Effects of Induced Streambed Infiltration on Water Levels in Well During Aquifer Tests

Abstract

The aquifer test, involving a production well and several observation wells near a stream, is one of the effects of induced streambed infiltration on water levels in aquifers. In presently available analytical treatments of flow problems associated with induced infiltration, recharge by the influent seepage of surface water is commonly simulated by use of a hypothetical image well. The assumption is made that water levels in the aquifer will behave the same whether recharge occurs over an area or through an image well. Electric analog computers, in which the streambed is simulated as an area of recharge in accordance with natural conditions instead of as a recharging image well, were used to appraise the accuracy of estimated effects of induced infiltration on water levels based on the image well theory. Electric analog computers for two aquifer test sites for which field data are available were constructed. The selected aquifer tests involve the two extreme aquifer-stream situations, i.e. the case where the cone of depression spreads beneath and beyond the entire streambed and the case where the cone of depression spreads only part way beneath the streambed. The analog computers consist of analog models and excitation-response apparatus. The analog models are regular arrays of resistors and capacitors and are scaled-down versions of aquifer-stream situations. The excitation-response apparatus consists of a power supply, waveform generator, pulse generator and an oscilloscope. It is concluded that, during induced infiltration aquifer tests, the image-well theory closely describes drawdowns on the land sides of streams with a high degree of accuracy whether the cone of depression spreads beneath and beyond or only part way beneath the streambed. Drawdowns beneath or beyond the streambed and the streambed areas of infiltration based on the image-well theory are not those which are observed in the field. However, the streambed infiltration rates per foot of head loss based on hypothetical drawdowns beneath streambeds and streambed areas of infiltration computed with the image-well theory seem to be empirically correct.