. 32: 253-284. 
Chesapeake Bay, 1950-2001: Long-term change in relation to nutrient loading 
and river flow. Estuaries 27: 634-658. 
Harvey, H. R., J. H. Tuttle, and J. T. Bell. 1995. Kinetics of phytoplankton dec
during simulated sedimentation:  changes in biochemical composition and 
microbial activity under oxic and anoxi
Cosmochimica Acta 59: 3367-3377. 
, D., G. W. Skyring, J. Orchardo, A. R. Longmore, G. J. Nicholson, and W. M. 
Berelson. 1999. Denitrification and denitrifying efficiencies in sedim
Port Phillip Bay: direct determination
with implications for water quality. Marine Freshwater Research 50: 589-596. 
v, A. C., and M. Holmer. 2001. Effects of benthic fauna on organic matter 
mineralization in fish-farm sediments: importance of size and abundance. Ices 
Journal of Marine Science 58: 427-434. 
 78 
 
 
Henriksen, K., and W. M. Kemp. 1988. Nitrification in estuarine and coastal marin
sediments, p. 207-250. In T. H. Blackburn and J. Sorensen [eds.], Nitrogen 
Cycling in Coastal Marine Environments
e 
. John Wiley & Sons. 
: 
- 
flux 
Mar. Ecol. Prog. Ser. 61: 87-96. 
tion 
Kana, T am, G. E. Bennett, and J. C. 
s. Anal. Chem. 
Kana, T
mnology and Oceanography 42: 334-339. 
Kana, T
n" by B. D. Eyre, 
S. Rysgaard, T. Daisgaard, and P. Bondo Christensen. 2002. Estuaries 25 : 
1077-1087. Estuaries 27: 173-176. 
Jenkins, M. C., and W. M. Kemp. 1984. The coupling of nitrification and 
denitrification in two estuarine sediments. Limnology and Oceanography 29
609-619. 
Jensen, M. H., E. Lomstein, and J. Sorensen. 1990. Benthic NH
4
+
 and NO
3
following sedimentation of a spring phytoplankton bloom in Aarhus Bight, 
Denmark. 
Joye, S. B., and J. T. Hollibaugh. 1995. Influence of sulfide inhibition of nitrifica
on nitrogen regeneration in sediments. Science 270: 623-625. 
. M., C. Darkangelo, M. D. Hunt, J. B. Oldh
Cornwell. 1994. Membrane inlet mass spectrometer for rapid high-precision 
determination of N
2
, O
2
, and Ar in environmental water sample
66: 4166-4170. 
. M., M. B. Sullivan, J. C. Cornwell, and K. Groszkowski. 1998. 
Denitrification in estuarine sediments determined by membrane inlet mass 
spectrometry. Li
 
. M., and D. L. Weiss. 2004. Comment on "Comparison of isotope pairing 
and N-2 : Ar methods for measuring sediment denitrificatio
 79 
 
 
Karlson, K., E. Bonsdorff, and R. Rosenberg. 2007. The impact of benthic 
macrofauna for nutrient fluxes from Baltic Sea sediments. Ambio 36: 161-
167. 
Kemp, W. M. and others 2005. Eutrophication of Chesapeake Bay: historical trends 
29. 
Kemp, . 
 Ammonium recycling versus denitrification in Chesapeake Bay 
Levin, 
Linton, dity of Capitella sp 
ss 
Llanso, R. J. 1991. Tolerance of low dissolved-oxygen and hydrogen-sulfide by the 
Luther, G. W., Iii, B. Sundby, B. L. Lewis, P. J. Brendel, and N. Silverburg. 1997. 
Mackin  1988. The effectes of iron reduction and 
nonsteady -state diagenesis on iodine, ammonium, and boron distributions in 
and ecological interactions. Marine Ecology-Progress Series 303: 1-
W. M., P. Sampou, J. Caffrey, M. Mayer, K. Henriksen, and W. R. Boynton
1990.
sediments. Limnology and Oceanography 35: 1545-1563. 
L. A. 1986. Effects of enrichment on reproduction in the opportunistic 
polychaete Streblospio-Benedicti (Webster) - a mesocosm study. Biological 
Bulletin 171: 143-160. 
 D. L., and G. L. Taghon. 2000. Feeding, growth, and fecun
I in relation to sediment organic concentration. Marine Ecology-Progre
Series 205: 229-240. 
polychaete Streblospio-Benedicti (Webster). Journal of Experimental Marine 
Biology and Ecology 153: 165-178. 
Interactions of manganese with the nitrogen cycle:  alternative pathways to 
dinitrogen. Geochimica et Cosmochimica Acta 61: 4043-4052. 
, J. E., R. C. Aller, and W. J. Ullman.
 80 
 
