American Society of Civil Engineers.

Transactions of the American Society of Civil Engineers (Volume 81) online

. (page 68 of 167)
Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 68 of 167)
Font size
QR-code for this ebook


cribs in a practically unchanged and polluted condition.

In 1905 and 1906, typhoid fever in Wilkinsburg became epidemic,
and just prior to the bringing of the suit it had the highest typhoid
fever case and death rate of any town in the United States. Very
heavy rates also prevailed in the 37th Ward of Pittsburgh, and in
Swissvale Borough, also supplied with water by the same company.
As a result of this alarming condition, protests were lodged, and,
finally, recourse was had to the Courts to compel the Water Company
to safeguard adequately the quality of the water. The case, recorded
as E. Z. PefPer et al. vs. Pennsylvania Water Company, was No. 396,
tried in the August term, 1906, of the Court of Common Pleas, l^o.
3, of Allegheny County, Pennsylvania, Miller, J., specially presiding.
The conclusions of law, in this case, as drawn by Judge Miller, were:

First. — The fact having been found that the water complained
of, as furnished by the defendant, is not reasonably pure and
wholesome, it follows that the plaintiff's bill must be sustained.

Second. — An interlocutory decree will be entered, directing it to
secure and provide a sufficient supply of pure and wholesome
water; further directing that it shall, within three months
from the date of decree, file a statement of the steps it has
taken, and purposes to take, in compliance with the require-
ments to furnish reasonably pure and wholesome water, upon
the submission of which the plaintiff may file a reply or
answer, as he may deem advisable. It will be further
decreed that this case be retained for such further proceedings



824: DISCUSSION : water supply of parkeesburg, w. va.

Mr. as may be necessary to insure its performance, and to enable

"^"^ ^^' the Court to exercise the jurisdiction conferred by this Act

of Assembly under which this action, is brought, with liberty

on the part of either of the parties to apply to Court for

further orders and decrees as may be necessary and just.

Third. — The defendant to pay the costs.

The meagerness of the data regarding the details of design, con-
struction, and operation of the Parkersburg plant, in Mr. Hall's paper,
makes extended discussion of it impossible. Those elements of design
which would give control over the process of filtration are scarcely
mentioned. Enough is stated, however, as to the diameters of the
main pipes leading from the strainer system, and as to the sizes of
the laterals and their perforations to indicate:

1. — That hydraulic conditions must exist which would be avoided
in a properly designed filter plant.

2. — That there can be almost no control over the effectiveness of
back-flushing, or washing of the filter areas.

3. — That the loss of head must vary between quite wide limits
at various parts of these areas during filtration, and after
partial clogging shall have taken place, and that, therefore,
parts of the areas must be passing the water at much higher
rates than others, and, consequently, must be doing less
effective work than planned.

4. — It is also evident that the back-flushing is likely to blow up
and overwash parts of the areas, and leave other parts badly
clogged.

5. — That floods in the river may scour, and probably have fre-
quently scoured, holes in the filters — possibly almost down to
the strainers.

6. — That the whole filter area is inaccessible, and utterly beyond
inspection and repair during fioods, the times when the water
is worst and the need of protection greatest.

The Parkersburg plant may be expected to give fair water when
the river is in its best condition, and, at such times, it is almost
good enough to use without filtration. When the river water is turbid
and dangerously polluted, the water delivered by the plant may or
may not be safe. Whether or not it is safe will be beyond the knowl-
edge of the men operating the works ; and, if it is unsafe, it is entirely
beyond their power to make it so by any agency or device inherent
to or part of the plant itself or of its design.

Judging by the discoveries of Mr. Leland, when making the in-
spections in September, 1916, referred to by Mr. Hall, the Parkers-
burg plant is experiencing difficulties similar to those which led



discussion: water supply of parkersburg, w. va. 825

ultimately to the virtual condemnation of the Wilkinsburg works Mr.
and to the erection of a properly built mechanical filter plant in its
stead.

