As a result of more than fifty year's aeromagnetic work, large-and medium-sized sedimentary basins and the total thickness of their sedimentary rocks have been discovered and delineated, the targets for oil and gas exploration in each sedimentary basin have been evaluated, and a series of important banded tectonic belts and block-like tectonic regions have been discovered, whose dislocation mosaic frameworks constitute the block tectonic pattern in China's geoscience. In this pattern, the nearly EW-trending Yinshan-Tianshan tectonic belt depicts Siberia-affinity tectonic domain (southern boundary of Junggar, Songliao, Jumusi block); West Kunlun-Qinling-Dabie tectonic belt constitutes the southern boundary of paleo-Chinese tectonic domain (Tarim, Qaidam-Alxa, North China-Jiaoliao block); Hoh Xil-Jinshajiang tectonic belt displays the northeastern boundary of Gandwana-affinity tectonic domain (Qiangtang, Ge'gyai-Baingoin, Gangdise-Lhasa, Himalaya block); Qinzhou-Hangzhou tectonic belt delimits the northern boundary of Cathaysian-South Sea tectonic domain (Cathaysian block, East Sea shelf, Taiwan Straits, Zhujiang estuary-Hainan Island block and South Sea basin). In the mosaic geoscience block, NS-trending rift tectonic belts, marginal curtain-style arc block tectonic belts and annular tectonic regions are distinctly displayed in the aeromagnetic anomaly map, and are closely related to the concealed oil-gas resources, coal fields and iron ore concentration areas, thus serving as the important targets in current nationwide ore-prospecting work. China's continental crust has experienced complex breakup, accretion, collage, gravity adjustment, and thermal isostatic evolution. These activities were controlled by the nonuniform action of the revolution of the Earth and mantle upwelling as well as subsidence dynamics.

In view of the fact that the northern South China Sea has complex geological structure, well-developed igneous rocks, and serious multiple wave interference, the authors used marine gravity and magnetic survey data, chose the United States FUGRO-LCT gravity magnetic seismic integrated interpretation system as the study platform in combination with seismic data to build a reasonable 2.5D geological-geophysical initial model and conduct synchronous gravity and magnetic forward and inversion calculation. According to the results of fitting modified seismic reflection fuzzy regional calibration position, the authors determined the deep formation properties and the structure of the crust. The results show that the Mesozoic formation density of the northeastern South China Sea is generally 2.40~2.57 g/cm³, and the density of some igneous rocks is less than that of Mesozoic sedimentary rocks. Some areas have magnetism, caused by sedimentary rock interbedded with igneous rock. Dongsha Island of the south slope and the basin at the junction of crustal structure are weak areas, where there may be Pre-Mesozoic strata. The region from the southern uplift area to the northeast area of the central basin remains the excessive crust.

In order to determine the geochemical characteristics of soil gas in the central south segment of Xiadian fault and evaluate its seismic hazard, the authors measured the concentrations of He, H₂, CO₂ and Rn in soil gas at 221 measuring points distributed in Qixinzhuang, Pangezhuang and Dadongguan sites across the fault. The geochemical characteristics, the background values and the anomaly thresholds of He, H₂, CO₂ and Rn in each sampling profile were identified. The data obtained indicate that various soil gas anomalies occur along the fault, and the utilization of various components of soil gas can be an effective way to identify the concealed fault. In addition, based on the anomaly amplitudes of He and Rn along each sampling profile, the authors infer that the activity of Xiadian fault in Pangezhuang is stronger than that in Qixinzhuang and Dadongguan. However, the decrease of the measured peak hydrogen concentrations in Xiadian fault and the seismic section data reveal that the activity of the central south segment of the fault was somewhat weakened.

