Liix.«.u States Environmental Protection Agency
CBP/TRS 15
November 1987
The Effect of pH on the
Release of Phosphorus from
Potomac River Sediment
MILChesapeake
Bay
Program
cm /11 -9-87/4202/epa report /en g/epacover B
-------
'D_
, rd-
53 i
U.S. E.P.A. Region HI
Information Resource Center
The Effect of pH on the
Release of Phosphorus from
Potomac River Sediments
Final Project Report
Sybil P. Seitzinger, Ph.D.
Report No. 86-8F
Division of Environmental Research
Academy of Natural Sciences of Philadelphia
19th and the Parkway Philadelphia, PA 19103
-28 October 1986
-------
ACKNOWLEDGEMENTS
Special thanks goes to Maryann Bucknavage who was respon-
sible for the development of the laboratory procedures for
maintaining constant pH, and who carried out all laboratory
measurements of nutrient fluxes as a function of pH. Clare E.
Casselberry made the denitrification measurements. Both con-
tributed to the preparation of this report. Editorial assis-
tance was provided by Robin L. Soltis. SCUBA diving support
was provided by Bill Yates, Kevin Braun, and Jamie Barnhard.
-------
TABLE OF CONTENTS
Page
ABSTRACT i
ACKNOWLEDGEMENTS 1
INTRODUCTION 2
METHODS 6
Sediment Collection 6
Methodology Development 6
RESULTS AND DISCUSSION 16
Phosphate Fluxes as a Function of pH 16
Other Factors Influencing Phosphate Fluxes 26
Nitrogen Fluxes as a Function of pH 29
Rate of pH Change ; 40
CONCLUSIONS 46
LITERATURE CITED 48
-------
ABSTRACT
Recurring algal blooms in the tidal freshwater portion of
the Potomac River indicate an association between chlorophyll
a, high phosphorus concentrations, and high pH of the water.
In the present study, the release of phosphorus from sediments,
as a function of overlying water pH was measured at eight loca-
tions. Phosphate release under aerobic conditions was found to
increase as a function of overlying water pH. Between pH 8 and
9 the sediment-water phosphate flux was low. Beginning with an
overlying water pH in the range of 9.0 to 9.5, the phosphate
flux increased. The increased release of phosphate at high pH
is likely due to solubilization from iron and aluminum phos-
phate complexes. The rate of release of phosphorus from the
sediments with an overlying water pH of 10 was similar to the
amount of phosphorus necessary to account for the excess phos-
phorus in the bloom area. The release of ammonia from Potomac
River sediments increased with increasing pH; the uptake of
nitrate by the sediments generally increased with increasing
pH. Denitrification appears to be an important sink for nitro-
gen in this portion of the Potomac River, and is removing an
amount of nitrogen equivalent to approximately 35% of the
nitrogen loading to this portion of the river.
-------
INTRODUCTION
For over two centuries the Potomac Estuary experienced
increasing eutrophication due primarily to discharges of domes-
tic wastewater from the metropolitan Washington, D.C. area.
During the past two decades more than one billion dollars has
been spent on remedial programs including upgrading the sewage
treatment facilities. This has markedly decreased phosphorus
inputs to the Potomac Estuary (Table 1) and greatly increased
water quality. However, extensive blue-green algal blooms are
still occurring, as observed during the summer in 1983 and 1985
and to a lesser extent in 1984. In order to protect the future
use of the Potomac Estuary as an aquatic recreational resource
and as a wildlife refuge and park area, the causes of these
recurring blooms must be identified and thoroughly understood
so that further appropriate remedial actions can be taken.
This research is an extension of pilot studies carried out
over the past two years, each of which strongly suggest that
the release of phosphorus from the bottom sediments is the
primary factor responsible for the recurrent algal blooms in
the Potomac Estuary. An extensive analysis of water column
data from the summer of 1983 by the Algal Bloom Expert Panel
(Thomann et al. 1985) shows that the 1983 bloom was centered
between river miles 20 and 40 (Fig. la). Total phosphorus (P)
concentrations in that area were also shown to be elevated
above levels outside the bloom area (Fig. lb). When these
elevated levels of phosphorus are used in the Potomac Eutrophi-
cation Model (PEM), chlorophyll levels similar to those found
during the 1983 bloom intensification (greater than 75 jjg chl
a/1) are generated. Preliminary studies (Seitzinger 1983,
1985) indicate that benthic sediments were the source of the
excess phosphorus observed in the water column during the
bloom. Generally, the release of phosphorus from the sediments
in this area is quite low (<10 pmol P m~2 h"1; Callendar and
Hammond 1982, Seitzinger 1983), and not of sufficient magnitude
to account for the excess phosphorus that accumulated in the
2
-------
POTOMAC RIVER - CHLa
AUGUST 19U
240 -
220 -
200 -
V
a
40 -
to -
40
40
20
0
RIVKH MttlS PflOM CHAIN 1*1001
POTOMAC RIVER - TOTAL P
AUGUST 1—3
0.U-
\
f 0.39 -
«
a
C 0.2 -
0.00 -
0
20
•o
*a
mvin wv.es from chain BMioac
POTOMAC RIVER - pH
AUGUST 1883
1
• -
0
20
40
•0
RIVER MILES FROM CHAIN SRIOGC
Figure 1. (a) Chlorophyll a,(b) tocal P and (c) pH in the Potomac Estuary
during August 1983. Data from Council of Governments (1984).
-------
Table 1. Wastewater loading trends (after treatment) in the
Potomac Estuary (From Thomann et al. 1985) .
Year
Waste Flow
(MGD)
5-day BOD
(lb/day)
Total Nitrogen
(lb/day)
Total Phosphorus
(lb/day)
1913
42
57,900
6,400
1,100
1932
75
102,900
11,500
2,200
1944
167
140,700
22,900
4,400
1954
195
199,600
31,700
5,500
1960
222
109,700
37,000
9,900
1970
370
140,700
59,900
24,000
1980
449
55,000
54,900
4,100
1983
446
18,400
58,400
2,100
3
-------
water column during the bloom. However, the pH of the river
water in the vicinity of the 1983 bloom was higher than usual
(pH 7 to 8), reaching 10.5 (Figure lc). Because the water
column is fairly well mixed, high pH was found in both the
surface and bottom water. Preliminary measurements of
phosphorus release rates from Gunston Cove sediments in
September 1984 showed that when the pH of the water over sedi-
ment cores was increased from 8 to 10, the release of phos-
phorus from the sediments increased by over an order of magni-
tude .
The present study was undertaken to examine in more detail
the relationship between pH and phosphorus release rates, pH
and NH4 and NOvj fluxes, and the spatial heterogeneity of the
sediment response from the Woodrow Wilson Bridge to Smith
Point. This report contains the results of the Potomac sedi-
ment-water nutrient flux studies funded by the EPA Chesapeake
Bay Program (all data exclusive of Woodrow Wilson Bridge) and
Metropolitan Washington Council of Governments (Woodrow Wilson
Bridge).
5
-------
METHODS
Sediment Collection
Sediment cores (approximately 7 - cm diameter and 15 - cm
deep) were collected by SCUBA-equipped divers using plastic
coring tubes at each of the locations in Table 2. Care was
taken during coring to avoid disturbance of the sediment sur-
face and the loss of flocculent material. At the time of
sediment collection, water was collected from each station in
acid-washed carboys. During transport to the laboratory, the
cores and water were kept dark and maintained at near ambient
river water temperature. The water over the cores was aerated.
Once in the laboratory, the cores were maintained in the dark
with water over each core continuously aerated and mixed by a
gentle stream of air. The water over the cores was changed
every 24 to 48 h with filtered (Gelman glass fiber, type A-E)
water from the site of sediment collection. While the ambient
river water temperature during the period of the studies ranged
from 7°C to 30°C, all experimental measurements in the labora-
tory were carried out at 22 ± 2°C to facilitate comparison of
results. Descriptions of the cores from each station are
included in Table 2.
Methodology Development
During preliminary pH - phosphorus release rate studies in
1984 (Seitzinger 1985), we found that maintaining a stable pH
greater than 8 of the water over sediments required the addi-
tion of base every 2 to 3 h, which made the measurements very
time-consuming. To decrease the labor involved in the present
studies, a partially automated method was developed to maintain
the pH at the desired levels. An hourly timer was set so that
for approximately 7 min/h a Technicon peristaltic pump added
base to each core. (Cores kept at pH 8 required no base addi-
tion). Various flow rate tubing (0.1-0.32 ml/min) and NaOH
6
-------
Table 2. Description of sediment cores collected from the Potomac River in
1985.
