Introduction

Paleokarst carbonatite reservoirs are widely distributed in China. With expanding applications of new technologies and methods for petroleum exploration, numerous paleokarst reservoirs have been identified in the Tarim Basin (Wu et al. 2012; Zhang et al. 2004), Qaidam (Feng et al. 2013), Ordos (Xia et al. 2007), and in the Sichuan Basin (Zhang et al. 2011). These findings have led to the development of various medium- and large-scale oil and gas fields, such as the Tahe, Lunnan, Changqing, and Weiyuan. This indicates that the paleokarst carbonatite reservoirs in China have great exploration potential (Jin and Cai 2006); therefore, expanding research into paleokarst reservoirs is very important for marine oil and gas exploration (Jia and Cai 2004; Liu 2007; Su et al. 2010).

Secondary karst reservoirs, such as corrosion pores, caves, and fissures of the Tarim Basin, the biggest petroliferous basin in China (Wu et al. 2012), were formed by paleokarstification and provide significant reservoir spaces for oil (Zhang et al. 2011; Wang and Al-Aasm 2002; Chen et al. 2007; Ni et al. 2009). Deep-burial karstic reservoirs are important for marine carbonatite oil and gas fields. These reservoirs have a long geological history of environmental change and are influenced by corrosion and infilling involving several types of fluids over a variety of periods and from various factors to form heterogeneous and complex structures (Li et al. 1996; Zhu et al. 2005). These structures create difficulties in predicting deeply buried paleokarst carbonatite reservoirs.

Previous studies on fluid inclusions have been applied mainly to research on mineral deposits. Marray (1957) discovered large oil and gas inclusions in self-formed quartz. From the late 1970s to the early 1980s, fluid inclusion analysis has been used widely as a research tool in petroleum geology and paleokarstic reservoirs (Roedder 1979; Pagel and Poty 1983; Haszeldine et al. 1984; Pagel et al. 1985; Burley et al. 1989), promoting worldwide research on paleokarst and diagenetic fluids.

Fluid inclusions reflect the temperature and pressure during the formation of minerals and the properties of the initial fluid. Homogenization temperature can be used to investigate the temperature of paleofluids and to predict the geotemperature and geothermal evolution histories of basins (Osborne and Haszeldine 1993; Huang et al. 2008). The salinity and composition of the inclusions can yield information on the properties and origins of mineralizing fluids in basins (Veizer et al. 1999; Munz 2001; Wen et al. 2014). Inclusion analysis also serves as an important tool for studying diagenetic periods and environments (Mafia 1986; Yan et al. 2005; Liu et al. 2012).

Halahatang oil field, a key exploration area in Tarim Basin, includes well-developed Ordovician paleokarst carbonatites. Scientists have conducted numerous studies on the geological evolution, seismic response, characteristics of oil and gas reservoirs, and mechanisms of karst development in this area (Ni et al. 2010; Zhang et al. 2012, 2014; Wang et al. 2013; Liu et al. 2014). Constrained by limited research methods, data, and knowledge, currently, only Wang et al. (2013) have conducted preliminary inclusion research on Ordovician oil reservoirs in the Halahatang area. In contrast, nearby areas, such as Tahe and Tazhong oil fields, have been studied extensively (e.g., He et al. 2010; Zhang et al. 2011; Li 2013; Liu et al. 2013). Based on the analysis of rock fabric, this study systematically analyzes the isotopic compositions of carbon and oxygen and conducts experimental analysis on fluid inclusions to understand the paleoenvironmental conditions of Ordovician carbonatite paleokarstification. The results of this study will provide geochemical evidence for the prediction and exploration of karst reservoirs.

Study area and methods

Study area

The Halahatang oil field is located on the Halahatang nose-like tectonic belt in the Ordovician anticline of the Lunnan low uplift, which is located in the central part of the northern Tarim uplift. It encompasses an area of about 4000 km2, neighboring the Luntai low uplift to the south, the Northern Depression to the north, the Yingmaili low uplift to the west, and the Lunnan low uplift to the east. This field is a favorable area for oil and gas migration from low to high potential areas, which favors petroleum accumulation (Fig. 1). It contains four blocks—the Ha6, Xinken, Rewapu, and Qiman—which are key exploration blocks of the Tarim Oilfield Company of China National Petroleum Corporation (CNPC) for Ordovician carbonatites in northern Tarim Basin (Zhang et al. 2012).

