Abstract Seismic interpretation of deepwater channel systems requires delineation of architectural elements (AEs) whose acoustic impedance contrasts are often subtle and whose amplitude responses can be non-unique. Standard amplitude displays conflate reflector geometry with lithologic properties, complicating facies predictions and architectural boundary mapping. This tutorial demonstrates a systematic corendering technique using two widely available attributes: cosine of instantaneous phase (cosφ) and relative acoustic impedance (RAI). These attributes address amplitude interpretation limitations by separating complimentary tasks: cos? provides amplitude-independent mapping of reflector terminations and (dis)continuous reflector patterns that define channel stratigraphy, whereas RAI enables property-based characterization of lateral and vertical facies variations. We present a five-step optimization workflow (attribute extraction, corendering methodology, colormap selection, sampling rate adjustments, and spectral balancing) applied to a deepwater channel complex in New Zealand’s Taranaki Basin. Results demonstrate significant improvement in architectural element visibility compared to conventional amplitude interpretation, enabling confident identification of erosion surfaces, channel stacking patterns, and stratigraphic terminations. This optimized visualization approach provides the foundation for systematic interpretation workflows applicable to exploration settings where limited well control and complexity of deepwater channel settings require transparent evaluation of interpretation confidence.
Abstract Ground-penetrating radar (GPR) is a rapid, nondestructive geophysical technique widely used for subsurface defect detection beneath road pavements. However, conventional single-frequency GPR systems face inherent trade-offs between penetration depth and resolution. To overcome these limitations, this study proposed a novel dual-frequency GPR approach for comprehensive subsurface assessment. Numerical simulations were performed to model the electromagnetic responses of typical road defects, such as voids and loose zones, and to establish their characteristic radar signatures. These simulations provided a theoretical framework for interpreting real-world GPR data with higher accuracy. Field validation was conducted using dual-frequency GPR data collected from a road section in Guangxi Province, China. The results demonstrated that high-frequency GPR signals (e.g., 1.6 GHz) achieved superior resolution for shallow defects (<3-m depth), enabling precise delineation of thin layers and small-scale anomalies. By contrast, low-frequency signals (e.g., 400 MHz) exhibited enhanced signal-to-noise ratios and improved detection capabilities for deeper defects (>3 m). By synergistically integrating both frequencies, the proposed system optimized the balance between depth penetration and vertical resolution, facilitating reliable identification of defect types, spatial distribution, burial depths, and lateral extents. This methodology offered a practical and efficient solution for urban road subsurface hazard detection, supporting proactive infrastructure maintenance and risk mitigation strategies.
Abstract The integration of machine learning (ML) into 3D seismic workflows has shown improved geologic interpretability of deepwater channels in the Taranaki Basin, New Zealand. The integration of two novel dimension reduction (DR) techniques, kernel principal component analysis (K-PCA) and uniform manifold approximation and projection (UMAP), showed improved interpretability for these channel systems. K-PCA yielded higher-resolution seismic images for identifying internal channel architectural elements (AEs), whereas UMAP demonstrated improved resolution for identifying lateral AEs when compared to an established DR technique, principal component analysis (PCA). Building on these findings, the focus shifted to determine how the implementation of DR on seismic attributes before conducting unsupervised clustering affects cluster performance. Our results demonstrated that pre-DR unsupervised k-means clustering produced nonunique clusters that exhibit low statistical performance, whereas post-DR clustering with K-PCA and UMAP showed noticeable improvement in statistical metrics and visual separation of clusters. The interpretation of cluster performance required careful consideration of dimensional perspective; data that appeared highly overlapped in 2D visualizations often reveals distinct cluster separation and better cohesion when examined in 3D space. A balanced evaluation framework was proposed that integrated both quantitative metrics (silhouette score and Davies–Bouldin Index) with a qualitative geologic assessment to ensure meaningful interpretation. This study presented a workflow for adopting and using advanced statistical methods from a geoscience perspective. As ML adoption grew within the field, workflows and interpretational considerations presented by geoscientists for geoscientists became increasingly valuable.
