Thermo-Poro-Elastic effects as hidden drivers of gravity signals in volcanic systems

Gravity observations are widely used in volcanic monitoring to infer subsurface mass redistributions, commonly interpreted in terms of magma intrusion. However, gravity changes may also arise from thermo-poro-elastic (TPE) processes associated with temperature and pore-pressure variations in fluid-saturated reservoirs. Neglecting these effects can lead to ambiguous or misleading interpretations of gravity signals during volcanic unrest.

The recent development of TPE inclusion models allows us to describe the mechanical fields induced by fluid-saturated rock volumes undergoing pore-pressure and temperature variations. These sources can coexist with magmatic sources within volcanic systems and are typically located at shallower depths than the deep magmatic reservoir, which acts as the primary engine by releasing hot fluids. These exsolved fluids rise from depth and either accumulate in, or migrate through, overlying brittle rock volumes, which respond to thermal and pore-pressure perturbations and therefore act as secondary sources of deformation and gravity change. In this work, we consider a disk-shaped TPE inclusion, a geometry that has been successfully applied in previous studies to represent deformation fields that are predominantly radial and associated with axisymmetric sources.

The results show that gravity variations induced by a TPE inclusion depend strongly on the fluid phase. Both liquid water and gaseous fluids can produce the same significant ground uplift, but lead to different gravity residuals: negative for liquid water and minor but positive for gaseous fluids. In contrast, condensation or vaporization of a thin layer near the surface can generate large gravity changes without notable deformation. As a result, heating and pressurization of a TPE inclusion can mask or weaken the gravitational signature of magma ascent, complicating the interpretation of gravity data and highlighting the need to account for hydrothermal effects when estimating magma volumes during unrest.

Gravity data collected over the past decades at the Campi Flegrei caldera (Italy) provide an ideal test site for applying our model and offer intriguing insights into both past and current unrest phases, although our results are applicable to any volcanic system with an active hydrothermal system. These findings highlight the importance of incorporating TPE effects into gravity data interpretation and integrated volcano monitoring strategies. Accounting for them improves our ability to distinguish between magmatic and hydrothermal contributions, leading to more robust assessments of subsurface dynamics and volcanic hazards.