3D Resistivity Inversion Research Articles

Coal Mine Goaf Interpretation: Survey, Passive Electromagnetic Methods and Case Study

Coal mine goaf detection remains confronted with the lack of fast, effective and low-cost exploration means, especially for the accurate prediction of mining threats primarily caused by hydraulic infiltration. The rapid popularization of passive electromagnetic methods has contributed greatly to improving the interpretation effects of different types of goafs. This paper, firstly, summarizes the pros and cons of various exploration methods in goaf detection. Then, the feasibility of goaf detection using novel passive electromagnetic methods (e.g., the super low frequency alternating magnetic component method (SLF) and audio frequency magnetotelluric method (AMT)) is proposed and further discussed. With well-designed geo-electrical goaf models, the theoretical results demonstrate that the semi-quantitative interpretation of SLF responses can be directly used for the delineation of the target layer in the estimated depth range. In contrast, 3D inversion provides more information about conductive targets with the appropriate initial model selection. Then, shallow, low-resistive targets can be more accurately allocated in the inversion maps. Moreover, the real data interpretation results from study areas demonstrate that the SLF method can utilize the magnetic component responses to effectively identify the fault structures, and indirectly contributes to judge the goaf collapse locations in favor of describing the potential distribution of fracture water infiltration. Combined with the three-dimensional (3D) resistivity inversion of AMT data, the low-resistive water-rich areas within the depth of 400 m were revealed. The inverted depth distributions are basically consistent with those of the water-filled goafs and surrounding layers, which were also confirmed by known logging data. The detailed delineations of water-control fracture zones can be inferred to relate to aquifers in some mining areas; this can reveal potential collapses that require successive mining planning. In specific working faces, goaf risks have been handled in advance by strengthening the continuous monitoring of the water level and water inflow. The above verification has laid a theoretical and practical foundation for passive electromagnetic interpretation methods for effectively predicting collapse-type risks or hydraulic threats in coal mine goafs.

Structural setting of the Zoujiashan-Julong’an region, Xiangshan volcanic basin, China, interpreted from modern CSAMT data

The Zoujiashan-Julong’an area is situated on the western side of the Xiangshan volcanic basin. The basin represents the central part of the Jiangxi Province and it is well endowed with uranium deposits. Ongoing brownfield exploration requires a better understanding of the deep-seated geological structures of the region so as to discover additional, but concealed, deposits at depth. Aiming at elucidating the deep structural and geological framework, 14 parallel controlled-source audio-frequency magnetotelluric (CSAMT) survey lines were conducted to document the relationships between lithological contacts, faults and uranium mineralization with a penetration depth of about 2 kms. By integrating the interpreted geological profiles with geological surface maps and drilling data, a 3D geological model was constructed by using GOCAD software. The 3D model illustrated the major lithological units, faults and concealed uranium mineralization, respectively. The obtained data show that: (1) The Ehuling Formation as well as the underlying metamorphic basement displays a medium-to-high resistivity, whereas the Daguding Formation has a low-resistivity. The unconformity surface between volcanic rocks and the metamorphic basement is represented by a distinct low-resistivity zone which is locally intersected by faults. The fault structure in the CSAMT inversed sections appears as a low resistivity zone or a high-low resistivity variation gradient zone cutting through different interfaces. (2) As is documented by the 3D geological model, the favorable sites for uranium mineralization are distributed in the depressions of the metamorphic rock basement, at the boundary of volcanic rock formations, and in the transitional zones between the depressions and uplifts, where the thickness of the strata increases (or decreases) and the composite faults exist. (3) The location and spatial distribution of the unexposed structures delineated by the 2D resistivity inversion results are consistent with the geological mapping results. The delineated interface between porphyritic lava and rhyodacite is basically consistent with the drilling data. The CSAMT detection results can provide reliable deep geological structure information for 3D geological mapping and deep metallogenic prediction in Xiangshan area.

