United States. Inland Waterways Commission.

Preliminary report of the Inland Waterways Commission. Message from the President transmitting a preliminary report online

. (page 51 of 83)
Online LibraryUnited States. Inland Waterways CommissionPreliminary report of the Inland Waterways Commission. Message from the President transmitting a preliminary report → online text (page 51 of 83)
Font size
QR-code for this ebook


abstract appears in the Report of the Chief of Engineers for 1873,
volume 1, pages 500 to 501, inclusive.

First. ''Six such sites can not be found on the headwaters of the
Ohio and an increased number of smaller reservoirs would be needed."

Mr. Roberts's criticism here is a reply to Mr. El wood Morris, an
engineer who, in support of the ideas of Mr. Ellet, calculated the
effect of such storage on Ohio River by assuming hypothetically the
existence of 6 reservoir sites, each capable of conserving the drainage
from 600 square miles. Indeed a large part of Mr. Roberts's report
is devoted to a demonstration of the fact that 6 such, sites are not
to be found in the Ohio basin. Mr. Morris merely assumed the possi-
l)ility of sites for these reservoirs for purposes t)f calculation.

Second. "These dams would destroy all raft or flatboat commerce
above them. Arrangements that might be made to provide a pas-
sage by chute would not answer when the reservoir was not full,
which might often happen."

The conditions of traffic that allowed tliis objection to be taken
seriously have now passed. At present the conditions of transpor-
tation are such that we are not giving much thought to raft and flat-
boat commerce on the highland tributaries of any stream.



WATER CONSERVATION AND FLOOD PREVENTION 453

Third. ''There would he ^reat danger of breakage from unforeseen
leaks or accidents. Reservoirs constructed on dry ground have some-
times broken and the danger is much greater in a running stream.
Should one such dam break it would probably carry all others lower
on the same stream and pour a fi-ightful wave of destruction on the
valley below. The possible loss of life and property that might
ensue is beyond all estimate. "

We laiow more about reservoirs than in those days. There are,
of course, possibilities of breakage, but this has not inliibited the
development of reservoir systems. Every jnece of construction made
by man is liable to failure. Life itself is hazardous. If the objections
here recorded were used consistently with reference to all modern
economic development, mankind would be crouching on the ground
in the open country, shaken by apprehension. Every feature of our
development makes it necessary to assume hazards, and modern
engineering has rendered the reservoir so safe that, as a agent of
violent death, it can not be placed in the same class with the ordinary
morning stroll down the streets of a great cit}^

Fourth. ''The damages on account of farms overflowed, towns,
oil wells, and manufacturing establishments submerged, railroads
compelled to rebuild their lines on higher levels so as to get above
the dams, and similar injuries would add very greatly to the cost of
the system."

It is true that there would be submergence, 3^et one is compelled
to reflect whether more farms, towns, oil wells, and manufacturing
establishments would be submerged beneath the reservoirs than are
annually submerged by the unrestrained Hoods. With reference to
railroad lines, the relocation of them does not appear to us so formid-
able as it did in those days. Indeed, it is an ordinary incident in
railway maintenance. The whole objection assumes that enormous
losses are to be entailed by such submergence, whereas, in point of
fact, reservoirs are constructed in regions not highly developed in
this way and wdiere the cases of submergence are merely incidental.
One might as well consider the cost of the large consumption of
ink as a reason for discontinuing the publication of newspapers as to
consider the above objection as fatal to economic reservoir construc-
tion. No sane man would advocate the construction of a reservoir
in places where the cost of submergence would be excessive.

Fifth. ''The utmost care would be necessary in the management
of these reservoirs to prevent injury by sudden floods, and terrible
disasters might ensue from carelessness or bad judgment."

This indictment is accepted. We are all subject to the good judg-
ment and management of our friends and associates in the con-
trolling conditions in our everyday life, and it is to be presumed
that each man will do his duty. Whether we travel or remain at
home we are constantly subject to the operation of utilities that
can not be made fool proof, and there is no reason for assuming that
the management of reservoirs is extrahazardous. On the contrary,
it is a comparatively simple matter.

Sixth. "The proper use of the reservoirs would be very difficult
in order to keep up the exact depth needed for navigation, as any
mistakes in regulating the supply would seriously embarrass naviga-
tion. Special telegraph lines would have to be constructed to all



454 REPORT OF THE INLAND WATERWAYS COMMISSION

reservoirs, and widely extended observations on the rainfall and the
conditions of all tributaries would be needed in order not to waste
nor to unduly economize the supply."

