28 EFFECTS OF WINDS AND OF 



CHANGE OF ELEVATION USED IN OBSERVATION EQUATIONS. 



Each observed daily elevation is the mean of 24 hourly elevations obtained 

 from the gage record. 



The inflow to Lake Michigan-Huron, through the St. Marys River, from 

 Lake Superior was determined each day by the U. S. Lake Survey by means 

 of a recording gage giving the elevation of the water surface at Sault Ste. 

 Marie above the rapids. Similarly, the outflow from Lake Michigan- 

 Huron to Lake Erie was determined for each day from the recorded elevation 

 of the water surface at Fort Gratiot and at St. Clair Flats in the St. Clair 

 River. In each of these cases the recorded elevations of water surface were 

 converted to volume of stream-flow by means of the known relations between 

 these quantities which had been established by river gagings made by the 

 U. S. Lake Survey. As these corrections for inflow and outflow are small 

 and are abundantly accurate for the purpose, the explanation in detail of 

 their computation is not here given. 



No correction was applied for run-off into Lake Michigan-Huron from the 

 surrounding land-drainage area. This quantity was very difficult to deter- 

 mine at the time this investigation was made. So, too, no correction is 

 applied for evaporation, for a similar reason. 



The amount of rise of water surface of the lake on each day, produced by 

 rainfall on the lake surface, was computed from a number of rain gages 

 operated by the U. S. Weather Bureau and the Weather Service of Canada. 



For convenience the three corrections for inflow, for outflow, and for 

 rainfall on the lake surface were grouped together in the computation as 

 shown in the third column of table No. 4. These three corrections are 

 given separately in the last three columns of the table. 



Note that the difference between outflow and inflow, or what might be 

 called net outflow, is no more than 0.008 foot on any day in September 1910. 

 Note that the maximum rise or fall during any day of the month due to 

 combined inflow, outflow, and rainfall on the lake surface was only 0.033 

 foot. It is clear, after comparing these values with the observed rise of the 

 lake surface at Milwaukee for each day, as shown in table No. 4, that the 

 observed rise is evidently due to causes other than these three, which are 

 minor in their effect upon the daily fluctuations of level as compared with 

 other causes which are operating. 



The corrections for wind effects as shown, at a maximum only 0.001 foot 

 for September 24 and September 25, were computed by methods shown in 

 detail later in this publication. 



In all columns except the second the change shown is for that from the 

 preceding day to the day indicated in the first column on that line. For 

 example, 0.180 foot is the rise of water surface from September 5 to Septem- 

 ber 6, namely, 580.07-580.25. Similarly, -0.033 foot is the correction for 

 rainfall, inflow, and outflow combined from September 5 to September 6. 



The observation equations were arranged in groups of one month each. 



