REPORT ON POLLUTION OF
THE MERRIMACK RIVER
AND CERTAIN TRIBUTARIES
part IV - Pilot Plant Study of
Ben thai Oxygen Demand
MASS.
U.S. DEPARTMENT OF THE INTERIOR
FEDERAL WATER POLLUTION CONTROL ADMINISTRATION
Merrimack River Project-Northeast Region
Lawrence, Massachusetts
August 1966
-------
REPORT OK
POLLUTION OF THE MERKTMACK RIVER
AND CERTAIN TRKER7CARIES
PART IV — PILOT PLANT STUDY OF BENTHIC OXYG35N DEMAND
by
Warren H. Oldaker
Alexis A. Burgum
Herbert R. Pahren
U. S. Department of the Interior
Federal Water Pollution Control Administration
Merrimack River Project
Northeast Region
Lawrence, Massachusetts
August, 1966
-------
TABLE OF COHTERTS
Page Ho.
IHTRODOCTION 1
MATERIALS AND METHODS 2
Description and Operation of the Pilot Plant 2
Water Supply and Sediments . 3
Chemical Analyses 6
Computations 6
EXPEKEMEHT&L RESULTS 9
DISCOSSIOW 10
SUMMAHT 13
CONCLUSIONS 13
KEEFEHERCI& ik
-i-
-------
LIST OF FIGURES
Pilot Plant Schematic 2
2 Area! Oxygen Demand of Recent and Aged Benthic
Sediments of 1.5 cm Depth 9
3 Area! Oxygen Demand of Aged Benthic Sediments. ... 9
k Calculated Iu versus Sediment Depth 9
5 Cumulative Area! Oxygen Demand 9
6 Comparison of Calculated and Observed Values of
L^ vith Recent Sediment of 1.5 cm Depth .... 9
7 Comparison of Calculated and Observed Values of
Ld vith Aged Sediment of 1.5 cm Depth 9
8 Comparison of Calculated and Observed Values of
L£ vith Aged Sediment of 10 cm Depth 9
9 Comparison of Calculated and Observed Values of
1^ vith Aged Sediment of 15 cm Depth 9
10 Comparison of Calculated and Observed Valves of
• Ld vith Aged Sediaent of 20 cm Depth 9
11 Variation of k^ vith Sediment Depth 9
12 Effect of nitrification on Oxygen Demand Rate
vith Recent Sediment of 1.5 cm Depth 9
13 Effect of nitrification on Oxygen Demand Rate
•with Aged Sediment of 1.5 cm Depth 9
-ii-
-------
LIST OF TABLES
Table Ho. Page No.
1 Sediment Type, Depth, and Volume at
Start-up of Pilot Plant. .......... k
Chemical and Biochemical Analyses of Sediments
at Start of Pilot Plant Run
-iii-
-------
INTRODUCTION
An estimate of the oxygea demand of the sediments which accumulate
In stream beds Is a practical necessity to the engineer engaged In the
Interpretation of oxygen levels In stream pollution surveys. Evaluating
this variable is a difficult task. Different empirical approaches for
estimating the sediment oxygen demand have been reported ' ' * ' .
Recently, a benthie respirometer has been developed **' to measure oxygen
levels over the sediments in situ. • . •
In this study, a pilot plant vas erected to operate under controlled
laboratory conditions using bottom sediments taken from different sites
in the Merrimack River bed downstream of certain municipal sever outfalls.
The sediments studied In the pilot plant include those solids which
normally would be removed in a primary settling device receiving raw
municipal sewage. The method outlined herein considered the variability
of the rate of biochemical assimilation of the residues and formulates
the findings in terms of the benthic equation of Camp * ' . In addition,
the effect of sediment depth on the area! oxygen demand of these sediments
was studied*
-1-
-------
MATERIALS AMD METHODS
Description and Operation of the Pilot Plant
The basic elements of the pilot plant used in this study were a
carbon filter (for chlorine removal), a sand filtration tank, galvanized
iron vater storage drums, a series of five 20-liter glass carboys con-
taining the sediments, and fire additional carboys to receive the effluent.
These elements vere connected by neans of glass and plastic tubing. A
schematic diagram of the pilot plant is given in Figure 1.
After passing through the sand and carbon filters, vater entered the
storage drums (vhich vere filled every day) and equilibrated at room
temperature (20-25°C). This is also the normal summer temperature in the
river from which the samples vere taken. Temperature of the vater in the
drums vas determined by a recording thermometer. From the storage drums
vater floved continuously through a manifold into the five sediment
carboys and from there discharged into the effluent carboys.
