*Wt jj>! 



m? • 



■jjfp j& . ?*S£p* 



1,.... - 



••I 



Figure 4. --Randomly placed standpipes at depths of 5, 10, 15, and 20 inches, experiment 1, Indian Creek, 



1958. 



When the four squares are considered in 

 a factorial analysis (table 6), there are sig- 

 nificant differences between depths, between 

 squares, and in interaction. The variance 

 ratio (F) is higher, however, for depths than 

 for the other sources of variation. Statisti- 

 cally significant interactions are due to normal 

 differences found between one point in a 

 streambed and another. 



Decrease in dissolved oxygen levels with 

 depth in the streambed is shown in figure 5, 

 where experiments 1, 2, and 3 are combined. 

 Because of this decrease, no deeper source 

 is indicated as contributing to high dissolved 

 oxygen content of intragravel water. This 

 further confirms that the primary source 

 of dissolved oxygen is the stream. 



Ground- Water Temperatures 



Ground-water temperatures were sampled at 

 the same points and usually at the same time 

 as dissolved oxygen (discussed on page 2). 

 Although not measured consecutively through- 

 out the year, ground-water temperatures at 

 various depths below the water table appeared 

 to be lower than stream temperatures in the 

 summer and higher in the winter (table 7). 

 This agrees with Benson's (1953) findings 

 for the Pigeon River, Mich., and with the 

 ground-water temperature regimen of Cabin 

 Creek in Southeastern Alaska, which was 

 investigated by Institute personnel from 1949 

 to 1952. 



