Eocene Paleoclimate Evolution under the Background of Warmhouse-Hothouse Conditions in the Continental Fushun Basin: Implications from Magnetic Susceptibility and Color Reflectance.

Paleocene-Eocene hyperthermal events are a current research focus in the fields of sedimentology and paleoclimatology. The Fushun Basin in northeast China contains continuous continental Eocene fine-grained rocks, and a series of Eocene hyperthermal events in the Fushun Basin have been identified. Because of the high cost of high-precision isotope data testing, it is necessary to find new and alternative paleoclimate parameters. In this study, Eocene coal and oil shale-bearing layers in the Fushun Basin are used as research objects. The high-precision data of magnetic susceptibility, color reflectance, rock composition, and cluster analyses are used to conduct a vertical comparison in the same category and compare that analysis with the identified Eocene hyperthermal events in the Fushun Basin. The preliminary results show that high-frequency-dependent susceptibility, high color reflectance a* (redness)/L* (lightness) values, and high kaolinite content in the study area have good correspondence with global hyperthermal events and can be used as effective parameters for the identification of continental basin hyperthermal events. The detailed magnetic susceptibility and color reflectance data also reveal that the Eocene strata in the Fushun Basin recorded the Late Lutetian Thermal Maximum (LLTM) and 13 short-term hyperthermal events during the Early Eocene Climatic Optimum (EECO). These results indicate that the parameters of rock physical properties can be used to study the evolution of the paleoclimate in geological history, and it has universal practicability in continental and marine fine-grained sedimentary rocks.

Introduction

Paleogene global paleoclimate change has always been a research focus in geological circles, among which the study of hyperthermal events is the most representative. During the late Paleocene and early Eocene, a series of transient global warming events occurred against the background of global warming.[1−4] These global warming events are called hyperthermal events. The most famous is the Paleocene–Eocene thermal maximum (PETM). In marine strata, the amount of research on Paleogene hyperthermal events is relatively high, and carbon isotope excursions (CIEs) in marine sediments can adequately indicate hyperthermal events. Frequent hyperthermal events during the Eocene (55–47 Ma) have been identified in marine records worldwide. Specifically, the absolute ages of approximately 28 short-term hyperthermal events recognized in the Eocene are based on δ13C data from marine sediments collected in the Integrated Ocean Drilling Program (IODP).[4−9] In recent years, many hyperthermal events similar to the PETM in nature have been found, such as the Eogene Thermal Maximum (ETM), J, L, M, N, O, P, Q, R, S, T, U, V, and W.[8−12] According to previous studies, the Cenozoic climate can be divided into four states: hothouse, warmhouse, coolhouse, and icehouse. The hothouse state lasted from 56 Ma to 47 Ma, and the temperature was 10 °C higher than that at present. The warmhouse states were from 66 Ma to 56 Ma and from 47 Ma to 34 Ma.[13]

Few studies have focused on Paleogene hyperthermal events in continental sedimentation. The identification of hyperthermal events has been carried out only in some areas on land.[14,15] In the Fushun Basin, northeastern China, according to the δ13C total organic carbon (TOC) values, the ages of CIEs and a comparison with the IODP sites, three Eocene continental hyperthermal events were recognized: ETM2, ETM3, and the long-term Early Eocene Climatic Optimum (EECO).[15] In southern China, the magnitude of the CIEs and the changes in temperature and precipitation during the PETM have been revealed by using carbonate-associated sulfate sulfur and oxygen isotopes and reconstructing the carbon isotopes (Δ) of C3 plants.[16,17] On the basis of lacustrine sediments from deep drill core SKD1 and mineralogical and geochemical evidence, a high-resolution record of the PETM event in the Jianghan Basin is proposed for the first time. The palynological record from the black shale of the lowermost Yangxi Formation exposed in the western Jianghan Basin has revealed the impact of the hyperthermal event on terrestrial ecosystems.[18] In Tibet, China, the study of shallow marine carbonates in the Tethys Himalaya of southern Tibet reveals the biotic response to sea level change in shallow marine environments during the PETM and the carbon isotope expression of the PETM in a shallow-water–carbonate platform.[19] The stratigraphic record of the Paleocene–Eocene Thermal Maximum in the Xigaze forearc basin (southern Tibet) and in NW Himalaya (India) represents one of the few records of the PETM in active continental margins worldwide.[20,21]

