Next Article in Journal
Determination of the Radius of the Ring Current in the Earth’s Core According to the Data of the INTERMAGNET Network Observatories
Previous Article in Journal
Quantification of Overlapping and Network Complexity in News: Assessment of Top2Vec and Fuzzy Topic Models
Previous Article in Special Issue
The Genesis of a Thin-Bedded Beach-Bar System Under the Strike-Slip Extensional Tectonic Framework: A Case Study in the Bohai Bay Basin
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 15
Issue 17
10.3390/app15179631
applsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Thin-Layer Siliceous Dolomite Prediction Based on Post-Stack Seismic Data Mining and Optimization

by
Shimin Liu
1,2,
Lu Zhou
1,2,*,
Baoshou Zhang
3,
Ruixue Dai
3,
Peng Lu
4,
Changxingyue He
3,
Yong Wu
1,2 and
Chunyong Ni
1,2
1
Sichuan Key Laboratory of Natural Gas Geology, School of Geosciences and Technology, Southtwest Petroleum University, Chengdu 610500, China
2
State Key Laboratory of Oil and Gas Reservior Geology and Exploitation, Southwest Petroleum University, Chengdu 610500, China
3
Exploration and Development Research Institute of PetroChina Southwest Oil and Gas Field Company, Chengdu 610041, China
4
Southwest Branch, CNPC Logging Company Limited, Chongqing 400021, China
*
Author to whom correspondence should be addressed.
Appl. Sci. 2025, 15(17), 9631; https://doi.org/10.3390/app15179631
Submission received: 24 June 2025 / Revised: 8 August 2025 / Accepted: 10 August 2025 / Published: 1 September 2025
(This article belongs to the Special Issue Advances in Reservoir Geology and Exploration and Exploitation)
Browse Figures
Versions Notes

Abstract

Siliceous dolomite is developed at the top of the fourth member of Dengying Formation in Penglai gas field, Sichuan Province, and can be used as a good cap rock to seal oil and gas. In this study, based on post-stack seismic data mining and optimization, sensitive logging curve analysis is found to be the optimal prediction method, subdividing small layers to finely characterize thin-layer siliceous dolomite. It is found that the navigation-pyramid two-stage classification seismic data and the waveform indication simulation method based on RXO parameters have a good recognition effect on the characterization of thin-layer siliceous dolomite. Studies have shown that faults can provide ascending channels for deep silicon-rich hydrothermal fluids, resulting in the development of siliceous dolomite mostly near the fault zone. The siliceous dolomite in the upper sub-member of the fourth member of Dengying Formation in Penglai area mainly develops two sets of siliceous dolomite in the vertical direction, and the lower siliceous rock develops on a larger scale. In the lateral direction, it is mainly distributed in the JT1-PS7-P8-PS9 well area in the north, in a continuous sheet shape. In the south, the development scale of siliceous dolomite is small and distributed in the shape of a strip.

1. Introduction

Siliceous dolomite is a sedimentary rock composed of dolomite (dolomite as the main mineral) and siliceous components (such as microcrystalline quartz, flint, and chalcedony) [1,2,3,4,5]. Ordinary dolomite is prone to dissolution to form intergranular pores, interparticle pores, dissolution pores, etc., providing space for oil and gas storage and migration. Due to the cryptocrystalline siliceous components (quartz or flint), siliceous dolomite is filled in the primary pores of dolomite particles, which significantly reduces the porosity and permeability, often leading to the densification of lithology. It can be used as a local interlayer to prevent the vertical seepage of fluid and form a layer of plugging, which is conducive to the stratification and accumulation of oil and gas. Therefore, the spatial distribution of siliceous dolomite will have a greater impact on the distribution of reservoirs [6,7,8].
Regarding the source of silica, there are different interpretations of the genesis and sedimentary environment of siliceous rocks. At present, there are three main understandings: biological origin, chemical precipitation, and diagenesis. Biogenetically, it is considered that a radiolarian or diatom in the deep-water environment provides the original source of siliceous deposition, forming in the syndepositional period [9,10,11]. According to the understanding of diagenetic metasomatism, siliceous rocks are formed by metasomatism of carbonate minerals by siliceous fluids in the later stage, and the migration channels mainly depend on faults. After metasomatism, chert nodules and chert clumps are formed. Chert nodules are often formed during the unconsolidated period of diagenetic sediments, and chert clumps are often formed after consolidation [12,13,14]. Based on the understanding of chemical precipitation, the siliceous source is mainly the hydrothermal fluid of submarine volcanic eruption. Because the solubility of siliceous rock increases rapidly with the increase in temperature, after the local hydrothermal fluid encounters cold water, the siliceous rock is precipitated in the supersaturated state due to the decrease in solubility [15,16,17,18]. According to the characteristics of cores, microscopic thin sections, major elements, and trace elements, Luo Wenjun et al. determined that the siliceous rocks in the Gaoshiti area are of hydrothermal sedimentary genesis related to submarine volcanic eruptions [19]. Because the Gaomo area is adjacent to the Penglai gas area, it is considered that the siliceous dolomite in the study area is also of hydrothermal sedimentary genesis.
The siliceous dolomite of the fourth member of Dengying Formation in the study area is a thin to medium-thickness layer. For ultra-deep strata, the conventional post-stack seismic data have poor resolution and data quality under conditions of high speed and low frequency, which can lead to multiple solutions. It is difficult to achieve accurate characterization of thin-layer siliceous rocks and their boundaries. The purpose of this study is to fully improve the accuracy and fineness of seismic interpretation, predict the spatial distribution characteristics of thin-layer siliceous dolomite, and characterize internal heterogeneity by mining post-stack high-resolution seismic data and subdividing small layers.

2. Overview of the Study Area

The study area is located in the eastern part of Eurasia, Suining City, central Sichuan Basin, China, near Yanting County in the north, Tongnan District in the south, Nanchong City in the east, and Santai County in the west, with an area of about 6410 km2 (Figure 1a). The sedimentary facies belt in this area is located in the platform margin on the east side of the Deyang–Anyue rift trough, mainly carbonate platform deposits [20,21,22,23]. Among them, the fourth member of Dengying Formation is mainly composed of clotted mud powder crystal dolomite and laminated dolomite, which is characterized by siliceous (quartz) dolomite, and a small amount of limestone is developed at the top (Figure 1b). Among them, siliceous dolomite is very important for the construction of reservoirs because of its low porosity and permeability, which can be used as a cap rock to prevent escape [24,25].
Structurally, the study area is located in the northern slope of the inherited paleo-uplift in central Sichuan and has undergone three uplifts of the Tongwan movement, forming multiple nonconforming interfaces [26,27]. Among them, the uplift intensity of the Tongwan one-scene movement is medium, resulting in the unconformity contact of the second section of the Dengying Formation and the third section of the Dengying Formation; the strongest motion energy of the second uplift in Tongwan resulted in the unconformity contact between the fourth member of Dengying Formation and Maidiping Formation. The energy of the third uplift in Tongwan was the weakest, leading to the unconformity contact of the Qiongzhusi Formation in the fourth member of the Dengying Formation [28,29,30]. Due to the influence of the Tongwan movement, the fourth member of the Dengying Formation was subjected to supergene karstification, forming weathering crust and karst reservoirs between the fourth member of the Dengying Formation–Maidiping Formation and the Maidiping Formation–Qiongzhusi Formation.

