Objective A frequency-domain induced polarization (FDIP) method enables the observation of critical electrical parameters such as frequency dispersivity and complex resistivity of geological bodies, mitigating the multiplicity of solutions in the interpretations of electrical anomalies. This method has emerged as a primary technique for electric method-based advance water detection during roadway tunneling in coal mines. However, the FDIP method primarily focuses on data observation in the direction opposite to roadway tunneling, exhibiting a limited capacity to capture geoelectric information in the roadway tunneling direction. This leads to practical challenges such as unclear orientations of water-bearing anomalous bodies. Therefore, exploring the advance response characteristics and anisotropy of the FDIP parameters of roadways holds critical theoretical and practical significance for improving data observation methods and further enhancing the identification accuracy of water-bearing anomalous bodies.

Methods First, a method for observing triaxial apparent IP parameters was developed based on the practical conditions of a coal mine roadway. Second, the response expressions of the triaxial parameters were derived by using an infinitely large tabular body in the whole space as the model of water-bearing bodies in the roadway tunneling direction. Third, the variations in the triaxial parameters with the model's attitude like azimuth, dip angle, and distance from the field source were analyzed using numerical calculations and theoretical analysis.

Results The results indicate that the curves of apparent frequency dispersivity and complex resistivity along the axial direction of the roadway exhibited K-type (low-high-low) and H-type (high-low-high) patterns, respectively, independent of changes in the model parameter, with electrical anomalies consistently demonstrating low resistivity and high dispersivity. The curves of apparent frequency dispersivity perpendicular to both sides of the roadway presented a K-type pattern when the model was set directly ahead of the roadway, and they appeared to be inversely proportional functions in other cases. The apparent complex resistivity curves displayed K- and H-type patterns when the model was arranged near the left and right sides of the roadway, respectively. The curves of apparent frequency dispersivity perpendicular to the roof and floor of the roadway exhibited a K-type pattern in the case where the model was upright, and they appeared to be inversely proportional functions in other cases. The apparent complex resistivity curves presented K- and H-type patterns when the model was inclined towards the front and rear of the roadway, respectively. The anomaly amplitude and detection ranges of the triaxial apparent IP parameters were significantly influenced by the model parameters. Notably, the detection ranges at extreme or step points changed significantly with the distance from the field source.

Conclusions The triaxial apparent IP parameters demonstrated pronounced anisotropic responses to the tabular water-bearing anomalous body model. The apparent IP parameters in the axial direction of the roadway manifested a relatively low sensitivity to the model's attitude, proving to be a primary cause of low detection of electrical anomaly accuracy. In contrast, the apparent IP parameters perpendicular to both sides of the roadway were sensitive to the model's azimuth, while those perpendicular to the roadway roof and floor are susceptible to the dip direction of the model. Compared to current observation methods, the triaxial observation method provides richer electrical information for detecting water-bearing anomalous bodies in the roadway tunneling direction, thereby enhancing the spatial positioning accuracy of water-bearing anomalous bodies.