Well logging evaluation of clastic reservoir productivity in the Wulanhua depression, Hailaer basin, China

Productivity evaluations are essential for reservoir characterization and development, particularly in low-porosity and low-permeability clastic systems. In the Wulanhua (WH) depression of the Hailaer basin in China, most reservoirs exhibit porosities below 15% and permeabilities lower than 100 × 10⁻³ μm², posing significant challenges for reliable productivity prediction. To address this, a multisource tiered evaluation approach is proposed herein that integrates mercury injection capillary pressure, nuclear magnetic resonance (NMR), and conventional well logging data. Quantitative analysis shows that reservoirs with mean pore-throat radii greater than 0.12 μm and displacement pressures below 5 MPa generally achieve natural productivities above 20 t/d, whereas those with radii below 0.05 μm require stimulation to reach industrial levels. NMR-based parameters, including an S₂ + S₃ pore fraction exceeding 65% and a T₂ geometric mean time greater than 20 ms, correspond to high-yield zones (>15 t/d). When only conventional logs are available, deep resistivity (>20 Ω·m), low natural gamma (<70 API), and a resistivity multiplication coefficient (AII) exceeding 1 × 10⁶ can effectively be used to discriminate productive intervals. Field validation of this approach demonstrates that the integrated scheme predicts the well productivity with an average deviation of ±15%. The novelty of this study lies in the establishment of a quantitative and multitiered evaluation framework that is adaptable to varying data availability, providing a robust reference for efficient development of tight clastic reservoirs in the WH depression and similar basins.

1 Introduction

The Wulanhua (WH) depression (Chen et al., 2014) is located in Ulanqab City of Inner Mongolia within the Hailaer basin in China; it is bordered by the Damiao depression to the east and Abqi depression to the north. It is classified as an intermontane depression and covers an area of approximately 600 km², with a maximum basement burial depth of approximately 3,000 m, as shown in Figure 1. Structurally, the WH depression comprises two sub-sags: a larger southern sag and a smaller northern sag. Four major positive structures are developed in the southern sag, namely, Tumu’er, Saiwusu, Hongjing, and Hongge’er. Four hydrocarbon-bearing reservoir systems have also been identified in this region, namely, the Tengyi formation clastic rocks, Arshan group clastic rocks, Arshan andesite, and Paleozoic granites. The Tengyi formation and Arshan group clastic sequences are primarily composed of conglomerates, conglomeratic sandstones, argillaceous sandstones, and calcareous sandstones. In the Tengyi formation, the porosity ranges from 0.8% to 38.1%, with the average value being 15.3% and main porosity distribution ranging between 14% and 22%; the permeability values range from 0.044 × 10⁻³ μm² to 1,619 × 10⁻³ μm², with the average value being 96.32 × 10⁻³ μm² and most other values concentrated in the range of (0.5–100) × 10^–3 μm². In the Arshan group, the porosity varies between 1.6% and 38.9%, with the average value being 13.5% and most values distributed within 9.1%–19%; the permeability ranges from 0.084 × 10⁻³ μm² to 349 × 10⁻³ μm², with the average value being 15.27 × 10⁻³ μm² and most other values in the range of (0.1–17) × 10^–3 μm². According to standard reservoir classification criteria, a large proportion of these reservoirs are categorized as low porosity and low permeability. The WH depression exhibits highly complex reservoir characteristics, which pose significant challenges to productivity evaluations (Ou, 1994). Therefore, the study of productivity evaluation methodologies is critically important for the exploration and development of the WH depression.

Figure 1

Figure 1. Oil and gas exploration deployment efforts in the Wulanhua (WH) depression.

The EAR reservoir productivity evaluation is a complex task because the productivity is a result of the combined effects of intrinsic and extrinsic factors. Intrinsically, the productivity is primarily controlled by parameters like movable oil saturation, effective thickness, porosity, and permeability; however, extrinsic factors like reservoir stimulation, production schemes, and engineering operations also exert significant influences on productivity (Elenius et al., 2018; Aminian et al., 2009). Extensive research has been conducted on productivity prediction. Many studies have proposed methods to predict productivity based on permeability (Ou, 1994; Mao and Li, 2000; Xu et al., 2014; Shi et al., 2020). Ge et al. (2003) developed a method incorporating porosity, permeability, oil saturation, and effective thickness to forecast the reservoir productivity. Zhang et al. (2005) introduced a new technique combining well logging data with cable formation testing for productivity prediction. Zheng et al. (2006) combined logging data with geochemical parameters for predictive purposes. Ju et al. (2005) used statistical analysis of the secondary porosity, pore-throat structure parameters, and total porosity to construct productivity prediction equations. Guan (1998) established a productivity interpretation model based on the cementation index and oil saturation, while Liu et al. (2004) highlighted the close correlation between productivity and abnormal pressure in certain blocks. Zhang et al. (2008) employed probabilistic neural networks to model the productivities of fractured low-permeability reservoirs, and Feng et al. (2012) developed a regional productivity empirical formula using block-level productivity indices. Clark et al. (2011) applied logistic growth models to ultralow-permeability reservoirs by incorporating known oil and gas physical volumes to forecast production. Alisheva et al. (2025) developed a three-dimensional hydrodynamic model to enhance reservoir performance prediction using data-driven approaches. Eydinov et al. (2009) proposed an algorithm to reasonably estimate the relative permeability curves, grid-block porosity, and permeability for productivity prediction. Rezaee et al. (2006) established relationships among permeability, porosity, and pore-throat size and further applied these relationships to forecast productivity. Geir et al. (2002) utilized Kalman filtering to update the reservoir models and improve predictive accuracy. Although these methods have achieved certain successes, they fall short of fully meeting the practical needs for objective prediction of reservoir productivity. With the development of artificial intelligence approaches, deep learning and machine learning have emerged as promising tools for productivity predictions (Okon et al., 2021; Wang et al., 2021; Hassan et al., 2020; Mohaghegh, 2011). However, the complex geological heterogeneities of reservoirs pose challenges to model convergence and predictive accuracy. In high-porosity and high-permeability reservoirs where the pore structures are relatively simple and homogeneous, the productivity evaluations are comparatively straightforward. In contrast, low-porosity and low-permeability reservoirs characterized by strong heterogeneities and complex pore structures (Li et al., 2024) present significant evaluation challenges, making it inappropriate to directly apply productivity models or standards from other blocks. Moreover, well logging data are inherently static, whereas productivity is a dynamic property, which necessitates the comprehensive integration of conventional logging data, core test analyses, and nuclear magnetic resonance (NMR) data to provide evaluations that reflect the actual productivities of low-porosity and low-permeability reservoirs more accurately (Liu J. et al., 2018; Zhan et al., 2024; Zhang et al., 2021).

