Regularizing the 3-D teleseismic wavefield for receiver function imaging using a radial basis function

Abstract

SUMMARY Receiver function is important for imaging crustal and upper-mantle discontinuities. However, sparsely scattered stations could introduce imaging artefacts or misinterpretations for complex structures. We regularize a 3-D teleseismic wavefield to reduce these artefacts using radial basis function interpolation. First, we evaluated the feasibility of wavefield regularization with several typical models using synthetic data. The results demonstrate the high reliability of our method for recovering local 2-D and 3-D structures, even when seismic stations intentionally miss 95 per cent of a uniform fine grid. Then, we applied this method to sparsely deployed stations in Northeast China. The waveforms reconstructed from surrounding stations show good consistency with the observed waveforms; furthermore, the imaging results using the regularized data are highly comparable with the reference results obtained by using a dense 2-D seismic array of 60 stations (with a spacing of 10–17 km), even though our input data were mainly contributed by only 9 stations (with an average spacing of ∼80 km). Our results show better continuity of the 3-D topography of the subsurface compared with that obtained by the traditional method. Our regularization method could significantly improve the spatial resolution of receiver function imaging for both sparse and dense distributions of seismic stations, especially for imaging relatively complex structures with lateral variations.