3. Data and Methods

3.1. Logging Data Analysis

The conventional seismic attribute analysis is based on the large time window for attribute calculation, and the extracted attributes are the overall reflection of the geological information of the target layer [31]. As shown in Figure 2, the thickness of siliceous rocks in the study area is mainly in the range of 2–24 m, which is not applicable to the fine interpretation of siliceous rocks in the study area. Therefore, considering the thinness of the siliceous rocks in the study area and the large difference in spatial distribution, the method of dividing small layers in the isochronous framework is adopted: taking the top and the upper and lower bottom portions of the fourth member of the Dengying Formation as the internal constraint boundary of the framework, well-to-well calibration is used to establish a unified sequence framework of wells and earthquakes in the upper part of the fourth member of the Dengying Formation. On this basis, the three-dimensional seismic data is used to divide the isochronous layers in the upper part of the fourth member of the Dengying Formation. Combined with the logging stratification data, the upper part of the fourth member of the Dengying Formation is divided into 10 small layers. Because of the relatively high thickness of siliceous dolomite in JT1, PS7, PS8, and PS9 wells and the relatively concentrated vertical distribution, the distribution characteristics of siliceous dolomite in the study area can be better observed. Therefore, the lithology profile of JT1-PS7-PS8-PS9 well is selected for comparative analysis. As shown in Figure 2, considering that the siliceous rock section is mainly developed at the top of the fourth member of Dengying Formation, it can be divided into upper and lower sets of siliceous rocks. The upper siliceous rock layer is mainly distributed in two small layers: small layer 2 and small layer 3; the lower siliceous rock layer is mainly distributed in two small layers: small layer 4 and small layer 6.

3.2. Principle of Method

Taking full account of the thinness of siliceous rocks, this study is based on the sublayer division method of the fourth member of Dengying Formation for establishing a high-resolution stratigraphic framework. However, this method has higher requirements for the accuracy and resolution of seismic data. In view of the low resolution of conventional seismic data, it is impossible to characterize thin siliceous rocks. Therefore, the seismic resolution is increased using navigation-pyramid classification and reconstruction, wavelet frequency division and reconstruction, etc.; then, through rock physics analysis, the logging curves sensitive to siliceous rocks are optimized. Through well-seismic matching, the dominant seismic inversion method is optimized, and the fine characterization of siliceous dolomite in the whole area is carried out. Based on the actual thickness of siliceous dolomite in each small layer underground, the accuracy of seismic interpretation method in predicting thickness is verified. Along with the distribution characteristics of faults in the study area, the genesis and distribution characteristics of siliceous dolomite in the whole area are analyzed (Figure 3).

3.3. Improving Resolution Processing of Seismic Data

3.3.1. Post-Stack Seismic Data Mining

In the process of well-seismic calibration, a synthetic seismogram is a bridge connecting underground geological information and ground seismic data. Its core idea is to simulate the propagation response of seismic waves in underground media through mathematical models. It is a convolution model based on reflection coefficient sequence and seismic wavelet, combined with certain noise information [32,33,34]. In theory, the reflection coefficient sequence can reflect the difference in the wave impedance of underground rock. The formation interface with a sudden change in wave impedance will produce the reflection coefficient. After convolution operation of the reflection coefficient and wavelet, a strong reflection axis will appear at the extreme point of reflection coefficient [35].
However, the seismic wavelet is a signal with limited bandwidth, and the earth is a low-pass filter, so the high-frequency components will attenuate more when propagating in the stratum, resulting in the effective frequency band of seismic wave narrowing [36,37]. Especially when the seismic wave propagates in the strata above 6000 m, under conditions of high speed and low frequency, a large number of high-frequency signals are attenuated; the energy of low-frequency signals is strong, and the energy of high-frequency signals is weak, resulting in some high-frequency signals being covered by low-frequency ones, which greatly reduces the resolution and credibility of seismic data.
The wavelet frequency division processing method is to preserve the high-frequency energy to the greatest extent, weaken the proportion of other frequency bands, and relatively increase the high-frequency energy to avoid being covered by the low-frequency signal, so as to excavate the high-frequency information. Therefore, the wavelet frequency division processing method can improve the seismic resolution in the time–frequency domain, highlight the detailed changes, obtain clearer post-stack data, and apply it to the fine characterization of thin-layer geological bodies (Formula (1)) [38,39,40]. As shown in Figure 4 and Figure 5, the single-frequency body with wavelet frequency divisions of 30 Hz, 40 Hz, and 50 Hz and 30 + 40 + 50 Hz reconstruction body are excavated, respectively. With the increase in the main frequency, the seismic resolution also increases.
R = λ 4 = v T 4 = v 4 f
R is the longitudinal resolution, λ is the main wavelength of the seismic wave, T and f are the main period and the main frequency, respectively, and v is the propagation velocity of the seismic wave in the medium. The vertical resolution limit of the root seismic data is λ/4, λ = VT = V/f. That is, the higher the dominant frequency, the smaller the limit, and the stronger the ability to distinguish geological features vertically.
The navigation-pyramid processing method primarily involves downsampling. In the frequency domain, the information of different scales is obtained by the radial filter and directional controllable filter, and then, the information of different directions is decomposed. Through the structure similar to the pyramid, the information of different scales is classified. The pyramid structure is bottom–up, the classification series is gradually increased, the image becomes more blurred, and the resolution is lower [41,42,43]. As shown in Figure 4 and Figure 5, the 1-level, 2-level, and 3-level seismic data from the navigation pyramid and 1-level + 2-level reconstruction data are mined, respectively. Among these, the lower the classification level is, the higher the seismic resolution is.

3.3.2. Post-Stack Seismic Data Optimization

For the accuracy of the excavated information, we can extract the wavelet of the well-side channel of different seismic data bodies and carry out convolution operation with the same reflection coefficient sequence. Through the simulated synthetic seismic record, it is compared with the corresponding processed types of seismic cross-well profiles. The higher the degree of conformity, the more reliable the information excavated by the corresponding seismic data.
Taking the original seismic data, wavelet frequency division 30 Hz, and navigation-pyramid level 2 as an example, the authenticity of the excavated information is verified by comparing the coincidence degree between the seismic trace and the seismic profile, and the best-advantage seismic data is selected. As shown in Figure 5, the original seismic data failed to effectively mine the trough information at I and the peak information at II, resulting in poor well-seismic coincidence rate and low resolution of seismic data. The resolution of wavelet frequency division 30 Hz seismic data improved, but it was still not enough. The trough information was effectively excavated at III, but the peak information at IV was not matched with the synthetic seismic record. The navigation-pyramid level-2 classification seismic data had the highest resolution and the highest coincidence rate with the synthetic seismic record on the well. Compared with the original data and frequency-division data, the trough event at V and the peak event at VI were effectively excavated. Briefly, as shown in Figure 4, the other frequency division, classification, and reconstruction bodies have the problem of low coincidence rate or low resolution. Finally, the navigation-pyramid level-2 data is preferred as the best-advantage data.

