SEPA
United States
Environmental Protection
Agency
EPA/600/R-16/202 | September 2016
www.epa.gov/ord
Modeling Benthic Sediment Processes
To Predict Water Quality
and Ecology in Narragansett Bay
Atmosphere
Water column
Particulate Organic
Matter
Flux to/from water
t i i
-S i ¦ ,cnn,
Layer 1 = 1 cm
(Oxic or Anoxic)
C
o
Phosphate
Ammonium
Niti
~~err
-------
-------
United States
Environmental Protection
Agency
EPA/600/R-16/202 | September 2016
www.epa.gov/ord
Modeling Benthic Sediment Processes To
Predict Water Quality
and Ecology in Narragansett Bay
Mohamed A. Abdelrhman
U.S. Environmental Protection Agency
Atlantic Ecology Division
NHEERL, ORD
27 Tarzwell Drive
Narragansett, RI 02882 USA
National Health and Environmental Effects Research Laboratory
Office of Research and Development
Narragansett, RI 02882 USA
by
-------
DISCLAIMER
This document is a final report. It has not been formally released by the U.S. Environmental
Protection Agency and should not at this stage be construed to represent Agency policy. It is
being circulated for comments on its technical merit and policy implications. Although the
material described here has been funded by the U.S. Environmental Protection Agency, it has not
been subject to Agency-level review and therefore does not necessarily reflect the views of the
Agency, nor does mentioning trade names or commercial products endorse or recommend them.
This report has the Tracking Number ORD-012538 of USEPA Office of Research and
Development, National Health and Environmental Effects Research Laboratory, Atlantic
Ecology Division.
11
-------
CONTENTS
DISCLAIMER ii
CONTENTS iii
FIGURES v
TABLES vii
ACKNOWLEDGMENTS viii
1. Introduction 1
1.1. Background 1
1.2. Objective 2
1.3. Approach 2
1.4. Sediment quality setting 3
1.5. Data 3
1.5.1. Historical data 3
2. Mass balance and temperature 10
2.1. Mass balance of ammonium, nitrate, phosphate, and sulfide/methane 10
2.2. Mass balance of silica 12
2.3. Temperature of benthic sediment 13
3. Modeled sediment processes 13
3.1. Depositional Flux 13
3.2. Diagenesis Flux 14
3.2.1. Ammonium Nitrogen 15
3.2.2. Nitrate Nitrogen 16
3.2.3. Phosphate Phosphorus 16
3.2.4. Carbon (Sulfide/methane and oxygen demand) 17
3.2.4.1. Sulfide 17
3.2.4.2. Sediment oxygen demand 17
3.2.4.3. Methane 17
3.3. Chemical Flux 18
4. Sediment model configuration 18
4.1. Boundary conditions 19
4.2. Initial conditions 19
4.3. Model coefficients and parameters 19
5. Calibration 22
6. Results 28
iii
-------
7. Summary and Conclusion 31
References 32
Appendix A: Values of sediment water quality parameters used for Narragansett Bay 34
Appendix B: Time series graphs of sediment water quality parameters and fluxes for
Narragansett Bay 39
iv
-------
FIGURES
Figure 1. General layout of Narragansett Bay with locations of data stations for
hydrodynamics, water quality, and benthic sediment data 6
Figure 2. General structure and processes of the benthic sediment diagenesis model 7
Figure 3. Sediments of Narragansett Bay (Raposa 2009) 8
Figure 4. Fulweiler and Nixon's model of SOD flux based on their Narragansett Bay
observations. (Modified from Fulweiler and Nixon, 2010) 9
Figure 5. Kremer and Nixon's model of P04, NH4, and Si fluxes based on their
Narragansett Bay observations. (Modified from Kremer and Nixon, 1978) 9
Figure 6. Layout of the model numerical grid and the data observation stations 21
Figure 7. Predicted sediment bed temperature (SMT) 23
Figure 8. Benthic flux rate of sediment oxygen demand (SOD) 23
Figure 9. Benthic flux rate of ammonium nitrogen (FNH4) 24
Figure 10. Benthic flux rate of phosphate phosphorus (FP04D) 24
Figure 11. Benthic flux rate of silica (FSAD) 25
Figure 12. Comparison between calculated and predicted trends of SOD with temperature 25
Figure 13. Comparison between calculated and predicted trends of ammonium nitrogen
with temperature 26
Figure 14. Comparison between calculated and predicted trends of phosphate phosphorus
with temperature 26
Figure 15. Comparison between calculated and predicted trends of silicate silica
with temperature 27
Figure 16. Time series of chlorophyll a concentrations in the water column (A and B)
and POC in the bed sediment (C) at stations CP and GD in 2009 29
Figure 17. Benthic fluxes of NH4, N03, P04, Si, SOD, and COD (A-F, respectively)
at stations CP and GD 30
Figure 18. Time series of predicted ammonium concentration in Layerl at all stations 41
Figure 19. Time series of predicted ammonium concentration in Layer2 at all stations 41
Figure 20. Time series of predicted ammonium flux from benthic sediment at all stations 42
Figure 21. Time series of predicted nitrate concentration in Layerl at all stations 42
Figure 22. Time series of predicted nitrate concentration in Layer2 at all stations 43
Figure 23. Time series of predicted nitrate flux from benthic sediment at all stations 43
Figure 24. Time series of predicted phosphate concentration in Layerl at all stations 44
Figure 25. Time series of predicted phosphate concentration in Layer2 at all stations 44
Figure 26. Time series of predicted phosphate flux from benthic sediment at all stations 45
Figure 27. Time series of predicted sulfide concentration in Layerl at all stations 45
Figure 28. Time series of predicted sulfide concentration in Layer2 at all stations 46
Figure 29. Time series of predicted oxygen flux into benthic sediment at all stations 46
Figure 30. Time series of predicted COD flux from benthic sediment at all stations 47
v
-------
Figure 31. Time series of predicted silica concentration in Layerl at all stations 47
Figure 32. Time series of predicted silica concentration in Layer2 at all stations 48
Figure 33. Time series of predicted silica flux from benthic sediment at all stations 48
Figure 34. Time series of predicted temperature of benthic sediment at all stations 49
Figure 35. Time series of predicted PON concentration in Layer2 at all stations 50
Figure 36. Time series of predicted POP concentration in Layer2 at all stations 50
Figure 37. Time series of predicted POC concentration in Layer2 at all stations 51
Figure 38. Time series of predicted Carbonaceous SOD flux into benthic sediment at all stations.
51
Figure 39. Time series of predicted nitrogenous SOD flux from benthic sediment at all stations.
52
Figure 40. Time series of predicted enhanced phosphate flux in Layerl at all stations 52
Figure 41. Time series of predicted enhanced silica flux in Layerl at all stations 53
Figure 42. Time series of predicted surface mass transfer coefficient of benthic sediment
at all stations 53
Figure 43. Time series of predicted nitrification flux from benthic sediment at all stations 54
Figure 44. Time series of predicted denitrifi cation flux from benthic sediment at all stations.... 54
Figure 45. Time series of predicted aqueous sulfide flux from benthic sediment at all stations.. 55
Figure 46. Time series of predicted gaseous methane flux from benthic sediment at all stations. 55
Figure 47. Time series of predicted diagenesis nitrogen flux from benthic sediment at all stations.
56
Figure 48. Time series of predicted diagenesis phosphorous flux from benthic sediment
at all stations 56
Figure 49. Time series of predicted diagenesis carbon flux from benthic sediment at all stations.
57
Figure 50. Time series of predicted nitrogen depositional flux to class G1 at all stations 57
Figure 51. Time series of predicted nitrogen depositional flux to class G2 at all stations 58
Figure 52. Time series of predicted nitrogen depositional flux to class G3 at all stations 58
Figure 53. Time series of predicted phosphate depositional flux to class G1 at all stations 59
Figure 54. Time series of predicted phosphate depositional flux to class G2 at all stations 59
Figure 55. Time series of predicted phosphate depositional flux to class G3 at all stations 60
Figure 56. Time series of predicted carbon depositional flux to class G1 at all stations 60
Figure 57. Time series of predicted carbon depositional flux to class G2 at all stations 61
Figure 58. Time series of predicted carbon depositional flux to class G3 at all stations 61
vi
-------
TABLES
Table 1. State variables in the benthic sediment model 5
Table 2. Stations and locations used for predicted benthic sediment parameters 20
Table 3. Initial values for benthic sediment model 20
vii
-------
ACKNOWLEDGMENTS
The author appreciates the effort by the late Dr. John Hamrick (Tetra Tech, Inc.) for providing
the executables for EFDC and the templates for water and sediment quality input files. The
author also thanks the in-house reviewers of this report, including Drs. Dan Campbell, Jason
Grear, and Henry Walker (USEPA-AED) for their technical reviews, insights, and constructive
comments.
viii
-------
1. Introduction
This report presents the methodology to apply and calibrate the built-in sediment processes
model provided with the Environmental Fluid Dynamics Code (EFDC). The advection and
dispersion mechanisms in the water column were generated simultaneously by the hydrodynamic
model for Narragansett Bay (NB, Bay, or system) (Abdelrhman, 2015). The methodology for
water quality in the Bay is presented in Abdelrhman (2016a) and the methodology to model total
suspended solids is covered in Abdelrhman (2016b). The sediment quality model receives
particulate and dissolved matter produced by the water quality model and it returns benthic
fluxes back into the water column. This report presents sediment processes which generate
benthic fluxes of various water quality loads into the overlying water column.
The organization of this report includes seven sections with relevant tables and figures at the end
of each. Section 1 presents an introduction with the background, objective, approach, sediment
model setting, and available data. Section 2 presents equations for the mass balance and
temperature in the sediment quality model. Section 3 describes the various sediment processes.
Section 4 covers model configuration for NB including initial conditions and boundary
exchanges. Section 5 presents model calibration. Section 6 presents results of the constituent
concentrations in the sediment and their exchanges with the water column. Section 7 presents the
summary and conclusion with a brief discussion of the use of the sediment quality model results
and some suggestions for future work to improve predictions. More data and results are
presented in appendices A and B.
1.1. Background
Contaminant and nutrient loadings enter the Narragansett Bay as point sources from rivers and
wastewater treatment plants (WWTPs) (Fig. 1) and as non-point sources through dry/wet
atmospheric deposition on the water surface and benthic fluxes from bed sediments to the
overlying water column. The benthic sediment acts as a huge reservoir of particulate and
dissolved material (within interstitial water) which contributes to loadings of contaminants and
nutrients to the water column. A benthic sediment model is presented in this report to identify
benthic fluxes of various chemicals in Narragansett Bay.
Field observations of benthic fluxes in Narragansett Bay covered their spatial and temporal
behaviors. Some of these studies included nitrification and denitrification rates (Fulweiler and
Heiss, 2014, Fulweiler and Nixon, 2012), sediment oxygen demand and nutrient fluxes
(Fulweiler and Nixon, 2010), rates of nitrification along the Bay (Berounsky and Nixon, 1993),
spatial and seasonal observations of nutrient fluxes in the Bay (Elderfield et al., 1981), and
relationships between benthic fluxes and temperature (Kremer and Nixon, 1978). The major two
outcomes from field observations were (1) there existed a seasonal variability with more activity
in benthic fluxes of various nutrients during the warmer summer season than the colder winter
months, and (2) benthic fluxes had decreasing gradient from north, where most organic deposits
existed over the silty bed, to the south where organic deposits were less over the more sandy bed.
