Multi-Scale Geophysics and Chemistry-Based Investigation of Alteration Evolution Mechanisms in Buried Hills of the Northern South China Sea

Abstract

Alteration is a common metamorphic process in igneous formations and recorded geological information in different times and spaces. Owing to its unique location, the igneous rocks of the buried hills in the northern South China Sea exhibit complex lithology and alteration patterns resulting from multi-phase tectonic, magmatic, and climatic influences. Here, we report buried hills igneous rock samples with both hydrothermal alteration and weathering leaching. Based on multi-scale geophysical–chemical data—including scanning electron microscopy, core slice identification, petrophysical–chemical experiments, zircon dating, wireline logs, element cutting logs, seismic profiles, and others—we analyzed the multi-scale alteration characteristics of buried hills igneous rocks and proposed a four-stage alteration model related to Earth activities. Results demonstrate that tectonic movements develop continuous cracks enabling hydrothermal alteration, while burial-hill uplift facilitates weathering leaching. We further find that multi-phase tectonic movements and associated magmatic activities not only influence global hydrothermal cycles but also govern elemental migration patterns, driving distinct alteration mechanisms in these igneous rocks—including plagioclase metasomatism, hornblende replacement, and carbonate dissolution. Additionally, we identify the Cretaceous arid–cold climate as the primary controller for generating chlorite-dominated hydrothermal alteration products. These multi-scale alteration characteristics confirm Late Jurassic Pacific Plate subduction and Cretaceous South China Plate orogeny and may indicate an earlier initial expansion of the South China Sea.

1. Introduction

Alteration refers to the chemical or physical changes in the composition, minerals, structure, or texture of rocks due to external factors [1]. Scientists consider alteration crucial for deciphering the Earth’s geological history, as altered minerals and textures record changes in temperature, pressure, fluid interactions, or other geological processes [2,3]. Analyzing chemical compositions of altered rocks further deepens our understanding of elemental redistribution during alteration, and this analysis lays a theoretical groundwork for interpreting diagenetic processes [4,5].

Alteration not only unlocks geological information but also guides natural resource exploration [6]. It significantly affects the porosity and permeability of rocks [7,8,9], potentially transforming them into viable hydrocarbon reservoirs. Alteration signatures often indicate economically valuable minerals. Hydrothermal alteration particularly forms gold, silver, and copper deposits [10]. Additionally, studying alteration processes offers insights into the climate system, as these processes record paleoclimatic shifts over geological timescales [8,11,12,13]. Thus, alteration models transcend mere petrographic description—they are pivotal keys to unlocking planetary history.

This paper studies alteration characteristics in the Pearl River Mouth Basin and explores the interplay between alterations and geological processes. While current research predominantly focuses Mesozoic magmatic and tectonic evolution on southern China’s inland and southeastern coast [14,15,16,17], researchers have also discovered an Andean-type continental arc from the Middle Jurassic–Early Cretaceous in the northern South China Sea’s Qiongdongnan Basin [18]. Note that in-depth studies targeting the northern South China Sea margin remain scarce. Meanwhile, the division of the South China Sea continent and igneous oceanic crust transformation represent current research hotspots [19]. Therefore, as a continental oceanic transition zone characterized by Mesozoic magmatic rock, the study of buried hills in the Pearl River Mouth Basin in the northern South China Sea fills this gap and provides evidence for continent–ocean transition evolution.

Tectonic activities have subjected Pre-Paleozoic igneous strata (classified as buried hills) to metamorphism and deformation [20,21,22,23,24], giving these strata a complex geological history. Intense Cretaceous magmatic activities and alterations further amplified the complexity, particularly within the Huizhou Depression [16,25,26]. To address this challenge, we employ a comprehensive multi-scale analytical framework integrating geophysics and geochemistry. Our approach combines data from different scales, including core thin section identification, petrophysics, mineral measurements, zircon geochronology, wireline logging, element cutting logging, and seismic profiles of pre-Paleozoic buried hills in the Huizhou Depression. These analyses aim to unravel intricate buried hills alteration patterns. Furthermore, we investigate how tectonic movements, magmatic activities, and regional paleoclimates affect igneous rock alteration and associated reservoirs. Then, we proposed a four stage alteration model that combines tectonic activity, regional climate, and alteration characteristics, and proposed the relationship between alteration characteristics and Earth activity. This work provides pivotal data and insights to enhance understanding of Mesozoic marine residual basin evolution, South China Sea continent–ocean transition processes, and ancient environmental reconstruction.

2. Geology and Methods

2.1. Geological Background

Straddling the boundary between the Pacific and Eurasian Plates [27,28,29], the Pearl River Mouth Basin stretches NE-SW across the continental shelf of the northern South China Sea, which is is located in the South China Sea. Scientists recognize it not only as the largest Mesozoic–Cenozoic sedimentary basin in the northern South China Sea but also as a significant part of the eastern margin of the South China Block [30,31]. The eastward extension of the Mesozoic fold basement from the eastern margin of the South China Block during the Indosinian period forms the main basement of the Pearl River Mouth Basin. A series of NW-oriented (inland) and NE-oriented (coastal) fault systems, prevalent within the Pre-Jurassic strata and formed during the Yanshanian orogeny (60–200 Ma), highlight the exceptionally complex geological history of the Pearl River Mouth Basin [32,33,34,35].

The Pearl River Mouth Basin can be divided from north to south into five NE-oriented primary structural units: the northern uplift belt, northern sag belt, central uplift belt, southern sag belt, and southern uplift belt, forming a unique geological setting termed the “two sags and three uplifts” [36]. The basin has undergone at least five Mesozoic magmatic tectonic movements, afterwards [24,37,38,39]. Due to its location at the border between the mainland and the northern edge of the South China Sea, it not only underwent continent–ocean transition, but may also be affected by the expansion of the South China Sea [18,40]. And these episodes of tectonic activity have resulted in the distinctive and complex geological setting of the Pearl River Mouth Basin.

