Statistical analysis of background noise in seismic prospecting |
| |
| Authors: | Tie Zhong Yue Li Ning Wu Pengfei Nie Baojun Yang |
| |
| Institution: | 1. Department of Information Engineering, Jilin University, Jilin, China;2. Department of Information Engineering, North University of China, Shanxi, China |
| |
| Abstract: | From a conventional viewpoint, seismic‐prospecting background noise is usually regarded as the product of a stationary and Gaussian stochastic process. In this paper, we use statistical methods to investigate the properties of the land‐seismic‐prospecting background noise on stationarity, Gaussianity, power spectral density, and spatial correlation. We use and analyse the passive noise records collected by receiver arrays at different typical geological environments (desert, steppe, and mountainous regions). Differences exist in the statistical properties of the background noise from different geological environments, but we still find some common characteristics. It is shown that the background noise is not strictly stationary and has different stationary properties over different timescales. Most of the noise records appear to be a Gaussian process when examined over a period of about 20 s but are found to be non‐Gaussian when examined over shorter periods of about 1 s. The background noise is a kind of colored noise, and its energy mainly concentrates in the low‐frequency bands. We also find that the spatial correlation of the background noise is weak. The results of this paper provide a scientific understanding about the properties of seismic‐prospecting background noise. |
| |
| Keywords: | Seismic‐prospecting background noise Statistical characterization Stationarity Gaussianity |
|
|