1
-------
Models of benthic sediment processes, including diagenesis and nutrient fluxes, were developed
to predict the spatial and temporal variability in loading of various chemicals and nutrients from
bed sediments (Ditoro and Fitzpatrick, 1993, DiToro, 2001). Applications of such benthic
sediment models covered estuarine systems (e.g., Chesapeake Bay, Park et al., 1995, Brady et
al., 2013, Testa et al., 2013) as well as freshwater systems (e.g., Lake Thunderbird, Oklahoma
Department of Environmental Quality-http://www.deq. state.ok.us/WQDnew/tmdl/thunderbird).
To the author's knowledge, predictive modeling of benthic sediment processes has never been
conducted for Narragansett Bay. This report provides description of the sediment process model
for Narragansett Bay.
The sediment process formulation in EFDC is from Park et al. (1995), which is based on the
model of DiToro and Fitzpatrick (1993) (see also DiToro, 2001). The sediment process module
was coupled with the water quality model (CE-QUAL-ICM, Cerco and Cole, 1994). The
sediment process model has 27 water-quality related state variables and fluxes (Table 1).
1.2. Objective
The main objective of this work is to numerically model concentrations and fluxes of various
chemical constituents in the bottom sediment as they change in space and time. The specific
objective is to identify loadings (fluxes) of dissolved water quality constituents from the
sediment to the water column in Narragansett Bay during the year 2009.
1.3. Approach
The three basic processes that are modeled in the sediment are: (1) depositional flux of
particulate organic matter (POM) and silica to the sediment, (2) diagenesis flux of POM within
the sediment, and (3) chemical flux from the sediment to the water column. The model tracks the
mass of an element (e.g., nitrogen) in the chemical compound in any form (dissolved or
particulate) (i.e., for nitrate-nitrogen, the model considers nitrogen mass in nitrate). The major
benthic fluxes from the sediment to the water column include ammonium-nitrogen (NH4-N),
phosphate-phosphorus (P04-P), silicate-silica (Si04-Si), and sediment oxygen demand (SOD-
02). Unless otherwise specified in this report, the mention of the chemical compound or the
relevant element are considered to be synonymous to the element mass. The terms for nitrate
state variables represent the sum of both nitrate (N03) and nitrite (N02) nitrogen. Sediment
temperature is also modeled because it moderates all reaction rates.
The depositional fluxes of POM is distributed between three classes (Gl, G2, and G3), which
represent labile fraction, with half-life of 20 days; refractory fraction, with half-life of one year;
and an inert (nonreactive) fraction, respectively. In the abbreviated notation of POM, the M is
used as a place holder for the actual type of particulates (i.e., M is replaced by C, N, or P for
particulate organic carbon, nitrogen, or phosphorous, respectively).
2
-------
1.4. Sediment quality setting
The benthic sediment model treats the bed as two vertical layers (Fig. 2). The upper layer, Layer
1 (with thickness Hi = ~ 0.01 m), can be oxic or anoxic based on dissolved oxygen concentration
in the overlying water column. The lower layer, Layer 2 (with thickness H2 = ~ 0.10 m for
Narragansett Bay), is always anoxic. Due to the negligible thickness of Layer 1, deposited POM
and particulate biogenic silica proceed directly from the water column to Layer 2, where the
diagenesis of POM by bacteria and the dissolution of silica take place.
The general mass balance formulation for the sediment model defines the concentration fields for
ammonium, nitrate, phosphate, sulfide/methane in space and time (Section 2.1). A separate mass
balance formulation is presented for silica (Sections 2.2). The presented mathematical forms and
parameters are the same as in the above-cited references, the presented descriptions are
reorganized to meet the general formulation for the hydrodynamics (Abdelrhman, 2015) and the
water quality (Abdelrhman, 2016a).
1.5. Data
The map of the existing sediment types in NB (McMaster, 1960) indicates the dominance of
clay-silt throughout the bay (Fig. 3). For convenience, it is assumed that the available sparsely
observed benthic processes (Section 1.5.1) are spatially uniform throughout the Bay. The
temporal trends from these observations are used only for comparison with model predictions
(Section 5). Beyond calibration, model predictions are completely independent from such data in
space and time (Section 6).
1.5.1. Historical data
Analysis of historical benthic flux data from various stations within Narragansett Bay indicated
that the general behavior of benthic fluxes is driven mainly by bottom temperature and that they
could be represented by the following exponential forms for sediment oxygen demand (SOD)
(Fulweiler and Nixon, 2010) (Fig. 4). The various sampling stations existed in Providence River
and Greenwich Bay with samples collected in various years during 1983-1984, 1975-1976, and
1972-1973. Benthic fluxes of phosphate-phosphorus (P04-P), ammonium-nitrogen (NH4-N),
and silicate-silica (Si04-Si) were collected at three stations in the upper, mid, and lower Bay in
1973-1974 as reported by Kremer and Nixon (1978) (see also Nixon et al., 1976) (Fig. 5).
Benthic fluxes of nitrate (N03) and nitrite (N02) did not show clear relation to temperature
(Fulweiler and Nixon, 2012; Seitzinger et al., 1984; Kremer and Nixon, 1978).
SOD = -9.83 e0 078 T (mg 02 m"2 h"1)
P04 = 0.029 e013 T (mg-at P m"2 d"1)
NH4 = 0.192 e015 T (mg-at N m"2 d"1)
Si04 = 1.015 e0,11 T (mg-at Si m"2 d"1)
(4)
(3)
(1)
(2)
3
-------
Where T is the bottom water temperature (°C). The flux of SOD (mg O2 m"2 h"1) is always
negative (out from the water column into the sediment). The fluxes of P04, NH4, and Si (mg-
atom m"2 d"1 = mmol m"2 d"1) are always positive (from sediment into the water column). All
benthic fluxes were converted to the same units of the model-predicted fluxes (i.e., g-at m"2 d"1)
using the molecular weights for P, N, and Si (i.e., 31 g mol"1, 14 g mol"1, and 28 g mol"1,
respectively) and dividing by 1000 g mg"1. Figures 4 and 5 present the recalculated exponential
relationships of the benthic fluxes using model units (g m"2 d"1). These relationships are used to
calibrate model predictions as described below.
4
-------
Table 1. State variables in the benthic sediment model
The three classes Gl, G2, and G3 for particulate organic matter represent labile fraction, with
half-life of -20 days; refractory fraction, with half-life of one year; and an inert (nonreactive)
fraction, respectively. The model assumes that there are two benthic sediment layers: the upper
layer, Layer 1 (with thickness Hi = ~ 0.01 m), can be oxic or anoxic and the lower layer, Layer 2
(with thickness H2 = ~ 0.10 m for Narragansett Bay) which is always anoxic.
State variable number and description
Name
(1) particulate organic carbon Gl class in layer 2
POC1
(2) particulate organic carbon G2 class in layer 2
POC2
(3) particulate organic carbon G3 class in layer 2
POC3
(4) particulate organic nitrogen Gl class in layer 2
PON1
(5) particulate organic nitrogen G2 class in layer 2
PON2
(6) particulate organic nitrogen G3 class in layer 2
PON3
(7) particulate organic phosphorus Gl class in layer 2
POP1
(8) particulate organic phosphorus G2 class in layer 2
POP2
(9) particulate organic phosphorus G3 class in layer 2
POP3
(10) particulate biogenic silica in layer 2
PSi
(11) sulfide/methane in layer 1
H2Si
(12) sulfide/methane in layer 2
H2S2
(13) ammonium nitrogen in layer 1
NH4i
(14) ammonium nitrogen in layer 2
NH42
(15) nitrate nitrogen in layer 1
N03i
(16) nitrate nitrogen in layer 2
N032
(17) phosphate phosphorus in layer 1
P04i
(18) phosphate phosphorus in layer 2
P042
(19) available silica in layer 1
SIi
(20) available silica in layer 2
SI2
(21) ammonium nitrogen flux
FNH4
(22) nitrate nitrogen flux
FN03
(23) phosphate phosphorus flux
FP04D
(24) silica flux
FSAD
(25) sediment oxygen demand (flux)
BF02
(26) release of chemical oxygen demand (flux)
BFCOD
(27) sediment temperature
SMT
5
-------
Blackstone River
•'Ten Mile River
PR A
Providence 2®
Warren/
Palmer River
Brayton'/ _ „
Pt •d 1 Fall R,ver
? A> FR1
Quonset
Point
AQP1
Aquidneck
Island
~ Stations
• WWTPs
-71°20'
-7no'
Figure 1. General layout of Narragansett Bay with locations of data stations for
hydrodynamics, water quality, and benthic sediment data.
6
-------
Nitrate
Phosphate
Ammonium
Layer 1 = 1 cm
(Oxic or Anoxic)
Atmosphere
Layer 2 = 10 cm
SOD
Deep layer
Figure 2. General structure and processes of the benthic sediment diagenesis model
(modified from DiToro, 2001). SOD fluxes contribute to ammonium nitrification and methane
oxidation. Flux of N2 gas from denitrification is lost to the atmosphere. Freshwater aqueous and
gaseous methane fluxes are not shown.
7
-------
Upland
Gravel
Sandy gravel
Gravel-sand-silt
Sand
Gravelly sand
Silty sand
Silt
Sandy silt
Clay-silt
Sand-silt-clay
Gravel-silt-clay
Rock
8 Kilometers
Figure 3. Sediments of Narragansett Bay (Raposa 2009).
All sediment data are from McMaster (1960) as presented in Rhode Island Geographic
Information System (RIGIS, www.edc.uri.edu/rigis) and Lee et al. (2000). Note the dominance
of clay-silt sediments in the mid- and upper Bay regions and the coarser sediments lower in
the Bay.
8
-------
Bed Temperature (°C)
0
30
0.00
-0.20
-0.40
-0.60
= -1.00
£ -1.20
-1.40
-1.60
-1.80
Figure 4. Fulweiler and Nixon's model of SOD flux based on their Narragansett Bay
observations. (Modified from Fulweiler and Nixon, 2010).
— NH4
• P04
Si
0.50
0.45
ZT 0.40
™ 0.35
E
up 0.30
J 0.25
u 0.20
5 0.15
m 0.10
0.05
0.00
/
/
/
10 15 20
Temperature (°C)
25
30
Figure 5. Kremer and Nixon's model of P04, NH4, and Si fluxes based on their
Narragansett Bay observations. (Modified from Kremer and Nixon, 1978).
9
-------
2. Mass balance and temperature
2.1. Mass balance of ammonium, nitrate, phosphate, and sulfide/methane
The general mass balance formulations for layers 1 and 2, as developed in DiToro and
Fitzpatrick (1993) (see also DiToro, 2001) and applied in Park et al. (1995), are presented below
in Equations (5) and (6), respectively. Equation (5) has an extra term to define the exchange
between the surface sediment layer and the water column. The last two terms in both equations
define the reactions and fluxes of each compound, as presented in Section 3, respectively.
In equation (5) below, the mass balance for ammonium, nitrate, phosphate, and sulfide/methane
in Layer 1 includes the following six terms on the right hand side: (1) exchange of the dissolved
fraction between Layer 1 and the overlying water column, (2) exchange of the dissolved fraction
between Layer 1 and Layer 2, (3) exchange of the particulate fraction between Layer 1 and Layer
2, (4) loss by burial to Layer 2, (5) removal (sink) by reaction, and (6) internal sinks/sources.