Date
Location Collected
Gunston Cove Sept. 9, 1985
Mainstem near Sept 30, 1985
Gunston Cove
Hallowing Point Sept. 30, 1985
W. Wilson Bridge Oct. 28, 1985
Comments
- Top layer (1-2 mm) of medium brown
sediment followed by a second layer
(2-3 mm) of light brown sediment.
Remainder of core was dark brown.
Sediment consisted of very fine par-
ticles. No animals visible.
- Thin layer (1-2 mm) light brown
sediment. Remainder of core medium
brown. Sediment consisted of very
fine particles. Clams present in
several cores.
Similiar to mainstem near Gunston
Cove cores.
- Top 3-4 mm had a medium brown color
and extremely fine sediment. It had
a flocky texture and large craters.
Sediment looked disturbed. Remainder
of core had brownish-greyish sedi-
ment and fine texture. Clams pre-
sent in all cores.
Mainstem near
Broad Creek
Oct. 28, 1985
Top 5 cm medium brown color and very
sandy. No distinct layers. A great
deal of debris (twigs, wood, and
stones) on top and mixed in with the
top sediment. Bottom sediment -
particles were smaller and brownish
- greyish in color. Clams present
in all cores.
Mattawoman Creek Dec. 4, 1985
No definite layers. Sediment light
to medium brown color. Fine sedi-
ment particles on surface, particles
became larger with depth. Many orga-
nisms on surface.
Indian Head
Dec. 4, 1985
No definite layers. Top 2 cm medium
brown color, increasing number of
black areas with depth. Sediment
particles were extremely fine.
Fragments of dead macrophytes on
surface.
Smith Point
Dec. 4, 1985
No definite layers. Top 3-4 cm
medium brown color. Areas of black
sediment mixed into remainder of
core. Particles were extremely
fine. Clams and other organisms
(similar to Mattawoman) present.
-------
concentrations (0.005-0.05 M) were used to maintain the differ-
ent pH levels. The base was added directly above the aeration
bubbles to ensure proper mixing. pH was monitored regularly
with a Fisher Accumet pH meter model 825 MP. The meter was
calibrated before each measurement with standard buffer solu-
tion. When necessary, manual adjustments in pH were made with
0.1 M HC1 or 0.05 M NaOH. The average pH and pH range of the
overlying water for each core during incubations are shown in
Table 3.
Detailed Relationship Between pH and Phosphorus Release Rates
Eight sediment cores were collected from each of four
locations: (1) Gunston Cove, (2) in the mainstem of the Poto-
mac River near Gunston Cove, (3) in the mainstem of the river
just upstream of Broad Creek and (4) near the Woodrow Wilson
Bridge (Fig. 2). All cores from a given location were col-
lected at one time. For each set of cores the sediment-water
exchange of soluble reactive phosphorus (hereafter referred to
as phosphate) was initially measured on all eight cores with
the pH of the overlying water at ~pH 8. (The procedure for
nutrient flux measurements is described below). Six of these
cores were chosen for pH experiments. Two of the selected
cores were then maintained at each of the following pH's for 5
days: 8, 9, and 10. At the end of the 5 days, the sediment-
water fluxes of phosphate, nitrite plus nitrate, and ammonium
were measured on the six cores. The water over one core from
each of the three pH treatments was then increased 0.5 pH units
(i.e. to either 8.5, 9.5, or 10.5). The other cores were kept
at pH 8, 9 or 10. Five days later the phosphate fluxes were
again measured on all six cores.
Spatial Variability in Phosphate Release Rates at High pH
Intersite variability in phosphate release rates from
sediments incubated with high pH overlying water was examined
8
-------
Table 3. Average (1 standard deviation) and range, during 24-h flux measurements, of pli of water overlying
sediment cores taken from the Potomac River.
£L of Water Column
Location
Core //
Target
PH
Flux
; 1
Target
pll
Flux
2
x ± S.D.
range
x 1 S.D.
range
Gunston Cove
1
8.0
8.15+0.05
8.09-8.19
8.0
8.14+0.06
8.08-0.20
3
8.0
8.17±0.08
8.08-8.22
8.5
8.4610.06
8.40-8.50
5
9.0
9.09+0.09
8.94-9.16
9.0
9.1010.08
9.01-9.11
7
9.0
9.08+0.05
9.01-9.12
9.5
9.5810.05
9.52-9.59
6
10.0
10.16+0.09
10.00-10.24
10.0
10.10+0.12
9.96-10.22
8
10.0
10.16±0.10
10.01-10.26
10.5
10.45+0.17
10.15-10.58
Mainstem
near Gunston
1
8.0
7.98±0.05
7.90-8.01
8.5
8.53+0.05
8.44-8.58
3
8.0
7.99±0.02
7.89-8.03
8.0
8.0310.06
7.92-8.08
5
9.0
9.0010.12
8.°2-9.07
9.0
9.0310.07
8.91-9.13
7
9.0
8.91+0.17
8.66-9.04
9.5
9.51+0.05
9.44-9.58
6
10.0
10.05+0.07
10.01-10.15
10.0
10.0210.06
9.97-10.15
8
10.0
10.09+0.06
10.03-10.18
10.5
10.53+0.12
10.25-10.59
Mainstem near
Broad Creek
1
8.0
8.28+0.07
8.22-8.36
8.0
8.13+0.13
7.95-8.23
2
8.0
8.29+0.09
8.23-8.39
8.5
8.5710.07
8.51-8.68
6
9.0
9.07+0.06
9.00-9.12
9.0
9.0410.05
8.97-9.11
5
9.0
9.01+0.05
8.96-9.08
9.5
9.5710.12
9.50-9.78
3
10.0
10.04+0.09
9.95-10.20
10.0
10.1910.11
10.07-10.33
4
10.0
10.06+0.10
10.04-10.20
10.5
10.5210.07
10.40-10.59
W.Wilson Bridge
2
8.0
8.22+0.15
8.05-8.32
8.0
8.14+0.09
8.00-8.20
7
8.0
8.19+0. 13
8.03-8.24
8.5
8.53+0.04
8.49-8.59
4
9.0
9.01+0. 12
8.97-9.16
9.0
9.0610.05
9.00-9.13
8
9.0
9.02±0.08
9.02-9.10
9.5
9.5510.06
9.54-9.65
1
10.0
10.04+0.10
10.03-10.18
10.0
10.0710.08
10.00-10.17
3
10.0
9.96+0. 16
9.84-10.10
10.5
10.5210.04
10.48-10.56
-------
Table 3. Average (± standard deviation) and range, during 24-h flux measurements, of pH of water overlying
sediment cores taken from the Potomac River.
£H_ of Water Column
Flux 1 Flux 2
Target _ Target
Location
Core #
PH
x ± S.D.
range
pH
x 1 S.D.
range
Hallowing Point
1
8.0
8.1010.01
8.10-8.11
10.0
9.98+0.12
9.86-10.11
2
8.0
8.1410.01
8.13-8.14
8.0
8.0110.07
7.95-8.09
3
8.0
8.1010.06
8.06-8.14
10.0
10.1110.08
10.05-10.20
4
8.0
8.1410.01
8.13-8.15
8.0
8.0210.07
7.98-8. 10
Indian Head
1
8.0
7.9910.01
7.80-7.98
10.0
10.0410.15
10.06-10.12
2
8.0
7.9810.04
7.95-8.00
8.0
7.9610.11
7.85-8.07
3
8.0
7.9010.13
7.81-7.97
8.0
7.9610.13
7.85-8.09
4
8.0
7.9010.13
7.81-7.97
10.0
10.0610.13
10.05-10.25
Mattawoman
1
8.0
7.7210.10
7.65-7.96
10.0
10.1410.16
10.00-10.41
2
8.0
7.7210.10
7.65-7.96
8.0
7.8610.13
7.70-8.00
3
8.0
7.6710.06
7.63-7.97
8.0
7.8710.12
7.74-7.98
4
8.0
7.6610.06
7.62-7.97
10.0
10.0610.15
10.05-10.25
Smith Point
1
8.0
7.8810.11
7.97-7.81
10.0
10.0310.12
10.03-10.13
2
8.0
7.8610.15
7.76-7.97
8.0
7.9610.16
7.81-8.10
3
8.0
7.8610.15
7.74-7.96
10.0
10.0210.12
10.07-10.11
4
8.0
7.9010.10
7.83-7.96
8.0
7.9410.13
7.80-8.05
-------
WASHINGTON. D.C
"ALEXANDRIA \ 4
ALEXANDRIA
LOCATION MAP
• 5
6^
OCCOOUAN
V BAY
NAUTICAL MILES
Figure 2. Sediment core collection locations in Potomac River; (X) sites
for detailed pH - phosphorus release rate studies, (®) addition-
al sites for intersite variability studies (Smith Point at river
mile 45.8 is not shown).