Geophysical and contamination assessment of soil spatial variability for sustainable precision agriculture in Omu-Aran farm, Northcentral Nigeria

The spatial and temporal variability of soil properties (fluid composition, structure, and water content) and hydrogeological properties employed for sustainable precision agriculture can be obtained from geoelectrical resistivity methods. For sustainable precision agricultural practices, site-specific information is paramount, especially during the planting season. An integrated one-dimensional (1D) and two-dimensional (2D) electrical resistivity survey have been adopted to characterize the subsoil parameters and delineate the aquifer unit of large farm areas, especially in precision agricultural practices. Also, contamination assessment reveals the soil quality status of farmlands. This study aims to determine the site-specific soil parameters of a commercial farm in Omu-Aran, Northcentral, Nigeria. The subsoil features from the geoelectrical resistivity surveys indicate 3 to 4 distinctive lithology to a depth of 43.4 m into the subsurface of the farm. The ID (Vertical Electrical Sounding) and 2D resistivity inversion models results have revealed the heterogeneity nature of the topsoil, also known as the stone zone comprising of reworked clayey soil and sandy gravelly soil, the weathered/saprolite zone (gravelly sandy/sandy soil), the fractured basement and the fresh basement rock. Contamination factor (Cf), pollution load index (PLI) and Nemerow integrated pollution index (NIPI) were used to assess the contamination index on the farmland. Toxic elements such as arsenic, cadmium, chromium, cobalt, lead, manganese, nickel, and zinc have low to moderate contamination in the farm. The depth of investigation (≤3m) covers the upper root zone of significant crops grown in the area. The findings can assess soil contamination, delineate basement features, subsoil variability, soil profiling, and determine the subsoil hydrological properties.

3D multi-scale resistivity inversion method applied in the tunnel face to borehole observations for tunnel-ahead prospecting

Many adverse geological structures, such as karst caves and faults, may exist in front of the tunnel face. If adverse geological structures ahead of the tunnel face cannot be detected clearly, geological hazards (i.e., the inrush of clay and water) may occur during excavation. In this study, a new ahead prospecting scheme was proposed to detect adverse geological structures during tunnel construction using probe holes. First, a new array was proposed: the potential electrodes were placed at the tunnel face, and the current electrodes were placed in the hole. The potential data were collected and used for inversion. Furthermore, a 3D resistivity multi-scale inversion method was proposed to locate and describe boundaries of geological structures with high accuracy. In this method, the 3D convolutional wavelet transform is used to extract wavelet parameter information from the resistivity model. On this basis, the inverted solution space is transformed from the spatial domain to the wavelet domain. The solution in the wavelet domain can be obtained using the inversion equation. Then, the solution in the spatial domain can be obtained from the wavelet domain using the 3D deconvolution transform. Finally, the effectiveness of the new ahead prospecting scheme was verified by numerical inversion tests. Furthermore, this paper describes the successful application of the new scheme to detect water-bearing structures in tunnel of the project “transfer water from Hongyan river to Shitou river”.

2D Electrical Resistivity Imaging of Bedrock Fissures in Ogbomoso, Southwestern Nigeria

2D electrical resistivity surveys were conducted around the site of the failed proposed Ogbomoso North Secretariat building with a view to examining the trend of suspected bedrock fissures and assessing the vulnerability of structures in the vicinity to potential failure. Electrical Resistivity Tomography data were acquired along ten traverses 80-200 m long each, using the dipole-dipole electrode configuration with station interval of 5 m and expansion factor, n, varied from 1 to 6. The data were processed by using 2D resistivity inversion technique in the DipproTM software package to generate 2D resistivity sections beneath the traverses. The 2D resistivity sections delineated 2-19 m thick typically clay overburden underlain by bedrock with resistivity ranging from 103 to 59767 Ωm, and anomalously low resistivity zones suspected to be fissures within the bedrock. The bedrock fissures are generally 5-20 m wide and occur at depths ranging from 5 m to ˃25 m beneath the traverses. The fissures trend southward toward the roundabout and front of the College of Health Sciences premises. The incessant road failures and groundwater seepages observed within the study area are attributable to the network of bedrock fissure.

3D spatial characterization of sand body for uranium reservoirs based on geostatistical resistivity inversion

The concern over climate change and a transition away from fossil fuels has reinvigorated interest in using nuclear energy in producing electricity. Uranium is the principal fuel used in nuclear power plants, and mineral deposits containing this element are of strategic importance. The successful development of sandstone uranium deposits benefits from 3D geophysical characterization of sand bodies in uranium reservoirs. In view of this, we have evaluated a method based on 3D geostatistical resistivity inversion. First, we analyze the application of that method to the problem at hand and introduce a workflow for analyzing the data. Second, through petrophysical sensitivity analysis, we identify the logging parameters that can characterize sandstone in this context and geostatistical inversion to estimate those parameters. We then use 3D geostatistical resistivity inversion to map a sandstone uranium reservoir and find it to provide accurate results. Finally, the 3D data of inversion are applied to 3D spatial characterization of a sand body in the Sifangtai uranium reservoir inverting a field data set in the western of Songliao Basin in China. We believe this geostatistical inversion method to be useful in determining the location of drilling wells for exploration and development of sandstone uranium deposits.

