mmmmmmmmmmm
mmmmmmmmmmm
mmmmmmmmmmm
ฐ/ mmmmmiTP
ฐVVฐ
$*J < !____
____________-. *m mmmmmmmmmmmmmmmmmmmmmmmmmm ...*... mmmmmmmmmmmmm' !
' ___ ."..*.*.*...."ป* mimmmmmmmmmmmmmmmmmmmmmmmmmmm !._ * ___' '_
--.--------^_iซ mmmmmmmmmmmmmmmmmmmm mm m m m m mmuu mm mm mm mm mm mm mm mm m m, m * m i
ENVIRONMENTAL PROTECTION AGENCY
&ซป
mm*
m m m
mm*
"""*B"B"ซ "* '
-------
DELAWARE ESTUARY COMPREHENSIVE STUDY
FINAL REPORT
CHAPTER I
HYDROLOGY
REGION II
ENVIRONMENTAL PROTECTION AGENCY
Edison, New Jersey
July 1971
-------
TABLE OF CONTENTS
Section Page
ESTUARY AND BASIN PRECIPITATION
RUNOFF TO ESTUARY 15
ESTUARINE DYNAMICS 36
-------
CHAPTER I
SECTION A
ESTUARY AND BASIN PRECIPITATION
Precipitation in the Delaware Estuary area is generally abundant, hav-
ing an annual average of approximately 42 inches. The average annual pre-
cipitation for the entire Delaware River Basin is slightly greater, being
about 44 inches.
The record of annual precipitation in the Delaware Estuary area is very
nearly a normal (Gaussian) distribution^). United States Weather Bureau
records(2' show that the average annual precipitation by sub-basins in the
Delaware River Basin varies from about 42 to 45 inches. Average precipi-
tation is greatest along the western edge of the basin and in the upper por-
tion of the basin (i.e., New York State).
Although there is a considerable variation in annual precipitation,
the variation in monthly precipitation is even more striking. For example,
at Philadelphia, the October 1963 recorded value was 0.09 inches as compared
to 5.21 inches for October 1943 and an average for the period of record of
2.78 inches. Extreme values are always associated with extreme weather
phenomena such as droughts and hurricanes.
Table 1 lists the normal monthly precipitation at Philadelphia, with
a range of 2.78 inches in October to 4.63 inches in August. Precipitation
is generally greatest during the summer and least during the winter.
The normal snowfall in the study area is approximately 21 inches.
Since the average temperature for the winter months is 33 F, much of the
winter precipitation occurs as rain rather than snow. The average annual
snowfall for the basin varies from about 15 inches in the Bay region to
over 70 inches in the headwaters area and the Pocono Mountains.
U. S. Weather Bureau records indicate that approximately 50% of the
average annual precipitation occurs between May and September. Summer
precipitation over the estuarine area is about 45% of annual. This propor-
tion increases from east to west, with the area west of the Delaware River
being almost 50%.
DROUGHT CONDITIONS
The Delaware River Basin has experienced extended periods of subnormal
precipitation, commonly referred to as droughts. One definition of a
drought is when precipitation is insufficient to meet the needs of estab-
This section was prepared by Darwin R. Wright, U.S. Environmental Protection
Agency, Washington, D.C.
- 1 -
-------
TABLE 1
Precipitation, Total Water Equivalent (in.)
(Climatological Standard Normals, 1931-1960)
(Precipitation During Drought, 1962-1966)
United States Weather Bureau
Philadelphia, Penna.
Month
January
February
March
April
May
June
July
August
September
October
November
December
Yr. Total
Deviation
from
normal
E
T
1962
2.95
3.51
3.91
3.69
1.85
7.40
2.30
6.58
2.77
0.95
4.60
2.11
42.62
+0.14
rought Period 1962-1966
otal Wa
1963
2.31
2.19
3.94
1.13
1.06
2.88
3.13
3.35
6.44
0.09
6.67
1.76
34.95
-7.53
ter Equi
1964
3.92
2.83
1.94
5.27
0.47
0.21
3.83
0.49
2.42
1.73
1.64
5.13
29.88
-12.60
valent (
1965
2.35
2.18
3.19
2.33
1.23
2.85
3.22
4.05
3.02
2.02
1.05
1.85
29.34
-13.14
in.)
1966
2.82
4.30
0.68
4.35
2.95
0.41
2.35
1.63
8.70
5.12
2.36
4.33
40.00
-2.48
Normal
Total
3.32
2.80
3.80
3.40
3.74
4.05
4.16
4.63
3.46
2.78
3.40
2.94
42.48
Period of Record 1931-1966
Maximum
(in.)
6.06
4.64
6.27
6.58
7.41
7.40
7.48
9.70
8.78
5.21
6.67
5.48
Monthly
Year
1949
1958
1953
1947
1948
1962
1959
1955
1960
1943
1963
1951
Minimum
(in.)
0.45
1.37
0.68
1.13
0.47
0.11
0.64
0.49
0.88
0.09
1.05
0.25
Monthly
Year
1955
1954
1966
1963
1964
1949
1957
1964
1941
1963
1965
1955
- 2 -
-------
lished human activities(3). Other definitions are also common, depending
on the water user (i.e., farmer, water supply operators, meteorologists,
etc.). Many other factors, such as effect on plant life, drying out of
soil, and drying up of streams are characteristic of droughts. Conse-
quently, a specific definition is not necessary if the effects are obvious,
The Delax^are River Basin Report{3) lists twelve periods of extended
subnormal precipitation. The 1930 drought stands out as the most severe,
both for duration and for deficiency of precipitation prior to the drought
of the 1960's. The beginning of the most recent drought varies, depending
on the source quoted, but it is generally agreed that the drought began
in 1962 and continued through 1966.
Table 1 lists the U. S. Weather Bureau precipitation records and
normals for Philadelphia, Penna. Thirty-five of the forty-eight months
of 1962-65 were below normal. During the drought, from September 1962
through December 1965, thirty-two of the forty months were below normal.
The total precipitation of 104.60" represents approximately 75% of the
normal. The severity of the drought is evidenced by:
1. The total precipitation during 1963-64 was only approxi-
mately 70% of the normal.
2. Five minimum monthly totals were recorded.
3. All twelve months of 1965 were below normal.
4. 1965 was the second driest year on record (29.34 inches).
5. 1964 was the third driest year on record (29.88 inches).
6. Precipitation during the period May through November 1964
was about 41% of normal.
7. 1964 was the driest summer (June to August) on record.
8. 1963 had the longest dry spell - October 4 to October 31, 28 days.
9. 1964 had the second driest May and August.
Not only have drought conditions persisted in the lower basin, as
is evidenced by the Weather Bureau records from Philadelphia, Pa., but
in general, the drought affected the entire basin. In the northern half
of the basin the drought caused extremely low streamflow, which in turn
resulted in low reservoir storage during 1965.
The Delaware River Basin may expect more severe dry spells for short
periods, in addition to extended periods of drought. However, the last
drought stands without precedent for its long duration and large preci-
pitation deficiency.
- 3 -
-------
UNITED STATES WEATHER BUREAU STATISTICAL ANALYSIS
In the analysis of stormwater overflows, precipitation records are
needed to develop such variables as frequency and duration of combined
sewer overflows and rainfall-runoff relationships. As a part of the
stormwater overflow study, a rain gage network was installed to provide
adequate precipitation records. (See Chapter IV F Stormwater Overflow).
The estuary region does not have a distinct wet or dry season, but
there is a pronounced difference in winter and summer precipitation.
Summer is characterized by local thundershowers, whereas winter precipi-
tation is generally more widespread and less intense. This variation in
precipitation is reflected in the flow and quality of stormwater overflows,
Most precipitation data are usually recorded as daily or even monthly
totals. Because of this, little can be concluded concerning stormwater
overflows from U. S. Weather Bureau daily precipitation records. In fact,
as shown in Table 2 which lists the number of days precipitation was re-
corded for the period 1961-1965, the differences between drought periods
(1963-65) and normal periods (1961-62) are not particularly striking.
TABLE 2
NUMBER OF DAYS PRECIPITATION RECORDED
UNITED STATES WEATHER BUREAU - PHILADELPHIA, PA.
Month
January
February
March
April
May
June
July
August
September
October
November
December
1961
11
14
7
11
8
10
7
10
8
10
8
8
TOTAL DAY 112
TOTAL PREC. (in.) 41.05
1962
7
12
12
14
13
10
10
9
5
7
8
12
119
42.62
YEAR
1963
12
9
12
7
9
9
8
10
8
1
10
7
102
34.95
1964
9
12
12
16
3
7
6
8
8
4
7
14
106
29.88
1965
8
10
10
11
9
11
7
14
4
6
9
7
106
29.34
- 4 -
-------
Such records therefore are of little value when attempting to evaluate
storrawater overflows. To adequately define and predict stortnwater overflows,
an analysis of each individual storm is needed. Since the study period for
measuring the effects of stormwater overflows was to last only approximately
two and one half years, a more general evaluation of "storm phenomena" was
desired. Consequently, a contract was let with the U. S. Weather Bureau
to furnish a statistical analysis of "storm phenomena" for Philadelphia,
Penna.
Precipitation records are available for Philadelphia, Penna., back to
1820, but only data from May 1890 included hourly precipitation values.
Consequently, the period from May 1890 through April 1964 was analyzed by
the Weather Bureau.
A storm was defined as (1) an hourly precipitation recording of more
than a trace, (2) beginning at that hour when the precipitation is a posi-
tive number and only a trace or zero for the preceding hour, and (3) ending
at any hour with zero or a trace if the preceding hour had positive precipi-
tation. Positive precipitation is defined as an amount greater than zero,
or a trace. The duration of a storm is the difference between its ending
and starting times. Similarly, the interval between one storm and the next
is the starting time of the second minus the ending time of the first.
The analyses furnished by the Weather Bureau included:
1. Bi-variate distribution of rainfall intensity per storm
(inches per hour) versus storm duration (hours). This
analysis was carried out for each month of the year for
the period May 1890 through April 1964 (74 years), result-
ing in 12 monthly bi-variate distributions.
For storms that extended from one month to the next,
e.g., August 31 thru September 1, the storm was assigned
to the month in which the storm lasted the longest. In
those cases where the number of hours of storm were equal
for both months, the storm was assigned to the month in
which it started.
2. A histogram analysis of the interval between successive
storms carried out for each month of the year for the
period May 1890 through April 1964 (74 years), resulting
in 12 histograms.
For "intervals between storms" that extended from one
month to the next, e.g., August 31 to September 1, the
interval was assigned to the month in which the interval
lasted the longest. If the number of hours of interval
were equal for both months, the interval was assigned to
the month in which it started.
5
-------
A copy of Official City hourly precipitation data for
April 24, 1890, through December 1948, and punched
computer cards giving hourly precipitation for Philadelphia
Weather Bureau-Airport (36-6889) for the period May 1, 1948
through June 30, 1964.
Copies of the twelve monthly bi-variate distributions
and the twelve monthly histograms are in Appendix 4.