The cost of construction and operation of the works, as built at
Parkersburg, are given by Mr, Hall as $284 171.22 and $11 892.46,
respectively. The annual cost was based on the pumpage of an average
of 3 000 000 gal. daily. In the Fuertes-Fuller report, the costs were
given as $165 945 and $24 000, respectively, the latter figure being the
estimated annual cost of operation for a daily yield of 4 000 000 gal.
Reduced to the basis of 3 000 000 gal. daily, for comparison with the
figures given by Mr. Hall for the Parkersburg plant, and adding to
the cost of construction certain items included by Mr, Hall, and with
interest at 7% in each case, the annual operating and maintenance
costs of the two plants would be practically equal. The Smith system,
therefore, has required at Parkersburg a greater investment, saves
nothing in annual expense, and gives to the city far less security
against water-carried diseases than would have resulted from the
establishment of a properly built and operated mechanical filter plant,
such as recommended and described in the Fuertes-Fuller report.

Mr. Hall gives, among the reasons advanced by Mr. Gray for
the adoption of the Smith system for Parkersburg, that,

"If properly built and operated, this system will furnish water of a
better quality than a mechanical filter plant, as regards steam-raising
purposes, on account of the slight increase in permanent hardness
which is caused by the use of a coagulant."

Although it is true that the use of aluminum sulphate as a coagu-
lant will result in the conversion of a portion of the alkalinity into the
sulphate or incrustant form of hardness, an examination of the
analyses in Appendix I shows that throughout January, 1914, the
total hardness of the water from the Smith system was about 87
parts per million, as against about 74 parts per million for the river
water, this increase being due to the increased alkalinity of the sub-
river ground-water. The increased quantity of soft scale due to the
extra 13 parts per million of hardness of this ground-water is probably
fully as disadvantageous from a steam-raising point of view as would
be the conversion of about 5 or 6 parts per million of the low
alkalinity of the river water into a proportional quantity of incrustants
by the application of aluminum sulphate used in connection with
mechanical filters. For summer conditions, when both the river water
and the water secured by the Smith system would be much harder than
reported in the January analyses, the advantages would probably
lie strongly with the mechanically filtered water.

And, further, with the mechanical filters, the operatives would
have access to and control over the filters and their efficiency at all



826 DiscussiON^ : watek supply of pakkeesburg, w. va.

Mr. times. With the Smith system, the whole delicate process of filtration
uer es. ^^ beyond reach, often buried under floods and subject to disturbance
by wash and scour, or to irregular deposition of mud on the filter
area, with consequent imperfect pTirification, due to varying and
excessive rates of filtration on parts of the area smaller than the total
required.

Taking everything into account, the speaker is not convinced, by
the data brought forward, that the proper decision was made at
Parkersburg when the Smith system was adopted in preference to
a properly designed and constructed mechanical filter plant.

It is much to be regretted that so little information has been
furnished as to the operation of and results obtained by the Parkers-
burg water-works plant. Fancied security has led to neglect of
vigilance, as represented by systematic, continuous, analytical studies
of the river and secured waters, and more frequent examinations of
this nature, as recommended by Mr. Hall, should certainly be made by
the authorities.

The ground-water studies, made at Parkersburg, and referred to
' by Mr. Hall, show a number of interesting facts. Figs. 3 to 12 were
selected from among 32 such diagrams in the Fuertes-FuUer report,
and seem to have been well chosen to exhibit the interesting
phenomena of filling and emptying the subterranean storage spaces
in the voids in the gravel underlying the flat lands along the river
north of Parkersburg.

It will be noticed that the water finds its way into the ground,
from the river, with comparative difiiculty, as judged by the steep
slope of the ground-water in the neighborhood of the river banks,
but that it makes its escape back into the river, when the latter
falls, with comparative freedom. This is as would be expected, the
surface of the river bottom and banks having become clogged with
mud (strained out of the river water as it filtered out through the
banks and bottom) on a rising river, but being cleansed, or washed
free of mud to a greater or lesser extent, on a falling river, by the
passage of the ground-water back again into the river.

The contours on Figs. 16, 17, and 18 show the gradual change in
direction of the flow of the ground-water away from and toward
the river above Parkersburg. Fig. 12, which was reproduced from
among the diagrams in the Fuertes-Fuller report, occupies a position
between Figs. 17 and 18. An examination of these diagrams shows the
following interesting conditions.

