U.S. ENVIRONMENTAL PROTECTION AGENCY
MIDDLE ATLANTIC REGION-III 6th and Walnut Streets, Philadelphia, Pennsylvania 19106
-------
"EPA-903/9-73-009"
THE BOD5/DO RATIO
A NEW ANALYTICAL TOOL
FOR WATER QUALITY EVALUATION
NORMAN W. MELVIN
Information Systems and
Q8£:Ci Analysis Branch
°l 39 O Surveillance & Analysis Division
and
RALPH H. GARDNER
ADP Support Branch
Management Services Division
Region III
Environmental Protection Agency
Philadelphia, Pennsylvania
-------
Table of Contents
Chapter Page
I Introduction 1
II BOD evaluation 2
III DO evaluation 3
IV Theory and Methodology 4
A. Theory 4
B. Methodology 5
V. Examples of BOD5/DO ratio in practice 8
A. Cleanwater stream - Jackson River 8
B. Clean water stream - White Oak
Creek 11
C. Partially degraded stream -
Susquehanna River 12
D. Partially degraded stream -
Missouri River 14
E. Degraded stream - Potomac River 15
F. Severely graded streams 16
G. Freshwater/esturine streams -
Delaware River 17
H. Specialized uses 19
VI. Stream classification using BODj/DO ratio 19
VII. Computer program (STORET) for processing
BOD /DO data 21
Summary 22
-------
Appendix
A. Computer program listings
1. BOD-/DO ratio
2. Statistical analysis
-------
I. Introduction
The basic definition of the biochemical oxygen demand (BOD)
is usually given as the amount of oxygen required by bacteria
while stabilizing decomposable organic matter under aerobic
conditions (1). This determination of BOD has been, and continues
to be, widely used to evaluate the oacygen-consuming strength of
municipal and industrial wastes being discharged into receiving
streams. Dissolved oxygen (D.O.) in stream waters is a measure
of the oxygen available for bacterial consumption in stabilizing
organic material. These two parameters are used by Sanitary
Engineers in determining oxygen-sag curves, Deoxygenation and
Reaeration rates as well as the ultimate oxygen consumption
load imposed on the stream.
In the past, the object of such analyses has been to ascertain
the various rate functions of the BOD and D.O. curves. This paper,
however, departs from such an approach and simply uses the two
parameters to determine the ratio between them at each sample site
and for each sample taken. This method is dependent upon two
basic premises. These are: 1) That the BOD test at the sample
station is reproducible within the normal limits of error for
the test and free from interfering substances and 2) That the
1. Sawyer, Clair N., Chemistry for Sanitary Engineers, McGraw-Hill,
New York, 1960 p.270
-------
period of record extends across several annual cycles on a monthly
sampling frequency at a minimum.
When a sufficient amount of data points have been accumulated,
then a linear trend for the data set can be established. If
this linear trend is horizontal (m=o) then the system is maintaining
itself as a steady state condition of dynamic equilibrium. A
condition of y=(mx+b) where m^o is unacceptable in that it
demonstrates a steadily degrading water quality condition in the
stream. A secondary method of analyzing the data is to examine
the fluctuations about the mean on both a daily and annual basis.
If the standard deviation value is small, then the stream is
demonstrating a normal situation. If these values are large,
then the stream is "nervous" indicating potentially unfavorable
conditions in the stream at the sample station.
This paper will demonstrate the postulations given above and
illustrate the various situations actually occurring at stream
stations across the conterminous United States.
II. BOD Evaluation
The 5-day BOD test (BOD ) repressats that amount of oxygen
demand exerted by biochemical activity over a 5-day time span.
This test is not fool proof and is subject to numerous sources of
-------
error. An attempt to rely too heavily upon the BOD values alone
must not be made. Instead, the BODc value should be used in
comparison with some other parameter; in this case the D.O.
value.
Because the kinetics of the BOD reaction are such that
approximately 70 to 80 percent of the ultimate BOD load value is
obtained in 5 days, it is logical to assume that a "safe"
ultimate BOD value in normal streams would be a value less than
the D.O. content. Therefore, a safe ratio between the BOD value
and the D.O. value measured at the same time at the same station
should be not more than 0.3 for normal streams. Such a value would
allow for the additional load imposed by chemical oxygen demand
as well as the added load from nitrification and still remain
less than the amount of D.O. present. Put in another way, it means
that the total debt (BOD loading) remains less than the reserve
(D.O. content) and one cannot, therefore, be "over drawn at the
bank".
III. D.O. Evaluation
The dissolved oxygen (D.0.) content of any given stream is
the sum- of a large number of interacting variables. Some of
these are inorganic, mechanical factors such as barometric pressure,
temperature, channel geometry and flow velocity. Others are
either directly organic as in the case of algal 02 production or
indirectly organic as in the case of nutrients such as viftamins, nitrates,
phosphates and trace elements.
