E3S Web of Conferences (Jan 2020)

Hot water injection in relation to 1982-84 microseismic events at Campi Flegrei Caldera by thermo-hydro-mechanical simulation

  • Akande Waheed Gbenga,
  • Gan Quan,
  • De Siena Luca,
  • Cornwell David G.

DOI
https://doi.org/10.1051/e3sconf/202020502008
Journal volume & issue
Vol. 205
p. 02008

Abstract

Read online

The repeated deformations and seismic unrests at Campi Flegrei caldera (southern Italy) have been identified due to a fluid injection source in volcano active area. We investigated the role of hot water injection in the seismic unrests of 1982-1984 with a view to gaining insights into the caldera’s dynamics, by applying coupled TOUGHREACT-FLAC3D simulator to our computational domain of 10 × 1 × 3 km with a single-phase steam isothermal (HM) and non-isothermal (THM) simulations comparison. The results indicated that the overlying caprock blocks the uprising hot water injection, leading to the building up of pore pressure and shear stress underneath over time. This process substantially modified the ambient stress regime, strained the caprock, led to microfractures formation which enhanced hydraulic properties of the system, and ultimately provided conductive migration pathways for hot fluids to pressurize the shallow overlying fault. The prolonged shear stress accumulation period culminated in shear fracturing, dilation and fault slips, whose displacements are in order of millimetres to few centimetres. The moment magnitudes of seismicity generated were generally in order of 3.0. The seismic slips occurred at ~7 and ~6 days under HM and THM cases, respectively – giving a time delay in the order of ~1 day in the isothermal case. A comparison of our results for isothermal and non-isothermal cases suggests that seismic slips occurred generally earlier at higher fluid injection rates and under non-isothermal conditions. Further examinations of thermal effects showed that fault slips generally occur earlier at higher injection temperatures and/or at high host rock/fluid injection temperature contrasts. At very low permeabilities (e.g., 10-19 m2), fault slips were delayed and the required time for shear stress accumulation to trigger slips is about 8 days. It is concluded that an interplay among variables such as injection rates, fluid injection temperatures and caprock permeabilities determines the timing of a seismic fault slip and associated seismicity at the caldera.