4. Experiment and Result

4.1. Parameter Sensitivity Analysis

Because of its high resolution and intuitive rock electrical characteristics, logging curve data is the basic data for reservoir prediction. According to previous research results, siliceous dolomite has logging response characteristics such as high resistance, low neutron (CNL), and low acoustic time difference. Based on this, the intersection analysis of sensitive curves is carried out. As shown in Figure 6a, the neutron and acoustic wave curves are not sensitive enough to siliceous dolomite and cannot be effectively distinguished from ordinary dolomite. Among them, the flushing zone resistivity value (RXO) reflects the rock resistivity value after the invasion of drilling fluid filtrate. As shown in Figure 6b, it can be seen from the log comprehensive histogram that siliceous dolomite is mostly developed in the upper part of the fourth member of the Dengying Formation in the study area. Although the quartz content is practically less than 50%, siliceous rocks are formed due to quartz itself. In high-resistivity minerals, combined with the low porosity and permeability of quartz particles caused by silicification, it is difficult for the intrusive conductive filtrate to enter the rock; hence, the siliceous dolomite has the characteristics of high resistivity (RXO). Most siliceous dolomite and ordinary dolomite can be distinguished by resistivity.
As shown in Figure 6, the cross analysis of the logging sensitive parameters, resistivity and density, was carried out on the dolomite, siliceous rock, and limestone in the fourth section of the Dengying Formation in the study area. It can be seen that the resistivity values of limestone and siliceous rock are above 20,000, and the resistivity value of dolomite is low, below 20,000. According to Figure 2, the limestone content in the upper part of the fourth member of the Dengying Formation is less, and it is mainly concentrated on the top of the fourth member of the Dengying Formation (sublayer 1), while siliceous dolomite is mainly distributed in sublayer 2–sublayer 6. Therefore, the prediction of siliceous rock with sublayer 2–sublayer 6 as the target layer can basically exclude the influence of limestone. Therefore, the resistivity curve is optimized as a sensitive logging parameter for the identification of siliceous rock.

4.2. Identification Effect of Thin Siliceous Rock

Seismic inversion is a process of solving the distribution characteristics of underground geological bodies by using pure wave seismic data, based on geological theory and objective function. Seismic inversion is a bridge connecting seismic data and geology. It can convert seismic profiles into lithologic cutting surfaces for geological interpretation and more directly show the spatial structure and physical properties of underground rock strata and reservoirs. It is one of the most effective reservoir prediction techniques for post-stack seismic data. As shown in Figure 7, through the comparative analysis of various inversion methods (waveform difference simulation, target probability simulation, waveform indication simulation), combined with the actual thickness of siliceous rock at the top of the fourth member of the underground Dengying Formation, the inversion method with the highest degree of conformity and the best resolution is selected.

4.2.1. Methodology

The frequency bands of seismic inversion are as follows: low frequency (0–8 Hz), medium frequency (8–80 Hz), and high frequency (above 80 Hz). Among them, the low-frequency band is obtained by logging information, the intermediate-frequency information is obtained by seismic wave impedance inversion, and the high-frequency information is obtained by waveform phase control simulation. The acquisition of high-frequency information is key to the realization of high-resolution inversion. Waveform indication simulation, waveform difference simulation, and target probability simulation are all inversion methods based on software for seismic waveform indication inversion (SMI5.0, Chengdu China) [44,45].
The waveform indication simulation involves inputting the logging information through the seismic waveform. Different from the waveform indication inversion, the input logging parameters are not limited to the wave impedance but can include any parameters sensitive to the reservoir, such as the resistivity curve and the corresponding seismic waveform. A matching relationship is established, the lateral variation of the seismic waveform is fully used to reflect the resistivity change characteristics, and then the phase change characteristics at high frequencies are simulated; then, the high-frequency information is gradually determined, which is suitable for reservoir prediction with strong heterogeneity [46,47,48].
Waveform difference simulation is primarily based on waveform similarity to optimize statistical samples, compare the predicted seismic trace waveform with the known seismic trace waveform, optimize the most similar well samples, establish the eigenvector variation function model, effectively establish the corresponding relationship between seismic waveform and sensitive parameters, and simulate the lateral variation characteristics of resistivity and the spatial distribution of siliceous dolomite [49,50].
The waveform probability simulation is based on the Bayesian theorem, which transforms the inversion problem into a probability distribution problem. It constrains the reasonable range of parameters through logging data (such as velocity, density, etc.), integrates geological knowledge to define the prior distribution, combines the Gaussian likelihood function to construct the matching degree between the target prediction data and the measured data, and constructs the probability profile through the random sampling algorithm to construct the posterior probability distribution of the model parameters [51,52].

4.2.2. Waveform Difference Simulation

The principle of the waveform difference simulation method is to optimize the underground model by minimizing the difference between the simulated waveform and the observed waveform. Its advantage is that it has high resolution and is suitable for simulation and interpretation of complex areas and complex geological bodies. In the simulated profile, siliceous dolomite is characterized by high resistivity inversion value, and ordinary dolomite is characterized by low resistivity inversion value.
As shown in Figure 7a, the waveform difference simulation method has a good recognition effect on siliceous dolomite in the PS8 and PS9 well areas, and has a poor recognition effect on the JT1 and PS7 wells. The prediction effect of the upper siliceous dolomite in the JT1 well is not obvious, and the effects of the upper siliceous rocks in the JT1 well area are intermittent, which does not conform to the characteristics of the continuous distribution of siliceous hydrothermal cooling deposits in the study area, and there is no logging interpretation conclusion of the lower siliceous dolomite in the JT1 well. This kind of inversion causes the error that there is less upper siliceous dolomite and more lower siliceous dolomite, which increases the ambiguity of the inversion results. In the PS7 well area, the recognition effect of the upper siliceous dolomite is poor, and it cannot be identified clearly. The lower siliceous dolomite is identified, but the overall recognition ability is poor and weakly displayed.

4.2.3. Target Probability Simulation

The principle of the target probability simulation method is based on the random simulation of the probability framework, which converts the seismic data into the probability distribution of the geological target. Its advantage is that it can fully combine the downhole interpretation conclusions, which is in good agreement with the well, but the randomness of the interwell simulation is strong. In the simulation profile, siliceous dolomite is characterized by high probability value, and ordinary dolomite is characterized by low probability value.
As shown in Figure 7b, the target probability simulation method has a good recognition effect on siliceous dolomite in the JT1 and PS8 well areas, and has a poor recognition effect on the PS7 and PS9 wells. For the PS7 well, the target probability simulation method has poor resolution; only one set of siliceous dolomite layers is identified, and the boundary of two upper and lower layers cannot be effectively identified. The ordinary dolomite layer between the two layers is interpreted as siliceous dolomite, resulting in a siliceous dolomite layer of high thickness. For the PS9 well, it can be seen that the upper and lower sets of siliceous dolomite from the PS8 well are stably extended to the PS9 well area. However, according to the logging interpretation conclusion, the PS9 well has only the lower siliceous dolomite layer. This method will lead to a thick layer of siliceous dolomite in the PS9 well area.

4.2.4. Waveform Indication Simulation

The principle of waveform indication simulation method directly uses seismic waveform to drive logging information and establishes the statistical relationship between geological parameters and waveform characteristics. Its advantage is that it can compensate for the lack of seismic wave frequency bands, effectively improve the inversion resolution, and restore the structural details. In the simulation profile, siliceous dolomite is characterized by high resistivity inversion values, and ordinary dolomite is characterized by low resistivity inversion values.
As shown in Figure 7c, the waveform indicator simulation method has a good recognition effect on siliceous dolomite in the JT1, PS7, PS8, and PS9 well areas, and has the highest coincidence rate with the interpretation conclusion of underground siliceous dolomite. The upper siliceous dolomite in the JT1 well area is accurately depicted, and the thickness is consistent with the logging interpretation. In the direction of the JT1 well to PS7 well, the siliceous dolomite gradually thickens, and two layers of siliceous dolomite begin to appear in the PS7 well area, which is also consistent with the conclusion of the downhole interpretation. In the direction of the PS7 well to PS8 well, siliceous dolomite is stably deposited, and there is a slight thickening trend. The simulation results are in good agreement with the conclusions of the PS8 downhole interpretation. In the direction of the PS8 well to PS9 well, the upper siliceous dolomite gradually becomes thinner, and it is not developed in the PS9 well, while the lower siliceous dolomite layer gradually becomes thinner, which is consistent with the conclusion of the logging interpretation.
Through the optimization of seismic data mining and inversion methods, the wavelet frequency division-2 seismic data volume and resistivity waveform indication simulation method are adopted. The coincidence rate between the high-resistivity section near the well and the underground siliceous rock section is high, and this is the main method adopted in this study for prediction of upper siliceous rock thickness. Figure 7c shows the resistivity waveform indication simulation profile of the fourth member of the Dengying Formation in the JT1-PS7-PS8-PS9 wells. The resistivity waveform indication simulation shows that the two sets of siliceous dolomite are mainly distributed in the PS7 and PS8 well areas, and the upper siliceous dolomite is mainly distributed in the JT1, PS7, and PS8 well areas. In the direction of the PS9 well area, the siliceous rocks gradually thin to undeveloped; the siliceous dolomite in the lower layer is mainly distributed in the PS9, PS8, and PS7 well areas. In the direction of the JT1 well, the siliceous rocks gradually thin to undeveloped.