Fig. 1
figure 1

Early Silurian geologic–tectonic background in the Halahatang area of northern Tarim Basin

The Ordovician strata denudates and pinches out from south to north, outcropping the upper Ordovician Sangtamu (O3 s), Lianglitage (O3 l), Tumuxiuke (O3 t), middle Ordovician Yijianfang (O2 yj), and lower Ordovician Yingshan formations (Fig. 2), respectively. The areas north and south of the pinch-out line of the Sangtamu Formation are dominated by a weathered crust karst of the paleo-hill and the buried karst overlain by clastic rocks, respectively.

Fig. 2
figure 2

South–north-trending early Silurian geological section I-I’ in the Halahatang area of northern Tarim Basin

Previous studies have shown that similar to the Tahe oil field (Zhang et al. 2012; Liu et al. 2013), the Halahatang area experienced three periods of depositional hiatus: between the end of Yijianfang Formation deposition and the beginning of the Tumuxiuke Formation, between the end of the Lianglitage Formation and the beginning of the Sangtamu Formation, and between the end of the Sangtamu Formation and the beginning of the Silurian Formation (Fig. 2).

In the study area, the upper Ordovician Sangtamu Formation (O3 s) is composed mainly of mudstone. The Lianglitage Formation (O3 l) is composed mainly of marlstone and nodular limestone interbedded with calcarenite, which can be divided into the Liang first section (O3 l 1), Liang second section (O3 l 2), and Liang third section (O3 l 3) from top to bottom. The Tumuxiuke Formation (O3 t) is dominated by taupe sparry limestone and marlstone. The middle Ordovician Yijianfang Formation (O2 yj) is composed mainly of taupe sparry calcarenite, bioclastic limestone, and oolitic limestone (Fig. 2a, b). The middle to lower Ordovician Yingshan Formation (O1−2 y) can be divided into four sections from top to bottom: middle Ordovician Tongying first and second sections (O1 y 1+2), which are composed of thick gray micrite interbedded with sparry calcarenite, and lower Ordovician Tongying third and fourth sections (O1 y 3+4), which are composed of limestone interbedded with dolomitic limestone. The lower Ordovician Penglai Formation (O1 p) is composed mainly of dolomite, although few wells are drilled to this stratum. The Yijianfang and Yingshan formations consist of strongly developed karst strata and well-developed thick limestone and serve as the main reservoirs of oil and gas in this area (Fig. 3).

Fig. 3
figure 3

Petrological characteristics of Ordovician carbonatite rocks from Halahatang. a Well T9, O3 l, microcrystalline silt bioclastic limestone, 10 × 2.5 light transmission (-), transverse diameter of the visual field 5.12 mm; b Well T601, O3 t, microcrystalline bioclastic limestone, 10 × 10 light transmission (-), transverse diameter of the visual field 1.27 mm; (c) Well T10, O2 yj, sparry bioclastic limestone, 10 × 2.5 light transmission (-), transverse diameter of the visual field 5.12 mm; d Well T7, O2 yj, sparry arenitic–oolitic limestone, 10 × 2.5 light transmission (-), transverse diameter of the visual field 5.12 mm; e Well T15, O1−2 y, sparry arenitic limestone, 10 × 2.5 light transmission (-), transverse diameter of the visual field 5.12 mm; f Well TQ1, O1 p, medium-grain dolostone, 10 × 2.5 light transmission (-), transverse diameter of the visual field 5.12 mm

Samples and experiments

The samples were collected mainly from the carbonate strata of the Ordovician Yingshan (O1−2 y) and Yjianfang (O2 yj) formations in the Halahatang area. These samples include calcites filling karst fissures and pores of the Ordovician carbonatites, which either partly or fully fill the fissures and holes of the paleokarst. The sampling depth ranged from 6685.8 to 6837.6 m, which is 0.5–155 m beneath the Ordovician paleo-weathered crust. From 10 wells, 32 samples were collected for carbon and oxygen isotopic analysis, including eight samples of infilling calcite from fissures and pores and four limestone samples. For analysis of inclusions, 15 samples were collected from which 350 calcite inclusions were examined using the THMSG600 Cooling–Heating Stage (Linkam Company, Britain). Analysis was conducted in the Zhongnan Supervision and Inspection Center of the Ministry of Land and Resources of the People’s Republic of China (MLR), which measures temperatures ranging from −190 to 600 °C and has a reproduction error <2 °C for the homogenization temperature and <0.2 °C for the freezing point.

