Preparation for the CCSD project can be traced back to 1988. Encouraged by the drilling projects of the Kola superdeep borehole and the Kontinentales Tiefbohrprogramm der Bundesrepublik Deutschland (KTB), Chinese scientists proposed a continental scientific drilling plan in mainland China in 1988. The Ministry of Geology and Mineral Resources of China, which was combined in the Ministry of Land Resources of China in 1998, started to investigate potential scientific drilling sites in 1991. After years of investigation and discussion, China founded the ICDP with Germany and the United States in February 1996. In July 1996, Chinese scientists hosted the ICDP workshop in Qingdao, eastern China, and proposed three UHP localities in the Dabie–Sulu orogenic belt from Jiangsu, Anhui, and Shandong provinces. Based on outcrops of UHP metamorphic rocks, seismic profiles, shallow boreholes, and technical feasibility, the Science Advisory Group of the ICDP agreed to select Maobei village in Donghai County (Jiangsu Province) as the drilling site of the CCSD project.

In June 1997, the CCSD project was listed as one of major scientific projects in the Ninth Five-Year Plan of China. This was an important step to get financial support from Chinese government. The first pre-pilot borehole CCSD-PP1 in the Zhimafang ultramafic massif in Donghai County was accomplished at a depth of 432 m in November 1997. Then in April 1998, the ICDP passed the drilling proposal of the CCSD project and funded USD 1.5 million in the form of drilling facilities. Drilling of the second pre-pilot borehole CCSD-PP2 at Maobei village in Donghai County started in December 1998, and reached a depth of 1028.68 m in June 1999. The CCSD-PP2 was the first scientific borehole with complete sampling and logging in China, and succeeded in core recovery of 92.8 %. Chinese engineers and technicians also used the CCSD-PP2 to test different drilling techniques and materials, as well as on-site measurements. In September 1999, Chinese government agreed the budget of the CCSD project. As the largest and most expensive geoscientific research project ever undertaken in China before 2005, the total cost of the CCSD project was CNY 176 million (about USD 21.3 million).

The CCSD project was the first scientific drilling project in an UHP terrane. The ICDP provided important help by inviting Chinese scientists and engineers to visit the KTB and attend training courses. Test of all drilling facilities and on-site measurements for the CCSD main borehole (CCSD-MH) began on 25 June 2001. Experts from the ICDP came to the drilling site at Maobei village to provide technical support. Chinese engineers developed new drilling and core recovery tools. With funding from the Ministry of Science and Technology, the Ministry of Land and Resources, the National Natural Science Foundation of China and the ICDP, the CCSD-MH was drilled to a depth of 5158 m from 4 August 2001 to 8 March 2005 (Fig. 1). The average core recovery of the CCSD-MH reaches 85.7 %. Meanwhile, the third pre-pilot borehole CCSD-PP3 penetrated 705 m into the Gangshang ultramafic massif in Ganyu County, with about 90 % core recovery.

The principal investigator (PI) of the CCSD project was Chinese Academician Zhiqin Xu, and the drilling project leader was Da Wang, a professor and senior engineer from the Geological Survey of China. More than 2000 people were involved in the CCSD project, including about 120 Chinese scientists and technicians and 30 scientists from the United States, Germany, France, and Japan.

The Dabie–Sulu orogenic belt was formed by northward subduction of the South China Block beneath the North China Block in the Triassic, and was then separated by the left-lateral strike-slip Tan-Lu fault into the Dabie Mountains in the west and the Sulu Terrane in the east (Fig. 2) (e.g., Cong, 1996; Xu et al., 2009a). The wide occurrence of coesite, microdiamond, UHP hydrous phases, and exsolution textures in eclogites, garnet peridotites, and their supracrustal country rocks indicates that the South China Block has been subducted to depths > 120 km at extremely low geotherms (for review see Liou et al., 2009). Most UHP metamorphic rocks in the Dabie–Sulu orogenic belt yield the Neoproterozoic protolith ages of 780–740 Ma and extremely low δ18O values, which suggests their tectonic affinity to the South China Block (Zheng, 2008, and references therein). It is generally believed that the UHP metamorphism in the Dabie–Sulu orogenic belt occurred in the Triassic, but different methods result in large variations: ∼ 245 Ma from zircon U–Pb ages (Hacker et al., 1998) or 226 ± 2 Ma from a mineral Sm–Nd isochron dating (Li et al., 2000). Sensitive high-resolution ion microprobe (SHRIMP) U–Pb dating on coesite-bearing domains of metamorphic zircon bracketed the UHP metamorphism between 240 and 225 Ma, and the amphibolite facies retrogression between 215 and 205 Ma (e.g., Liu et al., 2004a, b, 2006; Wu et al., 2006). This suggests that the subducted crustal rocks experienced a rapid syn-collisional exhumation at mantle depths.

