MPMatthew Jude Pranter

Matthew Jude Pranter

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Matthew Jude Pranter has a total of 131 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Semi-supervised Workflow to Generate Petrofacies Logs from Thin Sections and XRF Data

    by David Duarte-Coronado, Rafael Pires de Lima, Deepak Devegowda, Matthew J. Pranter

    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 3-D facies and petrophysical-property models. However, core data are often scarce, prompting us to explore 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, we developed novel Machine Learning (ML) methodologies. Our goal was to bridge the resolution gap between these two data types and identify subtle variations in rock properties. Specifically, we focused 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, we achieved a reliable classification of thin-section-based petrofacies. Remarkably, both approaches achieved accuracies exceeding 90% on Sycamore Formation data. Among the dimensionality reduction techniques tested. Uniform Manifold Approximation and Projection (UMAP) yielded the best results. The outcome? 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.

  • Revealing channelized features through multiscale workflows in a mixed carbonate siliciclastic setting, Grayburg and San Andres formations, Midland Basin, TX

    by Laura Ortiz-Sanguino, Heather Bedle, Matthew J. Pranter, Sumit Verma

    Mixed carbonate-siliciclastic channelized systems pose significant challenges for geologic characterization due to their high lithologic variability. Understanding the lithologic distribution is crucial for deciphering basin evolution and guiding ongoing drilling operations. An example of such a mixed channel system is identified in the San Andres and Grayburg formations in the Midland Basin, TX. These units feature siliciclastic-infilled channels cutting across predominantly carbonate shelves. An integrated study of core, well log, and seismic data was conducted to analyze the lithologic variability of the channelized interval and understand its geomorphological evolution. Seismic attributes such as coherent energy, sweetness, and spectral decomposition proved to be the most efficient at enhancing the contrast between clastic and carbonate elements, demonstrating the feasibility of depicting lithologic heterogeneity at a seismic scale. In addition, conventional seismic interpretation and geometric attributes revealed two categories of channel incisions: type I, characterized by V-shaped bases, straight, and mostly oriented in a northeast–southwest direction; and type II, characterized by U-shaped bases, slightly sinuous, and a northwest–southeast orientation. Well log-based litho-density techniques and core descriptions corroborated the seismic observations by illustrating heterogeneity and the siliciclastic nature of the channel in-fills. A 3D lithology model constrained by the previous analyses shows a dominance of siliciclastics in the Lower San Andres, whereas the Upper San Andres and Grayburg are limestone-rich with episodic siliciclastic events (i.e., channel incisions) and dolostone (in the Upper Grayburg). Geomorphological and stratigraphic variations were correlated to changes in sea level and source rock composition. Overall, these insights improve our understanding of siliciclastic deposition in these formations within the Midland Basin, potentially serving as analogs for the Brushy and Cherry Canyon formations in the Delaware Basin.

  • Characterization of seismic-scale petrofacies variability in the Arbuckle Group using supervised machine learning: Wellington Field, Kansas

    by A. B. Caf, D. Lubo-Robles, K. J. Marfurt, H. Bedle, M. J. Pranter

    The Arbuckle Group in southern Kansas has been investigated for carbon geosequestration-related studies. In this study, we evaluated the seismic-scale petrophysically defined facies variability of the Arbuckle Group at the Wellington Field, Kansas, using quantitative seismic interpretation and a supervised Random Forest classification approach. We first defined three petrophysics-based rock types (petrofacies) from core-derived porosity and permeability measurements using the flow-zone indicator approach. Then, using the artificial neural network, we classified these petrofacies in noncored intervals. We observed that petrofacies 1 corresponds to medium and coarse-grained dolomitic packstone, wackestone, and dolomitic breccia with up to 8% porosity and D-scale permeability values. Whereas petrofacies 2 and 3 correspond to argillaceous and fine-grained micritic dolomites and dolomitic mudstones with lower permeability values for a given porosity, with respect to petrofacies 1. Using the common reflection-point gathers, we performed prestack seismic inversion and calculated various amplitude-variation-with-offset (AVO) attribute volumes. We used these elastic properties and AVO attribute volumes as input for estimating supervised seismic-scale 3D petrofacies and petrofacies probability volumes using the Random Forest algorithm. Results reveal the complex distribution of the petrofacies in the candidate injection and baffle zones in the study area, where petrofacies 1 is mainly prevalent within the lower and upper portions of the Arbuckle group, whereas petrofacies 2 and 3 are mainly present in the middle Arbuckle interval. The workflow we present through this study provides the spatial variability of facies distribution that is reflective of the actual lithology and petrophysical properties of the Arbuckle group in the study area with limited well control.


