Measurement of Tilt in a Deep‐Level Gold Mine and its Relationship to Mining and Seismicity

Abstract

A mercury tube tiltmeter was installed in a small chamber at a depth of 3.2 km near the lowest level of the East Rand Proprietary Mines gold mine, South Africa; the site was near an advancing mine face. Changes in tilt were recorded for eight months during which a total change of 2.7 × 10−3 radians was measured. An array of seismometers was operated simultaneously in this general region of the mine; this allowed a unique study of rock deformation associated with seismic events within a radius of 200 m from the tiltmeter. Three types of tilt were recorded: (1) continuous tilt change at a rate of about 10−5 radians/day due to the enlargement of the mine excavation; (2) step changes in tilt caused by mine tremors with magnitudes ranging from about ‐ 1.5 to 3.75 at distances from 50 m to 2 km from the tiltmeter; the step tilts varied from 10−8 to about 3 × radians, were in either direction, and were often followed by unusually high rates of continuous tilting; (3) episodes of anomalous tilts which are currently unexplained. The time dependence of tilting can be explained, to a large extent, as logarithmic creep in response to a rapid change in the state of stress, caused either by a tremor or by advancing the mine face. The good correlation between the rate of tilt change and the rate of occurrence of tremors suggests that the probability of a tremor in a given period is proportional to the rate of strain change in the rock. The size of the step change in tilt for a tremor is approximately proportional to its seismic moment. Most of the deformation in the rock surrounding the mine excavation is due to large tremors and not to more continuous modes of strain change. On a time scale of a few minutes to a day tremors occurring near the tiltmeter are not preceded by any unusual tilts as large as 10−6 radians in amplitude. Copyright © 1975, Wiley Blackwell. All rights reserved