Search

The commonly-used tools for oil-source correlation, such as mass chromatograms of biomarkers and bivariate cross-plots of geochemical parameters, cannot deal with multiple geochemical parameters and plenty of samples simultaneously, leading to uncertainties in the results and even failures sometimes. In this paper, the discriminant analysis (DA) is selected from supervised machine learning algorithms, as it is superior to the commonly-used tools as well as other multivariate statistical methods, with the accumulation of geochemical data of source rocks and oils. In the slope of the Qikou Depression, the main source rocks in the third (Es3) and first members (Es1) of the Paleogene Shahejie Formation were deposited in similar depositional environments. The source rocks cannot be distinguished with the commonly-used tools. We firstly extended geochemical parameters and then used stepwise DA to select informative parameters and to develop a discriminant model for oil-source correlation. The 22 selected parameters are supported by geochemical characteristics of the source rocks in the study area. The DA of these parameters for oil-source correlation achieved a high correct rate of original validation (96.8%) and leave-one-out cross-validation (89.4%), indicating a sufficient discriminatory power. The oil-source correlation results with high posterior probabilities, showing strong similarity between the sources and oils, coincide with geological conditions and illustrate that there is still much exploration potential in the study area, especially for the Es3 petroleum system. All these illustrate that DA is one of the most useful tools for oil-source correlation with the accumulation of geochemical data of source rocks and oils.

In order to reveal the development mechanism of high-quality clastic rock reservoir, the basic characteristics of Sha-3 Member of the Shahejie Formation in the Raoyang sag, Bohai Bay Basin, are analyzed based on cores observation, thin-sections and SEM images, and petrophysical properties measurements as well. It is found that high-mature composition and texture, early oil charging, and dissolution are the main factors controlling the formation and preservation of pores in deep reservoirs. Compaction is the major factor destructing pores, whereas formation overpressure is conducive to the preservation of original pores, high compositional and medium textural maturity can enhance the resistance capacity to compaction and protect primary pores. Early oil charging could lead to temporary cessation of diagenesis and thus inhibit the cementation. When organic acids entered reservoir formations, considerable amounts of secondary pores were formed, leading to the local improvement of petrophysical properties. When predicting good quality belt in exploration of deep basin, it is recommended that the superimposing effects of the multiple factors (overpressure, early oil charging, compositional and textural maturity, diagenesis) be taken into consideration. Key words: Bohai Bay Basin, Raoyang sag, rifted basin, Shahejie Formation, diagenesis, high porosity zone, dissolution

Mercury Intrusion Capillary Pressure (MICP), Nuclear Magnetic Resonance (NMR), routine core analysis, thin sections, and scanning electron microscope (SEM) analysis were used to gain insight into the pore structure of the Eocene Sha-3 (the third member of Shahejie Formation) low permeability sandstones in Raoyang Sag, including pore type, pore geometry, pore size. Quantitative NMR parameters and petrophysical properties were integrated to build up the relationship between microscopic pore structure and macroscopic performance. The pore systems of Sha-3 sandstones are dominantly of residual intergranular pores, intragranular dissolution pores and intercrystallite micropores associated with authigenic clay minerals. The high threshold pressure and low mercury withdrawal efficiencies from MICP analysis indicate the poor pore connectivity and strong heterogeneous. Both uni- and bimodal T2 (transverse relaxation time) spectrum can be found because of the coexistence of small and large pores, and the T2 of majo...

The Niudong Buried Hill Field, which lies in the Baxian Depression of the Bohai Bay Basin, is the deepest oil/gas accumulation in eastern China. Its Precambrian dolomite reservoir occurs at burial depths of 5860 m–6027 m. This paper attempts to document the hydrocarbon charging and accumulation history in this field, which could greatly enhance the understanding of the mechanisms for the formation of deep hydrocarbon accumulations. Our previous study of oil trapped in fluid inclusions has demonstrated that the ratio parameters of the fluorescence spectral intensities at 425 nm and 433 nm (Q 425/433 ratio), and at 419 nm and 429 nm (Q 419/429 ratio) can be more effective for revealing hydrocarbon charging history than the previously-used fluorescence parameters such as Lambda max and red/green quotient as well as fluorescence colors. The hydrocarbon charging and accumulation history in the Niudong Buried Hill Field was studied with an integrated approach involving the application of these two spectral parameters of petroleum inclusion fluorescence as well as utilization of other data including homogenization temperatures of aqueous inclusions coeval with petroleum inclusions, and cross-cutting relationships of cements and “oil veins” in pores and fractures. The results indicate that the dolomite reservoir in the Niudong Buried Hill Field experienced three episodes of hydrocarbon charging. In the first two episodes (between 38.5Ma and 25Ma), the low mature and mature oils, which were derived from source rocks in the Sha-4 Member of the Eocene Shahejie Formation, migrated into the reservoir, but part of them leaked out due to normal faulting at the updip margin of the buried hill. These early-charged oils were preserved mainly in small pores in micritic dolomites by oil-wettability and capillary pressure. In the Neogene, the basin subsided as a whole and local faults at the updip margin became inactive and played a sealing role. By approximately 13Ma, the source rocks became highly mature and the generated hydrocarbons then migrated into the reservoir and accumulated. Therefore, the last charging is the most important for hydrocarbon accumulation in the Niudong Buried Hill Field.

