The temperature response of water-saturated rocks to stress changes is critical for understanding thermal anomalies in the crust, because most porous rocks are saturated with groundwater. In this study, we establish a theoretical basis of the adiabatic pressure derivative of the temperature of water-saturated rocks under both undrained (βwet_U) and drained (βwet_D) conditions. The value of βwet_U is linearly correlated with Skempton’s coefficient (B) and βwet_D increases nonlinearly as the pore water volume per unit volume of rock (ξ) increases. The theoretical calculations demonstrate that the thermal effects of pore water predominate in water-saturated rocks with medium to high porosity, especially under undrained conditions. In most cases, the temperature response of rocks with a porosity of ϕ > 0.05 under water-saturated and undrained conditions is greater than that under dry conditions. Experiments were also carried out on a water-saturated typical medium porosity sandstone (sample RJS, ϕ = 0.102) and on a compact limestone (sample L27, ϕ = 0.003) using an improved hydrostatic compression system. The experimental results confirm that the theoretical derivation is correct, and the calculated ranges of βwet_U and βwet_D are reliable for all 15 rocks. Consequently, this study increases our understanding of the thermal anomalies that occur after huge earthquakes, including the negative thermal anomalies, which are probably induced by co-seismic stress release, that were observed in the boreholes that penetrate seismic faults after the Chi-Chi Earthquake, the Wenchuan Earthquake, and the Tohoku Earthquake.

Depth profiles of sediment thermal conductivity are required for understanding the thermal structure in active seismogenic zones. During the Nankai Trough Seismogenic Zone Experiment (NanTroSEIZE), a scientific drilling project of the International Ocean Discovery Program, a borehole penetrated to a depth of 3262.5 meters below seafloor (mbsf) at site C0002. Because core samples obtained from below ~1100 mbsf in an accretionary prism are limited, a thermal conductivity profile over such depths usually determined by laboratory measurements using core samples is not available. To obtain the thermal conductivity profile at site C0002, we used core samples collected from sediments that overlay the in-coming subducting oceanic basement at NanTroSEIZE site C0012, which can be considered to have the same mineral composition as the accretional prism at site C0002. The thermal conductivity of the C0012 core samples was measured at high pressure to simulate subduction by reducing the sample porosity. We measured the thermal conductivity of six core samples from 144–518 mbsf at site C0012 up to a maximum effective pressure of ~50 MPa, corresponding to depths greater than ~4 kmbsf. We obtained an empirical relation between thermal conductivity and fractional porosity for the Nankai Trough accretionary prism as = exp(-1.09φ+0.977). Based on porosity data measured using core/cuttings samples and data derived from P-wave velocity logs, we estimate two consistent and complete thermal conductivity profiles down to ~3 kmbsf in the Nankai Trough accretionary prism. These profiles are consistent with the existing thermal conductivity data measured using limited core samples.