STORM INTENSITY ANALYSES
The marginal probability distributions were computed from the bi-
variate distributions. The values were tabulated on the bi-variate dis-
tributions in a total column or a total row. Using these values, the
monthly storm intensity and storm duration frequency distributions were
determined.
Storm intensity was defined as the sum of the hourly precipitation
values divided by the number of hours in the storm. Using the monthly
intensity marginal distributions, the probability distribution function
was plotted on semi-log paper. Figure I-A-1 is a sample plot showing
January, July and the entire year. For individual months, the data
approached a straight line. Consequently, the lines for January and July
represent a visual approximation of a line of best fit to the sets of data.
The other months were plotted using the same procedure. An exponential
distribution was assumed with the probability distribution function given
by:
-Ax
y = 1006 (1)
where y = percent of storms with intensity greater than the
stated value
x = storm intensity in in./hr.
A = constant parameter >o (hr./in.)
e = 2.718
Based on the assumption that the data is of the form of equation (1),
then I/A equals the mean and I/A equals the variance(4). The results of
this analysis by months are listed in Table 3.
- 6 -
-------
TABLE 3
STORM INTENSITY
UNITED STATES WEATHER BUREAU - PHILADELPHIA, PA.
May, 1890 - April, 1964
Month
I/A = Mean (in./hr.)
Variance (in./hr.)'
January
February
March
April
May
June
July
August
September
October
November
December
0.044
0.043
0.044
0.067
0.098
0.121
0.143
0.147
0.112
0.070
0.061
0.051
0.0019
0.0018
0.0020
0.0045
0.0095
0.0147
0.0205
0.0205
0.0216
0.0049
0.0037
0.0026
The mean storm intensity varies from 0.044 inches per hour during the
winter months to 0.147 inches per hour during the summer months. This is
the result of having long duration storms, including snowfall during winter
months, while the summer months are characterized by thundershower activity.
The variance is also greater during the summer months, which is an indication
of the wider range of intensities which result from thundershower activity.
The minimum intensity recorded was 0.01 inches per hour. Most of the
storms of this intensity are comprised of hourly precipitation values of
0.01 inches with the storm durations varying from one hour to several
hours. Of the 13,183 storms, 30.5% are comprised of a total rainfall of
0.01 inches. From a stormwater overflow point of view, these storms would
not produce bypasses to the stream. As expected, maximum storm intensities
are recorded during the summer months.
These maximum values should not be confused with actual values as
usually reported, because in this case, the storm duration is recorded to
the whole hour. For this reason, the maximum storm intensity was recorded
in the interval of 1.31 inches/hr. to 1.40 inches/hr. Table 4 lists
pertinent maximum storm intensity data.
- 7 -
-------
TABLE 4
MAXIMUM STORM INTENSITIES
UNITED STATES WEATHER BUREAU, PHILADELPHIA, PA.
May, 1890 - April, 1964
Maximum Storm
Month Intensity (in./hr.)
January
February
March
April
May
June
July
August
September
October
November
December
.26 -
.26 -
.36 -
.51 -
1.01 -
1.21 -
1.21 -
1.31 -
1.01 -
.81. -
.76 -
.56 -
.30
.30
.40
.55
1.10
1.30
1.30
1.40
1.10
.85
.80
.60
STORM DURATION ANALYSES
Storm duration was defined as the number of consecutive hours during
which at least 0.01 inches of precipitation was recorded. Using the monthly
duration marginal distributions, the probability distribution function was
plotted on semi-log paper. Figure I-A-2 is a plot showing January, July,
and the entire year. For the individual months, the data approached a
straight line. A visual approximation of a line of best fit to each set
of monthly data was drawn. Using equation (1), the X's were determined
and the mean and variance computed. Table 5 lists the results of this
analysis.
- 8 -
-------
TABLE 5
STORM DURATION
UNITED STATES WEATHER BUREAU, PHILADELPHIA, PA.
May, 1890 - April, 1964
Month
January
February
March
April
May
June
July
August
September
October
November
December
I/A = Mean (hr.)
5.04
5.17
4.28
3.65
2.89
2.73
2.08
2.78
3.39
3.47
4.34
4.91
I/ A2 = Variance (hr.)2
26.46
26.74
18.28
13.30
8.34
7.46
4.34
7.72
11.47
12.06
18.87
24.10
The mean storm duration ranges from 2.08 hours in July to 5.17 hours
in February. This again is an indication of longer duration storms during
the winter months and shower activity during the summer months. Obviously,
this storm duration analysis corresponds very closely with the storm in-
tensity analysis. That is, summer storms usually have a short duration -
high intensity, while winter storms are of a long duration - low intensity.
The variance shows the same trend as the storm intensity variance, indicat-
ing a wide range of storm durations.
The minimum storm duration was one hour and the maximum storm duration
interval was 45 to 46 hours. It should be noted that storm duration does
not indicate that precipitation was recorded continuously, but that accord-
ing to the Weather Bureau data, some precipitation was recorded during the
preceding hour for each hour of the storm. Also, what may appear to be a
continuous day of rain to the layman, may in fact, be a number of storms
because during some hours less than 0.01 inches of rain may have been
recorded.
The storm of one hour duration may represent two different types of
storms. The first type of storm is the one with a total precipitation of
0.01 inches. The second type is the thundershower, which prevails during
the summer months and usually last less than one hour. The percent of
one hour duration 0.01 inches total rainfall storms varies from 77% of
the total one hour storms for January to 40% of the total one hour storms
for July. The thundershower activity is evident when considering the per-
cent of one hour duration storms with an intensity greater than 0.11 -
- 9 -
-------
0.15 in./hr. interval. The range of these storms is from zero percent
in January to 12.4 percent in July. The one hour duration storms of vary-
ing intensity account for 38% of the total. The one hour duration 0.01
inches total rainfall storms account for 58% of the total one hour dura-
tion storms.
The maximum storm durations are recorded during the winter months.
Table 6 lists the maximum storm duration and the percent of storms with
a duration greater than twelve hours. While July has a maximum storm
duration of 19 hours, only 0.4% of the storms are greater than a 12 hour
duration. December, which has a maximum storm duration of 45-46 hours,
has 7.8% of its storms greater than twelve hour duration.
TABLE 6
STORM DURATION
UNITED STATES WEATHER BUREAU, PHILADELPHIA, PENNA.
May, 1890 - April, 1964
Maximum Storm % Storms
Month Duration Interval (Hours) Greater Than 12 Hours
January
February
March
April
May
June
July
August
September
October
November
December
33-34
39-40
41-42
35-36
39-40
21-22
19
29-30
27-28
23-24
35-36
45-46
9.0
9.4
5.5
3.9
1.6
1.4
0.4
1.5
2.2
2.6
5.6
7.8
ANALYSIS OF INTERVAL BETWEEN STORMS
The quality of combined sewer overflows has been found to be highly
variable, especially when measured by such parameters as BOD or solids.
One of the reasons for this variability may be the result of conditions
which prevail in the sewer between periods of overflow. Prolonged periods
of dry weather may result in the deposition of solids, which are later
flushed out if.sufficient storm flow results.
Prolonged periods of dry weather may also result in higher concen-
trations of pollutional materials being deposited on the surface, which
- 10 -
-------
enter the combined system from surface runoff. Studies have shown that
a large percentage of suspended solids in urban runoff could be attri-
buted to dustfall.d)
Consequently, the "interval between storms" may be a significant
factor in the quality variability of combined sewer stormwater overflows.
The Weather Bureau analysis showed that 56% of the "intervals between
storms" were less than 21 hours, with 42% less than 7 hours, and 20% equal
to or less than one hour. The 56% of the storm intervals (<21 hours) re-
present a different effect on overflow quality than the remaining 44% that
affect the quality of overflows. This difference is especially obvious
during thundershower activity. For example, if a thundershower of high
intensity is followed closely by another thundershower of high intensity,
the second overflow will probably have a lower concentration of total
solids, since both the surface and sewer have been flushed by the first
thundershower. The surface and sewer deposition does not have enough
time to build-up significant amounts in twenty hours or less.
The monthly histograms of the "interval between storms" are contained
in Appendix 4 - Stormwater Overflow Sampling and Analysis Program. The
number of storms of each interval is tabulated on the histograms. It
should be noted on these histograms that the size of the histogram cells
is not equal. The cell size for the range from one to twenty-four hours
is either one, two, or four hours. From one to six days, every 12 hours
and from six to fifteen days, every 24 hours. Then every five days,
(i.e., 15, 20, 25, 30) up to a maximum of 7200 hours or 30 days.
It was necessary to break down the small intervals because of the
definition of a storm and consequently the storm interval. Intervals
between storms of one to six hours are the result of intermittent
periods without precipitation during what would appear to be, for example,
a complete day of rain. The minimum and maximum "interval between storms"
are one hour and 601/720 hours, respectively.
The twelve histograms are all of the same general shape. For the
year, 80.5% of the "intervals between storms" are equal to or greater than
one hour; 42.0% are > 21/24 hours (1 day); 8.5% are > 133/144 hours (6 days);
0.4% are > 337/360 hours (15 days).
Since the "interval between storms" histograms did not appear to have
an exponential distribution, various analyses were tried. The month of
October was chosen for this analysis since it had recorded the greatest
"interval between storms" and also during October 1963 the longest dry
spell on record was experienced. Figure I-A-3 is a plot of interval between
storms (hr.) and the frequency of occurrence. The nine curves represent
the omission of selected storm intervals. With all intervals included,
the data is an S-shaped curve (A). However, with the intervals equal to
or less than 21-24 hours omitted, the data approach a straight line.
Considering the data in light of stormwater overflows, intervals less than
or equal to 21-24 hours are not as important as the longer intervals.
- 11 -
-------
Also, since the data greater than 24 hours appeared to approach a straight
line, it could be analyzed using the same procedure as storm intensity
and storm duration.
Using the monthly histograms, the probability density function was
determined and plotted on semi-log paper. Figure I-A-4 is a plot showing
these for January and October. For the twelve individual months, a
visual approximation of a line of best fit for each set of monthly data
was drawn. Using equation (1), the X's were determined and the mean
and variance computed. Table 7 lists the results of this analysis.
TABLE 7
"INTERVAL BETWEEN STORMS"
(INTERVALS GREATER THAN TWENTY HOURS)
UNITED STATES WEATHER BUREAU - PHILADELPHIA, PENNA.
May, 1890 - April, 1964
Month
January
February
March
April
May
June
July
August
September
October
November
December
I/A = Mean (Days)
3.15
2.93
2.74
3.34
3.04
3.13
3.17
3.34
4.69
5.17
4.26
3.36
2 2
1/X = Variance (Days)
9.91
8.59
7.48
11.18
9.24
9.77
10.05
11.18
22.00
26.71
18.11
11.32
For "intervals between storms" greater than 21 hours (21-24 interval),
the Weather Bureau analysis showed that the mean for October was the
longest, with 5.17 days and the mean for March was the shortest, with
2.74 days. The longest "interval between storms" was recorded in the
601-720 hour interval during September (1 occurrence) and October
(2 occurrences). October had the largest number of "intervals between
storms" greater than 360 hours (15 days) with 16 occurrences (13 occur-
rences were in the 361-480 hour interval). Table 8 lists the maximum
"interval between storms" and the percent of "interval between storms"
greater than 144 and 288 hours.