3be flow rate from each carboy vas Individually regulated twice a
day by means of screv clamps on rubber tubing. The rate vas set at
approximately 20 liters per day. The effluent volume vas determined each
2l4-hour period by measuring the height of vater in the effluent carboy
vith a yardstick. The actual volume vas then taken off a calibration
graph relating fluid height and volume.
-2-
-------
JL
9999
o
-hi
1
B
TAP WATER
FILTRATION TANK
2" FILTER SAND
2" GRAVEL
4" ROCK
WIRE EFFLUENT SCREEN
CARBON FILTER
ACTIVATED CARBON
GLASS WOOL
STORAGE TANKS
IN SERIES
TEMP. RECORDER
WATER LEVEL GAUGE
INFLUENT SAMPLE TAP
AIR TRAP
GAS TRAPS
SEDIMENT CARBOYS
R= RECENT
A=AGED
FLOW REGULATORS
EFFLUENT SAMPLE CARBOYS
PILOT PLANT SCHEMATIC
FIGURE
-------
-3-
Decomposition gases vere removed through gas traps which vere held
above the vater level in the storage tanks. Gases from the carboy with
10 en of sediment were analyzed after being measured in a Harvard manometer
in order to check for the presence or absence of anaerobic conditions.
Water Supply and Sediments
Tap water used in this study originated from the Merrimack River.
Raw river water was treated at the Lawrence, Massachusetts, municipal
plant by alum flocculation, activated carbon, sand filtration, aeration,
and chlorination.
Two types of sediments from the Merrimack River vere used. Although
they are physically and chemically characterized as shown in Tables 1 and 2,
arbitrary terms of "recent" and "aged" were selected to more generally
define them. Ideally, these sediments could be further characterized in
terms of their percent solids of sewage or non-sewage origin following a
method such as the organic analyses performed by Heukeleklan '5) . The
"recent" deposit was taken 100 yards downstream of a large outfall and was
assumed to consist largely of the municipal sewage sediments discharged to
the river. The "aged" sediments vere taken several miles downstream and
represented deposits in a more advanced stage of decomposition. Since the
rate of diffusion of oxidizable substances into the supernatant water,
rather than simply the sludge depth, is believed to control the rate of
oxygen demand of the sediment, no attempt was made in the field to collect
undisturbed sediment samples from different depths. Layering or sediment
depths are transient states affected by many variables such as stream
-------
hydraulics, and any reproduction in the laboratory of these layers of
sediments did not seem justified within the scope of this study.
Both sediments were nixed by impeller motor and by hand prior to
being poured into the carboys. The sediment type, along with the depth
and volume of deposits in the carboys, is given in Table 1. Physical and
chemical analyses of each sediment type were carried out at the beginning
of the study. These data are summarized in Table 2.
TABLE 1
SEDIMENT TYPE, DEPTH, AND 7GLBME
AT STAHT-0P OP PILOT PLAHT
Carboy Sediment Type
A recent
B aged
G aged
D aged
E aged
Depth (cm)
1.5
1.5
20
15
10
Volume (ml)
1,180
1,180
11,9*0
8,950
5,970
-------
TABLE 2
CHQGCAL AND BIOCHEMICAL ANALYSES OF SEDIMENTS AT START OF PILOT PLANT RUN
Residue, gins/50 ml
Sediment
recent
aged
Sediment
recent
aged
Sediment
recent
aged
pH % Moisture Specific Gravity
6.6 If0.5 1.6l
6.8 37.5 1.63
Total Fe HH3-H
mg/g Fixed Solids mg/g Total Solids ag/g
2.20 O.O2
3.04 0
Biochemical Oxygen Demand*, B
t, days: t » 2" t » 5
73.7 174
20.4 44.7
Total Volatile Fixed
47.70 i.j
50.75 I.1
OHG-N
Total Solids
3-30
0.86
ag 02/g_ Volatile
t •> 10
23L
7^.8
Ik 1*6.56
«) ^9.35
Total Mn
mg/g Fixed Solids
0.01
0.01
Solids
t * 15
21*
86.7
NOTE: *This test vas carried out by the dilution method given in Standard Methods (Eleventh Edition),
Part IV. The DO determinations vere made using the Wlnkler Method—azide modification.
-------
-6-
Chemical Analyses
Chemical analyses routinely employed in this study were: dissolved
oxygen (azide modification of Winkler method), nitrite, nitrate (phenol-
disulfonic acid method), ammonia (distillation procedure), organic nitro-
gen, and ferrous and total iron (o-phenanthroline method).
