Earthquakes cannot be predicted because we haven't been able to answer the question, Why should a fault, which has been inactive, start sliding unstably? The study of friction, which goes back to the times of da Vinci and probably before, established that the force required to slide an object is proportional to its weight. More recently, sensitive studies of rock friction in the laboratory have shown that the force also depends on time and on sliding velocity. Although time and velocity do not have a large effect on friction force, nevertheless the effect is important because it has allowed theoreticians to begin to study topics like stability, the triggering of sliding, etc. not only of earthquakes but also rock faces in mining and excavation settings. One difficulty that arises in the interpretation of experimental results is the problem of scale. The size of samples is measured in centimeters, whereas natural faults are many kilometers in length and depth; and the disparity in the scales of time that are characteristic of natural and laboratory processes is as great. The analysis that is being carried out is designed to bridge the gap caused by the disparity in time and length scales. The approach is to develop a theoretical model that simulates the behavior observed experimentally. The non-dimensional parameters that characterize the process are identified using linear viscoelastic theory. These parameters are then applied to natural faulting situations.