The Cambodian Kampong basin located in the east Gulf of Thailand, which is a Meso-Paleozoic basin and located in the east of the paleo-Thetis suture zone. The basin that recognized as a intraplate-craton basin until the Permian, became the intracontinental rift basin after the late Triassic, as the tectonic movement. At present, the exploration degree in this basin is very low. The hydrocarbon research doesn't carry out before, thus seismic data and well log data are lack. In order to analyze and evaluate the hydrocarbons potential, we integrated the seismic and high accuracy satellite gravity data (sample density: 0.02°×0.02°), to calculate Bouguer gravity anomaly, the gravity vertical second-order derivative and the gravity level total gradient anomaly, According to the gravity field theory and the global gravitational field model(EGM96).Finally, Based on seismic constrained gravity inversion and interpreted integration, the research forecasts the basement depth, analyzes the fault system and divide the tectonic unit, finally indicated the direction for the next step exploration research.

The obtaining of natural gas in borehole DK-1 in Muli area of Qilian Mountain in 2008 marks the fact that the study of natural gas hydrate in permafrost has entered into a new stage. However, there has been no effective survey technology, which constitutes the bottle neck of permafrost natural gas hydrate exploration. The authors examined the method and technique of geochemical exploration, and have reached the conclusion that acidolysis hydrocarbon, headspace gas and other geochemical indicators in shallow soil are obvious abnormal in places where natural gas can be obtained. The geochemical anomaly and natural gas hydrate are of homologous genesis, and hence this paper tentatively proposes a geochemical exploration model for natural gas hydrate in Muli area.

In 1^st member of Zhujiang Formation in A oilfield of Wenchang,the reservoir lithology and physical properties can not be recognized by measured acoustic wave,density log and the calculation of the wave impedance in this area,due to the effects of such factors as special lithology and wall collapse,thus leading to difficulties in predicting the reservoir.The multi-well logging and seismic attributes were utilized in integrated way to invert lithological characteristics for different target zones,and the reconstruction processing was conducted for sonic curve.The curve can represent the real features of the reservoirs and can also significantly enhance the resolution of acoustic impedance inversion profile and the capability of reservoir description and prediction.Practical application of this technique is very successful.

According to an analysis and sumarization of the geological, mineral resources and geochemical characteristics of the Xiaojiayingzi molybdenum ore district, the authors tentatively established a geological-geochemical prospecting model for porphyry-skarn type molybdenum ore deposits:the geological conditions are characterized by exposed strata of Wumishan chert bands-bearing dolomite and chert nodule-bearing dolomite, strong magmatic activity, and development of fractures before mineralization;the regional geochemical anomaly finds expression in the anomalies of Mo, W, Au, Cu, Pb and other elements, and each single element has its significant concentration center in the high value region; mineralization is partly expressed as significant zonation of elemental composition among ore body, contact zone and stratigraphic distribution, with Mo, W, Cu, Au serving as near-ore indicator elements combination. The establishment of the geological-geochemical prospecting model provides an important basis for the prospecting for the same type of deposits, the delineation of prospective areas, the selection of prospecting targets and the estimation of resources.

Located in the middle segment of the Middle and Lower Yangtze metallogenic belt and on the northern margin of Lu-Zong basin, the Xiwan iron-polymetallic ore pre-survey area of the Qingshuitang-Yueshan iron and lead-zinc ore-prospecting target is characterized by low-lying relief, dense network of rivers and thick overburden. In order to solve the problem of prospecting for iron-polymetallic deposits in thick overburden areas, the authors adopted the combined geological and integrated geophysical method to analyze gravity, magnetic and electric anomalies, thus making clear the ore-prospecting target and determining the favorable ore-prospecting position. Drill verification led to the discovery of a lead-zinc deposit at the position of IP anomaly, thus attaining ideal ore-prospecting result.

