High-resolution marine seismic data degraded by sea swell can be enhanced by swell correction. To correct for swell, two-way travel times are detected for sea bottom locations in seismic traces; linear predicted values are calculated from the previously detected travel times, and correct static shifts between the detected and calculated values before stacking. To pick sea bottom locations for very large numbers of seismic traces, automatic sea bottom detection (SBD) is normally required. However, sea bottom signal degradation by wave noise causes SBD failure. Our first SBD trial using trace-by-trace (TBT) picking from raw and energy ratio data was unsatisfactory, with many of the detected sea bottom values deviating from the correct values due to noise in the field data. However, we achieved reasonable results by considering source–receiver geometry and previously detected sea bottom values. We picked reliable sea bottom values within an expected hyperbolic range (EHR) calculated from the source–receiver offset and group interval. To prevent excessive correction for swell, we excluded unreliable picks that deviated greatly from the linear trend of previously detected sea bottom values, or from the immediately preceding value. Swell correction within the EHR significantly improved the quality of noisy data. We applied this method to noisy field seismic data acquired offshore Yeosu, Korea, and successfully detected the sea bottom, corrected for swell, and improved seismic section. Thus, high-resolution marine seismic data degraded by sea swell can be enhanced by pre-stack swell correction in shot gathers using a hyperbola calculated based on source–receiver geometry.

通过海浪修正可以增强因海浪而退化的高分辨率海洋地震数据。为了校正膨胀，在地震道中检测海底位置的双向行进时间。根据先前检测到的行驶时间计算线性预测值，并在叠加之前在检测到的值和计算出的值之间校正静态偏移。为了为大量地震痕迹选择海底位置，通常需要自动海底检测（SBD）。但是，由于海浪噪声引起的海底信号衰减会导致SBD失效。我们的第一个SBD试验使用的是从原始和能量比数据中逐条跟踪（TBT）进行拾取，但由于现场数据中的噪声，许多检测到的海底值偏离了正确的值，因此效果不理想。然而，通过考虑源-接收器的几何形状和先前检测到的海底值，我们取得了合理的结果。我们从源－接收器偏移量和组间隔计算出的预期双曲线范围（EHR）中选择了可靠的海底值。为防止过度校正涨潮，我们排除了不可靠的镐，这些镐与先前检测到的海底值的线性趋势或之前的值有很大出入。电子病历中的溶胀校正显着提高了噪声数据的质量。我们将此方法应用于在韩国丽水海上获取的嘈杂的现场地震数据，并成功地检测了海底，进行了涨潮校正并改善了地震剖面。因此，