Item

Abstract

Statics are time-shifts that occur in reflection seismic trace data and are generally considered to be mainly due to shallow velocity variations. Since the refraction static correction is most often based on first break picking and subsequent velocity model estimation, it is even today a labor-consuming and error-prone procedure. Time-lapse seismic also faces this issue in a temporal sense, since changes in statics, due to temporally variable near-surface conditions, are known to be first-order contributors to time-lapse noise. Considerable changes in the statics of repeated on-shore seismic surveys can occur due to precipitation-related changes in soil moisture and in the groundwater table, or may be due to man-made earthworks. Production-related or injection-related processes can cause considerable velocity changes, which leave time-shift imprints on time-lapse seismic data that can be very similar to that of near-surface velocity variations. In this context it is crucial to consider that refraction static corrections are in many cases of limited use, as they aim to enhance the stack coherency of the individual time-lapse data sets only. As an alternative, we propose a time-lapse difference (TLD) static correction that is focused on the accommodation of static changes between the time-lapse data sets. This TLD static correction decomposes the static differences that are determined from cross-correlations in a surface-consistent manner. It therefore does not require first break picking and inversion for velocities from repeat data sets. We tested the TLD static correction for a 4D case study from the Ketzin CO2 storage site, Germany. As a reference we used the results that were obtained from a recent processing in which refraction static corrections were performed individually on the time-lapse data sets. Although the TLD static corrections method is considerably less time-consuming, we found that it is providing a stack difference with enhanced S/N. This is particularly demonstrated for a 4D seismic signature that is proven to be due to injected CO2. This Ketzin case study shows further that the pattern of the TLD statics is highly consistent with patterns in the cumulative precipitation data. This observation confirms that near-surface velocity changes are due to changes in the soil-moisture saturation and that an efficient compensation for them can be achieved by the TLD static correction.