Pre-stack simultaneous inversion and CRP of pre-stack time migration provide an effective reservoir prediction approach for lithology and subtle reservoir exploration.The pre-stack simultaneous inversion formula is given according to the fundamental theory of elastic wave propagation and through simplifying the Zoeppritz equation for plane wave reflection and transmission.On the basis of well log data analysis and calculation of the elastic parameters,the utilization of P wave impedance and S wave impedance cross-plot analysis can distinguish lithology,the use of the P wave impedance and porosity can distinguish the reservoirs,and the employment of the reservoir and fluid factor λ cross-plot can distinguish oil and gas potentials of the reservoirs.The plane distributive and cross sectional characteristics of E_s1 gas and oil reservoir were discriminated directly with the inversion result,and the distributive and cross research shows that inversion is practical for the quantitative discrimination of the reservoir.

The combinational test of geophysical prospecting methods in the Wanglou ore district of Wugang City shows that the high precision magnetic method and the controlled source audio-magneto-telluric method seem to be the best combination in search for metamorphic iron deposits, that high precision magnetic method can be used to determine the existence or nonexistence of a deep underground magnetic geological body, and that the controlled source audio-magneto-telluric (CSAMT) can be utilized to determine the existence or nonexistence of a low resistance body. In combination with geological metallogenic regularity, we can judge whether the magnetic anomaly is caused by an iron ore body or not. According to the combined geophysical field characteristics, we can determine the spatial mode of occurrence of the iron ore body.

Gravity prospeting method has been one of the best way to solve geological problems. The efficiency of gravity point positioning affect the overall efficiency of the work progress, and the elevation accuracy determines the accuracy of the calculation of the gravity data processing. GPS, which can quickly obtain high-precision three-dimensional coordinates, accelerate the progress of the gravity prospecting method and improve work efficiency. RTK is the main method for gravity prospecting. But that need to set up the base station, and the further between the mobile stations and the base station, the worse of the accuracy. CORS makes up for the lack of RTK,because it can quickly survey high-precision coordinates of the WGS84 three-dimensional do not need to set up the station. The CORS system has been widely used in the fields of mapping, land, water, geological disasters, but not yet applied to the exploration of gravity. In this paper, the advantages of CORS will be discussed at first, then how to convert the results of the CORS to local coordinate system will be presented. By applying in Jiangsu 1：50 000 gravity survey, the plane accuracy is better than 1 cm, and refined quasi-geoid height accuracy is better than 5.2 cm.The accuracy meets the requirements of gravity prospecting.

"Step" is the main physical model for the interpretation of gravity gradient zone anomaly. The theoretical calculation has proved that the Bouguer gravity anomaly (Δg) directional derivative (V_xz) peak value indicates the ground projection position of the top edge of the step that forms the gravity gradient zone, but is less related to the angle of the step. Using the 3/4 extreme value position, the authors deduced another inversion formula and proved that this formula is better than the common formula which uses 1/4 extreme value position.

The criterion of aeromagnetic survey (DZ/T 1402-2010) provides specific measures for assessing the performance of digital magnetic compensation; nevertheless, it does not define the method of quality assessment. On the basis of the quality assessment method for aeromagnetic compensation apparatus (DSC-1), this paper proposes post-assessment method using uncompensated and compensated magnetic fields. The authors have also developed the performance assessment software for digital magnetic compensation on GeoProbe platform, which proves to be a valid tool to give uniform performance measures for digital magnetic compensation.

China's IP transmitter employs voltage source power supply but uses a steady flow source supply formula to calculate the polarizability, thus leading to a somewhat lower value, which can not represent the real ground polarization. The authors derived the formula for calculation of voltage source power supply polarizability and, on such a basis, developed the device for the measurement of the change rate of the current. The use of the corrective formula can get the real ground polarizability.

Terrain is an important factor affecting airborne gamma-ray spectrometry. The traditional height correction method for airborne gamma survey in complex topography will cause distortion of the correction result. In consideration of the features of dynamic survey and by using the DTM data acquired by airborne geophysical survey as well as the radiation field theory of point radiator and detailed terrain classification, the authors studied the implementation of a terrain correction method utilizing terrain coefficients. A comparison with the result of ground-based anomaly check shows that the terrain correction method applied has achieved satisfactory result in practical data processing and can basically eliminate the impact of terrain.

