<p>The attenuation of amplitude, phase, and frequency caused by stratigraphic absorption can be denoted by quality factor (Q). To address the sensitivity of Q estimation to waveform coupling in thin layers, we propose Q estimation based on multi-trace and corresponding travel time spectral ratio (M-CTSR) method. The spectral ratio (SR) method estimates Q values by quantifying the amplitude attenuation at different frequency bands. Given this, the amplitude-equalized data obtained from sub-spectrum balancing serve as the reference data. Using shaping regularization, we then estimate the effective Q at corresponding travel time, thereby reducing the influence of reflectivity on the amplitude spectrum. The instability of Q value fitting algorithm introduces wild amplitude noise. To quantify the resultant error, the discrepancy between the amplitude curve derived from spectrum division and its linear fitting line is computed. This yields an uncertainty matrix (or vector), which serves as an error parameter for assessing the accuracy of the estimated Q values. With the error parameter and the predefined normal range of Q values, anomalous Q values are identified. Then the estimation of Q for multiple traces, which is ultimately aimed at constructing Q-field, is formulated as irregular-sampled seismic data reconstruction. This is carried out by first modifying the sampling matrix to remove anomalous Q values, then applying spatial smoothing, which ultimately yields the Q‑field of the post‑stack seismic data. The synthetic models and the field data tests prove that the algorithm can, to some extent, reduce the errors of Q estimation caused by waveform coupling.</p>