4.3. Distribution Characteristics of Thin-Layer Siliceous Rocks

It can be seen from Figure 6b that there is a significant difference in the resistivity value between siliceous dolomite and dolomite rock. The resistivity value is greater than 20,000 in siliceous dolomite, and the resistivity value is less than 20,000 in dolomite (limestone is mainly distributed in sublayer 1, so it is neglected).
It can be seen from Figure 7c that the waveform indication simulation method establishes a matching relationship between the resistivity curve of each well and the corresponding seismic waveform, and makes full use of the lateral variation of the seismic waveform to simulate the resistivity variation characteristics of the study area. Then, on the basis of the resistivity waveform indication simulation, the threshold value of the waveform indication simulation value is set to 20,000 (the resistivity waveform indication simulation value is greater than 20,000 in the siliceous dolomite area, and less than 20,000 in the dolomite area). The inversion value within the threshold value is counted in each small layer, and the thickness of the siliceous rock in each small layer is calculated, which can intuitively show the thickness and distribution range of the siliceous rock on the plane.
Vertically, the upper siliceous rock is thin, and the siliceous rock in small layer 3 is slightly thicker than that in small layer 2. Horizontally, the high-value area of siliceous rock thickness in small layer 2 is developed in the JT1 well area, PS13 well area, and the southern part of the study area. The thickness of siliceous rock is low and distributed in contiguous areas (Figure 8a). The thickness of siliceous rocks in sublayer 3 has increased, and the plane distribution is mainly distributed in contiguous areas. The high-value areas of siliceous rocks are mainly distributed in the PS8 and PS106 well areas, and the strip distribution characteristics can be seen in the south of the study area (Figure 8b).
Vertically, the high-value area of the thickness of the lower siliceous rock is mainly developed in small layer 5 and small layer 6, and the thickness of the upper small layer 4 siliceous rock is the least. Horizontally, the siliceous rocks of sublayer 4 are mainly developed in the PS9 well area and the southwest of the study area, and are distributed in contiguous areas as a whole (Figure 9a). The siliceous rocks in the small layer 5 are the most developed and distributed in most areas of the whole study area. The PS7-PS8-PS9 well area is the most developed, and the distribution pattern is mainly irregular clumps (as shown in Figure 9b). The plane size of the siliceous rocks in the small layer 6 is slightly smaller, mainly developed in the PS7-PS8 well area, and the siliceous rocks are almost no longer developed in the southwest of the study area (as shown in Figure 9c).

5. Discussion

5.1. Accuracy Analysis

According to the comparative analysis of the actual siliceous rock thickness and the predicted thickness of each small layer at the top of the fourth section of the underground Dengying Formation, the absolute value of the difference between the actual siliceous rock thickness and the predicted thickness is used as the absolute error to verify the accuracy of the prediction results. As shown in Table 1, the thickness of siliceous rock in small layer 2 is generally small, only distributed in the PS7 and JT1 wells. The uniform lithology is conducive to improving the prediction accuracy, and the prediction error value is generally small. Only the JT1 well has a large thickness of siliceous rock, and the error value of small layer siliceous rock thickness is 2 m. As shown in Table 2, the thickness of the siliceous rock in small layer 3 is gradually increasing, mainly distributed in the PS106 and PS8 well areas, and the prediction error value is generally in the range of 0–1 m.
As shown in Table 3, the prediction error of the siliceous rock thickness of the small layer 4 is in the range of 0–1.6 m, which is still small compared with the siliceous rock thickness of the underground 0–15.1 m; as shown in Table 4, the thickness of siliceous rock in the fifth layer is the largest. Because of its large thickness of siliceous rock, it causes the difference in seismic waveform on seismic data. Therefore, the recognition effect is the best, the prediction error range is the smallest, and the error value is concentrated in the range of 0–1.1 m; as shown in Table 5, the thickness of the siliceous rock in the small layer 6 is reduced, and the overall thickness is 0–10.5 m. The error value in the northern part of the study area is still small. The DB1 well area in the south, due to its small thickness in the small layer 3, only 3 m, cannot reach the seismic data identification limit, so it cannot be identified. This shows that the method is reliable for predicting the thickness of the upper siliceous rock in the fourth member of Dengying Formation.

5.2. Analysis of Fracture Distribution

Typically, the fault can communicate with the deep magma, and the dissolved SiO2 carried in the hydrothermal solution can migrate upward along the fault and cool and precipitate in the dolomite to form siliceous bands [53]. Among them, the openness of the tensile fault is the highest, and this can provide an ascending channel for the silicon-rich hydrothermal fluid and form a large area of silicified zone. The poor opening of shear fracture will lead to limited fluid activity and the formation of local siliceous veins. The strike-slip fault has a good opening, and the secondary faults are developed in the local tension zone; the siliceous dolomite is usually distributed in an echelon [54,55].
Figure 10a shows a superposition diagram of coherent attributes and faults in the four sections of Dengying Formation. The coherent attribute is an attribute that calculates the similarity of seismic data in the horizontal direction. The white high-value area indicates that the adjacent seismic traces are similar in waveform, the strata are continuous and uniform, and the faults are not developed. The black low-value area indicates that the waveform of adjacent seismic traces is very different, which can be used as an indication of fracture. The fracture interpretation of the whole area is completed by combining the coherence attribute with the seismic profile.
Figure 10a shows the superposition diagram of the coherence attributes and faults of the four sections of the Dengying Formation. Most of the faults in the study area are NE-SW-trending, and SE-NW-trending small faults are locally developed. Most of the faults have steep dip angles, small fault distances, and strike-slip fault properties. In particular, the faults in the JT1-PS7-PS8-PS9 well area are more developed, and the strike is mainly NE-SW. Figure 10b shows a planar distribution and fault superposition diagram of siliceous dolomite. It can be seen that siliceous dolomite is mostly concentrated in the fault development area, mostly distributed in a straight line and an echelon along the fault, and the closer to the fault zone, the more developed the siliceous dolomite is.

