Page images

With the construction of a wall along the low-lying portions of the river bank at Pittsburgh, the flood commission believes that th's system would be sufficient to protect the city against all but the most severe floods. A larger system of 28 reservoirs, costing about $28.000,000, and a still larger one of 43 reservoirs, costing $34,000,000, have been outlined and suggested as alternatives.

The extent to which this system of reservoirs will benefit the navigation of the Monongahela, Allegheny, and Ohio Rivers depends upon the amount of water that can safely be retained in the reservoirs until needed for low-water assistance. The flood commission estimates that the equivalent of 50 per cent of the storage capacity of each reservoir could be retained for this purpose without impairing its efficiency for flood prevention. Col. Newcomer, of the Corps of Engineers, United States Army, after an examination of the report of the Pittsburgh Flood Commission, is of the opinion that the capacity of the system of 17 reservoirs which they recommend is not sufficient to prevent floods and at the same time afford much benefit to navigation. The difficulty of reaching a final conclusion on this point is due to the lack of complete and accurate statistics of rainfall and river discharge. The findings of the flood commission are, for this reason, based upon assumptions the accuracy of which it is difficult to determine. Mr. Leighton, however, suggests a plan of operating the reservoirs, by which the system of 17 could be effectively utilized both for flood prevention and for aiding navigation. His plan is that the reservoirs should be operated primarily for flood prevention from about November 1 to April 1, and entirely for aiding navigation from about April 1 to November 1. A study of past floods shows that 80 per cent occur between November 1 and April 1, and only 20 per cent in months from April to November. Furthermore, those occurring during the latter period are not as severe or as damaging. Therefore, Pittsburgh could well afford to remain subject to the occasional summer floods in view of the practically complete relief from the violent ones of winter and spring. After the 17 reservoirs have been constructed and operated for a sufficient length of time to demonstrate their effectiveness, additional reservoirs could be built if found necessary.

The communities interested in this project propose to raise the larger portion of the money required to carry it out by bond issues, and ask that the Federal Government, because of the benefit which will result to navigation, should cooperate with them by bearing a share of the expense, and by authorizing the Corps of ngineers to construct and operate this system of reservoirs. They assert that prompt action is necessary, or the most desirable reservoir sites which they have sel eted will be appropriated by private parties for the development of water power.

After careful consideration of the problem of utilizing storage reservoirs for flood prevention, the commission has arrived at the following conclusions:

1. As the country develops the necessity for controlling floods becomes of greater importance, both in respect to improved property in thickly populated districts and to valuable unimproved lands which are needed for agricultural or manufacturing purposes. Losses from floods are not confined alone to the destruction or damage of property. but also result from the inability to utilize large areas threatened

by floods. In the case of many streams the adoption of some means of flood prevention has already become most urgent, because of the constantly increasing losses due to floods.

2. The use of storage reservoirs as a means of controlling floods, although expensive, becomes more practicable where the value of property liable to damage is great and where the reservoirs can be used simultaneously for other beneficial purposes, such as power development and aiding navigation. The question of feasibility of storage reservoirs depends upon the relation between the cost of construction and the benefits to be derived in each particular case, and the benefits increase rapidly as the country develops. The time has already come, especially in the more thickly settled river valleys, when a stream must be considered with a view both to minimizing its harmful influences and to securing the maximum benefit from all its uses.

3. The lack of adequate information makes it impossible for the commission to specify on what streams the construction of reservoirs would result in benefits commensurate with the cost. In most cases little is known concerning stream flow and the physical conditions causing floods, or whether there exist reservoir sites suitable to afford the necessary relief. The extent of damages caused by floods on different streams has not, as a rule, been accurately determined, nor have investigations been made to ascertain the relation of the cost to the benefits that would be derived from the construction of reservoir systems. The commission is of the opinion that each case must be considered on its merits, after a thorough investigation of all the facts, and strongly urges the necessity of careful studies such as the one recently made by the Pittsburgh Flood Commission.