3D characterization of archaeological remains buried under obstacles using annular array electric resistivity tomography

When archaeological remains are buried under buildings or other obstacles, it is very difficult to detect them by performing the traditional electric resistivity tomography (ERT) which commonly sets out survey lines right above those remains. We presented an annular array ERT method to solve the above problem. All electrodes were arranged around the obstacles in an annular array form. The apparent resistivity data were acquired through a special electrode shifting design. The electrode moved in the same serial number order with the cross diagonal survey of a 3D rectangular array. ERT apparent resistivity data were inverted by a 3D resistivity inversion package named RESINVM3D. The proposed method was tested using a synthetic model and field data. Synthetic results show that the target buried under obstacles can be imaged through the annular array. The result of field case indicates grave chamber buried under grave mound can be characterized using the proposed method. We conclude that the annular array ERT is an effective and valuable scheme for the detection and characterization of archaeological remains buried under obstacles.

Geophysical and Contamination Assessment of Soil Spatial Variability for Sustainable Precision Agriculture in Omu-Aran Farm, Northcentral Nigeria

The Spatial and temporal variability of soil properties (fluid composition, structure and water content) and hydrogeological properties employed for sustainable precision agriculture can be obtained from geoelectrical resistivity methods. For sustainable precision agricultural practices, site-specific information is paramount, especially at the planting season. An integrated one-dimensional (1D) and two-dimensional (2D) electrical resistivity surveys have been adopted to characterize the subsoil parameters and delineate the aquifer unit of large farm area, especially in precision agricultural practices. Also, contamination assessment reveals the soil quality status of farmlands. This study aims to determine the site-specific soil parameters of a commercial farm in Omu-Aran, Northcentral, Nigeria. The subsoil features from the geoelectrical resistivity surveys indicate 3 to 4 distinctive lithology to a depth of 43.4 m into the subsurface of the farm. The ID (Vertical Electrical Sounding) and 2D resistivity inversion models results have revealed the heterogeneity nature of the topsoil also known as the stone zone comprising of reworked clayey soil and sandy gravelly soil, the weathered/saprolite zone (gravelly sandy/ sandy soil), the fractured basement and the fresh basement rock. Contamination factor (Cf), pollution load index (PLI) and Nemerow integrated pollution index (NIPI) were used to assess the contamination index on the farmland. Toxic elements such as arsenic, cadmium, chromium, cobalt, lead, manganese, nickel, and zinc have low to moderate contamination in the farm. The depth of investigation (≤ 3m) covers the upper root zone of significant crops grown in the area. The findings can assess soil contamination, delineate basement features, subsoil variability, soil profiling, and determine the subsoil hydrological properties.

Proterozoic tectonothermal processes imaged with magnetotellurics and seismic reflection in southern Australia

Over the last two decades, co-located seismic and magnetotelluric (MT) profiles provided fundamental geophysical data sets to image the Australian crust. Despite their complimentary nature, the data are processed and often interpreted separately without common processes in mind. We here qualitatively compare 2D resistivity inversion models derived from MT and seismic reflection profiles across a region of Archean–Proterozoic Australia to address the causes of variations in seismic response and anomalous conductivity in the crust. We find that there exists a spatial association between regions of low reflectivity in seismic sections and low resistivity in co-located 2D MT modelled sections. These relationships elucidate possible signatures of past magmatic and fluid-related events. Depending on their diffuse or discrete character, we hypothesize these signatures signify fossil melting of the crust due to mafic underplating, magma movement or hydrothermal fluid flow through the crust. The approach discussed herein is a process-oriented approach to interpretation of geophysical images and a significant extension to traditional geophysical methods which are primarily sensitive to a singular bulk rock property or state.

Imprints of tectonic processes imaged with magnetotellurics and seismic reflection

Co-located seismic and magnetotelluric (MT) profiles provide fundamental geophysical data sets to image the crust of Australia. Despite their overlapping nature, the data are processed and interpreted separately based on legacy workflows. We qualitatively compare 2D resistivity inversion models derived from MT and uninterpreted seismic reflection profiles across Proterozoic Australia to address the long-standing cross-cutting nature of interpreted seismic faults and low resistivity zones derived from MT. We find that a good correlation exists between high/low reflectivity in seismic sections and low resistivity in MT sections. These relationships elucidate signatures of past magmatic and fluid-related events and constrain zones of weakened rheology in the crust. Depending on their characteristics, these signatures may signify fossil melting of the crust due to underplating or magmatic invasion into the crust or reworking associated with redistribution of fluids along newly developed faults. These findings have implications for constraining mineral deposit genesis and location.