-------
Streams with steep gradients, high flow velocities, turbulent
flow and sediment-free, cold waters can be expected to assimilate
a greater loading of BOD than sluggish, meandering, low velocity
warm water streams. The level of nutrient is very important
because excessive nutrients allows an exaggerated fluctuation of
the diurnal D.O. curve. Such a condition is seen in the Tamiami
Trail area of Florida (2) where the D.O. concentration drops to
approximately zero each night under the influence of algal
respiration and rises to super-saturated values under energetic
photosynthetic activity of the plant community during the following
day. Under these circumstances fishkills can occur even under
"normal" stream conditions.
IV. Theory and Methodology
A. Theory
The BOD test should not be used in the exclusion
of other supporting tests such as total organic carbon (TOC)
or chemical oxygen demand (COD) due to the number of error sources
inherent in the BOD method. For the purposes of this paper, how-
ever, the most useful supporting tests for analytical purposes
appear to be total chlorophyll, total nitrates and total phosphates.
If long-term trends are required, the ratio of BODc/DO alone
may be plotted without need to refine the stream ecological
reactions further.
2. Little, John A., Robert F. Schneider and Bobby J. Carroll, A
Synoptic Survey of the Limnological Characteristics of the Big
Cypress Swamp, Florida, FWQA, Southeast Region, May 1970.
-------
STORE!, the Environmental Protection Agency's data storage
computer system, contains a large mass of data bits acquired at
various stations across the country. In order to evaluate the
validity of the BOD,./DO ratio as an indicator to stream health
in all types of streams, a computer program can be written to
retrieve BOD,- and DO data from each sample station, then to
divide D.O. into BOD5 and print the result. In addition, a
supplementary statistical linear regression subroutine program
is available to determine the slope of the linear trend of a
moving average plot. A value of m^o in the expression y=mx+-b
means the stream is either maintaining a dynamic equilibrium
(m=o) or becoming cleaner m
-------
plankton and rooted aquatic communities generates a diurnal
dissolved oxygen fluctation. The magnitude of the fluctuation
is dependent upon temperature, available nutrients, solar radiation
and type of stream flow. If the stream contains excess nutrients
and is sluggish, then the dissolved oxygen curve varies greatly
between the depressed D.O. levels seen during the evening hours
and the supersaturated conditions occurring during peak daylight
production (Figure 1). The condition for a normal, clean stream
is one of much less daily fluctuation of the dissolved oxygen
curve as also shown in Figure 1. Care must be exercised, however,
in the analyses of curves to account for the fluctuations in the
insolation rate caused by cloud cover.
The production of supersaturated dissolved oxygen conditions
also indicates large amounts of biodegradable material being
generated. This biomass, generally algal material, imposes a
biologically-derived oxygen demand upon the stream. Therefore,
samples taken during winter low temperature conditions should
reflect biological activity minima. If the BOD^/DO ratio diminishes
only slightly during winter months, then the greater proportion
of the observed BOD loading can be attributed to either man-made
or non-point source pollution. Under extreme conditions of very
low temperatures and frozen ground, the observed BOD loading will
be almost entirely man-made because input from all other sources
has been reduced essentially to a minimum. Conversely, a large
summer/winter BOD5/DO fluctuation would indicate the major portion
of the BOD load to be derived from biological activity and the
problem in the stream is one of the excess nutrients rather than
6
-------
Cose I Normal Stream
COM II Cutrophic Str«am
0600
1200
1800
2400
FIGURE I : Diurnal oxygen fluctuation curves for normal
versus eutrophic stream conditions
-------
municipal or industrial pollution. This supposition may be verified
by analyses for nutrients and chlorophyll. Lastly, there will be
a period when the BODc/DO ratio will increase toward unity thus
showing the normal upsurge of algal activity which occurs each
spring. A smaller peak may be detected in late summer for the same
reason. Figure 2 illustrates the conditions described above. In
addition, a stream which has been, and continues to be degraded
will show only minimal seasonal variation in the BOD5/DO ratio
but will show marked peaks and troughs reflecting the effect of
the influx of cleaner overland runoff from precipitation.
If a stream exhibits a BODc/DO ratio which is that of a clean
water system but which lacks normal bio-logical assemblages, then
the problem may be attributable to toxicity problems from heavy
metals and/or organic chemicals. The situation where the BOD5/DO
ratio occasionally drops to a very low value (BOD /DO £0.03)
seems to be due to suppression of the K-^ rate which is, in turn,
related to the inhibiting action of tannins and lignins. These
factors present special problems in assessing a stream BOD^ value
and tests for such materials should be performed during the early
spring and late fall to insure that a localized concentration of
such material (runoff from plant and tree litter) does not interfere
with the BODc test. An examination of the industrial discharges
in a given stream basin will readily identify tannins and lignins
-------
Cose I Stream polluted by man-mod*
discharges
Case II Eutrophic stream
Case III Clean stream
WINTER
SPRING
SUMMER
FALL
FIGURE 2 : Theoretical annual stream curves plotted from
BODC/DO ratios
-------
from such man-produced sources as pulp and paper mills.
