I enjoy making things. Here are a selection of projects that I have worked on over the years.
Different data types have different sensitivities to different anomalies. For example, ambient noise surface wave data has good lateral resolution in shallow depth, while teleseismic waveform and receiver function data have good vertical resolution. By combining multiple data types within the framework of full-waveform inversion (adjoint tomography) and jointly optimizing the velocity model, we can obtain a more reliable and accurate velocity structure.
Receiver function (RF) is widely used for imaging crustal and uppermost mantle velocity structure based on horizontal layered assumption. However this traditional technique has limitations on resolving non-layered anomalies, such as the dipping Moho, the subduction zone, and plumes. Thus, we develop a innovative technique to invert RFs based on adjoint waveform tomography method for high-resolution seismic array imaging.
Seispy is a graphical interface Python module for receiver function (RF) calculation and post-processing in seismological research. Automated workflows of RF calculations facilitate processing large volumes of different types of seismic data. The graphical user interface enables an intuitive and straightforward evaluation of RF quality. All parameters about the pre-processing for RF estimation can be adjusted based on user preference. Water-level frequency-domain deconvolution and iterative time-domain deconvolution for RF estimation are available in Seispy. The current version of Seispy contains five main modules for the post-processing of RF such as H-k stacking, crustal anisotropic estimation, harmonic decomposition, 2D and 3D Common Conversion Point (CCP) stacking. The CCP stacking in the different application scenarios can be handled by a rich collection of modules, such as time-to-depth conversion, 2D or 3D CCP stacking, and adaptive station or bin selection for CCP stacking profiles in a dense seismic array or a linear seismic array. As a Python module, functions in the Seispy can be called easily in python scripts for other purposes. The modular design allows new functionality to be added in a collaborative development environment. Licensed under GPLv3, Seispy allow users and developers to freely use, change, share and distribute copies of the package.