Due to the small thickness of Layer 1, a steady state is assumed to exist there, i.e.,
dCti
Hx = 0 = s(fd0 ¦ Ct0 — fdx ¦ Ctx) + KL(fd2 1 Ct2 — fdx ¦ Ctx) + oo(fp2 1 Ct2 — fpi1 Ctx)
-W-Cti-Ckf/sKti+Ji (5)
In equation (6) below, The mass balance for ammonium, nitrate, phosphate, and sulfide/methane
in Layer 2 includes the following five terms on the right hand side: (1) exchange of dissolved
fraction between Layer 1 and Layer 2, (2) exchange of the particulate fraction between Layer 1
and Layer 2, (3) loss by burial to deep sediment, (4) removal (sink) by reaction, and (5) internal
sinks/sources, i.e.,
H2^= —KL(fd2 ¦ Ct2 - fdi ¦ CtO - oo(fp2 ¦ Ct2 - fPl ¦ Ctl} + W(Ctl - Ct2)
-k2Ct2 + J2 (6)
The last two terms in the above two equations representing reactions and sinks/sources have
different mathematical formulation for ammonium, nitrate, phosphate, or sulfide/methane and
are discussed in Section 5.2. Definitions of the equation parameters are as follows.
Hi = thickness of Layer 1 (-0.01 m), (calculated from Eq. (7) below)
H2 = thickness of Layer 2 (assumed 0.10 m for Narragansett Bay)
Cti = total concentration in Layer 1 (g m"3)
Ct2 = total concentration in Layer 2 (g m"3)
Cto = total concentration in the overlying water column (g m"3)
s = surface mass transfer coefficient (m day"1)
KL = diffusion velocity for dissolved fraction between Layer 1 and 2 (m day"1)
co = particle mixing velocity between Layer 1 and 2 (m day"1)
fdo = dissolved fraction of total substance in the overlying water (0 < fdo < 1)
fdi = dissolved fraction of total substance in Layer 1 (0 < fdi < 1)
fd2 = dissolved fraction of total substance in Layer 2 (0 < fick < 1)
fpi = particulate fraction of total substance in Layer 1 (= 1 - fdi)
10
-------
fp2 = particulate fraction of total substance in Layer 2 (= 1 - fch)
ki = reaction velocity in Layer 1 (m day"1)
k2 = reaction velocity in Layer 2 (m day"1)
Ji = sum of all internal sources in Layer 1 (g m"2 day"1)
h = sum of all internal sources including diagenesis in Layer 2 (g m"2 day"1)
W = burial rate (m day"1)
The surface mass transfer coefficient, s (m day"1), is calculated from the sediment oxygen
demand and the oxygen concentration in the lower part of the overlying water column, i.e.,
D, SOD
S ~ ~ DO^ ^
Where
Di = diffusion coefficient in Layer 1 (m2 day"1)
SOD = sediment oxygen demand (g O2 m"2 d"1) (Section 5.2.4.2)
DOo = oxygen concentration in the overlying water column (g O2 m"3)
The partitioning of substance between dissolved and particulate phases in Layers 1 and 2 is
parameterized as follows
1
fdi = (8a)
1 +
fd2 = (8b)
1 + m2n2
Where
mi, m2 = solid concentrations in Layer 1 and 2, respectively (kg L"1)
711,712 = partition coefficients (ratio of particulate to dissolved fraction per unit solid
concentration) in Layer 1 and 2, respectively (per kg L"1)
The diffusion velocity for the dissolved fraction between Layer 1 and 2, KL (m day"1), is
represented by two terms to account for: (1) the diffusive exchange, and (2) the bio-irrigation by
benthic organisms, i.e.,
Dd " 0DTd~2O)
KL = „Dd + Rbi.bt ¦ <*> (9)
"2
Where
Dd = diffusion coefficient in pore water (m2 day-1)
0Dd = constant for temperature adjustment for Dd
Rbi,bt = ratio of bio-irrigation to bioturbation
The particle mixing velocity for the solids fraction between Layer 1 and 2, co (m day"1), is
represented by two terms: (1) the diffusive exchange, and (2) a minimum diffusion for particle
mixing, i.e.,
11
-------
DO0 Dp^
H2 Gpocr KMDp + DO,, ll2
Where
Dp = apparent diffusion coefficient for particle mixing (m2 day"1)
0dp = constant for temperature adjustment for Dp
Gpoc,i = particulate organic carbon in the labile class G1 (g C m"3)
Gi'oc.r = reference concentration for Gpoc,i (g C m"3)
KMdp = particle mixing half-saturation constant for oxygen (g O2 m"3)
Dpmin = minimum diffusion coefficient for particle mixing (m2 day"1)
f(ST) = effect of benthic stress on benthic biomass and thus particle mixing (unitless), given by
f(ST) = 1 - KST " ST (11)
With
ST = accumulated benthic stress (day) (see Park et al., 1995 for details)
Kst = first order decay rate for ST (day"1), given by
dST / DO0 \
= -Kst ¦ ST + 1 - —- , DO0 < KMDp (12a)
at 31 V KMDp
aST = -Kst " ST, DO0 > KMDp (12b)
at
2.2. Mass balance of silica
The following mass balance equation for particulate biogenic silica, as developed in DiToro and
Fitzpatrick (1993) (see also DiToro, 2001) and applied in Park et al. (1995), includes four terms
(on the right hand side) representing dissolution (i.e., production of dissolved silica, Si), burial,
and depositional and detrital fluxes from the overlying water column, i.e.
aPSi
H2 = —Ssi 1 H2 — W ¦ PSi + Jpsi + JDSi (13)
Where
PSi = concentration of particulate biogenic silica in the sediment (g Si m"3)
Jpsi = depositional flux of PSi (g Si m"2 day"1)
JDSi = detrital flux of PSi (g Si m"2 day"1) (not associated with algal flux)
Ssi = dissolution rate of PSi in Layer 2 (g Si m"3 day"1), given by
s* = K* • 9s^~2°^ psi + KMPSj (Si—t " fd«> • «») CM)
Where
Ksi = first order dissolution rate for PSi at 20 °C in Layer 2 (day"1)
0si = constant for temperature adjustment for Ksi
12
-------
KMpsi = silica dissolution half-saturation constant for PSi (g Si m"3)
Sisat = saturation concentration of silica in the pore water (g Si m"3)
Si2 = concentration of dissolved mineralized silica in Layer 2 (g Si m"3)
The mass balance equations for the dissolved mineralized silica, Si, follow the general equations
(5) and (6) for layers 1 and 2, respectively. The terms representing internal sinks/sources and
removal by reaction are given below.
Internal Si sinks/sources for Layers 1 and 2,
Ji,si = 0 (15)
J2 Si = ksi01T20) nc PL Sisat ¦ H2 (16)
' Sl PSi + KMpSi
Internal Si reactions for Layers 1 and 2,
ki,si = 0 (17)
PSi
k"' = k-0-"2O)psTTKM^fd"''H2 (18)
2.3. Temperature of benthic sediment
The sediment temperature is modeled based on heat diffusion between the water column and
sediment, i.e.,
dT Dt
g^Ow-T) (19)
T = sediment temperature
Dt = heat diffusion coefficient between the water column and sediment (1,8xl0"7 m2 sec"1)
Tw = temperature in the overlying water column (°C) (provided by the hydrodynamic model,
Abdelrhman 2015)
H = total thickness of sediment layers H1+H2 ~ H2 (m)
3. Modeled sediment processes
3.1. Depositional Flux
Depositional flux includes particulates settling from the three algal groups (dinoflagellates,
diatoms, and green algae, denoted as x = c, d, and g, respectively), refractory and labile
particulate organic carbon, particulate organic phosphorous, and particulate organic nitrogen, and
particulate biogenic silica. The proportion of each particulate organic type going to each group
(Gl, G2, and G3) is specified in by the user (Appendix A - Cards 6-8). All fluxes proceed from
13
-------
the overlying water column directly to Layer 2 where they are partitioned into three classes Gi (i
= 1, 2, 3) with depositional fluxes as:
Jpoc.i = FCLPj ¦ WSLP ¦ LPOC + FCRPi ¦ WSRP ¦ RPOC + ^ FCBx4 ¦ WSX ¦ Bx (20)
x=c,d,g
JpoN.i = FNLP; ¦ WSLP ¦ LPON + FNRP; ¦ WSRP ¦ RPON
+ ^ FNBx i ¦ ANCX ¦ WSX ¦ Bx (21)
x=c,d,g
Jpop.i = FPLPj ¦ WSLP ¦ LPOP + FPRPi ¦ WSRP ¦ RPOP + ^ FPBxi ¦ APCX ¦ WSX ¦ Bx
x=c,d,g
+ YiWSTSs 1 P04P (22)
Jpsi = WSd ¦ SU + ASCd ¦ WSd ¦ Bd + WStss ¦ SAp (23)
JpoM,i = depositional flux of POM (M = C, N, or P) routed into Gi class (g C, N, or P m"2 day"1)
Jpsi = depositional flux rate of PSi (g Si m"2 day"1)
i = counter for depositional classes Gi, i = 1, 2, 3
yi = 1 for i = 1; and yi = 0 for i = 2 or 3
FCLPi, FNLPi, FPLPi = fraction of water column labile POC, PON, and POP routed into Gi
class, Si FCLPi = Si FNLP; = Si FPLP; = 1
FCRPi, FNRPi, FPRP;= fraction of water column refractory POC, PON, and POP routed into Gi
class, Si FCRPi = Si FNRPi = Si FPRP; = 1
FCBx,i, FNBx,i, FPBxj = fraction of POC, PON, and POP, respectively, in the algal group x (x =
c, d, g) routed into Gi class, Si FCBx,i = Si FNBx,i = Si FPBx,i = 1
WS = settling velocity
LP, RP = referencing labile and refractory particulates, respectively
TSS = total suspended solids provided by the hydrodynamic model
(Abdelrhman, 2016b)
3.2. Diagenesis flux
Diagenesis of POM, as developed in DiToro and Fitzpatrick (1993) (see also DiToro, 2001) and
applied in Park et al. (1995), is considered to take place only in Layer 2 with the mass balance
represented by the following equation:
H2 = — KpoM,i0poM,i'') ' GpoM.i 1 H2 — W ¦ GP0M,i + JpOM.i (24)
GpoM,i = concentration of POM (M = C, N, or P) in Gi class in Layer 2 (g m"3)
Ki-om.i = decay rate of POM (M = C, N, or P) in Gi class in Layer 2 at 20 °C. Kpom,3 = 0
OpoM,i = constant for temperature adjustment for Ki-om.i
T = sediment temperature (°C)
14
-------
The diagenesis flux is calculated for the two reactive classes G1 and G2 from the following
general equation:
Jm = diagenesis flux (g C, N, or P m"2 day"1) of carbon (M = C), nitrogen (M = N), or
phosphorus (M = P)
As described below, the diagenesis fluxes are included in the sink/source terms in general mass
balance equations (5) and (6). In addition, relevant internal reactions are also counted in these
equations.