- 12 -
-------
TABLE 8
"INTERVAL BETWEEN STORMS"
(INTERVALS GREATER THAN TWENTY HOURS)
UNITED STATES WEATHER BUREAU - PHILADELPHIA, PENNA.
May, 1890 - April, 1964
Month
Maximum
"Interval Between Storms"
Interval, hours
% "Interval Between
Storms"
Greater than
144 hours 288 hours
January
February
March
April
May
June
July
August
September
October
November
December
361-480
481-600
337-480
481-600
337-360
481-600
361-480
481-600
601-720
601-720
481-600
361-480
17.1
14.6
12.9
16.0
16.5
17.4
16.2
18.0
29.2
32.7
26.3
20.5
1.6
1.5
0.9
2.3
1.7
1.3
1.1
1.3
7.3
9.6
5.1
2.5
The maximum "interval between storms" usually occurs in the summer and
fall months. Consequently, if the hydraulics of the system are susceptible
to solids deposition, sufficient solids may build up such that the quality
of the overflow is affected. Part of the difficulty in predicting overflow
quality is due to the fact that the duration and intensity of the storm after
a given interval also affects the quality. For example, a high intensity
storm will produce relatively high concentrations of solids, while for the
same preceding storm interval, a low intensity storm would have relatively
low concentration of solids.
SUMMARY
Precipitation in the Delaware Estuary is generally abundant and averages
approximately 42 inches per year. Precipitation is greatest during the sum-
mer (August = 4.63") months, and least during the winter months. The recent
drought (September, 1962 to August, 1966) stands without precedent for both
its long duration and its large precipitation deficiency.
The results of statistical analyses of 74 years of precipitation records
at Philadelphia, Pa. indicate that storm intensity and storm duration can be
described by an exponential type of probability distribution function. Mean
- 13 -
-------
storm intensity varied from 0.04 in./hr. in February to 0.15 in./hr. in
August. Mean storm duration varied from 5.2 hours in February to 2.1 hours
in July. This is an indication of long storms during the winter months and
shower activity in the summer months. Maximum storm intensity and duration
were 1.40 in./hr. and 46 hours respectively.
The "interval between storms" did not show any well defined probability
distribution. However, when all intervals less than or equal to twenty
hours were excluded, the data did exhibit an exponential distribution. The
"interval between storms" (>20 hours) varied from 2.74 days in March to
5.17 days in October. The maximum interval between storms was recorded
in the 601-720 hour interval during October. October had the largest number
of "intervals between storms" greater than 360 hours. Further studies are
needed to relate these analyses to the problems of stormwater overflows.
These analyses would be a valuable aid in the design of stormwater overflow
control facilities.
- 14 -
-------
STORM INTENSITY OCCURRENCE
FOR MONTHS OF
JANUARY,JULY AND FOR YEAR
JULY
\
\ N
\
YEAR\
A-
\
\
.01
.20
.40 .60 .80
STORM INTENSITY (HOURS)
1.00
1.20
1.40
Figure I-A-1
-------
yu.u
80.0
70.0
60.0
50.0
40.0
30.0
20.0
10.0
uj 9.0
3 8.0
< 7.0
> 6.0
ฃ 5.0
< 4.0
CO
!ฃ 3.0
i 2.0
i-
ae.
HI
(-
S 'ป
0 -9
1 :76
2 -5
Q -4
5 .3
ฃ
5 -2
Oฃ
O
V)
.1
.09
.08
.07
.06
.05
.04
.03
.02
.01
\ V
\x
\ X
V X
\
V
X
. \
.N
X
V'
X
V x
\
\
\
\
\
\
\
\
\
SI
J
X
N
V
"v
1 V
X. V
X N,
\
v
\
YEAR
,
\
\
\ .
\
\
\
\
\
\
'ORM DURATION OCCURR
FOR MONTHS OF
ANUARY,JULY AND FOR '
X
N .
y \
X0x
JULY
i
\
\
\
X
V
".
>
\
-.
v
X
X .
v
\
\
JANUARY
ENCE
/EAR
\
X
X
X
X
\
10 15 20
STORM DURATION (HOURS)
25
30
35
Figure I-A-2
-------
1,(
900
800
700
600
500
400
300
ซ/>
at
D
O
X.
v>
*
a:
O
ป
ซ/>
Z
UJ
UJ
*
UJ
CO
_l
<
>
Of
200
100
90
80
70
60
50
40
30
20
O> O> N OOXJO
FREQUENCY DISTRIBUTION
INTERVALS BETWEEN STORMS
OCTOBER
LEGEND
A ALL INTERVALS
B ONE HOUR INTERVAL OMITTED
C 1-2 HOUR INTERVAL OMITTED
D 1-3 HOUR INTERVAL OMITTED
E 1-4 HOUR INTERVAL OMITTED
F 1-5 HOUR INTERVAL OMITTED
G 1-6 HOUR INTERVAL OMITTED
H 1-12 HOUR INTERVAL OMITTED
I 1-24 HOUR INTERVAL OMITTED
10 15 20
30 40 50 60
PERCENTAGE
70 80 85 90
30
25
20
15
13
10
9
8
7
6
5
4
3/2
3
2
1/2
1
/2
/4
98
Figure I-A-3
-------
Z)
O
O
z
X
ป-
of.
UJ
t-
\u
O
vy
Of.
O
i
V)
ru.u
80.0
70.0
60.0
50.0
40.0
30.0
20.0
10.0
9.0
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
.9
.8
.7
.6
.5
.4
.3
.2
.1
VV'
>\
\x
v\
v
\
(
INTERVAL BETWEEN STORMS
DCCURRENCE FOR MONTHS <
JANUARY & JULY
DF
V (INTERVALS GREATER THAN 20 HOURS)
\
\\
\ \
\
\
\
\
\
\
\
\
y
\
\
\
\
JANUARY
*
\
\
\
\
\
\
\
\
\
\
\
\
X
\
\
OCTOBER
L
\
\
\
\
\
\
\
10 15 20 25
LENGTH OF INTERVAL (DAYS)
30
35
Figure I-A-4
-------
CHAPTER I
SECTION B
RUNOFF TO ESTUARY
Fresh water inflow to the Delaware Estuary is largely from drainage
of the upper and central sections of the Delaware River Basin, i.e., the
area above Trenton. The fresh water flow progresses over a series of rock
ledges at the Fall Line and becomes mixed with the waters of the Delaware
Estuary at Trenton, New Jersey. At this point the tidal motion of the
estuary becomes a limiting variable on the downstream movement of fresh
water inflow. In examining the dynamics of the estuary, the total fresh
water inflow to the estuary must be considered, i.e., from (1) the basin
above Trenton, (2) tributaries entering the estuary below Trenton and (3)
water and waste discharges not reflected in either (1) or (2).
To provide future estimates of expected inflow to the Delaware Estuary,
it is well to examine the existing conditions. Within the 86 mile reach
of the Delaware Estuary, there are approximately one hundred tributaries.
The major tributaries are gaged; however, the fresh water flow resulting
from the ungaged tributaries must be estimated. Future controls of inflow
due to upstream storage are discussed in the latter part of this section.
FRESH WATER INFLOW AT TRENTON
The Delaware River Basin above Trenton drains an area that extends as
far as the western slopes of the Catskill Mountains of New York State.
Runoff from the rugged terrain of Pennsylvania and New York is transported
to the mainstem of the Delaware River through the discharges of the numerous
tributaries. The collective flows from above the Fall Line enter the
estuary at Trenton, New Jersey. This flow is referred to hereafter as the
Trenton flow.
Gage height records of the Trenton flow are available from the courtesy
of the U. S. Department of Interior's Geological Survey. Daily discharges,
in cubic feet per second, are published annually by the Geological Survey.
For the 52 year period of record starting in 1913, the unadjusted Trenton
flow is 11,680 cfs'?). This unadjusted flow does not include the withdrawal
by the City of Trenton of approximately 50 cfs and the diversion for the
New York City water supply. Table 9 lists the respective mean annual un-
adjusted Trenton flows with corresponding recurrence interval. For flow
uniformity, the Trenton mean annual flows are based on the climatic year
(April-March). Low-flow frequency curves were constructed for the existing
data of the Trenton flow discharges.
This section was prepared by Wayne A. Blackard, U.S. Environmental Protection
Agency, Rockville, Md.
- 15 -
-------
TABLE 9
DELAWARE RIVER AT TRENTON, N. J.
MEAN ANNUAL FLOW
(cfs) (1913-1963)
(CLIMATIC YEAR)
Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Recurrence
Interval
(Years)
52.00
26.00
17.33
13.00
10.40
8.67
7.43
6.50
5.77
5.20
4.73
4.33
4.00
3.71
3.47
3.25
3.06
2.89
2.74
2.60
2.48
2.36
2.26
2.17
2.08
2.00
1.93
1.86
1.79
1.73
1.68
1.63
1.58
1.53
1.49
Probability
(% of Time)
<
1.9
3.8
5.7
7.7
9.6
11.5
13.5
15.4
17.3
19.2
21.1
23.1
25.0
27.0
28.8
30.8
32.7
34.6
36.5
38.5
40.3
42.4
44.2
46.1
48.1
50.0
51.8
53.8
55.9
57.8
59.5
61.3
63.3
65.4
67.1
Mean
Annual Flow*
(cfs)
5,755
7,695
7,820
7,970
8,327
9,051
9,169
9,339
9,528
9,649
9,723
9,741
9,750
9,751
9,940
10,062
10,142
10,205
10,268
10,389
10,688
10,738
10,935
11,453
11,611
11,646
11,701
11,739
11,751
11,970
11,989
12,082
12,417
12,606
12,706
*Unadjusted for diversions
- 16 -
-------
TABLE 9 (Cont'd)
DELAWARE RIVER AT TRENTON, N. J.
MEAN ANNUAL FLOW
(cfs) (1913-1963)
(CLIMATIC YEAR)
Rank
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
Recurrence
Interval
(Years )
1.44
1.41
1.36
1.33
1.30
1.27
1.24
1.21
1.18
1.16
1.13
1.11
1.08
1.06
1.04
1.02
Probability
(% of Time)
<
69.4
70.9
73.5
75.2
76.9
78.7
80.6
82.6
84.7
86.2
88.5
90.1
92.6
94.3
96.2
98.0
Mean
Annual Flow*
(cfs)
12,758
12,843
13,533
13,564
13,803
14,104
14,234
14,903
14,950
15,018
15,115
15,151
15 , 315
15,920
16,523
17,530
*Unadjusted for diversions
- 17 -
-------
These low-flow frequency curves were constructed as follows:
1. Rank the flows (Trenton mean annual) in ascending order
with reference to the climatic year.
