Skip to content
Frontiers
Search

Original Research

Frontiers in Earth Science ·

Volume 14 - 2026 | https://doi.org/10.3389/feart.2026.1930533

Present-day thermal regime of the Xihu Sag, East China Sea Shelf Basin: constraints from an updated well database and logging-derived thermal conductivity

  • Hui Diao1,2
  • Xiaobin Shi3,4
  • Xiong Cheng5,6,7*corresponding author
  • Zhongkun Yu2
  • Peng Zhao3,4
  • Jingqi Xu2
  • Fei Ding2
  • Qing Yu2
  1. 1State Key Laboratory of Marine Geology, Tongji University, Shanghai, China
  2. 2Shanghai Branch Company, China National Offshore Oil Corporation (CNOOC) Ltd., Shanghai, China
  3. 3State Key Laboratory of Tropical Oceanography, South China Sea Institute of Oceanology, Chinese Academy of Sciences, Guangzhou, China
  4. 4Laboratory of Ocean and Marginal Sea Geology, South China Sea Institute of Oceanology, Chinese Academy of Sciences, Guangzhou, China
  5. 5School of Energy Resources, China University of Geosciences, Beijing, China
  6. 6Beijing Key Laboratory of Unconventional Natural Gas Geological Evaluation and Development Engineering, Beijing, China
  7. 7Key Laboratory of Marine Reservoir Evolution and Hydrocarbon Accumulation Mechanism, Ministry of Education, Beijing, China
Article metrics0 viewsView details
PDFCiteMetrics

Abstract

Continental margin rift basins are characterized by highly variable present-day heat flow reflecting differential lithospheric extension. The East China Sea Shelf Basin, situated at the convergent boundary of multiple plates, represents a unique tectonic setting where Cenozoic rifting has produced an anomalously high thermal state that remains poorly documented. This study presents a logging-based workflow to predict thermal conductivity and compiles an updated geothermal database for the Xihu Sag of the East China Sea Shelf Basin. We compile temperature measurements from 102 wells and thermal conductivity data from 21 new core samples, supplemented by logging-predicted thermal conductivity from 62 wells using a back-propagation (BP) neural network. Bottom-hole temperatures (BHTs) were corrected using the Waples method and validated against drill-stem test (DST) temperatures. Geothermal gradients range from 25.7 °C to 39.8 °C/km (mean 34.0 °C ± 3.5 °C/km), increasing eastward from the Western Slope Belt (31.4 °C/km) to the Central Anticline Belt (38.0 °C/km). Heat flow values range from 44 to 82 mW/m 2 (mean 63 ± 8 mW/m 2), reaching 73 mW/m 2 in the eastern anticline zone. The observed eastward-increasing thermal gradient is dually controlled by deep lithospheric thinning and basement morphology-induced thermal refraction. This study provides a comprehensive updated geothermal database for the Xihu Sag and offers a reproducible workflow for analogous rift basins.

Keywords

1 Introduction

The Xihu Sag in the East China Sea Shelf Basin has emerged as a hotspot for offshore oil and gas exploration in China. With the discovery of multiple large-scale, thick-bedded, low-permeability to tight sandstone gas fields, the deep low-permeability to ultra-low permeability gas reservoir domain in the sag has stimulated renewed exploration interest (Zhang et al., 2023; Liu et al., 2024; Zhang et al., 2025). However, key issues, including the evolution of hydrocarbon generation from deep source rocks and the chronological relationship between reservoir densification and hydrocarbon charging remain poorly understood, significantly constraining exploration progress. These processes are intimately linked to the geothermal field.

Geothermal field studies in sedimentary basins provide a fundamental basis for petroleum resource assessment and hydrocarbon accumulation process research (Tang et al., 2014; Wang, 2015; Tang et al., 2016). Heat flow is a crucial parameter for understanding the present-day geothermal field of sedimentary basins and for reconstructing basin tectono-thermal evolution, thermal history of source rocks, and hydrocarbon generation history (Wang, 2015). Together with rock thermal properties, it forms the foundational data for geothermal research (Forster, 2001). The heat flow data is mainly calculated based on temperature measurement data from wells and their rock thermal conductivity (Liu et al., 2021; Xu et al., 2022; Zhu et al., 2022b). Previous researchers have carried out pioneering work on heat flow, rock thermal conductivity, and the present-day thermal state analysis of the Xihu Sag (Bi et al., 1994; Xu et al., 1995; Luan et al., 2002; Luan and Zhang, 2003; Yang et al., 2004), obtaining heat flow data from 15 wells and thermal conductivity data from 104 core samples (Xu et al., 1995; Yang et al., 2004). These data have provided an important foundation for studying the present-day thermal state and thermal history of the Xihu Sag. Nevertheless, the existing heat flow data remain relatively sparse, making it difficult to fully characterize the present-day thermal state (Yao et al., 2026).

Continental margin rift basins worldwide, including the North Sea, and South China Sea, exhibit highly variable present-day heat flow controlled by the degree of lithospheric extension and post-rift thermal subsidence (McKenzie, 1978; Allen and Allen, 2013). However, the East China Sea Shelf Basin, located at the convergent boundary of the Eurasian, Pacific, and Philippine Sea plates, represents a unique tectonic setting where multiple phases of Cenozoic rifting have produced an anomalously high thermal state that is still poorly documented. Previous modeling suggests mantle heat flow contributions in the Xihu Sag reach 73%–79% (Zhou et al., 2017), yet systematic heat flow mapping based on dense well networks has been lacking.

After more than 40 years of oil and gas exploration, the Xihu Sag has accumulated temperature measurement data from over 100 wells (Shen et al., 2020; Yu et al., 2023) and core samples suitable for measuring thermal conductivity and heat production rate. However, the distribution of cores from oil and gas wells is limited, making it difficult to reflect the variation of rock thermal conductivity with depth throughout the entire well. Previous studies have shown that rock thermal conductivity is mainly related to lithology, porosity, and rock fabric (Chen et al., 2020; Zhu et al., 2022a; Suft et al., 2025). Well logging data from oil and gas wells can reflect these rock characteristics and can be used to calculate the rock thermal conductivity varying with depth in the well. For example, (Goutorbe et al., 2006) used five types of logging response data (sonic transit time, density, compensated neutron, resistivity, natural gamma) and neural network technology to construct a numerical model to predict rock thermal conductivity at different depths in wells, and successfully applied it to geothermal studies of basins in the North Sea and the Australian continental margin. (Jiang et al., 2012) used this model to calculate thermal conductivity data from Ocean Drilling Program (ODP) wells in the South China Sea, obtaining prediction results that were in good agreement with measured thermal conductivity data, indicating that predicting thermal conductivity from well logging data is feasible. Therefore, in addition to collecting samples for thermal property measurements, well logging data accumulated over many years of exploration in the Xihu Sag can also be used to obtain more thermal conductivity data required for calculating heat flow values and basin thermal history.

This study aims to (1) establish a logging-based thermal conductivity prediction workflow applicable to the Xihu Sag and validate its accuracy against measured core data; (2) update the geothermal gradient and heat flow database using 102 wells with corrected temperature data; and (3) clarify the distribution characteristics and main controlling factors of the present-day geothermal field. The resulting dataset and methodology are intended to serve as a reproducible template for thermal characterization in analogous rift basins.

