American Society of Civil Engineers.

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

. (page 110 of 167)
Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 110 of 167)
Font size
QR-code for this ebook


In the upper reaches and head-waters, erosion usually predominates,
and flood conditions differ materially from those of either the lower
or middle reaches.

In many of the smaller drainage basins, such as those oi several
of the Ohio tributaries, all three conditions are represented within
comparatively short distances, and no one set of laws of erosion and
deposition will apply. The data derived from the Mississippi studies
may be applicable to the lower reaches of these streams, but there
is a decided deficiency of information as to the behavior and work
of the streams in their upper sections.

The result of this deficiency of knowledge of the behavior of the
swifter streams is illustrated by the destruction of many railroads at
various times in the past in the canyons of the West, especially along
the San Pedro line between Los Angeles and Salt Lake City, which
was rebuilt for long distances after its destruction by flood about
6 years ago.

The question of run-off needs more careful consideration, espe-
cially the effect on it of forestation, with a view of fixing quantitative
values that may be depended on with more certainty than is now
possible.

Flood depths and flood velocities in mountainous and hilly regions
need more extended investigation in order that the limits of danger
may be more definitely fixed. This is often even more important than
similar investigations in the flood-plain sections, as the floods are
more sudden, more luicontrollable, and more destructive than in
regions of the latter type.

The size and nature of the material available for transportation
are also important factors in floods. Structures that might withstand
the water alone may fail under a bombardment of boulders, often
hundreds and sometimes thousands of pounds in weight, which are
swept along by mountain torrents. More information is also needed
as to the effects of bends of various radii in the canyons and valleys,
as well as the influence on floods of narro wings of the valley walls.



Matthes



1378 DISCUSSION ON FLOODS AND FLOOD PREVENTION

Mr. Many of the factors mentioned are geological in their nature, and

Puller, .j-j-^g-j. investigation would naturally take place in connection with
that of the numerous other geological problems involved in flood regu-
lation or control.

Final Statement.

A few of the geological factors entering into the problem of flood
control and prevention have been stated briefly. Many others would
undoubtedly develop in the field. The writer believes that some of
these factors will be generally acknowledged as of unquestionable im-
portance. At any rate, it is not safe to pass them by without mature
consideration in any broad investigation of floods.
Mr. Gerard H. Matthes,* M. Am. Soc. C. E. (by letter). — The writer

endorses very heartily the recommendation contained in the Report
of the Special Committee on Floods and Flood Prevention, urging
the establishment and unification of systematic rainfall, run-off, and
flood observations covering the entire United States in far greater
detail than has yet been attempted ; also, the recommendation contained
in the Minority Report, urging the creation of a special agency, sup-
ported by adequate appropriations, for carrying on this work. Out
knowledge of floods in general is sorely in need of systematic develop-
ment; the recommendations referred to indicate the first logical step
to that end.

There is a crying need for carefully compiled chronological flood
data showing dates of occurrence, flood heights, distribution and quan-
tities of rainfall, rates of riin-off, and notes as to what tributaries were
directly responsible for the formation of flood crests, for all rivers of
importance, and including also the lesser streams which, owing to
geographical or commercial conditions, present special flood problems.
The writer believes that the Committee can render the Profession no
greater service than by outlining a plan of procedure for collecting,
collating, and publishing such facts systematically.

A review of available river stage, stream flow, and rainfall data
reveals a deplorable lack of accurate statistics pertaining to either
floods or to extreme low-water conditions. Yet, these two subjects are
of more economic importance than any others pertaining to the
regimen of streams. Floods are at present attracting unusual interest;
low-water stages and droughts will probably become of more and more
interest as time goes on. As matters now stand, the practising engi-
neer, who is required to study and report on means for abating flood
damage in a given locality, finds three distinct classes of information
at his command :

1. — River stage, stream flow, and rainfall observations, made under
the direction of Federal and State bureaus, and of certain

* Dayton, Ohio.