2. Compute the recurrence interval by the relationship:
M
1
where: T = Recurrence interval
P = Probability of a flow (Trenton mean
annual) occurring that is equal or less
than the predicted flow.
M = Rank in order of magnitude
N = Number of years record
The computations described for the Trenton mean annual flow are
presented in Table 9. The resulting probability graph (Figure I-B-1) is
the low-flow frequency curve for the Trenton mean annual flow. Table 10,
taken from the low-flow frequency curves, summarizes the Trenton mean
annual flows for the indicated recurrence interval.
MONTHLY FLOW SEQUENCES
It is often helpful to estimate the within-year variations of a given
mean annual flow. The method that will be discussed here was developed
by Frank H. Rainwater, FWPCA, and is one of many techniques used for
estimating monthly fluctuations of the annual flow.
This technique involves the following steps:
a. A low-flow frequency distribution for the annual mean stream
flow is constructed.
b. The median monthly mean flow for each month is computed.
c. The ratio of the median monthly flow to the annual mean
flow for each month is determined. See Table 11.
- 18 -
-------
d. These ratios are applied to the annual mean discharge at
the desired recurrence interval to obtain the flow
distribution.
The ratios as applied to the annual flows for the various recurrence
intervals are represented by Table 12. It should be recalled that the
monthly flow sequences are upper bound estimates, i.e., the flows would
be equal to or less than that shown with the given probability.
LOW-FLOW FREQUENCY CURVES - TRENTON FLOW
When estimating the interaction between fresh water flow and water
quality it is important to investigate low-flow regimes other than the
annual. Prolonged lower flows may exist that are not represented by
mean annual conditions. To describe the characteristics of the major
tributaries, low-flow frequency curves were computed for durations of
7, 30, and 120 days. Construction of these distributions was as pre-
viously described, with the substitution of the mean low-flow of the
desired duration for the mean annual flow. Table 13 and Figure I-B-2
present the 7, 30, and 120 day low-flow frequency curves for Trenton.
Table 14 summarizes the unadjusted mean low-flows for the respective
recurrence intervals.
TABLE 10
TRENTON MEAN ANNUAL LOW FLOW
Recurrence Probability of Occurrence Annual Average
Interval (% of time flows are equal Flow
(Years) to or less than) (cfs)
1.11 90 15,500
5 20 9,400
10 10 8,500
25 4 7,600
50 2 7,100
- 19 -
-------
TABLE 11
Ratio of Median Monthly Flow
to Annual Mean Flow at Trenton
(1913-1963 Climatic Year)
Month
Median Monthly Flow
Ratio =
Median Monthly Flow
4
5
6
7
8
9
10
11
12
1
2
3
(cfs)
23,690
11,900
6,970
5,120
4,050
3,850
4,515
9,630
10,500
11,200
11,000
20,970
Sum = 123,395
Annual Mean
2.30
1.16
0.68
0.50
0.39
0.37
0.44
0.94
1.02
1.09
1.07
2.04
12.00
Annual Mean
123,395
10,283 cfs
- 20 -
-------
TABLE 12
DELAWARE RIVER MONTHLY FLOWS
(RAINWATER COMPUTATIONAL METHOD)
Month
4
5
6
7
8
9
10
11
12
1
2
3
Recurrence Interval (Years)
1.11
5
10
25
50
Probability of Occurrence
(% of time flows are equal to or less than)
90
35,650
18,000
10,550
7,750
6,050
5,750
6,800
14,550
15,800
16,900
16,600
31,600
20
21,600
10,900
6,400
4,700
3,650
3,500
4,150
8,850
9,600
10,250
10,050
19,200
10
19,550
9,850
5,800
4,250
3,300
3,150
3,750
8,000
8,650
9,250
9,100
17,350
4
17,500
8,800
5,200
3,800
2,950
2,800
3,350
7,150
7,750
8,300
8,150
15,500
2
16,350
8,250
4,850
3,550
2,750
2,650
3,100
6,650
7,250
7,750
7,600
14,500
Flows rounded to the nearest 50 cfs.
- 21 -
-------
TABLE 13
DELAWARE RIVER AT TRENTON, N. J.
LOWEST CONSECUTIVE DAY MEAN DISCHARGE
(1913-1965 CLIMATIC YEAR)
Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Recurrence
Interval
(years)
54.00
27.00
18.00
14.00
11.00
9.00
7.70
6.70
6.00
5.40
4.90
4.50
4.20
3.80
3.60
3.40
3.20
3.00
2.80
2.70
2.60
2.40
2.30
2.25
2.16
2.10
2.00
1.90
1.86
1.80
1.74
1.68
1.63
1.58
1.54
Probability
(% of time)
<
1.9
3.7
5.6
7.1
9.1
11.1
13.0
14.9
16.7
18.5
20.4
22.2
23.8
26.3
27.8
29.4
31.3
33.3
35.7
37.0
38.5
41.7
43.5
44.4
46.3
47.6
50.0
52.6
53.8
55.6
57.5
59.5
61.3
63.3
64.9
Flow
7 -Day
1,309
1,340
1,351
1,423
1,493
1,560
1,614
1,622
1,627
1,639
1,661
1,669
1,670
1,697
1,737
1,781
1,836
1,871
1,890
1,951
1,951
2,000
2,000
2,019
2,021
2,050
2,119
2,238
2,244
2,390
2,563
2,570
2,691
2,721
2,762
Flow
30- Day
1,545
1,582
1,615
1,617
1,715
1,740
1,777
1,781
1,820
1,826
1,887
1,901
1,904
1,909
2,022
2,073
2,074
2,162
2,170
2,185
2,285
2,292
2,357
2,454
2,639
2,673
2,691
2,737
2,785
2,854
2,947
3,024
3,043
3,074
3,113
Flow
120-Day
1,966
2,150
2,233
2,312
2,353
2,355
2,446
2,525
2,544
2,546
2,607
2,742
2,777
2,784
2,845
2,944
2,974
3,145
3,222
3,377
3,606
3,810
3,901
3,929
4,104
4,161
4,281
4,283
4,343
4,394
4,752
4,774
4,889
5,143
5,535
- 22 -
-------
TABLE 13 (Cont'd)
DELAWARE RIVER AT TRENTON, N. J.
LOWEST CONSECUTIVE DAY MEAN DISCHARGE
(1913-1965 CLIMATIC YEAR)
Rank
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
Recurrence
Interval
(years)
1.50
1.45
1.42
1.38
1.35
1.31
1.28
1.25
1.22
1.20
1.17
1.14
1.12
1.10
1.08
1.05
1.03
1.01
Probability
(% of time)
<
66.7
69.0
70.4
72.5
74.1
76.3
78.1
80.0
82.0
83.3
85.5
87.7
89.3
90.9
92.6
95.2
97.1
99.0
Flow
7-Day
2,773
2,793
2,796
2,844
2,899
2,990
3,213
3,214
3,231
3,259
3,297
3,397
3,404
3,947
4,012
4,536
4,604
6,574
Flow
30- Day
3,161
3,196
3,285
3,327
3,433
3,438
3,467
3,794
3,937
4,003
4,027
4,037
4,252
5,056
5,794
5,867
5,939
7,974
Flow
120-Day
5,586
5,611
5,677
5,996
6,021
6,204
6,357
6,363
6,462
6,744
7,184
7,694
7,790
8,039
8,667
9,499
12,226
14,142
- 23 -
-------
TABLE 14
DELAWARE RIVER AT TRENTON, N. J.
MEAN LOW FLOW
(1913-1963 CLIMATIC YEAR)
Recurrence
Interval
(Years)
5
10
25
50
Probability of
Occurrence
(% of time flows
are equal to or
less than)
20
10
4
2
Mean Low Flow
(cfs)
7-Day
1,600
1,500
1,400
1,300
30-Day
1,900
1,700
1,600
1,500
120-Day
2,700
2,400
2,100
2,000
Flows rounded to 100 cfs.
- 24 -
-------
TIME SERIES ANALYSES - TRENTON FLOW
To better understand the nature of the flow variability of the Del-
aware at Trenton, several time series analyses were performed. Two groups
of data were analyzed; mean monthly flows from 1914-1963 (600 values) and
mean daily flows for 1960-1962 (1095 values). The latter group was ana-
lyzed to provide some insight into the day to day variability while the
former group was examined to obtain information on the month to month vari-
ations. By far the greater amount of information was obtained from the
monthly data.
The autocovariance function given by
N-T _ _
R(T) = I (qt-q) (qt+T - q)
t=l N-t (1),
where
qt = flow at time t; t = 1,2,3,...N
q = mean value of q
T = time lag, T = 0,l,2,...m
was computed for each group. The formal Fourier transformation of this
function, known as the power spectrum was also estimated for each group
utilizing the following relationships:
m-1 irir
V = Ai[R. + 2 ฃ R^ cos m +R cos rir] r=0,l,2,...m (2)
r 0 i=1 i
UQ = 1/2 (VQ + Vl)
U - 1/4 V . +1/2 V +1/4 V.., 1 = r = m-1
r r-1 r r+1
U = 1/4 V . +1/2 V
m m-1 m
where
and
V = the estimate of the "unsmoothed" record (contains all harmonics)
U = the "smoothed" estimate of the power spectrum (removal of some
harmonics).
- 25 -
-------
The power spectrum analysis basically distributes the total variance
of a given record into individual frequency "bands". As such, it provides
information on any predominant frequencies in the record that may not be
obvious from a mere visual inspection. Frequency as used here refers to a
time definition (cycles per time) as opposed to a probabilistic definition.
The reciprocal of frequency is therefore the period of the particular phe-
nomenon. More detailed information on the use of the power spectrum is
given in References 8 and 9.
For some of the analyses, the "removal" of the annual flow variation
was carried out via a harmonic analysis. The details of this procedure
are given in Chapter IVB - Statistical Analyses.
Figure I-B-3 presents the 100 lag autocovariance function and Figure
I-B-4 presents the power spectrum for the monthly flow series both before
and after the removal of the annual periodicity (the 50th harmonic). The
periodicity shown in the autocovariance function of the data before re-
moval results in the pronounced peak in the spectrum of Figure I-B-4 and
is a direct result of the annual periodicity in flow.
The annual harmonic which was removed from the record accounted for
about 23% of the total variance and had an amplitude about the mean
(12,340 cfs) of 6800 cfs and a time of maximum at 60 days from January 1
(about March 1). Removal of this harmonic had a pronounced effect on the
analyses as shown in Figures I-B-3 and I-B-4. The autocovariance function
after removal oscillates around zero without any noticeable features.