2 Geological background

The Xihu Sag, located in the northeastern part of the East China Sea Shelf Basin, is the largest, thickest, and most prospective hydrocarbon-rich sag in this basin (Zhou et al., 2019). The sag trends NNE, with a length of approximately 440 km, a width of approximately 110 km, and an area of approximately 5.9 × 104 km2. Influenced by multiple tectonic movements, the Xihu Sag has formed a structural pattern of “east-west zonation and north-south compartmentalization' (Figure 1). Tectonically, it can be divided from west to east into three first-order structural units: the Western Slope Belt, the Central Anticline Belt, and the Eastern Fault Terrace Belt (Figure 1).

Location of the East China Sea Shelf Basin **(a)**, structural units of the East China Sea Shelf Basin and the location of the Xihu Sag **(b)**, and structural elements of the Xihu Sag **(c)**. In **(c)**, dots indicate well locations, ①–⑨ refer to the Hangzhou slope belt, Pinghu slope belt, Tiantai slope belt, Jiaxing anticline belt, Ningbo anticline belt, Tiantai anticline belt, Zhenhai fault terrace belt, Huangyan fault terrace belt, and Jilong structural belt.
Figure 1. Location of the East China Sea Shelf Basin (a), structural units of the East China Sea Shelf Basin and the location of the Xihu Sag (b), and structural elements of the Xihu Sag (c). In (c), dots indicate well locations, ①–⑨ refer to the Hangzhou slope belt, Pinghu slope belt, Tiantai slope belt, Jiaxing anticline belt, Ningbo anticline belt, Tiantai anticline belt, Zhenhai fault terrace belt, Huangyan fault terrace belt, and Jilong structural belt.

Influenced by the subduction of the Pacific Plate beneath the Eurasian Plate, the eastward flow of mantle material caused by the collision between the Indian Plate and the Eurasian Plate, and deep Earth dynamics (Lee et al., 2006; Zhang et al., 2014), the Xihu Sag experienced multiple tectonic movements during the Cenozoic, with a complex dynamic mechanism. Since the Paleogene, the tectonic evolution of the Xihu Sag can be divided into three stages: the Paleocene–Eocene rifting stage, the Oligocene–Miocene depression and inversion stage, and the Pliocene–Quaternary regional subsidence stage (Zhang et al., 2012; Zhu et al., 2019; Liu et al., 2022; Yu et al., 2023).

Thick Cenozoic sediments developed upon the Late Cretaceous tectonic framework. The Cenozoic clastic sediment thickness is 9,000–15,000 m (Zheng et al., 2024). The Cenozoic strata from bottom to top include the Eocene Bajiaoting Formation, Baoshi Formation, and Pinghu Formation; the Oligocene Huagang Formation; the Miocene Longjing Formation, Yuquan Formation, and Liulang Formation; the Pliocene Santan Formation; and the Quaternary Donghai Group (Zhou et al., 2019) (Figure 2).

Generalized stratigraphic column of the Xihu Sag, revised from Yu et al. (2023) and Xu et al. (2024).
Figure 2. Generalized stratigraphic column of the Xihu Sag, revised from Yu et al. (2023) and Xu et al. (2024).

3 Thermal conductivity

3.1 Core sample collection and measurement

In this study, 21 core samples from 11 wells were collected (Table 1). Thermal conductivity was measured using a Hot Disk 2500 S thermal constants analyzer (Hot Disk AB, Sweden). The dry thermal conductivity of each sample was measured first, followed by saturated thermal conductivity after immersing the samples in water for seven days. Each sample was measured three times under both dry and saturated conditions, and the final result was set to the arithmetic mean. Measurement results are summarized in Table 1.

TABLE 1. Sample information and measurement results.

No.WellSampleDepth (m)StratumLithologyMeasured value (dry) λ d W/(m·K)Measured value (saturated) λ w W/(m·K)Value after temperature correction λ w , T W/(m·K)Value after temperature and anisotropy correction λ w , T , a W/(m·K)Predicted value λ B P W/(m·K)Error λ w , T , a ‐ λ B P / λ w , T , a (%)
1X1HY013,534HuagangSandstone2.793.602.962.692.651.4
2X2HY023,959HuagangSandstone3.084.273.333.033.185.1
3X3HY033,723HuagangSandstone3.144.123.313.01--
4X4NB014,053HuagangSandstone2.964.443.483.163.014.8
5X4NB024,350PinghuSandstone2.833.132.622.382.463.3
6X5NB044,051PinghuSandstone2.804.443.563.242.987.9
7X5NB054,062PinghuSandstone2.664.373.513.192.928.5
8X6NB074,563PinghuSandstone2.653.662.992.72--
9X6NB084,567BaoshiSandstone3.454.913.773.43--
10X7NB093,702HuagangSandstone3.194.133.242.953.032.8
11X7NB103,816HuagangSandstone3.294.253.303.002.6112.9
12X7NB114,015HuagangSandstone3.364.123.182.892.962.4
13X8TT013,677HuagangSandstone3.614.983.803.462.9414.9
14X8TT023,950PinghuSandstone2.783.773.012.732.4111.9
15X8TT034,070PinghuSandstone3.224.433.383.082.963.8
16X9TT043,671HuagangSandstone3.074.043.252.962.679.6
17X9TT054,217PinghuSandstone3.214.053.172.89--
18X9TT064,614PinghuSandstone3.394.473.363.06--
19X10603,426HuagangSandstone3.133.953.192.90--
20X11543,419HuagangSandstone2.963.713.202.91--
21X11743,421HuagangSandstone2.873.673.172.88--

The dry thermal conductivity ranges from 2.65 to 3.61 W/(mK), with an average of 3.1 ± 0.3 W/(mK). The saturated thermal conductivity ranges from 3.1 to 5.0 W/(mK), with an average of 4.1 ± 0.3 W/(mK). For the eight Pinghu Formation samples, the dry thermal conductivity ranges from 2.65 to 3.39 W/(mK), with an average of 2.94 ± 0.3 W/(mK), and the saturated thermal conductivity ranges from 3.1 to 4.5 W/(mK), with an average of 4.05 ± 0.5 W/(mK). For the 12 Huagang Formation samples, the dry thermal conductivity ranges from 2.79 to 3.61 W/(mK), with an average of 3.12 ± 0.2 W/(mK), and the saturated thermal conductivity ranges from 3.6 to 4.98 W/(mK), with an average of 4.1 ± 0.4 W/(mK). For the one Baoshi Formation sample, the dry thermal conductivity is 3.45 W/(mK) and the saturated thermal conductivity is 4.91 W/(mK). Overall, thermal conductivity increases with depth (Figure 3).

Variation of measured thermal conductivity with depth. **(a)** for dry samples. **(b)** for saturated samples.
Figure 3. Variation of measured thermal conductivity with depth. (a) for dry samples. (b) for saturated samples.

3.2 Predicted thermal conductivity from wells

Due to the limited core recovery rate in oil and gas wells, thermal property measurements based on core samples often fail to capture the full stratigraphic column. The measured thermal conductivity samples are still insufficient to represent the thermal conductivity of different formations in each well. To obtain the thermal conductivity distribution of the entire well section, this study adopts the method of predicting thermal conductivity from well logging data based on a BP neural network (Goutorbe et al., 2006; Jiang et al., 2012). The input parameters of the model include five types of logging data: sonic transit time, density, compensated neutron, resistivity, and natural gamma. The reliability of this model has been verified in ODP wells in the South China Sea, with an average prediction error for thermal conductivity lower than the maximum allowable error limit of 15% for rock thermal conductivity measurements (Jiang et al., 2012).