The present study addresses the technical challenges associated with evaluating the productivities of low-porosity and low-permeability reservoirs in the WH depression through several exploratory efforts. In terms of the evaluation methodology, we developed an integrated approach combining conventional well logging, mercury injection capillary pressure (MICP), and digital core NMR techniques to establish a relatively comprehensive technical framework for assessing the productivities of low-porosity and low-permeability reservoirs. This integrated approach improves upon the limited accuracies of traditional single-method evaluations by effectively combining qualitative and quantitative assessments. Regarding the application of technical parameters, we introduce array-induction-resistivity-related parameters as auxiliary indicators for reservoir productivity evaluations, thus enriching the conventional well logging methods for assessing low-porosity and low-permeability reservoirs. For quantitative prediction, we established a reservoir productivity calculation model based on the relationship between the pore-throat radius and reservoir productivity, which is applicable under varying drive pressure conditions while providing theoretical support for the quantitative forecasting of low-porosity and low-permeability reservoirs. In terms of the technologies applied, digital core NMR was employed for productivity evaluations, and a pore structure classification method based on the T₂ spectral characteristics and pore component features was developed to enhance understanding of the reservoir pore structure while improving the predictive accuracy. With regard to the evaluation standards, we established a system tailored to the geological characteristics of low-porosity and low-permeability sandstone and conglomerate reservoirs in the WH depression; this allowed categorization of reservoir productivity into different levels to better support practical development needs. Analyses of representative wells enabled verification of the feasibility and practicality of combining multiple methods. The evaluation methods and standards developed in this study demonstrate strong operability as well as provide valuable reference and guidance for assessing low-porosity and low-permeability reservoirs in the WH depression and similar regions.

2 Reservoir productivity evaluation using well logging

This study employs three productivity evaluation methods—conventional well logging, core analysis, and NMR data—to comprehensively perform qualitative and quantitative assessments of low-porosity, low-permeability reservoirs in the WH depression.

2.1 Reservoir productivity evaluation based on conventional well logging

For both exploration and development wells, the most readily available data are the conventional well logging curves (Lai et al., 2022; Wang et al., 2024). Analysis of the conventional log responses across reservoirs with different productivity levels indicates that high-productivity reservoirs typically exhibit the following logging characteristics. First, the baseline of the microresistivity curve is relatively low, reflecting a relatively clean lithology; conversely, a high microresistivity baseline often indicates higher calcareous content. Significant amplitude differences between the micropotential and microgradient curves reflect the reservoir permeability. Second, such reservoirs show a pronounced negative spontaneous potential anomaly, which also indicates good permeability. Third, low natural gamma-ray values in these reservoirs reflect the low clay content, whereas higher clay content is generally correlated with lower productivity and a relatively smooth curve. Fourth, the relatively high acoustic transit time and compensated neutron readings coupled with relatively low bulk density reflect porosity collectively; in the WH depression, the porosity is typically calculated using the acoustic transit time and compensated density. Fifth, high values of the deep resistivity effectively reflect the hydrocarbon saturation. With the widespread use of array induction resistivity logging, high-productivity reservoirs generally exhibit significant differences in the array resistivity values at various investigation depths, which are correlated with permeability. In practice, resistivity measurements at multiple investigation depths reflect radial variations of the formation fluids and invasion effects. Based on empirical analysis of numerous wells with confirmed productivities, we observed that productive reservoirs exhibited a steeper resistivity gradient from shallow to deep measurements. To quantify this behavior, a dimensionless array induction resistivity multiplication coefficient (AII) was proposed. The AII integrates resistivities from different depths into a multiplicative form, emphasizing their combined sensitivity to permeability and fluid saturation variations; accordingly, the AII is constructed as shown in Equation 1:

$AII=M2RX·\frac{M2RX}{M2R2}·\frac{M2RX}{M2R3}·\frac{M2RX}{M2R6}·\frac{M2RX}{M2R9},(1)$

where M2R2, M2R3, M2R6, M2R9, and M2RX represent the resistivity measurements at investigation depths of 20, 30, 60, 90, and 120 inches, respectively, in units of Ω·m. Generally, when the reservoir productivity is higher, the AII value is larger.

As an example, Figure 2 presents the well log interpretation results for the L1 well. Here, layers 94, 95, and 96 were evaluated using conventional flow tests with a pumping method, which yielded a daily oil production of 3.64 t and a cumulative production of 8.79 t, confirming the presence of oil. Examination of the well logs indicated that the natural gamma-ray values at the reservoir were relatively low at approximately 60 API, reflecting a clean lithology and low clay content. The spontaneous potential curve showed a moderate amplitude difference, suggesting some permeability. The micropotential and microgradient curves also displayed moderate amplitude differences, indicating a certain permeability. The microresistivity baseline was relatively high, suggesting possible calcareous content, while the deep resistivity values reached 18–27 Ω·m.

Figure 2

Figure 2. Well logging interpretation results for the L1 well.

Productivity levels based on the conventional well log data are typically determined using porosity, permeability, clay and calcareous contents, resistivity, and AII value, as summarized in Table 1. In Table 1, the productivity classification is based on whether the reservoir can achieve industrial oil and gas flows. Class I reservoirs are those that possess natural productivity or can produce industrial levels of oil and gas flows after fracturing. Class II reservoirs have relatively low natural productivity or can produce slightly below industrial levels of oil and gas flows after fracturing. Class III reservoirs exhibit low natural productivity and cannot achieve industrial levels of oil and gas flows even after fracturing.

Table 1

Table 1. Classification standards for reservoir productivity based on conventional well logging data.

Figure 3 shows the well logging interpretation results for the L8 well. The microelectrode and spontaneous potential logs indicate that the reservoir possesses a certain level of permeability, while the natural gamma-ray values reflect a relatively clean lithology (Cripps and McCann, 2000). However, the single-layer thickness is low and spontaneous potential curve exhibits negative anomalies. The acoustic transit time ranges from 263 to 340 μs/m, density ranges from 2.13 to 2.47 g/cm³, and resistivity ranges from 12.5 to 35.8 Ω·m, leading to an integrated interpretation as an oil-bearing layer. Oil testing was conducted at the depth interval of 623–667.6 m using pumping with a pump depth of 580 m, dynamic fluid level of 350 m, and frequency of 72 times per day, which resulted in a daily oil production of 44.35 t without water and indicated a high-yield oil layer. The core analysis at 623.4 m showed a density of 2.13 g/cm³, porosity of 19.5%, and permeability of 39.3 × 10⁻³ μm². The coring record at the same depth was used to identify the interval as “gray oil-stained sandstone-conglomerate,” with a clean lithology, minimal clay and calcareous contents, and high oil and gas displays. The high resistivity of the interval further indicates abundant oil saturation.

Figure 3

Figure 3. Well logging interpretation results for the L8 well.

The reservoir section of the Arshan group is mainly composed of conglomeratic sandstone. Owing to the relatively fewer lithological variations, the electrical responses are more significantly influenced by oil saturation. The high-yield fractured layers, industrial oil-flow layers, and low-yield fractured layers exhibit distinct differences in oil content that are clearly reflected in their electrical characteristics. Figure 4 presents the well logging interpretation results for the L5 well. The resistivity of the Arshan group oil layers in this well is notably higher than that of the corresponding oil layers in the L1 well shown in Figure 2 and generally ranges from 27 to 42 Ω·m. The core observations and thin-section analyses indicate that the reservoir is well sorted with minimal calcareous and clay cementation. Analysis of the well log electrical responses shows that the microelectrode readings are low with significant positive deviations, low natural gamma values reflecting a clean lithology, and “box-shaped” characteristics of the well log curves, indicating homogeneity within the layer, minor vertical variations, and good oil saturation. The core analysis at 1,752.79 m shows a density of 2.18 g/cm³, porosity of 17.8%, and permeability of 30.9 × 10⁻³ μm²; at 1,752.98 m, the core has a density of 2.16 g/cm³, porosity of 18.5%, and permeability of 29 × 10⁻³ μm².