We are ready to meet all or Mr. Roberts's requirements in this
matter. We have extended observations on the rainfall and the
conditions of all tributaries, and we could do even better than he
suggests in the way of communication, for we could, supply tele-
phones. It is readily recognized that this objection is antiquated
and not worth considering in the light of modern improvements.

Seventh. ''The cost of the reservoir system would be very great
and could only be estimated by careful examinations at each site
selected. Estimates based on a specially favorable site could not
be depended on."

All this is accepted as one of the trival incidents of reservoir con-
struction. Of the cost we shall speak later on.

Eighth. ''The whole thing is an enormously costly experiment, and
would require the continuation of the present system of improvement
l^y riprap dams and dikes in order to utilize the water delivered.

It is true that the system would be costly, yet, as will be shown later
on, the relative cost compared with the benefits derived would be
small. It is a well-known precept of modern business life that values
are not determined by cost, and nothing may be considered costly that
returns a reasonable income on the investment. Some things would
be costly if purchased for a farthing. With reference to the scheme
being an experiment, that which may truly have been considered an
experiment in the day of Mr. Roberts may now be a well-attested
reality. The scheme is not an experiment, as will be shown later on.
One of the purposes of this paper is to show that Mr. Roberts was
wrong in his assumption that complete channel improvement would
still be necessary if reservoirs were constructed. Indeed, the reservoir
is urged as a substitute for such works under certain conditions.

Wnile the objections above cited may have been important at the
time they were enunciated, it is clear that they are of little conse-
quence at the present time. The whole matter is brought down to a
question of fact, namely, is there available on the highland tribu-
taries of the Ohio sufficient water supply and storage capacity to pre-
vent floods and to maintain deep-water navigation during seasons of
low water. Mr. Ellet did not have the necessary information at his
disposal, and therefore could not prove his contentions. Mr. Roberts
confessed that he did not have it, but endeavored and, in fact, suc-
ceeded in discrediting the idea without it. We are now in possession
of it for certain areas, as a result of the investigations of the United
States Geological Survey and the United States Weather Bureau, and
can readily establish certain crucial points that were left in doubt in
the earlier discussion.

It will be appreciated on examination of this paper that the region
considered does not cover the entire basin. Therefore this presentation
can not do entire justice to the situation. Whatever results may appear
to be claimed as arising from the construction of these reservoirs with
reference to the efi'ects of floods and the maintenance of low-water
navigation on the Ohio, they do not represent the total possibilities of
the region, for, were surveys available on all the basin, it is manifest
that far greater storage facilities would be shown to be available.



WATER CONSERVATION AND FLOOD PREVENTION 455

Therefore the maximum effect of conservation would be much greater
than shown in the following pages.

It will be helpful now to consitler an objection that is frequently
made to the use of storage reservoirs for flood prevention purposes,
viz, that there is no way of predicting when floods may come, and it
would be certain that a flood would descend on the reservoirs when
they were filled to overflowing with the run-off from a previous flood.
Now, the records of floods during the past twenty or thirty years on
the Ohio indicate that, while they may not be j^redicted within a few
days, they do occur during certain seasons of the year, and therefore
the problem is narrowed down to that extent. But, supposing this
were not so, the effect of the reservoirs would not be diminished. It
will be noted in subsequent pages that the extent of drainage area
that can be conserved by various reservoirs has been determined.
The reservoirs will hold the entire year's run-off from a stated area, or,
in other words, if the gates of the reservoirs were allowed to remain
closed for an entire year the reservoirs would retain all the water
flowing from that territory for the entire period. Supposing now that
two floods should descend into the Ohio River, as they did in January
and March, 1907. The secontl flood could not descend on full reser-
voirs because the capacity of the reservoirs is sufficient to hold them
both. We have, for example, on the Monongahela storage facilities
of capacity sufficient to conserve the run-off of 38 per cent of the drain-
age area.

