Geochemical precursors to seismic activity

Studies of earthquake precursory phenomena during the last several decades have found that significant geophysical and geochemical changes can occur prior to intermediate and large earthquakes. Among the more intensely investigated geochemical phenomena have been: (1) changes in the concentrations of dissolved ions and gases in groundwaters and (2) variations in the concentrations of crustal and mantle volatiles in ground gases. The concentration changes have typically showed no consistent trend (either increasing or decreasing), and the spatial and temporal distribution of the observed anomalies have been highly variable. As a result, there is little agreement on the physical or chemical processes responsible for the observed anomalies. Mechanisms proposed to account for precursory groundwater anomalies include ultrasonic vibration, pressure sensitive solubility, pore volume collapse, fracture induced increases in reactive surfaces, and aquifer breaching/fluid mixing. Precursory changes in soil gas composition have been suggested to result from pore volume collapse, micro-fracture induced exposure of fresh reactive silicate surfaces, and breaching of buried gas-rich horizons. An analysis of the available field and laboratory data suggests that the aquifer breaching/fluid mixing (AB/FM) model can best account for many of the reported changes in temperature, dissolved ion and dissolved gas concentrations in groundwater. Ultrasonic vibration and pressure sensitive solubility models cannot reasonably account for the geochemical variations observed and, although the pore collapse model could explain some of the observed chemical changes in groundwater and ground gas, uncertainties remain regarding its ability to generate anomalies of the magnitude observed. Other geochemical anomalies, in particular those associated with hydrogen and radon, seem best accounted for by increases in reactive surface areas (IRSA model) that may accompany precursory deformation around the epicenter of an impending earthquake. Analysis of the probable response of these models to the earthquake preparation process, as well as to other environmental factors, suggests that geochemical monitoring programs can provide information that may be valuable in forecasting the probability of an earthquake; however, because of the complexity of the earthquake preparation process, the absolute prediction of seismic events using geochemical methods alone, does not presently appear to be feasible