4. The Federal Government has no constitutional authority to engage in works intended primarily for flood prevention or power development. Its activities are limited to the control and promotion of navigation and works incident thereto. The commission is of the opinion that flood prevention is primarily a local problem, and the work of controlling floods should in the first instance be undertaken by the minor political subdivisions, but that the Federal Government may very properly participate with the localities in carrying out such works on navigable streams, where a substantial and necessary improvement to navigation will result. Unless some such policy as this is adopted and adhered to, there is grave danger that the Federal Government may go outside its proper jurisdiction and become involved in enormous expenditures which are for local benefit. It has sometimes been urged that the Federal Covernment should undertake works for flood prevention on nonnavigable streams which happen to cross a State boundary line. It is clear that in such a case, if navigation is not concerned, the Federal Government should have nothing to do with flood prevention. A method is provided in the Constitution by which the States may cooperate for this purpose.

5. The extent to which the Federal Government should participate in the expense of constructing a reservoir system at the headwaters of a navigable stream should be determined in each particular case by an investigation of Government experts possessing the necessary training and facilities for undertaking a study of this nature. If such investigation shows that the promotion of navigation will require the reenforcement of the flow of a stream during the dry season through

the aid of storage reservoirs and shows the number and cost of reservoirs necessary for this purpose, the Federal Government will have a satisfactory basis for sharing in the expense of constructing a larger system intended also for preventing floods. In this connection it should also be noted that the prevention of floods will indirectly benefit navigation, but this alone is not sufficient reason for the participation of the Federal Government in reservoir projects.

6. The commission believes, for reasons given on pages 25 and 26, that the Federal Government is justified in investigating the proposition to construct a system of reservoirs on the Allegheny, Monongahela, and their tributaries, such as that recommended by the Pittsburgh Flood Commission after its extended study of means for securing protection against further damage from floods, with a view to determining whether the Federal Government should cooperate with the localities in its construction. The commission accordingly recommends, following the plan suggested in the preceding paragraph, that a preliminary investigation be made to determine whether such reservoirs are needed to supply sufficient water during dry seasons to operate the present and proposed systems of locks and dams in the Allegheny, Monongahela, and Ohio Rivers, and to what extent the Federal Government, on the basis of their benefit to navigation, is justified in participating in the expense of their construction. In order to ascertain these facts the commission recommends that a joint board of engineers, one to be appointed from the Corps of Engineers, United States Army, and one from the Geological Survey, be created by Congress to investigate this question and to report to Congress not later than the beginning of the next session, and that the sum of $5,000 be appropriated to defray the expenses of this board.




The commission has given special consideration to the influence of forests upon the navigability of streams. The benefits attributed to forests are greater regularity of stream flow, the prevention or amelioration of floods, and the prevention of erosion, and the consequent silting up of navigable channels.

In recent years this subject has been widely discussed and many conflicting opinions expressed. Members of the Corps of Engineers, United States Army, and metereologists, as a rule, are inclined to minimize the influence of forests, while geologists, foresters, and others are inclined to emphasize it, and civilian engineers are about equally divided. There is a prevalent impression that deforestation is alike responsible for droughts on the one hand and floods on the other. The prevailing difference of opinion on certain phases of this question is due primarily to the lack of accurate information, which makes it extremely difficult to reach any final conclusions. Hence one purpose of this report is to clarify, if possible, the discussion, and to encourage further investigation. Certain conclusions and recommendations are also presented as to the most salutary policy to be pursued in the future. An exhaustive report on Forests and Water in the Light of Scientific Investigation has been prepared for the

commission by Mr. Raphael Zon, chief of silvics, Forest Service, which brings together the results of investigations in this country and abroad, and contributes a large amount of valuable information not hitherto available. This is attached as Appendix V.

Effect of forests upon stream flow.-Forests are said to cause a greater uniformity of stream flow by reducing the amount of water that reaches the streams in wet weather and increasing it during dry periods. It is generally admitted that forests exert an influence of this kind, but to ascertain its extent is a very difficult task. Attempts to arrive at some conclusion inductively by a comparison of river gauge records with the changes in the forested area of the river basin, making due allowance for the amount of precipitation, have not as yet produced any definite results. Two investigations of late which are worthy of mention have failed to discover any appreciable connection between stream flow and forestation. Prof. D. W. Mead, of the University of Wisconsin, has recently made a careful study of the rivers of that State and finds no changes in their flow which are not due to precipitation. Deforestation in these river basins reached its maximum during the period 1882-1892, and since the latter date has declined considerably, but the river discharge records show no corresponding change. The second investigation was made by Lieut. Col. Edward Burr, of the Corps of Engineers, for the Merrimac River, in the States of Massachusetts and New Hampshire. In this case conditions were particularly favorable for an investigation because of the existence of accurate daily gauge readings since 1849, and also records of precipitation since 1858. The Merrimac River basin was extensively deforested from the time of the early settlements until the period 1860-1880. Since this period approximately 750,000 acres of abandoned farm lands have reverted to timber, resulting in an increase of more than 25 per cent in the forested area. Col. Burr, after a careful study of the records, finds no variations in stream flow which correspond at all to the period of deforestation, and later of reforestation. It would manifestly be incorrect, however, to maintain that these two investigations settle the question, particularly since in both cases there are natural reservoirs at the headwaters of the streams studied, which exert a potent influence upon the uniformity of their flow.