The use of geological-based geophysical surveys for groundwater distribution in crystalline basement terrain, SW Nigeria

This research involves the subsurface geological characterization for groundwater potential assessment within the campus of the Polytechnic of Ibadan, southwestern Nigeria. The study is directed towards groundwater resources exploration, development and management in the campus. Five 2D resistivity imaging traverses were conducted using Wenner array in addition to five VES surveys using Schlumberger array that provide layering information and geoelectrical parameters. Three geologic layers delineated from the 2D resistivity inversion models include predominantly clayey sand/sandy clay top soil (overburden), partly weathered or fractured basement and fresh basement. Their inverse model resistivity values ranges 6.68 − 98.6 Ωm, 68.0 − 929 Ωm and ≥ 2252 Ωm with bottom depths ranges 3.8 − 6.4 m and 6.4 − 10 m respectively. 1D model inversion from VES results also delineate three lithologies classifying both topsoil and some part of the partly weathered basement as overburden with resistivity and thickness range 483 − 1746.9 Ωm, 1.1 − 1.8 m; partly weathered or fractured basement 60.3 − 93.5 Ωm, 8.4 -12.9 m and fresh basement 984.6 − 2078.9 Ωm . The saturated portion of the partly weathered or fractured basement at depth will favour groundwater exploration and development in this area, while the relatively shallow overburden thickness would serve as the protective layer and recharge for the fractures.

Aquifer Characterization and Groundwater Potential Evaluation in Sedimentary Rock Formation

This study was conducted to characterize the aquifer and evaluate the ground water potential in the formation of sedimentary rocks. Electrical resistivity and drilling methods were used to develop subsurface soil profile for determining suitable location for tube well construction. The electrical resistivity method was used to infer the subsurface soil layer by use of three types of arrays, namely, the pole–dipole, Wenner, and Schlumberger arrays. The surveys were conducted using ABEM Terrameter LS System, and the results were analyzed using 2D resistivity inversion program (RES2DINV) software. The survey alignments were performed with maximum electrode spreads of 400 and 800 m by employing two different resistivity survey lines at the targeted zone. The images were presented in the form of 2D resistivity profiles to provide a clear view of the distribution of interbedded sandstone, siltstone, and shale as well as the potential groundwater zones. The potential groundwater zones identified from the resistivity results were confirmed using pumping, step drawdown, and recovery tests. The combination among the three arrays and the correlation between the well log and pumping test are reliable and successful in identifying potential favorable zones for obtaining groundwater in the study area.

A compressed sensing based 3D resistivity inversion algorithm for hydrogeological applications

Image reconstruction from discrete electrical responses pose a number of computational and mathematical challenges. Application of smoothness constrained regularized inversion from limited measurements may fail to detect resistivity anomalies and sharp interfaces separated by hydro stratigraphic units. Under favourable conditions, compressed sensing (CS) can be thought of an alternative to reconstruct the image features by finding sparse solutions to highly underdetermined linear systems. This paper deals with the development of a CS assisted, 3-D resistivity inversion algorithm for use with hydrogeologists and groundwater scientists. CS based l1-regularized least square algorithm was applied to solve the resistivity inversion problem. Sparseness in the model update vector is introduced through block oriented discrete cosine transformation, with recovery of the signal achieved through convex optimization. The equivalent quadratic program was solved using primal-dual interior point method. Applicability of the proposed algorithm was demonstrated using synthetic and field examples drawn from hydrogeology. The proposed algorithm has outperformed the conventional (smoothness constrained) least square method in recovering the model parameters with much fewer data, yet preserving the sharp resistivity fronts separated by geologic layers. Resistivity anomalies represented by discrete homogeneous blocks embedded in contrasting geologic layers were better imaged using the proposed algorithm. In comparison to conventional algorithm, CS has resulted in an efficient (an increase in R2 from 0.62 to 0.78; a decrease in RMSE from 125.14 Ω-m to 72.46 Ω-m), reliable, and fast converging (run time decreased by about 25%) solution.