3.2.1. Ammonium nitrogen
Internal sink/source in equations (5) and (6)
2
(25)
i =1
Jl,NH4 — 0 (26)
Internal reaction, nitrification flux (from Layer 1)
Where
k2 -
1,NH4 —
DO0
0 IV1V1NH4
KM
2 q(T 20) oq\
NH4°NH4
2. KMNh4,02 + DO0 KMNH4 + NH4!
^2,NH4 — 0 (30)
KMnH4,02
NH4i
KMNH4
kNH4
0NH4
JNit
nitrification half-saturation constant for dissolved oxygen (g O2 m"3)
total ammonium nitrogen concentration in Layer 1 (g N m"3)
nitrification half-saturation constant for ammonium (g N m"3)
optimal reaction velocity for nitrification at 20 °C (m day"1)
constant for temperature adjustment for kNH4
nitrification flux (g N m"2 day"1)
15
-------
3.2.2. Nitrate nitrogen
Internal sinks/sources in equations (5) and (6)
Jl,N03 = J Nit (31)
J2.N03 = 0 (32)
Internal reaction, denitrification flux
k2
IVl MHO
JN2(g)=^f21 ¦N031 + k2,N03-N032 (33)
Where
b-2 i_2 q(T—20) rr>A\
1,N03 — N03,l N03
^2,N03 = ^N03,2®N03 ^ (35)
Kno3,i = reaction velocity for denitrification in Layer 1 at 20 °C (m day"1)
Kno3,2 = reaction velocity for denitrification in Layer 2 at 20 °C (m day"1)
0NO3 = constant for temperature adjustment for Kno3,i and Kno3,2
JN2(g) = denitrification flux (g N m"2 day"1), gaseous (g)
N03i = total nitrate nitrogen concentration in Layer 1 (g N m"3)
N032 = total nitrate nitrogen concentration in Layer 2 (g N m"3)
3.2.3. Phosphate phosphorus
Internal sinks/sources in equations (5) and (6)
Ji,P04 = 0 (36)
2
J2.P04 = Jm = Jp = ^ KPOp,i1 0p^)P2i°') 1 GPOp,i1 H2 (37)
i =1
Internal reaction flux = 0
ki,po4 = k2,po4 = 0 (38)
As DOo approaches zero, the phosphate flux from the sediment increases. This is achieved by
enhancement of the partitioning between dissolved and particulate fractions in Layer 1 according
to the following
tt1,P04 = tt2,P04 1 (ATCp04,l) DO0 > (DO0)crit PQ4 (39a)
f i \ ^^o/C^^o)crit,P04 r\ ^ f r\ \ s t-\
tti,P04 — tt2,P04 1 (Atcpo^iJ DO0 < (DO0)critPo4 (39b)
(D0o)crit,po4 = critical DOo concentration for the enhanced phosphate flux from Layer 1
7ti,po4; 712,po4 = partition coefficients (ratio of particulate to dissolved fraction of phosphate per
unit solid concentration) in Layer 1 and 2, respectively (per kg Phosphate L"1)
AtcP04,i = amount of enhanced sorption of P04 in Layer 1
16
-------
3.2.4. Carbon (sulfide/methane and oxygen demand)
3.2.4.1. Sulfide
Sulfide applies only to non-freshwater systems including estuaries and coastal embayment.
Internal sulfide sinks/sources in equations (5) and (6)
Jl,H2S = 0 (40)
J2.H2S = a02,C " Jc — a02,N03 " JN2(g) (41)
Internal reaction, aqueous sulfide flux
Jaq,H2S = s(fdljH2S " H2SX - COD) (42)
Where
SOD
s=5o;
ao2.c = stoichiometric coefficient for carbon diagenesis consumed by sulfide oxidation
(2.6667 g 02-equivalents per g C)
ao2,N03 = stoichiometric coefficient for carbon diagenesis consumed by denitrification (2.8571 g
02-equivalents per g N)
H2Si = total sulfide concentration in layer 1 (g 02-equivalents m"3)
COD = chemical oxygen demand (g 02-equivalents m"3)
3.2.4.2. Sediment oxygen demand
k?
SOD = CSOD + NSOD = + a02,NH4 " Jwit (44)
kl,H2S = (^H2S,dl 1 fdi,H2S + kn2S.pl 1 fPl,H2s)0H2S ^ 9 k-m (45)
Z 1 i^ivlH2S,02
CSOD = carbonaceous sediment oxygen demand (g C m"2 d"1)
NSOD = nitrogenous sediment oxygen demand (g N m"2 d"1)
ao2,NH4 = stoichiometric coefficient for oxygen consumed by nitrification (4.33 g O2 per g N)
kH2s,di = reaction velocity for dissolved sulfide oxidation in Layer 1 at 20 °C (m day"1)
kH2s,Pi = reaction velocity for particulate sulfide oxidation in Layer 1 at 20 °C (m day"1)
0H2S = constant for temperature adjustment for kH2s,di and kms.pi
KMh2s,c>2 = constant to normalize the sulfide oxidation rate for oxygen (g O2 m"3)
3.2.4.3. Methane
Methane applies only to freshwater systems (i.e., lakes and rivers), but is included here for
possible future applications to lakes and reservoirs.
Internal methane sinks/sources in equations (5) and (6)
Jl,CH4 = 0 (46)
J2.CH4 = a02,C " Jc — a02,N03 " JN2(g) = CSOD + Jaq,CH4 + JcH4(g) (47)
17
-------
Where
Jaq,CH4 — CSODmax ¦ sech
(T—20)
CCH4 " ®CH4
= CSODmax - CSOD (48)
CSOD = CSODr
1 — sech
^CH4 °CH4
(T—20)
(49)
J aq,CH4
JcH4(g)
CSODn
kcH4
0CH4
CH4sat
h + H2
CSODmax = minimum (72 ¦ KL ¦ CH4sat ¦ ]2 Cm, ]2 Cm} (50)
CH4sat = 100 (l + h 1.024(20-t) (51)
: aqueous methane flux (g 02-equivalent m"2 day"1)
: gaseous methane flux (g 02-equivalent m"2 day"1)
: maximum CSOD occurring when all the dissolved methane transported to the
oxic Layer 1 is oxidized
: reaction velocity for dissolved methane oxidation in Layer 1 at 20 °C (m day"1)
: constant for temperature adjustment for kcH4
: saturation concentration of methane in the pore water (g 02-equivalent m"3)
: depth from water surface (m) that corrects for in situ pressure
3.3. Chemical flux
The aqueous sediment flux of ammonium, nitrate, phosphate, and sulfide/methane from Layer 1
to the overlying water column is given by the first term on the right hand side of equation (5)
Jaq = s(fdx ¦ Ctx - fd0 ¦ Cto) (52)
Jaq = aqueous flux (g m"2 d"1), positive from sediment to the overlying water
4. Sediment model configuration
Figure 6 presents the layout of the numerical segments used by the hydrodynamic and water
quality models for Narragansett Bay (Abdelrhman, 2015, 2016a,b). The benthic sediment exists
below the water column in each segment and it is coupled to it. Exchanges of fluxes between the
water column and bed sediments take place at the sediment-water interface. In this report,
benthic fluxes and chemical concentrations are reported at the same data stations for the water
quality model (Table 2).
18
-------
4.1. Boundary conditions
The sediment bed is assumed to have infinite lateral extent with no defined lateral boundaries.
The vertical structure of the model includes the lower part of the water column, two bottom
sediment layers (aerobic and anaerobic), and the deeper sediment (Fig. 2). The lower part of the
water column defines the upper boundary to the bottom sediment in each segment of the
numerical grid. Sediments deeper than the two defined bed layer thicknesses form the lower
boundary, which acts as a sink to particulates and chemicals through burial.
The water quality model (Abdelrhman, 2016a) provides all the required time series of boundary
conditions for the sediment model, which include depositional fluxes of J pom., and Jpsi as well as
the water column concentrations of all chemicals, Cto (Equation 5), and the water temperature,
Tw. All boundary conditions are updated at each time step of the hydrodynamic calculations.
4.2. Initial conditions
The initial conditions for the sediment model include the concentrations of particulates Gpotvy
and PSi in Layer 2 as well as the total concentrations of ammonium, nitrate, phosphate, and
sulfide/methane in the two layers, Cti and Ct2, in addition to the sediment temperature, T. In this
work, all initial values were assumed to have spatially uniform values commensurate with clay-
silt throughout the Bay (Table 3).
Assuming that chemicals in the Bay are in a long-term quasi-equilibrium condition, values at the
end of the year would be close to their initial values. This assumption was implemented when
visual inspection of model predictions indicated drastic difference between initial and ending
values. For such situations, initial values were set to match the values at the end of the simulation
on December, 31, 2016 at 24:00 h. All initial values were assumed to apply at 00:00 hour on
January 1st, 2009. Nonetheless, such rectification was not implemented as a new methodology
was applied to rerun the simulation for a specific year (e.g., 2009) a minimum of five times and
use the end-of-year results from the preceding run as restart values for the next run of the same
year.
4.3. Model coefficients and parameters
Appendix A presents values of sediment water quality parameters used for Narragansett Bay.
Presented values are based on previous applications of the presented sediment water quality
model to Chesapeake Bay (DiToro and Fitzpatrick, 1993) (see also DiToro, 2001).
19
-------
Table 2. Stations and locations used for predicted benthic sediment parameters
(Abdelrhman, 2015)
Model Total
Numerical
Station Name,
Latitude
Longitude
Water Depth
Location
Abbreviation
(North)
(West)
(m)
I,J
Conimicut Point, CP
41.7138°
71.3438°
-12.3
17,33
Greenwich Bay, GB
41.684833°
71.44603°
-2.7
3,30
T-Wharf, TW
41.57885°
71.32145°
-12.0
19,19
Phillipsdale, PD
41.84175°
71.3722°
-5.4
12,46
Bullock Reach, BR
41.740567°
71.374667°
-7.5
13,36
Sally Rock, SR
41.675283°
71.4240167°
-3.4
6,29
N. Prudence Island, NPI
41.6704°
71.354717°
-11.2
15,29
Poppasquash Point, PP
41.647433°
71.317467°
-9.0
20,27
Mount View, MV
41.638467°
71.383683°
-7.1
11,26
Quonset Point, QP1
41.587617°
71.380033°
-8.3
12,21
Mount Hope Bay, MH
41.681333°
71.215217°
-6.0
33,30
URI/GSO, GD
41.49183°
71.4188°
-8.0
6,11
Table 3. Initial values for benthic sediment model
Parameter
Default valuea
State Variable
Name
(g m"3)
(1) particulate organic carbon G1 class in layer 2
CPOC1
100.0
(2) particulate organic carbon G2 class in layer 2
CPOC2
1000.0
(3) particulate organic carbon G3 class in layer 2
CPOC3
2000.0
(4) particulate organic nitrogen G1 class in layer 2
CPON1
60.0
(5) particulate organic nitrogen G2 class in layer 2
CPON2
200.0
(6) particulate organic nitrogen G3 class in layer 2
CPON3
400.0
(7) particulate organic phosphorus G1 class in layer 2
CPOP1
30.0
(8) particulate organic phosphorus G2 class in layer 2
CPOP2
50.0
(9) particulate organic phosphorus G3 class in layer 2
CPOP3
100.0
(10) particulate biogenic silica in layer 2
CPSi
500.0
(12) sulfide in layer 2
C2H2S
30.0
(13) ammonium nitrogen in layer 1
C1NH4
2.5b
(14) ammonium nitrogen in layer 2
C2NH4
30.0b
(16) nitrate nitrogen in layer 2
C2N03
0.0163b
(18) phosphate phosphorus in layer 2
C2P04
94.0b
(20) available silica in layer 2
C2Si
500.0
(27) sediment temperature
GT
3.0 (°C)b
Accumulated benthic stress0
CBSt
0.0
a Used by EFDC developer (Tetra Tech, 2005).
b Assumed from end of one year values on December 31, 2009 at 24:00 h.