Dissolved oxygen determinations -were performed daily on the effluent
from the storage tanks and the effluent of each of the five sediment
carboys. Analyses for nitrite, nitrate, ammonia, organic nitrogen, and
ferrous and total iron were performed on 3- to 7-day composited samples
after acidification and storage at 10°C.
All methods and procedures used conformed to those listed in Standard
Methods, Eleventh Edition ^ .
Computations
To determine the sediment volume and to compute the surface area of
the sediments, it vas necessary to knov the inside radius of the glass
carboys. This vas determined by measuring the volume of water necessary
to bring the fluid level to a specific height. Solving for r In the
formula of a cylinder (V » TTr^h) gave a value of r a 13.8 em and a
surface area equal to 0.0596 square meters. A correction factor in the
sediment volume determination vas made for a slight irregularity in the
bottom of the carboy. The daily area! oxygen demand (BA.QD) vas computed
by using the following formula:
-------
-7-
MOD = Ve (DOi - DO ) 1
1000 A
where:
MOD « dally area! oxygen demand, gas Ojj/m /day
Ve « dally carboy effluent volume, liters/day
00.^ « dissolved oxygen of carboy Influent, mg/llter
D0e « dissolved oxygen of carboy effluent, ng/liter
A » surface area of deposit, 0.0596 square aeters
Camp's benthic equation ' ' includes the term L* which is defined
as the initial area! BOD of the bottom deposits at the start of decompo-
sition. This definition implies that L, is similar to the ultimate
ao
carbonaceous oxygen demand term L in the familiar BOD equation y »
L (1-10-**).
Using BOD data from Table 2, the rate constant was obtained by
/7\
Thomas's v ' ' graphical method. After solving for L in the BOD equation,
aad knowing the amount of the initial volatile solids and the volume and
surface area of the deposit, the initial ultimate areal oxygen demand LA
°o
is obtained.
L, - (IVS) (Vs) (L) 1 .......... 2
^o 1000 A
where:
LJ =• initial ultimate areal oxygen demand, gm/m2
IVS » initial volatile solids of the sediment, gm/ml
V8 a volume of sludge added to the carboy, ml
L - ultimate oxygen demand, mg/gm ITS
A = surface area of deposit = 0.0596 m2
-------
-8-
Evaluation of the tern Ld, the total areal BOD of the bottom deposits
in Camp's equation
............ 3
requires not only the value of the L^ but also the value of the areal
demand rate constant, k^.
Values of kj, for the various sediment depths were determined as the
slope b of the line, log y * log a + bx where y values were L^ values
obtained by pilot plant and z values were days after start-up of the
operation*
Experimental values of L^ were obtained as the difference between
the initial ultimate areal BOD, L*, and the cumulative areal oxygen
o
demand (!«<;)• This may be expressed as;
Experimental Ld = 1^ - LC • • • ....... ^
Finally, values of Ld calculated from Camp's equation and experi-
mental values of L obtained by pilot plant were plotted.
-------
EXPERIMENTAL RESULTS
The daily areal oxygen demands of the "recent" and "aged" deposits
at 1.5 cm versus time are shown in Figure 2, while similar data for the
"aged" sediments are presented in Figure 3* Flocculated material (probably
alum) in the storage drums was noted after 50 days. The pilot plant
operation was interrupted and the drums were cleaned.
A graph relating the values of the initial ultimate areal oxygen
demand with sediment depth is presented in Figure k. Thus, if all the
sediments of the same type would be completely oxidized, the oxygen
demand would be directly proportional to depth. Cumulative areal oxygen
demands for the five tests are indicated on Figure 5.
Comparisons of the calculated areal oxygen demand at different times,
using the value of k^ obtained from the slope of the line of best fit on
a semi-log graph, with the observed results are presented in Figures 6
through 10. These graphs show any deviations from the normal pattern.
The variation of k^ values with sediment depth is given in Figure 11.
Part of the oxygen demand attributed to the sediments was thought to
result from the oxidation of nitrogenous compounds. Graphs of the calcu-
lated oxygen demand derived from the oxidation of ammonia to nitrate are
presented in Figures 12 and 13, along with the observed daily oxygen
demand of the sediments. The nitrate-nitrogen oxygen equivalents were
based on the difference in nitrate-nitrogen values obtained on the
influent control water and the effluent. The chemical analyses were
carried out on composited samples. The observed oxygen demands are
based on averages of the dally demand for the same time period.