Here, this thermal conductivity prediction model was applied to obtain predictions for 62 wells, with representative results shown in Figure 4. To analyze the reliability of the predicted thermal conductivity, the measured values (including those newly obtained in this study and collected from the literature) were first subjected to in-situ correction. The in-situ correction includes saturation correction, in-situ temperature correction, and thermal conductivity anisotropy correction.

Comparison between predicted and measured thermal conductivity in sampling wells. **(a–g)** represent wells X1, X2, X4, X5, X7, X8, and X9, respectively. The green line is the predicted thermal conductivity (λ<sub>BP</sub>), the red line is temperature, and the black, blue, and red dots represent the measured saturated thermal conductivity (λ<sub>w</sub>), temperature-corrected thermal conductivity (λ<sub>w,T</sub>), and temperature- and anisotropy-corrected thermal conductivity (λ<sub>w,T,a</sub>), respectively.
Figure 4. Comparison between predicted and measured thermal conductivity in sampling wells. (a–g) represent wells X1, X2, X4, X5, X7, X8, and X9, respectively. The green line is the predicted thermal conductivity (λBP), the red line is temperature, and the black, blue, and red dots represent the measured saturated thermal conductivity (λw), temperature-corrected thermal conductivity (λw,T), and temperature- and anisotropy-corrected thermal conductivity (λw,T,a), respectively.

Since the newly obtained samples had already been measured for thermal conductivity under saturated conditions, saturation correction was not required. The temperature correction used the correction formula of Sekiguchi (1984):where, λ T is the thermal conductivity at the in-situ temperature (T), in W/(mK); λ m and T m are 1.8418 W/(mK) and 1473 K, respectively (Sekiguchi, 1984); λ 0 is the thermal conductivity measured at laboratory temperature T 0.

λ T = λ m + T 0 T m T m ‐ T 0 × λ 0 ‐ λ m × 1 T ‐ 1 T m

The samples collected in this study were cylindrical cores drilled parallel to the bedding direction (i.e., perpendicular to the heat flow direction). The measured thermal conductivity is the thermal conductivity parallel to the bedding plane (λ ∥), while the thermal conductivity used for calculating heat flow is that perpendicular to the bedding plane (λ ⊥). Consequently, anisotropy correction for thermal conductivity is required (Wang, 2015). The correction coefficient is defined as a = λ ∥/λ ⊥. For sandstone, the degree of anisotropy is relatively small, with reported correction coefficients typically ranging from 1.05 to 1.15 depending on compaction degree and clay content (Wang, 2015; Suft et al., 2025). A uniform correction coefficient of 1.1 was applied to all sandstone samples, representing the median value for moderately compacted sandstone at depths of 3,000–5,000 m (Wang, 2015). The measured thermal conductivities were then corrected for anisotropy as λ w , T , a = λ w/1.1 (see Table 1).

Altogether, thermal conductivity was measured for 21 samples from 11 wells, of which 7 wells had well logging data. The measured thermal conductivities (λ d and λ w) from the sampling wells are the averages of three or more experimental measurements. The saturated thermal conductivity was subjected to in-situ temperature correction (λ w , T) and thermal conductivity anisotropy correction (λ w , T , a), as shown in Table 1. The predicted thermal conductivity (λ B P) results for the seven sampling wells generally agree well with the measured thermal conductivity (λ w , T , a), as shown in Figure 4. The prediction errors for 13 samples ranged from 1.4% to 14.9%, of which 10 samples had prediction errors below 10%. The overall average prediction error, defined as λ w , T , a ‐ λ B P / λ w , T , a was approximately 6.9% (see Table 1), which is significantly lower than the maximum allowable error of 15% for rock thermal conductivity measurements, indicating that the thermal conductivity prediction model is applicable to the Xihu Sag study area.

3.3 Average thermal conductivity of well strata

Using the well log-derived thermal conductivity, combined with the stratigraphic division data of each well, the average thermal conductivity of each stratigraphic group can be calculated. Unfortunately, the well logging data for the wells is incomplete (data are missing at the top of some wells). For those formations with logging data, only partial sections have logging data for the uppermost and lowermost formations. Among these formations, only one well has logging data for the Santan Formation, with an average thermal conductivity of 1.5 W/(mK). For the Liulang Formation, there are only three data points, ranging from 1.65 to 1.77 W/(mK), with an average thermal conductivity λ(±σ) of 1.71 ± 0.06 W/(mK). The Yuquan Formation has 18 data points, ranging from 1.62 to 2.29 W/(mK), with an average of 1.92 ± 0.22 W/(mK). The Longjing Formation has 43 data points, ranging from 1.76 to 2.48 W/(mK), with an average of 2.07 ± 0.17 W/(mK). The Huagang Formation has 57 data points, ranging from 1.92 to 2.98 W/(mK), with an average of 2.28 ± 0.20 W/(mK). The Pinghu Formation has 44 data points, ranging from 2.08 to 3.0 W/(mK), with an average of 2.3 ± 0.20 W/(mK). The Baoshi Formation has 17 data points, ranging from 2.02 to 2.78 W/(mK), with an average of 2.29 ± 0.19 W/(mK). The frequency distribution for each stratigraphic group is shown in Figure 5, and the average thermal conductivity of each group generally shows an increasing trend with depth.

Histogram of average thermal conductivity distribution for each stratigraphic group. n is the number of statistical data points, λ and σ are the average thermal conductivity and standard deviation, respectively.
Figure 5. Histogram of average thermal conductivity distribution for each stratigraphic group. n is the number of statistical data points, λ and σ are the average thermal conductivity and standard deviation, respectively.

3.4 Average thermal conductivity of wells

To calculate the heat flow of a well, it is necessary to calculate the average thermal conductivity from the depth of the temperature measurement point to the seabed. The thermal conductivity varying with depth obtained from the logging data prediction model provides a reliable basis for calculating the average thermal conductivity of the well. However, the well logging data for the wells are incomplete, often missing logging data for the upper 1,500–2,500 m of the well, and for some wells, the missing depth interval without logging data is even larger. The processing method adopted in this study is as follows: For the depth interval from the seabed to 500 m below the seabed, due to the large variation in porosity and thus thermal conductivity, the average thermal conductivity of 1.28 W/(mK) is used as the average thermal conductivity for the 0–500 m interval. This value is based on the measured thermal conductivity from IODP Site 1,148 in the South China Sea (Wang et al., 2000), which represents unconsolidated to semi-consolidated marine sediments with porosities of 40%–60%. For the depth interval from 500 m below the seabed to the effective logging interval, the formula relating thermal conductivity to porosity is used. If the temperature measurement depth of the well is greater than the coverage range of the effective logging data, then the average thermal conductivity of the lower depth interval without logging data is set to the average of the predicted thermal conductivity from the effective logging data for the deepest 100 m (this situation is relatively rare).