  • Spatial variability of petrofacies using supervised machine learning and geostatistical modeling: Sycamore Formation, Sho-Vel-Tum Field, Oklahoma, USA

    by David Duarte, Rafael Pires de Lima, Javier Tellez, Matthew J. Pranter

    The Sycamore Formation at the Sho-Vel-Tum Field primarily consists of clay-rich mudstones (Mdst) and quartz-rich siltstones. The clay-rich Mdst are mainly composed of clays, quartz grains, some allochems, and detrital organic matter. The siltstones are structureless and are divided into two petrofacies: high porosity and permeability massive calcareous siltstones and low porosity and permeability massive calcite-cemented siltstones. Core and well-log data provide mineralogical, lithologic, and porosity information that is useful to define petrophysical facies (petrofacies) and to create facies logs within the Sycamore Formation. We used the data to establish the Sycamore Formation stratigraphic architecture and to map its spatial variability and reservoir properties. To classify the Sycamore Formation petrofacies in noncored wells, we developed a machine learning-based workflow that compares more than 1800 classification models and selects the best combination of well logs, algorithms, and hyperparameters to predict defined petrofacies. The process includes combinations of well logs that were optimized in four classification algorithms: artificial neural network, K-nearest neighbor, support vector machine, and random forest. To adjust each classifier, we used a grid search and a fivefold cross validation to find the best combination of three hyperparameters to improve results of each algorithm. This workflow allows for the efficient extraction of information from cores at a low cost. After we generated the petrofacies logs in noncored wells, we combined them with multiple constraints to create a 3D petrofacies model for the Sycamore Formation at the Sho-Vel-Tum Field and analyze the stratigraphic and diagenetic controls on petrofacies and its impact in reservoir quality.


  • Quantifying the sensitivity of seismic facies classification to seismic attribute selection: An explainable machine-learning study

    by David Lubo-Robles, Deepak Devegowda, Vikram Jayaram, Heather Bedle, Kurt J. Marfurt, Matthew J. Pranter

    During the past two decades, geoscientists have used machine learning (ML) to produce a more quantitative reservoir characterization and discover hidden patterns in their data. However, as the complexity of these models increases, the sensitivity of their results to the choice of the input data becomes more challenging. Measuring how the model uses the input data to perform either a classification or regression task provides an understanding of the data-to-geology relationships which indicates how confident we are in the prediction. To provide such insight, the ML community has developed local interpretable model-agnostic explanations (LIME) and Shapley additive explanations (SHAP) tools. In this study, we train a random forest architecture using a suite of seismic attributes as input to differentiate among mass transport deposits (MTDs), salt, and conformal siliciclastic sediments in a Gulf of Mexico data set. We apply SHAP to understand how the model uses the input seismic attributes to identify target seismic facies and examine in what manner variations in the input, such as adding band-limited random noise or applying a Kuwahara filter, impact the model predictions. During our global analysis, we find that the attribute importance is dynamic, and it changes based on the quality of the seismic attributes and the seismic facies analyzed. For our data volume and target facies, attributes measuring changes in dip and energy show the largest importance for all cases in our sensitivity analysis. We note that to discriminate between the seismic facies, the ML architecture learns a “set of rules” in multiattribute space, and overlap among MTDs, salt, and conformal sediments might exist based on the seismic attribute analyzed. Finally, using SHAP at a voxel scale, we understand why certain areas of interest were misclassified by the algorithm and perform an in-context interpretation to analyze how changes in the geology impacted the model predictions.