However, the spectrums indicate that both before and after removal of the
annual harmonic, some significant peaks appear to be present at the 2nd
(6 month) and 3rd (4 month) harmonics of the annual. In addition, some
low frequency (long period) components on the order of 2 and 5 years also
appear to be present. These peaks are not evidenced in the autocovariance
function and one may be misled into thinking that the spectrum of the func-
tion after removal shown in Figure I-B-3 should be completely flat. Figure
I-B-4 spectrums indicate that these peaks are still present. Indeed, it
is these harmonics (which as shown, have amplitudes of greater than 1000
cfs) which generate the "blockiness" in a flow record; i.e., the tendency
for records not to be exactly periodic but rather sharply "steeped" during
the spring and.fall. From a systems analysis viewpoint, one may visualize
these higher harmonics as the result of passing an annual variability in
rainfall through a nonlinear drainage basin; the non-linearities giving
rise to the higher harmonics.
The analyses of the daily data for the three year period (1960-1962)
indicated a predominance of low frequency components with the spectrum
decreasing in an exponential fashion with frequency. This is a reflection
of the general persistence of day to day type of flow phenomena.
- 26 -
-------
SYNTHETIC GENERATION OF FLOW AT TRENTON
Mathematical analysis of the time variability of quality in the estuary
requires relatively long periods of flow records which are utilized as in-
put data into the large dynamic dissolved oxygen model (see Chapter IV) .
The technique used to generate flow sequences for the Delaware at Trenton
is fully described in several references (10ปHป^ ' and will not be discussed
in detail here. Briefly, the historical flow record is used to obtain re-
gression coefficients between the flows in successive months. The first
step is to compute the linear regression coefficients between the flow in
the jth month and the flow in the j+lst month, designated as $j . This is
done for each of the twelve months. These coefficients are related to the
normalized autocovariance values at lag one discussed previously. In order
to preserve both the mean and the variance of the historical trace, a ran-
dom component must be added. The end result is the following equation which
is used to generate the synthetic flows :
-------
The four curves shown in Figure I-B-5 indicate that for all practical
purposes the spectrum is preserved in generating the flow sequence. There
are some minor deviations, especially in the low frequency end of the spec-
trum where several of the synthetic traces did not apparently preserve
the 2 and 5 year peaks although there is some hint of peaks in the first
one hundred year trace.
As indicated, the results of this synthetic generation will be used
more extensively in the simulation of long term water quality variations
under particular waste control schemes.
LOW-FLOW FREQUENCY CURVES FOR SCHUYLKILL RIVER
The major tributary of the Delaware Estuary is the Schuylkill River.
The discharges from the Schuylkill River enter the estuary at Philadelphia,
Penna. at mile 92.04 (confluence of Schuylkill River and Delaware Estuary).
Approximately 8.5 miles upstream from the confluence is a physical obstruc-
tion (Fairmount Dam) separating the estuarine and fresh water sections of
the Schuylkill River. The drainage area above Fairmount Dam is 1893 square
miles. For the 33 year period of record starting in 1931 the adjusted mean
annual flow of the Schuylkill River is 2900 cfs. The adjusted flow includes
the withdrawal of approximately 300 cfs by the City of Philadelphia.
To be consistent in comparing the respective stream flows, low-flow
frequency curves were constructed for the Schuylkill River. Tabulated in
Table 15 and presented in Figure I-B-6 are the 7, 30, and 120-day low-
flow frequency curves for the Schuylkill River.
Table 16 summarizes the Schuylkill River adjusted mean low flows.
CONTROL OF INFLOW BY UPSTREAM STORAGE
The inflow to the estuary at Trenton can be controlled by releases
from present and proposed upstream storage. Information concerning the
effect of various flow regimes on water quality can be obtained from
Chapter IV or from the report "Water Quality Control Study, Tocks Island
Reservoir".(13)
The primary inflow functions to the Delaware Estuary are: 1) the
Trenton flow as recorded at Trenton, New Jersey, and 2) the Schuylkill
flow as recorded at Philadelphia, Penna. Present (1965) minimum average
observed daily flows for these primary focal points of inflow are 2500 cfs
and 350 cfs, respectively. The minimum average daily flow at Trenton
(2500 cfs) is for the record drought (to 1963) and is based upon releases
from New York City reservoirs maintaining 1750 cfs at Montague. The Schuyl-
kill flows are not presently regulated by upstream storage; however, they
have been adjusted to include the diversions by the City of Philadelphia.
- 28 -
-------
TABLE 15
SCHUYLKILL RIVER AT PHILADELPHIA, PENNA.
LOWEST CONSECUTIVE DAY MEAN DISCHARGE (CFS)
(1932-1964)
(CLIMATIC YEAR)
Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Recurrence
Interval
(years)
34.00
17.00
11.33
8.50
6.80
5.67
4.86
4.25
3.78
3.40
3.09
2.83
2.62
2.43
2.27
2.13
2.00
1.89
1.79
1.70
1.62
1.54
1.48
1.42
1.36
1.31
1.26
1.21
1.17
1.13
1.10
1.06
1.03
Probability
(% of time)
<
2.9
5.9
8.8
11.8
14.7
17.6
20.6
23.5
26.5
29.4
32.4
35.3
38.2
41.2
44.1
46.9
50.0
52.9
55.9
58.8
61.7
64.9
67.6
70.4
73.5
76.3
79.4
82.6
85.5
88.5
90.9
94.3
97.1
Flow
7-Day
292
300
314
336
349
380
389
407
434
438
461
473
495
502
507
507
560
577
610
615
616
623
631
672
680
682
720
744
760
835
860
909
1,297
Flow
30-Day
324
332
346
357
452
459
460
503
512
517
537
544
569
570
603
611
688
726
730
751
756
770
784
788
798
827
838
1,072
1,079
1,136
1,205
1,212
1,571
Flow
120-Day
434
495
516
548
600
651
683
709
777
777
855
873
920
1,007
1,040
1,143
1,149
1,159
1,183
1,206
1,373
1,396
1,442
1,512
1,537
1,552
1,662
1,756
1,844
1,994
2,394
2,407
3,149
- 29 -
-------
TABLE 16
SCHUYLKILL RIVER AT PHILADELPHIA, PENNA.
ADJUSTED MEAN LOW FLOWS
(1932-1964)
Recurrence
Interval
(Years )
5
10
25
50
Probability of
Occurrence
(% of time flow
is equal to
or less than)
20
10
4
2
Mean Low Flow
(cfs)
7- Day
390
330
300
280
30- Day
440
360
330
320
120- Day
700
540
460
410
- 30 -
-------
Studies by the Delaware Estuary Comprehensive Project have indicated the
need for chloride control in the Delaware Estuary. Chloride control can be
obtained through the implementation of stream flow regulation, i.e., the
conservation of the Delaware River water resources.
The Corps of Engineers report (14) provides detailed descriptions of
the construction and operation of a number of dams and reservoirs on the
Delaware River and its tributaries. Further investigations on the effects
of fresh water inflow on chloride control are discussed in the locks Island
Report(I'). For the purposes described herein the incremental fresh water
flows due to the estimated storage of the proposed reservoirs are presented
in Table 17.
Upon completion of the presently proposed reservoirs, the incremental
flow additions through stream flow regulation is estimated to be 1194 cfs.
The fresh water increase for Trenton and the Schuylkill are 995 cfs and
199 cfs, respectively.
DISTRIBUTION OF RUNOFF ALONG ESTUARY
Incremental fresh water flows to the Delaware Estuary result from
tributary discharges. In some cases, the flows can be accurately de-
termined, i.e., for the gaged tributaries; however, estimates must be
made for the ungaged tributaries.
The Delaware Estuary Study area drainage basin was divided into
sub-basins as presented in Figure I-B-7. The sub-basins were chosen re-
lative to the sections of the estuary used for computational purposes
(see Chapter IVH). Application to the sub-basins of a runoff per
square mile coefficient provides the incremental flow per estuary sec-
tion.
For the gaged tributary areas (U.S.G.S. gaging Stations) the run-
off per square mile relationships were calculated from the mean annual
flow per square mile of drainage area. Applying these coefficients to
the ungaged areas results in their corresponding average annual dis-
charges. A summary of the mean annual discharges per estuary section
is tabulated in Table 18.
- 31 -
-------
TABLE 17
CUMULATIVE FLOWS
Corps of
Engineers
Project
None
Beltzville
Tocks Island
Blue Marsh
Trexler
Prompt on
Aquashicola
Maiden Creek
Bear Creek
Flow
Increments
CFS
-
94 (a)
530 (b)
65
55
57
63
134
196
Cumulative
Delaware
River
Trenton
2500 (c)
2594
3124
-
3179
3236
3299
-
3495
Minimum Flows - CFS
Schuylkill
River
@
Phi la.
350
-
-
415
-
-
-
549
_
Total
2850
2944
3474
3539
3594
3651
3714
3848
4044
(a) Assumes 56 cfs for water supply and 38 cfs for water quality
control.
(b) Of a total flow increment of 980 cfs, 450 cfs is planned for
out of the basin diversion and hence is not included.
(c) Minimum average daily flow during record drought (to 1963)
and is based upon releases from New York City reservoirs
maintaining 1750 cfs at Montague.
- 32 -
-------
TABLE 18
RUNOFF AND DRAINAGE AREAS PER ESTUARY SECTION
Estuary
Section
Above 1
1
1
1
2
2
2
3
3
4
4
5
5
5
6
6
6
6
7
7
8
8
9
9
10
10
Tributary
Delaware River
Assunpink Creek
Crosswicks Creek
Neshaminy Creek
N.B.Rancocas Creek
S.B.Rancocas Creek
Location
Trenton, N.J.
Trenton, N.J.
Penna .
New Jersey
Extonville,N.J.
Penna .
New Jersey
Penna .
New Jersey
Penna.
New Jersey
Langhorne,Pa.
Penna .
New Jersey
Pemberton,N.J.
Vincetown,N.J.
Penna.
New Jersey
Penna .
New Jersey
Penna .
New Jersey
Penna .
New Jersey
Penna .
New Jersey
Station
Code
0
3
1
2
5
4
6
7
8
9
10
11
12
13
16
17
14
15
18
19
20
21
22
23
24
25
Period
of
Record
1912-64
1923-64
1940-64
1934-64
1921-64
1961-64
Drainage
Area
(Sq.
Miles )
6,780
89.4
5.1
3.1
83.6
4.3
102.3
17.4
7.1
14.8
49.8
210
35.2
7.8
111
53.3
28.3
160.6
50.6
14.9
11.1
30.7
31.2
3.9
5.7
3.9
Mean
Annual
Flow
(cfs)
11,680
119
7.5
4.6
126
6.4
151.4
25.8
10.5
21.9
73.7
274
52.1
11.5
169
78.9
41.9
273.7
74.9
22.1
16.4
45.4
46.2
5.8
8.4
5.8
Runoff
per Sq.
Mile
(cfs)
1.72
1.33
(1.48)
(1.48)
1.51
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
1.31
(1.48)
(1.48)
1.52
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
Ace. Drain age
Area
(Sq. Mile)
6,780
6,877.6
7,067.8
7,092.3
7,156.9
7,409.9
7,763.1
7,828.6
7,870.4
7,905.5
7,915.1
Ace. Flows
Mean Annual
(cfs)
11,680
11,811
12,095
12,131
12,227
12,564
13,128
13,225
13,287
13,339
13,359
U)
OJ
-------
TABLE 18 (Cont'd.)
