Measurement of ground-water storage change and specific yield using the temporal-gravity method near Rillito Creek, Tucson, Arizona

The temporal-gravity method was used to estimate ground-water storage change and specific -yield values at wells near Rillito Creek, Tucson, Arizona, between early December 1992 and early January 1994. The method applies Newton's Law of Gravitation to measure changes in the local gravitational field of the Earth that are caused by changes in the mass and volume of ground water. Gravity at 50 stations in a 6-square-mile area was measured repeatedly relative to gravity at two bedrock stations. Ephemeral recharge through streamflow infiltration during the winter of 1992-93 resulted in water-level rises and gravity increases near Rillito Creek as the volume of ground water in storage increased. Water levels in wells rose as much as 30 feet, and gravity increased as much as 90 microgals. Water levels declined and gravity decreased near the stream after the last major winter flow but continued to rise and increase, respectively, in downgradient areas.

Water levels and gravity relative to bedrock were measured at 10 wells. Good linear correlations between water levels and gravity values at five wells nearest the stream allowed for the estimation of specific-yield values for corresponding stratigraphic units assuming the mass change occurred in an infinite horizonal slab of uniform thickness. Specific-yield values for the stream-channel deposits at three wells ranged from 0.15 to 0.34, and correlation coefficients ranged from 0.81 to 0.99. Specific-yield values for the Fort Lowell Formation at three wells ranged from 0.07 to 0.18, and correlation coefficients ranged from 0.82 to 0.93. Specific-yield values were not calculated for the five wells farthest from the stream because of insufficient water-level and gravity change or poor correlations between water level and gravity. Poor correlations between water levels and gravity resulted from ground-water storage change in perched aquifers and in the unsaturated zone near ephemeral streams.

Seasonal distributions of ground-water storage change since early December 1992 were evaluated from gravity change at all stations using Gauss's Law. Changes in the distribution of gravity are caused by the flow of water into or out of ground-water storage. Gravity along two profiles was measured frequently to evaluate spatial and temporal distributions of gravity change.

Gravity variations indicated preferential ground-water flow to the south in the western part of the study area where the saturate thickness of the aquifer is greatest. Storage changes from December 1992 through early March 1993, mid-May 1993, late August 1993, and early January 1994 were calculated as increases of 7,900, 8,000, 6,300, and 3,700 acre-feet, respectively. Seasonal variations in storage were caused by ground-water withdrawlas, ground-water flow across the boundaries of the gravity-station network, and streamflow infiltration from December 1992 through late April 1993. Most of the estimated recharge of 10,900 acre-feet occurred before mid-May 1993.