We propose a practical solution to overcome the frequency band limitation of conventional reverse time migration (RTM) through a least-squares migration approach. We first generate a high-frequency depth image by migrating deghosted marine seismic data using one-way wave equation migration (WEM). Then we use this highfrequency image as the initial model of band-limited leastsquares reverse time migration (LSRTM) and refine the image to enhance the low-frequency components and gradually recover the weak energy from complex structures. The original high-frequency components on the WEM image are maintained. Misplaced steep dips on the WEM image due to wide-angle approximation are also corrected after LSRTM. The final output image contains not only high-frequency and high-resolution horizons but also clear complex structures with sharp geologic boundaries. Compared to the method that directly computes broadband RTM images with high frequencies (greater than 40 Hz), our broadband LSRTM approach is more computationally affordable. We demonstrate the effectiveness of the proposed broadband LSRTM work flow with 2D and 3D real data examples.
[1]
D. Ristow,et al.
Fourier finite-difference migration
,
1994
.
[2]
Larry Lines,et al.
Comparison of Kirchhoff and reverse‐time migration methods with applications to prestack depth imaging of complex structures
,
1998
.
[3]
W. A. Mulder,et al.
A comparison between one-way and two-way wave-equation migration
,
2004
.
[4]
Gerard T. Schuster,et al.
Least-squares reverse time migration of marine data with frequency-selection encoding
,
2013
.
[5]
H. Masoomzadeh,et al.
Broadband Processing of Linear Streamer Data
,
2013
.
[7]
H. Masoomzadeh,et al.
Deghosted Least-squares RTM - Image Domain Broadband Solution for Complex Structures
,
2014
.