Earlier studies reveal that the Pearl River Mouth Basin has significant oil and gas potential [38,41,42,43]. The study area, the Huizhou Depression, lies in the central part of the Zhu I Sag, Pearl River Mouth Basin (Figure 1a), and represents one of the most hydrocarbon-rich depressions in the basin’s eastern part [28,44]. The Huizhou Depression features a diverse stratigraphic sequence (Figure 1b), with Pre-Cenozoic rocks acting as the basement overlain by the Wenchang, Enping, Zhuhai, Neogene Pearl River, Hanjiang, Yuehai, and Wanshan Formations, capped by Quaternary deposits [45]. The strata above burried-hill are composed of sandstone, mudstone, and conglomerate of different origins, and the main objective of this paper is to study the burried-hill (weathering crust and basement). It is widely known that the main hydrocarbon source rocks in the Huizhou Depression are primarily identified within Wenchang and Enping Formations [46]. Recent findings suggest that Pre-Cenozoic buried hills also have potential for oil and gas, this indicates that this study also has extremely high economic value.

The buried hills are primarily composed of various types of igneous rocks, to unravel the geological history of the Huizhou Depression, 149 igneous rock samples were collected from seven wells, as illustrated in Figure 1c. These wells are distributed across two key areas X-1 and X-2 (Figure 1d). Within the X-1 well area (Figure 1e), the samples comprise 39 cores from Well X-1-1, 18 cores from Well X-1-2, 10 cores from Well X-1-3, and 27 cores from Well X-1-4, summing up to 94 cores in total. Meanwhile, in the X-2 well area, the collection includes 18 cores from Well X-2-1, 2 cores from Well X-2-2, and 35 cores from Well X-2-3, totaling 55 cores. In addition to core samples, there are 18 rock debris samples collected: 8 from Well X-1-1, 2 from Well X-1-2, 1 from Well X-1-3, 6 from Well X-2-2, and 1 from Well X-2-3.

2.2. Methods and Experiment

2.2.1. Multi-Scale Data Analysis Framework

To achieve a comprehensive understanding of the geological processes in the Huizhou Depression, this study proposes a structured workflow framework including data collection, analysis, and interpretation of results, as illustrated in Figure 2.

Data collection involves conducting a broad spectrum of geophysical, mineralogy, and geochemical analyses on rock core and rock debris samples obtained from various depths within the buried hills. The analysis framework spans micro to nano-scale analyses, such as scanning electron microscopy (SEM) and zircon uranium-lead dating; centimeter-scale assessments, including core thin section analysis, porosity and permeability measurements, X-ray diffraction (XRD) for mineral identification, and X-ray fluorescence (XRF) for elemental composition analysis; meter-scale investigations, such as wireline logging and elemental cutting log, and upscaling-macro scale data such as seismic profiling. The samples used in the research are from the the Pearl River Mouth Basin in China. Except for the zircon dating instrument from Tongji University in Shanghai, China, all other equipment comes from Jilin University in Changchun, China.

To enhance the analytical techniques mentioned above, this study integrates advanced data analysis methods. We selected the K-nearest neighbor (KNN) algorithm for lithology identification due to its proven robustness in classifying complex igneous formations [47,48], achieving 95.6% test accuracy. Chemical Isoconcentration Line Analysis (ISOCON) is employed to study the elemental migration between rock samples and hydrothermal fluids [49]. This comprehensive methodological approach is designed to provide a detailed understanding of the complex geological history within the study area.

After determining the lithological information of the sample based on rock thin sections, typical rock cores of different lithologies were selected to carry out the above-mentioned multi-scale data collection work. These analyses aim to elucidate geological ages, physical properties, rock textures, mineralogy, and chemical compositions using multi-scale data. The combination of petrophysical and mineralogical data can be used to analyze the relationship between mineral changes and changes in rock porosity and permeability caused by different alteration. Mineralogy and elemental analysis can provide reliable evidence for the migration patterns of elements during alteration processes. The combination of laboratory core data and well data can restore the lithological distribution characteristics in the formation. By incorporating rock age information and up-scaling seismic data, this article can utilize multi-scale data to study the relationship between alteration characteristics and Earth activity.

2.2.2. Petrophysics Experiment

(1) Scanning electron microscope

Scanning Electron Microscopy (SEM) bridges the analytical capabilities of transmission electron microscopy and optical microscopy. Its straightforward sample preparation permits minimal processing of bulk or powdered specimens or direct imaging, allowing researchers to observe micro-morphology approximating the material’s natural state.

We utilized a ZEISS thermal field emission scanning electron microscope in this study. The instrument collects, amplifies, and reconstructs various physical signals of specimen into images to characterize rock micro-morphology. Before sample testing, it is necessary to perform electron beam calibration, detector calibration, stage and magnification calibration based on standard samples. In actual testing, it is necessary to evaluate the test results and obtain accurate results through more than 5 repeated experiments.

(2) Core thin section identification

Thin section analysis plays an essential role in characterizing reservoir properties. This technique involves mounting rock core slices approximately 0.03 mm thick (allowing visible light transmission) between glass slides for examination under an optical microscope. This process enables observation of pore structures and interstitial materials within the thin section.

This paper uses Olympus BX51 polarizing microscopes to identify polarizing films, which are based on the principles of Optics. After partial staining treatment of the rock slice, we select appropriate magnification under a polarizing microscope. By leveraging the distinct optical properties (particularly birefringence) of different minerals in the thin section, we identify mineral types while simultaneously examining grain size, sorting, and roundness, pore shape and size, distribution relationships of authigenic minerals within pores. Based on thin section images, obtain rock lithology information through multiple rounds of verification.