DISCUSSION OX FLOODS AND FLOOD PEEVENTION 1279

corporations, all of which observations are more or less readily Mr.

obtained in printed form. In this class belong also miscel- ^^"*^^^-

laneous data published in technical journals.
2. — Data pertaining to floods which antedate the periods covered

by existing records, and concerning which but little has been

published in technical publications.
3. — High-water marks.

Each class of data possesses limitations of its own, as will appear
from the following considerations. Data of Class 1 are generally re-
garded as aifording the most reliable information. Unfortunately,
river stage records are available for comparatively few streams, and
rainfall records are obtained at illogically distributed points. The
average river stage record does not cover more than 20 years, a period
entirely inadequate for studying floods with reference to their fre-
quency. Even 40-year records, of which a number are in existence, are
inconclusive in this respect, because they contain rarely more than
one extraordinary flood. The principal short-coming of such records,
however, is that they were obtained primarily to supply information
of a general nature, and not to afford specific information regarding
floods. This fault has been recognized, and is being remedied. The
United States Geological Survey and the United States Weather
Bureau, in recent years, have instructed their observers to record the
maximum or crest stage of each flood, and in case of flood stages
caused by back-water from ice jams, to make notes to this effect.
Previous to this there was no definite practice, some observers report-
ing readings taken at the customary hour, others reporting the crest
stage without stating in all instances the time of its passing. Others
took sufficient interest to obtain a series of readings during a flood,
but the Bureau, in publishing the record, contented itself with
averaging them and printing the meaningless figure thus obtained
as representing the average gauge height for the day. This practice
is still in vogue in some quarters. Many flood heights published by
the Signal Service and the Weather Bureau in their earlier reports,
on investigation, have proved to be distortions caused by ice jams,
and are not to be taken as indications of flood discharge. Failure on
the part of engineers to investigate and check up such records before
using them has resulted in the indiscriminate use of observations
which are not comparable. Some engineers believe that inaccuracies
and inconsistencies of the kind referred to are not likely to affect their
conclusions one way or another. Others, unaware of the defects, have
taken the records for gospel truth, and with painstaking care have
made them the basis of mathematical and graphical studies, assigning
to them a value which is wholly unwarranted.



1280 DISCUSSION ON FLOODS AND FLOOD PEEVENTION

Mr. There is an urgent need for a complete revision of all published

river stage records and stream flow data, in so far as they relate to
floods. The importance of this matter cannot be over-emphasized. A
revision of this kind was undertaken some time ago by the Water
Supply Commission of Pennsylvania for the streams of that State, as
a result of which many of the theretofore published maximum rates
of run-off have been materially increased. The writer's studies in
this field lead him to believe that probably 50% of the maximum run-off
figures published by various authorities, among them the much quoted
ones of the late Emil Kuichling, M. Am. Soc. C. E., do not represent
crest stages at all, but are based on unscientific observations of the
kind just alluded to, and are not even reliable 24-hour averages. The
seriousness of this state of affairs is not generally recognized, and de-
serves the earnest attention of the Committee.

It is generally recognized that the number of river and rainfall
stations should be greatly increased. Co-operation between the Federal
Government and individual States has given good results in some cases.
As an instance of what may be accomplished by a State acting inde-
pendently, may be cited Pennsylvania, which in 1907 took over from
the U. S. Geological Survey a score of river stations, and has increased
this number until at present observations are received from more than
100 stations.

The development of our knowledge of floods by continuing existing
records and starting new ones is only one phase of the subject; the
other phase is to gain knowledge concerning floods which took place
before regular observations were begun. This involves the collecting
and collating of data of Class 2.

