In the past two decades, surface wave imaging based on seismic ambient noise cross-correlation (CC) has been one of the most important technologies in the field of seismology. With the development of this technology, high-mode surface waves have received increasing attention, especially after the proposition of the frequency-Bessel transform (F-J) method, which can effectively extract multimode dispersion curves from ambient noise data. In the past few years, our research group has made many attempts to improve this method. We summarized these experiences and the corresponding algorithm for fast CC, and packaged them into a Python package called CC-FJpy. It is commonly understood that CC takes a good deal of time. However, we found that a simple reorganization of the CC logic can achieve computational acceleration by a multiple of tens or even hundreds in comparison with classical CC open-source programs for N stations. For the F-J method, we use Nvidia’s graphics processing unit (GPU) to speed up computation, and this approach achieves a hundreds-fold computational acceleration. We have encapsulated our experiences and technologies into CC-FJpy and submitted it to various types of data tests to ensure its speed and ease of use. We hope that providing the open source of CC-FJpy can benefit the development of surface wave studies and make it easier to start with high-mode surface waves. We look forward to your use and valuable suggestions.