(3) Porosity and permeability

Core porosity and permeability are critical parameters for reservoir evaluation. Consequently, their measurement constitutes a fundamental experiment for analyzing petrophysical properties.

The principle of porosity measurement is based on Boyle’s Law. During testing, gas undergoes isothermal expansion from a reference chamber of known volume at a set initial pressure into a core chamber at atmospheric pressure. As the gas diffuses into the rock pore space, the resulting pressure change, combined with the known volumes, allows calculation of the effective pore volume and grain volume of the core sample using the gas law equation. Porosity is then derived from these volumes.

Permeability determination employs the transient pressure pulse decay technique, a non-steady-state method. Pore pressure is applied to the core, followed by the transmission of a differential pressure pulse through it. As the pressure transient propagates, a computer data acquisition system records the pressure differential across the core, downstream pressure, and time. The software plots logarithmic curves of differential pressure and mean pressure versus time. Permeability is calculated via linear regression analysis of the pressure-time data, with results stored in data files. Using minimal pressure differentials minimizes non-Darcy flow effects. Conducting multiple measurements at varying pore pressures yields the Klinkenberg-corrected permeability.

This paper uses the AP-608 Automated Porosimeter-Permeameter for measurements. After calibrating the instrument with standard samples, we take subjected to >10 repeated measurements for each sample and calculate the absolute deviation of porosity and relative deviation of permeability separately. When deviations meet experimental requirements, we take the value with the smallest deviation as the final result. Porosity absolute deviation must not exceed 0.5 p.u., while permeability relative deviation for samples <1 mD must not exceed 15%.

2.2.3. Rock Mineralogical Analysis

X-ray diffraction (XRD) analysis technology is an important means of analyzing the mineral composition. It can analyze different types of crystalline (including quasicrystalline) substances and then qualitatively or quantitatively determine the types and contents of each phase. Mineral analysis through XD-6 X-ray diffractometer (XRD) have revealed significant differences between altered and unaltered igneous rocks.

When X-rays pass through a crystal, each crystalline substance has unique diffraction characteristics, which can be characterized by the interplanar spacing and relative intensity. The interplanar spacing is related to the shape and size of the unit cell, while the relative strength is related to the type of crystal particles and their positions in the unit cell. The diffraction data mesh spacing and relative intensity of any crystalline substance are inevitable reflections of its crystal structure, which will exhibit different diffraction angles and peak heights on the diffraction pattern, thereby identifying the composition of various crystalline substances. Before conducting the experiment, it is necessary to crush about 10 g of core into a powder sample of about 800 mesh, and then fill the powder sample into the measurement mold for measurement in an X-ray diffraction whole rock mineral analysis instrument. The instrument needs to undergo an empty sample stage scan before operation to complete zero background calibration. To avoid the influence of heterogeneity, the sample powder needs to be completely mixed and divided into five measurements. When the relative standard deviation of the test results is not higher than 1.5%, taking the result with the smallest deviation as the experimental result.

2.2.4. Elemental Analysis

X-ray fluorescence (XRF) core element analysis is a chemical analysis technique for core samples, with sample preparation requirements identical to those for XRD whole-rock mineral analysis. Unlike XRD, XRF utilizes the wave-particle duality of X-rays to measure the relative contents of major and trace elements in rocks.

Major element analysis was carried out using the EDX6000B X-ray fluorescence (XRF) spectrometer. During the operation of the instrument, when X-rays irradiate the core powder sample, characteristic X-rays of various wavelengths (X-ray fluorescence) can be excited in the sample. Then, the mixed X-rays are separated by energy (or wavelength), and the intensity of X-rays of different energies (or wavelengths) is measured separately to analyze the content of different elements in the spectrum. Prior to sample analysis, the instrument must undergo energy scale calibration and intensity response calibration using certified reference materials. Subsequently, a homogenized sample is subjected to >10 repeated measurements. The data point exhibiting the smallest deviation is adopted as the final experimental result when all elemental measurements satisfy predefined tolerance criteria.

2.2.5. Zircon Uranium Lead Dating Detection

Zircon U-Pb dating determines rock ages, provides chronological constraints for regional tectonic activities, effectively traces source area characteristics and backgrounds, and elucidates intrinsic relationships between basin subsidence and major thermo-tectonic events.

The fundamental principle of isotopic dating relies on the radioactive decay law. By measuring the concentrations of parent isotopes and their daughter isotopes produced through decay, formation ages are calculated using this decay law. Specifically, zircon U-Pb dating primarily targets radioactive elements U and Th (which decay to Pb isotopes) within zircon. Utilizing the abundant age information preserved in zircon crystals, this method identifies provenances of detrital materials from different periods within rock fragments.

Zircon U-Pb isotope dating was completed using Laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS) analysis. The laser ablation system is New Wave 213 nm, and the ICP-MS model is Thermo Elemental X-Series. For each sample analysis, data acquisition comprises approximately 25 seconds of blank signal measurement followed by 50 seconds of sample signal measurement. During U-Pb isotopic dating, zircon reference material 91500 (1065.4 ± 0.3 Ma) serves as an external standard for instrumental mass fractionation correction, with two analyses of 91500 performed following every six unknown sample analyses. Concurrently, zircon standard Plešovice (337.1 ± 0.4 Ma) is analyzed to monitor analytical precision.