By optimizing the detecting conditions, the authors obtained the best condition for analyzing carbon dioxide, oxygen and nitrogen in debris top gas by GC in a single run. Accuracy and precision were used to evaluate the method. It is shown that this method has high accuracy and precision. This method is characterized by simple chromatographic conditions, fast analysis speed and small gas volume and it has been proved to be effective in known areas. It can be used to analyze carbon dioxide, oxygen and nitrogen in geochemical subsurface exploration samples, simulated samples and natural gas samples.

In this paper, based on analyzing hydrocarbon gas in 213 soil and headspace gas samples collected from near-surface soil in Muli permafrost of Qinghai Province by such geochemical methods as acidolysis hydrocarbon, headspace gas and ΔC. The results indicate that the oil and gas geochemical exploration method has a positive meaning in land gas hydrate exploration and analysis of acidolysis hydrocarbon, headspace gas and content of ΔC indicators in shallow soil, which shows obvious top anomaly characteristics in the place where natural gas can be obtained. The authors hold that the "acidolysis hydrocarbon + headspace gas +ΔC" anomaly can serve as an important indicator to identify the occurrence of nature gas hydrate.

In order to determine the role of the phase state of elements in anomaly recognition, the authors studied the compositions of sulfide phase, oxide phase and combination phase of Cu in porphyry copper deposits and Zn in hydrothermal lead-zinc deposits. In the porphyry copper ore district, Cu exists mainly in sulfide phase and oxide phase, with the criteria of the ore-related anomaly being (w_{sulfide phase Cu} + w_{oxide phase} Cu)/ w_{combination phase Cu}≥21、(w_{sulfide phase Cu} + w_{oxide Cu})/ w_Cu≥96%; in the lead-zinc ore district, Zn in the background area mainly exists as combination phase, whereas Zn in the anomalous sectors (w_Zn≥ 200×10^-6) mainly exists as sulfide phase (including oxide phase), with the segments characterized by (w_{sulfide phase Zn} + w_{oxide phase Zn})/ w_Zn≥87%、(w_{sulfide phase Zn} + w_{oxide Zn})/ w_{combination phase Zn}≥7 being the favorable ore-prospecting segments and serving as the marks indicating exposed ores or concealed ores.

A low-velocity abnormal body upper strata under low-reduced zone is called conceal low-velocity body.If the body were enough wide and high,it were made travel time of seismic wave to delay.The delay of travel time of seismic wave will not be corrected as a result of deficiency of the delay in conventional static correction because the body is under the low-reduced zone.Thus shape of reflected events will be changed,and result in structure form will be changed.Analysis showed that it should belong remnant static time difference that change of the shape of reflected events brought by the conceal low-velocity body.But frequently the change were be mistook to bring by structure form changed the change because change of shapes of structure forms from different depth is different.If the remnant time differences were not be removed the seismic structure map were be distortion,so it will be perplex to map out the exploration oil and gas.

The subbottom profiler is one of the most widely used geophysical instruments for marine surveying and marine geological research.Although its vertical resolution becomes more and more high,there are some pitfalls in interpretation of the actual acoustic profile data.In this paper,based on the working principle of the subbottom profiler,the authors analyzed four factors causing the interpretation pitfalls.First,the understanding pitfall caused by the difference between acoustic layers and geological layers;Second,the shallow stratigraphic distortion caused by the horizontally separately placed transducer and hydrophone;Third,the spatial deformation of the acoustic interfaces in the data profiles reflected from special stratigraphic interfaces;The last,the layer thickness distortion caused by the change of the speed of acoustic wave propagation in marine sediments.The method concerning the elimination of these pitfalls during processing and interpreting the subbottom profile data is also discussed.