11
-------
using the phosphate release rate results from eight locations:
the four locations used in the detailed pH-phosphorus release
studies outlined above plus four additional locations. The
four additional locations were (1) mainstem off Smith Point,
(2) mainstem near Indian Head (3), shallows near Hallowing
Point, and (4) Mattawoman Creek (Fig. 2). Four cores were
collected from each of these locations and the release of
phosphate was measured on all four cores with the overlying
water at pH 8. The pH of the water over two cores from each
location was increased to 10. After 5 days, the release of
phosphate was again measured for all cores.
Intrasite variability in phosphate release rates at high
pH was examined using the phosphate release rates from dupli-
cate cores (pH 10 treatment) from the eight locations.
Buffering of the pH of the Overlying Water by Sediment
Processes
During preliminary studies of phosphorus release at pH 10
(Seitzinger 1985), it appeared that the sediments exerted a
"buffering action" on the overlying water pH (i.e., every few
hours it was necessary to add base (NaOH solution) to the water
over the sediment cores to keep the pH at 10, whereas much less
frequent additions were needed to maintain the pH at 10 in the
control water incubated without sediments).
The effect that the sediments have on decreasing the pH of
the bottom water in the river was examined using the sediment
cores from each of the eight locations. The rate of decrease
in pH of the overlying water due to sediment processes was
examined at the four locations used for detailed pH-phosphorus
release rate studies by measuring the time that it took the pH
to drop 0.5 pH units from pH 8.5, 9.5 and 10.5. Those measure-
ments were carried out on a day that nutrient flux measurements
were not made. Controls for pH changes due to water column
processes consisted of measurements of the rate of change in pH
of water incubated without sediment. The buffering action by
12
-------
the sediments was also examined at the four additional loca-
tions used in the spatial variability study. The time to
decrease 0.5 pH units below pH 10 was determined.
Denitrification Rates
Preliminary measurements of denitrification rates at pH 8
were made using one sediment core collected 2 October 1985 by
SCUBA divers from Gunston Cove and one from the mainstem of the
river near Indian Head. Denitrification was measured as a flux
of from those vertically intact sediment cores maintained
under aerobic conditions in gas-tight incubation chambers, as
described below. The details of that procedure are described
in Seitzinger et al. (1980) and Seitzinger, Nixon and Pilson
(1984).
The sediment cores (7-cm deep) were transferred to gas-
tight glass incubation chambers for measurement of , 0^, NH4,
N02 plus NOa, and P04 fluxes. The cores were incubated at
22±2°C in the dark with filtered (Gelman AE) river water (-650
ml) which was stirred continuously with a floating magnetic
stirring bar to facilitate the equilibration of dissolved gases
with the overlying gas phase (-70 ml). Water over the cores
was changed every 24 to 48 h with freshly prepared low-N^ river
water obtained by flushing water from the site of sediment
collection with a gas mixture of 21% 0-^ and 79% He.
Duplicate samples (50 (jl) of the gas phase were taken from
each chamber for and 0^ analysis approximately 24 h after
the water was changed, at 48 h, and for some cores, at 72 h.
The differences in concentration between sequential samples
within an incubation were used to calculate the net and 02
flux across the sediment-water interface. Gas samples were
injected directly into a gas chromatograph (Schimadzu, Model
GC-8A) equipped with a thermal conductivity detector (2-m x
0.318-cm o.d. stainless steel columns packed with 45/60 mesh
Moelcular Sieve 5A He carrier gas flow rate, 25 cm min'1)-
13
-------
Previous experiments (Seitzinger et al. 1980) show that
the N2 initially dissolved in the porewaters is equilibrated
with the low-N2 overlying water in about one week; only mea-
surements made after that period of time are reported here.
Four separate N2 flux measurements were made of each core.
Initial water samples for sediment-water nutrient fluxes
were taken from the chambers after the water was changed over a
core, and before the chambers were closed for Nv, measurements.
Final samples were taken 48 or 72 h later after the final gas
sample was collected. Samples from control treatments consist-
ing of water incubated without sediment were taken initially,
and final samples were taken 48 to 72 h later, at the same time
as the core samples. All samples were analyzed for ammonium,
nitrite plus nitrate and phosphate, according to the procedures
described below.
Sediment-Water Nutrient Flux Measurements
The sediment-water exchanges of phosphate, nitrite plus
nitrate, and ammonium were measured in the studies outlined
above according to the following procedures. Water pH was
adjusted using 0.05 M NaOH before introduction over the cores.
Approximately 1 h after the water was changed, an initial water
sample was collected, and approximately 24 h later a final
water sample was collected. A portion of each sample was
filtered through prerinsed glass fiber filters (Whatman, AH)
immediately after collection and frozen to be analyzed later
for nitrite plus nitrate (Technicon Industrial Systems 1977)
and soluble reactive phosphate (APHA, AWWA and WPCF 1981). Un-
filtered subsamples were analyzed immediately for ammonium
(Solorzano 1969).
Changes in nutrient concentrations due to water column
processes were measured by incubating water without sediment.
This water was obtained by drawing off approximately 100 ml of
water overlying each core after the initial core sample was
taken. These controls were treated and sampled using the same
14
-------
procedures as were used for sampling the water over the cores,
except that in all but the Gunston Cove and Hallowing Point
measurements and first flux measurements on cores 3, 7 and 8
from mainstem near Gunston Cove, the control final samples were
collected after a period of 2-4 h, not 24 h. This shorter
incubation time was necessary because the phosphate concentra-
tion decreased with time in the control water. To calculate
the rate of phosphate decrease, the final control sample had to
be collected before the concentration dropped below the level
of analytical detection.
15
-------
RESULTS AND DISCUSSION
Phosphate Fluxes as a Function of pH
The cores from all eight locations had low fluxes of
phosphate from the sediments (<25 |jmol P nT'^h-1) with the
overlying water at pH 8, except for one of the four cores
collected at Smith Point (44 nmol P m'^h"1; Table 4). This
range is consistent with values previously reported for Gunston
Cove and nearby Potomac River mainstem sediments. Callendar
and Hammond (1982) measured in situ phosphate benthic fluxes at
a Potomac River mainstem station off Piscataway Creek in August
1979 and reported values ranging from 0 to 8 (jmol P nT'^h""1.
Phosphate fluxes measured in September 1983 from Gunston Cove
sediments ranged from 0 to 19 |jmol P m'^h"1 (Seitzinger 1983)
and in September 1984 from -3 to 6 pmol P m'^h"1 (Seitzinger
1985). In July 1984 phosphate fluxes from sediments collected
in Gunston Cove and Potomac River mainstem between Broad Creek
and Indian Head ranged from 0 to 0.6 |jmol P nT^h"1 (Seitzinger
1985).
The effect of increasing pH on phosphate flux did not
occur until the overlying water pH was greater than 9.0 (Tables
4 and 5, Fig. 3a). Increased phosphate fluxes occurred at pH
9.5 at two locations, Gunston Cove and mainstem near Gunston
Cove. Increases in phosphate flux from the other two locations
were not noted until the overlying water pH was 10.0.
The phosphate fluxes at pH 10 varied between 30 and 120
|jmol P nT^h-1 among the eight locations (Figs. 3 and 4). The
highest flux of phosphate was from Gunston Cove sediments where
the fluxes ranged from 99 to 120 pmol P m~2h~1. This release
of phosphate is similar to, although slightly higher than,
rates measured at pH 10 in October 1984 in Gunston Cove sed-
iments (range of 57 - 79 jjmol P nT^h"1) (Seitzinger 1985).
Lowest fluxes of phosphate at pH 10 were from Indian Head
sediment. There is no direct evidence of the cause of the
16
-------
Table 4. Sediment water fluxes of phosphate (pmol P m z h ') from Potomac River sedimenL cores in-
cubated with varying overlying water pH. Each number represents the result of one flux
measurement.