Much valuable information concerning floods, their heights and
causes, is obtainable by consulting the files of historical societies,
public and private libraries, newspapers, old diaries, private records,
and unpublished matter of various kinds. Extensive researches by
the writer in this field have taught him that there is a vast quantity
of good material awaiting him who will take the trouble to unearth it.
Such material, when properly interpreted, and utilized with care, so
as to eliminate erroneous data, can be made of great value, covering
as it usually does the major portion of the history of any river, in point
of time. Obviously, a 150-year record, consisting of fairly complete
data relating to great floods, will be of much greater value in determin-
ing a future flood control policy, than a 40-year record of daily gauge
heights obtained on the same stream by a $5 per month uneducated
observer.

Some attempts have been made to compile flood records for a few of
the more important streams of the United States, the data extending
back to the earliest days of settlement. The reason more has not been



DISCUSSION ON FLOODS AND FLOOD PREVENTION 1281

done along this line is that it has not been made anybody's business to Mr. ;•
attend to it, and because, also, of a certain amount of prejudice generally
prevailing against the use of non-technical data. Federal and State
bureaus have made little effort to collect information of this kind,
principally for lack of appropriations, and partly because of inertia to
be overcome in stepping out of the beaten paths of regular routine.
The nearest approach to work of this kind was undertaken by the
Water Resources Branch of the U. S. Geological Survey, when, in
Water Supply Papers Nos. 96, 147, and 162, it published accounts of
destructive floods during the years, 1903, 1904, and 1905, respectively,
including many references to early floods. The Weather Bureau has
also listed early floods for a few streams. Without doubt, both these
bureaus possess a large quantity of material, which, if carefully checked
and amplified, would become most valuable contributions to flood
literature. In its present condition, such information lies dormant,
and is of no benefit to the Profession.

Reverting next to Class 3, High-water marks, it appears that man-
kind, since the earliest times, has manifested a keen interest in the
destructive action of rivers. Before the days of regular observations,
it had become a frequent practice to perpetuate by permanent marks
the height attained by great floods. The extent to which this has been
done along the streams in the early settled portions of the United
States is remarkable, as will be testified to by those who have had oc-
casion to make systematic search for such marks. Here, as elsewhere,
in dealing with information pertaining to floods, it is of the utmost
importance to guard against errors, and it is necessary to check the
marks, not only against each other, but by comparing the information
which they furnish with that obtained from accounts and other data.
When made on factories, mills, pumping stations, and bridges by the
mechanics, millers, or engineers thereof, such marks, as a rule, are well
recorded and reliable.

The present condition of our knowledge of floods is such that few
practising engineers are placed in a position where they can utilize
to advantage any one of the three classes of flood data here discussed.
Lack of facilities, and lack of access to old files or other sources of in-
formation stand in the way, and frequently the problem must be solved
by guesswork where reliable data might have been used had they been
available in conveniently accessible form. It should be made the
province of Federal or State bureaus to remedy this condition. To do
so calls for considerable research and academic work, neither of which
should devolve on the practising engineer, or become a source of ex-
pense to his client.

In conclusion, the writer wishes to state that it is to be regretted
that the duties of the Committee were limited to the investigation of
flood matters only; they should have included an investigation of low-



1282 DISCUSSION ON FLOODS AND FLOOD PKEVENTION

Mr. water conditions. The subjects are closely related, and the methods



Matthes.



Barrows



which may be utilized for systematizing the knowledge concerning the
one could without doubt be made to apply to the other. It is to be
hoped that this important feature will not be overlooked when an agency
is created such as suggested in the Minority Report. Such an agency
could handle advantageously matters relating to low as well as to high
water. Great floods have left their imprints where they may be seen
for generations to come ; but low-water stages have left no such records
behind them. Unless an effort is made soon to trace back the happen-
ings of this class which took place in years past, it will become increas-
ingly difficult to glean much on this important subject.

Mr. H. K. Barrows,* M. Am. Soc. C. E. (by letter). — The writer has

been much interested in the report of the Special Committee on Floods
and Flood Prevention, the Minority Report by Mr. Knowles submitted
therewith, and the subsequent discussions by Messrs. Eakin, Hill, Leigh-
ton, Grant, and Groat.