Once the thermal conductivity varying with depth for the well is obtained, the average thermal conductivity of the well is calculated using the thermal resistance method (Bullard method):

λ ¯ = ∑ Δ z i / ∑ Δ z i λ i

For wells with logging data but missing a large interval in the upper part, the thermal conductivity for the missing depth interval is borrowed from the thermal conductivity of adjacent wells at the corresponding depth interval. For wells with temperature measurement data but no logging data, the average thermal conductivity is also calculated by borrowing the predicted thermal conductivity from adjacent wells.

All geothermal gradient and heat flow data generated in this study are provided in Supplementary Table S1. Well logging data used for thermal conductivity prediction are proprietary and cannot be publicly released; however, the BP neural network architecture and training parameters are described in detail below to ensure methodological reproducibility.

4 Well temperature measurement data and geothermal gradient

4.1 Temperature correction

Temperature measurement data from 111 wells were compiled in this study (see Figure 1b for well locations). Among them, 63 wells have bottom-hole temperature (BHT) data, and 72 wells have drill-stem test (DST) temperature data, with 24 wells having both BHT and DST temperature data (111 unique wells in total). After quality screening, in which nine wells were excluded because each had only a single temperature measurement point and yielded an anomalous geothermal gradient, data from the remaining 102 wells were retained for the geothermal-gradient and heat-flow calculations (Supplementary Table S1). The well temperatures range from 70 °C to 210 °C, and the measurement depths range from 1,500 to 5,200 m, with most measurement depths between 2,200 and 4,800 m.

Generally, DST temperature data can approximately represent the pre-drilling temperature of the tested formation (Forster, 2001; Goutorbe et al., 2008) and do not require further correction. However, BHT temperature data are generally lower than the true undisturbed formation temperature because the geothermal field in the well environment is disturbed by drilling and circulation, and the temperature has not reached equilibrium at the time of measurement. Therefore, correction is required. The Waples method (Waples and Ramly, 2001; Waples et al., 2004) was selected because it explicitly accounts for the time elapsed since the cessation of mud circulation, is statistically calibrated against DST temperatures, and has been widely validated in rift basins including the Pearl River Mouth Basin (Tang et al., 2016).

The main alternative approaches were considered but found to be less suitable for the present dataset. The Horner plot method (Horner, 1951) requires at least three self-consistent temperature measurements at the same depth from successive logging runs, which are rarely available for the study wells. The empirical correction of Harrison et al. (1983) is a depth-only function calibrated on data from the Anadarko and Arkoma basins of Oklahoma over depths of 600–3,932 m; it does not consider the time elapsed since the cessation of mud circulation and is not necessarily transferable to an offshore rift basin with a different lithology and thermal history. In contrast, the Waples method statistically relates the correction factor to the time since the end of mud circulation (TSC) and, to a lesser extent, to depth, using DST temperatures as ground truth (Waples and Ramly, 2001; Waples et al., 2004), which matches the parameters available for the Xihu Sag wells.

Specifically, for measurement depths shallower than 3,500 m, the correction of Waples and Ramly (2001), calibrated using wells shallower than 3,500 m in the Malay Basin, was applied:where Tc is the corrected temperature (°C), T0 is the seabed temperature (18 °C), Tm is the measured BHT (°C), and the correction factor fs is given by:with TSC the time since the end of mud circulation (h) and z the measurement depth (m). For depths of 3,500–6,500 m, the modified equations of Waples et al. (2004), calibrated against DST temperatures in deep Gulf of Mexico wells, were used:

T c = T 0 + f s T m ‐ T 0

f s = 1.699 ‐ 0.1462 ⁡ ln T S C 0.572 z 0.075

T c = T 0 + f T m ‐ T 0 ‐ 0.001391 z ‐ 4498

f = 1.32866 ‐ 0.005289 T S C

The TSC values and mud-circulation information for each well were obtained from well-log headers and drilling reports; where TSC was not recorded, it was estimated from its trend with depth following (Waples et al., 2004). The measurement depths of the study wells (1,500–5,200 m) fall within the combined calibration range of the two formulations. We note that the Waples method is a statistical correction; its inherent scatter is quantified by the validation against DST temperatures below and is incorporated into the uncertainty analysis in Section 5.

Among the wells with temperature data, 24 wells have both BHT and DST temperature data; these wells constitute an independent validation set for the temperature-correction workflow. Because the BHT and DST measurements in the same well were generally acquired at different depths, a reference geotherm was established for each well by linear regression of its DST temperatures together with the seabed temperature (18 °C), and each corrected BHT was compared with the reference temperature at the same depth (Figure 6c). For the 84 valid BHT data points from the 24 wells, the corrected BHTs correlate well with the reference temperatures (R 2 = 0.87), with a root-mean-square error of RMSE = 10.3 °C and a mean absolute error of MAE = 8.2 °C. The corrected BHTs are systematically higher than the reference temperatures by +5.0 °C on average, which is attributed to two factors: the Waples correction tends to overcorrect BHT measurements with short shut-in times (Waples and Ramly, 2001), and the true geotherm is concave because thermal conductivity increases with depth (Section 3.1), so a linear reference geotherm anchored at the seabed slightly underestimates the formation temperature at depth. At the gradient level, which is the quantity actually used in the heat-flow calculation, the gradients derived from the corrected BHTs exceed the DST-derived gradients by +1.2 °C ± 1.5 °C/km on average (RMSE = 1.9 °C/km, i.e., approximately 6% in relative terms). Because this offset is approximately uniform across the study area, it does not affect the spatial pattern of the geothermal field, which is the main focus of this study; its magnitude is incorporated into the uncertainty analysis of heat flow in Section 5.

Comparison of Waples-corrected BHT temperatures with DST temperatures. Blue crosses are measured BHT temperatures; black solid circles are DST temperatures (except the solid circle at the seabed location); green open squares are corrected results using the Waples method; error bars represent two standard deviations. The solid line is the linear regression line of the effective temperature data and the seabed temperature (18 °C). **(a)** is for Well X12. **(b)** is for Well X1. Panel **(c)** is the cross-plot of the Waples-corrected BHTs versus the DST-constrained reference temperatures at the same depths for the 24 validation wells (n = 84): the dashed line is the 1:1 reference line, the dotted lines mark ±10 °C deviations, and the validation statistics (*R* <sup>2</sup>, RMSE, and MAE) are annotated in the figure.
Figure 6. Comparison of Waples-corrected BHT temperatures with DST temperatures. Blue crosses are measured BHT temperatures; black solid circles are DST temperatures (except the solid circle at the seabed location); green open squares are corrected results using the Waples method; error bars represent two standard deviations. The solid line is the linear regression line of the effective temperature data and the seabed temperature (18 °C). (a) is for Well X12. (b) is for Well X1. Panel (c) is the cross-plot of the Waples-corrected BHTs versus the DST-constrained reference temperatures at the same depths for the 24 validation wells (n = 84): the dashed line is the 1:1 reference line, the dotted lines mark ±10 °C deviations, and the validation statistics (R 2, RMSE, and MAE) are annotated in the figure.