There are many electricity penetration methods with different characteristics for detecting the floor-water hazard in the working face. Through the laboratory test by simulating monopole-dipole method, parallel bipolar-dipole resistivity method and parallel electrical method, the authors analyzed the advantages and disadvantages of these three kinds of detecting methods so as to provide reference for detecting the floor-water hazard in the working face. The results show that the three methods have their respective characteristics and reactions to anomalies. The monopole-dipole method has less convergence of the detection results, and the quantity of data collected by this method is limited; the parallel bipolar-dipole resistivity method characterized by relatively low data-collecting efficiency and long construction time is suitable for small range exploration by collecting massive data; the parallel 3D electrical technique of lane pairs is suitable for meticulous exploration because it can collect massive data, has good convergence of the target body and can accurately determine the spatial limits of the water-bearing district on the working face floor.

In this paper,the seismic inversion combined with the well logging,rock physics,seismic facies and the forward modeling was used to comprehensively study the characteristics of the igneous rocks.The rock physics experiment is a useful tool to study the physical property parameters of the igneous rocks.The analysis of the experimental data is helpful to predicting the igneous lithology and velocity. On such a basis,the seismic inversion is used to describe the igneous rocks by the seismic data.The environmental correction and standardization of the original logging curves is the prerequisite for high quality synthetic seismogram and inversion.The constrained sparse spike inversion,the model based inversion and the neural network inversion have higher resolution than the original seismic data.The spatial distribution of igneous rocks can be identified by them,with the model based inversion having the highest resolution.The GR characteristic inversion has higher resolution than the above-mentioned three methods.However,as a single characteristic can hardly be used to describe the characteristics of igneous rocks,the multi-parameter seismic inversion is an efficient way.The seismic facies is also used to obtain the igneous 3D velocity field.The precision of the velocity field is verified by the error analysis and forward modeling.

The utilization of the micro-seismic method to conduct real-time dynamic monitoring of the production in the production sweep area of the oilfield is a frontier research topic which has aroused much attention both in China and abroad.This authors deal with the mechanical mechanism producing the micro-seismic event as well as the principle and method of micro-seismic monitoring.The micro-seismic monitoring technique was used to conduct real-time monitoring of the large-scale fracturing process along BYHF1 shale horizontal well section of Biyang depression,the dipole logging and VSP data were used to establish the initial velocity model for the monitored well area,and correction was made through the perforating gun.Based on recognizing the effective micro-seismic event,the authors utilized the longitudinal and transverse wave information to calculate the position of the micro-seismic event,and described the direction,scale and spatial distribution of the fissures in the fracturing well area by using the visualization technology.The result of the monitoring is well consistent with that of the imaging logging.

The surface velocity structure can be estimated by inverting the dispersion curve extracted from the dispersion image result of the Rayleigh-wave information in shot gathers of near-surface exploration.The dispersion image results from the difference of Rayleigh-wave data in different offset ranges,and will affect the extraction and inversion results of the dispersion curve.In this paper,the authors studied the dispersion imaging results in different collection arrangements and offsets by dispersion imaging resolution and continuity using box wave in LJ area of the Shengli oilfield,and found that the long collection arrangement can improve the dispersion imaging resolution effectively,and that the dispersion imaging result in near offset is better.The authors also tested and verified the effectiveness and applicability of the optimum offset formula proposed by Zhang in 2004,and the results show that it can be used as a standard of the optimum offset in near-surface Rayleigh-wave exploration.

In order to understand the exploration detection capability of the transient electromagnetic method under some simple conditions, the authors conducted the physical simulation on the responses of the spherical anomalous body in different resistivity backgrounds, different forms of plate-like anomaly in the high resistance background, and spherical anomalous body at different depths. As a result, the authors obtained a number of induced voltage normalized curve profiles and apparent resistivity-time profiles, determined the best results of transient electromagnetic method for detecting the high resistance geological background conditions below the low resistance anomaly body, with the exploration depth being 2~3 times the side length of the frame. The results obtained by the authors provide some reference for the development of transient electromagnetic exploration.