V. Example of the BOD5/DO ratio in practice
A. Clean Water Stream-Jackson River above Covington, Virginia
This stream rises in a remote wooded area of the Virginia-
West Virginia boundary some 80 miles west of Charlottesville,
Virginia near Hot Springs, Virginia (Figure 3). The area is
undeveloped with a population density between 6 and 10 persons
per square mile. These conditions present an opportunity to observe
a clean-water stream system responding to naturally occurring
seasonal changes.
During 1971 and 1972, water samples were taken on a monthly
basis on the Jackson River above Covington, Virginia. These
samples, taken for a pre-impoundment water quality survey, were
analyzed for BOD and DO plus other water quality parameters. A
summary of the range of BODe/DO ratio data for the various stations
is shown on Table 1. Preparation of the final report for this
survey revealed that there seemed to be a general correlation
between the BOD /DO ratio at each station throughout the period
of study. The attempt to explain this apparent sameness led to
the investigation of the ratio in other areas of the country and
has culminated in the preparation of this report.
-------
* r Gothright
r V/Rt$»rvoir
CHARLOTTESVILLE
COVINGTON
N
0 5 10 20 30 40
FIGURE 3: Location map, Gathright Reservoir
Jackson River, Virginia
-------
f : I ! r
f 1 f ! I ! I I I 1 I 1 i 1 I 1 f !
Table 1. Jackson River BOD5/DO ratios above Covington, Virginia
7/8/71
10/6/71
11/18/71
12/13/71
2/24/72
4/6/72
5/4/72
6/8/72
8/1/72
9/8/72
Back Creek nr.
Mt. Grover. Va.
BOD5
4.5
1
1.1
0.8
1
—
1
1
1
1
DO
8.8
9.1
11.4
10.9
11.2
10.8
9.7
8.5
9.5
9.2
RATIO
0.511
0.110
0.096
0.073
0.089
—
0.103
0.118
0.105
0.109
Jackson River nr
Bacova, Va.
BOD
1.3
1
1.3
1.4
1
__
1
1
1
1
DO
9.3^
9.4
11.5
11.0
12.3
11.8
9.2
9.7
9.3
9.7
RATIO
0.140
0.106
0.113
0.127
0.081
-T
0.109
0.103
0.042
0.103
Jackson River @
Kelly Bridge
BOD
1.3
2
2.8
0.8
1
-f
1
1
1
1
DO
9.5
8.9
11.0
11.0
12.4
10.9
9.3
8.8
9.5
9.1
RATIO
0.137
0.225
0.255
0.073
0.081
—
0.108
0.114
0.105
0.110
Jackson River @
Natural Wells, Va.
BOD5
1.0
1
1.2
1.2
4
—
1
1
1
1
DO
10.3
9.6
11.3
11.3
12.7
12.2
9.2
9.4
9.3
9.3
RATIO
0.097
0.104
0.106
0.106
0.315
—
0.109
0.107
0.108
0.108
Cedar Creek nr
Callison Va.
BOD5
__
1
1.1
0.8
1
—
1
1
1
1
DO
_—
10.^
13. e
11. (
12.]
12.4
10.1
10.7
9.5
10.2
RATIO
— _
0.096
0.080
0.072
0.082
—
0.099
0.093
0.105
0.098
-------
Perhaps the fundamental consideration of water quality changes
along a river course is to reflect that the observed water quality
at any given point is the sum total of all of the changes and
influences acting on that stream from the headwaters to the point
of sample collection. These influences may be naturally occurring
factors such as topography, vegatation, soil cover, bedrock, rain-
fall, temperature, and stream geometry. Man-induced changes are
concerned with farming practices, suburban development, highway
construction, forest work, and industrial activity. The stream
water quality of any given stream is, at its headwater source,
generally free from pollution and contains a very low to non-
existant quantity of D.O. (ground water is essentially lacking
in dissolved oxygen). As the water flows downstream under the
influences of gravity, changes begin to occur. Such changes are
related to bedrock minerals, soil cover, foliage and channel
geometry. Organisms grow and die in these waters thereby adding
and subtracting oxygen, carbon dioxide and nutrient materials in
an endless ecological cycle. Because most streams flow continuously,
these changes become cumulative with time and each point along the
stream has its own more or less unique biochemical water quality
make-up. Further, because nutrients are being continually swept
downstream by the flowing water, a stream is generally nutrient
deficient unless large inputs of nutrients are derived from man-made
-------
sources such as sewage outfalls or non-point source runoff. When
the stream gradient diminishes and stream velocity decreases,
then conditions tend to become more unstable because of the
reduced self-purification potential, and finally, in estuaries,
there may be a build up of nutrients and an increase in ecological
diversity.
Therefore, a stream segment in a remote, thinly-populated area
should be carrying minimum BOD material and maximum dissolved
oxygen. The Jackson River fits such criteria and an examination
of the BOD^/DO ratio for all stations reveals a ratio of about
1:10. This ratio seems to be about the normal range for clean
mountain streams in the Blue Ridge area with the summer ratios
being higher than the winter values. Further, because the stream
is clean and pollution-free, the deviations of any given sample
from the 0.100 level are generally minimal. This reflects a
situation where overland runoff from rainfall is carrying essentially
the same BOD concentration as those waters in the receiving stream.