The Fushun Basin in northeast China contains complete continental Eocene strata that are mainly organic-rich, fine-grained rocks, which are the best carriers for establishing paleoclimate evolution archives. On the basis of previous studies on magnetic stratigraphy, biostratigraphy, and isotopic stratigraphy, the chronostratigraphic framework of the study layer is preliminarily constrained.[22,23] On the basis of the new Ar–Ar isotopic dating of the volcanic tuff of the Lizigou Formation, the possible oldest age at the bottom of the Guchengzi Formation coal seam is 55.07 ± 1.18 Ma.[14] Combined with the stable isotope analysis of organic matter carbon in thick coal seams, it is found that the sedimentary records include the Paleocene–Eocene thermal maximum (PETM), ETM2/H1, H2, I1, and I2. However, the sedimentary records are limited to a single coal seam, while the analysis of hyperthermal events has not been carried out for other coal-bearing strata and oil shale strata. In the study of Li et al. (2022),[15] based on U–Pb isotope dating (54.72 ± 0.20 Ma) and astronomical dating analysis, in the sedimentary periods of the Guchengzi Formation and Jijuntun Formation (54.72–46.5 Ma) in the Fushun Basin, the δ13CTOC and contemporaneous marine strata δ13C values were compared to identify the short-term hyperthermal events of ETM2 and ETM3, and the long-term warming events (long-term hyperthermal events) of the EECO (∼53–46.5 Ma) and 22 short-term hyperthermal events were identified within EECO events, namely, L, M, N, O, P, Q, R, S, T, U, V, W, C22nH3, C22nH4, C22nH5, C21rH1, C21rH2, C21rH3, C21rH4, C21rH5, C21nH1, and C21nH2.

According to the geological time scale (GTS2012),[24] the sedimentation of the upper Lizigou Formation to the Guchengzi Formation and Jijuntun Formation in the study area occurred from 54.72 to 47.8 Ma and from 47.8 to 41.2 Ma.[15] According to the global paleoclimate characteristics in the same period, the Fushun Basin in the Paleocene in this stage was in a warmhouse–hothouse background overall; however, high-precision isotope analysis is particularly expensive. On the basis of the multistage records of hyperthermal events in the Fushun Basin, this study uses the Eocene Guchengzi Formation coal, mudstone, carbonaceous mudstone, and Jijuntun Formation oil shale in the Fushun Basin as the research object and carries out the identification of hyperthermal events and the study of paleoclimate evolution based on systematic magnetic susceptibility, color reflectance, and mineral composition data, which are helpful to improve the Paleogene paleoclimate archives of northeast Asia and provide alternative indicators for studying paleoclimate evolution, such as hyperthermal events.

Geological Setting

The Fushun Basin, located in southern Fushun City, Liaoning Province, China, is a famous coal-bearing and oil shale-bearing basin in China. The Hun River is located in the northern part of the basin, and hills are located in the south. The basin starts from the Guchengzi River in the west and the Dongzhou River in the east. It is approximately 19 km long and 4 km wide, with a total area of approximately 76 km2 (Figure ).

Figure 1

Regional geological map of the Fushun Basin. (a) Fushun Basin global location (50 Ma).[3,15] (b) Fushun Basin location map.[22,25] (c) Distribution of the Fushun Basin in the Dunhua–Mishan fault zone.[22,25] (d) Stratigraphic column of the Fushun Basin.[22,25] (e) Geological structural unit of the Fushun Basin.[22]

The Fushun Basin is located on the western side of the Dunhua–Mishan (“DunMi”) fault zone, which is the northern extension of the Tanlu fault zone. It is a small continental strike-slip fault basin formed on Proterozoic gneiss that formed during the Paleogene rift cycle.[26] Its sedimentary fill comprises the Paleocene Laohutai and Lizigou Formations and the Eocene Guchengzi, Jijuntun, Xilutian, and Gengjiajie Formations. Oligocene strata are absent.[22] This research focuses on the Guchengzi Formation and Jijuntun Formation. The Guchengzi Formation consists of a thick coal seam intercalated with carbonaceous shale, and the thickness of the coal seam decreases from west to east and from south to north. The Jijuntun Formation is composed of thick oil shale and is stably distributed throughout the whole region.[27,28] The coal and carbonaceous mudstone of the Guchengzi Formation were deposited in a swamp environment, and the oil shale of the Jijuntun Formation was deposited in shallow and deep lake environments.[29]

