资源环境科技发展态势分析平台

Mechanisms by which hydrochemical changes occur after earthquakes are not well documented. We use the 2016-2017 central Italy seismic sequence, which caused notable hydrochemical transient variations in groundwater springs to address this topic, with special reference to effects on fractured carbonate aquifers. Hydrochemistry measured before and after the earthquakes at four springs at varying distances from the epicenters all showed immediate postmainshock peaks in trace element concentrations but little change in major elements. Most parameters returned to preearthquake values before the last events of the seismic sequence. The source of solutes, particularly trace elements, is longer residence time pore water stored in slow-moving fractures or abandoned karstic flow paths. These fluids were expelled into the main flow paths after an increase in pore pressure, hydraulic conductivity, and shaking from coseismic aquifer stress. The weak response to the later earthquakes is explained by progressive depletion of high solute fluids as earlier shocks flushed out the stored fluids in the fractures. Spring delta C-13(DIC) values closest to a deep magma source to the west became enriched relative to preearthquake values following the 24 August event. This enrichment indicates input from deeply sourced dissolved CO2 gas after dilation of specific fault conduits. Differences in carbon isotopic responses between springs are attributed to proximity to the deep gaseous CO2 source. Most of the transient chemical changes seen in the three fractured carbonate aquifers are attributed to local shaking and emptying of isolated pores and fractures and are not from rapid upward movement of deep fluids.

Plain Language Summary Understanding the mechanisms of chemical and fluid transport to groundwater springs after earthquakes is poorly understood. Recent earthquakes in the Apennine mountains of central Italy provide a unique example of pre and post earthquake changes to groundwater chemistry in different carbonate aquifers. Our data show that changes were temporary for about 4 months after the earthquakes and that distance from the earthquake gave different responses. We show that the cause of these temporary changes was caused by shaking of long residence time local water into the fast moving flow. Once this long residence time water was flushed from the system the chemistry returned to background concentrations. We found little evidence of precursor chemical changes and no change in arsenic, which had been reported elsewhere. We did find changes in carbon dioxide gas in some springs that indicated that where a deep source of this gas existed, that it could quickly be released to the springs. These findings have implications for water supply delivery immediately after earthquakes and show that local shaking is likely more important than fluids derived from deep sources.