DDD 2. REEXAMINATION OF ST. LAWRENCE RIVER. REPORT OF CAPT. SMITH S. LEACH, CORPS OF ENGINEERS, UNITED STATES ENGINEER OFFICE, Burlington, Vt., July 1, 1894. GENERAL: I have the honor to transmit herewith my annual report on the reexamination of the St. Lawrence River under an allotment from the appropriation for survey of Northern and Northwestern Lakes, 1891. Very respectfully, your obedient servant, Brig. Gen. THOMAS L. CASEY, Chief of Engineers, Ú. S. A. SMITH S. LEACH, Captain, Corps of Engineers. An allotment of $4,275 was made May 2, 1893, and became available on July 1 following. It was based on the estimated cost of a resurvey of the main ship channel for a width of 2,000 feet from Lake Ontario to the foot of the Brockville Narrows, at Morristown, a distance of 40 miles. Owing to the very uneven conformation of the bed of this part of the St. Lawrence the method of isolated soundings heretofore employed in all hydrographic surveys of a general character was inherently defective, and several shoals not disclosed by the original survey had been reported. It was desired to examine the part of the channel used by deep-draft vessels under such conditions as to leave no possibility of points of rock which could be touched by vessels remaining undiscovered. It was decided to employ the method of continuous sweeping, for many years in use in verifying the removal of rock to certain specified planes, but never before adapted to use on such a large scale. The apparatus devised and the method of working it are described in this report in a general way only, as the work remains unfinished and some details will be modified in future. A decked scow, 60 by 15 feet, was anchored near mid-channel. Two anchors, one backing the other, were used, and in placing the scow the anchors were let go, the proper length of cable paid out, and the tug made fast alongside, head downstream, and worked at full throttle until the anchors held the strain without dragging. If they failed to hold, they were raised and thrown again a little to one side of their first position. At the stern of the scow the end of a three-eighth-inch steel wire cable was made fast, and the cable was run out, with can buoys attached at intervals of 250 feet, until 2,750 feet were in the water. This part of the cable was called the permanent radius, and was the shortest line used until near the end of the season, when work was begun at 2,000 and finally at 1,700 feet from the scow. The last distance was found inconveniently short for a full sweep of 2,000 feet, but that or even less will do for narrow channels. At the lower end of this permanent radius a thimble was placed in the cable, and a second cable, called the variable radius, was made fast by a pair of sister hooks. The variable radius was arranged to take buoys every 175 feet, that being half the length covered at each sweep. This length was selected in order that the eyes permanently wired to the cable to receive the buoys might also be the distance graduations, to avoid the possibility of error. The sweep was composed of a float or raft of cedar, in sections 20 feet long, and of a line of 2-inch gas pipes of the same length, depending from the float by wire cables. The float sections were strongly and flexibly connected, and the gas pipes were joined by toggles. The joints of the pipe were vertically below those of the float, so that the whole system consisted of a series of flexible parallelograms, each length of pipe being always parallel to the corresponding section of the float. Each of the suspending cables turned 90° over a pulley and was lashed to a cable running the entire length of the float, called the "messenger." By hauling on the "messenger" all the suspending wires were lifted equally and simultaneously, or in other words the line of gas pipe was lifted parallel to its first position, but higher in the water. Two sweeps were used, having 9 and 10 sections, or 180 and 200 feet length. The tug was placed between them, the shorter one upstream, and having the radius cable attached to its upper end. The axes of the sweeps were parallel with and that of the tug athwart the current. Guy lines to bow and stern of the tug kept the system in the desired position. The space under the boat was filled by a length of pipe dropped over the bow and hanging from the gunwales, and which connected the two sweeps, making a line of pipe 390 feet long up and down stream and 21 feet below the low-water plane. At each swarth the radius cable was lengthened 350 feet, so that there was a lap of 40 feet to insure against gaps. By working the engine ahead or backward the entire system was moved across the channel, running parallel to itself and following the arc of a curve determined by the radius. The indicating device was simple and very efficient. At every second suspending cable a staff was placed, submerged about 4 feet and attached at its lower end by a spring-clip to the suspending cable. It was pivoted on the float in the plane of the cable and extended 6 feet above the water with a flag at the top. It thus prolonged and made