The writer is in general accord with the suggestions made by Mr.
Knowles in the way of amplifying the report of the Committee, and
particularly with regard to a special National agency for making gen-
eral studies of the subject of floods and flood prevention and procuring
systematic and comprehensive data necessary for carrying on such
studies.

Mr. Grant states the situation exactly when he says that data are
not now available with which to design intelligently and economically
the works to carry out the purposes contemplated by the enormous
appropriations for river regulation and water conservation which have
been urged continually before Congress.

The Water Resources Branch of the United States Geological
Survey is the only Government Bureau carrying on systematic river
measurements, and, up to the present time, its appropriation for this
purpose has been very limited. The entire amount appropriated
annually for gauging purposes is only $150 000, which is only partly
available for river measurements and must also be used for the admin-
istrative and other expenses of this Bureau. This sum is absurdly
small, and, in the writer's judgment, the Committee could well make
a specific recommendation to aid in the advancement of our knowledge
relative to floods and other river conditions by urging strongly upon
Congress that the amount of this appropriation be increased very mate-
rially. When the great value of such data is considered from the point
of view, not only of flood investigation, but of water power and water
supply in general, it would seem that the annual appropriation for
this purpose should be at least $500 000.

The discussion of the use of levees in flood prevention brings out
clearly the fact that on large interstate streams a comprehensive

* Boston, Mass.



' DISCUSSION" ON FLOODS AND FLOOD PEEVENTION 1283

scheme for the entire river should be developed, as the use of levees Mr.
often means a change in hydraulic gradient at flood times extending
many miles up stream from the point cf the works.

Although the necessity for the use of levees in the lower Missis-
sippi seems to have been proved, the writer is not greatly impressed
by the figure of 1 365 000 000 000 cu. ft. cited by the Committee as the
storage capacity of levees on the Mississippi between the mouths of
the Ohio and Red Rivers. Although the figures in themselves are
large, it must be kept in mind that the effect of storage capacity on
a stream depends on the tributary drainage area as well as the quantity
of storage. The total drainage area of the Mississippi above the Red
River is about 1259 000 sq. miles, and the 1365 000 000 000 cu. ft.
of storage capacity between levees corresponds to a little less than
^ in. in depth over this drainage area, a quantity which obviously of
itself is not of great importance in retarding flood waters. Considering
only the drainage area of about 329 000 sq. miles between the Ohio
and Red Rivers, the storage in the levees would be only about 1.8 in.
in depth over this area, or not enough to affect greatly the run-off
of even this smaller district. The writer agrees with Mr. Knowles
in this respect, that the potential storage over the surrounding country
and overflowed land is much greater than the volume confined between
levees. The important function of the levees is in increasing the
carrying capacity of the channel and confining the flow, rather than
in storage.

The recent report of the Miami Conservancy District is an excellent
example of an intelligent solution of the flood problem on one stream
wholly within one State. For the larger or more extended problems
involving interstate streams, obviously, studies should be made by
some National agency, and the writer believes that the Committee
might well take a more positive stand on this matter and urge the
creation of such an agency, supported by an adequate appropriation.

N. C. Grover,* M. Am. Soc. C. E. (by letter).— With the occu- Mr.
pancy of the river banks and bottom lands by cities, towns, and farms, ^''°^*^''-
the damages caused by floods have increased, until the problem of
their control has become of National importance. Recent disastrous
floods have led to the appointment of municipal. State, and Federal
commissions for the study of local conditions or special phases of
the flood problem. The appointment of a Special Committee of the
American Society of Civil Engineers for study and report on floods
and their prevention is a manifestation of the interest in and im-
portance of the problem.

The duty of the Society to make a broad, comprehensive, and
unbiased study of the flood situation having been recognized by the

* "Washington, D. C.