In the inclusion tests, the temperature was increased at varying speeds. The heating rate was increased by 3 °C/min in the lower temperatures <31 °C but was decreased to 0.2–0.5 °C/min when the temperature was close to the initial melting point, freezing point, and clathrate melting temperature of CO2. The heating rate generally ranged from 10 to 15 °C/min at medium or high temperatures >31 °C but decreased to 1 °C/min near the phase change temperature. The salinity of the inclusions was calculated on the basis of the salinity-freezing point equation of the H2O–NaCl system (Lu and Shan 2015):

$$ W = 1. 7 8T - 0.0 4 4 2T^{ 2} + 0.000 5 5 7T^{ 3} , $$
(1)

where W is the weight percentage of NaCl, and T is the decrease of the freezing point. The salinity of the H2O–CO2–NaCl three-phase inclusions was calculated on the basis of the temperature of clathrates:

$$ W_{\text{NaCl}} = 1 5. 5 20 2 2- 1 10 2 3 4 2T - 0 10 5 2 8 6T^{ 2} , $$
(2)

where W NaCl is the mass fraction of NaCl in the aqueous solution, and T is the clathrate melting temperature (Lu and Shan 2015).

Analysis of carbon and oxygen isotopes was conducted in the Supervision and Inspection Center on Karst Geological Resources and Environment at the MLR. The carbon isotopic values were analyzed online using the combined technology of Gasbench II and MAT253. Analytical accuracy was ensured by correction to the Chinese national carbonate standards (GBW04405, GBW04406, GBW04416, and GBW04417) and was monitored using the international standard calcite material (NBS18). The standard deviations of δ13C and δ18O were 0.005 and 0.07‰, respectively. The results are expressed in terms of the Pee Dee Belemnite (PDB) standard. During analysis, the temperature and humidity were 25 °C and 60%, respectively.

Results and analysis

Geochemical characteristics of carbon and oxygen isotopes

The abundance of 13C and 18O in marine carbonates is affected mainly by sea-level change, the origin and burial rate of organic carbon, and the oxidation–reduction conditions of the sedimentation–diagenetic environment (Zheng and Chen 2000; Zhang et al. 2011). Therefore, the composition of carbon and oxygen isotopes in the carbonatites varies with the sedimentation–diagenetic environment and the properties of diagenetic fluids. The carbon and oxygen isotopic results of the 32 calcite infillings of the Ordovician carbonatite from the Halahatang area are shown in Table 1.

Table 1 Carbon and oxygen isotopic compositions of the bedrock and calcites filling the karst fissures and pores

The large number of results indicates that the isotopic values of modern marine inorganic carbonatites are close to 0‰ for both δ13C and δ18O values (Liu et al. 2004; Gu 2000). However, the results indicate a broad range of δ13C and δ18O values of the fillings in the paleokarst fissures and pores of Ordovician carbonatite in the Halahatang area (Table 1). δ13C varied from 2.12 to −4.09‰ with an average of −0.65‰, and δ18O varied from −4.14 to −15.71‰ with an average of −9.95‰. In contrast, Gu (2000) reported that the average initial δ18O value of marine carbonatite from Devonian to Cambrian ranges from −4 to −5‰. Liu et al. (2008) found that the δ13C value of microcrystalline limestone in the Tahe oil field ranges from −1.697 to 0.921‰ with an average of −0.45‰ and that the δ18O value ranges from −8.64 to −4.153‰. Compared with these previous findings, the present results differ significantly with a considerably lower δ18O value.

However, the results of the present study are similar to those reported for the western Lungu area, in which δ13C ranges from −0.70 to −6.50‰ with an average of −1.76 and δ18O ranges from −3.75 to −17.10‰ with an average of −9.42‰ (Liu et al. 2004). Moreover, the Tahe oil field average δ13C is −0.807‰, and the average δ18O is −9.14‰, as derived from the analysis of 21 data points (Liu et al. 2008). The present results differ slightly from those reported for the eastern Lungu area in which the average δ13C is −1.40‰ and the average δ18O is −10.74‰, as derived from the analysis of 25 data points (Zhang et al. 2015). The results of the present study also differ from those reported for Well TZ12 in the central Tarim area in which the average δ13C is −1.08‰ and the average δ18O is −7.05‰, as derived from the analysis of four sampling sites. Only a few sampling sites show the isotopic values drifting toward the two ends.

