THE FUTURE WATER SUPPLY OF THE CITY OF PHILADELPHIA. By JAMES F. SMITH/'Gt-EI REPRINTED FROM THE JOURNAL OF THE FRANKLIN INSTITUTE, Tfor October, 1879. PHILADELPHIA! Merrihew & Son, Printers, 1 .‘i> North Third Street. 1879. [Reprinted trom the Journal of the Franklin Institute, October 1870.] THE FUTURE WATER SUPPLY OF THK CITY OF PHILADELPHIA. JAMES F. SMITH, C. E. So much has been said and written on the subject of the water sup- ply of Philadelphia that one may well feel a delicacy in making any further reference to it. The writer has, however, devoted some leisure time and study to its consideration, and, in common with others, appre- ciates the great importance to the city of Philadelphia of a never- failing and abundant supply of pure and wholesome water. In the year 1875, when the Commission of Engineers had under consideration the water supply of the city, the writer made certain suggestions, namely: First, the creation of a system of storage dams in the Schuylkill, where the river was not already occupied by the Schuylkill Navigation Company’s works, and the use of several dams of the upper navigation, and the reservoirs now existing or to be created in the valley of Tumbling Run, near Pottsvilie; second, util- izing more fully the water-power at Fairmount and at Flat Rock dams, by raising these dams and increasing the volume of water from the storage dams when required; third, the building of an aqueduct to lead the water from the Flat Rock dam to the pumping stations at Spring Garden, Belmont and Fairmount, so that pure water might be supplied free from the contaminations and impurities of the Fairmount pool. These plans were adapted to the wants of the existing water supply from the Schuylkill, and to the increase and improvement of the same. This supply is dependent upon steam-pumps with high lifts and inade- quate water-power. Coupled with these disadvantages is the liability to breakage and derangement of pumps and power; whereas the peo- ple of a great city with dependent manufacturing establishments need a constant, reliable and ample supply of water of a purity that may not be questioned. Purity of supply can only be had by a resort to streams of water originating in the hills or ridges where the popula- tion is sparse and the country not extensively un4er cultivation. For * A portion uf this article appeared in the North American. 2 Philadelphia, the Perkiomen creek and its tributaries, rising in the spurs and ridges of the South mountain in Berks county, and in the elevated water-sheds of Bucks county, midway between the Delaware and the Schuylkill, is the source from which the water must come, and gravity the mode of its conveyance. The citizens of Philadelphia have been made somewhat familiar with the question of a gravity supply from the Perkiomen by the consideration of Mr. Birkinbine’s plan as given in the “ Report of the Commission of Engineers” in 1875, and in his reports and papers on the subject. The plan proposed by Mr. Birkinbine for an impounding reservoir is, indeed, favorable for the creation of a large lake, by the erection of a dam seventy or seventy-five feet in height, except in one particular, namely, the great length of the mound dam that would be required. The Commission of Engineers regarded this project with considerable favor, and gave credit, not improperly, to Mr. Birkinbine for his fore- sight in suggesting it. But at the site selected the water surface of the proposed lake would be too low for the city, except for the supply of the East Park Reservoir and the basins below its level, namely, Schuylkill, Corinthian avenue, Delaware and Fairmount; and from what has been said by the Commission, it will be observed that the conduit from this lake was not designed to deliver water into a receiv- ing basin in the city at a greater elevation than 133 feet above city datum. This is manifest from what they say, namely, “ that thorough and careful surveys should be made of the location of the line of con- duit from the Perkiomen dam to the East Park Reservoir.” This deficiency in elevation, it must be admitted, is fatal to Mr. Birkinbine’s location, as it would not be wise to build an aqueduct line to bring in the water of the Perkiomen, only to deliver it into the East Park Reservoir, at an elevation too low to supply Belmont, Frankford, or the proposed Cambria basins. The surfaces of the two former of these are respectively 212 and 168*40 feet above city datum, or 217*38 and 173*78 feet above mean tide, Delaware river, and the proposed Cambria basin about 170 feet above city datum. My attention has but lately been called to an article on the subject of a gravity supply from the Perkiomen, by Chas. G. Darrach, C. E.