RUNOFF AND DRAINAGE AREAS PER ESTUARY SECTION
Estuary
Section
11
11
11
12
12
13
13
14
14
15
15
15
16
16
16
17
17
17
17
18
18
18
19
19
20
20
Tributary
Cooper River
Schuylkill River
Mantua Creek
Darby Creek
Cobbs Creek
Chester Creek
Location
Haddonfield.N.J.
Penna .
New Jersey
Penna .
New Jersey
Penna .
New Jersey
Penna.
New Jersey
Phila. , Penna.
Penna .
New Jersey
Pitman, N.J.
Penna .
New Jersey
Darby, Pa.
Darby, Pa.
Penna .
New Jersey
Chester, Pa.
Penna .
New Jersey
Penna. & Del.
New Jersey
Delaware
New Jersey
Station
Code
28
26
27
29
30
31
32
33
34
35
36
37
40
38
39
41
42
43
44
45
46
47
48
49
50
51
Period
of
Record
1963-64
1931-64
1940-64
1964-64
1964-64
1931-64
Drainage
Area
(Sq.
Miles)
17.4
4.8
30.3
3.4
0.8
2.5
10.0
0.8
62.0
1,893
19.9
14.5
6.8
6.2
55.1
37.4
22.0
81.7
19.0
61.1
14.0
58.8
14.3
46.6
15.0
3.9
Mean
Annual
Flow
(cfs)
25.8
7.1
44.8
5.0
1.2
3.7
14.8
1.2
91.8
2,900
29.5
21.5
11.5
9.2
81.5
55.4
32.6
120.9
28.1
80.8
20.7
87.0
21.2
69.0
22.2
57.7
Runoff
per Sq.
Mile
(cfs)
A
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
1.53
(1.48)
(1.48)
1.69
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
1.32
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
Ace. Drainage
Area
(Sq. Mile)
7,967.6
7,971.8
7,984.3
8,047.1
9,974.5
10,042.6
10,202.7
10,336.6
10,397.5
10,416.4
Ace . Flows
Mean Annual
(cfs)
13,392
13,398
13,416
13,509
16,460
16,563
16,800
16,988
17,078
17,158
I
LO
-------
TABLE 18 (Cont'd.)
RUNOFF AND DRAINAGE AREAS PER ESTUARY SECTION
Estuary
Section
21
21
21
21
21
21
22
22
23
23
24
24
25
25
26
26
27
27
28
28
28
29
29
29
30
30
Tributary
Brandywine River
Christina River
White Clay Creek
Red Clay Creek
Salem River
Alloway Creek
Location
Wilmington , Del .
Cooch's Bridge,
Del.
Newark , Del .
Wooddale, Del.
Delaware
New Jersey
Delaware
New Jersey
Delaware
New Jersey
Delaware
New Jersey
Delaware
New Jersey
Delaware
New Jersey
Delaware
New Jersey
Woods town, N.J.
Delaware
New Jersey
Alloway, N.J.
Delaware
New Jersey
Delaware
New Jersey
Station
Code
55
54
53
52
56
57
58
59
60
61
62
63
64
65
66
67
68
69
72
70
71
75
73
74
76
77
Period
of
Record
1946-64
1943-63
1931-63
1943-63
1940-64
1952-64
Drainage
Area
(Sq.
Miles)
314
20.5
87.8
47.0
79.4
2.0
2.7
2.1
3.0
1.9
3.9
2.7
12.2
1.3
20.1
2.0
1.1
4.1
14.6
29.4
100.6
21.9
14.6
43.3
80.6
5.5
Mean
Annual
Flow
(cfs)
453
26.2
108
63.1
117.5
3.0
4.0
3.1
4.4
2.1
5.8
4.0
18.1
1.9
29.7
3.0
1.6
6.1
19.0
43.5
148.9
24.4
21.6
64.1
119.3
8.1
Runoff
per Sq.
Mile
(cfs)
1.44
1.28
1.23
1.34
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
(1.48)
1.30
(1.48)
(1.48)
1.11
(1.48)
(1.48)
(1.48)
(1.48)
Ace . Drainage
Area
(Sq. Mile)
10,967.1
10,971.9
10,976.8
10,983.4
10,996.9
11,019.0
11,024.2
11,168.8
11,248.6
11,334.7
Ace. Flows
Mean Annual
(cfs)
17,929
17,936
17,943
17,952
17,972
18,005
18,013
18,224
18,334
18,462
tn
I
*(1.48) = Average of Gaged Runoff Coefficients.
-------
RECURRENCE INTERVAL IN YEARS
CD
C
-*
CD
CP
100,000
90,000
80,000
70,000
60,000
50,000
40,000
30,000
ฃ 20,000
u
IU
o
et
<
X
ซJ
^*
r^^
D
^
^>^
LOW FL
ELAWAF
f**\'
rr^
t
OW FREQUENCY CURVE .
IE RIVER AT TRENTON,N.J.
(1913-1963)
CLIMATIC YEAR
0.01 0.05 0.1 0.2 0.5 1
10 20 30 40 50 60 70 80 90 95
% OF TIME FLOW IS EQUAL TO OR LESS THAN
98 99
99.8 99.9 99.99
-------
RECURRENCE INTERVAL IN YEARS
(Q
C
-i
CD
DO
<
I
u
5
<
3E
100,000.
90,000
80,000
70,000
60,000
50,000
40,000
30,000
20,000
10,000
9,000
8,000
7,000
6,000
5,000
4,000
3,000
2,000
1,000
1,000 200100 50 20 10 5 2 1.25 1.05
*--'
~~~
^
*~
-P
~ป^
^^
f
S
S
^
,t
X
*2
s
^
r
C
/
s
s*-'
.
^X
*
sS
.S?-'
LOW FL
ELAWAI
i
>X
'
/
^
x
*
X
x
yS\
/
x
S*^
^
120 DAY
.^
''SO DAY
' ^^
7 DAY
OW FREQUENCY CURVE _
RE RIVER AT TRENTOKNJ
(1913-1965)
CLIMATIC YEAR
0.01 0.05 0.1 0.2 0.5 1
10 20 30 40 50 60 70 80 90 95
% OF TIME FLOW IS EQUAL TO OR LESS THAN
98 99
99.8 99.9
99.99
-------
1.0
.8
.6
.4
(Q 2
C oi
3 ?
F .2
CO 5T
-.2
-.4
I | I I I I I
AUTOCOVARIANCE FUNCTION FOR MEAN MONTHLY FLOW
AT TRENTON,N.J. 1914-1963
BEFORE REMOVAL OF ANNUAL HARMONIC
AFTER REMOVAL OF ANNUAL HARMONIC
10
20
30
40
50 60
(T-) MONTHS
70
80
90
100
-------
12.0
11.0
10.0
9.0
-o
2 8.0
3 I
C 5
ฎ ง 6.0
45^ >: 5.0
u
3.0
2.0
1.0
to
QC
UJ
>-
>o
to
at
10
Z
O
to
x
i-
z
O
s.
Z
o
SPECTRA MEAN MONTHLY FLOW
AT TRENTON,N.J. _
1914-1963
BEFORE REMOVAL OF ANNUAL HARMONIC
AFTER REMOVAL OF ANNUAL HARMONIC
20
40 50 60
CYCLES/200 MONTHS
70
80
90
100
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
1,000 200100 50 20 10
RECURRENCE INTERVAL IN YEARS
5 2 1.25
1.05
(Q
C
-i
0>
(>
9,000
8,000
7,000
6,000
5,000
4,000
3,000
2,000
"3T
u
o 1,000
5 900
5 800
2 700
o
Z 600
| 500
400
300
200
100
0
e
**
-ป
. - -
^
_j
^=BB-
^f
f-
~2
X
's
^^
/
^
y
/^
^
x
s
.s
r
X
.
>'
X
X
,x
sc
X
.*
X
-^
s*-
x
X
X
x .
^
^r-
LOW FL
:HUYLKII
X
x
^x^
X
X
X
X
/"
^
X
120 DAY
30 DAY
7 DAY
OW FREQUENCY CURVE
1 RIVER AT PHILADELPHIA
(1932-1964)
CLIMATIC YEAR
i ii ii
.01 0.05 0.1 0.2 0.5 1 2 5 10 20 30 40 50 60 70 80 90 95 98 99 99.8 99.9 99.
% OF TIME FLOW IS EQUAL TO OR LESS THAN
99
-------
GAGED AREAS
GAGED AREAS
( PARTIAL DATA)
GAGING STATIONS
-------
CHAPTER I
SECTION C
ESTUARINE DYNAMICS
TIDAL CURRENT INVESTIGATIONS
To fill some of the gaps in existing hydrodynamic data on the Delaware
Estuary, the DECS embarked on a tidal current measurement program. The
objectives of this program were to:
a', more accurately define the current pattern
over a tidal cycle;
b. relate long and short term changes in the tidal current pattern
to climatological data (wind, etc.) and fresh water inflow;
c. determine net downstream movement over a tidal cycle as a
function of fresh water inflow;
d. attempt to qualitatively evaluate turbulence patterns from
current velocity and direction data.
Experiments conducted at the U. S. Army Engineers Waterways Experiment
Station, Vicksburg, Miss., on the physical model of the Delaware Estuary
have aided in the development of information relative to tidal diffusion,
advection, time of travel and distribution of wastes(^^'. The U.S.G.S.
maintains data measuring stations on the Delaware Estuary to record
fresh water inflow (Delaware River at Trenton, Schuylkill River, other
tributaries) and tide height at several points along the estuary0-6).
Periodic measurements of tidal current velocities are also conducted at
selected stations to obtain correlations with the tide stage and slope
data for the computation of mass movement over a tidal cycle^?) (18) ^
The entire length of the Delaware Estuary from Trenton, N. J., to
Delaware Bay contains a shipping channel with an average depth of 35'-45'
and a width of 300'-1000'. Within this channel and its associated anchor-
ages moves the greater mass of water influenced by tidal action. Currents
in the estuary are almost entirely of tidal origin. Proximity to the
ocean through Delaware Bay permits the occurrence of regular tidal cycles
of 12 hours and 25 minutes duration. In the lower estuary the tidal
phenomenon produces a change in current direction of 180 every six hours
and 12 minutes, with current velocities for the flood and ebb tides being
approximately equal. As one progresses upstream, the physical configura-
tion of the estuary gradually alters the effects of the tidal phenomenon.
Localized shore to shore restrictions produce higher current velocities;
This section was prepared by Albert W. Bromberg, U.S. Environmental
Protection Agency, Edison, N.J.