The isotopic characteristics of carbon and oxygen in the minerals filling pores and fissures reflect the isotopic abundance of secondary minerals. The δ13C and δ18O values of calcites showed significantly different distributions in this study. The δ18O values were generally negative with an average of −9.95‰, indicating a paleokarstification changing from shallow burial to deep burial and later effects of corrosion and filling. The δ13C values can be either positive or negative, ranging from −1.697 to 0.921‰, which verifies the complexity of paleokarstification.

Characteristics of inclusions

The greatest difference between the exposed environment and the burial diagenetic environment of carbonatite is the variations in the properties of the pore water (Goldstein 2001). Precipitation of cement indicates diagenesis of carbonatites, and the saline inclusions trapped within the cements have a close relationship with the diagenetic environment. The homogenization temperature serves as an important indicator for distinguishing diagenetic stages and fluid activity. The salinity varies with sedimentary environments; thus, the salinity of saline inclusions provides information on the chemical evolution of the diagenetic fluids (Li et al. 1996; Andrew et al. 2000; Lu and Shan 2015).

Previous studies have identified various types of inclusions, such as hydrocarbon and water inclusions. Hydrocarbon inclusions include single-phase gaseous hydrocarbon, single-phase liquid hydrocarbon, and two-phase gaseous and liquid hydrocarbon inclusions. Water inclusions include single-phase saline, two-phase gaseous and liquid saline, and three-phase hydrocarbon-containing saline inclusions. The influence of hydrocarbons on hydrocarbon inclusions or hydrocarbon-containing inclusions might result in estimation errors of temperature and salinity during the formation of the fillings. Therefore, this study attempts to discuss the diagenetic environment of the fillings using single-phase saline inclusions and two-phase gaseous and liquid saline inclusions (Table 2).

Table 2 Results of two-phase gaseous and liquid inclusions

Physical characteristics of inclusions (Fig. 4a–i)

The shapes of the single-phase saline inclusions are generally round, square, oval, polygonal, or irregular. These inclusions are distributed in groups in a belt-like pattern or evenly in calcites. However, some inclusions are accompanied by hydrocarbon inclusions distributed in a line or in a belt along cemented microfissures in calcites. A no-bubble state in single-phase liquid inclusions is a quasi-stable phenomenon that in most instances indicates low-trapping temperatures (Chi and Lu 2008; Zajacz et al. 2011). These inclusions appear to be transparent colorless or light brown in plane light without fluorescence. They have a general size of 3–40 μm with some reaching 85 μm. Single-phase saline inclusions account for more than 60% of all inclusions.

Fig. 4
figure 4

Shapes and distribution characteristics of fluid inclusions in calcites. a Fluid inclusions distributed in small groups; rice-shaped, polygon, irregular; size 2–30 μm; b fluid inclusions distributed in small groups; rice-shaped, polygonal, rectangular, oval, irregular; size 2–30 μm; c fluid inclusions distributed throughout calcite, in small groups or in string of beads; rice-shaped, oval, polygonal, and rectangular; size 2–20 μm; d fluid inclusions distributed throughout calcite, in small groups or in string of beads, with slight directionality; rice-shaped, oval, polygon, and partly automorphic negative form; size 2–25 μm; e fluid inclusions distributed throughout calcite, in groups or in lines; rice-shaped, small rhombus, oval, polygonal, rectangular; size 2–20 μm; f fluid inclusions distributed throughout calcite, in groups or in lines; rice-shaped, oval, polygonal and rhombic; size 2–15 μm; g fluid inclusions distributed throughout calcite or in small groups; rice-shaped, rectangular, and polygonal; size 2–40 μm; h fluid inclusions distributed throughout calcite, in small groups or in lines; rice-shaped, oval, polygonal and rectangular; size 2–7 μm; i fluid inclusions distributed throughout calcite, in small groups or in lines; rice-shaped, oval, polygonal and rhombic; size 1–10 μm

The shapes of the two-phase gaseous and liquid saline inclusions are generally round, polygonal, or rectangular; some are irregular in a few instances. These inclusions are distributed in groups in a belt-like pattern or evenly in calcites, although a few are accompanied by hydrocarbon inclusions and are distributed in a line or belt along the cemented microfissures in the calcites. These inclusions appear to be transparent, colorless, or light brown in plane light and have no fluorescence; their general size is 2–50 μm.