6. Conclusions

The study shows that the siliceous dolomite of the fourth member of Dengying Formation in Penglai gas area has obvious layered distribution characteristics and shows strong heterogeneity both vertically and horizontally. The siliceous dolomite is mostly developed near the fault zone, which is similar to the plane distribution characteristics of the fault and is greatly affected by the fault activity.
Conventional siliceous rock interpretation is usually based on conventional post-stack seismic data to characterize the whole set of strata in the target layer. The improved reservoir imaging technology mainly focuses on the mining and optimization of post-stack seismic data, improving seismic resolution, and subdividing small layers and independent interpretation of each small layer. Based on the seismic data processed by the navigation-pyramid classification, it is often possible to avoid the reduction in seismic resolution due to the attenuation of high-frequency components, thereby improving the characterization accuracy of thin-layer dolomite. The navigation-pyramid secondary-classification seismic data can significantly improve the prediction accuracy of siliceous dolomite. At the same time, the method of subdividing the upper and lower sets of siliceous dolomite into small layers and interpreting each small layer independently can limitedly avoid the overall interpretation of the target layer. The obtained siliceous dolomite is a comprehensive reflection of the whole, lacking fine description of internal details, which helps improve the accurate characterization of the internal heterogeneity of siliceous dolomite.
In the vertical direction, the siliceous dolomite in Penglai gas area has the characteristics of layered distribution. Due to the cooling and condensation of deep hydrothermal fluid from the bottom to top, the thickness of the upper and lower layers is obviously different, and the extension distance of the lower layer is small. The thickness of siliceous dolomite is large and concentrated in the middle of the upper part of the fourth member of Dengying Formation. The upper extension distance is high, and the thickness of siliceous dolomite is small and concentrated in the upper part of the fourth member of Dengying Formation. Laterally, the siliceous dolomite in Penglai area is also relatively distributed in the area with more concentrated fault development, mainly distributed in the JT1-PS7-P8-PS9 well area in the north, showing a continuous sheet distribution. In the south, the fault development is relatively small, and the development scale of siliceous dolomite is also small, showing a strip-shaped distribution.
The limitation of this study is that it only depicts the spatial distribution characteristics of the two sets of siliceous dolomite in the upper and lower parts of the fourth member of the Dengying Formation, and finds that the spatial distribution of siliceous dolomite is similar to the distribution characteristics of faults. However, there is no specific analysis of the causes of siliceous dolomite formation, and further research and analysis are needed on the formation mechanism and influencing factors of siliceous dolomite.
To study the causes of siliceous dolomite formation, a subsequent field geological survey will be carried out to observe the occurrence of siliceous dolomite and the spatial distribution of faults and to identify faults and hydrothermal channels. Siliceous source analysis (such as Si-O isotope tracing) and the matching of hydrothermal age and regional tectonic events will be carried out to clarify whether the hydrothermal activity is consistent with the time of fault activity.

Author Contributions

Methodology, Y.W.; Software, C.H.; Validation, C.N.; Investigation, P.L.; Data curation, R.D.; Writing—original draft, S.L.; Visualization, B.Z.; Supervision, L.Z. All authors have read and agreed to the published version of the manuscript.

Funding

PetroChina Southwest Oil and Gas Field ‘Jie Bang Gua Shuai’ scientific and technological research project ‘Study on the Stereoscopic Accumulation Process and Mechanism of Sinian-Permian in Penglai Gas Area’ (JS2022-181).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data involving privacy is not shared.