On January 13th (Fig. 16) the river, after having reached its
lowest point from a previous rise, was just beginning another rise
which, by the 15th, had reached 9 ft. Observing the ground-water
contours on this diagram, it will be seen that the general trend of
the ground-water was down stream and toward the Ohio Kiver,



Discussiox: WATER SUPPLY OF pIrkersburg, w. va. 827




Mr.

FuerteB.



828



DISCUSSIOX : WATER SUPPLY OF PARKERSBUEG, W. VA.



Mr.
Fuertes.




Discrssiox: water supply of parkersbueg, w. va.



829




Mr.

Fuertes.



830 discussion: watek supply of parkersburg, w. va.
Mr. the 571-ft. contour being the farthest down stream and the 574-ft.



Fuertes.



one the farthest up stream, all these contours lying in approximately
parallel lines, and indicating a motion diagonally down stream toward
the river.

On January 17th, the river began another rise, and, within the
next few days, reached an elevation of 597.5 ft. On January 19th
(Fig. 17) it stood at Elevation 592. It will be observed that the
contours of the surface of the ground-water had changed entirely,
swinging around parallel to the shore with the highest contours
(Elevation 579) along the river bank, and the lowest still at Elevation
572, on a line parallel with and about 2 000 ft. distant from the river
bank. By the 24th, the river surface had fallen to 692.5 and the
ground-water near the river bank had risen to 584, or 14 ft. higher than
on January 13th, whereas, along a line practically parallel with, and
some 2 500 ft. east of, the river bank, the ground-water level had risen
only about 1 ft.

On February 5th (Fig. 18) the river still continued to fall, the
ground-water which had been piled up to Elevation 584 on January
24th had fallen at the river bank to an elevation of only 573, gradually
rising to an elevation of 575, about 2 000 ft. back from the river and
dropping to 574 a few hundred feet farther east. In this case the
water was slowly flowing in both directions from a line parallel to
and about 2 000 ft. back from the river bank.

It is very fortunate that this flood occurred during the time that
the tests were being carried on at Parkersburg, as the occasions on
which such phenomena can be observed are relatively rare.

Mr. ' Fuller has made the suggestion that in a short time the
results of studies for the more complete identification of certain forms
of bacteria will be made public. The present laboratory methods
for the quick determination of the probable presence of B. coli in
water, are, as Mr. Fuller has said, not only far from satisfactory, but
actually misleading under some conditions. An accurate and extensive
knowledge of the history of the water is always essential in deter-
mining the likelihood of the presence of these organisms, if shown
to be probably present by laboratory tests. The speaker calls to mind
an interesting experience of this sort in connection with the examina-
tion of the water supply of the Insane Asylum at Willard, N. Y., in
December, 1901, and January, 1902.

Diphtheria had been epidemic at the Willard Asylum since 1897,
and all efforts to eradicate it from the institution had proven unavail-
ing. The cases were confined mostly to the nurses, physicians, and
attendants, very few cases being noted among the inmates. The
superintendent of the asylum and his wife were both in quarantine with
diphtlieria while this investigation was going on. In connection with
the various studies made to determine the cause of the prevalence of



discussion: water supply of parkersbukg, w. va. 831

the disease, a thorough examination was made of the water supply and Mr.
sewerage of the institution. In order to eliminate possible sources of ^"^''*®^-
error, a great many samples of water, from widely separated locations,
were collected and examined.

The water supply for the institution was taken from Seneca Lake
at a point some 400 or 500 ft. from shore, where the water was about
30 ft. deep, and at a point about 1 800 ft. from where the main sewer
of the institution discharged into the lake.

An apparatus was devised for securing samples of lake water at
various depths, and these, after collection, were put in sterilized, glass
stoppered bottles, and examined promptly.

Samples were taken at the surface of the lake and at various
depths at numerous points, some of these being in the immediate
vicinity of the sewer, some at the water-works intake, others on the
opposite side of the lake, about 1^ miles from the intake, others at
the ends of the lake, more than 20 miles distant. The samples were
collected at various depths, from the surface to 300 ft. below the
surface.

