Methodological Evaluation of Coda Wave Interferometry for Concrete Monitoring: Numerical Assessment of Mechanical Load, Temperature, and Moisture Coupling

Coda Wave Interferometry (CWI) is a highly sensitive ultrasonic technique capable of detecting subtle changes in heterogeneous materials such as concrete. Due to multiple scattering, the coda portion of the ultrasonic wavefield is particularly responsive to small variations in elastic properties, which may arise from mechanical loading or environmental effects. In concrete structures, changes in stress, temperature, and moisture content are known to influence ultrasonic velocities in a coupled manner, making the interpretation of monitoring data challenging. In this study, the sensitivity of CWI to mechanical load, temperature, and moisture variations in concrete is investigated through numerical simulations. Realistic numerical models of concrete are employed to simulate ultrasonic wave propagation under controlled perturbations representing mechanical loading and environmentally induced changes in material properties. Relative velocity variations, quantified as relative velocity change (dv/v), are extracted using CWI and analyzed to assess the individual and combined influence of the considered parameters. Specifically, model-based findings indicate that within the investigated numerical configuration: (i) relative humidity variations between 35% and 95%, modeled via quality factor Q changes alone, induce negligible relative velocity shifts (<0.08%) while increasing waveform decorrelation; (ii) temperature variations between 0 °C and 40 °C produce a linear dv/v response of approximately −0.0188%/K via mortar matrix softening; and (iii) uniaxial tensile damage exhibits a strongly nonlinear, regime-dependent dv/v signature characterized by a steep drop during diffuse microcracking (D<0.08) followed by a flatter response in the post-peak softening regime (D>0.2). Experimentally validated trends reported in the literature are used as reference to guide model calibration and interpretation, while all analyses are performed in a fully numerical framework. The results demonstrate that CWI captures distinct responses associated with load-induced and environmentally induced changes, and they highlight the potential and limitations of numerical modeling for disentangling mechanical and environmental effects in concrete monitoring applications under the specified assumptions.