Si te";
Location
Gunston Cove
Mainstem near
Guston Cove
Mainstem near
Broad Creek
Approximate
Water Depth
(m)
W. Wilson Bridge
3.5
Date
collected
9/9/85
9/30/85
10/28/85
10/28/85
Core
Overlying
#
8.0
8.5
9.0
1
7,2,3
2""
8
3
3,3
2
4V'~"
18
5
2
7,4
6
16
7
4
7
8
6
1
1,7
11
2 **
0
3
1,2,4
4>v*
3
5
1
7,10
6
1
7
2
2
8
1
1
1,-2,14
2
0,3
-3
3
-2
4
-1
5
0
0
6
0
0,0
/ A A
3
8**
5
1
0
2
2,11,8
3
-1
4
1
17,20
5 " *
4
6""
7
7
3,-5
13
8
0
4
31
37
120,99
119
55,87
49
43,49
58
69,88
96
10.5
96
97
53
82
13
-------
Table 4 (continued). Sediment water
fluxes of phosphate
(pmol P hi 2 h ') from Potomac River sediment
cores incubated with varying
overlying
water pH.
Approximate
Water Depth
Date
Core
Overlying Water pH
Site"" Location (m)
collected
//
8.0 8.5 9.0 9.5 10.0 10.5
5 Hallowing Point 2
9/30/85
1
1 33
2
2,2
3
2 35
4
4,3
6 Indian Head 5
12/4/85
1
8 30
2
6,4
3
-2,5
4
9 31
7 Mattawoman Creek 1.5
12/4/85
1
14 30
2
10,5
3
15,0
4
17 48
8 Smith Point 2.5
12/4/85
1
11 35
2
16,12
3
44 58
4
25,23
"See Fig. 1 for station location.
"""Cores not used in further flux measurements.
-------
Table 5. Range and average of phosphate fluxes from Potomac River sediment cores incubated with overlying water at
pH indicated.
No. of
PO4 Flux
Si te
Treatment
Location pH
Flux
Measurements
Range
pmol P m 2 h 1
Average
|jmol P m 2 h 1
Average
mg P m A day 1
1
Gunston Cove
8.0
9
2-16
5
4
8.5
1
2
2
2
9.0
3
4-7
6
4
9.5
1
31
31
23
10.0
3
99-120
113
84
10.5
1
96
96
71
2
Mainstem near
8.0
9
1-7
2
2
Gunston Cove
8.5
1
11
11
8
9.0
3
2-10
6
4
9.5
1
37
37
28
10.0
3
49-87
64
48
10.5
1
97
97
72
3
Mainstem near
8.0
9
-2-14
2
2
Broad Creek
8.5
1
-3
-3
-2
9.0
3
0
0
0
9.5
1
4
4
3
10.0
3
43-58
50
37
10.5
1
53
53
39
4
W. Wilson Bridge
8.0
9
-5-11
3
2
8.5
1
13
13
10
9.0
3
4-20
14
10
9.5
1
13
13
10
10.0
3
69-96
84
62
10.5
1
82
82
61
-------
Table 5 (continued). Range and average of phosphate fluxes from Potomac River sediment cores incubated with
overlying water at pH indicated.
Site
Location
No. of
Treatment Flux
pH Measurements
P04 Flux
Range
|jmol P m 2 h 1
Average
(jmol P m h 1
Average
nig P m 2 day 1
Hallowing Point 8.0
10.0
Indian Head 8.0
10.0
8.0
]0.0
8.0
10.0
Mattawoman
Creek
Smith Point
6
2
6
2
6
2
6
2
1-4
33-35
-2-9
30-31
0-17
30-48
11-44
35-58
2
34
5
30
10
39
22
46
2
25
4
22
7
29
16
34
-------
1 20
1 00
90
SO
70
60
50
4.0
30
20
-10
s
9.2
9.6
10
8.8
10.8
pH of ovarlying *oUr
50
*0 -
30 -
20 -
l 0
/
— 1 i f 1 i 1 » 1 1—•
3 a.4 a.8 9.2 9.6 10 10.4
1 0.8
pH of ov«riying wortr
Figure 3. Average POt, flux at various pH levels.
(a) Detailed pH-p'nospnorus release rate
sites: Gunston Cove (A) / mainstem near
Gunston Cove (~), mainstem near Broad
Creek ( • ) , and near W. Wilson Bridge ( O) •
(b) Non-detailed sites: Hallowing Point
( ~ ) , Indian Head ( ~ ) , Mattawoman Creek
( ¦ ) , and Smith Point (0) • The points
on (b) are connected by a line to guide
the eye only;, --this is not meant to indi-
cate a straight line relationship.
21
-------
t 30 t
135--
105--
N
'6
to
£
o
£
a
x
3
o
a.
>
<
Figure 4. Average sediment-water flux of PO4 from Potomac River sediment
cores with overlying water at pH 10. Stations are, from left
to right, in order downstream from the Woodrow Wilson Bridge.
The Blue Plains sewage treatment facility is located just up-
stream from W. W. Bridge. Stations inside the • marks are
within the area of the 1983 algal bloom.
22
-------
i_jiiPO
Figure 5a. Solubility of iron, aluminum and calcium phosphates
as a function of pH (from Stumm and Morgan 1981).
L"*( MUD ASM
m
•0
s
CO
mU.dl'9 CAHTM
FCRRIC MTQUOXIOC
o
s
40
20
*
M
O
4
4
Figure 5b. Phosphate sorption by various substrates as a
function of pH (from MacPherson, Sinclair and
Hayes 1958).
24
-------
variability in phosphate fluxes at pH 10 between locations. It
does not appear to be directly related to distance downstream
from the Blue Plains plant (Fig. 4). There were no distinct
differences between the phosphate flux from sediments collected
in the 1983 bloom area and those collected outside the bloom
area. There is an indication that the date of sediment collec-
tion may have influenced the magnitude of the phosphate flux,
since some of the lowest fluxes were from sediments collected
last (Indian Head, Mattawoman Creek and Smith Point). However,
the relationship between the phosphate flux and the date of the
sediment collection was not statistically significant. It is
likely that a major factor influencing the magnitude of the
phosphate flux at high pH is the amount of phosphorus held in
the surface sediments in forms that are soluble at high pH.
We do not currently have data on the concentration of the
various forms of phosphorus in the sediments at the various
locations; fortunately, sediment samples were saved and could
be analyzed.
Phosphate in sediments can occur in a number of forms that
are pH dependent. Such forms include phosphate bound to Al, Fe
or Ca, or phosphate adsorbed to particles. Binding of phos-
phate to both Al [AlfOH)^] and Fe [Fe(OH)g] is strongest in
solutions with pH between 5 and 7 (Fig. 5a; Stumm and Morgan
1981). However, the binding of phosphate to aluminum and
ferric hydroxide in natural sediments may occur over a wider
range of pH due to the presence of humic acids, which form
colloids with metal-phosphate complexes (Ohle 1963, cited by
Andersen 1971). In contrast to Al and Fe phosphate minerals,
calcium phosphate minerals (apatites) decrease in solubility at
high pH (Stumm and Morgan 1981; Fig. 5a). In addition to these
changes in dissolution, increases in pH make the surface charge
of particles (clays and hydroxides) more negative, which causes
a decrease in the adsorption of negative phosphate ions.
Figure 5b shows some typical phosphate sorption curves for
bentonite (impure aluminum silicates), lake mud ash and ferric
hydroxide (MacPherson, Sinclair and Hayes 1958). Similar de-
23
-------
creases in sorption of phosphate have been demonstrated in
laboratory studies for alumina and kaolinite clay (Chen, Butler
and Stumm 1973), hermatite (Breeuwsma and Luklema 1973), and an
iron oxide and a soil (Obihara and Russell 1972). All demon-
strate decreased sorption of phosphate at approximately pH 6,
with sorption reduced more sharply between pH 8 and 10. Sev-
eral chemical species can therefore interact in a natural
system to provide the resultant effect of an increase in pH.
The results of such laboratory studies with artificial
substrates agree fairly well with the limited number of studies
using natural sediments from freshwater systems. The complexity
of the mineral interaction in sediments is demonstrated in a
study by Andersen (1975) in which the release, from the sedi-
ments of a eutrophic Danish lake, of phosphate over a range of
pH (8 to 11) was measured. Andersen found maximum phosphate
release (970 pmol nT^h"*1) at pH 9.5. Between pH 8 and 9.5 the
essentially linear increase in net release was attributed to an
increase in the exchange of phosphate sorbed to clay minerals
and to iron hydroxides. Above pH 9.5 there was a decrease in
liberation of phosphate as a result of precipitation of the
desorbed phosphate as hydroxyapatite. At pH 11, CaCOa was
precipitated from the water and subsequently formed hydroxya-
patite at the sediment surface, where phosphate concentrations
were high. Jacoby et al. (1982) observed a less dramatic
increase in phosphorus release in response to increased pH.