Under circumstances such as these, no dilution of BOD concentration
occurs in the stream subsequent to a rainfall event and the BOD,-/DO
concentration remains close to the 0.100 level over most of the
year. Further, because a clean stream carries only a limited
quantity of nutrients, no sharp upsurge of algal growth occurs
10
-------
in the spring. This condition results in little, if any,
fluctuation in the BOD /DO ratio over the period when algal
blooms could occur.
B. Clean water stream in farming area-White Oak Creek, N. C.
White Oak Creek near Bells, North Carolina is a small
tributary to the New Hope River which is, in turn, a tributary
to the Haw River. The area drained by the White Oak is essentially
open agricultural land with no present industrial or municipal
discharges above the point of sampling (Fig 4). Water quality
data indicate the stream is enriched in nutrients from agricultural
sources and this enrichment allows upsurges in algal populations
from time to time. A study made for the US Army Corps of Engineers
prior to the construction of the New Hope Reservoir revealed that
this stream, although receiving no direct effluent from either
municipal or industrial sources, still exhibited dissolved oxygen
stress from time to time. This stress was observed to coincide
with a rise in chlorophyll and turbidity values along with a
marked decrease in nutrient values under the influence of a rapidly
expanding algal population. Because of this algal upsurge, the
BOD /DO fluctuation for this stream differs from that observed
for the Jackson River in the previous section. The White Oak
BOD,./DO trare seems to approximate the curve shown in Figure 2
for a eutrophic stream but the fluctuations shown are more extreme
11
-------
HAW RIVEi
NEW HOPE RIVER
N
APEX
VH/TE OAK CREEK
NEAR BELLS, N.C.
FIGURE 4: Location map, White Oak Creek near
Bells, North Carolina
-------
(Figure 5). If a condition such as shown in Figure 5 is allowed
to progress in slowly moving water, a situation arises where
the oxygen content of the water is supersaturated during daylight
hours but nearly depleted under strong respiratory action during
the night. Examples of this type of reaction are observed
in the Florida Everglades and would be the case in White Oak
Creek were this stream impounded for any reason.
9
Because theee is an active current in the White Oak, the
upsurge of algae only occurs during the early summer when conditions
reach optimum levels of reproduction, allowing the population of
algae to maintain itself in spite of being continually swept
downstream. The BOD /DO ratio in White Oak Creek was observed
to fluctuate over a wide range of values and the low values
recorded in early 1968 and 1969 (winter conditions) indicate that
the high ratios of the summer months are due to algae activity
rather than a constant base line of municipal or industrial
BOD loading.
C. Partially degraded stream receiving municipal effluents
from upstream sources (low sediment yield area)
The main stem Susquehanna River at Berwick, Pennsylvania is
an example of a stream partially degraded by municipal effluents
from upstream sources; in this case from the Wyoming Valley populations
12
-------
0.6
0.5
0.4-
o
"9 0.3
ur
Q
O
CD
0.2
O.I
1968
1969
1970
FIGURE 5: BOD5/DO fluctuations at White Oak
Creek at Bells, North Carolina
-------
centers of Scranton and Wilkes-Barre, Pennsylvania (Figure 6).
The Susquehanna River at Berwick also contains contamination from
acid mine drainage but this effect does not modify appreciably
the BOD^/DO ratio observed. In addition, these data shown in Figure
7 extend through the period of maximum flood imposed by Hurricane
Agnes rains. The data shown in Figure 7 were taken at weekly intervals
for inclusion in an Environmental Impact Statement on the Pennsylvania
Power and Light Company nuclear generating plant at Berwick. The
effect of the flooding was to render the Wilkes-Barre sanitary
treatment plant inoperative and the effluent to be discharged
into the river without treatment save for chlorination. The peaks
observed in September and November 1971, prior to the flooding may
be due to periods of maintenance and/or poor operation on the
upstream plants. These peaks may also be due to flushing of the
bed load by higher flows runoff. In the case of the Susquehanna
River, the ratio of effluent to stream flow is small enough to
assign the fluctuations observed to the effect of scouring
bedload, rather than to treatment plant effluents, even if the
effluent is essentially untreated following plant shut down. The
case for BOD loading increases from overland runoff is strengthened
by the observation that large volumes of water create the illusion
of a clean stream (note the precipitous drop in the BOD /DO ratio
13
-------
N
0123456789
WILKES-BARRE
ISO
SAMPLE STATION
FIGURE 6: Location map, Susquchanna River
near Berwick, Pennsylvania
-------
0.5
0.4
Q 0.3
O
O
GO
0.2 ^
O.I
•p Hurricane Agnes rains, Wilkes-Barre
[_STP flooded out; row sewage effluent
to Susquehanna River
0 'Sept'Oci 'Nov'Dec
1971
Jan Feb Mar Apr May June July Aug
1972
FIGURE 7: BODs/DO ratio for Susquehanna River at Barwick, Pennsylvania
-------
subsequent to the Influx of Hurricane Agnes rains as shown
in Figure 7). Further, a partially degraded stream seems to be
much more sensitive to this type of condition than a clean stream
as shown by the narrow range of fluctuation in the Jackson River in
Case A.
