17:15 〜 18:45
[STT34-P04] Extracting Structural Boundary with High-Order Moments of Gravity Gradient Tensors
キーワード：重力勾配テンソル、構造境界、高次統計
The variation in gravity field on the Earth's surface is directly correlated with the uneven distribution of subsurface density. Gravity exploration has been commonly applied in mineral and oil/gas resource exploration, geological surveys, and investigations of faults and volcanoes. One of the main purposes of gravity exploration is to estimate the subsurface structures with density anomalies.
Besides direct measurements of gravity anomalies, the gravity gradient field data contains more high-frequence details that provides the directional derivatives of gravity potential field offering information about the source bodies from different perspectives. These anomaly data arise from the superposition of multiple source bodies with density anomalies underground, effectively reflecting the Earth's material distribution and stratigraphic structure. The separation of anomalies caused by the target body is fundamental for accurate interpretation for subsurface structures.
Traditional filtering methods are often employed to separate shallow and deep structure features or enhance specific horizontal boundary features, yielding visually straightforward results. However, these methods commonly rely on second-order statistical moments, with limited consideration given to higher-order statistical information. In this study, we employ forward modeling to construct gravity gradient data. It introduces higher-order statistical information to establish a new analytical method, attempting to extract target source bodies and highlight detailed information for subsurface structures. The objective is to enhance the signal-to-noise ratio and to extract additional structural features from the complex gravity gradient tensor fields.
Besides direct measurements of gravity anomalies, the gravity gradient field data contains more high-frequence details that provides the directional derivatives of gravity potential field offering information about the source bodies from different perspectives. These anomaly data arise from the superposition of multiple source bodies with density anomalies underground, effectively reflecting the Earth's material distribution and stratigraphic structure. The separation of anomalies caused by the target body is fundamental for accurate interpretation for subsurface structures.
Traditional filtering methods are often employed to separate shallow and deep structure features or enhance specific horizontal boundary features, yielding visually straightforward results. However, these methods commonly rely on second-order statistical moments, with limited consideration given to higher-order statistical information. In this study, we employ forward modeling to construct gravity gradient data. It introduces higher-order statistical information to establish a new analytical method, attempting to extract target source bodies and highlight detailed information for subsurface structures. The objective is to enhance the signal-to-noise ratio and to extract additional structural features from the complex gravity gradient tensor fields.