encountered over large bodies of water. The measurement site was 

 about 35 ft. from the shore in 4.5 ft. of water. Reeds along the 

 shore absorbed incident waves. The wind fetch was about 200 ft. for 

 each run analyzed. Data were collected at three wind speeds. 



1. Calm to light winds to serve as a base condition 

 with which to compare other measurements. 



2. Light winds, 2 to 4 meters per second. 



3. Moderate winds, 7 meters per second o 



Unstable conditions existed, at least locally over the lake, for 

 all runs up to the heights at which temperatures were measured. Con- 

 ditions were most unstable for the calm case and became less unstable 

 with increasing wind speed. Such conditions are not unusual at the 

 time of the year (late October, early November) these measurements 

 were made. Highest air temperatures occur in September and October 

 and maximum insolation occurs in November. 



RESULTS 



1. Calm to light winds 



The calm condition was interrupted at irregular intervals by 

 gentle gusts with almost immediate small wave generation. Often ap- 

 preciable temperature fluctuations accompanied the gusts. Temperature 

 fluctuations in excess of 0.5 C were observed with the float-mounted 

 thermistor and occasionally (Fig. 3A) changes of 1 C occurred in less 

 than one second. 



2. Light winds (October 30, 1968) 



The mean wind speed during this run was 2.8 m/sec and the 

 water temperature at a depth of 3 cm. was 17.7 C. The float-mounted 

 thermistor was 2.5 cm. above the wave surface and the fixed thermistor 

 4.5 cm. above the highest wave crests and slightly behind the float- 

 mounted thermistor. The significant wave height was about 4 cm. 



The two most striking aspects observed were a.) the general 

 agreement in the shape and phase of the two temperature records (Fig. 

 3B, Fig. 3C) and b.) the large temperature fluctuations measured by 

 the float-mounted thermistor compared to those measured by the fixed 

 thermistor. The magnitude of the temperature fluctuations is sur- 

 prising. At one point there is a change of over 4 C in less than a 

 fifth of a second. Such fluctuations were not uncommon with the float- 

 mounted thermistor but were not recorded by the fixed thermistor. A 

 correlation between the temperature signals and the waves (Fig. 3D) 

 is not obvious. At times there is a wave-like variation in the tem- 

 perature signal such as at "A" and "B". But there are also sections 



379 



