THE JOHNSTOWN DISASTER
JUST before going to press, we received the following important dispatch from Mr. H. W. Brinckerhoff, C. E., of our editorial staff, who is in the Conemaugh Valley making a personal investigation of the dam and its surroundings, in order that the readers of THE ENGINEERING AND BUILDING RECORD may have as soon as possible a reliable description of the dam, the failure of which caused such fearful ruin, from one whose experience and judgment gives weight to his conclusions.
Mr. Brinckerhoff telegraphs:
The dam appears to have been well built of good gravelly clay, about 8o feet high by 800 feet long on top, 20 feet wide on top, about 300 feet wide at bottom, both slopes well rip-rapped. Failure appears to have been due to overflow, caused by inadequacy of spill-way and aggravated by the fact that the dam was two or three feet lower at center than at ends, causing overflow to take place at weakest point. About 300 feet of the center of dam is gone. lts destruction took about half an hour from time serious overflow began.
We have italicized lines in the above dispatch which, taken in connection with Mr. Brendlinger’s account before the American Society of Civil Engineers, published on another page, of the manner in which the old break was being filled up, accounts for the rapid destruction of the dam, and renders it probable that the suggestion made at the meeting of the Society, that it was this new part which was carried away, was correct.
It seemed improbable that the Engineers of the State of Pennsylvania should have been guilty of negligence in building so important a structure, and our dispatch shows that what is left of the dam is well built-their part of the work. For the height, however, at which the dam was left, or to which it had settled during nearly 50 years, the spill-way was inadequate. And though the dam failed from the erosion of its face, and not from the pressure of the water, some one has an awful responsibility for not having made sure that the spill-way was of ample dimensions for any possible flood, and that the crest of an earthen dam holding such a large quantity of water, above a populous valley, was not entirely free from any danger of overflow.
This should, and undoubtedly will, be made a matter of judicial investigation, and it is to be hoped that the parties who are responsible, either through merely ignorant negligence, or through a possible culpable disregard of warnings received, will be held accountable, as far as is possible, for the damage occasioned by the failure to take such precautions as were demanded by the situation. And this demand for an accounting should proceed, so far that persons owning, or being in any way responsible for, dams whose failure would cause either loss of property or loss of life, will see it directly to their comfort and interest to secure proper supervision of their design, construction and maintenance.
THE JOHNSTOWN DISASTER.
Johnstown – and this name is intended to cover the suburbs, Cambria City, Kernville, Morrellville, Coopersdale, Woodvale and Conemaugh Borough-is 276 miles west from Philadelphia and 78 east from Pittsburgh, on the Pennsylvania Railroad, lies at the junction of the Little Conemaugh and Stoney Creek, and 10 miles below the dam. It was at the foot of the railroad that connected the Eastern and Western Divisions of the old Pennsylvania Canal, having been the point of transshipment from the cars to canal boats, and was 105 miles by the canal, which has long been disused, from Pittsburgh.
The dam on the Little Conemaugh, whose destruction has caused so much loss of life and property, was commenced in 1840 to give a supply of water to the Western Division, under the direction of the late William E. Morris, C. E. His plan was for a dam 850 feet long on top, to raise the water 62 feet and impound 480,000,000 cubic feet of water, flooding 420 acres, and he estimated for 110,000 cubic yards of good earth and 100,000 cubic yards of slate and stone; probably intending to build a dam of loose stone and face it with earth. The water was to be drawn off through iron pipes uniting in a masonry culvert.
The reason which induced Mr. Morris to select the Little Conemaugh instead of Rock Creek, which was the larger stream, was that the floods in the Rock Creek were much worse than in the smaller stream, and its descent more rapid. The site chosen, 10 miles above Johnstown, presented solid rock at each end of the dam, giving a safe spill-way, and “the channel below was narrow and protected from evaporation by woods and mountains.”
This dam stood until about the time that the Pennsylvania Canal was abandoned, having passed into the hands of the Pennsylvania Railroad Company. It was then broken down to within some thirty feet of the original surface of the valley, as has been said, to prevent just such a catastrophe as has occurred, and it is also said that it was breached by a flood.
About two years ago the property was bought by some persons who formed a club for hunting and fishing, repaired the dam, and made a very pleasant country resort for its members. Although the dam, as repaired, has been described as over 100 feet high, Mr. P. F. Brendlinger’s statement at the meeting of the Society of Civil Engineers, that the old spill-way was used for the repaired dam, renders it improbable that the water was raised more than was originally intended, and consequently we can calculate that the water set free amounted to 480,000,000 cubic feet, which the original dam was calculated to contain, with the quantity added by a superimposed height of from four to six feet on an area of 420 acres, or, say, somewhere near 600,000,000 cubic feet of water, which was about 300 feet above Johnstown.