Chemical characteristics of two-phase gaseous and liquid saline inclusions

The salinity, freezing point, and homogenization temperature of the two-phase gaseous and liquid inclusions are shown is Table 2. The crossplot of salinity and homogenization temperature (Fig. 5) shows that the salinity of the inclusions in the Halahatang area varies greatly. The salt mass fraction of NaCl ranges from 0.53 to 22.24%, indicating significant variations of karst fluid properties in this area. Overall, the inclusions in the Halahatang area can be divided into three types in terms of the salinity. Type I is low-salinity inclusion with a salinity mass fraction of NaCl ranging from 0.53 to 8.68%; this type is represented by samples 1-2, 4, 5, 8, 9-1, and 12. Type II is medium-salinity inclusion with a salt mass fraction of NaCl ranging from 8 to 18%; this type is represented by samples 7, 9-2, 10, 11, and 13. Type III is high-salinity inclusion with a salt mass fraction of NaCl ranging from 15 to 22.24%; this type is represented by Samples 1-1, 2, 3, and 6.

Fig. 5
figure 5

Cross-diagram of salinity and homogenization temperature for inclusions

Discussion

Environmental and geological significance of carbon and oxygen isotopes

The δ18O value of the bedrock and calcites filling the paleokarstic fissures and pores in the Halahatang area is generally negative, whereas the δ13C values fluctuate over a broad range. This can be attributed to the meteoric water eluviation of δ12C and to atmospheric CO32− sources of δ16O in an overall open environment resulting from the Ordovician uplift (Gasparrini et al. 2006; Azmy et al. 2009).

The crossplot of δ13C–δ18O (Fig. 6) indicates that the paleoenvironment can be categorized into four classes in terms of the calcites filling the karstic fissures and pores in the Halahatang Ordovician carbonatite (Table 3).

Fig. 6
figure 6

Crossplot of carbon and oxygen isotopes for filling materials in the Ordovician karst cave system

Table 3 Response characteristics of karst development stage and filling environment

Class I is a marine environment during the syndiagenetic stage, which reflects the isotope characteristics of the early paleokarstification environment filling. δ13C of the karst cave calcite ranges from 0.81 to 2.12‰, and δ18O ranges from −7.16 to −4.14‰. The burial depth, temperature, leaching, and corrosion of the sediments can all affect δ13C and δ18O values. The δ13C and δ18O values increase with an increase in salinity, and δ18O decreases with an increase in temperature. The δ18O values tend to be more negative with increasing temperature and burial depth (McCaig et al. 2000; Zhang et al. 2008). δ13C is affected mostly by paleo-salinity and less by temperature, indicating the existence of a marine environment. The δ13C values of limestone samples in this study are also within this range, suggesting a marine paleoenvironment.

Class II is weathered crust exposed in a karst freshwater environment. The δ13C values in the filling calcite range from −4.09 to −3.12, and the δ18O values range from −15.17 to −10.25‰. Owing to the influence of atmospheric freshwater (Gasparrini et al. 2006), the δ13C values, which were formed in an open environment, were apparently smaller than the bedrock carbon isotopic values. This environment lies above the water table; therefore, the corrosion-caused fissures and pores were filled with water and air, where water infiltrated and flowed only vertically. The δ18O value of the diagenetic fabric is determined by temperature, isotopic composition of bedrock, and meteoric water rich in 12C but poor in δ18O. Therefore, the δ13C value drifted negatively, and δ18O was strongly negative (Chen 1994; Shields et al. 2003).

Class III is a shallow-burial karst environment. The δ13C values in the karst cave filling calcite are between −1.92 and −1.59‰, and the δ18O values range from −10.19 to −6.34‰. δ18O shows obvious negative bias, although δ13C shows a slightly negative bias. Only one sample showed a positive δ13C value (PDB) of 1.59‰, which might be attributed to the decomposition and methanation (CH4) of organic matter caused by methane bacteria. The CH4 formed from biochemical processes is rich in δ12C, whereas the CO2 has abundant δ13C. Because CO2 released during the reaction of methane bacteria fermentation is rich in δ13C, the sedimentary filling has more positive values (Huang et al. 2008).

Class IV is a deep-burial high-temperature environment. δ13C in calcite is between −2.01 and 0.34‰, and the δ18O values range from −15.71 to −11.33‰. The δ18O value shows a remarkably negative bias, reflecting a significant relationship between the formation of calcite filling and hydrothermal activity. Because of the influence of heat loss in the transformation of deep δ18O in the burial stage (Lavoie and Chi 2006; Liu et al. 2012), the δ18O values are significantly more negative than those of other classes.