Case D. Partially degraded stream receiving municipal effluents
from upstream sources (high sediment yield area)
The Missouri River at Missouri City, Missouri is an example
of a stream receiving municipal and industrial effluents as well
as large amounts of sediment from eroding upstream areas. Such a
stream does not support expected algal concentrations for the
nutrient levels available due to suppression from both organic
(sewage) and inorganic (industrial effluents and silt) sources.
In the case of silts, however, the turbulence of the flow would
allow vertical mixing with the net result that the stream plankton
would achieve a steady-state production rate at some reduced
volume proportional to the concentration of the sediment present.
This net effect is to show mainly the municipal, industrial and
non-point source effluent BOD loading with a lesser loading input
from the plankton community. The sampling station (Figure 8)
is located downstream from the Kansas City area and shows the very
slight downward trend of the over-all trace of the BOD /DO curve.
14
-------
KANSAS CITY
MISSISSIPPI RIVER
0 50 100 ISO 200 250
SCALE OF MILES
FIGURE 8: Location map, Missouri River at
Kansas City, Missouri
-------
1964 ' 1965 ' 1966 ' 196? ' 1968
FIGURE 9: BODs/DO ratio trace for Missouri River, Missouri City, Missouri
1969
1970 ' 1871 ' 1972
-------
Such a trend is indicative of successful environmental clean-up
efforts, even though the total amount of ratio reduction is quite
small (Figure 9).
Case E. Degraded stream draining low sediment yield area
The Potomac River at Great Falls, Maryland (Figure 10) is an
example of a stream receiving municipal and industrial effluents,
from upstream sources, in quantities sufficient to cause a general
and continuing deterioration in the water quality at the observation
station. Much attention has been given to clean-up efforts of
the Potomac River water quality in the past but the stream water
remains in very poor condition (Figure 11). The trace of the BOD-/DO
ratio indicates a deteriorating condition with regards to water
quality since early 1968. Some improvement was noted in 1970 but
this gain was lost in 1971 with a return to unstable, oscillating
ratios.
However, an examination of the ratio trace indicates a strong
influence of BOD loading from plankton and rooted aquatics upstream
of the sample site due to optimum available nutrients. This
premise is based upon the climb toward unity shown during the
actively growing period which occurs through the warm-water summer
months. When cooler conditions prevail, the biological activity
declines and the apparent ratio approaches that observed for clean
15
-------
CABIN JOHN
GREAT
FALLS
DIFFICULT RUN
Maryland
ROCK CREEK
POTOMAC
RIVER
Washington, D.C.
FIGURE 10: Location map, Great Falls Potomac
River near Washington, D.C.
-------
13
\2
I.I
1.0
0.»
08
0.7
0.6
0.5
04
03
OZ
O.I-
0
1958
1959
I960
1961
FIGURE II: BODs/DO ratio trace. Great Falls Potomac River near Washington, DC.
-------
I ! t 1 I I f
f 1 f 1
f 1 f 1 I ! I I f 1
196Z ' 1863 ' ld«4
FIGURE II (continued): BODs/DO nitio trace, Great Falls Potomac River near Washington, D.C.
1965
-------
1966 ' 1067 ' I960
FIGURE fl (continued): BODs/DO ratio trace, Great Foils Potomac River near Washington, D.C
1969
-------
f i i i i i f i f i ? i ? i f ?
i r
f i i i i i i i i i f i i i f i
1970
1971
FIGURE II (continued): BODs/DO ratio-rrace. Great Falls
Potomac River near Washington, DC.
-------
streams. The problem, therefore, seems to be one of excess nutrients
derived from both sewage effluents and non-point inputs upstream.
Several very low values observed over the period of record are
taken as being examples of tannin and lignin inhibition of the BOD
curve. These materials are derived, it is believed, from effluents
being discharged from pulp and paper mills in the upper watershed
near Keyser, West Virginia and Cumberland, Maryland. The tannins
and lignins have the effect of delaying the assimilation curve rise
so that erroneously low values of BOD are recorded at the termination
of a normal five day test.
The basis for assigning a "degraded" label to the stream is the ob-
servation of BODc/DO ratios in excess of 0.6 and approaching 0.8
in some instances. The level of degradation in the Potomac River
is not excessive, however, when compared to those streams where the
BOD5/DO ratio is very much larger than 1.000. AT any rate, a BOD5/DO
ratio of 0.500 with a D.O. content of more than 4.0 mg/1 may still
be considered a degraded stream in that complete assimulation of
the BOD concentration would theoretically lower the D.O. content
below the recommended (3) lower limit for flowing streams.