Figure 4

Figure 4. Well logging interpretation results for the L5 well.

2.2 Reservoir productivity evaluation using core analysis data

The productivity evaluations using conventional well logging data tend to show good performances for reservoirs with high productivity levels; however, several challenges may arise when predicting productivity with conventional logging data. For example, spontaneous potential curves and microelectrode curves are key indicators for assessing reservoir permeability and consequent productivity. However, a large amplitude difference in the spontaneous potential curve does not always indicate high permeability. This discrepancy may be related to the logging instrument itself and attributable to the spontaneous potential responses arising from three different mechanisms, some of which are unrelated to permeability. Conversely, certain intervals with good permeability may exhibit small or negligible spontaneous potential amplitude differences, often due to the salinity of the mud filtrate being close to that of the formation water or the balance between the mud column and formation pressures. The microelectrode curves may present similar issues and may not reflect the permeability reliably. Other potential problems when using conventional logging data to predict productivity include situations where the porosity calculated from the acoustic transit time or bulk density is relatively high, yet the reservoir shows no natural productivity during the flow tests. In some cases, two reservoirs with similar porosities can exhibit vastly different productivity levels primarily owing to differences in the permeability, which fundamentally reflects pore structure variations. For instance, as illustrated in Figure 5, at a depth of 1,239.78 m in the L1 well, the porosity is 19.7% but permeability is only 3.49 × 10⁻³ µm²; however, at a depth of 1,112.2 m in the L11 well, the porosity is 19.1% and permeability reaches 8,900 × 10⁻³ µm². Such large discrepancies in the permeability at similar porosity values are attributable to significant differences in the pore structure. Therefore, it is not always reliable to calculate the porosity solely from the acoustic transit time or bulk density and then estimate permeability via regression formulas. The availability of reliable pore structure analysis data is particularly advantageous for accurate productivity evaluations.

Figure 5

Figure 5. Comparison of permeability for similar porosity values.

2.2.1 Changes and reactions of major elements

Multiple wells in the WH depression have undergone coring operations, including both drilling and sidewall cores. Drilling cores generally undergo full petrophysical analyses, while some sidewall cores are also analyzed for their petrophysical properties. When the core petrophysical data are available, reservoir productivity classification standards can be established based on the porosity and permeability measurements from the cores in combination with flow test and production data, as shown in Table 2.

Table 2

Table 2. Classification standards for reservoir productivity based on core porosity and permeability data.

The core-derived porosity and permeability data can be directly used to evaluate the pore structure of the core and reservoir productivity thereof (Nabawy, 2025). However, the availability of cored intervals is limited. Therefore, porosity–permeability correlations are often established from the cored sections to calculate the porosity and permeability values for uncored intervals to enable productivity evaluations. Owing to the strong heterogeneity of reservoirs in the WH depression, the porosity and permeability calculations must be conducted separately for different structural zones (Tumu’er, Saiwusu, Hongjing, and Hongge’er) and stratigraphic units (Tengyi formation and Arshan group). The Tengyi formation conglomerate reservoirs have relatively good petrophysical properties, with the porosity ranging from 12.1% to 17.1% and permeability ranging from 1.8 × 10⁻³ µm² to 28.8 × 10⁻³ µm². Among the four structural zones, Tumu’er and Hongge’er exhibit better reservoir quality and thicker sand bodies, whereas Saiwusu and Hongjing are relatively poorer. Based on the NMR logs, digital core analysis, and mercury injection experiments, the Tengyi formation reservoirs in the southern trough of the WH depression show two types of microscopic features: medium-to-low porosity with medium-to-high permeability and large pore throat; medium-to-high porosity with medium-to-low permeability and small pore throat.

The L11x well is located in the Hongge’er structural zone; its Tengyi formation reservoir is composed of conglomerates and shows well-sorted gravel textures in the core samples. Petrographic observations (Figure 6a) reveal that the rocks comprise subrounded to rounded quartz and feldspar grains with limited matrix and weak carbonate cementation. Intergranular pores dominate the reservoir space that is also supplemented by minor dissolution pores. The clay content is low and mainly contains illite–smectite mixed layers. Sedimentologically, the combination of good sorting, relatively high compositional maturity, and presence of medium-to-fine-grained sandstone suggests deposition in a fan-delta front environment. This interpretation is further supported by the upward-fining grain-size trend in the core as well as the presence of parallel bedding and low-angle cross bedding that are commonly associated with distributary-mouth-bar deposits. The pore structure is characterized by medium-to-low porosity, medium-to-high permeability, large pores, and medium-to-coarse pore throats, and the permeability is calculated using the Equation 2:

$k=0.0595e^{0.4844\phi }.(2)$

Figure 6

Figure 6. Petrographic characteristics under thin-section analyses: (a) L11x well of the Tengyi formation; (b) L1 well of the Arshan group.

The L1 and L5 wells are situated in the Saiwusu structural zone; their Tengyi formation reservoirs are also conglomerates but display poorly sorted unequal-grain structures in the core samples. Thin-section analyses of the samples show angular to subangular gravel composition with abundant matrix and stronger carbonate cementation. The pore types include small intergranular and limited dissolution pores, with the pore throats dominated by fine throats. These features, together with the coarser and more heterogeneous grain-size distribution, massive bedding, and debris-flow texture, indicate deposition in a fan-delta plain environment. Such environments typically include proximal distributary channels and sheet-flood deposits that are consistent with the observed poor sorting and high lithologic variability. The pore structure is characterized by medium-to-high porosity, medium-to-low permeability, small pores, and fine pore throats, and the permeability is calculated with the Equation 3

$k=0.2772e^{0.1409\phi }.(3)$

The clastic reservoirs of the Arshan group are mainly distributed in the Tumu’er and Saiwusu zones; their rock types, compositional maturity, and cement compositions are similar to those of the Tengyi formation. The reservoir space is dominated by secondary porosity, but the degree of dissolution is significantly lower than that in the Tengyi formation. Therefore, the conglomerate reservoirs of the Arshan group have relatively poorer petrophysical properties characterized by medium-to-low porosity, medium-to-low permeability, small pores, and fine pore throats. Figure 6b shows the petrographic thin section of the L1 well in the Arshan group, where tightly compacted clastic grains and limited dissolution pores can be observed; the permeability for this reservoir is calculated as with the Equation 4

$k=0.2506e^{0.193\phi },(4)$

where k is the permeability (in 10^–3 µm²) and ϕ is the porosity (in %).