Samples and Methods

Samples

The Guchengzi Formation (54.51–47.80 Ma) and Jijuntun Formation (47.8–41.2 Ma) in the Fushun Basin are organic-rich, fine-grained sediments that were continuously deposited in the Eocene,[15] and they are important records for studying Eocene hyperthermal events. Therefore, in this study, the coal and carbonaceous mudstone of the Guchengzi Formation and the oil shale of the Jijuntun Formation in the Liaofudi 1 well in the Fushun Basin were systematically sampled. The sampling interval was approximately 0.5 m. Sixty-eight samples were collected from the Guchengzi Formation, including 28 carbonaceous mudstones, 10 mudstones, and 30 coals. A total of 157 samples were collected from the Jijuntun Formation, including 2 mudstones and 155 oil shales. A total of 225 samples were collected, dried, and ground into powder (200-mesh). The samples were tested for color reflectance and magnetic susceptibility, and a core of the Jijuntun Formation was scanned by infrared spectroscopy. All experiments were completed in the Key-Lab for Oil Shale and Paragenetic Minerals of Jilin Province.

Principle and Experimental Method of Magnetic Susceptibility

In this paper, the magnetic susceptibility of 225 samples obtained from the target layer was analyzed. During the measurements, 5–6 g samples were placed in a nonmagnetic 2 cm × 2 cm × 2 cm polystyrene box for testing. The MS2B susceptibility tester of the British Bartington Company was used to carry out the magnetic susceptibility test and followed the standard of GB/Z 26082-2010 nanomaterial DC magnetic susceptibility (magnetic moment). The magnetic susceptibility test was repeated 10 times, and the error between data points from the same sample was less than 0.02 × 10–8 m3/kg. The low-field frequency was 0.47 kHz, and the high-field frequency was 4.7 kHz. Each sample was tested with a range of 0.1 10 times, and the averages of the low-field susceptibility (χlf), high-field susceptibility (χhf), and frequency-dependent susceptibility (χfd%) were obtained through Multisus software, where χfd% = (χlf – χhf)/χlf × 100%.

Principle and Experimental Method of Color Reflectance

The Commission International d’Eclairage 1976 L*a* and b* (CIELAB) color expression and measurement system specified by the International Lighting Commission in 1976 is quoted,[30] in which the L* (lightness) value represents brightness (L* = 0% represents black and L* = 100% represents white), a* (redness) value represents redness (a positive value tends toward red, and a negative value tends toward green), and b* (yellowness) value represents yellowness (a positive value tends toward yellow, and a negative value tends toward blue); these values indicate color reflectance characteristics.

A total of 225 samples were individually placed on the white background reference color plate, compacted, and flattened to ensure that the sample surfaces were uniform, smooth, and flat. A WR-18 color difference instrument (the test parameters were as follows: the light source was a CIED65 standard light source, and the aperture was 4 mm) was used to measure the color reflectance. Three positions were randomly selected on each sample for testing, and then the average value was calculated to obtain the color reflectance value of the sample. The color reflectance was followed by the SH/T 0168-92 petroleum product color reflectance determination standard, the color reflectance test was repeated three times, and the error between data points from the same sample was less than 0.05%.

Infrared Spectrum Scanning

This paper used a Hylogger core hyperspectral scanner to scan only the Jijuntun Formation core by infrared spectroscopy. On the basis of the reflection spectrum analysis technology, the reflection spectrum of the core in the wavelength range of 400–2500 nm was collected and measured by spectrometry, and different minerals were calculated and identified according to their spectral diagnostic characteristics to form mineralogical information. Because of the limitation of sample integrity, a total of 174 rock component data points were measured. The main minerals that were analyzed were kaolinite, quartz, smectite, carbonate, feldspar, and mica. It should be noted that due to the limited precision of the testing instruments, only high-content minerals could be identified.