0 Not a state variable
20
-------
Figure 6. Layout of the model numerical grid and the data observation stations
(Abdelrhman, 2015, 2016a,b).
21
-------
5. Calibration
The final calibration was performed after running the simulations for five years to spin up the
benthic sediment initial concentrations and reservoirs of various constituents (Section 4.2). The
model predicted sediment temperatures (Fig. 7) were used with the observed flux relationships
(Equations 1-4, Figs. 4-5) to produce the calculated values at the twelve stations in Table 2 (see
also Figs. 1 and 6) throughout the year 2009. The sediment model provided the predicted values
at the same stations during 2009.
Model parameters and coefficients were adjusted as needed to meet calibration needs for
Narragansett Bay. The enforced major calibration requirement was the quasi steady behavior
with similar values at the start and end of the year for all chemical concentrations in the sediment
and their fluxes to the overlying water column. Many parameters, however, were kept at their
default values that were used in previous applications (e.g., Chesapeake Bay/Lower Charles
River, Tetra Tech, 2005). Appendix A presents values of all the calibrated parameters and their
published ranges.
Comparisons between the time series of calculated benthic fluxes (Section 1.5.1) and calibrated
model predictions are presented in Figures 8-11. The vertical ranges in both the calculated and
the predicted values represent variations at the twelve stations at each time instance. Comparison
between observed and predicted trends of benthic fluxes with respect to sediment temperature
are shown in Figures 12-15. Trends of observed and predicted SOD are close with the use of an
enhancement coefficient (SODmult =2.5). The trends for silica are very also close. The
ammonium trend indicates a factor of two between observations and predictions. The trend of
observed phosphate fluxes indicates high spatial variability with temperatures at various stations
at the same time instant (-factor of two in the scatter range, Fig. 14). The observed trend is a
factor of four higher than the predicted trend.
Kremer and Nixon (1978) noted that relating observed benthic fluxes to temperature only is a
weak assumption, especially where heavy sediment and effluent loads exist. The range of
predicted values indicate that very weak benthic fluxes exist in the lower part of the Bay, where
sand content is high (Fig. 3), while higher fluxes are produced in the upper part of the Bay where
clay and silt dominate. The temporal pattern of benthic fluxes indicate correspondence not only
with temperature, but also with temporal behavior of organic particles (e.g., phytoplankton) as
they settle to the bed and contribute to such fluxes.
22
-------
30 ,
T C
Sec
diment Temp*
^rature
si 20
Cft CTi C
D o o c
N *Ns Xs
0 U-) "d- ^
H *H rH r
**• s ^
CO m c
r
XU/1H/U7
11/13/09
12/13/09
1/12/10
Figure 7. Predicted sediment bed temperature (SMT).
SOD Flux
Predicted BF02 Calculate BF02 (from Fulweiler& Mixon, 2010) (g/m2/d)
CO
CTi
0s!
cn
CTi
cn
0s!
cn
CTi
cn
CTi
cn
cn
O
O
O
o
o
O
o
•v.
o
o
O
o
o
O
O
rH
\
CO
r-.
ID
00
r^
r--
ID
UD
un
ro
ro
r\l
rH
rH
tH
rH
rH
rH
rH
rH
rH
rH
rH
rH
rH
rH
\
\
**>>
rN
\
rsl
tH
rsJ
m
^5*
LD
ID
r-.
00
cn
O
rH
rH
T—1
Date
rH
rH
rH
Figure 8. Benthic flux rate of sediment oxygen demand (SOD).
23
-------
Amonium Flux
0.18
0.16
0.14
0.12
Predicted FNH4
Calculated FNH4 (from K re trier & Nixon, 1978)(g/m2/d)
£ 0.10
-25 0.08
J 0.06
0.02
0.00
iji.LU=y
00
o
en
o
o
-------
Predicted FSAD
Silica Flux
Calculated FSAD (from Kremer& Nixon, 1978)(g/m2/d)
0.90
0.80
_ 0.70
TJ 0.60
(N
£ 0.50
3S 0.40
J 0.30
Ll_
0.20
0.10
0.00
Ieee
¦J—' 1 | |
cociObcncnaia^cntjitjiaiaiaio
ooooooooooooo^h
iD UD i-D ^
t—I t—I *—I t—I rH
"V "V "V
LO i£> f-* CO Cl
Date
m m rsi
Figure 11. Benthic flux rate of silica (FSAD).
Calculate BF02 (from Fulweiler& Nixon, 2010)
• Predicted SOD
Poly. (Calculate 8F02 (from Fulweiler& Nixon, 2010))
Poly. (Predicted SOD)
Temperature (oC)
0 5 10 15 20 25 30
0.00
-1.00
T -2.00
T3
rN
E -3.00
(N
O
QD -4.00
Q
° -5.00
-6.00
-7.00
y = -0.0019X2 - 0.0064X - 0.2535
y = -0.0026X2 - 0.0075X- 0.0482
R2 = 0.434
Figure 12. Comparison between calculated and predicted trends of SOD with temperature.
25
-------
• Calculated FNH4 (from Krerner & Nixon, 1978)
* Predicted FNH$
Expon (Calculated FNH4(from Kremer & Nixon, 1978))
Expon. (Predicted FNH$)
0.18
0.16
,14
£ 0.12
-22 o.io
= 0,0015eO1471*
Rz = 0.6905
0.08
0.06
53 0.04
CO
0.02
0.00
0
5
10
15
20
25
30
Temperature (°C)
Figure 13. Comparison between calculated and predicted trends of ammonium nitrogen
with temperature.
0.03
zr 0.025
~U
fN
E 0.02
35
a 0.015
• Calculated FP04D (from Kremer & Nixon, 1978)
• Predicted FP04D
Poly. (Calculated FP04D (from Kremer & Nixon, 1978))
Poly. (Predicted FP04D)
y = 4E-05x2 - 0.0002X + 0.0014
R21=0.9211
y = 8E-06x2 - 1E-05X + 0.0001
R21=0.4661
•;vi
m 0.005
10 15 20
Temperature (°C)
Figure 14. Comparison between calculated and predicted trends of phosphate phosphorus
with temperature.
26
-------
0.90
0.80
0.70
b 0.60
0.50
0.00
Calculated FSAD (from Kremer& Nixon, 1978)
Predicted FSAD
-Expon. (Calculated FSAD (from Kremer& Nixon, 1978))
-Expon. (Predicted FSAD)
V = O.0284eollJ<
Rz = 1
V = 0.0193e0123*
R1 = 0.768
^ 0.40
10 15 20
Temperature (°C)
25
30
Figure 15. Comparison between calculated and predicted trends of silicate silica
with temperature.
27
-------
6. Results
The major results from the sediment model include the spatial and temporal flux rates including
ammonium-nitrogen, nitrate-nitrogen, phosphate-phosphorus, silicate-silica, SOD, and chemical
oxygen demand (Table 1: state variables 21-26, respectively). Calibrated fluxes of SOD, NH4,
P04, and Si are presented in Figures 8-11. Other model results include the book-keeping of mass
concentrations and diagenesis fluxes of chemicals within the two sediment layers as shown in
Appendix B. The presented results are obtained after running the year 2009 for twenty seven
simulations during the calibration process, which readjusted the initial values of the various
pools of material specified in Table 3.
Phytoplankton particles represent the major contributor to particulate organic sediment which
replenishes the benthic sediments (Section 3.1). Three groups of phytoplankton (dinoflagellates,
diatoms, and greens) are modeled in the water column of Narragansett Bay during 2009
(Abdelrhman, 2016a). Figures 16A and 16B present time series of chlorophyll-a concentration of
the three phytoplankton groups in the water column at stations CP and GD, respectively
throughout the year. The north-south gradient of phytoplankton in the water column is evident
with more than an order of magnitude difference between concentrations at CP at GD. Results
from the benthic sediment model mimic the spatial and temporal behavior of phytoplankton with
a north-south gradient in POC concentration in the sediment (Fig. 16C). In general, the POC
concentration follows the temporal behavior of phytoplankton throughout the year, especially in
the northern part of the Bay (e.g., station CP).
Figure 17 presents samples of time series of benthic fluxes at the northern and southern stations
CP and GD, respectively. The following four general observations are clear. First, benthic fluxes
are minimal during the cold winter months and reach their peak values during the warm summer
months. Second, silicate-silica fluxes are the highest followed by ammonium-nitrogen and the
smallest was phosphate-phosphorus. This behavior is consistent with the observed relationships
presented in Figure 5. Silica fluxes were approximately two orders of magnitude higher than
nitrogen fluxes, and nitrogen fluxes were roughly one order of magnitude higher than
phosphorus fluxes. Second, during the summer, benthic fluxes of silica, nitrogen, and SOD show
gradients from higher values in the north (station CP) to lower values in the south (station GD).
Phosphate fluxes were very small and did not follow this north-south gradient. Fourth, the range
of variability in benthic fluxes is much higher in the northern part of the Bay (e.g., station CP)
than the southern part (e.g., station GD). The negative fluxes of SOD and nitrate indicate their
transport from the over-lying water column into the sediment.
Results presented in Appendix B illustrate the temporal behavior at all stations. The upper values
of the presented range relate to northern stations and its lower values represent the southern
stations. The presented graphs and their captions are self-explanatory. The quasi-steady
behaviors of chemical concentrations and their fluxes are demonstrated by the starting and
ending values for the year. Temporal correspondence between benthic results and chlorophyll-a
concentrations is clear. For example, the behavior of benthic silica concentration and flux shows
higher activity when diatoms are active during the period April-July (Figs. 11, 31-33). Similarly,
the deposited PON, POP, and POC show increase during the period when phytoplankton
concentration is high (Figs. 35-37).
28
-------
CP Chl-a
- Dinoflagellates
- Diatoms
Green algae
20
^ 18
E 16
c 14
0
•4= 12
ro
1 10
Date
(C)
Figure 16. Time series of chlorophyll a concentrations in the water column (A and B)
and POC in the bed sediment (C) at stations CP and GD in 2009.
29
-------
Date
Date
CP SOD GDSOD
0.012
CP COD GD COD
-0.50
0.008
"b
•jL 0.006
~ 0.004
_3
"¦ 0.002
-2.50
(E) (F)
Figure 17. Benthic fluxes of NH4, N03, P04, Si, SOD, and COD (A-F, respectively)
at stations CP and GD.
30
-------
7. Summary and Conclusion
The presented sediment model is fully coupled to the water quality model for the Bay
(Abdelrhman, 2016a). Particulate organic matter and particulate silica settle from the water
column to the bed sediment where some fractions of them are changed to soluble matter, by
bacterial action and dissolution, which then return back to the water column as benthic fluxes.