 
sediments from the Amazon Continental-Shelf. Continental Shelf Research 
363-386. 
, T. C., W. C. Boicourt, J. C. Cornwell, L. W. Harding, Jr., and J
8: 
Malone . C. 
Martin, . T. Banta. 1992. The measurement of sediment irrigation rates: a 
s. J. 
Mayer,
cts of sediment NH4+ and 
Mermil
ce of bioturbation by three benthic infaunal species 
ent. 
Newell, R. I. E., M. S. Owens, and J. C. Cornwell. 2002. Influence of simulated 
ics. 
Nielsen
62. 
 
Stevenson. 2003. The Choptank River:  A Mid-Chesapeake Bay Index Site for 
Evaluating Ecosystem Reponses to Nutrient Management.  EPA Agreement 
R826941-01-0. UMCES Horn Point Laboratory. 
 W., and G
comparison of the Br- tracer and 
222
Rn/
226
Ra disequilibrium technique
Mar. Res. 50: 125-154. 
 M. S., L. Schaffner, and W. M. Kemp. 1995. Nitrification potentials of 
benthic macrofaunal tubes and burrow walls:  effe
animal irrigation behavior. Mar. Ecol. Prog. Ser. 121: 157-169. 
lod-Blondin, F., R. Rosenberg, F. Francois-Carcaillet, K. Norling, and L. 
Mauclaire. 2004. Influen
on microbial communities and biogeochemical processes in marine sedim
Aquatic Microbial Ecology 36: 271-284. 
bivalve biodeposition and microphytobenthos on sediment nitrogen dynam
Limnology and Oceanography 47: 1367-1369. 
, L. P. 1992. Denitrification in sediment determined from nitrogen isotope 
pairing. FEMS Microbiol. Ecol. 86: 357-3
 81 
 
 
Nielsen, O. I., B. Gribsholt, E. Kristensen, and N. P. Revsbech. 2004. Microscale
distribution of oxygen and nitrate in sediment inhabited by Nereis 
diversicolor: spatial patterns and estimated reac
 
tion rates. Aquatic Microbial 
Nixon, ing in coastal marine 
ecosystems, p. 111-138. In B. J. Neilson and L. E. Cronin [eds.], Estuaries and 
Nixon, and-sea 
Orth, R. J., and K. A. Moore. 1984. Distribution and abundance of submerged aquatic 
531-540. 
Parsons
water Analysis. Pergamon Press. 
d 
Rysgaa Nielsen. 1994. 
 
1652. 
Ecology 34: 23-32. 
S. W. 1981. Remineralization and nutrient cycl
Nutrients. Humana Press. 
 S. W. and others 1996. The fate of nitrogen and phosphorus at the l
margin of the North Atlantic Ocean. Biogeochemistry 35: 141-180. 
vegetation in Chesapeake Bay: an historical perspective. Estuaries 4: 
, T. R., Y. Maita, and C. M. Lalli. 1984. A Manual of Chemical and 
Biological Methods for Sea
Revsbech, N. P. 1989. An oxygen microsensor with a guard cathode. Limnology an
Oceanography 34: 474-478. 
Rossi, F., and C. Lardicci. 2002. Role of the nutritive value of sediment in regulating 
population dynamics of the deposit-feeding polychaete Streblospio shrubsolii. 
Marine Biology 140: 1129-1138. 
rd, S., N. Risgaard-Petersen, N. P. Sloth, K. Jensen, and L. P. 
Oxygen regulation of nitrification and denitrification in sediments. Limnology
and Oceanography 39: 1643-
 82 
 