Cluster Analysis and Wavelet Analysis

Because of the large amount of data, many kinds of components, and high-frequency fluctuations in values, it is difficult to interpret the paleoclimate. To more intuitively reflect the paleoclimate change and carry out research on the corresponding relationship with the identified hyperthermal events, the mathematical statistical software SPSS24 was used to cluster indexes, such as the low-field susceptibility (χlf), high-field susceptibility (χhf), frequency-dependent susceptibility (χfd%), brightness (L*), redness (a*), and yellowness (b*) (Figure ). Cluster analysis was carried out by using the method of intergroup connection in systematic clustering, and then the clustering results were counted. The data classified into a class after cluster analysis were averaged, and the average value was used to replace all the data in this class. The change in the trend of the obtained result chart is more obvious than that of the untreated result chart, which is conducive to analysis and discussion. The frequency susceptibility was determined by the low-frequency magnetic susceptibility, and high-frequency susceptibility was obtained, which is more representative. Therefore, in this study, the frequency susceptibility was analyzed by MATLAB software (Figure ).

Figure 2

Cluster analysis process.

Figure 3

Magnetic susceptibility and color reflectance experimental results. (Fine dating scale of the Eocene strata in the Fushun Basin are according to Li et al. (2022);[15] paleotemperature data are according to Westerhold et al.;[13] the global temperature comes from the records of deep-sea benthic foraminifer oxygen isotopes. The record was first converted to a deep-sea temperature and then projected to the surface air temperature change.)

Results

In the time stratigraphic framework built by predecessors,[14,15,22,23] considering that the Guchengzi Formation (52.2–47.8 Ma) is mainly coal-bearing swamp deposits and the oil shale of the Jijuntun Formation (47.8–41.2 Ma) is stable shallow lake and deep lake deposits,[15] the Guchengzi Formation Jijuntun Formation can be divided into two stages according to the results of magnetic susceptibility and color reflectance: the early stage (52.2–47.8 Ma) and the late stage (47.8–41.2 Ma).

Rock Composition

The Guchengzi Formation is mainly composed of thick coal seams with mudstone and carbonaceous mudstone. The Jijuntun Formation is mainly composed of thick oil shale deposits. From 52 Ma to 47.8 Ma, which is the depositional stage of the Guchengzi Formation, from bottom to top, the rock composition gradually transforms from carbonaceous mudstone intercalated with coal seams and mudstone to thicker coal seam deposits, in which a small section of tuffaceous sandstone deposits are interspersed; from 47.8 Ma to 42.1 Ma, which is the depositional stage of the Jijuntun Formation, all thick oil shale deposition occurred, with alternating dark gray oil shale, gray black oil shale, and black oil shale.

Among these deposits, the measured rock component data show that the minimum kaolinite content is 15.9%, and the maximum content is 84.2%. From 47.8 Ma to 42.1 Ma, the kaolinite content declined, except at 47.5 Ma and 41.3 Ma, and the values fluctuated twice, with peaks of 70.5% and 61.4%, respectively. Then, the kaolinite content continued to decline until 41.2 Ma. The change in the trend of the mica content is generally opposite that of the kaolinite content; quartz, smectite, carbonate, feldspar, and other minerals appear intermittently, so their trends are not analyzed (Figure ).

Figure 6

Relationship between whole-rock analysis and paleoclimate (paleotemperature data are according to Westerhold et al., 2020[13]).

Magnetic susceptibility

Overall, the trends of χlf and χhf are basically the same. In the early stage (I, 52.2–47.8 Ma), from bottom to top, χlf and χhf show a downward trend overall, but at the end of this stage, the two parameters suddenly and rapidly increase, up to 32.1 × 10–8 m3/kg and 28.6 × 10–8 m3/kg (approximately 47.8 Ma), during which four high-value fluctuations occur, namely, 21.9 × 10–8 m3/kg, 19.3 × 10–8 m3/kg, 27.1 × 10–8 m3/kg, and 21.9 × 10–8 m3/kg for χlf and 18.8 × 10–8 m3/kg, 10.7 × 10–8 m3/kg, 26.8 × 10–8 m3/kg, and 19.6 × 10–8 m3/kg for χhf; χfd values show an increasing trend, but the overall range is 0.4–54%; the values reach the maximum value of 77.78% at approximately 48.4 Ma and then rapidly reduc to 8%.