Case F Severely degraded stream
No example of a severely degraded stEeam is given inasmuch as such
streams are readily identifiable by means other than the BOD /DO
ratio. Candidates for such dubious honors would be the Cuyahoga
River at Cleveland, Ohio, River Rouge at Detroit, Morgan Creek near
3) Water Quality Criteria, NTAS, Federal Water Pollution Control
Administration, April 1, 1968, Table II-l, p.20.
16
-------
PENNSYLVANIA
Rivtr-
Schuy/h'11 River
RMI 122.49
Roneoeos Rivtr
Brandywin* Crttk
RMI 110.70
PHILADELHIA
RMI
CHESTER
1NGTON
DEL.
RMI 70.96
RMI 60.55
NEW JERSEY
D*/awar* Bay
10
0
.1
10
i
20
i
30
i
SCALE OF MILES
40
i
FIGURE* 12: Sample station locations Delaware
River Basin Commission
-------
2.5-
2.0-
.2
1?
1.5-
m
Q
O
CD
1.0-
0.5-
50
Grossly degraded zone
on Delaware River
Summer low tide
slack
7-6-67
miner high tide
slack 8-1-61
Winter tow tide slack
12-4-67
60
Winter high tide slack
12-13-67
70 80 90 100
RIVER MILES ABOVE MOUTH
no
120
130
FIGURE 13: BODs/DO traces Delaware River between Fieldsboro, New Jersey
and Pea Patch Island, Delaware
-------
Chapel Hill, North Carolina, South Buffalo Creek at Greensboro,
North Carolina and the Lackawanna River at Scranton, Pennsylvania.
Criteria for a severely degraded stream would be one in which
the D.O. concentration was consistently 1.5 mg/1 or less with a
BOD of at least 6.5 mg/1. On the biological side, only a few
very pollution-tolerant species such as Tubificids, Chironomids
and air-breathers such as the pouch snail, Physa, will be present.
Under extreme conditions, all macroscopic forms are absent and only
bacterial forms, mainly anaerobes, will be found in samples.
An example of such a stream would be the Codorus above York,
Pennsylvania. In such instances, an occasional macroscopic form
may be recovered as an isolated, washed-in individual and not at
all representative of the actual ecosystem.
Case G. Fresh-water/estuarine system
Region III contains a number of very large, economically
important fresh-water estuarine systems. Among these are the
Delaware, Potomac, James, York and Rappahannock estuaries plus
Chesapeake Bay. There «re also several smaller estuaries on the
Virginia-Maryland Eastern Shore. Features such as these are the
result of a coastal subsidence inundating river valleys adjacent
to the sea. (Figure 12)
Although ttxe «atuarine system is a transition zone between fresh
and saline water, the BOD5/DO ratio holds as shown in Figure 13,
The curve begins with essentially near-normal stream conditions at
17
-------
Fieldsboro, New Jersey, shows degradation at the Philadelphia/
Camden section and then recovery at the Pea Patch Island section.
The curve also shows the net effect of the Philadelphia and Camden
sanitary treatment plant effluents at River Mile 103.96 and 97.88.
During the summer months, this effect is noticed almost immediately
in downstream stations but winter conditions inhibit the biological
assimilation rate (low temperature effects) and the maximum effluent
loading point occurs much further downstream. The loading effect
is concentrated by resistance to stream flow generated by high tide
as shown in the summer high tide slack curve. In addition, the
larger relative volume of water available during high tide allows
more dilution of river water, hence a better quality of water is
observed. During low tide slack, the stream behaves more normally
and the assimilation effects are noted much further downstream.
It is interesting to note that the Delaware River is only moderately
degraded at the Fieldsboro, New Jersey station (River mile 127.48)
and that assimilation of biochemical oxygen demanding material from
the Philadelphia area is nearly complete at Pea Patch Island (River
Mile 60.55). Use of the BOD /DO ratio, therefore, serves as a very
rapid means of assessing the relative health of any given section
of a stream as well as to detect any subtle changes occurring with
time. Further, the Delaware River curves show the relative stability
18
-------
of the better water quality stations (Stations 127.48, 122.49,
100.15 and 60.55) while the degraded sections during the year show
much greater fluctuations and instability of the BOD5/DO ratio.
Case H Specialized uses
In this category, the ratio can be applied to sanitary treatment
plant effluents to determine operational efficiency of the unit.
Such a test, of course, cannot be applied to a primary plant but
a secondary plant, if operating at an optimum level, should produce
effluent in which the BOD /DO ratio lies between O.6 and O.8.
Operator's manuals for such plants recommend a 5-day BOD value which
is 0.7 of the ultimate BOD figure. Therefore, if the BOD /DO ratio
is 'Bloserved to be 0.6 and O.8, the net carbonaceous portion of the BOD
loading effect of treatment plant effluent upon the receiving stream is
nearly zero. If the ratio between effluent and receiving stream is
small (small discharge, large stream) then the sum total of the
plant operation is for practical purposes, achieving zero discharges
insofar as the stream is concerned.
VI. Stream classification using the BOD5/DO ratio
Assignment of stream pollution categories by means of the BODg/DO
ratio can be achieved in a general way provided a sufficient backlog
of data points is available and the analyses are reasonably free from
error .