The main purpose of the benthic sediment model was to provide time varying benthic fluxes of
the major nutrients including ammonium, nitrate, phosphate, silica, and SOD, and chemical
oxygen demand throughout Narragansett Bay. In addition, the sediment model provided the
bookkeeping of the mass budgets of organic particulate fractions of carbon, nitrogen,
phosphorus, and biogenic silica in their labile, refractory, and inert pools. The amount of
sulfide and methane were also predicted by the model (Appendix B).
The sediment model was successfully calibrated at 12 buoy locations (Fig. 1) to qualitatively
reproduce reported flux values of SOD (from Fulweiler et al., 2010) and NH4, P04 and Si
(from Kremer and Nixon, 1978, Nixon et al., 1976). All reported fluxes were estimated as
functions of only sediment temperatures at the buoy locations in the Bay. This assumption
excluded other essential effects which contribute to the mass balance formulation (i.e., Equations
(5) and (6)) including: (1) the depositional flux of POM from the water column into the sediment
(e.g., Fig 16C), (2) the change in the vertical gradient of concentration between the water column
and the sediment and its impact on the diffusive fluxes between the sediment and the water
column, (3) the loss due to burial of particulates between the modeled surface sediment layer
(-10 cm) and the deeper sediment, and (4) the impact of external loading of POM from the
watershed. The shortcoming of this oversimplification was rectified by the benthic sediment
model as manifested in the calibration process and the presented results.
The results in Appendix B indicate that, in general, the Bay reflected a long-term quasi-steady
behavior of the various benthic materials where values at the end of a one-year simulation were
similar to their starting initial values at the beginning of that year. Inter-annual variability in such
behavior can take place due to, for example, wet and dry years which can impact phytoplankton
production and its contribution to POM (e.g., Fig. 16). Longer simulations (several years) may
be needed to study inter-annual benthic sediment fluxes as they reach their ultimate quasi-steady
behavior.
31
-------
References
Abdelrhman, M. A. 2015. Three-dimensional modeling of the hydrodynamics and transport in
Narragansett Bay. United States Environmental Protection Agency Report EPA/600/R-15/152,
ORD tracking number ORD-008162, pp. 171.
Abdelrhman, M. A. 2016a. Three-dimensional modeling of water quality and ecology in
Narragansett Bay. United States Environmental Protection Agency Report EPA/600/R-16/203,
ORD tracking number ORD-008354, pp. 125.
Abdelrhman, M. A. 2016b. Modeling total suspended solids (TSS) concentration in Narragansett
Bay. United States Environmental Protection Agency Report EPA/600/R-16/195, ORD tracking
number ORD-017732, pp. 76.
Brady, D. C., Testa, J. M., DiToro, D. M., Boynton. W. R., Kemp, W. M. 2013. Sediment flux
modeling: calibration and application for coastal systems. Estuarine, Coastal and Shelf Science
117:107-124.
Berounsky, V. M., Nixon, S. W. 1993. Rates of nitrification along an estuarine gradient in
Narragansett Bay. Estuaries 16(4) 718:730.
Cerco, C. F., Cole, T. 1994. Three-dimensional eutrophi cation model of Chesapeake Bay:
Volume 1, main report. Technical Report EL-94-4, US Army Engineer Waterways Experiment
Station, Vicksburg, MS.
DiToro, D. M. 2001. Sediment flux modeling. ISBN: 978-0-471-13535-7. J. Wily & Sons
Publishing. 656 p.
DiToro, D. M., J. J. Fitzpatrick. 1993. Chesapeake Bay sediment flux model. Contract Report
EL-93-2, US Army Engineer Waterways Experiment Station, Vicksburg, MS, 316pp.
Elderfield, H., Luedtke, N., McCaffrey, J., Bender, M. 1981. Benthic flux studies in Narragansett
Bay. American Journal of Science 281:768-787.
Fulweiler, R.W. and Heiss, E.M. 2014. (Nearly) a decade of directly measured sediment N2
fluxes: What can Narragansett Bay tell us about the global ocean nitrogen budget?
Oceanography 27(1): 184-195.
Fulweiler, R. W., S. W. Nixon. 2012. Net sediment N2 fluxes in a southern New England
estuary: variations in space and time. Biogeochemistry 111:111-124.
Fulweiler, R. W., S. W. Nixon, B. A. Buckley. 2010. Spatial and temporal variability of benthic
oxygen demand and nutrient regeneration in an anthropogenically impacted New England
Estuary. Estuaries and Coasts 33:1377-1390.
32
-------
Kremer, J. N., S. W. Nixon, 1978. A costal marine ecosystem simulation and analysis.
Ecological Studies 24, Springer-Verlag, NY.
Lee, V., D. Bonneau, S. Adamowicz, A. Beck, R. Greene, C. Deacutis, and C. Damon. 2000.
Narragansett Bay National Estuarine Research Reserve: Data rescue manual and CD-ROM. CRC
Coastal Management Report #2223. 18pp.
McMaster, R. L. 1960. Sediments of Narragansett Bay system and Rhode Island Sound, Rhode
Island. Journal of Sedimentary Petrology 30:249-274.
Nixon, S. W., Oviatt, C. A., Hale, S. S. 1976. Nitrogen regeneration and metabolism of coastal
marine bottom communities. In: Anderson, J. Macfayden (eds.). The role of terrestrial and
aquatic organisms in decomposition processes. London: Blackwell Sci. Pub. pp. 269-283.
Park, K., A. Y. Kuo, J. Shen, and J. M. Hamrick, 1995: A three-dimensional hydrodynamic
eutrophication model (HEM3D): description of water quality and sediment processes submodels.
The College of William and Mary, Virginia Institute of Marine Science. Special Report 327, 113.
Raposa, K. 2009. Ecological Geography of Narragansett Bay, Chapter 7 (page 84) in
Narragansett Bay National Estuarine Research Reserve. An Ecological Profile of the
Narragansett Bay National Estuarine Research Reserve. K.B. Raposa and M.L. Schwartz (eds.),
Rhode Island Sea Grant, Narragansett, R.I. 176pp.
Seitzinger, S.P., Nixon, S. W., Pilson, M. E. Q. 1984. Denitrification and nitrous-oxide
production in a coastal marine ecosystem. Limnol. Oceanogr. 29(l):73-83.
Testa, J. M., Brady, D. C., DiToro, D.M., Boynton, W. R., Cornwell, J. C., Kemp, W. M. 2013.
Sediment flux modeling: calibration and application for coastal systems. Estuarine, Coastal and
Shelf Science 117:107-124.
Tetra Tech 2005. A Hydrodynamic and Water Quality Model for the Lower Charles River Basin,
Massachusetts. Draft Report, prepared for US Environmental Protection Agency, Region 1.
Prepared by Tetra Tech, Inc. Eaton, Place, Fairfax, VA. 100pp.
33
-------
Appendix A: Values of sediment water quality parameters used for Narragansett Bay
Values are grouped in card-image format for the FORTRAN input file (WQ3DSD.INP) to the
EFDC model. Card images are separated by card number "C#". Values of parameters are
reported in Ditoro and Fitzpatrick (1993) and Cerco and Cole (1994) and presented by Park et al.
(1995). Calibrated parameter values are bolded. Uncalibrated parameters were assumed to be
similar to the Chesapeake Bay/Lower Charles River values.
Equation
Parameter
Card
Input
Parameter
Description and units
Value
C#2
iSMZ
Number of zones for spatially varying parameters
1
ilCI
initial conditions switch
2
iRST
write final spatial distributions to restart file WQSDRST.OUT
1
iHyst
activate hysteresis in benthic mixing
1
iZB
activate diagnostic output for FUNCTION ZBRENT
1
C#5
Dt
DifT
diffusion coefficient for sediment temperature (m2 s"1)
1.8xl0"7
C#6
FNBX1
FNBcl
fraction of PON from algae group 1 routed to G1 class
0.65
FNBc2
fraction of PON from algae group 1 routed to G2 class
0.3
FNBc3
fraction of PON from algae group 1 routed to G3 class
0.05
FNBdl
fraction of PON from algae group 2 routed to G1 class
0.65
FNBd2
fraction of PON from algae group 2 routed to G2 class
0.3
FNBd3
fraction of PON from algae group 2 routed to G3 class
0.05
FNBgl
fraction of PON from algae group 3 routed to G1 class
0.65
FNBg2
fraction of PON from algae group 3 routed to G2 class
0.3
FNBg3
fraction of PON from algae group 3 routed to G3 class
0.05
C#7
FPBX1
FPBcl
fraction of POP from algae group 1 routed to G1 class
0.65
FPBc2
fraction of POP from algae group 1 routed to G2 class
0.3
FPBc3
fraction of POP from algae group 1 routed to G3 class
0.05
FPBdl
fraction of POP from algae group 2 routed to G1 class
0.65
FPBd2
fraction of POP from algae group 2 routed to G2 class
0.3
FPBd3
fraction of POP from algae group 2 routed to G3 class
0.05
FPBgl
fraction of POP from algae group 3 routed to G1 class
0.65
FPBg2
fraction of POP from algae group 3 routed to G2 class
0.3
FPBg3
fraction of POP from algae group 3 routed to G3 class
0.05
C#8
FCBX>1
FCBcl
fraction of POC from algae group 1 routed to G1 class
0.65
FCBc2
fraction of POC from algae group 1 routed to G2 class
0.3
FCBc3
fraction of POC from algae group 1 routed to G3 class
0.05
FCBdl
fraction of POC from algae group 2 routed to G1 class
0.65
34
-------
Equation
Parameter
Card
Input
Parameter
Description and units
Value
FCBd2
fraction of POC from algae group 2 routed to G2 class
0.3
FCBd3
fraction of POC from algae group 2 routed to G3 class
0.05
FCBgl
fraction of POC from algae group 3 routed to G1 class
0.65
FCBg2
fraction of POC from algae group 3 routed to G2 class
0.3
FCBg3
fraction of POC from algae group 3 routed to G3 class
0.05
C#9
Diagenesis
KpoMi
KPON1
Decay rate of PON at 20 °C in Layer 2 for G1 class (d1)
0.055
KPON2
Decay rate of PON at 20 °C in Layer 2 for G2 class (d1)
0.0027
KPON3
Decay rate of PON at 20 °C in Layer 2 for G3 class (d1)
0.0
KPOP1
Decay rate of POP at 20 °C in Layer 2 for G1 class (d1)
0.055
KPOP2
Decay rate of POP at 20 °C in Layer 2 for G2 class (d1)
0.0027