- 36 -
-------
the gradual narrowing of the estuary channel from the bay to Trenton pro-
duces greater tidal variation with distance from the ocean; the influence
of surface water and ground water inflow alters the length and velocity of
the ebb versus the flood tide. Tidal height, as affected by the combined
influence of lunar and solar bodies, changes in relation to the position
of these bodies with each other and the earth. Table 19 indicates the
range of tidal height as experienced in the Delaware Estuary.(19)
The presence of large dominant weather systems in the eastern United
States also has an effect on tidal height. An extreme example of this
occurred in the estuary between Dec. 31, 1962 and Jan. 2, 1963.(2ฐ)
A deep low pressure system remained stationary off the east coast gen-
erating strong (average 35-45 MPH) northwest winds. These winds, blow-
ing approximately parallel to the axis of the estuary offset the tidal
phenomenon for the entire period. High tide levels were considerably below
normal low tide levels for the entire period, producing navigation hazards
and exposing many industrial water supply intakes. When the winds ceased,
permitting the return of normal tidal conditions, high salinity water en-
croached far beyond its seasonal normal), xhe estuary required approxi-
mately 30 days to return to normal salinity condition.
TABLE 19
TIDAL HEIGHT RANGES IN THE DELAWARE ESTUARY
Point Along
Delaware Estuary
Cape May, N.J.
Cape Henlopen, Del.
Liston Point, Del.
Philadelphia, Pa.
Trenton, N.J.
Miles from
Delaware Bay
Mouth
0
0
48.3
100.0
132.0
Tide Range
Mean Spring
(Feet) (Feet)
4.3
4.1
5.7
5.9
6.8
5.2
4.9
6.4
6.2
7.1
- 37 -
-------
CURRENT MEASUREMENT PROGRAM
The instrumentation chosen to measure current velocity was the Woods
Hole Recording Current Meter (See Figure I-C-1). It is a self-contained
digital recording meter capable of measuring current direction and speed
in a range of below 0.05 knots (.08 ft./sec.) to 5 knots (8.4 ft./sec).
All data are recorded on 16mm photographic film. A 100* roll of film is
capable of storing 9600 sets of rotor speed, vane and compass direction
readings at a film advance rate of 1/8 inch per minute. Recording con-
tinuously, the instrument can operate for six and one-half (6-1/2) days;
recording at one-hour intervals (a one minute reading per hour), it can
operate for four hundred (400) days.
All data are transmitted to the camera field of view via optical
fibers. Vane and compass are coded in Gray binary form and require seven
(7) channels each. Current speed is recorded as a series of light pulses
on two (2) channels, one each for the ones and tens count. Inclination
of the instrument from the vertical is recorded as elongated light pulses,
each pulse corresponding to 5 of tilt. Two (2) additional channels
provide a read pulse signal and a reference signal intensity.
Operational characteristics of the various instrument components are
as follows:
Savonius Rotor
Starting speed, .01 knots (.017'/sec.); maximum speed, 5.0 knots
(8.4'/sec.). Approximate rotation rate 80 rpm/knot. Accuracy
at .01 knot, ฑ 50%, greater than 0.3 knot, + 3%.
Directional Vane
Sensitive within 10ฐ at .01 knot, 2ฐ at .025 knot; resolution
2.5ฐ.
Compass
Sensitivity better than 2ฐ, resolution 2.5ฐ.
Maximum tilt angle before binding, 45 .
Inclinometer
Tilt error, 3% at 5ฐ, 6% at 10ฐ, 12% at 20ฐ, 20% at 30ฐ.
(Tilt of instrument causes low velocity readings).
The instrument is powered by a six (6) volt dry cell battery, which
is sufficient for a fourteen (14) month recording period. It is con-
structed of aluminum, weighing 90 Ibs. in air and 10 Ibs. in water. A
- 38 -
-------
tensile load of 7000 Ibs. may be applied across the Instrument in a moor-
ing system.
Heavy shipping traffic in the estuary necessitated the selection of
current measuring stations outside of the main channel but still in areas
where representative hydrodynamic conditions existed. In consultation with
the U. S. Army Corps of Engineers, the following guidelines were established
for placing semi-permanent instrument stations in the estuary:
a. Trenton to Philadelphia - no closer than 50' to the edge
of the ship channel.
b. Philadelphia to Delaware Bay - no closer than 100' to the
edge of the ship channel.
Using these guidelines, four current measurement stations were chosen along
the length of the estuary. (See Figure I-C-2).
Based on the records of the manufacturer as to the uses of the Wood's
Hole Current Meter, the DECS was the first to install them in an estuarine
system. The conditions which exist in such a system (high current velocity,
six foot tidal variation, heavy ship traffic, relatively shallow depth, etc.)
necessitated the design of a different type of mooring system. Utilizing
the equipment and facilities available through the U. S. Coast Guard,
Gloucester City, N. J., a set of mooring systems were fabricated and in-
stalled with satisfactory results.
The tidal current meters were an integral part of the mooring system.
They were suspended at a fixed depth below a surface buoy, the fixed depth
being equal to one half of the mean depth at the particular station. The
overall length of the mooring system was equal to the maximum expected depth
at high tide. (See Figure I-C-3).
Realizing the complexity of the phenomena which were under investigation,
the study attempted to measure estuary responses to input conditions which
covered as wide a range as possible. In particular, it was of interest to
measure the effect of transient peak fresh water inflows that occur at
Trenton during the spring. Unfortunately, natural occurrences prevented
the collection of data during the desired periods. The drought conditions
which prevailed in the northeastern U. S. during the study period (fall
1962 - spring 1965) minimized the occurrence of any well defined peak flows.
Although flows of greater than 60,000 cfs have occurred at Trenton in the
past, only spike flows of this magnitude were realized during the study
period. Instrumentation was not in place for either of these events because
they occurred during the winter and spring ice flow conditions. Other pro-
nounced peaks of lower magnitude occurred during the study period, but these
again were during the winter or masked by the overall flow regime.
Table 20 contains a log of all instrument installation during the study
period including the time and date of installation and removal and remarks
- 39 -
-------
pertinent to the instrument location, data recording interval and condition
of the film record.
A current measurement station was located at the Burlington-Bristol
Bridge (mile 117.5) early in the program, but was eliminated after the loss
of one instrument. Due to an instrument malfunction, no useful data were
gathered during the one recording period at this station.
The film processing and reading services provided by the instrument
manufacturer were utilized for processing all original data film records.
This service provided the following:
a. Point graph of rotor speed (knots) versus Direction (degrees).
(Figure I-C-4).
b. Polar coordinate histogram of current direction (degrees).
(Figure I-C-5).
c. Histogram of rotor speed (knots) versus number of occurrences.
(Figure I-C-6).
d. Analog strip chart plot of rotor speed, direction and
inclination.
e. Data transcribed to computer magnetic tape in IBM 7 channel
binary format.
Upon examination of the data records from each current meter station,
the degree of tilt of the instrument from the vertical during most of the
recording period was unusually high (20 -40 ). However, it was concluded
that this condition could not be avoided regardless of the type of mooring
system in view of the high velocities encountered (1.5'/sec. ave. - 3.0'/sec.
max.) based on actual current velocity measurement.
Although the study collected current data at only one point of the
cross-section at each station, data are available to obtain velocity rela-
tionships for part or all of the cross-section at the respective stations.
A current meter was on location approximately 800 yards upstream of
the Tacony-Palmyra Bridge recording continuously during the U.S.G.S. cali-
bration study at Palmyra, N. J.(18/, in May, 1963. Cross-sectional velocity
data has been collected by the U.S.G.S. at the Delaware Memorial Bridget17).
Two additional current meters were installed for a 24-hour period between
September 5-6, 1963, at the Fort Mifflin station for correlation purposes.
- 40 -
-------
TABLE 20
TIDAL CURRENT MEASURING STATIONS IN THE DELAWARE ESTUARY
LOCATION
Tacony-Palmyra Bridge
Latitude - 40ฐ 00' 50"
Longitude - 75ฐ 00' 10'
Mile 107.5
Fort Mifflin
Latitude - 39ฐ 52' 40"
Longitude - 75ฐ 12' 20"
Mile 91.8
TIME OF RECORD
1130 EST March 26, 1964 to
0830 EST May 11, 1964
0920 EST May 11, 1964 to
1620 EST May 19, 1964
0930 EDST June 9, 1964
0940 EST October 12, 1964 to
1810 EST December 21, 1964
1130 EST March 9, 1965 to
2110 EST May 26, 1965
June 4, 1963 to
June 11, 1963
1210 EST July 12, 1963 to
0730 EST December 18, 1963
1105 EST September 5, 1963
0940 EST September 6, 1963
REMARKS
Installed with directional vane.
Data recorded at one-half hour
intervals.
Installed with directional vane.
Data recorded continuously.
Instrument and Buoy reported lost
on June 22, 1964. Bottom searched
by diver with no results.
Data recorded at one-hour intervals.
Inclinometer malfunctioned for entire
recording period.
Data recorded at one-hour intervals.
Inclinometer malfunctioned for portion
of recording period.
Data recorded continuously. (Trial
run of instrument).
Data recorded at one hour intervals.
Inclinometer did not record for entire
period.
Data recorded continuously. Savonius
rotor located 9' below water surface.
No inclinometer. Special 24 hour study.
-------
TABLE 20 (Cont'd.)
TIDAL CURRENT MEASURING STATIONS IN THE DELAWARE ESTUARY
LOCATION
Fort Mifflin
Latitude - 39ฐ 52' 40"
Longitude - 75ฐ 12' 20"
Mile 91.8
Delaware Memorial Bridge
Latitude - 39ฐ 40' 25"
Longitude - 75ฐ 31' 15"
Mile 68.0
TIME OF RECORD
1120 EST September 5, 1963 to
0920 EST September 6, 1963
1815 EST March 26, 1964 to
0600 EST July 17, 1964
1120 EST October 12, 1964
0845 EST December 21, 1964
1540 EST March 30, 1965 to
0820 EST May 20, 1965
1540 EST March 26, 1964 to
1030 EST July 28, 1964
1330 EST October 12, 1964 to
1105 EST December 21, 1964
0915 EST March 9, 1965 to
1850 EST May 20, 1965
REMARKS
Data recorded continuously. No
inclinometer. Instrument located
S 38ฐ E True Bearing, 750 yards from
above position. Meter at mid-depth.
Special 24-hour study.
Data recorded at one-hour intervals.
Data recorded at one-hour intervals.
Data recorded at one-hour intervals.
Latter portion of record affected by
bad battery or a short in the camera
motor.
Data recorded at one-hour intervals.
Inclinometer record is spotty.
Data recorded at one-hour intervals.
Data recorded at one-hour intervals.
Inclinometer not operating properly-
Camera motor not operating properly
upon post-installation check.
Latter portion of data record may be
affected.
-------
RESULTS OF ANALYSES OF TIDAL CURRENT DATA
In order to analyze the data more efficiently, a computer program was
prepared which accepted the data in the form recorded by the meter and per-
formed the following operations: (a) for each location, all current speed
values within a 180ฐ arc facing downstream were designated positive (b) for
the 180ฐ arc facing upstream, the speed values were designated negative and
(c) for each value, the necessary corrections were performed to compensate
for the vertical tilt of the instrument at the time of recording. The
result of this operation was a time series of current values oscillating
around a zero value with positive values being oriented in a general down-
stream direction and negative values in a upstream direction.
