A semi-airborne transient electromagnetic data denoising method based on TOC-VMD and improved threshold

Abstract The semi-airborne transient electromagnetic method is commonly employed in geophysical exploration with complex geological conditions, so it is subject to noise interference from various sources. The multi-source noise in semi-airborne transient electromagnetic data frequently impacts the data quality and reduces the accuracy of subsequent inversion. To solve this problem, different types of noise are systematically classified by their source. Based on this classification, this study proposes a noise suppression method using the tornado optimizer with Coriolis force to optimize variational mode decomposition (TOC-VMD), combined with an improved threshold function, to suppress specific noise components. Firstly, the tornado optimizer with Coriolis force is employed to select the optimal combination of decomposition parameters, achieving adaptive decomposition of the noisy data. Subsequently, an improved Bhattacharyya distance is used to distinguish the effective signal modes and noise modes for signal reconstruction. Finally, an improved threshold function is applied to suppress residual noise. The simulation results show that the proposed method achieves superior noise suppression for multi-source noise compared to the conventional signal decomposition-based method. In the field experiment, this method reduces noise interference, enhancing data quality. The effectiveness of this method has been verified in field applications, demonstrating broad application prospects in the denoising of semi-airborne transient electromagnetic signals.