3. Results

3.1. Experimental Results

3.1.1. Petrography

(1) Lithological identification

The cross-polarized light thin sections of 149 cores from X-1 and X-2 well areas in the Huizhou Depression, Pearl River Mouth Basin, northern part of the South China Sea, indicate a diverse assembly of igneous rocks, ranging from mafic to felsic rock types (Figure 3). Signs of alteration are observed in approximately half of the samples. Notably, granite exhibits both hydrothermal and weathering alterations. This comprehensive examination has led to the categorization of the core samples into 12 distinct types: 10 samples of diorite, 32 tectonic schists, 2 diabase, 10 altered diabase, 4 basalt, 2 altered basalt, 31 andesite, 21 altered andesite, 14 granite, 5 altered granite, 6 weathered granite, and 12 granodiorite.

Diorite displays porphyroblastic and blastoporphyritic textures; its matrix contains fine-grained, columnar hornblende and exhibits schistose structure. Minor plagioclase and quartz occur within it (Figure 3a). Tectonic schist shows lepidoblastic texture and gneissose structure, dominated by plagioclase and quartz. K-feldspar appears sparingly, its surfaces displaying significant alteration (Figure 3b). Diabase exhibits crystalloblastic texture with micro-bedding structures, containing plagioclase, quartz, and pyroxene as primary constituents (Figure 3c). Altered diabase develops unequigranular texture and inhomogeneous structure, holding plagioclase, pyroxene, and chlorite with accessory pyrite (Figure 3d). Basalt possesses vitrophyric texture; plagioclase dominates its matrix (Figure 3e). Altered basalt maintains porphyritic texture with plagioclase-rich matrix. Minor pyroxene occurs throughout. Almond-shaped vesicles commonly contain chlorite fillings (Figure 3f). Andesite features porphyritic and intergranular textures. Its matrix comprises microcline and microcrystalline pyroxene. Plagioclase demonstrates distinct orientation, with interstitial pyroxene grains separating crystals (Figure 3g). Altered andesite presents irregular morphology, bearing pyroxene phenocrysts. Its matrix displays heterogeneity—alternating between typical andesitic and devitrified textures. Plagioclase and quartz predominate; chlorite and illite contribute significantly (Figure 3h). Granite has coarse texture and massive structure, consisting mainly of microcline, perthite, plagioclase, and quartz (Figure 3i). Altered granite retains coarse texture and massive structure. Noticeably altered feldspar and quartz characterize it alongside recrystallized matrix (Figure 3j). Weathered granite contains tectonic breccias and porphyroclasts, composed dominantly of microcline, plagioclase, and quartz, indicating deformation history (Figure 3k). Granodiorite possesses fine-grained texture, containing predominantly K-feldspar with plagioclase and quartz. Dark minerals occasionally intersperse (Figure 3l).

(2) Porosity and permeability

The porosity values for these rocks range from 1.5% to 14.8%, with an average of 5.52% (

Table 1. Core porosity and permeability in the research area.

Lithology	Porosity	Permeability
	%	mD
Diorite	1.7	0.002
	2.7	0.006
Tectonic schist	3.8	1.990
	4.3	0.029
Diabase	1.5	0.005
Altered diabase	10.4	8.850
	4.2	0.005
Basalt	6.9	0.004
	7	0.005
Altered basalt	14.8	0.002
Andesite	5.9	0.011
Altered andesite	6.7	0.007
	4.9	0.030
Granite	2.4	0.002
Altered granite	5.6	0.453
Weathered granite	4.4	0.014
Granodiorite	6.7	0.003

). The porosity values of most rocks fall below 10%, categorizing them within the low porosity range. Permeability values span from 0.002 millidarcy (mD) to 8.850 mD, with the average permeability standing at 0.672 mD (Table 1). Permeability values are predominantly lower than 1 mD, leading to an ultra-low permeability category.

(3) Mineral components

Altered rocks contain significantly more clay minerals than their unaltered counterparts. These clay minerals replace other mineral phases, demonstrating extensive alteration processes throughout the study area. Unaltered diorite consists primarily of hornblende (31.87%) and plagioclase (31.70%), with smaller quantities of quartz (12.73%), zeolite (3.23%), and minor clay minerals (4.67%). Tectonic schist (diorite’s altered equivalent) shows a sharp decrease in hornblende (15.57%) and plagioclase (22.97%), while hosting substantially more clay minerals (20.57%). This study consistently documents reduced primary minerals alongside increased clay content across multiple rock types, highlighting the transformative impact of alteration processes (Figure 4).

(4) Elemental composition

A significant loss of core samples during earlier analyses and the challenges associated with resampling prevented the measurement of two types of cores, diabase and altered basalt. Furthermore, since granodiorite does not have an altered counterpart, no major element measurements were performed on it. The mean values of 10 standardized major element oxides (${\mathrm{Na}}_2\mathrm{O}$, $\mathrm{MgO}$, ${\mathrm{Al}}_2{\mathrm{O}}_3$, $\mathrm{Si}{\mathrm{O}}_2$, ${\mathrm{P}}_2{\mathrm{O}}_5$, ${\mathrm{K}}_2\mathrm{O}$, $\mathrm{CaO}$, $\mathrm{Ti}{\mathrm{O}}_2$, $\mathrm{MnO}$, and ${\mathrm{Fe}}_2{\mathrm{O}}_3$) for nine different core types are reported in

Table 2. Mean major element compositions of different rock types sampled in Huizhou Depression.