With acquisition of geochemical data of oils and source rocks in mature basins, oil source re-identification is important for petroleum system analysis and subsequent exploration. In the Baxian Depression of the Bohai Bay Basin, the main source rocks are in the Sha-3 (Es3) and Sha-1 (Es1) Members of the Paleogene Shahejie Formation. These cannot be differentiated using the standard mass chromatograms and bivariate cross-plots of biomarker parameters obtained from the representative samples of the source rocks. In this paper, discriminant analysis (DA) was employed to highlight the difference between these two sets of the source rocks samples and to identify the sources of the oils in the multivariate space. Based on a comprehensive literature review, we selected twenty three candidate parameters that may reflect the oil sources, and excluded ones which are affected by the thermal maturity and biodegradation in the study area. Eight informative parameters were chosen with stepwise DA from these candidate parameters to establish a discriminant model for the oil-source correlation. The model achieved a high correct discrimination rate using leave-one-out cross-validation (90.7%) and original validation (94.4%). We found that these 8 parameters can describe the characteristics of the main sources and can effectively differentiate these sources. The high posterior probabilities in the oil-source correlation results indicate that the oils in the Wen'an Slope were mainly sourced from the source rock in Es3. These results, together with geological mapping of reservoir sand-bodies, suggest that more oil derived from the Es3 source rocks may exist below the thick shale and/or mudstone sequence in the Es1 than previously thought.

As one of the six subbasins in the lacustrine Bohai Bay basin, the Jizhong subbasin is characterized by a dominance of the petroleum reserves located in buried-hill traps of Paleozoic and Proterozoic marine carbonates, particularly the latter. This paper documents the revitalization of exploration of the buried-hill play and discusses other play fairways not previously considered through use of case studies. Discovery of the largest field, Renqiu field, in 1975 led to the establishment of the buried-hill play. In this play, oil derived from the Paleogene lacustrine source rocks charged into and accumulated in the underlying Proterozoic marine carbonate reservoirs. A number of buried-hill fields were discovered within a time span of ca. 10 yr in the subbasin. With more and more large buried-hill structures at depths of less than 5000 m (

Fractal geometry provides an effective method for characterization of the complex and irregular pore structure of Eocene Shahejie low permeability sandstones in the Raoyang Sag, the Bohai Bay Basin, China. Laboratory measurements including porosity, permeability, scanning electron microscope (SEM), thin sections, nuclear magnetic resonance measurements (NMR) and X-ray computed tomography (CT) technology are used to provide insights into the fractal characteristics of pore structure in sandstones of Eocene Shahejie Formation in the Bohai Bay Basin, China. Quantitative CT analysis reveals the pore radius is not always linear to T2 (transverse relaxation time) value obtained from the NMR tests, but instead power function of T2. Fractal analysis was performed on the T2 distribution using various fractal models, and the related fractal dimensions are calculated. The fractal dimensions calculated using various fractal models are correlated with NMR parameters and permeability. The fractal curves break into two segments at the T2cutoff (T2 separating the immovable and movable fluids) value or smaller when using fractal model Ⅰ and fractal model Ⅱ, and only the large-scale pore networks can be described by the fractal geometry. Mostly the entire pore size distributions (micro-pores to large-scale pore networks) can be described by the fractal model Ⅲ, and the calculated fractal dimensions are in accordance with the CT scanning and thin section data, and are strongly correlated with the T2gm (geometric mean of T2), permeability and BVI (bulk volume of immovable fluids). The fractal behaviors of pore size distributions from NMR analysis have implications for pore structure evaluation in low permeability sandstones with similar geological settings.