Total phosphorus release from Long Lake, Washington, sediments
increased from <1 prnol irT'^h'1 to 3 |jmol m~^h-1 when the pH was
increased from 6 to 10.
Specific studies of the various forms of phosphorus as a
function of pH at the eight locations in the Potomac River are
required to ascertain which chemical species are involved in
producing the observed phosphate response when pH is increased.
In the Potomac River sediments, the increased release of phos-
phate above pH 9.0 is probably due to increased exchange of
phosphate sorbed to clay minerals and to iron and aluminum
hydroxides. Characterization of the forms of inorganic phos-
25
-------
phorus in sediments is based on the ability of certain reagents
to dissolve various phosphate minerals (Chang and Jackson
1957). While the chemical fractionation scheme used is
somewhat limited by the inability of the procedures to quanti-
tatively remove discrete fractions, the analyses can provide
insight into the major forms present. Recent studies by
Broderick (1986), using Gunston Cove sediments maintained with
overlying water at pH 10, indicate that the phosphorus released
at high pH is associated mainly with inorganic phosphorus
mineral components that are soluble in 0.1 M NaOH. This is
indicative of phosphorus associated with iron oxides. Some of
the phosphorus released at high pH was also from inorganic
phosphorus minerals soluble in NH4F, which indicates amorphous
Al-P components.
Other Factors Influencing Phosphate Fluxes
A number of factors, in addition to pH, may influence the
phosphate fluxes. These include oxygen, phosphate concen-
tration and nitrate concentration. As the following paragraphs
demonstrate there is no evidence that any of these were respon-
sible for the increase in phosphate flux measured in the Poto-
mac River sediments.
The effect of oxygen-dependent redox potential on the
release of phosphate from sediments is demonstrated in the
classical studies by Mortimer (1941 and 1971). The increased
release of phosphate under anaerobic (reducing) conditions,
relative to aerobic conditions, is attributed mainly to the
decrease in the phosphate sorption capacity of ferrous hy-
droxide relative to ferric oxyhydroxide (Mortimer 1971).
Previous studies have demonstrated the effect of oxygen con-
centration on phosphate release from Potomac River sediments.
Rates of phosphate release from Potomac River sediments after
a week of exposure to anaerobic conditions ranged from 18 to
30.6 jjmol P riT^h-1 in July 1984 (Seitzinger 1985). In the
26
-------
present studies, oxygen concentrations were always near satura-
tion; therefore oxygen concentrations could not have been
responsible for the increased phosphate fluxes measured.
Equilibrium models suggest that the concentration of
phosphate is maintained in the water column through regulation
by the sediments. When the equilibrium concentration in the
water is exceeded, the excess phosphate moves into the sedi-
ments where it is retained. Low concentrations in the over-
lying water would produce a flux of phosphate out of the sed-
iments. Cerco (1985) calculated the equilibrium concentration
of phosphate in Gunston Cove to be 0.018 mg/L or roughly 0.6
(jM. While this may hold at the normal pH of the water (7 -
8), it would not be expected to remain constant as the pH
increases and the solubility of various mineral forms changes.
In the measurements of phosphate fluxes versus pH in Potomac
sediments, there were large fluxes of phosphate out of the
sediments when the phosphate concentration in the water over
the cores was as high as 20 (jM (Fig. 6). Clearly pH, not
phosphate concentration, was the primary factor controlling
phosphate release rates, although at pH 8, phosphate concen-
trations may have affected phosphate release rates.
Nitrate concentration has been suggested as one factor
controlling the release of phosphate from lake sediments
(Bostrom and Pettersson 1982; Andersen 1982), and from Gunston
Cove sediments (Cerco 1985). It is suggested that high water
column nitrate concentrations maintain higher pore water ni-
trate concentrations, and thus higher redox potentials deeper
in the sediments. This depresses the release of phosphorus
from the sediments. Nitrate concentrations below 1 mg NO^-N/L
are generally considered necessary to enhance phosphorus re-
lease. The nitrate concentrations in the water overlying
Potomac River sediments in the present study were always
greater than 1 mg NOa-N/L, which reflects the river water
concentrations during the study period. Therefore, no effect
of nitrate concentration on the phosphate fluxes was seen (Fig.
27
-------
P04 Flux vs P04 Concentration
K>
00
I
SL
(N
I
E
(0
J?
o
E
a.
X
3
o
Q_
(§1 ~
P04 concentration, yjM
Figure 6. PO!, flux versus average of the initial and final PO,, concentration of the
overlying water. P04 'fluxes over the entire pH range are included. These
data are from detailed sites only.
-------
7). It is possible that phosphate fluxes could have been
higher if the nitrate concentrations had been less than 1 mg
NO3-N/L. Studies of the combined interaction of nitrate and pH
on phosphate release should be conducted.
Initial and final phosphate, nitrate and ammonia concen-
trations, and the calculated fluxes for all the detailed sites
are shown in Table 6. Initial and final phosphate concen-
trations, and calculated fluxes for all non-detailed sites are
shown in Table 7.
Nitrogen Fluxes as a Function of pH
At pH 8 there was a positive, but small flux of ammonia
from sediment cores at the four locations examined (Table 8,
Fig. 8). Ammonia release rates were highest from the Woodrow
Wilson Bridge sediments. The flux of ammonia from the sedi-
ments increased with increased pH at all stations examined.
The pH at which this increased flux first occurred varied:
Woodrow Wilson Bridge at pH 9, mainstem Broad Creek at pH 10,
and Gunston Cove and mainstem near Gunston Cove at pH 10.5.
Uptake of nitrate by the sediments was, in general, ob-
served at all locations and over the experimental pH range
(Fig. 9; Tables 6 and 8). In a limited number of cases, the
calculated nitrate flux was out of the sediments (Table 6),
although there was actually a decrease in concentration between
the initial and final samples from the water over the sedi-
ments. The calculated positive flux of nitrate out of the
sediments in those cases reflects the relatively large decrease
in nitrate concentration in the final control sample.
The uptake of nitrate by the sediments generally increased
as the overlying water pH increased. The response was not as
distinct as that for phosphate or ammonia. An unusually large
uptake of nitrate at pH 9.5 was calculated for a core from the
mainstem near Gunston Cove (Table 8). This is due primarily to
the nitrate concentration of the control final sample, which
was unusually high compared to the initial concentration (Table
29
-------
P04 Flux vs N02 + N03 Concentration
U)
o
I
-C
-------
Table 6. Summary of core initial (I) and final (F) nutrient concentrations (yM) and calculated sediment-
water nutrient fluxes. Time interval between core initial and final samples was approximately
24 h. Time interval between initial and final control (CF) samples was 2-4 h, except for all
Gunston Cove controls and the first flux on controls 3, 7 and 8 from mainstem near Gunston Cove,
which were incubated for 24 h.