* Mr. Darrach proposes, instead of Mr. Birkinbine’s lake, an inter- * Proceedings of the Engineers’ Club of Philadelphia, Vol. 1, No. 3, p. 162. 3 cepting canal around the valley, the supply for which is proposed to lie taken from impounding dams to be erected on the main stream and its branches. This plan is ingenious and an improvement on that of Mr. Birkinbine in several particulars, namely, increased elevation, a delivery into the receiving basin of the city at an elevation of 170 feet instead of 133, and saving from destruction the Perkiomen rail- road and the improvements in the valley which would be submerged by the lake. Among the objections to the plan is still the want of elevation. A gravity supply to meet the future wants of the city of Philadelphia should deliver the water into the receiving basin at a much higher elevation than 170 feet of city datum, in view of the certain future extension of improvements and spreading of population over those parts of the city which are now suburban. The plan for a gravity supply by aqueduct from Perkiomen, which I now propose, meets the objection of the want of height of Mr. Birkinbine’s location. It will not require any alteration in the line of the Perkiomen Railroad, which would be necessary should a dam be erected near Schwenksville, and a large amount of valuable property in the valley of the main stream between that point and Green Lane would be saved from destruction. The Location. On the main stream of the Perkiomen, a few hundred yards above the Green Lane station of the Perkiomen Railroad in Montgomery county, eighteen miles above the mouth of the creek, there is an admirable site for a dam, at a point where the stream has broken through a ridge of hard rock, leaving a gap of little more than three hundred feet across with precipitous sides and solid foundations, just adapted to the erection of a safe and durable structure. At this place the dam may tie 90 to 100 feet high, backing the water several miles into a valley, with favorable slopes bounded by hills and ridges. From this dam I propose to carry the water by an aqueduct into the city, and the aqueduct would be so located as to command from 45 to 50 feet in depth of the storage water of the dam. The Course of the Aqueduct Line—its Length, Terminal Basin and Elevation. The route may be thus described ; Commencing at the storage dam 4 or lake, and crossing the Perkiomen Railroad at a sufficient elevation above it, the aqueduct pursues a course nearly straight to the Plymouth Valley, passing a little west of the dividing ground between the waters of the Wissahickon and Plymouth creek, about twenty miles from the starting point at Green Lane. Advancing from this point, after crossing the Plymouth Railroad, the line reaches Barren hill and passes through it into the Wissahickon valley; thence down that valley on the eastern slope of the ridge, crossing the stream, not far from Hermit’s lane, to the high ground in the vicinity, where a distributing reservoir, of very moderate dimen- sions, in connection with the several existing basins of the water-works and the East Park Reservoir when completed, would suffice, and fur- nish a storage capacity of nearly one billion of gallons. The length of the proposed aqueduct line is about 27J miles, and the location such that the water may be delivered into a receiving reservoir, the surface of which might be 240 feet above city datum, with a free delivery from the aqueduct itself, the water surface of which when running in its full capacity would be nearly 249 feet above city datum, or 254 feet above mean tide Delaware river. It may be stated here in regard to the line of the conduit, that whilst none of the valleys to be crossed will require long and elevated structures of masonry, there will be several tunnels, mostly through sandstone. Below Barren Hill the work will encounter mainly gneiss rock. This feature is deemed favorable, as tunneling at the present time, owing to improved methods and explosives, can be done with such great facility that in most cases the expense will be less than for an aqueduct of masonry, and especially so where the rock encountered is found to be of a character to justify the omission of interior brick linings. Tributaries of the Perkiomen Intercepted. In the first 10J miles a number of