2.2.2 Productivity evaluation using mercury injection data

One important approach to characterizing the pore structure is through the MICP curve (Liu et al., 2018; Jiang et al., 2018; Gu et al., 2024). The MICP analysis provides not only porosity and permeability data but also key parameters like the displacement pressure, mean throat radius, maximum throat radius, median pressure, sorting coefficient, mercury injection saturation, and mercury withdrawal saturation. Therefore, intervals with available MICP data should be given priority when evaluating the reservoir productivity. Based on extensive statistical data, the mean pore throat radius in the WH depression mainly ranges from 0.028 to 0.257 μm. Figure 7 shows the statistical analysis of the MICP data combined with the corresponding reservoir production test data. It is evident from Figure 7 that the productivity level, i.e., fluid production rate, is strongly influenced by the pore structure characteristics. The degree of development of large pores and pore radius have significant impacts on productivity, where higher productivity corresponds to larger pore radii and a higher proportion of large pores within the total pore volume.

Figure 7

Figure 7. Pore throat radius distribution of reservoirs with different productivity levels: (a) natural high-productivity reservoir; (b) natural low-productivity reservoir; (c) fracturing-responsive reservoir.

Analysis of extensive production test data from the WH depression indicates that the mean pore throat radius and displacement pressure of the reservoir are closely correlated with the daily oil production. A quantitative chart relating the mean pore throat radius with the daily oil production was constructed for reservoir productivity evaluation, as shown in Figure 8. From Figure 8, it is evident that the mean pore throat radius can be used to quantitatively assess reservoir productivity. It should be noted that the displacement pressure represents the minimum capillary pressure required for hydrocarbons to enter or move through the pore throats, thus reflecting the tightness of the pore system. A high displacement pressure of approximately 10 MPa indicates a tight reservoir with poor natural connectivity that usually requires hydraulic fracturing or other stimulation to achieve production; contrarily, a low displacement pressure of approximately 0.1 MPa characterizes a naturally productive reservoir. Therefore, two separate empirical models were established herein to describe the relationship between the mean pore throat radius and daily oil production under these distinct reservoir conditions.

Figure 8

Figure 8. Relationship between the mean pore throat radius and daily oil production.

By fitting the measured data, the relationship between daily oil production and mean pore throat radius can be expressed as follows:

$y=22.005x^2+6.7366x-2.6172,(5)$

$y=48.187x^{1.5837},(6)$

where x represents the mean pore throat radius (in μm) and y represents the daily oil production (in m³). Equation 5 is used to predict the daily oil production when the displacement pressure is 10 MPa, while Equation 6 is used when the displacement pressure is 0.1 MPa.

To ensure reliability of the empirical relationships established above, a quantitative error analysis and cross-validation were performed. The discrimination coefficients (R²) of Equations 5, 6 were 0.96 and 0.93, respectively, indicating strong fitting performance. The 95% confidence intervals of the regression coefficients were computed, and all terms were found to be statistically significant (p < 0.01). In addition, a leave-one-out cross-validation was conducted across five representative wells to assess predictive stability. The relative deviations between the predicted and measured daily oil production values were within ±12%, confirming good robustness of the model. Therefore, the established relationships can be considered to be statistically reliable and applicable for preliminary productivity evaluations in similar tight clastic reservoirs.

2.3 Reservoir productivity evaluation using NMR data

Using the MICP data to evaluate productivity (Jing-qiang et al., 2016) is indeed an effective method. However, most MICP core samples are sourced from drill cores and are relatively scarce, so many intervals lack MICP data. Moreover, MICP testing requires toxic mercury as the testing medium. However, NMR data provide more accurate values for parameters like porosity, permeability, and oil saturation compared to other logging methods, in addition to providing relatively precise pore structure information, thereby serving a role similar to MICP data. Currently, at least three approaches exist to obtain NMR data: NMR logging, laboratory NMR experiments, and digital core technology.

2.3.1 Productivity evaluation using digital core technology

Owing to logging costs and other factors, only approximately 30% of the exploration wells in the WH depression have undergone NMR logging. Thus, relying solely on field logging to obtain NMR data is clearly insufficient. Laboratory NMR experiments on cores are possible, but they require processing for oil and salt removal, which makes them time-consuming and low-efficiency efforts. In addition, only small cores can be tested in the laboratory.

Three-dimensional digital cores allow rock physical experiment simulations through numerical algorithms (Liu et al., 2009; Cao et al., 2022). Rock physical numerical simulations based on digital cores are a type of non-destructive core testing method. Once a digital core is established, it can be reused and enables numerical simulations of rock electrical properties, acoustic properties, NMR characteristics, and flow characteristics. Compared to traditional rock physical experiments, digital rock physical experiments have five advantages as follows: the simulations are fast and incur low costs; once a digital core describing the rock microstructure is established, multiple rock physical properties like resistivity, sonic velocity, permeability, and NMR response can be computed to establish relationships among the different physical properties; digital rock experiments can simulate physical quantities that are difficult to measure in conventional experiments, such as three-phase relative permeability; adjusting the microparameters of the digital cores allows studies on how the reservoir parameters affect the rock physical properties; for difficult-to-core rocks like fractured carbonates, shale, and oil sands, digital rock experiments can replace traditional methods to measure various rock physical properties.

Because the low-porosity and low-permeability reservoirs in the WH depression are highly heterogeneous, challenges remain even though 3D digital core NMR experiments have made significant progress. Logging-based digital core technology provides a rapid method for obtaining NMR data. In this approach, specialized equipment are used for coring at the well site and for conducting optical, acoustic, and NMR measurements to obtain various rock physical parameters that are later digitized to establish a core database. This approach allows fast coring, analysis, modeling, interpretation, and application. Extensive theoretical studies and laboratory experiments have indicated that microscopic features like the pore structure better reflect the essence of a reservoir rather than macroscopic properties like the porosity, and these features are also important for evaluating the reservoir productivity. For medium-to-high-porosity and medium-to-high-permeability reservoirs, the porosity has a relatively direct relationship with productivity; however, for low-porosity and low-permeability reservoirs, the pore structure influences on productivity must be studied. Therefore, the NMR T₂ spectra obtained from digital core technology reflect the pore structure characteristics, enabling reservoir classification and productivity evaluation. Regarding the acquisition of NMR data, when the data precision is equivalent, logging-based digital core technology has the following features: it is efficient and the data acquisition cycle is reduced to one-tenth of that for laboratory testing, thereby improving the efficiency tenfold; unlike one-time NMR logging, it is repeatable; it is cost-effective and incurs lower costs than NMR logging or laboratory NMR tests; it is flexible and allows adjustment of the testing plans on-site, retesting of the existing cores, or acquisition of new cores for testing; it is accurate, and although digital cores may be less precise for measuring oil saturation than NMR logging, they are is essentially equivalent for assessing the pore structure because the fluid effects can be minimized when measuring the NMR T₂ spectrum. In summary, logging-based digital core technology, NMR logging, and laboratory NMR experiments can complement each other.