Late stage (II, 47.8–41.2 Ma), χlf and χhf generally show a slight downward trend, with a small amplitude and low value fluctuations at approximately 45.0 Ma and 43.0 Ma, and the χlf minimum value is 3 × 10–8 m3/kg, 1.7 × 10–8 m3/kg, The lowest value of χhf can reach 1.4 × 10–8 m3/kg and 1.3 × 10–8 m3/kg; χfd has been kept low, ranging from 3.47–17.24%, and the highest value of 53.3% at this stage appears at approximately 45.0 Ma.

Color Reflectance

On the whole, the color reflectance L* value first increases and then decreases, and the a* value and b* value generally show an upward trend. In the early stage (I, 52.2–47.8 Ma), from bottom to top, L* increased from 44.44% to a stable high value, ranging from 54.66–68.68%, decreased to 40.54% after approximately 49.2 Ma, and rapidly increased to 55.85% at 47.8 Ma; a* and b* showed an upward trend, but b* fluctuated more violently. a* gradually increased from 3.36% to 7.66% and then decreased to 4.08% after approximately 49.2 Ma There was a small amplitude low value fluctuation at 48.8 Ma, with a minimum of 3.31%. The b* value slowly increased from the initial 4.15% to a stable high value, ranging from 10.05–13.44%. There was a small amplitude low value fluctuation at 49.4 Ma, with a minimum of 1.21%, which rapidly decreased to 2.50% at 47.8 Ma.

In the late stage (II, 47.8–41.2 Ma), the change in the trend of the overall color reflectance value was weak. From 47.8 Ma to 44.6 Ma, L* was relatively stable, showing a slight downward trend, with small fluctuations in the middle, and showing an upward trend at the end of this stage. The overall trends of the a* value and b* value were approximately the same. At this stage, the values showed a slight upward trend. After approximately 47.0 Ma, the values fluctuated stably, ranging from 3.92–5.65% and 3.22–10.81%, respectively.

Discussion

Detailed Analysis of the Paleoclimate

Magnetic Susceptibility, Color Reflectance, and Paleoclimate

Environmental magnetism has been successfully applied to reconstruct the climate and environmental evolution of marine and continental sediments.[31] The magnetic susceptibility is a measure of the difficulty of magnetizing a substance. It is mainly determined by the composition, content, and particle size of magnetic minerals.[32−34] Previous studies have shown that the magnetic minerals in the mineral composition of coal and oil shale in the Fushun Basin are mainly pyrite and siderite.[35] In other words, it is related to the depositional environment[36] and the sediment source.[37] The source rocks in the Eocene Fushun Basin are mainly of mixed felsic and mafic affinity[28] and remain unchanged for a long time. At the same time, there were only two sedimentary environments. The Guchengzi Formation formed in a swamp environment. The Jijuntun Formation formed in shallow and deep lake environments,[15] and we analyzed the magnetic susceptibility and chromaticity in these two units. Therefore, this analysis can reliably indicate the environmental change and climate change in one sedimentary environment. The magnetic minerals mainly come from the weathering process; when the climate is warm and humid, the increase in weathering intensity leads to more ultrafine magnetic particles in the source area, which may increase the frequency-dependent susceptibility (χfd%). Therefore, in a high-temperature and humid environment, the low-field susceptibility (χlf) is low, and χfd% is relatively high, while in dry and cold environments, χlf is relatively high, and χfd% is low.

The color reflectance mainly depends on the mineral composition (mainly carbonate) and organic matter content of sediments. It is the most intuitive feature of sediments, and the strata in this study are organic-rich fine granular rock deposits. The color reflectance can be used to understand climate change,[38,39] and it can indicate climate change and the redox degree under different climates. Previous studies have shown that in modern soils[36] and loess paleosols,[40] the changes in the a* values are very sensitive to warm and humid climates. The redness (a*) of fine-grained sediments is closely related to the temperature and precipitation conditions affecting sediment oxidation during weathering.[41] The L* value reflects the change in the carbonate content in sediments from arid to semiarid areas.[36] Under a dry and cold climate, the carbonate content is high, the temperature is low, the precipitation is low, the L* value of the sediment is high, and the a* value is low. In contrast, under humid and hot climate conditions, the carbonate content is low, the temperature and precipitation are high, the sediment L* value is low, and the a* value is high. Therefore, a higher a*/L* value indicates that the climate is relatively humid and warm; in contrast, lower a*/L* values indicate relatively dry and cold climatic conditions.[42,43] The correlation between the b* value, which represents yellowness (a positive value tends toward yellow, and a negative value tends toward blue), and the a* value is obvious. High values of b* and a* indicate humid and hot climates, and low values indicate relatively dry and cool climates. Compared with other methods, color reflectance is more convenient and faster and has wide application prospects in climate change research.[40]