Therefore, according to the adjusted capacities stated in the fol-
lowing pages, this per cent of the Monongahela drainage area could
be entirely cut off from the Ohio Valley for the period of one year.
Of course, this estimate is based on the records of mean flow, as
shown by observations extending over a series of years. There is
considerable variation from one year to another, so that if the reser-
voirs actually remain closed there are years in which the accumu-
lation of water would more than fill them and still other years in
which the accumulation of water would not suffice to fill them. But
the point is that this great capacity furnishes a wide margin on which
to work. The two floods of the spring of 1907, for example, would
not fill these reservoirs, but, assuming tiiat they remain closed for the
entire year, it is possible that the entire year's run-off would more
than fill them. But, with this wide margin of time, covering, indeed,
a low-water season, when the water would be needed in the Ohio,
there is ample time to draw off the water and prepare the reservoirs
for subsequent floods. Therefore, the criticism that floods might
descend upon reservoirs already filled is based on the hypothesis that
the reservoirs are small and their capacity is not commensurate with
the size of the basins, whereas, in point of fact, they are sufficiently
larcre for flood prevention. The whole matter therefore comes down
to mtelligent manipulation, with margins of safety so wide that only
the most flagrant stupidity could result in any misfortunate circum-
stance.

A further question now to be discussed is. How are we going to
manipulate the reservoirs above which there is a large drainage
area when their capacity is only sufficient to hold a portion of the
flood descending from that area? A glance at the tables in subse-
quent pages will show that there are many such. This is a mere



456 REPORT OF THE INLAND WATERWAYS COMMISSION

matter of intelligent manipulation. We will assmne, for example,
that there is, above a certain reservoir, a drainage area of 100 square
miles, while the reservoir itself has a capacity sufficient to conserve
the run-off from only 50 square miles. This does not make it neces-
sary that the run-off from the 100 square miles shall come down
and overwhelm the 50-square mile reservoir. The fact should be kept
in mind that this reservoir is to conserve the drainage from only 50
square miles and therefore as fast as the flow comes down into the
reservoir one-half of it should be released through the gates. The
release of one-half of the water may readily be accomplished by ad-
justing the size of the openmgs in the reservoir gates. The same
effect could be readily produced if the drainage area above the res-
ervoir were 1,000 square miles in extent and the reservoir capable
of conserving only 10 square miles. The flow into the reservoir
from the upper channel can readily be rated, and it can be known at
any time from observation of the gauge how much water is coming
into the reservoir. Knowing the proportion of the drainage area
which the reservoir will conserve, it is merely necessary to raise or
lower the gates, so that the portion which it is not intended to con-
serve may be discharged from the reservoir as fast as it comes in.
The computation of the proper amount is a simple engineering cal-
culation and would not require expert manipulation on the ground.
Simple directions based on these calculations could be given to any
intelligent man, so that he could manipulate the gates of the reser-
voir according to the amount of water coming in, in precisely the
same way as the unprofessional attendant of reservoirs on city water-
works is instructed by a few simple rules to manipulate the gates
according to the consumption. These considerations, it is believed,
will dispose of the oft-repeated assertion that a second flood need
descend after a first and upon full reservoirs.

Another preliminary statement should be made before passing to
the consideration of the detailed facts concerning the reservoir sys-
tem. It vnW be admitted that the reason why a river will overflow
its banks is because so great an amount of water is turned into the
channel mthin a short space of time. If the total annual flo\v of a
river were uniform there would be no floods. It is the great influx
of water derived from heavy rains and quickly melting snows that
overtaxes the channel. The question now arises, What are the fac-
tors that control the speed vnth which run-off water is discharged
into river channels? The answer is plain, namely. The slope of the
tributary country and the condition of the land surfaces. It is clear
that the water of precipitation will run off the side of a mountain more
quickly than off the surface of a plain. Therefore, it must also be
clear that the most destructive part of a flood originates in the steeper
part of the drainage area, and consequently, if the speed of this part
of the run-off could be inhibited the severity of floods woukl be
greatly reduced. Bearing this in mind, it will now be profitable to
consider what portions of the various (Irainage areas of Ohio River
tributaries it is proposed to conserve in the present instance. A
glance at the locations will show that almost mvariably it is these
quick-spifling upland areas above referred to, but it should be noted
that a reservoir system does not merely inhibit the run-off from these
steep areas, but actually holds and conserves it. It is the lower,
flatter portions of drainage areas that remain unobstructed. The



