This study identified the Pleistocene depositional succession of the group (A) (marine, estuarine, and fluvial depositional systems) of the Melor and Inas fields in the central Malay Basin from the seafloor to approximately −507 ms (522 m). During the last few years, hydrocarbon exploration in Malay Basin has moved to focus on stratigraphic traps, specifically those that existed with channel sands. These traps motivate carrying out this research to image and locate these kinds of traps. It can be difficult to determine if closely spaced-out channels and channel belts exist within several seismic sequences in map-view with proper seismic sequence geomorphic elements and stratigraphic surfaces seismic cross lines, or probably reinforce the auto-cyclic aggregational stacking of the avulsing rivers precisely. This analysis overcomes this challenge by combining well-log with three-dimensional (3D) seismic data to resolve the deposition stratigraphic discontinuities’ considerable resolution. Three-dimensional (3D) seismic volume and high-resolution two-dimensional (2D) seismic sections with several wells were utilized. A high-resolution seismic sequence stratigraphy framework of three main seismic sequences (3^rd order), four Parasequences sets (4^th order), and seven Parasequences (5^th order) have been established. The time slice images at consecutive two-way times display single meandering channels ranging in width from 170 to 900 m. Moreover, other geomorphological elements have been perfectly imaged, elements such as interfluves, incised valleys, chute cutoff, point bars, and extinction surfaces, providing proof of rapid growth and transformation of deposits. The high-resolution 2D sections with Cosine of Phase seismic attributes have facilitated identifying the reflection terminations against the stratigraphic amplitude. Several continuous and discontinuous channels, fluvial point bars, and marine sediments through the sequence stratigraphic framework have been addressed. The whole series reveals that almost all fluvial systems lay in the valleys at each depositional sequence’s bottom bars. The degradational stacking patterns are characterized by the fluvial channels with no evidence of fluvial aggradation. Moreover, the aggradation stage is restricted to marine sedimentation incursions. The 3D description of these deposits permits distinguishing seismic facies of the abandoned mud channel and the sand point bar deposits. The continuous meandering channel, which is filled by muddy deposits, may function as horizontal muddy barriers or baffles that might isolate the reservoir body into separate storage containers. The 3^rd, 4^th, and 5^th orders of the seismic sequences were established for the studied succession. The essential geomorphological elements have been imaged utilizing several seismic attributes.

The study aims to implement a high-resolution Extended Elastic Impedance (EEI) inversion to estimate the petrophysical properties (e.g., porosity, saturation and volume of shale) from seismic and well log data. The inversion resolves the pitfall of basic EEI inversion in inverting below-tuning seismic data. The resolution, dimensionality and absolute value of basic EEI inversion are improved by employing stochastic perturbation constrained by integrated energy spectra attribute in a Bayesian Markov Chain Monte Carlo framework. A general regression neural network (GRNN) is trained to learn and memorize the relationship between the stochastically perturbed EEI and the associated well petrophysical log data. The trained GRNN is then used to predict the petrophysical properties of any given stochastic processed EEI. The proposed inversion was successfully conducted to invert the volume of shale, porosity and water saturation of a 4.0 m thick gas sand reservoir in Sarawak Basin, Malaysia. The three petrophysical geobodies were successfully built using the discovery wells cut-off values, showing that the inverted petrophysical properties satisfactorily reconstruct the well petrophysical logs with sufficient resolution and an accurate absolute value at the well site and are laterally conformable with seismic data. Inversion provides reliable petrophysical properties prediction that potentially helps further reservoir development for the study field.

Fault detection is an important step in subsurface interpretation and reservoir characterization from 3D seismic images. Due to the numerous and complex fault structures in seismic images, manual seismic interpretation is time-consuming and requires intensive work. We applied a pre-trained CNN model to predict faults from the 3D seismic volume of the Poseidon field in the Browse Basin, Australia. This field is highly structured with complex normal faulting throughout the targeted Plover Formations. Our motivation for this work is to compare machine-learning-based fault prediction to user-interpreted fault identification supported by seismic variance attributes. We found reasonably satisfactory results using CNN with an improved fault probability volume that outperforms variance technology. Therefore, we propose that this workflow could reduce time and be able to predict faults quite accurately in most structurally complex areas.