Two other inductive studies of scientific value were made in 1907 by Mr. Leighton, of the Geological Survey, and Messrs. Hall and Maxwell, of the United States Forest Service, the purpose of which was to show that a number of the streams in the eastern part of the United States, such as the Ohio, Tennessee, and Connecticut, were becoming more irregular in their flow. This increasing irregularity was attributed in both studies mainly to the deforestation of the river basins of these streams, but no data showing the amount of deforestation were presented in support of this conclusion, and it must therefore be accepted simply as the opinion of the writers. It is probable that a number of other influences which naturally accompany an increase of population, such as encroachments along the shores, the building of roads, and the increased cultivation of farm lands in these river basins, have also been important factors.

1 The flow of streams and the factors that modify it, with special reference to Wisconsin conditions. (Bulletin of the University of Wisconsin, No. 425.) 2 House Document No. 9, 62d Cong., 1st sess.

* Report of National Conservation Commission, vol. 2, pp. 85-125.

In general, studies of an inductive nature are impossible because of the lack of accurate records of river discharge, precipitation, and the area of a watershed under forest cover at different periods. Furthermore, on a river basin of considerable size there are so many conflicting influences affecting stream flow that the influence of forests must be marked in order to be detected. Experiments are now being carried out on two streams in Switzerland, under the direction of the Hydrographic Institute of that country, and also by our Forest Service, in conjunction with the Weather Bureau, on two streams at Wagon Wheel Gap, Colo., to discover by a study of river records the effect of forests upon stream flow. In these cases streams of very small drainage area have been selected in order to avoid the difficulties which would be encountered on large river basins. The two streams under investigation in Switzerland have a watershed of 140 and 175 acres, respectively. They are similar in topography, geographical formation, soil, and latitude, but one is 90 per cent wooded and the other only 30 per cent. The plan of the experiments conducted by the Forest Service in Colorado is to collect precipitation and stream-flow records for a number of years for both streams as a basis of comparison, and then to deforest one of the tracts and note the results. This will require a considerable number of years before final results can be obtained.

Some experiments are also being conducted on the tributaries of the Pemigewasset in New Hampshire by the Geological Survey to discover what changes in the flow of these streams has resulted from deforestation and forest fires. In order to save time the plan is to compare the rate and volume of flow in the streams draining deforested and burned-over areas with those having forested watersheds. Eight different streams with basins presenting almost every conceivable condition from a virgin forest to complete denudation are being compared and the effect upon run-off and stream flow of each storm noted rather than the average effect over a long period of time. In this respect this investigation is a novel one, and its results will be awaited with much interest. Care has been exercised to obtain complete precipitation and snow records, and in the final results allowance is made for these, as well as for the difference in the slope and other factors that enter into the question.

A popular method of showing the influence of forests upon stream flow, which although inductive can not be considered scientific, is to compare the flow of two streams, one with a wooded and another with a deforested watershed, or to cite cases where springs have dried up after forests have been cut off, without any reference to the amount of precipitation in each case. A large number of examples of this kind were collected by the National Conservation Commission. Such illustrations, however, should not be regarded as conclusive evidence of the harmful effect of deforestation. They are of little scientific value unless records of precipitation are also given to make certain that the drying up was not due primarily to this cause. Such illustrations would also be more valuable if they were accompanied by definite information as to the extent and manner of deforestation, whether the land was entirely cleared or left covered with underbrush, or whether it had been burned over or broken up for agricultural purposes. There are examples of equal value where the flow of streams has improved as the result of deforestation and where

« PreviousContinue »