By comparing the magnetic susceptibility data and color reflectance data after clustering with the global average temperature given by Westerhold et al. (2020),[13] a good corresponding relationship with the paleotemperature can be obtained (Figure ). In the early stage (I, 52.2–47.8 Ma), the low-field susceptibility (χlf) at the beginning first decreased, and a relatively obvious increase until 50.8 Ma occurred. The a* values first increased, then fluctuated slightly at approximately 5.7%, and then decreased. The change in the trend of the b* value was basically the same as that of the a* value, which generally increased first and then decreased, indicating that the temperature increased first and then decreased at approximately 52.2–50.5 Ma. During 49.8–48.0 Ma, the low-field susceptibility (χlf) maintained a low value. The a*/L* values increased gradually at this stage. The frequency-dependent susceptibility (χfd) at the end increased significantly, and the a* value, b* value, and a*/L* value decreased rapidly. These results show that at this stage, the climate was generally warm and humid and finally became hot and humid, while the hot and humid climate at the initial stage was conducive to the formation of fine-grained ferromagnetic minerals; thus, the frequency-dependent susceptibility (χfd) was initially high and was significantly reduced at the end, which further indicates the transformation of the climate from hot and humid to warm and humid; this transformation can also be used as a sign of the transformation of the Fushun Basin paleoclimate from hothouse to warmhouse conditions. The paleotemperature also showed a process of first increasing, stabilizing, and finally decreasing from 52.2 to 47.8 Ma.

Figure 4

Relationship between magnetic susceptibility, color reflectance, and paleoclimate (paleotemperature data are according to Westerhold et al., 2020[13]).

In the late stage (II, 47.8–41.2 Ma), the low-field susceptibility (χlf) fluctuated within a certain range and decreased slightly overall. The a* and b* values decreased rapidly at first and then changed little, basically fluctuating at approximately 4.76% and 6.36%, respectively. Among these values, the b* values fluctuated the most, and the a*/L* values stabilized and decreased slightly at the end of this stage, corresponding to the decline in the paleotemperature. However, at 44.8 Ma and 43.1 Ma, the low-field susceptibility (χlf) appeared to be low. After the frequency-dependent susceptibility (χfd) dropped to a low value, it remained stable, with occasional high values, and a high value appeared at the corresponding position where the low-field susceptibility (χlf) appeared, which corresponds to the small warming process of paleotemperature under the background of warmhouse conditions. The high values of the frequency-dependent susceptibility (χfd), a* value, b* value, and a*/L* value at approximately 41.5 Ma can better correspond to the Late Lutetian Thermal Maximum (LLTM).

The comparative analysis shows that although there are several valley peak fluctuations in the middle of the magnetic susceptibility curve, which does not correspond well to the paleotemperature curve, the valley and peak values have good overall correspondence with the peak and valley values in the temperature curve. Furthermore, combined with the color reflectance curve and the paleotemperature curve, they have a good correspondence. Therefore, the magnetic susceptibility and color reflectance can be used to analyze the paleotemperature trend.

Generally, from 52.2 Ma to 41.2 Ma, the Fushun Basin changed from hot and humid to warm and humid. The sudden increase in the low-field susceptibility (χlf) and the sudden decrease in the infrequency-dependent susceptibility (χfd), as well as the a* and b* values at the boundary between the Guchengzi Formation and Jijuntun Formation (487 m (approximately 47.8 Ma)), can be used as a record of this transformation.

Identification of Hyperthermal Events

In conclusion, both the frequency-dependent susceptibility (χfd) and color reflectance can be used as paleoclimate indicators. During hyperthermal events, the temperature rises, the climate is hot and humid, and the color reflectance a*/L* values are high. In this climate, weathering is enhanced, which leads to an increase in ultrafine magnetic particles and high-frequency-dependent susceptibility (χfd). Therefore, when the paleoclimate background is known, the use of magnetic susceptibility and color reflectance can reveal hyperthermal events.