important streams are passed. Three of them are main branches of the Perkiomen, and three affluents of these branches. They originate chiefly in the hills dividing the waters of the Delaware from the Schuylkill in Montgomery and Bucks counties, more than 500 feet above tide, and extremely favorable for maximum precipitation. It is part of the plan that these streams shall be made tributary to the water supply by the erection of impounding dams or small lakes upon them, with dams of the heights and capacity 5 suited to their drainage areas and the quantity of water required to be stored. The cost for all of them will not, it it is believed, be more than for the dam, and the land and improvements that would be sub- merged and destroyed by the lake at Sellwenksvilie and the reconstruc- tion of ten miles of railroad would be saved. At first the construc- tion will be limited to the works at Green Lane and the aqueduct line to the city, leaving the improvement of the tributary streams to be undertaken one by one, in the future, as the increase in population of the city may demand more water for consumption. The water shed of the Perkiomen above Green Lane station is about seventy-three square miles, and with the tributary streams crossed by the aqueduct line added there is not less than 200 square miles. If 59 per cent, of the annual rainfall of 47*82 inches be taken as collect- able (being one-fifth greater than that of the year 1858 at Philadel- phia, and which Mr. W. J. McAlpine, C. E., has taken in his calcula- tions for the Perkiomen, vide report of Commission of Engineers, 1875), it would give daily 272,000,000 gallons. Of this quantity it may be assumed that 165,000,000 gallons daily average supply to the conduit may be taken, leaving, after deducting for soakage, evaporation and waste of the impounding dams, nearly 100,000,000 gallons to pass away as surplus or flood water. From the Perkiomen above Green Lane village 60,000,000 gallons daily of the supply may be furnished, and when the consumption of water is greatest, in July or August, 72,000,000. The drainage areas of the streams above the aqueduct line have been determined from measurements and calculations based upon a corrected map of the townships of Montgomery and parts of Bucks and Berks counties, which was specially prepared for the purpose. The results are believed to be fair approximations, and are given in the following table: Capacity of the Perkiomen AVater Shed. Table 1. Square miles. Perkiomen Creek at Green Lane, . . 73*00 Great Swamp Creek, . . . 34*47 Ridge Valley—affluent of Swamp Creek, . 9*64 East branch of Perkiomen, . . 46*00 Indian Run—affluent of east branch, . . 6*01 Skippack Creek, . . . 21*48 Towamencin—affluent of Skippack Creek, . 10*55 Total, . . . .201*18 6 Macoby Creek, a stream having a drainage area, above the aqueduct line, of 171 square miles, flows into the Perkiomen at Green Lane, south of the railroad, but is omitted to avoid any interference with the road, which occupies the valley for several miles. This stream may be utilized to the extent probably of 10 square miles of its drain- age area without injury to the railroad. In the annexed table is given both the annual and the average daily flow of the seven streams, based upon the utilization of 28‘41 inches of the rainfall upon the Perkiomen water shed; also the reservoir capacity required above the limit of draught in the impounding dams for a given consumption. Table 2. Name of Stream. Collectable annually— Daily average demand for Daily consumption Deficiencies to be provid- ed for bystor- Million U.S. average —Million age dams— gallons. gallons. gallons. Million galls. Perkiomen Creek, 36,041 98,742,466 60 3,996 Great Swamp Creek, 17,053 46,720,548 28 1,832 Ridge Valley, 4,751 13,016,438 62,287,671 8 560 Etist branch, 22,735 38 2,530 Indian Run, 2,959 8,106,849 5 365 Skippack Creek, 10,465 29,164,383 18 1,215 Towamencin, 5,193 14,227,397 8 592 99,377 272,265,752 165 11,090 This table is constructed upon the basis of 47’85 inches annual rain- tall, or one-fifth more than that of the year 1858 at Philadelphia, as assumed by Mr. McAlpine. There are no records of rainfall upon the Perkiomen in years past with which comparison might be made to establish the correctness of the above assumed data. We have, however, actual results obtained from storage reservoirs in constant use in Eastern Pennsylvania, one example of which will suffice. The Silver Creek reservoir of the Schuylkill Navigation, on the Broad Mountain