Weighted-average time-lapse seismic full-waveform inversion

Abstract

As seismic data can contain information over a large spatial area and are sensitive to changes in the properties of the subsurface, seismic imaging has become the standard geophysical monitoring method for many applications such as Carbon capture and storage and reservoir monitoring. The availability of practical tools such as full-waveform inversion (FWI) makes time-lapse seismic FWI a promising method for monitoring subsurface changes. However, FWI is a highly ill-posed problem thatcan generate artifacts. Because the changes in the earth's properties are typically small in terms of magnitude and spatial extent, discriminating the true time-lapse signature from noise can be challenging. Different strategies have been proposed to address these difficulties. In this study, we propose a weighted-average inversion to better control the effects of artifacts and differentiate them from the true 4D changes. We further compare ve different related strategies with synthetic tests on clean and noisy data. The effects of seawater-velocity variation on different strategies are also studied as one of the main sources of nonrepeatability. We tested different strategies of time-lapse FWI using the Marmousi and the SEAM Time-Lapse models. The results indicate that the weighted-average method can provide the best compromise between accuracy and computation time. This method also provides a range for the possible answer of other time-lapse FWI strategies.

Pyfwi: A Python Package for Full-Waveform Inversion and Reservoir Monitoring

“…PyFWI allows geophysicists to conduct a time-lapse study using seven methods such as parallel, double-difference [23], cascaded [24], cross-updating [25], simultaneous [25], central-difference [5], and weighted-average inversion [6]. These methods are discussed in more detail in [26].…”

Monitoring fluid saturation in reservoirs using time‐lapse full‐waveform inversion

“…Time-lapse FWI can be performed by considering different combinations of FWI runs. These combinations can be performed sequentially such as for the cases of the cascaded, cross-updating and weighted-average methods (Watanabe et al, 2005;Maharramov et al, 2016;Mardan et al, 2022cMardan et al, , 2023b or in parallel as in the case of the independent and central-difference methods (Plessix et al, 2010;Zhou & Lumley, 2021). In addition to these methods, Maharramov et al (2016) proposed a joint inversion of seismic data from different vintages by introducing the cost function as where subscripts 𝑏 and 𝑚 denote baseline and monitor models.…”

“…Simultaneous TL‐FWI is one of the most accurate strategies for time‐lapse studies (Mardan et al., 2023b). Incorporating a regularization term for minimizing the difference between two estimates (baseline and monitor models) helps this method to produce less artefacts.…”