Grover



1384 DISCUSSION OX floods and flood prevention

Mr. appointment of this Committee, a heavy burden of responsibility rests
on the membership in its attempt to guide the various governmental
organizations to an adequate solution of the flood problems. The
Society must now see to it that this responsibility is discharged prop-
erly and adequately, and its final action should be sound in principle,
broad in scope, and definite in application, so that it may serve as a
basis for safe, adequate, and unbiased legislation.

At the present time flood work is scattered through many organiza-
tions. Several States and municipalities have investigated local flood
problems. The United States Weather Bureau has developed an
efficient service for flood warnings on the principal rivers of the
country, and has collected many valuable records of river stage. The
United States Geological Survey has collected many records of flood
discharge, as a part of its work in systematic stream gauging, and
has made topographic maps of river basins and of reservoir sites that
are invaluable in a study of floods and their control. The Corps of
Engineers, United States Army, has charge of all construction on
navigable streams, and, incidentally, has collected much information
relative to floods on subh streams. Each of these various organizations
has conducted its work in full recognition of the work of the others,
but with little actual co-operation. There has been, however, no
organization to undertake a broad, comprehensive study of the whole
problem.

Flood problems range in complexity and importance from those
surrounding a local flood affecting only a small area in a single State
to those involving the welfare of the people of several States, or even
important international questions. Present methods of treatment are
almost invariably local and piecemeal, without proper consideration
of the general situation, of possible antagonistic results of different
projects and methods, or of the possibility of disastrovis eflfects of
projects on unimproved parts of the river system. They do not bring
about, therefore, that general orderly improvement in the condition
of river stage and erosion that should result from correct and adequate
methods of treatment. The co-ordination of projects according to
sound and harmonious methods should also eliminate useless efforts
and waste, and secure the desired results at a minimum cost.

The elements of a general programme for flood control, on which
recommendations should be made in the report of the Society, are:

1. — The development by appropriate research of a fuller knowledge
of the laws of river hydraulics and physiographic processes
which is prerequisite to a sound practice of river control;

3. — The expansion and improvement of methods of control, on
the basis of fuller scientific knowledge, including, possibly,
vertical as well as horizontal control of streams, with con-



DISCUSSION ON FLOODS AND FLOOD PREVENTION 1285

sequent reduction of necessary levee heights, contraction of Mr.

levee systems, and elimination of harmful reactions between

different projects and methods;
3. — The co-ordination and standardization of the collection of data

as to quantity of water, debris in transit, and consequent

adjustment of grades and channel forms;
4. — Surveys necessary as a basis for the design of regulatory works,

estimates of cost, and of the effects of such works ;
5. — Agencies to be used in collecting data and in building and

operating the necessary structures ;
6. — Division of cost of such structures among the organizations

co-operating, or the parties benefited;
7. — Possible combination of flood control, navigation, water power,

drainage, and irrigation, under the same general regulative

programme ;
8. — -The organization of a special Federal bureau equipped to

execute a full programme of river improvement.

The Report of the majority of the Special Committee on Floods
and Flood Prevention, which was presented at the Annual Meeting,
was discussed by the minority as a Progress Report. The Chair-
man, later, requested that the report be considered as the Final Report
of the Committee. It is unsatisfactory to the writer, however,
either as a progress or a final report, largely because of its apparent
lack of clear insight into several important phases of the flood question
and of definite suggestions or recommendations. It calls attention
only to the need for additional and standardized physical data, and
discusses certain suggested methods of flood control in a manner that
favors those now in vogue. It ignores the necessity for developing a
science of river hydraulics and improving the practice of river control,
for active co-operation of agencies in collecting data, in building and
operating structures, and in dividing costs, and for the co-ordination
of all flood work under one directing head and in one programme.
The Minority Report is a great improvement on that of the majority,
in many respects, but does not accomplish the purposes, as they



Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 110 of 167)