WATER CONSERVATION AND FLOOD PREVENTION 457

situation is ideal, because these lower portions are the least trouble-
some. The purport of the above suggestions is as follows: It has been
stated in a previous paragraph that the reservoir facilities on the
Monongahela basin -will, if developed, absolutely conserve the run-off
from 38 per cent of the drainage area. Other tilings being equal, it
might be stated that the reduction in flood height would be, broadly
speaking, a like proportion. Actually, however, the reduction will be
far greater than this, because, as above demonstrated, it is the quick-
spilling portion that will have been conserved and the more moderate
portion that will remain unobstructed. Therefore, in connection
with all statements hereinafter made concerning the reduction of flood
height by reservoirs it should be borne in mind that the proportion
stated is by no means as great as that which will occur in actual
practice.

RESERVOIR FACILITIES IN THE OHIO BASIN

Ohio River is formed by the junction of the Allegheny and Monon-
gahela rivers at Pittsburg, Pa. On the Allegheny basm the lack of
suitable surveys prevents a full estimate of reservoir possibilities, and
those here presented do not cover the entire basin and should not be
so interpreted. The Monongahela basin, on the other hand, has been
thoroughly surv^eyed and the amounts stated in the following para-
graphs probably represent maximum conservation. In other words,
on these two, as well as on the remainder of the tributaries, all the
available information is given, and deductions are here made on that
basis.

We ^^^ll now discuss briefly the characteristics of the various liigh-
land tributaries of the Ohio, their water resources, and their availa-
bility for storage-reservoir installation.

ALLEGHENY BASIN

The Allegheny basin is 11,400 square miles in extent. The greater
part of the area is steep and quick spilling and the deforestation
has been quite extensive. The river is subject to wide range
of stage and floods are frequent and severe. Except on the lower
25 miles of the river, the principal traffic consists in the doAvn-
stream transportation of timber and lumber products. The project
for improvement of the river contemplates the removal of bowl-
ders and snag obstructions, and the construction of low dams and
dikes to close secondary channels and concentrate the low-water flow
on shoals. The amount expended under this project to the close of
the year ending June 30, 1906, was $260,750, fully one-half of which
was applied to maintenance. The work done has not increased the
depth of water, generally speaking, except where dikes or dams have
been built, but tlie removal of the obstnictions has made it possible to
utilize the entire flow of the stream, so that now navigators may
operate safely on from 1 to 2 feet lower stages than formerly.

The annual tonnage of the river above slack-water improvements
wliich are now in course of construction is about 400,000, extending
over 230 miles from Tarentimi, Pa., to Olean, N. Y.

As above stated, a complete survey of the river basin is not avail-
able; therefore, no complete estimate can be made of the total possi-
bilities of preventing floods and maintaining navigable stages during

31673— S. Doc. 325, 60-1 30



458



EEPORT OF THE INLAND WATERWAYS COMMISSION



low water. The storage capacity that can now be computed is suf-
ficient to entirely conserve the run-off from an area of 1,774 square
miles, or 15.6 per cent of the total drainage area. The locations of
these storage reservoirs, together with facts concerning them, are
set forth in the following table:

Selected reservoir sites in the Allegheny River Basin



stream and location.



Conewango River, east of James-
town, N. Y.

Chautauqua Lake

French Creek, Marvin, N. Y

Dodge Creek, PortviUe, N. Y....

Haskell Creek, Olean, N. Y

Mahoning Creek, Mahoning, Pa .

Cowanshannock Creek, Valley
County, Pa.

Buffalo Creek, S.W.N. Buffalo
County, Pa.

Clarion River, Clarion, Pa

Loyal llanna, Saltsburg, Pa

Crooked Creek, Burrell Co., Pa...

Twollek Creek, Homer, Pa

Yellow Creek, Homer, Pa







Capacity of






Catchment
area above
reservoir.


Mean
annual flow


reservoir
in square


Capacity of
reservoir


Height


from catch-


miles of


(1.000,000


of dam.


ment area.


catchment


cubic feet) .








area.






Sq. miles.


Second-feet.






Feet.


280


560


275


17,590


38


190
86


148


27
86


1,740
5,677




75


46


92


46


2,990


100


30


60


30


1,950


70


370


636


370


22, 100


251


48


82


48


2,720


118


80


138


80


5,120


149


990


1,700


436


23,850


210


260


472


58


3,180


98


200


340


200


12, 415


110


68


116


52


2,860


122


66


112


66


3,900


150



Area of
flow line



A eves.
24,000

13,400
4,700
1,860
1,320
5,300
1,390

2,260

6,920
2,200
7,800
1,300
1,620



a Three feet on surface.

