接收函数曲波变换去噪与偏移成像

增强接收函数偏移图像的垂向分辨率意味着提高参与叠加的接收函数的频率,但是采用高频接收函数通常伴随着对接收函数质量和参考速度模型的更高要求.通过叠加处理可去除部分接收函数中的随机噪声干扰,但同一台站的接收函数之间经常存在难以通过简单叠加消除的噪声信号.压制接收函数随机噪声的干扰可加强成像效果和提高图像分辨率,对推进叠加偏移成像质量的提高有重要的实际意义.本文利用在川西地区布设的31个流动台站所记录的远震波形数据,使用曲波变换去噪后信噪比增强的接收函数进行共转换点叠加(CCP),获得沿北纬31°线下方800km深度范围内速度间断面图像.研究结果表明:(1)对接收函数进行曲波变换去噪,可压制随机噪声,增强转换震相的追踪性,提高数据信噪比;(2)通过去噪处理,大幅提高接收函数用于偏移成像的主频率;(3)偏移结果确认了接收函数反演得到的松潘和川滇块体下方具有厚度约10～20km的过渡性Moho的认识;(4)上地幔过渡带的结果预示在龙门山断裂带以西的小范围内有可能存在下地壳或上地幔物质的拆沉.

Denoising the receiver function through curvelet transforming and migration imaging

Improving the vertical resolution of receiver function (RF) migration means to increase the high-frequency components of RFs based on high-quality data and more accurate reference velocity model. The randomly distributed noise in the high-frequency RFs could contaminate the migration imaging, some details of the images were blurred by extraneous phases. Therefore, noise suppression for RF data can greatly improve the resolution of RF migration. We selected RF data from 31 stations of the array along the profile of 31°N in the western Sichuan province. We used curvelet transform to denoise the data after the moveout correction, and obtained the structure of the crust and upper mantle in this area. The results show that:(1) Our method can significantly improve the signal-noise ratio of RF data. (2) The high-frequency components of the RF data have been well preserved during the denoising procedure. (3) Our migration results confirmed the existence of a 10~20 km Moho transition zone beneath the eastern margin of the Qinghai-Tibet Plateau. (4) The upper mantle image indicates the delamination of the lower crust or upper mantle to the west of the Longmenshan fault.