P0«
N02 +¦ no3
NH,
Flux Flux Flux
Core I F CF I F CF (paol 1 F CF (jimol
Location pH No. (/iM) (pM) (pM) «"2h-' ) (pM) (pM) (pM) a ^h1) (pM) (pM) (pM) a"2!!"1
Gunston Cove
Ma 1 listen near
Gunston Cove
8.0
1
0.3
o
00
0.0
2
93
67
91
-67
1 . 2
0 . 5
0 . 5
0
8.0
1
0.4
1.1
0.0
3
91
71
95
-64
1 . 1
0 . 9
0. 1
2
8.0
3
0.2
1 . 0
0.0
3
95
82
105
-71
1 . 8
1 . 0
0 . 6
1
8.5
3
0.3
0.6
0.0
2
85
65
89
-73
1 . 1
0.4
0.2
1
9.0
5
0.5
2.7
0.0
7
90
68
99
-78
1 . 3
0 . 6
0 . 6
0
9.0
5
0.3
1. 8
0.0
4
84
61
87
-62
1 . 0
0. 5
0 . 2
1
9.0
7
0.3
2. 1
0.0
7
99
83
85
-5
1 . 2
0 . 6
0 . 6
0
9.5
7
1. 6
9.3
0.3
31
90
94
95
-5
1 . 4
1 . 2
0 . 2
4
10.0
6
3.6
27.6
0.6
120
97
94
88
30
3 . 0
5 . 5
0 . 5
22
10.0
6
5.7
30.6
6.0
99
95
90
94
-16
2.5
1 . 8
0 . 2
7
10.0
8
5.6
31 . 2
0 . 7
119
102
95
89
23
3. 3
9. 7
0.5
36
10.5
8
6.9
33. 3
5. 3
96
90
52
95
- 152
11.8
34 . 7
5. 1
103
8.0
3
0.2
0.4
0.0
2
145
123
154
-126
1 . 8
0 . b
0.5
1
8.0
3
0. 1
1 . 2
0. 1
4
149
122
153
-188
0.6
0.5
0 . 4
6
8.0
1
0.3
0.3
0.0
7
139
119
146
-287
I . 9
0 . 4
1 . 4
6
8.5
1
0.4
1 . 1
0. 1
11
152
13]
154
-178
0.8
0 . 6
0 . 2
20
9.0
5
0.2
0.9
0.0
7
138
108
138
-121
1 . 4
11.0
1 . 2
45
9.0
5
0.2
1 . 1
0.0
10
151
1 1 1
149
-100
0.6
0.7
0. 3
1 1
9.0
7
0. 1
0.5
0.0
2
143
1 10
139
-148
1 . 3
0 . 6
0 . 5
1
9.5
7
1.5
8.8
1 . 5
37
123
132
156
-1484
1 . 3
1 . 2
0.8
25
10.0
6
1 . 1
10.4
0.8
55
144
122
139
43
2 . 1
0 . 5
1 . 4
14
10.0
6
6.5
22. 2
6.3
87
147
1 38
157
-508
0 . 5
7 . 3
0. 6
30
10.0
8
1.2
12 . 5
0.6
49
144
120
142
-90
1 . 9
0 . 8
0 . 5
1
10.5
8
6. 3
29. 4
6.2
97
149
92
152
-339
7. 1
22.8
6.6
79
-------
Table 6 (continued). Summary of core initial (I) and final (F) nutrient concentrations (yM) and calculated
sediment-water nutrient fluxes. Time interval between core initial and final samples
was approximately 24 h. Time interval between initial and final control(CF) samples
was 2-4 h, except for all Gunston Cove controls and the first flux on controls 3, 7
and 8 from mainstem near Gunston Cove, which were incubated for 24 h.
P0«
N02 + NOa
NH4
Flux
flux
(lux
Core
I
F
CF
(fiaol
I
F
CF
( (1BO 1
1
F
CF
(fimo I
Location
pH
No.
(jlM)
(|iM)
(jiM)
or 2 h~ 1 )
(pM)
(jjM)
(fiM)
¦ ~ 2 h~1 )
(pM)
(»JM)
(f»M)
¦ ~ 2 h
Mainstea near
8.0
1
2.5
1 . 4
2.4
-2
152
138
155
-86
2. 3
1 . 0
2 . 0
1
Broad Creek
8.0
1
2.1
2.5
1 . 7
14
8.0
2
2.3
1.9
2.1
3
156
162
158
-39
2 . 3
1 . 0
1 . 7
1 1
8.5
2
2.4
1. 8
2.4
-3
9.0
6
2.0
1.6
2.0
0
135
114
138
-182
1 . 7
1 . 0
1 . 3
9
9.0
6
1.6
0.8
1.6
0
9.0
5
1 . 9
1 . 8
1 . 9
0
137
137
143
-214
1 . 6
1 . 3
1 . 9
-12
9.5
5
1. 5
2.0
1 . 4
4
10.0
3
6.2
13.7
6.2
43
132
91
140
-515
42 . 7
83. e
4 1.1
294
10.0
3
3. 1
11.0
3.0
49
10.0
4
5.8
12.8
5.2
58
134
95
140
-414
8.2
51.3
11.0
127
10.5
4
3.3
9.6
2.8
03^52.
Woodrow Wilson
8.0
2
2.2
1.4
1.7
11
261
198
256
-92
1 . 2
2 . 6
0 . 6
25
Bridge
8.0
2
1 .5
0.9
1 . 2
a
8. 0
7
2.2
2.2
2.3
-5
269
230
264
5
1 . 3
4 . 4
0.4
47
8.5
7
1 . 7
1.7
1.4
13
9.0
4
3.3
6. 1
3.0
17
265
183
259
-93
4 . 0
28. 3
2. 1
127
9.0
4
1.4
6.4
1.4
20
9.0
8
2.8
2.6
2.7
4
255
19b
249
-28
2 . 6
1 . 6
1 . 6
30
9.5
8
1 . 1
3.4
1.0
13
10.0
1
5.3
15.4
5.0
69
257
171
251
-207
34 . 1
74 . 0
31.2
345
10.0
1
6.4
19.0
6. 1
88
10.0
3
7. 1
23.9
6.6
96
264
136
253
-216
29. 3
78.0
29. 2
234
10.5
3
7.7
27.6
7.8
82
-------
Table 7. Summary of core initial and final PO4 concentrations (pM) and calcu-
lated sediment water fluxes for the non-detailed sites. Time inter-
val between core initial (I) and final (F) samples was approximately
24 h, and between control initial and final (CF) samples was 2-4 h except
Hallowing Point controls which were approximately 24 h.
Location
PH
Core
#
I
F
CF
Flux
-
(MM)
-
(pmol m z h 1)
Hallowing Point
8
2
0.3
0.5
0.1
2
8
2
0.3
0.5
0.0
2
8
4
0.2
0.8
0.1
4
8
4
0.3
0.6
0.0
3
8
1
0.2
0.5
0.3
1
8
3
0.2
0.4
0.1
2
10
1
0.7
5.8
0.3
33
10
3
0.7
5.2
0.2
35
Smith Point
8
2
1.5
2.3
1.3
16
8
2
1.4
1.4
1.1
12
8
4
1.9
3.1
1.4
25
8
4
1.6
2.4
1.1
23
8
1
1.8
2.2
1.5
11
8
3
2.9
5.8
1.4
44
10
1
3.3
10.2
3.4
35
10
3
5.0
15.4
4.6
58
Indian Head
8
2
0.7
0.2
0.6
6
8
2
0.4
0.0
0.2
4
8
3
0.4
0.5
0.5
-2
8
3
0.4
0.2
0.2
5
8
1
0.7
0.5
0.4
8
8
4
0.4
0.4
0.3
9
10
1
1.4
4.6
0.7
30
10
4
1.2
3.5
0.7
31
Mattawoman
Creek
8
2
0.7
0.8
0.4
10
8
2
0.4
0.8
0.2
5
8
3
0.6
1.1
0.4
15
8
3
0.5
0.2
0.4
0
8
1
0.6
0.6
0.4
14
8
4
0.8
1.0
0.4
17
10
1
2.1
6.4
2.0
30
10
4
1.6
8.1
1.2
48
33
-------
Table 8. Average NO^ + NO^ and average NH4 fluxes from Potomac River sediments cores incubated with
overlying water at pH indicated.
Ave. NO^ +_NOa_Flux
Ave NH4_F1ux
No. of Flux
Location
P»
pmol N m 2 h 1
[Jmol N m 2 h 1
Measurements
Gunston Cove
8.0
-67
1
3
8.5
-73
1
1
9.0
-48
0
3
9.5
-5
4
1
10.0
12
22
3
10.5
-152
103
1
Mainstem
near Gunston
8.0
-200
4
3
8.5
-178
20
1
9.0
-123
19
3
9.5
-1484
25
1
10.0
-185
15
3
10.5
-339
79
1
Mainstem near
8.0
-62
6
2
Broad Creek
9.0
-198
1
2
10.0
-464
210
2
W. Wilson Bridge
8.0
-48
36
2
9.0
-60
78
2
10.0
-211
290
2
-------
Ave. NH4 Flux vs pH
u>
Ln
I
_C
(N
I
£
M
o
E
a.
X
3
X
z
o
>
<
.500
280
-
260
-
240
-
220
-
200
-
180
-
160
-
1 40
-
1 20
-
100
-
80
-
60
-
40
4
20
-
0
4
8
9.2 9.6
pH of overlying water
l 1 1 r
10.4 10.8
Figure 8. Average NH4 flux at various pH levels for sediment cores collected from Cunston
Cove (A), mainstem near C.unston Cove (A), mainstem near Broad Creek (®), and
near Woodrow Wilson Bridge (Q).