-9-
-------
DISCUSSION
The graphs of the daily areal oxygen demand in Figure 2 indicate
that, except for the initial phase in the recent deposit, both of the
1.5 cm sediments followed a roughly parallel course, increasing in oxygen
demand up to 75 days and then tapering off. The recent deposit consist-
ently shoved a daily demand of about 0.25 gm/nr/day higher than the aged
sediment. The daily areal oxygen demands of the deeper sediments in
Figure 3 shov some fluctuations but tend to remain in the fairly narrow
range of 0.7-1-0 gm/nr/day. There was practically no difference in areal
oxygen demand in the 15 and 20 cm depths.
One factor influencing the observed fluctuations in daily oxygen
consumption could have been the oxidation of nitrogenous materials. When
the oxygen demand believed to have resulted from nitrification was plot-
ted along with the observed oxygen demand (Figures 12 and 13), there was
a definite correlation in the location and magnitude of the peaks. In the
two shallow deposits, the nitrate-nitrogen oxygen equivalent approached
nearly 37 to kk percent of the total areal oxygen demand at the maximum.
Both of these maximums occurred 70 to 80 days after start-up of the tests.
Although nitrification appeared to play an important role in the
areal oxygen demand of the shallow sediments, it did not appear to be
significant in the deeper sediments. For example, the nitrate-nitrogen
oxygen equivalent reached a maximum value of 5 percent of the total areal
oxygen demand in the 10 cm depth.
-10-
-------
20
40 60
TIME, DAYS
80
100
AREAL 02 DEMAND OF RECENT AND AGED
BENTHIC SEDIMENTS OF 1.5cm DEPTH
FIGURE 2
-------
20CM.
I5CM.
IOCM.
I.5CM.
20
40 60
TIME, DAYS
80
100
AREAL © DEMAND OF AGED
BENTHIC SEDIMENTS
FIGURE 3
-------
AGED SEDIMENT
5 10 15
SEDIMENT DEPTH, CM
20
CALCULATED LH
vs. d°
SEDIMENT DEPTH
FIGURE 4
-------
Od
UJ
Q
CVI
UJ
or
<
UJ
>
p
<
_l
o
5
D
o
I5CM
RECENT.
40 60
TIME .DAYS
80
100
CUMULATIVE AREAL 02 DEMAND
FIGURE 5
-------
200
(
100
OJ
5
X
in
5
13
| 60
<
2
u
o
0^40
O
I-
H
Z
LJ
03
20
10
<
»
•
•
v^
^>«
^1
">•
"«<^
^^
• OBSERN
— CALCUL
k4=0.00
i
^
^^
'ED DATA
ATED
51
•
•
•
<
'
*
•
•
(
»
D 20 40 60 80 100
TIME ,DAYS
COMPARISON OF CALCULATED AND
OBSERVED VALUES OF Lj 1.5cm
RECENT SEDIMENT
FIGURE 6
-------
100
80
60'
40
CVJ
5
"20
o
cf
z
g 10
cvi 8
O
o 6
I
l-
LL) 4
OQ
2
1
C
• OBSER^
— CALCUl
k^O.OC
*T
""•>-*•
* »^
<
^ED DATA
.ATED
33
*
•
•
•
<
)
•
•
•
4
1
) 20 40 60 80 100
TIME ,DAYS
COMPARISON OF CALCULATED
AND
OBSERVED VALUES OF Ld
1.5 CM AGED SEDIMENT
FIGURE 7
-------
600
200
0
40 60
TIME , DAYS
80
100
COMPARISON OF CALCULATED
AND
OBSERVED VALUES OF Ld
10 CM AGED SEDIMENT
FIGURE 8
-------
900
300
0
20
40 60
TIME, DAYS
80
COMPARISON OF CALCULATED
AND
OBSERVED VALUES OF Ld
15 CM AGED SEDIMENT
FIGURE 9
-------
900
300
40 60
TIME , DAYS
100
COMPARISON OF CALCULATED
AND
OBSERVED VALUES OF Ld
20 CM AGED SEDIMENT
FIGURE 10
-------
0.006
0.005
O.OO4
0.003
O.002
0.001
iRECENl
AGED
8 10 12
SEDIMENT DEPTH,CM
14 16 18 2O
K4vs SEDIMENT DEPTH
-------
o 1.0
E
o>
u 0.8
cr
o
z
<
5
UJ
Q
20
40
60 80
TIME (days)
100
EFFECT OF NITRIFICATION ON OXYGEN
DEMAND RATE OF 1.5cm RECENT SEDIMENT
120
FIGURE 12
-------
20
40 60 80
TIME (days)
100
120
EFFECT OF NITRIFICATION ON OXYGEN
DEMAND RATE OF 1.5cm AGED SEDIMENT
FIGURE 13
-------
-11-
Perlodlc release of gases of anaerobic decomposition from the deeper
sediments indicated a reducing environment. This was confirmed by analysis
of gases from the 10 cm sediment depth. These gases may have moderated
nitrification to the extent shown in the minor fluctuations of the areal
oxygen demand rate in Figure 3*
The plot of the cumulative areal oxygen demand of the aged deposits
(Figure 5) shovs that there is a fairly marked increase in demand with
sediment depth.