These analyses disclosed the interesting fact that the chlorine
content of the lake water varied from 36 to 37 parts per million, which
was about one-third of the quantity that might be expected in ordinary
town sewage. Organisms giving the reactions expected of B. coli
were found in the samples collected in the neighborhood of the sewer,
also, in samples collected 50 ft. below the surface on the other side
of the lake and about 600 ft. from shore, also, in some collected 50 ft.
below the surface, some 600 ft. off shore about 2 miles away, and in
a number of other places. Previous analyses made in 1899 had dis-
closed the suspected presence of B. coli communis in samples 1 300
ft. out from shore, on the line of the hospital's sewer, at a depth
of 20 ft.; J mile from shore at a depth of 150 ft.; J mile from shore
on a line with the water-works intake at a depth of 400 ft.; 500 ft.
deep off shore about 5 miles south of the Asylum, and in the center
of the lake at a depth of 400 ft.

In the analyses made in 1901-02, it was pretty well established
that the organisms giving the reactions for the differentiation of the
B. coli communis were probably washed into the lake from the culti-
vated fields along the sides, and bore a close relation to the rust on
grain.

The high chlorine in the lake water was caused by the presence
of underground salt beds in that vicinity; the normal chlorine for
that region would not be more than 7 parts per million.

Without knowing all the conditions, the analyses of Seneca Lake
water might easily have been mistaken for a highly sewage-polluted



833 DISCUSSION : water supply of parkersburg, w. va.

Mr. water; a conclusion which the most ordinary common sense would
Puertes. ^gf^gg ^q accept.

More positive methods for identification of sewage bacteria in waters
have long been needed, and water- works engineers will welcome the
greater certainty promised by Mr. Fuller's predictions.

MesBrs. MoRRis Knowles,* M. Am. Soc. C. E., and J. D. Stevenson * Assoc,

and M. Am. Soc. C. E. (by letter). — This paper is of especial interest
■ to the writers, because of the part they took, in 1908-09, as Consulting
and Resident Engineers, respectively, of the Water- Works Commission,
in the early studies of the water supply at Parkersburg and the
proposed new systems. They believe that a statement of some of the
considerations which led to the recommendation to develop a well
supply, instead of the system described by the author, will be of
interest. The author is to be commended for collating and bringing
before the Profession the various studies and reports which have
been made in reference to this interesting situation.

No discussion of this subject would be complete without mention
of the lively interest taken in these early investigations by the late
United States Senator J. N. Camden, and continued, at his death,
by his estate. The generous attitude of this public-spirited citizen
can be best stated by quoting from an open letter from him to the
city ofiicials, as follows :

"I, therefore, respectfully propose, in order to test the question
of natural filtration, that I will, at my own cost and expense, sink
and test wells, both on the river front and on the foot of Neal's
Island, for which I already have the permission of the owner, to
ascertain the quality and quantity of water that can be obtained by
natural filtration, and to commence as soon as the engineers selected
by the city authorities and water-works commission are ready to direct
and superintend the tests to be made. This will cost the city nothing
and will no doubt be of value in arriving at conclusions. I will also
add that, should the city desire to locate its plant or pumping station
or wells upon any ground owned by me, I will donate to the City all
the ground it may need for water-works purposes."

In pursuance of the engagement of the writers as engineers, a
report was made on August 24th, 1908, covering an investigation of
ten possible developments and including a recommendation that either
a mechanical filter plant or a system of wells be adopted as a source
of supply. A definite choice between these two projects was not
recommended at that time for two reasons: First, because the original
sum of money appropriated to the Commission did not permit of
drilling and testing wells; and, second, because after the money
donated by Senator Camden for this purpose was made available, it
was concluded "that the time was not propitious for either drilling

* Pittsburgh, Pa.



discussion: watek supply of paekersburg, w. va. 833

or testing wells, as both the river and the ground-water level were Messrs.
too high to justify any conclusion from a test as to safe yield." "and*^^

The investigation of the wells was only postponed, however, being Stevenson.
carried out in the succeeding warm, dry season, at a time when both
surface and ground-water levels were low. The conditions at the time
of the tests were most favorable to a safe conclusion, as the river
was lower than it had been for many seasons. The results of these
tests warranted the recommendation of a water supply to be drawn
from wells on the Camden Farm, which was reported on January
25th, 1909.