-------
Ave. N03 Flux vs pH
1 00
0
-100
-200
-300
-400 -
-500 -
-600
1 0.4
1 0.8
9.2 9.6
pH of overlying water
Figure 9. Average NO3 flux at various pH levels for sediment cores collected from
Gunston Cove (A), mainstem near Gunston Cove (A), mainstem near Broad
Creek (0), and near Woodrow Wilson Bridge (O) • The line is dashed be-
tween pH 9 and 10 for mainstem near Gunston Cove because the very large
(-1484) negative flux measured at pH 9.5 was off the scale of the graph.
-------
6). The reason for this is not clear; it is possible that the
control final sample was contaminated.
The magnitude of sediment-water nitrate fluxes measured at
pH 8 (5 to -287 [jmol N03~ nT^h"1 ) in the present study is
similar to previously reported nitrate flux measurements in
this portion of the Potomac River. In August, 1979 Callender
and Hammond (1982) reported sediment-water nitrate fluxes
ranging from -67 to +67 pmol N03~ m~zh_1 in the mainstem of the
Potomac River off Piscataway Creek. In late summer 1984, Cerco
(1985) reported sediment-water nitrate fluxes ranging from -190
to -726 |jmol N03~ m~^h_1 in Gunston Cove sediments (at 20-28
°C, 8 mg Oz/L). Ammonia fluxes reported by Callender and
Hammond (1982; 183 to 500 pmol NH4+ nT'^h-1) and Cerco (1985;
235 to 292 |jmol NH4+ m~2h-1) were higher than those measured in
the present study (0 to 47 (jmol NH4+ nT^h"1)- The reason for
this difference is not known.
Denitrification was an active process at both locations
examined (Table 9). Rates for Gunston Cove and Indian Head
sediments were both about 200 pmol N m'^h"1, and are similar to
rates measured in the tidal freshwater portion of the Delaware
River (Seitzinger and Casselberry, unpublished data). These
are the first denitrification measurements made in the tidal
freshwater portion of the Potomac River. Denitrification has
been recognized as a potential sink for nitrogen in the Potomac
River and included as a removal function for nitrogen in both
the PEM and Gunston Cove Model. The assumed rates used in
those models could be updated using the rates measured here.
Denitrification appears to be an important sink for ex-
ternal nitrogen inputs to the Potomac River at this time of
year. The nitrogen input to the tidal freshwater portion of
the river from upstream and point sources during September and
October 1985 was approximately 63,213 lbs/day (MWCOGS, unpubl.
data). The area of the tidal freshwater portion of the river,
including embayments, is 166 x 10b m* (Fitzpatrick, pers.
comm.). This results in a nitrogen loading rate for the tidal
freshwater portion of the river of 517 [jmol N nT^h-1. Assuming
37
-------
Table 9. Denitrification rates and sediment-water nutrient fluxes at pH
8 at two locations in the Potomac River.
Location Denitrification"" NH4 N0-^_+ NOv P04
(pg-at N m z h x) ((Jmol m 2 h *)
Indian Head 210(±31) + - NA - 24 1.4
#15 75 1.0
Gunston Cove 234(±38) + - NA - -94 1.1
# 3 -120 2.1
^Average of 4 N flux measurements ± standard deviation taken over a two
week period.
+Flux measured from 10/7/85 - 10/10/85
#Flux measured from 10/16/85 - 10/18/85
38
-------
that the denitrification rates of 200 pmol N irT^h"1 measured in
the sediments from Gunston Cove and in the mainstem off Indian
Head are representative of the denitrification rates in the
rest of the tidal freshwater portion of the river, then during
the fall denitrification is removing an amount of nitrogen
equivalent to approximately 35% of the nitrogen loading during
the same period of time to that portion of the river.
An adequate supply of nitrate is essential to drive the
denitrification process. There are two sources of this nit-
rate: (1) nitrate diffusing into the sediments from the water
column, and (2) nitrate produced in the sediments from mineral-
ization of organic matter. The flux of nitrate from the water
column into sediments is often used to estimate denitrification
rates. However, actual measurements of denitrification us-
ually demonstrate that this method greatly underestimates
denitrification rates because nitrification in the sediments
often supplies all, or a large percentage, of the nitrate that
is denitrified. This is the case in the Potomac River sedi-
ments. In Gunston Cove the flux of nitrate from the water into
the sediments (~100 pmol m'^h-1) (Table 9) was only about half
that required to supply the denitrification rates (-200 pmol N
nT^h-1) measured in the same core. At Indian Head there was a
positive flux of nitrate from the sediments, indicating that
all the nitrate used for the denitrification measured there was
from nitrification in the sediments.
The increased uptake of nitrate by the sediments, with
increasing pH> may reflect increased denitrification rates in
the sediments. However, data on denitrification rates in
Potomac River sediments, as a function of pH, are not avail-
able; denitrification rates were measured only at pH 8. Mea-
surements of denitrification rates in soils indicate that the
optimum pH range for denitrification is around 5 to 9 (Focht
and Verstraete 1977). While translation of those results to
Potomac River sediments is problematic, it is possible that the
increased flux of nitrate into the sediments reflected an
increase in denitrification rates.
39
-------
Rate of pH Change
The rate of decrease in the pH of water overlying sediment
cores and water incubated without sediment was similar at all
stations examined (Figs. 10, 11 and 12; Table 10). The rate of
decrease was generally 0.01 to 0.03 pH units per hour at all
locations except Hallowing Point, where the rate was approxi-
mately 0.07 to 0.08 pH units per hour. The reason for this
difference is not known.
40
-------
Figure 10. pH of water overlying sediment cores (A), and water incubated without
sediment (A), versus time. No acid or base was added during incuba
tion.
41
-------
Mattawoman Creek
CO rm I
Mattawoman Creek
10 3
10 1
Smith Point
co r« 1
Smith Point
cor* i
10.4
10 2
h
Figure 11. pH of water overlying sediment cores (A), and water incubated without
sediment (A), versus time. No acid or base was added during incubation.
42
-------
Hallowing Point
Cor« t
10.4 -
10 1
Hallowing Point
cor* 3
Indian Head
cor* i
Indian Head
IO. 5
10.4
tO 5
10 4 -
I0.J -
Figure 12. pH of water overlying sediment cores (A), and water incubated with-
out sediment (A), versus time. No acid or base was added during
incubation.
43
-------
Table 10. Linear regession analysis of pH versus time of water over sediment cores collected from the
Potomac River, or water incubated without sediment.
Corr. Coeff.
Location
Initial pH
Final pH
Time (h)
Slope
r
Gunston Cove
Core 1
8.52
8.22
7.25
-0.0430
0.8402
Control
8.49
8.20
7.25
-0.0349
0.8415
Core 2
9.47
9.26
7.25
-0.0276
0.8431
Control
9.47
9.34
7.25
-0.0159
0.8447
Core 3
10.48
10.25
7.25
-0.0331
0.8428
Control
10.47
10.27
7.25
-0.0248
0.8438
Mainstem near Gunston
Core 1
8.52
7.88
23
-0.0275
0.6688
Control
8.50
8.16
23
-0.0132
0.6771
Core 7
9.53
9.07
23
-0.0191
0.6749
Control
9.51
9.06
23
-0.0191
0.6749
Core 8
10.55
9.86
23
-0.0289
0.6713
Control
10.52
9.99
23
-0.0229
0.6/41
Mainstem near
Broad Creek
Core 2
8.57
8.00
32.5
-0.0191
0.7126
Control
8.57
8.47
32.5
-0.0008
0.7319
Core 5
9.58
9.23
32.5
-0.0088
0.7247
Control
9.58
9.19
32.5
-0.0102
0.7234
Core 4
10.56
10.12
32.5
-0.0116
0.7231
Control
10.56
9.98
32.5
-0.0160
0.7194
W. Wilson Bridge
Core 7
8.54
8.00
27
-0.0185
0.6609
Control
8.69
8.56
27
-0.0043
0.6805
Core 8
9.56
9.14
32.5
-0.0104
0.7232
Control
9.56
9.22
32.5
-0.0120
0.7228
-------
Table 10 (continued). Linear regression analysis of pH versus time of water over sediment cores
collected from the Potomac River, or water incubated without sediment.