in Schuylkill county, has a drainage area of square miles. This water shed is rolling table land, not rising in any part to the dignity of a hill, and therefore not very favorable to the flow of a large percentage of rainfall into the reservoir. Notwith- standing this fact it is filled at least twice every year, and this has been the experience of many years. Its capacity is 320,000,000 gallons, and the total quantity stored and utilized for supplying the navigation 7 is 320,000,000 X 2 = 640,000,000 gallons, = 28-82 inches of rain- fall over square miles. Upon the basis of this quantity utilized from 1 square mileSj applied to the Perkiomen drainage area of 201 square miles, we have 101,000 million gallons, as against 99,377 million gallons collectable from 28'41 inches of rainfall (Table 2). It may be said that the region of Silver Creek, being at an altitude of 1500 feet above the level of the sea, is more favorable for rainfall than the Perkiomen district. NAME OF STREAM. Distance from the reservoir.—Miles. Total distance (per map).—Miles. Total distance with allowance for de- viation.—Miles. Fall or descent.— Feet. Total fall.—Feet. Assumed level of water surf, in the aqueduct above C. D—Feet. West branch (Perkionien), 0 0 0 0 0 275- Great Swamp Creek, . 1-375 1-375 1-39 5¥55 2-159 262-84 Ridge Valley, 0-500 1-875 1-89 T575 2-445 272-55 East branch, . 1-875 3-750 3-80 T555 4-720 270-28 Indian Run, 1-250 5-000 5-05 T575 5-350 269-65 Skippack Creek, 3-875 8-875 8-95 5555 9-450 265 55 Towamencin Creek, 0-875 9-750 9-85 5 An 9-630 265-37 Stony Creek Railroad, 4-625 14-375 14-52 5557 13-690 261-31 1st branch crossing Stony Creek, 0-875 15-250 15-40 5A7 14-530 260-47 2d “ “ “ 1-000 16-250 16-40 5A7 15-470 259-53 3d “ “ 1-500 17-750 17-93 5557 16-910 258-09 Plymouth Railroad Summit, 2-250 20-000 20-20 5557 19-050 255 95 Wissahickon, . 6-875 26-875 2715 5A7 25 610 249-39 School Lane (high ground), 0-375 27-250 27-50 5557 25-940 249-06 Distances and Fall of Aqueduct Line (circular). I need only refer to so eminent an authority on climatology as Lorin Blodget in support of the contrary opinion. Mr. Blodget places the Schuylkill county region within the range of 40 inches mean rainfall, as against 42 inches in the Berks, Montgomery and Philadelphia dis- tricts, and, in an article on the climatology of Pennsylvania, says, fur- 8 ther: “ The heaviest rainfall is in the southeast, where the annual mean is 48 inches, and the greatest quantity in any one year is about 60 inches. This represents the first range of hills back from the Del- aware River, on which more rain falls than on the plains of New Jersey and Delaware, nearer the sea. West of the hills, and in the Susquehanna Valley, the quantity of rain falling is much less, only about 38 inches.” The first range of hills referred to by Mr. Blodget are those in which the Perkiomen and its branches take their rise, and in this con- nection it may be stated as a fact that freshets on this stream are more frequent, and higher in proportion to the drainage areaj> than on the upper Schuylkill. Quantity of water required for consumption itir the month of July,, hosed upon the rainfall of the year 185&v augmented by 4 for the Perkiomen, as per table, and increased from the average daily con- sumption by 120 per cent., being the assumed percentage for that month. NAME OF STREAM. Average daily con- stmptioti. — Mil- lion gallons. Quantity for con- sumption to lie carried by the aqueduct! —Mil- lion gallons. it it S.a H 11 H 2 t- 3 « **® s t © 4- a as v I'g'Si 1 Capacity of aque- duct running full. —Million gallons. Velocity per second. —Feet. - - ■ West branch (Perkiomen,), 6G 72-1 72-1 8-000 73,244 2-25600 * Great Swamp Creek, 28- 33-6 105-7 9-625 106,177 2-25800 > Ridge Valley, 8 9-6 115-3 10-042 115,545 2-25800 > East branch, 38 45-6 160-9 11-750 158,236 2-25800 ’ Indian Run, . 5 6-0 166-9 11-917 163,475 2-26800 > Creek, . 18 21-6 188-5 12-750 185,491 2-25388 ■ Towamencin Creek, 8 9*6 198-1 13184 195,000 2-25388 • 165 198-1 This table has been computed for the velocity, from the formula of Robert K. Martin, C. E., and used by him for the air-line conduit 12 feet diameter, for the permanent water supply of the city of Balti- more from the Great Gunpowder Falls. This work is a tunnel 36,495 feet long, from the dam on the Gunpowder River to an artificial reser- SHOWING The Line of the Proposed AQUEDUCT FROM THE PERKIOMEN CREEK TO PHILADELPHIA. DRAINAGE AREAS. SQUARE MILES. I. Perkiomen, . 7300 II. Great Swamp, 34 o0 III. Ridge Valley, . 