The pore structure evaluation method using logging-based digital core technology involves measuring the rock porosity and permeability through NMR and then calculating the pore structure parameters (Sun et al., 2021). The NMR T₂ spectrum is used to determine the proportions of pores of different sizes within the total porosity. In other words, logging-based digital core technology can be used to comprehensively evaluate the rock pore structure using porosity, proportions of pore types within the total porosity, and morphology of the T₂ spectrum. Consider that S1, S2, and S3 represent the proportions of small, medium, and large pores within the total porosity, respectively; clearly, a greater proportion of medium and large pores indicates better pore structure of the rock, meaning that a larger (S2 + S3) value indicates better pore structure.

Analysis of the core experimental data from the WH depression based on the porosity range, proportion of medium and large pores, and position of the main peak in the T₂ spectrum indicates that the reservoir pore structures can be divided into four grades, with Class I being the best and Class IV being the worst, as shown in Table 3.

Table 3

Table 3. Classification of reservoir pore structure.

By processing and analyzing the NMR experimental data from the rock samples, and by integrating the logging data with production test data, we established a qualitative reservoir productivity evaluation chart, as shown in Figure 9; the corresponding productivity evaluation criteria are presented in Table 4.

Figure 9

Figure 9. Reservoir productivity evaluation chart based on nuclear magnetic resonance (NMR) data.

Table 4

Table 4. Reservoir productivity evaluation criteria based on NMR data.

2.3.2 Productivity evaluation using pseudo capillary pressure curves

Reservoir productivity can also be evaluated using the pseudo capillary pressure curves derived from NMR logging data, which allow interpretation of parameters like the capillary pressure, displacement pressure, average pore throat radius, and maximum mercury saturation (Wu et al., 2021; Zhang et al., 2020). Figure 10 shows the NMR logging interpretation results for the H1 well, where track 6 represents the capillary pressure curve, track 7 is the displacement pressure, track 8 is the average pore throat radius, and track 9 is the maximum mercury saturation. The statistical results of the mercury injection parameters for the tested oil layers are listed in Table 5. The displacement pressure is an important parameter characterizing reservoir permeability; the lower the value of this parameter, the better is the permeability. From Table 5, the calculated displacement pressures range from 0.04 MPa to 0.2 MPa, with an average value of 0.1 MPa. The pore throat radius determines the connectivity of the reservoir pores, where larger values indicate better connectivity and thus permeability. The calculated average pore throat radii range from 1.0 μm to 3.9 μm, with an average value of 2.511 μm, corresponding to fine-to-micro-pore-throat levels. The maximum mercury saturation reflects the oil storage capacity of the reservoir to some extent, where larger values indicate stronger oil-bearing capabilities. The calculated maximum mercury saturation ranges from 38.5% to 58.1%, with an average value of 49.6%. In the H1 well, layers 8 and 12 exhibit good pore structures and contribute significantly to production; here, the displacement pressures are below 0.064 MPa, pore throat radii exceed 2 μm, and maximum mercury saturation exceeds 50%, yielding an average daily oil production of 30.71 t and indicating their classification as high-yield oil layers. Based on the NMR logging results from nine wells in the WH depression and their corresponding production data, we established a reservoir productivity evaluation standard for the reservoirs in the WH depression, as shown in Table 6.

Figure 10

Figure 10. NMR logging interpretation results for the H1 well.

Table 5

Table 5. Summary of mercury injection parameters for the tested oil layers in the H1 well.

Table 6

Table 6. Reservoir productivity evaluation criteria using the pseudo capillary pressure.

The pseudo capillary pressure curves are derived indirectly from the NMR data (Gray et al., 2021). Currently, the commonly used models for converting NMR data to pseudo capillary pressure curves exhibit some discrepancies with respect to the actual measurements. For downhole NMR data, oil and gas corrections should also be applied. Nevertheless, the pseudo capillary pressure curves can still serve as references for evaluating the reservoir productivity by reflecting the trends of variations in the pore structure.

3 Case study

The L18 well is an important exploration well located in the Hongge’er structural belt of the Nanwasuo depression in the WH depression. Here, the layers 58 and 60 (1,538.2–1,549.6 m) have developed into typical low-porosity and low-permeability conglomerate reservoirs, with the porosity ranging from 3% to 13.2% and permeability varying from 8.9 × 10⁻³ μm² to 276 × 10⁻³ μm². The oil and gas from these layers show fluorescence to oil-stained zones. Five core samples were acquired from this section to provide the material basis for validating multiple reservoir productivity evaluation methods.

3.1 Digital core NMR testing

The five cores underwent systematic digital core NMR testing to yield detailed T₂ spectrum distributions and pore structure parameters (the data and T₂ spectra are shown in Figure 11). Based on the evaluation methods and standards established earlier, the analysis result for each core is as follows:

1. The core from 1,540.4 m (fluorescent conglomerate) shows an NMR porosity of 13.21% and NMR permeability of 253.987 × 10⁻³ μm². The pore components S1, S2, and S3 account for 8.32%, 32.31%, and 59.37%, respectively, with (S2 + S3) being 91.68%. According to the classification in Table 3, this core has a Class I pore structure and the corresponding layer is predicted to have naturally high productivity according to Table 4.

2. The core from 1,549.7 m (oil-stained conglomerate) shows a porosity of 9.87% and permeability of 8.89 × 10⁻³ μm², with S1 = 44.93%, S2 = 39.46%, S3 = 15.61%, and (S2 + S3) = 55.07%; thus the layer can be classified as having a Class II pore structure with a predicted naturally low productivity.

3. The core from 1,539.2 m (oil-stained conglomerate) has a porosity of 7.42%, permeability of 56.408 × 10⁻³ μm², S1 = 31.44%, S2 = 43.50%, S3 = 25.07%, and (S2 + S3) = 68.57%, which indicates that the layer has a Class I pore structure and predicted natural productivity.

4. The core at 1,549.85 m (oil-stained conglomerate) has a porosity of 7.61%, permeability of 67.696 × 10⁻³ μm², S1 = 31.98%, S2 = 39.89%, S3 = 28.14%, and (S2 + S3) = 68.03%, which indicates that the layer has a Class I pore structure and predicted natural productivity.

5. The core from 1,548.0 m (oil-trace conglomerate) shows an extremely low porosity of 3.04% but abnormally high permeability of 276.211 × 10⁻³ μm², with S1 = 34.32%, S2 = 47.75%, S3 = 17.93%, and (S2 + S3) = 65.68%. This sample may contain fractures or very large pores; it has a Class I pore structure and is predicted to have natural productivity, although this may be limited by the very low porosity.

Figure 11

Figure 11. Comprehensive reservoir productivity evaluation chart of the L18 well.

3.2 Validation of NMR-based productivity evaluation

Using the NMR-based reservoir productivity standards outlined in Table 4, the layers 58 and 60 are predicted to achieve relatively high production under natural conditions. Since the cores from these layers underwent systematic MICP testing, the analyses show an average pore throat radius of approximately 0.55 μm and a displacement pressure of approximately 0.15 MPa, indicating relatively high-quality pore structures for the WH depression. Using the quantitative model established in Section 2.2 and Equation 6, we have Q = 145.67 × (r_mean)^1.326 with an average pore throat radius of r_mean = 0.55 μm, such that the predicted quantitative productivity is approximately 18.7 t/d.