Homogenization temperature and burial history analysis of the inclusions

A fluid inclusion is a closed micro-liquid or gaseous system trapped in mineral crystal. This fluid records information on the geological history of the rock and a relatively closed geochemical system; different periods of fluid inclusion filling in karst fissures and pores reflect different stages of mineralization (Nedkvite et al. 1993; Parnell et al. 2001).

The present study showed that the temperature of the two-phase gaseous and liquid saline inclusions is between 59 and 140 °C (Table 2). The Ordovician inclusions from different wells or from the same wells but at different depths showed different temperatures. For example, Inclusion 1-2 and Inclusion 2 from Sample 1 have maximum temperatures of 119 and 140 °C, respectively. Moreover, the gaseous and liquid inclusions from the same sample from the same well were different. For example, the homogenization temperatures of gaseous and liquid Inclusion 9-1 and Inclusion 9-2 from Sample 9 range from 59 to 83 °C and from 92 to 122 °C, respectively. These differences indicate that the Ordovician carbonatites in the Halahatang area experienced multi-stage diagenesis and modification at different temperatures.

The distribution frequency of the homogenization temperature divides the temperatures into three ranges of inclusion formation (Fig. 7). The first is 66–80 °C, during which time the greatest number of inclusions formed between 70 and 75 °C. The second is 81–95 °C, during which time the greatest number of inclusions formed between 86 and 90 °C. The third is 96–120 °C, during which time fewer inclusions formed than those during the two prior divisions. However, in the temperature range of 96–120 °C, the number of inclusions increased dramatically in the temperature range of 101–105 and 116–120 °C, indicating the existence of a weak filling stage. In this study, only a few inclusions showed temperatures >120 °C, with a maximum temperature of 14 °C.

Fig. 7
figure 7

Distribution of homogenization temperature for two-phase saline inclusions

However, the maximum homogenization temperature of the two-phase gaseous and liquid saline inclusions in the Tahe oil field in the eastern Halahatang area can reach 200 °C (He et al. 2010; Li 2013), suggesting the possibility of abnormally high-temperature fluids in this area. However, these fluids exerted only very limited influence.

The formation of widely distributed single-phase liquid inclusions may be affected by a number of factors (Ford 1995; Xiao et al. 2002; Loucks et al. 2004). Abundant single-phase liquid inclusions are distributed throughout the calcites filling paleokarst fissures and pores in the study area. As a result, it is possible that a portion of the inclusions was formed in a low-temperature environment in which the temperature was lower than 50 °C. These low-temperature single-phase liquid inclusions resulted from exposure and paleokarstification of Ordovician carbonatites by meteoric water during the Caledonian.

The paleo-geothermal gradient of the study area is 25 °C/km, which is lower than the current gradient (Wang et al. 2013). Given a surface temperature of 15 °C, the temperatures of inclusion formation were calculated and projected onto a burial history map, as shown in Fig. 8. Four large-scale paleokarstification events and filling periods were identified, including karstification by meteoric water during the Caledonian, burial karstification during the late Hercynian, burial karstification during the Indosinian–Yanshanian, and burial karstification during the Himalayan. These events are consistent with the carbon and oxygen isotopic results of the present study.

Fig. 8
figure 8

Filling history for filled calcite in the Ordovician karst cave system in the Halahatang area

Conclusions

The Ordovician carbonatites in the Halahatang area experienced three depositional hiatuses, i.e., pinching out of the strata from south to north, development of pure and thick limestone in the Yijianfang and Yingshan formations, and development of paleokarst. The carbon and oxygen isotopic analysis shows that the δ13C value varies widely from 2.12 to −4.09‰, with an average of −0.65‰; however, the δ18O values show obvious negative bias, ranging from −4.14 to −15.71‰ with an average of −9.95‰. These results indicate four classes of paleoenvironment of carbonatite karstification: a syndiagenetic marine environment, weathered crust exposed in a karst freshwater environment, a shallow-burial karst environment, and a deep-burial high-temperature environment.

The saline inclusions of calcites in the fissures and pores of the Ordovician carbonatite paleokarst can be divided into three types: low-salinity, medium-salinity, and high-salinity inclusions. Analysis of homogenization temperature indicates that the inclusions were formed mainly in three temperature ranges. These results are consistent with the carbon and oxygen isotopes and reveal four large-scale paleokarstification events and filling periods with burial karstification during the late Hercynian, the Indosinian-Yanshanian, and the Himalaya in addition to karstification by meteoric water during the Caledonian.