904 IV. East Branch, . 4(5'00 V. Indian Run, . 6 00 VI. Skippack Creek, 2P50 VII. Towamencin, . 10o0 9 voir (Lake Montebello, near Baltimore), having a storage capacity of 600 million gallons, and elevated above mean tide 163 feet. Mr. Martin’s formula is : v _ V180 X HXD l _ V 2gH 0-0074547 in which H = head, in feet I) = diameter, in inches l = length, in feet r = d 4 The diameters given in the above table are for a conduit running full. For the Perkiomen, another form is preferred, which will admit of a maximum flow limited to a surface sufficiently below the crown of the arch; miles to be of the full capacity for maximum delivery (about 12 feet wide by 13 feet in height), the remainder varying to meet requirements, beginning with 8 feet width and height. Swamp Creek, rising in Berks county, near Boyertown, and flowing into the Perkiomen below the village of Zieglerville, has not been been included in the general plan, because many years must elapse before it will be needed, but it can be utilized whenever required, and its waters brought into the main storage lake near Green Lane by means of a branch conduit. Its drainage area above the height of the Green Lane lake is about 30 square miles. Advantages of the Proposed Plan. Some of the advantages of this plan will appear from the following considerations: First. It will supply, from one point alone, a greater quantity of water than is now required for the present population of the city. Second. It will deliver the water to a basin at an elevation of 27 feet above Belmont, 72 feet above Wentz’s Farm, 69 feet above the pro- posed reservoir at Thirtieth and Cambria streets, 104 feet above East Park, 119 feet above Schuylkill and Corinthian, 125 feet above Dela- ware and 145 feet above the Fairmount basins. Third. It can supply Roxborough and Mount Airy basins by a pumping station at the aqueduct where it passes not far from the former, and save more than 10 200 feet in the lift encountered at the present Roxborongh pumping works, beside permitting them to be discontinued, and the long line of pipes in the pumping main taken up and utilized elsewhere. Fourth. It can supply a basin, which may be provided for the service of Man- ayunk, by a pipe line or branch conduit, tapping the aqueduct line in the Wissahickon Valley, and obviate the necessity for a supply, as at present, from the Roxborough reservoir. Fifth. It will obviate the necessity for all the present steam-pumping stations with their expen- sive and, in some cases, troublesome monster pumps, and leave Fair- mount and its water power, when disconnected from the basins, to be run moderately in the summer, to feed lakes and fountains in the Park or for flushing main sewers. It remains now only to say that the correctness of my statements as to the route for the aqueduct and the capacity and availability of the water sheds are subjects for investigation by submission to instrumental examinations by competent parties in the field, and when this is done, the cost, to which I have not referred, may be estimated very closely. The work, as far as possible, should be monumental in its character, and whilst no money need be spent for mere show or unnecessary ornamentation, yet nothing should be left undone that may contribute in the least degree to strength and durability. In the tunnels we shall have the everlasting rocks; along the slopes rock cuttings, and over streams and valleys permanent structures of stone or iron. The impounding dams must be of undoubted strength, as they certainly may be if founded upon the rock and built of solid impermeable materials, and so on with all other parts. The cost of the work would not be so large as that of the Croton Aqueduct— miles long—which cost originally $8,575,000; but as the completion for a full capacity of aqueduct will not be required for many years, neither will the full expenditure. The cost, whatever it may be, can be spread over several years, and will not be so much of a burden that it cannot be borne by a people who steadily look for- ward to the time when the city of Philadelphia will embrace within its limits a population of two million souls. 11 O C %> > g > g N- CfQ cd 0 0 i-S d cl E m 1~o m $L cr? i-S 5 A. vD p .'-s 5-b J s 3 3 • ... % P CT5 Rainfall at Philadelphia in CnCDOOhPCDCOCnOCDOtOCD OtOG5CD(40iOM*4CDCDO 1858. -1 Cn CD tO >—''Cn H‘ OI Ci 01 M to w Rainfall on Perkiomen, es- do tPdltodjcdcd^OCKM^JH‘ timated one-fifth greater. Cn OlPMCOMbOOH^HOlM Crt cDGooihP-tocohPcn as *—‘ to .to to M1 M t—‘ MM Consumption at an average CD b do CD CD t—1 to 0 CD as bn OS Cn of 60 millon gallons per day.