Electrical conductivity structure across the Parnaíba Basin, NE Brazil

Abstract Magnetotelluric (MT) data spanning from 0.001 to 50 000 s were acquired across the Parnaíba cratonic basin along an east–west profile approximately 1430 km long. Our preferred 3D resistivity inversion model revealed the presence of three distinct lithospheric blocks bounded by major electrical discontinuities. The western block is characterized by mainly resistive crust and upper mantle, comprising the Amazonian Craton and the western border of the Parnaíba Basin. The central block reveals a broad highly conductive anomaly extending to the upper mantle and is entirely within the basin limits. The eastern block extends from the eastern portion of the Parnaíba Basin towards the Borborema Province, and discloses two resistive profound vertical anomalies cutting through the crust and upper mantle, intercalated by a conductive zone. We associate this large conductive zone (>100 km long) with the signature of the Transbrazilian Lineament Shear Zone which appears to have served as palaeofluid pathways. The anomalous bulk conductivity increase in the crust beneath the central part of the Parnaíba Basin is credited to the impregnation of the lithosphere by conducting minerals related to tectonic events involving either Brazilian orogenic processes or dispersed magmatic residues associated with the continuous igneous intrusions that occurred during the Triassic and Cretaceous. Supplementary material: Additional forward modelling tests are available at https://doi.org/10.6084/m9.figshare.c.4157600

Comprehensive surface geophysical investigation of karst caves ahead of the tunnel face: A case study in the Xiaoheyan section of the water supply project from Songhua River, Jilin, China

This paper describes the application of a comprehensive surface geophysical investigation of underground karst systems ahead of the tunnel face in the Xiaoheyan section in the main line of the water supply project from Songhua River, located in Jilin, China. To make an accurate investigation, Surface Electrical Resistivity Tomography (S-ERT), Transient Electromagnetic Method (TEM), Geological Drilling (Geo-D) and Three-dimensional Cross-hole Electrical Resistivity Tomography (3D cross-hole ERT) were applied to gain a comprehensive interpretation. To begin with, S-ERT and TEM are adopted to detect and delineate the underground karst zone. Based on the detection results, surface and in-tunnel Geo-D are placed in major areas with more specific and accurate information gained. After that, survey lines of 3D cross-hole ERT are used to conduct detailed exploration towards underground karst system. In the comprehensive investigation, it is the major question to make the best of prior information so as to promote the quality of detection. The paper has put forward strategies to make the full use of effective information in data processing and the main ideas of those strategies include: (1) Take the resistivity distribution of the subsurface stratum gained by S-ERT inversion as the initial model of TEM inversion; (2) Arrange borehole positions with the results of S-ERT and TEM. After that, gain more accurate information about resistivity of subsurface stratum using those boreholes located; (3) Through the comprehensive analysis of the information about S-ERT, TEM and Geo-D, set the initial model of 3D cross-hole resistivity inversion and meanwhile, gain the variation range of stratum resistivity. At last, a 3D cross-hole resistivity inversion based on the incorporated initial model and inequality constraint is conducted. Constrained inversion and joint interpretation are realized by the effective use of prior information in comprehensive investigation, helping to suppress the non-uniqueness problem of inversion so as to raise its reliability. In this way, a 3D detailed model of underground karst system which is 30m ahead of tunnel face is finally formed. At the end of the paper, there is a geological sketch of the revealed karst caves, which illustrates the effectiveness of the presented strategy. To sum up, in the comprehensive investigation of underground karst caves, the integrated use of prior information can help to yield more accurate and detailed results.

2D magnetotelluric imaging of the Anqing-Guichi ore district, Yangtze metallogenic belt, eastern China: An insight into the crustal structure and tectonic units

Two parallel NW-trending magnetotelluric (MT) profiles were placed perpendicularly to the main structures of the Anqing-Guichi ore district, one of the seven ore districts in the middle–lower Yangtze River metallogenic belt of eastern China. In October–December 2013, the MT data acquisition was carried out at 117 sites with 0.5–1km site spacing. The MT data has a good quality in the frequency range between 320 and 0.01Hz. The dimensionality analysis and 2D resistivity inversion results indicate that: (1) the deep of the ore district with three-dimensional structural characteristics, but two-dimensional structural characteristics for shallow; (2) there is a clear correlation between resistivity and the main geological units of the ore district, as well as correlation with mapped surface faults; (3) the Gandan deep fault (GDF) and Jiangnan deep fault (JNF) extend from the surface to 10km deep, with dip of NW45°, and dip angles larger than 60°. A series of NE-trending acidic intrusive rocks were controlled by the GDF.