Abstract The highly undulating topography, complex surface geologic conditions, and challenging deep exploration environment in Nandan, Guangxi, China, have severely limited the research on shale gas reservoirs in this area. In this study, magnetotelluric (MT) surveys were conducted to investigate the enrichment conditions, controlling factors, and reservoir characteristics of shale gas in this area, with the survey depth reaching 4 km. The results indicated that the strata in the study area were relatively stable with gentle dip angles (<10°) and limited tectonic activity. The shale gas reservoir was hosted in the Lower Carboniferous Luzhai Formation (C1lz) mud shale, which had a thickness of 200–500 m, a burial depth of 300–3000 m, and a resistivity range of 1–7420 Ω·m (average 164 Ω·m). This formation had high organic carbon content (TOC >3% for 85.96% of samples, up to approximately 12%) and thermal maturity typical of southern marine shales, reaching the overmature stage and having large gas generation potential. Its considerable thickness and moderate burial depth provided favorable geologic conditions for shale gas accumulation. The caprocks of the first and second sections of the Luzhai Formation (C1lz) consisted of mudstone of the third section of the Luzhai Formation and tight limestone of the Baping Formation (C1-2b) with minimal erosion, forming effective sealing conditions for shale gas preservation. The strata in the study area experienced a low uplift rate (15.77 m/Ma in the Luzhai area), which prevented severe structural damage and favored the preservation of shale gas. The MT method is highly adaptable to the complex terrain and surface geologic conditions in the Nandan area, and can effectively delineate the deep geologic structure and shale gas reservoir distribution without artificial excitation. In summary, the application of magnetotelluric sounding for shale gas reservoir exploration in the Nandan area of Guangxi has proven to be effective and reliable, and provides important geophysical evidence for shale gas resource assessments and exploration in southern China.
Abstract Seismic waves experience absorption and attenuation by rock strata during propagation, resulting in high-frequency energy loss. A common method to solve this problem is to use a scaling factor to dynamically scale wavelets in the time–frequency domain and use a high-frequency wavelet for replacing the original wavelet to achieve dynamic inverse filtering. However, because a stable scaling factor is used, this method suffers from the problems of band shift and low-frequency energy loss after processing. To overcome these problems, an innovative time–frequency domain dynamic inverse filtering method using an adaptive scaling factor was proposed. This method performs broadband wavelet dynamic inverse filtering by establishing the correlation between the scaling factor and time and frequency and expanding the low-frequency and high-frequency parts of the wavelet. This method can take into account the low- and high-frequency parts of the wavelet during high-resolution processing and provides wide-frequency-band seismic data.
Abstract The geomorphology and internal architecture of a shallowly buried Pleistocene erosive-constructive channel-levee complex on the continental slope of the Niger Delta, Nigeria, named Channel-1A (Ch-1A), was studied using a combination of 3D seismic geomorphology and stratigraphy methods. Seismic facies description, seismic stratigraphy, channel morphology characterization, architectural elements (AEs) delineation, and reservoir modeling were carried out to establish Ch-1A geomorphology, internal architecture, and sediment fills. The study objectives also include the understanding of temporal and spatial evolution of Ch-1A’s internal architecture and the factors responsible for their variations, and demonstration of the impacts of internal architecture complexities on reservoir modeling and development strategies. Observations from this work suggest a strong interplay between channel evolution and a range of controlling factors such as relative sea-level change, the rate and type of sediment supply, and sea-floor morphology. Evolution of Ch-1A was categorized into five stages (from oldest to youngest): (1) channel initiation dominated by sediment bypass and slumping, (2) basal meander loops and lateral migration that resulted in lateral accretion packages (LAPs), (3) evolution into channels that are both meandering and aggrading, (4) late-stage channels that are dominantly aggrading and capped by turbidite mud fills and (5) hemipelagic draping. This study demonstrated the critical roles that a proper understanding of internal architectures and detailed AE delineation play in reservoir modeling. This is critical when building 3D geocellular models with grids, zonation, and cells sizing capable of capturing vertical and lateral variation in geomorphology and heterogeneities before and after upscaling.
Abstract The identification of volcanic lithology is crucial for the exploration and development of oil and gas reservoirs. Current methods primarily rely on core and thin-section data, which, despite their accuracy, are often inefficient, costly, and subjective. To address these limitations, conventional logging curves calibrated with core and thin-section data were used. A random forest (RF) intelligent identification model was established, and a novel ADASYN–KNN–RF ensemble model was further proposed for volcanic lithology identification. The sample numbers of various lithologies were first balanced using the ADASYN oversampling algorithm. Subsequently, KNN and RF models were trained independently. An optimal classification strategy was then determined by evaluating the performance of the two models on different lithologies, and their results were effectively integrated. When applied to actual data, the proposed ensemble model achieved a high identification accuracy of 93%. Comparative analysis demonstrated that the ensemble model outperforms individual models and proved to be more suitable for lithology identification.