4.2 Geothermal gradient

The average geothermal gradient of a well is obtained by linear regression of the seabed temperature and the DST temperatures and corrected BHT temperatures of the well. The seabed temperature is set to 18 °C. The study wells are located on the middle shelf of the East China Sea at water depths of approximately 50–150 m, where regional oceanographic observations indicate that the bottom-water temperature remains within approximately 15 °C–20 °C throughout the year; the adopted value represents the mean bottom-water temperature in this depth range and is consistent with the value taken by Yu et al. (2023). To evaluate the influence of this assumption, the seabed temperature was varied between 15 °C and 21 °C in a sensitivity analysis. The resulting change in the calculated geothermal gradient is less than ±2% for wells with measurement depths greater than 3,000 m and at most approximately ±7% for the shallowest measurements (∼1,500 m); the corresponding change in heat flow is generally smaller than ±5% and does not affect the eastward-increasing pattern or the conclusions of this study. After screening and correction, a total of 102 geothermal gradients were obtained in this study. These temperature gradients range from 25.7 °C to 39.8 °C/km, with an average of 34.0 °C ± 3.5 °C/km. The geothermal gradient in the Xihu Sag significantly increases from the Western Slope Belt (31.4 °C ± 2.5 °C/km) to the western part of the Central Anticline Belt (34.2 °C ± 1.8 °C/km) and to the southern and northern parts of the Central Anticline Belt (38.0 °C ± 1.8 °C/km) (Figure 7).

Temperature-depth variation diagram for different tectonic belts. The values on the dashed lines are the geothermal gradient values calculated for those dashed lines; the seabed temperature is set to 18 °C.
Figure 7. Temperature-depth variation diagram for different tectonic belts. The values on the dashed lines are the geothermal gradient values calculated for those dashed lines; the seabed temperature is set to 18 °C.

5 Heat flow

Based on the average geothermal gradient (G) and average thermal conductivity (λ) of the wells calculated above, the average heat flow value Q of the well can be calculated using the following formula:

Q = G ¯ × λ ¯

Heat flow values were calculated for the same 102 wells (Supplementary Table S1). The heat flow in the Xihu Sag ranges from 44 to 82 mW/m2, with an average of 63 ± 8 mW/m2. The heat flow in the western slope area varies considerably, ranging from 46 to 77 mW/m2, with two distinct peaks: a lower peak at 52 mW/m2 and a higher peak at about 63 mW/m2, with an average of 61 ± 7 mW/m2. Similar to the variation of the geothermal gradient, the heat flow gradually increases eastward from the Western Slope Belt. In the western part of the Central Anticline Belt, the heat flow ranges from 56 to 75 mW/m2, with an average increasing to 67 ± 5 mW/m2. Further east, in the northern and southern parts of the Central Anticline Belt, the heat flow ranges from 65 to 85 mW/m2, with an average reaching 73 ± 5 mW/m2 (Figures 8, 9b).

Histogram of heat flow distribution in each major tectonic belt. Q is the average heat flow and one standard deviation, in mW/m<sup>2</sup>; N is the number of heat flow data points. **(a)** Xihu Sag, **(b)** Central Anticline Belt (Northern+Southern), **(c)** Western Slope Belt, **(d)** Central Anticline Belt.
Figure 8. Histogram of heat flow distribution in each major tectonic belt. Q is the average heat flow and one standard deviation, in mW/m2; N is the number of heat flow data points. (a) Xihu Sag, (b) Central Anticline Belt (Northern+Southern), (c) Western Slope Belt, (d) Central Anticline Belt.
Contour maps of geothermal gradient **(a)** and heat flow **(b)** in the Xihu Sag.
Figure 9. Contour maps of geothermal gradient (a) and heat flow (b) in the Xihu Sag.

The uncertainty of the heat flow values was estimated by standard error propagation. Because heat flow is the product of the geothermal gradient and the average thermal conductivity, its relative uncertainty is the root sum of squares of the relative uncertainties of the two input parameters. The relative uncertainty of the geothermal gradient is approximately 6%, which is directly constrained by the 24-well validation (Section 4.1): the root-mean-square difference between the gradients derived from the corrected BHTs and those derived from the DST temperatures is 1.9 °C/km, approximately 6% of the mean gradient; this term also encompasses the sensitivity to the assumed seabed temperature (±2% for a ±2 °C change; Section 4.2). The relative uncertainty of the average thermal conductivity is approximately 7%, dominated by the mean prediction error of the BP neural network model (6.9%; Table 1). The combined relative uncertainty of the heat flow values is therefore approximately 9%, corresponding to ±6 mW/m2 at the mean heat flow of 63 mW/m2, which is within the standard deviation of ±8 mW/m2 reported above.

To validate the reliability of the updated workflow, the new results were systematically compared with previously published data. Yang et al. (2004) compiled borehole temperature data reported as corrected BHT and oil-fluid temperatures from 24 wells of the East China Sea Shelf Basin, and reported basin-scale geothermal gradients of 25 °C–43 °C/km (mean of 32.7 °C/km) and heat flow values of 55–88 mW/m2 (mean of 70.6 mW/m2; 70.8 mW/m2 for the Xihu Depression wells alone), with no significant difference between the Xihu and Taibei depressions. The mean geothermal gradient (34.0 °C ± 3.5 °C/km) and heat flow (63 ± 8 mW/m2) obtained in this study for the Xihu Sag fall within these previously reported ranges, differing from the previous mean values by approximately 4% and 10%, respectively, particularly considering that the previous compilation used earlier temperature-correction procedures and partly relied on oil-fluid temperatures. At the nine wells that could be matched between the previous and present datasets (Xu et al., 1995; Yang et al., 2004), the geothermal gradients obtained in this study are systematically higher than the previous values, with an average relative deviation of approximately 17% (positive for eight of the nine wells and zero for PX1), whereas the heat flow values agree more closely, with an average absolute relative deviation of approximately 14% (the new values being equal to or lower than the previous values for all nine wells, lower by approximately 14% on average). These systematic differences are consistent with the methodological differences between the two studies: the Waples correction applied here raises the geothermal gradients by approximately 14% on average relative to uncorrected BHTs, whereas the earlier correction procedure is not specified in detail and the previous data partly relied on oil-fluid temperatures; meanwhile, the previous thermal-conductivity values are systematically higher than the present logging-based estimates, so that the gradient and conductivity differences partly compensate in the heat flow values. A well-by-well comparison is provided in Supplementary Table S2. The close agreement at the regional scale (approximately 4% in mean gradient and 10% in mean heat flow) supports the reliability of the combined BP-prediction and temperature-correction workflow adopted in this study.

6 Discussions

6.1 Present-day geothermal field characteristics of the Xihu Sag

The present-day geothermal field of the Xihu Sag is characterized by a mean heat flow of 63 ± 8 mW/m2 and a geothermal gradient of 34.0 °C ± 3.5 °C/km, exhibiting a systematic eastward-increasing trend from the Western Slope Belt to the Central Anticline Belt (Figures 8, 9). The heat flow ranges from 44 to 82 mW/m2, with the Western Slope Belt averaging 61 ± 7 mW/m2 and the northern and southern parts of the Central Anticline Belt reaching 73 ± 5 mW/m2 (Figure 8). The geothermal gradient increases correspondingly from 31.4 °C ± 2.5 °C/km in the Western Slope Belt to 38.0 °C ± 1.8 °C/km in the Central Anticline Belt (Figure 7).