Corr. Coeff
Location
Initial pH
Final pH
Time (h)
Slope
r
W. Wilson Bridge
Core 3
10.57
10.05
32.5
-0.0138
0.7213
(cont1d)
Control
10.57
10.13
32.5
-0.0120
0.7228
Hallowing Point
Core 1
10.03
9.50
6.25
-0.0799
0.8179
Control
10.03
9.51
6.25
-0.0781
0.8181
Core 3
10.07
9.62
6.25
-0.0676
0.8194
Control
10.07
9.57
6.25
-0.0773
0.8183
Mattawoman Creek
Core 1
9.97
9.54
32
-0.0140
0.7006
Control
9.99
9.56
32
-0.0132
0.7013
Core 4
10.13
9.61
32
-0.0163
0.6988
Control
10.16
9.65
32
-0.0158
0.6992
Indian Head
Core 1
10.04
9.62
32
-0.0128
0.7018
Control
10.06
9.79
32
-0.0082
0.7059
Core 4
10.04
9.66
32
-0.0114
0.7032
Control
10.05
9.70
32
-0.0111
0.7033
Smith Point
Core 1
10.02
9.63
32
-0.0111
0.7033
Control
10.01
9.78
32
-0.0066
0.7073
Core 3
10.13
9.53
32
-0.0186
0.6966
Control
10.12
9.73
32
-0.0118
0.7027
-------
CONCLUSIONS
The results of the experiments described here clearly
demonstrate that the sediment-water flux of phosphate from
Potomac River sediments between the Woodrow Wilson Bridge and
Smith Point with aerobic overlying water is a function of pH.
Depending on the location, the phosphate release from the
sediments begins to increase when the overlying water pH is in
the range of 9.0 to 9.5. There was considerable spatial vari-
ability in the rate of phosphate release at pH 10, although
there was no obvious difference in the maximum release inside,
compared to outside, the 1983 bloom area. The highest release
of phosphate at pH 10 was from Guns ton Cove sediments and
sediments from the mainstem of the river near Gunston Cove.
Sediments from those two locations also showed an increase in
phosphate flux at a lower pH than sediments collected near the
Woodrow Wilson Bridge or Mattawoman Creek. The amount of
phosphorus released from sediments within the bloom area at pH
10 (22 - 84 mg P nT^d"1) is similar to the amount of phosphorus
necessary (40-80 mg P m~2d"1; Thomann et al. 1985) to account
for the "excess" phosphorus in the bloom area. Thus, it is
plausible that the recurring algal blooms in this portion of
the Potomac River are due, in part, to the folloing positive
feedback mechanism. An initial increase occurs in the pH of
the aerobic water column to ~9 due to enhanced photosynthesis
caused by specific hydrological and meteorological conditions
(Thomann et al. 1985). This increased pH causes an increased
release of phosphorus from the sediments which in turn leads to
a further increase in photosynthesis. The pH of the water is
further increased due to the enhanced photosynthesis, sediment
phosphorus release increases, etc.
The release of ammonia from Potomac River sediments in-
creased with increasing pH; the uptake of nitrate by the sedi-
ments generally increased with increasing pH. Denitrification
is an active process (at pH 8) in Gunston Cove sediments and in
sediments from the mainstem near Indian Head. Denitrification
46
-------
appears to be an important sink for nitrogen. The amount of
nitrogen removed by denitrification at pH 8 (-200 pmol N
is equivalent to approximately 35% of the nitrogen
loading from point and non-point sources to this area of the
river.
A more thorough understanding of the processes involved in
the release of phosphate as a function of water column pH and
the maximum amount of phosphate that could be released from the
sediments requires the following studies: (1) the forms and
amount of phosphorus in Potomac River sediments, (2) the
relationship between water column pH and pH in the sediments,
and (3) the original source of the phosphorus released (past or
present inputs).
47
-------
LITERATURE CITED
American Public Health Association, American Water Works Asso-
ciation and Water Pollution Control Federation (APHA,
AWWA, and WPCF). 1981. Standard methods for the exami-
nation of water and wastewater. 15th ed. American Public
Health Association, Washington, DC. 1134 pp.
Andersen, J.M. 1971. Nitrogen and phosphorus budgets and the
role of sediments in six shallow Danish lakes. Arch.
Hydrobiol. 74:528-550.
Andersen, J.M. 1975. Influence of pH on release of phosphorus
from lake sediments. Arch. Hydrobiol. 76:411-419.
Andersen, J.M. 1982. Effect of nitrate concentration in lake
water of phosphate release from the sediment. Water Res.
16:1119-1126.
Bostrom, B. and K. Pettersson. 1982. Different patterns of
phosphorus release from lake sediments in laboratory
experiments. Hydrobiologia 92:415-429.
Breeuwsma, A. and J. Luklema. 1973. Physical and chemical
adsorption of ions in the electric double layer on he-
matite (a-Fe^Oa). J. of Colloid and Interface Sci.
43:437 -448.
Broderick, T.A. 1986. Phosphorus release from tidal embayment
sediments under high pH, aerobic conditions. Virginia
Polytechnic Institute, MS Thesis, 78 pp.
Callendar, E. and D.E. Hammond. 1982. Nutrient exchange
across the sediment-water interface in the Potomac River
Estuary. Est. Coast. Shelf Sci. 15:395-413.
Cerco, C.F. 1985. Effect of temperature and dissolved oxygen
on sediment-water nutrient flux. Report to the Environ-
mental Engineering Division NSF. College of William and
Mary, Gloucester Pt., Va. 105 pp.
Chang, C.S. and M.L. Jackson. 1957. Fractionation of soil
phosphorus. Soil Sci. 84:133-144.
Chen, Y.S.R., J.N. Butler, and W. Stumm. 1973. Kinetic study
of phosphate reaction with aluminum oxide and Kaolinite.
Envir. Sci. and Tech. 7:327-332.
Council of Governments (COG). 1984. Appendices to the upper
Potomac estuary-a report on water quality data for 1983.
Metropolitan Washington Council of Governments.
48
-------
Focht, D.D. and W. Verstraete. 1977. Biochemical ecology of
nitrification and denitrification. Adv. Microb. Ecol.
1:135-214.
Jacoby, J.M., D.D. Lynch, E.B. Welch and M.A. Perkins. 1982.
Internal phosphorus loading in a shallow eutrophic lake.
Water Res. 16:911-919.
MacPherson, L.B., N.R. Sinclair and F.R. Hayes. 1958. Lake
water and sediment III. The effect of pH on the partition
of inorganic phosphate between water and oxidized mud or
its ash. Limnol. Oceanogr. 3:318-326.
Mortimer, C.H. 1941. The exchange of dissolved substances
between mud and water in lakes (Parts I and II). J. Ecol.
39:280-329.
. 1971. Chemical exchanges between sediments and
water in the Great Lakes-speculations on probable regula-
tory mechanisms. Limnol. Oceanogr. 16:387-404.
Obihara, C.H. and E.W. Russell. ' 1972. Specific adsorption of
silicate and phosphate by soils. J. of Soil Sci. 23:
105-117.
Ohle, W. 1963. Kolloid Komplexe als Kationen-und Anionenau-
stauscher in Binnengewassern. Vom Wasser 30:50-64. (This
article could not be obtained; information from this
article was cited by J.M. Andersen, 1971 and was used for
this report.)
Seitzinger, S.P. 1983. Sediment-water phosphorus exchanges in
the Potomac River and embayments. Final Project Report
for Limno-Techn. Inc. 12 pp.
. 1985. The effect of oxygen concentration and pH on
sediment-water nutrient fluxes in the Potomac River.
Final Report for Maryland Office Environmental Programs.
19 pp.
Seitzinger, S., S. Nixon, M.E.Q. Pilson, and S. Burke. 1980.
Denitrification and N^O production in near-shore marine
sediments. Geochim. Cosmochim. Acta 44:1853-1860.
Seitzinger, S.P. S.W. Nixon, and M.E.Q. Pilson. 1984. De-
nitrification and nitrous oxide production in coastal
marine ecosystems. Limnol. Oceanogr. 29:73-83.
Solorzano, L. 1969. Determination of ammonia in natural
waters by the phenolhypochlorite method. Limnol.
Oceanogr. 14:799-801.
Stumm, W. and J.J. Morgan. 1981. Aquatic chemistry. John
Wiley and Sons, Inc., New York. 580 pp.
49
-------
Technicon Industrial Systems. 1977. Nitrate and nitrite in
water and seawater. Industrial Method No. 158-71W/A.
Technicon Industrial Systems, Tarrytown, New York. 4 pp.
Thomann, R.V., N.J. Jaworski, S.W. Nixon, H.W. Paerl, and J.
Taft. 1985. The 1983 algal bloom in the Potomac Estuary.
Report prepared for the Potomac Strategy State/EPA Manage-
ment Committee.
50
------- |