In this paper, the magnetic susceptibility and color reflectance indexes in the period of 52.2 Ma to 47.0 Ma are selected and compared with δ13CTOC values in the same period.[3,17,44] The cluster analysis results of the two representative indexes, χfd and a*/L*, are compared with the δ13CTOC values in the same period (as shown in Figure ). The δ13CTOC values revealed short-term extreme heat events L, M, N, O, P, Q, R, S, T, U, V, W, C22nH3, C22nH4, C22nH5, C21rH1, C21rH2, C21rH3, C21rH4, C21rH5, C21nH1, and C21nH2 during the EECO. The high values of χfd and a*/L* have a good corresponding relationship, and the high values of the two parameters correspond to the hot and humid climate. These results show that the hyperthermal events can also be well recorded by the frequency-dependent susceptibility and color reflectance index. Therefore, the frequency-dependent susceptibility is applied, and the χfd and a*/L* values reveal the hyperthermal events; 13 short-term hyperthermal events are identified, including N, O, P, Q, T, U, V, C22nH3, C21rH1, C21rH2, C21rH3, C21rH5, and C21nH1. However, L, M, R, S, W, C22nH4, C22nH5, C21rH4, and C21nH2 are not recognized, which may be affected by factors such as the insufficient accuracy of magnetic susceptibility and color reflectance tests and poor sensitivity of the color reflectance index to paleoclimate change.

Figure 5

Comprehensive analysis chart of hyperthermal events (the GTS2012 is from Ogg, 2012;[24] δ13CTOC values are from Li et al., 2022;[15] Possagno is from Luciani et al., 2016;[45] site 1258 is from Turner et al., 2014;[3] hyperthermal events are from Westerhold et al., 2018[4] and Galeotti et al., 2019.[9] Among the hyperthermal events, the symbol before the letter “H” represents magnetochrons, such as C23r. The letter “H” and the number after it represent hyperthermal events, which are identified by CIEs.)

By comparing the frequency-dependent susceptibility (χfd) curve and spectrum after wavelet analysis with the identified hyperthermal events in the Fushun Basin and the global hyperthermal events (Figure ), cycles 1, 2, 4, 5, 6, and 7 in the χfd curve and the spectrum can well correspond to the fluctuation of the global average temperature, which increases first and then decreases. The Late Lutetian Thermal Maximum (LLTM)[13] in cycle 1 and 41.2 Ma to 42 Ma are relatively consistent, which can be used as the basis for LLTM identification. The identification of multistage hyperthermal events further shows that the Guchengzi Formation has a hothouse background. The identification also confirms that the paleoclimate of the Fushun Basin changed from hot and humid to warm and humid from 52.2 Ma to 41.2 Ma. The identification is also sufficient to prove that hyperthermal events can be completely recorded in the extremely stable terrestrial lake sediments of the Fushun Basin, and the frequency-dependent susceptibility (χfd) and color reflectance indicators also provide a good indication of the change in the paleoclimate. This indicator can be used as an alternative index for fine paleoclimate identification and climate event identification when the overall climate background is known.

In addition, there are differences and similarities in the recognition of hyperthermal events by color reflectance, magnetic susceptibility, and carbon isotopes. The difference is the accuracy of identification. For example, hyperthermal events can be directly identified by organic carbon isotopes, and the CO2 concentration and paleotemperature are also estimated quantitatively by bulk δ13CTOC. Magnetic susceptibility and color reflectance can indicate only the climate trend. The identification of hyperthermal events can be carried out only under a known climate background based on these parameters. The common point is that, in the identification markers, both negative carbon isotope excursions (CIEs) and positive excursions of magnetic susceptibility and color reflectance indicate that the climate became warm and humid.

Rock Composition and Paleoclimate

Because of shallow burial, the vitrinite reflectance (Ro) is less than 0.5%, weak diagenesis occurred in the study area,[29] and the fine-grained sedimentary rocks are basically primary clay minerals. The paleoclimate has an impact on organic matter and rock mineral composition.[46−50] Kaolinite is the main product of chemical weathering,[51−53] and a warm and humid climate is conducive to the formation and deposition of kaolinite.[54] The existence of smectites indicates cold climate characteristics;[55] quartz has a strong anticorrosion ability, and feldspar is far easier to weather than quartz. Under the background of a warm and humid paleoclimate, feldspar is less abundant and is easily transformed into clay.