The topography, geology, and history of the Ohio Valley give
evidence that, at one time, either the Little Kanawha River or Worth-
ington Creek, or both, flowed through a gap in the hills and into the
Ohio River at a point near the present settlement of Beechwood.
In later years, either the Little Kanawha River straightened its course
and shifted to its present location, while Worthington Creek occupied
in reverse flow what had been the Little Kanawha River bed; or else,
Worthington Creek cut a new course into the Little Kanawha River
and thence into the Ohio. In any case, the gap in the hills became
filled with sand and gravel, forming what are now known locally as
the "Sand Plains", separated from Parkersburg by the rocky promon-
tory known as Boremans Hill.

Evidence is conclusive that in times past the Ohio skirted the
West Virginia shore and, at one time, occupied the area now forming
the east shore terraces, on which are the Camden Farm and the settle-
ments of Beechwood. Vienna, and Brisco. It is also evident by the
slope of the rock that, in its former location, the Ohio River was at
a lower elevation than at present, the intervening space between the
present and the former beds having been filled with a water-bearing
sand and gravel formation. The trough made by the intersection
of the rock beds sloping from the two shores is under the east shore
terraces, and, at the Camden Farm and for many miles up stream,
probably has the same general course as Pond Run.

The flow of underground water normally follows the general in-
clination of the rock, and is outward from the surrounding hills.
When the flows from two hills meet, the resultant flow is upward, and
if it is abundant and the head is sufficient, an Artesian flow will
result. Such is the origin of much of the flow in Pond Run. An
inspection of the bottom and sides of the Run and feeders will disclose
many springs, and these have been found from above Neal's Island to
a point below the Camden Farm.

It would not be expected that there would be a large flow parallel
to the river in the sand and gravels directly over the rock, but rather
that something like an underground reservoir would be formed, over-



834 discussion: water supply of parkersburg, w. va.

Messrs. flowing into the river and to the surface of the terrace. However, in
ai^^ this particular case, the gravel beds are topped by impervious clay and
Stevenson. Iq^j^. As a result, the head due to the higher level of the ground-
water on the "Sand Plains" causes flow in the direction of the inclin-
ing floor of the gravel beds, which is not only toward the river, but
also down the valley. This was the basis for the proposed development
by two lines of wells, one parallel to, and another at right angles to,
the river, thus intercepting whatever water flowed toward or parallel
with the river.

It will be noted that the top of the gravel beds at the river's edge
is very little higher than the low-water level in the river. The
observations made by the writers indicate that the underground
reservoir is not fed to any great extent by infiltration from the river,
but probably receives most of its water from rainfall on the terraces,
on the "Sand Plains", and from the 26 sq. miles of drainage area
tributary to Worthington Creek, and probably some from entrance at
the rock riffles in the Ohio some distance up stream.

In Figs. 3 to 10, inclusive, it will be noted that the surface of the
ground-water as shown is always above the line of demarcation between
the water-bearing sand and gravel and the impervious loam and clay;
and that in no instance did the pumping lower the surface of the
ground-water into the sand and gravel. It may be concluded, there-
fore, that, even at high stages of the river, no large quantity of river
water would be forced into the sand and gravel; for these strata
would always be filled, and, as the impervious clay and loam above
would not absorb water in quantities rapidly, the rise in the river
would only increase the head on the water in the sand and gravel,
raising the level of the water in the observation wells correspondingly.

Eeferring to Pig. 11, the following explanation appears in the
writers' report:

"When pumping was checked in amount or stopped entirely, the
water quickly rose in the well being pumped, indicating a lack of
strainer openings for the free access of the water."

In the absence of long-continued tests, the quantity of water which
may be procured from any formation is largely a matter of speculation.
The planning of a water supply from wells in any undeveloped ter-
ritory, therefore, should provide, where possible, for permitting easy
future adaptation to other sources of supply, in case this should
become necessary. The conditions on the Camden Farm were par-
ticularly advantageous for such a water-works development; and,
although, as a result of the tests and the comparison of conditions
with those in somewhat similar deposits elsewhere, the conclusion



Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 68 of 167)