Existing deep learning-based seismic signal denoising methods primarily operate in the time domain. Those methods are ineffective when noise overlaps with the seismic signal in the time domain. Time–frequency domain-based deep learning methods are relatively rare and usually employ single loss function, resulting in suboptimal performance on low SNR signals and potential damage to P wave. This paper proposes a method based on generative adversarial networks (GANs). Compared to convolutional neural networks, the discriminator in GANs helps retain more true signal details by judging denoising performance. Additionally, an attention mechanism is introduced to fully extract signal features, and a perceptual loss is employed to evaluate the difference between the denoised result and the target’s high-level features. Experimental results show that this method can effectively improve SNR and ensure that the denoised result is close to the true signal. Furthermore, by comparing DeepDenoiser and ARDU, it is proven that the proposed method achieves better denoising performance, especially for low SNR signals, while causing less damage to the seismic signals.
Located mainly on thick lacustrine sediment, pumping ground water has led to pronounced subsidence in Mexico City, with a rate of nearly 0.35m per year in the worst affected locations. The load of new infrastructure exacerbates the problem. Monitoring the subsidence is crucial for understanding the mechanism and hazard prevention. While InSAR and other methods have provided highly resolved subsidence maps, they lack depth resolution.Seismic velocity, which reflects the state of the sediment (e.g. its stress, pore pressure, and other properties), will change with the compaction during the subsidence. Taking advantage of the omnipresent urban ambient noise recorded by the 30 broad band seismic stations in periods between 2010 and2021 around Mexico City, we calculate the cross correlations between different station pairs to determine the seismic velocity variations with coda wave interferometry in different frequency bands. We aim to obtain the spatial and depth distribution of seismic velocity variations using coda wave sensitivities and surface wave dispersion Previous studies have explained the long durations of ground-motion that has been observed during teleseismic and local earthquakes by higher-mode surface wave propagation in the competent rocks underneath the Mexico Basin, which has important implications for the depth sensitivity of any observed velocity changes.To determine the precise relationship between the subsidence and seismic velocity variations, we will further analysis the tectonic and environment effects. Our study aims to provide a more comprehensive understanding of the subsidence in Mexico City.