visible the direction of the cable extending from the float to the pipe. Itis plain that, if in moving across the channel the pipe met any obstruction, it would be held fast while the float moved on, so that the suspending wire, and consequently the staffs, would be inclined in the direction of motion. The effect was very pronounced, the "bowing" of the staff's being plainly and instantly visible. The boat was stopped in such cases and the messenger hauled in until the staff's resumed the vertical position, which they did suddenly and with a movement not to be mistaken. At that moment the messenger was stopped and the position of a zero point read on a scale which gave directly the depth of the pipes below the datum plane. That depth was recorded as the least depth on the shoal. At the same time a buoy was dropped on the highest point of the shoal. At first the buoy was located by transit intersections and quite independently of the sweeping apparatus. Observation of the accuracy with which the striking of known shoals could be predicted inspired such confidence in the sweep as a position indicator that one transit cut was abandoned, and locations were made by the arc described by the point of the sweep where the shoal struck and one transit observation. Under the latter method the transit station was always chosen so as to rake the channel, thus making the lateral, or most important coordinate, depend wholly on the transit. To check against any error from the dragging of the scow ancnor during the sweeping a tell-tale buoy was anchored alongside the scow, which showed any movement of the latter by casual observation. Preparations were begun early in July, and the party reached the point of beginning work at Sister Island on the 21st. After many vexatious delays, due to storms, discourtesy of captains of vessels, the novelty of the undertaking, and the incompetence of the crew of the chartered tug the work was closed on September 19 at the head of Brockville Narrows, 94 miles from the point of beginning. In this distance 14 new shoals were discovered, the positions of which were reported immediately after the close of fieldwork. Tri-daily observations were made at Charlotte and at Oswego, N. Y., on Lake Ontario, from. July 1, 1893, to June 30, 1894; at Erie Harbor, Pa., Ashtabula and Cleveland, Ohio, and Monroe, Mich., on Lake Erie; at Milwaukee, Wis., on Lake Michigan; and at Escanaba, Mich., on Green Bay, from July 1 to December 16, 1893, and from March 19 to June 30, 1894. Daily observations were made at Sand Beach, Mich., on Lake Huron, and at Sault Ste. Marie and Marquette, Mich., on Lake Superior, from July 1, 1893, to June 30, 1894. The accompanying table is a continuation of that published in the Annual Report of the Chief of Engineers for 1893, Part VI., p. 4381 : Monthly mean of water levels for the several stations below the planes of reference adopted Stations. 1893. in 1876. 1894. 3.31 Charlotto July. Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May. June. Fect. Feet. Feet. Feet. Feet. Fect. Feet, Feet, Feet. Feet. Feet. Fect. 3.73 3.85 3.54 3.30 3.06 3.01 2.83 2.30 3.01 3. 20 3.21 2.96 2.53 2.01 3.31 3.14 3.25 3.23 2.93 2.62 2.12 3.36 2.96 2.57 2.26 3.55 3.51 3.06 3.11 3.07 2.56 2.20 1.93 3.63 3.87 4. 11 4. 18 3.50 3.64 3.93 4.41 4.48 4.47 4. 62 Sault Ste. Marie. 2.84 2.78 2.87 2.90 3.06 3.29 3.45 WATER LEVEL OF LAKE ERIE. REPORT OF LIEUT. COL. JARED A. SMITH, CORPS OF ENGINEERS, FOR THE FISCAL YEAR ENDED JUNE 30, 1894. UNITED STATES ENGINEER OFFICE, Cleveland, Ohio, July 9, 1894. GENERAL: I have the honor to forward herewith record of water levels on Lake Erie for the fiscal year ending June 30, 1894. The records were taken at the light-house, Monroe, Mich., and in the harbors at Cleveland and Ashtabula, Ohio, and Erie, Pa. * a In connection with the record of water levels, I forward a report of Mr. William T. Blunt, U. S. assistant engineer, upon the levels of Lake Erie during the storm of October 14, 1893; also copy of the map indicated in Mr. Blunt's report. Very respectfully, your obedient servant, Brig. Gen. THOMAS L. CASEY, JARED A. SMITH, Lieut. Col., Corps of Engineers. Chief of Engineers, U. S. A. WATER LEVEL OBSERVATIONS FOR LAKE ERIE FOR THE FISCAL YEAR ENDING JUNE 30, 1894. Monthly mean water levels for Monroe, Cleveland, Ashtabula, and Eric harbors, expressed in feet below the plane of reference adopted in 1876; that plane being the surface of high water of 1838 and 2.34 feet above the mean level, 1860 to 1875, inclusive. 1893. 1891. Harbors at July. Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May. June. Feet. Feet. Feet. Feet. Fect. Feet. Feet. Feet. Feet. Feet. Feet. Feet. 1.98 2.26 REPORT OF MR. WILLIAM T. BLUNT, ASSISTANT ENGINEER. CLEVELAND, OHIO, June 20, 1894. SIR: I have the honor to submit the following report upon the variations in the surface of Lake Erie during the westerly gale of October 14, 1893: PRELIMINARY AND GENERAL. The extent of Lake Erie may be divided into three well-defined basins: The west basin, west of the "Islands,” containing about 1,200 square miles, and having a comparatively flat bottom at 5 to 6 fathoms when away from the immediate vicinity of the shore. The main basin, between the "Islands" on the west and the narrows at Erie and Long Point on the east, containing about 6,700 square miles, and having a marked shelving bottom deepening gradually to 14 fathoms. The east basin, east of the narrows, containing about 2,100 square miles, and having a deep depression of 30 fathoms just east from Long Point Island. Between the main and east basins lies an extensive flat at 11 fathoms depth, with only a narrow cut of 12 fathoms near the American shore. The general axis of the lake lies east northeast and west southwest, while that of the west basin makes a decided turn to west by north. It is a well-known fact that a westerly wind lowers the water surface at the west end of the lake and raises it at the cast end, while an easterly wind has the opposite effect. The amount and extent of fall or rise varies with the force and extent of the wind. A fresh local breeze will often change the level locally, while not affecting it materially in the open. A continued, general, and strong wind will have a general effect on the surface curve of the lake, lowering it considerably at the end from which the wind blows and raising it somewhat less at the opposite end. The variations due to this cause are most marked at the extreme ends of the lake, notably at Toledo, Monroe, and Buffalo. At the mouth of Detroit River they are tempered by the continuous supply from that river. At Toledo the record in the past eight years show an extreme fall of 74 feet and an extreme rise of 5 feet. As my data are more complete for the west end and for westerly storms, this report will deal more fully with westerly gales and consequent fall at west end of lake than with the opposite. The variation in the shoal and inclosed west basin in a continued gale is much greater than in the main basin. A high westerly wind for several hours will lower the water in the west basin 2 feet, as shown by gauge at West Sister Island, which is well toward its center. This same wind will lower the water east of the islands only a few tenths. This change of surface, due to heavy winds, has been many times remarked, usually in a general way, but I have no knowledge of its ever having been discussed on the basis of definite data. It would seem that the questions involved would not only be of great interest from a scientific standpoint, but would be of vital interest to navigators as enabling them to correctly judge of depths and currents during a severe storm. My own observations at the west end of the lake for the past eight years have convinced me that the subject should receive more than passing notice, and it is the purpose of this report to show a reason for that belief. About once in each year, usually in April, a heavy northeast storm occurs which raises the water 5 feet at the west end of the lake, and also once in each year, usually in October, a heavy westerly gale lowers the water 7 to 73 feet. These two storms are almost certain to come and to be attended by great loss of property and life. Never until last fall have circumstances permitted me to examine personally or to investigate generally the conditions attending such storms. CONDITIONS OCTOBER 14, 1893. On the morning of this day, while the steamer Swansea was tied up without steam, cleaning boiler, the wind freshened from northwest and all indications were for the annual low water. As often happens in such cases, the day was full of drawbacks, so that the boat could not leave the pier until 4 o'clock in the afternoon, at which time the water in the river, 5 miles from its mouth, had receded to 74 feet below mean level. A trip of unusual interest was then made to the bay. In the river, flats were showing where a few days before we had found 8 feet of water. The banks of the Straight Channel, where maps show 6 feet depth around Presque Isle, were 2 feet out of water, and for 2 miles these banks showed above water perfectly straight as if on a canal. Darkness came on as we reached the bay so that my intention of photographing the view was frustrated. As we reached the main crib in the middle of the bay we found the large Breymann dredge aground in the 17-foot channel and a reflex current rushing back against the gale with such force that the steamer could not be tuned and had to remain there over night. By 9 o'clock the water had set back to within 3 feet of its normal level, notwithstanding the gale continued. It so happened that in many harbors we had inspectors at the time, but it also unfortunately happened that none of them took special measurements of the stage of water, though I obtained from them, with the help of others, a very fair general idea of it. The general level of the lake before and after the storm was 0.7 foot below mean level of 1860-75 as used for our datum plane. This general level must of course be used in discussing the effects of this storm. The variations from this level at different points, together with notes showing their reliability, are given below: |