Abstract: Based on continuous waveform recording of vertical and horizontal component of broadband stations in June and July of 2016 in Fujian and its adjacent areas, we used the phase-weight stack algorithm to improve the signal to noise ratio (SNR). We obtain high quality Empirical Green Functions (EGF) of 108 stations. Applying time-frequency analysis technique from symmetrical component of EGF, we measured a large number of high quality Rayleigh and Love wave group velocity dispersion curves in the 1~20 s. To reduce error of straight ray-path assumption in 2D tomography of group velocity map, the travel-time field was computed with a finite-difference scheme and the ray-path was tracing on the travel-time field. The checkerboard test suggests the resolution of 2D tomography is up to 50 km. We obtain the Rayleigh and Love wave group velocity map of Fujian province and adjacent areas in the 1~20 s. The results show that the group velocity distribution of 1~10 s is consistent with the upper and middle geological features of the crust, and the Fuzhou and Zhangzhou basins are imaged as obvious low velocity bodies at the periods. The long period group velocity results suggest a high geothermal region at the northwest of Zhangzhou in the middle and lower crust, and the velocity difference between the two sides of Zhenghe—Dapu fault that suggest it maybe a deep and large fault and have an obvious velocity difference between east and west. The general pattern of our model crust is consistent with geological units in this region.

Key words: Tomography imaging, Ambient noise, Surface wave group velocity, Fujian area