Starting with the principles of IP logging and field methods, the authors analyzed the effects of polarization logging curve in electrode spacing, formation resistivity and residual polarization potential in combination with the measured curve. Some conclusions have been reached: the short distance of the electrode system should be selected in logging detailed stratification, the long distance of the electrode system ought to be selected for understanding electrical of rock and mineral layer, or the background value pole pitch electrode system should be chosen; the formation resistivity affects the visual polarizability logging curve of layering. In addition, in order to avoid the impact of residual polarization potential, we should decentralize the bottom of the gradient electrode system in logging and enhance the top of the gradient electrode system in logging.

It is very important to study the factors that affect the change of different forms of heavy metals in soil so as to understand their geochemical behaviors. In this paper, the typical soil in Yantai gold ore district and surrounding areas was described, organic matter, clay, the relationship and the ecological harm of magnetic susceptibility and pH in soil with different forms of heavy metals in soil were also discussed. Some conclusions have been reached: the main existing forms of heavy metals are residual state and Fe-Mn oxide state, whereas the main existing form of Cd is ionic exchangeable form;soil acidification may increase the content of Ni, Zn, Cd and Hg in activity form; if the soil is seriously polluted by Pb, the maintenance of an alkaline environment can reduce Pb harmfulness, and the prevention of salinization is an effective measure for eliminating As pollution in soil. The improvement of organic matter content may reduce the active state form content of Pb, Hg and Ni, whereas the increase of clay content may slightly reduce the percentage of ionic exchangeable form of Pb, Zn; soil magnetic susceptibility is a fairly ideal indicator for acidification situation and pollution degree of soil.

In order to find out the geochemical and hydrogeochemical background associated with high fluoride groundwater in Gaomi and to study the regularity governing sources, movement, and concentration of fluoride in groundwater, the authors collected 382 pieces of rock and soil samples, 300 pieces of groundwater samples, and analyzed the main components in them. The authors intensively studied the relationships of F^- with geotechnical category, pH of groundwater, Na⁺, Ca²⁺, Cl^-, SO₄^2-, and the horizontal and vertical distribution characteristics of F^-. Some conclusions have been reached: the main sources of fluorine in groundwater in this area are sandstones, conglomerate, mudstones, volcanic rock and pyroclastic rock in the south hill bedrock area; the horizontal distribution of soil fluorine content increases from south to north, and the vertical distribution of soil fluorine content decreases from top to bottom; the fluoride content in the high fluoride groundwater is positively correlated with Na⁺, and negatively correlated with Ca²⁺, and there is no significant correlation between fluoride and anions concentration and no significant correlation between fluoride and total dissolved solid; nevertheless, there exists weak positive correlation between fluoride and the value of pH. The result of the research on the geochemical and hydrogeochemical background in high fluoride areas can serve as a reference for the research on the formation, migration and accumulation of the high fluoride groundwater in the endemic fluorine disease area.

The studied site is located within an abandoned gas station which had nearly existed for 40 years. The authors used ground penetrating radar and high density resistivity method to detect the petroleum hydrocarbon contamination. At the same time, four monitoring wells were drilled in the gas station and samples of groundwater were analyzed so as to verify the accuracy of the two methods in detecting light non-aqueous phase liquids (LNAPLs) contamination. Comprehensive detection results showed that the source of pollution was the leaking pipes, the diffusion range reached the gas station boundary and the vertical distribution of con-tamination was about 0.5~2.8 m. Ground penetrating radar and high density resistivity method used in the investigation provided an excellent tool for environmental detection, suggesting that they can be applied to the study of similar areas.

In Tahe area,large-sized caves and fissure-caves are the main reservoir space types,which are characterized by complex geological characteristics and rapid change in lithology.The seismic response of these kinds of reservoirs is diffraction whose precision of images has a great impact on the reservoir prediction.Based on the scattering theory,the equivalent offset method (EOM ) can obviously improve the precision of diffraction imaging.The equivalent offset prestack migration method and its parallel method were introduced,and a model case was used to illustrate the efficiency of EOM.At last,this method was applied to the real data migration processing in Tahe area.A comparison with the conventional PSTM method shows that,in the EOM profile,the focusing of the carve-fracture diffractions,the images of faults and breakpoints and the lateral resolution have been improved.