The mean heat flow of the Xihu Sag is higher than that of the Bohai Bay Basin, where systematic well measurements indicate average values of approximately 59–65 mW/m2 (Lin et al., 2002; Chang, 2016) and geothermal gradients of approximately 31.6 °C/km (Chang, 2016). It is lower than that of the Pearl River Mouth Basin and the Qiongdongnan Basin in the northern South China Sea, which exhibit high background heat flow of 71.8 ± 13.6 mW/m2 and geothermal gradients reaching ∼42 °C/km (Yuan et al., 2009; Tang et al., 2014). This intermediate thermal state suggests that the Xihu Sag occupies a transitional position among the Cenozoic rift basins of eastern China.

The mean heat flow of the Xihu Sag (63 mW/m2) is close to the global continental average of ∼65 mW/m2 (Pollack et al., 1993) and falls within the range reported for classic passive-margin rift basins such as the North Sea (Allen and Allen, 2013), where heat flow values of 60–80 mW/m2 have been documented. However, the Xihu Sag differs from these passive-margin analogues in its pronounced eastward-increasing thermal gradient, a feature reflecting the specific structural configuration of an active-margin rift system.

6.2 Main controlling factors of the present-day geothermal field

The spatial distribution of the present-day geothermal field reflects the interplay of deep heat sources, basement morphology, sediment-cover thermal properties, and tectonic activity. The present-day geothermal field in the Xihu Sag exhibits a spatial distribution pattern of “high in the east and low in the west' (Figure 9), reflecting the complex interaction between deep heat sources and shallow geological structures.

Lithospheric thinning and asthenospheric upwelling play a dominant role in controlling the present-day thermal state. The Xihu Sag is located at the convergent boundary of the Eurasian Plate, the Pacific Plate, and the Philippine Sea Plate. Multiple phases of extensional rifting since the Cenozoic have led to significant lithospheric thinning and uplift of the asthenosphere–lithosphere boundary. Gao et al. (2006) showed that the lithospheric thickness in the East China Sea shelf area is significantly thinner than that in inland areas. Shen et al. (2020) calculated the crustal extension coefficient of the Xihu Sag to be 1.1–2.4 based on inversion of tectonic subsidence curves, confirming significant extensional thinning. Zhou et al. (2017) further revealed that the surface heat flow is approximately 71 mW/m2, with mantle heat flow density of 40–65 mW/m2, contributing 73%–79% to the surface heat flow. This mantle contribution exceeds the 60%–70% reported for the Bohai Bay Basin (Lin et al., 2002) and the 50%–60% for the northern South China Sea (Tang et al., 2014), reflecting more extreme lithospheric thinning than these adjacent basins. The mean heat flow of 63 ± 8 mW/m2 obtained in this study is comparable to the theoretical predictions of McKenzie (1978) for hyperextended rift basins. The average thermal lithosphere thickness is only approximately 66 km, exhibiting a “cold crust–hot mantle' structure that leads to a dominant mantle heat flow contribution, forming a high heat flow area in the Central Anticline Belt and a low heat flow area in the Western Slope Belt. While deep lithospheric thinning establishes the primary control, the observed eastward-increasing geothermal gradient is further amplified by shallow thermal refraction effects associated with basement morphology, as analyzed below.

The basement morphology exerts a significant control on the heat flow distribution pattern. He (2004) demonstrated that the bottom of the Pinghu Formation exhibits “downward truncation and upward onlap' against the underlying graben-like half-graben structure, creating the structural precondition for thermal refraction. (Shen et al., 2020) further demonstrated that the present-day geothermal gradient in the central structural belt (>34 °C/km) is higher than that in the slope belt (<32 °C/km), forming a distribution pattern negatively correlated with basement depth. (Lin et al., 2002) quantitatively showed that basement morphology can amplify geothermal gradient differences by a factor of 2–3 through thermal refraction. In the Xihu Sag, the basement depth difference between the two belts is approximately 2–4 km; given the thermal conductivity contrast between the graben fill and the basement highs, thermal refraction is expected to focus heat flow toward the structural highs of the Central Anticline Belt and thus to amplify the eastward-increasing geothermal gradient superimposed on the deep lithospheric control. A rigorous quantification of this contribution would require 2D or 3D finite-element thermal modeling constrained by basement depth and rock thermal properties, which remains a task for future work.

Figure 9a reveals several anomalously high geothermal gradients (>38 °C/km) within the Western Slope Belt, particularly in the Pinghu slope area. These local anomalies may reflect enhanced fluid circulation along fault zones or localized basement highs not resolved by the regional structural framework, warranting further investigation. Similar local thermal anomalies have been documented in the Pearl River Mouth Basin (Tang et al., 2016), where fault-controlled fluid flow creates transient thermal perturbations. Future work integrating 3D seismic fault interpretation with numerical fluid–heat coupling models is required to test these hypotheses.

7 Conclusion

This study adds 21 new measured thermal conductivity data and predicts thermal conductivity for 62 wells based on well logging data, verifying the feasibility of predicting thermal conductivity from well logs. BHT data were corrected using the Waples method, and on this basis, 102 new geothermal gradient and heat flow data were compiled. Combined with previous data, the present-day geothermal field characteristics of the Xihu Sag were systematically analyzed, and the main controlling factors affecting its distribution pattern were discussed. The main conclusions are as follows.

  • The geothermal gradient in the Xihu Sag ranges from 25.7 °C to 39.8 °C/km, with an average of 34.0 °C ± 3.5 °C/km. The geothermal gradient increases significantly from the Western Slope Belt (31.4 °C/km) to the western part of the Central Anticline Belt (34.2 °C/km), and to the southern and northern parts of the Central Anticline Belt (38.0 °C/km). The heat flow values range from 44 to 82 mW/m2, with an average of 63 ± 8 mW/m2. The heat flow gradually increases eastward from the western slope area, with the average reaching 73 mW/m2 in the northern and southern parts of the Central Anticline Belt.
  • The present-day geothermal field in the Xihu Sag exhibits a spatial pattern of “high in the east and low in the west”, which reflects the superposition of a primary and a secondary control. Deep lithospheric thinning, supported by the high mean heat flow (63 ± 8 mW/m2) and significant crustal extension, establishes the primary control; superimposed on this deep control, thermal refraction induced by basement morphology provides secondary amplification that enhances the eastward-increasing gradient.
  • The BP neural network-based thermal conductivity prediction from well logs, validated against 21 core samples with an average error of 6.9%, provides a reproducible method for basins with limited core recovery. The updated heat flow database establishes the largest systematic geothermal dataset for the Xihu Sag to date.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/feart.2026.1930533/full#supplementary-material

Supplementary material Supplementary material at the source

Supplementary material Supplementary material at the source

Footnotes

Edited by: Michael Short, Teesside University, United Kingdom

Reviewed by: Yigao Sun, Inner Mongolia University of Science and Technology, China

Hanqing Qiao, Chinese Academy of Geological Sciences, China

Statements

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This work is funded by the National Natural Science Foundation of China (No. 42302189),the research project of CNOOC (China) Co., Ltd. Shanghai Branch (The difference and favorable direction of gas reservoir accumulation in the core-wing of large-scale structure in the central anticline zone of Xihu Sag (No. SCKY-2025-JY09) and 15th Five-Year' accumulation prospect zoning of CNOOC (No. QGYQCCQH 2025-3)), and the National Natural Science Foundation of China Joint Fund for Enterprise Innovation and Development (No. U24B6002).