As indicated previously, one of the objectives of the current measure-
ment program was to "track" a mass of fresh water inflow through the estuary.
Unfortunately, hydrologic conditions during the investigations were such as
to preclude any definitive analyses of this type. However, some useful
information was obtained through the application of time series techniques
(see Runoff to Estuary, Chapter One, Section B). A number of different
analyses were performed on the data including harmonic analysis, filtering
of data and power spectral analysis. Fourier analyses of the record for
the semi-diurnal harmonic and a group of four side-band harmonics indicated
that this band accounted for about 85% of the total variance of the record.
The semi-diurnal amplitude was about 0.98 knots.
Figure I-C-7 illustrates the results from a power spectrum computation
on the residual 15% of the data obtained at the Ft. Mifflin station during
the period October 1964 to December 1964. The predominance of the semi-
diurnal tidal effect in this residual variance is quite striking but of
equal interest is the occurrence of the second and third harmonics of the
12.4 hour phenomena. Indeed, these latter two periodicities account for
as much variance as the diurnal harmonic indicating the relative significance
of these higher harmonics. The presence of these higher harmonics may be a
result of the non-cyclical nature of the tidal curve at the head end of an
estuary and reflect the dampening of the tidal wave and the influence of
changes in fresh water inflow. It is also interesting to note that for the
October-December, 1964 record, there is no significant long period component
reflecting the relative steady-state conditions in fresh water inflow.
(Flow conditions as measured at Trenton were between 2000-3000 cfs).
Figures I-C-8, I-C-9, and I-C-10 illustrate the results of power spec-
trum analyses on the residual variance for three stations for March-May, 1965.
The fresh water inflow during this time period decreased steadily with occa-
sional small transient increases. Flows at Trenton of about 13,000 cfs were
recorded at the beginning of the period, tapering off to about 3000 cfs at
the end of the period (Figure I-C-11). Although some difficulty was experi-
enced in recording the data at the three stations, the general patterns are
of interest. The presence of the higher harmonics can again be seen at the
Tacony Bridge and Ft. Mifflin stations but is not as evident at the Delaware
Memorial Bridge station except at the 4.16 hr. period. This may again reflect
- 43 -
-------
the influence of dampening of the tidal wave and fresh water flow on the
tidal velocity pattern. It would be valuable to attempt to measure the
current patterns during periods of high transient runoff and compare the
resultant spectrums with those obtained above under relatively steady-state
conditions. The influence of the long period effect (gradual decrease in
flow) is also most pronounced at the upstream Tacony Bridge station as shown
by the increase in power at the low frequency end.
The results of the above analyses thus indicate the relative importance
of examining the entire spectrum of tidal and fresh water phenomena and not
just a single isolated frequency such as the semi-diurnal.
- 44 -
-------
CHAPTER ONE
HYDROLOGY
REFERENCES
(1) Water Resources of the Delaware River Basin, Geological Survey
Professional Paper 381, United States Government Printing
Office, Washington, 1964.
(2) Third Water Resources Program, Delaware River Basin Commission
November, 1965.
(3) Delaware River Basin Report, Vol. VI, Appendix M, Hydrology,
December 1960.
(4) Modern Probability Theory and Its Applications, E. Parzen,
John Wiley & Sons, Inc., New York, N.Y., 1960.
(5) Urban Land Runoff as a Factor in Stream Pollution, S. R. Weibel,
R. J. Anderson, and R. L. Woodward. July 1964 Journal of Water
Pollution Control Federation.
(6) Stormwater Investigations at Northampton, A. L. H. Gameson &
R. V. Davidson, Water Pollution Research Laboratory, Stevenage,
Herts, England, June 1962.
(7) Surface Water Records of Pennsylvania and New Jersey,
U. S. Department of the Interior - Geological Survey, 1964.
(8) The Measurement of Power Spectra, Blackman, R.B. and Tukey, J.W.,
Dover Pub., Inc., New York, N. Y., 1958, 190 pp.
(9) An Introduction to the Theory of Random Signals and Noise, Davenport,
W.B., and Root, W.L., McGraw-Hill Book Co., Inc., New York, N. Y.,
1958, 393 pp.
(10) The Design of Water Resources Systems, Maass, Arthur A., et. al.,
Harvard Univ. Press, Cambridge, Mass., 1962.
(11) Queueing Theory and Simulation in Reservoir Design, Fiering, Myron B.,
Trans. ASCE, Hyd. Div., 127, 1962.
(12) A Multivariate Technique for Synthetic Hydrology, Fiering, Myron B.,
ASCE Journal, Hyd. Division, 90, September 1964.
(13) Water Quality Control Study, Tocks Island Reservoir, Delaware River Basin.
DHEW, PHS, New York, New York. December, 1966.
- 45 -
-------
CHAPTER ONE
HYDROLOGY
REFERENCES (Cont'd)
(14) Report on the Comprehensive Survey of the Water Resources of the
Delaware River Basin, U. S. Army Corps of Engineers, 1960.
(15) Dispersion Studies on the Delaware River Estuary Model and Potential
Applications Toward Stream Purification Capacity Evaluations, June
1961, Study Committee, Delaware, Pennsylvania, New Jersey, USPHS
Philadelphia, INCODEL.
(16) 1964 Surface Water Records of New Jersey, U.S.G.S
(17) Observations of Tidal Flow in the Delaware River, Miller, E.G.,
U.S.G.S., Water Supply Paper 1586-C, 1962.
(18) Calibration to Tidal Reach, Delaware River, Palmyra, New Jersey,
data in files of Trenton District Office, U.S.G.S.
(19) Tide Tables, East Coast of North and South America, Coast and
Geodetic Survey, Dept. of Commerce, 1965.
(20) Record Low Tide, December 31, 1962, Delaware River, Lendo, A.C.,
U.S.G.S. Water Supply Paper 1586-E, 1966.
(21) Estuarine Water Quality Management & Forecasting, R.V. Thomann &
M.J. Sobel., Jour. San. Engr. Div., ASCE, No. SA5, Oct. 1964,
pp. 9-36.
- 46 -
-------
WOODS HOLE CURRENT METER
Lifting Eye
Directional Vane
Vane Follower
Camera
Camera Motor
Camera Lens
Field of View
Timing Clock
Compass
Inclinometer
Pressure Case
Battery
Rotor Follower
Savonius Rotor
Figure I-C-1
-------
Anunpink
Creek
LOCATION MAP
TIDAL CURRENT
MEASUREMENT STATIONS
CURRENT METER STATIONS
Burlington-Bristol Bridge Mile 117.5
Tacony-Palmyra Bridge Mile 107.5
Fort Mifflin Mile 91.8
Delaware Memorial Bridge Mile 68.0
SCALE IN MILES
CIRCLED NUMBERS REFER TO DECS SECTIONS
Figure I-C-2
-------
CURRENT METER MOORING SYSTEM
Swivel
1,500 Ib. Cast Iron Anchor
6 Foot U.S. Coast Guard
Pencil Shaped Bouy
Direction Vane
Tidal Current Meter
Savonius Rotor
5 Foot Chain Length
River Bottom
Figure I-C-3
-------
PLOT OF ROTOR SPEED vs DIRECTION
DELAWARE MEMORIAL BRIDGE
10/12/64
2.00
1.50
O
z
*:
Z
'-00
at
O
O
at
0.50
45
90
135 180 225
DIRECTION IN DEGREES
270
315
360
Figure I-C-4
-------
POLAR COORDINATE HISTOGRAM PLOT OF DIRECTION
FOR DELAWARE MEMORIAL BRIDGE 10/12/64
315.0
0.0
45.0
270.0
90.0
225.0
180.0
135.0
1 lnch-16.4 Current Direction Readings
Figure I-C-5
-------
HISTOGRAM OF ROTOR SPEED
DELAWARE MEMORIAL BRIDGE
10/12/64
IUU
90
80
70
31 i 6ฐ
6- I 40
Z
30
20
10
0
llll
ll.lll
llll llll 1
0.2
ill
III,
0.4
1
II
1
i
0.6 0.8 1.0
1.2
I
1 ll II 1 .
ft ft
1.4 1.6 1.8 2.
ROTOR SPEED IN KNOTS
-------
(Q
c
n
.1
.09
.08
.07
.06
.05
.04
.03
.02
.01
.009
ฃ .008
3 .007
ฐ .006
o -005
^ .004
y .003
_ .002
CO
ป
o
Z
.001
.0009
.0008
.0007
.0006
.0005
.0004
.0003
.0002
.0001
1OCVI.-
o oo
CO
POWER SPECTRUM ANALYSIS
FT. MIFFLIN
10/12/64 TO 12/21/64
MEAN CURRENT VELOCITY=0.08 KNOTS
STANDARD DEVIATIONS.84 KNOTS
20
30
40
50
60 70
CYCLES/200 HOURS
80
90
100
110
120
130
-------
POWER SPECTRUM ANALYSIS
TACONY-PALMYRA BRIDGE
3/9/65 TO 5/26/65
MEAN CURRENT VELOCITY=.52 KNOTS
STANDARD DEVIATION=.72 KNOTS
.001
0009
0008
0007
.0001
60 70
CYCLES/248 HOURS
110
120
130
-------
POWER SPECTRUM ANALYSIS
TACONY-PALMYRA BRIDGE
3/9/65 TO 5/26/65
MEAN CURRENT VELOCITY=.52 KNOTS
STANDARD DEVIATION=.72 KNOTS
.001
.0009
.0008
.0007
.0001
20
30
40
50
60 70
CYCLES/248 HOURS
80
90
100
110
120
130
-------
(Q
c
I
CD
n
.1
.09
.08
.07
.06
.05
.04
.03
.02
.01
.009
" .008
O -007
i .006
2 .005
CN
U
O
z
.004
.003
.002
.001
.0009
.0008
.0007
.0006
.0005
.0004
.0003
.0002
.0001
K 00 CN
"0,0 CN
<"> -o o
POWER SPECTRUM ANALYSIS
FT. MIFFLIN
3/9/65 TO 5/26/65
MEAN CURRENT VELOCITY=.24 KNOTS
STANDARD DEVIATIONS.24 KNOTS
20
30
40
50
60 70
CYCLES/248 HOURS
80
90
100
110
120
130
-------
(Q
c
T
CD
*r*
O
POWER SPECTRUM ANALYSIS
DELAWARE MEMORIAL BRIDGE
4/5/65 TO 5/20/65
MEAN CURRENT VELOCITY.10 KNOTS
STANDARD DEVIATION=.91 KNOTS
.0002
.0001
130
CYCLES/248 HOURS
-------
17,000
16,000
WATER FLOW
USGS STATION 1-4635
DELAWARE RIVER AT TRENTON, N.J.
OCT., 1964 - SEPT., 1965
WATER YEAR
5 10 15 20 25
JANUARY
5 10 15 20 25
FEBRUARY
5 10 15 20 25
MARCH
1965
5 10 15 20 25
APRIL
5 10 15 20 25
MAY
Figure I-C-11
------- |