Lithology	Na2O	MgO	Al2O3	SiO2	P2O5	K2O	CaO	TiO2	MnO	Fe2O3
	wt%	wt%	wt%	wt%	wt%	wt%	wt%	wt%	wt%	wt%
Diorite	2.19	5.45	15.46	56.90	0.36	2.27	6.95	0.81	0.21	9.41
Tectonic schist	2.62	6.19	13.08	60.59	0.30	2.12	5.98	0.66	0.18	8.25
Altered diabase	2.57	5.29	12.78	68.39	0.09	2.47	2.84	0.50	0.10	4.97
Basalt	3.19	5.38	16.95	52.04	0.63	1.50	8.52	1.68	0.23	9.87
Andesite	1.82	2.33	19.30	59.53	0.20	2.42	6.84	0.71	0.12	6.73
Altered andesite	5.46	2.47	19.74	57.77	0.18	2.80	4.46	0.80	0.08	6.23
Granite	1.91	0.39	10.30	80.27	0.04	4.75	0.89	0.13	0.03	1.30
Altered granite	1.69	8.29	13.49	61.24	0.14	1.69	5.77	0.79	0.15	6.75
Weathered granite	2.72	0.24	9.62	82.20	0.02	3.05	0.50	0.12	0.02	1.50

.

3.1.2. Geochronology

Zircon dating of 20 rock debris offered critical insights into geological ages of rocks in the Huizhou Depression, which were selected samples based on lithology identification. After testing multiple age measurement points for each zircon sample, data points with larger errors are removed, and then grouped according to age to obtain age ranges and weighted average ages. The final determination of the geological age of different samples is shown in

Table 3. Zircon age determination results in the study area.

Well Name	Sample Number	Top Depth	Bottom Depth	Age Range	206Pb/238Uaverage Age
		m	m	Ma	Ma
X-1-1	1	3797	3807	108–126	116
				137–167	153
	2	3897	3907	98–124	110.8
				146–164	154.3
	3	3947	3957	126–2374	105.3
	4	3997	4007	101–141	118
				251–263	254
	5	4097	4107	151–174	162.7
	6	4147	4157	147–189	163.8
	7	4237	4247	126–181	155.4
	8	4287	4297	112–157	133.8
X-1-2	1	4120	4130	/	110
	2	4230	4240	/	144
X-1-4	1	4195	4210	/	116
X-2-1	1	4290	4300	57–73	61.6
				106–141	119.3
	2	4540	4550	106–135	116.5
X-2-2	1	3345	3354	100–120	112.4
	2	3369	3378	100–120	111.4
	3	3390	3399	100–140	114.7
				248	248
	4	3399	3408	50–70	58.9
				100–130	113.4
	5	3411	3420	90–130	111.3
	6	3450	3459	100–130	114.2
X-2-3	1	4860	4880	46–70	55
				151–160	155.3
				403–444	419
				893–895	894

.

For Well X-1-1, debris sample ages predominantly span 105–254 Ma, aligning with Cretaceous and Jurassic periods. The 3947–3957 m depth interval reveals notable age variation, indicating complex material sources and geological activities. Wells X-1-2 and X-1-4 yield geological ages of 110–144 Ma and ~116 Ma, respectively, primarily within Cretaceous–Jurassic strata and featuring a pronounced Yanshanian peak. Well X-2-1 encompasses 57–141 Ma (Paleogene–Cretaceous–Jurassic), while Well X-2-2 registers 50–140 Ma (mainly Cretaceous). Well X-2-3 exhibits the broadest age range (46–2715 Ma). This substantial variation likely reflects source effects, where 2599 Ma zircon grains likely ascended via deep-crustal magma upwelling, thereby recording deep crustal formation ages.

3.2. Drilling Data Processing

3.2.1. Lithology Identification in the Well

The core samples available were insufficient to fully reconstruct the lithology profiles within the wells. And the integration of logging data with the K-nearest neighbor (KNN) algorithm has been proven to effectively address complex lithological classification problems [50,51]. Using the rock lithology identification results verified by multiple geologists as labels, a total of 412 lithology data points were collected in this article, not solely from the wells of immediate interest but also from additional wells within the same study area, at various depths. The input wireline logging data for the KNN include nature gamma ray, compensated neutron, compensated density, acoustic time difference, and resistivity, along with magnesium and aluminum element contents from elemental cutting logging data. The lithology of each data point is determined by microscopic identification of core samples or through analysis of debris from wireline logging. The whole dataset is split into a training set and a test set, allocating 80% to training and 20% to testing. After comparing and analyzing different algorithms such as the Support vector machine and eXtreme Gradient Boosting, KNN is currently the most effective method for lithology recognition in this environment. The K value of 4 returns the highest accuracy, achieving 92.2% for the training set and 95.6% for the test set. These high accuracy rates underscore the reliability of KNN for lithology identification in this study. The KNN model was subsequently applied to the wells of interest, with logging data collected at a 0.125 m interval across all wells.

The lithology results based on KNN show that Wells X-1-1, X-1-2, and X-1-3, located in the southwest of the Huizhou Depression, exhibit a similar lithology, primarily consisting of altered diorite, diorite and altered granite, with interbedded thin layers of intruding altered diabase (Figure 5). The alteration within these wells is notably pronounced at depths ranging from 3800 to 4375 m. Well X-1-4, with its depth range of 4150 to 4350 m, mainly contains granodiorite, granite, and altered diabase with a lesser presence of altered and weathered granite. The X-2 series wells, situated in the northeast of the Huizhou Depression, span from 3300 to 5200 m depth, and have granite, andesite, altered andesite and basalt. Among these, Well X-2-1 mainly features altered andesite, interbedded with thin layers of basalt, and underlying granite. Well X-2-3 is dominated by andesite, and Well X-2-2 contains a variety of igneous rocks in interlayered forms.