Abstract Conventional fault detection methods based on the seismic-amplitude discontinuity face challenges in shale reservoirs. Such methods generally cannot balance antinoise performance and high sensitivity to small-scale faults, thus leading to poor spatial continuity of identified faults. To solve this problem, this paper proposed a planar fault detection method based on seismic horizons to make up for the deficiencies of amplitude-dependent approaches. This method was founded on the principle that the relative displacement of the same stratum on both sides of a fault plane is proportional to the local maximum standard deviation of the seismic horizon. The proposed method was implemented in three steps: step 1: extract the horizon surface from 3D seismic data; step 2: calculate the local standard deviation of the extracted surface; step 3: extract fault features using a multiscale vessel enhancement filtering algorithm based on the computed standard deviation. The proposed method achieved promising results in a shale reservoir in the Sichuan Basin, China. Compared with traditional amplitude-based technologies, this method had more definite geologic implications and enabled more robust identification of planar fault characteristics. The final results served as supplementary evidence for classifying complex faults into different levels.
Abstract A scheme for early amplitude versus angle (AVA) reconnaissance or screening was proposed using the near- and far-angle stacked seismic volumes as inputs when suitable migrated gathers and derived AVA attributes were unavailable. This new scheme, contrary to other screening methods, allowed for an objective and explicit identification of the most used AVA classes by exploiting the unique features of each AVA class under the scope of complex-trace seismic attribute analysis. The method did not require any user parameterization and yielded outcomes that preserved most of the original waveforms and amplitude dynamic range on each dedicated AVA class volume. It also generated a single categorical volume labeled with the AVA classes, which helped accelerate the AVA reconnaissance and reduced project turnaround. The scheme benchmarked positively in the presence of mild effects resulting from some phenomena that commonly degraded the AVA integrity of data, such as near-far event misalignments, signal bandwidth, angle-dependent frequency imbalances, and angle-uncorrelated random noise, as demonstrated on modeled data, achieving a prediction performance greater than 60%–70% in terms of the F1-score validation metric. The model and actual field applications showed that the scheme could be implemented for use on actual seismic data to enhance early amplitude-based interpretations.
Seismic data can provide an intuitive and accurate reflection of stratigraphic information. However, in areas with low-density seismic line coverage, relying solely on seismic profiles to accurately describe the spatial distribution characteristics of faults in the study area is not convincing. This study used two boundary identification methods of gravity and magnetic potential fields: analytical signal amplitude and mean normalized total horizontal derivative, to identify the boundaries of geological bodies in the southeastern Gulf of Mexico basin, based on the lateral heterogeneity of geological structures. Combined with the interpretation results of seismic profiles, the accuracy of the potential field boundary identification was verified, enhancing the rationality of joint gravity, magnetic, and seismic interpretation results for studying the spatial distribution characteristics of faults. The study confirmed that the analytical signal amplitude and the mean normalized total horizontal derivative methods can be effectively applied to fault characterization in areas with insufficient seismic coverage. Multiscale faults identified using various approaches controlled the stratigraphic deposition during the Jurassic and Early Cretaceous periods. This research implemented a method for enhancing the satellite gravity and magnetic anomalies and provided new insights into studying the sedimentary faults and regional tectonic evolution in the southeastern Gulf of Mexico basin.
The Kuqa Depression, a prolific hydrocarbon province in China's Tarim Basin, hosts a dual-sourcepetroleum system with Jurassic coal-bearing and Triassic lacustrine strata. However, the origins and accumulation mechanisms of hydrocarbons in the Dibei Structural Belt remain contentious, particularlyregarding contributions from Jurassic (J(2)kz, J1y) versus Triassic (T(3)h, T(2-3)k) source rocks, compoundedby ambiguities in biomarker interpretations and uncalibrated thermal models. Advanced geochemicalfingerprinting - including sterane distributions, gammacerane indices (GI; ratio of gammacerane to C30hopane), and delta & sup1;& sup3;C isotopes - was integrated with calibrated burial-thermal modeling and structural analysis,which allowed us to clarify source contributions, hydrocarbon charging history, and structural controlson accumulation. Geochemical results demonstrated that Yangxia Formation (Paleogene; a major regionalseal and secondary reservoir unit within the Kuqa Depression) oils exhibit "inverted-L" sterane patterns(C27 < C29) and delta & sup1;& sup3;C values of -26 parts per thousand to -23 parts per thousand, confirming derivation from Jurassic coal measures. Incontrast, Ahe Formation (Jurassic; a major regional sandstone reservoir unit within the Kuqa Depression)hydrocarbons displayed "V-shaped" sterane ratios (C27 > C29), delta & sup1;& sup3;C values of -32 parts per thousand to -30 parts per thousand, and elevatedGI (0.21-0.32), indicative of Triassic lacustrine sources. Notably, the identification of Triassic-sourcedhydrocarbons in deep Ahe reservoirs challenged previous Jurassic-centric models, resolving ambiguitiesthrough multiproxy integration. Burial-thermal modeling, constrained by fluid inclusion homogenizationtemperatures (80 degrees C-160 degrees C), revealed three charging phases: phase I (19-16 Ma, Jidike Fm.), phase II (16-12 Ma, Kangcun Fm.), and phase III (5-1 Ma, Kuqa Fm.), with phase III Himalayan tectonics criticallyreshaping paleo-accumulations into an inverted "gas-below-oil" stratification (phase reversal due to tectonicreorganization). Structural analysis revealed (1) two boundary-fault anticlinal traps (defined by opposednorth- and south-dipping thrust faults) with vertical gas migration in Ahe sandstones and (2) fault-sealed tightgas accumulations controlled by reservoir quality. These findings highlighted the dominant contribution ofTriassic sources to deep gas reservoirs in Dibei and underscored the importance of multiphase tectonicevolution and source-reservoir coupling. This study provided a predictive framework for hydrocarbonexploration in fold-and-thrust belts, advocating prioritized assessment of fault connectivity and Triassicsaline lacustrine source deposits.
Reservoir fluid movement analysis can be enhanced through the integration of 4D seismic attributes and principal component analysis (PCA), showcased through a detailed case study of the C Upper Sand reservoir in the Maui Field, New Zealand. The methodology demonstrates how combining instantaneous, spectral, and geometric attributes can reveal subtle reservoir characteristics that traditional 4D analysis might miss. The case study serves as a vehicle to illustrate the power of this technique, which successfully identifies previously unrecognized fluid barriers, tracks water encroachment patterns, and reveals reservoir compartmentalization affecting production performance. The workflow demonstrates how sweetness and envelope attributes effectively track gas depletion, whereas mean frequency analysis captures water encroachment with remarkable clarity. Spectral bandwidth reveals critical facies variations, including a northeast-southwest trending shoreface deposit acting as a barrier to fluid flow. When integrated through PCA, these attributes provide a comprehensive understanding of the relationship between reservoir heterogeneity and differential fluid movement. This attribute integration technique offers significant advantages over conventional workflows by linking diagenetic cementation patterns to production discrepancies and identifying zones of calcite cementation through the integration of spectral bandwidth and resistivity data. The approach provides a template for improved 4D seismic analysis applicable to similar reservoirs worldwide, with direct implications for optimizing field development planning, well placement, and production strategies.
Shale gas, a key unconventional energy resource, has become increasingly important in global energyproduction. In the current global context, as energy demands rise and the need for sustainable resourcesintensifies, the exploitation of shale gas presents a significant opportunity. In China, shale gas explorationholds great promise, especially in the Ordos Basin, which is considered a major contributor to the nation'sfuture energy portfolio. The Shanxi Formation in the Ordos Basin is rich in shale gas resources, withrecent exploration breakthroughs positioning it as a strategic replacement field for future reserve expansionand production enhancement. To analyze the geological characteristics of shale gas accumulation and itsmain controlling factors of gas-bearing capacity in the Shanxi Formation, this study systematically analyzedthe mineralogical, geochemical, reservoir properties, and adsorption-desorption characteristics of the shaleusing data from three new wells, including well logging, core samples, and experimental testing in theYan'an area. The findings revealed that the Shanxi Formation shale exhibits large single-layer thickness,excellent lateral continuity, and high organic matter abundance with Type III kerogen. Coupled with coalseams, it forms high-quality gas source rocks characterized by high thermal maturity. The shale demonstratedfavorable fracturability and gas storage capacity due to its high brittle mineral content (brittleness index:44.63%) and a well-developed nanoscale pore-fracture network (specific surface area: 3.84 m2/g; pore volume:0.0103 cm3/g). Field validation showed a desorbed gas content average of 1.137 m3/t, and horizontal welltests achieved an absolute open flow exceeding 5.3 & times; 104 m3/d, confirming its potential for commercialdevelopment. This study highlighted the critical roles of organic matter abundance, thermal maturity, andpore structure in controlling gas content, providing a scientific foundation for efficient shale gas explorationand hydraulic fracturing optimization in the Ordos Basin. These results hold significant implications foradvancing large-scale shale gas reserve expansion in the Yan'an area.
Uncertainty quantification (UQ) in well log prediction remains a significant challenge within geoscientific research because conventional methodologies yield single point estimates without addressing the associated predictive uncertainty The absence of uncertainty assessment may result in suboptimal decisions during exploration phases and reservoir characterization. This study addressed this limitation by applying the conformalized quantile regression (CQR) technique to predict missing AC (P-Sonic) acoustic log curves while simultaneously quantifying the uncertainty of these predictions. The analysis used nine Log ASCII Standard (LAS) files from the Volve Field, with exploratory data analysis revealing substantial missingness, particularly in AC (P-Sonic) logs. Pearson correlation analysis identified strong relationships between AC (P-Sonic), density (DEN), and neutron (NEU) logs. The conformal prediction framework partitions the dataset into training, validation, and test subsets, enabling the derivation of statistically valid prediction intervals. The CQR approach achieved AC (P-Sonic) log predictions with 95% coverage. This work demonstrated the value of UQ in well log prediction by providing accurate predictions accompanied by reliable prediction intervals, thereby supporting robust and informed decision-making in exploration and reservoir characterization. The methodology developed herein offers a comprehensive tool for risk management, enabling geoscientists to proactively identify and mitigate potential adverse outcomes associated with well-log data utilization.
Seismic modeling of carbonate reservoir analogs bridges the gap between outcrop and seismic scales. A forward modeling study based on the Cristal cave karst system (S & atilde;o Francisco Craton, Brazil) is presented. A digital outcrop model (DOM) of the Cristal cave, lithostratigraphic columns, and petrophysical measurements were used to build 2D models of cave systems. The shapes, dimensions, and horizontal spacing of the caves were varied so that 12 different scenarios were composed, which may also include silicified halos. The resulting raw seismic sections were processed following a conventional seismic workflow. Taking into consideration only the resolvability along horizontal and vertical directions, four patterns of cave responses are distinguished: (1) tabular string of bead responses (SBRs) caused by caves with relatively small sizes and horizontal spacings, as in the actual DOM of the Cristal cave, which cannot be resolved either horizontally or vertically; (2) short SBRs caused by caves with relatively small sizes and large horizontal spacing so that they can be resolved horizontally but not vertically; (3) superposed diffraction patterns caused by relatively large caves and small spacing so that they can be resolved vertically but not horizontally; and (4)isolated diffraction patterns caused by relatively large caves and large spacing so that they can be resolved horizontally and vertically. Pull-down effects are observed in most of these patterns. Silicified halos enhance the signals of the small caves but attenuate the signals of large caves. A seismic cube was simulated in the scenario of caves with relatively small sizes, but with large horizontal spacing. A strong correlation between the cave system and the SBRs is evidenced by the seismic cube. The insights gained from this study offer valuable guidance for interpreting cave systems in seismic images.
Core data provide valuable in situ information on the chemical and physical characteristics of subsurface formations. Thin sections, for instance, allow us to define rock types with similar mineral composition, lithologies, and pore types (i.e., petrofacies). Petrofacies logs illustrate the stratigraphic variability within a reservoir and are often used to constrain 3D facies and petrophysical-property models. However, core data are often scarce, prompting the exploration of alternative sources. Well logs, with their different vertical resolution, are commonly used to characterize mineralogy, lithology, and porosity. The challenge lies in reconciling the differences between thin sections and well logs. To address this, a machine learning-based workflow is developed. Our goal was to bridge the resolution gap between these two data types and identify subtle variations in rock properties. Specifically, the focus was on collocated cores that provided thin-section-based petrofacies and X-ray fluorescence (XRF) data. Our approach involved two semi-supervised methods: self-training and labeled clustering. By combining XRF data with dimensionality reduction techniques, a reliable classification of thin-section-based petrofacies is achieved. Remarkably, both approaches achieved accuracies exceeding 90% on Sycamore Formation data. Among the dimensionality reduction techniques tested, Uniform Manifold Approximation and Projection (UMAP) produced the most accurate results. This resulted in petrofacies logs that bridge the resolution gap between core-based thin sections and well-log data. Furthermore, integrating semi-supervised methods into routine core analysis offers substantial cost and time savings. These methods enhance stratigraphic correlation, aid in identifying target zones, design horizontal wells, and constrain subsurface models.
Seismic data can provide an intuitive and accurate reflection of stratigraphic information. However, inareas with low-density seismic line coverage, relying solely on seismic profiles to accurately describe thespatial distribution characteristics of faults in the study area is not convincing. This study used two boundaryidentification methods of gravity and magnetic potential fields: analytical signal amplitude and meannormalized total horizontal derivative, to identify the boundaries of geological bodies in the southeasternGulf of Mexico basin, based on the lateral heterogeneity of geological structures. Combined with theinterpretation results of seismic profiles, the accuracy of the potential field boundary identification wasverified, enhancing the rationality of joint gravity, magnetic, and seismic interpretation results for studyingthe spatial distribution characteristics of faults. The study confirmed that the analytical signal amplitude andthe mean normalized total horizontal derivative methods can be effectively applied to fault characterizationin areas with insufficient seismic coverage. Multiscale faults identified using various approaches controlledthe stratigraphic deposition during the Jurassic and Early Cretaceous periods. This research implemented amethod for enhancing the satellite gravity and magnetic anomalies and provided new insights into studyingthe sedimentary faults and regional tectonic evolution in the southeastern Gulf of Mexico basin
To analyze the deep geological structures of a copper-molybdenum polymetallic ore district located in the Shangri-La region, Yunnan Province, China, the audio-frequency magnetotelluric (AMT) sounding method was carried out. This approach accurately determined the electrical and structural characteristics of the subsurface in the mining area. Ore-prospecting indicators based on AMT data summarized the relationship between resistivity anomalies in AMT profiles and different types of orebodies: vein-like or columnar low-resistivity zones at shallow depths corresponding to intermediate- to low-temperature hydrothermal gold, lead, and zinc polymetallic orebodies; areas exhibiting columnar medium or medium-low resistivity characteristics indicating the presence resence of skarn-type or porphyry-type copper-molybdenum polymetallic orebodies, and intermediate-acid igneous rocks (e.g., diorite porphyrite and porphyry) displaying medium resistivity features. Several potential deep concealed orebodies were delineated, indicating that large-scale porphyry-type copper-molybdenum grebodies may exist at greater depths. The findings suggested that the AMT method is an effective geophytical tool for detecting concealed rock (ore) bodies in the Donglufang copper-molybdenum polymetallic and siruilar deposits. *
Water-rich goaf constitutes a primary hazard factor in coal mines, potentially triggering mine instability and ground subsidence. To detect water-rich goaf efficiently, economically, and non-destructively, the extended spatial autocorrelation (ESPAC) method was applied to perform microtremor surveys in the goaf of the mid-deep sections of the Renjiazhuang coal mine in Ningxia. Microtremor signals were obtained using a linear array, and the 2D distribution of subsurface shear wave velocity was inverted and verified against borehole data. The results showed that in the depth range of -200 to 600 m, the low-velocity zone (500-1500 m/s) is closely related to the fissure development and water-rich area, revealing the spatial distribution of hidden disaster-causing factors in the subsurface. Three low-velocity anomalies within the profile were successfully identified by microtremor probing and combined with borehole drainage validation, confirmed the presence of standing water within these anomalous areas. Enhanced application scope and depth of ESPAC methodology in coal mine water-rich goaf detection have been achieved, establishing comprehensive technical support and theoretical frameworks for subsequent water-rich zone risk assessments and mine safety monitoring systems.
Detecting and delineating thin channels and conduits within flow zones in a carbonate aquifer at field scale is a complex processt To achieve this task, an algorithm is applied that inverts sonic logs to determine the secondary porosity and pore aspect ratios of the carbonate structure. This information helps to characterize flow zones at the pore and borehole scales integrated with Formation microimager(FMI), permeability, and micro-resistivity logs. The inversion of the velocity logs for wells MF37 and EXPM1 from the South Florida Water Management District's Port Mayaca Aquifer site shows similar results. Data from both wells are combined to characterize the structure in the inter-well region of 1200 ft. Because the structure information at the well EXPMI defines detailed flow zone characteristics, the analysis is initiated at this well. In the inter-well region, the flow zones are defined by micro-resistivity anomalies and by relating the aspect ratio logs to the pore space observed in the FMI logs. The well logs are displayed and analyzed together to evaluate the connectivity of the flow zones in the inter-well region, where we identified and delineated three flow zones associated with water production zones. The FMI logs are used to describe the pore structure and the fractures intercepted by the wells. The Elemental Log Analysis porosity image delineates the flow zone structure, confirming the presence of the heterogeneous porous limestone formation at the field scale. This interpretation suggests a methodology for monitoring fluid transmission in flow zones by measuring changes in pore structure using seismic reflection.