Consideration of Figures 6 through 10 indicates that in the deeper
sediments (10 to 20 cm) the benthic oxygen demand Lg as calculated from
the benthic equation closely approximates the experimental values of LJ
throughout the test period.
In the 1.5 cm "recent" sediment, the benthic equation closely
approximates the benthal demand obtained only between 15 and 55 days.
For the first 15 days, the high values of Ld may have been due to the
oxidation of readily biodegradable sevage solids prior to the development
of benthic conditions. After 55 days, nitrification may have been the
cause of the Increasing rate of demand (Figure 12).
In the 1.5 cm "aged" deposit, the benthic equation closely approxi-
mates the benthic demand obtained in the pilot plant only for the first
ko days after vhich nitrification may have been the cause of the increas-
ing rate of demand (Figure 13).
-------
-12-
The effect of sediment depth is shown clearly in Figure 11 vhere
the benthic rate constant k^ decreases markedly vith an Increase of
depth up to about 10 cm* Above a depth of 15 cm, however, practically
no decrease in k^ vas observed. In the recent deposit, the presence of
more biodegradable solids probably resulted in the higher value of k^.
A decrease of k^ vith an increase in sediment depth vas also observed by
Fair et al ' ' . The change observed in the present study vas much acre
pronounced*
A comparison of Figures k and 5 indicates that after 100 days only a
saall fraction of the initial ultimate oxygen demand had been exerted in
the deeper sediments. This observation is similar to that of Fair vho
found that the ultimate oxygen demand vas not approached until after a
year*
The kj,. values obtained in this study are believed to be applicable
to other deposits if such deposits are of similar origin (primarily
domestic sevage) and if the BOD as determined by the Vinkler method is
in a similar range.
-------
SUMMARY
The areal oxygen demand of bottom sediments taken from the Merrlmack
River in Massachusetts -was determined by a small pilot plant similar in
design to that of Fair'et al (®) . Parameters in the benthic rate equa-
tion of Camp (**•) vere evaluated on the basis of the data obtained, arid
the effect of sediment depth on the benthic rate constant k^ was studied.
Chemical analyses of the influent and effluent water were carried out.
The effect of nitrification on the oxygen demand is discussed.
CONCLUSIONS
1. The value of the benthic rate constant k^ varies with the age and
depth of the deposit.
2. A marked decrease of kj, with increase in sediment depth occurred
between 1.5 and 10 cm. Above 15 cm no significant decrease in k.
was observed.
3. Only the upper 15 cm of sediment had any significant effect on the
areal oxygen demand.
k. The observed data were closely approximated by the equation Ld =
Lji 10"^ at all sediment depths except the 1.5 cm depth.
5. Nitrification was believed to play a role in the oxygen demand of
the sediments and was especially significant in the shallow depths
studied.
-13-
-------
REFERENCES
1. Tsivoglou, Ernest C., "The Significance of River Sludge Deposits,"
M. S. Thesis, University of Minnesota (19*18).
2. Benedict, Arthur Hove, "The Effects of Benthal Deposits on Streams
and Lakes," Thesis, Tufts University, Massachusetts (1965).
3. O'Connell, Richard L., "An In-Situ Benthic Respirometer," CB-SRBP
Technical Paper No. 6, FWPCA, Region III, U. S. Department of
Health, Education, and Welfare, Charlottesville, Virginia (1966).
If. Camp, Thomas R., "Water and Its Impurities," Pages 299-302,
Reinhold Publishing Corporation, New York (1963).
5. Heukelekian, H., and Balmat, J. L., "Chemical Composition of the
Particulate Fractions of Domestic Sewage," Sewage and Industrial
Wastes, 3JL, k, U13-423 (April, 1959).
6. Standard Methods for the Examination of Water, Sewage, and Indus-
trial Wastes, Eleventh Edition, American Public Health Association,
Inc., New York, I960.
7. Thomas, Harold A., Jr., "Graphical Determination of BOD Curve
Constants," Water and Sewage Works (March, 1950).
8. Fair, Gordon M., Moore, Edward W., and Thomas, Harold A., Jr.,
"The Natural Purification of River Muds and Pollutional Sediments,"
I-III Sewage Works Journal 13_, 2, 270-307 (March, 19^1) and
IV-V Sewage Works Journal 13, k, 756-779 (July,
------- |