A B S T R A C TWe present the chain of time-reverse modeling, image space wavefield decomposition and several imaging conditions as a migration-like algorithm called time-reverse imaging. The algorithm locates subsurface sources in passive seismic data and diffractors in active data. We use elastic propagators to capitalize on the full waveforms available in multicomponent data, although an acoustic example is presented as well.For the elastic case, we perform wavefield decomposition in the image domain with spatial derivatives to calculate P and S potentials. To locate sources, the time axis is collapsed by extracting the zero-lag of auto and cross-correlations to return images in physical space. The impulse response of the algorithm is very dependent on acquisition geometry and needs to be evaluated with point sources before processing field data. Band-limited data processed with these techniques image the radiation pattern of the source rather than just the location. We present several imaging conditions but we imagine others could be designed to investigate specific hypotheses concerning the nature of the source mechanism. We illustrate the flexible technique with synthetic 2D passive data examples and surface acquisition geometry specifically designed to investigate tremor type signals that are not easily identified or interpreted in the time domain.

Locating subsurface sources from passive seismic recordings is difficult when attempted with data that has no observable arrivals or a low signal-to-noise ratio. Using time-reversal techniques recorded energy can be focused at its source depth. However, when a focus cannot be matched to a particular event, it can be difficult to distinguish true focusing from artifacts. Artificial focusing could arise from numerous causes, including surface waves, local noise sources, acquisition geometry and velocity model effects. We present a method to more reliably locate subsurface sources that reduces the ambiguity of the results. Time-reverse imaging techniques are implemented on both the recorded data and a noise model. In the data domain, the noise model only approximates the energy of local noise sources. After imaging, however, the result also captures the effects of acquisition geometry and the velocity model. The noise image is then used to correct the data image to produce an estimate of the signal-to-noise ratio. Synthetic data examples show the versatility of this technique to varying amounts of noise and to challenging velocity models. A field data example shows how this technique can be used to locate the source of low-frequency energy collocated with an oil reservoir.

No abstract

No abstract

Microseismic event locations obtained from seismic monitoring data sets are often a primary means of determining the success of fluid-injection programs, such as hydraulic fracturing for oil and gas extraction, geothermal projects, and wastewater injection. Event locations help the decision makers to evaluate whether operations conform to expectations or parameters need to be changed and may be used to help assess and reduce the risk of induced seismicity. However, obtaining accurate event location estimates requires an accurate velocity model, which is not available at most injection sites. Common velocity updating techniques require picking arrivals on individual seismograms. This can be problematic in microseismic monitoring, particularly for surface acquisition, due to the low signal-to-noise ratio of the arrivals. We have developed a full-wavefield adjoint-state method for locating seismic events while inverting for P- and S-wave velocity models that optimally focus multiple complementary images of recorded seismic events. This method requires neither picking nor initial estimates of event location or origin time. Because the inversion relies on (image domain) residuals that satisfy the differential semblance criterion, there is no requirement that the starting model be close to the true velocity. We determine synthetic results derived from a model with conditions similar to a field-acquisition scenario in terms of the number and spatial sampling of receivers and recorded coherent and random noise levels. The results indicate the effectiveness of the methodology by demonstrating a significantly enhanced focusing of event images and a reduction of 95% in event location error from a reasonable initial model.