3.2.2. Formation MicroScanner Images

In this study, Formation MicroScanner images (FMI), which are detailed electrical images of rocks, were collected to reveal the fracture characteristics of the study area. FMI results show that fractures are predominantly developed in the X-1 well area. Specifically, diorite shows continuous fractures, partly open fractures, and dissolution fractures, with fracture aperture gradually increasing with depth. Tectonic schist primarily exhibits partly open fractures and continuous fractures, alongside dissolution fractures, with larger fracture aperture and porosity near the surface, which tend to decrease with increasing depth. Altered diabase is also marked by partly open fractures and continuous fractures. Granite shows development of micro-fractures. In Well X-1-1, altered granite displays massive and layered structures, featuring continuous fractures and partly open fractures with significant porosity. Conversely, altered granite in Well X-1-2 is primarily characterized by massive structures, exhibiting a high degree of fracturing and secondary dissolution. Weathered granite shows the most extensive development of secondary dissolution. Granodiorite has massive structures, displaying discontinuous fractures and partially filled fractures.

The X-2 well area shows a prevalence of natural fractures. Dissolution fractures and continuous fractures are observed in andesite. Altered andesite and basalt mainly shows massive structure with natural fractures while granite, primarily massive structure, exhibits fewer fractures.

4. Discussion

4.1. Alteration Characteristics of Buried Hills

4.1.1. Core Alteration Characteristics

The igneous rocks of the buried hills have undergone a complex history, likely influenced by magmatic and tectonic activities, as well as regional climate changes. Many researchers hold the view that hydrothermal alteration can reduce the porosity and permeability of the rocks [52,53], whereas weathering can enhance reservoir capability. Therefore, weathering crusts in buried hills have significant research interest [36,54,55]. Surprisingly, recent studies have also revealed the reservoir potential in the altered rocks of buried hills [56].

This study primarily observes hydrothermal alteration, with some instances of weathering. Altered rock types include diorite, diabase, basalt, andesite, and granite, with granite undergoing both hydrothermal alteration and weathering while hydrothermal alteration and metamorphism for diorite. The extent and nature of alteration vary significantly due to the composition of the original rocks and changes in external environment [1]. The degree of alteration is commonly assessed by the content of clay minerals in the altered rocks [57]. In this study, the clay minerals identified in altered core samples are predominantly chlorite and illite, with their content generally ranging between 10 and 30%, indicating moderate alteration.

The SEM results indicate that the clay in rocks experiencing hydrothermal alteration are predominantly sheet-like or petal-shaped chlorite and fibrous illite, the mineral combination of hydrothermal alteration in diorite is hornblende–plagioclase–mica (Figure 6a), in andesite is plagioclase (Figure 6b), in granite is K-feldspar alteration (Figure 6c), and granite weathering is accompanied by dissolution of K-feldspar and carbonate minerals (Figure 6d). This is consistent with the changing characteristics shown in mineralogical data. It is worth noting that SEM images and core mineral analysis did not show significant presence of kaolinite. Studies show that the temperature conditions favoring the formation of chlorite are lower than those required for kaolinite [58]. Additionally, chlorite and illite serve as indicators of a dry and cool climate [59,60]. This alteration feature, which is mainly composed of chlorite and illite, supports the study area being in a dry and cool climate during the alteration process, and a detail future basin models should integrate to predict clay-dominated alteration.

The alteration characteristics were further analyzed using an Isocon plot, which offers insights into elemental changes due to alteration. The results show an increase in potassium (K) content in altered diorite and andesite compared to their unaltered counterparts, as illustrated in Figure 6a,b. This increase is consistent with the augmentation of K-feldspar observed in the XRD results (Figure 4). Conversely, a decrease in K content is observed in both the altered and weathered granite (Figure 6c,d). In other words, the changes in geochemical composition of these igneous rocks may not solely be attributed to the interaction between fluids and rocks during hydrothermal alteration. They could also be the result of partial melting and crystal fractionation associated with magmatic activity in the northern South China Sea, as well as the heterogeneity of the mantle source [19].

4.1.2. Alteration Characteristics in Well

Both X-1 and X-2 well areas expose diverse igneous rocks spanning mafic to felsic compositions. The X-1 well area primarily contains diorite (Figure 7a), while the X-2 well area hosts andesite (Figure 7b). Weathered granite occurs in Well X-1-4; compared to altered granite in adjacent wells, FMIs reveal more dissolution fractures in this unit. Altered rocks in these wells develop more extensive fracture networks than their unaltered counterparts. Unaltered basalt and granite particularly lack fractures or show minimal fracturing. This fracture–alteration correlation underscores tectonic activity’s pivotal role in driving alteration processes. Faulting generates fractures, increasing rock susceptibility to alteration by enabling deeper hydrothermal fluid penetration. The similar alteration patterns observed in diorite and andesite and the similar depths they are situated at may suggest that these two rock types could have experienced the same geological event during the Jurassic–Cretaceous. The fracture openings and fracture porosity in the tectonic schist are relatively large, aligning with the observed increase in permeability. The porosity of the altered andesite changes slightly compared to unaltered andesite, but the permeability of the altered andesite notably increases, likely due to the fractures. Additionally, multiple fracture types are overserved in granite and diorite. Given that continuous fractures are of tectonic origin, it suggests that the same lithology in the study area may have undergone various episodes of tectonic events.

From a conventional perspective, the phenomenon of improved pore conditions in rocks after dissolution can be explained by the increase in dissolution cracks caused by hydrothermal alteration and weathering, especially secondary dissolution cracks caused by weathering. This is also consistent with the experimental results of porosity and permeability in this paper, which has proven the better petrophysical properties of altered igneous rocks (Table 1). Paradoxically, studies have shown that secondary kaolinite does not have such a significant improvement effect on rock pore structure [29], but is more prone to water absorption, expansion, and pore blockage. Although the igneous rocks in the research area did not show significant kaolinite mineralization, the content of chlorite and illite in the altered rocks cannot be ignored. This indicates that during the process of alteration, the infiltration channels induced by tectonic movements, the dissolution of primary minerals, and the precipitation of secondary clay jointly affect the development of geological pore structure. The difference in physical properties between altered and unaltered igneous rocks is a comprehensive response of the pore structure of the original rock before alteration and the pore structure after alteration. The development of cracks is also an important condition for the occurrence of alteration. After the development of fractures in igneous rocks, they are more likely to come into contact with the formation hydrothermal fluid and undergo alteration. The alteration process further transforms the pore space and seepage channels, jointly creating the complex pore structure of altered igneous rock formations. Based on the current results, the hydrothermal alteration igneous rock formations in this study also have storage space and energy potential, which is different from previous understanding.