—Million gallons. h—1 CO MQMBOlCO-vlOtCDOOCObO •~40hPbOCDtOOCOtOt—‘hP-Or to to ■—1 '—1 to to to to to 1—‘ t—i t—> 1 Total demand from reser- bt -IXOOtOCO^CD^B^Ci CD O^iOiBCJPaoOoWOMCn voir.—Million gallons. GO COOOIMOOP^JOOPPOP h“1 -v| tP i : : : 1—‘ to : —1 to od —J 4- cd : : • • cn do cn • Co O cd cd : : : : cn >—1 © : o*-1 Surplus.—Million gallons. as CO GC ; : ; ; COGOCO; t—‘O os CD h-± h-L ' COM4 GO -4 do do 00 00 ob do 00 Oi Oi Oi Oi Oi Oi Oi Rainfall on Perkiomen, I greater. Oi CD 05|M Oi Oi Oi Oi 01 Oi Oi CD CD CD CD CD CD CD ®|M 05|M ®|M 05|M 05IM 05|M 03|M Percentage of the rainfall collectable per annum. to 00 41 1—1 28*41 28-41 28-41 28-41 28-41 28-41 28-41 Inches of rainfall collect- able per annum. CD jr> do *4 -4 t—i to 1—1 CO Ox O JO JO _4m J Deficiency. — Million gal- lons, 13 Iii this connection there is now one other matter that deserves the most serious consideration, and which should be an incentive to early action by the people. It is this, that long before the population shall have reached the figure named, the waters of the Schuylkill will have become so contaminated by sewage, in consequence of the great increase in population along its banks, as to be totally unfit for domestic pur- poses ; even the next decade may produce a marked change, hence there is no time to be lost in determining the course that should be adopted to secure pure water for domestic as well as for manufacturing purposes. Much time must necessarily elapse, in any event, before a work adapted to the wants of the city can be brought into use. All the Delaware schemes are too costly, as was shown by the commission in 1875, and require no further consideration. So, after all that has been said and written upon the subject of the water supply for the future, if not notably for the present, there is no other stream, in my judgment, of equal purity with the Perkiomen within reasonable distance from which an adeequate, pure and never- failing supply can be brought; and in closing this paper I must remark that I very cheerfully resign to Mr. H. P. M. Birkinbine the credit of pointing out that stream, and for myself only claim the plan of tapping it at a higher and more favorable point and intercepting and utilizing the headwaters of its principal branches, which fortu- nately lie in the track and cross a direct aqueduct line to the city over high ground. THE JOURNAL OF THU FRANKLIN INSTITUTE. Devoted to Science and the Mechanic Arts, The Journal ok the Franklin Institute is issued in monthly numbers, of seventy-two pages each, largely illustrated, forming two volumes annually. Its object is to encourage original research, and disseminate useful knowledge in all matters relating to the practical application of science, hut more espe- cially to engineering and the mechanic arts. The number for December, 1878, completed the one hundred and sixth vol- ume of the Journal, and closed the fifty-third year of its existence, and its next volume has commenced under very favorable auspices. Under the direction of the Committee on Publication, with its list of able scientists and engineers, as contributors, largely increased, and with the fact that it is the only Technological Journal published in the United States without any private pecuniary interest, sufficient assurance is given that it will main- tain its high position as a leading organ of technology and a standard work of reference. Beside a great variety of matter of general interest, the Journal contains the proceedings of thje meetings, and has contributed in a very large degree to the usefulness of the Institute, which should especially commend it to the support of the members. The Committee on Publication -are desirous of increasing its circulation, being fully assured that it will more than repay the small outlay required to secure it. As the Journal circulates extensively among scientific men, engineers and manufacturers, advertisers will find it to their advantage to avail themselves of its advertising columns. SUBSCRIPTION PRICES. To Members, $3.00 per Year. Single Copies, 25 cents. To Non-Meuibers. 5.00 per Year Single Copies, 50 cents. RATES OF ADVERTISING. 1 Year. 8 Months. 3 Months. 1 MoNTn. 1 PAGE, $00.00 $32.00 18.00 $10.00 'A “ 32.00 1S.00 14.00 6.00 lA “ 18.00 14.00 10.00 4.00 Communications for the Journal ami business letters should be addressed to the Secretary of the Franklin Institute, Philadelphia, Pa.