Oscillations 



Winds and barometric -pressure gradients disturb 

 the equilibrium of an enclosed body of water and when 

 violent, they exert impulses that set up oscillations, 

 or seiches, within that body. Strong winds on Lake 

 Erie are often, but not always, accompanied bysteep 

 east -west barometric gradients. When the storm cen- 

 ter is northeast of Buffalo, the isobars are normal to 

 the east -west axis of the lake and form a downward 

 pressure gradient from Toledo at the west end to Buf- 

 falo at the east end. Such a condition tends to aug- 

 ment the effects of the westerly wind on the lake 

 level at Buffalo. When the storm center is northwest 

 of Buffalo, however, the isobars are parallel to the 

 east -west axis of the lake, although the general wind 

 direction over the whole lake may be southwest. In 

 this situation, the barometric pressure is not a direct 

 factor in changing the level at Buffalo. As an ex- 

 ample, at 8:00 a.m. on November 21, 1928, a low 

 pressure area northwest of Buffalo caused no east -west 

 pressure gradient on the lake. At this instant the 

 wind at Buffalo was southwest 40 m. p. h. This storm 

 caused the lake level to rise 3. 1 feet in 5 hours, at 

 the eastern end of the lake. Comparison of the 

 Weather Bureau records of strong westerly winds and 

 their effects on the lake level at Buffalo, indicated 

 but little difference in the amount of "pile -up" when 

 an east -west pressure gradient does or does not exist. 

 The wind, therefore, seems to be the principal factor 

 in producing the impulses which set up the main east- 

 west oscillation in the lake. 



There are several seiche areas in Lake Erie, and 

 each has its own definite boundaries and period. Fur- 

 thermore, the impulses of one area are transferred to 

 adjacent areas, where they tend to amplify or dampen 

 the oscillations occurring. Thus the system is ex- 

 tremely complicated. This overlapping of the seiches 

 is readily apparent from the continuous graphs at any 

 of the several gauges on the lake. Krecker (1928) 

 has discussed the seiches of the western part of Lake 

 Erie, and has prepared graphs of local oscillations in 

 the vicinity of Put -in -Bay, Ohio. 



Henry (1900) has concluded that this primary os- 

 cillation is stationary rather than progressive, and has 

 prepared an excellent graph showing the wind effect 

 on the lake level. The period of the oscillation has 

 been discussed by Hayford (1922). As some addi- 

 tional information was obtained this season, a further 

 discussion seems advisable at this time. 



Figure 19 shows the lake -level graphs at three 

 stations and the wind at Buffalo for the period May 

 1-17, 1929. The full line, broken line, and the 

 dash -dot line are the respective level curves at Buf- 

 falo, New York, Port Stanley, Ontario, and Put -in - 

 Bay, Ohio. The left-hand scale is the lake level in 

 feet (0 equals 572. 8 feet above mean sea level). 

 The wind at Buffalo is shown by a dotted line, with 

 the scale in miles per hour at the right. Buffalo 

 is at the extreme eastern end of the lake; Put -in -Bay 

 is on South Bass Island, 30 miles from the western 

 limit of the lake; and Port Stanley is less than 10 

 miles east of the line midway between the east and 

 west limits of the lake. The curve for Buffalo is from 

 the U. S. Engineers' gauge; that for Port Stanley is 

 from hourly heights from the Department of Marine 

 and Fisheries, Canada; and the one for Put -in-Bay is 

 from hourly and half -hourly staff readings. The wind 

 curve is from the hourly readings of the U. S. Weather 

 Bureau. 



It may be seen that the lake -level curves are close 

 together until 2 a, m. May 16, when the level at Buf- 

 falo commenced to rise rapidly while the Put -in -Bay 

 level fell correspondingly under the influence of west- 

 erly winds of high velocity. The maximum level at 

 Buffalo and the minimum at Put -in -Bay were reached 

 between 11 a. m. and 12 noon on the 16th, and during 

 the following 6 hours the level fell 4. 9 feet at Buffalo 

 and rose 3. 5 feet at Put -in-Bay. From 10 a. m. May 16 

 until 6 a.m. May 17, there were no observations of 

 the level at Put-in-Bay; but it is readily seen that the 

 maxima at Buffalo occur approximately at the times of 

 the minima at Put-in-Bay, and vice versa. The figure 

 shows excellently the original impulse set up by a 

 westerly wind of high velocity and the resulting east- 

 west oscillation which continues even though the wind 

 subsides. 



The oscillations at Buffalo (full line) and Port 

 Stanley (broken line) from May 7 to 11 are shown in 

 figure 20. The lake level at Buffalo rose about 1 foot 

 under the influence of a west wind which reached a 

 maximum of 40 m. p.h. on May 7, 1929. From this 

 date to May 11, the wind was light and variable and 

 therefore did not greatly distort the period or ampli- 

 tude of the east -west oscillation occasioned by the 

 impulse of May 7. The study of 180 of these more or 

 less undisturbed oscillations at Buffalo showed the 

 mean period for one complete oscillation to be 14. 1 

 hours. Henry (1900) referred to this oscillation as 

 having a period of from 12 to 16 hours, and Hayford 



64 