Conflicts of Interest

Authors Baoshou Zhang, Ruixue Dai and Changxingyue He were employed by the company Exploration and Development Research Institute of PetroChina Southwest Oil and Gas Field Company. Author Peng Lu was employed by the company Southwest Branch, CNPC Logging Company Limited. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Siddique, M.S.; Hashim, N.A.; Junaidi, M.U.M.; Ullah, A.; Yusoff, R.; Rabuni, M.F. Recent advances in the application of dolomite in membrane separation and beyond: A review on an abundant and versatile mineral. Mater. Today Sustain. 2024, 28, 100951. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:00129380690000 (accessed on 20 June 2025). [CrossRef]
  2. Yan, B.; Luo, F.; Cao, Y.; Cheng, L. Dolomitization and main controlling factors of Penglaiba formation reservoir in Tarim Basin. Xinjiang Pet. Geol. 2025, 46, 419–428, (In Chinese with English Abstract). [Google Scholar]
  3. Metz, P.; Milke, R. Mechanism and kinetics of forsterite formation in metamorphic siliceous dolomites: Findings from a rock-sample experiment. Eur. J. Mineral. 2012, 24, 771. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000307610600020 (accessed on 15 June 2025). [CrossRef]
  4. Thomas, E. Contact Metamorphism in Siliceous Limestone and Dolomite in Marble Canyon and Geology of Related Intrusion, Culberson County, Trans-Pecos Texas. Ph.D. Thesis, The University of Texas at Austin, Austin, TX, USA, 2005. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/PQDT:66224229 (accessed on 20 June 2025).
  5. Wang, Z.; Wang, X.; Kang, J. Stratigraphy subdivision and exploration implications of Cambrian Qiong-zhusi Formation in Deyang-Anyue aulacogen, Sichuan Basin. Lithol. Reserv. 2025, 37, 97–110, (In Chinese with English Abstract). [Google Scholar]
  6. Kernen, R.A.; Giles, K.A.; Poe, P.L.; Gannaway Dalton, C.E.; Rowan, M.G.; Fiduk, J.C.; Hearon, T.E. Origin of the Neoproterozoic rim dolomite as lateral carbonate caprock, Patawarta salt sheet, Flinders Ranges, South Australia. Aust. J. Earth Sci. 2019, 67, 815–832. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000470641900001 (accessed on 17 June 2025). [CrossRef]
  7. Wang, Q.; Zhang, L.; Lu, X.; Zhou, L.; Wang, R. Hydrocarbon accumulation types and distribution prediction of western section of frontal uplift of Kuqa foreland basin. Acta Pet. Sin. 2023, 44, 730–747, (In Chinese with English Abstract). [Google Scholar]
  8. Bourli, N.; Kokkaliari, M.; Iliopoulos, I.; Pe-Piper, G.; Piper, D.J.W.; Maravelis, A.G.; Zelilidis, A. Mineralogy of siliceous concretions, cretaceous of ionian zone, western Greece: Implication for diagenesis and porosity. Mar. Pet. Geol. 2019, 105, 45–63. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000469896400004 (accessed on 25 June 2025). [CrossRef]
  9. Felhi, M.; Saidi, R.; Fattah, N.; Tlili, A. Textural evidences for dissolution of silica-rich rocks of the Ypresian phosphatic series, Gafsa-Metlaoui basin, southwestern Tunisia: Implication of biogenic silica supply on genesis of fibrous clays. Arab. J. Geosci. 2016, 9, 695. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000387152300013 (accessed on 18 June 2025). [CrossRef]
  10. Joshi, P.; Sharma, R. Fluid inclusion and geochemical signatures of the talc deposits in Kanda area, Kumaun, India: Implications for genesis of carbonate hosted talc deposits in Lesser Himalaya. Carbonates Evaporite 2015, 30, 153–166. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000354992200005 (accessed on 20 June 2025). [CrossRef]
  11. Peng, B.; He, M.; Yang, M.; Liu, X.; Sui, X.; Sun, K.; Wu, S. Petrogenesis of Jian forsterite jade solely composed of end-member forsterite (Fo 99.8): Constrained by trace element and oxygen isotope. Ore Geol. Rev. 2022, 150, 105167. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000882199900002 (accessed on 19 June 2025). [CrossRef]
  12. Jain, R.; Bhu, H.; Purohit, R. Application of Thermal Remote Sensing Technique for Mapping of Ultramafic, Carbonate and Siliceous Rocks using ASTER Data in Southern Rajasthan, India. Curr. Sci. 2020, 119, 954. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000572677100026 (accessed on 16 June 2025). [CrossRef]
  13. Prasad, B.N.V.S.; Murthy, V.; Naik, S. Drillability predictions in Aravalli and Himalayan rocks–a petro-physico-mechanical approach. Curr. Sci. 2022, 122, 907–917. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000795008100009 (accessed on 20 June 2025). [CrossRef]
  14. Kuleshov, V.N.; Georgievskii, A.F.; Bugina, V.M. Isotopic Composition (δ13C, δ18O) and Genesis of Carbonates from Phosphorite Deposits in the Lesser Karatau (Kazakhstan). Lithol. Miner. Resour. 2020, 55, 111–130. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000520961300004 (accessed on 15 June 2025). [CrossRef]
  15. Giuntoli, F.; Engi, M. Internal geometry of the central Sesia Zone (Aosta Valley, Italy): HP tectonic assembly of continental slices. Swiss J. Geosci. 2016, 109, 1–27. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000390028300010 (accessed on 20 June 2025). [CrossRef]
  16. Zhang, Z.; Zhao, L.; Zhang, D.; Li, Q.; Chen, H.; Wen, L.; Zhang, B.; Zhou, G.; Zhong, Y.; Li, W. Diagenetic evolution and cementation mechanism in deep Carbonate reservoirs: A case study of Dengying Fm. 2 in Penglai, Sichuan Basin, China. Mar. Pet. Geol. 2024, 170, 107084. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:001319720700001 (accessed on 21 June 2025). [CrossRef]
  17. Tang, H.S.; Wu, G.; Lai, Y. The C-O isotope geochemistry and genesis of the Dashiqiao magnesite deposit, Liaoning Province, NE China. Acta Petrol. Sin. 2009, 25, 455–467. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000265945400019 (accessed on 16 June 2025).
  18. Feng, J.L.; Zhao, Z.H.; Chen, F.; Hu, H.-P. Rare earth elements in sinters from the geothermal waters (hot springs) on the Tibetan Plateau, China. J. Volcanol. Geotherm. Res. 2014, 287, 1–11. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000346551900001 (accessed on 18 June 2025). [CrossRef]
  19. Luo, W.; Xu, W.; Liu, Y. Origin of siliceous rocks in Sinian Dengying 4 Member, Gaoshiti area, Sichuan Basin. Nat. Gas Explor. Dev. 2019, 42, 1–9, (In Chinese with English Abstract). [Google Scholar]
  20. Shen, A.; Hu, A.; Tan, X.; Qiao, Z.; Zheng, J.; Pan, L.; She, M. Geneses of dolomites and dolomite reservoirs: Review, advances, and prospects. Oil Gas Geol. 2025, 46, 740–758, (In Chinese with English Abstract). [Google Scholar]
  21. Xia, M.; Wen, L.; Luo, B.; Xu, S.; Zhang, X.; Zhu, Y. Tectono-sedimentary mechanism and evolution model of the Deyang-Anyue intracratonic rift in Sichuan Basin. Nat. Gas Ind. 2025, 45, 50–67, (In Chinese with English Abstract). [Google Scholar]
  22. Li, G.; Ziran, S.; Xu, Y.; Li, H.; Li, Z.; Wang, Q. Prediction of structural fractures in Deng 4 member reservoirs in the Shehong-Yanting block, Penglai gas field, Sichuan Basin. Xinjiang Pet. Geol. 2025, 46, 136–143, (In Chinese with English Abstract). [Google Scholar]
  23. Wei, G.; Zhang, B.; Xie, Z. Main controlling factors for the formation of deep to ultra-deep carbonate giant gas fields: A case study of Anyue and Penglai gas fields in Sichuan Basin. China Pet. Explor. 2025, 30, 1–15, (In Chinese with English Abstract). [Google Scholar]
  24. Feng, M.; Shang, J.; Shen, A.; Wang, L.; Xu, X.; Liang, L.; Liu, F.; Liu, X. Episodic hydrothermal alteration on Middle Permian carbonate reservoirs and its geological significance in southwestern Sichuan Basin, SW China. Pet. Explor. Dev. 2024, 51, 81–96. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:001172881100001 (accessed on 17 June 2025). [CrossRef]
  25. Li, Y.; Xu, S. Reservoir fracture cave characteristics of middle–lower Ordovician carbonate rocks in Tahe oilfield in Tarim Basin, China. Indian J. Mar. Sci. 2017, 46, 10. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000423136300004 (accessed on 20 June 2025).
  26. Jing, N.; Gang, Z.; Luya, Z.Y.W.; Yiting, S. Diagenetic fluid evolution and reservoir formation effect of the second Member of Dengying Formation in thenorthern slope of central Sichuan paleo-uplift. Acta Geol. Sin. 2025, 99, 879–896, (In Chinese with English Abstract). [Google Scholar]
  27. Wang, L.J. Characteristics and controlling factors of high-quality reservoirs of the fourth member of Dengying Formation in northern Sichuan Basin. Lithol. Reserv. 2019, 31, 35–45, (In Chinese with English Abstract). [Google Scholar]