KPOP3
Decay rate of POP at 20 °C in Layer 2 for G3 class (d1)
0.0
KPOC1
Decay rate of POC at 20 °C in Layer 2 for G1 class (d1)
0.055
KPOC2
Decay rate of POC at 20 °C in Layer 2 for G2 class (d1)
0.0027
KPOC3
Decay rate of POC at 20 °C in Layer 2 for G3 class (d1)
0.0
C#10
0POM.1
ThKNl
Constant for temperature adjustment for KPON1 (unitless)
1.1
ThKN2
Constant for temperature adjustment for KPON2 (unitless)
1.15
ThKN3
Constant for temperature adjustment for KPON3 (unitless)
1.0
ThKPl
Constant for temperature adjustment for KPOP1 (unitless)
1.167
ThKP2
Constant for temperature adjustment for KPOP2 (unitless)
1.167
ThKP3
Constant for temperature adjustment for KPOP3 (unitless)
1.0
ThKCl
Constant for temperature adjustment for KPOC1 (unitless)
1.1
ThKC2
Constant for temperature adjustment for KPOC2 (unitless)
1.15
ThKC3
Constant for temperature adjustment for KPOC3 (unitless)
1.0
C#ll
mi
rMl
Solid concentrations in Layer 1 (Kg L1)
0.5
m2
rM2
Solid concentrations in Layer 2 (Kg L1)
0.5
0Dd
ThDd
Constant for temperature adjustment for Dd (unitless)
1.08
0Dp
ThDp
Constant for temperature adjustment for Dp (unitless)
1.117
Gpoc.r
GPOCr
Reference concentration for GPOC(l) (gC m 3)
50.0
KMdp
KMDp
Particle mixing half-saturation constant for oxygen (mg L1)
2.0
Kst
KST
First-order decay rate for accumulated benthic stress (d1)
0.03
Dpmin
DpMIN
Minimum diffusion coefficient for particle mixing (m2 d"1)
3xl0"6
Rbi.bt
RBIBT
Ratio of bio-irrigation to bioturbation (unitless)
1.0
C#12
02BSc
Critical overlying oxygen concentration below which benthic
1.0
hysteresis occurs (mg L"1)
TDMBS
Time duration for which the maximum or minimum stress is
60.0
retained (d)
35
-------
Equation
Parameter
Card
Input
Parameter
Description and units
Value
TCMBS
Critical hypoxia duration; if less than this value, no
14.0
hysteresis occurs (d)
C#13
7Il,NH4
P1NH4
Partition coefficient, ratio of particulate to dissolved NH4
1.0
in layer 1 (L Kg"1)
712,NH4
P2NH4
Partition coefficient, ratio of particulate to dissolved NH4
1.0
in layer 2 (L Kg"1)
KMNH4
KMNH4
Nitrification half-sat. constant for ammonium (gN m3)
1.2
KMnH4,02
KMNH40
2
Nitrification half-sat. constant for dissolved oxygen (g02 m"3)
2.0
0NH4
ThNH4
Constant for temperature adjustment for KNH4 (unitless)
1.17
0NO3
ThN03
Constant for temperature adjustment for KN031 and KN032
(unitless)
1.2
712,P04
P2P04
Partition coefficient, ratio of particulate to dissolved P04
10.0
in layer 2 (L Kg"1)
(DOo)crit,PO
4
D0cP04
Critical dissolved oxygen for P04 sorption (mg L"1)
3.0
C#14
7Il,H2S
P1H2S
Partition coefficient for H2S in Layer 1 (L Kg"1)
50.0
712,H2S
P2H2S
Partition coefficient for H2S in Layer 2 (L Kg"1)
50.0
kH2S,dl
KH2Sdl
Reaction velocity for dissolved sulfide oxidation in
0.2
Layer 1 at 20 °C (m d"1)
kms.pi
KH2Spl
Reaction velocity for particulate sulfide oxidation in
0.4
Layer 1 at 20 °C (m d"1)
0H2S
ThH2S
Constant for temperature adjustment for KH2Sdl and KH2Spl
(unitless)
1.08
KMh2S,02
KMH2S
Constant to normalize the sulfide oxidation rate for oxygen
(mg02 L"1)
2.0
kcH4
KCH4
Reaction velocity for methane oxidation in layer 1 at 20 °C (m
d"1)
0.2
0CH4
ThCH4
Constant for temperature adjustment for KCH4 (unitless)
1.08
cSHSCH
Critical salinity; less than this value CH4 is produced
1.0
C#15
ao2,c
a02C
Stoichiometric coefficient for carbon diagenesis consumed
2.6667
by H2S oxidation (g02 per gC)
ao2,N03
a02N03
Stoichiometric coefficient for carbon diagenesis consumed
2.85714
by denitritrification (g02 per gN)
ao2,NH4
a02NH4
Stoichiometric coefficient for carbon diagenesis consumed
4.5714
by nitrification (g02 per gN)
C#16
Ksi
KSi
First order dissolution rate for particulate biogenic silica
0.2
(PSi) at 20 °C in layer 2 (d1)
36
-------
Equation
Parameter
Card
Input
Parameter
Description and units
Value
0Si
ThSi
Constant for temperature adjustment of KSi (unitless)
1.1
KMPSl
KMPSi
Silica dissolution half-saturation constant for PSi (g Si m"3)
20.0
Slsat
SiSat
Saturation concentration of silica in pore water (g Si m~3)
44
712, Si
P2Si
Partition coefficient for Si in Layer 2, controls sorption
15.0
of dissolved silica to solids (L Kg1)
A7Il,Si
DPI Si
factor that enhances sorption of silica in layer 1 when D.O.
5.0
exceeds DOcSi (unitless)
DOcSi
Critical dissolved oxygen for silica sorption in layer 1 (mg L1)
2.0
JdSi
DetFPSi
Detrital flux of particulate biogenic silica from sources
0.1
other than diatom algae (gSi m~2 d"1)
cm
CPON1
Cone. Particulate Org. Nitrogen in G-class 1 (g m~3)
60.0
CPON2
Cone. Particulate Org. Nitrogen in G-class 2 (g m~3)
200.0
CPON3
Cone. Particulate Org. Nitrogen in G-class 3 (g m 3)
400.0
CPOP1
Cone. Particulate Org. Phosphorus in G-class 1 (g m~3)
60.0
CPOP2
Cone. Particulate Org. Phosphorus in G-class 2 (g m~3)
200.0
CPOP3
Cone. Particulate Org. Phosphorus in G-class 3 (g m~3)
400.0
CPOC1
Cone. Particulate Org. Carbon in G-class 1 (g m~3)
100.0
CPOC2
Cone. Particulate Org. Carbon in G-class 2 (g m~3)
1000.0
CPOC3
Cone. Particulate Org. Carbon in G-class 3 (g m~3)
2000.0
C#18
C1NH4
Cone. NH4-N in layer 1 (g m"3)
4.0
C2NH4
Cone. NH4-N in layer 2 (g m~3)
4.0
C2N03
Cone. N03-N in layer 2 (g m"3)
1.0
C2P04
Cone. P04-P in layer 2 (g m~3)
200.0
C2H2S
Cone. Sulfide (H2S) in layer 2 (g m"3)
30.0
CPSi
Cone. Particulate biogenic silica in layer 2 (g m~3)
500.0
C2Si
Cone. Dissolved available silica in layer 2 (g m~3)
500.0
ST
CBSt
Initial accumulated benthic stress (days)
0.0
T
GT
Initial sediment temperature (°C)
3.0
C#19
ISMz
zone for spatially variable parameters in SPM
1.0
h2
Hsed
Total active sediment thickness (m)
0.1
W
W2
sediment burial rate (cm year"1)
0.02
Dd
Dd
diffusion coefficient in pore water (m2 d"1)
0.6
Dp
Dp
apparent diffusion coefficient for particle mixing (m2 d"1)
0.001
kNH4
KNH4
optimal reaction velocity for nitrification at 20 °C (m d"1)
0.131
Kn03,1
KN031
reaction velocity for denitrification in layer 1 at 20 °C (m d"1)
0.1
Kn03,2
KN032
reaction velocity for denitrification in layer 2 at 20 °C (m d"1)
0.18
37
-------
Equation
Parameter
Card
Input
Parameter
Description and units
Value
A7I1.P04
DP1P04
factor to enhance sorption of P04 in layer 1 when DO >
D0cP04
100.0
SODmult
factor to enhance magnitude of sediment oxygen demand
(unitless)
2.5
C#20
ISMZ
zone index for spatially variable parameters
1
FNRPi
FNRPI
fraction of water column refractory PON routed to G-class 1
0.2
FNRP2
fraction of water column refractory PON routed to G-class 2
0.7
FNRP3
fraction of water column refractory PON routed to G-class 3
0.1
FPRPi
FPRPI
fraction of water column refractory POP routed to G-class 1
0.2
FPRP2
fraction of water column refractory POP routed to G-class 2
0.7
FPRP3
fraction of water column refractory POP routed to G-class 3
0.1
FPRPi
FCRP1
fraction of water column refractory POC routed to G-class 1
0.1
FCRP2
fraction of water column refractory POC routed to G-class 2
0.8
FCRP3
fraction of water column refractory POC routed to G-class 3
0.1
38
-------
Appendix B: Time series graphs of sediment water quality parameters and fluxes for
Narragansett Bay
The sediment water quality model outputs time series of many parameters and fluxes in two
output files WQSDTSl.OUT and WQSDTSl.OUT. The parameter description and its output
name are listed in the following table. Time series of the concentrations in the sediment and the
fluxes from the sediment to the overlying water column are presented in this appendix together
with other important parameters. At every time instance, the shown vertical range represents
values at all stations where the higher values relate to the northern stations and lower values
represent the southern stations.
Printout
Parameter in sediment in WQSDTS1 Name
Ammonium Nitrogen NH4 in Layer 1 NH41
Ammonium Nitrogen NH4 in Layer 2 NH42
Flux of Ammonium Nitrogen NH4 FNH4
Nitrate Nitrogen N03 in Layer 1 N031
Nitrate Nitrogen N03 in Layer 2 N032
Flux of Nitrate Nitrogen N03 FN03
Phosphate Phosphorus P04 in Layer 1 P041
Phosphate Phosphorus P04 in Layer 2 P042
Flux of Phosphate Phosphorus P04 FP04D
Sulfide H2S in Layer 1 H2S1
Sulfide H2S in Layer 2 H2S2
Benthic Flux of Oxygen 02 BF02
Benthic Flux of Chemical Oxygen Demand COD BFCOD
Silica SI in Layer 1 SI1
Silica SI in Layer 2 SI2
Flux of Silica FSAD
Sediment Model Temperature SMT
Benthic Stress BST
Particulate Organic Nitrogen PON
Particulate Organic Phosphorous POP
Particulate Organic Carbon POC
Printout
Parameter in sediment in WQSDTS2 Name
Flux of carbonaceous Sediment Oxygen Demand CSOD
Flux of nitrogenous Sediment Oxygen Demand NSOD
Enhanced P04 flux in layer 1 (D0>D0CP04) D1P04
Enhanced SI flux layer 1 (DO>DOCSI) D1SI
Surface mass transfer coefficient SS
Nitrification Flux JNIT
39
-------
Denitrification Flux (gaseous)
Flux of Aqueous Sulfide H2S
Flux of Gaseous Methane CH4
Diagenesis Flux of N
Diagenesis Flux of P
Diagenesis Flux of C
Depositional Flux of N to class
Depositional Flux of N to class
Depositional Flux of N to class
Depositional Flux of P to class
Depositional Flux of P to class
Depositional Flux of P to class
Depositional Flux of C to class
Depositional Flux of C to class
Depositional Flux of C to class
JDEN
JAQH2S
JGCH4
DGFN
DGFP
DGFC
G1
DFN1
G2
DFN2
G3
DFN3
G1
DFP1
G2
DFP2
G3
DFP3
G1
DFC1
G2
DFC2
G3
DFC3
40
-------
NH41
1.80
1.60
1.40
1.20
1.00
0.80
0.60
0.40
0.20
0.00
oo
o
oo
rH
i£>
rH
rsl
00
rH
m
r-
rH
r-
rH
W]
<£>
rH rH
<£> r--
Date
u->
rH
00
rH
G)
cn
cn
cn
O
o
o
\
O
rH
m
m
rsl
rH
rH
¦>
rH
rH
O
rH
rsl
rH
rH
rH
rH
Figure 18. Time series of predicted ammonium concentration in Layerl at all stations.