Both the Dabie Mountains and the Sulu Terrane consist of a series of high-pressure (HP) and UHP metamorphic slices bounded by ductile shear zones (Fig. 2a). The drilling site of the CCSD-MH is located in the northern UHP supracrustal zone in the Sulu Terrane, where eclogites and garnet peridotites appear as lenses, pods, or layers within coesite-bearing gneisses, quartzite, mica schists, and amphibolites. The northern UHP granitic zone is dominated by granitic gneisses and has been strongly modified by Cretaceous granitic plutons. Exhumation of the HP and UHP slices in the Triassic produced top-to-the-northwest (NW) ductile shear zones in the Sulu Terrane (Fig. 2b). Despite the post-collisional crustal extension and granitic intrusion, the structural relationship between the HP and UHP slices is well preserved in the Sulu Terrane. Therefore, the CCSD-MH provides us a natural laboratory to investigate physical and chemical properties and geodynamic processes in a continental collision zone.

The objectives of the CCSD project were as follows: (1) to obtain multi-parameter profiles of a 5158 m deep borehole in the Sulu Terrane, (2) to reconstruct the composition and structure of a deep continental orogenic root, (3) to reveal subduction and exhumation processes of UHP metamorphic terranes, (4) to search deep life in the borehole and constrain fluid–rock interaction, and (5) to establish a long-term observation station using the CCSD-MH.

The CCSD project has produced enormous amounts of data as well as technical innovation in drilling and core recovery. Combined with downhole logging, geophysical experiments, field survey, and laboratory measurements, fresh core samples from the CCSD project provide a valuable resource to study the structure, composition, deformation, physical properties, and material cycling in a continental subduction zone. These results made the Sulu Terrane a classic example to other continental collision zones such as the Himalayas and the Caledonides. In addition, a long-term observation station was set up in the CCSD-MH with continuous downhole monitoring on seismicity, fluid activity, stress, and temperature.

The main progress of the CCSD project is summarized below.

Multi-parameter profiles of the 5158 m deep CCSD-MH. A full range of physical and chemical measurements of core samples, as well as downhole logging, cuttings, and fluid measurements were carried out for the 5158 m deep CCSD-MH. About 60 precise and oriented profiles of the CCSD-MH were established, including petrology (Fig. 3), structure (Fig. 4), geochemistry, petrophysics, wellbore breakout, mineralization, fluids, geochronology, oxygen isotopes, heat production elements, downhole logging, and geophysical experiments. These multi-parameter profiles provide a new window into the deep root of a continental collision belt and set up the correlations between chemical and physical properties of UHP rocks.

The modern Indian summer monsoon (ISM) is characterized by exceptionally strong interhemispheric transport, indicating the importance of both Northern and Southern hemispheres processes driving monsoon variability. The first borehole of the CESD penetrated 666 m sedimentary sequence in the Heqing paleolake in the southeastern margin of the Tibetan Plateau in 2002. It is so far the longest high-resolution continental record for the ISM evolution.

High-resolution continental record from the Heqing Basin demonstrates the importance of interhemispheric forcing in driving ISM variability at the glacial–interglacial timescale. Interglacial ISM maxima are dominated by an enhanced Indian low associated with global ice volume minima. In contrast, the glacial ISM reaches a minimum, and actually begins to increase, before global ice volume reaches its maximum (Fig. 8). An et al. (2011) indicated that the changing ISM was driven by the relative dominance of the northern low-pressure and southern high-pressure systems during the Pleistocene; both are indispensable for understanding monsoon dynamics. They attribute this early strengthening to an increased cross-equatorial pressure gradient derived from Southern Hemisphere high-latitude cooling. This mechanism explains much of the non-orbital scale variance in the Pleistocene ISM record. The new insight into the ISM on the geological timescale is important for our understanding of global climate evolution in the past. It also provides background information for present and future monsoon changes under global warming (An et al., 2006).

As the second ICDP project in China, Lake Qinghai Drilling Project was conducted in 2005. A total of 323.23 m cores were acquired with GLAD800 drilling system by US Lake Drilling Company (DOSECC) at different sub-basins in Lake Qinghai. In addition, two onshore sites were drilled successfully at Erlangjian (1108.9 m with core recovery > 90 %) and Yilangjian (628.5 m) on the southeastern shore of the lake using Chinese equipment (An et al., 2006).