  28. Zhang, L.; Li, B.; Zhu, X.; Yang, Y.; Xu, Z.; Dai, L.; Zhang, W.; Xu, Y.; Hu, L. Key controls and accumulation processes of ultra-deep hydrocarbon reservoirs: A case study of the Dengying Formation in the Yuanba Area, northern Sichuan Basin. Pet. Sci. Bull. 2025, 10, 415–429, (In Chinese with English Abstract). [Google Scholar]
  29. Wang, W.; Yang, Y.; Wen, L. Calcareous space model of Sinian Deyang-Anyue rift in the Sichuan Basin and its evolution. Nat. Gas Ind. 2025, 45, 48–59, (In Chinese with English Abstract). [Google Scholar]
  30. Ma, X.; Jiang, L.; Lu, X. Paleoenvironmental study on the formation of stratigraphic unconformities on the top of the Sinian in the Southwest and North Sichuan Basin. World Pet. Ind. 2025, 32, 54–64, (In Chinese with English Abstract). [Google Scholar]
  31. Li, X.T. Application of Multi-Seismic Attributes Analysis in Prospective Oil and Gas Area Evaluation. Adv. Mater. Res. 2014, 868, 150–153. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000335880900034 (accessed on 19 June 2025). [CrossRef]
  32. Klimes, L. Comparison of ray-matrix and finite-difference methods in a simple 1-D velocity model. J. Neurosurg. Sci. 2019, 63, 247–256. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000469250500005 (accessed on 20 June 2025). [CrossRef]
  33. Wu, X.; Caumon, G. Simultaneous multiple well-seismic ties using flattened synthetic and real seismograms. Geophys. J. Soc. Explor. Geophys. 2017, 82, IM13–IM20. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000396469700015 (accessed on 15 June 2025). [CrossRef]
  34. Wu, H.; Li, Z.; Liu, N.; Zhang, B. Improved seismic well tie by integrating variable-size window resampling with well-tie net. J. Pet. Sci. Eng. 2021, 208, 109368. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000697359700089 (accessed on 20 June 2025). [CrossRef]
  35. Belous, A.A.; Korol’Kov, A.I.; Shanin, A.V. Experimental Estimation of the Frequency-Dependent Reflection Coefficient of a Sound-Absorbing Material at Oblique Incidence. Acoust. Phys. 2018, 64, 158–163. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000428998200004 (accessed on 17 June 2025). [CrossRef]
  36. Mukhopadhyay, S.; Sharma, J.; Del-Pezzo, E.; Kumar, A. Study of attenuation mechanism for Garwhal–Kumaun Himalayas from analysis of coda of local earthquakes. Phys. Earth Planet. Inter. 2010, 180, 7–15. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000278238400002 (accessed on 18 June 2025). [CrossRef]
  37. Deheuvels, M.; Faucher, F.; Brito, D. Numerical and experimental study of ultrasonic seismic waves propagation and attenuation on high-quality factor samples. Geophys. Prospect. 2024, 72, 2015–2031. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:001138266700001 (accessed on 20 June 2025). [CrossRef]
  38. Kumar, N.; Ishu. BER Analysis in Wavelet Based SC-FDMA for LTE Uplink Transmission. IEEE. 2015. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000380393400078 (accessed on 16 June 2025).
  39. Tian, Y.; Gao, J.; Wang, D. Synchrosqueezing Optimal Basic Wavelet Transform and Its Application on Sedimentary Cycle Division. IEEE Trans. Geosci. Remote Sens. 2022, 60, 5908413. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000756892900058 (accessed on 15 June 2025). [CrossRef]
  40. Amini, M.; Sadreazami, H. Image Watermarking through Joint Spatial Segmentation and Wavelet Packet Frequency Division. In Proceedings of the 2012 11th International Conference on Signal Processing (ICSP 2012), IEEE, Beijing, China, 21–25 October 2012. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000349102800144 (accessed on 17 June 2025).
  41. Ma, L.Y.; Xie, W.; Huang, H.B. Convolutional neural network based obstacle detection for unmanned surface vehicle. Math. Biosci. Eng. 2019, 17, 845–861. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000495897300045 (accessed on 20 June 2025). [CrossRef] [PubMed]
  42. Han, X.; Yuan, Y.; Zhong, J.; Deng, J.; Wu, N. Water Segmentation for Unmanned Ship Navigation Based on Multi-Scale Feature Fusion. Appl. Sci. 2025, 15, 2362. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:001442321100001 (accessed on 18 June 2025). [CrossRef]
  43. Plenge-Feidenhans’L, M.K.; Blanke, M. Open Water Detection for Autonomous In-harbor Navigation Using a Classification Network. IFAC-Pap. 2021, 54, 30–36. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000714877600006 (accessed on 20 June 2025). [CrossRef]
  44. Du, J.; Liu, Y.C.; Bai, J.B. Prediction of tight sandstonereservoir based on waveform indication simulation. Bull. Geol. Sci. Technol. 2022, 41, 94–100, (In Chinese with English Abstract). [Google Scholar]
  45. Du, W.W.; Jin, Z.J.; Di, Y.X. The application of seismic wave-form indicator inversion and characteristic parameter simulationto thin reservoir prediction. Chin. J. Eng. Geophys. 2017, 14, 56–61, (In Chinese with English Abstract). [Google Scholar]
  46. Gao, J.; Bi, J.J.; Zhao, H.S.; Fu, Z. Seismic waveform inversiontechnology and application of thinner reservoir prediction. Prog. Geophys. 2017, 32, 142–145, (In Chinese with English Abstract). [Google Scholar]
  47. Sheng, S.C.; Bi, J.J.; Li, W.Z. Research on the seismic waveform indication simulation inversion (SMI) method. Inn. Mong. Petrochem. Ind. 2015, 41, 147–151, (In Chinese with English Abstract). [Google Scholar]
  48. Zhu, S.; Zhang, C.; Zhou, W. Quality evaluation and prediction of low permeability fine-grained sandstone reservoir ofshallow marine gravity flow: A case study of the First Member of Huangliu Formation in Dongfang A area of Yinggehai Basin. Mar. Geol. Front. 2025, 41, 1–13, (In Chinese with English Abstract). [Google Scholar]
  49. Liu, G.; Mao, S.; Song, P.; Tan, J.; Qin, D.; Jiang, X. Multiparameter full waveform inversion based on fluid-solid coupling equations of acoustic-viscoelastic waves. Oil Geophys. Prospect. 2025, 60, 700–709, (In Chinese with English Abstract). [Google Scholar]
  50. You, S.; Yang, R.; Yang, L.; Duan, Y.; Liu, X.; Xiao, C.; Zhang, X.; Yang, Z.; Li, C.; Li, D. Forward Simulation of Unfavorable Geological Bodies and Analysis of Waveform Characteristics Based on time Domain Finite Difference Method. Pure Appl. Geophys. 2025, 182, 1983–1999. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:001478430000001 (accessed on 19 June 2025). [CrossRef]
  51. Sun, Y.; Lu, J.; Wan, Y. Research on Probability Simulation Method for Well Interference of Tight Sand Gas Reservoir Based on Metro Carlo. Sci. Technol. Eng. 2016, 16, 40–45, (In Chinese with English Abstract). [Google Scholar]
  52. Zhang, M.; Zhong, X. Application of Monte Carlo Simulation to Economic Evaluation of Oil and Gas Development. J. Southwest Pet. Univ. (Soc. Sci. Ed. ) 2012, 14, 6–10, (In Chinese with English Abstract). [Google Scholar]
  53. Xing, L.; Li, X.; Cao, P.; Luo, J.; Jiang, H. Stepwise extraction and utilization of silica and alumina from coal fly ash by mild hydrothermal process. Process Saf. Environ. Prot. Trans. Inst. Chem. Eng. Part B 2024, 182, 918–929. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:001145008600001 (accessed on 19 June 2025). [CrossRef]
  54. Wang, H.; Liu, S.; Hou, M.; Zhang, B.; Song, J.; Zhao, R.; Ding, Y.; Han, Y.; Li, Z. Petrological and micrometer-scale geochemical constraints on chert origins in the Dengying Formation, Yangtze Block, South China: Implications for Late Ediacaran hydrothermal activity and tectonic setting. Precambrian Res. 2022, 370, 106531. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000788097700001 (accessed on 17 June 2025). [CrossRef]
  55. Dong, Y.; Xu, S.; Wen, L.; Chen, H.; Fu, S.; Zhong, Y.; Wang, J.; Zhu, P.; Cui, Y. Tectonic control of Guadalupian-Lopingian cherts in northwestern Sichuan Basin, South China-Science Direct. Palaeogeogr. Palaeoclimatol. Palaeoecol. 2020, 557, 109915. Available online: https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000565723200009 (accessed on 20 June 2025). [CrossRef]
Applsci 15 09631 g001
Figure 1. The location of the study area (a) (according to Exploration and Development Research Institute of PetroChina Southwest Oil and gas field company, Chengdu, China), and JT1, PS7, PS8, PS9, etc., are the investigated wells; the comprehensive histogram of the upper part of the fourth member of Dengying Formation (b).
