Reservoir Fluid Identification Research Articles

Porosity estimation based on the shear modulus inversion of seismic shear wave

Reliable estimation of subsurface porosity is necessary for hydrocarbon reservoir characterization and fluid identification. Porosity estimations from seismic data can provide the lateral distribution of subsurface porosity, but the results may be highly nonunique because subsurface elastic properties (such as velocity and density) can be affected by both porosity and pore fluids. Since shear modulus is insensitive to pore fluid content, it can be effectively used to estimate porosity. We present a novel porosity estimation method using the reflected SH-wave (SH-SH wave), whose propagation characteristics depend mainly on shear modulus and S-wave velocity. We derive a new analytical expression for the SH-SH wave reflection coefficient, which acts as a function of shear modulus and S-wave velocity in its natural logarithm form. This new approximation has high accuracy, and both coefficients corresponding to shear modulus and S-wave velocity are “model-parameter independent”; thus, there is no need for prior estimation of any model parameter during inversion. Numerical analysis shows shear modulus inverted from the SH-SH wave has lower uncertainty, lower ill-conditioning, and lower requirements of data incident angle than that of the PP-wave. Furthermore, the highly correlated rock physics relationships between porosity and shear modulus facilitate accurate porosity estimation. A field data application shows that high-resolution porosities of fine structures can be reliably recovered using the proposed method.

Reservoir fluid identification based on multi-head attention with UMAP

Gas logging curves obtained during drilling, along with conventional well logs, typically serve as foundational data for identifying reservoir fluids. Traditional methods for identifying layers containing oil or gas often suffer from limitations such as inadequate feature extraction, underutilization of differences in data points, and limited applicability. These limitations lead to reduced accuracy in predicting reservoir fluid segmentation. To address the challenge of spatiotemporal feature extraction in logging curves, this paper presents a fluid identification method based on deep learning networks and multivariate sequence classification. The method takes into account both the sequential variations in geological strata and the relationships among different curves for fluid prediction. Our proposed method utilizes a Transformer multi-head attention mechanism and follows a series of steps. First, an improved Symbolic Aggregate approXimation (SAX) is employed to assess the response characteristics of the logging curves to the target fluid and identify sensitive curves based on information similarity. Subsequently, the semi-supervised learning method UMAP is applied to identify layers with high interpretability. Hierarchical training is then employed to mitigate the impact of noise and high-dimensional nonlinearity on identification results. Finally, the processed vector curves are input into the attention module layer, where features of each layer are extracted in an encoding-decoding manner. This method has been tested and validated using gas logging data from 12 wells in a gas field offshore China. The overall prediction accuracy can exceed 94%, and the prediction accuracy for a single gas layer is over 88%. When compared to other machine learning approaches, such as K-means, decision tree, and XGBoost, this method exhibits superior identification accuracy and generalization ability, especially for unbalanced samples.

Seismic data AVO analysis in time frequency domain using synchroextracting transform

Time-frequency analysis is a crucial tool for processing and interpreting seismic data in hydrocarbon exploration. Accurately determining the temporal and spectral characteristics is essential for applications such as reservoir characterization and fluid identification. High-resolution representation of seismic data time-frequency properties improves reliability in these applications. Recent advancements in post-processing techniques aim to enhance conventional methods. The synchroextracting transform (SET) is an innovative approach that combines a post-processing procedure with the short-time Fourier transform. It retains time-frequency information associated with spectral components while eliminating excessive energy dispersed in the time-frequency domain. This study evaluates the effectiveness of SET for seismic data time-frequency analysis, comparing it with contemporary post-processing techniques using synthetic signals and field seismic data. Results show that SET provides higher resolution for distinct signal components and effectively reduces energy dispersion. SET exhibits advantages such as robustness against random noise, reasonable computational cost, and offering more interpretable time-frequency information regarding attenuated amplitudes. Building on synthetic experiments, we demonstrate the capability of SET to generate high-resolution single frequency sections of seismic data, facilitating accurate analysis of amplitude variation with offset (AVO) in the time-frequency domain.