Lake Qinghai is located in the northeastern margin of the Tibetan Plateau. It is extremely sensitive to climate changes because it lies in a critical transitional zone between the humid climate region controlled by the East Asian monsoon and the dry inland region affected by westerly winds. High-resolution magnetostratigraphy of Yilangjian core samples provides a chronological record back to ∼ 5.1 Ma (Fig. 9). Analysis of lithofacies and depositional environments reveal that the change from eolian to lacustrine facies occurred ∼ 4.63 Ma, corresponding to a shift from an arid/semi-arid to a more humid climate, which resulted in the origin of Lake Qinghai (Fu et al., 2013). Changes in sediment lithology and mean grain size indicate that the lake level fluctuated considerably, superimposed on a long-term trend from higher to lower levels in response to East Asian Monsoon variations. This archive provides important information on regional and global environmental change, because so far records from northern China are mainly based on analysis of loess.

The longest core in the Lake Qinghai was obtained in the depositional center of the southwestern sub-basin. The results, constrained by high-resolution AMS 14C dating (An et al., 2012), show the variability of both the Westerlies and the Asian summer monsoon (ASM) since 32 ka. These records document the anti-phase relationship of the Westerlies and the ASM on both glacial–interglacial and glacial millennial timescales (Fig. 9). During the last glaciation, the influence of the Westerlies dominated. Prominent dust-rich intervals, correlated with Heinrich events, reflect intensified Westerlies linked to northern high-latitude climate. During the Holocene, the dominant ASM circulation, punctuated by weak events, indicates linkages of the ASM to orbital forcing, North Atlantic abrupt events, and perhaps solar activity changes (An et al., 2012).

On 12 May 2008, the Mw 7.9 Wenchuan earthquake happened along the northeast (NE)-striking Longmen Shan thrust belt in the eastern margin of the Tibetan Plateau and caused death of over 69 227 people and enormous economic loss. In contrast to extremely low GPS shortening rate in the Longmen Shan thrust belt before the earthquake (Zhang et al., 2004), the Wenchuan earthquake is characterized by > 4 m coseismic slip with comparable magnitude of reverse and right-slip components. It produced 270 km long surface rupture along the NE-striking Yingxiu–Beichuan fault, and 80 km long surface rupture along the Anxian–Guanxian fault simultaneously (Li et al., 2008; Liu-Zeng et al., 2009) (Fig. 10).

In order to better understand the fault mechanism and the physical and chemical characteristics of the eastern margin of the Tibetan Plateau, Chinese scientists carried out the WFSD project just 178 days after the earthquake, as an extremely rapid response to the devastating Wenchuan earthquake. The PI of the WFSD project was Zhqin Xu from Chinese Academy of Geological Sciences. The WFSD project eventually drilled six boreholes along the two main earthquake faults (Fig. 11). The first hole (WFSD-1) started on 6 November 2008 and was completed on 12 July 2009, with a total depth of 1201.15 m.

The scientific objectives of the WFSD project are as follows: (1) to determine the composition, texture, and structure of the fault zones; (2) to reconstruct the physical and chemical properties of the fault zones during the earthquake (frictional coefficient, pore pressure, stress, permeability, seismic velocities, mineral, and chemical compositions), energy budget, and rupture processes; (3) to improve our understanding of the transpressional behavior of this fault zone; (4) to provide key information for the early warning system of future large earthquakes; and (5) to establish a long-term earthquake monitoring station by putting instruments the boreholes.

So far the main results from the WFSD project are as follows:

Based on petrological and structural analysis of the core samples from the WFSD project, fault-related rocks (fault gouge, cataclasite, and fault breccia) in the Yingxiu–Beichuan fault and the principle slip zone (PSZ) of the Wenchuan earthquake were determined. Twelve of the 12 fault zones were identified in the entire core profile. At least 10 fault zones, including the Yingxiu–Beichuan fault zone, show a minimum thickness of ∼ 100 m. All major fault zones are characterized by high magnetic susceptibility, low density, and high porosity, with mostly low resistivity, high natural gamma ray, and sound wave velocity (Fig. 12). The high magnetic susceptibility values most likely result from the transformation of magnetic minerals by frictional heating due to the earthquake (Li et al., 2013).

3.