Figure 1. The location of the study area (a) (according to Exploration and Development Research Institute of PetroChina Southwest Oil and gas field company, Chengdu, China), and JT1, PS7, PS8, PS9, etc., are the investigated wells; the comprehensive histogram of the upper part of the fourth member of Dengying Formation (b).
Applsci 15 09631 g001
Applsci 15 09631 g002
Figure 2. Lithology comparison section of the fourth section of the Dengying Formation along the JT1-PS7-PS8-PS9 well line.
Figure 2. Lithology comparison section of the fourth section of the Dengying Formation along the JT1-PS7-PS8-PS9 well line.
Applsci 15 09631 g002
Applsci 15 09631 g003
Figure 3. Technical flow chart.
Figure 3. Technical flow chart.
Applsci 15 09631 g003
Applsci 15 09631 g004
Figure 4. Comparison of wavelet frequency division, navigation-pyramid classification, and reconstruction volume. (The yellow line represents the top interpretation horizon of the fourth member of Dengying Formation. The blue line represents the bottom interpretation horizon of the upper sub-member of the fourth member of the Dengying Formation.) (a) The 40 Hz wavelet frequency divider; (b) 50 Hz wavelet frequency divider; (c) 30 + 40 + 50 Hz wavelet frequency division reconstruction; (d) navigation-pyramid level-1 classification; (e) navigation-pyramid level-3 classification; (f) navigation-pyramid level-1 + 2 reconstruction.
Figure 4. Comparison of wavelet frequency division, navigation-pyramid classification, and reconstruction volume. (The yellow line represents the top interpretation horizon of the fourth member of Dengying Formation. The blue line represents the bottom interpretation horizon of the upper sub-member of the fourth member of the Dengying Formation.) (a) The 40 Hz wavelet frequency divider; (b) 50 Hz wavelet frequency divider; (c) 30 + 40 + 50 Hz wavelet frequency division reconstruction; (d) navigation-pyramid level-1 classification; (e) navigation-pyramid level-3 classification; (f) navigation-pyramid level-1 + 2 reconstruction.
Applsci 15 09631 g004
Applsci 15 09631 g005
Figure 5. Comparison of four different seismic data calibrations of the Dengying Formation in the study area.
Figure 5. Comparison of four different seismic data calibrations of the Dengying Formation in the study area.
Applsci 15 09631 g005
Applsci 15 09631 g006
Figure 6. The intersection diagram of the sensitive parameters of the upper part of the four sections of the light shadow group. (a) Acoustic–neutron crossplot. (b) Resistivity–density crossplot.
Figure 6. The intersection diagram of the sensitive parameters of the upper part of the four sections of the light shadow group. (a) Acoustic–neutron crossplot. (b) Resistivity–density crossplot.
Applsci 15 09631 g006
Applsci 15 09631 g007
Figure 7. Multi-method inversion profile of the upper part of the fourth member of Dengying Formation. (a) Waveform difference simulation; (b) target probability simulation; (c) waveform indication simulation.
Figure 7. Multi-method inversion profile of the upper part of the fourth member of Dengying Formation. (a) Waveform difference simulation; (b) target probability simulation; (c) waveform indication simulation.
Applsci 15 09631 g007
Applsci 15 09631 g008
Figure 8. The thickness prediction plan of the upper siliceous rock in the fourth member of Dengying Formation. (a) Sublayer 2; (b) sublayer 3.
Figure 8. The thickness prediction plan of the upper siliceous rock in the fourth member of Dengying Formation. (a) Sublayer 2; (b) sublayer 3.
Applsci 15 09631 g008
Applsci 15 09631 g009
Figure 9. The thickness prediction plan of the lower siliceous rock layer of the fourth member of Dengying Formation. (a) Sublayer 4; (b) sublayer 5; (c) sublayer 6.
Figure 9. The thickness prediction plan of the lower siliceous rock layer of the fourth member of Dengying Formation. (a) Sublayer 4; (b) sublayer 5; (c) sublayer 6.
Applsci 15 09631 g009
Applsci 15 09631 g010
Figure 10. Plane distribution and fault distribution map of siliceous dolomite in the fourth member of Dengying Formation. (a) The four-section coherent attributes and fracture superposition diagram of Dengying Formation. (b) Plane distribution and fault superposition diagram of siliceous dolomite.
Figure 10. Plane distribution and fault distribution map of siliceous dolomite in the fourth member of Dengying Formation. (a) The four-section coherent attributes and fracture superposition diagram of Dengying Formation. (b) Plane distribution and fault superposition diagram of siliceous dolomite.
Applsci 15 09631 g010
Table 1. Data table for prediction error of siliceous rock thickness in the upper small layer 2 of the fourth member of Dengying Formation at the main well location.
Table 1. Data table for prediction error of siliceous rock thickness in the upper small layer 2 of the fourth member of Dengying Formation at the main well location.
WellThickness of Well
Siliceous Rock (m)		Thickness of Seismic
Siliceous Rock (m)		Absolute
Error
Small
layer 2		PS1060.00.00.0
DB10.00.00.0
PS150.00.00.0
PS130.00.00.0
PS80.0 0.0 0.0
PS90.0 0.0 0.0
PS70.6 0.0 0.0
JT11816 2
Table 2. Data table for prediction error of siliceous rock thickness in the upper small layer 3 of the fourth member of Dengying Formation at the main well location.
Table 2. Data table for prediction error of siliceous rock thickness in the upper small layer 3 of the fourth member of Dengying Formation at the main well location.
WellThickness of Well
Siliceous Rock (m)		Thickness of Seismic
Siliceous Rock (m)		Absolute
Error
Small
layer 3		PS1068.0 7.01
DB10.0 0.00
PS150.0 0.0 0.0
PS130.0 0.0 0.0
PS812.0 11.5 0.5
PS90.0 0.00.0
PS700.00.0
JT10.0 0.00.0
Table 3. Data table for prediction error of siliceous rock thickness in the lower small layer 4 of the fourth member of Dengying Formation at the main well location.
Table 3. Data table for prediction error of siliceous rock thickness in the lower small layer 4 of the fourth member of Dengying Formation at the main well location.
WellThickness of Well
Siliceous Rock (m)		Thickness of Seismic
Siliceous Rock (m)		Absolute
Error
Small
layer 4		PS1060.0 0.0 0.0
DB10.0 0.0 0.0
PS150.0 0.0 0.0
PS130.0 0.0 0.0
PS81.28 01.28
PS915.113.5 1.6
PS711.5 101.5
JT10.0 0.00.0
Table 4. Data table for prediction error of siliceous rock thickness in the lower small layer 5 of the fourth member of Dengying Formation at the main well location.
Table 4. Data table for prediction error of siliceous rock thickness in the lower small layer 5 of the fourth member of Dengying Formation at the main well location.
WellThickness of Well
Siliceous Rock (m)		Thickness of Seismic
Siliceous Rock (m)		Absolute
Error
Small
layer 5		PS1060.0 0.0 0.0
DB10.00.00.0
PS150.0 0.0 0.0
PS130.0 0.0 0.0
PS816.615.5 1.1
PS94.9 4.70.2
PS712.5 11.90.6
JT10.00.00.0
Table 5. Data table for prediction error of siliceous rock thickness in the lower small layer 6 of the fourth member of Dengying Formation at the main well location.
Table 5. Data table for prediction error of siliceous rock thickness in the lower small layer 6 of the fourth member of Dengying Formation at the main well location.
WellThickness of Well
Siliceous Rock (m)		Thickness of Seismic
Siliceous Rock (m)		Absolute
Error
Small
layer 6		PS1060.0 0.0 0.0
DB13.00.03.0
PS150.0 0.0 0.0
PS130.0 0.0 0.0
PS86.1 5.8 0.3
PS90.0 0.00.0
PS710.5 9.6 0.9
JT10.00.050.05
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Liu, S.; Zhou, L.; Zhang, B.; Dai, R.; Lu, P.; He, C.; Wu, Y.; Ni, C. Thin-Layer Siliceous Dolomite Prediction Based on Post-Stack Seismic Data Mining and Optimization. Appl. Sci. 2025, 15, 9631. https://doi.org/10.3390/app15179631

AMA Style

Liu S, Zhou L, Zhang B, Dai R, Lu P, He C, Wu Y, Ni C. Thin-Layer Siliceous Dolomite Prediction Based on Post-Stack Seismic Data Mining and Optimization. Applied Sciences. 2025; 15(17):9631. https://doi.org/10.3390/app15179631

Chicago/Turabian Style

Liu, Shimin, Lu Zhou, Baoshou Zhang, Ruixue Dai, Peng Lu, Changxingyue He, Yong Wu, and Chunyong Ni. 2025. "Thin-Layer Siliceous Dolomite Prediction Based on Post-Stack Seismic Data Mining and Optimization" Applied Sciences 15, no. 17: 9631. https://doi.org/10.3390/app15179631

APA Style

Liu, S., Zhou, L., Zhang, B., Dai, R., Lu, P., He, C., Wu, Y., & Ni, C. (2025). Thin-Layer Siliceous Dolomite Prediction Based on Post-Stack Seismic Data Mining and Optimization. Applied Sciences, 15(17), 9631. https://doi.org/10.3390/app15179631

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop