UNDERFLOW METHODS 337 



The water-table method, more than any other, gives information on 

 the important question of the increase in the rate of recharge with in- 

 creased use. In southern California the United States Geological Survey 

 has maintained a series of measurements of water levels in 100 to 200 

 wells for 15 years, with the record for one well extending back 27 years. 

 The records of these wells show that the depleted supplies in the ground- 

 water reservoirs of that highly developed region were to an encouraging 

 extent restored in recent wet seasons, the recharge in some places during 

 these wet seasons being greater than would have been physically possible 

 if the reservoirs had already been nearly full. It was found that the 

 annual fluctuations were much less in Salinas Valley than in Santa Clara 

 Valley. This fact, however, does not mean a smaller supply available for 

 recharge, but, on the contrary, means that the supply in Salinas Valley 

 is so abundant that the ground-water reservoir is in all seasons kept 

 nearly full. 



UNDERFLOW METHODS 



The underflow methods consist of estimating the rate at which water is 

 percolating through a selected cross-section. They involve three factors : 

 (1) the area of the cross-section through Avhich percolation is taking 

 place, (2) the velocity of the percolating waters, and (3) the porosity, 

 or, rather, perhaps, the specific yield, of the sediments. The greatest 

 uncertainty is doubtless in the last factor. 



The rate of movement of ground-water was formerly estimated from 

 pumping tests and later by the use of salt or dyes, the salt being detected 

 in the downstream wells by chemical tests. The more practical and re- 

 liable electrolytic method for measuring velocities was devised by Charles 

 S. Slichter^* in 1901. 



The underflow methods have a rather restricted use for making deter- 

 minations of recharge and safe yield. Except where the water percolates 

 along a definite course, such as is afforded by the alluvium of a stream 

 valley, the conditions are too complex for its successful application. 

 Even in stream valleys it does not generally give maximum yield. For 

 example, the deposit of coarse, clean sand and gravel that underlies the 

 Arkansas Valley for hundreds of miles of its course is more important 

 as a reservoir than as a channel for ground-water. If the water-table is 

 lowered throughout this valley by heavy pumping, there will be recharge 

 in every locality, by rainfall and by percolation from the river and from 

 the porous deposits underlying the uplands adjacent to the valley. This 



** C. S. Sllchter : Field measurements of the rate of movement of underground water. 

 U. S. Geol. Survey Water-supply Paper 140, 1905. 



