Simulation Of Lithospheric Magnetic Field Over Chinese Mainland

Soil parent materials are formed by weathering of rocks of lithosphere, whose magnetic characteristic produces by permanent magnetization of lithospheric magnetic field, thereby affecting the formation and nature of the soil. The magnetization of the lithosphere is due to primarily magnetite with varying content of iron oxide in lithosphere, which could divide into permanent magnetization and induced magnetization, and consequently forms the lithospheric magnetic field (also called crustal magnetic field or magnetic anomaly field). There are lots of important theoretical and practical values in research magnetic characteristic of soil parent materials, distribution of lithospheric magnetic field and structure of crust through simulating and analyzing lithospheric magnetic field, and then make a connection with research of soil texture, formation of lithosphere, etc.In this study, we based on the models of the lithospheric magnetic field, and took Chinese mainland (18°N～54°N,73°E～136°E) as research region, analyzing three key problems during simulation processes after combining filtered data, global and regional models, which included selection of background field, determination of truncate level and control of boundary effects; Indicating the variation of power spectra and distribution of lithospheric magnetic field corresponding to different phase of coefficients of high degree model; Simulating and analyzing the spatiotemporal variation of lithospheric magnetic field over Chinese mainland by using global and regional models, which will be good conferences to the magnetic characteristic of soil parent materials and so on.Theoretically, the lithospheric magnetic field generally does not vary with time while there is no giant geological structure change. Due to the improvements of survey instruments, measurement technologies, number of survey points, distribution of survey points and modeling technologies, all of which results in the distribution differences of magnetic field in different epoch over Chinese mainland. In order to achieve relatively real lithospheric magnetic field, systematically research from theories is necessary. In this study, models of lithosphenc magnetic field over Chinese mainland were derived based on survey data in 1960,1970,1980,1990 and 2000 respectively by Taylor polynomial, Surface Spline and the fourth generation comprehensive model of geomagnetic field CM4(Comprehensive Model 4). Moreover, moreover, in order to research distribution and variation during period of 1960～2000, suppose the lithoshperic magnetic values do not vary with time, then all data of each epoch were put into 2000, the corresponding models were derived and analyzed. For the sake of justify the models we derived, four relatively modish models with high precision models——CM4, MF7(Lithospheric Magnetic Field Model 7), NGDC-720, EMM2010(Enhanced Magnetic Model 2010) were adopted.The results are as follows:1. Took survey data of 2000 as example, results showed that the background field made by CM4' core field was more reasonable, which has a good coincidence with global models in distribution outside the boundary and suit to research the geomagnetic anomaly in large scale; By contrary, the background field made by Taylor polynomial suit to research the geomagnetic anomaly in middle or small scale due to the distribution of positive and negative values comparatively uniform and reasonable. To the ideal determination of truncate level, RMS of each truncate level, RMS of IGRF in corresponding year and AIC values of different truncate level should be synthetically taken into account. In order to fundamentally control the boundary effects, different number of complementary points, contours distribution and boundary anomaly should be synthetically considered.2. The Lowes-Mauersberger power spectra were carried out for 14 global models to show the power spectra of lithospheric magnetic field. High degree model——EMM2010 was taken as a example, it is found that there were five basically phases of its lithospheric power spectra (degree 16～719):Degree 16～65, power slightly increased; Degree 66～120, power sharply increased; Degree 121～220, power trended stable; Degree 221～670, power slowly decreased, Degree 671～719, power slightly increased again. After simulating and analyzing the corresponding distributions of lithospheric magnetic field based on five phases, results showed that the main trend of magnetic field was represented by coefficients of degree 16～65, which maybe caused a great impact to the magnetic characteristic of soil parents materials, magnetic field distribution of relatively more shallow lithosphere was represented by coefficients of degree 66～120, which had more regions of positive and negative values, moreover, distribution of more shallow lithosphere was represented by coefficients of degree 121～719, especially the degree 121～220 and 221～670, which had fewer distribution regions of positive and negative values. However, coefficients of degree 671～719 resulted in the more distribution regions. Different depth of lithosphere produced magnetic field with different scale, which has different distribution.3. The lithospheric magnetic field in different epoch over Chinese mainland was simulated by three kinds of models above mentioned, the totally distribution trends during 1960～2000 were also researched. Results showed that Component X had totally negative values over Chinese mainland, there was a negative extreme point in interchange of Qinghai, Sichuan and Tibet, which intensity was around-100nT, amplitude ranged from around-20～-40nT while longitude increased; Component Y had positive values in most regions over Chinese mainland, there were negative values in northwest, southeast and northeast of mainland, as a boundary of around 105°E, western of which was positive value area, eastern of which was negative value area, amplitude ranges from around 40～0nT while longitude increased; Component Z and F had whole negative values over Chinese mainland, both of which had similar distribution. There was an extreme point about around-140nT of Z in interchange of Qinghai and Gansu, an extreme point about-160nT of F in Qinghai, amplitude of which ranged from around-40～-100nT and around-60～100nT while latitude increased; About Component H, both extreme locations and intensity changes were relatively similar with Component X; Distribution of component D was as much same as that of Y, amplitude ranged from around 4～0°while longitude increased; Component I had negative values over Chinese mainland, there were some positive value areas in northeast, northwest and southwest of mainland, amplitude ranged from around 0.5～-0.5°while longitude increased. After simulating and analyzing variation trends and distribution characteristics of lithospheric magnetic field in different epoch, which will be the conferences of deeply research on magnetic characteristics of soil parent materials, soil texture.4. The lithospheric magnetic field over Chinese mainland was calculated and simulated by four global models—MF7, NGDC-720, EMM2010 and CM4, which has a relatively similar distribution trend. Distribution of MF7, NGDC-720 and EMM2010 were more complex, which could reflect more information of lithospheric magnetic field. Due to characteristics of global models, combining distribution of magnetic rocks, soil composition and so on, the relation models could be derived. After comparing lithospheric magnetic fields simulated by Taylor polynomial and CM4, results showed there were some big differences between component X, Z, F, H, but the wholly distribution trends of component Y, D and I were similar, some differences were actually produced by truncate level and modeling theories. Because of the number of survey points were much more than that of four global models, the regional models used also adopted extensively, so that the models of lithospheric magnetic field over Chinese mainland made in this study have important conference values.5. The main reasons that cause the differences between global and regional models are data and modeling theories; The main reasons that cause the differences between Taylor and Surface Spline model are field sources the model represents and modeling theories, which were deeply analyzed in this study. According to the similar distribution of global models, combined with the magnetic characteristic of soil parent materials, magnetic susceptibility in research region, related comparison works could be done. The simulation precise of models could be improved if some works like geomagnetic measurement, data processing and model creation could be done better. According to the features of models selected, it will provide some valuable instructive hints to research on soil properties in different region, scale and type. As for the results of stimulation and deviation of models, statistic analysis was done by regression analysis,mean absolute percentage error (MAPE) and so on, results indicated that the period with more survey points could obtain better simulating effects of Taylor polynomial, MAPE values were respectively less than 110nT and 0.10°in each epoch.6. Through two examples—Project of 863 called "Research of technologies of geomagnetic navigation indoor and under water", and project "Measurement technologies of stress in petroleum pipes" that cooperated with CNPC (China National Petroleum Corporation). Both projects had verified the reliability of models, which indicated that these models had extensive applied values.