19
-------
The preliminary assignment of a stream classification based
upon the BOD../DO ratio is given in Figure 14.
This classification system, if refined by using a species
diversity index of the Shannon-Weiner type, can be used to assign
a stream to a meaningful spot in the environmental scale of viability.
Values obtained in this study indicate a stream to be nutrient deficient,
under winter conditions or having some interference with the BOD test
if the ratio falls below 0.075 for any extended period of sampling.
The normal stream ratio value seems to lie between 0.075 and 0.300.
Degradation begins to be observed between 0.3 and 0.5 and polluted
conditions occur between 0.5 and 0.65. Values consistently over
0.65 are seen in grossly1 polluted streams.
It must be kept in mind, however, that this scale is gradational
and the classification given above is considered a working approxi-
mation which may require some modification as more work on ratios is
carried out. One can say with certainty, nontheless, that any stream
with a BOD,-/DO ratio consistently around 0.8 is a stream with a very
real problem and in need of immediate remedial action.
20
-------
~3
f 1 f 1 I 1 I 1 f i I 1 I I I
Nutrient dtficwnt or
winter Mo or interfering Polluted biology
restricted
. . ,
substances present Degraded streams Grossly polluted streams
| tNormol strain range , stressed stream , j , H^W r«»*ncted to most tolerant fcrms
0 OJ 02 03 O4 OS O6 O7 O8 O9 LO o 00
BODs/DO ratio
FIGURE 14: Preliminary stream classification based on long-term average values of
BOD5/DO ratio
-------
VII. Example of computer program for retrieval of STORE! BOD./DO
data and ratio calculation
The program to extract, calculate and print the BOD /DO values
and ratio from the various station data stored in STORET is given
in Appendix A. This program will allow the data'-retrieval and
calculation of the ratio between BOD,, and DO for any given station.
The program is entered in the STORET system as FORT GCG and the
statistical linear regression package to determine the slope of the
line through the data points is called REG. This canned program can
be accessed by writing CALL REG at the proper location at the be-
ginning of the program.
21
-------
Summary
The BOD /DO ratio has been shown to be useful in evaluating
the general health of streams in a variety of settings and
conditions. Keeping in mind that a stream is a dynamic, quasi-
living organism under the influence of a large number of ever-
changing factors, then the value of acquiring a key to determine
and understand these changes will be appreciated.
Assignment of rating scale of ratios to streams can assist
workers in the water pollution field to understand the actual reaction
of a stream to clean-up efforts as well as to monitor any progressive
changes being imposed by man-made pollution sources. Data presented
indicate the ratio works well in differing stream types in many
areas of the county and can be used to evaluate the net ecological
activity in these streams. It can also be shown that stream$
exhibit varying ratios with both time and season but those streams
which are relatively clean exhibit only minor fluctuations about
a central mode. Conversely,streams which are degraded show a
much larger fluctuation amplitude and the BOD /DO trace is generally
erratic.
Evaluation of the long-term trend slope of the ratio trace will
give a good indication of the net effect of pollution-abatement efforts
in a watershed or, conversely, determine if a stream is actually in
need of a pollution abatement program.
22
-------
The result of this study is to show that an evaluation tool for
understanding one facet of a stream system has been developed.
However, additional analyses of stored water quality data as well
as new data being obtained today may show that some modification
of the classification system will be required. Because of the basic
simplicity of the two parameters involved in this ratio, a. large
number of workers may find this determination useful in the course
of their own studies and therefore benefit the general environmental
pollution abatement effort.
23
-------
Appendix A. BODs/DO ratio program with least squares linear regression
analysis.
How to Obtain BOD5/DO Data for Input into Ratio Program
Using the STORET system RET package, the type of retrieval
must be a PGM = PUNCH for the data input must be in exponential format.
Retrieve the agency, station, BOD5/DO, dates, etc., using the punch
program. Be certain to change (upward) the cards field of the job
card; this field is the actual number of cards not thousands as is the
lines field.
When the job is awaiting print do not route it to punch, but fetch
over your low-speed terminal with fetch xxx ddname 0 (alpha O). Save
the data in card or LRECL = 8O format.
Use the data set saved above as the input into the program following.
-------
C TO SUPHESS HOIMT I.IS1INCS, PLACE A CARD IN FRONT
•C (it- UAl'A .Jim 'I-'AST' 1U COLUMNS 1-4. -
C FUR JULIAN DATE AND RATIO LIST PLACE A CARD IN FRONT OF DATA
"C hITH ^DATE^TN^CDTDMR^nr^" ---- ~ — ---------------
__ INTj:GE_R -I' ATE(6).TI ME ( ?) _
DTMENSION IDSAVE(4"),ID(4), ~~
*KALDAY( !2>.Lh'APYR( I?)