3.3 Validation using conventional logging productivity standards

Combining the above results with the conventional logging productivity evaluation standards established earlier, we obtained a deep-reading resistivity of 15–35 Ω·m, notable negative anomalies in the spontaneous potential, and clear positive deviations in the microresistivity, all of which satisfy the Class I reservoir criteria in Table 1. The acoustic transit time of 260–340 μs/m and bulk density of 2.15–2.45 g/cm³ yield calculated porosities in the range of 8%–14%, which are consistent with the core measurements. According to the classification shown in Table 1 based on conventional logging data, all reservoir parameters meet Class I standards, supporting the high productivity evaluation conclusion.

3.4 Results

In October 2017, the 1,538.2–1,549.6 m section of well L18 was tested using a pumping-production method, which yielded a daily oil production of 20.8 t and a cumulative production of 37.66 t without water output, meeting the industrial oil flow standard.

In summary, the validation and comparison of multiple evaluation methods show that the digital core NMR prediction of natural high productivity was accurate, along with an actual daily production of 20.8 t. The MICP quantitative predictions yielded a productivity of 18.7 t/d, while the actual productivity was 20.8 t/d, resulting in a relative error of 10.1% that demonstrates high quantitative prediction accuracy. Conventional logging predicted that the reservoir level was Class I, and the actual tests achieved industrial oil flow, confirming the accuracy of the evaluation result.

Overall, the test results of the L18 well fully validate the reliability and practicality of the multiple reservoir productivity evaluation methods established in this study. They indicate that productivity predictions using digital core NMR data or MICP data are reliable and confirm the effectiveness of the proposed evaluation methods. The three methods explored herein showed good performances for the L18 well: digital core NMR achieved 100% accuracy for qualitative prediction, and the MICP quantitative prediction had only a 10.1% relative error, meeting the precision requirements for engineering applications. Therefore, the combined use of multiple methods can provide more comprehensive and accurate productivity evaluations in low-porosity and low-permeability reservoirs.

It should be noted that the tiered reservoir classification standards proposed in this study are derived from the lacustrine tight sandstone reservoirs of the WH depression. Although the quantitative thresholds (e.g., critical pore throat radius or displacement pressure) are influenced by specific sedimentary and diagenetic conditions, the general methodology linking the pore structure parameters with reservoir productivity remains applicable to other basins. For reservoirs with distinct depositional systems like deltaic or marine facies, we suppose that the same evaluation framework can be adopted after appropriate calibration using local core and test data. Therefore, the proposed approach provides a transferable methodology rather than fixed numerical criteria for flexible adaptation to diverse geological settings.

4 Discussion

Comparing the tight clastic reservoirs of the WH depression with other studies highlights both the similarities and differences in their petrophysical characteristics and productivity controls. In the Shushan Basin of Egypt, the Cretaceous and Cambro-Ordovician sandstones exhibit porosities below 10% and permeabilities under 1 mD, along with extensive silica cementation that significantly reduces pore connectivity (Fa et al., 2025; Farouk et al., 2024). Similar to the WH depression, the pore-throat geometry and effective porosity are critical for controlling deliverability. Moreover, geomechanical conditions like the dominant strike-slip stress regimes and azimuthal variations strongly influence wellbore stability and completion strategies (Chen et al., 2014; Elenius et al., 2018), which are in contrast with the WH depression where structural heterogeneities rather than stress orientations primarily affect the reservoir quality.

In the Risha field of northeastern Jordan, the Cambro-Ordovician sandstones show very low porosity (<6%) and permeability (<1 mD), along with late diagenetic silica cementation and chemical compaction that reduces the intergranular porosity (Zhang et al., 2025). However, the grain-coating clays and minor secondary porosity locally preserve the effective pore spaces to support limited hydrocarbon flows. This is comparable to the WH depression, where secondary dissolution contributes to productive pore networks in certain intervals, although the overall porosity is larger (up to ∼15%). These comparisons emphasize that both primary depositional texture and diagenetic alterations strongly control reservoir performances in tight sandstones.

Despite differences in the absolute quantitative values, some similarities are observed across all these basins in that the tight reservoir productivity is fundamentally controlled by the micro-to-meso porosity, pore throat distribution, and connectivity. While the workflow for the WH depression establishes specific thresholds for the pore throat radius and displacement pressure, the methodology integrating core, NMR, and logging data is transferable. With local calibrations, similar tiered evaluation frameworks may be applicable to deltaic, fluvial, or lacustrine tight sandstones to guide reservoir classification, completion design, and production forecasting.

5 Conclusion

In this study, we developed a multisource data evaluation approach to assess the productivity of low-porosity and low-permeability reservoirs in the WH depression by integrating MICP, NMR, and conventional logging analyses. The major conclusions of this work are as follows:

1. Quantitative MICP-based criteria: Analysis of the core mercury injection data shows that the mean pore throat radius is the most sensitive parameter controlling reservoir productivity. Reservoirs with mean pore throat radii exceeding 0.12 μm and displacement pressures less than 5 MPa generally exhibit natural productivity exceeding 20 t/d, whereas those with mean pore throat radii below 0.05 μm require stimulation to achieve industrial production levels.

2. NMR-based criteria: The NMR data analysis indicates that reservoirs with (S₂ + S₃) pore volume fractions exceeding 65% and T₂ geometric mean times exceeding 20 ms correspond to high-yield layers that typically produce over 15 t/d under natural conditions. Conversely, reservoirs with (S₂ + S₃) < 40% generally exhibit poor natural productivity and require fracturing.

3. Conventional log-based criteria: When only conventional logs are available, the reservoir productivity can be effectively estimated by combining the spontaneous potential, microresistivity, and array induction resistivity responses. High-productivity reservoirs typically feature deep resistivities >20 Ω·m, low natural gamma-ray values (<70 API), and AII values >1 × 10⁶.

4. Overall, the integrated approach presented herein that prioritizes MICP, followed by NMR and finally conventional logging interpretation provides a robust and practical method to evaluate the productivity potentials of low-porosity and low-permeability reservoirs quantitatively. This quantitative framework offers valuable guidance for reservoir classification, development planning, and well productivity predictions in the WH depression as well as similar tight oil systems.

Data availability statement

The original contributions presented in the study are included in the article/supplementary material, and any further inquiries may be directed to the corresponding author.

Author contributions

XH: Conceptualization, Formal analysis, Investigation, Methodology, Supervision, Validation, Visualization, Writing – original draft, Writing – review and editing. HF: Conceptualization, Data curation, Formal analysis, Project administration, Writing – review and editing. JT: Conceptualization, Methodology, Project administration, Writing – review and editing. JZ: Conceptualization, Data curation, Resources, Writing – review and editing. JL: Conceptualization, Data curation, Methodology, Validation, Writing – review and editing.

Funding

The author(s) declared that financial support was not received for this work and/or its publication.

Conflict of interest

Author JT was employed by China Oilfield Services Limited (Production Division). Authors JZ and JL were employed by China National Offshore Oil Corporation (CNOOC) Ltd. (Tianjin branch).