During 47.8–45.8 Ma, the kaolinite content decreased (from 84.2% to 61.4%), and a loss occurred near 47.6 Ma (Figure ), indicating that the climate changed from hot and humid to dry and cool, which corresponds to the overall decline in the global average temperature. From 45.8 to 44.5 Ma, the kaolinite content was generally low, approximately 60%, indicating that the climate was relatively dry and cool. The kaolinite content first increased and then decreased at approximately 45.4 Ma, which is reflected in the short relative high-temperature and high-humidity period against the background of relatively dry and cool conditions, which is similar to the variation in the global paleotemperature in the same period. During 44.5–43.3 Ma, the kaolinite content first decreased and then increased, indicating that the paleoclimate was first dry and cool, and then the temperature rose and initiated a humid and hot climate; this pattern corresponds to the global average temperature first decreasing and then increasing. In the period of 43.3–41.2 Ma, the kaolinite content showed a gentle “double peak”, indicating that the climate had two alternating upward and downward evolution processes; in addition, two high peaks correspond to the global average temperature, of which the last kaolinite content peak corresponds to the Late Lutetian Thermal Maximum. This stratum was deposited at 41.5 Ma, which is basically consistent with the occurrence time of the LLTM in marine strata, further confirming the response of kaolinite to hyperthermal events.

In addition, from 47.8 to 41.2 Ma, smectites appear only in the upper part of the Jijuntun Formation, which also shows that in the Jijuntun Formation, the climate changed from hothouse to warmhouse conditions, which also corresponds to the overall downward trend of the global average temperature. In general, the trend of the kaolinite content corresponds well with the global paleotemperature curve, which can be used as a parameter for fine-scale paleoclimate research.

Quantitative Characterization Parameters

Overall, the magnetic susceptibility has a baseline under warmhouse and hothouse conditions. The baselines of low-field susceptibility (χlf) are 13.6 × 10–8 m3/kg under warmhouse conditions and 2.69 × 10–8 m3/kg under hothouse conditions. The baselines of high-field susceptibility (χhf) are 11.14 × 10–8 m3/kg under warmhouse conditions and 1.76 × 10–8 m3/kg under hothouse conditions. The baselines of frequency-dependent susceptibility (χfd%) are 8.19% under warmhouse conditions and 26.81% under hothouse conditions. However, the overall range of the color reflectance values is small, so it is difficult to identify the baseline. These baselines may be used as the assessment index of warmhouse and hothouse conditions. The baselines of magnetic susceptibility and color reflectance vary in different regions due to differences in the clastic input, provenance type, etc. Moreover, only under the known general paleoclimate background can these parameters be used for fine paleoclimate analysis.

Because of the comparability of global hyperthermal events, this study provides alternative indicators for the study of paleoclimate evolution and has a certain guiding significance for understanding the evolutionary pattern of the modern Earth climate and future environmental predictions.

Conclusion

The Paleogene Guchengzi Formation and Jijuntun Formation in the Fushun Basin were deposited under warmhouse and hothouse conditions. In this study, relying on the high-precision data of magnetic susceptibility, color reflectance, and rock composition, the method of cluster analysis was used to reveal the fluctuation characteristics of the paleoclimate under the background of warmhouse–hothouse conditions.

The high-precision magnetic susceptibility values, color reflectance values, and clay mineral contents in the time period have a good corresponding relationship with the identified Eocene hyperthermal events and global average temperature in the Fushun Basin, which can be used as the basis for paleotemperature change and an effective alternative parameter for paleoclimate research.

During 42.0 Ma to 41.2 Ma, based on all data and according to the comparative analysis of δ13CTOC, 13 short-term hyperthermal events, including N, O, P, Q, T, U, V, C22nH3, C21rH1, C21rH2, C21rH3, C21rH5, and C21nH1, can be identified in the EECO long-term events.

From 52.2 Ma to 41.2 Ma, the paleoclimate generally presented a hot and humid to warm and humid trend. During the period of 50.0 Ma to 48.0 Ma, the paleotemperature still underwent a small warming stage against a hothouse background. The paleoclimate cooled rapidly at 47.8 Ma, the paleoclimate quickly changed from a hothouse background to a warmhouse background, and a small warming process occurred during the period of 42.0 Ma to 41.2 Ma.