4.1.3. Alteration Characteristics on Temporal Scale

The zircon dating results show that the igneous rocks in the study area were mainly formed from the Late Jurassic to the Early Middle Cretaceous, with intense magmatic activity around 160 Ma and 114 Ma (Table 3). Magmatic history can be categorized into multiple stages. The first stage involves volcanic eruptions and magma intrusions around 160 Ma, producing basalt and andesite. These rocks constitute the main lithology of the buried hills. The second stage comes with magma intrusions around 144 Ma and 133 Ma, forming diabase and new diorite, and extensive granodiorite and granite, respectively; the third stage features large-scale magma intrusions around 114 Ma, resulting in the extensive existence of granite and diorite. And andesite displays younger source information (~50 Ma), it indicates that there may be small-scale volcanic activity during this period.

To further analyze the control effect of geological tectonic movements on hydrothermal alteration in different geological periods, the tectonic schists with significantly different geological periods such as Well X-1-2 and the 3800–4000 m in Well X-1-1 were discussed separately, and compared with the diabase of Well X-1-4 and the andesite with complexer material sources in X-2-1 well (Figure 8). The results indicate that multiple hydrothermal alterations have occurred in the study area, and there are differences in the alteration characteristics of different geological ages.

Obviously, the diagenesis time of each igneous rock is different, and alteration usually occurs after the development of cracks, so there are also differences in hydrothermal alteration at different periods. The original rock of the 3800–4000 m tectonic schist in Well X-1-1 was more formed during the Cretaceous period (Figure 8a), the original rock of the tectonic schist in Well X-1-2 was more Jurassic diorite (Figure 8b), the andesite in Well X-2-1 contains Jurassic, Cretaceous, and Paleogene sources (Figure 8c), and the original rock of the altered diabase in Well X-1-4 was formed during the Cretaceous period (Figure 8d). Temporally, Late Jurassic large-scale tectonic movements fractured diorite and andesite after their consolidation. These fractures enabled relatively low-intensity hydrothermal metasomatism, producing minerals like mica and plagioclase (Figure 8a,b). Increased quartz in tectonic schists likely relates to coeval oceanic crust subduction (Figure 8b). Cretaceous volcanic activity generated multi-phase diorite, andesite, and diabase. Subsequent hydrothermal fluids intensely metasomatized hornblende. During the Paleogene, andesite acquired younger sources (Figure 8c). Although diorite and diabase incorporated no new sources, all three rock types experienced strong plagioclase replacement. Minor potassium feldspar increases in tectonic schist/altered diabase, and quartz increases in altered andesite/diabase display weak alteration signatures; we omit further discussion of these subtle features. These similarities and differences reflect combined effects of regional tectonics and spatial rock distribution variations, necessitating detailed analysis with regional tectonic history.

4.2. The Alteration Cause of Buried Hills

The tectonic movement in the research area is very complex, multiple pronounced faults are developed near the Wells X-1-1, X-1-2, and X-1-3 (Figure 9). Previous studies on the Mesozoic Paleo Tethys Plate and the ancient Pacific Plate, as well as multi-scale analysis conducted in this study, indicate that Huizhou Depression in the northern South China Sea experienced multiple episodes of tectonic activities and subsequent alteration, resulting in the complex buried hills’ structure depicted in Figure 9. To analyze the complex alteration mechanism in the study area, this paper proposes a four stage alteration model and conducts a detailed analysis and summary based on geological history.

The first stage features alteration within the late Cretaceous Pacific Plate subduction setting. Although the Pacific Plate’s initial subduction timing remains uncertain [61], subduction commenced by the mid-late Jurassic (~160 Ma). The ancient Pacific Plate subducted beneath the South China Plate, triggering large-scale volcanic eruptions and magma intrusions across the study area (Figure 10a). Subsequent fractional crystallization and crustal assimilation of these magmas generated the buried hills primary lithology (Figure 9, gray shaded area), comprising basalt, andesite, and diorite. During continuous Pacific Plate subduction, tectonic fractures developed in these igneous rocks. These fractures facilitated plagioclase alteration, transforming diorite into tectonic schist and establishing pathways for later fluid–rock interactions. The characteristic quartz enrichment after alteration reflects active continental margin arc magmatism following oceanic crust subduction. This evidence demonstrates that northern South China Sea arc-related igneous rocks include not only Early Jurassic magnesian granites and diorites [62] but also Late Jurassic alkaline series rocks (basalt, andesite, tectonic schist), indicating sustained arc magmatism throughout the Jurassic. Our zircon ages (~160 Ma) and quartz enrichment in schists align with subduction-driven metasomatism.

The second stage involves hydrothermal alteration during the South China Plate’s orogeny. In the Early Cretaceous (145–120 Ma), regional compression uplifted the South China Plate, forming mountains. This compression generated NE-striking thrust faults and NEE/NW-striking secondary strike-slip faults, subsequently facilitating granite, diabase, and diorite intrusions in the X-1 well area. Global-scale volcanic eruptions coincided with this magmatic activity. Zircon dating reveals at least three diorite intrusions in the X-1 well area, followed by two granite/granodiorite intrusions, with diabase forming the latest and smallest bodies. Hydrothermal alteration during this stage primarily chloritized hornblende. Tectonic compression fractured diorite and andesite extensively, creating continuous fracture networks that enhanced hydrothermal fluid access and intensified alteration reactions. The widespread chlorite and illite, alongside absent kaolinite, likely reflects Cretaceous cold–arid climatic conditions (Figure 10b).