The remaining author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Alisheva, Z., Al-Dujaili, A. N., Tileuberdi, N., Muratova, S., Omirzakova, E., Sanatbekov, M., et al. (2025). Modeling and analysis of filtration processes in oil reservoirs of small fields by reserves. Sci. Rep. 15, 11555. doi:10.1038/s41598-025-96797-8

PubMed Abstract | CrossRef Full Text | Google Scholar

Aminian, K., and Ameri, S. (2009). Predicting production performance of CBM reservoirs. J. Nat. Gas Sci. Eng. 1, 25–30. doi:10.1016/j.jngse.2009.03.003

CrossRef Full Text | Google Scholar

Cao, D., Hou, Z., Liu, Q., and Fu, F. (2022). Reconstruction of three-dimension digital rock guided by prior information with a combination of InfoGAN and style-based GAN. J. Petroleum Sci. Eng. 208, 109590. doi:10.1016/j.petrol.2021.109590

CrossRef Full Text | Google Scholar

Chen, J., Bai, X., Ye, Q., Li, H., Zhu, M., and Zhang, X. (2014). 3D seismic acquisition technology of igneous rock area in wulanhua depression of erlian basin. China Pet. Explor. 19, 65–72. doi:10.3969/j.issn.1672-7703.2014.05.007

CrossRef Full Text | Google Scholar

Clark, A. J., Lake, L. W., and Patzek, T. W. (2011). “Production forecasting with logistic growth models,” in Proceedings of the SPE annual technical conference and exhibition. doi:10.2118/144790-MS

CrossRef Full Text | Google Scholar

Cripps, A. C., and McCann, D. M. (2000). The use of the natural gamma log in engineering geological investigations. Eng. Geol. 55, 313–324. doi:10.1016/S0013-7952(99)00085-X

CrossRef Full Text | Google Scholar

Elenius, M., Skurtveit, E., Yarushina, V., Baig, I., Sundal, A., Wangen, M., et al. (2018). Assessment of CO₂ storage capacity based on sparse data: skade formation. Int. J. Greenh. Gas Control 79, 252–271. doi:10.1016/j.ijggc.2018.09.004

CrossRef Full Text | Google Scholar

Eydinov, D., Gao, G., Li, G., and Reynolds, A. C. (2009). Simultaneous estimation of relative permeability and porosity/permeability fields by history matching production data. J. Can. Petroleum Technol. 48, 13–25. doi:10.2118/132159-pa

CrossRef Full Text | Google Scholar

Farouk, S., Sen, S., Ghosh, S., and Mandal, P. P. (2025). In-situ stress State of the shushan basin, Western desert, Egypt: implications for developing tight Cretaceous and Paleozoic reservoirs. Arabian J. Sci. Eng. doi:10.1007/s13369-025-10766-x

CrossRef Full Text | Google Scholar

Farouk, S., Sarhan, M., Sen, S., Ahmad, F., Kahtany, K., and Reda, M. (2024). Evaluation of the Lower Cretaceous alam El Bueib sandstone reservoirs in shushan Basin, Egypt – implications for tight hydrocarbon reservoir potential. J. Afr. Earth Sci. 128, 105386. doi:10.1016/j.jafrearsci.2024.105386

CrossRef Full Text | Google Scholar

Feng, C., Lin, W., Cheng, Z., Zhang, W., Hou, Y., and Jing, S. (2012). Reservoir classification and productivity forecasting method for low prorosity and permeability reservoir with log data. Well Logging Technol. 36, 308–313. doi:10.16489/j.issn.1004-1338.2012.03.020

CrossRef Full Text | Google Scholar

Ge, B., Wen, Z., and Zheng, J. (2003). Evaluating and predicting method of oil layer’s natural production using well logging data. Petroleum Geol. and Oilfield Dev. Daqing 22, 54–56. doi:10.3969/j.issn.1000-3754.2003.01.021

CrossRef Full Text | Google Scholar

Geir, N., Mannseth, T., and Vefring, E. H. (2002). “Near-Well reservoir monitoring through ensemble kalman filter,” in Proceedings of the SPE/DOE improved oil recovery symposium. doi:10.2118/75235-MS

CrossRef Full Text | Google Scholar

Gray, M. E., Daniel, R., Kaldi, J., and Kulikowski, D. (2021). Generating pseudo-capillary pressure curves from nuclear magnetic resonance (NMR) data: a case study from the cooper basin, Australia. Aust. J. Earth Sci. 68, 83–98. doi:10.1080/08120099.2020.1750482

CrossRef Full Text | Google Scholar

Gu, Z., Wang, S., Guo, P., and Zhao, W. (2024). Utilizing differences in mercury injection capillary pressure and nuclear magnetic resonance pore size distributions for enhanced rock quality evaluation: a winland-style approach with physical meaning. Appl. Sci. 14, 1881. doi:10.3390/app14051881

CrossRef Full Text | Google Scholar

Guan, X. (1998). An interpretation model for reservoir production capacity based on log data and its application. J. Jianghan Petroleum Inst. 20, 54–57.

Google Scholar

Hassan, A., Mahmoud, M., Al-Majed, A., and Abdulraheem, A. (2020). “A new technique to quantify the productivity of complex Wells using artificial intelligence tools,” in Proceedings of the international petroleum technology conference. doi:10.2523/IPTC-19706-Abstract

CrossRef Full Text | Google Scholar

Jiang, Z., Mao, Z., Shi, Y., and Wang, D. (2018). Multifractal characteristics and classification of tight sandstone reservoirs: a case study from the Triassic yanchang formation, ordos basin, China. Energies 11, 2242. doi:10.3390/en11092242

CrossRef Full Text | Google Scholar

Jing-qiang, L., Chao-mo, Z., and Zhansong, Z. (2016). Combine the capillary pressure curve data with the porosity to improve the prediction precision of permeability of sandstone reservoir. J. Petroleum Sci. Eng. 139, 43–48. doi:10.1016/j.petrol.2015.12.018

CrossRef Full Text | Google Scholar

Ju, J., Wang, J., Gao, R., Sun, Y., Gao, S., and Liu, Y. (2005). Productivity forecast of the reservoir with low porosity and low permeability in Erlian oilfield. Well Logging Technol. 29, 379–381. doi:10.16489/j.issn.1004-1338.2005.04.031

CrossRef Full Text | Google Scholar

Li, W., Hu, D., Gong, C., Fan, T., Chen, Y., Li, Y. n., et al. (2024). Controlling factors of high-quality reservoirs in low permeability sandstone: a case study of the upper member of the lower ganchaigou formation, qaidam basin. Front. Earth Sci. 12, 12–2024. doi:10.3389/feart.2024.1396061

CrossRef Full Text | Google Scholar

Lai, J., Wang, G., Fan, Q., Pang, X., Li, H., Zhao, F., et al. (2022). Geophysical well-log evaluation in the era of unconventional hydrocarbon resources: a review on current status and prospects. Surv. Geophys. 43, 913–957. doi:10.1007/s10712-022-09705-4