  • Regional stratigraphy and proximal-to-distal variation of lithology and porosity within a mixed carbonate-siliciclastic system, Mississippian strata of northern and central Oklahoma

    by Katherine A. Drummond, Matthew J. Pranter, Michael G. Grammer

    Mississippian carbonate and silica-rich reservoirs of northern and central Oklahoma formed along a regionally extensive carbonate ramp to basin transect. The stratigraphy, lithology, and porosity characteristics of the Mississippian Meramec and Osage series significantly vary as older ramp carbonates prograde southward and transition into younger calcareous and quartz-rich silt deposits of the Anadarko Basin. Lithofacies identified within the northern carbonate-dominated portion of the system commonly include altered chert, skeletal grainstones, peloidal packstones-grainstones, bioturbated wackestones-packstones, bioturbated mudstones-wackestones, glauconitic sandstones, and siliceous shale. Lithofacies within the southern siliciclastic-dominated portion of the system include structureless to bioturbated sandstones, siltstones, and laminated mudstones, each with varying degrees of carbonate content. We have grouped these core-based lithofacies into dominant lithologies/rock types, which tie to well-log properties. Electrofacies classification methods including the artificial neural network (ANN) and k-means clustering predict lithologies in noncored wells. ANN yielded the highest overall prediction accuracy of 85% for lithologies. Core, well log, and lithology log data establish the regional stratigraphic framework. The Mississippian interval of interest subdivides into 16 stratigraphic zones. A depositional-dip-oriented cross section and the associated reservoir models illustrate the proximal to distal and stratigraphic variability of the lithology and porosity. Lithology trends moving from north to south, from older to younger strata, reveal a carbonate-dominated succession capped by diagenetically altered chert northward shifting into a siliciclastic-dominated interval, which increases in clay content southward. Northward, prospective conventional reservoirs developed near cycle tops within diagenetically replaced cherts and cherty limestones associated with subaerial exposure and sea-level fluctuations. Southward, higher total porosity associates with increased clay content linked to the suppression of calcite cement, forming prospective unconventional targets near the bases of depositional cycles.


  • Stratigraphic variability of Mississippian Meramec chemofacies and petrophysical properties using machine learning and geostatistical modeling, STACK trend, Anadarko Basin, Oklahoma

    by Laynie Hardisty, Matthew J. Pranter, Deepak Devegowda, Kurt J. Marfurt, Carl Sondergeld, Chandra Rai, Ishank Gupta, Heyleem Han, Son Dang, Chris T. McLain, et al.

    Mississippian Meramec deposits and reservoirs in the Sooner Trend in the Anadarko (Basin) in Canadian and Kingfisher counties (STACK) play of Oklahoma are comprised of silty limestones, calcareous sandstones, argillaceous-calcareous siltstones, argillaceous siltstones, and mudstones. We used core-derived X-ray fluorescence (XRF) data and established environmental proxies to evaluate the occurrence of specific elements (Al, K, Ti, Zr, Sr, Ca, and Si) and to illustrate their stratigraphic variability. For the Mississippian Meramec, six indicator elements or element ratios serve as proxies for clay (Al and K), detrital sediment (Ti and Zr), carbonate deposits (Sr and Ca), calcite cement (Sr/Ca), and quartz (Si/Ti and Si/Al). We used an unsupervised K-means classification to cluster elemental data from which we interpret three chemofacies: (1) calcareous sandstone, (2) argillaceous-calcareous siltstone, and (3) detrital mudstone. We used random forest to relate core-derived chemofacies to well logs and classify chemofacies in noncored wells with an accuracy of up to 83% based on blind tests. We integrated core-derived XRF, conventional well logs, and chemofacies logs to produce a cross-sectional chemofacies model that trends from the northwest to the southeast across the STACK trend. Chemofacies distribution indicates an upward increase of detrital mudstone from parasequences 1 to 3. Parasequence 3 is capped by a maximum flooding surface. From parasequences 4 to 5, an increase in argillaceous-calcareous siltstone and calcareous sandstone reflects the progradational stacking. Porosity is low in calcareous sandstones primarily due to calcite cement. Water saturation is high in argillaceous-calcareous siltstone, moderate in calcareous sandstone, and low in detrital mudstone. Interpreted biogenic quartz from element profiles corresponds to the calcareous sandstone chemofacies, which can be estimated from well logs and mapped. Effective porosity and water saturation models reflect the stratigraphic variability of chemofacies and rock types and can be predicted within the defined chemostratigraphic framework.