The dominant fractures in the core samples of the WFSD-1 above 600 m are dipping south with a dip angle of 60∘, and change to the NE dipping with a dip angle of 61–83∘ below 780 m, which represent the fracture characteristics in the Pengguan Complex and the Xujiahe Formation, respectively. Hence, the hanging wall and footwall of the Yingxiu–Beichuan fault have different stress fields, implying that the Pengguan Complex is allochthon (Li et al., 2014).

The SinoProbe Program is the Chinese government-funded Earth science program with the overall aim of exploring the composition, structure, and evolution of the continental lithosphere beneath China and the processes causing geo-hazards and natural resources (Dong et al., 2013). As one part of the SinoProbe Program, the Program of Selected Continental Scientific Drilling and Experiments (SCSDE) consisted of seven drilling projects in critical tectonic and mineral resource regions from 2008 to 2012. It includes the Jinchuan Cu–Ni sulfide deposits in Gansu Province, the Luobusa chromitite deposits in Tibet, the Tengchong volcano-thermal tectonic zone in Yunnan Province, the boundary of the North China and South China blocks in the Laiyang Basin of Shandong Province, the Yudu–Ganxian polymetallic deposits in South China, the Tongling polymetallic deposit, and the Luzhong volcanic basin and mineral deposit district in Anhui Province (Fig. 15). The PI of the SCSDE Program is Jingsui Yang from Institute of Geology, Chinese Academy of Geological Sciences.

Ophiolites along the 2000 km long Yarlung Zangbo suture zone in the Tibetan Plateau mark the collision boundary between the Indian and Eurasian plates. These ophiolites mainly consist of harzburgites and dunites, with minor gabbro dikes and basalts. The Luobusha ophiolite massif contains significant podiform chromitites, which appear as lenticular boudins of high-chrome chromite within dunites (Zhou et al., 1996). This ophiolite massif is believed to have formed initially at an oceanic spreading center at 176 Ma and then been modified by melt–rock reaction in a suprasubduction zone at 126 Ma. However, discovery of diamond and some UHP minerals in the Luobusha ophiolite challenges the traditional formation theory of ophiolites (Yang et al., 2007).

The Luobusa Scientific Drilling Project aims to reveal the formation conditions of the deep mantle minerals, the metallogenic mechanism and prospecting criteria of ophiolite-type chromite deposit. The first borehole LBSD-ZK1 ended at a depth of 1478.8 m with the average core recovery of 94 %. The second borehole LBSD-ZK2 has reached a depth of 1600 m with the average core recovery of > 90 %, and the target depth is 2000 m.

The lithological profile of the LBSD-ZK1 indicates that the upper 70 m is composed of Triassic sandstones, marbles, and chlorite schists, which are in a fault contact with the underlying ultramafic massif. Beneath the fault, there are 1400 m thick ultramafic rocks. The rest over 100 m cores are cumulates of dunites and pyroxenites, suggesting that the Luobusha massif is a tectonic slice with a reversed strata sequence (Fig. 16).

Recent studies found abundant diamonds from peridotites and chromitites in ophiolites along the Yarlung Zangbo suture zone, the early Paleozoic Ray–Iz ophiolite in polar Urals (Russia), and in Myanmar (Fig. 17). Chromitites from ophiolites contain crustal minerals such as zircon, and some highly reducing minerals such as carbides, nitrides, and metal alloys (Yang et al., 2014). Some of the minerals are very high-pressure phases, implying their origin depth near the top of the mantle transition zone. These new results suggest that crustal materials may be subducted to mantle transition zone and subsequently exhumed during the initiation of new subduction zones.

The Songliao Basin in northeastern China is one of the largest oil- and gas-bearing basins in China. Above a pre-Mesozoic basement, the Songliao Basin was filled predominantly with volcanoclastic, alluvial fan, fluvial, and lacustrine sediments of Late Jurassic, Cretaceous, and Paleogene ages. It can be divided into the six tectonic units: western ramp region, northern plunge region, central depression region, northeastern uplift region, southeastern uplift region, and southwestern uplift region (Fig. 18) (Wang et al., 2013a).

There is an increasing likelihood that our world is moving toward a greenhouse state due to increases in anthropogenic carbon dioxide in the atmosphere (Hay, 2011). This change provides a powerful motivation for understanding the dynamics of Earth's past “greenhouse” climate. Cretaceous is characterized by long-term climate stability with warm equable climates due to a higher atmospheric greenhouse gas content, and punctuated by rapid climate events related to global carbon and hydrological cycle (Bornemann et al., 2008). The Continental Scientific Drilling Project of the Cretaceous Songliao Basin, composed of SK-1 and SK-2 boreholes, was proposed to recover almost full Cretaceous strata in the Songliao Basin in eastern China (Feng et al., 2013). The PI of this project is Chengshan Wang from China University of Geosciences (Beijing).