This study focuses on the sequence stratigraphy and the dominated seismic facies in the Central Taranaki basin. Four regional seismic sequences namely SEQ4 to SEQ1 from bottom to top and four boundaries representing unconformities namely H4 to H1 from bottom to top have been traced based on the reflection terminations. This was validated using well logs information. An onlapping feature on the seismic section indicates a new perspective surface separated between the upper and lower Giant formation, which indicates a period of seawater encroachment. This study focused extensively on deposition units from SEQ4 to SEQ1. The seismic facies, isochron map, and depositional environment were determined, and the system tract was established. This study was also able to propose a new perspective sequence stratigraphy framework of the basin and probable hydrocarbon accumulations and from the general geological aspect, SA-Middle Giant Formation (SEQ3) could act as potential traps.

The application of geostatistics in seismic inversion techniques has been proven somewhat reliable in the delineation of reservoir properties and has recently attracted the attention of many geoscientists. However, there are cases where its prediction returned negative results after drilling. In this research, we re-evaluated a reservoir in Inas Field, whose geostatistical inversion result wrongly predicted sand continuity, resulting in the spudding of a dry hole. When a geostatistical seismic inversion is successfully applied, it provides an increase in seismic resolution and aids the prediction of sand continuity. Although this method relies more on the statistical data from a well because of the limitation of the seismic data in resolving thin geologic features, the spatial variation of reservoir parameters still depends on seismic data, which often have poor resolution quality. Therefore, to investigate the impact of bandlimited data on the geostatistical inversion, we harmonically extended the seismic bandwidth by applying a sparse-layer spectral inversion algorithm to the data. This algorithm increased the seismic data bandwidth from 80 Hz to 180 Hz, and its tuning thickness reduced from 32 m to 10 m at the reservoir interval. The resultant broadband (180 Hz), as well as the original seismic (narrowband of 80 Hz) data, were both used as input to build two separate geostatistical prediction models, respectively. Twenty (20) realizations of these models were generated, ranked into P10, P50, and P90, and the best case was selected for interpretation. These realizations were used to characterize the reservoir lithofacies distribution. When compared, the result of the broadband inversion, facies and sand distribution model showed that the reservoir facies changed towards the location of the dry well. The broadband geostatistical inversion efficiently improved the reservoir characterization process by not only producing an accurate estimation of the lateral extent of the reservoir heterogeneities but also generating outcomes that help us understand why other geostatistical inversion analyses of the target reservoir were misleading. Contrary to the popular assumption, it was discovered that the tuning effects of bandlimited data could affect the result of a geostatistical inversion and result in wrong facies predictions.

Predicting microporosity and permeability in clastic reservoirs is a challenge in reservoir quality assessment, especially in formations where direct measurements are difficult or expensive. These reservoir properties are fundamental in determining a reservoir’s capacity for fluid storage and transmission, yet conventional methods for evaluating them, such as mercury injection capillary pressure and scanning electron microscopy, are resource-intensive. The aim of this study was to develop a cost-effective machine learning model to predict complex reservoir properties using readily available field data and basic laboratory analyses. A random forest classifier was employed, utilizing key geological parameters such as porosity, grain size distribution, and spectral gamma-ray measurements. An uncertainty analysis was applied to account for natural variability, expanding the dataset, and enhancing the model’s robustness. The model achieved a high level of accuracy in predicting microporosity (93%) and permeability levels (88%) using cross-validation and independent holdout evaluation. By using easily obtainable data, this model reduces the reliance on expensive laboratory methods, making it a valuable tool for early-stage exploration, especially in remote or offshore environments. By incorporating uncertainty, the model reflects geological variability more realistically and improves confidence in predictions, particularly when decisions must be made with limited or noisy input data. The integration of machine learning with uncertainty analysis provides a reliable and cost-effective approach for evaluating key reservoir properties in siliciclastic formations. This model offers a practical solution to improve reservoir quality assessments, enhancing decision-making in exploration planning and risk reduction, especially where direct measurements are limited enabling more informed decision-making and optimizing exploration efforts.