  • Lithofacies, depositional, and diagenetic controls on the reservoir quality of the Mississippian mixed siliciclastic-carbonate system, eastern Anadarko Basin, Oklahoma, USA

    by Fnu Suriamin, Matthew J. Pranter

    In the eastern Anadarko Basin of central Oklahoma, the variability of Mississippian lithofacies, diagenetic products, and reservoir quality is critical for reservoir development. We have investigated lithofacies variability based on sedimentological characteristics and diagenetic alteration through integration of core and thin sections by using optical microscope, scanning electron microscope, energy dispersive X-ray diffraction spectroscope, and electron probe microanalyzer-cathodoluminescence analyses. Based on detailed descriptions of five cores (approximately 260 m [approximately 850 ft]) and analysis of 34 thin sections, we concluded that the Mississippian strata consists of eight lithofacies that represent wave-influenced nearshore with restricted embayment (lagoon) and channel or lobe deposits. We observed diagenetic alteration including calcite cementation, mechanical compaction, albitization, quartz cementation, silicification, dolomitization, Fe-dolomite cementation, pyritization, and dissolution. A paragenesis scheme suggests that quartz cementation occurred earlier compared to albitization and Fe-dolomite cementation. The Fe-dolomite is the latest authigenic mineral formed, whereas the quartz and calcite cement can be attributed to earlier diagenesis. The calcite, quartz, and Fe-dolomite cementation might have potentially increased the brittleness index and frackability of the rocks. The reservoir quality is relatively good in the channel or lobe deposits and is generally poor in the upper shoreface to upper offshore environments. The reservoir quality is significantly reduced by clays, calcite cement, and mechanical compaction. However, the dissolution of calcite cement and detrital grains tends to improve reservoir quality by forming secondary pores. We prove that understanding the characteristic of lithofacies variation, depositional environments, and diagenetic alterations of the Mississippian strata is crucial for optimal development of the Mississippian reservoirs in the eastern Anadarko Basin. We develop a predictive framework that aids in reservoir characterization.


  • Mechanical stratigraphy of Mississippian strata using machine learning and seismic-based reservoir characterization and modeling, Anadarko Basin, Oklahoma

    by Javier Tellez, Matthew J. Pranter, Carl Sondergeld, Chandra Rai, Jing Fu, Heyleem Han, Son Dang, Chris McLain

    The Sooner Trend in the Anadarko (Basin) in Canadian and Kingfisher counties play primarily produces oil and gas from Mississippian strata. The interval consists of interbedded argillaceous mudstones and calcareous siltstones. Such a contrast in rock composition is linked directly to the mechanical stratigraphy of the strata. Brittle (calcareous siltstones) and ductile beds (argillaceous mudstones) are related to the sequence-stratigraphic framework at different scales. We have used seismic and well-log data to estimate and map the geomechanical properties’ distribution and interpret the mechanical stratigraphy of rocks within the Mississippian strata. First, we defined the parasequences that form the main reservoir zones of the Meramecian-Mississippian strata. Once we established the stratigraphic framework, we estimated and compared rock brittleness index (BI) using two independent laboratory-based measurements from the core. The first method, the mineralogical-derived BI, uses mineralogical composition inverted from Fourier-transform infrared spectroscopy analyses, whereas the second method, the mechanical-derived BI, involves measurements of compressional and shear velocities from core plugs. We use the data from core-plug velocity measurements along with well logs and an artificial neural network approach to establish relationships among the geomechanical properties, well logs, and acoustic impedance values. We then applied these relationships to generate 3D geomechanical models constrained to seismic volumes. The resulting grid distributions illustrate the stratigraphic variability of the properties at the parasequence scale. Overall, brittle strata decrease in thickness and abundance basinward as the frequency of interbedded brittle and ductile zones increases and gradually transitions into thin calcite-cemented siltstones and clay-rich mudstones. Analysis of the production performance of selected horizontal wells drilled within the Mississippian strata indicates that the proportion of brittle and ductile rocks along the well path drilled and the drilled area vertical stacking pattern play a significant role in hydrocarbon production for these Mississippian units.