Improving imaging and repeatability on land using virtual source redatuming with shallow buried receivers Dmitry Alexandrov 1 , Andrey Bakulin 2 , Roy Burnstad 3 , and Boris Kashtan 1 ABSTRACT Time-lapse surface seismic monitoring typically suffers from different sources of nonrepeatability related to acquis- ition imperfections as well as due to complexity of the sub- surface. Placing sources and receivers below the surface can improve seismic data repeatability. However, it is not always possible to bury a large number of sources, and therefore the next best option is monitoring with surface sources and buried sensors. We have discovered that redatuming of surface sources to the shallow buried receivers produced a reliable image of target reflectors despite the fact that receivers were placed in the near-field zone of the source. We redatumed data with the virtual source method using crosscorrelation of the measured wavefields. We found that redatuming also reduced nonrepeatability of seismic data associated with changes in acquisition geometry, variable source coupling, and daily/ seasonal variations in the near surface. We developed these results with a synthetic case study using a realistic 1D elastic model with a free surface and acquisition geometry from an actual field experiment conducted in Saudi Arabia. INTRODUCTION Repeatability of time-lapse seismic data is the single most impor- tant factor influencing permanent reservoir monitoring. Onshore seismic monitoring can especially suffer from various sources of nonrepeatability. Nonrepeatability causes time delays and ampli- tude changes that are obscuring 4D reservoir changes. It may origi- nate from the source and receiver side as well as from variations in near-surface conditions. Schissele et al. (2009) show that these issues can be effectively addressed by burial of the sources and receivers. Although it can be feasible to bury a large number of receivers and a small number of sources in case of simple near sur- face (Forgues et al., 2011), a complex subsurface may still require significant fold to obtain a reliable image. Berron et al. (2012) present a case study of acquisition with sources and receivers buried at a small depth and demonstrate that imaging in a desert environ- ment with a karsted near surface can be particularly challenging. An alternative approach to burying the sensors and vibrators is to redatum the surface sources to the buried receiver positions. The simplest way to accomplish this without any velocity model is by applying the virtual source method (Bakulin and Calvert, 2006). This method reduces the influence of the complex near surface above the receivers, and it also improves repeatability (Bakulin et al., 2007; Mehta et al., 2008). A recent experiment in one of the onshore fields of Saudi Arabia (Bakulin et al., 2012) showed prom- ising results for improving the seismic image and repeatability with the virtual source method. The seismic data were acquired using a densely populated source array and shallow buried receivers placed at a 30-m depth. Previous studies did not analyze repeatability im- provements due to virtual source redatuming for such a shallow burial of receivers. In this study, we evaluate the imaging and re- peatability for shallow buried receivers using a realistic elastic 1D model with a free surface. Redatuming can also be accomplished with other methods such as multidimensional deconvolution (Wapenaar et al, 2010), which may also remove the free-surface multiples. This technique requires adequate receiver spacing, and in case of sparse receiver sampling, the redatuming can suffer from spatial aliasing (Hunziker et al., 2012). As such, we leave a trial of this approach for future studies. We use elastic finite-difference modeling and perform our analy- sis in two steps. First, we focus on imaging the deep target reflector and estimating the redatuming parameters that provide the best pos- sible image. To improve the image, we apply up-down wavefield decomposition at the receivers using dual-sensor summation (Barr, 1997). In addition, having the downgoing wavefield separated, we perform a detailed analysis of how the source-side ghosts affect Manuscript received by the Editor 11 August 2014; revised manuscript received 8 November 2014; published online 27 February 2015. 1 Saint Petersburg State University, Saint Petersburg, Russia. E-mail: dalexl30@gmail.com; bmkashtan@gmail.com. 2 Saudi Aramco, EXPEC ARC, Dhahran, Saudi Arabia. E-mail: andrey.bakulin@aramco.com. 3 Aramco Research Center, Houston, Texas, USA. E-mail: roy.burnstad@aramcoservices.com. © 2015 Society of Exploration Geophysicists. All rights reserved. Q15 GEOPHYSICS, VOL. 80, NO. 2 (MARCH-APRIL 2015); P. Q15–Q26, 20 FIGS., 1 TABLE. 10.1190/GEO2014-0373.1 Downloaded 10/16/17 to 51.36.185.77. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/