The successful use of near-surface anomalous signals, exhibited in all petroliferous basins, must lead to great development in petroleum exploration. The elementary equations and methods for objective and accurate separation of anomalous signals from background signals are the most critical requirements in these surface geochemical surveys. Therefore, we have worked up a series of fundamental equations and methods for eliminating surface interference and influences of geological conditions on anomaly intensities and recognizing multi-type anomalies of uni- and multi-variates, by using statistics, fractal geometry, wavelet analysis and artificial neural networks. These equations illustrate that the threshold (i.e. the boundary between background and micro-seepage anomalies) is determined by means, standard deviations and prior probabilities of background and micro-seepage anomalies when multi-normality is met, and that the traditional equation for the threshold (background mean plus one or two standard deviations) is not correct. In the face of many new equations and methods, it is important how to comprehensively use them to resolve the problems in geochemical surveys. A geochemical survey in the south slope of the Dongying depression, eastern China provides a good opportunity to address this issue. The anomalies obtained with the traditional equation do not reflect oil/gas fields. This study reveals three problems in the previous data-processing and anomaly recognition: (1) the surface interference caused by the variation of soil composition was not identified and removed, (2) the influence of caprock thickness on anomaly intensities was not revealed and eliminated, and (3) the micro-seepage (related to oil/gas pools) and seepage (related to faults) anomalies were not separated and correctly recognized. Furthermore, the interaction of these problems makes the data-processing and anomaly recognition more difficult. To ensure effective elimination of the surface interference and caprock thickness influence and correct recognition of uni-variate anomalies, we iteratively applied the wavelet-analysis-based methods to eliminate the surface interference and caprock thickness influence, and statistical methods with new fundamental equations to recognize uni-variate anomalies of seepage and microseepage. And then, multi-variate anomalies of seepage and microseepage were recognized with our fundamental equations and the corresponding methods. In the results, the seepage anomalies display a string-bead-shaped pattern and are distributed along faults, and the micro-seepage anomalies are ring-shaped and coincide with oil/gas pools, sandbodies or traps. Therefore, reprocessing of geochemical data with our fundamental equations and methods combined in an iterative manner can resolve the complicated problems in data-processing and anomaly recognition and thus greatly improve the predictive capability of surface geochemical survey.

All petroliferous basins exhibit near-surface anomalous signals. The successful use of these signals must lead to great development in petroleum exploration. However, oil/gas fields cannot be effectively predicted in the case of intensive influence of caprock thickness on intensities of anomalies in a geochemical survey for hydrocarbons. In this paper, we first study the features of the influence of caprock thickness and then establish digital conceptual models for the influence in one and two dimensions. Finally, we develop techniques for eliminating this influence with wavelet analysis, based on the digital conceptual models. The newly developed techniques are applied to the actual data of acid-extractable hydrocarbons of soils in the southern slope of the Dongying Depression, East China where the anomaly intensities are considerably influenced by caprock thickness. The results illustrate that this new approach enables us to satisfactorily eliminate the influence of caprock thickness on anomaly intensities and thus can greatly improve the predictive capability of the existing geochemical data.

The Ordos Basin is a large cratonic basin with an area of 250,000 km2 in central China. Upper Paleozoic coals and shales serve as gas source rocks with peak generation and migration at the end of the early Cretaceous. Recent exploration has verified the huge gas potential in the “basin-centered gas accumulation system” (BCGAS). However, the mechanism for the gas accumulation is controversial. With an integrated approach of thin-section petrography, ultra-violet fluorescence microscopy, fluid inclusion microthermometry, Raman microspectrometry, scanning electron microscopy, and X-ray diffractometry, we identified diagenetic trapping and evaluated the diagenetic history of sandstone reservoirs in the Yulin Gas Field in the central area, where structural, stratigraphic and/or sedimentary lithologic traps have not been found. It was revealed that three phases of diagenesis and hydrocarbon charging occurred, respectively, in the late Triassic, late Jurassic and at the end of the early Cretaceous. In the first two phases, acidic water entered the reservoir and caused dissolution and cementation, resulting in porosity increase. However, further subsidence and diagenesis, including compaction and cementation, markedly reduced the pore space. At the end of the early Cretaceous, the bulk of the gas migrated into the tight reservoirs, and the BCGAS trap was formed. In the updip portion of this system, cementation continued to occur due to low gas saturation and has provided effective seals to retain gas for a longer period of time than water block in the BCGAS. The mechanism for the gas entrapment was changed from water block by capillary pressure in the BCGAS to diagenetic sealing. The diagenetic seals in the updip portion of the sand body were formed after gas charging, which indicates that there is a large hydrocarbon exploration potential at the basin-centered area.

