Streaming seismic attributes

Abstract

Local seismic attributes play an important role in seismic processing and interpretation. However, the iterative regularized inversion required by the calculation of local seismic attributes makes it prohibitively expensive for real-time processing tasks. In this paper, we present an efficient method for estimating local seismic attributes, such as local frequency and local spectrum, using streaming computation. In our approach, the local attributes can be computed by updating the previously calculated attribute value using one new data point at a time, making it unnecessary to conduct the iterative inversion and thus significantly speeding up the computation for online usage. We apply our method to synthetic and field data to demonstrate its efficiency and effectiveness in accurately characterizing nonstationary seismic signals.