NH42
1.80
1.60
1.40
1.20
1.00
0.80
0.60
0.40
0.20
0.00
00
O
00
rH
fsT
O
O
i£>
rH
rsl
O
00
rH
m
CTi
O
r-
rH
CTi
O
r-
rH
U-)
CTi
O
(J)
O
(£> l£>
rH rH
<£> r--
Date
rH
rH
O
rH
rsl
rH
rH
rH
rH
Figure 19. Time series of predicted ammonium concentration in Layer2 at all stations.
41
-------
Predicted FNH4
0.18
0.16
0.14
ST 0.12
0.10
3B 0.08
| 0-06
0.04
0.02
0.00
.iljHJ
bt
J 4
it
AH
mi,,
00 (J)
o o
O
O
o
O
en
o
cn
o
O
rH
ro
rH
ro
rH
rH
rvT
rH
rH
Figure 20. Time series of predicted ammonium flux from benthic sediment at all stations.
N031
0.12
0.10
E
00 0.08
c
o
¦E 0,06
c
C
o
u
0.04
0.02
0.00
0^
O
O
O
O
rH
m
m
rH
rH
O
rH
-------
N032
00
o
O
O
G\
O
rH
rn*
rH
ro"
rH
r\l
rH
fN
T—1
T—1
(N
ro
to
r-*
00
aT
O*
rH
>
rH
fN
rH
rH
Date
Figure 22. Time series of predicted nitrate concentration in Layer2 at all stations.
FN03
Figure 23. Time series of predicted nitrate flux from benthic sediment at all stations.
43
-------
P041
4.00
3.50
3.00
2.50
2.00
1.50
o 1.00
0.50
0.00
00
o
O
O
G\
O
Date
Lfl
T—I
CO
*3"
rH
O
O
G\
O
Date
tn
T—I
00
rH
-------
Predicted FP04D
0.02
0.02
0.02
0.01
¦b
CN
0.01
£
0.01
M
X
0.01
3
Ll_
0.01
0.00
0.00
0.00
00
O
00*
r—I
fN
Mm 11' yt
Ol
o
o
to
rH
UD
rH
LT)
rH
<£>
r-
CO
Date
Figure 26. Time series of predicted phosphate flux from benthic sediment at all stations.
7.00
6.00
£ 5.00
Bp
J 4.00
to
i: 3.00
c
oo r- iDiDLort^rfTrofN
tH rH rH rH rH rH rH rH rH rH t—I tH rH rH
'S'^H(NriritLOtDr~-oo(TiO''iHr5'^H
rH rH rH rH
Date
Figure 27. Time series of predicted sulfide concentration in Layerl at all stations.
45
-------
H2S2
8.00
7.00
6.00
5.00
4.00
3.00
o 2.00
1.00
0.00
J
1
~
II
k\
lit
l|
¦
fcl
J
1
J
-md
MMta
Mm
COO^O^O^O^O^O^O^O^O^iO^O^O^O
OOOOOOOOOOOOOtH
KD 00
KD t£> LT
ro m
-------
BFCOD
0.25
E
ao
0.20
0.15
0.10
0.05
0.00
-0.05
COO^O^O^O^O^O^O^O^O^iO^O^O^
o o
o
00*
o
r*-
o o
00
o
r-
o
r-
o o o
\
VD Lf)
o
O
m ro (N
lti to r^-
Date
Figure 30. Time series of predicted COD flux from benthic sediment at all stations.
600.00
__ 500.00
m""
£
oo 400.00
c
'¦£ 300.00
c
o
<->
100.00
0.00
Sll
oo en
o o
0s) ID
T—I T—I
r--
Date
Figure 31. Time series of predicted silica concentration in Layerl at all stations.
47
-------
SI2
250.00
0.00
00
o
O
O
G\
O
t-H
rn*
T—1
ro"
T—1
(N
T—1
fN
T—I
T—1
(N
ro
to
t£»
r-*
00
aT
o*
T—1
>
T—1
fN
T—1
T—I
Date
Figure 32. Time series of predicted silica concentration in Layer2 at all stations.
Predicted FSAD
0.90
0.80
0.70
5" 0-60
£ 0.50
3B 0.40
I 0-30
0.20
0.10
0.00
H
ii 111^" I
OOOOOOOOOOOOO*—I
oor^uDOor^r^uDuDi-n'^-rtrnmrvj
t—It—It—It—It—It—It—It—It—ItHtHt—It—It—I
rvirHrNiro^LntDr^ooaiOrHrsiTH
Date
Figure 33. Time series of predicted silica flux from benthic sediment at all stations.
48
-------
Sediment Temperature
Figure 34. Time series of predicted temperature of benthic sediment at all stations.
49
-------
PON
200
180
160
140
120
100
80.
60.
40.
20.
0
.00
.00
.00
.00
.00
.00
00
00
00
00
00
00
o
O
O
G\
O
rH
rn*
*—i
ro*
T—1
(N
T—1
fN
rH1
T—1
(N
ro
to
Date
r-
00
aT
O*
T—1
>
rH
fN
rH
rH
Figure 35. Time series of predicted PON concentration in Layer2 at all stations.
POP
30.00
25.00
fn~"
£
20.00
c
¦¦5 15.00
c
O
O
G\
O
rH
rn*
rH
ro*
rH
r\l
rH
fN
rH1
rH
C\l
ro
to
Date
r-
00
aT
O*
rH
>
rH
fN
rH
rH
Figure 36. Time series of predicted POP concentration in Layer2 at all stations.
50
-------
POC
1,400.00
1,200.00
fn~"
£ 1,000.00
M
g 800.00
£ 600.00
c
co LT)
ro ro (N
-------
NSOD
0000
9000
8000
7000
0.6000
c 0.
ftD
X o.
ZJ
LI- 0.
0.
0.
0.
5000
4000
3000
2000
.1000
0000
00
o
O
O
O
rH
00
rH
rH
r-
rH
UD
rH
UD
rH
LO
rH
rH
>
rH
ro*
rH
fN
rH1
rH
(N
ro
LO
t£>
Date
r-
00
cT
rH
>
rH
01
o
rff
o
t-H
-
(N
Figure 39. Time series of predicted nitrogenous SOD flux from benthic sediment at all stations.
0.1200
o.iooo
OJ
E
aj
u
c
ra
0.0800
0.0600
0.0400
0.0200
0.0000
D1P04
00
r-.
Date
LO
T—I
00
rH
G)
rH
-¦V.
rH
O
rH
rsl
rH
rH
rH
rH
Figure 40. Time series of predicted enhanced phosphate flux in Layerl at all stations.
52
-------
D1SI
16.0000
14.0000
12.0000
+->
| 10.0000
u 8.0000
c
-f 6.0000
UJ
4,0000
2.0000
0.0000
COO^O^O^O^O^O^O^O^O^iO^O^O^O
OOOOOOOOOOOOO*—I
<£> 00
ID t£> LT
ro m rvj
r*- oo
-------
J NIT
0.0016
0.0014
0.0012
T3 0.0010
(N
£ 0.0008
00
0.0004
0.0002
0.0000
oo
o
-------
JAQH2S
"O
fN
£
0.2500
0.2000
0.1500
0.1000
x
- 0.0500
0.0000
-0.0500
00
o
G\
O
O
G\
O
rH
00
rH
r*-
rH
r-
rH
UD
rH
UD
rH
Ln
rH
rH
>
rH
ro*
rH
fN
rH1
T—1
(N
ro
LO
Date
r-
00
oT
O*
rH
>
rH
O
Figure 45. Time series of predicted aqueous sulfide flux from benthic sediment at all stations.
l.i
0.!
o.:
;r- o
¦b
0000
9000
8000
7000
0.6000
c 0.
no
x 0.
H 0.
0.
0.
0.
5000
4000
3000
2000
1000
0000
oo
o
JGCH4
i--
rH
rH
*-v.
rH
O
rH
rsl
rH
rH
rH
rH
Date
Figure 46. Time series of predicted gaseous methane flux from benthic sediment at all stations.
55
-------
DGFN
0.1400
0.1200
0.1000
0.0800
0.0600
0.0400
0.0200
0.0000
COO^O^O^O^O^O^O^O^O^iO^O^O^O
OOOOOOOOOOOOO*—I
<£> 00
id t£> in
ro m rvj
-------
0.9000
0.8000
0.7000
£" 0.6000
0.5000
25 0.4000
X
0.3000
0.2000
0.1000
0.0000
DGFC
¦Ml
ifcL*
OOOOOOOOOOOOO*—I
<£> 00
VD WD U")
ro m rvj
-------
DFN2
0.0400
0.0350
0.0300
5T 0.0250
0.0200
25 0.0150
X
0.0100
0.0050
0.0000
-0.0050
I
COO^O^O^O^O^O^O^O^O^iO^O^O^O
ooooooooooooo*—I
cor^uDOor^r^uDuDLn'^^mmrvj
t—It—It—It—It—It—It—It—It—ItHtHt—It—It—I
rvJrHrNiro^LntDr^ooaiOrHrsiTH
Date
Figure 51. Time series of predicted nitrogen depositional flux to class G2 at all stations.
DFN3
0.0060
0.0050
— 0.0040
¦b
1 0.0030
~ 0.0020
0.0010
0.0000
-0.0010
OOC^C^C^C^C^C^C^C^C^O^Q^Q^O
OOOOOOOOOOOOOt—I
oor^uDOor^r^uDuDLn^^rnrnrNi
t—It—It—It—It—It—It—It—It—ItHt—It—It—It—I
rvTr-ICNm^-iriOt--C001O"r:rrvrr-l
Date
Figure 52. Time series of predicted nitrogen depositional flux to class G3 at all stations.
58
-------
DFP1
0090
0080
0070
0060
.0050
0040
0030
0020
.0010
0000
0010
00
o
G\
O
O
G\
O
rH
00
rH
r*-
rH
r-
rH
UD
rH
UD
rH
Ln
rH
rH
>
rH
ro*
rH
fN
rH1
T—1
(N
rn
LO
Date
r-
00
oT
O*
rH
>
rH
01
o
Figure 53. Time series of predicted phosphate depositional flux to class G1 at all stations.
o.
o.
o.
? 0
I 0
3B o.
X
= 0
0.
0.
-0.
0040
0035
0030
0025
0020
.0015
.0010
0005
0000
0005
DFP2
00
rH
*-v.
rH
O
rH
rsl
rH
rH
rH
rH
Date
Figure 54. Time series of predicted phosphate depositional flux to class G2 at all stations.
59
-------
DFP3
E
ao
0007
0006
0005
0004
0003
0002
0001
0000
0001
IL.
COO^O^O^O^O^O^O^O^O^iO^O^O^O
OOOOOOOOOOOOO*—I
<£> 00
id
-------
DFC2
0.2500
0.2000
0.1500
0.1000
- 0.0500
0.0000
-0.0500
in 1^1 I—I—wm+mrnamt
COO^O^O^O^O^O^O^O^O^iO^O^O^O
OOOOOOOOOOOOO*—I
<£> 00
id t£> in
ro m rvj
------- |