Data availability

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

Author contributions

HD: Formal Analysis, Methodology, Writing – review and editing, Writing – original draft, Data curation, Resources. XiS: Writing – review and editing, Formal Analysis, Investigation, Data curation. XiC: Supervision, Funding acquisition, Writing – review and editing. ZY: Validation, Writing – review and editing, Investigation, Software. PZ: Formal Analysis, Data curation, Investigation, Software, Writing – review and editing. JX: Visualization, Writing – review and editing, Supervision, Formal Analysis. FD: Supervision, Formal Analysis, Writing – review and editing, Project administration. QY: Resources, Formal Analysis, Data curation, Writing – review and editing.

Acknowledgements

We would like to thank Xianjun Tang (Shanghai Branch Company of CNOOC) and Miss Xi Li (Shanghai Branch Company of CNOOC) for helping with the thermal conductivity data of the Xihu Sag. Our heartfelt gratitude is also extended to the editors and reviewers for their scientific and linguistic revisions of the manuscript.

References

  1. Allen P. A., Allen J. R. (2013). Basin Analysis: Principles and Application to Petroleum Play Assessment. Chichester: John Wiley and Sons.
  2. Bi C. X., Yang L., Wu G. H., Zhang M. (1994). Characteristics of heat flow and its geological significance in the East China Sea area. Jiangsu Geol. 18 (1), 30 – 36.
  3. Chang J. (2016). Present-day geothermal regime of the jizhong depression in Bohai Bay Basin,East China. Chin. J. Geophysics- Chin. 59 (3), 1003 – 1016. doi:10.6038/cjg20160322
  4. Chen C., Zhu C. Q., Tang B. N., Chen T. G. (2020). Progress in the study of the influencing factors of rock thermal conductivity. Prog. Geophys. 35 (6), 2047 – 2057. doi:10.6038/pg2020EE0013
  5. Forster A. (2001). Analysis of well temperature data in the northeast German basin: continuous logs versus bottom-hole temperatures. Pet. Geosci. 7 (3), 241 – 254. doi:10.1144/petgeo.7.3.241
  6. Gao D. Z., Zhao J. H., Bo Y. L., Tang J. (2006). A study on lithosphere 3D structure in the East China Sea and adjacent regions. Chin. J. Geol. 41 (1), 10 – 26.
  7. Goutorbe B., Lucazeau F., Bonneville A. (2006). Using neural networks to predict thermal conductivity from geophysical well logs. Geophys. J. Int. 166 (1), 115 – 125. doi:10.1111/j.1365-246x.2006.02924.x
  8. Goutorbe B., Lucazeau F., Bonneville A. (2008). The thermal regime of South African continental margins. Earth Planet. Sci. Lett. 267 (1-2), 256 – 265. doi:10.1016/j.epsl.2007.11.044
  9. Harrison W. E., Luza K. V., Prater M. L., Cheung P. K. (1983). “ Geothermal resource assessment in Oklahoma,”, 83-1. Norman, OK: Oklahoma Geological Survey Special Publication, 42.
  10. He J. Q. (2004). Tectonic-Thermal History of Xihu Depression, East China Sea. Shanghai, China: Tongji University.
  11. Horner D. R. (1951). “ Pressure build-up in wells,” in Proceedings of the Third World Petroleum Congress (The Hague, The Netherlands), 503 – 521.
  12. Jiang H. Y., Shi X. B., Yang X. Q., Shi H. C. (2012). Prediction of thermal conductivity of rocks through geophysical well logs based on BP neural network. Well Logging Technol. 36 (3), 304 – 307. doi:10.16489/j.issn.1004-1338.2012.03.019
  13. Lee G. H., Kim B., Shin K. S., Sunwoo D. (2006). Geologic evolution and aspects of the petroleum geology of the northern East China Sea shelf basin. AAPG Bull. 90 (2), 237 – 260. doi:10.1306/08010505020
  14. Lin C., Zhang Y., Li S., Liu J., Tong Z., Ding X. (2002). Quantitative modelling of multiphase lithospheric stretching and deep thermal history of some tertiary rift basins in eastern China. Acta Geol. Sin. 76 (3), 324 – 330.
  15. Liu F., Wang A., Wang G., Zhang W., Liao Y., Tong J. (2021). Characteristics of terrestrial heat flow and lithospheric thermal structure in typical intermountain basin at the eastern foot of yanshan Mountain, north China. Front. Earth Sci. 9, 758605. doi:10.3389/feart.2021.758605
  16. Liu X., Ge J. W., Zhao X. M., Yin G. F., Zhou X. S., Wang J. W. (2022). Time scale and quantitative identification of sequence boundaries for the Oligocene huagang formation in the Xihu Sag, East China Sea Shelf Basin. Oil and Gas Geol. 43 (4), 990 – 1004. doi:10.11743/ogg20220419
  17. Liu Y., Lin C. Y., Lin J. L., Huang X., Liu B. B. (2024). Pore structure characteristics and genesis analysis of deep tight sandstone in Xihu depression, East China Sea Basin. Nat. Gas. Geosci. 35 (3), 405 – 422. doi:10.11764/j.issn.1672-1926.2023.07.006
  18. Luan X., Zhang H. (2003). Heat flow measurement and distribution of East China Sea and ryukyu trench arc back-arc system. Prog. Geophys. 18 (4), 670 – 678.
  19. Luan X., Gao D., Yu P., Zhao J. (2002). Thermal conductivity of the cenozoic layer of East China Sea shelf. Oceanol. Limnologia Sinica 33 (2), 151 – 159.
  20. McKenzie D. (1978). Some remarks on the development of sedimentary basins. Earth Planet. Sci. Lett. 40 (1), 25 – 32. doi:10.1016/0012-821x(78)90071-7
  21. Pollack H. N., Hurter S. J., Johnson J. R. (1993). Heat flow from the Earth's interior: analysis of the global data set. Rev. Geophys. 31 (3), 267 – 280. doi:10.1029/93rg01249
  22. Sekiguchi K. (1984). A method for determining terrestrial heat flow in oil basinal areas. Tectonophysics 103 (1-4), 67 – 79. doi:10.1016/0040-1951(84)90075-1
  23. Shen W. F., Yu Z. K., Diao H., Zhang T. (2020). Simulation of heat flow evolution history in Xihu Sag and its significance for hydrocarbon accumulation. China Offshore Oil Gas 32 (1), 42 – 53. doi:10.11935/j.issn.1673-1506.2020.01.005
  24. Suft O., Hagenauer H., Bertermann D. (2025). Relationship between thermal conductivity, mineral composition and major element composition in rocks from central and south Germany. Geosciences 15 (1), 19. doi:10.3390/geosciences15010019
  25. Tang X., Hu S., Zhang G., Liang J., Yang S., Rao S. (2014). Geothermal characteristics and hydrocarbon accumulation of the northern marginal basins, South China Sea. Chin. J. Geophys. 57 (1), 64 – 78. doi:10.6038/cjg20140221
  26. Tang X., Huang S., Yang S., Jiang G., Hu S. (2016). Correcting on logging-derived temperatures of the pearl river mouth basin and characteristics of its present temperature field. Chin. J. Geophys. 59 (8), 2911 – 2921. doi:10.6038/cjg20160816
  27. Wang J. Y. (2015). Geothermics and its Applications. Beijing: Science Press.
  28. Wang P. X., Prell W., Blum P. e.a. (2000). Proceedings of the Ocean Drilling Program, Initial Reports 184. Texas: Texas A&M University.
  29. Waples D. W., Ramly M. (2001). A statistical method for correcting log-derived temperatures. Pet. Geosci. 7 (3), 231 – 240. doi:10.1144/petgeo.7.3.231
  30. Waples D. W., Pacheco J., Vera A. (2004). A method for correcting log-derived temperatures in deep wells, calibrated in the Gulf of Mexico. Pet. Geosci. 10 (3), 239 – 245. doi:10.1144/1354-079302-542
  31. Xu W. L., Jiao R. C., Yue J. Y., Wei D. W. (1995). Geothermal study on the continent shelf of the East China Sea. Prog. Geophys. 10 (2), 32 – 38.
  32. Xu W., Tang X., Cheng L., Dong Y., Zhang Y., Ke T. (2022). Heat flow and thermal source of the Xi’an depression, weihe Basin, central China. Front. Earth Sci. 9, 819659. doi:10.3389/feart.2021.819659
  33. Xu B., Wu Z., Cheng Y., Miocic J. M., Dai Y., Chu Y. (2024). Cenozoic structural and tectonic evolution in the Western Xihu Basin, East China Sea Shelf Basin. Int. J. Earth Sci. 113 (3), 713 – 732. doi:10.1007/s00531-024-02396-9
  34. Yang S., Hu S., Cai D., Feng X., Chen L., Gao L. (2004). Present-day heat flow, thermal history and tectonic subsidence of the East China Sea Basin. Mar. and Petroleum Geol. 21 (9), 1095 – 1105. doi:10.1016/j.marpetgeo.2004.05.007
  35. Yao Q. W., Diao H., Chang J., Zhang W., Zhan R. H., Gao H. Y. (2026). Cenozoic tectonic-thermal evolution of the Xihu Sag, East China Sea Basin: implications for hydrocarbon generation in continental margin basins. Geothermics 136, 103530. doi:10.1016/j.geothermics.2025.103530
  36. Yu Z. K., Hou D. J., Shen W. F., Zhao H., Ding F., Diao H. (2023). Tectonic-thermal evolution and hydrocarbon generation history of coaly source rocks in Xihu Sag of East China Sea Basin. J. Yangtze Univ. Nat. Sci. Ed. 20 (4), 43 – 52. doi:10.16772/j.cnki.1673-1409.2023.04.008
  37. Yuan Y., Zhu W., Mi L., Zhang G., He L. (2009). Uniform geothermal gradient and heat flow in the qiongdongnan and pearl river mouth basins of the South China Sea. Mar. Petroleum Geol. 26 (7), 1152 – 1162. doi:10.1016/j.marpetgeo.2008.08.008
  38. Zhang J., Xu F., Zhong T., Zhang T., Yu Y. (2012). Sequence stratigraphic models and sedimentary evolution of pinghu and huagang formations in Xihu trough. Marine Geology and Quaternary Geology 32 (1), 35 – 41. doi:10.3724/sp.j.1140.2012.01035
  39. Zhang J. P., Zhang T., Tang X. J. (2014). Basin type and dynamic environment in the East China Sea Shelf Basin. Acta Geol. Sin. 88 (11), 2033 – 2043. doi:10.19762/j.cnki.dizhixuebao.2014.11.002
  40. Zhang Y., Zou W., Chen Z., Jiang Y., Diao H. (2023). The mechanism of “ convergence ahead of accumulation” and its geological significance for gas reservoirs in Paleogene huagang formation across the central inverted structural zone of Xihu depression, East China Sea Shelf Basin. Oil and Gas Geol. 44 (5), 1256 – 1269. doi:10.11743/ogg20230514
  41. Zhang Y. Z., Jiang Y. M., Diao H., Li N., Duang D. P., Li S. (2025). Exploration and development practice, new geological understanding and resource potential of low to ultra-low permeability gas reservoirs in Xihu Sag of East China Sea Basin. Acta Pet. Sin. 46 (6), 1074 – 1088. doi:10.7623/syxb202506004
  42. Zheng W., Yin G., Sun L., Wei S., Wei X., Niu B. (2024). Sedimentary systems of the Oligocene huagang formation in the central anticline zone of the Xihu Depression,East China Sea Shelf Basin. Energy Geosci. 5 (1), 100150. doi:10.1016/j.engeos.2022.100150
  43. Zhou H., Jiang X., Li D., Xing J., Gong W. (2017). Thermal-rheological property of lithosphere beneath Xihu Sag, East China Sea Shelf Basin. Geotect. Metallogenia 41 (3), 481 – 490. doi:10.16539/j.ddgzyckx.2017.03.005
  44. Zhou X., Jiang Y., Tang X. (2019). Tectonic setting, prototype basin evolution and exploration enlightenment of Xihu sag in East China Sea basin. China Offshore Oil Gas 31 (3), 1 – 10. doi:10.11935/j.issn.1673-1506.2019.03.001
  45. Zhu W., Zhong K., Fu X., Chen C., Zhang M., Gao S. (2019). The formation and evolution of the East China Sea Shelf Basin: a new view. Earth-Science Rev. 190, 89 – 111. doi:10.1016/j.earscirev.2018.12.009
  46. Zhu C., Chen C., Yang Y., Qiu N. (2022a). Experimental study into the factors influencing rock thermal conductivity and their significance to geothermal resource assessment. Petroleum Sci. Bull. 7 (3), 321 – 333. doi:10.3969/j.issn.2096-1693.2022.03.029
  47. Zhu C., Xu T., Qiu N., Chen T., Xu M., Ding R. (2022b). Distribution characteristics of the deep geothermal field in the sichuan basin and its main controlling factors. Front. Earth Sci. 10, 824056. doi:10.3389/feart.2022.824056