The third stage encompasses hydrothermal and weathering alteration driven by buried hills uplift. During the Late Cretaceous (110–66 Ma), tectonic movements elevated buried hills, exposing granite at the Earth’s surface where it underwent weathering. This weathering process likely initiated granite reburial through sedimentation or subsequent tectonic activity, following which hydrothermal fluids altered the rock (Figure 10c). The main characteristics of weathering and alteration are the dissolution of potassium feldspar and carbonate minerals, followed by the clay alteration of potassium feldspar in hydrothermal alteration. Both the weathering and hydrothermal alteration processes contributed to the development of numerous dissolution fractures. This series of events resulted in the formation of weathered granite and a buried hills weathering crust, as depicted in Figure 9 (highlighted in the green shaded area).

The fourth stage is hydrothermal alteration during the Early Paleogene (65.5–45 Ma), The changes in this stage may be related to the initial expansion of the South China Sea and require more research. The current ocean drilling work shows that the process of the South China Sea margin rift is more complex and lengthy than previously thought, and the hydrothermal altered basalt during the South China Sea rift period shows chloritization and albitization of plagioclase, jointly leading to the loss of calcium elements, and a station with hydrothermal alteration aged over 33 Ma was also found near the study area [19,40]. Although the altered basalt in the Pearl River Mouth Basin lacks time evidence of alteration, its location is close to the northern margin of the South China Sea, and there are plagioclase chloritization characteristics consistent with the above research. At the same time, significant dissolution of potassium feldspar occurred in the altered basalt in the study area. Theoretically, the lost potassium element will upflow with the underlying hydrothermal fluid and metasomatize with the overlying andesite, resulting in a significant increase in potassium feldspar in the altered andesite in the X-2 well area. These changes suggest that the initial expansion of the South China Sea may have occurred earlier than expected.

5. Conclusion and Future Work

This paper employs an integrated multi-scale geophysical–geochemical framework to characterize alteration features in buried hills igneous rocks within the Pearl River Mouth Basin, northern South China Sea. We establish a four-stage alteration model based on mineralogical and zircon geochronological data, reconstructing the alteration evolution and controls for these igneous systems. This model not only guides reservoir potential assessment in the Huizhou Depression but also has important significance for the history of continent–ocean transformation and rift expansion in the northern edge of the South China Sea. Key findings are as follows.

The igneous rocks in the northern South China Sea mainly undergo clay alteration with chlorite as the main product, which is mainly controlled by the arid–cold ancient climate of the Late Jurassic and Cretaceous. The migration pattern of potassium elements during the alteration process is controlled by the variation of potassium feldspar content, which can indicate the altered strata. Unlike previous understandings, the improved pore structure conditions of altered igneous rocks indicate that in addition to the well-known weathering crust, altered igneous rock formations also have energy storage potential. Note that the characteristics of mineral changes in altered andesite and altered basalt, as well as the young source exhibited by andesite, are consistent with the hydrothermal alteration characteristics of altered basalt during the South China Sea rift period [40], which also suggests that the initial expansion of the South China Sea may have occurred earlier than previously thought. Paradoxically, our alteration signatures hint at the South China Sea rifting initiating earlier than current models suggest. Future ocean drilling could test this by targeting Paleogene hydrothermal stations near the basin’s northern margin.

The controlling factors of igneous rock alteration characteristics have also been deciphered in research. Tectonic and magmatic processes jointly govern igneous rock alteration in the study area. Tectonically generated fracture networks and stratigraphic uplift enable hydrothermal metasomatism and weathering leaching. Magmatism triggered by distinct tectonic events not only produces varied igneous lithologies but also modifies hydrothermal fluid geochemistry, making the igneous rocks in the study area exhibit complex alteration mechanisms regulated by different minerals such as plagioclase, K-feldspar, and hornblende. This also suggests that different alteration patterns can serve as evidence for the existence of different tectonic activities. For example, hornblende alteration represents a long-term high-temperature and high-pressure environment, and the coexistence of weathering and alteration is related to mountain building movements. Alteration is also a trace of Earth’s activity, and this study awaits more igneous rock samples from around the world to enrich.

Abbreviation

Petrophysics: it is a discipline that studies the properties of rocks (such as porosity, permeability, conductivity, etc.) and their interactions with fluids, mainly through experimental measurements, data analysis, and model construction methods, applied in fields such as oil and gas exploration and reservoir evaluation. Hydrothermal alteration: the interaction between rocks and formation hydrothermal fluids in closed or semi closed environments. Weathered crust: after a certain period of weathering and erosion, a clear weathering and erosion zone is formed, which is then buried and compacted to form a “shell” or “shell zone”. Wireline logging: a method of obtaining formation data through logging instruments in boreholes. Element cutting logging: a method for analyzing the composition of formation elements based on rock cuttings generated during drilling process. SEM: scanning electron microscopy. Core thin section identification: the method of preparing rocks into thin sections and identifying rock composition and lithology through their optical characteristics. XRD: X-ray diffraction for mineral identification. XRF: X-ray fluorescence for elemental composition analysis. KNN: K-nearest neighbor algorithm. ISOCON: Chemical Isoconcentration Line Analysis. LA-ICP-MS: laser ablation-inductively coupled plasma-mass spectrometry. FMI: Formation MicroScanner images, a logging method for stratigraphic imaging.