The newly finished SK-1 borehole has recovered Upper Cretaceous strata in the Songliao Basin, which offered an unprecedented chance to decipher the Cretaceous terrestrial climate change. The core samples reflect many facets of paleoenvironmental and climate changes during the Late Cretaceous in eastern China, which have attracted a wide range of Earth scientists to compare with other areas in the world.

Base-level cycles are driven either by relative sea level or by tectonic changes. In marine basins eustasy alters accommodation space as sea level rises and falls. In intracontinental basins tectonics drive both basin subsidence and depositional cycles. Core samples from the SK-1 indicate that the Songliao Basin was isolated from the ocean except for two brief periods in the late Turonian of the Qingshankou Formation and in the early Campanian in the lower part of the Nenjiang Formation (Xi et al., 2011). Paleoclimate research has shown that astronomical forcing drove global climate change in the Earth's history and may be recorded in both continental and marine sedimentary strata. The orbitally forced cycles are recorded in Cretaceous marine successions around the world, which provide a calibration tool of the Cretaceous timescale. Milankovitch cycles were recently identified in Late Cretaceous terrestrial strata in the Songliao Basin (Wang et al., 2013a).

High-resolution sequence stratigraphy of the Upper Cretaceous section cored in the SK-1 (Fig. 19) was based on centimeter-by-centimeter description integrated with well logs, seismic data, and well-to-well correlation in the Songliao Basin. Upper Cretaceous of the Songliao Basin can be subdivided into sequences of 2 second order, 6 third order, 72 fourth order, 527 fifth order and 1601 sixth order (Gao et al., 2009; Wang et al., 2013b).

Cyclostratigraphic analyses on the Quantou, Qingshankou, Yaojia, and Nenjiang Formations in the SK-1 south well confirmed the excellent preservation of astronomically driven sedimentary cycles, including long and short eccentricity, obliquity, and precession cycles constrained by four high-resolution SIMS U–Pb zircon ages (Fig. 20).

The climate history of the Songliao Basin is based on pollen and spore ratios (Fig. 21) using more than 20 000 samples from more than 500 cores. The Cretaceous vegetation landscape of the basin fluctuated between a conifer forest and an herbaceous-broadleaved forest. Climates of the Songliao Basin are compared with the marine oxygen isotopic data of Far East (Zakharov et al., 1999) (Fig. 21). The classification of temperature zones is based on the species of spore/pollen spectra that define the tropical, tropical–subtropical, subtropical, tropical–temperate, and temperate, as well as the oxygen isotopic trends of the Songliao Basin. The dryness and humidity are based on species of parent plants of spore/pollen fossils subdivided into xerophyte, mesophyte, hygrophyte, helophyte, and hydrophyte, which correspond to arid, semiarid, semihumid and semiarid, semihumid, and humid, respectively.

Partially based on the successful SK-1 borehole, the ICDP approved the SK-2 drilling proposal in September 2009. The SK-2 borehole is to recover Lower Cretaceous strata with an estimated drilling depth of nearly 7000 m. Hence, the SK-1 and SK-2 boreholes can provide continuous sedimentary records of the Cretaceous in the Songliao Basin. The drilling operation of SK-2 started in April 2014, because Chinese engineers developed new drilling equipment for a superdeep borehole of ∼ 10 000 m. The SK-2 borehole also adopted “one-well, two-hole” design (Fig. 18). The west hole SK-2w is located in Changyuan of Daqing City and will recover the Quantou–Denglouku Formation. The east hole SK-2e is located in the Xujiaweizi Fault Depression, and will recover the Yingcheng Formation–Huoshiling Formation. The two holes will be interconnected and correlated by the T4 seismic reflection interface. The east and west holes of SK-2 will become the main wells of the proposed deep underground laboratory. The target depth of the east hole SK-2e is 6400 m, which will be the deepest one in the Songliao Basin and in eastern China. More comprehensive understanding of paleoenvironmental and climatic change for the Cretaceous Songliao Basin would be achieved after the accomplishment of the SK-2. Compared with drilling projects in the ocean, loess plateau, and modern lakes, scientific drilling projects in giant paleolakes should be an indispensable part to understand the paleoenvironments and climatic changes from the terrestrial realm.