DATA
DATA ID/n*' '_,' _ //,SX_,SY,SXY,SX?,S_Y?/5*0./
DATA KAL!)AY/0,^I ,b9,90~, T?0,~l bT.Ta I . 212,24 i' ,T/J ^
_____ DA'IA LEAPYR/b2.S6,60.64 , oR_, 72 ,T6, 80, b4 , RP, 92,
" DATA N,JUMP.LAST,Kf)ATE/4*6/
C ENTER ID VALUE CHANGE
10 DO 20 LOOP=I,4
20 IDSAVE(LOOK)=ID(LOOH)
INPUT A DATA CARD
30 READ(b,40,t-.MD=70)ID,DATE,TIME,BOD,DOVAL
40 FORMAr(3A4.A.3.6AI ,2A2. IX.2ER.O)
C TEMP REVERSAL OF BOD ft DOVAL FIELDS BEGINS
HOLD=BOD
BOD=D()VAL
U)VAL=HOLD
C TEMP REVERSAL OF BOD & DOVAL FIELDS ENDS
_z
C CHECK WHETHER JULIAN DATE & RATJO LISTING~IS DESIRED
IF(ID( D.Nb.IDDATE)Gu TO 44
KDATE=I
GO To 30
C
C CHECK WHETHER POINT LISTING IS DESIRED
44 IF(ID(1).NE.IDFAST)GO TO 50
JUMP=I
GO TO 30
C COMPARE CURRENT ID TO PREVIOUS ID
50 DO 60 L()OP=I,4
IF(ID(LOOP).NE.IDSAVE(LOOP))GO TO 80
60 CONTINUE
GO To 170
C
70 LAST=I
C CHANGE 01- STATION ID
80 IF( IST.EQ.I)GO TO I
81 RErtlMD 10
DO 140 LO()P=I,N
«t-AD( I0.90)KYR.KMO,KDT,RATIO
90 FORMAT(3I2tF9.4)
C COMPUTE' JULIAN DATE
I=KMO
JULIAN=KALDAY(I)+KDT
IF(I.LT.3)GO TO 120
C ADJUST JULIA.'J DATE FOR LEAiJ YEAR
DO 100 1=1.7
1F(KYR.EQ.LEAPYR(I))GO TO 110
100 CONTINUE
GO TO 120
IIP JULIAN=JULUN+J
120 JULIAM=I000*KYR+JULIAN
X=JULIAN
Y=RAiIO
SY=SY+Y
_sxy_=MY __
i>X2=SX2+X**2
C JULIAN DATE UND hOD-DO VALUES ARE RETAINED
^, 13T) JULIAN, RATIO
130 H)RMAT< Ib, 1
140 COi-JTI^Ub
-------
IF(N.LT.2)GO TO I b3
C FORMULA COMPUTATIONS OF A,B, AND
COEN=N*SX2-oX**2
IF(CDtN.t-:0.0.")U) 1"0 Ib3
Af (SY*SX2-SX*SXY)/Cnfcli_
H=TN*"SXY-SX*SY)7CUEN"
tli-N2 = ( SX2-C SX**2/N ) ) * ( S Y2-( SY**2/N
lr(L)biM2.b(J.O.)UO 10 I 33
R=(SXY-((SX*SY)/N))/DEN
C hHITb THt EQUATION FUR THE LhAST SQUARE LINE
____ _ _ _
ANf)~ fHtf VALUc uF~"fw>: COMHLVflWJ "COEFFTcftrNr
n4ITb(6,
loO K>HWiAT(///' Y = '.E13.6,' + '.EI3.6,' * X',//' R = '.EI3.6,
*//' N = '.I?,' POINTS')
IH(KDATE.NE.1)00 10
149 FOHMAT(b(/),132,'JULIAN',T40,/RATIO',/T33,'DATE',//)
11
DO IbOO LOOH=I,N
REA»(Il.lbl)JULIAN.RATIO
Ib'l FORMATC Ib,F9.4)
rttiITfc(6.lb2)JULIAN.RATIO
Ib2 FOHMAT(T3I.Ib,T4b,F7.2)
IbOO CONTINUE
C FUTURE PLOTTER CALLS hILL APPEAR HERE
Ib3 IF(LAST.EQ.l)GO TO 220
REMIND 10
REi^IND 11
C
. C
' C CLEAH VAHlABLfcS
' N=0
bX=0
bY=0
bXY=0
SX2=0
HEADING
Ibb HMITEC6,160)ID
160 K)RMAT(IHI.6(/),T27.'SAMPLING STATION IDENTIFICATION NUMBER '. _
*3A4,A3,3(/),T21.'SAMPLE COLLECTION',TDI,'PARAMhTfcHS',T74.'BOP-00',
*/r24.'DArE'.T^2,yTIMfc',T48,'ROD'.T6l<'DO'.R ITEM 0,200)0 ATE, RATIO
200 KJRMATCOAI.F9.4)
N
IF(JUMP.EQ. I )G() 102 15
7TTOTD ATt \TTW7iTCrn; DOFAT7R ATTo"
210 t-ORMAV(?2X.6AI , 7X ,2 A2 , 4X , ?<4X,F7.2 .'0') ,8X ,FR. 3)
2lo 1S1=0
GO R) 10
220 STOP
tND
------- |