 
Sampou, P., and C. A. Oviatt. 1991. A carbon budget for a eutrophic marine 
ecosystem and the role of sulfur metabolism in sedimentary carbon, oxygen 
and energy dynamics. J. Mar. Res. 49: 825-844. 
e Delaware River Basin. Pennsylvania 
Seitzin  
 Limnology and 
Stevens
 
orking. J. 
ica Et 
 147-157. 
phic 
Seitzinger, S. P. 1988. Benthic nutrient cycling and oxygen consumption in the 
Delaware estuary, p. 133-147. In S. K. Majumdar, E. W. Miller and L. E. Sage 
[eds.], Ecology and Restoration of th
Academy of Science. 
ger, S. P., and S. W. Nixon. 1985. Eutophication and the rate of denitrification
and N2O production in coastal marine sediments.
Oceanography 30: 1332-1339. 
on, J. C., L. W. Staver, and K. W. Staver. 1993. Water quality associated with 
survival of submersed aquatic vegetation along an estuarine gradient. 
Estuaries 16: 346-361.
Sun, M., R. C. Aller, and C. Lee. 1991. Early diagenesis of chlorphyll-a in Long 
Island Sound sediments: a measure of carbon flux and particle rew
Mar. Res. 49: 379-401. 
Sun, M. Y., C. Lee, and R. C. Aller. 1993. Laboratory studies of oxic and anoxic 
degradation of chlorophyll-a in Long Island Sound sediments. Geochim
Cosmochimica Acta 57:
Svensson, J. M. 1998. Emission of N
2
O, nitrification and denitrification in a eutro
lake sediment bioturbated by Chironomus plumosus. Aquatic Microbial 
Ecology 14: 289-299. 
 83 
 
 
Taylor, D., S. Nixon, S. Granger, and B. Buckley. 1995. Nutrient limitation and th
eutrophication of coastal lagoons, p. 235-244. 
rup, B., and T. Dalsgaard. 2002. Pro
e 
Thamd duction of N-2 through anaerobic 
Trimm M. Stewart, and D. B. Nedwell. 1999. The spring 
5: 37-46. 
arinen. 
 
oporeia affinis. 
Twilley o 
esapeake Bay, p. 
Van He
Prog. Ser. 114: 303-313. 
g 
on 
ica Acta 66: 
457-465. 
ammonium oxidation coupled to nitrate reduction in marine sediments. 
Applied and Environmental Microbiology 68: 1312-1318. 
er, M., R. J. Gowen, B. 
bloom and its impact on benthic mineralisation rates in western Irish Sea 
sediments. Marine Ecology-Progress Series 18
Tuominen, L., K. Makela, K. K. Lehtonen, H. Haahti, S. Hietanen, and J. Kup
1999. Nutrient fluxes, porewater profiles and denitrification in sediment
influenced by algal sedimentation and bioturbation by Mon
Estuarine Coastal and Shelf Science 49: 83-97. 
, R. R., and W. M. Kemp. 1986. The relation of denitrification potentials t
selected physical and chemical factors in sediments of Ch
277-293. In D. Wolfe [ed.], Estuarine Variability. Academic. 
ukelem, L., A. J. Lewitus, T. M. Kana, and N. E. Craft. 1994. Improved 
separations of phytoplankton pigmets using temperature-controlled high 
performance liquid chromatorgrapy. Mar. Ecol. 
Van Mooy, B. A. S., R. G. Keil, and A. H. Devol. 2002. Impact of suboxia on sinkin
particulate organic carbon: Enhanced carbon flux and preferential degradati
of amino acids via denitrification. Geochimica Et Cosmochim
 84 
 
 
Vaquer-Sunyer, R., and C. M. Duarte. 2008. Thresholds of hypoxia for marine 
biodiversity. Proceedings of the National Academy of Sciences of the Un
States of America 105: 15452-15457. 
ited 
 
 85