Geothermal Fluid–Rock Interaction and Its Impact on Physicomechanical Behavior: Evidence from In-Situ Sinkhole Exposure Experiments
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Geothermal fluids can modify the physico-mechanical properties of rockson engineering-relevant timescales; however, field-calibrated modeling studiesremain scarce. We conducted an in situ exposure experiment within a karstsinkhole to quantify time-dependent degradation and relate hydrochemistry tochanges in strength, with direct implications for risk management in geothermal–karst settings. At the outset, NX-diameter cores were assigned to fourgroups: one group was tested in the laboratory as fresh rock, and the remainingthree groups were emplaced within the sinkhole and recovered sequentially, onegroup every three months. Field measurements included pH, temperature, electricalconductivity (EC)/salinity, and major-ion chemistry. Geophysical surveysand borehole logs from the authors’ previous studies in the same area were compiledand jointly interpreted with these data. Physical, mechanical, and strengthtests were conducted at baseline (fresh laboratory group) and at each subsequentretrieval from the sinkhole. The resulting data were statistically analyzed andevaluated with Phase2; four independent numerical models (fresh, 3, 6, and 9months) were run under saturated travertine conditions. Results show systematic declines in UCS and Vp with exposure, elevated ECwith near-neutral to mildly acidic pH at thermal outlets, and numerical responsesthat reproduce the weakening trend: service-state displacements (at SRF = 1.00)increase around the cavity, and the critical SRF decreases monotonically from1.86 (fresh) to 1.57 (9 months), with localization preferentially along mappeddiscontinuities. These findings indicate that geothermal fluid–rock interactionaccelerates karstification, reduces rock-mass strength, and narrows the stabilitymargin. To strengthen causality and improve forecasting, future work shouldimplement seasonal sampling of major–minor ions/oxides (e.g., Ca2+, Mg2+,Na+, K+, HCO−3 /CO2−3 , SO2−4 , Cl−, SiO2, Fe–Mn), determine alkalinity/DICand pCO2, compute saturation indices (SIcalcite, SIdolomite), and couple seepagewith mechanical modeling to relate chemical aggressivity to time-varying UCS,E, φ, c, and SRF.