In recent years, with the implementation of three-dimensional deep geological mapping and shale gas bearing basin prospective evaluation, magnetotelluric method, as one of geophysical methods, has become more and more important. There are a finite number of challenging problems in magnetotelluric data processing and interpretation such as impedance tensor rotation, TE and TM mode identification, static distortion correction, and selection of TE and TM for 2D inversion. Based on magnetotelluric module in WinGlink software, the authors adopted five impedance tensor rotation methods and three curve smoothing methods to process the magnetotelluric data gathered in Jarud Banner, Inner Mongolia. Three stations with different qualities in one profile were chosen to process. The results of five rotation methods and three smoothing methods were analyzed and assessed. Polar diagram analysis of three stations showed that the geological models under themselves were three-dimensional. The comparative study is critical for three-dimensional geological models, and is especially much instructive to the processing and interpretation of magnetotelluric data by using WingGlink software.

Aimed at solving the problem of precision calculation and assessment in median region terrain correction, this paper puts forward a method for computing median region terrain correction precision by rotating digital elevation model (DEM) data and measured data,and discusses the feasibility of the method and the precision of the rotated data. At last, the precision of the 1：50 000 DEM data is evaluated with a high accuracy of 1：10 000 DEM data. The Comparative analysis shows that the rotation method is feasible and practical, and that the precision of the terrain correction can meet the requirement of 1：50 000 gravity exploration.

1D forward of transient electromagnetic response of central loop configuration on sea floor was conducted first, in which the Hankle transform involved with the expression for vertical magnetic component in frequency domain was numerically calculated using filter coefficients of 124 points, and frequency-time domain transform was performed by using cosine transform method which could be adaptive within a wide range of time. And then, the algorithm of all-time apparent resistivity was employed, which is based on Continued Fractions expression. Several representative models were calculated and curves of apparent resistivity versus depth were drawn. The results show that these curves can generally describe the variation of resistivity under seabed, and the depth of conductivity/resistivity layers can be estimated according to the extreme points, thus providing a fast and efficient approach for the interpretation of marine TEM data and the assessment of original models.

This paper proposes a particle swarm optimization inversion algorithm for avoiding the dependency on initial model and local solution. This algorithm is applied to induction logging inversion on the models of different thickness layers, and yields consistent results with the models in the noise-free case. When noises of 5%, 10% and 20% are added to the models, the results of inversions remain fairly good. Numerical experiment results demonstrate that this particle swarm optimization inversion algorithm has advantages of being independent of initial models, capable of global optimization and anti-noise, and making induction logging data inversion more effective.

Among all MT Inversion methods used at present,there are two sorts of MT Inversion software which have mature system and have been widely used in engineering:One is the IMAGEM software applying inversion method of "Bostik", and the other is the WINGLINK software applying inversion method of "NLCG". There exist obvious differences between the two sorts of software in such aspects as inversion theory, operation system, smoothing pattern static correction and inversion result. On such a basis,the authors analyzed the inversion interpretation of EH-4 magnetotelluric data obtained in a tunnel of Lanyu Railway. The contrast and analysis of the data processed with the two sorts of software are demonstrated in this paper.

Combining the engineering practical experience with theoretical knowledge, this paper has summarized the problems that should be tackled when the ground penetrating radar is employed in such aspects as railway tunnel lining thickness, cavity behind the lining, internal reinforcement and steel frame detection, especially the detection accuracy of the lining thickness, the interpretation of the second rows of steel in the cases of double layers reinforcing and steel frame detection when the lining has been reinforced. This paper aims to show the actual detection capability of the ground penetrating radar at the current level of science and technology with the purpose of helping the engineers and technicians to have a further cognition of the problems that can be solved by radar.