Analysis of "black triangle" noise and its suppressing

In the single shot gather of vibroseis seismic data for 3D land seismic exploration, "black triangle" noise is distributed in a cone, with the shot point as the apex.Owing to its complex and variable characteristics, black triangle noise seriously reduces the signal-to-noise ratio of data gathered and the quality of imaging results.In this study, black triangle noise characteristics was analyzed and found to be mainly composed of Rayleigh-type surface waves and strong, scattered waves.A noise suppression scheme was then proposed and an adaptive data suppression method was developed.The key steps were as follows:first, the black triangle noise area was explicitly or implicitly separated.Second, linear move-out correction was performed on the data based on the maximum apparent velocity of the separated area.Third, a local window of the data (a time-space window) was defined and frequency-division robust Principal component analysis was performed to extract the linear signal.Finally, statistical filters were utilized to eliminate anomalous amplitude noise.Tests on field data showed that this method can suppress surface waves while preserving effective signals.