Citation

Diao, H., et al. (2026). Present-day thermal regime of the Xihu Sag, East China Sea Shelf Basin: constraints from an updated well database and logging-derived thermal conductivity. Frontiers in Earth Science, Volume 14. https://doi.org/
Received
Accepted
Published
Volume
14 - 2026

Copyright

Copyright © 2026 Diao, Shi, Cheng, Yu, Zhao, Xu, Ding and Yu.

License: https://creativecommons.org/licenses/by/4.0/

*Correspondence: Xiong Cheng

Article metrics

Views
0
Downloads
0
Citations
0
Shares
0

Related articles

Original ResearchMulti-constraint collaborative seismic imaging and interpretation method for fine characterization of small faults in complex coal mining areasNing Ma, Dong Yan, Liyong Qian, Pengcheng Niu, Xuefeng Tian, Jiandong Lin, Yong WangOriginal ResearchQuantified impact of projected climate change on groundwater recharge and river discharge leveraging the use of open access geospatial dataJulius Incillo Jimenez, Nitin Kumar Tripathi, Ashish Pandey, Sangam Shrestha, Kuo Chieh ChaoOriginal ResearchEffects of pressure on the stable carbon isotope composition of gas generated from continental type-I and type-III kerogen under semi-closed conditionsYanju Li, Hao Yang, Zhonghong Ning, Shengling Jiang, Biyang Feng, Xiulan Zhu, Wanchang Zhang, Xianghe Ma