Simulations of the interaction between formation water, source rocks and reservoir rocks suggest that the carbonate content of the source rock not only influences the chemical reactions in the system, but also affects the overall porosity and permeability of the reservoir rock itself. A great deal of organic acid is produced during the interaction of formation water with the source rocks, but if the source rock contains high amounts of carbonate, that organic acid will first react with the carbonate. This initial and primary interaction results in the consumption of a great deal of the acid; the end results of this consumptive process area a), a significant decrease in the overall amount of organic acid in the source rock's fluid content and b), a corresponding (past-saturation) increase in the detectable quantities of Ca and Sr. When those same depleted fluids are expelled from the source rocks into the reservoir rock, the small and residual amounts of organic acid will have little or no impact in their new environment; the diluted acid content is simply not strong enough to promote the dissolution of aluminosilicate and carbonate. Moreover, Ca and Sr precipitate as carbonate because of the high content of Ca and Sr in the fluid; this also serves to decrease the porosity and permeability of the reservoir rock.

Guidance for petroleum prospecting is provided by authentic geochemical anomalies, obtained after interference has been eliminated in the original data from a geochemical survey for hydrocarbons. However, this elimination is a complicated process. Few working models or effective methods for interference elimination have been proposed, hampering a wider acceptance of geochemical prospecting as a viable tool for hydrocarbon exploration. This study focuses on the models for different types of interference and developing techniques for interference elimination. We first restudied the geometric nature of hydrocarbon anomalies and interference on the basis of Zhang, L., and Lia, Z. (J. China Univ. Geosci., v. 9, no. 1, p. 72–80, 1998). From this, we determined that at least two types of interference are present in the data of a surface geochemical survey for hydrocarbons. One is the superimposition type originating mainly from microorganisms and contamination, and the other is the multiplication type resulting from variations in soil mineral components. We then established digital conceptual models for both types of interference based on the geometry of anomalies and interference. Finally, we used wavelet analysis to eliminate both types of interference in the models for testing. As Morlet wavelet and symmetric border treatment fit the geometric characteristics of the anomalies and interference, they were used in this study. We found that pre-wavelet analysis treatment is different for different types of interference. For the multiplication type, log-transformation and normalization are needed before wavelet analysis whereas only normalization is needed for the superimposition type. In practice, one needs to identify the type of interference. Therefore, the criteria for determining the interference type were first established geochemically in this study. The newly developed techniques were then applied to the data of acid-extractable hydrocarbons of soils in the Eastern Anan Sag, Inner Mongolia where traditional techniques had failed to eliminate the interference in these data. Using our new techniques, we determined the interference was the multiplication type. This new approach enabled us to satisfactorily eliminate the interference in the original data. It should be noted that our study indicates that the multiplication type of interference is much more common and severe than previously thought.

Traditional methods of multivariate statistical analyses, used to recognize anomalous hydrocarbon signatures in petroleum exploration, have at least four shortcomings: (1) it is difficult to isolate anomalies where the data are not normally distributed; (2) it is difficult to separate distinct anomaly populations corresponding to distinct formation mechanisms while separating anomalies from background; (3) it is not fitting to present illustrations of multivariate anomalies on contour maps; and (4) it is not suitable for preparation for multivariate pattern recognition. These present serious obstacles to the application of exploration geochemistry in hydrocarbon exploration. This study demonstrates that the Back Propagation Artificial Neural Network (BP-ANN) with logic multiplication cluster analysis (a new cluster analysis proposed in this paper) overcomes the difficulties exhibited by traditional multivariate methods. The logic multiplication cluster analysis was designed to produce a training set for the BP-ANN. This approach was established on the basis of geochemical characteristics and origin of the various populations in geochemical data, such as background, micro-seepage anomalies and seepage anomalies. The topology of the BP-ANN was optimized using the outputs of the BP-ANN and the correct rate. In order to illustrate the multivariate anomalies recognized using BP-ANN on contour maps, we designed the expression functions for BP-ANN application in this field. With traditional methods of anomaly recognition, acid-extractable hydrocarbons in soils have not proven to be efficient indicators for hydrocarbon potential in East Anan Sag, Inner Mongolia. However, the BP-ANN has indicated that these indicators are efficient and that areas of East Anan Sag have potential reserves of petroleum.