The bitumen development effect on the pore structure and electrical resistivity of carbonate reservoirs in the Sinian Dengying Formation

Abstract In the Sichuan Basin, bitumen widely distributes in the carbonate reservoir, and the bitumen occurrence morphology affects the pore structure and petrophysics. Two types of bitumen, boundary-filled and centre-filled, were reconstructed using a pore morphology-based method based on dolomite carbonate digital cores from X-ray computerized tomography. The pore structures of the reconstructed bitumen-filled digital core models were determined using an extracted pore network model. The finite element method simulated the electrical resistivity of the digital core model. The investigation showed that (i) boundary-developed bitumen increases the pore size whereas centre-developed bitumen decreases the pore size, and bitumen development increases the shape factor while reducing the coordination number; (ii) for electrical resistivity, boundary-filled bitumen increases the porosity exponent with increasing pore size, whereas centre-filled bitumen decreases the porosity exponent. Bitumen development caused a negative correlation between the porosity exponent and pore shape factor, a logarithmic increase between the porosity exponent and coordination number, and a power increase between the porosity exponent and porosity. Subsequently, a new porosity exponent model related to porosity was proposed based on the curve-fitting method. This new porosity exponent combines the Archie equations to calculate the saturation of the carbonate reservoir in the Sinian Dengying Formation in the Penglai gas area on the northern slope of the central Sichuan paleo-uplift, which fits the sealed core plug saturation well and provides the basis for further reservoir evaluation and fluid identification in the Penglai gas area.

Grounded source transient electromagnetic 3D forward modeling with the spectral-element method and its application in hydraulic fracturing monitoring

A long wire with large current source transient electromagnetic (TEM) monitoring, with a large detection depth, low cost, safety, and environmental protection, has unique advantages in the testing and identification of unconventional reservoir fluid and the evaluation of stimulated reservoir volume. So, the TEM 3D forward modeling method has become a research hotspot. Although the finite-element method (FEM) is a type of numerical algorithm that has been widely applied in three-dimensional (3D) electromagnetic field forward modeling, the efficiency and accuracy of FEM require further improvement in order to meet the demand of fast 3D inversion. By increasing the order of the basis function and adjusting the principle of mesh discretization, the precision of the mixed-order spectral-element (SEM) result will be increased. The backward Euler scheme is an unconditionally stable technique which can ignore the impact of the scale of the time step. To achieve a better description of the nonlinear electromagnetic (EM) response of the grounded source TEM method and to optimize the efficiency and accuracy/precision of the 3D TEM forward modeling method significantly, we proposed the use of 3D TEM forward modeling based on the mixed-order SEM and the backward Euler scheme, which can obtain more accurate EM results with fewer degrees of freedom. To check its accuracy and efficiency, the 1D and 3D layered models are applied to compare the SEM results with the semi-analytical and FEM solutions. In addition, we analyzed the accuracy and efficiency of the SEM method for different types of order basis functions. Finally, we calculated the long-wire source TEM response for a practical 3D earth model of a shale gas reservoir for fracturing monitoring and tested the feasibility of the TEM method in a hydraulic fracturing monitoring area to further demonstrate the flexibility of the SEM method.

Missing well-log reconstruction using a sequence self-attention deep-learning framework

Well logging is a critical tool for reservoir evaluation and fluid identification. However, due to borehole conditions, instrument failure, economic constraints, etc., some types of well logs are occasionally missing or unreliable. Existing logging curve reconstruction methods based on empirical formulas and fully connected deep neural networks (FCDNN) can only consider point-to-point mapping relationships. Recurrently structured neural networks can consider a multipoint correlation, but it is difficult to compute in parallel. To take into account the correlation between log sequences and achieve computational parallelism, we develop a novel deep-learning framework for missing well-log reconstruction based on state-of-the-art transformer architecture. The missing well-log transformer (MWLT) uses a self-attention mechanism instead of a circular recursive structure to model the global dependencies of the inputs and outputs. To use different usage requirements, we design the MWLT in three scales: small, base, and large, by adjusting the parameters in the network. A total of 8609 samples from 209 wells in the Sichuan Basin, China, are used for training and validation, and two additional blind wells are used for testing. The data augmentation strategy with random starting points is implemented to increase the robustness of the model. The results show that our proposed MWLT achieves a significant improvement in accuracy over the conventional Gardner’s equation and data-driven approaches such as FCDNN and bidirectional long short-term memory, on the validation data set and blind test wells. The MWLT-large and MWLT-base have lower prediction errors than MWLT-small but require more training time. Two wells in the Songliao Basin, China, are used to evaluate the cross-regional generalized performance of our method. The generalizability test results demonstrate that density logs reconstructed by MWLT remain the best match to the observed data compared with other methods. The parallelizable MWLT automatically learns the global dependence of the parameters of the subsurface reservoir, enabling an efficient missing well-log reconstruction performance.

Supervised machine learning for discriminating fluid saturation and presence in subsurface reservoirs

This paper describes a workflow that combines machine learning and frequency-related multi-attribute analyses to predict the occurrence of water, oil, and high or low gas saturations in three Miocene Gulf of Mexico clastic reservoirs, the Ursa, King Kong and Lisa Anne fields. The models are trained on labeled data from the Ursa Field dataset and validation is conducted on the King Kong and Lisa Anne Field datasets, comprising four wells with oil and varying gas saturations. We also analyze the impact of fluid-type and saturation on a suite of frequency-related attributes, including independent frequency magnitudes and spectral shape attributes.We also assess feature significance using SHAP Analysis and our results indicate that that high gas saturation is characterized by high magnitudes of low frequencies, increased bandwidth, and lower skewness and kurtosis values. Conversely, low gas saturation is distinguished by high magnitudes of medium frequencies, reduced bandwidth, and higher skewness and kurtosis values.Our workflow demonstrates the power and utility of supervised machine learning for reservoir fluid identification and saturation prediction. Human interpreters relying solely on post-stack seismic data may find the task of reservoir fluid predictions challenging and the results may be compromised by subjectivity. This study underscores the potential of machine learning to improve hydrocarbon exploration and production, providing a more robust approach to prospect risk assessment using post-stack seismic data and frequency attributes.

Noise Characteristics and Denoising Methods of Long-Offset Transient Electromagnetic Method

The advantages of the long-offset transient electromagnetic method include deep detection and sensitive response to resistivity anomalies. It is widely used in underground mineral resources exploration, fluid identification in petroleum reservoirs, hydraulic fracturing, and dynamic residual oil and gas monitoring. After the primary field signal is turned off, grounded electrodes or coils are used to observe the secondary eddy field. The secondary field signal decays quickly and has a large dynamic range and a wide frequency band but is easily affected by various natural and human electromagnetic interferences. Therefore, noise reduction and distortion correction are important issues in the processing of transient electromagnetic data. This paper proposes a systematic noise interference suppression process. Multi-period and positive–negative bipolar signal stackings were used to remove random noise and suppress DC offset signals. Then, a time-domain inverse digital recursive method was applied to remove characteristic frequency signals, e.g., power frequency signals and their harmonic interference. A standard noise-free signal was constructed through forward modeling simulation and verified by adding different types of noise. Finally, high-quality transient electromagnetic secondary field attenuation signals were obtained through overlapping windowing technology. We applied this algorithm to obtain electromagnetic data from dynamic monitoring of hydraulic fracturing in Fuling shale gas and from a copper–iron metal mine in Daye City, demonstrating its effectiveness.

An adaptive ensemble learning by opposite multiverse optimizer and its application in fluid identification for unconventional oil reservoirs

Unconventional reservoirs are rich in petroleum resources. Reservoir fluid property identification for these reservoirs is an essential process in unconventional oil reservoir evaluation methods, which is significant for enhancing the reservoir recovery ratio and economic efficiency. However, due to the mutual interference of several factors, identifying the properties of oil and water using traditional reservoir fluid identification methods or a single predictive model for unconventional oil reservoirs is inadequate in accuracy. In this paper, we propose a new ensemble learning model that combines 12 base learners using the multiverse optimizer to improve the accuracy of reservoir fluid identification for unconventional reservoirs. The experimental results show that the overall classification accuracy of the adaptive ensemble learning by opposite multiverse optimizer (AIL-OMO) is 0.85. Compared with six conventional reservoir fluid identification models, AIL-OMO achieved high accuracy on classifying dry layers, oil–water layers, and oil layers, with accuracy rates of 94.33%, 90.46%, and 90.66%. For each model, the identification of the water layer is not accurate enough, which may be due to the classification confusion caused by noise interference in the logging curves of the water layer in unconventional reservoirs.

Reflection dispersion of seismic waves at the ocean bottom due to mesoscopic-flow loss

Seismic reflections at the interface of heterogeneous porous media are significantly affected by the multi-scale fluid-flow losses. In this work, we study the seismic reflection dispersion from the ocean bottom separating seawater and heterogeneous porous seabed, described by the effective Biot theory where the mesoscopic loss is present. The analytical reflection coefficient is obtained based on the displacement potentials and suitable boundary conditions. Two reflection scenarios, including the water/water-bearing medium contact and water/gas-saturated medium interface, are considered. The shear-wave attenuation is also considered by using a Cole-Cole complex shear modulus. Numerical results showing the variation in reflection magnitude versus frequency and incidence angle are presented, revealing that the P-wave mesoscopic attenuation, complex shear modulus, boundary conditions and permeability can significantly affect the frequency and angle dependence of the reflection coefficients. Reflections at the water/gas-saturated medium interface exhibit quite different behaviors from those at the water/water-bearing medium contact. The open-pore interface gives frequency-dependent reflections, whereas in sealed-pore case, the reflection magnitudes remain unchanged for all frequencies, due to the fact that the interface degrades into a water/elastic solid one. The study can be relevant for reservoir prediction and fluid identification based on the reflected marine data.

Fluid identification by nonlinear frequency‐dependent amplitude variation with offset inversion based on scattering theory

AbstractPre‐stack seismic amplitude variation with offset inversion is a crucial technique in seismic exploration, employed to estimate reservoir elastic parameters and thus reservoir fluid properties. However, traditional amplitude variation with offset inversion methods are based on elastic theory and linear approximation, neglecting the inelasticity of medium and nonlinear theory. To overcome this limitation, a quadratic scattering coefficient equation of the viscoelastic fluid factor is derived, which provides the basis for the equation for amplitude variation with offset inversion. Traditional amplitude variation with offset inversion methods typically neglect seismic dispersion and attenuation, failing to account for the influence of the seismic wave velocity attenuation and frequency variation. Quality factors of P‐ and S‐waves represent the degree of attenuation of seismic waves. To comprehensively address the effects of seismic wave dispersion and attenuation, a novel method called pre‐stack seismic nonlinear frequency‐dependent amplitude variation with offset inversion has been developed. This method builds upon the new quadratic scattering coefficient and is utilized for reservoir fluid prediction. The reliability and stability of the method are verified through synthetic and filed data examples. Further analysis reveals that the method is more reasonable and accurate compared to the traditional linear amplitude variation with offset inversion method. The results demonstrate that the proposed pre‐stack seismic nonlinear frequency‐dependent amplitude variation with offset inversion method can effectively identify reservoir fluids, providing a novel solution for reservoir fluid identification.

Identification of an oil-bearing layer by formation water resistivity: A case study of the Jurassic reservoir, Southwest Ordos Basin

In the southwestern Ordos Basin, the formation water salinity and resistivity of Jurassic sandstone reservoirs are characterized by rapid changes and great variations, which lead to difficulties in the identification of reservoir fluid types and the development and exploration of the hydrocarbons in this area. To explore appropriate methods for reservoir fluid identification provided with complex formation water salinity and resistivity, we have conducted three research steps. First, using formation water testing and well-logging data, the formation water salinity, spontaneous potential, and reservoir resistivity are used to calculate the formation water resistivity. Second, the calculation results were used to separate the reservoir fluid types. Third, the calculated formation water resistivities are classified and the relationship between the formation water resistivity and the oil-producing interval is established. The results indicate that, for oil-producing intervals, the formation water resistivity calculated using the reservoir resistivity is the highest, whereas that using the formation water salinity is the lowest. Based on this understanding, two identification methods of reservoir fluid types are developed, including the qualitative curve overlap method and the quantitative ternary graph method. The curve overlap method can be applied to visually distinguish reservoir fluid types through the differentiation of formation water resistivity. Compared with some earlier quantitative methods, the ternary graph method can achieve an efficiency of 87.7% in identifying the type of reservoir fluid. Finally, the reliability of both methods is verified by case studies.

Azimuthal seismic reflection characteristics with quality factor Q in viscoelastic horizontal transverse isotropic media

Abstract In seismic exploration, a precise description of the seismic reflection property is critical for reservoir prediction and fluid identification. In the study of seismic wave transmission effects, taking into account both the viscoelastic and anisotropic properties of a medium is compatible with the features of the earth. Moreover, it is advantageous to the characterization of complicated reservoirs. According to the elastic medium foundation and the imaginary component with quality factor Q, seismic reflection properties of viscoelastic media are described in the complex domain. The complex wave number is expressed by phase velocity and Q. The attenuation angle is introduced when the complex wave number is represented by a propagation vector and an attenuation vector. The exact velocity and polarization direction of a viscoelastic medium are expressed using a complex stiffness matrix incorporating Q matrix elements. The quasi-Zoeppritz equation for viscoelastic horizontal transverse isotropic (VHTI) media is derived by using boundary conditions based on the wave function of viscoelastic media. The numerical simulation of reflectivity reveals that the reflection coefficient in a viscoelastic medium is clearly different from that of an elastic medium. Moreover, the difference in reflection coefficient in various orientations has distinct characteristics.

Petrophysical Acoustic Characterization of Weathered Granite: A Case Study of Mesozoic Granites in the Coastal Area of Fujian Province, China

In recent years, with the exploration and development of granite buried-hill oil and gas reservoirs, petrophysics research has played an important role in the study of reservoir characteristics and fluid identification. Through analysis of the relationship between the fluid-bearing petrophysical parameters and the reservoir, the seismic response changes caused by reservoir fluid changes can be determined. Mesozoic granites in the coastal area of Fujian Province in eastern China were investigated as the research object of this project. The mineral composition, density, porosity, P-wave velocity, and S-wave velocity of the granite were measured and analyzed by X-ray diffraction, rock density, rock porosity, and rock acoustics methods. Therefore, the granite’s petrophysical properties, fluid response characteristics, and gas sensitivity parameters were analyzed. The result of the study shows that the granite is predominantly monzogranite. According to the type of reservoir space assemblage, the samples can be divided into two types: those containing fracture-dissolution pores and those containing only dissolution pores. All the samples were characterized by medium to high densities and low to extra-low porosity. There was a linear correlation between the P-wave velocity and S-wave velocity under gas and water-saturated conditions. Factors such as P-wave to S-wave velocity ratio, Poisson’s ratio, Lame coefficient, and other parameters of the samples were analyzed, and the threshold values that distinguished the water and gas-saturated states of the samples were measured and determined. In addition, there were negative correlations between the P- and S-wave velocities and porosity. The sensitivities of the petrophysical parameters to the gas capacity from high to low are Ip2 − 2.03 Is2, λ − 0.03 μ, λ, λ/μ, E − 2.03 μ, σ, K/μ, K, Ip, Vp/Vs, Vp, E, μ, Vs, and Is. For granite-buried hill reservoirs, the variation ranges of the parameters, such as the density, porosity, and P-wave velocity, of the fracture-dissolution pore granite samples were larger than those of the dissolution pore samples. The bulk parameters (Ip, Vp, K, λ) and combination parameters (Ip2 − 2.03 Is2, K/μ, λ− 0.03 μ, E − 2.03 μ, λ/μ) of the dissolution pore samples were more sensitive to the gas capacity. The results of this study provide a basis for the geophysical identification of granite-buried hill reservoirs.

Shear wave velocity prediction using bidirectional recurrent gated graph convolutional network with total information embeddings

Shear wave velocity is an essential elastic rock parameter for reservoir characterization, fluid identification, and rock physics model building. However, S-wave velocity logging data are often missing due to economic reason. Machine learning approaches have been successfully adopted to overcome this limitation. However, they have shortcomings in extracting meaningful spatial and temporal relationships. We propose a supervised data-driven method to predict S-wave velocity using a graph convolutional network with a bidirectional gated recurrent unit (GCN-BiGRU). This method adopts the total information coefficient to capture non-linear dependencies among well-log data and uses graph embeddings to extract spatial dependencies. Additionally, the method employs a bidirectional gated mechanism to map depth relationships in both upward and backward directions. Furthermore, the prediction performance is increased by an unsupervised graph neural network to handle outliers and the generation of additional features by the complete ensemble empirical mode decomposition with additive noise method. Finally, the GCN-BiGRU network is compared with Castagna’s empirical velocity formula, support vector regression, long-short-term memory (LSTM), GRU, and BiGRU methods over the North Sea open dataset. The results show that the proposed method performs better predicting S-wave velocity than the other ML and empirical methods.

Thin Reservoir Identification Based on Logging Interpretation by Using the Support Vector Machine Method

A reservoir with a thickness less than 0.5 m is generally considered to be a thin reservoir, in which it is difficult to directly identify oil-water layers with conventional logging data, and the identify result coincidence rate is low. Therefore, a support vector machine method (SVM) is introduced in the field of oil-water-dry layer identification. The basic approach is to map the nonlinear problem (input space) to a new high-dimensional feature space through the introduction of a kernel function, and then construct the optimal decision surface in the high-dimensional feature space and conduct sample classification. There are plenty of thin reservoirs in Wangguantun oilfield. Therefore, 63 samples are established by integrating general logging data and oil testing data from the study area, including 42 learning samples and 21 prediction samples, which are normalized. Then, the kernel function is selected, based on previous experience, and the fluid identification model of the thin reservoir is built. The model is used to identify 21 prediction samples; 18 are correct, and the prediction accuracy reaches 85.7%. The results show that the SVM method is feasible for fluid identification in thin reservoirs.

Experimental Study of Complex Resistivity Characteristics of a Sandstone Reservoir under Different Measurement Conditions

Due to the variability of fluid properties and saturation of reservoirs, large differences in formation temperature and pressure, and the diversity of rock and mineral compositions, the petrophysical response of reservoirs is often complex. This study explored a new method of reservoir fluid identification and evaluation based on the complex resistivity response characteristics of sandstone reservoirs under different measurement conditions. The complex resistivity of the five sandstone samples was measured under normal temperature and pressure and variable pressure, temperature, and formation conditions and under different oil saturations. Furthermore, the reservoir was comprehensively analyzed and evaluated based on the mineral composition, porosity, and permeability parameters. The results show that the resistivity of the sandstone increases logarithmically with pressure and oil saturation but decreases logarithmically with temperature and depth. The polarizability decreases slightly with increasing pressure and increases slightly with increasing temperature. With increasing depth, the polarizability decreases obviously, and with increasing oil saturation, the polarizability decreases moderately. Under different measurement conditions, the complex resistivity data for the sandstone reservoir and the IP parameters extracted through inversion are regular. The results of this study provide a new method for the identification and evaluation of complex reservoir fluids and have important reference value.

An improved seismic fluid identification method incorporating squirt flow and frequency-dependent fluid-solid inversion

Joint inversion of seismic amplitude and frequency information for fluid discrimination works as a popular approach for reservoir fluid identification. However, ignorance of the factor caused by fluid flow and the viscosity of oil and gas in current frequency-dependent inversion approaches leads to the inaccuracy of oil and gas prediction. Therefore, based on squirt flow and viscoelastic theory, a new frequency-dependent viscoelastic squirt-flow fluid factor is established, which is compared with other conventional fluid factors to verify its advantages in fluid detection. In addition, a new linearized reflectivity equation related to frequency-dependent viscoelastic squirt-flow fluid factor is derived. Finally, to verify the feasibility of the new fluid factor in predicting oil/gas, a frequency-dependent elastic impedance inversion method is introduced. Synthetic data and field data examples find that frequency-dependent viscoelastic squirt-flow fluid factor can eliminate the interference of “false bright spots” in conventional fluid factors inversion and has a higher vertical resolution, which demonstrated the effectiveness and stability of our method in fluid prediction.