3D electrical resistivity inversion using prior spatial shape constraints

To minimize the number of solutions in 3D resistivity inversion, an inherent problem in inversion, the amount of data considered have to be large and prior constraints need to be applied. Geological and geophysical data regarding the extent of a geological anomaly are important prior information. We propose the use of shape constraints in 3D electrical resistivity inversion. Three weighted orthogonal vectors (a normal and two tangent vectors) were used to control the resistivity differences at the boundaries of the anomaly. The spatial shape of the anomaly and the constraints on the boundaries of the anomaly are thus established. We incorporated the spatial shape constraints in the objective function of the 3D resistivity inversion and constructed the 3D resistivity inversion equation with spatial shape constraints. Subsequently, we used numerical modeling based on prior spatial shape data to constrain the direction vectors and weights of the 3D resistivity inversion. We established a reasonable range between the direction vectors and weights, and verified the feasibility and effectiveness of using spatial shape prior constraints in reducing excessive structures and the number of solutions. We applied the prior spatially shape-constrained inversion method to locate the aquifer at the Guangzhou subway. The spatial shape constraints were taken from ground penetrating radar data. The inversion results for the location and shape of the aquifer agree well with drilling data, and the number of inversion solutions is significantly reduced.

3D DC Resistivity Inversion with Topography Based on Regularized Conjugate Gradient Method

During the past decades, we observed a strong interest in 3D DC resistivity inversion and imaging with complex topography. In this paper, we implemented 3D DC resistivity inversion based on regularized conjugate gradient method with FEM. The Fréchet derivative is assembled with the electric potential in order to speed up the inversion process based on the reciprocity theorem. In this study, we also analyzed the sensitivity of the electric potential on the earth’s surface to the conductivity in each cell underground and introduced an optimized weighting function to produce new sensitivity matrix. The synthetic model study shows that this optimized weighting function is helpful to improve the resolution of deep anomaly. By incorporating topography into inversion, the artificial anomaly which is actually caused by topography can be eliminated. As a result, this algorithm potentially can be applied to process the DC resistivity data collected in mountain area. Our synthetic model study also shows that the convergence and computation speed are very stable and fast.

2D and 3D resistivity inversion of Schlumberger vertical electrical soundings in Wadi El Natrun, Egypt: A case study

Abstract The Wadi El Natrun area is characterized by a very complicated geological and hydrogeological system. 45 vertical electrical soundings (Schlumberger array) were measured in the study area to elucidate the peculiarity of this unique regime, specifically the nature of waterless area. 2D and 3D resistivity inversion based on the finite element technique and regularization method were applied on the data set. 2D and 3D model resolution was investigated through the use of the Depth and Volume of Investigation Indexes. A very good matching was found between the zones of high resistivity, the waterless area, and the non-productive wells. The low resistivity zones (corresponding to Lower Pliocene clay) were also identified. The middle resistivity fresh water aquifer zones were recognized. Available results can assist in the aquifer management by selecting the most productive zone of groundwater.

3D resistivity inversion using an improved Genetic Algorithm based on control method of mutation direction

Abstract Traditional inversion method is the most commonly used procedure for three-dimensional (3D) resistivity inversion, which usually takes the linearization of the problem and accomplish it by iterations. However, its accuracy is often dependent on the initial model, which can make the inversion trapped in local optima, even cause a bad result. Non-linear method is a feasible way to eliminate the dependence on the initial model. However, for large problems such as 3D resistivity inversion with inversion parameters exceeding a thousand, main challenges of non-linear method are premature and quite low search efficiency. To deal with these problems, we present an improved Genetic Algorithm (GA) method. In the improved GA method, smooth constraint and inequality constraint are both applied on the object function, by which the degree of non-uniqueness and ill-conditioning is decreased. Some measures are adopted from others by reference to maintain the diversity and stability of GA, e.g. real-coded method, and the adaptive adjustment of crossover and mutation probabilities. Then a generation method of approximately uniform initial population is proposed in this paper, with which uniformly distributed initial generation can be produced and the dependence on initial model can be eliminated. Further, a mutation direction control method is presented based on the joint algorithm, in which the linearization method is embedded in GA. The update vector produced by linearization method is used as mutation increment to maintain a better search direction compared with the traditional GA with non-controlled mutation operation. By this method, the mutation direction is optimized and the search efficiency is improved greatly. The performance of improved GA is evaluated by comparing with traditional inversion results in synthetic example or with drilling columnar sections in practical example. The synthetic and practical examples illustrate that with the improved GA method we can eliminate the dependence on initial model, and obtain high-quality inversion results.