The highest run-oiT from the Allegheny basin, of which there is
authentic record, occurred during the flood that began March 22,
1905, when the highest discharge was 272,000 second-feet. The
total flowing past Pittsburg from both the Allegheny and the Monon-
gahela was 317,000 second-feet, showing that, in this region at least,
the Allegheny Avas the principal cause of the flood in the Ohio. This
is not usually the case, however, the Monongahela being more often
the principal offender.

MONONGAHELA BASIN

The drainage area of this river is 7,400 square miles in extent.
Excellent topographic surveys covering the entire basin make it
possible to determine with a fair degree of completion the storage
facilities aft'orded by the basin. The character of the drainage area
is somewhat similar to that of the Allegheny. The neatest con-
tributor to the flood of ^larch, 1907, was the Monongahela, and the
relation of this basin to the Ohio will be discussed in subsequent
pages.

The Monongahela is canalized from its mouth to Fairmont, W. Va.,
the slack-water svstem comprising 15 dams and 19 locks, the total
cost of which to^June 30, 1905, was $6,023,605. For the mainte-
nance of these locks up to the same date the United States has paid
$1,694,380. A slack-water system is now proposed for the Yough-
ioghenv, consisting of 3 locks and dams at an estimated cost of
$600,000.



WATER CONSERVATION AND FLOOD PREVENTION



459



Proper manipulation of the reservoirs that are shown to be avail-
able in the Monongahela basin would afford an increased stage above
that of low water at three sections, namely, the Morgantown bridge,
Brownsville, and McKeesport, as shown in the following tables:



SECTION BELOW MORGANTOWN BRIDGE



Increase of stage

Flow required

Reservoirs:

Full

Three-quarters full
One-half full





Duration of stage (days).


1.0


2.0


3.0


4.0


5.0


545


1,290


2,300


3,600


5,030


1,650


700


392


250


179


1,240


525


293


188


134


825


350


196


125


90



6.0
6,760

133

100
66



SECTION AT BROWNSVILLE, PA.



Increase of stage

Flow required

Reservoirs:

Full

Three-quarters full
One-half full



1.0


2.0


3.0


4.0


5.0


730


1,730


3,070


4,670


6,570


1,740


730


410


270


193


1,300


550


310


203


145


870


365


205


135


96



6.0
8,730

139
104
70



SECTION BELOW McKEESPORT, PA.



Increase of stage

Flow required

Reservoirs:

Full

Three-quarters full
One-half full



1.0


2.0


3.0


4.0


5.0


1,080


2,470


3,940


5,840


8,000


1,500


650


410


280


200


1,120


490


310


210


150


750


325


205


140


100



6.0
10,340



150
117



The above tables show that intelligent manipulation of avail-
able storage on the Monongahela would provide ample depth for
navigation of the river during low-water seasons. It is possible that
the money expended for canaHzation might have been spent to a
better purpose in the hills. Now, that the dams are erected, the
proper course is to erect the reservoirs and produce profitable power
on these dams, so that the system will be self-supporting.

The available storage in the Monongahela basin will absolutely
conserve 2,800 square miles, or about 38 per cent of the entire basin,
the result of which would be to (Uminish the height of the floods along
the lower reaches of the Monongahela by at least 50 per cent. Facts
concerning the available storage facilities in the basin are set forth
in the following table:

Selected reservoir sites in the Monongahela River basin



Stream and location.


Catchment

area
above res-
ervoir.


Mean an-
nual flow
from catch-
ment area.


Capacity
of reser-
voir in
square
miles of
catchment
area.


Capacity
of reser-
voir
(1,000,000
cubic feet).


Height
of dam.


Area of
flow line


iJunkard Creek, Cass County,
W. Va....


Sq. wile.';.
100

40
40
80
40


Second-feet.
270

70
70
140
60


150

40
40
80
40


8,300

2,400
2,800
6,200
2,600


Feel.
100

59
101
155
115


Acres.



Online LibraryUnited States. Inland Waterways CommissionPreliminary report of the Inland Waterways Commission. Message from the President transmitting a preliminary report → online text (page 51 of 83)