CrossRef Full Text | Google Scholar

Liu, Y., Wu, G., Yan, J., and Jin, Q. (2004). Identification of fluid property and productivity forecast in depression M. Well Logging Technol. 28, 458–461. doi:10.16489/j.issn.1004-1338.2004.05.031

CrossRef Full Text | Google Scholar

Liu, X., Sun, J., and Wang, H. (2009). Numerical simulation of rock electrical properties based on digital cores. Appl. Geophys. 6, 1–7. doi:10.1007/s11770-009-0001-6

CrossRef Full Text | Google Scholar

Liu, J., Liu, S., Zhang, F., Miao, B., Yuan, C., and Su, B. (2018a). A method for improving the evaluation of elemental concentrations measured by geochemical well logging. J. Radioanalytical Nucl. Chem. 317, 1113–1121. doi:10.1007/s10967-018-5986-y

CrossRef Full Text | Google Scholar

Liu, M., Xie, R., Wu, S., Zhu, R., Mao, Z., and Wang, C. (2018b). Permeability prediction from Mercury injection capillary pressure curves by partial least squares regression method in tight sandstone reservoirs. J. Petroleum Sci. Eng. 169, 135–145. doi:10.1016/j.petrol.2018.05.020

CrossRef Full Text | Google Scholar

Mao, Z., and Li, J. (2000). Method and models for productivity prediction of hydrocarbon reservoirs. Acta Pet. Sin. 21, 58–61. doi:10.7623/syxb200005012

CrossRef Full Text | Google Scholar

Mohaghegh, S. D. (2011). Reservoir simulation and modeling based on artificial intelligence and data mining (AI&DM). J. Nat. Gas Sci. Eng. 3, 697–705. doi:10.1016/j.jngse.2011.08.003

CrossRef Full Text | Google Scholar

Nabawy, B. S. (2025). New approaches in reservoir characterization utilizing conventional and special core analyses: a comprehensive review. J. Umm Al-Qura Univ. Appl. Sci. doi:10.1007/s43994-025-00225-6

CrossRef Full Text | Google Scholar

Okon, A. N., Adewole, S. E., and Uguma, E. M. (2021). Artificial neural network model for reservoir petrophysical properties: porosity, permeability and water saturation prediction. Model. Earth Syst. Environ. 7, 2373–2390. doi:10.1007/s40808-020-01012-4

CrossRef Full Text | Google Scholar

Ou, Y. (1994). Petrophysical interpretation and reservoir characterization. Beijing: Petroleum Industry Press.

Google Scholar

Rezaee, M. R., Jafari, A., and Kazemzadeh, E. (2006). Relationships between permeability, porosity and pore throat size in carbonate rocks using regression analysis and neural networks. J. Geophys. Eng. 3, 370–376. doi:10.1088/1742-2132/3/4/008

CrossRef Full Text | Google Scholar

Shi, X., Cui, Y., Xu, W., Zhang, J., and Guan, Y. (2020). Formation permeability evaluation and productivity prediction based on mobility from pressure measurement while drilling. Petroleum Explor. Dev. 47, 146–153. doi:10.1016/S1876-3804(20)60013-1

CrossRef Full Text | Google Scholar

Sun, J., Cai, J., Feng, P., Sun, F., Li, J., Lu, J., et al. (2021). Study on nuclear magnetic resonance logging T2 spectrum shape correction of sandstone reservoirs in oil-based mud Wells. Molecules 26, 6082. doi:10.3390/molecules26196082

PubMed Abstract | CrossRef Full Text | Google Scholar

Wang, Y., Cheng, S., Zhang, F., Feng, N., Li, L., Shen, X., et al. (2021). Big data technique in the reservoir parameters’ prediction and productivity evaluation: a field case in Western South China Sea. Gondwana Res. 96, 22–36. doi:10.1016/j.gr.2021.03.015

CrossRef Full Text | Google Scholar

Wang, X., Fu, G., Fan, B., Wang, S., Feng, L., and Peng, C. (2024). Integration of conventional well logs and core samples to predict porosity of tight reservoir: a case study from ordos basin. Q. J. Eng. Geol. Hydrogeology 57, qjegh2023-042. doi:10.1144/qjegh2023-042

CrossRef Full Text | Google Scholar

Wu, B., Xie, R., Xu, C., Wei, H., Wang, S., and Liu, J. (2021). A new method for predicting capillary pressure curves based on NMR echo data: sandstone as an example. J. Petroleum Sci. Eng. 202, 108581. doi:10.1016/j.petrol.2021.108581

CrossRef Full Text | Google Scholar

Xu, H., Tang, D. Z., Tang, S. H., Zhao, J. L., Meng, Y. J., and Tao, S. (2014). A dynamic prediction model for gas–water effective permeability based on coalbed methane production data. Int. J. Coal Geol. 121, 44–52. doi:10.1016/j.coal.2013.11.008

CrossRef Full Text | Google Scholar

Zhang, W., Zhuang, M., Liu, Q. Q., Shi, L., Sun, Y., Liu, Q. H., et al. (2021). Frequency domain spectral element method for modelling poroelastic waves in 3-D anisotropic, heterogeneous and attenuative porous media. Geophys. J. Int. 227, 1339–1353. doi:10.1093/gji/ggab269

CrossRef Full Text | Google Scholar

Zhan, W., Chen, Y., Liu, Q., Li, J., Sacchi, M. D., Zhuang, M., et al. (2024). Simultaneous prediction of petrophysical properties and formation layered thickness from Acoustic logging data using a modular cascading residual neural network (MCARNN) with physical constraints. J. Appl. Geophys. 224, 105362. doi:10.1016/j.jappgeo.2024.105362

CrossRef Full Text | Google Scholar

Zhang, L., Liu, G., Zhou, C., and Liu, Z. (2005). Reservoir productivity prediction by array induction logging data. Petroleum Explor. Dev. 32, 84–87.

Google Scholar

Zhang, C., Mao, Z., Li, G., and Han, Y. (2008). Method for productivity prediction of anpeng deep-seated reservoirs with low permeability and its application. Well Logging Technol. 32, 562–565. doi:10.16489/j.issn.1004-1338.2008.06.015

CrossRef Full Text | Google Scholar

Zhang, H., Li, G., Guo, H., Zhang, W., Wang, Y., Li, W., et al. (2020). Applications of nuclear magnetic resonance (NMR) logging in tight sandstone reservoir pore structure characterization. Arabian J. Geosciences 13, 572. doi:10.1007/s12517-020-05590-6

CrossRef Full Text | Google Scholar

Zhang, H., Li, G., Guo, H., Zhang, W., Wang, Y., Li, W., et al. (2025). Characterization of the gas-bearing tight Paleozoic sandstone reservoirs of the risha field, Jordan: inferences on reservoir quality and productivity. Arabian J. Sci. Eng. 50, 215–235. doi:10.1007/s13369-024-09000-x

CrossRef Full Text | Google Scholar

Zheng, L., Yao, Y., Zuo, X., Li, H., and He, T. (2006). Well log evaluation and productivity prediction method for heavy oil reservoir. Special Oil and Gas Reservoirs 13, 26–28.

Google Scholar