Through the possession of abundant supplies of coal a large manufacturing population had grown up about Johnstown, so that in the 20 miles from South Fork to Nineveh there were some 50,000 inhabitants, 30,000 of whom were at Johnstown. The Cambria Iron and Steel Works, which made 210,103 tons of Bessemer ingots, and 126,662 tons of steel rails in 1887, and, according to figures published by Dr. Cyrus Elder, their attorney, furnished more ton mileage of freights than any one of fourteen States, was the principal employer, though there was another steel works, woolen mills, etc., in the valley.
The various smelting furnaces had dumped a good deal of slag on the banks of the Conemaugb, narrowing its channel, and the Pennsylvania Railroad had raised quite a high embankment and built a very substantial stone bridge, with seven segmental arches, across the river, materially restricting the waterway.
Rock Creek had been in flood for some time, and had carried away dams and booms, raising the water and filling the river with logs and drift, which was doing some damage, and had filled portions of the towns with water, when the flood from the broken dam came down the valley with its added contribution of drift, which probably immediately clogged the arches of the bridge, for the waters rose rapidly 38 or 40 feet, as stated, washing the houses in and about Johnstown off from their foundations and piling them, in a confused and heaped up mass of wreckage, in the angle between the bridge and the shore, until it covered an area of six acres, as stated by Mr. Shinn. The inhabitants had generally kept in the houses, in spite of a warning conveyed by Mr. John G. Parke, and when an overturned stove or lamp set fire to the mass the possibility of saving any lives vanished.
The mass of wreckage turned the thread of the torrent through the Cambria Iron Works, the Gautier Steel Mill, and the best residence and business streets of the city; also wrecking the Pennsylvania Railroad Company’s roundhouse and carrying away some 28 locomotives. It also formed an eddy, so that persons in houses floated from their foundations in the lower part of the town were rescued far above the point from which they started.
If the Pennsylvania Railroad had been content with a long span iron bridge at that point, it is very probable the loss of life and destruction of property would have been much less than it was, for the bridge would have been carried off its abutments by the first houses that struck it, and there would have been no holocaust, as the houses would have passed between the piers ; and lastly, it does not seem possible that the water could have risen as high as it did, for no dam would have been formed. If this view is correct, though many houses would have been wrecked, a large part of the town would not have been lifted off its foundations.
W. P. SHINN, who had left Pittsburgh Tuesday morning gave the following particulars of the dam and its failure, not from personal examination, but as he explained, from information gathered at Pittsburgh: The dam was about 1,000 feet long, 70 feet high, 50 feet thick on top, and 300 feet on the bottom: The waste weir, excavated in the rock, was 75 feet long and four feet below the crest of the dam. The sides of the reservoir were rock, covered with shale and a little earth. The reservoir had never before been full enough to discharge over the crest of the dam.
John G. Parke, C.E., who was directing some drainage work for the club which owned the grounds, and had repaired the dam, saw in the morning that, though at night the streams were running ordinarily full, the heavy rainfall of a night had made a flood imminent. He immediately set thirty or forty men cutting a new spill-way 20 or 30 feet wide, which was sunk to the solid rock. In spite of this the water continued to rise at a rate of about ten inches per hour, and seeing that it would overtop the dam, he rode to the South Fork Station on the Pennsylvania Railroad about noon, and had a warning telegraphed down the valley. Although this warning was heeded in the upper part of the valley it was disregarded in Johnstown, where they had been alarmed before about the dam, and the water was then falling from a flood in Rocky Creek.
The dam broke between 2:45 and 3 o’clock, taking out a width of between 200 and 300 feet about two-thirds of the way down and about 100 wide to the bottom. All the water ran out in about forty-five minutes. Mr. Shinn said that the failure of the dam did not occur from pressure, but from denudation of its surface after the water had run over its crest. Where the valley was narrow the water swept everything away, and in Johnstown cut a swath 1,000 feet wide through the iron works and the principal business and residence portions of the town.
A new stone viaduct, which had been built by the Pennsylvania Railroad Company, probably caused a large percentage of the loss of life as it withstood the flood and caused a barrier to accumulate that made a powerful eddy, backing water up to the upper part of Johnstown and forming an obstacle against which houses, etc., were dashed.