Deep carbonate reservoir and gas prediction based on multicomponent seismic amplitude attributes — A case study

Multicomponent seismic exploration provides more valuable information for the prediction of underground structures, lithology, fluids, and fractures. Most studies on lithology estimation and fluid description using PP- and PS-wave seismic data mainly focus on relatively shallow targets with high porosity. Focusing on deep carbonate (dolomite) reservoirs with low porosity (generally less than 5%), we first analyze the characteristics of the seismic response difference between PP and PS waves, and then we combine logging data and seismic forward modeling to implement the identification of fluids using the amplitude difference between PP and PS waves. Data sets are acquired over grain beach dolomite reservoirs in the Cambrian Longwangmiao Formation, the central Sichuan Basin, Western China. The reservoirs are filled with different types of fluids (e.g., gas, water, bitumen, etc.). Based on comparisons between seismic responses of PP and PS waves, we observe that, at the location of the gas-bearing layer, PP-wave amplitude exhibits strong peaks and PS-wave amplitude finds relatively weak peaks, which is different from that obtained for the water-saturated layer. However, at the location of bitumen-filled reservoirs, PP- and PS-wave amplitudes exhibit strong peaks, and the reflection amplitude of PP waves is stronger than that of PS waves. Therefore, the amplitude difference of PP and PS waves provides valuable information and feasibility for fluid identification. We verify that the maximum peak amplitude attribute of PS waves may better characterize the distribution of porous dolomite reservoirs than that of PP waves, and using the attribute of maximum peak amplitude difference between PP and PS waves, we may distinguish gas-bearing and water-saturated layers. Comparing with actual drilling results, we conclude that our PP/PS interpretation technique is feasible for fluid identification in deep carbonate reservoirs.

Fluid Identification Derived from Formation Chlorine Measurements and Reservoir Characterization of Tight Carbonate in Sichuan Basin, China

Summary Natural gas production in the Sichuan Basin reached 30×109 m3 in 2020, but the shortfall between this and the production goal of 50×109 m3 in 2025 requires further exploration. The complex mineralogy and low porosity in tight carbonate reservoirs reduce the accuracy of formation water saturation calculations from Archie’s equation, which brings uncertainties to the reservoir characterization. Therefore, it is necessary to incorporate other methods as supplements to methods based on resistivities. In this paper, we outline a method that incorporates wireline-induced gamma spectroscopy, nuclear magnetic resonance (NMR), array dielectric, and borehole images. Spectroscopy is not only used to estimate the mineralogy of the reservoir, but it also provides measurements, such as chlorine concentration and thermal neutron capture cross section (sigma). The amount of chlorine in the formation is proportional to the water volume in the reservoir, hence formation water saturation. Sigma is also an indicator of the formation water saturation. It enables formation water saturation calculation without resistivity measurements. Case studies are presented from carbonate reservoirs in the Sichuan Basin, China. A robust and comprehensive petrophysical description of mineralogy, porosity, pore geometry, free fluid volume, rock type, and formation water saturation is presented. Calculation of formation water saturation from chlorine and sigma proves to be successful in both water-based mud and oil-based mud (OBM) environments. The depth of investigation (DOI) of chlorine from spectroscopy is about 8 to 10 in. for 90% of the signal. The various DOIs of different measurements must be considered when performing the fluid identification. Bound fluid saturation can reach more than 50% in tight carbonate reservoirs. Formation water saturation is not the only factor that determines the fluid type. Free fluid saturation from NMR must also be incorporated. Finally, a robust methodology integrating formation water saturation derived from dielectric and spectroscopy, and free fluid saturation derived from NMR shows great advantage in fluid identification in tight carbonate reservoirs. In this paper, we discuss a novel combination of wireline logging tools for fluid identification in a tight carbonate reservoir in the Sichuan Basin. It reduces the uncertainty when estimating formation water saturation and when resistivity measurements are suppressed in OBM environments. The gas zones identified by the new method have promising predictions of gas production. This workflow can also be applied to other tight carbonate plays in China.