As a multivariate statistical method, the Principal component analysis has been applied to many research fields. Recently, a seismological study successfully introduced the Principal component analysis into the teleseismic data, through which the three-dimensional crustal structure beneath the seismic station was derived.
The paper entitled 'Crustal structure study based on principal component analysis of receiver functions' was written by Ling Chen as the corresponding author of the new study, a researcher at the Institute of Geology and Geophysics, Chinese Academy of Sciences. The researchers used the principal component analysis to study the characteristics of teleseismic receiver function waveforms varying with back-azimuth, as a new method which has been proved be able to effectively constrain the crustal structure beneath the targeted station. The new method was further applied to the receiver function data from a station in the central part of the Sichuan Basin to constrain the crustal structure beneath the station.
The teleseismic receiver function is a time series containing the Ps conversions of the crust-upper mantle discontinuities and their multiples beneath the station. It has become an important observational data for studying the crustal structure. For a single station, the average one-dimensional crustal structure is usually derived by stacking the radial receiver functions from all back-azimuths, whereas structural variations (such as dipping discontinuities or anisotropy) can be constrained through analysis of waveform dependence on the back-azimuth of both the radial and tangential receiver functions. However, due to the common presence of noise in real data and the weak signals generated by the dipping discontinuity or anisotropy just being considered as a disturbance superimposed on the primary feature reflecting the velocity difference at the discontinuity, it is often difficult to study dipping discontinuities and/or anisotropic structures by using the original signals. Therefore, how to suppress the noise, improve the signal-to-noise ratio, and effectively separate the signals reflecting different structures has been a frontier issue in studies of the crustal structures.
In this study, the researchers applied principal component analysis to receiver function data, summarized the structural features reflected by individual principal components for different models based on a series of theoretical tests, and finally selected the appropriate principal components to reconstruct new receiver functions. The results showed that: (1) this method can effectively suppress the noise and improve the signal-to-noise ratio of the receiver functions. (2) It can effectively separate the receiver function signals that reflect structural variations from those representing the primary feature. Wherein the first principal component of the radial receiver functions contains the average structural information of the crust beneath the station, and the second principal component of the radial receiver functions and the first principal component of the tangential receiver functions both reflect the variation information of the crustal structure. (3) The reconstructed results for the second and third principal components of the radial receiver functions can effectively distinguish the isotropic dipping structure and anisotropy with a dipping symmetry axis (Figure 1).
The method not only significantly improves the signal-to-noise ratio of the receiver function data, but also makes up for the inadequacy of the traditional method which cannot effectively separate the signals that represent different structural information. In addition, the method provides a new idea to distinguish the isotropic dipping structure and anisotropic structure with a dipping symmetric axis. It is of vital significance in the scientific studies of the crustal structure.
This research was funded by the National Natural Science Foundation of China (No. 4188103) and the independent project of the State Key Laboratory of the Lithospheric Evolution, IGGCAS (No. SKL-Z201704-11712180).
See the article:
Zhang J Y, Chen L, Wang X. 2019. Crustal structure study based on principal component analysis of receiver functions. Science China Earth
Sciences, 62, https:/