The Study On Rapid Interpretation Techniques Of Gravity Anomaly And Gravity Gradient Tensor Data

Gravity prospecting data is composed of gravity anomaly data and full tensorgradient (FTG) data. It can be obtained by using satellite, airborne, ground and marinegravimeter and gradiometer. The gravity data reflect the information of spatialdistribution and density variation of geologic source. It can be used as the basis dataand starting point to interpret the geologic objects. In order to interpret the gravity andgravity gradient data rapidly, it is needed to select different gradient tensorcomponents or combine gravity anomaly to establish interpretation method fordifferent geological research objects. The response of different gradient tensorcomponents are not identical, which from the same geological source. The resolutionand detection ability are also different from each components. Therefore, in this paper,we aim to study on the interpretation method from theory and practice: to study on thedetectably of different data components, and to study on the rapid interpretationmethod from the combination of different components.Firstly, we evaluate and analyze the gravity anomaly and gravity gradient tensor(GGT) data by using forward model. These can provide the basis for the selection ofgravity and the gradient components. Based on the comparative research of therelationship between the gravity anomaly and the vertical derivative, we also make acomparative analysis of the resolution between the gravity anomaly and the gradientcomponents. And the study of resolution between each tensor components is alsoanalyzed. The results show that GGT components have higher resolution within thescope of the relatively shallow, but do not have advantage in the relatively deeperexploration. The superiorities of gradient tensor components lie in that they canprovide more details information of high frequency so that geological body can beexplained in detail. In addition, in this paper, we analyze the influence of the linelayout position and line spacing for the gravity anomaly and gradient components.The results show that with the increase of line spacing and line position deviationfrom the geologic sources, gradient tensor components can still give the relativelyintact geological information. The calculation results of the gradient tensorcomponents cannot provide more information. But they can give more detaileddescription of geological body from different sides. As a result, it can be seen in theinterpretation of the gravity anomaly and the gradient tensor data, gradient tensor data and multi-component joint interpretation method have more advantages.In view of this, in order to realize the perfect rapid interpretation method in thispaper, we establish the rapid interpretation method based on the traditionalinterpretation method to consider using more tensor component and optimize thegradient tensor component combination. To complete the interpretation work of thegravity anomaly and gradient tensor data, it is usually needed to obtain information ofgeologic source on the three aspects: horizontal position, depth scope and physicalparameters. Therefore, the mainly research theme of this thesis is about new methodsto determine this three parameters. Horizontal position usually is determined by edgedetection method, the depth scope parameters calculated by depth calculation method,and acquisition of physical parameters by traditional inversion method or by rapidimaging approach to obtain the related physical properties.The edge detection methods are widely used for they play an important role ingeologic interpretation, and the method has a great development in recent years. Themost commonly used traditional methods such as Tilt angle, Theta map method cangenerate lager error in deep geological edge detection, and they also lack the ability toextract more detail information. In this paper, we firstly analyze the relationshipbetween the two methods, and discuss the same theoretical framework, and then weuniform them as angle edge detection method. Based on this, we establish the otherangle edge detection methods such as arcsine angle edge detection method. Themodel tests show that all of these methods have same effect in edge detection. Inorder to improve the precision and the detectability of details, the angle edge detectionmethod is established based on a standard deviation calculation involving followingtwo steps: step1is calculation of standard deviation of the three directionalderivatives within windows, while step2is calculation of the trigonometric as normalangle edge detection. The model tests show that the new method can detect moredetail information which coincides with the real situation. Then we apply the methodon real field data, and good results have been achieved. In order to improve the offsetproblem of traditional angle method in deep geological edge detection, a new planefull tensor gradient angle edge detection method is established: Regarding horizontalplane as a whole relative to the vertical component, we perform a similar ratiocalculation of Tilt angle and Theta map, and propose a plane full tensor gradient angleedge detection and enhanced approach. Because of using more tensor component,made it more accurate to determine the edges of geologic sources. In this paper, theimproved inverse hyperbolic angle method overcomes the limitation of the original method can produce false edges when positive and negative residual density exist atthe same time. The new improved method is established by combining the gravityanomaly and gradient components. And the model tests and measured data processingresults demonstrate that the superiority of the improved method.There are many depth calculation methods for geologic sources, because of thedifferent types of geologic body and the research object need different depthcalculation methods. In this paper, we firstly establish an improved Tilt-depth method,which can be used to calculate the three-dimensional model such as sphere (pointsource) model. The model tests show that the method can be used to calculate thedepth accurately. To further display the advantage of gradient tensor componentinterpretation method, the curvature tensor matrix (CGTM) is introduced into thedepth calculation. We established a new depth calculation method based on thecurvature tensor matrix. The method is similar to tilt-depth method, and ourimprovement to this method is that the vertical derivative of the tilt angle is replacedby an eigenvalue of the CGTM. The improvement of horizontal resolution, so as tomake the depth calculation results more accurate. Both of the Tilt-depth method andthe CGTM-depth method are calculated separately in view of the positive or negativeresidual density body. In order to calculation the depth of geologic sources morequickly on the complicated geological conditions. We also use the gravity anomalycombined with the gravity gradient components, and improve the eigenvalues ofCGTM. Thus the improved approach can be used to directly calculate the depth ofgeologic sources exist both the positive and negative densities. Finally, we establisheda new depth calculation method based on the full tensor invariant. The depth range iscalculated using the ratio of gravity anomaly and the tensor invariant. It can amplifythe superiority of the tensor invariant in the gradient tensor data interpretation. Andthe new method also is a quick and simple method to obtain the depth of geologicsources. We use different model data and the filed data to test the method,respectively. And all of the results show that the method can get good applicationresults. They show the superiority and practicability of the new method.The rapid imaging method is usually used in the calculation of physicalparameters for the advantages such as they are fast and need small internal storage.Although these imaging method cannot give the real physical parameters, this resultswhich is related to the physical parameters are very useful in the geologicinterpretation. Therefore, in this paper, we firstly analyze the conventional DEXPmethod aimed to make certain the advantages, the causes of error. We find that it can get more little error when use the vertical derivative to calculate the imaging results.Based on this, the weighted factor is introduced into the generalized linear inversionimaging depth calculation, so as to make the generalized linear inversion imaging canbe more reasonable and accurate imaging results. To further develop the advantagesof gradient tensor component, based on the generalized linear inversion imaging, weestablish the gradient tensor generalized linear inversion imaging method. The modeltests and real data processing results show that its resolution is higher, and the result ismore accurate.All of the new interpretation methods are designed to realize the rapidinterpretation work of the gravity anomaly and the gradient tensor data. In this paper,the model data and real data analysis indicate that these methods can be used to makea rapid interpretation of the data from three aspects include the horizontal position,depth scope and physical property parameters of the geologic source. All of the newmethods can get high precision results, it demonstrated that the new methods havegood application value.